EX LIBRIS.

Bertram C. a. OTlintile,

ILil.Q., Q.Sc, m»^.(^o JF.ia.5*

DUBLIN REVIEW. S.R.B.
THE second edition of Dr Rice Holmes's monumental
work CcB sax's Conquest of Gaul (Oxford: at the
—
Clarendon Press. 191 1. pp. xl 872. Maps and Plans.
Price 24s. net) is a revised and largely rewritten successor
to the first issue which, when it appeared in 1899,
attra6led so much attention on the part of scholars. This
new issue is regarded byauthor as being final in its
its
present form. It consists of an account of the Conquest of
Gaul and of a number on points arising in
of excurses
connexion with the epoch of history, and especially in
connexion with Caesar's Commentaries^ on which any
description of the operations in question must be founded.
The first mentioned part of the book may be described
as a digest of Caesar's Gallic War, written in a lively and
fascinating style, illustrated by maps and plans, and made
so attractive to the reader that we could wish that it
might be published by itself and put into the hands of
boys who are about to be launched upon the study of the
Commentaries. For, if one may judge from one's own
experience, the human boy at that period of his career
generally fails to grasp the fa6f that the persons and
occurrences described in those immortal pages are real
beings and real events, but rather is inclined to look upon
them as tiresome and vexatious imaginings, probably
conceived, by some one of more than ordinary brutality,
expressly and solely for the wearying and exasperation of
youth. Hence his languid interest and his utter failure to
appreciate the real and extraordinary fascination in, for
example, the operations against Vercingetorix. We should
like to see the experiments tried of reading this account to
a class of boys with a large wall-map on which the opera-
tions could be easily followed, and then setting them to
work upon the Conunentaries as an ordinary Latin task.
Dull indeed must be the boy whose interest failed to be
roused by the spirited account w^hich forms the basis,
though much the smaller part, of Dr Rice Holmes's work.
It is to the excurses that scholars of all kinds will mainly
direft their attention and these will appeal to three classes
of persons. Classical scholars, of course, come first, and
they and historical students will find abundance of
material for consideration and for criticism. But, perhaps,
especially to anthropologists and ethnologists will these
portions of the work be of absorbing interest. For Dr Rice
Holmes grapples with all sorts of problems at present
subje6ts of fierce debate and wholly unsettles, and crosses
swords with experts like Sir John Rhys and Professor
Ridgeway. To touch upon these problems even briefly is
not possible within the limits of this notice, but enough
has been said to show in outline the intention of the book.
Like its companion work from the same pen, Ancient
Britain and the Invasions of Julius Ccesar which, on its
appearance, was welcomed by the Dublin Review, the
Conquest of Gaul is the outcome of the ripe knowledge and
judgement of a distinguished scholar; it is packed with
information and there are thousands of references to the
literature of the subje6f, and, in fine, it is a book which no
library of any institution devoted to learned nursnii-s, nor
any scholars' library either, can afford to be without.
B.C.A.W.
 Digitized by the Internet Archive
in 2011 with funding from
University of Toronto

http://www.archive.org/details/caesarsconquestoOOholm
U nlJuriiDotiUAl( .pl> -<-
 CAESAR'S
CONQUEST OE GAUL

BY

T. RICE HOLMES
Hon. LriT.D. (Dublin)

SECOND EDrnON
EEVISED THEOUGHOUT AXD LAEGELY EEWEITTEN

OXFORD
AT THE CLARENDON PRESS
1911
 APR 13
6807

HENRY FROWDE, M.A.

PUBLISHER TO THE UNIVERSITY OF OXFORD
LONDON, EDINBURGH, NEW YORK
TORONTO AND MELBOURNE
 PREFACE
The first edition of this book was exliausted in January,
1909 ; had been found useful by many scholars
and as it

both in this country and abroad, and had also attracted

general readers, it seemed advisable to re-issue it. But in
consequence of its own defects as well as of the progress of
research in some of the subjects with which it deals, it

needed revision ; and this task has required the almost

incessant labour of two years. Indeed, the book has been
not only revised, word by word, but largely rewritten.
The narrative is increased by thirty-seven pages, of which
twenty belong to the Introduction. In Part II several short
articles have been struck out altogether ; many passages
that either were originally or had become superfluous have
been deleted : many have been corrected or amended in
substance or in style. Certain articles have been lengthened,
either because it had become necessary to take note of fresh
information or of recent theories or because the argument
had not been completely thought out while others are ;

entirelv new. The instances in which conclusions reached

in the old edition have been altered are, however, rare.

I wish to express my gratitude to Dr. H. Meusel, the leading

Caesarian scholar of Germany, who kindly sent me a proof
of his important article in the Jahresberichte des yhilologi-
schen Vereins zu Berlin of 1910, and also a transcript of
the notes which he had v/ritten on the margin of his copy
of my first edition, thus enabling me to correct inaccuracies
and misprints, chiefly in footnotes to Dr. A. Gudeman, ;

who sent me an extract from the proof of one of his articles

in the Thesaurus linguae Latinae, which I have utilized on
page 663 ; and to i\Ii'. Stanley Hall, A.R.I.B.A., who wrote
iv PREFACE
for my use a paper, from which I have made exceipts
on pages 719-20, embodying his interpretation of a dis-
puted passage in Caesar's description of the bridge which
he built over the Rhine. Elsewhere I have done my best
to recommend to English readers M. Camille Jullian's
Histoire de la Gaule, which is in course of publication ; and
as sundry passages in which I have referred to it relate to
matters on which we differ, I am the more anxious to acknow-
ledge what I owe to him. Some of his footnotes have led
me had overlooked, to
to read books or articles which I
grasp the significance of words which I had not sufficiently
weighed, or to modify opinions which I had expressed before.
His plan is different from mine for his book contains
;

nothing which corresponds to Part II of this volume.

The book in its present form may be regarded as final ;

for the edition is large, and I do not expect that another

will be demanded while I live, or that I should have time
or inclination to make further alterations. Moreover,
although minute details are constantly being added to our
knowledge and although the criticism
of prehistoric Gaul,
of Caesar's text has been continued with zeal and not wholly
without profit, it is unlikely that we shall ever know much
more about Caesar's conquest, with which this book is mainly
concerned, than we do now.

11 DouRO Place, Kensington, W.

March 21,1911.
 PREFACE
TO THE FIRST EDITION
As this book has far outgrown my original conception,
I will explain how it came to assume its existing form.
Eleven years ago it occurred to me that an English narrative
of Caesar's conquest of Gaul might help to relieve the weari-
ness of the schoolboys whose lot it is to flounder, in ceaseless
conflict with the Ablative Absolute, through the pages of the
Commentaries ; might help them to realize that those pages
were not written for the purpose of inflicting mental torture,
but were the story of events which did really happen, and
many of which rival in interest the exploits of Cortes or of
Clive. hoped too that a few general readers might, if
I ' '

they could overcome their aversion to the title of the book,

find something to interest them in its contents. In my
ignorance I promised myself a comparatively easy task.
Certain chapters of history, which I had written before, had
cost me prolonged research and anxious toil. For the history
of the Gallic war, on the other hand, I imagined that vir-
tually the sole original authority was the Memoirs of the
conqueror. Virtually the sole original authority, but so
great a one that would be impossible, I thought, for
it

a man who honestly worked upon it to produce a really

bad book. So I said to myself, Let me once master the
Commentaries, and it will go hard with me if I cannot, witli

the aid of Napoleon's Histoire de Jules Cesar, and sundry

other books which I must of course consult, evolve from
such material a readable narrative. I shall be spared the
labour of searching through Blue Books, forgotten memoirs
and dusty bundles of MSS. It is needless to say that I soon
found out my. mistake. The list of the '
sundry other books '

DC
. C 3 H8
vi PREFACE TO THE EIRST EDITION
was continually lengthening. Though for the narrative as
a whole, Caesar is virtually the sole original authority —for
Plutarch and Suetonius, Dion Cassius, Florus and Orosius
do not count for much —yet, in order to understand his
military system and to supplement the information which
he gives on certain points, we are obliged to have recourse to
many other writers, ancient and mediaeval, historians,
geographers, chroniclers, compilers of itineraries. He has
left many questions obscure, — questions of geography, of
ethnology, of sociology, of religion, of politics, and of mili-
tary science. To throw light upon these questions, and to
explain the difficulties in his language, has engaged the
labour of a host of scholars, —geographers, antiquaries, an-
thropologists, ethnologists, archaeologists, military specialists,
philologists, learned editors ; and the works which they
have produced, the greater part of which are scattered in the

learned periodicals of foreign countries, would fill a large

library. If the bulk of these works are mainly controversial
or exegetical, if they are largely devoted to the discussion
and elucidation of ancient texts, yet on this point or on that
many of them are virtually original authorities. They con-
tain scraps of genuine information, which enable one to fill

up gaps in the memoirs of the conqueror. Excavators have

discovered disputed sites. Coins, inscriptions, rusty weapons,
and even skulls have added items to our store of know^ledge.

Soldier-scholars, trained to observe the geographical features

of a country, have travelled, Commentaries in hand, through
the length and breadth of France and Belgium and Alsace and
Switzerland and, if prejudiced zeal or local patriotism have
;

often misled them, theirunitedlabourshavenot been in vain.

Nor was this all. It was not enough for my purpose
merely to write a narrative of the conquest. I was obliged
of course also to write an introduction, in order to render

my narrative of the conquest intelligible ; and gradually it

became evident that, if I wished to avoid defrauding and
insulting the purchasers whom I hoped to attract, even this
 PREFACE TO THE FIRST EDITION vii

brief chapter could not be written without recourse to the

most recondite materials. Since the publication of the
standard histories of Thierry, Mommsen, Merivale, and
others, new had been thrown upon the ethnological
light
and other questions which I had set myself to handle. Some
opinion I must hazard regarding the degree of political
development which the Gauls had reached and, if it were ;

to be worth printing, I must form it at first hand. I had

no intention of writing a history of the Gauls : my subject
was only their conquest by Julius Caesar ; but I was bound
to take as much pains to understand their history as if

I had been ambitious of writing it. As I plunged deeper

and deeper into the slough, I saw that many of the problems
were insoluble but this did not absolve me from the duty
;

of grappling with them. Even if a historical or geographical

problem cannot be solved with mathematical certainty,
probability may be attainable ; and if one solution is as
good or as bad as another, the reader has a right to ask the
reason why. It is something even to fix precisely the extent
of one's ignorance. Either I must leave the subject alone,
or I must master it. If the study of Caesar is arduous, it is
fascinating. Year after year I read on and on, quite as
much for the delight of learning as with the ambition of
instructing. And I determined to do my best to produce
something which should not only be useful to teachers and
interesting to general readers, but should also be worthy of
the notice of scholars and of students of the art of war.

To praise the Commentaries of Caesar, laudatos toties

a laudatis, would be almost impertinent. But I may be

allowed to say why I hope that a better fate may yet be in
store forthem than to serve as a mere whetstone for gerund-
grinders. At present, I believe that the book is rarely used
in education, at least in this country, except by young boys,
and never read through by them. But, even if only one or
two of the seven Commentaries can be read, they can at all
events be read not merely as a lesson in construing but also
viii PREFACE TO THE FIRST EDITION
as history. Something, I gladly acknowledge, has already
been done to promote this object. Much, however, still

remains to be done. Unfortunately, the editions of the

Commentaries which have been published in this country-
are defective, especially in the department of geography.
Most of the editors are far too prone to submit to the autho-
rity of Xapoleon. them who have worked in the
Those of

most intelligent spirit, sometimes, for want of drudgery,

lead their readers farthest astray. I know of one who,

inspired by the hope of firing the imagination of youthful

scholars, embellished his edition with pictures with which
only one fault could be found, —that the greater number
represented places where Caesar had never been. If a little

knowledge is a dangerous thing, a little research is labour

thrown away. The fact is that, if a man professes to explain
the geography of the Gallic war, he must do one of two
things. Either he must go into the subject as an indepen-
dent inquirer, pursuing his researches whithersoever they
may lead him —and to do this requires an amount of labour
so enormous that it would not pay the editor of a school-
—
book to undertake it or he must take Xapoleon, or some
such writ;er, as his guide in which case he will assuredly
;

be led into a great many mistakes.

Nor is there any reason, apart from the consideration of
what subjects are most remunerative, why Caesar should
only be used as an elementarv' textbook. The reform which
I hope to see one day accomplished is that he should be

read by more advanced students as well. Bo^^s m the

highest class of a public school could easily read the whole
work through, side by side with other authors, in the course
of a couple of years. By doing so, their knowledge of Latin
would gain at least as much as
their knowledge of history.
I do not know whether educationists will consider this

ideal desirable. But is it even attainable ? Not certainly

at present. It does not '
pay '
to teach Caesar to the more
advanced scholars of public schools. If there is ever to be
 PREFACE TO THE FIRST EDITION ix

a reform, it must begin with the universities. And there is

another class of students for whom the Commentaries would

be peculiarly appropriate, —the candidates for the Royal
Military Academy and for the Royal Military College.
But book is not addressed only, not even primarily,
this
to teachers and for pupils, in its present form, it is of
;

course too costly and too large. The narrative is addressed

both to scholars and to those general readers, civil and
military, who are interested in history. The second part is

addressed in the first instance to scholars ; and if it wins

their approval, I hope that the labour spent upon it will not
repel other readers who are willing to be interested in the
subject. Of all that has been done in France, Germany,
Italy, and Belgium to solve the problems of Gallic history
nothing is known in this country, except to a few students.
And yet to those who care for history the study would be
full of entertainment. The story of the conquest of Gaul, if

that of any war of antiquity, is still worth reading ; for not

only were the operations intrinsically interesting, but their
results are of permanent unportance. Mr. Freeman was
right when he called the conquest ' one of the most important
events in the history of the world '} The war with Hannibal,
and it alone, rivals the Gallic war in interest. And the Gallic
war has this great advantage over the war with Hannibal,
that we know far more about it. Viewed simply as military
history, intelligible without being technical, the Commen-
taries are by far the most valuable work of antiquity : they
are among the most- valuable of any age. Let any soldier
who possesses a fair knowledge of Latm read Livy's de-
scription of the battle of Cannae : him then read
let
Caesar's description of the battle with the Nervii, and he
will have made up his mind. He will appreciate the differ-
ence between military history as written by a mere literary
and military history as written by a
artist literary artist
who was also a general.
* General Sketch of European History, 1874, p. 77.
X PREFACE TO THE FIRST EDITION
I said that I would not take upon myself to praise
the Coynmentaries ; but when one has derived great and
wholesome pleasure from a book, it is hard to refrain from
expressing one's gratitude and admiration. Xot to repeat
encomiums that who take any interest in
are familiar to all

the classics, I will orAy speak my own thoughts for I would ;

fain persuade all who have not wholly forgotten their Latin
— all who love good literature all who can appreciate an;

informing story well and truly told —to get a copy of Caesar,
and read him through from end to end. I sometimes wish
that the book had never been used, in the way it has been
used, as a school-book at all. For the reminiscences of the
Fourth Form are at once so vivid and so dreary, that even
classical scholars, many of them, pass through life without
reading this great classic. In boyhood they plodded through
the pages, chapter by chapter, forgetting one chapter before
they began the next, reading one book and missing the others,
and of the whole story or even of smgle episodes forming no
idea. Some critics say that the narrative is dull, cold, and
colourless. do not believe that any one would maintain
I
these charges if he read the book rapidly through and ;

otherwise no story can be fairly judged. Macaulay himself

might be dull, if he were read by a foreigner at the rate of
a single paragraph a day. Caesar certainly did not pour out
his spirit with the fervid passion of a Xapier. But if a man's
heart beats faster when he reads how Badajoz was stormed,
and how "
six thousand unconquerable British soldiers
'

fought their way up '

the fatal hill
'
of Albuera, he will not
be unmoved by Caesar's account of the battle with the
Nervii or of the last struggle of Vercingetorix. If his eyes
become dim when they light on Xapier's epitaph on Colonel
Ridge
—
And no man died that night with more glory j^et
'
;

many died, and there was much glory he wiU hardly keep
'

—
dowTi a tear when he reads how Sextius Baculus arose and
saved the camp at Aduatuca, *
facing fearful odds,' till he
was borne back fainting to his sick-bed. No, Caesar is not
 PREFACE TO THE FIRST EDITION xi

dull, except to minds enervated by sensational reading.

There is no tinsel in his narrative ; but it is not void of
colour. His style is severe ; but it is not frigid. Like
Thucydides and the historian of the Acts of the Apostles,
he has no sentimentality, but no lack of sentiment. His
passion never breaks from his control ; but it communicates
itself to us. Intent simply on tellmg his tale, he rises with-

out an effort, whenever the subject inspires, to genuine

eloquence. It is true that that swift narrative often baffles

curiosity, even when curiosity is legitimate ; but it is idle

to wish a Enough that this book

good book other than it is.

is worthy of its theme and of its author. We know on the

highest authority that even in our age the soldier who means
to study his profession cannot afford to neglect the Com-
rnentaries} And if a time should ever come when for purely
professional purposes they shall have lost their value, they
will still be worth reading for themselves.- They were
written, with a purpose no doubt but still in the main

honestly, by the greatest man of the world who has ever

lived and men of the world who are also lovers of literature
;

will best appreciate and most enjoy them. \Mioever cares

'
'The statement,' says General Maurice {War, 1891, p. 12), "of the most
brilHant and successful general of the British army of to-day appears to be
indisputable that a perusal of the words of even Caesar himself will suggest
to any thoughtful soldier who knows something also of modern war. reflections
that he may afterwards recall with advantage as applicable to modem cam-
paigns.' (See Lord Wolseley's The Soldier s Pocket-Book, oth ed., p. 286.)
The great Xapoleon. himself a diligent student of the Coimnentaries, recom-
mended all aspiring officers to read them {Memoires. vote-^ et melanges, ii,

lo.5).

^ 'La partie divine de Tart." writes Colonel Stoffel {Hist, de Jules Cesar,
Gnerre civile, i, v), '
est restee la meme et elle ne changera jamais . . . I'etude
des campagnes de Cesar est fertile en renseignements. On y trouvera 1' applica-
tion presque constante des vrais principes : tenir ses forces reunies, n'etre
vulnerable nulle part, marcher avec rapidite sur les points importants. s'en
rapporter aux moyens moraux, a la reputation de ses armes, a la crainte qu'on
inspire et aussi aux moyens politiques pour maintenir dans la fidelite ses allies,
dans I'obeissance les peuples conquis se donner toutes les chances possibles
:

pour s' assurer la victoire sur le champ de bataille pour cela faire, y reunir ;

toutes ses troupes. On y remarquera la promptitude dans T execution, I'habilete

k profiter de la victoire. Enfin on reconnaitra chez Cesar un chef ... en . . .

qui ni la bonne ni la mauvaise fortune . . . ne troublent Tequilibre.'

i
xii PREFACE TO THE FIRST EDITION
for a great book in a small compass, and will give it the
attention that it demands ; whoever can appreciate literary
qualities that have fallen out of fashion but will have their
turn again —masculine strength, simplicity, directness, reserve,
relevancy ; and, above all, the natural dignity that belongs to
*
the foremost man of all this world writing the history that
'

he had himself made —whoever cares for these things should

read Caesar's Commentaries, and he will have his reward.
Let me try to explain the scope of my own book. It does
not narrate the events of the conquest in precisely the same
detail, from first to last, m which Caesar narrated them ;

for such a narrative, even if it were skilfully composed, would

inevitably weary a modern reader ; and where it wearied, it

would also fail to instruct. Caesar doubtless knew, though

it was not his way to say so, that his book would be a KT?]\xa

h aei but he wrote, first of all, for his own generation

:
;

and, regarded as material for history, some of his matter,

if only a little, has lost its interest. Nothing, for instance,
would be gained by narrating in full detail the campaign of
Crassus in Aquitania. The general reader would be bored
by what he could not but regard as an anticlimax to the
more dramatic struggle of Caesar with the Veneti and the ;

student of Roman warfare would learn nothing that he

might not learn as well or better from a study of the opera-
tions which Caesar conducted in person. On the other hand,
of such events as the siege of Avaricum, the" blockade of
Alesia, the campaign of that great marshal, Labienus,
against the Parish, indeed of almost every operation of the
war, I have tried to give a full and clear account, which
might at once satisfy and interest general
professional
readers. Moreover, knowledge derived from personal ex-
ploration of the country, from the results of excavation,
from Cicero's letters and other ancient authorities, from the
researches of anthropologists, and from various monuments,
has made it possible, as the reader of the Second Part will
discover, to fill up certain gaps in Caesar's narrative. The
 PREFACE TO THE FIRST EDITION xiii

two expeditions to Britain I have, of course, not described

at all, but only made such a passing allusion to them as was
necessary to a right imderstandmg of my subject, the con- —
quest of Gaul. I do not profess to have followed the whole
of Caesar's track, because the thing is impossible : only
sections of the track can be traced with certainty, and we
often have to be content with the knowledge of the general
dhection of his march. But
have travelled long distances
I

in order to explore the known sites at which importan

events occurred. I hold that discussions on questions oi

evidence ought to be rigidly excluded from narrative and ;

my narrative therefore takes for granted the conclusions at

which I have arrived in the Second Part of the book. Let
me take the opportmiity of expressing my gratitude to
Colonel Stoffel, the principal collaborator of the late Emperor
Napoleon, who has sent me a most mteresting account, which
willbe found on pp. xxv-xxvii, of the method by which he
discovered Caesar's camps and entrenchments near Mont
Auxois (Alesia) and at other places ; and also to Major-
General J. F. Maurice and Major-General Sir Coleridge Grove,
who allowed me to consult them on certain military ques-
tions, which are discussed in Part II, and whose opmions,
I was glad to find, generally confii'med my own conclusions.
One word regarding the Second Part of this volume.
I dare say the impatient reader, who measm'es its length
against that of the narrative, will be inclmed to reverse
Prince Hal's dictum, and cry, '
Oh, monstrous ! but one
half -penny worth of sack to this intolerable deal of bread.'
But the remedy is in his own hands. It is not for me to
warrant the quality of my sack ; but whoever has no
appetite for the bread can leave it untouched. It happened
once at a dinner party that the lady whom I had taken in
asked me whether I had read an account of a certain battle
by a famous historian. I replied that I had not, but that,
if the critics were to be believed, it was most likely full of
mistakes. '
What does that matter,' rejoined my neighbour,
xiv PREFACE TO THE FIRST EDITION
'
makes a good battle of it ?
so long as he It was a delicious '

little speech and I verily believe that, if it had been

;

addressed to the late Mr. Fi'eeman, he would not have had

the heart to scold the lady. For my part, I have always
been grateful to her for her frank avowal. She made it so
clear to me that the majority of readers who take up a history
care nothing whether it is accurate or not, provided it is

interesting. Still, while I should like to think I had suc-

ceeded in '
making a good battle of it ', I do like to make
sure that this or that statement is true before committing
it to paper ; and so, for my own satisfaction and for the
satisfaction of scholars and the few general readers who are
not satisfied with results, but want to know the evidence on
which they are based, I have written my Appendix. Those
who are at all familiar with the difficulties of the subject will
not think that it has run to an undue length. For a writer
who deals with ancient history is at one great disadvantage as
compared with a writer whose period falls within more recent
times. He is obliged to spend years of labour in finding out
the truth on matters of geography, military science, and the
like, which his fellow- labourer finds ready to his hand.
My object in writing the Second Part has been to deter-
mine what can and what cannot be proved in regard to
those points which are still in dispute, and to furnish readers
with the materials for formmg their owti opinion. My
method has been not only to state my own reasons for the
opinions which I have formed, but also to present, in the
briefest possible compass, the reasons for the views from
which I dissent. It is true that a point can hardly be called
disputed when a decision, all but unanimously accepted, is

cavilled at b}^ a few crotcheteers. Astronomers do not waste

their time in defendmg the conclusions of Copernicus and
Kepler against the assaults of '
Parallax '
; and I once
thought that it would not be worth while to answer the
objections of the antiquaries who, even after the appearance
of the famous article by the Due d'Aumale in the Eevae des
 PREFACE TO THE FIRST EDITION xv

Deux Mondes, of the Dictionnaire archeologique de la Gaule,

of M. A. de Barthelemy's admirable article in the Revue des

Questions Historiques, and of Ernest Desjardins's candid re-
cantation, persisted in identifying Alesia with Alaise. But, for
reasons which I have given in the Appendix, I decided that
itwould be expedient to treat M. Quicherat and his school,
and even M. Maissiat, with more respect than Parallax ' '.

So far as I am aware, this is the only English narrative

which deals specially with Caesar's conquest of Gaul.
Narratives more or less detailed are to be found in

Mr. Froude's Caesar, in Mr. Warde Fowler's Julius Caesar,

in Colonel Dodge's Caesar, in Dean Merivale's History of the

Romans under theEmpire, in Long's Decline of the Roinan

Republic, and in the English translations of Mommsen's
Romische Geschichte, the late Emperor Napoleon's Histoire
de Jides Cesar, and Duruy's Histoire des Romains. None of
these writers, however, makes any systematic and com-
prehensive attempt to discuss doubtful points ; and even
the von Golers, father and son, in Cdsars Gallischer Krieg,
which has not been translated, have not regarded this task
as falling within their scope. Indeed there has not hitherto
appeared in any language a book which attempts to collect,

to co-ordinate,and to estimate the results of the innumer-

able researches which have aimed at throwing light upon
the problems of Gallic history, and most of which are prac-
tically inaccessible. Mommsen, strictly subordinating his
narrative to his great historical scheme, goes into details
hardly at all. Mr. Froude writes, not as a military historian
but as the biographer of Caesar ; and his brilliant sketch,
which has been as enthusiastically, if not as widely, admired
as his larger works, necessarily omits much that would
interest not only military but even general readers. On
geographical questions he almost invariably follows Napoleon;
and book would certainly have been not less trustworthy
his
than it is if he had never looked at any other commentary.
The scheme of his work and the rules of art compel him to
xvi PREFACE TO THE FIRST EDITION
dismiss battles, such as that with the Helve tii or Ariovistus,
in a single sentence ; even when he is describing such im-
portant operations as the siege of Avaricum or the attack
on Grergovia, he leaves very much to the imagination of his
readers ; and throughout his narrative he draws freely upon
his owii.^ Indeed, as he apparently wrote the entire work
in less than a year,^ it is safe to say that he did not waste
much time in investigation. Colonel Dodge's account, which,
like Mr. Fowler's brief sketch, did not appear until the
rough draft of my own narrative had been completed, is

sufficiently full : but he too, like Mr. Froude. is a faithful

follower of Napoleon and Napoleon, as I shall show,
;

makes many serious mistakes. The colonel claims credit

for having studied the works of the best recognized modern
'

critics ', and for having visited the theatre of Caesar's'

campaign and his many battle-fields '. But if a man wants

to find out what can and what cannot be known about the
Gallic war, he must not shrink from the labour of checking
the opinions of the best recognized modern critics by the
' '

works of unrecognized scholars who have wrought diligently

in the same field and, if I may be pardoned the Hiberni-
;

cism, it is of no use to visit battle-fields, unless it is certain

that battles were fought upon them. Merivale wrote before
the modern era of continental research had begun : he
worked upon a which forbade him to describe military
scale
operations in detail and I am obliged to say that whoever
;

compares his pages with the Commentaries will find that

some of hismost impressive passages are purely fictitious.
Long's narrative, which forms the bulk of his fourth volume,
is very full, —
too full perhaps in parts but Long had ;

a hearty contempt for the general reader. Moreover, his

^ See my article iu the Westminster Review of August, 1892, pi3. 174-81).

- In a letter, dated May 3, 1878, to Mr. Jolm Skelton, Froude t^ays, I am '

reading up Caesar and his times, with a view to writing a book about him,'
Iu a letter dated February 6, 1879, he says, " Caesar " is in the press.' The
'

book was publislicd some time before July of the same year. {Blackicvud's
Muijaziiie, December, 1894, pp. 772, 774.)
 PREFACE TO THE FIRST EDITION xvii

knowledge of Gallic geography, although thoroughly sound,

was very far from complete. Every student of Caesar is,
indeed, under the deepest obligations to him for no man ;

ever brought a stronger judgement to the study of the

problems which Caesar left us to solve. He knew his ancient

texts by heart he was perfectly familiar with the works of

:

such modern authorities as d'Anville, Walckenaer, RUstow,

and von Goler but of the enormous mass of articles which
:

are scattered among the transactions of the numerous

French archaeological societies and other periodical publica-
tions, as well as of the numberless monographs and pam-
phlets which have been published independently, and of the
mediaeval chronicles which bear upon the subject, he knew
very little. No doubt ninety-nine hundredths of the printed
matter contained in these works are valueless : but amid the
dross of verbiageand declamation with which too many of
them abound there lie embedded grains of solid information.
Moreover, since Long wrote, light has been throwTi upon
various matters which, in his time, were obscure.
It is to be wished rather than hoped that the appalling
mass of printed matter which, for four centuries, has been
accumulating round the Commentaries, may not be swelled
in the future by mere verbiage. If only the editors of
German periodicals would restrain the ardour of the emen-
dators who inundate them with futile conjectures, they
Avould be setting a good example. The Tabula Coniectura-
rum which Meusel prints at the end of his great Lexicon
Caesar ianum fills thirty-six pages super royal octavo, closely
printed in double columns and of all these conjectures
;

those which really deserve the name of emendations would

not fill a single page while those which have been unani-
;

mously adopted might be counted upon the fingers of one

hand. In the Greek state of Locri there was a rule that
whoever proposed a new law should do so with a rope
round his neck, and, if his proposal were rejected, should be
strangled on the spot. It would be a good thing if editors
1093 b
xviii PREFACE TO THE FIRST EDITION
would combine to deal with emendators in a like spirit.
Death would perhaps be an excessive penalty even for a bad
conjecture but whoever proposed an emendation which
;

failed within a certain period to win general acceptance

might be forbidden ever to contribute to a learned periodical
again.^ We have not yet got, nor will conjectural emenda-
tion give us, a final critical edition of the Commentaries ;

but for the purposes of history, in the most comprehensive

sense of the word, the text is good enough. Very few of the

passages in which it is uncertain offer a stumbling-block to

the historian and those mainly in points of minute detail.
;

Many of the geographical and other problems are now solved ;

and I hope that I have succeeded in contributing something

to the result. Others, as I have tried to show, are at present
insoluble, and must remain so unless and until fresh dis-
coveries throw light upon them. But excavation, carried
out regardless of cost and intelligently directed, has already
been so active in France that I doubt whether, for the period
of Caesar's campaigns, it has many surprises in store for us.
It is perhaps conceivable that the future may reveal some
lostmemoirs which may supplement Caesar's own narrative.
But even if our positive knowledge is not destined to be
increased, we know enough already for essential purposes ;

and the most that further research or happy chance can

bring to light is very little in comparison with what has been
already discovered. And when the catalogue of programs
'
'

and dissertations is complete, when modern research and

modern literary skill shall have combined to produce the
final history of the Gallic war, the unpretending little book

which Caesar wrote two thousand years ago in the scanty

leisure of a busy life will outlive them all.

'
If these remarks had not been misunderstood, I should have thought it

unnecessary to say that they were directed not against the use but against the
abuse of conjecture. [26.8.1903.]

11 DouRO Place, Kensington, W.

July 23, 1899.
 THE BUSTS OF JULIUS CAESAR
Whoever wishes to know all that can be known about the busts
of Caesar should read, first of all, Bernoulli's learned and beautifully
illustrated Romische Ikonographie. That work will tell him what
busts are generally regarded as authentic but what we really want
:

to know is which of the authentic busts offers the most faithful like-
ness and this is what neither Bernoulli nor any one else can certainly
;

tell. It comes to this, that every one must study for himself Caesar's
history, form his own idea of his character, and thei> use his own
judgement ; and if a man distrusts his own judgement and finds
a learned treatise tiresome, perhaps he might do worse than take
Mr. Baring Gould for his guide. It is true that the author of The
Tragedy of the Caesars sometimes lets his imagination run away with
him. He has, I think, idealized the character of Caesar, and read
his ideal in, or rather into, his favourite busts. But it is impossible
for him to take pen in hand without being interesting and, accurate ;

or not, a man of his calibre cannot fail to throw light upon any
subject with which he deals.
A portrait which has done duty in many works on Caesar is taken
from the colossal bust of Naples. This seems to me, not indeed, as
Mr. Baring Gould ^ thinks, characterless, but, at any rate, no true
presentment of the character of Caesar. The face is powerful, but
heavy if not brutal.^
Mr. Warde Fowler^ has suggested that the real Caesar may be
represented by the green basalt bust of Berlin. The breadth of skull
which characterizes the marble bust in the British Museum, and, in
varying degrees, all the others, is absent from this but Mr. Baring ;

Gould supposes that the block of basalt which the sculptor used
•*

may have been too narrow. Surely this is pushing conjecture too
far. M. Salomon Reinach,^ on the other hand, points out that the
type of the basalt bust is not to be found on any of the coins of
Caesar,^ and that it is similar to the type represented in the bust of
an Alexandrian Greek in the Imperial Museum of Vienna. Mr. J. C.
Ropes,"^ indeed, speaks of a mark by which one can generally recog-
'

nize the authentic busts of Caesar, namely a scar or furrow on the

left side of the face and he adds that this mark is to be found on
'
;

^ The Tragedy of the Caesars, i, 1892, pp. 3, 116.

^ The illustrations of this bust in Mongez's Iconographie romaine (ii, 1824,
pi.17 [1]) are idealized. Compare them with Taf. xiii in Bernoulli's book.
' Classical Review, vii, 1893,
p. 108.
* The Tragedy
of the Caesars, i, 106.
^ Gazette des Beaux- Arts, 3® per.,
vii, 1892, pp. 474-6.
® See the beautiful illustrations of the
coins in H. Cohen's Description generate
des monnaies de la republique romaine, 1857.
' Scribner's
Mag., i, 1887, pp. 132, 135.
b2
 XX THE BUSTS OF JULIUS CAESAR
the bust in the British Museum, and also on the basalt bust. There
is certainly a furrow on the left side of the bust in the British Museum
but there is a corresponding, though shorter, furrow on the right side ;

and I used to think that both of them simply represented lines such
as are to be seen on the faces of many men who have passed middle
life. I have, however, since noticed that some of the coins ^ show
a furrow on the right cheek with great distinctness. But, whatever
may be the worth of the furrows as evidence, Bernoulli, as well as
M. Beinach, questions the authenticity of the basalt bust and only;

an enthusiast could detect any similarity between it and any of the

other busts the authenticity of which is admitted.
M. Geffroy, the Director of the Ecole francaise de Kome,^ remarks
that Signor Barracco possesses a bust of Caesar, the genuineness of
which is proved by its bearing on the crown of the head the star
mentioned by Suetonius. Undoubtedly this bust was intended to
represent Caesar but what proof is there that the artist ever saw
;

Caesar, or even worked with an authentic portrait before him ? If

any one thinks this question vexatious, I beg him to suspend his
judgement until he has finished reading this note. Suetonius ^ says
that, on the occasion of the first games which Augustus held in
honour of Julius, a comet appeared that the comet was regarded
;

as a sign that Caesar's soul had been received into heaven and
;

that, in consequence, the image of a star was placed upon the head
of his bust. Now M. GefEroy cannot prove that the bust in Signor
Barracco's possession is the very bust of which Suetonius speaks, or
even a replica of it for it is probable that posthumous busts were
;

produced with a star upon the head and if Signor Barracco's bust
;

was posthumous, as he himself believes that it was,'* it must either

have been a copy of an original or simply a work of memory or of
imagination. It was found in the delta of the Nile; and two photo-
graphs of it are reproduced in a volume entitled La collection Barracco,
by G. Barracco and W. Helbig. The face is covered with a beard of
about a fortnight's growth.^ The shape of the head is strikingly
different from that of the bust in the British Museum, and its relative
breadth is much less though in both the forehead, as distinguished
;

from the head itself, is remarkably narrow. In expression the two

busts have hardly any resemblance.
Mr. Baring Gould has a very high opinion of the bust in the
British Museum so has Bernoulli ^ and, given the authenticity
: ;

^ See Bernoulli, Nos. 53 and 62.

Rev. arch., 3* ser., xx, 1892, p. 256.
- I do not know whether M. Geffroy is
stillDirector,
^ Divus luliiis, 88. Cf. Archaeol. Zeitung, xix, 1867, pp. 110-3.
* Nous pouvons conclure que la statue dont provient notre tete fut execut^e
'

apres la consecration de Cesar.' G. Barracco and W. Helbig, La collection

Barracco, 1893-4, p. 51.
^ The authors of La collection Barracco conjecture that Caesar had let his

beard grow as a sign of mourning for Pompey, just as, according to Suetonius
{Divus lulius, 67), he did while he was avenging the massacre at Atuatuca.
^ Among those busts,' says Bernoulli (p. 171), which recommend themselves
' '

by their resemblance to the coins, this is the one which most suggests Caesar.*
 THE BUSTS OF JULIUS CAESAR xxi

of the bust,^ I do not think that any one could doubt that it was the
work of a sculptor who, as Mr. Baring Gould says, knew Caesar '

and loved him,' or at least understood and admired him.^ But

Mr. Baring Gould tells us that Mr. Conrad Dressier, the sculptor,
who shares his admiration for the bust, has pronounced that Caesar
could not have sat to the artist, because the extraordinary breadth
of the skull above the ears is anatomically impossible.^ When I read
this it struck me as unlikely that a sculptor who is assumed to have
known Caesar well would have cared to model his bust from memory,
or that his memory would have been so defective and it seenied ;

incredible that a sculptor who was capable of producing such a work

of art should have lacked an elementary knowledge of anatomy.
I asked Mr. Hope Pinker, whose bust of Sir Henry Acland is a
speaking likeness, for his opinion. It confirmed my own. Have
Mr. Baring Gould and Mr. Dressier forgotten the bust of the youthful
Augustus which stands in the British Museum, within a few feet of
the bust of Caesar ? Let them look at it again, and they \^dll admit
that its breadth above the ears is just as remarkable as that of its
neighbour.'*
Mr. Baring Gould considers a bust in the Louvre, of which he gives
an illustration, as good in its way as the bust in our national collec-
tion but it seems to him to represent the militant rather than
;

the reflective side of Caesar's character.^ To my mind the bust in the

British Museum represents, as a bust should do, not one side of
the man's character, but the whole. The bust in the Louvre has
features of the Caesarian type but the expression is quite different.
;

Mr. Dressier has remarked that, in default of direct evidence, there is

no better test of the fidelity of a portrait than the impression which
it leaves upon the mind an intelligent observer.^ The test is
of
obviously imperfect ; but worth pages of discussion. Nor would
it is
I hesitate to apply that test, according to the measure of my in-
telligence, if only it were certain that the bust in the British Museum

The late Professor Adolf Furtwangler {Neuere Fahchungen von Antiken,

^

1899, p. 14) condemned the bust as a forgery,

—
a modern work with skilfully
'

imitated corrosion [of the marble], and an expression the very reverse of ancient.'
But Furtwangler was fond of laying down the law and Mr. A. H. Smith, the
;

Keeper of Greek and Roman Antiquities in the British Museum, who unhesi-
tatingly accepts the bust as a work of classical art, thinks that it has been de-
corroded. The marble,' he says, 'appears to have sufferedfrom a drastic cleaning
'

with acid, but portions at the back show the original surface {A Catalogue of '

Sculpture, &c., iii, 1904, p. 146). M. Salomon Reinach, in a review of the first
edition of this book, asserted that the bust is modern but Professor Percy
'
' ;

Gardner {Eng. Hist. Rev., xix, 1904, p. 326) still regards it as ancient. If the
sculptor was a forger, he was also a genius but no forger would have thought
;

of portraying that narrow forehead in combination with a broad head.

^ It is the man himself,' says Professor Ernest Gardner {Handbook of Greek
'

Sculpture, 1905, pp. 513-5), that the sculptor brings before us.'
'

^ The Tragedy
of the Caesars, i, 114-5.
* Only the other day I saw a whose head, extraordinarily broad,
child,
projected above the ears as much by the bust in the Museum.
as that depicted
[16.11.97.] The bust is not more brachycephalic than the heads of many
living Auvergnats and inhabitants of the department of the Jura.
* The Tragedy of the Caesars, i, 115. «
lb., pp. 9-10.
xxii THE BUSTS OF JULIUS CAESAR
is really an authentic bust of Caesar. But even this certainty is
wanting. There is not in existence a single bust of which it can be
said, with absolute certainty, both that the sculptor intended it to
be a portrait of Caesar, and also that either Caesar sat for the like-
ness or the sculptor had personal knowledge or an authentic likeness
to guide him. Some years ago I asked the late Dr. Alexander Murray,
who was then Keeper of Greek and Roman Antiquities in the British
Museum, whether there was any doubt of the authenticity of the
marble bust. Oh no,' he answered
' ! no doubt whatever.' But
;
'

he could not give me any proof. The bust was once believed to
represent Cicero. If physiognomy is any index to character, it is
certain that that calm face bore no resemblance to his but the ;

conjecture, absurd as it was, would never have been made if there

had been direct evidence that the bust was intended for Caesar.
IVidence, however, there is none for the authenticity of this or of
any one of the so-called busts of Caesar, except such evidence as is
to be got from the study of the texts and of the coins. The evidence
of the texts is very scanty and most of the coins differ widely
;

among themselves.^ The contemporary coins which bore Caesar's

—
efhgy were the work of five different agents, L. Aemilius Buca,
L. Flaminius Chilo, M. Mettius, P. Sepullius Macer, and C. Cossutius
Maridianus. None of them were struck before 44: B.C., the year of
Caesar's death. Others, known as the Voconian group, were executed
a few years later.^ In the Description of the collection of Ancient
Marbles in the British Museum ^ it is affirmed that there is an agree-
ment among the Aemilian and Voconian coins w^hich is perfectly'

satisfactory ', and that with all of them the bust in the Museum
'
exhibits a striking similarity '. Well, the reader should look through
Cohen's Description generale des monnaies de la republique romaine,
and judge for himself. The Aemilian coins are numbered II, 15,
16, and 17 on Plate ii the Voconian 1 and 2 on Plate xlii. No. 2
;

certainly resembles 15, but differs widely from 1 17 is about as

;

much like the others as Gladstone was like Beaconsfield and, in ;

expression, none of the six resembles any of the busts.^ All that can
be said is that, in profile, there is a general resemblance between
No. 15, No. 2, Nos. 2 and 3 on Plate xvi, 3 on Plate xviii, and 4 on
Plate xxxvii that the type of face depicted on these six coins is
;

not unlike that of the bust in the British Museum and that the
;

lean muscular neck shown in the former resembles that of the latter.
When one looks at different portraits of any well-known modern
face, one can always tell at a glance whom they were intended to
represent. Similarly, the portraits of Queen Elizabeth, for instance,
are all unmistakable. And, to go back to ancient times, it does not

1 The face on a coin in the British Museum, an illustration of which is given

inMr. Warde Fowler's Caesar, is that of an imbecile buffoon.
2
E. Babelon, Descr. hist, et chron. des monn. de la republique rom., 1886, i,
^ Part xi, 1861, pp. 39-40.
497, ii, 560.
*
Notice the varieties in form and expression in the Sepullian group,
xxxvii, 4-9.
 ;

THE BUSTS OF JULIUS CAESAR xxiii

need an expert to that the busts of Augustus were all intended

tell
to jDortray the same But the busts of Caesar differ from each
face.
other so much in expression, and some of them even in feature,
that, although there is a certain vague Caesarian type common
' '

to all, an untrained eye, if the inscriptions were removed, would

probably take them for portraits of different men. The conclusion
appears to be either that most of the sculptors were unable to catch
a likeness, or that most of them worked from memory or imagina-
tion, or, finally, that some of the busts were not meant to represent
Caesar at all. But this much is certain —
if the original of the bust
:

in the British Museum was not Caesar, he was a very great man ;

and who ? The experts cannot help us to arrive at a definite con-

clusion and for my part I am content to accept as the likeness of
;

Caesar the noble bust which has approved itself to Bernoulli, to

Mr. Baring Gould, and to other well-qualified judges.^
This bust represents, I venture to say, the strongest personality
that has ever lived, the strongest which poet or historian, painter or
sculptor has ever portrayed. In the profile it is impossible to detect
a flaw : if there is one in the full face, it is the narrowness of the

forehead as compared with the breadth of the skull. The face appears
that of a man in late middle age. He has lived every day of his life,
and he is beginning to weary of the strain but every faculty retains
;

its fullest vigour. The harmony of the nature is as impressive as its

strength. No one characteristic dominates the rest. Xot less re-
markable than the power of the countenance are its delicacy and
fastidious refinement. The man looks perfectly unscrupulous or, ;

if the phrase be apt to mislead, he looks as if no scruple could make

him falter in pursuit of his aim but his conduct is governed by

:

principle. Passion, without which, as Mommsen truly said, there

can be no genius, inspires his resolve and stimulates its execution ;

^ In the Description of the collection of Ancient Marbles in the British Museum

(Part xi, pp. 39-40) it is asserted that'
the general character of the features of
Caesar are as well known and as clearly marked as those of any personage
of Roman times,' and that '
the features of the marble bust agree with them.'
If this statement requires some qualification, it may, I think, be affirmed that
the marble bust agrees as well as any other with the coins. It is interesting to
compare it with the illustrations in Mongez's Iconographie romaine, ii, pi. 17
(1, 3), pi. 18 (1-2) of the Neapolitan, Capitoline, and St. Cloud busts. These
three, though they differ in expression, represent, I feel sure, the same man.
The lines of the forehead in them and in the British Museum bust are alike
and there is a certain resemblance in the profile and the shape of the head,
though the jaw in the St. Cloud bust is squarer, and the chin more prominent
than in the other three. The ear of the former is very like that of the British
Museum bust, and, like it, lies very close to the head.
Professor Percy Gardner [Eng. Hist. Rev., xix, 1904, p. 326) says that 'the
forehead [of the bust in the British Museum], which slopes backward, is certainly
not the forehead of Caesar. This,' he adds,'
is guaranteed by the coins . . .

it requires long familiarity with coins fully to appreciate their testimony.'

No doubt but even a layman may be allowed to point out that the angle
;

formed by the forehead differs widely in different coins ; and that, among
the coins illustrated by Cohen [Descr. gen., &c.) the slope in pi. ii, 15, pi. xviii,
3, pi. xxxvii, 4-8, and pi. xlii, 1, is as great or greater than in the marble bust,
the forehead of which could not fairly be called retreating.
xxiv THE BUSTS OF JULIUS CAESAR
but passion, in the narroAV sense, is never suffered to warp his action.
He is kindly and tolerant but, to avoid greater ills, he would shed
;

blood without remorse. The mild but inexorable yoke of Caesar,'

'

— so Mr. Strachan-Davidson ^ describes the ascendancy to which

Cicero reluctantly submitted and mild inexorability is apparent in
;

the expression of this man. He can be a charming companion to

men and, though he is no longer young, he knows how to win the
;

love of women. He sees facts as they are, accepts, and makes

the best of them. Knowledge of men has made him cjmical but the ;

cynicism is dashed by humour. Look at the profile from the left,

and you will note an expression of restrained amusement, as of one
who is good-naturedly observant of the weaknesses of his fellows.
If his outlook passes beyond mundane things and strains after the
unknown, he does not let us into the secret of his thoughts. But if
the ordinary observer is unable to discern that look of faith, that
'
far-off look which Mr. Baring Gould ^ loves to fancy that he can
'

read in the expression, he cannot fail to recognize the stamp not

only of will and of intellect, but also of nobility. The bust repre-
sents a man of the world, in the fullest meaning of the term. It
alone represents a man such as Caesar has revealed himself in his
writings, and as his contemporaries have revealed him in theirs ;

and that is why I have chosen it to illustrate this book.

[Mr. Frank J. Scott, of Toledo, U.S.A., has recently published
a book, called The Portraitures of Julius Caesar^ which contains
illustrations of all or nearly all the busts, coins, and gems that have
been regarded as meant to portray Caesar's features. That many
of them were so meant is certain but which of them was the best
;

likeness, and w^hether any one of them was executed from life, are
problems that remain unsolved.]
1 Cicero, 1894, p. 268. The Tragedy of the Caesars, i. 114-5.
''

^ For criticisms of Mr. Scott's book see Eng. Hist. Rev., xix, 1904, pp. 325-7,
and Class. Rev., xviii, 1904, pp. 183-5.
 COLOXEL STOFFELS EXCAVATIONS
The contributions which Napoleon III made to our knowledge of
the history of the Gallic war were due to excavations carried out by
his principal collaborator, the late Colonel Stoffel. I reprint here
part of a letter which the colonel wrote to me before the publication
of my first edition. Substantially, he confirmed my own preconceived
notions, and his method was identical with that which is followed
by Professor Haverfield and other well-known investigators. Vous
'

desirez,' he writes, savoir par quelle methode j'ai retrouve les

'

traces des camps que I'armee de Cesar construisit dans la guerre

des Gaules. II est necessaire de commencer a indiquer quelques

notions preHminaires. Les terrains dans lesquels ces camps furent

etablis presentent, comme tons les terrains cultives, une couche
superieure de terre vegetale, appelee humus, la quelle varie d'epaisseur
selon les differentes contrees, et pent avoir depuis un ou deux pieds
jusqu'a quatre ou cinq pieds et plus. Au-dessous de cette couche
de terre vegetale se trouve le terrain vierge (ou le sous-sol), qui est,
selon les contrees, ou marneux, ou siliceux, ou calcaire. A Alesia
(dans la plaine des Laumes) c'est de la marne epaisse et ferme ;

a Berry-au-Bac c'est une marne plus legere a la Roche-Blanche

;

(en face de Gergovia) c'est un calcaire ferme et blanc. Lorsque, apres

une bataille, ou apres un siege, I'armee romaine quittait son camp,
les habitants du pays en detruisaient les retranchements afin de
pouvoir de nouveau cultiver leurs champs. lis rejetaient les terres
du parapet dans le fosse. Ce fosse etait, de la sorte, plein d'wrie
composee de terre vegetale, de terre vierge, et souvent
terre me'langee,
d'objets que les soldats romains avaient pu laisser sur le parapet,
tels que debris d'armes, boulets en pierre, monnaies, ossements, etc.
Pendant quelque temps la partie superieure du fosse comble pre-
sentait la forme AB [slightly convex], a cause du foisonnement des
terres ; mais avec le temps, et grace a la culture de chaque annee,
elles se tassaient au niveau du sol avoisinant, ce qui fait que partout
xxvi COLONEL STOFFEL'S EXCAVATIONS
les traces des camps de Cesar out disparii. En tout cas, la terre de
remplissage des fosses est uiie terre meuhle et, fait important a re-
marquer, elle reste meuble, sans jamais reprendre la consistance du
terrain vierge, si bien qu'aujourd'hui, apres 2000 ans ecoules, elle
se detache aisement a la pioche. C'est la ce qui permet de retrouver
les fosses lorsqu'on a su determiner Templacement d'un camp.
C'est la, comme vous le dites tres bien, la premiere condition. II
faut done, avant tout, etudier le terrain oii on suppose que le camp
etait place, ce qui exige une connaissance parfaite des Commentaires
de Cesar et des connaissances militaires speciales.
Cela pose, voici comment j'ai toujours procede pour retrouver les
fosses d'un camp. Soit ABCD mie etendue de terrain dans laquelle
je supposais place le camp qu'il s'agissait de decouvrir et admet- ;

tons, pour fixer les idees, que la couche de terre vegetale ait 70 centi-
metres d'epaisseur. Je pla9ais les ouvriers, avec pelles et pioches,

sur plusieurs ^le^ffj] dans une direction perpendiculaire a un des

cotes supposes du camp, les ouvriers de chaque file a 20 ou 30 metres
les mis des autres. Chacun d'eux etait charge d'enlever la couche
de humus sur deux pieds de largeur. Si, apres avoir enleve cette
couche sur 70 centimetres de profondeur, ils sentaient que leurs
pioches frappaient un terrain resistant, c'est que celui-ci n'avait
jamais ete remue et qu'on n' etait pas sur le fosse romain. Les
ouvriers continuaient alors a avancer, et cela tant qu'il ne se pro-
duisait rien de nouveau. Mais lorsqu'ils arrivaient, sans s'en douter,
sur le fosse en xy, c' etait autre chose. Alors, apres avoir enleve la
terre vegetale jusqu'a la profondeur de 70 centimetres, ils ne trou-
vaient plus, comme precedemment, un sol vierge resistant an con- ;

traire, ils rencontraient une terre meuble Cjui se detachait facilement,

ce qui j^ermettait de supposer qu'elle avait ete autrefois remuee.
Je faisais alors elargir la tranchee en lui donnant six pieds de largeur
{cd) au lieu de deux pieds {xy), afin que les ouvriers pussent travailler
plus commodement ; et ils approfondissaient la tranchee jusqu'a
 COLONEL STOFFEL'S EXCAVATIONS XXVll

c-equ lis rencontrassent le sol naturel. D'ailleiirs on reconnaissait

bientot SI on etait, oui ou non, sur le fosse romain car, si on y etait
;

reellement, on distinguait sans peine

sur les deux bords ec et fd de la
tranchee, a droite et a gauche des
ouvriers, le profil du fosse qui se V)

detachait par la couleur de la terre

melee (celle de I'ancien parapet) sur
la couleur du terre vierge qui I'en- u O
cadrait. u
c
Je n'ai rien vu de plus curieux nt
u
-iJ
que les profils des petits fosses du K y
petit camp que j'ai mis a decouvert
d
sur la colline de la Roche-Blanche.
La, la couche de terre vegetale, u
epaisse tout au plus de 50 a 60 centi- s
metres (si j'ai bonne memoire),
repose sur un sol de calcaire dur et
blanc comme de la craie aussi les
:

fosses du camp, remplis d'une terre melangee de humus et de craie,

presentaient-ils des profils qui tranchaient sur la terre dont ils etaient
entoures aussi nettement
que le triangle ABC ci- B
contre tranche sur le papier
A
blanc. L'Empereur, qui '::fif;3r5^ ^p- :-.
:-:_-' ^i^~
etait
les
venu visiter mes fouil-
a Gergovia, fut tellement ^^^^W W^~
— -.
•

frappe et emerveille envoy- - ".Soln-vierp^e"^ ^^^ — Sblz^efge"_

ant ces profils, qu'il songea .
^^F^
a acheter la colline de la -c~ ^^--ZT"

Roche -Blanche pour les

conserver. II abandonna
cette idee lorsqu'il sut que les habitants desiraient ne pas etre
depossedes, et il m'ordonna de combler mes tranchees et de tout
mettre dans I'etat primitif.
Pour en revenir aux recherches necessaires pour determiner I'em-
placement d'un camp, il est a peine besoin d'ajouter que quand
j'etais parvenu a retrouver un des points par le profil du fosse,
je me bornais a en retrouver cincj ou six autres dans la longueur
de chacjue cote, ce qui suffisait pour delimiter le camp et en con-
naitre la forme exacte. ... A Alesia, les recherches ont dure plus de
deux ans, parce qu'il fallait retrouver non seulement les traces des
camps, mais encore celles des lignes de contrevallation et de cir-
convallation. J'y ai employe plus de 300 ouvriers.'
I do not at all mean to imply that Napoleon's maps are all trust-
worthy. On the contrary, it will be shown in Part II of this volume
that some of them are largely conjectural, and that several are
either wholly or in part erroneous.
 CONTENTS
Preface ....
Preface to the First Edition
PAGE
iii

V
The Busts of Julius Caesar xix
Colonel Stoffel's Excavations XXV
List of Illustrations . xl

PART I

CHAPTER I

IXTRODUCTION
Gallic invasion of Italy : battle of the Allia and its results 1

Gallic tribes assist the enemies of Rome 1

The Romans fight their way to the Po . 2
And conquer Cisalpine Gaul 2

Gaul and its inhabitants

Prehistory of Gaul
....
Formation of the Roman Province

.....
in Transalpine Gaul 3
4
5
The Ligurians and Iberians 11
The Celts
....
.......
Civilization of the Gauls
11
15
Coins
Bibracte
Stradonic
.......
......
17
19
20
Political and
.....
social organization

.......
Unifying influences
. 20
25
Religion
The Druids ...... 26
32
Invasions of the Cimbri and Teutoni
Invasion of Ariovistus
Revolt of the Allobroges
....
....
36
37
38

Consulship of Caesar .....

Threatened invasion of the Helvetii

How he attempted to provide against the Helvetian danger

38
40
40
He
His army
His intentions
......
is appointed Governor of Gaul

......
40
42
44
 CONTENTS XXIX

CHAPTER II

CAMPAIGNS AGAINST THE HELVETII AND ARIOVISTUS

PAGE
Caesar hears that the Helvetii are about to march through the Province 46
He hastens to Geneva and destroys the bridge 46
Helvetian envoys ask his leave to use the road through the Province 46

He
bank of the Rhone ......
He promises to reply in a fortnight, and meanwhile fortifies the left

prevents the Helvetii from crossing

47
47
The Sequani allow them to march through the Pas de I'Ecluse 48
Caesar goes back to Cisalpine Gaul, returns with reinforcements, and
encamps above the confluence of the Rhone and the Saone 48
The Aedui solicit his aid against the Helvetii 49

His passage of the Saone ......

He defeats and disperses the rearguard of the Helvetii

The Helvetii attempt to negotiate, but reject Caesar's terms

49
49
49
They march northward, followed by Caesar . 50
Caesar pressed for supplies, owing to the intrigues of Dumnorix 51
His abortive attempt to surprise the Helvetii 53
He
....
marclies for Bibracte (Mont Beuvray), to get supplies

....
Defeat of the Helvetii near Bibracte
53
55

Settlement of the Boi ......

Caesar's treatment of the fugitives 56
57

against Ariovistus ......

Envoys from Celtican Gaul congratulate Caesar, and

Failure of his attempts to negotiate with Ariovistus

solicit his aid
57
58
He marches against Ariovistus and seizes Vesontio (Besan9on 60
Roman army
Panic in the . 60
How Caesar restored confidence . 61
He resumes his march against Ariovistus 62
His conference with Ariovistus 63
Mission of Troucillus and Metius . 64
Ariovistus cuts Caesar's line of communication 64
How Caesar regained command of it 65

Caesar attacks
They are
them ....
The Germans from superstition delay to

defeated and expelled from Gaul

fight a pitched battle 66
66
67
Caesar quarters his legions at Vesontio 68
Significance of this step 68

CHAPTER III

THE FIRST CAMPAIGN AGAINST THE BELGAE

Results of the first campaign
The Belgae conspire against Caesar
......
..... 69
69
Caesar returns to Gaul and marches against them . 70
XXX CONTEXTS

The Remi submit, and help Caesar

He
.....
sends Diviciacus to ravage the lands of the Bellovaci
'AGE
70
71
Marches to encounter the advancing host, crosses the Aisne, and
encamps on rising ground near the bank
The Belgae attack Bibrax (Vieux-Laon)
Caesar sends his auxiliaries to the rescue
....
....
71
72
72
The Belgae encamp opposite Caesar
Caesar makes his position impregnable
.

....
. . . 72
72

They disperse .........

The Belgae attempt to cut his communications, but are defeated

......
Caesar's cavalry pursue them
73

......
He marches westward and receives the submission of the Suessiones

......
Bellovaci, and Ambiani 74

......
The Nervii resolve to resist
Caesar marches against them
75
75

the Sambre ........

He learns that they and their allies are encamped on the right bank of
7G

Battle of Xeuf-Mesnil .......

His pioneers mark out a camp on the heights of Neuf-Mesnil

....
76
77

He ....
Caesar treats the survivors with clemency

........
besieges the stronghold of the Atuatuci
80
81
They surrender
But afterwards make ....
a treacherous attack
81
82
Their punishment .

......
Galba's campaign in the Valais
. . . 82
82
Submission
Rejoicings at ........
of the tribes of Brittany
Rome
and Normandy 84
85

CHAPTER IV
CAMPAIGNS AGAINST THE MARITIME TRIBES AND THE
AQUITANI
Delusive prospects of peace .......
.... 86

Caesar prepares for a naval war .......

Rebellion of the Veneti, Coriosolites, and Esuvii

........
87
87
The conference at Luca
Caesar returns to Gaul ........
........
88
88
Preparations of the Veneti
The Roman weather-bound in the Loire
fleet ....
....
89
90
Caesars campaign against the Veneti
fruitless
Sea-fight between the Veneti and Brutus
Punishment of the Veneti . . .
.....
. . . . .91
90
90

Campaign of Sabinus against the northern allies

Brilliant campaign of Crassus in Aquitania
Fruitless campaign of Caesar against the Morini
.....
of the Veneti

....
. . 91
92
94
 CONTENTS XXXI

CHAPTER V
THE MASSACRE OF THE USIPETES AND TENCTERI
PAGE
The Usipetes and Tencteri invade Gaul 95
Caesar fears that some of the Gallic tribes may join them 96
He returns to Gaul and summons a Gallic council . 96
He marches against the Usipetes and Tencteri
And negotiates with their envoys .... 96
96
Their cavalry, in violation of a truce, attack his
He resolves to attack them at once .... 98
98
Arrests their chiefs,
And
who had come
virtually annihilates the host
His conduct condemned in the Senate
....
ostensibly to explain

....
98
99
99
He bridges the Rhine, punishes the Sugambri, and returns to Gaul 99

CHAPTER VI
THE DISASTER AT ATUATUCA AND* ITS RESULTS
Caesar's invasions of Britain
Intrigues of Dumnorix
.......
........ 101
102
His fate
The Gallic nobles in a dangerous mood
Distribution of the legions for the winter of 54-53 b.
..... c. . . .
103
103
105
Divide et impera . . . . . . . . .106
Assassination of King Tasgetius, Caesar's nominee, by the Carnutes . 106
Intrigues of Indutiomarus against Caesar . . . . .106
Sabinus and Cotta ........
The Eburones, under Ambiorix, make a futile attack on the camp of
107
Ambiorix advises Sabinus to withdraw to one of the nearer camps
The advice discussed in a council of war
In spite of the protests of Cotta, Sabinus decides to abandon the camp
.....
108
108
109
.

The Romans march out . . . .110

. . . .

They are surrounded by the Eburones . 110

. . . . .

And virtually annihilated . . . . . . . .112

Ambiorix persuades the Atuatuci and Nervii to join him in attacking
Q. Cicero
Siege of Cicero's
A messenger
camp ........
from Cicero carries a dispatch to Caesar . . ,115
112
112

Caesar marches to relieve Cicero . . . . . . .115

The Gauls abandon the
Defeat of the Gauls
Caesar joins Cicero
.........
.
siege,

.
and march to encounter him

. . . . .
.

.
.

.118
117
118

Immediate effects of his victory . . . . . . .118

Many of the nobles continue to intrigue . . . . .119
Schemes of Indutiomarus . . . . . . . .119
xxxu CONTENTS
PAGE
He isoutwitted by Labienus, defeated, and slain 120
Caesar raises two new legions, and borrows a third from Pompey 121
Continued troubles in North-Eastern Gaul 121
Caesar punishes the Nervii 121
Forces the Senones and Carnutes to submit 122
And prepares to punish Ambiorix 122
As a preliminary step, he crushes the Menapii 122
Labienus disperses the Treveri 123
Caesar again crosses the Rhine, and threatens the allies of Ambiorix 123
Returning unsuccessful to Gaul, he marches against Ambiorix
The Eburones keep up a guerrilla warfare
Caesar invites the neighbouring tribes to harry them
.... . 124
124
125
The Sugambri surprise Cicero 126
Caesar ravages the country of the Eburones 128
Ambiorix eludes pursuit 128

Execution of Acco ....

The legions distributed for the winter . 128
128

•
CHAPTER VII
THE REBELLION OF VERCINGETORIX
Agitation renewed .......
News of the murder of Clodius reaches Gaul .
129
129
Gallic chiefs encouraged to conspire against Caesar 130

Gergovia
.....
The Carnutes massacre Roman

.........
The news reaches the Arverni
citizens at Cenabum (Orleans) 130
131
131
Vercingetorix, notwithstanding the opposition of the Arvernian
government, rouses popular enthusiasm for rebellion 131
Most
elect him Commander-in-Chief ....
of the tribes between the Seine and the Garonne join him, and

...... 132
How he raised an army
The dissentient tribes ......
......
132
133
The mission of Lucterius
The Bituriges join Vercingetorix
Why did Labienus not take the field
..... ? .
133
134
134
Caesar returns with recruits to the Province . 134
How he rejoin his legions ?
shall . 134
He rescues the Province from a threatened invasion 135
Crosses the Cevennes, invades Auvergne, and forces Vercingetorix to
come to its relief . . . ... 135
Then seizes the opportunity
Vercingetorix besieges Gorgobina ....
to rejoin his legions 136
136
Caesar marches from Agedincum
Captures Vellaunodunum
Captures and punishes Cenabum
......
(»Sens) to relieve

....
Gorgobina 137
137
138
Crosses the Loire, and captures Noviodunum 138
 CONTENTS xxxiii
PAGE
And marches to besiege Avaricum . . . . . .138

..........
Vercingetorix persuades the Bituriges to burn their
granaries
towns and
138
The
Avaricum
Siege of .........
Bituriges, contrary to his advice, resolve to defend

........
Avaricum . 139
140
Storming of Avaricum
........
Indiscriminate massacre
......
Vercingetorix consoles his troops
145
146
146
He raises fresh levies . . . . . . . .147

.......
Caesar, at the request of the Aedui, decides between rival claimants
Vergobret
for the office of 148

marches himself to attack Gergovia .....

He sends Labienus to suppress rebellion in the basin of the Seine, and
148
He establishes a magazine at Noviodunum (Nevers)
Crosses the Allier by a stratagem
And encamps before Gergovia
......
.......
148
149
150
. . .

First operations at Gergovia .. . .151

. . . .

Defection of the Aeduan Vergobret . .151

. . . .

An Aeduan contingent, marching to join Caesar, persuaded by its

leader to declare for Vercingetorix . . . . .152
just in time to rescue his camp
Outrages of the Aedui against Roman citizens
......
Caesar makes a forced march, overawes the contingent, and returns

.... 153
154
Anxiety of Caesar . .

He attempts to take Gergovia by a cowp c?e man?

.

....
......
. . . . . .155
155
The attack repulsed with heavy loss
Caesar marches to rejoin Labienus
His critical position . . .
...... .159
. . . . .
158
158

Caesar from crossing the Loire ......

Eporedorix and Viridomarus seize Noviodunum, and try to prevent
159
He saves himself by a series of extraordinary marches
Labienus's campaign against the Parisii ..... 160 . . .

161
He extricates himself from a perilous
And marches to rejoin Caesar
The rebellion stimulated by the adhesion
.......
position

of the
by victory

Aedui .
.

.
.

.164
163
164

They claim the direction of the war . . . . . .164

Vercingetorix re-elected Commander-in-Chief by a general council 165 .

His plan of campaign . . . .165 . . . .

Caesar enlists German cavalry .......

He hounds on the neighbours of the Provincial tribes to attack them 165

...... 166
He marches to succour the Province
Vercingetorix attacks Caesar's cavalry
And retreats, beaten, to Alesia (Mont Auxois)
..... ....
167
168
169
Caesar invests Alesia . . . . . . . . 170
The Gallic cavalry make a
Vercingetorix sends them out
sortie, but are beaten
to fetch succour ..... . . .170
171
Caesar constructs lines of contravallation and circumvallation
Organization of an
1093
army of relief

C
...... . . 171
173
XXXIV CONTEXTS
PAGE
Famine in Alesia 175
Critognatus proposes cannibalism 175
The fate of the Mandubii 175
Arrival of the army of relief 176
The final struggle 176
The self-sacrifice of Vercingetorix 180
Surrender of the garrison 180
Vercingetorix and his place in history 181
Caesar distributes his legions for the winter 182

CHAPTER VIII
THE END OF THE STRUGGLE
Effects of Caesar's victory at Alesia . . . - . 184
Various tribes prepare to renew the struggle 184

Campaign against the Bellovaci ......

Caesar disperses the Bituriges and Carnutes

Caninius and Fabius compel Dumnacus to raise the siege

.

of Lemonum
184
185
189

Blockade of Uxellodunum .......

Drappes and Lucterius take refuge in Uxellodunum

.......
189
189
Execution of Cotuatus
Caesar marches for Uxellodunum ..... 191
191

Surrender of the garrison .......

He cuts off the garrison from their supply

........
of water 191
192
Their punishment
The fate of Lucterius ........
Caesar follows up coercion by conciliation ....
192
193
193

CHAPTER IX
Conclusion 194

PART II
QUESTIONS OF GALLIC AND GALLO-ROMAN HISTORY
RELATING TO THE FOREGOING NARRATIVE
Section Fundamental
The MSS.
When
of the Coynmentaries on
I.

the Gallic War

did Caesar WTite the Cominentaries on the Gallic War, and
.... 201

when were they published

The authorship of the Commentaries
The credibility of Caesars narrative
? . .

......
.
.

.
.

.
.

.
.

.
.

.211
202
209

Section II. — The People

The ethnology
I.

II.
of Gaul
Introduction
The
:

prehistoric races
........
....... 257
268
 . — .

CONTENTS XXXV
PAGE
III. The Ligurians 277
IV. The Iberians 287
V. TheCeltae 303
VI. TheBelgae 322
VII. WTio were the true Gauls ? 325

drusi
The population
.....
The nationaUty of the Eburones, Caerosi, Paemani, Segni, and Con-

of Gaul
338
340

Section III. — Purely Geographical

Caesar's want
The territories of client tribes
The map of Gaul
....
of precision in geographical statements

......
344
344
345
The Itinerary of Antonine and the Table of Peutinger 349
The Gallic league and the Roman mile 350
Geographical Index :

PAGE PAGE
Admagetobriga . 351 Bratuspantium 400
Aedui 351 Cadurci 402
Agedincum . 353 Caerosi 403
Alesia 354 Caleti 404
Allobroges . 363 Carnutes 404
Ambarri 365 Caturiges 404
Ambiani 366 Cenabum 405
Ambibarii . 366 Cenomani 415
Ambiliati . 367 Ceutrones (a) 415
Ambivareti . 367 Ceutrones (6) 415
Ambivariti . 368 Cocosates 415
Andes 370 Condrusi 415
Aremoricae (civitates) 370 Coriosolites 415
Arverni 371 Diablintes 418
At rebates . 371 Eburones 422
Atuatuca 371 Eburovices 422
Atuatuci 384 Eleuteti 422
Atuatucorum oppidum 387 Elusates 422
Aulerci Branno vices 393 Esuvii 422
Aulerci Cenomani 393 Gabali 424
Aulerci Eburovices 393 Garumni 425
Ausci . 394 Gates . 425
Belgae 394 Geidumni 425
Belgium 395 Gorgobina 425
Bellovaci 397 Graioceli 430
Bibracte 398 Grudii 432
Bibrax 398 Helvii 432
Bigerriones 400 Itius Portu^ 432
Bituriges 400 Latobrigi 438
Boi .
400 Lemovices 442
Brannovices 400 Lemovices Aremorici (?) 442
xxxvi CONTENTS

Leuci
Levaci
..... .
PAGE
444
444
Samarobriva
Samnitae .
PAGE
469
469
Lexovii . 444 Santoni 470
Lingones . 444 Seduni 470
Magetobriga . 445 Segni . 470
Mandubii . . 446 Segusiavi . 470
Mediomatrici . 447 Senones 471
Meldi . . 447 Sequani 473
Menapii . 449 Sibusates . 474
Morini ^ . 453 Sotiates 474
Namnetes . . 453 Suessiones . 477
Nantuates . . 453 Tarbelli 477
Nemetes . 455 Tarusates . 479
Nemetocenna . 456 Tigurini 480
Nervii . 456 Tolosates . 480
Nitiobroges . . 458 Treveri 480
Noviodunum (Biturigum) . 459 Triboci 481
Noviodunum (Haeduorum) . 464 Tulingi 483
Noviodunum (Suessionum) . 464 Turoni 483
Ocelum . 466 Uxellodunum 483
Osismi . 466 Vangiones . 493
Paemani . 467 Veliocasses . 494
Parisii . 467 Vellaunodunum 494
Petrocorii . 467 Vellavii 499
Pictones . 467 Venelli 499
Pleumoxii . 467 Veneti 499
Ptianii - 467 Veragri 499
Rauraci . 467 Viromandui 499
Pvedones . 467 Vocates 500
Remi . . 468 Vocontii 501
Ruteni . 469 Volcae 502

Monarchy in Gaul ........

Section IV. Social, Political, and Religious
504

Clanship, senates, and law in Gaul .....

Did two Vergobrets hold office in one state at the same time ?

Did the Gauls, in Caesar's time, recognize private property

.

in
505
507

land?
The Gallic nobiles ........
....
The power of the noble families in Gaul
509
512
513

Slavery in Gaul ........

On the meaning of the words amhacti, clientes, and obaerati

inter-tribal relations in ......

Gaul
514
517
517
Were the philo-Roman and the anti-Roman parties in Gaul identical
with the republicans and the adventurers respectively ? 520
The Druids 523
Was the rebellion of Vercingetorix a democratic movement ? . 529
 — — —

CONTENTS XXXVll

Section V. Relating to the Narrative of Chapter I

PAGE
The Celtic invasion of Italy
The Cimbri and Teutoni .... 542
546
When did Ariovistus arrive in Gaul
The battle of Magetobriga ....
Where was L. Cassius defeated by the Tigurini
?

?
553
554
555
Was Dumnorix Vergobret of the Aedui ? 555
When was Caesar born ? . . . . 556
How many legions did Caesar receive from the Senate and the Roman
People ? 557
Did Caesar intend, before he entered Gaul, to conquer it ? 558

Section VI. Military

The numerical strength of the Caesarian legion 559

The legati
The military tribunes
......
Who made the cohort the tactical unit of the Roman infantry

....
'^
563
563
565
Who were '
the centurions of the iirst rank '
? 567
The/a&ri
Caesar's cavalry
Caesar's artillery
.....
.....
579
579
582

The rations ......

The clothing and defensive armour of Caesar's regular infantry 584
585

Caesar's order of battle

The agger
....
The fortification of Caesar's camps

......
586
587
599

Hhe falx muralis

Caesar's bridges
.....
The vineuy the musculus^ the testudo, and the plutew

.....
608
611
612

Section VII. Relating to the Narrative of Caesar's Campaigns

The routes open to the Helvetii

Rhone
Caesar's lines on the
....
.... 613
614
What route did Caesar take when he marched in 58 b. c. via Ocelum
against the Helvetii ? . 615
Caesar's campaign against the Helvetii 616
What Gallic states were represented in the deputation which con
gratulated Caesar after the overthrow^ of the Helvetii 9 634
The probable length
Ariovistus
Caesar's
.......of

campaign against Ariovistus

Caesar's march from Vesontio against
635
636
Caesar's account of one of the motives of the Belgae for conspiring in
57 B. c 657
Did Adra succeed Galba as commander-in-chief of the Belgae
Where did the Belgae muster in 57 B. c? .... ? 658
658
xxxviii CONTENTS
PAGE
Where did Caesar cross the Aisne in his first campaign against the
Belgae ? 659
How long did Caesar take to march from his
to Noviodunum ?
The battle with the Nervii
....
....
camp on the Aisne

The assault on Galba's camp at Octodurus

To what tribe was Terrasidius sent in 56 b. c. '/

The theatre of the war with the Veneti

Caesar's operations against the Veneti .

Where did Sabinus encamp when he invaded the country of the

Venelli ?

CVassus's attack on the Aquitanian camp

Did Caesar attack the Morini or the Menapii in 56 b. c. ?
Where did the Usipetes and Tencteri cross the Rhine ? ~

On a difficulty raised by G. Long regarding one of Caesar's reasons

for attacking the Usipetes and Tencteri
Where were the Usipetes and Tencteri defeated ?
Wliere did Caesar make his first bridge over the Rhine
Caesar's bridge over the Rhine
Was Commius king of the Atrebates or of the Morini
.... '!

Was Cotta subordinate to Sabinus at Atuatuca ?

The meaning of a disputed passage in B. G., v, 31, ^ 5

Atuatuca ? .....
What was the formation of Sabinus's column, when it marched from

The meaning of orbis in B. G., v, 33, § 3

Q. Cicero's camp ....

The extent of the contravallation with which Ambiorix surrounded

The red-hot
Q. Cicero's
balls with
camp .....
Was the letter which Caesar sent to Q.
which the Nervii set fire to the huts in

Cicero written in Greek, or only

inGreek characters ? .

The meaning of novissimiLS in B. G., v, 56, § 2

The new levies which Caesar raised in 54-53 b. c.
Did Labienus encamp in the country of the Treveri in the winter of
54-53 b. c. ?
Did Caesar confound the Scheldt with the Sambre ?
The raid of the Sugambri in 53 b. c.
The chronology of B. G., vii, 1 .....
How was the news of the Carnutian raid on Cenabum spread abroad
.

........
\Miere did Caesar concentrate his legions at the outset of the Seventh
Campaign ?
The meaning of altero die {B. G., vii, 11, § 1 68, § ; 2)
The passage, qui turn primum comparahant {B.
. . . G., vii, 11, § 4)
On the passage vicos . . . videantur [B. G., vii, 14, § 5)
The meaning of misericordia viilgi {B. G., vii, 15, § 6)

Avaricum ? .........
Where did Vercingetorix make his first camp during the siege of
742
 CONTENTS xxxix
PAGE

awaited Caesar's attack

How Caesar was outwitted by Vercingetorix
?......•
WTiat was the hill near Avarieum on which the Gallic infantry

.... 743
744
\\Tiere did Caesar place his towers during the siege of Avarieum ?
The meaning of ag'S'^r in^. (?.,'vii, 22, § 4 ..... . 745
746
The Gallic wall
The dimensions of the agger at Avarieum ..... 746
748

Avarieum ? .........
How was the column of assault covered before the storming of
749

Gergovia ..........
The stratagem by which Caesar crossed the Allier on his march to
751

.......
Where did Caesar cross the Allier on his march to Gergovia ?

.......
Caesar's operations at Gergovia
Litaviccus's march to Gergovia
. . 754
756
767
The meaning of the passage ipsi ex finitimis transiri videretur , . .

{B. G., vii, 55, §§ 9-10) 769

The disputed passage Nam ut commutato timehat {B. G., vii, 56, § 2)
. . . 771

to rejoin Labienus ?........

Where did Caesar ford the Loire when he was marching from Gergovia

Labienus's campaign against the Senones and Parisii . . .

774
775
Where did Caesar, after his retreat from Gergovia, rejoin Labienus V
How long did he wait before setting out, and from what point
did he set out on his march to succour the Province ? . . 785
Why did Vercingetorix attack Caesar when the latter was marching
to succour the Province ? . .

What was the site of the battle between Caesar and Vercingetorix,
. . ... 790

which immediately preceded the blockade of Alesia ?

^¥\lJ did Vercingetorix retreat to Alesia ?..... . . 791
801
On the number of legions
The operations at Alesia ........
which Caesar had at Alesia . . . 802
804
The passage Tanta tamen
The execution of Vercingetorix
. . .

.......
incumber ent {B. G., vii, 76, § 2)

The attitude of the Aedui during the latter part of the rebellion
.

of
. 820
821

.821
Vercingetorix . . .

AVho wrote the Eighth Book of the Commentaries ?

Caesar's second campaign against the Bellovaci
. . .

...
....
. .

824
826
Where was the bridge by which Dumnacus crossed the Loire ? . .831
Who was Gutuater ?
' '
831
The duration of Caesar's proconsulship . 832
What was the height of the terrace which Caesar constructed at
Uxellodunum ? . . . , . 835
The date of the annexation of Gaul . 835
The spelling of Celtic names 839

.....
.

Addenda 851

......
.

Index . 853
 LIST OF ILLUSTRATIONS
Julius Caesar (from the marble bust in the British Museum) Frontispiece
Colonel Stoffel's methods of excavation . . pages xxv, xxvi, xxvii

....
Gaul in the time of Caesar
Defeat of the Helvetii
. . . . . . fo face page 1
53

....
Operations on the Aisne

......
Battle of Neuf-Mesnil
Gergovia
71
75
149

Alesia
Uxellodunum
......
Labienus's campaign against Camulogenus

.....
161
169
189
The agger at Avaricum, according to General de ReftVe 603
Diagram illustrative of the bridge over the Rhine . ixige 715

[I hope that readers who use the map and plans will turn to the article
in Part II on The Map of Gaul (pp. 345-8).
'
' Those who may wish to
test their accuracy should also consult the Geographical Index (pp. 351-
503) and various notes in Section VII (pp. 613 ff.), which are referred to
in footnotes to the Narrative.]
 :

CAESAR'S CO^WEST OF (lAUL

CHAPTER I

INTRODUCTION
Three centuries before the birth of Caesar, while patrician Gallic

was still struggling with plebeian, while both were still con- ofYtaly
tending with rival peoples for supremacy, the Gauls first battle of

encountered their destined conquerors. For a generation or audits

'^<''*"^^^-
more,^ the Celtic wanderers, whose kinsmen had already
overflowed Gaul, crossed the Pyrenees, and passed into
Britain and into Ireland, had been pouring, in a resistless
stream, down the passes of the Alps. They spread over
Lombardy. They drove the Etruscans from their strong-
holds in the north. They crossed the Po, and pushed further
and further southward into Etruria At length they
itself. 390 b.c.

overthrew a Roman army in the battle of the AUia, and

marched unopposed through the Colline Gate. The story of
the sack and burning of the city was noised throughout the
civilized world yet the disaster itself, though it was never
;

forgotten, hardly affected the history of Rome. It probably

tended to rivet the bonds of union between her and the other
cities of Latium, and to strengthen her claim to supremacy
in Italy. From time to time during the next century the
Gauls returned to plunder but their incursions were repelled
:
;

and the champion of Italian civilization was Rome.

But the Roman dread of the Gauls long remained and ;
Gallic

more than once Rome's enemies enlisted their services against assist the
her. In the last Samnite war, one of the most crucial events enemies of

of Roman history, Samnites, Etruscans, and Gauls made .395 ^ <.

a desperate effort to crush the rising power and after this

;

^ Regarding the date of the Gallic invasion of Italy, and the place from
which the invaders came, see pp. 542-6. .^^'^^^'c Ot" '^'^^''^rj/T**
yisX'^^^^..--—' ^^
1093 B
 2 TNTRODUfTIOX chap.

attempt had been frustrated, the Etruscans once again rose

in revolt, and their GaUic mercenaries destroyed a Roman
army under the Avails of Arretium. It was not until the
283 B.C. Senones had in their turn been defeated and expelled from
Italy, and the Boi, who hastened to avenge them, had been
282 B. c. crushed near the Lake of Vadimo, that the republic was
finally released from the fear of GaUic invasion.
The Years passed away. Rome became mistress of the penin-
tight their sula, and determined to vindicate her natural right to the

r^trM? rich plain on her own side of the Al2)ine barrier. The Gauls
I lie Po

offered a strenuous resistance, and even assumed the offensive.

Reinforced by a swarm of free-lances from the valley of the
upper Rhone, they boldly crossed the Apennines and plun-
dered EtiTjria. The Romans were taken by suqjrise but in :

225 B. c the great battle of Telamon they checked the invasion and ;

within two years they fought their way to the right bank of
the Po. The Insubres on the northern side still held out :

but before the outbreak of tlie second Punic Avar Mediolanum,

222 B.r. or Milan, their chief stronghold, Avas captured and the ;

fortresses of Placentia and Cremona Avere founded.

and con- But the Avork of conquest Avas only half completed Avhen
alpine^''
Hannibal descended into the plain, and the exasperated
Gaul. Gauls rallied round him. When Rome emerged, victorious,
218 B.C. fi-om lier great stmggle, they kncAV AA'liat Avas in store for
them, and made a last attempt to Avin back their liberty.
200 B.C. Placentia aa^s sacked, and Cremona Avas iuA^ested. The
Roman arm^^ Avliich marched to its rehef gained a victory,
199 B.C. but AA^as in its turn almost annihilated by the Insubres. The
Gauls, hoAvcA^er, could ncA^er long act together : their
countrymen beyond the Alps gaA^e them no help the league :

of the northern tribes Avas rent by discord and treachery ;

196 B.C. and the Insubres and Cenomani AA^ere compelled to accept
a peace, aaIucIi alloAved them indeed to retain their constitu-
tion, but forbade them to acquire the Roman citizenship.
South of the Po the Boi stroA^e frantically to hold their oaah :

but in a series of battles their fighting men Avere Avell nigh

191 B.C. exterminated : the Romans insisted upon the cession of half
their territory ; and on both sides of the ri\^er the surAavors
Avere gradually lost among Italian settlers.

I
 -
'

T INTRODUCTTOX 3

Eastward and southward and westward the empire of the Forma

Komans spread. They conquered Greece. They conquered Roman
Carthage. They conquered Spain. But between the central r^i^vmce
and the western peninsula they had no means of communica- alpine
^'*^'
tion by land save what was afforded by the Greek colony of
Massiha. It was an entreaty from the Massiliots for protec- [Mar-
^"^
tion that gave occasion to the wars which resulted in the
'^*^'

formation of the Province of Transalpine Gaul and the ;

natural wilhngness of the Senate to support their most

faitliful allies was doubtless stimulated by the desire to secure

possession of the indispensable strip of coast between the

Alps and the Pyrenees, partly also perhaps by the idea of
creating a Greater Italy for the growing ItaHan population.
In 155 B.C. the Romans stepped forward as the champions
of Massiha against the Ligurian tribes between the Maritime
Alps and the Rhone. The highlanders who inhabited the
forest-clad mountains above the Riviera were crushed in
a single campaign after an interval of thirty years their
;

western neighbours, the Salyes, Avere forced to submit and ;

their seaboard, like that of the other tribes, was given to the
Massiliots. But the Romans had come to stay. The Aedui,
who dwelt in the Nivernais and western Burgund}-, calculated
that the support of the Republic would help them to secure 123 b. c.
ascendancy over their rivals and by a treaty, fraught with
;

unforeseen issues, they were recognized as Friends and AUies

of the Roman people. The AUobroges, on the other hand,
whose home was between the Lake of Geneva, the Rhone,
and the Isere, refused to surrender the king of the Salyes,
Avho had claimed their protection and Bituitus, king of the
;

Arvemi, with all the hosts of his dependent tribes, marched 121 b.c
to support them. Just twenty years before the birth of
Caesar a great battle was fought at the confluence of the
Rhone and the Isere. ^ The GauLs were beaten and the ;

bridges over the Rhone broke down beneath the multitude

of the fugitives.
This victory was, in the strictest sense, decisive. The
Romans were now masters of the lower Rhone ; and if they
^ M. Jullian {Hist, de la Oaule, iii, 17, n. 4). rejecting the tradition, argues
that the battle took place on the Rhone at Pont-St. Esprit.

B 2
 4 TXTRODUCTION chap.

were ever to penetrate into Further Gaul, their base could

be advanced some hundreds of miles. The Arverni, whose
hegemony had extended to the Rhine and the Mediterranean,
had received a blow from which they never recovered.
The Province which was now formed stretched from the
Maritime Alps to the Rhone. Succeeding consuls rapidly
extended the frontier until it ran along the Cevennes and
the river Tarn down into the centre of the Pyrenees. The
tribeswere obliged to pay tribute and to furnish troops ;

and, although they were of course permitted to retain their

own forms of government, their subjection was assured by
the consti'uction of roads and fortresses. The heavy exac-
tions of the conquerors, aggravated by the greed of usurers,
provoked frequent insurrections but year by year the ;

Provincials became steadily Romanized. Roman nobles

acquired estates in the Province, and sent their stewards to
manage them. Roman merchants built warehouses and

[^^^''
bonne.]
,
counting-houses in the towns
tion of Rome began
°
and tlie language and ci\dliza-
to take root.^ Xarbo with its spacious
harbour was not only a powerful military station, but in
commerce the rival of Massilia. Nor was the activity of the
;

...
Romans confined to their proper sphere. Catamantaloedis,
king of the Sequani, whose territory lay north of the Allo-
broges, received from the Senate the title of Friend ; and
the same honour was bestowed upon an Aquitanian noble
and upon Ollovico, king of the Nitiobroges, who ruled the
upper valley of the Garonne.'^ For what services these dis-
tinctions were conferred, we do not know but events were ;

already paving the way for the conquest of the great country
that stretched beyond the Rhone and the Cevennes to the
Rhine and the Atlantic Ocean.
Caul and xiie aspect of this region was. of course, very different
tants. from that of the beautiful France Avith which w^e are famihar.
The land of gay cities, of picturesque old towns dominated
by awful cathedrals, of cornfields and vineyards and sunny
hamlets and smiling chateaux, was then covered in many

'
Cicero, Pro Fonteio, 5, § 11.
- B. G., i, 3, § 4 : iv, 12, § 4 ; vii, 31, § T). Cf. C. Jullian, Hist, de la Gaide,
iii. 28-9.
1 INTRODUCTION 5

places by dreary swamps and darkened by huge forests.

Gaul extended far beyond the limits of modern France,
including a large part of Switzerland, Alsace, Lorraine, and
the Rhenish Provinces, Belgium, and Southern Holland.
The people were divided into three groups, differing in race,
language, manners, and institutions. Between the Garonne
and the Pyrenees were the Aquitani, of whom certain tribes
were akin to the Iberians of Spain. North-east of the Seine
and the Marne, in the plains of Picardy, Artois, and
Champagne, on the mist-laden flats of the Scheldt and the
lower Rhine and in the vast forest of the Ardennes, dwelt the
Belgae, who may have partially mixed and were continually
at war with their German neighbours. The lowlands of
Switzerland, Alsace, Lorraine, and part of the Rhenish
Provinces, the great plains and the uplands of central France,
and the Atlantic seaboard, were occupied by the Celtae.
Modem science, however, has established a more detailed Prehistory

classification. Neither in Aquitania nor in Celtica nor in

the land of the Belgae were the people homogeneous.
During the last fifty years the classical texts, which were
once the only source of knowledge, have been supplemented
by geological, archaeological, and anthropological research ;

and it has become possible to reconstruct the prehistorj^ the

very existence of which had hardly been suspected, of every
European land. Skeletons have yielded information about
the physical characters of the people their implements
:

and weapons, their clothing and ornaments, their art, and

even their religion, have been revealed by relics extracted
from the river-drift, from caverns and sepulchres, from
\allages, hill-forts, and buried hoards. The Celts were but the
latest invaders of Gaul and their life was profoundly
;

influenced by the Ligurians, the Iberians, and the nameless

tribes who, during countless millenniums, had dwelt in Gaul
before them.^ To what era is to be assigned the first appear-

^ The very brief sketch of the prehistory of Gaul to whieh, in conformity

with the plan of this book, I have restricted myself, rests upon copious materials.
The evidence for the purely ethnological parts will be found in the essay on
'The Ethnology of Gaul (pp. 257-338). For the rest I have erased my biblio-
'

graphical notes since the appearance of M. J. Dechelette's Manuel d'archeologie

—
(1908, 1910), a synthetical work of the first rank, which deals in the fullest
6 IXTKODUCTION chap.

ance of man, is still a disputed question. Some enthusiasts

insist that even in the Tertiary Period, myriads of years
ago. the country was inhabited by men. if men they could
be called, who wrought for themselyes flint implements
which remain as their sole memorial but the more cautious ;

experts who haye handled these stones are unable to admit

that tliey were chipped, or hacked, b}' the hand of man.
Eyen after the close of that period our own countrs^was still
part of tlie continent, and the great ice-age had hardly
begun. Thenceforward uncertainty disappears. In tlie
Quaternary Period came the palaeolithic races, whose
existence is attested not only by their weapons but by their
own remains. These men maintained themselyes in Gaul
during tlie last interglacial epoch, and sheltered in cayes
throughout the countless centuries in which the glaciers
w^ere once more spreading and at length receding oyer the
uplands of Central Europe.^ Earliest of all were the Neander-
thal, or, as they are sometimes called, the Canstadt race,
A^th their low brutish foreheads and huge beetling brows,
whose skeletons haye been found in the basin of the Meuse
and between Croatia and the Dordogne. Although the oldest
is not earlier than the second of the stages, marked by

successiye t^^es of implements, into Avhich the Palaeolithic

Age has been divided, there is reason to believe that the
men who fashioned the more primitive tools belonged to the
same stock. That race saw the volcanoes of Auvergne,
Avhich during countless centurieshave slumbered, belching
forth flameand discharging lava - but for them all Xature ;

was animated, and lava and flame were demoniacal. Mam-

moths and rhinoceroses, lions, bears, and hyaenas, bisons,
gluttons, wolves Avere their fellows and over the vast ;

expanse of the forest-cumbered land, Adhere they roamed in

detail with Gaul. The fir.st volume, which was ably reviewed by M. Boule
ill V Anthropologie. xix, 1908, pp. 451-2. is dcAoted to the Palaeolithic and
the Xeolithic Age : the second, which is even better, to the Bronze Age. The
concluding volume, treating of the Iron Age, will probably be published before
the end of 1911.
1 See VAnthropologie, xix, 1908, pp. 5, 7-10, 12-3, 615.
* Bull, de la Soc. de geogr., xiii. 1906. pp. 291.365-7: Associ<tiion jran^. pciur
Vavancement de-s sc. 37'' ,sess", 1908. pp. 604, 692-4.
 I INTRODUCTION 7

quest of food, there was no sign, save their rude handiwork,

that they would rise superior to the beasts which the primi-
tive savage regards with mingled fear and veneration. ^ Yet
they buried their dead with scrupulous care, sometimes
placing tools beside them and we may perhaps infer that
;

they fancied that the soul would still endure. But they were
not the only race who then inhabited Gaul. In the caves of
Mentone have been found human remains little later than
the oldest of the Neanderthal stock, which show negroid
characteristics, and also gigantic skeletons with well-
developed skulls, belonging to a type whose most famous
representative was unearthed many years ago beneath the
rock-shelter of Cro-Magnon in the Dordogne ; and before
the end of the Quaternary Period there were living in the
caves of Laugerie-Basse and Chancelade a people who, if we
may judge from their well-formed and capacious skulls,
possessed an intellectual faculty not inferior to that of their
modern descendants. They have indeed left evidence of
their powers ; for late in the Palaeolithic Age appeared the
dawn of pictorial art. From the caves of the Tarn-et-Garonne
and the Dordogne have been recovered bones and antlers,
engraved or carved with likenesses of mammoths, reindeer,
and other animals, of fishes, and of men. Those who have
seen the specimens of their work preserved in the museums
of France, in which the essential alone is depicted with sure
touch, or the frescoes with which, by the dim light of their
rude lamps, they covered the walls of the Pyrenean caves,
"^

will hardly believe that they were not inspired by a love of

beauty for its own sake but it is not improbable that their
;

motive was also magical the animals which they portrayed

:

may in certain cases have been totems, and the sudden

extinction of palaeolithic art, which has caused so much
perplexity to archaeologists, may perhaps be explained by the
decay of totemism in neolithic times.^ The palaeolithic races
' See E. B. Tylor, Primitive Culture, 4th ed., 1903, i, 467-9 ; ii, 229.
* See Rev. arch. 4« ser., xiv, 1909, p. 126.
;

See Rice Holmes, Ancient Britain, &c., 1907, pp. 47-8, and J. Dechelette,
*

Manuel d' arcJieologie, i, 1908, pp. 268-71. There is an authoritative and

interesting article in Rev. arch., 4^ ser., xiii, 1909 (see especially pp. 407, 409-11),
on the successive phases of palaeolithic art.
t
 8 INTRODUCTION chap.

were all dolichocephalic : their heads, that is to say, were long

in proportion to their breadth ; and the same characteristic is

found in the skulls of the slender stunted people of THomme

Mort and Baumes-Chaudes, w^ho, though they were descended
from the older inhabitants, belonged to the Neolithic Age.
These peoples, who are called after the caverns in which the
first specimens were found, appear to have been diffused

over the length and breadth of Gaul. But as the new epoch
advanced, new races began to appear and the invaders who
;

came from the east, and gradually mingled with the abori-
gines, were a short but sturdy folk, characterized by brachy-
cephaly, or great breadth of skull. The palaeolithic hunters
had been forced to wander in search of game their successors, :

who learned to domesticate cattle and ultimately to till the

soil, settled in the tractswhich they had found. Among the
neolithic tribes were some whose chiefs erected dolmens, or
vast structures of stone, to cover the sepulchres of their
dead. Some are of enormous size, and could only have been
. erected by the toil of multitudes, controlled and organized
1 by chiefswhose motive was to honour and propitiate the
spirits that they believed to survive. At Perotte, in the
department of the Charente, a stone was set up which
weighed forty tons and had been quarried twenty miles away
the tumulus of Mont St. ^Michel in the Morbihan is a veritable
hill, and contains more tlian forty thousand cubic yards of

stone. The dolmens are not all of one pattern some of them :

contained implements of bronze as well as of flint and the ;

skeletons which have been found in them belong to more

than one race. The era in which they were constructed was
marked by considerable commercial activity for some of ;

them have yielded ornaments of a mineral resembling tur-

quoise, which must have been imported amber beads had
;

alreadv been conveved from the Baltic by wav of the Elbe,

the Moldau, and the Danube and flint from the factory of
;

the Grand-Pressigny in the Indre-et -Loire was diffused as far

as Switzerland. The huge stone monuments which Caesar
doubtless saw when his legions entered Brittany, were only
one of many groups which extended along the coast from the
Pyrenees to the Channel, and were scattered over Central
I INTRODUCTION 9

Gaul but not a single dolmen has been found on Gallic soil
;

east of the great barrier formed by the Jura and the Vosges.
Their source is still uncertain but their distribution in Gaul,
;

in Africa, and in the islands of the Mediterranean, combined

with resemblances which can only be explained by com-
munity of origin, suggests that the types were derived from
the Aegean region and reached Gaul by coastal navigation,
which even then linked west to east. In Gaul, as in Britain,
the investigator perplexed by rehcs of superstition which
is

appear also in far distant lands, but which cannot be ex-

plained by any theory of common origin. Red hands were
painted in Pyrenean, as they are in Australian, caves skele- ;

tons are disinterred in Europe and in America with the knees 1

drawn up towards the chin ; cup-markings appear on

megalithic monuments in every quarter of the globe.
Slow"ly, insensibly, civilization moved onward. There is
evidence to show that the Neolithic Age set in nearly ten
millenniums before our era ; the Bronze Age, Avhich succeeded
it,began before 2000 B.C. and it was not until more than
;i

a thousand years had passed that the culture which derives

its name from the Tyrolese settlement of Hallstatt, and in
which bronze, as material for tools and weapons, gradually
gave place to iron, spread westw^ard across the Rhine. '^

With the commencement of the Bronze Age, although

history is still unborn, our knowledge becomes more precise,
and absolute, not merely relative chronology may be
attained. The Bronze Age and the Hallstatt period are each
marked by well-defined stages, dated by the correspondence
and in later cases by the association of the objects that
represent them wdth works of Mediterranean craftsmen, the
dates of which are approximately known. The knowledge
of metals penetrated into Gaul by two routes, of which the
starting-point w^as in the Aegean. South-Eastern Gaul w^as
served by the route that led through Central Europe ;

^The early Bronze Age in Southern Gaul was, as might have been expected,
contemporary with the late XeoHthic in the north. This accounts for the
discovery of bronze in southern dolmens and its absence in those of Brittany.
^ Relics of
the earlier part of the Hallstatt period have been found in France
•only in the north-eastern and central districts. Probably, therefore, the Bronze
Age lasted in the rest of the country till about 700 b. c.
10 INTRODUCTION chap.

Western Gaul borrowed from Spain. It must not, indeed,

be supposed that the Gallic tribes were merely imitative.
The types which were introduced from the more favoured
East developed along national Hnes. But side by side with,
the products of their own manufacture they continued to
use many that came from abroad. Although the memory of
inter-tribal war is preserved by the earthworks and stone
forts which, even in the Neohthic Age, had been erected
upon the hills, commerce, internal and external, advanced
with rapid strides. Forests were gradually cleared and ;

trackways were laid out from village to village. Caravans

began to cross the Alps from the valley of the Po. Gold
crescent-shaped ornaments, intended to be worn round the
neck, and fancifully decorated with geometrical figures,
were brought from Ireland comparison of the tj^es of
:

pottery, of knives and axes, razors and swords, of bracelets,

pins, and brooches, shows that many were derived from
Italy and Germany and before the end of the Hallstatt
;

period trade was established with the Greeks, while w^ine

was imported and distributed by the merchants of Massilia.
But, whatever the cause may have been, the sterility of
art was remarkable in the land which had produced the
schools of the Dordogne and the Pyrenees. The art of the
Bronze Age. if art it can be called, represented, with insig-
movement nor life the decora-
nificant exceptions, neither :

tion was merely geometrical. Even sepulchral architecture

decayed. The chambered tombs of the Neolithic Age, which
m Greece gave birth to the noblest forms, w^ere succeeded
in Gaul and throughout the west by structureless mounds.
Such art as existed was inspired principally by rehgion.
Stone idols, quaintly sculptured in the form of female breast
and face,whose worship is traceable through Spain to the
Aegean, have been observed in the departments of the Gard,
the Aveyron, the Herault, and the Tarn.^ In Gaul, as in
other lands, though rehgion was doubtless intertwined with
sorcery, the forces of nature were adored and the worship ;

of the sun, which naturally arose with agriculture, was

symbolized in decorative forms. Sepulchral customs also
^ These idols probably belonged to the Copper Age.
1 INTRODUCTION 11

bear vague testimony to religious beliefs and it may be that

;

cremation, which prevailed throughout the greater part of

the Bronze Age, was partly due to a fancy that the burning f

of the body would hasten the liberation of the soul.

The earliest inhabitants of Gaul about whom history has Tlio Lig«-
1 . II ii X • A T• ji J
nans and
anythmg to tell were the Ligurians. Accordmg to the ibeiians.

ancient geographers, the land which originally belonged to

them in Gaul was the mountainous tract between the Rhone,
the Durance, and the Cottian and Maritime Alps but by :

the hfth century before Christ they were mingled with

Iberians on the west of the Rhone ; and from the evidence
of certain geographical names as well as of archaeology, it

would seem that they once possessed the whole of Eastern

Gaul as far north as the Marne. The culture of this region
in the Bronze Age differed from that of the west, but closely
resembled that of Northern Italy, where we know that
Ligurians lived. The vast number of sickles which have
been discovered in the south-east show that the Ligurians
were industrious tillers of the soil and they may have been
;

descended, at least in part, from Swiss lake-dwellers of the

Stone Age, who probably introduced cereals and domestic
animals into Gaul. The origin of the Iberians remains
uncertain but when they came under the notice of the
:

Greeks they occupied the eastern part of Spain as well as the

country between the Pyrenees and the Rhone ; and it should
seem that they had crossed the Pyrenees and made conquests
in Aquitania as well as on the Mediterranean coast. The
'
Iberian question is one of the problems which amuse and
'

baffle ethnologists for there can be little doubt that in the

;

land which belonged to the Iberians of history, in Spain as

well as in Southern Gaul, there once existed, besides Celtic,
at least two forms of speech, —
Basque and the uncouth,
undeciphered language or languages in which were engraven
the so-called Iberian inscriptions. But if the Iberians
were not one race, the bulk of them were small and dark,
and not unlike the neolithic people of 1' Homme Mort. In
Caesar's time Liguria, as well as the land of the Iberians,
was also peopled by the descendants of Celtic invaders. It The alts.
was perhaps in the seventh century before the Christian era
 12 IXTKODUCTIOX chap.

that the tali fair Celts began to cross the Rhine :

^ but it is

unlikely that even they were homogeneous ; and those to

whom belonged the characteristics which the ancient A^Titers
associated with the Gallic or Celtic type may have been
accompanied by the descendants of aliens who had joined
them during their long sojourn in Germany. Successive
swarms spread over the land, partly subduing and mingling
with the descendants of the palaeolithic peoples and of their
neolithic conquerors, partly perhaps driving them into the
mountainous tracts. Physically, they resembled the tall fair
Germans Avhom Caesar and Tacitus describe but they ;

differed from them in character and customs as well as in

speech.^ The Belgic Celts were the latest comers and among ;

the Belgae of Caesar's time the aboriginal elements were

comparatively small. If Caesar was rightly informed, the
languages of the Belgae and the Celtae were distinct. Both,
it is needless to say, were Celtic, and the difference may not

have been great for if a Goidelic dialect was spoken anywhere

;

in Gaul, the vestiges of Galhc that remain belong, for the

most part, to the Brythonic branch of the Celtic tongue.'^ In
Aquitania the natives remained comparatively pure, and
formed a separate group, which, in Caesar's time, stood
politically apart from the Celtae as well as from the Belgae.
They are generally spoken of as an Iberian people but the ;

name is The conquering Celts, as the evidence

misleading.
of nomenclature shows, had advanced, though probably in
small numbers, beyond the Garonne and evidence supplied
;

\)y recent measurements of living inhabitants appears to show

that in certain parts of x\quitania the round-headed element

V was considerable. But it is certain that the Celtic language
was not generally spoken in Aquitania and the Iberian type ;

was sufficiently conspicuous to give some colour to the

popular theorv\
Thus when Caesar entered Gaul, the groups whom he called
Belgae, Celtae, and Aquitani were each a medley of different
races. Tlie Belgae were the purest and the least civilized

'
See p. 274 and Rice Holmes, Anc. Britain, pp. 432-3.
- See mj^ essay on The Ethnology of Gaul,' pp. 325-37.
'

^ See pp. 281-2, 319-21.

T IXTRODUCTIOX 13

of the three and both in Belgic and in Cehican Gaul the

;

Celtic conquerors had imposed their language upon the

conquered peoples. Even in a political sense, the Belgae and
the Celtae were not separated by a hard and fast line for ;

the Celtican tribe of the Carnutes was among the chents

of the Belgic Remi, while on the other liand the Celtican
Aedui claimed supremacy over the Belgic Bellovaci. But
if not scientifically complete, the grouping adopted by Caesar

was sufficient for the purpose of his narrative. Just as a

modern conqueror, without troubling himself about recon-
dite questions of ethnology, might say that the people of
Great Britain were composed of Englishmen, Scotsmen, and
Welsh, so Caesar, knowing and caring nothing about ethnical
subdivisions, divided the people of Gaul into Belgae, Celtae,
and Aquitani.
But who would be content with the mere knowledge of
the physical characteristics of the races of which a people
was composed, or even of their arts and crafts, their imple-
ments and weapons, their dwellings and their tombs ?
Measurements of skulls, tables of stature, inventories of
archaeological discoveries, —
these things have their uses ;

but they leave our curiosity unsatisfied. Even the arrows

and the harpoons that have been found in the caves of
Perigord and the Dordogne, the potteiy, the tools, and the
ornaments that have been taken from the dolmens to enrich
—
the museums of France all the unwritten documents which
—
patient thought has classified have only enabled the most
dihgent of antiquaries to piece together an outline of the
culture of prehistoric men. I have attempted to describe its
course in our own island : here I desire only to show that
Gallic history had its roots in the remote past ; and the
primitive life of Gaul, in its main features, though more
advanced, was not very different from that of Britain.
Human interest, indeed, there is for the student whose
learning is quickened by imagination : but the personal
element is and when the races have amalgamated
lacking ;

into the three groups of Belgae, Celtae, and Aquitani, and

the epoch of Roman conquest is approaching, we desire to
know more. What manner of men were the inhabitants of
14 INTRODUCTTOX chap.

Gaul ?question can be answered, the answer can only

If this
come from a mind subtle and powerful no less than well-
informed. Every man has his own character. Yet, with all
the idiosyncrasies which distinguish them one from another,
Yorksliiremen have a common type of character which
differentiates them from the men of Kent Englishmen have :

a common type which differentiates them from Scotsmen ;

and finally Englishmen and Scotsmen have something in

common, which, in the eyes of foreign observers, differen-
tiates the people of Great Britain, morally and intellectually,
from the other nations of the earth. For in our own. as in
other lands, long association, intermarriage, the prolonged
influence of common conditions of life, have given to originally
distinct groups, without destroying the individuahty of any,
a common recognizable, if indefinable, mental, and even
physical, type. To some, though for ob\'ious reasons a less
degree, thesame causes must have operated in Gaul. Setting
aside the Aquitani. of whom Caesar had little to tell, and
perhaps also the Belgae, the medley of peoples whom he
called Galli had probably so far coalesced that they had
* '

acquired certain common Perhaps when

traits of character.
he described the features of the Gallic temperament which
had most impressed him in the course of the war, he took
little note of the lowest class, the cultivators and the shep-
herds, who had not much t do ^^'ith political life but we can
o ;

hardly suppose that his remarks apjDlied only to the ruling

class or to the purer Celts.^ To attempt the portrayal
of national character is often as misleading as it is tempting ;

but guided by Caesar's observations, we cannot go far astray

even if we do not go veiy far. The Gauls w^ere an interesting
people, enthusiastic, impulsive, quick- wdtted, versatile, vain-
glorious and ostentatious, childishly inquisitive and childishly
credulous, rash, sanguine, and inconstant, arrogant in victory
and despondent in defeat, submissive as women to their
priests, impatient of laAv and discipline, yet capable of loyalty
to a strong and sympathetic ruler.
The notices which Caesar and other writers have left of

1 See especially B. 0., ii, 1. § 3 ; iii, 19, § ; iv. .">.

§§ 2-3, 13, § 3 ; vii. 20-1 ;

ttnd compare Strabo, Geogr., iv, 4, §§ 2-6.

T TNTKODUCTIOX 15

their civilization have been supplemented by the more Civiliza-

informing evidence of archaeology. Five centuries before Qauls.

the birth of Christ the culture of Hallstatt had given place
to that which takes its name from the village of La Tene, at
the northern end of the lake of Xeuchatel, where, some sixty
years ago, was discovered a precious series of antiquities.
The art, essentially Celtic, characterized by the tasteful use
of curves, which was practised in the design and decoration
of these objects, Avas in part an outgrowth of that of Hallstatt,
but also owed much to classical and even to oriental influences.
Imported into Britain by the Br\i:honic invaders, it there
shook itself free from all trammels, and attained an even
higher level than in Gaul, culminating in the graceful and
exquisitely decorated shield of bronze and red enamel wliich
adorns the Central Saloon of our NationalMuseum. SpeciaHsts
have determined three periods, known as La Tene I, II, and
III, of which the last began about forty years before the
2)roconsulship of Caesar.^ By that time the Gallic peoples
had all risen far above the condition of barbarians and the ;

Celticans of the interior, many of Avhom had already fallen

under Roman had attained a certain degree of
influence,
civilization and even of luxury. Their trousers, from wliich
the Province took its name of Gallia Braccata, and their many-
coloured tartan shirts and cloaks excited the astonishment of
their conquerors. The chiefs wore rings and bracelets and
necklaces of gold and whenthose tall fair-haired warriors rode
;

forth to battle with their helmets wrought in the shape of some

lierce beast's head and surmounted by nodding plumes, their
chain armour, their long bucklers, and their clanking swords,
they made a splendid show\'- About fifty years before Caesar's
time, war-chariots, which had excited the astonishment of

^ Pending the i)ublication of the final a olume of 3[ Dechelette's Manuel

d'archeolofjie, readers who do not caie to consult the articles in the Hevue
arcJieohgique and other learned periodicals will tind a good account of the
culture of La Tene in the Guide to the Antiquities of the Early Iron Age (British
Museum), with which compare G. Dottin's ManufJ de Vantiquiti celtique, llKMi.
pp. 119-65.
- Diodorus Siculus, y, 28, § 3 30 Livy.
; ; vii, 10, §§ 7, 11 ; Virgil, Aen.. viii.
HBO. mi ; Strabo, iv. 4, §§ 3, 5; Tacitus, Hi.-it., ii, 20; Journal des Savanta,
1880, pp. 76-8 ; Bull, laonumental, 6""
ser., ii, 1886, pp. 175-96.
16 INTRODUCTION chap.

the Romans in the battle of Telamon/ and which were still

used in Britain, had fallen into disuse, probably because the

wealthy natives had begun to import horses powerful enough
for a charge of cavalry but from the older graves of the
;
-

Marne, which have yielded numerous remains of these cars,

bronze horse-trappings of most delicate open-work and
bronze flagons which had been fetched from Greece, have
been unearthed.^ The arts of building and of fortification had
made a Walled towns or large villages,
considerable advance.
the strongholds of the various tribes, were conspicuous on
numerous hi lis. ^ The plains were dotted by scores of open
hamlets. The houses, built of timber and wattle-work, were
large and well-thatched.^ Tweezers and ornamented mirrors
of bronze lay on the tables of Gallic dames. Painted pottery,
decorated with spirals or symmetrical curves, was used
everywhere, except, apparently, in the remote Aremorican
peninsula.^ The fields in summer were yellow with corn.
The vine was not yet cultivated but the merchants of :

Massiha imported wine from Italy and wealthy Gauls ;

would eagerly barter a slave for a jar."^ Roads, suitable for

wheeled traffic, ran from town to town. Rude bridges
spanned the rivers and barges, laden with merchandise,
;

floated along them. Ships, clumsy indeed but larger than

any that were seen on the Mediterranean, braved the storms
of the Bay of Biscay and carried cargoes between the ports
of Brittany and the coast of Britain. Tolls were exacted on
the goods which were transported on the great water-ways ;

and it was from the farming of these dues that the nobles
derived a large part of their wealth. ^ The Aeduans were
familiar with the art of enamelling. ^ The miners of Aquitaine,
of Auvergne, and of the Berri were celebrated for their skill. ^^

1 Polybius. ii, 28, §§ 5-6. Rice Holmes, Anc. Britain, pp. 342-3.
-

"Guide to the Ant. of the Early Iron Age (Brit. Museum), pp. 50-1 Association ;

J ran^. pour V avancement des sc, 36*^ sess**, 1907, pp. 281, 875-9.
See Cong res archeol. de France, Ixix^ sess", 1902 (1903), p. 181.
••

^ Strabo, iv, 4, § 3 ; A. Grenier, Habitations gauL, &c,, 1906, pp. 23-48.

^ Guide to the Ant. of the Early Iron Age (Brit. Museum), pp. 67-8, 75.
' Diodorus Siculus, v, 26, §§ 3-4. « B. G., i, 18,
§§ 3-4.
**
J. G. Bulliot and H. de Fontenay, L'art de V emaillerie chez le^ Eduens, 1875.
'
CL p. 20, infra. '» B. G., ill, 21, § 3 ; vii, 22, § 2.
 —
I INTRODUCTIOX 17

Every tribe had its coinage and the knowledge of writing,

;

in Greek and in Roman characters, was not confined to the

priests. Diodorus ^ remarks that the Gauls tlircAv letters,
addressed to the dead, on to the funeral pile and Caesar,
;

after he had defeated the Helvetii, found in their encamp-

ment a .scliedule, on wliich were recorded in Greek characters
the names of indi\4duals, the number of emigrants capable
of bearing arms, and the numbers of old men, women, and
children. It would seem, indeed, that some knowledge of
Latin had penetrated even to the rudest tribe of the Belgae.-
^^llen Caesar was marching to relieve Quintus Cicero, who
was besieged by the Nervii, he wrote to him in Greek charac-
ters, for fear the letter might be intercepted and read.^ At
an earher time there were natives, at least in the Province,
who acquired a smattering of Greek. Rich enthusiasts
resorted to Massiha as a school of learning, and became so
enamoured of Greek culture that they ^vrote their contracts
in the language of their teachers.^ Indeed in all that be-
longed to outward prosperity the peoples of Gaul had made
great strides since their kinsmen first came in contact with
Rome, and the enormous fortunes wliich Caesar and his
staff amassed are evidence of their wealth.
The coins which have just been mentioned require special Coin?

notice for none of the antiquities of the Later Iron Age have
;

thrown more Hght upon the culture of the Gauls. The oldest
were copied in the earlier half of the third century before
Christ from gold staters of Philip of Macedon, which had been
introduced through Massilia. For some time they bore no
inscription, except the name of Philip, more or less deformed ;

but about the middle of the following century more than —

a hundred years before the same change was made in our
—
island they began to be stamped with the names of the
rulers by whom they were issued, among whom are to be
recognized some who have been commemorated by Caesar,
notably the great Vercingetorix, whose coins are worth about

' V, 28, § G.

On the other hand, it nm;<t be lenieinbeied that Caesar conversed with

•^

Biviciacus through an interpreter {B. G., i, 19, § 3).

' .See
pp. 730-1. * Strabo, iv,
1, § 5.

1093 (J
18 INTRODUCTION chap.

fifty times their weight in gold. Greek characters are some-

times quaintly jumbled Avith Latin, which gradually became
familiar after the Romans had established their footing in the
land. Many Roman must have been circulated
coins, indeed,
in Gaul after the colonization of Narbo and Roman influence
;

is apparent on many Gallic coins, for example in a figure

of Pegasus, which appears on one that bears the name of

Tasgetius, king of the Carnutes. For many years gold coins,
Avhich gradually deteriorated in Aveight, and, through suc-
cessive copying, in form as well,^ were the only medium of
exchange ; and
but, as commercial needs increased, silver
bronze passed gradually into use,^ the coins of the latter
metal being imitated from those of Massilia, and, in the case
of certain Belgic specimens, even from those of Campania,
The coins, indeed, illustrate not only the commerce of the
Gauls, but also their inter-tribal relations, their manners
and customs, and perhaps occasionally their religion. Thus,
Avhile the extreme rarity of Arvernian coins in the great
mart of Bibracte may
perhaps be explained by the tra-
ditional enmity between the Arverni and the Aedui, the
discoveries of British coins in Gaul and of Gallic coins in
Britain attest the maritime trade which Caesar notices ^ ;

coins of Central Europe found as far Avest as Saintonge and

Gallic coins found in the Bohemian stronghold of Stradonic
prove that the Gauls had intercourse Avith the valley of the
Danube Massilian coins found in various parts of Gaul
;

bear witness to the enterprise of the Greek colony and ;

numerous hoards of silver coins of one type, apparently

copied from those of Rhoda, all of which have been found
in the basin of the Garonne, confirm the impression AA^hich
Ave derive from the Comynentaries that the relations of
Aquitania were mainly Avith Spain. Again, when we notice
that horses and swine are figured on Gallic coins more

• I'M. R. Forrer,' says M. Blaiicliet {Tniitc des monn. gauL, 1905, p. 186, n. 2),
'
a eu ringenieuse idee de faire executer par des cnfants des copies successives
de tetradrachnies de Tliasos. Les resultats obtenus sont tres analogues, pour
la barbarie, aux types des imitations frappees par les peuples du Danube,' &c.
- Silver and bronze coins were hardly used before the time of Caesar. See
V. Tourneur, La monnaie de bronze des Tonyrois, 1909, p. 10, and cf. Bev. celt.,
^ Cf. Rice Holmes, Anc. Britain,
xxxi, 1910, p. 57. p. 250.
1 INTKODUCTION 19

frequently than any other animals, we are reminded of the

passage which Caesar observes that the Gauls imported
^ in
well-bred horses at great cost, and of the passage in which
Strabo ^ speaks of the hams which the Sequani exported to
Italy. Shields and trumpets remind us of Diodorus's ^
description of Gallic arms and the lyre, which is figured
;

on certain coins, may represent the instrument with which

the bards accompanied their songs.* It is remarkable that
all the coins which have been found in the great strongholds

—
are of late date not earlier than about a hundred years
—
before the Christian era which tends to show that none
had been founded more than half a century before Caesar
entered Gaul. Probably Avaricum, Bibracte, Lutecia, and
the other towns which he mentions were fortified during the
invasion of the Cimbri and Teutoni, which devastated Gaul
between 113 and 109 B.c.^
Of all these towns the one which is best known to us was Bibiacic,
Bibracte, described by Caesar as by far the wealthiest and '

most important town of the Aedui ', which stood upon Mont
Beuvray, a few miles west of Autun. If Cicero had visited
it he might perhaps have spoken with less disdain of the

urban life of the Gauls. ^ Streets, workshops, ramparts have

been revealed by excavation. Fifteen hundred coins, nine-
tenths of which belonged to the period of independence,
testify to the manifold commercial relations of the inhabi-
tants. The houses show that the round conical wooden
huts which Strabo described were only the more primitive
productions of Gaulish domestic architecture. Like them,
indeed, the houses at Bibracte were partly subterranean,
this form having been adopted as a precaution against cold
on such a high altitude, and probably, like the modern
cottages of the Morvan, they were thatched with straw.
;

but their shape was rectangular, they were built of stone

1 i^. 6^., iv, 2, § 2. iv,3,§2; 4, § .3. '^

' V, 30, §§2-4. 31,§2. " /6.,

•'
8ee A. Blanchet's excellent work, Traite des monnaies gauloises, 1905,
especially pp. 30, 34-5, 46, 51, 68-9, 75-6, 94, 161-3, 165, 167, 178, 187, 193,
195-6. 224, n. 4, 476-7, 484, n. 1, 498, n., 3, 515, 517-20 ; and Rev. numism., 4"
ser., xi, 1907, p. 475.
® See Cicero's speech, De prov. cons., 12, § 29.

C 2
 20 INTRODUCTION chap.

compacted with and they were entered by an interior

clay,
staircase. The crucibles, moulds, and polishing-stones of
enamel- workers, broken tools, brooches, and pottery, all
belong, like the coins, to the latest period of the Celtic Iron
Age. Besides these relics of native workmanship were
painted vases, imported from Italy, which Gallic artificers
soon learned to imitate.^
Stradoiiio. Evcn morc remarkable than the revelation of the buried
industry of Bibracte the discovery of the fortified town of
is

Stradonic, about twenty miles south-east of Prague, which

is a duplicate of the Aeduan
hundred miles aw^ay. capital five
Pottery, enamel, implements, ornaments are the same and ;

the coins are closely analogous. The explanation of this

coincidence is still obscure ; but, since certain links between
the twin strongholds have been found in the valley of the
Rhine and in Switzerland, it seems not improbable that the
same industrial types were conveyed by commerce towards
the east and towards the west.-
But the growth of material prosperity had not been
matched by true national progress. The Aquitani, indeed,
the maritime tribes, and the Belgae were untouched by
foreign influences but the Celticans of the interior had been
;

enfeebled by contact with Roman civilization. Much non-

sense has been written about the enervating effect of luxury.
Its effect, however, when it is suddenly introduced among
a half-civilized people, is quite different from its effect when
it is a natural growth. The Gauls had lost the strength of
barbarism, and had not gained the strength of civilization.
They had once, as Caesar remarked, been more than a match
for the Germans but enervated by imported luxury, and
;

cowed by a succession of defeats, they no longer pretended

to be able to cope with them,
rolitical Their constitution was based upon the tribe, if that word
and social
j^^^y be applied to the political unit which Caesar called
organiza- ; .

tion. a civitas. The tribe was generally an aggregate, more or less

compact, of communities to which he gave the name of pagi,

'
J. (i. Bulliot. Fvidlles (lit Mont Beuvniy, 1809. i, 123-6, 129-40 ; ii, 3-4-4;

UAnthr., 1902, pp. 74-83.

xiii, Cf. Rev. arch., 4« ser., xv, 1910, p. 104.
- Congres archeol. de France, Ixvi** sess", 1899 (1901), pp. 119-82.
r TNTRODUOTTON 21

the members of which had originally been related by blood

or by near neighbourhood but it would seem that some of ;

the smaller tribes consisted each of one pagns only. Each

pagus, under its own magistrate, appears to have enjoyed
a certain measure of independence, and to have contributed
its separate contingent to the tribal host.^ Each tribe had
its council of elders, and had once had its king but in :

certain tribes the king was now superseded by an annually

elected magistrate ; while in others perhaps the council kept
the government to which prevailed among the
itself. A rule
Aedui illustrates the jealousy which was felt of monarchical
power. In that state the chief magistrate, who was known
as the Vergobret, was forbidden to stir beyond the frontiers
of the country, from which it may be inferred that it was not
lawful for him to command the host. The executive was
generally weak. Some
communities of which
of the smaller
a tribe was composed occasionally acted on their own account,
in opposition to the rest or to the policy of the tribal authori-
ties.^ Like the Anglo-Saxon thanes and the Norman barons,
the surrounded themselves with retainers, loyal
nobles —
followers or enslaved debtors ^ and none but those who ;

became their dependants could be sure of protection. On

the other hand, none but those who were strong enough to
protect could be sure of obedience. The oligarchies were no
more secure than the monarchs whom they had supplanted.
These men or their descendants sullenly plotted for the
restoration of their dynasties, and, reckless of the common
weal, they were in the mood to grasp the hand even of a
foreign conqueror, and to reign as his nominees. Here and
there some wealthy noble, like Pisistratus in Athens, armed
his retainers, hired a band of mercenaries, won the support

^ Sir Hemy Maine {Early Hist, of Institutions, 1875, p. 30) speaks of '
Caesar's
failure to note the natural divisions of the Celtic tribesmen, the families and septs
or sub-tribes'. See, however, F. de Coulanges, Hist, des inst. poL de Vancienne
France,— la 1891, pp. 8-9, and pp. 507-8, infra.
fJavle rom., As M. Jullian
has shown in a most interesting and suggestive article {Revue des etudes
anciennes, iii, 1901, pp. 77-97), the paqi were themselves '
natural divisions '.

' B. G., iv, 22, §§ 1, 5.

' B. G., i, 18, §§ 4-5 ii, 1, ; § 4 ; vi, 11, § 4; 13, §§ 1-2; 15 ; vii, 40, § 7. Cf.
V.
^
de Coulanges, Hist, des inst. pol. de Vancienne France, — la Gaule rom.,
])p. 37 8.
22 TNTRODUCTTON CHAr.

of the populace by eloquence and largess, and, overthrowing

the feeble oligarchy, usurped supreme power. Thus the
oligarchies lived in perpetual unrest if no one noble was :

conspicuously strong, there was intestine strife if one could ;

make himself supreme, the government was overthrown.

The populace were perhaps beginning to have some glimmer-
ing of their own latent strength ; but there is no evidence
that anywhere they had any definite political rights. The
Druids and the nobles or, as Caesar called them, the knights,
enjoyed a monopoly of power and consideration the bulk :

of the poorer freemen,ground down by taxation and strangled

with debt, had no choice but to become serfs.
And if in individual tribes there was anarchy, want of
unity was the bane of them all. It was not only that Belgian
and Aquitanian and Celtican were naturally distinct. This
distinction might have been as readily overcome as that
between English and Scotch and Welsh. But the evil was
more deeply seated. It is of course true that disunion is the
normal condition of half -civilized peoples. The Old English
tribes showed no genius for combination it was the strong :

hand of an Egbert, an Edgar, an Athelstan, that laid the

foundations of the English kingdom. Nor was the kingdom
united, except in the loosest sense, even on the eve of the
Norman Conquest. If Harold was formally king over all
England, his subjects felt themselves Yorkshiremen or men
of Kent rather than Englishmen. Moreover, the circum-
stances of the Gauls were peculiarly unfortunate. Their
patriotism, if it was latent, was real they were proud of :

what their fathers had achieved in war and the sense ;

of nationality was stirring in their hearts. Caesar himself

allows that some of the tribes were comparatively well
governed - and even clientship, which after all menaced our
;

own government until Henry the Seventh stamped it out,

had its noble side. Who does not respect the six hundred '

devoted followers of Adiatunnus,^ the four squires whom

'

neither fear nor favour could induce to betray Ambiorix,"^

>
C. Jullian, Hist, de la ilauh, ii, 82.
'
^
B. a., vi, 20. lb., iii, 22. ' Jb., vi, 43, § (i.
I INTRODUCTTON 23

and those attendants of Litaviccus who remembered that

'
Gallic custom brands it as shameful for retainers to
desert their lords even when
Assuredly the
all is lost '
? ^

Gauls had not, as Mommsen in his youth affirmed, reached '

their maximum of allotted culture.' - If they had been

unmolested or had been exposed to attack only from a single
enemy, it seems probable that, in the fullness of time, a
Vercingetorix would have welded them into a united nation.
But menaced as they were by the Germans on the one
hand and by the Romans on the other, their tendency to
disunion was increased. This much we may safely con-
clude, —
that the Gauls were not well fitted for developing
from their own resources a coherent polity. If the English-
man was provincial and unpatriotic, the Gaul was factious
and impracticable. Much glib generalization has been
hazarded regarding the hypothetical defects of the Celtic
character but only a very rash or a very discerning his-
;

torian would undertake to say how far the evil was due to
circumstances, how far to an inherited strain. Organism
and environment are for ever acting and reacting upon one
another. While, however, it is foolish to pass sweeping
judgements upon a people, of whom, except during the few
years that preceded the loss of their independence, we have
only the scantiest knowledge, it would be a great mistake
to leap to the conclusion that, in political capacity, one race
is as good as another. No one would deny that the Greeks
were endowed with a genius for art and literature which
their environment doubtless helped to develop and it may ;

be that the Celts were but poorly endowed with political

talent, and that circumstances had helped to stunt its
growth. The important fact is, explain it as we may, that
1 B. G., vii, 40, § 7.
''
Hist, of Rome, v, 1894, p. 28 {Rotn. Gesch., iii, 1889, p. 241). '
If,' says
Sir Henry Maine {Early Hist, of Institutions, 1875, pp. 54-5), '
the country
[Ireland] had been left to itself, one of the great Irish tribes would almost
certainly have conquered the rest. All the legal ideas which . . . come to us
from the existence of a strong central government lending its vigour to the
arm of justice would have made their way into the Brehon law and the ;

gap between the alleged civilization of England and the alleged barbarism of
Ireland during much of their history, which was in reality narrower than is
commonly supposed, would have almost wholly disappeared.'
2i INTOODUC^TION ( rLu\

the tribal rulers of Gaul had not achieved even that initial
step towards unity which the kings of Wessex, Mercia, and
Northumberland achieved when they swallowed up the
petty kingdoms of the heptarchic period. Or perhaps it
would be more true to say that, when the Romans first
established themselves on the west of the Alps, the Arvernian
king had achieved that step but that first his defeat on
;

the banks of the Rhone, and afterwards the revolution which

subverted the royal power, had broken the supremacy of his
house and dealt a fatal blow to the political development
of Gaul. There, as in Latium, the downfall of the monarch
inevitably weakened the power of the tribe and the oligar- ;

chies, if they had the power, were not granted the time to
work out their own salvation. It is true that able leaders
arose who attempted to follow the example of Bituitus.
Celtillus, the father of Vercingetorix, acquired a loose
supremacy over Gaul and Diviciacus, king of the Sues-
;

siones, who had already mastered the neighbouring Belgic

tribes, actually made himself overlord of South-Eastern
Britain. But from some unknown cause his successor failed
to retain his power and Celtillus fell a victim to his own
;

self-seeking. Individual tribes, such as the Aedui and the

Arverni, did indeed achieve some sort of supremacy over
their weaker neighbours and in certain cases two tribes,
;

for example the Senones and the Parish, formed one state.
There were leagues of the Belgae, the Aquitani, and the
maritime tribes. But supremacy had not hardened into
sovereignty ^ and the leagues were loose, occasional, and
;

uncertain. If some ]30werful baron, stimulated by ambition

or impressed by the evils of disunion, succeeded in clutching
the power of a Bretwalda, he was forthwith suspected by
his brother nobles of a design to revive the detested monarchy,
and was lucky if he escaped the stake. The country swarmed
with outlawed criminals, who had fled from justice, and
exiled adventurers, who had failed to execute coups d'etat.
Nobles and their clients lived sword in hand and hardly ;

^ Certain '
client ' tribes appear to liave paid tribute and rendered military
service. But hegemony was not iiriiily grasped, and client tribes transferred
tlieir allegiance from one overlord to another. »See pp. 517-1).
 1 INTRODUCTION 25

a year passed without some petty war. Every tribe, every

liamlet, nay every liousehold was riven by faction. One
was for tlie Romans and anotlier for tlie Helvetii one for :

the Aedui and another for the Arverni one for a Diviciacus :

and another for a Dumnorix one for the constitutional ;

oligarchy and another for the lawless adventurer. All, in

short, were for a party and none was for the state.^
;

One must not lightly acquiesce in the belief that the course
of history is always for the best but often it has happened ;

that a conquered people conserved what was most vital in

its own culture, and was quickened by the new. 'AttcoAo/uk^' '

«r,' said Themistocles, '

d //?/ a-noiX6\x€0a '
:
- like the English,
whom the Normans chastened, the Gauls needed the discipline
of foreign conquest.
Yet, besides the memory of their glorious past, which, as Unifying
i"""^"^^*^-
Caesar once remarked,^ both saddened the Gauls and spurred
them to desperate enterprises, there were certain influences
which tended to make every man feel that he and his fellows
belonged to one nation. That favoured land was formed by
Nature to tone down distinctions of race and to foster the
growth of nationality. If the French are the most united
of all peoples, they owe this fortune to their country, whose
unifying tendency has ever been the same. France, says
Vidal de la Blache,'* who of all geographers knows best how
to make between motherland and
his readers feel the tie
—
people, France is a country whose regions are naturally
connected, and whose inhabitants learned early to mingle
with and to know one another. No country of equal extent
comprises such diversities but they pass off into each
;

other by insensible gradations. There is ', says this writer, '

'
a beneficent force —
a genius loci which has guided our —
—
national life, an indefinable power which, without obliter-
ating varieties, has blended them in a harmonious whole.'
The wayfarer who roams from the sand-hills of the Channel
'
See various notes in Part II, Section IV.
-'We should liave l>een undone if we had not been undone/ Phitarch,
Themidodes, 29.
' n. G., V, r>4, § 5.

E. Lavisse, Hul. de la France, t. i, I (by P. Vidal de la Blache), pp. 49, 51-2

*

Bull, de (jeoyr. hist, el descr., 1902,

pp. 1 19, 124.
 26 TNTRODUCTTON ciiAr.

to the mountains of Auvergne, from the uplands of the

Morvan to the plam of the Berri, conveTsing with peasant
and townsman m turn, who is touched by the spirit of
prehistoric life wafted from the rude stone monuments of
Brittany and by the spirit of imperial Rome which broods
over the mediaeval glories of Bourges and over that ancient
town which is being revealed by the excavator on Mont
—
Auxois who feels how one influenced the other and both
—
survive in our Mechanical Age will comprehend what the
geographer means and for him the tale which Caesar told
;

will become real.

Religion. And England before the Norman Conquest,
in Gaul, as in
there was another influence which tended to make every
man feel that he and his fellows belonged to one nation,
community of religious ideas, controlled by one ecclesiastical
organization.^ Local cults of course abounded but the :

great gods whom Caesar noticed, however variously they

may have been conceived by various tribes, were common
to Gaul while every rite and every sacrifice was recognized
;

and regulated by Druidism.

The religion of the Gauls, like every other, was rooted in
—
animism, ^that habit of mind, common to savages and
children, which peoples the universe with spirits, which
ascribes to sun, moon, and stars, to earth and sea, to fire and
—
water and everything that moves even to many things that
—
are inert the life and will which they feel within themselves.
It was not merely the religion, influenced perhaps by contact
with the Greeks of Massilia, of the Celtic conquerors of Gaul :

much of it was Ligurian, Iberian, even aboriginal. Solar

symbols, as we have seen, pervaded the decoration of the
Bronze Age and to this day omelettes are offered to the
;

sun on the bridge in the village of Andrieux in Dauphiny,

when, on the 10th of February, he reappears, after four
months of gloom, above the mountains that imprison the
valley .2 The spirits of springs, of lakes, of rivers, of moun-

1 In regard to the religion of the Gauls in general see Rice Holmes, Anc.
Britain, pp. 271-80 0. Jullian, Hist, de la Gaule, ii, 113-81
; Trans.
; Third. . .

Internat. Congress Hist, of Religions, ii, 1908, pp. 218-20, 222; and S. Reinaoh,
. . .

Orpheus, 1909, pp. 162-83. - U Anthr., xix, 1908, p. 701.

 —
I INTRODUCTION 27

tains, and of woods were worshipped in the Iron Age

as they had been for millenniums before, and as they con-
tinued to be after what was called Christianity had become
the official creed. The Marne and the Dives were called
after the goddesses Matrona and Deva Epone in the ;

department of the Seine-et-Oise marks the cult of Epona ;

Bourbon and La Bourboule are derived from one of the spirits

which suggested Caesar's description of the Celtic Apollo,
Borvo, or the boiling one ', the god of hot springs. The
'

eponymous deities of towns were worshipped long before

the first foundations were laid the goddess Bibracte was
:

the spirit of a well reverenced by the peasants of the moun-

tain upon which the Aeduan capital was built .^ And perhaps
we may suppose that the heads of his slain enemies, which
the Celtic warrior fastened upon the walls of his cottage,
were dedicated to his household gods or to the spirits of his
ancestors.
was descended from the
Celtic religion, in so far as it
religion of the undivided Aryan stock, was fundamentally
one with the religions of Italy and Greece but our imper- ;

fect knowledge of the classical religions hardly helps us more

to understand the inwardness of Celtic religion than the
remark of Caesar, that about their deities their notions are '

much the same as those of other peoples '.^ What is certain

is that, like every other polytheistic religion, that of the Celts,
except perhaps where was moulded by Druidical doctrine,
it

had no definite theology, but was an ever-expanding, ever-

shifting, formless chaos, alike in its main developments in
Britain, Gaul, and Spain, yet differing in every tribe and
household, and in every age ^ that, on its practical side, it
;

was a performance of traditional rites and that it concerned

;

the individual most as a member of a family, a community,

or a tribe. Like all other polytheists, too, the Celts were
ready to believe in gods who were not theirs in the reign of :

Tiberius the boatmen of Paris set up an altar on which, side

by side with their own Esus and Tarvos Trigaranus, were

1 Rev.
celt., i, 1870-2, pp. 306-19. ^ ^ q^ ^^^ 17^ § 2.
See Rev. des etudes anc, vi, 1904, p. 329, and
'^
of. .Sir A. Lyall, Asiatic Studies,
1899, pp. 2-3, ().
28 TNTRODUOTION f-HAP.

and Vulcan. Perhaps the C'elts, like the

figured Jupiter
Romans,^ gave more thought to the ritual by which their
gods might be persuaded to grant tliem their hearts' desire
than to the gods themselves. Doubtless to the Celt, as to the
Roman, however little his religion may have fostered nobility
of life, dread of the mysterious was a salutary discipline.-
But what we want to apprehend is this, —wherein the spirit
of Celtic religion differed from that of the religion of ancient
Latium, of Greece, of the Semitic tribes and we may only ;

hope to attain a distant and hazy view.

Every student of Caesar knows the passage which
enumerates the national deities, whom alone he deemed
worthy of mention :
—
The god whom they most reverence
'

is Mercury, whose images abound. He is regarded as the

inventor of all arts and the pioneer and guide of travellers ;

and he is believed to be all-powerful in promoting commerce

and the acquisition of wealth. Next to him they reverence
Apollo, Mars, Jupiter, and Minerva. Their notions about
these deities are much the same as those of other peoples :

Apollo they regard as the dispeller of disease, Minerva as

the originator of industries and handicrafts, Jupiter as the
^
suzerain of the celestials, and Mars as the lord of war.'
Now one of the most sagacious of Celtic scholars has related
that when the musician, Felicien David, was invited by the
Viceroy of Egypt to instruct his wives, etiquette compelled
him to give the lessons to a eunuch, who passed them on as
best he could. Caesar, he remarks, was in the position of
the eunuch.'* No show how the Gauls thought
Gallic writings
of their own gods. Inscriptions and altars supply names of
deities which are names and nothing more, or bewilder us
by coupling with the name of a Roman god the names or
epithets of divers Celtic gods. Anonymous statues are
attributed to various deities by various archaeologists,
though some of them may not be deities at all. Inscriptions,
altars, and statues aUke belong to the period of the Roman
Empire, when the introduction of Roman gods and goddesses

>
W. Warde Fowler, The Raman Festivals, 1899, p. 333.
- Ih., p. 347. -'
B. (/., vi, 17, §§ 1-2.
^ H. (jiaidoz, Etudes de niytJiol. gaul., — Lt dieu guul. du sohil, 1880, p. 91.
I INTRODUCTION 29

liad thrown the Celtic pantheon into welinigh inextricable

confusion. And
merely to ascertain the Celtic names of the
five great gods would profit little unless the names were
significant of their attributes. It does not greatly matter

whether Teutates was Mercury, and Taranis Jupiter, and

Esus Mars, or whether, as some have persuaded themselves,
the famous whose honour Lucan says that dreadful
trio, in

rites were performed, were insignificant objects of local

worship.^ We must be content to know that there was a
Celtic Mars, a Jupiter, and a Mercury for we cannot learn ;

much that is worth knowing besides wliat Caesar has told.

Oiiginally, we may be sure, the great gods were simply the
elemental forces of Nature, as in his time they were still
among the Germans but gradualty they had come to be
;

personified.
Why was the god whom Caesar equated with Mercury
honoured above all others by the Gauls ? Some centuries
earlier, when the Celts were the most warlike
*
of all peoples \^
the war-god had been the most conspicuous figure in their
Olympus and
subsequent degradation is regarded,
; his
perhaps justly, as an indication of the progress which they
had made meantime in the arts of peace. ^ One religious
custom, however, of which Caesar witnessed examples, proves
that Mars, however inferior he may have been to Mercury,
had still many fervent worshippers in Gaul. When tribes-
men had made a successful raid, they used to sacrifice to
Mars the cattle which they had captured the rest of their ;

booty they erected in piles on consecrated ground. It

rarely happened that any one dared to keep back part of the
spoil and the wretch who defrauded the god was punished,
;

like Achan, by a terrible death."^

Borvo, the healing god, was certainly an equivalent of
Apollo ^ but in inscriptions Apollo's name is found also
;

—
with other epithets, Maponus, the ever youthful,' Grannus, '

1 Ittv. celt., xviii, 1897, pp. 140-1 ; Rev. arch., 4« ser., xi, 1908, p. 152. Cf.
Rev. des etudes anc, v, 1903, pp. 217-9, and C. Jullian, Hid. de la Gaule, ii,

2
118, n. 2. jj (J ^ 54^ § 5
^ Mercury was also reverenced more than any other god by the Germans
of whom Tacitus wrote {Germ. 19). * B. 6'., vi, 17,
§§ 3-5.
" See VAnthr., xx, 1909, p. 196.
30 INTRODUCTION chap.

'
the brilliant,' and Belenus, '
the shining one,' in other
.words, the sun. Again, among the gods who in imperial
times were assimilated to Jupiter, besides Taranis or Tanarus,
*
the Thunderer,' was one whose statues depict him as a man
bearing upon his shoulder or grasping a wheel and, since ;

the wheel was a solar symbol, it would seem that Jupiter

and Apollo were alike assimilated to the sun.
One must not, indeed, imagine that the great gods had
always been distinct, or even that in Caesar's time their
physiognomies, except perhaps in Druidical theology, were
sharply outlined. In polytheism attributes of deity tend to
become separate deities, and some of the Celtic epithets
which are attached to Minerva, Mars, and the rest may
mean that they were assimilated by this or that tribe to
topical divinities.^ Dis Pater, from whom the Gauls believed
themselves to be descended, was certainly near of kin to
Saturn ^ and Dis Pater and Teutates, the god of the
;
*

people,' who was perhaps primarily conceived as a kind of

Saturn, may once have been one. Again, if Teutates in
Britain remained Mars, while in Gaul the Romanized Celts
seem to have identified him here with Mars and there with
Mercury, one is tempted to conjecture that he may have
been the common ancestor of both.^
The worship of animals, to those who have not felt the
fascination of anthropology, appears merely unintelHgible
and absurd. Animals were worshipped because they were
formidable or wonderful or possessed powers that man had
not because men fancied that they were incarnations of
;

deity because they might be tenanted by the souls of fore-

;

fathers * and animal- worship, or a decaying remnant of

;

animal- worship, has left vestiges in Celtic The boar art.

was especially sacred. Like the Romans, the Gauls had
military standards, which were invested with sanctity : like
the Romans also, they carried not a flag but the figure of an
1 Rev. de8 etudes anc, iv, 1902, p. 221.
- lb., vi, 1904, pp. 1 11, n. 1, 134, n. 4 ; A. Holder, AU-celtischer Sprachschatz,
ii, 1805-C.
Eev. des etudes anc, iv, 1902, pp. 110-4.
•''

8ee E. B. Tylor, Prim. Culture, ii, 1903, pp. 229-34, and Mr.
' Edward
Clodd's admirable little book, Animism, 1905, pp. 75-8.
 —
I INTRODUCTION 31

animal, and with animal was commonly the boar.^

them this
Traces of animal- Avorship indeed survive in Gallic tribal
names the Brannovices were the people of the raven, the
:

'
Eburones the people of the boar. Sucellus, the hammerer '

— the local epithet of a god Avhose statue, discovered at

Sarrebourg, is attributed to Dis Pater was clad in a wolf- —
skin, and perhaps was originally thought of as a wolf. With
Borvo was associated the bull, probably because the activity
of the medicinal springs suggested the idea of a powerful
animal and, as Salomon Reinach has acutely seen,^ the
;

name of the goddess Epona the fountain of the mare finds — —

its counterpart in Hippocrene, the spring of Mount Helicon,

from which the horse of Apollo sprang. A relic of animal-

Avorship is also discernible in the horned head of Cernunnos,
a god who is figured on one of the altars of Paris, and
in Tarvos Trigaranus
—
the bull with the three cranes '
'

which fills the back of another.^ Originally perhaps Epona

was the spirit of some gushing spring, who came to be
symbolized as a swift steed at last, as her worship was ;

diffused, she was represented in statues by a woman riding

on a mare. And
would be figured with horns
so deities
because the horn was a symbol of strength.
But votive altars, statues, and temples, although they
enshrined older beliefs, belong, as we have seen, to the
period when the Celts had fallen under the dominion of
Rome. The Cisalpine Gauls, if Livy ^ and Poly bins ^ are
to be believed, worshipped in temples ; but, like the oldest
sanctuary of Jupiter on the Alban Mount, the holy places
of the Western Celts were groves.^ Caesar indeed says that
the Gallic Mercury was represented by numerous simulacra :

but if these were statues, it is inexplicable that none of them

has ever come to light ; and perhaps we may accept the

1 B. G., vii, 2, § 2 ; 88, § 4 ; S. Reinach, lUperloirt de la statuaire grecquc

ct rom., ii, 746-7 H. d'Arbois de Jubainville, La
; civilisation des Celtes, 1899,
pp. 390-1 ; Rev. des etudes anc, vi, 1904, p. 48.
2 Eev. arch., 4« ser., ii, 1903, p. 349.
^ Corpus inscr. Lat., xiii, 3026 b, c. Cf. Rice Holmes, Anc. Britain, p. 284, n. 5.
* xxii, 57, § 10 ; xxiii, 24, § 11. ^ ii, 32, § 6.
B. G., vi, 13, § 10 ; 17, § 5 ; Tacitus, Ann., xiv, 30 ; Dion Cassius, Ixii,
7, § 3. Cf. Rice Holmes, Anc, Britain, p. 285, n. 3.
 32 INTRODUCTION chap.

suggestion tliat Caesar was thinking of menhirs, whicli had

been erected long before the first Celt set foot in Gaul/ but
which, like the formless stones that the Greeks venerated a.';

figures of Hermes,"^ Avere, he supposed, regarded as possessed

by the spirit of the great national deity. On the menhir ol
Kernuz in Finistere a rude Mercury was sculptured in Roman
times. The conjecture may be well founded that the Druids,
"^

like the priests of Israel, condemned anthropomoq)hisni ;

-

but it is not needed to explain the lack of native statues of

Celtic gods. The Romans, according to Yarro, had for many

years no sacred images :
'^
like the Celts, like the Germans,
who also, even in the time of Tacitus,^ deemed it derogatory;
to the majesty of the gods to ascribe to them human form,
they were content to recognize manifestations of divine will
and even when their temples were adorned Avitli the creations
of Greek art, their ancient Vesta remained wrapt in awful
mystery. But, while the Druids may have been as hostile
as Israel to Gentile abominations, the Celts in general
were as receptive as the Romans, and, after they fell undei
Roman welcomed the services of foreign sculptors
influence,
The \ The question of the origin and affinities of Dniidism haf;
'

\i given rise to superabundant speculation, which has led tc

no certain result. Caesar was informed that the system was
believed to have been imported from Britain. At all events,
there is no evidence that it was known to the Celts oi
Cisalpine Gaul and the Germans, with whom the Celts were
;

long in contact and to whom they were ethnically akin, had

no Druids. If, then, Druidism originated in Gaul, it probabl}'
did not appear until after the Celtic invasion of Italy and ;

yet in Britain, the civilization of which was more backward

than that of Gaul, it was apparently of long standing. The
^ Rice Holmes, Anc. Britain, i). 285, n. 6; Rtv. celt., xiii, 1892, pp. 190-3.
M. Jullian {Bev. de-s etudes anc, iv, 190'2, pp. 284, n. G, 285, n. 1), refeiTing t€
the passage in which Lucan (iii, 412-3) describes the Druids' grove neai
Massilia, simulacraque niaesta deorum Arte carent cae^iisque cxdant inform la
irtincis, argues that Caesar's simulacra ne peut signiKer que des objets ayant
'

deja vaguement T aspect de forme humaine '. Cf. M. Jullian's Hist, de la Gaide.
ii, 153, 11. C, 154.
- Pausaiiias, vii, 22, § 4. ^ Btv. celt., xiii, 1892. p. 191.
* lb., p. 199. Cf. J. Dechelette, ManueUV ((rcheologie, ii, 409.
^ Augu.stine, De civ. dei, iv, 31. * Cierm., 9.
 I INTRODUCTION 33

prevalent opinion is that the Celtic conquerors of both Gaul

and Britain found it existing, and that it was strong enough
to secure terms, and finally to make itself supreme. Caesar's
words are sometimes explained in the sense that in his time
Druidism was more vigorous in Britain than in Gaul, and
that Gallic Druids travelled to Britain in order to be initiated
into its mysteries but no sufficient reason can be given for
:

not accepting them in their literal sense and if they are ;

true, the Celtic invaders of Britain did not create Druidism,

and, neolithic or aiot, it was at all events pre-Celtic.^ But
all that we know
about the Gallic branch of this
for certain
strange hierarchy we learn from the brief notices of Caesar
and other ancient writers and Caesar has told us all that
;

Avas essential for the subject of his narrative. The Druids

formed a corporation, admission to which was eagerly
sought they jealously guarded the secrecy of their lore
:
;

and full membership was only obtainable after a long

novitiate. They were ruled by a pope, who held office for
life and sometimes the succession to this dignity was dis-
;

puted by force of arms. They were exempt from ordinary

taxation ^ and from service in war. The ignorance and
superstition of the populace, their own organization and sub-
mission to one head, gave them a tremendous power. The
education of the aristocracy was in their hands. The doctrine
which they most strenuously inculcated, if Caesar was
not misinformed, was that of the transmigration of souls.
*
This belief,' he said, they regard as a powerful incentive
'

to valour, because it inspires a contempt for death.' ^ They

claimed the right of deciding questions of peace and war.
Among the Aedui, if not among other peoples, at all events
in certain circumstances, they exercised the right of appoint-
ing the chief magistrate. They laid hands on criminals, and,
in their default, even on the innocent, imprisoned them in
monstrous idols of wicker-work, and burned them alive as
'
See pp. 523-0.
M. Jullian {HisLde la Gaule, ii, 55 and n, 1), who translates neque (tributa)
^

una (cum reliquis pendunt [B. G., vi, 14, § 1]) by non dans le menie role que ',
'

and refers to H. Meusel, Lex. Caes. ii, 2365-7, holds th at the Druids did pay taxes,
,

but separately and at a lower rate than the laity. Cf. C. E. C. Schneider's
edition of B. G.> vol. ii, p. 230. '
B. G., vi, 14, § 5.

1093 D
34 INTRODUCTION chap.

a sacrifice to the gods. They immolated captives in order

to discover the divine will in the flow of their blood or their
palpitating entrails they lent their ministrations to men
;
^

prostrated by sickness or going forth to battle, who trusted

that heaven would spare their lives if human victims were
offered in their stead and one form of sacrifice which they
;

appear to have countenanced the slaughter of a child at —

the foundation of a monument, a fortress, or a bridge has —
left traces in European folklore and been practised in Africa,
Asia, and Polynesia in modern times. ^ They practically
monopolized both the civil and the criminal jurisdiction ;

and if this jurisdiction was irregular, if they had no legal

power of enforcing their judgements, they were none the
less obeyed. Every year they met to dispense justice in the
great plain above which now soar the spires of Chartres
cathedral.^ Those who disobeyed their decrees were excom-
municated and excommunication meant exclusion from
;

the civil community as well as from communion in religious

rites. Nor did their power extend only to individuals con- :

tumacious states were laid under an interdict, a warning —

that they had forfeited the protection of their gods.
Did the Druids owe their doctrine of immortality, as
Diodoms ^ and Timagenes ^ impty, to the influence of Pytha-
goras ? The testimony of these writers has been contemp-
tuously rejected but it seems not improbable that Druidism
:

may have absorbed tenets of Pythagorean origin through

the medium of the Greeks of Massilia ^ and this conjecture ;

gains some support from numismatic evidence. Coins have

been found both in Gaul and Britain, bearing on the reverse
side the flgure, formed by five interlacing lines, which if
known as the pentagram and was a well-known Pythagorear

^ Diodorus Siculus, v, 31, § 3 ; Tacitus, An)L, xiv, 30.

' Chronica minora, ed. Th. Mommsen, iii, 1898, p. 182, 11. 14-7. Cf. Sir A
Lyall, Asiatic Studies, ii, 1899, pp. 312-3, E. B. Tylor, Prim. Culture, i. 1903
pp. 104-8, and Rev. celt., xxvi, 1905, p. 289.
^ The disputes of humble litigants were probably settled by their lords ant ;

only suitors who could afford the expense of the journey would appeal to th(
Druids. See pp. 528-9.
* V, 28, § 6. ^ Ammianus Marcelhnus, xv, 9, § 8.
' H. Gaidoz, Esquisse de la rel. des Gaulois, 1879, p. 18.
1 INTRODUCTION 35

symbol.^ would seem, however, that if metempsychosis

It

was really a Druidical doctrine, it had no firm hold upon

the Celts in general and their sepulchral customs were not
;

consistent with it. They believed that there was an Elysium

somewhere in the west, Avhere they were to live again,
feasting, carousing, and duelling, a life like that which they
had lived before, but free from care.^ If the Druids, as
Caesar said, taught that souls passed from one body to '

another ', they meant perhaps that after death the soul
—
entered a new body, the ethereal counterpart of that which
it had left behind.^ Immortality was an idea, more or less
vague, common to many peoples for the Celts the Druids :

made it an article of faith. But what that theory was which,

as Caesar says, they inculcated in regard to the heavenly '

bodies and their motions, the size of the universe and of the
earth, the origin of all things,'^ it is useless to inquire. We
only know that as they traced the descent of the Gauls
to Dis Pater, they regarded night as older than day, and
reckoned time by nights and that in common with all the
;

peoples of antiquity, they computed years by the revolu-

tions of the moon.^
The which stand out as certain
facts of historical import
are these. Like the Brahmans, who, so long as their authority
is acknowledged, recognize the Protean manifestations of

Hindu religious fancy, the Druids kept control over the

manifold forms of aboriginal and Celtic worship. Being
a sacerdotal caste, not, like the priests of Rome, popularly
elected, but self -constituted and self-contained, they were
of course opposed to all innovation. It has been said that

J. Evans, Coins of the Anc. Britons, 1864, p. 98

^ A. Blanchet, Traite des ;

monn. gaul, pp. 331, 360, 378, 385-6. Perhaps Prof. E. B. Tylor was too
positive when he called the pentagram an interesting proof of tradition from
'

the Pj^thagoreans ' {Ericy. Brit., xv, 1883, p. 203). Cf. Globus, xcv, 1909,
pp. 7-9.
^ Rev. de Vhist. des religions, xiv, 1886, p. 61 ; G. Dottin, La rel. des C cites,
1904, pp. 35-7 ; Rev. celt., xxix, 1908, p. 103.
M. Salomon Reinach [Orpheus, p. 178) believes that the Druids did once
'

teach metempsychosis, which afterwards meant simply that souls migrated

to a region in the west. Cf. Rice Holmes, Anc. Britain, p. 295, n. 1.
* B. G., xi, 14, § 6.
' 76., 18, § 2, Pliny, Nat. Hist., xvi, 43 (95), § 250.

D 2
 36 INTRODUCTION chap.

ancient writers regarded as peculiar to tlie Druids beliefs and

practices which were common them and other priests of
to
antiquity ; but while every ancient people had its priests,
the Druids alone were a veritable clergy.^ Celtic religion,
in so far as it had the same ancestry Rome, would
as that of
easily harmonize with it ; but Druidism, with its more
definite theology, might be expected to counteract this
tendency, and would therefore be a danger to Roman
dominion." And it was Druidism that formed one of the
strongest ties between Gauls and Britons and between the
w^arring tribes of Gaul.
But though might perhaps foster the idea, it
religion
Invasions could not supply the instant need of political union. Over
^^^® y3ifii wooded plains of Germany fierce hordes were roam-
Cimbri
and ing, looking with hungry eyes towards the rich prize that
lay beyond the Rhine. Moreover, the danger of Gaul was
the danger of Italy. The invader who had been attracted
by the pleasant land of France would soon look south-
' '

ward over the cornfields, the vineyards, and the olive-gardens

of Lombardy. When Caesar was entermg public life, men
who were not yet old could remember the terror which had
been inspired by the Cimbri and Teutoni, those fair-haired —
giants who had come down, like an avalanche, from the
unknown lands that bordered on the northern sea. They
descended into the valley of the Danube. They overthrew a
113 b. c. Roman consul in Carinthia crossed the Rhine and threaded
;

the passes of the Jura and overran the whole of Celtican

;

109 B.C. Gaul. Four years after their first victory they defeated
another consul in the Province. Then they vanished but :

four years later they reappeared and two more armies were
;

105 B.C. destroyed on the banks of the Rhone. The panic-stricken

Italians dreaded another Allia : but, while Italy lay at their
mercy, the Cimbri turned aside and when, after three years'
;

wandering in Spain and Gaul, they rejoined the Teutoni, and

the tw^o swarms headed for the south, Marius was waiting for
them on the Rhone, and his brother consul in Cisalpine Gaul.
Once more the host divided ; and while the Teutoni encoun-
^ See Rice Holmes, Anc. Britain, p. 298, n. 4.
' See H. d' A. de Jubainville, Les Druides, 1906, pp. 60, 64.
I INTRODUCTION 37

tered Marius in the neighbourhood of Aix, the Cimbri threaded

the Brenner Pass, and descended the valley of the Adige.
The ghastly appellation of the Putrid Plain commemorated 102 b. c.

the slaughter of the Teutoni : the Cimbri were annihilated

at Vercellae, near the confluence of the Sesia and the Po.^ 101 b. c.

But if this danger had been averted, the movements of

the other German peoples might well cause anxiety. Press-
ing resolutely onward, they fought their way through the
outlying Celtic territory, up to the right bank of the Upper
Rhine. Some years before the conspiracy of Catiline ^ an 71 b.c.

opportunity was afforded them of making good their footing

in the heart of Gaul. A bitter enmity had for many years
existed between the Aedui and the Arverni, each of whom
were overlords of a group of tribes. The Aedui were the
stronger and they enjoyed the countenance of Rome. The
;

Arverni, in conjunction with the Sequani, hired the aid of Invasion

a German chieftain, Ariovistus, who crossed the Rhine with vistur
fifteen thousand men. They were enchanted with the
country, its abundance, and its comparative civilization ;

and fresh swarms were attracted by the good news. After

a long struggle the Aedui were decisively beaten, and had to
cede territory and give hostages to the Sequani, who
^

apparently usurped the hegemony which had been exercised

by the Arverni. One of the leading Aeduans, the famous
Druid, Diviciacus, went to Rome and implored the Senate
for help. His aim was not merely to get rid of Ariovistus
and to free his country from the yoke of the Sequani, but
also to regain his own influence, which had been eclipsed by
that of his younger brother, Dumnorix. He was treated
with marked distinction, made the acquaintance of Caesar,
and discussed religion and philosophy with Cicero * but ;

the Senate did not see their way to interfere on his behalf. 61 b.c.

All that they did was to pass a vague decree that whoever
might at any time be Governor of Gaul should, as far as

See pp. 548-9.

1 2
^^^ pp 553_4.
M. JuUian {Hist, de la Gaule, iii, 155) conjectures that the cession was of
^

territory on the left bank of the Saone which had been a subject of dispute
between the two peoples. See pp. 352-8, infra.
* Cicero, De Div.,i, -L\, §90.
 38 INTRODUCTION chap.

might be consistent with duty to the republic, make it

his
his business to protect the Aedui and the other allies of the
Roman people.^ Meanwhile the Sequani had found that
their ally was their master. He was not going to return to
the wilds of Germany when he could get a fertile territory
for the asking. He
compelled the Sequani to cede to him
the fertile plain of Alsace. At length they and their Gallic
allies, including, as it should seem, even the Aedui, mustered

all their forces and made a desperate

throw off the
effort to
60 B.C. yoke: but they sustained a crushing defeat; and their
conqueror was evidently determined to found a German
kingdom in Gaul.

Revolt of Meanwhile the Allobroges, who had never yet fairly

the Alio- their dependent condition, had risen in revolt.
accepted
^
broges. . .

Qi B c They were still embittered by defeat when the Roman

C)0 B. V. agents in the Province were alarmed by the appearance of
bands of marauders on the right bank of the Rhone. They
had been sent by the Helvetii, a warlike Celtic people, who
Threat- dwelt in that part of Switzerland which lies between the
^^^i^^^j ^1^^ Jura, the lake of Geneva, and the Upper Rhone.
vasion^of
the Hel- The Romans had already felt the weight of their arms. A
generation before, the Tigurini, one of the four Helvetian
tribes, had thrown with the Cimbri. They had
in their lot
107 B.C. spread desolation along the valley of the Rhone, defeated
a consular army, and compelled the survivors to pass under
the yoke. Now, in their turn, they were hard pressed b}^
the Germans they had reason to fear that the victorious
;

host of Ariovistus would sever them from their Celtic kins-

men ; and they had formed the resolution of abandoning
their country and seeking a new home in the fertile land
of Gaul.
The author of the movement was Orgetorix, the head of
the Helvetian baronage. His story throws a vivid light
upon the condition of the Gallic tribes. He persuaded his
brother nobles that they would be able to win the mastery
over Gaul, and undertook a diplomatic mission to the

^ I agree with Long {Decline of the Boman Republic, iii, 477) that the senatorial
decree was aimed against Ariovistus for there is no evidence that the Helvetii
;

entered Gaul before GO b. c.

I INTRODUCTION 39

leading Transalpine states. Two chiefs were ready to listen

to him, Casticus, whose father had been the last King of the
Sequani, and Dumnorix, brother of Diviciacus, who was
at that time the most powerful chieftain of the Aedui.^ If
Diviciacus saw the salvation of his country in dependence
upon Rome, his brother regarded the connexion with
abhorrence. He was able, ambitious, and rich and the ;

common people adored him. Orgetorix urged him and

Casticus to seize the royal power in their respective states,
as he intended to do in his, and promised them armed
support. The three entered into a formal compact for the
conquest and partition of Gaul and, if they had any aim
;

beyond their own aggrandizement, they may have hoped

that their success would not only checkmate Ariovistus, but
stop the anarchy which paralysed their country and avert
the encroachments of Rome.^ Their purpose threatened the
republic with a twofold danger.. Once they had gone,
the lands which they left vacant would be overrun by the
Germans, who would then be in dangerous proximity to
Italy and there was no telling what mischief they might
;

do in Gaul. Above the din of party strife at Rome the note

of warning was heard. ^ Men talked anxiously of the
prospects of war and the Senate sent commissioners to
;

dissuade the Gallic peoples from joining the invaders. But

the ambitious triumvirate had still to reckon with the
Helvetii. They heard that their envoy had broken his
trust, and immediately recalled him to answer for his con-
duct. He knew that if he were found guilty, he would be
burned alive and accordingly, when he appeared before
;

his judges, he was followed by his retainers and slaves,

numbering over ten thousand men. The magistrates,
determined to bring him to justice, called the militia to
arms but in the meantime the adventurer died, perhaps
;

by his own hand.

Thus the scheme of the triumvirate was destroyed and ;

probably Cicero was thinking of the death of Orgetorix when

he wrote to Atticus that tranquillity was restored in Gaul.
' See pp. 555-G. » Cf. C. Jullian, Hist, de la Gaule, iii, 162.
' Cicero, Alt., i, 19, § 2.
 40 INTRODUCTION chap.

Nevertheless the consul, Metellus, favoured a forward policy :

*
the only fault I have to find with him,' said Cicero, is that '

he is not much pleased with the news that Gaul is calm.' ^

But the idea which Orgetorix had conceived did not die.
The Helvetii had no intention of abandoning their enter-
prise, nor Dumnorix of abandoning his. He had married
a daughter of Orgetorix and he was quite ready to help
;

them if they would make it worth his while. They resolved

to spend two years in preparing for their emigration ;

bought up wagons and draught cattle and laid in large ;

supplies of corn. Diplomacy was powerless to shake the

purpose of a brave and desperate nation. Perhaps the
Senate failed to realize the gravity of the crisis. Perhaps
they shrank from putting the sword into the hands of the
man who might ultimately turn it against themselves.
Consul- But the hesitation of an effete Senate was soon to give
Caesar. way to the energy of a leader of men. One of the consuls
for the year 59 was Julius Caesar. About the time of the
election Ariovistus, who had already paid court to Caesar's
predecessor, Metellus,- made overtures for an alliance with
Rome and doubtless with the object of securing his
;

neutrality in view of the threatened Helvetian invasion, the

Senate conferred upon him the title of Friend of the Roman
People. They had already half promised to protect their
Gallic allies. They now practically guaranteed to the con->
How he queror of those allies the security of his conquest. And in

to provide
"^^^i^ latter policy Caesar, if we may believe his own word,
against fully concurred. He must have seen the impending troubles,
vetian But he was not yet free to encounter them and he doubtless ;

danger,
approved of any expedient for keeping the barbarian chief
inactive until he could go forth in person to confront him.
He is That time was at hand. In the year of his consulship Caesar
Governor ^^^ made Governor of Illyricum, or Dalmatia, and of Gaul,
of Gaul, that is to say of Gallia Cisalpina, or Piedmont and the Plain
of Lombardy, and of Gallia Braccata, or, as it was usually
called, the Province. If Suetonius ^ was rightly informed,

1 Att., 20, § 5.
2 Pliny, Nat. Hist., ii, 67, § 170. Cf. Hermes, xlii. 1907, pp. 509-10.
^ Divus lulins, 22,
I INTRODUCTION 41

his commission gave him tht^ right to inckide Gallia Comata

— the land of the long-haired Gauls
* that is to say the '
—
whole of independent Gaul north of the Province, within
his sphere of action.^ He had already gained distinction in
Spain both as a general and as an administrator but hitherto ;

he had had no chance of showing the full measure of his

powers. He was at this time forty -three years old.^ In person
he was and slight, but well-knit
tall and, if he was as ;

licentious as the mass of his contemporaries, his constitution,

fortified by abstemious habits, was capable of sustaining
prodigious efforts. His broad dome-like skull his calm and ;

penetrating eyes ; his aquiline nose massive yet finely

; his
moulded jaw, expressed, like no other human countenance, a
rich and harmonious nature, intellect, passion, will moving —
in accord. And, if his vices were common, his generosity,
his forbearance, his equanimity, his magnanimity were his
own. He believed, with an unwavering faith, that above
himself there was a power, without whose aid the strongest
judgement, the most diligent calculation might fail. That
power was Fortune and Caesar was assured that Fortune
;

was ever on his side.^ But it would be impertinent to this

^ It has
been objected {Athenaeum, Jan. 13, 1900, p. 42) that in another '

passage {Gram., c. 3) Suetonius applies the expression "Gallia Comata" to

a portion [only] of Transalpine Gaul'. Suetonius (ed. C. L. Roth, p. 289,
1. 23) there says that Munatius Plancus, when he was Governor of Gallia
'

Comata, founded Lugdunum' {JShinatius Plancus, cum Galliam regeret Comatam,

Luijdunum condidit), which surely does not prove that the Province could
properly be called Gallia Comata. Still Suetonius may have used the expres-
sion incorrectly.
See pp. 556-7.
' Cicero, AtL, x, 8 b ; Caesar, B. G., v, 58, § 6 ; vi, 30, § 4; 35, § 2 42, §§ 1-2
; ;

vii, 89, § 2 ; B. C, iii, 10, § 6 ; 68, § 1 ; 95, § 1, &c. In the Classical Review
of April 1903, pp. 153-6, Mr. did me Warde Fowler the honour of devoting
an statement which I made in the first edition
article to a criticism of the
of this book as to Caesar's belief in Fortune. In the present edition I have
allowed the statement to stand, merely substituting the words an unwavering '

faith for the faith of a devotee ', which was certainly open to criticism
'
'
;

and I believe that it is supported by the evidence to which I have referred.

I cannot help thinking that Mr. Warde Fowler misunderstood me. I really
was not so credulous as to suggest that Caesar believed in a personal goddess
called Fortuna. What I meant to convey was simply that, in common with
many men of action, he believed that fortune may wreck the most wisely
planned and most diligently executed operations, while she may rescue her
favourites from the penalties of their own mistakes and that, in common ;
 42 INTRODUCTION chap.

narrative to attempt to analyse the character — ^to which our

greatest poet has done less than justice — of the greatest man
of action who has ever lived. Whatever quality was lacking,
the want in no wise affected his fitness for the task which
he had now to perform.
His army. His appointment carried with it the command of an army
consisting of four legions, perhaps about twenty thousand
8th^9th^'
and 10th.] men.^ One of them was quartered in Transalpine Gaul the :

other three were at Aquikia, near the site of the modern

Trieste. He could also command the services of slingers from
the Balearic isles, of archers from Numidia and Crete, and of
cavalry from Spain but, as the narrative will show, he
;

raised the bulk of his cavalry year by year in Gaul itseP.

The number of the auxiliary infantry was perhaps generally
about one-tenth of that of the regulars ^ the number of the :

cavalry varied greatly, but four hundred for each legion

was near the average.^ Various military reforms had been
introduced by Marius ; and the legions of Caesar were, in
many respects, from those which had fought against
different
Hannibal. They were no longer a militia, but an army of
professional soldiers. Each legion consisted of ten cohorts ;

and the cohort, formed of three maniples or six centuries,^

had replaced the maniple as the tactical unit of the legion.
From the earliest times the legion had been commanded by
an officer called a military tribune. Six were assigned to
each legion and each one of the number held command
;

in turn. But they now^ often owed their appomtments to

interest rather than to merit and no tribune in Caesar's ;

army was ever placed at the head of a legion. They still

had administrative duties to perform, and exercised subor-
dinate commands. But the principal officers were the legati,
who might loosely be called generals of division. Their powers

with Sulla, Napoleon, and other great commanders, he had a firm faith, touched
perhaps by mysticism, in his own star.
I
See pp. 559-63. ''
Klio, vii, 1907, p. 318. Cf. B. C, iii, 4.
'
Cf. B. G., i, 15, § 1 with iv, 12, § 1 ; V, 8, § 2 ; 9, § 1 ; vi. 32. § 2.
* The writer of the article Cohors in the Diet. c?e.s ant. grecqiies et rom. (i, 1287-8)
of MM. Daremberg and Saglio believes that the reason for this change was that
the individual soldiers were generally inferior to the old burghers, and
that the only way of givingthem confidence was by forming them into
qomparatively large bodies.

I
I INTRODUCTION 43

were not but varied according to circum-

strictly defined,
stances and to the confidence which they deserved. A legatus
might be entrusted with the command of a legion or of an
army corps he might even, in the absence of his chief, be
;

entrusted with the command of the entire army. But he

was not yet, as such, the permanent commander of a legion.
The officers upon whom the efficiency of the troops mainly
depended were the centurions. They were chosen from the
ranks ; and their position has been roughly compared with
that of our own non-commissioned officers. But their duties
were, in some respects, at least as responsible as those of a
captain : the centurions of the first cohort were regularly
summoned to councils of war ; and the chief centurion of
a legion was actually in a position to offer respectful sugges-
tions to the legate himself.^ Every legion included in its

ranks a number of skilled artisans, called /aftr^, who have been

likened to the engineers in a modern army but they were not ;

permanently enrolled in a separate corps. ^ They fought in

the ranks like other soldiers ; but when their special services
were required, they were directed by staff-officers called
praefectifahrum. It was their duty to execute repairs of
every kind, to superintend the construction of permanent
camps, and to plan fortifications and bridges and it should ;

seem that they also had charge of the artillery,^ the ballistae —
and catapults, which hurled heavy stones and shot arrows
against the defences and the defenders of a besieged town.
The legionary wore a sleeveless woollen shirt, a leathern
tunic protected across breast and back by bands of metal,
strips of cloth wound round the thighs and legs, hob-nailed
shoes, and, in cold or wet weather, a kind of blanket or
military cloak. His defensive armour consisted of helmet,
shield, and greaves his weapons were a short, two-edged,
:

cut-and-thrust sword and a javelin, the blade of which,

behind the hardened point, was made of soft iron, so that,
when it struck home, it might bend and not be available for
return. These, however, formed only a part of the load which
he carried on the march. Over his left shoulder he bore
1 See B. 0., iii, 5, § 2. 2 g^e p. 579.
' See Long's Decline of the Boman Republic, ii, 19,
 44 INTRODUCTION chap.

a pole, to which was fastened m a bundle his ration of grain,

hiscooking vessel, saw, basket, hatchet, and spade. For it
was necessary that he should be a woodman and navvy as
well as a soldier. No Roman army ever halted for the night
without constructing a camp fortified with trench, rampart,
and palisade.
The column was of course accompanied by a host of non-
combatants. Each legion required at least five or six hundred
horses and mules to carry its baggage ^ and the drivers, ;

with the slaves who waited on the officers, formed a numerous

body. Among the camp-followers were also dealers who
supplied the wants of the army, and were ready to buy booty
of every kind.^
Hisinten- What Caesar intended to follow, he has not
line of policy
tions.
told us. While he was going forth to govern a distant land,
the government of his own was lapsing into anarchy. He
must have seen that the Germans would soon overrun Gaul
unless the Romans prevented them and that the presence
;

of the Germans would revive the peril from which Marius

had delivered Rome. We may feel sure that he had deter-
mined to teach them, by a rough lesson if necessary, that
they must advance no further into Gaul, nor venture to
cross the boundaries of the Province or of Italy. Confident
in himself and supported by his fellow triumvirs, he was
prepared to act without waiting for senatorial sanction and ;

it can hardly be doubted that he dreamed of adding a new

province to the empire, which should round off its frontier

See p. 585.
1

Caesar nowhere mentions that he used wagons or carts during the GalHc
-

war, though it seems certain that he must have used some, to carry artillery
and material for mantlets and the like. See Bell. A Jr., 9 B. 6'., iii, 42, § 3 ; ;

and Daremberg and Saglio, Diet, des ant. grecques et rom., i, 929. The larger
pieces of artillery were of course not conveyed entire, but in parts, which were
put together as occasion required (Sir R. Payne-Gallwey, The. Crossbow, &c.,
1903, p. 257).
^ W. Smith, Did. of Gk. and Roman Ant., i, 340; ii, 588-9; F. Frohlich.
Das Kriegswesen Cdsars, 1891, pp. 56-7, 62-4, 66-7, 75 ; Daremberg and Saglio,
Diet, des ant. grecques et rom., ii, 1005-6 iv, 482-3
; ; Frontinus. Strat., iv, 1,

§ 7. See also vario\is notes in Section VI. There is no evidence that there
was any medical staff in Caesar's army or under the Republic at all, though
itmay perhaps be inferred from a passage in Suetonius {Divus Augustus, 11)
that wealthy officers were attended by their private surgeons.
I INTRODUCTION 45

and add to its wealth. But whether he had definitely

resolved to attempt a conquest of such magnitude, or merely
intended to follow, as they appeared, the indications of
Fortune, it would be idle to conjecture. Ambitious though
he was, he only courted, he never tempted her. The greatest
statesman is, in a sense, an opportunist. When Caesar
should find, himself in Gaul, he would know best how to
shape his ends.
 CHAPTER II

CAMPAIGNS AGAINST THE HELVETII AND

ARIOVISTUS

58 B.C. About the middle of March a startling announcement

Caesar reached Caesar. The Helvetii had actually begun to move ;

the^He/-^^
and their hordes Avould soon be streaming over the Roman
vetiiare Province. Three neighbouring tribes, the Rauraci, the
march Tulingi, and the Latobrigi, and also the Boi, who had long
ago migrated into Germany, had been induced to join them ;
*IJ^°p^^^
vince. they had laid in sufficient flour to last for three months and, ;

to stimulate their resolution and enterprise, they had deli-

berately cut themselves off from all prospect of return by
burning their homes. On the 24th ^ of that very month the
whole vast multitude, numbering, according to their own
muster-rolls, three hundred and sixty-eight thousand,^ was
to assemble opposite Geneva, ready to cross the Rhone.
He hastens Cacsar instantly left Rome, and, hurrying northward
^ ninety miles a day,^ crossed the Alps, took command of the
j^j^(j^^^^^

stroys the Provincial legion, ordered a fresh levy, and reached Geneva

at the end of a week. He immediately destroyed the bridge

Helvetian by whicli the Hclvetii intended to cross the river. They
envoys ask
^^^^ ambassadors to sav that they only wanted to use the
his leave *;
.

to use road through the Province, and would promise to do no

throudi mischief. Would Caesar give them permission ? Caesar had
the Pro- of coursc no intention of granting their request but, as he ;

vmce.
wanted to gain time for his levies to assemble, he told the
ambassadors that he would think over what they had said,
and give them an answer on the 9th of the following month.

'
March 28 of the unreformed calendar. ^ See pp. 237-41.
' Plutarch, Caesar, 17. See also B. G., i, 7, § 1 ; Suetonius, Divtis hdius,
57 ; and the map of diaul.
^ F. Eyssenhardt {Ncue Jahrbilchcr fur Philologie und Paedagogik, Ixxxv,
18G2, p. 760) accepts Dion Cassius's statement (xxxviii, 31, § 4) that Caesar
 THE HELVETII AND ARIOVISTUS 47

He made good use of the interval. The legion was with him ;
58 B.C.

and the Provincial levies arrived in time to jom in executing

the design which he had formed. The road by which the
Helvetii desired to march led through Savoy and the river ;

was at certain pomts fordable. It should seem that they had

not yet had time to assemble in force. Along the southern
bank of the Rhone, between the lake and the Pas de FEcluse
— a distance of about seventeen miles —
Caesar threw up He pro-
mises to
lines of earthworks in the few places where the banks were reply in a
not so steep as to form a natural fortification.^ The soldiers fortnight,
and mean-
were posted in redoubts behind the works. When the while for-
tifies the
ambassadors returned, Caesar plainly told them that he left bank
would not allow the Helvetii to pass through the Province. of the
Rhdne.
Undeterred by this rebuff, the emigrants made several
attempts to force the passage of the river. Some of them He pro-
vents the
waded others made bridges of boats, and tried to storm the Helvetii
;

ramparts but, rapidly concentrating, the soldiers pelted from

:

crossing.
them with missiles and sent them stagg;3ring back.
Only one route now remained, the road that winded —
along the right bank of the Rhone, beneath the rocky steeps
of the Jura, through the Pas de I'Ecluse. The emigrants
might, it would seem, have made their way into Gaul by the
route that leads to Pontarlier or one of the other passes in
the Jura but either because they shrank from encountering
;

Ariovistus or for some other reason, of which Caesar took no

held out to the Helvetian envoys the hope that he would allow them to pass
through the Province. Otherwise, he insists, it is impossible to explain why
the Helvetii waited for the day which Caesar had appointed. Caesar neither
says nor implies that he did not hold out such a hope to the envoys. On his
own showing, indeed, he intended to deceive them. I suspect, however, that
this one of Dion's embellishments, because I believe that Caesar would
is

have kept the fact to himself instead of blurting it out to any of the excellent '

authorities' whom Dion is assumed to have followed (see pp. 215-7). But
Dion may have hit upon the truth. Caesar would of course have held out such
a hope to the Helvetii, if it had been worth his while to do so. As a nation,'
'

writes Lord Wolseley {Soldier's Pocket-hook, 5th ed., 1886, p. 169), we are '

bred up to feel it a disgrace even to succeed by falsehood ... we will keep

hammering along with the conviction that " honesty is the best policy " and
that truth always wins in the long run. Tiiese pretty little sentences do well
for a child's copy-book, but the man who acts upon them in war had better
sheathe his SAvord for ever.' Again, the general can, by spreading false news
'

among the gentlemen of the press, use them as a medium by which to deceive
an enemy '
{ib., 4th ed., p. 337). '
See pp. 614-5.
 48 CAMPAIGNS AGAINST THE chap.
58 B. c.

The account, these routes were out of the question.^ The road
aHow^'" that led through the Pas de I'Ecluse was so narrow that
them to there was barely room for a single wagon to move along it

througli ^'t ^ time beyond the : pass, it led into the territory of the
tl^ Pas de
Sequani and if they
; offered the slightest opposition, it

would be hopeless to attempt to get through. They refused

at first to grant a safe-conduct but Dumnorix, at the request ;

of the Helvetii, willingly acted as mediator. He had estab-

lished his influence with the Sequani by wholesale bribery
and, after a little negotiation, he succeeded in procuring for
his friends the favourwhich they sought. The Helvetian
leaders undertook to restrain their people from plundering ;

and hostages were exchanged for the fulfilment of the com-

pact. The ultimate object of the emigrants was to settle in
western Gaul, in the fertile basin of the Charente. Thence
they would be able to make raids upon the open corn-growing
districts of the Province and their mere presence, with
;

Dumnorix in the background, would be a standing menace

to Roman interests in Gaul. But first they would have to
make their way along the valley of the Rhone, across the
plain of Amber ieu, and over the plateau of Dombes to the
Caesar Saone. Caesar calculated that while their huge unwieldy
goes back
column was crawling along the muddy tracks, lie would
alpiue have time to raise a new army, strong enough to cope with
turns with
^^^^^- Leaving his ablest lieutenant, Labienus, to guard
reinforce- the lines on the Rhone, he hastened back to Cisalpine Gaul :

^ on his own responsibility

encamps enrolled two new legions with- ;

above the (Jrew tlic other three from their winter-quarters and ;
confluence i , i t i i i ^^ n i i i ^ 1

of the marched back by the road leadmg along tlie valley oi the
Sld^Jie
Dora Riparia and over Mont Genevre. The mountain tribes,
Saone. who doubtlcss hopcd to plunder his baggage-tram, attempted
[The to stop his advance but agam and again he dashed them
;

Graioceh,
^side until, descending into the valley of the Durance, he
Ceutroncs.
and Catu-
ii?.ii
puslicd ou tlirough the highlands ot Dauphiny, past i5rian9on,
tt-vi- t-»-

nges.]
Embrun, and Gap,^ crossed the Isere and the Rhone, and
'
See pp. 613-4.
* The 11th and 12th. We may suppose that, as ^L Jullian says {Hist, de la
Gallic, iii, 202), Caesar had given orders in advance for the assemblage of the

recruits.
'
Between Brian9on (Brigantio) and the Rhone the itinerary is not abso-
II HELVETJl AND ARIOVISTUS 49

encamped on the heiglits of SathonaVj near the pomt where 58 b.c.

the rushing current is swelled by the tranquil stream of the j^^^^^ , i

Saone.
He was only just in time. The bulk of the Helvetii had The Aedui
crossed the Saone and descended, like a swarm of locusts, hisai^
upon the cornfields and homesteads of the Aedui. Envoys against
the Hel-
came to beg Caesar to remember the loyalty of their country- vetii.

men, and help them to get rid of the invaders. His arrival
had wrought a change in Aeduan politics. Liscus, who
belonged to the Romanizing party, had been elected Vergo-
bret and Diviciacus had regained his prestige. Liscus
;

concluded an agreement with Caesar, and undertook to feed

his army. Labienus with his legion had already rejoined
him. The rearguard of the Helvetii, numbering about a
fourth of the entire host, were gathered on the eastern side
of the river, in the valley of the Formans, eleven miles to
the north. ^ Caesar left his camp soon after midnight,
marched quietly up the valley of the Saone over ground He defeats
which masked his approach, and launched his legions upon perses the
the unsuspecting multitude, as they were crowding into their rearguard
boats. Those who escaped the slaughter vanished in the Helvetii.
surrounding forests. They and their slain kinsfolk belonged
to the tribe called the Tigurini,^ by which, fifty years before,
a Roman army, under the consul Lucius Cassius, had been
defeated and compelled to pass under the yoke.
Within twenty-four hours Caesar had thrown a bridge of His pas-
^
boats * over the river, and transported his entire army to the l^^^^
right bank. The Helvetii, who had taken three weeks over Saone.
the passage, were greatly alarmed, and sent an embassy to ry,^
meet him. The principal envoy was an aged chief named Helvetii
Divico, who, in his youth, had commanded the army which negotmte!^
defeated Cassius. He said that his countrymen were willing but reject
to settle wherever Caesar pleased, if he would only leave them terms.
unmolested. But if he was bent upon war, they were ready ;

and he would do well to remember that they had already

lutely certain ; but Caesar must have gone either by tiio route indicated in
the text or by the valley of the Romanche and Grenoble. See Carte de France
(1 : 200,000), Sheet 60, and pp. (515-0.
' See Napoleon, Hi.'<t. de Jules Cesar, ii, 57, n. 2.
'-
See pp. (317-D. See p. 38.
•'
* See p. 0J2.
iai>3
^
 50 CAMPAIGNS AGAINST THE chap.

58 B. c. defeated a Roman army.

Caesar replied that he remembered
the treacherous exploit of which they boasted, and remem-
bered with indignation. Besides, even if he were inclined
it

to let bygones be bygones, he could not overlook the outrages

of which they had just been guilty. Still, he was ready to

make peace with them, upon certain conditions. They must

compensate the Aedui and the Allobroges for the damage
which they had done, and give hostages for their future good
behaviour. Divico haughtily replied that the Helvetii, as
the Romans had the best of reasons to know, were accus-
tomed to receive hostages, not to give them.
They Next day the emigrants broke up their encampment. To
north- reach the valley of the Charente, it was necessary to cross
^t5^' , the Loire. The Roanne.
direct line intersected that river near
followed
by Caesar. But the rugged country between the basins of the Saone and
the Loire was, in this direction, impassable and beyond ;

Roanne the mountains of Le Forez barred the way. The

only course was to move up the valley between the Saone
and the hills of Beaujolais until a practicable route could be
found. Caesar sent on his cavalry, four thousand strong,
to watch the enemy's movements. They were composed of
levies from the Province and from the Aedui and the ;

Aeduan contmgent was commanded by Dumnorix. Caesar,

if he was already acquainted with his character and aims,

probably did not think it expedient to challenge his appoint-

ment and indeed he may have welcomed the opportunity
;

of keeping him under his eye.^ The cavalry ventured too

near the Helvetian rearguard, and lost a few men in a
skirmish. As the victors were only five hundred, Caesar
must have suspected that something was amiss. For a
fortnight the two armies continued to advance, northward
and then north-westward, never more than five miles apart.
The Helvetii turned off from the Saone near Macon, and
moved up the valley of the Petit Grosne,^ along the Ime of
the road which leads from Macon to Autun; near Mont St.
Vincent they struck westward in the direction of Toulon-

'
See Class. Quartedy, iii, 1909, pp. 213-14, and C. Jullian, Hist, de la Gaule,
iii, 205-6.
- See pp. 620-1.
II HELVETII AND ARIOVTSTUS 51

sur-Arroux.^ Their vast column must have extended at 58 B. c.

least fifteen miles in The advanced guard was

lengtli.^

followed by the train of wagons, which was protected in

the rear by the Boi and Tulingi,^ and last of all came the ;

Helvetian fighting men.^ Elated by their recent success,

the Helvetii occasionally faced about and challenged their
pursuers but Caesar would not allow his men to be drawn
;

into a combat. He was looking for a favourable opportunity

to fight a decisive battle but for the time he had enougii
;

to do in trying to prevent the enemy from plundering his

allies. Nor was this his only anxiety. He depended upon
the Aedui for his supplies but day followed day, and no
;

supplies came. On the Saone indeed he had a flotilla of Caesar

pressed
barges laden with corn ; but the necessity of following the for sup-
Helvetii him far away from that river. The Aeduan
had led plies,
owing to
notables in his camp promised, protested, and poured forth the in-
trigues of
excuses, till he lost all patience and accused them of
Dumnorix.
deliberate breach of faith. This challenge elicited a full
disclosure. Liscus spoke on behalf of his brother chiefs. It
appeared that there were certain individuals whose power
was actually greater than that of the Government. They
had exerted their influence over the people to prevent them
from sending supplies, telling them that if the Romans
succeeded in defeating the Helvetii, they would use their
victory to enslave the Aedui as well as the other tribes.
Liscus concluded by he had revealed the
telling Caesar that
truth at the risk of his life, and had only spoken under
compulsion. Caesar had no doubt that by certain indi- '

viduals he meant Dumnorix.

'
But he had no intention of
discussing matters of state in the presence of men whose
discretion could not be trusted. He therefore told all the
chiefs, except Liscus, that they might go. Liscus then
spoke out frankly. He admitted that Dumnorix and no
other was the man. He had amassed great wealth, and had
See pp. 620-1, 624-7.
'

See Stoffel, Hist, de Jules Cesar,

^ —
Guerre civile, ii, 1887, p. 451, and my
essay on The Credibility of Caesar's Narrative (pp. 237-41).
'
'

See pp. 629-30.

''

^ May we suppose tliat some of the fighting men marched parallel with the
wagons? See Lord Wolseley, The Soldier's Pocket-book, 1886, p. 408.

& 2
 52 CAMPAIGNS AGAINST THE chap.

58 B. c. spent ifc lavishly in buying popular support. He had acquired

great influence witli the Bituriges and other tribes by
arranging marriages between the women of his family and
powerful chieftains. Not only was he politically connected
with the Helvetii, but he privately detested Caesar, because
Caesar had set him aside and restored his brother Diviciacus
to power. In his own country he was the leader of the anti-
Roman faction. The interests of the Helvetii were his
interests. they succeeded, they would help him to mount
If

the throne if they failed, he would be worse off than before.

;

He had kept them regularly supplied with information ;

and in the cavalry skirmish, a few days before, he had set

the example of flight.

Caesar hardly knew how to act. Dumnorix was evidently

one of the most powerful and implacable enemies whom he
had to fear. He could not afford to overlook such flagrant
hostility but he Avas afraid of offending Diviciacus, whom
;

he particularly desired to conciliate. Perhaps the astute

Druid had no great love for his brother but it would not ;

do to exasperate the whole patriotic party. Caesar sum-

moned him to his tent, and, addressing him through the
medium of Gains Valerius Troucillus, a distinguished Provin-
cial, and trusted friend, earnestly
his principal interpreter
pressed himto consent to hispunishing Dumnorix. Diviciacus,
with a burst of tears, begged him not to be too hard upon
his brother or it would be said that it was he who had
;

advised the infliction of the punishment, and public opinion

would brand him as a monster. Caesar pressed his hand
kindly, and bade him dismiss his fears. His regard for him,
he said, was so great that he was willing to condone the
insult which had been offered to his Government and the
provocation which he had himself received. The truth was
that he had no choice. He had not yet won the prestige
that would only come from victory and with powerful
;

enemies before him, and doubtful allies around him, upon

whose goodwill he depended for the means of subsistence,
it would be folly to raise a hornet's nest about his ears.

He contented himself therefore with sending for Dumnorix,

and giving him a severe rebuke and a stern warning. This
 DEP'EAT OF THE IIEIA ETir/ Sku jiagfji etS-e'lT
.(Ml rCa SI //c!

REFERENCE
S. . . . Suimnil of hill of Armpcy
C . . . .Enlrojichjiient lor protection of iin^J^RgV
RR . . .4 legions in-line of tattle Col£irc\ers rrLodifu:aJt.ione
HH . . .Helvelii orStoffeVs theory, -which are
H'H . . .Hrtvetii forced to retreat to n liUl od/}pt£ctbythe author undbv
TT . . .Belli)* Tiilliuli MJuUian are nhowK by recL
.

Ujieti. Seepaaes SZ6'6Z7.

r r . . .Romiui 7,'.^ line facintf Boii & Tullng;
hh . . . Helve tii renemng allacii
._..«_ Kuiiixu] line of march
^^_^^_^_ EIe.lvetian. „ „ ,,

The contoarB demote inlervBls in altitude of 10 m.etres.

Srnie 1 .'iti.OUO
Kilometres
11 HELVETII AND ARIOVISTUS 53

once, he said, for his brother's sake, his conduct should be 58b.c.
overlooked. At the same time he gave secret orders that
Dumnorix should be watched, and his movements reported.
Next morning Caesar made an attempt to surprise the Hia
enemy, which only failed through the stupidity of an officer, attempt to
They had encamped, his patrols reported, at the foot of a hill surprise

eight miles distant, probably Sanvigne, about six miles east yetii.

of the river Arroux.^ He at once sent a party to reconnoitre

the and ascertain whether it would be possible to ascend
hill,

it from the rear. They reported that such an ascent was

easily practicable. In the middle of the night Caesar sent
Labienus with two legions, under the guidance of the ex-
ploring party, to climb the hill and swoop down upon the
enemy's rear, while he should himself attack them in front,
About two hours after the departure of Labienus, he sent
forward his cavalry, and followed along the track by which
the enemy had advanced. Publius Considius, a centurion
of experience and reputation, was sent on ahead with patrols
to reconnoitre. Shortly before sunrise Caesar was within
a mile and a half of the enemy, who suspected nothing.
Suddenly Considius rode back at a gallop and told him that
all had gone wrong not Labienus, but the enemy occupied
:

the height he had recognized them by their arms and

;

crests, and was sure that he had made no mistake. Caesar

at once led his troops on to another hill close by, and formed
them in line of battle. Labienus meanwhile was wondering
why he did not come and when it was too late, Caesar
;

learned that Considius had been the dupe of his own fears.
The legions moved on in the afternoon, and encamped About
about three miles in the rear of the Helvetii, near the site ^"^^® ^^ '

of Toulon-sur-Arroux.2 The day after, as no corn-carts had He

appeared and only two days' rations were left, Caesar struck J|J^^°^^®^
off to the right, and marched for Bibracte, the capital of Bibracte,

the Aedui, about sixteen miles to the north, where he knew supplies.
that he would find granaries stored with corn. The route ran
along the watershed between the Arroux and one of its
affluents, a rivulet called the Auzon. The Helvetii were far
on their way, the head of the column having passed Luzy
' Stoffel, Guerre civile, ii, 445. ^ Seo pp. G25-7.
 54 CAMPAIGNS AGAINST THE chap.

58 B.C. and turned westward down the valley of the Alene, wlien
some deserters from Caesar's cavalry brought them the news.
Fancying that he was afraid of them, or hoping to prevent
him from reaching Bibracte, they turned likewise, marched
back rapidly, and attacked his rearguard near the hill of
Armecy, about three miles north of Toulon. Caesar sent
his cavalry to retard their advance, while he ordered the
infantry to retrace their steps and ascend the slopes of
Armecy. The whole movement must have occupied about
two hours. Half-way up the hill, the four veteran legions
Avere ranged in three lines of cohorts, each line being eight
men deep.^ The soldiers' packs were collected on the top,
under the protection of the auxiliaries and the two newly-
raised legions, who were ordered to entrench the position.
The baggage-train may either have been parked on the
ridge along which it was moving or have continued its march
towards Bibracte. It was exposed to no danger from the
Helvetii and a slender escort might have sufficed to protect
;

it against the Aeduans who favoured Dumnorix.^ The oppor-

tunity for Avhich Caesar had been waiting had at last come.
Although the enemy were now between him and Bibracte,
the hill of Armecy was the best position which he could have
chosen. If he won, the road would of course be open. If he
lost,— but he did not intend to lose. It was his first pitched
. battle ;and he knew that for him and his army defeat would
be destruction. The Helvetii would fight desperately his :

legions, except perhaps the 10th, had not yet come to know
him and he could not fully trust all his officers. He there-
;

fore dismounted and made his staff do the same, so that the
men might see that their officers shared their dangers and ;

then, in accordance with custom, he harangued the expectant

troops. The wagons of the Helvetii were parked, as they
came up, on rising ground to the left of the road and about ;

one o'clock in the afternoon the whole mighty host, con-

gregated in compact masses, flung back Caesar's horsemen
and with shields closely locked pressed up the hill against
the Roman line. The men in the front rank held their
shields before their bodies, while those behind bore theirs
1 See p. 588. » See p. 028.
 ^

11 HELVETII AND ARI0VISTU8 55

horizontally above their heads. ^ The legionaries in the front 58 b.c.

ranks stood with their javelins in their hands, ready to

throw. On the plateau above, recruits and auxiliaries were
hard at work with their entrenching tools. When the enemy
were within a few yards, the centurions gave the word.
Down flew a shower of javelins and the mass began to ;

break. The blades of the javelins, composed of soft iron,

had bent as the points penetrated the shields. Sword in
hand, the cohorts of the first line charged : many of the
Helvetii, finding their shields nailed together by the javelins,
which, pull and wrench as they might, were not to be torn
out, flungthem away, and parried the thrusts as best they
could but they were soon overborne, and fell back to a hill
;

about a mile south-west of Armecy on the further side of

the road by which they had marched. One cannot but
suspect that their first attack had been a feint, designed to
entice Caesar from his strong position ; for the Romans
were following when the Boi and Tulingi, who had just
arrived upon the field, rushed upon their right flank and
rear. The Helvetii took heart and returned to the attack ;

and, while the first two lines of the Romans closed with them,
the third faced about, and confronted their fresh assailants.
Long and fiercely the battle was fought out. In due time Defeat of
the men of the rear ranks reheved those in front, advancing
^etii near
between the files as the latter withdrew, and again the men Bibracte.

who had been relieved, relieved in turn their comrades ;

while the cohorts of the second line reinforced those of

the first, and the groups thus formed an unbroken whole.
Gradually the Helvetii were forced further up the hill while ;

the Boi and Tulingi retreated to their baggage. Standing

behind the wall of wagons, they hurled down stones and
darts upon the advancing Romans, and thrust at them
between the wheels with pikes when they attempted to
storm the laager. Even the women and children took part
in the defence.^ The struggle was prolonged far into the

^ See W. Smith, Did. of Greek and Roman Ant., ii, 808 ; and Stoffel, Guerre
de Cesar et d' Arioviste, 1890, p. 69.
* These details are conjectural. The reasons are given on pp. 588-98.
* Plutarch, Caesar, 18.
 56 CAMPAIGNS AGAINST THE CHAr.

58 B.C. night and meanwhile the Helvetii, covered by the resistance

;

of their alHes, were making good their retreat.^ At length

the legionaries burst through the barrier. Women and
children who could not escape were slaughtered and the ;

flying remnant of the invading host disappeared in the

darkness of night.
Before the sun went down, evil tidings must have reached
the non-combatants who were still wending their way
towards the field. It is certain that many of the wagons
never came into the laager.-^ What despair fell upon the
baffled emigrants how the jaded cattle were headed round
;

again towards the north, and goaded through that night ;

how those who escaped the slaughter tramped after, and told
the tale of the calamity the din, the confusion, the long
;

weariness of the retreat, —these things it is easy to imagine,

but those only who have shared the rout and ruin of a beaten
army can adequately realize.
Caesar's Caesar was unable to pursue. His troops were exhausted,
treatment ]^jg cavalry were untrustworthy and he had to give the ;

fugitives, wounded time to recover, and to bury the teeming corpses

that might have engendered a pestilence among his allies :

but he sent mounted messengers to warn the Lingones,

through whose country the fugitives would have to pass, to
give them no help. Tlie Lingones occupied the country
round Tonnerre and Bar-sur-Aube as well as tlie plateau of
Langres and the neighbourhood of Dijon. At the end of
three days Caesar started in pursuit. On tlie way he was
met by envoys, whom the Helvetii, now reduced to utter
destitution, had sent to arrange terms of surrender. He
bade them tell their countrymen to
and await his halt,
arrival. When he overtook them, he ordered them to give
hostages, and to surrender their arms and a number of slaves,
who had escaped to them. Six thousand Helvetians slipped
away in the night, and took the road towards the Rhine :

'
Cf. Col. H. l^iicher. Bibrude, 1904, p. 20.
^ If the estimate (see p. 46) of the number of the emigrants was
official
correct, and unless a considerable proportion had dispersed on the march,
over 100,000, as Colonel Stoffel calculates, must have perished in the battle- See
pp. 222-5. All questions relating to the campaign are discus.sed on pp. OJO-34.
'
Sec p. 239.
 ^

II HELVETII AND ARIOVISTUS 67

but Caesar sent peremptory orders to the inhabitants to 58 b. c.

hunt them down and bring them back and on their return, ;

they were treated as enemies '. The Boi were allowed, at

'
Settle-

the request of the Aedui, who appreciated their martial [^e^Boi.

qualities, to settle in Aeduan territory. It would seem that
the tract assigned to them was in the neighbourhood of
St. Parize-le-Chatel, between the and the Loire.
Allier
The Helvetii and the other tribes, who would be most useful
as a barrier between the Germans and the Province, were sent
back to their own land and the Allobroges were directed
;

to supply them with grain.

The news of this brilliant victory produced its natural Envoys
effect. The success of the Helvetii would have been a ^jcan Gaul
calamity to all, except Dumnorix and his following and ;
congratu-

this calamity Caesar had averted. He appeared as the Caesar,

solicit
conqueror, not of Gaul but of the invaders of Gaul ;
^^^
the oligarchies who dreaded ambitious upstarts welcomed against
Ariovistue.
his support and the Aedui might expect that their old
;

pre-eminence would be restored. At the worst, his rule

would be preferable to the tyranny of Ariovistus and he ;

would doubtless be glad to aid in expelling his rival. The

patriots in the tribal councils, if they offered any opposition,
were outvoted. Chieftains came from all parts of central
Gaul to congratulate the conqueror. They told him that
they had certain important proposals to lay before him ;

and, with his express sanction, they then and there convoked
a council to arrange details. The meeting took place some
days later. After the council had broken up, Caesar con-
sented, at the pressing request of the chiefs, to give them
a private interview. They earnestly begged him to keep
what they were going to say a close secret ; for if it were
to get abroad, they would be made to suffer cruelly. Divi-
ciacus, who spoke for them, related how Ariovistus had
established his footing in the land of the Sequani, defeated
the Aedui and their dependants, and finally overthrown the
combined and their respec-
forces of the Aedui, the Sequani,
tive allies.^ At that moment there were a hundred and
twenty thousand Germans in their midst and within a few ;

^ See pp. 425-30. ^ See pp. 554-5.

/ ^f 8T. NIICVHALL'S
 58 CAMPAIGNS AGAINST THE chap.

58 B.C. years the Gauls Avould be expelled from their own country.
The Sequani had already been forced to cede a third part
of their territory, and would soon be forced to give up another
third ; for a fresh horde, the Harudes, numbering four and
twenty thousand, had recently crossed the Rhine. Ariovistus
was a cruel bloodthirsty tyrant and, if Caesar would not
;

help them, they must all go forth, like the Helvetii, and seek
some new home.
Failure Caesar assured the chiefs that they might rely upon his
attempts ^^ppo^^^- Their interests indeed coincided with his. He
to nego- saw that it was absolutely necessary to stop the flow of
Ariovistus. German Like the Cimbri and Teutoni, these fierce
invasion.
hordes might, if they were not checked, soon overrun the
whole of Gaul, and thence pour into Italy. Moreover, the
interest as well as the honour of Rome required that she
should protect her allies and the Aedui were allies of long
;

standing, whose fidelity had been rewarded by the title of

'
Brethren ' . And there was another reason why Caesar
should interfere. Like Clive, when he found himself con-
fronted by Dupleix, he could not stand still. He must
cither advance or retreat. If he were to shrink from espous-
ing the cause of the Gauls, he w^ould lose the credit which his
victory had won, and perhaps force them to make common
cause with Ariovistus against him. If he boldly confronted
their oppressor, he himself would become the arbiter of
Gaul. Peaceful methods, however, might be tried first.
The Roman army was comparatively weak. Ario^^stus was
master of a formidable host and it would be foolhardy to
;

attack him without absolute need. He had been treated

with distinction by the Senate and there was just a chance
;

that he might listen to reason. Besides, it would be im-

politic for theproconsul to levy war against the King and
Friend upon whom those titles had been conferred with his
sanction, without prehminary diplomacy, which he must so
conduct as to justify liimseK before his countrymen. Ariovis-
tus was then probably in the plain of Upper Alsace and
;

Caesar sent ambassadors to ask him to name some inter-

mediate spot for a conference. He told them to say that if
their master wanted anything from him, he must take
11 HELVETII AND ARIOVISTUS 59

the trouble to come to him in person. He could not risk 58b.c

his safetyby moving outside his own territory without his
army and to move and feed his army would involve an
;

amount of exertion which he did not care to undergo. Mean-

while he should like to know what business Caesar had
in a country which the Germans had won by their own
swords.
Caesar now assumed a more peremptory tone. Ariovistus
had rejected his invitation. Very good Then these were !

his terms. Not another man must set foot across the Rhine :

the hostages of the Aedui must be restored and Ariovistus ;

must positively cease to molest that people or their allies.

If he obeyed, Caesar would be his friend. If not, he should

know how to avenge the wrongs of the Aedui. The Senate

had decreed, three years before, that the Governor of Gaul
for the time being should protect the Aedui and the other
allies of the Republic and he intended to obey his in-
;

structions.
Ariovistus haughtily replied that he was a conqueror ;

and, as a conqueror, he had a right to treat his subjects as

he pleased. The Romans invariably acted on the same
principle. He did not interfere with them what right, then, :

had they to interfere with him ? He would not molest the

Aedui so long as they paid their tribute but most certainly :

he would not give up the hostages and if the Aedui did ;

not pay, much good would their alHance with the Romans
do them For Caesar's threats he cared nothing. No man
!

had ever withstood Ariovistus and escaped destruction. Let

Caesar choose his own time for fighting. He would soon
find out what mettle there was in the unbeaten warriors of
Germany.
With message came the alarming news that a host
this
of Suebihad appeared on the eastern bank of the Rhine, and
that the Harudes were actually harrying the lands of the
Aedui. Caesar, the most reticent of writers, has told us that
he was seriously alarmed .^ The Gauls were waiting to see
whether he or Ariovistus was to be master. If he suffered
any reverse, they would probably rise in his rear and ;

1 B. Q., i, 37, § 4.
 60 CAMPAIGNS AGAINST THE CHAP.

58 b. c. between them and the Germans his army might perish.

Not a moment was to be lost if the formidable Suebi were
to be prevented from reinforcing the army of Ariovistus.
With all possible speed Caesar made arrangements with the
Aedui, the Sequani, and the Lingones for the forwarding
of supplies, and immediately put his army in motion. Three
He days later he heard that Ariovistus was hurrying to seize
against Vesontio, now Besan9on, the chief town of tlie Sequani,
Ariovistus a strong place well stored with all munitions of w^ar. March
and seizes . ., ..
Vesontio. mg night and day at his utmost speed to anticipate him, he
.,
i'i
encamped on the outskirts, and threw a strong garrison into
the town.
Vesontio, which now became Caesar's base, was an ideal
Gallic stronghold. The town stood on a sloping peninsula,
round which the Doubs swept in a curve that nearly formed
a circle while the isthmus, little more than five hundred
;

yards wide, rose from either bank into a steep and lofty hill,
girt by a wall, which gave it the strength of a citadel, and
connected it with the town. During the short time that
Caesar stayed there to collect supplies, his soldiers had plenty
of opportunities for gossiping. The people of the place, and
especially the traders, whose business had brought them into
contact with the Germans, told marvellous stories of their
great strength and superhuman valour one could not bear :
—
even to look them in the face, so terrible was the glare of
their piercing eyes. The Roman soldiers were brave but :

they remembered tlieir desperate struggle witli the Helvetii

they were liable to fits of panic and they were very ;

Panic in credulous. The idle chatter of their new acquaintances com-

the Ro-
man army pletely demoralized them. The mischief began with the
tribunes, the officers of the auxiliary coi'ps, and others who
formed the personal following of the General. Many of
them were soldiers only in name. Like every other Roman
governor, Caesar had been obliged, for political reasons, to
army for fashionable idlers and disappointed
find places in his
professional men, who had had no experience of war, and
simply wanted to mend their fortunes by looting.^ Now
that there was a prospect of real stern fighting, they began
'
B.G., i, 39, § 2. See also Cicero, Fain., vii, b-^, 8, 10, 18 ; Q. Jr., ii, 13, § 3.
II HELVETTI AND ARIOVTSTUS 61

to tremble. They whispered that the campaign was not 58b.c.

authorized by the Senate, but undertaken simply to gratify

Caesar's ambition.^ Some invented excuses for asking leave
of absence. bound, for very shame, to stay but
Others felt ;

they could not command their countenances enough to look

as if they were not afraid. Sometimes indeed, in spite of
themselves, they gave way to tears. Gradually even cen-
turions and seasoned veterans were infected by the general
alarm. Some of them indeed made an effort to disguise their
fears. They told each other that it was not the enemy, but
only the forests between them and the enemy and the
probable failure of supplies that they dreaded. All over the
camp men were making their wills ^ and Caesar was actually ;

told that, when he gave the order to march, the men would
refuse to obey.
He immediately and centurions, and How
sent for the tribunes
gave them a severe lecture. What business had they
o ^
to ask ^'^^^^^"
restored _

where he intended to march ? It was most unlikely that confidence.

Ariovistus would be mad enough to fight but supposing that ;

he did, what was there to be afraid of ? Had they lost all

confidence in themselves, all faith in their General ? What
had these terrible Germans ever really done ? The crushing
defeats which Marius had inflicted upon the Cimbri and
Teutoni, the defeats which had been inflicted upon the gladi-
ators, trained though they were in Roman discipline, in the
recent servile war, gave the real measure of their prowess.
Even the Helvetii had often beaten them and the Helvetii ;

had gone doAvn before the legions. No doubt Ariovistus

had defeated the Gauls but what of that ? He had tired
;

them out by avoiding a battle for months, and then attacked

^ Dion Cassius, xxxviii, 35, § 2. I do not believe that Dion is inventing here.
^ Meusel {Jahresb. Vereins zu Berlin, xxxvi, 1910, pp. 44-5), following
d. philol.

W. Paul, argues that the words vulgo toiis castris testamenta ohsignahanlur
{B. G., \, 39, § 4), my
sentence is founded, are an interpolation. The
on which
reason which he gives —that
they break the context and, in their existing
position, can only refer to non nulli (' a few ') in § 3, with which they are incon-
sistent —
appears to me inadequate. Any one who reads the chapter in my
translation of the Gallic War (pp. 34-5) will, I think, admit that the suspected
words tit naturally into their place. Besides, as A. Klotz observes {Caesnr-
studien, 1910, p. 24, n. 2), they are substantially repeated by Florus (i, 45,
§ 12), who probably copied Livy which suggests that they are genuine.
;
 62 CAMPAIGNS AGAINST THE chap.

58 B.C. them when they had dispersed and were off their guard. This
did not mean that Germans were braver than Gauls and ;

Ariovistus himself must know that Roman armies were not

to be trapped by such transparent devices. To talk about the
difficulty of the country or the difficulty of getting suppHes
was downright impertinence. It was as much as to assume
that the General did not know his own business. Supplies
were coming up to the front from the friendly tribes and ;

the croakers would soon see that their alarm about the
forests was absurd. As for the story that the army was going
to mutiny, he did not believe it. Armies did not mutiny
unless generals were incapable or dishonest. His integrity
had never been called in question and the late campaign ;

proved that he could command. Anyhow on the very next

night he intended to march and if nobody else would
;

follow him, he would go on Avith the 10th legion alone for ;

it, at all events, was faithful to its commander.

This vigorous little speech had a marvellous effect upon
the troops. From despair their spirits bounded to the
highest pitch of confidence and they were only impatient
;

to cross swords with the enemy. The men of the 10th,

flattered by Caesar's trust in them, sent him a message of
thanks through their officers ; while the other legions asked
him that they were sorry for what had occurred.
theirs to tell
[About At the hour which he had fixed Caesar struck his camp. He
ug. .J
j^£^ ^ detachment to hold Vesontio. Before him all was
He re-
sames his unknown but he had full faith in Diviciacus
: and Divi- ;

a^ainst ciacus undertook to be his guide. To avoid the broken

Ariovistus. vvooded country between Besan9on and Montbeliard, he

made a circuit northward and eastward, of about fifty miles,

and then, threading the pass of Belfort, debouched into the
plain of the Rhine, and pushed on rapidly past the eastern
slopes of the Vosges he reached a point Avithin twenty-
till

two miles of the German encampment. He has not told us

where he formed his own camp probably it was on the
:

river Fecht, between Ostheim and Gemar.^ Ariovistus, w^ho

was on the north, sent messengers to say that, as Caesar had
come nearer, he had no objection to meeting him. Caesar
1 See pp. 648-52.
ir HELVETIT AND ARIOVTSTUS 63

accepted his proposal ; and the conference was fixed for the 58 b. c.

fiftli day following. Ariovistus, who knew that Caesar's

cavalry were weak, pretended to be afraid of treachery from
the legions, and insisted that they should each bring with
them a cavalry escort only. Caesar was unwilling to raise
difficulties ; but, as all his cavalry were Gauls, and he did
not care to trust his safety to them, he mounted the 10th on
their horses. The place of meeting was a knoll,^ rising above
the plain, nearly equidistant from the Roman and the German
camp. Caesar stationed the bulk of his escort about three
hundred yards off Ariovistus did likewise and each rode
: ;

up with ten horsemen to the knoll. Ariovistus had stipu-

lated that they should hold the conference without dis-
mounting. Caesar began by reminding Ariovistus of the His con-

honours which the Senate had conferred upon him and ;

^^^^ ^^^0.
afterwards repeated the demands, which he had already vistus.

made through his envoys, on behalf of the Aedui. Ariovistus

replied that he had only crossed the Rhine in response to
Gallic appeals. The country which he occupied in Gaul had
been formally ceded to him by Gauls it was not he who :

had attacked them, but they who had attacked him. He had
overthrown their entire host in battle and, if they cared to
;

repeat the experiment, he was ready to fight them again.

As for the friendship of the Romans, it was only fair that
he should get some solid advantage out of it and if he ;

could only retain it by giving up the tribute which he received

from his subjects, he would fling it aside as readily as he
had asked for it. He had entered Gaul before the Romans.
Caesar was the first Roman Governor who had ever passed
beyond the frontier of the Province. What did he mean by
invading his dominions? That part of the country belonged
to him just as much as the Province belonged to Rome.
Caesar talked a great deal of the titles which the Senate had
bestowed upon the Aedui but he knew too much of the
;

world to be imposed upon by such shams. The Aedui had

not helped the Romans in the war with the AUobroges and ;

the Romans had not stirred a finger to help their Brethren

'
'

against liimself. He had good grounds for suspecting that

* Or perhaps an earthen mound. See pp. 639-40.
 64 CAMPAIGNS AGAINST THE chap.

58 b. c. the friendship which Caesar professed for him was another

sham, —a mere bhnd under cover of whicli Caesar was
plotting his ruin. He happened to know what was going on
in Rome and there were prominent men there who would
;

be glad to hear of Caesar's death. If Caesar did not with-

draw from his country, he would expel him by force of arms ;

but if he would only go away and leave him in peace, he

would show his gratitude. Caesar quietly answered that it
was impossible for him to go back from his word or to forsake
the allies of his country and, he added, if history were to
;

be appealed to, the claim of the Romans to supremacy in

Gaul was better founded than that of the Germans. He
was still speaking when a soldier rode up and warned him
that a number of Germans were edging up towards the
knoll and stoning his escort. Riding back to his men, he
withdrew them without attempting to retaliate for, though ;

he was confident that his splendid legion could easily beat

the Germans, he was determined not to give them any
pretext for accusing him of foul play.
Exasperated by this outrage, the Romans became more
than ever impatient for battle. Two days later Ariovistus
requested Caesar to meet him again, or else send one of his
generals. Caesar saw no reason for further discussion, and
did not care to expose his lieutenants to the tender mercies
of a treacherous barbarian but he sent his interpreter.
;

Mission of Troucillus,^ and a man called Metius, whom, as he believed,

Troucillus
Ariovistus could have no motive for injuring. They were
Metius. instructed to hear what Ariovistus had to say, and bring
back word. The moment he saw them, Ariovistus flew into
a passion. Why have you come here,' he shouted
'
to :
'

play the spy ? and when they attempted to explain, he

'

cut them short and put them under arrest.

Ariovistus On the same day he made a long march southward, and
Caesar's baited about six miles north of Caesar's camp, at the very
line of foot of the Vosgcs. He had
conceived a daring plan. Next
comniun]- • i •
i i i i i i i i
cation. morning his column ascended the lower slopes, marched
securely along them past the Roman army, and took up a
position two miles south of Caesar's camp. As he looked up
1 See p. 652.
 IT HELVETIl AND ARIOVISTUS 65

at the huge column winding leisurely by, Caesar saw that he 58 b.c.

was being outmanoeuvred to send the legions up the hill-

:

side would be to court destruction and he could only wait, ;

a passive spectator, while Ariovistus was cutting his com-

munications and barring the road by which he expected his
supplies.^
Next day Caesar formed up his army immediately in How
front of the camp, under the protection of his artillery. regTkied
Ariovistus might attack he liked but if he attacked, it command
if :

would be at liis peril if he declined the challenge, the

;

legionaries would be assured that the Germans were not

invincible.- Ariovistus remained where he was. On each
of the four following days Caesar offered battle but tho ;

enemy would not be provoked into leaving their camp.

Cavalry skirmishes indeed took place daily, but without any
decisive result. The Germans had light-armed active foot-
men, Avho accompanied the cavalry into action, each one of
them selected by the rider whom he attended they were :

trained to run by the horses' sides, holding on to their

manes and if the troopers were forced to retreat, they
;

supported them and protected the wounded. As the infantry

remained obstinately in their camp, and it was necessary for
Caesar to Avin back communication with his convoys, he
resolved to take the initiative. Forming his legions in three
parallel columns, prepared, at a moment's notice, to face
into line of battle, he marched back to a point about a
thousand yards south of Ariovistus's position, and there
marked out a site for a camp. One column fell to work
with their spades, while the other two formed in two lines to
protect them. Ariovistus sent a detachment to stop the
work but it was too late the fighting legions kept their
; :

assailants at bay, and the camp was made. Two legiona

with a corps of auxiliaries were left to hold it and the ;

8ee pp. (349-50.

^
Napoleon {Hist, dt Jules Cesar, ii, 89, n. 2) infers from
Caesar's narrative {B. G., i, 48, § 2) that Ariovistus only succeeded in cutting
Caesar's communication with the convoys that were
coming up from the Aedui
and the Sequani, not with those which he expected from the Leuci and the
Lingones.
* See Stoffel, Hist, de Jules Cesar, — Guerre civile, ii, 342-5 ; Guerre de Cesar
et d'Arioviste, p. 64 ; and Caesar, B. C, iii, 55, § 1 ; 84, § 2.

1093 y
 66 CAMPAIGNS AGAINST THE chap.

58 B. c. other four returned to the larger camp. Next day Caesar led
his men into the open, but not far from his camp, and again
offered battle. Ariovistus again declined the challenge ;

but, as soon as the legions had returned to their entrench-

The Ger- ments, he made a determined attempt to storm the smaller
superstT-'camp, and only drew off his forces at sunset. The Romans
tion delayhad Suffered as heavily as the Germans but Caesar now
to fighta, ,-
;

pitched learned irom prisoners that the enemy had been warned by
battle.
their ^dse women, whose divinations they accepted with
superstitious awe, that they could not gain the victory
Sept. 18. unless they postponed the battle until after the new moon.
Caesar Cacsar saw his opportunity. He waited till the following
them morning and then, leaving detachments to guard his two
;

camps, he formed his six legions, as usual, in three lines, and

marched against the enemy. They had no choice but to
defend themselves.^ Their wagons stood in a huge semi-
circle, closing their flanks and rear and, as they tramped ;

out, their women stretched out their hands and piteously

begged them not to suffer their wives to be made slaves.
The host was formed in seven distinct groups, each composed
of the warriors of a single tribe. As the Romans were
numerically weaker than their opponents, the auxiliaries
were drawn up in front of the smaller camp, to make a show
of strength. Each of the legati was placed at the head of a
legion, in order that every one might feel that his courage
in action would not be overlooked. Caesar commanded the
right wing in person, and, noticing that the enemy's left was
comparatively weak, directed against it his principal attack,
in the hope of overwhelming it speedily and thus disconcert-
ing the rest of the force. But before the Romans in the
front ranks could poise their javelins, the Germans Avere upon
them and they had barely a moment to draw their swords.
;

Quickly stiffening into compact masses, the Germans locked

their shields to receive the thrusts but some of the Romans
:

flung themselves right on to the phalanxes they tore the ;

shields from the grasp of their foes, and dug their swords
down into them and, after a close struggle, they broke the
;

formation, and their weapons got freer play. The unwieldy

'
See pp. 230, fioO.
II HELVETII AND ARIOVISTUS 67

masses, unable to manoeuvre or to deploy, reeled backward, 58 b. c.

dissolved, and fled. But the Roman left, overpowered by

numbers, was giving ground. Young Publius Crassus, son
of the celebrated triumvir, who was stationed in command of
the cavalry, outside the battle, saw the crisis, and promptly
sent the third line to the rescue. The victory was won, and They are
^
the whole beaten multitude fled towards the Rhine. But ^nd*
the Rhine was some fifteen miles away ^ the 111 had first to expelled
be crossed and; m
that weary night many tell under the
;

/»nii
from Gaul

lances of the cavalry. Only a few, among whom was Ario-

vistus, were lucky enough to swim the river or find boats.
Caesar, in the course of the pursuit, came upon his inter-
preter, who was being dragged along in chains by his captors,
and had only escaped death by the accident that, on drawing
lots, they had decided to postpone his execution. There is
nothing in Caesar's memoirs more full of human interest
than the passage in which, breaking his habitual reserve, he
tells us of the joy he felt on seeing this man, for whom he

had the greatest respect and regard, and unhurt. It

alive
gave him, he tells us, a pleasure as great as he had felt in
gaining the victory.
The victory was decisive. The Suebi, who were on the
point of crossing the Rhine, lost heart and set out home-
wards but the people in whose territory they had encamped,
;

and whom they had doubtless plundered, pursued them and

slaughtered many of the stragglers. And Caesar, where —
was he to go ? What use was he to make of his victory ?
If he had exceeded his authority in crossing the frontier of
the Province,^ it would be fatal to withdraw. That would be
to invite the German to attempt a new invasion to confess ;

weakness to the Gaul. Fortune beckoned him on. Gaul

was disunited : her foremost state was on his side ; and
others felt the spell of his success. To bring those gifted
peoples under the civilizing sway of Rome, to open their
—
broad lands to Italian enterprise, that was a work to satisfy
the most soaring ambition. For the present indeed he must
return to Cisalpine Gaul, to conduct the civil duties of his
government and watch the politics of Italy ; but leaving
* See pp. G55-7. ^ See pp. 40-1, 41, n. 1, and 220.
F 2
 68 THE HELVETII AND ARIOVISTUS
58 b. (;. his legions under the command of Labienus, he quartered
Caesar them for the winter in the stronghold of Vesontio.^ In that
^'^^^ ^^^ registration of a great resolve ;
hiriedons ^^^^ ^^^ ^^ ^^^ ^^^ ^^J
at Veson- and doubtless he reflected, as he travelled southward, upon

the magnitude of the undertaking to which he had committed

Signiii- himself. For to all who had eyes to see and ears to hear he
this step.
^^^^ made it evident that his purpose was nothing less than
the conquest of Gaul.

^ So Napoleon conjectures with probability {Hist, dc Jules Cesar, ii, 97) : w©

only know for certain that the winter-quarters were in the country of the
Scquani {Ii. G., \, 54, § 2). But Napoleon's conjecture is supported by the
fact that Caesar had garrisoned Vesontio (i6., 38, § 7).
 .

CHAPTER III

THE FIRST CAMPAIGN AGAINST THE BELGAE

The campaign which Caesar had just con-
results of the 57b.c.

eluded may be summed up in a few words he had secured, :

— Results of
til© tirsr'
at least for a time, the virtual submission of Central Gaul ; campaign.
he had paved the way for the conquest by destroying or
expelling the barbarian hordes who threatened to anticipate
him he had restored the credit of the Aedui, who would
;

serve him best and at Vesontio he had established a post,

;

commanding Central Gaul, from which he could advance

in any direction and to which, in case of need, he could
retreat.^
But the Gauls were not yet ready to bow their necks TheBelgae
beneath the Roman
Caesar's victories were doubtless agaiSr
yoke.
talked of in every village from the Rhine to the Atlantic Caesar. ;

and it needed than the Celtic quickness to perceive their

less
significance. Before the close of winter he heard rumours
that the warlike Belgae were conspiring and these rumours ;

were confirmed by dispatches from Labienus. The tribes

were binding each other by the interchange of hostages to
mutual fidelity. They were fearful that Caesar would first
conquer the rest of Gaul,^ and then conquer them. Moreover,
they were egged on to fight by certain influential chiefs from
Celtican Gaul. The motives of these counsellors were various.
Some simply desired to make their country free. It was all
very well, they argued, to have got rid of the Germans but ;

these new intruders were not a whit more welcome. If

Caesar had expelled Ariovistus, he was evidently determined
to take his place. The legions had settled down in the
' See G. Veith, Gesch. d. Feldzlhje C. J. Caesars, 1906, p. 93.
* Omni pacata Gallia {B. G., ii, 1, § 2). These words have sometimes been
taken to mean 'as the whole of Gaul was [already] subdued': but their
position in the sentence excludes such an interpretation and Meusel's lex. ;

Caes.,\, 717-20, shows that they mean 'if the whole of [Celtican] (Jaul were
subdued '
 70 THE FIRST CAMPAIGN chap.

f)? B. r. country and tliey intended to make the country support

;

them. Others, merely because they were Gauls, longed,

above all things, for revolution. Then there were princely
adventurers, who were plotting to seize royal power, and who
foresaw that, if Gaul became a Roman province, they would
be obliged to submit to law, and would no longer be allowed
to hire troops for the gratification of their ambition.
Caesar On liis own responsibility and at his own cost, Caesar
Gau["and ii^^^^^wtly raised two new legions in Cisalpine Gaul, and sent
marches them in the early spring to join Labienus. As soon as the
them. herbage was sufficiently forward to make it safe to take the
field, he crossed the Alps and rejoined his army at Vesontio.

The tribes nearest to the Belgae, whom he charged with

the duty of collecting information, reported that they were
busily raising and concentrating levies. Having arranged
for supplies of corn, Caesarpushed on and, after another
fortnight's marching, appeared on the northern bank of
the Marne.
The Belgae were taken completely by surprise. Engrossed
in their preparations against Caesar, they had never dreamed
TheRemi that Caesar might anticipate them. One tribe, the Remi,
and help ^^^o occupied the country round Reims, Laon, and Chalons,
Caesar. were slirewd enough to perceive that his patronage would
strengthen their own position. They were subject to the over-
lordship of their neighbours, the Suessiones, and wanted to
shake off the yoke.^ Two of their leading men, Iccius and
Andebrogius, presented themselves in Caesar's camp, and
not only submitted on behalf of the tribe, but promised to
render him every assistance. Nothing could have been more
opportune. He saw that it would be easy to establish in
the heart of Belgium a power as devoted to his interests as
the Aedui in Central Gaul. He gave the envoys a gracious
welcome, only stipulating that the Reman senate should
^ The Reman envoys told Caesar {B. 0., ii, that the Suessiones and
3, § 5)
the Remi formed one political community. NowGalba was the king of the
Suessiones ; and therefore, it should seem, had been on erlord of the Remi.
Mommsen then is doubtless right in affirming that the Remi discerned in this
'

invasion of the foreigners an opportunity to shake off the rule which their
neighbours, the Suessiones, exercised over them '. //<W. oj Rome, v, 1894, p. ")0
{Hum. Oesch., in. 1889, p. 259).
 OPERATKWS ON THE AISNP:(Ar.ordin{!U.{',ol.Sl»ffel)
Thi- altcrntiA-e sUe jxee pujjcs 6C0-b'66) i.s shown on iTu! cu'.compan\-uiq nu,p (Carle ,ii- L'RtaL MyoiilSVPOO], parf of sheet 3<l.j

The numbers denote the h^i^hts /n. inetrps ahovp the tevet n1' the sea .

Scale 1 60 000 Stxm^rtVs Geosl£stai)^

Rimaji Miles
 ^

Ill AGAINST THE BELGAE 71

present themselves before him, and that the sons of the 57 b. c.

leading men should be delivered up as hostages. The

envoys gave him full information. The Belgae, they said,
were full of confidence. They boasted that the Cimbri and
Teutoni, who had overrun the rest of Gaul, had never been
able to get a footing in their land. The Remi had done
their utmost to prevent the Suessiones from taking part
in the movement, but in vain indeed their king, Galba,
:

had been unanimously elected commander-in-chief. Every

other tribe had joined the league and Galba was prepared ;

to put nearly three hundred thousand men into the field.

The estimate was certainly exaggerated but Caesar could :

muster little more than forty thousand and his enemies ;

were the stoutest and the most stubborn of all the warriors of
Gaul. His only chance of success was to force their huge
host to divide. With this aim, he asked Diviciacus to raise
a levy of Aeduans and ravage the lands of the Bellovaci, He sends
which lay beyond
J the Oise, in the region now dominated by P^viciacua
-J
^
'
•/to ravage
the huge choir of Beauvais. The entire armament was now the lands
in full march against him. They were moving down a road Bellovaci
which led from La Fere, on the Oise, past Laon to Reims.
Caesar determined to choose his own battle-field. Marching Marches to
rapidly northward from Reims, he crossed the Aisne by a
theadv^ n-
bridge, and encamped on rising ground between that river cing host,
and a small morass. The site is still disputed we only know Aisne,\nd :

that it was either on the plateau of Pontavert or four miles encamps

1 • 1 T»/r
1
^^^ rising /• 1
higher up the stream on the gentle slope or Mauchamp, ground
about a mile and a half north-east of Berry-au-Bac. This "^^i,*'^®
bank. •^

might seem to be the point where Caesar would naturally

have crossed the Aisne but neither the slope nor the
;

trenches which have been discovered upon it correspond

satisfactorily with his description. The rampart of the
camp, eight feet high, was faced with sods and revetted with
timber, to keep its slope of the requisite steepness along :

the top of it was set a palisade of interlacing branches ^ and ;

the ditch which surrounded it was eighteen feet in width.

1 See pp. 241-2. See pp. 658-9.

^

=*
Rampart and palisade combined were 12 feet high [B. O., ii, 5, § 6, with
which cf. A. Klotz, Caesarstudien, p. 220, n. 2). See pp. 580-7.
 THE FIRST CAMPAIGN CHAP.

57 B. c. Caesar's rear was protected by the Aisne and his suppHes ;

could be brought up in safety by the Remi. At the northern

end of the bridge he cstabhshed a tele de poiit and, to guard ;

its furtlier extremity, he left a detachment about two

thousand strong under one of his generals, Titurius Sabinus.

TheBelgae Towards midnight a messenger came into camp with the
attack
Bibrax. news that the Belgae were making a furious attack upon
Bibrax, or Vieux-Laon, a Reman stronghold about seven
miles to the north, and that Iccius, who commanded the
Caesar garrison, despaired of being able to hold out unless he were
sends his
auxilia-
promptly reinforced. Caesar instantly dispatched his
ries to the Numidian and Cretan archers and Balearic slingers to the
rescue.
rescue. The Gauls knew nothing of the scientific methods
by which the Romans captured fortified towns. When their
numbers Avere sufficient 1}^ great, they used to drive the
defenders from the rampart by showers of missiles, and then,
locking their shields over their heads, to demolish a portion
of the wall.^ But Bibrax was defended on the south by
impregnable escaipments it would seem that Galba had
:

neglected to invest this side and when Caesar's light troops

;

appeared, the impatient and undisciplined host abandoned

their attempt.- They only lingered long enough to ravage
the lands and fire the hamlets within reach of the town.
The On the following night the sudden blaze of a line of watch-
Belgae
encamp fires, extending some seven miles in length beyond the further
opposite
side of the morass, revealed to Caesar their encampment.
Caesar.
Caesar So formidable was the appearance of the huge host, so
makes his
position
great was their reputation as fighting men, that Caesar did
impreg- not care to risk a battle until lie had seen enough to judge
nable.
whether he woidd have a reasonable chance of success. A
few cavalry skirmishes convinced him that he had nothing
to fear. The position which he had chosen may be described
in his own Avords. The hill on which the camp stood.
*

^ Here again Meusel {Jahred). d. phUol. Vertins zu Berlin, xxxvi. 1910.

pp. 40-1) sees an interpolation in Caesar's text {B. 6'., ii, G, §§ 2-3) ; but as he
admits that the sentences may nevertheless be true, and as I cannot conceive
tliat any interpolator would have invented them, I leave the text of my first

edition unchanged. [A, Klotz {Caesarstiidien, pp. 243-4), who generally agrees
with Meusel, does not suspect this passage.]
^ See p. 241. n. (i. and note on Bibrax, pp. 300-400.
 ;

Ill AGAINST THE BELGAE 73

rising gradually from the plain, extended, facing the enemy, 57 b.c
over the exact space which the line would occupy on :

either flank its sides descended rapidly while in front it

;

gradually merged in the plain by a gentle slope.' ^ The

legions were protected in front by the morass but the vast
;

numbers of the enemy might outflank them. To prevent

this, Caesar made his men dig two trenches, each about three
furlongs in length, transversely on either side of the hill
and at the extremity of each trench he caused forts to be
constructed and armed with hallistae and catapults. Along
the whole length of the hill, in front of the camp, he drew
up six of liis legions in battle array while the other two
;

remained to guard the camp. The enemy's masses were

ranged on the further side of the morass. Each of the two
armies obstinately waited for the other to cross. Meanwhile
Caesar's cavalry were scattering the Belgic squadrons. At
length, tired of waiting, he led his legions back into camp.
There Avas a ford on the Aisne, about two miles from the
tete de ^wnt, which he had either failed to notice or had not

thought it necessary to guard. Presently an orderly came

from Sabinus, who reported that a body of the enemy were
moving down to the bank, evidently intending to cross over,
attack his camp, and destroy the bridge. Even if they The
failed, the corn-fields of the Remi would be at their mercy :
^ftempt
the convoys would be cut oif and then the legions would to cut his
;

starve. Taking his cavalry, light-armed Numidians, archers, cations.

and slingers, Caesar hurried down the hill, crossed the bridge, ^^!^ *^"^
and pushed along the bank towards the ford. There were
the enemy, splashing through the water. The archers and
slingers attacked them, and did terrible execution. The
survivors clambered over the fallen bodies, and staggered
on, but were beaten back by showers of stones and arrows ;

while the leading division, who had crossed already, were

surrounded by the cavalry and cut to pieces.'-
The Belgae were thoroughly disheartened. They had no They
organized commissariat and their supplies were running "^'^P*^^'^^-
;

out. Galba had not the genius to control a vast multitude

> B. G., ii, 8, § 3.
'*
Regarding Caesar's operations on the Aisne see pp. 659-08,
 74 THE FIRST CAMPAIGN CHAP.

57 B.C. made up of hordes without discipHne, with conflicting

interests, and distracted by mutual jealousies. Caesar's
position was impregnable and he evidently had no intention
;

of quitting it. His allies would soon be swarming over the

frontier of the Bellovaci and the chiefs of that tribe
;

insisted on returning to defend their families. It was

decided, therefore, that each tribe should go back to its own
country, and that, whatever district the Romans might
invade, all should rally to its defence. But this resolution
was merely to save their self-respect. In the night the
whole multitude poured out of their encampment with great
uproar and confusion, each man struggling to get in front
of his fellows. Caesar at first suspected that this movement
was merely a ruse ; but at daybreak he received positive
Caesar's information that the enemy had really gone, and immediately
cavalry
pursue sent his cavalry, supported by three legions under Labienus,
them. in pursuit. The rear ranks, when they were overtaken, stood
at bay, and resisted resolutely ; but those in front, hearing
the shouts of the combatants, made haste to escape. The
He slaughter was instant ; and the pursuers raced on. As long
marches
westward, as dayhght lasted they hung on the rearguard, slaying,
and re- pursuing, and slaying again and at sunset they returned ;
ceives the
submis- to camp. Caesar left the disorganized host no time to rally.
sion of the
Sues-
The Suessiones, Bellovaci, and Ambiani had diverged from
siones, their natural route, in order to fetch their baggage and ;

Bellovaci,
and by a great effort he might be able to surprise their forts.^
Ambiani, Next morning he pushed on westward down the valley of
[Probably the Aisne. In a single forced march he reached Noviodunum,
Pommiers,
about near the modern Soissons, the chief stronghold of the Sues-
two and a siones, and at once attempted an assault - but though the ;
half miles
west of garrison was weak, the moat was so wide and the wall so
Soissons.]
high that his troops were repulsed. In spite of their fatigue,
they proceeded to fortify their camp and make preparations
for a siege. Sappers' huts were constinicted for protecting
^
the workers : earth and fascines were shot into the moat ;

1 See p. 070. * See pp. 464-0.

* Aggere iacto {B. 0., ii, 12, § 5) See Napoleon III, Hist, de Jules Cesar,
ii, 105 and n. 2. Von Goler {Caesars gall. Krieg, 1880, p. 73) takes aggere in
the sense of a terrace (cf. p. 81, infra), the construction of which would have
' '

required many days.

 THE BATTLE OF NEUF-MESNIL.
Th^ numbers heights ahove
denote the of in. metres the level the sea,.
To face paxie 75

OTISMIM (Jh
/ , A
V 1

^^Xj ,

r^^* iae

i^\ % r
.. ;«^^^\\V\^

Stajx^rd-'s Geog^I^'^tah^
KilcimftrHS Scale 1:4.0,000 Romeui Miles
»4 V-s V4 o
 ;

Ill AGAINST THE BELGAE 75

and movable towers were erected to carry the artillery 57 b.c.

which was to play upon the defenders of the wall> During

the night the whole host of the fugitive Suessiones thronged
into the town and reinforced the garrison but they were so
;

confounded by the formidable appearance of the siege works

that they surrendered without striking a blow.^ Marching
on westward, Caesar crossed the Oise. Bratuspantium, the [Breteuil?p

chief town of the Bellovaci, opened its gates on his approach;

and when he drew near Samarobriva, where now rises the
colossal pile of the cathedral of Amiens, the Ambiani likewise
tendered their submission. Caesar treated the three tribes
with equal clemency and firmness. He punished no one
but he disarmed the garrisons of Noviodunum and Bratus-
pantium, and required the surrender of hostages of noble
birth. Diviciacus, who had rejoined him, interceded for the
Bellovaci ; and, as his policy was to strengthen the influence
of the Aedui, he gave out that it was his regard for those
loyal allies which led him to show mercy. But now he
learned that his progress was about to be disputed. On the
north-east, among the inhospitable forests of the Sambre
and the marshes of the Scheldt, dwelt a tribe whose primitive
virtues had not yet been enfeebled by contact with civiliza-
tion. No traders were suffered to cross their frontier, for
fear the luxuries of which the rude warriors were still
ignorant might sap their manhood. Bitterly taunting their The Nervii
neighbour tribes for having so tamely surrendered, they JeSt!^
vowed that for their part they would accept no terms of
peace. This people, whom of all his enemies Caesar most
respected, and of whom he wrote with one of those rare
touches of enthusiasm that here and there relieve the severity
of his narrative, were the Nervii.
A couple of marches brought the legions to the Nervian Caesar
frontier. The road led through Hainaut, past the site of "ga^nsf
the modern Cambrai. Three days later Caesar gathered from t^iem.

some mstics, who had been taken prisoners, that the warriors

' Livy, xxi, 11,§7.

* Long {Decline of the
Roman Republic, iv, 52, note) thinks that there would
not have been time to complete the construction of the sappers' huts until
after the Suessiones had arrived. *
See pp. 400-2.
 76 THE FIRST CAMPAIGN chap.

57 B. c. of the tribewere encamped only nine miles off, on the further

He learns bank of the Sambre, with their allies, the Viromandui and
and th^r ^^^ Atrebates ; and that another tribe, the Atuatuci, were
allies are marching from the east to join them. They were waiting
on the for him near Maubeuge, where they expected that he would

^f^th
^^"^^ cross the river. ^ He immediately sent on a party of cen-
Sambre. turions and pioneers to choose a camping ground. It
happened that some of his prisoners had escaped to the enemy
in the night. They told them that each of the Roman
legions was separated on the march from the one that
followed it by a long baggage-train and that, when the
;

foremost legion, encumbered with their heavy packs,

reached the camping ground, it would be easy to overwhelm
them and plunder the baggage before the others could come
His to the rescue. The centurions selected for the site of the
mark m!t ^amp the heights of Neuf-Mesnil, about three miles south-
a camp west of Maubcugc. The ground sloped evenly and gently
heights of towards the left bank of the Sambre but from a point
;

^^"*' opposite the front of the camp to the bend of the river at
Boussieres, nearly a mile and a half further up, the slope
terminated near the water's edge in a steep escarpment. The
depth of the river was not more than three feet. From the
opposite bank an open meadow, over which were scattered a
few cavalry piquets, rose into a hill covered with woods.
The space for the camp was measured and marked out.
Meanwhile the Roman army was toiling up from behind,
its march being delayed by thick hedges, which the
inhabitants had planted long before to check the raids of
their neighbours' cavalry. The formation was different
from that which had been described to the Nervii for
;

when close to an enemy, Caesar always changed his order

of march. In front came six legions in column. Then
followed the entire baggage-train, protected by the two
newly raised legions, which closed the rear. The cavalry,
who had gone on in front, rode across the shallow stream,
and, supported by archers and slingers, engaged the enemy's
piquets. The piquets fell back into the wood, whither the
cavalry dared not follow them and there leisurely re-forming,
;

See p. ()7a.
 ;

Ill AGAINST THE BELGAE 77

they charged again and again. As the infantry arrived upon 57 b. c

the ground, some began to dig the trenches for the camp,
while others scattered over the country to cut down wood.
Caesar neglected to take the precaution of keeping a part of
his force under arms.^ At length the head of the baggage-
train appeared. Ambushed among the trees, the Gauls
caught sight of it. Suddenly they darted forth from the wood
and came pouring down the open their rush swept away the ;

terrified cavalry now they were across the river and racing
;

up the slope and now they fell upon the half -formed line.
;

The confusion was overwhelming. From the moment Battle of

when the onrushing host was seen there were hardly ten Mesnil.
minutes for preparation. The Romans flung aside their tools.
Caesar had to give all his orders in a breath. The red battle-
ensign was quickly hoisted over his tent. The blast of the
trumpet recalled the men who were working at the further
side of the camp, while messengers ran to fetch those who
had scattered far afield. They had not a moment even to
cram on their helmets or pull the coverings off their shields.
The generals were obliged to act without waiting for orders ;

and Caesar Avas glad that he had forbidden them to leave

their respective legions while the camp was being made.
He could not direct them ; for the hedges which crossed the
field obstiTicted his view. Want
time as well as the natureof
of the ground prevented them from forming a regular line
of battle : along the brow of the hill a number of isolated
combats were beginning at once and all that could be ;

done was to make each legion face its immediate assailants.

Disciplined, and self-reliant from the experience which they

^ As he had done when constructing his smaller canij) in presence of the

hostile force of Ariovistus {B. G., i, 49). The great Napoleon blames him for
having allowed himself to be surprised. II est vrai,' he says,
'
que sa cava- '

lerie et ses troupes legeres avaient passe la Sambre mais, du lieu ou il etait, il
;

s'apercevait qu'elles etaient arretees a 150 toises de lui, a la lisiere de la foret

il devait done ou tenir une partie de ses troupes sous les armes, ou attendre que

ses coureurs eussent traverse la foret et eclaire le pays. II se justifia en disant

que les bords de la Sambre etaient si escarpes qu'il se croyait en sOrete dans la
position ou il voulait camper.' Precis des (juerrts de Cesar, 1836, p. 45, with
which cf. Turpin de Crisse, Comni.. de Cesar, i, 1785, pp. 154-C. It should be
noted that 150 toises is a mistake
'
' the distance from the Roman camp to
;

the edge of the wood was about 7 furlongs.

 78 THE FIRST CAMPAIGN chap.

57 B. c. had gained, the soldiers instinctively grasped the situation :

they did not trouble themselves to join their respective

companies, but one after another, as they hastened up,
they fell into the ranks by the standards nearest them.
Hurrying down at haphazard to cheer them on, Caesar found
himself close to the left of the line. There was the 10th,
his favourite legion. 'Keep cool, men,' he cried, 'and re-
member the honour of the legion. Stand up against that rush !

He had no time to say more for the enemy were within

;

a javelin's cast, and, as he hurried on, both sides were engaged.

Hurling their javelins, the 10th and, on their left, the
9th fell, sword in hand, upon the At rebates, who, panting
from their headlong rush, soon gave way. Hunted down the
slope, they plunged into the stream, but the Romans dashed
after, sword in hand and when the survivors clambered up
;

the further bank and tried to rally, fell upon them again
and chased them up the hill. At the same time the 11th
and 8th drove the Viromandui from the front of the camp
right down to the water's edge. But the very success of
these four legions was disastrous to their comrades the —
12th and 7th —on the right. The left and front of the camp
were exposed Boduognatus, the commander-in-chief, seized
;

his opportunity ^ and the Nervii, compacted in one mighty

;

column, climbed the steep blufif, swarmed up the heights,

and while some outflanked the two legions on their right,
the rest pressed on for the defenceless camp. The beaten
cavalry came full upon them and again took to flight the :

officers' servants, who had gone out to plunder, looked back,

and ran for their lives the baggage-drivers, Avho were coming
:

up, scattered in all directions, shrieking with terror; and a body

of horse from the Treveri, who formed part of the auxiHary
force, rode off homewards to announce Caesar's defeat.
Caesar saw it all as he made his way from the left to the
right wing. The men of the 12th, who had never before
fought in a pitched battle, were huddled together so closely
that they could hardly use their swords ;and nearly every
officer was either killed or wounded. The standard-bearer
^ Capt. Veitli {Gesch. d. Feldziige G. J. Caesars, p. 109) compares his manoeuvre
to that of Epaminondas at Leuctra.
Ill AGAINST THE BELGAE 79

of the 4th cohort had fallen: the standard —to Roman 57b.c.

soldiers a holy emblem —was lost. Sextius Baculus, the

chief centurion of the legion, was so weakened by loss of
blood that he could no longer stand. From the rear ranks
men were slinking away to escape the showers of missiles.
There were no reserves and the numbers of the enemy were
;

inexhaustible. Fresh swarms kept pressing up the hill, and

closing in on either flank. Seizing a shield from a man in
the rearmost rank, Caesar pushed his way through to the
front : he called to his centurions by name he told the men :

to open up their ranks —

so they would be able to use their
—
swords better and charge. At the sound of his voice their
spirits rose and each man of them hoped that the General
;

would see how bravely he could fight. But the 7th also, on
their right, were hard pressed. Caesar told the tribunes to
bring the two legions gradually closer together, and form
them up so as to face the enemy on every side.^ And now,
as the men were relieved from the dread of being attacked in
the rear, they fought with renewed confidence. The two
legionswhich guarded the baggage had heard of the fight,
and were marching up at their utmost speed. Suddenly
above the ridge of Neuf-Mesnil they appeared and ;

presently the 10th, dispatched by Labienus, recrossed the

river, hurried up the hill-side, and threw themselves upon
the enemy's rear. The effect of their appearance was elec-
trical. Even the wounded leaned on their shields, and plied
their swords the scattered camp-followers plucked up
:

courage and turned upon the enemy while the cavalry did
;

all they could to atone for their flight. The Nervii in their
turn were hemmed in. But in their last agony they made
good their proud boast. Man by man, beneath the javelin
and the thrust of the short sword, their front ranks fell.
Higher rose the heap of prostrate bodies and leaping on to
;

them, the survivors snatched up the fallen javelins and flung

them back, till they too fell and all was still.^
;

'
8ee pp. 07G-7.
''
See pp. 671-7. CWsar'.s narrative {B. G., ii, 27, §§ 3-5, 28, §§ 1-2) implies
that a few of the Nervian contingent escaped ; but whether they ran away
from the fighting line or had not come into action at all. he does not say.
 80 THE FIRST CAMPAIGN chap.

57 B.C. So ended this wild fight, —a soldiers' battle, but withal

the battle of a great man. Within an hour it was over,
fought and wellnigh lost and won.^
The power of the Belgae was broken. What remained to
be done was only matter of detail. The old men of the
Nervian tribe, with the women and children, had gathered
before the battle in the midst of the marshes formed by the
estuary of the Scheldt. Within a few days a deputation
came from them to ask an audience of the conc[ueror. They
were shrewd enough to exaggerate their losses.- Their army,
they said, was all but annihilated. Only five hundred
fighting men remained out of sixty thousand and of six ;

Caesar hundred senators no more than three. Wishing to establish

survivors ^ reputation for clemency, Caesar permitted the survivors to
with clem- retain their lands and even their fortified villages, and Avarned
©ncv
the neighbouring tribes to refrain from molesting them. He
then marched eastward against the i\.tuatuci. This people
were dift'erent in origin from the rest of the Belgae. Fifty
years before, the Cimbri and Teutoni, marching for the south,
had left some of their number, under the protection of six
thousand warriors, in Belgic Gaul, to herd the cattle and
guard the booty which they could not take with them.
After the destruction of their kindred, these men and their
descendants had continued to maintain themselves against
the enemies who surrounded them they had achieved, by :

prolonged fighting, a commanding position and they now ;

occupied the broad plain of Hesbaye on the northern bank of

the Meuse.^ On hearing of the defeat of their allies, they
had returned home and concentrated in one town of great
strength, situated on Mont Falhize, opposite the modern
fortress of Huy.
The Meuse, winding in the shape of a
horse-shoe, flowed through the meadows beneath the southern
slopes of the hill and the town, perched above its rocky
;

heights, seemed inaccessible, save by one gentle ascent on

the north-east, where a high wall ^ frowned down upon the

'
See p. 077. * !See pp. 20o-7.
* In 57 B. c. they may also have possessed lands on the right bank. See pp.
;i85-7.
* Caesar {B. G., ii, 29, § 3) calls it
'
a double wall of great height' {duplici
Ill AGAINST THE BELGAE 81

besiegers. Heavy stones and pointed beams were ranged 57b.c.

upon the wall ; and in front of it was a deep moat. At first He be-
., . , •
f •
^ r-i J^ sieges the
the garrison made a succession or sorties :
,

but Caesar threw strong-

up a rampart from one reach of the river, round the north ^1^^^?^^^^^

of the hill, to the other ; and, as was usual in regular sieges,

a terrace, composed of a core of earth and timber, supported
by was built up at right angles
walls of logs piled cross-wise,
to the wall.^ On this terrace was erected one of the wooden
towers from the stories of which archers, slingers, and
artillery used to shower missiles among the defenders of a
besieged town. It was intended that, as soon as the terrace
approached the wall, a battering-ram should be employed to
effect a breach. The garrison, confident in the strength of
their fortress, watched these operations with ignorant con-
tempt. They despised the Romans for their small stature,
and asked them if they imagined that such pygmies as they
could get a huge tower like that on to the wall. But the
laugh was soon turned against them. When they saw the
tower actually moving on its rollers and steadily nearing
the wall, they fancied that there must be some supernatural
power at work, and in great alarm sent out envoys to beg
for terms. They would surrender, the envoys said only ;

they entreated to be allowed to keep their arms, without

which they could not defend themselves against their neigh-
bours. Caesar insisted on unconditional surrender. He They
would take care that their neighbours did not molest them. ^""^"
^^•

The chiefs could only submit and swords, spears, and shields
;

were pitched down into the moat until the heap almost
reached the top of the wall. Towards sunset all the Roman
soldiers who had gone into the town were withdrawn, for
fear they might commit any excesses. The garrison had
kept about a third of their weapons in reserve, and had
improvised shields of bark and wattle-work, covered with

aUissimo muro). The meaning of duplex has been much discussed ; and
Colonel De La Noe {Bull, de geogr. hist, et descr., 1887, p. 253) understood it as
simply gros, large, epais
'
but M. J. Dechelette [VAnthr., xvii, 1906, pp. 393-5)
' :

points out that M. Saint-Venant has discovered ancient forts in Provence and
the Maritime Alps, the ramparts of which were formed each of two distinct walls.
^ The difficult questions relating to the construction of the siege-terrace
{agger) are discussed on pp. 599-607. See also pp. 140, U4.
loss
Q
 82 THE FIRST CAMPAIGN CHAr.

57 B.C. skins. They calculated that the Romans would be oil their
guard, and laid their plans accordingly. The contra vallation
But after- was traced along rising ground. In the middle of the night
make a ^^^^ Atuatuci pourcd out of the gates, and advanced to attack
treacher- it where the ascent was easiest. But Caesar had provided
'
against the chance of treachery. Piles of wood, all ready
laid, were set ablaze and, warned by the signal, the troops
;

came streaming from the nearest redoubts. The Gauls

fought with the courage of despair but missiles rained down
;

upon them from the rampart and from the towers which had
been erected upon it and they were driven back with heavy
;

loss into the town. Next day the gates were burst open, and
the Romans rushed in. Caesar was neither vindictive nor
cruel but to those who defied him, and especially to those
;

Tiieiipuu- who broke faith, he was absolutely ruthless. Fifty -three

18 men .
—
thousand of the Atuatuci all avIio were found within the
—
town were sold as slaves.^
The campaign was over. The prestige which it had won
for Caesar was so great that more than one German tribe
sent envoys across the Rhine to offer submission. One
partial failure alone marred the general success. Amid the
clash of arms, Caesar did not forget the commercial advan-
(Jalba's tages whicli his conquest might secure for Rome. On liis
iiiThe^^^"
way back to Italy,'*^ he sent one of his generals, Servius Galba,
Valais. to open up the road leading from the Valais over the Great
St. Bernard into Italy, which traders had only been able to
use hitherto at great risk and by the payment of heavy tolls.
The tribes with whicli Galba had to deal were the Nantuates,
who occupied the Chablais and the southern bank of the
Rhone as far as St. Maurice the Veragri, whose chief town,
;

Octodurus, stood upon the site of Martigny, near the con-

Rhone and the Dranse ^ and the Seduni
fluence of the ;

whose name is preserved in the modern Sit ten, or Sion.

His force consisted only of the 12th legion, which had
suffered so severely in the battle with the Nervii, and a
body of cavalry. Skirting the northern shore of the Lake of
Geneva, the little column entered the broad valley of the

^ 8ee p. 44, and note on Atuatucorum orpiDUM, pp. 387-03.

* Suhueider':i Catdur, i, 210, note. * See pp. 077-S.
Ill AGAINST THE BELGAE 83

upper Rhone, walled in on right and left by wooded mountains. 57 b. c.

Having inflicted several defeats upon the mountaineers,

captured several of their strongholds, and compelled the chiefs
to surrender their sons as hostages, he posted two cohorts
in the neighbourhood of St. Maurice, and took up his own
quarters in Octodurus. The Dranse, which then flowed in
a different channel, down the middle of the valley, divided
the village into two parts, the eastern of which he allowed
the inhabitants to occupy, while lie reserved the vacant
quarter for the legion, and proceeded to fortify The it.

encampment thus formed was between Martigny-la-Ville and

the more southerly Martigny Bourg. Besides the two
cohorts which he had detached, Galba was obliged to send
out a number of small parties for supplies. The camp was
dominated on either side by the heights which border the
valley of the Dranse ; and the force which remained was
insufficient for its protection. The mountaineers resented
the deprivation of their children ; and, as Caesar half naively
remarked, they believed that the Romans, not content with
occupying the roads, intended to annex their country. One
morning Galba was informed that the heights were covered
by armed men. They were evidently determined to cut his
communications, and to bar his exit from the valley. The
fortifications were still unfinished, and the supply of corn
was insufficient ; for, as the mountaineers had submitted
and given hostages, Galba had never dreamed that he might
have to fight. A council of war was called. Some of the
officers urged Galba to abandon the baggage and fight his
way out ; but he resolved, with the concurrence of the
majority, to defend the camp. The troops had only just
time to man the rampart before the enemy rushed down to
the attack. They hurled stones and darts from every side.
The Romans offered a vigorous resistance and not a missile
;

which they threw from their commanding position missed

its mark. But the enemy's numbers enabled them to bring
down fresh men as often as they were wanted while the ;

Romans had to fight on without relief. For six hours they

fought at bay till their stock of missiles was nearly spent,
and the enemy were beginning to fill up the trench and to
G 2
 8i THE FIRST CAMPAIGN chap.

57 B.C. break down the rampart. Just in time, Sextius Baculus,

who had fought so gallantly on the Sambre, and a tribune
named Gaius Volusenus, ran to the chief, and convinced
him that the only chance of averting destruction was to
make a sortie. The men were told to stand quietly on the
defensive for a few minutes, and rest themselves. Suddenly,
at a given signal, four compact little columns dashed out
from all four gates, and cut their way through the loose
ranks of the astounded mountaineers. There was no time
to rally. Discipline prevailed over numbers and the ;

mountaineers were driven with heavy loss out of the plain,

and chased over the hills. But Octodurus was plainly
untenable and it appeared impossible to obtain supplies.
;

Next day therefore Galba burned all the houses in the

village, and returned to spend the winter in the country of
the AUobroges.
It is easy to find fault with Galba and nobody would
;

contend that he made the best of the situation. But his

circumstances were extremely difficult. The merest tiro
who spends a day in the valley of the Dranse w^ill see that
he could not have chosen a worse position than Octodurus.
But the force which Caesar allotted to him was inadequate :

it was hard to resist the temptation of occupying the village,

where substantial dwellings already existed and if he had

;

selected a suitable site, he might have been obliged to build

a large number of huts, which would have increased the
labours of his weakened legion.
The other had already been distributed in their
legions
winter-quarters. One, under Publius Crassus, the young
general whose promptitude had contributed so much to the
defeat of Ariovistus, had been sent, after the battle with the
Submis- Nervii, to receive the submission of the maritime tribes of

tHbes^of
'^ Normandy and Brittany.^ This legion and certain others
Brittany were cautoucd along the valley of the Loire, from Angers
mandv.^ to Orleans while the rest were quartered near the theatre
;

of the recent Belgic campaign.- Commius, a chief of the

Atrebates, w^hom Caesar had marked as a capable diplomatist,
^ The Veneti, Venelli, Osismi, Coiiosolites, Esuvii, Aulerci, and Redones.
-
The exact position of these legions cannot be fixed. See p. 709, n. 14.
Ill AGAINST THE BELGAE 85

was made king over his own people and apparently his
; 57 b. c.

appointment was acceptable to them.

In Italy the news of Caesar's victories Avas received with Rejoic-

an outburst of enthusiasm > Men felt that he had avenged Ro^^pl^

the disaster of the Allia and even the Senate gave expres-
;

sion to the popular feeling. After his dispatches had been

read, it was decided to hold a thanksgiving service of fifteen
—
days, an honour which no Roman citizen had ever received
before.
^ Plutarch, Caesar, 23.
 CHAPTER IV
CAMPAIGNS AGAINST THE MARITIME TRIBES AND
THE AQUITANI
56 B. c. The barbarian invaders of Gaul had been destroyed or
Delusive driven back the Belgae had been chastised and many of
; ;
prospect R
of peace. the other states had proffered their submission. The Aedui
and the Remi were still friendly and the countenance of
;

Caesar had greatly increased their renown, and therefore

the influence which they were able to exert on his behalf.
The Remi, indeed, had succeeded to the position from which
the Sequani had been deposed numerous tribes acknow-
:

ledged their ascendency and, though their prestige was

;

still inferior to that of the Aedui, a time might come when

they would be useful as a counterpoise to them. The Gallic

peoples had little consciousness of national unity they :

were familiar with the idea of Roman dominion and, while ;

Caesar did not interfere with their domestic affairs, they

were not prepared to make any serious effort to throw off
his supremacy. So confident was Caesar in the prospect of
tranquillity that he set out on a political tour to Illyricum,
—the most distant quarter of his province. But Gaul was
[The still a long way from bemg subdued. The legion under
country
of the
Publius Crassus had been quartered in the northern pait of
Andes.] Anjou. The most considerable of the neighbouring tribes
were the Veneti, who dwelt in the storm-beaten tract of
western Brittany which comprises the department of the
Morbihan and the southern part of the department of
Finistere. Like the modern Bretons, they were the stoutest
and the most skilful seamen in Gaul they had a powerful
;

fleet of large vessels, the model of which had, we may sup-

pose, been originally borrowed from that of the merchantmen

of the Carthaginians, whose commerce in the Atlantic and
 MARITIME TRIBES AND THE AQUITANI 87

in Britishwaters they had inherited and their prosperit}' 56B.r.

;

depended upon the carrying trade with Britain, of which

they possessed the monopoly. They, however, as well as
the more distant tribes of Brittany and Normandy, professed
to submit and Crassus sent a number of officers to arrange
;

with them for a supply of corn. But the chiefs of the Veneti
were beginning to repent of their tame surrender. Besides Rebellion
their natural impatience of foreign ascendency, they had, y^^^t-
we are told, a business-like motive for resistance. They had Corioso-

heard, should seem, that Caesar was contemplating an

it
Esmi?"*^
invasion of Britain and they were naturally determined
;

to prevent him from interfering with their trade. ^ The

rumour which had reached them was true. The subjection
of the Veneti was a necessary prelude to the invasion of
Britain. For Caesar could not safely embark his army
unless he had command of the Channel and at the time
;

when he planned the invasion the masters of the Channel

were the Veneti.^ Hoping to induce Crassus to restore their
hostages, they detained as prisoners the officers who had
come to them. With the rash precipitancy of Gauls, the
tribes of the C6tes-du-Nord and the Orne followed their [ThoCorio-
'^olitesand
example presently the whole north-western seaboard was
:

sworn to resist the encroachments of Rome and an embassy ;

was sent to Crassus, to demand the restoration of the

hostages.
Messengers were soon posting with dispatches for Caesar, Caesar
who was still in lUyricum. He had studied the character foTanaval
of the Gauls to some purpose and he knew that, if they war.
;

soon lost heart, their blood was up on the slightest stimulus.

Like other peoples, they preferred independence to subjection;
and, above all things, their restless spirit craved variety. If
he were to overlook the conduct of the Veneti, the other
tribes of Gaul would fancy that they might defy him with
impunity. The Belgae indeed were only half subdued and ;

they were said to have solicited the support of the Germans.

Accordingly Caesar sent instructions to his officers to have a

^
Strabo, Oeogr., iv, 4, § 1.
* See Rice Holmes. Anrient Britain and the. Invasions of Jvlivs Caesar,
pp. 301-3.
 88 CAMPAIGNS AGAINST THE chap.

56 B.C. the ports at the mouth of the Loire, to raise

fleet built in
oarsmen from the Province, and to collect as many pilots and
seamen as they could.
The con- Throughout his proconsulate Caesar was in a position
Luca!'^
^ different from that of a modern viceroy, who, if his work is
almost beyond his strength, may securely concentrate upon
it all the power of his mind. He was ever obliged to look
back towards Rome, to look forward to the uncertain but
stormy future, when he would have to struggle for political
supremacy and whenever an enemy attempted to weaken
;

his position, he was obliged to parry the blow. His old

enemy, Domitius Ahenobarbus, who was a candidate for
the consulship, openly declared that when he came into
office he would deprive him of his command. Even Cicero
manifested an inclination to oppose him and he had reason ;

to fear that Pompey would join Cicero. His term of office

would expire in about two years, on the 1st of March, 54 B.C.
If he were recalled then, his work in Gaul would be left
unfinished and he Avould go back, too soon, to chaos or
; —
civil war. From lUyricum he had returned to Ravenna,
where the elder Crassus met him. Hearing of Cicero's
measures, he moved southward, about the middle of April,
to Luca, and invited Pompey to come thither as his guest.
At this little town the fortunes of the world were decided.
Caesar offered terms of such startling liberality that an
agreement was come to at once. It was arranged that his
term of office should be prolonged for another five years,
and that Pompey and Crassus should exert their influence
with the Senate to secure to him the right of increasing his
legions to ten, and of charging the state-chest with the pay
of those which he had raised on his own responsibility.^
Caesar From Luca he hastened to join his army, and took up his
Gaul^^ quarters in the neighbourhood of Nantes. ^ His first step

1
Cicero, Fam., i, 7, § 10 ; 9, §§ 9-10 Q.fr., ii, 6, § 2
; ; Dc prov. cons., 11, § 28 ;

Suetonius, Divus lulius, 24 ; Appian, B. C, ii, 17 ; Plutarch, Crassus, 14,

Caesar, 21, Pompeius, 51.
' M. Jullian {Hist, de la Oaule, iii, 290) believes that Caesar, on returning
from Italy to Gaul, went to Belgium, where he rejoined the bulk of his army.'
'

But there is no evidence that the bulk of the army was in Belgium on the :

contrary, four legions at least out of the eight which then composed the army
 ;

IV MARITIME TRIBES AND THE AQUITANI 89

was to distribute the legions more widely. Labienus was 56 b. c.

sent to the country round Treves, to keep an eye upon the [The terri-

Belgae and to prevent the Germans, whose aid they were Tre^eri i

said to have solicited, from crossing the Rhine. Sabinus [The

was directed to disperse the allies of the Veneti in the y^^eHi^'
Calvados, the Cotentin, and the C6tes-du-Nord while ; and
Crassus marched for Aquitania. It is most unlikely that the HtJs.T^
Aquitanians would have taken up arms on behalf of their
alien neighbours but Caesar may not have been aware of
;

the want of sympathy between the two peoples and, with

;

or without provocation, he would of course have compelled

the former as well as the latter to acknowledge the supremacy
of Rome. The Veneti and their allies, who saw that they Prepara-
had irretrievably committed themselves, were equally active.^ ^^^^
They repaired and provisioned their fortresses, assembled Veneti.

their ships in the Venetian ports, and even sent across the
Channel to ask for help. But more significant of the alarm
Avhich Caesar's designs had aroused was that they succeeded
in securing the alliance of the Morini and the Menapii, two —
Belgic tribes whose territory was four hundred miles from
their own, but who commanded the north-eastern coast,
from which he would have to embark for Britain. The
Veneti knew the strength of their country, and had little
doubt of success. The coast of the Morbihan was pierced
by long estuaries and broken by numerous inlets, which
would greatly hinder the progress of an invading army.
Little corn was grown in those parts and the granaries
;

had been emptied to supply the forts. Want of food there-

fore must soon force the Romans to beat a retreat and, if;

the worst came to the worst, those born sailors knew that
they could take to the stout ships which had w^eathered so
many storms while the frail Roman vessels would be sure
;

were encamped elsewhere (see B. G., ii, 34, 35, § 3; iii, 11, § 2 ; and p. 84)
and why should Caesar have taken this very roundabout way of getting to
Venetia ? Surely he could have sent orders to Labienus {B. G., iii, 11, § 2)
instead of adding several hundred miles to the length of his march in order to
convey them in person.
M. A. Blanchet {Traite des monn. gauL, pp. 524-5, with which cf. 519-20)
^

infers from the discovery of forty-two Aremorican hoards of late coins that
the Veneti and their allies buried their treasure in view of the campaign.
 90 CAMPAIGNS AGAINST THE CHAP.

56 B.C. to run aground among the shoals, or to founder m the

tempestuous seas that buffeted the rock-bound shore.
The The Roman fleet, which included ships impressed from the
Roman
fleet maritime tribes ^ between the Loire and the Garonne, was
weather- soon assembled, under Decimus Binitus, in the estuary of
bound in
the Loire. the Loire but the weather was too stormy for it to put to
;

sea. Meanwhile Caesar crossed the river Vilaine and entered

the Morbihan, hoping, by the time the gales moderated, to
Caesar's get possession of the enemy's strongholds. This, hoAvever, as
fruitless
campaign he soon found, was a work of extreme difficulty. The forts
against were situated at the ends of spits or promontories, connected
the
Veneti. with the mainland by shoals, which, at high tide, were com-
pletely submerged. Caesar constructed dykes across the
shoals, along which the troops marched to attack the town.
Before they could deliver the assault, however, the garrison
took to their ships, and sailed away to the nearest fort. The
greater part of tlie summer was frittered away in these
tedious sieges and Caesar was obhged to confess that all his
;

labour had been expended in vain. Accordingly he resolved

to wait for his fleet, and encamped on the heights of St.
Gildas, south-east of Quiberon Bay. Hard by, in the river
Auray, which discharges itself into the bay, the whole
Venetian armada was assembled.^
Sea-figlit At length the wind moderated ; and one morning the
between long-looked-for fleet was descried in the offing. Forthwith,
the
Veneti ghding out from the mouth of the Auray, appeared the
and
BrntuR.
hostile squadron, numbering two hundred and twenty sail.
They stood out of the water like floating castles. The great
sails were made, not of canvas but of leather, to withstand

the force of the Atlantic gales. Clustering on the cliffs, the

legionaries had a good view of the two fleets as they ap-
proached one another. Brutus and his officers were at their
wits' ends to know what to do. The rams of the fight galleys
would fail to make any impression on those huge hulls. The

1 M. Jullian {Hist, de la Gaule, iii, 292) thinks

The Pictones and Santoni.
that the ships which Caesar borrowed from these two tribes were used
for transport [of grain ?] and auxiliary services ;but is it not likely that
he thought it advisable to include in his armada some vessels of native
build V

-
See pp. 670-8").
 -

IV MARITIME TRIBES AND THE AQUITANI 91

deck-turrets were run but even then the Romans were

up ; 56 b. c
overtopped by the lofty poops, and could not throw their
javelins with effect. But the Roman engineers had prepared
an ingenious contrivance. Two or more galleys rowed up
close to one of the enemy's ships. Then, with sharp hooks
fixed to the ends of long poles, the Romans caught hold of
the halyards, and pulled them taut the rowers plied their
:

oars with might and main and the sudden strain snapped
;

the ropes. Down fell the yards the troops clambered on to

:

the helpless hulk and the struggle was soon ended by the
;

short sword. When several ships had been thus captured,

the rest prepared to escape. But they had hardly been put
before the wind when there was a dead calm and, as they ;

had no oars,^ they could not stir. The swift little galle3^s ran
in and out among them, and captured them one after another.
When the evening breeze sprang up, a few slipped away in
the dusk, and ran for the shore but all the rest were taken.-
;

This battle decided the war. All the chiefs and all the Punish
°
warriors of western Brittany had taken part in it. They ^^^^
had no reserves. They had staked everything upon a single Venoti.

throw, and had lost. Deprived of their ships, the survivors

had no means of defending their forts. There was nothing
for them therefore but unconditional surrender. They had
made a very gallant fight for freedom and Caesar respected ;

a brave enemy but he always took the straightest path to

:

gain his end. He

determined to teach the whole Gallic
people, by a terrible lesson, that it was dangerous to rebel.
As the Venetian senate were responsible for the act of
violence which had led to the war, every man of them was
put to death and all the rest of the tribe, or all that could
;

be caught, were sold into slavery.

About the same time dispatches arrived from Sabinus. His Cam-
force consisted of three legions, probably about twelve yabmiis
thousand men, with a due proportion of cavalry, archers, against
and slingers. The allies of the Veneti, commanded by a chief northern
named Viridovix, had mustered in the peninsula of the ^j^^^^^^
Veneti.
Except perhaps sweeps, which they may have used occasionally
^ to help
them in tacking. See Class. Quarterly, iii, 1909, p. .37.
- See pp. 230-7.
 92 CAMPAIGNS AGAINST THE CHAP.

56 B. c. Cotentin. The tribes of the Calv^ados and the Eure, in their

[The feverish eagerness for war, had massacred their senators,
Lexovii
and the simply because they counselled peace. Bandits and des-
Aulerci peradoes from every part of Gaul flocked to join the host.
Eburo-
vices.] Sabinus encamped on a hill and, having a wholesome
;

come
respect for their numbers, he could not be provoked to
out and fight. The enemy put him down as a coward, and
his own men grumbled at his inaction. But he was simply
biding his time. He bribed a Gaul belonging to his auxiliary
corps to go over to the enemy, in the guise of a deserter, and
tell them that Caesar was and that he himself
in great straits,
was on the point of going to his assistance. The man had
a ready wit and a glib tongue, and played his part well. The
Gauls eagerly swallowed the tale, and clamoured to be led
to the attack. Their commissariat had, as usual, been
neglected and they were impatient to finish the campaign
;

at a blow. Viridovix and his brother chiefs were obliged to

let them have their way. Their plan was to fall upon the
Romans before they had time to man the ramparts. The
ascent from the plain to the camp was about a mile. The
Gauls ran up the slope at the top of their speed, each man
carrying an armful of brushwood to fill up the trench. But
Sabinus was ready for them. Sallying from the right and the
left gate,^ the disciplined cohorts fell upon the flanks of the

panting multitude, and sent them flying. The cavalry allowed

few to escape. No second blow was needed. The league fell
to pieces at once. As inconstant as they had been impetuous,
the tribes abandoned the struggle, and laid down their arms-
BrilUant Meanwhile Crassus Avas carrying all before him in Aqui-
campaign
of Crassus tania. He was determined not to forfeit the reputation
in Aqni- which he had gained in the battle with Ariovistus, and he
tania.
remembered that tAvo Roman generals, Lucius Valerius
Praeconinus and Lucius Manlius, had suffered defeat ^ in the
country which was to be the theatre of the campaign. Before
crossing the Garonne, he halted to complete his preparations
in the fertile lands of the Nitiobroges ^ ; and, unlike Galba, he

1 See Long's Caesar, p. 170, note. ' In the war with Sertorius.

^ See Rev. de Gascogne, xxxvi, 1895, pp. 230-4, 240, and B. G., iii, 20, § 2 ;

vii, 31, § 5.
IV MARITIME TRIBES AND THE AQUITANT 93

took the greatest pains to ensure the regular deKvery of 5G b.c.

suppHes. Caesar had only been able to spare him twelve

cohorts, or about five thousand men but he had a powerful
:

body of cavalry, which he now strengthened by fresh levies ;

and he also raised auxiliaries and summoned a number of

Roman veterans from Tolosa, Carcaso,^ and Narbo to join [Toulouse,
. Car-
liim. Crossing the Garonne, he defeated the Sotiates near casonne,
the source of the Ciron, and captured their stronghold, the ^"^ N^"
site of which is now occupied by the town of Sos. Thence
he penetrated into the basin of the Adour. The Aquitanians,
in great alarm, obtained reinforcements from their kinsmen,
the Iberians of the Pyrenees.^ The leaders who were chosen
had learned the art of war under the famous Sertorius, and
their operations showed some degree of skill. They carefully
selected a position for their encampment, and fortified it in
the orthodox fashion. They sent out detachments to block
the roads. Relying on their numbers, which were daily
augmented, they hoped to gain a bloodless victory by cutting
off the invader's supplies, and harassing his rear as soon as

he should be obliged to retreat. But Crassus had no inten-

tion of retreating. He could not spare a man to secure his
supplies, but he knew that sheer audacity will often work
wonders. His men were in great heart, emboldened by the
enemy's inaction, and confident in their young leader.
Having offered battle in vain, he boldly assaulted the
enemy's camp. They resisted stoutly, and threw their
javelins from the high rampart with great effect but they :

had neglected to secure the rear gate and some fresh;

In the $ MSS. the text stands Tolosa Carcasone et Narbona

^ in the a l^olosa
;

et Narbonae. H. Meusel, following /3, formerly supplied et for grammatical

reasons before Carcasone ; but now, influenced by Mommsen {Jahresb. d. philol.
Vereins zu Berlin, xx, 1894, p, 203), who thinks that the absence of et in {3 proves
that Carcasone was interpolated, he omits that name. A. Klotz, however
{Caesarstudien, p. 35, n. 3), very acutely remarks that x (see p. 202, infra) has
Tolosae instead of Tolosa, which suggests that in a something has dropped out.
I have no doubt that Caesar wrote et Carcasone.
* M. Jullian {Hist, de la Gaule, iii,
307, n. 2) believes that Caesar postdates
the preparations of the tribes of Southern Aquitania and their Spanish neigh-
bours against Crassus. Do not the opening sentences of B. 0., iii, 23, in which
—
Caesar states not only the time but also the motive the alarm of the tribes
on hearing that Crassus had captured the strong fort of the Sotiates militate —
against this view ?
 94 MARITIME TRIBES AND THE AQUfTANI
56 B.C. cohorts managed to get round by a circuitous way, break
down the feeble defences, and steal in unobserved while the
battle was raging at the opposite end. The imprisoned
Aquitanians and Spaniards rushed pell-mell out of the
entrenchment, and made a desperate effort to escape but :

the country was one vast open plain and they were ridden
;

down and slaughtered in thousands. Forthwith all except

the remoter tribes tendered their submission, and voluntarily
sent hostages.
FruiilesH The conquest of the maritime peoples was all but com-
of Caesar^ plete. The Morini and the Menapii —
those Belgic tribes who
against
the
j^^d —
formed an alliance with the Veneti alone refused to
Morini. Submit. Their country, which extended from the neighbour-
hood of Etaples to the lower Rhine, comprised the northern
parts of the Pas-de-Calais and the Nord, Flanders, Zeeland,
and North Brabant. Caesar had over four hundred miles
to march, and the summer was nearly at an end but he
;

felt confident that he would be able to subdue the recalci-

trant tribes in one brief campaign. He traversed Brittany

and Normandy, joining Sabinus on the way crossed the
;

Seine and the Somme and then pushed northward through

;

Artois. Tauglit by the sad experience of their impetuous

countrymen to avoid a pitched battle, the Morini sought
refuge, on the approach of the legions, in their vast forests.
While the legionaries were fortifying their camp, the enemy,
who had not yet been seen, suddenly dashed out of the woods
and attacked them and although they were driven back
;

Avith heavy loss, a few Romans, who chased them too far,
were cut off and killed. This mishap made the legionaries
more careful. They spent some days in cutting down the
trees, piling them up on both flanks, as they advanced, to
guard against surprise. The enemy's cattle and part of their
baggage fell into their hands. But now the wind blew and
the rain fell with such violence that the work of felhng the
trees had to be suspended the troops could no longer live
:

safely in tents and it was necessary to abandon the cam-

;

paign. The cultivated lands of the Morini Avere harried and

their liamlets burned and the legions returned to winter in
;

the newly conquered districts between the Seine and the Loire
 CHAPTER V
THE MASSACRE OF THE USTPETES AND
TENCTERI
Gaul was now, to all appearance, conquered. Throughout o5 b.c.

these three years the central tribes, influenced by the p^^^^s and
example of the Aedui, distracted by intestine rivalries, awed Tencteri

by the genius of the Roman Governor, had remained simply Qaul.^

passive. But it was not enough merely to conquer the :

conquest had also to be secured against foreign invasion. A

fresh incursion of hungry Germans was imminent. The
defeat of Ariovistus had alarmed the Teutonic races but ;

it had not stilled the inward throes by which they had so

long been convulsed. The Suebi had swept before them

the lesser tribes of the Usij)etes and Tencteri a land to
:

dwell in and food to eat the fugitives must needs obtain ;

and now, after three years' wandering, a vast horde of

emigrants appeared in the neighbourhood of Emmerich, on
the right bank of the lower Rhine. ^ The Menapii occupied
lands on both banks of the river. Those who dwelt on the
right bank, terrified by the appearance of the huge host,
hurriedly abandoned their villages, crossed to the southern
side, and, joining their kinsmen, prepared to dispute the
passage. Baffled in their attempts to cross, the Germans
made a feigned retreat, which lasted three days then :

marched rapidly back surprised and massacred the

;

Menapii, who had returned seized their boats and crossed

;

over and for the rest of the winter lived at free quarters in
;

the Menapian territory on the west of the Rhine.

The news reached Caesar in Cisalpine Gaul, while he was
discharging the civil duties of his government. He knew the
—
character of the Gauls, the frivolity and craving for excite-
ment that impelled them to rush blindly into new con-
1 See pp. 689-90.
 96 . THE MASSACRE OF THE chap.

55 B.C. nexions without counting the cost. There was, indeed, no

Caesar
reason why they should trouble themselves to repel
'^ ^ one
fears that .

some of invader for the benefit of another. But the chances were that
Gallic
some of the tribes might be impelled by jealousy of their
tribes may rivals or hostility to the Romans to welcome the new-comers.
"'
Determined to prevent such a coalition or crush it in the
He re- making, Caesar returned to Gaul earlier than usual, and
(lauf and
P^^o^eeded to join the legions, which had concentrated at
summons some point near the lower Seine, probably in the neighbour-

council, bood of Evreux.^ His apprehensions were justified. Certain

tribes had entered into negotiations with the Germans and ;

they had by this time moved as far southward as the terri-

tories of the Eburones and the Condrusi. The former
included portions of the provinces of Limbourg and Liege :

the Condrusi inhabited the district of Condroz, between the

Meuse and the Ourthe. Caesar summoned the Gallic chiefs,
including those who had committed themselves, to a council
and, pretending to be ignorant of the negotiations, told
them that he was going to make war upon the common
enemy, and called upon them to furnish their regular con-
tingents of cavalry. When the contingents arrived, he made
a selection from the whole number, and, having provided
He marched towards the distant
for the delivery of his supplies,

aga^inst*
country in which he heard that the Germans were encamped.
the Usi- It is impossible to say where he crossed the Meuse, or what

Tencteri; route lie but the general trend of his

followed afterwards ;

march was towards the neighbourhood of Coblenz. Appa-

rently the Germans were in no aggressive mood. Tired of
their enforced wanderings, they only wanted to settle down
andnego- peaceably in some fertile part of Gaul. When Caesar was
still some days' march from their encampment, their envoys
tlTe^TeT
voys. met him. The Germans, they said, had no desire to fight ;

but, if Caesar attacked them, they would not flinch. All

they asked was that he should assign them lands, or at all
events leave them to enjoy those which their swords had
^ The legions had wintered '
in the territories of the Aulerci, the Lexovii, and
the other tribes which had recently been in arms '
[B. G., iii, 29, § 3). M. Jullian
[Hist, de la Gaule, iii, 326, n. 2), assuming that Caesar met the Gallic council
at Samarobriva (Amiens), as he did in the following year {B. G., v, 24, § 1),

conjectures that the concentration was effected there.

V USIPETES AND TENCTERI 97

won. They acknowledged no superiors but the Suebi ; and 55b.c.

against the Suebi the gods themselves could not contend.
Caesar replied that he could make no terms with them while
they remained in Gaul. People who could not defend their
own country had no right to encroach upon others : besides,
there were no lands to spare in Gaul sufficiently extensive to
support so vast a multitude. They were welcome, however,
if they cared to recross the Rhine, to settle in the country

of the Ubii, who had just put themselves under the protec-
tion of —
Rome. The territory of this people the only German
tribe which had definitely submitted to Caesar ^ extended—
from the neighbourhood of Coblenz to the neighbourhood
of Bonn. The envoys said that they would refer Caesar's
proposal to their principals, and return with an answer in
three days. Till then they hoped that he would advance no
further. This request he rejected for he felt sure that it
;

was simply a pretext to gain time for the German cavalry,

who had crossed the Meuse in quest of corn and plunder, to
return.
Marching on steadily, he was only eleven miles from the
German head-quarters when the envoys returned. Again
they begged him to halt and again he refused. They then
;

asked for three days' grace, to arrange terms with the Ubii.
What they really wanted, as Caesar saw, was to gain more
time. He meant to do the same. He promised, however,
not to advance that day beyond a river, four miles distant,
where he intended to water and told them to come back
;

again on the morrow, that he might decide on their request,

and to bring with them as many of their leaders as could
come. What he desired was to get those leaders into his
power, so that their formidable host might be helpless in his
hands. Perhaps he knew that his offer to let the Germans
settle in the country of the Ubii was impracticable perhaps :

indeed he had only made that offer in order to gain time,

and to put the Germans off their guard certainly he
:

' No doubt
the Ubii were one of the (Jcrinan tribes whieh had sent eiivoyg
to Caesar two years before (see p. 82, and B. G., ii, 35, § I). As ho was on tiie
point of starting for Italy, he had then ordered the embassies to return in the
following spring but apparently the Ubii alone had obeyed.
;

1093 H
 98 THE MASSACRE OF THE CHAP.

55 B.C. believed that they were trying to outwit him, and hewas
—
determined to outwit them, determined, by hook or by
crook, to secure the essential object of ridding himseK
and Gaul of these dangerous immigrants, and to secure
it at the least possible cost to his own army. Mean-
while, at the urgent entreaty of the envoys, he sent orders
to his Gallic cavalry, who had gone on in advance, to refrain
from provoking a combat. The envoys took their leave.
The cavalry, five thousand strong, were riding quietly along,
on the faith of the truce, when, without a moment's warning,
Their a band of horsemen swept down, and scattered them right
cavalry,
in viola-
and left. As they tried to rally, the enemy leaped to the
tion of a ground, and stabbed their horses in the belly. An Aquitanian
truce,
attack his.
noble, named Piso, did his best to save the credit of the Galhc
cavalry, hazarding his life to rescue his brother, and when
he was unhorsed, fighting against desperate odds till he fell.
His brother, who had escaped, would not survive him, and
galloped back into the press to die. But their example was
wasted. The Gauls were six to one but they were thoroughly
:

unnerved and, while many lay dead, the rest galloped

;

away, and never drew rein till they came within sight of
the Roman column.
He re- made up his mind. Those Germans were treacherous
Caesar
solves to
attack savages and he saw no reason why he should make an}^
;

them at terms with them. Besides, this paltry triumph they had
once
stolen would make them heroes to the feather-pated Gauls.
To hold his hand until they were reinforced would be sheer
madness. Next morning the German chiefs came to his
—
camp, to apologize, as they said, for the unauthorized
attack by their cavalry. Caesar was delighted. He deter-
mined to end the business by a single blow, bloodlessly, for —
arrests his own men. He refused to hear what the chiefs had to say.
their
Believing, or professing to believe, that they only wanted to
chiefs,
who had cajole him into granting an extension of the truce, he ordered
come,
ostensibly them to be put under arrest, and then, placing his demoralized
to ox- cavalry in the rear, marched on rapidly against the Germans.
plain ;
They were taking their ease among their wagons, with their
wives and children, when the legions appeared. Confounded
by the sight, not knowing what had become of their leaders,
V USIPETES AND TENCTERI 99

they lost all presence of mind, and crying aloud in their 55b.c.
terror, ran hither and thither about the camp. The infuriated ^^^ ^^^^."'
n /-i T*'v anni- • 1
Eomans burst in. The few Germans who were quick enough hilates the
to seize their weapons, clustered behind the wagons and tried
to resist : but, distracted by piercing shrieks, they turned
and saw their wives and children flying before the em-
boldened cavalry ; and flinging aside
their arms, they rushed
pell-mell to overtake them. Many were slain in the pursuit.
Others scattered over the country and escaped. At length
the panting remnant reached the confluence of the Moselle
and the Rhine. ^ Worn out and desperate, they plunged in ;

and the swift current swept them away.

The conduct of Caesar was fiercely condemned by Cato Hiscon-
and others in the Roman Senate. The refusal to listen ^emneT"
to the explanation of the German chiefs ; their detention, in the
contrary, as it appeared, to the law of nations ; and then
the virtual extermination of an entire people, —these things
perhaps shocked sensitive consciences, and certainly gave a
handle to political opponents. Cato actually proposed that
the perfidious Governor should be given up to the Germans.^
Caesar pursued his course unmoved, and emphasized in
his narrative the comprehensiveness of the massacre the :
— '

host of women and children began to fiee in all directions ;

and Caesar sent his cavalry to hunt them down.' The object
for which their blood was spilt was gained. Thoroughly
cowed, the Germans thenceforward ceased to disturb the
tranquillity of Gaul.
But Caesar determined to make assurance doubly sure. He bridges
As the Germans thought so little of crossing the Rhine, he ^^i^j^e
would cross it too, and teach them that invaders might in punishes
their turn be liable to invasion. Besides, it was necessary Sugambri,
to chastise the Sugambri, the northern neighbours of the and re-

Ubii, in whose country the cavalry of the Usipetes and Gaul.

Tenet eri had just found a ready welcome. When he sent
to demand their surrender, the Sugambrian chiefs asked with
what face he, who complained so loudly of the Germans'
crossing the Rhine, could claim the right to dictate to the

1 See pp. 691-706.

^ Plutarch, Caesar, 22 ; Suetonius, Divus lulius, 24.
H 2
 100 MASSACRE OF USIPETES AND TENCTERI
55 B.C. Germans in their own The Ubii, on the other
country.
hand, besought him to come and help them against the
Suebi : his prestige, they said, was so great that the mere
appearance of his army would be enough to secure them
from attack and they would gladly undertake to find boats
;

to cross the stream. But Caesar did not think it safe to

trust to boats and he intended to make the passage in a way
;

that would produce a greater moral effect. Broad, deep, and

swift as the river was, he would throw a bridge across it, to
teach the Germans what Roman science could effect. He
selected for the spot a site between Coblenz and Andernach,
which was opposite the territory of the Ubii.^ The Roman
engineers were accustomed to bridge rivers but this was an ;

undertaking of unprecedented difficulty. But Caesar had

inspired every man with faith in his star and all ranks ;

worked with extraordinary energy. Within ten days from

the time when the first tree was felled, the great river was
spanned by a firm bridge of piles, buttressed to withstand
the force of the flood ^ and the legions were encamped on
;

the German bank. Leaving a strong guard at either end,

Caesar marched rapidly northward against the Sugambri.
Their country extended eastward of Crefeld, Diisseldorf and ,

Cologne. Envoys from various tribes met Caesar on the way,

and solicited his friendship. He answered them courteously,
and directed them to bring hostages to his camp. The
Sugambri, on the advice of the Usipetes and Tencteri, had
taken refuge in the outlying forests and, after burning
;

their villages and cutting their corn, Caesar returned to the

country of the Ubii. The Suebi had sent their wives and
children into the secure recesses of the vast forest of Central
Germany, and were banded together somewhere in the heart
of their country, ready for battle. But Caesar had neither
the force nor the inchnation to undertake the conquest of
Germany. Having accomplished every object for which
—
he had entered the country punished his enemies, reassured
his friends, and made the name of Rome respected he —
recrossed the Rhine and destroyed his bridge.
1 Sec pi>. 70G-10. ' See pp. 711^24.

^/tftARV
 CHAPTER VI

THE DISASTER AT ATUATUCA AND ITS RESULTS

Caesar's attention was now diverted for a time from the 55-54 b. c

affairs of Gaul. During the few weeks of summer that Caesar's

invasions
followed his passage of the Rhine and the latter part of JJf^grit ain.
the ensuing season he made his two famous expeditions to
Britain. He went to lUyricum in the intervening winter,
to punish a tribe which had been making raids upon Roman
territory, and did not return to Gaul until the close of
the following May. Quintus Cicero, a younger brother of the
orator, joined him on the road, and took up the post of
a legatus. Caesar had offered him the appointment, and he
accepted it in the hope of securing his political position but ;

he had no zest for the work. He complained to his brother

that his duties were onerous yet he had leisure in camp
;

to compose four tragedies in sixteen days.^ Caesar himself

often found time to write to the elder Cicero, and even to
read his verses. Cicero, indeed, who had atoned for his
former opposition by celebrating Caesar's victories in his
speech On the consular Provinces ', was at this time more
'

friendly to him than he had ever been and Caesar, knowing ;

the value of his support, neglected no opportunity of con-

cihating him. Their correspondence shows us what manner
of men Caesar had to entertain in his army when friends or
asked favours of him. Cicero begged him
political associates
to give a place of some sort to a lawyer named Trebatius ;

and Caesar, who knew how to render such appointments

innocuous, good-naturedly consented in a letter, the kindli-
ness and the humour of which are reflected in one which
Cicero wrote to Trebatius himself .^

^ See Mr. H. W. Garrod's interesting paper in Journal of Philology, xxxi, 1908,

pp. 79-80.
2 Cicero, Fam., vii, 5-6, 8, 10, 18 ; Q. Jr., u, 13 (15 a) ; iii, 5-G, § 7 ; 8, § I.
 102 THE DISASTER AT ATUATUCA chap.

54 B. c. Caesar's avowed objects in invading Britain were to inform

himself about the island and its inhabitants, and to punish
the southern tribes, who had helped kinsmen in Gaul
their
to resist him. Fortunately several clans of the Morini, from
whose country he was obliged to embark, spontaneously
submitted before his first voyage and although they
;

attacked an isolated body of his troops on his return, they

were instantly punished by Labienus. The first expedition
was a mere reconnaissance the second, made with a far
:

stronger force, was planned for conquest and although this

;

task was too great for the force and the time which Caesar
could spare, the avowed objects were effectually gained,
the Britons ceased to abet the resistance of their kinsmen
beyond the strait. On each occasion Caesar left behind him
a force sufficient to keep open his communications and to
overawe intending rebels and on the second expedition he
;

took with him all the chiefs whom he had the slightest
reason to suspect. The one of all others whom he had been
most careful to summon was the notorious Dumnorix, who
was as popular with the masses and as determined an enemy
of Rome as when he had been detected in his intrigues with
the Helvetii. Quite recently he had caused great alarm and
indignation to the Aeduan council by giving out that Caesar
intended to make him king.^ Nothing could have provoked
Caesar more ; depended largely
for the success of his policy
upon his keeping the Aeduan government in good humour.
Dumnorix was most reluctant to leave the country. He
doubtless saw that he might never again have such an
opportunity as Caesar's absence afforded of furthering his
schemes ;and he begged for leave to stay behind. He was
terrified, he said, at the prospect of crossing the sea besides, :

he had religious duties, which he could not fulfil unless he

remained in Gaul.^ Caesar was of course deaf to his entreaties

1 Various writers have suggested that Caesar really had made the offer to
Dumnorix, in order to purchase his support. It seems to me more likely
that, as Schneider conjectures {Caesar, ii, 26), Dumnorix had made the state-
ment in question in order to exasperate the Aedui against Caesar. Still, I am
not indisposed to believe that Caesar may have thrown out some vague hint
which led Dumnorix to expect that if he proved himself loyal he would be
rewarded. - See Schneider's Caesar, ii, 27.
 —

VI AND ITS RESULTS 103

and pretended scruples, Dumnorix then tried to induce

his 54 b o
his brother chiefs to join him in refusing to go. He assured
them that Caesar was only taking them to Britain that he
might put them all to death, and called upon them to swear
that they would unite with him in working for the interest of
Gaul. Caesar kept himself informed of his intrigues, and did
his best to prevent him from rushing on his doom. All this
time the was weather-bound in the Port as Itius, which
fleet

may possibly have lain between Cape Blanc-Nez and Cape

Gris-Nez, although one would not hesitate to identify it
with the harbour of Boulogne, from which Caesar sailed on
his first expedition, if his narrative did not suggest that
he had chosen a new point of departure, and if it were
certain that his enormous fleet could have cleared the
estuary without undue delay .^ At length the wind shifted ;

and Dumnorix took advantage of the confusion that attended

the embarkation to ride off with the Aeduan cavalry.
Instantly stopping the embarkation, Caesar sent a strong
body of horse in pursuit with orders to kill him at once if
he attempted to resist. He fought desperately for life and His fate,

liberty but the troopers failed to support him and he fell,

: ;

passionately asserting with his dying breath the independence

of his tribe.
The death was a temporary The Gallic
of this resolute adventurer
relief to the Roman Governor probably
r J helped
r to ^^^^^^ '^ ^
;
'
but it
dangerous
kindle into a flame the discontent which had long been mood.
smouldering in the breasts of the Gauls. Doubtless the
Aedui were glad enough to be rid of the Helvetii doubtless :

others besides the Aedui rejoiced at the overthrow of Ario-

vistus. But it was not to be expected that they should feel
any gratitude to Caesar. Individuals like Diviciacus, tribes
like the Remi, had of course gained something by his friend-
ship. But Gaul, as a whole, had so far gained nothing. Not
only were the constant presence of the legions and the end-
less requisitions of corn an intolerable burden,^ but to the

1 See pp. 432-8. The balance of probability is greatly in favour of Boulogne.

^ It may perhaps be inferred from B. 0., vii, 76, § 1 'pro quibus meritis
civitatem eius [Commu] i mm un em esse iusserat, iura legesque reddiderat — that
tribute also was exacted (see p. 838). Possibly, however, as M. Jullian remarks
{Hist, de la Gaule, iii, 278, n. 1), immunem may mean free from the burden of
 104 THE DISASTER AT ATUATUCA chap.

54 B.C. high-spirited Celtic knights the fact of subjection was more

gaUing still. They had indeed partly themselves to blame.
Weakness mutual jealousy, petty ambition had
of purpose,
been their bane. They had not realized, or had not valued
their national unity enough to make a united effort for its
preservation. The Nervii indeed had fought like heroes ;

but the bulk of the Belgae had been too selfish, too faint-
hearted, too distrustful of each other, above all, too feebly
organized to support them. The Veneti had made a gallant
resistance ; but the enthusiasm of their allies had vanished
at the first reverse. The states of the interior had acquiesced
in the domination of Caesar, without a blow, nay even
without a protest. It would, of course, be unjust to ignore
the difficulties with which they had to contend. If Caesar
was justified in the severity with which he criticized the
infirmities of their national character, it would have been
unreasonable to expect from a medley of tribes, which had
I
hardly had time to outgrow their political infancy, the
harmonious action which could only have been the fruit
of ages of discipline. Rome herself repelled invasion only
because fate deferred it until, favoured by geographical
position and after many centuries of struggle, she had been
able to unite the divers forces of Italy : the GaUic tribes
were attacked by a matured power while they were still in
the stage through which every great people has been forced
to travel before it could become a nation. They were heavily
weighted by the selfishness or the astuteness, call it which
one will, of the Aedui and the Remi and above all, no
;

leader had appeared whose personality was sufficiently com-

manding to rally the patriots of every state round his
standard. But the steady pressure of the conqueror was

paying tribute to some other Gallic tribe or may it mean free from the burden
;

of supplying corn ? What conclusion is to be drawn from the words iura

legesque reddiderai I cannot tell. Prima facie they would seem to imply that
Caesar had already begun to legislate for Gaul, which is hardly credible. They
clearly imply that the rights and laws' {iura legesque) of certain tribes had
'

been infringed, as they of course were in cases where oligarchy was superseded
in favour of a king. But perhaps the Atrebates, like the Remi {B. 0., ii,
3, § 5), had been subject to some other tribe; and if so, Caesar may only have
meant that he had restored their independence.
 ;

VI AND ITS RESULTS 105

producing its natural effect upon a people who, notwith- 543.0.

standing their intestine quarrels, were animated by the
recollection of ancient renown, who spoke one language, and
who venerated one divine ancestor. The chiefs were in
a dangerous mood and the populace were ready to back
;

them. Caesar was perfectly aware of their temper indeed :

he had returned prematurely from Britain because he was

warned that disturbances were impending. The harvest in Distribu-

Gaul this season was very scanty and he was obliged ; legions for
therefore, in order to ensure an adequate supply of grain, the winter

to distribute his legions for the winter over a wide extent

of territory. As the Belgic states appeared to be the most
restless, their country was selected for the occupation. One
legion, under Gains Fabius, was quartered among the Morini
another, under Quintus Cicero, who was allowed to choose,^
among the Nervii, in the neighbourhood of Namur a third, ;

under Labienus, not improbably at Mouzon, some ten miles

south-east of Sedan, near the western frontier of the Treveri.
Three, under Trebonius, Marcus Crassus,^ and Plancus respec-
tively, were stationed close together at Samarobriva and [Amiens.]

in the plain round Beauvais. One, consisting entirely of

recruits,^ with five veteran cohorts, was sent to Atuatuca,
in the country of the Eburones. The site of this famous camp
has never been determined for, although it is generally
;

identified with Tongres, Caesar's narrative would seem to

imply that it was east of the Meuse, and not very far from
Liege.* The garrison was commanded by Sabinus and
Aurunculeius Cotta, the former of whom, as the senior officer,
had the superior authority.^ One legion only, under Roscius,
was sent outside Belgic territory to the country of the
Esuvii, in the Orne. The services of the younger Crassus,
one of the most capable of Caesar's officers, who was very
near his heart, and whose noble nature captivated Cicero,^
were no longer available he had accompanied his father,
:

the triumvir, to Syria, where he was destined to perish on

^ Cicero, AtL, iv, 19, § 2.

2 Marcus Crassus was Caesar's quaestor and an elder brother of Publius.
'
See p. 732, n. 1. * See pp. 371-84. s
See p. 720.
« Fam., xiii, 16, § 1 ; Brulus, 81, § 282.
 106 THE DISASTER AT ATUATUCA CHAP.

54 B. c. the field of Carrhae. Caesar fixed his head-quarters at

Samarobriva ; and in view of the prevailing discontent, he
determined not to leave Gaul for the winter until the various
camps were fortified.
Divide et About this time an incident occurred which Caesar may-
impera.
have regarded as a sign of a coming storm. His motto was.

Divide et irapera. The Aedui and the Remi had both been
faithful to him and with the object of strengthening their
;

influence and thereby diminishing the chances of revolt, he

had always treated them with distinction. Moreover, he
had elevated chiefs who had done him service to the thrones
of their ancestors in states where monarchy had been over-
thrown by oligarchy his object doubtless being not only
;

to put a premium upon loyalty, but also to use the loyal as

instruments for keeping the anti-Roman party in check.
Assassina- One of his nominees, Tasgetius, had, for three years, been
tion of
King king of the Carnutes, a tribe which dwelt in the country
Tasgetius, round Orleans and Chart res. How he used his power, we
Caesar's
nominee, are not told ;but soon after Caesar's return from Britain
by the he was assassinated. Caesar instantly sent Plancus with his
Carnutes.
legion, to arrest all who were concerned in the deed, and to
terrorize intending rebels.
Intrigues All this time one chief in particular, whose pride Caesar
of Indu-
tiomarus had humbled, was busily intriguing against him. In the
against spring of every year he convened a diet of the Galhc chief-
Caesar.
tains, partly, it should seem, to test their temper, partly
to fix the strength of the cavalry contingents wliich their
respective tribes were to provide. Since the battle with the
Nervii, the Treveri, whose cavalry had witnessed the desperate
struggle of his legions, had refused to send their representa-
tives and it
; was said that they were intriguing with the
Germans. Just before the second expedition to Britain,
Caesar entered their country at the head of a strong
force with the view of re-establishing his authority. Two
chiefs, Cingetorix and his father-in-law, Indutiomarus, were
struggling for supremacy. Cingetorix at once presented
himself before Caesar, promised fidelity to Rome, and gave
full information of what was going on in the country.
1 See pp. 521-3.
VI AND ITS RESULTS 107

Indutiomarus collected levies, and prepared to fight. Many 54 b.c.

of the leading men, however, influenced by Cingetorix and
appreciating the power of the legions, came into Caesar's
camp and made terms for themselves. Indutiomarus soon
found that he had miscalculated his strength, and hastened
to excuse himself. Caesar, who had no time to spare, con-
tented himself with taking hostages for his good behaviour.
At the same time he of course did everything to strengthen
the influence of his supporter ; and Indutiomarus smarted
under the feeling that his credit with his countrymen was
gone. It is probable that during Caesar's absence he was
concocting schemes of revenge. The isolation of the various
camps gave him his opportunity. A few days after the
legions had taken up their quarters he instigated Ambiorix
and Catuvolcus, each of whom ruled one half of the country
of the Eburones, to attack the camp of Sabinus and Cotta.
Caesar was a hundred and fifty miles away the nearest
:

camp, that of Cicero, at least forty-five miles at Atuatuca

:

there were barely six thousand legionaries, all told, and

two-thirds of them were unseasoned men. Success seemed
certain. Ambiorix and Catuvolcus, who had only just taken
their quota of corn to the generals, mustered their tribesmen
in great force, surprised and overpowered a fatigue party,
who were engaged in felling wood outside the camp, and then
made a sudden onslaught upon the camp itself. But the The
camp was strongly fortified, and stood upon rising ground under
of great natural strength. The troops promptly manned the Ambiorix,
rampart a squadron of Spanish horse made a successful futile
:

^^
sally•^
and the assailants fell back in discomfiture. Their fj^^^^
;
'
the camp
leaders shouted out that they would like some one to come of Sabinus
and talk over matters, so that all disputes might be peace- ^^ ° ^'
ably settled. Two deputies accordingly, one of whom had
regularly conveyed instructions from Caesar to Ambiorix,
were sent out to hear what they had to say. Ambiorix had
made himself useful as a political agent and, in acknow-
;

ledgement of his services, Caesar had relieved him from the

burden of paying tribute to the Atuatuci, and had restored
to him his son and nephew, whom they had detained as
hostages. Ambiorix began by speaking of Caesar's kindness,
 108 THE DISASTER AT ATUATUCA chap.

54 B.C. and said that he was most anxious to prove his gratitude.
Ambionx jj^ protested that he had not attacked the camp of his own
advises ^ ^
^
^
Sabinus to free will,but simply because he could not resist the pressure
S one oT P^^ upon him by his tribesmen. Nor would they have stirred
the nearer if they had not been forced to join in the national movement.

His very weakness proved that he was speaking the truth.

He was not such a fool as to imagine that his feeble levies
could stand against the Romans. But the leading powers
of Gaul were banded together to recover their independence ;

and on that very day all the Roman camps were to be simul-
taneously attacked. He most earnestly entreated Sabinus
to be on his guard. A
German mercenaries had
host of
crossed the Rhine, and would be upon him in a couple of
days. If the two generals would take his advice, they would
abandon their camp at once, and make the best of their way
to the quarters of Cicero or of Labienus. He would pledge
his word that they should not be molested on the road. He
would not merely be making some return for Caesar's kind-
ness :it was to the interest of his people to be relieved from

the burden of supplying the camp.

The advice The deputies returned to camp, and reported what they
inacoim- ^^^ heard. Sabinus and Cotta were inclined to think that,
oil of war. whether Ambiorix were sincere or not in his professions of
friendship, his warning was not to be despised. One thing
was certain —
a single petty tribe like the Eburones would
:

never have dared to pit itself against the power of Rome

unless it had been strongly supported. The tribunes and
centurions of the first rank ^ were summoned to attend
a council of war. It took place in the middle of the camp,
in full view of the soldiers. Cotta spoke first. He argued
that, without Caesar's express command, they had no right
to leave the camp. Behind its defences they could defy any
force that could be brought against them. Had they not
already beaten off the enemy, and inflicted heavy loss upon i

them into the bargain ? They were not pressed for supplies ; i
and doubtless they would soon be relieved. Anyhow, nothing I
could be more unsoldierlike, more puerile, than to take a step
fraught with the gravest issues, by the advice of an enemy,
j

1 See pp. 567-79.

VI AND ITS RESULTS 109

Most warmly supported this view. But

of the officers 54 b. c.

Sabinus was only irritated by their unanimity. Speaking

loudly and passionately, he insisted that it was not a question
of being guided by the advice of an enemy, but by hard facts.
Caesar had doubtless gone back to Italy, or the Eburones
would never have attacked them so they need not expect:

help from him. The Rhine was close by. Both Germans
and Gauls had many an old score to wipe out and they ;

were naturally burning for revenge. The course which he

recommended was safe either way. If the whole thing turned
out to be a false alarm, then they risked nothing by going to
the nearest camp. If, on the other hand, Gauls and Germans
were really leagued against them, their one chance of safety
was to retreat at once. To follow Cotta's advice would in-
volve, at the best, the miseries of famine and blockade.
The dispute waxed warm. In spite of all that Sabinus
could say, Cotta and the centurions remained inflexible.
Sabinus rapidly lost all patience. Raising his voice so that
the men might hear, Have your own way,' he shouted,
'

'
have your own way Death has no terrors for me
! These !

men will judge between us, and, if any disaster happens,

they'll call you to account for it. If you would only let them,
they could reach the nearest camp the day after to-morrow,
and join hands with their comrades.' The generals stood up.
Their friends crowded round them, took them by the hand,
and entreated them not to quarrel. Go or stay, all would be
well if only they could agree. The strife of words was pro-
longed till midnight. At length, overborne by the authority in spite of
of his senior, Cotta gave up his point. All ranks were
Jg^^J'^'J"
warned that they would have to quit the camp at dawn. The Cotta,
soldiers spent the small hours in looking over their belong- decides to
ings to see what they could carry away, and told each other abandon
that, after all, Sabinus was in the right. They thought,' camp,
'

wrote Caesar, of every argument to persuade themselves

'

that they could not remain without danger, and that the
danger would be increased by protracted watches and conse-
quent exhaustion.' ^ The drivers had enough to do in loading
their cattle. Everybody was too agitated to think of sleep.
' B. G., V, 31, § 5. See pp. 72G-7.
 110 THE DISASTER AT ATUATUCA chap.

54 B.C. Meanwhile Ambiorix and his followers, hearing the hum

of voices in the camp, concluded that the Romans had deter-
mined to follow their advice. Whether Sabinus intended to
make for the camp of Labienus or for that of Cicero, the first
stage of his route would be the same.^ Ambiorix prepared to
execute his plan.
The Just as day was breaking, the Romans marched out of
march camp, an extended column encumbered by a heavy
in
^^^-
baggage-train. It seemed as if Sabinus had implicit confi-
dence in the good faith of Ambiorix for he could not have ;

adopted a more dangerous formation. He had decided to

make for the camp of Cicero. ^ After marching about two
miles, the head of the column plunged into a defile shut in
They are between wooded hills. Company after company tramped
rounded after. The last was just entering the valley when, rushing
by the from the woods, the Gauls threw themselves upon the van-
'
guard the rear was hustled forward
: before, behind, to :

right, to left, everywhere the enemy's masses were pouring

down. Sabinus hurried about from place to place, and
feebly attempted to make his dispositions. Cool and col-
lected, Cotta did his best to rally the men ; and, as the
length of the column made
unmanageable, he agreed with it

his colleague to abandon the baggage, and form in a hollow

squa,re.^ It was perhaps the only course to adopt yet the :

result was that the Romans lost heart, and the enemy were
emboldened for both knew that such an expedient could
;

only have been resorted to by leaders who despaired. Rough

soldiers were actually weeping confusion was worse con- :

founded and many contrived to slip away, and ran to save

;

their valuables in the baggage-train while there was yet

time. The Gauls, on the other hand, showed extraordinary
steadiness for their leaders told them that they had only to
;

win the battle, and they should have plunder to their hearts'
content. Still the square remained unbroken. Now and
again a cohort dashed out and beneath their short swords
;

many of the Gauls sank down. Observing this, Ambiorix

ordered his men to fall back some paces, and hurl their
'
See p. 373. ^ See
p. 373 and n. o.
^ The term '
square '
is ussd loosely. See the note on Orbis, p. 728.
 —

VI AND ITS RESULTS 111

missilesfrom a safe distance. He reminded them that they 54 b. c.

were in good training, and with their light equipment could

easily keep out of harm's way.^ If the Romans charged

them, they were to retreat when the Romans attempted

:

to return to the square, they were to pursue. Maddened by

the volleys which they were powerless to return for they —
—
had no slingers and no archers one cohort and then another
charged. Back darted the nimble Gauls. The right flank
of the Romans was exposed, and missiles rained in on their
unshielded bodies. The moment the baffled cohort retired,
the enemy swarmed all round it and then followed a swift
;

butchery. The rest stood shoulder to shoulder in the square :

but now their courage was of no avail the enemy would not :

come to close quarters and stones and arrows made havoc

;

in the dense ranks. Yet, facing such fearful odds, after seven
hours' fighting, they still held out ; and, as Caesar put it,

throughout that trying time they did nothing unworthy of

themselves. Quintus Lucanius, a centurion whom Caesar
singled out for special mention, was
attempting to killed in
rescue his own son. Cotta himself was struck in the face
as he was cheering on the men. The sun was sinking. The
battle could only end in one way and Sabinus, catching
;

sight of Ambiorix some way off haranguing his men, sent

his interpreter to ask for quarter. Ambiorix replied that
Sabinus might come and speak to him if he liked he would :

answer for his personal safety and he hoped that his men
;

might be prevailed upon to be merciful. Sabinus asked Cotta

to go with him but Cotta, true to Roman traditions, said that
;

nothing would induce him to treat with an armed enemy.

^ The words upon which this sentence is founded levitate armorum ei

cotidiana exercitatione nihil his noceri posse {B. G., v, 34, § 4) are certainly- —
suspicious. Levitas, says Meusel {Jahresb. d. philol. Vereins zu Berlin, xxxvi,
1910, p. 52), is not used in classical prose in the sense of lightness (of weight),
and, even if it were, Caesar would have written not levitate,but propter levi-
tatem. The former reason appears to me weak : brevitas,which Caesar (ii, .30,
§4) unquestionably uses in the sense of shortness (of stature), does not occur
elsewhere in that sense before the time of the elder Pliny (Thesaurus ling. Lat.,
ii, 2189-90), though Livy once (viii,
19, § 8) speaks of hrevitate loci; and when
I read B. G., ii, 1, § 3, iii, 21, § 3, I doubt whether Caesar would necessarily
have written propter levitatem. Still, as Meusel argues, the words levitate . . .

posse might well be dispensed with.

 112 THE DISASTER AT ATUATUCA chap.

54 B. c. Accordingly Sabinus and a few tribunes and centurions went

out alone. Sabinus was ordered to lay down his arms, and
obeyed, bidding his officers to do the same. A parley fol-
lowed ;and Ambiorix purposely spun out what he had to
say. While he was speaking, a number of Gauls crept
stealthily behind Sabinus ; and in a momejit he fell dead.
Then with a yell of triumph the Gauls rushed into the
exhausted legion ;and Cotta and the bulk of his men were
destroyed. The rest fled for the camp. The standard-
bearer, Lucius Petrosidius, finding himself hotly pursued,
flung his eagle inside the rampart, and died fighting like
and virtu- a Roman soldier. His surviving comrades defended them-
hilated.^^
selves till nightfall. Then, seeing that hope was gone, they
fell upon each others' swords.

A handful of men, more fortunate than their comrades,

had managed to escape into the woods. They made their
way to the camp of Labienus, and told him the whole story.
Ambiorix Ambiorix instantly followed up his victory. Bidding his
th?Atu^^
infantry follow, he rode off westward with the horsemen,
atuci and All that night and the day after he sped over the plateau of

join him in Herve and the plain of Hesbaye : just pausing to enlist the
attacking Atuatuci in the cause, he pressed on, and next day crossed
Q. Cicero.
' r '

^
the frontier of the Nervii. This people had not forgotten
how their brethren had been slaughtered, three years before,
on the banks of the Sambre. Ambiorix told the chiefs
exult ingly of his success. Here was such a chance as they
might never have again. Cicero's camp was close by. Why
—
should they not do as he had done, swoop down upon the
solitary legion, win back their independence for good, and
take a glorious revenge upon their persecutors. The chiefs
caught at the suggestion. The small tribes that owned their
sway flocked to join them : the Eburones, flushed with
victory, were there to help ; and the united host set out
with eager confidence for the Roman camp. Their horsemen,
hurrying on ahead, cut off a party of soldiers who were felluig
wood. Not the faintest rumour of the late disaster had
Siege of reached Cicero ; and the Gallic hordes burst upon him like
Cicero's a bolt from the sky. Their first onslaught was so violent
that even the disciplined courage of the Romans barely
VI AND ITS RESULTS 113

averted destruction. Messengers were instantly dispatched 54b.c.

to carry the news to Caesar and Cicero promised to reward

;

them well if they should succeed in delivering his letters.

Working all night with incessant energy, the legionaries
erected a large number of wooden towers on the rampart,
and made good the defects in the fortifications. The Gauls,
who meanwhile had been strongly reinforced, returned in the
morning to the attack. They succeeded in filling up the
trench but the garrison still managed to keep them at bay.
;

Day after day the siege continued and night after night
;

and all night long the Romans toiled to make ready for the
morrow's struggle. The towers were furnished with stories
and embattled breastworks of wattle-work :sharp stakes,
burnt and hardened at the ends, were prepared for hurling
at the besiegers, and huge pikes for stopping their rush if
they should attempt an assault. Even the sick and the
wounded had to lend a hand. Cicero himself was in poor
health but he worked night and day and it was not until
: ;

the men gathered round him and insisted on his sparing

himself, that he would take a little rest. His complaints, his
Epicurean studies, his abortive tragedies were forgotten he
;

remembered only that he was a Roman general. Meanwhile

the Nervian leaders, who had expected an easy triumph,
were becoming impatient. They asked Cicero to grant them
an interview. Some of them knew him personally and they
;

doubtless hoped that he would prove compliant. They

assailed him with the same arguments that Ambiorix had
found so successful with Sabinus. They tried to frighten
him by describing the massacre at Atuatuca, and assured
him that it was idle to hope for relief. But they would not
be hard upon him. All that they wanted was to stop the
inveterate custom of quartering the legions for the winter in
Gaul. If he and his army would only go, they might go in
peace whithersoever they pleased. Cicero calmly replied that
Romans never accepted terms from an armed enemy. They
must first lay down their arms then he would intercede
:

for them with Caesar. Caesar was always just, and would
doubtless grant their petition.
Disappointed though they were, the Gauls were not dis-
1093 T
 114 THE DISASTER AT ATUATUCA chap.

54 B. c. heartened. They determined to invest the camp in a scien-

tificmanner. From the experience of past campaigns they
had got a rough idea of the nature of Roman siege works ;

and now, with the quickness of their race, they proceeded

to imitate them. Some prisoners who had fallen into their
hands gave them hints. Having no proper tools, they were
obliged to cut the turf with their swords, and use their
hands and even their cloaks in piling the sods but the ;

workers swarmed in such prodigious numbers that in three

hours they had thrown up a rampart ten feet high and
nearly three miles in extent.^ They then proceeded, under
the guidance of the prisoners, to erect towers, and to make
sappers' huts, ladders, and poles fitted with hooks for tearing
down the rampart of the camp. The huts, which were
intended to protect the men who had to fill up the trench
and demolish the rampart, were partially closed in front, and
had sloping roofs, built of strong timbers, so as to resist the
crash of any stones which might be pitched on to them, and
probably covered with clay and raw hides, as a protection
against fire.^ On the seventh day of the siege there was
a great gale. The besiegers took advantage of it to fling
blazing darts and white-hot balls of clay,^ which lighted on
the straw thatch of the men's huts and the wind-swept ;

flames flew all over the enclosure. With a yell of exultation,

the enemy wheeled forward their towers and huts, and
planted their ladders in another moment they were swarm-
:

ing up but all along the rampart, their dark figures outlined
:

against the fiery background, the Romans were standing,

ready to hurl them down : harassed by showers of missiles,
half scorched by the fierce heat, regardless of the havoc
that the fiames were making in their property, every man of
them stood firm and hardly one so much as looked behind.
;

Their losses were heavier than on any previous day. The

Gauls too went down in scores for those in front could not ;

retreat because of the masses that pressed upon them from

behind. In one spot a tower was wheeled right up to the
rampart. The centurions of the 3rd cohort coolly withdi'ew
1 See pp. 728-9.
2 See Caesar, B. C, ii, 10, and pp. 609-10. ' See pp, 729-30.

I
VI AND ITS RESULTS 115

their men, and with voice and gesture dared the Gauls to 54b.c.

come on but none dared stir a step a shower of stones

: :

sent them flying and the deserted tower was set on fire.
;

Everywhere the result was the same. The assailants were

the bravest of the Gauls of death they had no fear but
: :

they had not the heart to hurl themselves upon that living
wall and, leaving their slain in heaps, they sullenly
;

withdrew.
Still the siege went on ; and to the wearied and weakened
legion its trials daily increased. Letters for Caesar were
sent out in more and more rapid succession. Some of the
messengers were caught in sight of the garrison, and tortured
to death. There was, however, in the camp a Nervian named
Vertico, who, just before the siege, had thrown himself upon
the protection of Cicero, and had been steadfastly true to
him. By lavish promises he induced one of his slaves to
face the dangers which to the messengers had proved Roman
fatal. The letter which he had to carry was fastened to
a javelin and concealed by the lashing. ^ He passed his A messen-

countrymen unnoticed, made his way safely to Samarobriva, cLero

carries a
and delivered his dispatch. None of the other messengers
had arrived and so close was the sympathy between the
; to Caesar.

peasants and the insurgents that Caesar had not heard a

rumour of the siege.
It was about four o'clock in the afternoon. Within a few Caesar
minutes messengers were spurring to the camps in the f^^^gjlg^e
surrounding country. Crassus was ordered to come in to Cicero.

Samarobriva at once, and take the General's place. It was

most important to leave Samarobriva in safe keeping for ;

there were collected the hostages of the various states, the

winter's supply of corn, the heavy baggage of the whole
army ,2 and the General's papers and accounts. Fabius was
^ B. G., V, 45, § 4. The reading of a (see p. 201) is in iaculo inligatas, of /3
iaculo inligatas. If, asLong suggests {Decline of the Roman Republic, iv, 227),
the letter was inserted in the hollowed shaft of a javelin, lashing {inligatas)
was obviously both superfluous and impossible.
^ Impedimenta exercitus [B. Q., v, 47, § 2). Perhaps the word 'material'
would be more accurate than heavy baggage '
for the troops at Atuatuca,
' ;

and doubtless also the legions in the other camps, had heavy baggage with
them. It is impossible to say with certainty what the impedimenta to which
Caesar alludes was but it may have included siege material.
;

12
 IIG THE DISASTER AT ATUATUCA chap.

54 B.C. to join Caesar on the road. Awent to Labienus,

letter
expressing the hope that he would be able to march direct
to the relief of the besieged camp but this able officer was
;

trusted to use his own discretion. Plancus and Roscius

were too far off to be able to help. About nine o'clock next
morning, hearing that Crassus was close at hand, Caesar set
out with Trebonius's legion and about four hundred cavalry.
No baggage-train accompanied the column the men carried :

all that they required upon their backs. The first march
was more than eighteen miles. Fabius joined his chief on
the way ;but Labienus did nob appear. An express came
from him instead, from which Caesar learned, for the first
time, the fate of Sabinus and Cotta. It is said that, in his
first burst of grief and wrath, he swore that he would not

shave his beard or cut his hair until he had avenged their
deaths.^ Labienus went on to say that he was himself hard
pressed by the Treveri, and thought it foolhardy to leave his
camp. Caesar approved his decision, though it left him with
barely seven thousand men. Everything now depended
upon speed. Passing through the Servian territory, Caesar
learned from some peasants who fell into his hands that
Cicero's situation was all but desperate immediately he
:

wrote a letter in Greek characters, assuring him of speedy

relief, and offered one of his Gallic horsemen a large reward

to deliver it. He told him, in case he should not be able to

get into the camp, to tie the letter to the thong of a javelin
and throw it Dreading the risk of apprehension, the
inside.
man did as Caesar had directed but the javelin stuck in
;

one of the towers, and remained unnoticed for two days.

A soldier then found it and took it to Cicero, who read
the letter to his exhausted troops. As they gazed over the
rampart, they saw clouds of smoke floating far away over
the west horizon, and knew that Caesar was approaching
and taking vengeance as he came.
That night Caesar received a dispatch ^ from Cicero, warn-

^ Suetonius, Divus lulius, 67.

2 This dispatch was carried by one of Vertico's slaves Cicero, says Caesar
;

{B. 0., V, 49, § 2), data facilitate Galium ah eodem Verticone quem supra demon-
stravimus repetit, qui litteras ad Caesarem deferat. These words imply that
VI AND ITS RESULTS 117

ing him that the Gauls had raised the siege, and gone off to 54b.c.

intercept him. . Notwithstanding their heavy losses, they "^^^ ^^"^^

numbered, it was said, some sixty thousand men.^ Caesar the siege,

made known the contents of the dispatch to the troops, and ^^rch to
encouraged them to nerve themselves for the approaching encounter

struggle. A short march in the early morning brought the

legions to a rivulet, running through a broad valley, beyond
which the enemy were encamped. Caesar had no intention
of fighting a battle against such heavy odds on unfavourable
ground. Cicero was in no danger and he was therefore
;

not pressed for time. He sent out scouts to look for a con-
venient place to cross the river. Meanwhile he marked out
his camp on a slope, and constructed it on the smallest
possible scale in the hope of seducing the enemy to attack
him. But the enemy were expecting reinforcements, and
remained where they were. At dawn their horsemen ven-
tured across the river, and attacked Caesar's cavalry, who
promptly retreated in obedience to orders. Sitting on their
horses, the Gauls could see inside the camp. An attempt
was apparently being made to increase the height of the
rampart, and to block the gateways. There was every
appearance of panic. Caesar had told his men what to do ;

and they were hurrying about the camp with a pretence of

nervous trepidation. The enemy hesitated no longer and ;

in a short time they were all across the stream. They had
to attack up hill but that mattered nothing against such
;

craven adversaries. Not even a sentry was standing on the

rampart. Criers were sent round the camp to say that if
any man cared to come out and join the Gauls, he would be
—
welcome, ^till eight o'clock. The gates looked too strong
to be forced, though there was really only a mock barricade
of sods, which could be knocked over in a moment. The
Gauls walked right up to the ditch, and began coolly filling
it up, and actually tearing down the rampart with their

—
hands, when from right and left and front the cohorts

the slave wlio carried Cicero's first dispatcli Iiad returned, just as the spy
Ungud returned again and again to Lucknow during the Mutiny. W. Nitsche,
however, needlessly supplies almm after Galium.
1 Tl
Tlie eslimate was doubtless greatly exaggerated. See pp. 242 and u. 10.
 118 THE DISASTER AT ATUATUCA CHAP.

5i B.C. charged : was a thunder of hoofs and reeling back-

there ;

Defeat of
the Gauls,
ward in amazement before a rush of cavalry, they flung
away their arms and fled.
Caesar Caesar prudently stopped the pursuit, lest his troops might
joins
Cicero. become entangled in the outlying woods and marshes
but ;

about three o'clock that afternoon the legions reached

Cicero's camp without the loss of a man. With keen interest
Caesar asked for details of the siege, and gazed with admiring
wonder at the enemy's deserted works. When the legion
was paraded, he found that not one man in ten was un-
wounded. Turning to Cicero, he heartily thanked him for
'

the magnificent stand which he had made, and then, calling

out, one by one, the officers whom he mentioned as having
shown especial bravery, he addressed to them a few words
of praise. From some prisoners, who had served under
Ambiorix, he gleaned details of the massacre at Atuatuca.
Next day he again assembled the men, and described to them
what had The culpable rashness of
befallen their comrades.
a general officer had entailed a disaster but they must not
;

be downhearted for Providence and their own good swords

;

had enabled them to repair the loss.^

Im- Meanwhile the news of the relief had spread like wildfire.
mediate
effects
Before midnight it was known in the neighbourhood of
of his Labienus's camp, more than fifty miles away. A number
victory.
of loyal Remans hurried to congratulate the general and ;

a shout of joy at the gates of his camp told him what had
occurred. Indutiomarus, who was on the point of attacking
him, beat a hasty retreat. A large force from the maritime
tribes of Brittany and Normandy was advancing against the
camp of Roscius, when an express came to warn them of
Caesar's victory, and they precipitately fled.
But even Caesar could not undo the effect of the annihila-

1 Caesar has been blamed for having quartered a comparatively weak force
under a bad general in the camp which w^as most exposed to attack. Probably
he knew that Sabinus, although he had done his work satisfactorily so far,
was not a strong man but it was necessary to employ him somewhere and
: ;

I doubt whether Caesar ought to be called a bad judge of character because

he did not foresee that Ambiorix, whom he had probably never met, would rebel.
But I admit that his temperament was over-sanguine. Indeed he virtually
admitted it himself {B. Q., iii, 7, § 1).
VI AND ITS RESULTS 119

tion of a Roman The Gauls lacked perseverance: 54b.c.

legion.
they wanted a great leader but they had broken the spell ^^ny of
;

of Roman success. Except among the iVedui and the Remi, continue
to in-
there was hardly a chieftain in Gaul who did not dream of trigue.
similar victories. Nocturnal meetings were held in secluded
places ; and embassies passed from tribe to tribe. As Caesar
frankly remarked, it was all perfectly natural the Gauls
:

had once been the most dreaded warriors in the world, and
to be forced to submit to Romans was most galling to their
self-esteem. The state of affairs was so alarming that Caesar
determined to break through his usual practice and spend
the winter in Gaul. He ordered Fabius to return to his camp
in the country of the Morini. His own quarters were at
Samarobriva ;and in the neighbourhood of that town he
cantoned in three separate camps the legion of Cicero, that
of Crassus, and the one with which he had gone to the relief
of Cicero. He sent for all the chiefs who were in any way
compromised, and when he had thoroughly frightened them
by letting them know that he was aware of their intrigues,
he tried to convince them that it was their interest to keep
the peace. The bulk of the tribes were thus deterred from
actually rebelling. The Senones, however, a powerful people
occupying the country round Sens and Montargis, were in
grim earnest. Their council condemned to death Cavarinus,
whom Caesar had set over them as king, and, as he contrived
to escape, declared him an exile. When Caesar ordered them
to come to Samarobriva and answer for this outrage, they
flatly refused to obey. But of all the malcontents the most Schen es
daring and the most dangerous was Indutiomarus. Rebuffed
^[o^*^j!^3
by the German chiefs, who answered his appeals for aid by
reminding him of the fate of Ariovistus and the Tencteri, he
raised troops and drilled them, bought up horses from the
neighbouring peoples, and offered rewards to all the outlaws
and exiles in Gaul who would join his standard. His prestige
rapidly increased and all the patriots began to look to him
;

for guidance. He summoned the warriors of his own tribe

to muster in arms at a stated place ; and, in accordance with
Gallic custom, the unhappy wretch who arrived last was
tortured to death in sight of his comrades. Indutiomarus
 120 THE DISASTER AT ATUATUCA chap.

54 B.C. began by declaring Cingetorix a public enemy, and confis-

cating his possessions. He then addressed the assembly.
His plan was to make a raid into the country of the Remi,
and punish them for their desertion of the national cause :

then to join the Carnutes and the Senones, and raise a revolt
in the heart of Gaul. First of all, however, he determined
to make one more attempt against Labienus. But the Roman
general was too strongly posted to fear any attack and he ;

determined to make an end of Indutiomarus and his schemes.

He upon the neighbouring tribes to furnish him with
called
cavalry, which were to arrive on a fixed date and, like
;

Caesar, he did his best to lure on the enemy by a pretence

of fear. Their horsemen rode up to the camp, hurled missiles
over the rampart, shouted every insulting epithet at the
Romans, and challenged them to come out if they dared.
Labienus would not allow his men to reply. The cavalry
which he had summoned arrived punctually and in the
;

night, thanks to the carelessness of the Treveran sentries,

they were admitted into the camp. Caesar afterwards noted
with admiration the extraordinary precautions which
Labienus had taken to prevent a single man from going
outside, lest the enemy should hear that he had been
He is out- reinforced. Next day, as usual, Indutiomarus and his men
LabfenuJ Spent their time in swaggering round the rampart and abusing
defeated, the Romans. In the evening, when the}^ were scattered and
off their guard, two of the gates were opened : the cavalry,
supported by several cohorts, charged ; and the astounded
Gauls fled. Labienus had given orders that every one should
pursue Indutiomarus, and him alone ;and he promised
a large reward to the man who should kill him. He was
caught in the act of fording a river and his head was cut off.
;

Forthwith the assembled bands of the Nervii and Eburones

dispersed and for a time Gaul was comparatively still.
;

Only for a time, however. Caesar had reason to believe

that the chiefs were hatching a more formidable conspiracy ;

and he saw that the best way to counteract it was to con-

vince them that, whatever successes they might gain, the
[The uth fighting strength of Italy was inexhaustible. He accordmgly
^'^
raised two new legions in Cisalpine Gaul, and asked Pompey,
 :

VI AND ITS RESULTS 121

with whom were still amicable, to lend him a 53b.c.

his relations
^^^^^^
third. Pornpey was at that time pro-consul, charged with
the government of Spain; but, as Caesar explained with two new
politic reticence, though vested with the command of an a^ncTbor-
'

army, he remained on public grounds in the neighbourhood rows a

of the capital.' Rome, whither Caesar must soon return, was from
convulsed by the throes of anarchy, and the civil war that Pompey.
was coming cast its shadow before but it was necessary ;

that he should shut out from his mind all distracting

thoughts, and perfect his work in Gaul.
Peace did not last out the winter. The Treveri, notwith- Continued
standing the death of Indutiomarus, succeeded in per- jjorth^^^
suading, by promises of gold, some of the more distant tribes Eastern
of Germany to join them, and made a formal alliance with
Ambiorix. The Nervii, the Atuatuci, the Menapii, and the
Eburones were all in arms the Senones and the Carnutes
:

were still defiant. But Caesar, as usual, was the first to

strike. While it was still winter, he left Samarobriva with
four legions made a sudden raid into the country of the Caesar
;

Nervii ;took numbers of prisoners before the bewildered ^^^

tribesmen could either muster their forces or flee drove Nervii ;

away their herds, ravaged their lands, and compelled the

cowed chiefs to submit. When he convened his annual
council at Samarobriva in the early spring, every Celtican
tribe except the Senones, the Carnutes, and the Treveri, sent
its representatives.^ The contumacy was com-
of the Treveri
paratively of little moment but Caesar has recorded what
he thought of the attitude of the other tribes
;

regarding :
— '

their absence as the first step in rebellion, he determined to

mark his sense of its paramount importance, and accordingly
transferred the council to Lutecia, a town belonging to the [Paris.]

Parish.' The Carnutes possessed the great cornfield from

which Caesar drew much of his supply and the hal- ;
'

lowed spot in their country where the Druids held their

'

annual synod was the religious centre of Gaul. The

rebellion must not spread. Immediately after the council
adjourned Caesar again took the field. A rapid march
southward so disconcerted the Senones that they surren-
'
8ec p. 395.
 122 THE DISASTER AT ATUATUCA CHAP.

53 B. c. dered at once, and begged the Aedui to intercede for them.

forces the
The Carnutes, without waiting to be attacked, induced
Senones
and Car- their overlords, the Remi, to do them a like service, and,
nutes to
submit
as time pressed, Caesar accepted, without inquiry, the
excuses of both peoples, took hostages for their good be-
haviour, and, after dispatching the business of the council,
turned northward to deal with the Treveri and the Eburones.
The exiled king of the Senones, with his contingent of
cavalry, accompanied the column ; would not
for Caesar
attempt to reinstate him, lest the odium which he had
incurred should provoke a fresh outbreak and impede the
legions in their punitive campaign. He had not forgotten
and pre- the shame and the suffering which Ambiorix had brought
pares to
punish upon his soldiers ;and he was determined to inflict upon
Ambiorix.
him a most signal and awful retribution.
As a pre- The first step was to deprive him of his allies, the Menapii,
liminary
step, he the Treveri, and the Germans. Caesar had ascertained that
crushes
the
he did not intend to fight and the object was to bar against
;

Menapii. him every way of escape. The Menapii, alone of all the
Gallic tribes, had never formally submitted to Rome. During
Caesar's first expedition to Britain, Sabinus and Cotta had
mercilessly ravaged their lands but it was impossible to
;

follow them into their fastnesses. Caesar took his measures

with extreme deliberation. He sent all the heavy baggage
to Labienus, and at the same time reinforced him with a
couple of legions. He then marched in overwhelming force
against the Menapii. Without attempting to resist, they
again took refuge in their forests and marshes but this
;

time they were not to escape. Caesar bridged the rivers,

constructed causeways over the marshes, and threw three
separate columns into their country and when their
;

flocks and herds were driven away, their villages ablaze, and
prisoners taken by scores, they were constrained to surrender.
Caesar left a body of horse to watch them under Commius,
the king of the At rebates, who had done good service in
Britain and warning them, as they valued the lives of their
;

hostages, to give no refuge to Ambiorix or his lieutenants,

he pushed southward to deal with the Treveri. Before he
could arrive, however, Labienus marched out to meet them.
VI AND ITS RESULTS 123

enticed them by a feigned flight across a river, and then, 53b.c.

suddenly wheeling round, sent them flying into the woods. Labienus

Their German allies, who had not had time to join them, the Tre-
'^®''^-
returned home and within a few days the whole tribe
;

submitted. Their leaders fled the country and Caesar's ;

adherent, Cingetorix, was appointed chief magistrate.

About this time Caesar joined Labienus and with the Caesar
;

twofold object of punishing the Germans and preventing crosses

Ambiorix from seeking an asylum in their country, he again *he
threw a bridge across the Rhine, a little above the site and
of the former one. He left a force to hold the Gallic end of ^{^reatens
the alhes
the bridge and keep the Treveri in awe. The Ubii imme- of Am-
diately sent envoys to explain that none of the Germans who
had reinforced the Treveri were countrymen of theirs, and,
offering to furnish more hostages if he required them, begged
him not to make the innocent suffer for the guilty. Finding
on inquiry that they had told the truth, he directed them to
inform him about the routes that led into the enemy's
country. A few days later he learned that the Suebi, who
had been active in sending reinforcements against Labienus,
were massing their warriors and warning their dependent
tribes to send in their contingents. He immediately en-
trenched himself in a strong position, and ordered the Ubii to
remove their stores from the open country into their strong-
holds, to drive in their cattle from the pastures, and to send
out scouts to watch the enemy's movements. His hope was
that, finding themselves short of supplies, they might be
enticed to venture a battle at a disadvantage ; but the
scouts, after a few days' absence, reported that the entire
host had fallen back to the outskirts of a huge forest near
the mountains of Thuringia. To follow them thither through
a wild country, where little or no cornwas to be had, would
simply be to court destruction. There was nothing for it
but to return. But, in order to keep the Germans in con-
stant fear of a fresh invasion, he only destroyed that part
of the bridge which touched their bank of the Rhine built ;

a wooden tower of four stories on its extremity ; and detailed

twelve cohorts ^ to hold the other end.
^ Caesar makes no further mention of these cohorts, which were probably
 124 THE DISASTER AT ATUATUCA chap.

53 B. c And now, having made every preparation that forethought

eould suggest, Caesar bent all his energies to destroy Ambi-
^n*^""""^
orix. The crops were beginning to ripen and if he could ;

not count upon the granaries of unfriendly tribes, there

would still be no danger of famine. The road ran westward
through the vast forest of the Ardennes. An officer named
Minucius Basilus was sent on ahead with the cavalry. His
orders were to march as rapidly as possible, so as to catch
the Eburones unprepared, and on no account to allow any
fires to be lighted in his camp, lest Ambiorix should be

warned of his approach. Caesar followed with the infantry

till he reached the deserted camp which, a few months before,

had witnessed the self-slaughter of the remnant of Cotta's

legion. The entrenchments were still intact. There he left
his heavy baggage and one of the newly raised legions to
guard it, under the command of Cicero. He promised to
return at the end of a week, and charged his lieutenant on
no account to allow a single man to venture out of camp
until then. The army was divided into three corps, each
consisting of three legions or, not counting auxiliaries, about
ten thousand men. Labienus was sent to the northern part
of the country of the Eburones, in the direction of the coast
between the Scheldt and the Rhme and Trebonius to the
;

south-western, in the direction of Huy. They were to harry

the enemy's country, to ascertain his designs, and to return,
if possible at the end of a week, to concert measures with

Caesar for a final campaign. Caesar himself marched

towards the lower Scheldt, in the hope of catching Ambiorix,
who was said to have retreated to the extremity of the
Ardennes.
Meanwhile that unhappy chief was being driven, like a
hunted animal, from lair to lair. Basilus and his cavalry,
guided by some peasants whom they had caught in the fields,
rode through a w^ood till they came to a cottage, m a small
clearing, where he was said to be hiding but his retainers,
;

detachments from various legions and I suppose that they were withdrawn
;

from the Rhine when the army went into winter-quarters. Their services
would certainly have been required in the seventh cai^paign. Guischard {Mem.
crll. ct hid., 177-4, p. 303) conjectures that they were supernumeraries; but
this is a mere guess.
VI AND ITS RESULTS 125

taking advantage of the contracted space, gallantly flung 53b.c,

themselves upon the Romans, while their chief threw himself

on horseback and disappeared among the trees. Catuvolcus,
the aged prince who had shared his counsels, was too infirm
to bear the hardships of a hunted fugitive, and, invoking
every curse upon the author of his ruin, committed suicide.
The Eburones were less civilized than their neighbours, and
had no walled towns to Ambiorix sent word
retreat to.
over the country-side that every one must shift for himself.
Many fled the country altogether : others dived into the
recesses of the forest : others lurked in the marshes or the
islets in Caesar found that there
the estuary of the Scheldt.
was no regular force to oppose him but every glen, every ;

bog, every clump of trees held its nest of armed skulkers.

Massed in their cohorts and companies, the legionaries were
powerless against such foes : the only way to get at them
was to send out small flying parties in every direction. But
in those narrow woodland tracks it was not easy for even the
smallest party to keep together. The enemy knew every
inch of the ground : they were wary and they were des-
;

perate : and a few legionaries who strayed in search of

plunder were cut off and
Always careful of his men's
killed.
lives, Caesar was especially careful now, when their thirst

for revenge tempted them to be rash. In order to spare Caesar

them as much as possible, he invited the surrounding tribes JJ^Ig^f

to come and destroy the Eburones, and to enrich themselves bouring
IribBS to
with booty. He intended, as he tells us, '
that Gauls should harry
*^®"^'
risk their lives in the forests, and at and not his legionaries,
the same time to surround the people with a mighty host,
and, in requital for their signal villainy, to destroy them,
root, branch, and name.' ^ Multitudes of eager plunderers
were attracted by the prospect and Caesar's old enemies, ;

the Sugambri, actually crossed the Rhine with two thousand

horse and their attendant light-armed footmen,^ in the
hope of sharing in the spoil. The wretched Eburones were
captured by scores, and their cattle driven off. But the
Sugambri were soon tempted by a richer prize. One of their
captives told them that Caesar was far away, and they need
1 B. G., vi, 34, § 8. 2 See pp. 735-6, and B. G., i, 48, §§ 5-7.
 126 THE DISASTER AT ATUATUCA chap.

53 B.C. not be afraid of him. Atuatuca was within a three hours'

march. Why should they not pounce upon Cicero's camp,
and carry off all the stores and the loot which it contained ?
It happened that on this very day Caesar was expected
in the camp. But Cicero had heard nothing of or from him,
and was beginning to fear that he would not be able to keep
his promise. Hitherto he had carefully obeyed his instruc-
tions, and had not allowed a man to stir outside the rampart.
But fresh rations were due there were cornfields within
:

three miles of the camp : it was absurd to suppose that the

persecuted Eburones would venture an attack, so near ;and

besides stung him to hear that the men were sneering at
it

his caution. Accordingly he allowed half the legion, with

three hundred convalescent veterans, who were under
a separate command, two hundred cavalry, and a number
The of slaves, to go out and cut corn. They were hardly out of
sight, when a host of horsemen broke from an outlying wood,
sun^riso
Cicero. swept down upon the camp, and tried to burst in through
the rear gate. The dealers who accompanied the army were
massacred in their tents outside the rampart and the
;

cohort on duty barely sustained the first shock. The enemy

spread round the camp, looking for an entrance and it was
;

all that the guards could do to prevent them from breaking

through the gates. The commanding nature of the site and

the strength of the fortifications forbade any attempt to
enter elsewhere. Withm, was confusion and panic and
all ;

the superstitious recruits remembered with horror that, on

the very spot where they stood, the soldiers of Cotta and
Sabinus had perished. Even Cicero lost his presence of
mind. But it happened that there was in the camp an
invalided centurion, whose deeds of daring Caesar was never
tired of extolling,— Sextius Baculus. Ill and weak, he had
not tasted food for ^ve days. As he lay in his tent, he heard
the uproar, and walked out to see what was the matter.
Without a moment's hesitation, he snatched sword and
shield from the men close by, and planted himself in the
nearest gateway. The centurions on guard rallied round him ;

and alone they kept the enemy at bay. Severely wounded,

Sextius fell down in a faint, and was with difficulty rescued :
VI AND ITS RESULTS 127

but his splendid courage shamed the trembling recruits into 53 b.c.

action ; and the camp was saved.

Meanwhile the harvesters were on their way back. They
heard the uproar. The cavalry rode on, and saw the enemy.
The rest followed. The recruits had never seen a sword
drawn in anger there was no cover near and they were
: ;

simply confounded by the apparition. They looked passively

to their officers for orders but the bravest of their officers
;

were for the moment unnerved. The Germans, descrying

infantry and cavalry in the distance, took them for Caesar's
legions and abandoned their attempt on the camp but ;

presently, seeing how few they had to deal with, rode off to
attack them. The slaves, who had rushed up a knoll for
refuge, were speedily dislodged, and, flying pell-mell into
the maniples, increased their alarm. A hurried consultation
was held. The recruits, in spite of all warnings, ended by
clustering together where they fancied that they
on a ridge,
might be safe. The handful of veterans who had accompanied
the detachment kept their presence of mind, and, followed by
the cavalry and slaves, charged boldly through the enemy's
loose array. The recruits stood watching them in helpless
hesitation. They could not make up their minds to stay
where they were and they knew that they could not follow
;

the example of the veterans. At length they tried to reach

the camp anyhow and two entire cohorts were surrounded
;

and slain. Those who escaped owed their lives to their

centurions, who threw themselves upon the enemy, for a
moment forced them back, and died, fighting to the last
man. The Germans rode away with the booty which they
had left in the woods. Caesar's advanced guard reached the
camp that night, and found the young soldiers almost beside
themselves with panic. They were positive that the General
himself and his army must have perished and nothing ;

could quiet them till they actually saw him arrive. But
nobody knew better than he how much fortune has to do
with war and he contented himself with telling Cicero
;

that he ought to have followed his instructions to the letter,

and not to have run the smallest risk.
One more effort was made to catch Ambiorix. Fresh
 128 THE DISASTER AT ATUATUCA
53 B. c. plunderers from the surrounding tribes were hounded on by
Caesar
ravages
the
imti
Caesar to hunt down his people and to harry his land. Every
it
hamlet, every buiidmg was burned down everjrthmg worth ;
-, - i

of the^
plundering was carried off and every ear of corn that was
;

Eburones. not sodden by the rain was devoured for ib was Caesar's ;

deliberate intention that everyman, woman, and child who

escaped the sword should perish of hunger. The soldiers
knew that he had set his heart upon getting Ambiorix into
his hands and they made incredible exertions to win his
;

favour. Cavalry in small parties scoured the country in

pursuit of the king. From time to time they captured
peasants, who declared that he was hardly out of sight.
Ambiorix But, in spite of the desperate efforts of his exasperated
pursuit, pursuers, he was never caught. With four retainers, who
would have suffered anything rather than betray him, he
was lost in the dark recesses of the Ardennes.
The The legions were distributed for the winter, —^two on the
legions
distri-
western frontier of the Treveri, two among the Lingones,
butedfor and the remaining six at Agedincum, now Sens, the chief
winter. town of the Senoncs.The twelve cohorts, detached from
various legions, which had guarded the bridge over the
Rhine, were doubtless now recalled. One other task Caesar
had to perform before he started for Italy. He summoned
a Gallic council to meet at Durocortorum, the modern Reims.
An inquiry was held regarding the rebellion, which at the
time he had necessarily condoned, of the Carnutes and the
Senones. Acco, a Senonian chieftain, was convicted of
Execution having originated the movement and, in accordance with ;

^^^'
° Roman custom, he was flogged to death.^
1 B. G., vi, 44, § 2; viii, 38, § 5 ; Suetonius, Nero, 49.
 CHAPTER VII

THE REBELLION OF VERCINGETORIX

'
Gaul was now tranquillized ; and Caesar, in accordance 62 b.c.

with his determination, started for Ita,ly to hold the assizes.'

So runs the first sentence of the book in which he narrated
the events of the most momentous year of his command. Agitation

But the stillness that lay upon Gaul was not peace and he ;

had hardly turned his back when agitation was renewed.

The Celts had not forgotten that they were the descendants
of Dis Pater, and that their kinsmen had once humbled
Rome. The fierceness with which the sporadic revolts of the
last two years had been repressed was at length constraining
tribes that were all embittered to combine and even in;

Central Gaul, which had submitted so tamely, the Senones

and Carnutes formed a nucleus of common Meetings
action.
were held in the recesses of forests and other secluded places.
The death of Acco was keenly discussed. The formality of
his execution seemed a sign that Caesar mtended to make
Gaul into a Roman province. Chieftains told each other
that their own turn might come They must make
next.
a supreme effort to save their unhappy country. Even
Commius was swept away on the wave of enthusiasm, and
attempted to engage the Belgic tribes but Labienus heard
;

of his intrigues, and sent Gains Volusenus, the tribune who

had won recognition at Octodurus, with a band of centurions
to assassinate him. He escaped with a wound but for the ;

time his plans had miscarried. His brother nobles were more
fortunate and early in the new year a gleam of hope shone News of
;

out. A rumour ran through Gaul that Rome was a prey to ^^® ^^^^'
sedition. The notorious Clodius had been murdered by Milo Clodius
and his bravoee. Furious riots followed temples were in Gaul,
;

flames and streets ran with blood. The story was of course
;

embellished by the eager imagination of the Gauls. They

1093 j£
 130 THE REBELLION OF VERCTNGETORrX chap.

52 B.C. persuaded themselves that Caesar would be detained in

Italy, and that his legions would be at their mercy. At the
final gathering a definite plan was formed. The great object
was to prevent Caesar from rejoining his army. The
conspirators persuaded tliemselves that there would be no
difficulty in doing this for the generals who commanded
;

the legions would not venture to leave their quarters in

Caesar's absence, and Caesar could not make his way to the
legions for Avant of a sufficient escort. At the worst, it would
be better to die in battle than to fail in recovering their
ancient renown and the freedom which was their heritage
from their forefathers. The question was put who w^ould :
—
take his life in his hand, and strike the first blow for father-
land and freedom ? He might count upon receiving a liberal
reward. The chiefs of the Carnutes instantly responded to
the appeal, and their declaration was received with loud
applause. All that they asked was a solemn assurance that
their brother chiefs would not leave them in the lurch. The
pledge would naturally have been confirmed by interchange
of hostages but this w^ould inevitably betray the secret and
; ;

it was resolved to invoke the sanction of religion. Making

—
a sheaf of their military standards a ceremony of the most
awful import —the assembled chiefs swore to be true to their
countrymen ; and a date was fixed for the insurrection to
begin.^
The Cenabum, one of the chief towns of the Carnutes, stood

nmssacre ^P^^ ^^^ ^^^^ ^^^^ occupied by Orleans. ^

was thus fitted
It
Roman to be the depot for the grain that came from the plain of
Clt'lZGIlS 3^t
Cenabum. La Beauce, and down the Loire from the fertile Limagne
d'Auvergne. Some Roman merchants were settled there,
and one of Caesar's commissariat officers. When the ap-
pointed day came round, a band of the Carnutes, led by two
desperadoes, Cotuatus and Conconnetodumnus, rushed into
the town, massacred the Romans, and plundered their stores. {

The tidings sped swiftly through the length and breadth of

Gaul ; for whenever an important event occurred, the by- ..

standers made it known by loud shouts, and those who heard

them passed on the cry over the country-side. When
^ See p. 736. ^ See pp. 405-15.

n
 ^

VII THE REBELLION OF VEROTNGETORIX LSI

Cenabum was was just sunrise. By eight o'clock

attacked, it 52 b.c.
The news
that night the news, flying from man to man, had reached
—
the country of the Arverni the modern Auvergne a hun- — the
Arvemi.
dred and forty miles to the south.
Gergovia, the chief town of this people, was situated on Gergovia.

a mountain, some two thousand four hundred feet above the

sea, about eight miles south-east of the Puy de Dome. It

was equally fitted for a place of refuge and for a capital.

Streamlets watered the meadows which compassed it round :

forage was abundant and the town commanded a view

;

ranging over a vast tract. Four miles to the north appeared

the gently sloping eminence above which now soar the
sombre lava spires of Clermont cathedral the vast plain of :

the Limagne, watered by the Allier and backed by the distant

range of the Forez, extended on the north-east above :

wooded hills and valleys on the west, its summit crowned by

the holiest sanctuary of Gallic worship,^ towered the huge
blunt cupola of the Puy de Dome and all around, as far as
;

the eye could reach, rose the cones of the volcanic land where
the Arvernian mountaineers had made their home.
At that time there was living in the town a young noble Vercinge-
""*"
named Vercingetorix, who had shared in the counsels of the ^^^|^'

conspirators, and had perhaps been chosen as their head, standing

Caesar had already discerned his ability and attempted to t/on oFthe
purchase his support. His father, Celtillus, had been the Arvernian
, govern-
most powerful chief in Gaul but he had tried to restore
; ment,
the detested monarchv, and had paid for his ambition with ^^^^^f
.
popular
his life. The boy, we may believe, like other youths of gentle enthusi-

birth, had been schooled in Druidical lore and doubtless ;

J'eJjeiHon
his imagination had been fired by the bards who at his
father's table sang the brave deeds of Arvernian heroes.^
For six years his fellow-tribesmen had watched the conquest
of Gaul, doing nothing which the conqueror thought worthy
of record, only sending their yearly contingent to swell his
cavalry.But the Arvernians had been the first to withstand
Roman invasion and he would lead them in a last effort
;

1 See pp. 73G-7.

^ See an interesting article in the Revue historique, xxxvi, 1888, pp. 1-28.
^ See Rice Holmes, Anc. Britain, p. 266 and n. 4.

K 2
 132 THE REBELLION OF VERCTNGETORIX chap.

52 B.C. to shake off tlie Roman yoke. We may be sure that, hke
Hannibal, he prayed his gods to bless his mission for ;

religion had its part in every act of Gallic life. When the
news from Cenabum arrived he summoned his retainers, and
communicated to them his plans. Their passions were easily
inflamed, and they answered with alacrity the call to arms.
The leading men, however, among whom was Gobannitio,
a brother of Celtillus, regarded the movement as quixotic,
and ordered the young chief to leave the town. But Ver-
cingetorix persevered. He took into his pay all the outcasts
and desperadoes in the district. He went from village to
village, and harangued the people and all who listened
;

caught the fire of his enthusiasm. At the head of his levies

he returned to Gergovia, and banished the chiefs who had
the lately banished him. His adherents saluted him as king.
tribes be-
jj^ ^^j^-^ ^^^ envovs
j^jg ^ in all directions j every
soon nearly :
^
tween the
Seine and tribe in Western Gaul from the Seine to the Garonne joined
Garonne ^^^ movement ; and the impressionable Celts, recognizing
join him, Vercingetorix as the man
who was to save their
of destiny

him Com- country, unanimously bestowed upon him the chief command.
mander-
in-Chief.
How he
Yie levied
iiii
from each state a definite quota of troops and of
c ir»-
hostages, and ordered each to manufacture a dennite quantity
raised an of weapons by a fixed day. He knew that the tribal miUtia-
army.
men would be of little use except for guerilla warfare, and
therefore devoted all his efforts to strengthening his cavalry.
Waverers and laggards he soon brought to their senses by
ruthless severity. Torture or the stake punished grave
breaches of discipline ;2 while minor offenders were sent home,
with their ears lopped off or an eye gouged out, to serve as
a warning to their neighbours. These methods were effective.
An army was speedily raised and the bulk of the Celtican
;

patriots were united, for the first time, under one great
leader.
It must not, however, be supposed that even now the

' Caesar does not say that the conspirators chose Vercingetorix as their head,
though his narrative suggests that they did. It is possible that mutual jealousy
may have prevented them from choosing any one if they chose Vercingetorix*
:

their choice was evidently confirmed by a general council.

* M. Jullian {Vercingetorix, p. 131), thinking of B. G., vi, 16,
§§ 4-.5, conjectures
that these victims were offered as a sacrifice to the gods.
VII THE REBELLION OP VERCINGETORIX 133

movement was general. Even in the insurgent tribes even — 62 b. c.

among the Senones and Carnutes—the Nationalists must "^^t-^^l"

have had to reckon with adherents of Rome. The Aedui, tribes,

jealous of their old rivals, the Arverni, and not prepared to

break with Caesar, still kept aloof the tribes who looked
:

up to them remained passive. The Remi and the Lingones

had long since made and from them and their
their choice ;

dependants no help was to be expected. The Treveri,

enfeebled by the chastisement which Labienus had inflicted
upon them, were distracted by German raids and the other ;

eastern tribes, if were with the insurgents, were

their hearts
severed from them by the Roman army. The Aquitanians
naturally took no heed of what was going on among the aliens
beyond the Garonne. The Belgae had been terribly punished
for their late rebellion and either for this reason or because
;

they were jealous of their Celtican kindred, they left them

alone. It remained to be seen whether Vercingetorix would
be able, by the spell of his personality, or by the victories
which he might gain, to rouse the whole people into united
action.
His first step was to send a chief, named Lucterius, tlie The
most daring of his lieutenants, to deal with the Ruteni, who LucteSus.
dwelt in the district, bordering on the Roman Province,
which is now called the Aveyron. If they and their neigh-
bours, the Nitiobroges, could be induced to join, the Province
would be exposed to invasion ; Caesar would be obliged to .

succour it ; and his separation from his army would

be indefinitely prolonged. Vercingetorix himself marched
northAvard, with the remainder of the force, into the great
plain of the Berri, which belonged to the Bituriges, hoping
to detach them from the Aedui, whose supremacy they
recognized, and to link them with the Carnutes, the Senones,
and his other allies in the north. The Bituriges at once sent
envoys to the Aedui to ask for help and the Aedui, acting
;

on the advice of Caesar's generals, dispatched a force of

infantry and cavalry to their assistance. The force marched
to the banks of the Loire, which separated the two peoples,
halted there for a few days, and then returned. They excused
themselves to the Roman generals, on the plea that they had
 134 THE REBELLION OF VERCINGETORIX chap.

52 B.C. had reason to fear that, if they crossed the river, the Bituriges
The Bitu- would combine with the Arverni to surround tliem. Caesar
riges join
Vercinge-
could never find out whether their plea was true or false.
torix. Directly after they had turned their backs the Bituriges
threw in their with Vercingetorix.
lot
Why did But what was Labienus doing ? Did he make no attempt
Labienus
not take to crush the rebellion before his chief could return ? No
the responsibility could appal him and, as the second-in-com-
;
Held ?

mand, he was entitled to act in the Governor's absence. He

had already paralysed the activity of Commius and if he ;

did not stir, it was not from lack of initiative. Perhaps, as

he expected that Caesar would arrive soon, he judged it
best to do nothing that might derange his plans. But there
is a sentence in the narrative of Hirtius, Caesar's friend,
which may supply the clue.^ At the very outset of the revolt,
Drappes, a chieftain of the Senones, called his slaves to arms,
reinforced them with a band of outlaws and brigands, and
succeeded in cutting off the Roman convoys. Caesar never
mentions him ; and he struck before Caesar
it is clear that
returned. May we infer that in the short space in which
Labienus was free to act he could not take the field because
the supplies on which he counted did not come ?
Caesar re- By the time that the news of the rebellion reached Italy,
turns with
recruits Rome, in the strong hands of Pompey, was quieting down ;

to the and Caesar was able to start for Gaul without delay. He
Province.
took with him a number of recruits, whom he had raised in
Cisalpine Gaul, to repair the losses of the late campaigns.
How shall His on arriving in the Province, was to rejoin
first difficulty,
he rejoin
his le- his army. The legions Avere quartered at Agedincum, on the
gions ?
plateau of Langres, and in the neighbourhood of Treves, two
hundred miles and more to the north. If he were to send
for them, they would be compelled to fight a battle as they
marched southwards and he was unwilling to trust the
;

issue to his lieutenants.- On the other hand, it would be

foolhardy for him, Avith only a slender escort, to attempt
to make his way to them. Even the Aedui were believed to

1 i5. (^.,viii,30, §1.

» See B. G., vii, 6, §§ 2-3. and A. von Goler, Oall. Krieg, 1880, p. 234. Die
Caesar distrust the generalship of Labienus ?
 :

VII THE REBELLION OF VERCINGETORIX 135

be untrustworthy; while Lucterius had just won over the 52b.c.

Ruteni and the Nitiobroges, taken hostages from the Gabali,
who were dependants of the Arverni, and, having raised
fresh levies, was threatening to cross the Tarn and descend
upon the opulent city of Narbo. Caesar saw that before all P^^' ,
^, ,
^ "^
bonne.)
things it was necessary to safeguard the Province. Hastening
to Narbo, he assured the anxious provincials that there was ^°
rGSCUGS
no cause for alarm, and posted detachments, drawn from the the Pro-
native troops
sr G who
garrisoned the Province, in the surround- V^^^
'
Irom a ^

ing country and also in the districts round Toulouse, Albi, threat-
and Nimes. Having thus checkmated Lucterius, he went vasionT
to join his new levies, which had been ordered to concentrate
who dwelt in
in the country of the Helvii, a Provincial tribe
the Vivarais, on the eastern side of the Cevennes. He now
saw his way to reach the army. Beyond the Cevennes lay
the country of Vercingetorix, —undefended, for Vercinge-
torix was in the Berri, a hundred miles away. But the
mountain tracks were buried beneath snow and no one had ;

ever before attempted the journey under such conditions.

Nevertheless Caesar advanced. Moving up the valley of the
Ardeche, he made for the watershed betw^een the sources of
the Allier and the Loire.^ By prodigious efforts the men crosses the

snow ^ and the Arverni, who had never

shovelled aside the ; invades
dreamed that any one would venture to cross their mountain Au-vergne,
•^
and forces
barrier, were astounded to see the Romans descending into Vercinge-
the plains. Caesar's horsemen swept over the country in come to
small parties, carrying fire and sword. The news soon spread its relief ;

and Vercingetorix, reluctantly yielding to the entreaties of

his tribesmen, hurried to the rescue. This was just what
Caesar had anticipated. Now that the rebel army was out
of the way, he might, with comparative safety, travel north-
ward to join his legions and so confident was he in the
;

soundness of his forecast that, before he learned that Ver-

^ See Archaeological Journal, xviii, 1861, p. 309, and Napoleon, Hist, de

Jules Cesar, ii, 244. M. Jullian {Hist, de la Oaule, iii, 430, n. 2) gives reasons
for believing that Caesar crossed the Cevennes by the Col du Pal.
^ There is a photograph
in the Kev. des etudes anciennes (xii, 1910, p. 85) of
a snow-drift, taken in the Col du Pal, which shows that we need not
suspect Caesar of exaggeration when he says {B. G., vii, 8, § 2) that the snow
was six feet deep.
 136 THE REBELLION OF VERCINGETORIX chap.

52 B.C. cingetorix had commenced his march, he acted as though he

had done so. He left Decimus Brutus, who had commanded
in the sea-fight with the Veneti, to occupy the enemy's
attention and for fear liis design might get abroad, he
;

announced that he was only going to procure reinforcements,

then seizes and would be back in three days. Then^ recrossing the
tunity^^^ Cevennes, he hastened to Vienna on the Rhone picked up ;

rejoin his there a body of Provincial cavalry,^ which he had sent on

legions. . .

to wait for him pushed on up the valley of the Saone as

;

swiftly as horses could carry him, hoping to elude the Aedui,

in case they were hostile rejoined the legions which he
;

I
Early in had left near Langres and, before Vercingetorix knew where
;

aici .J
1
YiQ was, concentrated the whole army in the neighbourhood

of Agedincum.^
Vercinge- Vercingetorix, however, quickly recovered from this sur-

besieges
prise. In the south of the Nievre, near the confluence of the
Gorgobina AUier and the Loire ,^ there was a town called Gorgobina,

le-Chatel?] belonging to the Boi, whom, it will be remembered, Caesar

had placed in dependence upon the Aedui. To strike at
Caesar's allies would be equivalent to striking at Caesar
himself. Vercingetorix accordingly prepared to besiege the
stronghold. Again Caesar was in a dilemma. If he left
Gorgobina to its fate, the tribes that still remained loyal
would conclude that be could not be relied upon to protect
his' friends, and would therefore probably join the rebels.

If, on the other hand, he undertook a campaign so early in

the year, the army would be in danger of starving for, ;

owing to the severity of the weather, it was very difficult to

transport supplies. But anything was better than to lose
the confidence of his allies. He must trust to the Aedui to
supply him with corn. Leaving two legions at Agedincum
1 See p. 581.
^ Caesar does not tell us what became of Brutus after he had fulfilled his
mission. Probably he retreated to the Province. He took part in the opera-
tions at Alesia, —
the closing scene of the campaign.
If Caesar was accompanied by his cavalry to Agedincum, his speed was
limited by theirs. I conjecture that he took the risk of pushing on in advance
and travelling by relays of horses, as he did when he crossed the Alps. Money
will always procure horses, and similar risks were taken repeatedly in the Indian
Mutiny.
^ See note on Gorgobina,
pp. 425-30.
VII THE REBELLION OF VERCINGETORIX 137

to guard his heavy baggage/ and sending messengers to tell 52 b. c.

the Boi that he was coming and encourage them to hold Caesar
T , r
marches 1 •

out, he marched for Gorgobina. Instead, however, of taking from

the direct route southward, he intended to go round by way f^°^^/gens)
of Cenabum ;for, although time was precious, it was of to relieve
paramount importance to punish, first of all, the people who bina;
had been the first to rebel, and who, by the massacre of
Roman citizens, had outraged the majesty of Rome.^ More-
over, by ravaging the lands of the Carnutes and Bituriges, he
might count on forcing Vercingetorix to relax his hold on
Gorgobina. His cavalry were comparatively weak, for some
of the tribes which in former years had furnished contingents
Avere now in revolt but he had reinforced his Gallic and
;

Spanish horsemen by four hundred Germans, whose value

he had recognized in the campaign against the Usipetes and
Tencteri. At the close of the second day's march he laid
siege to Vellaunodunum, a stronghold of the Senones,
probably on the site of the modern Montargis, in order to
avoid leaving an enemy in his rear, and to facilitate the
transport of his supplies. In three days the place surrendered, captures
and, leaving Trebonius to disarm the inhabitants and take dununfr
hostages for their good behaviour, he pushed on for Cenabum.
The road crossed the great forest of Orleans and Caesar
;

accomplished the distance in two long marches. It was

—
evening when he arrived, too late to begin the siege but ;

the troops at once began to make the necessary preparations.

The Loire was spanned by a bridge, the northern end of
which could only be reached from within the town. The
Carnutes, who had expected that Vellaunodunum would hold
out longer, were not prepared for resistance, and tried to
escape in the night over the bridge but Caesar, foreseeing
:

their attempt, had kept two legions under arms the gates ;

^ The recruits, who had been

temporarily left behind with Brutus in the
country of the Arverni, were ordered to march to Agedincum, though Caesar
does not say so, doubtless to learn their drill ; for Labienus left them there
when he started on his campaign against the Parisii and the Senones. See
p. 161,and B. G., vii, 57, § 1.
^ This seems a sufficient
explanation of Caesar's having made a ditotir (see
my note on Cenabum, p. 408). But it is also possible that, if there were
any bridges over the Loire above Cenabum, Vercingetorix had destroyed them.
 138 THE REBELLION OF VERCINGETORIX chap.

52 B.C. Avere instantly fired, and the town seized ; and, as the throng-
and^^^*^'^
ing masses were struggling forward through the narrow
punishes upon them, and almost all were taken
streets, the legions fell
^ ""^' The booty was given up to the soldiers
prisoners.^ the ;

town was set ablaze and the army passed over the bridge,
;

crosses and pushed on to relieve Gorgobina. Noviodunum, which

Loire, and 1^7 ^n their line of march, promptly surrendered. Hostages
captures j^^d been delivered up, and centurions Avere collecting arms
dunum and horses when the cavalry of Vercingetorix, who had
[Villate,
hurriedly raised the siege of Gorgobina, were descried in the
Neuvy- distance. Seizing the weapons which they had begun to pile,

geon vff
^
^^^ townsmen manned the walls and tried to shut the gates ;

but the centurions were too quick for them, and with swords
drawn withdrew their men unharmed. The rebel cavalry,
who were beginning overpower Caesar's Gallic levies,
to
scattered before the charge of the German squadron the ;

aiid baffled townsmen again surrendered and Caesar marched ;

to besiege Southward for Avaricum, the capital of the Bituriges, now

Avancum. occupied by the famous cathedral city of Bourges.
Vercingo- So far Vercingetorix had met with a succession of disasters.
suades the ^^t his Spirit was indomitable, and he knew how to learn
Bitunges from experience. He saw that the war must be conducted
to burn ^
, . ,

their on a totally different principle. Nothing was to be gained

!r?.r,!!fv,t^^^ by defending towns which could offer no resistance
gianaiies. '^ '—'
and ;

it Avas hopeless to encounter the Romans in the open field.

But he had thousands of light horse who could scour the

country and cut off their supplies. The grass was not yet
grown, nor the corn ripe and although Caesar had probably
;

found grain in the towns which he had captured,^ he could

only replenish his stores by sending out detached parties to
rifle the granaries. Vercingetorix called his officers together,
and told them his plans. They must hunt down the Roman

1
—
This seems to me the meaning of Caesar's words, (oppido potitur) perpaucis
ex hostium nuniero desideratis quin cuncti caperentur, quod pontis atque itinerum
angustiae multitudini fugam intercluserant {B. G., vii, 11, § 8) ; butM. Jullian
{Hist, de la Gaule, iii, 436, n. 12) thinks that the southern end of the bridge was
connected with the road beyond by a narrow causeway, which he identifies
with the itinerum angustiae. He does not explain why Caesar used the plural,
itinemm.
'
B. G., vii, 14, § 9. Cf C. JuUian, Hist, de la Gaule,
. iii, 441.

i
 ^

Ml THE REBELLION OF VERCTNGETORIX 139

foragers wherever they could find them, and attack tlie 52b.c.
baggage-train. They must make up tlieir minds to sacrifice
their own interests for the national weal. Every hamlet,
every barn where the enemy could find provender must be
burned to the ground. Even the towns must be destroyed,
save those which were impregnable, lest they should tempt
men who ought to be in the field to go to them for shelter,
and lest the Romans should plunder their stores. This might
sound very hard but it would be far harder for them to be
:

slain while their wives and children were sold into slavery ;

and, if they were beaten, this would inevitably be their

doom. This uncompromising speecli Avas greeted with
unanimous applause. For such a leader men Avould consent
to any sacrifice. Within a single day more than twenty
villages in the Berri were burned down. All round the great
plain, wherever the Romans looked, the sky was aglow.
The WTetched inhabitants told each other that they were
going to win, and would soon recover what they had lost.
But Vercingetorix could only govern by character and tact.
He had not the powers belonging to the ejeneral of an The Bitu-
rio'es con-
established commonwealth. He might venture to be severe ti^ry to ;

but he could not afford to lose his popularity. The question

J?^^^^^^^^^'
was raised, whether Avaricum should be defended, or defend
destroyed like the lesser towns. The Bituriges were not
restrained by the sense of discipline and their spokesmen
;

eloquently pleaded their cause. Their capital was the finest

town almost in the whole of Gaul and, moreover, its
;

position was so strong that they could easily defend it.

Vercingetorix strongly opposed their appealbut they ;

pleaded so pathetically, and their brother chiefs showed such

sympathy with them, that he was obliged to give way.
Following Caesar by easy stages, he finally halted about
fourteen miles north-east ^ of Avaricum, on a strong position,
from which he could communicate with the garrison and
harass the besiegers.
' M. Jullian {Hist, de la Gaule, iii, 441) remarks that Vercingetorix ought to
have adopted the plan of devastation at the outset of the campaign but could
;

—
he have induced even the Bituriges to make such a sacrifice and apparently
they were the only tribe who at that time did so (see p. 741, n. 0)—until they
had been taught by bitter experience ? ^
gg^ pp^ 742-3,
 140 THE REBELLION OF VERCINGETORIX chap.

52 B.C. Avaricum was surrounded, on every side except the south,

Siege of
j^y marshes intersected by sluggish
®^ streams. On the south-
Avaricum. "^ ^
•^

east it was approached by a natural causeway, which, about

a hundred yards from the wall, suddenly shelved down so as
to form a kind of huge moat.^ Behind this neck of land
Caesar pitched his camp. As the marshes rendered it im- .

possible to invest the position, he proceeded to construct

a terrace, by which picked troops were ultimately to advance
to the assault. The flanking parts were to serve as viaducts,
to carry the towers in which artillery were placed and
; it

is probable that the platform intended for the columns of

assault occupied only the front portion of the intervening
space. First of all, in order to provide a secure foundation,
the ground was cleared of obstructions and levelled as far as
possible by men working inside stout huts.^ The sides of
each viaduct were constructed of parallel tiers of logs, the
interstices between which were probably packed with earth
and rubble. The workmen brought up the material through
lines of sheds, which, being contiguous to one another and
open at both ends, formed covered galleries and they were ;

further protected in front by a fence of high wooden shields

moving on rollers. Between the walls of timber, which
served as lateral supports, they built up the core of the
viaduct, which was composed of earth, stones, and timber.
The artillerymen who manned the tower kept their catapults
playing upon the defenders of the wall. As the structure
rose daily higher, the elevation of the tower was correspond-
ingly augmented.^ When the viaduct was completed, the
tower could be moved backwards or forwards along the
surface while the sheds were ranged on either side, and
;

served as a means of safe communication. The central

mound was probably raised higher than the other two,** in
order to facilitate the assault and sheds were placed upon it
;

also, to screen the assailants from observation and attack,

'
See Napoleon, Hist, de Jules Cesar, ii, 255, and Planclie 20.
* See Stoffel, Hist, de Jules Cesar, — Guerre civile, ii, 357, and Caesar, B. C,
ii, 2, § 4. ^ See p. 746.
* Forming what is technically called a 'cavalier'. See my note on 'The
Agger \ pp. 002-0.

I
VII THE REBELLION OF VERCINGETORIX 141

Meanwhile the new policy was beginning

of Vercingetorix 52b.o.

to make itself felt. Organized patrols kept him informed of

Caesar's movements, and conveyed his instructions to the
garrison. Whenever the Romans went out to forage or
procure corn, his horsemen kept them in sight, and handled
them severely if they ever ventured to disperse. Caesar
did all that ingenuity could suggest to baffle him, sending
the men out at odd times and in varying directions ; but the
enemy seemed ubiquitous. Supplies were running short, and
Caesar called upon the Aedui and the Boi for corn but the
;

Aedui were half-hearted and the Boi, though they did their
;

best, had little to give. For several days the soldiers had no
bread, and were obliged to kill the cattle, driven in from dis-
tant villages, in order to subsist at all. Yet, as Caesar
proudly related, not one of them uttered a word that was
unworthy of their own victorious record or of the majesty
of the Roman people. The was steadily nearing the
terrace
wall, for gangs of legionaries, who took duty in turns, were
engaged upon it night and day. Caesar went among them
as they worked, and did all that he could to keep up their
spirits. He would abandon the siege, he told them, if they
found the pangs of hunger too hard to bear. But they would
not hear of such a thing. They reminded him that they had
fought under his command for six years with untarnished
honour, never abandoning any operation which they had
undertaken and they would cheerfully endure every hard-
;

ship if only they could avenge the massacre at Cenabum.

Vercingetorix, when his provender was consumed, moved
several miles nearer the town. Some prisoners reported
that he had left his infantry in their new encampment, and
gone with his cavalry to lie in wait for the Roman foragers
in the place where he expected that they would be found on
the following day. Caesar saw an opportunity, and marched
at midnight to attack the encampment. But the enemy
were well served by their patrols. They removed their
wagons and baggage out of harm's way into the recesses
of a wood and in the early morning Caesar found them
;

securely posted on a hill surrounded by a belt of morass, not

more than fifty feet wide. They had broken down the
 U2 THE REBELLrON OF VERCTNOETORTX chap.

52 B.C. causeways which spanned tlie morass, and posted piquets

opposite the places where it was fordable. The legionaries
clamoured for the signal to advance but Caesar told them
;

that victory could only be purchased by the slaughter of

many gallant men, and that their lives were more precious
to him than his own reputation. They allowed themselves
to be consoled :but now for the first time they had been
forced to abandon an operation which they had begun ; and
perhaps when Caesar led them back to resume the labour
of the siege, he suspected that he had been duped.
Vercingetorix, on returning to the encampment, was
accused of treachery. His officers told him to his face that
he would never have left them without a leader, exposed to
that well-timed attack, if he had not intended to betray them.
He ought never to have moved from his original position.
It was plain enough that he wanted to reign as Caesar's
creature, not by the choice of his countrymen. Vercingetorix
was at no loss for an answer. He had moved, he reminded
them, at their own request, simply in order to get forage.
They had not been in the slightest danger for the position
;

in which he had left them was impregnable. He had pur-

posely refrained from delegating his command to any one, for
fear they should worry his substitute into risking a battle ;

for he knew that they had not resolution enough to adhere

to a system of warfare which required patient toil. They
ought to be thankful that the Romans had tried to attack
them, because they could now see for themselves what
cowards the Romans were. He had no need to beg Caesar
for a kingdom which he could win for himself by the sword ;

and they might take back their gift if they imagined that
they were doing him a favour, and not indebted to him for
their safety. '
And now,' he said, to satisfy yourselves that
'

what I say is true, listen to Roman soldiers.' Some camp-

followers, whom he had captured a few days before, and had
kept on starvation diet, were told to step forward. They had
been carefully drilled in the part which they were to play.
Questioned by Vercingetorix, they stated that they were
legionaries, and had secretly left the camp in the hope of
1 See pp. 744-5.
 — —
VII THE REBELLION OF VERCINGETORIX 14.3

finding sometliing to eat ; that their comrades, one and all, 5-2 n.c
were half -starved, and too weak to get through their Avork ;

and that Caesar had made up his mind, unless within three
days he had achieved some tangible results, to abandon the
siege.^ '
You see,' said Vercingetorix, '
I — I whom you
call a traitor —have brought this mighty army, without the
loss ofa drop of your blood, to the verge of starvation. No
course is open to them but an ignominious retreat and ;

I have arranged that not a single tribe shall give them

refuge.' Clashing their weapons, as their custom was, the
tribesmen swore that Vercingetorix was the greatest of
generals, and that they would trust him through thick and
thin. They realized how much was staked upon the safety
of Avaricum and ten thousand picked men were sent into
;

the town. But jealousy had much to do with this decision.

If the Bituriges succeeded in holding the fortress ^ unaided,
the glory of the triumph would be theirs.
In devising expedients to baffle the operations of the
besiegers, the Gauls showed a resourcefulness which aston-
ished Caesar. '
They are a most ingenious people,' he
remarked, and always show the greatest aptitude in borrow-
'

ing and giving effect to ideas which they get from any one.'
The wall, compacted with transverse balks and longitudinal
beams of timber, was too tough, so to speak, to be breached
by the battering ram and, being also largely composed of
;

stone and rubble, it was proof against fire.^ The Roman

engineers used powerful hooks, riveted to stout poles, to
loosen and drag down the stones. These hooks the garrison
seized with nooses ; and then, by means of windlasses, pulled
them up over the wall. They made daily sorties, fired the
woodwork of the terrace, and harassed the Avorkers by
frequent attacks. They erected towers along the wall, in

^ A careful reader of B. G., vii, 20 will, I think, conclude that in the army of
Vercingetorix there were individuals who understood Latin.
have followed the reading penes eos {B. G., vii, 21, § 3), which is in all the
'^
I
good MSS. and is approved by Mommsen [Jahresb. d. philol. Vereins zn Berlin,
XX, 1894, p. 209).The other reading paene in eo (si id oppidum retinuissent)
is only found in two inferior MSS. If it were right, the meaning would be
that the Gauls reinforced the garrison because they realized that final success '

depended almost entirely upon their holding the town '. ^ See
pp. 74C-8.
 144 THE REBELLION OF VERCINGETORIX chap.

52 B.C. imitation of those of the besiegers, and filled them with

archers and slingers. They drove
under the terrace,
galleries
and dragged away the timber of which it was composed ;

and, assailing the Roman sappers with sharp stakes, heavy

stones, and boiling pitch, they stopped the galleries by which
they were approaching to undermine the wall.^
The siege had lasted twenty-five days and, m spitt? of;

numbing cold and drenching rains and harassing opposition,

the indefatigable Romans had built up the terrace, three
hundred and thirty wide and eighty feet high ,2 till it
feet
almost reached the wall. To complete the final section of
the work was always a difficult and troublesome operation.
It was no longer possible to rear a compact and uniform
structure, as the enemy, standing right above on the wall,
could pitch heavy stones and other missiles on to the work-
men. Huts of extraordinary strength, the sloping roofs of
which were protected against fire by bricks, clay, and raw
hides, were therefore placed near the edge of the terrace and, ;

screened by them, the men shot earth, timber, and fascines

into the vacant space until the mass reached the necessary
height.^ About midnight, when they were putting the
finishing touches to the work, a cloud of smoke was seen
rising above it. Some miners had burrowed underneath, and
set the woodwork on fire. A yell of exultation rang from
the town. Flaming brands shot down from the wall and
illumined the figures standing above pitch and logs were
:

flung on to the fire and the enemy's masses came streaming

;

through the gates. If the Romans were confused, it was

only for a moment. Caesar himself was on the spot for ;

he had been personally superintending the workmen. Two

legions regularly bivouacked in front of the camp, ready for
emergencies and while some cohorts threw^ themselves upon
;

the enemy, others drew back the towers out of reach of

the flames or dragged asunder the woodwork of the terrace
to save the hmder part of it from catching fire others again ;

ran to extinguish the flames. The small hours dragged

' See pp. 600-2. » See

pp. 748-9.
^ —
See pp. 60G and 608-10, and Stoffel, Hist, de Jules Cesar, Guerre civile,
ii, 359.
 ;

VII THE REBELLION OF VERCINGETORIX 145

by; and in the grey dawn the battle was still raging. The 62b.c.
mantlets that screened the workmen who moved the towers
had been burned and it was therefore hazardous to wheel
;

the towers to the front. ^ More than once it seemed that

the Gauls were winnmg and Caesar himself was moved
;

to admiration by their stubborn valour. He saw a man

taking lumps of fat and pitch from his comrades, and flinging
them into the flames. A bolt from a catapult pierced him ;

and he fell dead. Another man stepped across his prostrate

body, and took his place. He too was struck but in a :

moment a third was doing his work, and presently a fourth

nor was the post deserted until the Romans finally extin-
guished the flames, and the Gauls, beaten at every point,
were forced back into the town.
Vercingetorix was useless now to prolong the
knew that it

defence. He therefore sent word to the garrison to slip out

in the dark and come to his camp. They were confident that
the marshes would prevent the Romans from getting at them.
Night came on and the men, gathered in the streets and
;

open places, were just starting. Suddenly there was a rush

of women weepmg, they flung themselves at their hus-
:

bands' feet, and besought them not to abandon them and

the children who belonged to father and mother alike to the
vengeance of the Romans. Deaf to their entreaties, the
men pressed on. Frantic with terror, the women screamed
and gesticulated, to put the besiegers on their guard and ;

the men, fearing that the Roman cavalry would block the
roads, abandoned their attempt.
Next day Caesar completed the repair of the terrace,- and Storming
moved forward one of the towers. Rain fell in torrents and ;
?.^.:^^^^^*
cum.
noticing that the guards on the wall were posted carelessly,
he determined to deliver the assault. The workmen were
told to loiter, in order to put the garrison off their guard.
The troops were concealed within and hi the rear of the sheds

1 See pp. (510-1.

* I read perfectis (que operibus), which has good MS. support {B. G., vii,
27,
§ 1). If deredis is read, operibus must mean the sheds (vineae) and artillery
[tormenta), a meaning for which there is little or no authority ; and, more-
over, the words which he had begun
'
' {quae facere instituerat) would be
meaningless.
1093
L
 H6 THE REBELLION OF VERCINGETORIX chap.

52 B.C. which stood upon the terrace.^ Caesar harangued them, and
promised rewards to those who should be the first to mount
the wall. The artillerymen in the tower made play with
their engines, to give their comrades every chance.^ The
signal was given. Instantly the columns, darting forth from
their cover, streamed over the front of the terrace and
swarmed up the ladders and, panic-stricken and confounded,
;

the defenders were overborne and driven down on to the space

below. Quickly rallying, they formed up in compact wedge-
shaped masses, resolute to fight it out if they should be
attacked. But the Romans were too wary to attack them.
They lined the wall all round and not a man of them would;

come down. Throwing away their weapons, the Gauls ran

for their lives through the town to its furthest extremity
and there many, jostling one another in the narrow gateways,
were slaughtered, while others, who shouldered their way out
were cut down by the cavalry. Plunder was forgotten,
ludis- Exasperated by the long weariness of the siege, burning to
massacre, avenge the massacre at Cenabum, the Romans slew the aged,
they slew women and infants, and spared none. Some forty
thousand human bemgs all but eight hundred who made —
their way to the camp of Vercingetorix perished on that day. —
It was late at night when the fugitives approached the
camp. Vercingetorix had a turbulent host to control. They
were not a regular army, but an aggregate of tribal levies,
each commanded by their tribal chiefs. He had reason to
fear that the pitiable plight of the fugitives might excite
their emotions, and lead to disturbance and subversion of
discipline'. He therefore sent out his trusted friends and
the leading men of the several tribes to which the fugitives
belonged, who waited for them on the road, and conducted
them in separate groups to their several quarters in the
camp.
Vercinge- Next day Vercingetorix called the remnant of his people
Boles his"
together, and made them a speech. They must not, he said,
troops. be disheartened for no experienced warrior could expect
;

^ tSee pp. 749-51.

' See B. G., vii, 27, § 1 ; Stoffel, Hist, de Jule-s Cesar, — Guerre civile, li, 301 ;

and Guiscliard, Mem. mil. sur Ics Grecs d Ics Eomains, ii, 7.
VII THE REBELLION OF VERCTNGETORIX 147

invariable success. The Romans had not beaten them in 52 b.c.

fair fight they had merely stolen an advantage over them
:

by superior science. As they all knew, he had never ap-

proved of defending Avaricum. But he would soon repair
the loss. He would gain over all the dissentient tribes to
the cause and against an united Gaul the whole world
;

could not stand in arms. Meanwhile he had a right to expect

that in future they should adopt the Roman custom of
regularly fortifying their camps.
This speech made an excellent impression. The multitude
could not but admire the cheery courage of their leader :

they could not but admit that the event had proved his
foresight. They respected him too because he had had
the courage to confront them in the hour of defeat, when
another leader might not have dared to show his face. So
far then from lessening, the disaster only increased the
estimation in which he was held.
He immediately set to work to fulfil his promise. Agents, He raises
chosen for their eloquence and tact, bore lavish bribes and j^^^^^^
still more lavish promises to the dissentient chiefs. New
weapons and new clothing were provided for the survivors of
the siege. New levies, including large numbers of bowmen,
were speedily raised and Teutomatus, king of the Nitio-
;

broges, though his father had been honoured by the Senate,

hastened to join Vercingetorix with his own cavalry and with
others whom he had hired from the Aquitanians. Thus the
losses A\ hich had been incurred at Avaricum were made good ;

while those who had already fought under Vercingetorix

had learned a salutary lesson, and, in spite of their natural
laziness and impatience of discipline, were in the humour to
do or to suffer whatever he might command.
The hungry Romans found an abundance of corn in
Avaricum and Caesar remained there a few days to recruit
;

their strength. Winter was nearly over and he was about ;

to open his campaign in earnest. The Gauls, in their new-

born zeal, had entrenched their camp, and he was too
prudent to attack their strong position but he hoped either
;

to lure them into the open or else to blockade and force

them to surrender. Suddenly his attention was distracted
L 2
 148 THE REBELLION OF VERCINGETORIX chap.

62 B. c. by serious news from the Aedui. Two chiefs, Cotus and Con-
SiTre'^uest
were contending for the first
^ictolitavis, magistracy, each
of the insisting that he had been legally elected their retainers :

decides
wcre up in arms and a civil war was imminent. A deputa-
;

between tion of leading men begged Caesar to arbitrate. He saw that

claimants i^ was of vital importance to prevent the weaker side from
s-PPealijig for aid to Vercingetorix. Accordingly, though
Iffi^^^f
Vergobret. he was most reluctant to delay his operations, he summoned

the rivals and the council to meet him at Decetia, or Decize,

on the Loire. This town was in Aeduan territory, and nearly
sixty miles from Avaricum but it was illegal for the
:

Vergobret to cross the frontier and Caesar was too wise

;

to offer a needless slight to native custom. He was informed

that Cotus had been nominated by his brother, the late
Vergobret, in defiance of an Aeduan law which prescribed
that no man should hold office or even sit in the senate
while any member of his family who had done so survived.
He accoidingly settled the dispute in favour of Convic-
tolitavis, who, as was the custom when the magistracy
remained vacant, had been appointed by the Druids.^
Before dismissing the council, he urged them to forget their
differences, and told them that, if they wanted to share in
the spoils of victory, they must honestly help to put down
He sends the rebellion. ^ He should require ten thousand foot to guard
to^sup"^ his convoys, and all their cavalry. He then divided the
^ess army into two parts. Labienus was sent northward with
in the four legions, includmg the two that had been left at Agedin-
basin of
^^^1, to crush the rebellion of the Senones and Parish while ;
'
the Seme, '

and Caesar himself, with the remaming six, marched southward,

liimseirto ^P ^^^^ eastern
bank of the Allier, to strike a blow at Gergovia,
attack —the heart of the rebellion.
He^estlb- ^^ ^^® crowned by the cathedral of Nevers,
^^i^^ ^^^^^'

lishes a whicli rises above the Loire, in the peninsula formed by its
™t*Novio^ confluence with the Nievre, was an Aeduan town called
dunum Noviodunum. Caesar had marked the strength of the
( Nevers) ^
1
The meaning of the passage {B. G., vii, 33, § 4) on which the sentence in
the text based has been disputed. See p. 528, n. 2.
is
^ Leaving out of account tlie troops which Caesar raised in the Province, this

was apparently the only occasion in the war on which he employed Gallic
infantry.
(', EUGOVIA

Jfiirtyvrd-s li^c<f^ S.ttaU Lcndcn

SCALE I 30.000
Roman Milea
VII THE REBELLION OF VERCINGETORIX 149

position, and here he established his chief magazine ; but he 52 b.o.

could not spare a force sufficient to make it secure.

Vercingetorix was still on the western bank of the AUier.
As soon as he heard of Caesar's advance he broke down all

the bridges, but left the lower parts of the piles intact. The
two armies moved in full view of one another, with the river
between them. The Gallic patrols were so vigilant that
Caesar found it impossible to repair any of the bridges and ;

he began to fear that he might be barred by the river during

the entire summer. But Vercingetorix had not learned the
necessity of watching his rear. One evening, Caesar en- crosses the

camped on a wooded spot, opposite one of the bridges. ^ strata^

Next morning he took forty out of the sixty cohorts com- g^^^J

posing his force arrayed them in six divisions, so that, seen

;

from a distance, they would look like the six legions ^ and ;

ordered them to make a long march on with the entire

baggage-train. Vercingetorix suspected nothing. Caesar
remained behind with the rest of the force, waiting for the
hour when, as he calculated, the four legions and the enemy
should have encamped for the night. Then he set the men
whom he had kept behind, to work at the repair of the bridge.
When it was finished, he made them cross over, and sent for
the other cohorts. Vercingetorix, finding that he had been
outwitted, and unwilling to risk a battle, hurried on south-
ward by prodigious marches.
Caesar followed more leisurely and moving across the
:

level expanse of the Limagne, found himself, early on the

fifth day, approaching the mountain of Gergovia. He had
neglected or had failed to estimate the magnitude of the
enterprise to which he had committed himself.^ Rising on
his right front, fully twelve hundred feet above the plain,
the northern face, with its upper terraces broken here and
there by precipices, manifestly defied attack and, as he ;

moved on past the long spurs, he saw that the eastern side,
steep, rugged, and scored by deep ravines, was equall}-^
unassailable. Presently, observing on his left front a suitable
spot for a camp, he halted near the foot of the south-eastern
slope. His cavalry were soon engaged in a skirmish and ;

1 See pp. 751-4. Cf. C. JuUian, Hist, de la Gaule, iii, 460.

 150 THE REBELLION OF VERCTNOETORTX chap.

52 B.C. in the afternoon reconnoitred the stronghold from the

lie

south. The town stood on an oblong plateau, which formed

the summit, extending about seven furlongs from east to
west, and six hundred yards wide. The higher terraces, and
also the outlying heights of Risolles, linked by a col or saddle
to the south-western angle of the plateau, were bristling
with the tents of the Gauls ; and the camps were pro-
tribal
tected by a wall of loose stones, which, about half-way up
the slope, ran along the whole southern side. From the very
foot of the mountain, below the central point of the wall,
rose a low but steep hill, now called La Roche Blanche,
which projected southward at right angles, and terminated
in an almost sheer precipice. A small stream, the Auzon,
flowed eastward through the meadows which extended past
the base of the hill and two miles beyond the valley, on
;

the left as one looked up the stream, the view was closed by
a long ridge, the Montague de la Serre. Beyond the heights
of Risolles was the high pass of Opme, which at one point
gave access to them by a comparatively easy slope, and
separated them from the distant Puy Giroux.
The result of the reconnaissance was not encouraging.
The ascent to the stronghold appeared less difficult on the
south than on the other sides but even on the south the
;

ascent was not easy. Moreover, the Gauls held the whole
space between the outer wall and the town and their ap-;

pearance, as Caesar remarked, was truly formidable. Even

if the Romans could gain the col on the south-west, they

would still be confronted by a steep though short incline.

All round the plateau ran a natural glacis, to climb which.
in the face of a determined enemy, would be hopeless and ;

the usual expedient of approaching by a terrace was obviously

impracticable. To besiege or to assault the town was therefore
out of the question and Caesar resolved to make sure of his
;

supplies before proceeding even to blockade it. Meanwhile

and en- he pitched his camp on a low plateau north of the Auzon.
before about half a mile north-west of the modern village of Orcet
Gergovia. and three thousand yards from the south-eastern corner of
the town.
For some days no event occurred more important than
 — ^

VII THE REBELLION OF VERCTNGETORIX 151

a cavalry combat. Vercingetorix kept his troopers busy, 52b.c.

^^^^^
interspersing archers among their ranks ; and frequent
skirmishes took place in the plain between the south-eastern tionsat
^^^g^^'^'
spurs and the Roman camp. He made
the tribal chiefs
repair daily to his quarters before sunrise, to furnish their
reports and receive his instructions. But one detail escaped
his vigilance. Caesar had detected a weak point in the
enemy's position. The Roche Blanche, which commanded
the only descent from the town to the rich meadows of the
Auzon, was inadequately garrisoned ^ and his antagonist had
;

either neglected to fortify it or been prevented by want

of time. If only he could get possession of this hill, he would
cut off the Gauls from the chief source of their supplies.
The ascent on the eastern was practicable. In the dead
side
of night Caesar stole out of camp with two legions, drove out
the startled garrison before reinforcements could arrive, and
occupied the hill. There he constructed a small camp, and
connected it with the larger one by a pair of parallel trenches,
so that men might pass unobserved from camp to camp under
cover of the ramparts formed by the excavated earth.
Even now, however, he had cause for anxiety for his entire ;

force was hardly more than five-and-twenty thousand men,

too few to invest a position more than twelve miles in extent.
Just at this time the alarming news arrived that the Aedui Defection
were on the brink of revolt. They had not embraced the Aeduan
cause of Rome with the same unanimity, the same resolution Vergobret.
as the astute and far-seeing Remi. Diviciacus had been
Caesar's best friend but he had not been able to silence the
:

anti-Roman party and even the Caesarians were no longer

;

i=5taunch. If they adhered to Caesar, they would no doubt

—
be rewarded, if Caesar gained the day. But was it certain
that he would ? Vercingetorix was a formidable antagonist.

^B. 0., vii, 36, § 6. I accept the MS. reading, (praesidio) non nimis firmo,
on which see Schneider's Caesar, ii, 444-5. Meusel adopts Zucker's emenda-
tion, non infirmo, which would mean that the garrison was strong.
^ Napoleon, Hist, de Jules Cesar, ii, 271. '
Si Ton s'etonnait,' says Napoleon,
'
queles Romains eussent creusedeux petits fosses de 6 pieds de largeur chacun
et de 4 pieds de profondeur, au lieu d'en faire un seul de 8 de largeur sur 6 de
profondeur, ce qui aurait donne h meme deblai, on repondrait que les deux
petits fosses etaient bien plus vite faits qu'un seul grand fosse,'
 152 THE REBELLION OF VERCINGETORIX chap.

52 B. c. He might perhaps succeed after all and then their old rivals,
;

the Arverni, would supplant them. If, on the other hand,

they threw in their lot with him, their strength would surely
turn the scale. To them would belong the glory of liberating
Gaul from the invader and then they would hold sway, not
;

as his servile nominees, but as the champions of a great and

independent confederation. Caesar had suspected them from
the outset of the revolt but the story which he now heard
:

must have taken him by surprise. The ringleader was no

other than Con victolit avis, the Vergobret, whose election he
had himself secured. Vercingetorix had offered him a bribe ;

and he promptly responded to that most potent spur of Gallic

patriotism. He in turn talked over some of the younger
chiefs, and gave them part of the money. But the senate
An would certainly think twice before venturing to turn upon
contin-^
their powerful patron. The chiefs took counsel together.
gent, The mfantry contmgent, which Caesar had demanded, was
to join just starting for Gergovia. A chief named Litaviccus was
Caesar,
persuaded
placed
^
in command of it ; and, to avert suspicion,
^ his brothers
^ ^

by its were sent on ahead to join Caesar. About half-way to

dTclare^for
Gergovia, near the site of the modern village of Serbannes,^
Vercinge- Litaviccus halted the column, and delivered an inflammatory
torix
harangue. The troops were horrified to hear that all the
Aeduan cavalry with Caesar, and among them two chiefs
named Eporedorix and Viridomarus, had been massacred on
a trumped-up charge of treachery. Some men, who were in
the secret, came forward and swore that the story was true :

they themselves, they declared, were the sole survivors of the

massacre. The thoughtless Aeduans drank in the lying tale
and put themselves in the hands of their leader. It was
settled that as soon as they reached Gergovia, they should
join Vercingetorix and avenge the slaughter of their country-
men. Some Roman citizens were travelling under the
Aeduan escort with grain and stores for Caesar. Litaviccus
had them tortured and killed and, before resuming his
;

march, he sent off messengers to spread the news of the

pretended massacre among the Aedui, and urge them to arm.
Rumour flew fast. The intrigue was soon known at
1 See pp. 767-8.
VII THE REBELLION OF VERCINGETORIX 153

Gergovia. Eporedorix himself came to Caesar in the middle 52 b.c.

of the night, and told the whole story. He entreated him ^^^^^^Jg^
not to allow a few wrong-headed men to drag a friendly forced
people into revolt if Litaviccus and the ten thousand sue-
:
^^rawes
ceeded in joming Vercingetorix, the Aeduan authorities would the con-
have no choice but to support them. Caesar was intensely and
anxious but he did not hesitate. He determined to *^
;
'
go and ^^^ums
just in
intercept the deluded infantry at once, though he knew that time to
"^
the large camp would, in his absence, be exposed to a most p^mp^
serious risk. The camp on the Roche Blanche, in the hands
of a few resolute men, would be virtually impregnable.^
Before starting, Caesar ordered the arrest of Litaviccus 's
brothers but they had already fled. He took with him all
;

the cavalry and four legions, leaving two only to hold the
camps. The defence was entrusted to Fabius, who, two
years before, had joined in the relief of Cicero. Caesar told
his men that he must call upon them to make a most trying
effort but, he added, the occasion was urgent, and they
;

would not grumble. They were in the best of spirits and

ready for anything. They had marched twenty-three miles
down the valley of the AUier when the Aeduan column was
descried. Caesar sent on the cavalry to stop them, but
warned them to do violence to no man. At the same time
he made Eporedorix and Viridomarus show themselves and
converse with their countrymen. The Aedui were overawed ;

and they saw that they had been duped. They grounded
their arms and begged for mercy but Litaviccus managed
;

to escape with his retainers, and made his way to Gergovia.

Caesar knew that his action was sure to be misrepresented.
He therefore took the precaution of sending messengers to
give the Aeduan authorities a true account of what had
passed, and to impress upon them that he had treated the
mutinous contingent with forbearance. Three hours were
allowed for rest ,2 and then the Aedui went back quietly with
the legions. Darkness was now closing in. On the march
1 See pp. 757-8.
^ I read irihusque horis exercitui ad quietem datis castra ad Oergoviam movet
{B. G., vii, 41, § 1), deleting nortis after horis
with von Goler {Oall Krieg, 1880,
p. 274, n. 3) and Meusel. If the three hours' rest had been taken in the night,
there would not have been time to march back to Gergovia. ^.-^ —
„ rj-rrCT^--.

ubrarn
 154 THE REBELLION OF VERrTNGETORTX chap.

52 B. c. a party of horsemen came to meet the column, and reported

that Vercingetorix had been attacking the large camp with
desperate fury. The had alone enabled the little
artillery
garrison to hold out but many had been wounded by
;

sling-stones and arrows and Fabius was busily erecting

;

breastworks upon the rampart, in view of a renewed attack.

The news stimulated the tired men to do their utmost.
Pressing on through the small hours, Caesar reached the
all

camp before sunrise, having accomplished the extraordinary

march of forty-six miles in little more than twenty-four
hours, just in time to avert the destruction of his exhausted
legions.
Outrages Had his policy been miscalculated Perhaps the event
?
of tho
Aedui may have convinced him that he had been imprudent in
against trying to reconcile the Aeduan factions, and that, if he had
Roman
citizens. left them to fight out their quarrel and adhered to his purpose

of following up the success of Avaricum, one or the other

would have been forced to take his part. Perhaps he may
still have felt that, on the information which was available

at the time, he had not done wrong in trying to keep the

entire people on his side. But he must forget those things
that were behind. For the moment the danger w^as over ;

but there were unmistakable signs that the Aedui would

soon go over to the rebels. The ignorant populace took for
granted the truth of the news about the massacre of the
cavalry. Some were exasperated others simply rapacious.
;

They burst open the dwellings of Roman residents, robbed

them, murdered them, sold them as slaves. Convict olita vis
worked upon their passions. Once they had committed
themselves, he saw, they w^ould feel that Caesar would never
forgive them, and that they had everything to gain and
nothing to lose by taking up arms. The Aedui took care of
course to send apologies and explanations to Caesar, as soon
as they heard that their contingent was in his power. The
Government, they said, had not sanctioned the outrages
which had been committed the property of Litaviccus had
:

been confiscated and full restitution should be made. But

;

they had tasted the sweets of plunder they had little hope of
:

being forgiven and they secretly connnenced preparations for

;

I
VII THE REBELLION OF VERCTNGETORTX 155

war. Caesar received their envoys with all possible polite- 52b.c.
ness ; but he was not for a moment deceived. He doubtless
wished to leave the door of repentance open for his old allies.
There was perhaps just a chance that, if he affected to believe Anxiety of
^^^^^
that the authorities were not responsible for the excesses of
the rabble, they might be wise enough to draw back.
Meanwhile he would prepare for the worst. The defection
of so powerful a state would inevitably give a fresh stimulus
to the rebellion and it seemed probable that, if he delayed
;

where he was any longer, he might find himself hemmed in.

Yet, besides the humiliation of failure, to abandon the siege
would of itself encourage waverers to turn against him.
How was he to get away and rejoin Labienus without leaving
the fatal impression that he was obliged to flee ? .

While he was considering this problem, he ascended the

Roche Blanche in order to inspect the works of the camp.
Standing upon the plateau, he noticed with astonishment
that a hill forming part of the mass of Risolles was abandoned.
What could this mean ? Some deserters explained the
mystery. Vercingetorix was greatly alarmed for the safety
of the saddle which connected Risolles with Gergovia. If
the Romans captured this place as well as the hill on the
south which they already occupied, it would be hardly
possible for foragers to get out and the garrison would be
;

starved into surrender. Every available man therefore had

been called away to fortify the western approach to Risolles,
where alone the ascent was practicable.
Caesar immediately devised a stratagem. About mid- He
night he sent several squadrons of cavalry up the valley of to^take^^
the Auzon, whence they struck off to the left along the lower Gergovia
slopes of the Montague de la Serre, as though they intended dlmain!^
to make for the pass of Opme. In obedience to orders they
moved with a show of excitement and made a noise, in order
to attract attention. At daybreak a number of baggage-
drivers, equipped to look like troopers, rode after them,
accompanied by a few regular cavalry, who were to roam
further afield, in order to increase the effect. One of the
legions followed, and, after advancing a short distance,
crossed the Auzon, and concealed itself in a wood near the
 156 THE REBELLION OF VERCINGETORIX chap.

52 B.C. hamlet of Chanonat. Vercingetorix, who, from his com-

manding position, could just discern these movements,
became thoroughly alarmed, and sent the rest of his forces
to push on the work of fortification. Now was Caesar's
opportunity. He made the soldiers move in small parties,
so that they might not be observed, from the larger camp to
the foot of the Roche Blanche.^ Some cohorts of the 13th
legion were detailed for the protection of the smaller camp ;

while the 10th was to remain as a reserve under Caesar's

personal command, and the Aedui were dispatched from the
larger camp to ascend the mountain from the right. ^ When
allwas ready, Caesar explained his plans to his generals.
The ground, he said, being so unfavourable, he did not want
to fight a battle, but to effect a surprise : their one chance
of success was to ascend with all possible speed and he ;

particularly warned them not to allow the men, in their

eagerness for plunder, to get out of hand. Once in possession
of the camps, he doubtless hoped that they would have time
to cut off the Gallic troops from the town.
The legions were formed up on nearly level ground, on
the right of the Roche Blanche. Their path ascended a
hollow or gentle depression. From where they stood the
actual distance to the town was rather more than two
thousand yards while the place which the Gauls were
;

fortifying was barely five furlongs from the nearest gate.

The legionaries advanced rapidly until they came to the
outer wall over it they clambered, and took possession of
:

three of the camps. The few men who had been left in
them fled up the hill. The king of the Nitiobroges, roused
from his siesta, had but just time to spring up half naked,
scramble on to his horse, and gallop away. Caesar was with
the 10th legion on the hill-side, on the right of the valley
by which the column had ascended. Perhaps he had
reason to believe that it would be impossible to follow up his
advantage : possibly he intended to re-form the scattered
^ Though Caesar does not say so, I suppose that a sufficient force was left
to hold the largecamp and protect the baggage.
- M. Jullian {Hist, de la Gaule, iii, 474) supposes that this force was the
Aeduan cavalry Surely the words
! similiiudine armorum {B. 0., vii, 50, § 2)
prove that it was infantry.
 VII THE REBELLION OF VERCINGETORIX 157

legionaries, retain possession camps, and force

of the b-Z B. 0.

Vercingetorix to light anyhow he made his trumpeter sound

:

the recall.^ Separated from him by the valley, the troops

did not hear the blast of the trumpet, and, heedless of the
commands of their officers, pressed on still higher up the
slope, close to the southern gate of the town. A centurion,
named Lucius Fabius, had reminded his comrades of the
rewards which Caesar had offered before the assault of
Avaricum, and boasted that no one should get into Gergovia
before him. He was hoisted on to the wall by three of his
men, and then hauled them up in turn. A cry of terror
rose from the town. Women threw down money and clothes
to satisfy the soldiers, and, craning over with bare breasts and
outstretched hands, besought them not to treat them as they
had treated the women and children at Avaricum ; others
were let down from the wall and offered themselves a sacrifice
to lust ; many in the distant parts of the town,
while
fancying that the Romans were inside, ran for their lives.
Now, however, the men who had been engaged in fortifying
Risolles, hearing the uproar and stimulated by a succession
of messengers, came hurrymg back and formed up at the
foot of the wall. The women held up their little ones in
their arms and screamed to their men-folk to fight for them.
Standmg high above, these dense and ever-growing masses
I
were too much for the tired legionaries and they had to;

i
fight desperately to hold their ground. Anxiously watching
the struggle, Caesar sent an order to Sextius, the officer
I

whom he had left in command of the smaller camp, to lead

out his cohorts and form them up at the foot of Gergovia, so
that, in case the legions were repulsed, he might fall upon
the right flank of their pursuers. He himself moved with
the 10th a little nearer to the outer wall. Meanwhile the
panic m the town had subsided. The centurion and the
soldiers who had got in first were killed, and their bodies
pitched over the wall.Another centurion, Marcus Petronius,
while attempting to hew down one of the gates, was sur-
rounded and severely wounded. The men of his company
had followed him. I cannot save myself and you too,' he
'

^ Soe pp. 245-8,

 158 THE REBELLION OF VERCINGETORIX chap.

52 B.C. cried :
*
but it was I in my lust for glory who led you into
danger, and so help me Heaven, save you. You have
I'll

your chance : use it !

'
With these words, he flung himself
into the thick of theenemy, killed two of them, and beat off
the rest from the gate. His men rallied round him. It's '

useless,' he cried I am dying you cannot help me. Go

:
'
:

while you can, and return to your legion.' Fighting to the

last, Petronius fell but he saved his men. ;

The battle was still raging when the Romans caught sight
of a column moving over the shoulder of the hill on their right
The attack flank. It was the Aedui, whom Caesar had sent up the

with''^'^ eastern slope but the Romans,

to create a diversion:
heavy deceived by their armour, took them for enemies the Gauls :

loss. "^
. .

were closing in upon them on every side and now thoroughly ;

unnerved, they were hurled back, and fled headlong down the
valley. Blindly pursuing them, the Gauls were roughly
checked, on right and left, by the cohorts of Sextius, and by
the lOtli, who had moved lower down the slope. As soon as
they reached level ground, the runaways halted and faced the
enemy, who then moved off but forty-six centurions and ;

nearly seven hundred privates lay dead upon the hill.^

Caesar Next day Caesar assembled the troops, and lectured them
i^Teloin
severely for their disobedience. He admired their spirit, he
Labienus. told them but discipline was as necessary to a soldier as
;

courage and it was the height of presumption in them to

;

imagme that they knew how to gain a victory better than

their general. At the same time they must not be dis-
heartened for they had only been beaten because they had
;

been rash enough to fight on unfavourable ground. To give

effect to his words, he formed them up in line of battle on
the most advantageous position which he could select but ;

Vercingetorix naturall}^ refused to walk mto the trap. On

that day, however, and the next, there were slight cavalry
skirmishes, m which the Romans had the advantage. Then,
feeling that he had done enough to abate the exultation of
the enemy and to restore the confidence of his men, Caesar
abandoned the siege, and marched once more down the valley
of the AUier.
^ Regarding the operations at Gergovia. see pp. ToG-GT.
VII THE REBELLION OF VERCINGETORIX 159

The was serious indeed. The Gauls had found

situation 52 b. c.

out that he was not invincible. He had under-estimated Hjs

their force and deliberately weakened his own. The division position.

which he had assigned to Labienus, could not now crush

rebellion in the north the legions which he had kept for
:

himself were doomed from the first to waste their strength

against the strongest fortress in Gaul. For the first time in
all these years he had been beaten and his defeat would;

inevitably weaken his prestige and act like a tonic upon the
spirits of his enemies. Fortunately Vercingetorix did not
venture to pursue him he was too wary to hazard the fruits
:

of his victory by launching his levies against the legions on

the open expanse of the Limagne
and probably his pur-
;

pose was to let the Romans starve between his own force
and the insurgents of the north. ^ On the third day of the
retreat Caesar repaired one of the bridges over the Allier.
He had only just recrossed the river when Eporedorix and
Viridomarus told him that Litaviccus had left Gergovia with
the Gallic cavalry, and gone to recruit for Vercingetorix
among the Aedui. Might they go too ? It was of the last
importance that they should reach home first, so that they
might persuade their brother chi?fs to return to their
allegiance while there was yet time. Caesar was convinced
that the Aedui were lost irretrievably, and he believed that
the departure of the chiefs would precipitate the rupture ;

still he thought it best to let them go, as it would be wiser

not to betray any anxiety or to give the slightest ground for

saying that he had treated his allies as enemies. When they
took their leave, he reminded them of all that he had done
for their people, and made a last earnest appeal to their
loyalty. It is just possible that they may have meant what
they said but when they reached Noviodunum, and found
;

that Litaviccus had been officially welcomed at Bibracte and

that the Vergobret and the council had definitely declared
for Vercingetorix, they saw their opportunity. Two or Epore-

three days after their departure, Caesar learned that they virido-
had seized Noviodunum, where all his hostages, his stores, "^^ms

^ See C. Julliau, yerci7igetorix, p. 223, and G. Veitii, Gesch. der Feldziige C. J,

Caesars, p. 18G.
 160 THE REBELLION OF VERCINGETORIX chap.

52 B. c. treasure, and cavalry remounts were plundered andcollected,

seize
burned it to the ground, sent off all the hostages to Bibracte,
Novio-
dunum, thrown into the river all the corn which they could not carry
and try to
prevent away, and massacred the slender garrison and the Italian
Caesar traders who had settled in the town.^ Cavalry were scouring
from
crossing the country to cut off his supplies, and infantry threatening
the him from
Loire.
to prevent The water, swollen
crossing the Loire.
by the melting of the mountain snows, was rushing like
a torrent. Caesar saw that the crisis of the war had come.
The Aeduan infantry had deserted him. The Arverni,
elated by their victory, were on his rear on his left the ;

Bituriges, exasperated by the bitter memory of Avaricum :

the perfidious Aedui barred the road in front. His chief

magazine was destroyed ; and his supplies were fast running
out. The Province itself was insufficiently protected. The
object of the Aedui was to hem him in between the Allier
and the Loire, and there starve him into surrender or if ;

in desperation, he should make a dash for the Province, to

cut him
from the off easier way over the
Loire, and drive him
back towards the Cevennes into the clutches of Vercingetorix.
Retreat, however, was not to be thought of with the :

mountains barring the way, it would be very difficult as well

as disgraceful and above all, he could not leave Labienus
;

and his four legions to perish.'^ Probably it would be

necessary to build a bridge and at any cost, he must reach
;

the Loire before the Aedui had had time to assemble in

strengtii. They had not burned their granaries in accordance
with Vercingetorix's plan and he might perhaps get supplies
;

in their covmtry. Night and day he marched till he reached

He saves the river a few miles south of Nevers.^ Some troopers rode
himself by
a series of to look for a ford, and found one which was just practicable,

^
Merivale's narrative of this episode {History of the liomans under the Empire,
ii, 1850, p. 24) is remarkable. He says that Caesar arrived in front of Novio-
'

dunum in time to hear the last crash of the sinking bridge, and to see the
devouring flames rise triumphantly beyond it'. Now ajter Caesar heard
that Noviodunum had been burned, he made a series of forced marches in order
to reach the Loire. Yet, when he reached it, according to Merivale, he found
the fire still blazing and the bridge still falling There is not a word in the
!

Commentaries about a bridge at Noviodunum and there is no evidence that

;

Caesar went to Noviodunum at all after its destruction. See p. 774.

2 See * See
pp. 7()y-74. p. 774.
 Scale 1 K»0 000 (nenrK ]0 miles to 1 inch) 5 onf (few? EsttbLnl
Klloin ti es R mai Mile
VII THE REBELLION OF VERCINGETORIX 161

the water being breast-high. The cavahy rode into the river, 52 b. c.

and formed a from bank to bank, to break the force of

line '
^^^V^-
ordinary
^

the current then the infantry, holding their weapons above

:
^ marches,

their heads, waded across the stream. Once more Caesar was
saved by his marvellous speed. The Aedui were so confounded
by his unexpected arrival that they fled without attempting
to hinder the passage the soldiers took all the grain and all
:

the cattle that they needed and the army marched on

;

towards the valley of the Yonne to succour Labienus.

That officer meanwhile was in great peril. Leaving the Labie-

lieavy baggage at Agedincum in charge of the recruits who campaign

had accompanied Caesar from Italy, he had marched with against

his four legions down the left bank of the Yonne and of Parisii.

the Seine, for Lutecia, the capital of the Parisii. Master

of this central position, he would be able to overawe those
old offenders, the Senones and the Carnutes. A large force
assembled to oppose him. Their leader was Camulogenus,
an Aulercan from the neighbourhood of Evreux, who, though
weighed down by extreme old age, was looked up to as
a soldier of extraordinary skill. On the approach of the
Romans, he encamped on the edge of a far-reaching morass,
about twenty miles south of Paris, through which the
Essonne crept sluggishly to join the Seine. Labienus tried
to construct a causeway across the slush but finding this :

impossible in the face of the enemy, he silently quitted his

camp in the night marched back as far as Metlosedum, or
;

Melun, a town standing on an island in the Seine seized ;

some fifty barges and rapidly lashed them together threw ;

a detachment across chased away the panic-stricken

;

inhabitants repaired the bridge, which they had de-

;

molished transported his army to the opposite bank and

; ;

then moved down the valley in the direction whence he had

come. The townsmen who had fled from Metlosedum
hurried with the news to Camulogenus. He at once sent
messengers to order the destruction of Lutecia, and then
moved northward from the marsh. The barges accompanied
^ 1 am inclined to infer from a passage in the Civil War {B. C, i, 64, §§ 5-6)
that the cavahy may have been formed in two linos, one above tlio infantry,
the other below, to rescue any soldiers who might be carried off their feet.

1093 M
 162 THE REBELLION OF VERCINGETORIX CHAr.

52 B. c. the Roman column and with their aid Labienu.s crossed

;

the Marne. Lutecia was situate ujjon the island in the Seine
on which now stands the cathedral of Notre-Dame. When
Labienus arrived, the bridges had been broken down and the
town burned to the ground. He encamped just opposite the
island and the enemy established themselves over against
;

his army on the southern bank.

Just at this time the news arrived that Caesar had been
forced to retreat from Gergovia, and that the Aedui had
joined the rebellion. The story lost nothing in the telling.
Labienus was dependent on Gallic peasants for his informa-
tion and their statements were positive.^ Caesar had tried
;

to cross the Loire and had failed. He could get no supplies.

He was in full retreat for the Province. The Bellovaci im-
mediately rose in arms. Labienus found himself threatened
by this warlike people on the north-east on the south the :

Parisii and their allies confronted him while the broad ;

Hood of the Seine separated him from his base at Agedincum.

Back to that town he must somehow find his way for he ;

saw that, in his altered circumstances, it would be folly to

thmk of an offensive campaign. But how to return ? That
was a problem that would tax all the force of his mind ;

and, as Caesar said, who so appreciated his worth, he knew

that he must rely upon the force of his own mind alone. He
might have gone, as he had come, by the right bank of the
Seine but he had never yet fled before the face of an
:

enemy and to flee at such a crisis would shatter the

;

enfeebled prestige of the Roman arms. Besides, to reach

Agedincum, he must, sooner or later, recross the river and. ;

hurry as he might, cross where he would, the enemy would

be there to dispute his passage. There was nothing for it
but to cross there and then by some skilful stratagem and. ;

if he must fight, to clear the way by victory

.^

1 tSee C, E. C. Schneider's ed., vol.

ii, p. 515, Possibly he may be wioug iu
saying that the Romans
conversed [through iuterjireters] with non-com-
batants {cum pacatis) ; for B. C, iii, 48, § 2 suggests that the gossip to >^•hich
Caesar refers [B. O., vii, 59, § 1) may have passed between Labienus' s Gallic
cavalry and their countrymen of the enemy's outposts.
2 I doubt whether the tlirealening attitude of the Bellovaci deterred Labienus

from returnhig by the right bank of the Seine ; for he had a long start.
VII THE REBELLION OP VERCINGETORIX 163

In the evening he assembled his officers, and urged them 52 b. f.

to carry out his instructions to the letter. The barges were

lying under the bank, ready for use. A number of small
boats were also collected. Labienus placed each of the barges
under the charge of an officer, and ordered them to drop
down the stream about ten o'clock for a distance of four
miles, and there await his arrival. He left half a legion to
protect the camp sent the other half with the baggage-
;

train up the bank, bidding them make as much noise as

possible ;and ordered the boats to be rowed alongside of
them with a loud splashing of oars. Soon after midnight he
moved stealthily in the opposite direction with his remammg
legions, till he came to the spot where the barges were wait-
ing, near the southern end of the Bois de Boulogne. A furious
storm was sweeping over the valley and in the rush and
;

roar of wind and rain the enemy's outposts were surprised

and cut down and the troops were ferried across the river.
;

The stratagem, however, only partially succeeded. About

daybreak messengers hurried one after another into the
Gallic encampment, and reported that there was a great
uproar in the Roman camp, soldiers tramping and oars
splashing up the stream, barges crossmg below. Camulogenus
was perplexed. He fancied that the Romans were crossing
the river in three places, and would soon be in full retreat.
Sending a small detachment in the direction of Metlosedum,
and leaving another to watch the Roman camp, he marched
in person against Labienus.
was about half an hour before sunrise. The Roman
It Heextri-
general harangued his troops. He reminded them of the gelH^-oJ^
glorious victories which they had won in the past, and told a perilous
^°"
them that he expected them to fight as they would have ^y^^

fought if Caesar had been there to command them. The victory

Gallic leftbroke before the first charge : but the right fought
with extraordinary resolution and for ; a long time the issue
was doubtful. The aged Camulogenus was in the forefront
of the battle, cheering on his men. At length, however, the
victorious Roman right fell upon their rear. Even then not
a man would give way ; but all were surrounded and slain.
Camulogenus shared their fate. The troops which had been
M 2
 164 THE REBELLION OF VERCINGETORIX ciiAr.

i>2 B. C. detached to watch the Roman camp hurried to the rescue,

and established themselves on the hill of Mont Parnasse, but
were speedily dislodged, and, mingling with the runaways
from the left wing, were slaughtered — who failed to find
all

shelter in the woods and on the hills —by the Roman cavalry.
and The road to Agedincum was again open. Labienus returned
I'o^'cjohi
^liither to take up the heavy baggage and thence marched
;

Caesar. southward to rcjoin Caesar.^

Tho Still the rebellion was rapidly gaining ground. The
stiniu- defection of the Aedui was a turnmg-point in the war.
laicd by Other tribes were won over by their influence and their gold.
sionof the Wavcrers they terrified by threatenmg to put to death the
Aedui.
hostages whom Caesar had left at Noviodunum. But dis-
cord and jealousy even now made themselves felt. The Aedui
asked Vercingetorix to come to them and concert operations ;

They and he readily consented. Forthwith they claimed the right

the direc- of directing the campaign but their demand was disputed
:
;

"^ ^^^^ ^^^^ ^ general assembly was convened at Bibracte to settle

w?r
war.
the question. The Remi and the Lingones, who steadily
adhered to the stronger side, and the Treveri, who were
themselves hard pressed by the Germans, alone failed to
appear. All the other tribes, even the most distant, sent
their representatives to the mountain city.^ It was the
supreme moment in the life of Vercingetorix. A few weeks
before, while they were still smarting under defeat, he had
told his men that he would win over the rest of Gaul to the
cause, and that against an united Gaul the whole world could
not stand in arms. And now his promise seemed about to
be fulfilled. With a fraction of the people he had vanquished
the invincible conqueror and the whole people was rallying
;

to his side. Chieftains and their retainers defiled through

1 See pp. 775-85.

^ It will of course be understood that the Acj^uitauiau tribes were not repre-
sented, I am inclined to infer from certain significant omissions in the list

which sent contingents for the relief of Alesia that

{B. G., vii, 75) of the tribes
the Remi, the Lingones, and the Tre\eri were not the only tribes that were
unrepresented in the council at Bibracte, and tliat othcr>s, particularly the
Suessiones and the Leuci, were influenced by the two former to liold aloof. It
seems unlikely, moreover, that the Eburones and the other Cisrheuauc (Ger-
'

mans '
sent delegates.
 — —
viT THE REBELLION OF VEROINGETORTX 165

the busy streets, and thronged the open terrace where affairs 52 b.c.

were discussed. The question was put to the vote

of state ;
Y^r.

and, without one dissentient, the representatives of the re-elected

^^"^"
Oallic nation chose Vercingetorix
® as their General. Bitterly
^
*^ mander-
chagrined, the Aedui repented the rashness with which they in-Chief

liad flung aside the friendship of the Romans : but it was ^l^^aA
too late now to draw back ; and Eporedorix and Viridomarus council,

reluctantly submitted to the Arvernian king.

Vercingetorix determined to adhere to his origmal plan His plan of
of campaign. His numerical superiority was indeed largely ^^^^P^^S"-

illusory. Five years before, the huge Belgic host had melted
away and he knew that he must limit his force to the number
;

which could act without losing that mobility which was its
strength. 1 —
His infantry eighty thousand chosen men
were sufficient for a guerrilla warfare and he contented
;

himself with levying fifteen thousand horse from his new

allies. 2 Relying on his superiority in this arm, he intended
simply to cut off his enemy's supplies and once more he
;

appealed to his countrymen to destroy their crops and to

burn their granaries that they might achieve their liberty.
He forced the peoples who had just joined the movement to
give hostages for their fidelity.That he might have a strong-
hold to retreat to in case of necessity, he fortified and pro-
visioned Alesia,^ a town belonging to the Mandubii, which
covered the plateau of Mont Auxois, in the highlands of the
Gote-d'Or. But he intended also to carry the war into
the enemy's country. The Roman Province was a tempting
prize. he could seize it or could seduce the Provincials
If
to join him, would not the triumph of his cause be assured ?
He hounded on the neighbours * of the Helvii and the Volcae He
""
Arecomici to attack them and, believing that the Alio- ^^^^^^^^
;

broges were still smarting under the punishment which neigh-

Rome had inflicted upon them a few years before, he sent ^\^^ pj-^.
envoys to bribe the chiefs and to hold out to the government vincial
^ See G. Veith, Gesch. d. Feldznge C. J. Caesars, p. 496.
^ See B. G., vii, 64, § 1, omnes equites,XV milia numero, celeriter convenire
iuhel. I am
not quite sure whether Caesar means that Vercingetorix levied
l/),000 additional cavalry, or that the cavalry which he had already, amounted
with the new levies to 15,000. See B. G., vii, (JO, § 5 ; 7 J, § 7.
'^

* The Oabali, Arvcrni, Ruteni, and Cadurci.

166 THE REBELLION OF VERCTNGETORTX chap.

the prospect of supremacy over the Province, aiid raised a

levy of ten thousand Aeduans, whom he reinforced with eight
hundred cavalry, to coerce them if persuasion should fail.
It was a master-stroke and Caesar knew that, if it
;

succeeded, he would be in extreme peril. Everjrthing

depended upon the Allobroges. They had been badly
treated by former Governors and before Caesar entered
;

Gaul they had been the most disaffected subjects of Rome.

But (^aesar had rescued them from the Helvetii he had :

distinguished two of their leading men, who had rendered

him signal services, by special marks of favour ^ and, ;

doubtless by the exercise of his unerring tact, he had taught

them to believe that his cause was theirs. The Province was
fairly satisfied with Roman rule. The Allobroges formed
a chain of piquets on the Rhone and presented an impene-
trable front to the enemy while ten thousand men, raised
;

in tlie Province itself and commanded by Lucius Caesar,

a kinsman of the Governor, were posted at various points
on the threatened frontier. The Helvii, however, wlio risked
a battle, were defeated with heavy loss and driven into their
strongholds. Meanwhile Caesar contrived a plan for counter-
acting the enemy's superiority in cavalry. No reinforce-
ments could be expected from the Province for the roads ;

were blocked. He therefore sent across the Rhine to the

tribes which he had reduced to submission, ^ and procured
from them numbers of horsemen with their attendant light
infantry, who eagerly welcomed the chance of sharing in the
plunder of Gaul. But the German horses, though hardy,
were small and light and Caesar saw that his new allies
;

would be at a disadvantage when they encountered Ver-

cingetorix's well-mounted troopers in the shock of battle.
He therefore remounted them on the horses of his tribunes
and body-guard and of the time-expired centurions and
legionaries who, on his invitation, had volunteered for
service, and were accordingly privileged to ride on the
march.
Some weeks had passed since Caesar had rejoined Labienus.
The meeting had taken place on the south of Agedincum, ^
» 7?. a, iii, 59, § 3. « See p. 249. ffl
vTi THE REBELLION OF VERCTNGETORIX 167

near the confluence of Armangon and the Yonne and, as 52 b. c.

tlie ;

Asjedinciim
° itself had been abandoned, the united army took ^® ,
marches />

Up its quarters not far from Troyes, among the friendly to succour
the Pro-
Lingones.i was the most convenient breathing-place that
It
vince.
Caesar could have found. The Remi, steadily loyal to him
and steadily false to their countrymen, were close by on the
north, to support him and to receive his support the Aedui :

were on the south and, while he was near enough to watch

;

their movements, he could collect fresh stores and rest his

troops in comparative security. But the Province was still
threatened and he saw that he must march to its relief.
;

Probably he intended also to reinforce his troops there, and

then to return and make an end of the rebellion. Accord-
ingly he moved down the valley of the Tille, intending to
cross the Saone near St. Jean-de-Losne, and to take the
road through the country of the Sequani. Vercingetorix
with his infantry and his fresh hosts of horsemen moved
off from Alesia to intercept him, and took up a position
behind a stream, not far north of Dijon,^ about ten miles
from the spot where the Romans were encamped. His line
of retreat was secure and he was protected in front by the
;

slopes of the valley. He made up his mind to risk an action,

although, only a few weeks before, he had declared that he
would not tempt fortune so much harder is it to pursue
;

than to adopt a wise plan of campaign. It would be rash,

however, to affirm that he consciously departed from his
original resolution.^ He did not contemplate a regular
engagement. He was proud of his own cavalry and he was ;

perhaps ignorant that Caesar had been reinforced by those

doughty squadrons from beyond the Rhine. The legions
were of course too strong to be attacked but they were :

hampered by an immense baggage-train and they must ;

either lose precious time in defending it, or abandon it at

the cost of their honour, nay of their means of subsistence.
He would draw up encampment,
his infantry in front of his
to encourage his cavalryand to overawe the Romans. If he
allowed Caesar to reach the Province, he would soon come
back stronger than ever and then all hope of liberating
;

'
See pp. 785-90. 2 See pp. 791-801. ^^
See pp. 790-1.
 168 THE REBELLION OF VERCINGETORTX chap,

52 B. c Gaul would be at an end. vSuch, we are told, were the argu-

ments by which he tried to animate his officers. With one
voice they cried, in an outburst of enthusiasm, that every
man must be sworn, by a solemn oath, to ride twice through
the enemy's ranks, or never again be admitted to hearth and
home, never again be suffered to come nigh unto father or
mother or wife or child. Vercingetorix assented and the ;

oath was taken. Next morning the Roman column was

discerned beyond the stream, each legion followed by its
own baggage-train.i Vercingetorix must have exulted at
this sign of carelessness. But between him and Caesar was
a hill, which a leader gifted with the intuition that can
divine an enemy's intentions would have seized. Vercin-
Vercinge- getorix ranged his infantry in front of his encampment, in
tacks*
'
^^ imposing array while the cavalry swept down upon the
;

Caesar's Roman vanguard and on either flank. Caesar was surprised

as completely as in the battle on the Sambre. The lie of the
ground had prevented him from observing the approach of
the Gauls and, marching securely through a friendly
;

country, he had neglected to send out patrols. He made

his dispositions, however, with his usual calmness. He sent
his cavalry, in three divisions, to repel the triple attack ;

and the legions formed a hollow square outside the baggage,

ready to support them if they were hard pressed.^ For a time
the Gauls had a slight advantage but the legions prevented ;

them from following it up. And where were the Gallic

infantry, —
those eighty thousand chosen men, with whom
Vercingetorix had promised to support his cavalry ? Were
they inferior to the levies of the Seine, who had so borne
themselves against Labienus as to merit Caesar's praise ?
Why did Vercingetorix, since he had risked a battle, not
throw them into the scale ? We cannot tell. The knights,
who had tried to redeem their pledge, fought on without
1 Cf. B. G., ii, 17, § 2 and 19, § 2 with vii, OG, § 4, and see C. Jiillian, /7/.s^/.

dfla aan!e,u\,49S,n. 3.
- To effect this formation would of course have required a considerable
time and M. Masquelez {Spectateur militaire, 2" ser., xlvi, 1864, p. 54) infers that
;

the army was marching 'en plusieurscolonnessepareespar des intervalles dans

lesquels Jules Cesar fit entrer le=5 bagages '. This seems to me more tlum
improbable. See n. 1, supra, and C. Jullian, Hist, de la Gaule, iii, 500, n. 2.
ALE SIA
viT THE REBELLION OF VERCTNGETORTX 169

support. At length from the hill on the Roman right the 52 b. c

German horse came thundering down on their flank,i and
drove them with heavy loss towards the stream. The rest andre-

of the Gauls, fearful of being surrounded, galloped for their ^eTten, to

lives the infantry, passive spectators of the slaughter, fell Alesia

;

back upon their camps and Vercingetorix, ordering his

; Auxois).

l)aggage-drivers to follow him, hastened westward towards

Alesia. All that he had achieved in six months by sustained
effort was lost in a day. With his beaten force he could not
keep the field, lest his disheartened followers should fall
away and disperse.^ Either he must submit to the fate of
Ambiorix, or he must again plant himself in a strong-
hold and defy his enemy to dislodge him. But Caesar was
pressing upon his rear and at nightfall, when the pursuit
;

ceased, three thousand of the fugitives were slain.

Next day the Romans arrived at Alesia, where Vercin-
getorix was preparing to make his final stand. Their baggage,
which had been parked on a hill near the battlefield, was
following, escorted by two legions. The column descended
a valley closed on the right and the left by the hills of Bussy
and Pevenel. On their left front, connected with Pevenel
by a broad neck of land, rose a hill, much lower than Ger-
govia, but still too steep to be taken by assault. The Gauls
were swarming on the eastern slope, beneath the scarped
rocks of the plateau, on which stood the town and Ver- ;

cingetorix had made them build a wall and dig a ditch to

protect their encampment. Just at their feet the legions
saw a stream, the Oze, winding like a steely thread through
the greenery that fringed the north of the hill and beyond ;

its southern side, parallel to the Oze, but invisible, flowed

the little Moving down past the hill of Rea,
river Ozerain.
the soldiers came to a miniature plain, which extended, three
miles in length, beneath the western slope of Alesia, and
was bounded On its further side by a range of heights the :

river Brenne, which received the waters of the Oze and the
Ozerain, meandered through it from south to north and ;

^ '
The cliarge of 10 horsemen on the flank is more effective tlian that of 100
on the front.' Lord Wolseley, Tha Soldier s Porlel-hool; 88fi, p. 1^0.
1

- Seep. 801.
 17(1 THE REBELLION OF VERCTNGETORTX chap.

52 B. c. beyond the Ozerain the steep declivities of Flavigny com-

pleted the zone of hills.

Caesar Caesar harangued his troops and encouraged them to

Alesia!
bracc themselves for a toilsome effort. As it was evident
that the place could not be taken except by a blockade, he
drew a line of investment, fully nine miles in length, along
which a ring of camps was constructed. Those intended for
the cavalryi were on low ground, three in the plain and one —
in the valley of the Rabutin, which entered the Oze from the
north. The rest were strongly placed upon the slopes of the
outlying hills. Close to the camps redoubts or block-houses,
twenty-three in all, were thrown up and strong piquets ;

were placed in them, to guard against any sudden sortie.

The Gallic Soon after the commencement of the works, Vercing^torix
make? ^^^^^ ^H his cavalry down the hill and a desperate combat ;

sortie,
^as fought in the western plain. Caesar's Gallic and Spanish
niiirare
liorsc wcre soon m trouble
• n t t -i i /-n
beaten, and he sent his Germans to;

reinforce them. The legions were drawn up in front of their

camps, to support the cavalry and to deter the enemy's foot
from attempting a sortie.^ The Gauls were beaten, and
galloped back along the valleys of the Oze and the Ozerain,
hotly pursued by the Germans but the gates of the camp :

being too narrow, many of the thronging fugitives were cut

down ; while others threw themselves off their horses and
tried to scramble over the wall. The legions, by Caesar's
1 Commandant J. Colin {Pro Alesia, 1907, pp. 269-70), remarking that
cavalry has always needed the support of infantry, doubts whether any of
the camps that have been revealed by excavation were exclusively occupied
by cavalry. He admits, however, that the bulk of the cavalry encamped
on the plain, small detachments only being kept on the heights for reconnoitring.
- Commandant Colin (i6., p. 270) thinks that some of these castella {B. G.,

vii, 69, § 7) may have been block-houses, without trenches but cf. Stoffel,
;

Guerre civile, i, 225-6.

' Meusel {Jahresh. d. pMlol. Vereins zu Berlin, xxxvi, 1910, p. 54) may be
right in arguing that the words ne qua sitbito inruptio ah hostium peditatu fiat (' to
prevent any sudden attack by the enemy's infantry') in B. G., vii, 70, § 2,
are spurious ; for, as he remarks, the words that follow them praesidio le-
gionum addito nostris animus augetur {' supported by the legions, our men
—
gathered confidence') show that the legions were intended to support the
cavalry. Cf. vii, 66, § 4. But it does not follow that they were not also
intended to check any attempt which Vercingetorix might make to support
his cavalry and although the words in question are suspicious, it would
;

be rasli to delete them.

 :

VII THE REBELLION OF VERCINGETORIX 171

order, moved forward a little. The Gauls inside the wall 52 b. c

were smitten with panic :
'
To arms,' they cried, arms'
to '

many them fled helter-skelter

of up the hill-side and ;

Vercingetorix was obliged to shut the gates of the town,

for fear the camp should be left unprotected.
He saw with dismay that the toils were closing around Vercinge-

him. He had hoped, perhaps, that Caesar, who had failed them out
so ignominiously at Gergovia, would not be strong enough *« f^tch
/> •1111
to enforce a systematic blockade.
T"*
J3ut there
1
were now ten
succour.

legions instead of six and wherever he looked, over the

;
^

plain or down in the valleys, there were soldiers at work with

axe or spade. There was nothing for it but to appeal to the
whole Gallic people to extricate him from the trap in which
he was caught. The ring of redoubts was not yet complete ;

the Romans were far too few to guard the whole circuit of
the mountain and the cavalry might steal out in the dark
;

without attracting notice. He

charged them to go, each to
his own country, and bring back with them every man who
could wield a sword. He reminded them of all that he had
done for the good cause, and adjured them not to abandon
him to the vengeance of the Romans. Everything depended
on their using all speed if they left him to perish, the whole
:

garrison would perish with him. By reducing the rations,

he reckoned that he might make the provisions last a little
over a month. Silently up each river valley sped the
shadowy cavalcade, until it was lost to view.^
Caesar learned the whole story from some deserters. Its Caesar

only effect was to stimulate his inventive genius. If he unes of

could keep the army from breaking out, contraval-
of Vercingetorix
he could also keep the relieving force from breaking in and ; circumval-
^^*^*^"-
unless they could break in quickly, they would be forced to
disperse for want of food. The most vulnerable part of
his position was the open meadow on the western side
of the mountain. Across this expanse, from the Oze to
the Ozerain, a trench was dug, twenty feet wide with
perpendicular sides, to prevent the enemy from attacking
the troops while they were constructing the proper works.
Such trenches were extremely rare and that Caesar ;

1 8ee pp. 802-4. 2 g^e p. 806.

 172 THE REBELLION OF VERCINOETORIX chap.

52 B. c. should have imposed upon his men the enormous labour of

making the sides vertical shows how much depended upon
thepreliminary portion of the defences. About four hundred
yards behind the ends of this trench, but bending outwards,
was traced the line of contra valla tion, which was prolonged
so as to surround Alesia, and ran along the lower slopes of the
encircling hills and across the valley of the Rabutin. First
of all, two were dug, each fifteen feet wide
parallel trenches
and eight feet deep, the outer of which extended only across
the plain, while the inner, embracing the whole circuit of the
hill, was filled, where the level permitted, with water drawn

from the Ozerain and the Rabutin. Just behind the outer
trench, and also behind that portion of the other which
encompassed the rest of the position, a rampart was erected,
surmounted by a palisade, with an embattled fence of wattle-
work in front, from the bottom of which projected stout
forked branches. The combined height of rampart and
palisade was twelve feet. Wooden towers were erected upon
the western section of the rampart at intervals of one hundred
and thirty yards,^ and also at certain points along the rest
of the contra vallat ion.
To repel the reinforcements for which Vercingetorix had
sent, a line of works somewhat similar to these, forming the
circum vallat ion, was traced along the heights of Flavign}^
Pevenel, and Bussy, and across the intervening valleys and
the plain. The circuit of this line was about twelve miles.
But even these works were not deemed sufficient. The
Gauls made frequent and furious sallies. Comparatively few
of the Romans were available as combatants for many had ;

to go in quest of corn and timber, while others were labourmg

on the works. Caesar therefore invented various subsidiary
defences. Ditches, five feet deep, were dug just inside the
large moat that was filled with water and five rows of ;

strong boughs were fixed in each, with one end protruding

above ground, sharpened and with the branches projecting
so as to form a kind of abatis. In front of them and rising
a few inches above the ground, but purposely concealed by
brushwood, were sharp-pointed logs embedded in small pits.
1 See p. 810, n. 7.
VII THE REBELLION OF VERCINGETORIX 173

111 front of these again, concealed, but barely concealed, 52 b. g.

beneath the turf, were barbed spikes fixed in pieces of wood.

Fringed by these formidable defences, Caesar expected that
contravallation and circumvallation would be alike im-
pregnable.
Nevertheless, the struggle was likely to be prolonged ; and
it Avould certainly tax to the utmost the endurance and the
fighting power As soon as the relievmg army
of the men.
should arrive, the Romans would be hemmed in between
two desperate enemies. Every moment for preparation was
precious. Flying parties scoured the country for corn and
provender but they could not collect a sufficient supply
:
;

and the rations had to be reduced.^ Every day even by —

night, when the moon was up, or in the glow of the watch-
fires —
the besieged could see the indefatigable legionaries
labouring to finish their works before the time for the great
hazard should arrive.
Meanwhile Vercingetorix had abandoned his camp, and
withdrawn the troops who occupied it into the town. He
took every precaution to husband his scanty resources. He
ordered the whole of the grain to be thrown into one common
stock and brought to him for safe keeping and he let it be ;

known that disobedience would be punished with death.

From time to time each man received his scanty ration.
Meat was tolerably abundant for the Mandubii had driven
;

large numbers of cattle into the stronghold.

The appeal of Vercingetorix had meanwhile been answered. Orgauiza-

A council of chieftains met to consider the situation. Prob- annv of^

ably the Aeduan delegates were minded to wreck the plan ;
relief,

but there is no reason to doubt that the majority were whole-

hearted in supporting the leader who, a few weeks before,
had been unanimously acclaimed. Vercingetorix, in his great
need, had asked for a universal levy but the cooler judge-
;

ment of the council rejected his demand. So vast a multitude

would become unmanageable and it would be impossible
;

to find food for so many mouths. ^ It was resolved, therefore,

to call upon each tribe for a limited contmgent. Several,
however, were finally omitted, doubtless because it Avas
1 Cf. B, C, iii. 47, § 0.
-'
Sec p. 822.
174 THE REBELLION OF VERCINGETORIX CHAr.

known that they could not or would not help. The Treveri
were still struggling with the Germans. The Eburones had
been wellnigh exterminated the smaller tribes around
;

them had made their peace with Rome the Atuatuci and ;

the Menapii, who had suffered much, were naturally excused.

The Remi and the Lingones were of course obdurate the ;

Suessiones had been placed by Caesar in dependence upon

the Remi, and were content to follow their lead. But it is
surprising that the Viromandui, who had fought side by
side with the Nervii on the Sambre, now kept aloof, and
that while the Mediomatrici sent six thousand men, their
neighbours, the Leuci, whose territory was close to Alesia,
contributed none. Probably the Viromandui were influenced
by the Remi, and the Leuci by the Lingones. Forty-three
tribes in all were named. The summons was obeyed with
alacrity and from north and south and east and west, from
;

the Seine, the Loire, and the Garonne, from the marshes of
the Scheldt and the Sambre and the mount ams of the Vosges
and the Cevennes, from the Channel and the Atlantic Ocean,
horse and foot came swarming to save the hero of Gaul. His
fame had travelled to remote Helvetia and six thousand
;

—
devoted men a remnant of the host which had bled on the
hill of Armecy —
once more set their faces towards the west.
But even in this supreme moment, in one instance, tribal
jealousy prevailed over patriot if^m. The Bellovaci peremp-
torily refused to send a single man. They intended, they
said, to attack Caesar on their own account, and had no
intention of being dictated to by any one.^They consented,
however, as a personal favour to Commius, king of the
Atrebates, who had great influence with them, to dispatch
a small contingent. Four generals were chosen for, except ;

Vercingetorix himself, there was no one leader of sufficient

eminence to command universal respect. And, as if this
Aveakenmg of authority were not enough, the generals were
fettered by civil commissioners, whose instructions they were
to follow in the conduct of the campaign. One of the four

^
M. Victor Tounieur {Une monnaic de necessite des Bellovaqms, 1906, p. 10)
conjectures that tlie Bellovaci were also actuated by a desire to punish the
Rcnii for their desertion of the national cause.
 s

VII THE REBELLION OF VERCINGETORIX 175

was Commius, who had good reason to abhor the Roman 52 b. g.

name, and now saw a prospect of takmg his revenge.

His brother generals were Eporedorix and Viridomarus,
representing the Aedui, and Vercassivellaunus, a cousin of
Vercingetorix. The vast host mustered in the country of
the Aedui, eight thousand horsemen and nearly two hundred
and fifty thousand foot, and marched for Alesia in the
certain confidence of victory. .

By this time the garrison were in great straits. Their Famine in

^^^"^^'
grain was all consumed.^ Day after day they strained their
eyes, trying to catch a glimpse of the relieving army but ;

there was never a sign. At length the chieftains called

a council of war. Some advised surrender others were:

clamorous for a grand sortie but one proposal equalled in

:

atrocity the worst that has been told of Jerusalem or Samaria.

An Arvernian chieftain, called Critognatus, reminded his Ciitogna-
when driven into their fastnesses
hearers that their fathers, '
^"'^ P^^'
poses can-
by the Cimbri and Teutoni, had sustained life by feeding nlbalism.

upon the flesh of those who were useless for warfare and ;

he urged that, to give the garrison strength to hold out to

the last against the tyi^ants who made war only to enslave,
this glorious precedent should be followed. Finally it was
decided that all who were too old, too young, or too feeble
to fight should be expelled from the town that those who
;

remained should try every expedient before having recourse

to the desperate remedy of Critognatus but that, if the
;

relieving army failed to arrive in time, they should even

follow his counsel rather than surrender. Accordingly the The fate
Mandubii, to whom the town belonged, were compelled to ?J^^^ u-
depart, with their wives and children. They presented them-
selves before the Roman lines. Many of them were weeping.
They piteously begged the soldiers to receive them as slaves,
— only give them something to eat. To grant their prayer
was impossible and a line of guards, whom Caesar posted on
;

the rampart, forbade any attempt to escape.

But suspense was nearly at an end. It was just after
1 According to Napoleon I (Precis des guerres de Cesar,
1830, p. 110), more
than fifty days must have elapsed between the departure of Vercingetorix'
cavalry and the arrival of the relieving army.
 176 THE REBELLION OF VERCINGETORIX chap.

the expulsion of the Mandubii when the anxious watchers

on the saw, moving over the plam, a multitude of cavalry.
hill

The infantry were on the heights of Mussy-la-Fosse behind.

In a fever of exultation men ran to and
exchanging fro,
congratulations. The garrison descended the hill, prepared
for a sortie. Vercingetorix had forgotten nothing. His men
were provided with fascines for filling up the trenches, and
Tlio tiiial movable huts to protect their approach. Soon a fierce
combat of horse was raging over the plain. The legionaries
were posted, ready for emergencies, along the outer and the
inner lines. Archers were scattered among the Gallic ranks ;

and the arrows fell so thick and fast that scores of wounded
horsemen were seen riding off the field. Every man fought
like a hero for they knew that from the heights around
;

friends and enemies alike were anxiously watching. The

numbers of the Gauls began to tell and their country-;

men, behind and before, encouraged them by loud yells.

All through the afternoon the battle raged uncertain. But
Caesar had kept his best troops in reserve. Towards sunset
the ever-victorious Germans charged in a compact body,
and threw the division opposed to them into disorder the :

archers were exposed and killed the rout was general

:

and the besieged who had sallied forth turned in despair,

and reascended the hill.^
But Commius and his brother generals were still hopeful.
Next day their men were hard at work, making fascines,
scaling-ladders, and grappling-hooks for a grand assault on
the Roman lines. About midnight they quitted their camp,
and moved in silence across the plain. As they approached
the works, they raised a simultaneous shout, to put the
besieged on the alert and, as they flung their fascines into
;

the ditch, the trumpet was heard, calling the garrison to

arms. Stones flew from slings arrows whizzed through
:

' '
Cavalry encounters with Cavalry,' says Lord Wolseley {The Soldier's
In all

Pocket-book, 1886, p. 370), the side that is able to bring up a fresh reserve
'

when his opponent has exhausted all his, will, as a rule, win the day.'
It lias been said that the Callic cavalry, unaided, could not rescue
Vercingetorix, and that tlie infantry, on the day of tlicir arrival, never stirred.
This judgement seems hardly fair. The infantry could not hope to storm the
Roman lines before they had made the necessary fascines and other implements.
VTi THE REBELLION OF VERCTNGETORIX 177

the air ; and, though the Romans too plied their slings, and 52 b.c.

supports hurried from the neighbouring redoubts to the

relief of any point that was too hardly pressed, the enemy
were too many them, and they suffered heavily
for but :

when those ghostlike companies pressed on to storm the

entrenchment, they trod upon the spikes, or, stumbling into
the pits, impaled themselves on the pointed logs, while heavy
pikes were hurled from rampart and towers into the seething
multitude. The Roman artillery made great havoc. The
losses on either side were very heavy for they were fighting
;

in the dark, and shields were of little use. Towards dawn the
Gauls retreated, fearing an attack in flank and the besieged,;

who had lost much valuable time in attempting to cross the

outer trenches, 1 went back before they could strike a blow.
One more chance remained. The leaders of the relieving
army questioned the rustics about the lie of the ground on
the north and the nature of the Roman defences. Mont
Rea, which bounded the plain and rose above the further
bank of the Oze, extended so far to the north that Caesar
had not been able to enclose it in his line of circumvallation.^
On the southern slope, close to the stream, stood one of the
Roman camps. It was held by two legions perhaps about —
—
eight thousand men under Reginus and Caninius. In order
to avoid observation, it would be necessary to approach the
camp by a wide detour. The Gauls sent scouts to recon-
noitre. It appeared that Mont Rea was connected by a
ridge with a further group of heights. Soon after dark sixty
thousand picked men, under the command of Vercassivel-
launus, left the Gallic camp, and, passing right round the
sweep of the northern hills, halted before daybreak fox a rest
in a hollow north-east of Mont Rea. About noon, just as
they were moving down on the camp, the cavalry, by a pre-
concerted arrangement, streamed over the plain towards the
Roman the rest of the infantry showed themselves in
lines :

front of their encampment and Vercingetorix, observing

;

these movements from the citadel, descended the hill and

moved towards the plain.
This time there was no delay. The inner trench had been
' See pp. 813-5. 2 g^e pp. 361-2.
1093 N
 178 THE REBELLION OF VEROINGETORTX chat.

52 p. c. filled up, where necessary, with earth and fascmes : stout

sappers' huts, destined to protect the men when they should
approach to storm the lines, poles fitted with hooks for
tearing down the rampart, long pikes, and other implements
which Vercingetorix had provided, were carried across and ;

the besieged moved on to make their last effort.

A desperate struggle then began. Wherever there was
a weak spot in the defences, the Gauls threw themselves
upon it and the Romans, comparatively few in numbers,
;

and scattered owing to the vast extent of their lines, found

great difficulty in massing themselves upon the exposed
points. Moreover, they were painfully distracted by the
roar of battle in their rear for both on the inner and the
;

outer line men felt, as they fought, that they must perish if
their comrades behind suffered the enemy to break through.
Yet, agitated as they were, they combated with a nervous
eager energy and the besieged struggled as desperately as
;

they for both knew that that day's fight would decide all
;
:

the Gauls were lost unless they could break the line the ;

Romans, if they could but hold that line, saw their long toil at
an end. From the slope of Flavigny, south of the Ozerain, the
view from which embraced the whole plain, Caesar directed
the battle, and sent supports to every point where he saw
his men hard pressed. The attack on the circumvallation
in the plain was comparatively feeble for the bulk of the
;

relieving force was formidable only in numbers. Nor were

those numbers wisely directed. The Aedui may have been
treacherous the generals may have disagreed, or they may
:

have been fettered by the civil commissioners anyhow ;

the Gauls made no serious attempt, except on Mont Rea and

in the plain.^ The fighting was fiercest on Mont Rea. The
Gauls were so numerous that Vercassivellaunus could always
send fresh men to relieve their comrades. Coming down on
the camp from a higher level, the assailants hurled their
missiles with fatal momentum they shot earth in heaps
:

over the pointed logs and the spikes, and, locking their
shields over their heads, passed unscathed to the rampart

1 It is not absolutely certain that the relieving army did more than make
a demonstration even in the plain. 8ee p. 810.
VII THE REBELLION OF VEROINGETORIX 170

and then their numbers began to tell. The Roman cavalry 52 b.o,

were unable to create a diversion they were kept inac-

; for
tive by the Gallic troops in the plain.^ Suddenly a galloper
rode up and told Caesar that the garrison were worn out, and
their stock of missiles f ailing. ^ He immediately sent Labienus
with six cohorts to the rescue, telling him to hold on as long
as he could, and, when he could hold on no longer, to sally
forth, and fight it out in the open. Then, riding down
between the lines on to the plain, he harangued his weary
soldiers and adjured them not to give in just one short :

hour, and the prize was won. At last the besieged abandoned
in despair the attempt to break through, and, wheeling to
the left, crossed the Ozerain, and flung themselves against
the works at the foot of Flavigny. They drove the artillery-
men from the towers with volleys of missiles they shot ;

earth and fascines into the ditch, and made their way across ;

they tore down rampart and breastworks with their grap-

pling-hooks six ^ cohorts, then seven ^ more were sent down
;

—
to help, and still they pressed on, till Caesar himself hurried
to the spot with fresh reinforcements, and drove them away.
Everywhere, except at Mont Rea, the victory was won.
Caesar called out four cohorts from the nearest redoubt, told
his cavalry to follow him, and sent a horseman galloping to
the northern cavalry camp to send another detachment down
upon the enemy's rear.^ They were now swarming over the
rampart and, as a last resource, Labienus summoned every
;

available man from the neighbouring redoubts to his aid.

By good luck these reinforcements amounted to eleven
cohorts, —perhaps thousand men.
four And now, con-
spicuous in his crimson cloak, Caesar was descried, hurrying
across the plain. The enemy made a supreme effort but ;

they were left without support. Labienus and his men took
heart, and rushed into the thick of the stormers. As Caesar
approached, he heard the shouts of the combatants he sav/ :

the camp abandoned and the short swords flashing over the
slopes beyond. Suddenly the cavalry appeared on the
1 See Pro Alesia, 1909, p. 582. 2
gg^ p^ 819.
^ The numbers are uncertain, and those which I have given have no other
authority than the Aldine edition. * See
pp. 818-9.
N 2
 180 THE REBELLION OF VERCTNGETORIX chap.

52 B.C. heights above the enemy's rear : Caesar's reserves came up

to attack them in front ; and they fled inbewilderment,
into the midst of the hostile squadrons.^ Vercassivellaunus
himself was captured, and seventy-four standards ; and of
the sixty thousand chosen men who had marched out of
camp the night before only a remnant returned. The whole
scene was visible from the town and in despair the officers
;

left in command sent to recall their comrades from below.

The vast host without vanished in the gathering darkness.
The legions were too tired to follow, or all might have been
destroyed but at midnight the cavalry were sent in pursuit
:
;

and when day broke, they were still hunting the fugitives
and capturing or slaying them in scores.
The self- All was lost so Vercingetorix clearly saw.
: In the night
Vercinse^ he formed his resolve. Next morning he gathered the tribal
torix. chiefs around him. He told them that he had fought, not
for himself but for national liberty and, since they must ;

needs all bow to fortune, he was ready to place himself at

their disposal, —to die, if they wished to appease the Romans
by his death, or to yield himself up as a prisoner of war.
They accepted his offer, and consented to purchase life by
sacrificing the leader of their own choice. Ambassadors were
sent to learn the pleasure of the conqueror. He ordered the
chiefs of the garrison to be brought out, and all the arms to
be surrendered. The chiefs were led forth and Caesar, ;

Surrender seated on his tribunal, received their submission. Vercinge-

garrison
torix, mounted on a gaily caparisoned charger, rode round
the tribunal, and then, leaping to the ground, took off his
armour, laid down his sword, and bowed himself at Caesar's
feet.^ He was sent to Rome, and imprisoned in a dungeon.
Six years later he was brought out, to adorn Caesar's
triumph and then he was put to death.*
;

1 M. Pernet {Pro Alesia, 1909, p. 582), remarking that the Roman cavalry
in the northern camp (G) were left free to act, conjectures that the leaders
of the relieving that the camp existed. But even if this
army were unaware
squadron had been neutralized, Vercingetorix would not have been saved.
2 All questions relating to the operations at Alesia are discussed on
pp. 804-20.
3 See p. 820. M. Jullian {Vercingetorix, pp. 306, 310) regards Vercingetorix'
surrender as un acte de devotion religieuse. ... II s'offrit a Cesar et aux dioux
'

suivant le rite mysterieux des expiations volontaires '. * See p. 821.

vii THE REBELLION OF VERCINGETORIX 181

Two thousand away; and still the

years have passed 52b.c!.

name of Vercmgetorix retains its hold upon the imagination. J[)|!j^^"fj^'

Our neighbours think of him as the Germans think of his place

Arminius and the Scots of Wallace and the traveller who

;
^ory!^
stands upon the wind-swept plateau of Gergovia and looks
down upon the vmeyards that cover the slopes over which
he drove Caesar's legions, or, speeding on his way to the Swiss
mountains, looks out, as the train whirls him past the
station of LesLaumes, upon the colossal statue which marks
the western promontory of Mont Auxois, must be dull indeed
if he does not sympathize with the nation's veneration for

the great Gaul. Looking back across that vast gulf of time,
we behold him, as he appears by the testimony of his con-
queror, not only a chivalrous patriot, but also a born leader
of men. In this character he is the equal of Caesar himself.
The Gauls and their descendants have sometimes mistaken
a charlatan for a hero but the hero to whom they are loyal
;

while they are smarting under a defeat, must be a hero

still

indeed. When Vercmgetorix at Avaricum regained his

ascendancy over the fickle Celtic multitude, he showed a
knowledge of human nature as profound as Caesar when he
quelled the mutiny of the Tenth Legion. If he knew how
to use flattery as an instrument for fortifying self-respect,
he never condescended to the arts of the demagogue he :

could tell wholesome truths, however unpalatable and with

;

the most winning persuasiveness he possessed a capacity for

being terribly severe. He recognized the softness of moral
fibre, the mollities animi, which in the Gauls coexisted with
personal bravery ;and with springing energy he stimulated
them to transmute that weakness into strength, to undergo
toils from which they had ever shrunk, and to sacrifice their

particular interests for the national weal. Who shall

—
imagine the intensity with which he lived ? within that
year the youth became a veteran. Those only who have
some experience of affairs can appreciate the genius for
organization, the unremitting toil, the sleepless vigilance
that were needed to force those diverse levies into the field,
to arm and clothe and feed them, to direct their opera-
tions, to procure information, to raise money, to negotiate, to
 182 THE REBELLION OF VEKCINGETOKIX chai'.

52 B. a. bribo, to persuade. must, moreover, be remembered that

It
his power depended upon sheer unaided foree of charaeter :

he might eontrol only so long as he could please his com- :

mission was held at the pleasure, nay the caprice, of the

most inconstant of the races of men. Yet, alone among
the Gallic leaders, he united the discordant elements of the
greater part of Celtican Gaul and, by his tact in gaining
;

over the dissentient tribes, he drove one of the greatest

generals of the world, whose army was numbers
in all but
far superior to his, to the point of withdrawing from the
theatre of war. That he was compelled at la-st to '
bow to
fortune was due to the jealousy of rivals, to the self-seeking
'

of the Remi, to the treachery of the Aedui, who prevented *

Gaul from making sure of success ', to the littleness of those

who could not face the sacrifice that would have forced the
Romans either to retreat or to starve, to the master-stroke
by which Labienus saved from the results of his own
his chief
error, above all to the fatal weakness that opened the way
to Alesia. Then, and then only, Verchigetorix was not
himself ;and the errors of the general in his first and last
campaign are forgotten when we contemplate the man. In
him idealism was steadied by vision of the real imagmation ;

was served by mastery of detail enthusiasm Avas sustahied ;

by resolution the brain was fellow-worker with the heart.

;

He fought to make Gaul free by making her one and what ;

though this endeavour, almost successful, did ultimately fail?

Gaul had to work out her salvation in another way, because
she was not ripe to obey her leader's voice. To be great, and
yet to fail, is tragical but Nature sees to it that her noblest
;

children shall not waste their strength. If Vercingetorix,

like Hannibal, was the hero of a lost cause, his influence also
is inexhaustible. For the achievement is less than the quest
of the ideal and no heroic deed was ever done in vain.
;

• • • • •

Caesar Caesar determined, instead of going to Italy, to spend the

^1
ciistfi*
butes his winter in the Aeduan capital for he knew that though Gaul
;

Icsjions for ^r^s sore smitten, it was not yet tranquillized. For the
(ho win-
lor. moment, all was still. The Aedui wcvc ready and eager to M
return to their allegiance. The Arverni, ^^'llo had given iiu '
 ^
;

MI THE REBELLION OF VERCINGETORIX 183

trouble former years, were quite cowed, and promised

iii 62 b. c
implicit obedience for the future. Caesar was too politic
to bear hardly upon either. He therefore restored to them
the prisoners whom he had made, though he demanded
a large number of hostages. But the soldiers had to be
rewarded for their protracted labours and every man ;

received, by way of booty, a prisoner, whom he might sell

as a slave. Caesar was generous as well as politic and ;

doubtless his officers were not overlooked. The fortunes

which had been amassed by Labienus and the notorious
Mamurra disgusted Cicero,^ though Caesar was willing to
provide for any friend whom he might recommend. ^ For
himself, there was no law of prize to limit the general's share.
When he came to Gaul, he was poor and in debt when he :

quitted Gaul, he was rich enough to lend and to bribe.^

The legions were quartered for the winter among the Remi,
the Sequani, the Aedui, the Ambivareti, the Bituriges, and
the Ruteni, that is to say, around Reims, Besan9on, Mont
Beuvray, Chalon and Macon, Bourges, and Rodez.^ By this
arrangement the friendly Remi would be protected from 'the
vengeance of the Bellovaci the submission of the Aedui
:

was assured the legions quartered among them could easily

:

communicate, on the east, through the territory of the friendly

Lingones, with their comrades in Sequania, on the north-east
with those quartered among the Remi the Arverni were:

hemmed in on the north by the legion which menaced the

Bituriges, on the south by that which watched the Ruteni
and this last was on the borders of the Province, whence it
could, if necessary, summon aid. Thus the troops were
distributed in such a way as to safeguard the loyal, to
overawe the disaffected, to cover the Province, and to be
ready for mutual support.
1 AtL, vii, 7, § 6. 2 F^m., vii, 8, § 1;17, § 3 Q. jr., ii, 14, § 3.
;

* In law of course the whole booty belonged to the state

strict but the
:

general had the right to retain what he deemed necessary for the successful
conduct of the war and the results were what might have been expected.
;

See Daremberg and Saglio, Diet, desant. grecques et rom., iv, 610-1, s.v. Praeda,
and cf. Cicero, De prov. cons., 11, § 28.
' !Scc Long's Decline of the Roman Ilepublic, v, 475 Cicero, All., vii, 3, § 11
; ;

7, § () ; and ISuctonius, Divus lulius, Qi.

" Tlio habitat of the Ajnbivarcti is uncertain. 8cc p. 308.
 CHAPTER VIII

THE END OF THE STRUGGLE

52 B. c. The victory at Alesia was decisive. Their great leader
Effects of gone, their entire host shattered, like a billow surging against
Caesar's
victory at a rock, by the little army which it had marched to destroy,
Alosia.
the confederacy was dissolved as quickly as it had been

formed.
Various Nevertheless some of the more resolute patriots were pre-
tribes pre-
pare to
paring to renew the struggle.They knew, indeed, that all
I'enew the the men whom they could muster had no chance of standing
struggle.
against Caesar in a pitched battle ; but they allowed them-
selves to hope that, if they all rose simultaneously, his forces
would not be strong enough to engage them all at once in
detail. Such is the account, based probably upon the reports
of Caesar's spies, which Aulus Hirtius ^ has given us. But
it may perhaps be doubted by those who have analysed

his narrative whether the rebellious tribes had any such

definite and concerted plan. It is probable that they
were actuated, not jointly but severally, by sheer abhor-
rence of a foreign yoke, by imllen despair, by desire for
plunder, perhaj^s by the vague hope that when Caesar
was gone, his successor would leave such obstinate rebels
to themselves.
Caesar The Bituriges, who had not forgotten the slaughter at
disperses
the
Avaricum, were the first to stir. The single legion which
Bituriges had been quartered in their country was powerless to restrain
and Car-
nutes. them. Caesar was anxious to give a long rest to his soldiers,
51 B.C. who were tired out by the extraordinary duration and severity
of the late campaign but before the year was out he took
:

the field and while the chiefs were still talkmg over their
;

^
The last hook of the Cunittuntaries on the Gallic War was written, not by
Caesar, but by his friend Aulus Hirtius.
 THE END OF THE STRUGGLE 185

plans, another legionwas upon them. Thousands of peasants 51 b.o.

were captured while they were working in the fields others :

had just time to flee but hurry where they might, Caesar
:

was too quick for them and his swiftness so impressed men's
;

minds that the friendly tribes saw that it was their interest
to remain loyal to a Governor who was strong enough both
to protect and to punish, while waverers hastened to sue for
peace. Caesar sent the legions back to quarters with the
promise of a substantial present for every officer and man ;

while he himself returned to his civil work at Bibracte.

But in little more than a fortnight his rest was interrupted.
When the humbled Bituriges begged for his aid against
the Carnutes, who had turned upon them, he called out the
legions at Chalon and Macon and, on the mere rumour of
;

his coming, the Carnutes fled in every direction. Chased

from place to place by cavalry and auxiliary infantry,
numbed by the cold and drenched by the rains, they finally
dispersed among the neighbouring tribes and their pursuers ;

returned, laden with plunder. The lesson sufficed for the

time ; but the legions were left at Cenabum, to keep the
tribesmen in awe.
Still, there was another tribe to be reckoned with, the Campaign
warlike Bellovaci, who, six years before, had headed the Belgic t^e^^^
league. They had some grudge against the Suessiones, whom Bellovaci.

Caesar had placed in dependence upon his steady allies,

the Remi, and were mustering their forces and those of the
neighbouring tribes to attack them. The confederacy com-
prised the Atrebates, the Ambiani, the Veliocasses, the Caleti,
and the Eburo vices, who inhabited the districts round Arras,
Amiens, Rouen, Lillebonne, and Evreux. The leaders were
a Bellovacan chief called Correus, and Commius, whose spirit
was not subdued by his defeat at Alesia. On Caesar's
approach they established themselves in the forest of Com-
piegne, on Mont St. Marc, a hill protected by a marshy
watercourse, which oozed northward into the river Aisne.^
Caesar's force consisted of four legions, which, without
reckoning auxiliaries, probably numbered about fifteen thou-
sand men ;and, remembering the tactics of Ariovistus, he
1 8ee i)p. 820-30.
180 ^JMiE END OF THE STRUGGLE chap.

had hired ii corps of German light mfantry to support his

cavahy. One legion had already served against the Bituriges
two, under Fabius, were withdrawn from the neighbouring
camp in the country of the Remi while the other had ;

marched from Labienus's quarters at Vesontio. Caesar

was very anxious to bring on a battle, and advanced in
a formation which partly concealed his strength but the :

enemy were too wary to quit their vantage ground their ;

numbers were great and the hill, rising abruptly above the
;

further side of the deep valley, was hard to ascend. Accord-

ingly he encamped on Mont St. Pierre, the height just
opposite theirs. The fortifications which he constructed were
of extraordinary strength ; for he hoped that the enemy
would be emboldened by his caution to attack him, and, as
his foragers were obliged to go long distances, it was necessary
that the camp should be defensible by a comparatively small
force. During the next few days frequent skirmishes took
place with varying results but nothing woidd mduce the
;

enemy to come out and liazard a general action. They

succeeded, however, m
surrounding some foraging parties,
which were unavoidably isolated. These reverses were magni-
fied by rumours, which gladdened Caesar's enemies at Rome.
They had been annihilated,
told each other that his cavalry
tJiat the 7th legion had suffered a defeat, and that lie

himself was hemmed in by the Bellovaci. The Gauls

indeed were elated by their success and their elation was ;

increased by the arrival of five hundred German horsemen,

whom Commius had induced to serve. It was impossible
to storm the camp without excessive loss ; and, as a large
forcewas needed to invest it, Caesar sent for the three legions
which he had left at Cenabum and in the country of the
Bituriges.
When the rebel leaders heard of their approach ,i they
remembered the dismal fate of Alesia, and determined to
send non-combatants and baggage in the night.
off their
The long line of wagons was barely in motion when day
broke, and the Romans caught sight of them. The enemy
formed up in front of their camp to co^'er the retreat,
'
Cicero, Fani., viii, 1,^4.

I
vm THE END OF THE STRUGGLE 187

iiitoiidiiig to follow as soon as possible. Caesar was too 51 b.c.

wary to attempt to fight his way up that steep aseent but ;

he determined not to let the enemy move off unseathed.

On their left and separated from their eamp only by a narrow
depression, was a plateau with gently sloping sides. Caesar
rapidly bridged the marsh, led his troops across, ascended
the plateau, and just on its edge placed engines to throw
missiles against the enemy's masses. They dared not send
off their troops, for fear they might become confused as they
broke into detachments, and fall victims to the Roman
cavalry. For some hours, therefore, they remained under
arms. Caesar made a new camp on the plateau, formed up
the legions in front of it, and kept the troop-horses bridled,
ready to charge at a moment's notice. Towards nightfall, as
the enemy could not remain where they were any longer
without food, they had recourse to a stratagem. Bundles of
straw and sticks were laid in front of the line and set ablaze.
In a moment a vast wall of flame hid the entire multitude,
and they instantly fled. Suspecting, though he could not
see what they had done, Caesar made the legions advance
cautiously, and sent his cavalry up the hill in pursuit. But
the cavalry were afraid to ride through the fiery barrier ;

and a few bold troopers who spurred in could hardly see

their horses' heads for the smoke. Meanwhile the enemy
were well on their way down the valley of the Aisne and ;

having crossed the Oise, of which it is a tributary, they

encamped on Mont Ganelon in the plain beyond.
On the southern bank of the Aisne, in the angle formed
by its confluence with the Oise, there was a large meadow,
the luxuriance of which, Correus expected, would attract the
Roman foragers. In the woods which encompassed this
meadow he posted a strong force of horse and foot. Having
learned his design from a prisoner, Caesar sent his foragers
down the valley of the Aisne, escorted by cavalry and light-
armed auxiliaries, and, following in support with the infantry,
took post as near the meadow as he could without being
perceived. Discerning the cavalr^r as they approached, the
Gauls rode out from the wood and charged but the dis-
:

ciplined squadrons sustained the shock with admirable

 188 THE END OF THE STRUGGLE chap.

51 B.C. coolness supported by the auxiliaries, they baftted every

;

effort to outflank them and they had already won the day
;

when the infantry appeared. The flying Gauls, caught m

their own trap, were hunted down and slaughtered in the
woods and by the banks of the Oise. But Correus would
neither yield nor fly. Standing alone upon the field, refusing
to accept he struck fiercely at his opponents
quarter,
and wounded numbers of them, until, infuriated by his
obstinacy, they hurled a volley of javelins into his body, and
he fell dead.
This was the expiring effort of the Bellovaci. Commius
escaped to wage a guerrilla warfare, but ultimately made his
peace with the conqueror, stipulating only that, as a con-
cession to his well-grounded fears, he might never again look
upon the face of a Roman.^ Those who had remained in
camp appealed to Caesar's clemency, and obtained a con-
temptuous forgiveness. Their excuse was that Correus had
stirred up the populace to rebel, in defiance of the senate.
Caesar reminded them that they had borne arms against him
before it was easy to blame the dead, but no single man
:

could raise a revolt with the support of a mere rabble if the

friends of order were determined to prevent him. The allies
of the Bellovaci, who had been waiting to see what terms
they would obtain, submitted likewise and gave hostages.
From many parts people were actually emigrating, so intense
was their reluctance to submit to the authority of Rome ;

but Caesar distributed his legions in such a way as to bar their

escape. Caninius, who had wintered near Rodez, had been
transferred to Poitou, the country of the Pictones, where
with two weak legions he could barely hold his own and ;

Fabius was sent with twenty-five cohorts to reinforce him.

Caesar himself, accompanied by Labienus, and reinforced
by a legion under Mark Antony, marched against the
remnant of the Eburones, and sent out flying columns
everywhere to ravage, burn, and slay. Ambiorix evidently
w^as not to be captured but Caesar resolved that the
;

wretched man should never dare to show his face again

'
An account of the aucccssful clo80 of liiis career will be found in my
Anc. Britain, pp. 361, 305-0, 371.
UXELLODTINUM
 °®

VIII THE END OF THE STRUGGLE 189

among the people upon whom he had brought such a terrible 51 b. c.

doom.
The end was at hand. Labienus, who had been com- Caninius

missioned to deal the irreconcilable Treveri a final blow, pabius

captured their chieftains, and among them an Aeduan noble, compel
Surus, who, scornmg the tergiversations oi his countrymen, custo
remained a brave, unwavering, incorruptible patriot. The g^^^^ ^f
most warlike states were now subdued or overawed only Lemo- :

^^^'
some tribes in the west were still restless. A rebel chief
named Dumnacus, with a motley force from Brittany and the
country round Orleans and Chart res, was besieging Lemonum,
on the site of the modern Poitiers, in which an adherent of
Caesar had taken refuge. Caninius, who was too weak to
relieve the town, entrenched himself hard by and Dumnacus
;

wasted several days in trying to dislodge him. After he

had returned to Lemonum, Fabius, who had received the
submission of various tribes, marched up rapidly, compelled
him to raise the siege, and while he was hurrying to escape
across the Loire, pounced upon him and defeated him twice
with heavy loss. Flying alone from the battlefield, he
wandered to the uttermost parts of Gaul and history knows
;

him no more. Two thousand of the fugitives, rallied by the

Senonian adventurer, Drappes, and Lucterius, the chief who Drappes
IT
had so ably supported Vercingetorix, went off to plunder the teriustake
Province but, finding themselves hotly pursued by Caninius,
;
refuge in

they took refuge in the fortress of Uxellodunum, the modern dunum.

Puy d'lssolu,^ of which, before the great rebellion, Lucterius
had been the over-lord.
They had hardly shut the gates before their pursuers Blockade

arrived. The hill overlooked the left bank of the river Tour- dunum.
mente, which, about two miles to the south-west, emptied
itself into the Dordogne. It rose fully six hundred feet
above the valley ; steep rocks on every side forbade any
attempt to assault and to approach by an embankment
:

was utterly impracticable. Caninius, therefore, proceeded

to invest the town. On the west, rising above the valley of
the Tourmente, and on the north-east, linked to the strong-
hold by a broad neck of land, there were hills of considerable
1
See pp. 483-93.
 190 THE END OF THE STRUGGLE chap.

51 B.C. height. CaninmR made two camps on the former and one
on the latter, and began to connect them by a line of con-
tra vallat ion. Watching the progress of the works, the
garrison remembered tlie story of Alesia Lucterius had been
:

there, and knew how Vercingetorix and his people had

suffered ; unless his own men bestirred themselves at once,
they too would be starved into surrender. It was agreed
that Lucterius and Drappes should make an attempt to
procure supplies. On the following night, leaving two
thousand men to hold the town, they stole out with the
rest of the force. For several days they scoured the sur-
rounding country, collecting corn. During this time they
occasionally attacked the Romans by night with such vigour
that Caninius was obliged to suspend the construction of
his lines. One morning, in the early twilight, the Roman
sentries heard an unusual noise scouts were sent out, and
:

returned with the news that a string of pack-horses was

moving up a narrow path leading to the town. The troops
instantly turned out the drivers rushed helter-skelter down
:

the hill and the escort were slaughtered almost to a man.

;

Lucterius with a few followers escaped. Within a few hours

another division under Drappes, encamped about ten miles
off, was surprised and every man who escaped the sword was
;

made prisoner. But Drappes did not remain a captive long.

Resenting the indignity of bonds, or dreading the penalty of
obstinate rebellion, he starved himself to death.
Meanwhile Caninius was reinforced by the legions of
Fabius, who had just concluded a most successful expedition
along the valley of the Loire. Promptly following up his
victory over Dumnacus, he had fallen upon the Carnutes,
who, having suffered severely in that battle, were ill pre-
pared to resist. This warlike people, who had never been
thoroughly subdued, were now completely cowed and forced
to give hostages and the maritime states of Brittany,
;

which, like them, had supported Dumnacus, hastened to

follow their example. Caesar, who had been making a
political progress, and trying to humbled chiefs,
conciliate the
was now at Cenabum while Mark Antony, whom he had
;

left behind with five thousand men, was holding the Belgae in
VIII THE END OF THE STRUGGLE 191

awe. The Carnutes were still uneasy at the remembranee 51 b.c

of the provocation which they had given in the great revolt ;

and it seemed likely that despair might drive them to fresh

excesses. Caesar saw that the only way to restore their
confidence was to make an example of the chief who had led
them astray, and frankly to forgive the rest. He therefore
demanded that Cotuatus, who had been the author of the
massacre at Cenabum in the preceding year, should be
delivered up to him for punishment and the people, eager
;

to purchase the favour of the conqueror, hunted him down

and brought him a prisoner into the Roman camp. Caesar,
if Hirtius is to be believed, was unwilling to order his exe-

cution, but could not afford to disregard the clamours of

the soldiery. But Caesar knew how to
any clamour silence ; Execu-
and, if he had told the story himself, he would have told it
^^°J^ ^J
without excuse. The wretched man was flogged till he was
insensible and his head was cut off.
;

Caesar now received a series of dispatches informing him Caesar

of the obstinate resistance of Uxellodunum.
Contemptible f^orUxeflo-
as were the numbers of the rebels, their example might dumim.
encourage other states to renew the wearing struggle. Only
one more summer had to pass, as the malcontents had doubt-
less reckoned, and his government would be at an end.^ But
Caesar determined that, before that time, they should be for
ever subdued. Taking his cavalry with him, he hurried
southward, followed by two legions, for Uxellodunum.
He instantly detected the weak point in the enemy's He cuts
position.
^ His lieutenants had merely intended a blockade. ^^^^^
*^
,
garrison
But the garrison were amply provisioned ;
^ and the only from their
effectual way of reducing them was to cut off their supply of ^^^/
water. Archers, slingers, and artillery were posted on the
western bank of the Tourmente, so as to command every
approach to the stream. Thus menaced, the enemy were
afraid to descend and thenceforward they could get no
;

water except from a spring on the western slope of the hill.

1 See pp. 832-4.

^ It must be remembered that, although the attempt to procure fresh
supplies had failed, the mimbers of the garrison had been greatly reduced,
and therefore there were far fewer people to feed.
 192 THE END OF THE STRUGGLE chap.

51 B.C. Opposite this spring, Caesar proceeded to construct a terrace.

From the heights above, the enemy hurled down missiles,
and many of the Romans were struck : but the rest toiled
doggedly on and the terrace was built up nearer and nearer
;

still. A tower was erected upon it, of the extraordinary

height of ten stories, high enough to overtop the spring and ;

the garrison dared not approach under the shower of stones

and arrows which its engines rained down. Men and cattle
alike were parched by thirst. Torture and death stared them
in the face. But there was the spring, still gushing forth.
As a last resource, number of
the garrison set fire to a
barrels, filled with pitch, grease, and shavings, and rolled
them on to the terrace. The woodwork and the sheds were
presently in a blaze. The garrison with desperate energy
flung down missiles to deter the Romans from advancing to
put out the fire. But right up against the roar'ng flames,
undaunted by the missiles, unheeding the sight of their
fallingcomrades, the Roman soldiers pressed steadily on :

with a mighty shout they answered their enemy's yells and ;

each man, eager that his valour should be observed, fought

as he had never fought before. Still the flames shot up ;

and precious lives were sacrificed in vain. In this extremity,

Caesar sent a number of cohorts to climb the hill and feign
an assault upon the town. Panic-stricken, the garrison
recalled their comrades from below and the moment they
;

had turned their backs, the Romans ran forward and

extinguished the flames. Still the Gauls held out ; for the
spring itself untouched. At length, however, a
was still

party of sappers crept through a gallery which had been

secretly driven into the hill-side to the source of the spring.
Surrender and diverted its flow. Then at last, feeling that Heaven was
fighting against them, the garrison surrendered.
Caesar saw that, if these rebellions were to break forth
again and again, his work would never be at an end. He
was aware, says Hirtius, that his clemency was notorious,
and had no fear that any measures which he might be forced
to adopt would be misunderstood. He determined, there-
fore, to inflict upon the garrison a punishment so appalling
that all malcontents should in future remain quiet. He

I
VIII THE END OF THE STRUGGLE 193

would not put he did, their

his prisoners to death, because, if 51 B.a
fate, thougli it might be talked of for a time, would soon be
forgotten. They were to remain as a living warning to
intending rebels. He ordered their hands to be cut off, and
sent them forth to exist as they best might.
One notable survivor of the great rebellion was still at The fate of
large. Lucterius, the lieutenant of Vercingetorix, a man who,
as Caesar said, was ready to dare anything, had wandered
far from Uxellodunum. He knew that for him there was
no forgiveness and he went from place to place in fear of
;

betrayal. At length he fell into the hands of a renegade

Arvernian, who brought him in chains to Caesar and what ;

was his fate we can only guess.

But Caesar knew that conquest can never be complete Caesar

until coercion has been followed by conciliation. In little coercion^^

more than a year he would be leaving the country and he ;
by con-
.
, r^i . . 1 1
ciliation.
must contrive to leave it at peace. Ihe time had not come,
nor had he the authority, to organize a government it would :

be enough if his successors could enter upon that task with-

out encountering opposition. He had no wish to oppress the
Gauls, or to hurt their national pride on the contrary, he :

desired that they should learn to feel themselves really citizens

of Rome. He fixed their tribute at a moderate amount .^ He
did not interfere with their institutions, though he doubtless
used his influence to promote his own adherents to power.
He loaded the chiefs with presents he won their hearts by :

the charm of his address and when he quitted Gaul, and

;

threw down the gauntlet, on a wider arena, to a mightier foe,

they sent their bravest warriors to fight under his flag.^

'
40,000,000 sesterces or about £400,000. See p. 838.
2 B.G.,ym,4d; Cicero, ^^^, ix, 13 ; 5. C, i, 39, §2. See pp. 835-8.

1093 O
 CHAPTER IX
CONCLUSION
The
conquest of Gaul, fraught with ilUmitable issues, was
at last complete.^ Destiny had decided that Gaul was to
be either German or Roman and Caesar did not hesitate
;

to grasp the gift of destiny for Rome. The Gallic warriors

were perhaps as brave, man
man, as the Roman legiona-
for
ries and their numbers were far greater. But, whatever
;

may have been their political capacity, when Caesar came

among them they were only feeling after political union :

they did not combine to expel him imtil it w^as too late, and
not with a whole heart even then. With all their dash and
nervous enthusiasm, they lacked the tenacity of the Roman :

rushing vehemently to the attack, they fell away at the first

reverse. This weakness, which Caesar so often notices, may
have been inherent in the race it may have been wholly or
:

in part the result of a want of mutual confidence ^ but :

whatever the cause, the fact remained. Their naive credulity

was observed by Posidonius ^ and, unwarned by experience,
;

^ This statement will naturally be taken in a general sense. The subjugation

of the north-western part of the country was doubtless, as Mommsen says {R<'>m.
Gesch., V, 1894, p. 72 [Hist, of Borne, —
The Provinces, i, 79]), comparatively
superficial : there was fighting in Aquitania in 38 and 28-27 b. c. and there ;

was a partial insurrection in the reign of Tiberius. Still, tlie thoroughness

with which Caesar had done his work was demonstrated, first b}' the peace

which prevailed during the civil war, when Gaul was almost entirely denuded
of troops, and secondly by the facts that, during the long reign of Augustus,
notwithstanding the disturbances in Germany, Gaul remained submissive, and
that, as Mommsen puts it {ib., pp. 73-4 [80-1 J), Vercingetorix found no suc-
cessor. See also F. de Coulanges, Hist, des inst. pol. de Vancienyie France, la —
Gaule rom., pp. 71-84, and Desjardins, Geogr. de la Gaule rom., iii, 49-50.
* The numerous host was defeated by Sir Charles
of the warlike Baluchis
Napier's little it was a loose aggregate of
force at Miani principally because
tribal levies which had not been trained to act in concert (see my article on
the battle of Miani in MacmillarCs Magazine, January, 1900) and it is prob- ;

able that the defeats which the Gauls suffered were partly due to the same
defect. 3 Strabo, iv, 4, §§ 2, 5.
 CONCLUSION 195

they precipitated themselves eagerly, again and again, into

the simplest traps. Nor, for the most part, were the hetero-
geneous levies who opposed Caesar the equals of the purer
Gauls who had routed a Roman army on the banks of the
Allia. The Helvetii, the Parisii, the Senones, and a few of
the Belgic tribes alone maintained the ancient renown of the
Celtic infantry. The Gauls had no regular army they had :

no science they had no discipline and, until Vercingetorix

: ;

arose, they had no great leader. Their conqueror, on the

other hand, was master of a compact, disciplined, and well-
equipped army, the finest in the world ^ he was free to :

pursue a definite aim in opposition to the sporadic efforts of

his enemies and, while he became a general only to achieve
;

higher ends, he was one of the greatest generals that have

ever lived. His writings leave so much to the intelligence
of the expert that few can conceive how hard it was to
conduct the operations which, in the narrative, appear so
easy ; what resolution was needed to adhere, in the face
of unforeseen obstacles, to plans readily formed, to banish
distracting doubts, to preserve equanimity under the friction
of accumulating difficulties, to sustain the military virtue
of the army in privation and in the bitterness of defeat, to
carry out combinations when calculations were disturbed.^
How Caesar did these things the war-bred soldier can alone
realize but we can all form some conception if we rightly
;

study what he wrote. He knew that a well-organized com-

missariat is the foundation of success in war and the truth ;

of this maxim is borne in at every turn upon the reader of

his memoirs. While his enemies were more than once obliged
to strike prematurely or to disperse because they had not

^ It has been asserted that the legionaries with whom Caesar conquered Gaul
were themselves Gauls. No one could make a statement so misleading who
had any knowledge of ethnology, or who had noted the emphasis with which
Caesar marks the distinction, in regard to stature, between the Gauls and his
legionaries {B. G., ii, 30, § 4). All the legions which he raised during the
Gallic war, with one exception (see pp. 802-3), were levied from the mixed
population, composed of Italian, Gallic, Ligurian, and doubtless also Etruscan
and aboriginal elements, which inhabited Piedmont and the Plain of Lombardy.
^ '
Everything,' says Clausewitz {On War, translated by Col. J. J. Graham,
i, 1873, p. 40), everything is very simple in war, but the simplest thing is
'

difficult.'

2
196 CONCLUSION chap.

secured their means of snbsisteneo, he was always able to keep

his army together and to choose his own time. For a few days'
raid the legionaries could carry their food on their backs
but whenever his operations were likely to be protracted, he
stored his grain in magazines and provided for its transport
and protection. His secret service was perfectly organized.
His geographical intuition was as unerring as that of
Napoleon. He knew both how to govern and how to
fascinate his soldiers, so that they would strain every nerve
to win his praise, —all the more because they saw that he

was more careful of their lives than of his OAvn. Emergencies

the most sudden and alarming, even when they resulted
from his own mistakes, seemed only to make him more calm.
He was not only master of all the science of his time, but
he showed an inexhaustible fertility in inventing expedients.
He concentrated his strength upon the vital point he was :

always ready to put everything to the hazard for a great end.

He knew the rashness of his enemy, and lured him on by an
affectation of fear. He confounded him by the swiftness of
his marches :he used his reserves with decisive effect and ;

when he had won the victory, he followed it up with an

energy that overwhelmed.
Nor would it be just to forget the support which the
general received from his lieutenants. Few of them failed
to do what was required ;and one may fairly rank among the
great marshals of the world. The genius of Labienus has
not been adequately appreciated but it needs little insight
;

to see that Caesar placed him in a class by himself. Caesar

trusted him to the full and, so long as his engagement
;

lasted, that faithless man was true. The most difficult enter-
prises were imposed upon him and he accomplished them
;

all. He fulfilled his instructions to the letter he assumed

:

responsibilities without fear. Beset by dangers the most

appalling, his judgement was unerring, his decision unfalter-
ing. In the crisis of the most critical campaign he avenged
his chief's defeat by victory in the crisis of Alesia he repelled
;

the fiercest onslaught, and struck the decisive blow and ;

throughout those eight years, from first to last, he never

made a single mistake.
IX CONCLUSION 197

But Caesar's was the directing miiid. And Caesar was

much more than a great general. He was a far-seeing states-
man and withal a dexterous politician. Many historians
have affirmed that the oligarchies in the Gallic states sup-
ported him, and that the adventurers who aimed at winning
royal power were his opponents. There is some truth in this
view ; but it needs qualification. No generalization can be
safely made about the attitude of the various parties in Gaul.
Caesar shaped his policy according to circumstances and ;

if Dumnorix and Indutiomarus were his enemies, he himself,

as we have seen, set up kings, perhaps not always wisely, in

various republican states. In two notable instances, indeed,
this policy miscarried : but the Senones and Carnutes were
Caesar's most persistent a,ntagonists and we do not know ;

enough to judge whether by any other means he could have

kept them on his side. Commius served him well until the
great rebellion and he had no reason to repent of the choice
;

which he had made. With cool calculation he took advantage

of the fears, the necessities, the jealousies, the intestine
broils, the spasmodic revolutions, the petty ambitions of
those incoherent multitudes. For it must never be forgotten
that, as we conquered India with the aid of Indians, Caesar
conquered Gaul with the aid of Gauls. Fortune, which, as he
so often said, ' power in war, as in all other affairs,'
is a great
was no fickle friend to him. The aristocracies broke up the
triumvirate which threatened to anticipate him, and deprived
the Helvetii, who, unaided, were almost a match for him, of
allies who would have been irresistible. At first he was
welcomed as a deliverer and when he had expelled the
;

Helvetii and the Germans, it is doubtful whether he was

generally feared as a conqueror. It was only when the
presence of his legions was felt as a burden, and when
ambitious chieftains saw reason to fear that he would blast
their schemes, that he awakened partial opposition. The
Romanized Provincials supported him throughout and in ;

the great rebellion their assistance was indispensable. The

Gauls were not devoid of patriotism but it was choked by :

the tares of jealousy and when Vercingetorix was fighting

;

for the fatherland, it is probable that there were many who

198 CONCLUSION chap.

had much to fear from his success as from his faihire.

as
Those who courted Caesar's friendf;hip and espoused his cause
were distinguished by every mark of favour, and might reckon
with certainty upon his support. The Aedui adhered to him
for six years, and when they changed their mmds they found
that they had served his turn ^ the Remi saw from the first
:

that he was going to win, and, having made their choice, they
abided by it The Aquitanians cared nothing for
to the end.
the Gauls, and their isolated resistance was paralysed in a
single campaign. The Arvernian oligarchy executed the
chief who had restored the traditional glory of Bituitus and
Luernius, and while his son was still too young to play his
part, reduced the state which had been the rival of Rome's
strongest ally to such insignificance that for five years it

remained unnoticed in the record of the war. Indutiomarus

and Ambiorix and Acco struck too soon and when their ;

help was needed, one had vanished and two were slain.
The Celticans, with the exception of the Senones, the
('arnutes, and the maritime tribes, submitting, for the
most part, without an effort, looked on, with folded hands,
until, at the eleventh hour, Vercingetorix roused them
to a convulsive resistance ; and then the Belgae, who had
liitherto borne the brunt of the struggle, held aloof until
it was too late. Lucterius and Drappes, Correus and
Commius, all the chiefs who had been inspired by Vercinge-
torix, were doomed to fail for to reanimate enthusiasm would
;

liave needed one who had inherited a double portion of his

spirit and the Gauls, who were but men, had not the heart
;

to make that sacrifice which, on the required scale, even

he could not enforce. And there was one other factor in
Caesar's success which we must never forget. Even his
fortune and his genius might have failed against Vercin-
getorix if the hereditary enemies of Gaul had not crossed
the Rhine to his aid.
It has been said that it is impossible to conquer a people
who are determined to be free. Perhaps, in our modern
^ Ces chefs eduens,' says M. Jullian {Vercingetorix, p. 236),
*
qui n'em-
'

brassaient une cause que pour en regretter une autre, etaient toujours traitres
k la trahison merae.'
IX CONCLUSION 199

age ; every age, when the people dwell in

and doubtless iii

a country which nature has fortified, and when they are

brave, numerous, and of one mind. But Caesar succeeded, as
William the Conqueror succeeded, not merely because the
people with whom
he had to deal were disunited, but also
because he was prepared to go any lengths rather than fail.
The Gauls were willing to sacrifice myriads of lives, so they
might preserve their liberty ? Then he would slay a million,
aye and slay women and children, and ravage their lands,
and burn their houses over their heads, and lop off their
limbs, so he might at last subdue them And, though he !

was ruthless, he was also merciful.^ When he had beaten

down opposition, he held out his hand in friendship and ;

the Gauls took it, and bore him no grudge.

And when he had gone, what motive had they to rebel ?
Many of the states retained administrative independence and ;

none had exchanged independence for servitude. National

independence they had never had for they had never been
;

a united nation. As a nation, they could make no effort to

throw off the Roman yoke for there was none among them
;

who could command the confidence of the nation, or weld it

into a coherent whole. Many of the smaller peoples had
already been in subjection to powerful neighbours ; and it
was less humiliating to obey an alien master than one of
their own Rome was distant and her glory wrought
race. ;

upon the imagination. Rome was the resistless power which,

for centuries, had been bringing, one after another, the
nations of the earth within her empire. Jealousies were
hushed beneath her sway. Her yoke was easy and her rule ;

brought peace, security, and prosperity. If adventurers in

Gaul, as in India, regretted the good old days when they could
win thrones by their wits and their swords, the many gained
more than they had lost and so it happened that the few
;

spasmodic outbreaks which followed Caesar's departure were

foredoomed to failure, and that his conquest was effected once
for all.

^ In Caesare hate sunt : mitis clemensque natura. So wrote Cicero in 46 b. c.

[Fam., vi, 6, § 8).
 PART II

QUESTIONS OF GALLIC AND GALLO-ROMAN

HISTORY RELATING TO THE FOREGOING
NARRATIVE
SECTION L—FUNDAMENTAL
THE MSS. OF THE COMMENTARIES ON THE
GALLIC WAR
For my own satisfaction I have studied the correlation and the
value of the various MSS. but, as I am not here editing the Com-
;

mentaries, I should be wasting time and space if I were to discuss the

question in detail. For the number of passages in which, for his-
torical purposes, textual criticism is important, is small and all ;

those passages will be found fully discussed in subsequent articles.

It is sufficient to say that the MSS. which are alone worth consider-
ing are generally divided into two classes known as a and j3, derived
from a common original or archetype, not now extant, called X.
The 13 MSS.' were rated very low by C. Nipperdey in his Quaestiones
Caesarianae (pp. 37-46), which form the introduction to his famous
edition of the Commentaries but in framing his text he himself was
:

often compelled to have recourse to them and the best modern

;

critics, including H. J. Heller, H. Walther, R. Richter, Rudolf

Schneider, B. Kiibler, and H. Meusel, agree that Nipperdey greatly
under-estimated their value .^
In referring to the MSS. I generally use the symbols adopted by
Meusel, namely
^= codex Bongarsianus (or Amstelodamensis 81) of the 9th or 10th
century.

^ None of the extant /8 MSS. were written before the eleventh century;
but it is certain that Orosius, who wrote in the early part of the fifth century,
used a MS. which belonged to a time when a and 13 did not yet exist, but
which commonly agreed with the latter. See R. Schneider in Jahresb. d.
philol. Vereins zu Berlin, xi, 1885, p. 154; H. Meusel {ih., xx, 1894, p. 216) and
;

A. Klotz, Caesar studien, 1910, p. 215. When Orosius (vi, 7, § 2) said that the
authority whom he followed for the Gallic War was Suetonius, he did not,
I think, mean, as Prof. J. S. Reid supposes {Classical Philology, iii, 1908, p. 443),
that he had used an epitome of the Commentaries :surely he was misled by the
superscription of his copy of Caesar, which must have been identical with, or
—
analogous to that of A, mcipit liber Suetonii.
202 MSS. OF THE COMMENTARIES
B = Parisinua I (Paris, Bibliotlieque iiationale, 5703, 9th or 10th century]
31 = Vaticanus (Vatican, 3864, 10th century).
Q = Moysiacensis (Paris, Bibl. nat., 5056, 12th century).
= Ashburnhamianus (Bibl. Laurent. R. 33, 10th century).
iSf

a = Parisinus II or Thuaneus (Paris, Bibl. nat., 5764, 11th century).

/= Vindobonensis I (Bibl. Vindob. [Vienna], 95, 12th century).
A = Ursinianu8 (Vatican, 3324, 11th century).
= Riccardianus (Bibl. Riccard. [Florence], 541, 11th or 12th century).
?

Meusel traces the pedigree of these MSS. as follows ^ :

X
a (3

X ^ IT
p

A ~Q B~M~S a"7 ^^l

Ihave myself collated with Meusel's edition the best MS. in the
British —
Museum Add. MSS. 10,084 of the eleventh century (known —
as Lovaniensis and quoted as L), which agrees generally, though not
invariably with a, and belongs to the group cf).^ Meusel's great Lexicon
Caesarianum (1887-93) I have found absolutely invaluable.
Those who wish to study the problems presented by the MSS.
should read C. E. C. Schneider's edition, vol. i, 1840, pp. xxxvi-lii,
0. Nipperdey's edition, 1847, pp. 37-49, B. Kiibler's edition, 1893,
pp. iii-x, H. Meusel's edition, 1894, pp. v-ix, the prefaces to vols, i
and ii of Meusel's Lexicon, Philologus, xvii, 1861, pp. 492-509,
H. Walther, De Caesaris codicihus interpolatis, 1885, R. Richter,
Kritische Bemerkungen zu Caesars Comm. VII. de h. G., 1889, four
articles by H. Meusel in Jahreshericlite d. philol. Vereins zu Berlin,
xi, 1885, pp. 173-204, xii, 1886, pp. 262-93, xx, 1894, pp. 214-
398, and xxxvi, 1910, pp. 1-75, H. Schiller, Vher Entstehung und
EcJitheit d. Corpus Caes., 1899, and A. Klotz's article in Bheinisches
Museum, N.F., Ixiv, 1909, pp. 224-34.

WHEN DID CAESAR WRITE THE COMMENTARIES

ON THE GALLIC WAR, AND WHEN WERE
THEY PUBLISHED ?

The common view is that Caesar wrote the Commentaries on the

Gallic War after the conclusion of his Seventh Campaign, which took
place in 52 b. c.^ It is certain that the book w^as published not later
than 46 b. c. ; for it is noticed in Cicero's Brutus (75, § 262), which

1 A. Klotz, in a very interesting paper {Rhein. Mus., N.F., Ixiv, 1909, pp. 224-
34), has constructed a somewhat different pedigree, which I hope to notice
in an edition of the Bellum Gallicum, but which does not affect any of the
questions that are discussed in the following pages.
^ Meusel in his admirable edition of the Civil War
(p. viii) calls Lovaniensis
the twin-brother' of Ashburnhamianus, and so it is in the Civil War; but,
'

as he has acknowledged after reading my collation, not in the Gallic. The

collation will, at his suggestion, be printed in the Classical Quarterly.
'
See Jahresb. d. philol. Vereins zu Berlin, xvii, 1891, p 258.
 WHEN DID CAESAR WRITE HIS MEMOIRS? 203

appeared in that year. There isno direct evidence that it was either
written or published earlier : but it is most unlikely that Caesar
would have had time or inclination to write it during the intense
labour and distraction of the civil war and probably, as Mommsen
;

says,^ it was intended [at least in part] to justify '

before the . . .

public the formally unconstitutional enterprise of Caesar in con-

quering a great country and constantly increasing his army for that
object without instructions.' ^
Schneider argues^ that if Caesar had been writing his Commen-
taries during the successive years of the Gallic war, Cicero, whose
brother, Quintus, and whose intimate friend, Trebatius, were serving
on Caesar's staff, and who himself corresponded with Caesar, could
not have been kept in ignorance of the fact and that, if Cicero had
;

been aware of it, he would certainly have let it out. To which

I answer that Caesar knew how to keep his own counsel.
Schneider also argues that Caesar probably published his book in
the spring of 51 b. c, as a counter-stroke to the attack of M. Marcellus,
who had just brought forward a motion in the Senate that he should
be superseded before the expiration of his term of office. I agree
with Nipperdey that this argument is weak. Schneider concludes "*

that the book was written in the latter part of 52 and the early
part of 51.
M. P. Fabia ^ holds with Schneider that the Commentaries were
written after the Seventh Campaign, in the winter of 52-51 B. c. first, ;

because Caesar then had abundant leisure secondly, because his

;

enemies at Rome were then very active, and he would have been
anxious to counteract their machinations and thirdly, because, on
;

this hypothesis, we can understand why he omitted to describe the

Eighth Campaign. None of these reasons is conclusive. From the
very beginning of 51, Caesar was hard at work, campaigning against
the Bituriges, the Carnutes, the Bellovaci, and the Cadurci. In the
last two or three months of 52 he was comparatively at leisure ;

but the winter must have been one of the busiest that he spent
during the whole war. Secondly, if his enemies were active then,
they were also active before and afterwards. Thirdly, to account for
his not having described the Eighth Campaign, we need only suppose
that he was called away to more pressing duties when he had only
just finished his narrative of the Seventh.
Nipperdey ^ believes that the book was written after the war, in
the leisure of winter- quarters, and, so to speak, in one heat and in ;

support of this opinion, he quotes Hirtius, who writes ceteri enim

quam bene atque eynendate, nos eliam quam facile atque celeriter eos
perfecerit scimus? The quotation is, I think, apposite for if the ;

'
statement that the Com^mentaries were written easily and rapidly '

1 Hist, of Rome, v, 1894, p. 499 {Rom. Gesch., iii, 1889, p. 615).

* See, however,my essay on The Credibility of Caesar's Narrative
'
', p. 220.

' Wachler's Philomathie von Freunden d. Wissenschaft und Kunst, i, 1818,

pp. 180-2. < Caesar, vol. i, p. xxxi.

" De orationibus quae sunt in

Gomm. Gaesaris de B. G., 1889, pp. 19-20.
' Gaesar, pp. 3-5. ? B. G., viii, Praef. § 6.
204 WHEN DID CAESAR WRITE HIS MEMOIRS ?

is not absolutely inconsistent with the view that each book was
written during the comparative leisure of the winter follo^ving the
campaign which it described, the natural meaning is that the whole
work was the result of a continuous effort.
Nipperdeyi goes on to say that, even during the winter of 52-51 B. c,
which followed the rebellion of Vercingetorix, Caesar had his hands
full 2 but that in the year 50 he had nothing to disturb him.
; He
concludes that the first seven books were written during the year
that preceded the outbreak of the civil war and, believing that ;

Caesar would certainly have finished his narrative if he had had

time, he argues that he was interrupted by the outbreak of hostilities,
and published the book in the first stage of the war.
Schneider ^ maintains that no one who has read the remarks which
Caesar made about Pompey in his Commentaries on the Civil War can
believe that he would have had the magnanimity to praise him, as he
did, in his Seventh Commentary on the Gallic Tl'ar,* if he had written
the latter after his breach with Pompey. I confess that I have
a higher opinion of Caesar's magnanimity than Schneider.^ The
true reasons for deciding that the Commentaries on the Gallic TT'ar
were written before the outbreak of the civil war are, first that
Caesar would not have had time to write them during that war, and
secondly that it was to his interest to bring them out before the war
began.
Professor E. G. Sihler*' argues that the book could not have been
written before the winter of 52-51 B. c, because nearly all the speeches
are thrown into the form of Oratio Ohliqua,
—
a literary peculiarity '

utterly at variance with the literary habit of Greece and Rome and
with the culture and training of the day.' The one important excep-
tion, he continues —
the speech of Critognatus was a recent matter — '

when Caesar wrote '. This is a very weak argument. Caesar was no
more able to report Critognatus's speech accuiately verbatiin than those
of an}^ of the other persons the gist of whose utterances he professed
to give. Does Sihler suppose that he had a special correspondent,
who could write short-hand, in Alesia, or that he could have learned
more from prisoners than the drift of what Critognatus said ? He
1 Caesar, p. 4.
* I am
not convinced of the truth of this view. Tliroughout the autumn of
1842 and the first six months of 1843 Sir Charles Napier was occupied in Sind
with political and military work of the most engrossing kind. He was an old
man, and he often complained that his power of work was not what it had been.
Yet, over and above the anxious and heavy labour of negotiation and cam-
paigning, which, for some months, was carried on in a most trying climate,
over and above the task of writing frequent dispatches to Lord Ellenborough,
he found time to write up an exhaustive journal and long letters to his brother
William. Caesar was in the prime of life his power of work was enormous
: :

he had at all events some weeks of comparative leisure at Bibracte and ;

therefore I doubt whether Nipperdey is justified in saying that he could not

have found time to write his book in the winter of 52-51.
» Caesar, ii, 342. ' 6, § 1.
So also, 1 find, has Nipperdey. See pp. 3-4 of his Quaestiones.
* Caesar does
not abuse Pompey in the Civil War he simply narrates.
:

« Class, Rev., May, 1890, p. 199.

I
 —
WHEN DTD CAESAR WRITE HIS MEMOIRS ? 205

wrote this speech in Oratio Recta not because he remembered or had

ever known what Critognatus said, but because he thought lit, once
in a way, to study rhetorical effect. As to the other speeches, he
could easily have thrown them into the form of Oratio Recta, if
he had cared to do so.^
Long,2 on the other hand, believes that Caesar wrote his Com-
mentaries during his campaigns. There are inconsistencies in the
'

work,' he says, which are

'
not inconsistencies if the books were

written as the events happened. There is a brevity, sometimes an

incompleteness in the narrative, which I have observed in many
writers, who are writing of things before them, which are plain
enough to them then, but would not seem so plain to them if they
wrote afterwards. They would feel certain difficulties themselves
and try to remove them for others. It is impossible to understand
the attack on Gergovia, unless a man has seen the place or has
a perfect map of it.' Nor is this his only argument. Speaking of the
three chapters^ which Caesar devotes to the geography of Britain
and the manners and customs of its inhabitants, he says, It has '

been remarked that this digression would have been just as appro-
priate in the fourth book as here and so it would if Caesar had
;

been writing a history. But he wrote his Commentarii as the events

occurred, and according to the plan of his work it would have been
absurd to insert in the fourth book what he did not know when
he wrote it. If we compare the little that he could learn about
Britain before he sailed on his first expedition (iv. 20) with what he
tells us here, it is plain that in his fourth book he wrote down what
he knew at that time and in this, his fifth, he wrote down what he
;

learned in his second expedition.' ^ Alluding to Caesar's report of

the exaggerated statement of the Nervian old men regarding the
losses which their people had suffered in the battle on the Sambre,
Long says, In ii. 23 it is said that nearly all the fighting men of
'

the nation w^ere destroyed. Schneider thinks that what is said in the
second book could hardly have been written by a man who knew the
fact of this fresh rising of the Nervii and remembered it. But the
evidence that the same man v/rote both is the same as the evidence
that he wrote either, or any other part of these Commentarii. The
true conclusion is, that he wrote both at the time of the events. In
the second book he wrote that he had nearly destroyed the Nervii,
and he might suppose so.' ^ Finally, commenting on a passage which
occurs in Caesar's description of his decisive movement at Alesia
Eius adventu ex colore vestitus cognito turmisque equitum et co-
. . .

hortihus visis, quas se sequi iusserat, ut de locis superiorihus haec

declivia et devexa cernehantur, hostes proelium committunt Long —
observes that the phrase haec declivia, &c., 'is the true expression of

^ See my essay on 'The Credibility of Caesar's Narrative ', p. 213. There

are brief speeches in Oratio Recta in B. 0., iv, 25, § 3 ; v, 30, §§ 2-3 ; 44, § 3 ;

vi, 8, §§ 3-4 ; 35, §§ 8-9 ; vii, 20, §§ 8, 12, 38, §§ 1-3, 7-8 60, §§ 4, 6.
;

^ Caesar, p. xiii. ^ B. G., v. 12-4.

* Long's Caesar, pp. 229-30. ^ ib., p. 254,
206 WHEN DTD CAESAR WRITE HIS MEMOIRS ?

a man who writes with the facts fresh in his recollection : he speaks
of thesemovements along the descent to the level ground being seen
bv the enemy from the higher ground.' ^

Long's arguments are always interesting but I do not think that

;

these are conclusive. Caesar does not say himself in his Second Book
that the Nervian militia had been reduced from 60,000 to 500 he ;

only says that their representatives said so. It is true that he says
that the Nervian people had been well-nigh exterminated {prope ad
internecione^n gente ac nomine Nerviorum redacto) but perhaps this
;

was only a flourish.^ Again, whether he wrote his Commentaries year

by year, or all at once, he did not profess to write a regular history ;

and it does not seem to me incredible that, if he wrote them rapidly

in one heat, he should have described the manners and customs of
the Britons in his Fifth iiistead of in his Fourth Book. And, with
regard to the passage in the Seventh Book, on which Long lays so
nmch stress, it must be remembered that, even if Caesar wrote all
the Commentaries at once, the facts connected with the siege of
Alesia, which took place in 52 b. c, must have been fresh in his
memory.
De Belloguet ^ takes the same view as Long. He asserts that
various statements which Caesar makes about Britain * are incon-
sistent. I can only discover one instance of inconsistency and that ;

one is, I think, more apparent than real. In B. G. iv, 20, § 3, Caesar
says that none of the Gauls readily undertook a journey to Britain,
except merchants (neque enim temere praeter mercatores illo adit quis-
quam). In vi, 13, § 11, he says that those who desired to make
a special study of Druidical lore generally went to Britain for the
purpose (disciplina in Britannia reperta atque inde in Galliam translata
existitnatur, et nunc qui diliqentius eam rem cognoscere volunt plerum-
que illo discendi causa proficiscuntiir). But there is no real incon-
sistency between these statements. No one undertook the journey
readily {temere) or as an ordinary affair except the merchants.
Students did undertake the journey, but not temere they had :

a serious object in view. Besides, even if there were any incon-

sistency between the two statements, that would only prove that in
the former, Caesar, writing rapidly, as he did, made a slip.
Again, in support of the theory of intermittent authorship, de
Belloguet says that Caesar's narrative does not bear out the state-
ment which he makes, in his opening chapter, regarding the difference
^ Long's Caesar, p. 405.
- must be remembered that many of the Nervians who fought
It at the end
of 54 B. c. were doubtless too young to fight at the beginning of 57.
Ethnogenie gauloise, 1858-68, iii, 157-8.
^

B. G., ii, 4, § 7
' iii, 8, § 1
; iv, 20-1
; ; v, 12 ; vi, 13, § 11. Meusel {Jahresb
d. philol. Vereins zu Berlin, xxxvi, 1910, pp. 29-31) argues that the description
of Britain in B. G., v, 12-4, is an interpolation. It would be irrelevant to
examine his arguments here :but the fact that numerous hoards of the iron
'
currency-bars '{taleis ferrets) which are mentioned in 12, § 4, have been
discovered seems to me to suggest that the writer was Caesar for it is more
;

than unlikely that the}' were in circulation at the very late date to which Meusel
ascribes the alleged interpolation. See Rice Holmes, Anc. Britain, pp. 250-1.

I
 —'

WHEN DID CAESAR WRITE HIS MEMOIRS? 207

between the Celtae and the Belgae and therefore that we may
;

conchide that, when he made that statement, he was ill informed.

But it may be replied, first, that we have no right to expect that
Caesar should furnish instances in detail of the difference which he
states as a broad fact and secondly that, in regard to some at least
;

of the Belgic tribes —

the Nervii, the Morini, the Menapii, and the
—
Eburones his narrative does bear out that statement.
A. Kohler ^ has revived the theory of intermittent authorship, in
a modified form. Caesar, he says, in the Fourth Book, describes as
peculiar to the Suebi manners and customs which, in the Sixth, he
describes as common to the Germans generally. As, moreover, there
are [apparent] inconsistencies, regarding the Nervii, between the
Second Book and the Fifth and Sixth, Kohler concludes that the
Commentaries were written in two instalments, the former compris-
ing the first four books, and the latter the last three.
The argument relating to the Nervii has been already noticed.
There is certainly a resemblance between Caesar's descriptions of
the Suebi and of the Germans generally but the description in the
;

Sixth Book omits statements which are contained in the Fourth,

and adds others which are not contained in it.
F. Vogel 2 has lately attempted to reinforce the arguments
which have just been reviewed. Pointing to the passage in
which Caesar, after remarking on the want of cavalry among the
Nervii, says that to this day they pay no attention to that
'

arm, their whole strength being in infantry {neque enim ad '

hoc tempus ei rei student, sed quicquid possunt, pedestrihus valent

copiis^), he observes that in the last sentence of the same book
'
a thanksgiving service of 15 days was appointed to celebrate
his achievements, an honour which had not hitherto fallen to the
lot of any one' (dierum XV
supplicatio decreta est, quod ante id
—
tempus accidit nulli^) the words ante id tempus explain the pre-
ceding ad hoc tempus, and at first sight look like proof that Caesar
did not write the book until after the conclusion of the war. But
only at first sight ad hoc tempus must be regarded as a late inter-
:

polation ; for how could

Caesar, at the beginning of a description
which ends with the statement that the Nervii had been virtually
annihilated, have said that to this day '
their whole strength is
. . .

in infantry ? I have answered this argument by anticipation

'
but ;

Vogel has others in reserve. He notices that in B. G. iv, 6, § 4,

Caesar says that the Eburones are dependents of the Treveri
'

{sunt Treverorum clientes), whereas in vi, 43 their annihilation is

described. But is it ? Let Vogel read again, and he will see that it
was desired and anticipated, but not described and in fact two ;

years later (viii, 24, §4; 25, § 1) it was still incomplete. When
Vogel contrasts the two descriptions of the Ardennes (the latter of

1 Blatter f. d. hayer. Gym., s-xvii, 710-5, reviewed by R. Schneider in Jahresb.

d. philol.Vereins zu Berlin, xvii, 1891, pp. 257-9.
^ Neue Jahrb.
f. d. Mass. Altertum, v, 1900, p. 219.
^ B. G., ii,
17, § 4. 4 ih,, 35, § 4.
208 WHEN DTD CAESAR WRITE HIS MEMOIRS ?

which, by the way, Meusel ^ regarda as an interpolation) in v, 3, § 4

and vi, 29, § 4, I ask him to look at his map, which will show that
they are not inconsistent when he contrasts Caesar's almost
:
'

tender treatment of Sabinus in iii, 17-9 with

' the relentless '

manner in which he is pilloried '

in v, 33, § 1 and 52, § 6, I reply
that on the former occasion Sabinus succeeded, on the latter he
failed.^
doubt whether restless disputants will ever hold their peace
I ;

but no difficulty in assenting to the conclusion that has satis-

I find
fied the best critics. The arguments that have been advanced in
favour of intermittent composition are inconclusive for even real ;

inconsistencies, which are very few, can be fully explained by thought-

less handling of discordant materials, lapse of memory, or simple
carelessness. On the other hand, what has been called the strongest
argument in support of the prevalent theory is equally abortive. In
B. G., i, 28, § 5, Caesar says that the Aedui assigned lands to the Boi, and
afterwards admitted them to equal rights. If the First Commentarij
was written immediately after the first campaign, this passage
must have been added at some later time.^ But is there the slightest

1 Jahresh. d. philol. Vereins zu Berlin, xxxvi, 1910, pp. 31-2. Cf. A. Klotz,
Caesarstudien, 1910, pp. 53-4.
" H. Waltlier {tjber d. Echtheil und Ahjassung d. Schr. d. Corpus Cats., 1903)

has recently argued that the Commentaries were composed year by j'ear ;

but H. Schiller [Berl. phil. Woch., 1903, col. 1417) points out that many of the
discrepancies on which he relies are only apparent, and that others are con-
sistent with the theory that the book was written continuously. A. Klotz
also {Caesarstudien, 1910, pp. 18-25), noticing Chr. Ebert's Uber die Entstthuny
—
von Caesars B. G. the most recent statement of the case for the theory of
—
intermittent authorship easily confutes various objections, similar to tliose
which I have examined, that have been raised against the view which is
authorized by Hirtius. Cf. Blatter f. d. Gy mnasialschulwesen, xlvi, 1910, p. 369.
There is an inconsistency between vi, 2, § 3 and 31, §§ 1-2 which, if \ve
accepted Vogel's reasoning, would tend to show that Caesar wrote this or that
chapter whenever he had an hour to spare. In the former passage Caesar says
that all the Cisrhenane Germans ', who included the Segni and Condrusi, were in
'

arms against him ; in the latter he implies that the Segni and Condrusi proved
their innocence. But this inconsistency should warn us not to draw hasty
conclusions as to Caesar's method of composition it only proves that he did
:

not thoroughly revise his work.

^ Some critics indeed, notably H. Schiller {Uber EntsttMing iind Echtheit
d. Corpus Caes., 1899, pj). 28-32), pointing to the unusual repetition of the
relative pronoun in a demonstrative sense, contend that the passage was not
written by Caesar at all, but by an early editor. If so, remarks Schiller, the
strongest argument that has been advanced for continuous composition
collapses. Strongest indeed ? But Schiller cannot even slay the slain and ;

I am glad to see that his criticism has not imposed upon Meusel, "whose know-
ledge of Caesar's diction is supreme. Granted that the passage is inelegant,
do great writers never fall below their own standard ? In Matthew Arnold's
Obermann Once More {Poems, vol. ii, 1869, p. 243) the following deplorable
stanza occurs in the midst of many noble lines. I quote the stanza which
precedes it as well :

(VVellnigh two thousand years have brought

Their load and gone away,
Since last on earth there lived and wrought
A world like ours to-day.)
 WHEN DID CAESAR WRITE HIS MEMOIRS? 209

supposing that it was ? ^ In default of internal evidence,

difficulty in
then, let us accept the only original testimony that we have, the —
testimony of Caesar's most intimate friend. Hirtius's words show
that he had himself witnessed the rapidity with which Caesar
wrote his Commentaries therefore it is morally certain that
:

they were written at one time for the other theory, besides
;

involving a forced interpretation of the Latin, would oblige us to

assume that Hirtius regularly spent his winters with Caesar in
Cisalpine Gaul.
Mommsen argues that the work must have been not only written
'^

but published before the end of 51 B. c, because in B. G., vii, 6, § 1,

Caesar 'approves the exceptional laws of 702 (52 B.C.). This he
might and could not but do, so long as he sought to bring about
a peaceful accommodation with Pompeius, but not after the rupture,
when he reversed the condemnations that took place on the basis of
those laws injurious for him '. But why should not the publication
have taken place in 50 b. c. ? Even in 49 Caesar tried to bring about
a peaceful accommodation.^

THE AUTHORSHIP OF THE COMMENTARIES

Recent German criticism has been nmch exercised in striving to
determine how far Caesar can be credited with the authorship of
the books which for many centuries had been attributed to him
without question. Did he incorporate, more or less verbally, the
reports of his lieutenants in his narrative, or did he use them as
a modern historian uses such materials ? Was he the actual writer
of the Commentaries, or were the records wrought into shape under
his supervision by a secretary ? To discuss the literature of these
questions would be irrelevant to the purpose of this book, because
they hardly affect the value of the Comynentaries as evidence. Max
Strack in a recent article* recommends the perusal of the various
—
works which he enumerates, as a soporific.^ His own paper is lively
enough. He remarks that the five books which constitute the Corpus
—
Caesarianum the Bellum, Gallicum and Bellum civile, the Bellum
Alexandrinmii, the Bellum Africanum, and the Bellum Hispaniense —
have this in common, that they are generally limited to the narrative
of military events ^ he points out that Hirtius wrote an invective
:
"^

Like ours it looked in outward air !

But inward prize,

of that
Soul, that we take more count and care,
Ah there our future lies.
1

Some centuries hence perhaps this stanza will be deleted or bracketed by

restless editors ; but we know that it was written by the author of The Forsaken
Merman.
'
A. Klotz {Caesarstudien, p. 18) agrees with me.
2 Hist, of Rome, v, 1894, p. 499, n. 1 [ROrn. Gesch., iii, 1889, p. 616, note).
' B. C.,i, 9-10; 24, §§ 5-6, &c.
* Bonner Jahrhi'icher, cxviii, 1909, pp. 139-57.
^ Ih., p. 147. « Ih., p. 144. •>
Ib.i p. 151.
1093 P
210 THE AUTHORSHIP OF THE COMMENTARIES
against which Cicero ^ called a TrpoTrAatr/xa a model or
Cato, —
prototype —of
Caesar's Anticato, and argues that he was regarded
by Cicero as Caesar's literary mouthpiece he urges that Caesar, :

under the enormous pressure of his work, had need of a literary

secretary as much as of officers to execute the movements which he
planned ^ and he observes ^ that Hirtius was Caesar's most intimate
;

friend,* that he accompanied him in many of his campaigns, that

his ability is proved by Caesar's having designated him as consul for
the year 43 b. c, and yet that he was never appointed, though others
of inferior capacity were, to a military command. Strack concludes
that Hirtius w^as wanted for other work. He did for Caesar what
Lothar Bucher did for Bismarck in the editing of his Gedanken und
Erinnerungen. He collected and assorted dates, names, facts of
every kind, and submitted this material to Caesar, who winnowed
out the essential, and gave it form. Hirtius threw into semi-final
shape what Caesar dictated, and Caesar afterwards corrected his
work. In spirit and largely in actual words the result was Caesar's
own. If Hirtius was the editor, the unity of the Corpus Caesarianum
is explained.^ But Strack, who is troubled by a sense of humour,
can criticize his own conjectures and he concludes that we
;

shall never know the nature of the relations between Caesar and
his friend.^
Much that Strack has said is probable enough but does he not ;

go too far
? He admits, nay he insists, that the Eighth Book of the
Gallic War, of which Hirtius was the sole author, is far inferior to
the first seven. But, says he, this only shows that the mind which
created the Eighth Book was more concerned with petty details
than with what was really important. Does it not also show that
Hirtius was in no sense the author of the other seven, but only, at
the most, Caesar's collaborator ? We may well believe that he
'
devilled for Caesar
'
he may conceivably have compiled a pre-
:

liminary narrative of the war but to a man who has made the
;

Commentaries his intimate companion nothing can be more certain

than this :
—
apart from a few passages which were probably in-
terpolated, the first seven books are characterized by one style, and
bear the impress of one master mind. The material once collected
and arranged, it would have given Caesar less trouble to write the
whole narrative himself than to reconstruct, and prune, and vivify
the work of Hirtius. And did not Hirtius himself testify that the
actual writer was Caesar ?

1 Alt., xii, 41, § 4. Cf. Tyrrell and Purser, The Correspondence of Cicero,
vol. vi, p. ciii.
- Op. cif., p. 152. 3 lb., p. 148.
* Cicero, Att., vii, 4, § 2. * Op. cit., pp. 15G-7.
«
/6.,p. 157.
 211

THE CREDIBILITY OF CAESAR'S

NARRATIVE

For the history of the first seven years of Caesar's conquest of

Gaul our principal authority is Caesar himself. It is, of course, as
I have shown in the Preface, impossible to grasp the full meaning
of his narrative unless one has studied the works of the modern
scholars who have contributed so much to the task of solving the
various problems which the Commentaries present. It is true, more-
over, as I have also remarked, that later writers, such as Suetonius,
Plutarch, and Dion Cassius, make certain statements, true or false,
which are not to be found in Caesar. But Caesar is the authority ;

and if he is a bad one, all the commentators in the world can do

little to mend his work. It becomes necessary, then, to inquire how
far Caesar's narrative is worthy of credit. From this point of view,
the earliest extant criticism is that of Asinius Pollio, who died A. D. 4.
According to Suetonius/ Asinius Pollio thought that the Commen-
taries were written carelessly and with little regard for truth that;

Caesar accepted, without due inquiry, the accounts of his lieutenants ;

and that, either intentionally or from failure of memory, he was

inaccurate in describing what he had done himself. Of course it is
impossible to say with certainty how^ far Pollio was right ^ but one
;

who has had considerable experience in writing contemporary history

from original sources may perhaps form a tolerably just idea of the
significance of his opinion. When Sir William Napier published his
History of the Peninsular War, officers who had taken part in the
events which he described hastened to point out mistakes in his
narrative ; and he himself, half humorously, half despairingly re-
marked that, after all the care which he had taken, his book was full '

of lies '.^ Many of his brother-officers, however, bore testimony to

his general accuracy. Sir John Kaye, Mr. Kinglake, all in fact who
have written history which was open to the criticisms of the actors,
must have had a similar experience. Possibly the opinion of Pollio,
who took no part himself in the Gallic war, was founded upon
^ —
Divus lulius, ch. 56. Pollio Asinius parum diligenter parumque integra
veritate compositos putat, cum Caesar pleraque et quae per alios erant gesta
teinere crediderit et quae per se vel consulto vel etiam memoria lapsus perperam
ediderit.
^
F. Seek {De C. I. Caesaris comm. fide, partii, 1864, pp. 8-10) maintains that
Pollio' s criticismson Livy and Sallust (for which see C. L. Roth's edition of
Suetonius, 1893, pp. 261-2) were fair, and argues that we may therefore accept
his criticism of Caesar. But the two former judgements refer only to style ;

and we do not even know whether the criticism of Caesar was directed against
the Gallic War.
^
Life of Gen. Sir William Napier, edited by H. A. Bruce, 1864, i, 448.
P ?.
212 THE CREDIBILITY OF CAESAR'S NARRATIVE
criticisms like these. Caesar inevitably made mistakes and Pollio ;

may have conversed with eye-witnesses who pointed out these

mistakes, who were perhaps aggrieved by them, and who, exaggerat-
ing their significance, as men devoid of the sense of historical pro-
portion will always do, shrugged their shoulders and exclaimed,
'
Such is history.' It may even be true, as Mezger ^ thinks, that
Pollio's criticism was only directed against the Commentaries on the
Civil War, the history of which he himself wrote. Anyhow, as
Mezger goes on to observe,^ the only other contemporary criticisms
of the Gallic War that we possess —
those of Cicero and Hirtius are —
favourable and if Caesar had been a dishonest or grossly inaccurate
;

writer, it would be remarkable that there shoiUd be absolutely no

evidence that any of his statements was ever expressly contradicted.
His accuracy, as a narrator of details, has been confirmed, on various
points, by modern investigations. When he is writing military
history, pure and simple, for example chapters 39 to 84 of the First
Book of the Civil War, his general trustworthiness is as self-evident
as his skill. No one who reads his book can deny that he was gifted,
in an extraordinary degree, with the faculty of observation. Most
of the operations which he describes were performed under his own
eye he had opportunities for observing what took place in a battle
:

or a siege which a modern general, whose operations extend over

a vast area, cannot have and it is important to notice that he rarely
;

indulges in that sort of detailed description which gives rise to most

of the mistakes that occur in modern military history. This is
'
a point which I could not make perfectly clear to a general reader '

unless he would bear with me while I explained to him the labour

which I have myself undergone in writing an account of a modern
battle, and the process by which I have been enabled to correct
mistakes which had crept into my original draft. But all who have
tried their hands at writing military history from original sources will
understand what I mean. As a rule, Caesar gives us only the outline
of a battle, —
he tells us just so much as may enable us to understand
the moves, to differentiate his account from his accounts of other
battles, and no more.
It may perhaps be impertinent to remark that lapse of time does
not breed mistakes in a historical narrative. But nothing would be
more natural than for the general reader, especially if he had been
taught to believe that history is a Mississippi of falsehood ', to say
'

in his haste, How can we know anything about the details of battles
that happened 2000 years ago ? Well, if Caesar was an honest writer,
when we are reading his Commentaries, we are in much the same
position as we should be if, for our knowledge of the Peninsular war,
we depended mainly upon a volume of Memoirs by the Duke of
Wellington. Caesar sent dispatches to the Senate, and it may be
assumed that he kept copies of them his generals sent dispatches
:

to him and if he did not write his commentary on each campaign

;

* ijher d. Abfassungszeit von Casars Co)n)n. i'lber d. gall. Krity, 1874-5, p. 5.

- 16. See Cicero, Brutus, 15, § 262, and B. G., viii, Praef. § 5.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 213

as he fought it, he certainly finished the whole work within two years
after the close of the war.^ His account, therefore, was strictly a con-
temporary account by the eye-wdtness who had the best eyes and
the most favourable point of view and it remains as accurate
;

to-day as when it was first penned.

II
But one may accept his narrative as generally trustworthy without
claiming for him immunity from error. The speeches which he puts
into the mouths of Vercingetorix and others may have been composed
from information supplied by Vercingetorix himself or by prisoners ;

but when one remembers the licence which was assumed by ancient
historians in this respect, one naturally asks whether all his speeches
are even substantially true. P. Fabia ^ argues that, as Caesar
undoubtedly drew upon his imagination in writing some of them,
for instance the one which he put into the mouth of Critognatus,^
we are justified in believing that he did not always tell the simple
truth even when he knew it. But there is no proof that he drew
upon his imagination for the outline even in writing the speech of
Critognatus. The words in which he clothed it being, contrary to
his custom, cast in the form of Oratio Recta, were of course his own.
But the substance is another matter. Fabia says that Caesar could
not have knowai anything of what Critognatus said, except his pro-
posal to kill the non-combatants for food.^ But, if he was informed
of this, why should he not also have been informed of the drift of the
orator's arguments ? As Fabia himself admits, it was not Caesar's
manner to set down anything merely for literary effect ;
^ and his
critic forgetsthat he could have questioned Vercingetorix, who had
been in Alesia with Critognatus, and who w^ould doubtless have been
glad, after his surrender, to talk over the events of the war. Still
I can conceive that, on this occasion, Caesar did give the rein to his
imagination. He was conscious of his own oratorical powers and ;

very likely he was tempted to show, for once, that he could do some-
thing in the style of Thucydides, w^hich even Cicero might read with
admiration.
Again, Fabia contends that Caesar, in writing his speeches, must
have trusted almost entirely to his memory, because, if he had
caused any memoranda to be made, they were certainly very brief.*^
This is a pure assumption and are not the speeches themselves verv
;

brief ?

At the same time, Fabia remarks with justice that, in com-

posing his speeches, Caesar was far more careful of truth than any
other ancient historian, as may be inferred from the fact that most
of the speeches were written in Oratio Ohliqua, which proves that he
only professed to give the drift of what was said.'^
'
See pp. 202-9.
- De orationihus quae sunt in Comm. Caesaris de B. G., 1889, pp. 9] -2.
* B. 0., vii, 77. * De orationihus, &c., pp. 2H-4.
6 Ih., pp. 13, 39. «/?>., p. 21. ' lb., p. 92.
214 THE CREDIBILITY OF CAESAR'S NARRATIVE

III
So much for the speeches. More serious charges have been brought
against the general tone of his narrative. It has been alleged that he
wrote with a political purpose and that he was consequently led to
;

omit facts which would have told against himself, to invent plausible
motives for his more questionable actions, and to bring false charges
against his enemies. The popular judgement at Rome, argues the
Due d'Aumale, whose criticism is singularly fair and on the whole
highly favourable, was sure to be indulgent to and not over-critical
of an account of victory over Rome's ancient enemies while Caesar's
;

lieutenants, even those who opposed him in the civil war, etaient
'

interesses a ne pas diminuer la valeur d'un livre qui etait pour eux
aussi un monument de gloire.' ^ The critic, I may remark in passing,
fails to notice that, if the popular judgement was sure to be so favour-
able, that was the very reason why Caesar could afford to tell the
truth. I am quite ready to believe that one of his motives was to
conciliate public opinion in view of the civil war. But I believe that
this object was to be attained by a truthful just as well as by a dis-
torted narrative.'^
In the first edition of this book I adopted a method of testing the
credibility of Caesar's memoirs which, by some critics in this country,
though not on the Continent, was condemned as over-scrupulous.
I took every charge, at all colourable, of inaccuracy, of misrepresenta-
tion, or of mendacity that had been brought against him, and had
been supported by argument, and examined it as impartially as
I could. I thought that it would be useless to select a few charges
and dismiss the rest, on the ground that the names of their authors
carried no weight, because, though hardly one of the assailants of
Caesar's veracity has a European or even a national reputation, their
united assaults have made an impression. Although the utility of
this minute investigation has been spontaneously and generously
acknowledged by a French historian,^ I intend now to take account
only of such specimens of destructive criticism as may seem either
well grounded or not safely to be ignored.
Readers who take an interest in the question will of course not rest
satisfied with merely examining charges of untrustworthiness. They
will read Caesar's Commentaries and his letters for themselves, read
the letters that Cicero wrote to and about him, and form at first
hand an opinion of the character of the man. They will of course
discriminate between the narrative which he composed from per-
sonal knowledge and those chapters which were based upon the
reports of his marshals. But their judgement will be liable to error
unless they have some knowledge of the operations of war. Not
even a trained historian, unless he is also a soldier, or unless his
training has included diligent study of military history, fortified by

^ Rev. des Deux Mondes, 2^ per., xv, 1858, p. 120.

- See p. 224, n. 1.
^ C. Jullian, Hist, de la Oaule, iii, 1909, pp. loO-l.

I
 THE CREDIBILITY OF CAESAR'S NARRATIVE 215

intimate converse with military men, can appreciate, unaided, the

military sections of Caesar's narrative.
The charges which have been brought against Caesar's good faith
may be grouped in two classes, according to the motives which his
accusers impute to him. These motives are, first, the desire to put
the best construction upon unconstitutional or unrighteous acts, and
secondly, the desire to magnify his own exploits, to take to himself
the credit of the exploits of his lieutenants, and to conceal everything
that might have injuriously affected his reputation as a general.

IV
The assailants of Caesar's credibility, for the most part, attach
a high value to the authority of Plutarch, Appian, Florus, Orosius,
Eutropius, and above all Dion Cassius, whenever they differ from
or supplement Caesar. Their theory is that these later writers used
other sources of information besides the Commentaries. About the
first five there is very little to be said. Anybody who prefers the
authority of Plutarch to that of Caesar must be so credulous or so
wrong-headed that it would be useless to argue with him. Plutarch
is a delightful writer, who is read not because he was a critical his-

torian, but because he is readable. But, like some other readable

biographers, he was a romancer and, as F. Eyssenhardt,i one of
;

Caesar's sternest critics, admits, his Lives abound with monstrous

blunders.
Florus dismissed the whole Gallic war in a single rhetorical chapter
filling about two pages and the value of that scrap may be gauged
;

from the fact that he confounded Gergovia with Alesia.*^

The epitome of Appian's narrative of the Gallic war^ is shorter
even than that of Florus and although it is not so pretentious, it
;

is a wretched blundering piece of work. For example, Appian says

that the Nervii were descended from the Cimbri and Teutoni that;

Caesar defeated the Allobroges and that the Sugambri with 500
;

routed Caesar's 5,000 Gallic troopers.

Orosius was a Spanish presbyter of the fifth century, who compiled
a history of the world from the creation, in defence of Christianity.
That part of it which deals with the Gallic war is contained in the
seventh and the four following chapters of the Sixth Book. In this
small space Orosius makes several gross mistakes. Immediately
after reproducing from the Commentaries the list of the Belgic con-
tingents which Caesar encountered in 57 b. c, he says that they
emerged from ambush in a wood and threw Caesar's army into con-
fusion, but were subsequently defeated and all but annihilated.'*
Here he is evidently confounding the entire Belgic host with the
^ NeueJahrb.f. Philologie, &c., Ixxxv, 1862, pp. 760-1.
'-
i,45 (iii, 10).
^ Geltica, i, 2-5.
* His repente silva erumpentibus, exercitus Caesaris perturbatus atque in
fugam actus, plurimis suoruin
amissis, tandem hortatu ducis restitit, victoresquo
aggressus usque ad internecionem paene delevit. vi, 7, § 16.
216 THE CREDTBTLTTY OF CAESAR'S NARRATIVE
Nervii.i Speaking of the leaders of the Veneti, he says, with a
Froudian embellishment, that Caesar tortured them to death.
Caesar's object, he says, in crossing the Rhine was to relieve the
—
Sugambri the very people whom he marched to attack from —
blockade.^ Lastly, he confounds Cenabum with Avaricum and
Gergovia with Alesia."* But Orosius is not an irritating romancer
like Dion Cassius. Excepting that one little flourish about the
Veneti, he does not try to dress up and embellish Caesar's narrative.
In general, he follows the Commentaries closely enough and there ;

is only one passage in which I can discover any trace of his having
consulted an authority who differed from or attempted to correct
Caesar. This passage, which deals with the numbers of the Helvetii,
will be referred to later on.
Eutropius's account of the Grallic war is comprised in one para-
graph of his epitome of Roman history (vi, 17 [14]). As a specimen .

of his accuracy, I will quote one sentence,—' Caesar attacked the

Germans on the further side of the Rhine, and defeated them in
a series of bloody battles {Germanos trans Rhenum aggressus im-
'

manissimis proeliis vicit).

Dion Cassius was rated at his true value by George Long.^ He
was nearly as great a romancer, in his way, as Plutarch but his ;

way was dull. How far he used other authorities besides Caesar
for his narrative of the Gallic war, we shall never know ^ but at all ;

events Caesar was his chief authority. In following him he gave as

free play to his imagination as Mr. Froude, but with this difference :

Mr. Fronde's was the imagination of an artist, who had a sense for
the fitness of things, Dion's of a tasteless newspaper reporter Dion ;

wrote bad Greek, and Mr. Froude wrote good English. As I shall
often have occasion in this book to examine Dion's statements, I shall
give a few specimens of his work. First, instead of summarizing or
reproducing the vigorous little speech with which Caesar^ quieted the
panic that seized his army before their campaign against Ario\'istus,
he puts into Caesar's mouth a sermon, which fills eleven chapters of
his book,s which would have had no effect in reassuring his hearers,
and which Long ^ aptly characterized as a rambling and unmeaning
'

piece of fustian, worthy of Dion's age '. Secondly, describing the

attack which Viridovix made upon the camp of Sabinus, he says
that the Gauls took firew^ood and other wood, some on their shoulders
'

and some dragging it after them, in order to burn the Romans '.^^
Caesar says simply that the Gauls collected brushwood and faggots
'

to fill up the Roman trenches .^^ Thirdly, Dion, by way of embellishing

'

^ See B. G.,u, 19-28.

^ Cunctis principibus per tormenta interfectis. vi, 8, § 17.
3 vi, 9, § 1. vi, 11, §§ 3-4, 7.
* ^ Caesar, p. xi. n. 5.
* See M. Jullian's Hist, de la Gaide, iii, 151, note.
' B. 0.,
i, 40. 8 xxxviii, 36-46. " Caesar,
p. 85.
^^ (ppvyava nal ^v\a ra fxtv apajXfvoi, r^ b\ ecpeXfioixfvot, ujs kqi KaTairpTjcroi'Tfi
aiiTovi . . TTpo(T((3a\ov (xxxix, 45, § 4).
. See Long's Caesar, p. 165.
^^ Sarmentis
virgultisque collectis, quibus fossas Roiuanorum compleant.
B. G., iii, 18, § 8.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 217

his account of Caesar's campaign against the Morini, says that

there were great mountains ^ in their country, that is, great
' '
—
mountains in the neighbourhood of Calais. Fourthly, whereas
Caesar ^ tells us that Sabinus and Cotta quitted Atuatuca at dawn,
Dion^ changes dawn into evening. Fifthly, he says that when
Caesar heard the news of the fate of Sabinus and Cotta, he was on
his way to Italy .^ But Caesar's narrative ^ shows that he was at
Amiens (Samarobriva) and Dion's blunder was evidently the out-
;

come of a vague reminiscence of the words which Caesar puts into

the mouth of Sabinus :
—
My belief is that Caesar has started for
'

Italy (Caesarem arhitrari profectum in Italiam).^ Sixthly, Dion

'

says that Caesar crossed the Allier on rafts whereas we know ;

"^

that he crossed by a bridge, which he had repaired.^ These instances

— and the list might be considerably extended are sufficient to —
justify the opinion which I have expressed of Dion's work. A dull
accurate historian and an interesting inaccurate historian both have
their uses but tedious fiction is intolerable.^
;

To sum up, it is clear that Plutarch, Appian, Florus, Orosius,

Eutropius, and DionCassius folio wed Caesar, more or less inaccurately,
as their authority. It is possible that some of them may have re-
ferred, on certain points which will be noted in the course of this
essay, to some other authority or authorities as well. But if they
did, there is not the slightest reason for preferring those unknown
authorities to Caesar. To appeal to Orosius or Plutarch against
Caesar is much the same thing as it would be to appeal against
Wellington's despatches to John Richard Green. If Caesar is to be
arraigned for mendacity, the appeal must be to internal evidence
alone.^^

V
The principal accusations which fall within what
have called the I
first groupie are
(1) that Caesar misrepresented the motives which
prompted the Helvetii to emigrate and those which impelled him to
attack Ariovistus (2) that he invented the story of the attempt of
;

the Helvetii to cross the Rhone (3) that, in relating the intention
;

of the Helvetii to settle in the country of the Santoni, he minimized

the distance between the frontier of that tribe and the frontier of
the Tolosates, in order to make the danger that threatened the
Province appear greater than it really was and (4) that the whole
;
'

drift of his own report is that he was continually forced to enter

f^^^ y^P ^ ii-(ii(jo.p nal Is avra to, oprj ttju vKtjv refivoov -npoxajpriGai'
^ eiTfx^'ipV^^

aircnrwv Se, 5ta re to fxeyidos avrujv . diraveaTi] (xxxix, 44, § 2).

. ,

2 B. G., V, 31, § 6. '

xl, 6. ^ xl,
9, § 1.
' B. O., V, 46-7. Cf. Long's Caesar, p. 261.
« B. G., V, 29, § 2. 7 xl, 35, § 3. ' B. G., vii,
35, §§ 5-6.
^ See, in confirmation of my estimate
of Dion, Heller in Pkilologus, xxii,
1865, pp. 108-11; Neue Jakrh. f. Philologie, &c., cliii, 1896, pp. 269-71;
Rev. celt., xxii, 1901, p. 87; and p. 237, n. 3, infra.
^^ I am not speaking
here of the Commentaries on the Civil War, from which
we can appeal, on certain points, to Cicero's Letters.
''
See p. 215,
218 THE CREDIBILITY OF CAESAR'S NARRATIVE
upon new enterprises and campaigns against his own wish and
expectation '.

1. The charge has been formulated in various terms again

first
and again but only one of the advocates has gained a wide hearing,
;

— the well-known historian, Guglielmo Ferrero, who has conducted

his case with more ability than his predecessors and on entirely new
lines. His method involved a drastic reconstruction, which I criti-
cized in the Classical Quarterly of July, 1909,^ of Caesar's First
Commentary. I need only repeat here that his whole case depends
upon four assumptions —
first, that the main object of the Helvetii
:

in invading Gaul was to drive out Ariovistus secondly, that while ;

Caesar's countrymen would have approved his attacking the Helvetii

in order to prevent them from founding a Gallic empire ', they '

would have condemned him for preventing them from seizing a

fertile tract in Gaul and from dispossessing its occupants thirdly, ;

that when he charged them with having raided the lands of the
Aedui, the Ambarri, and the Allobroges, he lied and lastly, that ;

he made an alliance with Ariovistus in order to purchase his aid

' '

against the Helvetii.^ In the Classical Quarterly of January, 1910,

Signor Ferrero replied to my criticism,^ He thinks that I failed to
seize his main point, which, he says, is represented by this question :

Why did Caesar conclude an alliance with Ariovistus in 59 B.C. and

break it in the following year ? * Any one who may have read my
article with care will have seen that I recognized that this point
was capital.
In the first edition of this book I conjectured that the Senate
'^

conferred upon Ariovistus the title of Friend of the Roman People

'
with the object of securing his neutrality in view of the threatened
Helvetian invasion ', and in the article to which Signor Ferrero
replies I said that Caesar foresaw that when he went to Gaul he
'

would have to deal both with the Helvetii and with Ariovistus ;

and to dispose of two formidable hosts separately would be quite

as much as he could manage '.^ L'explication,' Signor Ferrero
'

remarks, parait d'abord satisfaisante

'
mais la lecture attentive
;

des Commentaires nous prouve que ni les Helvetes ni Arioviste ne

desiraient que de vivre en paix avec Cesar (cf. B. G. i, 13 i, 44), ;

et bien loin d'avoir songe a une alliance contre Cesar, regardaient une
guerre contre lui comme une chose presqu'impossible.'
"^

The passages to which Signor Ferrero refers are speeches, which

Caesar puts into the mouths of Divico, the Helvetian leader, and
Ariovistus respectively, and in which the former declared that if '

the Roman People would make peace with the Helvetii, they would
go wherever Caesar fixed their abode ', but warned him not to

Pp. 203-15.
^

Cf. F. Frohlich, Die Glauhwiirdigkeit Caesars in seinem Bericht uher d.

^

Feldzug gegen d. Helvetier, 1903, pp. 4-7.

^ What follows is reprinted, with certain omissions and additions, from my

rejoinder {Class. Quarterly, October, 1910).

^ C. Q., 1910, p. 28. ' 5 P. 21. ^ C. Q., 1909. p. 213.
' C. Q., 1910, p. 28.

i
 THE CREDIBILITY OF CAESAR'S NARRATIVE 219

'
suffer the place where they stood to derive its name from a Roman
reverse and from the annihilation of a Roman army ', and the latter,
after telling Caesar that unless he took his departure and withdrew
'

his army from the neighbourhood, he should treat him not as a friend
but as an enemy ', and hinting that he could purchase the gratitude
'

and friendship of many of the nobles and leading men of Rome

'
' '

'
by killing him ', ended by genially assuring him that if he with- '

drew and left him in undisturbed possession of Gaul, he would re-

ward him handsomely '. Signor Ferrero apparently now thinks that
Caesar ought to have accepted these assurances, although in the
article which I criticized he maintained that Caesar ought to have
adopted an anti-German policy. At all events the passages which
' '

he quotes are hardly sufficient to refute my contention, that Caesar

bestowed a title upon Ariovistus with the object of securing his
'

neutrality in view of the threatened Helvetian invasion '. But, says

Signor Ferrero, Caesar, by making an alliance with Ariovistus, risked
losing the support of the philo -Roman party in Gaul. Nothing, he
adds, in the First Book of the Cotmnentaries is more striking than
the weakness of that party for the words of Diviciacus when he
;

begged Caesar not to punish Dumnorix, as otherwise the result '

would be that the feeling of the whole country would turn against
him (qua ex re futurum uti totius Galliae animi a se auerterentur)}
'

prove that le parti nationaliste representait la presque totalite de

'

la Gaule '. Why, asks Signor Ferrero, were the philo -Romans re-
duced to impotence ? L'alliance avec Arioviste est la seule ex-
'

plication possible mais c'est une explication suffisante. En s'alliant

;

avec I'ennemi national de la Gaule, Rome s'alienait toutes les sym-

pathies des Gaulois. La question de savoir pourquoi Cesar a fait
cette alliance est done plus complexe et difficile que M. Holmes ne
^
le suppose.'
I confess that Signor Ferrero seems to me to build too much upon
the rhetorical phrase, totius Galliae. Still I am willing to admit, for
the sake of argument, that the philo -Roman party was temporarily
weakened by the fact that a title had been bestowed upon Ario-
vistus.^ But what then ? Caesar knew that he must run risks in
order to solve a difficult political problem and he solved it trium- ;

phantly. Signor Ferrero thinks that he ought to have rejected the

overtures of Ariovistus, and resolutely embraced the cause of the
—
pro-Helvetian party, in other words, of the Nationalists. But the
inevitable result would have been that he could only have got rid of
Ariovistus, who, remember, according to Signor Ferrero, was only
anxious to live on friendly terms with him the Helvetii would have :

remained, and he would have been obliged to tolerate their per-

manent settlement in Gaul. Consider what that would have in-
volved. Dumnorix was in league with the Helvetii, and intended

1 B. G., i, 20, § 4. ^ C. Q., 1910, p. 29.

^As I remarked in my former article (C Q., 1909, p. 211), Caesar did exactly
—
what the philo-Roman party wanted, first expelled the Helvetii, and then
Ariovistus.
220 THE CREDIBILITY OF CAESAR'S NARRATIVE
to use them as instruments of his ambition.^ Caesar would have
been obliged either to look on helplessly at the development of their
schemes, or to attack the Helvetii at some future and less propitious
time. I cannot help thinking that he understood Gallo-Roman
politics better than his critic. The radical difference between Signor
Ferrero's view and mine is this he believes that both the Helvetii
:

and Ariovistus only desired to live in peace with Caesar ', but that
'

Caesar ought nevertheless to have joined with the Helvetii and the
Nationalists in expelling iVriovistus from Gaul I believe, as Caesar's
;

narrative obliges me to do, that Caesar regarded the presence both

of the Helvetii and of Ariovistus as dangerous to Roman interests.
Is Signor Ferrero really prepared to maintain that Caesar would
have served Rome and Gaul better if in 58 b. c. he had only expelled
Ariovistus, and had permitted the Helvetii to remain ? If he is,
I can only admire his hardihood. If he is not, how can he help
admitting that Caesar's policy was right ?
Signor Ferrero tells me that I have forgotten that Caesar was not
authorized by the laws of his country to defend Roman prestige in
Gaul according to his personal views that the right of declaring
;

war belonged not to him, but to the Senate that Ariovistus was
;

a Friend of the Roman People and that consequently the war

;

which Caesar waged against him was illegal?" I have never denied
any of these propositions but I do deny their relevancy. Granted
;

that Caesar's action was illegal it does not follow that his sole
;

motive for attacking Ariovistus was to recover the sympathies of

the Gauls. If he was prepared to break the law in order to please
the Gauls, why should it be deemed incredible that he was prepared
to break it in order to do what Signor Ferrero himself insists ought
to have been done, namely to rid Gaul and Rome of a dangerous
enemy ? Were not illegal and unconstitutional acts frequent in those
revolutionary times ? Did not Pompey show his contempt for
legality in 52 b. c. ? ^ Does Signor Ferrero maintain that Caesar
consulted the Senate before he undertook the campaigns of 57 b. c.
and the six years that followed ? Did he not defend Roman pres- '

tige in Gaul ', and make war from first to last according to his own '

personal views just as the Marquess Wellesley did in India ?

'

But it is at least questionable whether Caesar's action was even

illegal. The decree of the Senate to which he appealed was surely
a sufficient justification. If he could prove that he had been called
upon, in accordance with that decree, to protect the Aedui '
the '
"*

charge of illegality w^ould collapse. Ariovistus himself, by his head-

strong arrogance, flung away the advantage which his title had
given him, and played into the hands of his wary antagonist.
There is one fact which alone pulverizes Signor Ferrero's recon-
struction. If Caesar conferred titles upon Ariovistus, not, as I main-
tain, in order to secure his neutrality, but in order to obtain his
active co-operation against the Helvetii, why did he never avail

'
B. G., i, 18, § 9. - C. Q., 1910, p. 29.
-'
Dion Cassius, xl, 56. ^ B. G., i, 35, § 4.
 THE CREDIBILITY OF CAESAR'8 NARRATIVE 221

himself of it It was as much as he could do, unaided, to defeat

?

the Helvetii why, then, did he not call upon Ariovistus to join
:

him ? Obviously because he had never contemplated a proceeding

which would have paralysed his policy. If he had accepted the
assistance of Ariovistus against the Helvetii, it would have been
impossible for him, however cynical he may have been, to turn
round afterwards and expel Ariovistus from Gaul. And that was
what he intended to do.^
Signor Ferrero attributes to Caesar a lack of foresight and a sim-
plicity — —
nay, innocence which are both ludicrous and incredible.
He says that after the Helvetian war Caesar saw, for the first time,
that if he rested satisfied with his victory, he would only have played
the game of Ariovistus, forfeited the sympathies of the philo -Roman
])arty, and destroyed Roman prestige and that, as he had not yet
;

thought of conquering Gaul, the result would have been a veritable

disaster, to avoid which he was forced to turn against his own ally.^
How does Signor Ferrero know that Caesar had not yet thought of
conquering Gaul ? Is it conceivable that he would not have fore-
seen that sooner or later he would have to deal with Ariovistus ?
Signor Ferrero, as is well known, goes so far as to reject Caesar's
statement that the Helvetii intended to settle in the country of the
Santoni. Confessing that he does not know how or when they pro-
posed to drive Ariovistus into Germany, he goes on to say, Ce qui '

me semble le plus probable, c'est que le parti de Dumnorix cherchait

a transporter cette population belliqueuse dans des territoires situes
dans le nord-est de la Gaule, de I'etablir dans ces territoires, pour
pouvoir recruter un jour dans cette population une armee contre
Arioviste.' ^ But is not this an afterthought ? In the article which
I criticized Signor Ferrero did not say that the Helvetii were march-
ing towards the north-east of Gaul with the intention of settling
there and at some future time attacking Ariovistus he said, the
:
'

Helvetii had already [after their defeat] turned eastwards. The

Helvetii therefore were marching towards the Rhine and he added
'
;

that they marched towards the country in Avhich the army of

'

Ariovistus was encamped ? '

The army of Ariovistus was not
'^

encamped in the north-east of Gaul, but in the plain of Alsace. To

march from Geneva towards North -Eastern Gaul by way of the
valley of the Saone and Mont Beuvray would have been a sufficiently

^ According to Appian {Geliica, 16), Ariovistus, after he had finally defeated

the Aedui, promised not to molest them further, and in return received the
title which he sought. This may be true ; but Appian is a bad authority
(of. C. Jullian, Hist, de la Gaule, iii, 158, n. 3). M. Jullian, who agrees with me
in rejecting Signor Ferrero's reconstruction, nevertheless supposes {ib., p. 165)
that it was intended, in case of need, to utilize the forces of Ariovistus against
the Helvetii. For the reasons which I have given this seems to me more than
unlikely, unless indeed the Senate alone was responsible for the grant of the
title. Mommsen apparently thinks tliat it was for lie attributes the grant
;

of the title to *
ignorance and laziness in statesmen [Hisf. of Rome, v, 1894,
'

p. 36 [Rom. Gesch., iii, 1889, p. 247 n. **]). But Caesar assumed joint
responsibihty. 2 G.
Q., 1910, p. 30. =>
lb., p. 31.
* The Greatness and Decline
of Rome, ii, 345-6.
222 THE CREDIBILITY OF CAESAR'S NARRATIVE
inexplicable proceeding but to inarch by the same route, with the
;

intention of turning eastwards '

towards the Rhine and to-
' '
'
'

'
wards the country in which the army of Ariovistus was encamped
would have been more Gilbertian still. Accordingly I made an
objection, which Signor Ferrero quotes and endeavours to answer.
'
Ariovistus,' I said, was in the plain of Alsace. Let Signor Ferrero
'

look at his map, and ask himself whether a movement more insane
than a trek from Geneva to the neighbourhood of Lyons, across the
Saone, up the valley of the Saone to the neighbourhood of Milcon,
then westward towards Toulon-sur-Arroux, then northward to the
plateau of Langres, and then back again eastward a hundred miles
or more " towards the Rhine " —
a trek in bullock-carts full of non-
combatants, against a powerful host which there was no motive for
—
attacking was ever planned outside Bedlam.' ^ Appealing from his
former to his present self, Signor Ferrero says, II n'-est point neces- '

saire de supposer que 1' emigration des Helvetes allait, comme une
arniee en formation complete, attaquer immediatement Arioviste :

elle allait occuper les territoires qu'on lui donnait, comme les Ger-
mains qu' Arioviste faisait venir d'outre-Rhin, pour s'y tenir prete
a fournir une armee. II n'est done non plus necessaire de supposer
que les Helvetes se dirigeaient vers I'endroit ou Arioviste campait.'
Very well but how does he explain the startlingly devious character
;

of the trek ? Read on.

' '
Quant a la route prise par les Helvetes,
'

elle pent sembler " insane " a celui qui I'etudie sur une carte de
I'Europe moderne. Evidemment ceux qui veulent aujourd'hui se
rendre du territoire situe entre les Alpes et le Jura dans la France
du nord-est n'ont pas besoin de passer par Macon, Autun, et Langres.
Pourquoi les Helvetes ont-ils pris, il y a dix-neuf siecles, ce chemin ?
Parce qu'ils voulaient passer par le riche territoire des Eduens, qui
etaient leurs amis et qui pouvaient leur donner des vivres.' ^
It appears, then, that although we must not believe Caesar when
he says that the Helvetii forcibly took supplies from the Aedui,
'
although Dumnorix's popularity would be difficult to understand '

if they had done any such tiling,^ the Aedui were ready and willing

to give them supplies. But if so, why did the Helvetii move westward
from the Saone when ex hypothesi their goal was in North-Eastern
Gaul ? If they had pushed on up the valley of the Saone, would
they not still have been in Aeduan territory, and would not supplies
have been still forthcoming ? And since Signor Ferrero assures us
that all the Nationalists in Gaul were their friends, w^hy did they not
draw their supplies from friends who were not remote from their
alleged natural route ? Why go a hundred miles out of the way to
feed at the expense of the Aedui when the eastern and north-eastern
tribes were available ? No, Signor Ferrero, your explanation is
a little too far-fetched. If Divico really planned that circuitous
'
trek ', a Bedlamite he must have been !

But, despite the stubborn fact that the Helvetii struck westward
from the Saone in the last stage of the march that preceded their
'
a Q., 1909, p. 209. » lb., 1910. p. 31. ='
lb,, p. 33.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 223

defeat, Sigiior Ferrero contends that their goal could not have
been the country of the Santoni, because Si les Helvetes allaient
'

leur chemin avant la bataille, quand ils etaient talonnes par les
Romains, il est au moins temeraire de supposer qu'ils se sont
jetes a I'aventure, dans la Gaule, sur la premiere route qu'ils ont
trouvee, quand ils avaient reussi a immobiliser i'ennemi pour trois
jours et a le distancer de toute la marche qu'ils avaient pu accomplir
en trois jours,' ^
I supposed no such thing the Helvetii did not march at random.
:

After their defeat they abandoned the hopeless enterprise of march-

ing to Saintonge they knew that Caesar could, as he did, overhaul
:

them, and their prestige was shattered. From the direction which
they took after their defeat Signor Ferrero formerly concluded that
they were marching towards the Rhine with the object of attack-
' '

ing Ariovistus ;he is now forced to amend this conclusion and to

say that they were marching towards North -Eastern Caul is the
:

second inference more valuable than the first ? And, once more,
I ask, if Caesar had had any interest in deceiving his countrymen as
to the place in which they originally intended to settle, would he
have hit upon the country of the Santoni, the mention of which was
not unlikely to excite suspicion ? Would it not have served the pur-
pose of a liar better to say that they intended to settle in the country
of the Aedui ?
M. Holmes,' says Signor Ferrero, me demandera probablement
' '

sur quoi je m'appuie pour affirmer que Cesar a alt ere la verite, quand
il nous a decrit les Eduens implorant son aide centre les Helvetes

qui mena9aient de omnes fortunas sociorum consumere. Sur ce fait :

que quelques chapitres plus loin (au XVII'i^e) Cesar est oblige
d'admettre que Dumnorix, le protecteur des Helvetes, etait si popu-
laire, qu'il tenait en echec tout le gouvernement et que celui qui
toucherait a lui serait deteste par toute la Gaule. II serait difficile
de comprendre une telle popularite, si les Helvetes avaient mis la
Gaule a fer et a feu, comme on le raconte dans le onzieme chapitre.'
Then does Signor Ferrero also deny the truth of B. G., i, 15, § 4,
where Caesar us that while he was following the Helvetii through
tells
Aeduan territory he thought it enough to prevent the enemy from
'

looting, foraging, and ravaging the country (satis hahehat in jprae-

'

sentia hostem rapinis pahulationibus populationihusque prohihere) ?

Does Signor Ferrero believe that a host numbering over three hundred
thousand souls could have been prevented by their leader from
attempting to plunder ? In his eleventh chapter Caesar says nothing
about fire and sword
'
he only says that the Helvetii ravaged
'
:

(populabantur) Aeduan territory. According to Signor Ferrero, they

had gone far out of their way to that territory for the sole purpose
of procuring food, which the Aedui were willing to give them but ;

ifthey had plundered it (for aught we know, from philo-Romans),

Dumnorix's popularity would be incomprehensible Yet it would
!

C. Q., 1910, p. 32.

'
I explain on pp. 50, 232, why the Helvetii, in order to
reach Saintonge, marched through Aeduan territory.
224 THE CREDIBILITY OF CAESAR'S NARRATIVE
seem that he managed somehow to retain his popularity with the
unhappy people of North-Eastern Gaul, whose fertile lands, accord-
ing to Signor Ferrero, he had promised to assign to them.^
I venture to' say that the famous historian has entered a path
M'hich does not lead to historical truth. It is dangerous to mate
scepticism with imagination for the offspring thereof will be ille-
;

gitimate fiction.
2. Caesar tells us that the date [according to the unreformed
calendar] fixed for the muster of the Helvetii on the right bank of
the EhAne w^as the 28th of March, 58 b. c. When he reached the
neighbourhood of Geneva, Helvetian envoys presented themselves,
and asked him to allow the host to march through the Province.
Wishing to gain time, he told them that he would take a few days
to consider their request, and would give them an answer on the
13th of April. Meanwhile he constructed entrenchTuents along the
left bank of the Rhone between Geneva and the Pas dc I'Eclusp.
On the 13tli of April the envoys returned and he told them that he
;

could not allow the host to pass through the Province .^

Now, observes H. Rauchenstein, according to Caesar's narrative,
the Helvetii were fools. They must have seen his entrenchments in
process of construction. Yet they were so simple as to wait patiently
for the date which he had fixed. He lies when he says that they
tried to storm his entrenchments. Dion Cassius does not mention
anything of the kind and the Helvetii w^ould not have needlessly
;

exasperated the Romans and endangered the success of their enter-

prise by defying Caesar. The truth is that he invented the story of
their attempt to cross the Rhone in order to excuse himself for
having afterwards attacked them.^
I reply that, on Rauchenstein's theory, Caesar was a fool and, ;

whether he was a knave or not, he was assuredly a great captain. He

would not have constructed his lines unless he had had good reason
to believe that he could either do so unobserved or in spite of detec-
tion, and good reason to fear that otherwise the Helvetii might

'
Mr. Heitland has sent me a MS. note, which ends with the words,. I have
'

only to express my belief that in rejecting the reconstruction of Caesar's story

by Signor Ferrero Mr. Rice Holmes is right. That story 1 believe to be true
in the main, though I agree with the clever Italian critic that in it the instances
of artful colouring are not far to seek.' Granted. Caesar naturally made out
the best case that he could for himself, not falsifying facts, but emphasizing,
as he had a right to do, impersonal motives. No doubt,' as Mr, Heitland
'

says, it suits his purpose to lay stress on the order of the Senate commending
'

the Aedui ... to the protection of the Governor of Roman Gaul. Still, unless
this order be a fiction, there it was. . .On nearer acquaintance Ariovist turned
.

out to be a Friend whose presence in Gaul was incompatible with Roman

overlordship. Therefore he had to go, and the pitiful entreaties of the pro-
Roman nobles served as an excuse or pretext for undertaking what had in any
case to be done if Rome was to justify her position as protector.' Perhaps
A. Klotz {Caesarstiidien, pp. 22-5) is right in saying that Caesar took pains
to justify his first two campaigns because in crossing the frontier of the Province
he had acted without senatorial sanction. The campaigns of 57-52 b. c. followed
as a matter of course.
* B. G., i, 6, § 4
; 7, §§ 3-G
; 8, §§ 1-3.
" Der Feldzug Ccisars gegen d. Helvetier, 1882,
pp. 51-4.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 225

succeed in forcing a passage. Napoleon III has shown that the lines
could have been made in two or three days. Rauchenstein assumes
that the Helvetii assembled on the 28th of March, the date which —
their leaders had fixed. But this is not proved nor is it proved
:

that, even if they did, they were aware that the lines were in process
of construction for, as Colonel Stoffel has explained,^ the banks of
;

the Rhone, in the first nine miles ^ of its course below Geneva,
were so steep that no lines were there required. Dion Cassius does
not, it is true, expressly say that the Helvetii attacked the lines :

but his narrative ^ implies either that they did attack them or
that they did not know of their existence until they saw them ;

and either alternative is fatal to Rauchenstein's theory. If they

had shrunk from offending the Romans, they would not have
emigrated at all and, as I have already shown, it was quite
;

unnecessary for Caesar to resort to fiction in order to justify his

attack upon them.*
3. With regard to the third charge, what Caesar wrote was, Caesari
nuntiatur Helvetiis esse in animo per agrum Sequanorum et Haeduorum
iter in Santonum, fines facere, qui non longe a T olosatium finihus absunt,
quae civitas est in provincia. Id sifieret, intellegebat magno cum peri-
culo provinciae futurum ut homines bellicosos, populi Romani inimicos,
locis patentibus maximeque frumentariis finitimos haberet (' Caesar
was informed that the Helvetii intended to march through the terri-
tory of the Sequani and the Aedui, and to make for the country of
the Santoni, which is not far from that of the Tolosates, a Provincial
tribe. He saw that, if this happened, it would be very dangerous to
the Province to have a warlike population, hostile to the Roman
People, close to its rich and defenceless cornfields ').^ Non longe,'

indeed !
says Professor Sihler.
' About 200 miles.' ® The map
'

shows that the distance from the nearest frontier of the Santoni to
Tolosa (Toulouse) is 210 kilometres, or 130 miles."^ But let that pass.
The point is that, as Long says,^ there is no obstacle to an army
'

marching from Saintes to Toulouse the road is open

; and the ;

Helvetii would have been dangerous neighbours to the Provincia, if

they had planted themselves on the lower Garonne.' The Gauls had,
for centuries, been the dreaded enemies of Rome. To say nothing of
the probability that Caesar may have been misinformed as to the
distance, the Senate had just done its best, by diplomatic means, to
compass the exclusion of the Helvetii from Transalpine Gaul.^ Can

^ Napoleon III, Hist, de Jules Cesar, ii, 49-53, note. Cf. F. FrohHch, Die
Glaubwurdigkeit Caesars, &c., p. 20.
^ 14 kilometres, or 8| miles nearly. ^ xxxviii,
32, § 1.
* M. Jullian {Hist, de la Gaule, iii, 199 and n. 5) accepts Caesar's statement

literally, and remarks that le mur a pu etre place sur le rebord des plateaux
'

et servir surtout a relier les differents corps depuis Geneve jusqu'a I'eperon
du Vuache, face au pas de I'Ecluse.' But was such a rampart required ? See
pp. 614-5, infra.
5 B. G., i, 10,
§§ 1-2. « Class. Rev., iv, 1890, p. 154.
' See Desjardins, Geogr. de la Gaule rom., ii, 603, n. 2, and Napoleon, Hist,

de Jules Cesar, ii, 55, n. 3.

8 Caesar,
p. 49. " Cicero, AtL, i, 19, §2.
1093
Q
226 THE CREDTBTLTTY OF CAESAR'S NARRATIVE
any one honestly argue, on the bare evidence of the words non longe,
that Caesar was so afraid lest public opinion at Rome should
condemn him if he repelled the invasion of the Helvetii, that
he thought it necessary to tell a lie in order to smooth over
his iniquity ?

But M. JuUian,^ who defends the general trustworthiness of the

Commentaries, scoffs at this particular passage. Were not the
Arverni, he asks, as dangerous neighbours as the Helvetii would
have been ? Were the Helvetii less dangerous behind the lake of
Geneva than they would have been near the Atlantic ? Toutes ces '

paroles de Cesar n'etaient que des formules a I'usage du peuple

romain.' But was the presence of one danger a reason for ignoring
another ? The time had not yet come for dealing with the Arverni :

while the Helvetii remained beyond the lake they had enough to do
in repelling the Germans and so long as they did not threaten
;

Rome, a Roman general could not attack them. Was Caesar to

look on passively while they and Dumnorix were maturing their
schemes ?

4. Finally, are told that the whole drift of his own report
we '
. . .

is that he was continually /orced^ to enter upon new enterprises and

campaigns against his own wish and expectation.' ^ This criticism
has no point unless it means that his own report was so far false,
'
'

and that he hypocritically endeavoured to disguise the fact that his

purpose from first to last was to conquer Gaul. Whether, when he
first crossed the Alps in the spring of 58 b. c, he had definitely made
up his mind to undertake the conquest, we cannot tell. But that,
before the end of the same year, he saw his way clear to doing so,
he has left us in no doubt. Has he not
told us that he quartered his
legions, for the winter of 58-57
c, at Vesontio (Besancon), outside
b.
the Roman Province ? Has he not frankly avowed or rather —
—
related as a matter of course that in 57 B. c. he sent one of his
lieutenants to receive the submission of a group of tribes which had
offered him no opposition, but which his policy required him to
subdue. In B. G., ii, 34 I read, Eodem tem'pore a P. Crasso, quern cum
leqioneuna miserat ad Venetos, Venellos, Osismos, Coriosolitas, Esuvios,
Aulercos, Redones, quae sunt maritim^e civitates Oceanumque attin-
gunt, certior factus est oynnes eas civitates in dicionem fotestatemque
fopuli Boynani esse redactas (' At the same time he was informed by
P. Crassus, whom he had sent with a single legion to the territories
of the Veneti, Venelli, Osismi, Coriosolites, Esuvii, Aulerci, and
Redones, maritime tribes, whose country reaches the Ocean, that all
of them had been brought completely under the dominion of the
Roman People'). Could anything be plainer ? Is there the least
attempt, in this part of his report ', to make us believe that he
' '

was forced to enter upon a new enterprise against his own wish and
expectation ? Therefore, when he tells us, as he sometimes does,
'

that he was forced to undertake a campaign unexpectedly, it seems

natural to conclude, not that he is lying, but that he is showing us

1 Hist, de la Gmde, iii, 201. '

Class. Rev., iv, 1890, p. 198.
 —
THE CREDIBILITY OF CAESAR'S NARRATIVE 227

how much more arduous the conquest of Gaul turned out than
might have been anticipated. Having made up his mind to conquer
the country, he naturally desired to do so at the least possible ex-
penditure of valuable lives. Why then should we suspect him of
hypocrisy when he tells us that, after the whole country had ap-
parently submitted, he was again forced to take the field ? Does
his accuser mean to argue that he was not forced to enter upon
'

fresh campaigns at all, in other words, that his descriptions of the

'

circumstances which led to the campaigns are fictitious or distorted ?

If so, the charge is simply gratuitous. Or does he, without impugn-
ing the accuracy of Caesar's narrative, simply mean that he intended,
whether he were forced or not, to conquer Gaul ? If so, he has
' '

told us no more than Caesar has enabled us to see for ourselves.

Supposing that Caesar had directly avowed his design, instead of
leaving us to infer it, would there be anything surprising in his having
told us that he was forced to undertake fresh campaigns against his
wish and expectation ? Let me give a brief analysis, which any one
can verify for himself, of the causes which he assigns for his succes-
sive campaigns. He tells us that the Helvetii invaded Gaul and
refused to give hostages for their good behaviour and he implies
;

that he thereupon determined to eject them. He tells us that Ario-

vistus intended to settle permanently in Gaul, and refused to restore
the hostages whom he had taken from the Aedui, or to pledge himself
to bring no more Germans across the Rhine. He tells us that the
Belgae conspired against him because they were afraid that he in-
tended to conquer their country, and because they were egged on
to fight by influential Gauls who objected to the Roman legions'
wintering in Gaul, and by others who feared that the establishment
of Roman rule would upset their plans for seizing supreme power in
their respective states. He frankly avows, as I have already re-
marked, that he sent Publius Crassus to reduce the maritime tribes
to submission and he does not pretend that those tribes had given
;

him any provocation. He goes on to relate that they revolted that ;

he suppressed their revolt without mercy that he sent Crassus to

;

prevent the Aquitanians from helping, if they intended to help, the

tribes of Gaul that the Aquitanians combined to resist Crassus
;
;

and that Crassus defeated them.^ In the Fourth Book he relates

^ Mommsen remarks {Hist,

of Rome, v, 1894, p. 500, note [Rom. Gesch., iii,
1889, p. 616, note]) that Caesar's attempt to justify Crassus's invasion 'as
a defensive measure which the state of things had rendered inevitable breaks
'

down. Read Caesar's statement (-S, (?., iii, 11, § 3) of his reason for having sent
Crassus on this errand P. Grassum
: in Aquitaniam proficisci iubet, ne ex
. . .

his nationibus auxilia in Galliam mittantur ac tantae nationes coniungantur

(' P. Crassus ... was directed to march for Aquitania, in order to prevent the

dispatch of reinforcements to Gaul from the peoples of that country, and the
junction of two powerful races'). It is generally assumed that the danger
which Caesar professed to fear did not exist, because the Aquitani had apparently
no political connexion with the Celtae or the Belgae, and were ethnologically
distinct from both ; but is it quite certain that the alleged danger, which, as may
be gathered from B. 0„ vii, 31, § 5, did exist, was not removed by the severe lesson
which Crassus gave the Aquitani ? and, if the danger was really imaginary,
Q 2
 228 THE CREDIBTLTTY OF CAESAR'S NARRATIVE
that the Usipetes and Tencteri, expelled from their own country,
crossed the Rhine with the intention of settling in Gaul that he ;

warned them to recross the Rhine and that, in consequence of their ;

having violated a truce, he attacked and destroyed them. The re-

^mainder of his work, in so far as it relates to Gaul, is devoted to
a narrative of the insurrection of the Eburones and their allies in
54-53 B. c. and of the great rebellion of 52 b. c.
This is a fair summary of the reasons which Caesar gives for
having undertaken his successive campaigns. Those reasons appear
to me perfectly natural and perfectly consistent with the theory that
what he intended from first to last was the conquest of Gaul. I can
hardly believe that those who have impugned his good faith imagine
that the moral sense of Roman society was likely to be shocked by
the conquest, and that therefore it was necessary for him to veil it
as decently as he could. There are Little Englanders but if there ;

were Little Romans, I never heard of them. The conquest awoke

the deepest enthusiasm in Italy. Possibly Cato and his followers
may have been genuinely indignant at such acts as the slaughter of
the Usipetes and the Tencteri but it Avas precisely his most appall-
;

ing acts of severity which Caesar related with the greatest emphasis
and the greatest precision. The Commentaries were an apologia :

they were not an apology.

It must be remembered, moreover, that Caesar had already taken
care, by his negotiations with Pompey and Crassus at Luca, to secure
absolution from the Senate even while the war was going on. Momm-
sen ^ thinks that the Commentaries were designed partly to justify '

as well as possible before the public the formally unconstitutional

enterprise of conquering a great country and constantly increas-
. . .

ing his army without instructions '. This may be true

. . . but in ;

the conference at Luca Caesar had played his cards so well that the
Senate had easily been induced to vote pay for the legions which
he had raised on his own responsibility,^ and Cicero, in his speech
De Provinciis Consularibus, had celebrated the glories of Caesar's
exploits in Gaul. Too much has been made even of Caesar's anxiety
to defend himself in his Memoirs on purely constitutional grounds.
His book was not likely to win him many new adherents. He relied
principally upon other means, gold and the powerful advocates —
whom gold kept on his side. When Mommsen insists upon the
necessity under which Caesar found himself of arguing that he had
been justified in acting without the sanction of the Senate, does he
not forget that the Senate, by the repeated thanksgivings ' which '

it had decreed in honour of his victories,^ had virtually sanctioned

is it knew it to be so ? Mommsen may be right but

certain that Caesar ;

if so, why
did Caesar take no pains to apologize for having sent Crassus
to reduce the maritime tribes, who had offered him no provocation, to sub-
mission ? Why did he simply record the fact ? Surel}^ because it never
occurred to him that his conduct required an apology.
1 Hist,
of Rome, \, 1894, p. 499 {Rom. Gesch., iii, 1889, p. 615).
- Cicero, Fam., i, 7, § 10.
^ B. G., ii, 35, § 3 ; iv, 38, § 5 ; vii, 90, § 8.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 229

advance V ^ Such argument was really needed only in

his action in
connexion with the campaigns of 58 b. c. and, as I have already
;

shown, Caesar was able to justify himself without departing from

the truth.

VI
Now to deal with the critics who complain that Caesar
lied from
motives of vanity. The Due d'Aumale observes that Caesar was not
as candid as Turenne, who frankly told an indiscreet questioner that
he had lost the battles of Mariendal and Rethel by his own fault.^
I freely admit it. Caesar did not think it necessary to anticipate the
censure of military critics and he was quite right. He made mis-
;

takes, like every other general and if he does not call our attention
;

to them, neither does he conceal them. He tells us quite enough to

enable us, if we know our business, to see where he went astray.^
The charges which I propose now to examine are (1) that Caesar —
concealed certain reverses which he had sustained (2) that he took
;

to himself the credit, which really belonged to Labienus, of having

defeated the Tigurini (3) that his account of the later stage of his
;

battle with the Helvetii is incredible (4) that he suppressed im-

;

portant facts in his account of the sea-fight with the Veneti (5) that ;

he constantly exaggerated the numbers of his enemies (6) that his;

account of the predatory expedition of the Sugambri in 53 b. c. is

garbled (7) that he disguised the defeat w^hich he had sustained at
;

Gergovia (8) that he pretended to have subdued several German

;

peoples, whereas only the Ubii had in fact submitted and finally, ;

that he was unjust and ungenerous in his notices of the services of

his lieutenants.
1. Rauchenstein contrasts the story which Caesar tells in B. G.,
i,48-51 of the operations which preceded the defeat of Ariovistus
and of the outset of the battle itself with Dion Cassius's narrative.
Caesar says that, for five days after Ariovistus made the flank march
by which he hoped to cut off the Romans from their supplies, he
daily offered battle, but that Ariovistus remained shut up in his
camp and only ventured upon cavalry skirmishes that afterwards,
;

when he himself had regained communication with his convoys by

constructing a smaller camp, Ariovistus made an unsuccessful
attempt to storm it, in which both sides suffered heavy loss and ;

that, having learned that Ariovistus had been warned by the German

^ A generous reviewer of the

first edition {Athenoeum, Jan. 13, 1900, p. 41)
who acknowledges the substantial fairness of this essay, remarks that the
'
'
'

honour of the supplicatio was sometimes good-humouredly granted to com-

manders who were known to have hugely exaggerated their achievements'.
No doubt ; but the supplicationes which were granted to Caesar were pro
tanto a justification of his policy. Moreover, if, as Suetonius apparently means
(see pp. 40-1), the Senate had given Caesar the whole of Transalpine Gaul for
his sphere of action, the grant, as Long says {Decline of the Roman Eepuhlic,
iii, 433), was equivalent to a commission to make war in that country '.
'

Rev. des Deux Moiides, 2" per., xv, 1858, p. 119.

•^

For an instance in which an acute critic has detected that Caesar was
'

outwitted by Vercingctorix sec ])p. 744-5.

230 THE CKEDIBILITY OF CAESAR'S NARRATIVE
matrons not to fight a pitched battle before the new moon, lie forced
Ariovistus's hand by attacking him on the very next day. Dion
Cassins, on the contrary, says Rauchenstein,^ tells us that the
Romans were defeated in minor engagements by Ariovistus, and
that, setting at nought the warnings of the matrons, he attacked
them when they were drawn up in battle array. What Dion
Cassius really says is that Ariovistus greatly harassed {L(Txvp(h<s iXvireL)
the Romans with his cavalry that he nearly took Caesar's smaller
;

camp and that, emboldened by this success, he paid little attention

;

to the warnings of the matrons, and on the next day, when the
Romans moved out of camp and formed in line of battle, he made
his troops do the same.^ Having regard to the general character of
Dion's narrative, which is throughout, for the most part, obviously
a condensed paraphrase of Caesar's, to his love of rhetorical em-
bellishment and to the monstrous blunders into which this taste
occasionally hurries him, I see no reason to believe that, in the
present instance, he used any independent authority. The phrase
Icrxvpv? iXvTTu and the statement that Ariovistus nearly took Caesar's
smaller camp I regard as mere flourishes or hasty inferences.
Rauchenstein says that C^aesar's narrative of the commencement
of his battle with Ariovistus is not only improbable but impossible.
Caesar tells us that he marched right up to the German camp [usque
ad eastra liostkmi accessit) and that the Germans then perforce
;

(necessario) led their troops out of camp, and formed them up in

groups composed of their several tribes.^ Upon which Rauchenstein
remarks, Caesar is so polite as to wait until his opponents form up.
'

I cannot see why the Germans were obliged to come out and attack
Caesar. Why could not Ariovistus have waited securely in his camp
and left the initiative to Caesar ? ' '*

Now all this is mere quibbling. Rauchenstein lays undue stress

on the words usque ad (eastra). Caesar does not mean that the
Germans waited until the Romans were near enough to be able to
touch the wagons. When they saw the Romans advancing with the
evident intention of forcing on a battle, they were compelled to
defend themselves. Being brave, in spite of their superstition, and
seeing that they could no longer defer the battle, as their wise women
had advised, they determined, as fight they must, to put themselves
between the Romans and their helpless women and children, to fight
manfully in the open, and not merely to defend themselves but to
beat the Romans. Rauchenstein forgets that their wagons were on
their flanks and in their rear and therefore Caesar would not have
;

been obliged to storm a laager.^

According to Rauchenstein, Caesar's account of his operations

against the Atuatuciis equally disingenuous. Caesar says that when

^ Der Feldzug Gdsars gegen d. Helvetier, pp. 26-7.

" xxxviii, 48-9. ^ B. a., i,
51, § 2.
* Cf. G. FciTcro, Grandezza e decudenza di lioma, il, 28, note (The Greuincdb
and Decline of Borne, ii, 24, n. f). ^ ^- G., i, 51, § 2. Cf. p. 053.

J
 —
THE CKED1J3ILITY OF CAESAR'S NARRATIVE 231

his army encamped before tlicir stroiigliold, they made frequent

first
and liad skirmishes {2)arvulis proeliis)
sorties (crehras excursiones)
with his troops but that afterwards, when he had regularly invested
;

the place, they did not venture out any niore.^ Dion Cassius says
that they repulsed Caesar's attacks for many days until he had
recourse to the construction of regular siege works.^ Dion's account,
says Rauchenstein, is the more credible of the two, because it was
only natural that Caesar should try to storm the town before he
undertook a regular siege.^
Here is a cobbler who flings his last out of window, and must
needs set up for a professor of the art of war. The idea that Caesar
should have tried to storm out of hand a town strongly situated
upon a rocky height and defended, in the only place where it was
open to attack, by two high walls, is so delicious that it almost ceases
to be absurd. Caligula would not have done such a thing And !

Rauchenstein asks us to believe that Caesar, not content with making

one attempt and getting badly beaten, tried again day after day !

Now let us turn to Dion Cassius. Any one who is familiar with the
Greek's book can see that here, as usual, he is simply paraphrasing
and trying to embellish Caesar's plain narrative. He infers from
Caesar's expression, jparvulis proeliis, that he tried to take the town
anyhow and he puts in, as a touch of his own, that he was
;

repulsed.

2. According to Appian * and Plutarch,^ it was not Caesar who

defeated the Tigurini, but Labienus ; and Rauchenstein argues that
—
Plutarch's words ovk avTo<s dXAa Aa/3t7;vos show that he intended
to correct Caesar. Caesar, says Rauchenstein, according to his
own account, acted like a fool. His object was to prevent the
Helvetii from reaching the country of the Santoni. On his return
from Italy he encamped in the neighbourhood of .Lyons. The
vanguard of the Helvetii was much further south for, 20 days
;

previously, it had crossed the Saone about 9 miles north of Lyons ;

and the column must in that time have marched a consider-

able distance in a south-westerly direction. Yet Caesar, according
to his own account, now marched northward, destroyed the Tigurini,
crossed the Saone, and then made a flank march of at least two days
in order to force the Helvetii to turn towards the north. This would
have been a strategical absurdity. What really happened was this.
While Caesar remained encamped on the right bank of the Rhone
in the neighbourhood of Vienne, Labienus attacked and defeated
the Tigurini, crossed the Saone, and marched on in pursuit of the
Helvetii. Caesar barred the south-western route towards the country
of the Santoni and the Helvetii, finding themselves outmanoeuvred,
;

sent Divico to negotiate.^

Now Caesar gave all his lieutenants, and especially Labienus, full
'
B. G., ii, 30, §§ 1-2. - xxxix, 4, § 2.
^ Der Feldzwj, &c., p. 28. . ' Celtica, 1, § 3.
'"
Caesar, 18. ''
Der Feldzuy, &c., pp. 05-70.
232 THE CREDIBILITY OF CAESAR'S NARRATIVE
credit for their exploits. Even if he had been minded to rob Labienus
of his due, he must have known that every officer who had served
under him would detect his lie, and would make the truth known
privately if not publicly in short that he would gain nothing and
;

lose much.
Rauchenstein's argument is based upon erroneous assumptions.
First there is no evidence that Caesar encamped near Vienne or at
any point on the right bank of the Rhone, below its confluence with
the Saone; and on pages 617-9 I show that he encamped, probably on
the heights of Sathonay and certainly in the angle formed by the
confluence of the Saone and the Rhone. Napoleon conjectures that
he posted a detachment on the right bank of the Saone, at or near
Lyons, to intercept the road which led into the Province. But it is
more than doubtful whether the Helvetii would have taken this road
in any case ^ as Long and Napoleon have shown, the south-western
:

was far more difficult than the north-western route, by which they
marched when Caesar was pursuing them and Napoleon observes ;

that at an epoch very near to our own, before the construction of

'

the railways, the public conveyances, to go from Lyons to La Rochelle

. .took the direction to the north-west, to Autun, and thence to
.

Nevers, in the valley of the Loire '.^ Moreover, during the whole of
the 20 days which had elapsed since the head of the Helvetian
column gained the right bank of the Saone successive sections of
the column were crossing the river and it is very unlikely that the
;

vanguard would have endangered its own safety by moving away

before the whole column got safely across.
I believe that Plutarch and Appian either drew hasty inferences
from the fact that Labienus had been left by Caesar on the east
of the Saone,^ or that, like Rauchenstein, they made the mistake of
assuming that Caesar was encamped on the west of that river.
Blunders far more gross than theirs are to be found in Mr. Fronde's
Caesar.'^ At all events on this point, Dion Cassius,^ to whom Rauchen-
stein so often appeals against Caesar, confirms him and, as Rauchen-
;

stein maintains that Dion used other authorities besides Caesar for
his narrative of the Gallic war, one would think that, if Caesar
falsified this part of his narrative, Dion would have been able and
eager to correct him.^ Moreover, it must be remembered that
Labienus had been left with only one legion, whereas the attack on
the Tigurini was made with three. Therefore on Rauchenstein's
theory, Caesar must have sent two legions from the right to the left

' See Long's Decline of the Roman Republic, iv, 11.

2 Hist, de Jules Cesar, ii, 61. ^ B. G., i, 10,
§ 3.
See Westminster Review, Aug., 1892, pp. 174-89.
* ^ xxxviii, 32, §4.

The defeat of the Tigurini is mentioned not only in the epitome of Appian's
**

narrative of the Gallic war, but also in an excerpt from Appian {Celtica, 15).
In this passage he says that the Helvetii and the Tigurini were two distinct
nations, whereas we know that the Tigurini were only one of the four Helvetian
tribes and he says that the attack on the Tigurini look place after and in
;

consequence of the failure of Caesar's negotiations with the Helvetii. These

two gross blunders do not dispose one to belicxe Appian when he contradicts
Caesar.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 233

bank of the Saone to co-operate with Labienus. In order to cross

the river, they must have made a bridge. Then why did they make
another in order to return ? Kanchenstein's theory refutes itself.^
3. Rauchenstein regards Caesar's account of the later stages of
the battle with the Helvetii ^ as purely fictitious. He says that,
according to von Goler and Napoleon, Caesar's auxiliaries probably
numbered not less than 30,000 men,^ while the two legions which
were detached with them to defend the Roman baggage must have
numbered about 10,000. If Caesar is to be believed, says Rauchen-
stein, these 40,000 men did not stir hand or foot to help their com-
rades, the four veteran legions, in their hard-fought struggle. Caesar
tells us that the Helvetii, after the repulse of their attack on the
four veteran legions, retreated to a hill about a mile off, and there
renewed the battle. This second episode, says Rauchenstein, Caesar
dismisses in a few words. He does not say that the Helvetii were
driven from their position and from his own statement that through-
;

out the whole battle not one of the enemy turned his back upon the
Romans, we may infer that they were not. It therefore appears that
they looked quietly on while Caesar's third line was destroying the
Boi and Tulingi, who had taken refuge behind the laager of wagons.
As, Rauchenstein continues, Caesar's account of the retreat of the
Helvetii to the hill is questionable, it follows that his account of
the capture of the laager is incredible. According to that account, the
Romans surrounded and stormed the laager. Yet, although sur-
rounded, 130,000 persons managed to escape. No prisoners were
made, except two of Orgetorix's children. The Romans were so
obliging as to open their ranks, and let all the rest pass through !

Finally, Rauchenstein oflers us a narrative of his own in place of

Caesar's fiction. (Has the worthy man no sense of humour ?) Caesar's
auxiliaries, he tells us, as soon as they saw the attack of the Boi and
Tulingi, assumed that the battle was lost, and ran for their lives.
Thereupon Caesar retreated to his camp on the hill. The Helvetii
returned to the attack, but failed to storm the camp, and accordingly
marched for Bibracte. Subsequently the Verbigeni thoughtlessly
separated themselves from the main body of the Helvetii and ;

Caesar destroyed them. But why did the Helvetii, in spite of the
fair measure of success which they had gained, return home? Rauchen-
stein finds no difficulty in devising an explanation. They had already
become sick of wandering in Gaul. They had consumed all their
provisions ; and, although they had not been defeated, they had
suffered heavy losses in the battle. They were afraid that Caesar
would get reinforcements and revenge himself for his failure and ;

therefore they decided that their wisest course would be to abandon

'
See B. 0.,i, 8, § 1 ; 10, § 3 ; 12, § 2; 13, § 1, and cf. F. Frolilich, Die
Claubwiirdigkeit Caesars, &c., pp. 21-4. Is it not possible that, as M. Jiillian
suggests {Hist, de la Gaule, iii, 207, n, 4), Labienus may have accompanied
Caesar or commanded a division under him ?
''
See B. a.,i, 24-G.
^ Napoleon {Uisl. de dales Cesar, ii, 73, n. I) really says 20,000, — on the
evidence of Appian ! 2,000 would be nearer tiie luark : see p. 42.

^,^of ^^OMr^^

:^X'' cT wJCF.At^'S
UJ
y
234 THE CREDIBILITY OF CAESAR'S NARRATIVE
their expedition. Finally Rauclicnstein assures us that Caesar's
officers and would not have contradicted his account of
legionaries
the campaign, because they would have been delighted at the clever-
ness with which he managed, by a few dexterous phrases, to trans-
form a drawn battle into a brilliant victory.^
It might be a sufficient answer to Rauchenstein to say that if
Caesar had really sustained a severe check at the hands of the Helvetii
and had concocted a false account of the battle, the facts must, in
spite of his cleverness, have leaked out. Surely Dion Cassius, by
whom Rauchenstein sets such store, would not have failed to give
us the true version.^ But I will examine Rauchenstein's arguments.
There is no evidence, and it is to the last degree improbable that
Caesar's auxiliaries numbered 30,000 men,^ but that point is quite
innnaterial. If they and the newly raised legions merely protected
the baggage, the reason was that Caesar did not think it wise to
expose them in his first battle, or that he felt able to win the battle
without their help, or, possibly, that he had reason to fear that the
hostile party among the Aedui might attack the baggage.^ Secondly,
Caesar implies, if he does not expressly say, that the Helvetii were
driven from the hill. He describes the last stage of the action in
a few vivid sentences. Two battles, he says, went on simultaneously,
— one between the first two Roman lines and the Helvetii, the other
between the third Roman line and the Boi and Tulingi. The Helvetii
were forced further and further up the hill the Boi and Tulingi
:

withdrew to their wagons, from which they were at last driven ;

and about 130,000 souls in all survived the battle. These 130,00()
were not all in the laager, as Rauchenstein absurdly pretends :

they included the Helvetii. Otherwise, what became of the Helvetii ?

The statement of Caesar, to which Rauchenstein refers, that through-
out the whole battle not one of the enemy turned his back upon the
Romans, interpreted as he perversely chooses to interpret it, would
prove too much it would prove that neither the Helvetii nor the
:

Boi nor the Tulingi ever retreated from the battle-field at all. Ob-
viously Caesar means that, while the battle lasted, the Gauls all
fought like men he does not mean that, when they were at last
:

overpowered, they even then refused to retreat for he says that

;

they did retreat. He does not say that the Romans surrounded the
laager he merely says that they stormed it and therefore Rauchen-
: ;

stein's sarcasm falls flat. Nor does he say that no prisoners were
taken, except Orgetorix's children. He simply mentions those two,
because they were worthy of mention.^ According to Rauchenstein's
estimate, Caesar's force equalled, if it did not outnumber the force of
^ Der Feldzug, &c., pp. 92-101. Cf. G. Ferrero, Grandezza e decadenza di
Roma, ii, 16, n. 2 {The Greatness and Decline of Rome, ii, 14, n. I).
^ I find that R. Schneider {Jaliresh. d. philol. Vereins zu Berlin, xiii, 1887,
pp. 38G-7), in a review of Petsch's Die hist. Glauhivardigkeit d. Comm. Cusars
vom gall. Kriege, 1885-6 (which I have tried in vain to procure), remarks that
Petschhas rightly pointed out that the later hi«torians, witli various deviations,
arrive at the same results as the Coinmtnlaries, which tends to show that where
they dilforcd from Caesar, they misunderstood or perverted his statements.
' Sec p. 42. ' Sec
p. 628. B. G., i. 24-0.
'"
 2 —
THE CREDIBILITY OF CAESAR'S NARRATIVE 235

llic (jicUils was better armed and better disciplined and it was
: it ;

connnanded by one of the great generals of the world. Is it incredible

that, after a hard-fought battle, it should have gained a decisive
victory ?

hardly necessary to point out the absurdity of Rauchenstein's

It is
guess about the Verbigeni. Why should they have separated them-
selves from the main body of the Helvetii, if they had not intended,
as Caesar says, to make a dash for liberty, after the main body had
surrendered ? If Caesar destroyed the Verbigeni and Rauchenstein —
—
admits that he did it is obvious that he pursued the Helvetii. Is
it credible that he would have pursued a host which had virtually
beaten him ?
Caesar alleges, as his reason for not having pursued the Helvetii at
once, that he was obliged to remain three days on the field of battle,
in order to bury the dead and to allow his wounded time to recover.^
Rauchenstein refuses to accept this reason. It was Caesar's custom,
he says, to follow up a victory and a single legion would have
;

sufficed to bury the dead. He remained on the battle-field although,

according to his own statement, he had only enough food for a single
day and if he had found stores in the deserted Helvetian encamp-
;

ment, he would have said so. Evidently therefore he must have

feared that the Helvetii would rally, and, finding themselves un-
pursued, would march to Bibracte and there lay in fresh stores of
pro visions.
Now Caesar frankly avows that he suffered heavy loss in the
battle his cavalry were weak and untrustworthy ^ and, as Stoffel
: ;

points out, he was in the country of his allies, the Aedui, and was
therefore bound, out of consideration for them, to bury the vast
heaps of dead, in order to avoid a pestilence.* These appear to me
sufficient reasons for his inaction and evidently he implies that
;

after a hardly won victory his troops were unfit for a fresh effort.
He unquestionably remained master of the field of battle and he ;

knew that he could overtake the unwieldy and shattered host when-
ever he pleased. It is probable that, during his three days' stay on
the battle-field, he obtained stores from Bibracte and, if he had ;

found stores in the deserted Helvetian encampment, it would have

been quite in accordance with his laconic manner to omit all mention
of the fact. Anyhow that he did get stores somewhere and somehow
needs no demonstration. Finally, the causes which Rauchenstein
invents for the return of the Helvetii are totally inadequate.^
Colonel Bircher,^ indeed, asserts that troops who fight as the
^ B.G., i, 26, § 5. 2 Der Feldzug, &c., pp. 92-4.
^ ^. G^., i, 15, § 1 26, § 5.
; 18, § 10 « Guerre civile, ii, 452.
;

^ Cf. F. Frohlich, Die Glauhwi'irdigkeit Caesars, &c., p. 38.

^ Bibracte, 1904, p. 27. Signer Ferrero [Grandezza e decadenza di Roma,
ii, [The Greatness avd Decline of Rome, ii, 15, n. -j-]) agrees witli Bircher
19, n. 1 :

'
The conditions of peace whicli Caesar says that he itnposed upon the
. . .

Helvetii are such as to belie his whole account of the war. It is altogether
unhivcly that tlic Helvetii surrendered because tlie Lingoncs refused to grant . . .

them su])i)lies. Tliey were clearly in a position to take what was not given
thcni.' Now the standing corn was not yet ripe (7>. G., i, 16, § 2) and during ;
 236 THE CREDIBILITY OF CAESAR'S NARRATIVE
Helvetii fouglit do not run away, especially when they are not
pursued. But they do retreat the colonel himself admits that the
:

Helvetii retreated and when he rejects Caesar's account of their

;

surrender and substitutes a story of his own about negotiation,

I cannot follow him.
4. Caesar says that the Roman ships had no advantage over those
of the Veneti, except in speed and in the use of oars.^ I quote from
my translation his account of the battle.^ Brutus, who commanded — '

the fleet, and the tribunes and centurions, each of whom had been
entrusted with a single ship, did not quite know what to do, or what
tactics to adopt. They had ascertained that it w^as impossible to
injure the enemy's ships by ramming. The turrets were run up ;

but even then they were overtopped by the foreigners' lofty sterns,
so that, from the lower position, it was impossible to throw javelins
with effect, while the missiles thrown by the Gauls fell with increased
momentum. Our men, however, had a very effective contrivance
—
ready, namely, hooks, sharpened at the ends and fixed to long
poles. .. By means of these the
. halyards were seized and pulled
taut the galley rowed hard
: and the ropes snapped. When they
;

were cut, the yards of course fell down. When, as we have said,
. . .

the yards fell down, the Roman ships, two or three at a time, closed
round one of the enemy's and the legionaries clambered aboard
;

with the utmost vigour. Several ships had been captured, when the
natives, seeing what was happening and realizing that there was no
help for it, hastened to save themselves by flight. And now, just as
the ships had been put before the wind, there was suddenly a dead
calm, and they could not stir. Our men gave chase and captured
. . .

the ships one after another,' &c.

M. Le Moyne de la Borderie ^ objects that in order to cut the
halyards, the Romans would have had to use hooks fastened to poles
of prodigious length and that so long as the wind lasted, other
;

Venetian ships could easily have come to the rescue of those which
were attacked, and rammed Caesar's small galleys. Evidently, he
concludes, the account of Dion Cassius is the true one. Dion's
account * is much longer than Caesar's. The part which is relevant
to the present discussion may be summarized as follows Brutus :
—
came iK rr}? h'SoO^v OaXd(T(Ty]^, that is to say from the Mediter-
ranean, with the swift ships '. The Veneti were confident that they
'

would be able to sink these ships tol^ kovto2<5, with the boat- — '

hooks.' As long as the w4nd lasted, Brutus dared not attack in :

fact he intended to abandon his ships and repel the enemy's onslaught
on land. Suddenly, however, the wind dropped. Thereupon Brutus
proceeded to attack the Venetian fleet in detail. In some cases he

the few days that elapsed before Caesar overtook the shattered and demoralized
host the Lingones could very well prevent them from seizing sufficient grain
to enable them to pursue their retreat. Anyhow, unless they were prepared to
fight again, they were not in a position to negotiate. Signor Ferrero should
leave cavilling to smaller men. ^ B. G., iii, 13, § 7.
- Caesar''^ Coinni. . Iratidutcd into jbJiKjliali, ItJOS, pp. 8G-8 (Z>. G., iii, 14-o).
. .

'
Hist, de Brdaync, i, 18'JG, pp. 74-0. * xxxix, 40-3.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 237

surrounded one of the hostile ships with several of his own in others :

he attacked several ships with an equal or even a less number.

Some of the Venetian ships were rammed, shattered, and sunk :

others were burnt and some of the Veneti committed suicide.

;

Finally, the Romans cut the rigging of the Venetian ships from '

a distance (Troppw^ei/) with hooks fixed to long poles.

'

Now the reader will have observed that the very statement of
Caesar which M. de la Borderie refuses to accept is repeated by
Dion The Romans, according to both writers, did cut the rigging
!

of the Venetian ships. Is it likely that Brutus, who had sailed from
the mouth of the Loire for the very purpose of attacking the Veneti,
would have determined, as soon as he sighted them, to abandon his
own ships because the wind was still blowing ? Is it conceivable
that the Veneti would have been mad enough to believe that they
could sink the Roman galleys with boat-hooks ? Is it credible that
the light galleys would have been able to ram, shatter, and sink the
Venetian ships, which, as Dion, following Caesar, himself admits, were
far superior to them both in size and strength ? If M. de la Borderie
had known anything about the rigging of ancient ^ or modern ships,
he would not have committed himself to the statement that it was
impossible for the Romans to reach the Venetian rigging with their
grappling-hooks. As Mr. Froude, a practical yachtsman, remarks,
'
It was not difficult to do if, as is probable, the halyards were made
fast, not to the mast, but to the gunwale ^ and Lieut. K. Foote,
'
;

R.N., in conversation with me, has endorsed Mr. Froude's statement.

When M. de la Borderie asserts that other Venetian ships could have
come to the rescue of those which were attacked, he forgets that the
Roman ships had the advantage of the enemy in speed, and, as they
had oars while the enemy had none, could attack them from which-
ever side they pleased. It is certain, and M. de la Borderie admits,
that Caesar won the battle. What conceivable motive could he have
had for telling a lie about the manoeuvre by which he gained the
^
victory ?

5. Max Eichheim, who has achieved notoriety by the eccentricity

and virulence of his invective, insists in a comparatively lucid interval
that Caesar's estimate of the numbers of the Helvetii and their
allies is grossly exaggerated. The length of a column numbering
368,000 persons would have been about 99 miles ^ they could not :

have found provender for their cattle and they could not have
;

quitted their encampment, as Caesar says in B. G., i, 22 that they did,

in the short space of three hours. Again, Caesar says that, when
the decisive battle was fought, all the Helvetian wagons were parked
in one place, —
an obvious lie, because, on his estimate of the numbers
of the host, the wagons would have numbered 36,000, which could

See C. Torr's Ancient Ships, 1894, pp. 78, 81, 94.

^

Caesar : a Sketch, ed. 1886, p. 290.

2

^ I am glad to find that my criticism

of Dion has been anticipated by J. Melber
{Commentationes Woelfflinianae, 1891, pp. 291-7), who tears his account of the
battle to shreds and shows that it has no independent value.
*
161 kilometres.
238 THE CREDIBILITY OF CAESAR'S NARRATIVE
not have assembled on one spot in less than three days.^ Rauchen*
stein, who, in the main, agrees with Eichheim, calculates that, if two
wagons had moved abreast, the length of the column would have
been 165 kilometres, or about 103 miles. Caesar, he says, could
easily have attacked such a column in flank, and broken it up before
the enemy could come to the rescue. According to Plutarch, the
emigrants numbered 300,000 according to Appian, 200,000 accord-
; ;

ing to Strabo, about 408,000 according to Orosius, 157,000.

;

Eauchenstein concludes that Orosius's estimate is the most credible.^

He also argues that if the Helvetii had emigrated en masse, they
must have known that hordes of Germans would promptly occupy
their country and, as Caesar does not say that they had to reconquer
;

it when they were sent back, he concludes that a large number

never emigrated at all.^
I admit that there is some force both in Rauchenstein's and in
Eichheim's arguments. It is difficult to believe that the Helvetii
left their country absolutely deserted. Still, it must be remembered
that other ancient writers besides Caesar describe universal migra-
tions * nor does Caesar's narrative compel us to assume that the
;

Helvetii were the only inhabitants of Helvetia. He may conceivably

have left out of account tribes with which they had not coalesced.^
Moreover, even if the ordinary interpretation of his narrative is
correct, it is not certain that the Grerman neighbours of the Helvetii
would or could have taken possession of their territory in the few
months that elapsed between their departure and their return. If the
Helvetii required two years to complete the preparations for their
emigration, is it likely that the Germans could have transferred
themselves, their women and children, their cattle, and their goods
from one country to another in three months ? If Caesar's narrative
was incorrect, either he intended to mislead his readers, or he was
led astray by false information, or, when he said that the Helvetii
were induced to undertake a national emigration (ut de finibus suis
cum, omnibus suis copiis exirent) he used the word omnibus,^ as he
often did, loosely. It must be remembered that he does not say that
the Helvetii did actually emigrate en masse and I cannot see what
;

motive he could have had for trying to deceive. On the other hand,
nothing is more likely than that he should have been misled by
the rhetorical or blundering statements of his spies. I have myself
worked through many reports embodying the information which had
been furnished to intelligence officers in India by spies and, on the ;

faith of such reports, erroneous statements were frequently made by

^ Die Kdmpfe der Helvetier und Sueben gegen C. J. Ccisar, 1876, p. 31, nn. 13, 20.
2 Orosius generally follows Caesar's numbers with scrupulous accuracy and ;

it is curious that his statement (vi, 7, § 5) of the number of the Helvetii who
— —
returned home 110,000 agrees exactly with Caesar's.
^ Der Feldzug Cdsars gegen d. Helvetier^
pp. 44-5, 47.
"^
See, for instance, Strabo, ii, 3, § 6. Ces emigrations en masse,' says the
'

writer of the article Helvetii in Diet. arch, de la Gaule, ii, 15, entraient evidem-
'

ment dans les habitudes traditionnelles des Gaulois.' The writer refers to
Poly bins, ii, 17.
& Cf. Die*, arch, de la Gaule, ii, 15. « B. G., i, 2, § 1.
 THE OREDIBTLTTY OF CAESAR'S NARRATIVE 239

British officers, for example by Sir Charles Napier. We do not accuse

them of mendacity because they were misled.
According to Napoleon's calculations,^ based on the reasoned
conjecture that the Helvetii had 8,500 wagons, the length of such
a column as Caesar describes would have been 128 kilometres, when
the wagons were moving in single file. But it was not necessary for the
wagons to move in single file, except when they were threading the
Pas de rEcluse.2 Professor Delbriick,^ indeed, who admits that
several wagons might have moved abreast, nevertheless reminds me
that when they met with an obstacle such as a bridge, they must
have moved in single file. Obviously this would depend upon the
width of the defile and after it had been passed the column would
;

resume its usual formation. Besides, a moment's thought will show

that the crossing of a bridge would simply have caused delay :

it would not have increased the length of the column by one yard
more than the length of the bridge. After the Helvetii emerged from
the country of the Sequani, they provided for their cattle by plunder A
while they were in the country of the Sequani, we must assume that
they arranged with that people for a supply. In B. G., i, 21, § 1,
Caesar says that they encamped at the foot of a hill. Eichheim, in his
comment on the passage, absurdly attributes to Caesar the statement
that the Helvetian wagons were all parked there. If he had known
anything of military movements, he would have understood that
most of the wagons were in front, and that only the fighting men
who formed the rearguard, were encamped at the foot of the hill.^
Moreover, Caesar says nothing about three hours '
he only says ' :

that late in the day ', or when it was broad day

' '
(multo die) he '

learned that the Helvetii had struck their camp. His narrative of
—
the battle certainly presents a difficulty, to those who are deter-
mined to make difficulties. What he says is that the Helvetian infantry
'
with all their wagons (cum omnibus suis carris ^) followed their
'

cavalry, and parked their baggage in one spot. But this is not the
same thing as saying that all the wagons were parked in one spot.
Wagons were of course coming up all through the day and doubtless ;

many wagons had not come up when the battle was over.'' Caesar
wrote for sensible readers. They know that thousands of wagons
cannot arrive at one spot simultaneously and, when the sense is
;

clear, they do not demand rigid precision of statement in every line.

To return to Bauchenstein. For the sake of argument I will accept
his figures. He holds that the entire host, after the defeat of the
Tigurini, numbered only 100,000 souls. On his own showing, the
length of the column would then have been at least 43 kilometres,
or about 27 miles. Is he prepared to argue that Caesar could have
broken up a column 103 miles long, but could not have broken up

^ Hist, de Jules Cesar, ii, 59, note.

2 See Lord Wolseley's The Soldier's Pocket-Book, 5th ed., p. 408.
^ Gesdi. d. Kriegskunst, 1900, p. 442.
i, « J5. (?., i, 15, § 4.
5 See Stoffel, Guerre civile, ii, 445, and B. 0., \, 15, § 3 ; 25, § 6.
« B. G., i, 24, § 4.
^ See Stoffel, Guerre civile, ii, 451.
240 THE CREDIBILITY OF CAESAR'S NARRATIVE
a column 27 miles long ? The truth
as Napoleon has pointed out,^
is,

that Caesar could not venture to attack the column at all, because
the country through which it was moving was too broken and confined
to admit of his attempting any offensive operation and because, ;

as Rauchenstein has failed to see, before he could have struck

a single blow at the wagon-train, he would have had to reckon with
the tens of thousands of fighting men who composed its rearguard.
It is idle for Rauchenstein to appeal to Orosius for it requires ;

no critical acumen to see that, where Plutarch and Appian and '

Strabo and Orosius differ so widely, one w^ould not be justified in

accepting Orosius's estimate merely because it is the lowest. Rau-
chenstein's estimate, based upon the absurd assumption that the
wagons moved in single file, of the length to which a column composed
of 368,000 persons must have extended, is a gross exaggeration.
Stoffel, who knows what he is writing about, estimates the length
of the column, after the loss of one-fourth of its number by the
defeat of the Tigurini, at not less than 30 kilometres, or about
19 miles and he accepts Caesar's figures.^ Veith,*^ who also holds
;

that the length of the column has been exaggerated, remarks that
the Helvetii remained in the neighbourhood of Geneva at least
six weeks,* and that their stock of provisions was therefore reduced
by at least one-half (and a corresponding number of wagons rendered
superfluous) before they began to move. Even these, he believes,
were probably abandoned after the Helvetii were disheartened by
the defeat of the Tigurini and thenceforward they may have
;

supported themselves by requisitions alone.^ If so, all the wagons,

except such as were necessary to form the laager and convey those
who could not walk, would have been left behind. At all events
the reasons which Eichheim and Rauchenstein have adduced are
not sufficient to prove that Caesar's figures are incorrect.
Caesar tell us that the host which he sent back to Helvetia num-
bered 110,000.^ Let us provisionally accept his figures. Assuming
that the Tigurini amounted to one-fourth of the entire host, the
number of the remainder would have been 276,000. The rate of
mortality in a vast multitude travelling under such conditions would,
I suppose, have been high and we may safely assume that in the
;

two months which elapsed between their departure from Switzerland

and the battle at least 2,000 died. The Verbigeni, numbering 6,000,
were killed or sold as slaves after the battle.' The Boi, who originally
numbered 32,000, were allowed to settle in the country of the Aedui.*^
126,000 remain to be accounted for ^ and, if Caesar's figures are
;

correct, this number must have perished in the battle or have dis-
^ Hist, de Jules Cesar, ii, 63, note.
'^
Guerre de Cesar 1890, p. 36.
et d' Arioviste,
^ Gesch. d. Feldzilge C. J. Caesars, 1906, p. 493.
* This estimate is apparently based upon ^. (?., i, 6, § 4 ; 7, § 6 ; and 9.
' Cf. B. G., i, 11, §§ 1-5; 15, § 4. « lb.,
29, § 3. i,

' lb., 28, § 1. « lb., 28, § 5 ; 29, § 2.

^Caesar himself says {ib., 26, § 5) that about 130,000 survived the battle,
which would leave 144,000 to be accounted for. But of course a good many
of the Boi perished in the battle.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 241

persed, besides those of the Boi who were slain. This is an enormous
number. Stoffel finds no difficulty in believing that 143,000, includ-
ing old men, women, and children, were killed in the battle.^ I can
hardly share his faith and, if Caesar's figures are correct, I can only
;

assume that large numbers must have dispersed on the march before
the battle was fought. Professor C. Wachsmuth,^ however, regards
it as suspicious that the original number of the fighting men,
according to Caesar's statement, was exactly one-fourth of the
whole and he concludes that the latter was merely a rough estimate,
;

based upon the ascertained number of the men, whose individual

names were recorded.^ Of course, as the professor observes, Caesar
was not himself responsible for the calculations, and if the estimate
was too high, his good faith remains unimpaired.
There remains the possibility that some part of the host whoso
numbers were recorded in the schedule which was found in the
Helvetian encampment after the battle had, at the eleventh hour,
resolved to remain in Helvetia. At all events, the number of the
emigrants must have been very large advantages ; for with all his
in generalship, discipline, it was all
and superiority of weapons,
that Caesar could do to that, from
win the battle. I cannot believe
the mere desire of enhancing his own glory, he fabricated an exag-
gerated estimate of their numbers.

Caesar, says Eichheim,^ estimates the numbers of the confederate

Belgae, whom he encountered in 57 b. c, at 296,000, a gross ex- —
aggeration for a wooded and merely cattle-rearing country could
;

not have sustained such an enormous host.

Now it is not true that Caesar offers any estimate of the numbers of
the confederates he merely reports the estimate that was furnished
:

to him by the Remi ^ and it is probable that the numbers which

;

actually took the field fell short of those which the confederates
had promised to furnish. It is not true that the country of
the Belgae was merely cattle-rearing
'
and it is certain that the '
;

population of Gaul, considering the condition of the country, was

large.^ Still, I do not deny that there is some ground for suspecting

^ Guerre de Cesar et d^ Ariovisie, p. 77.

2 Klio, 1903, pp. 283, 287.
iii,
^ . quihus in tabulis nominatim ratio confecta erat, qui numerus domo exisset
. .

eorum qui armaferre possent, et item separatim (quot) pueri, &c. B. G., i, 29, § 1.
B. A. Miiller {Klio, ix, 1909, p. 73), relying upon modern statistics, concludes that
the percentage of fighting men would have been between 28-4 and 31. Assuming
that it was 30, the entire host, as the number of fighting men was 92,000, would
have been originally 307,000. I agree with J. Beloch [Rhein. Mus., N. F., liv,
1899, p. 423) that there is not the slightest reason for suspecting the accuracy
of the number 92,000.
^ Die Kampfe, &c., 1866,
pp. 71, 72, n. 1 ; 1876, p. 88.
' B.G.,ii,4.
« See pp. 340-3. Ihnc {Rom. Gesch., vi, 414, n. 2) remarks that if the 296,000
Belgae had boon in earnest, the light-armed troops whom Caesar sent to the
relief of Bibrax {B. G., ii, 7, §§ 1-2) could not have succeeded in their mission.
He means of course to imply that the Reman estimate of the Belgic force was
109U R
242 THE CREDIBILITY OF CAESAR'S NARRATIVE
that the numbers of the Bellovaci, the Suessiones, and the Nervii
may have been exaggerated for the contingent of the Bellovaci
;

would have represented a population almost as great as that which

now inhabits their country ^ and, moreover, as Beloch'-^ observes,
;

the contingents of the three tribes in question are said to have

amounted to more than half of the entire Belgic force, whereas their
territories were only about one-fourth of the whole. Allowance
must, however, be made for greater fertility and if the figures ;

are wrong, the Remi were responsible.^

General Lamarque ^ denies that the country of the Nervii, Atuatuci,

and Eburones, covered as it was with dense forests, could have
supported a population of half a million, which, he argues, is implied
])y Caesar's statement that the camp of Quintus Cicero was be-
leaguered by 60,000 men of those tribes ^ and he thinks it equally
;

incredible that such an army would have fled before Caesar's seven
thousand. He also denies ^ that the barren country of Les Landes
could have sustained 50,000 warriors, the host which, according —
to Caesar,"^ was encountered in the Third Campaign by Crassus.
In the case of half-barbarous tribes, the raising of a host of 60,000
men docs not necessarily imply the existence of a po2)ulation of half
a million ^ and the expression covered with forests is a gross
;
'
'

exaggeration. The host which Crassus encountered came from many

other places besides Les Landes.^ Still I have little doubt that the
numbers, in both cases, are far in excess of the truth. Caesar must
have derived them from the reports of Cicero and of Crassus respec-
tively, and, if they did not guess, they probably got their information
from hearsay ; for it is not likely that their enemies kept muster-rolls.^^

very miicli in excess of the truth. Still, the great size of the Belgic cncainp-
incni and the elaborate precautions which Caesar took to render his position
on the Aisne impregnable, prove that the Belgae were very numerous. How
then was Bibrax relieved V The explanation is not difhcult. Napoleon points
out that it would have been easy for the relieving force to enter the stronghold
(whicli he identifies with Vieux-Laon) on the south, as on this side the Belgae
could not have attacked it with any hope of success. The archers and slingera
could have effectively replied to the Belgic missiles. The Belgae had no
scientific methods of besieging a fortress and unless they could succeed
;

in driving the garrison from the wall by their missiles, they were helpless.
They were undisciplined and impatient to move on against Caesar their :

commissariat was ill organized and therefore thc}^ abandoned the attempt
;

to take the town. Or again, if Bibrax is to be identified not with Vieux-Laon

but with Beaurieux (see pp. 398-9), to which Napoleon's remarks would not
apply, it may be that, as M. Jullian suggests, the relieving force were able under
cover of night to penetrate the lines of the careless assailants. A modern his-
torian would have explained these things Caesar left them to the intelligence
:

of his readers. Ithein. Miis., 1899, p. 423.

^ ^ Ih.
^ A. Klotz {Caesarstudien, p. 101, n. 2) thinks that Caesar reproduced the
statements of the Remi in order to impress his readers.
* Spectateur mil., iii, 1827, pp. 263-4. ^ B. G., v, 49,
§§ 1-2.
^ Spectateur mil,, iii, 1827, ' B. G., iii, 26,
p. 267. § 6.
« Seei6., i, 29. » 76., iii, 26, § 6 ; 27.

^^ Col. G. F. R. Henderson {Stonewall Jackson, i,

1898, p. 158) says that
in the American Civil War Patterson rejjortod to his Government that he had
'

been opposed by 3,500 men, exactly ton times Jackson's actual number'.
 ;

THE CREDIBILITY OF CAESAR'S NARRATIVE 243

Commenting on the implied statement of Caesar ^ that nearly

430,000 persons, exclusive of the cavalry who took refuge with the
Sugambri, perished in the rout of the Usipetes and Tencteri, Des jar-
dins 2 observes toujours la meme exageration et la meme invraisem-
'

blance dans les chiilres des Commentaires. A nioins que les manuscrits,
au lieu de XXXXIII, ne portent par erreur CCCCXXX.'
Perhaps Desjardins would have been wiser if he had confined
himself to scepticism, and not attempted conjectural emendation.

According to a speech ^ which Caesar puts into the mouth of

Vercingetorix, 80,000 men, exclusive of the Mandubii, remained in
Alesia after the departure of the Gallic cavalry. The great Napoleon,
who, if he is not foully calumniated, lied so systematically himself
that he was not likely to give Caesar credit for disinterested accuracy,
gives professional reasons for rejecting his figures. He says that if
Vercingetorix' s force had been so strong, he would have sent out
60,000 men along with his cavalry, as the remaining 20,000 would
have been able to hold the fortress. The 60,000, he says, could have
harassed the besiegers and the provisions of the garrison would
;

have been virtually multiplied fourfold.*

It looks presumptuous to question the opinion of such an authority
but is it certain that the 60,000 would have been able to get out ?
The Due d'Aumale remarks that Vercingetorix could not have fore-
seen that even the cavalry would escape unobserved, although he
might reasonably hope that, if they encountered one of the Roman
piquets, their speed would enable them to get away. But the Due
d'Aumale himself gives reasons for believing that Caesar exaggerated
the forces of his enemy. He finds it hard to believe that Vercin-
getorix could have collected enough corn to feed 80,000 men, as well
as the Mandubii, for nearly two months. He also remarks that,
if he had had so large a force, he would have harassed the besieging

army far more than he did, and that, after the arrival of the relieving
army, he would not have been so foolish as to concentrate the whole
of his available force along that small section of the contra vallation
which crossed the plain, but would have made simultaneous sorties
at various points. He concludes that, after the manner of con-
querors, Caesar exaggerated the numbers of his enemy .^
None of these objections appears to be conclusive. Vercingetorix
had plenty of time to collect stores, and abundant resources to draw
upon. He did make frequent and furious sallies ^ but Roman :

discipline and Roman science kept him at bay and surely his only
;

chance of breaking the lines was to concentrate his strength. Anyhow

Veith'' has acutely seen that the passage in which Caesar relates
how he disposed of the captives after the fall of Alesia ^ confirms his
^ B. 6r., iv, 15, §§ 1-3
; 16, § 2. ^ Geogr. de la Gaule rom., ii, 652.

=>
B. G., vii, 71, § 3. Cf. 77, § 8. « Precis des guerrcs de Cesar, pp. 109-10.
5 Mev. des Deux Mondes, 2'' per., xv, 1858,
pp. 110-7.
^ Non niuuquaiu opera nostra Galli teiuptare at(pie cruptioneiu ox oppido

plui'ibuy portis Humiiia vi facero conabantur. B. G., vii, 73, § 1.

' Gesch. d. FeldziUje C. J. Caesars, p. 498. « B. G., vii, 89, § 5
; 90, § 3.
R 2
211 THE CREDIBILITY OF CAESAR'S NARRATIVE
estimate of the numbers of the garrison. He gave every man in his
force, which comprised ten legions, besides cavalry, archers, and
slingers, one prisoner by way of prize and from motives of policy
;

he restored 20,000 prisoners to the Aedui and Arverni. As his army

could hardly have numbered less than 40,000 men, 60,000 prisoners
are thus accounted for and the garrison had doubtless suffered
;

heavy loss.^
6. argues that Caesar's account of the predatory expedition
Ihne '^

of the Sugambri
in 53 b. c. is a perversion of the truth. Caesar,''

he says, represents that the invitation to the neighbouring peoples

'

to lay waste the land of the Eburones had reached the other side of
the Rhine. Thereupon two thousand Sugambri crossed the Rhine
in boats and took part in the plundering, and were then informed by
an Eburonian prisoner that the Roman camp contained much richer
booty and this it was that caused them to make the attack. This
;

account is highly improbable for the following reasons. If Caesar

needed seven days for the raid '
—
in the direction of the Scheldt
'
and the invitation to the neighbours of the Eburones preceded the
march of the Romans by one or two days, yet this time is not sufficient
to organize and carry out so considerable a campaign from the east
of the Rhine to the west of the Meuse. It must be assumed that the
Sugambri were already on the left bank of the Rhine when C-aesar
made his raid into the land of the Eburones. From this it follows
that their intention was some other than to plunder the Eburones.
Their attack on the camp and on the five cohorts outside the camp,
from which no booty was to be expected,^ has much rather the appear-
ance that it was made in secret understanding with the Eburones
than with the view of injuring them. If this was the case, it is a
proof that the Germans had been in no way intimidated by C^aesar's
inroad into their land, in other words that the campaign beyond the
Rhine had been without result. Caesar wished to hush this up, and
hence the misrepresentation.'
I reply, first, that the Sugambri, who, as Caesar says,^ were born
freebooters, no more needed a considerable time to organize a border
raid than did the moss-troopers of the Middle Ages.^ Secondly, to
assume that 2,000 undisciplined horsemen would have deliberately
crossed the Rhine with the purpose of attacking a Roman legion
entrenched within a fortified camp, is to assume what is both
groundless and absurd. Thirdly, if the Sugambri had come to
help the Eburones, Caesar would have detailed a force to destroy

^ —
It has also been argued that the area of the plateau of Alesia U7 hectares,
—
or about 240 acres was too small to accommodate 80,000 men ; but this
objection has been easily disposed of by General Creuly {Bev. orch.,no\i\. scr.,
viii, 1863, pp. 507-8) and A. dc Barthclemy {Rev. des quest, hid., iii, 1807,
p. 54). Cf. p. 586, infra.
- Ri'mi. Gesdi.. vi, 503, n. 1.

^ To say nothing of tlic corn, there were several hundred horses to be taken.

«eei)\(;.,vi,36, §3; 3U, § 1.

The rapidity with which the Boers mustered to repel

'^
Jamesons raid will be
fresh in the recollection of readers.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 245

them. Fourthly, nobody, except Ihne himself, ever suggested that

'
the attack on Cicero's camp was made with the view of injuring '

the Eburones it was made with the view of enriching the Sugambri.^
:

Fifthly, all the evidence we have goes to prove that the German
'
encroachments were emphatically checked by the terror which
'

Caesar's punishment of the Usipetes and Tencteri inspired.^ Lastly,

if, as Ihne maintains, C^aesar had failed to terrorize the Germans and

had wished to hush up his failure, surely his wisest course would
' '

have been to say nothing.

How, asks Eichheim, did Caesar learn that an Eburonian captive

guided the Sugambrian freebooters to Atuatuca, since not one of the
Sugambri fell into his hands, and the guide could scarcely have
'

^
come back to him for a tip.'
There no proof that none of the Sugambri or of their captives
is

fell into Caesar's hands but Eichheim's joke is not so bad

: and :

it may be that here, for once, he has hit a nail on the head. Caesar
may have assumed that the Sugambri must have been guided to
Atuatuca by some one, and if so, surely by a captive and the little ;

speech ^ which he puts into the mouth of the guide may have been
an invention. But if it was, such a trifle does not affect the general
trustworthiness of his narrative. Ancient writers were not scrupulous
about inventing speeches but Caesar, as one of his assailants
;

admits,^ allowed himself far less latitude in this respect than his
predecessors.
But I must say a good word for Eichheim before
take leave of him. I
Unlike Caesar's other assailants, the fellow has a sense of humour ;

and for this much may be forgiven him. Caesar, he says, must '

have had more hostages than soldiers (ilberhaupt miisste Cdsar '

meJir Geiseln als Soldaten qehabt hahen).^ Well, he certainly had a

good many, though it was not his habit to carry them about with
him and he could have had no conceivable motive for exaggerating
;

their number.
7. It has often been alleged that Caesar endeavoured to disguise his
repulse at Gergovia.'^ I have described this episode in detail on pages
155-8 of this book. It is enough to say here that Caesar, having
learned that Vercingetorix had sent a large body of Gauls to fortify
the western approach to the heights of RisoUes, which were linked by
a col to the south-western angle of the plateau of Gergovia, succeeded
in making him believe that he intended to attempt an attack on
that side, and when the Gallic encampment on the southern slope
of Gergovia was practically deserted, sent a column up that slope •

'
to effect a surprise '. At the same time the Aeduan auxiliary
'
6'.,vi,35,§§4-5.
J5.
2 Mommsen, Hist, of Rome, v, 1894, p. 61 {R<>m. Oesch., iii, 1889, p. 268).
=«
Die Kampfe, &c., 1866, p. 145, n. 1. * B. G., vi, 35, § 8.
5 See
p. 213. e
Di^ Kcimpfe, &c., 1866, p. 113, n. 1.
' See Napoleon's Hid. de Jules Cesar y ii, 281-2 W. C. Compton's Caesar's
;

Seventh Campaign in Gaul, pp. 89, 92-5, 97 ; W. Dnimann's Gescliichte Rows,

iii, 349 (312 of tlie new edition), &c.
246 THE CREDIBTLTTY OF CAESAR'S NARRATIVE
infantry were sent up the eastern slope, to create a diversion. The
point from which the Roman column started was rather less than
2,000 yards, in a direct line, from the wall of the towji and the ;

position which the Gauls were fortifying was about 5 furlongs from
the col. The column ascended the hill crossed a wall which ;

Vercingetorix had erected along the hill-side about half-way up ;

and captured three of the Gallic camps. Thereupon C^aesar, who was
with the ]Oth legion in reserve, sounded the recall (or retreat V)
but the troops, not hearing the sound of the trumpet and disregarding
the orders of their officers, pressed on, in pursuit of the few Gauls
who had been left in the camps, right up to the wall, from which,
after a fierce struggle, they were driven with heavy loss by the
Gauls who had hurried back from Risolles.
Mr. Compton, referring to the last sentence of B. G., vii, 43,
which, in his edition, runs ifse, maiorem Galliae motum exspectans,
:

ne ah omnibus civitatihus circmjisisteretur, consilia inihat, quemad-

7nodum a Gergovia discederet ac rursus omnem exercitum contraheret,
ne 'profectio nata a timore defectionis similis fugae videreiur, insists
that to make his failure appear part of a preconcerted plan, he
'

[Caesar] draws a fine distinction between profectio and fuga,'' &c.

Mr. C^ompton would not have made this charge if he had
—
adopted the true reading (ne profectio nata a timore defectionis
similis)5'we (fugae videretur) for in that case his remarks would
;

have lost any force that they may have had. Is there the least
sign of disingenuousness in this translation Anticipating that the :
— '

insurrection in Gaul would spread, and desiring to avoid being

surrounded by all the tribes, he began to think out a plan for with-
drawing from the neighbourhood of Gergovia and once more con-
centrating the whole army in such a way that his departure might
'
not be attributed to fear of a general defection and resemble a flight ?
Mr. Compton ought to have remembered that C^aesar uses a similar
phrase in four other passages with reference to the Belgae, the
Aremorican tribes, and Labienus.^
At the beginning of chapter 47, Caesar, having described the
capture of the three camps, writes, consecutus id quod animo pro-
posuerat, Caesar receptui cani iussit. Whereupon Mr. Compton says,
'
it cannot be seriously believed that Caesar never intended to do
more than capture three empty encampments ... is it to be credited
that the retreat was sounded before Caesar saw the hopelessness of
his undertaking ? If the signal was really given at this juncture,
it must have been because he already knew that the alarm would be
given and his attempt frustrated.' This criticism is substantially
identical with that of Napoleon ^ and Drumann,^ both of whom have
been answered by Long.^ If, he says, Caesar's design was only to '

surprise the camps ... he knew better than his critic whether he
gained any advantage by doing so '. He admits, however, in aiiother

' B. C, ii, II, § 1 ; V, 47, § 4 ; .53, § 7 ; vi, 7, § 8.

2 Hist, de Jules Cesar, ii, 281-2. ^ Gcsch. Boms, iii,340.
* Decline oj the Roman liepuhlic, iv, 319-21.
 THE CREDIBTLETY OF CAESAR'S NARRATIVE 247

passage,^ that Caesar's ultimate object

'
was to take the place '.

'
It seems probable,' he suggests, that he
'
had a further design . . .

to plaut his troops on the high ground immediately west of Gergovia,

and the first thing necessary was to disperse the enemy who were
below the walls on the south side of the city. . When Caesar had
. .

got possession of the site of the Gallic camps, he was very near to
the ridge at the west end of Gergovia, and there was, as far as we
can judge, a possibility that he might have seized this place, and kept
it against any attack.' ^ Not, surely, unless he pushed on without
delay ; for, as he himself says, the difficulty presented by the
unfavourable nature of the ground could only be overcome by speed,
and it was certain that the Gauls would hurry back to seize the col :

not, certainly, if, as Long says, he gave the signal for the legions
who had occupied the camps to return, and thus withdrew them to
a considerable distance from the ridge.
Let us examine Caesar's narrative dispassionately, and provisionally
accept his statements, (a) As he sent the Aedui to climb the moun-
tain by a different path on the right, in order to effect a diversion,
it seems clear that he intended to do something more than capture
the three camps if that had been his only object, the services of
:

the Aedui would not have been required. We may perhaps conclude
that he hoped either to take the town by a cowp de main, or to plant
his troops on the col, and thus to cut off the Gauls from the town.
The only possible alternative seems to be that he knew that it would
be impossible to seize either the town or the col before the Gauls
came to the rescue, and merely intended to engage Vercingetorix,
while the Aedui distracted his attention, on the upper hill-side.
(6) As he instructed his officers beforehand not to let the men advance
too far from over-eagerness for fighting or love of plunder, it is clear
that he did not wish them to advance, at least in the first instance,
as far as the town or the col : doubtless he intended that, as soon as
they had captured the encampment, they should push on at once
if there were then reason to believe that they were likely to succeed,

(c) As he gave the signal for recall when the troops had only captured

the three camps, it is difficult to resist the conclusion that he saw

that it would be impossible to capture the town by a couf de main
or to seize the col before the Gauls returned thither for surely the
;

only chance of gaining either of these objects would have been to

push on with all possible speed. ^ But on the other hand, by the
capture of the three camps he had gained a solid advantage, if only
he could follow it up it would perhaps have been impossible for the
:

Gauls to dislodge him from this position it is possible that, as Long-

;

maintains, he intended, when he sounded the recall, to form his men

again in order ; and, if he had then thought it advisable to advance
liigher up the slope, it would be rash to deny that he might have
worsted Vercingetorix in a fight under the wall of the town.

Caesar, p. 371.
' ^ See
pp. 156-7 of my narrative.
Unless the Gauls in the town kept a very bad look out, they must liave sent
^

warning to their comrades the moment they saw the legions ascending tlie hil].
248 THE CREDIBILITY OF CAESAR'S NARRATIVE
The truth is that everything turns upon the meaning of the words
consecutus id quod animo firoposuerat and receptui cani iussit. The
obvious meaning of tlie former expression is having achieved his '

purpose the latter may either mean

'
: ordered a retreat to be '

sounded or ordered a recall to be sounded '.^ Now, as I have

' '

already shown, it is certain that Caesar had not achieved the whole
of his purpose for the Aedui were employed for something more
;

important than the capture of the three camps. The question, then,
is whether, at the time when he receptui cani iussit, he had, as Long
suggests, so far succeeded as to have achieved the first step towards
the accomplishment of his real object, gained a point d^appui for
further operations. If so, receptui cani iussit means ordered a recall '

to be sounded ', Caesar's object being to re-form the legionaries,

who had scattered over the camps. That he should have expected
any one to believe that he had accomplished the whole of his purpose
when he had only captured three half-empty camps is hardly ad-
missible. But if he did already see the hopelessness of his under-
'

taking ', we can only suppose that he used the vague expression
consecutus id quod animo proposuerat in the hope that his readers,
like Long, might credit him with having intended to follow up his
trivial success. It may, however, be said that he had in some
measure achieved his purpose for it was probably better to try
;

and fail than to sneak away from Gergovia without trying at all.
In spite of their heavy loss the men did not lose heart when Vercin- :

getorix on the following day declined to come down and fight, they
were doubtless encouraged and Caesar knew how to fortify their
;

self-respect.2 But, honest or not, his narrative is certainly unsatis-

factory and it is a pity that he did not think fit to say exactly what
;

he intended to do.
Mr. Compton goes on to assert, or at least imply, that the legions
were only 100 yards from Caesar's trumpeter at the moment when he
sounded the recall. If,' he says, the advanced part of the army
' '

failed to hear the trumpet call from so short a distance, they must
have been out of sight in the hollow, not beyond it. Either " inter-
cedebat " is not accurate, or the signal was not actually sounded
as stated.' Apparently Mr. Compton forgets that, whether the
troops were out of sight in the hollow or beyond it ', they were
' ' '

on the further side of the wall which bounded the Gallic camps on
the south, and, according to his own map, von Kampen's, and
Napoleon's, at least 400 yards from Caesar's legion.
Finally, Mr. Compton insinuates that Caesar deliberately under-
stated his losses. The number of officers killed,' he says, (one out
' '

of every four) would be quite out of proportion to the whole loss

' See Mr. C. E. Moberly's note {Caesar, pp. 321-2) Bell. Alex., 47, § 1 ;
;

Bell. Aft:, 40, §5 and Livy, xxx, 34, § 11. It has been asserted that the
;

words niean 'ordered a half to be sounded' but this interpretation is in-

;

consistent with the meaning of receptus. See B. C, iii, 45, §§ 4-5 40, §§ 1-4. ;

" See
Professor Spenser U'ilkinson's remarks in the Mornimj Post, Jan. 7,
1904, p. 2, col. 2.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 249

given below at 700.' But in the first place it is incredible that Caesar,
having stated the losses of his officers truly, should then, by a pitiful
half-lie, have minimized the casualties among the rank and file ;

and secondly, as every student of military history knows, the loss

of officers in a hard-fought battle is often out of all proportion to
the loss of privates.^
8. C\aesar says that, before the blockade of Alesia, he raised cavalry
from the German tribes which he had reduced to submission in
'

former years '.^ Numerous critics have insisted that this is an

intentional exaggeration for, they say, Caesar tells us himself in
;

another passage ^ that of all the Transrhenane peoples the Ubii alone
had submitted to him. But the critics forget that in a subsequent
chapter* Caesar states that, when he was marching to punish the
Sugambri, several tribes made their submission.
9. In a long-winded and carping dissertation,^ the tediousness of
which is only relieved by one passage in which, Phormio-like, he takes
Caesar to task for bad generalship, Otto SumpfE accuses him of
having treated the services of his legati, in the Commentaries, in a
manner equally ungenerous and unjust. The only generals, he says,
whom Caesar does not blame are the two who died during the war,^
young Crassus and Cotta. The former was no longer to be feared
as a rival the latter was praised simply in order to heighten the
:

discredit which Caesar fastened upon his unfortunate colleague,

Sabinus."^
SumpfE's inaccuracy is astounding. Two of the lieutenants and
two only ^ are blamed and one of the two is let off with a reproof
;

so gentle that it can hardly be called blame. For while the facts
show that Cicero wholly failed in his duty at Atuatuca, Caesar merely
tells him that he ought not to have let any troops leave the camp,
and does his best to exonerate him by throwing the larger share of
the blame upon fortune.^ Sabinus is certainly blamed for his conduct
at Atuatuca but no lieutenant who was responsible for a great
;

disaster was ever less harshly spoken of by his chief. Caesar's actual
comment upon his behaviour limits itself to this, that he was want- —
ing in foresight that he lost his head and showed nervous trepidation
;

in the action and that the disaster was caused by his rashness.^^
;

The bare recital of the facts was his sufficient condemnation and ;

in Caesar's language there is not a trace of resentment. It must

* Mr. Compton ought to have remembered that in the battle of PharsaHa
Caesar lost not more than 200 privates and yet about 30 centurions, and that
he is careful to emphasize the disproportion. B. C, iii. 99, § 1.
B. G., vii, 65, § 4.
2 ^ lb., iv,
16, § 5. lb., 18, § 2. ->

Caesars Beurteilung seiner Offiziere in den Cornmentarien vom gallischen

^

Kriege, z welter Teil, 1893. I have not been able to obtain the first part, which
I do not regret.
« Die zwei Offiziere ohne Tadel sind zwei Tote, &c. Op. ciL, p. 33.
' lb.
* Or three, if it was blaming Galba to say that he had not made sufficient

provision for his commissariat, because, as the tribes of the Valais had sur-
rendered and given hostages, he did not expect that they would fight. B. O.,
iii, 3, § 1.
« lb., vi, 42, §§ 1-2. 10
76,^ V, 33, § 1 ; 52, § 6.
 250 THE CREDIBILITY OF CAESAR'S NARRATIVE
be remembered, moreover, that when Caesar was describing Sabinus's
earlier campaign in Normandy, he was careful to imply his approval
of the cautious tactics which he had adopted,^ and thereby to supply
an antidote, as it were, to any prejudiced judgement which might
arise in the mind of the reader. If Caesar deviated from the line
of strict impartiality in narrating the actions of his lieutenants, it
was on the side of lenity. But of all those who served under him, the
one to whom the greatest compliment is paid is Labienus. Even
Georg Mezger, one of the sanest critics of the Commentaries, cannot
forgive Caesar for the cold and businesslike tone of the narrative
' '

of Labienus's campaign against the Parisii.^ If by cold Mezger '

'

means restrained, I accept the epithet. Caesar never suffers his

enthusiasm to escape him but it communicates its force to us,
;

if our minds are attuned to his. Whenever he mentions Labienus,

he makes us feel that Labienus stood in a class by himself, the —
greatest of his lieutenants and one of the great marshals of history.
But it is precisely in his narrative of this campaign that he impresses
the features of the man's character most deeply upon the imagination.
In four terse sentences he puts before us the difficulties, all but over-
whelming, by which Labienus was beset. And then he says, With '

these formidable difliculties suddenly confronting him, he saw that

he must look for aid to the force of his own character (tantis suhito '

difficuUatihus ohiectis, ah animi virtute auxilium petendutn viclehat)?

We are made to feel that the character will be equal to the strain ;

and the next three chapters show us that it was. Could any finer
compliment be desired by a soldier from his chief, the more telling —
because it is conveyed not by praise but by suggestion ? ^

VII
The protagonist in the destructive criticism of the Commentaries
was General Warnery. For the most part his work may safely be
ignored for, as Long says,^ the foundation of nearly all his comment
;
'

is a misunderstanding of the text.' He was hardly capable of con-

struing easy Latin. But there were moments when he could discern
the obvious. He denies that 120 towers could have been required
for the defence of Cicero's camp when it was beleaguered by the
Nervii and their allies and I believe that this objection is valid.
;
^'

1 B. G., iii, 17, § 7.

- Ueber d. Abfassungszeit von Ccisar^s Comm. iiber d. gall. Krieg, 1874-5, p. 18.
^ B.G., vii, 59, § 6.
* Read what Caesar says of his great enemy after the fall of Avaricum
also :

'
And while a reverse weakens the authority of commanders in general,
so,
his prestige, on the contrary, in consequence of the disaster, waxed daily greater
[ilaque id reliquorum hnperatoritrn res adversae aucioritatein ruinuunt, sic huius ex
rontrario dignUas incommodo acceplo in dies augebatur [vii, 130, §3]). Yet Professor
Tyrrell [The Correspondence of Cicero, vol. iv, p. xlvii) tinds it amazing that he '

seems to have completely failed to recognize the nobleness of Vercingetorix '.

Who, then, enabled the professor to recognize it ?
° Caesar, p. xiv.

Melanges de renHirques, surlovl sur Cesar, 17S2, pp. 12-3.

•^
1
 )

THE CREDIBFLTTY OF CAESAR'S NARRATIVE 251

Assuming that the towers were 80 feet apart, as at Alesia,i the peri-
meter of the camp, exchisive of the space required by the towers
themselves, would have been 119 x 80, or 9,480 feet and allowing ;

for the space occupied by the towers, it could not have been less
than two miles. The camp then would have covered an area of
]60 acres, which is much too large for a single legion. But I see no
reason to question Caesar's good faith. To exaggerate the number
of the towers which the legionaries erected could hardly have en-
hanced the glory of the defence. I suspect then that the error,
if there is one, is due to a copyist or to Cicero.'^

VIII
A who undertakes to measure the credibility of the Gallic
writer
TFar bound to ask himself whether the Civil War does not throw
is
light upon his subject. Now German critics, followed respectfully
by one or two in this country, have for some years been busy picking
holes in Caesar's later work.
1. In B. C, i, 6, §§ 7-8, Caesar says that, in 49 b. c, the newly

appointed provincial governors, after performing the usual religious

ceremonies in the Capitol, left Rome, wearing the dress of a military
commander, although the legal form of giving them the power to
command troops had been omitted and that, contrary to all
;

precedent, the consuls left the city although, being without this
power, they were really only private individuals (neque exspectant
[quod superiorihus annis acciderat,] ut de eorum imperio ad populum
feratur, paludatique votis nuncupatis exeunt. Consules [, quod ante
id tempus accidit numquam,] ex urbe proficiscuntur, lictoresque hahent
in urbe et Capitolio privati contra omnia vetustatis exempla). I follow
the reading of the MSS. but the words enclosed in square brackets
;

are rejected by Nipperdey ^ as spurious. Mr. Peskett,* following

a conjecture of Davies, but omitting to say that it is a conjecture,
prints quod numquam in the first sentence, immediately after
. . .

exeunt and at the same time he only half follows Davies, for Davies,
;

without any authority changed exeunt into the subjunctive, exeant.

Mr. Peskett says that the statement quod ante id tempus accidit
numquam is so demonstrably and even ludicrously untrue that Voss,
'

Nipperdey and others would eject the words altogether. But Caesar
... no doubt relied on his readers having short memories, and I do
not see why he should be less likely to make a false statement, if

^ If, as A. Klotz argues (see p. 810, n. 5, infra), passus ought to be substituted

in B. G., vii, 81, § 5, for pedes, the towers at Alesia were 400 feet apart.
^ M. Jullian {Hist, de la Gaule, iii, 385, n.
4) accepts the number CXX,
remarking that the towers were sans doute tres rapprochees et reunies par des
'

ponts (cf. B. G., viii, 9, §§ 3-4). But M. Jullian supposes {op. cit., p. 386, n. 1
'

that the perimeter of the camp was 1,730 metres. The interval between any
two towers, then, would have been only 14 metres, or about 46 feet, minus the
breadth of the tower. Where did the men who stood on tlie rampart {B. G.y
V, 39, § 3 ; 43, § 4) find room ?
* (hipsar, pp. 132-3. Meiisel puts quod . . . numquam. after proficiscuntur.
* B. 6'., i, ed. A. G. Peskett, 1890, p. 60.
252 THE CREDIBILITY OF CAESAR'S NARRATIVE
itsuited his purpose to do so, than a modern Christian statesman.'
Nor do I but I find it difficult to believe that a marvellously adroit
;

and worldly-wise statesman would have made a statement which

was ludicrously untrue '.
'
Before the time of Sulla,' continues
'

Mr. Peskett, it had been the regular thing for the consuls to leave
'

the city during their term of office, and although in 81 the lex Cornelia
de ffovinciis enacted that they should not leave it till the expiration
of their year, yet between that date and 49 there had been five or
six instances of the rule being contravened. Caesar,- however, for his
own purposes chooses to ignore these.' To which Dr. J. S. Reid
appends this little note ^
—
It was the going in and out of the city
: '

and yet retaining the imferium which Caesar declared unparalleled.

He would never have said that breaking a usage which had only
existed since Sulla was contra o?nni(t vetustatis exempla. By this
recrossing of the pomerium, after leaving it in military array, they
became frivati,^
Has not Mr. Peskett discovered
a mare's nest ? The text of the
Civil War is, as he himself remarks,^ in many places corrupt. If
the words quod ante id tenipus accidit numquam were not inter-
polated, and if they were written either before or after ex urhe
froficiscuntur, Caesar was evidently thinking of Sulla's enactment,
and either forgot or ignored the exceptional cases in which the
Senate had disregarded it. This, I freely admit, was not what an
impartial historian would have done.^ If, on the other hand, quod . . .

numquam is arbitrarily inserted in the preceding sentence, after

exeunt, Caesar is made to write like a madman, unless the words
are taken as referring to his charge that the magistrates had not had
the imperium conferred upon them, or unless, as B. Kiibler con-
jectures,* Caesar wrote non nuncupatis and in either of these cases
;

the other charge against him disappears. He had a right to say that,
by going in and out of the city without having had the imperium
conferred upon them, or without the observance of the usual religious
rites, the consuls were acting contrary to all precedent.^
2. 0. E. Schmidt ^ accuses Caesar of having distorted the sequence
of the events which he narrates in B. C, i, 8-11, in order to make it
appear that he did not resort to force until Pompey had made it

'
C, i, ed. A. G. Peskett, p. 60.
B. ^
j^^ ^ 47^
Meusel in his edition of the Civil War (p. 23, note) points out that consuls
^

had only been permitted [or enjoined] by the Senate to leave the city in
exceptional circumstances. See Cicero, Farn., viii, 10, § 2,
* Fhilologus, Iv, 1896,
pp. 157-8. Oudendorp, as Kiibler remarks, had antici-
pated this conjecture. Kiibler quotes a parallel passage from Cicero, Phil., v, 9,
§ 24, Post autem, neque sacrificiis solennibus factis, neque votis nuncupatis, non
profectus est, sed profug it paludatus. Caesar's complaint would have lacked
point, if he had said that the magistrates had fulfilled their religious obligations.
See Nipperdey, pp. 131-4, Kiibler, in his edition of the Civil War (p. xvi),
'"

abandons the passage in despair as miserably corrupt '. Mommsen, he says,

'

has communicated to him the following conjecture, which perhaps substantially

represents what Caesar wrote Consules, quod ante id tempus accidit numquam,
:

anlequam ex urbe proficiscuntur, Uctores habent in vrbe et Capiiolio paludatos,

contra omnia vetustatis exempla.
« Rliein. Mus., N.R, xlvii, 1892, p. 263, n. 2.
 ;

THE CREDIBILITY OF CAESAR'S NARRATIVE 253

evident that he would not consent to any reasonable accommodation.

In chapter 10 we read that Caesar's envoy, Roscius, delivered his
ultimatum to Pompey and the consuls, and brought back Pompey's
answer to Caesar, who was at Ariminum. In chapter 11 Caesar
states his objections to Pompey's proposed terms, and goes on to
say that he sent Mark Antony to occupy Arretium, and three
cohorts to occupy Pisaurum, Fanum, and Ancona respectively.
Now Pompey's letter did not reach Caesar before January 29, 49 B. c.
and we know from letters of Cicero ^ that the news of the capture
of Arretium, Pisaurum, Fanum, and Ancona had reached Rome on
or before January 18. The facts are certain. Either Caesar made
a mistake from haste or lapse of memory, as Stoffel assumes,'^ or he
lied if you like, told
; a monstrous lie {ungeheuerliche Liige), as
' '

Schmidt puts it. Schmidt ^ points out, in reply to Stoffel, that Caesar
mentions Ariminum three times ^ and he insists that no one who ;

does not wilfully close his eyes can be blind to the elaborate art with
which he told his story. Those who hold with Stoffel that he would
have amended this portion of his work if he had had time to revise it,
might argue that his case was too strong in itself to require embellish-
ment. But it may be that anxiety to exhibit his own conduct in the
most favourable light led him to represent the occupation of Arretium
and the other towns as the consequence of Pompey's stubbornness,
whereas it was really a justifiable measure of precaution.
3. Caesar says that when he was at Brundisium he was greatly
astonished that a certain Magius, whom he had sent to Pompey with
overtures for peace, was not sent back to him that he accordingly ;

sent Caninius Rebilus to Scribonius Libo on a similar errand but ;

that Libo sent back word that Pompey could not entertain any pro-
posals for peace in the absence of the consuls.^ But a letter of Caesar's
to Cicero is extant, in which he says, I reached Brundisium. on the '

9th of March. Pompey is at Brundisium. He has sent N. Magius

to me to treat of peace. I have replied as I thought fit. As soon . . .

as I see a chance of being able to come to a settlement, I will let you

know at once.' ^ F. Eyssenhardt regards this letter as a proof that,
in the Civil War, Caesar told a deliberate falsehood with the intention
of persuading his readers that he had been honestly anxious to avoid
bloodshed, but that his efforts had been frustrated by the obstinacy of
Pompey.'^
Long has written a note on this question, which is worth consider-
ing. Caesar,' he says,
'
arrived before Brundisium on the 9th of
'

March, and Pompeius then sent Magius to him. Caesar, as he says

in the letter, returned an answer to Pompeius by Magius. Now when
Caesar (B, C, i, 24) let Magius go and gave him a message to Pom-
peius, he was on his march to Brundisium, and, as far as we are told,

1
Alt., vii, 14, § 1 ; ix. 10, § 4 ; Fa7n., xvi, 12, § 2.
''
Guerre civile, i, 213.
^
Der Briefwechsel des M. T. Cicero, 1893, p. 382.
*
i?. 6\, i, 10, §3; 11, §§1,4. ' i6.,i,26,§§2-6.
^
Cicero, Ait., ix, 13 a, § 1.
'
NeiLe Jahrh, f. Philoloyie, &c., Ixxxv, 18G2, pp. 703-4.
 254 THE CREDIBILITY OF CAESAR'S NARRATIVE
he received no answer until he had pitched his camp before Brun-
disium, when Magius came with the message to which alludes C
in his letter. C. gave his answer to this message, but he had no reply.
This may be the true explanation, and there is no contradiction.' ^
Schmidt, who does not in this case charge Caesar with misrepre-
sentation, says A few days after Caesar's arrival at Brundisiuni,
'

about March 12, Pompey sent Magius to Caesar, plainly only to gain
time for his preparation for sailing. Probably Caesar requited
. . .

a personal interview, but his opponent was too proud to grant it.'
''^

I am prepared to admit that Caesar's narrative of the civil war

may be in certain respects disingenuous. If it is, we are justified
in saying that he did not love truth with entire devotion, which I, —
for one, have never for a moment supposed that he did but we are;

not justified in throwing doubt upon the honesty of his narrative

of the Gallic war, unless it can be shown that he had an adequate
motive for misrepresentation. Furthermore, I believe that any
competent critic who reads the Civil War rapidly through for it is —
impossible to judge it fairly piecemeal will form a high opinion—
not only of the magnanimity, but also of the general trustworthiness
of the writer.

IX
There is one fact in the history of Caesarian criticism which ajipears
to me not without significance. So far as I am aware, no great writer,
no great historian, no great statesman or general has ever thrown
"
serious discredit upon the Commentaries. It is one thing to say that
Caesar exaggerated the numbers of his enemies, that his narrative
was not so impartial as that of a disinterested historian, and that he
may have concealed facts which he thought it imprudent to disclose :

it is quite another thing to charge him with a systematic mendacity

that would deprive his memoirs of all historic value. To do this has
been reserved for the Ihnes, the Eichheims, and the Rauchensteins :

the Montaignes, the Mommsens, the Napoleons are satisfied of Caesar's

veracity.
But after all, if we may believe— with the necessary reservation
on occasional points of detail which we must make, in a greater or
less
degree, in reading every historian —the particular statements of
definite fact which he makes,^ as distinguished from his statements of
motives and from the general tendency of his narrative, Caesar's
standard of literary ethics is, for the present discussion, a matter of
indifference. If any one believes that he made a clumsy effort to
delude us into the belief that he was reluctant to conquer Gaul, we
are not deluded and, whatever we may think of his sincerity, the
;

^ Decline of the Boman Republic, v, 33, n. 2.

- Rheiyi. Mus., N. Schmidt's Der Briejwcchsel des
F., xlvii, 1892, p. 267. Cf.
M. T. Cicero, p. 152, where, however, he remarks that B. C, i. 20 'contains
at least a gross inaccuracy '. I woukl ask him to read Long's note.
^ M. Jiillian, in a paragraph which is well worth reading
{Rev. des eludes one.,
X, l'J08, p. 355), has called attention to various incidents in the recent French
campaign in Morocco analogous to some recorded by Caesar the truth of which
has been gainsaid.
 THE CREDIBILITY OF CAESAR'S NARRATIVE 255

cause of history does not suffer. If he meant to suggest what was

false, he did it so clumsily that, even if he succeeded in deceiving his
contemporaries, he has utterly failed, so they tell us, to deceive his
modern critics. If, without saying what was untrue, he suppressed
what was true, the suppression could hardly have been of such a kind
as to affect our estimate of his conduct ;for he said quite enough
to shock modern humanitarians and modern professors of political
morality ; and he said it without a blush.
But the reader must not run away with the idea that I am so sim2)le
as to regard the Commentaries as absolutely true. No history is
absolutely true and Caesar assuredly made mistakes. He is often
;

laconic to a fault he often writes with a looseness of expression

:

which was natural in a busy man who did not write for cavillers,
who made large demands upon the intelligence of his readers, and
who, moreover, had not the fear of modern critics before his mind :

he was sometimes either uncritical or careless in reproducing the state-

ments of his lieutenants writing as a politician, not as a historian,
:

he may have thought it discreet to withhold valuable and interesting

information he doubtless exaggerated, consciously or unconsciously,
:

the numbers of his enemies and the losses which he or his lieutenants
had inflicted upon them he may have glossed OA^er a mistake or
:

two he may have concocted a partial narrative of the one defeat

;

which he himself sustained and I am willing to believe that his

;

memoirs leave upon the mind an impression of his prowess, if not

of his character, more favourable than would have been produced
by the narrative of an impartial and well-informed historian. I am
also willing to believe that, if he had had a solid political object
to gain, he would have had recourse, as we are told that Bismarck
had recourse, to brazen mendacity. Mendacity is a weapon which,
in this wicked world, no statesman can afford to despise. I do
not claim for Caesar that he had the passion for truth that inspired
Mr. Rawson Gardiner. Even Mr. Baring Gould would hardly
maintain that if Caesar could have armed himself for his duel with
Pompey by garbling the history of the Gallic war, he would have
resisted the temptation. Only the temptation was not there. On
the whole, he could afford to tell the truth. He did full justice to
his lieutenants :he wrote generously of his enemies ; and I see no
reason for believing that he was ashamed of anything that he had
done. But, as it is quite certain that my criticism is imperfect,
I will quote the Avords with which a large-minded scholar, a man
of arms, of affairs, and of the world, concludes a brief discussion
of the same subject. Nous n'avons cherche,' says the Due d'Aumale,
'

'
a etablir qu'une chose, c'est que, sans etre nullement detracteur de
Cesar, en restant admirateur declare, non seulement de ses grandes
actions, mais de la f a9on dont il les raconte, il etait permis de soumettre
a I'analyse certains passages de ses Commentaires et de discuter
quelques-unes de ses assertions, alors qu'il avait un interet evident a
dissimuler ou a exagerer la verite mais nous ne croyons pas que cette
;

facultc puisse s'etendrc jusqu'a changer le caractere de I'inter-

pretation rigoureusc du texte dans tout cc qui regarde les descriptions
256 THE CREDIBILITY OF CAESAR'S NARRATIVE
de lieux,d'ouvrages, de mouvemens.' ^ On ne peut contester,'
'

affirms the same critic, que ses recits rcspirent la sincerite.' ^

'

Finally Montaigne, in a note written on the margin of his copy of

the Commentaries, calls the author le plus net, le plus disert, et le
'

plus sincere historien qui fut jamais.' ^ Perhaps we shall hit the
exact truth if we add the comment of the Due d'Aumale le plus
— '

sincere de ceux qui ont ecrit leur propre histoire.' *

—
Note. The foregoing essay lias dealt simply with the credibility of Caesar's
narrative. The credibility of some of his statements about Druidism and the
status of the Gallic plebs, in regard to which it has been alleged that he erred
from defective information, will be dealt with in subsequent articles. Sec
pp. 523-41.
'
Jiev. des Deux Mondes, 2" per., xv, 1858, p. 121.
- //;., p. HI). ' lb.,
p. 118. 1 lb.
 SECTION II,— THE PEOPLE
THE ETHNOLOGY OF GAUL
I

INTRODUCTION
Gaul, taken as a whole, is divided into three parts, one of which
'

is inhabited by the Belgae, another by the Aquitani, and the third

by a people who call themselves Celts, and whom we call Gauls.
These peoples differ from one another in language, institutions, and
laws. The Gauls are separated from the Aquitani by the Garonne,
from the Belgae by the Marne and the Seine.' With these words
Caesar begins his narrative of the Gallic war and every schoolboy,
;

not only in Macaulay's but in a literal sense, is familiar with them.

But Caesar was not an ethnologist. It was not his business to ask
himself or to inform his readers whether the Belgae, the Celtae, and
the Aquitani were homogeneous groups, or whether each group was
an aggregate or a mixture of different races. We all know that the
populations of England and Scotland, of Ireland and Wales, are each
composed of various races, not one of which has remained pure ;

but for the purposes of daily life we speak of Englishmen and Scotch-
men and Irish and Welsh.
Nevertheless, the problems which Caesar had no call to solve, and
indeed no means of solving, are full of interest and a modern
;

historian of the conquest of Gaul is bound to grapple with them.

At the very outset of our inquiry we are confronted by a difficulty.
—
Caesar tells us that the Gauls and by the Gauls he means the Belgae
—
and the Celtae were big men.^ Every other ancient writer who
describes the physical features of the Gauls says the same and;

some add that they had fair hair and blue eyes. But when we travel
in France, we find that tall fair-haired men, though not uncommon in
the north and north-east, are everywhere in a minority and that, in
;

certain districts, nearly every one is short and dark. Anthropologists

assure us that the contrast between the physical features which
we see and those of which we read cannot be ascribed, except in
a limited sense, to the influence of climate or of physical environ-
ment.2 History tells us that, since Caesar's time, the majority of

1 B. G., ii, 30, § 4.

M. Topinard expresses the orthodox opinion when he says that, generally
^

speaking, les grandes tailles sont la consequence d'une heredite de race, qui
'

pent etre legerement modifiee par les milieux.' Stature is, as M. Mondiere
has shown, undoubtedly influenced to some extent by social conditions: the
1093 S
258 THE ETHNOLOGY OF GAUL
the invaders and immigrants who have settled in those parts of
France which were inhabited by the Belgae and the Celtae have
belonged to the Teutonic race, of which the prevailing type was,
like that of the Gauls of history, tall and fair. The conclusion is
irresistible. Besides the Gallic warriors, who seemed so tall to the
sturdy little legionaries, and whose red hair and fair skins attracted
the attention of travellers who belonged to a swarthier stock, there
must have been a short dark people, who passed comparatively
unnoticed and a third type had doubtless already developed itself
;

from intermarriage between the two. These main conclusions, as we

shall presently see, history, archaeology, and anthropology alike
confirm.
Nevertheless, it is certain that, in Caesar's time, the tall fair men
were more numerous in proportion to the whole population than
they are at present. No modern invader could possibly describe
the population of France as he and other ancient writers described the
population of Gaul. We should expect, then, to find not only
that the tall fair inhabitants were proportionally more numerous
than they are now, but that, instead of being scattered among
people of different types, they were, so to speak, grouped in masses,
so as to leave a sharper impression upon the mind; and it may perhaps
be inferred that the people of whom Caesar took special note, and
whose physical features he described, were those with whom he was
chiefly thrown in contact, that is to say the warrior class. On the
other hand, it is not necessary to suppose that the armies which he
encountered were composed exclusively, or even principally, of tall
fair men ; for the experience of anthropologists has shown that,
when untrained observers enter a strange country, they take special
note of the individuals whose physical features leave the deepest
impression upon their minds, and ignore the rest.^ Thus a modern
English traveller hastily remarks that Frenchmen are dark, that
Germans are fair, and that Scotsmen have high cheek-bones and red
beards ; while a trained observer, having carefully written down all
the observations that he has been able to make, reports that a con-
siderable proportion of Frenchmen are fair, that the fair Germans
are in a minority, ^ and that in certain districts the majority of Scots-
men are dark. But how are we to account for the fact that the
proportion of tall fair men among the population of ancient Gaul
was greater than among the population of modern France ? First

average height of males of tlie well-to-do classes in America was said, in 1883,
to be 5 feet 91 inches, of the proletariat 5 feet 6^ inclies. Did. des sc. antlir.,
1883, p. 1035. See also Mem. de la Soc. d'anthr. de Paris, ii, 1865, pp. 232-3;
ib., 3*' ser., i, 1895, p. 89 -Rev. mensuelle de V Ecole d'anihr., 189G, pp. 51-6
; ;

U Anthr., xi, 1900, p. 700 Middlesex Hospital Journal, xii, 1908, p. 43

; ; and
Association franc, pour V avancement des sc, 35*" sess"., 1906, pp. 683-90.
^Journ. Anthr. Inst., ii, 1873, p. 19. Cf. J. Beddoe, The Races of Britain,
1885, p. 244, and J. T. Thurnam and J. B. Davis, Crania Britannica, 1865,
pp. 163, 207.
- Virchow reports that of 5,000,000 German school-children, 33 per cent were
found to be blonde, 13 per cent daik, and 54 per cent of mixed type. Congres
inlernat. d' anthr. et d'arch. prehist., 1876 (1877), p. 579.
 THE ETHNOLOGY OF GAUL 259

of must be remembered that a great many Gauls perished in

all, it

Caesar's wars or were sold into slavery and it is probable that of

;

those who were thus lost to the country a number disproportionately

large belonged to the race by whose great stature he was so impressed.
Secondly, it is certain that, except in cold and comparatively cold
climates, the tall fair typeis less successful in the struggle for existence

than the dark. Thirdly, it is believed by Dr. Beddoe that the tall
fair type is less able than the dark to resist the unsanitary conditions
of crowded cities. Fourthly, it is suggested by the same authority that ^| -
the proportion of fair men in France, as well as in other countries
of Europe, may have been diminished by the influence of what he
calls conjugal selection in other words, his researches have led
;

him to believe that the proportion of dark women who are married
is greater than the proportion of fair women. Fifthly, in families S —
of which one parent is dark and the other fair, the proportion of
dark children is generally greater than the proportion of fair.^ In
a word, there can be little doubt that, since the time of Caesar, the
dark type in the country which was once called Gaul has been steadily
gaining ground upon the fair.
The questions which I propose to ask myself are these First, :
—
what were the ethnic elements of the prehistoric population of Gaul ?

Secondly, what was the physical type of the Ligurians, and what
portion of Gaul did they occupy ? Thirdly, whether the Iberians
were, in any sense of the word, a race if so, whether they are now
:

represented by the Basques and what portion of Gaul they occu-

;

pied ? Fourthly, who were the Aquitani ? Fifthly, who were the
Celtae ? Sixthly, who were the Belgae ? Lastly, were the Gauls,
—
properly so called, whether Belgic or Celtic the tall fair conquerors,
whose physical features are described by ancient historians and
—
geographers ethnically identical with or akin to the Germans ?
The subject of the ethnology of Gaul is one of great difficulty. The
literature is vast in amount, and the greater portion is scattered in the
periodical publications of learned societies. In spite of the enormous
labour which has been expended in collecting facts, the facts are
insufficient ;and regarding the conclusions which are to be drawn
from them, there is much difference of opinion. But this is not all.
The student is constantly encountering the word Celt and this
'
' ;

word is used in so many different senses that he is obliged to exercise

unflagging vigilance to avoid being misled. Before he began his
studies, he had taken for granted that Irishmen and Welshmen and
Scottish Highlanders were Celts. As his reading becomes more
extensive, he learns that Celtic dialects are spoken by peoples of
divers physical types that the word
; Celt ', as used by certain
'

ancient writers, was hardly less vague than American or Indian

' ' ' '
;

that Polybius and others used it as a synonym for Gaul that

'
' ;

Caesar, while remarking that the people of Transalpine Gaul who

Bull, de la Soc. d'anthr. de Paris, ii, 1861, p. 408 ; Anthr. Rev., i, 1863,
'

pp. 311-2; Did. encycl. des sciences medicales, xiii, 1873, p. 766; Scottish
Review, xix, 1892, p. 418; V
Anthr., vi, 1900, pp. 702-3 Journ. Anthr. Inst.,
;

XXXV, 1905, p. 237.

S 2
260 THE ETHNOLOGY OF GAUL
called themselves Celts
'
were called by the Romans
' Gauls ',
'

applied the former term only to a section of the population of Gaul,

and not to the Belgae, who also spoke a Celtic language, and whom
he also called Gauls and that, even after his time, Dion Cassius spoke
;

of the Germans as Celts. Next he finds that, according to certain

French ethnologists, there have never been any Celts in the British
Isles ; that the word Celtic ', as applied to the dialects in question,
'

is a misnomer and that Caesar's Celtae, the people who lived

;

between the Seine and the Garonne and were not a pure race at all,
were the only true Celts. Finally, he is warned to purge his mind
of the notion that there has ever been, at least within historic times,
a pure Celtic race, in the biological sense of the word race ', or indeed
'

any other pure race at all and he is solemnly assured that the
;
'

word " Kelt " has long ceased to have any ethnical significance 'J
But if the perplexed student has the patience to reconsider these
theories without bias, he will gradually discover that there is more
of agreement among them than he had at first supposed, and that
the disagreement is about words as much as about things and if ;

he has the courage to collect and sift all the facts that have been
discovered, he will find that the data, though not sufficient for the
solution of all the problems which confront him, are yet sufficient
to establish important conclusions.
But the problems cannot even be approached until one has collated
evidence drawn from every available source. Historians have mar-
shalled every ancient text which bears upon the subject, and have
offered solutions which Celtic scholars and anthropologists have
contemptuously rejected. Celtic scholars have propounded theories
which other Celtic scholars have refused to accept. Anthropologists,
travelling, notebook in hand, from department to department and
from town to town, climbing mountains and descending into plains,
have jotted down and tabulated the physical characteristics of the
unheeding individuals whom they passed others have explored
:

caves and groped in barrows, collected skulls, measured them, and

referred them to various types others again have recorded the
;

stature and examined the heads of groups of living men and all ;

these observers have been told that their data are insufficient and
their generalizations premature. The inquirer who is resolved to
succeed will press into his service every science that bears upon his
subject. He willbear in mind that the historians and geographers
of antiquity were that some of them wrote at second-hand,
fallible ;

and that none were trained in the exact methods of a scientific age ;

that the philological conclusions of one decade are often questioned,

if not disproved, in the next and that physical anthropology is a
;

science which has hardly outgrown its infancy. On the other hand, he
will remember that scepticism has its dangers as well as credulity :

he will find that, side by side with much that is uncertain, there are
facts established as securely as the conclusions of Kepler and Ne^vton ;

and he will gather in the testimony of geographer, historian, philo-

1
A. H. Keane, Ethnology, 2nd ed., 1896, p. 397.
 THE ETHNOLOGY OF GAUL 261

lugist, archaeologist, and an open and yet guarded

craniologist with
mind. Nor need he be afraid, even if he is not a Celtic scholar or
a professional anthropologist, to form an opinion of his own. For
he will observe that the specialists, in so far as they differ among
themselves, are simply drawing their own conclusions from ascer-
tained facts w^hich are accessible to all.
Anthropologists are obliged to make use of technical terms, more
or less uncouth and they are guided in their observations by very
;

precise and minute rules, framed with the object of eliminating, as

far as possible, the chance of error. But it is unnecessary for my
purpose to trouble the reader with more than a few of these things.
What I shall have to say about stature,^ complexion, hair, and eyes,
will need no explanation and in regard to the skull I shall, as a rule,
;

only have to deal with that measurement which fixes the proportion
between its length and its breadth. In this measurement the length
is represented by 100 and the proportion which the breadth bears
;

to the length is called the cephalic index. Thus, if the breadth is

four-fifths of the length, the index is 80. According to the system
formulated by the great French anthropologist, Paul Broca,^ skulls
are grouped, according to their cephalic index, in five classes. Skulls
w^hose index exceeds 83-33 are brachycephalic those whose index ;

falls between 83-33 and 80 are sub-brachycephalic those between 80 ;

and 77-77 mesaticephalic those between 77-77 and 75 sub-dolicho-

;

cephalic; and those below 75 dolichocephalic.^ It is necessary to bear

in mind that measurements of living heads invariably yield a higher
—
cephalic index the average difference being as much as 2 than —
those of skulls.* The orbital index, which is also important, denotes

^ The stature of prehistoric men, when their skeletons are found, can only be

estimated by calculating the relations between the lengths of certain bones and
the actual height of the individual and since these relations are obviously
;

variable, the calculation can only lead to approximately true results. The
error would no doubt be insignificant if the average relations were certain ;

but various anthropologists have adopted various methods of calculation,

which are explained in my Ancient Britain (pp. 379 and n. 2, 740). The most
satisfactory seems to me to be that of Dr. Beddoe [Journ. Anthr. Inst., xvii,
1888, p. 204),
—I take away from the length of the femur [or thigh-bone]
'

one-quarter of the excess over 13 inches up to 19, and thereafter only one-
eighth, and then multiply by four but in the following pages I have of course
'
;

accepted the calculations of French anthropologists. Even if, as Dr. Beddoe

thinks, they are generally rather below the mark, the error is, for our purpose,
unimportant.
2 3Iem. d' anthr., iv, 1883,
p. 243.
^ According to the notation which is prevalent in this country, skulls whose
indices exceed 80 are called brachycephalic, those between 80 and 75 mesati-
cephalic, and those under 75 dolichocephalic. Professor Gustaf Retzius {Journ.
Anthr. Inst., xxxix, 1909, pp. 290-1), who adheres to the simple classification
established by the famous Anders Retzius, under which brachycephaly was
separated from dolichocephaly by the index 80, considers that Broca's minute
subdivision has made the whole study of the question complicated and
'

obscure'. Many French anthropologists follow a system as simple as ours,

the only difference being that they take 77 instead of 75 as the superior limit
of dolichocephaly.
* Professor W. Z. Ripley, however, thinks that the difference is nearer 1-5
than 2 {U Anthr., vii, 1896, pp. 516-9) ; while Mr. Gray {Man, ii, 1902, no. 41,
262 THE ETHNOLOGY OF GAUL
the relation of the breadth of the orbit to its length and another
;

important character of the skull or head is gnathism, that is to say,

the degree of projection of the upper jaw. The word orthognathous
denotes that this projection is comparatively slight; for, as Professor
A. H. Keane observes, absolute orthognathism does not exist.
'

All races are more or less prognathous.' ^ The remaining technical

terms which it is necessary for general readers to understand are
those which describe the structure of the nasal skeleton. Platyr-
rhinian means that it is wide, mesorrhinian intermediate, and
leptorrhinian narrow.
The value of the cephalic index was for many years taken for
granted in all ethnological treatises and many anthropologists
;

still lay great stress upon it.^ But there has lately been a reaction.^
Sergi ^ scoffs at the old and discredited method of the cephalic
'

index, which only indicates artificial and conventional distinctions ',

and tells us that it is the forms alone that we have to take into
'

consideration ',^ and that indices may serve to approximate the

'

most diverse forms and to separate the most homogeneous '.' This
last remark is unquestionably true as Huxley said, in nine cases
:
'

out of ten you may diagnose an Australian skull [among other

dolichocephalic skulls] with certainty.'^ Nevertheless the cephalic
index, used with discrimination, retains the value which Broca,
Beddoe, Collignon, Turner, and other anthropologists of the old
school ascribe to it and those who are familiar with Sergi's WTitings
;

will not be surprised to learn that, when it suits his purpose, he lays
great stress upon the distinction between dolichocephalic and
brachycephalic skulls.^ He considerably modifies his view when he

pp. 50-1) regards the method of subtracting 2 as illogical, and would subtract
8 mm. from the breadth and 10 mm. from the length.
1 Ethnology, p. 182. See UAntlir., xx, 1909, p]). 35-50, 175-88.
^ At the meeting of the Congres international d'anthwpologie. &c., held at

Monaco in 1906, suggestions were formulated for an international unification of

cranial measurements, and adopted unanimously {L'Anthr.. xvii, 1906, p. 572 ;

Bev. de V Ecole d'anthr., xvii, 1908, pp. 47-59).

' For instance, Professor Ripley {The Races of Europe, 1900, p. 37), Dr. Beddoe
{Journ. Anthr. hist., xxx, 1900, no. 93), Sir W. Turner {Trans. Boy. Soc. Edin-
burgh, xl, part iii, 1903, pp. 547-614), Professor Symington {Report of . the
. .

British Association, 1903, p. 796) and Professor G. Retzius {Journ. Anthr.

;

Inst., xxxix, 1909, p. 288).

* See Professor C. S. Myers's article {ib., xxxiii, 1903, pp. 36-40), and Man,
iii, 1903, no. 13, pp. 28-32.
5 The Mediterranean Bace, 1901, p. 102. « lb.,
p. 104.
' lb., p. 195. « Proc. Soc. Ant., 2nd ser., iii, 1864-7, p. 284.

» For instance, on pages 136, 138, 143, 160-2, 189-92, and 238. So also
Dr. William Wright, after affirming {Middlesex Hospital Journal, xi, 1907,
p. 263) that it is the shape which is instructive
*
and {ib., xii, 1908, p. 40) that
'

'the cephalic index . has no intrinsic value', admits on second thoughts

. .

{ib., p. 42) that '

both the cranial shape and index have a certain stability',
and, like Sergi, appeals in support of his arguments to the cephalic index over
and over again {ih., pp. 45, 49, &c.). Moreover, when he finds it necessary to
'

remind its adherents that unless we accept a dual origin of man [which, by the
way, Klaatsch does] we must admit that all skulls are the result of variation
from a single type' {ib., p. 41), may not the said adherents, who really do not
need this reminder, reply that if the admission which they are bound to make
 2

THE ETHNOLOGY OF GAUL 263

affirms the truism that we cannot accept the evidence of the cephalic
'

index when that evidence is contradicted by other important facts ^ '

;

but if any one will spend a few days in walking through the depart-
ment of the Jura or the mountainous parts of Auvergne, the contrast
between the round heads which he will see everywhere and the totally
different type which he has been accustomed to in his own country
'
will convince him that the cephalic index has been discredited '

in vain.
The truth is that in general neither Sergi's method nor cranial
measurement is sufficient in itself ^ the rivals should combine.
:

The former, as A. Mochi has pointed out,"* when it is used exclusively,

begets confusion, since different observers, judging by the eye alone,
may disagree as to the particular one of Sergi's numerous categories ^
to which this or that skull should be referred and, moreover, ;

Sergi generally takes account of the norma verticalis the view of the —
skull from above —
alone in other words, he characterizes a solid
:

by one of its faces. Now Mochi has measured, with the object of
ascertaining how far the two methods agreed, the skulls which had
been previously studied by Sergi or his pupils and his researches
;

have shown that on the whole the correspondence is close, but that
—
two of Sergi's groups the ellipsoids and the ovoids are not, —
apparently, homogeneous, the cephalic indices of the former ranging
from 61 to 82.
But there are writers who regard the study both of indices and of
cranial shapes as little better than waste of time. M. Salomon
Reinach has observed that the hope of reconstructing the history
of prehistoric migrations by means of skulls was one of the illusions
of the nineteenth century ^ and if the remark does not contain
;

quite as much truth as he intended to put into it, it is not to be

wholly ignored. Broadly speaking, there are two anthropological
schools, one of which believes in heredity, the other in the influence
of environment. Professor Ridgeway, who tells us that he holds

tells against reliance upon the cephalic index, it tells equally against reliance upon
the shape ? Any one who has a clear head will see that it tells against neither.
The doctor allows {ib., p. 42) that, unless the environment alters, a long-
'

headed people breeds on the whole long-headed, a round-headed people round-

headed descendants.' Precisely and that is why the cephalic index remains
;

useful, and will not mislead provided that we are careful to note the cases in
which skulls that fall within the same group of indices differ, and those that fall
within different groups are alike, in regard to shape.
^ The Mediterranean Race,
pp. 199-200.
'^
Cf.U Anthr., xix, 1908, pp. 478-9. Even Sergi, as Mr. Myers observes
{Journ. Anthr. Inst., xxxiii, 1903, p. 37), shows signs of yielding the isolated
'

position which he originally took up as to the utter worthlessness of indices.'

" In France Sergi's grouping of skulls
according to their shapes as viewed from
above is generally discredited {U Anthr., xvii, 1906, p. 207), while Professor H.
Klaatsch (Weltall und Menschheit, ed. H. Kraemer, ii, 1902, pp. 32-4) criticizes
his methods and those of the old school with impartial vigour.
* In Archivio per V antrapologia, xxxviii, 1908, pp. 87-124, a valuable —
paper, ably summarized by Dr. Rivet in U
Anthr., xx, 1909, pp. 115-8.
* Ovoid, pentagonoid, sphenoid, ellipsoid, rhomboid, beloid, platycephalous,

spheroid, and cuboid.

« Rev. arch., 4« ser., ix, 1907, p. 486. Cf. ih., xv, 1910, pp. 446-7.
264 THE ETHNOLOGY OF GAUL
'
very strongly the doctrine of heredity ',^ nevertheless pushes the
doctrine of environment to its extreme ^ for him the three so-called
:

—
European races Mediterranean, Alpine, and Nordic are not several —
groups, each sprung from a common stock, but simply products of
environment; and cranial measurements and shapes are ethno-
logically useless. He observes that the change in the type of the
'

American of New England from that of his English ancestor and his
approximation to the hatchet face and thin, scraggy beard of the
Red Indian has long been remarked '. Again, Starting from the '

Mediterranean, we meet in the lower parts a melanochrous race ;

but gradually, as we advance upwards, the population as a whole

is growing less dark, until finally, along the shores of the Baltic, we
meet the tallest and most light-complexioned race in the world . . .

it is difficult to believe that the movements up or down of the people

from the southern side of the Alps, or of those from the shores of
the Baltic, have been so nicely proportioned as to give the general
steady change from north to south in coloration without the aid of
some other force. The case of America, which I have just cited, is in
itselfenough to raise a suspicion that climatic influences are at work
all the time, and that environment is in reality the chief factor in
the variation of both stature and pigmentation from the Mediter-
ranean to the Baltic. The white race of the north is of the same
proximate ancestry as the dark-complexioned peoples of the northern
shores of the Mediterranean. ... If we turn from man to the other
animals we find a complete demonstration of this doctrine. For
instance, the conditions which have produced a blonde race on the
Baltic have probably produced the white hare, white bears, and the
tendency of the stoat and the ptarmigan to turn white in the winter.
It may be objected that the Lapps and Eskimo are not tall and
blonde, but on the contrary short and dark but they live within
;

the Arctic circle in regions where the sun does not shine at all for
a great part of the year, and consequently they are quite outside
the conditions of environment under which the tall, blonde race
of North Germany has long dwelt.' Further, the blonde element '

in the Berbers is not a survival from invasions they rather owe

. . .

their fair complexions and light-coloured eyes to the circumstance

that they were cradled in a cool, mountainous region, and not along
the low-lying border of the Mediterranean like their dark-coloured
relations whose language and customs they share although the . . .

fair-haired race Oi Upper Europe has age after age kept pouring
over the Alps into Italy and the other southern peninsulas,^ and have
{sic) constantly intermixed with the indigenous populations, it is
only in the upper part of Italy that the blonde race is able to hold
its own.' And, lastly, the Alpine type of skull does not mean
' ' '

any racial difference ', but, on the analogy of the changes in the
'

^ Science Progress, 1910, p. 131.

» Nature, Sept. 24, 1908, pp. 526-9.
* Into Spain at all events ' the fair-haired race '
trickled rather than poured
Cf pp. 282-3.
.
 THE ETHNOLOGY OF GAUL 265

osteology of the Equidae the professor urges that the roundness

' '

of the skulls was simply due to [Alpine] environment '.

Is the type of the American of New England really approximating
to that of the Red Indian ? May I suspect that if such New Eng-
landers exist, they have Red Indian blood in their veins ? Two years
ago I heard Professor Ripley of Harvard University, who, as a trained
observer, is presumably familiar with the features of his countrymen,
deliver the Huxley Memorial Lecture before the Royal Anthropo-
logical Institute. Professor Ridge way, the President of that society,
was in the chair, and paid his American colleague many graceful
compliments. It is amusing,' said Professor Ripley,
'
to read in '

the older books on ethnology, and even in the files of this learned
body, of the undoubted effect of the American climate upon Europeans
in tending to produce the black wiry hair, the bronze skin and the
aquiline features of the American Indian. Such conclusions are,
of course, now understood to be a product, not of climate but of
vivid imagination.' ^ The way in which Professor Ridgeway struggles
to account for the inconvenient presence of the Lapps in Northern
Scandinavia is hardly satisfactory :
—
they live within the Arctic
'

circle in regions where the sun does not shine at all for a great part
of the year.' It would appear, then, that the great heat and brilliant
sunshine of the Mediterranean produced darkness flus dolichoce-
phaly that in the same southern zone brilliant sunshine, tempered
;

by mountain breezes, produced fairness plus dolichocephaly, although,

according to the professor, mountains elsewhere produce brachy-
cephaly that the coolness and exiguous sunshine of mountainous
;

Scandinavia produce fairness plus tall stature plus dolichocephaly ;

that either the very exiguous sunshine or the temporary absence of

sunshine of Lapland produces darkness plus dwarf -like stature plus
extreme brachycephaly and that the very exiguous sunshine of
;

'
Greenland's icy mountains produces darkness plus dolichocephaly
'
;

while the Polar bears owe their whiteness to the same conditions,
intensified, that have produced the swarthiness of the Eskimos and
the Lapps. Is the professor serious ? Is he not aware that side by
side with the swarthy Lapps in the far north live blonde Norwegians ;
and that 500 miles south of Lapland, along the shores of the Sogne
Fjord, an intensely dark group, whose presence is a standing puzzle
to their blonde neighbours, has lived from time immemorial ?
Let us first examine the anthropological chart of Europe with
an eye to stature. There is no such regularity in the distribution
as the professor's theory requires. There are two great areas of
tall stature one in the south-east, embracing Southern Dalmatia,
;

Bosnia, Herzegovina, and Servia, the other in the north and north-
west, including Scotland, Ireland, Sweden, parts of Norway, and
of the Baltic provinces, the south-west of Finland, the Netherlands,
and Schleswig-Holstein. Again, there are three regions of short
stature the southern comprises Sicily and Southern Italy, Corsica,
;

1 Journ. Anthr. Inst., xxxviii, 1908, pp. 221-2. Cf. Nature, Nov. 3, 1910,
p. 11. The effect of the cHmate has not yet been soientitically tested.
266 THE ETHNOLOGY OF GAUL
and Sardinia, South-western France, and part of Spain the central :

comprises Hungary and Poland the northern North-east Russia

;

and Northern Scandinavia. In the rest of Europe stature is medium ;

but there are islets, so to speak, where the population is generally tall,
and others where it is generally short.^
In regard to colouring and cranial shape Professor Ridgeway's
generalization is disturbed by numerous exceptions. To say nothing
of the dolichocephalic Berbers, whose fairness he attributes to a
mountainous country, which, on his own theory, ought to have
made them brachycephalic, fairness is not uncommon in Corsica,
where also, contrary to the general rule, the tallest inhabitants are
mountaineers, whose stature, as M. Verneau remarks,- cannot be
attributed to environment. In the small area which coincides with
Latium there are two types, differing from one another in every
respect one is dark, small, and dolichocephalic the other brachy-
:
;

cephalic, tall, and comparatively fair.^ When we find that in

America environment has not produced the least modification in
the negro type * that intensely dark groups are dw^elling and have
;

been dwelling for millenniums in foggy England as well as in sunny

Sicily that short dark brachycephali abound in the mountains of
;

Auvergne and tall fair dolichocephali in the mountains of Corsica ;

that tall fair men are conspicuous in the district round Limoges,
while in the adjacent department of the Dordogne the population
is short and dark ^ that in the Perigord there is an isolated group
;

of intensely dark dolichocephali;^ that the people of Zeeland are

dark and have round skulls, while those of Friesland and North
Holland are 'extremely blonde' ;'^ that the Walloon frontier separates
tall dolichocephali from short brachycephali ^ that the territory of
;

the Czechs is surrounded on the chart by 'lighter-coloured tracts';^

that the mountainous region in Central Europe is occupied by round-
heads, in the southern peninsulas by long-heads ^^ and, finally, that
;

in Great Britain M. Deniker's shades of colouring are one and all

—
represented ;^^ we may, for our purpose, leave environment out of
the question. I need hardly add that anthropologists who have
specially studied the question of the Berbers, or Kabyles, have
concluded that they are descendants of prehistoric European
invaders who occupied the tracts that suited them best.^- The
^
U Antlir., XX, 1908, pp. 481-2 ; Bull, et mem. de la Sac. dCanthr. de Paris,
S*" ser., ix, 1908, pp. 456-8. No anthropological map of Europe can adequately
denote the complexity of the various groups and M. Deniker's map, being on
;

a very small scale, would mislead readers who were not on their guard.
Bull, et mem. de la Soc. d^anthr., 5*^ ser., iii, 1902, pp. 360-1.
^

Atti d. Soc. Bom. d. antropologia, xii, 1906, pp. 32-117 L^Anthr., xviii, 1907,
^
;

p. 730.
* Bull, et
mem. de la Soc. d'anthr., 5*" ser., ii, 1901, pp. 152, 157.
5 Annales de geogr., v, 1896, « lb., xii, 1903, p. 210.
p. 164.
Journ. Anthr. Inst., xxxv, 1905, pp. 230-1.
' ^ lb.
* lb., p. 226. 10 lb., xxxiv,
1904, p. 206. " lb., p. 206 and pi. viii-ix.
^^
Association franf. pour V avancement des sc, 36*^ sess"., 1907, p. 1040 ;
Zeitschr.f. Ethn., xl, 1908, pp. 525-7; Archiv f. Anthr., N. F., viii, 1909,
pp. 249-86. Cf. Bull, de la Soc. d'anthr., 2« ser.. ix, 1874, pp. 125-6 ;xi, 1876.
pp. 467-8.
 THE ETHNOLOGY OF GAUL 207

truth is that Professor Ridgeway's generalization is both hasty and

superficial.
But I must take note of a reply which he has made to an incom-
petent critic. Answering the objection that a dark race exists in
the British Isles, he says, Yes but very different from the dark
'
. . .

race as seen in the southern peninsulas the skin is beautifully . . .

fair, and the eyes are often blue. ... In this type I maintain that
we have the transition stage between the full, melanochrous and . . .

the tall blonde Scandinavian.'^ In other words, give the little dark
people time, and (although the poor melanochrous Lapps will ' '

remain melanochrous) they will become like tall fair Norwegians.

I suppose that this is what the professor means for the tall fair ;

Scandinavians who migrated into Scotland more than a thousand

years ago remain tall and fair so presumably their environment is
;

favourable. Well, time will be wanted, —

time running into millen-
niums. For, although the professor assures us that osteological
'
variations take place within very short periods ',^ although he
affirms —
perhaps on somewhat unsubstantial evidence that the —
physical type of pygmies who had left the forests of the Congo was
changed in three generations,^ although less than three centuries
have apparently sufficed to transform the Anglo-Saxon into the
semblance of a hatchet-faced Red Indian, the weaker aboriginal '

dark race ^ of the British Isles has remained dark ever since the
'

'
beginning of the Neolithic Age. Admit that their beautifully fair '

skin is due to environment, though it may be due partly to inter-

marriage still, that is hardly sufficient to prove that
; the blond '

tall race ... is identical in origin with the small dark long-headed
'
race .^

Here, as elsewhere, the via media is the path that will lead to truth.
Imperfect though our knowledge is, no impartial anthropologist
would deny that the influence of environment is considerable ^ ;

but, as Dr. Beddoe says, if geographical

'
position be the warp,
heredity, race, inherited type is the woof, and its threads are extremely
tough and lasting.' Short dark men were not transmuted into tall
'^

fair men by the climate of Scandinavia but tall fair men found the
;

climate of Scandinavia congenial. Assuming that, as Dr. Wright

affirms,^ the so-called three races of Europe are in the main the
'

result of variation from a common European stock, a variation due

to isolation and natural selection,' my argument is unaffected for ;

at the outset of the Neolithic Age, if not earlier, these three races did
exist as distinct and permanent entities.

'
Science Progress, 1910, p. 129. 2 Jb., p. 128.
'
lb., p. 133. * lb., p. 137. « lb., p. 127.
« See Nature, Nov. 3, 1910, pp. 11-2.
' Journ. Anthr. Inst., xxxv, 1905, p. 226.
^ Middlesex Hospital Journal, xii, 1908, p. 44.
268 THE ETHNOLOGY OF GAUL

n
THE PREHISTORIC RACES
Until a very recent date not a single human skull had been found
in any part of the world which could, with reasonable probability,
be referred to an earlier period than the Quaternary, or Pleisto-
cene. Even the Pithecanthropus, or monkey-man ', of Java the '
—
famous missing link whos3 remains were discovered in 1894 by
' '

—
Dr. Dubois although it was certainly of Pliocene Age,^ is not unani-
mouslyregarded as an intermediate form between the simian precursor
of Man and Man himself.^ For many years past, indeed, certain
geologists have insisted that flints of various well-marked forms
which are known as eoliths ', or stone implements of a dawning age,
'

and which have been found in Tertiary deposits in France and

perhaps also in England, show marks of intelligent workmanship ;

but this claim has never won general acceptance, and by many
experts has been treated with contempt.^ Nevertheless, just five
years ago the eminent palaeontologist. Monsieur Marcellin Boule,
avowed that he believed firmly in the existence of Tertiary Man :

'
I do not doubt,' he added, 'that some day his traces will be found. '^
Three years later we were startled by the announcement that this
prophecy was fulfilled. Six miles south-east of Heidelberg, in a bed
of sand which marks an abandoned channel of the river Neckar,
was discovered an enormously massive lower jaw-bone, which
resembles in profile the corresponding bone of a gorilla. Its age
seemed certain it was unhesitatingly assigned to the Pliocene,
:

or latest division of the Tertiary Period. Had it, however, been less
perfectly preserved, the anatomists who examined it would hardly
have allowed their enthusiasm to seduce them into giving it the name
by which it will be known, Homo Ileidelhergensis for some of its ;

characters are ape-like, and it is more primitive than any other

known human specimen. But fortunately the teeth had escaped
destruction ; and their human character was beyond all doubt.^
But what about its age ? Professor Boule himself does not claim
for it a date anterior to the early Pleistocene ^ Klaatsch, asking
:

^ V Anihr., xx, 1909, pp. 373-7. Tins paper corrects another in an earUer
number, — xix, 1908, pp. 260-9.
2 See Archivf. Anthr., N. F., v, 1906, p. 225, Science Progress, iii, 1908, p. 345,
and Dr. A. C. Haddon's Hist, of Anthropology, 1910, pp. 76-7.
^ Rice Hohnes, A)ic. Britai7i,-p-p. 25-30. See also Archiv f. Anthr., N. F., iv,
1905, pp. 75-86; Association franc, pour Vavanceynent des sc, 35^ sess°., 1906,
pp. 603-28; Man, viii, 1908, no.' 26, pp. 49-53; and J. Dechelette, Manuel
cCarcheologie, i, 26, n. 3. There is also an article by Hoernes in Mitteil. d. natur-
wiss. Vereines f. Steiermark, xlv, 1909, which I have not yet been able to see.
* r
Anthr., xvi, 1905, p. 267.
^ O. Schoetensack, Der Unterkiefer d. Homo Heidelbergensis, 1908. Cf. Rev.
de VEcole d' anthr., xix, 1909, pp. 105-8, and U
Anthr., xx, 1909, pp. 81-6.
6 V
Anthr., XX, 1909, pp. 266-7.
 '

THE ETHNOLOGY OF GAUL 269

whether it is to be assigned to that period or to the late Tertiary,

remarks that there is no sure boundary between the two.^ All that
is certain is that Homo Heidelbergensis is the oldest known human
fossil ; and it is not unreasonable to suppose that in those incalculably
—
remote days his fellow men if men they are to be called lived and —
hunted on the western side of the Rhine.^
Of the people who were living in Gaul in the earliest period of the
Palaeolithic Age no fossil remains, except perhaps a skull unearthed
from the volcano of Denise,^ have yet come to light our oldest :

examples belong to the second period, which derives its name from the
cave of Le Moustier in the valley of the Vezere. Most of them are
called after that famous specimen which was discovered about half
a century ago in the valley of the Neander in Rhenish Prussia.^
The skulls of this type are dolichocephalic ^ and the people to whom;

they belonged had extremely low and retreating foreheads, heavy

and projecting lower jaws, and amazingly prominent brow ridges,
and w^ere short, big-boned, and muscular. Representatives of the
race have been found in the cavern of Spy in Belgium, as far east as
Croatia, as far west as the valley of the Dordogne.^ A skeleton which
was found in the cavern of La Chapelle-aux-Saints in the Correze
was in a nearly perfect state of preservation : like the other fossils
of the Neanderthal type, it had a defective chin and the stature ;

was barely 1 metre 60 centimetres, or about 5 feet 3 inches. Between

it and Homo Heidelbergensis, says Professor Boule, there are resem- '

blances more considerable than one could have expected, and perhaps
actual relationship.' ^
At the end of the nineteenth century the Neanderthal skull was
accepted by the most eminent anthropologists of Europe and
America as the type of the most ancient of the known races of men.
But in 1901 a German anthropologist, Dr. G. Schwalbe, wrote an
article of appalling length,^ which disturbed settled convictions.
Huxley had pronounced the Neanderthal to be the most ape-like
Zeitschr.f. Etlin., xli, 1909, pp. 538-41.
1

Mr. J. Reid Moir {Times, Oct. 17, 1910, p. 8, col. 2) states that bo
2

has discovered flint implements which were unhesitatingly acknowledged

'

—
by Miss Nina Layard a well-known investigator and other experts, in —
undisturbed Pliocene strata near Ipswich. This claim must be further tested.
3 BuJl. de la Soc. de geogr., xiii, 1906.
pp. 289-91, 366.
* The age of the Neanderthal skull itself is uncertain. See L'Anthr., xvii,
1906, pp. 70-3.
5 It has been said that one of the Mousterian skulls which have been found
at Krapina in Northern Croatia, and which, in prominence of the supraciliary
ridges, resemble the Neanderthal type, has an index of over 80 ; but these skulls
were too fragmentary for accurate measurement {L'Anthr., xvi, 1905, pp. 17-8).
« L'Anthr., xvii, 1906, pp. 156-8 xix, 1908, pp. 106-7, 519-25 ; xx, 1909,
;

pp. 220, 603-4 xxi, 1910, p. 85 ; Philosophical Transactions, cxcix b, 1907,

;

pp. 281-339 Archiv f. Anthr., N. F., vii, 1909, pp. 287-97.

;

^ Comptes rendus de V Acad, des sc, cxlvii, 1908, no. 24, pp. 1349-52 ; no. 25,
pp. 1414-5; L'Anthr., xx, 1909, pp. 257-71. See also ib., pp. 576-7; xxi,
1910, p. 84 ; and Archiv f. Anthr., N. F., vii, 1909, pp. 287-97.
^ Bonner Jahrbiicher, cvi, 1901, pp. 1-72. See also Globus, Ixxx, 1901,
pp.' 217-22 Ixxxi, 1902, pp. 165-74
; Man, ii, 1902, no. 129, pp. 186-9 and
; ;

L'Anthr., xiii, 1902, pp. 356-8 ; xvii, 1906, pp. 67-72.

270 THE ETHNOLOGY OF GAUL
of all known human Schwalbe refused to regard it as human,
skulls :

in the accepted sense, at all. For him it represents, or represented,

a distinct species, Homo primic/enius,^ intermediate between Pithec-
anthrojms and Homo sapiens. In the same class he placed the skulls
of Spy (those of the Correze had not then come to light) and he ;

insisted that all the other human palaeolithic skulls of Europe, how-
ever closely they might appear to resemble these, were in reality
different.
Now it is universally admitted that between Homo- sapiens and the
men of Neanderthal, Spy, and the Correze there is no generic dis-
tinction whether there is any specific difference depends upon the
:

meaning which is to be attached to the word species.' ^ Having ' '

regard,' says Dr. William Wright, to the bones other than the
'
. . .

cranium [and] to the teeth, I prefer to speak of the skeletons con-

stituting a type rather than a species a type, however, characterized —
by numerous well-defined simian traits.' ^
Did the men of the Neanderthal race leave descendants whose
remains are recognizable among the later skulls of Gaul ? Opinions
differ.** Professor Gustaf Retzius ^ remarks that as yet no genuinely '

transitional forms between the Neanderthal race and

. . the
. . . .

race branches in Europe at the present day have been discovered ',
and that the trend of opinion
'
favours the assumption that the
. . .

real Neanderthal race became extinct '. It is true that, as Dr. Wright
observes, none of the later skulls which approximate to the type ^
'
fulfil all the conditions necessary for their admission but who ' ; ''

would have expected that the type should have remained pure ?
Since other palaeolithic races are unquestionably to be reckoned
among the ancestors of modern Frenchmen, is it conceivable that the
men of the Correze and the Dordogne should have utterly died out ?
But whether there was or was not any specific difference between
the men of the Correze and Homo sapiens, there is not the slightest
doubt that, despite the skill with which they fashioned their tools,
they were intellectually far inferior not only to the earliest men of
whom we read in history, not only to Man as he had become before
the close of the Palaeolithic Age, but also to men who were their con-
temporaries. persons to whom anthropology is Abracadabra
Many
have heard the Baousse-Rousse caves near Mentone.
of Deep
down in the Grotte des Enfants were found the skeletons of an aged
woman and of a youth who were living in Mousterian times. Although
^ Professor G. Retzius [Journ. Anthr. Inst., xxxix, 1909, p. 290, note) tliinks
the term primigenius premature, as the Neanderthal race probably belonged '

to a lateral branch of the main stem and were preceded by other forms which
would rather have merited that name '.
- Comptes rendus de V Acad, des sc, cxlvii, 1908, no. 24, pp. 1349-52 no. 25, ;

pp. 1414-5. Cf. r

Anthr., xix, 1908, p. 215.
^ Middlesex Hospital Journal, xi, 1907, p. 98.

^ U
Anthr., xix, 1908, p. 469. Cf. Archiv f. Anthr., N. R, vii, 1909, p. 247.
* Journ. Anthr. Inst., xxxix, 1909, pp. 297-8.
® See Scottish Review, xx, 1892, pp. 148, 152-3, and Rice Holmes, Anc.
Britain, pp. 385, 397.
"
Middlesex Hospital Journal, xi. 1907, pp. 97-8.
 ;

THE ETHNOLOGY OF GAUL 271

the skulls of both had certain negroid characteristics, which reappear

in a few dolichocephalic skulls of the Neolithic Age,^ they were of
a markedly higher kind than the skulls of Neanderthal and the
Correze. The chin, indeed, was feebly developed but the brow-:

ridges were unobtrusive the cranium was symmetrically formed

;

and the forehead was almost vertical.'^ Thus comparatively early

in the Palaeolithic Age there were living in Gaul men of two strongly
contrasted types.
Before the end of the Quaternary Period the zoological evolution
of Man had come to its end.^ The dolichocephalic skulls that have
been discovered at Laugerie-Basse and Chancelade in the valley
of the Lozere are as different from those of the hunters of the Correze
as the skull of Cuvier from that of a Hottentot.^ More capacious than
those of an average Frenchman of to-day, they have high and broad '

foreheads and brow-ridges which are hardly perceptible '. From their
mere contour one would be justified in concluding that the intellectual
power of the race which they represent was not inferior to our own ^ ;

and the work of that race remains to confirm this conclusion. For
it is to the period which has been called after the cave of La Madeleine
that belong the finest specimens of the carved antlers, the engraved
stones and bones, and the frescoes, drawn on the walls of caverns,
which are recognized by modern artists as true works of art.^
The rock-shelter of Cro-Magnon in Perigord has given a name to
a type of men, related to that of Laugerie-Basse, representatives
of which have been found in many parts of France and also in other
European lands."^ The famous skeleton which is known as the old '

man of Cro-Magnon belonged to the later part of the Palaeolithic

'

Age.^ The Cro-Magnon race was long-headed and, unlike that of

Laugerie-Basse, tall, the old man's stature having been estimated
at 1 metre 90, or about 6 feet 2| inches.^ His cranium was large,
his forehead high with slight supraciliary ridges, and his chin
well developed. ^^ The characteristics of the type had probably
been acquired at a time earlier than his for in the Grotte des
;

Enfants, about 16 feet below the level of the uppermost interment,

^ Bull, et mem. de la Soc. d'anthr. de Paris, 5^ ser., v, 1904, pp. 119-24.

2 UAnthr., xiii, 1902, pp. 561-85 xvii, 1906, pp. 109, 291-320 xviii, 1907,
; ;

p. 625. Middlesex Hospital Journal, xi, 1907, pp. 99-100. ^ lb.,

p. 103.
* The average height of the people of Laugerie-Basse and Chancelade was

only 1 ni. 60, or about 5 ft. 3 in. {Bull, et mem. de la Soc. d'anthr., 5*" ser., viii,
1907, p. 214).
^ See Anthropology and the Classics, ed. R. R. Marett, 1908, p. 18.
^ See p. 7. Palaeolithic art, which advanced through successive stages,
began in the first division of the later Quaternary Period, known as the Auri-
gnacian, developed in the Solutrean, and culminated in that of La Madeleine.
' Professor G. Retzius {Journ. Anthr. Inst., xxxix, 1909, pp. 297-8) believes
that the tall North European race is descended from that of Cro-Magnon.
^ Strictly speaking, to the earliest of the three later divisions, which is called

after Aurignac in the Haute-Garonne.

" According toTopinard,M.Rollet estimated it at Im.
80, or about 5 ft. 10 J in.,
and M. Manouvrier at only 1 m. 75, or about 5 ft. 8f*^ in. See Mem. de la Soc.
d'anthr., 2" ser., iv, 1893, p. 389 Rice Holmes, Anc. Britain, p. 379, n. 3.
;

"» Bull, de la Soc. d'anthr., 2'- ser.,

iii, 1868, pp. 352-91.
272 THE ETHNOLOGY OF GAUL
which was itself palaeolithic, and little more than two feet above
the negroid Mousterians, was found a gigantic skeleton, closely
resembling the old man' ^ and similar skeletons were discovered
'
;

in adjoining caves.^ Indeed the chief value of the Mentone caves

lies in the demonstration which they have furnished of an intimate
connexion between the palaeolithic and the neolithic race in Southern
France. The newest skeleton in the Grotte des Enfants approximates
to the dolichocephalic type, presently to be considered, of the
Neolithic Age.^
After the artists of South-western Gaul had ceased to flourish
two widely different races, both of short or medium stature, one
brachycephalic, the other dolichocephalic, w^ere diffused over the
length and breadth of the land. The type to w^hich the latter be-
longed, characterized by remarkable elegance, gentle contours, and
orthognathism, is called after two groups of skeletons which were
discovered in the caverns of I'Homme Mort and Baumes-Chaudes
in the department of the Lozere. The latter group comprised
35 skulls, every one of which was dolichocephalic, the indices
ranging from 64-3 to 76 T in the former two were mesaticephalic,
:

having indices of 78-8 and 78-5 respectively, while the remaining

seventeen varied between 68-2 and 76-7. The average height of
each group, according to M. Manouvrier's method, was 1 metre 60,
or barely 5 feet 3 inches.'* Similar remains have been found in various
parts of Britain, in Central and in Southern Europe.^ Most English
anthropologists, following Sergi, include these people under the so-
called Mediterranean race but in France this term is branded as
:

misleading ^ and French ethnologists with good reason regard the

;

cave-dwellers of Baumes-Chaudes and I'Homme Mort as descended

from the Quaternary race of Laugerie-Basse.'^ Whatever may have

1 UAnthr., xvii, 1906, pp. 297-302.

* The average height of these men was 1 m. 87, or about 6ft. liin. Ih.,
pp. 110-1.
^ It has been —
maintained that another the so-called steatopygous race —
existed in Gaul in late palaeolithic times. The existence of this people is
inferred from the discovery of certain 'statuettes' at Brassempouy in the
department of the Landes [UAnthr., vi, 1895, pp. 129-51) and near Mentone.
I have not seen them :but when I saw the woodcut of one which was selected
for illustration {Bull, et mem. de la Soc. d'anthr., 5* ser., iii, 1902, p. 775, fig. 4),
it seemed to me that the carving was so villainous that no scientific conclusion
could be drawn from it ;and I found that this was also the opinion of
M. Manouvrier {ih., p. 778). See, however, J. Dechelette, Manuel d'archeologie,
i, 217. Supposing that tiiere was a steatopygous race in Southern Gaul, is
there any reason to suppose that it was not identical with the negroid race ?
* Rev. d'anthr., ii, 1873, pp. 1-53 ; Bull, de la Soc. d'anthr., 3^ ser., i, 1878,
pp. 213-4 ; llein. de la Soc. danthr., 2*^ ser., iv, 1893, p. 388 ; Rev. mensuelle
de rEcole d'anthr., v, 1895, pp. 162-4, 171 ; Bull, et mem. de la Soc. d'anthr.,
5« ser., viii, 1907, pp. 150-2, 174, 305, 540-1 ix, 1908, pp. 723-5.
;

« Rice Holmes, ^«c. Britain, pp. 393-401; UAnthr., xx, 1909, p. 212;
Archivf. Anthr., vii, 1909, pp. 249-56, 266-7.
» Bull, et mem. de la Soc. d'anthr., 5^ ser., ix, 1908,
p. 166.
^ Rev. mensuelle de I'Ecole d'anthr., v, 1895, p. 408. The old theory of
a '
hiatus between the Palaeolithic and the Neolithic Age in Gaul has long
'

since been exploded. See J. Dechelette, Manuel d'archeologie, i, 309-12.

 THE ETHNOLOGY OF GAUL 273

been the original home of that race, its oldest known representatives
lived, should seem, in the neighbourhood of Mentone.^
it

In a tumulus, apparently of the Neolithic Age, on Cape Blanc-Nez,

at Escalles, in the department of the Pas-de-Calais, a dolichocephalic
skull has been found of a type widely different from the skulls of
I'Homme Mort. The index is only 71-9 the skull has a strongly-
;

marked occipital protuberance, a projecting chin, and prominent

supraciliary ridges.^ In fact, tall dolichocephalic people undoubtedly
existed in France in the Neolithic Age ^ and M. Verneau, whom one
;

specimen reminded of the rugged skulls of Borreby in the Danish

island of Falster, remarked, after he had examined the collection to
which it belonged, that races already formed an almost inextricable
'

medley '.^
The brachycephalic people, w^ho are generally called after Grenelle,
near Paris, where six typical specimens have been discovered,^ were
for the most part extremely short,^ and differed widely, even in the
shape of their skulls, from the comparatively tall and moderately
brachycephalic round-headed people of the British Bronze Age,''
who are generally known as the Round Barrow race. The character-
istics of their modern descendants, which must have been noticed
by many travellers, are clearly described by Dr. Collignon, taille
— '

plutot petite, cheveux fonces, tete globuleuse, face ronde, courte,

large, plate, nez large et court.' ^ They undoubtedly mingled with
the dolichocephalic population for in many of the neolithic settle-
;

ments that have yielded human remains dolichocephalic, mesati-

cephalic, and brachycephalic skulls were associated, although in
some only long skulls, in others only broad were to be found.^ The
'
U Anthr., vii, 1896, p. 687 ;xvii, 1906, p. 292. Cf. p. 272, supra, and Rice
Holmes, Anc. Britain, p. 382. 8ome neolithic people of this type may have
been immigrants.
•^
Congres internat. cCanthr. et cCarch. prehist., 1872 (1873), p. 308 ; Bull, de la
Soc. d' anthr. de Paris, 2" ser., xi, 1876, pp. 140-1.
Some ethnologists have argued that the tall dolichocephali of the Neolithic
•^

Age belonged to an early wave of blonde invaders, and were akin to the
Tamahu, whose features are portrayed on the Egyptian wall-paintings which
were executed about 1500 b. c. 8ee J. Beddoe, The Races of Britain, p. 14.
* U Anthr., i,1890, pp. 171-2, 184. Tall dolichocephalic individuals also
existed in the Neolithic Age in our island (Rice Holmes, Anc. Britain, p. 395).
Various dolichocephalic skulls, characterized by considerable anterior'
elongation without the occipital protuberance which is common in skulls of the
Homme Mort type, and belonging to individuals of medium stature, are classed
by French anthropologists under the type de Genay {Bull, de la Soc. d' anthr.,
' '

2* ser., iv, 1869, pp. 89-98). M. Herve [Rev. mensuelle de VEcole d' anthr., v,
1895, p. 145) assimilates them to the Hohberg and Row Grave skulls (see p. 327,
infra) ;and they are commonly regarded as of northern origin.
5 A. de Quatref ages and E.T. Hamy, Crania Ethnica,
1882, pp. 116, 120-4, 141.
« Rev. mensuelle de V^cole d'anthr., iv,
1894, pp. 396, 400; vi, 1896, p. 105.
Still, in one station, where races were intermingled, the average height of
33 adult males was 1 m. 654 (about 5 ft. 5f in.), the limits being 1 m. 541
(about 5 ft. fin.) and 1 m. 752 (nearly 5 ft. 9 in.). Bull, et mem. de la Soc.
d'anthr., 5" ser., viii, 1907, p. 563.
' The Grenelle race was represented, though not largely, in the Bronze
Age in Britain. See p. 329. » Ann. degeogr., v,
1896, p. 164.
® Rev. mensuelle de V Ecole d'anthr., v, 1895, pp. 155, 161, 407. See also
1093 X
274 THE ETHNOLOGY OF GAUL
departments in which brachycephalic skulls were fewest are west
of a line drawn from Normandy to the sources of the Garonne,
virtually coinciding with the area in which dolmens are most abun-
dant while skulls of this type were most numerous, on the one
;

hand in Belgium and the valleys of the Seine and Marne, on the
other in Alsace, the neighbourhood of the lake of Neufchatel, and
the valley of the Rhone below Geneva and accordingly M. Philippe
:

Salmon concludes that a brachycephalic race migrated into Gaul by

two routes, one of which passed through Belgium, the other through
Savoy .^
In a cave called the Trou du Frontal, near Furfooz, in the valley
of the Lesse, which flows into the Meuse just below Dinant, two
skulls have been found, one male, the other female, with indices
of 79-31 and 81-39.2 ^\^q male skull was rather low, with a low
and retreating forehead and moderately developed supraciUary
ridges and the stature was only 1 metre 53, or about 5 feet J inch.
;

For many years hardly any similar specimens were discovered

outside the basin of the Meuse and it was commonly believed that
;

they were a mere variety, due to crossing between the races of

Grenelle and I'Homme Mort ^ but early in this century about 60
:

neolithic skeletons were found in the Grotte des Bas-Moulins, near

Monaco, of which the great majority belonged to the Furfooz type ;

and MM. R. Verneau and L. de Villeneuve, who examined them,

concluded th it they represented a distinct race.'*
Owing to the prevalence of cremation in the Bronze Age, the data
for ascertaining the physical type of that period are scanty and ;

if fresh immigrants appeared, there is no reason to suppose that

they represented any new stock.^ At the beginning of the Hallstatt

Period, however, which was contemporaneous with the late Bronze
Age in AVestern Gaul, tall men, whose skeletons have been found in
tumuli, were living in the Jura and the Doubs. Some of them, like
the Celtic invaders, of whom they may have been the advanced

Bull, et mem. de la Soc. cCanthr. de Paris, 5" ser., v, 1904, pp. 73-6 viii, 1907,
;

pp. 150-2, 170, 174, 307-9, 539-41, 563. M. Philippe Salmon {Rev. mensuelk,
&c., V, 1895, p. 159) shows that of 688 neolithic skulls found in Gaul 397 were
dolichocephalic, 145 mesaticephalic, and 146 brachycephalic. Of 140 inter-
ments 55 contained only dolichocephalic skulls, and '20 only brachycephalic.
^ Rev. mensuelle de lEcole d'anthr., v, 1895, p. 184.
2 Crania Ethnica, pp. 105-9 ; A. de Quatrefages, Hist. yen. des races hum.,
1887-9, pp. 72-3, 448. The Furfooz skulls, which were wrongly assigned by
de Quatrefages to the Quaternary Period, are now known to be neolithic.
^ Rev. mensuelle deV E cole d'anthr., v, 1895,
p. 100.
* UAnthr., xii, 1901, pp. 1-27. In the first edition (p. 254, n. 4) I criticized
a theory, propounded by the late Canon Isaac Taylor {The Origin of the Aryans,
1889, pp. 66, 113, 115, 118-9), which implied that the Aquitanians, the Celtae,
and the Belgae were each identical with one of three races whose remains
'
'

have been found in neolithic tombs, and that in the early neolithic age
'

Gaul was conquered by the ancestors of the Belgic Gauls

'
but as it is unlikely
'
;

that the theory has now any adherents, I refrain from reproducing the criticism.
Everybody knows that the Aquitanians, the Celtae, and the Belgae were not
races at all, but mixtures of races.
^ See Rev. mensuelle de VEcole d'anthr., vi,
1896, pp. 102-3; UAnthr., xviii,
1907, p. 128; and Rev. arch., ^ ser., xiii, 1909, pp. 217-9.
 — -

THE ETHNOLOGY OF GAUL 275

guard, were dolichocephalic others, apparently including nidividuals

;

who resembled the tall Round Barrow race of Britain, as brachy-

' '

cephalic as the people of Grenelle.^

The prehistoric ethnology of Switzerland has been exhaustively
treated by M. A. Scheiik.^ No palaeolithic skulls are known.
Eighteen skeletons have been found in neolithic cists at Chamblandes,
near Lausanne, comprising dolichocephalic and mesaticephalic
specimens, the average height of which was 1 metre 582, or about
—
5 feet 2J inches. The skulls represented three types, that of Baumes-
Chaudes, a negroid type resembling that of Mentone, and another
which is believed to be of northern origin. ^ Twenty-seven skeletons
all either dolichocephalic or mesaticephalic —
have been found near
Schafihausen, and several others at Dachsenbiiel in the canton of
Schaffhausen. The two latter groups included individuals of very
short stature, who, however, must not be called pygmies.* Forty-
three skulls of the Neolithic Age have been found in lake-dwellings,
of which 6 belong to the earliest period, 20 to the period when
the art of fashioning stone implements was at its zenith, and
17 to the following period of transition. The earliest group is
entirely of the brachycephalic Grenelle type the second is composed
:

of 10 Grenelle skulls, belonging to individuals whose stature,

calculated by M. Manouvrier's method, varied from 1 metre 50
(barely 5 feet) to 1 metre^GO (about 5 feet 3 inches), 2 mesaticephalic
and 8 dolichocephalic the third is likewise mixed, but here dolicho-
;

cephalic skulls are predominant.^ The Bronze Age is represented

by 51 skulls of various types, dolichocephalic, akin to those of the
Neolithic Age, being again the most numerous but in the period
;

of transition, which has left 14 of the specimens, pure brachycephalic

skulls once more form the majority.
That the original brachycephalic inhabitants of Gaul were invaders
who came from the east is with most French anthropologists an
article of faith.^ M. Zaborowski observes that Sicily, which was
already inhabited by a dolichocephalic people, was entered in the
Copper Age by round-headed immigrants, coarse -visaged and short
nosed, like those who are figured on Assyrian monuments of about
3700 B. c.,'^ and that people of similar type appeared in Crete in the

1 VAnthr., xi, 1900, pp. 382-4, 386, 388-9, 400.

^ Rev. de V^cole d'anthr., xiv, 1904, pp. 335-78 ; xv, 1905, pp. 389-407 ;

Bull, et mem. de la Soc. d'anthr. de Paris, 5" ser., viii, 1907, pp. 212-28. Cf. Rev.
mensuelle de VEcole d'anthr., v, 1895, pp. 137-54, and UAnthr., x, 1899,
pp. 281-9 ; xvii, 1906, pp. 547-57 ; xix, 1908, pp. 283-4.
' See * UAnthr., xvii, 1906,
p. 273, n. 4. pp. 78-9.
^ The latest Swiss brachycephalic skulls of the Neolithic Age belong to the

—
Disentis type so called after a town in the canton of Grisons —
which is simply
a variety of the brachycephalic type of Middle Europe. They are very short
and broad, having higher indices than those of the Grenelle type.
" Rev. mensuelle de VEcole d'anthr., vi, 1896, p. 105; viii, 1898, pp. 201-8;
xiv, 1904, p. 378 ; xviii, 1908, pp. 394-6; Bull, et mem. de la Soc. d'anthr.,
5^= ser., viii, 1907.
p. 227
; UAnthr., xx, 1909, p. 232. See also Geogr. Journal,
xxviii, 1906, p. 554 ; Archivf. Anthr., N. F., vii, 1909, pp. 266-7.
' Rev. de VBcole d'anthr., xviii, 1908, pp. 394-6.
T 2
 276 THE ETHNOLOGY OF GAUL
' '
—
Eneolithic period ^ the period of transition between the Neohthic
—
and the Copper Age which corresponded with the Neolithic Age
of Northern Gaul.^ Remarking that similar skulls have been found
on the eastern coast of Spain, associated with objects of Asiatic
origin, he concludes that the round-headed inhabitants of Spain and
those of Central Europe were of Asiatic descent.^
Certain anthropologists, however, hold that the origin of the
so-called Alpine race is uncertain * while others, notably Dr. William
;

Wright, believe that the type was evolved on European soil '.
'

Dr. Wright, indeed, apparently believes that brachvcephaly was

indigenous in Gaul. Admitting that its existence in Europe in the '

Palaeolithic period is doubtful ',^ he points out that some of the '

skulls from Furfooz, Crenelle, and the kitchen-middens of Portugal

(all of which he apparently regards as mesolithic) are brachycephalic ',
and that nine of the eighteen Chamblandes skulls, which he would
like to assign to the same period, are mesaticephalic and he concludes
;

that on the whole the evidence points to the presence of brachy-

'

cephalic individuals in Europe in Mesolithic, if not in Palaeolithic

time \^ Why the Chamblandes skulls should be quoted, I cannot see ;

for Dr. Wright himself justly admits that two of the implements
that were found with them prove that they belong to the Neolithic '

or polished-stone Age '.'^ He goes on to remark that 'if Europe

be charted according to the cephalic index of its inhabitants, it is
found that a wedge-shaped area, having its base to the East and its
apex to the West, contains the great mass of the round-headed
population of Europe but, allowing that the shape of the area
'
;
'

i^ certainly suggestive of an intrusion from the East ', he nevertheless

insists that his opponents have fallen into the error of believing '

that the cephalic index was distributed in prehistoric times as it is

to-day.'^ As he observes, 'the great bulwark'
— or one of the —
bulwarks of the theory is the statement that the earliest Swiss
'

lake-dwellers w^ere round-headed but he urges that if instead

'
;
'

of trusting to the cephalic index we take note of the shape of the

brachycephalic crania, wc find that they are pentagonoid not . . .

round '.^ Here speaks a loyal disciple of Sergi. His conclusion is

that historical invasions from Asia 'are mainly responsible for the high
cephalic index possessed by the Eastern parts of Central Europe '.^^

Rev. de PEcoIe d'anthr., xviii, 1908, p. 397.

^

Rev. arch., 4^ ser., xii, 1908, pp. 230-1.

2

Rev. de VEcole d'anthr.. xviii, 1908, pp. 2, 18-9.

•''

' ArcMvf. Anihr., N. F., vii, 1908, pp. 172-5.

^ Middlesex Hospital Journal, xi, 1907. p. 190. « Ih.. pp. 189-90.
' lb., p. 180. Dr. Wright says in another place {ib., xii. 1908, p. 46) that
'
it is 80 difficult to assign a dwelling to its proper period . that it would be
. .

unwise to place too much reliance on the dating of various Swiss lake-dwellings'.
But is not this what he does himself in regard to mesolithic dwellings ? For
' '

proof that the Chamblandes skulls belonged to the close of the Neolithic Age
see Bull, et 7nem. de la Soc. d'anthr., 5*" ser., v, 1904, p. 614, and J. Dechelette,
Manuel dUircheologie, i, 464.
« Middlesex Hospital Journal, xii. 1908, p. 45. » lb., p. 46.
^^ lb.,
p. 47. In a review, which appeared in Man, viii. 1908, no. 109.
pp. 191-2, of my Ancieiit Britain, Dr. Wright insists, further, on the presence
'
 THE ETHNOLOGY OF GAUL 277

Perhaps ;but obviously not for that of Gaul. The brachycephalic

skulls of the earliest Swiss lake-dwellings are unanimously assigned
by Continental anthropologists to the Grenelle type in other words, ;

they belong to the Alpine race. Moreover, it seems difficult to

' '

reconcile Dr. Wright's theory with the facts that in the two oldest
—
known neolithic stations in Gaul those of Baumes-Chaudes and
r Homme Mort —
almost every skull was dolichocephalic and not
one brachycephalic, and that of other neolithic sites some yielded
only brachycephalic, others only dolichocephalic skulls.^ These
facts seem to me and to all anthropologists, except Dr. Wright and
Professor Ridgeway, to prove that in Gaul, as in Britain, the round-
heads, wherever their cradle may have been, were originally invaders.^

HI
THE LIGURIANS
From the statements of the ancient geographers and historians
1.

itappears that the Ligurians in Transalpine Gaul inhabited a tract

between the Maritime Alps and the Rhone that they were inter- ;

mingled with the Iberians on the west of the Rhone and that their :

country was invaded by Celts.^

In the time of Hesiod the islands of Hyeres were called AiyiJo-rtScc.^

of a sub-bracliycephalic population in North Wales in Mesolithic time. This

population', he says, was almost certainly earlier than that whose dead are
'

found in the long barrows of Wiltshire, and yet it was well on its way towards
brachycephaly.' But is it not a little rash to infer the presence of a sub- '

brachycephalic population from the existence of three skulls, only one of

'

which had an index as high as 80 ? These skulls, moreover, are generally

classed as neolithic {Anc. Britain, pp. 395-6) and even if they were locally
;

mesolithic, it does not follow, considering the remoteness of the district in

which they were found, that they were older than the long barrows.
See p. 273.
^

^ In many neolithic interments of France skeletons have been found side

by side which differed from one another in every respect. Par suite,' says
'

M. R. Verneau {VAnthr., xii, 1901, pp. 547-8), on ne s'explique pas que le

'

meme milieu ait agi sur les uns de maniere a les modifier dans un sens et sur les
autres de fa9on a les modifier dans un sens oppose.' And again, Lorsqu'on '

reporte sur une carte de I'Europe les stations neolithiques dans lesquelles ont
6t6 trouves des cranes brachycephales, on voit se dessiner bien nettement
un courant oriente a peu pres de Test a I'ouest. Dans cette derniere direction
on constate avec non moins de nettete que le courant se divise et qu'entre
ses branches existent des zones franchement dolichocephales. Peut-on sup-
poser un seul instant que le milieu ait presente alors des variations absolu-
ment desordonnees sur un espace restreint ? . C'est la une hypothese absolu-
. .

ment inadmissible, tandis que la theorie des courants de migration accepte'e

par presque tous les anthropologistes explique ces faits de la fa9on la plus
satisfaisante . . .ce qu'on pent dire c'est qu'il est infiniment vraisemblable
que ce berceau [of brachycephaly] ne saurait etre place en France.' See also
VAnthr., xviii, 1907, p. 127, and Journ. Anthr. Inst., xxxix, 1909, p. 299.
^ The earliest extant mention of the
Ligurians, contained in a line borrowed
by Eratosthenes from a Periplus of the 6th century b. c. and quoted by Strafco
(vii, 3, § 7), is too vague to help us.
Aryonautica, ed. L>idot, bk. iv, 554-5.
''
278 THE ETHNOLOGY OF GAUL
According to Festus Avienus,^ a writer of the fourth century, whose
Ora maritima was based ultimately upon the Carthaginian account
of the voyage of Himilco,^ the Rhone formed the boundary between
the Ligurians and the Iberians but Hecataeus,^ who wrote about
;

500 B.C., says that the Elisyces, who dwelt in the neighbourhood
of Narbonne, were Ligurians. In another passage, which will be
examined presently, Avienus says that the northern part of Gaul
also, and therefore presumably the whole, was once occupied by
Ligurians, who were expelled by Celts.^ Scylax ^ says that the
Ligurians inhabited the country east of the Rhone as far as Antium
(in Italy), and that the country west of the Rhone w^as occupied by
Ligurians mixed w4th Iberians.^ According to Aristotle,*^ Ligurians
dwelt in the neighbourhood of Bellegarde, some 20 miles south-
west of Geneva. Scymnus of Chios ^ says that Massilia was in
Liguria, and that the easternmost town of Liguria in Gaul was
Antipolis, or Antibes. Strabo ^ implies that in this part of Gaul
Celts were mingled with Ligurians and this is also proved by the dis-
;

covery of Celtic inscriptions at Nimes and Saint-Remy.^^ According

to an inscription quoted by Gruter,^^ the Vocontii, who dwelt in the

^ Ora maritima, 608-10.

^ H. d'A. de Jubainville, Principaux auteurs a consulfer svr Vhist. des
. . .

Celtes, &c., 1902, p. 42. Cf. Bheiii. iMvs., N. F..1&95, pp. 321-47.
1,

Hist. Graec. fragm., ed. Didot, t. i, p. 2, fr. 20. M. Philipon {Les Ihercs, 1009,
=>

p. 121, n. 2) argues that this passage is an inter})o]ation for, lie says, appealing
;

to Herodotus, vii, 165, § 1, the Elisyces were not Ligurians, but Iberians. But
Herodotus sa^'^s no sucli thing he differentiates the Elisyces both from the
:

Iberians and from the Ligurians and M. d'Arbois de Jubainville (Les premiers
;

hahitants de V Europe, i, 1889, p. 377) infers that they were a medley of Ligurians
and Iberians. * Ora maritima, 129-35. * Geogr. Graec. w/??., ed. Didot, i, 15-7.

^ It is important to decide whetlier in the country which was occupied

by Ligurians mingled with Iberians the Iberians had invaded Ligurian or the
Ligurians Iberian territory. M. Jullian {Hist, de la Gaitle, i, 265. nn. 2, 4)
remarks that Hecataeus, who wrote about 500 B.C., mentions no Iberians on
the north of the Pyrenees, whereas only a few years later, when the Peri pi us
which Avienus copied appeared, Roussillon and the neighbourhood of Narbonne
were inhabited by Iberians. But when we consider the admitted brevity of
the interval which separated the Periphis from the work of Hecataeus, and the
impossibility of fixing the date of either with precision, M. Jullian's inference
seems to rest upon unsubstantial grounds. He appeals to Avienus {Ora
maritima, 586-8) and Herodotus {Fragm. hist. Graec, ed. Didot, vol. ii. p. 34)
to show that the Ligurian kingdom of the Elisyces was conquered by Iberians;
but in the former text the Iberians, in the latter the Elisyces are not mentioned.
Scylax, as we have seen, states that Ligurians were mingled with Iberians
west of the Rhone whereas Avienus says that the Rhone separated the t\\o
;

peoples. Must we not conclude from a comparison of these two texts (with
which compare H. d'Arbois de Jubainville, Les premiers hahitants de V Europe.
i, 1889, p. 375) that Ligurians crossed the Rhone after the date of Avienus's
Periplus and encroached upon Iberian territory ? If we are to escape this
conclusion, wo must supposeeither that Avienus's statement was not intended
to be taken literally, or that it was derived from some authority later tlian the
Carthaginian Periplus. See Bhein. Mus., 1, 1895, pp. 321-47. and Rice Holmes,
Anc, Britain, p. 490. ' Meteorol, i, 13, § 30.
^ Oeogr. Graec. min., ed. Didot, i, 204, vv. 211, 216.

8 ii,
5, § 28 iv, 6, § 3.
; Cf. J. C. Zeuss, Die Deutschen, 1837. p. 163.
>» Rev. celt., xviii,
1897, p. 320.
^^ Corpus inscr., ed. Graevius, t. i, pars
2, p. ccxcviii.
 THE ETHNOLOGY OF GAUL 279

department of the Drome, were Ligurians but the names of the :

towns in their territory were Celtic Strabo ^ clearly implies that ;

they were not Ligurians and, though there was doubtless a Ligurian
;

stratum in the population, the Celts were indisputably predominant.

Strabo ^ says that the Salluvii, in whose country Marseilles is situated,
were called Ligurians by the old Greek writers and Pliny ^ calls them ;

by the same name. Liyy,^ not troubling himself about ethnological

distinctions, speaks of them as Gauls and Strabo ^ says that, in ;

o
later times, they were known as Celt -Ligurians. Similarly Plutarch ^
says that the Ligurians were intermixed with Gauls and Iberians.
Strabo also says that the Taurini and the inhabitants of the kingdom
'^

of Cottius, that is to say the Alpine country between Embrun and

Drubiaglio, were Ligurians and Ligurians are said to have colonized
;

Spain,^ Corsica,^ and Sicily .^^ Finally, Diodorus Siculus ^^ describes

the Ligurians as small and lean, but robust.
The passage in which Avienus says that the northern part of
Gaul was once occupied by Ligurians is worthless. Desjardins,^^ -^j^q
has minutely analysed it, remarks with grim humour that Avienus
would have smiled if he could have foreseen that Miillenhoff and other
scholars would take these lines seriously. When we remember how
vaguely the word Celt was sometimes used by ancient writers,
'
'

we cannot be sure that the meaning of the word Ligurian was ' '

constant.^^ Moreover, the lines are not strengthened by the statement

which Avienus makes in another poem, that the whole of Gaul was
once occupied by Iberians ^'* and they are inconsistent with the state-
;

ment, which is supported by other writers, that (about 500 b. c,

before the Celts reached South-eastern Gaul) the western frontier
of the Ligurians was the Rhone.
2. M. d'Arbois de Jubainville ^^ endeavours to prove, from the
evidence of toponymy, that the dominion of the Ligurians in Gaul
extended much further than is generally believed. Remarking the
frequency with which the suffixes -asco, -asca, -usco, -usca, -osco, and
-osca are found in the province of Cuneo in Piedmont, which was
undoubtedly inhabited by Ligurians, he infers that these suffixes

2
' iv, 6, § 4. iv, 6, § 3.
3 Nat. Hist., iii, 5 (7), § 47. * Epit., 60. Cf. xxi, 26, § 3.
* iv, 6, § 3. Aemilius Pauhis, 6. «

' iv, 6, § 6. Dr. Lagneau says that, according to Strabo, the Salassi, the
Veragri, and the Nantuates were Ligurians but, as Desjardins points out :

{Geogr. de la Gaule mm., ii, 92-3, n. 5), Strabo nowhere makes this statement;
and Polybius(ii, 15, § 8) calls these peoples Gauls. Still, their territory may have
been occupied by Ligurians before Celts invaded it.
8 See p. 280, n. 1. Seneca, Dial, xii, 7, §§ 8-9.
»

^° Dionysius of Halicarnassus, i, 22, 2§ Silius Italicus, xiv, 37.

; M. J. De-
chelette {Manuel d'archeologie, ii, 8, n. 1) believes that their statements are
confirmed by archaeological evidence.
'^ iv,
20, § 1. 12
Geogr. de la Gaule rom., ii, 52-5, note.
1^ See the significant
remarks of Strabo, i, 2, § 27.
'* Seep.
281, n. L
15 Rev. celt., xi,
1890, p. 156; Les premiers habitants de VEurope, ii, 1894,
p. 115.
280 THE ETHNOLOGY OF GAUL
were Ligurian ^ and, after a further search, he finds that they
;

occur also in the departments of the Alpes-Maritimes, Var, Bouches-

du-Rhone, Gard, Herault, Basses-Alpes, Vaucluse, Hautes-Alpes,
Drome, Ardeche, Savoie, Isere, Ain, Rhone, Jura, Saone-et-Loire,
Cote-d'Or, Doubs, Haute-Saone, Yonne, Aube, Marne, Haute-Loire,
Aveyron, and Ariege.^
M. d'Arbois ^ draws a further argument from the name Rhodanus
(Rhone), which, like K. Miillenhoff,* he regards as Ligurian. The
name, he remarks, was borrowed bv the Greeks from their country-
men, the colonists of Massilia, who, as late as the fourth century B. c,
had only Ligurians for neighbours.^ If it were objected that Rhodanus
might have been the name of the Rhone before the Ligurians entered
Gaul, he would reply that the word conies from the root Rot or Rod,
which is found also in Rodmnna, a word undoubtedly Ligurian.^
M. d'Arbois then points to the existence of des Rhodanus near ' '

Treves, Cahors, and Le Mans, and infers that, at some time or other,
the districts which surround those towns were inhabited by Ligurians.

'
M. E. Philipon {Les Iberes, 1909, pp. 150-1), who holds that Iberian and
Ligurian were closely related, affirms tliat the suffixes in question were
common to the two languages for, he observes, they are found in Spain as
;

well as in Gaul, and Ligurians never entered Spain. This reckless assertion
implicitly contradicts Thucydides, vi, 2, § 2 and M. d'Arbois de Jubainville,
;

who gives a curious reason (see p. 281, n. 1) for rejecting his statement,
nevertheless infers that there were Ligurians in Spain from Avienus, Ora
maritvma, 284-5, with which cf. Strabo, ii, 1, § 40. M. Dechelette, liowever
{Manuel d^archeoloyie, ii, 8, n. 1), argues on archaeological grounds that the
Ligurian element w^as small.
•^
M. Maximin Deloche {Rev. celt., xviii, 1897, pp. 366-71) argues in a similar
strain. He says that in the department of the Basses-Pyrenees there is a hill
called Legorre that two forests in the basin of the Dordogne and the depart-
;

ment of the Aisne were respectively called, in the Middle Ages, silva de Lignrio
and Ligurium, and so on and he agrees with M. d'A. de Jubainville {Lea
;

premiers habitants de V Europe, ii, 207) in regarding the word Liger (the Loire)
as probably of Ligurian origin. But, as one of his reviewers sensibly remarks
{Rev. arch., S" ser., xxx, 1897, p. 424), the fact tliat the site of the ancient
Epidaurus is now called Ligourio warns us to view such arguments with
suspicion and M. d'Arbois de Jubainville himself gives reasons {Rev. arch.,
;

nouv. ser., xxxi, 1876, pp. 379-88) for rejecting the derivation proposed by
Artemidorus (cited by Stephanus of Byzantium, ed. A. Meineke, i, 416), of
the name of the Ligurians from the river Aiyvpos (Loire).
^ Les premiers habitants de V Europe, ii, 1894,
pp. 124-5.
^ Deutsche Altertumskunde, i, 193-4.

'"
Scylax, § 4, in Oeogr. Graec. min., ed. Miiller, i, 17-8.
M. Philipon {Les Iberes, pp. 104, 119, 121) asserts that Rhodanus is not
*

a Ligurian, but an Iberian word for, he argues, the Ligurians would liave
;

written not Rhodanus, but Rhodenus, as in A-qn^vva, the Greek form of tlie —
Ligurian name of the Lake of Geneva (see A. Holder, Alt-celtischer ^prach-
schatz, ii, and, moreover, the Ligurians did not reach the Rhone until
172-3) ;

about 400 whereas the word Rhodanus was mentioned by Aeschylus and
B.C.,
Avienus. Certainly it was but Aeschylus (quoted by Strabo, iv, 1, § 7)
:

implies that Provence was inhabited by Ligurians, and, as we have seen,

Avienus expressly says that the Rhone was the boundary between the Iberians
and the Ligurians. So also Herodotus (v, 9, § 3) speaks of the Ligurians who
dwell above Massilia (A('7i'6s- ol dvoj vvlp MaaaaXlijs olKiovTa). The mode
in which M. Philipon manipulates texts is not one of his strong points. As
to his linguistic argument, what right has he to deny that in a language of
which we know hardly anj'thing the termination -ayios was inadmissible ':
 THE ETHNOLOGY OF GAUL 281

Sequana, the Gallic name of the Seine, is also, he maintains, a Ligurian

word ^ for if it were Celtic, we may be sure, from our knowledge
;

of a phonetic law which influenced the development of Celtic speech,

that the q would have been replaced by a 2? at a date anterior to the
first Celtic invasion of Gaul.^ The Ligurians, unlike the Celts, did
not change q into j) ^ witness Quadiates,^ the name of a Ligurian
:

tribe who dwelt in the Cottian Alps.

Sir John Rhys, who formerly regarded Sequana not as Ligurian
but as Celtican ', that is to say, belonging to a dialect akin to
'

Goidelic, the ancestor of Gaelic, now thinks that Celtican was

' '

'
practically the same language as that which M. dArbois de Jubain-
ville calls Ligurian '. Obviously it was, whether it was akin to
Goidelic or not but the question is whether it was originally spoken
;

by Ligurians and not by Celts. Sir John Rhys's reason appears to

be simply that the same dialect was at a later time spoken, or at
all events written, in territory which was inhabited by Ligurians.
But, as we shall afterwards see, he still holds that this dialect was
spoken by the Celtae before they were conquered by those Celts
who imposed upon them the Gaulish (or Gallo-Brythonic) language,
which was virtually identical with the Brythonic speech from which
Welsh and Breton are sprung and it would therefore seem that in
;

his opinion the Celtae, or rather that branch of the Celts who formed
the conquering group among the Celtae, learned Celtican from
' '

Ligurians.^ This, as we shall afterwards see, is contrary to all

^ It is amusing, if at first sight somewhat perplexing, to find that M. Philipon

{Les Iberes, pp. 127-9) insists that Sequana must be an Iberian word, because it
ends in -ana (see the preceding note) and, appealing to Avienus {Descriptio orhis
;

terrae, 416-8), he argues that the Iberian empire(!) 's'etendait a travers la Gaule
jusqu'a la mer du Nord.' Thus the authority of Avienus is invoked, on the
one hand to show that the Ligurians occupied the whole of Gaul in the sixth
century b. c, on the other hand to show that at the same period it was subject
to Iberians ! Avienus' s Descriptio was based upon the Perieyesis of Dionysius.
who flourished about 300 a.d., and whose work was itself derived largely
from Eratosthenes.
M. d'Arbois de Jubainville {Les premiers habitants de VEurope, i, 1889,
pp. 28-9) likewise holds that the Iberians, in the wider sense of the word, were
ortce masters of Gaul. He thinks that it would not be rash to search in Gaul
for the river Sicanus, mentioned by Thucydides (vi, 2, § 2), and to identify
it with the Sequana, —
the Seine. The Sicanians of Sicily, as Thucydides
informs us, were Iberians, and were driven from the river Sicanus in Iberia
'

by Ligurians'. Thus, according to M. d'Arbois, Gallia was first Iberia and

afterwards Liguria Sicanus was the equivalent of Sequana
: and Sequana
;

was a Ligurian word Are not these hypotheses as futile as they are bewilder-
!

ing ? Cf. Desjardins, Geogr. de la Gaule rom., ii, 45, n. 3, and E. Hiibner,
Monumentaling. Iber., 1893, p. Ixxxv.
2 Les premiers habitants de VEurope, ii, Les Celtes,
1894, pp. 132-3, 282 ;

1904, pp. 17-9. M. d'Arbois holds, however, that although the Celts had
— —
changed q into p for instance, *Qarisii into Parisii they had recovered the '

faculty of pronouncing q ', and that accordingly the Sequani called themselves
after the Ligurian name of the Seine.
^ See * Corpus inscr. Lat., v, 7231, § 14.
p. 319.
5 Or else that Ligurians and
Q Celts were one The Celts were sharply
'
' !

distinguished by the ancient writers from the Ligurians would Sir John Rhys
:

contend that the language of the Celticans', when they invaded (Taul, was
'

identical with the language of the Ligurians .'

282 THE ETHNOLOGY OF GAUL
analogy and I would ask Sir John Ehys whether he believes that
;

the Goidels who invaded the British Isles learned Goidelic from
Ligurians. I cannot help thinking that his partial recantation is
weak and with great respect I would suggest that he has thought
;

himself into a muddle. Considering the great dearth of linguistic '

remains belonging to the older and subjugated people on which '

he rightly insists, is it not more likely that Sequana was Ligurian

and f re-Celtic than that a solitary tribe of Q Celts should have ' '

appeared in the midst of P Celts ', and retained their distinctive

'

name although the existence of a town called Efomanduodurum ^

proves that even among them P Celts Avere or became pre- '
'

2
dominant ?

And now, the reader is not too weary, let him listen to M. Camille
if

JuUian.^ This eminent scholar argues that the termination -brig a,

which is commonly regarded as Celtic, is probably Ligurian for, ;

first, most of the place-names which end in -hriga belong to the

Spanish peninsula, where the Celtic element was weaker than in any
other part of Western Europe secondly, many of these names are;

found in lands where the names of Celt or Celtiberian were never

'

uttered thirdly, -hriga, unlike -dunum, -durum, and -magus, is

'
;

hardly ever compounded with a word which is unmistakably Celtic ;

fourthly, in the country of the Cantabri -hriga is associated with
Latin words, as in Juliohriga, which proves that it belonged to t
language spoken in that country, and, says M. Jullian, if that
language had been Celtic, Mela and Pliny would have said so and, ;

lastly, -hriga is extremely rare in Gaul. May I remark in passing

that this last statement hardly supports M. Jiillian's contention that
in pre-Celtic times the masters of Gaul were Ligurians, and, further,
that -hriga is extremely rare even in the area which was unquestion-
ably Ligurian ?
All over Gaul, however, and likewise in Spain, Corsica, Italy,

^ I tin. Ant., ed. Wesseling, p. 386.

* Sir John Rhys, in a paper {Proc. Brit. Acad., vol. iv) which has just come
into my liands, has given a fresh reason for beheving that the famous calendar
of Coligny (see p. 319, n. II, infra) is in a Celtic language [which he still identi-
'

fies with Ligurian] other than Gaulish.' See Addenda.

M. Alfred Maury [Comptes rendus de V Acad, des inscr., &c., 4^ set., v, 1877,
pp. 208-11) argues that Ligurian was a dialect of Celtic, because Bodincus, the
Ligurian name of the Po (Pliny, Nat. Hist., iii, 16, § 122), reaenihles Agediticum,
and the names of various Ligurian tribes, mentioned by Pliny, affectent '

une physionomie celtique.' This is rather vague but Celts had settled in :

Ligurian territory and indeed M. Maury himself says (pp. 210-1), 11 nous
;
'

parait liors de doute que les Ligyens etaient une population celtique ou tout
au moins qui avait ete celtise.' As to Bodincus, Pliny made his statement on
the authority of a Greek, who said that the word meant bottomless and '
' ;

if this is true, it is hard to see how Bodincus and Agedincmn can have anything
in common. Cf Bull. bill. . du 21 usee beige, 1908, p. 69.
. . ,

Since the appearance of M. d'Arbois de Jubainville's Xe5 premiers habitants

de V Europe the prevailing opinion has been that Ligurian was at all events an
Indo-European language but M. Jullian {Hist, de la Gaule, i, 125) exercises
;

a judicious reserve when he says that 'I'origine indo-europeenne de cettelangue

est la moins invraisemblable des conjectures qu'elle a suggerees '.
2 Rev. des etudes anc, viii, 1906,
pp. -47-50, with which cf. ix, 1907, p. 178.
 —

THE ETHNOLOGY OF GAUL 283

Sicily,Germany, and Britain, M. Jullian ^ finds traces of the Ligurian

language. A thousand names or more of rivers, lakes, springs, and
mountains in France are of Ligurian origin, for instance, Ardennes,
Bievre, Dive, and Divonne and so are various names of towns, such
;

as Alesia, Bibracte, Bibrax, Geneva, and Nemausus.^

But M. Jullian's conclusions have not been allowed to pass un-
challenged. M. Loth ^ tells him that few words are more certainly
Celtic than -briga M. Philipon, justly remarking that if all the
;

forty-one names in the Spanish peninsula which ended in -brica or

'briga were Celtic, it would be necessary to admit, against historical
evidence, that the Celts were once masters of the whole country,
argues that the Celtic -briga, which appears in Magetobriga, must be
distinguished from another -briga, which was merely the Latin
adaptation of an Iberian word, meaning town ' * while M. d'Arbois^ '
;

and Dr. Victor Tourneur ^ insist that Ardennes, Bievre, Dive, Divonne,
and the rest are not Ligurian, but Celtic and M. Salomon Eeinach, ;

remarking that Pan-Ligurism is an illusion, asks whether the

'
'

predecessors of the Ligurians were dumb.*^

Still, it is not sufficient to show that some of the names which
M. Jullian cites are Celtic. He insists, with some exaggeration but
not wholly without reason, that his examples se retrouvent presque '

tous dans des regions ou I'influence celtique n'a jamais penetre'.^

Thus a Deva, which he equates with Dive and Dee, flowed through
the territory of the Spanish Basques ^ in the Portuguese Douro :

he sees an equivalent of the Italian Duria, which is likewise the

name of a tributary of the Danube ^^ Isara is the common name ;

of the Isere, the Oise, and the Bavarian Isar ^^ and Cebenna, the ;

Cevennes, otherwise Cimenice regio}^ is the counterpart of saltus

Ceminii,^^ the forests north of the Tiber. But if these names were
pre-Celtic, who can show that they were Ligurian ?

Sieglin^^ contends that the Aquitanian tribal names ending in

-ates are Ligurian ; and although Schuchardt ^^
suggests that they
may be Celtic, it is remarkable that another group of such names
Deciates, Desuviates, Ednates, Nantuates, and Quariates — exists in
Liguria, and nowhere else in Gaul.
The truth is that the linguistic data are insufficient for, as ;

Schuchardt ^^ says, hardly anything of the Ligurian language is

1 Hist, de la Gaiile, i, 111-7. Cf. H. d'Arbois de Jubainville, Les premiers
habitants de V Europe, ii, 1894, pp. 46-215.
2 Hist, de la Gaule, ii, 254.
=*
Rev. celt., xxviii, 1907, p. 339. See also p. 356.
Les Iberes, pp. 158-64. M. Vendryes {Mev. celt., xxx, 1909, p. 204) com-
mends M. Philipon' s suggestion. ^ Bev. celt., xxix, 1908, p. 81.
« Bull. bibl. du Musee beige, 1908, pp. 62, 69.
. . .

' Bev. arch., 4« ser., xi, 1908, pp. 302-3.

^ Hist, de la Gaule, i,
114, n. 3.
^ Ptolemy, Geogr., ii, 6, § 8. Cf. 3, § 4.
'» Pliny, Nat. Hist.,
iv, 12 (25), § 81.
^1 A. Holder,
Alt-celtischer Sprachschatz, ii, 72-5.
^'^
Avienus, Ora maritima, 622. ^^ Livy, ix,
36, § 7.
'* Sitzungsberichte d. Kl,nigl.
preuss. AJcad. d. Wissenschaften, 1896, p. 446, n. 3.
^^
Die iber. DeUination, p. 66. ^^
/j^
284 THE ETHNOLOGY OF GAUL
known, and unlikely that rival philologists will ever agree.
it is

Conjecture is of course necessary for the progress of linguistic science ;

but when conjecture has no material to work upon, it reminds

one of the scholastic chmiaera bombinans in vacuo. The force of
the argument from toponymy must not indeed be underrated ;

but if rigorously tested, I do not think that we can safely

it is
rely upon any names except Rhodanus, the terminations in -asco
and the like,^ and perhaps Sequana, Isara, and the tribal
names in -ates. This evidence would lead to the conclusion
that, besides the territory which the ancient writers assigned
to them, the Ligurians also possessed the eastern region as far
north as the Marne, which comprised Franche-Comte, Alsace and
Lorraine, Burgundy, and a part of Champagne, and perhaps also
Aquitania.2
3. We
have next to inquire, with a view to solving the problem
which we have just been considering, what were the physical charac-
ters of the Ligurians. Diodorus Siculus, as we have seen, says that
they were small and lean, but strong and the epithet lean may
;
'
'

possibly suggest that the individuals whom he or his informants had

observed belonged to the slender long-headed stock rather than to
the round-heads, who, though they were also small, were certainly
of sturdy build.^ A. Hovelacque has measured 70 modern skulls
found in the mountains of Savoy, the inhabitants of which, he main-
tains, have been free from all admixture of foreign blood."* The
average cephalic index of these skulls was 8541, and the indices
of seven exceeded 90, —
that is to say, they were more brachy-
cephalic than 88 Auvergnat skulls taken from St. Nectaire-
le-Haut, which Broca examined.^ There are other differences
between the two series. The Savoyard skulls are mesorrhinian,
those of Auvergne leptorrhinian.^ The orbital index of the
Savoyard skulls is 89-41, that of the Auvergnat 86-5 and the ;

cranial capacity of the former is considerably less than that of

the latter. The hair of the modern inhabitants of Liguria, adds
M. Zaborowski,^ is very dark brown, darker than that of the
inhabitants of Central France, and so are their eyes. Li fine,
says this writer, they are a type of the small, dark, brachycephalic
race at its purest.^
All this is the hastiest generalization. The Ligurians of Savoy
were conquered by Celts and Hovelacque's skulls, if they prove
;

anything, can only be regarded as typical of the Ligurians or the

^If Festus Avienus, on whose testimony M. d' Arbois relies, were right, would
it not be surprising that these names occur only in the eastern region, south
of the Marne ?

M. C. Jullian's article, De la necessite d'un Corpus topographique du

'

nionde ancien' {Klio, ii, 1902, pp. 1-13), is well worth reading.
- Cf. p. 297, and J. Dechelette, Manud
d'archeologie, ii, 8-11.
^ Mem. de la Soc. d'anthr. de Paris, 2^ ser., i, 1873, pp. 285-6.
* Rev. d'anthr., vi, 1877, pp. 226-52.
' See p. 309. " See p. 2(32.
' Diet, des sc. antkr., p. 247.
* See also Rev. de V Ecole d'anthr., xvii, 1907, p. 365.
 —

THE ETHNOLOGY OF GAUL 285

Celto-Ligurians. of a certain region.^ R. Livi ^ has measured 7,951

soldiers — —
natives of Italian Liguria whose mean index barely
—
exceeded 82, equivalent to 80 on the bare skull. It is impossible
to tell how much or how little mixture of foreign blood there may
have been in certain inhabitants of Italian Liguria only eight ^ —
in all —
whose skulls have been measured by MM. Nicolucci and Vogt ;

and even admitting that they were pure Ligurians, it is impossible

to tell what was the exact shade of their hair and eyes. Look at
the three tables, giving particulars of the stature, cephalic index, and
colouring of the inhabitants of the French departments, which have
been prepared by MM. Broca, Coliignon, and Topinard.* Taking the
departments in which the Ligurian element, in Caesar's time, was
probably strongest, we find that the Alpes-Maritimes, Var, and
Bouches-du-Rhone have indices of 82-85, 82-77, and 81-43 respec-
tively. The measurements on which these figures are based were all,
it must be remembered, taken from the heads of living men none :

of the figures represents a genuinely brachycephalic, and two only

a sub -brachycephalic type of skull. ^ All these Ligurian departments
are less brachycephalic than many of the departments of Central
France. They are, indeed, marked by M. Topinard as very dark ' '
;

but the table of stature lends little support to the popular theory.
The inhabitants of the Ligurian departments ought to be shorter
than those of the Celtic. They are taller. The Var is 39th com- —
—
paratively high and the Bouches-du-Rhone 19th while the Allier
;

is 78th and the Puy-de-Dome 84th. Taking the other departments

which were inhabited by Ligurians, we obtain results more surprising
still. Haute-Savoie, Savoie, the Isere, and the Drome, have indeed
high cephalic indices— 87-39, 86-25, 85-32, and 84-89— but all four rank
as fair the Vaucluse, which is also fair, has the low index of 81-53
:
;

the Hautes-Alpes and Basses -Alpes, which are dark, have indices
of 84-37 and 83-67 respectively.^ If these departments are repre-
sentatively Ligurian, it is clear that the Ligurians were greatly
mixed if, as is probable, their higher indices are due to an influx
:

of the Celtae or the old round-headed population, the argument

that the Ligurians were brachycephalic breaks down. In the three
Ligurian departments of the coast there have settled, at different
times and in different proportions, Greeks, Gauls, Italians, Jews,
^ Pruner-Bey has written at great length about three Ligurian skulls,
'
'

found in the neighbourhood of Hyeres. But even supposing that they did
belong to Ligurians, there is nothing to be learned from them. The index
of one was 84-50 of another 80
; and the measurements of the third are not
;

stated. See Bull, de la Soc. d'anthr. de Paris, vi, 1865, pp. 458-74 2^ set.,
;

i, 1866, pp. 442-67.

* Antropometriamil., 1896, pp. 139-41, 251.
^ Or Dr. Beddoe is right in saying that Nicolucci found an average
fifteen, if
index of 86-7 in ten old skulls from Liguria. The Races of Britain, p. 28, note.
Cf. Bull, de la Soc. d'anthr., vi, 1865, pp. 259-61 ; 2" ser., i, 1866, p. 84.
* See
p. 305, n. 5. ^ I am using Broca' s terminology.
" The four departments —
the Gard, Herault, Aude, and Pyrenees-Orientales
in which Iberians were mingled with Ligurians, are all marked as very dark,
and, as might have been expected, have comparatively low indices, ranging
from 83-12 to 78-24.
286 THE ETHNOLOGY OF GAUL
Phoenicians, Saracens, and other peoples ^ and it may be that this
;

immigration has depressed the cephalic index below that of the

old Ligurian population but to say more would be to mislead.^
:

Recently, indeed, attempts have been made to prove that the

Ligurians were a dolichocephalic people. M. Pompeo Castelfranco
affirms that two dolichocephalic skulls of the Cro-Magnon type have
been found in the PoUera cavern in the commune of Finale, and
20 similar skulls in the cavern des Arene Candide.' ^ Both of
'

these caverns are in Italian Liguria and M. Castelfranco quotes

;

M. Morelli, the discoverer of the Pollera skulls, who argues that the
whole collection probably belonged to genuine Ligurians, because,
according to Florus ^ and Diodorus Siculus,^ the Ligurians, in the
second century B.C. and in the time of Caesar, dwelt partly in caverns.^
Moreover, Sergi has exhumed 59 skulls, all dolichocephalic, in the
valley of the Po and he argues that if it is true that prehistoric
;
'

Italy was occupied by the Mediterranean race and by two branches

—
Ligurian and Pelasgian of that race, the ancient inhabitants of the
Po valley, now exhumed in those 59 skulls, were Ligurian '.'^ Thus
the 8 (or possibly 15) brachycephalic skulls of Nicolucci and Vogt
are confronted with the 22 dolichocephalic skulls of Morelli and the
59 dolichocephalic skulls of Sergi. If the three groups were all
Ligurian, the only conclusion is that the Ligurians were greatly
mixed. If the collections of Morelli and Sergi alone deserve con-
sideration, the most that we can safely infer is that dolichocephaly
characterized the Ligurians of Italy.
4. And now, who were the Ligurians ? According to Broca, their
descendants are more brachycephalic than any of the palaeolithic or
neolithic peoples and he accordingly concludes that the Ligurian
;

invaders did not appear in Gaul until towards the close of the
Neolithic Age.^ I do not question the conclusion but it is needless
;

to insist that Broca's arguments do not prove it. M. Dechelette^

is inclined, for archaeological reasons, to regard the Ligurians as
akin to the Illyrians and the Thracians. On the other hand, Sergi
dogmatically affirms that they were simply a branch of the same
'
Mediterranean stock from which the Iberians sprang. If he is
'

right, the Ligurians did not speak an Indo-European language and ;

^^
I venture to say that his confidence is at least premature.

Rev. d'anthr., 2° eer., ii, 1879, p. 484.

'
Cf. Atti d. Soc. ligure d. sioria patria,
xl, 1908, pp. 215-8.
2 See Bull de la Soc. d\(nthr. de Paris, 4:" eer., i, 1890, pp. 748-9, vi, 1895,
p. 342, with which cf. C. Jullian, Hist, de la Gaule, i, 125-6. But see also Bull,
de la Soc. d' etude des sc. nat. de Nimes, xxx, 1902 (1903), p. 50.
3 Rev. d'anthr., 4« ser., i, 1890,
pp. 593-9. Cf. Bull, de la Soc. d'anthr., 4" ser.,
i, 1890, p. 749. * ii, 3. ' v, 39,
§ 5.
« The Pollera and Arene Candide skulla are assigned by Signor A. Issel
{Atti d. Soc. ligure d. storia patria, xl, 1908, pp. 113-4, 650-3) to the Neolithic
Age and the Bronze Age. ^ Arii e Italici, 1898, p. 60.
8 Bull, de la Soc. d\nithr., 2* ser., ix, 1874, p. 713 ; Rev. d'anthr., ii, 1873.
pp. 597-8. ® Mayiuel d'archJologie, ii, 7, 16-9.
M. J. Dechelette {Manuel d'archeologie, ii, 27, n. 2) shows that the theory
^•^

that the Ligurians and the Iberians were two branches of the same stock.
 THE ETHNOLOGY OF GAUL 287

My general conclusions are these :

—Ligurians undoubtedly lived in
South-eastern Gaul, where they were found at least as far north
as Bellegarde in the department of the Ain and, mingled more or
;

less with Iberians, in the departments of the Gard, Herault, Aude,

and Pyrenees-Orientales. Most probably they had once occupied
the whole eastern region as far north as the Marne,^ but had been
submerged by Celts and perhaps they had also pushed westward
;

as far as Aquitania. Were it possible to regard the theory of

MM. d'Arbois de Jubainville and Jullian as more than an interesting
hypothesis, we should have to conclude that the Ligurians were
simply the long-headed and short-headed peoples who, reinforced
perhaps from time to time by hordes of immigrants, had inhabited
the whole of Gaul since the Neolithic Age, and of whom the former,
or many of them, were descended from palaeolithic hunters in ;

other words, that they were the same people who, after they had
been conquered by, or had coalesced with, the Celtic invaders, called
themselves Celtae but to say which of them were first known as Ligu-
:

rians or introduced the Ligurian language would be utterly hopeless.

Finally, the little evidence that we possess tends to show that the
people called Ligurians, when they became known to the Greek
writers who described them, were a medley of different races.

IV
THE IBERIANS
L The '
the most complicated and difficult of
Iberian question '
is

all the problems of Gallic ethnology .^ The w^ord Iberian is used '
'

by different writers in different senses and unless one can rely

;

upon being followed with close attention as well as quick intelligence,

one is occasionally obliged to pause and explain which of these senses
is meant.
which has been defended by B. Modestov {Intr. a Vhist. rom., 1907, pp. 126-43)
and Mr. T. E. Peet [The Stone and Bronze Ages in Italy, 1909, pp. 167-74), is
not confirmed by archaeological evidence. Nor, as we shall see, by the evidence
of language.
Mr. Peet, who refers to Sergi in support of his conclusions, argues that all the
neolithic peoples of Europe ' originally belonged to one vast race ', because
—
they had four sepulchral customs in common, painting the skeleton with
red ochre, interment in the contracted position, breaking implements which
were deposited in the grave, and secondary burial. But the contracted position,
which is still adopted by many savages, was almost universal in prehistoric
times ; and both it and the use of red ochre were common in Gaul even in
the Palaeolithic Age. See J. Dechelette, Manuel d'archeologie, i, 204, 301.
^ C. Mehlis {Archiv f. Anthr., xxvi, 1899, pp. 71-94) comes to a somewhat
similar conclusion. With his remarks on pp. 83-5 cf, J. Dechelette, Manuel
d' archeologie, ii, 23, note.
^ The origin of the Iberians was a subject of controversy 2,000 years
ago,
and is so still. Of the ancient geographers some, whom M. E. Philipon [Lea
Iberes, pp. 82-134) endeavours to support, believed that they were an offshoot
from the Iberians of Asia ; others held the contrary opinion others again
;

thought that the two peoples had nothing in common except their name. See
H. d'A. de Jubainville, Les premiers habitants de V Europe, i, 1889, pp. 24-5.
288 THE ETHNOLOGY OF GAUL
2. According to Festus Avienus,^ the Iberians occupied the north-
eastern part of the Spanish peninsula and the coast of the Mediter-
ranean between the Pyrenees and the Rhone. Herodorus of Heraclea,
who wrote in the fifth century b. c, appears to regard the Cynetes,
who lived near Cape St. Vincent, as the most remote of the Iberian
peoples and he too makes their territory extend eastward as far
;

as the Rhone. 2 Scylax, whose Periplus is generally referred to the

middle of the fourth century B.C., Ogives them the country between
the Rhone and the Straits of Gibraltar, though he tells us that, along
the coast of Gaul, west of the Rhone, Ligurians were mingled with
Iberians.^ Scymnus of Chios, who wrote about 90 B.C., locates the
Iberians between the Rhone and the country of the Tartessi, which
was north of Gibraltar and in the lower valley of the Guadalquivir.'^
Hecataeus ^ and Herodotus also distinguish the Iberians from the
"^

Tartessi while Herodorus loosely includes the latter among the

;

Iberian tribes. Eratosthenes, whose views on geography are severely

criticized by Strabo, describes the Spanish peninsula as the Ligu- '

rian {Xtyva-TLKrjv), and in another place as 'that to which Iberia

'

belongs '.^ Strabo tells us that the writers of his day (about 19 b. c.)
restricted the name Iberia to the Spanish peninsula, but that it had
formerly been applied also to the country between the Pyrenees and
the Rh(5ne and he appears to affirm that the name was once con-
;

fined to the eastern part of the peninsula, north of the Ebro.^ He

also remarks ^^ that the people of Aquitania spoke a language akin
to that of the Iberians, and resembled them rather than the Gauls
in appearance. Finally Tacitus ^^ says that the Silures of South Wales,
who were dark and had curly hair, resembled the Iberians.
The conclusions which I draw from the foregoing items of evidence
are these. The name Iberian was probably applied, in the first
' '

instance, only to the people who dwelt between the Ebro and the
Pyrenees. The Iberians once occupied the seaboard of Gaul between
the Rhone and the Pyrenees but Ligurians encroached upon this
;

part of their territorv.^'^ They also probably occupied the whole

eastern region of the Spanish peninsula. But we must bear in mind
that the data are both insufficient and uncertain. Were the various
languages or dialects that have been recognized in the so-called
Iberian inscriptions all spoken by Iberians, properly so called, or
did those inscriptions which have been found in the country of the
Tartessi belong to them and not to Iberians who conquered them ?
Was the language of the Tartessi akin to that of the Iberians ?
Much depends upon where the Iberians came from and that we ;

cannot tell.

'
Ora maritima, 248-53, 472, 552, 608-10. See p. 278, supra
^ Fragm. hist. Graec, ed. Didot, ii, 34, fragm. 20.
^ See Desjardins, Geogr. dc la Gaide rom., ii, 31, and Miiller, Prolegoiuena to
Geogr. Graec. mm., ed. Didot, pp. xxx-li. *
Geogr. Graec. wi?j.,i, 15-7.
' 'lb., i,203-4, vv. 198-202. « lb., 1-2. fr. 4-5, 11-8.
' i, 163, § 1. C'f. Diodorus Siculus, xxv, 10, § 1.
" Strabo, ii, 1, § 40 4, § 8.
;
» lb., iii. 4, § 19.
'0 lb., iii, 1, § 6. " Agricola, 11. ''
See p. 278, n. 6.
 THE ETHNOLOGY OF GAUL 289

3. Li spite of the coldness or positive hostility with which it has

been received by certain scholars, the theory regarding the Iberians
which still dominates the popular imagination is that with which
Humboldt,^ nearly a century ago, electrified the learned world.
The great German scholar collected from the itineraries and the works
of the ancient geographers all the names of towns, mountains, and
the like, belonging to the Spanish peninsula and Southern France :

he succeeded, as he thought, in proving that many of these names

can be explained from the Basque language, and that some of them
are analogous to modern names of places in the districts in which
Basque is spoken and he concluded that the whole peninsula;
;

except those tracts in which Celts settled, and the adjacent part
of France had formerly been occupied by a people who spoke one
language or various closely allied dialects, the modern representative
of which is Basque, and that this was the language of the Iberians.
But Humboldt's theory has of course been modified even by those
who are on his side. All modern scholars are agreed that his know-
ledge of Basque was inadequate, and that many of the etymological
explanations which he gave were forced. M. Vinson, in 1870,
admitted, on the other hand, that many of those explanations were
convincing, and that it was possible, though not proved, that the
Basque country is the last refuge of the Euskarian peoples '.^
'

M. Van Eys also, in 1874, admitted that Basque might possibly be

descended, more or less directly from Iberian.^ But M. Vinson now
—
holds that it is impossible to explain Iberian by which he means the
language whose remains are preserved in the Iberian inscriptions
' '

' '
—
and the legends on the Iberian coins by means of Basque and ;

neither the legends on the coins nor the inscriptions have yet been
interpreted.'* On the other hand, an Austrian scholar, G. Phillips,
M. Luchaire, Emile Desjardins, and the German epigraphist, Emil
Hiibner, accepted the essential part of Humboldt's theory that is ;

to say, while admitting that Humboldt made mistakes in detail,

they all held that Basque was descended from the language, of
which there were doubtless several dialects, spoken by the Iberians
of Spain and Southern France. On the whole,' says Hiibner,^
'

'
Humboldt succeeded, by a most lucid train of reasoning, in estab-
lishing what he had set himself to prove, that is to say that names
of places and of men which are undoubtedly Iberian are to be
explained from the language of the Basques.' ^ The latest successors

'Frujutuj der Untersuchungen iiber die Urbewohner Hispaniens vermittelst der

vaskischen Sprache, 1821.
Eev. de linguistique, iv, 1870, pp. 58, 62.
•^

^ Association franr. pour Vavancement des sciences, Lille,

1874, pp. 544-5.
* Even E. Hiibner, the learned editor of Monumenta linguae Ibericae, dce^
'

not confess to even an inkling of the meaning of more than some half a dozen
words.' See Class. Rev., viii, 1894, pp. 375-9.
^ Mon. ling. I her.,
pp. xxiv-xxv.
^ In the first edition
(pp. 257-62) I examined the theory to which M. J. F.
Blade, one of Humboldt's principal opponents, devoted a large part of his
Etudes sur Vorigine des Basques (1869) ; but I have decided not to reprint my
criticism. M. Blade's view was that the word Iberia', as applied to Spain,
'

1093 XJ
290 THE ETHNOLOGY OF GAUL
of Humboldt, however, have taken up a somewhat different position.
They do not regard Basque as a direct descendant of Iberian as :

M. Vinson remarks, the connexion which they endeavour to estabUsh

between the two is rather analogous to that which subsists between
English and Gothic ;
^ and Schuchardt is disposed to regard Basque
as an offshoot from Aquitanian.^
4. I will begin by quoting a sagacious remark of Mr. Wentworth
Webster. A great deal of the opposition to Humboldt's conclusions,'
'

he says, arises from not observing the difference of the conditions

'

of two very different problems,' —

(1) whether certain names, for
example Iliberris, in Spain and Southern France, which we find in
a Greek and Latin dress in classical itineraries, &c., are Basque ;

and (2) what is the language of the so-called Iberian inscriptions.

'
The two questions are often confounded, as if they must stand or
fall together whereas they are wholly distinct.' ^
;

The inscriptions are divided by M. Vinson into three grou])s,

those of the south-western part of the peninsula, which are written
in Roman characters those of the south, written in characters
;

adapted from the Phoenician alphabet, from right to left and ;

those of the north and east, written in the same characters, but
from left to right. M. Vinson now believes that each group represents
a different language."*
Let us now see what modern scholars have to say in support of

was only a geographical term ', and that no proper Iberian race ever existed ;
'
'
'

but, warning his readers not to strain his meaning, he says {op. cit., p. 155, n. 1 ),
'
J'admets, pour les temps prehistoriques ot pour I'antiquite, la predominance
en Espagne et dans la Claule meridionale de la race brune, de mediocre stature,
et aux cheveux frises ou ondes, a laquelle les ethnologues donnent le nom
d'iberiennc. ... A cette race s'appliqucnt la plupart des indications fournics par
les auteurs classiques sur les populations de I'Espagne et de Ligurie.' I freely
admit that M. Blade has succeeded in proving that Iberia was simply a geo-
graphical expression but this fact required no proof ; and it was a sad
:

waste of time and labour to expend half a volume in establishing it. No fair
critic would, however, allow that this scholar has succeeded in proving that
'
no proper Iberian race ev^er existed', or that those inhabitants of Southern
Gaul whom the ancient geographers called Iberians were Celts or some other
people. The word 'Iberian' was sometimes used to cover the Tartessi and
other tribes with whom the Iberians had mingled ; but I have already warned
the reader to be on his guard against this confusion. The ancient geographers
made mistakes and from the standpoint of modern science the best of them
;

were no doubt sadly deficient. But they were not fools.

But I must try to make my meaning When I follow the
perfectly clear.
ancient geographers and say that the southern seaboard of Gaul stretching
westward from the mouths of the Rhone was inhabited by Iberians, I by no
means intend to imply that the Iberians were a homogeneous people. There
is no such thing as a homogeneous people and if there ever was, tlie time
;

was very remote. When we say that England is inhabited by the English
and France by the French, w^e do not mean that the English and the French
are homogeneous peoples. Nevertheless, the English and the French, mongrel
though they both are, are each characterized by certain features which dilleren-
tiate them, for the most part, from all other peoples.
^ Eev. de linguistique, xl, 1907,
p. 210.
* Die iberische Deklinalion, 1907.
^ Amdemy, 1874, p. 588.
* Rtv. de linguistique, xl, 1907, pp. 5, 211.
 -
^

THE ETHNOLOGY OF GAUL 291

Hiuiiboldt's theory. Desjardins ^ points out the similarity between

Ausk-i (spelt by Caesar Ausci)^ the name of one of the Aquitanian
p3oples, and EusJcara, the name by which the Basques call them-
selves and the well-known Basque scholar, H. Schuchardt,^ derives
;

the latter from the former. Other Aquitanian tribal names, more-
over, namely Tarbelli, Lactorates, Elusates, Bigerriones, Beneharnenses,
lluronenses (which is obviously connected with lluro, the name of an
ancient town in Spain) and especially Oscidates, have une physio '

nomie ibero-euskarienne qu'on ne retrouve nulle part dans la Gaule

chevelue' but, adds Desjardins,^ they have not been identified with
:

Basque words and Luchaire refuses to regard Tarbelli as a word

;
'*

of Basque origin. These words therefore, except perhaps Oscidates,

which Schuchardt"^ assimilates to Ausci, do not amount to proof; and
proof is what the sceptics, MM. Van Eys and Vinson, very properly
require. Now there is a word which occurs, under various forms,
nearly as frequently in Spain and Southern Gaul as Noviodunum in
Celtic and Belgic Gaul. This word is Iliberris. The word iri in
Basque means town and berri means new.^ The letter r is often
' '
;

changed into /. The obvious meaning oi Iliberris is 'New Town ',

or, in Gallic, Noviodunon, which Caesar Latinized into Noviodunum.
There was an llliberris in Roussillon, an Elimberri in Auch, and
an Illiberri in Grenada (the ancient Baetica). Ces trois noms,' '

says M. Luchaire,' suffiraient a eux seuls pour etablir que le basque

'

fut parle jadis dans TAndalousie, en Gascogne et en Eoussillon.

Mais, si I'ancien nom llliberris represente exactement le nom
. . .

euskarien moderne Iriberri, il devait signifier la meme chose (ville-

neuve).' Moreover, the root Hi or iri is found in many other geo-
graphical names scattered over the map of the Iberian peninsula.^
I am bound, however, to point out that even this solitary example —
Iliberris —
is rejected by M. Van Eys. II est vrai,' he remarks,^ '

'
que I est quelquefois pour r mais Hi ne se trouve jamais, autant
;

que nous sachions, pour in'.' On which Mr. Webster very sensibly
comments,^^ Perhaps not in Basque phonetics, but that is not the
'

question, but whether the Greek or Latin authors would have

represented the Basque Iri by the Greek and Latin 'lAAt, 'IX?, Illi . . .

We have ... no right to expect a much greater amount of literal

Geogr. de la Gaule rom., ii, 403.
^

'^
Die
iberische Deklination, 1907, p. 11. M. Vinson {Eev. de linguistique,
xl, 1907, p. 235) urges that certain names a apparence basque ', found in'

inscriptions in the region of the Pyrenees, which have been cited by Schuchardt
{op. cit., p. 10) to prove that Basque was formerly spoken over a Avider area
than it is now, may be only the work of travellers. Granting that this may
be true of some of them, it is significant that M. Vinson virtually admits
that the apparence basque is real.
'
'

^ Geogr. de la Gaule rom., ii. 404. * De lingua Aquitaiiica, 1877,

p. 10.
5 Die iberische Deklination, p. 66.
« W. J.Van Eys, Diet. basque-Jranrais, 1873, p. 208.
^ Bull, de la Sac. des sciences, lettres et arts de Pau, 1875, quoted by Desjardins,
ii, 38-9.
8 8ee Desjardins, ii, 49-50, and G. Phillips, Sitzungsberichtc d. Kaiserl. Akad.
d. Wissenschaften, Ixvii, 1871, pp. 354-400.
» Rev. de linguistique,
1874, p. 7. '"
Academy, \i, 1874, pp. 588-9.
U 2
292 THE ETHNOLOGY OF GAUL
accuracy in the Greek and Latin transcription of Basque names than
we find in the French, German and EngUsh transcription of Oriental
names. Even a word like Punjab ... is written Punjab, Pendjab
and Pandschab, in English, French and German atlases, besides
irmumerable variations.' And I would ask M. Van Eys this question.
Would he deny that in the word Iliherris the element berris is Basque ?
And if it is Basque, as it assuredly is, what does Hi mean if it does
not mean the same as iri ? It will hardly be maintained that Iliherris
is compounded of a Basque word and one which is not Basque.
If it is admitted that herris is Basque, all that MM. Desjardins and
Luchaire contend for is proved.^
But the strife which centred round Iliherris in the years that pre-
ceded the original publication of this book has lately broken out
afresh. The specialists are ranged in opposite camps their warfare
:

is entertaining to the onlooker, even though he suspects that they

will never agree and he is reminded of the controversy which is
;

still raging between Professor Lowell and those other astronomers

who insist that the Martian canals are the offspring of imagination.
' '

And, if 1 am not mistaken, Webster's warning is not less needed now

than when he wrote is there not still a tendency to confuse those
:

—
two questions which are distinct whether Iliherris and other place-
names prove that a language akin to Basque was once spoken in
Southern Gaul and a great part of the Spanish peninsula and ;

whether it was the language of the Iberian inscriptions ?

' '

M. E. Philipon, the author of an interesting volume,^ whose main

object was to give a death-blow to the theory of Humboldt, affirms,
first, that Humboldt's modern supporters have arbitrarily divided
the word Illiherri into Illi-herri, whereas a study of the place-names
of Iberia proves that it ought to be divided into Il-liheri secondly,;

that the first i in the Basque iri is long and alternates with u (iri =
uri), w^hereas the first i of Illiheris (sic) was short and alternated
with e, Illiherri having a variant Eliherri and, lastly, that the
;

Illiberri of Roussillon, so far from corresponding with the Basque

'
It is unnecessary to frame a complete list of the names in Spain and the
eouth of France which Humboldt explains from Basque sources. Two or three
specimens, however, may be given. Acha and aitza, he says, mean 'rock";
and asta is a form of the same word. Livy (xxviii. 22, § 2) mentions Astapa
in Baetica, and (xxxix, 21, § 2) Asta in the country of the Turdetani. (iSee
Humboldt, pp. 23-4). Humboldt and M. Luchaire also regard Calagorris in
Aquitania, which is mentioned in the liinerury of Antonine (p. 457) as a Basque
word. It should be noted that there was a town called Calagurris in Spain
as well as in Aquitania. On the w^hole, it is sufficient to say that, while admitting
the paucity of such words in Portugal, Humboldt (pp. 12G-9) infers from
toponymy that Basque was once spoken over the w^iole Iberian peninsula.
Hiibner also argues (pp. Iviii-lix) from the similarity between geographical names
which have a Basque-like physiognomy and are certainly not Celtic, that one
— —
language split up, it may be, into various diiilects was spoken over the
length and breadth of the peninsula ; and, he says. if Artemidorus. as reported
by Strabo (iii, 1, § 0) appears to say the contrary, Artemidorus was thinking of
the Phoenician, Greek, and perhaps Celtic languages, which were also spoken
in the same area ; and perhaps he may have mistaken dialects for distinct
* Les Iberes,
languages. pp. 17-21.
 THE ETHNOLOGY OF GAVL 293

iri-herri, derived its name from the river on the banks of which it
stood. Schuchardt, who has subjected his arguments to a merciless
analysis,^ affirms that the place-names beginning with //- prove that
//«(berri) not /^(liberi) is the true form, and roundly tells him that
in regard to the river the truth is just the reverse.^
It happens that in one of the inscriptions an Iberian equivalent
' '

of Iliberris is recognized. As transcribed by Hlibner,^ the word is

ilurir but Schuchardt, w^ho holds that the language of the inscrip-
;

tions was akin to Basque, maintains that for ilurir should be sub-
stituted ildurir. This M. Vinson is willing to admit but he cannot
;

understand how either ildurir or ilidurir, which he suggests as an

alternative, could have given birth to the Basque word iliheri^
It would be useless to summarize his arguments or those of his
antagonist for whether the Iberian ilurir or ildurir can or cannot
;

be equated with the Basque iri-herri is not the question ^ all that :

really matters is whether such an equation can be established between

iri-herri and Iliherris ^ and those who believe that it can, will
;

welcome the support of Schuchardt as a set-off against the scepticism

of M. Vinson. The name by which Iliherris was called in the language
which was the ancestor of Basque may only have borne the same
relation to ildurir as Dover to Douvres or London to Londres.
Now supposing that there is sufficient reason for recognizing
a relationship between iri-herri and Iliherris, what are we entitled
to infer ? Only that Basque was once spoken in that part of Gaul
which was inhabited by Iberians and in that part of Spain which
was inhabited by the Tartessi. This, indeed, is no more than might
have been expected for no one who compares the present restriction
;

of the Celtic languages with their ancient extension would contend

that the Basques were always confined wuthin a small region of the
Western Pyrenees. Is it conceivable, asked Broca, that le basque '

ait pu naitre, vivre, se developper isolement dans un coin de I'Europe

sans qu'aucune langue voisine ait eu d'affinite avec lui ? ^ Humboldt '

^ Die iberische Deklination, pp. 1-16. Cf. M. J. Vendryes's review in Rev.

celt.,XXX, 1909, pp. 200-4.
^ See also Miillenhoff's
Deutsche Altertumslunde, 1, 182, 184.
^ Mon. ling. Iber.,
pp. 113-4. * Rev. de linguistique, xl, 1907, p. 212.

* M. Philipon observes {op. cit.,

pp. 77, 185) that Mela (iii, 15, § 1), who
transcribed Iberian names, abandoned in despair the attempt to reproduce
those of the Cantabri, doubtless because the latter were the last survivors of
'

the Euskarian race'. This may show that the language of the 'Iberian'
inscriptions was not related to Basque ; but it does not show that the
'
Euskarian race had not founded Iliberris.
'

M. Philipon also argues (pp. 26-8, 182-93) that names of places and peoples
in the lands that were occupied by Iberians were derived and declined like
Indo-European names, whereas Basque is an agglutinative language, that is —
to say, a language in which two or more words are united in one compound
vocable but M. Vinson, who agrees with M. Philipon in denying the affinity
;

of Iliberris to Basque, regards the languages that were spoken in Northern and
Eastern Spain as agglutinative {Rev. de linguistique, xl, 1907, pp. 21-2).
« MM. Jullian {Rev. des
etudes anc, viii, 1906, pp. 322-3) and J. Vend ryes
{R(^v. celt., XXX, 1909, p. 201) are among those who remain convinced that
Iliberris is equivalent to iri-berri.
' Association Jraui^. pour V avancement des sc, 1874, p. 548.
294 THE ETHNOLOGY OF GAUL
has a similar argument. There are at present eight main dialects
of Basque, which fall into three groups indeed there are probably
;

no two villages where it is spoken exactly alike.^ This infinite

variety, says Humboldt, would be inexplicable if the Basque nation
had not been composed of numerous tribes, dispersed over a vast
area. 2 But if it is true that names of Basque origin are to be found
within the country which we know to have been occupied by Iberians,
it is also true that the Iberian coins which we possess come only
' '

from the eastern, the southern, and the northern paints of the Spanish
peninsula, and from the iieighbourhood of Narbonne, that is to say
from those which the names in question occur, and tliat
districts in
most Iberian ' inscriptions come from the same places
of the '
;

while in Spain Celtic geographical names and Celtic names of men are
only to be found in those districts which we know to have been in-
habited by Celts and Celtiberians.^
Hiibner claims to have proved that the Iberian language, though
doubtless split into various dialects, was really one that it was ;

spoken all over the peninsula and in the adjacent parts of Southern
Gaul which were once inhabited by Iberians that it was not an
;

Indo-European language and that it was the ancestor of modern

;

Basque (reMat una quae de linguae Ibericae vetustae origine et indole

quaestionem absolvere possit lingua Vasconum hodierna).^
But in a review^ of Hiibner's book M. Vinson has directly chal-
lenged this conclusion. Speaking of the most famous of the Iberian
inscriptions — —
that of Castellon he says that none of the readings
'

which have been proposed allow one to discover in the inscription

either Basque or any other known idiom '. Again, when I study '

the Iberian language thus partially restored

—
'
—
by Hiibner's collection
of inscriptions I am unable, with the best will in the world, to dis-
'

cover therein a trace of Basque.' Hiibner himself does not attempt

to demonstrate by philological reasoning a connexion between the
two and in the teeth of expert evidence like that of M. Vinson,
;

I do not see how one is justified in maintaining that the theory which
correlates Basque with the language of the Iberian inscriptions is
proved.^ Even if Schuchardt's equation of iri-herri with ildurir
should be established, we could only infer that Basque was related
^ See Rev. de linguisiique, xlii, 1909, p. 87.
2 Pr liftingder Untersuchungen i'lher die Urbeioohner Hispaniens, &c., p. 128.
^ E, Hiibner, Mon. ling. Iber., pp. xxvi, Iviii-lix, cxvii-cxxi.
*
76., pp. cxli, Iviii-lix. ^ Rev. de linguisiique, xxvii, 1894.
pp. 248-53.
^ The inscription of Castellon, like most of the Iberian inscriptions, is in

peculiar characters, wliicli the Spaniards call desconocidas. — des adaptations."

'

as M. Vinson says, de 1' alphabet phenicien.' According to the reading of

'

Hiibner [Mon. ling. Iber., pp. 155-6), the inscription of Castellon is as follows : —
{z)irtaims. airieimth. sinektn. urcecerere. aurunikiceaiasthkiceaie. ecariu. aduniu.
kduei. ithsm. eosu. shsinpuru. krkrhniu. qshiu. iithgm. kricarsense. ulttlicraicase
argtco. aicag. ilcepuraies. iithsiniecarse. M. V. Stempf, devising a totally
different reading from that of Hiibner, has attempted to prove that the inscrip-
tion is Basque [Rev. de linguistique, xxx, 1897, pp. 97-111): but M. Vinson
{ih., pp. 112-25) has demolished his argument ; and, as he says, Hiibner is our
best guide. [M. Vinson [Rev. de linguistique, xl, 1907, pp. 8, 13) has recently
made a fresh study of this famous inscription. It contains, as he admits, only
 — —;

THE ETHNOLOGY OF GAUL 295

to one of the three known Iberian dialects. M. Jullian,^ indeed,

who is inclined to accept the very novel doctrine that Basque is,
after all, an Indo-European language, has avowed his conviction
that itsmysteries will one day be explained by Iberian and Ligurian ;
'
but since he rightly insists upon our almost absolute ignorance
'

of both these languages,^ I fear that he is too sanguine. Very few

Iberian names can be explained from Basque notwithstanding
:

diligent search no trace of Basque has been found in the British Isles,
where, according to M. Jullian, the Ligurian element was strong ;

and if there is a similarity between certain place-names in the Iberian

peninsula and others in Sardinia and Corsica,^ no vestige of Basque
has been discovered in those islands.*
In the theory of Humboldt there was this much truth. He and.
his successors have proved that in the Spanish peninsula and the
south of France, which were inhabited by Iberians, a language akin
to Basque was once spoken far more extensively than it is now
but that this was the only language of the Iberians (exclusive of the
Celts and Celtiberians, who by Strabo were loosely called Iberians),
they have failed to prove. The conclusion at which M. Vinson
arrives is as follows :
—
Before the arrival of the Phoenicians, the
'

Celts, and the Komans in Spain, there was a number of indigenous

peoples, each with a language of its own. One of these languages
is still represented by Basque another by the Iberian inscriptions.'
:

Perhaps a myriad years hence ethnologists will be busy theorizing

about the races which inhabit our own island. Scattered over its
length and breadth they w411 find coins and inscriptions in the
English language, and names of towns and rivers both English and

one word argtco —

of which the meaning is almost certain but he cannot
;

temptation to attempt to translate the whole. Nor can I resist the

resist the
temptation to quote Professor Tylor's description of etymologists,
—
a race '

not wanting in effrontery' {Primitive Culture, 4th ed., 1903, i, 397).]

1 Hist, de la Gaule,i, 276. 2 /^^^
p_ 274.
^ The instances which Hiibner gives {Mon. ling. Iher.y pp. Ixxxv-lxxxvi)
Celsitani (Ptol., Geogr., iii, 3, § 6) in Sardinia, and Tarabenii and Palania {ih.,
2, §§ 7-8) in Corsica —are not convincing. Cf. Desjardins, Geogr. de la Gaule
rom., ii, 44, and E. Philipon, Les Iberes, p. 117-9.
* The late Canon Isaac Taylor argued in The Origin
of the Aryans, 1889, p. 222»
that Basque was the language of the Ligurians. His anthropological argu-
ments, which are now obsolete, were examined in the first edition of this book
(pp. 263-5) ; but he also affirmed that asia (Pliny, Nat. Hist., xviii, 16 [40],
§ 141), meaning grain of some kind
—
'the one undoubted Ligurian word
[except proper names] which has come down to us '
—
had as yet been
'

only explained from Basque sources'. Whitley Stokes, however, had already
suggested that for asiam should be read sasiam, which would thus be equated
with the Sanskrit sasya, corn (J. Rhys, Lectures on Welsh Philology, 2nd ed.,
' '

1879, p. 8) and if so, the word could hardly have belonged to a non -Aryan
;

language. Moreover, there was a town in North Africa called Tillahari or

Tillihari {Itin. Ant., ed. Wesseling, p. 75), which M. Salomon Reinach {Bev.
arch., 4® ser., xi, 1908, pp. 302-3) is disposed to assimilate to Illiberri; and
if he is right, Tillihari was probably akin to Basque. But has any one ever
contended that there were Ligurians in Africa ? And can any one produce
a single name which resembles Basque in Gallic or in Italian Liguria ? It is
now generally admitted that Ligurian was an Indo-European language. See p.
282, n. 2.
296 THE ETHNOLOGY OF GAUL
Celtic. They will conclude perhaps that English and Celtic were
both spoken throughout the length and breadth of the island. But
will they be able to tell that in the nineteenth century Celtic was
spoken only in certain districts of the north and west, and that
English was spoken in many towns whose names were of Celtic
origin ? May not the explanation be this ? Since the great majority
of the place-names in the countries that were inhabited by Iberians
are not explicable from Basque, it is surely probable that the language
of the Iberians, properly so called, was not related to Basque. Is it
not likely that Basques were in Spain before the Iberians invaded it ;

that they were the founders of Iliberris, of Elimberri, and of lUiberri ;

but that before the time of Caesar they had been driven into the
regions of the north-west ?
5. Let us now approach the question from the standpoint of
physical anthropology. The characteristics of the Iberian type,
according to most ethnologists, were and are short stature, dark hair
eyes and complexion, orthognathism, and a doHchocephalic skull.
But ethnologists commonly use the word Iberian in a loose sense.
' '

'
Iberian is for them simply a term by which they find it convenient
'

to designate a particular type of man. Were these the characteristics

of the Iberians who, as the ancient writers tell us, lived in Southern
Gaul ?

The ancient writers have told us very little about the physical
characteristics of the Iberians. Tacitus implies that they were short
and dark. Jornandes says the same. Strabo implies that their
physical type differed from that of the Gauls, who were tall and
fair. No other ancient writer says anything about the matter. And
when these writers spoke of the Iberians, they meant the mass of
—
the inhabitants of Iberia ', the Spanish peninsula.
'

Let us now see what is the physical type of the modern inhabitants
of the Gallic territory which was once inhabited by Iberians. This
territory, it must be remembered, was also inhabited by Ligurians :

it had been occupied since the Neolithic Age by nameless tribes,

among whom there may have been a brachycephalic element it ;

was invaded by Celts, who, it should seem, were relatively few and, ;

since Caesar's day, there have settled in it, in small numbers, Saracens,
Visigoths, and Jews.^ Excluding, for the present, C^aesar's Aquitania,
the Iberian departments of France are the Bouches-du-Ehone, Gard,
Herault, Aude, Pyrenees-Orientales, Ariege, and Haute-Garonne.
Speaking generally, the inhabitants of all these departments are
very dark those of all, except perhaps the Bouches-du-Rhone, are
;

short ; and those of all, except the Pyrenees-Orientales, which has

the lowest cephalic index of any department in France, are mesati-
cephalic or sub -brachycephalic but their index is considerably
:

lower than that of many departments in the region which belonged

to the Celtae.2 In respect of stature and colouring, these departments
correspond with what the ancient writers tell us about the Iberians :

—
their comparatively high index high as compared with that of

1 Rev. cVanthr., ii, 1879, p. 485. ~ See pp. 30C-7.

 THE ETHNOLOGY OF GAUL 297

—
Spain is possibly to be accounted for by the fact that they were
inhabited by Ligurians as well as Iberians.^ The question is whether
their comparatively low index —
low as compared with that of Central
—
France is to be accounted for by the fact that they were inhabited
by Iberians as well as Ligurians. In the department of the Pyrenees-
Orientales the prevailing type of skull is said to resemble that of
Cro-Magnon. On the whole, the facts would seem to suggest that
in Gaul, as in Spain,- the Iberian type of skull was narrow and ;

it may be that if the Iberian element could be isolated, the indices

would be still lower than they are.
I have come to the conclusion that Basque was spoken in certain
parts, at all events, of Aquitania and I will assume provisionally
;

that either the language of the Iberians was related to Basque or

Basque was once spoken by some of the Iberians of Gaul. The
former assumption is, I need hardly repeat, at least questionable ;

but, as it represents the dominant opinion, we must not leave it

out of account. The departments which correspond with Caesar's
Aquitania are the Landes, Basses-Pyrenees, Hautes-Pyrenees, Gers,
and the western part of the Lot-et-Garonne. M. Collignon has
published an article ^ which gives fuller details than his general
table. While he emphasizes the fact that the majority of the in-
habitants hardly differ from those of Central France, he shows that in
the mountainous districts the index falls below 82. He concludes that
the small dolichocephalic inhabitants were driven by brachycephalic
invaders to take refuge in the mountains and he points out that
;

the districts which show evident traces of having been inhabited

by peoples of the type which ethnologists call Iberian are those
' '

which were occupied by the Tarbelli, the Cocosates, the Iluronenses,

and the Oscidates. Now there is no historical evidence that there
were Iberians in Aquitania at all, except Strabo's statement that
the Aquitanians resembled the Iberians rather than the Gauls and
spoke a language akin to that of the former. M. Jullian * indeed
argues that the Iberians invaded Aquitania, which he believes to
have been already inhabited by Ligurians, and that they took
possession of the district of Bordeaux and of the region which extends
round Oloron, Pau, Tarbes, and Auch. His conclusion is based
upon toponymy. Following Sieglin,^ he regards the tribes whose
names end in -ates as Ligurian, and also the Medulli of the Medoc,
because their name is identical with that of a tribe in the Maurienne ;

while the names Iluro (Oloron), Bigerriones (Bigorre), EUherre or

Elimherri (Auch), and Calagorris (near Toulouse) are, he says, un-
doubtedly Iberian. Bigerriones and Elimherri are, as Schuchardt ^
holds, related to Basque —
and, except Illiherris another form of
;

—
Elimherri not one of the names occurs either in that part of Gaul
which was certainly occupied by Iberians or in the territory which
^ See, however, p. 285. 2 Scottish Review, xxi, 1893,
p. 352, n f.
^ Mem. de la 80c. d'anthr. de Paris, 3- ser., i, 1895, pp. 67-129, especially
pp. 73-4, 78-80, 83, 110-1. .^ - — .

' Hist, de la Gaule, i, 265, n. 1. 277-8.

..--''•^•r [if^D; \if
^^
6 See
p. 283. e
x»ie iberische DeUination, p. 66. y^ .^IJ^l-J-^ "

^/ ET. MiCV.-^-^'^
298 THE ETHNOLOGY OF GAUL
was originally occupied by the Iberians of Spain. The region in
which they are found ])e]onged almost exclusively to the Tartessi
and the language of the Tartessi may have differed from that of the
Iberians. Therefore it seems possible that although Aquitania was
undoubtedly invaded by tribes which had inhabited the country
that was occupied in historical times by Iberians, the invaders did
not all belong to the original Iberian stock. I do not dispute
M. JuUian's view, indeed I think it highly probable but it is hardly ;

susceptible of proof .^ Anyhow it is there were Iberians,

clear that, if

properly so called, in Aquitania, and if the Iberians were dolicho-

cephalic, they were largely mixed with the dark brachycephahc
people of the Grenelle type for, since Caesar's time, there has been
;

no brachycephahc immigration into Aquitania.

It remains to examine the types of the French and the Spanish
Basques. The well-known researches of Broca on this subject have
been supplemented by Dr. Collignon. From a measurement of
60 skulls found in a cemetery at Zaraus in Guipuzcoa, the most
brachycephahc of which showed an index of 83-24, while the average
index was 77-67, Broca concluded that the typical Spanish Basque
skull was dolichocephalic. These skulls exactly resembled six others
taken from three different places in Bilbao and measured by Virchow ^ ;

and the results are confirmed by the measurements made by

A. d'Abbadie of the heads of living Spanish Basques.^ M. Collignon,
though he insists that Zaraus was uneville cosmopolite par excellence
'

depuis plusieurs siecles,' ^ affirms, like Broca, that the Spanish

Basques are dolichocephalic and differ, in respect of head-form, from
all other European peoples.^ The dolichocephaly is due, Broca
points out, to the development of the posterior portion of the skull,
the front part being but slightly developed.
Dr. Telesforo de Aranzadi y Unamuno has written a monograph
on the ethnology of the Spanish Basques. He finds that their average
cephalic index (corrected) is 77-1 that their average stature is
;

from 5 feet 3 to 5 feet 6 inches that 19-2 per cent have blue eyes,
;

2-8 grey, 17-6 green, 18 greenish hazel, 0-8 bluish hazel, and 41-6
brown and that 23 per cent have blonde hair, 13 medium, 40 dark
;

brown, and 24 black."^ M. G. Herve regards them as a mixed race ', '

the principal factor of which is 'the (so-called) Iberian race, the old —
race of Baumes-Chaudes '.^

1 Cf. C. Jullian, Hist, de la Gaule, i, 258, n. 6.

- The only immigrants have been Spanish Basques, Saracens, gipsies, Jews,
Irish, English, and a very few Dutch and Flemings. Rev. d'anthr., i, 1872,
pp. 603-4.
^ Bull, de la Soc. d'anthr. de Paris, iii, 1862,
pp. 579-82 ; Congres internaf.
d'anthr. et d'arch. prehist., 1867 (1868), p. 374.
* Diet, des sc. anthr.,
p. 165.
^ The Rev.
W. Webster says much the same both of Zaraus and of Bilbao.
Journ. Anthr. Inst., v, 1876, p. 10.
« V
Anthr., v, 1894, pp. 286-7.
' El pueblo euskalduna, See also Scottish Review, xxi, 1893,
1889, pp. 27-30.
p. 351.
* Rtv. mensuelle de VEcole d'anthr., x, 1900, p. 220.
 THE ETHNOLOGY OF GAUL 299

The French Basques, on the other hand, are for the most part
sub-brachyceplialic, and their cranial capacity is considerably less
than that of their Spanish brethren.^ From an examination of
732 recruits in the French Basque-speaking cantons M. Collignon ^
—
found a cephalic index of 83-02, 'sur le vivant.' This comparative
brachycephaly, however, he regarded as factitious and accidental,
due to the head being prodigieusement gonfle au-dessus des tempes,
'

precisement au niveau du point ou se prend le diametre transversal

maxillairCj —
caractere absolument propre a cette race.' The type
of face was very long and narrow, resembling an inverted triangle,
the forehead being narrow at the lower part and broad above. The
hair of these recruits was, for the most part, brown and wavy. Their
striking characteristics, which M. Collignon regards as de veritables '

caracteres ethniques ', were le renflement de crane au niveau des

'

tempes et le prodigieux retrecissement de la face vers le menton'.

In the heart of the Basque country the pure type was found in
41 per cent of the recruits examined outside the linguistic frontier,
:

says M. Collignon, the race does not exist. M. Collignon remarks,

further, that both French and Spanish Basques differ, in certain
respects, from all other European peoples and that those features ;

in regard to which the Spanish Basques differ from the French, are
precisely those in regard to which they resemble the Spaniards
generally. Finally, M. Collignon is inclined to assimilate the Basques
to the Berber type.^
Mr. Wentworth Webster gives some interesting particulars regard-
ing the French Basques. He points out that even at Saint- Jean de
Luz, where the infusion of French, Gascon, and gipsy blood must have
tended to darken the original tint, M. Argellies found 22 out of 47
persons who had blue, green, or grey eyes that Arthur Young and ; '

Sir William Napier, who have no scientific theory to support, call

the Basques a fair race that, according to his own observations,
'
;

carried on all over the French Pays Basque, the fair type especially '
—
with blue or grey or very light-brown eyes, with somewhat darkish
—
hair is the distinctive Basque type and that it will be found more
. . .

numerous in proportion to the distance from the neighbourhood of

the sea and the great roads, where the chances of admixture are the
greatest that the French Basques generally are taller than the
;
' '

average Frenchman and than the inhabitants of the surrounding

plains and that even the dark Basques of Spain are not so dark
'
;
'

as their Spanish neighbours '. If,' he concludes,

'
the Basques had '

been originally a dark people, whence could they have obtained their
present fairness ? The infusion of English blood is manifestly
. . .

^ Did. des
sc. anthr., p. 165.
2 L' Anthr., v, 1894, pp. 276-87.
=»
See Bull, de la Soc. d' anthr. de Paris, 4« ser., vii, 1896, pp. 666-71. If
any one should be tempted to regard G. von der Gabelentz's work, Die Ver-
ivandtschajt d. Baskischen mil d. Berbersprachen Nord-Afrikas nachgewiesen,
as supporting this conchision, let him note that in the opinion of M. Vinson
{Rev. de linguistique, xxxviii, 1905, p. Ill) 'no more absurd book on Bas(|ue
has appeared of late years '.
300 THE ETHNOLOGY OF GAUL
inadequate to acroiint for it.' ^ Mr. Webster's observations, combined
with those which have estabHshed the existence of a large blonde
element in the Spanish Basques, may possibly support M. Collignon's
assimilation of the Basque to the Berber type for, as Dr. Beddoe
;

says,^ fairness has been recognized among the inhabitants of

'

Northern Africa for more than 3,000 years '.

Do the French Basques, as described by M. CoUignon, represent
the Iberian type ? To this question only one answer can be given.
If the French Basque type is nowhere to be found outside the limits
of the F'rench Basque-speaking country, it can hardly represent the
type of a people who formed the bulk of the population in the eastern
half, at all events, of Spain and a large part of the population in
Southern France and, as we have seen, there is sufficient evidence
;

that the mass of that population belonged to a quite different type.

But if M. Collignon is right in thinking that the type which he
—
describes is the original Basque type ^ and I do not see how this
—
can be proved it may have characterized a considerable number of
the people whom the ancients called Iberians. This is the view
of M. Collignon himself but he denies that the Basques have any
;

claim to be regarded comme le prototype de I'lbere \^ The type

'

which he describes is obviously different from that which ethnologists

call Iberian
'
and M. G. Herve, who offers certain tentative
' ;

suggestions as to the origin of the Basques, affirms that their physical

characters entitle them to rank as a fourth European race ', different
'
—
from the Iberian or Mediterranean ', the Alpine or Grenelle ',
' ' ' ' ' '

and the fair North European.^ How it arose whether from a fusion—
of dolichocephalic aborigines with brachycephalic invaders, or from
—
some other cause it would be idle to conjecture.
Dr. Beddoe ^ and others have pointed out the close resemblance
which exists between the physiognomy of many of the dark inhabi-
tants of South Wales and that of the Spanish Basques. If we accept
the statement of Tacitus regarding the resemblance between the
Silures and the Iberians, this similarity may perhaps lend some
support to the view that the Spanish Basques represent, approxi-
mately, the type of the ancient Iberians. But it must not be forgotten

^ Journ. Anthr. Inst., ii, 1873, pp. 154-5 v, 1870, pp. 12, 14-5 ; W. Webster,
;

Les d'un etranger au pays basque, 1901, pp. 14-6.

loisirs
- Journ. Anthr. Inst., v, 1876,
p. 24,
^ M. Vinson, remarking justly that there are great analogies between the
French and the Spanish Basque types, says that it is impossible to decide which
of the two is really Basque (Diet, des sc. anthr., p. 165). Is either ?
* Mem.de la Soc. d'anthr. de Paris, 'i'' ser., i, 1895, p. 74. It should perhaps
be mentioned that M. Collignon, noting the absence of the French Basque
type in Beam and Gascony, argues that the ancestors of the French Basques
did not dwell in Aquitania in Caesar's time, but that they were identical with the
Vascones, who, coming from the north of Spain, invaded Aquitania in the sixth
century of our era. If so, the fact is of course no proof that Basque was not
spoken in Aquitania long before. M. Philipon, on the other hand {Le3 Iberes,
pp. 29-31), argues that the ancestors of the French Basques were the Vaccaei,
who accompanied the Vascones.
^ Bev. mensueUe de VEcole d'anthr., x, 1900, pp. 225-7.
^ The Races
of Britain, pp. 25-6.

i
 ;-

THE ETHNOLOGY OF GAUL 301

that, according to Dr. C^olJignon as well as Broca, the Spanish as well

as the French Basques differ, in certain respects, from all other
European peoples. Nor must
be forgotten that a considerable
it

proportion of the Spanish Basques are fair, and that Tacitus used
the word Iberians loosely.
' '

Reviewing the evidence collected by philologists and by cranio

logists, it seems to me probable that the Iberians comprised both
people who spoke, or whose ancestors had spoken,^ Basque and
people who spoke the language or languages of the Iberian inscrip-
' '

tions ; that to observers who had not learned to measure skulls

and knew nothing of modern scientific methods, they appeared to be
homogeneous that the prevailing type was that which is now called
;

Iberian and is seen at its purest in Sardinia, Corsica, and Sicily ;

but that a certain proportion of the whole population may have been
characterized by physical features more or less closely resembling
—
those which the modern Basques French and Spanish possess in —
common, and which, as MM. Broca and Collignon tell us, distinguish
them from all other European peoples. Finally, it seems probable
that the true Iberians were the people who spoke the languages of
the inscriptions, and that Basque was spoken by a people who
occupied Spain and Southern Gaul before the Iberians arrived. But
unless and until the key to those appalling inscriptions is found,
the problem will never be solved.
6. We have next to inquire whether the Iberians occupied any
other part of Gaul besides that which the ancient writers assigned
to them. And first of all, let us consider the evidence of toponymy.
Evidence of this kind there is none, except in that part of France
which corresponds with Caesar's Aquitania.^ Professor Boyd
Dawkins,^ however, argues that an ethnological connexion between
'

Aquitaine and Brittany may be inferred from the remark of Pliny,

" Aquitania Aremorica ante dicta." An ethnological connexion
' '
'

of some sort there may have been but Pliny's authority is worthless.*
;

If the primitive name of Aquitania had been Aremorica —

a Celtic
—
word the fact would tend to show that the primitive inhabitants
of Aquitania had been Celts, which the professor certainly would not
admit and as, in Caesar's time, the name was certainly Aquitania,
;

we should be driven, on the professor's theory, to conclude that the

Celts had been conquered by Iberians The only safe conclusion
!

is that there is no sufficient evidence, linguistic or historical, that

Iberians ever occupied any part of Gaul except the Mediterranean

coast and Aquitania.
7. But was there elsewhere a people physically akin to the Iberians?
This question has been answ^ered already by implication. If the
prevalent Iberian type was long-headed and orthognathous as well

'
Assuming that lUiherris and Illiherri were Basque names, it does not
follow that Basque was still spoken in Roussillon and Baetica in Caesar's time.
^ See Desjardins, Geogr. de la Gaule roin., ii, 44 ; Man. liivj. Ibcr., p. xxvii
Academy, xl, 1891, p. 2(58 and E. Philipon, Les Iberes, pp. 79-81 Cf. pp. 297-8,
; .

supra.
^ Early Man in Britain, Seep. 371.
1880, p. 320. "
302 THE ETHNOLOGY OF GAUL
as short, there were in the Neolithic Age —
we cannot tell what
—
proportion of them survived in Caesar's time people who resembled
the Iberians everywhere in Gaul. Even now there are traces of
the same population.^ The departments of the Charente, Dordogne,
and Haute-Vienne, between the Cher and the Gironde, are inhabited
by a people who are among the shortest in France, dark, though not
extremely dark, and relatively dolichocephalic, their indices bein^
as low as 80*93, 79-17, and 79-70. Dr. Beddoe conjectures that they
represent the primitive dolichocephali, moderately crossed by
'
Celtae and by the blonde Gauls from the north. Professor Bo}d
'

Dawkins goes further, and argues that the reason why Augustus
added the district between the Garonne and the Loire to Aquitania
was that the population were more akin to the Aquitani than to
the Celtae. Twenty out of these twenty-five departments, says
Professor Boyd Dawkins, were very dark. But the professor had
only Broca's researches to guide him for he wrote before MM. Collig-
;

non and Topinard had published their maps. His conclusion is not
borne out by modern statistics. As far as skull-form, stature, and
colour go, the bulk of the modern inhabitants of Caesar's Aquitania
differ but little from the descendants of the Celtae. Does the pro-
fessor mean that the Aquitani were darker than the Celtae, or more
dolichocephalic, or both ? If he means any of these things, the facts
do not support his conclusion. The Aquitani of Caesar's Aquitania,
if we may judge from the indices of their modern descendants, were,

as I have shown, for the most part, brachycephalic or sub-brachv-

cephalic. They were darker than the inhabitants of some of the
'
departments which are not included in Augustus's Aquitania,
Celtic '

and not darker than others more brachycephalic than some, and
;

less brachycephalic than others. The twenty-four departments show

such various results that it is idle to group them together. Their
cephalic indices range from 79-70 to 87-87 and their colouring
;

varies as much. Some are very dark, others dark, and three are
relatively fair.-
But enough of the Iberian question. If my conclusions are wrong,
I have stated the essential facts correctly ; and some critical reader
may discern their true bearing.

See Rev. d\mthr., ii, 1879, p. 46"2.

'

It would be interesting to learn how Professor Boyd Dawkins reconciles

^

his theory with the fact, revealed by the famous inscription of Hasparron,
that the Novem Populi of Caesar's Aquitania sought and obtained in the third
century imperial recognition of their existence as a group distinct from the
rest of Uaul. Sec Desjardins, Geogr. de hi Gaule rotn., ii, 164 iii, 157, n. '2.
;

For the reasons which led Augustus to enlarge the province of Aquitania see
E. Lavisso, Hist, de la France, i. i (by G. Bloch), p. 130.
 '

THE ETHNOLOGY OF GAUL 303

V
THE CELTAE
L Caesar says that that part of Gaul which lay between the
Garonne on the south-west and the Marne andthe Seine on the north-
east was inhabited by a people who call themselves Celts and whom
'

we call Gauls '. He further defines the area inhabited by the

Celtae by tellijig us that it included the territories of the Sec^uani
and the Helvetii ^ and he either excludes or ignores the Roman
;

Province. The Treveri, who inhabited the country round Treves,

are also generally believed to have been Celtae but it is not abso-
;

lutely certain in which group Caesar included them.^

In B. G., iij 30, § 4, Caesar remarks that the Gauls in general
w^ere tall men.^ I shall afterwards examine the value of this state-
ment for the present it is enough to say that by omnibus Gallis
:

he evidently meant the Celtae as well as the Belgae for first, if he;

had been speaking of the Belgae only, he would certainly have said
omnibus Belgis and secondly, while he sometimes uses the word
;

Gain in an extended sense, which includes Belgae, the people whom,

in his opening chapter, he expressly designates as Galli are the Celtae.
Diodorus Siculus, who lived in the time of Augustus, says, in the
chapter in which he describes the manners and customs of the Gauls,*
'
It may be well to emphasize a fact of which many people are
ignorant. The Celts are the peoples who live beyond Massilia, in
the interior of the country, near the Alps and on the north of the
Pyrenees. The peoples settled on the north of Celtica and those
who inhabit all the countries extending along the ocean and the
Hercynian forest as far as Scythia, are termed Gauls.' This passage,
as I shall afterwards show, has given rise to much misapprehension.
Pausanias ^ says that the Gauls were originally called Celts by
themselves and by their neighbours, and that the name of Gauls did
not become general till later.
Strabo frequently speaks of the whole of Gaul as Celtica '

{rj KeXTLKrj) but, like Diodorus, he also uses the word Celtae in
;
' '

a limited sense. The inhabitants of Gallia Narbonensis,' he says,

'

'
were formerly called Celts and I believe that the name was trans-
;

ferred by the Greeks from them to the Gauls in general.' ^ In

another chapter ' he says, quoting Caesar as his authority, that the
Celtae were separated from the Aquitani by the Garonne and the

^ i^. (?., i, 1, §§ 2, 4. 2 ycc pp. 394-5.

'plerumque omnibus Gallis prae magnitudine corporuiu
^ suoi'uiu brevitas
nostra contomptui est.' * v, 32,
§ 1. ^ i, i, ^ I.
ravra fxev vnep twv vefxaixivcuv rrju NapliwyiTiv kuiicpaTHav Kfyojfxev, ovs ol irpoTipuv
^

KfAras <jJv6p.aC,ov' dvo tovtwv d' oifj.ai ical tovs avpLvavras VaKaras KeArous vvu tSjv
'EWrjvojv -npoaayopivOfjvai (iv, 1, § 14).
' iv, 1, § 1.
304 THE ETHNOLOGY OF GAUL
Cevennes, which were at right angles with the Pyrenees, and that
they inhabited the country on the east of the Cevennes as far as
the sea in the neighbourhood of Massilia and Narbo, and as far
as the Alps. All the other inhabitants of Gaul, he says, appealing to the
same authority, were Belgae In two other passages he says that
!

the Belgae inhabited the country between the Rhine, the Loire, the
ocean, and the central plain.^ Needless to say, scholars generally
recognize that, in his distribution of the Celtae and the Belgae,
he blundered grossly .^ It is important to notice that, while he
distinguishes the Belgae from the Celtae, he says that all the inhabi-
tants of Gaul, except the Aquitani, have a Gallic exterior, though they
do not all speak the same language.^
Polybius,* like Caesar, describes the Gauls as tall Virgil,^ Mani- :

lius,^ Silius Italicus,"^ and Claudian ^ describe them as fair or red-

haired and Livy,^ Ammianus Marcellinus,^^ Diodorus Siculus,^^ and
;

Strabo ^^ describe them as tall and fair. Strabo, however, says

that they were less tall and less fair than the Germans and Manilius ;

that they were less red. Polybius and Livy were only describing the
Gallic invaders of Italy but Diodorus Siculus, Strabo, and Ammianus
;

referred to the Gauls of Transalpine Gaul. The qualifying remarks

of Strabo and Manilius only mean, if I am not mistaken, that fairness
and tall stature were less common among the people of Gaul than
among the people of Germany.
2. The observations of MM. Broca, Boudin, Collignon, and Topi-

nard on the stature, skull-form, and colouring of modern French-

men have thrown light on the ethnology of Gallia Celtica but it ;

would be misleading to examine the results of their observations

without first inquiring what admixture of foreign blood the popula-
tion of that country has received since Caesar's day.
Franks settled 'in a thin stratum', as Dr. Beddoe puts it,^'*^ over
most of the country north of Burgundy and of the Loire, except
Brittany and both they and the Visigoths have left traces in
;

Auvergne.i^ Colonies of the Alani, who, according to Ammianus

Marcellinus,^^ were tall and moderately fair, settled near the mouth of
the Sevre. Saxons settled in large numbers in the country round
Bayeux and Caen, between the rivers Orne and Dive, and in the
1 lb., 3, § 1 ; 4, § 3.
-
M. Jullian i, 323, n. 1), who accepts Strabo's testimony,
[Hist, de la Gaule,
says that the westward extension of the Belgae as far as the Atlantic may
also be inferred from Diodorus Siculus, v, 32, § 1, Pliny, Nat. Hist., xvi, 3^
(()4), § 158, and Plutarch, Camillus, 15. I could not accept their evidence if
it contradicted that of Caesar; but I doubt whether any one who examines
the texts will endorse M. Jullian's conclusion. Besides, what about the remain-
ing lands which Strabo assigned to the Belgae ?
3 iv, 1, § 1. " ii, 15, §7.
'"
Am., viii, 659.
^ Astronomiai, iv, 710-11. ' iv, 200. ^ xxii,241.

» xxxviii,17,§3; 21,§9. ">xv,12,§l.. iiv,28,§l.

*'-
iv, 5, § 2 vii, 1, § 2.
; M. Bcrtrand {Les Celtcs d Ics Gaidois dans hs vnllccs
dii P6 et du Danube, p. 30) has given one or two more references but I have ;

given enough.
" Scottish lieview, xxi, 1893, pp. 177-8.
'* UAnthr., xi, 1900, p. 695. '* xxxi, 2, § 21.
 —

THE ETHNOLOGY OF GAUL 305

peninsula of Batz. Burgundians occupied the department of the

Jura, the country on the east of the Saone, the districts of Geneva
and Lyons, and the country to which they gave their name. Britons
invaded Brittany in the fifth century. Normans, described as tall,
whose chiefs had red hair, settled in large numbers in Normandy, and
in less proportions in the north-eastern part of Brittany and on the
southern bank of the middle Loire. Saracens settled near Chatelus
in the department of the Creuse.^ Italians of course settled in the
principal Gallic towns. The influence of Jews and other foreigners
who have settled in France in modern times may safely be neglected ;
but it must not be forgotten that, early in the imperial epoch, the
Ubii, the Vangiones, the Nemetes, and the Triboci German tribes —
which may, however, have been mixed with Celts were established —
on the left bank of the Rhine from the neighbourhood of Cologne to
that of Strasburg.
3. I will now state the evidence which has been accumulated by
the discovery of human remains and by the observations of modern
ethnologists on living persons.
I have already spoken of the prehistoric races who inhabited this
part of Gaul, and have shown that they were still represented among
the Celtae in Caesar's time. As far as I have been able to discover, the
skeletons, unearthed on French soil, that have been generally identified
as the remains of warriors of the Celtic-speaking conquering race, and
were sufficiently well preserved to be susceptible of measurement,
were all found in the basin of the Marne either in Belgic terri- —
tory or close to the common frontier of the Belgae and the Celtae
in the Cote-d'Or, in Franche-Comte, and in Lorraine. They belonged
to tall dolichocephalic men ^ and two of them may be seen in Salles ix
;

and X of the Musee de St. Germain, the former having been buried
with his war-chariot, iron helmet, and long iron sword. But it should
be noted that tumuli and cemeteries of the Iron Age, similar to those
in which these skeletons were discovered, are quite as numerous in
the territory of the Celtae as in that of the Belgae.^
1 now come to the observations which anthropologists have made
upon the population of that part of modern France which was
inhabited by the Celtae.'* For the purpose of comparison, I shall also

^ Ammianus Marcellinus, xxxi, 2, § 21 ; Beddoe, The Races of Britain, pp. 73-

4,78 Diet, des sc. anthr., p. 504 ; Rev. d'anthr., 2^ ser., ii, 1879, pp. 478, 481-3,
;

485 ; Mem. de la Soc. d' anthr. de Paris, 3*^ ser., i, 1895, pp. 113, 115-6 ; Sidonius
Apollinaris, lib. viii, epist. ix, p. 316 ; Notitia dignitatum, ed. 0. Seeck, 1876,
p.204 (xxxvii, 14) Gregory of Tours {Hist. Franc., ii, 272, lib. v, cap.^xxvii,
;

ed.Guadet and Taranne) Capitularia Regum Francorum, ii, 69 (tit. xiv), ed.
;

Baluze Bouquet, Recueil des hist, des Gaules, vi, 50-1

; Wace, Le Roman de ;

Ron, i, 126-7, vv. 2509-10, ed. F. Pluquet.

2 See
pp. 310-1.
^ See A. Bertrand's .4 re/?, celt, et gaul., 2nd ed., 1889, map facing p. 264;
and Rev. de VEcoled' anthr., xv, 1905, pp. 217-30.
Broca's article on stature will be found in Mem. de la Soc. d* anthr. de Paris,
"^

iii, 1868, pp. 147-209 Boudin's on the same subject in another volume of the
;

same periodical (ii, 1865, pp. 221-59) Topinard's on colouring in Rev. d'anthr.,
;

T ser., iv, 1889, pp. 513-30 and Collignon's on head-form in Anthr., i, 1890,
; U
pp. 201-24. Collignon has published a valuable supplementary article in Annales
1093 X
306 THE ETHNOLOGY OF GAUL
refer to the departments which lie within the tracts that correspond
respectively with Belgic Gaul, Aquitania, and the Roman Province.
But there is one fact which must be kept constantly in mind, because,
as we shall see, it introduces an element of uncertainty into the
results of the inquiries of Broca, Collignon, and Topinard, as far
as they relate to particular districts, and, in one case at least, tends
to stultify them. Not only has the French population been affected
since Caesar's time by the infusion of foreign blood, but its distribu-
tion has undergone considerable change.
Roughly speaking, the inhabitants of the country on the north of
a line drawn from Savoy to the extremity of Finistere are blonde or
relatively blonde those on the south of the same line are brown or
;

relatively brown. This general statement is subject to two excep-

tions the inhabitants of the Cote-d'Or are brown
: those of the
;

Charente-Inferieure fair.
Roughly speaking, again, the average height of the population is
certainly greatest in the northern and eastern departments and least
in those of the centre, the south, and the west. This statement is based
upon the returns for thirty years (1831-60) of the number of males
in every department who were exempted from serving in the army
on the ground that their stature was below the minimum height
of 1 metre 56, or about 5 feet 1 J inches and the assumption is that,
;

roughly speaking, the average height of the whole population varied

according to the number of exemptions. Returns have also been
published for the years 1836-40 of the number of recruits in every
department whose height exceeded 1 metre 732, or about 5 feet
7yJ inches. These returns correspond up to a certain point with the
others but there are striking discrepancies here and there. Thus the
;

Doubs, the department which has the least percentage of exemptions,

has also proportionally the greatest number of recruits above 1 metre
732 and the Haute-Vienne, the department which has the greatest
;

percentage of exemptions, has also the smallest number of recruits

above 1 metre 732 but, on the other hand, the Cote-d'Or, which
:

is 2nd in the former list, is only 25th in the latter the Loiret, which
;

is as low as 55th in the former, is actually 14th in the latter and, ;

more remarkable still, the Meurthe and the Vaucluse, which are
respectively 26th and 27th in the former, are 5th and 54th in the
latter.
As regards cephalic index, if we adopt Broca's terminology,
not a single department in the whole of France is dolichocephalic :

but the adjacent departments of the Haute-Vienne and Dordogne

are sub-dolichocephalic, with indices of 79-70 and 79-17 respectively,
and the Charente, which is conterminous with both of them, is
mesaticephalic, showing the only slightly higher index of 80-93 ^ ;

while the adjacent departments of the Calvados (81 '62) and Eure

de Geogr., v, 1896, pp. 156-66 on the departments of the Charente-Inferieure,

Charente, Correze, Creuse, Dordogne, Gironde, and Haute-Vienne. See Addenda.
^ M. Collignon' s measurements were made sur ]es vivants', and therefore
'

all the figures which I am about to extract from his table must be reduced bv 2.
See p. 261. '
 THE ETHNOLOGY OF GAUL 307

(81-34) in Normandy, and of the Seine-et-Oise (81-57) and Seine

(81'59), Charente (80-93), and, in Central France, the Cher (81-77)
and Indre-et-Loire (81-40) are also mesaticephalic. No less than
thirty -two Celtican departments are sub-brachycephalic. The
remaining Celtican departments are all brach3Tephalic, having indices
which range between 85-50 and 88-20. They may be described as
the southern central and the eastern group. Each of them covers
a large tract of country, and is absolutely unbroken and they are
;

separated from one another by the sub-brachycephalic department

of the Loiret. The southern central group comprises the departments
of the Haute-Loire (87-57), Lot (85-92), Lot-et-Garonne (86-66),
Lozere (87-87), Puy-de-Dome (85-53), Aveyron (85-50), and Cantal
(87-08) the eastern group comprises the Meurthe-et-Moselle (85-64),
:

Vosges (86-68), Haute-Saone (87-37), Doubs (86-05), Jura (88-20),

Ehone (86-01), Saone-et-Loire (87-11), and Ain (86-72).
Let us now inquire what relation subsists between the tables of
Broca, CoUignon, and Topinard. Taken in the mass, the people of the
Ain, Aube, Doubs, Jura, Haute-Marne, Orne, and Haute-Saone are
comparatively tall and fair, and brachycephalic or sub-brachy-
cephalic while tall, fair, and mesaticephalic people are found in
;

the Calvados and Eure. The familiar combination of short stature,

dark hair and complexion, and brachycephaly is found in the
Aveyron, Cantal, Haute-Loire, Lot, Lot-et-Garonne, Lozere, and
—
Puy-de-Dome, in other words, in a continuous tract of south-
central Gaul. Medium or short stature, dark hair and complexion,
and mesaticephalic or sub-brachycephalic skulls characterize the
inhabitants of the Charente, Deux-Sevres, Gironde, Indre, Indre-et-
Loire, Vienne, and Haute- Vienne. But various cross-divisions have
to be noted. The tall groups are not always fair for instance, the
;

coast between St. Malo and St. Brieuc is inhabited by tall dark
men with sub-brachycephalic skulls. Nor are the short groups always
dark. Short stature is found in union with a mesaticephalic or
sub-brachycephalic type of skull and fair or relatively fair hair and
skin in the Charente-Inferieure, Cher, Cotes-du-Nord, Creuse, Eure-
et-Loir, Finistere and Morbihan, Loiret, Manche, Nievre, Seine, and
Yonne. Even the statement, generally true, that the short round-
skulled people of Central France are generally dark, must not be
taken too literally for, if their hair is dark, their eyes are often
;

grey or green ^ and on the whole their darkness is much less intense
;

than that of the less brachycephalic people of the Mediterranean.^

^ V. de St. Martin, Nouv. Diet, de Geogr. univ., ii, 1884, p. 345 ; Bull, de la
Soc. d'anthr. de Paris, 3^ ser., v, 1882, p. 151.
2 Bev. d'anthr., 2* ser., ii, 1879, p. 195. There is a significant difference,
though it is not enough to affect this inquiry, between the results of Topinard's
observations and those recorded by Dr. A. Bouchereau {UAnthr., xi, 1900.
p. 696). According to the
latter, the index of nigrescence of the departments
'
'

of the Loire, Puy-de-Dome, Allier, Cantal, Correze, Aveyron, and Dordogne

increases in the order named. Topinard, on the other hand, classes the Puy-
de-Dome and the Cantal among the twenty-two darkest departments, and the
remaining five among the twenty-two which in nigrescence fall immediately
below them.
X 2
308 THE ETHNOLOGY OF GAUL
Recent observations have shown that individuals of Mongoloid
type are to be found on both slopes of the C'evennes,^ where they form
probably between 5 and 10 ])er cent, of the population, and also
in Auvergne and Brittany.^ M. Herve regards them as pure repre-
—
sentatives of what he calls the Ligurian, that is, the Grenelle type.
Finally, it must be borne in mind that, as Topinard remarks, types
are found in every district which do not properly belong to them, and
everywhere individuals are to be seen who combine in their own
persons the characteristics of different types thus the tall stature
:

of the so-called Kymric type is to be found associated with the

' '

features of the short sturdy Auvergnat and the black hair of the
southerner and the fair hair and complexion of the Norman with
;

the aquiline nose of the Jew or the round skull of the Savoyard. In
the towns especially, as might be expected, the characters of different
types are inextricably confused, and the most various characters
are exemplified within single families. Of two brothers, one will
have black eyes and the other blue one will have a long, and the
:

other a round skull the father and mother will be tall, and the
;

children short.^
Is it then impossible to construct a theory out of the available data,
which shall satisfy the reason ? I do not think so. Broca made the
attempt and if the explanation which he offered was not satis-
;

factory, we may be able, with the aid of information which has since
been accumulated, to amend it.
Broca's theory has been so often misrepresented,^ or at least stated
in a way which was sure to mislead those who had not studied his
writings, that it is necessary to be very careful in pointing out exactly
what he said. In a paper called Qu'est-ce que les Celtes ? ^ he
' '

began by remarking that he aimed at putting a stop to the continual

confusion which had been caused by the manifold senses in which
the word Celt had been used both by ancient and modern writers.
'
'

The true Celts, the Celts of history, he concluded, were the mixed
race whom Caesar called Celtae.
Broca's theory regarding the Celtae and their relation with the
Belgae may be put briefly as follows. He remarks, referring to the
official returns for 1831-60, which I have already quoted, that the
height of recruits is greatest in the departments of the north and
east, including parts of the country inhabited by the Belgae, and
least in the departments of the centre, the south, and the west
while between these two groups there is a zone inhabited by men
of middle height.^ On the maps,' he says, on which I have noted
' '

the variations in stature by departments, the line of demarcation

between tall men and short men reproduces exactly the line of demar-
cation fixed by Caesar between Belgae and Celtae. Consequently
the two peoples so clearly distinguished by Caesar differed in respect
1
Rev. de VEcole d'anthr., xvii, 1907, pp. 264-5.
pp. 201-8. Noiiv. Diet, de Geog. univ., ii, .340.
'^
- Ih., viii, 1898,
* I find that what I have said is confirmed by M. Collignon {Bull, de la Sac.

d'anthr. de Paris, vi, 1895, p. 342).

5 Mem. d'anthr., \, 1871, pp. 391-2.
pp. 370-6. ' Ih.,
 THE ETHNOLOGY OF GAUL 309

of stature. The were short, and the Belgae were tall (.

CV^ltac obser- . .

vation teaehes us that blondes predominate in the country once

inhabited by the Belgae and brunets in the country once inhabited
by the Celtae).'^ But of course this division between big and little
men is not to be taken literally. Broca himself corrects it by pointing '^

out that between .the two groups there is a zone inhabited by men
of middle height and this fact, he remarks, is naturally to be
;

explained by a mixture of the two races.^ In support of this explana-

tion, he argues that, whereas in the Belgic departments of the
Seine-et-Oise, Oise, Aisne, Somme, and Marne, the prevalent form
of skull is sub -dolichocephalic, and in the Celtic departments of
the Puy-de-D6me, Cantal, Lozere, C6tes-du-Nord, and Finistere
brachycephalic or sub-brachycephalic, Paris in the intermediate zone
is characterized by a mesaticephalic type.* He admits, indeed, that
'
there are districts in Celtic Gaul in which, owing to a Kymric '

admixture or to the survival of some prehistoric race, a dolicho-

cephalic (or, strictly speaking, sub-dolichocephalic) type predomi-
nates but, he insists, dans la plus grande partie de cette region
;
'

c'est le type brachycephalique qui a prevalu, et il est d'autant plus

prononce, que les populations celtiques ont re9U moins d' elements
. . .

etrangers.' ^
To and enforce his theory, Broca examined four groups
illustrate
of skulls,taken from Auvergne, from Brittany, from Paris, and from
the department of the Marne respectively. The Auvergnat skulls,
125 in number, came from the village of St. Nectaire-le-Haut in
the department of the Puy-de-Dome and from the isolated situation
;

of this village Broca concluded that its population had received

practically no foreign admixture since the Gallic epoch. The value
of this conclusion will be presently examined. The average index
of the whole number was 84-07, and of the male skulls 84-45. The
Auvergnats, says Broca, are the purest, because they are the most
brachycephalic, of the living representatives of the Celtae.^ Whether
they are the purest, the reader will soon be able to judge but, :

according to the measurements of M. Collignon, which had not been

made public when Broca wrote, they are certainly not the most
brachycephalic for the departments of the Lozere (87-87) and Jura
;

(88-20) show higher indices than any of the departments of Auvergne.

The Breton skulls, numbering 136, were collected by Dr. Guibert
from the department of the C6tes-du-Nord. A line, running from
north to south about 20 kilometres west of St. Brieuc, divides the
Celtic-speaking from the French-speaking inhabitants of the depart-
ment. The average index of all the skulls taken from the former
district, which is in Basse-Bretagne, was 81-34, of those of males
81-71 ;while in the French district, which is in Haute-Bretagne, the

^ Bull, de la Soc. d'anthr. de Paris, 2"^

1873, pp. 247-8.
ser., viii.,
2 Mem. d'anthr., i, 392. ^ ^g^
ii, 1873, pp. i-f.!), 689.
d'anthr.,
* Theso statements are not strictly accurate. According to Broca' s own
standard, tlie tScinc-ct-Oiso, Aisne, and Hojnnie are mesaticephalic, and llie
Marne and Oise sub-braciiycephalic,
' licv. d'ant/ir., ii, 1873, p. 590. " Ih., pp. 598, (i()8.
310 THE ETHNOLOGY OF GAUL
figures were 82-9 and 82-54 respectively. In these results Broca sees
a confirmation of his theory. Haute- Bretagne. he remarks, was
indeed successively invaded by Alani, Visigoths, Frisians, Saxons,
and Normans but it is probable that the influence of these successive
;

infusions of foreign blood upon the original population was weaker

than that of the British invasion of Basse-Bretagne, because I'inten- '

site des effets du croisement depend avant tout du nombre relatif

des etrangers qui viennent, a. un certain moment, contracter des
alliances,' ^ &c. Therefore, Broca thinks, it is natural that the
average index of the skulls taken from Basse-Bretagne, which was
invaded, a un certain moment,^ by dolichocephalic Britons, should
'

be lower than that of the skulls taken from Haute-Bretagne, which

was invaded, at five successive periods, by dolichocephalic Saxons
and other Germanic peoples'. '

The Parisian skulls, numbering 125, which were taken from the
'
cimetiere de I'Ouest ', showed an average index of 79 and, as the ;

same number, taken from a Parisian cemetery of the twelfth century,

yielded an almost identical average (79-18), Broca inferred that 79
was approximately the index which would have been yielded by any
group of Parisian skulls in the time of Caesar. While, he concludes,
the Auvergnat group represents the Celtic race almost at its ' '

purest, and that of the Cotes-du-Nord displays the characters of

a mixed population, in which the Celtic element predominated, '
'

but the Kymric formed a considerable minority, the Parisian series

' '

shows the results of a fusion in which the two elements were almost
equally represented.
The Kymric skulls were only 38 in immber, 27 being male and
' '

11 female. The mean index of the former is 78-49, of the latter

77-02 : but Broca assumes that the index of skulls of the 2)urest
'
Kymric type would have been considerably lower for, he argues,
'
;

as the Gauls of the Marne lived very near the frontier of the Celtae,
they must have been fortement croises de Celtes brachycephales '.
'

His conclusions have recently been reinforced by Dr. E. T. Hamy,

who in three papers on the earliest Gallic invaders of the Iron Age ^
The canton of Marans, near the mouth of the Sevre, whicli was colonized
^

by a group of Alani, will serve to illustrate the effect which nia}', in certain
circumstances, be produced by the immigration of an alien people. Marans,
though 25 percent, of its population is blonde, is one of the most brachycephalic
cantons of Saintonge. Yet the Alani were certainly a dolichocephalic people.
M. Collignon concludes that, not having been accompanied by women, the}'
would soon have been absorbed by the original population. In his opinion,
'

what has modified the Auvergnat type in this canton is the tall, fair, Gallic
'
'
'

type. See Mhn. de la Soc. d'anthr. de Paris, 3" ser., i, 1895. pp. 115-6, \'2'2.
• Bull, du Museum dliist. nat., &c.,
1902, p. 178 ; Antlir., xvii, 1906, pp. 7, V
10, 16-7, 25 ; xviii, 1907, pp. 127-39. See also Bull, et mem. de la Soc.
d'anihr., 5*^ ser., ii, 1901, pp. 721-3, and Association fran^. pour Vavanceynent
des sc, 36" sess"., 1907, p. 876. l^nfortunately, although the skeletons
of the Iron Age which have been found in France were very numerous, those
which were suflficiently well preserved to be susceptible of measurement were
very few ; and I have not here taken account of four skulls, found in the
Grotte de Courchapon in the department of the Doubs {LAnthr., xviii, 1907,
p. 128), as they appear to ha\e belonged to the end of the Bronze Age. Bci.
arch., 4:'' ser., xiii, 1909, pj). 228-9.

I
 THE ETHNOLOGY OF GAUL 311

shows that in the Cotc-d'Or dolichocepha,li predoiniiiated that the ;

cephalic indices of the avaikible skulls from the Chatillonnais and

the aiTondissement of Beaune range between 73-1 and 76-59, while
the average stature was 1 metre 757, or just over 5 feet 9J inches ;

and that in Franche-Comte and Lorraine the same race, dolicho-

cephalic or mesaticephalic ^ and tall, had asserted its superiority,
although the old brachycephalous population survived.
Finally, Broca is careful to explain the sense in which he applies
the term race to the Celtae and the Belgae.
' '
These two peoples, he
warns his readers, were not des races primitives et pures '.^ What
'

happened, he believes, was this. The tall, fair, Kymric race took ' '

possession of the whole of Gaul, except Aquitania.^ Between the

Seine, the Marne, and the Scheldt, that is to say in the western part
of the country of the Belgae, the Kymri remained almost entirely '
'

pure until the Merovingian epoch, when they received an infusion of

German blood * the Celtae were a mixed race resulting from the
:

crossing of the Kymri with the small, dark, aboriginal peoples.

'
'

M. Hovelacque,^ Dr. Lagneau,^ and others have endeavoured to

support Broca's theory by referring to ancient writers. Diodorus
Siculus, they argue, distinguished the Gauls from the Celts he :

described the former as tall and fair the inference is obvious that
;

the latter were short and dark. And, says Dr. Lagneau, if Caesar
tells us that the people who dwelt between the Belgae and the
Aquitani called themselves Celts, while the Komans called them
Gauls, the explanation is easy :
—
il semble que les Celtes, confondus
'

sous la denomination de leurs vainqueurs Gaels, aient cherche a pro-

tester encore contre cette denomination en persistant a se donner
encore a eux-memes le nom de Celtes.'
Neither of these arguments has any force. I have already quoted
the passage in which Diodorus distinguishes between the Celts and
the Gauls.''' Anybody who carefully reads the chapters in which he
describes the inhabitants of Gaul will see that he habitually uses the
word TaXoLTai not in the restricted but in the general sense, includ-
ing both TaXdrai and KeXrot'. Therefore when he describes the
Gauls as tall and fair, he means that description to apply both to

The indices varied from 71-5 to 78-8. Dr. Hamy {L'Anthr., xviii, 1907,
1

p. 135) regards the mesaticephali as the result of intermarriage between brachy-

cephalic natives and Celtic invaders.
2 3Iem. d'anthr., i, 394-6
; Bull, de la Soc. d'anthr. de Paris, 2" ser., ix, 1874,
p. 713.
^ Mem. d'anthr., i, 395. Broca believed that the fair invaders came in two
hordes, the second of which entered Gaul more than 1,000 years before
the first. But I need hardly say that no scholar would admit that the first
—
invasion of the Celts I use the word in its strict sense, not in the sense which
—
Broca arbitrarily attached to it took place in 1500 B.C.
* lb.,
pp. 394-6. Elsewhere Broca says that la race celtique '
avait i:res
. , .

probablement autrefois occupe au moins la partie meridionale de la Gaule

belgique, ou la race Kymrique . avait fini par la supplanter en I'absorbant
. .

dans un melange inegal.'

Bull, de la Soc. d'anthr. de Paris, 2° scr., ix, 1874, p. 705.
^

Rev. d'anthr., i, 1872, pp. 613-4 ; Diet, encycl. des sc. medicales, xiii, 1873,
'''

pp. 704-5. 7 See p. 303.

312 THE ETHNOLOGY OF GAUL
the people wlioiu he calls Celts and to the })eople wiiojii he calls
Gauls. If I am wrong in this conclusion, he does not describe the
Celts at all for in describing the people of Gaul he invariably speaks
;

of them as Gauls. But it is certain that my conclusion is right for, ;

after describing ^ the manners and customs of the Gauls, as he calls

them, he says I have now spoken enough about the Celts and will
'

proceed to describe the Celtiberians ' (rjfjLu<; 8' apKoi'iro^? Trepl twv
KeArwv elprjKOTe'; /xeraySt/^ttcro/xev rrjv laroptav Ittl tovs KeXrt/Jr^pa?).
In fact, though he thinks it necessary to w^arn his readers that
the Celtae were geographically distinct from the Galli, he draws
no physical distinction between them ^ and, in conformity with
;

ancient usage, he as a rule uses the two terms indifferently. It is

hardly necessary to add that he was an uncritical writer, and that
the territory to which he restricts the Celtae is not coextensive with
that which Caesar assigns to them. The argument that the CVltae
called themselves by this name in order to mark their protest against
the domination of their Gallic masters, is too fanciful to require an
answer. When Caesar tells us that the people whom the Romans
called Galli called themselves Celtae, he is not speaking of the con-
quered people only but of the entire population of the great central
division of Gaul, including the conquerors, who, by the admission of
Broca and his disciples, were Gauls properly so called.*"^

V, 25-32.
'

Since I wrote these words I have found that Prichard said much the same
" :

'
It is plain,' he remarked, that this distinction laid down by Diodonis is
'

founded on no ethnographical limitation.' Physical History of Manki)id, iii,

1841, p. 49, note.
3 Sir John Rhys {Celtae and Galli, 1905. p. 57 [Pwc. Brit. Acad., vol. ii])
affirming that inscriptions prove that Celtae and Galli spoke different lan-
'

guages in the central district between the Seine and the Garonne', argues
that as we are forced to admit the fact of a cleavage, one cannot help accept-
'

ing the names Celtae and Galli as marking the lines of that cleavage '. But did
not Caesar take pains to make it clear that Celtae and (J alii were identical ? Tlic
inscriptions show that a language which Sir Jolin Rhys regards as Celtic, but
which was certainly not Gallo-Brythonic (see pp. 319-21, infra), was probably
once spoken in the department of tlie Ain and in the department of the Deux-
St'vrcs. These two languages point to tlie existence of two groups of invaders ;

but wliy assume, in defiance of Caesar's statement, that they were respectiveh'-
Celtae and Galli ? Sir John Rhys, as I have remarked elsewhere {Anc. Britain,
p. 438, n. 3), argues from the comparative dearth of '
Celtican '
linguistic
remains that the Celtae were conquered by the Galli. Possibly Goidelic Celts,
or 'Celticans', and the prehistoric peoples with whom they had coalesced,
were conquered by Gallo-Brythonic Celts ;but if so, it remains certain that
both conquered and conquerors, after their amalgamation, were by themselves
collectively called Celtae. How could the name of the conquered supersede
that of the conquerors if, as Sir John Rhys of course maintains, the lan-
guage of the conquerors prevailed ? See also Bev. de synthese hist., iii, 1901,
pp. 32-3; Bev. de VEcole d'anthr., xv, 1905, pp. 21G-30 ; and Bev. celt., xxvii,
1906, pp. 109-10.
Sir John Rh3's formerly argued that a distinction between Celtae and Galli
was established by the distribution of dolmens and tmmdi in Gaul, the former
of which he attributed to the Celtae and tlie latter to the Galli, and also by
a familiar passage in Sulpicius Se\erus {Dialogus, i, 27), who puts into the
mouth of Postumianus, one of the characters in his dialogues, the words Tu vera
vel Celtice aut, si mavis, Gallice loquen, which Sir John translated by '
Speak
 THE ETHNOLOGY OE GAUL 3L3
'
The Bruca's definition of the C^elts oi history
riidical errors in '

arc these :
—
first, he cahnly assumes that no classical writer's testi-

mony, except Caesar's, is of any value and secondly, he fails to see ;

'
that Caesar, by saying that the people who called themselves Celts '

were called by the Romans Gauls ', makes it as clear as noon-day

'

that for him and for his countrymen, as for Polybius and Pausanias,
the words Celt and Gaul w^ere synonymous. Broca admits that
'
'
'
'

the older population of Gallia Celtica was conquered by men of the

same race as the Gauls or Celts who captured Rome. Therefore it is
absolutely certain that the Celtae of Transalpine Gaul were called
after their conquerors. The truth is that Broca, while he aimed at
putting an end to confusion, only made confusion worse confounded.
Moreover, throughout his discussion, he simply ignores the Helvetii,
v/ho, according to Caesar,^ were included among the Celtae.
But these points are of minor importance and I have only touched ;

upon them in order to clear away the misconceptions to which Broca's

theory has given birth. The real question is whether his account of
the ethnology of Gallia Celtica is accurate, in other words, whether —
the bulk of the population in Caesar's time was short, dark, and

Celtic or Gallic, if you prefer it'. I do not suppose that Sir John would
now insist upon these arguments. Celtice aid, si mavis, Gallice simply means
'
in the Celtic or, if you prefer to call it so, the Gallic tongue and
' ;

it is matter of common knowledge that the dolmens were not built by Celts
at all, and, moreover, that the map referred to by Sir John, which purported
to represent the distribution of dolmens and tumuli, was based upon in-
sufficient data.
M. Jullian {Hist, de la Gaule, i, 319, n. 4), who says that he is substantially
in agreement with Sir John Rhys, admits that the ancient writers finally
extended to the Celtae the name Galli, but argues that the reason why the
Celtae did not call themselves Galli was that the latter name properly
belonged to the Belgae. If he means that it never properly belonged to the
Gallo-Brythonic Celts who had amalgamated with and become predominant
among the mixed population which Caesar called Celtae, he differs essentially
from Sir John.
The theories of Sir John RJiys and M. Jullian are both traceable to the writings
of the late Alexandre Bertrand {Rev. cTanthr., ii, 1873, pp. 235-50, 422-35,
629-43 ;Eev. arch., 3« ser., i, 1876, pp. 1-24, 73-98, 153-61 Archeologie celt, ;

et gauL, 2'^ ed., 1889 Les Celtes

; dans les vallees du P6 et du Danube, 1894,
. . .

especially pp. 1-63, 124-5, 132-3, 142, 149-51, 156-7, 177-8 La religion des ;

—
Gaulois, 1897, pp. 8, 10-3, 313, &c.). He maintained that the Gauls, properly
so called les Gaulois or les Galates
' ' '
'
—
formed a group apart from the Celts :

nevertheless he held that the two groups were, in physical characters, identical,
and spoke dialects of the same language. The invaders who captured Rome
were, he maintained, Gauls, properly so called but, long before they threaded ;

the passes of the Alps, a peaceful Celtic population had been settled in the
' '

Plain of Lombardy. Similarly, he argued that Celtic immigrants had '

'

established themselves in Transalpine Gaul five or six centuries before the

true Gauls arrived. He pointed, in support of his theory, to various archaeo-
logical discoveries, and he laboured at prodigious length to prove that it had
the support of Polybius. But the archaeological evidence is now excluded as
irrelevant and it has been demonstrated by M. d'Arbois de Jubainville {Les
;

premiers habitants de F Europe, ii, 1894, pp. 394-409) and by Professor Zupitza
{Zeitschr. f. celt. Philologic, iv, 1903, pp. 3-6) that the terms Celtae and Galli,
as used by the ancient writers, including Polybius, were, generally speaking,
synouynious. Cf. A. Holder, Alt-celtischcr Sprachschafz, i, 892.
' B.G.,i, 1,§4.
314 THE ETHNOLOGY OF GAUL
bracliyccplialic, as it is at the present day. There is certainly
a solid substratum of truth in Broca's theory, for sepulchral evidence
proves that the number of roundheads was considerable but I be- ;

lieve that, partly perhaps from the characteristically French love of

logical precision, partly from the want of sufficient data, he pushed
it too far. In his time the supplementary tables of MM. Collignon
and Topinard had not been published.
There are several departments the returns for which appear to
clash with or, at all events, are not to be accounted for by Broca's
theory. Thus Finistcre and the Morbihan are both in the lowest
class as regards stature and yet the former ranks as comparatively
;

fair, the latter as very fair in Topinard's list. Again, the eastern
group of departments presents features which neither Broca nor any
other ethnologist has satisfactorily explained. The people of the
Doubs, Jura, Haute-Marne, and Haute-Saone are respectively 1st,
3rd, 4th, and 9th in the table of stature the first three appear in the
:

group which Topinard styles les plus blonds ', and the fourth is
'

comparatively fair yet all four are extraordinarily brachycephalic.

;

Dr. Collignon^ and others explain this combination as the result of

a cross between the tall fair Burgundians and the short, dark,
brachycephalic people whom they conquered and Dr. Collignon
;

remarks that the modern Lorrainers are much shorter than genuine
blondes like the Scandinavians. But the Burgundians were extremely
dolichocephalic and it is difficult to see how a cross between a
;

dolichocephalic and a brachycephalic race should result in a popu-

lation even more brachycephalic than the most brachycephalic
of their ancestors. I can only suggest that, if Dr. Collignon had
measured a relatively large number of skulls in the Burgundian
departments, their indices might have turned out lower.^
Dr. Guibert has made an elaborate investigation into the ethno-
logy of the Cotes-du-Nord,^ which was supplemented and corrected
by Dr. Collignon.* They found that the conditions of the problem
were far more complex than Broca had supposed. For example,
the average stature is greatest in the maritime districts, not of the
Breton-speaking but of the French-speaking portion and it is in ;

these very districts that the proportion of blue eyes and fair hair is
lowest. Indeed, the researches of Collignon, although they do not
disturb Broca's general theory, point to such a medley of races as
must make it impossible to draw any valuable inference from the
ethnology of the C6tes-du-Nord regarding that of Gallia Celtica as
a whole.
But it was in his treatment of the ethnology of Auvergne that
1 UAnihr., i, 1890, p. 213. See also Bev. mensueUe de VEcole d'nnthr., 1896,
p. 218. where it is pointed out that les croisements ethniques out souvent pour
'

resultat non pas la fusion, mais I'echange des caracteres '.

^ M. J. Deniker {Association franc, pour V avancement des sc, 36'' sess"., 1907,

1 ^ partie, p. 277) says that people of the tall, dark, brachycephalic type which

he calls Adriatic (see p. 325, infra) are to be found in Alsace, Lorraine, and
the region of the Vosges.
^
Bull, dc la Soc. d\inthr. dc Paris, 2" scr., v, 1870. pp. 252-65.
* lb., 4." st^r., i, 1890, pp. 750-83.
 ;

THE ETHNOLOGY OF GAUL 315

Broca was least fortunate. Dr. A. Bouchercau^ has recently shown

that the Auvergnats in general are extraordinarily mixed,'^ and his
researches suggest that the villagers of St. Nectaire-le-Haut, whom
Broca singled out on account of their supposed purity, may have
been very different from their Gallic predecessors. It is true that
the number of ancient skulls which Bouchereau has been able to
collect is very small but among those of the Neolithic Age and the
:

Iron Age he could not find one specimen that was not dolichocephalic
and in the cemetery of St. Floret he found that while the mean
index of the most recent skulls was 83-73, that of the lowest layer,
some six centuries older, was less than 80.
At allevents, it would seem that Broca did not sufficiently allow
for the influence which the various Germanic conquerors of Gaul must
have exercised in modifying the physical characters of the people
among whom they settled.^ It is probable that those conquerors
were, generally speaking, tall, fair, and dolichocephalic and it is ;

precisely in those parts of France in which they settled always —

excepting the inadequately explained brachycephaly of Burgundy,
—
Franche-Comte, and Lorraine that the closest approximation to
those characteristics is found among modern Frenchmen. Who shall
decide what proportion of the result is to be ascribed to these Ger-
manic peoples, and what to the tall fair Gauls of Caesar's time ?
When Broca and his disciples maintain that the Celtae were, for the
most part, markedly different in physical characters from the Belgae,
they forget or disregard the evidence of Strabo,^ who says that the
'
Belgae and the Celtae participated in the same Gaulish exterior
'

— that is to say, that they were both alike tall and fair and the —
evidence of Caesar, who says that omnes Galli, that is to say, Belgae
and Celtae alike, were tall. On the other hand, if we attempt to use
this evidence against Broca, we are confronted by the fact that a
large proportion of the Celtae and some certainly of the Belgae were
demonstrably short and dark, and by the fact that we cannot tell
whether Strabo was only repeating what he had heard or read, and
whether Caesar did not exaggerate, like many unscientific travellers,
the prevalence of those characters by which he was most impressed.
Furthermore, in the well-known passage ^ in which he tells us that
the inhabitants of Vesontio (Besan9on) frightened the legionaries by
telling them of the huge stature of the Germans, he certainly seems
to imply that tallness was much less common among the people of
Gaul than among the people on the east of the Rhine.
^ VAnthr., xi, 1900, pp. 691-706 ; Association fran:. pour V avancement des
sc„ 37« sess"., 1908, p. 700.
^ It is remarkable that the red-haired individuals whom lie noticed were not,

as might have been expected, dolichocephalic, but the most brachycephalic of all.
^ In one of his articles, however {Mem. d'anthr., i,
287), ho lays great stress
upon that influence ' nous voyons
: entrelaMeuseetle Rhin
. . . les Kymris . . .

presque entierement germanises par suite de la predominance de la population

franke . entre la Seine, la Mouse ct I'Escaut, Ics memos Kymris, rcstes a
. .

peu pros purs jusqu'a I'epoiiuo mcrovingienne, et meles dcpuis lors, en propor-
tion notable, aux conquerants germaniques,' &c.
iv, 1, § 1. o
y^. (,'.,i, 39, §1.
 316 THE ETHNOLOGY OF GAUL
I will )iow try to use the data which I have collected. First, it is
undeniable, and is admitted by every one who has studied the
question, that a large proportion of the people whom Caesar called
Celtae were short, dark, and brachycephalic that is to say, that
;

they differed essentially in physical characters from the Gauls or

Celts, whom the ancient writers with one voice describe as tall, and
whom nearly all of them describe as fair. It is also probable that
a considerable portion of them, in certain districts, were short, dark,
and relatively dolichocephalic. Furthermore, M. Collignon has re-
marked that la brachycephalic semble croitre avec I'altitude du
'

lieu, et que ses maxima repondent aux massifs montagneux les plus
eleves,' ^ from which he concludes that the older inhabitants were,
in some measure, driven out of the more fertile tracts by blonde
dolichocephalic invaders. So much modern research has added to
the knowledge which we derive from the ancients. But of that
research the statistics that have been collected regarding the modern
population are not perhaps the most valuable part. They confirm
the evidence of history and of prehistoric graves but they do not
;

add to it much. The question that remains is why the existing traces
of the tall fair Gauls in the country (exclusive of Helvetia) that was
once occupied by the Celtae are so slight. M. d'Arbois de Jubainville
offers a peculiar explanation.^ The Gauls, he says, w^re very few
and even those few were almost annihilated in the Gallic war. The
true Gauls, he insists, were merely the aristocratic caste whom Caesar
describes^ as equites and a passage in B. 6^, vii, G4, § 1, proves that
;

they only numbered 15,000 fighting men, from which it may be

inferred that the whole population, including women and children,
was not more than 60,000 There is no force in this argument.
!

First, it entirely ignores the statement of Caesar that the Gauls

generally, including the Celtae {ornnes Galli),weie tall, as w^ell as the
statements of Strabo and Diodorus, to which I have already referred.
Secondly, it assumes that Caesar ahvays uses the word equites in the
—
same sense, that the Gallic cavalry w-ere entirely composed of the
'
knights ' or nobles whom Caesar describes in his general account of
the manners and customs of the Gauls whereas he expressly says
;

that Dumnorix maintained a large number of equites, who could not

have been knights at his own expense.^ Thirdly, it leaves out of
' ',

account the fact that, even in the last year of the Gallic war, there
were, at the very least, 8,000 equites ^ besides those mentioned in
'
B. G., vii, 64, § 1. Fourthly, it assumes that these same knights '

had for centuries preserved their organization as a close corporation,

—
and that none of the plebs belonged to the tall fair race, an assump-
tion which is contrary to all reason. Moreover, there is no evidence
that the equites were almost annihilated in the Gallic war
' '
and it;

A. Bertrand, La Gmih avcmt les Gavhis, pp. 323-4. Cf. AnnaJes de gcogr.,
^

V, 1896, p. 192. Dr. Bouchereau's researches, however, would lead one to

suspect tliat Collignon's generalization may not be universally applicable to
Caesar's time.
^ Les premiers hahilanis de C Europe, ii, 1894,
pp. 7, 10.
3 B. G., vi, 13, * lb., i, 18, § 5. lb., vii, 76, § 3.
§ 3
^
15, § 1.
;
 THE ETHNOLOGY OF GAUL 317

is hard to understand why the cavalry should have been annihilated

any more than the infantry.^
By way of explaining the existing rarity of blondes in Celtican
Gaul, Dr. Beddoe remarks ^ that we do not know exactly what the
ancient writers meant by certain shades of colour, nor can we esti-
mate the personal equation of those of them who wrote from personal
observation. What is darkish brown,' he remarks, to most Eng-
' '

lishmen would be chestnut in the nomenclature of most Parisians

... an ancient Roman might have called it
. . . eYenJlavus.^ This
. . .

argument is to my mind invalidated by three facts the meaning :

of flavus and rutilus is fixed by numerous passages ^ the ancients ;

described the Germans, whom Dr. Beddoe admits to have been

genuine blondes, by practically the same epithets as the Gauls and ;

Diodorus * describes the extreme fairness of Gallic children and the

gradual darkening of their hair as they grow older in terms which
would be applicable to Norwegian children in the present day :^ ra 8e —
Trat^ta irap avrots Ik yeveTrjs V7rdp)^€i TroXta Kara to irXudroV Trpo/Saivovra
Sk rat? r]XiKiaL<s et? to twv TraTepwv \pii)fxa TaL<5 )(poaLS /xerao-^^v^jaaTi^erat.
Without resorting to M. d'Arbois's explanation or to Dr. Beddoe's,
it is not difficult to account satisfactorily for the general disappear-
ance from Gallia Celtica of the type which Caesar, Strabo, and
Diodorus describe. Men of this type were, there can be little doubt,
less numerous, even in Caesar's time, than the dark races and even ;

then intermarriage had probably been going on for generations

between them and the peoples whom their forefathers had conquered.
The losses which they suffered during the Gallic war were no doubt
disproportionately heavy it is, as Broca has pointed out, the con-
:

stant tendency of a mixed race to revert to the type which was, at

the outset, numerically in the ascendant ^ and, as Penka ; has '^

1 Mr. W. H. (Bullock) Hall {The Romans on the Riviera, &c., 1898, p. 4)

found reason to hope', after an interview with M. d'Arbois, that he was
'

'
prepared to modify his extreme views '.
The Races of Britain, p. 3.
^

See Forcellini, Totius latinitatis lexicon, ui, 1865, pp. 98-9; v, 1871, pp.
^

279-80. « V, 32, §2.

^ In the public gardens of Clermont-Ferrand, which is in the typically

(?)
'
Celtic department of the Puy-de-D6me, almost every child whom I saw was
'

fair,and many had flaxen hair. H. Martin remarks {Rev. d'anthr., 2® ser., ii,
1879, pp. 194-5) that immense numbers of children are born blonde, and
darken, which, he argues, proves that the Gallic element among the Celtae
'
'

was larger than is commonly supposed. The members of the Association

franfaise pour V avancement des sciences, when they met at Clermont, made
an excursion into the mountains and were astonished to find an almost pure '

race with fair hair, blue eyes, and white skin' {UAnthr., xi, 1900, p. 695).
Dr. A. Bouchereau {ih., p. 698) affirms that a blonde element also exists in the
departments of the Creuse, Correze, Cantal, and AUier.
Broca lays it down as an axiom that when two races, numerically very
**

unequal, mix, the less numerous is rapidly absorbed, and that hybrids tend to
revert to the type of the more numerous. 8ee Rev. d'anthr., ii, 1873, p. 619.
Again, Dr. Beddoe remarks {Scottish Review, xix, 1892, p. 416) that tall fair
children cannot stand the insanitary conditions of urban life as well as short
dark ones. See also L. Vanderkindere, Recherches sur V ethnologic de la Belgique,
1872, p. 62.
' See The New Princeton Review, v, 1888, pp. 12-3. Cf. p. 259, supra.
318 THE ETHNOLOGY OF GAUL
shown, the tall blonde races have, owing to climatic reasons, never
been able to maintain their original proportion in C'entral or Southern
Europe. When we consider all these things, and bear in mind further
that, since Caesar wrote, racial amalgamation has been going on for
nearly 2,000 years, it is not to be wondered at that so few specimens
of the classical Gallic type are to be found now in the country which
corresponds with Celtican Gaul.
But we must beware of exaggerating the rarity of the type. The
three tables published by MM. Broca, Topinard, and Collignon are
apt to suggest to an unwary reader that the people of this or that
department are uniformly short, dark, and brachycephalic. Any
traveller who kept his eyes open would soon find out that this was
a mistake. The tables only profess to give general results and the ;

numbers of individuals on which those of MM. Topinard and Collignon

were based were comparatively small. The well-known traveller,
William Edwards,^ saw numerous examples of what he called the
'
Kymric type between Geneva and Macon, in Burgundy, and
'

between the mouths of the Somme and the Seine. The charac-
teristics of thistype he described as la tete longue, le front large et
'

eleve, le nez recourbe, la poi.nte en bas, et les ailes du nez relevees, le

menton fortement prononce et saillant, la stature haute '. I saw it
myself fourteen years ago, combined with fair hair and complexion,
in two carters, near Royat, in the department of the Puy-de-Dome.-
We must not overlook the Helvetii. Caesar, it will be remem-
bered, includes them among the Celtae. Now, although, as we have
seen,^ the older inhabitants of Helvetia, like those of the rest of
Gallia Celtica, were a medley of roundheads and long-heads, and
doubtless, for the most part, dark, there is abundant evidence that
the Helvetii, properly so called, were tall powerful Gauls. In the
Hallstatt Period tall dolichocephalic invaders appear to have sub-
dued the lake-dwellers in the period of La Tene, corresponding with
;

that which in our own island is called Late Celtic, the Helvetii, who
belonged to the same stock, made themselves masters of Switzerland.^
It is universally admitted that a certain proportion, great or small,
of the people whom Caesar called Celtae were tall and fair, and were
ethnically identical with or akin to the Gauls who captured Rome.
Were these men identical in race with the tall fair Belgae, and did
they originally belong to one ethnical group or to two ?
Perhaps the evidence of language may help us. Of the language
or languages that were spoken in Gaul nothing remains except names
of men and of peoples, geographical names, a few names of things

^ Les caracteres physiol. des races humaines, 1829, pp. 58, 61, 66-7.
2 M. P. Gouy {Rev. de VEcole d'anthr., xvii, 1907, pp. 266-8) remarks that
many persons of Kymric type are to be seen in the Ardeche. They are
'
'

often fair and generally dolichocephalic ; many of them exceed 1 m. 80 (nearly

5 ft. 11 in.) in height, and some reach 1 m. 90 (nearly 6 ft. 3 in.). Dr. Collignon
{Annales de geogr., v, 1896, p. 164) tells us that the same type is noticeable
in the rich district round Limoaes. See pp. 331-5, iiifm.
=^
See p. 275.
4 Bull, et mem. de la Soc. d'anthr., 5" ser., viii, 1907,
pp. 226-7 ix, 1908.
;

pp. 743-5.
 THE ETHNOLOGY OF GAUL 319

which passed into Latin, and a few inscriptions.^ Here it may be

well to state certain elementary facts of Celtic phonology, although
I dare say that to most of those who may read these pages they are
already familiar.
The Gauls, for the most part, including the Belgae, and the Bry-
thons, or Celts who invaded Britain, from whose dialect modern
Welsh is descended, are commonly called the P Celts ^ while the
;

Goidels, whose dialect was the ancestor of Gaelic, Irish, and Manx,
are known as the Q Celts. The reason of this distinction is that the
Gauls and Brythons changed the original sound of qu into jOj while
the Goidels retained it, and in the sixth century of our era modified
it into c? It has been afhrmed, however, on the evidence of the
'
formularies of Marcellus of Bordeaux, who was physician to
'

Theodosius the Great, that some of the Western Gauls in the fourth
century spoke a dialect which was akin to Goidelic * and Sir John ;

Rhys and Mr. E. W. B. Nicholson ^ regard the words Seqiiani and

Sequana (the Gallic name of the Seine) as proving that this dialect
was not confined to the west but, as we have seen,^ M. d'Arbois de
:

Jubainville refuses to admit that these names are Celtic and he ;

contemptuously denies that the formularies are to be taken seriously.''

Sir John Rhys^ and Mr. Nicholson ^ also infer from certain inscrip-
tions found in the departments of the Ain and Deux-Sevres, which
probably belong respectively to the first and the fourth century of
our era, that a dialect akin to Goidelic was spoken in those localities :

but here again M. d'Arbois dissents ^^ and he remarks that an inscrip-

;

tion found at Geligneux in the department of the Ain contains a word,

petrudecametos,^^ which belongs to the language of the P Celts. Sir
John Rhys urges that the presence of monuments in the language oc-
'

cupying the subordinate position maybe taken as evidence presump-

tive of its being the vernacular in the immediate neighbourhood':^"^

^ See Desjardins, Geogr. de la Gaule rom., u, 580, and Sir J. Rhys's papers
in Proc. Brit. Acad., vol. ii.
2 Mr. E. W. B. Nicholson {Keltic Researches, 1904) has attempted to prove

that the Belgae were Goidels, a theory which is refuted in my

Arte. Britain,
pp. 449-52.
3 SeeJ. Rhys, Celtic Britain, 3rd ed., 1904, pp. 215-6.
* Zeuss, in the preface (dated 1853) to his Grammatica Celtica (pp. xxxii-
xxxiii, ed. 1871), denied that there were any Celtic words in the formularies;
but, according to Prof. W. K. Sullivan (E. 0' Curry's Manners and Custoins of
the Ancient Irish, i, 1873, p. Ix), he afterwards '
fully admitted the Celtic
character of the Marcellian formulae in a letter to Jacob Grimm '. Cf. J. Rhys,
Celtae and Galli, 1905, p. 50, note {Proc. Brit. Acad., vol. ii).
5 Keltic Researches, pp. 6, 128, 149, 167. « See
p. 281.
' Rev. celt., xxv, 1904, pp. 351-3 ; xxvii, 1906, pp. 107-8.
^ Celtae and Galli,
pp. 1-2, 46, 55-64. See p. 282, n. 2, supra.
^ Keltic Researches, pp. 116-53.
i«
Rev. celt., xxv, 1904, pp. 351-3 ; xxvii, 1906, p. 107.
^^ Corpus inscr. Lat., xiii,
2494. M. Loth regards the word petrudecametos
as a proof that the other inscription found in the same department the —
—
famous calendar of Coligny cannot be Ligurian, But M. d'Arbois is never
at a loss for an answer :
—
may not the calendar have been written in an archaic
language, just as the ordinal used by Catholic priests is in Latin ? See
'
'

Rev. cell., xxx, 1909, p. 214. '^

q^h^^^ ^^^^^ g^fjn^ p^ 62.
 320 THE ETHNOLOGY OF GAUL
but a bearing a Goidelic inscription, has been found at Sil-
pillar,
chester,where the vernacular was undoubtedly Brythonic ^ and the ;

explanation may be that the inscription was the work of a stranger.

M. d'Arbois,^ moreover, unlike Sir John Rhys, maintains that when
the Celts first invaded Britain, the Celtic language everywhere was
one and the same according to him, none of the Celts had then
:

changed q into f, but that change was made at a later date by the
Celts who conquered Gaul, and some of whose descendants after-
wards conquered Britain.
But, as the reader will have gathered from an earlier paragraph,^
the antagonism between M. dArbois and Sir John Rhys is less sharp
than it might at first sight appear. The former regards Sequani,
Sequana, and the language of the inscriptions as Ligurian the latter ;

is now inclined to believe that Ligurian was the Continental idiom

'

akin to Goidelic as Gaulish was to Brythonic '.* On what does he

base this assimilation of a language of w^hich we know hardly any-
thing to a language of which, as he himself says, we know little ? ^
Simply, it would seem, on the fact that certain inscriptions which
he is greatly inclined to regard as Celtican \^ but which M. dArbois
'

refuses to regard as Celtic at all, have been found on both banks of

the Lower Rhone in territory w^hich M. Jullian has shown to have
been occupied by Ligurians. But M. Jullian, who by emphasizing
this fact (which, by the way, was already public property) has in
Sir John's opinion gone some way to solve the Celto -ligurian ques-
'

tion V insists that before the fourth century B.C. the Ligurian lan-
guage had been supplanted by Celtic.^
The conclusion appears to be this. It is admitted that in Caesar's
time the bulk of the Celtae spoke a language Gaulish or Gallo- —
—
Brythonic w^hich was also that of the Belgae, and therefore that,
as far as we can judge from linguistic evidence,^ there was no ethno-
logical distinction between the Belgae and the conquerors who, as
'
I shall afterwards prove,^^ had imposed that language in Celtican '

Gaul. It is admitted that at a later time inscriptions were erected

in a language which, whether it was Ligurian or Celtican or both, ' '

was different from Gaulish for it retained qu and Indo-European p,

;

both of which Gaulish had lost. If, as has been suggested,^^ the in-
scriptions were written in a dead language, still it had been spoken
in Gaul at an earlier time. If it was Ligurian, it throws no light upon
Celtic ethnology unless we are to adopt the desperate solution that
Ligurians and Celts, or rather Celticans ', were linguistically one.
'

If it was Celtican and not Ligurian, it points to the existence of an

earlier influx of Celtic conquerors who differed in speech from the

^ See F. Haverfield, The Romanizaiion of Roman Britain, 1905, p. 29 {Proc.

Brit. Acad., vol. ii).
2 Les premiers habitants de V Europe, ii, 1894, pp. 255-82 ; Les Celtes, pp. 17-9.
3 See pp. 281-2.
* Celtic Inscriptions in France and Italy, 190G, pp. 99-100 {Proc. Brit. Acad.,
6
vol. ii). 5
/^.^ p, 79, 11^^ p. 81. 7
ijj^^ pp, 77_8i, 99.
^ Melanges H. d'Arhois de Jnhainville, 1900, p. 108.
9 See J. Rhys, Celtae and Galli, p. 60. '« See pp. 328-31.
1'
Seep. 319, n. 11.
 THE ETHNOLOGY OF GAUL 321

later Gallo-Brythons. If it was Ligurian and not Celtican, and if

accordingly the Celtic language, when the Celts first invaded Gaul,
was everywhere the same, the Goidels and Brythons of later times
sprang from the same stock. And even if the first invaders of Gaul
were Goidels, there is no reason to suppose that they were physically
different from the Brythons who conquered them.
It is certain that the conquering people among the Belgae and the
conquering people among the Celtae were, ethnically, one and the
same.i They spoke the same language or dialects of the same lan-
guage their physical features are described by the ancient writers
:

in terms which are virtually identical remains of a common civiliza-

:

tion are found in the territories of both peoples. The chief reason
for raising the question which I have been discussing is that, accord-
ing to the Reman envoys who visited Caesar in 57 b. c, the Belgae
were of German origin (ortos a Germanis) ^ and we shall presently
' '
;

see that this statement does not shake the orthodox conclusion, that
the Belgae, or rather the Belgic conquerors, were, for the most part
at all events, a Gallic and Celtic-speaking people.
One word more, and I bid adieu to the Celtae. It has been asserted
that the word " Kelt " has long ceased to have any ethnical signifi-
'

cance '.^ If so, the reason is that writers on ethnology have not kept
their heads clear. Broca, as we have seen, laid it down that * the
Celts of history were the mixed population of Gallia Celtica whom
'

Caesar called Celtae and if for the Celts of history we substitute

;
' '

*
the Celts of history as written by Caesar ', his definition is a truism.
But one important point Broca overlooked. Just as the French are
called after one conquering people, the Franks just as the English
:

are called after one conquering people, the Angles so the hetero-
;

geneous Celtae of Transalpine Gaul were called after one conquering

people and that people were the Celts, or rather a branch of the
;

Celts, in the true sense of the word. The Celts, in short, were the
people who introduced the Celtic language into Gaul, into Asia Minor,
and into Britain the people who included the victors of the Allia,
;

the conquerors of Gallia Celtica, and the conquerors of Gallia Belgica ;

the people whom Polybius called indifferently Gauls and Celts the ;

people who, as Pausanias said, were originally called Celts and after-
wards called Gauls. If certain ancient writers confounded the tall
fair Celts who spoke Celtic with the tall fair Germans who spoke
German, the ancient writers who were better informed avoided such
a mistake. The popular instinct in this matter has been right. Let
us therefore restore to the word Celt the ethnical significance which
'
'

of right belongs to it.^

^ Of course I do not mean to affirm that either the conquering Belgae or the
conquering Celtae were homogeneous or to deny that the prevailing types of
;

the two groups, of whom the Belgae were the later comers, may, from various
causes, have become more or less differentiated I only mean that the purer
:

Belgae and the purer Celtae sprang from the same stock.
^ B. G., ii, 4, § 1.
' A. H. Keane, Ethnology, 1896, p. 397.
* I find that my conclusions were sanctioned in advance by Dr. Collignon in
an article {Annales de geogr., v, 1896, pp. 156-66) which I had not seen when
1093 Y
322 THE ETHNOLOGY OF GAUL

VI
THE BELGAE
1. What Caesar, Strabo, and Diodorus Siculus tell us about the

physical characters of the Belgae, I have virtually stated already.

They were tall, and, according to the two later writers, fair and they ;

did not differ noticeably from the Celtae. Caesar, as everybody knows,
tells us that the Belgae,*according to his Reman informants, were, for
the most part, of Cerman origin, and that the Condrusi, Eburones,
Caerosi, and Paemani were designated, in his time, as Germans;^ and,
in one passage,^ he himself describes the Segni and Condrusi as Germans
by race. The significance of these statements I shalLpresently discuss.
2. The great majority of the people who have settled in Belgic
Gaul since Caesar wrote have belonged to the German or kindred
races. Franks settled in large numbers in Flanders and Brabant
and also in the neighbourhood of Laon and Soissons afterwards :

Saxons and Frisians completely Germanized Flanders and Brabant.^

3. Of the skulls, prehistoric and Celtic, which have been found in
Belgic territory, I have already spoken.^
The evidence collected by MM. Broca, Collignon, and Topinard is
available only for the western and southern portions of the Belgic
territory, which comprise the departments of the Aisne, Ardennes,
Marne, Nord, Oise, Pas-de-Calais, Seine-Inferieure, Seine-et-Marne,
and Somme. In every one of these departments the mass of the
people are fair or relatively fair in all, except the Seine-Inferieure,
:

where the average height is medium, they are relatively tall while ;

in the Aisne, Nord, Pas-de-Calais, Seine-Inferieure, and Somme, they

are mesaticephalic, and in the remainder sub-brachycephalic.
With regard to that part of the Belgic territory which is included
in Belgium and Holland, we are mainly dependent upon the re-
searches of L.Vanderkindere,^Dr. Beddoe,^ and M. J. Deniker. The
the first edition of this book appeared. He speaks (p. 159) of the population
'

pre-gauloise que Broca nommait a tort les Celtes and affirms that les Celtes
'
'

de I'histoire sont en elfet aussi des blonds de haute taille'. Since the first edition
was published I have also lighted upon a passage {Bull, de la Soc. d'anthr., ii,
1861, pp. 508-9) in which Broca himself justifies my argument and uses the word
'
Celt in the same sense which I attach to it.
' The Celtae of Gaul, he remarks,
*
were already mixed before the arrival of the Kimris [or Gallo-Brythonic
invaders], since the name [Celtae] under which they appeared for the first time
in history had been imposed upon them by the conquering race of the Celts
properly so called, which, like the Kimris [who of course were also Celts] and the
Germans, came from the east, and like them, was dolichocephalic' Cette'

premiere opinion,' says Dr. L. Wilser {UAnihr., xiv, 1903, pp. 496-7), oubliee '

plus tard par son auteur et ses disciples, etait juste.' Cf. Anc. Britain, pp. 4-36—40.
1 5. 2
(?., ii, 4, §§ 1, 10. 76., vi, 32, § 1.
' Scottish Review, xxi,
1893, pp. 177-8. ^ See
pp. 269,272-4, 305.
^ Recherckes sur V ethnologie de la Belgique, 1872, pp. 21-2, 59, 63, 65, 68 ;

Nouvelles recherches sur V ethnologie de la Belgique, 1879, pp. 41, 44-5.

« The Races
of Britain, pp. 16, 21, 72-3, 94; Scottish Review, xxi, 1893,
pp. 178-9.
 THE ETHNOLOGY OF GAUL 323

line which marks the separation between the Flemish and Walloon
languages also marks the separation between two physical types,
the Flemings, whose territory was conquered, and has often been re-
peopled by Germans, being taller and fairer than the Walloons. The
Walloons, however, in the strip of territory which extends along the
valley of the Lys in Western Flanders, are taller than the Flemings
in the basin of the Yser. The inhabitants of the Flemish provinces
are also more dolichocephalic than those of the Walloon, the cephalic
index of the former ranging between 76-7 and 78-31, that of the
latter between 78-51 and 81- 17.^ Although the Walloons, generally
speaking, are dark, the purest, according to Vanderkindere for —
—
example, those of Namur are fair. The inhabitants of the Dutch
provinces of Zeeland, North Brabant, and Limburg, belong for the
most part to Broca's Celtic type,^ which is also found in the district
'
'

of Mons ^ while tall, dark, brachycephalic individuals, belonging to

;

the type which M. Deniker calls Adriatic, may be seen in Luxem-

bourg as well as in the department of the Marne.'*
4. I have already, in speaking of the Celtae, mentioned the view
which Broca took of the Belgae. He believed that the great mass of
the people, in Caesar's time, were Kymric —
that is to say, tall fair
' '

—
Gauls of the classical type intermixed with or living among a
minority of the short, dark, brachycephalic type which he called
Celtic. But it is certain that the comparatively high stature, the
comparatively low cephalic index, and the fairness of the existing
population are largely due to the German immigrants who have
settled in the country since Caesar's time and it is possible, as
;

Dr. Beddoe ^ suggests, that the cephalic index may be due not only
to Kymric and German elements, but also to the presence, in
' '

ancient times, of the small, dark, dolichocephalic people of the

Homme Mort type, whom ethnologists loosely call Iberian. That
Broca's Celtic element existed in considerable force, is proved not
' '

only by the continued existence, in certain parts of the country, of

people of the Grenelle type, but also by the fact that so large a pro-
portion of the population are now relatively brachycephalic. The
round-headed element suggests a question which I cannot answer.
Caesar's informants told him that the Belgae had expelled the Gauls
whom they found in possession, and who certainly included numerous
representatives of the old brachycephalic race. If the expulsion was
even approximately complete, it would seem that the Belgae them-
selves included a large proportion of round-heads, which, indeed, is
not unlikely anyhow round-heads must have been numerous in
:

the Belgic population of Caesar's time. Nevertheless I believe that

Broca's view, if it was pushed too far, contained a considerable
element of truth, and that in the northern, western, and relatively
dolichocephalic districts, at all events, the tall, blonde, Gallic element
^ According to M. E. Houze {Rev. d'anthr., 2^ ser., v, 1882, pp. 527-30), tho
average cephaHc index of 166 Flemings was 76-84, of 75 Walloons 81-10.
•^
Bull, et mem. de la Soc. d'anthr. de Paris, 5® ser., v, 1904, p. 579.
^ Association
frauf. pour V avancement c?&ssc., 36" sess"., 1907, V^ partie, p. 277.
* lb 6 The Races of Britain, p. 94.
Y 2
324 THE ETHNOLOGY OF GAUL
was proportionately stronger, in Caesar's time, among the Belgae
than among the Celtae. For it is certain, and indeed obvious that
the dark brachycephalic element is proportionately much smaller
now in Belgic than in Celtic Gaul and, moreover, a large number of
;

the taller modern inhabitants, especially among the Walloons and in

the neighbourhood of Reims, present a type which is markedly non-
Germanic, and which, as I shall show in the proper place, there is
strong reason to regard as a modified form of the true Gallic type.
Before I come to this question, it will be well to- examine the view
which Dr. Beddoe has published regarding the composition of the
Belgae.^
Besides the short, dark, brachycephalic people, the so-called
Iberians, and other representatives of prehistoric races. Dr. Beddoe
holds, as I understand, that the bulk of the Belgae proper consisted
of a large proportion of tall dark people, and a small proportion of
tall fair people, who formed the dominant caste. He arrives at this
conclusion by the following reasoning. The purest representatives
of the Belgae are, he believes, the Walloons, who, as Thurnam re-
marked,2 pride themselves upon being of the most ancient Gaulish
'

lineage.' ^ Doubtless, Dr. Beddoe admits, the Walloons are not,

strictly speaking, a pure race. There is probably a Germanic and an
*
Iberian element in their composition. Still, their type is markedly
'

different from that of the true Germans and the blonde Flemings.
They and the inhabitants of Reims and Epernay, who closely re-
semble them, have the tall frames, square foreheads, and long
'

sharply drawn features which constitute W. Edwards's Kymric

type'. This type prevails throughout the country extending north-
eastward from the neighbourhood of Reims through the Ardennes
to Liege and Verviers. But the moment we cross the Walloon
frontier, we find ourselves among a totally different people, whose
features and complexions are German. People of the Walloon
type are found in other districts, where Gauls or Celtic-speaking
peoples have lived. They are found everywhere in Brittany, and
especially in the district of Leonais, where the British immigrants
of the fifth century are believed to have landed they abound in :

Normandy, in Northern Italy, and in Cornwall and the type seems ;

'

to constitute an element of more or less importance in the population

of most parts of the British isles '. This hatchet-faced Walloon
'

type,' says Dr. Beddoe, is the same which

'
Broca .connects
. . . . .

with tall stature and fair hair.' But the point upon which the
doctor, speaking from personal observation and with the authority
of an expert whose reputation among anthropologists is European,
lays special stress, is that in the country of the Belgae the type is
not connected with fair, but with dark hair. All the 14 Flemish
arrondissements of Belgium, he insists, show more blondes and
1 The Races of Britain, pp. 21-2 ; Journ. Anthr. Inst., ii, 1873, pp. 18-20.
^ Crania Britannica, pp. 164-5.
^ According to Vanderkindere {Recherches, &c.,
p. 66) the Walloons of Naiuur
and Luxembourg are ethnically identical with the people of North -Eastern
France, and especially of Champagne.
 THE ETHNOLOGY OF GAUL 325

fewer brunets than any of the 12 Walloon

' '
and his conclusion
;

is that though the chiefs, the true Galatae, were fair, the mass of
'

the old Belgae was of old something like what it is now '.
Vanderkindere ^ is of a different opinion. He admits that the
Walloons in general are dark, but he holds that Dr. Beddoe is wrong
in concluding from this fact that the mass of the Belgae resembled
them in colouring. The darkness of the Walloons is due, he thinks,
to the crossing of their ancestors with the prehistoric dark popula-
tion and the proof is that the purest Walloons, for example those
;

of Namur, are fair.

It is impossible to decide the question. Dr. Beddoe himself infers
from the skull-form of the Walloons that their ancestors were largely-
crossed by the dark Iberians '. Vanderkindere thinks that their
'

comparatively high cephalic index is due to admixture with the

dark brachycephalic people,^ and Dr. Beddoe himself^ ascribes it
to partial descent from the same stock. Possibly both may be
right. It is impossible to say whether the coalescence of races had
already proceeded so far before Caesar's time that even then, as
Dr. Beddoe thinks, the mass of the Belgae were dark, or whether it
was mainly deferred till a later time. The tallest men in Dalmatia
are the darkest ^ and the tallest representatives of the tallest
;

—
population in Great Britain the men of Upper Galloway, whose
average height is nearly 5 feet lOJ inches, or 1 metre 79 are also —
the darkest ^ though the fact that their dark-brown hair often
;

shows a substratum of red, and is generally accompanied by grey

eyes, seems to show that they had fair as well as dark ancestors.
But considering the implied statements of the ancient writers re-
garding the prevalent fairness of the Belgae, I find it difficult to
believe that this attribute was virtually confined to a few chiefs.

VII

WHO WERE THE TRUE GAULS ?

have now to grapple with one of the most difficult problems of

I
Gallic ethnology, —
who were the tall fair conquerors of Gaul, the
Gauls properly so called, the latest element in each of the two groups
which were known as Celtae and Belgae ? Are we to identify them
with the ancient Germans, or did they constitute a distinct group ? ^

Recherches sur V ethnologie de la Belgique, p. 65.

^

So also M. Houze. See Rev. d'anthr., 2" ser., v, 1882, p. 530.

^

^ The Races o/ Britain,

p. 42.
* Scottish Review, xx,
1892, p. 378.
Beddoe, The Races of Britain, p. 249.
•*

* The absurd but widely accepted theory that the Goidels, for whom
some
have actually substituted the Belgae, were identical with the tall Round
Barrow race of Britain is refuted in Aiic. Britain, pp. 429-33, on which
'
'

a well-informed writer in Nature, April 30, 1908, p. 602, remarks that perhaps
'

the most valuable pages of the book are those in which he demonstrates that
the short-headed race was not Celtic '.
326 THE ETHNOLOGY OF GAUL
It ought not now to be necessary to warn the reader that, when
I speak of the tall fair conquerors of Gaul, the Gauls properly so
'

called ', I do not mean to imply that any group of Gauls, even the
Gauls who captured Eome, all belonged to one type. On the con-
trary, if anything in ethnology is certain, we may be sure that even
they were more or less mixed. So, Dr. Beddoe warns us not to believe
that there was ever a period when, for example, all the Caledonians
were red-haired.^ I only mean to imply that, among the conquerors
of Gaul, tallness and fairness were the prevailing, or at least the most
noticeable characteristics.
The ancient writers, who are unanimous in describing the physical
type of the Gauls, tell us nothing about the shape of their skulls.
Broca, as I have already observed, thought that it was dolicho-
cephalic, although he admitted that the authentic Gallic skulls
which were then available were too few to generalize from. Since he
wrote, however, further discoveries, which I have already noted,^ have
proved that he was right the Celts who invaded Gaul in the Iron
:

Age were for the most part a long-headed people and the mor- ;

phology of their skulls was identical with that of those which have
been found in Merovingian graves. Or rather I should say that such
were the invaders as we know them from sepulchral evidence for ;

the humbler classes may not have been represented in the graves
from which that evidence is derived.
Some 50 years ago a Belgian general, M. J. B. Renard, published a
treatise^ to prove that the Gauls and the Germans were, anthropo-
logically, the same people. Many of his arguments are now com-
pletely out of date and M. Vanderkindere ^ speaks of his theory
;

as one which it is no longer worth while to refute but there are still
:

eminent authorities who hold it, at all events in a modified form.

Broca ^ maintained that the Gallic conquerors and the Germans were
closely akin, if not identical Dr. Lagneau^ grouped them along
:

with the Germans, the Franks, the Normans, the Goths, and the
Burgundians, under the common designation of la race germanique
'

septentrionale ' Huxley affirmed that the typical Gauls were the
;
'^ '

close allies, by blood, customs and language, of the ancient Germans'

and M. JuUian, who is unable to draw a sharp distinction between
German and Celt, regards the Celtic conquerors of Gaul as des '

Cimbres qui ont reussi'.^

The testimony of the ancient writers has been quoted so often that

The Races of Britain, p. 245.

^

See pp. 310-1, and Rice Holmes, Anc. Britain, pp. 434-6, 440.
^

^ De Vxdentite de race des Gaulois et des Oermains, 1856. Cf. A. Holtzmann,

Kelten und Germanen, 1855, and M. Jullian's bibliographical notes {Hist, de la
Oaule, i, 231, nn. 1, 2).
* Recherches sur V ethnologic de la Belgique, p. 11,
^ Bull, de la Soc. d'anthr. de Paris, iv, 1893,
p. 51.
* Diet. ency. des so. medicales, xiii, 1863,
p. 767.
' Critiques and Addresses, 1873, p. 180. See also F. Delisle's article in Did.
des sc. anthr., p. 503.
8 Rev. hist, Ixxxv, 1904, p. 320. Cf. C. JuUian, Hist, de la Gaule, i, 231-2,
243, n. 3.
 ^ ^

THE ETHNOLOGY OF GAUL 327

it is unnecessary to repeat it. With absolute unanimity they describe

the Germans in terms which are virtually identical with those which
they use in describing the Gauls. The Germans, like the Gauls, were
tall and fair :that is the sum and substance of their evidence.
Attempts have been made to show that they draw distinctions in
detail between the two peoples, but without much success. Thus
Suetonius ^ tells us that Caligula tried to palm off Gauls for Germans
by picking out the tallest and dyeing their hair red, the inference
being that the Gauls generally were shorter and darker than the
Germans ^ but this does not prove that the pure Germans were
:

redder than the pure Gauls for in Caligula's time the Gauls were
;

very much mixed. Again, Tacitus, after saying that the Caledonians
had the huge stature and the red hair of Germans, goes on to say
that the people of Southern Britain were more like the Gauls.*
Assuming the accuracy of Tacitus's statement, it should seem that
the Gauls, or rather the inhabitants of Gaul generally in his time,
were less fair than the Germans whom the Romans knew but this ;

does not prove that the purest Gauls were less fair. Strabo ^ also
says that the Germans, though like the Gauls, were taller and fairer ;

but the only inference which can fairly be made from his statement,
as from those of Tacitus and Suetonius, is that, in the first century
of our era, tall stature and fair hair were less common among the
Gauls than among the Germans. Moreover, there is no evidence,
except the remark of Tacitus, for the theory that the Caledonians
were Germans they certainly spoke Celtic
: and even if they or
;

their ancestors had migrated from Germany, the fact would be quite
consistent with the view that they were Celts.
So much for the testimony of the ancients. We may also learn
something from modern research. Although in Southern Germany,
as in Gaul, the older population was brachycephalic, it is certain that
the Germans whose physical features the Romans described were
a dolichocephalic race. This statement is true of all the races who
are generally known as Germanic. The Merovingian skulls have
been already mentioned. Numerous skulls, which are known as the
Row Grave (Reihen-Graber) skulls, have been found in South-
western Germany, and are assigned to Frankish and Alemannic
'

warriors of the fifth and following centuries '. Their average index
is 71-3 and skulls of the same type have been found over the whole
;

area which was conquered by the Goths, the Franks, the Burgundians,
and the Saxons."^ Modern Germans with skulls of this type are
nearly always fair and, as Dr. Beddoe observes, this fact is a further
;

proof that the Row Grave type of skull was that of the Germans of

^ See Prichard, Physical History of Mankind, iii, 391-2, who gives all the
references. ^ Caligula, 47.
^ Huxley
observes {Critiques and Addresses, p. 171) that the Germans were
in the habit of artificially reddening their hair. But so were the Gauls (see
Diodorus Siculus, v, 28, § 1).
* Agricola, 11. « vii,
1, § 2.
^ Rice Holmes, Anc. Britain, p. 418, n. 1.
' De Quatrefages, Hist. gen. des races humaines, 1887-9, pp. 492-4.
328 THE ETHNOLOGY OF GAUL
classical times. Again, Dr. Beddoe tells us that 70 Anglo-Saxon
skulls discovered in Britain show a mean index of 74-59 and these ;

skulls resemble the well-known Hohberg type, which has been proved
to be that of the Burgundians.^
It may be taken as proved, then, that the typical Gaul and the
typical German were alike tall, fair, and dolichocephalic. But was
their fairness of the same kind ? For instance, were the Gauls like
the fair red-bearded Highlanders of Perthshire, and the Germans like
the yellow-haired and yellow-bearded Scandinavians ? Were there
any other features which differentiated them ? Judging by the '

testimony of the ancient writers, who did not draw nice distinctions,
we should say that there was practically no difference between the
two peoples. Let us, then, turn to other sources of information.
It is important to decide what people introduced the Celtic language
into Gaul. The prevailing view, which indeed is taken for granted
by some writers, is that the Gauls imposed their language upon the
peoples whom they conquered but M. A. Hovelacque, who assumes
;

the ethnic identity of the Gauls and the Germans, has tried to prove
that the language of the Gallic conquerors was German. ^ His argu-
ment is substantially this —
wherever the Gauls pushed their con-
:

quests, they found a Celtic population —

using the word Celtic in ' '

the erroneous sense of the dark brachycephalic people in posses- —

sion ; but the Gauls did not penetrate into every country which was
occupied by C^elts '. It is clear that the Gauls did not impose their
'

language upon peoples among whom they never settled while on ;

the other hand it is easy to understand that in the countries in which

they did settle they adopted the language of the inhabitants. En '

somme,' triumphantly concludes M. Hovelacque, 1*^ tons les blonds '

de haute taille parlaient jadis des idiomes teutoniques

. . . 2" la ;

partie de cette race (Galates) qui penetra sur le territoire occupe par
des Celtes perdit la propre langue et parla celtique.'
I must say that this argument is one of the most amazing instances
of sheer confusion of thought that I have ever come across. Of two
things one. By les Celtes M. Hovelacque either means the dark
' '

brachycephalic Grenelle race, or he means Celtic -speaking peoples.

As far as I can see, he means the former. If so, there is no evidence
that the Gauls, wherever they pushed their conquests, found a
'
Celtic people in possession
'
and there is evidence that in some
;

countries they did not. It is certain that the people whom the Belgic
immigrants found in possession when they settled in the British
Isles were for the most part ethnologically distinct from the dark
brachycephalic race whom the Gallic conquerors found in Gaul nor ;

is there any evidence that the Gallic conquerors of Galatia found

in possession a people of this race. If, on the other hand, M. Hove-
lacque means by les Celtes Celtic-speaking peoples, he begs the
' '

whole question !

That the Gauls imposed the Celtic language upon the peoples
' Scottish Review, xxi, 1893, pp. 167-8 ; LAnthr., v, 1894, p. 518.
* Rev. de linguistique, xviii, 1885, pp. 194-5.
 -

THE ETHNOLOGY OF GAUL 329

whom they found in possession of Gaul, can be proved to demonstra-

tion. Assuming, what is not proved, that the Gallic conquerors
were greatly inferior in numbers to the people whom they conquered,
there is abundant evidence that a conquering minority may and
frequently has imposed its language upon a subject population. It
is quite true that there are plenty of instances on the other side the ;

Normans who conquered England, the Goths, and the Burgundians

learned the languages of their subjects. But in these cases the con-
querors, besides being numerically inferior, were also either less
civilized or not more civilized than the peoples whom they conquered.
The inhabitants of Gaul were far superior in numbers to the Roman
conquerors who settled among them but their language is a Ro-
;

mance language. But what I have said only shows that the Gauls
might have imposed their language upon their subjects. There is
abundant evidence that they did. Putting aside certain geographical
names, such as Sequana, which may be Ligurian, the vast majority
of the names of towns and people in Gaul are Celtic. Is it credible
that the chiefs of the conquering race should have been called by
names which were not their own but those of their subjects ? Wher-
ever history tells us that the Gauls or the Celts (I use the word not
in M. Hovelacque's sense but in the sense of Polybius) conquered or
settled, there we find traces of the Celtic tongue. The Gauls made
conquests in Germany and Switzerland, and there we find abundant
linguistic traces of their occupation. The Celts settled in certain
parts of Spain and names like Celtici, Celtica} and Celticoflavia,^
;

if not Nemetohriga, bear witness to their presence. They settled in

Asia Minor and they spoke Celtic, and their chiefs bore Celtic
;

names. Goidels, Brythons, and Belgae settled in Britain and ;

Celtic is still spoken in the British Isles. Yet Broca insisted that
there never were any Celts, in the sense in which he used the word,
in Britain ^ and although recent discoveries have shown that he
;

was too positive, the invaders of the Crenelle type who reached our
country were few.'* As M. Zaborowski remarks,^ Celtic names exist
in places where notre type celtique, celui des anthropologistes, n'a
'

jamais existe '.

Professor Ridge way, indeed, who believes that the Celtic con-
querors of the British Isles learned the language of the Gaelic '

speaking melanochrous Aborigines ',^ argues that even when the con- '

querors bring with them some women of their own race the . . .

invaders are liable to drop their own language and practically adopt
that of the natives ', and that the adoption of the language of the
'

conqueror by the conquered, except in the most favourable circum-

stances, is not common, and only takes place by a very gradual
process, as is seen in the case of Ireland and he complains that'
;

'
Pliny, Nat. Hist., §§ 13-4.
iii, 1 (3),
^
Corpus 880.
inscr. Lat., ii,
3
Rev. d'anthr., ii, 1873, p. 625.
*
Rice Holmes, Anc. Britain, pp. 425-30, 455.
^
Diet, des sc. anthr., pp. 246-7.
^
Who were the Romans ? 1907, p. 22 {Froc Brit. Acad., vol. iii).
 330 THE ETHNOLOGY OF GAUL
Sir John Rhys has assumed that it was possible for the aborigines
'

[of Britain] to have been so completely Celticized as to have adopted

the Aryan tense system, as well as the Aryan vocabulary, in its
fullness in the interval between the sixth or fifth century and the
second century B. c' Yet, he says, although English has been the
'

master speech in Britain for many centuries, and that, too, when
reading and writing have been commonly practised Gaelic still. . .

survives, while Welsh not only survives, but flourishes. It is there-

fore simply incredible that such a complete transformation as that
postulated could have taken place in three or four centuries in an
age when writing and literature can be hardly said to have existed
'.^
in these islands
Now, as the reader will have seen, the position against which
Professor Ridgeway's battery is chiefly directed is that the language
of the people whom the Celtic conquerors found in possession was
non-Aryan. Admit that it may have been Aryan all that I am
:

concerned to show is that the conquerors did not learn Celtic, but
spoke it already and imposed it. To begin with, it is certain that
the Celtic conquerors of Gaul did bring with them not only some '

women but all their women. This was the regular practice both of
'

the Celts and of their kinsfolk, the Germans. ^ The time in which,
according to Professor Ridgeway, it is simply incredible that the
' '

Celtic language should have become dominant in Britain, and pre-

sumably also in Gaul, w^as considerably longer than that in which,
on his own showing, the language of a small minority of English
settlers became dominant in Ireland. Leaving out of account the
rapidity with which the language of Rome took root in Britain ,2
Gaul,* and Spain,^ I ask Professor Ridgeway to reflect upon what is
involved in the theory that the Celtic conquerors learned Celtic from
the peoples whom they conquered. First, this startling paradox
requires us to assume either that the Ligurians had always been con-
fined to the south-east of Gaul, or that the Ligurian language was
Celtic : if the former alternative is true, the Celtic conquerors imposed

the Celtic language, which ex hypothesi they had only just learned,
upon the southern Ligurians, for the latter spoke Celtic in the fourth
century B. c. ^ if, as Sir John Rhys is now inclined to believe, the
;

—
Ligurian language was Celtic in other w^ords, the twin-brother of
—
Goidelic it was itself supplanted by Gallo-Brythonic.'^ Secondly,
Professor Ridge w^ay's theory would oblige us to assume that Celtic
was the language which the Celtic conquerors found spoken not
only in Gaul and in Britain, but also in Helvetia and in Spain is :

' Nature, Sept. 24, 1908, pp. 529-30.

- Strabo, iv, 4, § 2. Cf. Caesar, .B. (?., i, 29, § 1
; 51, § 3.
3 F. Haverfield, The Bomanization of Roman Britain, 1905, pp. 10-1, 27
{Proc. Brit. Acad., vol. ii).
* E. Lavisse, Hist, de France, t. i (by G. Bloch), 1900,
pp. 385-6. Cf. Rev.
xxviii, 1907, p. 374.
celt.,
Professor Ridgeway himself remarks ( Who were the Romans ? p. 24) that
^

'
the Latin language got a hold on all Eastern Spain with an astonishing
rapidity after the Roman conquest'.
« Milanges H. d' A. de Juhainville,
p. 108. ^ See pp. 281-2, 319-20.
 —;;

THE ETHNOLOGY OF GAUL 331

such a remarkable diffusion of Celtic among peoples widely separated

and anthropologically different credible ? Thirdly, is Professor
Ridgeway prepared to maintain that Celtic was spoken by the
—
aborigines I use the word not loosely, but in its strictest sense
in Gaul, in Britain, and in Helvetia ? If not, he must admit that in
those countries some invaders had imposed it. Fourthly, if the
'
aborigines of the British Isles and of Gaul spoke Gaelic, Gaelic
'

must have branched off from the parent Indo-European language

in the Palaeolithic Age ^ is Professor Ridgeway prepared to main-
:

tain this ? Fifthly, if the aborigines of the British Isles spoke

' '

Gaelic, why did not the Brythons, who conquered them, learn Gaelic
from them ? Sixthly, if the Celts did not speak Celtic when they
invaded Gaul, Britain, and Helvetia, how are the Celtic place-names
in Germany ^ to be accounted for ? Lastly, it is admitted that the
language of the Belgae was Celtic they certainly did not learn it
:

from the Gauls whom they found in possession, for they incontinently
expelled them ^ therefore they must have spoken it when they
;

invaded Gaul. Is not the inference inevitable that their kinsmen

who had already conquered the rest of Gaul did the same ? The
Gallo-Brythonic language unquestionably prevailed, wholly or in
great part, over that Q language which has provoked so much con-
troversy in Caesar's time it was the language not only of the
:

Belgae but also of all, or almost all the Celtae and not even the;

most perverse ingenuity can plausibly argue that conquerors learned

it from conquered. It is likely enough that both in Britain and in
Gaul aboriginal languages, Aryan or non-Aryan, survived here and
there even in Caesar's time but let Professor Ridgeway think
;

again, let him remember how Etruscan vanished before Latin, and
haply he will no longer find it incredible that the language of the
conquerors, who were accompanied by their women, should have
prevailed.
1 regard it, then, as certain that when the Gallic conquerors
entered Gaul, they brought the Celtic language with them and, ;

inasmuch as Celtic is more closely akin to Latin than to German, it

is clear that the tall fair Gauls, if they had been originally one with

the tall fair Germans, had long since branched off from them and ;

it is therefore not improbable that the physical types of the two

peoples had become to some degree differentiated. N. Freret^
remarks that Celts and Germans must have become greatly inter-
mixed during the long sojourn of the former in Germany and the ;

same thought had often presented itself to my own mind before

I read Freret's book. But is it absolutely certain that when the
^ See 0. Schrader, Prehist. Ant.
of the Anjan Peoples, pp. 358-9, 362-3, 367-8
Rice Holmes, Anc. Britain, p. 433, n, 5 Rev. de VEcole d'anthr., xviii, 1908,
;

p. 406 and J Deehelette, Manuel d' archeologie, ii, 9, n. 2.

;

2 See p. 339.
^ B.
0., ii, 4, § 1. No doubt Caesar's words are not to be taken literally
but admitting this, all analogy,' as Zupitza says {Zeitschr. f. celt. Philologie,
'

iv, 1903, pp. 18-9), is opposed to the assumption that the Belgae did not
speak Celtic before they crossed the Rhine.
* (Euvres completes, v, 1896, pp. 7-8.
332 THE ETHNOLOGY OF GAUL
Celts began to migrate from Germany into Gaul, the tall fair Ger-
mans had long established themselves in Germany ? Is it certain
that the pressure of their invasion was not the motive of the Celtic
emigration ? ^ Still, we must not forget that Hamy could detect no
essential difference between the skulls of the early La Tene period
which he examined, and those of the Merovingian type and perhaps ;

it was not long before Caesar's time that the aspect of the descen-
dants of the Gallic conquerors had become noticeably different from
that of their German neighbours.
have argued that the Belgae must have spoken Celtic before
1
they arrived in Gaul and I do not waver when I recall those famous
;

words flerosque Belgas esse ortos a Germanis (' most of the Belgae
were of German origin '). Caesar merely records the statement of the
Reman envoys without endorsing it ^ and the fact that he himself,
;

rightly or wrongly, specifies four Belgic peoples —

the Eburones, the
Segni, the Condrusi, and the Paemani —
as German,^ would seem to
imply that he had reason to believe that the rest of the Belgae were
not.* Tacitus, who, as Dr. Beddoe observes,^ evidently aimed at '

accuracy in the matter, allows only the Triboci, the Nemetes, and the
Vangiones ... to be "hand dubie Germanorum populi " ;^ and '

none of the three were Belgae at all. The Treveri and the Nervii,
according to Tacitus,*^ wished to be thought Germans but, if Tacitus ;

was rightly informed, this very fact would appear to show that they
were not what they professed to be. Strabo says that the Nervii were
Germans ^ but his unsupported statement does not count for much
:
;

and the names of Nervian and Treveran individuals, as well as the

geographical names of both peoples,^ were Celtic. So also were the
names of the Eburones and their two kings, Ambiorix and Catu- —
volcus. Hirtius, while noting the resemblance of the Treveri to the
Germans in manners and customs, says that it was due simply to
geographical propinquity ^^ and M. Vanderkindere points out ^^ that
;

Caesar draws a distinction between the Menapii and the Germans.^^

It may be that, as M. Jullian^^ suggests, there were Celticized Ger-

^
d' A. de Jubainville gives Hnguistic reasons for beheving that the Germans
M.
had lived for some centuries in subjection to the Celts, and hemmed in between
them, the Slavs and the North Sea. He points to the word Teutoni —
mot '

'

germanique conserve intact par les Celtes, qui n'en ont celtise que la desinence
— and to various military and other words qui sont communs aux Celtes et
'

aux Germains', and which, he affirms, were borrowed by the latter from the
former {Les premiers habitants de V Europe, ii, 1894, pp. 325-73). Is the evidence
sufficient to sustain the theory ? M. d'Arbois argues further that une opposi-
'

'

tion religieuse chez les Germains empeche leur absorption par les Celtes
{ih., pp. 373-83) ; whereas Sir John Rhys {Celtic Heathendom, p. 41) is impressed
by the striking similarity between the ancient theologies of Celts and Teutons.'
'

2 B. 0., ii, 4, §§ 1-2. ' Ih., ii, 4, § 10 ; vi, 32, § 1.

* See pp. 338-40. ^ The Races of Britain, ^^.20-\.
« Germania, 28. ' lb. » iv,
3, § 4.
8 Seep. 329. '» B. G., viii,
25, § 2.
^^
Recherchcs sur V ethnologic de la Belgique, p. 16, n. 3. Dr. Beddoe {The
Races of Britain, p. 21) argues that, as the modern inhabitants of the country
which was once occupied by the Nervii are generally Walloon in language and in
physical type, although the country was invaded by Franks, the Nervii were not
of German origin. ^^
B. G., iv, 4. " Hist, de la Gaule, ii, 464-5.
 THE ETHNOLOGY OF GAUL 333

mans among the Nervii and the Treveri ^ but, unless we know what ;

the Reman envoys meant by the word Germani, their statement

that the Belgae were of German origin proves nothing.^ The word
' '

Germanus may or may not have been Celtic ^ but there is no proof ;

that the Gauls meant by Germani a Teutonic people who spoke

a Teutonic language. My own conviction is that when the Reman
envoys told Caesar that the Belgae were of German origin ', they '

spoke the truth but that they only meant that the Belgae were
;

the descendants of a people who had once dwelt on the east of the
Rhine.4
Assuming for the present that, notwithstanding their general
similarity, there was a physical distinction of some sort between the
Gauls who conquered Gaul and the Germans, it may be that an
observation of the Celtic-speaking peoples of the British Isles will
help us to form an approximately just idea of the Gallic type. This
inquiry will of course demand great care, for the purest specimens of
the Scottish Highlanders and the Irish Celts are greatly mixed but ;

I do not think that the difficulties are insuperable. The Gaels in

Ireland and Scotland, says Dr. Beddoe,^ are probably Iberians ', '

crossed with a long-faced, harsh-featured, red-haired race, who con-

'

tributed the language and much of the character '. The dolicho- '

cephalous Celt of the Scottish Highlands, says the same writer,

'

comprehends both Galatic and Iberian elements, if not others.

Some of the leading points of this type, he goes on to say, are fre-
valent ivherever Gaelic is known to have been spoken. Now, what are
the physical characteristics of the dolichocephalous Celt ? Taking '
'

colour first. Dr. Beddoe found that of 48 individuals 5 had red hair,
4 fair, 3 lightish brown, 11 brown, 17 dark brown, 3 brown-black,
and 3 coal-black, but that the eyes were generally light. And, he
adds, the figures for the Highlands generally are much the same.^
According to Hector Maclean, the dolichocephalous Celt the ' '
—
truly Celtic type, as he regards it, to which belonged the Galli of
—
'

the old Roman writers and the Celtae of Caesar is often tall he '
' ;

is of various complexion ranging from a ruddy white to a swarthy

. . .

hue. . The face is frequently long

. . the lips are usually full . . . . . .

cheek bones large and prominent eyes most frequently grey or ;

bluish grey hair reddish yellow, yellowish red, but more fre-
. . .

quently of various shades of brown, of which yellow (" I," notes

^ If the Cimbri and Teuton! were Germans

(see pp. 549-53), so were their
descendants, the Atuatuci {B. G., ii, 29, § 4), though they must have become
Celticized in speech, and largely in blood.
^ Professor G. Kossinna {Korrespondenz-Blatt d. deutsch. Gesellschaft f.
Anthr., &c., xxxviii, 1907, p. 58) attributes the fact that in the period known
as La Tene III, which began about 100 b. c, cremation appears to have been
general in North-Eastern Gaul, to German influence but he admits (p. 59) ;

—
that in Alsace the territory of Ariovistus and his German host German —
graves have not been found.
^ See K. Miillenhoff, Deutsche Altertumskunde, ii, 200-5 A. Holder, Alt- ;

celtischer Sprachschatz, i, 204 ; and Bev.

xxviii, 1907, p. 414,
celt.,
* Cf. Bull, bibl du Jihisee beige, 1908, p. 71.
• The Races of Britain, pp. 270-1. « lb., pp. 244-5.
334 THE ETHNOLOGY OF GAUL
Dr. Beddoe, " should read here red or reddish yellow ") is the ground
colour.'
Dr. Beddoe remarks further that the Strathclyde Brythons and
many Cornishmen have a considerable resemblance to the Walloons ;

that the features of the Belgae depicted on the monument of

Jovinus at Reims were not unlike those of modern Gaels and that ;

the people of the Aran Isles in Galway Bay have much the same
'

long-headed long-featured type which is common in the Belgic

districts of North-Eastern France '.^
Now, making every allowance for the admixture of other blood
which must have considerably modified the type of the original
Celtic or Gallic invaders of these islands, we are struck by the fact
that among all our Celtic-speaking fellow-subjects there are to be
found numerous specimens of a type which also exists in those parts
of Brittany which were colonized by British invaders and in those
parts of Gaul in which the Gallic invaders appear to have settled
most thickly, as well as in Northern Italy, where the Celtic invaders
were once dominant and also by the fact that this type, even among
;

the most blonde representatives of it, is strikingly different, to the casual

as ivell as to the scientific observer, from that of the purest representatives

of the ancient Germans.^ The well-known picture by Sir David Wilkie

— —
Reading of the Waterloo Gazette illustrates, as Daniel Wilson re-
marked, the difference between the two types. Put a Perthshire
Highlander side by side with a Sussex farmer. Both will be fair :

but the red hair and beard of the Scot will be in marked contrast
with the fair hair of the Englishman ; and their features will differ
still more markedly. I remember seeing two gamekeepers in a rail-
way carriage running from Inverness to Lairg. They were tall,
athletic, fair men, evidently belonging to the Scandinavian type,
which, as Dr. Beddoe says, is so common in the extreme north of
Scotland ;but both in colouring and in general aspect they were
utterly different from the tall fair Highlanders whom I had seen in
Perthshire. There was not a trace of red in their hair, their long
beards being absolutely yellow. The prevalence of red among the
Celtic-speaking peoples is, it seems to me, a most striking charac-
teristic.^ Not only do we find 11 men in every 100, whose hair is
absolutely red, but underlying the blacks and the dark browns the
same tint is to be discerned. In France again the proportion of red-
haired individuals is greatest (5-32 per 100) not in Normandy or
the north-eastern departments, where the proportion of Germanic
immigrants was greatest, but in Finistere,^ where many of the
'
Kymric invaders from Britain landed. It is true that M. Topinard
'

1 The Races of
Britain, pp. 21, 25, 28, 245.
^ W. Edwards {Les caracteres physiol. des races humaines,
It cannot, argues
pp. 6-7), have been either Burgundian or Norman, because it is found in places
where neither Burgundians nor Normans have ever existed.
^ So Dr. Beddoe, I find, regards the high proportion of red hair in the Bur-
gundian cantons of Switzerland as a legacy of the Helvetii rather than of the
'

Burgundians' [The Races of Britain, p. 78). He also remarks that the people
of Central Wales resemble the Scottish Highlanders in the frequency of red hair.
* UAnthr., iv, 1893, pp. 584, 590-1.
 —'

THE ETHNOLOGY OF GAUL 335

regards red hair as a mere variety of blonde, without racial signifi-

cance but if so, it is difficult to understand why it is so frequent
;

—
judging from my own observations among Jews, and comparatively
so rare among the modern Germans ^ and the English. My investiga-
tions lead me to believe that red-haired people, living in England,
would, in nearly all cases, prove to be of Scotch, Irish, Welsh, or
Jewish origin.2 I know well that it is impossible to break up the
Highland type and the Anglo-Saxon into their component parts, and
then to isolate the Celtic element in the one and the Germ.an in the
other, and compare or contrast them but, considering that, wher-
;

ever the Celtic language is spoken, we find the same Kymric type, ' '

I think that what I have said is enough to establish at least a proba-

bility that the Celtic element and the German, notwithstanding
their general resemblance, differed one from the other. The Kymric '

type of face may be due to crossing between Gaul and Iberian ' '
;

but is the striking difference in colouring between the red High-

landers of Perthshire and the blonde North German or Scandinavian
to be explained by any intermixture in the blood of the former ? ^
Perhaps it is. For the most that I have succeeded in proving is
that the Celts had become differentiated from the Germans some
centuries after they had parted from them and what we want to
;

learn is whether any difference had arisen when they first entered
Gaul. The tall Gaul and the tall German were undoubtedly de-
scended from a common xanthochroid stock.*
' '

^ Scottish Review, xix, 1892, p. 418.

- Some years ago I found that 34 out of about 620 boys in one of the great
pubhc schools had red hair of various shades. I ascertained the nationality of
31. Of these 10 were either wholly or partly Scotch, 8 (including one of the
previous group) wholly or partly Irish, 7 (including one of the last-named
group) wholly or partly Welsh, 1 Lancastrian, 1 Salopian, and 3 Jewish. Of
the 3 whose nationality I omitted to inquire, 2 had names which are common
in Scotland and Ireland.
* In regard to the cause of red hair see Rice Holmes, Anc. Britain, p. 440,

n. 6 ; Globus, xciii, 1908, pp. 309-12 ; Rev. de VEcole d'anthr., xviii, 1908,
p. 281 ; and Man, viii, 1908, no. 27, pp. 54-8. It looks,' says Mr. J. Gray
'

{ib., p. 57), as if red hair were evolved from dark brown by converting a certain
'

percentage of its black pigment into orange pigment. It would follow that . . .

red hair should be of rare occurrence in a very blonde population. This . . .

conclusion is confirmed bj'' the observations of Virchow in North Germany,

where only a very small percentage of red hair was found.
* In the first edition
(pp. 316-7) I tried to show further that the tempera-
ment of the Celtic invaders of Gaul was different from that of the Germans.
But I was groping in the dark. The many-faceted Celtic temperament is
'
'

a fascinating object of study but it is not everywhere the same. Let me

;

repeat what I have said elsewhere {Anc. Britain, p. 457),

—
Study it in Wales,
'

in Man, in the Scottish Highlands, in Ireland, in Cornwall, in France, and you

will find that while it is Celtic everywhere, everywhere it is diiferent that ;

everywhere it has become what it is, because it is compounded, in different

degrees, not only of Celtic, not only of pre-Celtic and pre-Aryan, but also of
post-Celtic elements. And all these elements have been modified and moulded
by different geographical and climatic influences and by adventitious circum-
stances too numerous to be particularized and too elusive to be estimated.'
The Celtic character, as described by Caesar and other ancient writers,
'
'

was the character of a mixed people, whose Celtic ancestors had parted from
the Germans centuries before. We cannot compare the Celts who first crossed
336 THE ETHNOLOGY OF GAUL
I will now which I have arrived.
briefly recapitulate the results at
It isnot improbable that man existed in Gaul even in the immeasur-
ably remote Tertiary Period. The palaeolithic peoples, who lived
there in the Quaternary Period, were all, as far as is known, dolicho-
cephalic :some of them belonged to the Neanderthal type others ;

had certain negroid characteristics others again belonged to the

;

highly developed types which are represented by the skeleton of

Cro-Magnon, by similar skeletons in the caves of Mentone, and by
those of Laugerie-Basse and Chancelade. There is not enough
evidence either to confirm or to disprove the hypothesis that the
Neanderthal race became extinct but at all events before the close
;

of the Palaeolithic Age the zoological evolution of man was complete.

Of the neolithic races the earlier, represented by the people of
I'Homme Mort and Baumes-Chaudes, was of medium stature, slender,
and dolichocephalic, and descended from the people of Cro-Magnon
and Laugerie-Basse the later was for the most part brachycephalic,
:

and entered Gaul from the east. If new immigrants entered Gaul in
the Bronze Age, they probably belonged for the most part to this
same stock but early in the Hallstatt Period a tall dolichocephalic
;

race appeared in the Jura and the Doubs, who may have been the
advanced guard of the Celts. The Ligurians, when they first came
under the notice of the Greek geographers, occupied the region
bounded by Italy, the Mediterranean, and the Rhone, and after-
wards the country between the Rhone and the Pyrenees but there :

is reason for believing that Ligurians had also once possessed the
whole eastern region as far north as the Marne, if not Aquitania ;

and if so, their descendants must have survived in considerable

numbers among the Celtae. The Iberians dwelt between the Rhone
and the Pyrenees, where, in the fourth century b. c. and doubtless
later, they were mingled with Ligurians, and also most probably in
Aquitania. Their country had once been occupied by people who
spoke a language akin to Basque and it may be that descendants
;

of this people survived among the Iberians who are noticed by the
ancient writers. The Iberians were, for the most part, short, dark,
and dolichocephalic. There is no evidence that, in Gaul, any Iberians,

so called, dwelt outside the limits which the ancient writers assigned
to that people but it is certain that men who physically resembled
;

—
them dwelt in various other parts of Gaul, in considerable numbers
in the Neolithic Age, here and there at a later time. The original type
of the Ligurians, if it ever existed, cannot be certainly defined but :

they were undoubtedly both short and dark and there is some
;

reason to believe that at least in North Italy, they were mainly

dolichocephalic. When Caesar entered Gaul, Celtic-speaking immi-
grants, who first crossed the Rhine about the seventh century b. c,
had long been settled both in the country of the Iberians and in
'
that of the Ligurians. The people who called themselves Celtae '

were very much mixed. They comprised descendants of palaeolithic

the Rhine with the Germans whom they had left behind : if we could, perhaps
we should find that they had much in common.
 ;-

THE ETHNOLOGY OF GAUL 337

and neolithic races and the latest comers, the conquering Celtic
;

speaking Gauls, were those who had given their name to the entire
group. These conquerors, who doubtless intermarried to some
extent with the peoples whom they subdued, were, for the most
part, tall and fair
; but, although the proportion of men of this type
must have been far greater among the Celtae than it is among the
modern inhabitants of the same country, the majority of the Celtae
were, even in Caesar's time, brachycephalic, short, and dark. The
conquerors of Belgic Gaul belonged, in a large measure at all events,
to the same race as the conquerors among the Celtae ; and, like the
Celtae, the entire group of peoples whom Caesar called Belgae in-
cluded descendants of indigenous races. There is perhaps some
reason to believe that the Celtic-speaking conquerors of Belgic Gaul
included tall dark men ;but this is uncertain. The Gauls, properly
so called— the Celtic-speaking conquerors of Belgic and of Celtic
—
Gaul like their kinsmen, who conquered Lombardy and Piedmont,
closely resembled the Germans both in stature and in colouring, and
were, like them, generally dolichocephalic. But, as the Gauls differed
from the Germans in custom, and as their language was more nearly
related to Latin than to German, it is certain that, although the two
peoples had sprung from the same stock, they had branched off into
two groups, practically distinct. There is, I would suggest, some
reason to believe that red hair was a common characteristic of the
purer Gauls, yellow or flaxen hair of the Germans. When the
Reman delegates told Caesar that the Belgae were of German '

origin ', they probably meant only that the ancestors of the Belgic
conquerors had formerly dwelt in Germany, and this is equally true
of the ancestors of the Gauls who gave their name to the Celtae
but, on the other hand, it is quite possible that in the veins of some
of the Belgae there flowed the blood of genuine German forefathers.
And now I have done. To me the writing of this essay and the
long study upon which it is based have been deeply interesting ;

but I fear that it will be interesting to hardly any one else, except
perhaps the few professed students of ethnology, who will be quick
to detect its many faults. And I do not wonder. For the main
interest of these studies is the certaminis gaudium,— the interest that
belongs to every attempt to solve a difficult problem, the interest
that Adams and Le Verrier felt when they were fighting their way
through the long calculations, which tasked all their powers to the
utmost, that led to the discovery of the planet Neptune. And to
deal successfully with the problems of ethnology requires powers
—
hardly less than theirs such a combination of moral and mental
qualities as is hardly to be found in any one man — enthusiasm, in-
defatigable zest for research, sagacity, judgement, common sense,
perfect clearness of head, lucidity in exposition. Even if all these
qualities could be brought to bear upon the investigation of the
problems which I have set myself, some of those problems would, for
want of evidence, remain, in a scientific sense, insoluble. And even
if they could all be solved, if we could describe exactly the physical

1093 2.
338 THE ETHNOLOGY OF GAUL
type of the Iberians, the Ligurians, and the Gauls, if we could tell
exactly the proportion which each bore in the population of ancient
Gaul, the scoffer would still say. What then? Of what value are your
conclusions ? What do I care whether the Iberians were or were not
the ancestors of the Basques, whether the Ligurians were or were
not the same as the brachy cephalic Celtae ', w^hether this people
'

had broad skulls and dark hair, and that people long skulls and fair
hair ? What I ask is that anthropologists should discuss ethnical
questions in their bearing upon national character for there is no
;

subject upon which more clap -trap has been written by glib para-
graphists ignorant of the rudiments of ethnology. Can anthropology
analyse the elements of the Celtic temperament, refer them to the
' '

various groups that mingled to form the population of Gaul, and

show how far they were due to heredity, how far to environment ?
Can it help to explain why Koman culture so rapidly assimilated the
culture of Gaul ? When it can do all this my interest will awake.^
Such are the posers that the scoffer might set and I am not
;

prepared to say that he would have no reason on his side.

THE NATIONALITY OF THE EBURONES,

CAEROSI, PAEMANI, SEGNI, AND CONDRUSI
The nationality of the Eburones, Caerosi, Paemani, Segni, and
Condrusi is a separate question. Among the Belgic tribes that con-
tributed their contingents to the confederate army which Caesar
dispersed in 57 b. c, the Reman envoys, he says, enumerated Con-
drusos, Eburones, Caerosos, Paemanos, qui uno nomine Germani
appellanturr Caesar himself, in another passage,^ calls the Segni
and Condrusi Germans. MiillenhoS,* however, insists that all these
tribes were Celts. Caesar himself, he remarks, includes them among
the Belgae (B. G., i, 1, § 2) and, according to B. G., v, 27, §§ 4-6,
;

they allowed themselves to be classed as Gauls. All their racial and

individual names, all ancient names of rivers and places within their
territory, are Celtic ^ and whoever classes them among the Germans
;

proper must also class the Walloons, who dwell in the western part
of their territory, among peoples of Germanic descent, whereas they
are really Romanized Gauls. Again, pointing out that, according
to the testimony of Caesar (B. G., vi, 32-4), the customs of the
Eburones were more primitive than those of the rest of the Gauls,
Miillenhoff observes that this persistent conservatism may account
for the fact that the more civilized Belgae of the south-west

^ See the interesting remarks of M. J. Deniker in U Annee psi/chologique,

xiii, 1907, pp. 292-4, and of Professor G. Retziusin Journ. Anthr. Inst., xxxix,
1909, pp. 301-2.
2 B. (?.,ii,4, § 10. ' /6., vi,
32, § 1.
* Deutsche Altertmnsku7ide, ii, 196-204.

^ See also Rev. hist., xxx, 1886, p. 39.

 —

THE NATIONALITY OF THE EBURONES 339

distinguished their neighbours and kinsmen from themselves by the

surname '
Germani '.

Miillenhoff's is a great name but I do not think that his argu-

;

ments establish his theory. The mere fact that the racial and one or
two individual names of the peoples in question and the ancient
names of rivers and places within their territory are Celtic, does not
prove that they themselves were not partly Germanic. Granted
that Ambiorix is a Celtic name so is Boiorix, the name of a leader
:

of the Cimbri yet

;
Miillenhoff, with nearly every modern scholar,
believes that the Cimbri were Germans. The prevalence of Celtic
local names might possibly be accounted for by the fact that the
Eburones and their neighbours had intermingled with an older
Celtic population. Celtic local names were current in Germany
long after the time of Caesar, from which fact it has been inferred
that in Germany the Celts were once predominant but Miillenhoff
;

knew as well as anybody else that their predominance (if it ever

existed) had ceased before the conquest of Gaul. Kent is a Celtic
name but the fact does not prove that the present inhabitants of
;

Kent are Celts.

Miillenhoff quotes B. G., v, 27, § 6, to prove that the Eburones
allowed themselves to be classed as Gauls. Admitting that Am-
biorix really made the statement which Caesar puts into his mouth,
that statement proves nothing. The Eburones, whether Gauls or
not, had thrown in their lot with the Gauls and Ambiorix might
;

speak of them as Gauls in the same sense that a Dutch broker, living
in Fifth Avenue, would call himself an American, or a Jewish
alderman of the City of London an Englishman.
Now, us see exactly what Caesar says. He certainly classes the
let
Eburones, the Condrusi, the Caerosi, and the Paemani among the
Belgae. But in B. G., ii, 4, § 10, he calls them Germani in vi, 2,
:

§ 3, he describes them by the same name and in vi, 32, § 1, he says

;

that the envoys of the Segni and Condrusi begged him not to conclude
— —
that all the Cisrhenane Germans including themselves were of the
same way of thinking Segni Condrusique, ex gente et numero
:

Germanorum, qui sunt inter Eburones Treverosque, legato s ad Caesarem

miserunt oratum ne se in hostiwn numero duceret neve omnium
Germanorum, qui essent citra Rhenum, unam esse causam iudicaret.
It appears, then, that not only did the Keman envoys in 57 b. c.
describe the Condrusi as Germans, but the envoys of the Condrusi
called themselves by the same name. It is also clear that Caesar
believed them. He emphasizes the distinction between the Germans
and the Gauls both in regard to religion and manners {B. G., vi, 21-4)
and in regard to language (i, 47, § 4). His informants had probably,
his interpreters had certainly both seen and conversed with Germans
about whose nationality there could be no dispute.^
'
I do not attach any importance to the argument of M. Plot, that the con-
versation which Caesar records {B. G., vi, 35, §§ 8-9) between the German
Sugambri and their Eburonian captives proves that the Eburones spoke
German ; for the leaders of the Sugambri might have learned, like Ariovistus
(i, 47, § 4), to speak the Gallic tongue,

Z 2
340 THE NATIONALITY OF THE EBURONES
But I am not trying to prove that these tribes were German ; only
to show that there are objections, which Miillenhoff has not fairly
met, to the theory that they were Celtic. Indeed, I believe that he
was, in the main, right. For the truth is, as I have remarked in the
preceding article,^ that we do not know what meaning Caesar's
Belgic informants attached to the word Germanus. Probably they
only meant that the Cisrhenane Germans were descended from
'
'

people who had dwelt on the east of the Rhine and a large tract
;

of what we call Germany had, for a long period; been inhabited by

Celts.2 If the Eburones and their neighbours were called by Caesar
Germani in a special sense, as distinct from the rest of the Belgae,
who also claimed to be of German origin, the explanation may be
that the former were the latest immigrants. It may be that a certain
'
proportion of the Eburones and the other Cisrhenane Germans
'

were descendants of German-speaking ancestors of genuine Teutonic

blood ^ but if so, they had certainly been absorbed, as the evidence
;

of nomenclature shows, by a more numerous Celtic and pre-Celtic

population.

THE POPULATION OF GAUL

Napoleon III, Mommsen,^ and others have attempted to estimate
approximately the population of Gaul in Caesar's time. The data
are as follows. Caesar tells us that from certain tablets which fell
into his hands after the Helvetian war he ascertained that the
Helvctii, who apparently had emigrated en masse, numbered 263,000,
and, along with their allies, 368,000 souls and that of the latter
;

number 92,000, or exactly one-fourth, were fighting men.^ He also

tells us, on the authority of the Remi, that when he marched against
the Belgae, the Belgic tribes, exclusive of the Remi, had undertaken

Seep. 333.
1 2
ggg p 331
M. Jullian {Hist, de la Gaule, ii, 466-7) appears to think that the 'Cisrhenane
^

Germans were really Germans who had become Celticized in culture and
'

speech. But is it likely that they, alone among the Belgic tribes (see p. 331,
supra), learned Celtic in Gaul ? Certainly they could not have learned it
from mere contact with neighbouring tribes; and if they were originally
German-speaking conquerors who became Celticized, they must have been
unaccompanied by women and far inferior in numbers to the Celtic peoples
whom they subdued, and who, with them, formed the Cisrhenane Germans'.
'

Such an assumption is most improbable. See p. 330. The detachment

of the Cimbri and Teutoni, however, who formed the nucleus of the Atuatuci
were, it should seem, left without women. See p. 333, n. 1.
* Former editions of the authorized English translation (iii, 228 [ed. 1881])
of Mommsen's Bdmische GeschicJite contained serious mistakes, due to
a confusion of the German with the English square mile, which were pointed
out in my first edition (pp. 325-7), and corrected accordingly by the late
Professor Dickson in his reprint of 1901.
^ B. G., i, 29,
§§ 2-3. M. Jullian {Hist, de la Gaule, ii, 6), referring to B. G.,
i, 4, § 2, says that, besides the 263,000 Helvetii, numerous slaves must be f

reckoned. But can we be sure that they were ignored in the Helvetian census ?
The number 368,000, which was probably an estimate, based upon the
ascertained number 92,000, may be an exaggeration. See p. 241, supra.
 THE POPULATION OF GAUL 341

to put 296,000 men into the field and he implies that this force was
;

considerably less than their full fighting strength, for he remarks

that the Bellovaci only contributed 60,000 out of an available 100,000.1
Finally, he says that the army which was dispatched to the relief
of Vercingetorix, and to which every Gallic tribe, except the Atuatuci,
Ambarri(?), Ambiliati, Ambivariti, Caerosi, Condrusi, Eburones,
Esuvii (?), Latobrigi, Leuci, Lingones, Mandubii, Meldi, Menapii,
Paemani, Remi, Segni, Suessiones, Treveri, and Viromandui, the
Aquitani generally, and possibly the Diablintes and Namnetes,
contributed, amounted to about 258,000.^ It is obvious that the
statements regarding the Belgic army and the army of relief are
useless for our purpose for in neither of the two cases have we any
;

means ofknowing what proportion the army bore to the male popu-
lation who remained at home. If we might assume that the Helvetian
territory was peopled with exactly or nearly the same density as the
rest of Gaul, we might form an approximate estimate of the whole
Gallic population. But of course we have no right to make any
such assumption. Nor indeed can we be sure that Caesar's state-
ment that the Helvetii emigrated en masse is literally correct.
I agree, therefore, with Desjardins ^ that to calculate the population
of Gaul even approximately is impossible.
J. Beloch, basing his calculations on the Reman estimate, as reported
by Caesar, of the Belgic levies, and assuming that the territory of the
Belgae was less thickly peopled than the rest of Gaul, originally
estimated the whole population, exclusive of that of the Roman
Province, at 3,390,000 * but, after revising his calculations, he
;

substituted for this figure 4,500,000.^

J. Levasseur ^ remarks that, according to Diodorus Siculus,^ the
chief Gallic tribes numbered each about 200,000 and the smallest
50,000 souls. It is probable, he argues, that the average was 100,000
accordingly he estimates the whole population at 6,000,000, pas — '
:

comme un fait certain, ni meme comme un fait probable, mais

comme une simple hypothese.' This caution is certainly justified ;
but to make such hypotheses at all is futile.
It may, however, be admitted that M. Jullian,^ who rarely handles
a problem without illuminating it, has adduced reasons for suspecting
that the highest of the foregoing estimates is too low. According
to Frontinus,^ the Lingones in Domitian's time had 70,000 fighting
—
men, say that they numbered 280,000 all told the present popula-
:

tion of their territory is 580,000. The 100,000 warriors of the

Bellovaci would have represented 400,000 inhabitants,^^ which,
M. Jullian remarks,^^ is the actual population of the Beauvaisis.
' B. G., ii, 4, §§ 5-10. ' lb., vii, 76, § 3.
^ Geogr. de la Gaule rom., ii, 552-3.
* Die Bev'dlkerung der griech.-rom. Welt,
1886, pp. 453-60.
5 Rhein. Mus., N.',F., liv, 1899, pp. 429, 438.
' La population See also H. Molliere, Re-
franfaise, i, 1889, pp. 99-102.
cherches sur revaluation de la population des Gaules, &c., 1892, pp. 22-3.
' V, 25, § 1. 8 Hisi^ de la Gaule, ii, 3-8.

" iv, 10 See, however,

3, § 14. p. 242.
'^ Hist,
de la Gaule, ii, 7 and n. 4.
342 THE POPULATION OF GAUL
The contingent which they were required to furnish in 52 b. c.^ was
only one-tenth of what their effective strength had been five years
before if the other tribes were assessed proportionally, their effective
:

strength in 58 b. c. must have been 275,000 x 10, which would imply

a population of eleven millions.^ Plutarch ^ reckoned that the Gauls
whom Caesar encountered in the field numbered three millions, while
Appian's * estimate was four millions, numbers which presuppose —
a population of from twelve to sixteen millions. The area, comprising
the departments of the Nord, Pas-de-Calais, Somme, Seine-Inferieure,
and Oise,5 and inhabited by the Bellovaci, Suessiones, Ambiani,
Caleti, Veliocasses, Viromandui, Atrebates, and Morini, which in
57 B. c. contributed 180,000 men to the Belgic host,^ was only one-
twentieth of Gaul. The 180,000 men, if the several contingents bore
the same ratio to the whole available strength as the contingent
of the Bellovaci, were selected from 300,000, representing a popula-
tion of 1,200,000 multiply this by 20, and you will see that the
:

Gallic population, not including slaves, amounted to 24,000,000.

Now what first strikes the sceptical reader is that these divers
methods of calculation not only rest upon unverifiable and, in some
cases, most improbable hypotheses, but also lead to very different
results. M. Jullian insists that the numerical statements of the
writers whom he quotes are trustworthy. But the estimates which
he deduces from them range from eleven to thirty millions and is ;

it not evident that to multiply 1,200,000 by 20 w^hich implies that —

the whole of Gaul could sustain as dense a population as the very '

fertile territories of the Suessiones

'
'^
—
can only result in gross exag-
geration ? Did not the mere fact that the 100,000 warriors of the
Bellovaci would have represented a population as great, or nearly as
great, as the actual population of the Beauvaisis suggest to M. Jullian
a suspicion that the number 100,000 was wrong ? ^ There is nothing '

more difficult in war,' says Colonel G. F. R. Henderson,^ than to '

get an accurate estimate of the enemy's numbers, especially when

civilians are the chief sources of information.' Either Plutarch or
Appian was certainly misinformed, for there is a discrepancy of a
million between their estimates and when we see that, according
;

to the former, Caesar took more than 800 GaUic towns by storm,
whereas he himself only records the capture of eight, our faith
receives a shock. Perhaps inevitable error will be minimized if,
making a due reduction in view of the probable excess in the number
ascribed to the Bellovaci at their full strength, we provisionally
accept the figure based upon Caesar's enumeration of the contingents
that were raised for the relief of Vercingetorix. Considering what

1 B. G., vii, 75, § 3. - C. JulHan, Hist, de la Ganle, ii, 8, n. 3.

3 Caesar, 15. •»
Celtica, 2.
These appear to be the departments which M. Jullian {Hist, de la Gaule,
^

ii, means by the cinq departements du nord '. But the Morini also
8, n. 3) '

inhabited a part of Belgium, and the Suessiones a part of the department of the
Aisne. ^ B. G., ii, 4,
§§ 5, 7, 9.
' Suessiones . . . latissimos feracissimosque agros possidere {ib., §§ 6-7).
^ Seep. 242. * Stonewall Jaclson, i, 1898, p. 401.
 THE POPULATION OF GAUL 343

Strabo ^ tells us of the fecundity of the Gallic women, and that the
population of Kent, the conditions of which closely resembled those
of Gaul, appeared to Caesar, or to the pseudo-Caesar ', immense
' ' 2
;
'

remembering, on the other hand, that great cities did not then
exist in Gaul, and that much land which is now cultivated had not
then been reclaimed,^ we may be inclined to acquiesce in the belief
that M. Jullian's lowest estimate is not very far above the truth.
But I would ask him, before he pins his faith to the statements
of the ancient writers, to consider an illustration drawn from modern
history. In 1857 William Tayler, the Commissioner of Patna,
—
observed that the population of that city the capital of his Division
— was estimated at 400,000 '. An official holding such a position
'

in Gaul would certainly have been regarded by M. Jullian as an

unimpeachable authority. But, according to the census taken in
1872, the population of Patna was 158,900.*

Mv, 1, §2. ^. G^.,v, 12, §3

2 14, § 1.
; See p. 206, n. 4.
^ M. Jullian himself i, 283) emphasizes the comparative
{Hist, de la Gaule,
smallness of the habitable area of Gaul.
* See Rice Holmes's Hid. oj the Indian Mutiny, 5th ed., 1898 (or the reprint

of 1904), p. 179, n. 1.
 SECTION III,—PURELY GEOGRAPHICAL
CAESAR'S WANT OF PRECISION IN GEOGRAPHICAL
STATEMENTS
The Commission de la topographic des Gaules remark that Caesar
must have known that the Bituriges Vivisci occupied both banks of
the Garonne.^ It is clear, then, they argue, that when he spoke
of the Garonne as the common frontier of Aquitania and Gallia
Celtica,^ he did not intend to speak with literal accuracy. Again,
part of the territory of the Veliocasses, who were a Belgic people,
was on the left bank of the Seine and the Seine and the Marne,
;

according to Caesar,^ formed the boundary between the Celtae and

the Belgae. These inaccuracies are worth noting, because some critics
have gone astray by basing investigations on the assumption that
every geographical statement in Caesar must necessarily be regarded
as rigidly precise. But the Commentaries are quite accurate enough
for every practical purpose, if we exercise a little common sense in
interpreting them.

THE TERRITORIES OF CLIENT TRIBES

It is necessary, in order to define the territories of certain Gallic
tribes, to ascertain whether Caesar included within their limits the
territories occupied by tribes which he calls clientes. Deloche takes '*

Walckenaer and B. Guerard to task for reckoning client tribes as

pagi, or mere subdivisions of the tribes whose overlordship they
recognized. This view, he argues, finds no support in the Commen-
taries for Caesar does not apply the word pagus to any of the tribes
;

which he calls clientes and the Carnutes and other client peoples
;

are expressly designated by him as civitates. Now there can be no

doubt that, as a general rule, Caesar distinguishes the territories
of client tribes from the territories of their overlords. Thus he says
that the Usipetes and Tencteri made their way into the territories
'

of the Eburones and the Condrusi, who are clients of the Treveri
{in fines Eburonum et Condrusorum, qui sunt Treverorum clientes,
pervenerant) ^ and he says that the Carnutes were clients of the
;

Remi,^ from whom they were separated by the Suessiones, the

1 Diet. arch, de la Oaule, i, 71. =^

B. 0., i, 1, § 2.
=>
lb.
•*
31 em. frhentis par divers savants a VAcad. des Inscr., 2® s^r., iv, 1860,
pp. 365-70. 6 B.O., iv, 6, § 4. « lb., vi, 4, § 5.
 THE TERRITORIES OF CLIENT TRIBES 345

Meldi, and the Parisii. On the other hand, the passage ^ in which
he remarks that the Cevennes separates the Helvii from the Arverni
can only be explained on the assumption that he included in the
territory of the Arverni that of their clients, the Vellavii.^

THE MAP OF GAUL

I. I warn those who may read the articles in my Geographical
Index or those which deal with the geography of Caesar's campaigns
that it is absolutely impossible to make this book self-sufficient in the
matter of maps. For in the course of the articles I have often been
obliged to mention streams, hills, and villages which are not marked
in any maps except the sheets of the Carte de VEtat-Major, the
Topographische Atlas der Schweiz, or the Carte topographique de la
Belgique. My own maps are intended merely to illustrate the
narrative not to illustrate the arguments by which I have arrived at
;

the results which the narrative records. Those who may wish to
control the statements in my articles must buy or borrow the various
sheets to which I refer therein for themselves. The scale of the
Carte de VEtat Major is goioo' ^^ about one mile and a quarter to an
inch.^ Based upon it, is a map drawn on the scale of 3 2 o^o o o ^^^ ^^® >

Service geographique de Varmee have also issued a third map on the

scale of ^ooVoo ^^^ scale of the Topographische Atlas der Schweiz
•

is ^05^0 ^^^ of ^he Carte topographique de la Belgique 20000* "^^^

•

these maps are beautifully executed. An admirable little book, called

La Carte de V Etat-Major : Guide pour sa lecture, has been written by
Captain J. Molard. It only costs ^d., and enables any one who
masters it to read the map perfectly.
II. The map of Gaul in this book will fail to satisfy minds which
are intolerant of an avowal of ignorance. Headers of this kind will
find what they want in the maps of Napoleon, von Goler, von Kampen,
Kiepert, and others and
they find discrepancies between the con-
; if

clusions of those authorities, they will have no cause to complain

that any of them is an agnostic. But any one who has the patience
to read my geographical notes will find that even Kiepert has marked
on his maps towns, camps, and battle-fields, the sites of which it is
impossible, with our present materials, to determine and that he ;

has marked them without warning the reader that his identifications
are purely conjectural. It is simply impossible to construct a com-
plete map to illustrate the Gallic War which shall not be misleading.
A map which, like Kiepert's and von Goler's, traces the whole network
of Caesar's lines of march certainly looks much prettier than one
which omits many of them but it is not scientific
: and whoever ;

pubHshes such a map ought to put his readers on their guard.

HI. Generally speaking, it is impossible to determine the frontiers
^ £. G'.,vii, 8, §2.
2 Diet. arch, de la Gaule, ii, 17. Cf. p. 518, n. 3.
3 A new edition on the scale of
^^^ is now being published.
346 THE MAP OF GAUL
of the Gallic states with certainty. With one
or two exceptions, one
may indicate their position on the map one may determine approxi-
:

mately the extent of country which they occupied and in a few ;

cases some portion of the actual frontier can be traced with precision.
But if the reader finds that maps which, like those of Napoleon III,
have the boundaries of the states marked, help him to realize the
story more vividly, he should distinctly understand that those maps
are so far for the most part conjectural. I am aware that d'Anville
believed that it was possible to reconstruct the map of Gaul even in
this detail. En general,' he affirmed, le gouvernement Eccl6siae-
' '

tique en France a ete regie sur le gouvernement Civil, tel qu'il etoit
lors de I'etabHssement du Christianisme dans les provinces de la
Gaule en sorte que les anciens Dioceses repondent aux territoires
;

des anciens peuples.' ^ By way of proving his thesis, he says that

there is a town called Feins on the confines of the dioceses of Orleans
and Blois, another town of the same name on the confines of tlie
dioceses of Orleans and Chartres, and a third on the confines of
the dioceses of Sens and Auxerre that each of these towns stands
;

upon the site of one of the numerous Gallo-Roman frontier-stations

called Fines and that consequently les limites de ces Dioceses
;
'

sont les memes que les Fines des Cites de Chartres, d'Orleans, de
Sens et d' Auxerre sous I'Empire Romain '. He also says that between
Alise-Sainte-Reine and Montbard there is a town called Fins, situated
on the confines of the dioceses of Autun and Langres, a sure proof
— '

that the Aedui and the Lingones had the same limits under the
Roman Empire and that the Fines of the Table, near Aquis-
;
'

Segeste, is on the confines of the dioceses of Orleans and Sens, which

proves that the boundaries of those dioceses are identical with the
'

boundaries of the Carnutes and Senones '.^ D'Anville's conclusion

is certainly very plausible but it would be rash to conclude that
;

because the territories of certain Gallo-Roman peoples coincided with

certain ancient French dioceses, therefore the territories of all the
Gallo-Roman peoples might be similarly defined. It would be still
more rash to assume that all the Gallo-Roman provinces exactly
coincided with the territories of the Gallic states that preceded them.
^
It is true that the stations called Fines were frontier-stations ;

but, says a French writer, Malheureusement I'emplacement de ces

'

Fines n'est pas tou jours facile a determiner, et, quand on parvieut
k le faire, il ne Concorde pas toujours avec les limites des anciens
dioceses.' *
Desjardins admits that, in principle, the boundaries of the Gallic
^ £!claircissemens sur Vancienne Gaule, 1741, pp. 234-5. See also A. Longnon,
Atlas hist, de la France, 1884, p. iii. The Eclaircissemens, which I have just
quoted, are commonly attributed by French scholars to the Abbe Belley but ;

M. Jullian {Rev. des etudes anc, x, 1908, p. 351), with whom I agree, is inclined
to believe that they were really written by d'Anville, under whose name they
were published. In liis Notice de Vancienne Gaule he frequently refers both to
the Eclaircissemens and to Belley, but never attributes the former, with which
he invariably agrees, to the latter.
^ {Eclaircissemens, pp. 191, 453.
•*
Diet. arch, de la Gaule, i, 397. ^ Ih., and 399, no. viii.

1
 THE MAP OF GAUL 347

peoples whom Caesar subdued served to define the territories of the

sixty civitates of Augustus and that subsequently the boundaries
;

of those civitates served, in principle, to define the areas of episcopal

jurisdiction. Mais,' he proceeds, dans le passage de Tetat autonome
'
'

de la Gaule a I'organisation romaine il faut se garder de croire

. . .

que les delimitations des Soixante cites de I'epoque romaine corre-

spondissent exactement a celles des peuples, dont les frontieres
politiques etaient souvent vagues et indeterminees : des peuplades
entieres avaient ete exterminees comme celles des Eburons et des
Aduatuques ; d'autres, sensiblement diminuees par la guerre, se
trouverent fondues, sur differents points, dans les cites voisines on ;

en voit surgir d'autres qui proviennent d'un dedoublement opere,

sinon par Auguste, du moins par ses successeurs, afin de propor-
tionner a peu pres les territoires entre eux le pays des Eduens et de ;

leurs sujets etant juge beaucoup trop vaste pour former une seule
cite, on dut en detacher celui des Segusiavi leurs anciens clients. . . .

C'est pour les memes causes sans doute que la cite des Tricasses . . .

fut creee plus tard dans une portion detachee des Senones ou des . . .

Remi.^ ^ In another place Desjardins speaks of the difl^culty occa-

sioned by ces liens de patronage et de clienteles si repandus dans la
'

Gaule, liens qui empechent d'isoler ces peuples et parfois meme de

les distinguer entre eux.' ^ Thus the Aulerci Brannovices were
'
clients '
of the Aedui, and, as clients, possessed, in Caesar's time,
a territory of their own. But was afterwards merged
this territory
in one of the dioceses which are generally assigned to the Aedui ;

and impossible to say how much of that diocese belonged to the

it is
Aedui, how much to the Aulerci. Again, Caesar makes no mention
of certain peoples, for example the Silvanectes and the Tricasses,
whose territories afterwards formed separate dioceses and it is ;

impossible to decide whether these were client peoples or pagi,^ and

to which, if to any of the better known states their territories
belonged. Moreover, it is certain that several of the states mentioned
in the Notitia 'provinciarum were not formed into dioceses, and that
several dioceses even of the fifth century do not figure as states in
the Notitia.^ Finally, Deloche points out that bishops sometimes
encroached upon the dioceses of their neighbours.^
Nevertheless, it would be a great mistake to jump to the conclusion
that our knowledge of the boundaries of the ancient dioceses is of no
use to the student of Gallic geography. Those dioceses do probably
often correspond exactly or nearly with the territories of Gallic states ;

for there is no reason to suppose that, except in certain cases, which

will be mentioned in the proper places, the areas of the Gallo-Roman
differed materially from those of the Gallic territories. The cases in
which the evidence of the dioceses fails us ^ will also be duly noted in
^ Oiogr. de la Oaule rom., ii, 428-9. ^ lb., ii, 217.
3 See p. 344.
* See J. Loth, U
Emigration bretonne en Armorique, 1883, p. 49.
^ 3Iem. presentes par divers savants a V Acad, des inscr., iv, 1800, pp. 326-7.
« See K. Thomann's Der Jranzosische Atlas zu CUsars gall. Kriege, 1868,
pp. 3-4, and Rev. du Lyonnais, 3^ ser., i, 1866, p. 191.
348 THE MAP OF GAUL
the Geographical Index. To the student of Caesar the matter is of no
practical importance, except in so far as it affects the search for the
sites of places which are mentioned in Caesar's narrative. Only it is
not always safe, in that search, to place absolute reliance upon argu-
ments drawn from the areas of the ancient dioceses.^ Moreover,
although Caesar often speaks of the frontiers of Gallic tribes,^ it seems
probable that, as in India in the early days of the British occupation,
these frontiers were more or less variable, portions of territory being
periodically won or lost as the result of intertribal wars, and the
possession of river-valleys, which involved the power of collecting
tolls, being constantly in dispute.-*^ Thus the territories of a tribe
would be the area in which it was able for the time being to collect
revenue.^ Caesar, however, of course described the frontiers as they
existed in his time.
IV. One more point demands consideration. If, in drawing a map
of Gaul, one traces the coast-line as it appears in the map of contem-
porary France, the map will be so far misleading for the coast has
;

undergone considerable changes. Accordingly Desjardins, Kiepert,

and A. Longnon have attempted to trace the coast-line as it existed
in the time of Caesar. But even their maps cannot but be misleading
to a certain extent for although it has been proved that the sea
;

has in some places gained upon the land, and that certain tracts of
land have been won from the sea, it is, generally speaking, impossible
to say exactly what was the dividing line between sea and land in
the time of Caesar and Desjardins himself admitted that his maps
;

were largely conjectural. Nevertheless, even a conjectural map,

based upon a study of the numerous monographs that have appeared
upon the subject, should be more trustworthy than a mere reproduc-
tion of the modern map. Moreover, the tracing of the coast-line, for
the most part, has no interest except for physical geographers :

in three instances only, which will be duly mentioned in the notes,

has it any bearing upon the study of Caesar's campaigns.^

^ M. J. F. Blade {Annales de la Faculte des lettres de Bordeaux, 1893,

p. 115)
argues that it is impossible to establish a correspondence between the area of
any given diocese and the territory of any independent Aquitanian tribe.
For independent Aquitania was represented under the late Empire by ten
civiiaies, six of which are generally supposed to correspond with the territories
of the following six independent tribes — Elusates, Ausci, Tarbelli, Vasates,
Bigerriones, and Boi [or Boiates]. But independent Aquitania contained
twenty-seven tribes in all. Therefore, concludes M. Blade, the territories of
twenty-one of them would be represented by four civitates. But may not
some of the twenty-seven have been included in the other six (see my article
on the Tarbelli [pp. 477-9]) ? Still, it is impossible to prove that any given diocese
did not include the territory of some insignificant tribe besides that of the
tribe with which it is commonly associated.
^ B. a, i,
6, §§ 2-3; 8, § 1 ;ii, 15, § 3
; v, 24, § 2
; 26, § 2; 54, § 2; vi, 44,
3 Seep. 353.
§3; vii,5, §4.
^ These remarks have been suggested to me by conversation with Sir Alfred

Lyall.
5 Any one who may wish to study the question should refer to A. E. E. Des-

jardins, Geogr. de la Gaule rom., i, 188-398 ; A. Longnon, Atlas hist, de la

France, -p-p. i-ii ; Ilim. couronnes par V Acad. Roy. des sciences etbelles-lettres de
Bruxelles, vi, 1827, pp. 23-4, 26, 74-7, 149-50, 171 ; A. Belpaire, ^tude sur la
 349

THE ITINERARY OF ANTONINE AND THE TABLE

OF PEUTINGER
I shall often have occasion in the following notes to refer to the
Itinerary of Antonine and the Table of Peutinger. The compilation of
the former was probably begun in the reign of Antoninus (Caracalla) ;

but, as it mentions Diocletianopolis and Constantinopolis, it can

hardly have been completed before the middle of the fourth century.
The figures contained in that part of it which relates to Gaul are
usually followed by the letters P M M
{milia 'plus minus), except in
—
two routes from Fines Helvetiorum to Argentorate (Strasbourg)
and from Lugdunum (Lyons) to Gesoriacum (Boulogne), where the
distances are indicated both in Roman miles and in Gallic leagues ;

—
and in three others from Co Ionia Agrippina (Cologne) to Castra
leg(ionis) XXX
(Birten ?), from Durocortorum (Reims) to Treveri
(Treves), and from Treves to Cologne —
where they are indicated in
Gallic leagues only. Nevertheless, it is universally admitted that,
except in the case of the Province, the figures actually denote Gallic
leagues .1 The edition which I have habitually used is that of
Wesseling but the edition of Parthey and Pinder (1848) may also
;

be consulted.
The so-called Table of Peutinger, in the form in which it has come
down to us, was the work of an anonymous writer, the monk of '

Colmar,' who was living in 1265. It was discovered at Worms in

1494, and in 1508 passed into the hands of Conrad Peutinger. The
original document upon which the monk worked belonged to the
imperial epoch but its date cannot be ascertained. The best
;

edition is that of Desjardins. See Geogr. de la Gaule rom., iv,

72-3, and Desjardins's La Table de Peutinger, p. 76 ff.

formation de la plaine maritime depuis Boulogne jusqu' an DanemarJc, 1855, pp. 197,
200-1 ; A. de Laveleye, Affaissement du sol et envasement des fleuves, 1859, p. 8 ;

}i. Dehrsiy, Etude geol. . du littoral flamayid, 1813 Bull, de la 8oc.de geogr.
. . ;

d'Anvers, i, 1877, pp. 155-88 Rev. des Deux Mondes, 2® per., Ixxix, 1869,
;

pp. 429-53 ; Mem. du Congres scientifique de France, ii, 1872, pp. 451-60 ;
Bull, de la Soc. de geogr., G^ ser., x, 1875, pp. 225-41 Bull, de geogr. hist, et
;

descr., 1901, pp. 313-41 Bull, de la Soc. geol. de France, 4^ ser., vi, 1906,
;

pp. 142-7 ; Rice Holmes, Anc. Britain, &c., pp. 517-8 C. Jullian, Hist,
;

de la Gaule, i, 6-12, 22, n. 6 and Rev. des etudes anc, xi, 1909, p. 361. See
;

also pp. 681-3, 691-6, infra. I am inclined to think that in constructing my

map twelve years ago I in some cases accepted evidence for changes of coast-line
which was inconclusive but for the reason implied in the text I leave the
;

errors uncorrected.
^ Desjardins, Geogr. de la Gaule
rom., iv. 37-8 I tin. Ant., ed. Wesseling,
;

pp. 251-2, 254-6, 358-63, 365-7, 372-3. See also Corpus inacr. Lat., xiii,
pars ii, fasc. 2, p. 646.
 350

THE GALLIC LEAGUE AND THE ROMAN MILE

I. It is generally admitted, on the strength of certain statements
of Ammianus Marcellinus ^ and Jornandes ^ and of certain passages
in the Itinerary of Antonine,^ that the Gallic league was equivalent
to 2,222 metres, or one Roman mile and a half and the orthodox;

view is that, whenever in the Itinerary or the Table distances are

computed in leagues, a league of this length is meant. In reality
one Roman mile and a half is equivalent to 2,217| metres. But
according to T.Pistollet de St.Ferjeux,^ this league was only the Gallic
league as officially recognized by the Romans the true Gallic league,
:

he maintained, was 2,415 metres.

If I were writing a treatise upon Gallic geography, it would be my
duty to discuss his theory as carefully as I have studied it but, ;

as I am only concerned with Gallic geography in so far as it is related

to the history of Caesar's conquest and of the events which led up
thereto, I may leave it alone. For Pistollet does not dispute any of
the identifications that have been proposed of the towns which are
mentioned in Caesar's narrative.
II. The Roman mile was 1,617 yards, or 1,478J metres. In
W. Smith's Dictionary of Antiquities, ii, 159-60, will be found a full
account of the various methods by which the length of the Roman
foot has been computed it was -2957 metre.^ From this result the
:

length of the mile has been determined.

1 XV, 11, § 17; xvi, 12, §8.

- Gelica, xxxvi, 192 [Monumenta Germaniaehist., auctores antiquissimi v, 1882.
,

p. 108).
'
Ed. Wesseliiig, 1735, pp. 359-63.
* Mem. sur Vancienne lieue gaiiloise,'pp. 9-13, 18-20. iSee also Bcv. rfes Soc.
saviintes, S^ ser., ii, 1803, pp. 186-91 iv, 1864, p. 449
; J/em. de V Acad, du
;

Gard, 1863-4, pp. 110-20; 1866-7, pp. 109-19; Rev. arch., nouv. ser., vii,
1863, pp. 344-9 xiv, 1866, pp. 194-7
; xv, 1867, pp. 444-6
; La Table de
;

Peutinger, pp. 10, col. 2 25, col. 3

; 33, col. 2-3
; Desjardins, Geogr. de la
;

Gaiile rom., iv, 23-5 and C. Jullian, Hist, de la Gaule, ii, 395, nn. 3-4. M. Jullian
;

{Rev. des etudes anc, ix, 1907, p. 189) suggests that there is an allusion to
the Gallic league in Posidonius [Fragin. hist. Graec.,ed. Didot,iii, 1849, pp. 260-1,
fr. 25), who says that the Arvernian king, Luernius, had enclosures, 12 stades
square, made for festivals 12 stades, M. Jullian points out, are equivalent to
:

2,220 metres.
^ Cf. Daremberg and Saglio, Diet, des ant. grecques et rom., iii, 1730. Tlie
leading authority is F. Hultsch, Griech. und rijm. Metrologie, 1882, pp. 88-98.
Cf. Mem. de la Soc. nat. des ant. de France, Ixviii, 1908 (1909), pp. 444-6.
 GEOGRAPHICAL INDEX '

Admagetobriga. See Magetobriga.

Aedui. —
The territory of the Aedui, if we include in it, as most
geographers do, the territories of their clients,^ the Aulerci Branno-
vices and the Mandubii (q.v.), comprised, roughly speaking, at all
events, the dioceses of Autun (Augustodunum), Chalon-sur-Saone
(Cavillonum), M'lcon (Matisco), and Nevers (Noviodunum), the last
three of which were severed from the primitive diocese of Autun .^
This territory corresponded with the departments of the Sa6ne-et-
Loire and Nievre and parts of the Cote-d'Or and the Allier. Desjar-
dins has shown in his posthumous volume that in the third century of
our era, if not in Caesar's time, the Aedui also possessed the territory
which belonged to the civitas Autessiodurum (Auxerre).*
When Caesar said that the Loire separated the Aedui from the
Bituriges, he evidently meant that it served as the common frontier
of the two peoples in that part of its course to which the Aeduan
force came in 52 b. c, when they had been sent to the assistance
of the Bituriges.^ This part of the river, it should seem, lay some-
where between its junction with the Allier and Sancerre. But the
diocese of Nevers embraced a narrow strip of territory on the left
bank of the Loire, south of a point about 3 miles north of
La Charite, which is itself about 12 miles south-east of Sancerre.^
If, then, as would seem probable, the Aeduan force approached the

Loire between Nevers and La Charite, Caesar's delimitation of the

frontier does not coincide with the western boundary of the diocese
of Nevers. M. Jullian, however, pointing out that intertribal frontiers
in Gaul were rarely formed by rivers, argues that Caesar's statement
must not be taken literally^ I cannot agree with him. Consider
what Caesar says :
—
When they [the Aeduan force] reached the
'

—
Loire the boundary between the Bituriges and the Aedui they —
lingered there a few days, and then turned back without venturing
to cross the river (qui cum aijiumen Ligerim venissent, quod Bituriges
'

'
I generally pass over places like Avaricum and Lutecia, the sites of which
have either never been disputed or have been finally identified with such
certainty that they are no longer disputed even by charlatans, though, where
it seemed necessary, I have briefly given the reasons for the identification ;

and, in speaking of the territories of Gallic peoples about the geographical

position and the extent of which there is no dispute, I simply state the facts
and give the necessary references.
^ See pp. 344-5, and C. Jullian, Hist, de la Gaule, ii, 538, n. 5.
^ DAnville, Notice de Vancienne Gaule, p. 35 Walckenaer, Qeogr. des Oaules,
;

i, 325 ; Diet. arch, de la Gaule, i, 14.

* See GoRGOBiNA and Senones. ^ B. G., vii, 5, §§ 3-4.
^ See M. Clairefond's map in Bull, de la Soc. d' emulation du departement de
V Allier, vii, 1859, p. 284, and cf. the map facing p. 1 of the twelfth volume
of Gallia Christiana.
^ Festschr. zu 0. Hirschfelds sechzigstem Geburtstage, 1903, p. 217.
352 AEDUI
ah Haeduis faucos dies ihi morati neque flwnen transire ausi
dividit,
domum revertuntur). To my mind these words clearly imply that the
Aedui, if they had crossed over, would have found themselves in
the territory of the Bituriges.i Either they approached the Loire
more than 3 miles north of La Charite at Sancerre or even —
further down the valley or the Aedui did not possess that strip —
of territory west of the Loire.
According to Strabo,^ the Saone separated the Aedui from the
Sequani but the French Commission disregard his statement and
;

follow the indications of the dioceses, remarking that Strabo's text

'
n'a rien d'absolument affirmatif quant a la non-discontinuite de la
limite le long du fleuve, comme il est facile de s'en convaincre en
examinant 1' ensemble de la phrase '.^ Strabo may have made a mis-
take, but his language seems precise enough pet Sk kuI 6 "Apap ck :
—
TMv "A\7reo)y, opit,(Dv '^rjKoavov's re kol AlSovovs- Ptolemy * also says that
the Saone was the eastern boundary of the Aedui. Caesar says that
the Saone flows through the territories of the Aedui and the Sequani,
'

and discharges itself into the Rhone' (per fines Haeduorum et Sequa-
no7'um in RJwdanum influit).^ From these words also von Goler ^
infers that the Saone formed the boundary between the Aedui and
the Sequani but, as R. Schneider says, he mistranslates the '^

—
:

passage and Napoleon refers to three others in the Gallic War

;

inter fines Helvetiorum et Allohrojum Rhodanus fiuit (i. 6, § 2),

. . .

cum Sequanos a p'ovincia nostra Rhodanus divideret (i, 33, § 4), and
flumen Ligerim quod Bituriges ah Haeduis dividit (vii, 5, § 4) to
. . . —
show that if Caesar had meant what von G5ler says, he would have
expressed himself differently. Caesar simply meant that the Saone
flowed through the territories of the Aedui and the Sequani considered
as one tract.^ P. Guillemot,^ differing from von Goler, holds that
*
la Bresse chalonnaise belonged to the Aedui, arguing that when
'

they complained to Caesar that the Helvetii were ravaging their

lands,^^ the Helvetii had not yet crossed the Saone. I do not think
that this inference can legitimately be drawn from Caesar's narrative ;

for he appears to have attacked the Tigurini directly after the

^
Bituriges acknowledged the overlordship of the Aedui
The (Z?. G., vii,
and therefore presumably did not dispute their right to
5, § 2), collect tolls
on the river. If so, the Aedui would not have been tempted to encroach
upon the western bank. See p. 353.
- Oeogr., iv, 1, § 11 ^ Diet. arch, de la Gaule, i, 14.
; 3, § 2.
4 Geogr., ii, 8, § 12. B. G., i, 12, § 1. « Gall. Krieg, 1880, p. 15.
^

^ Jahresb. d. philol. Vereins zu Berlin, xii, 1886,

pp. 240-1.
^ Thomann [Der franzbsische Atlas zu COsars gall. Kriege, 1868, p. 9), com-

menting on Napoleon's remarks, asks, Are we then to conclude that the '

tribes mentioned in B. G., iv, 10, lived on the right as well as on the left bank
of tlie Rhine ? The well-known passage to which Thomami refers, and
'

which, as Meusel has shown (see p. 692), is probably spurious, runs as follows :

Rhenus .per fines Nantuatium, Helvetiorum, Sequannrum, Mediomatricorum,

. .

Tribocorum, Treverorum citatus fertur. In my note on the Triboci I show that,

unless the writer made a slip, some portion of the territory of that people must
have been on the right bank, as otherwise per must mean the same as practer
(' past or' along ')
' and there is no evidence that per ever bears this meaning.
;

® Excursions dans les montagnes de la Cote-d'Or, 1861, p. xx.

^0 5. (?.,i, ll,§§2-3.
 ;

AGEDINCUM 353

xlcdui made their complaint and, at the moment when he attacked

;

the Tigurini, three-fourths of the Helvetii had crossed the river.^

M. JulUan ^ argues that the reason why rivers were rarely frontiers
is implied in Strabo's ^ further statement that the Aedui and the
Sequani disputed about the collection of tolls on the Saone. The
possession of both banks of a river was necessary, he maintains, in
order to enforce payment. No doubt but if the Aedui had per-
:

manently possessed both banks, would there have been a struggle ?

Strabo's statement surely implies that the riparian territory v/as
debatable land.
My own belief is that no certain conclusion can be drawn from
any of the ancient authorities as to the eastern boundary of the
Aedui. Caesar's words, even though they do not prove that it was
formed by the Saone, are not irreconcilable with the view that it
was : on the other hand, Strabo and Ptolemy, if they did not
misunderstand Caesar, may only have meant that the Saone was,
roughly speaking, the boundary. On the whole, it seems to me safer
to prefer the evidence of the dioceses to their uncertain statements.
It seems probable that after Caesar had restored the hegemony of the
Aedui they acquired permanent possession of the disputed territory.
Agedincum has always been generally identified with Sens but ;

a few commentators have decided for Provins, which is about

25 miles, as the crow flies, from Sens, and nearly due north of it.
This theory, however, is no longer defended. Provins is on the
right bank of the Seine and Labienus, when he decided to abandon
;

his campaign against the Parisii, in order to return to Agedincum,'*

crossed from the right to the left bank. For Caesar, describing the
position in which Labienus found himself when he was about to
return, says,^ on one side he was menaced by the Bellovaci
'
. . .

on the other Camulogenus held the field with a well-found army,

ready for action {altera ex farte Bellovaci
'
. instabant ; alteram
. .

Camulogenus farato atque instructo exercitu tenehat). This passage

proves that Labienus was, at that moment, on the right bank of
the Seine. For he was on one bank, and Camulogenus on the other.
Therefore, if he was on the left bank, Camulogenus was on the right
and Camulogenus and the Bellovaci were both on his right whereas ;

Caesar says that he was between them. Achaintre makes a desperate

attempt to anticipate and turn this argument. Caesar, he urges,
was only speaking generally
'
the Bellovaci were preparing for
'
:

war the Parisii, on the other hand, were already prepared. But
;

that Caesar used the words, altera ex parte, &c., in a strictly geo-
graphical sense, is proved by his having, in the same breath, used
the phrase alteram (partem) Camulogenus 2)arato atque instructo
exercitu tenehat.
then, that Agedincum was not Provins.
It is clear, There is
evidence to show that it was Sens and as this evidence is similar
;

' B. G., i, 12, 13, §§ 1-2.

^ Festschr. zu 0. Hirschfelds sechzirjstem Gehartdlayc, 1903, p. 218, n. 1.
' iv,
3, § 2. ^ B. G., vii, 59-62. = lb., 59,
§ 5.
1093 Aa
354 AGEDINCUM
to that by which the sites of Lutecia, Avaricum, and many other
Gallic towns, were identified centuries ago, it may be well to state it.
Caesar says that Agedincum was in the territory of the Senones.i
Now the chief town of the Senones, in the early days of the Church,
was Sens, or, as it was then called, Senones and the chief town of
;

the Senones was, according to Ptolemy ,2 'Ayr/StvoV. Again, in the

Itinerary of Antonine,^ Agredicum is marked as 13 Gallic leagues,
or 19 J Roman miles, from Condate (Montereau-sur-Yonne), and as
30 Gallic leagues, or 46 Roman miles, from Augustobona (Troyes)
and the actual distances, according to Cassini,^ are 21 and 43 J Roman
miles respectively.^ Finally, an inscription of the year 250 a. d.,
in which the name AGIED (incensium) occurs, was discovered, in
1837, at Sens.6
A. dc Barthelemy has, however, suggested "^
that, although a
Gallo-Roman town Agedincum undoubtedly stood upon the
called
site of Sens, the Gallic town of the same name may have l^een
situated elsewhere. Just as Gergovia was succeeded by the Gallo-
Roman Augustonemetum, and Bibracte by the Gallo-Roman Augus-
todunum, so, he suggests, the Agedincum which stood upon the site
of Sens may have been the Gallo-Roman successor of the Agedincum
mentioned by Caesar. I cannot accept this suggestion for if the ;

Gallic Agedincum had yielded its position as the chief town of the
Senones to a Gallo-Roman foundation, the latter would not have
been called by a Celtic name, but by some name of which Caesar, as
in Caesarodunum, or Augustus, as in Augustonemetum, would have
formed an element.
Similar groundless suggestions have been made in regard to
Durocortorum (Reims) and Lemonum (Poitiers).
—
Alesia. The site of Alesia is absolutely certain. It covered the
plateau of Mont Auxois, on the south-western slope of which now
stands the village of Alise-Sainte-Reine. But, as it was the scene of
the most famous event in the Gallic war, the question of its where-
abouts has given rise to a controversy, the echoes of which have not
yet died away. M. Ruelle, in his Bibliographie generale des Gaules
(pp. 163-72), enumerated 158 works bearing uj)on the subject and ;

since 1870, the date which he fixed as his limit, new pens have been
busy with the same theme. The controversy was not closed by the
publication of the results of Napoleon's investigations. Jules
Quicherat, who was, in his day, one of the best known of French
antiquaries, remained unconvinced C. Miiller, in the Atlas which
:

he published in 1880, to illustrate Strabo's Geography, marked

Alesia on the site of Alaise ^ and other writers of less note, while
;

differing among themselves as to the true site, agree in rejecting

Napoleon's conclusion. I was therefore bound, by the principles

'
B. G., vi, 44, § 3. 2 Geogr., ii, 8, § 9. « Ed. Wessehng, p. 383.
* Walckenaer, Geogr. des Gaules, iii, 54.
^ Sec also La Table de Peidinger, ed. Desjardins, p. 26, col. 2.
^ Rev. de philologie, ii, 1847, p. 355
; Corpus inscr. Lai., xiii. 2949.
7 Rev. celt., viii, 1887, p. 398.
^ Strabonis Geographicorum Tabulae XV, Praefatio, p. vi.
 ALESIA 355

which guided me in writing this book, to discuss the question in the

first edition ; and
notwithstanding the multitude of treatises
as,
which had appeared, the whole of the arguments had never been
marshalled within any one woik, I thought that it might save future
inquirers trouble if I supplied the want. Within the last few years,
and especially since the commencement of the excavations, still in
progress, which have proved that a and afterwards a Gallo-
Gallic,
Roman town stood upon Mont Auxois, various writers have attempted
to reopen the question. Like the enthusiasts who have formed a
society to prove that the earth is flat, such writers will never be
convinced and perhaps a generation will pass before the truth is
;

left unassailed. But although the question is now regarded by all

scholars as closed, intelligent readers will naturally want to know
the reasons upon which the truth is founded. I will therefore
re-state them as briefly as I can.
Besides Mont Auxois, seven sites have been proposed, namely
Alaise in the department of the Doubs, Alais in the department of
the Gard, Novalaise and the plateau of La Crusille in Savoy, Izernore
in the department of the Ain, a place, the name of which I cannot
discover, in Auvergne,^ and Aluze in the department of the Saone-et-
Loire.2 Izernore and La Crusille are dismissed by A. de Barthelemy,
in an admirable article on Alesia,^ as unworthy of discussion but ;

as Izernore was advocated by a laborious student, and still finds

support,"* I shall examine its claims here. The pretensions of Alais,^
Novalaise, and the above-mentioned (or any) site in Auvergne, are
so wildly absurd that it is difficult to believe that any one could have
advocated any of the three, except from a love of singularity. Aluze
is a new candidate, having only made its appearance in 1906 but ;

its supporter is not deterred by the mere fact that during centuries
of discussion it had been overlooked. The real controversy, however,
has always been between the champions of Mont Auxois and those
of Alaise.
Caesar gives a description of Alesia, of the camp which Vercin-
getorix formed upon its eastern slope, and of his own camps and
lines of contravallation and circumvallation, the gist of which I
have embodied in my narrative ^ and need not reproduce here. The
additional data which he supplies for determining the geographical
position of the stronghold may be stated in a few words. When
he was passing through the furthest part of the territory of the
'

Lingones {fer extremos Lingonum fines) towards the country of

'

the Sequani, Vercingetorix encamped at a distance of 10 Roman

miles from his camp. Next morning Vercingetorix attacked the
^ Comptes rendus de V Acad, des inscr., 4® ser., xvii, 1889 (1890), p. 410.
2 Pro Alesia, 1908, pp. 274, 276.
^ Eev. des questions hist., iii, 1867, pp. 43-4.
* Comptes rendus de V Acad, des inscr., 1906, pp. 724-5.
^ Alais, which was advocated in 1696 by Des Ours de Maiidajors (Nouvellea
decoiivertes sur Vetat de Vancienne Gaule dit temps de Cesar) is at least 200 miles,
as the crow flies, from the nearest frontier of the Lingones, in whose comitry
Caesar was the day before he reached Alesia {B. G., vii, 66-8).
« See pp. 169-73.
A a 2
356 ALESIA
Roman coliann, when it had advanced to within a short distance
from his own camp. Beaten in this engagement, he retreated to
Alesia ; and the Romans arrived there on the day after the battle.^
I. Alesia, according to J. Maissiat, stood upon the site of Izernore,
near the Perte du Rhone and 9 kilometres, or between 5 and 6 miles,
north-west of Nantua. The whole weight of his theory rests upon his
interpretation of the famous passage in which Caesar describes his
own position and that of Vercingetorix on the night before the
action which immediately preceded the blockade of Alesia, cum
Caesar in Sequanos per extremos Lingonum fines iter faceret, quo
facilius subsidium provinciae ferre posset, circiter milia passuutn ah X
Romanis trinis castris Vercingetorix conseditP' I discuss, on pages79 1-4,
the meaning of per extremos Lingonum fines, in so far as those words
bear upon the question where the action was fought. The passage
means that, at the time of which Caesar wrote, he was marching
through the country of the Lingones towards the country of the
Sequani. Nothing of the kind, says Maissiat Caesar : indique '

seulement la direction de sa marche actuelle, a savoir, du pays des

Lingons, chez les Sequanes et vers la Province '.^ Again, Maissiat
says that in this passage in does not mean vers or dans la direction
' ' '

de for, in order that in should have either of these meanings,

'
;

'
il nous parait indispensable que le texte presente quelque raison

accessoire et particuliere, comme pourrait etre la nature du verbe

employe par I'auteur,' ^ as, for instance, instituit, in the passage
iter in Senones facere instituit.^ Now there is one passage in the
Commentaries which proves conclusively that Maissiat is wrong. In
jB. G., ii, 29, § 4, Caesar, speaking of the Atuatuci, whites ipsi erant :

ex Cimhris Teutonisque prognati, qui, cum iter in provinciam nostram

atque Italiam facerent, iis itnpeditnentis quae secum agere ac portare
non poterant citra flumen Rhenum depositis custodiae ac praesidio VI
milia hominum una reliquerunt. At the time of which Caesar wrote,
the Cimbri and Teutoni were marching in provinciam nostram atque
Italiam ; and they had not yet come anywhere near either of those
places. Therefore in, without any raison accessoire et particuliere ',
'

does mean towards '

and cum Caesar in Sequanos per extremos
'
;

Lingonum fines iter faceret means that Caesar w^as marching through
the country of the Lingones towards the country of the Sequani.^
Thus the foundation of Maissiat's theory is undermined and it is ;

needless to examine the buttresses by which he strives to sustain the

shattered structure but perhaps I should mention that the Gallic
:

name of Izernore was not Alesia but Isarnoduros.'

1 B.G.>
vii, 66, § 2 ; 67-8. ^
/^^ ^jj^ qq^ § 2.
^J2des Cesar en Gaule, ii, 269.
* lb., 271-2. 5 js Q^ vii, 06, § 5.
" M. A. de Barthelemy justly argues that, if Caesar had already penetrated

into Sequania, he would have written, not Her faceret but iter fee isset. Rev. des
quest, hist,, iii, 1867, p. 50. Further proof, if it be required, will be supplied by
a comparison of B. C, i, 39, § 3 with i, 60, § 5.
' A. Holder, Alt-celtischer iSprachschatz, ii, 76. It is incredible that this
name sliould have been given to the stronghold after the conquest if it had
ever borne the famous name of Alesia.
 . -

ALESIA 357

II. I now come

to the knot of the question. I have sufficiently
established the meaning of "per extremos Lingonuni fines. There may
be room for doubt as to the exact spot in the country of the
Lingones to which Caesar referred but it is absolutely certain that
;

on the night before the battle he was somev/here within that country.
As the battle-field was not more than 10 Koman miles from this
point, and as, before the battle began, Caesar was marching towards
the country of the Sequani, it is clear that the battle-field was either
within the country of the Lingones or only just south of the Saone,
which separated their country from that of the Sequani. I show
on page 800, and it is universally admitted, that Alesia could hardly
have been more than 35, or perhaps at the very outside 40 miles
from the battle-field. Now, on the south of the Saone the only
—
conceivable site the only site that has ever been suggested, answer-
—
ing to these conditions is Alaise. On the north of the Saone, the
—
only conceivable site the only site that has ever been suggested
—
at all is Mont Auxois. Between these two the choice must lie.
Quicherat tries to show that the names of various localities in the
neighbourhood of Alaise recall various scenes in the drama of Alesia,
and that the topography of Alaise corresponds with Caesar's descrip-
tion of the topography of Alesia. Now even when the meaning of
names of places is certain, their evidence should be used with great
caution. But when a writer, determined to make out his case by
hook or by crook, arbitrarily attaches this or that meaning to the
name of a place, his arguments may be safely ignored. En regie '

generale,' says M. A. de Barthelemy,i je ne me fie pas plus aux

'

lieux-dits qu'aux traditions. Ces sources alterees a chaque siecle par

I'influence de I'imagination populaire, aidee de ce que les erudits
peuvent y ajouter en passant, ne peuvent que faire composer une
histoire fantastique.' ^ As for topography, the following reasons are
sufficient to convince any one who has a good map and can read it
that Alesia was not Alaise. First, Alaise is not a hill at all, but
a mountain mass, rising into numerous hills, various in form and
elevation, nearly all wooded, and separated from one another by
ravines. Secondly, the plain, about 3 miles long, which extended
'
on the west of Alesia, does not exist at Alaise. The only plain '

which the advocates of Alaise can point to is situated in the vallej''

of the Todeure,^ and forms an inclined plane, not a plain, about
1,000 yards long and 165 yards wide the cavalry battle which
:

Caesar describes could not possibly have taken place here. Thirdly,
Alaise is not surrounded, as Caesar's description requires, by hills
about as high as itself, but on the north, the west, and the south
by hills of very unequal altitudes, diverse forms, and chaotic arrange

^ Rev. des quest, hist., iii, 18G7, p. 43.

^ Here are a few instances of the way in which Quicherat manipulates lieux- *

dits'. Mouniot, he says, is derived from munitorium, Chateley from castellare,

Chataillon from castellio, &c. &c. {Melanges <farc7i. et (Vliist., \, 518-20). There-
fore Alaise= Alesia !

•'
In the Carte de VEtat-lIajor (Feuille 126) the Todeure is called Ruisseau de
'

Conche '
358 ALESIA
ment, and on the east by the vast plateau of Amancey, which rises
to a height of 707 metres. Fourthly, if the Gallic army had been
encamped on the eastern slopes of Alaise, the fortifications which
Caesar describes would have been superfluous for nature had forti- ;

fied the site with huge precipices.^ Fifthly, the smallest line of
contravallation which could have been drawn round Alaise would
have been more than 22 kilometres in extent w^hereas the line of ;

contravallation which Caesar drew round Alesia was not more than
10 2 Roman miles, or less than 15 kilometre^.^ Sixthly, the labour
of constructing and of defending such elaborate lines of contravalla-
tion and circumvallation as Caesar describes, on this vast scale,
would have been beyond the power even of Caesar's army. Seventhly,
this stupendous labour would have been labour thrown away for ;

the mountain mass of Alaise was easily accessible, on the west and
south, by the pass of la maison Pourtalis and Montfordes
'
and '
;

Caesar could thus have gained possession of the western and southern
parts of the mountain, which dominated the site of the alleged
oppidum.^ Lastly, Quicherat, finding it impossible to discover on
the north of Alaise a hill corresponding with the description which

^ Chataillon, where the camp has been placed, is far too small and, as it is ;

protected by the precipitous banks of the Lison, the Gallic fortifications would
have been unnecessary or, if they had been made, they would have looked not
;

towards the east but towards the west, facing the town. Quicherat, recognizing
the objections to this site, placed the camp in the neighbourhood of Saraz but, ;

as the Due d'Aumale points out, Saraz is not on the east but on the south of
Alaise. See Rev. des Deux Mondes, 2" per., xv, 1858, p. 133, note.
^ Or, according to the a MSS., 11. Caesar does not expressly say that
the contravallation was 10 miles in extent he only says that his original
:

line of investment was and it is not certain that the two were identical. This
;

suggestion, which I made in the first edition (p. 373), is supported by M. Pernet's
description of the excavations {Pro Alesia, 1907, p. 280).
^ Perverse and wrong-headed as he was, Quicherat was no fool, and he saw

clearly enough that the mountain mass of Alaise was too large to have been
surrounded by the lines of contravallation and circumvallation which Caesar
described. Accordingly he set to work to distort and force, if by any means
it were possible, the narrative of Caesar into some sort of agreement with the
topography of Alaise. He insists {Melanges d'arch. et d'hist., i, 530-9) that,
after Vercingetorix withdrew his army into the town, Caesar entirely changed
his plan of blockade. From Caesar's description, he says, on se croirait '

revenu au moment ou le blocus vient d'etre decide.' The works which Caesar
describes in B. G., vii, 72 did not surround Alesia, but, as we may gather
—
from the topography of Alaise observe how naively Quicherat begs the
—
question only one side of it, namely that on which is situated Charfoinge ;

for on the other sides natvire had done his work for him. Why, then, I may
ask, did Caesar change his plan of attack ? Had he neglected to examine
the ground before he originally invested Alesia ? Caesar, continues Quicherat,
speaks of the lines of contravallation and circumvallation as his own work ;

whereas he speaks of the entrenchments that defended the camp of Rebilus

and Reginus as the work of his men {nostri), —
premier indice que ce camp
'

n'adherait pas aux lignes qui enveloppaient la ville.' This, as Dr. Johnson
would have said, is sad stuff and no unbiased person who has read the Com -
;

mentaries would thank me for refuting it. If anything about Alesia is certain,
it is that Caesar's lines surrounded the place entirely, and that every stroke of
work was done by his men {nostri) under his supervision.
* See Desjardins, Geogr. de la Gaule rom., ii, 696, n. 3, 697, nn. 1-3, 699, n. 1,

700, n. 1, and Bev. des Deux Mondes, 2'' per., xv, 1858, pp. 122-34.
 ALESIA 359

Caesar gives of that on the southern slope of which was situated

the camp that Vercassivellaunus attacked, coolly identifies it with
the plateau of Amancey. La colline au nord,' comments the Due
'

d'Aumale, scrait le plateau d' Amancey, qui est situe a Test, et qui
'

'
a soixante-quatre kilometres de tour !

These facts prove that Alaise does not correspond with Caesar's
description of Alesia but if any one is not convinced, let him
;

read pages 125-34 of the Due d'Aumale's article in the Revue des
Deux Mondes, and I warrant that he will make up his mind.
But there are other arguments equally conclusive. The geogra-
phical position of Alaise suits Caesar's description no better than its
physical features. Alesia had been provisioned in advance. If, then,
Alesia was Alaise, it must be admitted that Vercingetorix had known
Caesar's intentions for weeks before Caesar started on his march
for the Province and, what is more, that he knew exactly what
;

route Caesar would take. That he possessed such foreknowledge is,

as the Due d'Aumale argues, incredible. It may be replied that the
assumed identity of Alesia with Mont Auxois is open to the same
objection. But the two cases are widely different. For at Mont
Auxois Vercingetorix was in a central position from which he could
strike at Caesar, whatever route Caesar might take in marching to
the relief of the Province whereas if Vercingetorix established
;

himself at Alaise, he must have expected that Caesar would take

that route through the country of the Sequani which would lead
him to the Jura and this was just the route which Caesar would
;

have avoided. Again, it is obvious that, in taking up his position

at Alaise, Vercingetorix would have left the all-important city of
— —
Bibracte the political centre of Gaul exposed to Caesar's attacks.
As the Due d'Aumale says, Le proconsul pouvait par une marche
'

rapide fondre sur cette ville, peut-etre I'enleyer par un coup de main,
peut-etre detacher du parti national les Eduens mecontents. En
tout cas, la position excentrique prise par I'armee gauloise aurait
laisse le champ libre au genie de Cesar.' ^
On the theory that Alesia was Alaise, it is impossible to discover
any site for the cavalry action which immediately preceded the
blockade of Alesia. I show, on pages 791-4, that, on the night
before the action, Caesar encamped somewhere on the north of the
Sa6ne.2 Nobody will believe that Vercingetorix would not have
attacked him while he was endeavouring to cross that river and ;

to cross a river in the face of an enemy is one of the most difficult

^ Rev. des Deux Mondes, 2® per., xv, 1858, pp. 84-5, 90, 94-5. Heller {Philologus,
xiii, 1858, p. 595) argues, further, that if Vercingetorix had intended to take
up a position in Sequania at all, he would probably have occupied its strong
capital, Vesontio (Besan^on), not Alaise. But is it not possible that, a a
G. Veith suggests {Gesch. der Feldziige C. J. Caesars, p. 498, n. 1), Caesar may
have continued to hold this important position smce 58 B.C. ?
^ Quicherat simply drives a coach and four through this difficulty. He mis-
translates yer extremos Lingonum fines, and he mistranslates circiter milia
passuum X ah Romanis trinis castris Vercingetorix consedit au moment ou
:
'

Cesar passait de la frontiere des Lingons en Sequanie . Vercingetorix se tint,

. .

pendant trois etapes, a environ dix milles des Romains.'

360 ALESIA
of military operations. A. Delacroix, however, one of the most
notorious champions of Alaise, places the battle-field in the neigh-
bourhood of Cugney, near the northern bank of the Ognon. But,
objects the Due d'Aumale, if we accept this site, w^e still have to
admit that Vercingetorix, the greatest general of Gaul, was so
imbecile that he sat idle in his camp and allowed the Eomans to
cross the Saone without molestation. As this assumption is incredible,
we must look for the battle-field south of the Ognon. Between
Pontailler and the forest of Serre there is a site which might perhaps,
by a prejudiced inquirer, be regarded as agreeing with Caesar's
description of the battle-field. But even on this theory, the Romans
would have virtually crossed the Saone in presence of the enemy ;

for Vercingetorix would have been quite near enough to dispute the
passage. In any case, then, to quote the Due d'Aumale, we must
admit that, within the space of two days, Caesar crossed three, or even
—
four rivers the Saone, the Ognon, the Doubs, and the Loue that ;

he crossed the Saone on the morning of the battle and in the

presence of an apathetic enemy that of all these rivers he only
;

mentioned one and that within those two days he also fought a
;

battle, and marched at least 60 kilometres, or 36 miles, over a broken

and wooded country. Such an admission would be absurd. There-
fore Alesia was not Alaise.^
One word more. After the fall of Alesia, Caesar went to the
country of the Aedui. Either he took his army or a part thereof with
him or he went alone. In any case, after the surrender of Vercin-
getorix, he sent Labienus, with two legions and the cavalry, into the
country of the Sequani.2 Alaise is in the country which belonged
to the Sequani. If, then, Alesia was Alaise, and if Caesar took his
army with him, he needlessly marched Labienus's large force, as well
as his other eight legions, across the Saone, and then sent it back
again. If he left Labienus at Alesia, it was at Alesia that Labienus
received orders to go to Sequania. But in that case, if Alesia was
Alaise, Labienus received orders to go from Sequania into Sequania,
which is absurd.
III. If the foregoing arguments are sound, Alesia can only have
stood upon the plateau of Mont Auxois. Intrinsically, the reasons for
accepting this view, w^hich was adopted at least as early as the ninth
century, are (1) that the geographical position of Mont Auxois alone
agrees with the indication which Caesar gives of his own whereabouts
on the day before he reached Alesia (2) that its physical features
;

alone agree with Caesar's description of Alesia (3) that excavation

;

has revealed (a) remains of the Callic town on the site of which the
recently discovered Gallo-Roman one was built, (h) unmistakable
traces of Caesar's lines both of contra vallation and of circumvallation,
(c) numbers of weapons both Roman and Gallic, as w^ell as numerous
bones of men and of horses, (d) five barbed spikes such as Caesar
describes, in his inventory of the siege-works, under the name of

^ Eev. des Deux Mondes, 2* per., xv, 1858, pp. 90-3.

2 B. G., vii, 90, §§ 1-4.
 ,

ALESIA 361

stimuli, (e) Gallic pottery, and (/) —

619 coins, none of which are of
a later date than 52 B. c, the year of the siege, and one of which
bears the image and superscription of Vercingetorix ^ (4) that a ;

Gallic inscription, in which occurs the word Ali/sia has been dis-
covered at Alise-Sainte-Keine ^ (5) that in the territory which the
;

advocates of Mont Auxois, but not those of any other site, assign
to the Mandubii, there has been found an inscription, now preserved
at Dijon, containing the name Mandu-bilos,^ a spelling which is
supported by the MavSi/^ovkoyy of Strabo,^ and (6) that the geogra-
phical position of Mont Auxois alone agrees with the description
which C-aesar gives of his distribution of the legions, after the capture
of Alesia, into winter- quarters. To quote M. A. de Barthelemy,^
'
de toutes les localites oii Ton a propose de placer Alesia, Alise-
Sainte-Reine est la seule qui presente les caracteres les plus certains :

le texte de Cesar, la description des travaux de siege, et les decouvertes

archeologiques et numismatiques s'accordent a faire accepter cette
identification.' ^
The coins are fully described on pages 555-61 of the second volume
of Napoleon's Histoire de Jules Cesar. One hundred and thirty-two
were Roman four hundred and eighty-seven Gallic. It is significant
;

that the latter include specimens belonging to twenty-four different

states, —
more than half of those which sent contingents to the reliev-
ing army that no less than one hundred and three of the whole
;

—
number belonged to the Arverni, the countrymen of Vercingetorix ;

and that, like most of the bones, they were discovered in the trenches
of the very camp — —
the camp on Mont Rea which, assuming the
identity of Alesia with Mont Auxois, must have been the scene of
the final struggle.'
1 proceed to deal with objections.^
1. The most rational comes from the pen of Captain Gallotti ^ but ;

Rev. arch., nouv. ser., i, 1860, p. 271 ; Napoleon III, Hist, de Jules Cesar
^

ii, 555-61
; A. Blanchet, Traite des monn. gaul., p. 495, n. 4 ; Pro Alesia, 1906,
p. 31 ; 1907, pp. 122-5, 157-9, 173-5, 207-8, 248-50, 277-81 ; 1908, pp. 300-3,
379-80 ; 1909, pp. 505-7, 534-7, 596 ; Juin, 1910, p. 2 of cover.
2 Corpus inscr. Lat., xiii, pars i, no. 2880.
M. A. T. Vercoutre {Pro Alesia,
1907, p. 193) points out that, according to Consentius {Ars . . de harharismis,
.

&c., ed. H. Keil, 1868, p. 394), the Gauls gave to i a sound intermediate between
the Latin e and i. Thus they would have pronounced Aosiia (=Alisea)
in away which Caesar would have reproduced as Alesia.
R. Mowat, Inscr. de la cite des Lingons, F^ partie, p. 35, no. 37, quoted by
^

H. d'A. de Jubainville, Les noms gaulois cliez Cesar et Hirtius de Bello Gallico,
1891, p. 128.
* Ed. Miiller and Diibner, iv,
2, § 3 and p. 962.
^ Eev. des quest, hist., iii,
1867, p. 66.
® The various antiquities discovered at Mont Auxois by the French Com-

mission are to be seen in the ' salle d' Alesia of the Musee de St. Germain.
'

See also Diet. arch, de la Ga%de, i, 36-9, and Journal des Savants, 1880, pp. 561,
563-4.
' See Rev. des quest, hist., iii, 1867, p. 65.
^ I ignore various
perverse and frivolous objections, made by Quicherat,
which I refuted in the first edition (pp. 370-2).
^ 3rem. de la Soc. d' emulation dn
Doubs, 4*^ ser., i, 1866, pp. 361-75, and
especially 364-5, 368, 370, and 374.
362 ALESIA
it is safe to say that he would have withdrawn it if he had known that
the choice lay between Mont Auxois and Alaise, and that the objec-
tions to Alaise were unanswerable. If Alesia was on Mont Auxois,
it is certain that the hill which, Caesar says, extended so far to the
north that he had not been able to include it within his circum-
vallation, was Mont Rea. Now, observes Gallotti, Rea is connected
with the plateau of Menetreux by a col the circumvallation would :

naturally have crossed the col and in order to give it this extension,
;

Caesar would only have had to increase its length by two kilometres.
Let any one look at the map, and he will see that Gallotti's statement
is correct. But surely the natural conclusion to be drawn from
Caesar's words is that he had not had time, before the arrival of the
relieving army, to increase the length of his circumvallation by
a fraction which would have been about one-eighth of the whole.
Again, Gallotti argues that Vercassivellaunus would not have
required 10 hours to march the 4 kilometres which separated his
camp from Mont Rea but he forgets that it was necessary for
:

Vercassivellaunus to make a long detour, in order to avoid observa-

tion ^ for his chance of storming the camp of Caninius would be
;

greatly increased if he could attack it unawares.^ Nor can I see any

force in Gallotti's objection that Caesar would not have called the
defences of the camp on Mont Rea superiores munitiones? They were
higher than the contravallation and, as Caesar himself says that
;

the camp was on gently sloping ground (leniter dedivi loco ^), we
' '

may reasonably infer that it was not high up the hill. If the lines
of contravallation and circumvallation which the excavations re-
vealed were not Caesar's, one must admit that a position which
corresponds exactly with Caesar's description of Alesia was blockaded
in a manner which corresponded exactly with his description of the
blockade that the blockader defended himself, like Caesar, against
;

a relieving army and that of these vast operations no record

;

remains. The chances against such a coincidence are almost infinite.

2. Maissiat denies that Vercingetorix could have got water on
Mont Auxois. 5 But, to quote Napoleon, ^ Near the western summit '

of the mountain two abundant springs arise and there is another ;

on the eastern side. Besides, manifest traces of a great number

. . .

of wells are visible on the plateau, so that it is evident the besieged

can never have wanted water, besides w^hich they could always
descend to the two rivers.' See also Pro Alesia, 1907, pp. 124-5 ;

1910, p. 627.
I have now proved the assertion with which I began this note.
No sane man who has studied the subject will ever again deny that
Alesia stood upon the plateau of Mont Auxois. The question should

^ See Napoleon, Hist, de Jules Cesar, 309-10, and Pro Alesia, 1909, p. 581.
ii.
^ The distance from the nearest point camp to the rear of Mont
of the Gallic
Rea, where Vercassivellaunus halted {B. G., vii, 83, § 7) is 5|, not 4 kilometres ;
and the time which he spent on his circuitous march from some period in the —
first watch to daybreak —
may not have been more than 7 hours.
'
3 B.G., vii, 85, § 4. * lb., 83, § 2.
° Jules Cesar en Gaule, iii, 52-3. " Hist, de Jules Cesar, ii, 300, n. 1.
 ^

ALESIA 363

now be closed. Andany man with an eye for a country, who

if

knows his Caesar and happens to be travelling to or from Dijon,

will get out at the station of Les Laumes, walk up to the top of Mont
Auxois, and look about him, he will marvel that the question should
ever have arisen.
—
AUobroges. We learn from Caesar that the Allobroges were
separated from the Helvetii by the Rhone that they possessed
;

certain lands on its right bank that their territory was conterminous
;

with the territories of the Vocontii, the Segusiavi, and the Nantuates ;

and that the most northerly town in their country was Geneva .^
Their other important towns were Vienna (Vienne) and Cularo
(Grenoble).^ It is clear, then, that the greater part of their territory
lay between the Rhone, the Isere, and the Lake of Geneva. Roughly
speaking, it comprised, as the French Commission remark,* the
territories of the civitas Viennensium, the civitas GratianopoUtana,
and the civitas Genavensium. But in the opinion of the Commission,
in order to determine their frontiers, it is not enough to unite the
dioceses of Vienne, Grenoble, and Geneva, because nous nous '

trouvons ... en presence de diverses petites populations reunies

a de plus grandes sans que nous puissions bien saisir la loi qui a preside
a ces reunions.' For example, the dioceses of Geneva and Grenoble
included the territories of the Nantuates, Veragri, Ceutrones, Medulli,
and Tricorii while it is probable that of the diocese of Lyons certain
;

portions on the left bank of the Rhone belonged to the Allobroges.

These statements, however, are not altogether accurate at all :

events, I can see no reason for including either the Nantuates (as
a whole) or the Veragri in the diocese of Geneva. That diocese
extended eastward along the southern bank of the lake to a point
about midway between £]vian and its eastern extremity, and did
not include St. Maurice, which belonged to the Nantuates while ;

we learn from Caesar that the Veragri possessed territory on the

southern bank of the upper Rhone. At all events, the reasons
given by the Commission need not prevent us from tracing approxi-
mately the frontiers of the Allobroges.
Eastward their territory extended as far as Evian, or a little
beyond it.^ The Nantuates and Veragri, who were their eastern
neighbours, certainly possessed the towns of Tauretunum, which was
close to the point where the Rhone enters the lake,^ St. Maurice, and
Martigny ;and the eastern boundary of the diocese of Geneva
may serve to mark, at least approximately, the eastern boundary
^
M. R. Bouton has just published a history of the controversy in Mem. de la
Soc. d' emulation du Douhs, 8*^ ser., iii, 1908 (1909), pp. 43-83.
^ B.G.,i,Q>,%2; 11, § 5 ;10, § 5 6, § 3
; iii, 1, § 1
; G, § 5.
;

3 Ptolemy, Geogr., ii, 10, § 7 ;Cicero, Fam., x, 23, § 10.

* Diet. arch, de la Gaule, i, 41-2.
^ Desjardins {Geogr. de la Gaule rom., ii, 242) thinks that the Dranse or the
spurs of the Alps, which abut on Evian and Thonon, would have formed
a natural boundary.
® See Gregorj'' of Tours, Hist. Franc, iv, 31 {Patrologiae cursus completus, ed.
Migne, Ixxi, 294), Marius of Avenches {ib., Ixxii, 799), and Desjardins, Geogr.
de la Gaule rom., ii, 234, 242.
364 ALLOBROGES
of the Allobroges. The Ceutrones, who occupied the Tarentaise,
were their neighbours on the south and the famous inscription
;

of Forclaz, which was discovered in the valley of the Arve, between

Chamonix and Sallanches, enables us to trace the frontier of the
Ceutrones, as itwas in the time of Vespasian. The tracing of the
frontier which took place then was probably, as Desjardins says, an
official delimitation of that which had existed before. It ran along
the valley of the Arly to its confluence with the Isere, and thence
along the ridge which separates the Tarentaise from the Maurienne.
The last-named valley belonged to the Medulli, who, like the Ceu-
trones, were conterminous with the Allobroges. So too were the
Uceni and the Tricorii, who occupied respectively the valleys of
the Romanche and the Drac.^ Desjardins, who does not believe
that rivers generally served as frontiers, makes the southern frontier
of the Allobroges leave the Isere on the south, and extend across the
two last-named valleys along the foot of the hills but he believes
;

that it may have followed the line of the Isere from the mountains
of St. Nizier to the Rhone.^ Who can tell ? Rivers sometimes did
serve as frontiers, for instance between the territories of the Aedui
and the Bituriges.^
De Valois, remarking that part of the diocese of Vienne was in the
Vivarais, argued that the Transrhodane possessions of the Allo-
broges, which Caesar mentioned, were on the frontier of the Helvii
(q.v.), below the confluence of the Saone and the Rhone.'* But these
Transrhodane possessions were ravaged by the Helvetii ^ and, as ;

d'Anville points out,^ the Helvetii would not have marched in that
direction, as they were bound for the country of the Aedui. Still, it
is possible, that, besides the Transrhodane possessions w^hich Caesar
mentioned, the Allobroges had others in the tract indicated by de
Valois Debombourg, who finds fault with Napoleon for tracing
."^

^ I can see no reason for including the territories of the Uceni and Tricorii, as
M. Longnon does {Atlas hist, de la France, pi, 1 ), apparently because they were
in tlie diocese of Grenoble, in the territory of the Allobroges. '
Clients they
'

may have been ; but the Tricorii, at all events, were plainly regarded by Livy
(xxi, 31, § 9) as an independent people. See myarticle on the Vocontii.
2 Geogr. de la Gaule rom., ii, 236.
3 B. G., vii,
5, § 4. See, however, pp. 351-3, supra. Debombourg {Rev. du
Lyonn-ais, 3'' ser., i, 1806, p. 456) argues that a letter from Plancus to Cicero
{Fam., X, 23, § 7) proves that Cularo was on the frontier of the Allobroges, and
therefore that the Isere, in a part of its course, was their southern frontier,
— namely from Vinay and St. Marcellin on its right bank and St. Gervais and
Iserononits left asfaras its confluence with the Rhone. Between these limits
the Isere is a natural barrier, difficult to cross. Not so in the upper part of its
course there the Allobroges possessed both banks as far as the foot of the
:

mountains of Belladonne. Debombourg may be right but Plancus only said

;

—
that Cularo was ex finihus Allobroqum, in the territory (not on the frontier)
of the Allobroges. * Notitia Galliarum, p. 608.

5 B. G., i, ® Notice de Vancienne Gaule, p. 54.

11, § 5.
7 Debombourg {Rev. du Lyonnais, 3® ser., i, 1866, pp. 455-6) infers from
Strabo (iv, 3, § 4) that the territory of the Allobroges extended on the west of
the Rhone as far as the chain of the Vivarais and, he adds, Strabo' s statement
;

is confirmed by the fact that the diocese of Vienne extended on the right bank of
the Rhone as far as Doux, near Tournon. But Strabo only says that the
 ALLOBROGES 365

the comuioii frontier of the Sequani and Allobroges without regard to

natural boundaries, gives the Allobroges trans Rhodanum the archi- '

pretres ' of Belley, Virieu-le-Grand, and Arbignieu, that is to say,

a triangular tract of land of which the apex is near Tenay, and which
is bounded by the chainon d'Inimont and the chainon de Parve '.^
'
'
'

The French Commission agree with dAnville in assigning them

those parts of the dioceses of Geneva and Belley which are on the
right bank Rhone, that is to say, a strip of land extending
of the
from Bellegarde to Mar, comprising the val Romei, the districts
of Chatillon and Michaille, and the Bugey.^ On this theory, however,
the Helvetii, who traversed the country of the Sequani on their
march from Geneva to the Saone,^ would have passed out of their
territory into that of the Allobroges almost immediately after they
emerged from the Pas de I'Ecluse. Desjardins, having regard to
Caesar's statement ^ about the complaints which the Transrhodane
Allobroges made to him of the devastation of their lands by the
Helvetii, and to the fact that Caesar was then in the angle between
the Rhone and the Saone, infers that their territory lay further
westward, and gives them the cantons of Amberieux, Menemieux, and
Montluel.^ See Nantuates and Sequani.
—
Ambarri. The name of the Ambarri appears to be preserved in
Amberieu and Arabronay. Their territory formed that part of the
civitas Lugdunensium which lies between the Rhone and the Saone,
excepting the small tract which belonged to the Segusiavi (q.v.).
The French Commission ^ remarks that, after defining the territories
of the surrounding peoples —
the Aedui, the Sequani, the Helvetii,
—
the Allobroges, and the Segusiavi we find remaining for the Am-
barri a district which corresponds roughly with the department of
the Ain. The Commission, however, do not admit that the Segusiavi
possessed any territory between the Rhone and the Saone. According
to dAnville,'^ the northern boundary of the Ambarri reached its
northernmost point near Macon. But the Commission, following the
traces of the ancient dioceses, extend it to the river Seille, and trace
its other boundaries along the Saone, the Rhone, and the mountains
of Bugey. They admit, however, that the northern frontier, as
traced in their map, is quite conjectural, and that dAnville may be
right. For my part, I cannot see why they do not firmly adhere
to the principle by which they are usually guided. I believe therefore
that, besides the central and western parts of the Ain, the Ambarri
possessed the south-eastern part of the Saone-et-Loire. Desjardins,^

peoples beyond the Rhone and the Saone and between the Loire and the Seine
are adjacent to [napaKfiTai) the Allobroges ;and, as he includes the Carnutes
among these peoples, it is plain that he used the word TrapaKfiTai loosely.
The argument based on the western extension of the diocese is, however,
reasonable enough.
^ Rev. du Lyonnais, 3*' ser., iv, 1807, pp. 10-2.
''
Diet. arch, de la Gaule, i, 43. ^
j^ g^^ j^ y .
k)^ ^
j;
n, § l.
.

* Ih., 5 Qc,o(jr, de la Gaule roni., ii, 005.

10, § 5.
" Diet. arch, de la
Gaale, i, 48.
^ Notice de Vaneienne Gaule, maj) facing p. 1.
^ Geogr. de la Gaule
rom., ii, 005.
366 AMBARRI
who restricts their territory by extending that
of the Segusiavi on
the eastern bank of the Saone as far north as Mr'>con, is, I believe,
so far wrong. M. d'Arbois de Jubainville ^ appears to think that the
Ambarri, originally at all events, possessed territory on both banks
of the Saone for he derives their name from Arnbi-arari,
; ceux qui — '

habitent sur les deux rives de I'Arar ' (Saone). Debombourg ^

assigns to the Ambarri the archipretre ' of Anse on the western
'

bank of the Saone, which is usually assigiied to the Segusiavi, and

the archipretre ' of Morestel on the southern bank of the Rhone,
'

which is usually assigned to the Allobroges (q.v.). To sceptics,' he '

says, I should reply by pointing to Amberieux in Anse and Ambla-

'

gnieu on the southern bank of the Rhone.' But whatever may be

the etymology of Amberieux, it is evident from Caesar's narrative ,3
read in conjunction with Ptolemy,* that Anse, which is nearly
opposite Trevoux, belonged, at least in his time, to the Segusiavi ;

and the very faint resemblance between the names Amblagnieu and '
'

*
Ambarri will hardly prevail against the evidence which goes to
'

prove that the country between the Rhone and the Isere belonged
to the Allobroges.
The parts of the frontier of the Ambarri which can be defined with
least certainty are those which separated them from the Segusiavi
and from the Transrhodane Allobroges.
—
Ambiani. The Ambiani occupied the diocese of Amiens, which
nearly corresponds with the department of the Somme.^ See
Bellovaci.
—
Ambibarii. The Ambibarii, who are mentioned only by Caesar,^
appear in his list of the Aremorican or maritime states. The French
Commission place them in the diocese of Avranches, which subse-
"^

quently belonged, wholly or in part, to the Abrincatui,^ because, they

argue, after the territories of the other Aremorican states have been
fixed, this diocese alone remains unoccupied. Desjardins, who agrees
with the Commission, adds that the Abrincatui are not mentioned by
Caesar, while the Ambibarii are not mentioned by Ptolemy or Pliny.
But this is hardly a sufficient reason for assuming that the Ambibarii
were identical with the Abrincatui and the Commission regard them
;

as distinct though conterminous peoples. Moreover, it is impossible,

as I show elsewhere (see Diablintes), to determine the frontiers of
all the Aremorican states. The Ambibarii, in the opinion of the
Commission,^ may have also possessed a part of the diocese of
Coutances, as the southern frontier of the Venelli (q.v.) is very
uncertain.
Desjardins,^^ comparing Caesar's two lists of maritime states, notes

^ Les noms gaulois chez Cesar ef Hid ins de Bcllo Gallico, 1891, p. 38.
2 Bev. du Lyonnais, S'' ser i, 18GC, pp. 183-97 (esp. 189-90).
,

=*
B.G.,i, 10-2. " Geogr., ii, 8, § 11. ^ Diet. arch, de la Gatde, i, 49.

^ B. G., vii, 75, § 4. There are various readings but Ambibarii has the most
:

support. ' Diet. areh. de la Gaule, i, 49.

8 Ptolemy, Geogr., ed. C. Miillcr, ii, 8, § 8, pp. 214-5.
^ Eeu. arch., iiouv. ser., ix, 1864, pp. 400-7.
^^ Geogr. de la Gaule ram., ii, 487-91.
 —

AMBIBARII 367

that the Ambibarii, whom he wrongly calls Ambivariti,i are men-

tioned only in the latter, the Aulerci only in the former.^ As none
of the known Aulercan tribes were maritime, he infers that the
Ambibarii belonged to the Aulercan group.
—
Ambiliati. The Ambiliati are named once only in the Commen-
taries,^ among the tribes whom the Veneti engaged to join them in
their war against Caesar. The /? MSS. read Amhianos. Orosius,
mentioning in one and the same place * all the other tribes that are
enumerated in this passage by Caesar, wrote Amhivaritos. The
French Commission, considering the unsupported, testimony of
Orosius insufficient, and having regard to the fact that, except in
certain MSS. of one solitary passage of Caesar, the name Ambiliati
does not occur, prefer the reading Amhianos.^ They consider it
probable that the Ambiani joined the Venetian alliance, as their
near n<^ighbours, the Morini, did so and they call attention to the
;

fact that those very neighbours are mentioned by Caesar, in B. G.,

ii, 4, § 9, immediately after the Ambiani. Thomann ^ reads A^n-
bivaritos, but in addition to, not instead of Ambiliatos. He says that
'^
the Ambiliati were evidently the same as the Ambilatri of Pliny ;

and, he continues, as Pliny observes geographical order in his enumera-

tion of the Grallic peoples, the Ambiliati were on the left bank of the
lower Loire. But Pliny often departs from geographical order ;

for instance, he mentions the Veneti immediately after the Caleti,

the Abrincatui immediately after the Veneti, and the Osismi imme-
diately after the Abrincatui.^ Walckenaer,^ who accepts the reading
Ambiliatos, places them in the environs of Lamballe, in the diocese
of St. Brieuc, because, he says, there is no other place for them.
But, say the French Commission, ils seraient la en plein pays
'

curiosolite.' Nor is it certain that there is no other place for them ;

'
'

for we cannot determine the frontiers of all the Aremorican peoples.

M. Jullian ^^ thinks that Ambiliati may mean those who dwelt on both'

banks of the Ille ', in the department of the Ille-et-Vilaine. Ditten-

berger^^ suggests that Caesar may have written Ambibarios (q.v.).
I follow the a MSS. in accepting the reading Ambiliatos but the ;

problem of their whereabouts is insoluble.

Ambivareti. Ambibareti is the form found in the MSS. of the
name of a people mentioned in B. G., vii, 90, § 6, in whose country
Caesar quartered a legion after the fall of Alesia. Schneider ^^
identifies this people with the Ambluareti —
to adopt the reading of
the a MSS. —
who are mentioned in B. G., vii, 75, § 2, among the
clients of the Aedui. All the well-known editors, except Meusel,
likewise hold that in both passages Caesar is speaking of the

^ Geogr. de la Gaule rom., p. 490. ^ii, 34 ;

B. G., vii, 75.
» Ih., iii, 9, § 10. " vi, 8, § 8. arch, de la Gaule, i, 50.
5 j)i(.t^
^ Der franz. Atlas zii Cdsars gall. Kriege, Desjardins {Geogr.
1871, p. 17.
de la Gaule rom., ii, 487-8) also gives the Ambiliati, whom
he wrongly identifies
with the Ambibarii of B. G., vii, 75, § 4, the misnomer, Ambivariti.
' Nat. Hist., iv, 19 » lb., iv, 18
(33), § 108. (32), § 107.
^ Geogr. des Guides, i, 382. ^o
Hist, de la Gaule, ii, 490, n. 1.
^' Caesar, 15th ed., p. 315. 12
Caesar, ii, 641.
368 AMBIVARETI
same people but most of them adopt, with Gliick,! Amhivareti,
;

— the reading of p. Ni2)perdey,2 however, thinks it probable

tliat Amhluareti and Amhihareti (or, according to the reading
which he follows, Amhilareti) are corrupt forms of Ambarri, as
the Ambarri were intimately connected with the Aedui, and
are not mentioned in B. G., vii, 75. If, however, the Ambluareti
(so-called) were identical with the Ambarri, the Ambarri must
have been clients of the Aedui and we are not told that they
;

were. The French Commission^ 'also identify the Ambluareti with

the Ambarri, but think it doubtful whether the Ambivareti were
the Ambarri or the Ambibarii (q.v.). Long, who also includes the
Ambivareti among the clients of the Aedui, places them east of
the Loire and west of Bibracte. He remarks that in the winter of
52-51 B. c. Caesar left Bibracte, joined the legion which was quartered
among the Bituriges, and joined to it the 11th legion, which was
nearest. This legion, says Long, could not have been either of those
which were quartered in the valley of the Saone, for they were the
8th and 14th it could not have been the legion at Bibracte, for
:

Caesar Bibracte with only a cavalry escort

left it must, therefore, :

have been the legion which was quartered in the country of the
Ambivareti. That legion must have been on the west of Bibracte :

it must, being quartered among clients of the Aedui, have been

near their country and therefore it was probably on the east of
;

the Loire.*
Napoleon, perhaps following Dr. Noelas,^ places the Ambivareti, or,
as he calls them, the Ambluareti, on the west of the Loire, in the
neighbourhood of Ambierle, where a Roman camp has been dis-
covered, in the arrondissement of Roanne. Maissiat ^ suggests that
they may have possessed a subdivision of the territory of the Ambarri,
and that their name may survive in Vavre, Vavrette, and Varambou,
— places in the neighbourhood of the confluence of the Ain and the
Suran. But the truth is that, unless we identify them with the Am-
barri, which we have no right to do, it is useless to look for their
territory.
Ambivariti. — The geographical position of the Ambivariti cannot
be determined. No ancient writer mentions them except Caesar.' He
says that, when the envoys of the Usipetes and Tencteri first met him,
their cavalry were in the country of the Ambivariti, on the further
bank of the Meuse and the common opinion is that the further
;

1 See
p. 840.
Caesar, p. 106.
•^
Mommsen {Jahresb. d. philol. Vereins zu Berlin^ xx, 1894,
p. 21 1) also changes Amhivaretis —
the Ambluaretis of the MSS. into Amharri'^, —
partly for the reason given by Nipperdey, partly because the Ambivareti, who,
he says, dwelt near the Menapii [!], could not have been clients of the Aedui,
This curious argument is evidently due to a lapse of memory not the Am- :

bivareti but the Ambivariti dwelt near the Menapii.

^ Diet. arch, de la Gaule, i, 50. * Long's Caemr,
pp. 407-8.
5 Rev. dii Lyomiais, 3® ser., iii, 18()7,
pp. 261-79. Noelas's paper was not
published till after the appearance of the Hist, de Jules Cesar but he may have
;

communicated his views to the Emperor before.

" Jules Cesar en Gaule, i, 81. ' i>. 0., iv, 9, § 3.

I
 AMBIVARITT 369

bank means the left This opinion is based upon the

or western bank.
facts that the Usipetes and Tencteri had, before Caesar marched
against them, reached the territories of the Eburones and Condrusi,^
and that the greater part of the territory of the Eburones ^ and the
whole of the territory of the Condrusi (q.v.) were on the right bank
of the Meuse. Various writers place the Ambivariti in the environs
of Antwerp (An vers) others more to the south, near Givet, in a
;

district where there is a place called Amberive.^ Napoleon asserts,

without giving any reasons, that they were established on the left '

bank of the Meuse, west of Roermond, and south of the marshes of

Peel '.* General von Veith ^ finds them in the country round Weert,
about 12 miles west-north-west of Roermond. Desjardins, who
supposes that the name Ambivariti had a purely local significance,
suggests that it denoted the Eburones (!) situes des deux cotes de la '

Meuse, comme Amhitrehius signifie le pays situe des deux cotes de

la Trebie.' ^ But if so, why Ambivariti ? M. JuUian conjectures '^

that they dwelt on both banks of a river called ^Ivara either in Dutch
Brabant or in the low lands of the Dutch province of Limburg.^
The French Commission endeavour to prove that the common view
is erroneous, and that the Ambivariti were between the Meuse and the

Rhine.^ They argue that if the Roman army had been on the right
bank of the Meuse, Caesar would have been able to prevent the
cavalry of the Usipetes and Tencteri from taking refuge in the country
of the Sugambri.io This argument will not bear examination. The
rout of the Usipetes and Tencteri took place either near the confluence
of the Meuse and the Waal, which was at Gorkum or Fort St. Andries,
or near the confluence of the Rhine and the Moselle.^i In either
case there is nothing to show that the Ambivariti, if their territory
was anywhere near the left bank of the Meuse, were not as near as
Caesar to the Sugambri. If so, how was he to prevent the German
cavalry from reaching Sugambrian territory ? Even if he had had the
shorter distance to march, he might have failed to intercept them ;

for, as his cavalry were no match for the Germans,^^ he would have
been obliged to march against them with his infantry and the light ;

German horse would probably have been too quick for him. It is
even probable that, when the rout of the Usipetes and Tencteri
took place, the German cavalry had quitted the country of the
Ambivariti, and were actually on their way to rejoin their country-
men. Besides, as I have already remarked, Caesar says that when
he began to march against the Usipetes and Tencteri, they had
reached the territory of the Condrusi, which was on the right bank
1 B. G., iv, 6, § 4. 2 Ih., V, 24, § 4.
^ Diet. arch, de la Gaule, i, 51. * Hist, de Jules Cesar,
ii, 140, n. 1.
5 Jahresh. d. Geschichtswissenschaft, iii, 1880, ii, 2.
^ Geogr. de la Gaule rom., ii, 651, n. 2.

^ Hist, de la Gaule, ii, 475, note.

* The latter alternative corresponds with Napoleon's view.

^ Diet. arch, de la Gaule, i, 51-2. So also Walckenaer {Geogr. des Gaules, i,

508), who places them on the south of Liege, in the plain watered by the
Ambleve.
^° B. G., iv, 16, § 2.
11
See pp. 691-706. " ^ q^^ j^^ 12.
1093 , B b
370 AMBIVARITI
of the Meuse. Therefore the territory of theAmbivariti, which, from
the standpoint of the Usipetes and Tencteri was trans Mosam, must
have been on the left bank. General Creuly ^ assumes that when
Caesar said that the Usipetes and Tencteri had advanced as far as
the country of the Condrusi, he was speaking of a reconnoitring party
only ;but, in order to support the theory that the Ambivariti dwelt
between the Meuse and the Rhine, he is obliged to make the further
assumption that the main body remained throughout the campaign
on the western side of the Meuse. He believes that they had originally
crossed the western arm of the Rhine below the (alleged) confluence
of the Waal and the Meuse at Gorkum for otherwise, he argues,
;

when the Germans marched towards the country of the Treveri,

their course would have been parallel with the Rhine, and the words
uti ah Rheno discederent ^ in Caesar's account would be meaningless.
But Caesar says that they marched towards the country of the
Condrusi, not that of the Treveri and the map shows that, marching
;

between the Meuse and the Rhine towards the country of the
Condrusi, they would have moved away from the Rhine.'*^
The conclusion of the matter is this. The Ambivariti dwelt some-
where on the left bank of the Meuse and, as they are not mentioned
;

by any ancient writer, except Caesar, and do not appear in the

Notitia provinciarum, it is probable that they were dependants of
some more powerful people. But there is not the faintest e^ddence
for fixing their geographical position and to mark them on the
;

map would be simply to mislead.

—
Andes. The Andes occupied the whole of the diocese of Angers,
excepting the canton of Mauges, that is to say, the department of
the Maine-et-Loire and part of the Sarthe.*
—
Aremoricae (civitates). Walckenaer ^ and Desjardins ^ under-
stand by Aremorican simply 'maritime' while D'Anville*^ observes
' '
;

that Caesar appears to apply the term specially to the peoples between
the Seine and the Loire. The only passage in which Caesar mentions
by name any of the Aremorican peoples, as such, occurs in B, G.,
vii, 75, § 4, in the enumeration of the states which sent contingents
for the relief of Alesia. In this passage he says that 30,000 ^ men were
levied universis civitatihus quae Oceanum attingunt quaeque eorum
consuetudine Aremoricae appellantur, quo sunt in numero Coriosolites,
Redones, Amhiharii^ Caletes, Osismi, Veneti, Lexovii, Venelli.^ All
these peoples dwelt between the Seine and the Loire, except the

^ Rev. arch., nouv. ser., 1863, pp. 20-8.

viii, ^ B. G., iv, 6,
§ 3.
^ The theory of General Creulyand of the French Commission, on which their
location of the Ambivariti depends, that the Usipetes and Tencteri crossed tlie
Meuse (regarded as the western arm of the Rhine) below Gorkum, is refuted
on pp. 689-90. See also Heller in Philologus, xxii, 1865, p. 132.
* D'Anville, Notice de Vancienne Gaule, pp. 67-8 ; Walckenaer, i, 376.
^ Geogr. des Gaules, i, 437.

^ Geogr. de la Gaule rom., ii, 437, 461, 489.

^
Notice de Vancienne Gaule, p. 103.
^ Schneider {Caesar, ii, 587) argues that the MS. reading is wrong, and
that Caesar probably wrote X
(milia). See, however, Rhein. Mus., N. F., liv,
« Cf. B. G., ii, 34.
1899, p. 421.
 ^ —

AREMORICAE 371

Caletes, who were on

the right bank of the estuary of the Seine ;

and as the Morini, the Atrebates, and the Ambiani, maritime peoples
who dwelt on the east of the Seine, and the Pictones and the Santoni,
maritime peoples who dwelt on the south of the Loire, are mentioned
in the enumeration, it is clear that Caesar did not reckon them as
Aremorican.
According to Pliny,i the country between the Garonne and the
Pyrenees was called by the name Aremorica ', before it was called
'

'Aquitania '. But Pliny could hardly have had any authority for this
statement and, as he does not apply the term Aremorican to the
;
' '

tribes which Caesar called by that name, Long concludes that he

simply made a mis take.
—
Arverni. The territory of the Arverni, exclusive of the territories
of their clients, is represented by the diocese of Clermont and part of
that of St. Flour, which was severed from the parent diocese in the
fourteenth century. This territory comprises the departments of
the Cantal and Puy-de-D6me and parts of those of the Allier and
Loire-Superieure. The Gabali, one of the client peoples of the Arverni,
occupied the remaining part of the diocese of St. Flour.^
—
Atrebates. The Atrebates occupied the diocese of Arras, that
is to say the south-eastern part of the department of the Pas-de-
Calais and the adjacent part of the department of the Nord.^
Atuatuca (an(^the winter camps of 54-53 B.C.). 'The researches,'
—
says Napoleon, Wu.ich Major Cohausen kindly made for me, and those
'

of MM. Stoffel and Locqueyssie have enabled me to determine

approximately the winter- quarters.' ^ But Napoleon vouchsafes not
a word to tell us what these researches were, or where or how they
were carried on and, as neither Caesar nor any other ancient
;

writer gives us the least help towards determining any of the sites,
except those of Atuatuca and the camps of Cicero, Labienus, and
Trebonius, and only the least possible help towards determining
the first two, no researches could, except in regard to these four
places, be of any use.
I. The first question is how to interpret the famous passage,

harum tamen omnium legionum hiherna, praeter eam quam L. Roscio in

facatissimam et quietissimam "partem ducendam dederat, milibus
passuum C continebantur.^ This apparently means that no two camps
were more than 100 Eoman miles apart and, as the camp at Atua-
;

tuca, wherever Atuatuca may have been, was confessedly much more
than 100 miles from Samarobriva (Amiens), where the camp of
Trebonius was placed,*^ it follows that either there is an error in the
Nat. Hist.,iv, 17 (31), § 105.
1

^ Diet, of Greek and Roman Geogr., i, 218.

W. Smith's Cf. C. Jullian, Hist,
de la Gaule, ii, 455, n. 3, and Annates de la Faculte des lettres de Bordeaux,
1893, p. 99.
^ See d'Anville,
pp. 104-5 ; Walckenaer, i, 340 ; and Diet. arch, de la Gaule,
i, 83.
* Walckenaer,
i, 433 ; Diet. arch, de la Gaule, i, 89.
Hist, de Jules Cesar, ii, 200, n. 1.
^ ^ B.G.,v, 24,
§ 7.
' Schneider {Caesar, ii, 15G-8) holds that Caesar, after he had dispatched
messengers to order Crassus, Fabius, and Labienus to join him in marching to
B b 2
372 ATUATUCA
text or Caesar made a mistake. Schneider,^ Long,^ and Desjardins ^
interpret the Latin as I have done and Long observes that Caesar
;
'

could not know the distance accurately, and we can never trust the
numerals in the manuscripts '.'* But it is hard to believe that Caesar
could have made so gross a mistake and therefore it seems just
;

possible that he wrote CC, and that a C has dropped out. Still, it is
unlikely that he would have told us that the greatest distance
between any two camps was 200 miles, since he evidently means
that the camps were not so very far apart, after all. Napoleon,^
apparently determined to make the MS. reading square with the
facts, says these different winter- quarters were all included within
'

a circle of 100 miles' radius '. But this meaning cannot be got out of
the Latin. Von Goler ^ remarks that the camps formed two groups,
an eastern group and a western group, which is, in a sense, true and ;

he holds that between the most easterly camp of the western group
and the most westerly camp of the eastern group there was an interval
of 100 miles. But of all this Caesar says nothing. M. Jullian thinks "^

that the passage means that no camp was more than 100 miles
—
from the one nearest to it, an interpretation which seems to me
unwarrantable. Moreover, it is evident that according to Caesar's
informants, no camp was much more than 50 miles from the one

reheve Cicero, moved himself from Samarobriva to the camp of Trebonius

returned thence to Samarobriva with Trebonius' s legion "xnd, leaving it there
;

under the command of Crassus, marched with Crassus's o^ion to the rescue of
Cicero. Schneider completely misunderstands Caesar's narrative [B. G., v,
4(1-7), the meaning of which a clear-headed child could hardly fail to grasp.
Caesar says that when he received Cicero's request for help, in the afternoon,
he sent a messenger to order Crassus to come to join him ;that he sent a
messenger to order Fabius to lead his legion into the country of the Atrebates,
through which lie would himself have to march to Cicero's camp and that he;

wrote to order Labienus to march, if he could safely do so, into the country of the
Nervii. The rest of the army, that is to say, the legions of Plancus and Roscius,
were, he says, too far off to be able to help him. Next morning he was
informed that Crassus was approaching. Thereupon he placed Crassus in
command of Samarobriva assigned him a legion for its defence, because
;

he was leaving there the heavy baggage of the army, his hostages and state-
papers, and the winter's supply of corn ; marched himself with one legion to
join Fabius ; and advanced 20 Roman miles on the same day. The only
reasonable conclusion is that the camp of Trebonius was at or in the immediate
neighbourhood of Samarobriva, and that, as soon as Caesar knew that Crassus
had approached sufficiently near Samarobriva to secure it, he left the place
with Trebonius' s legion. For (1) if Schneider was right, Samarobriva, the
importance of which was so great that Caesar was obliged to detail an entire
legion for its protection even when he required all the troops that he could
get for the relief of Cicero, was absolutely defenceless before Caesar received
Cicero's dispatch and (2) if Trebonius had been, as Schneider maintains,
;

20 miles from Samarobriva, Caesar would not have made a useless journey from
Samarobriva to Trebonius' s camp and back, but would have summoned
Trebonius to join him nor (3) would he have said that all the legions, except
;

those of Crassus, Fabius and Labienus, were too far off to be able to help him.
,

He only said this because Trebonius' s legion was with him at Samarobriva.
^ Caesar, ii, 97. ^ Caesar, p. 243.
^ Geogr. de la Gaule rom., ii, 658, n. 1.
* Decline of the Roman Republic, iv, 216.
5 Hist, de Jules Cesar, ii, 202. « Gall. Krieg, 1880, pp. 168-71.
^ Hist, de la Gaule, iii, 371, n, 8.
 .

ATUATUCA 373

nearest to it.^ However, the difficulty presented by the MSS. is

unimportant for, whatever Caesar wrote, it is beyond question that
;

Atuatuca was much more than 100 Roman miles from Amiens.
II. Let us see first of all what Caesar has to say about Atuatuca.
(1) He says that the camp of Sabinus and Cotta, which, as we learn
from B, G., vi, 32, §§ 3-4, was at Atuatuca, was situated in the
country of the Eburones, the greater part of whose territory is
'

between the Meuse and the Rhine (quorum pars maxima est inter
'

Mosam ac Rhenum) ^ (2) he says (according to the usual interpreta-

;

tion of the Latin) that Atuatuca was nearly in the centre of the
'

country of the Eburones (fere in mediis Ehuronum finibus) ^

'
;

(3) he remarks that its position w^as naturally strong, and gives
particulars, for which I may refer to pages 107, 110, and 126-7, of
my own narrative, regarding the country in its immediate vicinity ;

(4) he apparently implies that, when Sabinus quitted it, he marched

in the direction of Cicero's camp, which was in the country of the
Nervii for he makes Ambiorix say that the camp of Cicero was
;

nearer to Atuatuca than the camp of Labienus * and he makes ;

Sabinus say that the best plan would be to march to the nearer
camp.^ Furthermore, as it is not likely that Ambiorix knew which
of the two camps Sabinus intended to join, we may perhaps infer
with General Creuly that in either case the first part of his route
would have led through the valley which Ambiorix occupied.^
1. D'Anville, Walckenaer, Mommsen, Long, Napoleon, the French

Commission, Desjardins, Kiepert, and most of the modern editors

identify Atuatuca with Tongres, in the valley of the Geer, which
flows into the Meuse on its left or western bank. Desjardins affirms
that the identification is proved by the itineraries. The Atuaca of
the Table —
the 'Atovoltovkov of Ptolemy ^ was indisputably upon
'^
—
the site of Tongres, and was the chief town of the Tungri. The Tungri
are mentioned for the first time by Pliny,^ and afterwards by Tacitus ;i^
and it is inferred that they succeeded to the territory of the Eburones,
who appear to have been virtually exterminated, as a people, by
Caesar. General Creuly remarks that it is unlikely that in the
country of the Eburones there were two places of the same name ;

^ B. 0., V, 27, § 9.
2 See Schneider's Caesar, ii, 96, note.
Ih., V, 24, § 4.
3 5. G^.,vi,32,§4. * /6.,v, 27, §9.

^ Ih., 29,
§ 6 ; 30, § 3, M. A. de Vlaminck, on the contrary {Messager des
sciences hist, de Belgique, 1887, p. 396), argues that Sabinus must have marched
in the direction of Labienus' s camp, because pas un soul des legionnaires
'

echappes au massacre n'arriva jusqu'a Ciceron'. I do not think that this

argument will prevail against the one which I have drawn from Caesar's
narrative nor do I think that it is intrinsically strong. The few legionaries
;

who escaped the massacre may have found that retreat in the direction of
Cicero's camp would be cut off by the army of Ambiorix ; and, as they were not
an organized body, if Atuatuca was between the Meuse and the Rhine, they
may have shrunk from the prospect of having to cross the Meuse.
^ Eev. arch., nouv. ser., viii,
1863, pp. 143-4.
''
La Table de Peutinger, ed. Desjardins, p. 12, col. 1
* Geogr., ii, 9, § 5.
' Nat. Hist., iv, 17 (31), § 106. '» lltst., iv, o5, 79, &c.
374 ATUATUCA
and accordingly concludes that the Atuaca of the Table was
lie
identical with the Atuatuca of Caesar.^ But the general forgets that
he has himself argued in another place that there were probably
S3veral towns called Uxellodunum in the country of the Cadurci.^
T. Fuss ^ remarks that Atuatuca must have been, like Tongres, in
a plain and on a low-lying site, since the cavalry of the Sugambri rode
up at a rapid pace from the neighbouring woods to attack the
decuman gate of the camp.^ This fact may prove that the decuman
gate was very little, if at all higher than the level of the woods :

but it does not prove that the camp did not stand on ground which
shelved down in other directions and, if it did, it would not
;

prove that Atuatuca stood upon the site of Tongres it would :

—
merely remove one of the many objections which have been urged
against that site.
First, whenCaesar, after mentioning that the camp of Sabinus and
Cotta was in the country of the Eburones, says, in the same breath,
that the greater part of that country lay between the Meuse and the
Rhine, one is naturally inclined to infer that the camp was somewhere
between those two rivers. Otherwise, Caesar's remark would be
pointless and out of place. But Tongres is on the left bank of the
Meuse. And the inference from Caesar's statement is no less obvious
even if he meant, as has been argued, not that the greater part of the
territory was situated, but only that the greater part of the population
dwelt between the two rivers. Fuss struggles to get over this difficulty
by suggesting that Caesar, not having previously mentioned the
whereabouts of the Eburones, took this opportunity of doing so.
If, he pleads, Caesar had intended to indicate the site of the camp of
Sabinus and Cotta, he would have done so with more j^recision.^ But
what if Caesar did take this opportunity of mentioning the where-
' '

abouts of the Eburones ? The point is that, in the same breath in

which he mentioned that the greater part of the territory of the
Eburones was between the Meuse and the Rhine, he said that the
camp of Sabinus was in their territory. And, as Fuss ought to have
known, Caesar does not, except very rarely, describe sites with
precision. In a later passage he does intend to indicate the site of
'

the camp of Sabinus ', and says vaguely that it was fere in mediis
Ehuronum finibus.
Secondly, it has been objected that it is useless to try to reconcile
the actual position of Tongres with Caesar's statement that Atuatuca
was fere in 7neiiis Eburonum finibus, if that statement means nearly '

in the centre of the country of the Eburones.' Attempts have, how-

ever, been made to explain away the difficulty. Tongres est situe,'
'

says Napoleon,^ in mediis finibus Eburonum,, ce qui signifie en plein

'

^ Rev. arch., nouv. ser., viii, 1863, p. 138.

2 Examen hist, et topogr. des lieux proposes pour representer Uxelloduuum,
1860, pp. 15-6, 27.
^ Bull, de la Soc. sc. et litt. da Linihoitnj, ii, 1854, p. 175.
' 5. 6«.,vi, 37, §1.
5 Bull, de la Soc. sc. et litt. du Lunhourg, ii, 1854, p. 169.
" Hist, de J ides Cesar, ii, 201.
 ATUATUCA 375

pays des Eburons ct noii aii centre du pays.' I cannot see how this
meaning is to be got out of the Latin if Caesar had meant what
:

Napoleon says, he would simply have written in finihus Ehuronum ;

for the word/ere would have had no point. Mr. C. E. Moberly ^ offers
a curious explanation. A military man,' he says, would take his
' '

idea of a central position more from the roads than from the general
country which he occupies. Thus Tongres would be a medius locus, as
lying on the great arterial road of the Meuse, by which alone the
baggage of the army could be transported.' It is not proved that
there was no other road through the territory of the Eburones by
which baggage could be transported but let us assume the truth of
;

Mr. Moberly 's assertion. On his theory, if an invading army, having

effected a landing at Brighton, were to march direct on London,
Redhill would be in mediis Anglorum finihus. On his theory, too,
the word /ere in Caesar's statement would be superfluous. Lastly it
has been argued that the country of the Eburones extended westward
as far as the Scheldt, and therefore that Tongres is in the centre of
that country. D'apres le recit meme des Commentaires,' says a
'

member of the French Commission,^ les Eburons s'etendaient a '

gauche jusqu'a I'Escaut.' The passage on which the argument is

based is in B. G., vi^, 33, § 3. Describing the measures which he took
in 53 B. c, after returning from Germany, to apprehend Ambiorix,
Caesar says that he determined to march in the direction of the
Scheldt and the most distant parts of the Ardennes, whither he
heard that Ambiorix had gone (ifse cum reliquis III [legionibus]
ad flumen Scaldim, quod influit ifi Mosam, extremasque Arduennae
partes ire constituit, quo profectum Ambiorigem audiehat). This
. . .

passage proves nothing as to the extent of the Eburonian country,

for Caesar docs not say that Ambiorix was in that country he may, :

for aught we know, have escaped beyond his own frontier, as Caesar
expressly says that many of his subjects did.^ General Creuly^
says that 210 kilometres, or 130 miles, the distance from Tongres to
the nearest point of the Scheldt and back, is just what Caesar would
have accomplished in seven ordinary marches. But, to say nothing
of the fact that 30 kilometres is considered by most authorities too
much for an ordinary march,^ Caesar does not say that he went all
the way to the Scheldt and when he told Cicero that he would return
;

from his expedition in seven days, he may not have known how far
off the Scheldt was. Moreover, even if he did march as far as the
Scheldt, even if the Scheldt bounded the territory of the Eburones,
Tongres was not in the centre of that territory for it was certainly ;

close to their southern boundary.

One can conceive, however, that Caesar may have meant to
convey a meaning different from that which has commonly been
ascribed to his words. It has been suggested that finihus should be
1 In his edition of the B. G., 1880,p. 301.
'^
Diet. arch, de la Gaule, i, 11.
^ Multi ex suis tinibus cgrcssi, so suaiiuc omnia alicnissimis crediderunt.
B. a., vi, 31, § 4.
^ Bev. arch., nouv. scr., vii, 1803, p. 380. ^ Hoc p. 035.
376 ATUATUCA
translated by '
There are at least three passages ^ in the
frontier '.

Gallic War where fines obviously means frontier and one of them
' '
;

certainly suggests that Atuatuca was on or very near the southern

frontier of the Eburones. In the passage to which I refer 2 Caesar
says that Ambiorix and Catuvolcus met Sabinus and Cotta near the
frontier of their kingdom, and brought their quota of corn to the
Roman camp initium repentini tumultus ac defectionis ortum est
:

ah Ambiorige et Catuvolco ; qui, cum ad fines regni sui Sabino

Cottaeque praesto fuissent frumentumque in hiberna com-
portavissent, kc. But I do not believe that fines, in Caesar, ever
means frontier, except where that meaning is unmistakably indicated
by the context and if frontier is the meaning here, it would
;
' '

seem that the words hoc fere est in mediis Eburonum finibus are
meaningless for a frontier, considered as a whole, has no centre.
;

Pondering over this, it occurred to me some time ago that Caesar

may have written merldianis, not mediis and von Goler ^ makes ;

the same suggestion. But Meusel in a MS. note urges that

Caesar would have written meridiano (hne), not merldianis (finibus);
and the longer I study the Commentaries the more convinced
I become that in nine cases out of ten the passages which it
is sought to amend require no emendation. Caesar tells us that
Ambiorix was king of one half of the country of the Eburones,
and Catuvolcus of the other."* I am inclined to believe that what
he meant (though he expressed his meaning rather loosely) was, not
that Atuatuca was nearly in the centre of the whole Eburonian
territory, but that it was nearly in the middle of that territory,
in the sense that it was near the common frontier of the two kingdoms
of which the whole territory was composed. Still, I find it difficult
to suppose that he would have used the words in mediis Eburo7ium
finibus to indicate a site which lay 15 kilometres west of the Meuse
and yet belonged to a people the greater part of whose territory is
'

between the Meuse and the Rhine '.

There are several other passages in Caesar which point to the
conclusion that Atuatuca was between the Meuse and the Rhine. He
tells us that Ambiorix, in his colloquy with Gains Arpineius and
Quintus Junius, stated that a body of Germans, who were coming to
the assistance of the Gallic insurgents, had crossed the Rhine, and
would be at Atuatuca in a couple of days.^ Sabinus, in the council of
war which was called to discuss the statements of Ambiorix, remarked
that the Rhine was close by (subesse Rhenum),^ a phrase which he
would hardly have used if the Meuse had intervened between the
Rhine and Atuatuca. When the Sugambri invaded the country of
the Eburones, we are told that they crossed the Rhine but we are ;
'^

not told that, in order to reach Atuatuca, they crossed the Meuse.
When they left Atuatuca, we are told that they retreated across the
Rhine (trans Rhenum se receperunt) ^ and this phrase would certainly
;

1 B. G., V, 26, § 2 54, § 2

; ; vi, 44, § 3. ^ lb., v, 26, § 2.
^ Gall. Krieg, 1880, p. 175. '
B. G., v, 24, § 4 ; vi, 31, § 5.
' /6., v,27, §8. «
/6.,29,§3.
«
' /6., vi, 35, § 6. /6.,41,§1.
 ATUATUCA 377

be misleading if they had first had to cross so important a river as the

Meuse.
Heller, who took the road leading
believes, as I do, that Sabinus
to Cicero's camp, says that he would not have attempted to cross
the Meuse, as he must have intended to do, if Atuatuca was on the
right bank, in order to reach his destination.^ It is amusing to find
that Heller himself points out that the Sugambri, in order to reach
Tongres, which he identifies with Atuatuca, could easily have crossed
the Meuse at Lixhe or Maestricht.^ If the Sugambri, why not
Sabinus and his cohorts ? And why should Sabinus have hesitated
to cross the Meuse when he did not fear opposition ? Besides, on
Heller's theory, the handful of fugitives who escaped to Labienus's
camp must have crossed the Meuse.
I do not forget that M. Jullian ^ has affirmed the identity of
Atuatuca with Tongres and I rarely differ from him without
;

searchings of heart. But M. Jullian is also convinced that Labienus's

camp was at Mouzon, some 10 miles south-east of Sedan. Now
Mouzon is about 85 English, or 93 Koman, miles in a straight line
from Tongres and Caesar says that Labienus's camp was rather
;

more than 50 miles from Atuatuca.* It can hardly be supposed that

the number L is due to a copyist's mistake for Caesar also puts ;

into Sabinus's mouth the statement that the brigade quartered at

Atuatuca would be able to reach the camp of Cicero or of Labienus
in a couple of days.^ He could not have reached Mouzon in less than
four.
So much for the geographical position of Tongres let us consider its :

topography. Napoleon, after the publication of his History, sent

Major de Locqueyssie to study the environs of Tongres and this ;

officer decided against identifying it with Atuatuca.^ Having explored

the position myself, I fail to see how it can be satisfactorily reconciled
with Caesar's description. Several years ago I showed that the only
part of the adjoining valley which could be identified with the
magna convallis was close to Koninxheim and M. Jullian has
;
'^

adopted this view. He holds that the combat took place between
Koninxheim and Eusson and he adds that excavations carried
;

out in Tongres and its environs by M. F. Huybrigts add fresh proofs

to those furnished by the texts and by study of the ground, for
hundreds of Roman coins earlier than 54 B.C. have been found
during the last twenty years between Koninxheim and Lowaige.^
But I should like to know how many later Roman coins have been
found in the same valley and I would suggest that since coins
;

1 Philologus, xxii, 1865, pp. 139-40. 2

/^^^ p^ 140.
3 Hist, de la Gaule, iii, 376, n. 6. « B. G., v, 27, § 9.
' lb., 30, § 3. See p. 383.
^ Bull, de VAmd. Roy. de Belgique, xxxvii, 1874, p. 118.
' Bev. des etudes anc, x, 1908, p. 362. Cf. M. Jullian' s Hist, de la Gaule, iii,
379, n. 4.
" F. Huybrigts, La Torujrie et ses antiquiies, 1C07, pp. 9-10. M. Huybrigts
himself says (p. 9) that the coins have been found cntre Tongres et Koninxheim,
'

ettoutju-esde Tongres', and again (p. 10) aux abords de Tongres vers I'ouest
'

et tou jours dans la meme vallee'.

-<; of MEDMfy'^
®
^/^ ET. ^'^lc^^'-'-
378 ATUATUCA
minted in 54 B.C. or earlier remained in circulation for many years,
these coins (assuming that they are described correctly) may have
been lost by Eornan troops quartered at Atuaca after Sabinus's
death. Moreover, we are merely assured that hundreds of coins ' '

have been found they are not catalogued in M. Huybrigts's tract

:

and no details of their discovery are given. Indeed M. Huybrigts's

account seemed to me so unsatisfactory that I asked a well-known
numismatist, attached to the Cabinet des Medailles in the Biblio-
theque royale of Brussels, to enlighten me. It would, I fear, be
indiscreet to quote his reply in full but while he emphasizes the
;

'
perfect good faith of M. Huybrigts, he assures me that no conclusion
'

regarding the site of Atuatuca can be based upon his statements.

For, he says, on n'a pas fait de trouvaille importante de monnaies
'

con.sulaires entre Tongres et Koninxheim vous trouverez dans . . .

le Bulletin de la Societe scientifique et littcraire du Limhourg, t. xvii,

1897, p. 36, une liste de huit deniers consulaires dans la collection
de M. Huybrigts. II y accuse egalement " 60 monnaies de bronze
anterieures a I'annee 700 de Rome " mais ce sont pour la plupart
;

des pieces frustes de I'epoque imperiale.' Again, I would ask any

man who has seen Tongres whether it could have been described by
Caesar as a naturally strong position. It is approached by an ascent,
which is very gentle, on the south and south-east only and it was ;

of course on this side that the first attack of the Sugambri would
have been directed. How, then, could Caesar have said that the '

strength of the position as well as the entrenchments forbade any

attempt to enter elsewhere (reliquos aditus locus ipse per se muni-
'

tioque defendit^i)^ M. Jullian, indeed, argues that Tongres was

partly protected by marshes but where could they have been,
;

except between the south-east and the south-west, where they may
have been formed by the Geer ? And even they would not have
extended up to the hypothetical site of the camp. Besides, Tongres
is at least 15 kilometres, or more than 9 miles, in a straight line from
the nearest point of the Meuse and no one will deny that the
;

Sugambri must have been on the eastern bank when they were
advised to attack Atuatuca. The Meuse is a very broad and deep
'^
river; and it is hardly credible that the obscure and insignificant '

Eburones should have bridged it anywhere near Liege. If not, the

Sugambri could only have crossed it, as they crossed the Rhine,
in boats or on rafts are we to believe that they collected boats,
:

crossed, and marched to Tongres in three hours ? ^ And if so, what

did they do with the numerous cattle which they had looted ? Is it
not equally difficult to believe that they ferried them across the
Meuse, and that if they had left them on the right bank, they would
have found them on their return ?
General Creuly observes that the Atuaca which stood upon the
' '

site of Tongres is described on the famous mile-stone of that town as

a castellmn, and adds that it is worthy of remark that Caesar
'

designates Aduatuca as a castellmn, while everywhere else he calls

' B. G„ vi, 37, § 5. - lb., y, 28, § 1. ' lb., vi, 3o, § «.

 ATUATUCA 379

the strongholds of the Gauls oppida '.^ But Caesar says that the
Atuatuci and the Nantuates had castella.'^
The conclusion of the matter is this. For the assumed identity of
Atuatuca with Tongres there are only two strong arguments, the —
fact that a Gallo-Roman town, which stood upon the site of Tongres,
was called Atuaca, and the fact that this town was situated at the
junction of great roads.^ A. Wauters* warns us that, unless we
accept Tongres, we must remain in complete uncertainty as to the
site. No doubt But what then ? It is unpleasant to be forced
!

to confess that, after prolonged research, one has failed to identify

the scene of one of the most famous episodes in the Gallic war but ;

even Wauters would probably admit that it is better to be un-

certain than to deceive oneself and one's readers.^
Nevertheless, of the numerous other sites that have been proposed,
there are two or three which have found advocates of sufficient repute
to make it worth while to examine their claims. Supposing that
Atuatuca was east of the Meuse, it could not have been far from the
right bank if it was not more than 50 Roman miles from Cicero's
camp in the country of the Nervii.^ 2. The plateau of Embourg or
Embour, near Liege and between the Ourthe and the Vesdre, answers,
it is said, to all the requirements of Caesar's narrative and von ;

Cohausen, Thomann, Grandgagnage, and von Kampen agree in

adopting it. A place called La Hasette on the plateau is, in their
opinion, the site of the castellum? On three points it is protected
by rocks, thus answering to Caesar's description ^ of the natural
strength of the position. On the north it touches a plain and it ;

was here, Grandgagnage believes, that the Sugambri attacked it.

Woods still exist in the neighbourhood a Roman road leads to
:

La Hasette on the north and south there are two hills, either of
:

which might have been the collis to which Caesar alludes in his
description of the attack on Cicero's camp;^ and there is more than
one majna convallis in the neighbourhood.

^ Rev. arch., nouv. ser., iii, 1861,

pp. 412-3. I am rather inclined io suspect
that the words id casteUi nomen est {B. G., vi, 32, § 4) were interpolated by-
some one who knew that the Atuaca of the Table was a castellum, and that the
following hoc was originally haec. I doubt whether Caesar would have made
such an irrelevant remark.
^ B. G., ii,
29, § 2 iii, 1, § 4.
; Von Goler {Gall Krieg, 1880, p. 3) says that
a castellum was a stronghold intended for military purposes only, whereas an
oppidtim was permanently inhabited ; and I believe that he is right.
^ See C. Jullian, Hist, de la Gaule, iii, 376, n. 6. But M. Jullian himself
(lb., p. 406) says that the camp was '
sottement expose '.
* Bull, de VAcad. Roy. de Bruxelles, xv,
1863, p. 286.
^ Professor L. Halkin {Bull. bibl. du Musee beige, ix, 1905, p. 446) and
Dr. V, Tourneur {ib., xii, 1908, p. 72, n. 1 ) hold that I have disproved the identity
of Atuatuca with Tongres but I do not claim to have done so much ; for
:

I cannot offer an alternative which is wholly satisfactory.

« B. G., V,
27, § 9.
' See Jahrb. d. Vereius von Alterthunisfreunden im Rhcinlande, xliii, 1867,
pp. 26-41, and A. von Kampen' s Qaiudecirn ad Cacsarls de h. G. comm. tabulae,
viii.
« B. G., vi, 37, § 5. 9 lb., 36, § 2.
380 ATUATUCA
These ar^unicnts are not conclusive. The suggestion that the
name Embourg is derived from Eburones is hardly convincing.
' '

Moreover, there is one objection to Embourg which is absolutely

fatal. The magna convallis is assumed to have been a part of the
valley of the Mosbeux, a rivulet which flows into the Vesdre. This
valley is not two but five miles from Embourg, Von Kampen naively
admits that the words in silvis oportuno atque occulto loco a milibus
'

passuum circiter duobus can only apply if the forests extended . . .

to within that distance of the camp, which is not unlikely in itself,

if only we assume further, that the hostile army lying in ambush

there, first allowed the Romans to pass them, in order to pursue and
fall upon them with more chance of success in the magna convallis
'
some four miles farther '.^ In other words, if only we assume '

what flatly contradicts Caesar's narrative !

3. Von Goler ^ finds the camp of Labienus in the neighbourhood of

Chiny, on the river Semoy. This place, he says, is 100 miles from the
most easterly of the western group of camps, which he places at
Soissons, and was connected with Reims by a Roman road. The
camp of Cicero he places at Namur, which he regards as the best
place that Caesar could have selected, both because of its strength
and because it would have afforded easy communication, on the one
side by the valley of the Meuse with the (assumed) camp of Labienus
and on the other by the valley of the Sambre with Caesar's camp at
Samarobriva, which he identifies not with Amiens, but with Bray ^ I

Fifty miles eastward of Namur, he finds in Limbourg a site which he

regards as conforming to Caesar's description of Atuatuca.'*
This is a fair specimen of von G5ler's laborious and ingenious, but
futile method. Labienus's camp may have been near Chiny but if ;

von Goler was really convinced that he had found the site, he must
have been sadly deficient in the sense of humour. Caesar's camp was
not at Bray, but at Amiens.^ There is no evidence that the camp of
Plancus was at Soissons and if it was, there is no evidence that it was
;

100 miles from the camp of Labienus. There is no evidence, and it is

most unlikely that the camp of Cicero was at Namur.
Since von Goler wrote, Limbourg has found new advocates, General
von Veith ^ and M. E. Harroy,*^ the latter of whom has said everything
that can be said in its favour. Limbourg, which stands upon a high
rock, about 270 feet above the river Vesdre, is some 15 miles, as the
crow flies, east of Liege. The magna convallis, in w^hich the disaster
occurred, is identified by M. Harroy w^ith the valley of Dolhain, the
nearest entrance to which is about one mile, measured along the
winding road which leads to it, from Limbourg itself. This valley
is about 1,640 yards ^ long, and its average breadth is from a quarter

of a mile to 550 yards.^ It forms a plain, hemmed in on the north

^ Quindecim ad Caesaris de b. G. comm. tabulae, viii.

2 Gall. Kriecj, 1880, p. 173. =^
lb. ^ lb., pp. 172-5.
^ See my note on Samarobriva.
^ Fick' s 3Ionatsschriftf. d. Gesch. Westdeulschlands, 1878, pp. 419-27.
^ Les Eburons a Limbourg, J 889.
« 1,500 metres. » 400 to 500 metres.
 ATUATUCA 381

by an escarpment, which rises fully 300 feet above it, and on the south
by wooded heights of lower elevation and more gentle slope. At its
western, as at its eastern entrance, the valley narrows into a gorge.
The point at which the Roman advanced guard was checked would
have been opposite the gorge of Bilstain, which opens into the valley
from the north and the doomed cohorts would have made their last
;

—
stand on the Pave du Diable, the widest part of the valley. The
field to which Cicero's foragers were sent would be on the north-east
of Limbourg, between Baelen and Honthem, and separated from it,
as Caesar's narrative requires,^ by a hill and the Sugambri would ;

have approached the rocky height of Limbourg from the south-east

by the slope which leads past Eupen, Membach, and Goe.^ To the
possible objection that the Vesdre flows through the valley of
Dolhain, and that Caesar, in his description of the disaster, says
nothing about a river, M. Harroy replies that within his own recollec-
tion the Vesdre, in that part of its course, was a rivulet a mere —
—
thread of water which one could cross without wetting one's feet ^ ;

and that, as Caesar tells us, the summer had been characterized by
extraordinary drought. M. Harroy also insists that Limbourg and
"*

its environs correspond, in every detail, with Caesar's narrative.

I can find no fault with the field to which he sends Cicero's foragers
nor with the valley of Dolhain, except that it seems to me rather too
short to have held a column which could hardly have numbered less
than 5,000 fighting men, besides cavalry and servants and a long
—
baggage train, a column the formation of which Caesar condemns
because of its excessive length.^ But be this as it may, I have two
objections to make to M. Harroy's theory. Caesar says that the
Eburones lay in wait for the Romans, in two divisions, at a distance
of about two Roman miles from the Roman camp {conlocatis insidiis
in silvis oportuno atque occulto loco a milihus passuum circiter duobus).^
This is not perfectly clear but, as Long justly remarks, the simplest
;
'

explanation is this, that a points out the termination of 2,000 paces

from the camp as the place where the ambuscade began. Thus it was
from or after the space of 2,000 paces that the ambuscade was seen
and felt.' Now M. Harroy is only able to bring his theory into a
'^

show of harmony with Caesar's statement by reckoning the distance

of two miles from the farthest extremity of the valley of Dolhain.^
I press this objection because Caesar saw Atuatuca himself,^ and we
cannot doubt that he visited the scene of a disaster which he never
allowed himself to forget. Secondly, although Atuatuca was un-
doubtedly a place of natural strength, I find it difficult to believe
that it stood upon a high and almost impregnable rock like Limbourg.i^
Caesar would hardly have failed to inform us of so remarkable

'
B. G., vi, 36, § 2.
^
Les Eburons a Limbourg, pp. 17, 64, 71, 74.
''

lb., p. 80. * B. 0., V, 24, § 1.

5 7^,.^ 31^ § g
«
lb., 32, § 1. ' Caesar, p. 250.
^
Les Eburons a Limbourg, p. 65.
«
B. G., vi, 32, §§ 2-5 ; 41, § 4.
'"
See Carte to2}ogr. de la Belgique, Feuille 43, pi. 5.
382 ATUATUCA
a circumstance and it is difficult to understand how the Sugambri,
;

with their inadequate force, could have had the temerity to attack
so formidable a position. I fear, therefore, that M. Harroy's interest-
ing brochure must be pronounced inconclusive .^
4. It is hardly necessary to mention any of the other conjectures.
Colonel P. Henrard ^ identifies Atuatuca with Vieux-Virton, which
is about 80 miles due south of Liege M. de Vlaminck with
!

Aix-la-Chapelle ^ M. Caumartin * with Hontem, near the eastern

:

bank of the Meuse and the ford of Na vague, and about 15 miles
north-east of Liege : with Rheinbach, about 11 miles
B. Schottler ^

south-west of Bonn, which is very much too far from the country
of the Nervii to correspond with Caesar's narrative. Besides Atuatuca
and Aduatuca, in B. G., vi, 35, § 8, there are, in inferior MSS,, various
—
readings, ad Vatucam or, according to Frigell, at Vatucarn, which is
—
absolutely meaningless and ad Varucam ^ and accordingly, from ;

the resemblance of the names, the camp has also been placed by
some at Wittem,'' about 8 miles west-north-west of Aix-la-Chapelle,
and by others at Waroux,^ on the west of the Meuse, about 3 miles
from Liege. But it is needless to say that Caesar, being an educated
man, would not have written ad Vatucam or ad Varucam ^ (venire) ;

and besides it needs no critical acumen to see that ad Vatucam and

at Vatucam are simply Aduatucam and Atuatucam, miswritten by
a dunce. Finally, Alfred Holder ^^ places Atuatuca at Vetschau,!^
about 3 miles north-west of Aix-la-Chapelle but I have not been
;

able to discover on what grounds he bases his opinion.^^

The truth is that the data are insufficient. But, although the
arguments against identifying Atuatuca with Tongres are many and
strong, it must be admitted that, after the most searching exploration,

^ M. Jullian is right in placing Labienus's camp at Mouzon (see p. 384,

If
infra),Limbourg can hardly be identified with Atuatuca for it is about 80 miles
;

from Mouzon. But this objection applies equally to Tongres. M. Jullian

{Hist, de la Gaule, iii, 376, n. 6) objects, further, that the few Roman soldiers
who escaped would not have been able to make their way to the camp of
Labienus a travers les montagnes et les bois'. But does not Caesar say that
'

they went by the woodland tracks {incertis itineribus per silvas [B. G., v, 37,
'

2 Mem. couronnes . puhlies par V Acad. Roy. de Belgique, xxxiii, 1882,

. .

pp. 10, 21, 37-40, &c. Cf. pp. 54, 57.

^ Messager des sciences hist, de Belgique, 1882,
pp. 413-4.
* lb., 1883, pp. 238-9.
s
Tiber d. Lage d. geschichtl. Orte Aduatuca, 1889.
^ Schneider's Caesar, ii, 299.

^ J. B. Renard's Hist. pol. et mil. de la Belgique, p. 442 ; Mem. de V Acad.

Roy. ..de Bruxelles, ii, 1822, pp. 261-2.
.
« lb.,
pp. 258-61.
^ A reviewer in the Athenaeum (Jan. 13, 1900) takes exception to this remark
;

but I would remind him that ad Vatucam (venire) would only be admissible if it
meant (to come) to the neighbourhood of Vatuca '.
'

^" Alt-celtischer Sprachschatz, i, 47-8.

11
See Sheets 2965-6 of the German Government Map (1 25,000). :

12
Essneux, Fauquemont, Fouron le Comte, Gressenich, Herve, Huy, Jule-
mont, Juliers, Liege, Maestricht, Mortroux, Rolduc, Valkenberg, Verviers,
Wandre, and the plateau de Falise', at the confluence of the Ourthe and the
'

rivulet of Laval, have also been proposed. See Jalirb. d. Vereins von Alter-
thum^freunden im Rheinlande, xliii, 1867, p. 18.
 ATUATUCA 383

nobody has succeeded in finding another site which is not open to

grave objection. Moreover, Atuatuca was in the kingdom of Am-
biorix, which formed the western part of the territory of the
Eburones and it cannot be denied that the Meuse might have
;

formed a natural boundary between his dominions and those of

Catuvolcus.
III. Napoleon ^ places Cicero's camp at Charleroi, which, he says,
'
is situated near the Roman road from Amiens to Tongres, and, as

the Latin text requires, at 50 miles from the latter town. On the
high part of Charleroi, where the camp was no doubt established,
we command the valley of the Sambre, and we can see, in the distance
towards the west, the country through which Caesar arrived. ^ More-
over, the valley of the Haine and Mont Sainte-Aldegonde, above the
village of Carnieres, agree perfectly with the details of the combat in
which the Gauls were defeated.' ^
M. Jullian^ points to Binche, about a dozen miles west of
Charleroi, which, assuming that Atuatuca was at Tongres, is open
to no objection, —
save what M. Jullian himself calls the grosse '

objection that it is about 66 Roman miles from Tongres, and

'

almost 80 from Mouzon, where he places the camp of Labienus.

My friend strives to set our minds at rest by arguing that the distances
which Caesar gives were given on the authority of Gauls, and there-
fore that not Roman miles were meant, but Gallic leagues. Possibly.
But Caesar was not credulous and he states the distance from the
;

camp of Cicero to that of Labienus on his own authority.^

Desroches ^ places Cicero's camp at Assche Dewez and others at
;

Mons von Goler ^ and von Kampen at Namur. But it would be as

;
'^

tedious as it is needless to prolong the list of guesses. Until the

position of Atuatuca has been fixed it will be absolutely hopeless
to search for Cicero's camp. For we know nothing about its position,
except that, if Caesar's, or rather Ambiorix's, figures are accurate,
it was about 50 Roman miles from Atuatuca, was within a short
march and on the east of a rivulet which flowed through a broad
valley, and was probably close to the road that led from Amiens to
Atuatuca. Regarding the camp of Labienus, which the omniscient
Napoleon places at Lavacherie on the Ourthe, we are somewhat
better informed. It was rather more than 50 Roman miles from
Atuatuca ^ it was near the common frontier of the Treveri and the
;

^ Hist, de Jules Cesar, ii, 214, n. 2. 2 ^

g^ y, 48, § 10.
3 lb., 48-52. Hist, de la Gaule, iii, 383, n. 2.
*

^ See
p. 377. '
De plus,' says M. Jullian, ' comparez v, 53, 1 a vii, 3, 3,
et vous verrez que le chiffre de 60 milles, entre les camps de Ciceron et de
Labienus, est evidemment trop faible.' M. Jullian means that if news was
transmitted over a distance of 160 Roman miles from Cenabum in about
12 hours, it ought to have travelled 60 Roman miles in less than 9. But how
can he tell that the circumstances were the same ? See pp. 736-7.
® Nouveaux mem. de V Acad. Roy. de Bruxelles, ii, 1822, pp. 241, 243.
. . .

' lb.,
pp. 241-4. 8 Gall. Krieg, 1880,
pp. 170, 173-4.
Caesar {B. G., v, 27, § 9) makes Ambiorix say, in his interview with Arpineius
**

and Junius, that the camp of Cicero was about 50 Roman miles from Atuatuca,
and that of Labienus a little more {ipsorum esse consilium velintne prius quam
 384 ATUATUCA
Remi ; it was close to a river, which was fordable ; and if it was
identical with the camp which Labienus occupied
53 B.C., that in
river, or at all events a river in the neighbourhood, had steep banks.
In 53 B. c. Labienus quitted his camp, and marched 14 miles against
the Treveri, who were separated from him by the river with steep
banks. On his side of the river and between him and his old camp,
there was a knoll (tumulus), on which he parked his baggage.^ M.
Jullian has not the slightest doubt that the old camp was at Mouzon,
on the mont de Brune ', where, he says, the Meuse was fordable
'

and he identifies the knoll with a hill between Mouzon and Izel
and the river with steep banks with the Semoy.^ I do not gainsay
my friend's discovery but is the district between Mouzon and Izel
;

the only one in Luxemburg which corresponds with Caesar's narrative?

And is it not difficult to explain away the fact that Mouzon is about
85 miles, in a straight line, from Tongres, which he identifies with
Atuatuca ?
Atuatucl. — The
Atuatuci and the Eburones were, in the opinion
of Desjardins and others, practically one people
^ but it is difficult ;

to understand how any one who knows the Commentaries can maintain
such a paradox. Caesar clearly implies, again and again, that the
two peoples and their territories were distinct. Ambiorix, one of the
two kings of the Eburones, says that he has been relieved by Caesar
from the obligation of paying tribute to the Atuatuci, his neighbours.*
Again, after his victory over Sabinus, Ambiorix marched into the
'
country of the Atuatuci, who were conterminous with his kingdom
'

(qui erant eius regno finitimi),^ and persuaded them to join him in
attacking Cicero. In the following year (53 B.C.) Trebonius was
sent to ravage that part of the country of the Eburones which
bordered on the country of the Atuatuci.^ Lastly, Caesar says that

finitimi sentiant eductos ex hibernis milites aid ad Ciceronem aut ad Labiemim

deducere, quorum alter milia passuum circiter L, alter pauh amplius ab iis absit).
General Creuly, referring to B. G., vii, 17, § 2 {de re frumentaria Boios atque
Haeduos adhortari non destitit ; quorum alteri^ quod nullo studio agebant, non
Tnultum adiuvabant ; alteri non magnis facultatibus quod civitas erat exigua et
infirma, celeriter quod habuerunt consumpserunt) argues that the first alter refers
to Labienus and the second to Cicero. But the two passages are not parallel.
In the second it is made clear by the context that the first alteri refers to the
Aedui and the second to the Boi whereas in the other passage the context
;

does not help us, and to prove his point Creuly ought to be able to show that
Caesar habitually uses alter alter in an inverted order.
. . . But Heller {Philo-
logus, xxii, 1865, p. 154) caps his quotation by another from B. G., v, 54, § 4
(ut praeter Haeduos et Memos, quos praecipuo semper honore Caesar habuit, alteros
pro vetere ac perpetua erga popidum Romanum fide, alteros pro recentibus Gallici
belli officiis, nulla fere civitas fuerit non suspecta nobis). In this passage it is
unquestionable that the first alteros refers to the Aedui and the second to the
Remi. On Creuly' s theory, in v, 27 Caesar's meaning would have been doubtful
even to Roman readers but it is clear that as, in that passage, the context
;

does not fix his meaning, he used alter alter in their natural order.
. . .

1 B. G„ V,
24, § 2 ; 58, § 6 vi, 7, § 5 ; 8, § 3.
;

2 Rev. des etudes anc, x, 1908, p. 266. Cf. M. Jullian's Hist, de la Gaule,
iii, 394, n. 6, where he adds that Mouzon (Mosomagus) is on the great road from

Reims to Treves.
3 Geogr. de la Gaule rom., ii, 437-8, 457. " B.G., v, 27,
§ 2.
5 76., 38, § 1. 6 /6.,vi,33,§2.
 ATUATUCI 385

the Atuatiici had numerous strongholds (o^ppda) and that the

Eburones had none.^
I. According to Dion Cassius,^ who is no authority, the territory of

the Atuatuci bordered on that of the Nervii but I have no doubt ;

that he is right here, for, according to Caesar,^ a part at least of their

territorywas between that of the Nervii and that of the Eburones.
The Eburones had some territory on the west of the Meuse, east of
the Nervii and south of the Menapii but the greater part was ;

between the Meuse and the Rhine.'* South of the Meuse, in the
neighbourhood of Condroz, were the Condrusi. Accordingly the
Atuatuci are generally placed in the valley of the Meuse, principally
along its left or northern bank, between the Nervii and the Condrusi.
The French Commission, who identify Atuatuca with Tongres,
consider that the name Atuatuca proves that the Atuatuci possessed
the vast plain which is dominated by Tongres.^ But Caesar expressly
says that Atuatuca was in the country not of the Atuatuci but of
the Eburones ^ and, moreover, as I have shown in the preceding
;

note, it is at least doubtful whether Atuatuca was on the site of

Tongres. The Commission also argue that the Atuatuci must have
possessed Namur, because numerous coins bearing the legend
AVAVCIA, which Caignart de Saulcy attributed to them, have been
found there The attribution is imaginary ^ but, assuming its
.'^
;

correctness, what would the discovery of the coins prove ? Coins

belonging to the Helvetii, the Santoni, the Sequani, the Carnutes,
the Senones, the Lingones, the Bituriges, and the Atrebates have
been found on Mount Beuvray ^ (Bibracte) but these peoples did ;

not live in Aeduan territory. Still, the orthodox faith is that the
Atuatuci occupied Namur and its neighbourhood,!^ that is to say,
the district of Hesbaye, on the northern bank of the Meuse, and
perhaps also some little territory in the western part of Condroz,
on the southern bank of the same river.^^
A novel view regarding the geographical position of the Atuatuci
has been propounded by A. de Vlaminck, and accepted by M. A.
Longnon.^2 According to this view, as originally stated, they
occupied that part of the country between the Meuse and the Rhine
which subsequently belonged to the Ubii. De Vlaminck took his stand
upon the well-known passage in which Caesar says that the Atuatuci
were descended from the Cimbri and Teutoni, who, when they were
about to invade the Province and Italy, had left a detachment on

' B. G., 29, §§ 1-2 ; vi, 34, §

ii, 1. ^ ^xxix, 4, § 1.
' B.G, V, 38, §§ 1-2. * Ih., 24, § 4.
5 Diet. arch, de la Gaule, i, 11. ^ B. G., vi, 32, §§ 3-4.
^ Did. arch, de la Gaule, i, 11.
« 8ee Rev. celt, xxx, M. Blanchet {ih., xxxi, 1910, has
1909, p. 194. p. 57)
no doubt that Avaucia was the name of a monetary agent.
^ Diet. arch, de la Gaule, t. i.
^^ D'Anville, Notice de Vancienne Gaule, p. 33.
^^
M. Jullian {Hist, de la Gaule, ii, 465, n. 4) assigns Hesbaye to the Eburones
and the country round Namur to the Atuatuci. But how much of the country
round Namur ? If the Atuatuci did not possess Hesbaye, they must have
been very badly off. ^2
j^fias hist, de la Gaule, p. 4.
im C C
386 ATUATUCI
the left bank of the Rhine to protect all the impedimenta that they
were unable to take with them.^ As, argued de Vlaminck, the detach-
ment evidently had to keep open the communications of the Cimbri
and Teutoni, and to cover their retreat in case of a disaster, it would
be absurd to suppose that their territory was so far removed from the
Rhine as is generally supposed. He also insisted that so powerful
a tribe would never have been confined in the narrow space between
the Eburones and the Nervii.^
The first of these arguments is childish the Atuatuci were not
:

required to keep open the communications of their kinsmen for half

a century. In reply to the second, it may be said that, judging from
the size of the contingent which they furnished to the Belgic army
in 57 B. c, they were far inferior in strength even then to the Bellovaci,
the Suessiones, and the Nervii ^ and also that it is quite possible that,
;

after the crushing blow which Caesar inflicted upon them in that
year, when he slew 4,000 of them, sold 53,000 into slavery, and
released the Eburones from the obligation of paying them tribute,*
he may have transferred to the Eburones a portion of their territory,
including the fastness of Atuatuca.
De Vlaminck's arguments naturally provoked hostile criticism ;

and he subsequently modified his theory.^ It would, however, be

waste of time to examine the fresh arguments by which he endea-
voured to prove that they never crossed the Meuse ^ for a single ;

passage in the Commentaries stultifies them. In B. G., v, 38, §§ 1-2,

describing how Ambiorix followed up his victory over Sabinus and
Cotta, Caesar writes Hac victoria suhlatus Amhiorix statim cum
:

equitatu in Atuatucos, qui erant eius regno finitimi, proficiscitur :

neque noctem neque diem intermittit peditatumque se suhsequi iuhet.
Re demonstrata Atuatucisque concitatis, postero die in Nervios pervenit,
&c. After reading the first of de Vlaminck's later papers, I wrote,
'
this passage proves that the Atuatuci were between the Eburones,
from whose territory Ambiorix started, and the Nervii.' But by
extending the territory of the Atuatuci westward a little beyond
Embourg and placing Atuatuca at Aix-la-Chapelle, de Vlaminck
labours to force his theory into conformity with Caesar's narrative.
Yet even so, Ambiorix would only have passed across the extreme
western corner of the country of the Atuatuci, and would then have
had to recross the territory of the Eburones or to cross the territory
of the Condrusi before reaching that of the Nervii. But no unbiassed
reader can fail to see that, taken in their natural sense, Caesar's
words imply that the territory of the Atuatuci intervened between
that of Ambiorix and that of the Nervii.
Nevertheless, I believe that in de Vlaminck's pages there is a grain
of truth. He wholly fails to prove that the Atuatuci did not possess
the territory on the western bank of the Meuse which, until he wrote,
1 B. G., a, 29, § 4. ^-
La Menapie, &c., 1879, pp. 42-4.
' B. O., ii, 4, §§ 5-9. * Ih., 33, § 7 ; v, 27, § 2.
* Messager des sciences hist, de Belgique,
1882, pp. 391-2 ; 1884, p. 282 ; 1887,
pp. 39, 55-6, 351-2.
" I did this work of supererogation in the
first edition (pp. 350-2).
 — ^ ;

ATUATUCI 387

geographers had generally assigned to them. But it is probable, or

at least possible, that, before their treachery provoked the vengeance
of Caesar, they also possessed a tract between the Meuse and the
Rhine.
II. The Eburones are not mentioned by any ancient writer except
Caesar. The greater part of their country was between the Rhine
and the Meuse, from which it may be inferred that the remaining
part was west of the Meuse. Their neighbours on the south were
the Treveri, the Segni, and the Condrusi on the north the Menapii
;

and on the south-west, I believe, the Atuatuci. Their territory may

also have been conterminous on the west with that of the Nervii.^
But it is impossible to define the frontiers of any of these peoples,
except perhaps the Nervii and therefore it is impossible to define
;

the frontiers of the Eburones. General Creuly endeavours to prove

that their territory extended northward as far as the sea, and separ-
ated the Menapii from the Morini.^ This theory contradicts the
evidence of Strabo, who says that the territories of the Menapii and
the Morini were conterminous.^ But Creuly relies on Caesar. Caesar
says that in the sixth year of the war those of the Eburones who
dwelt nearest the sea took refuge in the islands which were formed
at high tide.* These, says Creuly, must have been the islands at the
mouths of the Scheldt. They must have belonged to the Eburones
because Caesar, after describing his campaign, says, haec in omnibus
Eburonum finibus gerebantur.^ But Creuly unduly strains Caesar's
text. I have no doubt that his statement refers only to the events
described in B. G., vi, 33-4. If it refers to chapters 31 and 32 as well,
it obviously includes events which did not take place in the country
of the Eburones for in chapter 32 he describes what the Segni and
;

Condrusi were doing and in chapter 31, on which Creuly lays great
;

stress, he says that many of the Eburones fled beyond their own
borders. I believe, then, .that those Eburones who took refuge in
the islands were not at the time in their own country and I con- ;

clude that Strabo was right. See also page 375.

—
Atuatucorum Oppiduni. According to Caesar, the stronghold
of the Atuatuci was steep and rocky on every side, except at one
place, where it was approached by a gentle slope not more than
200 feet wide cunctis oppidis castellisque desertis, he writes, sua
:

omnia in unum oppidum egregie natura munitum contulerunt. Quod

cum ex omnibus in circuitu partibus altissimas rupes despectusque
kaberet, una ex parte leniter acclivis aditus in latitudinem non amplius
pedum CC It was large enough to shelter at least
relinquebatur.^
57,000 people and the line of contra vallation with which Caesar
;
^

surrounded it was traced along ground comparatively high but of

varying elevation, and measured 15 or, according to another in-
terpretation of the text, only 3 Roman miles.

' B. G., V, 24, § 4 ; vi, 32, § 1 ; 33, §§ 1-2.

^ Rev. arch., nouv. ser., vii, 1863, pp. 385-6.
''
See p. 450. '^
B. G'., vi, 31, § 3.
' lb., 35, § 1. « lb., a, 29, § 3.
' lb., 33, §§ 5-7. « lb., 30, § 2. See pp. 390-1.
C C 2
388 ATUATUCORUM OPPIDUM
A score of sites or more have been suggested but the majority ;

are not worth examining. Either they do not correspond with

Caesar's description, or they are in territory which could not have
belonged to the Atuatuci.^
The great Napoleon, probably following d'Anville, decided for
Falais, near Huy, a height which is nearly surrounded by the river
Mehaigne.2 Des Roches and others ^ have suggested Hastedon, which
is quite close to Namur Napoleon III pronounced for the citadel of
:

Namur von Goler, the French Commission,

: K,. Thomann, von
Kampen, and Kiepert for Mont Falhize.
1. According to the researches,' says Napoleon, which have
' '

been carried on in the country supposed to have been formerly

occupied by the Atuatuci, two localities only, Mont Falhize and the
part of the mountain of Namur on which the citadel is built, appear
to agree with the site of the oppidum of the Atuatuci. But Mont
Falhize is not surrounded with rocks on all sides, as the Latin text
requires. The contravallation would have had a development of
more than 15,000 feet, and it would have twice crossed the Meuse,
which is difficult to admit. We therefore adopt the citadel of . . .

Namur.' ^

Long, in his edition of Caesar,^ rejected Namur, on the ground

that it is on the Maas, a circumstance incompatible with Caesar's
'

description, for he mentions no river '. But, in his History,^ he

accepted Mont Falhize, which is also on the Meuse, and therefore
virtually withdrew his objection. Long also argued, as d'Anville
"^

had done before him, that Namur is too small for its area is only ;

12 hectares,^ or about 27 acres. M. Jullian contends that additional

room might have been gained if some of the garrison had encamped
on the flanks of the hill but even so it would hardly have been
;

possible to find a greater space than 52 acres, or 40 square feet for

each man.^ If the plateau had been covered by houses several
storeys high, the garrison might have found accommodation but ;

can we conceive that they were packed for days, like the crowd at
the boat-race, along with numerous cattle,^^ nearly four times as
tightly as the garrison of Alesia ? ^^ Dewez, indeed, urges i- that,

^ Diet. arch, de la Gaule, p. 1 1

2 —
See A. E. E. Desjardins, Alesia, suivie (Tun appendice renfermant des notes
inedites ecrites de la main de Nap. I., &c., and d'Anville, Notice de Vancienne
Gaule, J).
34.
^ Nouveaux mem. de V Acad, Roy. des sciences de Bruxelles, ii, 1822,
. . .

pp. 251-2. * Hist, de Jules Cesar, ii, 116, n. 1.

^ Page 135. ^ Decline

of the Roman Republic, iv, 63—1.
' Notice de Vancienne Gaule,
pp. 33-4.
* C. JulHan, Hist, de la Gaule, iii, 270, n. 2.
'^
See Napoleon's Planche 11 and A. von Goler, Gall. Krieg, 1880, p. 91, n. 4.
M. Jullian himself {op. cit., iii, 556, note) rejects the proposed identification
of Uxellodunum with Luzech partly because the area of the latter is only
8 hectares but the garrison of Uxellodunum, exclusive of the townsfolk, was,
;

on the highest estimate, less than one-tenth of that of the stronghold of tlie
Atuatuci.
1" 5. G^.,ii,33,§2. " Seep. 243.
'2 Noiiv. mem. de VAcad. Roy. de Bruxelles, ii, 1822, p. 247.
 ATUATUCORUM OPPIDUM 389

although the citadel of Nainur is too small in itself, it communicates

with other rocky heights, and that all of them taken together would
have been sufficiently large. Caesar, however, does not describe
a group of hills, but a single hill and even if the chain of heights
;

the easternmost of which is occupied by the citadel could be regarded

as a single hill, their length and extent are such that they could not
have formed one stronghold, and they are nowhere approached by
a gentle slope, 200 feet wide. The writer of the article Aduatuci in
the Dictionnaire arcMologique de la Gaule rejects Namur, not only
because it is too small, but also because (1) the gentle slope is not to
be found ^ and (2) if the contravallation had been made to follow
;

the windings of the Sambre, which, in his opinion, would not have
been a sufficient obstacle, it would have required a development of
9 kilometres, or 6 Roman miles ^ whereas, if Caesar had merely
;

drawn the line over the heights from the bank of the Meuse to the
bank of the Sambre, two kilometres, or barely a mile and a quarter,^
would have sufficed. As a matter of fact, Napoleon, in his plan
(Planche 11), adopts the latter alternative, thereby contradicting
his own interpretation of AF
milium\ But if any one will ascend
'

the hill on which the citadel is built, or even inspect the model,
executed by M. Locqueyssie for Napoleon, which is in the museum
of Namur, he will see that other reasons militate against Napoleon's
choice. Caesar's contravallation would necessarily have been carried
up the steep and rocky flanks of the hill and over ground consider-
ably higher than that on which the oppidum is supposed to have
stood. Why should the Atuatuci have restricted the ofpidum
within absurdly narrow limits ? Why should they have left the
higher ground unoccupied ? According to M. Locqueyssie, remains
of an old earthwork, which he describes with a note of interrogation
as gallo -beige ', were discovered about 500 metres south of the
'

slope which, according to Napoleon, led up to the double wall of

the oppidum. Suppose that this earthwork was really erected by the
Atuatuci. In that case, if Namur represents the oppidum, the slope
did not exist the wall, which would have to be identified with the
:

earthwork, was about ten times as wide as that which Caesar de-
scribes and he would still have been obliged to carry his contra-
;

vallation right up and over the rocky heights. Again, if we accept

Napoleon's theory, in what direction could the Atuatuci have made
their desperate sortie ?* Not down the gentle slope, which ex hypothesi
was already occupied by the Roman agger. As far as I can see, they
would either have headed north-westward along the low ground
parallel with the Sambre, or southward along the eastern flank of
the hill or the narrow strip of ground below, parallel with the Meuse.
But in either case, if they had succeeded in storming the contra-
vallation, would they not have been trapped between river and hill ?
2. The writer of the article, already referred to, in the Dictionnaire

^ Napoleon marks the slope in his plan (pi. 11) but is it minime arduus ?
;

^ This is an exaggeration 4^ kilometres would be nearer the mark.

:

^ The contravallation which Napoleon traces in his plan (pi. 11) is even less.
' B. G., a, 33, § 2.
390 ATUATUCORUM OPPIDUM
arcMologique de la Gaule says that of all the localities which have
been proposed Beaumont, Namur, and Mont Falhize alone deserve
serious consideration. he rejects Namur. Beau-
As we have seen,
mont he also rejects because only 25 kilometres, or about 15 miles,
it is
from Neuf-Mesnil, the site of the battle with the Nervii,^ and, as the
Atrebates and the Viromandui, who also took part in the battle,
had to travel respectively 80 and 100 kilometres, the Atuatuci, if
they had only had so short a distance to go, could have easily arrived
in time. But there is no reason to assume that, because the Atuatuci
took refuge against Caesar in a certain stronghold, therefore that
stronghold must have been the point from which they marched to
join the Nervii. There are, however, other reasons for rejecting
Beaumont. First, the plateau of the alleged oppidmn is smaller
even than the plateau of Namur ^ and secondly, Beaumont is not
;

in the territory which belonged to the Atuatuci.

3. General von Goler,^ who carefully explored Mont Falhize, con-
sidered that it corresponded exactly with Caesar's description of the
stronghold and General Creuly, after examining the Carte topo-
;

graphique de la Belgique, came to the same conclusion, and also re-

corded his belief that no other site in the whole of Belgium fulfilled
the conditions of the problem.^ Napoleon's objections to this site
have been already stated. In answer to them it may be said (1) that
quod cum omnibus in circuitu partibus altissimas rupes despectusque
haberet describes, as the writer of the article in the Dictionnaire says,
a hill with scarped flanks, d'ou la vue dominait sur la campagne
'

environnante,' ^ which description applies to Mont Falhize and ;

(2) that the contravallation need not have had a development of '

more than 15,000 feet ', because it need not have crossed the Meuse
at all. As the writer of the article in the Dictionnaire observes, the
Meuse, which encloses the southern slopes, is, on that side, a sufficient
obstacle and Caesar need only have left a corps of observation on
;

the right bank at Huy, to watch it. Moreover, the word circummuniti
does not necessarily mean that the contravallation entirely sur-
rounded the town ^ and in fact Napoleon's Plan does not make it
;

surround Namur or cross the Meuse.

The extent of the contravallation is, however, uncertain. The
reading of p, which Meusel adopts, is vallo p. XII in circuitu XV
milium crebrisque castellis circummuniti. Schneider supplies passuum
with XV milium and, if he is right, the extent of the rampart was
;

15 Roman miles. As XII is omitted in a, FrigelH believes that the

true reading is vallo passuum in circuitu milium. A. Holder ^ XV
^ See pp. 671-7.
2 See Carte topogr. de la Belgique (1 : 20,000), Feuille 52, pi. 6.
^ Gall. Krieg, 1880, pp. 91-5.
* Rev. arch., nouv. ser., viii, 1863,
p. 392. See also O. Bocquet in Bull, de
Vinst. arch, liegeois, v, 1862, pp. 167-76, and A. Hock, Etudes sur quelques
campagnes de Jules Cesar dans la Gaule helgique, 1897.
5 Cf Schneider's Caesar, i, 191.
" Cf. —
B. G., i, 38, § 4, flumen Dubis ut circino circumductum paene to turn
oppidum cingit.
' Caesar, i, 38 ; ii, 23. « Caesar, p. 50.
 —
ATUATUCORUM OPPIDUM 391

prefers the reading of a, vallo pedum in circuitu XV

milium, that
is to say,with a rampart 15,000 feet in extent
'
but whoever
' ;

collates the relevant passages in the Commentaries will conclude that

Caesar would not have reckoned miles in terms of feet.^ Creuly and
Bertrand, who are followed by Napoleon,^ adopt the reading of (3,
but supply pedum after XV
milium, remarking that Caesar nowhere
else uses milia or its cases, without passuum, to express miles '.^ '

But this is an astounding blunder as, in a very incomplete search,

;

I have found no less than three passages where he does so.* Schneider
says that to supply pedum after XV
milium would be contrary to
the usage of Latin but Heller
; believes that the word pedum is
employed, as a joint factor
'
', to be coupled both with XII and
with XV
milium.^ I am sure that he is wrong but at all events it
;

is certain that to construct a line of contra vallation 15 Roman miles

in extent round either Namur or Mont Falhize would have been
contrary to the usage of a sane, not to say a great, general.
Other objections have been urged against Mont Falhize. The
Atuatuci, we are told, would not have built their principal strong-
hold near their north-eastern frontier, which separated them from
the Condrusi. To this General Creuly replies that we do not know
what that frontier was, and that if the Condrusi occupied the right
bank of the Meuse, that was the very reason why the Atuatuci
should have established a stronghold on the opposite bank. More-
over, the general considers that of all the places which they could
have chosen for a city of refuge Mont Falhize was naturally the
best.6
4. The same objection applies to Falais as to Mont Falhize, it is —
nearly surrounded by a river and in other respects it does not
;

conform so closely to Caesar's description.

5. MM. G. Arnould and de Radigues consider that the plateau of
'^

Hastedon, about a mile and a half north-west of Namur, corre-

sponds with Caesar's description more nearly than any other place
which has been suggested and they claim that its area, 13 hectares
;

or about 32 acres, is not too small. In France, they remark, on '

admet comme surface minima, pour le campement de I'infanterie

sous la tente, un rectangle de 48 m. carres, pour 15 hommes . . .

soit 3 m
2. 20 par homme. Admettons 13 hectares 25 ares pour
la superficied'Hastedon on trouve 2; m
2. 45 par individu.' They
omit to say that, as even soldiers must have room to move about,
the tents in a camp do not cover anything like the whole area and ;

even so they allot much less space to the unhappy Atuatuci than
the surface minima which they mention. But without elaborate
' '

^ See p. 729. ^ Hist, de Jules Cesar, ii, 117, n. 1.

Rev. arch., 2® ser., iv, 1861, p. 459.

•^

* 5. (?.,i,41,§4; iii,
17, §5; iv, 14,§L See also Meusel's Lex. Cae-?., ii, 609-10.
^ Philologus, xxvi, 1867, pp. 665-7. Meusel does not agree with Heller,
and in a MS. note suggests that F is corrupt.Z
« Rev. arch., 2" scr., vii, 1863,
p. 392.
' Congres internal, d'anthr. et d'arch. prehist. (Bruxellcs, 1872), 1873, pp. 318-
26 ; Annates de la Soc. arch, de Namur, xii, 1872-3, pp. 229-39.
392 ATUATUCORUM OPPIDUM
calculation any one can see that to pack 57,000 men into a space
of 32 acres and keep them there for several days would be impossible.
In conclusion, I ought to say that M. de Vlaminck, followed by
M. A. Longnon,! considers that nearly all the attempts that have
been made to identify the stronghold were foredoomed to failure,
because they were based upon the assumption that the territory of
the Atuatuci was confined to the region of Namur, whereas it really
extended nearly as far eastward as the Rhine. I have examined
elsewhere ^ M. de Vlaminck's theory of the habitat of the Atuatuci.
He himself identifies the oppidum with Embourg, on the east of the
Meuse, near the junction of the Vesdre and the Ourthe ^ but this ;

place does not correspond with Caesar's description. Cohausen and

others have assumed that the stronghold is identical with Atuatuca,^
the site of the camp in which Sabinus and Cotta were quartered in
the autumn of 54 b. c. It is hardly necessary to refute this assump-
tion, for which there is not a particle of evidence * but one fact is
sufficient to demonstrate its absurdity. The stronghold of the
Atuatuci was situated upon a steep and rocky hill, which even
Caesar's powerful army could only approach, with the aid of a
terrace, by one narrow slope Atuatuca, although a strong place,
:

was so situated that in 53 b. c. 2,000 German cavalry nearly carried

it by a coup de mainJ*
To sum up, Mont Falhize conforms perfectly, in the judgement of
trained professional observers, to Caesar's description. On the other
hand, he does not mention the Meuse or any other river ^ no :

antiquities had been found on Mont Falhize up to the date (1868)

when the first part of the Dictionnaire archeologique de la Gaule was
published, and I have not been able to discover that any have been
found since the place was not, so far as we know^ occupied in the
;

Middle Ages, nor has it been occupied in modern times. A more

serious objection presented itself to me when I visited Mont Falhize :

I was unable to detect the narrow gentle slope on the eastern side,
the existence of which von Goler affirms. Possibly it may have been
obliterated by the earth and rubble of which the ajger would have
been largely composed but if it exists at all, it is insignificant.''
;

^ Atlas hist, de la France, p. 4, - See pp. 385-7.

3 Messager des sciences hist, de Belgique, 1887, pp. 46, 49. Cf. Jahrb. d.
Vereins von Alterthumsfreunden im Rheinlande, xliii, 1867, pp. 26-36.
*
See A. von Kampen, Quindecim ad Caesaris de h. G. comm. tabulae, ii.
6 B. G., ii, 29, § 3 ; vi, 37-8.
^ It has-been argued that he did not mention the Meuse because it played no
part in the siege (A. Hock, Etudes sur quelques campagnes de Jules Cesar, &c.,
1897, p. 22) ; but surely it would have played the part of a natural line of
contra vallation.
' Von Goler {Gall. Krieg, 1880, p. 94) says that the ground falls eastward about
20 feet {Oestlich gegen das Gehillze von Huy hin senkt sich das Terrain um ungefahr
20 Fuss). M. Jullian {Hist, de la Gaule, iii, 270, n. 2) says, Je trouve bien
'

a Falhize le col de 200 pieds dont parle Cesar . . mais il ne presente pas une
.

inclinaison ou une depression assez sensible pour justifier ce qu'il en dit.'

As at Namur, the contravallation would necessarily have been carried up
and down, but only in a small fraction of its extent. The minime arduus
ascensus would have been on the north.
 ATUATUCORUM OPPIDUM 393

Moreover, I agree with M. de Vlaminck that the territory of the

Atuatuci, before they came into collision with Caesar, extended
further eastward than is commonly supposed.^ Therefore, although
I mark the oppidum upon the map, I only do so tentatively.
—
Aulerci Brannovices. The Aulerci Branno vices and Blannovii
are only mentioned by Caesar, and only once by him and he tells ;

us nothing about them, except that they were clients of the Aedui.^
To add to the difficulty of identifying their territories, the MSS.
offer a variety of readings. A has Blannovicibus instead of Branno-
vicibus. Nipperdey follows one of the old editions in reading
Brannoviis, instead of Blannoviis, Ciacconius deleted Blannoviis on
'

the ground that its insertion was probably due to dittography and ;

Meusel ^ now follows his example.

The French Commission * agree with d'Anville ^ that the territory
of the Brannovices may be represented by the canton of Brionnais,
on the eastern bank of the upper Loire. Walckenaer, who accepts
the reading (Aulercis) Blannovicibus, Blannoviis, places the Blanno-
vices in the neighbourhood of Blannot in the arrondissement of Macon,
and the Blannovii in the neighbourhood of another Blannot in the
arrondissement of Beaune.^ M. Longnon proposes to identify the
territory of the Brannovices with the diocese of Auxerre, which
he assigns to the Aedui (q.v.). He bases his conjecture on la '

mention d'un " quidam Aeolercus ", que la legende du premier

eveque d' Auxerre montre elevant un temple paien a Entrains
(Nievre) '?
The whole matter is quite uncertain, as we have no evidence
except that of nomenclature. I follow d'Anville, but doubtfully, in
assigning to the Brannovices the canton of Brionnais. As to the
Blannovii, supposing that they existed, nobody can tell where they
lived for there are two Blannots, and one might as well toss up as
;

attempt to decide between them.

—
Aulerci Cenomani. The Cenomani occupied that part of the
diocese of Le Mans w^hich did not belong to the Diablintes (q.v.)
and the Arvii but the frontier can only be conjecturally traced.
;

Their territory corresponded, roughly, with the department of the

Sarthe.^
Aulerci Eburovices. — The Aulerci Eburones appear in the MSS.
among the tribes who sent troops to the relief of Vercingetorix.^
De Valois,!^ however, conjectured that iov Eburonibus should be read
Eburovicibus, (1) because the Aulerci Eburovices appear in B.G.,
iii,17, §3, side by side with the Lexovii, immediately after whom,
the Aulerci Eburones are mentioned in the passage which I am dis-
cussing and (2) because the Aulerci Eburones are mentioned
;

^
See the preceding article. ^ B. G., vii, 75, § 2.
^
Jahresb. d. philol. Vereins zu Berlin, xxxvi, 1910, p. 60.
* Did. arch, de la Gaule, i, 93. ^ Notice de Vancienne Gaule, p. 129,
^
Geogr. des Gaules, i, 331. ' Atlas hist, de la France, p. 4.
^
D'Anville, Notice de Vancienne Gaule, p. 130 ; Diet. arch, de la Gaule, i. 93.
Cf. F. Liger, La Cenomanie rom., 1904, pp. 3-4.
^
B. G., vii, 75, § 3. »« Notitia Galliarum,
p. C6.
394 AULERCI EBUROVICES
nowhere else by Caesar and by no other writer, whereas the Eburo-
vices are mentioned not only by Caesar, but also by Pliny ,i Orosius,^
and later writers. De Valois's emendation is accepted by most
editors but Schneider retains the reading of the MSS., believing
;

that if there was a mistake, it was made by Caesar in consequence

of his having been misinformed. I think that de Valois's arguments
are sufficient and it is significant that in B. (r., iii, 17, § 3, where
;

Caesar mentions the Eburovices, one inferior MS. has Eburones.^

The Eburovices occupied the diocese of Evreux, or the central and
the southern part of the department of the Eure.^
—
Ausci. The Ausci occupied the southern part of the diocese of
Auch, that is to say, the central and the southern part of the
department of the Gers.^
—
Belgae. Are the Treveri to be included among the Belgae ?
Caesar's introductory chapter, taken by itself (unless, as Meusel ^
contends, the last three sentences were interpolated), certainly
implies that this question is to be answered in the affirmative.
After telling us that the Gauls, or Celtae, are separated from the '

Belgae by the Marne and the Seine,' he says, that part of the '

whole country which, as we have said, is occupied by the Gauls,

begins at the river Rhone, and is bounded by the Garonne, the
Ocean, and the country of the Belgae it extends, moreover,
: in
'
the region occupiedby the Sequani and the Helvetii, to the Rhine
{eorum una fars, quam Gallos ohtinere dictum est, initium ca'pit
a flumine Rhodano : continetur Garumna flumine, Oceano, finibus
Belgarum attingit etiam ab Sequanis et Helvetiis flumen
;

Rhenum)? Now he says elsewhere that the territory of the

Treveri extended to the Rhine.^ But if they had been Celtae, not
Belgae, he ought to have written, in the passage which I have

1 Nat. Hist., iv, 18 (32), § 107. ^ vi,8, § 18.

* Schneider' s Caesar, i,263. Diet. arch, de la Oaule, i, 94.
*

* Walckenaer, i, 288 ; Diet. arch, de la Gaule, i, 97-8.

^ Jahresb. d. philol. Vereins zu Berlin, xxxvi, 1910, pp. 20-3. With one
exception, Meusel' s arguments appear to me inconclusive, for they imply that
Caesar was incapable of writing carelessly, and certain objections which he
makes on the score of orientation would hold good against the authorship of an
interpolator as well as against that of Caesar but if he is right in maintaining
;

that the words (Belgae) a b extremis Galliae finibus oriuntur and the singular
septentrio (which is used by Livy, xxxii, 13, § 3) could not have been written
by Caesar, the passage must be condemned. Still, if it is an interpolation,
it must have been derived from a writer earlier than Augustus for the boun- ;

daries which it assigns to Aquitania are those of Caesar's time. Is it likely that
the interpolator, who, according to Meusel [op. cit., pp. 73-4), lived in the third
or the fourth century, would have been ignorant of the boundary as it existed in
his own day ? [A. Klotz {Caesar studien, pp. 27-30) also condemns the passage
for reasons which are much the same as those of Meusel. He argues that
initium capit (in the sense in which it is used in § 5), ab (Sequanis), spectant
in and spectant inter, and (Pyrenaeos) montes instead of saltiis are not expres-
sions which Caesar would have used. Apparently he believes that the statement
about the northern boundary of Aquitania continetur Garumna fi,umine was —
derived from Timagenes.]
' B. G., i,
1, §§ 2, 5.
^ lb., vi, 9,
§ 5. The authenticity of iv, 10 is more than doubtful. See p. 692

i
 —
BELGAE 395

just quoted, attingit etiam ah Treveris, Sequanis, Helvetiis Jlumen

Rhenum. Moreover, as tlie Condrusi, clients and neighbours of
the Treveri, were undoubtedly Belgae,^ it might be inferred
that the Treveri themselves belonged to the same group. On
the other hand, the Renii told Caesar in 57 B. c. that all the
Belgae, except themselves, were in arms and the Treveri are ;

not mentioned in the list which they furnished of the various

contingents.2 Again, there is a passage in B. G., vi, 3, § 4,
which might suggest that Caesar reckoned the Treveri as Celtae.
He says that when he held his annual council of Gallic chiefs
in the spring of 53 b. c, all the states, except the Senones, the
Carnutes, and the Treveri, sent their representatives {concilio Galliae
prinio vere, ut instituerat, indicto, cum reliqui praeter Senones, Carnutes
Treverosque venissent, &c.) and, as the preceding chapter proves
;

that the Nervii, the Atuatuci, and the Menapii all Belgic peoples —
— were not represented. Long ^ infers that Caesar was only speak-
ing of the states of Gallia Celtica. After looking at both sides of
the question, it seems reasonable to conclude, from the fact that the
Treveri do not appear in the list of the Belgic tribes and from the
fact that they actually assisted Caesar in his first campaign against
the Belgae,^ that, according to his informants, they were reckoned
among the Celtae. Still, there were striking differences, which
Hirtius noticed, between the Celtae and the Treveri.^
—
Belgium. Did Caesar use this word in B. G. v, 24, § 2, and if so,
in what sense ? He mentions Belgium three times (v, 12, § 2 24, ;

§ 2 ;25, § 4), or, according to the a MSS., twice (12, 25). In the
—
former of these two chapters I may here assume that it is genuine,
although Meusel and A. Klotz regard it as an interpolation^ he —
says that the maritime part of Britain had been colonized by immi-
grants from Belgium, almost all of them being called after the
'

tribes from whom the first comers were an offshoot maritima ' :

'pars (incolitur) ah Us qui praedae ac helli inferendi causa ex Belgio

transierunt qui omnes fere Us nominihus civitatum appellantur, quihus
orti ex civitatihus eo pervenerunt. The natural conclusion is that
Belgium meant the country of the Belgae for, if it had meant only
;
"^

a particular part of that country, how could Caesar's readers have

known what part he meant ? In the second of the two chapters (25)
he says that he ordered Plancus to march with his legion from
Belgium to the country of the Carnutes and this statement, taken
;

by itself, would lead to the same conclusion. But in the preceding

chapter (24) there occurs a passage which has led most commen-
tators to adopt a different opinion. Describing the distribution of
the legions which he made in the autumn of 54 B. c, Caesar says that
he placed three, under Crassus, Plancus, and Trebonius respectively,
in the country of the Belgae, or, according to the /3 MSS., in Belgium •

— tres (legiones) in Belgis (or Belgio) conlocavit : his M. Crassum

' B. G., ii, 4 ;iv, 6 ; vi, 32, § 1. ' Ih., ii, 3, § 4 ; 4.
^ Caesar, p. 288. " B. G., 24, § 4.
ii,
^ lb., viii, 25, § 2. « See p. 206, n. 4.
^ Cf Schneider' s Caesar,
. ii, 56.
396 BELGIUM
quaestorem et L. Munatium Plancum et C. Trebonium legatos fraefecit.
Now all the other legions, except one, were also quartered in the
country of the Belgae and accordingly the reading Belgis would
;

seem to have no point. For this reason A. Holder, following Cluver,^

reads Bellovacis while d'Anville,^ Walckenaer,^ the French Com-
;

mission,* and Meusel (in the re-issue of his school edition) read Belgio.
Assuming that this is the true reading, it is obvious that, in this
passage at all events, unless Caesar was very careless, Belgium can
only mean a part of the whole territory of the Belgae. If so, what
was that part ? Now Hirtius ^ mentions Belgium four times but ;

we only learn from him that within the limits of Belgium was
Nemetocenna, the chief town of the Atrebates. A passage in v, 46, § 1
— Caesar . in Bellovacos ad M. Crassmn quaestorem mittit, cuius
. .

hiberna aberant ab eo milia passuum XXV—

proves that the camp of
Crassus, one of the three camps which, according to the passage in
V, 24, were in Belgium, was in the territory of the Bellovaci. As it
is clear, then, says d'Anville, that Belgium comprised the territories
of the Bellovaci and the Atrebates, it must have also comprised the
territory of the Ambiani, which was between the other two. This is
the orthodox view and if, in the doubtful passage (v, 24), the true
;

reading is Belgio, it is right.

But is Belgio the true reading ? It would be quite natural that
Caesar should have called the country of the Belgae Belgiwn but, ;

although he used the word Gallia both in a general and in a re-

stricted sense, it seems unlikely that he should have also used the
word Belgium to denote a portion only of the country of the Belgae.
On the other hand, the emendation Bellovacis does not mend matters ;

for it is certain that Trebonius's legion was at Samarobriva (Amiens)

in the country of the Ambiani .^ Caesar must have forgotten this,
if he wrote Bellovacis. If he wrote Belgis or Belgio, he must have
made a slip, unless the term Belgium was really used in a restricted
sense, to describe the territory of the Bellovaci, the Ambiani, and
the Atrebates, as well as in a general sense, to describe the whole
territory of the Belgae.
To sum up, if we adopt the reading Bellovacis, we adopt a purely
conjectural reading, and we have to admit that Caesar made a slip.
If we adopt either of the readings, Belgis and Belgio, we have the
support of good MSS. and although it seems probable that, in
;

either of these cases, Caesar made a slip, it is at least possible that in

the latter he did not. Taking probability as our guide, we must,
I think, decide in favour of Belgio. It may, indeed, be argued that it
is unlikely that Caesar, departing from his habitual practice, would
have said that certain legions encamped in Belgium, instead of
mentioning the peoples among whom they encamped and Nipper- ;

dey, who reads Belgis, remarks that, in v, 24, when we see that the
'

Belgae are mentioned after certain Belgic tribes, w^e must conclude
^
Oermania antiqua, 1631. p. 341.
^
Notice de Vancienne Gaule, pp. 147-8.
2
Geogr. des Gaules, i, 420-2. * Diet. arch, del a Gaule, i, 138.
5
B. G., viii, 46, §§ 3, 6 49, § 1
; 54,
; § 4. « See p. 371, n. 7.

J
 — ;

BELGIUM 397

either that Caesar thought it superfluous to state what Belgic tribes

he meant by Belgae or that he simply forgot to do so.' ^ But the
retort is obvious. If we read Belgio, we must conclude either that
Caesar thought it superfluous to state what part of the territory of
the Belgae he meant by Belgium or that he simply forgot to do so.
Again, it seems more likely that the word Belgium should have been
used in a restricted as well as in a general sense than that Caesar
should have used the word Belgae, the meaning of which he had
himself defined at the outset of his work, in a way which would
have been certain to mislead. Moreover, it is possible that the word
Belgium was only used in a restricted sense for in the chapter ;

(v, 12) in which we read that immigrants from Belgium settled in

Britain, there is a various reading —
found, it is true, only in inferior
MSS. Belgis. See Nipperdey's Caesar, page 355.
—
Bellovaci. The Bellovaci certainly possessed the diocese of
Beauvais. D'Anville,^ Walckenaer,^ and the French Commission*
also assign them the diocese of Senlis, corresponding with the terri-
tory of the Silvanectes,^ a people who are not mentioned by Caesar,
and who may have been one of their pagi. The Commission observe
that the diocese of Beauvais alone would not have been sufficient
for so important a state. But by the same reasoning the diocese of
Senlis might be, as it has been, assigned to the Suessiones, who,
according to Caesar,^ had a very extensive territory. To one or the
other of the two tribes, it has been argued, this intermediate tract
must have belonged for Caesar implies that their territories were
;
"^

conterminous. His statement is, indeed, not absolutely inconsistent

with the theory that the Silvanectes were an independent people ;

for in any case the Bellovaci and the Suessiones would have been
conterminous in the country north of the Silvanectes. Still, as he
does not mention the Silvanectes in his list of the Belgic tribes, I am
inclined to believe that they were dependants of or included among
the Suessiones.
The Commission further propose to extend the territory of the
Bellovaci to the sea, giving them the tongue of land which separates
the river Bresle from Arques, as this tract cannot be assigned with
certainty either to the Ambiani or to the Caletes^ (q-v.). But neither
can it be assigned with certainty to the Bellovaci and it seems to ;

me safer to follow the indications of the dioceses.^

L. Maziere ^^ believes that the Bellovaci also possessed the Noyon-
nais. The district in question is generally assigned to the Viromandui

^ See Nipperdey's Caesar, p. 79. It must be remembered that Nipperdey was

prejudiced against the )3 MSS.
- Notice de Vancienne Gaule, p. 148.
^ Geogr. des Gaules, i, 429. * Did. arch, de la Ganle, \, 140.

^ Ptolemy, Geogr., ii, Pliny, Nat. Hist., iv, 17 (31), § 106.

9, § 6 ;

« J5. (?., ii, 4, § 6. ' Ih., 13, § 1.

® The district in question forms part of the diocese of Rouen. See the map
in Gallia Christiana, t. xi.
^ Cf. C. Jullian, Hist, de la Gaule, ii, 482, and V. Leblond's paper in Congres

archeol. de France, 1905 (1906), pp. 318-26.

^" Comptes
rendus et mem. du comite arch, de Noyon, iii, 18G8, pp. 34-9.
398 BELLOVACI
and, in support of this opinion, it is pointed out that in 531 the see
of the bishopric of Vermand was transferred to Noyon. Maziere,
ho.wever, infers from Caesar's statement that the Bellovaci numbered
100,000 fighting men,i that their entire population must have
amounted to at least 400,000 and he argues that, without the
;

Noyonnais, their territory would have been much too small to sup-
port such a multitude. T admit that, if it was confined to the diocese
of Beauvais, it would have been disproportionately small but, if :

we are to disregard the law of the dioceses, we rnust have stronger

evidence and, as I have shown on page 242, the number, 100,000,
;

is probably exaggerated. See Veltocasses.

—
Bibraete. In the first edition of this book,^ it was demonstrated
that Bibraete was identical, not, as was formerly supposed, with
—
Augustodunum the Gallo -Roman town which stood upon the site
of Autun —
but with the great manufacturing town on Mont Beuvray,
12 miles west of Autun, which preserves the ancient name, and whose
defences, houses, and workshops have been revealed by excavation.
Although the old error was repeated by well-known writers up to
the close of the last century and was not corrected even by Mommsen
in his latest edition, I need not reprint my original article for the ;

claims of Mont Beuvray, unlike those of Alise-Sainte-Reine (Alesia),

are no longer disputed. Bibraete was abandoned early in the imperial
era and succeeded by Augustodunum, which was built upon virgin
soil,and, as Freeman said, was a new city on a new site, deliberately
'

^
laid out from the beginning on a great scale.'
—
Bibrax. Bibrax was 8 Roman miles from the camp which Caesar
made in 57 b. c, immediately after crossing the Aisne, and was
situated on or near the road by which the Belgae advanced against
him.4 Many places have been proposed for the site ^ but, as Caesar
;

crossed the Aisne either at Berry-au-Bac or at Pontavert,^ there are

only two which answer sufficiently to his description to call for dis-
cussion. They are Beaurieux and Vieux-Laon.'^
1. Von Goler ^ pronounced for Beaurieux, which is about 8 Roman
miles west of the hill of Mauchamp, where, if Caesar crossed at
Berry-au-Bac, he encamped, and about 2 miles north of the Aisne.
General Creuly ^ objects that Beaurieux was so situated that it could

1
B.O., ii, 4, § 5.
2
pp 387-94.
3 Hist. Essays, 4th ser., pp. 103-5. Cf. Mem. de la Soc. eduenne, i, 1872,
pp. 311-7, 356, 367, and Sitzungsberichte d. Konigl. preuss. Akad. d. WisseJi-
schajten, 1897, pp. 1104, n. 1, 1113, n. 5. ^ B. G., ii,
6, § 1.
5 See Bull, de la Soc. acad. de Laon, xix, 1869-70,
pp. 265-76.
« See pp. 659-66.
' See Caisjnart de Saulcy, Les campagnes de Jules Cesar dans les Gaules, 1862,
p. 110. D'Anville indeed, who believed that Caesar crossed the Aisne at
Pontavert, decided for Bievre, qui conserve evidemment le nom de Bibrax
'

{Notice de Vancienne Gaule, pp. 159-60). But the hill of Bievre is scarped
on every side, except the north, which would have prevented the Belgae from
attacking it and it is almost 16 kilometres, or about 11 Roman miles, from
;

the Aisne. Therefore, if it was the site of Bibrax, Caesar's camp must have
been on the heights of Craonne and this, as I show on p. 666, is impossible.
;

See de Saulcy, p. 120, and von Goler, Gall. Krieg, 1880, p. 67, n. 1 and jd. 272.
8 Gall. Krieg, p. 67- ^ Bev. arch., nouv. ser., vii, 1863,
pp. 299-300.

i
 — —

BIBRAX 399

have been entirely surrounded by the Belgae and that the reUeving
;

force sent by Caesar would therefore have been useless. Von G5ler
assumes, however, that the heights on which the town stands were
then washed by the Aisne, and he argues that this assumption ex-
plains the fact that Caesar's light-armed troops were able to relieve
Bibrax (by crossing the river) although it was blockaded by the
Belgae. M. Jullian,i who accepts von Goler's choice, but of course
rejects his hypothesis about the Aisne, maintains that in the night
Caesar's troops could have made their way into the town but does ;

not this argument ascribe to the Belgae a degree of carelessness

which is hardly credible ? Moreover, Beaurieux was not on the road
by which the Belgae marched against Caesar, unless they came from
Soissons and, as I show on pages 658-9, this is most improbable. It
;

need hardly be added that the identification of Bibrax with Beaurieux

depends on the assumption that Caesar crossed the Aisne at Berry-
au-Bac.2
2. Napoleon ^ adopted the plateau of Vieux-Laon, which Thillois ^
and de Saulcy ^ had selected before him. A little to the south of this
plateau and dominated by it, is the village of Berrieux, or, as it was
once called, Bebrieux.^ It has been conjectured that in the fifth
century a town which stood upon the plateau was abandoned by its
inhabitants, who took up their abode upon the hill of Laon that ;

they in turn built a settlement upon the site of Berrieux and that ;

they gave to this settlement the name of Bibrax, their abandoned

home.'' Anyhow, it is certain that the hill of Laon was once called
Bibrax ^ Laon is much too far from the Aisne to have been the
:

Bibrax of the Commentaries it is therefore probable that the name

:

Bihrax was given to it because the inhabitants of the Bibrax of the

Commentaries had migrated to it this probability is increased by
:

the fact that the entrenchments of St. Thomas, which stood on the
hill of Vieux Laon, were called in the charter of the abbey of St.
Vincent, in 1213, Vetus Landunum ^ and it is a reasonable con-
;

jecture that the Bibrax of the Commentaries stood on or close to the

hill which is now called Vieux Laon. Napoleon ^^ observes that, on
its southern side, it was unassailable by the Gallic method of assault ^^ ;

^ Hist, de la Gaule, iii, 253, n. 2, 254, n. 2. 2 g^e

pp. 659-66.
^ Hist, de Jules Cesar, ii, 101.
* Bull, de la Soc. acad. de Laon, xix,
1869-70, pp. 263-76
^ Les campagnes de Jules Cesar dans les
Gaules, p. 1 10.
« Bull, de la Soc. acad. de Laon, xix, 1869-70,
pp. 273-4. ' lb.

^ See Acta Sanctorum, June 20, Landunum montem qui antiquo nomine
Bibrax nuncupabatur. M. Wauters {Bull, de V Acad. Roy. de Bruxelles, "^^ ser.,
i, 1881, p. 367, with which cf. C. Jullian, Hist, de la Gaule, iii, 253, n, 2) does not

beheve that Laon was ever called Bibrax, because he considers it unlikely that
one place should have had two Gallic names Bibrax and Landunum at the —
same time. But it is not certain that the names Bibrax and Landunum were
applied to Laon at the same time ; and M. Wauters might have remembered
that the Saone was called by two Gallic names, Sauconna and Arar (Ammianus
Marcellinus, xv, ii, § 17).
* A. Piette, Itin. gallo-rom. dans
le dcpartemen de V Aisne, 1856-62, pp. 263-4.
^» Hist, de
Jules Cesar, ii, 101, n. 1. " B. G., i 6, §§ 2-3.
,
400 BIBRAX
and he argues that the Belgae, with the carelessness of a half bar-
barous people, would have neglected to invest it on that side. But,
objects Long,i Napoleon assumes that Caesar's light troops entered
'

Bibrax, which is not distinctly stated by Caesar '. This is true ;

but, if the troops did not actually enter Bibrax, it is probable that
they were prepared to do so.^
—
Bigerriones. The Bigerriones, whose name is preserved in Bi-
gorre, inhabited the diocese of Tarbes, or, roughly speaking, the
department of the Hautes-Pyrenees.^
—
Bituriges. The Bituriges Cubi occupied the diocese of Bourges,
which included the departments of the Cher and Indre and the
north-western part of the department of the Allier.^
The Bituriges Vivisci, who are not mentioned by Caesar, occupied
the diocese of Bordeaux, that is to say, the greater part of the depart-
ment of the Gironde.^ See Santoni.
Boi. —
See GoRaoBiNA.
—
Brannovices. See Aulerci.
—
Bratuspantium. Bratuspantium cannot be identified with cer-
tainty. Caesar mentions the town once only ^ and no other ancient ;

writer mentions it at all. Caesar merely says that it was a strong-

hold of the Bellovaci, who surrendered it to him in 57 B.C., when he
was marching from Noviodunum (Suessionum) into the country of
the Ambiani, whose capital was Samarobriva, or Amiens. Novio-
dunum (q.v.) was certainly close to Soissons, and was most probably
on the hill of Pommiers, 4 kilometres, or about 2| miles, north-west
of that town.
Napoleon,'^ following the Abbe
Devic,^ adopts a site, close to
Breteuil, which, he remarks, is just 25 miles —
the distance, he says,
—
mentioned by the Commentaries from Amiens. What there is to be
said for this site is that it is on the Roman road from Soissons to
Amiens that the ground is suitable for a stronghold and that in
; ;

the neighbouring valley of Vandeuil a few Gallic and numerous Gallo-

Roman antiquities have been found.^ But when Napoleon said that
25 miles was the distance from Bratuspantium to Samarobriva,
'
mentioned by the Commentaries,'' he must have been dreaming for ;

the Commentaries do not mention the distance at all. Walckenaer ^^

considers it unlikely that the chief oppidu7n of the Bellovaci should
have been situated, like Breteuil, close to their frontier, instead of in
the heart of their territory but this objection is not conclusive.
;

It used to be asserted that Breteuil itself, or the neighbourhood of

Breteuil, had, from time immemorial, been called Bratuspance ^^ :

'
Decline of the Roman Republic, iv, 48, n. 2.
2 See p. 241, n. 6. ^ Did. arch, de la Gaule, i, 160.

D'Anville, Notice de Vancienne Gaule, p, 1G2.

"

5 lb.,
p. 163, and Diet. arch, de la Gaule, i, 165. « B. G., ii, 13, § 2.

' Hist, de Jules Cesar, ii, 106, n. 1.

^ Dissertation sur Bratuspantium, 1843; Etude sur les Comm. de Cesar,
. . .

1865, pp. 20—57.

^ Diet. arch, de la Gaule, i, 193. ^^ Geogr. des Gaules, i, 42, 7-8.

^^ L.
d'Allonville, Dissertation sur les camps romains du dep* de la Somme,
1828, p. 155.
 BRATUSPANTIUM 40i

lut that argument has long been abandoned, even by the advocates
f From time immemorial,' in questions of this kind,
Breteuil. '

enerally means from the time when some forgotten antiquary

tar ted a theory which grew into a pseudo -tradition and, as Walcke- ;

aer observes, if Breteuil was ever called Bratuspance, it probably

cgan to be so called in 1574, when a certain cure of Breteuil reported
the prince of Conde in favour of the probable identity of Breteuil
liih Bratuspantium.
Another conjecture is that of Perrot d'Ablancourt, who placed
2.
>ratuspantium at Gratepanse, a hamlet about 5 miles cast of Mont-
idier.i Gratepanse must not be confounded with Gratepanche in
tie diocese of Amiens, with which Bratuspantium has also been

bsurdly identified. ^ The hamlet has disappeared, but was still

banding at the beginning of the last century. It has been said that
etween the words Bratuspantium and Gratepanse there is a ' '

?.semblance that the hill upon which Gratepanse stood was suitable
;

3r a Gallic stronghold that various roads met at the hamlet, from

;

T^hich it might be inferred that the place had once been an important
entre that in 1687 the remains of massive walls were discovered
;

bere and that since then Roman coins have been found.^ But it
;

5 needless to say that all these facts taken together do not amount
anything like proof. Besides, the hill on which the hamlet stood
; only about 600 metres long and between 400 and 500 broad and ;

bis is too small to have been the site of an important oppidum like
bratuspantium.
3. De Valois* identified Bratuspantium with Beauvais, on the
ite of which stood Caesaromagus, the Gallo-Roman capital of the
)ellovaci but it has been proved that no Gallic town existed there
;

>efore.^
4.M. de Lepinois conjectures that Bratuspantium was on the
ite Clermont but, as Dr. Leblond ^ observes, the area of this
of ;

ill is only 20 hectares, or about 49 acres, and hardly any antiquities

ave been found upon it.

5. Von Goler decides for Montdidier, which is about 12 miles, in
direct line, east of Breteuil and his son, endorsing his decision,
;

ays that Montdidier was the place selected for the camp of Crassus
a the winter of 54-53 b.c."^ But there is not a tittle of evidence that
Crassus was ever at Montdidier and even if he was, that is no reason
;

or inferring that Montdidier was the site of Bratuspantium. In

act there is nothing to be said for Montdidier, except that it was
ituated on or near the route which it may be conjectured that
Caesar took.^

V. de Beauvillc, Hist, de la ville de Montdidier, 1857 (2nd ed., 1875), i, 23-5.

^

Comptes rendus et rriem. du Comitd arch, de Noyon, 18G2, pp. 198-9.

'^

^ lb.,
p. 200 De Beauville, op. cit., i. 23-5. Dr. V. Leblond, however,
;

ays, je ne saclie
'
pas que d'importantes trouvailles gauloises ou gallo-
omaines y aient jamais eto faites' {L'oppidum Bratuspantium, &c., 1909, p. 10).
* Notitia Gain arum, ^. 113. ^ Diet. arch, de la Gaule,i, IdS.

« Op. cit., p. 24. 7 Gall. Krieg, 1880, p. 73, n. 2.

^
^ Comptes rendus et mem. du Comite arch, de Senlis, 2" scr., iv, 1878, p. 19.
1003 D d
402 BRATUSPANTIUM
6. Perhaps the most plausible conjecture is that of M. Seymour de
Ricci,^ who, accepting the identification of Noviodunum (Suessionum)
with Pommiers/'^ argues that just as Pommiers is close to Soissons,
(the Gallo-Poman capital of the Suessiones). so Bratuspantium was
close to Beauvais. The only admissible site, in his opinion, is Mont-
Cesar in the commune of Bailleul-sur-Therain, 11 kilometres, or
between 6 and 7 miles, ESE. of Beauvais. It may be added that
Augustonemetum (Clermont-Ferrand) and Augustodunum (Autun)
— the Gallo-Poman capitals of the Arverni and' the Aedui respec-
tively —were built near the old strongholds, Gergovia and Bibracte.
But does not Caesar's statement ^, that when he was encamping on
the outskirts of Bratuspantium the women and children stretched '

out their hands from the wall and begged the Romans for peace ',
. . ,

suggest that it 78 metres above the plain ? ^

was not situated on a hill
M. Jullian in despair proposes sous toutes reserves, quelque croupe
^ '

dans les environs immediats de Beauvais '.

The reader now knows all that can be known about the site of
—
Bratuspantium, that is nothing but he also knows why nothing :

can be known and that is something gained. There is perhaps

;

more to be said for the site near Breteuil than for any other site
which has been suggested but if Bratuspantium is to be marked
;

upon the map at that point, it should be with a note of interroga-

tion.
Cadurci. — The Cadurci were dependants of the Arverni ;^ and
their territory corresponded with a part at least of the diocese of
Cahors, in other words of the department of the Lot."^ Caesar men-
tions them three times ^ and in the last passage he writes Eleutetis
;

Cadurcis, or, according to tt, Heleutetis et Cadurcis. Napoleon,^

apparently assuming that Eleuteti [or, according to an inferior MS.,
Eleutheri] was the Celtic equivalent of iXivSepoi, conjectures that
Eleutheri Cadurci means the independent Cadurci
' that the ' ;

Eleutheri Cadurci occupied the northern part of the whole territory

of the Cadurci and that the southern part was, even in Caesar's
;

time, under the dominion of Rome. This conjecture is simply a bad

guess. If the southern part had been under Roman rule, it would
have formed a part of the Province and when Caesar was writing, ;

of the danger that was likely to threaten the Province if the Helvetii
were allowed to settle in the country of the Santoni,^^ he would have!
given additional point to his statement by saying that the countryj
of the Santoni was not far from the country of the Cadurci, insteac
of saying that it was not far from the country of the Tolosates.;
Long^^ suggests that Eleutheri, like Aulerci, was a generic name but ;

^ Rev. arch., 4" ser., ix, 1907, p. 403. Dr. Leblond {op. cit., pp. -20-32)
agrees with M. de Ricci, principally because le Mont-Cesar apparait un '

emplacement parfait comme camp d'observation et comme lieu de refuge.'

^ See pp. 465-6. » B. G., ii, 13.
§ 3.
* See Carte de VEtat-Major
(1 : 80,000), Sheet 32.
^ Hist,
de la Gaule, iii, 259, note. B. G., vii, 75, § 3. *^

^ Diet. arch, de la Gaule, i, 215. ^ B. G., vii, 4, § 6 64, § 6 ; 75, § 2. ;

» Hist, de Jules Cesar, ii, 24, n. 2. i»

B. G., i, 10, §§ 1-2. .
'I
Caesar, p. 393.
J
 CADURCI 403

if so, why was it not applied by any ancient writer to any other
people besides the Cadurci ? I believe that if Caesar wrote Eleuteti
or any such name, the Eleuteti were distinct from the Cadurci.
M. JulUan ^ thinks that Eleuteti perhaps denotes those of the Ruteni,
or of a part thereof, who were not subject to Rome but are not the ;

independent Ruteni mentioned by Caesar in the next sentence of

the same chapter ? Gliick,^ following F. A. Ukert,^ proposes to read
Helviis instead of Eleutetis ^ but Heller ^ rightly objects that the
;

Helvii, as far as we can judge, rendered no assistance to Vercingc-

torix, whereas the Eleuteti appear among the peoples who sent con-
tingents to his relief. The Helvii, indeed, just before the blockade of
Alesia, had taken an active part against him.^
—
Caerosi. The Caerosi, the Condrusi, the Paemani, and the Segni
are mentioned by no ancient writer except Caesar. He groups the
Condrusi with the Eburones, the Caerosi and the Paemani he im- :

plies that the Eburones and the Condrusi were conterminous and ;

he says that the Segni and Condrusi were between the Eburones
and the Treveri.'^ From these statements, supplemented by a com-
parison of the names of the several tribes with modern local names,
attempts have been made to determine their respective territories.
1. Sanson 8 is inclined to place the Caerosi either in the neigh-
bourhood of a village called Sire, not far from Liege, or in the
neighbourhood of Bouillon, near the river Chiers, which enters the
Meuse between Mouzon and Sedan. A. Wauters,^ arguing that their
powerful neighbours, the Treveri, would have kept the best parts
of the country for themselves, and driven them into the mountainous
regions of the Ardennes, assigns them a tract which was called in
the eighth century pagus Caros or Carascus, north of Treves, on the
banks of the Prum and this view is adopted by the French Com-
;

mission. ^^
2. The Condrusi were placed by d'Anville ^^ in a district which, in
the ninth century, was called Condrustum and this identification is
;

generally accepted. Their name is still preserved in Condroz, a tract

which extends along the right bank of the Meuse, between Liege and
Dinant, Namur, and Huy.
Cluver^'^ insists that the rout of the Usipetes and Tencteri must
have taken place in the country of the Condrusi, because Caesar
says that the Usipetes and Tencteri had reached their territory and
that of the Eburones when he began his march against them he :

places the scene of the rout near Coblenz and accordingly he assigns
;

^
Hist, de la Gaule, iii, 51G, n. 2.
- Die bei C. J. Caesar vorkommenden keltischen Namen, 1857, pp. 111-2.
" Geogr. der Griechen und Romer, 1816-46, ii, 2, 265,
•
Meusel, in the re-issue of his school edition, also reads Helviis. Has he nor
overlooked B. G., vii, 65, § 2 ?
5 Philologus, xvii, 1861, p. 282. « B. G., vii, 64, § 6 65, ; § 2.
^ 76., ii, 10
4, § iv, 6, § 4 ; vi, 32, § 1.
;

^ Les Comm. de Cesar, 3rd ed., 1658, p. 25.

^ Bull, de V Acad. Roy. de Briixelles, xiii, 1862, p. 393.
'" Diet. arch, de la Gaule, i, 217. " Notice, &c., p. 240.
^^
Germania antiqua, 1631, p. 397.
D d2
404 CAEROSI
the Condrusi the country on the left bank of the Rhine from its con-
fluence with the Moselle northward as far as the mouth of the 8ieg.
This argument is worthless for Caesar does not imply that the rout
;

of the Usipetes and Tencteri took place in the country of the Con-
drusi. Besides, on Cluver's theory, their rout took place both in
the territory of the Eburones and in that of the Condrusi
3. The Paemani are placed, unanimously I believe, in the Pays de
Famenne, a district which appears to retain their name, and also
adjoins Cbndroz, the assumed territory of the Condrusi.^
4. The name of the Segni is believed to be preserved in that of
Sinei or Signi, a town in the county of Namur, near Condroz.2
Caleti. — The Caleti occupied that part of the diocese of Rouen
which did not belong to the Veliocasses and their territory included
;

the Pays de Caux (pagus Caletus), or the western and central portion
of the department of the Scine-Inferieure. Their precise limits, how-
ever, cannot be traced. D'Anville gives them, besides the arch-
deaconries of Grand Caux and Petit Caux, a part of the archdeaconry
of Rouen, because, he says, their chief town, Juliobona (Lillebonne),
was situated in the archdeaconry. But, according to the French
Commission, Lillebonne is in the archdeaconry of Grand Caux.
Longnon adds to the archdeaconries of Grand Caux and Petit Caux
the archdeaconry of Eu.^
—
Carnutes. The Carnutes possessed the dioceses of Chartres,
Orleans, and Blois, or the greater part of the departments of the
Eure-et-Loire, Loiret, and Loire-et-Cher. The diocese of Blois was
formerly incorporated in that of Orleans.'*
—
Caturiges. Caesar mentions the Caturiges once only. He says
that they, as well as the Ceutrones and the GraioceH, occupied the
heights when he was crossing the Alps in 58 B.C., on his way from
Cisalpine to Transalpine Gaul, and that they attacked him. The
route which he followed led over Mont Genevre.^
Ptolemy ^ mentions only one of the towns of the Caturiges,
Ebrodunum, or Embrun, which, according to Strabo,' was on their
western frontier. Another, Caturigae or Caturigomagus, which
is mentioned in the itineraries, has been identified with Chorges.
If it belonged to the Caturiges in Caesar's time, their western
frontier, as a glance at the map will show, was almost certainly
west of Ebrodunum. Walckenaer,^ indeed, believes that their
territory extended westward of Vapincum, or Gap, to a place
called Fines, a name which, in Gallo -Roman geography, always
marked a boundary. This place is identified by the measurements
in the Itinerary of Jerusalem with Blaynie-Sept-Fons. The French

1 D'Anville, pp. 511-2 Walckeiiacr, i, 508.

;

•^
See D'Anville, p. 591, and generally, on the position of all four tribes,
K. Miillenhoff's Deutsche Altertumskunde, ii, 196-7.
3 D'Anville,
pp. 192-3; Walckenaer, i, 434-5 ; Did. arch, dela Gaule, 1, 219
A. Longnon, Atlas hist, de la France, p. 5. Eu is just in the territory which
d'Anville's map assigns to the Caleti.
* Diet. arch, de la Gaule, i, 232.

' B. G., i, 10,

§§ 3-4. See pp. 615-0. « Geogr., iii, 1, § 35.

' lb., iv,

1, § 3. * Geogr. des Gaules, i, 541.
 ;

CATURIGES 405

Commission, however, do not believe that Vapincum belonged to the

Caturiges. Remarking that it has never belonged to the same
ecclesiastical province as Embrmi, they prefer to assign it to the
Vocontii (q.v.) or one of the small neighbouring tribes which they
regard as their dependants. Fines, on their theory, marked the
frontier, not of the Caturiges, but of the small tribe to which, they
assume, Vapincum originally belonged. Besides, they argue, there
is no ground for supposing that the diocese of Gap was split into two
parts, one of which, on the west of Fines, belonged to the Vocontii,
and the other, on the east, to the Caturiges yet, if Fines marked
;

the western boundary of the Caturiges, this assumption is necessary.

Vapincum is not mentioned by any authority earlier than the
'
itineraries of Antonine and Jerusalem but Desjardins ^ holds with
:

good reason that it belonged to a small tribe called the Avantici

for, as he observes, the name of this people is preserved in Avangon
—
and St. ^^tienne d'Avangon, the names of two communes in the
neighbourhood and accordingly he believes that Fines marked the
;

western boundary of the Avantici. But the question remains whether

the Avantici were a pagus or a dependent tribe of the Caturiges or
of the Vocontii.
D'Anville ^ believes that Brigantio, or Briangon, which Ptolemy ^
assigns to the Segusiani, really belonged to the Caturiges, as it lay
outside the natural limits of the Segusiani, whose other town was
Segusio (Susa) but Walckenaer maintains that Ptolemy was right,
;

because, in his opinion, Ptolemy used, in preparing his book, a

description of Italy written before the new divisions were made by
Augustus. This is not a convincing argument and whoever reads
;

Caesar's notice of the Caturiges side by side with a good map ^ will
agree with me that Brigantio must have belonged to that tribe.^
See Ceutrones, Graioceli, and Vocontii.
—
Cenabum. Where was Cenabum ? This is one of the most vexed
questions of Gallic geography and it has been a vexed question for
;

centuries. Napoleon identifiesCenabum, or, as he calls it, Gena-

bum,6 with Gien. Mr. Froude, it is needless to say, follows Napoleon :

so do Colonel Dodge, Captain G. Veith,"^ and Dr. Sieglerschmidt ^ ;

and so do some of the modern English editors. But I suspect

that they all follow blindly ;and they certainly follow a blind
guide. Most of them seem to fancy that Napoleon has made a nev/

^ Geogr, de la Oaule rom., ii, 228.

^ Notice de Vancienne Gaule, p. 174. ' Geogr., iii, 1,
§ 36.
•*
I recommend Sheet 60 of the Carte de France (1 200,000).
:

^ Perhaps Desjardins is right in affirming that, in Caesar's time, 'ils s'eten-

daient aussi sur le versant oriental des Alpes, dans la vallee superieiire de la
Duria' {Geogr. de la Gaule rom., i, 84-5).
" The right form is Cenabum. See p. 841.
' The young Austrian officer, whose
knowledge of the literature of his sub-
ject is inadequate, makes the remarkable statement {GescJi. d. Feldzllge C. J.
Caesars, p. 509) that the Gallo-Roman town which stood upon the site of
Orleans was called, not Genabum (sic) simply, but Genabum Aurelianum.
^ Rev. arch., 4" ser., vi,
1905, p. 258, n. 1. Dr. Sieglerschmidt is a professor
in the military school of Gross-Lichterfelde.
406 CENABUM
discovery whereas the truth is that many antiquaries had argued
;

for Gien, or rather Gien-le-Vieux, which is about two kilometres, or

a mile and a quarter, north-west of Gien, before Napoleon, and had
argued much more forcibly than he. The tradition which identified
Cenabum with Orleans appears to have been unbroken until, in the
sixteenth century, some citizen of Gien endeavoured to make out
a claim on behalf of his own town. Early in the eighteenth century,
the famous Abbe Lebeuf attempted to sustain this claim but, after ;

Lancelot and d'Anville had refuted his arguments, he virtually recanted

his error. The authority of d'Anville was practically unquestioned for
more than a century. But while Napoleon was collecting the materials
for his work, MM. Brean and Petit renewed the battle on behalf of
Gien, or rather Gien-le-Vieux. Their arguments were criticized with
relentless acumen by a commission, appointed by the SociHe arcMo-
logique de VOrUanais to consider the question but meanwhile the
:

Emperor's book appeared and his verdict was' given for Gien.
;

Besides Mr. Froude and the other writers to whom I have alluded,
Desjardins, StofEel, Alfred Holder ^, and Meusel - scholars whose —
—
opinion has weight agree with Napoleon. The literature of the
subject is very bulky. I have worked through it all, totally in-
different to the result to which my researches might lead me and ;

I undertake to prove to demonstration that Cenabum stood, not

upon the site of Gien, nor upon the site of Gien-le-Vieux, but upon
the site of Orleans.
I. First of all, let us hear Caesar. He says that Cenabum belonged
to the Carnutes, and that a number of Eoman traders had settled
there. The town was on the right bank of the Loire, and about 160
Roman miles from some point in the territory of the Arverni. It
was seized by Carnutian rebels in the early part of 52 b. c. and they ;

massacred the traders. A few weeks later Caesar marched with

eight legions from Agedincum (Sens) for the country of the Boi,
intending to relieve their town, Gorgobina, which was besieged by
Vercingetorix. He left his heavy baggage at Sens, and directed the
Aedui and the Boi to keep him supplied with corn. On the day after
his departure from Sens he reached Vellaunodunum, a town belong-
ing to the Senones, and captured it on the third day following.
Thence he marched to Cenabum, and arrived there on the evening
of the following day but, as it was late, he was obliged to postpone
;

the siege. He implies that he marched from Vellaunodunum to

Cenabum as quickly as possible.^ Expecting that the inhabitants
would try to escape in the night by a bridge over the Loire, he kept
two legions under arms. He says distinctly that the bridge was in
actual contact with the town.* About midnight he learned from his

^ Alt-celtischer Sprachschatz, i, 979, 1997.

''
In his school edition, 1908, p. 252. ^ See p. 497.

* I assume, with all the best modern editors, except Nipperdey, that in tlie

passage oppidum Cenabum pons fluminis Ligeris contingehat, Caesar wrote con-
tingebat — —
the reading of the /3 MSS. not continebat, which is found in the
a MSS. Schneider (ii, 359) demonstrates that continebat is wrong. It has
been defended on the assumption that Cenabum consisted of two })arts. one on
 CENABUM 407

patrols that the inhabitants were trying to escape. He accordingly

fired the gates,and sent the two legions into the town and, as the ;

fugitives were checked by the narrowness of the streets and bridge,

they were nearly captured. After this Caesar crossed the bridge,
all
and made his way
into the country of the Bituriges.
Hirtius says that when Caesar marched in the winter of 52-51 B.C.
to punish the Carnutes for having attacked the Bituriges, he quartered
some of his troops at Cenabum in the houses of the Gauls from ;

which statement it is clear that some, at all events, of the houses

were still standing. Hirtius adds that, after this expedition, Caesar
left two legions at Cenabum, evidently to keep the Carnutes in awe.^
II. I will make one observation to clear the ground. It has been
proved, so conclusively as to win the assent of Napoleon himself,
that on the site of Orleans there was a Gallo-Boman, if not a Gallic
town, called Cenabum. The proof is furnished by the testimony of
ancient and mediaeval writers,'^ by the itineraries,^ by an inscription
which was discovered in 1846 in the Faubourg St. Vincent at Orleans,^

the right, the other on the left bank of the river, which the bridge joined {con-
tinebat). But Caesar's narrative shows that Cenabum was situated entirely
on the right bank. Vossius, who accepts the reading ccntirebat, exjAsiins that
the bridge linked the town to the opposite bank of the river [urbem adversae
'
'

fiuminis ripae adiurigit, aut quasi unum cum ilia continentem facit), an explana-
tion which, as Schneider shows, introduces ambiguity and obscurity into
Caesar's text. Anyhow, if the town had been only near the bridge and not
in actual contact with it, Caesar could have barred the inhabitants from all
access to the bridge. See von Goler's Gall. Krieg, 1880, p. 239, n. 5.
1 B. G., vii, 3 10, §§ 3-4
; 11; viii, 5, § 2
; ; 6, § 1.
^ Ptolemy, Geogr., ii,
8, § 10, with which cf. Eev. arch., nouv. ser., viii, 1863,
p. 392 ; Hist. Franc, scriptores, ed. F. Duchesne, iii, p. 5 a ; Hugo, Hist. eccL,
lib. V {Patrologiae cursus completus, ed. J. P. Migne, clxiii, col. 836 b) A. de
;

Valois, Notitia GaUiarum, p. 226. —

Ptolemy's estimate of the latitude and
longitude of Cenabum is very nearly right for Orleans, while for Gien it is very
far from the truth.
^ Itin. Ant., ed. Wesseling,
pp. 367-8 La Table de Peutinger, ed. Desjardins,
;

pp. 26, cols. 2-3; 33, col. 3; 34, col. 1. Cf. d'Anville, Eclair cissemens sur
Vancienne Gaule, p. 180 J. B. P. Jollois, Mem. sur les ant. diL depf- du
;

Loiret, 1836, p. 14 ;and Mem. de la Soc. arch, de VOrleanais, xi, 1868,

pp. 234-73, especially 260-5. The distances of Cenabum from Brivodurum
(Briare), Agedincum (Sens), Masava (Mesves), Lutecia (Paris), and Caesaro-
dunum (Tours) respectively are either absolutely or approximately right if
Cenabum was at Orleans, utterly wrong if it was at Gien.
J. Lebeuf {Eecueil de divers ecrits pour servir d' eclaircissemens a Vhist. de
France, ii, 1738, pp. 225-8) argues that as the Itinerary of Antonine, in the state
in which it has come down to us, dates from a period later than that of Con-
stantine, Cenabum ought, if it had been Orleans, to have been called in the
Itinerary 'Aureliani', —
the name by which Orleans was called after the
Emperor Aurelian. But, at the time when the Itinerary was compiled, most
of the towns of Gaul had two names, —the old Gallic name and the name
of the people in whose territory the town was situated ; and in the Itinerary
the old Gallic names were generally used. Thus we find Samarobriva, not
Ambiani, Agedincum, not Senones, Cularo, not Gratianopolis, &c. Varioas
other arguments of Lebeuf, which are not worth mention, are demolished by
d'Anville in his Eclair cissemens.
* Desjardins, Geogr. de la Gaule rem., ii, 477, n. 1. Cf. Mem. de V Acad, des
inscr., xxvi, 1867, pp. 119-36. The inscription is referred to the period of
Augustus.
408 CENABUM
and by the discovery of Gallic and Gallo -Roman coins.^ But while
admitting this, Napoleon, StofEel, Desjardins, and others deny that
the Cenabum which stood on the site of Orleans was the Cenabum
captured by Caesar.
III. 1. Napoleon bases his first argument on the hypothesis,
which he believes himself to have proved, that the site of Gorgobina
is St. Parize-le-Chatel, near the confluence of the Allicr and the Loire,
He cannot believe that Caesar, marching from Sens to relieve Gorgo-
bina, would have gone so far out of his way, even in order to attack
a rebellious town, as the site of the modern Orleans, the route by
which is nearly 60 miles longer than the route by Gien.^
Now it is not proved that Gorgobina was at St. Parize-le-Chatel ;

but wherever it was, if Caesar marched by way of Orleans, he cer-

tainly went out of his way in order to reach it.^ But there is no
reason why Caesar, even though his ultimate object was to relieve
Gorgobina, should not have gone some distance out of his way in
order to capture a town so important as Cenabum. Since I wrote
the foregoing sentence, I have found that d'Anville enforces the
same argument. S'il avoit une ville alliee,' he writes,
'
a secourir, '

il devoit avant tout venger la majeste du nom Romain viole par le

massacre de Genabum c'etoit I'unique moyen de conserver les

:

peuples qui restoient encore fidelcs.' * Besides, how can we tell

that, to the east of Orleans, any bridge spanned the Loire in that
part of its course which crossed Caesar's line of march ? Or, if there
were bridges, how can we tell that Vercingetorix had not destroyed
them ? 5

2. Secondly, says Napoleon, the distance from Cenabum to the

country of the Arverni was, according to Caesar, 160 Roman miles.
Assuming that Caesar is alluding to Gergovia, the capital of the
Arverni, Napoleon points out that this distance corresponds exactly
with the distance from Gergovia to Gien whereas the distance from
;

Gergovia to Orleans is nearly 40 miles more.

This argument has no weight for how can we be sure that Caesar's
;

estimate of the distance was strictly correct, or that when he spoke

of the distance from Cenabum to the territory of the Arverni, he
meant the distance from Cenabum to Gergovia ? Moreover, nearly '

40 is an exaggeration
'
the difference is about 25.
:

3. Thirdly, Napoleon says that, after crossing the Loire at Cena-

bum, Caesar found himself in the country of the Bituriges. This, he
remarks, is true if he crossed at Gien, but false if he crossed at
Orleans.
This argument is worthless. Caesar does not say that, after cross-
ing the Loire, he found himself in the country of the Bituriges. On
^ Mem.
de la Soc. arch, de VOrleanais, xv, 1876, pp. 115, 123-5, 133-47 ;

xviii, 1884, pp.172-5 A. Blancliet, Traite des monn. gauL, p. 492. In the
;

former edition I stated the proofs in detail ; but it is no longer necessary to

do this, as the identity of the Gallo-Roman Cenabum with OrleajUs is admitted
by those scholars who still deny its identity with tlie Cenabum of Caesar.
- Hist, de Jules Cesar,
ii, 247. n. 1 (and following pages),

^^ee pp. 425-G.

* Eclair cissemens sur Vancieiine Gaule,
p. 212. See B. G., vii, 34, § 3.
''
 —
CENABUM 409

the contrary, he impHes that he did not. He simply says that he '

threw his army across the Loire, and made his way into the country
of the Bituriges ' {exercitum Ligerim traducit atque in Biturigum
fines fervenit i). Napoleon does not know the meaning of the word
fervenit?-
4. Fourthly, argues Napoleon, the site of Orleans, not being a hill,
failsto answer the requirements of a Gallic op'pidum.
The answer is that even of the Gallic oppida which were fortresses,
some, such as Avaricum, were not built upon hills and that we ;

may gather from Caesar's narrative and from Strabo ^ that Cenabum
was rather a trading town than a stronghold, and was not strongly
placed. If it had been, would the Carnutes have run away from it
without making the slightest attempt to stand a siege ?
5. Finally, Napoleon gets over the difficulty presented by the
discovery of the inscription at Orleans, by assuming that, after the
destruction of Cenabum, its surviving inhabitants built a new town
on the site of Orleans, and called it by the name of the old one.
In reply to this conjecture, I have only to say that there is not
a particle of evidence for it.
6. If, says Brean,* Cenabum was at Orleans, Caesar, after leaving
Cenabum, must have had the infuriated Carnutes on his rear, the
Bituriges on his right, and Vercingetorix in his front. He could
not have recrossed the Loire, and he was too far from the Aedui to
get supplies. The least check, therefore, must have been disastrous
to him.
It is amazing that Brean does not see that every one of his argu-
ments recoils against himself. Let the reader look at his map. He
will then see that, if Cenabum was at Gien or at Gien-le-Vieux,
Caesar, after leaving Cenabum, would also have had the infuriated
Carnutes on his rear, the Bituriges on his right, and Vercingetorix in
his front. If he could not have recrossed the Loire in the one case,
neither could he have recrossed it in the other and as for supplies,
;

when he marched to Avaricum, he was necessarily at a considerable

distance from the Aedui. But of course the dangers which Brean

^ B. 0., vii, 11, § 9. Cf. Festschr. zu Otto Hirschfelds sechzigstem Geburtstage,

1903, p. 217.
^ I have not the doubt that every competent scholar will agree with
slightest
me that Caesar meant that, had crossed the Loire, he had some distance,
after he
great or small, to travel before he could reach the frontier of the Bituriges.
There is a passage in B. G., i, II, % I, which is ahnost exactly parallel with the
one which I am discussing. I print the two side by side :

Helvetii iam per angustias et fines exercitum Ligerhn traducit atque in

Sequanorum suas copias traduxerant Biturigum fines pervenit.
et in Haeduorum fines pervenerant.
Now Napoleon ought,if he were consistent, to hold that the moment after
the Helvetii stepped across the Sequanian frontier they were in Aeduan
territory. But, as a matter of fact, he holds rightly that, before they reached
Aeduan territory, they crossed tlie territory of the Transrhodane Allobroges
and that of the Ambarri.
^ Strabo (iv,
2, § 3) calls Cenabum the emporium of the Carnutes.
**
Itin. de F expedition de Cesar d' Agendicum a Gergovia-Boiorum, &c., 18G5,
pp. 3-4.
410 CENABUM
conjures up were no more than the risks which every invader must
face. The infuriated Carnutes had been soundly thrashed, and
' '

were wise enough to keep their fury bottled up the Bituriges, who,
:

wherever Cenabum may have been, were on Caesar's right, aye and
on every side of him, took good care to keep out of his way and ;

Vercingetorix, as soon as he ventured a battle, was beaten. Caesar

was actually, as he tells us} obliged to do without the supplies which
he expected from the Aedui yet he survived the deprivation.
:

7. One question remains. Would Caesar have been able to march

from Sens to Orleans in the time which he says that it took him to
march from Agedincum to Cenabum ? ^ The distance from Sens to
Orleans was, by the road mentioned in the Table ^ 88J Roman miles,
or 131 kilometres by the other Roman road, which, according to
:

Napoleon,* can never have been a Gallic road, 108 kilometres. The
time which he occupied on the march was four days. Admit, with
Napoleon, that he must have gone by the longer road. We are then
obliged to assume that he marched at the rate of 22 Roman miles
a day. This is a high rate of marching.^ But Caesar gives us to
understand that he was doing his best and it must be remembered
;

that the army was encumbered by very little baggage. There is,
however, no proof that the other Roman road was not made upon
the line of a Gallic road. It is incredible, says A. Challe,^ that there
should not have been a Gallic road running in a direct line from the
important town of Agedincum into the heart of the rich country of
the Carnutes, and following almost the same direction as the Roman
road which is still traceable between Sens and Beaune. Boutet de
Monvel, indeed, and other opponents of Orleans deny that Caesar
could have performed the journey, even by the shortest road, in
four days but it is beyond dispute that Caesar marched 25 Roman
:

miles down and 25 Roman miles up the valley of the AUier in about
28 hours and practical soldiers like General Creuly, Colonel Stoffel,
;
"^

and the Due d'Aumale consider that he could easily have marched
27 kilometres, or 18 Roman miles a day for several days at a stretch.
As a matter of fact, he marched from Corfinium to Brundisium, a
distance of 465 kilometres, in 17 days, or at the rate of more than
27 kilometres a day,^
Thus every objection that has been brought against the view
which identifies the Cenabum of the Gallo-Roman period with the
Cenabum of Caesar, falls to the ground. I proceed to examine the
arguments that have been devised to prove that the C^enabum of
Caesar stood on the site of Gien or of Gien-le-Vieux.
IV. 1. Five roads, we are told, meet at Gien-le-Vieux, namely
from Sens (Agedincum), Autun (Augustodunum), Chartres (Autricmn),

" Ih., vii, 10-11,

1
B. G., vii, 17, §§ 2-3.
^
La Table de Peutinger, ed. Desjardins, p. 2G, cols. 2-3,
*
Hist, de Jules Cesar, ii, 249, note. ' See
p. G3.5.
^
Bull, de la Soc. des sciences hist.
. de VYonne. xx, 18G6, p. 142.
. .

'
B.G., vii, 40-1.
*^
See p. 635. According to 0. E. Schmidt's itinerary [Briefuechfiel d.

M. T. Cicero, 1893, pp. 149-50, 379), the distance was 485 kilometres
 — *

CENABUM 411

Bourges (Avaricum), and Sancerre and the supporters of Gien-le-

;

Vieux labour to prove that these roads were Gallic.^ But even if
the so-called proof could be accepted, it would only show that there
was an important Gallic town at Gien-le-Vieux it would not show :

that that town belonged to the Carnutes, still less that it was Cenabum,
2. According to Brean,^ an old bridge spanned the Loire opposite
Gien-le-Vieux and he infers that this was the very bridge by which
;

'
Caesar crossed the Loire after capturing Cenabum. The proofs '

which he offers of the former existence of this bridge are, that various
inhabitants of Gien and the neighbourhood told him that they had
seen the ruined piles of the bridge in the bed of the river when the
water was low, and that he had himself discovered a massif of ' '

stone close to Port Gallier (a quay on the southern bank), which had
evidently formed a part of the bridge. Moreover, a number of
antiquities, of which Brean gives a list, were discovered at Gien-le-
Vieux and he asserts that among them were Gallic coins of a period
;

anterior to the Roman conquest. Finally, on the spot to which

Brean points as the site of Cenabum charred remains were discovered ;

and he maintains that they are a relic of the fire which Caesar kindled
in 52 B. c. !

Now a commission was appointed by the Societe archeologique de

rOrleanais to investigate the value of Brean's discoveries and the ;

report of the commission is embodied in two papers published in

volume ix (1866) of the Society's Memoirs, Rapport sur les com-
munications de M. Brean (pp. 234-52) by M. Marchand, and Question
de Genabum (pp. 253-90) by M. Collin.^ The certificates, written by
inhabitants of Gien, which Brean printed in support of his theory
regarding the alleged bridge, are demolished by the evidence of their
authors No less than 2,090 soundings, in which Brean himself took
!

part, were made in the bed of the Loire, with the object of discover-
ing the alleged remains of the bridge but all in vain.
;

Some years ago a coin of Pope Clement VIII was discovered

adhering to the masonry of the so-called bridge, the massif to — ' '

which Brean pins his faith. It is almost certain that the coin could
only have been left where it was found by the workmen who built
the structure. If so, the structure itself cannot be assigned to any
date earlier than the pontificate of Clement VIII.
It has also been affirmed by Marchand, representing the Com-
mission, that all the objects in Brean's cabinet that were discovered
at Gien-le-Vieux were Gallo-Roman, not Gallic ^ and, according ;

to the French Commission, no Gallic antiquities have been found

either at Gien or at Gien-le-Vieux.^ Nevertheless, I will assume that
a Celtic oppidum did really exist upon the site in question. Still, it
cannot be proved that that oppidum was Cenabum and it can be ;

^ A. Brean, Itineraire, &c., pp. 47-8 ; A. Petit, Dissertation sur Genahnm-

Gien, &c., 1863, pp. 74-5.
- Itineraire, pp. 33-44.
'
See also L. A. Marchand, Hist, de la ville . de Gien, pp. 6-7.
. .

* Mem. de la Soc. arch, de VOrleanais, ix,

1866, p. 290.
^ lb., p. 247. ^ Did. arch, de la Gaule, i, 446.
412 CENABUM
proved that it was as to the charred remains
not. First of all, :

the fact that coins of Tetricus (about A. d. 273) were found among
them proves that the fire which we are asked to believe was kindled
by Caesar took place not less than three centuries after Caesar's
death.i Secondly, the dimensions of the Gallo-Roman town, as far
as they can be ascertained by the results of the excavations, were far
too small for an oppidum of the importance of Cenabum.^ Now it is
proved by Caesar's words oppidum Cenahum pons fluminis Ligeris
—
contingebat that the town of Cenabum was in actual contact with
the bridge over the Loire. Accordingly the advocates of Gien-le-
Vieux are compelled to assume that the small settlement which the
excavations have revealed was only a minute fraction of the entire
Cenabum, which must have extended right down to the bank of
the Loire. But this imaginary Cenabum is as much too large as the
other is too small. La ville actuelle d'Orleans,' says Marchand,^
'

'
tournerait a I'aise dans I'enceinte assignee a Gien-le-Vieux.'
3. There are traces of a Roman camp in the wood des Marceaux ', '

near Gien-le-Vieux.
So says Brean.^ But, under the cold scrutiny of the commission,
the camp ' turns out to be of quite modern construction, and to
'

have been made for a very peaceful and even prosaic purpose. Ces '

dossees,' says Marchand, ont ete faites pour arreter les moutons
'

des Merceaux, et non les Carnutes de Gien-le-Vieux.' ^ Readers of

The Antiquary will remember the shrewd aside of Edie Ochiltree, as
he listened to the old pedant dilating on the construction of the
'
camp on the Kaim of Kinprunes
'
Praetorian here. Praetorian :
— '

there I mind the bigging o't.' ^

!

4. Gien contains a street called a la Genabye ', which leads, not '

.''
towards Orleans, but towards the higher part of the town itself
Now there is no evidence that the street in question was called by
this name before the seventeenth century and it is probable that, ;

as A. de Valois says,^ the name was given to it, from motives of

local patriotism, in consequence of the theory that Gien was identical
with Cenabum. Moreover, there is, or was in the time of Walcke-
naer^ a faubourg in Orleans which was also called by the name of
'
Genabie '.

5. Cenabum was not Gien-le-Vieux, because

Stoffel 1^ sees that
Cenabum must have touched the right bank of the Loire but he ;

maintains with Napoleon that Cenabum was le Gien actuel and '
' ;

these are his reasons. Vellaunodunum, he says, was at Toucy.

C^aesar marched from Vellaunodunum to Cenabum in two days.
^
Mem.
de la Soc. arch, de VOrUanais, ix, 1866, p. 247.
-
pp. 237-8.
lb., ^ Ih., p. 245. ^ Itineraire, &c.,
pp. 45-6.
^ Mem. de la Soc. arch, de VOrUanais, ix, 1866,
p. 241.
^ Abbotsford edition,
p. 28.
' Napoleon, Hist, de Jules Cesar, ii, 250, note, De Monvel {Mem. de la
Soc. d' agriculture d'Orleans, vii, 1863, p. 47) calls the street
. . . rue dc la '

Genabie '

* Notitia Galliarum, p. 226. Cf. Mem. de la Soc. arch, de VOrUanais, ix, 1866,
p. 168.
^ Geogr. des GauUs, i, 401. ^" Guerre de Cesar etd'Arioviste, pp. 149, 156.

I
 ;

CENABUM 413

'
D'apres cela, et etapes a trente kilometres
en evaluant toujours les
... on est conduit a placer Genabum
sur la Loire, a I'ouest et a
soixante kilometres de Toucy. Cette distance est exactement celle
de Toucy a Gien. ... II en resulte que Genabum correspond a la
ville actuelle de Gien.'
But, as I show elsewliere,^ there is no proof, there is no evidence
that Vellaunodunum was at Toucy.
That, reader, is the case for Gien and Gien-le-Vieux and thus, ;

at every point, it breaks down. I have proved that a Gallo-Roman

town called Cenabum, in the country of the Carnutes, was at Orleans.
I have refuted the theory that this Cenabum was different from the
Cenabum of Caesar. I have shown that there is no evidence for the
theory that the Cenabum of Caesar was at Gien or at Gien-le-Vieux.
I shall now prove, by independent arguments, that it was not at
either of those places.
V. 1. It is certain that, after the conquest, both Gien and Gien-
le-Vieux were situated not in the territory of the Carnutes, but in
that of the Senones, or possibly in that of the Aedui ^ and, in the ;

absence of any evidence to the contrary, it may be confidently

affirmed that this statement is equally applicable to the time of
Caesar. As this fact, if it is a fact, alone decides the whole question,
a desperate effort has been made to disprove it. We may assume,
say the advocates of Gien, that Caesar deprived the Carnutes of part
of their territory as a punishment, and transferred it to the Senones
and this assumption would explain the fact that Gien belonged to
the diocese of Auxerre.^
Yes, we may assume whatever we like but assumption is not
;

proof.
2. The argument from tradition against Gien is very strong.
Although we have Latin documents, which relate to Gien, ranging
from the sixth to the seventeenth century, Gien was never called
either Genahum or Cenabum in any of them, except once by Seguier,
bishop of Auxerre, w^ho, in 1634, headed a document with the words
Datum Gcnahi. This solitary exception proves no more than the
bishop's private opinion. To quote a modern antiquary, unc '

tradition qui sommeille durant onze siecles n'a plus le droit de se

reveiller.' ^ And, asks Marchand,^ if Gien was Cenabum, why did
not St. Aunaire and St. Tetrice, bishops of Auxerre, call it so instead
of Giemus ?
3. Hirtius says that in the winter of 52-51 B.C. Caesar, after
punishing the Carnutes for having attacked their neighbours the
Bituriges, left Trebonius at Cenabum with two legions to overawe
them.^ Now, if Cenabum was at Gien-le-Vieux, even assuming that
Gien-le-Vieux was in the country of the Carnutes, Trebonius was left
on the extreme eastern limit of the Carnutian territory. But from
such a position how would he have been able to keep the Carnutes
' See pp. 497-8. ''
See p. 472.
^ Caesar, iv, 204.
'See Acliaintre,
'
Mem. de la Soc. arch, de VOrleanais, ix, 18GC, pp. 101, 100.
« lb., p. 250. « B. G., viii, 0, § 1.
414 CENABUM
in check, or to prevent them from making another raid across the
frontier of the BiturigesSurely Caesar would have had the common
?

sense to assign him a more central position.

Every argument which tells against the identification of Cenabum
with Gien-le-Vieux tells equally against its identification with Gien.
Even if the narrative of Caesar did not make it clear that Cenabum
was not a hill-fort, it would be impossible to identify the hill of Gien
with the site of Cenabum ;for the hill is too small, and the very
name of Gien-le-Vieux tends to show that it is older than Gien.^
VI. Desjardins ^ proposes a compromise. Genahum, he says (not
quite correctly), is the reading of all the MSS. for the town which
Caesar captured whereas the town which stood upon the site of
;

Orleans was Cenabum. Desjardins agrees with Napoleon that

Orleans was too far from Agedincum to have been the place captured
by Caesar." He therefore suggests that the town plundered by the
Carnutes was Orleans, and that Genabum was Gien.
If this theory had been propounded by any one of less eminence
than the late French geographer, I should not notice it. I do not
believe that any one of Caesar's readers during the last 2,000 years,
except Desjardins, ever dreamed of doubting that the Genabum of
B. (r., vii, 3 was identical with the Genabum of chapter 11.^ Every
scholar knows that c and g were often interchanged. The town of
the Carnutes which Caesar captured was confessedly identical with
the town of the Carnutes in which he encamped in the following
year ; and Hirtius calls that town Cenabum, which, as I show on
page 841, is the right form. Desjardins would of course have main-
tained that Gien must have been in the country of the Carnutes.
Granted that it may have been, though the evidence is all the other
way. Still, as Desjardins maintains that, besides Cenabum, there
was a town called Genabum, on him lies the burden of proving that
this assumed place was at Gien. But he has not attempted to do
this. As I have shown, there is not a particle of evidence for the
theory that either Gien or Gien-le-Vieux was ever called Genabum :

and the argument which Desjardins bases upon the distance of

Orleans from Sens I have already refuted.
To conclude. There is no evidence at all that Cenabum or Gena-
— —
bum call it which you will stood upon the site either of Gien or
of Gien-le-Vieux and there is conclusive evidence that it did not.
;

That Cenabum stood where Orleans stands now is proved by evidence

strong enough to convince Lord Collins or Professor Huxley. It is
proved by the evidence of Caesar and Hirtius, by the evidence
of Ptolemy, by the evidence of the itineraries, by the evidence of
a tradition which is at least as old as the ninth century, by the
evidence of coins and other antiquities, by the evidence of the in-
scription of the Faubourg St. Vincent. The conclusion is not merely
—
probable, but certain, as certain as that Lutecia stood upon the
^ See Brean, Itineraire, &c., pp. 19-25.
- Geogr. de la Gaide rom., ii, 477, n. 1, 478, 480.
^ In both passages the reading of the MSS. is not Cenabum, but Genabum,

and (in vii, 3, § 1) Getiahim or Genabiti.

 CENABUM 415

island of Notre-Dame, and Alesia upon Mont Auxois. It is the con-

clusion to which the most sagacious students of the Commentaries
and the most eminent geographers and antiquaries in France, in
Germany, and in Great Britain have come. It is the conclusion to
which diligent study has led the Commission de la topographie des
Gaules, the Academie des Inscriptions, the Comits des travaux his-
toriques.^ If Napoleon came to a different conclusion, it was because
he allowed himself to be misled by writers whose local patriotism
was stronger than their judgement. It is a pity that Colonel Stoffel,
Mr. Froude, Desjardins, and other writers of less distinction, should
have helped to propagate the error which he revived.
Cenomani. See Aulerci Cenomani.
—
Ceutrones (a). The Alpine Ceutrones occupied the valley of
Tarentaise and the adjoining mountains.^
Ceutrones (6). See Nervii.
—
Cocosates. The Cocosates are mentioned by no ancient writer
except Caesar 2 and Pliny ,^ neither of whom gives the slightest
indication as to their geographical position. DAnville ^ conjectures
that Coequosa, for which he proposes to read Cocosa, a place men-
tioned in the Itinerary of Antonine, 16 Gallic leagues north of Dax,
on the road to Bordeaux, was their capital and the French Com-
;

mission,^ by the process of exclusion, arrive at a result substantially

the same. After marking on the map all the peoples whose terri-
tories are approximately known, they find remaining for the Cocosates
a strip of land in the department of Les Landes, between Castetz and
Mimizan.
Condrusi. See Caerosi.
Coriosolites. —
The neighbours of the Coriosolites were the Osismi
on the west, the Veneti on the south, and the Redones on the east.
They dwelt in the country round Corseul, which preserves their
name. D'Anville,"^ with whom Walckenaer ^ agrees, believes that
their territory extended westward as far as the neighbourhood of
St. Brieuc, where, he maintains, a place called Finiac marks an
ancient frontier as evidently as Fins (Fines) in other parts of France.
But the so-called Finiac is really Ifhniac and the name has no
' '
;

etymological connexion with Fines. The French Commission ^ are

inclined to believe that the diocese of Treguier, or a part of it, as
well as the dioceses of Aleth and Dol, belonged to the Coriosolites.
With regard to the diocese of Aleth (which is identical with that of
St. Malo) they have no doubt. The position of Aletum, or St. Servan,
which commands the mouth of the Ranee on its right bank, was,
they argue, so important that the Coriosolites would not have allowed
it to pass out of their hands. M. Kerviler, on the other hand, insists
that the Ranee, being a natural frontier, must have separated them

'
See Bull, de la Soc. arch, et hist, de VOrleayiais, ix, 1887, p. 81,
" D'Anville, p. 221 Desjardins, Geo(jr. de la Gaule rom., i, 78.
;

^ B. G., iii, 27, § 2. * Nat. Hid., iv, 19 (33), § 108.

'•
Notice de Vancienne Gaule, p. 229. "*
Diet. arch, de la Gaule, i, 295.
^ Notice de Vancienne Gaule, p. 259. **
Geoijr. den Gaulen, i, 381-2.
* Did. arch, de la Gaule, i, 332.
416 CORIOSOLITES
from the Redones.^ Certainly, the possession of Aletum would have
been just as essential to the Redones as to the Ooriosolites and as ;

the Redones were a maritime peoplej^ and in any case only possessed
a small seaboard, it seems likely that their territory extended west-
ward as far as the Ranee. If so, the Ooriosolites did not occupy the
diocese of Dol, or even the whole of that of Aleth. But it is impos-
sible to define their frontiers with any approach to certainty. The
geography of this part of Gaul, as d'Anville^ observes, is most
obscure. The number of dioceses greatly exceeds the number of
states those who, like the French Commission, are guided by the
:
*

indications of the dioceses, are obliged to admit that it is impossible

to say with certainty whether the whole or only a part of the diocese
of Treguier belonged to the Coriosolites the diocese of Rennes does
:

not touch the sea, and yet Caesar says that the Redones were a mari-
time people and finally M. J. Loth ^ has proved that of the dioceses
;

in the Aremorican peninsula only those of Vannes, Nantes, and

Rennes were Gallo-Roman that all the rest were created by the
;

invaders from Britain and that when those invaders formed their
;

dioceses, they took no account of the existing political divisions. On

the south, it seems probable that the territory of the Coriosolites was
limited by the natural barrier of the Montagues Noires, and the
French Commission ^ hold that it extended eastward as far as Feins,
which they identify with the Fines of the Redones
"^
but all that
;

we can say with certainty is that it corresponded more or less closely

with the department of the Cotes-du-Nord.
II. Ptolemy ^ mentions a people called the Arvii ('Apovioi)- Des-
jardins ^ believes that they were identical with the Coriosolites,
whom Ptolemy does not name. In the environs of Corseul there
were discovered in 1707 the remains of a temple, which is locally
known, or was known as recently as 1849, as the Temple de Mars.^^
In the Table^^ is mentioned a place called Fanum Martis, the distance
of which from Condate (Rennes) appears to identify its site with that
of Corseul. Desjardins infers that the temple is that from which
Fanum Martis took its name. He argues that 'Apoviot is only
another form of "Apjtot (of or belonging to "Apvys [Mars]) and that ;

'ApovLOL ought to be translated, not by Arvii, but, as it was trans-

lated in the fourth century, by MartensesP- This reasoning is
^ Bull. arch, de VAss"' breionne, 3*^ ser., iv, 1885, pp. 225-8. See, however,
351, 353. ^ B. G., ii, 34. * Notice de Vancienne Gaule, p. 508.
I)p.
* See Rev. arch., nouv. ser., ix, 1804, p. 325.

^ UEmigration hretonne en Armorique, 1883, pp. xxi, 50-1, 75-82.

" Diet. arch, de la Gaule, i, 399.

' I tin. Ant., p. 387. « Geogr., ii, 8,

§ 7.
" Geogr. de la Gaule rom., i, 322-4 ; Rev. arch., 1849, p. 228.
1" A. de la Borderie, Hist, de Bretagne, i, M. Burgault {Bull, de
1896, p. 114.
la Soc. polymath, du Morhihan, 1875, p. 72) denies that there is any proof that
the debris exhumed in the neighbourhood of Corseul were those of a temple ;
and he adds triumphantly that, according to the Itinerary of Anfomne (p. 387),
Fanum Martis was in Gallia Belgica. A Fanum Martis certainly ! But not
the Fanum Martis of the Table. Were there not many Noviodunums in Gaul '/

" La Table de Peutinger, ed. Desjardins, p. 28, col. 2.

'2 Notitia dignilatum,
ed. O. Seeck, p. 205 (xxxvii, 19).
 CORIOSOLITES 417

accepted by M. Loiigiioii.^ To my mind it is unsatisfactory. Ptolemy

places the Arvii between the Diablintes (q.v.) and the Veliocasses
(q.v.). If he was right, it is evident that they dwelt far east of
Corseul and d'Anville ^ conjectures that they inhabited the valley
;

of the Arve or Erve, a tributary of the Sarthe.

On the other hand, the Commission disbelieve in the existence of '^

the 'A/oomoi, (1) because no ancient writer, except Ptolemy, men-

tions them (2) because the name Arve, which has been supposed
;

to be derived from their name, is common to numerous rivers and ;

(3) because the word 'Apomot is probably only another form of

'Ecro-oviot (the Esuvii of Caesar). None of these reasons, except the
second, which I am unable to verify,* appears to me to have any
weight. Several Gallic tribes, whose existence has never been dis-
puted, are mentioned by only one ancient writer and it is in- ;

credible that Ptolemy should have invented the name. The third
reason is simply a bad guess.
III. According to M. Longnon,^ the Coriosolites disappeared as
an independent people, before the publication of the Notitia pro-
vmciarum, that is to say before the fifth century, and were absorbed
by the Diablintes. Although I am not concerned directly with the
state of Gaul after the time of Caesar, it is my business to examine
the theory of M. Longnon, because he makes use of his conclusion to
argue that the Diablintes, and not the Coriosolites, occupied the
diocese of Aleth.
Certain MSS. of the Notitia have the form Coriosolitum, others
Corisopitum ; ^ and M. Longnon maintains that the latter is the
"^

true reading, because it is found in a MS. of the sixth century, which

is probably to be assigned to the year 570. Therefore, he concludes,
the Notitia does not mention the Coriosolites at all.
C^an the reading Corisopitum be accounted for ? A. de Courson^
conjectures that emigrants from Corstopitum^ (for which some MSS.
have Corisopito), or Corchester, on the Tyne, founded Corisopitum,
or Quimper, in Brittany. 'It is unUkely, says A. de la Borderie,^o who
agrees with this conclusion, that the Coriosolites existed, as an
independent people, after the immigration from Britain in 514. Ac-
cordingly he pictures the copyist who transcribed the MS. of the
Notitia mentioned by M. Longnon, saying to himself, as he read his
original, " Civitas Co-ri-o-so-li-tum, qu'est-ce que cela pent etre ?
'

—
Je n'ai jamais ou'i ce nom-la." Puis tout-a-coup il se frappe le front :

" Ah j'y suis, c'est en Bretagne, c'est un eveche breton, seulement

!

'
ici ils se sont trompes d'une lettre."

^ Atlas hist, de la France, p. 5. ^ Notice de Vancienne Gaule, p. 105.

^ Diet. arch, de la Gaule, i, 85.
* Only one '
Arve' is mentioned in Joanne's Diet, geogr, . . . de la France,
i, 1890, p. 177.
5 Geogr. de la Gaule au VI^ siecle, 1878, p. 316.
^ C&riosopolum, according to M. Loth {U Emigration hretonne, p. 57).
^ 31 em. du Congres scientijique de France, 38^ scss"., 1872, p. 397.
^ Bull, de la Soc. de geogr., 4^ scr., xx,
1860, pp. 264-5.
^ 8ec Itin. Ant., pp. 464-5, and cf. A. Holder, Alt-celtischer Sprachschatz,
ij 1136. 1" Geogr. gallo-rom. de
V Armorique, pj). 27-8.
1093 E e
418 COmOSOLITES
Ue explanation may conceivably be right
la Borderie's but of :

course a mere guess ^ and there is no proof that Corisopitum

it is ;

was founded in the fifth century by British innnigrants. The bishop

of Quimper was, it is true, called episcopus Corisopitensis in the
eleventh century but the earliest previous mention of the name
;

Corisopitum, as applied to Quimper, was in the ninth century. The

Breton name of this place in the fifth and the four following centuries
was Kemper .2 But it is unnecessary to have recourse to conjecture
in order to refute M. Longnon for, save the very doubtful evidence
;

of the Notitia, there is absolutely no evidence that there were any

Corisopites in Gaul either in Caesar's time or at the time when the
Notitia first appeared. The natural conclusion is that the reading
Corisopitum (or rather, according to M. Loth, Coriosopotmn) in one
of the MSS. of the Notitia is a copyist's blunder.^ Ne lit-on j^as '

aussi,' asks M. Loth,^ immediatement au-dessus Namtniium pour

'

Namnetum, dans ce manuscrit 12,097, qui semble u'n livre sacre pour
'
certains erudits ?

—
Diablintes. Caesar ^ mentions the Diablintes once only, among
the allies whose services the Veneti enlisted in 56 B.C. Ptolemy ^
describes them thus iv 8e rfj /xecroyeta Twv jjilv Ov€V€Tiov elaiv avaro-
'.

XlK(j)T€.pOi AvXi fJKLOL 01 Ata/^XtVat, (Lv TToAt? ISoLoSoVl'OV. N0I080ITOV

is generally identified with the Nu-Dionnum of the Tabled which was
situated on the road from Le Mans to Bayeux.
The foregoing evidence supports d'Anville's ^ theory that the
Diablintes occupied the neighbourhood of Jublains,^ near Mayenne,
where the remains of a Gallo-Roman town have been discovered.
Nor is this all. Long remarks that a document of the seventh cen-
tury speaks of condita Diablintica '
as situated
'
in pago Ceno- '

mannico ' thus, says Long, ' we obtain ... an explanation of the
;

fact of the name Aulerci being given in Ptolemy both to the Dia-
blintes and Cenomani.' Another document,' he observes, of the
' '

seventh century speaks of " oppidum Diablintes juxta ripam Araenae

fluvioli " and the Arena is recognized as the Aron, a branch of the
;

Mayenne.' lo

The principal opponent of the received view is M. A. Longnon.^^

He admits that the ancient name of Jublains was Diablintes but ;

he will not allow that Diablintes was the chief town of the civitas
Diahlintum. It was simply, in his opinion, a colony founded by the
Diablintes. He cites a case which he regards as parallel. Exmes,

^ See Professor Haverfield' s remarks in Proc. Brit. Acad., ii, 215, u. 1.

^ See E. Halleguen, La Cornouaille et Corisopitum, 1861, pp. 20, 26, 30, and
J. Loth, U
Emigratio7i bretonne, p. 59.
^ The MSS. of the Notitia offer many variants, e.
g. Consuliium, ConisoUlum,
Corosopitum, Coiisolitum, Corisolitum, CoriosoUtiun, and Corisuletum.
* U
Emigration bretonne, p. 57.
5 B. G., iii, 9, § 10. « Geogr., ii, 8, § 7.
' La Table de Peutinger, ed. Desjardins, p. 23, col. 1.
^ Notice de Vancienne Gaule, pp. 266-7.
^ Diet. arch, de la Gaule, i, 339-40.
" W. Smith, Diet, of Greek and Roman Geogr., i, 772.
^^ Geogr. de la Guide an VP siecle, 1878, p. 318.
 —
DIABLINTES 419

lie Oxmensis, in the diocese of Seez, doubtless owes

says, the paijus
its name But nobody would contend that it was their
to the Osismi.
capital for it was not in their proper territory. It was probably
;

only an Osismian colony. The territory of the Diablintes proper was

in the neighbourhood of Aleth (St. Servan) between the territories
of the Redones, the Veneti, and the Coriosolites Aleth is a con- :

traction of Dialeth (Dialetum) and Dialetum was the capital of the

;

Diablintes.
M. Longnon's arguments have been answered by the distinguished
Breton antiquary, A. de la Borderie.^ As he remarks, M. Longnon's
theory rests upon the hypothesis that the Coriosolites disappeared as
a civitas, before the publication of the Notitia provinciarum, that is
to say before the fifth century, and that they were absorbed by the
Diablintes and, as I have shown, this is a mistake. Nevertheless,
;

I will examine M. Longnon's arguments.

1. The only passage in which Caesar mentions the Diablintes
socias sihi ad id helium (Veneti) Osismos, Lexovios, Namnetes, Am-
biUatoSy Morinos, Diablintes, Menapios, adsciscunt ; auxilia ex
Britannia, quae contra eas regiones posita est, arcessunt^ proves, in —
M. Longnon's opinion,^ that the Diablintes were a maritime people.
M. de la Borderie * replies that the statement of Ptolemy, which
I have quoted above, proves that the territory of the Diabhntes
was not maritime. It is true that Ptolemy is sometimes mistaken ;
and I do not think that his statement, taken by itself, amounts to
proof. But it is supported by the fact, admitted by M. Longnon,
that some Diablintes lived in Jublains and the neighbourhood. More-
over, Caesar's statement that Britain was opposite a number of states,
of which the Diablintes were one, does not prove that the Diablintes
actually possessed a seaboard. It has, indeed, been argued that they
must have dwelt on the coast, because Caesar includes the Aulerci in
his list of maritime tribes,^ and none of the other known Aulercan
peoples possessed a seaboard. But, assuming that Caesar did not
carelessly include the Diablintes, who were certainly near the sea,
among the maritime tribes strictly so called, the Aulerci may have
included other peoples besides the Cenomani, the Brannovices, the
Eburovices, and the Diablintes moreover, it was not Caesar but
:

Ptolemy who included the Diablintes among the Aulerci. Again, in

support of Ptolemy's statement, the French Commission ^ point out
that, while Caesar twice enumerates the maritime (Aremorican)
states,' he does not include the Diablintes among them. Finally, it
is possible that the Diablintes, even if we place them next the Ceno-
mani, may have possessed a strip of coast.
2. M. Longnon^ maintains that if Jublains had been the chief
town of the Diablintes, it would have been the see of a bishop. But
M. de la Borderie replies that it is unreasonable to assume that
^
Geoyr. gallo-rom. de V Arinorique. B. G., iii, 9, § 10.
'^

* Mhn. du Congres scientijique de France, 38® «ess"., 1872, p. 430.

* Geogr. gallo-rom. de V Armorique, pp. 4-5. * B. G., ii, 34.
•*
Diet. arch, de la Gaiile, i, 340. B. G., ii, 34 ; vii, 75, § 4.
''

^ Mem. du Congres sc. de France, 1872, p. 429.

E e 2
420 DIABLINTES
because, as a general rule, Gallo-Roman states became dioceses, they
did so without exception.
3. M. Longnon quotes a passage from the Acta Sanctorum (October,
t. vii, pars 2, p. 1098), from which it appears that in the ninth and

tenth centuries the metropolitan see of Dol and its seven subordinate
sees, St. Pol de Leon, Vannes, Carhaix, Quimperle, Quimper, Portus
Saliocan, and Diallentic, were pillaged by the Northmen. This
appears to him a proof that the territory of the Diablintes was
situated in the Aremorican peninsula. De la Borderie ^ replies first,
that the date of the document quoted by M. Longnon deprives it of
all value and secondly, that there is no proof that by Diallentic
;

the writer meant Aleth.

4. A MS. of the Notitia provinciarum, belonging to the tenth
century, has the gloss (civitas Diablintum) quae alio nomine Aliud
vel Adala vocatur ; and another MS. of the same century has the
gloss (civitas Diablintum) id est Carifes. This, s'ays M. Longnon,'*
proves that, in the Middle Ages, the civitas Diablintum was not
regarded as having formed part of the diocese of Le Mans for the ;

gloss id est Carifes evidently points to one of the numerous Breton

local names of which the radical prefix is the Breton her (a house),
Replying to this argument, M. de la Borderie ^ says, We are told '

that the least practised eye must recognize Aleth in Aliud and that
Adala means Dol. Admitting this provisionally, it should seem
. . .

that the Diablintes must have had three capitals Aleth, Dol, and —
—
''CWifes" and as this is two too many, the only conclusion to
which we can come is that the copyists of the Notitia did not know
which was the real capital, and wrote these names at haphazard. . . .

The gloss Carifes is to be regarded simply as a blunder on the part

of the copyists, analogous to that which, in the same chapter of the
Notitia, ascribes to Vannes the name of Ciancti, civitas Cianctium
... a name which Vannes has never borne.'
5. Comparing the province of (Gallia) Lugdunensis Tertia, as
described in the Notitia provinciariwi, with the ecclesiastical province
of Tours, we find, says M. Longnon,^ that the first seven states of the
former correspond with seven dioceses of the latter :

Landranen, bishop of Tours Metropolis civitas Torinoruin

Aldric, ,, Mans Civitas Cenomannorum
Gernobrius, ,, Reiines ,, Redonuin
Dodon, ,, Angers ,, Andicavoruni
Actard, ,, Nantes ,, Namnetuni

^ II n'est pas vrai,' says M. de la Borderie (pp. 31-2), ' que tout«s les cites de
'

la Notice soient devenues des dioceses. Outre les Diablintes, il en est quatre
tout au moins d'oa Ton ne voit sortir nul eveclie, savoir la civitas Boatium : . . .

civitas Rigomagensium (Chorges), civitas SoUinicntiuni (Seillaus), civitas

Eqiiestrium . D' autre part, en dehors de la troisienie Lyonnaise, on trouve
. .

dans les Gaules au moins une dizaine d'cveches qui ne figurent point conime
. . .

cites dans la Notice/ o. g. Nevers, Laon, Maurienne, Toulon, and Carpentras.

(See also J. Loth, U
Emigration hretonne, p. 49.
^ Geogr. gallo-roni. de V Armorique, pp. 13-4.
^ Mem. du Congres sc. de France, 1872, p. 434.
* Geogr. gallo-rom. de V Armoriqae, pp. 29-30.
^ Mem. du Congres sc. de France, 1872, pp. 400, 420.

i
 —
DJABLINTE8 421

Felix, episcopus Corisopitensis Civitas Corisopitum

8usannus, bishop of Vannes ,, Venetum
Liberalis, episcopus Oximensis ,, Ossismorum
Salacon, episcopus Aletensis or Dialetensis ,, Diablintum
Besides those seven states Lugdunensis Tertia included the civitas
Ossisfnormn and the civitas Diablintum besides those seven dioceses
:

the province of Tours included in 848 the episcopatus Oximensis and

the episcopatus Aletensis. In the face of these facts, says M. Longnon,
no man who is not blinded by prejudice can fail to see that the
episcopatus Aletensis (or Dialetensis), out of which were formed
the dioceses of St. Brieuc, St. Malo, and Do), was identical with the
civitas Diablintum.
This argument perhaps looks plausible but it rests upon the
;

assumption, which in my article on the Coriosolites I have refuted,

that the compiler of the Notitia wrote Corisopitum and not Corioso-
litum, and upon the assumptions that the territory of the Cenomani
comprised the whole of the diocese of Le Mans, that the territory of
the Eedones was identical with the diocese of Rennes, and that the
territory of the Osismi did not include, besides the episcopatus Oxi-
mensis, some part of the episcopatus Corisopitensis or diocese of
Quimper. Moreover, it must be remembered that in the fifth century,
when the Notitia was published, only three of the dioceses in the
—
Aremorican peninsula namely Vannes, Nantes, and Rennes
existed : there was no episcopatus Corisopitensis, no episcopatus
Oximensis, and no episcopatus Aletensis.^ The two lists which
M. Longnon prints side by side are easily explicable on the orthodox
theory. The civitas Diablintum occupied a part of the diocese of
Le Mans the territory, or a part of the territory ,2 which subsequently
:

corresponded with the episcopatus Aletensis belonged to the civitas

Coriosolitum. There is no evidence that the diocese of Le Mans
corresponded exactly with the territory of the Cenomani, or that
the Redones possessed only the diocese of Rennes. What shadow of
proof is there, then, that the territory of the Diablintes corresponded
with the episcopatus Aletensis ? Absolutely none, save the imaginary
resemblance between the names Diablintes and Aletensis.
Writing in 1885,^ M. Kerviler, who had formerly supported
M. Longnon, frankly admitted his mistake but he insists that his ;

conversion is due solely to M. Loth. This eminent Celtic scholar

shows in his work, UEmigtation bretonne, that the dioceses of Breton,
as distinguished from those of Gallo-Roman origin in Brittany were
not formed out of Gallo-Roman states. The Aremorican peninsula,
completely Romanized in the fifth century, was completely Bretonized
by the British invasion which followed. The invaders, in forming
their dioceses, took no account of the Gallo-Roman states. It is
therefore idle to look for a complete correspondence between the
political divisions mentioned in the Notitia and the ecclesiastical
divisions mentioned in 848 by the council of Tours.

^ See J. Loth, L' Emigration hretonne, pp. 75-82.

^ See pp. 415-6.
^ Bull. arch, de Z'^ss" bretonne, 3*"
ser., iv, 1885, pp. 225-8.
422 DTABLTNTES
In his Atlas historique de la France (p. 4) M. Longnon says les '

arguments de nos adversaires ne nous ont pas convaincu. Cependant

nous n'avons pas voulu, pour une question ainsi controversee, abuser
de Toccasion qui nous est offerte aujourd'hui pour faire penetrer en
quelque sorte notre opinion dans le domaine public, et nous avons
marque le nom des Diablintes aupres de la ville romaine de Jublains
(Diahlintes)\ &c. If I may say so without offence, M. Longnon's
modesty is out of place. He has not hesitated on other controverted
points, for example, the position of the Atuatuci (q.v.), to faire '

penetrer son opinion dans le domaine pubHc'. If he believes that the

Dia])lintes did not occupy the neighbourhood of Jublains, he is
doing his best to mislead his readers by marking on his map the name
Diahlintes near Jublains. If his map represents his real opinion, he
ought to recant his error.
Eburones. See Atuatuci.
Eburovices. See Aulerci Eburovices.
Eleutet'. See Cadurci.
—
Elusates. The Elusates occupied the country round Eauze.
Their frontiers cannot be traced but they apparently possessed the
;

north-western part of the department of the Gers and a fraction of

the southern part of that of the Lot-et-Garonne.^
—
Esuvii. Caesar is the only ancient writer who mentions the
Esuvii and, if I am not mistaken, their name, under one form or
;

another, occurs three times in the Commentaries.'^ In the first pass-

age,^ they are mentioned together with the Veneti, the Venelli, the
Osismi, the Coriosolites, the Aulerci, and the Redones, which,' says '

Caesar, are maritime tribes, whose country reaches the ocean ' in
' :

the second ^ they are mentioned together with the Coriosolites

and the Veneti in the third ^ they appear as the people into whose
:

territory the legion of Roscius was sent to winter in 54 B. c. and ;

from a subsequent chapter ^ we learn that Roscius was threatened

by the maritime states. It is therefore certain that, although the
MSS. offer a large choice of readings, Caesar is speaking, in all three
passages, of the same people.
In the first passage all the good MSS. have Sesuvios, and in the
third Essuos in the second the a MSS. have Esuhios.
: Schneider
reads Sesuvios in the first two passages and Essuos in the third, re-
marking that Sesuvios might easily have been corrupted into Esuhios,
but that Sesuvios could hardly have been evolved by any copyist
out of Essuos. Nipperdey and Meusel read Esuvios in all three
passages and it is probable that either Esuvios or Esuios is the
;

* Did. arch, de la Oaule, i, 368 Desjardins, Geogr. de la Gaule rom., ii, 367
; ;

Longnon, Atlas hist, de la France, p. iv.

^ Four times, according to Mommsen {Jahresb. d. philol. Vereiiis zu Berlin, xx,

1894, p. 21 1 ), who holds that in vii, 75, § 3 Esuviis should replace Lexoviis and if ;

Lemovices [Aremoricani] (q.v.) is wrong in§ 4, 1 have no doubt that he is right.

One would expect, he says, to find the Lexovii mentioned in § 4 among the
Aremorican tribes and the omission of the Esuvii [unless, indeed, thej^ were
;

included, though not named, among those tribes] would be startling.

3 B. G., ii, 34. * lb., iii, 7, § 4.
« V, 24, § 2. 6
//;., ji,^^ 53^ § o.
 ESUVII 423

right form, because these forms are found on coins,i while the first
letter of Sesuvios is obviously due to dittography. the careless —
repetition of the last s of the preceding word, Coriosolitas. M. Long-
non,2 however, believes that in all three passages Caesar wrote
Lexovios. Others again distinguish the Essui from the Sesuvii, but,
like M. Longnon, identify the Sesuvii with the Lexovii. They urge
that of all the Aremorican peoples the Lexovii were one of the most
important and the most frequently jnentioned. Therefore, they say,
it w^ould have been extraordinary if they had been omitted from the
list of the peoples whose submission Crassus was sent to receive.
Finally, they point out that in the list of the tribes which were
called upon to furnish contingents for the relief of Vercingetorix ^
the name Sesuvii is omitted and this, they argue, is another reason
;

for identifying them with the De

Valois ^ gives an inde-
Lexovii.^
pendent reason for doing the same. He observes that in B. G., ii, 34
Caesar writes Sesuvios immediately after Coriosolitas, and that in
B. G., iii, 11, § 4 he writes Coriosolitas Lexoviosque.
The French Commission ^ reply first, that Lexovii is found, without
any various reading, in five other passages of the Commentaries,'^
besides the two under discussion, and that it is therefore unlikely
that if in those two passages Caesar had written Lexovios, the various
readings should have been so many ; secondly, that, if Lexovios
were read in B, G., iii, 7, § 4, its presence in iii, 9, § 10 would be
inexplicable, for it would be absurd to say that the Veneti allied
themselves with the Lexovii, when the Lexovii were their allies
already thirdly, that it would be inaccurate to describe the Lexovii
;

as neighbours ' (finitimi ^) of the Veneti. This last argument is

'

worthless: the diocese of Seez, where the Commission place the

Esuvii, is very nearly as far from Venetia as the country of the
Lexovii. In support of the other two arguments, it may be said
that the theory which identifies the Sesuvii with the Lexovii is
shattered (1) by the coins to which I have referred, and (2) by the
fact that in none of the three passages does any single MS. give
the reading Lexovios and I therefore regard it as certain that there
;

was, in the time of Caesar, a tribe called Esuvii, and that that tribe was
the one which the MSS. call Sesuvii. It remains to inquire where
their territory was situated.
Now, if Caesar was the only ancient writer who mentioned the

^ Desjardins {Geogr. de la Gaule rom., ii, 491, n. 1) says that M. C. Robert sent
him a Gallic coin found in Jersey, bearing the inscription Esvvios. But
Desjardins mis-spells the name. The name on the coin in question is spelt
Esvios. This form, however, in the opinion of Eugene Hucher, is the proto-
type of Esvvivs, which is found on two coins of Tetricus. Melanges de numismU'
tique, i, 1875, pp. 321-2. Cf. Corpus inscr. Lat., xiii, pars ii, fasc. 2, p. 669,
and A. Blanchet, Traite des moyin. gaul., pp. 81, 319, 400.
^ Atlas hist, de la France, ^ B. O., vii, 75.
p. 5.
* Rev. arch., nouv. ser., ix,
1864, pp. 411-2.
^ Notitia Galliarum, p. 494.

•^
Rev. arch., nouv. ser., ix, 1864, p. 412.
' B. G., iii,
9, § 10 ; 11, § 4 ; 17, § 3 ; 29, § 3 vii, 75, § 3.
» B. G., iii, 8, § 3.
424 ESUVII
Esuvii, there were two maritime peoples, the Baiocasses and the Vidu-
casses, whom he did not mention at all. Augustodnrns, the chief
town of the Baiocasses, was indisputably Bayeux. Arigenus or
Aregenua, the chief town of the Viducasses, appears in the Table as
Araegenue, which was indisputably Vieux, near Caen.^ The Baio-
casses appear for the first time, as a civitas, in the Notitia frovincia-
rum^^ which does not mention the Viducasses at all. Both tribes, it
is true, figure in Pliny ^ but Pliny mentions several tribes which
:

did not rank as civitates and Desjardins considers that, at the time
;
•*

when Pliny wrote, the Baiocasses, or, as Pliny calls them, Bodio-
casses, were merely clients of the Viducasses. The Notitia frovincia-
rum ^ mentions a people called the Saii, who do not figure in the
Commentaries, and whose territory is identified with the diocese of
Seez. As Caesar mentions the Esuvii side by side with the Aulerci,
who occupied the dioceses of Le Mans and Evreux, Walckenaer ^
infers that they occupied the diocese of Seez, which is conterminous
with those two dioceses. The French Commission arrive at the "^

same conclusion. After assigning to the various peoples of Normandy

on the west of the Seine, namely the Lexovii, the Aulerci Eburovices.
the Ambibarii, and the Venelli, the dioceses which belonged to them,
they find the diocese of Seez unoccupied, and accordingly assign it
to the Esuvii.
Desjardins goes further. Having regard to the fact that the Esuvii
are mentioned only by Caesar, and that Caesar does not mention
either the Baiocasses or the Viducasses, he identifies the joint terri-
tories of the two latter tribes, which, he supposes, included the
diocese of Seez, with the territory of the Esuvii. If he is right, the
Esuvii possessed the central and the western parts of the departments
of the Calvados and Orne.
It is clear that we cannot get beyond conjecture but the con-;

jecture of Desjardins appears to me the most reasonable that we

have. Just as the Helvetii were divided into four pagi, so it is prob-
able that the Baiocasses, the Viducasses, and the Saii were 'pagi, if
they were not clients, of the Esuvii. The French Commission,^ in-
deed, include the territories of the Baiocasses and Viducasses in the
territory of the Lexovii (q.v.) but, unless we accept Desjardins's
;

suggestion, it must be admitted that Caesar was mistaken in describ-

ing the Esuvii as a maritime people.
A. Bertrand, having regard to the position in which Ptolemy
places the Arvii, suggests that 'Apovtoi is probably only a corrupt
form of 'Ea-a-ovLOi (Essui).^ Yes, corrupt indeed Suggestions of
!

this kind lead to nothing. See Coriosolites.

Gabali. See Arverni.

'
Hist, de VAcad. des inscr., xxxi, 1768, p. 235.
-
Ed. O. Seeck, p. 263 (ii, 3). ^
^j^f j^^^f^ ^^^ ig (32), § 107.
*
Geogr. de la Gaule rom., i, 338 ii, 492-3.
;

5
Ed. 0. Seeck, p. 263 (ii, 6). « Geogr. des Gauhs, i, 391.
'
Rev. arch., nouv. ser., ix, 1864, p. 409.
^ Diet. arch, de la GauJe, ii, 90-1.
^
JRev. arch., nouv. ser., ix, 1864, p. 413.
 GARUMNI 425

—
Garumni. The Garuinni are mentioned by no ancient writer,
except Caesar.i D'Anville " agrees with de Valois in placing them
in the valley of the Garonne, below St. Bertrand-de-Comminges ;

and Walckenaer, the French Commission, and Desjardins accept

this conjecture.
—
Gates. The Gates are placed by d'Anville,^ who follows Sanson
and de Valois, between the Elusates and the Ausci, in the comte de '

Gaure '. But, as the French Commission observe,^ this conjecture

was based merely upon the resemblance of the name Gaure to ' '

Garites, the name by which the tribe in question was designated in

the old editions of the Commentaries. Now the reading Garites is
only found in a few bad MSS.^ the best have Gates. Walckenaer
:

makes a similar guess, which is not worth transcribing.

Geidumni. See Nervii.
—
Gorgobina. Gorgobina was the stronghold, or the chief strong-
hold, of the Boi.^ In order to find its site, it is first necessary to
find the district which the Boi occupied.
Caesar says that the Boi, after the conclusion of the Helvetian
campaign, were allowed by him, at the request of the Aedui, to
settle in Aeduan territory and he remarks that the Aedui made this
;

request because they knew that the Boi were a brave people.'^ We
may reasonably conclude that the Boi were established in the western
part of the Aeduan territory, where they might perhaps serve as an
—
outpost against those old rivals of the Aedui, the Arverni.^ Indeed,
on any other hypothesis, Caesar's narrative of his march from Cena-
bum (Orleans) to Noviodunum and thence to Avaricum (Bourges),
taken in conjunction with his statement that Vercingetorix, march-
ing from the country of the Arverni to Gorgobina, passed through
the country of the Bituriges, is inexplicable. I may add that, during
the siege of Avaricum, Caesar expected a supply of corn from the
Boi ;and he would hardly have done so if they had not been near
enough to forward the supply, that is, somewhere in the western
part of the Aeduan territory.^
According to one of the earliest editors of Caesar, Raimondus
Marlianus,!^ the Boi settled in the Bourbonnais. This was also the
view of d'Anville.ii He premised that the Boi must have dwelt
somewhere in the Bourbonnais, because the route which Caesar
followed from Agedincum (Sens) by way of Orleans across the Berri
must have led into that country. He went on to quote passages from
early mediaeval writers, showing, in his opinion, that that part of the
Bourbonnais which lay on the west of the Allier had belonged to
the Bituriges and the Arverni. Therefore, he concluded, the Boi
must have occupied the part between the Allier and the Loire. If this

^ B. G., 27, § 2.
iii, ^ j^Qtice de Vancienne Gaule,
p. 342,
^ lb., p.340. « Diet. arch, de la Gaule, i, 436.

5 Schneider's Caesar, i, 290. « E.G., vii, 9, ' lb. and i, 28,

§ 6. § 5.
^ See Bull, de la Soc. d' emulation du dep^
de V Allier, viii, 1859, pp. 288-9.
» B. 6r'., vii, 12-13; 17, § 2.
9, § 6;
'" Veterum Galliae locorum
. descriptio, ed. 1544.
. .

" Eclaircissemens sur Vancienne Gaule, pp. 203-6.

426 GORGOBTNA
conclusion is wrong, then, unless d'Anville is mistaken in holding
that the Aedui possessed no territory on the west of the Allier, the
Boi must have been somewhere east of the Loire, which, below its
confluence with the Allier, formed the boundary between the Bituriges
and the Aedui.^ But such a position is open to the objection that
the Aedui had no reason for establishing an outpost against the
Bituriges, who were in a state of friendly dependence upon them ;

and, as far as I know, there is nothing to be said in its favour. Never-

theless, this opinion has found adherents. Walckenaer^ supposes
that Caesar would have restored to the Boi their original territory,
which he assumes to have been occupied by the Aedui and accord- ;

ingly he places them in the district which afterwards became the

diocese of Auxerre. He observes that near Entrains, in the centre
of that diocese, there is a place called Boui, which, in the Middle
Ages, was known as Boiacum. But the name Buy ', between which
'

and Boi the curious may also discover an affinity, is found more than
once in the district between the Allier and the Loire. Moreover,
there is no reason for assuming that the diocese of Auxerre was the
original home of the Boi, except the fact, if it is a fact, that the
diocese of Auxerre was adjacent to the territory of the Senones,^ and
that the Boian settlers in Italy were neighbours of the Senones.
And this reason is purely fanciful.
In support of Walckenaer's view, a statement of Pliny * has been
quoted intus autem Aedui foederati, Carnutini foederati, Boi,
:

Senones, &c. But there is not the slightest proof that Pliny meant
to enumerate the states in question according to their geographical
order ^ and if he did, his words would seem to mean that the Boi
;

were wedged in between the Carnutes and the Senones, which is

absurd.
On the other hand, it is impossible to prove that the Boi did not
dwell in the diocese of Auxerre. For it has been proved that that
diocese, or a part of it, belonged to the Aedui in a.d. 245 ^ ;

and it is probable that it did so in Caesar's time as well. The route

which Caesar followed from Sens by way of Orleans across the
Berri would have served his purpose even if the Boi had been east
of the Loire ;for he would have diverted Vercingetorix from attack-
ing them by merely threatening Avaricum. The one strong argu-
ment against placing the Boi in the diocese of Auxerre is the argument
which I have already stated, namely that it is more reasonable to
suppose that the Aedui should have established them as an outpost
against the Arverni than against the Bituriges.
I conclude, then, that, while it is not certain that the country of
the Boi was between the Allier and the Loire, it is more probable
that it was there than that it was anywhere else.''

1 See B. G., vii, 5, §§ 3-4, and pp. 351-2. ^ Qeogr. des Ganles, i, 82-4.

3 See pp. 471-3. * Nat. Hist, iv, 18 (32), § 107.

^ In the same chapter Pliny mentions the Andegavi (Andes), who Hved in

the neighbourhood of Angers, immediately after the Tricasses, who Hved in the
neighbourhood of Troyes !
'^
Desjardins, Geogr. de la Gaule rom., iv, 184-5.
' Brugiere de Lamotte argues that, if the territory of the Boi had been on
 GORGOBINA 427

But if we cannot positively determine the territory of the Boi,

II.
how can we fix the site of their chief town ? Needless to say that
there has been no lack of guesses. M. Clairefond ^ easily disposes of
most of them, showing that the proposed sites are either outside the
Aeduan frontier or otherwise irreconcilable with Caesar's narrative.
There is really nothing, except doubtful military considerations, to
guide the inquirer but it may be useful to examine the conjectures
;

which the best known commentators have made.

1. Gorgobina has been identified with Moulins. But Moulins is
not an old town ^ and besides, it is so far south that we may doubt
;

whether Vercingetorix, after hearing the news of Caesar's departure

from Cenabum (Orleans), would have been able to reach Noviodunum
in time to fight a battle for its relief. For it must be remembered
that, while the news was travelling from Cenabum to Gorgobina,
Caesar was marching southward from Cenabum towards Novio-
dunum.-^
2. Napoleon,^ following M. Crosnier,^ believes that he has found
the site at St. Parize-le-Chatel. Crosnier says that this place was
once known as le village de Gentili \ or, as it was called in the
'

legends of St. Patrice, Pagus Gentilicus

'
and that the people of
'
;

this pagus remained idolaters until the middle of the sixth century,
that is to say, for two centuries after the neighbouring peoples had
accepted Christianity. This, argues Napoleon, is what we might
expect from a tribe settled in a foreign country as the Boi were,
'

who would retain their customs and religion for a longer time un-
changed '. Perhaps. But precisely the same argument has been
advanced in favour of placing the Boi in the neighbourhood of
Sancerre.^ At Buy in the neighbourhood of St. Parize-le-Chatel, says
Crosnier, there is a decided bend in the Roman road from Augusto-
dunum (Autun) to Avaricum (Bourges) and from this circumstance
;

he concludes that on or near the bend in the road, that is, at or near
Buy, was situated the opfidum of the Boi. The only argument of
any weight which Napoleon adds is that the site of St. Parize-le-
Chatel is better adapted for a Gallic stronghold than any other which
has been proposed. The choice has the qualified support of Des-
jardins, who speaks of Gorgobina
'^ '
que I'auteur de la Vie de
. . .

Cesar place, avec vraisemblance, a Saint-Parize-le-Chatel '.

the right bank of the Alher, they could not have brought corn to Caesar at
Avaricum (Bourges), or Vercingetorix would have intercepted them. But
Vercingetorix could have intercepted them just as well if they had been on
the left bank.
Again, Brugiere insists that, if Gorgobina was between the Allier and the
Loire, Vercingetorix, in order to get from Gorgobina to Noviodunum, would
have had to cross the Allier, which was not fordable. But he forgets that the
Allier was spanned by several bridges {B. G., vii, 34, § 3). His arguments
may be found in Bull, de la Soc. d'emulation du dep^ de V Allier, ix, 1864,
pp. 434-5, 437-9, 444, 454, 473-5.
1 lb., vii,
1859, pp. 294-303.
^ D'Anville, Eclair cissemens sur Vancienne Gaule,
pp. 209, 236-7.
. ^ B. G., vii, 12-3. * Hist, de Jules Cesar, ii, 247-8, note.
^ Bull, de la Soc. nivernaise, viii,
,
1880, pp. 104-6,108-9.
® Diet. arch, de la Gaule, i, 171. ' Geogr. de la Gaule rom., ii, 478.
428 GORGOBINA
3. Bonniardplaces Gorgobina quite close to Boui, on the site of
^

the ruins of a Gallic or Gallo-Roman town near St. Reverien in the

department of the Nievre. But this conjecture rests simply upon
the resemblance between the names Boui and Boi '. '
'
'

4. General Creuly refuses to look for the site anywhere in the

angle formed by the Allier and the Loire. He objects that Vercin-
getorix would not, in order to enter this district, have crossed the
territory of the Bituriges on his march from that of the Arverni ^ ;

and that Caesar would not, in order to succour a town situated

between the Allier and the Loire, have crossed the j^oire so far to the
west as Cenabum (Orleans), when he might easily have gone direct
by Nevers. Accordingly Creuly thinks that the site must be looked
for somewhere on the west of the Allier and of the Loire. Starting
from this hypothesis, he affirms that no place in the neighbourhood
in question is so well adapted for defence as Sancerre.^ If, he says,
his suggestion is accepted, one can easily understand why Vercin-
getorix established himself at Gorgobina. His object was to prevent
Caesar from crossing the Loire, all the bridges over which that lay on
or near the probable direction of his march had doubtless been
demolished by the Carnutes and the Bituriges but Caesar upset his ;

calculations by making a detour by way of Cenabum.^ CVeuly sug-

gests that when Caesar said that the Loire separated the Bituriges
from the Aedui, he only meant that part of the Loire which was in
the neighbourhood of Nevers for in Creuly 's opinion it was prob-
;

ably at Nevers that the Aedui crossed the Loire when, as Caesar
relates in B. G., vii, 5, they entered the country of the Bituriges. It
seems to me much more probable that Caesar meant that all that
part of the Aeduan territory which extended northward of the con-
fluence of the Allier with the Loire, was separated from the territory
of the Bituriges by the Loire.^ Nor have Creuly's other arguments
any real weight. It is true that if Vercingetorix had started from
the neighbourhood of Gergovia with the intention of marching direct
to the district round St. Parize-le-Chatel, he would naturally have
gone down the valley of the Allier, and therefore need not have
entered the territory of the Bituriges. But it is not necessary to
assume that he did march direct to Gorgobina he may have had :

reasons, of which we know nothing, for going back first into the
country of the Bituriges and, indeed, Caesar's words Vercingetorix
;

rursus in Bituriges exercitum reducit at que inde profectus Gorgo-

hinam . . . —
oppugyiare instituit^ would seem to suggest that from
the country of the Bituriges he started off in a new direction. The
objection that Caesar, marching from Agedincum, would not have
gone so far to the west as Cenabum in order to relieve a town situated
within the angle formed by the confluence of the Allier and the Loire,
has been answered by anticipation in the note on Cenabum. Besides,
if Cenabum lay far out of the way from Sens to St. Parize-le-Chatel,

^ Bull, de la Soc. d' emulation du dep^ de V Allier, viii, 1859, p. 298.

2 B. G., vii, 9, § 6. » See Carte de VEtat-Major (1 80,000), Sheet
: 123.
* Rev. arch., nouv. ser., viii, 1863, pp. 398-9.
5 See pp. 351-2. « B. G., vii, 9, § 6.

i
 GORGOBINA 429

it lay just as far out of the way from Sens to Sancerre. And Oreuly
himself maintains that Caesar made a detour.
M. Jacques Soyer,i has, however, recently supported General
Creuly's choice. He argues that the form Gorgohina is bizarre,
— —
whereas Gortona the reading of the fS MSS. is familiar in ancient
toponymy and he affirms that Gortona is the primitive name of
;

Sancerre. As a matter of fact Sancerre has had, at successive epochs,

various names, the earliest known of which was Castrmn GordonicumJ'^
The name Gortona, of which M. Soyer does not produce a single
instance (though it may be equivalent to Gortona), is not more
familiar than Gorgohina and Gorgohina has been accepted without
;

question by Celtic scholars. However, if Caesar wrote Gortona, it is

more than probable that the fort which Vercingetorix besieged was
Sancerre. M. Jullian^ objects, first, that Sancerre is in the ancient
diocese of Bourges, and, secondly, that the Aedui would never have
entrusted so strong a fort to the keeping of the Boi. The former
objection is valid. As to the latter, it seems to me hardly probable
that if the Aedui, as M. JuUian himself maintains,'* had entrusted
the Boi with the defence of Gorgohina as an advanced post against
the Bituriges, they would have deliberately assigned to them a weak
rather than a strong position. What had they to fear from their
feeble dependants ? Still, I agree with M. Jullian in rejecting
Sancerre for leaving out of account the fact that it is in the diocese
;

of Bourges, I cannot accept the reading Gortona and I doubt

;

whether the mission of the Boi was to check the Bituriges.^

5. The French Commission originally decided for St. Pierre-le-
Moutier but this place is ill situated for defence ^ and the Com-^
: ;

mission afterwards recanted, and professed themselves convinced by

the arguments of Creuly.'
6. Von Goler ^ identifies Gorgohina with La Guerche-sur-l' Aubois,
which is 17 kilometres, in a straight line, west by south of Nevers ;
but General Creuly ^ remarks that the name has only a deceptive
resemblance to Gorgohina (or, as he calls it, Gergovina). M. Jullian,
however, defends von Goler's hypothesis on different grounds.
Arguing that Gorgohina must have been between the Aedui and the
Bituriges, but south of that part of the Loire which separated those
two peoples,^^ he remarks that it must also have been in fertile
territory ,11 and concludes that these conditions are satisfied by

Bull, degeogr. hist, etdescr., 1904, pp. 157-9.

^

" Mem.
de la Soc. hist, du Cher, 3^ ser., ii, 1882, p. 318. A. Chazaud {Bull,
de la Soc. du dep*^ de V Allier, viii, 1859, pp. 90-1), who also adopts the read-
ing Gortona, identifies the stronghold with St. Satur, between Sancerre and
St. Thibaut, which, he says, was formerly called Gortonis castrum ; but
M. Hippolyte Boyer {Mem. de la Soc. hist, du Cher, ii, 1882, pp. 301 ff.) has
proved that the attribution of this name to St. Satur is unfounded.
^ Hist, de la Gaule, iii, 434, note.

* Eev. des etudes anc, v, 1903, p. 30, n. 1. ^ Sec

p. 420.
" Hist, de Jules Cesar, ii, 248, note. ' Diet. arch, de la Gaule, i, 453-4.
^ Gall. Krieg, 1880, p. 237 and
n. 3. ^ Hist, de la Gaule, ii'i, 434, note.

*" See pp. 351-2. It must

be remembered that M. Jullian does not believe
that the Loire, strictly speaking, separated the Aedui from the Bituriges.
11 B. a,vii,
17, §3.
430 GORGOBINA
La The site of the stronghold would then have been the
Clueiclie.
rising ground on which the church stands. But, as M. JulUan truly
says, the country between the Allier and the Loire, which, in agreement
with me, he attributes to the Boi, is likewise fertile and that Gorgobina
;

must have been intended as an outpost against the Bituriges I cannot

admit. Besides, as I have already shown,i it is milikely that the
Aedui possessed any territory west of the Allier and of the Loire ;

and that they should have assigned to the Boi both this territory
and that between the two rivers seems to me more unlikely still.
The reader has now before him the pith of what has been written
upon the subject. The conclusion of the whole matter, in my judge-
ment, is that there is not sufficient evidence for fixing the site of
Gorgobina but that there is more to be said for St. Parize-le-Chatel
;

than for any other site which has been proposed. Li short, I claim
only a negative value for this note it shows that cartographers who
:

mark on their maps the territory of the Boi and the site of Gorgobina
as if their geographical positions were certain, are not to be trusted.
—
Graioceli. The Graioceli are mentioned only by Caesar,^ immedi-
ately after the Ceutrones (q.v.) and immediately before the Caturiges
(q.v.). These three Alpine tribes attacked him in 58 B.C., when he
was returning from Italy to Transalpine Gaul, to deal with the
Helvetii. Describing his march, he says that the last, that is to say,
the westernmost town in the Citerior Provincia is Ocelum, and that
he marched thence into the country of the Vocontii (q.v.) in seven
days. Ocelum has been identified with Exilles, Uxeau or Usseau,
Usseglio, and Aosta.^ But there are only two sites for which any
real evidence can be adduced and for one or the other of these two
;

the evidence is conclusive.

Desjardins * maintains that the name Graioceli proves that the
territory of the tribe was on the eastern slopes of the Graian Alps.
Strabo,^ he points out, says that Ocelum was 99 Roman miles from
Epeprodunum (Embrun) and, within a mile or two, this is the
;

distance from Embrun, along the Roman road which ran past
Brigantio, or Briancon, and over Mont Genevre, to Avigliana. Four
vases, on each of which an itinerary is inscribed, have been discovered
at Bagni di Vicarello, the ancient Aquae Apollinares. Three of them
place Ocelum at 20 Roman miles, or 29J kilometres, from Turin, and
two of them at the same distance from Susa.^ These figures corre-
1 See pp. 351-2.
^ B. G., i, 10, § 4. Mommsen {Jahresb. d. philol. Vereins zu Berlin, xx, 1894,
p. 200) conjectures that Caesar wrote Grai Oceli. The Graioceli, he observes, are
not mentioned anywhere else, while the Grai are mentioned by Pliny {Nat.
Hist., in, 20 [24], § 134) as inhabitants of the Graian Alps ; and he argues that
by breaking up Graioceli the sense is made clear. I confess that it seems to
me clear as the text stands, but with Mommsen' s alteration obscure. More-
over, the Esuvii are not mentioned by any ancient writer except Caesar.
^ N. Sanson, Les Comm. de Cesar, 3rd ed,, 1658, p. 64 ; A. von Goler,
Gall. Krieg, 1880, p. 13 ; d'Anville, Notice de Vancienne Gaule, pp. 500-1 ;

Napoleon III, Hist, de Jules Cesar^ ii, 56, n. 5 ; Walckenaer, Geogr. dcs
Gaules, 538-9, 542-4.
* Geogr. de la Gaule rom., i, 84-5. ^ iv, 1, § 3.
"^
The third, by an obviousmistake,places0celumat27 Roman milesf rom Susa.
 —

GRAIOCELI 431

spond approximately with the actual distances of Drubiaglio and of

Avigliana from Turin and from Susa respectively, Drubiaglio being
on the northern, and Avigliana opposite it on the southern bank of
the Dora Riparia. According to the fourth itinerary, there was
a station called Ad Fines, 23 Roman miles from Turin. The same
station, according to one passage in the Itinerary ofAntonine (p. 341),
was 18 Roman miles from Turin and 33, which is evidently a mistake
for 23, from Susa according to another (pp. 356-7) 16 Roman miles
;

from Turin and 23 from Susa according to the Jerusalem Itinerary

;

(p. 556) 16 Roman miles from Turin and 24 from Susa. At Avighana
an inscription has been discovered, containing the words Finib(ws)
CoTTi ;and Strabo speaks of Ocelum as the boundary of Cottius's
kingdom.! The inscription ^ in question and another found at the
same place prove that at the town which stood upon the site of
Avigliana was collected the duty of 2J per cent {quadragesima or one-
fortieth) which was levied upon merchandise and on the fourth vase
;

Ad Fines is called Ad
Fines XXX
X, which, as Desjardins explains,
means Ad Fines quadragesimae. From this evidence and from the
evidence of the itineraries Desjardins^ identifies Ad Fines with
Avigliana and from the fact that important antiquities have been
;

discovered at Drubiaglio,* as well as from the evidence of the itiner-

aries and of Strabo, he identifies Ocelum with Drubiaglio. It is '

indubitable,' he says, from the vestiges of the ancient roads and the
'

mile-stones found in situ that between Turin and Susa there were two
roads, one on the left bank of the Dora Riparia and the other on the
right.' ^ I can find no evidence for this assertion, except the discovery
of antiquities at Drubiaglio and Mommsen ^ denies it. Desjardins's
;

opinion, he maintains, is refuted not only by the topography but also

by the fact that the anonymous geographer of Ravenna ' places Ad Fines
and Ocelum on the same road. These two towns, therefore, he con-
cludes, were very close to one another, the former being nearer to Turin.^

^ Rev. arch., nouv. ser., v, 1862, pp. 254-8 ; xxii, 1871, pp. 124-9.
^ The inscription {Corpus inscr. Lat., v, 7213) runs as follows :

PVDENS soc .

PVBL XL SER
• •

7SCE, • FINIB
COTTI VOVIT •

ARCAR LVGVC •

S'L'M',
that is to say, Pudens, soc(iorum) publiici) quadragesimae se/'(vus) contra-
scr(iptor) Finib{\is) Gotti vovit arcar{i\is) Lugud{\ini) 6(oIvit) Z(ibens) w(erito),
or to quote Desjardins's translation, '
Pudens, esolave des fermiers associes de
Timpot indirect du quarantieme des Gaules, controleur a la station de Fines
de Vancien royaume de Cottius, a voue ce monument. Devenu trcsorier de la
douane a Lyon, il a acquitte son vceu de grand coeur.'
^ Geogr. de la Gaule rorn., iv, 19-20.
* See Carlo Promis, Storia delV antico Torino, 1869, pp. 56, 129, 238.
^ Eev. arch., nouv. ser., xxii, 1871, p. 125.

^ Corpus inscr. Lai., v, pars ii,

pp. 811-2.
' iv, 30 (ed. M. Pinder and G. Parthey, 1860,
p. 250, 1-3).
^ Mommseu's criticism notwithstanding, Desjardins {Geogr. de la Gaule
rom.,
ii, 604, iii, 319-20) adhered to his opinion. My own has been modified.
432 GRAIOCELI
M. E. Celesia, in a note which he conununicated to Napoleon,^
argued that besides the Ocelum of the itineraries, there were others
at Usseglio and at Usseau.^ UssegUo is in the valley of Viu, 40 kilo-
metres north-west of Turin Usseau is in the valley of the Clusone,
:

between Pignerol and the col de Fenestrelle '. Now, it is certain

'

that Caesar's Ocelum was neither at Usseglio nor at Usseau. For —

to say nothing of the evidence which I have already adduced —
Caesar must have taken the route by Mont Genevre and, to reach
;

Mont Genevre, he would surely have followed the well-defined route

along the Dora Riparia,^ instead of going out of his way through the
valley of the Stura or the valley of the Clusone.
Grudii. See Nervii.
—
Helvii. The Helvii occupied the Vivarais, which forms the
southern part of the department of the Ardeche.'* See Allobroges.
—
Itius Portus. In the first edition of this book I argued, incon-
-^

clusively, that Portus Itius was the former harbour of Wissant,

between Cape Gris-Nez and Cape Blanc-Nez. In Ancient Britain ^
I again approached the question, with knowledge which was all but
complete, and endeavoured to demonstrate the identity of the
port with the harbour of Boulogne. My arguments were accepted
by Mr. A. G. Peskett,' M. Salomon Reinach,^ Professor Walter
Dennison,^ and other critics as conclusive but two scholars who
:

were convinced, and one of whom was converted, by the article in

the same volume which deals wdth the question of Caesar's landing-
place in Britain, told me that the article on Portus Itius did not
seem to them to achieve demonstration and a few days before the
;

publication of Ancient Britain, when all the sheets had been printed,
I myself saw that there were defects in the reasoning. While I was
preparing the article I felt the need of trustworthy and detailed
information regarding the experiments that were made in order to
ascertain the time in which the main division of the first Napoleon's
fiotilla could clear the port of Boulogne but I failed until too late
;

to put my hands upon the authoritative work —

Captain E. Desbriere's
Projets et tentatives de deharquement aux ties hritanniques in which —
it is contained.
It w^ll not, however, be denied by any critic who has even an
elementary knowledge of seamanship or is willing to accept the
unanimous testimony of nautical experts that the article made one
contribution to knowledge it proved that the port from which
:

Caesar sailed in his first expedition was Boulogne.^^ It is now gener-

ally admitted that Portus Itius was either Boulogne or Wissant.
If, then, Caesar sailed from the same port on both his expeditions,

^ Hist, de Jules Cesar, ii, 56, n. 5.

^ Cf. C. Promis, Storia delV antico Torino, p. 129.
^ Eev. arch., nouv. ser., vii, 1863, p. 255.
* Diet. arch, de la Gaule, ii, 17. ^ Pp. 433-43.
« Pp. 552-95. ' Class. Rev., xxii, 1908, p. 94.

8 Rev. arch., 4'' ser., xi, 1908, p. 307. ^ Class. Philology, iii, 1908, p. 457.

^^Ancient Britain and the Invasions of Julius Caesar, pp. 581-3. I am glad
to lind that Mr, Stuart Jones {Eng. Hist. Rev., xxiv, 1909, p. 115), with many
other competent critics, accepts this conclusion.
 ITIUS PORTUS 433

Portus Itius was Boulogne. The difficulty is that he did not say
that on his first expedition he started from Portus Itius and while
;

I was at work upon the article I felt obstinate questionings ' in

'

regard to his having only mentioned the harbour in connexion with

the second expedition. The drift of my argument was that Boulogne
was in all respects more convenient as a starting-point than Wissant,
and that Caesar, having had experience of the superior advantages
of Boulogne in 55 b. c, would not have abandoned it in the following
year. But, for want of the information which I afterwards found,
I failed to see that Boulogne, with all its superior advantages,
may have had, for the second expedition, one serious drawback.
I have explained this matter in a paper which appeared in the
Classical Review of May, 1909.
As the question of Portus Itius is more closely related to the
story of Caesar's British expeditions than to the conquest of Gaul,
it would be out of place in this volume to discuss it in all its bear-
ings. The article in Ancient Britain contains, with one exception, all
the information that is required for forming a decision and it is
;

admitted by those who detected its shortcomings that the general

superiority of Boulogne as a port of departure for Caesar and its
general correspondence with the requirements of his narrative are
therein established. I will therefore only reprint, with certain
alterations, so much of the paper in the Classical Review as may
be necessary, pointing out the flaws in the article in Ancient Britain
that escaped the notice of reviewers, and asking scholars to consider
—
that one aspect of the question the most important of all which —
has hitherto been neglected.
'
On page 569 ^ of Ancient Britain I asked "if eight hundred ships
had been beached at Wissant [during the twenty-five days for which
Caesar was windbound at Portus Itius in 54 B.C.], would it not have
been necessary, in order to protect them from storm-driven spring
tides, to construct an enormous naval camp, the earth necessary for
w^hich did not exist ? " I asked the question because, as I have
shown on pages 566-7, there was no harbour, properly so called, at
Wissant except a creek formed by the mouth of the rivulet called
the Rieu d'Herlan, and possibly a small anchorage partially sheltered
by a shoal. The answer is that to construct a naval camp would not
have been necessary if the ships could be hauled up beyond the highest
high-water mark of spring tides. Supposing that the dune which
extends from the " ruisseau de Guiptun ", near Tardinghem, to the
*'
ruisseau d'Herlan ", at Wissant, and which did not exist in the
time of Caesar,^ were bodily removed, I am not sure that it would be
possible now to haul up ships beyond this mark ^ but if we may
;

suppose that the subsidence which, in the opinion of French geo-

logists, has taken place since Roman times between Sangatte and
Dunkirk extended to Wissant, there must in 54 B.C. have been
a fringe of beach immediately below the high ground wide enough

1 See also p. 574. 2 ^^^ Britain, &c.,

p. SGO.^^^'^^^t OF ^^'
^ This question could only be settled by taking levels. ""
^cS^^^"""^
^
I ^/ ST. MICHAEL ^.

( ':^ I COUUEGE
434 ITTUS PORTUS
to allow eight hundred ships to remain high and dry at all states of
the tide.i Hypothesis indeed is unnecessary for the foundations
;

of the mediaeval Wissant have been discovered in the dune.^

On page 571, note 2, 1 hardly allowed sufficient weight to the fact
'

that the author of Bellum Africanum (10, § 1) applies the name of

'portus to a mere anchorage, —
that of Monastir (the ancient Ruspina),
which is protected from northerly and westerly winds, but otherwise
exposed.^
On page 584 I argued, as Desjardins had done before, that the
'

sixty ships which Labienus built during Caesar's absence in Britain

could not have been built at Wissant, where there were certainly no
dockyards and whither it would have been very dijfi&cult to convey
the necessary timber whereas the material could have been carried
;

bath by road and river to Boulogne. But I overlooked a passage in

Caesar (v, 8, § 1) to which I had on an earlier page called attention.
He tells us that he directed Labienus " to protect the ports " (ut
portus tueretur), which implies that he thought it necessary to keep
more than one port under control. If, then, Portus Itius was Wissant,
the ships were doubtless built at Boulogne.
On page 585 I showed that, according to " seafaring men, both
'

English and French, who have practical experience of the winds

and the currents in the Channel ", " the passage for sailing-vessels
from Boulogne to the south-eastern part of Britain is, and always
has been, in circumstances such as Caesar described, not only very
convenient but by far the most convenient ". But Caesar had to
think of the start and of the arrival as well as of the passage and ;

this consideration brings me to the question on which the whole

—
controversy really turns, could Caesar's fleet have started from
Boulogne without becoming unduly scattered ?
It must of course be remembered that the port of Boulogne in the
'

time of Napoleon was less spacious and less deep than it was 2000
years ago, because it had been largely silted up.'* Still, the map
in which Desjardins ^ attempts to depict the state of the Liane in
Caesar's time and represents it as navigable for sea-going ships as
^ The eminent geologist, M. Charles Barrois, of the University of Lille, has

very kindly written to me on this question. Je ne crois pas,' he says, que

' '

nous ayons encore des documents assez precis pour arriver a une connaissance
decisive et absolue de la question topographique qui vous interesse. II faudrait
pour cela faire une serie de levees topographiques et de nivellements precis
qui n'ont pu etre faits encore.
'
Je ne puis done vous donner que mon impression que les conclusions de
M. Gosselet [that the coast between Sangatte and Dunkirk extended con-
siderably further seaward in Roman times than now {Anc. Britain, p. 566)]
me paraissent appuyees sur des bases solides, qui n'ont pas ete refutees, et
doivent entrainer I'assentiment, la cote s'etendant plus loin a I'epoque ro-
maine. . Je n'ai rien a ajouter a vos connaissances bibliographiques, qui me
. .

paraissent fort completes.'

^ Annales de la Soc. geol. du Nord, xxviii, 1899, Cf. Anc. Britain,
pp. 84-92.
&c., p. 566.
^ Cf. Stoffel, Hist, de Jules Cesar,— Guerre civile, ii, 110-1, and pi. 20.
* Anc. Britain, &c.,
pp. 586-7. Cf. Boulogne-snr-mer et la region bonlonnaise,
i, 1899, p. 31.
^ Qeogr. de la Gaule rom., i, 1876, pi. xv.
 —

ITIUS PORTUS 435

far as Isques —
7 kilometres from the mouth is of course in part —
conjectural and presumably ho only intended to represent the
;

harbour as it was at high water.^ Moreover, although we know

that Boulogne was from the time of Augustus the regular starting-
point for ships sailing from North-Eastern Gaul to Britain and
the naval station of the Roman Channel Fleet, we have no infor-
mation as to the largest number of ships which ever started from
it at one time. There is not even direct evidence that Aulus Plautius
sailed from Boulogne ^ if he did, some of his ships may have sailed
:

from Ambleteuse and we do not know how many he had. All

;

we know is that Caesar sailed from Boulogne in 55 b. c. with about

80 transports and a few galleys and it is probable that even this
;

comparatively small fleet was inconveniently strung out.^ Philip

Augustus is said to have assembled 1500 ships at Boulogne in 1213
for his contemplated invasion of England * but the attempt was ;

abandoned.
Captain Desbriere's researches have shown that one of the in-
'

superable difficulties with which Napoleon had to contend was this :

— was impossible, in the most favourable circumstances, to float

it
more than 100 vessels out of Boulogne harbour in one tide ^ and ;

therefore it would have been necessary for each successive relay of

ships to anchor in the roadstead until the whole flotilla had cleared
the harbour. But experience proved that it was dangerous to keep
more ships in the roadstead than would be able, in case an unfavour-
able wind sprang up, to return for shelter into the estuary and that ;

westerly and south-westerly winds, which were favourable for the

voyage, generally made the roadstead unsafe.^ Owing to the rapidity
of the current, vessels could not safely begin to move out of the port
until hall an hour before high tide ; and even those which were
rowed could not continue the operation later than two hours after the
tidebegan to fall, and then only if the wind was not against them,''
My point then is this. Although we know that the estuary of
'

^ I can hardly believe that, even at high water, the Liane, which, less than
two miles above Boulogne, is now only 8 or 10 metres wide, was then many
times wider.
2 Mr. H. G. Evelyn-White {Class. Rev., xxii, 1908, p. 205, n. 9) thinks
that it can probably be inferred [that Plautius started from Boulogne] from
'

Suetonius, v, 17 ', Quare a Massilia Gesoriacum usque pedestri itinere confedo

inde [Claudius] transmisit, &c. But Claudius was not accompanied by an
army and it is questionable whether he would have started from Boulogne
;

if he had had to get 800 ships out of the harbour. Moreover, Plautius's fleet
sailed in three divisions (Dion Cassius, Ix, 19, § 4).
3 Cf. B. G., iv,
23, § 2 with § 4.
* M. Luchaire (E. Lavisse, Hist, de France, t. iii, L^ partie, 1901,
p. 162)
appears sceptical as to the number.
* Projets et tentatives, &c., iii, 1902, pp. 451, 566.
* 91,94-5 iii, 141. Between the 1st of May and the 1st of November,
76., iv, ;

1804, more than 150 vessels were on three several occasions anchored in the
roadstead for six or seven successive days but on each occasion, when they
;

were returning into the harbour, some of them were dispersed or injured {ib.,
iv, 145). Except in the very narrow space formed by the channel of the Liane,
which at low water nowhere exceeded 40 metres in breadth and was in many
places not more than 20, the ships were generally aground {ib., iii, pp. 147-8).
The vessels of least draught could only cross the bar even at spring tides during
four hours. '
Ih., p. 144.

Ff 2
 436 ITIUS PORTUS
the Liane was larger and deeper in Caesar's time than in Napoleon's,
we can hardly be sure that Caesar would have been able to get eight
times as many ships out of it in one tide as Napoleon and even ;

if he could have done so, they would probably have been obliged

to anchor in the roadstead as they emerged until the whole flotilla

had cleared the harbour. For, unless the harbour was as extensive
and as deep as Desjardins maintained, this operation might have
required several hours ^ in that case, if the ships had sailed on as they
:

emerged from the estuary, the leading division would have been off
the British coast at daybreak 2 before the rearmost had begun their
voyage and it is clear from Caesar's words that the start was
;

virtually simultaneous.^ But there is another point to mark. I have

said that the ships, as they came out of the harbour, might have been
obliged to anchor in the roadstead. But it is doubtful whether
they would have anchored in the open roadstead. Possibly they
would have been attached by hawsers to the shore, and anchored
as well. For Caesar describes his start by the words naves solvitA
Now, as Professor J. S. Eeid has written to me, " the natural
meaning of the expression [navem solvere] is ... to free the ship from
all her fastenings " and it commonly connotes the operation of
;

—
unmooring, letting go a hawser and putting off from shore or quay.
Perhaps, if the ships were merely riding at anchor, the expression
might, as Professor Reid admits, " be loosely extended to lifting the
anchor " but it seems unlikely that Caesar uses it in this sense,
;

for he repeatedly describes the operation of weighing anchor by the

words sublatis ancorisJ* If, then, Portus Itius was Boulogne, his
narrative would suggest that the ships, as they passed out of the
harbour, were moored alongshore outside until the signal was given
for the whole fleet to set sail. Now, however closely they may
have been moored, we can hardly allow a less breadth of front
for each than 7 yards.^ The ships then would have extended in
a row more than 5,600 yards long, about 5 kilometres, or more —
than 3 miles in other words, they would have reached from the
;

mouth of the Liane two -thirds of the way to Ambleteuse ! . . .

Although it would not have been possible to make as good a run

"

1 In the Classical Review (p. 79) I stated positively that to clear the harbour
'
would have required not much less than ten hours '. It would not have done
so if the harbour was free from obstructions and as capacious as Desjardins
supposed. My statement was a mistaken inference from an answer which
Captain Iron, the harbour-master of Dover, had given to a question {ih., n. 6).
2 B. G., V, 8, § 2. « lb., §§ 2, 5-6. * lb., § 2.
s
lb., iv, 23, § 6 ; B. C, i, 31, § 3 ; ii, 22, § 3 ; 25, § 7.
* The ships were small, but comparatively broad 540 of them carried ;

5 legions with their auxiliaries, camp equipage and stores, 2,000 troopers, 2,000
cavalry horses, remounts, and baggage cattle {B. G., x, I, ^ 2 2, § 2 ; 5, § 2 ; ;

8, §§ 1-2). Their breadth of beam cannot have been less than 15 feet and was
probably rather more. The breadth of one of the great merchant-ships of the
Mediterranean, the dimensions of which have been recorded by Lucian, was,
as Mr. Torr points out {Ancient Ships, 1894, p. 24), slightly more than a fourth '

of the length and Caesar says that the breadth of the ships which he
; '

designed was proportionally greater than that of the Mediterranean craft.

The breadth of Napoleon's 'bateaux canonniers', which were CO feet long
and drew only 41 feet of water, was 14 feet (E. Desbriere, Projets d tentatives,
&c., iii, 1902, p.^90}.
 —

ITIUS PORTUS 437

to Britain before a south-west wind from Wissant as from Boulogne,

although the labour of hauling up and hauling down the ships at
Wissant would have been great, these disadvantages may not have
been considered too high a price to pay for the advantages of security,
certainty, and a simultaneous start.
Professor Camille Jullian, who has seen the rough draft of this
'

paper, writes to me, " Je ne vois pas en faveur de Wissant que

le nomhre donne par Cesar [nor do I, with one exception, on which,
as the reader will see, I do not lay undue stress], et je me demande
si le pays est assez peuple, assez fertile, assez pres des bonnes
routes pour nourrir une armee de 10 [read 8] legions." The con-
siderations which my friend adduces were emphasized in Ancient
Britain. But the comparative infertihty of the country, its sparse
population, and its want of good roads would not have been fatal if
it was possible, as it surely was, to provision the army for a few
weeks by sea or by pack-horses, which could have moved on tracks
that would have been impracticable for wagons.'

Again, the fact that Caesar only mentions Portus Itius in con-
nexion with his second expedition does suggest that he did not sail
from it on his first and the impression is not removed by the
;

words he ordered all the ships to converge on the Itian harbour,

'

from ivhich he had ascertained that the passage to Britain was most
convenient^ (owmcs ad portum Itium convenire iuhet, quo ex for tic
commodissimum in Britanniatn traiectum esse cognoverat^).
A. Klotz,2 indeed, who actually identifies Portus Itius with the
'further harbour' {portus ulterior^), in which Caesar's cavalry trans-
ports assembled in 55 B.C., remarks that when Caesar mentions the
'

two harbours in the Fourth Book he gives no name, because the name
had no significance for his readers ', whereas in the Fifth Book,
if he had not given the name, he would have been forced to indicate

by a clumsy circumlocution which of the two he meant. But he

did not indicate this by giving the name nor, indeed, did he make
;

it clear that he meant either of the two.* Still, his words may suggest
more than he intended and it would not be judicial to rely upon the
;

mere fact that in the Fourth Commentary Portus Itius is not named.
My only aim has been to show that the case for Boulogne cannot
be regarded as absolutely proved, because, if there is only one real
objection, that objection may not safely be ignored. But if there
were not one gap in our knowledge, it would perhaps disappear.
If we but knew precisely the ancient conditions of the estuary of the
Liane, we might be able to identify Portus Itius as certainly as the
port from which Caesar sailed in 55. He got under weigh {naves

^ B. G., V, 2,
§ 3. Caesar uses an analogous expression in v, 8, § 3, remis con-
tendit partem insulae caperet qua optimuin esse egressnm anperiore
iit earn,

a est ate cognoverat; and the place where he landed in 54 b. c. was a little north
of the coast between Walmer and Deal, on which ho had landed in 55. See
Anc. Britain, pp. 002, 004-5. ^ Cuesarstndien,
p. 22. ^ B. G., iv, 23, § 1.
* The portus ulterior was
Aniblcteuso ; and the notion that Caesar assembled
800 vo.'^scls in that tiny port is loo absurd for discussion, ^Sce Anc. Britain,
pp. 503-4.
438 ITIUS PORTUS
solvit) about sunset.^ If these words could have been used, as
Professor Keid suggests, in an extended sense, referring to ships
which were anchored in the roadstead though of course this —
assumption would imply that the operation of clearing the harbour
—
had begun several hours before sunset I should still regard it as
not merely probable, but almost certain that Portus Itius was
Boulogne. Indeed, as the whole fleet could most probably have
cleared the harbour in one tide, and as the ships would then have
remained at anchor not more than a few hours, I only scruple
because Caesar, like Napoleon, may have feared that freshening
winds would make the anchorage unsafe. But his words appear to
mean that the fleet began to move out of the harbour about sunset :

if so, assuming that Portus Itius was Boulogne, the ships must all

have sailed on, without anchoring, as they cleared the estuary ;

and the whole business of clearing the harbour must have been
completed within an hour and a half, or the fleet would have been
strung out over a space of more than 7 nautical miles .^ Perhaps,
if the estuary was as large and as deep as Desjardins maintained,

such a feat would have been just possible.^ But unless the question
is settled by exc ivating for traces of the Roman camps, an element

of doubt, however slight, will remain.*

—
Latobrigi. It will be convenient to treat of the Latobrigi in con-
junction with the Rauraci and the Tulingi. All three peoples were
nsar neighbours of the Helvetii ^ and, according to Ptolemy,^ the
;

chief towns of the Rauraci were Augusta (Augst), about 7 miles east by
south of Basle, and Argentovaria, which was probably near Heidols-
heim, close to the common frontier of Upper and Lower Alsace.
'
B. G., V,
8, § 2. 2 ^(33 4,j(. Britain, p. 576, n. 1.
^ My
friend, Mr. W. H. Stuart Garnett, who is a thorough seaman, thinks
that, given sufficient depth and extent of water, the feat might have been
accompHshed within two hours. He suggests that the vessels may have been
ranged along both banks of the estuary at intervals of about 25 feet, so as
to allow room for the oars. They would not have drawn more than 5 feet of
water May, 1909, p. 81.
{Class. Rev., note). As, however, they would have been
obliged to keep some distance apart in order to avoid collisions, they would not
have had room to move simultaneously unless they had been formed in four
or five parallel lines. This of course would only have been possible if there was
water enough outside the narrow channel of the Liane (see p. 435, n. 6) other- :

wise they could only have moved in single file and when Caesar naves solvit
;

it was three or four hours before high tide (see Anc. Britain, pp. 729-30).
Moreover, the current would have been against them Captain Iron thinks that
;

they could not have been rowed much more than two miles an hour ; and it
must not be forgotten that they were accompanied by some of the clumsy old
transports {B. ^., v, 1, § 1). I believe that for convenience of embarkation
they would have been moored as closely as possible, like the trawlers and
'
hookers in Plymouth Sound, near the mouth of the harbour, in which case
'

considerably more time would have been required.

" Meusel {Jahresb. d. philol. Vereins zu Berlin, xxxvii, 1911, p. 99) is among

the critics who believe that in Anc. Britain the identity of Portus Itius was finally
e ;tablished. I can only claim to have brought the question as near solution as,
in the present state of knowledge, is possible. ^ B. G., i,
5, § 4.
« Geogr., ed. C. Miiller, ii,
9, § 9. Cf. Miiller's note (p. 231) with Diet. arcJi.
de la Gaule, i, 77, d'Anvillc's Notice sur Vaticienne Gaule, pp. 97-9, and Walckc-
nacr' s G'eof/r. des Gatdes, i, 521. Artzenheini, near Markolshcim, which d'Anville
and Walckonaor identify with Argentovaria, is about 5 miles south of Hoidols-
heim.
 LATOBRIGI 439

J. D. Schoepflin ^ remarks that Pliny ^ places the Rauraci between

the Sequani and the Helvetii and he infers that they occupied the
;

slopes of the Jura, their territory being bounded on the east by the
Aar and the Rhine, on the south by the ridge of the Jura between
the sources of the Birse and the Pierre -Pertuis, and on the west by
the branch of the Jura known as Lomont. As a matter of fact, Pliny
only mentions the Rauraci between the Sequani and the Helvetii,
which, as any careful reader of the Natural History will admit, is no
proof that they separated the two peoples and Caesar distinctly
;

says that the Jura separated the Helvetii from the Sequani.^ Ptolemy,
Schoepflin argues, is mistaken in assigning Argentovaria to the
Rauraci their territory did not extend so far northward
: other- ;

wise the territory of the Sequani would not have touched the Rhine ;

and Caesar says that it did. The passage which Schoepflin has in
mind * is perhaps an interpolation ^ but let us assume that it is
;

true. Even so, if the territory on the north of Argentovaria is

rightly assigned to the Triboci (q.v.), and if they crossed the Rhine
with Ariovistus, Schoepfhn's argument collapses for Ariovistus's
;

followers settled in the territory of the Sequani.^

C. Martin argues that the Rauraci could not have been clients
'^

of the Sequani, because Caesar, in his enumeration of the tribal levies

which were raised for the relief of Alesia, couples the Rauraci with
the Boi and mentions them quite apart from the Sequani whereas ;

he mentions the clients of the Aedui and of the Arverni respectively

in the same breath, as it were, with those two tribes.^ This reason-
ing appears to me to be sound, and also to prove that the Rauraci
were not, as W. Gisi^ supposes, a pagus of the Sequani. On the
other hand, L. W. Ravenez ^^ argues that they must have been clients
of the Sequani, because Caesar does not mention them in B. G.,
iv, 10 among the tribes who dwelt on the banks of the Rhine. This
famous chapter is probably spurious ^^ nor does the writer mention
;

among those tribes the Nemetes or the Eburones or the Menapii.

In B. G., vi, 25, §§ 1-2, which also appears to be an interpolation,!^
we read that the Hercynian forest, which was wholly in Germany,
was conterminous with the territory of the Rauraci and Schneider i^ ;

infers that when they left their country to join the Helvetian emigra-
tion, some of them remained at home. But the writer who described
the Hercynian forest also says that it was conterminous with the
territory of the Helvetii, which, if we may accept Caesar's express
statement,!* was wholly on the left bank of the Rhine and therefore, ;

^ Alsatia illustrata, i, 1751, p. 37, § vi.

2 Nat. Hist, iv, 17 (31), § 106.
' E.G., 1,2,^3. "/&., 1,§5.
' See p. 394, n. 6. « B. Q., i, 31,
§ 10.
' Questions alsaciennes, 1867, p. 18.
« B. G., vii,
75, §§ 2-3.
' Anzeiger filr schweizerischc Alterthumskunde, 1884,
p. 82.
i»
L' Alsace illustree, i, 1849, pp. 380-1. ^^ See
p. 692.
*^ Jahresb.
d. philoL Vcreins zu Berlin, xxxvi, 1910, pp. 27-9. Cf. A. Klotz.
Caesarstudicn, pp. 50-4, 145.
'^
Caesar, i, 57. ^*
B. G., i, 2, § 3.
440 LATOBRIGI
if any conclusion can be drawn from the doubtful passage, it is
was wholly on the same
possible that the territory of the Rauraci
bank.
Martin argues further that the Latobrigi, Rauraci, and Tulingi
must all have been conterminous with the Helvetii,^ because Caesar
calls them neighbours (finitimi) of that people that they were ;

Gauls, not Germans and therefore that their territories must have
;

been entirely on the left bank of the Rhine. The fact, he goes on to
say, that Caesar expressly says that the Boi, who were associated
with the Latobrigi, the Rauraci, and the Tulingi in their emigration,
dwelt on the eastern bank of the Rhine, proves by implication that
the other three tribes did not. Following the indications of Ptolemy,
he says that the territory of the Rauraci extended along the western
bank of the Rhine, from the Helvetian frontier as far as the northern
frontier of Upper Alsace, and was bounded on the west by the 111.
He admits that there are no texts which show directly the where-
abouts of the Latobrigi and Tulingi but he thinks that local names,
:

such as Larg, Oberlarg, Largitzen, Thur, Thuringheim, and Thur-

bourg, indicate their position and he maintains that in the tenth
;

century the Tulingi were called Thuringi. Accordingly he places the

Tulingi in the valley of the Thur, between the 111, the Doller, the
Vosges, and the Eckenbach and the Latobrigi between the Vosges,
;

the Doller, and the 111.^

These arguments are worth nothing. Caesar often uses the word
finitimus loosely."' If one of the three tribes had been conterminous
with the Helvetii, and the other two had been at all near that one,
he would not have hesitated to call them all three finitimi of the
Helvetii. Indeed, if none of the three had been, strictly speaking,
conterminous with the Helvetii, he might have called them finitimi
of that people for he calls the Santoni, who dwelt in the Charente-
;

Inferieure, finitimi of the Provincia.* In the argument based upon

his statement about the Boi there is a contradiction, which Martin
overlooks if the Boi, who, on his own showing, were undoubtedly
:

Gauls, dwelt on the eastern bank of the Rhine, why should not the
other tribes have done so too ? Besides, Martin forgets that the
Menapii, w^ho were also Gauls, had territory on both banks of the
Rhine.^ Doubtful similarity in names proves nothing. There is not
much resemblance between Larg and Latobrigi^ and there is no proof
that the Thuringi were the same people as the Tulingi, or that, if they
were, the Thuringi of the tenth century occupied the same territory
as the Tulingi of Caesar's time. Walckenaer,^ who is followed by
Napoleon,' places the Tulingi in the south of the Grand Duchy of
Baden, and believes that Stllhlingen, a town near Schaffhausen,

^ Martin does not mean exactly what lie says for on his own showing, the
;

Latobrigi and the Tulingi were not conterminous with the Helvetii.
^ Questions alsaciennes,
pp. 4-5, 8-12, 16.
See B. G., i, 10, § 2
=*
iii, 7, § 3 ; 20, § 2
; vii, 7, § 5. ;

« lb., i, 10, § i^- , If^; iv, 4, § 3.

^ 6 Geogr. ties Gaules, i, 559.
' Hist, dc Jules Cesar, ii, 46, n. 3. Cf. Ccntralblatt fUr Anthropologic, Ethno-
logie und Urgeschichte, 1897, p. 69.
 ;

LATOBRIGI 441

derives its name from theirs. Walckenaer and Martin cannot both
be right, and may both be wrong.
Walckenaer,! arguing from similarity of names, places the Lato-
brigi in the neighbourhood of Brugge, on the rivers Brege and
Briggach, tributaries of the Danube but the French Commission,
;
'-^

perhaps more wisely, confess ignorance. The form Latohrigi, on

which Walckenaer's conjecture is based, is not certain Gliick ^ :

prefers Latovici A. Holder, who adopted that form in his edition

;
"*

of Caesar, now reads Latohrigi. Cluver ^ rejects Walckenaer's view

(which had been advocated long before Walckenaer's time) for, he ;

says, it is difficult to see who the Germans that were conterminous

with the Helvetii ^ could have been unless they dwelt in the valley
of the upper Ehine, between Lake Constance and the Aar, that is to
say in the district which Walckenaer assigns to the Latohrigi. He
goes on to argue that there was no room for either the Tulingi or the
Latohrigi beyond the Jura, and therefore that they must have dwelt
either on the east or on the south of the Helvetii. He concludes that
the Latohrigi dwelt in that part of the Valais which was not occupied
by the Nantuates, Veragri, and Seduni, namely on the east of the
last-named people, in the district round Brige and the only place
;

which he can find left for the Tulingi is in the valley of the Khine,
above Lake Constance.
I am not concerned to defend Walckenaer, who simply made
a guess but Cluver's argument does not refute him. If the Lato-
;

hrigi were a small tribe, there was room for them as well as the
Germans between Lake Constance and the Aar and when Caesar ;

said that the Ehine separated the Helvetii from the Germans, he
may have been thinking of the Germans who dwelt on the north of
Lake Constance, through which the Rhine flows. Besides, Caesar's
geographical statements are often loose As for the Tulingi, they may
."^

have been where Cluver places them or they may not. And if Lato-
vici, not Latohrigi, is the true form, Cluver's Brige helps us no ' '

more than Walckenaer's Brege '. '

To conclude. It is impossible to do more than guess at the positions

of the Latohrigi and Tulingi but the conjecture which assigns to
;

the latter the country round Stiihlingen appears to me probable. As

for the Rauraci, the evidence of Ptolemy is not decisive for, like ;

the Ubii and other Rhenish peoples, they may have changed their
abode before Ptolemy wrote or their territory may have been
;

smaller in Caesar's time than in Ptolemy's. The solution of the

puzzle may be that, as Napoleon holds, they dwelt, in Caesar's time,
on both banks of the Rhine. I find it difficult to believe that any of
the three peoples could have been clients or fagi of the Sequani

'
Geogr. des Gaules, i, 559-00.
^ Diet. arch, de la Gaule, ii, 13.
^ Die bei Cdsar vorkommenden keltischen Namen, See p. 844,
1857, p. 112.
infra, and P. Goycr in Jahresb. d. pliilol. Vereins zu Berlin, v, 1879, j)- 333.
* Alt-cellischer ^pracliscluilz, ii,
155.
^ Germania antiqua, 1031, pp, 358-9.
« B. G., i, 2, § 3. ' See p. 344.
442 LATOBRIGI
for if they had been, it is unlikely that the Sequani would have

acquiesced in their emigration.

—
Lemovices. The Lemovices 23ossessed the dioceses of Limoges
and of Tulle, which was severed in 1318 from the ancient diocese of
Limoges.^ This territory corresponded roughly with the depart-
ments of the Haute-Vienne, Correze, and Creuse.
—
Lemovices Aremorici (?). In the enumeration of the states
that were called upon to furnish contingents for the relief of Ver-
cingetorix are mentioned two tribes which bear the name Lemovices.
The position of one has just been defined. The name of the other
occurs in the enumeration of the Aremorican tribes."^ This second
tribe is mentioned nowhere else by Caesar, and not at all by any
other writer and the appearance of the name in this passage has
;

given rise to much discussion. Davies, Oudendorp, and Schneider

retain the word because it is found in all the MSS. Nipperdey ^ con-
siders that Lemovices was written by some copyist in mistake for
Lexovii. Desjardins,* following Frigell, proposes to substitute
Namnetes for Lemovices and Veneti, which appear side by side in
Caesar's list, —
an idle conjecture, which leaves the question exactly
where it was. De Valois ^ is inclined to read Leonnenses, the assumed
name of the Gallic inhabitants of the district of Leon in southern
Brittany and d'Anville ^ approves of the conjecture, though for
;

Leonnenses he would substitute Leonnices, pour s'ecarter d'autant

'

moins de ce qui est ecrit Lemovices.'* Walckenaer remarks that some '^

of the MSS. of Ptolemy assign the town of Ratiatum to the Aljxovlkol

(Lemovices), while others give the Pictones two chief towns Limo- —
—
nuni (Poitiers) and Ratiatum and mention Awjustoritum (Limoges)
as the capital of the Lemovices.^ Ne doit-on pas croire,' he asks
'

'
d'apres cela, que les Limovici etaient probablement mentionnes
deux fois dans Ptolemee comme dans Cesar, et que les copistes,
yyant considere cette double mention comme une faute, auront fait
'
disparaitre un des deux Limovici ?
Maximin Deloche ^ develops Walckenaer's suggestion. He rejects
the emendations of de Valois and dAnville, because there is no
evidence of the existence of any ancient Gallic tribe called Leon-
nenses or Leonnices. He goes on to say, in pursuance of Walckenaer's
argument, that the MS. of Ptolemy known as A (No. 1401 of the
Bibliotheque nationale) wrongly assigns Ratiatum to the Lemovices
and both Limonum and Augustoritum to the Pictones. Ces trans- '

positions,' he observes, sont d'autant plus etranges, que le para-

'

graphe relatif aux Pictones (§ 5) est separe du paragraphe des Limo-

vices (§ 9) par les paragraphes 6, 7 et 8, et qu'il a fallu un motif

1 See d'Anville, Notice de Vancienne Gaule, p, 407, and Did. arch, de la Gaule,
ii, 82.
2 B. G., vii, 75, §§ 3-4. ^ Caesar,
pp. 107-8.
* Geogr. de la Gaule rom., ii, 705-6. * Notitia Galliarum, p. 269.
^ Notice de Vancienne Gaule, p. 408. ' Gcogr. des Gaules, 1, 369.

« Geogr., ii, 7, §§ 5, 9. See C. Miillcr's ed., pp. 202-4.

'•*
Mem. de la i^oc. impcriale des ant. de France, xxiii, 1857, pp. 51, 56-63,
67-78.
 LEMO VICES AREMORICI (?) 443

particulier pour que le copiste du manuscrit A placat Ratiatum chez

les Limovices ; et on ne peut s'expliquer cette circonstance qu'en
admettant, suivant la conjecture de M. Walckenaer, que les Limo-
vices se trouvaient nommes dans le paragraphe des Pictones.' This
explanation seems to me unconvincing. If, as Walckenaer and
Deloche maintain, Ptolemy mentioned a people called Lemovices in
§ 5 and another people of the same name in § 9, how are we to account
for the fact that none of the MSS. exhibit, in these two sections, the
names of any towns except Ratiatum and Limonum, which unques-
tionably belonged to the Pictones, and Augustoritum, which un-
questionably belonged to the Lemovices of Limoges ? To quote
C. Miiller, the best editor of Ptolemy (p. 202), Confusio ista forte
'

e tabula fluxit in qua Ratiati et Augustoriti nomina permutata erant.

Similes errores Aquitaniae tabula habet in codice A, ubi positioni in
Cadurcorum finibus notatae Mediolanum (quod est Santonum opp.)
adscribitur, dum vera Mediolani positio indicatur quidem, sed nomine
caret,' &c.
Deloche then proceeds to search for the Aremorican Lemovices.
From the fact that Caesar has not distinguished them by any surname
analogous to that of the Aulerci Cenomani, he concludes that they
were merely a branch of the inland Lemovices. We learn, he says,
from the anonymous author of La Vie de St. Waast, who flourished
about A. D. 667, that there was once in western Gaul a people called
Leuci. Scattered over a strip of territory which extends from a little
to the east of Limoges along a chain of heights in a westerly, and
then a north-westerly direction to the Bay of Biscay, near St. Jean-
des-Monts, are to be found vestiges of this people and proofs of their
Lemovician origin. Such are the names La Mothe- Limousin, Le Buy
Limousin, La Limouziniere, &c. Appealing to the anthropological
observations of J. M. Dufour,i Deloche maintains that the aforesaid
chain of heights was inhabited by a race different from the neigh-
bouring Pictones ; and from the evidence of nomenclature, speci-
mens of which I have just given, he concludes that this race was the
Lemovices Aremorici. Finally, to the anticipated objection that
this theory requires a curtailment of the territory of the Pictones,
he replies that it ought to be curtailed for the Pictones only con-
;

tributed 8,000 men to the army destined for the relief of Vercinge-
torix, whereas the Aremorican Lemovices contributed 10,000.
Deloche's anthropological argument is absolutely worthless and it ;

is certain that, if he had studied the ethnology of Gaul as a whole, he

would never have adduced it. Still, if there were two tribes of
Bituriges and three tribes of Aulerci, there may also have been two
tribes of Lemovices. Limousin, the name of the province which
roughly corresponded with the territory of the inland Lemovices,
is unquestionably derived from their name and it is certainly
;

remarkable that in the territory which Deloche assigns to their

alleged Aremorican kinsmen there are two places bearing a name of
which the same word forms a part, as well as a third of which the

'
DcVancicn Fuitou, 1820, pp. 111-2.
444 LEMOVICES AREMORICI(?)
name the obviously analogous Limouziniere. But Deloche has not
is
proved his case and it is not capable of proof.^ I have therefore
;

not marked the territory of the alleged Aremorican Lemovices on

my map.
Leuci. —
The Leuci, in the opinion of the French Commission,
possessed the ancient dioceses of Verdun and Toul, as they were
before the dioceses of St. Die and Nancy were severed from them, or
the department of the Vosges and the southern parts of the depart-
ments of the Meuse and Meurthe-et-Moselle.^ The diocese of Verdun
corresponded with the territory of the Veroduneiises, who are not
mentioned by Caesar and I am inclined to agree with Walckenaer
;

and M. Longnon that they were a pagus or clients of the Mediomatrici

(q.v.), not of the Leuci but it is impossible to decide the question.
:

—
Levaci. See Nervii.
—
Lexovii. The Lexovii certainly possessed the diocese of Lisieux ^ :

the only question is whether they did not possess something more.
The members of the French Commission were not unanimous the ;
'*

majority decided that to the diocese of Lisieux must be added the

dioceses of Bayeux and Seez, because Caesar does not mention either
the Viducasses or the Baiocasses,^ both of whom must therefore, in
their opinion, have been clients of the Lexovii and because he ;

couples the Lexovii, without mentioning any intermediate people,

with the Venelli, while between the known territory of the Venelli
and the diocese of Lisieux intervene the territories of the Viducasses
and Baiocasses. M. Longnon is inclined to agree.^ He also considers
that the Lexovii are identical with the Esuvii but in my notice of
;

the Esuvii (q.v.) I have combated this opinion. Desjardins identifies

the joint territories of the Viducasses and Baiocasses with the
territory of the Esuvii. He argues further that, as Caesar does not
mention the Lexovii in B. G., vii, 75 among the maritime peoples,
their territory probably did not touch the sea.'' But this view is
refuted (1) by the fact that no known people intervened between
the Lexovii and the sea, and (2) by the fact that the Lexovii are
placed by Strabo ^ and Ptolemy ^ at the mouth of the Seine, and by
Strabo, in another passage, on the sea-coast. Moreover, it is doubtful
whether Caesar did not mention the Lexovii in the passage to which
Desjardins refers.^^
—
Lingones. The Lingones possessed the diocese of Langres, as it
was before the diocese of Dijon was severed from it.^^ To this territory

^ The French Commission {Diet. arch, de la Gaule, ii, 82) agree with Deloche.
^ Diet. arch, de la Gaule, ii, 87.
* D'Anville, Notice de Vancienne Gaule, p. 413 ; Walckenaer, Geogr. des
Gaules, i, 394-5.
^ Diet. arch, de la Gaule, i, 114, ii, 90-1 ; Rev. arch., nouv. ser., ix, 1864, p. 408.

^ Tlie Baiocasses are not mentioned under that name by any author before the

compiler of the Notitia provinciarum, ; but Pliny {Xat. Hist., iv, 18 [32], § 107)
doubtless referred to them when he wrote Bodiocasses.
^ Atlas hist, de la France, ' Geogr. de la Gaule rom., ii, 489.
p. 5.
«
14; 3, §5.
iv, 1, § « 6'eoi/r., ii, 8, § 5.
'"
Nipperdey's Caesar, pp. 107-8, and my article on the Lemovices
»Sce
Aremorici. ^^ D'Anville, Notice de Vancienne Gaule, p. 417.
 ;

LINGONES 445

M. Longnoni adds the dioceses of Troyes and Chalons-sur-Marne,

representing the territories of the Tricasses and the Catiivellauni,
two tribes not mentioned by Caesar, whom, contrary to the prevailing
opinion, he regards as cHents of the Lingones. He argues that his view
is the only one that can be reconciled with the passage in which

Strabo ^ says that the Lingones were conterminous on the north with
the Mediomatrici (q.v.). I am not sure that Strabo's words necessarily
mean this but if they do, he may have made a mistake, as Long
;

believes.^ For the territory of the Catuvellauni is generally assigned

to the Remi (q.v.) the Tricasses were probably either clients of the
;

Remi or a pagus of the Senones (q.v.) and, on M. Longnon's theory,

;

the territory of the Remi appears to be unduly small * and that of

the Lingones unduly large.
Von Goler ^ maintains that the diocese of Dijon must have be-
longed, in Caesar's time, not to the Lingones but to the Mandubii
for otherwise, he argues, the Helvetii would not have taken four days
to march from the neighbourhood of Bibracte to the frontier of the
Lingones. But von Goler made the mistake of placing the scene
of the Helvetian defeat at Chateau Chinon, on the north, instead of
on the south of Bibracte. From the site which Colonel Stoffel rightly
identifies with the battle-field ^ to the nearest point of the frontier
of the Lingones as ordinarily traced, the distance, in a direct line, is
over 50 miles. The distance actually traversed would have been, say,
a fifth longer than the distance in a direct line and there is no
;

evidence that the Helvetii marched towards the nearest point of the
frontier.'^ Von Goler's argument therefore collapses.
—
Magetobriga. Regarding the site of Magetobriga, the scene of
the decisive victory which Ariovistus gained over the Aedui and their
allies,^ enough has been written to stock a small library. But it all
amounts to guess-work, more or less ingenious for Caesar gives us no
;

indication whatever of the site, and thus there is nothing to go upon,

except the name.^
D An ville,^^ following F. Dunod,^ identifies Amagetobriga (sic) with
'

Broie, close to a marsh called Moigte-de-Broie, near the confluence

of the Oignon and the Saone. At Moigte-de-Broie, says d'Anville,
a piece of pottery was found, bearing the inscription Mag. Etob :

but this piece of evidence, such as it was, has long been generally dis-
credited for, if Walckenaer ^^ is to be believed, on a eu soin de [le]
;
'

perdre presque aussitot apres I'avoir trouve.' Not content, however,

with exposing the futility of d'Anville's conjecture, Walckenaer must
needs make one of his own and his choice falls upon Amage, near
;

Luxeuil. M. A. Berget^^ has recently attempted to revive the theory,

^ Atlas hist, de la France, pp. 5, 6.
^ vnep Tuiv MediofxaTpLKaiv AevKoi koi tmv Aiyyovcuv ti ixepo7. iv, 3, § 4.
^ W. Smith's Diet, of Greek and Rom. Geogr., ii, 194.
^ See B. G., vi, 12, §§ 7-9. & Gall. Krieg, 1880,
pp. 331-2.
« See pp. 625-7. ' See
pp. 631-4. « B. G., i, 31,
§ 12.
i"
^ See pp. 844-5. Eclair cissemens sur Vancienne GavJe, p. 165,
^^ Hist, des Sequanais, i, 1735, pp. 92-5.
^2 Geogr. des Gaules, i, 319-20.
^^ Bull, archeol. du Comile des travaux hist., &c., 1908, pp. 108-16.
446 MAGETOBRTGA
propounded more than a century ago, that Admagetobriga {sic) was
a fort on Mont Ardov, on the left bank of the Saone, close to Pont-
ailler. He urges that Gallic antiquities have been discovered on the
site, and that the speech which Caesar delivered at Vesontio^ justifies
the inference that Admagetobriga was a stronghold, probably near
the common frontier of the Aedui and the Lingones, and on the
banks of the SaOne. Does it ? Adtnagetobrigae, if that were the right
reading, would mean in Magetobriga ^ '
and, as M. d'Arbois de
'
;

Jubainville dryly remarks, il parait peu vraisemblable que la grande

'

bataille ait ete livree dans I'interieur d'une ville gauloise.' ^

. . .

Most absurd of all, Desjardins, who is never tired of ridiculing the

antiquaries who waste their time and ingenuity in the conjectural
restoration of Gallic geography, dogmatically asserts that Admageto-
briga was close to Broie, without troubling himself to produce any
evidence or to offer any argument in support of his. assertion.*
And now, absurd as it may seem, I am going to have my say. Of
all the guesses d'Anville's is the least likely to be wrong. I give
reasons, on page 845, for believing that the true form of the name is
not Admagetobriga but Magetobriga or Magetobria. That questionable
piece of pottery may after all have been really lost and if it ever
;

existed, it was evidence of a sort.^ Anyhow there is no other, good,

bad or indifferent.^ But to my mind the mere fact that Ariovistus
defeated the Gauls is enough for Caesar does not describe the battle,
;

he merely registers it.

—
Mandubii. The Mandubii possessed the stronghold of Alesia (q.v.),
or Mont Auxois. Their territory, therefore, comprised part of the
department of the Cote-d'Or, but how much it is impossible to tell.
Strabo says that they were neighbours of the Arverni, an obvious
'^
—
blunder.^ A. de Barthelemy ^ thinks that the manner in which
Vercingetorix was received by the Mandubii in Alesia proves that
they were an independent people. I am quite unable to understand
this argument. The Mandubii admitted Vercingetorix into Alesia
either because they had voluntarily joined in the rebellion, or because
they were a pagus or clients of the Aedui or under their influence,
or because Vercingetorix compelled them to admit him. The P'rench

B. G., i, 40, § 8.
1

See Jahresb. d. philol. Vereins zu Berlin, xx, 1894, p. 287.

2

Rev. celt., xxx, 1909, pp. 112-3.

=»

* Geogr. de la Gaule rom., ii, 355.

^ F. Fiedler indeed says {Geogr. d. transalpin. Galliens, 1828, pp. 45-6), appa-

rently on the authority of some members of the Academy of Dijon, that the
piece of pottery was still preserved at that town at the time when he wrote
(1828 ?) but I cannot find any confirmation of his statement.
;

^ Cf. C. Jullian, Hist, de la Gaule, iii, 157, n. 4. ' iv,

2, § 3.
^ Desjardins {Geogr. de la Gaule row., ii, 468) suggests that the Mandubii may

have been neighbours of the Arverni at the time when Vercingetorix was at the
height of his power, and when he may have extended the hegemony of the
Arverni over the territory of the Aedui. But it is much more Hkely that Strabo
made a mistake. On Desjardins's theory, if Vercingetorix had extended his
hegemony over North -Western Gaul, the Osisnii, who lived near Brest, niiglit
have been called neighbours of the Arverni.
^ Bev, des questions hist,, iii, 1867, p. 46.
 —

MANDUBII 447

Commission,! following the principle of the dioceses, includes their

territory in that of the Aedui. D'Anville,^ in support of the same
view, cites Hericus, who says, referring to Alesia,
Te fines Aeduos et limina sacra tuentem.^

The geographical accuracy of a mediaeval monk, who wrote in verse,

may be questioned but, assuming that Hericus was well informed,
;

his words may only mean that Alesia was near the Aeduan frontier,
not necessarily that it was in their territory. It seems morally
certain, however, that the Mandubii were clients of the Aedui.*
—
Mediomatrici. The Mediomatrici are mentioned by the writer
of B. G., iv, 10,^ between the Sequani and the Triboci, among the
peoples whose territories bordered on the Khine but if that state- ;

ment holds good for the time of Caesar, they must have possessed
the country round Worms and Spires, which, in Ptolemy's time, was
occupied by the Vangiones and the Nemetes. Their chief town
was Divodurum ^ (Metz). Their neighbours on the north were the
Treveri (q.v.), on the west the Kemi, on the south the Leuci and the
Sequani. D'Anville' believes that they were separated from the
Remi by a people called the Verodunenses, whose name survives in
'
Verdun ', and who are not mentioned in any document earlier than
the Notitia provinciarum for, he remarks, a place called
; Fines ', '

between Virodunum (Verdun) and Divodurum, is noticed in the

Itinerary of Antonine ^ but, as the Verodunenses were not men-
:

tioned by Caesar, their territory is generally included in that of the

Mediomatrici. According to Strabo,^ the Triboci (q.v.) had settled in
the country of the Mediomatrici but the Triboci were one of the
:

tribes who fought under Ariovistus and Caesar says that the
;

followers of Ariovistus had settled in the country of the Sequani.^^

—
Meldi. A tribe called Meldi is mentioned by Strabo,i! Pliny ,^2
Ptolemy,!^ and in the Notitia provinciarum i* and it is universally ;

admitted that this people occupied the diocese of Meaux, that is to

say, the northern part of the department of the Seine-et-Marne and
a fraction of the south-eastern part of the department of the Oise.
It has, however, been denied that Caesar's Meldi were the same
people. Strabo puts the Meldi next the Lexovii and says that they
were a maritime tribe, a blunder into which he may have been led
by misunderstanding the passage in Caesar which I am about to
discuss. Caesar mentions the Meldi once only, when describing the
preparations which he made for his second expedition against

1Did. arch, de la Gaule, i, 14. ^ Notice de Vancienne Gaule,

p. 431.
^Patrologiae cursus completus, ed. J. P. Migne, t. cxxiv, p. 1178, 1. 106.
* Cf. C. JulHan, Hist, de la Gaule, ii, 538, n. 5. 0. Hirschfeld {Corpus inscr.
Lat., xiii, pars i, p. 439), referring to B. G., vii, 90, § 1 his rebus confectis (i. e.
after the capture of Alesia) [Caesar] in Haeduos proficiscitur argues that the —
Mandubii were not in Aeduan territory.
5 See p. 692. « Ptolemy, Geogr., ii, 9, § 7.
' Notice de Vancienne Gaule, p. 692. ^ Ed. Wesseling, p. 364.
' iv, 3, §4. '« 5. 6^., i, 31, §10; 51, §2.
" iv, 3, § 5. 12
]^af fjigi^ i^^ Ig (32)^ § 107.
" Geogr,, ii, 8, § 11.
J*
Ed. O. Seeck, p. 265 (iv, 9).
448 MELDT
Britain.^ On arriving at the Portus Itius (q.v.), where he had ordered
the whole of his fleet to assemble, he found that 60 ships, which had
been built in the country of the Meldi,^ had been prevented by
contrary winds from making the harbour, and had returned to the
place from which they had started. The Portus Itius must be
identified either with Boulogne or with Wissant. Strabo ^ says that
the naval arsenal was at the mouth of the Seine and as the legions, ;

by which the ships had been built, had wintered in the country of the
Belgae,* it is certain that all the ships had been built or repaired on
or east of the Seine. Long ^ argues that, as the wind had prevented
those ships which had been built in the country of the Meldi, but not
the others, from reaching the Portus Itius, and as the bulk of the
fleet must have been constructed south of that harbour, we may look
for the Meldi to the north of it. He goes on to say that the hypo- '

thesis of these ships being built on the Marne and carried down the
Seine is inadmissible. If Caesar had built ships on the Seine, he
would have built them low^er down. These ships of the Meldi
. . .

returned to the place from which they set sail and it is absurd to ;

suppose that they sailed back up the Seine and the Marne to the
country of the Meldi.' But Long's argument notwithstanding, these
ships may have sailed from the same side of the Portus Itius as the
rest ;for the wind which blew them back may not have arisen until
after the others had reached port.^ On the other hand, it certainly
seems unlikely that the ships should have been built so far from the
sea as in the neighbourhood of Meaux and it is more than unlikely
;

that, after having put out to sea, they should have sailed all the way
back again up the Seine and the Marne. P. N. Bonamy, how^ever,
—
remarked that in his time the middle of the eighteenth century
timber used in the construction of barges at Kouen came down the
Marne from the neighbourhood of Meaux. Moreover, the reader will
observe that Caesar does not say that the ships returned to any point
in the country of the Meldi. He simply says that they returned to
the point from which they had set sail (eodem unde erant frofedae
revertisse) and it seems possible that this point w^as at the mouth
;

of the Seine, where they may have remained for a time after they
had dropped down the river from the neighbourhood of Meaux. If,
however, this conjecture is correct, and if the rest of the ships were
also assembled at the mouth of the Seine, the ships that were driven

^ Certain inferior MSS. read Belgis (Schneider's Caesar, ii, 22) but this :

cannot be right for there would have been no point in saying that the ships in
;

question were built in the country of the Belgae, when the rest of the ships,
with which they were contrasted, were built there also. N. Sanson {Les Comm.
de Cesar, p. 55) makes the absurd conjecture that Caesar wrote not Meldis but
Venellis in other words, that he had ships built in the Cotentin, far away in
;

the west As Mommsen {Jahresh. d. philol. Vereins zu Berlin, xx, 1894, p. 205)
!

justly remarks, it is difficult to believe that Meldis the name of a people who—
—
are seldom mentioned should have been interpolated.
' iv, 3, § 3. * B. G., iv, 38, § 4. ^ Caesar,
pp. 221-2.
® I find that Bonamy used the same argument more than a century ago.
See 3Iem. de litt, tires des registres de V Acad, des inscr., &c., xxxi, 1708, p. 2"2().
 MELDI 449

back by the wind could not have sailed until after the rest, which
did not encounter contrary winds, had put to sea.^
D'Anville ^ conjectures that the Meldi inhabited the country round
Meld-felt, vulgairement Maldeghem-velt,' which, he affirms, means
'

Meldicus campus and nous transmet le nom des Meldi sans aucune
'

alteration'. I cannot find Meld-felt in any map but Maldeghem,

;

of which d'Anville was thinking, is about 10 miles east of Bruges.

Unhappily for his theory, Heller^ observes that, according to the
Academy of Amsterdam, Maldeghem is derived from the Dutch
plant-name, melde.
Creuly,'* who also regards the Meldi as a pagus, or sub -tribe, of the
Morini, advances the following arguments against identifying them
with the Meldi of Meaux. The Roman army, he says, was cantoned
in the neighbourhood of Cassel, Therouanne, and Amiens ^ they :

had plenty of timber on the spot to build the fleet and there was no
;

lack of ports along the coast of the North Sea. What reason, then,
asks Creuly, could there have been for having any ships built at
Meaux, nearly 400 miles away ? I cannot answer this question :
but Caesar may have had some reason, which he did not think it
necessary to state ^ and I should like to know what were the ports
;

which Creuly had in mind. Anyhow the notion that Caesar would
have established a dockyard on the North Sea in the neighbourhood
of Bruges is absurd and if the Meldi were only a pagus of the
;

Morini, why did not Caesar write Morinis ?

Reviewing the arguments on both sides of the question, I conclude
that Caesar's Meldi were identical with the Meldi of Meaux.'
Menapii. I. — We
learn from Caesar ^ and from Strabo ^ that tlie

^ Heller {Philologus, xxii, 1865, pp. 129-30) has an argument to prove that
the ships came from the mouth of the Seine, which rests upon the assumption
that the wind which drove them back was the Corus, Caesar {B. 0., v, 7, § 3)
says that, after he reached the Portus Itius, he was prevented from sailing
for Britain for between three and four weeks by the corus ventus, which'

commonly blows throughout a great part of the year on these coasts {qiii '

magnam 'partem omnis temporis in his locis flare consuevit). The corus, according
to Pliny {Nat. Hist., ii, 47 [46], § 119), blew from the quarter where the sun
sets at the solstice apparently it was from a point between NW. and W. by
:

N. ^ N. (Rice Holmes, Anc. Britain, p. 555, n. 2). Obviously it is impossible

to prove that this was the wind that blew back the 60 ships. Assuming that
it was, Heller remarks that such a wind would not have driven back ships
coming from the neighbourhood of Ostend to the Pas de Calais, but that it
would have blown full against ships which were trying to get out of the mouth
of the Seine. But no ancient ships could have sailed from Ostend to the
Pas de Calais in the teeth of a NW. or WNW. wind.
^ Notice de Vancienne Gaule,
p. 452.
^ Zeitschr. f. allg. Erdkunde, xviii, 1865, p. 186.
* Rev. arch., nouv. ser., viii, 1863, pp. 385, 387.
^ They were certainly cantoned in the country of the Belgae, along or near

the northern coast.

" No commentator, as far as I know, has remembered, in discussing this

question, that during the civil war Caesar had ships built at Hispalis, the modern
Seville, which is quite 70 Roman miles by river from the sea. B. C, ii, 18, § 1.
' In the first edition
(p. 456) I was rather inclined to adopt the opposite
opinion but, as I now see, I had not completely thought the matter out.
;

« B. G., iv, « iv,

1, § 1 ; 4, § 2. 3, § 4.
1093 G g
450 MENAPII
Menapii possessed land on the right hank of the Rhine, not far from
the sea, as well as on the left, and from Caesar ^ that their territory
was conterminous on the soutli with that of the Eburones according :

to Strabo,^ Pliny ,^ and Dion Cassius,^ they were neighbours of the

Morini ^ while, according to Ptolemy ,6 the eastern frontier of the
;

Morini was the river Tabula. Walckenaer"^ argues that this was
the Aa, which, he remarks, was actually the frontier in the seventh
century. Every other ancient writer, he adds, who mentions the
Scheldt, calls it Scaldis and Ptolemy is the only pne who mentions
;

the Tabula. Walckenaer's theory, however, is combated by Plot ^

and Desjardins.^ The former remarks (1) that the Aa was always
called Agnio or Agniona and (2) that Ptolemy placed the Tungri ^^
;

on the right bank of the Tabula, a statement which, he says, is true

of the Scheldt, but not of the Aa. Desjardins explains that most
modern geographers identify the Tabula with the Scheldt, (1) because
Ptolemy places its mouth between Gesoriacum (Boulogne) and the
estuary of the Meuse, thus :

Gesoriacum 22° 30', 53° 30'

Tabulae 11. ost 23° 30', 53° 30'
Mosae fl. ost 24° 40', 53° 30'

(2) because, on any other hypothesis, Ptolemy does not mention the
Scheldt at all and (3) because the geographer, Ortelius, found in
;

mediaeval documents the name Tabula applied to the Scheldt. ^^ The

balance of probability, then, is in favour of identifying the Tabula
with the Scheldt. Pliny places the Menapii on the west of the Scheldt,
and the Toxandri on the north of it.^^ But Caesar does not mention
the Toxandri at all and it is certain that in his time the Menapii,
;

whether they possessed any land on the west of the Scheldt or not,
did possess land on the east of it. My conclusion is that the ancient
writers do not help us much to trace the boundary between the
Menapii and the Morini. Let us examine the other evidence.
II. Castellum Menapiorum, which is mentioned in the TahUP was
undoubtedly Cassel, in the department of the Nord, east of the Aa,
and about 11 miles north-east of St. Omer. Long^* insists that for
1 B. G., vi, 5, § 4. ^ iv,
3, § 5.
3 Nat. Hist., iv, 17 (31), § 106. * xxxix, 44,
§ 1.
^ This may be also gathered from Caesar, though he does not say it in so

many words. See B. G., ii, 4, § 9 iii, 28, § 1 ; iv, 22, § 5 34, §§ 1-3. Creuly
; ;

(Rev. arch., nouv. ser., vii, 1863, pp. 385-6) argues that the Menapii were
separated from the Morini by the Eburones, a view which I have examined —
in my note on the latter people.
* G^eog'r.,ii,9, §§ 4-5. ' Geogr. des Gaules, ii, 4i4:Q-l
^ Annales de la Soc. d^ emulation de Bruges, iv, 1869, p. 290, n. 1.
^ Geogr. de la Gaule rom., i, 137.
1° The Tungri occupied the territory which had formerly belonged to the
Eburones. See La Gazette numismatique, xii, 1907, pp. 41-6.
^^ What Ortelius actually says is this huic [i. e. by Ptolemy] (Scaldis)
:
'

TABVDA, TajSoCSa nominatur. Tabul, et Tabula etiam, apud scriptores media

a-etatis reperio.' Thesaurus geographims, 1587, under Scaldis.
^^ A Scalde incolunt extera Toxandri
'
Deinde Menapii, Morini.'
. . .

^^ La Table de Peuiinger, ed. Desjardins,

p. 13, col. 2.
^^
W. Smith's Diet, of Greek and Boman Geogr., i, 561. Cf. J. Ghesquiere,
Acta 88. Belgii, v, 1789, p. 603 (121).
 MENAPII 451

Menapioruni we should read Morinorum. If,' he says, '

we were to '

admit that the Menapii extended so far (westward J as Cassel, which is

improbable, we should not expect to find their Castellum there and ;

it is just the place where we might expect to find the Castellum of the
Morini.' This is hardly a sufficient reason for making such a correc-
tion as Long proposes and, as Walckenaer observes, it is stated
;

in the archives of the church of St. Pierre at Cassel that the town
was m
pago Memfisco. The statement in the Table harmonizes with
Walckenaer's theory that the Tabula— the eastern frontier, according
to Ptolemy, of the Morini was the Aa — but on the other hand,
:

Ptolemy mentions the Castellum of the Menapii and, as he makes ;

the Meuse the western frontier of the Menapii, and places the Tungri
between them and the Morini, it is clear that, unless he defined the
position of the Menapii wrongly, the Castellum which he mentions
was not Cassel.i Some writers believe that it was Kessel, on the left
bank of the Meuse, between Roermond and Venloo ^ but even on this ;

theory, Ptolemy's tracing of the frontier was wrong. At all events

the Morini possessed, in Ptolemy's time,^ Gesoriacum (Boulogne)
and Taruana (Therouanne) that is to say, their territory comprised
;

Bononiensis pagus and Tervanensis pagus. The latter extended

a little beyond the western and the southern frontiers of the modern
diocese of St. Omer on the other hand, it did not extend so far
:

as the eastern frontier of the original diocese, which comprised

Bourbourg and other places included within the pagus Mempiscus.
The diocese of Ypres, Walckenaer points out, also belonged to the
pagus Mempiscus. He therefore concludes that the Aa, in its whole
course, formed the eastern boundary of the Morini, and that their
southern boundary was formed by that of the diocese of St. Omer.*
A. G. B. Schayes ^ remarks that there was probably no definite
frontier between two tribes so little civilized as the Morini and the
Menapii. In another place,^ however, he maintains that, on the side
of the Morini and the Atrebates, the frontier of the Menapii must have
been marked by the rivers Scarpe, Deule, and Lys, which formed the
boundary of the pagus Mempiscus in the early Middle Ages.
A. de Vlaminck^ maintains that, in Caesar's time, the territory

^ Ptolemy places the Tungri on the east of the Tabula, and the Menapii

'beyond the Meuse' {ficTo. rov Moaav). General Creuly {Rev. arch., nouv. ser.,
viii, 1863, pp. 27-8) infers from these words that the Meuse was the eastern
boundary of the Menapii. But, to say nothing of the fact that, according to
Caesar, the Menapii possessed lands on both sides of the Rhine, Creuly mis-
translates Ptolemy. Just before mentioning the Menapii, Ptolemy says that
the Morini were ^uera the Ambiani,and that next to the Morini, /xerd the river
Tabula, were the Tungri. iura the river Tabula confessedly means on its
eastern bank surely then, when Ptolemy immediately afterwards says that
:

the Menapii were Aterd the Meuse, he means that they were on the east of that
river. See Miiller's ed. of Ptolemy, note to p. 223.
2 Diet, of Greek and Roman Geogr., i, 561. ^ Geogr., ii, 9, §§ 1, 4.
* See Walckenaer, Geogr. des Gaules, i, 441-3.
5 La Belgique et les Pays-Bas avant
'pendant la domination romaine, i,
et
1887, p. 38. 6
/^.^ p, 402.
' La Menapie, &c., 1879, pp. 13, 16, 23, 26-7, 78-81 ; Messager des sciences
hist, de Belgique, 1882, p. 427 ; 1884, pp. 441-2.

Gg2
452 MENAPIl
of the Menapii was much smaller than is commonly believed, and
accordingly he refuses to trace their western frontier along the
western boundary of the pagus Mempiscus. He refers to the small
number of the contingent, only 7,000 men,i which was levied from
the Menapii in 57 b. c, in comparison with the 25,000 contributed by
the Morini, and remarks that no authentic document older than the
Table of Peutinger indicates that the Menapii were established in
Flanders. A considerable part of the pagus Mempiscus belonged to
the diocese of Therouanne, that is to say to the country of the Morini.
The inference, he says, is that it was not until after Caesar's time that
the Menapii established themselves in that part of the country. He
also denies that there is any proof that the country round Cleve,
which Napoleon assigns to the Menapii, ever belonged to them on :

the contrary, he says, Caesar gives us to understand that it belonged

to the Eburones, the greater part of whose territory is between
'

the Meuse and the Rhine.' But Caesar's vague statement is quite
consistent with Napoleon's view and on page 690 I give reasons
;

for believing that Napoleon is right. De Vlaminck quotes Folquin

de Lobbes, a writer of the eleventh centuiy, who says that Flanders
belonged to the Morini and remarking that the supremacy which
;

Arras claimed in the Middle Ages over the communes of Flanders

doubtless originated in the transference by Caesar of the sovereignty
of the Morini to Commius, King of the Atrebates,^ he finds in this
circumstance a confirmation of the statement of Folquin. Finally,
he conjectures that the Morini, as a punishment for their repeated
revolts in the time of Caesar and in 29 b. c, were deprived of the
northern part of their original territory ;that the Castellum Mena-
piorum was not built by the Menapii until after their immigration
into these northern districts and therefore that no safe conclusion
;

can be drawn from the geographical position of Cassel as to the

original westward extension of the Menapian territory.
In a subsequent paper,^ de Vlaminck reinforces these arguments.
As Caesar, he remarks, informs us that, when invading the territory
of the Menapii in 53 b. c, he was obliged to build bridges,^ we may
conclude that the Menapii did not occupy Flanders, the southern
frontier of which is not protected by any large rivers. This is a feeble
argument. Caesar does not mention rivers, large or small. He only
says that he invaded the country of the Menapii, celeriter ejfectis
pontihus. Even small rivers would have had to be bridged, unless
they were fordable and Caesar says that the Menapii trusted for
;

protection to woods and marshes. Pontihus may mean cause- '

ways '5 and even if Caesar did invade the country on the east of
;

Flanders, the fact does not prove that Flanders also did not belong
to the Menapii.
Wauters, on the other hand, argues that when Caesar invaded the
country of the Menapii in 53 B. c, he must have gone to the west of

1 B. (?., ii,4, §9. - Ih., vii, 76, § 1.

^ Messager, &c., 1887, pp. 352-5. * B. G., vi, 6, § 1.
5 See B. G„ vii, 19, § 2.
 ;

MENAPII 453

the Scheldt, because he could not have made his bridges rapidly
enough in the country on the east of that river.^ But why not ? If
Caesar operated on the east of the Scheldt, we are not obliged to
assume that he crossed the Meuse and the Rhine and he would
;

have had no more difficulty in bridging the smaller streams on the

east of the Scheldt than those on the west. Wauters's argument is as
futile as his opponent's.
III. My conclusion is that it is impossible to trace the common
frontier of the Morini and the Menapii with certainty. If the Tabula
was the Scheldt, the evidence of Ptolemy is at variance with the
evidence of the Table and with the evidence of mediaeval documents.
I cannot see any sufficient reason for denying that the Castellum
Menafiorum of the Table really belonged to the Menapii but there ;

is no proof that it belonged to them in Caesar's time. De Vlaminck

has not proved his case but it:may be that, as he argues, the Morini,
after Caesar's time, were deprived of a part of their territory and ;

if this was the case, we can no more follow the western frontier of the

pajus Memfiscus than we can follow the eastern frontier of the

diocese of Therouanne. My own belief is that the Scheldt, as a natural
boundary, formed the common frontier of the two peoples. But all
that can be said with certainty is that the Menapii, in Caesar's time,
possessed lands on both sides of the Rhine, above its first bifurcation,
where the Usipetes and Tencteri encountered them,^ that is in the
neighbourhood of Cleve and Xanten that as the insula Batavorum,
;

which is mentioned by the writer of B. G., iv, 10, was apparently

not invaded by the Batavi before the time of Augustus,^ they
probably occupied it as well that their territory extended at least
;

as far westward as the Scheldt that it was bounded by the territory

;

of the Eburones on the south and by that of the Morini on the west
and that the territory of the Morini was bounded on the west by that
of the Ambiani and on the south by that of the Atrebates and possibly
also by that of the Nervii, all of them fairly well ascertained.
Morini. —§ee Menapii.
Namnetes. —The Namnetes occupied the ancient diocese of
Nantes, or that portion of the department of the Loire-Inferieure
which lies on the right bank of the Loire and is bounded on the north-
east by the river Semnon.* See Samnitae and Veneti.
—
Nantuates. The Nantuates, the Veragri, and the Seduni, reckon-
^ VAthenceum beige, 1883, p. 77.
^ B. G., iv, 1, §§ 1-2; 4. ^ g^e
p. 692, n. 2.
* D'Anville, Notice de Vancienne Gaule,
pp. 471-2; J. Loth, L^ Emigration
bretonne, p. 51. M. Jullian {Festschr. zu 0. Hirschfelds sechzigsiem Geburtstage,
1903, p. 214), observing that in Gaul rivers were very rarely frontiers, is inclined
to believe that the district of Retz, south of the Loire, also belonged to the
Namnetes. The statement of Strabo (iv, 1, § 1) that the Loire bounded [Gallo-
Roman] Aquitania, may, he thinks, have been only intended as approximately
true. But Strabo {ib., 2, § 1) also says that the Loire entered the sea between
the Namnetes and the Pictones : it seems unlikely that the Pictones, who
were much stronger than the Namnetes, would have suffered them to encroach
upon the southern bank ;and, even supposing that Strabo' s statements are
both inaccurate, I would ask M. Jullian to consider whether the Pictones
may not rather have possessed a strip of territory north of the I/oire.
454 NANTUATES
ing from west to east, dwelt in the valley of the upper Rhone. The
Nantuates occupied the territory which extended on the south of the
lake of Geneva as far west as the frontier of the Allobroges^ (q.v.).
What that frontier was is uncertain but, speaking roughly, the
;

Nantuates probably possessed the eastern part of the Chablais, or

that part of Upper Savoy which lies between the Valais and the river
Dranse, as well as the north-western part of the Valais, including
St. Maurice (Agaunum), which was their chief town.^ The Veragri
occupied the western part of the Valais, their chief town, Octodurus,
being situated between Martigny-la-Ville and Martigny-Bourg. The
name of the Seduni is preserved by the town of Sion.^
In B. G., iv, 10, § 3, it is said that the Rhine rises in the country
of the Lepontii, and flows through the countries of the Nantuates,
Helvetii, Sequani, Mediomatrici, Triboci, and Treveri {Rhenus autem
oritur ex Lepontiis, qui Alpes incolunt, et longo spatio per fines Nantua-
tiu?n, Ilelvetiorum, Sequanorum, Mediomatricorum, Tribocorum, Tre-
verorum citatus fertur). This passage, which is probably spurious,*
has caused the commentators much needless vexation of spirit. For
it is absolutely certain that, if the unknown writer did not make
a slip or a gross blunder, and if the Rhine did really flow through the
country of the Nantuates, these Nantuates were different different —
—
as a group, if not in blood from the. Nantuates w^hom I have just
mentioned and there is no evidence for fixing their whereabouts.^
;

Among the various readings which are found instead of Nantuatium

is Nemetum and if the interpolator did not appear to be enumerating
;

the tribes in question in geographical order from south to north,

I should say that this was what he wrote. Schneider ^ tries to recon-
cile the two statements in B. G., iii, 1 and iv, 10 by assuming that
the Nantuates inhabited a long strip of land extending eastward
from the lake of Geneva to the head- waters of the Rhine. But Caesar
clearly implies that the Nantuates whom he mentions in B. G., iii, 1,
§§1,4, and 6, § 5, were on the west of the Veragri. Desjardins sug-
gests that the Nantuates of the Rhine may have emigrated from the
Valais at the time when the Tigurini, who defeated Cassius Longinus,
emigrated from their original home." So they may ; but who can tell ?

' J5. 6?.,iii, 1,§L

- G. Deboinbourg {Rev. du Lyonnais, 3^ scr., ii, 1866, pp. 7-10) maintains
that the basin of the Dranse, which flows into the lake of Geneva between
Thonon and Evian, belonged not to the Nantuates but to the Allobroges, and
that the Nantuates dwelt not in tlie Chablais but only in the Valais. It is
unlikely, he argues, that they occupied both the eastern and the western slopes
of the mountains which separate the Valais from the Chablais and he affirms
;

that the western slope has always belonged to the diocese of Geneva, and the
eastern to the diocese of Martigny (Octodurus). But Martignj'- belonged to the
Veragri, not to the Nantuates and if the Nantuates possessed territory which
;

afterwards belonged to the diocese of Martigny, why should they not have
possessed territory which afterwards belonged to the diocese of Geneva ? See
my note on the Allobroges.
^ D'Anville, Notice de Vancienne Gaule,
pp. 472-3, 589-90, 639 ; Desjardins,
Geoyr. de la Gaule rom., ii, 234, 241-2. * See
p. 692.
^ Cf. C. Jullian, Hist, de la Gaule, ii, 521, n. 1.
^ Oaenar, i, 328 ^ Geogr. de la Gaule rum., ii, 239-41.
 NANTUATES 455

Mommsen offered a drastic solution of the difficulty. According

to him, Caesar thought that the upper Rhone, where it flows through
the Valais, was the Rhine, and that the Rhone had its source in the
lake of Geneva ^ The only argument that can be urged in support
!

of this astounding conjecture is that Caesar says that the lake of

Geneva flowed into the Rhone (lacu Lemanno, qui inflmnen Rhodanum
influit 2) and this was only his way of saying that the surplus waters
;

of the lake flowed off into the lower Rhone.^ In the first edition of
this book I criticized Mommsen's article, and the justice of the
criticism was acknowledged ;
"*
but I shall not reproduce it, for
Meusel us that
^ tells Mommsen, towards the end of his life, virtually
admitted his error.
—
Nemetes. The Nemetes are mentioned by Caesar^ among the
tribes who fought in the army of Ariovistus and he relates that ;

the few persons who survived the battle and the retreat recrossed the
Rhine7 They were established on the left bank in the neighbourhood
of Spires, in the time of Pliny ,s of Tacitus,^ and of Ptolemy ^^ but it ;

is possible that in Caesar's time none remained on the left bank after

the defeat of Ariovistus, because otherwise it would be necessary to

admit that the Nemetes who formed part of Ariovistus' s host
were only a fraction of those who followed him into Gaul. The
Nemetes mentioned by the later writers may only have been immi-
grants, descended from a portion of the tribe, which Ariovistus may
have left behind.
Desjardins ^^ says that, as the territories of the Sequani, the Medio-
matrici, and the Treveri extended to the left bank of the Rhine, it is
impossible to find room for the Nemetes on that side but, even if ;

we admit the doubtful evidence of the interpolated chapter i- in which

the Cisrhenane tribes are catalogued, this argument is not con-
clusive, because the host of Ariovistus, of whom the Nemetes formed
a part, had settled, before they encountered Caesar, on the territory
of the Sequani .1^
Mommsen ^^ says that Caesar '
left the Germans settled by
. . .

—
Ariovistus along the left bank of the Rhine the Triboci about Strass-
burg, the Nemetes about Spires, the Vangiones about Worms in —
possession of their new abodes, and entrusted them with the guard-
ing of the Rhine-frontier against their countrymen.' This view he

^ Hermes, xvi, 1881, pp. 445-6.

—
Es scheint vielmehr, dass fiir ihn der
'

Rhodanus aus deiii Genfer See kommt iind die obero Rhone ihm zwar bekannb
war, aber als der oberste Theil des Rheines gait.'
2 B. G., i, 8, § 1.
^ Similarly Caesar says {B. O., vii, 57, § 4) that the waters of a marsh or
marshy stream drained into the Seine. Various unnecessary attempts have
been made to amend the passage on which Mommsen bases his theory.
* Athenaeum, Jan. 13, 1900,
p. 42.
^ Jahresb. d. philol. Vereins zu Berlin, xxxvi,
1910, p. 25.
« B. G., i,
51, § 2. 7
7&,^ 53^ §§ 1_2.
« Nat. Hist., iv, 17 Germ., 28.
(31), § 106.
'>

'» Geogr., ii, ii

9, § 9. Gev(jr. de la Gaule rom., ii, 439, n. 5, 445.
12 B. G., iv, 10.
See p. 692.
'^ B. G., i, 31, § 10. See note on the Triboci.
1* Hist, uj Borne, v,
1894, pp. 48-9 {Rom. Gesc/i., iii, 1889, pp. 257-8).
456 NEMETES
defends in the following note :
—
That Ariovistus settled these
'

peoples on the middle Khine is probable, because they fight in his

army and do not appear earlier that Caesar left them in possession
;

of their settlements is probable, because he in presence of Ariovistus

declared himself ready to tolerate the Germans already settled in
Gaul [Caes., i, 35, 43), and because we find them afterwards in these
abodes.' I venture to say that these reasons are insufficient. Caesar,
in the presence of Divico, virtually declared himself ready to '

tolerate the Helvetii, provided that they gave him hostages for their
'

good behaviour ^ but after he had defeated them, h-e sent them back
;

to their own country. And if we find the Nemetes and the Triboci
'
afterwards in these abodes on the left bank of the Rhine, w^e also
'

find the Ubii afterwards on the left bank, whereas in Caesar's time
they were on the right.
I have not marked the Nemetes on my map, because I only profess
to represent Gaul as it was in the time of Caesar and while there is ;

no evidence to show what territory was occupied by the Nemetes

before the defeat of Ariovistus, it is doubtful whether any consider-
able number of them remained in Gaul after his defeat for we are :

told that his entire host took part in the campaign,^ and, as I have
said before, the few who survived the battle and the retreat recrossed
the Rhine. Still Caesar's words cum suis omnibus copiis may only —
mean '
with all his cannot agree with
[available] forces '
; and I
Walckenaer that the silence of Strabo proves that in Caesar's time
there were no Nemetes in Gaul after the defeat of Ariovistus. See
Triboci.
—
Nemetocenna. Nemetocenna, where Hirtius says that Caesar
wintered after his last campaign,^ is usually identified with Neme-
tacum,'* which stood upon the site of Arras. Desjardins asserts that
Nemetacum was distinct from Nemetocenna ^ but M. d'Arbois de ;

Jubainville shows that the former is simply a diminutive of the

latter.<5
Nervii. — The Nervii are not mentioned in the Notitia provinciarum :

but in their stead we find the civitas Camaracensium and accord- ;

"^

ingly it has been concluded that their territory corresponded with the
ancient diocese of Cambrai, which comprised Hainault, that part of
Brabant which lies west of the Demer and the Dyle, East Flanders,
and part of the province of Antwerp.^ D'Anville ^ makes the Nervian
territory extend to the sea round the mouth of the Scheldt, thus
separating the Morini from the Menapii. He also observes that, in the
Notitia dignitatum^^Nervicanustractusis mentioned as a continuation
'
B. G., i, 14, § 6. 2
ji,^ i^ 38^ § 1^ 3
/^,^ ^-iij^ 40, § 6
^ Ptolemy, Geogr., ii, 9, § 4 (Miiller's ed., i, 222).
^ Geogr. de la Gaule rom., ii, 724-5.
^ Mem. de Jn Soc. de Ungnistique, ix, 1896, p. 190. Cf. A. Holder, Alt-
celtischer Sprachschatz, ii, 708, 711.
'Ed. 0. Seeck. p. 2G6 (vi, 7).
Walckenaer, Geogr. des Gaules, i, 470-1.
^'

" Notice de Vancienne Gaule,

p. 482-3.
^"
Ed. 0. Seeck, p. 204 (xxxvii, 13). The Notitia actually mentions traclus
Armoricani et Nervicani.
 NERVII 457

of Armoricanus tractus. But Walckenaer replies that the term Ner-

vicanus tractus was loosely applied to the entire north-eastern sea-
board of Gaul, including the coasts of the Morini and the Menapii :

d'Anville's theory is opposed to a statement of Strabo,^ from which

we learn that the territories of the Morini and the Menapii were
conterminous and the testimony of Strabo is confirmed by Pliny .^
;

Still, it is possible that the territory of the Nervii may have extended
as far northward as the head of the estuary of the Scheldt, in which
case it might not have been considered as breaking the continuity
of the Morini and the Menapii. It might be urged, in support of this
view, that Caesar tells us that the Nervii, before encountering him
in 57 B. c, sent their non-combatants for safety in aestuaria,^ which
can only mean the low-lying tracts bordering the estuary of the
Scheldt.^ But it is not proved that the aestuaria were in Nervian
territory.

The
clients of the Nervii, namely the Ceutrones, Geidumni, Grudii,
Levaci, and Pleumoxii, are mentioned by no ancient writer, except
Caesar, and only once by him.^ He says that the Nervii, just before
they marched, on the instigation of Ambiorix, against Quintus Cicero,
sent messengers to summon their clients to join them, and advanced
with them to the attack. He also says that time was precious
to the Gauls ^ and from this it may possibly be inferred that the
;

territories of the client tribes were situated close together somewhere

in the neighbourhood of Cicero's camp. But if this inference is
correct, the statement upon which it is founded is the sole hint
which Caesar gives towards fixing their geographical position. The
aid of etymology has, however, been invoked. (1) Wauters places
the Ceutrones south of Chimai, in the neighbourhood of a hamlet
called Cendron while the French Commission ^ prefer to avow
;
"^

ignorance. (2) The position of the Geidumni is, so Wauters formerly

considered, indicated by the village of Gourdinne, in the canton of
Walcourt.^ But he admitted that this identification required that
the name of the tribe should be differently spelled and accordingly
;

for Geidumni he substituted Gorduni, a reading which is found in

certain inferior MSS.^o Subsequently, reverting to the reading
Geidumni, he selected Geidines, near Dinant and Desjardins ^^
;

tentatively follows him. D'Anville ^^ places the Gorduni, as he calls

them, near the dunes between Dunkirk and Ostend. But if he is
right, the Geidumni, though clients of the Nervii, dwelt in territory
that belonged either to the Morini or to the Menapii and the position
;

which he assigns to them is too far from any place that could be

Mv, 3, § 5; 4, § 2. ^ Nat. Hist., iv, 17

(31), § 106.
^ B. G., u, 28, § 1. See pp. 674-5.
" « B. G., v,
39, §§ 1, 4.
^ Omnem spem hostes in celeritate ponebant.
' Bull, de V Acad. Roy. de Bmxelles, xiii, 1862, pp. 396-7.
^ Diet. arch, de la Gaule, i, 249.
^ Bull, de VAcad. Roy. de Bruxelles, xiii, 1862, pp. 396-7.
'" See Sclineider's Caesar, ii, 133. " Geogr. de la Gaule mm., ii, 436, n. 8.
^^ Notice de Vanchnne Gaule, pp. 357-8.
458 NERVII
identified with the site of Cicero's camp.^ (3) The Grudii are placed
by d'Aiiville ^ which
in the canton of Groede, north-east of the tract
he assigns to the Geidumni and the French Commission ^ think
;

this a probable conjecture but it is open to the same objection

:

as d'Anville's choice in the case of the Geidumni, which the Com-

mission reject.'* Wauters finds their territory in the neighbourhood
of Graux and Desjardins ^ suggests that what Caesar wrote was
;

not Grudii but Gradii. (4) Sanson,^ who believes that the clients of
the Nervii were "pagi or sub -tribes of the Morini, assigns to the Levaci
—
the district of Loeuve, wherever that may be for Sanson's spell- ;

ing is probably different from the modern, and I cannot find the
place in the map. D'Anville ' finds an analogy between their name
and that of the river Lieva, which joins the Scheldt at Ghent. Others
point to the resemblance between their name and Louvain.^ Wauters^
places them entre I'Entre-Sambre-et-Meuse, oii on rencontre Lesves,
'

et le Brabant, oii plusieurs localites presentent la syllabe Lew '.

(5) The same writer ^^ considers that the name Pleiwioxii is preserved
in Moxhe and Moxheron, villages situated near the river Mehaignc.
A. Malengraen^i places this people in the neighbourhood of the
hill of Pleumont, near Chimay. Sanson ^^ gives them the country
round Peule, in which name his keen eye detects a resemblance to
Pleumoxii.
These conjectures must be taken for what they are worth. Some
of them may haply be right but they all rest upon a frail founda-
;

tion.13
Nitiobroges. — The
Nitiobroges occupied the diocese of Agen
(Aginuum and perhaps also that of Condom, which was severed
1^)

from it, that is to say, the greater part of the department of the
Lot-et-Garonne and a small fraction of that of the Tarn-et-Garonne.^^
The abbe A. Breuils, however, holds ^^ that they had no territory
south of the Garonne for, he argues, the Bituriges Vivisci possessed
;

lands on its left bank, and therefore if the river in its central course
did not separate the Celtae from the Aquitani, it did so nowhere,
and Caesar's statement in B. G., i, 1, § 2 is stultified. I rather doubt
wiiether this reason justifies the abbe in setting aside the principle
of the dioceses for, as I have already remarked,^'^ Caesar's state-
;

ments about frontiers were not always precise. It is true, indeed,

that if the Nitiobroges possessed the diocese of Condom, the territory
^ 8ee p. 383. ^ Notice de Vancienne Gaule, p. 362.
^ Diet. arch, de la Gaule, i, 471-2. * lb., p. 437.
^ Geogr. de la Gaule mm., ii, 436, n. 5.
Les Comm. de Cesar, pp. 40-1.
•*
Notice de Vancienne Gaule, pp. 411-2.
''

N. L. Achaintre, Caesar, iv, 299.

»

» Bull, de VAcad. Roy. de Bruxelles, xiii,

1862, pp. 396-7. ^^ lb.

" Annales du Cercle arch, de lions, x, 469.

'- Les
Comm. de Cesar, pp. 40-1
'^ The omission
of the clients of the Nervii from the list of tribes which
contributed to the army raised for the relief of iVlesia {B. G., vii, 75) seems
to show that they were regarded as incorporated with the Nervii.
" Ptolemy, Geo<jr., ii, 7, § 11.
"^ D'Anville,
Notice de fancienne Gaule, pp. 485-6.
'" liev. dc Gascogne, xxxvi, 1895, pp. 235-6. ^' fSec
p. 344.
 —
NITIOBROGES 459

of tlie Sotiatcs (q.v.)was very small but 8trabo ^ emphasizes the

;

general sinallness and insignificance of the Aquitanian tribes.

—
NoviodTinum (Bituriguni).^ The only clues that Caesar gives
as to the position of Noviodunum are, that it was situated on the
road from Cenabum, that is to say from Orleans, to Gorgobina ;

that Vercingetorix, on hearing that Caesar was marching from

Cenabum to relieve Gorgobina, abandoned the siege of that place,
and marched against Caesar that his cavalry, who had moved on
;

in advance of his column, appeared before Noviodunum on the day

on which Caesar reached it, and apparently a few hours after Caesar's
arrival that Caesar, after beating off this cavalry and receiving
;

the surrender of Noviodunum, marched against Avaricum (Bourges) ;

and that Vercingetorix followed him by easy stages. The passage

in Caesar ^ runs thus exercitum Ligerim traducit atque in Bituri-
:

gum fines pervenit. Vercingetorix, uhi de Caesaris adventu cognovit,

oppugnatione desistit atque obviam Caesari proficiscitur. Ille oppidum
Biturigum positum in via Noviodunum oppugnare instituerat,
&c. The words Biturigum positum in via Noviodunum are omitted
in the a MSS. but these MSS. are obviously at fault, as Caesar
;

would never have used such a vague expression as Ille oppidum

oppugnare instituerat, without giving any intimation as to what or
where the oppidum was. The word Noviodunum at all events must
be genuine, as it is repeated in chapter 14. In L and also in certain
inferior MSS. Biturigum positum in via is omitted, while Noviodunum
is retained. Accordingly it has been suggested that Noviodunum
was identical with the Noviodunum of the Aedui,^ the modern
Nevers. This astounding blunder, which has been adopted by
Mr. Froude,^ implies that Caesar would have crossed and recrossed
the Loire, without saying a word about it, in order to attack a town
belonging to a people with whom he had every motive to remain
friendly. Besides, any one who reads chapter 55 in connexion with
the chapters that immediately precede it, will see that Caesar is
there mentioning the Aeduan Noviodunum for the first time. The
words Biturigum positum in via are certainly genuine or, if they ;

are not, they express Caesar's meaning. For, as the Noviodunum of

chapters 12 and 14 was certainly not the Aeduan town of the same
name, and as, immediately before mentioning it, Caesar says that he
entered the territory of the Bituriges, it is plain that it was in their
territory. It is also certain that this Noviodunum was positum in
via, that is to say, situated on the road from Cenabum to Gorgobina,
because, when Caesar laid siege to it, he was marching from Cenabum
to relieve Gorgobina, and did not yet know that Vercingetorix had
raised the siege of Gorgobina and marched to meet him. It is prob-
able, however, that he intended to march to Gorgobina by way of
Avaricum for by attacking or threatening so important a town as
;

Avaricum he would have compelled Vercingetorix to abandon the siege.

' iv, 2, § 1.
^ Sheet 18 of the Carle de. France (1 : 320,000) will be found useful for the
study of this question. ^ B. G., vii, 11-2.

5 Gaemr, a ISkelch, cd. 1886,

« /6., 55, § 1. p. 354.
460 NOVIODUNUM (BITURIGUM)
Not to mention mere guesses, Noviodunum has been placed by
Napoleon at Sancerre by von Goler and Heller at Nouan-le-Fuzelier
; ;

by d'Anville at Nouan, east-south-east of Bourges by a writer ;

whose name I cannot discover at Argent, on the Grande-Sauldre,

north-west of Sancerre by de Monvel at Neuvy-en-Sullias
; by ;

J. Dumontet at Nohan-en-Gra9ay by various writers at Pierrefitte-

;

sur-Sauldre by General Creuly at some unknown point in the

;

neighbourhood of Chatillon-sur-Loire by M. Jacques Soyer at;

Neung-sur-Beuvron and by the French Commission and various

;

independent writers at Neuvy-sur-Barangeon, or, with more proba-

bility, at the village of Villate, about 3 kilometres east by south of
Neuvy-sur-Barangeon.
1. The reasons which Napoleon ^ gives are that Sancerre is about
half-way between Gien, which he identifies with Cenabum, and the
confluence of the Allier and the Loire, about 10 miles south-east of
which, at St. Parize-le-Chatel, he places Gorgobina- that Sancerre ;

is at a sufficient distance from Avaricum to correspond with Caesar's

narrative that it is situated on a hill, as, he maintains, Noviodunum
;

must have been in order that its inhabitants might see in the dis- '

tance, from the top of their walls, the cavalry of Vercingetorix '
;

that the ground in the neighbourhood was suitable for a cavalry

engagement and that the remains of a Gallo-Roman town have
;

been discovered at the foot of the hill of Sancerre.^

Not one of these reasons is convincing. The Gallo-Roman town
may have been any town but the successor of Noviodunum. It is
unnecessary to assume that Noviodunum was on a hill ^ for from ;

the top of an ordinary town-wall the range of vision, if the country

were flat, would extend over several miles, and Lord Wolseley ^ tells
us that good eyesight can hardly distinguish cavalry from infantry
at a greater distance than 1,200 yards. Indeed, as the inhabitants
made no attempt to defend themselves, it is probable that the town
was not strongly situated.^ The argument that Sancerre is situated
at the right distance between Gien and St. Parize-le-Chatel is worth-
less for it is not proved that Gorgobina was at St. Parize-le-ChAtel,
;

and Cenabum was not at Gien but at Orleans. Lastly, the earliest
known name of Sancerre was not Noviodunum, but Castrum Gordo-
nicum.^
2. Von argues that Caesar must have taken two days to
Goler '^

march from Cenabum to Noviodunum, and must have spent another

day at Noviodunum before the cavalry of Vercingetorix appeared ;

that the news of Caesar's approach must have taken one day to reach
Vercingetorix, and that Vercingetorix must have taken two days to

^ Hist, de Jules Cesar, ii, 252, note.

- Napoleon's arguments are identical with those of A. Brean {Itineraire de
V expedition de Cesar d' Agendicum it Gergovia-Boiorum, &c., 1865, pp. 83-4).
3 See
pp. 4G3-4.
^ The Soldier's PocJcet-BooJc, 5th ed.,
p. 491. Cf. p. 760, infra.
^ I find that General Creuly makes the same remark. Bev. arch., nouv. ser.,
viii, 1863, p. 400.
« See p. 429. ' Gall. Krieg, 1880, pp. 240 and 241, n. 1.
 NOVIODUNUM (BITURIGUM) 461

march from Gorgobina to Novioduniim. The geographical position

of Nouan-le-Fuzelier, he affirms, satisfies these conditions. But, as
General Creuly observes,^ all this calculation is mere guess-work.
Caesar does not say how long he took to march from Cenabum to
Noviodunum he does not say what time elapsed between his
;

arrival and the appearance of Vercingetorix's cavalry and more- ;

over, on von Goler's theory, the cavalry of Vercingetorix would have

—
marched 107 kilometres the distance from La Guerche, which von
Goler identifies with Gorgobina, to Nouan-le-Fuzelier in the same —
time in which Caesar is assumed to have marched 43 kilometres, the —
distance from Orleans to Nouan-le-Fuzelier which is very improb- ;

able. Moreover, it is not proved that Gorgobina stood upon the site
of La Guerche.
Among other reasons for rejecting Nouan-le-Fuzelier, d'Anville-
observes that it was in the ancient diocese of Orleans, and therefore
would have belonged not, as Caesar's text requires, to the Bituriges,
but to the Carnutes. It may be added that no remains, Celtic or
Eoman, have been discovered on the site.^
3. In favour of Nouan,* his own selection, d'Anville ^ urges that
it is at the right distance from the southern frontier of the Carnutes
and from Moulins, which he tentatively identifies with Gorgobina ;

that its name is derived from iVo^'io-dunum and that its geographical ;

position tallies with Caesar's account of the march of his army,

followed by Vercingetorix, from Noviodunum to Avaricum.
I know of only one positive objection to d'Anville's view but ;

that one is fatal. Nouan is only 10 miles from Avaricum, and not
more than 15 miles from any point within a radius of 16 (Roman)
miles from Avaricum, where Vercingetorix can be assumed, accord-
ing to Caesar's narrative, to have encamped. Now Vercingetorix
could not have taken several stages ^ to traverse this short distance ;

and a glance at the map will show that he could not have followed
Caesar at all because, as Caesar encamped just outside Avaricum,
;

and Vercingetorix 16 Roman miles from the same town,"^ Vercin-

getorix, marching from a point only 10 miles east of the town, must
necessarily have taken a different route. Moreover, Moulins is
certainly not identical with Gorgobina and even if there were no;

positive objection to d'Anville's view, the reasons which he gives

would be inadequate to prove it.
4. Argent, according to de Monvel,^ shows evident traces of
Roman occupation. But what of that ?
5. At Neuvy-en-Sullias Gallo-Roman remains are said to have

^ Rev. arch., nouv. ser., viii, 1863, pp. 394-5.

^ Edaircissemeris, &c., p. 240.
^ Quelques notes sur Noviodunum Biturigum (anon.),
1850, pp. 8-9 ; Mem. lus
a la Sorhonne, 1866 (1867), pp. 114-5.
* Nohant-le-Gout, according to the Government
Map (1 320,000), Sheet 18. :

^ Eclaircissemens,
pp. 236-9.
* '
Vercingetorix minoribus Caesarem itineribus subsequitur, et locum castris
deligit,' &c. B. Q., vii, 16, § 1.
'
lb., and 17, § 1.
^ Mem. de la Sac. d' agriculture . . . d' Orleans, vii, 1863, p. 66.
462 NOVIODUNUM (BITURIGUM)
been found ^ but there is nothing else to be said for it. Besides, it
;

isnot in the territory which is generally assigned to the Bituriges.

6. Nohan-en-Gra9ay is about 38 kilometres, or 24 miles, north by
west of Avaricum that is to say, it is far to the west of the road
;

which Caesar would have followed in marching either to Gorgobina

or to Avaricum, and so nullifies the words fositum in via. Moreover,
in order to reach it and then to get to Avaricum, both Caesar and
Vercingetorix must have crossed and recrossed the Cher.
7. Pierrefitte may be rejected at once. It is, indeed, on the Roman
road from Orleans to Bourges it is at a reasonable distance from :

Bourges and it possesses Roman remains.^ But it is not in the

;

country of the Bituriges.

8. General Creuly^ selects the neighbourhood of Chatillon-sur-
Loire because such a position harmonizes with his identification of
Gorgobina with Sancerre. But this identification is most probably
wrong.
9. Neung-sur-Beuvron, the chief town of the arrondissement of
Romorantin in the department of the Loir-et-Cher, is in the country
of the Carnutes, and is not situated upon the line of march
'
but, '
:

says M. Soyer,^ rien ne prouve que Noviodunum fut sur le territoire

'

des Bituriges for is it not easy to sweep away those troublesome

'
;

words Biturigum positum in via ? Besides, on a trouve a Neung '

des traces de fortification gauloise,' and most important of all —

Noviodunum gave birth to Neeun (forme primitive hypothetique),
'

puis Neun, et enfin Neung.^ Thus everything is explained, or ex-

—
plained away, except the hard fact that Neung is too near Orleans
and too far from any site with which Gorgobina has been or can be
identified.
10. There remains only Villate, near Neuvy-sur-Barangeon for ;

Neuvy-sur-Barangeon itself has been rightly rejected. It is not on

the Roman road, but about 3 kilometres west of it and no anti- ;

quities have been found in its environs.^ Moreover, the name Neuvy
is derived not from Noviodunum, but from Noviacum? On the other
hand, local tradition, or what passes for local tradition, and long-
established opinion, whatever they may be worth, are in favour of
Villate. It is on the Roman road from Orleans to Bourges and ;

though- it is only 30 kilometres from the northern side of Bourges,

Caesar probably marched a much greater distance by a circuitous
route, in order to avoid the forests on the north and the marshes
which nearly surrounded the town, and to reach his camping-ground
on the south-eastern side ^ therefore the words Vercingetorix :

minorihus itinerihus Caesarem suhsequitur would not lose their

^ Mem. de la Soc. d* agriculture . . . (^Orleans, vii, 1863, p. 66.
^
Saint-Hypohte, Recherches sur quelques points hist, relatifs au siege deBourges,
&e., 1842, pp. 7-8.
^ Rev. arch., nouv. ser., viii, 1863,
p. 400. * See pp. 428-9.
^ Bull, de geogr. hist, et descr., 1904,
pp. 153-4.
Quelques notes sur Noviodimum Biturigum, (anon.), p. 10 ; Mem. I us a la
**

Sorhonne, 1866 (1867), pp. 123-30.

' Desjardins, Geor/r. de la Gaule rom., ii, 673, n. 4.

« B. G., vii, 15, §5; 17, § 1.

 NOVIODUNUM (BITURIGUM) 463

force. Neuvy-sur-Barangeon was in the Middle Ages called Novus

Vicus ;and it has been argued ^ that this name implies the existence
of an older town, which may have been Noviodunmn. Celtic remains,
Roman coins belonging to the period comprised between the reigns
of Augustus and Gratian, the ruins of a building, which appears to
have been a Gallo -Roman theatre, and inscriptions which have been
referred to the Gallic epoch, have been discovered at the place in
question.2 On the other hand, d'Anville^ maintains that Neuvy-
sur-Barangeon is too near Orleans. But marching rapidly from
St. Parize-le-Chatel —
though not from the neighbourhood of Mou-
lins, where d'Anville tentatively places Gorgobina —
the cavalry of
Vercingetorix could certainly have reached Villate in time and ;

therefore the objection disappears. D'Anville, however, adds that

Vercingetorix, coming from Gorgobina, could not have followed
Caesar from Neuvy-sur-Barangeon (or rather, as he ought to have
said, Villate) to Avaricum. This objection has no more weight than
the other. Vercingetorix may have been present in person in the
combat before Noviodunum and even if he was not, he doubtless
;

marched up with his infantry to join his cavalry. Then, as Caesar

tells us,^ he held a council of war. By the time it was over, Caesar
was well on his way to Avaricum. It is therefore no offence against
language to say that Vercingetorix followed him thither.
Lancelot ^ objects that Neuvy-sur-Barangeon is not situated upon
a hill ;and the objection would also apply, though with less force,
to Villate, which is on gently rising ground.^ The termination
dunum, says Lancelot, implies an eminence. Caesarodunum is no
exception, for the original foundation must have been on one of
the eminences near Tours, not on the low-lying site of the modern
town. Plutarch, speaking of Lugdunum, confirms the view that
dunum means a hill and all other ancient towns in France, Germany,
;

and England, the names of which ended in dunum, were situated on

hills. Similarly the Abbe Fenel quotes Hericus, a monk who wrote
'^

in the reign of Charles the Bald, according to whom

Augustidunum demum turn coepta vocari,
August! montem transfert quod celtica lingua.^

These arguments are hardly convincing. Lancelot's assertion

about the site of Caesarodunum is not supported by any evidence ^ ;

and if it is true, it does not prove his case. The statements of Plutarch
and Hericus on a question of Celtic etymology will not be taken
seriously. There is certainly one instance of a low-lying Celtic
—
town, the name of which ended in dunum, Lugdunum Batavorum
(Leyden). Fenel urges that Ptolemy called it Kovyoheivov but the ;

^ Quelques notes, &c., p. 10. ^

75^ pp 13_32.
^ Edaircissemens, &c., p. 240. ^ B. G., vii, 14.
"^
Mem. de lilt, tirez des registres de V Acad. Roy. des inscr., &g., vi, 1718-25,
pp. 640-2.
« 8ee Carte de V Mat-Major (1 : 80,000), Sheet 122.
' Mem. de litt. tires des registres de V Acad. Roy. des inscr., &c., xx, 1753, p. 44.
**
Acta Sanctorum., Julii, vii, 29.
^ See J. J. Bourrasse, Ln Tonraine,
1855, p. 08,
464 NOVIODUNUM (BTTURIGUM)
best reading is AovyoSowov.^ Moreover, Sir John Khys ^ remarks
that diln- is of the same etymology as the famiUar English word
'

town and the German Zaun, " a hedge or field-fence " and Zeuss ^ '
;

says that its proper meaning is that of a fortified position, not a hill
(' Munitum enim locum proprie significat vox celtica dun, non
eminentem vel montem
. . . sunt etiam oppida quaedam eadem
:

voce nominata non in monte sed in planicie sita '). The fact that
Celtic strongholds were, as a rule, naturally built upon hills is no
proof that the Celts would not have applied the word dun to a forti-
fied town situated upon low ground.
The reader who has read so far will naturally exclaim with Des-
jardins,* Qui a raison ? il est bien probable qu'on ne le saura jamais.'
'

The reader will be right. It has been proved that nearly every place
which has been identified with Noviodunum was not Noviodunum ;

and it cannot be proved that any of the other places was Noviodunum.
But, as every conceivable site would appear to have been proposed,
and as there is more to be said for Villate than for any other, I
mark Noviodunum there on my map with a note of interrogation.
Noviodunum (Haeduorum). The modern editors speak of the —
identity of Noviodunum with Nevers as a thing absolutely certain.
But Caesar is the only ancient author who mentions this Novio-
dunum and he simply describes the place as oppiduin Haeduorum
;

ad ripas Ligeris oportuno loco positum.^ How^ever, although there is

no direct evidence, the probability that Noviodunum stood upon the
site of Nevers is sufficient to justify one in marking it upon the map.
It is proved by the itineraries ^ that the Gallo-Roman town of
Nivernum was situated at Nevers. Aimoin, a monk of the tenth
century, remarks that Nivernum is said to be identical with Novio-
dunum and Hugo, a monk of Fleury, writing in 1109, unhesitatingly
;

affirmed that identity.'^ The evidence, then, is simply the evidence

of tradition. But this evidence is supported by the strong probability
that an important place, like Noviodunum, would not have perished,
but would have developed into a Gallo-Roman town. If so, we may
be sure that Noviodunum was Nivernum, because there w^as no
other Gallo-Roman town with which Noviodunum can be identified.
Moreover, it seems possible that; as Noviomaqus was corrupted into
Noviomum, so Nivernum is a corruption of Noviodunum. Finally,
the great strength of the site of Nevers oportuno loco positum —
would certainly have recommended it to Caesar and this is the ;

strongest argument of all.

Noviodunum (Suessionum). — The position of this Noviodunum
cannot be fixed with absolute certainty, but was close to Soissons.
Caesar tells us that Noviodunum was a strongly fortified town ^ :

1 Qeogr., ed. C. Miiller, i, 229. ^ (j^m^ Heathendom, 1888, p. 34.

3 Gramm. Celt., 2nd ed., 1871, p. 52 (64), note.
* Geogr. de la Gaule rom., ii, 479. ® B. G., vii, 55, § 1.
^ 367
Itin. Ant., p. La Table de Peutinger, ed. Desjardins, p. 32, col. 3.
;

' D'Anville, Edaircissemens sur Vancienne Gaule, p. 409-10.

^ Propter latitudinem fossae murique altitudinem panels defendentibus

expugnare non potnit. B. G., ii, 12, § 2.
 NOVIODUNUM (SUESSIONUM) 465

wc may, as I show on pages 668-70, gather from his narrative that it

was a long day's march from his camp on the Aisne ^ and we may ;

reasonably infer that it was the chief stronghold of the Suessiones.

Caesar's camp on the Aisne was either on the hill of Mauchamp,
about a mile and a half NNE. of Berry-au-Bac, or on the plateau
of Chaudardes, some five miles nearer Soissons.^
1. Napoleon^ and von Goler,* following most of the best known

commentators of the sixteenth, seventeenth, and eighteenth cen-

turies, identify Noviodunum with Soissons but Augusta Suessionum, ;

the Gallo-Roman capital of the Suessiones, was, as its name suggests,

an entirely new town.^
2. The Abbe Le Beuf ^ decided in favour of Noyant, which is on
a hill a little south of Soissons. His argument is that Noviodunum
must have been near Soissons that the hill of Noyant is only
;

two kilometres off and that the name Noyant is derived from
;

Noviodunum. Perhaps. Who can tell ?

3. Dom Lelong"^ adopted Noyon. But Noyon, as the Itinerary
of Antonine ^ proves, was not Noviodunum, but Noviomagus.
4. Wauters thinks that Noviodunum is to be identified with
^

Nouvion-le-Vieux, in the department of the Aisne for, says he, ;

'
Nouvion est la contraction la plus naturelle qu'il soit possible de
trouver de la denomination gauloise.' This is a fair sample of the
'
evidence which often satisfies antiquaries of vast learning and high
'

authority.
Peigne-Delacourt ^^ proposes Mont de Noyon, west of the Oise,
5.
which, he says, presente tons les caracteres des oppides gaulois,'
'

and is situated on a Gallic road leading from Pontarcy to Caply,

near Breteuil. But Mont de Noyon was not in the territory of the
Suessiones the plateau is only 3 hectares, or about 7 J acres, in extent;
;

and the site is far beyond the furthest point which Caesar could have
reached in the longest day's march.
6. M. 0. Vauville has excavated an important Gallic oppidu7)i on
the hill of Pommiers, which is on the northern bank of the Aisne,
about 2 J miles north-west of Soissons.^i The ruins have yielded over
2,500 Gallic coins and 19 Roman ones earlier than 57 B.C. ^^ and ;

A. de Barthelemy ^^ suggests that this offidum was the old capital

of the Suessiones, which was succeeded by the Gallo-Roman Augusta

^ Or perhaps from another camp, a little lower down the valley. See p. 670.
'^
See pp. 659-66. ^ j{igf,^ ^q Jules Cesar, ii, 105.
* Gall. Krieg, 1880, p. 72.
^ and Rice Holmes, Anc. Britain, pp. 255, 704.
Cf. p. 401,
Mercure de France, Avril, 1736, p. 637.
«

' Hist. eccl. et civ. du diocese de Laon,

1783, p. 10, note. Cf. Bull, de la Soc.
arch. de Soissons, xii, 1903-4 (1907), pp. 196-9.
. . .

^ Ed. Wesseling,
p. 262.
^ Bull, de V Acad. Roy. de Bruxelles, 3" ser., i, 1881, pp. 564-5.
^^ Eecherches sur la position de Noviodunnm Suessionum (1856).
" See Carte de VEtat-Major(1 80,000), Sheet 33.
:

Bull, de la Soc. arch.

12 de Soissoris, xii, 1903-4 (1907), pp. 317-61 ; Mem.
. . .

de la Soc. nat. des ant. de France, Ixv, 1904-5 (1906), pp. 45-90 ; Ixvi, 1906
(1907), pp. 1-26. " Rev. celt., viii, 1887, p. 398.
1093 Hh
 466 NOVIODUNUM (SUE8SI0NUM)
Suessionum, just as Gergovia was succeeded by Augustoiiemetum
and Bibracte by Augustodunum. Moreover, M. Vauville has re-
vealed the entrenchments of a Roman camp quite close to and on
the east of Pommiers and he is probably justified in concluding
;

that this was the camp which Caesar made.^

—
Ocelum. See Graioceli.
—
Osismi. The Osismi dwelt in the department of Finistere. Pom-
ponius Mela ^ says that the island of Sena, or Sein,^ was opposite
territory which belonged to them on the other hand, Ptolemy ^
:

implies that the territory of the Veneti extended northward as far

as the Gobaean promontory, or the Pointe du Raz and, as Sein is
;

nearly due west of the Pointe du Raz, the two statements are only
reconcilable by the assumption that Mela was referring to the
southernmost part of the country of the Osismi. We shall not go
far wrong if we assume that the Montagues Noires a natural —
—
boundary divided the two peoples.^ One of the towns of the Osismi
1 M. Jullian {Hist, de la Gaule, iii, 258, n. 2), who agrees with me in identify-

ing Noviodunum with Pommiers, believes that Caesar made his camp on the
. north.
^ iii, 6, § 48.
" An attempt has been made to prove that Sein is not the same as Sena.
According to R. F. Le Men {Rev. arch., nouv. scr., xxiii, 1872, pp. 51-4), the
earliest docmnent in which Sein is mentioned is un acte du cartulaire de
'

Landevennec ', belonging to the eleventh century, in which it is called Tile '

de Seidhun'. In the sixteenth and eighteenth centuries, says Le Men, it was

generally called 'Sizun'. 'Seidhun', or 'Seithyn', was, he says, the name
of a Breton chief, who probably gave his name to the island. It has also been
argued that as, according to Mela, Sena was situated in the Oceanus Britnnnicun,
while Sein is not in the English Channel, Sein cannot be the same as Sena.
But Le Men's theory has been demolished by MM. A. de la Borderie and
J. Loth. The resemblance between the names Sena and Sein is obvious. The
cape opposite Sein is called Sizun. If, M. de la Borderie asks, the island itself
was originally so called, why has its name become Sein, while the name of
the cape remains unaltered ? Finally, he remarks that the opponents of the
identification of Sena with Sein identify it with Ouessant, a name which is
much less like Sena than even Seidhun {Geogr. gallo-rom. de V Armor ique, 1881,
pp. 20-2).
M. Loth, a distinguished Celtic scholar, says that the name Sena has nothing
to do with Seidhun. He also observes that as Sena, according to the Itinerary
of Antonine, was between Uxantis (Ouessant) and Vindilis (Belle-Ile), it can
only have been Sein. Further, he remarks, with perfect truth, that Oceanus
Britanniciis designated not merely the English Channel, but the Bay of Biscay
as well, which the ancient geographers regarded as opposite the coast of Britain
{Rev. celt., x, 1889, p. 352 ; U
Emigration brefonne, p. 54. Cf. Strabonis Geogr.
tabulae XV, instruxit C. Miiller, 1880, p. 5 and tab. iv ; B. G., v, 13, §§ 1-2 ;

Mela, ii, 6, § 85 and Vibius Sequester, De fiuminibus quorum, apud poetas Jit
;

meniio, ed. J. J. Oberlin, 1778, p. 13).

' Geogr., ii, 8, §§ 5-6.
^ A. Bertrand, remarking that the author of the legend of St. Menulfe calls
St. Corentin, who was bishop of Quimper, bishop of the Osismi, infers that they
possessed the whole of the territory which corresponded with the diocese of
Quimper {Rev. arch., nouv. ser., ix, 18G4, 324). Now, as I have shown already
(p. 421), the diocese of Quimper was one of those dioceses which were founded
without reference to the boundaries of Gallo-Roman states and no safe
;

conclusion as to the area of the Osismian territory can be drawn from a docu-
ment so late as the legend in question (Bolland, Ada ISanctoram, Julii. iii,
292, F).
 OSISMI 467

was Vorgum or Vorgiimi, which is identified with Carhaix.i See

CoRiosoLiTES and Veneti.
Paemani. See Caerosi.—
Parisii. —
The territory of the Parisii corresponded with the
modern diocese of Paris,^ that is to say, the department of the Seine
and part of the department of the Seine-et-Oise.
—
Petrocorii. The territory of the Petrocorii comprised the modern
diocese of Perigueux and that of Sarlat, which was severed from the
ancient diocese in the sixteenth century.'*^ This territory is nearly
identical with the department of the Dordogne.
—
Pictones. The Pictones occupied the ancient diocese of Poitiers,
from which the cantons of Retz and Manges were transferred to the
dioceses of Nantes and Angers respectively, and the dioceses of Lucon
and Maillezais were severed in 1317.^ From this territory, how-
ever, if we accept the conjecture of Maximin Deloche regarding the
Lemo vices Aremorici (q.v.), must be deducted the southern part of
the Loire-Inferieure and the western part of the Maine-et-Loire.
Roughly speaking, there would then remain for the Pictones the
departments of the Vendee, Deux-Sevres, and Vienne.
Pleumoxii. See Nervii. —
—
Ptianii. The position of the Ptianii cannot be determined.
Walckenaer ^ remarks that Caesar mentions them immediately after
the Bigerriones and that, after placing the surrounding peoples,
;

there remains only the country whose chief town was Beneharnum.
Following a suggestion of Sanson,^ who observes that Beam was
divided into six districts called parsans, he inclines to place them in
this country. It is difficult to see what connexion parsans can have
with Ptianii but Walckenaer, who adopted the erroneous reading
;

Preciani, insists that Sanson's conjecture brings the text of Caesar

into harmony with the works of later writers, none of whom places
any people in the country referred to.
—
Ranraci. See LATOBRiai.
—
Redones. The Redones are mentioned twice by Caesar, among
the maritime states {quae Oceanum attingunt)? D'Anville ^ considers
that their territory extended beyond the diocese of Rennes, and
included the diocese of Dol and part of that of St. Malo ^ for other- ;

^ La Table de Peutimjer, ed. Desjardins, p. 29, col. 2 Bull, de la 8oc. arch.

;

di(,Finistere, ii, 1874-5, pp. 18-72. Corpus inscr. Led., xiii, pars ii, fasc. 2,
p. 679. F. Liger {Les Osismiens, 1907, pp. 20, 69) insists that Vorganium,
not Vorgium, must be identified with Carhaix ; but Desjardins {Geogr. de la
Gaule rom., i, 317-20) proved that Vorganium was at Castell A' oh, some 20 kilo-
metres, or about 12 miles, north of Brest. See also Corpus inscr. Ltd., xiii,
pars i, p. 490, where Otto Hirschfeld confutes Liger.
" Walckenaer, Gkogr. des Gaules, i, 404.

^ D'Anville, Notice de Vancienne Gaule, 517 Walckenaer, Geogr. des

p. ;

Gaules, i, 361.
* DWnville, pp. 519-20 ; Walckenaer, i, 365-6 A. Longnon, Atlas hist, de la
;

France, p. 6. See also Rev. arch., d^ ser., xviii, 1891, pp. 260-1.
^ Geoyr. des Gaules, i, 293-5.
^ Les Comm. de Cesar, ed. 1650, p. 65.
^ B. G., ii, 34 vii, 75, § 4.
;
» Notice de Vancienne Gaule,
p. 542.
^ It should bo noted that these dioceses were not founded until after the Gallo-
H h 2
468 REDONES
wise, he remarks, Caesar would liave been wrong in saying that they
were a maritime people. The French Commission ^ reject this view
(1) because Caesar reckons the Aulerci, who had no sea -board, as
maritime states (2) because in their opinion Aletum, or Aleth, and
;

therefore probably also the diocese of Aleth, the see of which was
transferred in the ninth century to St. Malo, belonged to the Corioso-
lites and (3) because after the tenth century Dol and St. Malo were
;

closely connected with the Breton-speaking part of Brittany, while

Eennes remained French. The Commission are further of opinion
that if the territory of the Redones actually communicated with the
S3a, it did so along the valley of the Vilaine, which the Veneti could
have afforded to let them retain. But it is not proved that no
Aulercan state possessed a sea -board it is not proved that Aletum
:

belonged to the Coriosolites, and as the diocese of Aleth was not

formed until after they had ceased to exist as a Gallo-Roman state,
there is no evidence that it corresponded with their territory it :

is incredible that the powerful Veneti, who, moreover, were con-

terminous on the south with the Namnctes, would have left the
valley of the Vilaine in the possession of the Redones and, as ;

Caesar says twice over that the Redones were a maritime people,
we must assume that he was right. MM, Kerviler ^ and J. Loth ^
give them the strip of coast between the Ranee and the Couesnon,
which was included in the diocese of Dol. See Coriosolites.
—
Remi. The territory of the Remi, according to Walckenaer ^ and
Dcsjardins,^ included not only the diocese of Reims and that of Laon,
which was severed from it in the fifth century, but also that of
Clialons. The diocese of Chalons was formed out of the territory
of the Catuvellauni, who are not mentioned by Caesar, or ))y any
writer before the time of Constantine. Walckenaer remarks that,
according to Caesar, the territory of the Remi was conterminous with
Celtican Gaul ^ and he maintains that this would not have been
;

true unless they had possessed the diocese of Chalons. But Caesar's
statement would be equally true if the diocese in question had
belonged, as M. Longnon maintains, to the Lingones.
'^
Still, I
believe that the diocese of Chalons did form part of the territory of
the Remi, or that the Catuvellauni were one of their client peoples ;

for Caesar tells us ^ that, in consequence of the favour which he

showed the Remi, various tribes enrolled themselves as their clients.
D'Anville ^ believes that the Remi 'possessed only a part of the
diocese of Laon, and that the other part belonged to the Viromandui.
He thinks it likely that this part belonged originally not to the
diocese of Reims, but to that of Noyon, which was formed out of the
territory of the Viromandui and he finds it difficult to believe that
;

Roman period. The diocese of St. Malo included the broad headland cast
of the Kance. ^ Bev. arch., nouv. ser., ix, 1864, pp. 328-30.
^
Bull. arch, de VAss''^ bretonne, 3" ser., iv, 1885, pp. 225-8.
^ UEmigration hretonne, p. 52. * Geogr. des Guides, i, 487-8.

^ Geogr. de la Guide rom., ii, 455. ® B. G., ii, 3, § 1.

' Atlas hist, de la Gaule, pp. 5-G. See luy note on the Lingones.
8 B, G., vi, 12, § 7. * Notice de Vunciennt Gaule, p. 693.
 REMI 469

the Viromandui were confined within such narrow limits on the south
of their chief town, Augusta (St. Quentin), as they must have been
if they did not possess a part of the diocese of Laon. But there is
no evidence for d'Anville's conjecture.
[See A. Piette's deHmitation of the territory of the Remi (Itin.
gallo-rom. dans Je de/p^ de V Aisne, 1856-62, pp. 29-31)].
—
Ruteni. The territory of the Ruteni was identical with the
ancient diocese of Rodez, that is to say, the greater part of
the department of the Aveyron and the northern part of that
of the Tarn.i
Samarobriva has always been generally and rightly identified
with Amiens and the arguments by which von Goler ^ and others
;

attempted to upset this conclusion have never been taken seriously.^

It is established not only by the evidence of the itineraries,* of the
mile-stone of Tongres,^ and of mediaeval writers,^ but also by a well-
known law of nomenclature. We
have seen that at the time when
the Itinerary of Antonine was compiled most of the towns of Gaul
—
had two names, the old Gallic name and the name of the people
in whose territory the town was situated.'^ The chief town of the
Ambiani, which, in the Table, is called Samarobriva, is mentioned in
the Itinerary of Antonine under the name of Ambiani. The names
of the capital and of most of the chief provincial towns of France are
derived, not from the old Gallic names, but from the names of the
tribes to which the towns belonged. Thus we have Paris (Parisii),
Soissons (Suessiones), Reims {Remi), Bourges (Bituriges), and so on.
Now Amiens is, beyond all doubt, derived from Ambiani.
—
Samnitae. Strabo ^ says that, according to Posidonius, an island
opposite the mouth of the Loire was inhabited by the women of '

the Samnitae ', and that from time to time they visited their hus-
bands, who dwelt on the mainland. Ptolemy ^ says that below ', '

that is to say, south of the Veneti, and in the neighbourhood of the

Loire, dwelt the Samnitae. A few lines further on,^^ he mentions
the Namnetes (q.v.), placing them erroneously between the Cenomani
and the Abrincatui.ii Marcian ^^ speaks of the ^a^vlrai. C. Miiller ^^
remarks that Ptolemy erroneously distinguishes ^a^vZrat from
Na/xj/ryrai and Desjardins ^- takes the same view, observing that
;

^ D'Anville, Notice de Vancienne Gaule, p. 693.

2 Gall. Krieg, 1880, p. 168, n. 1.
^ I refuted them in my first edition, pp. 477-8.
* Itin. Ant., ed. Wesseling, pp. 362, 380 ; La Table de Peidinger. ed. Des-
jardins, p. 14, col. 2-3 ; p. 15, col. 3.
5 See Rev. arch., nouv. ser., iii, 1861, pp. 408-13.
« See L. P. Colliette, Mem. sur le Vermandois, 1771-3, p. 12.
>
See p. 407, n. 3. » iv, 4, § 6.
» Geogr., lo
ii, 8, § 6. Ih., § 8.
'^ For a list of the blunders which Ptolemy made in dealing with Gallic
geography see Hist, de V Acad, des inscr., xxxi, 1768, pp. 265-6,
12 Geogr. Graec. min.,
ed. C. Miiller, i, 552, T. 18. According to M. de la
Monneraye {Bull. arch, de VAss^^ bretonne, 3" ser., iv, 1885, p. 271, n. 3), a good
MS. of Marcian gives the reading HafxviTai.
" Ptolemy, i, 213, note,
^* Geogr. de la Gaule rom., i, 143, n. 4 ; ii, 140, n, 2.
470 SAMNITAE
the error probal)ly crept into the text of Strabo owing to the simi-
larity between S, which was often written M, and N, and also to the
fact that the name Samnitae was familiar to Italian ears and that ;

Ptolemy doubtless copied Strabo, and Marcian Ptolemy. Accord-

ingly Desjardins concludes that in Strabo for Sa/xvirwi/ (yvvatKos)
we ought to read Na/xj/erwv and this conclusion is supported by
;

the statements of Strabo ^ and Polybius,^ that the Loire entered the
sea between the Pictones and the Namnetes.^
Some coins belonging to the Aremorican type, stamped on the
reverse side with the letter 'X, were found some years ago at Cande
and at Ancenis, in the valley of the Loire. These places are east of
the comitry which belonged to the Samnitae, if there ever was such
a people but M. Parenteau^ believes that ^ stands for ^a/xvirat.
;

Unbiased critics will, I think, agree that this argument is worthless.

To conclude, I am strongly inclined to believe, w^ith C. Miiller,
Desjardins, and the majority of scholars, that the Samnitae never
existed.^
—
Santoni. The Santoni occupied the dioceses of Saintes and
Angouleme and the Pays d'Aunis in other words, the departments
;

of the Charente and C^harente-Inferieure and part of the Gironde.^

To this M. Longnon proposes to add the territory of the Bituriges
"^

Vivisci (q.v.), who are not mentioned by Caesar, and may conceiv-
ably have been clients of the Santoni. He points out that, if his
conjecture were adopted, Caesar's statement,^ that the territory of
the Santoni was not far from that of the Tolosates, would be less
open to objection. But, as I have already shown ,^ Caesar's state-
ment throws no real doubt upon his good faith and I do not see;

the use of making conjectures which there is no evidence to support,

and which can neither be established nor refuted.
—
Seduni. See Nantuates.
—
Segni. See Caerosi.
—
Segusiavi. According to all the good MSS., the people whom
Caesar mentions in B. G., i, 10, § 5, were the Sebusiani the people ;

whom he mentions in vii, 64, § 4, 75, § 2, the Segusiavi. It is now

agreed that the two were identical. ^^
In B. G., i, 10, Caesar says that, after crossing the Alps from Italy
in 58 B. c, he passed into the country of the Vocontii, thence into
that of the Allobroges, and thence into that of the Segusia^^. He

Mv, 2, § 1. 2 xxxiv, 10, § 6.

^De la Monneraye {Bull. arch, de VAss'" hreionne, 3''
1885, p. 271)
ser., iv,
observes, further, that Ptolemy makes the Samnitae neighbours of the
Andecavi, or Andes ; and he argues that this proves that Ptolemy confounded
them with the Namnetes.
* Bull, de la Soc. arch, de Nantes, ii, 1862,
pp. 115-6.
^ The Samnitae are identified by Mommsen {Bbm. GescJi., v, 1885,
p. 87, note
\Hist. of Borne, —
the Provinces, i, 96, note]) with the Coriosolites !

^ D'Anville, Notice de Vancienne Gavle,

pp. 576-7 ; Walckenaer, Geogr. des
Gaules, i, 363.
' Atlas hist, de la Gaule, p. 6. « B. G., i, 10,
§ 1. » See
pp. 225-0.
'° Cf. Nipperdey's Caesar,
p. 792, with A. Bernard, Descf^ du pays des
Segusiaves, 1858, pp. 42, 46.
 SEGUSIAVI 471

goes on to say that the Segusiavi were the first people outside the
'

Province, beyond the Ehone '.^ It is clear from B. G., i, 12-3 that
when he entered their country he was on the eastern bank of
the Saone. The greater part of their territory, however, was on the
western bank for two of their towns, Eodumna (Roanne) and
;

Forum Segusiavorum (Feurs), which Ptolemy ^ mentions, were on

that side, and Rodumna was actually west of the Loire. Their chief
town, Lugdunum (Lyons), is placed by Ptolemy in Aeduan territory,
doubtless because they were dependants of the Aedui. The western
part of their territory appears to have comprised that part of the
diocese of Lyons which extended on the right bank of the Rhone
and of the Saone.
It is impossible to define the extent of the tract which the Segusiavi
possessed in the angle between the Rhone and the Saone. Desjar-
dins 3 thinks that it may have extended as far north as Mixon but :

Valentin Smith* argues that Trevoux, which is 25 miles south of

Macon, being situated between two places called Amberieux, must
have belonged to the Ambarri and accordingly he restricts the
;

eastern Segusiavi to a very small tract. A. Bernard ^ gives them

a considerable territory, extending in a north-westerly direction from
a point near the confluence of the Ain and the Rhone to Thoissey.
Caesar speaks of the Segusiavi as clients of the Aedui,^ and of the
Ambarri as necessarii et consanguinei Haeduorum? Desjardins holds
that we must understand the clientship of the Segusiavi in the same
sense as that of the Cadurci, the Gabali, and the Vellavii, who were
under the hegemony of the Arverni,^ and yet ranked as states. It is
clear, he remarks, that Caesar regarded the Segusiavi as forming
a distinct state (civitas), because he says that the territories of the
Aedui and the Segusiavi bordered on the Province {finitimi frovin-
ciae),^ which proves that the Ambarri were regarded as forming part
of the civitas Haeduorum, for the territory of the Aedui, properly
so called, did not touch the Province. Besides, he goes on to say,
the Segusiavi ranked as one of the sixty civitates of Augustus.^® There
is no doubt that the territory of. the Segusiavi was distinct from

that of the Aedui but the argument by which Desjardins tries to

:

prove that it was is wrong for Caesar sometimes uses the word
;

finitimus loosely. (See my note on the Latobrigi.)

B. Guerard,^^ who wrongly holds that client peoples were merely
pagi, or subdivisions, of the people to whom they stood in that
relation, includes the territory of the Segusiavi in that of the Aedui,
—
Senones. The Senones undoubtedly occupied the diocese of Sens :

^
Hi sunt extra provinciam trans Rhodanum primi.
^ Geogr., ii, 8, § 11. ^ Geogr. de la Gaule rom., ii, 605.
* Fouilles dans la vallee du Formans en 1S62, p. 5.
^ Description du pays des Segusiaves, pp. 41, 46.
« B. G*., vii, 75, § 2. ' 7&.,i, 11,§4.
« 76., vii, 75, § 2. » Ih., 64, § 4.
'" Geogr. de la Gaule rom..,
ii, 468-9.
^^ Essai sur le systeme des divisions territoriales de la Gaule, 1832, p. 8. Cf.
pp. 344-5, supra, and Mem. presentes par divers savants a V Acad, des inscr.,
2" ser., iv, 1860, p. 366.
472 SENONES
the diocese of Auxerre (civitas Autessiodurmn) is generally attributed
to them and, according to Walckenaer,i they occupied the dioceses
;

of Troyes and Meaux as well. That is to say, in Walckenaer's

opinion, the Tricasses, who are not mentioned by Caesar, and the
Meldi were clients (or ])agi) of the Senones. It has been argued that
the territory of the Tricasses was not included in that of the Senones,
because Pliny ^ and Ptolemy ^ mention the two peoples separately
but Walckenaer replies (1) that from the silence of Caesar and Strabo
in regard to the Tricasses we may infer that they and the Senones were
practically one in the time of Caesar, and were only separated by
Augustus, whose policy was to break up states that were unduly
powerful (2) that as, according to Caesar,* the Senones were neigh-
;

bours of the Belgae, the territories of the Tricasses and the Meldi
must have been included in theirs and (3) that such a union
;

harmonizes with Caesar's remark about the great power of the

Senones.^
These arguments are not conclusive. It is obvious that the first
and the second might be used to prove that the Tricasses and the
Meldi were clients, not of the Senones but of the Remi or the Sues-
siones and in fact Desjardins ^ does infer that the Tricasses were
;

clients either of the Senones or of the Remi. The conclusion is

probable but which of the alternatives is right ? Walckenaer offers
;

a further argument in support of his. He quotes from an inscription

found at Auxerre, in which there is mention of civiiatis Senonum,
Tricassinormn, Meldorum, Parisiorum, et civitatis Aeduorum ; and he
argues that the omission of Autessiodurum proves that Auxerre too
was included in the country of the Senones. But, as Long remarks, '^

'
it is difficult to see what conclusion can be drawn from this inscrip-
tion.' It would seem to imply that in the Gallo-Roman period the
Senones, the Tricasses, the Meldi, and the Parisii formed only one
civitas, which is certainly not true.^ M. Longnon, as I have observed
in my note on the Lingones, holds that the Tricasses were clients
of that people. I should say that they were clients either of the
Senones or of the Remi but, as these were both powerful tribes,^
;

it would be useless to attempt to decide between them. M. Jullian,!^

indeed, unhesitatingly includes the Tricasses among the Senones, on
the ground that the Belgae were separated from the Celtae by
the Marne but the Remi acquired new dependencies through the
;

influence of Caesar,^^ and their clients may not all have been in Belgic
territory. If, however, the Tricasses were merely a pagus, M. JuUian's
argument is sound.
It is rather puzzling to find that while Walckenaer argues that
Auxerre was included in the country of the Senones, he maintains

1
Geogr. des Gaules, i, 407. ^ ^r^^ jj^^f^
^^^ jg (32)^ § 107.
3 Geogr., ii, 8, §§ 10-1. * B. G., ii, 2, § 3.
^ lb., V, 54, § 2. Geogr. de la GauJe rom., ii, 470.
«

^ W. Smith, Diet, of Greek and Roman Geogr., i, 346.

8 See Pliny, Nat. Hist., iv, 18 (32), § 107.
» B. G., V, 54, § 2 ; vi, 12, §§ 7-9.
10 Hist, de la Gaule, ii, 524, n. 6. " B. G., vi, 12, § 7.
 SENONES 473

that the diocese of Auxerre belonged to the Boi, whose territory was
situated within the frontiers of the Aedui. Autessiodurum is not
mentioned by any author earlier than Ammianus Marcellinus but ;

it appears in the Notitia frovinciarum} where it is assigned to the

Senones. Walckenaer^ thinks it natural to attribute the territory
embraced by the diocese of Auxerre au seul peuple celebre de la '

Gaule dont I'emplacement n'est pas clairement indique par les

auteurs,' that is to say, to the Boi. In my article on Gorgobina
I have given reasons for dissenting from this view.
Desjardins ^ remarks that the Aedui, the Bituriges, and the
Carnutes were all neighbours of the Senones :
'
par consequent
I'Auxerrois faisait partie du domaine immediat des Senones.'' But
the conclusion does not follow. Substitute Aedui (q.v.) for Senones ;

and the argument remains as good.

[Since I wrote this article the last (and posthumous) volume of
Desjardins's work has appeared. I learn that a mile-stone has been
discovered, bearing the date 245, which was erected at the distance of
72 Roman miles from Augustodunum, on the road to Autessiodurum.
It stood in Aeduan territory (in Aeduormn finihus) and also in the
territory of iVutessiodurum and, as Desjardins * points out, it proves
;

that, at all events at the time when it was erected, the territory which
afterwards belonged to the civitas Autessiodurum formed a part of the
territory of the Aedui. Thus Desjardins recants his former argu-
ment.^]
—
Sequani. The territory of the Sequani was bounded on the east by
the Jura, which separated them from the Helvetii, and, according
to the writer, or writers, of 5. (r., i, 1, § 5, and iv, 10, § 3,^ by the
Rhine they held the Pas de I'Ecluse between the Jura and Mont
:

Vuache westward from the Pas de I'Ecluse their territory was

:

bounded, in part, on the south by the Rhone but a part of the ;

country between the Rhone and the Saone was occupied by the
Allobroges, the Segusiavi, and the Ambarri.*^ The western boundary
of Sequania was, according to Strabo^ and Ptolemy,^ the Saone and ;

on the north-west and north their neighbours were the Lingones,

the Leuci (from whom, as d'Anville ^^ says, they were doubtless
separated by the Vosges), and the Mediomatrici. D'Anville argues
that those parts of the dioceses of Chalon (Cabillonum) and Macon
—
(Matisco) both Aeduan towns which are on the east of the Saone,—
1 Ed. 0. Seeck, p. 265 (iv, 4). 2 g^^^g^,^
^^^ Gaules, i, 83-4.
^ Geogr. de la Gaule rom., ii, 471-2.
* Ih., iv, 184-5. Cf. Corpus inscr. Lat., xiii, pars 2, fasc. 2, p. 711.
^ M. Jullian {Hist, de la Gaule, ii, 526, n.
2) argues that the Auxerrois probably
belonged to the Senones, (1) because Caesar, when he was near Nevers, was
probably not far from the Senones {B. G., vii, 56, § 5) (2) because the ;

Senones were a very powerful people {ih., v, 54, § 2) ; and (3) because Auxerre
belonged later to the Senones but he is still doubtful. Naturally ; for were
:

not the Aedui also a very powerful people ?

« See
pp. 394, n. 6, and 692.
' ^. 6'., i, 1, § 5 ; 2, § 3 ; 6, § 1 ; 8, § 1 ; iv, 10, § 3.
' iv, 3, § 2. 9 Geogr., ii, 8, § 12.
^" Notice de Vancienne Gaule, p. 599. •
474 SEQUANI
belonged to the Sequani, not the Aedui. The proof which he offers
is that, according to Fredegaire, the monastery of St. Marcel, which

was in the neighbourhood of Chalon and on the eastern bank of

the Saone, was in Sequanian territory .^ The French Commission ^
reject his view, preferring to follow the indications of the dioceses.
Besides the diocese of Besangon, the Sequani may have possessed
a portion of the northern part of that of Belley.^ Moreover, if, as
might be inferred from B. 6r., i, 31, § 10, 51, §2, the Triboci had
settled before the defeat of Ariovistus in the territory of the Sequani,
and if they then occupied, as they subsequently did, the territory
which corresponded with the diocese of Strasbourg, that territory
also belonged originally to the Sequani. According to Strabo,*
however, the Triboci were established in the territory of the Medio
matrici (q.v.).
Stoffel ^ thinks that the river Ecken-Bach, between Gemar and
Schlettstadt, was the boundary betw^een the Sequani and the
Mediomatrici for, he observes, a strip of country bounded by two
;

parallel lines running through Gemar and Schlettstadt respectively

at right angles to the general direction of the Ehine, presents all the
features which, in ancient times, would have constituted the natural
obstacles desirable in a borderland.^ See Aedui, Allobroges,
Latobrigi, Rauraci, and Triboci.
—
Sibusates. The Sibusates have been conjecturally placed in the
neighbourhood of Sobusse, between Dax and Bayonne but there ;
'^

is no evidence except the resemblance between the names.

—
Sotiates. The Sotiates occupied the northern part of the civitas
Elusatium, or the country round Sos in the department of the Lot-et-
Garonne. The evidence for this, which is the common view, is put
together by d'Anville.^ He observes that Sos was in the Middle Ages
called Sotium and that the Jerusalem Itinerary ^ mentions a place
;

called Scittium, the distances between which and Elusa and Vasata
respectively very nearly corresponded with the actual distances
between Sos and Eauze and Bazas. Scit\t\ium, he suggests, was
due to a wrong reading of Sotium, the scribe mistaking an ill-formed
for ci. He also points out that Crassus, having passed through
'

the country of the Santoni, entered Aquitania by the north, and the
Sotiates would be the first tribe on whom he fell \^^ which condition

' Monumenta Germaniae historica, ed. B. Krusch, ii, 1888, p. 124.

^ Diet. arch, de la Gaule, i, 14.
^ Anzeigerfiir schweizerische AlterthumsTcunde, 1885, p. 1 12. Cf p. 365, supra.
.

* iv, 3, § 4.
5 Guerre de Cesar et d'Arioviste, pp. 89-90.
^ Regarding the territory of the Sequani see also Congres archeol. de France,
Iviii'^ sess., 1890 (1893),
l28. p.
' A. de Valois, Notitia Galliarum, p. 524. See Desjardins, Geogr. de la
Gaule rom., ii, 360-1. M. Jullian {Hist, de la Gaule, ii, 452, n. 6), remarking
that among the various readings of 5. G., iii, 27, § 1, we find Sibtilates, identifies
the Sibusates with the Sybillates of Pliny {Nat. Hist., iv, 19 [33], § 108), who
are supposed to have occupied the valley of the Soule in the department of
tlie Basses -Pyrenees.
^ Notice de Vancienne
Gaule, pp. 611-3. ® Ed. Wesseling, p. 550.

^° See
W. Smith's Diet, of Greek and Roman Geogr., ii, 1024.
 —
SOTIATES 475

is satisfied by the geographical position of Sos. This reasoning is

sound enougli ^ but many writers have refused to accept it.

;

1. A. Garrigou" places the Sotiates in the pays de Foix ', in the '

department of the Ariege. Now the pays de Foix was occupied

' '

by the Consoranni but this difhculty in no way disconcerts Garrigou.

;

Evidently, he says,^ Caesar, impelled by a barbarous thirst for

revenge, deprived the tribe of the Sotiates of their individuality;
for when he mentions the general submission of the Aquitanian
tribes which followed the victory of Grassus,^ he says nothing about
the Sotiates. It is true that they had submitted before ^ and that
Caesar tells us that his list of the tribes which submitted is incom-
plete ^ but trifles like these do not shake Garrigou's conviction.
;

Again, the pays de Foix ', if it was in Caesar's Aquitania at all,

'

was the very furthest part from the northern frontier, by which he
invaded the country while the Sotiates were the very first people
;

whom he encountered This difficulty seems staggering

! but ;

Garrigou fancies that he can dispose of it. He claims to have proved

that L. Manlius was defeated by the Aquitani in the valley of the
Ariege."^ Now there is no evidence whatever to show where Manlius
was defeated but I am willing to assume, for the sake of argument,
;

that Garrigou is right. Well, he continues, Caesar says that Crassus

was campaigning in the same country in which Manlius had been
defeated, and that it was the Sotiates who had defeated Manlius
(' Cesar nous donne ce fait comme positif ') and to make assurance
;

doubly sure, Caesar tells us that the country of the Sotiates was
conterminous with the territories of Tolosa, Carcaso, and Narbo,
Toulouse, Carcassonne, and Narbonne. The conclusion, says Garri-
gou, is inevitable that the territory of the Sotiates was in Ariege.
1 reply that Caesar tells us none of these things. What he says
is that Crassus, when he arrived in Aquitania, was aware that he

would have to fight in a country from which Lucius Manlius had

retreated in disorder. (P. Crassus, cum in Aquitaniam pervenisset . . .

cum intellegeret in iis locis sihi helium gerendum unde L. Manlius . . .

. impedime^itis amissis frofugisset.)^ Crassus waged war in different

. .

parts of Aquitania, first in the country of the Sotiates and afterwards

in that of the Vocates and Tarusates and Caesar's statement
;

proves nothing, except that Manlius had been defeated somewhere in

Aquitania. So far from stating as a positive fact ' that Manlius
'

had been defeated by the Sotiates, Caesar never mentions the

Sotiates or any other people in connexion with the disaster. Nor
does he say that the country of the Sotiates was conterminous with
the territories of Tolosa, Carcaso, and Narbo he merely says that :

Crassus, on arriving in Aquitania, summoned to his standard a large '

number of excellent soldiers from Tolosa, Carcaso, and Narbo

Cf. C. JuUian, Hist, de la Qaule, iii, 305, n. 1.

^

£ltudes hist, stir Vancien pays de Foix et le Conseran, 1863.

2

« lb.,
pp. 46-7, 58. « B. G., iii,
27, § 1. « Ih., 22, § 4.
* maxima pars Aquitaniae sese Crasso dedidit quo in numero fiieriint
. . .

TarbelH, &c. 76., 27, 5 1.

' Etudes, &c.,
pp. 35-6. ^
s
s. G., iii, 20, § 1.
476 SOTIATES
states in the Province, adjacent to Aquitania —and marched his
army into the country of the Sotiates '
(multis 'praeterea viris fortibns
Tolosa et Carcasone et Narbone, quae sunt civitates Galliae frovinciae
finitimae his regionihus, nominatim evocatis in Sotiatium fines exercitum
introduxit)?- Garrigou assumes that Crassus marched unmolested
right through Aquitania into the Province to fetch these reinforce-
ments, instead of sending for them and he asserts that the words
;

in Sotiatium fines exercitum introduxit prove that the country of

the Sotiates was conterminous with the Province !

2. The fact that the Sotiates were the first enemy whom Crassus
encountered after crossing the northern frontier of Aquitania is
also fatal to the theory of B. de la Greze,^ who identifies the principal
stronghold of the Sotiates with a rock, on which now stands a ruined
castle, at Lourdes, in the department of the Hautes-Pyrenees. The
only argument worth noticing which he advances is that there are
mines in the neighbourhood of Lourdes and none in the neighbour-
hood of Sos.^ But Caesar does not say that there were mines in the
country of the Sotiates he only says that the Sotiates undermined
:

the terrace (agger) which Crassus constructed in besieging their

stronghold, and then remarks parenthetically that the Aquitanians '

are very skilled in operations of this kind, as mining works exist in

many parts of their country (cuius rei sunt longe peritissimi Aquitani,
'

froperea quod multis locis apud eos aerariae secturaeque sunt).^ He

learned from Crassus's dispatch that the agger had been undermined :

he knew that mines were worked in Aquitania and he connected

;

the two facts. That was all.

3. M. Camoreyt ^ has advocated the claims of Lectoure, a town
about 18 miles due south of Agen. His paper, which attracted undue
attention, has been thoroughly examined by the Abbe A. Breuils,^
who remarks that Lectoure belonged not to the Sotiates but to the
Lactorates. M. Camoreyt's theory rests upon the assumption
that Crassus did not trust Teutomatus, the king of the Nitiobroges
who helped Vercingetorix four years later,"^ and that he there-
fore entered Aquitania, after a long detour, from the side of the
Province. Is it necessary to reply that we do not know whether
Teutomatus was reigning in 56 b. c. that, on M. Camoreyt's assump-
;

tion, Crassus would not have trusted the Pictones or Caesar the Aedui
or that, if Crassus had had reason to distrust the Nitiobroges, he
would not have left them unsubdued on his rear ? Two of M. Camo-
reyt's arguments, however, are worth considering. First, the plateau
of Sos only covers 14 hectares, or about 34 acres, an area which, he
says, would not have sufficed to accommodate more than 12,000
souls, and therefore not more than 3,000 fighting-men. But the

^ B.Q., iii, 20, § 2.

2 Mem.de la Soc. des ant. de France, 2^ ser., x, 1850, pp. 284-304.
•^
lb., pp. 290-1. * B. G., iii, 21,
§ 3.
^ L' emplacement de Voppidum des Sotiates, 1883.

« Rev. de Gascogne, xxxvi, 1895, pp. 225-44, 280-5. Cf. J. F. Blade in

Annates de la Faculte des lettres de Bordeaux, 1893, pp. 127-8.
^ See B. G., vii,
31, § 5.
 SOTIATES 477

host of Vercingetorix in Alesia enormously outnumbered the Man-

dubii who inhabited the town and it is not unreasonable to suppose
;

that more than half of the 12,000 would have been soldiers, while
Crassus's infantry were only 12 cohorts, say about 5,000 men. More-
over, as the Abbe Breuils points out,^ it is quite possible that the
oppidum extended beyond the limits of the modern town. Secondly,
the principal Gallic towns did not exchange their names for those
of the tribes to whom they belonged (for example, Lutecia became
Parisii) before the close of the third century, and by that time the
Sotiates were incorporated in the civitas Elusatium M. Camoreyt
:

concludes that the name of the town which Crassus captured can
never have been eiTaced by Sotiates, and therefore that the town was
not identical with Sos. Breuils replies that the well-known law of
nomenclature to which M. Camoreyt refers is subject to exceptions :

for instance, while Avaricum became Bourges (Bituriges), Bordeaux

is derived from Burdigala, —
the ancient name of the chief town of the
Bituriges Vivisci.
—
Suessiones. The Suessiones possessed the diocese of Soissons, that
is to say, the greater part of the department of the Aisne, and possibly

something more.^ The diocese of Senlis (civitas Silvanectum), as I have

shown in my article on the Bellovaci, probably belonged to the
Suessiones, or, if it did not form part of their proper territory, was
under their overlordship.^ Desjardins * argues further that as,
according to the description which the envoys of the Remi gave to
Caesar,^ the territory of the Suessiones was of great extent, it probably
extended beyond the northern bank of the Oise, and thus included
part of the diocese of Noyon, which is usually assigned in its entirety
to the Viromandui (q.v.). This is a guess and nothing more. Long-
non^ assigns the diocese of Meaux [civitas Meldorum [q.v.]) to the
Suessiones, but gives no reasons. See Bellovaci and Senones.
—
Tarbelli. The territory of the TarbeUi, according to Ptolemy,^
extended southward from the southern frontier of the Bituriges
Vivisci (q.v.) to the Pyrenees and they are generally considered
;

to have occupied that portion of the dioceses of Aquensis (Dax) and

Lapurdensis (Bayonne) which did not belong to their clients, the
Cocosates and the Sibusates. This territory comprised the western
part of the departments of Les Landes and Basses-Pyrenees.^
M. Longnon^ holds that the Tarbelli also possessed the diocese of
Rev. de Gascogne, xxxvi, 1895, pp. 273-4.
'
Cf. Carte de VEtat- Major
(1 80,000), Sheet 216.
: ^ Walckonaer, Geogr. des Gaules, i, 486.

^ See Rev. des etudes anc, v, 1903,

pp. 31-2.
* Geogr. de la Gaule roin., ii, 452-3. ^ ]3, G., ii, 4, § 6.
* Atlas hist, de la France, Cf. C. JulHan, Hist, de la Gaule, ii, 483, n. 3.
p. 7.
' Geogr., ii, 7, § 8. M. J. F. Blade {Annales de la Faculie des lettres de Bor-
deaux, 1893, p. 119) remarks tliat Ptolemy was wrong in placing the Tarbelli
immediately south of the Bituriges Vivisci, from whom they were actually
separated by the Boiates. It is possible, however, that in Ptolemy's time
the Boiates were included in the group of tribes to which was given the official
name of Tarbelli. See p. 479.
^ D'Anville, Notice de Vancienne Gaule, p. 632 ; Desjardins, Geogr. dc la
Gaule rom,., ii, 362, n. 6.
Alias hist, dc la France, p. 7.
''
478 TARBELLI
Aire (civitas Aturensiunt), which is generally assigned to the Tarusates
(q.v.), that of Beam (civitas Benarnensium), and that of Oloron
(civitas Iluronensium). He quotes Ausonius ^ to prove that Paiilinus,
a traveller, returning from Zaragoza (Caesaraugusta) to Hebroinagus
on the Garonne, found himself, on leaving Spain, in the territory of
the Tarbelli. Hebromagus is generally identified with Bram, about
12 miles north-west of Carcassonne.^ But according to P. de Marca,*^
there was another Hebromagus, w^hich stood upon the site of Embrau,
near Blaye, on the right bank of the Gironde and this is the place
:

to which he believes that Ausonius referred. There is not, however,

the slightest evidence that more than one Hebromagus ever existed.
Desjardins, in his latest utterance on the subject,* expresses his
agreement with M. Longnon. M. Longnon, he says, has shown that
of the twenty-two peoples enumerated by Ptolemy as forming the
imperial province of Aquitania, seventeen were on the right bank of
the Garonne. The remaining five therefore occupied the whole
of the Aquitania of Caesar and this conclusion is confirmed by
;

the fact that none of the ancient geographers who wrote between the
time of Pliny and the time of Diocletian mentioned any other people
besides those five as having dwelt in the Aquitania of Caesar. Those
live, of whom Caesar mentions only two, the Ausci and the Tarbelli,
were the Tarbelli, the Vassarii, the Datii ^ (whom M. Longnon
identifies with the Lactorates, because the latter are not mentioned
by any other writer), the Ausci, and the Convenae. (I may remark,
in passing, that Ptolemy^ only mentions seventeen Aquitanian
peoples in all). Now according to the well-known inscription of
Hasparren, this smaller Aquitania was occupied by nine peoples;
and, according to M. Longnon, these nine peoples were the five
already mentioned and the Boiates, the Elusates, the Bigerriones, and
the Consoranni. But the inscription of Hasparren cannot be referred
to an earlier period than the end of the third century, when the
province of Novempopulana was created."^ In the time of Theodosius
this province comprised twelve peoples, —
the nine just mentioned
and the Aturenses, Benarnenses, and Iluronenses. Accordingly,

^ The passage in Ausonius (cd. R. Peiper, 1866, pp. 281-2, vv. 123-9) runs as
follows :

Ecquando iste meas impellet nuntius auras ?

'
Ecce tuus Paulinus adest iam ninguida liuquit
:

Oppida Hiberoruin, Tarbelhca iam tenet arva,

Hebroniagi iam tecta subit, iam praedia fratris
Vicina ingreditur, iam labitur amne secundo
lamque in conspectu est iam prora obvertitur amni
:

Ingressusque sui celebrata per ostia portus.'

- D'Anville, Notice de Vanrienne Gaule, pp. 363-4.
3 Hist, dc Beam, 1640, p. 32 (x).
* Rev. arch., 3" scr., ii, 1883,
pp. 217-20.
•^
M. J. F. Blade {Annales de la Faculte des lettres de Bordeaux, 1893, p. 103)
believes that the Datii were not in the Aquitania of Caesar's time.
^ Geogr., ii, 7.
^
8ee Desjardins, Geogr. de la Gaule roni., iii, 157, n. '2, and facsimile facing
p. 158. The judgement regarding the date of the inscription is based upon
the form of the letters. fSce also liev. arch., 3^ scr., ii, 1883, pp. 216-7.
 TARBELLI 470

having regard to their geographical position, M. Longnon considers

that, in Caesar's time, the territories of the Aturenses (or Tarusates),
the Benarnenses, and^the Iluronenses, who, so he infers from Ptolemy's
not having mentioned them, were dependent peoples, were included
in the territory of the Tarbelli.
M. Longnon may
be right for Tibullus ^ and Lucan ^ allude to
;

the Tarbelli as a great people but M. Jullian,^ pointing out that the
:

Aquitanian tribes, for the most part, were much smaller than those
of the rest of Gaul,^ suggests that it was only under the Empire that
several smaller tribes were annexed to the Tarbelli and grouped with
them under their name.
—
Tarusates. The Tarusates appear to have been neighbours of the
Vocates,^ and are generally believed to have been identical with
the Aturenses of the Notitia provinciarwn.^ If so, they occupied the
country round Aire (Atura),'^ or the eastern part of the department
of the Landes and the western part of that of the Gers. D'Anville
and Walckenaer only assign part of the diocese of Aire the vicomte — '

de Tursan '
—
to the Tarusates, leaving the rest for the Oscidates
Campestres of Pliny ,^ who, says d'Anville, if we may judge from the
order in which the peoples among whom their name occurs are
enumerated, lived on the common frontier of the dioceses of Auch,
Bazas, and Aire.^ M. Longnon ^^ regards the Tarusates as dependants
of the TarbeUi (q.v.).
Desjardins^i says that Caesar's words, Grassus in fines Vocatium et
Tarusatium profectus est}"^ appear to show that the territories of the
Vocates (q.v.) and the Tarusates were not separated by the territory
of the Sotiates (q.v.), as they must have been if the Vocates dwelt
in the neighbourhood of Bazas, and the Tarusates in the neighbour-
hood of Aire. We must, then, he says, have the courage to admit
that we do not know where the territories of the Vocates and the
Tarusates were situated. I confess that I cannot follow this argu-
ment for, as Sos lies well to the east of both Bazas and Aire, the
;

Vocates and the Tarusates need not, in the case which Desjardins
supposes, have been separated by the Sotiates. Still, if Caesar's
statement was correct, the Vocates and the Tarusates were not
separated at all.

^ i, 7, 9. ^ i, 421. ^ Hist, de la Gaule, ii, 454 and nn, 5-G.

* Not counting the Datii, at least twenty-seven tribes occupied the pmall
area of independent Aquitania. See Amiales de la Facidle des lettres de Bor-
deaux, 1893, p. 104.
5 B. G., iii, 23, 27, « Ed. 0. Seeck,
§ 1. p. 271 (xiv, 9).
' '
According to all appearance,' says d'Anville {Notice de Vancienne Gaule,
p. 034), the name of the Tarusates is preserved in a district of the diocese of
'

Aire, under the name of Tursan.' Desjardins, however, says {Geogr. de la

Gaule rom., ii, (345, n. 2) that the identification of Tarusates with Aturenses
violates the most elementary phonetic rules.
« Nat. Hist., iv, 19
(33), § 108.
® Notice de Vancienne Gaule, Cf. Desjardins, Geoyr. de la Gaule rom.,
p. 510.
ii, 370-1. Pliny's order of enumeration proves nothing. See p. 420, supra.
^" Atlas hist, de la France,
p. 7.
" Ge.of/r. de la Gaule rom., ii, 045, n. 2.
''
B. G., iii, 23, § 1.
480 TARUSATES
P. do Marca,! who wrongly
assigns the diocese of Aire to the
round Tartas,^ which is
Sotiates, places the Tarusates in the country
on the Adour, about 15 miles north-east of Dax. He is guided,
I suppose, by the resemblance between the names Tarusates and
Tartas. See Vocates.
—
Tigurini. The Tigurini were one of the four tribes (pagi) of the
Helvetii.^ Caesar gives no clue to their geographical position. The
geographers of the seventeenth and eighteenth centuries Sanson, de —
Valois, d'Anville, and the rest* —
believed, for reasons which are not
worth mentioning, that they dwelt in the neighbourhood of Zurich.
But an inscription, containing the words PAai^i) TiuoR(inorum),^\\s.^
been discovered in the neighbourhood of Avenches, south of the lake
of Morat ;and accordingly most modern scholars hold that the
Tigurini dwelt in this part of Switzerland. Desjardins,^ however,
adheres to the old view. The fact that the inscription was found in
the neighbourhood of Avenches does not, he says, prove" that Avenches
was in the pagus Tigurinus. No doubt but it makes it more
;

probable that it was in that pagus than in any of the other three.
M. d'Arbois de Jubainville says that the Tigurini have left a trace
'^

of their settlement in the name of Tegernau, which is in the canton

of St. Gall. There is another Tegernau, about 15 miles NNE. of
Basle, where he believes that they had a settlement before the
Helvetii invaded Switzerland.
—
Tolosates. The Tolosates occupied a part of the territory of the
Volcae Tectosages, corresponding with the ancient diocese of Toulouse,
that is to say, the greater part of the department of the Haute-
Garonne and part of the department of the Gers.^
—
Treveri. The position of the Treveri is roughly indicated by
Caesar. He implies that their territory extended northward of that
of the Mediomatrici (q.v.) along the left bank of the Rhine he says :

that it was conterminous with that of the Remi (q.v.) and he adds
;

that it was separated (in part) from the Eburones (q.v.) by the
territories of the Segni and Condrusi (q.v.).^
Strabo ^^ says that the Treveri were conterminous with the Nervii
1 Hist, de Beam, 1640, pp. 35-6. Cf, C. Jullian, Hist, de la Gaule, iii, 307, n. 3.
- So also tentatively does M. Jullian, ib., ii, 451, n. 1.
« B. G., i, 12, § 4.
* See d'Anville, Notice de Vancienne Gaule, p. 643.
^ Th. Mommsen, Inscr. confoederationis Helveticae latinae, p. 29 {Mittheil. d.

ant. Gesellschaft in Zurich, x, 1854). Mommsen{ib., pp. 26-9) says that the
passage [B. G., i, 12) in which Caesar describes his defeat of the Tigurini proves
that they dwelt on the common frontier of the Helvetii and the Sequani,
namely in the neighbourhood of Avenches but any one who reads the passage
;

in question will see that it proves nothing as to the habitat of the Tigurini.
At the time of their defeat they and the rest of the Helvetii had emigrated from
their original abode, and they had left not only the territory of the Helvetii
but also that of the Sequani behind them. The one sound argument for
placing the Tigurini in the neighbourhood of Avenches is based on the discovery
of the inscription which 1 have mentioned in the text.
^ Geoijr. de la Gaule rom., ii, 240-1, n. 6 and 463.

^ Les Premiers Habitants de V Europe, ii, 1894,

pp. 75-6.
^ D'Anville, Notice, &c.,
pp. 648-9 A. Longnon, Atlas hist, de la France, p. v.
;

» B. G., X, 3, ^o
§ 4 ; 24, § 2; vi, 32, § 1. iv, 3, § 4.
 TREVERI 481

but in Caesar's time they certainly were not. The territories of the
Segni, Condrusi, Paemani, and Atuatuci separated the two peoples.
D'Anville i and Walckenaer ^ assign the Treveri that part of the
diocese of Treves which lies on the west of the Ehine but their :

limits cannot be defined, except on the south, in part on the east,

;

in part, where their territory was limited by the Rhine and on the ;

west, where it was conterminous with the territory of the Remi. The
difficulty of tracing the rest of their frontier arises from the fact that
we cannot tell whether they or the Mediomatrici possessed the
tract which afterwards belonged to the Vangiones (q.v.), and from
the fact that we cannot exactly define the territories of the Eburones,
Segni, Condrusi, Caerosi, and Paemani, with which no dioceses
correspond. Long ^ thinks that the rugged valley of the Ahr would
'

be a natural boundary of the Treveri on the north '. Zangemeister,^

on the other hand, argues that their northern boundary was the little
river Vinxt bach, between the Ahr and the Moselle. Roughly speaking,
their territory may be said to have comprised the greater part of the
province of Luxembourg, the Grand Duchy of Luxembourg, and the
southern part of Rhenish Prussia.
—
Triboci. The Triboci are mentioned by the unknown writer of
B. G., iv, 10 5 among the tribes who dwelt on the banks of the Rhine ;

and, according to Strabo,^ "they had taken up their abode on the

territory of the Mediomatrici (q.v.). D'Anville,' remarking that
Ptolemy s places them on the north of the Rauraci (q.v.), concludes
that they possessed the diocese of Strasbourg and Argentorate,
;

which stood upon the site of Strasbourg, was in their country.^

Near the common frontier of the dioceses of Basle and Strasbourg,
he observes, there is a place called Markolsheim, which means the
same as the French Feins (Fines). But it is not certain that in the
time of Caesar the Triboci occupied the same territory as in the time
of Strabo or in the time of Ptolemy and Desjardins ^^ thinks it
;

useless to attempt to fix their position, puisque Cesar nous les

'

montre sans cesse en mouvement.' Caesar, however, does no such

thing either directly or by implication. He only mentions the
Triboci once,^^' —
in his list of the tribes that fought in the host of
Ariovistus, which had for some time been settled in the northern

^ Notice de Vancienne Gaule, p. 652. ^ Geogr. des Gaules, i, 510.

^ W. Smith's Diet, of Greek and Roman Geogr., ii, 1227.

* Westdeutsche Zeitschrift/ iii, 317 (quoted by de Vlaminck, Messager des
sciences hist, de Belgique, 1887, p. 40, n. 1).
5 See p. 692.

* iv, 3, § 4. De Golbery [Mem. de la Soc. Roy. des ant. de France, v, 1S23,

pp. 142-3) thinks that the writer of B. G., iv, 10, in his enumeration of the
peoples past whose territories the Rhine flowed, mentioned the Triboci after the
Sequani and the Mediomatrici because they occupied portions of the territories
of both the two latter peoples ; and in support of this view he appeals to Strabo,
who says that 'SrjKoavol kcI MeSio/xaTpiKol KaroiKovffi rov 'Vrivov, kv oh i'dpyrai
TepfiaviKov e6vo^ . Tpi0oKxoi.
. , oh is generally referred to MeSiofxaTpiKoi. only ;

but de Golbery refers it to 'S.-qKoavoi as well.

' Notice de Vancienne Gaule,
pp. 653-4. ^ Geogr. , ii, 9, § 9.

9 See d'Anville, p. 96, and C. Miiller's ed. of Ptolemy, p. 230.

^^ Geogr. de la Gaule rom., ii, 621, n. 1. " B. G., i, 51, § 2.
1093 I i
 482 TRIBOCI
part of the territory of the Sequani. The doii))tfiil passage in whicli
they figure among the peoples who dwelt on the banks of the Rhine
gives us no help for who can tell when it was written ?
; After the
defeat of Ariovistus, the remnant of his host were driven across the
Rhine ;and, in the absence of any evidence, we have no right to
assume that, in Caesar's time, they returned. If, as I am inclined,
in view of Strabo's testimony, to believe, the Triboci remained upon
the left bank when Caesar wrote, then those who fought in the army
of Ariovistus were only a portion of the tribe but I cannot explain
;

why the remaining portion did not throw in their lot with their
brethren,^ and if they subsequently appear on the left bank, so do the
Ubii, who, in Caesar's time, were certainly on the right bank.^
Desjardins ^ argues that the Triboci could not have been on the left
bank in Caesar's time, because the territories of the Sequani, the
Mediomatrici, and the Treveri touched the left bank, and therefore
there could have been no room for the Triboci. But this is a bad
argument for if the Triboci were on the left bank, either they
;

occupied a part of the territory which Ariovistus had wrested from

the Sequani (see Nemetes) or, as Strabo ^ says, they had settled
in the country of the Mediomatrici. Cluver ^ maintains that they
had settled in Gaul before the invasion of Ariovistus, a theory for —
which there is not a particle of evidence. Stoffel,^ however, argues
that it enables us to understand how it was that some of the Triboci
remained in Gaul after the defeat of Ariovistus. They were, he
suggests, entirely distinct from the later invaders who fought
under the German king.
In any case, it is unsafe to attempt to determine the frontiers of the
Triboci, as they existed in the time of Caesar.'^ See Mediomatrici
and Sequani.
* See pp. 455-6.
^ Napoleon (Planche 2) may be right in assigning to the Triboci lands on both
banks of the river. Perhaps he was guided by the fact that the diocese of
Strasbourg (see Gallia Christiana, vol. v) included a tract on the east of the
Rhine. ^ Geogr. de la Gaulerom., ii, 445.

5 Gerrnania antiqua, ed. 1631,

* iv,
3, § 4. p. 365.
8 Guerre de Cesar et d* Arioviste,
pp. 88-9.
' In B. G., iv, 10 we read that the Rhine '
flows swiftly through the territories
of the Nantuates, Helvetii, Sequani, Mediomatrici, Triboci, and Treveri
{per fines Nantuatium, Helvetiorum, Sequanorum, Mediomatricorum, Tribo-
corum, Treverorum cifatus fertur). The word Nantuatium is probably corrupt
(see p. 454). Setting aside the Triboci, all the other peoples undoubtedly
lived on the left bank of the Rhine, and are mentioned in geographical order,
from south to north. But there is one word which, so far as I could discover,
had escaped the criticism of all the commentators before I called attention to it
in my first edition, —
I mean the word per. If the Triboci were all on the left
bank of the Rhine, not one of the peoples (with the doubtful exception of
N antuatium) whom the writer mentions were on the right bank and therefore
;

it would appear that he used per not in the sense of through ', but in the sense
'

of past or along '.

'
'
'
But there is not one single other instance in the whole
of Latin literature in which per is used in this sense. The inevitable conclusion
is that the Triboci were wholly or partly on the right bank, or that the writer
made a mistake or, finally, that he wrote praeter, the abbreviated form of
;

which was sometimes pter. I confess myself unable to decide between tliese
three possibilities.
 ;

UXELLODUNUM 483

Tulingi. — See Latobrigi.

Turoni. — The territory of the Turoni corresponded with the diocese
of Tours, roughly speaking, the department of the Indre-et-Loire.^
or,
Uxellodunum. — Hirtius says that Uxellodunum was in the
country of the Cadurci, which is represented by the modern department
of the Lot that the stronghold was protected on all sides by steep
;

rocks; that the ascent was difficult even whenunopposed; thatthrough

the valley, which nearly surrounded the hill, there flowed a stream ;

that the stream flowed at the foot of the hill {in infimis radicihus
mantis) in such a way that it was impossible to divert its course
that the descent to it for the townspeople was difficult and steep ;

that below the stronghold itself, on that part of the hill which, for
a space of about 300 feet, was not surrounded by the stream in —
other words, on a part which overlooked the isthmus there was —
a spring and that the Roman engineers drove subterranean galleries
;

towards the source of the spring, and diverted its flow.^ Orosius^
says that the river was of considerable size but his unsupported
;

testimony on such a point is of no value.

Now of the various sites that have been proposed several may be
dismissed at once, because their pretensions have been shown to be
absolutely groundless. Cahors * and Puy I'^veque have no longer any
supporters Uzerche ^ has never received any but the faintest support,
;

and besides it is in the Limousin, not the Quercy, and was therefore
outside the limits of the territory of the Cadurci. There are only four
about which there has ever been any serious discussion, Capdenac, —
Ussel, Luzech, and Puy d'Issolu.
I. To Capdenac,^ which is on the river Lot, about 35 miles east of
Cahors, there are several objections. (1) The place has never, so far
as is known, borne a name at all resembling that of Uxellodunum.
Champollion-Figeac ' affirms indeed that, according to a mediaeval
charter, which was preserved in the archives of Capdenac, the town
was formerly called Ucce-Lugdunum but he only cites the charter
;

as evidence that in 1320 tradition identified Capdenac with Uxello-

dunum. Moreover, MM. Creuly and A. Jacobs could learn nothing
of the charter at Capdenac itself the keeper of the archives at
:

Cahors, whom they also consulted, could not enlighten them and ;

it should seem that the charter was a forgery .^ (2) The hill is only

^ See Walckenaer, Geogr. des Gaules, i, 374.

2 B. G., viii, 32, §§ 1-2
; 33, § 1 ; 40, §§ 2-4 ;41, §§1,4; 43, § 4.
^ vi, 11, §21.

* N. Sanson {Les Comm. de Cesar,

pp. 90-4) advocated the claims of Cahors ;

but the Galhc town which stood upon the site of Cahors was, as every scholar
now admits, Divona. See A. de Valois, Not. Gall., p. Ill, and Diet. arch, de
la Gaule, i, 345. Moreover, the isthmus, if it can be so called, is not 300 feet,
but 700 metres or about 2,300 feet wide.
5 Uzerche is in the department of the Correze, between the river Vezere and

the river Bradascon. The width of the isthmus is about 500 metres, or more
than five times too great.
« See Carte de VEtat-Major (1 : 80,000), Sheet 195.
' Nouvelles reeherches sur la ville gauloise d' Uxellodunum,
1820, pp. 96-7, 110-1.
* Exam en des lieux proposes pour representer Uxellodunum, 18G0,
pp. 16-7 ;
Rev. d' Aquiiaine, ix, 1865, pp. 101-2.
I i2
 484 UXELLODUNUM
protected by steep rocks on the east and the west and on the ;

north there was nothing to prevent the Roman army from under-
taking a regular siege. (3) Assuming that Uxellodunum was at or
near Capdenac, the description of Hirtius would lead one to suppose
that the stronghold was at Vic, which is at all events nearly sur-
rounded by the river Lot.^ But Vic is completely dominated by
Capdenac and accordingly Champollion is compelled to place the
;

stronghold there. In this case, however, the statement of Hirtius,

that the stronghold was nearly surrounded by a stream, must be
disregarded while the isthmus which separates Capdenac from Vic
;

is more than twice too broad.^ Moreover, the area of the plateau of
Capdenac is only 3 hectares, or about 7| acres ^ and this is, of ;

course, too small to have admitted of the existence of an ofpidum

which accommodated more than 5,000 men.^ (4) The only spring
which Champollion could find to identify with the spring described
by Hirtius is more than 100 yards from the nearest point through
which the wall of the town could have passed.^ It would have been
completely exposed to the fire of the Roman artillery and archers ;

and the terrace which Caesar constructed with such difficultv would 4/

have been unnecessary.^

^ Examen, &c., pp. 18-21.

In order to force the narrative of Hirtius into conformity with the geography
^

of Capdenac, Champollion (pp. 50, 81) perverts the meaning of the famous
passage, magnus fons aquae prorumpebat ah ea parte quae fere pedum CCC
intervallo a fluminis circuihi vacabaf, in a manner which is simply astounding.
He says that when Hirtius spoke of the isthmus as 300 feet wide, he was only
thinking of the terrain on which the Roman trooj^s could manoeuvre and
construct their siege works (' Si Ton admet que I'historien latin n'a pu
. . .

parler que de I'etendue du terrain sur lequel les troupes romaines pouvroient
manceuvrer et les travaux de siege s'executer, ce sera de la superficie de I'isthme
qu'il faudra connoitre I'etendue. Selon Cesar {sic), I'isthme d' Uxellodunum
etoit de 300 pieds celui de Luzech n'en a plus de 40 ').
;

^ Champollion-Figeac, p. 74.

* The garrison of Uxellodunum consisted of the oppidani, or townsfolk, who

were there when Lucterius arrived, and of the force which he brought with him.
Hirtius does not estimate the number of the oppidani. He says that when
Drappes and Lucterius stole out of the fort to get corn, they left 2,000 armed
men behind and took the rest with them [B. G., viii, 34, § 2) and the words ;

dnohus milibus armatorum relictis seem to imply that there was an indefinite
number of non-combatants as well indeed, it is obvious that this must have
:

been the case. Hirtius also says that when Drappes and Lucterius started on
the flight that led them to Uxellodunum, they had not more than 5,000 men
'
'

{non amplius liominum milibus ex fuga quinque coUectis. lb., 30, § 1) or,
according to the iS MSS., 2,000. Accepting the smaller estimate, and remem-
bering that a sufficient force must have been left to hold the fort in the absence
of Drappes and Lucterius, I conclude that the whole garrison, including non-
combatants, numbered at the very least 5,000.
Nouvelles recherches, &c., pp. 81, 83.
5

The Comte de Caylus (A. C. P. de Tubieres de Grimoard), overlooking all the

^

other objections which I have stated, says, La seule difficulte est que cette
'

fontaine etoit situee dans la partie de la montagne qui n'etoit point environnee
de la riviere dans un espace d' environ 300 pieds. ... La fontaine de Capdenac
. .n'est pas precisement dans cette situation
. raais on pent croire que cette
;

fontaine ayant ete coupee et detournee par les Remains sa source aura . . .

change de place.' Eecueil d'ant. egyptiennes, &c., 1752-G7, v, 280. One can
believe anything, when the wish is father to the thought,
 — ;

UXELLODUNUM 485

II. Ussel, like Uzerclie, is in the Limousin, and was therefore ahnost
certainly outside the country of the Cadurci. Creuly and Jacobs,
who personally examined every single locality that could conceivably
be identified with Uxellodunum, considered that it in no way corre-
sponded with the description of Hirtius, and that its claims were not
worth discussing.^ Still, as it is just possible that the territory of
the Cadurci may have extended beyond the boundaries of the
diocese of Cahors, and as, since Creuly and Jacobs wrote, the claims
of Ussel have been twice advocated, it may be well to examine them.
— —
Ussel or rather the plateau of Peyrol near Ussel is described as
follows by Colonel A. Sarrette.^ The plateau, which is 2,190 feet
high, 1,54:0 yards long from north to south, and from 440 to 770 yards
wide, is protected by sheer scarped rocks, and almost completely
isolated by a deep valley. On the west flows a little stream, the
Sarsonne, and on the east un tres-petit affluent, faible file d'eau '.
'

The assumed oppidum was only accessible by a narrow ridge on the

north-east. On this ridge was a spring, which gushed forth at the
foot of the assumed wall. The three camps of Caninius^ were on
the north of the oppidum, on the heights of Sarsonne and Teil,
which dominate Peyrol. I need hardly add that, in the environs of
Ussel, the inevitable antiquites gauloises et gallo-romaines have
'
'

been discovered.
Now there is one feature in this description which is fatal to the
assumed identity of Uxellodunum and Peyrol. The position of the
spring cannot, by any ingenuity, be forced to agree with the descrip-
tion of Hirtius, majnus fans aquae prormnpehat ah e a parte quae
fere pedum CCC intervallo flufninis circuitu vacahat. Colonel
Sarrette struggles to meet this objection by saying that the word
Jiuminis was substituted by some blundering copyist for vallis.^ But
this will not do. Unless a text is repugnant to reason or to undeniable
and essential facts, one has no right to alter it, merely because it
refuses to square with one's pet theory. Moreover, at Peyrol there
is no isthmus and a tunnel, by which the source of the spring
;

was cut, has been discovered at Puy d'Issolu, and has not been
discovered at Peyrol. In the face of these objections, I do not see
how any unbiased inquirer can identify Uxellodunum with Peyrol.
III. Luzech was the site adopted by General Creuly and Alfred
Jacobs,^ acting as the representatives of the French Commission.
The town of Luzech is about 7 miles west-north-west of Cahors.
It is situated on an isthmus at the foot and on tbe north of a hill
which rises 87 metres, or 287 feet, above the level of the Lot, and is
nearly surrounded by that river. The isthmus, according to Creuly
and Jacobs, is 330 feet wide but this estimate is, I believe, too
;

low.6 The hill itself, according to the same authorities, occupies

^ Examen, Sec, pp. 14-5.
^ Quelques pages des comm. de Cesar, 1863, pp. 228, 244-5, 263.
3 B.G., viii, 33, § 2.
Quelques pages, &c., p. 237. ^ Examen, &c.,
*
pp. 27-32.
'^
J. B. Cesyao, who holds a brief for tho Puy dTssolu, asserts {Uxellodunum^
— aper^us critiques, &c., 1862, p. 44) that the width is 177 metres, or 621 feet
486 UXELLODUNUM
some 17 hectares, or 42 acres, —
not more than one-third of the pcnin-
siila.i The spring described by Hirtius is, they maintain^ represented
by a slight oozing, so to speak, of water, which overlooks the isthmus.
Finally, it has been asserted that trenches have been discovered,
uniting those points on the north, east, and west of the hill, where the
three camps of Caninius have been conjecturally placed but I have
;

not been able to discover any proof of this assertion.

In many respects Luzech does not conform to the description of
Uxellodunum. The one point Avhich tells strongly in its favour is

that, alone among

the sites that have been proposed, it is almost
all
entirely surrounded by a
river, and that the breadth of the isthmus
approximately corresponds with the statement of Hirtius. At first
sight, indeed, as has been remarked by a skilled observer ,2 the traveller
might fancy that he saw Uxellodunum rising before him but a ;

closer inspection suggests numerous doubts. First, the hill is only

scarped on its northern side the eastern face, indeed, is steep
: but ;

on the south and west the slopes are so gentle that even carriages can
ascend them without difficulty.^ If it is objected that the hill may,
in Caesar's time, have been as steep as Uxellodunum, the answer is
that the rocks of which it is composed are so hard that they cannot
have suffered any considerable change of form.'* But, it has been
replied, Hirtius does not distinctly say that the hill itself was steep
{praeruptum) he only says that the oppidum which stood upon the
:

'
hill was praeruptum, and that it was protected by very steep rocks
'

(praeruptissimis saxis munitum). Von Goler,^ remarking that saxis

is the word which Hirtius uses, not rupibiis, argues that he could not

have meant to describe a naturally scarped plateau (like that of the

Puy d'Issolu), but rather boulders laid by the hand of man. The
distinction between saxum and rupes is, however, not invariable ^ ;

and Caesar does not always observe it.*^ Whether Hirtius did, we
cannot tell. Secondly, the river does not flow, as the text requires,
at the very foot of the hill ^ the least distance which separates it
;

from the bank is over 100, the greatest 500 yards. What becomes,
then, of Hirtius's statement that the garrison could only approach
the river by a steep and difficult descent ? To this objection also
von Goler ^ has an answer. He understands the words of Hirtius as
applying, not to the descent from the hill to the bank of the river,
but to the descent from the bank to the water. Such an interpreta-
tion is perhaps just possible but it is certainly not the one which
;

would naturally be put upon Hirtius's words. Thirdly, the summit

of the hill is not a plateau at all, but, so to speak, a platform, only
44 yards long by 15 or 16 wide, and barely capable of accommodating

but he is mistaken, as any one who examines Sheet 194 of the Carte de V Etat-
Major (1 80,000) may see for himself.
:

Cf. the next page and p. 388, n. 9.

- ReVi d' Aquifaine, x, 1806, p. 14. ^ lb., ix, 1865, p. 251.

* Jb., X, 1866, p. 16. 5 QaU. Krieg, 1880, p. 359.

^ See Forcellini, Toliud iuliniialin lexicon, \, 1871, p. 355.

'
Sec B. C, i, 68, § 2. ^ in imis radicibus moutis.

« Gall. Kricj, 1880, p. 365, n. 1.

 UXELLODUNUM 487

200 men. Evidently this narrow space cannot have been the site
of the oppidum. Nor could the oppidum have been placed, even
partially, upon the flanks of the platform for they are too steep.
;

The platform, then, could only have been a citadel, and the oppidum
itself could only have stood upon the lower plateau, called La Pistoule.^
Fourthly, Captain Gallotti maintains that, at the point which Creuly
and Jacobs indicate as the site of the spring, and indeed at any
point of the hill facing the isthmus, it is geologically impossible that
there could ever have been a spring. La partie du monticule,' he
'

says, faisant face a I'isthme ne presente pas un talus d'une certaine

'

largeur, mais simplement une arete anguleuse comme le tranchant

d'un soc de charrue. Toutes les couches qui composent la masse
sont inclinees du nord au sud, dans le sens oppose a I'isthme. II
y a solution de continuite entre Ces couches et celles qui leur corres-
pondent dans le massif voisin. Malgre ces faits, il fallait absolument
trouver I'emplacement de la fontaine vis-a-vis de I'isthme, et cela
ne pouvait se faire qu'en la pla9ant sur ou contre I'arete anguleuse
dont nous venons de parler. non-seulement en un point presque
. .

isole dans I'espace, n'ayant aucune couche de terre soit au-dessus,

soit a cote de lui, mais encore se trouvant situe au point le plus eleve
du banc calcaire auquel il appartient, lequel banc ne communique
a aucun autre du terrain circonvoisin. D'oii I'eau a-t-elle jamais pu
venir pour alimenter la une fontaine ahondante ? Peut-etre,
. . .

objectera-t-on, qu'il a pu exister la une source artesienne naturelle

tirant ses eaux du massif du nord. C'est une objection a laquelle la
stratification du sol repond victorieusement. Ce phenomene ne pour-
rait exister que dans le cas ou les couches sedimentaires du nord,
s'inclinant vers I'isthme, se redresseraient ensuite pour former le
monticule. .Mais ... les bancs constitutifsdu sol sont horizontaux
. .

dans le massif du nord et inclines vers le sud dans celui de Luzecli,

ce qui rend la supposition impossible.' - But I should warn the reader
that Sir Archibald Geikie, who was so kind as to consider this argu-
ment at my request, does not regard it as conclusive. There may '

of course,' he writes, be features of geological structure, not alluded

'

to in the quotation you send me, which would show that the water
could only flow along the planes of bedding. In that case, the argu-
ment would be valid enough. But in most limestone countries the
water not only flows along the planes of bedding but rises through
joints and gradually dissolves among their sides, forming open
fissures, caverns and tunnels. It seems to me therefore quite possible
that, supposing the form of the surrounding ground be favourable,
a considerable body of water might issue from a line of joint on the
spot indicated in your diagram.' Fifthly, even assuming that there
was a spring at the point in question, Gallotti denies ^ that it could
have been the spring which Hirtius describes for the Gauls could;

not have come down in force to attack their assailants, as the only
approach was a narrow arete ; the assumed spring was not more

^ Rev. (T Aquitimie, x, 1800, pp. 17-8.

'
- lb., pp. 180-1. lb., p. 188.
488 UXELLODUNUM
than 13 or 14 feet above the level of the isthmus and the gigantic ;

works which the Romans constructed ^ in order to place themselves

on a level with it would therefore have been unnecessary. Sixthly,
there is a hill immediately north of Luzech and considerably higher,
on which are the remains of a Gallic fortress, known as Impernal,
which stands in the same relation to Luzech as Capdenac does to
Vic, and the close proximity of which makes it most unlikely that
Luzech was a stronghold at all.'^ The seventh objection has not, so
far as I know, been urged before but it seems to me, if not un-
;

answerable, the most weighty of all. If Uxellodunum was on the

peninsula of Luzech, or on any peninsula with an isthmus only 300
feet wide, how could 3,000 men have got out of it, how could a long
string of pack-horses have got into it again, when it was invested by
two Roman legions ?
Four other objections have been brought against this site but ;

they seem to me to have comparatively little weight. The hill of

Luzech, we are told, is far too low to have been described as a mons ;

and, as some Roman cohorts climbed the slopes of Uxellodunum,

with the object of making a feigned attack upon the fortress, the
whole circuit of the river must have been fordable, whereas the Lot
could not have been forded.^ In answer to the former objection,
I may point out that Caesar describes the hill which he occupied in
the battle with the Helvetii as a mons,^ and that this hill was certainly
not more than 300 feet above the level of the plain.^ With regard to
the other objection, it has been strenuously maintained,^ and as
strenuously denied,'^ that the Lot, in the sweep which it makes round
the hill of Luzech, is fordable in the autumn, the disputants appa-
rently forgetting that the siege took place in the summer.^ But was
it necessary that the assaulting cohorts should cross the river at
all ? Could they not have made their way into the peninsula by
way of the isthmus ? Again, it has been objected that the Lot, in
that part of its course w^hich encircles the peninsula of Luzech, is
—
more than 150 yards broad, so broad that the archers of whom
Hirtius speaks could not have shot across it with effect. But with
the Turkish bow flights of from 600 to 800 yards have been made :

—
even the English long bow a much less powerful weapon is known —
to have attained a range of 310 yards and Sir Ralph Payne- Gall wey
;

admits that our mediaeval archers could shoot 230 or perhaps 250

^ B. G., 41-2.
viii,
^
Some writers have contended that Impernal was Uxellodunum
itself
(P. Joanne, Diet, giogr. . . . de la France, iv, 1896, p. 2367, and
s.v. Luzech) ;

as conceivable that Hirtius, if

it is he had not seen the stronghold, might
have so far misunderstood his authority as to say that Impernal, which is
washed by the Lot on two sides, was almost entirely surrounded by a stream,
this theory might be defensible but for one stubborn fact, there was nothing —
to prevent Caesar from undertaking a regular siege and erecting an agger on
the northern side. See Congres arcMol. de France, 1874 (1875), pp. 444-6.
^ Rev. d^ Aquitaine, ix, 1865,
pp. 240-1.
' E.G., i,
24, § 3. 6 See pp. 625-7.
" Fxamen, &c., 30-1.
pp.
' Rev. d' Aquiiaine, ix, 1865, p. 243, B. G., vii, 46, §§ 1-2.
'^
 -

UXELLODUNUM 489

yards.i Lastly, R. Perie points out that the struggle for the posses-
'^

sion of the spring took place on an elevated position {excelso loco) ^ ;

whereas the isthmus at Luzech is nothing of the sort. But the

Romans were fighting on the agger, which might fairly have been
called an excelsus locus.
IV. The Puy d'Issolu, near Vayrac, is isolated on every side except
the north-east, where it is connected by a col with the Pech Demont,
and risesabout 650 feet above the valleys. Its summit is an un-
dulating plateau, covering 80 hectares,^ or about 198 acres, and
nearly surrounded by steep escarpments of rock. On the north, east,
and south, the plateau is practically impregnable while the western ;

slopes, where, between the hamlets of Loulie and Leguillat, there is

a considerable break in the escarpments, are nevertheless sufficiently
steep to justify the statement of Hirtius that, even if no resistance
had been offered, the ascent would have been difficult for armed
men. On this side, about 72 feet below the line which the walls of
the alleged ofpidmn would have followed, there was a spring. The
Dordogne flows past the south of the hill, about a mile from the
nearest point of the plateau, and 1,200 yards from the foot of the
slope. The western slopes are bathed by a stream, about 10 metres,
or 11 yards, wide, called the Tourmente, which, flowing through
a narrow valley, hemmed in on the right by another steep hill,
empties itself into the Dordogne, and is itself fed by a rivulet, which
bathes the north-western side of the hill. The southern and south-
eastern slopes are washed by a little stream, the Sourdoire, which,
like the Tourmente, empties itself into the Dordogne.
The hill itself corresponds almost exactly with the description of
Hirtius. It has received the greatest number of suffrages its name :

closely resembles the Uxello in Uxellodunum, and its character is

exactly described by that word, which means a fort on a pre- '

cipitous hill ' a charter of the abbey of Tulle, dated 944, proves
:

that it was then called Uxellodunum,^ and, as J. B. Cessac remarks,

there is no evidence that any other place in the country of the
Cadurci was ever called by any such name ^ a tradition, which is
:

at all events some centuries old, identifies the Puy with Uxello

^ Sir R. Payne-Gallwey, A Treatise on . Turkish and other Oriental Bows,

. .

pp. 19-20 (the second part of his Projectile-throwing Engines, &c., 1907). Cf.
L'Anthr., xviii, 1907, p. 734 (summary of a paper in Globus, xci, 1907, no. 21).
2 Hist, du Quercy,
1861, pp. 54-62. » B. 0., viii,
42, § 4.
* Napoleon, Hist, de Jules Cesar,
ii, 343.
^ Inter praecipuas Veiracum (Vayrac), Mayronam et Wogaironum, in quorum

vicinia, scilicet in podio vocato Uxelloduno, ubi olim civitas Romanorum

obsidione nota. D. Bouquet, Recueil des hist, des Gaules, ix, 580 d.
« Mem. lus a la Sorhonne, 1866 Mr. E. H. Barker says
(1867), p. 91, note.
{Two Summers in Guyenne, 1894, p. 75) that an educated native of Vayrac told
him that the stream
' '
—
I suppose the Tourmente
—
where it issues near the
'

base of the rocky height has been known in the neighbourhood from time
—
immemorial as " lo foun Conino ", Conino's fountain. Conino is a natural
Romance corruption of Caninius.' I attach less importance to this tradition
than Mr. Barker appears to do. Time immemorial is a vague expression
'
'
;

and the stream may have been named after Caesar's lieutenant by some anti-
quary of the seventeenth century.
490 UXELLODUNUM
dunuiii numerous coins have been found on or quite close to it ^
: :

it is a position of great strength it commands the northern entrance

:

to the territory of the Cadurci and therefore it is the very strong-

;

hold which Lucterius might have been expected to choose. Paths

such as those by which Hirtius says that the rebel chief endeavoured
to throw his convoys into the town are easily recognizable. The
Tourmente could not have been diverted at any point of its course
which the besieged garrison could have reached.^ The Sourdoire
could not have been approached by the garrison at all, because the
sheer rocks which overlooked it were practically inaccessible, and if
the garrison had attempted to reach it by a circuitous route, they
must have been exposed, going and coming, to the fire of the Roman
archers and artillery the Dordogne was too far off
: and access to ;

it would have been barred by the Roman Hues. Traces of those lines
or, at all events, of some lines of investment, have been discovered ;

and also a gallery driven through the western side of the hill to the
source of the spring.^ Creuly and Jacobs, indeed, deny that an army
so small as that of Lucterius would have taken refuge on a hill so
large as the Puy * but in this argument there is no force
; the Puy :

was so strong that 5,000 soldiers, backed by an indefinite number of

townsmen, could have easily defended it. It might also perhaps be
objected that if the Romans had made a feint of storming the Puy

on the garrison, unless fear had deprived them of their

all sides,
senses, would only have laughed at their efforts. If, then, Uxello-

dunum stood upon the Puy d'Issolu, the words ex omnibus partibus,
in Hirtius's 4:3rd chapter, must be inaccurate for the assailants ;

could hardly have mounted except by the paths of Loulie, Leguillat,

and les Tourneries, and by the farm of Ronjon on the north-east.^
But Hirtius, like Caesar, might naturally have used the word omnibus
rhetorically and it must be remembered that the garrison would
;

have been small in proportion to the size of the phiteau. On the

other hand, it is conceivable that the works of which traces have
Qimw discovered were not made by Caesar and his lieutenants at all
and no unprejudiced inquirer can deny that, in two res])ects at least,
the site does not conform to the description of Hirtius. He ex-
})ressly says that Uxellodunum was almost entirely surrounded by
a river and this cannot be said of the Puy d'Issolu. P. Bial, in-
;

deed, affirms that every year, about the beginning of October, the
^ N. L. Achaintre, Caesar, iv, 381, 386-7 ; A. Blanchet, Traite des monn.
ijanl., p. 288, n, 1.
- Walckeuaer {Gl'oyr. des Gaules, i,
355) says that the Tourmente could have
been diverted but look at Sheet 183 of the Carte de r£!tat-Major (1 80,000),
; :

and read Napoleon, Hist, de Jules Cesar, ii, 343-4.

^ See Napoleon, Hist, de Jules Cesar, ii, 343-7, and Planches 31-2. This gallery,
which was minutely explored by Cessac, was already known in the sixteenth
century {Bew des eludes anc, x, 1908, p. 352). *
Examen, &c., p. 25.
5 C. Jidlian, Hist, de la Gaule, iii, 557, n. 3; 5G2, n. 3. M. Tamizey de
Larroque, quoting the Abbe de Vayrac, says that at the Puy d'Issolu the
rocks are tellement escarpes, qu'on n'y pent monter qu'engrimpant et parde
'

petits scntiers qu'on a pratiques dans Ic roc en quclqucs endroits carpartout ;

ailleurs, les rochers sont aussi perpendiculaires que les tours de Notre-Dame
de Paris.' Rev. d'Aquitaine, ix, 1805, p. 108.
 — ;

UXELLODUNUM 491

Toui'inente overflows its banks, increases the volume of a little rivulet

called the Hierle, and thereby mingles its waters with those of the
Sourdoire, in such a way as to form une seule echarpe d'eau encei-
'

gnant la montagne ^ but Creuly and Jacobs say that, when they
'
;

visited the Puy, in November, the Tourmente and the Sourdoire

did not commingle their waters ^ and no unbiased critic would lay
;

any stress upon Bial's argument.-^ Again, Hirtius describes the

spring as situated ab ea parte quae fere pedum CCC intervallo fluminis
circuitu vacahat. Napoleon explains this passage as meaning that
the spring was 300 feet from the Tourmente * but he has to admit :

that the spring was 1,000 feet from the Tourmente and therefore ;

he is forced to read passuum CC instead of pedum CCC. Moreover,

no scholar could accept his translation. The Latin can only mean
that the spring was on that side of the hill which, for an interval of
300 feet, was not surrounded by the stream. Von Kampen,^ remark-
ing that, just above the spring, there is a break of 300 feet in the
escarpment, proposes to substitute rupium ior fluminis. But in that
case Hirtius's statement would have no point. Another emendation
has been proposed by Rudolf Schneider (ab ea parte quae fere):
—
passuum CC intervallum a fluminis circuitu habebat.^ In a passage
which is supported by the unanimous authority of all the MSS.
Schneider alters four words out of six and puts in another He !

changes pedum into passuum, CCC into CC, intervallo into inter-
vallum, vacabat into habebat and he puts in a. Now Hirtius may
;

not have been a stylist but he did not write schoolboy's Latin
:
;

and if there were no fault to find with Schneider's Latinity, he would

still say to himself, if he were a man of humour, It is long odds '

that this emendation of mine, although it may make people say

that I am a clever fellow, is not what Hirtius wrote.' Moreover, "^

I would ask Schneider what Avould have been the relevancy of the
statement which he attributes to the historian. Is it not obvious

^ Mem. de la iSuc. (T e viiilaiion du Doubs, 3'"

ser., iii, 1859, p. GOO.
- Examen, &c., pp. 30-1.
^ Cessac {Uxellodimum, —
a per fus criiiqaes, &c., p. G7) argues, on geological
grounds, that the Dordogne itself once surrounded tlie whole of the Puy except
the col on the north-east. But it is absurd to suggest that it did so in 51 b. c.
and even supposing that it did, the spring, according to Hirtius, must have
been on the col, whereas Cessac, as the result of his excavations, identified the
spring mentioned by Hirtius with the one on the western slope !

* Hist, de Jules Cesar, ii, 344.

^ Quindecim ad Caesaris de b. G. Comm. tabulae, xv. As a matter of fact, the

extent of the break is much more than 300 feet. See Ch. Lentheric, La Grece et
r Orient en Provence, 1878, pi. ii.
^ Jahresb. d. philol. Vereins zu Berlin,
xiii, 1887, pp. 362-4.
' Cessac
( Uxellodunum,
—
observations touchant les fouilles executees a Luzech,
1863, pp. 12-3) appeals to the authority of Orosius as a proof that there was
no isthmus at Uxellodunum. Orosius (vi, 11, § 21) describes the position in
these words : hoc oppidum in editissima rnontis arce pendehat, duabus partibus
per abrupta latera non parvo flumine cingebatur. Duabus partibus are the
'
'

words on whicli Cessac relies. I believe that Orosius's description was merely
a loose and rlietorical paraphrase of tliat of Hirtius but at all events it must
;

be clear to any one wJio has the slightest critical faculty that the authority of
Orosius cannot be weighed against the autliority of Hirtius.
492 UXELLODUNUM
that what Hirtius intended to convey was that the gap in the circuit
formed by the river enabled Caesar to construct a terrace there,
and that he could not have done so anywhere else ? To any unpreju-
diced mind it must be clear that either Hirtius made a gross
blunder or the Puy d'Issolu is not Uxellodunum. Now Long suggests
that Hirtius may never have seen Uxellodunum, and that his in-
formation or his interpretation of it may have been at fault. ^ Creuly
and Jacobs, on the contrary, insist that Hirtius's narrative is worthy
of the fullest confidence. His Preface, they argue, proves that he
was a writer of scrupulous accuracy for he tells us that he would
;

not undertake to describe the African and Alexandrian wars because

he took no part in either of them.^ What Hirtius really said was
that he took no part either in the Alexandrian or in the African war ;

and that, although he had gained some knowledge of them from

conversation with Caesar, he listened to Caesar's story for the
pleasure of hearing him, not with the object of writing history.^
But, if this statement implies that Hirtius did take part in the
eighth campaign of the Gallic war, he could not have been present
in every operation of that campaign and there is no proof that he
;

ever saw Uxellodunum.

Creuly and Jacobs also maintain that we must either accept
Hirtius's narrative as it stands, literally in every detail, or else can-
didly admit that the question is insoluble. If, they argue, Hirtius
was mistaken in describing the isthmus, how can we tell that he was not
also mistaken in saying that Uxellodunum was in the country of the
Cadurci ? * No experienced student will accept the principle which
the authors lay down. Every historian makes mistakes and in ;

describing facts of a certain class it is very difficult to avoid error,

while in narrating others it is almost impossible to go astray. For
instance, an historian of the Indian Mutiny might, if he had never
seen Delhi, make some slight mistakes in describing its principal
features. He might place the Jamma Masjid too near the Jumna ;

but he would hardly say that the Jumna almost entirely surrounded
the town. In short, it is not necessary to fling a book to the other
end of the room because one suspects the writer of having made
mistakes and it is necessary, in reading the most careful writer, to
;

use one's critical faculty. At the same time I freely admit that the
particular mistake which an advocate of the Puy d'Issolu must, if
he honestly translates the Latin, believe Hirtius to have made, is one
which is very difficult to understand.
And yet it is as certain as the Binomial Theorem that Uxellodunum
did stand upon some one of the sites which I have examined. Scan
closely the sheets of the great Carte de V Etat- Major which embrace
the department of the Lot and the surrounding country, and you will
not be able to find any other site which is even worth discussing.
Every conceivable site has been carefully examined by keen eyes
and of the whole number there are, as we have seen, only four which

^ Decline uj the Human Republic, iv, 385. " Exanicii, etc., p. 24

^ B. G., viii, Praef., § 8. •*
Exanien, &c., p. 23.
 UXELLODUNUM 493

have secured any real support. Two of the four have been, on closer
examination, unhesitatingly rejected and the final choice lies
;

between Luzech and the Puy d'Issolu. Luzech is the only place
where there is an isthmus even approximately corresponding with
the description of Hirtius the Puy d'Issolu is the only place which,
:

in other respects, corresponds or even approximately corresponds

with his description. It seems almost incredible that he should have
described Uxellodunum as a peninsula surrounded by a river on
every side, except an isthmus only 300 feet wide, if there was no
isthmus at all. The mistakes which he is assumed to have made are
nothing less than monstrous. If he had merely said that the hill was
almost entirely surrounded by a river, I should say that there were
ten chances to one against his having described the Puy d'Issolu in
such terms. But when I consider that he made the further state-
ment that the spring was just above the isthmus, I am inclined to
say that there are ten times ten chances to one against his having
been mistaken. On the other hand, I am obliged to admit that, if
Luzech was Uxellodunum, he made gross mistakes in his description.
Yet, compared with those other assumed mistakes, they are perhaps
easily explicable. But again, I am constrained to assume that
Caninius, whose generalship throughout the campaign had been
admirable,^ was so careless, or rather so imbecile as to let Lucterius
get out of the place with 3,000 men, when he could have shut him in
with the greatest ease.^ Lastly, I am obliged to admit that the spring
which he described is not to be found ^ and that, if it was there,
;

his account of the struggle which took place for its possession is
absolutely incomprehensible. When I turn again to Cessac's cir-
cumstantial account of the way in w^hich the spring on the Puy
d'Issolu was diverted when I consider the combined force of all
;

the other arguments that have been urged by him and his fellow-
—
advocates the argument from tradition, from the charter which
mentions Uxellodunum, from the discovery of lines of invest-
ment, from the great strength of the Puy d'Issolu, the difficulty
of ascending it, and the geographical position, which would have
recommended it to Lucterius, lastly, the obvious resemblance of its
—
name to Uxellodunum when I consider all these things, I can
no longer hesitate. Hirtius did make that mistake which seems
all but incredible and Uxellodunum is to be identified with the
;

Puy d'Issolu.
Vangiones. —The Vangiones, one of the German peoples who con-
tributed a contingent to the host of Ariovistus,* occupied, in the time
of Ptolemy,^ the country round Borbetomagus (Worms). It is

1 Cf. B. C, ii, 34, § 4.

^ I am
glad to find that R. Schneider has anticipated me in this argument.
See Jahresb. d. philol. Vereins zu Berlin, xiii, 1887, p. 362.
^ Luzech,' sarcastically remarks Cessac {Uxellodunum retrouve, 1865, p. 13),
'

'
eut la prudence de ne pas jeter en pature aux hasards de la pioche le subside
complementaire mis a sa disposition par un membre de la commission de la
carte des Gaules.' « B. G., i,
51, § 2.
5 Geogr., ii, 0,
§ 9. See d'Anville, Notice de Vancienne Gmde, p. 171.
494 VANGIONES
doubtful, however, whether they were settled there in the time of
Caesar. See Nemetes and Tribooi.
—
Veliocasses. The Veliocasses dwelt in that part of the diocese of
Rouen which did not belong to the Caleti (q.v.), namely the southern
part of the department of the Seine-Inferieure and the eastern part
of the department of the Eure. Their name survived in the old
'
Vexin '.1

—
Vellaunodunum. Caesar says but little that can help us to find
Vellaunodunum. He tells us distinctly why he laid siege to it he :

wished to avoid leaving an enemy in his rear, in order that the sup-
plies which he expected might reach him more quickly. He says
that Vellaunodunum was a stronghold of the Senones that it was
;

on the road which he took from Agedincum (Sens) to Gorgobina, the

stronghold of the Boi and that, after leaving his heavy baggage at
;

Agedincum, he reached Vellaunodunum altero die. As the garrison

held out for two days, while he was making a line of contra vallation,
we may perhaps infer that the place was fairly strong. Finally he
says that, after receiving the submission of Vellaunodunum, he
marched thence in two days to Cenabum (Orleans). On pages
'^

—
738-40 I prove that altero die means the day after ', that if, for
'

example, Caesar left Agedincum on the 1st of the month, he reached

Vellaunodunum on the 2nd. It might be said that we are none the
wiser for this knowledge for Caesar may have arrived at Vellauno-
;

dunum early on the day which he calls altero die, or he may have
arrived late and the difference in time would involve a consider-
:

able difference in distance. Still, as Caesar marched the whole way

from Agedincum to Cenabum in four days, and as C^enabum was not
less than 108 kilometres, or about 67 miles, from Agedincum,^ it is
not likely that he made a short march on any one of the four days ;

and we may therefore perhaps infer that he did not reach Vellauno-
dunum On the other hand, as
until comparatively late in the day.
he took two whole days to march from Vellaunodunum to Cenabum,
and arrived at Cenabum too late in the afternoon to begin the siege,
it is tolerably certain that Vellaunodunum was at least as far from
Cenabum as it was from Agedincum and, as he says that, when he
;

arrived at Vellaunodunum, he oppugnare instituit, which apparently

means that he began his preparations for the siege on the day of his
arrival,General Creuly^ may be justified in inferring that Vellauno-
dunum was nearer to Agedincum than to Cenabum. It would be
more to the purpose to ascertain the road by which Caesar
marched. There was a Roman road from Agedincum to Cenabum,
108 kilometres long, which passed by Chateau-Landon, Sceaux,
and Beaune. The road mentioned in the Table ^ passed through
Aquae Segeste and a little to the north of Gien its length, accord-
:

ing to the MSS., was 59 Gallic leagues, or about 131 kilometres

^ See d'Anville, p. 684, Walckenaer, i, 397-8, and Rev. des ehtdes anc, viii,
1906, p. 172.
2 B.G., vii, ^ gee
10, § 4 ; 11, §§ 1-4. p. 410.
* Rev. arch., noiiv. ser., viii, 1863, p. 393.

5 La Table de Peutinger, ed. Desjardins, p. 26, cols. 1-2.

 ;

VELLAUNODUNUM 495

but Desjardins believes that LIX ought to be reduced to LIV,

Creuly ^ mentions a third
—I'ancien chemin
' —
of 116 kilometres, or
'

about 72 miles, which passed through Montargis and Ladon. The

distance from Sens to Chateau-Landon is 44 kilometres, or about
27 miles ; from Sens to Sceaux 52 kilometres, or rather more than
32 miles ; from Sens to Beaune 66 kilometres, or about 41 miles
from Sens to Montargis 50 kilometres, or about 31 miles from Sens;

to Ladon 65 kilometres, or rather more than 40 miles.

Now there is no proof that any of these three roads existed in the
time of Caesar. Neither the road by Chateau-Landon nor that by
Montargis is marked in the Itinerary ofAntonine or in the Table and ;

it has been argued that the Gauls would not have made a road
through the Forest of Orleans and across the marsh of Sceaux.
D'Anville,^ it is true, says that the road by Chateau-Landon was
commonly called the Chemin de Cesar but evidence like this, as
;

— —
the reader if there is such a person of the countless French mono-
graphs on questions of Gallic geography knows to his cost, is brought
forward for almost every site that ingenious antiquaries have ever
proposed.^ The exact direction of the road mentioned in the Table
cannot, except in part, be ascertained."*
On the other hand, though it is not proved that the road which
runs by way of Sceaux, Chateau-Landon, and Beaune was a Gallic
road, there is nothing to show that it was not. Though not men-
tioned in the itineraries, it was certainly a Roman road and. ;

Napoleon notwithstanding, it is probable that Caesar followed a

route as direct or nearly as direct as this, because otherwise he
could not have accomplished the march from Sens to Orleans in four
days unless he had made forced marches. For the same reason,
although it is not proved that a Gallic road passed through Montargis,
the absence of proof is not a sufficient reason for denying that
Vellaunodunum lay on the ancien chemin '.
'

I now come to the various conjectures that have been made as to

the site of Vellaunodunum. 1. Napoleon,^ who identifies Cenabum
with Gien, pins his faith to Trigueres, on the road from Sens to
Gien-le-Vieux. His arguments appear to be an imperfect summary
of those of an antiquary, A. Petit,^ whom he does not mention. As
Cenabum (q.v.) stood upon the site of Orleans, not of Gien, it is, as
any reader who keeps his map open will see, useless to examine
these arguments. For Caesar would have had to march in two
days 56 miles, the distance from Trigueres, by the road which
Napoleon makes him take, to Orleans. Such a march would not,
indeed, have been absolutely impossible but, if Caesar had made
;

1 Rev. arch., nouv. ser. viii^ 1863, p. 392.

2 Eclaircissemens sur Vancienne
Gaule, p. 189.
3 According to J. B, Jollois {Mem. sur les ant. du dep' du Loiret, 1836,
pp. 16-20), that part of the road which lies between Sens and Nancre is called
Chemin de Cesar while its continuation, which crosses the Forest of Orleans,
;

is called Chemin Perre.

* Diet. arch, de la Gaule,
\, 68.
^ Hist, de Jules Cesar, ii,
251, note.
6 Dissertation sur Genahum-Oien,
&c., 1863, pp. 43-4.
496 VELLAUNODUNUM
two such extraordinary marches on two successive days, he would
certainly have said so.^ Therefore we must reject Trigueres.
2. D'Anville ^ proves that Lebeuf ^ is wrong in selecting for the site
the village of Vellan, near Auxerre but I need not summarize his
:

arguments for Lebeuf argued on the untenable hypothesis that

;

Cenabum was on the site of Gien.

3. D'Anville himself decides for Beaune. He argues that the
distance, both from Sens and from Orleans, answers to Caesar's
narrative that the place
; was well situated for the interception of
the supplies which Caesar expected that it was in the diocese of
;

Sens, and must therefore have been in the country of the Senones ;

that it was on the direct road from Sens to Orleans and fmally ;

that the name Beaune is etymologically connected with Vellauno-

' '

dunum. To prove this, he contends that Vellauna would have been

contracted into Velna and that Velna would have been changed
;

into Belna, as Vesuntio was changed into Bisuntio (Besangon).

'
Beaune,' he concludes, est nomme Belna des le neuvieme siecle.'
'

But I believe that no Gallic or even Gallo-Koman antiquities have

been found at Beaune * and there is really no positive evidence in
;•

its favour, save the very doubtful evidence of etymology, which

cannot be also cited in favour of Montargis. Beaune, moreover, is,
as General Creuly ^ points out, considerably nearer to Orleans than
to Sens, whereas Vellaunodunum was probably nearer to Sens than
to Orleans.
4. about a mile and a half east of Sceaux,
Jollois ^ decided for a site
which is on one of the Roman
roads between Sens and Orleans, and
has yielded Celtic remains and Roman coins to the excavator^ It
has been objected to on the ground that the site is low-lying, and was
therefore not suited for a dunum. But the word dunum means a '

town or strong place ',^ not necessarily a stronghold built upon

' '

a hill. Like Avaricum (Bourges), Sceaux stands upon marshy

ground and some of the Gallic oppida were no better situated for
;

defence.
5.Walckenaer^ is inclined to place Vellaunodunum at CVan-et-
Cheneviere, between Chatillon-sur-Loing and Chateau- Renard, where,
according to a MS. Memoir by Jollois, have been discovered the ruins
of an ancient town. But of course Walckenaer is only guessing ;

and Cran-et-Cheneviere, which is some 10 miles south-east of Mont-

argis, is not situated upon the road by which Caesar would have
marched to Cenabum.
6. Von Gdler ^^ decides for Ladon but Ladon is 14 kilometres
;

^ He makes
a point of telling us when he made forced marches. For a long
passages see H. Meusel, Lex. Caes.,n, 370-1.
list of
- Eclaircissemens, &c.,
pp. 221-4.
^ Becueil de divers ecrits pour servir d^ eclaircissemens a Vhist. de France, ii,

1738, pp. 179-207.

* A. Petit, Dissertation sur Genahum-Oien,
pp. 87-8.
^ Rev. arch., nouv. ser., viii, 1863, p. 393.

^ Mem. sur les ant. du dep^ du Loiret,

pp. 22-30.
^ See
' Rev. numismatique, xxii, 1852, pp. 313-6. pp. 463-4.
9 Geogr. des Gaules, i, 410. ^^ Gall. Krieg, 1880, p. 239.
 VELLAUNODUNUM 497

nearer to Orleans than to Sens, and must therefore, as General Creuly

says, be rejected.
7. Creuly 1 thinks that the choice lies between Chateau-Landon
and Montargis. Both, he remarks, are at a reasonable distance from
Agedincum (Sens) and from Cenabum (Orleans) but he considers ;

—
the site of Montargis the most defensible in a generally flat district
— better adapted for a Gallic oppidum. If Lebeuf ^ is to be believed,
Chateau-Landon can hardly be the place that we are looking for. II '

est constant,' he asserts, 'par les Actes de saint Severin, qu'encore de

son tems, c'est-a-dire sous le regne de Clovis I, Chateau-Landon
n'etoit qu'un bois et qu'une forest.' ^ To Montargis, in the environs
of which Roman antiquities have
been discovered,"* I know no serious
objection. been argued that Montargis is too far
It has, indeed,
from Orleans but I cannot see that the objection is valid. Mon-
;

targis is not more than 66 kilometres, or about 41 English miles,

from Orleans Caesar implies that it was his object to march as fast
:

as he conveniently could ^ and 20 miles, though a more than

;

ordinary, was not an impracticable day's march. Still, there is no

sufficient reason for definitely accepting Creuly' s choice.
8. has persuaded himself that he has solved the problem
StofEel ^
which has amused and baffled so many generations of inquirers. If
he had not a deservedly high reputation as a Caesarian scholar,
I should not notice his arguments for they are based upon the
;

untenable hypothesis that Cenabum stood upon the site of Gien :

but, as a specimen of the reasoning which satisfied Napoleon's ablest

collaborator, they have a certain interest. Vellaunodunum, he says,
was on the site of Toucy, which stands on the direct road from Sens
to Nevers. His main arguments are these. First, Caesar's words
prove that he marched by the direct road from Sens to Nevers :

'
Les mots ad Boios proficiscitur (B. G.j vii, 10) ont un sens trop
. . .

Eev. arch., viii, 1863, p. 393.

^

Recueil de divers ecrits pour servir d" eclair cissemens a VJiist. de France, ii,
2

208 ; Mem. de litt. tirez des registres de V Acad, des inscr., &c., ix, 1731, p. 373.
At the time of which Lebeuf speaks Chateau-Landon was called Castrum
Nantonis.
^ A fanciful etymological argument which has been offered in its favour

{Comptes rendus dela Soc. du Berry, 1863-4, p. 349) may safely be neglected.
. . .

'
Vellaunos,' says Sir John Rhys {Celtic Britain, 3rd ed., 1904, p. 289), probably '

meant a prince or one who reigned, and so Vellaunodunum would have meant
"the King's fort"'. According to A. Holder {Alt-celtischer Sprachschatz, iii^
149), it meant the fort of Vellaunos or perhaps ' the safe fort '.
'

* Jollois, op. cit.,

pp. 123-7.
^ Ea qui conficeret, C. Trebonium legatum [Vellaunoduni] relinquit, ipse ut

quam primum iter facer et, Cenabum Carnutum proficiscitur. B. G., vii, 11, §§ 3-4.
Meusel {Jahresb. d. philol. Vereins zu Berlin, xxxvi, 1910, p. 53) deletes ut . . .

faceret as an interpolation, remarking that it is absurd to say that the reason

for Caesar's marching to Cenabum was the intention tit quam primum iter
conficeret (v. 1. faceret). I cannot see where the absurdity comes in. Is there
anything absurd in this translation, —
He left Gains Fabius to give ellect to
'

these orders, and, being anxious to finish his march as soon as possible, pushed
on for Cenabum,' &c. ? In other words, he loft Trebonius at Vellaunodunum
instead of staying there himself, because he wished to reach Cenabum without
delay. ^ Guerre de Cesar et d' Arioviste, ])]). 146-9, 154-5.

1093 Kk
498 VELLAUNODUNUM
net pour qu'il ne soit pas evident que Cesar marcha directement vers
le pays des Boiens par la route d'Agedincum a Novioduiium
. . .

(Nevers).' Secondly, when Caesar says that, after leaving Age-

dincum, he arrived at Vellaunodunum altero die, he means that he
arrived there on the second day after his departure. The proof of
this is that the battle which Caesar fought with Vercingetorix
immediately before the siege of Alesia undoubtedly took place on
the river Vingeanne, at a distance of 72 kilometres from Alesia that ;

Caesar arrived at Alesia altero die and that, as he could not have
;

compassed that distance before nightfall on the day after the battle,
lie evidently did not reach Alesia until the second day after. Thirdly,
Toucy is 61 kilometres from Sens. Caesar, being pressed for time,
marched, on an average, 30 kilometres a day. That is to say, he
marched from Sens, along the direct road on which Toucy stands,
00 kilometres before he reached Vellaunodimum. Therefore Toucy
and Vellaunodunum are identical.
In the whole of this chain of reasoning there is not a single sound
link. First, ad Boios proficiscitur does not necessarily mean that
Caesar took the shortest road to the country of the Boi. Read in
connexion with the context, the words mean that he marched to
relieve the Boi but that does not imply that he did not intend to
:

punish Cenabum en route nor is it certain that there was a bridge

;

over the Loire nearer than Cenabum. Secondly, as I shall hereafter

prove,^ there is no evidence that the battle which preceded the siege
of Alesia was fought on the Vingeanne, while there is strong evidence
that it was not and, as I prove on pages 738-40, the meaning which
;

Stoffel, arguing from this questionable premiss, gives to altero die is

entirely wrong. Thirdly, even if that meaning were right, even if it
were certain that Caesar marched, on this particular occasion, at the
—
rate of 30 kilometres a day no more and no less Stofiel's conclu- —
sion would not be justified for he has no right to assume that Caesar
;

made two complete marches, no less and no more. Caesar might

have marched 20 kilometres on the day on Avhich he left Agedincum ;

30 on the second day and 20 on the third

;
.'^

The conclusion of the whole matter is that there is no decisive

evidence for any one of the sites which I have mentioned. Beaune,
Cran-et-Cheneviere, Ladon, Trigueres, and Toucy must be rejected ;

and so apparently must Chateau-Landon. Montargis and the site

near Sceaux remain. I believe that there is more to be said for the
former than for the latter. But unless some decisive inscription
—
should hereafter be discovered and the chance of such a discovery is
—
almost infinitesimal we shall never get beyond conjecture.^
1 .See pp. 791-801.
^ As makes Caesar reach Vellaunodunum on the second day after
Stoffel
hi« departure from Sens, and yet only accomphsh two marches, I presume that,
Hke Napoleon {Hist, de Jules Cesar, ii, 251, note), he holds that the day of
Caesar's departure was spent in concentrating the troops, whom Napoleon
assumes to have been echeloned in the environs of .Sens, and possibly in crossing
the Yonne. This hypothesis is inconsistent with the meaning of altero die.
The concentration, if it took place at all, must have taken place before.
^ M. Jullian {Hist, de la Oaide, iii,
435, n. 4), who agrees with me in provision-
 VELLAVII 499

—
Vellavii. The Vellavii were clients of the Arveriii.^ Their
territory corresponded, roughly, with the ancient Velay, or the
department of the Loire-Superieiire."
—
Venelli. The Venelli dwelt in the Cotentin (the department of
the Manche). The French Commission believe that they only occu-
pied a part of the diocese of Coutances, because it is cut in two by
marshes and they suggest that the Ambibarii (q.v.), who perhaps
;

occupied the southern part and possibly also the adjacent part of
the diocese of Avranches, may have been originally their clients.^
Walckenaer^ believes that the diocese of Avranches, which was
formed out of the territory of the Abrincatui (probably, he thinks,
the same people as the Ambibarii), belonged to the Venelli ; and
M. Longnon takes the same view.^
—
Veneti. The Veneti, according to the common opinion,^ occupied
the diocese of Vannes, or, roughly speaking, the department of the
Morbihan. M. Longnon,'^ however, also gives them the alleged
civitas Coriosopotum of the Notitia provinciarum,^ which, if it ever
existed, was perhaps in the territory of the Osismi ^ and Desjardins^^
;

gives them, besides the Morbihan, the arrondissement of St. Nazaire,

that is to say, the peninsula of Guerande.
Ptolemy ^1 places the Veneti north of the Samnitae (q.v.) and south
of the Osismi, whose southern boundary, he says, was marked by the
Gobaean promontory, that is to say, the Pointe du Raz. On pages
469-70 I have shown reason for believing that the Samnitae never '
'

existed. Strabo^"^ says that the Loire enters the sea between the
Pictones and the Namnetes (q.v.), who were the southern neighbours
of the Veneti. Pomponius Mela ^^ places the Osismi opposite the
island of Sena, which is generally identified with Sein.^'*
The northern frontier of the Veneti cannot be traced with cer-
tainty. Unless Ptolemy made a mistake, it could not have coincided
with the northern frontier of the diocese of Vannes, which struck
the coast far south of the Pointe du Raz ^^ and all that can be safely
;

said is that Venetia did not extend further northward than that
promontory and the natural boundary formed by the Montagues
Noires. See Osismi.
The arguments that have been advanced to prove that the Veneti
occupied the peninsula of Guerande are refuted on pages 680-3.
—
Veragri. See Nantuates.
—
Viromandui. The Viromandui, whose name was preserved in the
old name of Vermandois, occupied the diocese of Noyon, or the

ally accepting Montargis, records one or two other guesses, which are not worth
discussing. ^ B. G., vii,
75, § 2.
^ D'Anville, Notice de Vancienne Gaule, p. 685 ; Walckenaer, Geogr. des
Gaules, i, 344.
^ Rev. arch., nouv. ser., ix, 1864,
pp. 404-6.
* Geogr. des Gaules, i, 385-7. ^ Atlas hist, de la France, p. 7.

« Walckenaer, Geogr. des Gaules, i, 378. ' Atlas hist, de la France, p. 7.

» tSee note on the Coriosolites. ^ See
p. 421.
1" Geogr. de la Gaule rom., ii, 485. " Geogr., ii, 8, §§ 5-6.
12 1^
iv, 2, § 1 ; 4, § 0. iii, 0, § 48.
1* See p. 466, n. 3. ^^ Gallia Christiana, xiv, map facing p. 1.

K k2
 500 VIROMANDUI
northern part of the department of the Aisne and the eastern part
of that of the Somme. See Walckenaer, Geogr. des Gaules, i, 431-2,
and also Remi and Suessiones.
—
Vocates. The Vocates are generally believed to have occupied
the diocese of Bazas, or the south-eastern part of the department of
the Gironde.i Caesar says that Crassus, after he had defeated the
Sotiates, marched for the country of the Vocates and Tarusates.^
The Sotiates occupied the country round Sos,and the Tarusates (q.v.)
dwelt either in the district of Aire or in that of Tartas. Ptolemy^
mentions a people called the Vassarii {OvaadpLoi), .whose chief town,
Cossio, stood upon the site of Bazas. The people of Bazas are called
in the Notitia * Vasates and perhaps Ptolemy really wrote Ot-a-
;

(TOLTLOL. Desjardins^ identifies the Vocates with the Vasates and the
Vassarii. Pliny ^ mentions the Basabocates, a name which, as Meusel
holds,"^ is merely an amalgamation of the forms Vasates and Vocates.
The diocese of Bazas is divided by the Garonne, on the north of
which Walckenaer places the Vocates and on the south the Vasates.
This conjecture is certainly wrong for how can any one believe that
;

Crassus, who had invaded Aquitania from the north, would, after he
—
had defeated the Sotiates the first tribe that he encountered have —
marched back from the neighbourhood of Sos to the north of the
Garonne, in order to attack another people, whose allies came from
the Pyrenees ? Besides, on Walckenaer's theory, the Vocates were
S3parated by a considerable tract from the Tarusates whereas it is
;

clear from Caesar's narrative that the two peoples were conterminous.
The problem of determining the habitat of the Vocates is further
complicated by the fact that Pliny mentions, immediately after the
Basabocates, another people called the Vassei. I am inclined to
believe that the Vocates and the Vasates were identical but if so, ;

it will be evident to any one who looks at the map that if Crassus
invaded their country as well as that of the Tarusates, it must have
extended south of Bazas into the department of Les Landes.
[Since the foregoing paragraph was written, the current theory, ex-
pounded by MM. Blade and Jullian, has been that the Basabocates
were really the Basates (or Vasates) and the Boiates, who are assumed
to have been identical with the Vocates that the former occupied
;

the country round Bazas, and the latter the district of Buch near
Arcachon.^ The question is obscure but I adhere to my former
;

opinion because Caesar's narrative is irreconcilable with the view

that Crassus invaded a country so far from Sos and from the Pyrenees
(whence the allies of the Vocates and Tarusates came) as Buch.
M. Jullian, indeed, as I learn from his latest volume,^ now virtually
agrees with me for he holds that the Basates were probably
;

included by Caesar with the Vocates and he also observes, as I have

;

^ B. G., iii,
^ Walckenaer, Geogr. des Gaules, i, 302. 23, § 1.
^ Geogr., ii, 7, § 11. See note to vol. i, 205, of C. Miiller's edition.
p.
^ Ed. 0. 8eock, p. 271 (xiv, 10). ^ Geogr. de la Gaule rom., ii, 3G2-3.

« Nat. Hist., iv, 19 (33), § 108. ^ MS. note sent to me by Meusel.

^ Annates de la Facidte des lettres de Bordeaux, 1893, p. 118; C. Jullian,

Hist, de la Gaule, ii, 450, n. 9. ^ lb., iii, 307, n. 3.
 VOCATES 501

done, that, according to Caesar, the Tarusates (whom he places near

Tartas) and the Vocates were conterminous. At the same time he
justly remarks that Caesar could not have simultaneously marched
towards Bazas and Tartas and, as I have done in my narrative,^
;

he prefers the latter route. The Vocates had probably marched

southward to reinforce the Tarusates and Caesar may have meant ;

that Crassus marched into the territories of the Vocates and Tarusates
considered as a whole.]
—
Vocontii. Caesar mentions the Vocontii once only.^ He says
that, after crossing the Alps —
doubtless by Mont Genevre^ he —
entered their territory, and passed thence into that of the Allobroges.
Strabo * also describes the Vocontii as conterminous with the Allo-
broges and he implies that Ebrodunum, or Embrun, marked
;

their eastern boundary. According to Ptolemy,^ Ebrodunum be-

longed to the Caturiges (q.v.) and Long suggests that Strabo pro-
;

bably meant that it was just on the borders of the Vocontii '.^
'

Ptolemy says that the chief town of the Vocontii was Vasio, which
occupied the site of Vaison Pliny gives them another town, Lucus
:
'^

Augusti, which is identified with Luc and from the Tahle we learn
;

that they had a third, Dea, the modern Die. From these data it has
been inferred that their territory comprised the dioceses of Vaison
and Die. D'Anville ^ believes that they also occupied a part of the
diocese of Gap, dans lequel on ne connoit point d'aucun peuple en
'

particulier,' as well as Val-Benois, that part of the diocese of Sisteron

which borders on Vaison and, if Strabo is to be believed, they
;

probably also possessed territory extending as far eastward as the

river Durance. D'Anville, who traces their eastern frontier far to
the west of the Durance, ignores or overlooks the testimony of
Strabo. But Chorges or Caturigomagus, which belonged to the
Caturiges, is nearly due west of, and about 12 miles from Embrun.
If, then, Chorges belonged to the Caturiges in the time of Strabo, he
was wrong in making Embrun the eastern limit of the Vocontii and ;

their territory could not have extended to the Durance, except

perhaps below Tallard.^
Maissiat,io remarking that, according to Strabo, the Vocontii were
conterminous with the Allobroges and above the Cavares, infers that
their territory extended northward as far as the Isere, and that they
possessed the valley of the Graisivaudan as far eastward as Mont-
melian. But it does not follow that because the Vocontii were con-
terminous with the Allobroges, they were conterminous along their
whole northern frontier on Maissiat's theory the territory of the
:

Vocontii must have included the territories of the Tricorii and

the Uceni, and part of that of the Medulli as well and these small ;

tribes may just as likely have been cHents of the Allobroges.

^ See p. 93. 2 b. O., {, 10, § 5. ^ g^e pp. 430-1.

* iv, 1, § 3. 5 Geogr., iii, 1, § 35.
" W. Smith's Did. of Greek and Roman Geogr., i, 798.
' Nat. Hist., iii, 4 (5), § 37. » Notice de Vancienne Gaule,
p. 715.
" Cf. C. JulHan, Hist, de la Gaule, ii, 517, n. 2.
^^ Jules Cesar en Gaule, i, 1865, pp. 46-7, 334.
502 VOCONTII
The task of determining the frontiers of the Vocontii is indeed
rendered peculiarly difficult by the fact that the territory which un-
doubtedly belonged to them was surrounded by the territories of
minor peoples, who may possibly have been fagi of theirs, actually
incorporated within their proper territory, or may only have been
connected with them by the loose tie of clientship while some of ;

these peoples may have been j)agi or clients of the Caturiges or of

the Allobroges. The minor peoples to whom I refer were the Memini,
the Vulgientes, the Quariates, the Bodiontici, the Avantici, the
Tricorii, and the Uceni. Desjardins^ remarks that their insignificance
must have caused them to be absorbed in the dientela of the Vocontii,
although, if I understand him aright, he distinguishes their terri-
tories from those of the stronger people. But it seems more probable
that the Uceni and the Tricorii, whose territories the valleys of the —
—
Romanche and the Drac respectively formed part of the diocese of
Grenoble, were clients of the Allobroges (q.v.) and Desjardins him- ;

self elsewhere ^ says, rightly in my opinion, that the Memini must

have been clients [if not a pagus] of the Cavares. For the district of
Carpentras (Carpentoracte), which belonged to the Memini,^ separates
that of Cavaillon (Cahellio) from that of Orange {Arausio), both of
which formed part of the territory which Ptolemy'* assigns to the
Cavares. The exact position of the Quariates, who were on the east
of the Vocontii, is uncertain ^ but there is no doubt that the Bodion-
:

tici dwelt in the neighbourhood of Digne w^hile, as Desjardins ;

remarks, there is a striking resemblance between Avantici and '

'

'
Avance ', Avangon ', and St. Etienne d'Avanyon ', the names of
' '
—
a river and of two communes which are to be found on the west
of Chorges. It seems to me probable that these three peoples, as well
as the Vulgientes, who dwelt in the neighbourhood of Apt, were
clients of the Vocontii.^ Desjardins thinks that the eastern boundary
of the Vocontii would naturally have been formed by the Devoluv,
the Montague d'Aurouze, and the chain popularly known as Monts '

de France '.
—
Volcae. The Volcae (Arecomici and Tectosages) occupied the
country comprised between the Rhone, the Cevennes, and the
Garonne and the territory of the Tectosages included that of
;

the Tolosates (q.v.). According to Strabo,*^ Narbo belonged to the

1 Geogr. de la Gaule row., ii, 232. ^ 76.. pp. 226-7.
^ Ptolemy, Geogr., ii, 10, § 8 (C. Muller's ed.. p. 246).
* Ih. (Miiller, pp. 243-4).
^Desjardins (ii, 228) is inclined to place them near Forcalquier.
^ D. Long, who has devoted a monograph [Mem. presentes par divers
J.
savants a V Acad, des inscr., 2^ ser., ii, 1849, pp. 284-313) to the Vocontii,
holds that their territory comprised the dioceses of Die, Vai.son, and Sisteron.
part of the diocese of Gap, and part of the diocese of Grenoble (on the left bank
of the Isere). But, seeing that Cularo (Grenoble) was one of the chief towns
of the Allobroges, why should part of the diocese be assigned to the Vocontii ?
The passage in one of Cicero's letters Isara, flumine maximo, quod in finibus
est Allobrogum {Fam., x, 15, § 3) —
on which Long (p. 296) relies, only proves that
the Isere was in the territory of the Allobroges ior fnibus means territory',
;
'

not, as Long thinks, '

frontier'. Cf, p. 364, n. 3, svpra.
' iv, 1, § 12.
 —
VOLCAE 503

Arecomici according to Ptolemy,^ to the Tectosages. D'Anville ^

;

thinks that so long as Narbo was the capital of the Province, it

probably belonged, as Strabo says, to the Arecomici, and afterwards
to neither of the two peoples. A passage in Caesar praesidia in
Rutenis provincialibus, Voids Arecomicis, Tolosatibus, circumque
Narhonem constituit ^
. . . —
seems, as C. Miiller remarks in his edition
of Ptolemy (p. 241), to imply that Narbo was not, at all events in
52 B. c, in the country of the Arecomici. Walckenaer,^ who accepts
d'Anville's theory, assigns to the Arecomici the dioceses of Agde,
Lodeve, Montpellier, d'Uzes, Nimes, and Alais, a territory corre- —
sponding roughly with the departments of the Gard and the Herault;
while the country of the Tectosages corresponded with the arch-
bishopric of Toulouse and all that part of the diocese of Montauban
which lies on the east of the Garonne. Desjardins ^ conjectures, in
spite of Strabo, that the river Herault, or else a line passing between
the Herault and the Orb, separated the two peoples.

^ Qeogr., ii, 10, § G, ^ Notice de Vancienne Gaule, p. 717.

=»
B. G., vii, 7, § 4. * Geogr. des Gaules, i, 253.
^ Geogr. de la Gaule rom., ii, 215.
 SECTION JV.— SOCIAL, POLITICAL, AND
RELIGIOUS
MONARCHY IN GAUL
FusTEL DE CouLANGEsi Gaul only attained
asserts that kings in
power by the will of the majority words which
; and he refers to the
Caesar puts into the mouth of Ambiorix, king of the Eburones
sua esse eiusmodi imperia ut non minus haheret iuris in se multitudo
quam ifse in multitudinem,'^ (' My authority is so far limited that the
multitude has no less power over me than
I over them ') to show —
that their power was extremely restricted. He holds that monarchy
in Graul was not a traditional institution, but rather a revolutionary
power, created by a faction, Avhich arose in troublous times.^
Let us examine these conclusions. Caesar mentions in all eleven
Gallic kings, namely Catamantaloedis, king of the Sequani Divicia- ;

cus, king of the Suessiones Galba, king of the Suessiones Commius,

; ;

king of the Atrebates Tasgetius, king of the Carnutes

; Moritasgus, ;

king of the Senones Cavarinus, king of the Senones

; Ambiorix and ;

Catuvolcus, each king of one half of the Eburones Vercingetorix, ;

king of the Arverni and Teutomatus, king of the Nitiobroges.^

;

Three of these kings, Diviciacus, Galba, and Commius were Belgic.

Diviciacus and Catamantaloedis had ceased to reign before Caesar
arrived in Gaul. Commius, Tasgetius, and Cavarinus were nominees
of Caesar the ancestors of Tasgetius and Cavarinus had reigned
:

before them, and Moritasgus, the brother of Cavarinus, was reigning

when Caesar first entered Gaul Tasgetius was murdered by enemies
;

belonging to his own country, and Cavarinus incurred bitter odium

during his reign. Orgetorix, Casticus, Dumnorix, and Celtillus, the
father of Vercingetorix, aimed at seizing the royal power Celtillus ;

was put to death by his countrymen for having done so and an ;

attempt was made to execute Orgetorix for the same ofience.^

These facts prove that among those peoples which had discarded it,
any attempt to revive the royal power was regarded by the repubhcan
party with extreme jealousy ^ and as among the Aedui, if not among
;

^ Hist, des inst. pol. de Vane. France, —La Gaule rom., 1891, pp. 12-3.
2 B. G., V, 27, § 3.
^ He refers to B. G., vi, 8, § 9, Cingetorigi, quern db initio permansisse in
officio demonstravimus, principatus atque imperium est traditum. But we are not
told that Cingetorix was a king at all.
^ B. G., i, 54, § 2
3, § 5 ii, 4, § 7 ; iv, 21, § 7
; v, 25, § 1 ; vi, 31, § 5
; vii, ; ;

4, §5; 31, §5.

'o
lb., V, 25, § 3 vi, 5, § 2 ; i, 2, § 1
; 3, § 5 4
; 9, § 3 vii, 4, § 1.
; ; ;

« Cf. Sir J. Rhys' s Celtic Britain, 3rd ed., 1904, p. 58.

 — — ;

MONARCHY IN GAUL 505

other peoples, a supreme executive magistrate called the Vergobret

was ammally elected, and Caesar frequently mentions the senates of
Gallic peoples in such a way as to imply that there were no kings to
limit their power,i it has been concluded that, in his time, a monar-
chical form of government was the exception, and a republican the rule.
The conclusion to which this analysis leads is that monarchy
had once prevailed throughout the whole of Gaul, but that by some
revolution or series of revolutions, like those of which we read in the
histories of Greece and Rome, it had, in the majority of states,
perished. Among the Belgae, however, it should seem that the
dislike of monarchy was less active and the proportion of kings
greater than in Celtican Gaul. Powerful nobles, Caesar tells us,^ still
frequently seized royal power, or rather, as Sir John Rhys puts it,
made themselves despots, just as Pisistratus did at Athens but ;

they attained this power not by the will of the majority but by the
help of armed clients and mercenaries, and by ingratiating themselves
with the masses.

DID TWO VERGOBRETS HOLD OFFICE IN ONE

STATE AT THE SAME TIME ?
It is events among the Aedui, only one
generally held that, at all
Vergobret could legally hold at a time. This opinion is based
office
upon the well-known passage ^ summo esse in periculo rem, quod,
:

cum singuli magistratus antiquitus creari atque regiam fotestatem annum

ohtinere consuessent, duo magistratum gerant et se uterque eorum legibus
creatum dicat. On the other hand, Caesar writes convocatis eorum :

principihis, quorum magnam copiam in castris hahehat, in his Diviciaco

et Lisco, qui summo magistratui praeeraln^t, quem vergohretum appellant

Haedui,qui creatur annuus et vitae necisque in suos hahet potestatem,^ &c.

Praeerant is found in all the MSS. but since Lipsius published his
;

edition, the editors, following Nicasius, have almost unanimously

substituted for it praeerat, on the ground that the other passage proves
that there can only have been one Vergobret.^ But M. Robert Mowat
has disputed this opinion ^ and Sir John Rhys agrees with him."^
;

There is a coin of the Lexovii, bearing the inscription cisiambos •

CATTOS VERGOBRETO. The words Cisiambos Cattos, M. Mowat argues,

•

must be the names not of one man, but of two (1) because the word ;

which follows them would otherwise be vergobretos, not vergobreto

and (2) because by Gallic custom no man had more than one name,
sometimes followed by a patronymic. Therefore to Cisiambos and

1 B.G.,i,lQ,^5; 31, §6; ii, 5, § 1 ; 28, § 2 ; iii, 16, § 4 ; 17, § 3 ; v, 54, § 3 ;

vii, 32, § 5 ; 33, § 2 ; 55, § 4.

2 Ih., ii, 1, § 4. in Gallia a potentioribus atque iis qui ad conducendos homines
facultates habebant vulgo regna occupabantur. Cf. i, 4, § 2 ; 17, §§ 1-2 ;

18, §§ 3-6. 3 7^,^ vii, 32, § 3.

* lb., i, 16, § 5. 5 ^QQ Schneider's Caesar, i, 35.
6 Eev. celt., v, 1881-3, pp. 121-4. ' 3rd ed., 1904, p. 59.
Celtic Britain,
506 THE OFFICE OF VERGOBRET
Cattos belongs in common the title vergohreto, which must be either
plural or, more probably, dual.
M. Mowat's argument is controverted by M. P. Ch. Robert. He
does not deny that Cisiambos and Cattos were two persons but, :

remarking that s is often dropped in Gallic, he explains vergobreto

as an abridged form of the nominative singular vergobretos ^ and ;

M. E. Ernault, who agrees with him, says, in opposition to M. Mowat,

that the dual would be not vergobreto but vergobretu.^
Sir John Rhys admits the difficulty presented by the former of the
two passages which I have quoted but he struggles to get over it
;

by suggesting that the two offices were not filled at the same time
'

of the year '. I do not believe that anybody, reading the passage
with an unbiased mind, could agree with him. How can the
(assumed) two offices have been filled at different times of the year
when Gaesar distinctly says that the Vergobret held office for a whole
year (singuli magistratus . regiam folestatem annum obtinere con-
. .

suessent) ? And if Caesar really wrote (summo magi'stratui) praeerant^

as Sir John Rhys maintains, is it not obvious that the (assumed) two
Vergobrets held office simultaneously ? Also, is it not extraordinary
that Sir John should have failed to see that if two Vergobrets
were elected every year, of whom one held office during one part of
the year and the other during the remaining part, the magistracy
was not dual but single ? ^ Besides, on Sir John's theory, how is one to
explain the other passage convocatis eorum 'principibus .
:
in his . .

Diviciaco et Lisco, qui sumyno magistratui praeera[n]t, quern vergo-

b return appellant Ilaedui, qui creatur annuus, &c.? What can be
the antecedent to quern, if it is not the preceding clause qui . . .

praeera[n]t ? And if praeerant is the right reading, what becomes of

the grammar of the sentence ? Surely Caesar would have written
quos vergobretos (appellant Haedui).
[Since I wrote the foregoing note, I have seen in the Dictionnaire
archeologique de la Gaule (1©^ fascicule, Planches desmonnaies, No. 78)
an illustration of the coin in question.^ It bears on the reverse side
this inscription :

\/x \^
CO

4^
Speaking of the Vergobret, Cotus, the writer of the article ,
Aedui
says,^ '
comme au moment de sa magistrature la cite des Eduens
exercait une suprematie de fait sur la Celtique entiere, nous trouvons
1 Comptes rendus de V Acad, des inscr., xiii, 1885, pp. 283-4. - Ih.

^ See M. d'A. de Jubainville's remarks in Bev. celt., viii, 1887, p. 222, n. 8.

* No. 7103 in the Cat. des monn. (jaul. de la Bibl. Nat., by E. Muret and
^ Did. arch, de la Gaule, i. 15.
M. A. Chabouillet.
 — —:

THE OFFICE OF VERGOBRET 507

de beaux semis frappes chez les Lixoviates, par le chef Cisiambos,

avec la mention purement honorifique dii vergobret Cottus' (sic).
To the same effect writes the imaginative numismatist, Caignart de
Saulcy,^ who is surely wrong in identifying Cattos with the Cotus
whom Caesar displaced in favour of Convictolitavis.^
M. Robert offers a different explanation.^ The Gauls, he remarks,
formed their coinage upon Greek models. Now many Greek coins
bear the name both of a magistrate and of a monetary agent. Pos-
sibly, then, Cisiambos may have been an official of this kind.]

CLANSHIP, SENATES, AND LAW IN GAUL

J. G. Bulliot * holds that the so-called states (civitates) of Gaul
' '

—
the Aedui, the Arverni, and the rest were merely aggregates of
clans ; for, he maintains, when Caesar uses the word familia or
clientela, he means clan '. The pagus was the territory of the clan.
'

Each clan was ruled by its own chief and all these chiefs were
;

subject to the tribal chief, whether he was an elected magistrate, like

the Vergobret, or an hereditary ruler. This tribal chief, however,
was himself virtually subject to the control of the senate, which was
not a definite body, but comprised all the free landowners, in fact —
all the influential men in the community. In a body politic of this
kind the clan was everything, the state little or nothing ' (le clan
'

etait tout, la cite rien ou pen de chose). Written law did not exist
everything was regulated by custom more fatrio, more maiorum —
expressions which se retrouvent a chaque ligne [!] dans Cesar, dans
'

Strabon, dans Diodore, dans Tacite.' The word lex, when used of
Gaul, means custom ', as we may gather from Caesar's statement
'

that Germani multum ah hac consuetudine dijferunt ; from his remark

that an armed gathering more Gallorum initium est belli, quo lege
communi omnes puheres armati convenire consuerunt and from his ;

saying that Convictolitavis, who, as well as his rival Cotus, claimed

to be legally Vergobret, was afterwards duly elected by the priests,
in accordance with Aeduan custom.^
Fustel de Coulanges,^ on the other hand, remarks that neither
Caesar, nor Strabo, nor Diodorus Siculus makes any mention of clans
in Gaul.'*' There is only one passage in Caesar in which clientela can
—
possibly mean clan \^ the passage in which he describes, on the
'

^ Ann. de num., 1867, p. 11. The article Aedui in the DicHonnaire

la Soc.
was inspired, not written, by de Saulcy.
if
2 B. G., vii, 32-3. See A. Blanchet, Traite des monn. gauL, p. 83.
^ Comptes rendus de V Acad, des inscr., xiii,
1885, pp. 283-4.
* La cite gauloise, 1879,
pp. 51-2. 62, 193-4, 198, 204-6, 209-2.
= B.G., vi, 21, V, 56, § 2 vii, 33, § 3
§ 1 34, § 3.
;

^ Hist, des inst. pol. de Vane. France,

;

— ;

la Gaule rom., pp. 8-9.

^
I have already (p. 21, n. 1) quoted the passage in which 8ir H. Maine
records his conviction that Caesar failed to notice the existence of the natural '

divisions of the Celtic tribesmen, the families and septs or sub-tribes'.

^ See Meusel's Lex. Caes,, i, 562-3.
508 CLANSHIP, SENATES, AND LAW
authority of the Aediian envoys, the struggle between Cotus and
Convictolitavis (civitate^n esse o^nnem in armis ; divisum senatum,
divisum populu7n, suas cuiusque eorum clientelas)}- Here clientelas
obviously means groups of clientes
'
and there is nothing to show
'
;

that a Gallic chief might not have clients who were not his clansmen.^
Similarly there is only one passage in which it could with any plausi-
bility be argued thsitfamilia means —
clan ', the passage in which
'

Caesar says that Orgetorix assembled his familia and debtors to

overawe his judges (Orgetorix ad indicium oynnem suamfamiliam ad
hominum milia decern undique coegit, et omnes clientes ohaeratosque
suos . eodem conduxit)? It is curious that M. Bulliot should not
. .

have noticed that familia, in this passage, is distinguished from

clientes, and that if, as he maintains, clientes were clansmen, familia
was necessarily something different. If Caesar used the latter word
in its usual sense, Orgetorix' s/ami7m comprised his slaves.
Nevertheless, M. d'Arbois de Jubainville holds that the clan
system did prevail in Gaul.* After pointing out that the number
of senators in each state was large,^ and that, among the Aedui at
all events, two members of one family might not sit in the senate
simultaneously,^ he remarks that II semble que ces deux faits nous
'

mettent en presence d'une organisation analogue de Eome primitive.

En Gaule la race dominante dans chaque civitas se compose d'un
certain nombre de gentes representee chacune au senat par son chef '.
And again, Chaque peuple gaulois
'
etait forme de gentes ou clans,
. . .

au-dessus de chacun desquels s'elevait une famille plus noble et

plus riche que les autres, c'est-a-dire un petit groupe aristocratique
qu'entouraient des clients de race moins distinguee.' Does not
all depend upon what is to be understood by the word clan ? The ' '

primitive social, political, and military unit in Gaul was the 'pagus,
which M. Jullian calls the tribu '. Pagi may, as he remarks,''' have
'

been originally groups of families, descended, or traditionally sup-

posed to have been descended from a common ancestor for the ;

Helvetian pagus Verhigenus bore a real or mythical ancestral name :

but in Caesar's time the pagus was simply a community, which had
its own standards in war and common worship.^
When M. Bulliot says that the state was little or nothing ', he
'

exaggerates strife existed between the pagus and the civitas, and
:

pagi, as we have seen,^ occasionally acted independently in their

relations with Caesar but, however imperfectly welded the pagi
;

may have been, their aggregate, the state ', was the unit with
'

which Caesar had to reckon. As for the senates, such a passage as

that in which Caesar tells us that the Nervian elders informed him

^ B. G., vii, 32, § 5. ^ gee

pp. 514-6.
=*
B. G., i, 4, § 2. See Meusel's Lex. Caes., i, 1280-2.
* Becherches s^ir Vorigine de la propriete fonciere, &c., 1890,
pp. 51-2. Cette
'

theorie,' says M. A. Blanchet {Traite des monn. gaul., p. 225, note), Concorde
'

parfaitement avec I'explication que je propose pour le monnayage infiniment

varie de la Gaule.' ^ b. G., ii, 28, § 1.

« Ih., vii, 33, ' Hist, de la Gaule, ii, 14-6.

§ 3.
^ B. G., i, 12,
§§ 2-7 ; vii, G4, § G
; 88, § 4. ^ 8ee
p. 21.
 CLANSHIP, SENATES, AND LAW 509

that their senators had been reduced in number from 600 to 3,i would
seem to point to the conclusion that they were definite bodies and ;

Fustel, while admitting that we do not know how they were com-
posed, conjectures that they comprised, in each state, all the members
of the class which Caesar called nobiles? Probably, as M. Jullian
suggests,^ they were the chiefs of villages and the magistrates of jjagi.
M. Bulliot's view of the meaning of lex, when used by Caesar in
speaking of the Gauls, is certainly not established by the passages
which he quotes but there is not enough evidence to determine
;

the question, though, as the Druids were unwilling to allow their '

doctrine to become common property ', and therefore would not

commit it to writing, I believe that Gallic law was not codified.'* It
is, however, worth while to point out that of the two expressions,

more fatrio and inore maiorum, which, M. Bulliot says, are found
'
a chaque ligne dans Cesar ', the former is never used by Caesar at all,
while the latter is used once only, and then with reference not to the
Gauls, but to the Romans.^

DID THE GAULS, IN CAESAR'S TIME, RECOGNIZE

PRIVATE PROPERTY IN LAND ?
M. d'Arbois de Jubainville ^ holds that, in Caesar's time, private
property in land did not exist in Gaul. Each community, he says,
had over the whole of its land a right analogous to that which the
Roman people had over their ager publicus. He remarks (1) that,
according to Polybius,''' the Cisalpine Gauls did not cultivate the
soil, but that their property consisted in cattle and gold, which
could be easily moved from place to place (2) that the Helvetii ;

never would have consented to emigrate if they had possessed private

property in land (3) that the tribute (stifendium) which the Aedui
;

exacted from the Boi whom they invited to settle in their country ^
was simply rent due for a share of the ager publicus (4) that when ;
'
Caesar said that the Druids decided disputes regarding inheritance '

{de hereditate),^ he used the word hereditas in the sense of I'heritage '

de la royaute ', and that the author of Bellum Alexandrinum (^Q, § 6)

speaks of hereditas regni (5) that when Caesar said that the Druids
;

'
decided boundary disputes (de finibus), he meant by boundaries '

1B. G., ii, 28, § 1.

2Hist, des inst. pol. de Vane. France, &c., pp. 13-4.
^ Hist, de la Gaule, ii, 49-50.

* Cf. C. JuHian, Hist, de la Gaule, ii, 379, n. 1, and see J. Bryce, Studies in

Hist, and Jurisprudence, i, 1901, p. 89.

5 B. G., vi, 44, § 2. See Meusel's Lex. Caes., ii, 641-2. De Coulanges {Hist,
des inst. pol. de Vane. France, &c., p. 15 and n. 2) is disposed to infer from
B. G., vi, 20, that the GalHc laws were written.
® Comptes rendus de V Acad, des inscr., 4® ser., xv, 1887,
pp. 66-9, 74, 79-
83 RecTierches sur Vorigine de la propriete fonciere, &c., 1890, pp. xxiii-xxxi, 61.
;

' ii, 17.

« B. G., i, 28, § 5 ; vii, 10, § 1. » lb., vi, 13, § 5.
510 PRIVATE PROPERTY IN LAND
the frontiers of states and (6) that, according to Caesar, it was the
;

custom for a Gaul, when he married, to add to his wife's dowry an

equivalent from his own personal estate, and to administer the whole
as a joint piece of property, which, with its accumulated produce
(frudus), went to the survivor.^ M. d'Arbois's reason for denying
that the dowry or its equivalent could have been land will be given
presently.
Fustel de Ooulanges and M. Ch. Lecrivain ^ have no difficulty in
'^

disposing of these arguments. To the first Fustel replies that Poly-

bius was only speaking of the Gauls who had invaded Italy several
centuries before the time of Caesar, and were in a nomadic state. In
reply to the argument based upon the emigration of the Helvetii, he
points out that they emigrated simply because they preferred the
fertile plains of the Charente to their own wild and mountainous
country .4 Have peasant proprietors, he asks, never been known to
emigrate in order to seek more productive property elsewhere ? To
M. d'Arbois's third argument he replies that the Aedui might well
have had sufficient unoccupied public land to accommodate the small
Boian community, and yet have been themselves landed proprietors.
And, asks M. Lecrivain, why should we infer from the fact that the
Aedui assigned a portion of their territory to the Boi, that the whole
of that territory was ager publicus ? Is it not more likely that with
the Aedui, as with the Italians, private and public property in land
existed side by side ? In answer to the fourth argument, Fustel
remarks that, if the author of Bellum Alexandrinum uses the phrase
hereditas regni, he only does so because the word hereditas, by itself,
could not be taken as meaning I'heritage de la royaute '. Besides,
'

he points out, M. d'Arbois has failed to notice that, while Caesar

speaks again and again of sons who desired to succeed to the king-
doms of their fathers, their claims were never referred to Druids.
Again, says Fustel, Caesar, in his digression on the manners and
customs of the Gauls and Germans, twice uses fines in the obvious
sense of boundaries of landed estates ^ and whenever he uses the ;

word in the sense of frontiers, that meaning is unmistakably defined

by the name of the state in question or by a phrase of equivalent
meaning. Replying to the sixth argument, Fustel remarks that
M. d'Arbois tries to prove that the joint property of husband and
wiie, to which Caesar refers, could not have consisted of land, because
the produce of the land must have been stored in granaries and kept
there till husband or wife died, or, if it had been sold, the proceeds
would simply have been hoarded in a strong box,^ and the joint owners
would have perished of hunger The joint property, according to
!

M. d'Arbois, must have consisted of cattle and the jfructus of which ;

Caesar speaks Avere simply le croit des troupeaux ', the cattle
'
—
1 B. G., vi, 19, §§ 1, 2. - Question.^ hist., ed. 1893, pp. 104-1'2.
^ Annates de la Faculte des lellres de Bordeaux, 1889, pp. 182-94, csp. 18-4.
* B. G., i, 5 ; 10, § 1.
3, §
5 lb., vi,
13, § 5 ; 22, § 2. Fustel iiiiglit have added vi, 22, § 3.
" Comptes rendus de V Aaid. des inscr., 4^ ser., xv, 1887, p. 74. Cf. Rev. celt.,
xxii, 1896, p. 329.
 —
PRIVATE PROPERTY IN LAND 511

which were bred from the original stock. But, says Fustel, M. d'Ar-
bois forgets that fructus means profits made out of property of
every kind, cultivated land, cattle, money lent at interest, and
what not. The dowry and its equivalent may have consisted, wholly
or in part, of cattle but if so, the fact is no evidence that the Gauls
;

did not recognize private property in land.

Independently of the arguments which I have summarized, Fustel
proves conclusively that private property in land was recognized in
Gaul. It is significant, he remarks, that Caesar, while professing to
tell us in what respects the customs of the Germans differed from
those of the Gauls, says expressly that the former did not recognize
private property in land,^ and implies, in the passage Si de finibus
controversia est, that the latter did. Again, the existence in Gaul of
the institution of clientship, and the oft- quoted passage, in omni
Gallia eorum hominum qui aliquo sunt numero atque honore genera
sunt duo ; nam flehes paene servorum habetur loco^ (' Everywhere in
Gaul two classes only are of any account or enjoy any distinction ;

for the masses are regarded almost as slaves ') clearly point to a state
of society in which the land belonged to the rich. I may add that
Caesar expressly says that the Germans refused to sanction private pro-
perty in land 'to prevent the growth of SLvance'' {nequaoriaturfecuniae
cupiditas) and to keep the masses contented
'
{ut animi aequitate'

flebem contineant), neither of which objects was attained in Gaul ^ ;

and that it is simply inconceivable that a people who had a coinage,

a wealthy class, and an extensive commerce, and with whom slavery
was an institution, should not have recognized private property in
* See an article by F. de Coulanges, entitled Recherches sur cette question
'
:

Los Germains connaissaient-ils la propriete des terres ? in Seances et travaux de

'

V Acad, des sciences morales et poL, nouv. ser., xxiv, 1885, pp. 5-10.
- B. G., vi, 13,
§ 1.
^ lb., 22,
§ 4. M. d'Arbois indeed asserts {Comptes rendus de V Acad, des inscr.,
4* ser., XV, 1887, pp. 80-1) that when Caesar says that none of the Germans
have agri ^noduni certum aut fines proprios, he means that none of them enjoyed
the perpetual possession of a definite number of acres of ager puhlicus for, he
;

says, modus agri was the regular term used at Rome to denote the extent of ager
puhlicus which a citizen might occupy ai\d fines proprios obviously means the
;

—
boundaries of possessiones a word which Caesar uses almost in the same
—
breath 'c'est-a-dire, de champs qui font partie de Yager puhlicus.^ I call this
a triumph of special pleading. M. d'Arbois refers to Livy, vi, 35, and Corpus
inscr. Lat., \, pp. 79-86, neither of which authorities proves his point. They
simply show that the term modus agri was used in regard to ager puhlicus they :

do not show that it was not used in regard to land owned by individuals. Listen
to Fustel. '
I doubt,' he says, whether M. d'Arbois has grasped the exact
'

meaning of ager puhlicus. Ager puhlicus was not common landed property,
. . .

but property belonging to the state, which existed side by side with private
property in land. . Where did M. d'Arbois learn that modus agri was the
. .

regular term used in speaking of ager puhlicus ? In Varro {De re rustica,

i, 15) he will find the words de modo agri used unmistakably to denote the extent

of a private estate. He will also find in Varro (i, 18) the words agri modum
certum in a passage in which the writer says that the number of slave labourers
ought to be duly proportioned to the size of the estate.' Here are the passages
[

to which Fustel refers Igilur primurn haec, quaedixi, qualluor videnda agricolae,
:

de fundi forma, de terrae nalura, de modo agri, definihus tuendis (M. T. Varronis
Rerwm rusticarum i, 15). De familia Cato dirigit ad duas metas, ad certum
modum agri el genus sationis {ih., i, 18, § 1).]
512 PRIVATE PROPERTY IN LAND
land. See my quotation, on page 516, from Sir H. Maine, Early
History of Institutions, pp. 167-9.
[In a review i of the first edition of this book M. d'Arbois says that
he is in no way convinced by my arguments or by those of M. Lecri-
vain. I did not expect that he would be but I am not sure that
;

his conversion has not begun.^ The possessors of the public land
were, he now says, separated from one another by boundaries (fines)
and transmitted their possessions by heredity. If so, except that
they could not sell the land, they were proprietors did private :

property in land not exist under the laAV of primogeniture ? I now

have the additional support of M. JuUian,^ who reinforces the argu-
ments of Fustel and Lccrivain. Is it, he asks, credible that Ambiorix
was not the owner of the land which surrounded his country house ;
'*

or that Lucterius owned no land though the whole of Uxellodunum

formed part of his clientela ^ or that the authority of the state would
;

have been as weak as it undoubtedly was^ if it had controlled

"^
cultivation and disposed of the crops ?]

THE GALLIC NOBILES

Fustel de Coulanges ^ holds that the Gauls whom Caesar describes
as nohiles formed a class apart, superior to the equites but this view ;

is, I think, refuted by Caesar's express statement that in Gaul there

were two classes and only two who were held in any esteem, namely
Druids and equites [In omni Gallia eorum hominum qui aliquo sunt
numero atque honore genera sunt duo Sed de his duobus generibus
. . .

alterum est druidum, alterum equitum. B. G., vi, 13, §§ 1-3). Tlie
passages in the Gallic War which might seem to lend some support
to Fustel's view are the one in which Diviciacus is made to say that
the Aedui had lost omnem senatum, oninem
nobilitatem, omnem
equitatuni {ib., i, 31, § 6) and the one in which Litaviccus is made
to say that all the Aeduan equitatus and nobilitas had perished
{ib., vii, 38, § 2). But in these passages nobilitas does not mean a class
of men who were superior to the equites as such it simply means :

'
men of rank '. The proof is that between the two sentences,
quoted above, in which Caesar says (1) that there w^ere only two classes
which were held in any esteem, and (2) that of these two classes one
consisted of equites and the other of Druids, there occurs the folio A^ang,
Plerique cum aut acre alieno aut magnitudine tributorum aut iniuria
'potentiorum preniuntur, sese in servitutem dicant nobilibus. Is it not
clear that if the nobiles had formed a definite class, superior to the
1 Bev. celt., xxi, 1900, pp. 107-8.
- I regret that since the words in the text were written M. d'Arbois has died.
'^
Hist, de la Gaiile, ii, 71-4, 407-8.
* B. G., vi, 30, § 3. ^ lb., viii, 32, § 2.
« lb., i,4, § ;217, ; 18, § 3 ; vii, 4, § 4.
§ 1
' See also Bcv. crit. dliist. et de lilt., nouv. ser., xxx, 1890, pp. 441-2 ; and
G. Dottin, Manuel pour servir aV etude de Vant. celt., pp. 184-6.
^ Hist, des inst. pol. —
de Vane. France,- La Gaule rom., p. 14, n. 1.
 THE GALLIC NO BILE 8 513

equites, Caesar would have written, In omni Gallia eorum hominum

qui aliquo sunt numero atque honore, genera sunt Iria . . . Sed de his
trihus generibus j)rimum est druidum, alterum nohilium, tertium
equitum ? The men whom Caesar described as nohiles were simply
the most prominent, in birth or in power or in both, of the equites.^
It is worth noticing that the equites were not a caste in Caesar's :

time a man could rise by ability or influence from a humble position '

to the highest dignity '.^ Probably, as M. Jullian^ says, there were

wealthy plebeians in the Gallic cavalry but when he goes on to ;

suggest that the possession of a war-horse was a mark of nobility,

I find it difficult to follow him. Does not Caesar use the word equites
in relation to the Gauls in two senses, in the sense of cavalry and —
in that of knights ?* This seems to result from a comparison, of B. G.,
i, 18, § 5, with vi, 15, § 2. In the former passage Caesar says that
Dumnorix maintained a large number of equites (which I translate
by horsemen') at his own expense
'
in the latter he says that ;

when war breaks out

'
the knights ... all take the field and
. . .

surround themselves with as many armed servants [?] and retainers

as their birth and resources permit '. If the Gallic cavalry were all
knights, it is obvious that when a knight took the field his armed
retainers were all on foot. But Dumnorix, himself a knight, evi-
dently took the field with his large number of equites ', who were '

certainly mounted. Was each of these equites followed by his own

armed retainers on foot ? Would a knight have condescended to
follow another knight for pay ? Is it not clear that Dumnorix's
equites were, or at least included, simple troopers and were his
'
armed retainers '
?
^

THE POWER OF THE NOBLE FAMILIES IN GAUL

'
The leading families,' says Mommsen, '
of the different clans
were closely connected, and through intermarriages and special
treaties formed virtually a compact league, in presence of which the
single clan was powerless ^ and he goes on to say that the nobility ;
'

were sufficiently powerful to allow no king and no canton to accom-

'

plish the work of union There is not sufficient evidence for these
'."^

sweeping statements. Before the invasion of the Helvetii we find

their chieftain, Orgetorix, trying to establish a triumvirate with
Dumnorix and a Sequanian chief called Casticus. Orgetorix was
brought to trial for this by the Helvetii but the three chiefs were :

'
Cf. C. Julliaii, Hist, de la Gaule, ii, 69-71.
- B. G., vii, 39, § J, — Viridomarus
qucm Caesar ab Diviciaco sibi traditum
. . .

ex liumili loco ad summam

dignitatem perduxerat.
'"
Hist, de la Gaule, ii, 68. * B. G., vi,
13, § 3 15, § 1. ;

^ I find that in Vercingetorix, p. 99, M. Jullian himself endorses what I have

said.
« Hist, of Home, v, 1894, p. 21 {JUm. Gesch., iii, 1889, p. 235). The word
'
clan '
is here used in the sense of civitas or community.
^
H.,i). 24 {libm. Gesch., iii, 1889, p. 237).
1093 L 1
514 THE NOBLE FAMILIES
]iot combining against any king or against any one clan ^ and ' '
;

their combination was broken up by the action of a single people,

the Helvetii. Celtillus, the father of Vercingetorix, was put to death
for trying to make himself king but he w^as put to death by the
:

nobles of his own tribe ^ and apparently he brought his doom upon
;

himself by trying to make his power hereditary in his own family. Of

course the nobles of different tribes combined against Caesar but ;

what evidence is there that they combined to form a separate '

alliance hostile to the power of the community ? If they did, how '

are we to explain the internecine war which prevailed among them ?

'"^

It is quite true that Dumnorix formed family alliances with the

nobles of other tribes but evidently he did this simply in order to
;

strengthen his own position.^ Again, what evidence is there that

the nobility attempted to prevent any king or any canton from
accomplishing the work of union
'
or that any canton or any king,
'
;

except Bituitus, the Arvernian, whose power was broken in 121 B.C.,
and j)erhaps Vercingetorix, attempted to accomplish that work V
But I am not denying that Mommsen's theory contains a kernel
of truth. The dread of tyranny which led to the overthrow of
monarchy and impelled the Aedui and doubtless also other peoples
to restrict the powers of their magistrates would of course have
hampered any statesman who attempted to unite the tribes of Oaul.
Only I cannot see that this check would have operated through the
combination of nobles of different tribes.

ON THE MEANING OF THE WORDS AMBACTI,

CLIENTES, AND OBAERATI
Caesar says that the Gallic equites surrounded themselves with
I.
'
and clients (eorum [sc. equitum] ut quisque est genere
arnhacti '

opiisque amplissimus, ita 'plurimos circum se amhactos cUentesque

habet).^ According to the epitomator of Festus^, amhactus apud
Ermium Gallica lingua servus appellatur, which I take to mean that
Ennius used the Gallic word amhactus, learned by the Romans from
the Cisalpine Gauls, in the sense of servus. Polybius,'^ in a passage
which, as M. JuUian observes, might seem to have been copied
by Caesar, evidently translates amhacti by a-v/jLTreptcfiepoixevovs, the —
entourage or retainers of a lord. But were all these (ri'/x7rept<^€po/xa'oi
on the same footing ? If Ennius was right, servus, as the equivalent
of amhactus, clearly means, not a slave, but, according to a gloss ^

' B.O., i, 3-4. 2 lb., vii, 4, § 1. 3

11^^ ^.j^ 15, § 1.
' 7ft.,i, 3, §5; y, §3; 18, § 7.
« lb., vi, 15, § 2.
^ De
signijicatione verborum, od. Miiller, 1839, p. 4.
' 17, § 12,
ii, ncpl Se Tcii kraipdas [the Cisalpine C-aul«J i.i€yi(jTi]v onovbiju
knoiovvTo, (5ta to Kal ijiofiipojTaTov K'\\ bwarwraTOV elvai irap' avTols ruvTOV, vs av
irXdaTovs e'xf"' Sokt) rovs Oepanevovras ical avunepiipepo/xivovs avToi.
" G. Locwc, Corpus yloss. Lai., ii, 1888,
p. 1G3.
 ;

AMBAGTI, GLIENTES, AND OBAERATI 515

quoted in the editions of Festus, 8o{)A.o9 /xio-^wro?, a paid servant —

and this is the meaning which Mommsen^ and others attach to
amhadus in other words, they regard ambacti as retainers inferior
:

to cUentes.'^ Other writers, including M. JulHan,^ hold that ambacti

were clients of the highest order, akin to the Aquitanian soldurii,^
and bonnd to their lord by ties of gratitude, honour, and religion.
There is no evidence for this and the evidence of Festus, whatever
;

it may be worth, is against it.^ Caesar in another passage draws

a distinction between servi and clientes. Describing Gallic funerals,
he says, shortly before our time slaves [?] and retainers who were
'

known to have been beloved by their masters were burned along

with them after the conclusion of the regular obsequies {paulo supra '

hanc memoriam servi et clientes quos ab iis dilectos esse coiistabat iustis
funebribus confectis una cremabantur).^ I am not quite sure whether
servi here means slaves or, as A. Holder thinks, SouAot fxtcrOwToi',
' ' '^

that is, ambacti. As Schneider says, Caesar appears to use ambactus

as an established Latin word, though it was of Gallic origin, and does
not explain its meaning, as he does that of the Aquitanian word
soldurius and this militates against M. d'Arbois's theory, that he
;

wrote clientesque as a gloss upon ambactos.^

M. d'Arbois, in another work,^ draws a distinction between the
words ambactus and cliens. The Gallic word ambactus,'' he says,
'

'
had no exact equivalent in Latin its meaning was intermediate
;

between that of servus and that of cliens the tie which bound an ;

ambactus to his lord was looser than that which bound a servus to his
owner, closer than that which bound a Roman client to his patron.'
M. d'Arbois is here substantially in agreement with those who believe
that ambacti were inferior retainers. Schneider thinks that the term
ambactus applies to those who, as Caesar says, entered the service
of nobles because they were oppressed by debt or taxation or wronged
by powerful individuals, and over whom their respective lords had
'
all the rights that masters have over their slaves ^^ while he '
;

identifies the clientes with the retainers of whom Caesar says in

another passage ^^ that if their lord failed to protect them against
oppression, he lost all authority over them. I am not sure that
there is enough evidence to establish this distinction but I so far ;

agree with Schneider that I incline to regard ambacti as retainers

of a lower grade than clientes and this is the only theory that has
;

any documentary support.

n. Caesar says that Orgetorix, in order to resist arrest, assembled
his familia, his clients and his obaerati or debtors (omnem suam
' '

familiam ad hominum fnilia decern undique coegit, et omnes clientes

* E'nm. Gesch., 1889, p. 234, note {Hist, of Rome, v, 1894, p. 20, note).
iii,
^ Daremberg and kSaglio, Did. des ant. grecques et rom., i, 223.
^ Hist, de la Oaule, ii, 77-8. * B. G., iii, 22,
§§ 1-3.
^ M. Jullian insists that ambactus was traduit a tort par servus chez Ics
'

lexicographes ', that is, Fabius.

« B. G., vi, 19, § 4. ' Alt-ceUischcr Sprachschatz, i, 115.
* Cf. A. Klotz, Caesar studien,
p. 40.
9 Reclierches sur Vorigine de la propriete
fancier e, &c., 1890, i). 04, n. 1.
1" B. G., vi,
13, §2. u lb., 11, §4.
Ll 2
516 AMBACTI, CLIENTES, AND OBAERATI
ohaeratosque suos eodem conduxit).^ By obaerati, says J. Flach,^
. . .

Caesar means not only enslaved debtors, but also clients who rendered
military service in return for grants of cattle. This seems to me
a gratuitous assertion. Possibly in the phrase clientes ohaeratosque
the latter word might grammatically be regarded as explanatory of
the former but it seems more natural to identify the obaerati with
;

the debtors,^ mentioned in B. G., vi, 13, § 2. In this passage, as I have

already observed, Caesar says that it was usual for men who were
burdened with debt or taxes or wronged by powerful individuals to
take refuge in servitude to some noble and, he adds, the nobles
;

possess the same rights over them that masters have over their
slaves (plerique, cum aut acre alieno aut mainitudine tributorum aul
iniuria potentiorum premuntur, sese in servitutem dicant nobilibus :
in has eadem omnia sunt iura quae dominis in servos).^ M. d'Arbois
de Jubainville,^ however, concludes from the passage in which Caesar
describes the following of Orgetorix that a Gallic chieftain had two
classes of retainers, free and non-free and assuming that Orgetorix's
;

familia included his slaves, and remarking that Caesar distinguishes

the familia from the clientes and obaerati, he infers that the obaerati
Avere free. The free retainers (vassaux) were the companions of their
lord, and, although poor, held a rank almost equal to his they :

resembled the soer-cheli of Ireland. The other class of vassaux ', '

who resembled the Irish doer-cheU, comprised shepherds and rural

labourers, whose condition was analogous to that of Roman slaves.
I do not altogether agree with M. dArbois. He appears to me to
overlook the fact, expressly stated by Caesar, that those debtors
who sought the protection of a lord, and who surely are to be identi-
fied with the obaerati, were virtually his slaves, even though they
may have been distinct from the slaves, born in slavery or purchased,
w^hom Caesar calls simply servi. If I am right, the free retainers,
whose position was analogous to that of the Irish sOer-chcli, were not
the clientes and the obaerati, but the clientes only.^

1 i?. (?., i, 4, § 2.
^ Les origines de Vancienne France, 1884, i, 57.
^ P. Geyer {Jahresb. d. philol. Vereins zu Berlin, v, 1879, p. 341) assumes that
these debtors were clientes but Caesar does not call them by that name and
:

there is nothing to show that it was applicable to them.

« B. a.,xi, 13, §§2-3.
^ Eecherches sur Vorigine de la propriete fonciere, &c., jjp. 04-5.
® The following remarks of Sir Henry Maine {Early Hist, of Institutions.
pp. 107-9) illustrate the position of the obaerati
— We obtain from the
'

[Irish] law-tracts a picture of an aristocracy of wealth in its most primitive

form and we see that the possession of this wealth gave the nobles an immense
;

power over the non-noble freemen who had nothing but their land. Caesar
seems to me to be clearly referring to the same state of relations in the Celtic
sister society, when he speaks of the Gaulish chiefs, the Equites, having one
principal source of their influence in the number of their debtors {B. G., i, 4;
B. G., vi, 13). Now you will remember how uniformly, when our knowledge
of the ancient world commences, we find plebeian classes deeply indebted to
aristocratic orders. At the beginning of Athenian history we find the Athenian
commonalty the bond-slaves through debt of the Eupatrids at the beginning
;

of Roman history we find the Roman commons in money bondage to the

Patricians. The fact has been accounted for in many ways, and it has been
 517

SLAVERY IN GAUL
It has been denied that slavery existed in Gaul but, as Fustel ;

de Coulanges ^ remarks, three passages in the Commentaries, namely

V, 45, § 3,2 vi, 19, § 4,3 and viii, 30, § 1,4 prove that it did.

INTER-TRIBAL RELATIONS IN GAUL

C. A. Serrure ^ infers from certain passages in the Commentaries
that inter-tribal relationship in Gaul was of three kinds, namely
(1) simple alliance {in amicitia esse, in fide esse), (2) protectorate {in
dientela), and dependence {suh imperio esse). According to the
(3)
Aeduan, Diviciacus, the Bellovaci had always been in fide atque
amicitia civitatis Haeduae in 53 B.C. the Senones sued for Caesar's
:

pardon per Haeduos quorum antiquitus erat in fide civitas and in the ;

following year the Bituriges asked the aid of the Aedui, quorum erant
in fide, against Vercingetorix.^ It is clear to my mind that in fide
denotes something more than mere alliance it is usually, and I think
:

rightly, translated under the protection of

'
after the submission
'
:

of the Bellovaci Caesar says that, out of regard for the Aedui and
Diviciacus, who had interceded for them, he will himself receive them
in fidem^ —under his protection
' and Cicero says, speaking of the
'
;

assassins who had been employed by Chrysogonus, quaere in cuius

fide sint et dientela,^ Long ^ is, I believe, right in saying that the
Bituriges (and of course he meant the Senones and the Bellovaci as

plausibly suggested that it was the occurrence of repeated bad seasons which
placed the small farmers of the Attic and Roman territory at the mercy of
wealthy nobles. But the explanation is imperfect unless we keep in mind the
chief lesson of these Brehon tracts, and recollect that the relative importance
of land and capital has been altering throughout history. ... In very ancient
times land was a drug, while capital was extremely perishable, added to with
the greatest difficulty, and lodged in very few hands. The ownership of the
. . .

instruments of tillage other than the land itself was thus, in early agricultural
communities, a power of the first order, and, as it may be believed that a stock
of the primitive capital larger than usual was very generally obtained by
plunder, we can understand that these stocks were mostly in the hands of noble
classes whose occupation was war, and who at all events had a monopoly
of the profits of office. The advance of capital at usurious interest, and
the helpless degradation of the borrowers, were the natural results of such
economical conditions.'
^ —
Hist, des inst. pol. de Vane. France, La Gaule rom., p. 22.
^ hie (Vertico) servo spe libertatis magnisque persuadet praemiis ut litteras
ad Caesarem deferat.
^ Funera sunt pro cultu Gallorum magnifica ac paulo supra hanc memo-
. . .

riam servi et clientes, quos ab iis dilectos esse constabat, iustis funebribus
confectis una cremabantur.
* Qua ex fuga cum constaret Drappetem Senonem, qui, ut primum defecerat

Gallia, collectis undique perditis hominibus, servis ad libertatem vocatis, &c.

^ Etude sur la numismatique gaul. des Comm. de Cesar, 1885,
p. 35.
« B. G., ii, 14,
§ 2 vi, 4, § 2
; vii, 5, § 2.
; lb., ii, 15, § 1. ->

® Pro Boscio Amerino, ^ Caesar,

33, § 93. p. 335.
518 INTER -TRIBAL RELATIONS
well) were among the veleres dientes of the Acflui mentioned in
B. G., vi, 12. It is clear, then, that to be in fide was consistent with
being in clientela.
In 53 B.C., the Carnutes, when soliciting Caesar's pardon, availed
themselves of the intercession of the Remi, quorum erant in clientela^
To my mind in clientela here denotes virtually the same relationship
as in fide.
Certain passages in the Commentaries prove that clientela, which
Serrure distinguishes from the state of dependence denoted by the
words sub imperio esse, was sometimes identical with it. The dientes
of the Nervii were under their imperium and the new clients who ;

placed themselves under the protection of the Aedui after Caesar's

arrival in Gaul found that they enjoyed in consequence aequiore
imperio.^
The truth is that theword clientela was elastic. A state which had
clients exercised over them whatever power it could and some ;

clients were less dependent upon the same overlord than others.
The Bellovaci, the Bituriges, and the Senones were in one sense no
doubt clients of the Aedui and the Carnutes of the Remi but they ;

were not under their imperium, for the Senones and the Carnutes
rebelled when the Aedui and the Remi remained loyal. In B. G., vii,
75, § 2, Caesar specifies those clients of the Aedui who contributed
along with them contingents for the relief of Vercingetorix, and there-
fore were presumably under their imperium they were the Segu- :

siavi, the Ambivareti, and the Aulerci Branno vices. The Eleuteti,
the Cadurci, the Gabali, and the Vellavii stood in a similar relation
to the Arverni.^
One wishes, however, that Caesar had or could have told us more
of the nature of the hegemony exercised by the Aedui and their
principal rivals. Mommsen says that A powerful canton induced '

a weaker to become subordinate on such a footing that the leading

canton acted for the other as well as for itself in its external relations
and stipulated for it in state treaties, while the dependent canton
bound itself to render military service, and sometimes also to pay
a tribute '.* At all events the dependent canton did sometimes
render military service and we find the Eburones paying tribute to
;

the Atuatuci.^ The dependent peoples evidently managed their own

internal affairs and, as we learn from B. G., vi, 12, §§ 6-8, client
;

tribes, even when they were under imperium, occasionally trans-

ferred their allegiance from one overlord to another.
AVhat again was the nature of the imperium which Diviciacus, king
of the Suessiones, exercised over South-Eastern Britain,^ and how

B. G., vi,4, § 5. 2 7^^ V,

1
39, §§ 1, 3 vi, 12, § 6. ;

^ M. JulHan {Hist, de la Gaule, ii, 440) distinguishes between the Vellavii,

who, he holds, were the subjects of the Arverni, and the Gabali, who perhaps
'
'

only paid them tribute and owed military service. I cannot see any sufficient
reason for this distinction for Caesar says that all three tribes were under
;
'

the sway' of the Arverni {Cadurcis, Gahalis, Vellaviis, qui sub imperio Arrer-
norum esse consuerunt [B. G., vii, 75, § 2]). See, however, p. 345, supra.
* HisL of Rome, v, 1894, p. 24 {Rem. Ge.'^ch., iii. 1889, p. 238).
s
B. G., i, 31, § 6 ; v, 27, § 2 ; 39, § 1. « Ih., ii, 4, § 7.
 —
INTER-TRIBAL RELATIONS 519

had itbeen acquired ? One cannot believe that he had invaded

Britain and conquered his Belgic kinsmen. May we suppose that
after he had made himself overlord of a large part of the Belgic '

territory ', some of the British tribes had sought his aid against
their rivals, and had purchased it by recognizing his supremacy and
perhaps also by paying tribute ? When Caesar came to Gaul the
tribes of South-Eastern Britain were divided into antagonistic groups,
headed respectively by the Catuvellauni and the Trinovantes and ;

perhaps this rivalry may explain both the supremacy of Diviciacus

and its decay.^
M. d'Arbois de Jubainville warns us not to confound les litats
^ '

clients avec les pewples clients qui ont cesse de former un Etat separe
et dont I'armee est fondue dans celle de I'Etat sous I'autorite ou impe-
riuni duquel ils se sont places. Ces clients de second ordre ou sujets ne
re9oivent pas dans les Commentaires le titre d'Etat, civitas, donne
. . .

par I'auteur aux clients de premier ordre.' The contingents, he goes

on to say, of the Aeduan client peoples the Segusiavi, the Ambi- —
vareti, and the Aulerci Branno vices —
who formed part of the host
which marched to relieve Alesia,^ se confondaient avec celui des
'

Aedui.' This is true but the contingent of the Cadurci was also
:

lumped with that of the Arverni yet, as M. d'Arbois has to admit,

;

Caesar calls the Cadurci a civitas^

One other inter-tribal tie requires notice, that which Caesar —
denotes by the words fratres consanguineosque ^ (brethren and kins-
men), which united the Remi and the Suessiones, and by the words
necessarii et consanguinei ^ (friends and kinsmen), which united the
Aedui and the Ambarri. The Reman ambassadors told Caesar that
the Suessiones had the same rights and laws as themselves and
'

jointly owned the authority of one and the same magistrate ', who
in their case was a king {qui eodem iure et isdem legihus utantur, unum
imperium unumque magistratiim cum ipsis haheant) and the con- ;

nexion between the Senones and the Parisii had once apparently
been similar. It is remarkable that Caesar uses the same words
"^

—
fratres consanguineosque to denote the relation into which the
Aedui had entered in the second century B. c. with the Roman
People.^

^ Rice Holmes, Anc. Britain, pp. 299-300.

2 Recherches sur Vorigine de la propriete fonciire, pp. 30-2, 34.
' B. G., vii, 75, § 2. « Ih., vii, 6, § 7.
5 Jh., ii, 3, § 5. « Ih., i, 11, § 4. ' lb., vi, 3, § 5.
8 lb., i, 33, § 2. Cf. 44, § 0. 0. Hirsehfekl {Sitzungsberichte d. Kdnigl.
preuss. Akad. d. WissenscJiaften, 1897, pp. 1110-2) devotes some interesting
pages to this form of alliance.
 520

WERE THE PHILO-ROMAN AND THE ANTI-ROMAN

PARTIES IN GAUL IDENTICAL WITH THE
REPUBLICANS AND THE ADVENTURERS RE-
SPECTIVELY ?
Fustel de Coulanges ^ thinks that the philo -Roman party among
the Gauls was identical with the supporters of republican institu-
tions, and the party hostile to Caesar with the powerful adventurers,
such as Dumnorix, who aimed at making themselves kings, and with
the clients and members of the lower orders who supported them.
' '

There is no inconsistency, he maintains, between this general prin-

ciple and the individual instances in which Caesar set up a king over
such and such a people.
Desjardins, on the other hand, says that the object of Caesar in
setting up these kings was to curb the democratic spirit and to
paralyse the aristocratic leagues.^
I show in another article that no democratic spirit worth noticing,
in the sense in which Desjardins understands the term, existed
in Gaul in Caesar's time ^ and I do not know what he means by the
;

aristocratic leagues. Caesar set up three kings, Commius, Tasgetius,

and Cavarinus and his object was to reward useful adherents and
;

to strengthen his own hold upon the country."*

For the rest, Dumnorix certainly was hostile to Caesar so was :

Vercingetorix and so in general were the powerful adventurers

;

who aimed at making themselves kings but I do not think that :

Fustel has proved that the philo -Roman party if there was such —
—
a "party was identical with the supporters of republican institutions.
Tasgetius and Cavarinus both belonged to royal families, and were
no doubt opposed to the supporters of republicanism in their re-
spective states before Caesar appointed them. In Caesar's first cam-
paign there is no evidence that any opposition was offered to him by
any of the Gauls, except Dumnorix and his followers. In his second
campaign all classes among the Belgae, with the exception of the
Bemi, appear to have been unanimous in opposing him. In the
history of the campaign against the maritime tribes there is no trace
of any philo-Boman party, except perhaps among the Eburovices
and the Lexovii, whose senates, republican no doubt, were opposed
to war ;and it is reasonable to suppose that they wished to keep the
peace because they had the good sense to j^erceive that they had no
chance of contending successfully against Caesar. In the fifth and
sixth campaigns, with the exception of the persistently servile Bemi,
Tasgetius, Cavarinus, and Cingetorix are the only philo -Romans who
are mentioned, and we do not know whether Cingetorix was a repub-
lican or not while Ambiorix and Catuvolcus, both constitutional
;

^ Hist, des inst. pol. de Vane. France, — la Gaule rom., pp. 52-3.
^ Geogr. de la Gaule rotn., ii, 548. ^ See
pp. 529-41.
' B. G., iv, 21, § 7 ; v, 25, § 2 ; 54, § 2.
 ;

PHILO-ROMAN AND ANTI-ROMAN 521

kings, rebel and after the disaster at Atnatnca, Caesar has reason
;

to suspect every Gallic people, except the Aedui and the Kemi,
of rebellious designs. Even then no doubt he had interested ad-
herents among other peoples besides the Aedui and the Kemi but he ;

does not mention them. The opponents of Tasgetius and Cavarinus

were his opponents, and surely they were republicans the Senones :

and Carnutes, apparently of all parties, were anti-Roman. In the

seventh campaign, who were philo -Romans, except the Arvernian
—
opponents of Vercingetorix, republicans, it is true and their friend-
ship to Rome was a compound of shrewd calculation and hatred of
—
Vercingetorix a section of the Aedui, the Remi, and the Lingones ?
Towards the close of the campaign so intense was the unanimous
'

determination of the entire Gallic people to establish their liberty

and to recover their ancient military renown that no favours, no re-
collection of former friendship had any influence with them, but all
'
devoted their energies and resources to the prosecution of the war
{tanta tamen universae Galliae consensio fait libertatis vindicandae et
fristinae belli laudis recwperandae, ut neque heneficiis neque amicitiae
inemoria moverentur, omnesque et animo et opihus in id helium incum-
herent [vii, 76, § 2]). A passage in Hirtius's narrative of the eighth
campaign, however, supports the theory of Fustel the constitu-
:

tional party among the Bellovaci appear to have been opposed to

hostilities at least they throw the blame of rebellion upon Correus,
;

the adventurer and leader of the flehs.^

I believe that there is some truth in Fustel's theory but it is too
;

broadly stated. Our knowledge amounts to little more than this :

— Caesar's friends were his friends, republican or royalist, as the case

might be all whom he could gain over by favours or expectation of
;

favours, all who hoped by his support to triumph over or to pay off
old scores against enemies among their own countrymen, all who
were shrewd enough to see that he was going to win.
It is remarkable that, with the exception of the Veneti and their
allies, and of the Senones, Carnutes, and Treveri, not a single Celtican
tribe rose in rebellion against Caesar until the seventh year of the
war. I have tried in my narrative to account for this as far as
the evidence would allow but it is possible that if Caesar had chosen
;

to take us into his confidence, there would have been more to say.^
Signor Ferrero has recently made a characteristic contribution to the
discussion of the subject of this article. He asserts that Caesar after his
second campaign learnt that the Diets and Assemblies of Notables
'

[that is to say, the tribal councils] were feeble and decadent

bodies ... he made up his mind to abandon the Conservatives . . .

and to depend upon the popular party, which had been hitherto
steadily opposed to him '? Let us analyse the evidence upon
which this theory rests. After the overthrow of the Atrebates in
1 B. G., i, 17-20 ii, 1-4, 24, § 4
; iii, 8, 16, 17, § 3
; ; v, 3-4, 6-7, 26, § 1
53-4 vi, 3-4, 8, § 9
; vii, 4, § 2
; 33, § 1
; ;63, § 7; viii, 21, § 4.
^ See B. C, iii, 59, which gives us an inkling
of his methods.
^ Grandezza e decadenza di Roma, ii, 85-6
{The Greatness and Decline of
Rome, ii, 61-2).
522 PHILO-ROMAN AND ANTI-ROMAN
57 B.C. Caesar nominated Commius as tlieir kinir,i and in acknow-
ledgement of his diplomatic services in Britain granted his tribe '

immunity from taxation, restored to it its rights and laws, and

placed the Morini under his authority '.^ So far as we know, the
appointment was not unacceptable to the Atrebates. Tasgetius was
a Carnutian of noble birth'
whose ancestors had held sovereignty
. . .

in their own country, and to whom Caesar, in recognition of his

energy and devotion (for in all his campaigns he had found his
services exceptionally valuable) had restored their position'. He
was made king in 56 B.C., and was assassinated two years later with '

the avowed sanction of many of the citizens '.^ Cavarinus's brother, '

Moritasgus, had held sovereignty [over the Senones] when Caesar

came to Gaul, and his ancestors before him'. In 54 B.C. Cavarinus
was sentenced to death by the council of the Senones, and, as he
contrived to escape, was formally dethroned and banished.* Signor
Ferrero, referring to the assassination of Tasgetius, remarks that it '

looked as if the Nationalists intended to make his assassination the

beginning of a series of reprisals against all the Gallic leaders who
had consented to recognize the Roman dominion '.^ Evidently, then,
if Signor Ferrero is right, Caesar had only succeeded in inducing

Tasgetius to betray the Nationalists in his own country, and had

completely failed to gain the support of the popular party '. But '

what are the facts ? There is not the least evidence that Tasgetius
was ever a Nationalist he was from the first an ardent philo-
:

Roman. The citizens who sanctioned his assassination may

have been members of the Diet themselves '
and it is clear
' ;

that if Cavarinus was sentenced to death by the Nationalists, the

Nationalists and the council were identical, and the action of the
council, which Signor Ferrero assumes to have been republican
and conservative ', was neither decadent nor feeble ', but
' '
'
'

rather headstrong. About the state of parties among the Atrebates

we know absolutely nothing Commius was, for aught that we
:

know, not only originally a philo-Roman himself but the most in-
fluential representative of the philo-Roman party. M. Jullian,^ who
does not commit himself to any theory about the Nationalists, thinks
that in promoting Tasgetius and Cavarinus, Caesar so far renounced
his policy of protecting the oligarchies. But we do not know whether
that had been his settled rule and, as I have just remarked, the
;

Senonian oligarchy at all events was vigorously anti-Roman.

It all comes to this. There is not one tittle of evidence that Caesar
ever changed his policy. He appointed Tasgetius and Cavarinus
because they seemed the best instruments that he could find to keep
the Carnutes and the Senones respectively under control and, as ;

far as we can tell, they succeeded for a time in doing so. Those two
tribes were the most irreconcilable of all Caesar's enemies and their ;

'
feeble and decadent councils were evidently plus nationalistes que
'
'

'.
les nationalistes

' E.G., iv, 21,^1. 2 76., vii, 70, § 1. ' lb., V, 25, U- * 76., 54, §2.
5 The. Greatness and Decline of Borne, ii. SI. The passage not in the
is
Italian 6(1. of 1902. « Ifid. clela Gaule,u\,3l^
 PHILO-ROMAN AND ANTT-ROMAN 523

Nothing is easier for a clever man than to compose a plausible

and interesting account of la haute politique in Gaul, as practised by
Caesar, ifhe is ready to ignore facts and give the rein to his imagina-
tion. But even with all the help that we can get from Cicero's
letters,we know too little of the course of politics in Italy ; and of
the course of politics in Gaul we know hardly anything.

THE DRUIDS I

1.Where and when did Druidism originate ? Caesar in a well-

known passage remarks that it was believed to have been discovered
in Britain andhave been imported thence into Gaul - and some
to ;

scholars accept this tradition as literally true. The earliest extant

mention of Druids, the substance of which was reproduced in Dio-
genes Laertius's Lives of the Philosophers,^ was made by Sotion of
Alexandria * about the commencement of the second century before
Christ, —
not long after the Belgic conquest of Britain began and it ;

has been supposed that the conquerors found Druidism flourishing

there, and made it known in the land from which they had set out.
But the Belgae were not the first Celtic conquerors of Britain and ;

it is reasonable to suppose that if Druidism was of British origin, it

had been imported into Gaul before. Maximin Deloche argued that
it could not have originated in Gaul, since the Gallic invaders of
Italy knew nothing of it ^ but A. Reville replied, first, that Caesar
:

did not afhrm its British origin, but only recorded the belief which
he found prevalent among the Gauls and, secondly, that if the
;

Gauls who invaded Italy had no Druids, Druids may nevertheless

have existed in Gaul before the invasion a I'etat d'humble com-
'

pagnie de sorciers-medecins '.^ This of course is a mere conjecture ;

but the assumption that Brennus and his followers were not accom-
panied by Druids is only an inference from the silence of the his-
torians and from the statement of Pausanias ^ that the Gallic chief
who attacked Delphi did not employ any Greek priests or diviners ^ '
'
;

and in the third century B.C. at all events the Cisalpine Boi had
priests.^ Indeed the prevalent theory is that Druidism was pre-
Celtic and developed in the Neolithic Age in Gaul or in Britain,
possibly in both and if this belief were well founded, it would
;

^ The etymology of the word Druid is still disputed. M. d'Arbois de

'
'

Jubainville {Rev. celt, xxix, 1908, p. 83) defends Thurneysen's view that —
it was derived from dru (fortement), uid (savant)
'
—
against M. Jullian {Hist,
'

de la Gaule, ii, 85, n. 7).

2 B. G., vi, 13,
§ 11, —
Disciplina in Britannia reperta atque inde in Galliam
translata existimatur.
3 Ed. Didot,
p. 1, 11. 9-12.
* Or possibly in the treatise called to MayiKov, once erroneously ascribed

to Aristotle. Diogenes cites both.

s Rev.
des Deux Mondes, 3® per., xxiv, 1877, p. 46G.
« 76.,
pp. 473-5. ^ X, 21, § 1.
8 Cf. C. Jullian, Hist, de la Gaule, ii, 89. « Livy, xxiii,
24, § 12.
 524 THE DRUIDS
follow that, unless Druidism was a late importation from Britain,
the Gallic invaders of Italy must have known it before they left
their own country. M. Salomon Eeinachi attributes the megalithic
monuments of Gaul to Druidical influence, arguing that their con-
struction is inexplicable except on the hypothesis of a religious '

aristocracy exercising an almost absolute dominion over a numerous

population '. But is it necessary to suppose that this aristocracy
was a hierarchy ? And if it is a fair conclusion that the supposed
hierarchy was composed of Druids, might it not be argued that
Druidism was a world-wide institution, or at least co-extensive with
rude stone monuments ?2 M. Jullian,^ remarking that the Germans,
who were congeners of the Celts and had always been more or less
closely connected with them, had no priests in Caesar's time,^ but
had a century later,^ argues that among the Celts also the priestly
ofhce had once been exercised by the tribal kings, but had naturally
become specialized in other words, he holds that Druidism was
:

a Celtic institution, which made its appearance comparatively late,

probably in the third century B.C. But Druidism, as we may infer
from Caesar's words, had flourished in Britain long before his time,
and, as we may infer from the Annals ^ of Tacitus, especially in the
western and less civilized districts and it may be doubted whether at
;

the time of its birth remote British tribes were politically or socially
more advanced than the German contemporaries of Caesar. M. Jullian
indeed urges that if the Druidical doctrine originated in Britain,
Druids may nevertheless have first appeared in Gaul for, he says, ;

'
la doctrine et le clerge d'une religion peuvent avoir des berceaux
fort differents voyez le Christianisme.
: But, replies M. d'Arbois
' '^

de Jubainville, Christianity was introduced into the Roman Empire

—
by Jews, the disciples of Jesus besides, how^ could an unwritten
:

doctrine like Druidism have been imported into Gaul from Britain
unless Druids had brought it ? ^ M. Jullian' s theory implies that in
Britain Druidism was a religion without a clergy that after it was ;

imported into Gaul, a clergy was formed, who were called Druids ;

and that subsequently Britain, which had given Druidism to Gaul,

borrowed from Gaul the idea of appointing Druids. Is not this very
improbable ?

^ Compfes rendus de V Acad, des inscr., 4" ser., xx, 1892,

pp. C-7. See also
Rice Holmes, Ayic. Britain, p. 291, n. 2.
2 M. Reinach {Orpheus, 1909, p. 177) is now inclined to believe that Druidism

originated in the Neolithic Age in the British Isles that it flourished principally
;

in Ireland and that it early spread to the Continent. Julius Pokorny {Celtic
;

Review, v, 1908-9, p. 19) insists that the fact that the oak plays no part in
'

the life of the Irish Druid and a very small one in the popular superstition can
be explained only if we assume that the Druids were originally the priests of
a people who did not know the oak worship', that is {ib., p. 6) of 'the pre-
Celtic aborigines of the British islands'. I cannot follow this argument. If
Pokorny has stated the facts correctly, they only prove that oak-worship was
unknown to the Irish aborigines, and that the Celtic invaders of Ireland, who
undoubtedly knew it, were unable or unwilling to impose it upon them.
3 Hist, de la Gaule, ii, 88-9. * 5. G'., vi, 21, § 1.
^ Tacitus, Germania, 10. ^ xiv, 30.
'
Hist, de la Gaule, ii. 87, n. 3. « Eev. celt., xxix, 1908, p. 81.
 THE DRUIDS 525

My own always be room for hypotheses

belief is that there will ;

but that we can never get behind Caesar's report of the Gallic
— —
tradition that the disciflina the Druidical doctrine originated in
Britain, and, in the absence of other positive testimony, had better
cling to it. Caesar's statement is sometimes explained in the sense
that in his time Druidism was more vigorous in Britain than in Gaul,
and that Gallic Druids therefore travelled to Britain in order to be
initiated into its mysteries ^ but why should we not accept it in
;

its natural sense ?

2. Desjardins ^ points out that there is no evidence that Druidism
existed in Caesar's time in Aquitania or in the Koman Province ;

and, he says, it seems clear from Caesar's narrative that the Romans
came in contact with Druids for the first time when they had passed
beyond the northern boundary of Roman territory. But neither is
there any evidence that Druidism did not exist in the Province and it ;

seems clear that it must have existed there before the inhabitants
became Romanized. Sir John Rhys^ maintains further, that there
is no evidence that Druidism was ever the religion of any Brythonic
people and since he assigns almost the whole of Britain south of
;

the firths of Forth and Clyde to the Brythons, he appears to restrict

the British area of Druidism to a narrow western fringe. This hardly
accords with Caesar's statement that Britain was the stronghold of
Druidism. Moreover, when Caesar tells us that the Druids were the
religious aristocracy of the Gauls, he plainly gives us to understand
that Druidism was common to all the peoples who lived between the
Seine and the Garonne and it is certain that among these peoples
;

the Gallo- Brythonic element was predominant. Indeed, although it

is commonly assumed that the Belgae had no Druids, there is abso-
lutely no ground for the assumption. Caesar often used the word
Gain in a comprehensive sense, including the Belgae and it is not
;

improbable that when he was describing the manners and customs

of the Gauls, and Druidism, which was their most remarkable in-
stitution, he intended his description to apply to the Belgae.^
3. It has often been alleged that Caesar overestimated the power
of the Druids.^ Reville, observing that there is no evidence that
Vercingetorix received any support from them, holds that Caesar was
misled by Diviciacus, who hoped by exaggerating the importance

^ See Neue Jahrb. f. Philologie, &c., cxlv, 1892, p. 770.

^ Geogr. de la Gaule rom., ii, 515.
3 Celtic Britain, 3rd ed., 1904,
p. 69.
* The political condition of the people of Brythonic Britain,' says Sir
'

John Rhys {Celtic Britain, 3rd ed., 1904, j^p. 57, 01), 'towards the end of the
Early Iron Age and the close of their independence, is best studied in connection
with that of Gaul as described by Caesar. . .The state of things, politically
.

speaking, which existed in Gaul, existed most likely among the Belgic tribes
in Britain.' That is to say, Sir John accepts the political part of Caesar's
description as applying to the Belgic and the other Brythonic tribes of both
Gaul and Britain. Yet he insists that that part of the same description which
deals with Druidism, and which is indissolubly connected with the political
part, has nothing to do either with the Belgae or the other Brythons.
5 Cf. Roget de Belloguet, Etknoyenie gauloise, 1858-68, iii, 310, and Diet. arch,
de la Gaule, i, 430.
526 THE DRUIDS
of owji order^ to secure for himself I'liegciiiouic spirituelle dc
liis '

la Giiiilo '."^ J )esja.rdins,^ however, points out that by such luis-

rcpresentatio]! Diviciacus would have defeated his own object and ;

he believes that Vercingetorix regarded the Druids as antagonistic

to his own power and therefore would have nothing to do with them.
Desjardins's view is that Caesar's account is correct as far as it goes,
but that it refers to a state of things anterior to the Roman conquest,
especially as regards the custom of human sacrifice. This is a pure
guess, based, I suppose, upon the notion that the Gauls, in Caesar's
time, were too civilized to sacrifice human victims. But Caesar tells
us that it was their custom to torture to death the warrior who was
the last to present himself at the general muster which preceded
a miUtary expedition ^ and if the Romans, in the second Punic war,
;

offered human victims to the gods,^ if suttee prevailed two genera-

tions ago in India, why should it be incredible that human sacrifice
was practised in Gaul ? Moreover, TertuUian ^ affirms that the
Druids offered human sacrifices to Mercury, and Suetonius ' speaks
of the superstition of the Druids with its terrible cruelty '. Sir John
'

Rhys ^ says that in Ireland

' druidism and the kingship went hand
. . .

in hand ;nor is it improbable that it was the same in Gaul, so that

when the one fell, the other suffered to some extent likewise' while ;

Duruy ^ affirms that the nobles had dealt a fatal blow at the power
of the Druids. I cannot find a scrap of evidence in support of this
assertion on the contrary, Caesar expressly says that in 52 B. c. the
;

Aeduan Vergobret, or chief magistrate, was elected by the priests.^^

Mommsen ^^ takes a different view from Duruy's. It may readily '

be conceived,' he says, that such a priesthood attempted to usurp,

'

Fustel de Coulanges {Hist, des inst. pol. de Vane. France, la Gaule rom.
^ —
p. 29, 11. 2) remarks that Caesar, who knew Diviciacus intimately, does not say

that he was a Druid. But Cicero, who conversed with him on matters of
religion, says that he was and Caesar's silence proves nothing. When one reads
;

the passage in Cicero, his testimony appears conclusive eaque divinationum

:

ratio ne in barharis quidem gentihus neglecta est, siquidem et in Gallia Druidae

sunt, e quibus ipse Divitiacum Haediium coynovi, qui et naturae rationem,
, . .

quam. ijwaioXoyiav Graeci appellant, notam esse sibi projitebatur et partim auguriis,
parlim coniectura, quae essent Jutura dicebat, &c. De Divin., i, 41, § 90.
M. Salomon Reinach {Orpheus, 1909, p. 179) thinks that we must distinguish
between nobles who had been merely educated by Druids, comme etait sans '

doute ce Divitiac,' and the sacerdotal body properly so called ; but the fact
' '

of Diviciacus' s having received a Druidical education does not invalidate

Cicero's statement, — that he was himself a Druid.
^ Rev. des Deux MondeSf 3^ per., xxii,
1877, p. 849.
^ Geogr. de la Gaule rom., ii. 532.

* B. G., v,
5(5, § 2. 6 Livy^ xxii,
57, §§ 10.
Apologeticus, 9, Adversus Gnosticos, 7, quoted by M. M. P. Monccaux in
®

XXXV, 1887, p. 255.

liev. hist.,
' Claudius, 25,
— Druidarum religionem dirae immanitatis.' A. Bert rand,
'

remarking {La religion des Gaulois, p. 252) that there is no trace of human
sacrifice in the history of Ireland
—
' le pays druidique par excellence argues
'
—
(pp. 08-73, 380) that the Druids did not originate, but merely tolerated and
sanctioned this rite, which he believes to have been a survival of prehistoric
times. ^ Celtic Heathendom. 1888, p. 231.

» Hist, des Romains, iii, 1889, 119. 131. '» B. G., vii,
33, § 3.
pp.
" Hist, oj Rome, v, 1894, p. 20 {R6m Gesch., iii, 1889, p. 237).
 THE DRUIDS 527

as it partially did usurp, tlie secular government. . The Gauls were

. .

not much removed from an ecclesiastical state with its pope and
councils, its immunities, interdicts, and spiritual courts.' If we are
entitled to infer from Caesar's silence that the Druids did not play
the part of a national priesthood in the Gallic war, does the truth of
the inference prove that their power had been broken by the nobles ?^
But the silence of Caesar regarding the part which the Druids
played in the great rebellion has exercised the ingenuity of many
commentators. A singularly powerful priesthood,' says Professor
'

Haverfield,^ numbering political leaders, like Divitiacus, among its

'

ranks, might be expected in a national crisis to take some definite

line, requiring notice in the Commentaries. Yet omit two chapters,
and so far as the Commentaries go, the Druids might never have
existed.' M. Jullian ^ argues that they did take an active part in the
rebellion, but that Caesar chose to ignore the fact Caesar a :
'
laicise
a outrance V esprit et Vhistoire de la Gaule Nul ne croira que la
. . .

Gaule 71' ait pas appele pretres et dieux a son secours '. Professor
Haverfield, who naturally asks What motive had Caesar for this ?
'
'

suggests that an analogy to these powerful non-political priests

'
. . .

is provided by various priestly collegia at Rome, which include

political teachers, but which in their augural or other capacity take
no political action ', and maintains that the Druids, as Druids, '

uttered no word against Caesar or for him.' But if so, why, at

a time when their power had certainly diminished, did they aid and
abet the insurrection of Civilis ? * To my mind Caesar's attitude
hardly needs explanation. Unless he had a motive for concealment,
he omitted nothing which had an important bearing upon his opera-
tions, but he omitted everything else. I agree with M. JuUian that
Druids, as Druids or as individuals, sided with or against the rebels
in the movement of 52 b. c. it is inconceivable that they held aloof .^
:

That they interfered as a corporation is, however, most impro-

bable. It is incredible that in the inter-tribal politics of Gaul they
should have habitually acted Avith unanimity otherwise, in the
;

event, for instance, of a war between the Aedui and the Sequani,
the Druids in one state or the other would have been ranged against
their fellow-countrymen. Fronde's assertion ^ that, So far as can '

be seen, the Druids were on the Roman side,' is, I suppose, based
upon the facts that Caesar's friend and ally, Diviciacus, was a Druid,
and that ConvictoHtavis, whose candidature for the office of Vergo-
bret Caesar supported, was the nominee of the Druids.'^ But one
swallow does not make a summer and even if the Aeduan Druids
;

1 Michelet {Hid. de France, i, G3,

ed. 1835) fancies tliat B. (/., vii, 1-2 show
that Druids i^layed an important part in initiating the great rebellion but,
—
as Fustel remarks {Hist, des inst. pol. de i'anc. France, la Gaule rom., p. 31,
;

n. 2), this whimsical notion is unsuj^ported by any evidence.

2 Eng. Hist. Rev., xviii,
1903, p. 336.
3 Vercingetorix,
pp. 107-11. " Tacitus, Hist., iv, 54.
5 Commentators generally forget that '
Ootuatus, one of the two desperadoes '

who sacked Cenabum, was a priest {giduater). fScc pp. 831-2.

" Caesar, a Sketch, ed. 1886, p. 24.
' F. do Coulangcs, indeed, says {Hist, des inst. pol. de Vane. France— la
528 THE DRUIDS
had been opposed to Caesar, I am not sure that he would have
thought worth while to accentuate their hostility and outrage
it
public opinion by deliberately setting at naught the cojistitution of
his principal allies.^ I conclude that in the rebellion, save in the
case of Convictolitavis,^ neither the Druids nor any individual Druid
exercised an influence which, at all events so far as Caesar knew,
sensibly affected the result.^
4. Fustel* remarks that Caesar does not say that the Druids
acted as judges in all suits, but only in almost all (de omnibus fere
controversiis) ^ and also that he does not say that their jurisdiction,
;

Gaule rom., that there is no evidence that the priests by whom

p. 27, n. 2)
Convictolitavis was elected were Druids but I take leave to say that there
:

is, and that of the most convincing kind. Caesar distinctly says {B. G., vi,
13, §§ 1-3) that in the whole of Celtican Gaul there were only two classes of
men who were held in any esteem, the knights {equites) and the Druids.
'
'

He mentions no other priests, except the Druids and he implies {ih., 13,
;

§§ 4-7, 10) that there were no other. The Aedui were in Celtican Gaul. The
priests who elected Convictolitavis were certainly important personages.
Is it to be supposed that the Druids, to whom Caesar ascribes such importance,
would have permitted any other priests, if there were any, to oust them ?
The only other Gallic word that denoted a priest was gutiiatros, which occurs
(see p. 832, infra) in Gallo-Roman inscriptions, and the Latin form of which
was gutuater. Gutuatri, as M. Salomon Reinach says [Orpheus, 1909, p. 179),
regulated local cults. I doubt whether it is possible to prove that in pre-
lloman times they were not Druids ; but if they were a distinct, they were
also an inferior order.
^
M. d'Arbois de Jubainville {Rev. arch., nouv. ser., xxxviii, 1879, p. 378)
asserts that Caesar avait triomphe des chevaliers, grace a I'appui du sacerdoce
'

(ju'il etait parvenu a detacher de la cause nationale '. There is simply no evi-
dence for this sweeping assertion except the fact that Diviciacus was on Caesar's
side ; and if the Druids were so powerful that their alleged support enabled
Caesar to triumph over the 'knights', why were they powerless to prevent
or even to put the least drag on the insurrection in 52 b. c. ?
^ The procedure of the Druids in the election of Convictolitavis is somewhat

obscure. Caesar says that he '

authorized Convictolitavis, who had been
appointed by the priests, in accordance with tribal custom in a period of inter-
regnum (or " when the magistrates failed to attend" [?]), to continue to hold
office'
{Convictolitavem, qui per sacerdotes tnore civitatis intermissis magistratibua
csset crealus, potestatem ohtinere iussit \^B. G., vii, 33, § 4J). Schneider {Cae-mr,
ii, 430) interprets intermissis magistrafibus in the sense that there was regu-

larly/ a period of interregnum after the expiration of the Vergobret's term

of office, in which the priests, acting as interreges, appointed his successor
but the words leave on my mind the impression that the right of appointment
did not ordinarily belong to the priests, and, moreover, per sacerdotes is different
from a sacerdotibus. M. Jullian {Hist, de la Gaule, ii, 48, 103) appears to hold
that the right of appointment belonged to the magistrates, or headmen, of the
various clans, and that in their default it was exercised by the priests. But
why should the magistrates have waived their right ? M. Jullian' s inter-
pretation of intermissis seems to me to conflict with Caesar's use of the word
(see Meusol's Lex. Caes., ii, 221-2). Intermissis magistratibus must, I think,
mean ' when the magistracy was (or remained) vacant ; but who the regular
'

electors were is doubtful.

^ M. Jullian in his Hist, de la Gaule, iii, 373-6, suggests that the prominent
part which the Carnutes played against Caesar not only in the great rebellion
but also in the two years that preceded it may have been due to the instigation
of the Druids, whose annual council was held in their territory {B. G., vi, 13, § 10).
Perhaps but how little wc know
—
; !

* Hist, des inst. pol. de Vane. France, la Gaule rom., pp. 19-20.
3 B. G., vi, 13,
§ 5. Cf. H. d'A. de Jubainville in Bev. celt., viii, 1887, p. 519,
 THE DRUIDS 529

as far as it extended, was obligatory, but rather leaves on our minds

the impression that litigants sought it voluntarily. M. G. Bloch^
argues that the suitors who appealed to Druids probably all belonged
to the upper class, who, having unlimited rights over their depen-
dants,2 doubtless decided their disputes. Moreover, the Aeduan
Vergobret had the power of life and death over his countrymen ',^
'

just as fathers had over their wives and children ^ and the chief ;

magistrate adjudicated on offences against the state,^ though the

punishment which he inflicted may have been sanctioned and
superintended by Druids. Fustel infers, moreover, from Caesar's
statement that every one submitted to the verdicts of the Druids,^
that they had no power of enforcing them. No power, though they
could and did excommunicate those who were refractory Besides, !

Caesar distinctly says that those who were excommunicated were

outside the pale of the law {neque his petentibus ius redditur)? Whether
the law here spoken of was that administered by the Druids them-
selves or by the secular authorities, matters nothing. Practically, if
Caesar was not misinformed, the Druids had the most irresistible
power of enforcing their judgements.
5. Desjardins ^ and others hold that the Druids and the oligarchical
nobles were leagued together to repress the democratic aspirations of
the populace but, as Fustel points out,^ there is no evidence what-
;

ever in support of this view. Nor, I may add, is there any evidence
that such aspirations existed,^^ at all events that any leader had arisen
to formulate and sustain them. As the knights, or many of them,
sent their sons to be educated by Druids, and as many Druids must
have belonged to noble families, it may be reasonably inferred that
the two orders had common interests and when an ambitious noble
;

endeavoured to raise the populace for his own purposes, the Druids
may have used their spiritual powers against him but all this is :

mere conjecture.

WAS THE REBELLION OF VERCINGETORIX

A DEMOCRATIC MOVEMENT ?

Albert Reville has written two very interesting and very ingenious
articles upon Vercingetorix in the Itevue des Deux Mondes for August
and September, 1877. One of his objects is to show that the rebellion
of which Vercingetorix was the leader was not merely a national

and Les premiers hahitanU de V Europe, ii, 1894, p. 375, and C. Jullian, Hist,
de la Gaule, ii, lOO-l. M. Jullian remarks (p. 101, n. 3) that la formation des '

cites [civitates] a du contribuer a affaiblir I'autorite judiciaire des druides'.

. . .

But were not many of the cites formed by the union, more or less loose, of
'
'

pagi before the late date to which M. Jullian assigns the rise of Druidism ?
1 Rev. internal, de V enseignement, Aout
1895, p. 151.
2 B. G., vi, 13, §§ 3, 5. 3 lb., i, 16, § 5. * lb., vi, 19, § 3.
« lb., i, 4
V, 56, § 3 ; vii, 4, § 1.
; « lb., vi, 13, § 10. ' lb.,
§§ 6-7.
8 Geogr. de la Gaule rom., ii, 529. Cf. C. Jullian, Hist, de la Gaule, 107.
* Hist, des inst. pol. de Vane. France, — Gaule rom.,
la p. 31, n. 1.
ii,

1" See pp. 529-41.

1093 Mm
530 ALLEGED SIGNS OF
movement for the overthrow of the Roman dominion, but also
a movement for the vindication of popular rights. Notre demo- '

cratie,' he says, remonte par ses origines premieres jusqu'an parti

'

egalitaire, impatient de Toligarchie, deja national, qui permit a

Vercingetorix de grouper un instant sous ses ordres les forces de la
Gaule entiere.' ^ Similarly, Mommsen, who asserts that most of the
Gallic attempts to throw off the Roman yoke were to an undue
'

extent the work of certain prominent nobles ', appears to think that
in 52 B.C. the situation was reversed. Describing how rapidly the
insurrectionary movement spread after Vercingetorix had declared
himself, he says, where the common council made any difficulty,
'

the multitude compelled it to join the movement.' ^ Desjardins takes

the same view. Of the nobles and the Druids he says that the re-
bellion of Vercingetorix was un mouvement national accompli sans
'

eux et malgre eux '. Indeed he appears to consider that the demo-
cratic movement had begun or was beginning when Caesar first set
foot in Gaul. He tries to prove that the state of society described by
Caesar in B. G., vi, 13, 15, in which the lower classes had no political
power, belonged to a period anterior to Caesar's arrival. Pendant
'

les huit annees qu'il passa dans les Gaules, il semble qu'une revolu-
tion fut, sinon accomplie, du moins en voie de s'accomplir, car cette
inferiorite politique du peuple s'etait deja profondement modifiee au
temps de Vercingetorix, que nous voyons en effet, avec I'aide de ses
clients, et malgre la noblesse arverne, malgre sa famille, appeler son
pays aux armes il n'est pas possible de se meprendre sur le double
. . .

caractere du soulevement qui se produisit alors, non pas seulement

—
en faveur de la patrie de la patrie mal defendue jusque-la par la
—
noblesse mais encore en faveur de la liberte.' In another passage,
alluding to the statement which Ambiorix, king of the Eburones, is
reported to have made regarding the popular rights by which his
own authority was limited suaque esse eius modi imperia ut non
minus haberet iuris in se multitudo quam ipse in muUitudinem Des- —
jardins says, il faut done admettre que, dans certaines cites de la
'

—
Gaule, exceptionnellement, peut-etre meme seulement chez les
nations qui cachaient et defendaient si bien leur liberte dans les
epaisses profondeurs de I'Ardenne, il existait, —
chez un peuple au
—
moins, avant la revolution politique accomplie par le heros arverne,
— une constitution admettant la classe inferieure au partage de
^
certains droits publics.'
Two questions, then, of considerable historical importance have to
be discussed. First, whether a democratic revolution was in progress
in Gaul, or was in process of inception, during the first six years of
Caesar's proconsulship in other words, whether Caesar's account
;

of the condition of the Gallic plebs is to be regarded as apphdng

Eev. des Deux Mondes, 3® per., xxiii, 1877, p. 75.

^

Hist, of Rome, v, 1894, pp. 76, 99 {Rr,m. Gesch., iii, 1889, pp. 280, 299).
2

^ Geogr. de la Gaule rom., ii, 539-43. I omit the arguments of F. Monnier

(Vercingetorix et Vindependance gauloise, 3rd ed., 1883, pp. 174, 282), based
upon a gross mistranslation of B. G., vii, 29, § 6 and 43, § 1, which I refuted in
the first edition (p. 537).
 DEMOCRACY IN GAUL 531

not to those six years but to an earlier period. Secondly, whether

such a revolution began during, and was stimulated by, the rebellion
of Vercingetorix. I shall quote and examine seriatim every single
passage in Caesar's narrative that bears upon either question.
(1) In i, 2-4 Caesar describes the origin of the Helvetian emigra-
tion. Orgetorix apud Ilelvetios longe nohilissimus et ditissimus con-
cocted a plan with his brother nobles {coniurationem nohilitatis fecit),
—
and persuaded the state (civitati yersuasit) which, if it does not
mean the government, must mean either the whole comnmnity,
whom an individual would not have had time to persuade, or those
—
leading men who could influence the community to emigrate en
masse. Orgetorix was chosen to superintend the necessary arrange-
ments. He undertook a diplomatic mission to Gaul, and secretly
formed a triumvirate with Dumnorix and Casticus, son of a former
king of the Sequani. The plan devised by Orgetorix was that the
three should seize the disused royal power in their respective states ;

that Orgetorix should help the other two to secure their thrones ;

and finally that the three should divide between them the supremacy
of Gaul. The Helve tii got information of the conspiracy, and sum-
moned Orgetorix to answer for his conduct. He appeared before his
judges with an army of clients, debtors, and slaves. Then comes
'
the passage which we are in search of. It tells us that the state '

was provoked to assert its authority by force of arms ; that the '

magistrates raised a posse from the country side and that Orgetorix
'
;

perished {cum civ it as oh eam rem incitata armis ius suum exsequi
conaretur, multitudinemque hominum ex agris magistratus cogerent,
Orgetorix mortuus est). Civitas here plainly means the state or the
leading men as represented by the magistrates, who take the initiative.
There is nothing in the three chapters that indicates anything like
a popular revolution, anything different from the state of society
depicted by Caesar, in which the plehs are without political power.
Even if civitas in the expression civitati persuasit means the whole '

community ', including the flehs, that does not prove that the plehs
had begun to enjoy political rights for of course a whole people
;

could not be coerced into leaving their country.

(2) Some passages in i, 17-20 may throw light upon the question.
Liscus, Vergobret of the Aedui, when he and his brother chiefs were
taxed by Caesar with neglecting to fulfil their engagement to supply
him with corn, replied that there were certain individuals who had
'

great influence with the masses '

{esse nonnullos quorum auctoritas
apud plehem plurimum valeat), which simply implies that the Aeduan
plehs, like the mass of the people in every nation, whether they enjoy
definite political power or not, had to be reckoned with, and that
clever leaders could Avork upon them for their own purposes. So in
chapter 18 we read that Dumnorix was extremely popular with the
'

masses from his open-handedness {magna apud plehem propter

'

liheralitatem gratia). In chapter 20, Diviciacus, begging Caesar not

to punish Dumnorix too severely for his treachery, pleads that
'
'
public opinion as well as fraternal affection had weight with him
{sese et amorefraterno et existimatione vulgi commoveri). But this only
M m2
532 ALLEGED SIGNS OF
proves that, even in a country governed by a powerful aristocracy, it
is dangerous to ride roughshod over pronounced popular sentiment
and this needs no proof.
(3) The next passage tells more in favour of Desjardins's theory,
though, strange to say, he has overlooked it Aulerci Ehurovices
:

Lexoviique senatu auo interfeclo, quod auctores belli esse nolehant}

&c. Who put the senate to death ? One would be inclined to say
that they fell victims to an outburst of popular fury and this is
;

how Mommsen ^ understands the passage. I do not think that such

an interpretation can be regarded as certain because there is no
;

proof that the senate comprised the whole of the equites and those ;

of that class who dissented from the policy of the senate may have
been the authors or the instigators of the massacre. Obviously the
actual murderers must have been comparatively few in number ;

and the natural conclusion is that they were hounded on by

influential men. But assuming that what Caesar describes was a
spontaneous outburst of popular fury, should we be warranted in
concluding that une revolution fut, sinon accomplie, du moins en
'

voie de s'accomplir ? '

(4) When the motley host of Viridovix had been duped by Sabinus's
emissary, they would not suffer Viridovix and the other leaders to
'

leave the assembly until they had agreed to let them arm and make
a dash for the camp {non prius Viridovicem reliquosque duces ex
'

concilio dimittunt quam ah Us sit concessum arma uti capiant et ad

castra contendant).^ This passage proves that Viridovix and his
officers were incapable of restraining their undisciplined followers ;

but it no more invalidates Caesar's general statement about the

political nullity of the Gallic populace than the attack which the
53rd Regiment made at the Kali Naddi in defiance of Sir Colin
Campbell^ proves that the British army in 1857 was administered
on democratic principles.
(5) The only remaining scrap of evidence is the well-known state-
ment which Caesar attributes to Ambiorix, sua esse eiusmodi
imperia ut non minus haheret iuris in se multitudo quam ipse in multi-
tudinem. But in the first place, Caesar can only have written his
account of Ambiorix's speech at second, or rather at third hand,
from the report of the report of the deputies who were sent by Sabinus
to confer with Ambiorix ^ and in the second place, supposing that
;

Ambiorix was correctly reported, is it not possible that, as his speech

was confessedly intended to deceive Sabinus and Cotta, this par-
ticular statement of his may have been intended to give an air of
plausibility to his arguments ? I will assume, however, that, among
the lies which Ambiorix told, this particular statement was true ;
that, among the Eburones, the multitude if, —
indeed, multitudo
necessarily means multitude ', and not simply the equites, who alone,
'

1 B. Q., iii, 17, § 3.

2 Horn. Gesch., iii, 1889, p. 263 {Hist, of Home, v, 1894, p. 55).
'
B. G.,iu, 18, §7. Seep. 92.
* Rice Holmes, Hist, of the Indian Mutiny, ed. 1904, p. 430.
^ B. G., V, 27, § 3 ; 37, § 9.
 —
DEMOCKACY IN GAUL 533

according to Caesar's express statement, were held in any account,

—
and their cUents did enjoy poUtical power.i How little, after all, the
statement proves Desjardins, who begins by affirming that a
!

popular revolution was, if not accomplished, at least in a fair way of

being accomplished, can find nothing more to say, after examining
his evidence, than this, —
that there existed, in one state at least,
before the rebellion of Vercingetorix, a constitution which admitted
the j)lebs to a share in political power. In one state out of more than
sixty Truly this is a lame and impotent conclusion
! !

I now proceed to examine the passages in the Seventh Book of the

Gallic War that bear upon the question whether a democratic revolu-
tion in Gaul began during, and was stimulated by, the rebellion of
Vercingetorix. (1) The first is in chapter 1. Indictis inter se prin-
cipes Galliae conciliis queruntur de Acconis morte ; posse hunc
. . .

casmri ad ipsos recidere demonstrant omnibus pollicitationibus ac . . .

praemiis deposcunt qui belli initium faciant, &c. In this passage, the
nobles (principes), of whom Desjardins says that the rebeUion took
place sans eux et malgre eux ', are represented by Caesar as having
'

^
been its authors !

^ That multitudo does not necessarily mean the multitude in the sense of '
'

'
the masses', is proved by a passage in vii, (53, § (>, which I shall examine in its
turn.
^ The meaning of the words prmceps and principaius, as used in the fUillic

War, has been the subject of a good deal of discussion. Deloche {Mem. pre-
stntes par divers savants a V Acad, des inscr., 2*^ ser., iv^ 1860, p. 308) thinks
that the term princeps, as used in such passages as indictis inter se principes
Galliae conciliis . . queruntur de Acconis morte (vii, 1, § 4) and (Dumnorix)
.

principes Galliae soUicitare coepit (v, 6, § 4), denoted a magistrate

. . . to ;

which M. d'Arbois dc Jubainville {Bev. celt., viii, 1887, p. 226, n. 1) replies that
the passage on which Deloche relies in pace nullus est communis magistratus,
sed principes regionum atque pagoriim inter sues ius dicunt controversiasque
—
minuunt (vi, 23, § 5) refers to Germany and not Gaul. But the truth is that
the word is used in several different senses by Caesar, as any one may convince
himself by studying Meusel's Lexicon Caesarianum, ii, 1196-1203 andaltbougli ;

some of the principes mentioned by Caesar were certainly magistrates, it is

doubtful whether the word, as such, bears that meaning, except in a very few
passages. When used absolutely, principes would, as a rule, be most safely
translated by 'leading men'. Deloche {Mem. presentes, &c., p. 308, n. 3),
admits that it sometimes has this meaning but he nevertheless insists that
;

princeps frequently denotes the holder of an office. He cites the passage

(vii, 39, § 2) in which Caesar describes the struggle between Convictolitavis
and Cotus for the principatus of the Aedui the passage in which we read ;

that Celtillus once held the principatus of Gaul {ib., 4, § I) and the passage ;

in which Caesar describes Sedulius as dux et princeps Lemovicum {ib., 88, § 4).
'
Since,' he argues, dux evidently denotes the military chief, princeps naturally
'

denotes the civil magistrate.' This I freely admit and I would say the same of ;

C. Valeria Domnotauro principe civitatis (vii, 65, § 2) and of Vertisco, principe

. . .

civitatis (viii, 12, § 4). Moreover, when we read that after the death of Indutio-
marus Cingetorigi . principatus atque imperium est traditum {\i, 8, ^ 9),
. .

we can hardly deny that the principatus transferred to Cingetorix was the first
magistracy. On the other hand, in vi, 22, § 2 the magistratus of the Germans
are differentiated from those who were only principes while in vii, 39, § 2 ;

{his [Eporcdorix and Viridomarus] cral iuier se dc principalu contentio) jjrinci-

patus is obviously vague, for Convictolitavis had already been appointed chief
magistrate, and the juoaning is simply that the two men were rivals for power'. '
534 ALLEGED SIGNS OF
(2) Li chapter 4 we read that Vercingetorix. on hearing that the
Carnutes had actually commenced the insurrection, convocatis suis
clientihus facile incendit.Cognito eius consilio, ad arma concurritur.
Prohibetur a Gobannitione patruo suo, reliquisque princijjibus, qui
hanc temptandam fortunam non existimabant ; expellitur ex
oppido Gergovia ; non desistit taynen, atque in agris habet delectum egen-
tium ac perditorum. Ilac coacta manu, quoscumque adit ex civitate ad
suam sententiam perducit magnisque coactis copiis, adversarios suos,
. . .

a quibus paullo ante erat eiectus, expellit ex civitate. Rex ab suis

appellatur. Dimittit quoque versus legationes .. .omnium consensu
ad eum defertur imperiurn. These last words cannot, except by a
most forced interpretation, refer to a popular election. They probably
mean that Vercingetorix was elected commander by a general council
of the chiefs of the insurgent states, or recognized as such by the
senate of each several state : they may also imply that this arrange-
ment was sanctioned by popular sentiment.^ The rest of the passage
contains nothing to warrant us in supposing that the populace were
asserting their rights against the nobles it only shows that Vercin-
:

getorix, at the head of his clients and his popular levies, banished from
Gergovia his brother chiefs, who objected to rebellion. Nor is there

Finally, passage in which Celtillus is mentioned tells directly against

tlie
Deloche ; one who has the most elementary knowledge of Gallic
for every
history knows that the principatus tolius Galliae was not a definite office. The
Se(|uani at one time held that 'principaius (vi, 12. §§ 4, (i), and at another
time the Aedui (i, 4:i, § 7). Caesar only means that Celtillus, as the most
powerful chief of the Arverni, exercised at one time a kind of loose and indefinite'
supremacy over Gaul. Again, C. Valerius Troucillus was certainly not the
chief civil authority of the Province, though Caesar {ib., 19, § 3) calls him
principem Galliae provinciae and Cingetorix, the Treveran, who is called
;

alterius principem faclionis (v, 50, § 3), was sim])ly tlie leader of the party
opposed to Indutiomarus, which does not prove tiiat he then held any civil
appointment.
G. Braumann {Die Principes der Gallier und Germanen bei Casnr und
Tacitus, reviewed in Bursian's Jahresbericht, Ixviii, 1808, p. 73, with which
cf. H. d'A. de Jubainville's Recherches sur Vorigine de la propriete fonciere, &c.,
1890, pp. 40-9) agrees with the opinion, now modified, which I expressed
in my first edition the principes, he holds, were simply the most prominent
t

among the nobiles or equites, and had no special office as such. With the
few exceptions which T have noted, I still hold that this, as far as we can
ascertain, is true but it is possible that the principes civitatum whom Caesar
;

occasionally mentions were, or included, the magistrates of their respective

states or pagi. Whether the word principes denoted their official character
is another question. Cf. C. Jullian, Hist, de la Gaule, ii, 39, n. 40, n. 1. ;

^ Desjardins {Geogr. de la Gaule rom., ii, 543) admits that, in this passage,
there is no allusion to a regular concilium but he goes on to say, il y eut
;
'

du moins acclamation populaire.' Very likely but popular acclamation does

;

not imply popular revolution. Desjardins believes that Strabo was thinking
of this passage when he said (iv, 4, § 3) that Gallic generals were elected by
popular suffrage. This may be true though it seems more likely that Strabo
;

was thinking of a passage in B. G., vii, 03, § 0, to which I shall presently refer.
At all events there is no better authority in Caesar for Strabo' s statement
and commentators have attached to it an importance which it does not deserve.
If such a law or custom had really existed, Caesar would certainly have men-
tioned it but, as a matter of fact, Strabo' s statement, if by to -rrXfjOos he meant
;

plcbs, is absolutely irreconcilable with what Caesar says about the condition
of the lower orders.
 — — .

DEMOCRACY IN GAUL 535

any evidence that the course of events in the country of the Arverni
had its counterpart in any other state.
(3) Chapter 14 relates that Vercingetorix, tot continuis incom-
modis Vellaunodimi, Cenahi, Novioduni acceptis suos ad concilium
convocat, and explained to them that thenceforward they must adopt
a guerrilla warfare. In the next chapter (15) we are told that this
plan was approved ofunium consensu. Caesar goes on to relate how
dcliberatur de Avarico in communi concilio, incendi placeat an
defendi. Procumbunt omnibus Gallis ad pedes Bituriges. . .

Datur petentibus venia, dissuadente primo Vercingetorige, post concedente

et precibus ipsorum et misericordia vulgi. Probably the council
mentioned in this chapter was identical with that mentioned in the
fourteenth. I will assume, however, that there were two. The
former concilium is obviously a council of war.^ Suos cannot possibly
mean all the troops it means the insurgent chiefs under the com-
:

mand of Vercingetorix. The same may be said of omnium consensu

and of omnibus Gallis. As for misericordia vulgi, if vulgi does not
simply mean the mass of the officers present at the council,^ the
sense must be that Vercingetorix was influenced by the general sym-
pathy which was felt for the Bituriges in fact that, like other com-;

manders, he could not ignore the pronounced sentiment of his army.

(4) Chapter 20 contains the famous speech in which Vercingetorix
defends himself from the charge of treason Vercingetorix cum ad :

suos rediisset, proditionis insimulatus regnum ilium Galliae

—
. . .

malle Caesaris concessu quam ipsorum habere benejicio tali modo

accusatus ad haec respondit, &c. and chapter 21 describes the
;

effect produced by his vindication: conclamat omnis multitudo^

. .summum esse Vercingetorigem ducetn

. statuunt,^ &c. The . . .

expression omnis multitudo may refer to the entire host, or as

many of them as could hear what Vercingetorix said. But I cannot see
that the fact of their having shouted their approval of their general's
speech, proves that a popular revolution was in progress. Similar
scenes must often have been enacted in Gallic history long before.
Nothing is proved except that the most popular and powerful noble
could not afford to disregard the wishes of his followers.
(5) Chapter 29 reports the speech which Vercingetorix made after
the fall of Avaricum. Concilio convocato, writes Caesar, consolatus
cohortatusque est. Desjardins himself calls this a concilium des '

officiers ' ^ and so it obviously was.

;

(6) Chapter 31 tells us that Vercingetorix tried to gain over the

^ Meusel bids me remember that concilium is never " a council of war",

'

the word for which is consilium \ But does not all depend upon what is to
be understood by a council of war ? Caesar calls Roman councils of war
'
'

consilia but Meusel himself rightly speaks of concilia principuyn

: it is self- ;

evident that when Vercingetorix had to discuss grave matters of state {B. G.,
vii, 14, § 1 77, § 1) he did not summon 80,000 men to the discussion, but only
;

his officers and surely a council assembled to consider military operations

;

may be called a council of war. M. Jullian {Vercingetorix, pp. 166, 169) agrees
with mc. Cf. Meusel' s Lex. Caes., i, 628-30.
^ Cf. C. Jullian, Vercimjctorix, p. 169.
^ Geogr. dc la Gaule rom., ii, 541.
536 ALLEGED SIGNS OF
dissentient tribes, and bribed their chiefs or leading men to juin liini,i

and that the king of the Nitiobroges came over to his side.
(7) Li chapter 32 Caesar relates that when Cotus and Convicto-
litavis were contending for the office of Vergobret, the people were '

all up in arms the council and the commonalty were divided

; and ;

the rivals were each supported by their own retainers,' ^ who were
instruments in their hands.
(8) Chapter 36 shows Vercingetorix at Gergovia in daily consulta-
tion with the frincifes civitatium '.
'

(9) In chapter 37 we find Convictolitavis trying, in conjunction

with others, who belonged to a very illustrious family,' ^ to get up
'

a rebellion.
(10) In chapter 42 the Aedui rob and nmrder Roman citizens ;

and Convictolitavis, still taking the lead, adds fuel to the flame
'

and hounds on the masses to frenzy.' ^

(11) From chapter 55 we learn that Convictolitavis has gained
over a large proportion of the [Aeduan] council '. Eporedorix, a
' '

young Aeduan of noble birth and commanding influence in his own

country,' ^ and Viridomarus, whom Caesar had raised to the highest '

dignity ',^ now join the rebellion. Among the Aedui, at all events,
the flebs are merely instruments in the hands of the nobles.
(12) Chapter 57 says that, when the Parisii and their allies assem-
bled to oppose Labienus, sujnma imj>erii traditur Camulogeno. It
is conceivable perhaj^s that this may mean that Camulogenus was
elected by a plebiscite but it is, in my judgement, certain that the
;

election rested with the chiefs or the equites. I am aware that Strabo
says, in a passage which has been quoted ad nauseam, that Gallic
generals used to be elected by the multitude
'
(eis iroXeiJiov ct?
'

VTTO Tov irXyjOovi aTre^ecKvuTo crTparrjyos) ' but I am strongly

',

inclined to believe that Strabo simply based this statement upon

the following passage in B. G., vii, 63, § 6 Petunt a Vercin-
:

getorige Haedui ut ad se veniat rationesque belli gerendi communicet.

Re impetrala contendunt ut ipsis summa imperii tradatur ; et re in
controversiam deducta totius Galliae concilium Bibracte indicitur.
Conveniunt undiquefrequentes.Multitudinis suffragiis res
permittitur : ad unum omnes V ercingetorigem probant im-
peratorem. Generally, when Caesar speaks of totius Galliae concilium,
he means a council of chiefs ^ and undoubtedly that is what he
;

means here.^ For it is simply absurd to suppose that concilium denotes

the whole aggregate of tribal contingents which composed the army of
Vercingetorix and when Caesar goes on to say that the Remi, the
;

^ animo laborabat ut reliquas civitates adiungeret, atque earum principes

donis pollicitationibusque adliciebat,

^ civitatem esse omnem in armis, divisum senatum, divisum populum, suas

cuiusque eorum clientelas.

^ amplissima familia nati.

"
adiuvat rem proclinatam Convictolitavis plebemque ad furorem impellit.
° siunnio loco natus adiilescens ct siimniac domi poteutiae.

" ad sununani dignitalein. ' iv,

4, § 3.
« B. ti., i. 30.
§ 4;V, 24, § 1 vi, 3, § 4
; 44, § 1
; vii, 03, § 5
; 75, § 1. ;

" Here also Mcusel disagrees with me.

 — —
DEMOCKACY IN GAUL 537

Lingoncs, and the Treveri kept aluuf from the council, he evidently
implies that it consisted of delegates from the various states. Is it
credible that those delegates included representatives of the masses,
Avho could not have borne the expenses of the journey, and who, we are
distinctly told, nuUi adhibetur consilio ? At the most, the ynultitudo
can only have comprised the equites who were present and the clients
who may have accompanied them. [I find that Desjardins regards
this co7icilium also as composed of chefs '.^] '

(13) In chapter 75 the principes unmistakably take the lead :

Dum haec ad Alesiam geruntur, Galli concilio frincipum indicto

non omnes qui arma ferre possent, ut censuit Vercingetorix, convo-
candos statuunt, sed,^ &c.
(14) In chapter 76 we read haec in Haeduorum finibus :

recensebantur numerusque inibatur. Praefecti constituebantur.

Commio siimma imperii traditur. His delecti ex civitatibus
. . .

attribuuntur quorum consilio bellum administraretur.^ Considering

that the number of troops assembled in the country of the Aedui
exceeded 250,000, it is clear that both the officers and the civil
delegates were chosen, not by a plebiscite but by the chiefs.
The above list comprises every single passage in Caesar's Seventh
Commentary that bears upon the question at issue. That, with all
these passages before him, Desjardins should have asserted that the
insurrection of 52 B. c. was un mouvement national accompli sans
—the nobles—
'

eux '
malgre eux,' only shows how the judgement
' et
even of a man may be warped by prejudice. A
of vast learning
national movement the insurrection assuredly was but, so far from ;

having been accomplished sans et malgre the nobles, it could '

'

never have been begun, much less accomplished, without them. It

was the nobles of Gaul who planned it. It was the Carnutian nobles
who undertook to strike the first blow. It was an Arvernian noble
who thenceforth became the life and soul of the movement. At
Gergovia he regularly consulted with all the principes in his force.
When he had proved his natural right to command, the majority of
the Aeduan nobles joined him and their action encouraged a multi- ;

tude of waverers. Finally, when he was in extremest peril, the

nobles of all Gaul raised and organized the host who marched to
rescue him. Yet we are told that the insurrection was accomplished
'
sans eux et malgre eux The passages which I have quoted prove
'
!

that it w^as accomplished malgre the Arvernian nobles, all except

' '
—
^Oeogr, de la Gaule rom., ii, 542. Judge then of the amazement with which
I have read this passage in his book (p. 691) '
Chefs et soldats s'y rendirent en
:

foule, et, par un vote populaire —

on pourrait meme dire par le suffrage universel
— le commandement fut confirme a Vercingetorix.' Evidently the distin-
guished geographer did not know his own mind.
^ While this was going on at Alesia, the Gauls convened a council of their
'

leading men, who decided not to adopt Vercingetorix' s plan of assembling all
who could bear arms, but,' &c.
^ The levies (destined for the relief of Alesia) were reviewed and numbered
'
' '

in the country of the Acdui, and their officers ajipointed. The chief command
was entrusted to Conuuius. Delegates from the various tribes were associated
. . .

witli them, in accordance witli whose advice they were to conduct the cam-
paign.'
538 ALLEGED SIGNS OF
Vcrciiigetoi'ix, his cousin Vercassivellaunus, and Critognatus but ;

that does not greatly support the thesis of Desjardins and Reville.
—
There is nothing which proves nothing which even renders it in the
—
smallest measure probable that a democratic revolution had begun.
The statements which Caesar makes about the condition of the
plebs are so emphatic and so precise that I cannot see what right
any commentator has to modify them without the very strongest
evidence In onmi Gallia eorum hominum qui aliquo sunt numero
:

atque honore genera sunt duo ; nam flehes paene servorum habetur
loco, quae nihil audet per se,nulli adhibetur consilio. Plerique, cum
aut acre alieno aut magnitudine tributorum aut iftiuria potentiorum
prefiiuntur, sese in servitutem dicant nobilibus ; in hos eadem omnia
sunt iura quae dominis in servos. Sed de his duobus generibus alterum
est druidum, alterum equitum} We are told that Caesar's words
refer to a period anterior to his arrival in Gaul. This cannot be
proved and it is intrinsically most improbable. It is clear at all
;

events that he intended to describe the Gauls of his own time.

Partly no doubt he got his information from Gauls whom he trusted,
like Diviciacus partly, keen observer as he was, he saw and judged
:

for himself. I cannot see what motives his informants could have
had for trying to mislead him still less can I see how, after he had
:

passed eight seasons in Gaul and had enjoyed ample opportunities

for observation, he should have allowed himself to be misled. It is
possible, though it is not proved, that, among the Eburones, the
power of the king was limited by the power of the people. But it
would be very bad logic to generalize from a doubtful particular, in
defiance of the plain statement of our best authority. Still, it is one
thing to deny that the Gallic plehs had begun, in Caesar's time, to
acquire definite political power quite another thing to deny that
;

they had the power of making their wishes felt. It may be that,
as Sir John Rhys supposes, the common people were collectively
'

beginning to acquire influence, and already here and there to

understand their own power, though they had not yet taken the
initiative.' "^
Nor would
deny that, if there was a democratic
I
tendency '
became stronger during the rebellion of
in the air ', it
Vercingetorix I simply deny that there is any sufficient evidence
:

to decide the question. And when a great historian, whose word

with the majority of his readers is law, says of the rebellion of Vercin-
getorix that, where the common council made any difficulty, the
'

multitude compelled it to join the movement,' ^ it is necessary to say

that for such a statement there is no evidence at all.'*
1 Everywhere in Gaul there are two classes only who are of any account or
'

have any distinction for the masses are looked upon almost as slaves, never
;

venture to act on their own initiative, and are not admitted to any council.
Generally when crushed by debt or heavy taxation, or ill treated by powerful
individuals, they bind themselves to serve men of rank, who exercise over
them all the rights that masters have over their slaves. One of the two classes
consists of the Druids, the other of the Knights.' B. G., vi, 13, §§ 1-3,
''
Celtic Britain, 3rd cd., 1904. pp. GO-1.
^ Mommscn, Hist,
of Home, v, 1894, p. 70 (Horn. Gesch., iii, 1889, p. '280).
^ Since I wrote the rough draft of this article I have come across a paper bj'
 DEMOCRACY IN GAUL 539

I have lately been reading the chapters in which Fustel de Cou-

langes describes the political condition of Gaul in Caesar's time. On
the whole, he appears to agree with Reville and Desjardins. La '

question,' he says, qui divisait le plus la Gaule, a cette epoque,

'

etait celle de la democratic.' ^ Chaque fois,' he insists, qu'un

' '

peuple est vaincu, nous voyons les principaux personnages de ce

peuple se presenter devant Cesar, I'assurer qu'ils ont combattu
malgre eux, et rejeter la responsabiHte de la guerre sur " la multi-
tude " ;
2 and in support of this assertion he refers to B. 6r., ii, 13-4,
'

V, 27, vi, 13, and vii, 43. He remarks further that Caesar almost
always spoke with contempt of the Gallic forces which opposed him,
as composed of the dregs of the population and he holds that la ;
'

monarchic democratique 'of Vercingetorix was viewed with distrust

and even hatred by the aristocracy, and that he counted so little
on willing obedience that he required all the GalHc states to give him
hostages. II ne regnait qu'a force de se faire craindre.' ^
'

Let me examine these arguments. The first is based upon a serious

exaggeration. It is not true that every time a Gallic people was
conquered the leading men of the state assured Caesar that the '

multitude had set at naught their authority and insisted upon

'

making war. The Bellovaci did so in the last campaign, and the
Aedui assured Caesar that the outrages which followed the defec-
tion of Litaviccus had been perpetrated without the sanction of
the government but Caesar refused to accept the excuses of the
:

Bellovaci and only pretended to accept the excuses of the Aedui,

and I see no reason why we should be more credulous while in ;

both cases it is clear that the multitude was a mere tool in the
' '

hands of leading men. What Fustel represents as having happened

on every occasion did not happen in the Belgic war of 57 b. c. or in ;

the war with the maritime states in the following year or, excepting ;

the case of Ambiorix and the Eburones, in the campaigns of 54 and

53 B. c. or, except in the cases of the Aedui and the Bellovaci, in the
;

great rebellion or in the final campaign.

Caesar says that the army of Viridovix was reinforced by a rabble ;

that the army of Indutiomarus was reinforced by exiles and con- '

demned criminals ', who most probably had belonged to the upper
classes and that the nucleus of Vercingetorix's Arvernian contingent
;

consisted fartially of needy and desperate men '. Hirtius tells us

'

that the host of Drappes and Lucterius was reinforced by broken '

men ', emancipated slaves, exiles, and bandits. That is all. On the
other hand, the huge Belgic host, in the second year of the war, was
composed of regular levies the host of the Veneti and their allies
;

included all the foremost men of each state the army of Vercin- ;

getorix was almost entirely composed of regular levies, raised, certainly

in part and perhaps altogether, by the principes of the various states ;

and the huge host which marched to the relief of Vercingetorix

M. d'Arboisde Jubainvillc {Bev. celt., vii, 1886, p. 9), which supports the view
for whicli 1 am contending.
* Hist, des insf. pol. de Vane. Franze, —
la Guide rom., ]). 43.
- lb., pp. 54-0. lb., p. 58. =J
540 ALLEGED SIGNS OF
was, we are expressly told, similarly conqwsed and similarly raised.^
Moreover, the fact that a small fraction of the vast multitude
of Gauls who fought against Caesar consisted of the dregs of the
])opulation in no way proves that democracy was in the ascendant
it only proves that the leaders enhsted every man whom they

could get.
Finally, if Vercingetorix required hostages from the states which
joined the insurrection of 52 B.C., he only followed an established
custom.2 His monarchy may have been democratic in the sense that
he was the idol of the populace and he may have been distrusted ;

by those aristocrats who supported ohgarchical institutions. But

that does not prove that democracy, properly so called, was even in
process of inception.
Fustel refers to various other passages, which I have discussed, and
to one which I overlooked, —
the passage from which we learn that
Indutiomarus, the Treveran, excused himself to Caesar for not haviiig
repaired to his camp by the plea that he had desired to keep the '

tribe loyal ', as there was a danger that if all the men of rank '

abandoned them, the masses in their ignorance might fall away \^

Indutiomarus was confessedly making a false excuse but, accepting ;

his statement, where is the evidence that a democratic movement was

afoot ? The ignorant populace of any country occupied by an in-
vader, even if they were as destitute of political power as the Russian
serfs under Nicholas I, might readily fall to committing excesses if
they were left to themselves.
I have considered Fustel's arguments with the attention which
every utterance of his deserves but I cannot see that they lend
;

any support to the theory of MM. Reville and Desjardins. After all,
a good deal depends upon the meaning which one attaches to the
Avord democracy '. If a state in which a rich adventurer with a glib
'

tongue can ingratiate himself with the populace and hire their bows
and spears and thereby exalt himself to power is governed by demo-
cracy, then democracy flourished in Gaul and flourished in Greece
also long before the constitution of Cleisthenes or even of Solon.
But, for all that I can see, such a democracy is not inconsistent with
the state of society described by Caesar, in which the masses are '

regarded almost as slaves, never venture to act on their own initiative,

and are not admitted to any council '. And if there is any incon-
sistency in Caesar's narrative, if among the Eburones, say, the
plehs really did enjoy political power, the inconsistency is easily
explained. Caesar did not profess to be a constitutional historian.
He had neither the time nor the inclination, if he had the knowledge,
to modify his general statements.
M. Jullian's view differs somewhat from my own (for he attaches

1 B. a, ii, 4, §§ 4-10 ; iii, 16, § 2 ; 17, §§ 2-4 ; v, 55, § 3 ; vii, 4, §§ 3, 7 ; 31,

§4; 75, U; viii,30, §1.
- Cf. B.
G^., vii, 2, §2.

inter se noii po.ssint, &c.

—
principcs, quoniam in praesentia obsidibus cavcrc
. . .

^ quo fac'ilius civitatcni in oUicio contincrot, ne omuis nobilitaiis disceseu

plcbs propter imprudentiam laberetur. B. G., v, 3, § 6.

 DEMOCRACY TN GAUL 541

more importance than I do to the passage in Strabo), but not very

seriously. While he emphasizes the toute-puissance des grands
'

dans la cite '} he holds, quoting B. G., iii, 17, § 3, and vii, 32, § 5,
that the aristocracies degeneraient sans cesse, sous la puissance des
'

-
vieilles habitudes militaires, en democraties bruyantes et brutales '
;

and that 'lorsque Thomme se dctachait de sa demeure, qu'il marchait

et combattait pour tuer ou mourir, il recouvrait quelques-uns de ses
droits naturels, il reprenait une part de sa liberte in fact that in
' ;
'^

time of war c'etait une democratic tumultuaire qui rempla9ait le

'

gouvernement normal par le senat et les princes de la cite '."* The

word democratic is, I think, misleading.
' '

1 Hist, de la Gaule, ii, 79-83. " 76., p. 51.

3 Jb., p. 53. * lb., p. 203.
 SECTION V,— RELATING TO THE
NARRATIVE OF CHAPTER I
THE CELTIC INVASION OF ITALY
I.The date of the Celtic invasion of Italy has been the subject of
much controversy. According to Livy,i it was contemporaneous with
the foundation of Massilia, which took place in 606 b. c. Mommsen ^
rejects this date. The association,' he argues, of the migration of
' '

Bellovesus with the founding of Massilia, by which the former is

chronologically fixed down to the middle of the second century of
Rome, undoubtedly belongs, not to the native legend which of course
specified no dates, but to later chronological research and it deserves ;

no credit. Isolated incursions may have taken place at a very

. . .

early period but the great overflowing of Northern Italy by the

;

Celts cannot be placed before the age of the decay of the Etruscan
power, that is, not before the second half of the third century of
the city.'
Desjardins,^ on the other hand, argues that even during the con-
tinuance of the Etruscan power the Celts may have been established
in the country between the Alps and the Po for a century and a half
before they invaded the country of the Senones, from which they
started on their march against Rome. He remarks further that
Livy * mentions four distinct Celtic invasions of Italy, namely those
of Bellovesus, of Elitovius, of ih.^ Boi and the Lingones, and of the
Senones and he maintains that three-quarters of a century,
;

the period which elapsed between the foundation of Massilia and the
—
64th Olympiad (527-524) the date assigned by Dionysius of
Halicarnassus ^ for the fall of the Etruscan power is not too much —
to allow for these successive immigrations.
M. d'Arbois de Jubainville ^ declines to accept Desjardins's reason-
ing. He argues that the date which Livy assigns to the invasion
is irreconcilable with the fact that the Etruscan hegemony in the
country north of the Po was, according to Polybius,'^ contemporaneous
with their supremacy in Campania, which ended in the last quarter
of the fifth century B. c. and he rightly observes that the passage in
;

Dionysius to which Desjardins refers only means that the fall of the
Etruscan power was later than the 64th Olympiad. The Gauls, he

1
V, 34.
- Hist, of Rome, i, 1894, p. 423, n. 1 {Rim. Gescli., i, 1888, p. 327, n. **).
^ Oeogr. de la Gaule rom., ii, 68, n. 1, 203-4.
* V, 35, §§ 1-3. 5 vii^ 3.
6 Rev. celt., iii, 1876-8, p. 471. ' ii, 17, §§ 1-3.
 THE CELTIC INVASION OF ITALY 543

maintains,! conquered Northern Italy from the Etruscans about

396 B. 0. for the most ancient recorded event of the conquest was
;

the capture of Melpum, which took place on the day on which the
Romans captured Veii.- Polybius^ tell us that the Etruscans were
expelled from Lombardy only a short time before the Gauls captured
Rome. According to Appian,* the Celts invaded Italy in the 97th
Olympiad (392-389 b. c.) and according to Diodorus Siculus,^ in
;

the second year of the 98th (387 B. c). Both indeed are wrong as
to the exact date for they suppose that the first operation of the
;

war was to attack Clusium, and the capture of Melpum, which they
overlook, belonged to the year 396 b. c. but Dionysius of Hali- :

carnassus ^ also makes the period which elapsed between the invasion
of Italy and the capture of Rome very short.'
Alexandre Bertrand,^ who substantially agrees with M. d'Arbois,
infers from a passage in which Livy ^ describes the invaders who
captured Clusium as 'an unfamiliar and unheard-of enemy' (invisitato
atque inaudito hoste) that the invasion took place only a short time
before the capture of Melpum and the battle of the Allia. But
Bertrand did not, perhaps, consider Livy's narrative as a whole.
In this matter Livy handled his authorities in a way with which
his more critical readers are painfully familiar. He was perfectly
aware !^ of the chronology which was adopted by Diodorus, and
afterwards by Appian he took pains to explain that he preferred
:

another chronology, which made the first Gallic invasion of Italy

two centuries earlier ^i yet, having, from impatience of drudgery,
;

failed to think the problem out, he inadvertently committed himself

in more than one passage to the chronology of Diodorus.^^ The
invaders whom he described as unfamiliar and unheard-of '
the '
—
captors of Clusium and victors of the Allia were on his theory —
merely the last of four bands whose incursions had been spread over
two centuries.
It appears to me that the data are insufficient. If we can trust
our authorities at all, the invasion which preceded the capture of
Melpum occurred shortly before 396 b. c. M. d'Arbois supposes that
the chronology which Livy preferred was invented by his con-
temporary, Timagenes i^ but are we to suppose that Livy's authority
;

also concocted the details of the earlier invasions which Livy

describes ? Unless we are to follow Bertrand and reject three-fourths
of Livy's account altogether, we must admit that the invasion which
^ Les premiers habitants de V Europe, i, 1889, p. 166.
2 See Plinv, Nat. Hist., iii, 17 (21), § 125.
^ ii,17, § 3 ; 18, § 1. 4
Didot), pp. 25-6.
Celtica, 2, § 1 (ed.
5 xiv, 107, § 1 ; HI, § 1 ; 113, §§ 1-4.
« xiii, 10-2.
' See H. d'A. de Jubainville's Les premiers habitants de VEurope, ii, 1894,
pp. 297-9, and Bev. celt., xx, 1899, pp. 376-80.
* Les Geltes dans les vallees du Po et du Danube,
1894, pp. 20-8.
8 V, 37, § 2. Cf. 35, § 6.
^» V, 33, §§ 1-4. " lb., §§ 5-6. 12
/&., 35, § 4 37, § 2. ;

" Otto Hirschfeld rejects M. d'Arbois's theory and gives reasons for believ-
ing that Livy relied upon a lost geographical work of Cornelius Nepos.
See M. Salomon Reinach's article in Les Geltes dans les vallees du Po, &c
pp. 205-12.
544 THE CELTIC INVASION OF ITALY
was immediately followed by the capture of Melpum had been
preceded by others, the dates of which it is useless to attempt to fix ;

and this assumption, which is intrinsically not improbable, is perhaps

supported by Polybius,^ who seems to imply that the Celts by whom
the Etruscans were expelled from the valley of the Po had been
for some time their neighbours, and that the invasion of the Boi,
Lingones, and Senones was subsequent to that of the Insubres and
Cenomani.2
II. Bertrand rejects in toto Livy's statement, that the Gallic
invaders of Italy came from Transalpine Gaul. He argues (1) that
it may be inferred from Polybius ^ that they came from the country
north of the Alps, that is to say from the valleys of the Inn and the
—
Danube (2) that of the six tribes the Aedui, Ambarri, Arverni,
;

Aulerci, Bituriges, and Carnutes —

which, according to Livy, emi-
grated under Bellovesus from Gaul, not one finds mention in the
histories of Polybius or in any history of Cisalpine Gaul * (3) that, ;

according to Livy, Gallia Celtica was occupied about 600 B. c. by

the same tribes which occupied it in the time of Caesar, which is not
credible (4) that the population of Gaul could not, as Livy says,
;

have been so great as to necessitate emigration and (5) that the

;

king of the Bituriges could not have been supreme over the whole of
Gallia Celtica and certainly could not have induced 600,000 of his
subjects to quit their fertile country.^
These reasons do not appear to me sufficient to warrant the absolute
rejection of the legend reported by Livy. It is not necessarily to be
inferred from Polybius that the invaders came from the country north
of the Alps. In the passages to which Bertrand refers he states that
the slopes of the Alps on the north towards the Rhone (ctti t6v
' '

'VoSavov Kol ra? apKTovs) are inhabited by Gauls called Transalpine \^

'

and he describes the whole course of the Rhone, and particularly that
part of its course which is in the Valais.'' These passages lend no sup-
port to Bertrand's theory; for Polybius knew perfectly well that there
were Transalpine Gauls in Gaul properly so called as well as in the
Valais. I cannot see anything absurd in the notion that Gallia
Celtica should have been occupied, at the time of the Gallic invasion
of Italy, by some of the tribes which occupied it in the time of

1 ii, 17.
'^
M. JulHan {Hist, de la Gaule, i, 281, n. 2) would explain the discrepancy
between the two chronological systems thus :the invasion of Bellovesus was
contemporary with a threatened attack by the Salyes on Massilia (cf. Livy,
V, 34, §§ 7-8) and it was supposed that the threat was a consequence of the
:

landing of the Phocaeans who founded the town, and that the Celts had assisted
them to make their settlement. M. Jullian also thinks it probable that the
invasion was synchronized with the foundation of Massilia owing to the obscu-
rity of some Greek text, which mentioned the Phocaean emigrants as the founders,
and was taken to mean that they were attacked when they were about to
lay the first stone.
« ii, 15,
§ 8 ; iii, 47, §§ 2-3.
* The Aulerci are mentioned by Polybius under the name of Cenomani.
^ Bev. d'anthr., ii, 1873,
pp. 430, C43, n. 1 ; Les Celtes dans Its valUes du
P6 et du Dayiuhe, pp. 19-27.
« ii, 15, § 8. ' iii, 47, § 2.
 THE CELTIC INVASION OF ITALY 545

Caesar ; and we know that Pytheas in the latter half of the fourth
century b. c. found the Osismi in the same peninsula which they
occupied in 56 b. c. It must be remembered that even in the time of
Caesar large parts of Gaul were overgrown by dense forests ^ and ;

if the Helvetii, the Usipetes and Tencteri, the Cimbri and the Teutoni

emigrated from their respective abodes, why should not Gallic tribes
have emigrated from Gaul ? If Bituitus, king of the Arverni, was
overlord of half Transalpine Gaul in the second century b. c, it is
hard to see why the king of the Bituriges should not have ruled over
as wide an area and Bertrand appears to forget that Caesar,^ as
;

well as Livy, reports that Gauls had emigrated, in bygone times,

from Transalpine Gaul, properly so called, and that he also assigns
as a reason for their emigrations over-population. It may, however,
be admitted that the argument which Bertrand bases upon the
absence of any mention of the Aedui, Ambarri, Arverni, Bituriges,
and Carnutes in the later history of Cisalpine Gaul does militate
against the complete acceptance of Livy's account but M. Jullian^
;

suggests that Insuhres may have been adopted as a federative name

by the various tribes whom Livy mentions.
Professor Niese,* however, attempts to reinforce the arguments of
Bertrand. He remarks that among the allies of the Insubres and the
Boi at the battle of Telamon were the Taurisci^ and the Gaesatae, who,
according to Polybius,^ lived 'about the Alps and on the Rhone' (Kara
ra? "AAttcis kol tov 'FoSavov TroTafxov). This, says Niese, points to the
conclusion that they came from the north of the Alps and the valley of
the Danube. For Polybius^ says that the Rhone rises north-west
of the Adriatic, on the northern slopes of the Alps, and flowing south-
westward, discharges itself into the Sardinian sea (6 Se 'FoSavos ex^i ra?
fiev TT^yyas vTrep tov ASptaTLKov [xv^ov Trpos rY]v kcnripav vevovcra?, ej/ rot?
aTro/cAiVovcrt /xepeortrwv AA-Trewi/ ws Trpos ras apKTOv;' p^l Se Trpos ra?
^et/xepivas 8i;(rei9, CK^aAXet 8' cl<s to %ap8(oov TreAayo?). In other
words, says Niese, the Rhone, according to Polybius, did not rise
in the Alps, but the Alps lay between the valley of the Rhone and
the plain of Lombardy. It is easy, he continues, to see how this
false conception arose in South-Eastern Gaul the Rhone does
:

really follow the Alpine chain and Polybius assumed that moun-
;

tains and river were in close proximity throughout. When he says ^

that Transalpine Gauls inhabited the parts of the Alps on the north '

towards the Rhone (cTrt tov 'FoSavov koI ra? apKTov^), he means,
'

in the language of reality, the country north of the Alps. In fact,

the relations of the Italian Celts were always with the north they :

received no help from the Gauls of Southern Gaul.

Now the Taurisci inhabited the southern slopes of the Alps, over-
looking the Cisalpine plain :how Polybius's vague geographical
1 L. F. A. Maury, Hist, des grandes forets de la Gaule et de Vancienne France,
1850, p. 166.
2 B. 3 jji^i ^g ;^ g^ui^^
0., vi, 24, § 1. i^ 291, n. 6.

f. deutsches AUerthum, xlii, 1898, p. 146.

* Zeitschr.

5 Polybius, ii,
28, § 4.
« ii, 22, » iii, 47, § 2. « ii, 15,
§ 1. § 8.
1093 N n
546 THE CELTIC INVASION OF ITALY
statement about the Gaesatae points to the ronehision that they^
came from the valley of the Danube, is to me incomprehensible ;

and, supposing that the Italian Celts were assisted by Danubian

mercenaries long after they themselves appeared in Italy, how does
that affect the question of their own origin ? Polybius's description
of the course of the Rhone has been said to ruin his credit as a
geographer but is it, after all, so very faulty ? The Rhone does
;

rise north-west of the Adriatic its source is north of that section

:

of the Alpine chain which was known to the Romans and its ;

general direction, so far as it can be indicated in, a few words, may

be loosely called south-westward. If Polybius (evidently referring to
the tribes of the Valais whom Galba encountered) ^ said that Trans-
alpine Gauls inhabited the slopes of the Alps on the north towards '

the Rhone ', he does not say that they were the invaders of Italy
and, I repeat, he knew perfectly well that other Gauls lived in Trans-
alpine Gaul properly so called.
Desjardins, whose views on the question are conservative, may
possibly be right when he suggests that the invaders may have come
from the valley of the Danube as well as from Transalpine Gaul.^

THE CIMBRI AND TEUTONI

Where did the Teutoni and the Cimbri respectively dwell imme-
diately before they set out on the wanderings which first brought them
into contact with the Romans ? When did the Teutoni first join the
Cimbri ? To what races did the Teutoni and the Cimbri respectively
belong ? All these questions have provoked much controversy.
I. The Cimbri, according to the common opinion, came from

Jutland, or at all events from the peninsula formed by Jutland,

Schleswig, and Holstein. This view is based upon the evidence of
Augustus,* Strabo,^ Pliny,^ Pomponius Mela,'' Tacitus,^ Florus,^ and
Ptolemy 10 but of these writers Strabo was the only one who affirmed
;

that the Cimbri still occupied the lands from which the emigrants
whom Marius encountered had come. Strabo vaguely located them
in Northern Germany, between the mouths of the Rhine and the
Elbe ;
^^ Tacitus, in a remote corner of Germany, bordering on the
'

^ Cf. A. Bertrand and S. Reinach, Les Celtes dans les vallees du P6 et du

Danube, pp. 142, 144.
- B. G., iii, 1-6. ^ Geogr. de la Gaule rom., ii, 207.
* Res gestae divi Augusti, ed. Th. Mommsen, 1883, 5, 11, 14-8 (p. Ixxxxii).
^ vii, 2, § 1.
« Nat. Hist., ii, 16 (67), § 167 ; iv, 13 (27), §§ 96-7.
' iii, 3, § 32. « Germ., 37.

'

1^
iii,3, §1.
Mommsen {Res gestae divi Augusti, 1883, p.
...
103) thinks that Strabo misplaced
'" G^eo^r., ii, 11, §7.

the Cimbri between the mouths of the Rhine and the Elbe because, when they
sent their embassy to Augustus, the Romans had not yet crossed the Elbe.
It is clear, he adds, from the monument of Ancyra that the Cimbri lived
— —
beyond that is, east of the Elbe ; for at that time Roman dominion was
limited by the river.
 THE CIMBRI AND TEUTONI 547

Ocean '
Pliny called Jutland the Cimbric peninsula
:
'
Mela and '
;

Ptolemy imply that this was their abode and Florus says
; that
the host which overran Germany, Gaul, and Spain came from the '

uttermost parts of Gallia ', by which he meant Northern Germany.

Florus states, further, that they were compelled to abandon their
country by a flood while Appian^ attributes their emigration to an
;

earthquake, which might have caused a tidal wave. We learn from

the monument of Ancyra ^ that the Cimbri and several other German
tribes sent envoys to Augustus, soliciting his friendship and that of
the Roman People, and that a Roman fleet sailed eastward from the
mouth of the Rhine to the country of the Cimbri while Strabo adds ;

that the Cimbri offered Augustus one of their sacred vessels and
implored pardon for the outrages of their forefathers. All this
evidence, however, is set aside by Karl Miillenhoff.^ Remarking
that Posidonius ^ rejected the story of the flood as absurd, and that,
according to Timagenes,^ the Druids taught that some of the Gauls
had been forced by an inundation to leave their homes beyond the
Rhine, he conjectured that when the Cimbri appeared in 113 B.C.
'
they were supposed to be Celts, and that the legend of the flood
'

had been transferred to them. But, as J. F. Marcks observes,^ to

stigmatize the story as a legend is pure assertion there are numerous ;

well-attested instances of disastrous inundations in mediaeval and

modern times and unless good reason can be given for rejecting the
;

tradition, it must be regarded as historical. Besides, Miillenhoff

has to account for the facts that Jutland was known to Pliny and
Ptolemy as the Cimbric Chersonese, and that Roman officers found
the Cimbri inhabiting it. He extricates himself from this impasse
with characteristic ingenuity. Reminding us that Columbus called
the natives of the Bahamas Indians, he affirms that the Roman
officers, misled by Greek geographers, fancied when they reached
Jutland that they had got to the extremity of Celtica, and that the
people whom they saw in Jutland were Cimbri and Augustus, ;

wishing the Romans to feel that the memory of old defeats had been
wiped out by the humiliation of the Cimbri, let them believe that the
ambassadors from Jutland belonged to that people. But Columbus
erred because, groping, as it were, in the dark, he hoped to discover
an Oriental people by a westward route, the direction and length
of which he had miscalculated as Marcks points out,''' the victories
:

of Marius had sufficed to soothe Roman self-love and, moreover, ;

Miillenhofl's desperate conjecture fails to explain why so remote

a tribe should have sent their envoys across Europe, and requires us
to believe that they asked forgiveness for outrages which their
forefathers had not committed. He considers that the original home
of the Cimbri is plainly indicated by the first known fact in their

1 Illyr., 4.
2 Bes gestae divi Augusti, ed. Th. Mommsen, 1883, 5, 11. 14-8 (p. Ixxxxii),
Cf. Strabo, vii, 2, § 1.
^ Deutsche Altertumsktmde, ii, 282-90.
* Strabo, vii, 2, § 2. ^ Ammianus Marcellinus, xv, 9, §§ 2, 4.
" Bonner Jahrbiicher, xcv, 1894, pp. 3G-7. ' lb., p. 40.

Nn 2
548 THE OTMBRT AND TEUTONI
history, namely their attack on the Boi,^ who Hved in the upper
valley of the Elbe. Accordingly he contends that they migrated
from the central valley of the Elbe, that is to say, Saxony. But the
fact that they had attacked the Boi in the course of their wanderings
does not prove that they had not marched from the north before
they encountered them.
Georg Wilke ^ has recently endeavoured to support Miillenhofi's
theory by archaeological arguments. He observes that the kingdom
of Saxony, the western part of Niederlausitz, and the south-eastern
part of the province of Saxony are rich in finds of the periods of the
Early Iron Age known as La Tene I and La Tene II, while remains
of the final period — —
La Tene III are very rare that in the northern ;

part of Brandenburg and the western part of the province of Saxony

this difference is not noticeable and that in the eastern part of
;

' '
—
our area (unseres Gehietes) which, if I do not misunderstand a
vague expression, means the area comprising all the above-mentioned
—
lands antiquities of La Tene III are very numerous. The pheno-
mena cannot, he insists, be accidental, and can only be explained
by the assumption that towards the end of La Tene II, between 150
and 100 B.C., the population of the region in which finds of La
Tene III are rare was considerably diminished by a great emigration.
The argument is ingenious but no one who is familiar with archaeo-
;

logical maps —
no one who remembers that up to the end of 1907
only one interment of the Early Iron Age had been found in Scotland
— will admit that it outweighs the testimony of the ancient writers.
Notwithstanding the general vagueness of their statements, it is
clear that they believed the Cimbri to have set out from Northern
Germany, and that at the beginning of our era the remnant of the
nation was domiciled in Jutland.^
The Teutoni, according to Miillenhoff and almost every other
writer, ancient* and modern, originally dwelt somewhere in Northern
Germany. But the Teutoni were probably neighbours of the Cimbri.
Accordingly Dr. Kossinna, w^ho accepts Miillenhoff's theory regarding
the latter, logically develops it, and places the Teutoni in Northern
Bavaria.^ As his premiss is questionable, it is needless to discuss his
conclusion.^
II. Regarding the chronology of the Teutonic invasion, the ancient
authorities are not agreed. The portion of Livy's work which
narrated the history of the period is lost. According to Plutarch,"^
the Teutoni and Cimbri made their first appearance side by side, and
pursued their wanderings together until after Marius had been
appointed to take command against them. But Mommsen ^ attaches
1Strabo, vii, 2, § 2.
2Deutsche GeschichtshldUer, vii, 1906, pp. 291-303, especially 292-3 and 300.
' The article on the Cimbri in VsuvXy'' sReal-Encyclopadie,u\, 1899, col. 2449-

52, is well worth reading. Cf. C. Jullian, Hist, de la GauJe, iii, 53, n. 1.
* Pliny, Nat. Hist., xxxvii, 2
(11), § 35 ; Mela, iii, 4, § 32 ; 6, § 54 Ptolemy,;

Gcogr., ii, 11, § 9.

^ Westdeutsche Zeitschr. f. Gesch. und Kunst, ix, 1890, p. 213.
« See Rev. celt., xii, 1891, pp. 11-2. 'Marius, 11-2, 15, 19, 25.
« Bom. Gesch., ii, 1889, p. 183, n.* {Hist, of Borne, iii, 1894, p. 444, n. 1).
 THE CIMBRI AND TEUTONI 549

more weight to the Epitome of Livy,i in which the Teutoni are

nientioiied for the first time in the narrative of the events of 102 B. c.
Mommsen's view has, of course, been widely accepted ; but Miillen-
hofE^ dissents from it. His argument is as follows. The Epitome,
in recording the battles of 113, 109, and 105 b. c, mentions the Cimbri,
but not the Teutoni. In all three cases, however, Livy himself must
have mentioned the Teutoni as well for, according to Julius ;

Obsequens^ and Velleius Paterculus,* the Teutoni fought in the

battle of 113 according to Florus ^ and Velleius Paterculus,*' they
;

fought in the battle of 109 and according to Florus, Eutropius,^

;

Orosius,^ Valerius Maximus,^ and Velleius Paterculus, they fought

in the battle of 105 and the authority of all these writers was
;

probably Livy. Moreover, two passages in Plutarch ^^ appear to

show that Posidonius, who was probably Livy's authority, had
mentioned the Teutoni in his allusion to the battle of 105.ii
There is a passage in Caesar's narrative which may help us to
understand the silence of the Epitome. In the speech which he puts
into the mouth of the Arvernian chief, Critognatus,^^ he makes him
allude to the war which the combined forces of the Cimbri and
Teutoni waged against the Gauls but a few lines further on he
;

makes him say, without mentioning the Teutoni, that the Cimbri
left Gaul to invade other lands. So I believe that the epitomizer,
in the passages in which he mentioned the C^imbri alone, hastily
used the word Cimhri as a general expression for the united hosts
of the two peoples.i^
III. Dr. Kossinna maintains that Teutoni is a Celtic word, for,
he argues, it could only be German if it belonged to the time, long
anterior to the first century B. c, when the transmutation of
mute consonants (under Grimm's law) had not yet begun and he ;

infers, in opposition to the general opinion, that the Teutoni were

not a Germanic but a Celtic people .^^ M. d'Arbois de Jubainville,
however, remarking that the Celts knew the Germans before that
change occurred, and that the Gallic alphabet had no th, suggests
that Teutoni was only the Gallic form of a German name Theudands,
'
Quand nous Fran9ais,' he says, nous disons Cologne au lieu de '

Koeln, Mayence au lieu de Mainz, nous obeissons a une sorte d'instinct

conservateur, comparable a celui qui, vers la fin du second siecle
avant J.-C, faisait prononcer par les Gaulois Teutoni le nom de . . .

peuple transforme en Theudands depuis un siecle ou deux sur la

rive droite du Ehin inferieur.' ^^
The prevalent view that the Cimbri were Germans. ^^
is Festus

^ c. 63, 65, 67-8. 2 Deutsche Altertumskunde, ii, 292, 295, 298.

3 c. 98, 103. 4 Ed, I. C. H. Krause, ii, 8, § 3 ; 12, § 2.
5 i, 38 (iii, 3, §§ 1-4). 6 ii, 12.
' V, 1. « V, 16, § 9. 9 iv, 7, § 3.
1" Marius, 16 Sertorius, 3.
;
" See Rev. celt., xii, 1891, pp. 3-7.
" B. O., vii, 77, §§ 12, 14.
1^ See also Desjardins, Geo(jr. de la Oaule rom., ii, 309-10, 316-7.
1* Wasldeulsche Zeitschr. f. Oe-sch., &c., ix, 1890, p. 213, 11. 37.
>5 Rev. celt., xii, 1891, p. 16.
^« De signijicatione verhorum, od. MUller, 1839, p. 30.
550 THE CIMBRI AND TEUTONI
says that Cimhrl was a Gallic word, meaning freebooters [Civthri ' '

lingua gallica latrones dicuntur) ; Plutarch ^ that freebooters were

called Cimbri by the Germans (Kt/>t/3pov9 cvrovo/xa^ovo-t Vipixxxvot tovs
\r)(TTd<s) but the word Gallicus was used so loosely that the two
:

statements are not necessarily irreconcilable. ^ Most of the arguments

on either side have been stated by Canon Rawlinson.^ The argu-
ments for the prevalent view are as follows (a) the name Cimbri
:
—
is believed to be of Teutonic origin. It has been variously derived
from Kaniffer (a warrior), from chempho (a champion), and from
the Scandinavian kimpari (to rob).* Rawlinson, however, pleads
that the Latin form Cimhri could not have been evolved either
from Kdmpfer or from chempho, but that from Kcimpfer would
have come Camfi or Camferi ', from chempho Camfones or
'
'
' ' '

'
Camphi '. But, if there is no known German stem from which
Cimhri with the meaning of freebooters could be derived, Cimhri
' '

is nevertheless regarded by scholars as a German word.^ (h) Jutland,

which is generally believed to have been the original home of the
C^imbri, was, we are told, bordered by lands the inhabitants of which
were mainly German. But it is admitted that Celtic tribes once
occupied the valleys of the Rhine and upper Danube. Why then,
asks Rawlinson, may we not suppose that Jutland also was peopled
by Celts ? (c) As the Cimbri were allied with the Teutoni, and as
the Teutoni were confessedly German, it is reasonable to infer that the
Cimbri were German also. The answer is that the Cimbri were also
allied with the Helvetii, and that the Helvetii were Gauls, (d) The
Cimbri had blue eyes and fair hair. This argument, it is replied, is
w^orthless for, if our authorities are to be believed, the Gauls had
;

the same, (e) The host of the Cimbri was accompanied, and their
movements were directed, by priestesses, not by priests. Now
Caesar tell us that the Gauls had priests, and that the movements of
the German army of Ariovistus were directed by women. But Caesar
does not say that there were no priestesses in Gaul and, if Pom- ;

ponius Mela is to be believed, there were. Moreover, Rawlinson

argues that Strabo's account ^ of the priestesses of the Cimbri agrees
'
better with w^hat we know of the bloody rites of the Druids than . . .

with what the most trustworthy writers tell us of the religious

temper of the ancient Germans '. But Rawlinson had apparently
forgotten the bloody ', or at least fiery rites with which Ariovistus
'

and his priestesses intended to sacrifice Troucillus.'^ (/) Caesar,^

^ Marius, 11, § 5. In CamUlns, 15, § 1, Plutarch calls the Cimbri Celts.

This confusion, as every scholar knows, was common.
^ See J. C. Zeuss, Die Deutschen,kc., 1837, p. 141, note, and Bonner Jahrbiicher,

c, 1891, p. 12.
Journ. Anihr. Inst., vi, 1877, pp. 150-8.
3

See J. C. Zeuss, Die Deutschen, &c., p. 141, note.

*

5 See A. Holder, AU-celtischer Sprachschatz, 1, 1015. Fr. Matthias connects

Cimhri with Kimtn (old Germ. Kimha), meaning bank', or 'coast', so that
'

Cimhri would mean 'dwellers on the shore'. Sec Mitthcil. aus d. hist. Lit.,
xxxiii, 1905, pp. 7-8.
« vii, 4, § 3.
7 B. C, i, 53, § 7. Sec p. 07. ^ lb., 40, § 5.
 THE ClMBRi AND TEUTONI 551

and Tacitus^ agree in calling the

Strabo,' Augustus,- Pliny j*^ Seneca,'*
Cinibi'iGermans. To this argument Rawlinson replies that if Caesar
had been personally familiar with the Cimbri, great weight might
be attached to his testimony, but that there is no proof that he had
ever seen a single individual of the Cimbrian nation.^ As for Strabo,
Tacitus, and Pliny, in the canon's opinion, the terms " Gaul " and '

" German " are with them geographical rather than ethnological.'
I believe, however, that, as one may infer from the care with which
'
Tacitus discriminated between those tribes who only affected '

a Germanic origin and those whom he regarded as genuinely German,^

he aimed, at all events, at ethnological accuracy.
To prove that the Cimbri were Celts, it has been argued (a) that
the word Cimbri is a Latin corruption of Cymry. But this argu-
'
'

ment, although Professor Ridge way has rushed into the linguistic
arena to revive it,^ is obsolete. Whether the Kymry have ethno-
'

logically anything to do with the Cimbri or not, the names,' sa}S

Sir John Rhys,^ have absolutely nothing in common, in spite of
'

what charlatans continue to say to the contrary ' : (6) Sallust,^^

Florus,^! Appian,i2 Diodorus Siculus,^^ and Dion Cassius,^* agree in
calling the and, says Rawlinson, the importance of
Cimbri Celts ;
'

these witnesses is the greater because they evidently intend their

assertions ethnically.' Perhaps they do but how can we tell that ;

their ethnological information was correct ? Sallust was a grossly

inaccurate writer.^^ Florus was a blundering rhetorical compiler,
whose worthless chapter on the Gallic war ^^ gives the measure of his
authority. Dion, as every scholar knows, confused the Celts with
the Germans. As M. d'Arbois de Jubainville ^' observes, Caesar was
the first writer who drew a marked distinction between the two
peoples and even after his time there persisted, in the writings of
;

Appian and others, par une sorte de routine, I'usage de la confusion

'

entre les Celtes et les Germains.' (c) Boiorix, the name of the

' vii, 1, § 3.
^ Res gestae div'i Augusti, ed. Th. Mommsen, 5, 14-8 (p. Ixxxxii).
^ Nat. Hist., iv, 13 (28), § 99.
^ adHelv.,7,^2. s
Qerm., 37.
® to the sense in which the word Germani was used by the Reman envo^ s
As
who told Caesar that the Belgae were of ' German origin, see pp. 332-3. '

' Germ., 28.

8 Cambridge University Reporter, March 3, 1908,

p. 649.
9 Celtic Britain, 3rd ed., 1904, p. 281. Cf. Rev. celt., xxix, 1908, pp. 217-8 ;
XXX, 1909, pp. 384-91. Nobody would call Professor Ridgeway a charlatan ;

but he has rashly associated himself with charlatans.

'^Jug.,lU. iiiii, 3.
1'^
B. C,29 Celtica, 2.
i, ;
'^
v, 32, §§ 4-5. '* xliv, 42, § 4.
^5 Henri Martin {Bull, de de Paris, 2« ser., xii, 1877, p. 487)
la Soc. d'anthr.
laid stress upon the fact, if it is a fact, that Sallust had seen Cimbrian
prisoners in the streets of Rome, But unless Sallust was also familiar with
the appearance of Gauls, and was as acute an observer as Dr. Beddoe, it does
not follow that his conclusion was correct and Pliny, who visited the;

country of the Cimbri, described them as Germans.

^« i, 45 (iii, 10).
^^ Rev. celt., xii, 1891, p. 16. See also the same writer's Les premiers habitants
de r Europe, ii, 1894, pp. 401, 419-20.
552 THE CIMBRI AND TEUTONI
Cimbrian leader, is admittedly Celtic. This argument looks strong ;

but may not the true inference be that the Cimbrian host was com-
posed of mingled Celtic and Teutonic elements, and that, as Mommsen^
suggests, the leader may have arisen from the C/clts who joined the
Cimbri on the march ? (d) The Romans, in the war with the Cimbri,
employed Celts as spies and it has been inferred that the Cimbri
;

must have spoken a Celtic dialect. But Mr. Hyde Clarke considers
that Celtic spies would have been employed even if the Cimbri had
been Germans, as they would have had more practice in communi-
'

cating with them than the Romans, who were not at that time in
contact with the Germans '.^ (e) The Cimbrian system of warfare
was, as Mommsen
himself admits, 'substantially that of the Celts;'**
and Rawlinson adds that the same may be said of the Cimbrian use of
wagons and the Cimbrian practice of sacrificing prisoners to the
gods. But the Cimbrian system of warfare was likewise substan- '

tially that of Ariovistus and his German host. As to wagons, what

'

would the canon propose as a substitute ? At all events, Ariovistus

and his Germans w^ere, like the Cimbri, sensible enough to use
wagons * and Ariovistus and his Germans sacrificed prisoners to the
;

gods.^ Arguments like these might be used to prove that all Celts
were Germans, and allGermans Celts.
Professor Rolleston ^was inclined to identify the Cimbri with the
Celts on craniological grounds. The skulls found in the tombs of the
Neolithic Age in Denmark closely resemble those found in the British
round barrows, which are or were believed by many ethnologists
ignorant of archaeology and history to have belonged to Belgae :

they also resemble the skulls of the modern Walloons, the assumed
descendants of the continental Belgae, and those of the Sion type,
which the Swiss ethnologists. His and Rutimeyer, ascribed to the
Helvetii. But it is certain that the men whose skeletons have been
found in our round barrows were not Belgae it is certain that
;
'^

the Sion skulls did not belong to the Helvetii properly so called ;

and it is not certain that the Cimbri themselves were of the same race
as the men whose skulls have been found in the neolithic tombs of
Denmark.
Dr. J. Thurnam ^ adduced, as an argument, the resemblance of '

the armour of the Cimbri, as described by Plutarch {Marins, 25),

to that of the Gauls as described by Diodorus (v, 30), and its dis-
similarity to that of the Germans as described by Tacitus.' He
remarks further that a passage in Pliny ^ seems to show that the

^ Rm. Gesch., ii, 1889, p. 172 {Hist, of Borne, iii, 1894, p. 431). According
to Miillenliofif {Deutsche AltertumsJcunde, ii, 120), Boiorix is simply a German
'
'

word Baiarik Gallicized. But there was another Boiorix, king of the Boi,
'
'

a Celtic people. See Livy, xxxiy, 46, § 4.

* Journ. Anthr. Inst., vi, 1877,
p. 157.
" Hist,
of Borne, iii, 431 {Bom. Gesch., ii, 173).
* J5.(?.,i,5],§2. J*
76., 53, §7.
^ Grcenwell's British Barrows, 1877, p. 032, note.
^ Sec Rice Holmes, A)ic. Britain, pp. 429-30.
^ Crania Britannica, pp. 88, n. f, 147, n. §.
9 Nat. Hist., iv, 13 (27), § 95.
 THE CIMBRI AND TEUTONI 553

language of tlie Cimbri was Celtic :

— ' Morimarusam a Cimbris
vocari hoc
; est, mortimm mare.' (Mor y marw in Welsh = Dead
Sea.i) But Much holds that marus, meaning dead was an old '
',

Grerman word and A. Holder says the same of mori in other

;
^ ^ :

words, Morimarusa, so far as we know, may just as probably be

German as Celtic.
When we ask ourselves whether the Cimbri were Celts or Germans,
we should be careful to settle exactly what we mean. Two thousand
years hence, perhaps, ethnologists will be disputing as to whether
the English were a Celtic or a Teutonic people. English know We
that we are neither one nor the other, but both ; and that the blood
of ancestors who were neither Celtic nor Teutonic is also coursing
in our veins. We may indeed be sure that the Cimbri were not as
composite a people as ourselves but we may be equally sure that
;

they were not homogeneous. It would only be true in a limited

sense to say that even the Arverni, for instance, were a Celtic people.
In point of fact, they were a Celtic-speaking people, among whom
the aristocracy were men of more or less pure Celtic blood, while the
populace was a medley of races. Even assuming that the familiar
descriptions of the physical characteristics of the Cimbri applied to
the entire people, the people might have been composed both of Celtic
and Teutonic elements for the physical characteristics of the
;

Celts and the Germans are described by the ancient writers in terms
which are practically identical. If the Cimbri all spoke a Celtic,
or if they all spoke a Teutonic dialect, Celts or Teutons may, for
aught we know, have formed the conquering, and Teutons or Celts
the conquered section of the population. We know that Celts lived
for centuries in Germany and they may have become intermingled
;

with Teutons.'* I do not deny that the Cimbri were mainly a Teutonic,
or that they were mainly a Celtic people. I simply deny that either
the one theory or the other can ever be proved. But, as the only
ancient writers whose opinions on such a point are worth considering
are unanimous in calling the Cimbri Germani, I incline to accept,
with the necessary limitations, the orthodox view.^

WHEN DID ARIOVISTUS ARRIVE IN GAUL ?

Caesar puts into Ariovistus's mouth the statement that for fourteen
years his Germans had not sought shelter beneath a roof .^ It has been
inferred that Ariovistus entered Gaul in 71 B.C. and Mommsen'^ ;

endorses this conclusion. But, Schneider asks, is it certain that the

wanderings of Ariovistus and his host had not begun in Germany
itself ? Ariovistus, in his interview with Caesar, tauntingly remarked

^ Cf. Zeuss, Gram. Celt., pp. 13, 785.

'^
A. Holder, Alt-cellischer Sprachschatz, ii, 449. ^ j(j^ 628-9.

* See pp. 331-2, 340 ; and cf. C. Jullian, Hist, de la Oaule, i, 244, note.
5 Cf. ib., iii, 53, 11. 3. « B. G., i,
30, § 7.
' Rom. Gesch., iii, 1889, p. 247, n.* {Hist, of Home, v, 1894, p. 35, n. 1).
554 AK10V18TUS IN GAUL
that tlio Acdui had not helped the Roiuaii.s in the recent war with '

the Allobroges '.1 This war took phice in 61 B.C. Merivale- infers
that it must have taken place before the Aedui were menaced by
'

the 8uevi ', and therefore that the date of the arrival of the Germans
'

cannot be placed earlier than 61 b.c' He means of course that the

Aedui would not have been in a position to help the Romans after
they were themselves threatened by Ariovistus but I doubt whether ;

this consideration would have prevented Ariovistus from making

what he thought a telling remark. I am inclined to believe that he
arrived in Gaul, if not as early as 71 B.C., at all events before the
rebellion of the AUobroges. For, according to Caesar, Ariovistus
afhrmed that he would never have left his own home and his kindred
without strong inducements ^ and this statement, if true, is hardly
;

consistent with the theory that his wanderings had begun in Germany.
The general impression which the First Commentanj leaves upon my
mind is that his stay in Gaul had been prolonged.

THE BATTLE OF MAGETOBRIGA

Stoffel * infers from a passage in one that the of Cicero's letters ^

battle of Magetobriga took place in 60 B.C. following Mommsen,^

the same authority, and also referring to B. G., i, 35, § 4, says 61.
I agree with Stoffel. Cicero, in the letter referred to (written in
60 B.C.), tells Atticus that the Aedui have recently suffered a defeat
[evidently at the hands of Ariovistus] while Caesar says that in ;

the consulships of Messala and Piso, that is to say in 61, the Senate
had decreed that the Governor of Gaul for the time being, w^hoevcr
he might be, should protect the friends and allies of the Roman people.
This decree, as Long ^ remarks, appears to have been made on the '

occasion of the ineffectual efforts of the Aedui to oppose Ariovistus.'

But, as it appears from Cicero's letter that the Gauls were still at
w^ar with Ariovistus in the following year (60), and as the battle
of Magetobriga appears to have been the closing event in the war,
I infer that it took place in 60 and not in 61 b.c.^
Mommsen ^ says that this battle was fought between Ariovistus and
the Aedui and their clients only Thierry ,i^ on the other hand, thinks
:

that Ariovistus encountered the united forces of the Aedui and the

1 £.(?., i, 44, § 9.
^
Hist, of the Romans under the Empire, i, 273, note.
^ —
B. G., i, 44, § 2 non sine magna spe magnisque praemiis domum propin-
quosque reliquisse.
* Guerre de Cesar et d" Arioviste, p. 29, n. 1.
5 Alt., i, 19, § 2. The text
uncertain. See Tyrrell and Purser, The
is
Correspondence of Cicero, vol. i, 3rd ed., 1904, p. 426.
6 Rom. Gesch., iii, 1889, p. 247 {Hist, oj Rome, v, 1894, p. 35).

' Caesar, p. 35.

* M. Jullian {Hist, de la Gaulc, iii, 158, n. 1) takes the same view, but some-

what doubtfully.
^ Ron. Gesch., iii. 247 {Hist,
of Rome, v, 35).
^^ Hist, des Gaulois, G"^ ed., ii, 1
800, p. 70.
 THE BATTLE OF MAGETOBRIGA 555

Scquaiii. According to Caesar,^ Ariovistiis, as the ally of the Sequani,

defeated the Aedui and their dependants repeatedly '. But,' says
' '

Diviciacus, whose speech Caesar reports,'^ a worse fate had befallen '

the victorious Sequani than the beaten Aedui and he goes on to ;

'

say that Ariovistus, having defeated the united Gallic forces in

'

one battle, which took place at Magetobriga \^ &c. The use of the
words semel (once) as opposed to semel atque iterum (armis contendisse)
and proeliis compluribus and of Gallorum as opposed to Haeduos
eorumque dientes would seem to imply that the Sequani, finding that
they had not profited by their victories, had joined forces with the
Aedui and made a desperate effort to get rid of their masterful ally ;

and this view is supported by the fact that Caesar, in the speech *
which he made at Vesontio (Besan9on), alluded to a single battle not —
—
one of a series in which Ariovistus defeated the Galli. Evidently this
was the battle of Magetobriga. I believe, therefore, that Thierry is
right.

WHERE WAS L. CASSIUS DEFEATED BY THE

TIGURINI ?

According to the MSS. of the Epitome of Livy (ch. 65),^ Lucius

Cassius was defeated in the country of the Nitiobroges, who dwelt in
the departments of the Lot-et- Garonne and Tarn-et- Garonne and ;

although the editio princeps has AUobrogum instead of Nitiohrogum^

the MS. reading is supported by Orosius,^ who places the battle-
field in the west of Gaul. The alleged pursuit of the Tigurini as far
as the basin of the Garonne by a general whose mission was simply
to protect the Province, may appear improbable but it must be ;

remembered that the Helvetii, who included the Tigurini, intended

to settle in the country of the Santoni and Mr. W. E. Heitland ;

suggests to me that they may have been influenced by the recollection

of the enterprise of the Tigurini. A. de Valois conjectured that '^

what Orosius really wrote was usque ad Rhodanum.

WAS DUMNORIX VERGOBRET OF THE AEDUI ?

Merivale affirms that Dumnorix had succeeded Diviciacus in the '

Vergobret '.^ Now there is no evidence that Diviciacus had

office of
himself been Vergobret and if he had been, Dumnorix could not
;

have succeeded him without violation of the law.^ Merivale of course

1 B. G., i, 31, § C ; vi, 12, § 3. ^ lb., i, 31, § 10.

=*
lb., § 12. « lb., 40, § 8.
5 T. Livii periochae, ed. 0. Jahn, 1853, p. 69.
. . .

^ L. Cassius consul in Gallia Tigurinos usque Oceanuni persecutus rursusque

ab isdeni insidiis circumvcntus occisus est. v, 15, §§ 23-4. Desjardins {Geogr.
dc la Gaule rum., ii, 241, note, 311, n, 4) argues that the testimony of OrosiuK
cannot outweigh that of Livy ' Notitia Galliarum, p. 553.
!

**
Hist, of the Romans under the Empire, i, 281. " B. G., vii, 33, § 3.
556 DUMNORIX VERGOBRET OF THE AEDUI
admits this but lie suggests that the law was relaxed because Duin-
;

iiorix wasa popular favourite '. Thus does one assumption beget
'

another. Besides, Caesar only says that Dumnorix held the princi-
patus at the time of Orgetorix's mission and principatus does not
;

necessarily mean the office of Vergobret '.^ In B. G., vi, 8, § 9, it

'

denotes the chief magistracy

'
but in vii, 39, § 2, where we learn
' ;

that between Eporedorix and Viridomarus there was de princi'patu

contentio, the meaning is simply that they were rivals for power
' '
;

for the Vergobret was then Convictolitavis. Still, I am inclined to

think that Dumnorix was Vergobret for otherwise we should have
;

to assume that at the time to which Caesar refers he was strong

enough to overpower the Vergobret and if so, he would probably
;

have made himself king.

WHEN WAS CAESAR BORN ?

According to Velleius Paterculus^ Caesar was about 18 in the

year when Sulla became supreme, that is to say in 82 B.C. Plutarch ^
says that he died at the age of 56. Suetonius ^ and Appian ^ say
that he died in his 56th year. Eutropius ^ says that he was 56 at the
time of the battle of Munda, which was fought in March, 45 B.C.
Macrobius says that he was born in July. Until Mommsen wrote,
'^

the date of his birth was fixed, on the evidence of Suetonius and
Appian, as 100 B.C.; but Mommsen ^ proposed 102. Mommsen's
reasoning has been developed and supported by the Comte de Salis ;
'-^

and it is only necessary to summarize his arguments.

Cicero was born on the 3rd of January, 106 B.c.,iOand became
consul in 63. It may be inferred from a passage in his speech De
lege agraria^^ that a man could not legally become consul until he
had entered his forty-third year. This rule was broken in the case
of Pompey, who became consul in 70, w^hen he was only 35 but he :

had not served as quaestor or as praetor and his election was ;

irregular. Caesar became consul in 59, four years after Cicero and, ;

as there is no direct evidence that he was elected before he had

reached the legal age, we may infer that he was born not later than
102, four years after the birth of Cicero. Again, Caesar and Cicero
were both praetors three years before their respective consulships,
and aediles three years before their respective praetorships. It there-
fore seems clear that, if any exception was made in Caesar's favour,
which permitted him to stand for the consulship before the legal age,

1 B. G., i, 3, § 5. See p. 533, n. 2. " ii, 41, § 2.

* Caesar^ 69. * Divus lulius, 88.
5 B. C, ii, 149. 6 vi, 24.
' Sat., 12, § 34.
i,
^ Rom.
Gesch., in, 1889, p. 16, note {Hist, of Rome, v, 1894, p. 278, n. 1).
" Rev. arch., iiouv. ser., xiv, 1866, pp. 17-22. A
futile objection which
Napoleon III makes to Mommsen's view is disposed of by Long, Decline of (he
Roman Republic, ii, 377, note.
>" Aulus Gellius, xv, 28. " ii, 2.
 WHEN WAS CAESAR BORN ? 557

that exception must have been made at least six years in advance,
before he stood for the aedileship. L'histoire,' says the Comte de
'

SaUs, ne nous dit rien d'une pareille exception, sans raison ni

'

precedent, qui n'aurait servi qu'a avancer d'un an ou deux I'acces

a ces charges.' Furthermore, certain coins of Caesar (numbered
25-29 by E. Babelon in his Description historique et chronologique
des monnaies de la repuhlique romaine, ii, 18-9) bear the number LII.
These coins were probably struck in 49 B.C., immediately after
Caesar's entry into Rome and the Comte de Salis, with whom
;

M. Babelon agrees, believes that the number indicates Caesar's age.

Similarly, M. Babelon remarks, coins of Mark Antony bear the
numbers XL
and XLII, indiquant I'age du futur triumvir.'
'

It is impossible to fix the date of Caesar's birth with certainty but ;

the weight of evidence is in favour of the view that he was born in

102 B.C.

HOW MANY LEGIONS DID CAESAR RECEIVE FROM

THE SENATE AND THE ROMAN PEOPLE ?

According to Appian ^ and Dion Cassius,^ the Roman People and

the Senate only gave Caesar four of the legions which served under
him in Gaul. Desjardins,^ who prefers the authority of Orosius,^ says
that Caesar obtained seven legions from the Roman People, in virtue
of the Vatinian Law. Fustel de Coulanges ^ remarks that, if we
compare the eighth with the tenth chapter of the First Commentary,
we can see that Caesar only got four legions from the Senate for, ;

after he had raised two new ones on his own responsibility, he only
had six. But the question is whether he did raise the two new ones
on his own responsibility. In strict accuracy, he received three
legions from the people and the Senate added another. In 57 B.C.
;

he raised two more legions and thus, at the time of the conference
;

at Luca, when his term of office was prolonged, he was master of

eight. That the four which he raised himself were raised without
special authority from the Senate, is proved by the facts that it
was agreed, in the conference at Luca, that he should receive a grant
for the payment of the legions which he had raised, and that this
grant was duly voted by the Senate.^ Willems,"^ however, supposes
that the Vatinian Law authorized him to raise new legions in Cisalpine
Gaul.

1 B.C, ii, 13. 2 xxxviii,

8, § 5.
^ Geogr. de la Gaule rom., ii, 355. * vi,
7, § 1.
^ Hist, des inst. pol, de Fane. France, —la Gaule rom., p. 45, n. 2.
8 Cicero, Fam., i, 7, § 10. Cf. Mommsen, Rom. Gesch.y iii, 1889, p. 320
{Hist, of Rome, v, 1894, p. 126).
' Le senat de la repuhlique rom., ii, 1883,
p. 651.
 558

DID CAESAR INTEND, BEFORE HE ENTERED

GAUL, TO CONQUER IT ?
Long ^ says that Caesar's object
'
—
before he set foot in Gaul
'

'
and the policy of the Senate was the complete subjugation of Gallia'
and in support of this assertion he refers to B. G., i, 35, § 4, and
Cicero, De Prov. Cons., 13, § 32. But Caesar merely says that, if
Ariovistus declines to accept his terms, then, in accordance with the
'

resolution which the Senate had passed in the consulship of Marcus

—
Messala and Marcus Piso that the Governor of Gaul for the time
being should, so far as the public interest would permit, protect the
—
Aedui and the other friends of the Roman people Caesar would not
suffer the wrongs of the Aedui to go unavenged '
(sese, quoniam
M. Messala, M. Pisone consulihus senatus censuisset, uti quicumque
Galliam provinciam ohtineret, quod commodo rei puhlicae facere posset,
Haeduos ceterosque amicos populi Romani defenderet, se Haeduorum
iniurias non neglecturum) and Cicero merely says that he con-
;
'

cluded that he would not only have to wage war with the tribes
who were already visibly in arms against the Roman People, but
also to bring the whole of Gaul beneath our sway {non enim sihi'

solum cum eis quos iam armatos contra populmn Romanum videbat
bellandum esse duxit, sed totam esse Galliam in nostram dicionem
redigendam). There is nothing in either of these passages to show
what policy, if any, Caesar had sketched out before he started
for Gaul.
Fustel de Coulanges, on the other hand, holds that, at the time
when Caesar was made Governor, he did not meditate the conquest
of Gaul :
—le jour oii il se trouva en presence de 200,000 Helvetes,
'

il avait si pen songe a la guerre qu'il ne disposait que d'une seule

legion.' 2 At all events, it should seem that he was surprised by the

sudden emigration of the Helvetii for he had to return from Geneva
;

to Italy to fetch reinforcements.^

But whether the conquest was premeditated or was not definitely
planned before the campaigns of the first year had made it practicable,
Caesar was ambitious his ambition needed riches and an army
:

proved in the field and even to a pacific governor the Helvetian

;

and the German peril would have presaged a fresh outburst of that
—
perennial fount of war, the need of securing a frontier which could
only be secured by indefinite extension.

^ Caesar, p. 36.
'^
Hist, des inst. pol. de Vane. France, —
la Gaule rom., pp. 45-6.
» B. G., \, 10, § 3. Cf. Class. Quarterly, iii, 1909, p. 208.
 ;

SECTION VI.— MILITARY

THE NUMERICAL STRENGTH OF THE CAESARIAN
LEGION
Colonel Stoffel pours ridicule on Frohlich and other scholars for
wasting their time in trying to calculate the normal strength of the
legion. Qu'entendent-ils,' he asks,
'
par I'effectif normal de la
'

legion ? Peuvent-ils afhrmer qu'il y en eut un ? ^ Colonel Stoffel'

does not indeed, as I understand him, mean to deny that the legion
had a fixed numerical standard on paper, just as an English battalion
of infantry, on a war footing, is supposed to number 1,096 men of all
ranks.2 What he means, I suppose, is that the effective strength of
this or that legion varied from time to time, and that the legions
of Caesar had no normal effective, as distinguished from an ideal
strength. But all this Frohlich admits by ' Normalstarke he
:
'

simply means the fixed numerical standard which the legion, at its
full strength, was supposed to attain.
C. C. L. Lange ^ arsjues that the normal strength of the legion in
the time of Marius was 6,000 (or, according to Festus,* 6,200) men,
because when Mithridates reorganized his army on the Roman model
he fixed the strength of the cohort at 600.^ The legions of Sulla ^ and
of Lucullus were, Lange points out, of the same strength. But he
'^

holds that in Caesar's time the normal strength had fallen below this
standard, first because those of his legions which were at their full
strength numbered less than 6,000 men ^ and secondly because ;

28 legions of Octavian, which were at their full strength, together

with the auxiliaries (/^era tmv G-vvTaa-a-ojxiviav), numbered only
170,000.^ But how can Lange tell what legions of Caesar or of
Octavian were at their full strength ? ^^

^ Rev. de philologie, xv, 1891, p. 139.

^ Lord Wolseley, The Soldier's Pocket-Booh, 5th ed., p. 183.
^ Hist, mutationum rei mil. Bom., 1846, p. 18.
* S. P. Festi De verhorum significatione, ed. C. 0. Miiller, p. 336.
6 Appian, Mithr., 87, 108.
« Plutarch, Sulla, 9 ; Marius, 35. ' Appian, Mithr., 72.

8 Appian, B. C, i, 82 ; ii, 32 ; Caesar, B. C, 1, 7-8 ; Plutarch, Caesar, 32,

44 ; Pompey, 60, 71.
Appian, B. C, iv, 108; v, 5.
9

^" Guischard {Mem. crit. et hist.,

in, 295), referring to the statement of Appian,
says It is clear that Antony, in reckoning the number of soldiers at 170,000,
'

did not form his estimate with reference to the existing strength of the legions
(for they were all much weakened by the recent war), but reduced the whole
number of survivors to the corresponding number of legions at their full strength
so that, reckoning all his effective troops, he only estimated them as forming
560 THE NUMERICAL STRENGTH
Le Beau ^ remarks tliat Cicero had two legions when he was
proconsul in Cilicia, and that this force amounted to 12,000 men.
80 Plutarch 2 says but Cicero himself complains that he only has
;

'
the nominal command of two skeleton legions (me nomen habere '

duarum legionum exilium)?

At the outbreak of the civil war,' says Mr. H. P. Judson,^ Caesar
' '

had with him only the 13th legion.^ But Plutarch ^ says that
. . .

Caesar had at that time 5,000 men. So we may fairly assume that
that number was in round numbers the strength of a legion when its
ranks were full.' This is a very hasty inference. Plutarch is not
a trustworthy authority, least of all about numbers. His estimate
of the strength of Cicero's army, flatly contradicted, as we have seen,
by Cicero himself, proves this, and proves also that he believed the
full strength of a legion to be not 5,000, but 6,000 men. Frohlich ^
undertakes to demonstrate that he mis-stated the strength of the
13th legion. He shows that Plutarch's figures are sometimes wrong ;

and he shows, on the authority of Caesar himself,^ that the legions

had suffered heavy losses in the Gallic war. The 13th legion took
part in the campaigns of 54 b. c. and the three following years there- :

fore it could not, at the outbreak of the civil war, have numbered
as many as 5,000 men. But Frohlich is as hasty as Mr. Judson.
How can he tell that the losses of the 13th legion had not been made
good by fresh drafts {supplementa) ?^ And if the normal strength
of a legion was 6,000, how can he tell that the 13th had lost more
than 1,000 men ?
Mr. Judson 10 endeavours to support Plutarch's statement by a
calculation founded upon certain statements in the Fourth Book of
the Gallic War. 'In the return from Britain in 55 B.C.,' he says, two '

transports came to land below the main port, and the soldiers
debarked and marched overland. From these two ships 300 soldiers
landed. Assuming the two transports to have been of about the
same size, that would average 150 men to a ship. Now Caesar had
80 transports and an unknown number of galleys. He lost 12 vessels
in the storm. It seems likely that those 12 were transports, as they
lay at anchor, and hence would be more exposed to the storm than
the galleys, which were hauled up on the beach. Then at that rate
the 68 transports remaining carried 10,200 men. Allowing for staff-
ofhcers and servants, the two legions must have averaged somewhat
less than 5,000 men.' All this is highly ingenious but it is simply;

twenty-eight complete legions, instead of the forty-three which the triumvirs

had at the outset of the war.' But Guischard estimates the normal strength
of the legion at 6,000. How then does he account for the auxiliaries who were
included in the 170,000 men ?
^ Mem. de litt. tires des registres de V Acad. Boy. des inscr. et belles-lettres,
'

1752-4 (1759), pp. 488-9. ^ Cicero,3Q.

^ Att., V,
15, § 1. Cf. R. Y. Tyrrell and L. C. Purser, The Correspondence of
Cicero, iii, 1890, p. 60. ^ Caesar's Army, pp. 5-6.
5 B.C., i, 7, § 7. « Caesar, 32.
' Das Kriegswesen Cdsars, 1891, p. 9. ^ B. C, iii, 2, § 3.

^ I find that Stoffel {Guerre civile, i, 203) has anticipated this suggestion.
1" Caesar's Army, p. 5 ; B. G., iv, 22, § 3 ; 29 ; 31, § 3 ; 36, § 4 ; 37, §§ 1-2.
 OF THE CAESARIAN LEGION 561

labour thrown away. For (1) Mr. Judson has no right to assume that
'
the two transports '
—
still less all the transports were of about— '

the same size '

(2) he forgets to mention that, according to Caesar,
;

not 300, but about 300 soldiers {milites circiter CCG) landed from
the two ships, and that, according to the same authority, there were
originally not 80 but about 80 transports (navihus circiter LXXX
onerariis coactis, &c.) (3) he forgets to mention that part of the
;

two legions were carried in the galleys ^ and (4) he cannot tell
;

whether the two legions were or were not at their full strength.
But if Mr. Judson is a bad advocate, it does not follow that his case
is bad and Stoffel ^ maintains that Plutarch was probably right
;

in saying that the 13th legion, when it crossed the Rubicon, numbered
5,000 men. Frohlich indeed says that the only swppleynentum which
Caesar received in Gaul formed a separate corps ^ but, says Stoifel,
:

Frohlich must not assume that, because Caesar only mentioned one
supplementum,'^ he only received one and furthermore, this par-
;

ticular supplementum did not form a separate corps, but was incor-
porated in the rest of the force. It was, indeed, sent to Agedincum
(Sens), the grand depot of the Roman army, probably to be drilled,
armed, and equipped but, though Caesar does not mention the
;

fact, we may be sure that it was subsequently distributed among

the legions.^
Frohlich ^ adduces evidence to show that, even in Caesar's time, the
normal strength was 6,000. Cicero, in one of his letters to Atticus,^
tells him that Pompey had crossed the Adriatic with 30,000 milites.
Milites, says Frohlich, means legionaries Caesar says that Pompey
:

crossed with five legions ^ therefore each legion averaged 6,000 men.
;

Frohlich goes on to argue that, even if, under the head of milites,
Cicero included cavalry and light-armed auxiliaries, the average
strength of the five legions could not have fallen much below 6,000 ;

because the light troops whom Pompey employed to cover his retreat
from Brundisium could not have been numerous and the archers
;

and slingers whom he employed in Greece, coming as they did from

Pontus and Syria,^ could not have joined his army until after he
left Italy.
I do not think that Frohlich succeeds in proving his case, because
the evidence of Cicero is not enough to go upon. He may have written
from hearsay. Even if Pompey himself had been Cicero's authority,
there would be room for doubt for modern generals have been
;

known to make statements in their letters regarding the numbers

of the armies under their command which the adjutant-general's
returns have subsequently shown to be incorrect. Stoffel ^^ ridicules

Meusel in a MS. note objects that B. G., iv, 29, § 2 {longas yiaves
^
. curaverat)
. .

is certainly corrupt. No doubt ; but it is not less certain that the galleys
carried troops. Cf. iv, 25, § 1. ^ ^^^
^^ philologie, xv, 1891, pp. 140-1.
^ Das Krieijswesen Ciisars,
p. 9. * B. G., vii,
7, § 5 ; 57, § 1.
^ This view is supported by the fact that after i>. G., vii,
57, § 1, we hear no
more of the sitpplonentiun, while in chap. 90 all the legions are accounted for.
® Das Kriegswesen Casars,
p. 10.
' ix, 0, § 3. 8 B. C\, iii,
4, § 1.
' lb., iii, ^o
4, § 3. Rev. de philologie, xv, 1891, p. 140.
1093 O O
562 THE NUMERICAL STRENGTH
Frohlicli's calculations. Pompey's legions could not, he insists,
have numbered, at Brundisium, 6,000 men apiece for at Pharsalia ;

their average strength was only 4,000, and they could not have lost
a third of their number in the interval. Still, the view for which
Frohlich contends is probable in itself there is no evidence against
:

it; and, as he points out, it is supported by the fact that each of

the two cohorts of slingers which Pompey commanded in Greece
numbered 600 men.^
Sextius Rufus^says that Caesar conquered Gaul with ten legions,
each numbering 3,000 men but this statement, is of no use. The
;

number is suspiciously round '

and, for aught that we can tell,
'
;

it may merely represent the writer's own rough estimate, based upon
such data as we possess ourselves. Besides, the statement is hope-
lessly vague. It may mean that Caesar kept his legions at an average
effective strength of 3,000 by regularly filling up with fresh drafts the
gaps caused by casualties or disease or it may mean something
;

quite different.
Weare told, in such a way as to suggest that the number was
remarkably small, that, in the fifth year of the Gallic war, two legions,
including, it should seem, the 400 cavalry that accompanied them,
numbered barely 7,000 men ^ and we know that the average strength
;

of the eight legions which Caesar commanded at Pharsalia was 2,750.'*

As he tells us that these eight legions had suffered heavy losses
since the conclusion of the Gallic war, I infer that, during that war,
their average effective strength never fell below 3,000 men, and was
generally a good deal higher.
My conclusions are these. The ideal strengtb of the Marian legion
was 6,000.^ There is perhaps no positive proof that that of the
Caesarian legion was the same but neither is there any reason to
:

believe that it had changed and the evidence of Cicero, such as it

;

is, as well as the inference which Frohlich draws from the recorded

strength of Pompey's cohorts of slingers, goes to show that it had

not. But, as Mr. Judson sensibly remarks, From the experience '

of modern armies we know that the number of effectives ready for

duty in the field always falls considerably below the number on
the rolls and again, that even the number on the rolls rarely
;

approximates very closely to the full strength of any organization

as prescribed by the tactics.' ^ What we want to know is the effective
force which Caesar brought into action at any and every given time.
Now Caesar, as we have seen, twice gives us the information that we
require. When he fails us, there is but one way of obtaining it for
ourselves. It is idle to say. The average effective strength of a legion

^ B. C, iii, 4, § 3. ^ Breviariiun, cli. C.

3 B. G., V, 49, § 7. ^ B.C., iii, 89, § 2.
^ Even in pre-Marian days, when the ideal strength of the legion was 4,2(X)

(Livy, xxii, 36, § 3), legions were occasionally raised of 5,000 men {ih.), 6,000
{ib., xlii, 31, § 2 ; xliii, 12, § 4 ; xliv, 21, § 8), or even 6,200 {ib., xxxv, 2, § 4).
^ Caesars Army, '
General JSheriuan,' remarks Lord Wolseley {The
p. 5.
Soldier's Pocket-Book, 5th cd., pji. 118-9), 'says that all experience proves
that in a large organized modern army not more than 66 per cent, of the total
force can be reckoned upon for actual battle.'
 OF THE CAESARIAN LEGION 563

was so iiiucli multiply this amount by the nmnbcr of legions. The

:

only sound method is to consider what legions were employed at

any given time, how long they had been enlisted, what service they
had seen, and how far, if at all, the losses which they had sustained
had been made good and it is obvious that, for want of sufficient
;

information, even this method can only lead, at the best, to fairly
approximate results.

WHO MADE THE COHORT THE TACTICAL UNIT

OF THE ROMAN INFANTRY ?
There is good reason, as Marquardt ^ remarks, to believe that it was
Marius who made the cohort the tactical unit of the Roman army ;

for, as far as we know, Metellus, in the Jugurthine war, was the last
Roman general who employed the manipular organization ^ and ;

from the time of Marius the size of Roman armies is reckoned by

both Greek and Roman historians in cohorts.^

THE LEGATI
Legati were officers of senatorial rank, appointed by the Senate *
and immediately responsible to the proconsul to whom they were
assigned. They were his lieutenants, as their name implies, and
were expected to perform any duty with which he might entrust them.
On Monday a legatus might be placed in command of a legion and
lead it in battle on Tuesday he might be charged with the duty
:

of raising a fresh levy of troops. 0. Sumpif ^ observes that the chief

difference between legati and our generals of division consists in '

the want of permanence of their position '.

Stoffel,^ remarking that a passage in B. G., i, 52, § 1, proves that the
legati were not the regular commanders of legions in the first year
of the Gallic war, suggests that each legion was probably then
ordinarily commanded by the principal tribune, and that the other
five tribunes, acting under his orders, commanded groups of cohorts.
In B. G., ii, 26, § 1, Caesar says that, in the battle with the Nervii,
he ordered the tribunes to bring the 7tli and 12th legions together.
From this Stoifel infers that in the second year of the war the position
of the legati was unchanged but he believes that subsequently they
;

became veritables chefs de legion '. There is no doubt that in

'

^ Rom. Staatsverwaltung, 2nd ed., ii, 1884, p. 435.

'^
8ce Sallust, Jug., 49, § 6.
^ See C. C. L. Lange, Hist. 7rmtationum rei mil. Rom.,
pp. 10-7, and Pauly'b
Real-Enajclopudie, vi, 1909, col. 1(300.
* Cicero, Fani., i,
7, § 10. Cf. Appian, Milhr., 94, and Plutarch, Pomp., 25.
* Caesars Beurteilung seiner
Offizierc in den Oomm. vom gall. Kriege, Zweiter
Toil, 1893, p. 10.
^ Guerre de Cesar el d'Arioviste, pp. 120-7.
2
564 THE LEGATI
Caesar's time tlie office of legatus was passing through a transitional
stage and gradually tending to crystallize into the form which it
assumed under the empire, when the legatus became a legatus legionis.
But Stoffel is not quite right. No legatus was a veritable chef de '

legion ', that is to say a permanent chef ', in Caesar's time. If the
'

legion had any recognized commander, that commander w^as a

tribune, who was, however, of course frequently superseded by
a legatus and after the second campaign in Gaul, as before, any
;

legatus who commanded a legion was specially appointed to his

command by Caesar and held it only so long as Caesar pleased.
Stoffel has overlooked several passages which prove this. If Caesar
placed individual legati in command of individual legions in the
battle with Ariovistus, he did the same in his attack on Gergovia ^ ;

and the legati who commanded legions in the winter of 55-54 B.C.
and in the winter of 54-53 were specially appointed by him to their
commands for those two periods ^ in other words, when they received
;

their commands, they were not chefs de legion '.

'

It isworth mentioning that Caesar sometimes appointed a legatus

whom he specially trusted to the command of several legions.^
Caesar mentions the following officers as legati, C. Antistius —
Reginus, M. Antonius, L. Aurunculeius Cotta, L. Caesar, C. Caninius
Rebilus, T. Labienus, L. Munatius Plancus, Q. Pedius, T. Sextius,
M. Silanus, P. Sulpicius Rufus, Q. Titurius Sabinus, and C. Trebonius ;

while, according to Hirtius, Q. Calenus and P. Vatinius, who are not

mentioned by Caesar at all, held the same rank. Besides these,
Quintus Cicero, whom Caesar does not call a legatus, is designated
as such by his brother, Marcus ^ and that Servius Galba was one
;

we know from Hirtius.^ The rank of several other officers, however,

who exercised independent commands in Gaul, namely, D. Brutus,
P. Crassus, L. Minucius Basilus, L. Roscius, M. Sempronius Rutilus,
and C. Volcacius Tullus, is doubtful. Brutus was apparently a
legatus at the outset of the civil war,^ but he could not have been
one before, unless the Gabinian Law had been either violated or
superseded for he was not yet a senator.
; Publius Crassus, who
commanded Caesar's cavalry in the battle with Ariovistus and was
"^
commander-in-chief in the Aquitanian campaign, is said by Dion
to have been a legatus in the latter year (56 B.C.), and Caesar's
silence does not prove that he was not but he too was not a senator.^
;

Caesar, without giving Basilus any title, mentions him in the same
breath with Fabius, whom he calls a legatus ^ but this proves nothing,;

for Caesar does the same in the case of Cicero.^^ From the other
passage, however, in which Basilus is mentioned, and in which
Caesar says that he was sent in command of all the cavalry to hunt

^ legatis, quos singulis legionibus praefecerat, quid fieri velit, ostcndit. i>. G.,
vii, 45, § 7. lb., V, 1, § 1 ; 25, § 5.
"

« /6.,i,54,§2; ii,ll,§§3-4; iii, 11,H; v, 17,§2; vii, 34, § 2, &c.

* Fam., i, 9, § 21. ' B. G., vii, 50, § 4.
« Livy, EptL, 110. ' xxxix, 31, § 2.
® Cf. P. Willenis, Le saial de la rep. rom.. ii. 1883, p. 014.
i»
9 B. G., vii, 90, § 5. lb., V, 24, § 2.
 —
THE LEGATI 565

down Ambiorix,! Nipperdey ^ infers that he was simply a prae/ectus

equitum, or commander of cavalry. Roscius is designated by Caesar
in V, 53, § 6, as a legatus according to the a MSS., as a quaestor
according to the /?. P. Groebe^ thinks it unlikely that he was
quaestor, because Marcus Crassus held this office in the same year.
But, unless all the MSS. are wrong, Caesar had more than one quaestor
in that year,^ though he only had one in 55 ^ and in 58 b. c.^ Rutilus,
who is only mentioned by Caesar in one passage,"^ is there said to
have been placed under the command of Labienus and we may ;

perhaps infer that he was not a legatus. Volcacius Tullus was

probably not a legatus in 53 b.c.,^ because he was not yet a senator.
Groebe,^ however, thinks that he would not otherwise have been
entrusted with the command of twelve cohorts, an argument which —
would apply equally to Brutus and Publius Crassus. It has been
suggested that he may be alluded to by Hirtius in a passage in which
Tullius (sic) is mentioned among other legati}^ Groebe^i denies that
this Tullius could have been Quintus Tullius Cicero for, he says, ;

at the time of which Hirtius was speaking 51 B.C. Quintus was — —

in Cilicia. This argument is sound for, although it is not absolutely
;

certain that Quintus arrived in Cilicia before January, 50,^2 {j^ ig more
than unlikely that he would have quitted his winter camp in Gaul
at the close of the previous year, immediately after he had been
placed in command of it.
Irregular things were done in those times and I would suggest ;

that if Brutus and Crassus were not already legati when Caesar
entrusted them with important commands, he may have conferred
the titles upon them.

THE MILITARY TRIBUNES

We are generally taught that Caesar chose his military tribunes,
as a rule, for political reasons rather than for military efficiency ;

that he was therefore obliged to relegate them to a position of com-

parative insignificance and that the legati and centurions gained in
;

importance at their expense. Thus Long ^^ says It is plain from '

Caesar's Commentaries that the tribuni were not employed by him

to command legions ... we may infer that, though he could not
remove the tribunes ^^ ... he reduced them to insignificance and, '
;

^ B.G., vi, 29, § 4.

^ Caesar, p. HI. See also Schneider's Caesar, ii, 640.
""
W. Drumann, GescJi. Roms, iii, 1906, pp. 697-8.
* B. G., V, 25, 5 e
§ 5. /5.^ iv, 13, § 3. 75., j, 52, § 1.
' lb., vii, 90, § 4. 8
75^ vi^ 29, § 3.
8 W. Drumann, op. cit., p. 699. "> B. G., viii, 46, § 4.
" W. Drumann, op. cit., p. 699. 12
Cicero, Fam., xv, 4, § 8.
^^ Decline of the Roman Republic, ii, 26.
'* That Caesar could not remove tribunes whom he had himself appointed
and he apparently had the right or at all events the power of choice is —
incredible. Cf. P. Willems, Le sevat, &e., ii, 63.3 ; C. Jullian, Hi.^f. dp la Ganle, iii,
185, n. 12, and p. 566, n. 2, infra.
566 THE MILITARY TRIBUNES
according to Kraner-Dittenberger,^ their duties in Caesar's arm}- '

are insignificant with the solitary exception of C. Vohisenus

;

Quadratus they commanded only small detachments ... or were

. . .

employed in various administrative duties.'

There is some truth in this view but it is, as a rule, stated too
;

broadly. The only direct evidence in the Commentaries for the asser-
tion that Caesar chose his tribunes without regard to their military
efficiency is his statement that the panic which seized his army at
Vesontio (Besan9on) before his campaign against Ariovistus began '

with the tribunes, the auxiliary officers, and others who had left the
capital to follow Caesar in the hope of winning his favour, and had
little experience in war (hie [timor] primum ortus est a tribunis
'

militum, fraefectis reliquisque qui ex urbe amicitiae causa Caesarem

secuti non magnum in re militari usum hahebant). But if qui refers
to tribunis militum^ it also refers to fraefectis and it is obviously ;

incredible that auxiliary officers and tribunes aHke were generally

chosen without regard to efficiency. It is quite untrue that the
duties of the tribunes were insignificant or that they commanded
'
'
'

only small detachments '. The Commentaries prove that their duties
were most important and it is evident that Caesar must have taken
;

care, with due regard no doubt to political exigencies, to choose

the best men that he could get for the places. The tribunes, like the
centurions of the first rank, attended councils of war.^ In the battle
on the Sambre Caesar instructs the tribunes, not the centurions,
of the 7th and 12th legions to bring them closer together * and as ;

each legion in that battle was apparently commanded by a legatus,

I think we must infer that the tribunes commanded groups of cohorts.
In the sea-fight with the Veneti each ship was commanded by a
tribune or by a centurion.^ When Caesar was about to land for the
first time in Britain, he gave special instructions to the legati and to
the tribunes ^ but of the centurions he says nothing.
; Certain

'
Caesar, ed. 1890, p. 47. See also G. Veitli's remarks in Klio, vii, 1907,
p. 322.
2 B. G., i, 39, § 2. G. Hubo {Neue Jahrbiicher f. Philologie, &c., cxlix, 1894,
pp. 272-4), who has such a high opinion of the tribunes that he refuses to
beheve that the panic began with them, proposes to read tironihus (mihtum,
praefectis, &c.). This conjecture is rightly rejected by Wesener [ib., p. 576),
wlio remarks that some tribunes, e. g. Volusenus, were experienced, others the
reverse. Anyhow, the conjecture is unnecessary and if the text is wrong, ;

nobody can put it right. The centurions and veterans, whose bravery and
experience no one denies, yielded to panic why then should Hubo refuse to
:

believe that the tribunes did the same ? Lange observes {Hist, mntationum rei
mil. Rom., p. 22) that although [in Caesar's time] military tribunes were
'

sometimes, as formerly, appointed by the people (Sallust, Jug., 63 Plutarch, ;

Cato min., 8 Caes., 5

; Suetonius, Caes., 5), it should seem that they were
;

usually nominated by the generals themselves'. The evidence which he cites

for this view is to be found in Cicero, Aft., vi, 3, § 5, and Fam., vii, 5, § 3, from
which it would appear that Caesar offered Trebatius a tribuneship. Cicero
also tells his brother {Q. fr., ii, 13 [15 a], § 3) that he asked Caesar for a tribune-
ship for M. Curtius. It is clear then that whether Caesar appointed all his
tribunes or not, he had tribuneships in his gift.
3 B. G., v, 28, §§ 3-4 vi, 7, § 8. ;
* lb., ii,
26, § ]
s
lb., iii, 14, §§ 3-4. " Jb., iv, 23, § 5.
 ^ ? —
THE MILITARY TRIBUNES 567

tribunes were personally thanked by Caesar for the gallantry which

they had shown in the defence of Quintus Cicero's campj When
the foraging detachment which Cicero sent out from his camp at
Atuatuca was attacked by the Sugambri, the young recruits looked
for orders to. the tribune, who probably commanded the whole
detachment, as well as to the centurions. ^ When Caesar attempted
to surprise Gergovia, each legion was, it is true, commanded by
a legatus but the tribunes evidently played a more important part
:

in the action than the centurions for in describing the efforts which
;

the officers made to keep the troops in hand, Caesar couples the
tribunes with the legati, and does not mention the centurions.*^
Finally, in the battle of Lutecia, the tribunes jointly commanded the
7th legion ^ and in this connexion Caesar makes no mention either
;

of a legatus or of centurions. Still, it remains certain that the position

of the tribunes was not what
had once been Caesar was ready to
it :

grant sinecure tribuneships to men who had no experience of war,

in order to oblige political associates ^ and there is no evidence that
;

any one tribune ever commanded an entire legion in action in any

of Caesar's battles in Gaul.
From the fact that Caesar, in the Civil War, speaks of tribuni
cohortium,^ it has been argued that, in his time, each cohort was
commanded by a tribune but, as Frohlich points out, it would
:
"^

follow from this view that the number of tribunes in each legion had
been raised from six to ten, of which there is no evidence and Caesar ;

was only speaking of tribunes who had been placed in command of six
cohorts which were sent to Cadiz.

WHO WERE THE CENTURIONS '

OF THE FIRST
RANK '

'
The frimorum ordinum centuriones were the centurions who ranked
highest in a legion. This obvious statement is almost the only one
that can be made about them with absolute certainty. do not We
know, for certain, how many centurions of the first rank (primi
ordines) there were in each legion. We do not know, for certain, to
what cohorts and to what maniples the centurions of the first rank
belonged. We —
do not even know at least Mommsen^ does not
whether the frimorum ordinum centuriones formed, in Caesar's time,
a definite class, or whether their number was fixed. Our information
being so scanty, it is only natural that many theories should have
been formed upon the subject.' Since these words were originally
prnited the conclusion which I reached in the first edition that —
^ 2
lb., V, 52, § 4. 11,^ ^i^ 39^ I 2.
^ lb., vii, 47, § 2. —ab
tribunis militiim legatisque, lit erat a Caesare praece-
ptiim, [mihtes] retinebantiu*. .

* 76., vii, 62, § 0. « Cicero, Fam., vii, 8, § 1.

* B. C, ii, 20, § 2. ' Das Kriegswesen Cdsars, p. 21.

8 i?.C.,ti,18,§2. » Ephemeris EjnymiMca, iv, 1881, pp. 238-9.
568 THE CENTURIONS
'
the centurions of the rank were the centurions of the 1st cohort
first '

— has been confirmed by A. von Domaszewski.i But do not take it

upon trust. Before examining the other theories, I will set down
a few essential facts.
1. The ten cohorts in the legion were numbered. As the 1st cohort
ranked above the rest,^ it appears to me morally certain that they
took rank according to their numbers ^ but this view, as will presently
;

appear, is not universally accepted. 2. The 1st centurion of the

1st maniple in each cohort was called 'pilus ])rior, the 2nd jjilus
posterior * the 1st of the 2nd maniple frinceps prior, the 2nd
:

princeps posterior the 1st of the 3rd maniple hastatus prior, the
;

2nd hastatus posterior. The 1st centurion of the 1st maniple of the
1st cohort was called primus pilus or primipilus, and definitely
ranked as the chief of the 60 centurions of the legion.^ 3. Whether
the primi ordines did or, as Mommsen thinks, did not form a definite
class in the time of Caesar, they certainly did so, as he admits, after
the time of Hadrian. 4. If it had not been disputed, I should also
say that it was indisputable that there were at least eight (and there-
fore obviously ten) definite classes of centurions in the time of Caesar ;

for he relates that, for gallantry at Dyrrachium, a centurion named

Scaeva was promoted from the 8th class to the rank of primipilus
(ah octavis ordinibus ad primipilum).^ 5. If we may trust a passage
in Tacitus,"^ there were, in the time of Galba, at least six primorum
ordinum centuriones in the 7th legion. Describing the battle between
Vitellius and Antonius Primus, he writes, Urguehatur maxime septima
legio, nuper a Galba conscripta. Occisi sex primorum ordinum centu-
riones, &c. But if Mommsen's view is correct, it does not necessarily

follow that there were six primorum ordinum centuriones in the other
legions, even in Galba's day.
But Mommsen's view is so strained that, if it were not sanctioned
by his great name, it would hardly be worth noticing. In B. G., i, 41,
§§ 1-3, we read that the legions, ashamed of having yielded to panic
at the prospect of encountering Ariovistus and his Germans, asked the
tribunes and the primorum ordinum centuriones to make their excuses

^ Bonner JdhrbOcher, cxvii, 1908, pp. 90-5.

2 '
At least from the time of Hadrian,' says Mommsen {Eph. Epigr., iv, 230).
But a passage in Caesar which, I think, proves that, in his time, the
there is
1st cohort of the legion was also the highest in rank duabus viissis ,siihsidio
:

cohortibiis a Caesare atqne his primis legionum duarum. B. G., v, 15, § 4.

^ See
pp. 575-7.
* Under the Empire, if not before, there was no pilus posterior in the 1st cohort.

See the next note. The 1st cohort contained five centuries only (although,
like the rest, it had six centurions [Tacitus, Ann., i, 32]) under the Empire ;

and A. von Domaszewski {Bonner Jahrhiicher, cxvii, 1908, pp. 91-2) gives
reasons for believing that this division probably originated at tlie time when the
manipular organization was superseded.
5 Primus centurio erat, quem nunc primi pili appellant. Livy, vii, 41, § 5.
The centurions of the 1st cohort under the Empire were called priynus pilus
(there being two with this title), princeps, hastatus, princeps posterior, and
hastatus posterior. Corpus inscr. Lat., viii, suppl., pars ii, 18,005, 18.072, and
Bonner Jahrhiicher, cxvii, 1908, pp. 90, 92, 97.
6 B. C, iii, 53, § 5. 7 Hist., iii, 22.
 — s

OF THE FIRST RANK 569

to Caesar.i This, as Marquardt observes,^ proves that the privates

recognized the "primorum ordinum centuriones as such no less clearly
than they recognized the tribunes, and therefore that the "priynorum
ordinum centuriones formed a definite class. Moreover, whenever
Caesar alludes to the frimi ordines, he does so in such a way as to
leave no reasonable doubt that they formed a definite class. Describ-
ing the defeat of Sabinus and Cotta, he writes, Turn T. Balventio, qui
superiore anno primum pilum duxerat utrumque femur tragula
. . .

traicitur : Q. Lucanius, eiusdem ordinis interficitur : L. Cotta

. . .

Jegatus omnes coTiortes ordinesque adhortans vuJneratur? It has . . .

never been denied that the primipilus was one of the primi ordines.
Would Lucanius have been described as eiusdem ordinis unless the
description implied that that ordo was a definite class ? Again we
Tesid,Erant in ealegionefortissimiviri, centuriones, qui iam primi
ordinihus adpropinquarentA Would the words qui adpropin- . . .

quarent have been used if the primi ordines had been, not a definite
class, but merely those centurions who stood highest in general
estimation ? A passage in B. G., vi, 40, § 7 Centuriones, quorum
non nulli ex inferiorihus ordinihus reliquarum legionum virtutis causa
in superiores erant ordines huius legionis traducti, &c. is meaningless —
unless it means that the centurions in question had been promoted
to definite higher grades, the highest of which was of course composed
of the primorum ordinum centuriones. Tacitus would not have used
the words sex primorum ordinum centuriones unless the primi ordines
had formed a definite class. Besides, the primi ordines, as well as the
iribuni militum and the legati, were called to councils of war.^ Would
it not have been invidious to summon them if they had been simply
the centurions of the greatest weight and reputation in the legion,
and had not attained a definite rank, which gave them a formal right
ex officio to attend ?
For all these reasons, I unhesitatingly state as a fact that the primi
ordines formed, not only after the time of Hadrian, but also in the
time of Caesar, a definite class.
I shall now proceed to examine the various theories that have
been constructed upon these facts, or upon such of them as the
theorist took into account.
1. H. Bruncke ^ holds that the only difference in rank among the

centurions was between primi ordines and inferiores ordines but he ;

also holds that the primorum ordinum centuriones were the centurions
of the 1st cohort. In this I am sure that he is right but the other ;

part of his theory, which assumes that the centurions of all the
cohorts below the 1st were of equal rank, appears to me inconsistent
with his identification of the primorum ordinum centuriones. In
B. 6-'., V, 44, § 1, Caesar writes, erant in ea legione centuriones, qui . . .

^ Reliquae legionescum tribunis

militum et primorum ordinum centurionibus
egerunt uti Caesari satis facerent.
- R'6m. Staafsvenvaltung, \i, 1884, p. 371, n. 3.
' B. G., V, 35, §§ 6-7. ' Ib„
44, § 1.
' lb., 28, § 3 ; 30, § 1 ; vi, 7, § 8. See also v, 37, § 1 and B. C, i, 74, § 3.
^ Die Banyordnung der Centurionen, 1884.
 570 THE CENTURIONS
iam primis ordinibus adpropinquarent ; andin5. C,ii, 35, § 1,
FahiuR Paeliqnus quidam ex infimis ordinibus, &c. I hold that
Caesar, who uses the comparative inferiores ordines, would not have
used the superlative infimi ordines in exactly the same sense, and
would not have used that expression to denote nine cohorts of equal
rank, to which only one, the 1st, was superior. The expression qui iam
primis ordinibus adpropinquarent obviously means that the centurions
in question had already gained one or more steps in rank, and would
soon be promoted to the highest class of all.^ This meaning is
established in the very next sentence, in which Caesar says that
these two centurions every year contended for promotion with the
'

greatest acrimony ' {omnibus annis de locis summis simuUatibus

contendebant). Again, the expression ab octavis ordinibus, &c., which
I have already quoted, can only mean one of two things. Either it
means, as I hold, that Scaeva was promoted from the 8th class, that
is to say, the 8th cohort or it means that he was promoted from
;

the 8th cohort, the six centurions of which belonged to different classes.
In either case, Scaeva, before his promotion, was below the rank of
primi ordines. But why should Caesar have taken the trouble
to indicate his original rank, if all the centurions below the primi
ordines had been on a footing of equality ? Moreover, under the
Empire the 10th cohort was certainly the lowest for an evocatus,;

or time-expired volunteer ^ was promoted to be a centurion therein.^

Finally, the fact that there were superiores ordines as well as primi
ordines simply pulverizes Bruncke's theory that all the centurions ex-
cept those of the 1st cohort were grouped together in inferiores ordines.^
^ Bruncke actually says {Die Rancjordnung , &c., p. 19), Only by accepting
'

the view that there were two classes, can we explain why Pullo and Vorenns
contended with one another quinam anteferretur, i. e. who should be first
admitted to the primi ordines.'' This is simply begging the question. Caesar's
words can be just as well explained on the hypothesis that Pullo and Vorenus,
having reached the 2nd class, contended with one another who should be first
admitted to tiie primi ordines ; and Bruncke ought to have said, Only by '

accepting the view that there were more than two classes can we explain why
Caesar, in describing the rivalry of Pullo and Vorenus, said that " they were
getting close to the first grade " (primis ordinibus adpropinquarent), and why
he added that " every year they contended for promotion " (omnibus annis
de locis. .contendebant).*
.

2 See
p. 166.
^ Corpus inscr. Lat., xiii, pars ii, fasc. 1, 6728. Cf. Bonner Jahrbi'icJier, cxvii,
1908, p. 90.
* Bruncke calmly denies that superiores ordines are ever mentioned (' dieser

Ausdruck existiert nicht.' Die Rangordnung, &c., p. 19, n. 9). Let him turn
to B. G., vi, 40, § 7, and he will find his mistake. Bruncke also maintains
(pp. 18-9) that 'if promotion really took place according to cohorts' —
that
is to say, that if, for example, the centurions of the 9th cohort ranked above
those of the 10th — one would think that a newly appointed centurion would
•'

have entered as decimus hastatus posterior.^ But, he continues, from Corpns

inscr. Lat., v, 7004, we learn that an optio, or sub-centurion, on being promoted
to a centurionship, became ocfavus pilus prior. On the theory which Bruncke
combats, the 17 centurions below would, he maintains, have had reason to feel
aggrieved. This argument is unavailing. If these centurions would liave had
reason to feel aggrieved, so would those over whose heads Scaeva was promoted.
This optio may have had great merit ; and besides, for aught we know, there
may have been numerous vacancies. Cf. B. G., vi, 40, § 7.
 ——
OF THE FIRST RANK 571

2. Marquardt ^ groups the centurions in six classes, the 'primi

ordines being the ten fill 'pr lores of the ten cohorts, the 2nd class the
ten frinci'pes priores, the 3rd class the ten hastati priores, and so on.
This scheme is plausible at the first glance, because, as each cohort
must have been commanded by some one, it might be argued that the
pilus prior of each cohort must have commanded the cohort ^ and ;

one would certainly say a priori that the centurions who commanded
cohorts must have been higher in rank than any of the other cen-
turions, and therefore must have been the primi ordines.
Nevertheless, Marquardt's scheme must be rejected, because it
flatly contradicts the fact, attested by Caesar himself, that there were
at least eight (and therefore naturally ten) classes of centurions.
It is true that octavis ordinibus, in the passage to which I refer, has
bsen differently interpreted. Kraner^ agrees with Marquardt in
thinking that the primi ordines were the ten pili priores. Yet, with
manifest inconsistency, he says that the centurion who was promoted
ab octavis ordinibus ad primum pilwn was octavus pilus prior. On
his own showing, this man, being a pilus prior, had belonged to the
primi ordines. Therefore primi ordines and octavi ordines w^ere
identical The only way of escaping from this absurdity is to assume
!

that in the expressions primi ordines and octavis ordinibus the word
ordo is used in two different senses, —
that primi ordines means. the '

1st class ', and octavis ordinibus the 8th cohort without reference
'
'

to any class."*
Marquardt's explanation has been condemned, but on different
grounds, by Mommsen. Referring to the passage in the Civil War
to which I have alluded quem Caesar ab octavis ordinibus ad
. . .

primipilum se traducere pronuntiavit he says, — si octavus pilus,

'

octavus princeps, octavus hastatus dignationis tam diversae fuissent

quam fuisse eos adversarii statuunt, non octavos ordines Caesar
ponere debuit, sed eum in quo Scaeva erat gradum ^ and Dr. L. C.;
'

Purser^ echoing this objection says, On Marquardt's theory

'
. . .

octavis ordinibus would be 8th, 18th, 28th, &c.' ^

The well-known passage in which Vegetius describes the system of
promotion has, however, been quoted in favour of Marquardt's view.
The passage runs thus Nam quasi in orbem quemdam per diversas
:

scholas milites promoventur, ita ut ex prima cohorte ad gradum quem-

piam promotus vadat ad decimam cohortem ; et rursus ab ea, crescen-
tibus stipendiis, cum maiore gradu per alias recurrit ad primam!^ But,

^ Rim. Staatsverwaltung, ii,

1884, pp. 368-72. ^ See, however,
p. 576.
^ Caesar, ed. Kraner-Dittenberger, 15th ed., pp. 48-9.
* As Bruncke
(p. 14) justly remarks, if the primi ordines had been the ten
leaders of the ten cohorts, there would have been no room in the legion for octavi
ordines, seeing that there were only 60 centurions in all. And, he adds, if Caesar
had meant to convey that Scaeva was octavus pit us prior, he would have written
not ab octavis ordinibus but ab octavo pilo priore.
5 Eph. Epigr., iv, 230, n. 2. « W. Smith's Diet,
of Ant., i, 799.
' Soldiers, as they advance in rank, proceed as it were by rotation through
'

the different grades [of the several cohorts], in such a manner that one who
is promoted passes from the first cohort to the tenth, and returns again regularly
through the others ... to the iirst.' Clarke's translation, p. 77 (ii, 21), modified.
572 THE CENTURIONS
as Dr. Purser observes, '
this only means that a common soldier
of the 1st cohort, if advanced to be a centurion, begins at the bottom
of the centurions of the 10th cohort and works his way up.' ^ That
Dr. Purser is right is proved by another passage in which Vegetius
says that the primus princeps was regularly promoted to the post
of primipilus?- Except in rare instances, such as that of Scaeva, when
a man was promoted over the heads of his fellows to the post of chief
centurion, or when casualties or disease had caused several simul-
taneous vacancies, the officer who was chosen to succeed to this post
must have been the 2nd centurion of the legion. Therefore, if
Vegetius was right, the 2nd centurion of the legion was the 2nd
centurion of the 1st cohort. But on the theory of Marquardt the
2nd centurion of the legion was the 1st centurion of the 2nd cohort.
Von Goler^ tries to wriggle out of this impasse by insisting that
Vegetius was only speaking of the earlier period when the tactical
unit of the legion was not the cohort, but the maniple. This is a mere
assumption and an absurd one for Vegetius goes on, in the same
;

chapter, to describe the organization of the 1st cohort when it was

twice as strong as any of the remaining nine, that is to say, as it
existed in the time of Hadrian, if not later. Moreover, the statement
of Vegetius is supported by three inscriptions,* two of which show that,
in the time of Augustus, a primus princeps became primus pilus ;

while a fourth ^ shows that a primus hastatus was promoted to the

rank of primus princeps.
There is yet another objection to Marquardt's theory. I have
shown that the 1st cohort, even in Caesar's time, ranked above all
the other cohorts in the legion. That being the case, is it conceivable
that the 6th centurion of the principal cohort should have been only
the 51st in the legion ?

3. Von Goler's scheme ^ embraces twelve classes, the first of which

comprised the ten pili priores the 2nd the ten pili posteriores
;
;

the 3rd the ten principes priores and so on down to the 6th. The last
;

six classes are composed of sub-centurions, or optio7ies, the 7th class

consisting of ten sub-centurions serving respectively under the ten
pili priores, and so on.
This theory must also be rejected, not only because its arrangement
of the primi ordines is identical with that of Marquardt, but also
because, although it does provide for the necessary eight classes and
more, it only does so by including optiones. Now although there is
evidence that an optio might rise to be a centurion,'^ optiones were not
centurions, and Caesar never mentions them. Von Goler, indeed,

^ Diet, of Ant., i, 799. See also Bruncke, pp. 15-6.

^ Vetiis consuetude tenuit ut ex primo principe legionis promoveretur
centurio primipih. ii, 8. Cf. Jahresb. d. philol. Vereins zu Berlin, 188(), p. 250.
=»
Gall. Krieg, 1880, ii, 228.
^ Inscr. regni Neapolitani (ed. Mommsen), 5712, and Corpus inscr. Lot., viii,

2768, 2941. See Bruncke, pp. 16-7.

^ Eph. Epigr., iv, 231. Cf. Bruncke, p. 16.
« Gall. Krieg, 1880, ii, 222-6.

' (Jorpn,'^ inscr. Lat.. iii, 3445. See also v. 7004, and Bonner Jahrhiicher,
cxvii, 1908, pp. 43, 95.
 OF THE FIRST RANK 573

argues tliat Caesar miglit have spoken loosely of centuriones when he

meant we speak loosely of colonels when we mean
optiones, just as
lieutenant-colonels. But this is an unwarrantable assumption. If
the optiones formed the last six classes, Caesar must have been
thinking of them when he wrote of the centurions who had risen
ab inferiorihus ordinibus and this is more than unlikely.
;

L. Miiller^ thinks that the primi ordines were only three,

4.
namely primus pilus, primus princeps, and primus hastatus. This
theory is irreconcilable with the implied statement of Tacitus that
there were not less than six primorum ordinum centuriones in the
legion.
5. F. Giesing groups the centurions of each legion in three classes,
the 2nd and 3rd of which have each three subdivisions. The 1st
class consisted of the three primi priores, —
the primus pilus prior,
primus princeps prior, and primus hastatus prior. These three were
the primi ordines. The 2nd class consisted of all the remaining
priores, the 1st subdivision comprising the jnli priores of the nine
cohorts bslow the 1st, the 2nd the principes priores and the 3rd the
hastati priores of the same nine cohorts. The 3rd class consisted of
all the thirty posteriores, the first subdivision comprising the ten jnli
posteriores, the 2nd the ten principes posteriores, and the third the
ten hastati posteriores.^ As regards the primi ordines, this view is
identical with that of L. Miiller.
In support of his theory Giesing cites a well-known passage in
which Vegetius^ says that the 1st cohort comprised ten centuries,
which were officered by five ordinarii that in the same cohort there
;

were also ten centurions, each of whom had charge of one century;
that each of the remaining cohorts had five centurions and that ;

there were fifty-five in the whole legion. Giesing concludes that the
five ordinarii and the ten other centurions did not constitute a
'
closed rank-class that the former were the primi ordines
'
; and ;

therefore that their prototypes, who, he assumes, were the three

primi priores of Caesar's time, were also the primi ordines. I do not
believe that any conclusion can be safely drawn from this passage.
To begin with, there were, as we have seen,* not five only, but six
centurions in the 1st cohort under the Empire ^ and there were ;

not five but six centurions in each of the other cohorts. Moreover,
according to Vegetius, two of the ordinarii commanded each one
century and a half, and yet the ten centurions commanded each one
century. Well might Mommsen call the passage a locus perturbatus.^
If there is any truth in it, the ten centurions of the 1st cohort were
certainly inferior to the ordinarii even if they were not optiones.
But this tells against Giesing' s theory, not in its favour. For in

^ De re 7nil. Rom. quaedam e Caesaris comm. excerpta, p. 9.

^
Nem Jahrb. f. Philologie, &c., cxlv, 1892, pp. 495-503.
^ Deremil., ii, 8. * Sec
p. 568, nn. 4-5.
^ AsA. von Doniaszcvvski shows {Bonner Jahrb., cxvii, 1908, pp. 91-2),
Moiiimsen {Eph. Epi(jr.,iY,221-8) was wrong in concluding from Corpus inscr.
Lai., viii, 18,072, that in Hadrian's time there were only live centurions in the
1st cohort. 8 Corpus inscr. Lat., v, 8295.
574 THE CENTURIONS
Caesar's legions there were also optiones and in Caesar's legions
;

the 1st cohort was officered by six centurions. Therefore the five
ordinarii, who commanded all the centurions of the 1st cohort, corre-
sponded, not with the first three centurions of the 1st cohort of
the Caesarian legion, but with all six centurions of that cohort.
Therefore, if the five ordinarii were the primi ordines, so were the
six centurions of the 1st cohort of the Caesarian legion.
Giesing does not ignore the passage in which Tacitus says that sex
primorum ordinum centuriones of one legion were killed in a single
battle, or the passage in which Scaeva was said to have been pro-
moted ah octavis ordinihus but he believes that both these passages
;

can be reconciled with his theory. Like Madvig,i he holds that

Tacitus was not speaking of the prmii ordines as a definite class,
but used the phrase in a wide sense in other words, that by pri-
;

morum ordinum he meant superiorum ordinum. But to explain away

is not to explain'; and I hold that we are bound to believe that
Tacitus meant what he said. Ciesing's explanation of the passage in
Caesar is at least ingenious. Octavis ordinibus, he says, refers not to
the 8th cohort of Scaeva' s legion, but to several legions Scaeva :

'
' —
was the leader of one of the 8th centuries (Zug) that is to say, of
the 2nd century of the 2nd cohort. In other words, he was secundus
princeps prior, or, on Giesing's theory, the 13th centurion of the
legion. How this meaning is to be dragged out of the Latin, I am
unable to discover. And- even if Giesing is right, even if Caesar used
such a ridiculous expression as from the 8th centuries (of the
'
'

several legions), to only one of which 8th centuries Scaeva could

' '

have belonged, Giesing contradicts himself. For if the 8th century

of a legion was the 2nd century of the 2nd cohort, surely all the six
centuries of the 1st cohort ranked above all the centuries of all the
remaining cohorts, which is what Giesing denies.
6. B. de Launay2 reckons no less than fifteen primi ordines,
namely, all six centurions of the 1st cohort and the chief centurions
of the other nine cohorts. The first three places he assigns to the
primipilus, primus princeps prior, and primus Jiastatus prior after :

them he places the nine pili priores of the nine remaining cohorts ;

and last of all the primus pilus posterior, primus princeps posterior,
and primus hastatus posterior. He maintains that, when Caesar sj^oke
of octavi ordines, he meant the 3rd cohort but that, when he spoke
;

of primi ordines, superiores ordines, inferiores ordines, and infimi

ordines, he designated a hierarchy, the several grades of which did
not correspond with the numbers of the cohorts.
This view, in so far as it agrees with that of Marquardt, is open to
the same objection. Moreover, the reader will have no hesitation in
rejecting it when he has reflected on what I shall have to say of
Riistow's scheme.
7. According to Lange,^ there were seven classes. He holds

^ Kleine philologische 1875, ]3. 515, u. 1.

ticlirijlen,
^ Uordre en halaille el lescenturions a Pepoque de Jules Cesar, 1873, pp. 20-32.
^ Hist, mutalionum rei mil. Bom,, pp. 20-2.
 . —
or THE FIRST RANK 575

that Caesar took over the relations of rank from the manipular
organization, making, however, one change owing to the growing
:

importance of the 1st cohort, he placed its centurions in a class by

themselves. Like Eiistow, he identifies the "priini ordines with the six
centurions of the 1st cohort the 2nd class he identifies by a cross
:

division with the nine joili priores of the nine remaining cohorts the ;

3rd with the nine fili posterior es and so on. I agree with his
;

identification of the frimi ordines : the other part of his scheme is

open to the objections which are fatal to the schemes of Marquardt

and von Goler.
8. Riistow,! who
followed by F. Frohhch,- believes that there
is
were ten classes of centurions, each composed of the six centurions
of a single cohort and consequently that the primi ordines were the
;

six centurions of the 1st cohort.^ This theory clashes with none of
the proved facts but it is open to two objections. First, it provides
;

only for six centurions of the first rank in each legion. Now Tacitus,
as we have seen, says that in the battle between Antonius Primus
and Vitellius, six centurions of the first rank were killed in the
7th legion alone. On Riistow's theory, therefore, unless the number
of priini ordines had bsen increased between the time of Caesar and
the time of Galba,^ all the centurions of the first rank in one legion
were killed in one battle. But, however improbable this may seem,
it is quite possible. Caesar's famous battle with the Nervii is a
parallel case. Speaking of the 12th legion, Caesar writes, quartae
cohortis omnihus centurionibus occisis reliquarum cohortium omnibus
. .

fere centurionibus aut vulneratis aut occisis.^ Still, even if we admit

that all the centurions of the first rank in a single legion may have
been killed in a single battle, would not Tacitus have written not sex
but omnes (primorum ordinum centuriones) ? ^ The two difficulties
combined appear to me almost insuperable unless we assume that —
Tacitus made a mistake. But this assumption is not extravagant.
The most accurate historian, unless he had access to carefully com-
piled returns, might blunder on such a point.^
The other objection has been stated by Marquardt.^ Suppose that
a centurion had reached the rank of secundus pilus prior that is, —
of 1st centurion of the 2nd cohort. He would then, according to
Heerwesen mid Krieyfilhrung C. J. Cdsars, 1855, pp. 8-10.
^

Das Kriegswesen Giisars, pp. 23-8.

^

^ This theory is nowadays generally

spoken of as Riistow's, but Lij)sius
{Opera, 1637, iii, 52-3) advocated it nearly three centuries ago.
See PJiilologus, xxxviii, 1877, p. 142.
*

^ B.
G., ii, 25, § 1. Cf. B. G., iii, 64, § 4, where we read that live centurions
of the 1st cohort of the 9th legion were killed in one engagement, aquila
conservatur omnibus primae cohortis centurionibus interfectis praeter principeiii
priorem.
* 1 find that Madvig has made the same remark. Kleine philoloyische
Schriffen, p. 515, n. 1.
Assuming that there were only six centurions of the first rank in each
'

legion, Tacitusmight perhaps have been ignorant of the fact, and accordingly
have written sex instead of omnes. Many a modern historian would be puzzled
if he were asked how many captains there are in a regiment.
^ Eom. StaalsverwaUung, 1884, pp. 371-2.
576 THE CENTURIONS
Marquardt, have commanded the cohort. But, according to Riistow,
on his next promotion, he would, supposing that he only gained one
step, become fvimus hastatus 'posterior that is to say, while rising
;

to the coveted grade of primi ordines, he would sink from the position
of commander of a cohort to that of centurion of the lowest maniple
of another cohort. But this objection may, I think, be satisfactorily
answered. First, it is certain that, little more than a century before
Caesar's time, centurions were sometimes called upon to serve in
grades lower than those to which they had attained in previous
campaigns. Livy^ relates that in 171 B.C. 23 centurions qui primos
piles duxerant appealed to the tribunes of the plebs against having
to serve on such terms, but that they were induced to desist from
their appeal. He also states that in 341 B.C. a law was passed ne
quis, uhi tribunus ynilitum fuisset, postea ordinum ductor esset. Of
course I do not mean to say that the appointment of a centurion, on
public grounds, to a post lower than one which he had previously
filled, is analogous to a system of promotion which might involve
descent from the command of a cohort to the command of a maniple.
Still, the facts which Livy relates tend to show that such a system of
promotion would not have been as startling to a Roman as it sounds
to us. Secondly, even if we must infer that the pilus prior of a cohort
commanded that cohort, no ancient writer mentions that the cohort,
as such, had any commander at all groups of cohorts, as I have
:
'^

shown in a former note, were commanded by tribunes ^ and it is ;

probable that the pilus prior of a cohort was more highly esteemed
as pilus prior than as ex ejfficio commander of the cohort. Lastly,
whatever may be the force of Marquardt's objection to Riistow's
scheme, it cannot be sustained unless we accept Marquardt's own
scheme or some one of the other schemes which have already been
examined and condemned.
There are, on the other hand, very strong arguments in favour of
Riistow's scheme. Mommsen, in the article from which I have
quoted, agrees with the theory of rank on which it is based, that is
to say, he regards each and every centurion of any cohort as superior
to all the centurions of the cohort or cohorts below.'* His reasoning
amounts in brief to this. The 1st cohort ranked above all the other
nine, certainly in Hadrian's time, and probably long before. It is
therefore simply incredible that the 2nd centurion of the 1st cohort
should have been, as Marquardt tries to make out, only the 11th
centurion of the legion. By M^ay of further proof, Mommsen adds
that the bulk of the inscriptions referring to centurions which throw
light upon the matter contain the names of primi principes and
primi and that very few inscriptions mention the cohortes
hastati,
Moreover, putting aside a passage in B. C, iii, 64, § 4,
poster iores.
and two inscriptions (numbered 49 and 56 in his article), he shows
that all centurions except the first three of the 1st cohort are
1 xlii, 32-5.
^ Cf. Diet, of Ant., i, 709. Bruiicke (p. 15) argues that the cohort no more
had a apecial commander than the maniple had had in earlier times.
'•"

See p. 50(5. ' Eph. Epujr., iv, 229, n. 1, 230-1, 235.

 OF THE FIRST RANK 577

desigiititcd by the numbers of their respective cohorts and the grades

which they held therein, for example ce7iturio legione 111 Cyrenaica,
cohorte V princeps posterior and from this fact he concludes that
;

the first three centurions of the 1st cohort were distinguished above
all their fellows. Finally he refers to an inscription (numbered 50 in
his article) from which it appears that a centurion named Modestus,
after serving for eighteen years in four grades of rank, held the
position of hastatus posterior in the 3rd cohort. It is incredible, he
argues, that Modestus, after such a long service, should have been
only the 53rd centurion of his legion but he may well have been
;

the 17th, as he would have been if all the centurions of any one
cohort had ranked above all the centurions of the next.
Giesing objects to Riistow's theory that it compels us to assume
that the cohorts in the third line of the acies triplex the army —
—
formed in order of battle were commanded by the youngest and
least experienced centurions. What an idiotic expenditure,' he
'

exclaims, of the best materials in the first line at the cost of the
'

reserve.' ^ To this I reply first, that the reserve would sometimes

not come into action at all secondly, that the cohorts of the third
;

line, or rather groups of those cohorts, were commanded by tribunes,

who sometimes acted under the orders of an experienced legatus ;

thirdly, that the younger centurions must have been in command

somewhere, and that it would be more natural to look for them in
the reserve than anywhere else and fourthly, that there is no reason
;

to suppose that the younger centurions were inefficient, while there

is abundant evidence that they did their work thoroughly well.
Again, unless we are to accept von Goler's theory or Giesing's,
both of which have been shown to be absolutely inadmissible, octavis
ordinibus can only mean the six centurions of the 8th cohort.^ The
conclusion is irresistible that the six centurions of the 1st cohort were
primi ordines.
I have already, in discussing Marquardt's theory, adduced evi-
dence from Vegetius and from inscriptions, which proves that the
primus pilus prior, primus princeps prior, and priynus hastatus prior
were the first three centurions of the legion, and therefore primi
ordifies. If there were any others, those others must, as every scholar
admits, have been the 4th, 5th and 6th centurions of the 1st cohort,
or the nine chief centurions of the nine remaining cohorts, or both
these two groups. But if the first three centurions of the 1st cohort
ranked above all the centurions of all the other cohorts, it seems
logical to infer that the 4th, 5th and 6th centurions of the 1st cohort
did the same and if so, why should not all the centurions of the
;

2nd cohort have ranked above all the centurions of the remaining
eight cohorts, and so on ? These arguments complete the proof that
the primi ordines were the centurions of the 1st cohort.
9. One other scheme, however, appears possible. It would be
identical with that of Riistow, except that any centurion who had

^ Neue Jahib. f. Fkilulugie, &c., cxl\^, 18U2, p. .1:98.

^ !So also argues Brunckc, p. 14.
10«J3 p p
578 THE CENTURIONS
once filled the office of primipilus or had belonged to the 1st cohort,
would rank with the primi ordines. AVe have already seen that
a little more than a century before Caesar's time a centurion who had
been the first of his legion might be called upon afterwards to serve
in a lower grade. May we explain this apparent anomaly by sup-
posing that it might sometimes have been advisable to place a cen-
turion of proved capacity in command of a maniple or century of
low rank and consisting mainly of raw recruits ? It seems possible,
at all events, that the same liability existed in Caesar's time. Other-
wise, how is one to explain the following passages \^Tum T. Balventio,
qui superiore anno priynum pilurn duxerat, viroforti et macjnae
auctoritatis} &c., and (perhaps also) Puhlius Sextius Baculus, qui
primum pilum apud Caesarem duxerat,^ Scc^i It will bs replied
that Balvenfcius and Baculus were evocati in other words, that they
;

had completed their term of service, and were serving again as

volunteers.^ This is not absolutely certain but, assuming the truth
;

of the conjecture, is it likely that two centurions who had been the first
in their respective legions and who belonged to the highly privileged
evocati, should have ceased to rank with the primi ordines ? Caesar
calls Balventius a man of commanding influence
'
(vir magnae '

auctoritatis) and Baculus, the Hector Macdonald of the Roman

;

army, was the most distinguished of all his centurions. Both men,
if evocati, must still, like the volunteer, Crastinus, at the battle of

Pharsalia, who had, in the previous year, been the chief centurion
of the 10th legion, have held positions of trust whether they com-
:

manded other evocati, who, as at Pharsalia, may have been dispersed

among the ranks to stiffen them, or were employed in some other
' '

way, matters little. My suggestion is that centurions qui primos pilos

duxerant or who had balonged to the 1st cohort and had, whether as
evocati or not, bsen appointed on public grounds to some other com-
mand, probably still ranked with the primi ordinesA If so, there
might sometimes have been more than six centurions of the first rank
in a legion. At all events, Riistow's scheme, or the modification of
it which I have suggested, is the only one that agrees with all the
known facts, and is open to no real objection.
One word more. It is certain that those who reached or approached
the rank of primi ordines were not obliged to pass through each and
every grade in the list of centurions for if that had been the case,
;

the primipilus would have had to climb no less than 59 successive

steps before he reached the top of the ladder. No doubt the majority

1 B. G., V, 35, § 6.
''
lb., vi, 38, § 1.
^ Cf. tSchneider's Caesar, ii, 127. Long
{Caesar, p. 252) suggests that Caesar
may mean that Balventius had only received his promotion in the preceding year
and was still primipilus. I believe that if Caesar had meant this, he would have
written qui superiore anno ad primipilum promotus erat. Cf. B. C, i, 40, § 4.
Moreover, in B. C, in, 91, § 1, he says that Crastinus, qui superiore anno apud
eum primum pilum duxerat, was an evocatus.
* I find that the author of the article Evocati in Darembcrg and Saglio's Diet,

des ant. grecques et rom., ii, 867, remarks that il est bien difficile de supposer
'

qu'ils [primi pili] aient accepte, en reprenant du service, un rang infericur

a celui qu'ils avaicnt en le quittant.' Cf. Eph. Epigr., v, 1884, p. 143.
 — ;

OF THE FIRST RANK 579

of centurions never reached the rank of frimi ordines at all. Some-

times several steps of rank may have been gained at a time, owing
to vacancies caused by death or superannuation. Sometimes, as in
the case of Q. Fulginius ^ and Scaeva,^ an officer of exceptional merit
may have risen at one bound from nearly the lowest to one of the
highest or even to the very highest rung of the ladder. Anyhow it
seems tolerably certain that the primipilus, in the course of his
career, had seldom served in more than six or seven grades. To quote
Mommsen once more, It is clear that the conditions of service were
'

such that a man would reach the rank of frimipilus before his fiftieth
year, and generally after passing through six or seven grades of rank.
I am aware that many scholars nowadays are of opinion that a cen-
turion had regularly to pass through the whole of the sixty grades
of rank. But this view is at once refuted by the inscriptions, which
mention a much smaller number of grades, and is condemned by
comnion sense for who would allow himself to be convinced that
;

a centurion only remained some six months in each grade ? ^ '

THE FABRI
Some writers assert that the fabri, in Caesar's time, still formed
a separate corps * but there is no evidence for this statement
: and ;

it is refuted by a passage ^ in which Caesar says that, in order to

repair the ships which had been damaged on his second expedition
to Britain, he selected /a^ri from the various legions.^

CAESAR'S CAVALRY
Caesar's cavalry consisted entirely of foreigners,^ —
Gauls, Spaniards,
and Germans. They were organized in alae or squadrons of from 300
to 400 men, divided into turmae or troops, and were commanded by
praefecti equitum, who were often their national chiefs.^
It is commonly asserted that legionary cavalry, that is to say
cavalry organized as permanent corps and attached to the legions,
no longer existed in Caesar's time and that this arm w^as not revived
;

before the imperial epoch .^ Nipperdey, however, maintains that this

view is erroneous. In Bellum Africanum, 51, § 7, 52, § 2 the follow-
ing passage occurs Dum haec opera, quae ante dixi, fiehant a
:

legionibus, interim pars acie ante opus instructa sub hoste stabat
equites barbari levisque armaturae proeliis minutis comminus
1 B.C., i, 40, § 4. ' Ih., iii,
53, §§ 4-5. ^ Eph. Epigr., iv, 235-6.

* E.g. Long, Decline of the Roman Republic, ii, 11).

5 B.G.,Y, 11, §3.

^ See Darenibcrg and Saglio, Diet, des ant. grecqiies et rom., ii, 957-8;
Frohlich, Das Kriecjswcsen Camrs, pp. 51-3 and Pauly's Real-Encyclopddic, vi,
;

1909, col. 1919-20. ' See, however,

p. 581. » B. G., viii,
12, § 4.
" See, for instance, P. Geyer in Jahresh.
d. philol. Verelns zic Berlin, v, 1879,
p. 345 ; and C. C. L. Lango, Hist, mutatloniim rei mil. Rom., p. 13.-
P p 2
580 CAESAR'S CAVALRY
dimicabanl. Caesar ah eo opere cum iam sub vesperunt copias in caslra
reduceret, ntagno incursu cum omni equitatu levique amialura luba,
Scipio, Labienus in legionarios impetumfecerunt. Equites Caesa-
rian i vi universae subitaeque hostium muUitudinis pulsi parumper
cesserunt. According to this passage, as it stands, the equites Caesa-
riani were apparently identical with the legionarii, though StofEel,^
if I do not misunderstand him, takes legionarios as meaning the '

legions followed by the cavalry '. Da vies, however, remarking that

legionarii properly means legionary infantry ', concluded that the
'

words in legionarios impetumfecerunt were corrupt and R. Schneider ;

in his edition of Bell. Afr. reads (in legionarios impetum fecerunt)

equites. Caesariani, &c. but Nipperdey,'^ who accepts the text,
:

remarks that legionarios must mean legionarii equites, as opposed to

equites barbari levisque armaturae. He points out that, according to
Plutarch,^ there were legionary cavalry in the army of Antonius ;

and Frohlich ^ quotes a passage from Appian,^ froiii which it should

seem that there was a similar force in the army of Pompey. Madvig ^
and 0. Schambach endorse the argument of Nipperdey
'^
and ;

Schambach undertakes to prove from the Commentaries that in the

Gallic War also a permanent corps of cavalry was attached to the
legion. He quotes three passages, from B. G., v, 2, § 4, 5, § 3, 8, § 1.
In the first passage Caesar says that he marched (in 54 B.C.) for the
country of the Treveri with four legions and 800 cavalry (ipse cum
legionibus expeditis IV et equitibus DCCC in fines Treverorum pro-
ficiscitur) in the second he says that the cavalry of Gaul, to the
:

number of 4,000, assembled at the Portus Itius before he set sail for
Britain (eodem equitatus totius Galliae convenit numero milium I V) ;

and in the third he says that when he set sail, he left Labienus
bahind with three legions and 2,000 cavalry, while he himself took
the same number of cavalry and five legions (Labieno in continente
cum III legionibus et equitum milibus duobus relicto ipse cum . . .

V legionibus et pari numero equitum quem in continenti relinque-

bat . naves solvit). From the first two passages, says Schambach,
. .

one would expect to find that Caesar had 4,800 cavalry, not 4,000
only, to divide between himself and Labienus. It is clear that none
of the 4,000 Gallic cavalry took part in the expedition against the
Treveri. It follows that Caesar immediately before his embarkation
nmst have had at least 4,800 cavalry under his command. Since he
ignores the odd 800, we may gather that their connexion with the
'

legion was already firmly established and was taken for granted.' ^
Guischard,'^ however, says that Caesar's account of his interview
with Ariovistus proves that at that time (58 B.C.) legionary cavalry
no longer existed, as he had none to form his escort, and was obliged
^ Guerre civile, ii, 129. ^ Caesar, p. 216.
^ Anton., 37. * Das Kriegswesen Casars, p. 38, n. 7.
^ B. C, ii, 49. Ho^iTTjio) b( TTiure /.ilu [re\j/] l£ 'iraXias . . . Kal tovtois oaoi
ovi'iTaaaovTo innus. Kleine 2)hilol. Schrijten, p. 502, note.
**

' Die Reiterei hei Caesar, 1881 Biirsian'a Jaliresb. ilher d. Fortschritte d.
;

class. AUerthionsivissenscJutft, xxxvi, 1883, pp. 258-9.

^ In B. C, iii,
29, § 2, vvc find 800 cavalry mentioned in connexion with four
legions. ^ 3Iem. cril. el hist., iii, 310.
 CAESAR'S CAVALRY 581

mount the soldiers of the 10th legion on the horses of his Gallic
to
cavalry, to whom he dared not entrust his safety. The fact may
prove that there were no legionary cavalry in this, the first year of
the Gallic War but I am not sure that it even proves this
: for, as ;

Caesar's escort numbered 4,000, the legionary cavalry, if there were

any, would have been far too few.
Nevertheless, it appears to me that Schambach has failed to prove
his case for it is by no means clear that the 4,000 cavalry who
;

assembled at the Portus Itius did not include the 800 who had
accompanied Caesar to the country of the Treveri and he did not ;

'
take for granted the existence of those 800 when he described the
'

expedition on which they accompanied him, but thought it neces-

sary to mention them expressly. Apparently it was his practice to
raise his Gallic cavalry at the outset of each campaign and the ;

passage in which he says that he assigned Quintus Cicero 200 horse

when he left him with the 14th legion at Atuatuca ^ seems to imply
that he had no legionary cavalry in Gaul.^ Since, on the other hand,
it is probable that his Gallic cavalry, or the bulk of them, used to
return home at the close of each campaign, the Spanish and German
cavalry, who remained with the legions, may possibly have been
brigaded with them.
In the seventh campaign some of Caesar's cavalry may have been
Italian. After he left Brutus in the country of the Arverni he hastened
to Vienne, and there picked up his cavalry, which he had sent on
'

a considerable time before, in good condition '.^ This corps had cer-
tainly not been raised from the independent Gallic tribes, many of
whom were already in revolt nor were they part of the cavalry
;

which he had had in the previous campaign, for he had not yet
succeeded in opening communication with his legions. It is there-
fore possible that he had raised them in Italy ,^ foreseeing that he
would not be able to complete his usual levy in Gaul but probably
;
^'

they belonged to the Province.^

Guischard says that Caesar had raised the 400 German cavalry
'^

whom he employed against Vercingetorix in the combat at Novio-

dunum, in the first year of the war. Caesar speaks of them as equites
. .quos ah initio secum habere instituerat? As Long ^ says, it is
.
'

not clear what he means by " ab initio " but it is quite clear that
'
:

he did not employ these troops in his first campaign and we may ;

be sure that he did not raise them until after he had learned their
value in his campaign against Ariovistus. Either ah initio means
'
from the outset ' (of the seventh campaign) or, more probably, it
is used in a loose sense, —
from the time when he first employed
German cavalry, recognizing their value.

1 B.O., vi, 32, § 6.

2 5 ; iv, 0, § 5 ; 7, § 1 ; 35, § 1 ; and Bonner Jahrhucher, cxiv^-
See ib., i, 42, §
cxv, 190G, p. 182. 3 B. G., vii, 7-9. * Cf. ib.,
65, § 4.
B, G., vii, 13, § 2, however, proves that Caesar was not without Oallie
'^

cavalry even in the first stage of the seventh campaign.

« Cf. ih.,1,
§ 5. ' Mem. crit. d hist., p. 305.
« B. G., vii, 13, 9 Caesar, j). 338.
§ 1.
 582

CAESAR'S ARTILLERY
Neither ballistae nor catafultae are ever mentioned in the Gallic
War : but both are perhaps included under the generic name of
tormenta, which Caesar mentions often and, as that name suggests,
;

both derived their power from the recoil of tightly twisted cordage.^
How the Roman engines were constructed, we are not told but it ;

is generally assumed that they were identical or virtually identical

with those of the Greeks. The latter are described by Rudolf
Schneider, the highest authority, in Pauly's Beal-Encyclopadie, vii,
1910, col. 1297-1322, with which compare Schneider's monograph
in Erg.'inzungs-IIeft zum Jahrbiich der Gesellschaft fiir lothringsche
Geschichte, &c., ii, 1907.
As the Greek engineer, Heron,^ explained, the idea of both was sug-
gested by the ordinary long bow. Roughly speaking, they resembled
huge crossbows, the main difference being that instead of one bow there
were two arms, connected by a rope which formed the bowstring ;

and both catapult and ballista could discharge either stones or

feathered javelins.^ It is impossible, however, to state exactly what
the difference between them was.'* Dictionaries of antiquities explain
that they were identical in principle, but that the pipe or groove of
the ballista, which directed the flight of its projectile, was inclined
at an angle of 45° to the ground, whereas the catapult could shoot
either point-blank or at any required angle. But Caesar^ relates
that at the siege of Massilia the defenders used ballistae which dis-
charged javelins with such force as to penetrate the vineae, or sheds,
that screened the besiegers and it is obvious that these engines
;

could not have been used with effect unless their trajectory could
have been altered at will.
Ammianus Marcellinus ^ describes an engine, called the onager,
which differed radically both from the catapult and from the ballista.
Unlike them, it had nothing in common with a bow it had no
:

groove to direct the flight of its missile, which it necessarily cast at

a high elevation ;and it could only discharge stones. Its principal
feature was one enormous arm or beam, mounted on a frame-work
of timber, and carrying a sling, in which was placed the projectile.
Before the sling was loaded the arm was drawn backwards and held
by a catch when this was withdrawn the released arm flew forward
:

and the missile sped on its path.

Caesar, in his description of the siege of Avaricum, mentions
a piece of artillery called scorfio^ which was used for shooting at

1 Sir R. Payne-Gallwey, The Crossbow, ^q., 1903, p. 250.

2 Ed. C. Wescher, 1867, p. 75.
^ Polybius, xi, 11,
§ 3; Caesar, B. C, ii, 9, § 3 ; Diodorus Siculus, xx, 48,
§ 3 ; Vitruvius, x, 11, § 3 ; Josephus, Bell, hid., iii, 7, 9, § 167 ; Tacitus,
Hist., iv, 23.
* See Paulv's Beal-Encyclopadie, vii. 1309. The word ballista does not exist
' B. C, ii, 2,
in Greek. ^
§§ 1-2.
« xxiii,
4, § 5. ' B. G., vii, 25, §§ 2-3.
 '

CAESAR'S ARTILLERY 583

individuals. Polybius,i says that in the siege of Syracuse scorpions '

were used in conjunction with archers by Archimedes for picking off

marines. Livy^ distinguishes scorpions from catapults and bal- '
'

listas the writer of Bellum Africanum^ and Vitruvius* from cata-

;

pults. Vegetius^ calls the 'scorpion' a hand-ballista and says ' '

that it shot small arrows while Ammianus Marcellinus ^ affirms

;

that it threw stones and identifies it with the onager According to !

Heron it was a catapult. As Caesar's scorfio discharged missiles

'^

which pierced individuals in rapid succession, I conclude that it

was, so to speak, a quick-firing weapon with two arms, or what
Livy s calls a scorpio minor. ^ That it was identified both with the
catapult and the ballista is easy to understand, when we consider
the general resemblance between those weapons. It need hardly
be added that Ammianus was mistaken for how could a man have ;

been pierced by a stone, and how could the most skilful artillery-
' '

man have taken aim with an onager at so small a mark ?

I have shown in my narrative (page 65 and note 2) that Caesar
occasionally used, or was prepared to use, artillery in the field as
well as in the defence or attack of fortified places.
The Roman weapons were probably inferior to their Greek models ;io
but the latter are said to have attained ranges which would be
hardly credible if Sir Ralph Payne-Callwey had not verified them
by experimenting with models of his own construction. Josephus^i
states that at the siege of Jerusalem engines threw stones weighing
a talent, or about 57 pounds, more than 400 yards and, according ;

to Agesistratus ^^, a range of 4 stades over 800 yards was occasion- — —

ally reached. Sir Ralph Payne-Gallwey has himself succeeded
with an onager which he describes as much smaller and less '

powerful than the best ancient machines in hurling a stone ball

'

'.^^
weighing 8 pounds '
from 450 to nearly 500 yards

I
viii, 7, § 6.
•'
xxiv, 34, § 9 ; xxvi, 47, §§ 5-6.
' ' X, 15, § 4.
31, § 6.
5 ii, 13 ; iv, 22. « xviii, 7, § 7 ; xxiii, 4, §§ 4-7 ; xxxi, 15, § 2.
7 Ed. C. Wescher, p. 74. « xxvi, 47, § 6.

^ Mr. Judson's illustration of the scorpio {Caesar's Army, p. 24) appears to

represent a crossbow, which, so far as we can tell, was unknown in ancient

warfare.
'" Caesar's artillery was evidently no match for that of the Massiliots {B. C,

ii, 2, § 5 9, § 3) ; and for the siege of Alexandria he was obliged to import

;

engines from Greece or Asia {Bell. Alex., 1, § 1).

II
Bell. lud., V, 6, 3, § 270.
1'^
Quoted by Athenaeus, ed. C. Wescher, 1867, p. 8.
13 Projedile-throiving Engines, &c.,
1907, p. 9. Sir Ralph Payne-Gallwey's
books would have been more useful if he had taken a scholar into partnership.
Such value as they possess is due to his having made practical experiments
which illustrate the power of ancient artillery.
 584

THE CLOTHING AND DEFENSIVE ARMOUR OF

CAESAR'S REGULAR INFANTRY
I. Daremberg and give a woodcut, after Trajan's column,
Saglio ^

representing soldiers wearing hraccae, which resembled tightly-fitting

drawers and reached down to the middle of the calf. There is no
direct evidence that they were worn by Caesar's troops but they ;

appear to have been part of the equipment of Roman soldiers gener-

ally when they were serving in the comparatively cold climates of
Central Europe. Bandages (fasciae) were also worn in the imperial
epoch and I should think that they were probably used, as they
;

are now, in order to prevent or to support varicose veins.

IL Greaves were worn, according to Livy,'^ in the early republican
period and Polybius^ mentions them in his inventory of the arms
;

of the legionaries. Numerous monuments prove that under the

Empire they were worn by centurions ^ but, although Lampridius ^ ;

says that Alexander Severus rewarded deserving soldiers with

presents of greaves, there is no trace on the columns of Trajan and
Antonine that privates wore them,^ nor are they mentioned in the
Notitia Dignitatum'^ in the list of arms which were made in the
imperial factories. Nevertheless, as E. Hiibner^ remarks, we must
not assume, in the absence of express testimony, that their use was
abandoned in the time of Caesar.
Livy, in the passage already referred to, implies that greaves were
worn on both legs and this, according to Lebeau,^ is confirmed by
;

various monuments. Arrian,^^ on the other hand, says that Roman

soldiers wore greaves on their right legs only, because^ when they
were fighting, their right legs were advanced, and their left legs pro-
tected by their shields. Vegetius^^ supports this statement and ;

Lebeau infers that the soldiers represented in the monuments were

either foreigners, officers, or gladiators.^^
III. Caesar nowhere mentions the lorica or cuirass but it is sup- ;

posed that in his time legionaries wore cuirasses consisting of bands

of leather covered with metal. On Trajan's column legionaries are
depicted, wearing cuirasses of this kind. According to Polybius,
a metal breast-plate was worn by the private soldiers without any
lorica ;i^ but there is no evidence that it was still in use in Caesar's

^ Diet, des ant. grecques et rom., i, 746, See also Rev. celt., xi, 1890, pp. 3-i-6,
' 3
i, 43, § 2. vi^ 23, § 8.
* Corpus inscr. Lat., v, 3374 ; vii, 90 ; Hermes, xvi, 1881, pp. 304-5. Cf.
PMlologus, xxxiii, 1874, p. 651 ; 1881, p. 249.
xl, « Vita Alex., 40.
^ Daremberg and Saglio, Diet, des ant. grecques et rom., iv, 149.
' Ed. 0. Seeck, pp. 31-3 (c. xi), 144-6 "(c. ix).
^ Hermes, xvi, 1881, pp. 304-5,
^ Mem. de litt. tires des registres de VAcad. Roy. des inscr., &c., xxxix, 1770-2,
pp. 475-6.
^» Ars tactica, 3, § 5. " De re mil., I 20.
^^ Gladiators, however, sometimes wore only one greave (Daremberg and
Saglio, Diet, des ayit. grecques et rom., iv, 149).
" Marquardt (R'6m. Staatsverwaltung, ii, 1884, pp. 336-7) says that the
 CLOTHING AND DEFENSIVE ARMOUR 585

time. Loricae of other kinds are portrayed on monuments on the :

cohmm of Antonine there is one with scales like those of a serpent

{O(opa^ 4>o\lSo)t6?) on the arch of Trajan there is one with
:

plates like a bird's feathers on the column of Antonine there is

:

a representation of a shirt of chain-mail and on a monument pre-

;

served in the museum of Verona a centurion is represented wearing

a cuirass with scales like those of a fish (lorica squa7nata)^ Cuirasses
of this kind and coats of mail were worn in the time of the Republic
as well as under the Empire.
Lebeau remarks that at Dyrrachium Caesar's soldiers improvised
cuirasses of cloth and leather, in order to protect themselves from
the arrows of the Pompeians, which, he says, proves that their
ordinary cuirasses were not missile-proof.^ Perhaps the arrows had
penetrated between the joints of the loricae.

THE RATIONS
According to Cicero,^ Ammianus MarceUinus,* and Lampridius,^
the legionary used to carry rations for 16 or 17 days. The soldiers
of Afranius, indeed, in their retreat from Ilerda, are said to have
carried supplies for 22 days ;^ but the number XXII in the MSS.
has been variously corrected by suspicious editors."^ Rudolf Schneider^
shows that if ordinary provisions for such long periods had been
carried, the men would have been overburdened, and that the rations
in question must therefore have been compressed food, which, as
we learn from Ammianus Marcellinus,^ the Romans knew how to
prepare. It is of course obvious that such rations would only have
been served out in exceptional cases, when, for example, it was un-
desirable or impossible for pack-horses to accompany the column .^^
Josephus 11 speaks of a 3 days' ration and common sense suggests
;

that the amount must have varied according to circumstances.

breast-plate was worn under a cuirass ; but this is a mistake. See Polybius,
vi, 23, § 14.
I
Mem. de lift, tires des registres de V Acad. Roy. des inscr., &c., xxxix, 1770-2,
pp. 465-6, 468 ; Hermes, xvi, 1881, p. 304 ; Daremberg and Saglio, op. cit.,
iii, 1314. Cf. Proc. Soc. Ant., N.S., xxi, 1905-6, p. 135.
^ Omnes fere milites aut ex coactis aut ex centonibus aut ex coriis tunicas

aut tegimenta fecerant, quibus tela vitarent. B. C, iii, 44, § 7.

3 Tusc, ii, 16, § 37. * xvii, 9, § 2.
5 Alex. Sev., 47, § 1. ^ B. C, i, 78, § 1.

' Meusel's Lex. Caes., vol. ii {Tabula coniedurarum, p. 51), and Kiibler's
edition, p. xxiv.
® Jahresb. d. philol. Vereins zu Berlin, xix, 1893, pp. 279-85.
^ xvii, 8,
§ 2. Cf. Thesaurus ling. Lat., ii, 2228, s.v. huccellatiim.
^" See Vegetius, De re mil., i,
19.
II
Bell. lud., iii, 5, 5, § 95.
 586

THE FORTIFICATION OF CAESAR'S CAMPS

It is unnecessary, for the purpose of this book, to describe the in-
terior arrangement of a Roman camp. The only questions with
which I am concerned are the average size, which it is necessary to
know in order to decide whether certain camps discovered on French
soil are really the camps of Caesar with which they have been identi-
and the nature of the
fied, fortifications. On the former point what
we know amounts to this :
—the camp described by Polybius ^ and
intended for 18,400 foot and 2,400 horse was 2, 150 Roman feet square,
or about 106 acres in extent and the camp described by Hyginus,'^
;

and intended for at least 40,000 men, was 2,320 x 1,620 feet or about
86 acres.
Caesar's camps were of two kinds, temporary and permanent.
The former were constructed at the close of each day's march the :

latter were occupied during the winter, and occasionally in the course
of a campaign. Vegetius ^ says that the rampart of the temporary
camp was made of sods of turf, cut in the form of large bricks, or,
when the earth was too loose, of earth simply, and surmounted with
stakes. Permanent camps were, as he observes, of course fortified
more elaborately. When it was necessary to guard against attacks
in force the ditch was 12 feet broad and 9 deep, backed by a rampart
4 feet high, on the edge of which was planted a stout palisade, made
of stakes, which the soldiers regularly carried.^ In another passage
Vegetius ^ says that the ditches were 9, 11, 13, or even 17 feet wide ;

for, he remarks, an uneven number was prescribed by custom (though

probably he was unaware that in this, as in so many other instances,
custom embodied a survival of superstition) and the rampart, which, ;

as in temporary camps, was no doubt compacted of sods, was faced

with fascines or branches of trees, to ensure its stability, and finished
by an embattled parapet. Such a parapet, which Caesar erected
upon the rampart in his line of contravallation round Alesia, is
minutely described in B. G., vii, 72, § 4. The dimensions of the ditch
and rampart of course varied according to circumstances. The camp
which Caesar constructed on the Aisne, and which he occupied for
several days, had a rampart 12 feet high ^ and a ditch 18 feet wide.
It should be noted, however, that when he mentions the height of
the vallum, he means the combined height of the rampart, properly
so called, and of the palisade."^ Guischard,^ referring to Polybius as

1 vi, 27-37, 41.

^ De munitionibus castrorum, § 21 (ed. A. von Domaszewski, 1887).
3
iii, 8. * i, 24. 5 iii^ 8,
^ B. 0., ii, 5, § 6. P. Bial {Chemins, habitations et oppidum de la Gaule au
temps de Cesar, 1864, pp. 214-5) holds that when Caesar mentioned the height
of a rampart, he reckoned not from the level of the ground, but from the
bottom of the trench. He refers to B. C, iii, 63, § 1, which does not prove
his point. Here is the passage Erat eo loco fossa pedum
: et vallnm contra XV
hostem in altitudinem pedum X, tantundemque eius valli agger in latitudinem
patebat. Cf. A. Klotz, Caesarstudien, p. 220, n. 2.
' Seep. 71, n. 1. « Mem. crit. et hist., 1774, p. 468.
 —
FORTIFICATION OF CAESAR'S CAMPS 587

his authority, says that the palisade was generally planted on the
edge of the ditch and at the foot of the rampart but there is nothing
:

in Polybius which supports this view and it is irreconcilable with

;

the description of Vegetius, and, as it seems to me, with common

sense. Polybius gives an interesting description of the palisade, as
^

it was made in his time. 'As soon as they fix their stakes,' he says,
'
they interlace them in such a manner that it is not easy to know
to which of the stems fixed in the ground the branches belong, nor
on which of these branches the smaller shoots are growing. More-
over, it is impossible to insert the hand and grasp them owing to
the closeness of the interlacing of the branches and the way they lie
one upon another, and because the main branches are also carefully
cut so as to have sharp ends.' ^
Hyginus ^ describes two kinds of trenches, -fossa fastigata, of
which both the scarp and the counterscarp were sloping, and/bs^a
funica, of which the scarp was sloping and the counterscarp vertical.
Caesar does not tell us what was the usual form of his trenches but :

as he once mentions a trench, evidently exceptional, both the scarp

and the counterscarp of which were vertical (directis laterihus^), we
may infer that his trenches were usually either fastigatae or punicae ;

and Colonel Stoffel's excavations have proved that they weie fasti-
gatae.^
In conclusion, it should be noted that, although the normal form
of the Roman camps was oblong, they were sometimes, owing to the
lie of the ground, necessarily irregular in outline.^

CAESAR'S ORDER OF BATTLE

Caesar, when about to fight a battle, formed his army sometimes
in two lines {duplex acies), once, for a special purpose, in four (qua-
druflex acies), but generally in three (triplex acies) y The questions
^ xviii, 18 (E. S. Shuckburgh's translation, modified, ii, 217).
2 Commandant J. CoHn{Pro Alesia, 1907, pp. 149-52), after quoting Vegetius,
in so far as Vegetius supports his own conclusions, ignores him when his
testimony is inconvenient. He argues, if I have succeeded in grasping his
meaning, that there was no paHsade on the rampart but his reasoning depends
;

on the assumptions that the distinction commonly drawn between vallum

(the rampart with its palisade) and vallus (the palisade alone) is imaginary,
and that the word vallum (except as the accusative of vallus) does not exist.
Vallus, he says, generally denotes 1' ensemble du retranchement, ou au moins
'

du rempart ', and, when its sense is restricted, le parement exterieur, Vescarpe
'

du rempart' or, in other words, ''Vescarpe qui fait face a I'ennemi.' If the
commandant will turn to Varro, De lingua Latina, v, 24, § 117, he will acknow-
ledge that he was hasty in denying the existence of the word vallum. Cf. A.
Klotz, Caesar studien, p. 220, n. 2.
^ De munitionihus castrorum, * B. G., vii, 72,
§ 49. § 1.
5 See Atlas (Planches 9, 22, 27-8) to Napoleon's Hist, de Jules Cesar.

® Daremberg and
Saglio, Diet, des ant. grecques et rom., i, 950.
^ Von Goler Krieg, &c., 1880, ii, 214-5,267-71), while admitting that
{Gall.
Caesar's army was generally formed for battle in three lines, rejects the obvious
interpretation of the phrases duplex acies, triplex acies, and quadruplex acies.
A triplex acies, according to him, was so called because it contained three
588 CAESAR'S ORDER OF BATTLE
which we must try to answer are (1) what was the normal depth of
each line ; (2)what interval separated each cohort from the one
next to it ; and (3) how was the relief or the reinforcement of the
fighting line effected during a battle ?
1. Von Goler^ believes that the three maniples of each cohort
stood one behind another, and that the two centuries of each maniple
stood side by side, in two ranks. According to this arrangement,
the cohort, that is to say the line, would have been six men deep.
Stoffel 2 considers that this depth would not have been sufficient to
stand the shock of ancient warfare. Frontinus^ says that in the
battle of Pharsalia Pompey ranged his cohorts ten men deep.
Stoffel maintains, I think rightly, that Frontinus mentions this depth
as something exceptional and he concludes that the normal depth
;

of the cohort was eight men. The conclusion appears probable but ;

the point to notice is that if the maniples of each cohort were arranged
in the manner indicated by von Goler, then in the battle of Pharsalia
they could not have been of equal depth. But, argues Frohlich,^ as
they were of equal strength, they must have been of equal depth :

therefore von Gofer's theory must be rejected. It must be remem-

bered, of course, that, in point of fact, the different maniples must
often have been of unequal effective strength. Nevertheless, Froh-
lich's argument holds good for it is most improbable that in every
;

cohort, all along each of the three lines, one maniple should have
been of greater or less depth than the other two. It is then, if not
certain, at least in the highest degree probable that the three maniples
in each cohort were placed side by side ^ and if so, it is probable that
;

the two centuries in each m.aniple were placed one behind the other.
2. The theory, based upon a famous passage in Livy,^ which finds

divisions in the same Hne, namely a right wing {cornu dextrum), a centre {acies
media), and a left wing {cornu sinistrum) ; while a duplex acies consisted only of
a right and a left wing. This theory has, so far as I know, gained no converts
and the common sense of most readers would reject it but it may be worth
:

while to prove that it is false. First, it fails to explain the formation known
as simplex acies, or the single line, which Caesar once employed in Africa, and
which, as the writer of Bellum Africanum (13, § 2) expressly says, included
a right and a left wing. Secondly, the duplex acies which Crassus formed when
he offered battle to the Aquitani {B. G., iii, 24, § 1), had a centre (media acies),
and therefore ought, on von Goler' s theory, to have been a triplex acies.
To quote Mr. Judson, 'it seems plain enough that Caesar used the terms right
wing, left wing, and centre quite as they are used of a modern army ; applying
them in an indefinite way to those parts of a line of battle, but not necessarily
implying distinct divisions under separate commanders ' {Caesar's Army,
pp. 44-5. Compare Frohlich's Das Kriegswesen Cdsars, p. 150). Thirdly, it
is clear from Caesar's accoimt of the battle with Ariovistus {B. G., i, 52, § 7)
that the third line {tertia acies) acted as a reserve, and was therefore not brought
into action unless and until its services were required. Fourthly, we read
{Bell. Afr.y 60, § 3) that on one occasion Caesar's left wing was triplex. 8ee
Neue Jahrb. f. Philologie, &c., Ixxxv, 1892, pp. 214-6, and cf. Frontinus,
Strat., ii, 3, §§ 16, 22, and Caesar, B. C, i, 41, which, by itself, overthrows von
Goler' s theory.
1 Gall. Krieg, 1880, ii, 216-8. ^ Guerre civile, ii, 327.

* Das Kriegswesen Ccisars,

^ ii,
3, § 22. p. 144.
5 Cf. Neue Jahrb.
f. d. klass. Altertum, &C., ix, 1002. p. 187.
6 viii, 8,
§§ 5, 9-13.
 —
CAESAR'S ORDER OF BATTLE 589

favour ill dictionaries of antiquities and handbooks is that, in the

})eriod of the manipular organization, the maniples in each of the
three lines were separated from one another, not only before but
during close combat, by intervals equal to the breadth of a maniple ;

and that the maniples of the second line stood immediately behind
the intervals of the first. The passage in Livy runs as follows :

'
The first line consisted of the hastati, fifteen maniples, separated —
by intervals of moderate size. The hastati began the battle. If
. . .

they failed to overthrow the enemy, they stepped slowly back, the
princi'pes receiving them into the intervals between their companies.
The principes then came into action, followed by the hastati while ;

the triarii remained with their colours. ... If the principes also failed
to win the battle, they fell back gradually from their frontal position
to the triarii hence originated the proverb, applied to men in
:

a difficult position, " the last resource is in the triarii.''^ The triarii^
springing to their feet, and receiving principes and hastati into the
intervals between their companies, quickly closed their ranks, shut
the passages as it were, and, with no hope to fall back upon, in one
compact column fell upon the enemy (prima acies hastati erant, '

manipuli quindecim, distayites inter se modicwn spaiiiwi hastati . . .

omnium primi pugnam inihant. Si hastati profligare hostent non

possent, pede presso eos retro cedentes in intervalla ordinum principes
recipiebant. Turn principum pugna erat ; hastati sequehantur. Triarii
sub vexillis considebant. Si apud principes quoque haud satis
. . .

prospere esset pugnatum, a prima acie ad triarios sensim referebantur.

Inde rem ad triarios redisse, cum laboratur, proverbio increbuit. Triarii
consurgentes, ubi in intervalla ordinum suorum principes et hastatos
recepissent, extemplo conpressis ordinibus velut claudebant vias, unoque
continenti agmine iam nulla spe post relicta in hostem incidebant).
The principal modern exponent of the theory is RiJstow.i From
certain statements in Caesar he inferred that during the later period,
when the tactical unit was the cohort, the formation of the army
in battle was the same. He argues, first, that Caesar once states
that there was an interval between two cohorts standing in the same
line ; secondly, that the expression
'^
cohortes (^isponere ^ proves ' '

that such intervals were usual and thirdly, that Caesar appears to
;

have always regarded it as an evil that cohorts in the same line

should have been huddled together.
None of these arguments has any value. In the first passage
quoted by Riistow * Caesar describes an attack made by a number of
Britons upon a Roman force on guard in front of a camp in process

^ Heerwesen und Kriegfilhrung Clisars, 1855, p. 45. Since the appearance of

Delbriick's article, referred to on p. 591, the tendency of German writers has
—
been to adopt his view, that, after the light troops had done their work,
the closed formation was general.
2 B. G., V, 15,
§§ 3-4.
'
lb., 33, § 1. _
* At illi imprudentibus nostris atque occupatis in munitione castroriun,
. . .

svibito se ox silvis eiccerunt impetuqiie in cos facto qui erant in statione pro
castris conlocati, acritor pugnavcrunt, duabustjue jnissis subsidio cohortibus
a Cacsare .cum hae xi^rexiguo intcrniisso loci spatio inter se constitisscnt,
. .

novo genere pugnae pcrtcrritis nostris. ]}qv medios audacissimc pcrrupej-unt, &c.
590 CAESAR'S ORDER OF BATTLE
of construction. Hesent two cohorts to the rescue and the Britons
;

broke through the very narrow space that separated the two
' '

cohorts. Now, first of all, it is absurd to argue from the fact that
two cohorts, fighting in the same line, were on one occasion separated
from each other by a very narrow space ', that, as a general rule,
'

a large number of cohorts fighting in the same line, in a pitched

battle, were separated from each other by intervals each equal to
the front of a cohort secondly, Caesar may have had some special
:

reason for placing 'these cohorts apart and thirdly, the two cohorts
;

may have issued from two opposite gates of the camp.i

In the second passage Caesar describes the efforts which Titurius
Sabinus made to repel the troops of Ambiorix. The Roman column
was attacked in front and in rear, on right and left, at the same
moment. Sabinus had to arrange his cohorts in such a way as to
repel this attack ;and cohortes disponere simply means to make '

his dispositions ', not necessarily, as Riistow imagines, to arrange

'

the cohorts with regular intervals between them.'

With regard to Riistow's third argument, it is of course perfectly
true that Caesar regarded it as an evil that his troops should be
huddled together but the passages which Riistow quotes only
:

show that the evil was that the soldiers composing each maniple
were so crowded that they had no room to strike they do not show
:

that Caesar considered it necessary to leave intervals between the

several cohorts.
The theory of intervals has been strenuously opposed, in the —
eighteenth century by Guischard, afterwards by Renard, and of late
years by Delbriick, Giesing, Stoffel, and Rudolf Schneider. These
writers indeed admit that, in the period of the manipular organiza-
tion, there were intervals between the maniples, to allow of the free
passage to and fro of the velites or light-armed troops, who opened
the action, up to the moment when close fighting began 2 but they ;

insist that from that moment the intervals were closed, either by the
extension of the ranks in the several maniples, or, as Guischard
thought, by the advance of the maniples in the second line into the
intervals in the first.^ Schneider* also, though he admits that in
Caesar's army the auxiliary troops played a minor part, remarks,
referring to the 41st chapter of Plutarch's Antonius, that even then
intervals were left, before close fighting began, for the passage of
these light-armed troops and he infers that in Caesar's time, as in
;

the period of the manipular organization, the intervals, after they

^ See Frohlich's Das Kriegswesen Cllsars, p. 156.

^ R. Schneider, Legion und Phalanx, 1893, pp. 135-6. Schneider (p. 1-46)
holds that when the acies duplex (see p. 587, supra) was adopted, the light-
armed troops did not begin the battle, but that the first line was closed through-
out, while the second served as the reserve. But even in the acies triplex the
light-armed troops did not always begin the battle : in fact in the BeUuin
Gallicum they are never said to have done so. See p. 594.
^ 3Iem. mil, ii, 91. See also R. Schneider, op. cit., pp. 137-8. I am sure that
Guischard was wrong on this point. His view is irreconcilable with Appian,
B. C, iv, 128. Cf. Klio, vii, 1907, pp. 323-4, 330-2, and pp. 594-5, 597, infra.
* Op. cit.,
p. 139.
 —
CAESAR'S ORDER OF BATTLE 591

luid served their purpose, disappeared.^ If, says Guischard, intervals

liad been left in the inevitable result would have

the fighting line,
been that the enemy would have rushed through, attacked the
maniples in flank and rear, and destroyed the whole formation."^
When, he adds, Livy says that in the Latin War the Roman first
and second lines retreated between the intervals of the third, he
simply exposes his own ignorance.^ The Roman tacticians did not
tie themselves down to any one formation and the intervals were ;

merely intended to facilitate the movements necessary for taking up

any order which circumstances might require. The German writers
whom I have mentioned take the same view and Delbrlick ^ adds ;

that it would have been impossible, in the stress of battle, to preserve

the prescribed intervals and further, that the maniples of the
;

second line could not have relieved those of the first, when they
were tired, by advancing through the assumed intervals, since,
when the tired maniples began to fall back, the enemy would have
pressed after them and thrown the whole array into confusion.
Renard ^ quotes from Livy the passage inde tribunis centurioni-
husque imperat ut viam equitibus fatefaciant jpanduntur inter . . .

ordines viae ^ (' he then ordered the tribunes and centurions to make
room for the cavalry and passages were opened between the
. . .

centuries ') and he remarks that since it was necessary to make

;

these intervals in the heat of battle, they evidently did not exist
bsfore. Riistow's opponents also point to the famous passage in
w^hich Livy says that Scipio, before the battle of Zama, did not form '

his cohorts in close order, but drew up his maniples at considerable

intervals, in order that there might be room for the enemy's elephants
to pass without breaking the formation' {non confertas autem co-
hortes .instruebat, sed manipulos aliquantum inter se distantes,
. .

ut esset sfotium quo elephanti hostium acccpti nihil ordines tur-

barent)^ from which it is inferred that the normal intervals between
maniples or cohorts were only sufficient to mark their individuality.
Schneider ^ remarks, further, that it is absurd to conclude from the
unsupported testimony of Livy that only the first line was utilized
in the earlier part of the battle evidently, he insists, the first two
:

^ Op. cit., p. 141. ^ Stoffel {Guerre civile, ii, 328) says much the same.
2 Renard, on the other hand {Hist. pol. et mil. de la Belgique, 1847, p. 311,
note), observing that, according to Livy, the maniples stood at '
moderate
distances apart' {distantes inter se modicum spatium), argues that what he
meant vi^as, not that the maniples of the first line retreated between the intervals
which separated the maniples of the second, but that the several files of the
first line retreated between the files of the second but this interpretation is
;

stultified by the later sentence in whicli Livy says that if tlie principcs also
failed, the triarii, or third line, received both them and the hastafi into the
intervals between their ordines, and that then, suppressing the intervals {viae)
and closing the ordines, the whole army in one unbroken line {uno contlnenti
Ufjmine) fell upon the enemy.
* Hist. Zeifschrift, N. F., xv,
1883, pp. 240-3.
5 Hist. pol. et 7nil. de la Belyique,
pp. 310-3. " x, 41,
§§ 8-9.
' XXX, 33, § 1. Frontinus {Strat., ii, 3, § 16) says much the same,
nee continuas consiruxit cohortes sed manipulis inter so distantibus spatium
;

dodit, &c. 8 Op, cit.,

pp. 13(5-7, 143.
592 CAESAR'S ORDER OF BATTLE
linos vvcic ill tictioiithroughout, and only the third served as a reserve.
To prove this, he refers to various passages in the Commentaries —
B. G., i, 49, § 2 ; 52, § 6 ; B. C, i, 41, § 4 iii, 94— which shall be
;

presently considered.
The theory that the legion fought in groups has, however, recently
gained a powerful advocate, Captain G. Veibh,i who attacks Stoffel,
Delbriick, and Rudolf Schneider with impartial vigour. The young
Austrian officer begins by reminding presumptuous historians and
philological pedants that they are incompetent to solve the problems
of ancient warfare unaided, and must ask practical soldiers like
himself to help them. But the historian, however patiently he may
submit to being put in his proper place, cannot help reminding
Captain Veith that doctors are not the only professional men who
disagree and when he finds that three such practical soldiers as
;

Riistow, Stoffel, and the captain are at variance on vital points,'^ and,
moreover, remembers that the captain is the only one of the three
who has not seen active service, he feels that, after he has listened
humbly to professional opinions, he must still use his own judgement.
It must not be supposed that Veith maintains that the quincunx
formation of 4 cohorts in the first line, 3 in the second, and 3 in the
the third was unalterable, or that the intervals were necessarily
equal to the breadth of a cohort, or even that intervals in all circum-
stances existed.^ On the contrary, he argues that as the battle
neared its decisive phase, the fighting line must have tended, by the
successive arrival of reserves, to approach continuity.* The source
of his inspiration is the famous passage in which Poly bins ^ explains
why and how the legion defeated the degenerate Macedonian phalanx.
Elasticity, says Polybius, was the secret of Roman victory the :

essence of Roman tactics, says Veith,^ was to dispose the cohorts

in a way which would make it possible to bring into action groups
of any required strength, and in any desirable formation at any
point and at any moment and, he asks,"^ what would have been
;

the use of adhering slavishly to a prescribed scheme when the utility

of the Roman system lay in freedom from restriction ? This, as the
reader will have observed, is substantially what Guischard said
a century and a half ago and indeed Veith is careful to point out
;

that the difference between his theory and that of the advocates of
a continuous formation is not so great as might at first sight appear.^
Veith begins by endeavouring to show that the danger of leaving
intervals between cohorts is imaginary, while it would have been
very dangerous for the enemy to penetrate between tw^o intact cohorts
— far more so than to break a continuous line especially as there —
was a third in reserve before him. Moreover, Veith continues, every
scholar admits that the continuity of the line was broken by intervals
of some sort, however small why were they not dangerous if larger
:

ones were ? ^ Again, Polybius explains that Roman tactics enabled

1 Kilo, vii, 1907, pp. 303-34.
- lb., p. 325. ^ lb., pp. 309-10. ' lb., p. 328, u. 1.
' xviii, 29-32 (12-5). « Klio, vii. 327. ' lb., p. 333.
8 Cf. R. Schneider, op. cU., pp. 138, 1-19. » Klio., vii, 300-7.
 —
CAESAR'S ORDER OF BATTLE 593

the legion to utilize the features of the country and in broken ;

country a continuous line could not have acted.^ The opponents

of Riistow's theory forget that even in Alexander's army there were
intervals between the Macedonian phalanx and the auxiliary corps
which accompanied it.^ Caesar's statement^ that at Pharsalia
Crastinus and his company were the first to charge alone pulverizes
the theory that the line of battle was continuous ^ and the contrast, ;

so sharply emphasized by Polybius, between the legion and the

phalanx, would have been insignificant if there had been no intervals
in the former.^ In short, says Veith,^ the continuous formation
imagined by Delbriick would have been merely a bastard phalanx,
a phalanx with all its defects and none of its advantages. For let
us suppose that such a line encountered a genuine phalanx. What
would happen? The phalanx, by dint of its immensely superior
weight, would inevitably break the line intervals would be formed :

involuntarily and the fragmentary groups, overpowered, driven

;

back, and shattered, would prevent their own reserves from acting.
On the other hand, if intervals had been left intentionally, any
portions of the phalanx that attempted to penetrate would be
exposed to attack in flank and rear, if not to disaster.*^ Unlike
Schneider, Veith believes that the second line as well as the third
acted as a reserve.
And now, reader, having profited by Veith's professional knowledge
and Schneider's learning, let us endeavour, with due humility, to
draw our own conclusions. To begin with, we must take account
of the arguments, drawn from Livy, which Veith has left unanswered.
Assuming Livy's accuracy, the words inde tribunis centurionibusque
imferat ut viam equitibus fatefaciant panduntur inter ordines viae^
. . .

may prove that before the order was given intervals had not existed ;

but nobody maintains that intervals were an invariable feature of the

Roman order of battle. Livy's description of the peculiar order which

1 Klio, 309.
vii,
^ '
Philip and Alexander,' says Professor Oman {Companion to Greek
Under
Studies, ed. L. Whibley, 1905, p. 468), the phalanx had still some mobility, '

and its various Ta^eis could act independently of each other and execute
individual movements. But in the third century [b. c] it became a single
clump of spears of most unwieldy size. Pyrrhus seems to have been the only
. . .

general among the Epigoni [the inheritors of Alexander's dominions] who tried
to keep the phalanx mobile in his Roman wars we read that he drew it up
:

in a row of small columns, not in one mass,' &c.

3 B.C., iii, 91, § 3. * Klio, vii, 313.

^ G. Veith, Qesch. d. Feldziige C. J. Caesars, p. 484.

« Klio, vii, 312.
' 314.
lb., p. Th. Steinwender {Philologus, Ixviii, 1909, p. 262) says that
the maniples [or cohorts] in the second line would have been unable to encounter
the portions of the phalanx that penetrated through the intervals because they
were too far behind the first line. Surely this is a weak argument. The
second line was never more than 50 or 60 yards behind the first, and could
have advanced in a few seconds to stem a threatened attack.
Veith seems over-anxious to prove his case and I have omitted some of his ;

weaker arguments, e.g. (pp. 308, 311) those based upon B. G., v, 34 and
B. C, iii, 88-9, which, as any intelligent reader will see, are inconclusive.
* See p. 591. ^^>** ,
..^_, . ^
594 CAESAR'S ORDER OF BATTLE
Scipio adopted at the battle of Zama ' certainly suggests that he
believed that in 202 B.C. the usual formation was a closed line but ;

if he did, he is contradicted by Polybius, who says that behind the

hastati Scipio stationed the principes, their maniples not covering '

the intervals between those of the hastati, as the Roman custom is,
but being posted immediately behind them at some distance, on
account of the multitude of the enemy's elephants ' (cttI Sk tovtol^
rov<S TrpiyKLTras, rivets ras o-Tretpa?, ov Kara to twv Trpwrwv ar]fxaL(x)V Std-
o-Trjfia, KaOoLTrep e^os ecrrt tol^s Pw/xatot?, dXXa Kar dXXr]X.ovq iv aTroo-Tacrei,
8ta TO 7rXrjOo<; tmv irapd I'do not infer from
tols IvavTiOiq iXecfjdvTwv).^
the words '
as the Roman custom
that the intervals usually
is '

remained open after the light-armed troops had done their work ;

but there is sufficient evidence that they sometimes did. Schneider

is mistaken in asserting that Livy is the only writer who says that
the second line did not come into action until after the first stage of
close fighting ended for Polybius distinctly states that this happened
;

at Zama.^ There, as in Livy's famous chapter, after the preliminary

skirmishing the hastati separately attacked the Carthaginians then ;

the frincipes were ordered to advance and finally hastati, principes,

;

and triarii fought in unison. Scipio, says the historian, ordered the '

principes and triarii to take close order and deploy into line with . . .

the hastati on either flank.' Do not the words to take close order
'
'

show that before their order had not been close ? As to the period
which followed that of the manipular organization, when Schneider
affirms that the light-armed auxiliaries regularly opened the action,
he forgets that in the -battle with the Helvetii and apparently also in
the battle with Ariovistus the auxiliaries took no active part nor ;

is there a single word in the Conmientaries which confirms his theory.

Again, the passages which he quotes from Caesar do not prove that
the second line was from the commencement of close fighting incor-
porated with the first they only prove that the first two lines formed
:

in some sort a whole as distinct from the third, which remained in

reserve. Consider the description of the battle with the Helvetii,
where the usual triplex acies was adopted. On Schneider's theory
intervals did not exist even at the outset of this battle, because the
auxiliaries, who, as he believes, would normally have filled them,
were far removed from the fighting line if, then, as he insists, the :

second line was from the commencement of the battle incorporated

with the first, why was it formed at all, and what was the sense of
the expression triplex acies ? Evidently no triplex acies existed and ;

Caesar ought to have written duplicem aciem instruxit Yet it is !

1 See
p. 591. Weissenborn in his edition of Livy (1858, vol. vi, p. 382)
observes that cohortes is not to be understood in its later sense, but was used
instead of manipulos for the sake of variety. But is it not conceivable that
Livy may have been guilty of an anachronism and have meant that the
maniples of each (imaginary) cohort were not close together, but separated by
intervals ? ^ xv,
9, § 7.
^ crvi'€x^o.v ras twv aardTcuv arjuaiar ol fxev yap twu irpiyKL-nojv yye-
(irtTTCffovTcs
ix6v(:s awOeaoafxevoi to yeyovus, kitiaTrjcrav ras avrojv ra^eis rovs 8' eTTidiwKOVTas. . .

TMv dardTOJu dvaKaXeadpn'os rovs 5e npiyKmas nal rpiapiovs irvKvwaas e<^' eKarepov
. . .

TO Kepas irpodyeiv naprjyyeiXe 5m twv I'eKpwv (xv, 13, § 7 ; 14, §§ 3-5).

 CAESAR'S ORDER OF BATTLE 595

clear from his narrative tliat even after the first stage of the battle
was over the first two lines still existed as such. Again, take Poly-
bius's account of the battle of Cannae. He says that Varro stationed '

the maniples closer together than usual (n-vKvoTipa<s r/ 7rp6cr6ev ra? '

o-?;/xata9KaOtaTaviov),^ and that he placed the light troops in advance '

of the whole army (Trao-r/? 8e r-^? Svvdixeo)<; irpoia-Trjo-i tov<; eu^wvov?

eV aTToarda-ei).'^ The light troops, therefore, did not occupy
the diminished intervals before close fighting began and unless ;

Polybius means that the maniples stood closer than usual when and
after close fighting began, his remark has no point. For, according
to Schneider, the maniples habitually stood close together after
hand-to-hand fighting had begun and what would have been gained ;

by placing them closer than usual if, immediately after hand-to-

hand fighting had begun, the intervals had disappeared?
There is, indeed, another passage in Polybius which may appear
to support the theory that the line of battle was generally continuous.
At the end of his account of the battle of Zama he remarks that
'
though the Roman line is hard to break, yet each individual soldier
and each company can fight in any direction (ovarj^ yap Sva-Sia-
. . .
'

(TTrdcrTov r^s Pw/xatwv ra^ews koi Swa^ew?, tov dvSpa crvve/Sr] kol
KaOoXov KOL Kara fxepyj ixd)(€(T6aL Trpo? rrdaa^ ras eTTK^aveia?).^
These words might suggest that the Roman line, though it was
composed of units which could act independently, was normally
continuous but Polybius's account of the battle of Zama (not to
:

mention Cannae) shows that on that occasion it was not connected

until the final stage when the hastati, princi'pes, and triarii formed
a compact line and he may only have meant that it was hard to
;

break because the second and third lines were always ready to support
the first. Let us examine his comparison of the Roman line with
the degenerate Macedonian phalanx. The Romans,' he says, do ' '

not attempt to extend their front to equal that of a phalanx, and

then charge it directly with their whole force but some of their ;

divisions are kept in reserve, while others encounter the enemy.

Now, whether the phalanx in its charge drives its opponents from
their ground, or is itself driven back, in either case its peculiar order
is dislocated for whether in following the retiring, or flying from
;

the advancing enemy, they quit the remaining divisions of their

force and when this takes place, the enemy's reserves can occupy
;

the space thus left and the ground which the phalanx had just before
been holding, and fall upon them on their flank and rear (Ov ydp
. . . '

i^La-d)(TavT€S TTfjv irapdra^LV TrdcrLV ajxa (TvpL^aXXovai TOt<s (rr/aaroTreSois

/xercoTnySoi/ Trpos ras ^aXayya?, dXX.d to. fxkv icfieSpevet twv jxepwv avrots,
TO, 8e crvjjLfjiLayeL rot? TroXe/xtoi?. Aolttov, av t iKmeaoicnv ol <5f)aAayytTat
Tov<s Ka$ avTov<; 7rpocr/?aAXoj/TC9, av r iKTriecrOoidiv vtto rovTOiv, XeXvTai to
Trjq (f>d\ayyo's lSlov. '
H ydp €7r6fxevoL rot? v7ro)((ji)pov(TLV, tj (f>evyovT€S tov<;
7rpo(TKeLfxivov<i, (ZTroAciTrovcrt to, AotTra fJ^^prj rrjs OLKeca'S SvvdjJi€Oi<i. Ov
yevojxevov, SeSorai rots icl>€Sp€vov(TL rwv iroXefXLWv Stacrr^z/xa Kal totto^, ov
ovTOL KaT€i)(ov, TTpo? . . . TrapetaTricrovTa'S TrAayt'ous TraptorTacrOaL Kal Kara
1 ill, 113, §3. 2 ji,^ § 4 3 XV, 15, § 7.

Q q2
596 CAESAR'S ORDER OF BATTLE
vwTov TOi<s (^aXayytVais).' ^ What
the meaning of the words The
is '

Romans do not attempt to extend their front to equal that of a pha-

lanx'? Either that the Roman line, though continuous, was over-
lapped on right and left by the phalanx or that, although the Roman
;

wings were opposite those of the enemy, there were intervals in their
line which made the combined width of the several maniples less than
the width of the phalanx. Prima facie I should say that the former
was the meaning and if so, Polybius here supports Schneider but
; :

Caesar's description of his battle with Ariovistus proves that his wings
were not overlapped by the German phalanx, and seems to me to imply
that his army began by fighting in groups. The auxiliaries were
certainly not posted in intervals between the cohorts, but, as Caesar
expressly says,^ were massed in front of his smaller camp. Schneider
therefore holds that the second line was from the outset incorporated
with the first. That, for the reasons which I have given, seems to
me out of the question. Stoffel, on the other hand, who agrees with
Schneider that there were no considerable intervals, insists that the
second line was posted behind the first, and relieved not reinforced —
it as occasion required. But Stoffel surely forgot that, as the Germans
outnumbered the Romans, the first line would in this case have been
less than one-third as strong as the German phalanx, and must have
been speedily overpowered. And if the first line was, as he holds,
continuous, evidently it could not have been reinforced for want of
room. I am therefore inclined to agree with Veith that the Roman
lines fought in groups. It might, indeed, be objected that if the
intervals had been considerable, either the greater part of the phalanx
would have done absolutely nothing, which is incredible, or they
would have penetrated the intervals, in which case the second
line would of course have advanced to meet them, and the chequer
formation would, temporarily or permanently, have disappeared.
But it would not have disappeared until it had done its work and ;

surely it is significant that Philopoemen in the battle of Mantinea, as

Polybius^ is careful to point out, deliberately left intervals, which
were not intended for the passage of light troops, between the
component parts of his phalanx.
It is true that Vegetius,* describing an order of battle that was
customary at some period of the Empire, says nothing about intervals,
while the words which he uses in explaining the collocation of the
several cohorts coniungitur and adnectitur suggest that the line —
was then continuous and in another passage he observes that panic
;

must ensue if the line is broken and the enemy are able to attack
it in the rear.^ But may we not suppose that this continuous forma-
tion was a symptom of decadence, analogous to that which we have
already noticed in the case of the later Macedonian phalanx?
Now let us test Veith's argument. Neither Schneider nor any other
writer argues that the Roman line was invariably continuous and ;

therefore Veith's remark about Crastinus, which, by the way, does

1 xviii, 32 (15), §§ 2-5.

^ ^ q^ j^ 51^ § 5^
3 ^i, 11, § 6.
^ De re mil, ii, 15. ^ /j,.^ j^ 2G.
 CAESAR'S ORDER OF BATTLE 597

not prove the existence of lateral intervals, is irrelevant. Even if the

Roman was generally continuous, it was still sharply contrasted
line
with the Macedonian phalanx for its continuity was not rigid but
;

elastic, and its component parts were capable of independent action.

But these are minor points my only serious complaint against
:

Veith is that he is too fond of tilting at windmills, that he has neglected

passages which tell in favour of his theory, and that he has left
certain objections unanswered.
I have tried to show that the arguments of Veith and Schneider are
both vulnerable and my own conclusions are not as dogmatical
;

as theirs. I only claim to have called attention to matters which have

been overlooked, and to have shown that the whole question is more
difficult than Delbriick and Schneider on the one hand, Veith on
the other, appear to suppose. Unhappily no ancient writer has left
a description of the tactics of the Caesarian as complete as that which
Polybius gives of the manipular legion, which itself leaves much to
be desired we can only have recourse to scattered scraps of informa-
:

tion. But this much is clear —

there is no evidence that in Caesar's
:

time the auxiliaries ever opened the battle and certainly they did
;

not in the battle with the Helvetii or in the battle with Ariovistus.
The first two lines appear to have maintained their individuality
even after close fighting had begun, though, as the battle neared its
end, the second may have become gradually incorporated with the first.
As for intervals, I believe, for the reasons which I have given, that
Veith is nearer the truth than Schneider, though both are right in
denying that, when they existed, their width was invariable. No
doubt the Roman line was never absolutely unbroken but the size ;

of the intervals must have depended upon circumstances which we

cannot ascertain. The main point is that, as Polybius says, the Roman
order of battle was flexible for every Roman, once armed and on
'
;

the field, is equally well equipped for every place and time, and for
every appearance of the enemy. He is, moreover, quite ready and
needs to make no change, whether he is called upon to fight in the
main body, or in a detachment, or in a single maniple, or even by
himself (r; Se 'VfiifxaLMV [o-wra^t?] €V)(^pr](TTO<;. Ila? yap 'Pw/xato?, oTav
oLTra^ KaOoTrXicrdil'S opfjLrjarYj Trpo? rrjv ^petar, ofxoio)^ ypfiocTTat Trpo? Travra
TOTTOV /cat Katpov, Kat 7rpo<s Traaav lirK^aveLav Kat /xrjv eroip-o? Icttl^ Kat
T7]v avrrjv €;^et av re /xera iravTOiv Sirj KLvSwevetv, av re /JLera
SlolOco-lv,

fxepov?^ av re Kara cr-jy/xaiav, av re /cat /car' avSpa).^

3. The next problem which we have to solve is this how were :

the men in the fighting line relieved ? Frohlich,^ postulating that

the tactical unity of each cohort must be preserved, originally held
that the ranks of the tired cohorts closed up, and thus created
intervals for fresh cohorts to enter and take their places. The tired
cohorts retired fighting, while the ranks of the relieving cohorts
immediately spread out on both sides and filled up the intervals.
Soltau,^ improving a little upon this theory, held that, instead of
1 xviii, 32(15), §§9-11.
•^
Realistischcs and Sfilidlsches zu Cdmr, &c., 1887, pp. 11-4.
^ Hermes, xx, 1885, pp. 264-7.
598 CAESAR'S ORDER OF BATTLE
the relieving cohort's waiting patiently until the whole interval was
ready for it to enter, the men of whom it was composed gradually
insinuated themselves into the interval the moment it began to be
formed. Then and not till then, Soltau thought, the tired cohort
began to fall back. The thing sounds easy, says Giesing ^ but there ;

is one difficulty to be reckoned with. The moment the tired cohort

begins to fall back, the plaguy enemy press after
'
the relieving
'
:

cohort finds itself surrounded and the line of battle is destroyed.

;

Giesing's theory is as follows. In hard fighting the foremost ranks

of the first line inevitably become thinned the gaps are filled up :

by the advance of the rear ranks between the files of those in front
and this process goes on until the depth of the first line is in danger
of becoming unduly thinned. Should it be necessary to relieve any
of the cohorts as a whole, the relief is accomplished in the same way :

the ranks of the relieving cohort advance between the files of the
tired cohort; and the enemy never gets a chance of breaking the
formation.^ The theory that entire groups were relieved has, how-
ever, I believe, been generally abandoned.^ Veith,* remarking that
no modern commander withdraws troops from the fighting line when
he reinforces them, argues that in Caesar's time at all events relief
had ceased to be a regular feature of Roman tactics and was replaced
by reinforcement in other words, he believes, if I do not misunder-
:

stand him,^ that fresh troops were brought into the fighting line
through the intervals, which thus, as the battle neared its end,
disappeared or tended to disappear. But this suggestion, although
it is supported by Caesar's account of the last stage in the battle of
Pharsalia,^ does not remove all perplexities for modern warfare is not ;

characterized by prolonged hand-to-hand fighting. I can only suggest

that, according to circumstances, relief or reinforcement or both
were adopted when they were practicable, and effected in the way
which under the circumstances seemed best and I agree with ;

Schneider that for the relief of individual soldiers the method sug-
gested by Giesing is the best. Superintendent Froest, of the Criminal
Investigation Department, tells me that it would be adopted by the
police in street fighting and indeed no other method appears
;

practicable. When, as in the battle with Ariovistus, the third line

^ Neue JaJirb. f. Philologie, &c., cxxxvii, 1888, pp. 852-9.

^ So also Stoffel {Guerre de Cesar et d' Arioviste, pp. 120-1) writes, '
comme le
legionnaire etait pesamment arme et que les combats etaient des luttes d'homme
a homme, il est a croire qu'au bout de quinze a vingt minutes la premiere ligne
avait besoin d'etre soutenue ou relevee. Alors les soldats de la deuxieme
ligne passaient dans les intervalles des combattants pendant que ceux
de la premiere se retiraient7 «S:c. I ought to say that Frohlich {Das Kriecjs-
wesen Giisars, p. 164) was converted by Giesing's arguments.
^ See R. Schneider, op. cit., p. 147. Stoffel consistently adhered to the
theory, which, as we have seen (p. 589), is supported by Livy, that entire
cohorts were relieved.
* Klio, vii, 316.
^ Ih., p. 328, n. 1.
^ Eodem tempore tertiam aciom Caesar . . . procurrere iussit. Ita cum
rcccntes atquc intcgri dcfcssis succesi^isscnt . . . sustinere Pompeiani nun potue-
runt, &c.
 CAESAR'S ORDER OE BATTLE 599

was brought into action, the cohorts that composed it may have been
directed against one or both of the enemy's flanks or on his rear.
It may be worth while to add that FrohHchi affirms, on the
strength of certain passages in Plutarch ^ and Appian,^ that, after the
first two ranks of the first line had thrown their javelins and while
they were using their swords, the rear ranks threw their javelins over
the heads of the first two. I do not dispute this statement but I can ;

find nothing in the passages referred to which proves it. Stoffel *

simply remarks that the first two ranks opened the battle by throwing
their javelins, the second rank throwing theirs a travers les inter-
'

valles du premier '.

THE AGGER
I. No ancient writer has left any detailed account of the mode in

which the agger, or siege mound, as used by Caesar, was constructed ;

but there is evidence enough to enable us to piece together an accu-

rate, if not complete description of its material, form, and mode of
construction.
Before the construction of the agger could be begun, the ground
upon which it was to be erected had to be levelled, or if, as at Avari-
cum, it formed an irregular hollow or depression, which could not
be reduced to one plane surface, it was probably levelled, so to speak,
in step-like sections ^ and this was done by men working under the
;

cover of a testudo,^ or sapper's hut. The agger was made largely,

if not mainly, of wood and sometimes, if not always, it contained
;

earth and rubble as well."^ The woodwork consisted of logs, piled

in successive layers, the logs in each layer being laid at right angles
with those in the layer below. When, as at Avaricum and Massilia,
the agger was very large, this elaborately constructed woodwork
probably served only as a wall on either side, to prevent the more
loosely heaped interior from scattering.^ The agger was sometimes
undermined and fired by the enemy. The workmen were protected
from the enemy's missiles by vineaep or sheds. ^^
^ Das Kriegswesen Ccisars, p. 148. ^ Sulla, 18.
3 B. C, iv, 128. *Guerre civile, ii, 339.
^ See pi. 10 of the Atlas accompanying Stoffel's Guerre civile.

^ See
pp. 609-10, and B. G., ii, 2, § 4. Stoffel {Guerre civile, ii, 358) asserts
that when, as in the Gallic war, the besieged had no artillery, the workmen
were protected merely by mantlets, ranged in their front and on their flanks.
How about arrows ?
' Stoffel {ib., ii, 356) thinks that timber, being much lighter than earth or
stones, would, in places where it was sufficiently abundant, have been used
ahnost exclusively, earth being only employed a combler les espaces vides,
'

a egaliser et a affermir la masse'. See Thucydides, ii, 75, § 2; Lucan, iii,

394-8.
8 Thucydides, ii, 75, § 2. » See p. 608.

1" B. G., ii,

2, § 4 B. G., vii, 24, §§ 2-3
; ; Thucydides, ii, 75 ; Lucan, iii,
394-8, 455 Ai)pian, Milhr., 30 ; Siiius italicus, xiii, 109-10.
; See also Stofl'el,
Guerre civile, ii, 355-6, and Uarcmbcrg and Saglio, Diet, den ant. yrecques et
ruin., i, 141.
600 THE AGGER
11. Mr. Judson,! following Rustow,^ has propounded a theory
1.

regarding the construction of the agger, which it would be useless to

examine, because it is based upon the fantastic assumption that the
structure was not solid, but contained galleries. Von Goler,^ who
agrees with Riistow, cites a passage in which Caesar, describing the
siege of Avaricum, says that the Gauls endeavoured, by means of
sharp stakes, molten pitch, and huge stones, to prevent the Romans
from bringing their open (?) galleries up to the wall (apertos cuniculos
praeasta et praeacuta materia et pice fervefacta et maximi ponderis
saxis morahantur moenihusque adpropinquare prohibebant)A This
passage is omitted in the ^ MSS. :^ but I have little doubt that Caesar
wrote it for I cannot frame any theory to account for its having
;

been interpolated. Von Goler holds that by apertos cuniculos Caesar

meant the ends, open towards the town, of galleries which ran through
the whole length of the agger. Frohlich,^ on the other hand, believes
that these cuniculi were subterranean galleries, by which Caesar
intended to undermine the enemy's wall or to make his way into the
town and this view is supported by a passage in Curtius,"^ from
;

which we learn that Alexander the Great, while constructing an

agger during the siege of Tyre, made galleries for the purpose of
undermining the wall. But if the galleries of which Caesar speaks
were subterranean and had not yet reached the wall, how could he
have called them open ? Long^ says, If the text is right .'.
' ' '
.

the Galli contrived to work into the Roman " cuniculi ", to open
. . .

into them,' &c. This explanation is supported by a passage in which

Appian ^ describes how, when Sulla was besieging Athens, the mines
of the besiegers and of the besieged met but it seems doubtful ;

whether apertos can be taken as a participle.!^ Those commentators

who hold that the cuniculi were galleries in the agger believe that
they were intended to protect the workmen but, as protection ;

^ Caesar' s Army, pp. 93-5.

^ Heerwesen und Kriegjvhrunrj C. J Casars, 1855, pp. 147-51..

3 Gall. Krieg, 1880, p. 252. ^ B. G., vii,

22, § 5.
Meusel's Lex. Caes., i, 800. ^ Das Kriegswesen Casars,
^ p. 254.
' De rebus gestis Alexandri Magni, iv, C, § 21. Alexander aggerem quo . . .

moeniuin altitudinem aequaret, exstruxit, et pluribus cuniculis niuros subrui

iussit. Cf. Livy, xxiii, 18, §§ 8-9 ; xxxviii, 7. Meusel in a MS. note asks,
'
Must not these cuniculi, then, have been in the agger ? Surely not Of '
!

what use would such cuniculi have been for undermining the wall ?
^ Caesar, p. 346. See also Schneider's Caesar, ii, 399.
9 Mithr., 36.

^^ See W. Paul in Zeitschr. /. d. Gymnasialwesen, 1878, pp. 168-9. Paul,

who holds that the cuniculi were subterranean, conjectures that Caesar wrote
repertos. P. Geyer {Jahresh. d. philol. Vereins zu Berlin, v, 1879, p. 353),
objecting to Paul's emendation, argues that the cuniculi must have been
galleries in the agger, (1) because the second agger which the Romans
constructed at Massilia contained galleries, and (2) because the use of boiling
pitch by the besieged would have been more applicable to such galleries than
to underground ones. But tiie second agger at Massilia contained one gallery
only it consisted of two walls of brick, 6 feet thick, and a planking of timber
:
'

laid across those walls Caesar expressly says that it was of a novel and
' :
'

unheard-of kind {aggerem novi generis atqjie inauditum

' facere instituerunt . . .

\B. C, ii, 15, § 1]) and, as far as I can see, boiling pitch would have been as
;

inconvenient to workmen underground as above.

 THE AGGER 601

could have been obtained just as well by the use of vineae, it is clear
that if the agger was made with galleries, the object must have
been to save material. Such a saving, however, would have been more
than counterbalanced by the enormous increase of labour that would
have been entailed by making an agger with galleries sufficiently
strong to carry the host of soldiers, the vineae, and the huge towers
that stood upon it. Indeed, it is hardly credible that, without bricks,
the Romans could have made such an agger. There is absolutely
no evidence for the view that aggeres were ever built with galleries,
except the passage which I have quoted from B. G., vii, 22 and of ;

the scholars who admit the genuineness of that passage the majority
understand by the word cuniculus not galleries in the agger but
subterranean galleries. It is true that the writer of the article
Agger in Daremberg and Saglio's Dictionnaire des antiquites grecques
et romaines ^ points to a drawing which he reproduces from S. Bartoli's

Colonne Trajane as evidence that the Roman agger did contain

galleries. Dans la figure,' he says, on voit des soldats romains
' '

occupes a construire un agger et entassant des troncs d'arbres entre-

croises. On remarque une sorte de voute formee de poutres disposees
en arc-boutant et peut-etre destinee a proteger un de ces chemins
converts (cuniculi) par lesquels on pouvait faire avancer des sapes et
des mines jusqu'aux murs.' But it is very doubtful whether the
drawing in question represents an agger at all. Mr. Judson argues
that the agger could not have been destroyed by fire unless it had
been built with galleries. Thinking over this argument, it occurred
to me that Caesar would hardly have played into his enemies' hands
by affording them facilities for destroying his handiwork and when ;

I consulted scientific men and members of the London Fire Brigade,

I found that they did not agree with Mr. Judson. If he had read his
Josephus, he would have understood how an agger could be set on fire,
even though it had no galleries. In the defence of Jerusalem John '

undermined the ground between the citadel and the banks [aggeres']
and propped up the works above [namely the banks] by beams
placed at intervals in the mines. Then he brought in wood smeared
with pitch and bitumen and set fire to it the beams were consumed
:
;

the mine all at once collapsed, and the banks fell into it with a
tremendous noise. At first there arose thick smoke and dust, as the
fire was choked by the fall but when the wood which pressed it
;

down was burnt through, a flame was seen bursting forth '.2 Stoffel,^
who ridicules the notion that the agger contained galleries, sensibly
remarks that at Avaricum the besiegers only learned that the agger
was on fire by seeing smoke rising from its surface.* If, he argues,
there had been galleries in the agger, on se serait aper9u de I'accident
'

plus tot,' &c. Moreover, there is direct evidence that some aggeres,
at all events, had no galleries. The agger which the Lacedaemonians
constructed at the siege of Plataea was solid, and had tiers of logs
on either side to prevent the interior wood, earth, and stones from

'
i, 142. 2
^^11 j^^^ yii^ 4^ §§ 469-71.
^ Guerre civile, ii, 303. ^ B. G., vii, 24, §§ 2-3.
602 THE AGGER
scattering.! Finally, Lucan says that the first agcjer [or rather one
of the two original aggeres] constructed at Massilia was made of a
core of earth and faggots supported by walls of timber on either
side :

tunc omnia late

Procunibunt nemora, et spoliantiir robore silvae,
Ut cum terra levis mecliam virgultaque molem
Suspendant, structa laterum compage ligatam
Arctet humuni, pressus ne cedat turribus agger.^

Ciacconius,^ contrasting apertos cuniculos with the tectos cuniculos *

('covered galleries ') which were driven into the rock at Uxellodunum,
in order to divert a spring, maintains that the former were open
trenches and this view is perhaps supported by a passage in which
;

Ammianus Marcellinus ^ describes besiegers as attempting under the

protection of glutei, or mantlets, to undermine a wall. I am rather
doubtful whether the word cuniculus could be used of an open trench ;

but if so, Ciacconius's explanation is probable enough. Whatever

the true explanation may be, I am absolutely certain that the agger
was never built with galleries and my conviction was shared not
;

only by Stoffel but also by the late General McLeod Innes, R.E., V.C.,
who had had great experience in military mining.
2. Colonel Stoffel has devised a theory entirely different from that
of Riistow. He holds that Caesar's terraces were of two kinds, the —
^
terrasse-viaduc and the terrasse -cavalier '. The former, which was
'
'

perpendicular to the wall of the besieged town and carried only one
tower, was used, he explains, when the attack was to be directed sur '

—
un seul point determine ', when it was intended to breach the wall
with the battering-ram, and not to take the town by escalade. To
this type the colonel refers the terraces which Caesar constructed
in besieging the stronghold of the Atuatuci, Uxellodunum, and
Massilia. When, on the other hand, the besieged town was situated,
like Avaricum, in a plain, and when (as was the case throughout
the Gallic war) the besieged had no artillery, he holds that the
terrace was a parallel with the wall.^
'
cavalier ',

Now, the colonel means that there was no terrasse-viaduc at

if
' '

Avaricum,'^ Caesar's narrative does not support his contention.

Though Caesar does not say in so many words that the terrace at
Avaricum was built at right angles to the wall, he implies that
those sections of it which carried the towers were. Describing the
siege, he uses the words cum iam muro tunes adpropinquassent,^
&c. (the towers had now got dose to the wall). The towers stood on

1 Thucydides,
ii, 75, § 2. ^ iii, 394-8.

Caesar, ed. G. Jungermann, 1606, p. 288.

^

* B. (9.,viii,41, §4. ^ xxi, 12, § 6.

Guerre civile, ii, 354-61. The colonel means, as I understand, that the
•^

dimension of the terrace from front to back was less than its dimension parallel
with the wall ; but I am not sure whether he means that it was as narrow
as the cavalier shown in Napoleon's Planclie 20.
'
'

^ One would certainly infer that this was his meaning from pages 354-5
of the second a olume of his Guerre civile but when I look at Planche 10 of
;

his Atlas, I am doubtful. B. G., vii, 18, § 1.

**
ace pivje 603

Plan des travayx d'attaque

Coupe suivant C D

'Echclle de o'!^oo8poui'
10 5 O 10
Pied.s
20
romains
lo

SO
metres
M)

lOO

PLAN OF THE AGGER AT AVARICUM

AccoRDiNo TO Napoleon III. and General De Reffye
 THE AGGER 603

the terrace ^ and the word ad'pro'pinquassent would be misleadiii<i;,

;

not to say meaningless, if the terrace had been merely a cavalier ', '

parallel with the wall. Again, describing the measures which he

took to repel the sortie from Avaricum, Caesar says that some of his
men drew back the towers (turres reducereni),^ in order to prevent
the fire which the Gallic miners had kindled under the terrace from
destroying them. The word reducerent speaks for itself.
I conclude, then, that the agger at Avaricum was a terrasse- '

viaduc ', in so far that it carried two towers, on the right and on the
left, which, as the work progressed, gradually approached the wall.
But it was also a terrasse-cavalier
'
for its width, or extension
'
;

parallel to the wall, was 330 feet, and it served as a platform over
which the legionaries advanced to storm the town.^ The question
is whether the cavalier
'
occupied the entire space between the
'

two viaducts or only the front part of that space. Napoleon adopts
the latter view. According to General de RefEye,* whose explanation
he borrows, the agger consisted of two parallel viaducts, with an
empty space between, on each of which, flanked by two rows of
vineae, moved one of the two towers. The 'cavalier', which joined the
two viaducts, was, from front to rear, very narrow it was ascended :

from the rear by an elaborate system of steps and along its whole
;

extension, parallel with the wall, stood two rows of vineae, the vineae
in each row being placed end to end. Now I cannot understand how,
on this theory, the assault was delivered at all. The steps, or rather
staircases, are of course conjectural. There would have been little
or no room on the cavalier for any troops except those who stood
' '

under the vineae if the sides of the vineae which faced the wall were
:

closed, the troops could not have got out of them and if they were
;

open, the vineae would have afforded no protection! I have always

understood from Caesar's description of the assault that the vineae,
the ends of which were open, stood end to end in a direction perpen-
dicular to the wall. Moreover, on Napoleon's theory, the gap between
the edge of the cavalier and the top of the wall of Avaricum was
' '

at least 50 feet and I cannot understand how the assaulting columns

;

were to get across that space.

Nevertheless, I am inclined to think that there is some truth in
Napoleon's view. He estimates the length of each viaduct at about
250 feet ;and to construct a terrace of which the dimensions were
330 feet by 250 would have been a work of enormous labour. Would
it have been necessary? An engineer whom I have consulted agrees
with me in thinking that the cavalier would probably only have
' '

occupied so much space as was requisite to afford room for the leading
1 See p. 745. 2
^
g^ ^ii^ 24, § 5.
^ lb., 27, §§ 2-3. Caesar took Avaricum by escalade ; but it is evident
from vii, 23, § 5, where he says that the timber in the wall ' protects itagainst
the ram {ah nriete materia dej'endii), that, unless he had previously ascertained
'

that such walls could not be breached, he must have attcmi)ted to batter down
the wall of Avaricum and failed.
* Tlicrc i3 a model of the agger at Avaricum by Clcnoral dc Keli'yc in
Salle Xlll of the Muisce de St. (;!ormain. See also II. Ochhr a BiUler- Alias za
Ciimrs Bikhern de B. G., 2nd cd., 11)07, pp. 77-8 and iig. 02.
604 THE AGGER
companies of the assailants, and that fresh troops could have moved
up to support them through the rows of vineae which stood upon the
viaducts.
Stoffel does not attempt to explain in detail the construction
of the '
which he believes to have been used at
terrasse-cavalier
',

Avaricum but he offers a very elaborate explanation of the terrasse-

;
'

viaduc '.1 The following is a summary. Stofiel assumes that the

ground sloped, as at Avaricum, towards the town. As soon as the
inequalities of the surface had been removed, logs were passed from
hand to hand to a group of workers, who arranged them in layers,
descending like huge steps, those of each tier being laid crosswise
on those of the tier immediately underneath. As soon as the first
series, so to speak, of layers, was finished, two galleries of sheds
(vineae) were placed upon the rearmost pile. Under cover of these,
workmen passed fresh logs to others standing upon the pile imme-
diately below and so on. The third and fourth- series were laid in
;

the same way.

Nothing could be gained by criticizing this theory. Right or wrong,
it can neither be disproved nor proved.^ But Stoffel does not explain
how the work would have been executed when the ground, as at
Uxellodunum, sloped upward in the direction of the town.
M. Jullian, remarking that no ancient text supports the distinction
which Stoffel draws between the terrasse-cavalier and the terrasse-
'
'
'

viaduc ', refuses to accept it. It is a mistake, he says, to imagine

that terraces were ever long and narrow on the contrary, the
:

agger was a vast embankment, sometimes as broad as it was long '.^

'

But if no text directly supports StofEel's contention, numerous texts

do so indirectly. The width of the agger must have depended upon
its object and it is self-evident that if the object was simply to
;

make a breach or to batter down a tower, a vast embankment,

as broad as it was long, would have been superfluous. M. Jullian's
criticism is involved in the singular theory, which he has defended
with great ingenuity as well as learning,^ that in the siege of
Massilia there was originally only one agger. Caesar tells us that
at the outset of the siege he proceeded to construct a terrace,
'

to form a line of sheds, and to bring towers to bear against the

—
town at two points, one close to the harbour and the docks, the
other near the gate which is approached from the side of Gaul and
Spain (duahus ex partibus aggerem, vineas turresque ad oppidum
'

agere instituit. Una erat proxima portui navalibusque, altera ad

portam qua est aditus ex Gallia atque Hispania).^ M. Jullian argues

^ Guerre civile, ii, 357-9, See also de Folard, Hist, de Polyhe, ii, 1727,
pp. 488-508, and especially 495-7.
2 Stoffel refers to Planches 89 and 90 of Frohner's La Colonne Trajone
for illustrations of an agger in course of construction. But Frohner himself
describes these ilkistrations as construction d'un camp et rempart de troncs
'

d'arbres et catapultes' ; and what Stoffel calls an agger Frohner regards as

'
un enorme rempart de troncs d'arbres derriere lequel I'armee s'est retranchee'.
See also S. Reinach, La Colonne Trajane au mnsee de Saint-Germain, 1886. p. 56.
^ Rev. des etudes anc, ii, 1900, p. 337, n. 2.

' B. C, ii,
* lb.,
pp. 331-8. 1, §§ 1-2.
 THE AGGER 605

that Lucan mentions only one terrace that Caesar speaks of one
;

agger (agger, aggerem, or aggeri)} not two (aggeres) that he specifies ;

the height of only one agger ^ that, in

; alluding to the right attack,
he speaks of the soldiers who had charge of the right-hand portion
'

of the work' (qui dextram partem operis administrahant),^

whereas, if there had been two aggeres, he would have written qui
dextrum opus administrahant; that if there had been two a^/^eres, the
labour of the troops would have been more than doubled, because the
imaginary second agger would necessarily have been constructed on
marshy ground near the harbour and that in no other siege did
;

Caesar erect more than one agger. Accordingly, M. Jullian under-

stands by the words he proceeded to construct a terrace ... at
'

two points that at those two points the construction of one

'

terrace was commenced simultaneously. Meusel,* who has examined

M. Jullian's arguments with his usual dispassionate fairness, admits
their force, but nevertheless concludes that they are irreconcilable
with the passage from which we learn that on the day after the
Massiliots had destroyed one agger and its movable tower by fire
they set fire to the other tower and the other agger ^ M. Jullian
'

does not refer to this passage, but perhaps he has it in mind when
he says ^ that Caesar's agger, v/hile it only formed one system of
construction, was really double and although it would seem
'
;

to be straining language to contend that alterum aggerem meant the '

other viaduct of the whole agger ', such an expression is perhaps con-
ceivable. Nevertheless the agger which was burnt on the second
day was entirely distinct from that which had been already destroyed.
For, as the one object of the besiegers was to batter down the stone
towers opposite to each of their own movable towers, what would
have been the use of constructing an embankment 300 feet wide "^

between the two viaducts ? Surely M. Jullian, when he lays stress

upon the labour that would have been involved in building two
aggeres, forgets that it would have been far easier to build two
'
terrasses-viaducs than two such viaducts plus a useless embank-
'

ment. Another proof that this embankment would have been super-
fluous is furnished by the description of the brick agger (aggerem
novi generis atque inauditum) ^ which was constructed after the fire.
It was of the same breadth as the original agger (or aggeres) ^ it :

consisted of two brick walls, each six feet thick, connected by a roof

1
/6.,§4; 2,§6; 14,§2; 15, § 1. ^ /&., 1, § 4.
""
Ih., 8, § 1. * C. I. Caesaris comm. de b. c, 1906, pp. 301-3.
^ B. C. ii, 14, §§ 2, 5, —
Hunc [ignem] sic distulit ventus uti uno tempore
agger, plutei, testudo, turris, tormenta flammam conciperent Temptaverunt . . .

hoc idem Massilienses postero die. Eandem nacti tempestatem maiore cum
fiducia ad alteram turrem aggeremque eruptione pugnaverunt multumque ignem
intulerunt. Rev. des etudes anc, ii, 337.
"^

' lb.,
p. 340. « B. C, ii, 15.

^ Unless indeed Nipperdey {Caesar, p. 149), whose arguments have been

reinforced by Ph. Kraus {Berl. phil. Woch., 1899, col. 729-31), was right in
adopting Stephanus's conjecture, altitudine, instead of the MS. reading latitudine.
But anyhow my argument remains unshaken for Kraus holds that the brick
;

agger was narrower than the original.

606 THE AGGEB
or deck oi timber and obviously it was no wider than each of the
;

two viaducts had been. Fiirtliermore, Thucydides's account of

the siege of Plataea seems to show that an acifjer which was only
constructed to carry a battering-ram was a terrasse-viaduc ', long
'
—
and narrow for he says ^ that the Plataeans
;
'
made a hole in . . .

that part of the wall against which the mound pressed and drew
in the earth '.^ Meusel^ accounts for Caesar's having omitted to
describe one of the two original aggeres by the supposition that his
account of the siege of Massilia merely embodies the reports of
Trebonius, Decimus Brutus, and the engineers and he believes that;

in the passages in which agger and turres are mentioned together* the
text is corrupt, and that both words ought to be either singular or
plural.
Finally, I would ask M. Jullian whether he has not forgotten that
in the siege of Jerusalem there were not two aggeres only, but four
[jxeyiiTTa yapra reVtrapa ^ ^oj/xara] ).
i)(^ioa6r) |

III. Von Goler holds with Lipsius that the terrace must have
^ "^

sloped gradually upwards as it approached the wall of the besieged

town. They infer this from Caesar's statement that the daily rise
in the height of the terrace at Avaricum raised, in a corresponding
degree, the elevation of the towers.^ There is a drawing in the
Journal asiatique (4© serie, tome v, 1845, planche Hi) taken from
a bas-relief found at Khorsabad, which represents an agger forming
an inclined plane.^
IV. Guischard,!^ pointing to a passage in which Livy^i says that an
agger collapsed into a trench because it was ill compacted, maintains
that the front part of the agger was, as a rule, strongly revetted and
constructed with especial care. But the agger of which Livy speaks
was merely the rampart of a camp Moreover, it is hard to see how
!

the work which Guischard imagines could have been done unless
a wide gap was left between the agger and the wall and if so, how ;

was the storming party to get into the town ? StoffeU^ maintains,
on the contrary, that the agger, in the last 20 feet of its length, could
only be made by shooting material into the vacant space in the
manner which I have described on page 144 but I am not quite ;

sure whether he is referring to terraces in general or only to the

'.
'
terrasse-viaduc
V. As to how the vineae, or sheds, were used, it is impossible to

give a full and satisfactory account. Guischard ^^ explains the phrase

' ii, 75, § 6.

^
B. Jowett, Thucydides translated, into English, i, 1881, p. 144.
^ G. I. Caesaris comm. de h. c, p. 303.

^ ^. C, ii, 5 Josephus, Bell. lud., v, 11, 4, § 467.

1, § 1 ; 2, § 6.
« Gall. Krieg, 1880,
p. 251 and n. 4. ' Opera, 1637, iii, 295.

^ B. G., vii,
22, § 4. Cf. p. 746. According to Napoleon's Plan (20), which
I reproduce, the whole rise in the elevation of the towers would not have been
more than 6 or 7 feet.
^ In Planche 10 of the Atlas to Stoffel's Guerre civile there is an illustration

of a terrasse-viaduc ', which forms an inclined plane.

'

10 Mem. mil, ii, 7. " x, 5, § 11.

1^ Guerre civile, -p. 359. ^^ Mem.mil.,u,5-d.
 —
THE AGGEB 607

vineas agere as follows :

—
two rows of vineae inclined gradually
inwards until each reached a certain point, where they were joined
])y a third row, which served as a protection to the front of the
aqger. But on this theory, after the agger had risen to the very
moderate height of the vineae, the latter would have been useless.
Caesar 1 says that the men who brought up the material for the
construction of the first agger at Massilia passed it from hand to
hand under the protection of vineae but he does not say how the
;

vineae were placed when the agger was actually being constructed.
We read in his description of the siege of Avaricum^ (though the
passage may be an interpolation) ^ that vineae were placed on the
agger but were they placed there to protect columns of assault or
;

to protect workmen or for both these purposes ? I believe that the

last is the true explanation ^ and Stoffel, as I understand him,
;

takes the same view.^

VI. Riistow ^ says that the surface of the agger was on a level
with the top of the enemy's wall, or even higher. That this was not
always the case is proved by Caesar's statement that the troops
who assaulted Avaricum had to climb the wall,*^ doubtless by —
ladders. At the same time Stoffel ^ is wrong when he concludes
that the agger never reached the top of the wall for Lipsius quotes ;

instances to the contrary from Josephus and Zosimus. The banks,' '

says Josephus, erected by the Romans

' '
—
at the siege of Jotapata
'
overtopped the wall.' ^
VII. Stoffel^o asserts that when a town was assaulted by the
aid of a terrace, such as he believes to have been constructed at
Avaricum, drawbridges were let down from the towers on to the
wall. Vegetius ^^ says that drawbridges were used in this way :

but he also says that scaling-ladders were used in storming towns ;

and we learn from Josephus ^^ that they were used in the assault of
Jotapata. Caesar merely says that at Avaricum his soldiers scaled
the wall (murum ascendissent) ^^ and this statement stultifies
;

Napoleon's illustration, which represents the cavalier as actually ' '

higher than the wall and, moreover, separated from it by a gap at

least 50 feet wide !

The whole subject is very difficult and neither Marquardt nor

;

Daremberg and Saglio nor, as far as I know, any other writer has
given an adequate explanation. Stoffel's is the most satisfactory ;

but I wish he had explained more fully his views as to the way in
which the agger at Avaricum was constructed.

B.C., n,2,r^-
1 ' B.G.,vn,21,^2.

See pp. 749-51.

=^ " See
pp. 749-51.
^ Guischard, on the other hand, asserts {Mem. mil., ii,
6) that les Galeries '

etant trop basses pour proteger ce travail, lorsque I'ouvrage approchoit de son
etendue et de sa hauteur projettees, on avoit recours aux Mantelets.'
® Heerwesen und Kriegfiihrung, &c., pp. 147-8.
' B. G., vii, 27, § 2. s
Querre civile, ii, 362.
lo
9 Bell. lud., iii, 7,33, § 316. Guerre civile, ii, 361.
'^
De re mil., iv, 21.
''
Bell. lud., iii, 7, 24, § 257. " B. G., vii, 27, § 2.
 608

THE VINEA, THE MUSCULUS, THE TE8TUD0,

AND THE PLUTEUS
I. The vinea, as described by Vegetius,i was a stout movable

wooden hut, 16 feet long, 8 feet high, and 7 feet wide. The sides were
defended by wicker-work and the roof was made of planks and
;

protected against fire by raw hides. According to A. Rich,^ one only

of the four sides of the vinea was open but constructed in this way,
;

it would have been useless. The vineae mentioned in the Gallic War
were intended to protect soldiers while they were constructing the
agger and perhaps also while they were forming on the agger prior
to delivering the assault ^ and it seems clear that they were placed
;

end to end in a row. Therefore, in order to enable soldiers to move

from one to another, they must have been open at the ends. Mar-
quardt ^ and Stoffel ^ represent them as open both at the ends and
on one side, the roof being supported by the solid side and by posts
at the corners. Rich, who ignores the purpose for which vineae were
used in constructing the agger, says that a sufficient number of them '

were joined together in a line, and run up close to the walls, so that
the ram .could be securely plied
. . underneath them '. On this
. . .

theory, if the vineae were built in the way which Rich describes,
there would have been one ram for each vinea, and there would have
been no room to work the ram.^ StofEeF holds that vineae were
occasionally placed parallel to the wall and possibly Vegetius ^ ;

may have meant this when he said that several vineae are joined '

together in a row and afford safe shelter to the besiegers as they

advance to undermine the foundations of the walls but if so, '
:

these vineae must have been far stronger than those which he describes
in the same chapter. Is it not more probable that the besiegers
advanced through a row of vineae placed at right angles to the wall,
and that the work of undermining was done under cover of a testudo
^
or a musculus ?

Vineae must occasionally have been built of very stout timbers ;

for they served to protect the ynusculus which was used in the siege
of Massilia while it was in process of construction \vithin short range
of powerful artillery and for this purpose vineae like those which
;

Vegetius describes would have been absolutely useless.

II. The musculus is only mentioned once in the Gallic ^Yar and ;

there the reading is open to suspicion.^^ If it is right, Caesar says

^ De
re mil., iv, 15. ^ Diet,
of Rom. and Gk. Ant., 4th ed., p. 727.
^ See pp. 749-51. * R'6m. Staatsverwaltung, ii, 1884, p. 530.

5 Guerre civile, Atlas, Planche 10.

^ Of course the ram was really worked under the cover of a testudo or a

miLsculus, not of vineae, which would have been much too weak.
' Guerre civile, ii, 352. ^ De re mil., iv, 15.

8 See Guischard, Mem. crit. et hist., 1774,

pp. 455-08, and Stofifel, Guerre
ii, 352.
civile,
1"B. G., vii, 84, § 1. Musculos is only found in AQ : (p^ have 7nulculos,
which occurs nowhere else.
 —
THE SAPPERS' HUTS 609

that Verciiigetorix brought out musculi from Alcsia when he was

about to make his final attempt to break through the Roman lines.
The }inisculus,'ds used at the siege of Massilia,is described fully in B. 6'.,
ii, 10. It was a sapper's hut, 60 feet ^ long, 4 feet wide, and 5 feet high,
made of timbers so strong that no heavy weights thrown upon its
roof could break through. The roof was two-sided and sloping,
so that missiles might roll off. It was covered by a layer of unbaked
bricks and of clay :hides were spread over the bricks, to prevent
their being dissolved by water ;and above the hides again wet
mattresses or cushions, to guard against fire and to break the impact
of stones. This musculus was used for protecting Trebonius's soldiers
while they were trying to undermine the wall of Massilia, and perhaps
in the first instance for protecting the battering-ram.-^ As Caesar
describes it so fully and gives its dimensions, I believe that it was
a novel kind, devised for a special purpose. Vercingetorix must have
used his, if he used musculi at all, to protect his men while they were
attempting to fill up Caesar's trenches.^
III. The testudo used for protecting soldiers when they were filling
up ditches {quae ad congesHonem fossarum paratur), as described by
Vitruvius,* and by Athenaeus and Apollodorus, who call it x^^^^^^V
Xwo-rpts,^ was 25 feet square and mounted upon rollers and the
;

men who worked inside it were protected in front by a sloping roof

or, so to speak, shutter, which almost reached the ground, and
descended from the line joining the corners of the two sloping sides
of the roof, properly so called. Another kind (x€Xo)vr) SiopvKTpt^),
described by the same writers,^ was used for protecting soldiers
when they were undermining a wall. According to Marquardt,'^
the front was perpendicular, to enable the testudo to be rolled
right up against the wall ;and the line of the roof formed by the
two sides, which themselves sloped down to the ground, declined
backwards, in order that stones dropped on to the testudo by the
besieged might slide off harmlessly.

^ Lipsius {Opera, 1637, iii, 281) believes that Caesar wrote IX not LX. He
argues that it would have been impossible to procure beams 60 feet long, and
that the number IX would harmonize with the statement of Vegetius (iv, 16),
musculos dicunt minores macMnas. But Lipsius was not a practical man. It
was not necessary that the individual beams should be as long as the musculus ;
and the musculus itself was evidently of a special kind.
2 B. C, ii, 10, 11,
§§ 1-3 ; Stoffel, Guerre civile, i, 296-9.
^ It has been urged that he could not have transported sappers' huts from

his camp to the contravallation ; but did not the Nervii bring up testudines
to the attack of Q. Cicero's camp {B. G., v, 43, § 3) V
* X, 14.
^ The
illustration of the x^^<^^l x^(^^p'^^ hi Smith's Dictionary of Anti-
quilics, 808, which differs from that given by Marquardt {op. cit., p. 531), is
ii,

identical with the one in A. de Rochas d'Aiglun's Traitc de Fortification,

d'attaque et de defense des places, par Philon de Byzanco {Mem. de la Soc.
d' emulation du Doubs, 4" scr., vi, 1872, pp. 286, 365), and is taken from Heron
of Byzantium. See C. Wescher's PoUorcetique des Grecs, 1867, p. 211. The
editor of the Dictionary of Antiquities oii'ers an illustration oi the x^^'^'''?
diopvicTpis identical with that given by Marquardt (p. 529), which is borrowed
from Rustow and Koclily {Gesch. d. yriech. Kriegswesens, 1852, p. 207, fig. 83).
^ See C. Weschor, op. cit.,
pp. 19-20. ' Op. cit., p. 529.
1093 B, r
610 THE SAPPERS' HUTS
The Nervii used testudines when they were attacking Quintus
Cicero's camp.^ It is only in reference to this episode that Caesar
mentions testudines in the Gallic War but in the civil war Trebonius :

used a testudo 60 feet long to protect his men in levelling the ground
for the construction of the agger at Massilia ^ and Caesar may have ;

used one for the same purpose in Gaul.

I have done my best but I am aware that the result is unsatis-
;

factory. What I wanted to find out was (1) the nature of Caesar's
vineae (2) of the musculi which Vercingetorix is supposed to have
;

used ; and the testudines employed by the Nervii. About the

(3) of
vineae we do know enough but it is extremely unlikely that Vercin-
:

getorix's musculus was like the one which Caesar describes and the ;

testudines of the Nervii may, for aught we know, have been diilerent
from those described by Vitruvius. However, the testudines of the
Nervii and the musculi of Vercingetorix were intended to serve
practically the same purpose, —
namely to protect men in the attempt
to fill up trenches and to tear down ramparts the Nervii built :

their testudines under the direction of Eoman captives and it is ;

probable that they were built and protected against damage on the
same general principles, the nature of which has been sufficiently
explained. More detailed information it is impossible to obtain.
Again, the dimensions of the testudo which Caesar used at Massilia
differed widely from those given by Vitruvius Vitruvius borrowed :

his description from a Greek, Philo there is no evidence that :

Caesar's testudines were identical with those of the Greeks and ;

the descriptions of Vitruvius, of Athenaeus, and of Apollodorus are

so obscure in certain points that commentators differ widely in their
interpretation. All these difficulties are ignored, perhaps wisely, by
the compilers of dictionaries of antiquities and hand-books.
IV. The pluteus, as described by Vegetius,^ which was used to
protect soldiers when they were constructing siege works, was a
convex wicker shield with an arched roof, covered with hides and
running on three rollers."*
But although Caesar doubtless used plutei of this sort, it is not
certain that he uses the word in this sense in the Gallic War. In
B. G., vii, 25, § 1, he speaks of pluteos turrium, which Napoleon ^
explains as les mantelets protegeant les approches des tours '.
'

Long^ says that they were the planks on the " turres ", the breast-
'

works which protected the soldiers * and in support of this view he

;

refers to B. G., vii, 72, § 4, where Caesar uses the word to designate
the breastworks which protected the rampart in his line of contra
valla tion round Alesia. But there is no evidence that these plutei
were identical with the plutei turrium. Kochly,'' as far as I under-
1 i5. (?., V, 42, § 5. 5. C, ii, 2, § 4.
2 ^ DeremilA\,\o.

* Rich {Diet, of Ant., 4th ed., p. <513), who iiiisunderstaiids

Rom. and Gk.
Vcgetius, defines the pluteus which he mentions as a movable tower with a roof
'

overhead fixed upon wheels'.

. . .
^ Hist, de Jules Cesar, ii, 2G1.

« Caesar,
p. 349. Cf. A. von Goler {Gall. Krieg, 1880, p. 255, n. 9), who
supposes that the breastworks were themselves protected as far as possible
against tire by raw hide^.
' kScc von Goler, Gall. Krieg, 1880, p. 25G, note.
 THE SAPPERS' HUTS 611

stand liim, thinks that these latter were separate from the towers,
and placed some in front and others along the sides of the terrace,
in order to protect the workers. But if so, why were they called
plutei turrium V Koclily gets over this difficulty by deleting turrium,
which is too drastic a remedy for my taste. Curtius^ speaks of
plutei which were intended to protect the men who moved towers ;

and accordingly Schneider ^ concludes that the plutei used at Avari-

cum were identical with those which Vegetius describes. We must
not, he argues, regard them as breastworks attached to the stories
of the tower itself, because Caesar says that, in consequence
of the burning of the plutei, it was difficult for the Romans adire
apertos ad auxiliandum. I do not regard this argument as con-
clusive, because the men who are described as aperti might have been
the men who were stationed in the towers, and not those who moved
them. I am inclined, however, to believe with Schneider and
Napoleon that the plutei turrium were of the kind to which Curtius
alludes, because, if they had been breastworks attached to the towers
and had been burned, the flames would probably have consumed the
towers themselves.
In B. G., vii, 41, § 4, 72, § 4, Caesar uses the word pluteus in the
sense of breastworks, which were used as an extraordinary defence.
In the former chapter we read that Fabius, whom Caesar left in
command of his camp at Gergovia when he went to meet the Aeduan
infantry, added plutei to the rampart. In the other passage Caesar,
describing the line of contravallation which he constructed round
Alesia, says that he constructed a rampart and palisade 12 feet high,
'

which he strengthened by an embattled breastwork, with large

forked branches projecting along the line where the breastwork
joined the rampart' {aggerem ac vallum XII pedum extruxit. Huic
loricam pinnasque adiecit grandibus cervis eminentibus ad commis-
suras pluteorum atque aggeris). These plutei were identical with the
lorica pinnaeque, and perhaps also with the palisade.

THE FALX MURALI8

The falx muralis was used for loosening and dragging down the
stones and timbers in the wall of a besieged town,^ and was worked by
men who were protected by a testudo or musculusA According to
Vegetius,^ it was a wooden beam with a piece of iron at the end, bent
into the shape of a hook. A specimen, of which Daremberg and
Saglio give an illustration, was discovered in 1862 in the Gallic wall
of Vesontio (Besan9on).^

^ Cacsa materia pluteis faciendis, ut qui turrcs admoverent extra

. . . teli
ictum esseiit. De rebus (jedis Ahxaridri Mcujni, v, 3, § 7.
2 Caesar, ii, 407-8. 3 B. G., vii, 22,
§ 2.
* 8ee Stott'el, Guerre civile, ii, 354, and his Atlas, Planche 10.
* De re mil., iv, 14. « Did. des ant. yrccques el rom., ii, 970.

R r 2
 612

CAESAR'S BRIDGES
Caesar does not describe any of his bridges, except tlic one wliicli
he built over the Rhine in 55 b. c. He tells us, however, that Labienus
crossed the arm of the Seine which separated him from the island on
which Metlosedum was situated by lashing boats together and he ;

also says that he himself threw his army across the Saone (not far
from its junction with the Rhone) in a single day.i We may infer that
he generally used bridges of boats of some sort.^ His legionaries
crossed a narrow tributary of the Nile by means of long hollowed
out trunks of trees, which stretched from bank to bank, the hollows
being filled with a^.^er— probably earth or rubble to render them —
easily passable and during his second Spanish campaign the river
;

Baetis was hastily spanned by an impromptu bridge, which was

kept in position by means of baskets weighted with stones.^

' i^. 6'., i, 13, §§1-2; vii,58,§4.

'^
Cf. Frohlich, Das Krieyswesen Ciimrs, pp. "213-4, and Guischard, Mem.
crit. el hist., i, 45-52.
^ Bell. Alex., 29, § 4; Bell. 11 is p., 5, § 1. See also Fiolilich, p. 213.
 SECTION VII.—RELATING TO THE
NARRATIVE OF CAESAR'S CAMPAIGNS
THE ROUTES OPEN TO THE HELVETII
Caesar says that '
there were only two routes by which it was
possible for the Helvetii to leave their country {erant omnino itinera
'

duo quibus itinerihus domo exire possent),^ namely the route through
the Pas de I'Ecluse, which they actually took, and the route leading
across the Rhone into the Roman Province .2 On this question
Long ^ writes a sensible note. It has been objected,' he says,
'

'
to Caesar's text, that there are other practicable passes through the
Jura but that is nothing to the purpose. All he says is that there
;

was only one road through the Jura by which the Helvetii could
leave the country, encumbered with their women, children and
wagons. The Helvetii had formed their plan to go through the
Provincia, as the shortest and easiest way, and accordingly had
mustered ... in the neighbourhood of Geneva. If they had antici-
pated opposition, they might have mustered somewhere else, and
crossed by the road that leads to Pontarlier ... or by any other,
if there was any other wagon-road at that time, and at this season

of the year. But they would choose the shortest route to the San-
tones, and not the longest.' Desjardins, however, who believes that
the Helvetii dared not go through any of the more northerly passes,
for fear of Ariovistus, insists that detachments of the host went
through the more southerly. ' Le texte,' he argues, n'exige '

B. G., i, 6, § 1.
1

A. Klotz {Caesarstudien, p. 31) roundly asserts that Mommsen {Jahresb.

2

d. philol. Vereins zu Berlin, xx, 1894, p. 200) was undoubtedly right in

'
'

identifying the unum iter of B. 0., i, 6, § 1— the first of the two routes which
—
Caesar indicates with the pass of Pontarlier ; and he goes on to remark
that Mommsen was accordingly obliged to delete the words, inter montem
luram et fluinen Rhodanum, which identify the route with the Pas de I'Ecluse.
Meusel, as Klotz regretfully observes, is here conservative. He is, very wisely.
Behold the corner in which the revolutionary Klotz finds himself pinned.
Mommsen maintained that the Helvetii did actually march through the pass
of Pontarlier. But, objects Klotz {op. cit., pp. 33-4), in this case they could
not have ravaged the lands of the Allobroges and Caesar (i, 11, § 5) says that
;

they did. Therefore we must admit that they went through the Pas de I'Ecluse
[although we have just been told that Caesar pointed to the pass of Pontarlier
as the only alternative to the route through the Province !] ; and accordingly
the interpolator, who had an accurate knowledge of the events of the cam-
paign, but not of the geography, inserted the words, inter montem luram . . .

Rhodanum. Was ever conclusion lamer or more impotent ? And all because
Klotz submissively bowed to Mommsen' s pontifical dogmatism.
* Caesar, p. 44.
614 THE ROUTES OPEN TO THE HELVETII
nullement que la sortie ait ete accomplie en entier par le pas de
TEcluse.' 1 Desjardins is certainly wrong. What can be clearer
'

than the text ? Erant omnino itinera duo Belinquehatur una

. . .

'per Sequanos via, qua Sequanis invitis propter angustias ire non
poterant.

CAESAR'S LINES ON THE RHONE

I.Caesar says- that the distance from the Lake of Geneva ad
7nontem luram was 19 Roman miles, or about 28 kilometres. Des-
jardins,^ who assumes that the terminus of Caesar's entrenchment
was the Pas de I'Ecluse, finds fault with Napoleon for saying that
the distance is 28 kilometres, measured along the bank, and says that
it is really 28 kilometres in a straight line. Desjardins, however, is
himself mistaken for, as any one may see who will examine Sheets
;

150 and 160 of the Carte de VEtat-Major, the distance in a straight

line is only 20J kilometres. According to StofEel,* the distance,
'
following the sinuosities of the river,' is 32 kilometres. My own
measurement agrees more closely with Napoleon's but at all events :

Caesar was thinking of the distance following

—
'
as far as it was
'
—
necessary to do so the sinuosities of the river
' for how was he to ;
'

measure the distance in a straight line ? His measurements were

rough and ready, and we have no right to assume that by ad montem
luram he meant the Pas de I'Ecluse exactly but he was not far
;

wrong.
II. A lacu Lemanno, writes Caesar, qui in fiumen Rhodanum influit,
ad montem luram milia passuum XV 11 II murum in altitudinem
. . .

pedum sedecim fossamque perducit. J5. 6^., i, 8, § 1.

Napoleon points out ^ that Caesar's description is not to be under-
stood literally, because it would have been impossible for the Helvetii
to attempt the passage of the Rhone, between Geneva and the Pas de
I'Ecluse, with any prospect of success, except at a few places. Those
places are described in full detail by Napoleon, in a summary of the
report of Colonel Stofiel, who was sent by him to examine the ground ;

and they are marked accordingly in his map.

Long believes, in spite of Stoffel, that the lines were continuous,
(1) because Caesar says so, and (2) because continuous lines would
have been a better protection, as the Helvetii might otherwise have
climbed the banks.^ But how could they have done so, where the
banks were precipitous, with their wagons ? And, supposing that
some of them had crossed the river without the wagons and had
climbed the precipitous banks, they would also have been able to
climb the (assumed) rampart unless Roman soldiers had been there to
defend it while if they had been there, the bank would have served
;

as a natural rampart, and the Helvetii would not have been allowed

^ Geogr. de la Guide rom., ii, 601-2. - J5. G^., i, 8, § 1.

^ Geogr. de la Gaule rom., ii, 598, n. 5.
* Hist, de Jnles Cesar, ii, 49-51, note. ^ lb., ii, 48.
* Decline oj the Roman Bepuhlic, iv, 2, note. Cf. p. 225, n. 4, supra.
 CAESARS LINES ON THE RHONE 615

to climb. Caesar was not writing a technical treatise, but a popular

narrative and he doubtless expressed himself loosely, as he did, on
;

certain points of detail, in describing his works at Alesia. Dion

Cassius had the wit to perceive his meaning for he tells us that
;

Caesar fortified the most important points (ra eVt/catpoTaTa Siera-

cfypevae Koi dTreret^C^crer, XXXviii, 31, § 4).

WPTAT ROUTE DID CAESAR TAKE WHEN HE

MARCHED IN 58 b.c. VIA OCELUM AGAINST
THE HELVETII?
Caesar 1 says that he took the shortest route over the Alps into
Transalpine Gaul that the Ceutrones, Graioceli, and Caturiges
;

attacked him on the way that seven days after he left Ocelum he
;

reached the territory of the Vocontii and that he made his way
;

thence into the country of the AUobroges, and thence into the country
of the Segusiavi.
I have shown on pages 430-1 that Ocelum was close to Avigliana,
from which it follows that, in the first stage of his march, Caesar
moved along the valley of the Dora Riparia. On this point von
Goler and Napoleon are wrong. But they and almost all other
modern commentators are agreed that Caesar crossed the Mont
Genevre, and passed by Brigantio (Brian9on) and, as he went by
;

way of the Dora Riparia, he must have done so. From Brigantio
divergence begins. According to von Goler,^ Caesar subsequently
advanced by the left bank of the Romanche and the right bank of
the Drac to Cularo (Grenoble), where he crossed the Isere, and thence
to Vienna (Vienne) and Lugdunum (Lyons). This itinerary, says
General Creuly,^ contradicts Caesar's implied statement that he
crossed the country of the Vocontii and Long,* who agrees with
;

this argument, concludes that Caesar took the comparatively easy

—
'

road '
—
from Brigantio through Embrun, Gap, Die (Dea) to
'

Valence (Valentia) on the Rhone.' The route traced by Kiepert ^

leads from Brigantio to Cularo along a slightly different line from
that indicated by von Goler from Cularo to Lugdunum his route
:

and von Goler's coincide. Napoleon ^ also takes Caesar to Cularo

;

but thence he leads him by a nearly direct route to a point on the

Rhone a few miles above Lugdunum. His route and Kiepert's are
both open to the objection which Creuly makes to the theory of
von Goler.
As far as Brigantio, I repeat, the route is certain. As Caesar's
objective was that part of the country of the Segusiavi which ex-
tended between the Rhone and the Saone, in the neighbourhood of
1
B.G.,\, 10, §§ 3-5. 2 Q^^ll Xrieg,
1880, p. 13 and Taf. I.
^
Rev. arch., nouv. ser., viii, 1863, p. 255.
* Decline of the Roman Republic, iv, 0.
* Tabula Galliae Cisalpinae el Transalpinae in usum scholarum descripta.
* Hist, de Jules Cesar, ii, 57, note, and Planche 19.
616 CAESAR'S ROUTE VIA OCELUM
Lyons, be obvious to any one who consults a good map ^ that
it will
his shortest route would have led past Grenoble, i£ between Briangon
and Grrenoble there was a practicable road and, at the time when
;

the Peutinger Table was constructed, there certainly was.^ From

Brianyon to Grenoble the modern traveller must go either by the
valleys of the Guisane, the Romanche, and the Drac, or, by a longer
road, past Embrun, Chorges, and Gap, and thence down the valley
of the Drac. But it is doubtful whether either of these routes would
have led Caesar into the country of the Vocontii.^ I am therefore
disposed to believe that he did not pass Grenoble at all, but took
the road, indicated by Long, which leads past Embrun^ Chorges, Gap,
and Die.

CAESAR'S CAMPAIGN AGAINST THE HELVETII ^

I. The route which the Helvetii pursued, after threading the Pas

de I'Ecluse, to the Saone, cannot be traced exactly. We only know

that they passed through the outlying territory of the Allobroges ^
on the northern bank of the Rhone, and that they struck the Saone
not far north of Lyons. Napoleon, following Valentin-Smith,^
assumes that they moved along the right bank of the Rhone as far
as Culoz, and then struck off to the west, along the line of the Roman
road which led past Virieu-le-Grand, Tenay, and St. Rambert, and
across the plateau of Dombes to the Saone. By following this route,
says A. Bernard,"^ they would have avoided incommoding the Sequani.
Spreading out, as Valentin-Smith thinks, in the latter part of their
route, they crossed the Saone at various points between St. Bernard
and Montmerle, or, according to C. Cadot^ and Thomann,^ at Ville-
franche, Messimy, Montmerle, and Belleville. Cadot points out that
above Belleville access to the river would have been barred by des '

forets inextricables, dont le defrichement, assez recent, n'est meme

pas completement acheve ', and he argues that if the Helvetii had
crossed the Saone at one spot, they would have opposed Caesar's
passage. A. Senault ^^ also brings them to St. Bernard, but by a
different route, namely Nantua, La Cluse, and Brion. He maintains
that they could not have marched along the bank of the Rhone,
because, if they had done so, Caesar, instead of going to Italy for
reinforcements, would have attacked them in flank. How Caesar

1 The best is Feuille 60 of the Carte de France (1 200,000).

:

'
Desjardins, Geogr. de la Gauh rom., vol. ii (map facing p. 224), vol. iv, p. 15").
3 See pp. 501-2.
* See Carte de France (1 200,000), Sheets 41 and 47, and Carte de VEtat-
:

Major (1 : 80,000), Sheet 136, NE. and SE.

5 See pp. 363-5.
^ Fouilles dans la vallee du Formans en 1862, map facing p. 6.
' Mem. de la Soc. Boy. des antiguaires de France, nouv. ser., xviii, 1846,
pp. 372-3.
^ Note sur V invasion des Helvetes, 1862,
pp. 6-8, 11.
^ Neue Jalirh. /. Philologie, &c., xci, 1865, p. 697.
i"
UCEuvre de Jacques Maissiat, 1892, pp. 24-5, 28-9.
 CAMPAIGN AGAINST THE HELVETII 617

was to attack in flank a vast host from which he was separated by

a broad and rapid river, I cannot see nor, even if such an operation
;

had been practicable, would he have attempted it with his insignifi-

cant force.
M. Jullian,^ on the other hand, believes that the Helvetii followed
the direct route, which passes through Chatillon, Nantua, and Bourg,
and crossed the Saone at Macon. He argues that if they had taken
the road indicated by Napoleon, the Allobroges would have been
able to stop them from plundering their villages, and that unless
they had moved away from the Rhone after threading the Pas de
I'jfecluse, Caesar would not have felt safe in leaving Geneva. He
selects Macon as the place of crossing, first because it is on the direct
road from the Pas de I'Ecluse to Toulon-sur-Arroux, near which (as
we shall presently see) the decisive battle of the campaign was
fought secondly, because the Helvetii could only have found the
;

necessary boats at a frequented spot and thirdly, because the main

;

roads that lead into the interior of the Aeduan territory converge at
Macon. Moreover, he doubts whether the road by which the Helvetii
must have approached the Saone if they crossed it between Belleville
and Trevoux passed through Aeduan territory. I do not think that
these arguments are valid. Why should the Allobroges have been
able to prevent a powerful army, which, even when it had lost
a fourth of its strength, Caesar only overcame after a desperate
struggle, from plundering their villages in one direction rather than
in another ? Caesar was obliged to leave Geneva for he could not ;

attack the Helvetii until he reinforced his feeble army. Macon is

certainly on the direct route from the Pas de TEcluse to Toulon ;

but, as we shall presently see, there is reason to believe that the

Helvetii, after they crossed the Saone, moved up its right bank until
they reached Macon. Surely the necessary boats might have been
found between Belleville and Villefranche, which are both on great
roads ; and such boats as the Helvetii found were not sufficient, for
they used rafts as well.^ Why should we suppose that they entered
Aeduan territory before they crossed the Saone ? And if they had
crossed it at a place so renowned as Macon, which Caesar mentions in
vii, 90, § 7, would he not have said so ? But the most cogent reason
for deciding that they crossed it south of Macon must be reserved
for the next section of this article.
11. Where did Caesar defeat the Tigurini the rearguard of the —
—
Helvetii and from what place did he march against them ? These
questions are closely connected with that which I have just dis-
cussed.
De Saulcy makes Caesar cross the Rhone at Vienne in his march
from Italy, then cross the Saone, and finally recross it after defeating
the Tigurini.^ But this itinerary is obviously wrong. Why should
Caesar have three times performed the troublesome operation of
crossing a river when it was only necessary to do so twice? De

^ HU. de la Gaule, iii, 205, n. 1 ; 207, n. 1. 2 ^ q^ |^ 12^ § 1_

2 Les campagnes de Jules Cesar dans les Gaules, 18G2, p. 287.
618 CAESAR'S CAMPAIGN
Saulcy misunderstands the passage (Caesar) ah AUobrogihus in Segu-
siavos exercitum ducit. Hi sunt extra provinciam trans Bhodanum
frimi^ These words mean that Caesar, after crossing the Rlione,
found himself in that part of the territory of the Segusiavi which lay
between the Rhone and the Saone. General Creuly, however, agrees
with de Saulcy. He says that in Segusiavos must mean le veritable '

pays segusiave, le Forez ', even if the Segusiavi possessed any terri-
tory on the eastern bank of the SaOne.^ I have proved on pages 470-1
that they did possess territory on this bank. Why should not that
part of their territory have been veritable as well as the other ?
' '

Creuly 's theory leads him to an absurd conclusion. He is forced to

make Caesar cross the Saone in order to attack the Tigurini, although
Caesar does not say one word about any such crossing. If Caesar
did cross the Saone, he must either have made a bridge or found one.
But if so, he would surely have recrossed the Saone, after destroying
the Tigurini, by the same bridge. Yet he tells us that, in order to
cross from the eastern to the western bank, he was obliged to make
a bridge.^ It is therefore clear that he crossed the SaOne once only,
and that that part of the country of the Segusiavi in which he
encamped on his return from Italy was on the eastern bank of
the river.
When he set out on his night march to attack the Tigurini he
was somewhere in this territory. Valentin-Smith argues that his
camp could not have been so far north as Trevoux, because Trevoux,
being situated between two places called Amberieux, must have been
in the country of the Ambarri ^ and this may be a reasonable con-
;

clusion. Now in the country south of Trevoux the most suitable

spot for a camp is on the heights which command Sathonay. The
Tigurini were attempting to cross the Sadne at some point where it
was so sluggish that one could hardly tell, by merely looking at it,
in which direction it was flowing. Caesar left his camp soon after
midnight, doubtless with the intention of attacking the Tigurini
before sunrise, and succeeded in surprising them as they were attempt-
ing to cross the Saone .^ We may infer, then, that he had not far to
march, and that the lie of the ground had masked his approach.
1. De Saulcy believes that the defeat of the Tigurini took place

a little north of Macon. ^ But that it did not take place so far north
as this, still less at Chalon, where it has also been placed, is proved
by the following facts. First, the current of the Saone answers most
closely to Caesar's description in that part of its course which lies
between Trevoux and Thoissey secondly, Caesar would not have
;
'^

taken the trouble to leave his camp in the country of the Segusiavi
at midnight, in order to march to a spot which he could hardly have
reached at any time on the following day and thirdly, if the Helvetii
;

had started from a point so far north as Macon on their march

^ B.G.,i, 10, § 5,
- Rev. arch., nouv. ser., viii, 18C3, p. 256.
^ B. G., i, 13, § 1. * Fouilles dans la vallee du Formans, p. 5.
» B. G., i, 12, §§ 1-3.
® Les campagnes de Jules Cesar dans les Ganles, pp. 289-94,
' Ch. Cadot, Note sur Vinvasion des Helvetes, pp. G-8, 10.
 AGAINST THE HELVETII 619

towards a point south-east of Mont Beuvray, they would not

have taken a fortnight or more to accompHsh so short a distance.^
M. Jullian,2 indeed, argues that when Caesar began his march against
the Tigurini he was perhaps not in the country of the Segusiavi, but
had already made one or two marches northward from the camp
which he had occupied there. This seems to me utterly inconsistent
with his narrative if the Helvetii had crossed the SaOne, and the
:

Tigurini had been about to cross it, at Macon, surely he would have
written in Amharros, not in Segusiavos (exercitum ducit).
2. Napoleon places the attempted passage of the SaOne by the
Tigurini at the point where it is joined by the Formans.^ If their
encampment was not more than a few miles north of Trevoux, the
route by which they had approached the Saone must have been the
valley of this stream. This valley is dominated on the left by hills
which would have screened the Roman column from observation as
it marched from Sathonay. Napoleon, however, simply asserts that
'
the excavations carried on in 1862 leave no doubt of the place of
this defeat '. The excavations were carried on in the valley of the
Formans, on the plateau of St. Bernard, and at the hamlet of Cormoz.
The finds, which included cinders, human bones, fragments of flint
weapons, pottery, bronze bracelets, a bronze sword, and a couple
of iron weapons, belonged to all periods from the neolithic to that of
La Tene * and no archaeologist would allow that they lend more
;

than the flimsiest support to Napoleon's theory. Moreover, Des-

jardins objects that, if the Tigurini had been encamped near Trevoux,
Caesar, who, according to Napoleon, was only 18 kilometres off at
Sathonay, would not have needed scouts to tell him what they were
doing and that the Tigurini would not have been such fools as to
;

attempt to cross the Saone under the eyes, so to speak, of six Roman
'

legions '.^ But these objections have no force. Caesar could not
have seen what the Tigurini were doing 18 kilometres off and if he ;

wanted to know, he had no choice but to employ scouts. If the

Tigurini had attempted to move their unwieldy wagons further up
the bank, they would have gained nothing, for Caesar could have
overtaken them whenever he pleased and, whether they were fools
;

or not, they certainly attempted to cross the Saone at a point which

was within a night's march from Caesar's camp. By far the strongest
argument, indeed, which can be urged in favour of Napoleon's site
is the purely geographical one which I have already stated.

3. General Creuly believes with Heller that Caesar crossed the

Saone at Belleville ^ but it is much more likely that the Tigurini
;

had approached the river by the valley of the Formans.

III. Assuming that Caesar defeated the Tigurini in the valley of
the Formans, and there crossed the Saone, we have next to inquire
by what route the Helvetii marched to the scene of their final over-
throw. Their object was to reach the country of the Santoni, that —
1
B. G., i, 15, § 5 ; 16, §§ 1-3 ; 23, § 1. « Hist, de la Gaule, iii, 207, n. 3.
^
Hist, de Jules Cesar, ii, 60, 61, n. 1.
*
UAiithr., xvii, 1906, p. 336. ^ Geogr. de la Ganle mm., ii, 606.
"
Rev. arch., noiiv. ser., viii, 1863, pp. 257-8.
620 CAESAR'S CAMPAIGN
isto say, the valley of the Charente. After telling us that thev
marched for a fortnight about five miles ahead of his vanguard, Caesar
remarks that they had moved away from the SaOne, iter ah Arare
averterant. He does not say when they began to move away but ;

the phrase which he uses seems to imply that for some time they had
marched parallel with the river. When he had reached a point not
more than 18 Eoman miles from Bibracte (Mont Beuvray), he
changed his direction and marched towards Bibracte. It is there-
fore clear that the general direction of the march up to that point
had been towards the north-west.
1. Napoleon's route leads by way of Belleville, over the Col
d'Avenas, through the valley of the Grosne, past Cluny to St. Vallier,
thence westward across the river Arroux, about three miles south of
Toulon, past Issy-l'fiveque and Mont Tauffrin to Remilly on the
Alene. From a point near Remilly, he thinks, Caesar struck off for
Bibracte but he does not say by what road the Helvetii marched
;

back to attack Caesar.^

Heller, who, like Napoleon, makes Caesar diverge from the Saone
near Belleville, argues, in a criticism of the map published by the
Commission de la topographie des Gaules, that if he had marched up
the valley as far as Chalon, he could have procured grain from his
boats on the river and would never have felt any anxiety about his
supplies, since, for the short distance between Chalon and the scene
of the final overthrow of the Helvetii, the legionaries could have
carried enough food on their backs.^
2. General Creuly so far agrees with Heller that he admits the
absurdity of the theory which makes Caesar march up the valley of
the Saone as far as Chalon but he holds that he must have gone
;

as far as Macon, remarking that his words, quod iter ah Arare Helvetii
averterant, imply thafc he had pursued the road along the valley for
a considerable distance after crossing the river. Moreover, if the
Helvetii, whom Caesar followed, had diverged from the Saone at
Belleville, they would have found themselves walled in between
abrupt hills, on the flanks of which it would have been impossible
to deploy.^ From Macon Creuly suggests that Caesar followed the
Helvetii by way of Cluny, Joncy, and St. Eusebe or Blanzy.* The
latter part of this route, as the reader will presently see, is too
far east.
3. Stoffel believes that the Helvetii marched up the right bank of
the Saone till they neared Macon then struck ofi in a north-westerly
;

direction towards Prisse followed the line of the modern road lead-
;

ing from Macon to Autun by way of Cluny, Salornay, and Mont

St. Vincent, where they were at the lowest point of the mountains
which separate the valley of the Saone from the valley of the Loire ;

and thence turned westward, past Sanvigne to Toulon-sur- Arroux.

^
Hist, de Jules Cesar, ii, 62-6 and Planche 4.
^ Philologus, xix, 1863, pp. 558-9. See also Ch. Cadot, op. cit., p. 13.
' 76., pp. 12-3. * Bev. arch., nouv. ser., viii, 1863,
pp. 259-60.
Guerre civile, ii, 443-4. The route traced by Cadot {op. cit., pp. 12-7)
*

coincides with that of Stoffel as far as Mont St. Vincent. In tracing tlie
 —
AGAINST THE HELVETII 621

The tracing of the last few miles of the route must depend upon the
identification of the battle-field.
has indicated the route accurately
I believe that Stoffel for, as ;

I shall presently show, his identification of the battle-field, certain

topographical details excepted, is almost certainly correct and if ;

the Helvetii moved up the Saone as far as Macon, as they must have
done unless they quitted it at Belleville,^ no other route was avail-
able. It remains to look for the scene of the battle.
IV. Caesar, in his description of the battle, says that, when the
Roman army, formed in three lines, was pursuing the Helvetii, after
their first repulse, the Boi and Tulingi attacked the Romans on
their exposed flank, ex itinere nostras {ah) latere aperto adgressi
circumvenireP' The meaning of the words (ah) latere aferto has been
much discussed, because it affects the question of the identity of the
battle-field.^
are usually taken to mean on the right flank ',
The words '
un- —
protected, because the shield was worn on the left arm and Napoleon ;

quotes, in support of this interpretation, a passage in which Livy

describes the escape of 600 Roman soldiers from Cannae. The 600
were formed in the shape of a wedge and, says Livy, ; As the '

Numidians aimed at their right flank, which was exposed, they

changed their shields to their right arms, and escaped to the number
of six hundred into the greater camp * {cum in latus dextrum, quod
'

patebat, Numidae iacularentur, translatis in dextrum scutis, in maiora

castra ad sexcentos evaserunt).^ I do not regard this passage as
decisive, because the 600 were not fighting in line, but in the shape
of a wedge and therefore they were of course completely exposed,
;

on the unshielded side, along the whole depth of the formation.

F. Frohlich ^ quotes two passages in support of the orthodox view.
In the former we read that when the legions were fighting outside
the wall of Gergovia, they were panic-stricken by the sight of an
Aeduan force, which they mistook for an enemy, suddenly appearing
on their exposed flank (suhito sunt Haedui visi ah latere nostris aperto)?
In the latter we read that, at Pharsalia, Pompey determined to
attack Caesar's right wing on its exposed flank (persuasi equitihus
nostris .ut
. . dextrum Caesar is cornu ah latere aperto adgrederentur).^
. . .

later stages, however, he argued on the untenable hypothesis that Bibracte

was on the Autun.
site of
1 See Carte de France (1 200,000), Sheet 47.
:

2 B. G., i, 25, §§ 6-7.

^ In the first edition
(p. 017) I remarked that Caesar, if the MSS. are right,
'

does not use the phrase ah latere aperto, but simply latere aperto\ This,
I suggested, might mean that the Boi and Tulingi attacked the Romans
'

in flank, as their flank was exposed ; and I added that if so,

' there would
'

be nothing to show whether the exposed flank was the right or the left'.
Professor H. Delbriick {Qesch. d. Kriegskunst, i, 440) admits the possibility
of this tentative interpretation but I have no longer any doubt that Meusel
;

{Jahresb. d. jjhilol. Vereins zti Berlin, xi, 1885, p. 201) is right in supplying
ab before latere aperto.
* I modify Messrs. Church and Brodribb's translation.

^ xxii, 50,
§ 11. 6 jjdg Kriegsivesen Gcisars,
p. 225.
' B. 0., vii, 50, « B. C, iii,
§ 1. 8(5, § 3.
622 CAESAR'S CAMPAIGN
The exposed flank ', in each of these cases, is universally admitted
'

to have been the right flank and Frohlich infers, perhaps hastily,
;

that ah latere aperto always means on the right flank '. '

Stoffel ^ At Gergovia, he observes,

ridicules Frohlich's arguments.
the Roman was covered by Sextius's legion, and therefore the
left
right, which was unprotected, was of course^ on this particular occasion,
the latus apertmn. In the battle of Pharsalia, Caesar's left rested on
the Enipeus, and his right was exposed in the Pharsalian plain
'^
;

therefore here too his right flank was the exposed flank, not because
it was the right, but simply because it happened to be exposed.
Again, Stoflel asks, if the right flank of a line of battle rested upon
a river and the left were uncovered in a plain, how could the soldiers'
shields prevent an enemy from turning that flank ? Every soldier
would call the left flank of an army so situated its exposed flank.
To prove his point, Frohlich ought to have shown that no Latin
writer ever called the left flank of an army latus apertum.
With all respect, however, for the professional knowledge of
Stoflel, his reply, at least in so far as it relates to Gergovia, is incon-
clusive. Let him look at Napoleon's (in other words, his own) map
of Gergovia, and he must admit that, if the Roman left was covered
by the legion of Sextius, the Roman right was equally covered by
the 10th legion. As a matter of fact, neither the left nor the right
was, strictly speaking, covered at all, except during the retreat.
But, if Stoflel's reply is inconclusive, so is Frohlich's argument, at
least in so far as it relates to Pharsalia. Heller ,3 indeed, argues that
the right flank of Caesar's right wing was covered by his cavalry,
and yet was called latus apertumA But to this I reply that it was not
until Caesar's right wing had become exposed by the rout of his
cavalry that Pompey's cavalry attacked the right wing ab latere
aperto.
Nevertheless, there are several passages in Caesar which show
that ab latere aperto does mean on the right flank '. In B. G., v,
'

35, § 2, describing the destruction of Sabinus's division by the

Eburones, he tells us that one of the Roman cohorts occasionally
charged forth from the hollow square in which the division was
formed. Meanwhile,' he continues,
'
the cohort was necessarily
'

exposed, and missiles fell on its exposed flank (interim earn partem '

nudari necesse erat et ab latere aperto tela recipi). Now here is a fact
which Stoflel would find it difficult either to contradict or to explain
away that cohort was equally uncovered, in the modern sense of
:

the word, on its left flank and on its right. Therefore, unless ab
latere aperto means ab utroque latere (on both flanks), either it must
have had a technical meaning, which Caesar's Roman readers would
have at once understood, or it must have conveyed no meaning what-
ever Ab latere aperto may mean ', says Long,^ that the cohort
!
' '

was altogether exposed after leaving the " orbis ".' Of course it
^ Eev. de philologie, xv, 1891, pp. 139, 144-5.
^
Cf. my article in Class. Quarterly, ii, 1908, pp. 271-92.
^ Philologus, xxvi, 1867, p. 659.
' B.C., iii, 93, §§ 3-4. 6 Caesar, p. 252.
 AGAINST THE HELVETII 623

was, — as regarded position. ah latere aperto meant what Long

But if

suggests, why did not Caesar write ah utroque latere (or ah laterihus)
and make his meaning clear ?

In B. G., vii, 82, § 2, Caesar describes the night attack which the
GaUic army of relief made upon the Roman line of circumvallation
in the plain on the west of Alesia. Towards daybreak the Cauls
retreated, for fear they might be attacked on their exposed flank by
a Roman force sallying forth from one of the camps on the high ground
(veriti ne ah latere aperto ex superiorihus castris eruptione circum-
venirentur). The Gallic left was exposed, as regarded position, no
less than the Gallic right. There were Roman camps on high ground
on the left and also on the right. Either, then, ah latere aperto signified
the right and unshielded flank, or it signified nothing. There is no
escape from this conclusion unless ah latere aperto means ah utroque
latere or ah laterihus.
But if Caesar had meant ah utroque latere or ah laterihus, surely he
would have said so ? It would have been so easy to make his mean-
ing clear and he does use the phrases ah utroque latere and ah
;

laterihus'^ when they are required. According to StofTel, ah latere

aperto means either on the left flank or on the right flank ', as
' ' '

the case may be in other words, Caesar does not take the trouble
:

to say which flank he means. Perhaps. But it would have been so

easy to say ah latere dextro or ah latere sinistro and he does use the ;

j)hrase ah dextro latere.^ Is there not some ground, then, for arguing
that ah latere aperto had a fixed technical meaning ?
One word more. Thucydides,^ in his description of the battle of
Mantinea, tells us that, All armies, when engaging, are apt to
'

thrust outwards their right wing and either of the opposing forces
;

tends to outflank his enemy's left with his own right, because every
soldier individually fears for his exposed side, which he tries to cover
with the shield of his comrade on the right, conceiving that the closer
he draws in the better he will be protected. The first man in the
front rank of the right wing is originally responsible for the deflec-
tion, for he always wants to withdraw from the enemy his own
exposed side, and the rest of the army, from a like fear, follow his
example.' * Does not this passage lend some support to the view
that ah latere aperto means on the right flank ?' '

Lastly, Lucan, in his description of the rash attack which some of

Caesar's troops made upon those of Afranius and Petreius at Ilerda,
calls the left flank of the Roman cavalry
'
their protected flank'

(munitum latus),^ which is an additional argument in support of my

view, that apertum latus was a technical phrase, meaning the right '

flank '.6
' B. G., ii, 8, §§ 3-4 and
vii, 24, § 3.
'^
lb., 49, § 1. A
third passage, which is strictly analogous to the two whioh
I have just examined, is to be found in iv, 26, § 3.
' V, 71,§1.
* B. Jowett, Thucydides, Translated into Endish, i, 389.
5 iv, 43-5.

^ Since my first edition was published Frohlich [Die Glaubwlirdujkeit Cacsar^\

&c., p. 29) and H. Dclbriick [Gesch. d. Kricgskimst, i, 440) have remarked

 624 CAESAR'S CAMPAIGN
•
The according to Caesar,^ was not more than 18
battle-tield,
Roman miles irom Bibracte, or Mont Beuvray, and close to a point
where a road leading to that town diverged from the route by which
the Helvetii would have marched to Saintonge.^ This is all that is
certainly known but we may be sure that it was somewhere to the
;

south-east, south, or south-west of Bibracte, because the Helvetii

could have had no motive for passing by the east and north of that
town, in order to reach the Loire.
1. De Saulcy, writing before the results of M. Bulliot's investiga-
tions had been published, identified Bibracte with Autun, and
placed the battle-field near Cussy-la-Colonne, 16 miles, in a direct
line, north-east by east of Autun and about 25 miles east-north-east
of Mont Beuvray .2
2. Other writers, also assuming the identity of Bibracte with
Autun, indicated various sites more than 18 Roman miles from
Beuvray, and remote from the route which Caesar must have followed.^
3. Von Goler,^ who also argues on the untenable hypothesis that
Bibracte was at Autun, concludes that when Caesar struck off towards
Bibracte, he had reached Chateau-Chinon. But Chateau- Chinon is
at least 9 miles further north than Beuvray and than Autun.
Kiepert,^ who makes the Helvetii march along the right bank
4.
of the Saone as far north as Chalon, and then turn sharply to the
west, places the battle-field about 8 miles east by south of Autun.

that I have demonstrated, against Stoffel, that ab latere aperto means on '

the right flank '. in the first edition, however, I overlooked a passage which
might perhaps be regarded as supporting Stoffel' s view. In the operations near
ilerda, wlien Afranius and Petreius were encamped on the hill of Gardeny,
Caesar, who was encamped opposite them, probably on their north-west,
attempted to seize a knoll, now called the Puig JBordel, about midway between
Ilerda and the Pompeian camp. Accordingly he formed three legions in lino
of battle, and sent a picked force of antesiynani, belonging to the lith legion,
to seize the knoll {B. C, i, 43, § 3 ; 46, § 4). The enemy, however, were too
tj[uick for him ; and their outposts beat off the antesignani and drove them
back upon the legions. Caesar goes on to say that his men were afraid of
being attacked on their exposed flank {ab aperto latere) ; that the antesigimni
were thrown into confusion ; that thereupon the legion posted on that wing'
'

{qaae in eo cornu constiterat) abandoned its post and took refuge on a hill close
by ; and that he then sent the 9th legion to the rescue {ib., 43-4, 45, § 1). Stoffel
{Guerre civile, i, 51, 269) affirms that the 14th legion was on the left of Caesar's
line, and that the 9th was therefore on its right. If so, what was the hill to
which the 14th fled ? Stoffel {ib., i, 52) says la hauteur de las Collades",
'

which on his ma^^ (j)!. 5) is invisible. It has been argued that as the right
flank of the 14th was covered by the 9th, its latus apertum must have been its
left (A. von Goler, Gall. Krieg, 1880, ii, 39, n. 1). But at the moment when the
antesiynani feared for their exposed Hank they were isolated from the rest of
the line.
1 B. G.y i, Cf. C. Jullian, Hist, de la Gaule, iii, 213, n. 1.
23, § 1.
I"
^ Les campagnes de Jules Cesar dans les Gaules,
pp. 317-74. Even Signor
Ferrero {Grandezza e decadenza di Roma, ii, 1902, p. 14), misled by his anxiety
to reconstruct Caesar's narrative, has committed himself to this absurdity-,
although he rightly identifies Bibracte with Beuvray.
* Sec Rev. des tioc. savanies, 3" scr., iv, 1864,
pp. 120 ii., and M. JuUian's
bibliography in Hist, de la Gaule, iii, 213, n. 1.
5 Gall. Krieg, 1880, p. 24.
•^
Galliae Cisalpinat et Transulpinat . . . tabula in usum schularuin dcacripia.
 AGAINST THE HELVETII 625

But the Helvetii would have needlessly increased the length of their
journey by taking this route and I have proved that they did not
;

follow the line of the Saone as far as Chalon.

5. Heller insists that the battle must have taken place on the east
of Bibracte for, he argues, if the Helvetii had got westward of that
;

town by the time when Caesar moved off towards it, they would have
tried to push on as far ahead of him as possible, and would not
have turned back, and if they had been on the west of Caesar when
they changed their line of march, they would not have brought back
their wagons. His view is that they only attacked Caesar in order
to force a passage towards the west.^ But he overlooks several
important considerations. First, the Helvetii must have known that
Caesar could easily overtake their unwieldy host. Secondly, they
would have taken their wagons with them to the battle-field in any
case partly because they would not have been so foolish as to leave
:

them alone and unprotected with the women and children partly, ;

as we may gather from Caesar's description of the battle, to serve as

a laager. Thirdly, even if they had been on the east of Bibracte,
they would not have been obliged to attack Caesar when he moved
:

off towards Bibracte, they would only have had to moVe on and
leave him to his own devices. Finally, if they were actuated by the
motive which Heller imputes to them, why did Caesar impute to
them a motive wholly different ?"^
6. Napoleon ^ points to a site on the rivulet of La Roche, about
7 miles, in a direct line, south-south-west of Mont Beuvray and he ;

identifies the proximus collis, on which Caesar formed his line of

battle, with a hill between the villages of Grand-Marie and Petit-
Marie. But 7 miles would hardly have been called not more than '

18 miles and, as M. Jullian observes,* Napoleon's site was not,

'
;

apparently, on a natural route.

7. Stoffel claims to have established the identity of the site
beyond all doubt and he has certainly written a most valuable
;

and interesting essay on the campaign.^ He tells us that he was

commissioned by Napoleon to search for the battle-field that, after ;

a careful study of the country round Mont Beuvray, he selected two

possible sites, —
the one which Napoleon adopted, and Montmort,
about 3 miles north-west of Toulon-sur-Arroux ^ that he himself
;

very decidedly preferred the latter but that, as his choice involved
:

a heterodox interpretation of the phrase (ah) latere aperto, he deferred

to the authority of certain scholars who assured the emperor that
the phrase in question could only mean on the right flank '. He
'

goes on to say that in later years he became so strongly convinced

of the truth of his original opinion that he determined to put it to
a practical test. He believed that the hill on which Caesar formed

1 Philologus, xxvi, 1867, p. 658. 2 jg

q^^ i^ 23, § 3.
Hist, de Jules Cesar, ii, 68, 71, n. 1.
•''

* Hist, de la Gaule, iii, 213, n. 1. ^ Guerre civile, ii, 439-52.

^ I suspect that Stoffel was indebted to Xavier Garenne, who had aheady

[Bibracte, 1867, pi. 1 [p. 16] and pp. 34-62) pointed to Montmort as the battle-
field.
1093 S S
626 CAESAR'S CAMPAIGN
his line of battlewas Armecy, just south of Montmort, and that the
hill to which the Helvetii fell back after the failure of their first
attack was just north of that village. Concluding from Caesar's
narrative ^ that an entrenchment had been thrown up on the plateau
of Armecy, he in 1886 set navvies to work, who presently discovered
the remains of an entrenchment. West of Montmort, in the neigh-
bourhood of a farm called La Bretache, numerous fragments of
Gallic pottery were found and in 1889 on the hill of Armecy itself
;

were discovered nine trenches, filled with ashes and charcoal, and
containing bones, which crumbled under the touch.
Now there is not the slightest doubt about the genuineness of
Colonel Stoffel's discovery. His good faith is above suspicion his :

skill as an excavator has been proved beyond question.^ The en-

trenchment which he discovered is not that of a camp it can only :

have been intended to serve a temporary purpose. Its shape is that

of a crescent, the width from horn to horn being about 300 yards.
The colonel points out that the work of entrenching must have been
hurriedly going on while the battle was raging on the lower slope of the
hill ;and, he remarks, ainsi s'explique que les fosses soient simples
'

sur divers points, doubles sur d'autres.' The depth of the trenches
was only 1 metre 50, or about 4 feet 11 inches. If the entrench-
ment was not that which Caesar mentions, it is difficult to account
for its existence. Since the colonel made his discovery, M. Carion,
mayor of Montmort, has found calcined bones and debris of Gallic
weapons hard by the entrenchment.^ The distance of the hill of
Armecy from Mont Beuvray tallies with Caesar's statement, that on
the morning of the battle he was not more than 18 Roman miles from
Bibracte. But can Stoifel's view be reconciled with the words {ah)
latere aperto ? If my interpretation of those words, which has been
accepted on the Continent as conclusive, is correct, then either
(1), as I suggested to General Sir Frederick Maurice, who was in-
clined to agree with me, the Boi and Tulingi may have worked round
to their left so as to strike Caesar's right flank or (2) Stoffel's theory,
;

so far as it concerns the hill to which the Helvetii fell back, must be
wrong.
Frohlich ^ admits that the entrenchment proves that a battle was
fought hard by but he maintains that it was not made by Caesar.
;

It would have been useless, he argues, as a protection against the

Helvetii, for they could easily have turned it if they had beaten
;

Caesar's veteran legions, they would have attacked the hill from the
rear as well as in front and therefore it may be assumed that the
;

entrenchment which Caesar constructed was not crescent-shaped but

a complete enclosure. I would, however, ask Frohlich to consider
whether the men who threw it up, and w^ho would certainly have
begun it on the side which was most exposed to attack, may not
^ Sarcinas in unum locum coiiferri et eum ab his qui in superiore acie con-
sbiterant muniri iussit. B. G., i, 24, § 3.
- See pp. xxv-xxvii.
^ Mem. de la Soc. eduenne, xx, 1892, pp. 304-5.
* Die Glaitbwurdiykeit Caesars, &c., p. 30.
 AGAINST THE HELVETII 627

have stopped work when they saw that the battle was going in
favour of their comrades.
Colonel H. Bircher,i on the other hand, holds with Stoffel that
the irregularities in the entrenchment show that it was constructed
in haste and accordingly he regards the battle-field as determined.
;

This conclusion, indeed, is now generally accepted. Bircher, how-

ever, agrees with me and Frohlich in considering Stoffel's theory
irreconcilable with the words (ah) latere aperto and, after a thorough
;

examination of the ground, he informed Frohlich that the hill of

Armecy was steeper on the north-western side than it appears in
Stoffel's map. On StofEel's theory, therefore, the extremity of the
Roman right wing would have been posted on a steep declivity,
w^hereas the Romans preferred a gentle slope. For these reasons
Bircher concluded that the veteran legions were posted on the lower
slopes, facing west-south-west, and that the hill to which the Helvetii
retreated was not, as Stoffel supposed, the hill near Montmort, but
on the further side of the valley through which runs the road from
Toulon-sur-Arroux to Luzy. This was the road by which the Boi
and Tulingi would have marched to reinforce their comrades and ;

thus they would have struck the Roman army on its right flank 2
{ah latere aperto).
But Professor Walter Dennison has come to the rescue of Stoffel.
^

Bircher's theory, he argues, robs the word circumvenire (which

Caesar uses in his description of the manoeuvre of the Boi and
Tulingi) of all point and it is demolished by the fact that the
;

Helvetii finally retreated northward, whereas, according to Bircher,

when they withdrew, after the first stage of the battle, to the hill
about a mile off, they moved southward it is contrary,' says the
:
'

professor, '
to the recognized laws of the human mind and of self-
preservation to suppose that the enemy should have turned about
arbitrarily and fled almost in the face of the Roman soldiers.'
. . .

But if Professor Dennison will look at his map again, he will, I think,
acknowledge that he has somewhat misrepresented Bircher's theory.
The Helvetii, according to Bircher,* withdrew after their first repulse,
not southward, but west-south-west and neither he nor Captain
;

Veith, who, after exploring the country, independently reached the

same conclusion,^ saw anything improbable in a final northward
retreat of the beaten force from this position.^ As to the word
circumvenire, it does not, as Professor Dennison supposes, imply that
the Boi and Tulingi made a detour it simply means that they not
:

only made a lateral attack on the Roman right, but also attempted
to lap round it so as to strike it from the rear.'

1 Bibrade, 1904, pp. 22-8.

^ See C. Jullian, Hist, de la Gaule, iii, 215, n. 4.
=»
Glass. Philology, 1909, pp. 200-1. * Bibrade, Taf. III.

^ Oesch. d. Feldzfige C. J. Caesars, p. 508, note.

'^
M. Jullian {Hist, de la Gaule, iii, 218, n. 1) supposes that the Helvetii may
have retreated in the first instance south-eastward to Toulon-sur-Arroux, and
thence struck north-eastward towards Dijon, See pp. 633-4, infra.
' See H. Meusel, Lex. Caes., i, 536.
S S 2
628 CAESAR'S CAMPAIGN
V. Describing the dispositions which he made for the battle,
Caesar says that he had the sarcinae stacked on a hill, fortified the
spot on which they lay, and detailed two legions and his auxiliary
corps for their protection (ij)se interim in colle medio triplicem aciem
instruxit . in summo iugo duas legiones
. . et omnia auxilia conlocari
. . .

. .sarcinas in unum locum conferri et eum ah iis qui in superiore acie

.

constiterant muniri iussit)}- As sarcinae generally means, not heavy

baggage (impedimenta), but only the bundles which the soldiers
carried, von Goler ^ and Stoffel ^ infer that the heavy baggage had
been sent on, under a slender escort, towards Bibracte. Napoleon,*
on the contrary, maintains that sarcinas here includes impedimenta ;

but the phrase sarcinaria iumenta,^ to which he refers, does not prove
that sarcinae can be used in such an extended sense and when he
;

goes on to argue that if Caesar had sent on the baggage-train in

advance, he would have sent two legions to escort it, he apparently
forgets that the road to Bibracte led through the country of the
Aedui, who were his allies. It is true that many of them sympathized
with the Helvetii but they were not likely to attempt a serious
;

attack. Two legions, however, and the auxiliaries would seem to

have been a large force to detail merely for the protection of the
men's bundles and if Caesar had sent on the heavy baggage, we
;

should have expected him to say so. Moreover, as his army appa-
rently remained near the battle-field for three days after the victory,
it seems reasonable to suppose that they must have wanted some of
their baggage. On the other hand, it has occurred to me that the
baggage-cattle, or some of them, may have been sent on to Bibracte
in order to fetch a supply of corn, as the legions had only two days'
rations left. May we suppose that the necessary baggage was left
upon the hill, and that the cattle and their drivers were sent on to
Bibracte ? 6
VI. StofEel, whom Bircher and I have followed, places the Helvetian
laager on the plateau of La Bretache, west of Montmort. M. Jullian,'
on the contrary, argues that the fact that Caesar attacked and took
'

possession of the Gallic camp proves that it was not on high ground '.
Why ? Did not Caesar attack and defeat the Helvetii when they
were on a hill ? And is it not likely that they would have formed
their laager on a strong position ? I see no reason to differ from
Stoffel.8

1B. G., i, 24, §§ 2-3. ^ Qall. Krieg, 1880, p. 26.

'Guerre civile, ii, 449. * Hist, de Jules Cesar, ii, 68, n. 2.

' £. (7.,i, 81,§6.

® I doubt whether the entrenchment on the hill of Arniecy would have been

large enough to hold the entire baggage-train. Cf. H. Bircher, Bibracte, p. 24.
' Hist, de la Gaule, iii, 214, n. 3 217, n. 1.
;

* Miss
(?) Elizabeth Reed {Classical Journal, iii, 1908, pp. 192-3) quaintly
raises the question whether the baggage and the camp {itnpedimentis cas-
'
'

trisque [B. G., i, 26, § 4]) which Caesar captured were 'in the same place' or
whether castris means the earlier camp which the Helvetians had left on the
'

morning of the battle ? Quite correctly she decides for the former alternative
'
;

but why did she suggest a doubt ? The camp which the Helvetians had left
'

on the morning of the battle' ceased to exist as a camp when the Helvetians
marched on it was simply a wagon laager.
:
 AGAINST THE HELVETII 629

VII. Describing the stage of the battle, Caesar says

first The :
'

Gauls were greatly hampered in action by the fact that in many

cases several shields were transfixed and pinned together by the
impact of one javelin and as the iron bent, they could not pull
;

the javelins out or fight properly with their left arms encumbered ;

so that many, after repeated jerks, preferred to drop their shields,

and fight bare (Gallis magno ad fugnam erat impedimento quod
'

'plurihus eorum scutis uno ictu pilorum transfixis et conligatis, cum

ferrum se inflexisset, neque evellere neque sinistra impedita satis com-
mode pugnare poterant, multi ut diu iactato bracchio fraeoptarent
scutum manu emitters et nudo corpore pugnare).^ V. Wenning finds
serious difficulties in this passage. Must not plurihus, he asks, mean
at least three ? Could one javelin have pierced three shields ? And
"^
could plurihus have been used in a distributive sense ?
1 think we had better accept the passage as it stands for Caesar ;

could have had no motive for misrepresentation or invention. No

doubt plurihus strictly means at least three but we are not obliged
;

to suppose that, in every instance of the occurrence which Caesar

describes, more than two shields got fastened together. One javelin
might certainly have pierced two light shields and stuck in a third ;

and plurihus is obviously used in a distributive sense. Wenning has

overlooked a passage in which Livy ^ describes a similar incident.
VIII. Adverting to the words pedem referre, which Caesar uses in
describing the conclusion of the first stage of the battle. Long ^ says
that the Helvetii drew back, still facing the Romans, to a hill
'

a mile distant '. Apparently he supposes that they backed for

a mile, like dehutantes in the presence of royalty.
IX. Describing the flank attack which the Boi and Tulingi made
upon the Romans, Caesar says, Boi et Tulingi, qui agmen ho-
. . .

stium claudehant, et novissimis praesidio erant,^ &c. Stoffel

understands this passage to mean that the Boi and Tulingi habitually
served as the rearguard of the enemy's train of wagons, but in front
of the Helvetian column of fighting men, and therefore on the day
of the battle, after the emigrants had begun their retrograde move-
ment, marched in rear of the Helvetian column. In other words,
he takes qui novissimis praesidio erant as referring to the habitual
order of the whole column, and agmen hostium claudehant as referring
to the position of the Boi and Tulingi on the day of the battle.^
Now, apart from the obscurity with which it charges Caesar, there
are two objections to this interpretation. First, it compels Stoffel to
assume that the Boi and Tulingi were [needlessly] separated, on the
march, from the Helvetii by a space, to traverse which required the
whole time that was occupied by the first portion of the battle and

^ B.G.,u '25, §§ 3-4.

2 Correspondenz-Blalt f. d. Gelehrten- und RealscJiulen WilrUemhergs, 1881, p. 80.
^ xxxviii, 22, § 9. —
In cos qui portas clausorant legionum antesignani pila
. . .

conicccrunt. Hi vcro noii vuliiorabantur, sed, transvorberatis scutis, plcrique

inter so conscrti liacrcbant.
* Decline of Ihe Ronmn Republic, iv, 10.
3 B. G., i, 25, « Guerre civile, ii, •450.
§ ().
630 CAESAR'S CAMPAIGN
the retreat of the llelvetii to the hill on which they rallied. Stoffcl
niaiiitaiiisthat this time was short. I maintain, on the contrary,
that the words dlu and tandem in Caesar's description ol the lirst
portion of the battle suggest that it was considerable. Secondly,
Stoffel's interpretation appears opposed to Caesar's implied statement
that the wagons immediately followed the Helvetian fighting column,
and began to encamp as the action was about to begin.i If the order
of Caesar's narrative does not prove that the Boi and Tulingi arrived
upon the field after the wagons began to be parked, the narrative is
obscure. One might suppose, then, that the words qui novissimis
praesidio erant mean that, after the emigrants retraced their steps,
the Boi and Tulingi served as the rearguard of the wagon-train,
which they marched past in order to come into action. This
interpretation, however, which is generally accepted, presents a
difficulty. According to Stoffel's calculation of the length of the
entire column,^ the nearest company of the Boi and Tulingi nnist
have been nearly 15 miles from the battle-field when the column
began to retrace its steps. The battle began about one o'clock,
between two and three hours, as Stoffel argues, after the Helvetian
vanguard began to harass the Komans would the first stage of :

the fighting have lasted long enough to enable the Boi to hurry up
on to the field ? M. JuUian ^ agrees with StofTel that the Boi and
Tulingi marched in front of the wagons, but holds apparently that they
were not separated from the Helvetii by any considerable interval
and he understands the words novissimis praesidio erant in the sense
that the Helvetii left the Boi and Tulingi on the road, to guard the
wagons. Notwithstanding the obscurity w^hich it imputes to Caesar,
this explanation seems probable but if it is right, the wagons
;

must evidently have been also protected in the rear by another

force, which never came into action.
X. Describing the last stage of the battle, w^hich followed the
attack made by the Boi and Tulingi upon the Romans, Caesar writes :

'
Thus two battles went on at once and the fighting was prolonged;

and fierce. When the enemy could no longer withstand the on-
slaughts of the Romans, one division drew back, in continuation of
their original movement, up the hill, w^hile the other withdrew to
their baggage and wagons {ita ancifiti proelio diu atque acriter
'

pugnatum est. Diutius cum sustinere nostrorum im/petus non possent,

alteri se, ut coeperant, in montem receperunt, alteri ad impedimenta et
carros suos se contulerunt).^ The common view is that the first alteri
denotes the Helvetii, the second the Boi and Tulingi. Schneider,^
however, holds that some of the Helvetii joined the Boi and Tulingi
at the wagon laager.^ He argues that Caesar would not have praised

1 i?. (^.,i,24,§4.
30 kilometres {Guerre de Cesar et d^ Ariovisle, p. 3G).
2 G. Veitli (see p. 240,
supra) apparently believes that Stoffel overestimated the length of the column.
* Hist, de la (J aide, iii, 214.
' 13. a., i, 2(5, § 1. 6 Caesar, i, 53.
'^
A. Hug {liheiu. Mus., N. ¥., xv, 18G0, pp. 480-1) conjectures, for reasons
which are not worth discussing, that Diutius . . . contulerunt is an interpolation.
 AGAINST THE HELVETII G31

if they had kept away from the

the valour of the Helvetii, as he did,
light at the hiager,and remarks that carros suos, if written only with
reference to the Boi and Tulingi, would have been inaccurate, as the
wagons belonged to the whole host. But the Helvetii could have
fought just as bravely on the hill as at the laager and, as Caesar ;

distinctly says that tw^o separate battles some of them at went on,
all events must have remained on the hill. He had a perfect right
to say that the Boi and Tulingi retreated ad carros suos, even though
only some of the wagons belonged to them and if Biutius ; . . .

contulerunt is taken in its plain sense, the first alteri can only refer
to the Helvetii and the second to the Boi and Tulingi. Bircher holds
that the Helvetii made good their retreat under cover of the resis-
tance which their allies offered in the laager.
XI. Stoffel estimates that the Helvetii had only about 32,000
men actually engaged in the battle before the Boi and Tulingi came
into action but his calculations,^ which are very elaborate, are
;

based upon insufficient data.

XII. What route did the Helvetii take after their defeat ? What
route did Caesar take, after he had overtaken them, in his march to
Vesontio (Besan9on) ?
Caesar says that they fled all night without stopping, and reached
the territory of the Lingones on the fourth day (ex eo froelio circiter
'
'

milia Jiominum CXXX superfuerunt eaque tot a nocte continenter

ierunt : [nullam partem noctis itinere intermisso] in fines
Lingonum die quarto pervenerunt).^ Before starting in pursuit
he was obliged to remain three days on the battle-field. The fugi-
tives sent envoys to meet him, and halted, in obedience to his orders,
at the spot which they had reached when the envoys returned. After
he had overtaken them, a portion of their host made a vain attempt
to escape in the direction of the Rhine. After this, Caesar received
a deputation from the states of Celtican Caul, and then resolved to
march against Ariovistus, who was in the country of the Sequani
(Alsace). He advanced by forced marches for three days, and then,
hearing that Ariovistus was hastening to occupy Besancon, pushed
on at the top of his speed (niagnis nocturnis diurnisque itinerihus),
in order to anticipate him.^ The phrase which I have just quoted
proves that, after he was informed of the movement of Ariovistus,
Caesar must have marched, at the very least, two days and tw^o
nights before he reached Besan9on. It is therefore clear that his
journey was sufficiently long to occupy three days of forced and at
least two days of extraordinarily rapid marching. We shall be well
within the mark if we assume that its extent was not less than
110 Roman miles.
1. The route which Napoleon traces leads by way of Moulins-
Engilbert, Lormes, and Avallon to Tonnerre, where he believes
C^aesar to have overtaken the Helvetii thence by way of Tanlay,;

Cland, Laignes, i]trochey, and Dancevoir to Arc-en- Barrois, where

'
Guerre civile, ii, -450-1 ; Guerre de Cesar et d' Ariovide, p. 77.
' ' lb., 27, 30-8.
B. 0., i, 20, § 5.
632 CAESAR'S CAMPAIGN
he believes Caesar to have received the news of Ariovistus's advance ;

and thence by way of Langres, Grenant, Seveux, and Oiselay to

Besan9on.i
Now, as regards the flight of the Helvetii, everything depends
upon the meaning of the words die quarto. Kraner, arguing that the
Helvetii would not have required four days and nights to reach the
country of the Lingones, proposed to regard these w^ords as a gloss or
to substitute for them postero die^ (' on the next day'), which shows
that he did not know how far the country of the Lingones was from
the battle-field and A. Hug ^ conjectured that Caesar had written
;

die orto (' at daybreak '). But this part of the text, at all events,
requires no alteration and it is certain that when we translate die
;

quarto by on the fourth day ', we must, remembering the Roman

'

method of reckoning, regard the day of the battle as the first day.
Thus, if the battle was fought on a Sunday, the Helvetii reached the
country of the Lingones on Wednesday. According to Napoleon,
they would have reached it on Thursday. But Napoleon made the
same mistake here which he made in his note on the meaning of
altero dieA Another question is whether the w^ords nullam "partem
noctis itinere intermisso (supposing that they are genuine) are simply
a reaffirmation of ea tota nocte continenter ierunt or whether they
—
convey fresh information. Li other words, did the Helvetii march
all Sunday night, and then march on on Monday, Tuesday, and
Wednesday or did they, throughout their journey, march in the
;

night only and rest in the day-time ? Napoleon's view, that they
marched without interruption day and night ', may be set aside as
'

absurd.^ Schneider ^ holds that Caesar WTote the words nuUam

partem noctis itinere intermisso in order to emphasize the fact that it
was only because the Helvetii marched throughout the whole of the
first night that they reached the territory of the Lingones on the
fourth day :— quod si ilia nocte vel paululum quievissent, die
'

quarto fines Lingonum non attigissent.' B. Miiller,' who calls this

a very naive explanation, holds that the Helvetii marched by night
only, and rested in the day-time. After marching the whole of the
first night, they naturally, he argues, had to rest the next day. In
the evening they made a fresh start and so on till they reached the
;

country of the Lingones. I am sure that Miiller is wTong for ;

1 Hist, de Jules Cesar, ii, 70, 80.

2 Caesar, ed. Dittenberger-Kraner, 1890, p. 87.
3 Rhein. Mus., N. F., xv, 1860, pp. 477-8. « See
pp. 738-40.
5 It is hardly necessary to point out that, even if the Helvetii marched on
four consecutive days and nights, they must have rested during a large part
of each day and of each night, except the first ; or that when Caesar says of
his own march in 52 B.C., neque diiirno neque nocfurno itinere intermisso [B. G.,
vii. 9, § 4), lie did the same. B. Miiller (Znr Kritik und Erhlarinu] von Caemrs
gall. Kriege, 1877, pp. '23-4) points out that, to say nothing of the superhuman
endurance with which Napoleon credits the Helvetii, if Caesar had meant to
describe a march by day and night, he would have expressed himself as he has
done in B. 0., vii, 9, § 4, and 50, § 3 {adtnodum magnis diurn id nocturnisqut
iiinerib as confect is . ad Ligtrim pcrvenit).
. .

" Caesar, i, Ot).

^ Zur Krilik und ErkUirung von Catsara gall. Kriege, p. 23.

 —
AGAINST THE HELVETII 633

a succession of night marches seems very unlikely and I believe

;

that if Caesar had meant to describe such a thing, he would not

have laid stress on the fact, if it had been a fact, that the Helvetii
marched all night (nullam partem noctis itinere intermisso) he would, :

I believe, have used some such expression as nocturnis itineribus

confectis or simply nocturnis itineribus. Moreover, as Meusel says,
he would have written, not (nullam partem) noctis, but noctium.^
I have no doubt that Morus was right in deleting nullam inter- . . .

misso as a gloss for the words are absolutely superfluous, and

;

although Caesar has a way of repeating statements which he wishes

to emphasize, he only does so, as Meusel says, in order to make his
meaning more clear. But this by the way. The important point is
that, as Napoleon exaggerates the duration of the flight and also the
duration of each day's march, the Helvetii could hardly, as he
believes, have arrived on the fourth day at Tonnerre.^
' '

2. Von Goler ^ thinks that Caesar overtook the Helvetii in the

neighbourhood of Langres, Avhich is as far from the battle-field as
Tonnerre and if there was no course open to them but to return to
;

Switzerland, what would have been the use of marching far out of
their way ?

3. According to Kiepert,* Caesarfirst marched to Dijon, then

struck a south-easterly direction, crossed the Saone at Sego-
off in
bodium (Seveux), and marched thence to Besangon. But on this
theory the distance which he had to march from the country of the
Lingones to Besangon was so short that he would not have required
at least five days' forced marching to perform it. Still, Dijon was
the point in the country of the Lingones for which the Helvetii would
naturally have made if they already intended to return to Switzer-
land and, if M. Jullian ^ is right in believing that the Pan-Gallic
;

council which was held with Caesar's permission after their defeat ^
met, not at the place where they were overtaken, but at Bibracte,
he is perhaps also right in believing that they retreated to Dijon,

^ Jahresb. d. philol. Vereins zu Berlin, xxxvi, 1910, p. 39.

^ Stoflfel {Guerre de Cesar et d' Arioviste, p. 78) takes the sensible view that the
Helvetii arrived in the country of the Lingones on the fourth day, counting that
on which the battle took place as the first. Yet his map agrees with that of
Napoleon ; and accordingly he requires from the unhappy fugitives and their
still more unhappy cattle exertions even more herculean than had satisfied his
imperial master. The retreat, he says, lasted about 60 hours the length
:

of the retreat was 160 kilometres or about 100 miles ; and the Helvetian
wagons were tugged, for the most part, by oxen. Was any bullock ever yet
required to drag a cart 100 miles in two days and a half ? It is safe to say
that if the Helvetian oxen had been goaded in this way, not one of them would
have reached Tonnerre alive.
=»
Gall. Krieg, 1880, pp. 31, 332.
" Transalpinae
Galliae Cif^alpinue et iahnla in usiim ficholarnin descripta.
. .

5 Hist, de la Gavle, 222, n. 2 223, n. 2. M. Jullian's reason that all

iii, ; —
the other Pan-Gallic councils which Caesar mentions took place in to\\ns
seems hardly sufficient. The council in question was exceptional, not one of —
the annual councils summoned by Caesar, but held with his permission at
the request of the delegates themselves and we can only be sure that it was
;

hold at a place which suited his convenience.

« B. G., i, 30, §§ 4-5.
634 CAMPAIGN AGAINST THE HELVETII
for the distance from Bibractc to Besanoon corresponds closely
enough with Caesar's account of his march. His narrative suggests,
indeed, that the council was held in the country of the Lingones at
the place where he received the submission of the Helvetii and where
the Gallic deputies congratulated him but political reasons may
;

have influenced him to go to Bibracte, the capital of his subservient

allies, and to emphasize by his presence there at the head of his
victorious army the fact that he was the protector and the master of
Gaul.
I am is one of our insoluble problems.
afraid that this H we
cannot where Caesar overtook the Helvetii, or where the council
tell
was held, neither can we tell by what route Caesar marched to
Besan9on. This conclusion will not please those who require positive
results. But
I did not promise to solve all the problems connected
with the Gallic war and to know that a problem is insoluble is the
;

next best thing to solving it.

WHAT GALLIC STATES WERE REPRESENTED IN

THE DEPUTATION WHICH CONGRATULATED
CAESAR AFTER THE OVERTHROW OF THE
HELVETII ?
Bello Helvetiorum confeeto, writes Caesar, totius fere Galliae Jegati,
principes civitatum, ad Caesarem gratulatum convenerunt.^ On this
—
Long 2 remarks that Gallia here means Celtica,' that is to say, the
'

whole of Gaul, except the Province, Aquitania, and the country of

the Belgae.^ Desjardins,* who expresses the same opinion, excludes
the Belgae, on the ground that they made war upon Caesar in the
following year (57 B.C.). Mommsen goes further, and calls the con-
cilium totius Galliae,^ which met, with Caesar's sanction, just after
the conclusion of the Helvetian campaign, a diet of the Celtic tribes
'

of Central Gaul.' ^ My belief is that Mommsen is right, because we

may gather from B. G., ii, 34 that the maritime states between the
Seine and the Loire did not acknowledge Caesar's authority until
towards the end of 57 B. c. Besides, the envoys of the more distant
tribes would not have had time to reach Caesar.''

'
B. G.,
i, 30, § 1. 2 Caesar,
p. 70. ^ See B. G., i, 1.
*
Geogr. de la Gaule rom., ii, 61G. See, however, ih., p. 542.
5 B. G., i, 30, § 4.

6 Hist,
of Borne, v, 1894, p. 45 [Bom. Gesch., iii, 1889, p. 255).
' It might perhaps be inferred from B. G., iii, 11 that the Pictones and
Santoni, who assisted Caesar in 56 B.C., but are not mentioned in ii, 34, had
acknowledged the supremacy of Rome in 58 B.C.; but they may only have
done so under the influence of the victorious campaign of 57.
The words omni pacaia Gallia {B. G., ii, 1, §2), which 1 quoted in the lirst
edition (p. 020), arc irrelevant. Sec p. 09, u. 2,
 J

635

THE PKOBABLE LENGTH OE CAESAR'S MARCH

FROM VESONTIO AGAINST ARIOVISTUS
Vegetius^ says that Roman soldiers used to march 20 miles in full
armour, by way of practice, three times a month and from this
;

statement Rustow 2 and others infer that the length of an ordinary

day's march was 20 Roman miles, or between 29 and 30 kilometres.
This estimate is certainly exaggerated for when Caesar marched
;

from Besan§on to the Belgic frontier in 57 B. c. in about a fort-

'

night ',3 his average daily march, allowing two days for rest, was not
more than 20 kilometres and in the last five days of his march
;

from Decetia to Gergovia, of which, it is true, the first stage and the
fifth may have been comparatively short, he accomplished, at the
most, no more than 100 kilometres.* On the other hand, he once
made a forced march of 50 Roman miles, or 74 kilometres, with four
legions in less than 30 hours ^ and he marched from Agedincum (Sens)
;

to Cenabum (Orleans), a distance of at least 108 kilometres, in two

periods of two days each, separated by an interval of three days,
during which he captured Vellaunodunum.^ The Commentaries on
the Civil War supply another instance. On the 21st of February,
49 B.C., Caesar left Cbrfinium, and marched thence to Brundisium,
a distance of 465 kilometres, arriving on the 9th of March. If he
marched on every one of these 17 days, his daily average was 27
kilometres if he allowed his troops two days' rest, it was 31.'^
:

From these data it may be concluded with certainty that, although

Caesar's usual daily march was probably much less than 30 kilo-
metres, he could have marched, in an open country, at least 27
kilometres a day, for seven days, if he had wished to do so. The
Due d'Aumale, an experienced soldier, holds that as many as ten
legions could have easily marched 28 kilometres a day.^

' De re mil., i, 9.
2 Heenvesen und Kriegfiihrung C, J. Cdsars, 1855, pp. 92-3.
3 B.G., a, 2, § 6. ^ See
pp. 754-5.
6 B. G., vii, 40-1. « See
pp. 410, 494-5.
^
See Stoffel's Guerre civile, i, 196-7, and F. Frohlich, Das Kriegswesen Cdsars,
p. 207. According to 0. E. Schmidt's itinerary {Der Briefwechsel des M. T.
Cicero, 1893, pp. 149-50, 379), the averages were respectively 28 ^'-W and 32J.
^ Rev. des Deux Mondes, 2® per., xv, 1858, p. 95. According to Lord Wolseley
{The Soldier's Pocket-Book, 5th ed., p. 322) the length of ordinary marches,
'

for a force not stronger than one division, moving by one road, should be from
12 to 16 miles a day, for 5 days out of 6, or at most 6 days out of 7'. Guischard,
however, remarks {Mem. crit. et hist., 1774, pp. 40-3) that ancient armies
could march faster than modern because ils etoient dispenses d'un grand
'

nombre de besoins, que nous nous sommes rendus necessaires, et delivres par
consequent de I'obligation de trainer apres eux tout cet attirail de guerre,
et ce grand train d'equipagcs, qui no peuventqu'embarrasser Ics mouvemens
de nos armees'.
 636

CAESAR'S CAMPAIGN AGAINST ARIOVISTUSi

I. The data for drawing a map of Caesar's campaign against
Ariovistus are very scanty. The gist of his narrative is as follows.
He marched from the place where he received the Gallic deputies
after his victory over the Helvetii, to encounter Ariovistus. After
he had marched for three days he heard that Ariovistus was hasten-
ing with his whole force to seize Vesontio (Besan9on), and (if the
words triduique viam a suis finibus frocessisse in B. G., i, 38, § 1 are
—
not an interpolation^) that he had apparently at the time when
—
Caesar's informant started on his errand advanced three days'
journey beyond his own frontier. Thereupon Caesar pushed rapidly
on, and making forced marches by day and night, seized Besan9on
himself. He remained a few days in the neighbourhood. Marching
from Besan9on, he reached his final camping ground on the seventh
day and he marched on every one of those seven days. After leav-
;

ing Besan9on, he took a circuitous route, in order to gain the advan-

tage of moving in an open country. I shall discuss in the proper
place the disputed passage in which he describes this movement.
When he reached his camping ground, he was 24 Roman or about
22 English miles from Ariovistus. A few days later the two leaders
had an interview at or on an earthen mound or a knoll of consider-
able size, nearly equidistant between the two camps and situated in
a great plain. Two days after the interview Ariovistus broke up his
encamj^ment, marched to a point 6 Roman miles from Caesar's
camp, and encamped at the foot of a mountain. Next day he
marched past Caesar's camp and encamped two Roman miles beyond
it, with the object of intercepting the supplies which were being

brought up to the Romans from the territories of the Aedui and the
Sequani. Five days later Caesar made a retrograde march past
Ariovistus's camp and constructed a smaller camp about 1,000 yards
{circiter passus DC) from it, in order to re-establish his communica-
tions. Two days later he marched against Ariovistus. Thereupon
the Germans moved out to fight, leaving their wagons ranged in rear
of their line, in order to deprive themselves of all hope of flight.
The Germans were beaten and fled, and did not cease their flight
'

until they reached the Rhine ', which, according to the Commentaries,
was 5 Roman miles from the battle -field.^ For reasons which I shall
presently examine the word quinque (5) has been altered by some
editors to quinquaginta (50).
II. I have already argued * that it is impossible to trace Caesar's
route with certainty from the place where he received the Gallic
deputies toBesancon; and it therefore appears to me safest to make

1 See Carte de France (1 200,000), Sheets 28, 3C, 55, and Carte de V Etat-
:

Major (I :80,000), Sheet 101.

^ Meusel {Jahresb. d. yhilol. Vereins lu Berlin, xxxvi, lUlO, p. 44) argues
that they were. See pp. 037-8, infra.
^ B. G., i, 37-53. * See
pp. 031-4.
 —
CAMPAIGN AGAINST ARIOVISTUS 637

Besan9on the starting-point in any attempt to trace his line of march

against Ariovistus. The first question is, what was the territory
from which Ariovistus marched to seize Besancon.
III. According to the speech which Caesar puts into the mouth of
Diviciacus,! Ariovistus annexed one-third of the Sequanian territory,
'
which is the best land in the whole of GauL' L. W. Ravenez,^ like
Napoleon, holds that the territory which Ariovistus annexed was
Upper Alsace, which, he remarks, is the most fertile part of Sequania.^
It may be added that he would naturally have settled in the part
from which he could most easily communicate with his countrymen
beyond the Rhine, and further that, as I shall presently show, it is
impossible, on any other theory, to explain Caesar's narrative of the
campaign."* All commentators, indeed, now agree that Ariovistus
had settled in the plain of the Rhine but it is surely probable
;

that he had annexed not only Upper, but also Lower Alsace, the —
northern part of the country.
Were the words triduique viam a suis finibus processisse written by
Caesar ? Meusel contends that they were not. Read the whole
sentence Cum tridui viam processisset, nuntiatum est ei Ariovistum
:

cum suis omnibus copiis ad occupandum Vesontionem, quod est oppi-

dum maximum Sequanorum, contendere triduique viam a suis finibus
processisse. Id ne accideret, magnopere sibi praecavendum Caesar
existimabat. (' After a march of three days he received news that

Ariovistus was hurrying with all his forces to seize Vesontio, the
largest town of the Sequani, and had advanced three days' journey
beyond his own frontier. Caesar felt it necessary to make a great
effort to forestall him.') Meusel argues, first, that triduique viam . . .

processisse, following (cum) tridui viam processisset, is suspicious

(which I admit) secondly, that the words break the connexion
;

between ad occupandum Vesontionem contendere and id ne acci-

. . .

deret, —
a fault of style of which Caesar would never have been guilty ;

and thirdly, that as the fines of Ariovistus were simply the territories
of the Sequani, he could not have made a three days' march beyond
them in order to seize Vesontio, which was in them. The last reason
is unsound, for Meusel apparently forgets that Ariovistus had only

annexed one-third of the country of the Sequani ^ and I do not ;

think that the others are more than enough to justify suspicion.
Caesar was a great writer but he had no time for revision, and
;

his style was occasionally careless. Moreover, I find it difficult to

conceive that an interpolator would have been guilty of an invention
even more puerile than the marginal exclamations of some of
Mr. Mudie's subscribers.
M. Jullian ^ supposes that a suis finibus means from Ariovistus's

1 B. G., i, 31, § 10. 2 r Alsace illustree,

1849, i, 382-3.
^ C. Martin {Questions alsaciennes, &c., 19-20) maintains that
1867, pp.
Ariovistus's territory was bounded by the Vosges, the Jura, and the frontiers of
the Aedui and the Lingones but any one who may take the trouble to examine
;

his argument will see that it recoils against himself.

* See also StofiEel, Guerre de Cesar et
(T Arioviste, pp. 88-90.
^ j5. (?., i, 31, § 10. « Hist, de la Gaule, iii, 227, n. 3 ; 231, n. 4.
 B38 CAESAR'S CAMPAIGN
territory on the German side of the Rhine but Caesar's narrative
;

shows that Ariovistus had long since left this territory i what :

reason is there to suppose that he had returned? Surely suisfinibus

can only mean the territory which he had wrested from the Sequani.
Stof£el,2 indeed, argues that it means the country of the Triboci,
one of the tribes which fought under Ariovistus and that they
;

already possessed that part of the country of the Mediomatrici just

—
north of the river Ecken-Bach which they occupied later.^ But is
—
it not evident that a suis finibus processisse must mean that Ario-
vistus had left all his territories behind ? If a king of Scotland,
about to invade England, had made three marches beyond the
southern boundary of Sutherland, would he have marched tridui
viam a suis finibus ?
StofTel assumes that Caesar was at Arc-en-Barrois, 125 kilometres
from Vesontio, when he heard that Ariovistus was marching to seize
it. But this is a mere conjecture we only know that he was at
:

—
a distance of at least two very long marches say "between 50 and
—
60 miles from Vesontio ^ and we may conclude that Ariovistus
;

was nearer. Assuming that my explanation of suis finibus is right,

and that he possessed Upper Alsace, the words triduique viam . . .

frocessisse can be explained, if we may suppose that Ariovistus had

only just completed the three days' journey when Caesar began his
'
great effort to forestall him '. But, as I have already said, the
passage would seem to imply that he had already completed
the journey when Caesar's informant started on his errand. If so, he
would nearly have reached Besan9on when the messenger arrived ;

and the suspected words would be inexplicable. Therefore, although

I can frame no theory to account for the supposed interpolation,
Meusel may be right. Anyhow, as we shall presently see, Ariovistus,
after he had failed to seize Besancon, must have marched back into
the plain.
IV. Speaking of the circuitous route which he took after leaving
Vesontio, Caesar says itinere exquisito per Diviciacum ut milium
. . .

amplius quinquaginta circuitu locis apertis exercitum duceret,^ &c. It

has been argued that these words mean that the whole length of the
march was rather more than 50 Roman miles. But it is incredible
that Caesar should have marched barely seven English miles a day
through an open country. Von Goler ^ indeed reads XC instead of
L but L or quinquaginta is found in all the MSS. It has also been
;

suggested that Caesar meant that the whole length of the march was
50 miles more than it would have been if he had taken the direct
."^
road. But it is impossible to get this meaning out of the Latin
The words can only mean that the circuitous part of the march was
1B. G., i, 31 ff., especially 31, § 10
; 34, § 4
; 44, § 2.
^Guerre de Cesar et d' Arioviste, p. 90.
" See
» See p. 482. p. 631.
« Gall Krieg, 1880,
5 5. 6'., i, 41,
§ 4. p. 46, n. 3.
' Napoleon {Hist, de Jules Cesar, ii, 84, note) concludes from a passage in
B. C, i, 64, § 8 ac tantum fuit in militihus studii ut, milium VI ad iter addito
—
circuitu, &c. that when Caesar means to speak of a turn of road to be added
'

to tlie total length of the route, he is careful to indicate it'.

 ;

AGAINST ARTOVISTUS 639

50 miles long and this is just what it would have been if Caesar
;

had taken the natural circuitous route which Napoleon and Stoffel
trace on their maps.
1. According to Napoleon,^ Caesar marched by way of Penne-
sieres, Vallerois-le-Bois, Villersexel, and Belfort, to a point about
a mile and a half south-west of Cernay, and there encamped. Ario-
vistus was, he supposes, encamped at the time near Colmar. The
interview between Caesar and Ariovistus took place at a knoll near
Feldkirch. Two days afterwards Ariovistus made his long march
southward, and encamped between Hartmanns wilier and Roeders-
heim. Next day he made his flank march, which, according to
Napoleon, led him by an immense detour, south-eastward and then
westward, past Pulversheim and Pfasstadt, to a hill about a mile
north-east of Schweighausen. Caesar made his smaller camp on
rising ground west of Ariovistus's camp and north of Schweighausen.
But von Kampen^ remarks that Caesar would not have pitched his
larger camp in the plain, as Napoleon makes him do and that the ;

site which Napoleon selects for the smaller camp is dominated by

a hill which Ariovistus might have easily occupied. 2. Accordingly
von Kampen places the larger camp on rising ground just south of
the river Thur and about a mile and a half west-south-west of Cernay
the final camp of Ariovistus on a hill just north-east of Nieder
Aspach on the river Klein-Doller and Caesar's smaller camp on
;

a hill half a mile south-west of Ariovistus's camp. According to him,

Ariovistus made his flank march along the line of an old Roman road,
which crossed the Thur between Cernay and Wittelsheim.
Caesar, as we have seen, says that, at the end of his seven days'
march from Vesontio, he encamped at a distance of 24 Roman miles
from Ariovistus.^ A week later Ariovistus broke up his camp and in
a single day marched to within 6 Roman miles from Caesar's.'* It '

is probable,' observes Napoleon,^ that, during the negotiations,

'

Ariovistus had approached nearer to the Roman camp, in order to

facilitate intercommunication for if he had remained at a distance
;

of 36 kilometres from Caesar, we should be obliged to admit that the

German army, which subsequently advanced towards the Roman
camp, in a single day, to within 9 kilometres, had made a march of
25 kilometres at least, which is not probable when we consider that
it dragged after it wagons and women and children.' ^ But Napoleon
was obliged to assume that during the negotiations, Ariovistus had
'

advanced nearer to the Roman camp ', in order to make the knoll
which he identified with the tumulus terrenus^ at which the con-
ference between Caesar and Ariovistus took place, equidistant
between the Roman and the German camp.
It is generally taken for granted that the tumulus was a natural

^ Hist, de Jules Cesar, ii, 84-5.

^ Quindecim ad Caesaris de b. 0. comm. tabulae, ii.
' B.G.,i,4l,^5. " 76.,
48, §§ 1-2.
' Hist, de Jules Cesar, ii, 85, n. 1.
•'
Stoffel {Ouerre de Cesar et d' Arioviste. pp. 62-3) seef5 nothing improbable
in such <a march. '
^_ q^^ j^ 43^ § 1^
640 CAESAR'S CAMPAIGN
feature, — a knoll. so, why did Caesar describe it as terrenus
But if

(earthen) ? He mentions
a tumulus near the camp which Labienus
occupied in the country of the Treveri another near Atuatuca ;
;

a third near Ilerda a fourth near Dyrrachium

; but to none of :

them does he apply the epithet terrenus.^ It is certainly possible

that the tumulus terrenus was simply an earthen mound or barrow.^
But, wherever the tumulus may havebeen, I cannot see how, on
either Napoleon's theory or von Kampen's, the flank march of
Ariovistus is to be explained.
J. Schlumberger^, indeed, appeals to
Dion Cassius,* who
says that the German cavalry engaged and
defeated Caesar's, and that Ariovistus thus succeeded in passing
Caesar's camp unscathed. But the unsupported testimony of Dion
is not to be trusted Caesar says that Ariovistus, before making the
:

flank march, encamped at the foot of a mountain (suh monte ^) ;

and both von Kampen and Napoleon rob these words of their
significance.
3. Von Goler^ conducts Caesar from Besan9on by way of Vesoul,
Lure, and Belfort to Damerkirch, near which he assumes that he
encamped after his seven days' march. He identifies the tumulus
with the gently rising ground immediately north of Nieder Aspach.
He makes Caesar continue his march on the day after his conference
with Ariovistus, and encamp about a mile south-east of Cernay ;

he places the smaller camp on the site which von Kampen chooses
for the final camp of Ariovistus, and he places the latter in the plain
on the left bank of the Klein-DoUer and about half a mile north of
Ober Aspach. The flank march of Ariovistus he traces along the
lower slopes of the Vosges, north of the Thur from Soultz to Alt-
Thann and thence southward to his camp.
The assumption that Caesar, after his interview with Ariovistus,
marched from Damerkirch to the neighbourhood of Cernay is un-
authorized. Von Goler admits that this additional march is not
mentioned in the Commentaries but he argues that it must never-
;

theless have taken place, or else Ariovistus could not have approached
to within 6 miles of Caesar's camp in a single march. Napoleon, as
we have seen, gets over the difficulty, which is hardly serious, by
assuming that Ariovistus had himself approached nearer to the '

Koman camp and perhaps neither of the two assumptions is

'
;

right. Von Gofer's explanation of the flank march is perfectly reason-

able :but I do not understand why he made Ariovistus encamp in
the plain instead of on a commanding position and nobody will ;

believe that the tumulus terrenus was the site of Caesar's smaller
camp. Napoleon argues further that von Goler is wrong in making
Caesar fight the battle with his back to the Rhine. It would be '

impossible,' he says, to understand in this case how, after their

'

1 B, G., vi, 8, § 3 40, § 1 B. C, i, 43, § 2 ; iii, 51, § 8.

; ;

" It is true that Livy (xxxviii, 20, § 4) speaks of colles terrenos ; but he is
contrasting them with rocky heights.
^ Caesar und Ariovistus, 1877,
p. 168.
* xxxviii,
48, § 2. ' B. G., i, 48, § 1.

« Gall. Krieg, 1880,

pp. 47-51, and Taf. I and III.
 AGAINST ARIOVISTUS 641

defeat, the Germans would have been able to fly towards that river,
Caesar cutting off their retreat or how Ariovistus, reckoning upon
;

the arrival of the Suevi, should have put Caesar between him and
the reinforcements which he expected.' ^ This objection does not
appear to me conclusive. If the Germans had fought, as Napoleon
holds, with their backs to the Rhine, it is evident that, their rear
and perhaps their flanks being closed by their line of wagons, they
could only have commenced their flight in a northerly or southerly
direction. If they had fought facing the Rhine, they would have
done just the same. As to the expected arrival of the Suebi, it must
be remembered that the great object of Ariovistus was to cut Caesar's
line of communication. In order to do this, it may have been neces-
sary for him, as Stoffel thinks, to hold to the line of the Vosges.
Moreover, if a man compares von Goler's map with Napoleon's, he
will, I think, find it hard to believe that Ariovistus would have found
it more difficult to effect a junction with the Suebi in one case than
in the other.
Stoffel,2 following Plutarch,^ places the camp of Ariovistus on the
slope of a hill. Plutarch is certainly no authority on a point of
this kind. He may have been thinking of Caesar's statement that
Ariovistus sub monte consedit, and have forgotten that Ariovistus
marched 8 miles further on the day after he encamped there. Still
it is very likely that Ariovistus did encamp on a hill and if that
;

hill was a spur of the Vosges, he probably fought the battle with
his face towards the Rhine.
C. Martin, who denies that the territory of Ariovistus was in
Alsace, gives various other reasons to show that every explanation
of the campaign which is based upon the hypothesis that the decisive
battle took place in the plain of Alsace, is wrong. First, he says,
Caesar tells us that, while he was encamped in the country of the
Lingones, and before he started on his march against Ariovistus,
the Harudes devastated the territory of the Aedui. If, he concludes,
Ariovistus had encamped in the plain of Alsace, the Harudes could
not have rejoined him before the battle without being destroyed by
Caesar on the march.*
The conclusion is inconclusive. The Harudes may have got the
start of Caesar, while he was waiting at Besangon or, if they were
;

afraid to go through the pass of Belfort, they may have gone through
one of the passes in the Vosges.
Secondly, says Martin, on Napoleon's theory, it is impossible to
discover the earthen mound or knoll {tumulus terrenus satis grandis)
at which Caesar says that the interview between himself and Ario-
vistus took place. Napoleon thinks that the tumulus was either near
Feldkirch or between Wittenheim and Ensisheim. But, objects
Martin, the alleged tumulus near Feldkirch n'existe meme pas
'
:

nous avons ete sur les lieux '. And the one between Wittenheim and

Hist, de Jules Cesar, ii, 91, n. 2. (J. Veith {Gesch. d. Feldzuge C. J. Caesars,
^

p. 508) endorses Napoleon's objection, which would also apply to Stoffel's

theory (pp. 648-51, infra). ^ Guerre de Cesar et d' Arioviste,
pp. 10-1.
^ Caesar * Questions alsaciennesy
i 19. p. 25.
T t
642 CAESAR'S CAMPAIGN
Ensisheim does not deserve the name of knoll for its summit is
;

only 2 metres above the level of the plain at its base ^ !

Martin's statement is, I believe, correct. I was at Miihlhausen in

September, 1895, but was unable, from want of time, to make a per-
sonal exploration of the country. I was so fortunate, however, as
to meet an English gentleman, who had lived at Miihlhausen for six
years and was in the habit of cycling in the plain of Alsace, which he
knew thoroughly. He assured me that there was no such thing as
a knoll at either of the places which Napoleon mentions. Von
Kampen indeed, who professes to have constructed his map (Tab. 2)
on the basis of that of the German military staff, marks two knolls
near Feldkirch, one just outside Wittelsheim, and a fourth near
Ensisheim. The last, which, I suppose, is the one to which Martin
alludes, is marked on the German Government map (1 25,000), :

Sheet 3678 but its height is not indicated

: and I can discover no
;

traces of the others on Sheet 3677. Still, Martin has, at the most,
only succeeded in discrediting Napoleon's map of the campaign :

the theory which finds the theatre of the war in the plain ol Alsace
remains unshaken. And if it is true, as I have suggested, that the
tumulus may have been not a natural feature at all, but a mound or
barrow, he has not shaken even the theory that the battle was fought
near Cernay.
Thirdly, says Martin, on Napoleon's theory, the flank march by
which Ariovistus endeavoured to cut off Caesar's communication
with the convoys which he expected from the Sequani and the
Aedui would have been useless for Caesar could still have com-
;

municated with the Sequani by the valley of the Lauch.^

This argument breaks down unless Martin can prove that the
Sequanian convoy would have taken the route through the valley
of the Lauch whereas it is much more likely that it was coming
;

by the pass of Belfort.

4. Schlumberger 3 takes the words ut milium amplius quinqua-

^ Questions alsaciennes, pp. 32-3.

^ From Caesar' s statement that he made his second camp ne diutius
lb., p. 36.
commeaiii pwhiberetiir (' in order to reopen communication with his convoys
without delay') Thomann {Der franzdsische Atlas zii Cdsars gall. Kriege,
pp. 12-3) argues that his communications were, for a time, completely broken.
With the Sequani and the Aedui certainly ;but I do not see that we can
draw any conclusion about the other two peoples.
Mommsen {Hist, of Borne, v, 1894, p. 47, n. 1 [Rnni. Gescli., iii, 1889, p. 250,
note]) says, * The corn exported from the Sequani, Leuci, and Lingones was not
to come to the Roman army in the course of their march against Ariovistus,
but to be delivered at Besanfon before their departure and taken by the troops
along with them ; as is clearly apparent from the fact that Caesar, while
pointing his troops to those supplies, comforts them at the same time with
the hope of corn to be brought in on the route.' What Caesar says on the
matter is merely this frumentuyn Sequanos, Leucos, Lingones subminisfrare,
:

iamque esse in agris frumenta matura [B. G., i, 40, § 11). Caesar quitted
Besan^on the night after he gave this assurance to the troops {ib., § 14),
so that he did not give much time for Mommsen' s imaginary ' delivery of the
'

corn. Moreover, Mommsen notwithstanding, Caesar distinctly says {ih., 48, § 2)

that the Sequani were to supply him with corn during the campaign.
' Caesar mid Ariovistus, id. 148.
 AGAINST ARIOVISTUS 643

(jinta circuitu locis apertis exercitiim duceret to mean that the distance
which Caesar actually covered in his seven days' march was 50
Koman miles more than he would have had to do if he had taken
the direct route. This interpretation compels him to take Caesar
quite unnecessarily out of his way. He makes him, after leaving
Besangon, march by way of Cussey on the Oignon to the Saone, and
by Seveux to Port-au-Saone, where he found the road which led by
way of Lure, Chalonvilliers, and Belfort into Alsace. Schlumberger
assumes that Caesar marched at the rate of 20 Roman miles a day,
and that his march lasted not seven whole days, but six days and
a fraction, say 6 J, 6 J, or 6 J. On this hypothesis, the whole distance
marched would have been 122 J, 125, or 130 Roman miles and the ;

direct distance, being about 50 Roman miles shorter, would have

been 72J, 75, or 80 Roman miles. Schlumberger reckons that, after
travelling 72J Roman miles by the shortest road from Besanyon
into Alsace, one would arrive at La Chapelle-sous-Rougemont, on
the rivulet St. Nicholas, about 50 Roman miles from the Rhine ^ ;

and in this neighbourhood accordingly he looks for the battle-field.

The elaborate calculations which lead Schlumberger to this result
are useless, because {a) he misunderstands the most important
passage in Caesar's narrative ; (6) when he assumes that Caesar
marched 20 Roman miles, or nearly 30 kilometres a day, he almost
certainly exaggerates ;and (c) he cannot tell whether he made a full
march or only a fraction of a march on the last day. Besides, where
are we to look for that great plain ? Schlumberger admits, indeed
' '

insists, that it was the plain of Alsace but from the fact that Caesar
;

only mentioned it in relation to his interview with Ariovistus, he

infers that it was not visible from his camp.^ The inference seems
to me strained. But is the site which Schlumberger selects otherwise
objectionable ? I think it is, because, rightly interpreted, the words
ut milium amplius quinquaginta circuitu locis apertis exercitum duceret
show that Caesar must have marched considerably further north-
ward than La Chapelle-sous-Rougemont.^
5. The absence of the great plain is only one among many reasons
for rejecting the innumerable attempts that have been made to find
the battle-field in Franche-Comte. One of these attempts, however,
which appeared in the Revue archeologique of July, 1898, has attracted
considerable attention and, as M. Salomon Reinach has recom-
;

mended it as an admirable study ', I am bound to examine its

'

claims. M. Colomb begins by insisting that his extraordinary know-

ledge of the whole theatre of the war places him in a position to
speak with authority and his conclusion is that the defeat of Ario-
;

vistus took place not in the plain of Alsace but between Arcey and
Presentevillers. Readers who do not know the country as intimately
as M. Colomb will find that Sheets 101 and 114 of the Carte de VEtat-
Major (qo^oo) ^^^^ enable them to control his arguments.

^ In a straight jinc La Chapelle-sous-Rougemont is barely 40 kilometres

(about 27 Roman miles) from the Rhine.
^ Caesar und Ariovistus,
pp. 152-3. ^ See p. (349.
T t 2
644 CAESAR'S CAMPAIGN
M. Colomb makes Caesar advance from Vesontio (Besan9on) by
way of Oiselay to Pennesieres, that is to say, by a longer and more
westerly route than the one adopted by Napoleon, and, as we shall
see, by StofEel but from Pennesieres to Arcey the route which he
;

adopts coincides with theirs. He argues that Caesar marched by

way of Oiselay in order to approach the river Saone, by which the '

Aedui and the Lingones were forwarding him supplies.' ^ But

a glance at the map will show that, by following the route indicated
by M. Colomb, Caesar would, in the most favourable circumstances,
only have begun to receive supplies from the Aedui and the Lingones
one day earlier than if he had gone by Vovay, Rioz, and Filaine,
the route adopted by Stoffel for the first few days his troops
:

unquestionably carried their food with them and the Aedui and ;

the Lingones were obliged to forward supplies right up to the actual

theatre of war.
According to Napoleon and StofEel, Caesar marched on from Arcey
through the pass of Belfort into the plain of Alsace according to :

M. Colomb,^ his march terminated at Arcey. M. Colomb defends his

view by the following arguments —
(1) Caesar, he insists, could not
:

venture to advance beyond Arcey either north-eastward in the

direction of Hericourt, or eastward in the direction of Montbeliard,
because, if he had taken either of these routes, Ariovistus would have
seized the other, planted himself in the rear of the Romans, and thus
severed their line of communication. (2) The distance from Besan^on
by Oiselay to Arcey is 90 kilometres, and M. Colomb argues that
Caesar would not have marched more than this in seven days. He
says that in 52 B.C. Caesar took four days to march from Sens to Gien
by way of Trigueres, that is to say, that he marched not more than
25 kilometres a day at the very outside and he infers that from
;

Besanyon to Arcey he only marched 14 kilometres a day. He admits

that Caesar marched from Sens to Gien very early in the year, when
the roads were in bad condition, whereas he marched against Ario-
vistus at the most favourable season but he says that the road
;

from Besan9on to Arcey must, at the best of times, have been bad,
and he maintains that Caesar had no motive for hurrying. (3) He
points out that in the Hungarian invasion of a. d. 929 and in Bour-
baki's campaign of 1871 fighting took place along the line Villersexel
— —
Arcey Montbeliard and he holds that these examples prove
;

that this is the natural route for all invasions coming from the east
and for all attacks coming from France and having the pass of Belfort
as their objective.
The argument depends upon the unverifiable assumption that
first
Ariovistus waited for Caesar in the pass of Belfort but I am willing, ;

for the sake of argument, to grant the assumption. Now if Ario-

vistus had attempted, with his whole force, to cut Caesar's line of
communication, he would have played a dangerous game for, by ;

doing so, he would have found himself cut off from his own dominions
in the plain of Alsace. If, in the case which M. Colomb supposes,

^ Rev. arch., d'' ser., xxxiii, 1898, p. 36. ' lb., pp. 34-5, 40-5.
 A

AGAINST ARIOVISTUS 645

Caesar had advanced beyond Arcey, he would have left detachments

to guard Arcey, or the gorge of Presentevillers
on the road leading to
Montbeliard, or both, and would have advanced himself by way of
Hericourt. Now, supposing that Ariovistus had been so rash as to
quit Belfort and advance by the Montbeliard road in order to seize
Arcey, what would have happened ? In the gorge of Presentevillers
he would have found a force ready to dispute his passage. Mean-
while would Caesar have neglected his opportunity ? Turning to the
right, he would have hotly pursued the German column, and Ario-
vistus would have found himself caught inextricably in a trap. If
he had merely sent a detachment to operate against Caesar's com-
munications, he would evidently have had no prospect of success.
Besides, as we shall presently see, M. Colomb, contradicting himself,
holds that Caesar did advance a few kilometres from Arcey in the
direction of Hericourt, and did leave Arcey undefended The second !

argument depends upon a string of blunders. Caesar, as I have

demonstrated elsewhere,^ never went near Trigueres or Gien he :

marched from Sens not to Gien but to Orleans, a distance of at least

108 kilometres. Moreover, the argument that because Caesar
marched 25 (or rather 27) kilometres a day on a bad road in the
winter, therefore he did not march more than 14 kilometres a day
on a bad road in the summer, is one which I find rather difficult to
follow. I maintain, in opposition to M. Colomb, that Caesar marched
against Ariovistus as fast as he conveniently could otherwise, why
:

did he make a point of telling us that he marched for seven con-

secutive days without allowing one day for rest (septimo die cum iter
non intermitteret, &c.) ? ^ The historical precedents which M. Colomb
quotes might perhaps have weight if it could be proved that Ario-
vistus waited for Caesar in the pass of Belfort but I find it difficult
;

to believe that Ariovistus would ever have committed himself to an

offensive movement against Caesar westward of the pass.^
But when we come to scrutinize the kernel of M. Colomb's argu-
ment, we find that his case completely breaks down. Caesar says
that his conference with Ariovistus took place at a tumulus terrenus
in a great plain. I have argued that the great plain was the plain
of Alsace ; and I agree with Stoffel, who is not a bad topographer,
that there is no other great plain in which the conference can possibly
be supposed to have taken place. No says M. Colomb
! the great :

plain was that in which Montbeliard is situated it was between

;

the Savoureuse and the Lisaine, which flow into the Allan, and it
was bounded on the south by the Doubs. Its extent from east to
west was more than 6 kilometres, and from north to south nearly 7
Well, I will not quarrel about measurements, although, if M. Colomb's
description is just, Port Meadow, near Oxford, might fairly be called
a great plain. But is M. Colomb's great plain a plain at all ? Cer-
tainly it looks like one in M. Colomb's sketch-map he contrives to
:

2 5
'
pp. 405-15, 494-8. q^ j^ 41^ § 5,
^ H. Delbriick {Gesch. der Kriegskunst, i, 451) remarks that Ariovistus
deliberately waited for Caesar in the plain of Alsace because he was conscious
of his own superiority in cavalry. * Eev. arch., xxxiii, 1898,
p. 49.
646 CAESAR'S CAMPAIGN
make it do so by the simple process of leaving the area blank and
shading the snrrounding hills. By a similar process I could produce
a map in which the Matterhorn would look like a plain. If the
reader will take my advice, he will check M. Colomb's map by
Sheet 114 of the Carte de V]ttat-Major. He will there find that the
entire area of M. Colomb's plain is covered by hill-shading. The
tumulus terrenus, according to M. Colomb, was the hill called La
Chaux. M. Colomb observes that, viewed from the summit of this
hill, the plain semble etre rigoureusement plate '. I can only
'

reply that within a fraction of the area, not including La Chaux

itself, I find the following different elevations, expressed in terms of
metres above the level of the sea,— 312, 320, 347, 349, 366. Is not
this great little plain somewhat uneven ?
'
'

Let us now examine M. Colomb's explanation of the flank march

by which Ariovistus succeeded in temporarily cutting Caesar's line
of communication. According to M. Colomb, the hill at the foot of
which Ariovistus halted on the night before he made this march
was a hill overlooking Montbeliard Caesar's camp was on the north-:

west, between Semondans and Desandans, and on the road leading

from Arcey to Hericourt and Ariovistus advanced through the
;

gorge of Presentevillers, passed Ste-Marie, and encamped at Arcey.

When the reader looks at the map, he will w^ant to know how Ario-
vistus came to undertake so desperately hazardous a movement, and
why Caesar tamely allowed him to execute it. But M. Colomb ^ is
ready with an answer. He shall speak for himself Cesar ouhlie :
— '

assez volontiers de raconter les evenements qui n'ont pas tourne a son
honneur. Dion Cassius dit en effet qu'il y eut une lutte acharnee
. . .

dans la quelle la nombreuse ca valeric germaine ayant fait eprouver . . .

de grandes pertes aux Komains, les for9a a se renfermer dans leur

camp et a y demeurer spectateurs impuissants de la marche hardie
qui, conduisant Arioviste a Arcey meme, c'est-a-dire a I'orifice
superieur du col de Granvillars et au point de croisement de toutes
les routes de Sequanie, coupait Cesar et I'isolait.' I take leave to say
that Dion Cassius says nothing of the kind. What he says is that
Ariovistus, having been warned by his wise women not to fight ' '

a pitched battle before the new moon,^ contented himself at first,

although the Eoman infantry challenged him, wdth engaging in
cavalry combats, in which he handled the Komans severely and ;

that, in consequence of this success, he conceived a contempt for

the Eomans, and occupied a position beyond their camp, &c. (Sia
TOVTO 6 AptoovLCTTOS 01'^ OLTrdar] evOis rfj Swa/xei Kanoi tiov Poj/xat'tov
. • .

7rpoKa\ovfJi€V(i)v a<f>a<s crvve/JiL^ev, dXA.a tovs iTTTrea? /xera twv (tvvt€t ayjxeviov
acf>L(Ti Tre^wv /xorors CKTre/XTrwv
la)(ypC)<s avTOvs iXvTrei. kolk tovtov
rov Ta(fip€vpLaTO<s (T(f>iov KaraXa/SeLV i7r€)(€tp7]a€
KaTa<f>povrjcra<s ^inptov tl virep
KOL Ka.TC(T)(€ fxkv ovTo, dvTLKaTaXa^ovTOJV 8e KOi iK€ivo)V erepov, &C.^). This
is obviously an inaccurate paraphrase of Caesar's narrative ; for
the challenges of the Roman infantry and the cavalry combats took

^ Rev. arch., xxxiii, 1898, p. 53.

2 Cf. B.G.,i, 50, §§ 4-5. =»
xxxviii, 48, § 2.
 ;

AGAINST ARIOVISTUS 647

place not before but after Ariovistus occupied the position in ques-
tion 1 but even if Dion's account were correct, it would lend no
:

support to M. Colomb's theory, that Ariovistus succeeded in forcing

his way through the gorge of Presentevillers and accomplishing his
flank march by dint of a single lutte acharnee ' in which he defeated
'

Caesar's cavalry. The notion that Caesar would have attempted

to stop his march with cavalry alone, while the legions looked idly
on, is truly comical Caesar at all events had no scruples about
:

employing his infantry before the new moon. The whole episode of
the march, as conceived by M. Colomb, is absolutely incredible. If
Ariovistus had attempted it, he must inevitably have been driven
back with heavy loss. Nothing would have been easier for Caesar
than to seize the commanding position at Arcey, which Ariovistus
is assumed to have occupied, when the head of Ariovistus's column
began to debouch from the gorge of Presentevillers, even if he had
not secured it before nothing, I say, would have been easier, except
:

to destroy the unwieldy column as it was slowly emerging from

the gorge. M. Colomb asks us to believe that Caesar, who, a few
days later, utterly defeated Ariovistus in a pitched battle, was so
imbecile as to allow him to execute a movement v/hich any intelli-
gent centurion would have known how to frustrate.^
M. Colomb points triumphantly, in support of his theory, to
Caesar's statement of the distance which separated the battle-field
from the Rhine les Commentaires,' he asserts, disent que le champ
:
' '

de bataille se trouve a 50,000 pas du Rhin.' ^ M. Colomb will pardon

1 B.G.,i, 48-50.
2 Delbriick {Gesch. der Kniegskunst, i, 452) makes substantially the same
objection to M. Colomb's theory of the turning movement that I have done;
and M. Jullian {Hist, de la Gaule, iii, 231-2, note) gives other good reasons
for rejecting his explanation of the campaign.
The site of the battle has also been placed by various writers (a) at the
Pas de Ronchamp {Mem. de la Soc. d* emulation du Douhs, v^ ser., i, 1876, pp. 442-
55) ; (6) within a triangle whose points are formed by Belfort, Montbeliard,
and Lure (Desjardins, Geogr. de la Gaule rom., ii, 621) ;
(c) just west of Gon-
villars, about midway between Malbouhans and Bavans {Hem. de la Soc.
d' emulation de Montbeliard, xiii, 1881, pp. 17-9, 21-4, 29-31, 36-8, 41-4);
{d) about 3 miles west of Luxeuil (see the map in the last-mentioned periodical
in which are also marked the following sites) (e) between Ronchamp and
;

Champagney (/) just south of Arcey {g) just west of Bavans {h) just south
; ; ;

of Bavans and of the Doubs (P. Cluver, Germania antiqua, ed. 1631, p. 361)
{i) just east of Porrentruy ; and (j) 5 miles from Basle, between Miihlhausen
and the Rhine (Beatus Rhenanus, Rerum Germ, lihri tres, ed. 1610, p. 13). All
these guesses are wrong. All, except the second and the two last, are irrecon-
cilable with the words ut milium amplius quinquaginta locis apertis exercitum
duceret all, except the last, are irreconcilable with the statement that the
:

meeting of Caesar with Ariovistus took place in a great plain ; and the last
is irreconcilable with the statement that Ariovistus halted, on the night before
ho marched past Caesar's camp, at the foot of a mountain, unless we are to
assume that the mountain was one of the low hills between Altkirch, Miihl-
'
'

hausen, and Basle, or that it was one of the northern heights of the Jura;
and in either case it will be o.bvious to any one who can read a map that Ario-
vistus would not have succeeded in cutting Caesar's line of communication with
the source of his supplies. Some other futile guesses are recorded in Rev.
d' Alsace, 1905, pp. 194-6.
^ Rev. arch. ,xxxiii, 1898, p. 44. See also p. 61.
648 CAESAR'S CAMPAIGN
me for correcting him. Milia "passuum quinquaginta (50,000 paces,
or 50 Roman miles) does not occur in any MS. of the Commentaries :

milia fassuum circiter quinque (' about five miles ') occurs in almost
all. But on this question I must refer to a later paragraph.
6. Riistow 1 originally fixed the site in the valley of the upper
Saar, a view which nobody who looks at the map will consider
worthy of refutation, but which Riistow himself subsequently refuted.
According to his revised conclusion,^ the battle was fought in the
plain of Alsace, between Ostheim and Sigolsheim and in order to
;

reach this spot, Caesar marched, not through the pass of Belfort, but
across the Vosges, travelling at the rate of 30 kilometres a day for
seven days
7. C. Winkler has recently discovered a camp with remains which
he calls Roman, near Epfig, about 7 miles north of Schlettstadt.
The carhp is 700 metres long and from 310 to 333 broad, with ditches
3 metres broad and 1 metre 20 to 1 metre 50 deep.-'^ Its area not —
—
more than 22J hectares, or about 55 acres is much too small for
6 legions 4 and, as it has not yet been thoroughly excavated, evidence
;

is still wanting that it was made in Caesar's time.^ Moreover, the

site is so far north that in order to reach it Caesar would have been
obliged to march almost 30 kilometres a day.
All the theories which I have hitherto examined are not only
incapable of proof, but open to serious objection. Two only remain ;

and one or the other is almost certainly right.

8. Colonel Stoffel's itinerary agrees with that of Napoleon, as far
as the latter goes.^ But the battle-field to which he ultimately brings
Caesar is much further north than the sites selected by von Goler,
Napoleon, and von Kampen and his explanation of the opera-
;

tions described in B. G., i, 48-53 differs radically from each of

theirs.
Caesar says that he marched from Besangon to the spot where he
encamped immediately before his interview^ with Ariovistus, in seven
days. How far did he march in that time ? Napoleon says 20 kilo-
metres or 12 miles a day. But, according to Stoffel, 20 kilometres
was too short a distance for the average day's march. Arguing from
Caesar's narrative that he must have advanced rapidly, and yet not
by forced marches, he assumes an average daily march of 27 kilo-
metres, and accordingly conducts Caesar to a point 189 kilometres
from Besan9on, between Ostheim and Gemar, on the left bank of
the Fecht. The tumulus, at or on which the interview took place, he
identifies with the tertre de Plettig ', a hill w^hich rises to the height
'

of 55 metres, or about 180 feet, above the plain. The slopes of this

^ Einleitung zu Clisars Comm., 1857, p. 117.

^ See Thomann, Der franzosische Atlas zu Casars gall. Kriege, pp. 13-5.
^ Mitteiluncjen d. V ereinigung d. Saalbnrgfreunde, March 22, 1908,
pp. 250-1.
* See
"

p. 664. ' Lit. Zentralblatt, 1907,

p. 1429.
^ M. Jullian {Hist, de la Qaule, iii, 230, n.
5) thinks that the circuitous part
of the route, as traced by Stoffel, past Voray, Rioz, Filain, and Vallerois-le-
Bois, was needlessly long, and that Caesar marched by way of Voray and the
valley of the Oignon to Villersexel, from which point to Belfort M. Jullian'
itinerary and Stoffel' s coincide. But see p. 639, supra.
 AGAINST ARIOVISTUS 649

hill are very gentle and although it is close to the eastern side of
;

the Vosges, il se presente, surtout a qui s'en approche par le nord

'

ou par le sud, comme entierement isole dans la plaine.' Subse-

. . .

quently the colonel examines Caesar's description of the manoeuvre

Ly which Ariovistus cut his line of communication with the Aedui
and the Sequani. Caesar's description^ is even more laconic than
usual. Ariovistus broke up his encampment, marched southward to
a point 6 Roman miles north of Caesar's camp, and encamped at
the foot of a mountain {sub monte consedit). Next day he marched
past Caesar's camp, and encamped two (Roman) miles beyond it.
The key to this description, says Stoffel, is in the words suh monte
consedit. Why, he asks, did not Caesar attack Ariovistus while he
was making this flank march ? Evidently because Ariovistus was
protected by the nature of the ground along which he was marching.
To secure this protection, he must have marched either along high
ground or across a plain covered by forests or marshes. But the
words suh monte consedit prove, as every contemporary reader of the
Commentaries would have understood, that he marched along high
—
ground, that is to say, along the lower slopes of the Vosges. Now
in the whole length of the chain, between Cernay and Schlettstadt,
the only part in which Ariovistus could have executed his manoeuvre
is that comprised between the defiles of the Weiss and the Strengbach.
Having marched past Caesar's camp along the heights of Zellenberg,
he encamped between the defiles, on rising ground which dominated
the road by which Caesar's supplies were coming up. We
have next
to fix the spot on which Caesar, after re-establishing his communi-
cations, constructed his smaller camp. Caesar describes it as an
—
idoneus locus, doubtless a strong position commanding the same
road. Stoffel finds it on a spur of the Vosges, between Bebelnheim
and Mittelweier, about 900 metres, or 980 yards, south of the ex-
treme right of the assumed German position.^
Now if Stoffel's determination of the site of the battle rested
solely upon his calculation of the distance which Caesar covered in
hismarch from Besancon, it would not be worth examining. We
may, indeed, infer from Caesar's narrative that, after quitting
Besangon, he marched at more than his usual speed. But it would
be very rash to fix his daily average at exactly 27 kilometres. Besides,
he does not tell us how far he marched on the seventh day. That
last march may, for aught we know, have been a short one. All that
we are justified in saying is that, as he was moving in an open country,
he could have marched at the rate which StofEel fixes.^
But it is upon his other arguments that Stoffel mainly rests his
case. The deduction which he draws from the words suh monte
consedit appears to me wellnigh irresistible. It is a truism to say
that Caesar was economical of words. He would hardly have told
us that Ariovistus encamped at the foot of the mountain, unless the

1 B. G., i, 48, §§ 1-2.

2 Guerre de Cesar et d'Arioviste, pp. 8-10, 13-5, 19, 64-5, 94-6, 98-100.
^ See p. 635.
650 CAESAR'S CAMPAIGN
statement had been essential to his narrative nor would the state- ;

ment have been essential unless it had implied that Ariovistus, after
encamping there, ascended the slopes in order to execute his flank
march without the risk of being attacked. If we may accept Stoffel's
further statement, that the only part of the Vosges along which the
flank march would have been practicable is the part between Hesten-
holz and the defile of the Weiss, his map of the campaign must be
regarded, at least in its main features, as the most satisfactory which
we have. Now Stoffel was a trained observer, and he studied the
ground carefully but I am not sure that the map bears out his
;

contention.! Would not the lower slopes of the Vosges between

Cernay and Thann (where Ariovistus would also have been protected
by the river Thur) and between Thann and Koderen have been
practicable for a flank march ? 2 Stoffel insists, moreover, that the
'
tertre de Plettig corresponds exactly with the description, tumulus
'

terrenus satis grandis, and that between Cernay and Barr no other '

height, separated from the chain of the Vosges, is to be found '.^

This last statement is true and unless, as I have suggested, the ;

tumulus was artificial, it settles the question. It has been objected,

however, to Stoffel's view that the Germans could not have fled
eastward after a battle in which they had faced the east * but ;

Delbriick,^ without denying the force of the objection, argues that it

can be removed. It is quite possible, he says, that the Germans did
not begin the battle in front of their camp, but had previously made
a movement which enabled them to face south and he thinks that ;

this movement may be inferred from Caesar's observation that they

had closed their flanks and rear with their wagons.^ But the answer,
as far as Stoffel is concerned, is simply that he makes the Germans
flee northward.' Another objection is raised by Colomb, who, observ-
ing that if Caesar marched 27 kilometres a day, he must have had
an urgent motive for speed, says that it is unintelligible that he
should then have halted 24 Eoman miles, or 36 kilometres, from
Ariovistus, instead of attacking him at once. Delbriick, however,

1 Carte de France (1 200,000), Sheets 28 and 36.

: The English gentle-
man whom I met at Miihlhausen (see p. 642) confirmed Colonel Stoffel'
statement.
2 When I expressed the contrary opinion in the first edition I had not examined

the map of the country near Cernay with sufficient care. But I have not been
able to explore this region and Stoffel may be right. ;

^ Guerre de Cesar et d" Arioviste,

pp. 14-5.
* See also
pp. 640-1.
Gesch. d. KriegsTcunst, i, 449.
^

« 450.
Ih., p. I agree with Veith {Gesch. d. Feldziige C. J. Caesars, pp. 508-9)
that Delbriick' s suggestion is improbable.
' Like Napoleon, Stoffel traces the German line of retreat down the valley

of the 111 and he remarks that the distance from the site which he identi-
;

fies with the battle-field to the confluence of the 111 and the Rhine, is just
50 Roman miles. The 111 now joins the Rhine 12 kilometres north of Strasburg.
If M. Reclus {Nouv. Geogr. Univ., iii, 1878, pp. 514-5) is to be believed, the
confluence in Caesar's time was above Strasburg. Un proverbe patois,' he adds, '

'
de la haute Alsace dit " Die Ell geht wo sie well "
. . . (The 111 floweth
. . .
'

where it listeth). However, if Caesar wrote quinquaginia (see p. G55, infra), he

qualified it by circiter (about).
 AGAINST ARIOVISTUS 651

reminds us that many errors crept into the memoirs of Frederick the
Great in places where he had no motive for falsification, and that
Caesar also may [as Pollio said] have had lapses of memory. For
my part I cannot see the difficulty. Doubtless Caesar had a motive
for speed, for he expressly says that he marched seven days without
a halt :^ but he also tells us ^ that Ariovistus, on learning his approach,
sent envoys to propose an interview and to reject this proposal
;

would obviously have been bad policy.

Apart from these objections, Stoffel's explanation of the cam-
paign is probable enough but I fear that it is not conclusive,
;

lict us examine the last attempt that has been made to super-
sede it.

9. M. Jullian, who has explored the theatre of the campaign,

acutely remarks that commentators have sometimes been led astray
by the assumption that Gallo -Roman roads were generally laid down
on the lines of old Gallic roads whereas in certain cases, for example
;

when a Gallic town was abandoned in favour of a new Gallo -Roman

capital, they necessarily assumed new directions. Arguing from this
premiss, he observes that while the neighbourhood of Cernay, which
was selected by Napoleon III and von Goler, would have been the
most convenient for Caesar's purpose, he would have marched, not
by the so-called Roman road to which they pointed, but by a Gallic
road on a higher elevation and closer to the Vosges. M. Jullian con-
cludes that Caesar encamped about a mile and a half south-w^est of
—
Cernay that is, on the site selected by von Kampen where he —
would have been protected both by marshes and by the river Thur ;

that Ariovistus, after halting suh monte, north-east of Wattwiller,

marched on the northern bank of the river, along the foot of the
Vosges from Cernay to Thann and thence southward to Roderen,
where he encamped, in order to cut Caesar's communications, near
the bench-mark (signal) fixed by the surveyors who mapped the
'
'

country and that Caesar re-established his communications by

;

pitching his smaller camp on the plateau between Michelbach and

Guerwenheim. M. Jullian notes that his theory is analogous to that
of Stoffel, which corresponds so closely to the principal data of
'

the problem that it very rapidly became popular but he claims

' ;

that his own site has the advantage of being at the place of inter-
'

section of various routes and that it is not in a remote corner of

'

Alsace, but at the principal point of the opening which leads into
'

the plain '.^ It appears to me that his theory is probable enough,

but not more probable than that of Stofcel. For M. Jullian argues
as though Caesar was free to choose his own battle-field, whereas the
choice evidently belonged to Ariovistus and it is not improbable
;

that Ariovistus, after he had failed to seize Vesontio, may have

thought it expedient to lure Caesar as far as possible from his base
1 5. 6^., i, 41, § 5. 2
76.,42,§i.
^ Rev. des etudes anc, xi, 1909, pp. 351-6. Cf. C. Jullian, Hist, de la Gaule,
iii, 232, note. M. Jullian says that the tmnnlns terrenus was near Merxheim ;

but I cannot find it either in the German map (1 25,000), 8heet 3609, or in
:

the Carte de VEtat-Major (1 80,000), Sheet 101.

:
652 CAESAR'S CAMPAIGN
into that remote corner of Alsace '. Between Stoffel and Jiillian
'

excavation alone could decide.^

V. Meusel, to whom I am indebted for an elaborate and most
valuable review ^ of the first edition of this book, tells me that I am
mistaken in identifying Procillus, whom Caesar sent to confer with
Ariovistus, with Troucillus, the interpreter through whom he com-
municated with Diviciacus. I was certainly wrong in calling the
interpreter Procillus, in doing which I adopted the emendation of
Manutius for in the passage ^ in which he is mentioned the MS.
;

readings are Troucillum, TroaciUum, and Traucillum, and the

accuracy of Troucillum is confirmed by inscriptions.'* The question,
however, remains whether the man who was sent to Ariovistus was
not Troucillus. In the two passages ^ in which he is mentioned he
is designated as Procillus in all the MSS., except Vind. I, which
calls him Troicillus. Dittenberger ^ says that he was not Troucillus,
arguing that the way in which Caesar first describes him shows that '^

he had not been mentioned before and also that Caesar calls him
;

a young man (adulescens), whereas he evidently implies that Troucillus

was well advanced in years.^ The reader will draw his own infer-
ences from Caesar's language I will only observe that the mere fact
:

that Troucillus was called a princeps does not prove that he was old.
Assuming that the interpreter and the adulescens were two different
men, it is a remarkable coincidence that both were named Gains
Valerius that both belonged to the Provincia that Caesar had the
; ;

utmost confidence in both and that he described each of them as

;

familiar em suum. In these circumstances I am inclined to conclude

that Troucillus and Procillus were one and the same.
VI. Napoleon^ and Stoffel^^ assume that Ariovistus took the
auxiliaries, whom Caesar drew up on the morning of the battle in
front of the smaller camp, for the two legions which had previously
^ Stoffel was not able to test his theory by excavation, as the site was
covered by vineyards {Guerre de Cesar et cT Arioviste, p. 19).
Caesar says {B. G., i, 54, § 2) that, after the rout of Ariovistus, he led his
army into the country of the Sequani, into winter-quarters {in hiherna in
Sequanos exercituni deduxit) and from this expression it has been inferred
;

that the battle-field was not in the country of the Sequani. Stoffel, however,
argues {Guerre de Cesar et d' Arioviste, p. 117) that if, as he believes, it was in
the extreme north of Sequania, Caesar might have used the words in question.
But, be this as it may, there is no need to suppose that the battle-field was,
strictly speaking, in the territory of the Sequani. It was most probably
in that of Ariovistus (J5. (r., i, 31, § 10), which he had wrested from the Sequani.
See p. 637.
2 Berl phil Woch., Jan. 12, 1901, col. 39-45. ^ B. (?., i, 19,
§ 3.
* Corpus inscr. Lat., iii, 5037 v, 7269, 7287.
;

•^
B. G., i, 47, § 4 53, § 5.
;

**
C. lulii Caesaris comm. de b. G., 15th ed., 1890, p. 394.
' Commodissimum visum est C. Valerium Procillum, C. Valeri Caburi filium,
summa virtute et humanitate adulescentem, cuius pater a C. Valerio Flacco
civitate donatus erat, et propter fidem et propter linguae Gallicae scientiam . .

ad eum mittere, &c. B. G., i, 47, § 4.

® Diviciacum ad se vocari iubet et per C. Valerium Troucillum, principem
. . .

Galliaeprovinciae,familiarem suum,cui summam omnium rerum fidem habebat,

cum eo conloquitur. lb., 19, § 3.
^ Hist, de Jules Cesar, ii, 90-1. "^ Guerre de Cesar etd' Arioviste, -p. 112.
 AGAINST ARIOVISTUS 653

occupied it. nothing in Caesar to justify such an assump-

But there is

tion. What that he posted all his auxiliaries, in view of

he says is '

the enemy, in front of the smaller camp, with the object of creating
a moral effect, as his regular infantry, compared with the enemy,
were numerically rather weak (alarios omnes in conspectu hostium
'

pro castris minor ihus constituit, quod minus multitudine militum

legionariorum pro hostium numero valehat, ut ad speciem alariis
uteretur) ^ and, as Long remarks,^ Ariovistus would not have been
;

duped. I may add that, if Caesar had intended to dupe him, he

would have had to distribute and form up the men of the six legions
which he brought into the line of battle, so as to make them look like
four just as he afterwards, in order to deceive Vercingetorix, dis-
;

tributed the men of four legions in order to make them look like six.^
But the circumstances were very different. Vercingetorix was only
deceived because a considerable distance separated him from Caesar's
army, which was marching in column, parallel with his own. More-
over, the auxiliaries were dressed and equipped differently from the
legionaries. Caesar's object was simply to make as imposing a display
as possible.*
VII. Stoffel conjectures that some of Caesar's light-armed troops
ascended the slope on which he believes the German encampment to
have stood, and, by hurling missiles into the encampment, provoked
the Germans to descend.^ This conjecture is founded upon the state-
ment of Plutarch that Caesar attacked their entrenchments and
'

the hills upon which they were posted which provoked them to
;

such a degree that they descended in fury to the plain '.^ It is not
safe to follow Plutarch but it may be true that Caesar adopted
;

this method of enticing the Germans from a strong position which

he could not have safely attempted to storm.
VIII. Caesar says that he began the battle by attacking the enemy's
left, because he noticed that that part of their line was the weakest
(ipse a dextro cornu, quod earn partem minime firmam hostimn esse
animadverterat, proelium commisit)? These words are generally
explained in the sense that the Roman right wing, commanded by
Caesar in person, struck the first blow. But Stoffel denies this.
'
A moins de circonstances exceptionnelles,' he says, il etait de '

' B. G., i, 51, § 1.

^ Decline of the Roman Republic, iv, 41. ^ B. G., vii, 35, §§ 3-4.
Guischard [Mem. crit. et hist., 1774, p. 304), who believes that Ariovistus
*

did take the auxiliaries for legionaries, says, as I have said, that he would
not have been duped unless they had been armed and accoutred like legionaries ;

and accordingly he assumes that they were not auxiliaries at all, but the
nucleus of the legion which Caesar called Alauda but it is needless to tell
:

scholars that this is nonsense.

M. Jullian [Hist, de la Gaule, iii, 237, n. 3) believes that the auxiliaries {alarios)
were simply Caesar's cavalry. I do not agree with him. In B. C., i, 73, § 3
(with which cf. ii, 18, § 1), cohortes alariae are expressly distinguished from
cavalry ; and it is matter of common knowledge that after the Social War
the term alarii denoted auxiliary infantry, unless (as in Tacitus, Ann., iii, 39;
iv, 73 ; xii, 27 ; xiii, 35) the word equites was added.
^ Guerre de Cesar et d Arioviste,
p. 68.
8 Caesar, 19. ^ B. G., i,
52, § 2.
654 CAESAR'S CAMPAIGN
priucipe, dans les combats de Tantiquite, que rattaquc se fit en . . .

meme temps sur tous les points de la ligne.' ^ He takes the passage
to mean that Caesar reserved the command of the right wing for
himself dans I'espoir de remporter un prompt succes qui deconcer-
'

terait les Germains de I'aile droite et lui faciliterait la victoire com-

plete '. But may not Caesar have temporarily refused his left wing
in the manner described by Vegetius (iii, 20) cu7n instructae acies :

ad congressum veniunt, tunc tu sin,istram alam tuam a dextra adversarii

loncjius separahis dextram autem alam tuam sinistrae alae illius
. . .

iunge, et ihi primum inchoa proelium ^

IX. Caesar, in his description of the battle, says that many of '

our men actually leaped on to the phalanx [or phalanxes], tore the
'
shields out of their enemies' hands, and stabbed them from above
{reperti sunt complures nostri qui in phalanga [v.l. phalangas] insilirent,
et scuta manihus revellerent, et desuper vulnerarent).'^ Meusel*^ main-
tains that the reading phalangas is wrong, because a phalanx fought
not in divisions but in one mighty column '
and Caesar, after '
;

saying that the Germans formed their host at equal intervals in '

tribal groups' (generatimque constituerunt paribus intervallis),^ adds

that in the battle they formed in a phalanx {phalange facta,^ not
phalangihus factis). This reasoning leaves me unconvinced. Very
likely Caesar wrote phalanga but if so, he only meant that Ariovistus
:

adopted the phalanx formation, not that the phalanx was one and
undivided for what would have been the use of elaborately forming
;

up the tribal groups at equal intervals' if the intervals were im-

'

mediately afterwards to be suppressed? Moreover, as M. Jullian^

observes, it may be inferred from Tacitus ^ that the Germans fought in
divisions and, as I have remarked before,^ the Macedonian phalanx,
;

before it degenerated, did the same.

X. Allowing for the losses which Caesar had suffered in the battle
with the Helvetii, Stoffel calculates that he had about 27,000 legiona-
ries in the battle with Ariovistus.^ This calculation rests on the
assumption that when he began his campaign against the Helvetii,
the strength of each of his six legions was about 5,000 men but ;

I have given reasons for believing that, if those legions, two of which
were newly raised, were of normal strength, they were each 6,000
strong.i^ It is, however, impossible to tell whether the four veteran
legions had been brought up to the normal standard when Caesar
took command of them. Nor can we tell what loss Caesar had suffered
in the battle with the Helvetii we only know that it was heavy.:

Of the assumed 27,000 men Stoffel believes that six cohorts, or,
according to his estimate, about 2,700 men, were left to guard the
large camp that there were about 300 invalids
; and therefore that ;

^ B. G.,
^ Guerre de Cesar et d'Arioviste, p. 113. i, 52, § 5.
^ Jahresb. d. philol. Verems zu Berliri, xxiv, 1894, pp. 229-30.
' lb., 52, § 4.
* B. G., i, 51, § 2.
« Hist, de la Gaide, iii, 42, n. 10, 238, n. 5. Cf. Stoffel, Guerre de Cesar
et d'Arioviste, p. 09. ' Ann., ii, 45 ; Hist., iv, 20.
^ See p. 593, aud cf. Daremberg and Saglio, Diet, des ant. grecques et rom.,
iv, 425. ^ Guerre de Cesar et d'Arioviste,
p. 83.
'" See
pp. 559-63.
 '

AGAINST ARIOVISTUS 655

there were some 24,000 legionaries in the line of battle. The small
camp, he assumes, was guarded by some of the auxiliaries. When
Caesar joined Fabius near Ilerda, and was about to march towards
the camp of Afranius, he left six cohorts to guard Fabius's camp
and the lower bridge over the Sicoris and Stoffel infers that he ;

detached the same number of cohorts in the battle with Ariovistus,

in which he had the same number of legions.^ But during the battle
with Ariovistus there was no bridge to be guarded. Still, the colonel's
estimate may be not far from the truth for in the battle of Pharsalia,
;

where Caesar had 80 weakened cohorts, or about 22,000 men, in

action, he left seven cohorts to hold his camp.^
XI. StofEel believes that Ariovistus had about 36,000 men in the
battle.^ He
points out that, according to Caesar,^ the entire German
host amounted to 144,000 souls and from the estimate which
;

Caesar gives of the Helvetian forces ^ he infers that one -fourth of these
were fighting men, of whom 6,000 were cavalry.^ Of course we cannot
implicitly trust the figures with which Diviciacus supplied Caesar ;

but it should seem from his account of the battle that the Germans
outnumbered the Romans.
XII It has often struck me that there is an apparent inconsistency
.

between the fact that the Germans escaped from the battle-field
and Caesar's statement that, before the battle, they closed their
rear with a semicircle of wagons, to do away with all hope of escape
'

(ne qua spes in fuga relinqueretur)? Frontinus ^ offers a solution

of the puzzle, but we cannot tell on what authority. As the '

Germans,' he says, being hemmed in, were fighting desperately,

'

Caesar ordered that they should be allowed an exit, and fell upon
them when they were fleeing {Caesar Germanos indusos, ex despera-
'

tione fortius pugnantes, emitti iussit fugientesque aggressus est).

XIII. The enemy,' writes Caesar, all turned tail, and did not
' '

cease their flight until they reached the Rhine, about five miles from
the battle-field (omnes hostes terga verterunt, nee prius fugere desti-
'

terunt quam ad flumen Rhenum milia passuum ex eo loco circiter

—
V pervenerunt).^ All the MSS. except Vind. C, which has //,
a number often confounded by the copyists with V have the reading —
quinque or V. Plutarch,!^ on the other hand, writes araScovs rerpa-
Koo-Lovs (50 miles), or, according to some MSS., rptaKoa-Lovs (37J
miles), and Orosius^^ and Eutropius^^ quinquaginta milia pas-
suum and all of these writers must, one would think, have followed
;

some MS. of Caesar. If so, it is certain that MSS., several centuries

older than any which are now extant, had the reading quinquaginta.
Schneider 1^ thinks that the word circiter ('about') agrees better
with quinquaginta than with quinque. Mommsen ^^ places the battle-
^ B. C, See Guerre de Cesar et d' Arioviste, pp. 110-1.
I, 41.
'^
B. C, 89, § 2.
iii, ^ Q^erre de Cesar et d'Arioviste,
pp. 84-5.
« B. a., i, 31, §§ 5, 10. ^ Ih., i,
29, §§ 2-3. See, however, p. 241.
^ lb., 48, § 5. Dion Cassius, xxxviii, 47, § 5 Plutarch, Caesar, 19;
See also ;

and Appian, Celtica, 1, § 3. ' 5. (r., i,

51, § 2.
« ii, 3,
§ G. » B. G., i, 53, ^» Caesar, 19.
§ 1.
11 12 13
vi, 7, § 10. vi^ 17^ Caesar, i, 115.
1* Rom. Gesch., iii, 1889, p. 256, note {Hist, of Rome, v, 1894, p. 47, n. 1).
656 CAESAR'S CAMPAIGN
field at Cernay, the spot fixed upon by von Goler and Napoleon,
which is at least 15 miles from the Rhine, as the crow flies he does :

not say that the channel of the Rhine, in the latitude of Cernay,
was different then from what it is now and yet he argues that
;

'
the whole description of the pursuit continued as far as the Rhine,
and evidently not lasting for several days but ending on the very
—
day of the battle, decides the authority of tradition being equally
—
balanced in favour of the view that the battle was fought five, not
fifty, miles from the Rhine Long ^ says, A flight or pursuit of
' !
'

fifty miles after a hard battle is impossible and again, Caesar

'
;
'

says nothing of night coming on which would have stopped the

;

pursuit in an unknown country even if fatigue did not. But he says

that the routed Germans did not stop and the horsemen pursued
them to the Rhine and some of the Germans swam the river
:
;

a thing impossible after a heat of fifty miles.' Napoleon ^ insists

that the distance was 50 miles, because (1) Caesar would not have
used the words neque prius fugere destiterunt (' did not cease their
flight') of a flight of merely a few miles
'
and (2) the Germans
'
;

probably retreated down the valley of the 111, which they had
'

previously ascended '.

All that I have to say is this. Schneider's argument that circiter
harmonizes better with quinquaginta than with quinque is futile for ;

in B. G., ii, 13, § 2, Caesar uses the phrase circiter milia fassuum V.
On the other hand, assuming that the battle took place at any point
west of the 111, the distance cannot have been only five miles, unless
the 111 was then regarded as an arm of the Rhine. As to Long's
argument, the pursuit after the siege of Alesia was not stopped by
night, for it began just after midnight. The country was open the :

Germans, at any rate, knew it and the Romans only had to follow.
;

But there is considerable force in Long's other objection and there ;

is very little force in the argument which Napoleon bases upon the
phrase neque prius fugere destiterunt. Caesar uses the very same
phrase in B. G., iv, 12, § 2, where he is describing a flight which could
not have extended over anything near 50 miles. Von Goler asserts,
on the authority of a pamphlet by an engineer named Tulla, that the
111 was at that time a branch of the Rhine .^ As Long says,^ this '

hypothesis or assumption settles all difficulties, and makes Caesar's

narrative perfectly intelligible.' Yes, if it is true but is it ? It is
:

scarcely credible that Caesar should not have known that a very few
miles east of the 111 flowed the main stream of the Rhine and if he did ;

know this and yet spoke of the 111 as the Rhine, he used the word
Rhenus in two consecutive chapters to describe two different streams.
For in chapter 53 he describes the flight of the Germans, and in 54 he
tells us that, on hearing of the defeat of their countrymen, the Suebi,
who had advanced to the banks of the Rhine, returned home and ;

^ Caesar, pp. 97-8.

^ Hist, de Jules Cesar, ii, 93, n. 1. Napoleon assumes that the fugitives
crossed the 111 first, and afterwards crossed the Rhine at Rheinau.
3 Gall. Krieg, 1880,
p. 58 and n. 2.
* Decline
of the Roman Republic, iv, 43.
 AGAINST ARIOVISTUS 657

by the Rhine in this passage he could not have meant the 111. If
I am Caesar could not have written quinque unless he made
right,
a slip, because the Rhine is nowhere nearer than 12 miles from any
site with which the battle-field can be identified. But I agree with
Long that it is hardly credible that the Germans should, after a
desperate battle, have fled 50 miles in one heat,^ still less that some
of them should have then swum the Rhine I cannot see why they :

should have fled 50 miles when the Rhine, to cross which was their
only object, was not more than 12 to 15 miles away ^ and I can ;

therefore only make the lame suggestion that Caesar may have
written XV.
XIV. The battle with Ariovistus was fought in September for
we may gather from a statement in B. G., i, 40, § 11 the standing — '
;

corn was already ripe (ia7n esse in agris frumenta matura)

'
that —
Caesar was at Vesontio during the latter half of August. It was
new moon on September 18 and the battle was fought before
;

the new moon. The date of the battle was therefore, Stoffel infers,
about September 14.^

CAESAR'S ACCOUNT OF ONE OF THE MOTIVES OF

THE BELGAE FOR CONSPIRING IN 57 b.c.
Coniurandi has esse causas quod ah non nullis Gallis solli-
. . .

citarentur, partim qui mobilitate et levitate animi novis imperiis

. . .

studehant.^ Schneider ^ understands Caesar to mean that some

of those who had (virtually) submitted to the Romans were tired of
their supremacy and wished to exchange it for that of the Belgae.^
This is the sense in which I had myself understood the passage. But
perhaps Caesar's meaning is not as definite as this. He may perhaps
only mean that the Gauls, being tired of Roman supremacy, were
bent on making a revolution of some sort."^

^ Even the exhausting retreat from Waterloo to Charleroi was not more than

25 miles. See, however, E, Desjardins, Alesia, suivie (Tun appendice ren-—

fermant des notes inedites ecrites de la main de Napoleon I^'' sur les campagnes
de Jules Cesar, p. 148.
- M, Jullian {Hist, de la Gaule, iii, 231, n.
4) thinks that the Germans fled to
Strasburg because, except Basle, it was the nearest of the principaux lieux '

de passage but there must have been boats elsewhere

' : and apparently ;

very few found them {B. G., i, 53, § 2). May we suppose that some fled to
the nearest point of the Rhine, and others, who escaped pursuit, towards
Strasburg ?
^ Guerre de Cesar et d' Arioviste,
p. 80.
* B. G.,\i, " Caesar, \, 121.
1, §2.
« Dr. W. G. Rutherford {Caesar, Bk. ii, refers to B. 6^., i, 17, § 3, where
p. 47)
the Aeduan malcontents are said to have argued that it was better ... to '

have Gauls for their masters than Romans ' {praestare . . . Gallorum quam
Romanorum imperia perferre).
' See Long's Decline of the Roman Republic, iv, 44.

1093 U U
 658

DID ADRA SUCCEED GALEA AS COMMANDER-IN-

CHIEF OF THE BELGAE?
According to Dion Cassius, who, if the MSS. are right, does not
mention Galba, the commander-in-chief of the Belgae in 57 B. c. was
named Adra.^ De Saulcy affirmed that a Gallic coin had been found
at Pasly, bearing the name Adra and he identified the personage
;

whose name appears on the coin with the Adra of Dion Cassius, and
suggested that he might have succeeded Galba. The name on the coin,
however, is not Adra, but Arda? Moreover, A. Michaux pointed out
that, according to Dion Cassius, Adra was appointed at the beginning
of the war, and argued that he commanded the Bellovaci alone.
If so, Dion Cassius made a mistake but it seems more likely that
;

the MSS. are at fault.

WHERE DID THE BELGAE MUSTER IN 57 b.c. ?

Von Goler ^ believes that the Belgae mustered in the country of the
Suessiones, apparently at or near Noviodunum, which was close to
Soissons. The reason which he gives is that their commander-in-
chief was the king of the Suessiones ^ and M. Jullian ^ adds that
;

the valley of the Aisne was the natural route for the Bellovaci and the
Suessiones, who furnished the strongest contingents to the confederate
army. On the other hand, as General Creuly points out,"^ the king
may have wished to avoid the injury which the presence of such
a huge host could not but have inflicted upon his standing crops.
Besides, there is a passage in Caesar which suggests that the muster
could not have taken place anywhere in the valley of the Aisne.
Describing his own movements at the outset of the campaign, he
writes, Finding that all the Belgic forces had concentrated and were
'

marching against him, and learning from the reconnoitring parties

which he had sent out and from the Remi that they were not far off, he
pushed on rapidly, crossed the Aisne and encamped near its
. . .

banks (fostquam omnes Belgarum copias in unum locum coactas ad

'

se venire vidit, neque iam longe ahesse ah iis, quos miserat, explora-
torihus et ah Remis cognovit, flumen Axonam exercitum traducere
. . .

maturavit atque ihi castra fosuit).^ He was indisputably on the south

of the Aisne when he began his march. If the Belgae had mustered

^ xxxix, 1, § 2. Xylander and Bekker read TaK^av instead of 'Ahpav.

^ See Cat. des inonnaies gaul., ed. Muret and Chabouillet, pp. 204-5, n^^
8841-57, and H. de la Tour's Atlas de rnonnaies gaul., 1892, pi. xxxvi. The
coins of Arda are assigned in the Catalogue and the Atlas to the Treveri, who,
in this war, were on the side of Caesar. See B. G., ii, 24, § 4.
^ B^dl. de la Soc. arch, de Soissons, ix, 1878,
pp. 64-71.
* Gall. Krieg, 1880, p. 67. ' E.G., ii,
4, § 7.
^ Hist, de la Gaule, iii, 251, n. 5.

' Rev. arcli., nouv. ser., viii, 1863, p. 298. ^ B. G., ii, 5,
§ 4.
 WHERE DID THE BELGAE MUSTER ? 659

in the valley of the Aisne, is it likely that they would have marched,
as they undoubtedly did, on the north of that river, to attackhim ? i
Nay, they would hardly have done so even if he had already encamped
on the north for, if they had marched on the south, they would,
;

by the mere fact of doing so, have compelled him to abandon his
position for fear his communications should be cut and M. Jullian ;
'^

himself admits that if they had concentrated on the Aisne, their

natural line of march would have been the road from Soissons to
Reims. Moreover, it is, as I have shown,^ impossible to find a satis-
factory site for Bibrax, the stronghold which they attacked when
they were marching against him, at any point on or near the road
leading from Soissons to the place where he crossed the Aisne.^
Finally, as I show on page 670, it is difficult to explain why the
fugitive Suessiones, after the defeat of the Belgae, reached Novio-
dunum some hours later than Caesar, although they had begun their
flight long before he began his march, except on the hypothesis that
in the early part of their flight they had moved northward from the
Aisne. It is probable, therefore, that the point where the Belgae
concentrated was somewhere north of the Aisne, and at a considerable
distance from it. Assuming that Caesar crossed the Aisne at Berry-au-
Bac, Creuly infers that it must have been somewhere north of Laon ^ ;

for otherwise, he argues, their line of march would have been

threatened by Berry-au-Bac, and Caesar would not have hesitated to
seize that point at the earliest possible moment. We must conclude,
he continues, that they were at least as far north as the neighbour-
hood of La Fere on the Oise, whence they could march against Caesar
and Remi, by way either of Laon or of Soissons and
his allies, the ;

that it was only when Caesar had ascertained from his scouts which
route they had decided upon, that he put his own troops in motion.
Although we cannot be absolutely sure that Caesar was ready to
march before he actually did so. General Creuly's conclusion appears
to me well grounded for it may be inferred from Caesar's narrative
;

that the Belgae had begun to move against him before he marched
to meet them he probably had to make a whole day's march at
;

least before he could reach his camping-ground and on the night

;

of his arrival the Belgae were still at Bibrax, eight Roman miles away.

WHERE DID CAESAR CROSS THE AISNE IN HIS

FIRST CAMPAIGN AGAINST THE BELGAE ?

The number of pamphlets and articles that have been published

upon this question is bewildering. It has been discussed as hotly as
the question of Cenabum or the question of Alesia.

2 Rev. arch., nouv. ser., viii, 1863, p. 298. » See

pp. 398-400.
* —
Unless that place was Berry-au-Bac and this, as I shall show in the next
article, is doubtful —
Bibrax was certainly not in the valley of the Aisne.
'-'

Creuly's argument would hold good if Caesar crossed the Aisne at Pontavert.
U U 2
 660 WHERE DID CAESAR
I. 1. Since Napoleon published his Jlistoire de Jules Cesar, the

prevalent view has been that Caesar crossed the Aisne at Berry-au-
Bac, and pitched his camp near Mauchamp, on a hill between the
Aisne and the Miette, a small stream which flows into the Aisne
on its northern bank. Napoleon was not, indeed, the first to adopt
this site ;for von Goler and more than one French antiquary ^ had
anticipated him but he claimed that the results of the excavations
:

which had been undertaken by his orders had established its identity.
Besides Berry-au-Bac, however, Conde-sur-Suippe, which is about
3 miles higher up, and Pontavert and Pontarcy,. which are respec-
tively about 4 and 11 miles, as the crow flies, lower down the Aisne,
have found ardent champions.
Caesar tells us that he marched from the country of the Sequani to
the frontier of the Belgae, entered the territory of the Remi, and
remained there some days. It is probable that, during this time, his
head-quarters were at or near their chief town, Durocortorum
(Reims). Hearing that the hostile Belgae were marching against
him, he put his own army in motion, crossed the Aisne by a bridge
at a point which was within an easy march of the eastern frontier
of the Suessiones, and * there (ihi)'
—that is to say, close to the right
—
bank pitched his camp on a hill, while on the left bank he stationed
six cohorts to guard his communications.- He describes the hill and
the measures which he took to render it impregnable in the following
sentences Ubi nostros non esse inferiores intellexit, loco 'pro castris
:

ad aciem instruendam natura oportuno atque idoneo, quod is collis uhi

castra posita erant paululum ex planitie editus tantum adversus in
latitudinem patehat quantum loci acies instructa occupare poterat, atque
ex utraque parte lateris deiectus habebat et in fronte leniter fastigatus
paulatim ad playiitiem redibat, ab utroque latere eius collis transversarji
fossam obduxit circiter passuum CCCC et ad extremas fossas castella
constituit ibique tormenta conlocavit, ne, cum aciem instruxisset, hostes,
quod tantum multitudine poterant, ab later ibus pugnantes suos circum-
venire possent. Hoc facto duabus legionibus quas proxime conscripserat
in castris relictis ut, si quo opus esset, subsidio duci possent, reliquas VI
legiones pro castris in acie constituit. Hostes item suas copias ex castris
eductasinstruxerunt? According to my unbiased comprehension of this
passage, the hill rose gradually from level ground on the right bank
of the Aisne it descended gradually to level ground in front
: its :

flanks, on the right and left, did not insensibly merge into the plain,
but descended to it, so to speak, with a strongly marked slope ;

and its length, or extension from right to left, was just sufficient to
allow six legions to be drawn up on it in line of battle. Between
this hill and the enemy's camp, which was in front of it, was a small
marsh. In order to prevent the enemy from outflanking him,
Caesar drew a trench, about 400 paces, or 650 yards long, crosswise
—
that is, at right angles with the extension of the hill on either flank
of it ; and at each end of each ditch he constructed a redoubt. Then,
^ e.g.A. Piette in Bidl. de la Soc. acad. de Laon, viii, 1858, p. 188.
2 B. G., i, 1, § 2 ; ii, 2, § 6 3-5.
;
^ j^^ g, §§ 3-5.
 s

CROSS THE AISNE ? 661

leaving two legions in reserve in his camp, he drew up the other six
in line of battle in front of his camp.
I am certain that the foregoing description would agree substan-
tially with the interpretation of Caesar's narrative that any unpreju-
diced scholar, examining the passage without reference to other
sources of information, would give and I find that it is substantially
;

identical with the interpretation given by Long.^ If it represents

Caesar's meaning, Napoleon's Plan,^ which traces one of the trenches
from the north-western angle of the camp to the Miette, and the other
from the south-eastern angle to the Aisne, is obviously wrong.^
Accordingly Dr. Gunion Rutherford offers a different interpretation ;

and, what is more, he maintains that he alone of all the editors has
explained Caesar's narrative correctly .* The key of this description,'
'

he says, is fro castris, which proves that Caesar was looking west-
'

w^ard towards the Aisne along the axis of the hill.' Where the proof
is, I cannot see. In fact Rutherford himself supplies disTpiooi. In
his Vocabulary (p. 124) he translates pro castris by in front of the '

camp.' The front of the camp, it is needless to say, was that side of
it which faced the enemy and the side of this particular camp which
;

faced the enemy was confessedly the north. Rutherford's Vocabulary

is justified by Caesar's usage in every other passage in which he
:

writes pro castris he means on the side of the camp which faced the
'

enemy (see B. G., i, 48, § 3 51, § 1 iv, 35, § 1 v, 15, § 3 16, § 1

'
; ; ; ;
;

37, § 5 vii, 24, § 5

; 66, § 6
; 68, § 1
; 70, § 2 83, § 8 ;89, § 4 ; ; ;

B. C, i, 43, § 4 iii, 56, § 2). The key which Rutherford found only
;
'
'

opens the door to fresh mistakes. Having mistranslated pro castris, he

is compelled to mistranslate infronte. In fronted he says, refers to
' '

that end of the hill's ridge furthest removed from the camp.' A scholar
like Rutherford could not have made such a mistake as this if he had
not been biased by Napoleon's Plan. Let any one look at the Plan, and
he will see at once that the front of the hill can only be that side of it
which faced the enemy. Again, Rutherford's translation of infronte
obviously involves a forced interpretation of the words ex utraque
parte [collis] lateris deiectus habebat (' on either side its flanks descended
abruptly '). Moreover, he forgets that, according to Caesar, the hill,
from its right to its left flank, was just wide enough to enable the
line of battle to be formed along it whereas, according to his
;

interpretation of Caesar and according to Napoleon's Plan, the hill

was wide enough to allow the line of battle to be formed upon it
alongside of the camp, that is to say, wider, by the length of one side
of the camp, than Caesar says. It is perfectly clear that, if Caesar''
narrative was not misleading, the line of battle was formed in front
of, not alongside of, the camp. Finally, Caesar says that his object
in constructing the two trenches was to prevent the enemy from
surrounding his troops on their flanks {ne, cum aciem instruxisset,
hostes . ah lateribus pugnantes suos circumvenire possent). Would
. .

* Caesar, p. 115. See also Schneider's Caesar, i, 140-1.

^ Hist, de Jules Cesar, Planche 8. Sec Plan facing p. 71 of this volume.
' Cf. (J. Long, Decline of the Roman Republic, iv, 50, n. 5.
* Caesar, Books II-III, Preface and pp. 55-0.
662 WHERE DID CAESAR
he have used the plural lalerihus if he had unly nieaiit his right
flank ?

And now, having acted as Advocatus Diaholi, I will say all that can
be said for the defence.^ It might possibly be argued that Caesar
loosely used the phrase a laterihus (on the flanks) instead of a dextro
latere (on the right flank), because he uses the words rifas and in ripis
more than once when he is only speaking of one bank of a river.^
Again, a glance at the Plan will show that the camp near Berry-au-Bac
(assuming that it was made by Caesar) could only have faced the
extreme left of the Roman line the Roman troops would naturally
:

have been drawn up on the left of the camp and, as they would have ;

been confronting the enemy and at the same time resting upon the
support of the camp, even though not, so to speak, between it and
the enemy, it might be pleaded that the ground which they occupied
could have fairly been described as loco pro castris. But I fear that
the plea would be brushed aside for Caesar says clearly that the
;

hill extended, facing, the enemy, over the exact space which
'

the line would occupy and Rutherford's interpretation of infronte

'
;

and ex utraque parte could by no ingenuity be upheld.^

Dittenberger ^ suggests that Caesar may have forgotten the nature
of the ground and the direction of the trenches. To my mind such
a lapse of memory is inconceivable and it is safe to say that Ruther-
;

ford would never have explained Caesar's text as he did if he had

not assumed the correctness of Napoleon's Plan and been guided
by it.

I shall now examine the other objections that have been made to
Napoleon's view.
(1) The western slope of the hill of Mauchamp, where it descends
towards the Miette, is so extremely gentle that it could not be
described by the words lateris delectus.^ This objection does not
directly touch Rutherford's interpretation of the text for, as we ;

have seen, he identifies the lateris delectus with the northern and
southern sides of the hill if he is wrong, the objection, as any one
:

who has seen the ground will admit, can only be removed by

^ In the first edition (pp. 646-8, 651), ahhough I printed a criticism of

Rutherford's explanation, which I had written after I first read it, and which
I have just presented anew in a sHghtly modified form, I tried to show that he
was substantially right: but I was biased by the belief that Stoflfel's discovery
was conclusive ;and my defence, which, except one unsound argument, was
also virtually identical with what I have said here on Rutherford's behalf,
involved the unwarrantable assumption that Caesar's description was mis-
leading. Professor Lehmann's articles, to which I refer on p. 666, first led
me to reconsider the question.
^ H. Meusel, Lex. Caes., ii, 1750-1.

^ M. Jullian {Hist, de la Gaule, iii, 255, n.

1), who refers his readers to my
former defence of Rutherford, interprets in froute as du cote regardant
'

Beaurieux', which he identifies with Bibrax. But Bibrax was no longer

important ; and Caesar was not regarding Beaurieux, but the enemy, who
'
'

had quitted Bibrax and were on his north {B. G.,u, 7-8). M. Jullian's explana-
tion could only be accepted if Caesar'w lino of battle had faced Beaurieux.
* Caesar, 1890, p. 122.
^ Bull, de la /Soc. acad. de Laon, xiii, 1863, p. 177.
 ;

CROSS THE AISNE ? 663

supposing that the word delectus could be apphed to the slightest per-
ceptible slope. I doubt whether such an interpretation is admissible ;

and Professor A. Gudeman, the learned author of the article Delectus

in the Thesaurus linguae Latinae, agrees with me.i I am convinced,' '

he writes, that you are absolutely correct in taking delectus of

'

a sharp, fairly steep gradient. This interpretation is confirmed by

the use of deicio, which, apart from its tropical uses, invariably
denotes " praecipitem or praeceps dare ".' And, even supposing that
Rutherford's explanation can be accepted, the northern and southern
slopes of the hill are hardly less gentle than the others.
(2) The trench which, according to Napoleon's Plan, touched the
Aisne, is only 400 metres long, whereas, according to Caesar, each
trench measured about 400 paces, or nearly 592 metres and it shows ;

no trace of a redoubt.^ To this, E. Fleury replied by anticipation

that the Aisne had changed its course since Caesar's day and thereby
destroyed all traces of the end of the trench and of the redoubt.^
But, according to M. Jullian (who nevertheless accepts the results
of Stoifel's excavations), there is no evidence that the Aisne has
changed its course.^ A. E. Poquet ^ indeed affirms that the change
is attested by the archives of the commune of Juvincourt for 1791,
in which it is stated that the Aisne a forme une ile et change I'etat
'

des lieux and he goes on to speak of son ancien lit qu'on reconnait
'
;
'

parfaitement encore but I can find no evidence that its course

'
:

in Caesar's day corresponded with Napoleon's Plan.

(3) The northern trench shown in Napoleon's Plan is cut, 80 metres
from its termination, by four small trenches, which were supposed
by Stoffel to be the remains of the northern redoubt but if so, why ;

was the northern trench prolonged 80 metres beyond them? ^ And

since the small trenches were at the most one metre, and in some
places only 20 centimetres deep, and the potsherds and other
antiquities found near them in pits were post-Caesarian, how can
they be identified with one of Caesar's redoubts Poquet replies V
that the prolongation of the trench was simply intended to complete
'

the defence of the redoubt but this argument is unsound, for

' ;

Caesar says distinctly that the redoubts were at the ends of the '

trenches (ad extremas fossas). On the other hand, the contents of

'

—
the pits fragments of jars (amphorae), a piece of wrought iron, two
coins of the Remi, and a stone axe ^ though the circumstances of —
^ The professor kindly sent me a list, taken from a proof of the article, of the

—
passages in which deiectus occurs, Caesar, B. G., ii, 8, § 3 ; 22, § 1 29, § 3 ;

Livy, ix, 2, § 9 Ovid, Met., i, 571

; Vitruvius, vi, 3, § 1 ; Valerius Maximus,
;

iii, 2, § 1;Seneca, Dial., v, 1, § 5 vi, 8, § 14 Nat., iii, 11, § 6 ; Epist., 56, § 3

; ; ;

Pliny, Nat. Hist., ii, § 179 ; xxxiii, ^ 75 ; xxxvii, § 88 ; Statius, Theb., iv, 272 ;
Sidonius, Epist., v, 13, 1.
2 Bull, de la Soc. acad. de Loon, xiv, lb., xiii, 1863, p. 185.
1864, p. 103. ^

* Oral communication from M. Jullian. iii, 253,

Cf. his Hist, de la Gaule,
note.
''
Bull, de la Soc. hist. de Soisso)ts, xvii, 1863, j). 362, n. 1.
. , .

^ Bull, de la ISoc. acad. de Laon, xiii, 1863,

p. 179.
7 lb., xiv, 1864, p. 104.
'^
Bull, de la Soc. hist. de Soissons, xvii, 1863, p. 361 ; Bev. arch., nouv.
. . .

scr., v, 1862, p. 360.

664 WHERE DID CAESAR
their discovery were not recorded with modern precision, do not
appear to preclude the attribution of the works to Caesar.
The remaining objections are unimportant. It has been said that
the camp at Mauchamp, being only 41 hectares, or about 101 acres, in
area, was insufficient for Caesar's six legions ;
^ that the trenches
were only from 4 to 5 metres (about 13 to 16 feet) wide, not, as
Caesar's statement requires, 18 feet and that the camp had five
;

gates, whereas Roman camps had only four.^ But a smaller camp,
described by Hyginus, was intended for three legions, which, with
the foreign auxiliaries and the camp-followers who accompanied
them, amounted to at least 40,000 men ^ in digging the trenches
;

Caesar's instructions may not have been punctiliously observed ;

and we do not know whether the number of gates in a Roman

camp was invariable. Again, it has been objected, on the one
hand, that the Miette never could have flowed through a marsh,
for one of the transverse trenches actually touches it * on the ;

other, that if the marsh which Caesar described had been traversed
by a stream, he would have mentioned it.^ But the surrounding
land, according to Poquet,^ is still known as the marais de la '

Miette ', a name which, as a local innkeeper assured me, it fre-

quently deserves and Caesar did not mention the Essonne, the
; —
stream that undoubtedly traversed the marsh which Labienus
attempted to cross in 52 b. c."^ Besides, Poquet affirms that the
channel of the Miette itself was dug in order to drain the stagnant
waters.
Let us now test the claims of the other places that have been
proposed for Caesar's passage of the Aisne.
2. On behalf of Pontarcy all that can be said has been said by
Caignart de Saulcy. After the publication of the results of Stoffel's
excavations he acknowledged that he was wrong ^ but it may be ;

well, in order finally to settle the question, to prove that his former
opinion was untenable. First, Pontarcy has always formed part of
the diocese of Soissons, and was therefore probably in the country
of the Suessiones, not, as Caesar's narrative requires, in that of
the Remi.
Secondly, according to A. Piette,^ who made a special study of the
ancient roads in the department of the Aisne, no ancient road
passed through Pontarcy.
Thirdly, from Caesar's statement that one side of his camp was
covered by the Aisne, de Saulcy infers that the hill on which the

^ Bull, de la Soc. acad. de Loon, xiv, 18C4, p. 103.

* lb., xiii, 1863, pp. 182-3.
^ Hyginus, De munitionibus castrorum, § 21 (ed. A. von Domaszewski, 1887).
Cf. Daremberg and Saglio, Diet, des ant. grecques et rom., i, 951.
* Bull, de la Soc. acad. de Laon, xiv, 18(54, p. 104.
^ Neue Jahrh. f. d. Mass. Altertum, vii, 1901, pp. 506-9. I have ignored
one or two objections which seem to me frivolous.
® Jules Cesar et son entree dans la Gaule belgique, 1864, p. 64.
' B. G., vii, 57, § 4 ; 58, § 1. Cf. pp. 775-8.
* Journal de V Aisne, 7 mai, 1862.
^ Itin. gallo-rom. dans le dep^ de V Aisne. 1862, map at end of volume.
 CROSS THE AISNE ? 665

camp stood s'etendait perpendiculairement au cours de I'Aisne et

'

dominant terrain compris entre elle et la riviere, ne laissait pas

le
d'attaque a craindre dans cette direction '.^ He was driven to
adopt this singular interpretation of Caesar's words because the
plateau of Comin, the only hill which he could find near Pontarcy,
extends in a direction perpendicular to the course of the river.
Caesar meant of course that the hill extended in a direction parallel
with the Aisne. Moreover, as Piette observes,^ the hill of Comin
is about 200 feet above the level of the valley beneath it, and is
surrounded by steep escarpments whereas the hill which Caesar
;

describes rose only a little above the plain, and its slope was very
gentle.
Fourthly, the slope of the hill descending towards the valley
which de Saulcy believed to have separated the Romans from the
Belgae is bounded, he says, by two parallel trenches. These trenches
he identified with Caesar's transversae fossae, which, I need hardly
say, would have been filled up by the inhabitants after his departure.
The distance between them is about 450 metres, or nearly 500 yards.
This space corresponded, on de Saulcy's theory, with the width of the
Roman line of battle. There were, he assumes, 800 men in each rank,^
and, as there were six legions in order of battle, or perhaps about
24,000 men, who were probably arranged in three lines, each eight
men deep, this estimate is not very far from the truth. Now
observe that, on de Saulcy's theory, each man would have had
a space of just five-eighths of a yard, or 22J inches, to stand in !

In other words, the men would have been packed like sardines
in a box, not like Roman soldiers, who had to use their swords
and shields.
Before Napoleon's book appeared, the spot which most anti-
3.
quaries favoured was Pontavert,* where a Gallic road from Duro-
cortorum (Reims) to Bagacum (Bavay) is believed to have crossed
the river by a bridge.^ If this was the bridge which Caesar mentions,
the hill on which he encamped can only have been the plateau of
Chaudardes but this site was carefully examined by General Creuly
;

and various members of the Societe academique de Laon, and unhesi-

tatingly rejected.^ Professor Konrad Lehmann has, however,

^ Les campagnes de Jules Cesar dans les Gaules, 1862, p. 95.

^ I tin. gallo-rom. dans, le dep^ de V Aisne, p. 251.
^ Les campagnes, &c.,
pp. 98, 107.
* Bull, de la Soc. hist, et arch, de Soissons, xv,
1860, pp. 127, 129.
^ Bull, de la Soc. acad. de Laon, vii,
1858, p. 188. M. Melleville {Le passage
de r Aisne par J. Cesar, 1864, p. 19) asserts that this was the only ancient road
within the line of the Aisne from Neufchatel to Pontavert ; but A. Piette, the
principal authority on the ancient roads of the department of the Aisne, main-
tains {Itin. gallo-rom. dansledep* de V Aisne, p. 92) that the road which leads
from Reims to Arras by way of Berry-au-Bac existed in the Gallo-Roman period.
® A. E. E. Poquot, J ides Cesar, &c., pp. 69-70. Melleville {Le passage
de V Aisne, &c., pp. 21-2, 25, 27-8) placed Caesar's camp on the hill of
St. Thomas, which is at least 6 miles from the nearest point of the Aisne ! Of
the various objections that are fatal to this theory one is so obvious that I
wonder that it did not occur to Melleville himself : —how could Caesar have
said that one side of his camp was protected by the Aisne {latus unum castrorum
666 WHERE DID CAESAR
recently attempted to vindicate its claims ^ and Creuly's con- ;

demnation nmst not be accepted as final. Still, when I went over

the ground I noted one or two objections. Although definitely
sloping flanks (lateris delectus) are there, the plateau did not appear
to me to correspond in other respects with Caesar's description as
closely as the hill of Mauchamp. The western end, where his left
wing would have been posted, does not gradually merge in the'

plain by a gentle slope (in fronte leniter fastigatus faulatim ad

'

flanitiem redibat),^ but is actually lower than the ground immediately

north of it which would have been occupied by the Belgae the ;

northward slope of the central and eastern parts of the plateau is

perhaps rather too marked and I did not feel sure that the marsh
;

would have been suitable for cavalry. Lehmann,"^ affirming that

the plateau was exactly wide enough to admit of the formation of
the Koman line, assigns 1,100 metres for the aggregate breadth
of the cohorts and 600 for the intervals between them and for the
cavalry. But he can only allow a front of one metre, or about
3 feet 3 inches, for each man, which, as Polybius'* and common
sense alike show, was little more than half of the necessary room.
StofEel^ shows that one legion of only 3,600 men would have occupied
a front of 348 metres and the intervals between the legions have
;

to be allowed for. This objection can only be removed by assuming

that Caesar's line was unusually deep. Finally, Piette ^ argues
that Caesar would not have attempted to cross the Aisne at
Pontavert, as to reach it he would have had to traverse the broken
country which stretches north-westward from Keims. But if there
was a road,'^ does not this objection disappear ?
The choice lies between Chaudardes and Mauchamp. The topo-
graphy of the latter, with the very important exception of (lateris)
delectus, conforms perhaps somewhat more closely to Caesar's
description but the results of StofEel's excavations cannot be
;

reconciled with Caesar's text except by Rutherford's forced inter-

])retation. Let the advocates of Chaudardes excavate in their
turn.
II. One or two minor points remain to be settled. Piette says
that Caesar probably marched by the route along which subsequently
ran the Roman road from Reims to Bavay, and which crossed the
Aisne about 4 kilometres east of the confluence of the Suippe and
the Aisne that he left Sabinus in the so-called camp of Conde ',
;
'

close to the confluence and that he encamped himself on the hill

;

ripis fluminis muniehat) if the camp had been 6 miles from that river? Craomie
has also been suggested ; but Thillois {Bull, de la Soc. acad. de Laon, xix,
1869-70, p. 272) points out that it is scarped on all sides, and that it is
6 kilometres, or nearly 4 miles, from the river.
^ Neue Jahrb. f. d. Mass. Altertum, iv, 1901, pp. 50G-9 ; Klio, vi, 1906,
pp. 237-48.
^ It will be observed that I take in fronte in its natural sense, not in that

which Rutherford attributes to it. ^ Klio, vi, 1906, p. 240.

* xviii, 30
(13), §§ (j-S. Guerre civile, ii, 327-8.
'•

^ Itineraires, &c., pp. 97-8. tSce JSheet 34 of the Carle de V Elat-Major

(1 : 80,000). ' See p. 665.
 —
^
-

CROSS THE AISNE ? 667

of Maui'liainp, that is to say, in the position selected by Napoleon.

But whether Caesar encamped on the hill of Mauchamp or not,
Piette is For the camp of Conde, if it is a
certainly mistaken.
Roman camp at all, is far too large to have been occupied by the
small force of Sabinus ^ it is tolerably clear from Caesar's narrative
:

that he encamped just north of the point where he crossed the Aisne,
not several miles to the west of it, and that the camp of Sabinus
was immediately south of and opposite to the tete de font, not
4 kilometres to the west of it if we may believe Napoleon,
: the '

entrenchments of this tcte de pont are still visible at Berry

. . .

;
au-Bac ^ and, as Piette believes that the road leading from
'

Reims to Arras by way of Berry-au-Bac existed in the Callo-

Roman period, I cannot see why he refuses to believe that Caesar
marched by it.
III. Referring to the tHe de pont which is shown in Napoleon's
Plan, Long says, this tete-de-pont would be of little or no use
'

on the north side of the river and if the " praesidium " of c. 5
;

and the " castellum " of c. 9 are the same, as I think they are,
Caesar placed the tete-de-pont on the south side of the Aisne.' ^
This note is unworthy of Long. How could he have thought
that the praesidium of chapter 5 and the castellum, which Sabinus
occupied, of chapter 9 were the same, after reading this passage :

In eo flumine pons erat. Ihi praesidium ponit et in altera parte

jiuminis Q. Titurium Sahinum legatum cum sex coJiortibus relinquit ? ^
These words immediately follow Caesar's description of the position
which he himself took up on the northern side of the Aisne. Altera
parte Jiuminis, then, must mean the southern side and the prae- ;

sidium must have been on the northern. Besides, from chapter 9

it is clear that the castellum was on the southern side.^ Therefore,
as the praesidium was on the north, the castellum and the praesidium
were not the same.
Of course Long made a great mistake when he asserted that
'
this tete-de-pont would be of little or no use on the north side
of the river '. The tete de pont made it absolutely impossible for any
hostile force to cross the river by the bridge.
IV. According to Napoleon and von Kampen, the Belgae
attempted to cross the Aisne by the ford of Gernicourt, between
Berry-au-Bac and Pontavert but Piette thinks that they marched
;

to attack Sabinus by the north-east, concealed themselves behind

the hill of Proviseux, and then gained the ford of St. Pierre, near
Guignicourt.'^ The map shows that the route indicated by Napoleon
^Bull, de la Soc. acad. de Laon, vii, 1858, p. 188.
^See A. E. Poquet, Jules Cesar, &o., pp. 28-9.
^ Hist, de Jules Cesar, ii, 100, n. 1.

* Decline
of the Roman Republic, iv, 51, n. 6.
^ The river was spanned by a bridge, at the head of which he established
'

a strong post, while on the other side of the river he left six cohorts under one
of his generals, Quintus Titurius Sabinus.' B. 0., ii, 5, § 0.
^ Ibi vadi.s repertis partem suarum copiariun traducere conati sunt, uo
consiho ut, si posscnt, castellum . oxpugnarcnt, »&c.
. .

^ Itin. (jallu-rom. dans le dip' de C Aisne, pp. 100-1.

668 WHERE DID CAESAR CROSS THE AISNE ?

was the easier but of course Piette was obliged to choose the
;

other because he had made the mistake of placing Sabinus's camp,

which the Belgae intended to attack, at Conde-sur-Suippe. And
if Caesar's camp was on the plateau of Chaudardes, they could only

have crossed by the ford of Gernicourt.

HOW LONG DID CAESAR TAKE TO MARCH FROM

HIS CAMP ON THE AISNE TO NOyiODUNUM?
Caesar's words are Postridie eius diei Caesar . . . in fines Sues-
sionum, qui proximi Remis erant, exercitum duxit et magno itinere
[confecto] ad oppidum Noviodunum contendit.^ The question is
whether he reached Noviodunum on the day of the magnum iter
or on the next day.
Nipperdey^ says that the words magno itinere confecto would be
absurd unless some place were indicated as the terminus of the mag-
num iter and Schneider ^ virtually takes the same view, remarking
;

that in other passages where Caesar uses the words itinere confecto,
the general who performed the march is said to have reached some
definite place, or to have fallen upon his enemy.* But as, in this
particular passage, no place is, in Schneider's opinion, expressly
indicated as the terminus of the iter, he concludes that, after confecto,
we must understand in fines Suessionum pervenit that is to say, ;

that Caesar reached the territory of the Suessiones by a forced

march, and then marched on for Noviodunum. But, to say nothing
of the fact that Caesar had already told us that he had entered
the territory of the Suessiones (in fines Suessionum exercitum . . .

duxit), the point from which he started was close to that territory ;

and therefore, on Schneider's theory, the word parvo would be more

applicable than magno. We are to assume, then, says Nipperdey,
who waxes very sarcastic at his brother editor's expense, that
Caesar marched about to amuse himself, and deliberately wasted
his time for want of something better to do {scilicet animi causa
ambulavisse Caesarem putabimus et de industria tempus trivisse in
summa rerum gerendarum inopia). Nipperdey concludes that the
word confecto is spurious, and that Caesar reached Noviodunum in
a single march. Long,^ on the other hand, while admitting with
Schneider that contendere ad properly means to march towards
'

a place ', suggests that it does not seem to exclude the notion
'

of reaching it also'. No it does not exclude that notion; but

!

in order that Long's suggestion should have any weight, it would

be necessary that contendit should in this passage mean the same
as pervenit. Of course, if this were so, the difficulty presented by
the suspected word confecto would disappear and the passage ;

1 B. G., ii, 12, § 1. 2 Caesar, pp. GO-1. = Caesar, i, 148-9.

* B. G., iv, 4, § 5 ; 14, § 1 ; vi, 30, § I ; vii, 50, § 3 ; 83, § 7.
^ Caesar, p. 119.
 CAESAR'S MARCH TO NOVIODUNUM 669

would mean Caesar led his army into the country of the Suessiones,
'

and, after making a forced march, arrived at Noviodunum '. But

contendit cannot mean the same as pervenit.^
Still, Nipperdey's argument involves one difficulty. Caesar began
to attack Noviodunum immediately after his arrival, and, having
failed in an attempt to take it by storm, proceeded on the same
evening to make preparations for a regular siege. Could he, it
might be asked, have done all this if he had already marched 28, or
even 24 miles ^ on the same day ? It may, however, be replied
that, as Schneider says, the word fostridie proves that Caesar
hastened to Noviodunum {ad oppidum Noviodunum contendit) on
the same day on which he made the magnum iter. Similarly Dr.
W. G. Rutherford,^ who also retains the word confecto, translates
the passage thus, the long march accomplished, he hastens to
'

Noviodunum.' It is astonishing that neither of the two editors

should have seen that Caesar could not have said that, after having
made a forced m^arch (magno itinere confecto), he hastened on the ' '

same day to Noviodunum seeing that the hastening would itself

;

have formed a part of the forced march, and would have changed
the magnum iter into a maximum iter.
But another explanation has been proposed. Perhaps,' remarks '

Mr. Peskett,^ as Vielhaber suggests, this word [confecto^ is to be

'

explained by the consideration that we occasionally find ahlativi

''

consequentiae formed with the participle perfect to express a cir-

cumstance which does not precede but accompanies or follows the
main action ", Madvig, L. G., § 431, obs. 2 in the present passage :

the forced march is not prior to his hastening to Noviodunum but

synchronous with it. " Caesar hastened to Noviodunum, making
a forced march." If Caesar reached Noviodunum on the same
'

—
day on which he made the magnum iter and, considering the force
of postridie, we must admit that he did —
then either we must accept
Vielhaber's explanation or with Nipperdey and Meusel we must
expunge confecto. Whoever is familiar with Caesar's language or
will consult the relevant passages in Meusel's lexicon ^ will conclude,
without any hesitation that Vielhaber's suggestion is inadmissible.
But there remains the difficulty of understanding how, at the end
of a magnum iter, Caesar's troops could have attempted to take
Noviodunum by storm, and then begun their preparations for
a regular siege.

^ In the first edition I put this question Supposing that Caesar had
:
'

wished to say that he made a forced march, which did not take him the whole
way to Noviodunum, and that on the following day he pushed on for Novio-
dunum, would not the words magno itinere confecto ad oppidum Noviodunum
contendit have adequately expressed his meaning ? And how was he to indicate
the terminus of the iter, if it had no name ? I now see that, in the case
'

which I supposed, Caesar would have written (magno itinere confecto) proximo
(or altero or postero) die ad oppidum Noviodunum contendit.
2 The march would have been shortened
by about 4 miles if Caesar encamped
near Pontavert. » Q^^sar,
p. 59.
* Caesar, B. 0., ii, 1888, Notes, pp. 26-7.
^ Lex. Caes., \, 640-2.
670 CAESAR'S MARCH TO NOVIODUNUM
[Since I wrote the rough draft of this note, I have remembered
that during the Indian Mutiny, Colonel Greathed's column fought
the battle of Agra almost immediately after making a forced march
of 44 miles, and that the famous corps of Guides went into action
at Delhi two hours after they had finished their wonderful march of
580 miles in 22 days. See my History of the Indian Mutiny, 5th ed.,
1898 (or the reprint of 1904),' pp. 339, 392-3.]
The following suggestion may be worth considering. Is it certain
that the starting-point of the forced march was Caesar's original
camp on the Aisne ? Immediately before describing the march he
says that towards sunset the force which he had detached in
'
'

pursuit of the Belgae stopped and returned, in obedience to instruc-

'

tions, to camp {suh occasum solis sequi destiterimt seque in castra,

'

ut erat imperatum, receferunt).^ Consider what is involved in the

assumption that they returned to the original camp. If their
pursuit had been directed down the valley of the Aisne, they would
have been obliged, after an extraordinarily long day's work and
immediately before another prodigious march, to return the whole
distance which they had covered between dawn and sunset while;

Caesar, although his object was to reach Noviodunum as soon as

possible, would needlessly have imposed this heavy labour upon
them and imposed both upon them and the rest of the army a march
of nearly twice the ordinary length on the next day. Is it not
more likely that, instead of sitting idle in camp while the detach-
ment was hard at work, he would have marched down the valley
to within a short distance of the frontier of the Suessiones ?
But why did he order the detachment to return at all ? If the
pursuit was directed down the valley, I cannot answer this question.
If, on the other hand, the Suessiones fled, as I have before suggested,

down the road towards Laon,^ the answer is easy. While Caesar
moved a few miles down the valley and thus shortened the inevitable
magnum iter, the detachment would have rejoined him by a road
leading to Pontavert or by a road leading to Beaurieux. M. JulHan,*
however, holds that the various Belgic contingents dispersed at the
beginning of their flight, the Bellovaci and the Suessiones going
down the valley and all the others towards Laon. The former, he
believes, marched by the chemin des Dames ', on the heights
'

parallel with the right bank of the river. But if so, how could they
have failed to detect that Caesar in the valley below was overtaking
and outstripping them, and why should they have allowed him
to do so ?

1 B. G., ii, 11, § 6. 2 See

pp. 658-9.
^ General Creuly {Bull, de la Soc. acad. de Laon, xv, 1865, p. 12) conjectures
that the Suessiones and the other contingents retreated to La Fere, where
(see p. 659, supra) he believes that they had concentrated before they began
their final march against Caesar; but M. Jullian [Hist, de la Gaule, iii, 257,
n. 5) rightly objects that the Suessiones would not have had time to make so
long a circuit. * 76., 256, n. 2.
 ^ ;

671

THE BATTLE WITH THE NERVII

1. 1. I have followed Creuly and Bertrand,! who carefully examined
the course of the Sambre, and on whose notes Napoleon evidently
relied, in placing this battle on the left bank, on the heights
of Neuf-Mesnil, opposite Hautmont and before the publication of
;

Napoleon's book von Goler ^ had selected the same site. Cette '

position,' says Desjardins, convient en effet assez bien, mais ce

'

n'est pas la seule du cours superieur de la Sambre qui presente

les conditions exigees par le texte nous n'avons done pas plus de
;

certitude a cet egard que pour le reste, le texte de Cesar etant,

comme tou jours, d'une desesperante sobriete de renseignements
topographiques.' ^ But Caesar's description, the substance of which
I have embodied in pages 76-79 of my narrative, is perfectly clear
and, besides Creuly and Bertrand, Long and Camille Jullian give
reasons for believing that the site adopted by von Goler is the
only one which exactly corresponds with it. Let us, however, hear
what other commentators say.
2. J. Des Roches * and V. Gantier ^ find the battle-field at Presles.
This place, which is about 8 kilometres, or 5 miles, east of Charleroi,
is on the right bank of the Sambre, not, like Napoleon's site, on

the left. Gantier maintains that if Caesar had marched against the
Nervii by the left bank, he must have crossed the country of
the Atrebates before entering that of the Nervii, which would be
contrary to his statement ^ that the country of the Nervii, where he
entered it, bordered on that of the Ambiani. Furthermore, says
Gantier, the western frontier of the Nervii was probably the Scheldt,
not, as Napoleon, who is anxious to allow space enough for a march
of three days across the Nervian territory to Neuf-Mesnil, main-
"^

tains, close to Fin and Bapaume. Hautmont, he continues, is only

50 kilometres from the Scheldt even from Bapaume it is only 80
:
;

and this distance is not enough to answer the requirements of

Caesar's narrative. Presles satisfies all the conditions tradition is :

in favour of Presles and at Presles, as at Hautmont, antiquarian

;

discoveries have been made. Dewez adds that the name Presles ', '

or, as he spells it, Prele,' is derived from froelium

'
and that ;

human bones have been found on the spot. There is no record,

he says, of a later battle having been fought there and we may ;

therefore conclude that the bones are those of men who were killed
in the battle with the Nervii. Finally, Des Roches says that the
lie of the ground at Presles and the depth of the river correspond
with Caesar's description.

'
Rev. arch., 2« ser., iv, 1861, pp. 453-67.
2 Gall. Krieg, 1880, pp. 75-7.
^ Geogr. de la Gaule rom., ii, 635-6.
* Hist. anc. des Pays-Bas autricMens, ii, 1787, pp. 37-40.
^ La conquete de la Belgique par Jules Cesar, 1882, pp. 113-4, 156-04.
« B.G.,n,l5,%2. ' lb., 16, § 1.
* No^iv. mem. de V Acad. Roy. de Bruxelles, ii, 1822, pp. 238-9.
 672 THE BATTLE WITH THE NERVII
These argumentswill not bear examination. It is quite possible
i
that, as d'Anville argued, Caesar's statement that the country of
the Nervii bordered on that of the Ambiani is not to be taken literally
when he was marching from Samarobriva (Amiens) into the country
of the Nervii in 54 B.C. he did cross the country of the Atrebates ^ ;

and, moreover, on Gantier's theory, he must have crossed the

country of the Viromandui before entering that of the Nervii.
Finally, it is strange that Gantier should have overlooked the fact,
directly pointed out by Caesar,^ that the legions could not have
marched at their ordinary speed across a country intersected by
thick hedges which had to be cut through.
The arguments of Des Roches and Dewez are not more convincing
than those of Gantier. If the lie of the ground at Presles corresponds
with Caesar's description, so does the lie of the ground at Neuf-
Mesnil the name Presles ' is common to eight places in France
:
'

alone, and,asDinaux sensibly remarks, 'nous serious bien embarrasses

de leur assigner huit champs de bataille ' and the argument drawn
;
'*

from the alleged discovery of bones has been so often used by rival
antiquaries on behalf of their pet sites that cautious investigators
have come to regard it with suspicion. Besides, ancient weapons,
human remains, and rows of graves have also been discovered in the
wood of Quesnoy, near Hautmont.^
3. who observes that Caesar did not mention
Colonel A. Sarrette,^
Samarobriva in his account of this campaign, and rashly concludes
that he did not pass through it, takes him from Breteuil (which he
identifies with Bratuspantium) direct to Peronne and thence to
Wassigny there he makes him turn sharply to the right, round the
:

headwaters of the Sambre, advance to Avesnes, and thence move on

to Hautmont. In other words, he places the battle-field at the same
point of the Sambre as von Goler and Napoleon, but on the opposite
bank. Caesar was obliged, he says, to march down the right bank,
for fear his communications with the Remi, from whom he drew his
supplies, should be cut.
I Remi by the Sambre, Caesar
reply that, separated from the
would have been no more danger than he was when he penetrated
in
into the hostile country of Ariovistus, when he invaded the country
of the Veneti, when he marched into the heart of Britain, when, in
the seventh campaign, he again and again ran risks about supply to
gain a great end or than his lieutenant, Crassus, was when he
;

penetrated far into hostile Aquitania finally, that, on Colonel

;

Sarrette's theory, his convoys would equally have been exposed to

attack from the Viromandui. Besides, as General Creuly asks, what
business had Caesar on the right bank of the Sambre ? Why should
he have gone out of his way to avoid the country which he wished

^ Notice de Cancienne Gaule, p. 62. ^ B. G., v, 46, § 3.

^ lb., ii, 17, §§ 4-5.
* Archives hist, et litt. du Nord de la France et du Midi de la Belgique, S** ser.,
iii, 1852, p. 187.
^ Von Goler, Gall. Krieg,
1880, p. 86.
^ Quelques pages des comm. de Cesar, 1863, pp. 102-4 and map facing p. 75.
 ;

THE BATTLE WITH THE NERVll 67^

to invade ? His objective was probably Bagacum (Bavay), the

Nervian capital,^ which lay on the direct road from Samarobriva to
the Sambre. Von Moltke, in 1870, might as well have marched for
Marseilles or Toulouse, instead of Paris.
on the
4. V. Ganchez ^ finds the battle-field near Hantes-Wiheries,
rightbank of the Sambre, about 20 kilometres, or 12 miles, below
Hautmont. He bases his choice on the hypothesis that Caesar
marched 20 Roman miles a day and in justification of this hypo- ;

thesis he refers to B.G., v,47-8. tells us that in 54 B.C.,

Caesar there
on an occasion when everything depended upon speed,^ he marched
20 miles in one day through the territory of the Nervii. So we are to
assume that in 57 B.C., when he had no motive for making haste, he
marched 60 miles in three days !

5. Nicholas Le Long points to Landrecies, but gives no reasons.*

Nor can any be given for the country near Landrecies does not
;

correspond with Caesar's description.

6. A. de Vlaminck,^ following P. G. Baert,^ decides for the village
of La Buissiere, about 7 kilometres, or 4J miles, south-west of Thuin.
There is nothing to be said for this site or for Thuin itself, where the
battle has also been placed ' and^ as Thuin is about 14 and La
;

Buissiere about 13 miles further down the Sambre than Neuf-Mesnil,

it is unlikely that Caesar could have marched to either place, across
such a difficult country, in the time which he mentions.
7. L. Caudet makes Caesar march, after the submission of Samaro-
briva, to Vermand, establish there a magazine, and then begin
his three days' march fer Nerviorum fines, along the right bank
of the Oise. The battle took place, according to Caudet, at Catillon-
sur- Sambre .8
All this simply wild imagination. Caudet ignorantly translates
is

per fines by along the frontier ', whereas of course fines means, in
'

this as in almost every other passage where it occurs in Caesar,

'
territory,' and per means not along but through '.^ ' ' '

8. Achaintre,^^ Le Glay,ii and others believe that Caesar wrote

Sabim by mistake for Scaldem, and that the battle was fought on the
banks of the Scheldt. Achaintre argues, first, that Caesar, marching
from the country of the Ambiani against the Nervii, would have
traversed the district in which are situated Cambrai, Bouchain, and
Valenciennes, and thus would necessarily have come to the Scheldt
secondly, that it is absurd to suppose that, if he had marched to
^
Rev. arch., nouv. ser., viii, 1863, p. 36.
^
Annates de V Acad, d'arch. de Belgique, 3" ser., viii, 1882, pp. 422-7.
^
unum salutis auxilium in celeritate ponebat.
*
Hist. eccl. et civ. du diocese de Laon, 1783, p. 12.
^
Messager des sciences hist, de Belgique, 1882, p. 385.
"
Mem. sur les campagnes de Cesar dans la Belgique, 1833, p. 58.
^
Archives hist, ellitt. du Nordde la France et du Midi de la Belgiqite, 3® ser.,
V, 1856, p. 326 ; N. Lc Long, Hist. eccl. et civ. du diocese de Laon, p. 12.
^
Cotnptes rendus et mem. du Comite arch, de jSenlis, 2^ ser., iv, 1878, pp. 28-9,
34.
«

^»
Sec B, o., i,

Caesar, i, 82.
6, H ; y, § ^ ; 11, § 1 ; 10, § l ; 28, § i, &c.

^^ Mem. de la ISoc. d' emulation de Camhrai, 1830, p. 87.

1093 XX
674 THE BATTLE WITH THE NERVII
Neuf-Mesnil, the Nervii would have abandoned the left bank of the
Sambre and waited for him on the other side ^ and thirdly, that in
;

B. G., vi, 33, § 3, Caesar confounds the Sambre with the Scheldt.
It is difficult to say which of these reasons is the worst. To argue
that Caesar, marching from the country of the Ambiani into that of
the Nervii, '
would necessarily come to the Scheldt and not the '

Sambre, is about on a par with arguing that an invader having

effected a landing in England, would necessarily
'
come to the
'

Severn and not the Thames. The second reason is not much better.
The Nervii chose the strongest position that th<ey could find. If
Caesar had determined to attack it, he would first have had to cross
a river. They attacked him from ambush and it was all that he
;

could do to avoid defeat. As for the third reason, there is no proof

that Caesar ever confounded the Sambre with the Scheldt ^ and ;

it is not credible that he should have mistaken the name of a river

which was the scene of one of his most memorable exploits.
Le Glay,^ who finds the battle-field on the hill of Bona vis, opposite
Vaucelles, reasons no better than Achaintre. He says that the
Nervii, being a brave people, would have defended the threshold of
their territory. The answer is that they did not defend the threshold
of their territory for Caesar says that, when he was still 10 Roman
;

miles from the battle-field, he had already marched for three days
across their territory. In order to get over the difficulty presented
by the extreme nearness of the site which he selects to Caesar's
starting-point, Le Glay asserts that the Romans marched with ex-
treme slowness in order to avoid being surprised. No doubt they
marched more slowly than usual, owing to the difficulty of the
country ; but I will not believe that they crawled, like old ladies
out for an airing.
It has been argued, as Isidore Lebeau remarks, that the Sambre,
'^

in that part of its course near which the battle must have taken
place, if it took place on the Sambre at all, was not wide {latissi-
mum) and that there were no marshes in the country of the Nervii
;

such as those in which Caesar says that the non-combatants took

refuge, except near the sea-coast. But Lebeau replies that, in Caesar's
time, when the climate was much damper than it is now, the country
in the neighbourhood of the Sambre must have been constantly
flooded, and that the epithet latissmimn has a purely relative significa-
tion. This, however, is no answer to the one reasonable argument
which has been brought against the theory that the battle was fought
on the Sambre. For Caesar does not only say that the non-com-
batants took refuge in marshes he also says that they took refuge
:

in aestuaria ; and there is no evidence that the word aestuaria can

be used of marshes formed by a river which does not flow into the

'
So also A. Eberz in Neue Jahrh. f. Philoloijie, &c., Ixxxv, 18G2, p. 221.
^ See pp. 734-5. Since my first edition was printed the Sabis has actually
been identified with the Selle ! See C. Jullian, Hidt. de la Gaule, iii, 261, n. 2.
^ Mem. de la Soc. d' emulation de Camhrai,
1829, pp. 87, 92-3.
* Archives hist, et litt. du Nord de la France et
du Midi de la Belgique, 3° ser.,
V, 1856, pp. 315-7.
 THE BATTLE WITH THE NERVII 675

sea. It must therefore, I think, be admitted that the aestuaria to

which Caesar were formed by the tide. But there is nothing
refers
incredible in the hypothesis that the Nervii sent some of their non-
combatants to a place of refuge in the remotest parts of their country,
while they themselves prepared to defend the all-important line of
the Sambre ^ and it is incredible that they abandoned the basin
;

of the Sambre, which was the heart of their territory.

The battle, then, was undoubtedly fought somewhere on the left
bank of the upper Sambre. But is there any proof that it was fought
opposite Hautmont ? Long offers a most ingenious and, I think,
almost conclusive proof. The banks of the river,' he says,^ opposite
' '

to the enemy's left were, as Caesar describes them (c. 27), very high,
a statement which is the strongest proof that the site of this great
battle has been truly determined. The heights of Neuf-Mesnil . . .

descend to the river with a uniform slope but at Boussieres, a ;

little farther up the stream, the heights which are connected with
Neuf-Mesnil terminate on the river in escarpments from sixteen to
about fifty feet high, which are not accessible at Boussieres, but may
be scaled lower down. The bank of the river on the right side
opposite to Boussieres is flat.' ^ I should add that, as M. Jullian *
acutely observes, the battle must have been fought at a place where
Caesar would have been obliged to cross the Sambre, for the Nervii
were awaiting him and at Maubeuge, which has always been the
;

strategical point of the river, the valley is crossed by the old Roman
road from Bavay to Reims. To every site that has been proposed,
except the site opposite Hautmont, the objections are fatal. There
is no objection worth considering to the latter, except that, accord-
ing to Caesar, the Sambre opposite the battle-field was only three feet
deep, whereas the depth opposite Hautmont is much more. But the
Sambre at Hautmont is deeper now than it was in Caesar's time,
because it has been canalized.^ The site may therefore be regarded
as fixed beyond all reasonable doubt.
n. Li his description of the battle, Caesar states that he could
not see all the legions at once, because thick hedges interrupted the
view (saepibusque densissimis, ut ante demonstravimus, interiectis pro-
spectus impediretur ^). A. Eberz contends that this passage is either
a gloss or a fiction. Caesar's own words, he says, prove that there
were no hedges on the battle-field. He was not stopped by hedges
when he moved from the 10th legion to the right wing nor were ;

the 9th and 10th legions when they drove the Atrebates down to and
across the Sambre.'''
Now any one who examines the Plan of the battle-field may con-
vince himself that the statement in the Commentaries is not necessarily

^
I find that General Creuly has a similar argument in Bev. arch., nouv. ser.,
viii, 1863, p. 36.
^ Decline of the Roman Republic, iv, 59.
*I do not know whether Long wrote from personal observation or followed
von Goler {Gall. Krieg, 1880, p. 90).
* Hist, de la Gaule, iii, 261, n. 2. '
Reo. arch., nouv. ser., iv, 1861, p. 457.
^ B. G., a, 22, § 1. ^ Neue Juhrb. /.Philologie, &c., Ixxxv, 1862, p. 221.
X X 2
676 THE BATTLE WITH THE NERVII
false. there was a hedge in Caesar's way, as he was moving from
If
one part of the field to another, he might have turned it or, if that ;

was impossible, a few sappers could have cut a passage for him in
a trice ;and hedges might have run in such a way as to intercept
his view and yet not to hinder the free passage of troops moving
down or up the slope, to or from the Sanibre. It is just possible that
the passage of which Eberz complains may be a gloss but I do not :

believe that it is a fiction for I cannot see that Caesar could have
;

had any motive for telling an untruth. The rest of his narrative
made it clear that he had allowed himself to be surprised, and that,
hedges" or no hedges, the difficulties which he had brought upon
himself were serious enough. He would not have exaggerated his
difficulties by making a statement the falsity of which would have
been noticed at once by every officer who had served under him.
in. According to von Goler, Napoleon, and von Kampen, the
lOtli legion stood on the extreme left of the Roman line. There is
no direct evidence for this view and M. Crain ^ adduces an argu-
;

ment against it. He points out that in chapter 23 Caesar mentions

the six legions in three pairs, the 9th and 10th, the 11th and 8th,
the 12th and 7th ;and he draws attention to the fact that in the
first pair the smaller number takes precedence, in the other two the
larger. Now it is clear from chapters 25 and 26 that the 7th legion
was on the extreme right and it is therefore probable that the 9th
;

was on the extreme left.^

IV. On page 79 of my narrative I have written Caesar told '

the tribunes to bring the two legions gradually closer together, and
form them up so as to face the enemy on every side '. The passage
on which this sentence is based runs as follows Caesar, cum VII.
:

legionem, quae iuxta constiterat, item urgeri ah hoste vidisset, tribunos

militum monuit ut paulatim sese legiones coniungerent el conversa slgna
in hostes inferrent. Quo facto cum aliis alii suhsidimn fervent neque
timerent ne aversi ah hoste circumvenirentur, audacius resistere ac
fortius pugnare coeperunt? The exact meaning of the words conversa
signa in hostes inferrent is doubtful. In B. G., ii, 24, § 4 Caesar says
that, before he ordered the manoeuvre which I have briefly described,
the 7th and 12th legions were almost surrounded {paene circum-
ventas), and in 25, § 1, that the Nervii were attacking the 12th legion
in front and on either flank (hostes neque a fronte ex inferiore loco
suheuntes intermittere et ah utroque latere instare). It seems certain,
therefore, that the object of the formation Avhich he describes was
to enable the two legions to face the enemy on all sides. They
certainly had to repel attacks in front and on either flank and ;

Caesar says that when they had effected the movement which he
ordered, they no longer feared an attack in the rear. Roesch gives
an ingenious description, which Long* reproduces with a plan, of
^ Zeitschriftf. d. Gymnasialwcaen, ISOO, pp. 485-G.
''
It is true that in the battle of Pharwaha the 10th was on the extreme left
{B. C, iii, 89, § 1) ; but that docs not prove that it occupied the same position
in this disorderly fight.
3 B. G., ii,
26, §§ 1-2. * Caedur, pp. 144-5.
 THE BATTLE WITH THE NERVII 677

a complicated evolution by which the movement may have been

effected. Frohlich ^ supposes that the 7th legion moved to the rear
of the 12th, and stood back to back with it. Giesing'^ ridicules this
conjecture. This going one behind the other,' he says,
'
of two '

legions, which were hemmed in, in front and on the flanks, by a

victorious enemy, is quite impossible.' It was only necessary, he
adds, for the rear companies of the two legions to turn round and
for the wing companies to make a quarter-turn, thus forming a closed
parallelogram. I believe that Giesing is right but it is enough to
;

get a clear general idea of Caesar's meaning.

V. The battle must have been as brief as it was eventful. The
Nervii began their charge when they saw the head of the approach-
ing Roman baggage-train, which immediately followed the legions
that were already on the ground the decisive blow was struck by
:

the two legions which had immediately followed the train and ;

they marched up at their utmost speed to join in the action .^

THE ASSAULT ON GALEA'S CAMP AT OCTODURUS

I. Everybody admits that Octodurus was between Martigny-la-
Ville and Martigny-Bourg but the topographical details of the
;

struggle are disputed. De Saulcy maintains that Galba encamped

on the right, or eastern, bank of the Dranse, and the Gauls on the
left first, because on the right bank, but not on the left, the Romans
;

would have been near the timber which they required for the camp ;

secondly, because on the right bank Galba would have cut the com-
munications of the Gauls with the rest of the Veragri and with the
Seduni thirdly, because on the right bank, he would probably have
;

been on the road which led over the Pennine Alps and fourthly,
;

because the Gauls could not have attacked him from the steep heights
that overlook the left bank.*
These reasons have no weight. The mere advantage of being
a few hundred yards nearer timber would not have led Galba to
encamp on the right bank unless it had been advisable, on more
important grounds, to do so. Besides, there is timber on the left
bank now (1893), south of the vineyards and why should there not
;

have been then? If on the right bank Galba would have cut the
communications of the Gauls with the rest of the Veragri and with
—
the Seduni and a glance at the map will show that he could not
—
have done so without quitting his camp the Gauls, on the left
bank, would have effectually cut his communications with the two
cohorts which he had left among the Nantuates and in the event
;

of an attack he would have been compelled to cross the Dranse in

the face of an enemy, or else to surrender. Finally, the Gauls could

^ Das Kriegswesen Cdsars',

p. 183, with which cf. A. von Goler, Qall. Krieg^
1880, p. 84.
2 Neue Jahrb.
f. Philologie, &c., cxlv, 1892, p. 495.
3 B. 0., ii, 19, 26-7 * Rev. arch., nouv. ser., iv, 18C1,
pp. 4-7.
678 THE ASSAULT OF OCTODURUS
have attacked him from the heights overlooking the left bank.
I arrived at this conclusion after examining Blatt 526 of the Topo-
graphischer Atlas der Schweiz (1 50,000) and my conclusion was
: ;

confirmed by a visit to Martigny. It is true that for a space of

about 1,000 yards leading southward up the valley of the Dranse
from the so-called Roman tower, the slope of the hill is very steep,
and, at certain points, too steep and too rocky to be rapidly de-
scended by armed men but south of this space the gradient is no
:

steeper than that of the opposite mountain while in the immediate;

vicinity of the tower there is a slope which can be. easily descended.
I think it probable, however, that men were posted on the eastern
as well as on the western hills, in order to cut off the Romans from
all possibility of escape and in fact Caesar says that almost all the
;

high ground which dominated the valley was occupied (omnia fere
superiora loca multitudine armatorum completa conspicerentur).^ Des-
jardins holds that Galba must have compelled the Veragri to encamp
on the right bank not only of the Dranse, but also of the Rhone
"'
;

but Caesar says nothing about this. When he says that Galba left
one bank of the river, that is of the Dranse, for the Veragri, he doubt-
less means only the Veragri who inhabited Octodurus.
II. Caesar says that Galba destroyed 10,000 of the enemy .^ De
Saulcy 4 and Desjardins ^ with good reason declare that this is
a gross exaggeration. The former remarks that, as the entire popula-
tion of the country occupied by the Nantuates, Veragri, Seduni, and
Viberi amounted, at the time when he wrote (1861), to no more than
81,559 souls, the Seduni and Veragri could not, in Caesar's time,
have numbered more than 40,000 and that, deducting women and
;

children, they could not have put more than 10,000 men into the
field. Galba, says Desjardins, must have misled Caesar in his report.
I should say that he was also misled himself.^

TO WHAT TRIBE WAS TERRASIDIUS SENT IN

56 B.C.?

Speaking of the envoys whom

Crassus sent in 56 B.C. to make
requisitions of corn, Caesar says, according to the a MSS., quo in
numero est T. Terrasidius missus in Esuhios, M. Trebius Gallus in
Curiosolitas, Q. Velanius cum T. Silio in Venetos? Instead of Esuhios
the p MSS. have Unellos Sesuvios. Napoleon ^ reads Unellos simply,
on the ground that the geographical position of the Unelli (or rather
Venelli), who dwelt in the Cotentin, agrees better with Caesar's
narrative than that of the Esuvii, who probably dwelt in the depart-
ment of the Orne. I cannot understand this argument, unless
Napoleon means that the Venelli played a prominent part in the
^
B. G., iii, 3, § 2, ^ Geogr. de la Gaule rom., ii, 610.
^ B. G., iii, 6, § 2. * Rev. arch., nouv. ser., iv, 1861, p. 8.
^ Geogr. de la Gaide rom., ii, 610. ^ See
p. 242, n. 10, and p. 342.
7 B. G., iii, 7, § 4. « Hist, de Jules Cesar, ii, 121, n. 1.
 THE MISSION OF TERRASIDIUS 679

campaign, and that the Esiivii did not. Heller points out that the
reading of /5 cannot be right, first because it is not to be believed that
Terrasidius was sent to two peoples, and secondly because Unellos
Sesuvios, without a conjunction, is nonsense. It is unlikely, he con-
tinues, thatany copyist would have added (S)esuvios, as this people
are not mentioned again in the narrative of the campaign. On the
other hand, somebody might have inserted Unellos for the same
reason which I assume to have influenced Napoleon. Therefore it is
probable that Caesar wrote Esuvios. Besides, argues Heller, the
country of the Esuvii (q.v.) is a rich corn-growing district and it ;

was probably for this reason that Caesar sent Roscius and his legion
there in 54 B.C., when there was a drought in Gaul.i

THE THEATRE OF THE WAR WITH THE VENETI2

I.Caesar says that the Veneti were by far the most influential
'

of all the maritime peoples in that part of the country that all'
;

the ports on the coast, which was exposed to the full force of the
open sea, were in their hands and that almost all the tribes who
;

navigated the sea paid them toll. He goes on to say that, on hear-
ing that the Veneti had committed an act of rebellion, he ordered
ships of war to be built on the Loire and that, as soon as the season
;

was sufficiently advanced, he joined his army, which had been

cantoned in the territories of the Carnutes, the Turoni, and the
—
Andes, that is, along the valley of the Loire, from the neighbour-
hood of Orleans to the neighbourhood of Angers.^ The Veneti, he
says, assembled their fleet in Venetia and they were confident of
;

victory, because they knew that the Romans would find it much
harder to navigate the vast open Atlantic than the Mediterranean.
The roads running through their country were interrupted by estua -
ries and their strongholds were situated upon promontories and
;

tongues of land, which were insulated at high water. Caesar, as

soon as he had completed his arrangements, made his way with
his army into the country of the Veneti, ordering Decimus Brutus
to follow as soon as possible with the fleet and also with a number
of Gallic ships, borrowed from the Pictones, the Santoni, and other
peoples. He besieged and captured a considerable number of the
Venetian strongholds but, as he found that his labour was futile,
;

he determined to wait for his fleet, which had been detained by storms
and by the difficulty of navigating in a vast and open sea, where
there were hardly any harbours. When the fleet did at last arrive
and was sighted by the enemy, they sailed out of port and the ;

decisive battle, which immediately followed, was witnessed by the

Roman legions, who, from the position which they occupied on the
cliffs, had a full view of every detail of the fighting.'*

^ Philologus, xxvi, 1867, p. 667.

' Sheets 88, 89, 103, and 104 of the Carte de V Fifat-llajor (1 : 80,000) may be
consulted. ^ See
p. 709, n. 14.
* B. G., ii, 35, § 3 ; iii, 8, § 1 ; 9, §§ 1-8 ; 11, § 5 ; 12, 14, §§ 1-2, 9.
680 THEATRE OF THE WAR
n. According to Napoleon,i whose theory is founded upon a paper
written by a naval officer, the Comte de Grandprej^ the strongholds
of the Veneti which Caesar attacked were situated on the coast of
the Morbihan the Venetian fleet issued from the river Auray to
:

fight the battle the battle took place in Quiberon Bay off Point
:

St. Jacques and the Roman army was encamped meanwhile on the
;

heights of St. Gildas.

The chief opponents of Napoleon's system are MM. Sioc'han de
Kersabiec, Kcrvilcr, Blanchard, and Desjardins. They contend that
the theatre of the war was the peninsula of Guerande, which lies
between the Loire and the Vilaine and they naturally endeavour
;

to show that the territory of the Veneti extended southward as far

as the Loire.
M. de Kersabiec^ insists that, according to Caesar, the Veneti
possessed all the harbours on the coast of Brittany as far south as
the estuary of the Loire. What Caesar really says is that the sea '

being very stormy and open, with only a few scattered harbours,
which they keep under their control, they compel almost all who sail
those waters to pay toll (in rnagno mipetu maris (vasti) atque aperti,
'

faucis foriihus interiectis, quos tenent ipsi, omnesfere qui eo mari uti
consuerunt hahent vectigales).^ But it does not follow that the Veneti
possessed the seaboard between the Vilaine and the Loire. The
natural conclusion to be drawn from Caesar's statement is that they
possessed, or were able, owing to their naval strength, to blockade
harbours in territory which was not theirs. There would have been
no point in saying that they were masters of the harbours in their
own country, unless those were the only harbours on the coast to
which Caesar alludes. And if they were, de Kersabiec's argument
falls flat.
Desjardins ^ urges that writers of the early Middle Ages constantly
speak of the country of the Veneti as having extended as far south
as the Loire. But, although the Veneti, retreating before the British
invaders, may in the fifth century have taken possession of the
southern bank of the Vilaine, we have no right to assume that they
possessed it in the time of Caesar.
According to Kerviler, the name of insulae Veneticae, which, he
affirms, is applied by Pliny to the group of islands that extends from
Belle-Ile to Noirmoutier, tends to prove that the Veneti possessed
the peninsula of Guerande, which is opposite the southern islands of
the group. 6 But there is no reason for including Noirmoutier among

^ Hist, de Jules Cesar, ii, 124-5, 126, n, 2.

2 Mem. de la Soc. des ant. de France, 1820, pp. 325 ff. See also A. M.
Blancho, Guerre maritime de Cesar, &c., 1890, pp. 12-5, 17.
^ Etudes arclieol., &c., 1868,
pp. 62-3. I oinit the refutation, which I gave
in the first edition (pp. 664-5), of worthless arguments, depending upon the
assumption that Strabo, when he says (iv, 2, § 1) that the Loire discharges
between the Pictones and the Namnetes {yura^v Hiktovcov re nal Ha/^viTun'
eK^aWei), does not mean 'discharges itself (into the Ocean)', and upon
his statement about the priestesses of the so-called Samnitae (q.v.).
* B.C., iii, **
Geogr. de la Gaule rom., ii, 284-5.
8, § 1.
^ Comptes rendus de VAss""^ bretonne, 1874,
p. 46.
 —
WITH THE VENETI 681

the insulae Veneticae i and whoever looks at the map will see that, if
;

the Vilaine had been the southern boundary of the Veneti, the name
might well have been applied to the more northerly islands.
To sum up. There is no evidence that the territory of the Veneti
extended, in Caesar's time, as far south as the Loire and there is
;

evidence that it did not. Nevertheless, if Caesar was correctly in-

formed, and if he meant precisely what he said, the Veneti did con-
trol the harbours, if any existed, on the coast of Guerande. But
Caesar says distinctly that he invaded the actual territory of the
Veneti and therefore I conclude that he invaded not the peninsula
;

of Guerande, but only Venetia proper, which lay to the north of it.
Still, although I have come to this conclusion, I will examine the
arguments by which de Kersabiec and his school have tried to prove
that Caesar did not invade the country north of the Vilaine.
in. De Kersabiec ^ argues that the theatre of the war, in its first
stage, was an ancient group of quasi-insular headlands on the northern
bank of the estuary of the Loire. The modern explorer will look in
vain for these headlands, and find nothing in their place but the
plain of La Grande-Briere, which is only thinly covered even by the
floods of winter but according to de Kersabiec, the configuration
;

of the coast has completely changed. He holds that Caesar crossed

the (assumed) gulf in flat-bottomed boats and captured the (assumed)
strongholds that afterwards, moving westward into the peninsula
;

of Guerande, he captured successively the (assumed) strongholds of

Guerande, Saille, Batz, and Pen-chateau that the Veneti finally
;

took refuge in Le Croisic that their fleet anchored in the Grand-

;

Trait that the Koman fleet, which had hitherto been weather-bound
;

in a harbour opposite the Grande-Briere, sailed to join Caesar and ;

that the decisive battle took place off the promontory of Castelli.
Napoleon's theory, that the sea-fight took place ofi the mouth of the
Auray, that is to say, some 50 miles from the mouth of the Loire,
he covers with ridicule. Comment '
he exclaims,
!
'
ces deux '

flottes qui n'ont aucune communication entre elles, sans s'etre donne
le mot et sans se voir, sont parties toutes les deux le meme jour, a la
meme heure, ont mis le meme temps pour franchir une tres-grande
distance, et sans desemparer et se reposer, se sont rangees en bataille,
se sont battues, et tout a ete fait en une demi-journee.' All this
rhetoric, however, is directed merely against a suggestion of
Napoleon's which in no way touches his real argument and it ;

may safely be disregarded.

But further proof is offered that the theatre of the war could not
have been the gulf of the Morbihan. It is nothing less than this,
that in Caesar's time the gulf of the Morbihan did not exist If, !

argues Desjardins,^ it had existed then, it would have had a Koman

name. Jules Girard, moreover, affirms that it has been proved by
the investigations of MM. Ariondeau and de Closmadeuc that the

'
See d'Anville, Notice de Vancienne Gaule, p. 687.
2 Etudes arcMol, pp. 78, 80-1, 84-7, 89-90.
^ Geogr. de la Gaule mm., i, 281-3, .304.
682 THEATRE OF THE WAR
bed of the gulf has undergone a subsidence of some 5 or 6 metres.
They found knives of flint in a part of the bed which is never exposed
even at the lowest tideA
Now I freely admit that there has been subsidence since the time
when flint knives were dropped and megalithic monuments were
erected in Brittany. Any tourist can verify the fact for himself.
There are two stone circles on the islet of Er-Lamic in the gulf of the
Morbihan, one of which is only visible when the tide is exceptionally
low.2 But this does not prove that the gulf did not exist in 56 B.C.
The flint knives and the stone circles were probably two thousand
years old or more, even in Caesar's time. Obviously the facts are
consistent with the hypothesis that some of the islands which now stud
the surface of the gulf were then headlands insulated at high tide. As
Colonel de la Noe suggested, the promontories which Caesar described
have disappeared, partly from subsidence, partly from erosion.^
The theory that the plain of the Orande Briere was in Caesar's
time covered by a gulf is as baseless as the theory that the gulf of
the Morbihan did not then exist. The plain of the Grande-Briere
produced, at some remote epoch, abundant vegetation. Upon a bed
of clay there is a layer of peat 2 metres thick. The surface of the
peat is 85 centimetres above the mean level of the sea. The subsoil
therefore is hardly more than a metre below this level, and, if the
peat did not exist, would only be partially covered during some
hours of each tide.*
About 500 yards north-east of Breca, there is a menhir 5 feet
3 inches high, which is submerged by the floods of winter and ;

E. Orieux points out that on the borders of the plain of the Grande-
Briere are various monuments of stone, notablv the menhirs of Clos
d'Orange and La Vacherie.^ Est-ce,' he asks, que sous les eaux on
' '

1 Bull, de la Soc. de geogr., Sept., 1875, p. 234.

^ See Bull, de la Soc. polym. du Morhilian, 1882, pp. 8-24 ; and Homme, U
i, 1884, p. 421. I do not know whether Professor C. Vallaux is referring to
these monuments when he argues {Annales de geogr., xii, 1903, p. 27) that
no subsidence can be inferred from the fact that certain megaliths
'
are now
. . .

below the level of high tides ', and that, as they were originally built on a low '

coast, almost at the level of the sea', their present position is due simply to
'
erosion '. If so, I cannot follow him. The lower of the two circles mentioned
in the text was only discovered (in 1872) when the tide was extraordinarily
low, and is between 5 and 6 metres below the level of high tides. As M. de Clos-
madeuc says {Bull, de la Soc. polym. du Morbihan, 1882, p. 16), the gently
sloping line which is now formed by the southern declivity of the islet of Er-
Lamic has no sensible break, and is prolonged uninterruptedly to the base of
the stones which form the circle. Therefore, he concludes, the submergence
of the circle has not been caused merely by encroachment of the sea, but also
by subsidence. See also Bull, de la Soc. geol. de France, 4*^ ser., vi, 1906,
pp. 142, 147, and J. Dechelette, Manuel d'archeologie, i, 444.
^ Bull, de geogr. hist, et descr., 1889
(1890), pp. 29-30. Before the publica-
tion of my first edition I found that M. Le Moyne de la Borderie {Hist, de
Bretagne, i, 7) had rejected Desjardins's argument.
* Bull, de geogr. hist, et descr., 1889 (1890),
pp. 29-30. Cf. Rev. des Deux
Mondes, 2« per., Ixxix, 1869, p. 432 Bull, de la Soc. arch, de Nantes, xxi, 1882,
;

p. 189 and P. de Lisle du Dreneuc, Des Gaulois venetes, 1887, pp. 3, 5-7.
;

Bull, de la Soc. arch, de Nantes, xxi, 1882, pp. 185-6.

'"
The positions of these
monuments are marked on Sheet 104 of the Carte de VEtat-Major (1 80.000). :
 WITH THE VENETI 683

a construit ces monuments de pierre ? ^ Orieux's argument of

'

course is that the level of the Grande -Brie re, so far from having
risen, as de Kersabiec maintains, since Caesar's time, has actually
sunk at all events it is lower now than it was when the monuments
;

in question were erected.

The Grande-Briere, then, is out of the question and, with the
;

possible exception of the islands of Batz and Le Croisic, there was

not a single stronghold on the coast of Guerande which Caesar could
not have approached by land at high tide.^ But even if the campaign
began in the peninsula of Guerande, it did not end there. For Caesar
distinctly says that he captured several of the Venetian strongholds,
and that, before his fleet could join him, it had to sail across a vast '

open sea '. His statements can only mean that he penetrated far
northward into Venetia. Therefore, even if Venetia included the
peninsula of Guerande, it is very probable that he pushed far to the
north of it and, as it is more than unlikely that Venetia did include
;

the peninsula, the probability amounts to a certainty.

IV. C. de la Monneraye has proposed a compromise.^ He too con-
tends that Caesar must have penetrated far into Venetia but he ;

then brings him back again far to the south, and places the sea-fight
off the miniature peninsula enclosed by the entrenchments of St. Ly-
phard. He argues that when Caesar decided to wait for his fleet,
he must have chosen for his encampment some spot near the place
—
where the fleet was to assemble, that is to say, near the mouth of
the Loire for otherwise he could not have got supplies. The best
;

place, continues de la Monneraye, would have been the peninsula

enclosed by the entrenchments of St. Lyphard. These, he argues,
were probably the very fortifications which the legions constructed.
He admits that Caesar's statement regarding the difficulty which
his admiral experienced in joining him appears to militate against
his theory but he disposes of this difficulty by appealing to the
;

authority of Dion Cassius. When, he argues, Caesar says that the

fleet was weather-bound, he is probably referring to the ships which,
according to Dion, were coming from the Mediterranean. Therefore
there is no need to assume that there was a vast open sea between
' '

Caesar's camp and the mouth of the Loire. The sea in question was
the sea between the mouth of the Loire and the Straits of Gibraltar.
This theory will not bear examination. No unbiased mind could
detect in Caesar's narrative anything to warrant the assumption that
he marched back from the northernmost of the forts which he cap-
tured into the peninsula of Guerande. His commissariat was always
perfectly organized and if he was able to feed his army during the
;

time which he spent in reducing the forts of the Veneti, there was no
reason why he should not be able to feed it during the time that he
passed in waiting for his fleet. To assume that the retranche- '

ments of St. Lyphard are the remains of his camp is simply a wild
'

^ Bull. arch, de V Ass^^ hretonne, T ser., vii, 1887, p. 27. See also BiiU.
de la Soc. arch, de Nantes, xix, 1880, pp. 01-4 ; xxi, 1882, p. 215.
2 76., xix,
1880, p. 63.
^ Geogr. de la peninsule armoricaine,
pp. 149-52, 156, 166-7.
684 THEATRE OF THE WAR
guess. Finally, the authority of Dion Cassius, where it contradicts
the authority of Caesar, is worth nothing. For these reasons I believe
not only that Caesar penetrated into Venetia far to the north of
Guerande, biit also that the sea-fight took place in some higher
latitude.
V. This is also the view of P. de Lisle du Dreneuc.i He maintains
that cliffs answering to Caesar's description of those from which the
Koman legions watched the battle are not to be found until, moving
towards the north, one approaches the Pointc du Raz, the seaward
termination of the northern frontier of Venetia. Accordingly he
maintains that the battle took place off the Pointe de la Cornouaille '.
'

But Caesar's description of the place from which he watched the battle
is vague he simply tells us that he watched it from high ground
:

overlooking the sea, and that in the vicinity of his camp there was
a harbour. As he says that, after the battle, the survivors had no
means of defending the strongholds which had not yet been captured,
one might be inclined to argue that he had not advanced nearly as
far as the northern frontier of the Veneti. On the other hand, it is
possible that the strongholds of which he speaks did not belong to
the Veneti, but to their allies.^ Be this, however, as it may, I find it
difficult to reconcile the theory that Brutus had sailed so far north-
ward as the Pointe de la Cornouaille with Caesar's narrative.
After studying the whole literature of the question, I conclude that
the theory which Napoleon borrowed from the Comte de Grandpre,
although its truth cannot be demonstrated, is highly probable.-"^
M. Jullian,^ however, although he admits that it may be right even
in detail, differs from Napoleon on one important point. Contrary to
the prevailing opinion, he believes that Caesar penetrated into Venetia
not from the south but from the north of the gulf of the Morbihan :

he thinks that Caesar watched the battle not from the heights
of St. Gildas but from the peninsula of Locmariaquer, on the
opposite bank of the estuary of the Auray and he suggests that the
;

Venetian fleet may have put out to sea not from the Auray but from
—
Port Navalo, a roadstead off the peninsula of Sarzeau and between
St. Gildas and Locmariaquer. He argues, first, that the northern
district — —
Vannes, Auray, and Hennebont has always been the centre
of the economic, political, and religious life of Venetia secondly, ;

that if Caesar had come from the south, he would have risked being
blockaded in the peninsula of Sarzeau thirdly, that the spits and
;
'

headlands (Imgulis frofnunturiisque) ^ on which the strongholds

'

which he attacked were situated, were principally on the northern

coast of the gulf fourthly, that the mention of the shoals, har-
;
'

bours, and islands (vada, portus, insulas) ^ which were the theatre
'

^ Des Gaulois
venetes, pp. 8-9. ^ Cf. B. G., iii, 9,
§ 3.
* It may
or may not be true that, as Tranois affirmed [Mem. de la Soc. arch.
. .des Cotes-du-Nord, i, 1852, pp. 363, 365, 367), traces of the dykes which
.

Caesar built when he was besieging the strongholds of the Veneti still existed
in 1852 at the islands of Conlo, Goalabre, and Gavernis.
* Hist, de la Gaule, iii, 295, n. 6, 297, n. 7.

' B. G., iii, « lb.,

12, § 1. 9, § 6.
 —
;

WITH THE VENETI 685

of the of the gulf rather than of the coast of

war makes us think
Sarzeau ; roads interrupted by estuaries {pedestria
fifthly, that the ' '

itiyiera concisa aestuariis^) correspond with the route which leads

from Blain past Rieux and Elven to Vannes, Auray, and Hennebont
and, lastly, that the topography of Locmariaquer and Port-Navalo
agrees better than the sites adopted by Napoleon with certain details
in Dion Cassius's description of the battle,^ which M. Jullian prefers
to that of Caesar.
I do not share M. JuUian's appreciation of Dion's description ;

but his other arguments demand attention. None of them, as far

as I can see, is really opposed to Napoleon's opinion. Whatever
route Caesar may have followed, he doubtless intended originally to
penetrate to the political centre of Venetia I question whether he ;

would have run the slightest risk of being blockaded in the peninsula
of Sarzeau, for he was too strong, and the Veneti, who belonged to
the Blue Water School ', had staked their fortunes on the sea
'
he ;

was originally prepared to attack every stronghold that might defy

him, and there Avas nothing to prevent him from attacking the
strongholds north of the gulf even if he entered Venetia from the
south the Vilaine had a considerable estuary, and if he had pushed
;

far into Venetia, he would necessarily, even if he had come from the
south, have met with other estuaries, and may have actually done so
before he desisted from the fruitless labour of besieging elusive
garrisons. No certain decision, as M. Jullian wisely acknowledges,
is attainable but what leads me to adhere to the current opinion
;

is the belief that Caesar must have wished to keep in touch with
Brutus and the fleet, and therefore would not have cut himself of?
from them by marching far away from the Loire in order to enter
Venetia from the north.

CAESAR'S OPERATIONS AGAINST THE VENETI

Caesar describes the dykes which he constructed for the pur-
I.

pose of capturing the strongholds of the Veneti, in these words :

si quando, fnagnitudine operis forte superati, extruso mari aggere ac

molihus atque his oppidi moenibus adaequatis suis fortunis desperare
coeperant,^ &c. Napoleon,^ following von Goler, interprets this
passage as follows :
—
the Romans constructed two parallel dykes.
'

. .During the process of construction, the space between the two

.

dykes was regularly inundated at high tide but as soon as the ;

besiegers had brought them into contact with the stronghold, the
water could no longer find its way into the space, and it served
them as a kind of '' place d'armes ".' Thomann,^ however, agrees
with Riistow in questioning the necessity for two dykes, and justly
remarks that Caesar's text does not support von Goler's view.
1 B. G., iii, 9, § 4. 2 xxxix, 41-3.
;
B. G., iii, 12, § 3. " Hid. de Jules Cesar, ii, 124-5.
^ Der franzosischc Alias zu Ciisars yall. Krieye, p. 22.
686 OPERATIONS AGAINST THE VENETI
II. Did Caesar employ any ships while he was besieging the
Venetian strongholds ?
'
The Romans,' says Long,^ describing the battle between the
Roman fleet and that previous conflicts had dis-
of the Veneti, '
m '^

covered that they could not injure the enemy's ships by the beaks
of their vessels.' The passages upon which he bases this statement
are, rostro enim noceri non posse cognoverant, which occurs in Caesar's
—
narrative of the battle,^ and in the general description of the
Venetian ships ^ neque enim iis nostrae rostro nocere poterant, tanta
in iis erat firmitudo and accordingly, when Caesar writes that, after
;

capturing a number of the Venetian strongholds, he determined to

wait for his fleet,^ Long explains, He means all his fleet, the com-
'

plete fleet.' Of course he does but he does not mean that he

:

already had a part of the fleet with him for he says distinctly ^ ;

that the weather was too stormy for his ships to put to sea. Also,
after telling us that he placed Decimus Brutus in command of the
fleet, which he had ordered to assemble in the estuary of the Loire,
he says that he marched in person for Venetia with the land forces
(D. Brutum adulescentem classi Gallicisque navihus, quas ex Pictonibus
et Santonis reliquisque pacatis regionibus convenire iusserat, prae-

ficit et cum primum possit, in Venetos profi^isci iubet. Ipse eo pede-

stribus copiis contendit)? Not a word to show that he sent on any
ships to meet him. Besides, if he had been foolish enough to employ
a part of his fleet, any that escaped shipwreck would have been
destroyed by the powerful fleet of the enemy. His meaning is clear
enough. In the two passages which Long quotes he is simply stating,
as a fact, that the enemy's ships were too stout to be rammed by his
light galleys. Immediately before the second passage he writes that
the only advantage which his fleet had over the enemy's ships was
in speed {cum his navibus nostrae classi eius 7nodi congressus erat ut
una celeritate et pulsu remorum praestaret).^ It is clear that this is
only a general statement, and does not mean that the enemy's fleet
and Caesar's fleet, or any part of it, came to blows before the de-
cisive battle because a few lines further on, Caesar, using the word
;

classis again, says that he was obliged to wait for his fleet. A moment's
reflection might have convinced Long that the word cognoverant does
not support his argument for it is obvious that Brutus and his
;

officers, who had been weather-bound for weeks and had only just
arrived upon the scene of action, could not have discovered in '

previous conflicts that they could not injure the enemy's ships by
the beaks of their vessels '. They, at all events, had not engaged in
any previous conflicts
'
and the same storms which had prevented
'
;

them from putting to sea would probably have also prevented any
of the imaginary ships which might have engaged in previous con- '

flicts from communicating with them.' Is it credible that, if Caesar

'

^
Decline of the Roman Republic, iv, 112. ^ The itahcs are mine.
'
B. G., iii, 14, § 4. * lb., 13, § 8.
'
Compluribus expugnatis oppidis Caesar . . . statuit expectandam classem.
lb., 14, § 1.
« lb., 12, §5. ' lb., 11, §5. « lb., 13, § 7.
 OPERATIONS AGAINST THE VENETI 687

had used ships when he was besieging the Venetian forts, he would
have said that the Veneti were able to escape in their ships from
one port to another because the Roman ships were prevented by
storms from joining him [quod nostrae naves tempestatibus detine-
hantur) ? And if the fleet, as a whole, was unable to put to sea,
how could individual ships belonging to that fleet have encountered
the storm with impunity ? Cognoverant means not they had dis- '

covered (in previous conflicts) ', but they had ascertained '
(by '

inquiry or by inspecting Gallic ships in the estuary of the Loire).

1

Blanchard, indeed, says that we may gather from the narrative of

Dion Cassius that Caesar did emplo}^ ships when he was besieging the
Venetian oppida.^ But Caesar himself distinctly says that it would
have been impossible to do so.^ What Dion says is that when Caesar
marched for Venetia, he conveyed a number of ships, specially con-
structed to stand the ebb of the tide, which he had caused to be built
'
in the interior ', down the Loire.* But Dion, as usual, misunder-
stood Caesar, and invented. Caesar says that when he marched for
Venetia, he put Brutus in command of the Roman fleet and of the
ships of the Pictones and Santoni, and ordered him to set sail as
soon as he could.^ If Dion is right, the ships which were built on
the Loire set sail at once in spite of the storms, while Brutus re-
mained weather-bound with the rest of the fleet. As a matter of
fact, it is plain from Caesar's narrative that Brutus's fleet included
the ships which had been built on the Loire. Dion, however, says
that after Caesar had spent nearly the whole summer in besieging
the Venetian strongholds, Brutus arrived with his fleet from the
Mediterranean This ridiculous fiction flatly contradicts Caesar's
!

narrative. Evidently Dion imagined that the ships which were

built on the Loire were flat-bottomed, like those of the Veneti ;

whereas Caesar expressly says that they were galleys {naves longae).^
III. Napoleon says that the Venetian fleet was more numerous
'^

than the Roman Thomann ^ holds that Caesar's statement that

:

certain Venetian ships were hemmed in, each by two and, in some
cases, three Roman ones,^ proves the contrary. Thomann's argu-
ment is futile for the Romans may have destroyed their unwieldy
;

opponents in detail but there is nothing to show which of the two

;

fleets was numerically the stronger.

^Schneider {Caesar, i, 255) denies that cognosco can bear this meaning, unless
it relatesto an event which has actually happened {nisi ad rem factam relatum).
But refer to B. G., v, 19, §§ 1-2, and you will see that Schneider is wrong.
^ Bull, de la Soc. arch, de Nantes, xxii,
1883, p. 166.
^ Erant eius modi fere situs oppidorum, ut posita in extremis lingulis pro-

munturiisque neque pedibus aditum haberent neque navibus, quod rursus

. . .

minuente aestu naves in vadis adflictarentur. B. G., iii, 12, § 1. On the force
of the subjunctive adflictarentur see Schneider's Caesar, i, 245.
* avTos km Tovs Ovfverovs rjKaoe' /cm irXoTa ki' ttj ixiaoyeiq, a rjKovev iJTiTrjbda

TTfos Tr}v Tov diKeavov iraXippoiav ejpai, KaraCKivdaas, 5id re tov Aiypov voraixov
KaTfKo/^iae. xxxix, 40, § 3.
° B.G., iii, 11, §5. 6 /6., 9, §1.
' Hist, de Jules Cesar, ii, 126, n. 2.
^ Der franzosische Atlas, &c., pp. 23-4.
**
cum singulas binae ac ternae naves circumsteterant. B, G., iii, 15, § 1.
688 OPERATIONS AGAINST THE VENETI
IV. Dion's account of the battle ^ differs, in certain respects,
from Caesar's and when they differ, MM. de la Monneraye and
;

Jullian 2 prefer the testimony of Dion. In my essay on The Credi- '

bility of Caesar's Narrative (pages 236-237) I have criticized Dion's

'

account.

WHERE DID SABINUS ENCAMP WHEN HE

INVADED THE COUNTRY OF THE. VENELLI ?
Napoleon ^ places the camp of Sabinus on a hill [Chatelier], '

about 4 miles east of Avranches, belonging to the line of heights

which separates the basin of the See from that of the Celune '.
A Roman camp has been discovered there but Napoleon admits ;

that it was probably made at a time later than that of the conquest
of Gaul. L. Fallue says that local tradition identifies the camp with
Montcastre ^ but local tradition is easily manufactured, and usually
;

originates in the opinion of some local antiquary. Moreover, the area

of this camp is 40 hectares, or about 100 acres, which, tested by the
standard of Caesar's camp at Gergovia, would have been much too
large for Sabinus's three legions. Sabinus,' says Caesar, made his
' '

way into the country of the Venelli ' {in fines Venellorum fervenit^).
L. Mayeux-Doual,^ misunderstanding the meaning oi fines, says that
the camp must have been on the frontier of the Venelli, and places
it at Champrepus, which, he says, answers exactly to Caesar's descrip-
tion. Perhaps it does but so do other places in the country of the
;

Venelli. It is useless to attempt to fix the site for Caesar tells us

;
'^

nothing about the camp, except that it was in the country of the
Venelli, and on high ground, which sloped gently down for the
distance of about one Roman mile to the plain.^

CRASSUS'S ATTACK ON THE AQUITANIAN CAMP

W. Paul, remarking that the four cohorts which attacked the
Aquitanian camp in the rear made their way thither quickly, in
spite of the long detour which they took in order to avoid observa-
tion, argues that they must have ridden to the scene of action on
horseback.^ Caesar's account runs as follows Crassus equitum :

praefectos cohortatiis ut magnis praemiis pollicitationibusque suos ex-

citarent, quid fieri vellet ostendit. Illi, ut erat wiperatum, eductis lis
cohortibus, quae praesidio castris relictae intritae ab labore erant, et

^ xxxix, 41-3. 2 jjigi dg i(j^ Qaule, iii, 297, n. 4.

^ Hist, de Jules Cesar, ii, 129, 130, note 1.
* Etudes arch, sur Vhist. de Jules Cesar par . Napoleon III, 18G7, p. 30.
. .

' B. G., iii, « Mem. hist., 1870,

17, § 1. pp. 259-72.
' M. Jullian {Hist, de la Gaule, iii, 301, n.
6) offers us a choice among three,
including Chatelier and Cliamprepus, but liimself prefers Vire
—
grand carrcfour
'

de routes '
—about 33 miles oast of Granville.
« B. G., iii, 17, § 1, 19, § 1. » Berl. phil. Woch., v, 1885, col. 1185-0.
 AN IMAGINARY DIFFICULTY 689

longiore itinere circumductis, ne ex hostium castris conspici possent,

omnium oculis mentihusque ad pugnam intentis, celeriter ad eas quas
diximus munitiones pervenerunt} &c. On this Paul observes that
Caesar never uses educo in the Gallic War in its present sense without
mentioning the point of departure accordingly he proposes, as an
:

emendation, devectis.
This is a striking instance of the perverted ingenuity which is
responsible for many of the conjectural emendations that swarm in
German periodicals. Celeriter is a relative term, and simply means
that the cohorts made haste. And if Caesar does not use educo in
the sense required in any other passage in the Gallic War without
mentioning the point of departure, he does so in no less than eight
passages in the Civil War, namely in i, 41, § 2 64, § 6 81, §4
; ; ;

iii,41,§l; 54,§2; 64, § 6 67, § 3

; 81, § 4
; and 85, § 4.
;

[Meusel holds that the words Illi, ut erat imperatum, eductis iis
cohortibus quae, &c., are inconsistent with Crassus equitum prae-
fectos ostendit, for it seems to him inexplicable that cavalry
. . .

officers should have commanded infantry and accordingly he sup-

;

poses that between imperatum and eductis some words have been
lost.2 But is not Schneider's explanation {Caesar, ii, 286-7) satis-
factory ? Crassus, he says, requested his cavalry officers to pick out
the troopers who knew the ground best, to employ them as guides
for the cohorts, and to stimulate their zeal by rewards and promises.]

DID CAESAR ATTACK THE MORINI OR THE

MENAPII IN 56 B.C.?
Caesar says that he intended to attack both the Morini and the
Menapii ^ and Napoleon ^ carelessly infers from his narrative that
;

he actually did so. But close study will show that he invaded the
territory of one only of the two tribes and a passage in B, G., iv,
;

38, §§ 1-2,^ proves that, as we might expect from their geographical

position, it was the Morini whom he attacked, not the Menapii.

WHERE DID THE USIPETES AND TENCTERI

CROSS THE RHINE ?

Caesar says that the Usipetes and Tencteri crossed the Rhine not '

far from the sea (non longe a mari), and that, at the point where
'

they crossed it, the country on both banks belonged to the Menapii.^
'
B.G., iii, 26, §§ 1-2.
^ Jahresb. d. philol. Vereins zu Berlin, xi, 1885, p. 201.
^ 5. 6^., iii, 28, ^ Hist, de Jules Cesar, ii, 134-5.
§ 1.
'"
Caesar postero die T. Labienum legatum cum iis legionibus quas ex Bri-
tannia reduxerat in Morinos qui rebellionem fecerant misit. Qui cum propter
siccitates paludum quo se reciperent non haberent, quo perfugio superiore anno
erant usi, omnes fere in potestatem Labieni venerunt,
« B. 0., iv,
1, § 1 ; 4, § 2.
1093 Yy
690 THE ROUT OF
General Creuly ^ holds that by Rhenum Caesar meant the Meuse
below its junction with the Waal at Gorkum, a view which I have
combated on page 370.2 If the Usipetes and Tencteri crossed the
Rhine anywhere below the point where the Waal diverged from the
main stream, they must have crossed either the Waal or, if they
crossed the lower Meuse at the point which Greuly indicates, the
Meuse as well but of two passages Caesar says nothing. Folir re-
;

marks that if Caesar used the word Rhenum in its strict sense, the
passage must have taken place above the first bifurcation of the
Rhine ^ and this is the common opinion. It has been objected
;

that, in that case, the words non longe a mari would be inaccurate ;

but longe is a relative term, and, in relation to the whole length of

the Rhine, Emmerich or Cleve might fairly be described as not far '

from the sea '. On the left bank of the Rhine, above its first bifurca-
tion, between Xanten and Nymegen, there is a chain of heights.
The only practicable points of passage for the Germans would have
been at Xanten itself and lower down, near Cleve. Napoleon'*
asserts that they crossed at both these points. It would appear,
however, from Caesar's narrative ^ that they crossed at one point
only ; and, having regard to the words non longe a mari, it seems
reasonable to look for that point as near the sea as possible. I feel
little hesitation, then, in concluding that the Usipetes and Tencteri
crossed the Rhine near Cleve.^

ON A DIFFICULTY RAISED BY G. LONG RE-

GARDING ONE OF CAESAR'S REASONS FOR
ATTACKING THE USIPETES AND TENCTERI
Stating his reasons for having determined, after the treacherous
attack of their cavalry upon his, to march against the Usipetes and
Tencteri without further delay, Caesar writes cognita Gallorum
:
"^

injirmitate, quantum iam apud eos hostes uno proelio auctoritatis

essent consecuti sentiehat quihus ad consilia capienda nihil spatii
;
dandum existimahat. Gallorum, according to Long,^ means Caesar's
Gallic cavalry, for '
the defeat of the cavalry could have no
. . .

immediate effect on those Galli who were at a distance.' But it is

contrary to Caesar's practice to speak of his Gallic cavalry by the

^ Rev. arch., nouv. scr., viii, 1863, pp. 27-8.

2 Heller {Philologus, xxii, 1865, pp. 131-2) remarks that the territory on the
right bank of the lower Meuse belonged not to the Menapii but to the Batavi
but this argument is obsolete. It is very doubtful whether the Batavi occupied
this country in Caesar's time at all events, it is most improbable that the only
:

passage in the Commentaries in which they are mentioned iv, 10, § 1 was — —
written by Caesar. See p. 692, n. 2.
* Rhein. Archiv, iv, 1811, p. 235.

* Hist, de Jules Cesar, ii, 137, n. 1. '"

B. G., iv, 1, § 1, 4.
« Cf. von Goler, Gall. Krieg, 1880, p. 113. M. Jullian {Hist, de la Gaule, iii,
325, n. 3 ; 328, n. 5) decides for Xanten.
' B. G., iv, 13, § 3. ^ Decline oj the Roman RepuhUc, iv, 155.

1
 —
THE USIPETES AND TENCTERI 691

generic name of Galli ^ they were not likely to side with the
'
' :

invaders by whom they had just been so roughly handled if they :

had attempted to do so, the legions would have crushed them in five
minutes and, as Caesar tells us in a later book ^ that the news of
;

an important event could travel 160 (Roman) miles between sunrise

on a winter's day and eight o'clock at night, the defeat of the cavalry
could have had a very immediate effect on those Galli who were at
'

a distance '.

WHERE WERE THE USIPETES AND TENCTERI

DEFEATED ?

The question which I am going to discuss is the most complicated

and difficult that Caesar's memoirs present. Broadly speaking,
three solutions have been propounded (1) that the Usipetes and :
—
Tencteri fled to one or the other of the known confluences of the
Meuse and the Waal or to some point in the neighbourhood of one or
the other (2) in spite of the statement of the text,^ that they fled
;

to the confluence of the Rhine and the Moselle and (3) that they ;

fled to the confluence of the Meuse with some river unknown. Let
us see, first of all, what Caesar has to say.
I. The Usipetes and Tencteri crossed the Rhine, not far from the
sea, took possession of all the buildings of the Menapii, passed the
winter in their country, and fed on their stores. On hearing of their
incursion, Caesar proceeded to join his legions and, as soon as he
reached their quarters, learned that the Germans had wandered
further south and reached the territories of the Eburones and Con-
drusi. When his preparations were complete, he marched towards
the district where he heard that the Germans were and when he ;

was within a few days' march of their whereabouts, their envoys

met him. The envoys undertook to convey his ultimatum to their
principals,and to return with their answer in three days. Caesar
was aware, at the time when he was parleying with the envoys, that
the German cavalry were in the country of the Ambivariti, on the
opposite side of the Meuse, whither they had gone to plunder and
collect corn.'* Immediately after the passage in which he mentions
this movement
the narrative is interrupted by the well-known para-
graph which describes the course of the Meuse. In the good MSS.
it runs as follows Mosa frofluit ex monte Vosego, qui est in finihus
:

Lingonum, et "parte quadam ex Rheno recepta, quae appellatur Vacalus

insulamque efficit Batavorum, in Oceanum influit neque longius ah
Oceano milihus passuum LXXX
in Rhenum influit. The words que and
in Oceanum influit are absent from certain inferior MSS. instead :

^ I can only discover one passage {B. G., vii, 13, § 2) in which he does so ;

and there his meaning is unmistakable. See Meusel's Lex. Caes., i, 16G4-0.
' B. G., vii, 3, §§ 2-3.
^ Cermani . . . cum ad confluentem Mosae et Rheni pervenissent, &c. Jb.,
iv, 15, § 2. * lb., 1, § 1 ; 4-9.
Yy 2
 692 THE ROUT OF
of ah Oceano the Aldine edition has ab eo and Vind D. ab Rheno ;

and both Vind D, and the Aldine edition have in Oceanum instead
of in Rhenum. Accordingly Nipperdey ^ conjectures that Caesar
wrote insulam efficit Batavorum neque longius ab Rheno milibus
fassuum LXXX in Oceanum influit. The authenticity of the descrip-
tion has, however, long been suspected and Meusel has recently
;

given reasons for rejecting it which seem wellnigh conclusive.^ Still,

it would hardly be safe to set it aside as having no historical value.
The various readings only concern us in so far as they affect the
question where the Meuse was joined by the Waal.^ If the inter-

Caesar, pp. 75-6.

^

Jahresb. d. philol. Vereins zu Berlin, xxxvi, 1910, pp. 23-6.

2 Meusel argues
that the description of the Rhine is irrelevant that the Meuse rises not in
;

the Vosges, but on the plateau of Langres, both of which regions Caesar had
visited that the Vosges is not in jinihus Lingonum ; that F. Stolle has given
;

strong reasons for believing that the Batavi did not occupy the delta of the
Rhine before the time of Augustus that the Rhine did- not flow through the
;

country of the Nantuates (cf. pp. 454-5, supra) that the geographical position
;

of the Triboci is mis-stated ; that in the list of the Cisrhenane tribes the Rauraci,
Nemetes, and Vangiones are ignored ; that the statement that the Rhine,
as it approaches the sea, branches off into several channels and forms numerous
'

large islands' {in plures dijjfluit partes multis ingentibiisqiie insulis ejfectis) is
absurdly false ; that, as the Menapii occupied both banks of the Rhine not '

far from the sea', there would have been no room for the Batavi and the
'
fierce rude tribes ', who are said to have occupied many of the numerous
'
'
'

large islands ; that Caesar would not have written parte quadam ex Rheno, but
'

parte Rheni ; and, finally, that caput jlurninis elsewhere [in prose ?] means not
the mouth, but the source of a stream. Caput is, indeed, used in the sense of
ostium in Livy, xxxiii, 41, § 7, xxxvii, 18, § 6, 37, § 3, not to mention poets
(see Thesaurus ling. Lat.,\\\, 410) ; but in the face of Meusel's array of objections
it will not be denied that the chapter is at least open to grave suspicion. I am
rather inclined, however, to believe that Dion (xxxix, 49) may have found the
passage in his copy of Caesar ; and if so, the interpolation, assuming that it
existed, must have been much earlier than Meusel believes. I should add that
Strabo names the same tribes, except the Nantuates and the Batavi, in con-
nexion with the river as the writer of B. G., iv, 10, and names them in the same
order and he remarks that Asinius Pollio blamed those writers who said that
;

the Rliine had more than two mouths. Also, like the writer of iv, 10, he
observes that the Rhine is swift. Meusel, emphasizing Strabo' s omissions,
points out that he does mention the Menapii ; and he urges that there is no
need to suppose that either of the two writers copied the other, that the Latin
writer may have copied Strabo, and that Asinius Pollio may not have referred
to Caesar. All this is true but while Meusel is justified in concluding that
:

'
the proof of the authenticity of the chapter is not made good ', it is equally
true that spuriousness is not absolutely proved.
[A. Klotz {Caesarstudien, pp. 36-43, 135-8), who argues in much the same
sense as Meusel, thinks that the chapter was derived mainly, through Strabo,
from Timagenes. He remarks
that Caesar does not couple the Batavi in
B. G., 28, § 1 with the Menapii and the Morini,
iii, —
the tribes which, while all '

the rest of Gaul was tranquillized, remained in arms that he would have
' ;

written longius immediately before milibus passum LXXX; that he would not
have used the quasi-poetical expression, cifatus fertur, or the passive, incolitur;
and above all, that the territory of the Mediomatrici did not touch the Rhine.
I do not underrate the combined force of these assaults, though I regard the
linguistic arguments as individually weak :but the Menapii might well have
had lands on both sides of the Rhine and a seaboard without possessing the
insula Batavorum and it is impossible to prove that the Mediomatrici were
;

separated from the Rhine.]

^ In the first edition
(p. 680, n. 4) I minutely examined the arguments by
 —
THE USIPETES AND TENCTERI 693

passuum LXXX in Oceanum

polator wrote neque longius ah eo milibus
influit, eo means Vacalo} and he meant that from the point where
the Waal joined the Meuse, the Meuse flowed 80 Roman miles to the
sea. If he wrote neque longius ah Oceano milihus passuum LXXX in
Rhenum influit, he may possibly have meant that the Meuse, after
joining the Waal at Fort St. Andries, flowed on till it rejoined the
Waal (considered as a branch of the Rhine) at Gorkum, and thence
(under the name of the Rhine ^) flowed 80 miles to the sea but it :

is not certain that the confluence at Gorkum existed in Caesar's

time and Napoleon^ concludes from a study of the deserted beds
;
'

of the Rhine that the Waal then joined the Meuse at Fort St.
'

Andries, and that the channel of the Waal which is now prolonged
to the east of Fort St. Andries did not then exist. Walckenaer main-
tains that the words in Rhenum the junction of the
influit refer to
Meuse with the Waal at Fort St. Andries, and that, from the point
of junction, the Meuse flowed 80 miles to the sea. The former inter-
pretation is consistent with the theory that the Meuse in Caesar's
time, as before 1856, joined the Waal by a connecting channel at
Fort St. Andries * and again joined it at Gorkum, if we assume that —
the interpolator made a mistake in speaking of 80 miles but it is :

inconsistent with the statements of Tacitus, who says (1) that the

which the various readings have been defended but, considering that the
;

passage is most probably spurious, I do not think it necessary for historical

purposes to reprint the note.
^
C. Miiller, in his edition of Ptolemy {Geogr., i, 221), holds that eo means
Rheno, and that the writer was speaking of the distance from the mouth of the
Meuse to the western mouth of the Rhine. In this opinion he is, I believe,
alone.
2 General Creuly {Rev. arch., nouv. ser., viii, 1863,
pp. 27-8), remarking that
lacitus {Ann., iii, 6) speaks of the Meuse, below its junction with the Waal, as
Mosa, argues that Caesar's statement Mosa in Oceanum influit, neque
. . .

longius ah Oceano milibus passuum LXXX

in Rhenum influit proves that it —
was also called Rhenus. '
La plume rapide de Tecrivain,' he says, verse la '

Meuse dans 1' Ocean, puis elle se rectifie en faisant couler d'abord la Meuse dans
le Rhin.' Creuly refers to Tacitus {Hist., v. 23), who says Mosae fluminis
IS amnem Rhenum oceano ajfundit.
^ Hist, de Jules Cesar, ii, 143, n. 1 and Planche 14. M. Reclus {Nouv. Geogr.
Univ., iv, 218) as well as Napoleon denies that the Meuse, in Caesar's time,
joined the Waal at Gorkum :it flowed on, he says, towards the west in the
bed of the Oude-Maas. Ci.Rev. d'hist. et d'arch., i, 1859, pp. 296-303. On the
3ther hand, A. Belpaire {Etude stir la formation de la plaine maritime depuis
Boulogne jusqu'a Danemark, 1855, pp. 200-1) concludes that the Meuse s'est '

toujours dirige vers Gorcum . ou que la branche, dirigee par Heusden

. . . . .

a dii s'obliterer a une epoquo bien anterieure a celle de 1421 [the year of the
. . .

^reat inundation] de maniere a etre definitivement abandonnce par les eaux

1 cette epoque '. His argument is as follows Ce qui est de nature a jeter le
:
'

plus grand doute sur la realite de I'ancien cours de la Meuse par Heusden et
Geertruidenberg c'est que si I'inondation de 1421 avait eu pour effet d'attirer
, . les eaux du Waal vers le Hollandsdiep, a bien plus forte raison les eaux
.

ie la Meuse qui auraient dans ce cas ete traversee^ par I'inondation, auraient
iu etre attirees du meme cote et par consequent le cours pretendu par Heusden,
a,u lieu de se fermer comme on soutient qu'il I'a fait a la suite de I'inondation
ie 1421, aurait du s'accroitre do manicrc a devcnir le seul dcbouchc dcs eaux
ie la Mouse. Est-il croyable quo ce soit Tinvcrso qui ait eu lieu, et que la
branche (^ui s'cioignait do I'inondation se soit accrue au dcpens de cello qui s'y
'
rendait directonient ? ^
8ee p. 695.
694 THE ROUT OF
sea bounded the western side of the insula Batavorum, and the Rhine
its rear and sides (insulam quam mare Oceanus afronte, Rhenus
. . .

amnis tergum et latera circumluit), and (2) that, at the beginning or

eastern extremity of the island, the Rhine divides into two branches ;

that the northern branch flows direct to the sea that the southern
;

branch, under the name of Vahalis, flows on until it joins the Meuse ;

and that the Meuse flows on until it joins the sea {Rhenus apud . . .

fvincifium agri Batavi velut in duos amnes dividitur, servatque nomen

et violentiam cursus, qua Germaniam fraevehitur, donee Oceano mi-

sceatur : ad Gallicam ripain latior et placidior adjluens ; verso cog-

nomine Vahalem accolae dicunt ; mox id quoque vocabulum mutat Mosa
flumine, eiusque immenso ore eundem in Oceanum effunditur) ) In
other words, the statement of Tacitus, except in the fact that it does
not specify any distance, harmonizes with the reading neque longius ah
eo [that is Vacalo'] milihus passuum LXXX
in Oceanum influit and ;

I agree with Long,^ who says that the meaning of B. G., iv, 10, § 1,
'
appears to be that the island of the Batavi was formed by the
Waal the Mosa, and the main stream of the Rhine '. But the
. . .

point which I wish to emphasize is this whatever reading Ave

:

adopt, unless it be Nipperdey's conjecture, the interpolator places

the junction, or a junction, of the Meuse and the Waal at a point
about 80 Roman miles from the sea.^
C^ontinuing his narrative, Caesar says that he had got within
12 Roman miles of the Germans, when the envoys returned to him.
They asked him to give them three days' grace, in order that they
might send messengers to the Ubii, to find out whether the Usipetes
and Tencteri might safely settle in their country. Within three days
these messengers would undertake to go to the country of the Ubii
and return to Caesar. Caesar told them in reply that he would only
march four miles further on that day, in order to get water and he ;

ordered them to return to him on the following day that he might

take their request into consideration. Next day he marched against
and attacked the Germans in their encampment, and drove them in
rout ad confluentem Mosae et Rheni,^
What does confluentem mean ? It is commonly translated by con- '

fluence but Heller,^ affirming that the word only has this meaning
'
;

in late Latin, argues that Caesar, if he had meant to denote a con-

fluence of the Rhine with the Meuse, would have written (ad) con
fluentes Rhenum et Mosam. Moreover, Heller continues, the Rhine
did not flow into the Meuse, and after Caesar had mentioned the
Waal, he would not have spoken of it as the Rhine. With con-
fluentem we must supply fluvium Caesar called the Waal co7i-
:

fluentem, or connecting stream ', in order to show that the fugitives

'

^ Ann., ii, 6.
2 W. Smith, Diet, of Greek and Roman Geogr., i, 381-2.
^ Walckenaer indeed says {Geogr. des Gaules, i, 494) that the text accepted

before the appearance of Oiidendorp's edition neque longius ab eo in . . .

—
Oceanum transit might mean that the Waal, from the point where it leaves
the Rliino to its junction with the Meuse, has a course of 80 miles but it ;

is impossible to extract this sense from any reading which has been proposed.
* B. G., iv, U-o. ^ Fhilologua, xxii, 1805, pp. 132-3.
 THE USIPETES AND TENCTERI 695

were hemmed
in on their right by the Rhine, in front by the Waal,
on by the Meuse. But confluens was used in the sense of
their left
'
confluence by Pliny ^ only a century after Caesar's death
' and ;

seeing that both confluens and confluentes are very rare, it is surely
arbitrary to deny that Caesar might have used the former word with
the same meaning. Nevertheless, I admit that Heller's translation
may be right.
n. The points that we have first to determine are these (1) :
—
Where did the Waal, in Caesar's time, join the Meuse ? (2) Where
were the Germans when they had reached the territories of the
'

Eburones and Condrusi ? (3) Had they all, or had only their
'

advanced guard reached that territory ? (4) Had they moved from
that position when Caesar attacked them ? (5) On which bank of
the Meuse were the Ambivariti ? (6) How far were the German
envoys from the country of the Ubii when they promised to go
thither and return with an answer to Caesar in three days ? (7) How
far from the river in which the remnant of the fugitives perished was
the camp which Caesar attacked ?
(1) The Meuse at present joins the Waal near Gorkum. Before
1856 it was joined to the Waal by a connecting channel at Fort St.
Andries as well.^ According fo P. Cluver,^ who is followed by Des-
jardins * and Kiepert,^ it did not join the Waal, in the time of Caesar,
either at Fort St. Andries or at Gorkum. Those two junctions were,
Cluver maintained, due to modern canalization. In Caesar's time
the Meuse quitted its present bed at Megen, flowed past Battenburg,
Heusden, Waelwyck, Gertruidenberg, Maasdam, Westmaas, Simons-
haven, and Biert, and joined the western branch of the Rhine at
Geervliet, only 7 miles from the sea. Cluver's argument is that the
channel of an old river runs from the neighbourhood of Bockhoven in
the direction which I have just indicated that this channel, in the
;

eastern part of its course, is called the Hedickse Maas, and thence to
its western extremity the Oude Maas or Old Meuse '
and that this '
;

name proves that the Meuse, in Caesar's time, flowed in the channel
in question. But in Cluver's map and in Desjardins's the Oude Maas
quits the channel of the modern Meuse at Megen in the Descriptio
:

Fluminum Rheni, Vahalis et Mosae, a map which, in the Catalogue

of the British Museum, is dated 1630, it is not traced further eastward
than a point about two miles south-east of Heusden and with this ;

M. Reclus's map agrees.^ The Hedickse Maas, moreover, appears in

the Descriftio as a streamlet, which enters the Meuse between Bock-
hoven and Hedickhuysen. It is true that in sheets 39 and 45 of
Nat. Hist., vi, 17 (21), § 63
1
; 28 (32), § 145.
See E. Reclus, Nouv. Geoyr. Univ., iv, 218. Numerous maps of the sixteenth
2

and seventeenth centuries in the British Museum show the junction of the
Meuse and the Waal at Fort St. Andries, e.g. Fluviorum Rheni, Mosae . . .

descriptio, cmendata per F. de Witt.

^ Germania antiqua, 1031, lib. ii,
pp. 459-00; De tribus Rheni alveis, 1011,
pp. 34-8. « Geogr. de la Gaule rom., i, 118-9, 124-6.

' Galliae Cimlpinac et Truiisalpinae

. tabula in iisum scholarum descripta.
. .

« Nouv. Gcofjr. Univ., iv, 217. Cf. Kuait van de Rivicr de Bovcn Maas, &c.,
(1 : 10,000), Blad32.

'<^roF^^
'v^/ ST. MICH
-^ \ COLL

IL/BRAR^
696 THE ROUT OF
a map called Topographische en mililaire Kaart van liet Koningrijk
der Nederlanden (1 50,000) the Oude Maas is depicted in the neigh-
:

bourhood of Megen but it is simply a sheet of water which extends

;

about two miles in a south-easterly direction from a point about half

a mile from the bank of the Meuse. It should seem, then, that for
tracing the channel of the Oude Maas as far eastward as Megen
Cluver had no authority except the existence of this isolated sheet
of water nor does he adduce any proof of his assertion that the
;

Meuse did not, in Caesar's time, join the Waal at Fort St. Andries.^
Finally, Desjardins's map is absolutely inconsistent with B. G., iv, 10;
his theory forces him to maintain that both the writer of that chapter
and Tacitus were wrong in saying that the Meuse helped to form the
insula Batavorum and he is therefore driven to assert that the
;

former knew nothing of the matter which he undertook to describe.^

D'Anville^ conjectured that in Caesar's time the confluence of the
Meuse and the Waal was at Dordrecht, about 30 miles from the sea ;

but while he is justified in rejecting Cluver's theory, he cannot

establish his own. I can only conclude that it is impossible to prove
that the Meuse did not receive the Waal near Gorkum and it may ;

have been linked to the Waal at Fort St. Andries as well.* (2) When
the Germans had reached the territories of the Eburones and
'

Condrusi they must have advanced at least as far southward as the

'

latitude of Liege. (3) General Creuly insists that there is not a single
word in Caesar's narrative which goes to show that they retreated
before him as he advanced against them. Therefore, he argues,
those who had advanced as far as the territories of the Eburones
and Condrusi were only a party of cavalry who had been sent to
reconnoitre the country in which the host proposed to settle.^ Now
Caesar is often desperately concise and this part of his narrative;

is confessedly the most obscure in the whole of his book. If Creuly

is right, Caesar omitted to mention that those Germans who had
penetrated into the country of the Condrusi were only a reconnoitring
party therefore in either case Caesar was silent on an important
:

matter. When we read that Germani had advanced as far as the

'
'

territories of the Eburones and the Condrusi, and that Caesar began
to march towards the place where he heard that Germani ' were,^ '

^ Tout nous porte a croire,' says Desjardins {Oeogr. de la Gauh rom., i, 122),
'

who professes himself a faithful disciple of Cluver, que la reunion au fort

'
. . .

St. Andre est moderne.'

. . . Tout nous porte a croire is not proof. Where is
the record ? After a prolonged search I cannot find any. Nor, it should seem,
can M. Reclus. Nor could Johannes Pontanus, a diligent geographer of the
seventeenth century. See his Disceptationes chorographicae de JRheni divortiis,
&c., 1614, p. 32, and cf. Jahrh. d. Vereitis von Altertliumsfreunden im Rheinlande,
X, 1847, p. 58. 2 Oeogr. de In Gauh rom., i, 122.

^ Notice de Vancienne Gaule,

pp. 407-8, 6G8-9.
* Having patiently studied all the monographs that have been written on the

subject, I believe that it is impossible to trace with certainty the courses of

the lower Meuse, the lower Rhine, and the Waal, as they existed in the time of
Caesar. The Belgian geologist, M. A. Rutot (Le inouvemenf geogr. [a periodical].
1897, col. 99), concludes that the physical history of the Meuse closes with
'

the Quaternary Period '

^ Rev. arch., nouv. ser., viii, 18(53, p. 29. « B. G., iv, 7, § 1.

 —
THE USIPETES AND TENCTERl 697

we naturally take for granted that German! means the entire '
'

German host for it is hardly credible that Caesar would have

;

marched against a mere reconnoitring party. A. Dederich,i quoting

Caesar's words iter in ea loca facer e coepit quibus in locis esse Ger-
manos audiebat — lays stress upon his use of the imperfect tense, and
argues, as I understand him, that Caesar was speaking of successive
reports which reached him on his march and accordingly he infers ;

that by the time Caesar crossed the Meuse, the (assumed) recon-
noitring party may have retreated to join the main body. Possibly ;

but the meaning which Dederich attaches to audiebat seems irre-

concilable with Caesar's use of the word coepit and I believe that ;

he was speaking of reports which had reached him before he started.

All that seems certain is that if the whole host moved away from the
territories of the Eburones and the Condrusi, they did so before or
soon after he quitted the winter- quarters of his army for his ;

narrative shows unmistakably that he moved nearer and nearer to

a fixed, not a receding point. But the natural meaning of Caesar's
words appears to be that the Germans as a body, not merely their
advanced guard, had reached the territories of the Eburones and
the Condrusi.2 (4) If so, they must have moved from that country
before Caesar attacked them unless their flight occupied more than
one day. For it is generally ^ admitted that they fled either to
the confluence of the Meuse and the Waal or to the confluence of the
Rhine and the Moselle. In either case their final destruction would
have taken place at a considerable distance from the territories of
the Eburones and Condrusi. (5) I have already given reasons for
believing that the Ambivariti dwelt, as is generally held, on the left
bank of the Meuse.* (6) Caesar makes it clear that, when the German
envoys proposed to him to send a message to the Ubii, they were not
more than 12 Roman miles from the German camp, which he attacked
on the following day. If, then, the defeat of the Germans took place
near the confluence of the Meuse and the Waal, and if the said con-
fluence was near the site of Fort St. Andries, the envoys were at least
60 miles, in a direct line, from the nearest frontier of the Ubii.^ If,
on the other hand, the defeat took place near the confluence of the
Moselle and the Rhine, the German messengers would only have had
^ Julius Ccisar am Bhein, 1870, pp. 19, 28.
^ Meusel wrote on the margin of his copy of my first edition that Caesar
meant by Gennanos sicher das ganze Heer (undoubtedly the entire host).
'
'

^ Generally, but not universally. Von Cohausen, A. Dederich, von Kampen,

and T. Bergk think otherwise, as will presently appear but I engage to prove ;

that they are all wrong. ^ See

pp. 368-70.
^ I assume that the territory of the Ubii
extended no further northward than
Cologne or, more probably, the mouth of the Sieg, near Bonn. The name Sieg
is generally connected with that of the Sugambri, who dwelt between that river
and the Lippe. It would appear, then, that the territory of the Ubii was bounded
on the north by the Sieg but Heller (Philologus, xxii, 1865, pp. 160-2),
;

believes that they possessed a strip of land north of the Sieg as far as Cologne,
and between the Rhine and the Sugambri. If so, the Sugambri must have
crossed their territory in 53 b. c. when they made their raid into the country
of the Eburones [B. G., vi, 35, §§ 4-6). But do not Caesar's words, Sugambri
qui muil proximi Eheno, prove that they crossed the Rhine immediately from
their own territory ? Cf. B. G., i, 54, § 1, and iii, 11, § 1.
698 THE ROUT OF
to travel the few miles that separated them from the Rhine, and,
after crossing that river, they would have found themselves in Ubian
territory. (7) It is generally taken for granted that the scene of the
rout of the Usipetes and Tencteri was comparatively near the con-
fluence whether of the Waal and the Meuse or of the Rhine and the
Moselle. Levesque de la Ravaliere, however, infers from Caesar's
saying that the Germans were tired out when they reached the con-
fluence of the Rhine and the Meuse, that their flight had extended
over a considerable distance.^ This is also the opinion of Achaintrc.-
As Caesar does not say that the Usipetes and Tencteri had moved
away from the country of the Condrusi, he conjectures that their
defeat took place somewhere in the neighbourhood of Aix-la-Chapelle.
I believe, however, that almost every one who reads his Caesar
attentively will conclude that Aix-la-Chapelle is much too far from
and I am sure that the considerable distance
'
either confluence
' '
;
'

of which de la Ravaliere speaks was performed within a single day.

Moreover considerable distance is an elastic expression. On the
' '

theory of Napoleon, the scene of the rout, 8 miles north-west of

the river Niers, was quite 20 miles from the alleged confluence of the
Meuse and the Waal at Fort St. Andries. Surely this distance was
enough to tire the Germans.*^
III. As we have seen, if the MSS. are correct, Caesar says that the
Usipetes and Tencteri fled to the confluence of the Mosa (Meuse) and
the Rhine. Merivale and Long, however, as well as Cluver, de Valois,
Sanson, d'Anville, and von Goler, have argued that by Mosae (or
whatever Caesar may have written) we ought to understand not the
Meuse but the Moselle. I say, whatever Caesar may have written,'
'

because Cluver* would alter Mosae to Mosulae, and Long's sugges-

tion 5 that the two rivers had the same name is improbable. The
Latin name of the Moselle was Mosula or Mosella. There is no
evidence that the Moselle was ever called Mosa and if it had been,
;

Caesar would not have used the word Mosa to describe the Moselle in
B. G., iv, 15, § 2, when he had used the very same word to describe
the Meuse in chapters 9 and 12.
Cluver decides for the Moselle because, as I have already shown,
^ Hist, deVAcad. Roy. des inscr. et belles-lettres, xviii, 1744-6 (1753), p. 216.
- Caesar, i, 145.
^ M. Jullian {Hist, de la Gaule, iii, 328, n.
5) places the German camp on
'

the wooded heights behind Sonsbeck less than 10 kilometres [about 6 miles]
. . .

from Xanten', and {ib., 329, n. 5) conjectures that the Germans lied, not to
the confluence of the Meuse and the Waal or anywhere near it, but to Gennep,
30 kilometres, or about 18 miles from the battle-field. I cannot understand
how he reconciles the former opinion with Caesar's statement that the Germans,
before he marched against them, '
were now wandering further afield {latins
iam vagabantur) and had reached the territories of the Eburones and Condrusi',
or the latter with the statement that they fled ad confiuentem Mosae et Rheni.
'
Je ne crois pas necessaire,' he says, ' de conclure du textc de Cesar, ad con-
fiuentem Mosae et Rheni, que les fugitifs soient arrives jusqu'au confluent.'
Perhaps not ; but how can any one conclude that they fled to Gcmicp, which
is on the Mouse, about 30 miles from Fort St. Andries, and has no more con-
nexion with the Rhine than Guildford with the Thames ? ^^'ould M. Jullian
maintain that Mcaux is ad confiuentem Matronae et Sequanae ?
* De tribus Rheni alveis, ^ Caesar,
p. 38. p. lUl.
 —
THE USIPETES AND TENCTERI 699

he believes liimself to have proved that the Meuse did not, in Caesar's
time, join the Waal or the Khine at all until it reached Geervliet.
If Cluver's premiss is right, so is his conclusion but I have shown ;

that his premiss is very doubtful. Even so, however, his conclusion
may possibly be right.
Merivale ^ holds that the fact that the Germans only required '

three days ^ to send a message to the Ubii (on the right bank of the
Rhine, between Cologne and Coblenz) ... is quite inconsistent ' with
the statement that the battle was fought near the confluence of the
Meuse with the Rhine.
Long says that Caesar tells us nothing of a long march up
'

the Rhine to make his bridge '.^ And again, Before Caesar saw the '

Germans, they had left the Rhine and advanced south of Liege, and
when the Romans crossed the Maas and approached them, they
could not move westward, nor would they move northwards into
the country where they had wintered and fed on the stores of the
Menapii and as they finally fled to the Rhine, it is plain that the
;

junction of the Rhine and Mosaisthe junction of the Rhine and Mosel.'*
Napoleon, on the contrary, thinks that the country between the'

Meuse and the Rhine, to the south of Aix-la-Chapelle, is too much

broken, and too barren to have allowed the German emigration, com-
posed of 430,000 ^ individuals with wagons to move and subsist
. . .

in it. Moreover, it contains no trace of ancient roads, and if Caesar

had taken this direction, he must have crossed the Ardennes, . . .

a circumstance of which he would not have failed to inform us.

Besides ... on the news of the approach of Caesar, instead of direct-
ing their march towards the Ubii, who were not favourable to them,
the Germans .would have concentrated themselves towards the
. .

most distant part of the fertile country on which they had seized,
that of the Menapii '.^
Napoleon places the battle-field 8 miles north of Goch, which is on
the river Niers. Caesar states that on the day before the battle he
intended to march 4 miles to get water and Napoleon's argument is
;

that since, to the north of the Roer, there exists, between the Rhine
'

and the Meuse, no other watercourse but the Niers, he [Caesar] was
evidently obliged to advance to that river to find water '."^
Desjardins, who accepts Cluver's theory regarding the course of the
Meuse, nevertheless places the scene of the rout near Fort St. Andries ;

but he does not discuss the question. He merely observes, following

Cluver, that the Meuse and the Waal probably approached each
other above Bommel without actually mingling their waters, and
then separated, to flow, each in its own course, to the sea.^ But
if the rivers merely approached each other, Caesar would not have

written ad confluentem Mosae et Rheni.

^ Hist, of the Romans under the Empire, i, 453, note.

'
- B. G., iv, 11, § 3. Caesar, p. 191.
* Decline of the Romati Republic, pp. 157-8.
'"
The number is probably a gross exaggeration. fSee p. 243.
'^
Hist, de Jules Cesar, ii, 138-9, note.
s
' lb., p. 141. Oeoyr, dt la Gaule rum., ii, 052.
700 THE ROUT OF
Mr. A. G. Peskett's note ^ is worth transcribing.
—
Florus and'

Dion Cassius have been quoted in support of the Mosel but Florus :

only says iterum de Germano Tencteri querebantur. hie vero iam Caesar
ultro Mosellam navali ponte transgreditur ipsumque Rhenum et Her-
cyniis hostem quaerit in silvis, i, 45, § 14 (Halm) Dion Cassius says
:

that the Tencteri and Usipetes t6v re 'Pf/vov SiejSrjcrav koI h rrjv
Twv Tpyjovipwv ive/3aXov.^^ Florus says absolutely nothing about the
site of the battle, and Dion may have been thinking of B. G., iv, 6,
where Caesar says that the Germans got as far as the Condrusi,
who were the clients of the Treveri. . The Tencteri and Usipetes
. .

. . crossed the Rhine and wintered among the Menapii ... in

.

[ch.] 6 we read Germani latins vagabantur et in fines Eburonum

. . .

et Condrusorum, qui sunt Treverorum clientes, pervenerant taking ;

this passage in connexion with 12, where it is stated that they had
sent a great part of their cavalry on a foraging expedition among the
Menapii on the left bank of the Meuse, we naturally conclude that
the Germans were still somewhere in the neighbourhood of the
Meuse. . Caesar, in the passage above quoted, is doubtless speak-
. .

ing, as Kraner suggests, not of the whole body of the Germans but
of wandering predatory bands and there is no need to suppose that
;

even these had advanced further south than Aix-la-Chapelle, which

would be on the northern border of the Condrusi. Yet though every
hint given by Caesar up to the present chapter (15) points to the
Meuse as the scene of hostilities, we are suddenly transferred by
G5ler and Mr. Long to the Mayenfeld near Cologne though the
. . .

greater part of the cavalry, which must have formed no slight

element of strength in these vast predatory hordes, were left about
100 miles away among the Ambivariti.'
Now when Mr. Peskett concludes that the Germans were still
'

somewhere in the neighbourhood of the Meuse ', he is confusing

dates. So long as they remained in the country of the Condrusi,
they were no doubt in the neighbourhood of the Meuse but, as ;

I have already shown, whether the battle was fought near the con-
fluence of the Meuse and the Waal or near the confluence of the
Moselle and the Rhine, either the Germans had moved away from
the country of the Condrusi, or their main body had never gone near
that country. Undoubtedly they were in the neighbourhood of the
Meuse, if the rout took place near the confluence of the Meuse and
the Waal. But this is the very point which Mr. Peskett has to
prove. Moreover, Aix-la-Chapelle was nowhere near the Condrusi,
whose nearest territory was a good 25 miles south-west of it and ;

no unprejudiced critic can fail to see that when Caesar said that the
Germans had reached the country of the Eburones and Condrusi, he
meant the whole German host.
Still, if Mr. Peskett fails as an advocate, it does not follow that

^ Caesar, Bks. iv-v, pp. GO-1. Parts (which I have omitted) both of
Mr. Peskett's note and of the argument which he criticizes have been rendered
obsolete by the arguments of Meusel and Klotz, which show that B. G., iv, 10,
is most probably spurious.
^ crossed the Rhine and invaded the country of the Treveri."
'
 THE USIPETES AND TENCTERI 701

his cause is remains for me to deal with the arguments

had. It
which he has left Assuming, with Napoleon,^ that when
unnoticed.
the German convoys asked for three days' grace, they were at
Straelen, the distance, in a straight line, to the frontier of the Ubii
and back was not less than 100 miles,^ which, if the envoys had
meant what they said, their mounted messengers might have covered
in three days. But this calculation assumes that the chiefs of the
Ubii would have been found waiting on their northern frontier, and
that the business of negotiation could have been settled off-hand,
both of which assumptions are absurd. Mr. C. E. Moberly, indeed,
says,^ The difficulty would be hardly less on the other supposition,
'

as the German cavalry would then have had to get from the lower
Meuse to Coblenz in the same time.' But the German cavalry had
been expected for days past and the envoys, in asking for three
;

days' grace, might take for granted that they were already on the
road. It is possible,' Mr. Moberly proceeds, either that ad eas res
' '

conficiendas ^ may mean " in order to arrange for the embassy ", or
that the ambassadors wildly pressed for these three days without
considering what they promised to accomplish in them.' Both of
these alternatives appear to me extremely far-fetched.
Long's argument, that Caesar could not have attacked the Germans
near the confluence of the Meuse and the Waal, because he tells us '

nothing of a long march up the Rhine to make his bridge ', has con-
siderable force. It is true, as Napoleon points out, that, if he beat
the Germans near the confluence of the Moselle and the Rhine, he
must have made a long march across the Ardennes, and that of
that he tells us nothing either. Still, Caesar, who disregards facts that
are not essential to his narrative, might have omitted to mention
that he crossed the Ardennes for he tells us that he was marching
;

steadily and continuously on, and he gives us to understand that he

was marching towards the Rhine and he may have thought this
;

information sufficient. But he certainly leaves on our minds the

impression that he crossed the Rhine near the spot where he defeated
the Usipetes and Tencteri ^ it is certain that he crossed at least as
:

far south as Cologne ^ and we should have expected^him to tell us

;

that he had to make a long march southward to get there.

IV. And now I will do my best to seize the vital points in this
discussion. If we decide that the Germans were defeated near the
confluence of the Meuse and the Waal, (1) our decision harmonizes
with the reading ad confluentem Mosae et Rheni, which is found in
all the MSS. if (2) we assume that Rheni is equivalent to Vacali,
;

or else that confluentem means, as Heller supposes,"^ the connecting '

link between the Rhine and the Meuse, that is to say, the Waal
'
;

(3) we are obliged to assume that the Germans retreated before

Caesar, or on receiving the first news of his approach, from the
positions which they had taken up in the territories of the Eburones

^ Hist, de Jules Cesar, ii, 141. ^ See p. 697, n. 5.

3 Caesar, p. 2C6. « 5. G*., iv, 11, § 3.
' Jb., 15-0. « See pp. 706-10. ' See pp. 694-5.
702 THE ROUT OF
and Condriisi which they had recently
to the country of the Menapii,
left unless indeed
; we accept General Creuly's desperate theory,
that those who had penetrated into the territories of the Eburones
and Condrusi were only a reconnoitring party (4) we are obliged
;

to assume that when Dion Cassius said that the Germans invaded
the country of the Treveri, he either included the Condrusi and
Eburones among the Treveri or simply made a blunder ^ (5) we ;

are obliged, unless we accept Merivale's far-fetched suggestion, to

assume that, when the German envoys offered to send a message
to the Ubii and to get an answer in three days, they were offering to
perform a manifest impossibility, and that, unless they were talking
wildly, their sole object was to gain time, —
in which case they would
surely have asked for more than three days, unless indeed they knew
that their cavalry would rejoin them within that time (6) we are
;

obliged to assume that, after the defeat of the Germans, the Ubii
asked Caesar to march at least 70 miles up the valley of the Rhine
and to cross the river into their territory that he did so that he
; ;

then marched northward again into the country of the Sugambri

that he then marched southward again into the country of the Ubii
and all this in spite of the fact that his narrative leaves on our minds
the impression that he crossed the Rhine near the spot where he had
defeated the Germans (7) we are obliged (unless we accept Heller's
;

translation of confluentem) to assume, with Napoleon, that the con-

fluence of the Waal with the Meuse was at Fort St. Andries, because,
if it was at Gorkum or at Dordrecht or at Geervliet, the difficulty

regarding the proposed message to the Ubii and the difficulty regard-
ing Caesar's passage of the Rhine and his unrecorded march are
enormously increased.
The objections to the above-mentioned theory which I have
numbered 2, 3, 5, and 6 seem to me to be serious.
If, on the other hand, we decide that the defeat took place near
the confluence of the Moselle and the Rhine, (1) our decision flatly
contradicts the MSS. of the Commentaries. We
must assume either
that Caesar wrote Mosae when he meant Mosellae, or that he wrote
Mosellae, but that some blundering copyist put Mosae instead ;

(2) the difficulty about Rheni disappears we need not assume that
:

Caesar wrote Rheni when he meant Vacali or that confluentem means

connecting link (3) we must assume that the Germans moved
' '
;

away from the positions which they had taken up in the territories

Zeuss indeed {Die Deutschen und die Nachharstamme, 1837, i)p. 84, 216)
^

infersfrom Caesar's statement, that the forest of the Ardennes extended from
the Rhine through the heart of the territory of the Treveri to the frontier of the
Remi [silvam Arduennnm, quae per medios fines Tnverorum aflumine Rheno
. . .

ad initium Remorum pertinet. B. G., v, 3, § 4), that Bonn and Cologne may
have been in Treveran territory and accordingly Dederich argues that Dion
;

may have been right in saying that the Usipetes and Tencteri penetrated into
the country of the Treveri. But Zeuss' s remark is quite unfounded for the
;

Ardennes, as we know it, is partly in Luxemburg far to the south of Bonn and
Cologne, and the forest may, in Caesar's time, have extended even to Coblenz ;

and, moreover, it is improbable that the territory of the Treveri extended so

far northward even as Bonn.
 1

THE USIPETES AND TENCTEKI 703

of the Eburones and Condrusi, and that, instead of retreating in

the direction whence they had come, they strnck off, through the
Ardennes, towards the south-east, and phmged into the land of
the Treveri. (4) Wemust assume that the Germans had taken up
a position about 70 miles, in a direct line, from the nearest point of
the Meuse, to the further side of which they had sent their cavalry
on a foraging expedition (5) the difficulty regarding the offer of
;

the German envoys to send a message to the Ubii disappears ;

(6) the difficulty regarding the unrecorded marches of Caesar up and

down the valley of the Rhine also disappears.^
But to this view also there are serious objections. It is, of course,
conceivable that Caesar did write ad confluentem Mosellae et Rheni,
and that some meddling copyist altered Mosellae into Mosae but ;

it needs great confidence in one's own judgement, even with the

support, such as it is, of Dion Cassius and Florus, to assume that he
did so. The objection which I have numbered 4 appears to me
hardly less grave.
Nevertheless, I am inclined to conclude, not without hesitation,
that the confluence to which the routed Germans fled was the con-
fluence of the Moselle and the Rhine. What has principally led me
to this conclusion is the conviction, which deepens with each succes-
sive reading of Caesar's narrative, that the rout took place opposite
the country of the Ubii. First, we see the German hosts marching
southward into the district of Condroz and the neighbouring country
of the Eburones. A few marches further southward would bring
them to the neighbourhood of Coblenz. It seems to me quite in
accord with Caesar's manner that, having already told us that they
had moved far from the country of the Menapii, he should have
omitted to mention these marches but I cannot conceive that if
;

they had retreated northward 90 Roman miles,^ he would have

neglected to say so. Then their envoys ask for three days' grace to
negotiate with the Ubii and the inevitable conclusion is that the
;

Ubii were close by. Finally, the Ubii ask Caesar to bring his army
into their country and it is hardly credible that they should have
;

expected him to march 90 Roman miles before he could cross the

Rhine.^ Again, it is natural to suppose that, if the rout had taken
place near the confluence of the Meuse and the Waal, Caesar would
have written confluentem Mosae et Vacali and, after all, the
;

substitution of Mosae for Mosellae is only what might have been

expected from a meddlesome scribe who had read chapters 9 and 1

1 In the first edition

(p. 691) I said that
' we are obliged to assume that Caesar
mentioned by name, for the first and only time, so important a river as the
Moselle without one word of introduction'. But this difficulty has vanished.
For Caesar also mentioned the Meuse in chapter 9 without introduction and, ;

as he probably did not write chapter 10, the explanation which that chapter
would have supplied is wanting.
^ Napoleon III, Hist, de Jules Cesar, Planche 14,

^ lb. The march would have been diminished by nearly 20 miles if Caesar
made his first bridge not, as Napoleon thought, at Bonn, but at the most
northerly point at which it can on any reasonable theory be placed, namely,
Cologne.
704 THE EOUT OF
and probably also the interpolated chapter 10. Finally, Caesar tells
us that the fugitives who plunged into the stream were overwhelmed
'
by the force of the current (vi fluminis)
' and these words are
;

absurdly inapplicable to the waters that flow past Fort St. Andries.
My conclusion is merely tentative but of one thing I am sure
;
:

the Germans did not flee to the confluence of the Waal and the
Meuse. For they would have fled, not westward to Fort St. Andries
or Gorkum, but to the nearest point of the Waal. If, then, Caesar
wrote Mosae and meant the Meuse, we must accept Heller's explana-
tion of confluentem. In the next article, however, it will be seen
that the choice between Mosae and Mosellae is inextricably involved
in the question where Caesar bridged the Rhine.
V. It remains to notice the views of those commentators who
refuse to believe that the rout of the Germans took place either in
the neighbourhood of Fort St. Andries or in that of Coblenz. And
first of those who accept the reading ad confluentem. Mosae et Rheni.
1. According to the French Commission,^ the Germans, when they

were attacked, were in the neighbourhood of Gorkum, on the left

bank of the Meuse, and had the estuary of Bies-Bosch on their left.
This theory, which is based on the erroneous assumption that the
Germans crossed the Meuse (regarded as a branch of the Rhine)
below Gorkum, requires us to place the Ambivariti on the right bank
of the Meuse, a view which I have shown to be wrong.^ Moreover, I
doubt whether Caesar's cavalry could have acted in a region which was
at that time probably a vast swamp and on the theory of the Com-
;

mission the difficulties involved in the assumption that the rout took
place at a point remote from the country of the Ubii are greatly increased
2. According to von Cohausen, who is followed by von Kampen,
the German encampment was near Wissen, about 10 miles south-
east of Goch ;and the beaten host fled to the Rhine, which they
reached at a point near the Cranenburger Bucht, about 5 miles west
of Cleve. At this point, says von Cohausen, the Meuse appeared to
join the Rhine, because the Bucht was inundated by water from the
latter.^ But von Cohausen's suggestion that Caesar was misled on
this point by the report of his cavalry is nothing but a wild guess.
According to Dederich,^ indeed, the Cranenburger Bucht is actually
inundated, early in the year, by water from the Rhine and the
;

inundation is only divided from the Meuse by a spit of land not

more than 1,000 paces wide :^ but granting the correctness of his
statement, it is hardly necessary to point out that Caesar would not
have described the Cranenburger Bucht as confluens Mosae et Rheni
or, indeed, as a confluence of anything with anything else.
3. Dederich ^ infers from a statement of Dion Cassius that the
'^

German camp, which, like Napoleon, he places north of Goch, must

Diet. arch, de la Gaule, i, 52.
^ ^ See
pp. 368-70.
Jahrb. d. Vereins von Alterthumsfreunden im Bheinlande, xliii, 1867, pp. 44, 50.
^

* Julius Caesar am Rhein,

p. 35.
^ See Neue Jahrb.
J. Philologie, &c., xcv, 1867, p. 43.
^ Julius Caesar am Rhein,
pp. 8, 28-31, 33-4.
' xxxix, 48, § 2. Dion, whose authority is nil, merely says that Caesar
reached the German camp at midday.
 THE USIPETES AND TENCTERI 705

have been within one and a half Stunde ', or about 4 Eonian miles,
'

from the corifluens Mosae et Rheni to which the Germans fled. It

will be objected, he says, that at this distance there is no confluence
of the Meuse and the Khine. But the answer is easy. It was not
Caesar himself but his cavalry who pursued the Germans. Caesar
mistook the bifurcation of the Ehine for a confluence of the Meuse
and the Khine the Meuse was near the Rhine, and this led him
:

astray.
The objection which Dederich anticipates and attempts to answer
is unanswerable. How any man in his senses could delude himself
into the fancy that Caesar should have imagined that the bifurca-
tion of the Rhine was a confluence of the Meuse with the Rhine,
passes one's power of conception. Dederich's theory can only be
reconciled with Caesar's narrative by assuming with Heller that
confluentem Mosae et Rheni means the Waal.
4. T. Bergk, who argues that confluentem meant a river which
'

flows into another river ', has made an attempt to solve the problem,
which is at all events original and ingenious.^ He holds that what
Caesar really wrote was simply cum ad confluentem Mosae pervenis-
sent (' when they had reached a point where a river flows into
the Meuse ') and that some commentator, unable to understand the
;

meaning of these words, added et Rheni. If, he says, it be objected

that Caesar would have given the name of the other river, the answer
is easy. Such reticence was one of Caesar's characteristics. He
knew the weaknesses of the general reader and he was too wise to
;

trouble him with outlandish names.

Having amended the text to his own satisfaction, Bergk proceeds
to identify his confluens. The German camp, he says, was at Heins-
berg, on the left bank of the Roer and the Germans fled to the
;

confluence of the Roer and the Meuse. In the narrowing peninsula

formed by these two rivers Caesar's cavalry could have acted with
effect and they could not have done so near the confluence of the
;

Meuse with any other stream which joins it in the country of

the Condrusi or of the Eburones.
General von Veith agrees with Bergk in deleting et Rheni after ad
confluentem Mosae, but differs from him in details. He believes
that the scene of the German rout was Louveignez, between the
Ourthe, which joins the Meuse at Liege, and the Vesdre, which flows
into the Ourthe about a mile and a half south-east of the same
town.2
Now the reason why Bergk thought it necessary to amend the text
was simply that he regarded it as an inevitable inference from Caesar's
narrative that the battle took place in the country of the Eburones
or in the country of the Condrusi and I have shown that this
;

inference is not inevitable. I freely admit that Caesar was sparing in

his use of geographical names but Bergk has wholly failed to prove
;

'
Zur
(JeacJi. und ToiMxjr. d. llhtinlande in rdm. ZeU, pp. 7-8.
Pick's Monalsschrijl /. d. (Jcicli. WcstdeiUscJd., vi, 100-30, referred to in
2

Jahresh. d. Gedchicht^ivis-scnuchaft, Jahrg. 3, 1880, ii, 2.

1003 Z Z
706 WHERE DID CAESAR
that conjluens was ever used in the sense which he claims for it or
that it was not used in classical Latin in the sense of confluence '. '

In fact, he begs the question, which is whether Caesar did not him-
self use the word in this very sense.

WHERE DID CAESAR MAKE HIS FIRST BRIDGE

OVER THE RHINE ?
This question is inseparably connected with that which I have
discussed in the preceding note. If the rout of the Usipetes and
Tencteri took place near the confluence of the Meuse and the Waal,
it is natural to look for the site of the bridge at a position as near
that spot as may be consistent with Caesar's narrative. If, on the
other hand, the rout took place near the confluence of the Moselle
and the Rhine, the bridge must have been made in the neighbour-
hood of Coblenz.
1. Napoleon^ fixes on Bonn, for the following reasons. {a) The
bridge joined the western bank to the territory of the Ubii.^ Two
years later Caesar built another bridge faulum supra a little higher —
—
up the river than the site of the former one which connected the
territory of the Treveri with that of the Ubii.^ Therefore it is almost
certain that the western end of the older bridge also touched the
territory of the Treveri. (6) Those critics who have chosen Andernach
as the site of the former bridge argue on the hypothesis, which
Napoleon claims to have disproved, that the defeat of the Usipetes
and Tencteri took place near the confluence of the Moselle and the
Rhine, (c) Cologne, which has also been adopted, is too far north
to agree with the Commentaries for when, two years later, Caesar
;

had recrossed bis second bridge, he traversed the Ardennes, passed

near the country of the Segni and Condrusi, and marched for Atua-
tuca,^ which Napoleon identifies with Tongres but, if he had started
:

—
from Cologne or, as Napoleon ought to say, a little above (paulum
—
supra) Cologne he would not have passed near the country of those
tribes, (d) South of Bonn, as far as Mainz, the Rhine flows upon
'

a rocky bed, where the piles could not have been driven in, and
presents, owing to the mountains which border it, no favourable
point of passage.' (e) Less than fifty years after Caesar's campaigns,
'

Drusus, in order to proceed against the Sicambri that is, against —

the same people whom Caesar intended to combat crossed the —
Rhine at Bonn.'
The first of these arguments tells against Napoleon's theory for ;

it is unlikely that the territory of the Treveri extended as far north-

ward as Bonn,^ whereas they certainly did possess the western bank
of the Rhine between Coblenz and Andernach. The second I have

' Hist, de Jules Cesar, ii, 143, n. 3. ^ b. G., iv, 16, §§ 5-8 ; 19.
3 lb., vi, 9, 29, § 2. * Ih., vi, 29, § 4 ; 32, §§ 1-3.
^ See p. 481.
 ^

BRIDGE THE RHINE ? 707

dealt with in the preceding note. The third, which rests upon a
doubtful hypothesis,! is irrelevant for, supposing that the first
;

bridge was at Cologne, the second was probably at Bonn and if ;

Caesar had marched from Bonn to Tongres, he would have skirted

the country of the Segni and Condrusi. Napoleon would, I think,
have been wiser if he had based his objection to Cologne on the
ground that it was probably opposite the territory, not of the Ubii
but of the Sugambri. And when he asserts that the bridge could
not have been constructed south of Bonn, he surely forgets that, on
his own theory, Caesar's second bridge was a little south (pauluni
supra) of Bonn. Moreover, it is certain that the bed of the Rhine
between Coblenz and Andernach would admit of the construction of
a bridge such as Caesar describes.
2. M. F. Essellen ^ argues that the first bridge was built at or near
Cologne. He says first, that it may be regarded as certain that in
Caesar's time the Ubii possessed a strip of land between the mouth
of the Wupper and the mouth of the Sieg and between the Rhine and
the territory of the Sugambri and secondly, that if the bridge had
;

been built at any point between Cologne and Coblenz, Caesar, in

order to reach the country of the Sugambri, would have been obliged
to take the road leading through narrow passes past Linz, Unkel,
and Konigswinter, thus exposing himself to attack from the Suebi,
who inhabited the mountains on the right. Neither of these argu-
ments has any value. Essellen offers no evidence or even argument
in support of his assertion that the territory of the Ubii extended as
far northward as the mouth of the Wupper and if Caesar had
;

crossed the Rhine at Bonn, he would not have had to march a single
yard by the road to which Essellen alludes, Konigswinter, the most
northerly of the three places which he mentions, being at least
6 miles south-east of Bonn. Moreover, the mountains in question
belonged, not to the Suebi but to the Ubii, who were Caesar's humble
servants.^
M. Jullian,^ who has revived the case for Cologne, argues that
Caesar did not build his bridge on the territory of the Treveri, whom '

he does not mention and in whom he had no great confidence '

;

that, coming from the north, he would not have built it higher up
the Rhine than was necessary and that he would have chosen
;

a site as close as possible to the southern frontier of the Sugambri.

Certainly he had no great confidence in the Treveri, who, as he says,
'
would not attend his councils or submit to his authority ^ never- '
:

"^
theless he unquestionably built his second bridge in their territory ;

and the fact that he does not mention them in connexion with the
first proves nothing, for he does not mention any Gallic tribe.

1
See pp. 371-83.
*
See von Goler's Gall. Krieg, 1880,
p. 123, and Long's Decline of the Roman
Republic, iv, 161, n. 8.
3
Zur Fragcy wo Julius Cdsar die beiden Rheinbn'icken achlagen liess, 1804,
pp. 9, 11.
'
E.G., iv, 8, § 3 ; IG, §§ 5-8. ^ ^,^-^.^
^^ i^ Gaule, iii, 331, n. 9.
«
B. G., V, 2, § 4. 7
11,^ vi, 9, § 5.
Z Z 2
708 WHERE DID CAESAR
M. JuUiaii's other arguments follow naturally from his conclusion
that the Usipetes and Tencteri were defeated near the Meuse. But
he has left one grave objection unanswered. He places the second
bridge at Bonn ^ and on his theory it could not be placed further
;

up the stream." But it is most improbable that the territory of the

Treveri extended so far northward and M. Jullian himself, in his
;

SK'ond volume,^ supposes that it did not extend beyond Andernach.

3. Long,* who holds that the rout of the Usipetes and Tencteri
took place near the confluence of the Moselle and the Rhine, natu-
rally goes on to express his belief that the bridge .was made between
(.•oblenz and Andernach. Caesar,' he says, could not cross above
' '

Coblenz, if the battle was fought where I have placed it, without
crossing the Mosel nor is the river practicable above C'oblenz
; . . .

nor could he cross below Andernach till he came to Bonn. He must

therefore have crossed between Andernach and Coblenz or at some
place near Bonn, or lower down. But he crossed in the country
. . .

of the Treveri, and we cannot make the Treveri extend further north
than Andernach, or, at the most, the valley of the Ahr.'
Assuming that the battle took place near the confluence of the
Moselle and the Rhine, there is no fault to find with any of these
arguments, except the one which is founded upon the assumption
that Caesar crossed the Rhine from the country of the Treveri, and
that the country of the Treveri extended no further north than the
valley of the Ahr. It is true that when Caesar made his second
bridge, he was in the country of the Treveri but it is not certain
;

that he was there when he made the first. It is also probable that
the Ahr, if not the Vinxtbach, was the northern frontier of the
Treveri ^ but we cannot be sure. But even if Long's assumptions
;

are wrong, his error is unimportant. Everybody who accepts his

view regarding the place of the rout of the Usipetes and Tencteri will
admit that Caesar would not have needlessly marched as far north
as Bonn, when, between Coblenz and Andernach, a site, or rather
a variety of sites equally practicable, lay ready to his hand.
M. Jullian,^ who does not accept Long's premiss, argues that if
the second bridge had been built near Coblenz, it would have been
too far from Atuatuca, which he identifies with Tongres and regards
as Caesar's base of operations, and that the German territory opposite
Coblenz was not rich enough to support his army. But how can
Atuatuca, which was in the country of the Eburones, be regarded
as Caesar's base of operations, seeing that immediately before he
crossed the Rhine he had marched from the country of the Menapii
'

to that of the Treveri '.'^ Major-General Wolf, to whom M. Jullian

appeals, says that in order to reach the Rhine near Andernach,
^ Hist, de la Gaule, iii, 400, n. 3. * See B. G., vi, 10, § 3.

3
p. 478, n. 1. Major-General Wolf {Beihefle zum Militdr-Wockenhlait, 1901,
pp. 42-4), who places the second bridge only 3 kilometres above Cologne,
actually argues that the territory of the Treveri was conterminous with that
of the Menapii
« Caesar, p. 104. ^ JSee p. 481.
6 Hist, de la Gaule, iii, 400, n. 3.
' B. G., .vi, 9, § 1.
 —
BRIDGE THE RHINE ? 709

Caesar would have had to pass through a defile between mountain

and stream, and that, cut off from communication with his rear, he
would have been unable to feed his troops. In these circumstances,
he says, it is inconceivable that he would not have been attacked,
in Germany by the Suebi, on his return by the Treveri.^ I do not
deny the force of these objections but I would point out that
:

when Caesar had crossed the Rhine he was in the territory of the
friendly Ubii, upon whose resources he could count that he did ;

actually fear that he would be unable to procure sufficient corn, and

accordingly gave up his intention of penetrating far into Germany ^ ;

that he took every precaution to secure his rear ^ and that it was ;

obviously not the policy of the Suebi to attack him, but rather to
lure him on.* What impels me to prefer, though doubtfully, the
neighbourhood of Coblenz is the fact that Caesar's chief object in
invading Germany in 53 B.C. was to prevent the Germans from
reinforcing the Treveri, whose territory most probably did not extend
as far northward as Bonn, and whose political centre, Treves, was
far south of Bonn, and even south of Coblenz. M. Jullian himself,^
rightly in my opinion, holds that Labienus, when he was sent in
56 B.C. to prevent the Germans from joining the Treveri, took up
his quarters at Treves.^
4. Von Goler"^ and Heller ^ believe that the second bridge was
built at some point in the course of the river where there is an island.
General Creuly ^ remarks that, on this theory, there would have been
two bridges, not one, a fact which Caesar would have mentioned ;

and he adds that when Caesar says that, after recrossing the bridge,
he broke down, for the length of 200 feet, that part of it which
touched the German bank, and constructed a tower on the end in
the stream,io he plainly gives us to understand that there was no
island. I do not agree with him.^^
5. A. von Cohausen i^ holds the amazing opinion that the first
bridge was made at the foot of the Fiirstenberg, near Xanten. This
place is at least 90 miles, in a direct line, from Neuwied, where
Cohausen believes that Caesar made the second bridge and Caesar ;

says that this second bridge was only a little above ' the first one.
'

But this does not shake the conviction of Cohausen. As Caesar says
that the Usipetes and Tencteri crossed the Rhine not far from the '

sea '
(non longe a mari, quo Rhenus injiuit)^^ and as, according to
Cohausen, the point where they crossed could not have been less
than 90 miles from the sea, he concludes that Caesar might have
described Neuwied as only a little above Xanten. i* The answer
' '

^ Beihefte zum Militar-Wochenhlatt, 1901, pp. 42-4. Wolf was referring to

the first bridge but his arguments would apply equally to the second.
;

E.G., vi, 29, § 1. 3

ji,^ 9^ § 5_ 4
ji^^ 10, §§ 4-5.
I
'^
Hist, de la Gaule, iii, 311, n. 5.
6 B. G., iii, 11, § 1. 7 Gall. Krieg, 1880,
p. 214.
8 PUlologus, xlix, 1890, p. 695.
9 Rev. arch., nouv. ser., viii, " B. G., vi, 29,
1863, p. 388. § 3.
" 8ee Long's Decline of the Roman Republic, iv, 243, note.
12 Casar's Rheinbriicken, '" B. G., iv,
1867, pp. 7-11. 1, § 1.
1* Cohausen
tries to bolster up his theory by referring to a passage in B. G.^ii,
35, § 3, which has given commentators a great deal of unnecessary trouble. The
710 WHERE DTD CAESAR BRIDGE THE RHINE ?

is (1)that, in relation to the whole length of the Rhine, the point

where the Usipetes and Tenctcri crossed might be said to be not '

far from the sea and (2) that Caesar's information about the lower
'
;

Rhine was confessedly vague. Moreover, there is one fact which

simply pulverizes Cohausen's theory. It is certain that Caesar's first
bridge touched the western frontier of the Ubii. Everything points
to this conclusion. The Ubii begged Caesar to cross the Rhine into
their country. After he had punished the Sugambri, he returned to
the country of the Ubii (se in fines Ubiorum rece^nt) and, having;

obtained from them information as to the movements of the Suebi,

he immediately recrossed the Rhine.^ It is clear then that, when he
was about to set foot on the bridge to recross the Rhine, he w^as
standing upon Ubian territory. Now the northern frontier of the
Ubii, trace it as far north as we possibly can, was many miles south
of Xanten.
I conclude, doubtfulty, that both the first and the second bridge
were built between Andernach and Coblenz, but that there is no
evidence for defining their positions more exactly .^

passage, as it stands in (jt-n, runs as follows

: ipse in Carnutes, Andes, Turonosque,
quae civitates propinquae his locis erant uhi helium gesserat, legionibus in hiber-
nacula deductis, in Italiam profectus est. Now Caesar had been making war
in tlie country of the Belgae ; and the nearest of the territories which he men-
tions in this passage was more than 40 miles from the nearest point of the
theatre of the war. Therefore, triumphantly concludes Cohausen, panlum
supra may mean 90 miles. To which I reply that, if Caesar really wrote the
passage as I have printed it, either his memory or his geographical knowledge
was defective, unless he was only thinking of the distance between the south-
western /rowifter of the Bellovaci and the north-eastern frontier of the Carnutes,
which is not more than 12 miles. It has been suggested that if we read (Turonos)
quaeque, which is found in xp? for quae, the difficulty disappears. K. Thomann,
however {Der franzosische Atlas zu Cdsars gall. Kriege, pp. 12-3), thinks that
the reading quaeque is unsatisfactory for, he argues, if Caesar had sent his legions
;

into the territories of any other peoples besides the Carnutes, Andes, and
Turoni, he would have mentioned them by name. If Thomann had looked
at B. G., iii, 27, he might have withdrawn this objection. His own suggestion
that uhi helium gesserat means uhi helium et per se et per Crassum gesserat is futile ;
for there is no evidence that Crassus had made war at all. He appears simply
to have received the submission of the maritime peoples {B. G.,u, 34). M. Jullian
{Hist, de la Gaule, iii, 270, n. 1), who adopts the reading quaeque, holds that
Caesar was referring to the Ambiani, Atrebates, Bellovaci, and Suessiones.
If so, his locis uhi helium gesserat does not mean the whole theatre of the Belgic
campaign, but only points to those tribes in whose territories there was actual
fighting, namely the Remi, the Nervii, and the Atuatuci.
'
B. G., iv, 16, §§ 5-8 ; 19.
2 In the smaller edition of this book
(p. 75, n. 1) I incautiously committed
myself to the view that a Roman camp recently discovered on the left bank
of the Rhine, near Neuwied {Bonner Jahrh., civ, 1899, pp. 1-55), was the camp
which Caesar constructed after his second passage of the river ; but Professor
Haverfield pointed out to me that later investigations had suggested grave
doubts. In tlie first edition (pp. C95-6) I mentioned the discoveries or alleged
discoveries of the remains of old bridges near Bonn {Philologus, xlix, 1890,
p. C95) and at Engers, between Coblenz and Andernach (J. N, Hontheim,
Prodromus Hist. Trev., 1757, i, 209) ; but there is no evidence that either of
them was made by Caesar.
 711

CAESAR'S BRIDGE OVER THE RHINE

The bridge was constructed as follows. First of all, a couple of
I.
balks, a footand a half square in the thickness, and sharpened at the
lower ends, were fastened together by cross-pieces, which kept them
exactly two feet apart. They were then driven by pile-drivers into
the bed of the river, in a line at right angles to the bank, and sloping
in the direction of the current. Forty feet lower down, another
couple of balks, similarly joined together, were driven in exactly
opposite to them, and sloping towards them. A beam two feet wide
was laid across from one couple to the other, being let into the space,
which it exactly filled, between the two balks of each couple. The
couples were kept apart by means of two pairs of braces {fibulae), each
pair being attached to the upper ends of the balks and to the ex-
tremity of the beam. The effect of the whole arrangement was that,
the greater the force of the current, the more firmly were all the
timbers held together. Similar couples of balks, with similar trans-
verse beams, and similarly kept apart by braces, were driven, in
a straight line with the two original couples, right across the stream.
Then, to form the roadway, timbers were laid, in the direction of the
bridge, over the transverse beams, and covered by a second layer of
smaller joists laid transversely, above which was placed a third layer
of fascines. The bridge was strong enough already but, to make
;

assurance doubly sure, piles were driven vertically into the bed of
the river along the down-stream side of the bridge. They were
arranged in diagonal groups corresponding with the several couples
of balks in the main structure, and were connected therewith. On
the up-stream side, piles were driven in vertically at a short distance
from the bridge, to protect it from the shock of any timber or vessels
which the Germans might float down to destroy it. These piles were
arranged in triangular groups, each group being exactly opposite to
each couple of balks in the main structure.^
The foregoing description represents the view which I have myself
formed, after studying Caesar's description and the remarks of the
numerous editors and modern writers on the subject. But to a
modern reader Caesar's description is, on certain points, obscure ;

and for the present I have necessarily left my explanation incom-

plete.
A. Rheinhard, finding Caesar's description wanting in lucidity,
argues that he had no knowledge of engineering that, his express
;

statement notwithstanding, he did not himself design the structure ;

that he borrowed the materials of his description from his chief

engineer, Mamurra and that,
; as he failed to understand them, it was
jejune, loose, and merely intended for popular reading.^ Lieutenant

1 B. G., iv, 17.

2 Caesar's Rheinhriiche, 1883, pp. 6-9. There is no evidence that Mamurra
was Caesar's chief engineer '. He was, as far as we know, simply one of the
'

praefecti fabrum.
712 CAESAR'S BRIDGE
A. Schleussinger replies that Caesar would not have dared, even if
he had wished, to rob Mamurra of his credit, for fear the truth should
come out that he did, as a matter of fact, give his lieutenants the
;

credit which was their due and that he must have known that pro-
;

fessional readers at Rome would study his description in a critical

spirit.^ I agree with Schleussinger and it appears to me most
;

unlikely that a man like Caesar should have been ignorant of the
principles of a most important branch of his business.
II. J. Rondelet holds the singular opinion that the balks of each
couple in the main structure were not two feet apart the whole way
down, but only at the top. He bases his opinion on the words j)rone
acfastigate. The word prone, he thinks, shows that the balks sloped
in the direction of or against the current, as the case might be and ;

the word fastigate that they sloped inwards towards each other.^
But this theory is disproved, first, by the fact that Caesar says that
the balks were joined by cross-pieces at an interval of two feet ^ ;

and secondly, by the fact that Rondelet's arrangement would have

greatly increased the difficulty of driving them into the river-bed.^
As Schneider ^ says, explaining/as/i^a^e, each couple of balks was so
placed that it formed, with the opposite couple, a figure resembling
a gable or sloping roof.
The theory of 0. Pohl,^ who represented the two balks of each
couple as crossing each other, two feet above the water in the shape
of an X, and that of T. Maurer,*^ who argued that they were placed
one behind the other, hardly require even a passing mention.
III. M. Sonntag^ contends that all the plans of the bridge, except
his own, exaggerate the inclination of the balks. The result, he says,
of their having inclined towards each other at the angles which most
of the plans represent, would have been (1) that the pile-drivers
would have worked less effectively (2) that the balks would not
;

have been firmly fixed and (3) that the obtuse angles which they
;

formed with the transverse beam would have been too great to admit
of the insertion of the fibulae. (It will be observed that Sonntag here
begs the question, how the fibulae were fixed.) The inclination, he
maintains, would have been such that, if the distance along the
surface of the water between the opposite couples of balks had been,
as Caesar says, 40 feet, the distance, 6 feet from the surface, would
have been 39.
^ Studie zu Caesar s Rheinhrilcke, 1884, pp. 25-30.
^ Traite theorique et pratique de Fart de hdtir, 1812-4, iv, 305-7.
^ Tigna bina sesquipedalia paulum ab irao praeacuta dimensa ad altitiidinem

fluminis intervallo pedum duoruni inter se iungebat. Haec ciini machina-

tionibus immissa in flumen defixerat fistucisque adegerat, non subhcae modo
derecte ad perpendieulum, sed prone ac fastigate, ut secundum naturam
fluminis procumberent, iis item contraria duo ad eundem modum iuncta
intervallo pedum quadragenum ab inferiore parte contra vim atque impetum
fluminis conversa statuebat.
* Glass. Rev., xvi, 1902, ^ Caesar, i, 353.
p. 30.
^ Festschr. d. Realgymnasiums am Zwinger zu Breslau, 1886, reviewed in
Jahresb. d. philol. Vereins zu Berlin, xiii, 1887, p. 367.
' Noch einmal J. Cdsars Briicke iiber d. Rhein, 1883, p. 4.
^ Beinerhungen zu Caesar
de B. 0., iv, 17, 1890, pp. 4-5.
 ^

OVER THE RHINE 713

IV. According to Rheinhard,i the balks were lowered separately

into the water after they had been rammed in, the cross-pieces were
:

added then the transverse beam (trahs), and then the hnxces (fibulae).
;

Most commentators, however, hold that the balks were coupled before
they were lowered into the bed of the river and this appears to be
;

the only conclusion which can be drawn from Caesar's words.

V. Again, there is a difference of opinion as to the distance between
the opposite couples of balks. After describing the couple which was
first lowered into the water, Caesar writes that he had another
'

couple, similarly joined together, planted opposite them, on the lower

side, at a distance of 40 feet (iis item contraria duo ad eundem 7nodum
'

iuncta intervallo 'pedum quadragenum ah inferior e parte statuehat).

. . .

G. Hubo argues that the distance of 40 feet could not have been
measured along the bed of the river, on account of the difference
between the depth near the bank and in mid-stream, which would
have caused a variation either in the width of the roadway or in
the angle formed by the balks. Nor, he argues, could the measure-
ment have been made between the tops of the balks, because trans-
verse beams 40 feet long would have been difficult to procure and
could not have sustained the weight which the bridge had to bear,
and because the difficulty of building the bridge on such a scale
would have been very great. He concludes that the distance must
have been measured along the surface of the water. But, he adds,
Napoleon is wrong in measuring the 40 feet so as to include the
thicknesses of the two opposite couples of balks. In that case
the intervallum of which Caesar speaks would have been only 37, not
40 feet and it is clear from Caesar's words, tigna hina sesquipedalia
;

paulum ah imo praeacuta dimensa ad altitudinem fiuminis intervallo

pedum duorum inter se iungehat,^ that he used intervallum in its
strict sense.
Now there can be no doubt that if Hubo is right in his main con-
tention, his correction of Napoleon is justified ; and that his first
argument is sound is proved by Caesar's statement that the balks
Were made of varying lengths to suit the varying depth of the stream
{dimensa ad altitudinem. fiuminis). According to Mr. E. Kitson
Clark,* 40 feet was the distance between the lower ends of the balks
sunk in the bed of the river. This view is of course equally open to
Hubo's criticism, which cannot be rebutted unless we assume that
40 feet was only a rough general measurement. A. Eberz,^ indeed,
holds that the distance of 40 feet was measured between the tops of
the balks, that is to say that the roadway of the bridge was 40 feet
wide, because, according to Riistow,^ the breadth of a cohort marching

^
Caesar's Rheinhriicke, 1883.
2 '
He took a couple of balks a footand a half thick, had them sharpened to
a point from a little above the lower end and adapted in length to the [varying]
depth of the river, and fastened them together at an interval of two feet.'
=*
Neue Jahrb. f. Philologie, &c., cxlv, 1892, pp. 485-92.
* Class. Rev., xxii, 1908^ p. 145.
•^
Neue Jahrb. f. Philologie, &c., Ixxv, 1857, pp. 849-50.
•^
Heerwesen und Kriegfi\hrung C. Julius Ciisars, 1855, p. 01.
714 CAESAR'S BRIDGE
in column was just 40 feet. A weaker argument could hardly be
imagined. Caesar was not bound by red tape it was necessary for
:

him to build his bridge at the least possible cost of time and labour,
and not at all necessary to adhere pedantically to rules of formation ;

and nothing could have been easier than to contract the breadth of
the column while it was marching across the bridge. Moreover, it
cannot be proved that the breadth of a cohort marching in column
was 40 feet. But there is not much force in Hubo's second argu-
ment for if the distance, measured along the surface of the water,
;

between the opposite couples of balks had been 40 feet, the length
of the transverse beams would only have been a little less and the ;

difficulty of finding beams of the required length would surely have

been almost as great as if they had been fully 40 feet long.
VI. The next difficulty concerns the braces, which Caesar calls
fibulae, —
their exact nature, number, and arrangement.
— I pro-
visionally translate his statement beams two feet wide, fitting
:
'

into the interval between the balks of each couple, were laid across ;

and the two couples were kept apart by a pair of braces on either
side at the extremity. As they were thus kept asunder, and on the
other hand lashed together, the strength of the structure was so
great and its principle so ordered that, the greater the force of the
current, the more closely were the timbers held together (haec '

utraque insuper hipedalibus trahihus immissis quantum eorum tigno-

rum iunctura distahat, hinis utrimque fihulis ah extrema parte dis-
tinehantur (v. 1. destinahantur) ; quihus disclusis atque in contrariam
partem revinctis tanta erat operis firmitudo atque ea rerum natura
ut quo inaior vis aquae se incitavisset, hoc artius inliqata tenerentur)?-
The various reading is comparatively unimportant. Caesar may
have written destinahantur ('were made fast'), the reading of E — \

but anyhow the opposite couples of balks could not have been made
fast unless they were kept at the right distance apart.
Mr. J. H. Taylor insists that distinehantur would yield no useful
'

meaning, whether it was held to be predicated of the tigna which

form the twin-pier [or couple of balks], or of the two twin-piers
which one trahs joined because the former were already securely
;

fastened, two feet apart, by their iuncturae the latter were suffi-
;

ciently separated by being driven into the river-bed '.^ But this
reasoning cannot be accepted for, as the
; accompanying diagram,
prepared by my friend Mr. Stanley Hall, A.R.I.B.A,, will show, the
tendency of the current would have been to make the upper couple
of balks AB, and, in a less degree, the lower couple CD rotate about
the fulcra A, D, thus depressing B, elevating C, and diminishing the
horizontal distance between the upper ends of the couples.
The question of the fihulae has been debated, sometimes with
acrimony, since the Renaissance and, as P. de la Ramee said in
;

the days of Queen Elizabeth, in the whole construction of the

'

^ As will be seen presently, the meaning not only of fihulis, but also of binis,
ufrimque, ab extrema parte, quihus, and in contrariam partem has been disputed.
Class. Rev., xvi, 1902, p. 33.
''

I
 — '

OVER THE RHINE 715

bridge nothing is more obscnre '

:
^ but the truth has gradually been
brought to light.
1. Lipsius- conceived of the fibulae as iron bolts, driven right

through the transverse beam, one on the inner side, the other on
the outer side of each couple of balks.
2. Kraner and Heller believe that there were four bolts to each
couple of balks. Kraner^ holds that they were driven, at each end
of the transverse beam, through the transverse beam itself into the
'
cross-piece ' immediately underneath it. Heller considers that two
were driven horizontally through the transverse beam on the outer
side, and two on the inner side of each couple of balks. He defends
his view by the following reasoning utrimque must be understood
:

as referring to the inner and the outer side of each couple of balks ;

for on any other theory it must be regarded as a mere repetition of

(haec) utraque. The opposite couples of balks were kept apart (distine-
hantur), he explains, by the inner fibulae while they were prevented ;

by the outer fibulae from moving too far apart (revinciebantur) .*

Heller is a thorough Caesarian scholar but on this point he is ;

mistaken. Assuming that Caesar wrote distinebantur Heller has no ,

right to limit the application of the word to the inner fibulae for ' '
;

Caesar says that the desired efEect was produced by all. As to

utrimque^ who can doubt that Caesar used the word in order
to make it clear that each couple of balks was provided with two
fibulae ? ^ Read the sentence, in this sense, without the word, and
you will see that it is needed.
3. It is amusing to read Frohlich ^ after reading Heller. Caesar,
he argues, did not write distinebantur, but destinabantur but the ;

reason which he gives is unsound. Distinebantur, he insists, makes

P. Rami
^ liber de Caesaris militia, 1559, pp. 17-8.
. . .

Opera, od. 1637, iii, 306 and illustration on p. 308.

^

3 Caesar, 1855,
pp. 158-9.
* Philologus, Suppl. Band v, 1889, pp. 386-8, and Philologus, x, 1855,
pp. 732-4.
^ Mr. Roby {Class. Rev., i,
1887, p. 242) takes the same view, which makes
Heller exclaim {Philologus, xlix, 1890, p. 694), it is inexplicable how the English
'

"
scholar, when he finds in Caesar's text the words " binis utrimque fibulis
can get out of them the meaning "one on each side"'. Mr. Roby might
retort that one on each side (of each couple of balks) means hvo at each end
'
' '

(of the transverse beam). Where Mr. Roby really differs from Heller is in his
interpretation of utrimque ; and if, as I believe, his interpretation of utrimque
is right, so is his interpretation of binis.
^ Das Kriegswesen Cilsars, 1890,
pp. 216-9.
716 CAESAR'S BRIDGE
no srtiise for, it' ( ^aesar had written it, we should be obliged to con-
;

clude that flhulae were placed on the inner side of the balks, in order
to keep the opposite couples apart. But this conclusion is negative d
by the words ah extrema "parte, which can only mean on the outer '

side ' (of the balks). In support of this view, Frohlich says that in
the other passages in which Caesar uses the word extremus, a double '

standpoint cannot be understood. The fibulae therefore, which, as

'

the derivation of the word proves, were bolts, were inserted in the
transverse beam, outside the balks, two to each couple. Their object
was to prevent the couples of balks from moving apart. The couples
could not move towards each other, because the transverse beam was
notched at the points of contact. In fine, the transverse beam kept
the couples of balks apart (quibus disclusis) while the fibulae kept
;

them sufficiently close together.

Now the unbiased reader who is familiar with Caesar's use of the
word extremus will have no hesitation in concluding that by extrema
parte he simply meant at the top of the balks ', or (which comes to
'

the same thing), at, or near, the end of the transverse beam.' ^ Any
'

one who has seen the collections of fibulae in the British Museum
will smile at the suggestion that the derivation of the word proves
that Caesar's fibulae were bolts. And if the transverse beam was
notched at the point of contact with the balks on the inner side of
them, it is hard to see why it was not notched on the outer side as
well ;and why, if notching was sufficient, fibulae were used at all.
I should add that an engineer whom I have consulted regards every
theory which identifies the fibulae with bolts as wholly unpractical,
and, as Sonntag justly remarks, the pressure of the current would
'^

have made bolts not firmer, but looser.

4. A. von Cohausen,^ who holds that there were only two fibulae
to each couple of balks, thinks that one of the two served as a cross-
piece outside the couple, for the transverse beam to rest upon that ;

the other was let into a groove cut in the upper surface of the trans-
verse beam and on the inner side of the couple and that balks,;

transverse beam, and fibulae were firmly lashed together by rope.

5. F. Zimmerhaeckel,'* a Prussian officer of engineers, has devised
a modification of Cohausen's arrangement, which is clearly sum-
marized as follows by Mr. A. H. Allcroft^ —
Zimmerhaeckel believes
: '

^ One would have thought that the meaning of ab extrema parte was plain
enough but Heller {Philologus, x, 1855, pp. 732-4), like Frohlich, misunderstands
;

it. These words, he says, were essential to Caesar's description, as the reader
will perceive if he will reflect that the fibulae were placed on the inner and the
outer edge of each couple of balks : otherwise we should infallibly have
received the false impression that they had been driven through the middle
of the two balks of each couple as well as through the transverse beam. But
if Caesar had meant to convey the meaning which Heller, alone among all

commentators, has read into his words, would he not have written ab exteriore
or ab externa (parte) ? See B. G., ii, 8, § 4 ; II, § 4 iii, 12, § 1 ;
; 29, § 2 ;

vi, 29, § 3 and Meusel's Lex. Gaes., i, 1242-3.

;

'^
Bemerkungen zu Gaesar de B. G., iv, 17, pp. 2-3. Cf. Jahrb. d. AUerihum<i-
freunde im, Rheinlande, Ixxx, 188G, p. 122.
^ Giisays Rheinhriicken, 1807,
p. 47 (Fig. 19).
* G. Julius Caesars Rheinbrilcke, 1899. Glass. Rev., xiii, 1899, p. 408.
''

J
 OVER THE RHINE 717

that the fibulae were merely bars of undressed wood laid in the
alternate acute angles of the X-cross formed by the junction of
the horizontal trahs [beam] with the sloping tigna [balks] viz. one ;

fibula in the outer and lower angle, the other in the inner and upper
angle. To hold them in place they were stoutly roped to each other ;

the essential point is that while remaining firmly connected, they

should nevertheless move with freedom. This, he says, is a recog-
nized method in modern bridge-building.' Now the force of the current
would obviously tend to lessen the acute angles in which the up-stream
fibulae were placed, and therefore to bind the timbers more closely
together on that side but, not to mention other objections, which
;

will presently appear, on the down-stream side the effect would be

the very reverse.
6. Maxa,i Rheinhard,^ and Schleussinger ^ hold, like Cohausen, that
the balks were grooved for the reception of the fibulae. According
to Rheinhard, the fibulae bit into grooves in the transverse beam
and in the balks. Their arrangement was triangular that is to say, :

they rested against the outer sides of the balks, and, sloping up-
wards, lapped over and met above the transverse beam.*
7. Napoleon ^ interprets the word fibulae in a wholly different
sense. He holds that the opposite couples of balks were kept at the
right distance apart by two pairs of wooden tie-beams, each pair
being bolted in the shape of the figure
balks.
x to the outer sides of the

Professor E. V. Arnold virtually agrees with Napoleon, only dif-

fering from him as to the way in which the fibulae were made fast to
the balks. The word distinebantur,^ he says, shows that the main
' '

object of the fibulae was to prevent the couples of piles [or balks]
from falling together inwards and this end would most simply be
;

attained by notching the fibulae or by letting them into the beams

at each end.' He interprets in contrariam fartem revinctis as
—
meaning 'they were linked together crosswise, i.e. the right-hand
upper pile to the left-hand lower pile, and vice versa '.^
Now there is one fatal objection to both Napoleon's theory and
Professor Arnold's neither the one nor the other agrees with Caesar's
:

description. Fibulae were not tie-beams in contrariam parte^n

;

cannot possibly bear the meaning which Professor Arnold gives

it ; and the words utrimque and ab extrema parte can only mean that
the fibulae were at the upper ends of the opposite couples of balks
and at the ends of the transverse beam."^ Every writer who has com-
bined a practical knowledge of the principles of the construction of
bridges with adequate scholarship has come to this conclusion.

* Zeitschr.f. d. osterreich. Oymnasien, xxxi, 1880, picture on p. 493.

^ Caesar's Bheinhrilcke, p. 14.
^ Studie zu Caesar's RheinbriicJcc, p. 30. * See Rhcinhard's i^lan and p. 1 4.
'^
Hist, de Jules Cesar, ii, 145-0 and Planclic 15.
^ Class. Rev., i, 1887,
pp. 108-0. The professor takes quibus as an instru-
mental ablative. Tiiis view Mr. Koby [ib., p. 242) combats, because disclusia
would then liave no subject expressed.
^ Cf. Heller in Philoloyus, xxvi, 1807, pp. 009-70.
718 CAESAR'S BRIDGE
8. Soiintag thinks that the system oi fibulae devised by Cohauseii
^

would have answered for the upper balks, but not for the lower ;

for, he says, the pressure of the water would have tended to depress
the upper and to raise the lower balks, thereby enlarging the acute
and diminishing the obtuse angle. He holds that quihus must
grammatically refer not to utraque (tigna) but to fibulis, in other
words, that disclusis agrees with quihus (fibulis), not with quihus
(tignis), or with tignis understood and Mr. Taylor, who agrees with
;

him, remarks that to take quihus as having for its antecedent

'

tignis . does not yield any sense such as to explain or justify

. .

Caesar's statement about the effect of increased strength of the

current '.^ If they are right, Caesar meant, not that the balks were
kept apart by the fihulae, but that the fihulae were themselves kept
open. Sonntag goes on to argue that if, as appears certain, the plan
of Caesar's bridge was new, and if fihula was not a word used in the
terminology of bridge-construction, Caesar must have used it in
a sense with which his readers would be familiar. Now the words
disclusis (' kept apart ') and revinctis (' made fast ') could have been
predicated of the fihula, or safety-pin, which was used to fasten
clothes, revinctis because the pin which is inserted in the catch
'

cannot move far from the shaft disclusis because shaft and i3in
' ;
'

are kept apart by folds of clothing '. Caesar's ^61^/06, Sonntag con-
cludes, were two beams, adjusted in the angles formed by the balks
and transverse beam and held together at the ends by iron bands.
But, according to the analogy, did not the two beams form one
fihula ? Wait and see.
9. Mr. Taylor, in an interesting article which appeared in the
Classical Review of January, 1902, offers an explanation of the
fihulae, which, as he acknowledges, was anticipated by Colonel Emy,*^
and is substantially identical with that of Sonntag.* Like the German
scholar, he takes quihus as referring to fihulis and agreeing with
disclusis, and translates quihus disclusis atque in corUrariam partem
revinctis by since these fihulae were separated (viz. by the trahs and
'

tigna hina) and lashed together in direction opposite to their separa-

tion '. His theory, which expressly identifies the fihulae with safety-
pins, will be best explained in his own words.
—
Suppose the fihidae
'

to have been poles of six or seven feet in length, placed in the acute
angles between the up-stream tigna and trahes, and in the obtuse
angles between the down-stream tigna and trahes, and that the ends
of each pair of fihulae which projected beyond the tigna on either
side of the trahs were firmly lashed together. Consider the effect
. . .

of this arrangement. Any movement caused by the current would. . .

be of the nature of a slight rotation of the twin-piers about transverse

' Bemerkungen zu Caesar de B. G., iv, 17, p. 4.

2 Class. Rev., xvi, 1902, pp. 32-3.
^ Traite de Vart de la charpenterie, ii, 1841, pp. 369-70 and Atlas, pi. 129,
fig. 4.
* The only difference is that, according to Mr. Taylor, the fibulae were
fastened by ropes, which Sonntag {Bemerkumjen, &c., pp. 2-3) thinks would
have been too weak.
 OVER THE RHINE 719

axes ill the bed of the river. This rotation would tend to diminish
the angles in which the fibulae were placed, and therefore to thrust
them outwards from the vertices of those angles and ipso facto to
tighten the lashing and bind all the timbers firmer together. . . .

Hence it is clear that it is a mistake to figure such fibulae placed in

the acute angles on the down-stream side of the bridge for in that ;

case the lashings on that side would be slackened by any movement

induced by the current.' Mr. Taylor adds that he verified his theory
by constructing a model .^
Now when I read Mr. Taylor's article I wrote on the margin,
'
Taylor's binae fibulae plus the lashing only constitute one " safety-
pin " 'and Mr. Stanley Hall endorses this objection.
; The whole '

significance,' he writes, of Mr. Taylor's translation must lie in the

'

fact that the two poles plus the lashings constituted a single fibula.
In fact he admits as much. ... In the second place, even allowing
this unsurmountable difficulty, can it be claimed that, with his
fibulae, the greater the force of the current, the firmer is the structure ?
I think not rather the reverse is the case. True, as the angles tend
:

to diminish, the lashings become more taut the effect of the strain
:

upon them is nevertheless to stretch the lashings and the best that ;

can be said for such a construction is that the lashings, in spite of

being stretched, will nevertheless remain taut. But Mr. Taylor's
fibulae still fail because he has not appreciated the importance of
keeping apart (distinebantur) the top ends of the balks. The trabs
bipedalis is a beam in compression, to use a mechanical term and ;

joints at the points B and C, such as he describes, would allow the

ligna bina to slide to some degree, owing to the tendency to rotate
about the point A, caused by the force of the stream. In fact, the
strain put upon both the lashings and the members of the structure
is infinitely greater than in a triangulated construction. Further, . . .

to take disclusis as referring to fibulis and translate it " since they

were separated " (viz. by the trabs and tigna bina) seems to me to be
forcing the Latin. [Mr. Taylor] says " the words quibus disclusis
. . .

will aptly represent the unclasping of the safety-pins before passing

them round the woodwork ". This is all very excellent when applied
to a single safety-pin but it is quite a different thing when applied to
;

a couple of poles tied at both ends. It has only to be stated to show

that the meaning will not hold :
—
take two poles, unclasp them (!)
round the tops of the tigna, and then bind them up in direction
'
opposite to their separation !

10. Mr. Hall himself has devised a theory which tallies perfectly
with Caesar's description. Its novelty is that it supposes the results
described by the words disclusis and revinctis to have been produced
not by the fibulae alone, but also by a tie. The fibulae in Mr. Hall's
opinion were iron^ dogs ', such as are universally used now. They
'

are shaped somewhat like a broad angular croquet-hoop, and are

1 Class. Rev., xvi, 1902, pp. 31-2.

2 It has been objected that Caesar would not have carried about a sufficient
quantity of iron ; but this objection is disposed of by B. G., vii, 73, § 9.
720 CAESAR'S BRIDGE
about 12 inches across, with the ends bent at right angles and pointed.
They would have been fixed diagonally, one of the points being
driven into the transverse beam and the other into the top of one of
the balks. If, says Mr. Hall, any one should object that such hold-
fasts do not resemble safety-pins, it is a perfectly fair and adequate
'

answer to say that they are as much like Roman brooches as their
English equivalent is like English dogs.'
Now for Mr. Hall's original idea. Referring to his diagram, he
remarks that since the points B and C are made fast [by the fibulae],
'

the whole structure ABCD will tend to take up the position AB'C'D,
and the effect of this will be to lengthen the distance between the
points A and G. The most elementary engineer would know that
this tendency would be counteracted by a tie
— one tie, remember,
'
—
not, as in Napoleon's theory, two between the point C and a point
'

on AB as near as possible to A. This tie triangulates the structure

ABO, and as long as there is a constant pressure in the direction of
—
the stream, the bridge provided always that the various members
of the structure have been designed to meet the various stresses put
—
upon them must be absolutely rigid. ... It will be seen,' he adds,
'
that I take quibus as referring to haec utraque, that is, both pairs of
balks disclusis as summing up and repeating binis utrimque fibulis
;

ab extrema parte distinebantur and in contrariam partem revinctis as

;

meaning having a diagonal tie '. It does not seem to me in the

'

smallest degree surprising that Caesar should add the last member of
the structure by an ablative absolute clause any engineer would at
:

once assume that there would be this tie.' In regard to the words
tanta erat operis firmitudo atque ea rerum natura ut quo maior vis
aquae se incitavisset, hoc artius inligata tenerentur, Mr. Hall thinks
that the meaning of the Latin is sufficiently brought out if we
'

understand Caesar to mean that the constant pressure of the stream

was an essential factor in the stability of the bridge '.
11. Mr. E. Kitson Clark,i remarking that the horizontal beam'

[trabs] was continued beyond its junction with the piers [tigna], '

affirms that between the end of this [ab extrema parte] and the top
'

of the piers braces [fibulae] were attached in such a manner as to

hold the horizontal and uprights [tigna] rigidly in position. The
ends of these were thus thrust apart [disclusis] and could not close,
and tied together [in contrariam partem revinctis] and could not
open. In short,' Mr. Clark continues, by the insertion and method
'

of attachment of the fibula, a member combining the properties of

strut and tie was introduced, which formed the base of a strong
triangle whose apex was the crossing of the main members a tri- —
angle which would resist any distortion of the angle at the crossing.'
Mr. Taylor, in a MS. comment, objects that no single thing can '

combine the properties of strut and tie for, he asks,

'
; how can '

com2)ression and tension coexist in one piece ? Besides, as we have

'

seen, the two tigna of each couple were already securely fastened,
and therefore could neither close nor open.

^ Clasd. liev., xxii, 1908, pp. 144-5.

 —

OVER THE RHINE 721

The impartial reader will perhaps agree with me that Mr. Hall, if

my one, has solved the problem.

VII. The only remaining difficulty is about the supports and
iefences which were added after the bridge was built. Piles,' says '

Daesar, ' were also driven in diagonally on the down-stream side,

which were connected with the entire structure and planted below
ike a buttress, so as to break the force of the stream. Other piles
were likewise planted a little above the bridge, so that in case the
natives floated down trunks of trees or barges to demolish the struc-
ture, their force might be weakened by these bulwarks, and they
might not injure the bridge (suhlicae et ad inferiorem partem fluminis
'

ihlique agehantur, quae pro ariete suhiectae et cum omni opere con-
iunctae vim fluminis exciperent, et aliae item supra pontem mediocri
^patio, ut si arhorum trunci sive naves deiciendi operis causa essent
I barharis missae, his defensorihus earum rerum vis minueretur neu
X)onti nocerent).
1. Long infers from the words et aliae item supra that the suhlicae '

below and above the bridge were fixed in the same manner while '
;

Prom Caesar's saying that the balks of the bridge itself were not
iriven in, like suhlicae, perpendicularly, but sloping (non suhlicae
modo derecte ad perpendiculum, sed prone ac fastigate), he gathers
that all the suhlicae were perpendicular. He explains ohlique age-
hantur thus :
—
the suhlicae were placed in a triangular form ' (that
'

is to say, arranged in a series of triangles), and those on the down-

stream side were connected with the balks, and presented as they '

were viewed from the lower side of the bridge the appearance of
a head or solid angle.' His Plan ^ would seem to show that he con-
sidered each triangle of suhlicae on the down-stream side of the
bridge to have enclosed a separate couple of balks. I believe, how-
ever, that his Plan does not represent his meaning for he goes on ;

to say, 'it may be said that it would have been better if the " suh-
licae " on the lower side of the stream had been placed with the
ram's head ("aries"^) towards the stream, and inside and under the
bridge ^ but Caesar seems to place this work in the lower part on
;

the outside, as he certainly does place it outside in the upper part,

for there the " suhlicae " were not connected with the piles [or
balks], but placed in front of them.'
2. Schneider, who, like Long, believes that the suhlicae were
vertical, and agrees with him in the interpretation of ohlique, holds
that from a point close to and below the centre of the bridge two
rows of suhlicae extended obliquely to either bank, being arranged
in the form of an obtuse angle with its apex pointing up-stream. He

Facing page 194 of Long's Caesar.

^

Mr. Peskett {Caesar, Books iv-v, 1887, p. 65) says that it is not easy to
2 '

see what meaning Mr. Long attaches to pro ariete'. Long's meaning is em-
bodied in Schneider's note ad aquas rumpendas sic suhiectae, ut qui vulgo
:

vacatur aries ad muros diruendos adhiheri solet Neque inepta ad illam, quam
. . .

statuimus, suhlicarum dispositionem arietis comparatio videtur. Queynadmodum

enim aries murum, sic illae undarum vim frangehant.
^ This is the very way in which Long's Plan represents them as having been

placed.
1093 3 A
722 CAESAR'S BRIDGE
argues that when Caesar says that the suhlicae were cum omni opere
coniunctae, he only means that they were close to, not in actual
contact with, the inain structure of the bridge and he contends ;

that if the force of the stream, after it had just passed the lower
side of the bridge, had been checked, the force of the water above
the balks must have been checked also.^ No doubt but, as anv :

mathematician or engineer would have told him, only to a slight

extent. This objection tells equally against the theory of Long ;

and both are inconsistent with the statement that the piles were
designed ut vim fluminis exciperent.^
3. Mr. A. H. Allcroft^ has modified Long's theory in a way which
appears to Mr. Peskett unintelligible '. He holds that the stockades
'

'
were actually beneath the piers [or balks], the latter sloping over
and resting upon them, so that the stockade served as a buttress
(pro ariete suhiectae), and pier and stockade could be securely fastened
to one another' {cum omni opere coniunctae). Remembering,' says '

Mr. Peskett,^ that a so-called pier consisted of two large sloping

'

tigna each a foot and a half thick, separated by an interval of two

feet, one would like to see a diagram showing how a pier of this kind
could slope over and rest on a V-shaped stockade of perpendicular
piles.' I have no doubt that Mr. Allcroft could gratify Mr. Peskett's
wish but his theory does not agree with the words cum omni opere
;

coniunctae for the balks could not have rested upon the piles which
:

he describes unless the latter had been driven in first.

4. Mr. Kitson Clark, who adopts a quasi-V-shaped formation
(the figure \/ being substituted for a V), places the down-stream
suhlicae on either side of, and in contact with, each of the couples
of balks, and explains pro ariete as meaning like a lateral buttress '.^ '

Arranged in this way, the suhlicae might, indeed, have been described
as cum omni opere coniunctae but Mr. Hall agrees with me that
;

they would have been more effective if thev had formed a com-
plete V.
5. Napoleon holds that the lower suhlicae leaned, like a buttress,
against each of the balks.^ But if Caesar had meant to describe
such an arrangement as this, he would surely have written not ohlique
but prone. Ohlique means that the suhlicae were driven into the bed
of the river, not in a line parallel with that of the bridge, but slant-
wise with regard to that line. Caesar expressly says, in the earlier
part of his description of the bridge, that suhlicae were fixed

^ Schneider's Caesar, i, 358-9.

2Mr. Allcroft, who translates excipere vim fluminis by 'to intercept the rush
of the current ', affirms that the phrase is used only of an object which, being
'

itself higher up the stream, covers and protects an object lower down' and he ;

refers to Livy, xxi, 28 (for which read 27, § 8). Mr. Peskett {CIom. Bev., xiii,
1899, p. 462) stigmatizes his assertion as easy to make but difficult to prove'
'
;

but it is supported not only by the quotation from Livy, but also by the other
relevant passages B. G., i, 52, § 4 iii, 5, § 3 B. C, iii, 93, § 2 in which
; ; —
Caesar uses the word. See p. 723, infra.
3 Class. Rev., xiii, 1899, p. 409.
* 76., p. 462. " lb., xxii, 1908, pp. 145-6.
« Hist, de Jules Cesar, ii, 146 and Planche 15.
 OVER THE RHINE 723

perpendicularly. If, then, he had used the word suhlicae of piles which

sloped, he would have been guilty of inconsistency and obscurity.

Moreover, as Long points out, the suhlicae were placed in position
after the main structure of the bridge was finished and it would ;

have been impossible, he maintains, in these circumstances, to drive

them into the bed of the river in the way in which they are repre-
'

sented in the common plates '.^

Nevertheless Mr. Peskett will have it that the suhlicae did slope.
'
I believe,' he says,^ that the whole theory of V-shaped stockades
'

is wrong, because I cannot conceive that Caesar's powers of expres-

sion were so limited that he would have described such an arrange-
ment simply by the words suhlicae oblique agerentur quae fro
. . .

ariete suhiectae et cum omni opere coniunctae vim jiuminis exciperenl.^

Evidently on this point agreement is hopeless but can Mr. Peskett ;

conceive that Caesar's powers of expression were so limited that,

after expressly stating that suhlicae were fixed perpendicularly, he
should have proceeded in the same paragraph to say that they were
fixed oblique, if by oblique he meant to denote or to connote a position
which was not perpendicular ? Moreover, if Mr. Peskett will think
again, he will admit that Caesar's powers of expression would
' '

have hopelessly failed if he had said that sloping buttresses, such

as those imagined by Napoleon, were intended to intercept the force '

of the stream '. Obviously the stream would have struck the tigna
with undiminished force and the buttresses would only have sup-
;

ported them against it. But in this case Caesar would surely have
written, not vim Jiuminis exciperent, but pontem (or tigna) jlrmarent?

^ R. Menge [Philologus, xliv, 1885, pp. 288-9) has shown that it would
have been possible to drive piles sloping into the bed of the river in such a way
that they would have been in contact with the bridge and supported it but ;

how can his explanation avail against Caesar's express statement that suhlicae
were perpendicular ?
In the first edition (p. 707) I argued that, as there was no evidence that
oblique was ever used in the sense of prone, if the suhlicae were props or
'

buttresses, they leaned ohliquely against the bridge in other words, there
;

were two suhlicae to each couple of balks, both sloping forward and at the
—
same time obliquely against the stream, one from right to left, the other from
left to right 'and I added that an engineer, to whom I suggested this
;
'

arrangement, approved of it as the most effective ', and that, leaning in this '

manner like buttresses against the bridge, the suhlicae would have indirectly
broken the force of the current'. In this case the word aries ('ram') in
Caesar's description might possibly be analogous to capreoli (literally wild '

goats '), as used in B. C, ii, 10, § 3, where it denotes pieces of timber forming
the framework of the roof of a sapper's hut. But I ought to have held fast to
the argument, the force of which I of course admitted, that Napoleon's view
obliges us to assume that Caesar was unpardonably inconsistent in his use of
the word suhlicae. A suhlica,' says Mr, Peskett {Class. Rev., xiii, 1899,
'

p. 462), of course remains a suhlica whether it is driven in straight or

'

crooked'. The tower of Pisa remains a tower although it leans; but the
normal position of towers and of suhlicae is perpendicular.
Class. Rev., xiii, 1899, p. 462.
2

It is not, perhaps, absolutely necessary to

•*
assume that the arrangement
of the suhlicae was V-shaped. If,' says Mr.
'
Taylor {ih., xvi, 1902, p. 38),
'
Caesar's bridge was ... at, or just below, a bend in the river ... a diagonal
line of piles across the front of each pier would have been all that was required
to protect them and this may be what Caesar meant by ohlique.'
;

3 A2
724 CAESAR'S BRIDGE OVER THE RHINE
6. Napoleon, after describing the lower suhlicae, says, Other piles '

were similarly driven in at a little distance above the bridge. The

'

words similarly might suggest that, in his opinion, the upper piles
' '

were also driven in slantwise but his Plan (15) represents them as
;

vertical and arranged in the triangular form which I have described.

I am told that wooden bridges in Norway are protected in the same
way now.
7. Cohausen thinks that the upper suhlicae w6re not driven into
the bed vertically, but were placed at an angle more acute than that
formed by the piles of the bridge, and leaned against them.^ This
arrangement is inconsistent with the natural meaning of mediocri
spatio. Schleussinger, however, defends Cohausen's view. That the
upper suhlicae were in contact with the bridge, is, he maintains,
proved by Caesar's having used the word minueretur, instead of
frangeretur, to express that the object of these piles was to protect
the bridge from the shock of any timber or vessels which the Germans
might float down to destroy it and also by the word item, which,
;

he insists, evidently refers to ohlique agehantur. Besides, he argues,

if the upper suhlicae had been upright, Caesar would certainly have

said so for those on the lower side of the bridge were unquestion-
;

ably sloping.2 I see no force in these arguments. The first implies

that Caesar deliberately preferred a less to a more effective and
easier method of protecting his bridge if minueretur is to be pressed,
:

it can only mean that Caesar calculated that the suhlicae would
minimize the damaging force of the floating timber, even if they did
not absolutely prevent its reaching the bridge. The second recoils
against Schleussinger it is not unquestionable that the lower
:

suhlicae were sloping on the contrary, it is morally certain that

;

they were vertical.^

8. According to Alberti,* there were two separate upright suhlicae
to each couple of balks on the upper side of the bridge according to
:

Scamozzi,^ two separate suhlicae sloping towards, but not touching the
bridge. Neither of these contrivances, however, would have answered
Caesar's purpose so well as that suggested by Napoleon Scamozzi's :

is inconsistent with Caesar's general statement about suhlicae and ;

both appear to me inconsistent with the words ohlique and item.

WAS COMMIUS KING OF THE ATREBATES OR OF

THE MORINI?
Caesar relates that in 55 B.C. he sent on a mission to Britain
Co7nmium, quem ipse Atrehatihus superatis regem ihi constituerat,^
which, in conformity with the usual interpretation of the passage,
^ Cdsar's RheinhrilcJcen, pp. 27 (fig. 10), 44-5.
2 Studie zu Gaesafs Rheinbrncke, pp. 13-4.
=»
R. Menge {Philologus, xHv, 1885, p. 289) thinks that item may refer to
ogebantur only, and not to ohlique. I do not agree with him.
* J. Rondelet, Traite theorique et pratique de Vart de bdtir, t. vii, pi. cxxxviii,

fig. 1. ' lb., fig. 3, « B. G., iv, 21, § 7.

 WAS COMMIUS KING OF THE ATREBATES ? 725

I translate by Commius, whom, after the overthrow of the Atre-

'

bates, he had set up as king over that people '. M. Jullian,i however,
maintains that ibi refers not to the Atrebates, but to the Morini,
who are mentioned in an earlier section ^ of the same chapter. He
argues, first, that Caesar's later statement, that in acknowledge-
ment of Commius's he had granted his tribe immunity
services, '

from taxation [?], restored to it its rights and laws, and placed the
Morini under his authority (pro quihus mentis civitatem eius im-
'

munem esse iusserat, iura legesque reddiderat atque ipsi Morinos

attrihuerat)^ seems incompatible with the view that he had imposed
a king upon the Atrebates ; secondly, that Commius is never called
rex Atrehatium, but always Atrehas and thirdly, that in the passage
;

just quoted ipsi apparently means Commius. Certainly it does ;

but why should not Caesar have placed the Morini, who had more
than once rebelled, in dependence upon Commius, whom he trusted,
even though Commius was king of the Atrebates ? M. Jullian's
second argument does not convince me, for Caesar, as I understand
him, had stated once for all that he had made Commius king of the
Atrebates and the words pro quihus meritis, &c., are quite com-
;

patible with the received view, for they only mean that Caesar
relieved the Atrebates from the obligation of paying tribute (or
supplying corn ?), and left them free to manage their own internal
affairs. Moreover, there are cogent reasons for dissenting from
M. Jullian's conclusion. First, the obvious meaning of ibi, as the
consensus of other commentators shows, is certainly apud Atrebates :

to make it refer to the Morini is to charge Caesar with unpardonable

obscurity, for, if he had intended to refer to them, he would surely
have written apud Morinos. Secondly, to make ibi equivalent to
apud Morinos deprives the words Atrebatibus superatis of all point.
Thirdly, the assumption that Commius was king of the Morini im-
plies that he was wholly unable to control his subjects for in the ;

year after the subjection of the Atrebates the Morini refused to sue
for peace and successfully resisted Caesar,* and in the following year
part of them sent envoys to Caesar,^ apparently without any authority
from Commius, and others rebelled again although a few weeks before
troops had been dispatched against them.^ Lastly, if M. Jullian will
read the words pro quibus meritis ipsi Morinos attribuerat again,
. . .

he will acknowledge that they alone prove that the Morini were not
placed under Commius's authority until after the second invasion
of Britain for the merita in question were the services which he
;

had rendered in the two British expeditions. It follows that the

people over whom he was made king in 57 B. c. were the Atrebates.

^ Hist, de la Gaule, iii, 315, n. 4. 2

§ 3^
^ 5. G'., vii, 76, § 1. * /6., iii, 28-9.
' lb., iv, 22, § 1. 6
/6.^ 37, 38, § 1.
 726

WAS COTTA SUBORDINATE TO SABINUS AT

ATUATUCA?
Mommsen ^ holds that Cotta was probably directed, in the event
'

of a difference, to yield to Sabinus. The reasons which he gives

'

are, first, that Sabinus had already performed important services

secondly, that whenever the two are mentioned together in the
Commentaries, except in B. G., vi, 37, § 8, Sabinus takes precedence ;

thirdly, that Caesar would not have placed two officers over one
camp without providing for the contingency of a difference of opinion
and fourthly, that Caesar's narrative of the disaster at Atuatuca
confirms the foregoing arguments.^

THE MEANING OF A DISPUTED PASSAGE IN

B. G., V, 31, § 5

Caesar, describing how the troops of Sabinus and Cotta spent the
night after it had been decided to abandon Atuatuca, writes, omnia
excogitantur quare nee sine periculo maneatur et languore miliium et
vigiliis periculum augeatur. Long^ explains the meaning of this
perfectly. The passage,' he says, seems to mean " every reason
' '

is suggested why there was no staying without danger, and why the
danger would be increased by the lassitude and watchesof the soldiers."
C'aesar puts it in the most general way. It was settled that they
must go, and everybody, at least those who were in favour of going,
thinks of every possible reason to confirm his opinion, and to con-
vince others that if they stayed in the camp, it was not without
risk,' &c. F. Liidecke* thinks that the passage is out of place, and
suggests that it originally stood in § 3 after ferducitur, representing
an argument which was urged by Sabinus and his party at the
council of war.^ But this conjecture is quite uncalled for. To a
reader gifted with the least historical imagination the passage, where
1 Hist, oj Rome, v, 1894, p. 68, note [Rdm. Gesch., iii, 1889, p. 274, note).
2 B. G., iv, 22, § 5 ; 38, § 3 ; v, 24, § 5 ; 26, § 2 ; 27-37, 52, § 4 ; vi, 32, § 4.
^ Caesar, p. 249.
* Ne^Le Jahrb. f. Philologie, &c., cxi, 1875, pp. 429-32.
^ W. Paul, who has wasted more time and ingenuity than any other scholar

in trying to amend the text of the Commentaries, deletes the whole passage.
Zeitschr. f. d. Gymnasialwesen, xxxv, 1881, pp. 281-4. [I am very sorry to
see that Meusel {Jahresh. d. philol. Vereins zu Berlin, xxxvi, 1910, pp. 46-7),
though he interprets the sentence as I have done, also condemns it as an inter-
polation. In its existing context, he says, it must mean, as Mommsen (who
changes maneatur into eatiir) puts it, Everything is done in order to make
'

the marcli as daiigerous as possible and to tire out the soldiers beforehand ' !

He also remarks, following the restless Paul, that Caesar nowhere else uses
cxcogitare in the Gallic War, or languor at all. But what then ? On the very
lirst page of Mcu.scr.s Lcxicou, Caesariannni, which 1 open at random —
i, 742

1 find three words which Caesar only uses once, copiosus, copula, and cor on :

the next (i, 1 198) two, cxcello and cxccplo ; and on the next (i, 1201-2) four,
excubitor, cxculco, cxcursio, and excursus !]
 A DISPUTED PASSAGE IN B. G., V, 31 727

it stands, will appear far more natural and effective than anywhere
else. Moreover, as F. Klein ^ points out, Caesar represents Sabinus
as unsupported in was Cotta who was backed up by the
argument : it
other officers. After Sabinus had announced his decision, the soldiers
would naturally have talked the matter over among themselves.
I subjoin my own translation of a part of the chapter, to show
how naturally the passage reads in its proper place The dispute :
— '

dragged on till midnight. At length Cotta gave way Sabinus's . . . :

view prevailed. An order was issued that the troops were to march
at daybreak. The men stayed up for the rest of the night, every
one looking about to see what he could take with him, what part
of his winter's kit he would be forced to leave behind. Men thought
of every argument to persuade themselves that they could not
remain without danger, and that the danger would be increased by
protracted watches and consequent exhaustion. At dawn they
marched out of camp,' &c.

WHAT WAS THE FORMATION OF SABINUS'S

COLUMN, WHEN IT MARCHED FROM ATUATUCA ?

Caesar describes it thus :

— 'At dawn they marched out of camp,
like men convinced that no enemy had counselled them, but the
best of friends, in a column of extraordinary length and with an
immense baggage-train (Prima luce sic ex castris proficiscuntur ut
'

quibus esset persuasum non ab hoste, sed ab homine amicissimo con-

silium datum, longissimo agmine maximisque impedifnentis)."^ Ob-
viously Caesar implies that Sabinus was to blame for adopting such
a formation and the blame appears to be defined by the two super-
;

latives, longissi^no and maxiynis. If so, Caesar's censure was directed

simply against the needless length of the column and the excessive
amount of baggage which accompanied it. He tells us nothing about
the arrangement of the baggage and perhaps he had no fault to
;

find on that score. But Long^ appears to think that the arrange-
ment was similar to that which Caesar himself adopted when marching
against the Nervii in 57 B.C., and which he discarded when making
his final march before the battle.^ If so, we may perhaps surmise
that the legion and the five cohorts under Sabinus's command were
separated from each other, each division being followed by its own
baggage-train. But there is not a word of this in Caesar's narrative.
Schneider^ suggests that the order of march was practically
identical with that which Caesar adopted when marching against
the Nervii, the only difference being that, on this occasion, each
cohort was followed by its own little baggage-train. This is very
ingenious, but very improbable. Censuring Sabinus as he did for
the undue length of his column, Caesar may have thought that,

^ Neat Jahrb., &c'., cxi, 1875, pp. 854-0.

- B. (/., V, 31, § G. ^ Cae6ar, p. 250.
* ISco p. 7(5, supra, and IJ. (/., ii, 17, § 2 ; 11), §§ 2-3.
'
Caesar, ii, 118.
728 SABINUS'S COLUMN
having foolishly decided to abandon his camp, he ought to have
formed his force in what was called a quadratum agmen, that is to
say, a hollow square or parallelogram, and placed the baggage in
the centre.^ The Emperor Julian once adopted a similar formation,
which is minutely described by Ammianus Marcellinus.^ But we
cannot get beyond conjecture.

THE MEANING OF ORBIS IN B. G., v, 33, § 3

Aulus Gellius^ gives a list, in which orhis occurs, of the words by

which the various military formations were described and he adds ;

that the words are derived from the things themselves, which are
'

strictly so called ', tralata autem sunt ah ipsis rebus, quae ita proprie
nominantur : earumque rerum in acie instruenda sui cuiusque vo-
cahuli imagines ostenduntur. This passage seems to me to support
the view that when Caesar uses the word orhis, he means a circle ', '

or rather a figure approximating thereto.

A friend of mine, a very able officer of the Royal Artillery, sug-
gests that the word orhis should be understood loosely as an irregular
figure, such as the men would have been able to form in the circum-
stances described by Caesar for, as he points out, to dress the line
;

properly, in such hurry and confusion and with the enemy pressing
on to the attack, would have been impossible. Colonel Dodge ^
suggests that the term orhis may have come from the natural habit
'

of flattening out the corners of such a square as a legion in three

'

lines could have readily formed for easier defence '.

'
It is difficult,' '

he continues, to imagine the manoeuvre by which a legion ployed

'

into anything approaching an actual circle and again deployed

into line. It may have been an irregular half-square, half-circle,
according to the accentuation of the ground or to the conditions
demanding a defensive formation.' Having regard to the passage
in Aulus Gellius, I am inclined to believe that this is the true
explanation.

THE EXTENT OF THE CONTRA VALLATION WITH

WHICH AMBIORIX SURROUNDED Q. CICERO'S
CAMP
. minus horis tribus milium passuum XV in circuitu munitionem
. .

perfecerunt (B. G., v, 42, § 4). Mr. A. G. Peskett^ says, In the '

place of passuum the best [he means the a] MSS. have p, which may
stand for either passuum or pedum all agree in the number XV.'
;

The latter remark is not correct, for the reading of the p MSS. is

^ I find that this remark has been anticipated by Turpin de Crisse, Cornm. de

Cesar, i, 1785, p. 358. "

xxiv, 1, §§ 2-4. Nod. AtL, x, 9.
=»

* Caemr, 1892, p. 303. Caesar, Books iv-v, 1887, p. 107.

'"
 —

PASSU UM OR PEDUM ? 729

X milium in circuitu munitionem pedum XV, It would of course

have been absurd to surround Cicero's small camp, the perimeter
of which could hardly have been much more than a mile,i with a
contra vallation of 15 Eoman miles and Napoleon, Thomann, and
;

others agree in holding that (milium) pedum (XV) should be read.

Desjardins,2 however, as well as Schneider and Nipperdey, adopt
the reading (milium) passuum (XV) but they give no reason for
;

their choice. Probably they were influenced by the fact that Orosius ^
described the rampart and ditch as vallum pedum decem et fossam
pedum quindecim per milia passuum quindecim in circuitu. Von
Goler * follows the p MSS. but a rampart 10 miles in extent would
;

have been as unnecessary as one of 15 miles. R. Menge ^ conjectures

that Caesar wrote (milium) passuum III, which would make the
extent of the wall identical with that specified by the assumed read-
ing (milium) pedum XV. I do not believe that Caesar wrote (milium)
pedum XV
for although in B. C, i, 82, § 4, he mentions a distance
;

of 2,000 feet,^ he nowhere computes a distance of more than one

Roman mile in feet, except possibly in B. G., ii, 30, § 2 (postea vallo
pedum XII in circuitu XV
milium, &c.), where the reading is also
doubtful and this is hardly an exception to his rule.'
;

THE RED-HOT BALLS WITH WHICH THE NERVII

SET FIRE TO THE HUTS IN Q. CICERO'S CAMP
The passage in which Caesar describes this operation runs as
follows maximo coorto vento ferventes fusili ex arcjilla glandes fundis
:

. .in casas, quae more Gallico stramentis erant tectae, iacere coeperunt.^
.

All the MSS. hsive fusili, the ordinary meaning of which is molten ' '
;

but clay cannot be melted. Holder, following C. Wagener,^ reads

fusilis (es) but this conjecture only makes matters worse
: the ;

difficulty of explaining the meaning of the passage remains, and, as

Scholl points out, ferventes fusiles glandes is bad grammar .^^ Let us
then accept the reading of the MSS., and translate /MS^7^ ex argilla

' See p. 251 and n. 2.

'^
Geogr. de la Gaule rom., ii, G60.
=»
vi, 10, § 3. * Gall, Krieg, 1880, p. 190, ii. 3.
See Meusel's Tabula Coniedurarum, p. 21 {Lex. Caes., vol. ii, pars ii).
^

® A. Klotz {Caesar studien, p. 219) proposes to substitute passuum in B. C,

i, 82, § 4 for pedum, on the ground that 2,000 feet was too narrow an interval

between two camps. I do not agree witli him. Caesar emphasizes the narrow-
ness. Cf. Stoffel, Guerre civile, i, 70-1 and B. G., i, 49.
' See
pp. 390-1, and A. Klotz, Caesar studien, pp. 218-9.
« ^. (?.,v,43,§l.

^ Neue Jahrb. Wagener argues that

f. Philologie, &c., cxxi, 1880, p. 624.
Caesar would never have written fusili ex argilla glandes because, as a rule,
he places ex before botli substantive and attribute, except in relative clauses.
But the rule, if it can be so called, has numerous exceptions (Meusel's Lex.
Caes., i, 1161-2). As far as I can make out, Wagener attaches to his imaginary
fusiles the meaning of capable of being thrown',
'
—
a meaning which is mean-
ingless and destitute of all support.
^" Blatter
f. d. buyer. Gymnasialschulweseii, xvii, 1881.
730 THE RED-HOT BALLS
by of softened (or plastic) clay '. But, it has been objected, balls of
'

clay cannot be so heated as to make an object, however inflammable,

which they may strike, take fire. If so, the balls must have been
made of something else besides clay. Eaumer pointed out in 1830
that in the valley of the Sambre people still burned balls made of clay
mixed with small coal as fuel and he suggested that the balls of
;

which Caesar spoke were similar to these. Von Goler, however,

affirms that he has proved by experiment that balls of clay can be
sufficiently heated to set straw on fire ^ and innumerable bullets of
;

baked clay have been found in ancient British forts.^

The commentators,' says Long, are troubled about these hot
' '

balls cast from leather slings, ... If the barbarians did really throw
hot clay balls with slings, let us be satisfied with the fact.' ^ Certainly,
'
if But the difficulty disappears if we assume the barbarians
. '

to have had their leather slings lined with metal.*

WAS THE LETTER WHICH CAESAR SENT TO

Q. CICERO WRITTEN IN GREEK, OR ONLY
IN GREEK CHARACTERS ?

Caesar describes the letter as Graecis conscriptam litteris ^ and,

taken by themselves, these words can only mean that it was written
;

I
in Greek characters. Dion Cassius ^ distinctly says that the letter
was Avritten in Greek but his testimony on such a point proves
;

nothing^ The only plausible reason for concluding that Caesar

wrote in Greek is that he would have taken every precaution to
prevent the contents of his letter from becoming known and ;

although it is unlikely that the half -barbarous Nervii and their allies
were acquainted with Greek characters, there may have been
individuals among them who were for (unless Caesar feared that
:

Roman prisoners might be forced to read the letter) is it not evident

that some of them knew Latin ?
0. Hirschfeld argues that Caesar merely wTote in Greek charac-
ters for, he says, although inscriptions written in Greek characters
;

have been found in southern France, all those found in the central
and the northern districts are in Roman.^ If Caesar had meant to

^ See Schneider's Caesar, ii, 144, and von Goler's Gall. Krieg, 1880, p. 191,
n. 2.
^ Rice Holmes, Anc. Britain, p. 268 and n. 6. ^ Caesar, p. 258.
* I findthat this remark has been anticipated. See von Goler, Gall. Krieg,
1880, p. 191, n. 2.
' B. G., V, « xl, 9,
48, § 4. § 3.
' It has been said that Polyaenus (viii, 23, § fi) says the same ; but this
is a mistake. Polyaenus professes to give Caesar's words, Kaiaap KiKipwvi
dappciv. TT-poo-Sf'xcu fiorjOeiau
: but as he himself wrote his treatise in Greek, it is
natural that lie should have given Caesar's alleged letter in the same language.
^ Silzunijsbcridila d. kaiscrl. Akad. d. WisscnscJiaJlen, Wien, cxiii, 1883,
p. 277, note. Inscriptions in (Jrcck letters on coins have, however, been found
in certain parts of Northern France (A. Blanchet, Traitc dea monn. gauL, p. 93).
 —

CAESAR'S LETTER TO CICERO 731

say that tlie letter was written in Greek, surely he would have said
Graece simply.^ As M. Blanchet ^ points out, he uses the expression
Graecis litteris in two other passages B. G., i, 29, § 1 and vi, 14, § 3
— where the meaning '
Greek characters '
is indisputable.

THE MEANING OF NOVISSIMUS IN B. G., v, 56, § 2

Describing the armed assembly which Indutiomarus summoned

towards the end of 54 B.C., Caesar writes, 'This, by Gallic usage, is
tantamount to a declaration of war. By inter-tribal law all adult
males are obliged to attend the muster under arms and the last ;

comer is tortured to death in sight of the host (Hoc more Gallorum '

est initium belli, quo lege communi omnes puheres armati convenire
coguntur : qui ex iis novissimus venit in conspectu multitudinis omni-
bus cruciatibus adfectus necatur). On this Long remarks,^ The words '

" qui ex iis novissimus venit " must mean, if any man came late and
did not arrive in company with the rest.' I do not see how this in-
terpretation can be got out of the Latin. Novissimus does not mean
'
late ', but last '.* If Caesar had meant late ', would he not have
' '

written si quis ex iis serius venit ?

THE NEW LEVIES WHICH CAESAR RAISED IN

54-53 B.C.
Speaking of the new levies which he made in the winter of 54-
53 B. c, in order to increase his army after the destruction of the
legion and the five cohorts commanded by Sabinus and Cotta, Caesar
says tribus ante exactam hiemem et constitutis et adductis legionibus,
daplicatoque earum cohortiu7n numero quas cum Q. Titurio
amiserat,^ &c. The words dwplicatoque amiserat are generally . . .

taken as explanatory of tribus legionibus in other words, Caesar

. . . ;

is understood to have meant that his three new legions doubled the

legion and five cohorts which he had lost. 0. Schambach, however,

argues that the words convey a fresh piece of information that is to ;

say, that Caesar procured not only three new legions but also 5x2
cohorts, which were not embodied in any legion. Caesar, he says, is
describing how he made good the loss which he had sustained (1) in
respect of troops embodied in a legion and (2) in respect of Sabinus's
five cohorts, which do not appear to have belonged to any legion.^

1 Cf. Cicero, De Or., i, 34, § 155 ; De Off., iii, 32, § 115 ; and Tusc, i, 8, § 15.
^ Traite des monn. gaul., p. 92, n. 4.
^ Decline of the Roman Republic, iv, 234, n. 4.
* Was the last comer regarded as accursed l That his execution involved
a religious ceremony I have little doubt. ^ B. G.,\i,
1, § 4.
« Rhein. ifws., N.F., xxxi, 187G,
pp. 308-9. Mcusel. on the other hand {Jahre.sh.
d. philol. Verein>i zn Berlin, xxxvi, 1910, p. 09) deletes que alter duplicato on
the ground that if que vvcjo genuine, avc should recei\e the false impression
that Caesar raised three new legions plufi 15x2 cohorts, that is, six new legions. —
iSomohovv I nexer receixcd this impression.
732 CAESAR'S LEVIES IN 54-53 B. C.

According to Schambacli, Caesar did not speak of the ten cohorts

which doubled those five as a legion, because they were not Romans,
'
'

but Transalpine Gauls.

Schambach is certainly mistaken. Before the destruction of
Sabinus's force, Caesar had eight legions and five cohorts. Sabinus's
force consisted of one legion and five cohorts and at the end of
;

53 B.C. Caesar had ten legions. This number corresponds with the
received view, that he increased his force by three new legions and
no more. If he raised ten new cohorts as well, he never again
mentioned them, and, as far as we know, never used them.i

DID LABIENUS ENCAMP IN THE COUNTRY OF

THE TREVERI IN THE WINTER OF 54-53 B.C.?
Caesar, in describing the distribution of the legions for the winter
of 54-53 B.C., says that he ordered Labienus to winter with one
legion in the country of the Remi just on the frontier of the Treveri ^
'
'
;

but in a later chapter he says that the Treveri were preparing to

'

attack Labienus and the single legion which was wintering in their

1 Mommsen [Hist, of Rome, v, 1894, p. 68, n. 1 IRom. Gesch., iii, 1889, p. 274,
note]) says the five cohorts are not counted as part of a legion any more than
'

the twelve cohorts at the Rhine bridge (vi, 29), and appear to have consisted of
detachments of other portions of the army '. If Mommsen is right, Caesar had
eight legions and no more before the destruction of vSabinus's force. Schneider,
however, argues {Caesar, ii, 95-6) that he had nine, because in 57 b. c. he had
eight {B. G., ii, 8, § 5), and he speaks of the entire legion which he gave to
tSabinus as that q^iam proxime trans Padum conscripserat {ib., v, 24, § 4).
Schneider and Meusel {Lex, Caes., ii, 1258) take proxime as meaning nuper, —
an interpretation which is supported by a comparison of B. G., vi, 1 with
32, § 5 and, if they are right, the legion in question was raised in the winter
;

of 55-54 B.C. This is also the view of von Goler {Gall. Krieg, 1880, p. 169),
who, however, suggests that the five cohorts in Sabinus's brigade belonged to
one of the old legions and that the remaining five of the same legion were
utilized for repairing losses in the army. This theory is ojjen to one objection.
If it is right, the newly raised legion must have marched into Belgium to
join Sabinus notwithstanding the scarcity of corn which Caesar so strongly
emphasized. P. Groebe (W. Drumann, Gesch. Roms, iii, 1906, p. 702) thinks
this more than unlikely
'
'but, after all, the legion, even if it had not been
;

brought into Gaul until after the winter of 54-53, would still have had to be
fed on the proceeds of the scanty harvest of the preceding summer. The
five cohorts were certainly veterans, and, if von Goler is wrong, they must have
been detachments from other legions (cf. B. G., v, 9, ^ I) ; for otherwise it
is incredible that the fact of their having been raised should not have been
recorded. In other words, if von Goler is wrong, Caesar, although he had"
nominally ten legions in 53 B.C., only had the equivalent of nine and a half,
five of the ten having each only nine cohorts. It seems to me more probable
that von Goler is right.
M. Jullian {Hist, de la Gaule, iii, 372) argues that the five cohorts had only
just been raised comme ces cinq cohortes ne font partie des legions
:
'

—
'
—
surely
this is begging the question ce ne peuvent etre que des nouvelles recrues.' I
'

find it very hard to believe that Caesar, after saying that the legion had been
recently raised, would have omitted to add that the cohorts were raw recruits.
^ quartam in Remis cum Tito Labieno in continio Treverorum hicmare iussit.

B. G., V, 24, § 2.
 —

LABTENUS'S CAMP IN 54-53 B.C. 733

'.i
territory Napoleon,^ perhaps following Schneider,^ explains this
apparent contradiction by suggesting that the country in which he '

[Labienus] encamped was either on the boundary of the two

countries, or ground of which the Remi and the Treveri disputed
the possession'. Labienus, he insists, had remained, throughout the
winter, in his original camp, wherever it was for, he asks, is it ;
'

not evident that, after the catastrophe of Aduatuca and the in-
surrection of the people seduced by Ambiorix, everything dictated
to Lnbienus the necessity of engaging himself no further in a hostile
country, by separating himself from the other legions?''* General
Creuly, who agrees with Napoleon, holds that the camp was in the
country of the Treveri and that in B. G., vi, 5 Caesar corrected the
;

mistake which he had made in v, 24.^ Besides, he argues, it is reason-

able to believe that Labienus, instead of following his instructions
to the letter and encamping within the territory of the Remi but close
to the Treveran frontier, found a more convenient site within the
territory of the Treveri.^
Von on the other hand, maintains that Labienus did change
Groler,'^
his quarters,and conjectures that he wished to relieve the friendly
Remi of the burden of having to feed his army. This view is certainly
defensible. If Caesar's narrative is to be interpreted literally, it is
clear that in the autumn of 54 b. c, after Caesar had relieved Cicero's
camp, Labienus was still in the country of the Remi for Caesar ;

says ^ that, on hearing the news of the relief, Indutiomarus, who had
determined to attack the camp of Labienus, returned into the country
of the Treveri (copias omnes in Treveros reducit). But it would seem
that in the early part of the following year Labienus was no longer
in the country of the Remi, but in that of the Treveri. Caesar tells
us this twice. ^ He says that he sent the baggage of the whole army
into the country of the Treveri to Labienus {totius exercitus impedi-
menta ad Lahienum in Treveros mittit) and he says that Labienus
;

was wintering in their territory. He also says that, after he had

finished his campaign against the Menapii, he set out for the country
of the Treveri, and that he arrived there ;^^ and we may reasonably
conclude that when he arrived he rejoined Labienus.
Schambach quotes approvingly an argument of Kochly andRiistow,

^ Treveri . . . Lahienum cum una

legione, quae in eorum finihus hiemahat,
adoriri parabant. B, G., vi, 7, § 1. —
Hiemaverat the reading of the a MSS.
is obviously wrong ; imply that Labienus had wintered in the
for it would
country of the Treveri, but was there no longer. But at the time of which
Caesar speaks Labienus was evidently in the country of the Treveri and, ;

moreover, the winter was not yet over. See also Jahresb. d. philol. Vereins
zu Berlin, xi, 1885, p. 193.
2 Hist, de Jules Cesar, ii, 227, note. ^ Caesar, ii, 202.

* Long takes the same view in a note on

p. 290 of his edition of Caesar,
and the opposite one on p. 412. In his history {Decline of the Roman Republic^
iv, 240) he reverts to Napoleon's opinion.
^ A. Klotz maintains, if I do not misunderstand him {Caesarstudien,
pp. 55-6),
that Caesar made a mistake not in v, 24, but in vi, 5 and 7.
^ Rev. arch., nouv. ser., viii, 1863,
p. 38.
7 Gall. Krieg, 1880, p. 209, n. 2. ^ B. G., v, 53, § 2. See also § 1.
« /6.,vi,5,§6; 7,§1. '»
76., vi, 6, §4; 9, § 1.
734 LABIENUS'S CAMP IN 54-.53 B. C.

who say that, if Labieniis had really moved into the country of the
Treveri, it would be very remarkable that Caesar should not have
explicitly stated the fact, seeing that he described the movements of
Labienus in detail and, adds Schambach, Caesar would not have
;

sent his baggage, as he did, to Labienus if Labienus had abandoned

his original camp for the second camp would not have been fortified
;

so thoroughly as the first.^ This is a groundless assumption. If

Labienus had made a second camp, he was passing the winter in it
and, as he was a thorough soldier, we may be certain that he had
fortified it with all possible care. As to the argument of Kochly and
Riistow, I admit that Caesar's silence would be (or is) remarkable
but the reticence of the Commentaries is frequently remarkable.
Schambach offers to extricate us from the difficulty by the usual
and futile expedient, an emendation — and, as it happens, the
;

emendation which he proposes was made two centuries before he

was born, by Cluver.^ In B. G., vi, 7, § 1, says Schambach, for eorum
read Remorum the false reading eorum, you are to understand, led
:

some copyist to interpolate in Treveros before ad Labienum in vi, 5,

§ 6. I don't believe it.
Heller, assuming that Labienus remained throughout the winter
in the same camp, suggests that it was situated in the territory of
some small tribe, perhaps the Segni, between the proper territories
of theRemi and the Treveri, which may have been subject to one or the
other of those two peoples.^ But if the camp was in the country of
the Segni, whom Caesar mentions by name, why did he not say so ?
Nothing is gained by making conjectures of this sort.
I admit that there is room for doubt but I am strongly inclined
;

to believe that Schneider and Napoleon are right, and that the
explanation of Caesar's apparent inconsistency lies in the fact that
Labienus's camp was very near the common frontier of the Treveri
and the Remi.^

DTD CAESAR CONFOUND THE SCHELDT WITH

THE SAMBRE ?

Caesar says that in 53 B.C. he left Quintus Cicero at Atuatuca and

marched towards the river Scaldis (Scheldt) quod influit in Mosam,
and the most distant parts of the Ardennes [extremasque Arduennae
fartes^). Stephanus, who has been followed by many other com-
mentators, altered Scaldim into Sahim (the Sambre). In support of
this conjecture it has been argued (1) that the Scheldt does not flow
into the Meuse ; (2) that if Caesar had marched in the direction of
^ Neue Jahrb. f. Philologie, &c., cxxv, 1882, pp. 218-20.
- Meusel's Tabula Coniecturarum, p, 23 {Lex. Gaes., vol. ii, pars ii).
^ Philologus, xxii, 18G5,
pp. 159-60.
4 M. Jullian {Eev. des etudes anc, x, 1908,
pp. 266-7, with which cf. pp. 383-4,
supra) has discovered a site which he believes to correspond with Caesar's
account of Labienus's movements both in 54 and in 53 b. c,
^ B. G., vi,
33, § 3.
 — —

SCHELDT OR SAMBRE ? 735

the Scheldt, he would not have marched towards the most distant '

parts of the Ardennes ' and (3) that he could not have marched
;

from Atuatuca to the Scheldt and back in seven days. In reply to

the first of these arguments Des Roches ^ affirms that the Old Meuse
(Oude Maas) did communicate with the Scheldt near the island of
Tholen ; and, according to Ortelius's map of Brabantia in his '
'

Theatrum Orhis Terrarum (ed. 1573), this statement is true. In reply

to the second it may be said that Caesar does not imply that the
Ardennes extended as far as the Scheldt he merely says that he
:

intended to march towards the Scheldt and the most distant parts
of the Ardennes, which extended north-westward as far as the
country of the Nervii ^ and if he marched towards the Scheldt, he
;

may have been marching towards the most distant parts of the
Ardennes. The third argument is worthless for Caesar does not
;

say that he marched as far as the Scheldt and if he intended to do

;

so, he probably did not know the distance.

T. Bergk, who is nothing if not original, conjectures that Caesar
wrote (ad flumen) Calhem (quod infiuit in) Mosellam. Calhem, he
means the Kyll, which Ausonius called Galhis? I need not
asserts,
discuss this guess.
It has also been suggested that Caesar confounded the Scheldt
with the Sambre in other words, that although he wrote Scaldim,
;

he intended to march towards the Sambre and believed that it was

the Scheldt. I have refuted this suggestion on page 674.

THE RAID OF THE SUGAMBRI IN 53 B.C.

Caesar says that the Sugambri raised 2,000 horsemen and crossed
the Rhine cogunt equitum duo milia Sugambri, &c. (B. G., vi, 35,
:

§ 5). I originally understood this passage in the accepted sense,

namely that the expeditionary force of the Sugambri consisted of
2,000 horse, and no more. A. de Vlaminck, however, maintains ^
that this interpretation is wrong, and that the cavalry formed only
a portion of the force. This, he says, is proved by the statement of
Caesar ^ that, when the Sugambri saw the foraging party which
Cicero had allowed to go out of camp, returning, they advanced to
attack them, despising the smallness of their numbers
'
despecta ' :

paucitate ex omnibus partibus impetum faciunt. The foraging party

consisted of five cohorts, 300 convalescents, a large number of camp-

1Archives hist, et Hit. du Nord de la France et du Midi de la Belgique, 3" ser.,

V, 1856, pp. 339-40. In support of the correctness of the statement in the
Commentaries, it has also been remarked that Pliny {Nai. Hist., iv, 14 [28], § 100),
while mentioning that the Meuse and the Rhine flow directly into the sea, does
not say the same of the Scheldt ; but see § 98. A writer in Bull, de la Soc. de
geogr. d'Anvers, 1877, pp. 175, 185, believes that he has discovered a deserted
bed by which the Scheldt once joined the Meuse.
E.G., vi, 29, § 4.
•'

Zur Gesch. und Topogr. d. Rheinlande in r6m,. Zeit, ] 882, pp. 33-4.
=•

* La Menapie, &c., 1879, pp. .50-2. B. G., vi, 39, § 4.

•'
736 THE RAID OF THE SUGAMBRI
followers, and not more than 200 horsemen,^ probably considerably—
more than 2,000 men all told. As the Germans despised the smallness
of this force, it is reasonable to infer that they outnmnbered it and ;

de Vlaminck is probably right. The explanation may be that,

according to German custom,^ the cavalry were accompanied by
a number of light-armed footmen.

THE CHRONOLOGY OF B. G., vii, 1

Caesar's narrative, which I followed without misgiving in the first

edition (pages 101-2), suggests, if it does not imply, that in the
winter of 53-52 B.C. the Gallic conspirators did not begin to form
their plans until after they had heard of the murder of Clodius.
But M. Jullian ^ is probably right in arguing that Caesar disregarded
the order of events. The news of the murder had to travel 800 miles
and more before it reached the conspirators. The conspiracy was
not matured in a day. Caesar did not leave Italy until he heard of
the fall of Cenabum. Then followed the movements which I have
described on pages 134-47. Caesar marched to Narbonne from ;

Narbonne to the Vivarais thence across the Cevennes into Auvergne

;
;

thence by way of Vienne to Sens thence by way of Orleans to Avari-

;

cum (Bourges). The siege of Avaricum lasted 27 days and when it ;

ended winter was hardly over. Is it not probable that the news of
the murder of Clodius only gave the final stimulus to a movement
which was already in train ? Indeed Caesar himself seems to hint
that this was what happened when he says that the Gauls were '

already smarting under their subjection to the Roman People, and

'
they now began, unreservedly and boldly, to form projects for war
(qui iam ante se populi Romani imperio suhiectos dolerent, liherius
at q lie audacius de hello consilia inire incipiunt)^

HOW WAS THE NEWS OF THE CARNUTIAN RAID

ON CENABUM SPREAD ABROAD ?

'
Caesar, spread swiftly to all the tribes of Gaul
The news,' says '

for whenever an event of signal importance occurs, the people make

it known by loud cries over the countryside, and others in turn take
up the cry and pass it on to their neighbours. So it happened on
this occasion (celeriter ad 07nnes Galliae civitates fama ferfertur ;
'

nam uhi quae maior atque inlustrior incidit res, clamore 'per agros
regionesque significant ; hunc alii deinceps excipiunt et proximis
tradunt, ut turn accidit. B. G., vii, 3, § 2). On this Napoleon remarks
^

that '
An ancient manuscript belonging to Upper Auvergne, the
manuscript of Drugeac, informs us that the custom continued long
1 B.G., vi, 36, §§ 2-3, and Schneider's Caesar, ii, 301, note.

=See B. G., i, 48, §§ 5-7. ^ Hist, de la Gaule, iii, 415, n. 1.

* B. G., vii, 1, § 3/ ^ Hist, de Jules Cesar, ii, 241, n. 1.
 ;

HOW THE GAULS SPREAD NEWS 737

in use and tliat it still existed in the Middle Ages. Kough towers
were built for this purpose on the heights, 400 or 500 metres apart
watchmen were placed in them, who transmitted the news from one
to another by sonorous monosyllables. A certain number of these
towers still exist in the Cantal. If the wind prevented this mode of
transmission, they had recourse to fire. It is evident that criers had
been posted beforehand from Genabum to Gergovia, since it was
agreed that the Carnutes should give the signal of the war.' No
such agreement is mentioned by Caesar. He merely says that it was
agreed that the Carnutes should strike the first blow.^ If Napoleon
is right, criers must have been posted beforehand along every route
' '

leading from Cenabum for Caesar says that the news flew rapidly
;

to all the states of Gaul. Can anything be more grotesque than this
notion, of criers posted beforehand ', standing expectant on all
'

the great thoroughfares of Gaul, and bawling out news from one
to another ?
Colonel T. A. Dodge fails to see that Napoleon was not such
a fool as to commit himself to the statement that the towers existed
in Caesar's time, and naively remarks that the news was passed '

along by men stationed in towers on convenient hills, who gave out

shouts of peculiar kinds'.- Colonel Dodge's narrative is hardly less
peculiar. A country in which convenient hills ', dotted with towers,
'

are to be found, at regular intervals of 500 yards, must be worth

exploring.
Long^ examines the passage with his usual common sense. '
What
Caesar describes,' he says, is simple enough.
'
The country was
populous, and great news was quickly carried from one spot to
another ... to make it like a telegraph, there must have been stationed
persons in readiness, always waiting for the news, and at distances
within which the human voice could clearly convey intelligence ;

not a trace of which is there in Caesar.'*

WHERE DID CAESAR CONCENTRATE HIS LEGIONS

AT THE OUTSET OF THE SEVENTH CAMPAIGN ?
Caesar says that, after leaving Decimus Brutus in the country of
the Arverni, he made his way into the country of the Lingones,
where two of his legions were wintering and that, on his arrival,
;

he sent messengers to the other legions, two of which were quartered

near the western frontier of the Treveri, and the remaining six at
Agedincum (Sens), and concentrated the whole ten in one spot before
Vercingetorix, who was marching from the Berri into Auvergne, was
informed of his arrival.^ Vercingetorix, as soon as he heard what
Caesar had done, determined to besiege Gorgobina whereupon ;

' B. a., vii, 2, §§ 1-2. . 2 Caesar, p. 230. ^ Caesar, p. 330.

* M. Julhan {Vercingetorix, p. 119), like Napoleon, affirms that Des crieurs, '

tout autour de Genabum, avaient ete disposes a travers champs et forets

jusqu'aux extre mites de la Gaulo but as I remain sceptical, I let my original
'
;

coriiment stand. '

B. G., vii, 0.
1093 3 B
738 CONCENTRATION OF THE LEGIONS
Caesar, leaving, as he says,^ two of his legions and all his heavy
baggage at Agedincum, marched to relieve the threatened town.
From this last statement of C^^aesar it has generally been inferred
that the spot at which the ten legions were concentrated was Agedin-
cum but the statement only proves that they were there imme-
;

diately before Caesar set out for Gorgobina.

Colonel StofEel ^ combats the common view, on the ground that,
first, Caesar himself ^ gives us to understand that the concentration
was effected somewhere in the country of the Lingones secondly, ;

time would have been saved by concentrating on the central one of

the three posts occupied by the legions rather than on either of the
other two ^ and thirdly, if Agedincum had been the place of con-
;

centration, Caesar would have written ad reliquas legiones mittit,

priusque omnes Agedincum (instead of in unum locum) cogit,^ &c.
I believe that Stoffel is right. When we read that Caesar sent
messages from the camp of the two legions in the country of the
Lingones to the other two divisions in the country of the Treveri and
at Agedincum respectively, and concentrated them in one spot, we
are justified in inferring that that spot was not Agedincum for the ;

meaning of his words appears to be that he sent/or the two divisions.^

The other two arguments appear to me not less sound.
Stoffel conjectures that the central post was Chatillon-sur-Seine ;

but it is obviously impossible to get beyond conjecture.

ON THE MEANING OF ALTERO DIE (B. G., vii,

1L§1; 68, §2)
I discuss this question because, unless one can find an answer to
it, it is impossible to search, with any prospect of success, for the
site either of Vellaunodunum or of the battle-field on which Caesar
defeated Vercingetorix just before the blockade of Alesia.'
Napoleon believes that altero die means the second day after '.
'

The former of the two passages in the Commentaries on the Gallic

War in which the expression occurs, runs thus duabus Agedinci
:

legionibus atque impedimentis totius exercitus relictis ad Boios pro-

ficiscitur. Altero die cum ad oppidum Senonum V eUaunodunmn
venisset, &c. Supposing that Caesar left Agedincum on March 1,
Napoleon would say that he reached Vellaunodunum on March 3.
Napoleon's arguments may be thus summarized ^ (1) he quotes :
—
a passage in which, speaking of the events that followed a certain
session of the Senate, Cicero says, proximo, altero, tertio, denique
1 B. G., vii, 10, § 4.
2 Guerre de Cesar et (VArioviste, pp. 142-3. ^ B. G.. vii,
9, §§ 4-5.
^ Remember that when Caesar gave the order for concentration he did not

yet know what the next move of Vercingetorix would be.

^ B. G., vii,
9, § 5.
® M. Jullian {Hist, de la Gaule, iii, 431, n. 4) disposes of the word mittit by

denying that reliquas includes the six legions at Agedincum. I cannot follow
him. >
See B. G., vii, 10, § 4; 11, § 1 68, § 3.
;

^ Hist, de J ides Cesar, \\, 246, n. 1.

 —

ALTERO DIE 739

reliquis consecutis diehus} &c. Is it not evident,' asks Napoleon,

'

'
that here aUero die signifies the second day which followed the
session of the Senate, or two days after that session ? He goes on '

to quote several other passages, which I omit, because they only

prove, what nobody denies, that alter means second '. (2) He '

states that in the whole of the Commentaries the expression fostero

die is to be found sixty-three times, 'proximo die thirty-six, insequenti
die ten, postridie eius diei or pridie eius diei eleven, and altero die
only five. Does it not appear certain,' he asks, that, if Caesar
' '

had arrived at Vellaunodunum the day after his departure from

Agedincum, he would have written, Postero die (or proximo die) cum
ad oppidum Senonum Vellaunodunum venisset ? &c. (3) He asserts '

that Caesar defeated the cavalry of Vercingetorix on the Vingeanne,

and arrived altero die at Alesia, 40 miles off, which distance he —
could not have completed until the second day after the battle.
No weight can be attached to any of these reasons. The quotation
—
from Cicero only proves w^hat everybody admits that altero means —
'
second '. In the quotation, altero means second in relation to ' '

proximo. Therefore, says Napoleon, in B. G., vii, 11, § 1, it means

'
second in relation to the day following that on which Caesar left
'

Agedincum. No his opponents reply, it only means

! second in ' '

relation to the day on which he left Agedincum. In other words,

in the quotation from Cicero, the use of the word proximo makes all
the difference. Or, as Long puts it,^ in the quotations which Napoleon
gives, there is a word which expresses directly or indirectly, " first,"
'

and so " alter " of course means " second ".'

Napoleon's second argument is equally weak. Supposing that some
future commentator were to discover that, in So-and-So's History
of the Decline and Fall of the British Empire, the expression on the '

next day occurred ten times, and the expression on the following
' '

day a hundred times would that prove that on the next day '
' :
'

and on the following day meant two different things ?

' '
On
Napoleon's theory, postridie would mean something different from
postero die.
The third argument is absolutely worthless. There is no proof
that the cavalry combat between Caesar and Vercingetorix was
fought on the Vingeanne.^ In fact, in order to bolster up the feeble
evidence which he adduces in support of that site. Napoleon is
obliged to bring forward his interpretation of altero die !

In Virgil's Eighth Eclogue occurs this line, Alter ah undecimo

tum me iam acceperat annus. Authorities,' says Conington,^ were
' '

"
at one time divided on the question whether " alter ab undecimo
meant the twelfth or the thirteenth. Modern editors have found . . .
"
in deciding it to be the twelfth, considering " alter
little difficulty
to be convertible with " secundus ", but following the inclusive
mode of counting.' With this passage Conington compares alter ah

' Phil., i, 13, § 32.

^ Decline of the Roman Republic, iv, 334, n. G.
^ See pp. 795-6. * Vergil, \, 87.
3b2
740 ALTERO DIE
and heros ah AcJiille secundus? He might have added that the
illo 1
Eomans called the 29th of June the third, not the second day before the
1st of July, and that in Greek ttj ere/aa means the same as rfj va-repaLa.
Furthermore, in one of the three passages in which altera die
occurs in De Bello Civili (iii, 19, § 3) Napoleon might have found
proof positive that his theory was wrong. Caesar is describing the
colloquies that passed between his troops and the Pompeians on the
banks of the Apsus. Speaking of his lieutenant, P. Vatinius, he says,
Multa suppliciter locutus est, ut de sua atqiie omnium salute debebat,
silentioque ah utrisque militihus auditus. Responsum est ah altera
parte, Aulum Varronem profiteri se altera die ad conloquium ven-
turum atque eundem visurum quern ad modum tuto legati venire ei
quae vellent exponere possent ; certmnque ei rei tempus constituitur.
Quo eum esset postero die ventum, magna utrimque multitudo con-
venit, magnaque erat exspectatio eius rei, atque omnium animi intenti
esse ad pacem videhantur, &c. Is it not clear that altero die and
postero die in this passage mean the same thing ? [Since I wrote the
above, I have found that Heller (Pkilologus, xxvi, 1867, 682-4)
quotes the same passage to prove the same point and his argu- ;

ments are endorsed by Meusel (Lex. Caes., i, 239).]

THE PASSAGE, QUI TUM PRIMUM . . . COM-

PARABANT (B. G., vii, 11, § 4)

(Caesar) pushed on for Cenabum in the country of the Carnutes.

'

Believing that the siege of Vellaunodunum, news of which had only

just reached them, would be protracted, they were beginning to
collect troops to send to the protection of Cenabum (ipse '
. . .

Cenahum Carnutum proficiscitur. Qui turn primum adlato nuntio

de oppugnatione Vellaunoduni, cum longius earn rem ductum iri
existimarent, praesidium Cenahi tuendi causa, quod eo mitterent, cot)i-
parahant). Mommsen^ regards tuendi causa as an interpolation,
and explains eo as referring to Vellaunodunum. The praesidium, he
argues, was being assembled at Cenabum, not for its protection but
for that of Vellaunodunum. As if the Carnutes were likely to
perform this act of altruism They were of course collecting the
!

praesidium for their own protection ; and from over-confidence

they had neglected to do so before.

ON THE PASSAGE VICOS . . . VIDEANTUR (B. G.,

vii, 14, § 5)

Describing the arguments by which Vercingetorix persuaded his

followers to burn their towns and granaries, Caesar writes vicos atque
aedificia incendi oportere hoc spatio [a Boia] quoque versus, quo pabu-

'
Virgil, Eel, v, 49. * Horace, Sat., ii, 3, 193,
^ Jahresb. d. pliiloL Vereins zu Berlin, xx, 1894, p. 208,

i
 —

A notA OR AB VIA ? 741

landi causa adire posse videantur. The words which I have bracketed
appear in various forms ab oia, aboia and a hoia in the best MSS. —
Boia^ on the analogy of Venetia} would mean the territory of the
Boi. Scaliger bracketed the word other scholars regard it as
:

a marginal gloss, which crept into the text ^ and the French Com- ;

mission^ justly remark that it adds nothing to the clearness of

Caesar's narrative. Why should Vercingetorix have decided to burn
the towns all round Boia, and yet to spare Boia itself, from which
the Romans expected supplies ? Von Goler, indeed,, remarks that the
Boi were in a position to defend themselves, and would not have
submitted quietly, on Caesar's approach, to the burning of their
towns.4 But it seems clear that if Vercingetorix was master of all
the country between Caesar's camp at Avaricum and the frontier of
the Boi, he could, if he had wished to do so, have devastated the
country of that feeble tribe ^ as well. What likelihood was there
that Caesar, who was encamped at Avaricum, in the country of the
Bituriges, would send out foraging parties to the further side of Boia,
which w^as in the country of the Aedui ? And if Caesar was alluding
in this passage to ravages which Vercingetorix ordered to be carried
out in the territories of the neighbouring states,^ why should Ver-
cingetorix have specified Boia as the central district round which
this devastation was to be enacted ? Besides, if the country on all
sides of Boia was devastated, the Aedui, in whose country Boia was
situated, must have burned their towns but it is certain that they
;

did not, for they had not yet joined Vercingetorix. A. Holder reads
ah via, an emendation of Madvig,'' which is also found in the old
edition referred to by Schneider as Ven. c and Dittenberger adopts
;

Hoffman's emendation ohvia. H. Schickinger ^ suggests that the true

reading may be (vicos atque aedificia incendi oportere) ah hoc spatio
omnia (quoque versus). He thinks that oia was an abbreviation,
misunderstood by a copyist, for omnia, and that ah was transposed

1 B. G., iii, 9, § 9.
^ See Schneider, ii, 367-8, and Nipperdey, pp. 89-90.
^ Did. arch, de la Gaule, i, 170. * Gall. Krieg, 1880, p. 242, n. o.
^ civitas erat exigua et infirma. B. G., vii, 17, § 2.
^ Hoc idem fit in reliquis civitatibus. lb., 15, § 1. Since the publication
of my first edition I have noticed that Mommsen {Jahresb. d. phitol. Vereins
zu Berlin, xx, 1894, p. 208) insists that hoc idem fit in reliquis civitatibus
(' the same thing happened in the territories of the other tribes '), following the

sentence in which Caesar says that in a single day more than twenty towns
'

belonging to the Bituriges were set ablaze' {uno die amplius XX

urbes Biturigum
incenduntur), must be an interpolation for, he argues, it is inconceivable that
;

all the tribes of Gaul should have been required to burn their towns ; and the
words iyi omnibus partibus incendia conspiciuntur (' the whole country was
a scene of conflagration') can only refer to the Bituriges. But there is no
(iuestion of all the tribes of Gaul
'
' obviously in reliquis civitatibus only means,
:

as in B. G., iii, 9, § 3, the other tribes immediately concerned.

' '
Still Mommsen
is probably right; for, although M. Jullian {Hist, de la Gaule, iii, 441, n. 5)
remarks that reliquis civitatibus may denote the Carnutes and the Senones,
tires in their territories could not have been seen from the neighbourhood of
Avaricum; Q,nd conspiciuntur presumably refers only to what Caesar himself
saw. -
Adv. crif., ii, 1873, pp. 250-7.
* Wiener Studien,xx\'i, 1904, pp. 342-3.
742 A BOIA OR AB VIA ?

from its proper place. Various other absurd conjectures have been
made. If one must have a conjecture, nothing could be better
than Madvig's but for historical purposes the sense is perfectly
;

clear ;and a hoia is perhaps an interpolation. [Meusel in a marginal

note on his copy of my first edition remarked that Madvig's con-
jecture is inadmissible, because ah is never found before v but do ;

we not find ah Venere (Corpus inscr. Lat., vi, 2274), ab veste (ib., 5197),
ap via {ib., ii, 4926), and ah Nat. Hist., xvii, 22, § 196) ?
vite (Pliny,
It must, however, be admitted that Caesar never, as far as we know,
wrote ab before v.]

THE MEANING OF MISERICORDIA VULGI (B. G.,

vii, 15, § 6)

Datur petentibus venia, dissuadente primo Vercingetorige, post conce-

dente et precibus ipsormn et misericordia vulgi. Long^ interprets
misericordia vulgi as pity for the common sort ', who, he adds,
'

'
would have been turned out of their houses in the winter.' But
this, asSchneider ^ shows, is certainly wrong. Since Vercingetorix,' '

he writes, as may be gathered from his own speech, felt no pity,

'

and the others who were present at the council do not appear to have
considered the common sort any more than the whole body of the
inhabitants of Avaricum, the word vulgus must be understood
. . .

of the majority of the council, from whom a few, possibly the leading
men, dissented.' There is, moreover, a passage in ch. 28 quos iUe
multa iam nocte silentio ex fuga excepit, veritus ne qua in castris ex
eorum concursu et misericordia vulgi seditio oreretur which suggests —
that in ch. 15 vulgi is a subjective, not an objective genitive, and
that misericordia vulgi means pity felt by the mass (of those present
'

at the council).' Another passage (ch. 29) completes the proof. In

the speech which he made to encourage his men after the fall of
Avaricum, Vercingetorix is reported to have said, sibi numquam
placuisse Avaricum defendi, cuius rei testes ipsos haberet ; sed factum
imprudentia Biturigwn et nimia ohsequentia reliquorum uti hoc in-
commodum acciperetur. Nimia
ohsequentia reliquorum obviously
describes the same feeling as misericordia vulgi. Great men of action
do not allow themselves to be turned away from a wise policy by
'
pity for the common sort or pity for any one else.
'

WHERE DID VERCINGETORIX MAKE HIS FIRST

CAMP DURING THE SIEGE OF AVARICUM ?

Caesar tells us that Vercingetorix, following Caesar by easy stages

'

[from the neighbourhood of Noviodunum, that is from the north],

selected for his encampment a spot protected by marshes and woods,
^
Caesar, p. 340, and Decline of the Roman Republic, iv, 290.
^ Caesar, ii. 374.
 GALLIC CAMPS NEAR AVARICUM 743

16 miles from Avaricmn '.^ Von Goler,^ who holds that Vercin-
getorix would not have encamped between Caesar and the Boi, fixes
on a site near Vierzon, north-west of Avaricum but General Creuly
:

makes light of this argument, remarking that the Boi were not
strong enough to attack the rebels;^ and, assuming that Vercinge-
torix started from the neighbourhood of Chatillon-sur-Loire, which
he identifies with Noviodunum,*he believes that Vercingetorix halted
on the heights behind Morogues, about 15 miles north-east of Avari-
cum.^ Napoleon thinks that he encamped about a mile and a half
north of Dun-le-Roi, to the south of Avaricum. It was indeed
'

natural,' he says, that he should place himself between the Roman

'

army and the land of the Arverni, whence probably he drew his
provisions.' ^ But why should he not have drawn them from the
Bituriges, who were his allies and were on the spot ? Colonel St. Hy-
polite argues with good reason that he would not have encamped
'^

on the south of the town, because the Romans attacked it on that

side, and because the marshes, by which he communicated with the
garrison, were on the north. Accordingly he selects for the site of
the Gallic encampment Allean, which is near Baugy.
It is difiicult, with such slender data as Caesar gives, to determine
the site exactly. But there are indications in his narrative which
point to the conclusion that Vercingetorix encamped on the north
or north-east. When Avaricum was stormed, the garrison made '

a rush for the furthest quarter of the town {ultimas oppidi partes
'

continenti impetu petiverunt).^ As Caesar attacked from the south,

the few who escaped the massacre and rejoined Vercingetorix pre-
sumably got out of the town on the north or north-east. Moreover,
as M. Jullian points out, Vercingetorix could have communicated
with Avaricum (as he did every day ^) more easily from the north-
east than from the north-west, because on the former side the marshes
are broken and comparatively small, on the latter the road is often
inundated. 10

WHAT WAS THE HILL NEAR AVARICUM ON WHICH

THE GALLIC INFANTRY AWAITED CAESAR'S
ATTACK ?

Caesar says that the hill was nearer to Avaricum than Vercin-
getorix's original encampment, which was 16 Roman miles off that ;

its slope was gentle (collis leniter ah infimo acclivis) that it was ;

2 Q^ii Xrieg, 1880,

1
5. (7., vii, 16, § 1. p. 242, ii. 5.
Eev. arch., nouv. ser., viii, 1863, p. 396. * See
' p. 462.
^ Rev. arch., viii, 1863,
p. 400. ^ Hist, de Jules Cesar, ii, 255, note.

' Becherches sur qnelqnes points hist, relaiifs au siege de Bourges execute par

Cesar, 1842, pp. 10-1. « B. G., vii,'28, § 2. ^'ib., 16, § 2.

^" C. JulHan, Vercingetorix,
p. 359 and n. 4 ; Hist, de la Gaule, iii, 442, n. 2.
M. Jullian also argues that Caesar was cut off from communication with the
north-east, as he got no supplies from the Senoncs. I doubt whether, at that
stage of the campaign, the Senones, who were hostile {B. (J., vii, 4, § 6), would
have supplied him even if the road had been open.
744 GALLIC CAMPS NEAR AVARICUM
almost entirely surrounded by a niarsli not more than 50 feet wide ;

and that there were dense woods close by.^ If I am right in con-
cluding that Vercingetorix's first camp was north or north-east of
Avaricum, the hill must of course be looked for on the same side.
Von Goler^ places it on the right bank of the Yevre, between
Vignoux and Mehun but General Creuly ^ denies that this position
:

is a collis at all ^ and he affirms that the Yevre, in this part of its
;

course, is too swift to have been called a marsh (palus), unless it

overflowed its banks, in which case the marsh formed by it must
have been far more than 50 feet wide. He himself fixes upon Baugy.
Napoleon places the hill at La CTieneviere, east of Caesar's camp. .

Von Kampen ^ and Mr. W. C. Compton^ agree in placing it on

the left bank of the Auron, opposite St. Just,
—
a position,' says
'
\

Mr. Compton, corresponding with the description in chapter 19.'

'

Colonel St. Hypolite finds it at Maubranches. M. Jullian ^ tenta-

'^

tively proposes a site on the same side of the town, between Les
Aix and Rians, which corresponds exactly with Caesar's description ;

but he admits that St. Hypolite's choice may be right.

Evidently the position can only be guessed at and in this case
;

it matters little whether our guesses are right or wrong.

HOW CAESAR WAS OUTWITTED BY

VERCINGETORIX
Paul Menge ^ has recently argued that in the affair near Avaricum
which is described in B. G., vii, 18-9 and on pages 141-2 of this book,
Caesar was duped by his youthful antagonist. Menge points out
that he acted simply on the statements of prisoners, although as
a rule he avoided taking any step on their unsupported testimony .^^
If, he says, Caesar had ascertained beforehand the nature of the
position which Vercingetorix's infantry had occupied the position —
—
which is the subject of the preceding article he would never have
dreamed of attacking it. He might have reflected that Vercin-
getorix would not have left his infantry unprotected in order to
throw his cavalry against the Roman foragers, whose whereabouts he
might not be able to divine.^^ Clearly his object was to entice Caesar
to attack the encampment in order that its occupants might be

' B. G., vii, 18, 19, §§ 1-2. 2 Gall. Krieg, 1880,

p. 247.
^ Bev. arch., nouv. ser., viii. 1863, p. 397.
* It is represented as a hill in von Goler's map, vvhicli, however, as Creuly

remarks, is imaginary. See Sheet 122 of the Carte de VEtat-Major (1 80,000). :

'
Quindecim ad Caesaris de b. G. comm. tabulae, ix.
^ Caesaris Seventh Campaign in Gaul,
p. 80.
^ Becherches sur qnelques points hist, relatifs au siege de Bourges, &c., 1842,

* Vercingetorur, p. 175 ;de la Gaule, iii, 444, n. 0.

Ilisi.
^ Neue Jahrb.f. Alterium, &c., xv, 1905, pp. 520-3.
d. klass.
'» lb., p. 520, Cf. H. Mcusel, Lex. Cues., i, 449-50, and B. C, ii. 39, ^ 3.
" Cf. B. G., vii, 18, § 1 with IC, § 3.
 — —

CAESAR OUTWITTED BY VERCINGETORIX 745

encouraged by Caesar's ignominious failure for he had organized ;

a service of patrols along the road between the encampment and

Avaricum and the speech in which he vindicated his conduct ^
;

proves that he had not only set a trap for Caesar, but, like a master
of mise en scene, had prepared for the outcry which he foresaw would
follow his return. Caesar's story is literally correct but, as he ;

naturally did not choose to proclaim his own humiliation, it must

be read between the lines.
Menge's paper deserves careful study.

WHERE DID CAESAR PLACE HIS TOWERS DURING

THE SIEGE OF AVARICUM?
Different opinions have been expressed regarding the position of
the towers which Caesar mentions in his description of the siege of
Avaricumi.2 Long ^ says that the agger was protected by two towers,
'

one on each side of the terrace and not on it.' Von Goler,* Napoleon,^
and Colonel Stoffel ^ think otherwise and indeed there is no evidence
;

in any of the chapters in which Caesar mentions the towers to show

that they were on each side of the terrace
'
while there is evidence
'
;

in B. G., vii, 22 and 24 to show that they were on it. The passage in
the latter chapter runs celeriter factum est ut alii tunes reducerent . . .

aggeremque inter scinderent (meaning that the towers were drawn back
out of reach of the fire which the Gauls had kindled in the front part
of the agger) while in the former Caesar says that, day by day
;

during the siege, as his own towers were raised to a higher level by
the daily increase in the height of the agger, the enemy added fresh
stories to the towers w^hich they had themselves erected upon the
wall.''' Hirtius also says that a tower was erected upon the agger
which the Romans made at Uxellodunum extruiiur agger in :

ahitudine^n fedum LX, conlocatur in eo turris,^ &c. To clinch the

proof, Stoffel remarks that the towers could not have been moved
except on the levelled surface of the terrace, and that, as it was the
business of the artillerymen who manned them to drive the enemy
from the wall of the besieged town, they must have dominated the
wall.9

'
B.Q., vii, 20, §§ G, 9.
•^
B. 0., vii, 17, § 1
; 22, § 4
; 24, §§ 3, 5 ; 25, §§ 1-2.
18, § 1 ;

^ Decline of the Boman

Bepublic, iv, 300, n, 9.
* Gall. Krieg, 1880,
p. 251.
^ Hist, de Jules Cesar, ii, 261. « Guerre civile, ii, 361.

' See the next article. 8 B. G., viii, 41, § 5.

^ Besides the passages which I have quoted, 8toffel refers to Lucan, iii,
394-8, 455-7. Lucan writes :

Stellatis axihus agger

Erigiiur, geminasque acquanles moenia iurres
Accipil.
 746

THE MEANING OF AGGER IN B. G., vii, 22, § 4

Describing the expedients devised by the garrison of Avaricum,

Caesar says, nostrarum turrium altitudinem, quantum has cotidianus
agger expresserat, comrnissis suarum turrium malts adaequahant. A
plain man would understand by this that, as the Roman towers rose
daily higher, owing to the rise in the height of the terrace on which
they stood, the Gauls matched their height by adding new stories to
their own towers. But Riistow assumes that agger does not here
mean the terrace ', but the material (timber) which was used for
'

increasing the height of the Roman towers.^ Of course agger does

sometimes mean material '. But in all the other passages in which
'

the word occurs with that meaning, the meaning is unmistakable ^ ;

whereas, if agger, in the passage under discussion, means material '

the meaning is certainly both liable and likely to be mistaken.

Besides, if Riistow is right, the Romans set the example in raising
their towers daily higher by successive stories whereas it is pro-
;

bable that, at the time of which Caesar speaks, they were already
completed.^ Finally, I cannot conceive that Caesar would have
denoted the operation of building additional stories by the word
exfresserat.

THE GALLIC WALL

Caesar, as I understand him, describes the construction of the
Gallic \vall as follows."* —
At right angles to the direction of the in-
tended wall, balks of timber were laid on the ground, parallel to each
other and two feet apart. These balks were braced on the inner side
by beams, 40 feet long, which were laid upon them transversely, and
doubtless mortised into them.^ The intervals between the balks were
tilled up, on the inner side of the wall, with earth or rubble on the ;

outer side, with large stones. The first layer of the wall was now
complete. On the top of it was placed a second layer, exactly like
it. Layer was placed above layer until the required height was
reached. The structure was protected from fire by the earth and
stones, and was so firmly held together by the 40-foot beams that
the battering ram was powerless to destroy it.
Caesar's account, however, presents various difficulties. Speaking
of the second layer, he says, When the balks [of the first layer] are
'

fixed in their places and fastened together, a fresh row is laid on

the top of them, in such a way that the same interval is kept and the

^ Heerwesen und Kriegfuhrung C. J. Cdsnrs, 1855, p. 146,

2 See Meusel's Lex. Cues., i, 211-2.
^ My view is confirmed by Stoffel {Guerre civile, ii, 361-2).
Cf. p. 606, supra.
* B. G., vii, 23.
''
De Saulcy, describing the remains of a Gallic wall, which were discovered at
Mursceint, says that, at the place where the balks were crossed by the beams,
the fastening was secured by enormous bolts of iron. Journ. des Savants, 1880,
p. 025.
 —

THE GALLIC WALL 747

balks do not touch each other, but are separated by similar intervals,
into each of which a stone is thrust, and thus are kept firmly in
position (his conlocatis et coagmentatis, alius insuper ordo additur,
'

ut idem illud intervallum servetur neque inter se contingant trahes,

sed paribus intermissis spatiis, singulae singulis saxis interiectis, arte
contineantur).^ On this Mr. A. G. Peskett remarks,^ It is quite '

possible that the successive layers were so laid, that beam rested on
beam, interval on interval ... if it be so, trahes here will not denote
any two beams, but the several whole vertical lines of beams, which
were prevented from touching each other by the intervening lines of
rubble and stones.' (For beams in the above read balks '.)
' ' '

Schneider ^ interprets the passage in the same way but, unlike

;

Mr. Peskett, he denies that any other interpretation is admissible.

Long, however, points out * that the clause, ut idem illud intervallum
servetur, expresses the intervals between the balks in the second
'

tier '
and he argues that the second clause, neque inter se contingant
;

trahes, is unnecessary unless it means something else.'

'
The balks '

of the second,' he proceeds, were not laid on the balks of the first
'

tier, but on the stones, so that no balks touched one another. . . .

The beams [read balks '] would be also better protected against
'

fire in this way than if they were on one another.'

The arrangement may have been what Long describes Caesar :

does not inform us. But what Long fails to see is that, if his in-
terpretation of the Latin is correct, either the trahes which are the
subject of contineantur denote something different from the trahes
which are the subject of contingant, the latter denoting the balks
both of the first and the second row, the former those of the second
only or, if in both cases trahes denotes the balks both of the first
;

and the second row, paribus spatiis does not mean intervals similar
to those in the first row, but intervals similar to one another. Such
an interpretation seems to me inadmissible.
Remains of Gallic walls have been discovered at Mursceint, in the
department of the Lot, on Mont Beuvray, and in the hill-fort of
Lnpernal, just north of Luzech.^ In the wall of Mursceint there are
three or four stones between the balks of each layer, and three layers
of stones between each pair of layers of mingled stones and balks.
The wall of Mont Beuvray conforms more closely to Caesar's de-
scription.
Now Caesar does not guarantee the absolute and invariable
accuracy of his description. He only professes to describe the
general principles of construction Muri omnes Gallici hac fere
:

forma sunt. Is it not possible that, as the walls which have been
discovered are not all alike, and as Caesar does not mention any layer
of stones, the wall of Avaricum had no such layer ?

1 B. 0., vii, 23, § 3. 2 Caesar, bk. vii,

p. 65.
^ Caesar, ii, 402. * Caesar, p. 347

^ Desjardins, Geogr. de la Gaule rom., ii,

119 ; Bev. arch., nouv. ser., xx, 1869,
pp. 400-8 and pi. xix ; xxi, 1870, pi. vii, x Congres archeol. de France, 1874
;

(1875), pp. 451-85; Soc. prehisl. de France, 1907 (10** rai^port mensucl dc la
comm. d'ctude des enceintes pruhist., p. 2).
748 THE GALLIC WALL
There is another point in Caesar's description which has been
misunderstood. The stone,' he writes, secures it [the wall] against
' '

fire, and the woodwork, which is braced on the inner side by beams,
generally 40 feet long, running right across, and so can neither be
'
broken through nor pulled to pieces, protects it against the ram
{et ah incendio lapis et ah ariete materia defendit, quae perpetuis
trahihus pedum quadrage7iuni plerumque irUrorsus revincta neque
perrumpi neque distrahi potest). I have explained these perpetuae
trahes as beams laid in the direction of the wall and mortised into
the balks. Schneider i and Long ^ identify them with the balks
themselves. But this theory is refuted by the fact that in the Gallic
walls which have been discovered, the 40-foot beams are actually
laid in the way
w^hich I have described. In the wall of Mursceint
there are two beams, laid parallel with each other and the balks, ;

which are about 3 yards ^ apart and 23 feet * long, occupy the whole
thickness of the wall. Moreover, even if no Gallic walls had been
discovered, a moment's reflection might have convinced Long of
the absurdity of his interpretation. In § 2 of the chapter in ques-
tion Caesar says that the trahes derectae distantes inter se hinos pedes
(the balks two feet apart) revinciuntur introrsus (are bound together
on the inner side). In § 5 he describes the materia (the balks) as
being perpetuis trahihus pedmn quadragenum introrsus revincta (bound
together on the inner side with beams 40 feet long). According to
Long, the trahes derectae were identical with the trahes pedum quadra-
genum with which they were bound together !

THE DIMENSIONS OF THE AGGEE AT AVARICTOI

Aggerem, latum pedes CCCXXX, ahum pedes LXXX, extruxerunt.^
Such are the numbers given by all the MSS. and it is better to trust ;

them than to make conjectures. The agger which Trebonius made

at the siege of Massilia was also 80 feet high ^ and here again all ;

the MSS. agree. Napoleon accounts for the great height of the
''

agger at Avaricum by pointing to the depression of the ground in

front of the wall. As to the other dimension, Kiistow^ proposes longum

^ Caesar, ii, 403. " Caesar, pp. 347-8.

^ 2 m. 70 c. m. M ^ B. G., vii, 24,
§ 1.
* Huiiis quoque spatii pars ea quae ad arcetn perlinet loci natiira et vulle

altissima munita lojigarii et difficiJem habet oppugnationem. Ad ea perficienda

opera C. Trebonius vimina materiamque comportari iubet. Quibus comparatis
. . .

rebus, aggerem in altitudinem pedum LXXX

exstruit. B. C, ii, 1,
§§ 3-4. Commandant Rouby, in his study on the siege of MassiHa [Spectafeur
mil., 3" ser., xxxv, 1874, p. 173, n. 1), remarks that some MvSS. have latitudinem
instead of altitudinem. This is a mistake some old editions have latitudinem,
:

but no known MSS. Rouby defends altitudinem on the ground that it cor-
responds better witli the words valle altissima, and that the alleged height
of 80 feet agrees perfectly with the features of the ground on which the agger
must have been constructed. Cf. Stofifel, Guerre civile, i, 291.
' Hist, de Jules Cesar, ii, 25G, n. 1. Cf. C. Jullian, Vcrcingelorix, p. 302, n. 5.
* Heerwesen und Kriegfuhrung G. J. Cllsars, 1855, p. 143.

1
 ;

THE AGGER AT AVARTCUM 749

instead of latum. I am sure that Caesar did not write longum first, :

because it is much more likely that he laid stress upon the width of
the terrace, as allowing room for the assaulting columns, than upon
its length, —
or rather its extension from front to rear and secondly, ;

because it is more likely that, if there is a mistake in the MSS., the

numerals are wrong than the word latum. Practical men like
Guischard and Stoifel find no difficulty in accepting the statement
of the MSS. See my article on the agger (pages 599-607).

HOW WAS THE COLUMN OF ASSAULT COVERED

BEFORE THE STORMING OF AVARICUM ?
'
The legions [unobserved] got ready for action [under cover of
the sheds]. Caesar gave the troops the signal. Suddenly they
. . .

darted forth from every point and swiftly lined the wall (legionesque '

[intra vineas in occulta'] expeditas cohortatus militihusque signum

. . .

dedit. Illi suhito ex omnibus partihus evolaverunt murumque celeriter

com'pleverunt).^ Intra vineas in occulto is the reading of ^. The read-
ings of a are extra vineas (in x)> ^xtra castra vineas (in </>), and extra
castra vineasque in L, in occulto being omitted in both ^ and <^ but
not in L. None of these readings can be right for the first and the;

third are pointless and the second is nonsense. Meusel says that
expeditas can hardly be a participle,^ because expeditus in connexion
with legiones, cohortes, &c., is regularly an adjective and that, if it ;

is an adjective, in occulto is impossible. He points out that intra

vineas must mean either inside individual vineae or within a space
enclosed by vineae, and that troops could only have been concealed
by vineae if they had been inside them and he scoffs at the notion
;

that vineae would have sufficed for the concealment of eight legions.^
Accordingly he brackets intra vineas in occulto.^ But Caesar did not
always write with precision and if only the foremost part of the
;

force was simultaneously concealed intra vineas, he might perhaps

have spoken loosely of the legions as intra vineas. Heller,^ in order
to get over the difficulty, invents or borrows from Achaintre ^
the emendation inter castra vineasque, which so nearly agrees with
the reading of L that it deserves consideration. But, if the
troops had been standing only where Achaintre and Heller place
1 B.G., vii, 27, § 2.
- In his Lexicon, however (i, 1226), he took it as such.
^ Jahresh. d. philol. Vereins zu Berlin, xx, 1894, pp. 32G-8.
* It does not matter for our present purpose whether we read legionibusque
expeditis, which Meusel adopted in his critical edition, or legionesque expeditas,
the reading of 0, which he adopts now. He rejects the former because if, as he
concludes, in occulto is spurious and expeditus is here used adjectivally, the
ablative absolute, legionibus expeditis, would plainly be inadmissible.
^ Philologus, xix, 1863, p. 534.
® Caesar, i, 317.
^
Meusel {Jahresh. d. philol. Vereins zu Berlin, xi, 1885, p, 202) observes
that the double alteration which Heller introduces into the text is suspicious
but que is found in L, which I have recently collated.
750 INTRA VINEAS
them, would not Caesar have written inter casira acjgeremque ? ^ Instead
of intra vineas —
von Goler- suggests interea, a much more impro-
bable conjecture even than Heller's. Guischard/^ who accepts the
reading intra vineas, believes that there were two towers, one on
either side of the agger, and connected with it by vineae that some
;

of the troops belonging to the assaulting force were formed up, as

Heller believes, between the agger and the camp and that the rest
;

were formed up under and immediately behind the connecting vineae.

This suggestion, he says, makes everything intelligible. To my mind,
everything is intelligible without the suggestion and if there is any
;

difficulty, the suggestion only makes matters worse. For Caesar

says that the towers were on the agger, not on either side of it * ;

and the imaginary connecting galleries, in order to be of any use for

the purpose of communicating between the agger and the imaginary
towers, would have had to be closed on the side exposed to the town.
How then were the assaulting columns to get out of them ? And
how were they to escalade the wall unless they first mounted the
terrace ?

The readingintra vineas in occulto derives some support from the

— —
statement which Caesar makes in the next sentence that the soldiers
'
suddenly darted forth from every point and swiftly lined the wall
(suhito ex omnibus partibus evolaverunt murumque celeriter com-
fleverunt) but I do not deny that Meusel is justified in bracketing
;

the words.^
Whatever the right reading may be, this much is clear. Caesar
intended to surprise the enemy and therefore he concealed his troops
;

somehow. If they had only been concealed behind the agger and had
become visible as soon as they set foot upon it, the enemy would

^ In the first edition (p. 732) I argued that the words in occulto, if they
are genuine, throw additional discredit on Heller's emendation, as it is difficult
'

to see how, from the point of view of the sentries standing on the wall, the
storming columns could have been hidden unless they had been inside, or
within a space enclosed by, the vineae . . even if they had not been seen by
.

those sentries who were standing on the wall right opposite the agger, they
would surely have been seen by those who were standing on the wall on either
side of the agger.' This argument was perhaps unsound for the agger was
;

doubtless so close to the camp that it might have been impossible from any
part of the wall to see troops who were just behind it, and although the fact that
the Romans had to ascend the wall (vii, 27, § 2) proves that the wall was higher
than the agger, the difference of level was slight, for they evidently ascended
it easily. See also the last sentence of the article.
M. Jullian {Hist, de la Gaule, iii, 452, n. 5) attempts to improve Heller's
emendation by substituting intra for inter (castra vineasque). But emendation
will not solve the riddle.
2 Gall. Krieg, 1880, p. 258, n. 1. » 3Iem. mil, ii, 22.

* B.G., vii, 22, § 4. See p. 745.

^ To account for the various readings, Meusel makes the following ingenious

suggestion. Some reader may have written in occulto on his copy in the belief
that the assaulting columns must have been concealed ; another may have
added intra vineas as an explanation ; and a third, recognizing that there would
not have been enough room intra vineas for legions, may have written extra
above intra and castra above vineas, meaning that the legions must anyhow
have been extra vineas, but may have been intra castra. Jahresb. d. philol.
Vereins zu Berlin, xi, 1885, p. 202 ; xxxvi, 1910, p. 48.
 ——

INTRA VINEAS 751

have had warning. Therefore the head of the cohimn, at all events,
was probably formed up on the agger and the only means of con-
;

cealing them was to place them inside the vineae or within a space
enclosed by vineae or both.i The rest of the troops would have been
concealed either in the space between the viaducts (if Napoleon is
right in believing that the central portion of the agger occupied only
the front of that space ^) or in the space between the agger and the
camp.

THE STRATAGEM BY WHICH CAESAR CROSSED

THE ALLIER ON HIS MARCH TO GERGOVIA
The passage in which Caesar describes his stratagem runs as
follows :
Cum
uterque utrique esset exercitui in cons^pectu, fereque
e regione castris castra poneret,^ dispositis exploratoribus, necuhi
effecto fonte Romani copias traducerent, erat in magnis Caesari
difficultatihus res, ne maiorem aestatis partem flumine impediretur,
quod non fere ante autumnum Elaver vado transiri solet. Itaque,
ne id accideret, silvestri loco castris positis, e regione unius eorum
pontium, quos Vercingetorix rescindendos curaverat, poster o die cum
duahus legionihus in occulto restitit ; reliquas copias cum omnibus
impedimentis, ut consuerat, misit, c apt is quihusdam cohortihus,
uti numerus legionum constare videretur. His quam longissime possent
progredi iussis, cum iam ex diei tempore coniecturam caperet in castra
perventum, isdem suhlicis, quarum pars inferior integra remanehat,
pontem reficere coepit. Celeriter effecto opere legionihusque traductis,
et loco castris idoneo delecto, reliquas copias revocavit.^ Now this
much is evident. Caesar either kept back two entire legions and
sent on four, or selected 20 cohorts out of the 60 of which his legions
were composed, and sent on 40 and he so arranged the four legions
;

or the 40 cohorts which he sent on that, seen from the opposite

bank of the Allier, they looked like six legions. To me it seems
clear that he meant what he said, —
kept back two entire legions
and sent on four otherwise, he would, I believe, have written
:

(postero die cum) viginti cohortihus (in occulto restitit).

Captis quihusdam cohortihus is found in all the MSS., except Leid.
tert., Dorn., and Duk., which have demptis quihusdam cohortihus ;

And., which has captis quartis quihusque cohortihus Oxon., which

;

has captis quartis quidem cohortihus and Leid. sec, which has captis
;

quartis cohortihus.^ The authority of these MSS. is nil.

' I find that what

I have written is supported by Stoffel, who says {Gnerre
civile, ii, Cesar
363), '
disposa ses legions dans les vineae.'
. . .

- See
pp. 603-4.
^ I follow the reading of )3, which A. Klotz {Caesar siudien,
pp. 256-8) slightly
modifies thus, cum uterque utrique exercitus esset in conspeciu, fereque e regione
cast{rorum Caesa)ris castra (Vercingetorix) poneret. As the accuracy of this
part of the text is for my present purpose unimportant, I need only refer
to his ingenious argument.
* B. 0., vii, 35. 5 Caesar, ed. Nipperdey,
p. 93.
752 CAPTIS QUIBU8DAM COHORTIBUS
The modern editors, with the single exception of Frigell, are
unanimous in holding that Caesar did not write captis, I suppose on
the ground that the word cannot be used in the only sense which
can be here attributed to it. Schneider ^ follows Wendel, who reads
carptis. Carptis quibusdam cohortihus would mean, I apprehend,
'
breaking up a certain number of cohorts into their constituent
parts,' namely maniples and Wendel points out that Livy (xxvi,
;

38, § 2) writes in 7nuUas parvasque partes carpere exercitum, where

some MSS. have capere. Van der Mey ^ refuses to believe that Caesar
wrote carptis qiiihusdam (cohortibus) for, he argues, if Caesar had
;

made 40 cohorts look like 60 by breaking up cohorts into their con-

stituent maniples, he would have been obliged to break up the whole
40 and not merely some (quibusdam). Therefore, if Caesar wrote
carptis, and used it in the sense of breaking up \ he could not have
'

written quibusdam. I cannot see the force of this argument. Why

should not Caesar have sent on three of the four legions in their
usual formation, broken up the 10 cohorts of the remaining legion
into their 30 maniples, and, by diminishing the breadth of each of
those maniples, extended their length so as to make them look like
cohorts ? It would have been impossible for Vercingetorix to detect
that the breadth of the sham cohorts had been diminished for he ;

was probably separated by a distance of several hundred yards, at

least, from Caesar's column.^
Holder,^ following an emendation of H. Deiter, reads ita apertis'

quibusdam cohortibus ', which means much the same as Wendel's

reading. Dittenberger,^ following an emendation of Vielhaber, reads
partitis quibusdam cohortibus but, as partitis is virtually identical
:

in meaning with carptis, and much more unlike captis, the reading
of the best MSS., I cannot see what is to be gained by adopting it.
For the same reason, I wonder that Meusel should have adopted
B. Miiller's conjecture, distractis (quibusdam cohortibus).
Nipperdey ^ proposes maniplis singulis demptis cohortibus, Since,' '

he writes, Caesar kept back two out of his six legions, the remain-
'

ing four had to be arranged in such wise as to look like six. Now,
as there were 30 maniples in each legion, and three maniples in
each cohort, the only way of making six [apparent] legions out of
four legions or 120 maniples was to assign 20 maniples instead of 30
to each legion. The most convenient way of doing this was to with-
draw one maniple from each cohort for in this way the number of
;

the cohorts remained the same.' But Nipperdey's emendation intro-

duces a new word maniplis into the text, which is not found in any
of the MSS. ;and Caesar could have effected his purpose just as
well by sending on 40 entire cohorts, or 120 maniples, and rearranging

Caesar, ii, 435-6.

^ ^ Mnemosyne, xii, 1884,
pp. 226-7.
See Eos, ii, 1866, p. 136, and Carte de VEtat-Major (1 80,000), Sheets 135,
3 :

146. Meusel does not agree with me ; but has he reflected that ordinary eye-
sight cannot distinguish cavalry from infantry at a greater distance than 1,200
yards (see p. 460) ? Anyhow his objection would tell equally against the
emendation {distractis) which he has himself adopted and against any other.
* Caesar,
p. 167. ^ Caesar,
pp. 297, 406-7. ^ Caesar,
pp. 93-4.
 —

CAPTIS QUIBUSDAM COHOETIBUS 753

them so that they should look like 60 cohorts, as by sending on

60 mutilated cohorts, each composed of two maniples.
Other conjectural emendations, which German scholars have
amused themselves by devising, are sectis, laxatis} interceptis, sic
and ita positis (quibusdam cohorti-
aptatis, cavis, dvmidiatis, distractis,
bus) ; quibusque cohortibus)
detractis (quartis and demptis tertiis ;

(quibusque cohortibus). I have no doubt that a good scholar, with

a fertile imagination and an hour to waste, might make considerable
additions to this list.
Long 2 has recourse to one of the inferior MSS. Following Feld-
bausch, he suggests that if Caesar wrote captis quartis quibusque
cohortibus, he may have meant that he took from his 60 cohorts the
first, fourth, seventh and so on, and sent on the rest. Thus,' he '

adds, instead of taking two entire legions, which the enemy might
'

from some circumstance or other have missed, he preserved the six

legions with diminished numbers.' I am certain that Feldbausch
was wrong, for these reasons. First, I believe that Caesar meant
—
what he said, that he kept back two entire legions and not every '

fourth cohort quartis quibusque cohortibus is surely bad Latin.

' :

Why not quarta quaque cohorte ? Secondly, on Feldbausch's theory,

if Caesar had written captis quartis quibusque cohortibus, he would

simply have been repeating, clumsily and obscurely, his previous

statement, cum duabus legionibus in occulto restitit. Thirdly, there
would have been no sense in saying, the rest of the force he sent on
'

as usual, with all the baggage, picking out every fourth cohort, so
that the number of legions might appear unchanged.' Picking '

out,' that is to say, keeping back, certain cohorts would not have
made the number of legions appear unchanged by itself, it would :

have had the opposite effect in order to make the four legions look
;

like six, it was necessary to rearrange them.^ If, then, as M. Jullian *

maintains, captis is defensible, the meaning can only be that certain
cohorts of the four legions were picked out in order that they
'
'

might be rearranged and the four legions might thus look like six.
But is it likely that Caesar would have emphasized the mere picking
out of the cohorts and left their rearrangement to the reader's
imagination ? The text, then, is almost certainly wrong ; and if so,

^ F. Kindscher {Zeitschr. f. d. Gymnasialwesen, 1860, p. 426) defends laxatis

by the argument that Caesar would have made the ranks march further apart
than usual. But this is just what he would not have done, for fear of arousing
Vercingetorix's suspicions. He would have made the breadth of the column
narrower than usual, because from the opposite bank Vercingetorix could not
have seen that he had done so. ^ Caesar,
p. 356.
^ Messrs. Bond and Walpole {Caesar, if captis is the
p. 363) suggest that,
'

right reading, the text may mean " that although some cohorts had been
removed, yet," &c. But we shoiild have expected uti to come before captis.'
Certainly we should and what then would be the sense of iiti numerus legiovum
;

constare videretur What would be the sense of saying, the rest of the troops
'! '

he sent on as usual with all the baggage, in order that, although some cohorts
had been removed, the number of the legions might appear unchanged V '

* Hist, de la Caule, iii, 459, n. 5. After a careful examination of the Thesaurus

liuyuac Latinae I can only conclude that captis cannot be used in the required
sense, though capio is often eciuivalent to eliyo {Thes. liny. Lat., iii, 335).
1093 3 C
 754 CAPTIS QUIBU8DAM COHOETIBUS
we cannot do without an emendation of some sort. Now Professor
Tyrrell ^ has remarked, with much good sense, that the first duty'

of an editor is to see that the ms. tradition is not put aside, unless
it is quite clear that it is wrong, and cannot be reasonably defended.
His next duty is to keep as close as he can to the mss. when he is
obliged to desert them, and never to put forward a conjecture with-
out a theory to account for the corruption '. And again, the best '

copyists ... on meeting an unfamiliar word almost invariably write

down, instead of it, that common word which most closely resembles
it in form, without in the least troubling themselves about the mean-
ing of the sentence.' Not one of the emendations of the passage in
question fulfils the necessary conditions, except carptis. Carptifi, as
I have shown, makes perfectly good sense, and it enables us to form
a perfectly intelligible theory of Caesar's stratagem. Therefore,
unless it is to be condemned as unclassical, carptis is the emendation
for me.'^ But, for purely historical purposes, it matters very little
what Caesar wrote for even if all the MSS. and all the emendations
;

are wrong, it is impossible to mistake his drift.

WHERE DID CAESAR CROSS THE ALLIER ON HIS

MARCH TO GERGOVIA ?
Napoleon*^ fixes uj)on Varennes as the site of the bridge which
1.

C'aesar repaired. He argues that (a) the existence of the bridge

proves that there must have been a road leading to it (6) only two
;

Roman roads, and therefore probably only two Gallic roads led to
—
the Allier below Moulins, one at Varennes, the other at Vichy
(c) the distance of Varennes from Gergovia, nearly 77 kilometres,
or about 48 miles, is just what Caesar would have accompHshed
through a strange country in five marches, the first of which, owing
to the fatigue of the four legions which had done double work on the
previous day, was probably short, and likewise the fifth, because
Caesar arrived at Gergovia early on that day ^ and (d) Vichy is
;

only 55 kilometres, or about 33 miles, from Gergovia.

I do not think that the strangeness of the country would have
interfered with the speed of Caesar's marching, any more than it did
^ The Correspondence of Cicero, vol. ii, pp. vii-viii.
* H. Schickinger {Rhein. J/hs., N.F. Ix, 1905, pp. 639-40) has recently proposed
,

coartatis qiiidem (cohortibus). —

He ascribes to coartatis a verb which is only
—
used in one other passage {B. G., vii, 70, § 3) by Caesar the unusual meaning
of reduced in size '. P. Menge, commenting on his emendation, supports
'

my conclusion, and gives additional instances from Livy and Cicero of the
use of carptis in the sense which I have attributed to it' {Rhein. Mii"^., Ixi,
1906, pp. 306-7. Cf. W. Schott's note in Blatter f. d. buyer. Gymnasialschul-
wesen, 1908, p. 72). Meusel, indeed, warns me that carpo, in the sense of
discerpo, is not admissible unless it is conjoined with in (partes, &c.). Not, as
far as we know, unless Livy's carpere tnidtifar lain vires (iii, 5, § 1) may be
regarded as an exception ; but does not Cicero {De or., iii, 49, § 190) say that
saepe carpenda membris minutioribus oratio est ? And, seeing that the use of
carpo in this sense is extremely rare, are we safe in generalizing ?
' Hist, de Jules Cesar, ii, 267, n. 3. * See B. G., vii, 35-6.
 THE PASSAGE OF THE ALLIER 755

when he was making his rapid march against Ariovistus ;

^ for if

guides were necessary, he had probably procured them. Desjardins

'-^

thinks that the data upon which Napoleon's calculations are based
are too slight to be of any value.
2. Caesar says that, three days after he left Gergovia to rejoin
Labienus, he repaired a bridge over the Allier and crossed by it/"^
D' Anville * infers that this was the same bridge by which Caesar had
crossed the river when he was marching to Gergovia. He assumes
that Caesar, on his return march, accomplished 28 kilometres a day ;

and accordingly he places the bridge a little above Moulins. There

is nothing in this argument. The fact that Caesar had to repair the
bridge by which he crossed the Allier on his return march goes to
prove that it was not the same as the one by which he had crossed
before ; for there is no reason to suppose that the latter had mean-
while been injured. And it is impossible to say how far Caesar
marched in the three days.
3. General Creuly ^ thinks the distance from Varennes to Gergovia
too short to have required five marches. He believes that when
Caesar went to Decetia, or Decize, to settle the dispute between the
rival candidates for the office of Vergobret,^ he left his army at
Noviodunum, or Nevers, where he established an important depot."^
If so, he observes, Caesar must have begun his march for Gergovia
near the point where the Allier flows into the Loire and therefore,
;

as a glance at the map will show, he might have crossed the Allier,
not at Varennes but at Moulins, which is about 18 miles, as the crow
flies, lower down the river. On the other hand, unless he began his
march at Nevers, it is obvious that he could not have crossed the
river so low down as Moulins for, if he started from Decize, he
;

struck the iVllier only just above Moulins and his narrative ^ shows
;

that he had marched a considerable distance up the valley before

he crossed the stream.
Desjardins is quite right in saying that the data are insufficient :

but Napoleon's choice, Varennes, appears to me the most probable ;

^^

for Caesar's remark, that patrols were thrown out [by Vercin-
'

getorix] to prevent the Eomans from making a bridge ',i^ suggests

that he had been obliged to pass one of the broken bridges.

1 B. 0., i, 41, §§ 4-5. 2 Q^og^.^ ^g la, Qaule rom., ii, C76-7, n. 5.

' B. G., vii, 53, § 4. Schneider {Caesar, ii, 494) thinks that when Caesar said
that he reached the Allier tertio die, he reckoned the time from the day on whicli
he rebuked the legions for their disobedience during the attack on Gergovia
{B. G., vii, 52). I believe, on the contrary, that by tertio die he meant the third
day of his march. * Eclair cissemens sur Vancienne Gaide,
p. 444.
^ Eev. arch., nouv. ser., viii,
1863, p. 401. « B. G., vii, 33,
§§ 1-2.
7 lb., 55,
§§ 1-3. Cf. C. Jullian, Hist, de la Gaule, iii, 459, n. 2.
« B. G., vii, 35,
§§ 1-2.
® Von Goler {Gall. Krieg, 1880,
p. 289, n. 1) thinks that Caesar crossed the
river higher up, as being narrower, when he was retreating from Gergovia than
when he was marching to it. But as he crossed by a bridge, which he only
had to repair, it would not have mattered much whether the river was wider
or narrower. The truth is that we are as mucli in the dark on the one point as
on the other. Caesar crossed the Allier where it suited him to cross it, and that
is all that we shall ever know. ^o
B. G., vii, 35, § 1.
3 C 2
 756

CAESAR'S OPERATIONS AT GERGOVIA

Since the publication of Napoleon's Hisloire de Jules Cesar, Caesar's
description of his operations at Gergovia has become more intelli-
gible. The emperor's main conclusions had, indeed, been anticipated
by Pasumot, von Goler, and others ^ but they had not, like him,
;

unlimited funds, skilled officers, and gangs of excavators to help

them. Whoever reads the letter from Colonel Stoffel which I have
printed on pages xxv-xxvii, will be convinced that he really dis-
covered two Roman camps, connected by a pair of parallel trenches,
on the south-east of Gergovia and on the Roche Blanche and that ;

the attempts which have been made to discredit his discovery are
absurd. Even now, however, the commentators are not all of one
mind and no man can discuss the question Avith authority who has
;

not carefully examined the ground with his own eyes.

I. It is no longer necessary to analyse the arguments of J. B.
Bouillet,^ who, notwithstanding Stoffel's discoveries, identified the
larger camp with a camp at Condole, five kilometres and a half, or
nearly three miles and a half, east of Gergovia, and located the smaller
camp on the hill of Bonne val, north-east of Gergovia, and 1,780 metres,
or nearly 2,000 yards, from the plateau. Nobody will believe that
Caesar would have encamped three miles and a half from the town
which he hoped to take. It is true that he describes the hill on
which the smaller camp was situated as scarped on every side,
whereas the Roche Blanche, according to Bouillet, is scarped only on
the south but the hill of Bonneval, as Bouillet himself admits, is
:

only scarped on the south and west and, as Napoleon says,^ The
;
'

Roche Blanche, which presents in its southern part an escarpment

almost as perpendicular as a wall, has lost on the sides its abrupt
form by successive landslips, the last of which took place within
the memory of the inhabitants.' I can myself testify that the Roche
Blanche extremely steep on the southern part of its western side
is ;

and it is plain that Caesar, when he described the hill as ex omni

parte circumcisus could not have meant to say that it was literally
scarped on all sides, but only that it was steep for if it had been
;

as steep on every side as it is on the south, it would have been im-

possible for him to capture it, nor would the Gauls, even if they had
been as skilful mountaineers as Sir Martin Conway, have attempted
to occupy it. It would have required twice as much labour to dig
a trench from the camp at Condole to the hill of Bonneval as from
Napoleon's camp to the Roche Blanche and the Roche Blanche is
;

'
M. Jullian {Vercingetorix, pp. 365-6) has justly reminded us that we must
not let the results of modern excavation obliterate the record of good work
done in the remote past.
2 J/em. de VAcad. des sciences de Clermont-Ferrand, nouv. scr.. xvii,
. , .

1875, pp. 45-9. Cf. P. P. Mathieu, Nouvclles observations sur les camps rom.
dc Gergovia, 1863. In the first edition (pp. 738-9) I gave a detailed refutatiou,
which was perhaps even then a work of supererogation, of Bouillet's theory.
* Hist, de Jides Cesar, ii, 270, n. 1.
 CAESAR'S OPERATIONS AT GERGOVTA 757

the only hill which is e regione oppidi, —

opposite the town 'of '

—
Gergovia that is to say, on a line perpendicular to the plateau on
which Gergovia stood.^
II. Colonel T. A. Dodge ^ says, '
The north
wont to be slope is

described as impossible to capture. It is not so. The slope is not

steep, though it is long. ... It is probable that in Caesar's day the
slope was concealed by woods and that he did not reconnoitre it
thoroughly. But the position could have been surprised on the
north far more easily than assaulted on the south.' These remarks
are altogether misleading. Steep is a relative term
'
but local
'
:

authorities'^ agree in describing the northern slope of Gergovia as

too steep to have been attacked with any chance of success and ;

the Carte de VEtat-Major, Sheet 166, confirms their statements.

I have myself ascended Gergovia from the north and from the south-
west and I can say positively that the ascent is much more difficult
;

on the north. Caesar tells us that he reconnoitred the whole position;*

and it is absurd to suppose that he did his work carelessly or that he
threw away his best chance of capturing Gergovia. The truth is that,
as he was not prepared to blockade the stronghold, there was only
one way of attacking it which offered the least chance of success ;

and that was to seize the Col des Goules, which linked the south-
western part to the outlying heights of RisoUes. ^
III. Caesar says that Vercingetorix had encamped near the town '

and grouped the contingents of the several tribes at moderate dis-

tances from one another and round his own quarters (castris prope '

oppidum in monte positis mediocrihus circum se intervallis separa-

\im singularum civitatum copias conlocaverat).^ From the words
"circum se von Goler inferred that the Gallic camps stood in a circle
on the slopes of the mountain all round the town ' but, as Napoleon ^ ;

^hows, this is a mistake. The words only imply that the camps of
the contingents surrounded the camp of Vercingetorix. Vercin-
getorix could have had no object in placing any of the camps on the
northern or the eastern face of the mountain and they doubtless ;

5tood on the southern slope and on the heights of Risolles.

IV. Long insists that when Caesar temporarily quitted Gergovia,
n order to intercept Litaviccus, he left his smaller camp undefended.
'
Caesar,' he says, took four legions out of the larger camp, and left
'

jWo legions to defend it. These two legions were therefore transferred
from the small camp to the large camp, and the small camp . . .

night have been seized by the Galli, but it probably contained

See pp. 779-80. Various other absurd guesses, of which A. OUeris {Examen
^

les diverses opinions emises sur le siege de Gergovia, 1861, pp. 14-5) gives a Hst,
lave been hazarded as to the site of the smaller camp.
2 Caesar,
pp. 254-5.
2 See A. Olleris, Examen, &c., p. 8, and A. d'Aigueperse, QEuvres arch, et Hit.,

[862, i, 3-4. * 5. 6^., vii, 36, § 1.

^ In the first edition
(pp. 824-5) I refuted the arguments by which Mr. Stock
Caesar de b. G., 1898, pp. 315-6) attempted to prove that Caesar attacked
Jergovia from the north. It does not seem necessary to reprint the refutation.
« B. G., vii, 36, § 2. ^ QaJl. Krieg, 1880, Taf. ix.
* Hist, de Jules Cesar, ii, 270, n. 1.
758 CAESAR'S OPERATIONS AT GERGOVIA
nothing that they would vahie, and they would not have been able
to defend it when Caesar returned.' ^ This is a sorry argument. It
was not what the small camp contained that the Gauls would have
valued, but the commanding position on which the camp stood, the
possession of which enabled Caesar to cut them off from the principal
source of their supplies of water and forage. He had only been able to
seize this position originally because it was inadequately garrisoned ;2
and if the Gauls had once recovered it, they would have taken care
not to lose it a second time. Caesar simply says that he left two
legions to defend his castra (C. Fahium cu7n legionihus duahus castris
. . .

'praesidio relinquit) and I believe that castris in this passage, as

;
^

in vii, 82, § 2, 83, § 1, means camps ', not camp '. Anyhow,
' '

Caesar would have been mad if he had left the smaller camp un-
defended.
—
V. The identity of the collis nudatus the hill which, on the day
before his final repulse, Caesar observed to be bared of defenders is —
still an open question. Caesar says that it was visible from the
smaller camp, that is to say from the Roche Blanche and that ;

there was a ridge (dorsum) belonging to eius iugi, which was nearly
level and, in a place where it gave access to the further side of the
'

town ', wooded and narrow. Finally, he says that the Gauls who had
previously occupied the hill had been withdrawn by Vercingetorix
to fortify 'this place ', that is to say, the place which gave access to
'
the further side of the town.* Ems iugi must mean either the
'

whole range formed by the mountain of Gergovia and the heights of

Risolles or, more probably, the latter only and nothing can be
;

more certain than that the dorsum, or at least that part of it which
was wooded and narrow, was the Col des Goules.^
1. Von Kampen identifies the collis nudatus with that part of the

mass of Risolles which is marked in the map as 723 metres high ;

but Mr. W. C. Compton ^ truly observes that the Puy de Jussat '

partially obscures the view of Risolles from the Roche Blanche '.
2. Napoleon, with whom Mr. Compton agrees, identifies the coUis
with a hill, marked A in his Plan (No. 21), 692 metres high, which
forms a part of the mass of Risolles, and is about 550 yards south-
west of the nearest part of the plateau of Gergovia."

^Decline of the Roman Republic, iv, 312, note.

2B. 0., vii, 36, §§ 5-6. 2 76., 40, § 3.

* Cum in minora castra operis perspiciendi causa venisset, animadvertit collem,

qui ab hostibus tenebatur, nudatum hominibus, qui superioribus diebus vix jjrae
mukitudine cerni potuerat. Admiratus quaerit ex perfugis causam. Con-
. . .

stabat inter omnes, quod iam ipse Caesar per exploratores cognoverat, dorsum
esse eius iugi prope aequum, sed silvestre et angustum qua esset aditus ad
alteram partem oppidi; vehementer huic illos loco timere, nee iam aliter sentire,
uno colle ab Roraanis occupato, si alterum amisissent, quin paene circumvallati
atque omni exitu et pabulatione interchisi viderentur ad hunc muniendum
:

locum omnes a Vercingetorige evocatos. Ih., 44.

This is now universally admitted.
'"

* Caesar's Sei^enth Campaign in Gaul, p. 93.

' So also A. Fischer in Annates scientifiques de V Auvergne, xxviii, 1855,

map facing p. 416, M. JuUian {Vercingetorix, p. 198) says that dense woods '

covered the Col des Goules and the heights of Risolles '.' The Col des C^oules
 CAESAR'S OPERATIONS AT GERGOVIA 759

3. Desjardins says tliat the collis was sans doute le puy de

'

Jussat '.1 Of the hill which Napoleon chooses he says, elle nous '

semble trop haute, trop pres de I'acces du plateau, c'est-a-dire du col

des Goules, pour avoir ete degarnie des troupes, puisque le systeme
du general gaulois aurait ete au contraire de mieux defendre cet
accfes.' Of course But in order to defend the acces ', Vercin-
!
'

getorix was obliged to remove troops temporarily from the collis to

fortify it. Whatever may be said of Napoleon's choice, Desjardins's
is certainly wrong. It is incredible that Vercingetorix would have
occupied a hill the possession of which would have been useless to
him and which was so far from Gergovia as the Puy de Jussat.
Moreover, any one who has seen the puy will admit that it would
' '

have been impossible for Vercingetorix to occupy any part of it

which was visible from the Roche Blanche, except the lower slopes,
with more than a handful of men.
4. According to von Goler,^ the collis nudatus was simply that
part of the southern slope of Gergovia which lies between the plateau
and the Gallic wall of loose stones and he identifies the place which
;

the Gauls attempted to fortify with Mont Rognon, the nearest point
of which is at least two kilometres, or about a mile and a quarter,
from the nearest point of the plateau It is difficult to understand
!

how a soldier like von Goler could have blundered in this fashion ;

and if he had ever seen the country, he would have realized that for
Vercingetorix to fortify Mont Rognon would have been about as
useful as to fortify the Puy de Dome. His son frankly admits that
he was wrong.^
5. M. Jullian,'* who of course agrees with Napoleon about the
place which the Gauls attempted to fortify, nevertheless follows
von Goler in identifying the collis with Gergovia. I am obliged to
differ from him, first, because Caesar would have seen that the
southern slope of Gergovia was abandoned before he ascended
the Roche Blanche, and, secondly, because he says that one of the
results of the stratagem which he devised after he saw that the collis
was abandoned was that all the [Gallic] troops were withdrawn [from
their former positions] by Vercingetorix to assist in the work of
fortifying ^ [the approach to Risolles and the Col des Goules], which

certainly we have Caesar's word for it but what is the evidence for Risolles ?
: ;

According toCassini'sOarfe de la France, t. i, f. 52, which M. Jullian (p. 370, n. 2)

cites to prove that the Puy de Jussat was wooded, there was not one tree on
Risolles ; and Caesar's words — 'the ridge to which the hill [the collis nudatus]
belonged was nearly level, but where it gave access to the further side of the
town wooded and narrow {dorsum esse eius iugi prope aequum, sed silvestre
'

et angustum qua esset aditus ad alteram partem oppidi) —

surely imply that only
the col was wooded. Perhaps M. Jullian has changed his mind for in his
;

Hist, de la Oaule, iii, 468, n. 3, he rightly includes among the high points of the
'

mountain mass which the Gallic troops occupied {omnibus eius iugi coUibus
'

occupatis \B. G., vii, 36, § 2]) the heights of Risolles.

^ Geogr. de la Gaule rom., ii, 682.
2 Gall. Krieg, 1880,
pp. 277-9, 281.
* lb., p. 277, nn. 3-4. See also Napoleon's Hist, de Jules Cesar, ii, 275, note,
for an elaborate and conclusive refutation of von Goler.
* Vercingetorix, p. 373. ^ B. G., vii,
45, § G.
760 CAESAR'S OPERATIONS AT GERGOVIA
seems to show that the southern slope of Gergovia had not been
abandoned before.
Myconchision is that Napoleon's choice was almost certainly
right. That he traced correctly the line which the Gauls fortified,
no one who has seen the ground will deny.
VI. Describing the stratagem by which he endeavoured to dis-
tract the attention of Vercingetorix before attempting to take
Gergoyia, Caesar says that he sent several squadrons of cavalry
'

thither [that is, towards the place which Vercingetorix was fortify-
ing] about midnight, ordering them to rove all over the country and
make a good deal of noise. At daybreak he ordered a large number
of pack-horses and mules to be taken out of camp, the pack-saddles
to be taken off, and the drivers to put on helmets, so as to look like
troopers, and ride round over the hills [collihus). He sent a few
regular cavalry with them, with orders to wander further afield, so
as to increase the effect. All were to make a wide circuit and head
towards the same goal.' ^ M. Jullian,^ if I do not misunderstand him,
traces the route of the horsemen along the north of the Auzon but ;

the words further afield (lafius) and wide circuit (longo circuitu)
' ' '
'

surely point to the high ground beyond, that is south of the stream.
The only conceivable objection to this view is that at such a distance
— —
nearly a mile and a half from Vercingetorix's camp the cavalry
would not have been recognizable as such.^ But men who have
lived their lives in the open country can see further than our town-
bred troops * and, moreover, Caesar says that although these
;
'

movements could be seen, far off, from the town ... it was impossible,
at such a distance, to make out exactly what they meant,' ^ a re- —
mark which would have been quite inapplicable if the cavalry had
kept on the northern bank of the Auzon, and if collihus had denoted
the lower slopes of the mountain-mass of Gergovia.
VII. After describing the movements of the cavalry Caesar says
that a single legion was sent along the same chain
'
and after it ;

had advanced a little way it was stationed on lower ground and con-
cealed in the woods {legionem unam eodem iugo mittit et faulum
'

frogressam inferiore constituit loco silvisque occultat ^.) It has been

conjectured that eodem iugo means the hill which Vercingetorix was
fortifying but it is needless to say that eodem iugo cannot mean
:

the same as ad idem ingum (' to the same hill ') and the context
;

shows that the legion was not sent to this place. If Caesar wrote
eodem iugo, the words must mean that the legion was sent along the
same line of high ground by which the horsemen had gone, that is
to say, the northern slopes of the Montague de la Serre.'^ Von Goler,
' B. G., vii, 45, §§ 1-3. Vercingetorix, pp. 210-1, 370.
-

^ '
Good eye-sight,' says Lord Wolseley {The
Soldier's Pocket-Book, 5th ed.,
1886, p. 491), ' can distinguish bodies of troops at 2,000 yards at 1,200 yards
. . .

eavahy is distinguished from infantry,' &c.

* See my
Anc. Britain, p. 608, n. 3. My
friend, Captain Motard, tells me that
he knows men who can
distinguish cavalry from infantry at over 2 kilometres
(2,186 yards). ' B. G., vii,
45, § 4. ^ lb.. 45, § 5.

' So Heller {Philohqus, xxvi, 1867, p. 686). Dittenberger quotes from Li%\y
(ii, 50, § 10) -id jugo circummissus Veiens in verticem collis evasissel ; and a
 — —

CAESAR'S OPERATIONS AT GERGOVIA 761

however, believes that Caesar wrote eodem, illo, just towards that '

side.' 1 There is no necessity for this emendation and in fact, as ;

Steinberg remarks,^ von Goler only made it because he had fallen

into the mistake of supposing that the place which Vercingetorix
fortified was Mont Eognon.
It is to trace the exact route by which
impossible and unnecessary
this legion marched.
According to Napoleon, after leaving the large
camp, it crossed the Auzon about a mile east of the foot of the Roche
Blanche and recrossed it near the village of Chanonat. Mr. Compton ^^

thinks that it moved along the northern bank of the Auzon through-
out but he fails to see that his theory is stultified by the words
:

eodem iugo * and, according to his Plan, the legion, instead of

;

descending to lower ground before it halted, as Caesar says that it

did, ascended to higher ground. That part, at all events, of the
route which Napoleon traced along the slope of the Montague de la
Serre is indicated correctly, at least in its general direction.
Paul ^ believes that Napoleon has made the legion advance further
westward than Caesar's words, faulum progressam ^, entitle him to do.
He considers that it halted on the southern or south-eastern edge of
the Puy de Jussat. The point is of infinitesimal importance ])ut ;

I believe that Paul is mistaken. Caesar's words are vague and on ;

a question like this the opinion of a military expert like Stoffel is

more likely to approximate to the truth than that of the most
learned of scholars. Napoleon, following the colonel, made the legion
halt just north of Chanonat.
VIII. The various positions occupied by the 10th legion and by
Sextius during the attack on Gergovia cannot be exactly fixed. It
'^

passage in B. C, i, 70, §4, eo consilio, uti ipse iugis Octogesam perveniret,

. . .

is decisive. Paul argues {Berl. phil. Woch., iii, 1883, col. 563) that iugo must
mean along the ridge, not along the slope of the hill. This is an arbitrary
'

assertion and it is certain that the legion did not go along the ridge
;
' '
;

for no general who was not a lunatic would have sent them on such an errand.
Paul also contends that Caesar would have mentioned the time at whicli
he dispatched the legion and accordingly he conjectures that he wrote
;

(legionem unam eodem) luce (mittit), &c.,

—
Caesar sent a single legion in the
'

same direction in broad daylight.' This is one of the many instances in which
this scholar has wasted his time and his ingenuity in correcting what requires
no correction. There was no reason why Caesar should mention any time.
It was evidently daylight when he sent the legion on its journey for in an earlier;

section of the same chapter he mentioned that he had sent a body of horsemen
on a similar errand at dawn {prima luce). Therefore he would have conveyed
little new information by writing luce afterwards, [I am glad to find that
Mommsen {Jahresb. d. philol. Vereins zu Berlin, xx, 1894, p. 209) rejects both
luce and another conjecture of Paul's, decimam (legionem) for unam [B. G.,
vii, 45, § 5).]
1 Gall. Krieg, 1880, p. 281, note. ^ puiologus, xxxiii,
1874, p. 452.
Caesar's Seventh Campaign in Gaul, Plan facing p. 30.
^

* Are not these words also fatal to the theory of M. JuUian, according to

whom {Vercingetorix, p. 370), the legion, 'apres avoir remonte la vallee entre
Gergovie et la Roche-Blanche,' turned to the left and disappeared in the woods
behind the Puy de Jussat ? In his Hist, de la Gaule, iii, 474, he sends the legion
'
le long de 1' Auzon ', but does not say on which bank.
^ Berl. phil. Woch.,
iii, 1883, col. 5G3-4.
e
B. G., vii, 45, § 5. ' lb., 47, 49, 51.
762 CAESAR'S OPERATIONS AT GERGOVIA
is certain, however, that the 10th was posted on the left and Sextius's
division on the right of the Roman cheek the
line of retreat, so as to
pursuing Gauls.
Caesar says in chapter 46 that about half-way up the mountain
of Gergovia the Gauls had made a wall of loose stones and the line ;

of this wall is drawn in all maps about 300 yards north of the village
of Gergovie, which was formerly called Merdogne. In chapter 47
Caesar says that after the other legions had passed the wall and
taken three camps, he sounded the trumpet for them to retreat, or
fall back {receptui cani iussit), and made the 10th legion, which he
commanded in person, halt but he adds that the other legions did
;

not hear the signal because a wide valley intervened between him
and them. This valley is generally identified with the hollow or
depression, which has been largely filled up, on the west of Gergovie.^
M. Jullian,2 however, holds that it v/as the one ou il se trouvait '

—
lui-meme ', between the Roche Blanche and the hill of Gergovia.
He adds that presque tons les ecrivains placent a ce moment la
'

X© legion sur le flanc de la montagne gergovienne, et pas loin du

. . .

village and he refers to page 744 of my first edition. I adhere

'
;

provisionally to my opinion, for the following reason. Caesar says^

that when he saw that the assaulting column was in difficulties, he
ordered Sextius, whom he had left in charge of the camp on the
Roche Blanche, to take up a position with some cohorts of the 13th
legion at the foot of Gergovia, and that he himself advanced a little
with his own legion from the position which he had taken up (ad
T. Sextium legatum, queni minorihus castris praesidio reliquerat, misit,
ut cohortes ex castris celeriter educeret et sub infimo colle ah dextro
latere hostium constitueret. Ipse paulum ex eo loco cum legione pro-
. . .

f/ressus uhi const it erat, eventum pugnae expectahat). According to

M. Jullian, Sextius was to take the place which Caesar had vacated.
It appears to me, on the other hand, that the words suh infimo colle
are contrasted with eo loco uhi constiterat Moreover, when Caesar .

says, just before mentioning the considerable valley ', that after
'

the capture of the three Gallic camps he immediately halted the 10th
'

legion, which he commanded in person' (legionisque X., quacum

erat, continuo signa constiterunt [v.l. constituit]),'^ he evidently im-
plies that the legion had up to that moment been in motion whereas, ;

if the point of departure of the column of assault had been, as

M. Jullian says,^ the smaller camp, and if, as he also maintains,^ the
legion, at the moment when Caesar sounded the recall, was immedi-
ately north of the smaller camp, it would surely have been at rest.
Heller maintains that when Caesar gave the order for the legions

^ If this depression was not the satis magna vallis, Caesar (unless M. Jullian's
identification, noticed in the text, is right) must have referred to the very
broken valley or gorge on the east of Gergovie ; and although von Goler
{Gall. Krieg, 1880, Taf. ix) takes this view, nobody who has seen the ground
will admit that Caesar would have taken up his position on the east of this gorge.
2 Vercinqetorix, pp. 213, 373-4, 374, n. 3.
^ B. G., vii, 49. * See
p. 763, n. 1.
^ Verciiigetorix, p. 373. * lb., pp. 213, 373-4, 374, n. 1.
 •

CAESAR'S OPEUATTONS AT GERGOVIA 763

to retreat, lie set the example by retreating himself.^

of obedience
Steinberg, however, justly observes that it was not necessary for
him to do this. As he was himself outside and south of the Gallic
stone wall, he could conveniently remain where he was until the
other legions should have fallen back.^ If, as Long believes,^ he only
ordered the legions to halt, or fall back and re-form, it is of course
plain that he did not retreat himself.
Napoleon* fixes the first halt of the 10th legion on a knoll on
which stands the south-western part of the village of Gergovie :

von Kampen ^ and Mr. Compton,^ who has studied the ground, agree
with him and the position is in my opinion right or very nearly so.
;

Von Goler' selects a spot about 400 yards north-east of Gergovie,

which istoo far from the valley, and where, moreover, as any one
who has seen the ground will admit, it would have been very difficult
for the 10th legion to assist the others.
The second position of the 10th legion is fixed by Napoleon about
a quarter of a mile east-south-east of the first. He bases his choice
upon the reading regressus in chapter 49, which von Goler and he
substitute, without any authority, for progressus. The 10th legion,'
'

says Napoleon, must, in the presence of a combat the issue of

'

which was uncertain, have taken up a position behind rather than

towards the front.' ^ In other words, the legion which acted as
a reserve must have retreated just when its services seemed likely to
be required Von Goler defends regressus on the ground that as
!

Gaesar had caused the 10th legion to halt immediately after the

^ Philologus, xxvi,
1867, p. 686. Caesar's text here presents a difficulty ; and
Heller selects the spot where he believes the 10th legion to have stood, in accor-
dance with an emendation of his own. According to the MSS., Caesar wrote : —
Caesar receptui cani iussit legionisque {v. I. legionique) X., quacum erat contionatu/^,
signa constiterunt [v. I. constituit).' Schneider {Caesar, ii, 477-9) defends con-
tionatus, remarking that Caesar was referring to the speech which, as he tells
us in chapter 45, he made to his lieutenants before sending them to attempt
the capture of Gergovia, and suggesting that he addressed the assembled
legions — —
and particularly the 10th to the same effect. Nipperdey {Caesar,
p. 95), who puts a comma after erat, takes contionatus absolutely, and says,
'
there is no obscurity as to the subject of Caesar's harangue :he announced
that, as he had achieved his purpose, he intended to retreat '. Other scholars,
despairing of extracting any satisfactory meaning from contionatus, have made
attempts, more or less futile, to amend the text. Paul {Berl. phil. Woch., iii,
1883, col. 601) argues that Caesar would have been too busy to harangue the
10th legion at such a crisis, and proposes to read (quacum) ierat G. Trehonius
legatus{\) Whitte simply deleted co7?i?'or?a^ws ; Heller {P7^^7oZo<7'^^?, xix, 1863,
p. 540), who is followed by Holder, proposed clivom nactus : and von Goler,
who is followed by Dittenberger, Meusel, and others, contmuo. This seems
to me the most probable emendation. See Meusel's Tabula Coniecturanim,
p. 31 {Lex. Caes., vol. ii, pars ii).
2 Philologus, xxxiii, 1874, p. 455.

; Decline of the Roman Republic, iv, 319-21. See pp. 246-8.

^ Caesar, p. 367

supra.
* Hist, de Jules Cesar, ii, 278.

^ Quindecim ad Caesaris de b. Q. comm. tahiilae, x.

^ Caesar's Seventh Campaign in Gaul, Plan facing

p. 30.
'
Gall. Krieg, 1880, Taf. ix.
" Hist, de Jules Cesar, ii, 279, note.
 764 CAESAR'S OPERATIONS AT GERGOVTA
capture of the three camps, he would not liave advanced nearer to
the town, when his object was to be ready to support the legions,
which were hard pressed by the Gauls.^ But the circumstances were
now changed, for the legions, in defiance of orders, had not halted,
but had pushed on to the wall of Gergovia 'progressus would not
:

necessarily imply that Caesar advanced nearer to the town and he ;

certainly would have advanced nearer to the legions. It is perhaps

doubtful whether progressus means advancing nearer
'
(to the '

town) or advancing nearer (to the hollow by which the legions

' '

had climbed the hill). Von Kampen and Compton adopt the latter
meaning. They find the second position about 150 yards south-west
of Gergovie, which seems to me unobjectionable.
As to the third position of the 10th legion, Caesar simply says,
'
Overborne at every point, the Romans were driven from their
position with the loss of 46 centurions. The Gauls were relentlessly
pursuing when the 10th legion, which had taken post in reserve on
comparatively favourable ground, checked them' (nostri, cumundique
premerentur, XLVI centurionibus amissis deiecti sunt loco. Sed in-
toJerantius Gallos insequentes legioX. tardavit, quae pro suhsidio paullo
aequiore loco constiterat). Where the comparatively favourable'

ground was, depends upon the line of retreat taken by the defeated
'

legions. Napoleon fixes upon a position about a quarter of a mile

east-south-east of the second position, as indicated in his Plan ;

and about 600 yards east of this third position, on the Puy de Mar-
mant, he and von Goler find the second position of Sextius, of whom
(aesar says, The 10th was in its turn supported by the cohorts of
'

the 13th, which, under T. Sextius, had quitted the smaller camp and
occupied a commanding position (hanc [legionem decimam] rursus
'

XIII. legionis cohortes exceperunt, quae ex castris minorihus eductae

cum T . Sextio legato ceperant locum superiorem). Napoleon mis-
understands the words locum superiorem, which do not mean a high '

hill '. In B. G., iii, 4, § 2, Caesar describes the rampart of a camp

as a locus superior and, as A. Eberz points out, the locus superior
;

which Sextus occupied was only superior in relation to the position

at the foot of the mountain of Gergovia, which he had occupied
])efore.2 Von Kampen ^ holds that the flight of the Gauls assuredly
'

took place along the valley which runs from Merdogne [Gergovie] to
Donnezat, and not across the hills, the less so because the Romans
had regarded the Aedui, who appeared on their right flank, as
enemies, whom therefore they would certainly have not hastened to
meet '. Heller, objecting to the third position which Napoleon assigns
to the 10th legion, says that Caesar would there have been separated
from the other legions by the valley of Gergovie, and therefore could
not have supported them. Accordingly he puts the 10th legion on
the west of the valley.* I do not agree with this view. If the line

1 Gall. Krieg, 1880, p. 286, n. 1.

- Neue. Jahrb. f. Philologie, &c., Ixxv., 18")7, p. 854.
^ Quiiidecim ad Caesaris de h. G. comm. iahuJae, x.
^ Philologus, xxvi, 1807, pp. G87-8.
 ;

CAESAR'S OPERATIONS AT GERGOVIA 765

of retreat of the other legions lay along the valley, Caesar could have
supported them from the east just as well as from the west. More-
over, Sextius was unquestionably on the right of the line of retreat
and it is clear that Caesar remained on the side opposite to that of
Sextius. Von Kampen finds the third position of the 10th legion
immediately north-east of Donnezat, and the second position of
Sextius immediately north-east of the village of La Roche Blanche.
The latter of these positions is perhaps too far south for von ;

Kampen makes the retreating Romans cross the parallel trenches

which connected the larger with the smaller camp. If he is right in
tracing the line of retreat, I believe that the second position of Sextius
was on the eastern slope of the Roche Blanche, and the third position
of the 10th legion just opposite, on the north-east of Donnezat. I do
not feel certain about the line of retreat but Caesar's narrative
;

seems to me to suggest that the Romans retreated, as they had

advanced, along the valley of Gergovie. It might be argued in
favour of Napoleon's view, that they would have retreated towards
their large camp for fear of exposing it to attack but Caesar had
:

doubtless left a force to guard it and if the Gauls had ventured to

;

attack it, they would have been compelled to abandon their strong
position on the high ground, and to fight a battle with the Romans
in the plain .^
Regarding the first position of Sextius, there is no room for doubt.
Caesar says that he sent an order to T. Sextius, the general whom
'

he had left in charge of the smaller camp, to take his cohorts out
quickly and form them up at the foot of the hill, on the enemy's
right flank, so as to check their pursuit in case he saw our men
driven from their position {ad T. Sextium legatum, quefn minorihiis
'

castris praesidio reliquerat, misit ut cohortes ex castris celeriter educeret

et suh injimo colle ah dextro latere hostium constitueret, ut, si nostras

loco depulsos vidisset, quo minus libere hostes insequerentur, terreret).

Sextius was certainly in the hollow between the Roche Blanche and the
hill of Gergovia and Napoleon, von Kampen, and Steinberg ^ agree
;

in placing him there, just on the right of the Roman line of retreat.
M. Jullian, who, as we have seen,^ differs from most commentators
regarding the identity of the wide valley ', naturally differs from
'

them also regarding the successive positions of the legions.'* Tons '

^ Caesar says {B. G., vii, 51, § 3) that the legions, as soon as they reached the
plain, halted and stood fronting and threatening their pursuers. Heller
{Philologus, xix, 1863, p. 539) actually holds that the plain was the narrow
depression between the mountain of Gergovia and the Roche Blanche ; and in
support of this view, which he never could have adopted if he had seen the
ground, he refers to the passage in which Caesar says that the distance, in
a straight line, from the plain and the starting-point of the ascent to the wall
of Gergovia was 1,200 paces (about 1,940 yards). But the distance from rbe
depression between the Roche Blanche and the mountain of Gergovia to the
wall of the town, is considerably less than 1,200 paces. Probably, as Napoleon
holds, the legions started on their ascent from the low ground between the
Roche Blanche and the Puy de Marmant. The ground on the east of the
Roche Blanche and just north of Donnezat might fairly be called a planilus,
^ Philoloijus, xxxiii, 1874,
p. 457. * p. 702.

* Hist, de la Gaiile, iii, 477, n. 5.

766 CAESAR'S OPERATIONS AT GERGOVIA
ces mouvements des legions,' he says, '
me paraissent succeder sur
line meme ligne, le chemin actuel de la Roche Blanche
et naturel
au village.' Accordingly he fixes the second position of the 10th
legion east of the village of Gergovie at the point the elevation of
which is marked on the map as 536 metres, and the third position
on the site of the village itself while he supposes that Sextius took
;

np his second position on what had been the second position of the
10th. If this conjecture is right, Caesar, who h^d before stationed
Sextius on the enemy's right flank, so as to check their pursuit in
'

case he saw the Romans driven from their position ',i must have
deliberately transferred him for the same purpose to their left.
IX. The path followed by the Aeduan contingent, w^hom the
legionaries, at the time when the fighting under the wall of Gergovia
was going on, mistook for enemies, cannot be fixed with absolute
precision. Caesar says that they appeared on the exposed (that is
the right) flank of the legions, and that he had sent them by a path
on the right, to distract the attention of the garrison.^ The legions
had climbed the southern slope of Gergovia. It is clear, then, that
the Aedui ascended the mountain by some path on its eastern or
south-eastern flank. According to Napoleon and von Kampen, they
went up the south-eastern slope, taking the shortest road from the
larger Roman camp, and passing about 500 yards north of the Puy
de Marmant but while von Kampen brings them to a point on the
;

north-west of Gergovie and about 160 yards south of the outer Gallic
wall. Napoleon makes them turn off to the left from a point about
550 yards due east of Gergovie, and strike off in a south-westerly
direction. I am sure that Napoleon is wrong, because, on his theory,
the Aedui would never have created a diversion at all and it was ;

for this reason that Caesar had sent them up the hill. Mr. Compton
takes them to the same goal as von Kampen by a long detour, passing
northward of the long spur which Gergovia throws out on the east,
the height of which is marked as 445 metres and this is in part the
;

route indicated by von Goler but von Goler makes the Aedui stop
:

short at a point about 900 yards east of the northern extremity of

Gergovie, in which case the Romans would not have noticed them
until they were themselves in full retreat. Mr. Compton ^ says, A '

second survey of the ground, after some years' interval, has only
sufficed to confirm the impression that the point at which the Aedui
are seen is the ufper shoulder, the lower position being out of sight
from the ravine under the south gate, where the fight took place
(c\ 48). But in order to reach the upper shoulder, a troop of armed
men w^ould be obliged to make a longer circuit to the north, the
south-east corner of the hill being practically inaccessible.'
I believe that Napoleon, von Goler, von Kampen, and Mr. Compton
are all mistaken, but that Mr. Compton goes nearest to the truth.
It is obvious that the Aedui Avould not have fulfilled their mission
of creating a diversion unless they had penetrated the space between

'
B. G., vii, 49, § 1- . .
* lb., 50, § 1.
^ Caesuras /Seventh Campaign in Gaul, p. 95.
 CAESAR'S OPERATIONS AT GERGOVIA 767

the town and the outer wall and it is clear from Caesar's narrative
;

that the legions, when they caught sight of them, were still striving
to hold their ground under the wall of the town. Besides, it is very
unlikely that if the Aedui had been below the outer wall when they
were descried, the Romans would have taken them for Vercin-
getorix's troops. What conceivable object could the latter have had in
voluntarily descending from their strong position ? ^ I believe, there-
fore, that the Aedui, having climbed the hill on the east by the route
which Mr. Compton indicates, and encountering no opposition,
marched westward along the south side of the hill and on the north
of the outer wall.

spent nine afternoons in examining Gergovia

In 1896 and 1898 I
iind its environs. I was already familiar
Avith the whole literature of
the subject and had studied the best maps but I found that I had
:

a great deal to learn and I must admit that a great deal remains
;

uncertain.
X. According Compton,^ a story is told that in the fight
to Mr. '

Caesar himself lost his sword, which he afterwards was shown in

a temple of the Arverni.' No such story is told, except by Mr.
Compton, Napoleon, and other inaccurate modern Avriters. Plutarch,'^
indeed, says that Caesar lost a sword, but in the cavalry combat
that preceded the blockade of Alesia. Another apocryphal story is
told by Napoleon in such a way as to suggest that the incident which
it describes took place at Gergovia. Servius,' he says,
'
relates
'

that .when Caesar was taken away prisoner by the Gauls, one of
. .

them began to cry out Caecos Caesar, which signifies in Gaulish let
him go, and thus he escaped.' * But Servius only says that this
happened in one of Caesar's battles in Gaul —
Caius lulius Caesar,
:
'

cum dimicaret in Gallia et ab hoste raptus equo eius portaretur

armatus, occurrit quidam ex hostibus qui eum nosset, et insultans
ait Caesar Caesar [v. 1. Cecos Caesar'], quod Gallorum lingua dimitte
significat et ita factum est ut dimitteretur. Hoc autem ipse Caesar
;

in Ephemeride sua dixit.' ^

LITAVICCUS'S MARCH TO GERGOVIA

Caesar says that when Litaviccus was marching from the country
of the Aedui to Gergovia, he halted about 30 Roman miles from that
place to harangue his troops, and told them that they must push on
and join Vercingetorix, and not Caesar. On the following night,
probably just before dawn, Caesar started to intercept Litaviccus,
^ These reasons tell equally against the view of M. Jullian {Hist, de la Gaulc,
iii,474, n. 6), who makes the Aedui ascend the path that leads from the large
camp to the farm of Gl ergo vie, and tlien move along tlie path that leads to
(Jergovie itself.
'^
Caesar^s Seventh Campaiyn in Gaul, p. 97.
^ Caesar, 29. * Hist, de Jules Cesar, ii, 282, n. 1.

' H. A. Lion's ed.

(1820), ii, 48. Of. Class. Philology, iii, 1908, p. 445.
768 LITAVICCUS'S MARCH
and encountered his force 25 Roman miles from Gergovia. There-
upon Litaviccus and his personal retainers made their escape to
Gergovia .1
According to Napoleon, ^ Litaviccus marched towards Gergovia
along the left or western bank of the Allier, halted to harangue his
troops near Serbannes, and encountered Caesar near Randan.
(J. Hartung^ objects that, on this theory, Litaviccus would only

have marched 5 miles, the distance between Serbannes and Randan,

while the news of his approach was travelling to Gergovia and Caesar
was marching 25 Roman miles from Gergovia to meet him. Hartung
argues further that Litaviccus would have marched by the right or
eastern bank of the Allier, in order to avoid falling into Caesar's
clutches. He supposes that Litaviccus harangued the troops near
Vichy, which is about 8 miles, as the crow flies, from Randan, and
then marched in an easterly direction, so as to avoid Caesar and reach
Gergovia by a detour. Caesar, however, so Hartung tells us, had
foreseen this move. Accordingly he crossed from the left to the
right bank of the Allier made his legions march on a widely ex-
;

tended line, so as to avoid all possibility of missing Litaviccus and ;

encountered him near Thiers, east-north-east of Gergovia.

A. Hug * admits that there is some force in Hartung's objection ;

but he holds that Hartung's theory presents greater difficulties than

Napoleon's. It is quite inconceivable, he argues, that at a moment
when everything depended upon speed Caesar should not have re-
garded the passage of the Allier as a hindrance and, he asks, if ;

Caesar commanded the passage of the river, why did he allow

Litaviccus to cross it ?
I agree with Hug's first argument, though Litaviccus might have
got across the Allier before Caesar could stop him. Anyhow Har-
tung's theory is simply a bad guess. Caesar does not say a word
about having marched on a widely-extended line and if he had ;

done so, he would have lost the all-important advantage of present-

ing himself before the Aedui in irresistible force. If Eporedorix had
not told him by what route Litaviccus was advancing, his own
cavalry could have procured the necessary information. Litaviccus
might have kept out of Caesar's way by moving away from the
Allier in a westerly direction and entering Gergovia by its north-
western gate. Caesar's narrative, interpreted in its natural sense,
inevitably leads to the conclusion that Litaviccus, '\fter leaving his
halting-place, marched straight on and the conclusion is perfectly
;

credible, even if he did not advance more than 5 miles while Caesar
marched 25. Litaviccus may have nearly finished his day's march
when he halted to harangue his troops Caesar merely says that he
:

halted about {circiter) 30 miles from Gergovia he himself was march- :

ing at the utmost speed of which his men were capable and he may ;

have been misinformed about the distance.

1 B. a., vii, 38-40. - ///sV. de Jules Ccmr, ii. "iTl-'J.

^ Philoloijus, xxxii, 1873, j^p. 309-71.
^ Bursian's Jahreshericht, i, 1873, pp. 1107-8.
 769

THE MEANING OF THE PASSAGE IPSI EX FINI-

TIMIS TRAN SIRE VIDERETUR (B. G., vii,
. . .

55, §§ 9-10)

Ipsi ex finiiimis regionibiis capias cogere, praesidia custodiasque ad

ripas Ligeris disponere, cquitatumque omnibus locis inicie^idi timoris
causa ostentare coeperunt, si ah re fnmientaria Romanos exchidere aut
adductos inopia ex provincia expeUere possent. Quam ad spern multimi
cos adiuvabat, quod Liger ex nivibus creveraf, ut omnino vado non
posse transiri videretur.^ 8o runs this famous passage in the a MSS. :

the reading of ^ is (inopia) provincia excludere. Nipperdey^ reads m

provinciam expeUere, an emendation suggested by Nicasius and if ;

Caesar wrote expeUere, the emendation is obviously required for the :

Aedui could not have expelled Caesar from the provincia, unless pro-
vincia meant, not the Province properly so called, but the rest of
Gaul and, as a matter of fact, he never uses the word in this sense.'^
;

Achaintre,* indeed, who reads expeUere, contends that provincia means

the country of the Aedui, and refers to a passage ^ in which Caesar
makes Ariovistus speak of the Gallic territory which he had annexed
as provincia sua. But in this passage the meaning of provincia is
unmistakable and the use of the word gives additional point to
;

Ariovistus's argument. In the passage which I am discussing every

reader of Caesar, except Achaintre, has always understood that by
provincia he meant the Roman Province and if he had meant the
;

country of the Aedui, he would have written not ex provincia but

exfinibus suis. Schneider, who regards the reading of a as obviously
absurd, follows Morus and deletes aut adductos inopia provincia
excludere.^ Eporedorix and Viridomarus, he argues, had no prospect
of being able to exclude Caesar from the province, and would have
been delighted to see him go thither. On the other hand, they could
not have been so foolish as to imagine that they could compel him
to do so therefore the emendation of Nicasius does not mend
:

matters. Men, he adds, who are said to have been influenced by

hunger (adductos inopia), would be said to have been influenced to do
something, not to suffer. It would be absurd to say that a man was
influenced to be excluded from a place. Besides, the repetition in the
same sentence, of the word excludere would be extremely harsh. We
may therefore conclude, says Schneider, that the words aut adductos
inopia provincia excludere (or ex provincia expeUere) are a gloss.

' '
They
[the Aedui] then proceeded to raise forces from the neighbouring
districts,estabhshing detachments and piquets along the banks of the Loire, and
throwing out cavahy in all directions, to terrorize the Romans, in the hope of
being able to prevent them from getting corn or to drive them, under stress
of destitution, from the province. It was a strong point in their favour that
the Loire was swollen from the melting of the snow, so that, to all appearance,
it was quite unfordablc.'
- Caesar, p. 9(3. •'
Sec H. Meuscl's Lex. Caes., ii, 1279-84.
* Caesar, i, 348. B.''
C, 44,
i, § 8.
" iSoc Oudendorp's Caesar, i, 401, and Schneider's Caesar, ii, 502.
1093 3 D
770 A DISPUTED PASSAGE IN B. G., VII, 55

Long 1 remarks that '

the sentence quam ad sfem ynultum eos
adiuvabat, quod Liger ex nivibus creverat, ut omnino vado non posse
transiri videretur shows that the direct object of the Aedui was to
prevent Caesar from crossing the river. If,' he continues, they '

kept him on the west side, he would have to levy contributions in

a country which he had already passed through, and which was
probably devastated. He had also Vercingetorix in his rear. There
was corn on the east side of the Loire, for Caesar got it as soon as he
crossed. " Quam ad spem " therefore would be quite intelligible if
the direct object of the Aedui was to prevent his getting supplies,
by preventing him from crossing the river. If he crossed the river,
however, he would readily get into the " provincia " by a much
easier and shorter road than by crossing the Cevennes. It seems
likely then that the Aedui wished to prevent his getting supplies
and also to prevent his getting into the " provincia " by crossing the
Loire, for Caesar would plunder all the country of the Aedui, if he
"
crossed the river. Davies's conjecture, " in provinciam repellere
is without any authority and it is absurd. If they simply drove
;

him into the " provincia ", Avhere he could get supplies, and from
Italy too, they might expect to see him among them again, and the
country of the Aedui would be the first to suffer. Their plan was to
starve him where he was, between the Allier and the Loire, or if he
retired into the " provincia ", to compel him to pass south through
the Cevennes. I conclude that the text is corrupt but there is
;

enough to show what was meant.'

I agree with Long that there is enough to show what was meant
and I believe that his explanation of the meaning is right neverthe- :

less, his condemnation of Davies's conjecture, which is virtually

identical with Nipperdey's, is hasty and unjust. For both Davies's
conjecture and Nipperdey's do really harmonize with Long's ex-
planation of Caesar's meaning and of the readings found in the
;

MSS., provincia exdudere, following adductos inopia, is, as Schneider

points out, ungrammatical Avhile ex provincia expellere is, as I have
;

shown, absurd. The MSS., therefore, being plainly corrupt, we are

driven to conjecture. The only question, for historical purposes, is
whether the Aedui desired to prevent Caesar from reaching the Pro-
vince, or whether they desired to drive him into it. The only satis-
factory answer is that they desired, if possible, to prevent him from
reaching the Province, by starving him out between the Loire and
the Allier ; and if this were impossible, to drive him into it by the
most difficult way. This is exactly what Caesar meant, according to
Long and virtually what he said, according to Davies and Nipper-
;

dey. There are two other passages which go to prove that the Aedui
wished to exclude the Eomans from the Province. In B. G., vii, 66,
§§ 3-4, we find Vercingetorix telling his officers that Caesar was trying
to make his way into the Province. If he does so,' Vercingetorix
'

goes on to say, '

we shall be free for the moment but we shall
;

have done little to secure lasting peace, for he is sure to return with
^ Caesar, p. 374.
 —

A DISPUTED PASSAGE IN B. G., VII, 55 771

reiiiforeeinents, and continue the war indefinitely.' And in cliapter 65,

Caesar says that he sent to Germany for reinforcements of cavalry,
qaod interdusis omnibus itinerihus, nulla re ex provincia atque
. . .

Italia suhlevari poterat, because, as all the roads were blocked, he

'

could not get any assistance from the Province and Italy.' I admit
that this passage does not prove my point, because it is not certain
that the roads were blocked at the time when Caesar was marching
to join Labienus but it certainly suggests that the object of the
;

Aedui, as of Vercingetorix, was to cut off Caesar from the Province.

On the other hand, there is another passage which suggests that the
Aedui expected that Caesar would find himself obliged to make for
the Province. Describing the rumour which reached Labienus after
he had arrived before Lutecia, he says, Galli in conloquiis inter-
clusum itinere et Ligeri Caesar em inopia frumenti coactum in pro-
vinciam contendisse confirmahant} the Gauls — affirmed that
' . . .

Caesar was prevented from pursuing his march and from crossing
the Loire, and that want of corn had forced him to make a dash for
the Province.'
To conclude. Either we must delete the doubtful words, or, read-
ing expellere, assume that provincia here means the whole of Gaul
north of the Koman Province, or accept the emendation in pro-
vinciam. The first course is to my mind hazardous,^ and the second
is unwarrantable but, for historical purposes, the second and the
;

third lead to the same result, and the first is not inconsistent with it.^

THE DISPUTED PASSAGE NAM UT GOMMUTATO

TIMEBAT {B. G., vii, 56, § 2)
. . .

I am
not here editing the Commentaries and I only discuss this
;

i^assage because I want to find out whether Caesar intended to convey

that some of his officers were in favour of retreating to the Province,
or whether he simply meant that, notwithstanding the difficulties of
his position, such a course was out of the question.
Schneider* reads Nam ut commutato consilio iter in provinciam con-
verteret, ut (non) nemo tunc quidem necessario faciundum exist i-
mabat, cum infamia atque indignitas rei et oppositus mons Cebenna via-
rumque difficultas impediebat, tum maxime quod abiuncto Labieno atque
iis legionibus quas una miserat vehementer tiynebat. The words in open
italics are those that are doubtful. The a MSS. have ut ne metu (qui-
dem), the p MSS. ut nemo tunc, and certain inferior MSS. id ne metu.^
1 B.G., vii, 59, 1.
* Variety of readings, however, commonly accompanies, and sometimes
points to, an interpolation.
^ A. Klotz {(Jaesarstvdien,
pp. 258-9) adopts the conjecture, in provinciam
expellere. < Caesar, ii, 504-0.
" Reading id
ne nietn quidem and inserting quod before infamia, the passage
moans To change his whole plan of campaign and march for the Province,
'

that, he deemed, was a course to which he ought not to allow even the pressure
of fear to force him for the disgrace and the humiliation of retreat, the barrier
;

3 D -2
 772 A DISPUTED PASSAGE IN B. G., VII, 56

Nipperdey ^ adoj)ts Ciacconius's conjecture, ut nemo non turn

quidem. He also follows Elberling in substituting ne (commutato)
for ut (commutato) — the all the MSS.
reading of
because, with —
his reading, ut, followed by
would be ungrammatical, and,
impediebat,
as he points out, ut and ne were often interchanged by copyists.
Ut ne metu quidem is certainly not what Caesar wrote. I defy any
one who adopts this reading to make sense out of the passage. Ut
nemo tunc quidem has little or no point and I agree with Schneider
;

that if we accept this reading, we must supply non. Nipperdey's

version makes perfectly good sense but that does not prove that
;

it is right. Schneider says in defence of his, which, but for the un-
grammatical ut, is nearly identical with Nipperdey's, I have inserted '

non BEFORE nemo, because it is more likely that the course which to
Caesar appeared to involve shame and humiliation as well as great
difficulty, should have appeared inevitable to some of those whom
he was accustomed to consult than to a?/.' He also makes a strenuous
effort to prove that, admitting Ciacconius's conjecture, ut (com-
mutato, &c.) is not ungrammatical. He explains ut converteret
—
. . .

by referring to two other passages in the Gallic War, damnatum

poenam sequi oportebat, ut igni concremaretur (i, 4, § 1), and cum id,
quod ipsi diebus XX
aegerrime confecerant, ut flumen transirent, ilium
uno die fecisse intellegerent (i, 13, § 2).2 I may be mistaken; but
I think it will presently appear that while these passages are strictly
analogous to the one which I am discussing if we accept the reading
id ne metu quidem, the analogy disappears if we accept Ciacconius's
conjecture. In other words, I believe that if we accept that con-
jecture, ut (commutato, &c.) must depend upon impediebat, and
therefore must be ungrammatical. Schneider indeed, while he is
careful to explain that he does not make ut (commutato, &c.) depend
upon impediebat, argues that a passage in Cicero dii prohibeant,
indices, ut . existimetur {Pro Roscio Amer., 52, § 151)
. . proves —
that it might do so. But this is the only passage in Cicero in which
prohibeo is followed by ut and there is not a single passage in
;

classical Latin in which ut follows impedio. I believe therefore that

if we accept Ciacconius's conjecture, w^e must follow Nipperdey, and

read ne for ut.

Desperate remedies occasionally succeed and if we are to make;

sense out of this passage we must have recourse to conjecture. Long

rejects Ciacconius'semendation on the ground that to tell us what
others thought in a crisis in which he had to decide is not what we
should expect from Caesar.^ But if he wrote ut nemo non ex- . . .

istimabat, he was probably alluding to the opinion not of his officers

interposed by the Cevennes, and the condition of the roads forbade liini to
attempt it ; and above all he was intensely anxious for Labicnus, who
was separated from liim, and for the legions which he had placed under his
command'. See J. N. Madvig's Adversaria crilica, 1873, ii, 258.
^ Caesar, pp. 9G-7.
- J. C. Held takes id converteret as concessive (' assuming that he were to
. . .

change his whole plan of campaign', &c.), a view which Schneider effectually,
but unnecessarily, demolishes. ^ !See Long's Caesar,
p. 375.
 A DISPUTED PASSAGE IN B. G., VII, 56 773

only, but of the Gauls as well. Long adopts the reading which,
before the appearance of Oudendorp's edition, was the accepted one,
— Nam ut commutato consilio iter in provinciam converteret, id
'

netum quidem necessario faciendum existimabat,' &c. for, although ;

it has no MS. authority, he holds that it is the only reading that '

fits the context'; and he argues that it gets rid of the difficulty of
'

the construction " ut .converteret

. . impediebat " '.^ But he
. . .

has overlooked the obvious fact that, while it gets rid of this difficulty,
it introduces another, which can only be removed by supplying quod
between cum and infamia? This correction was made by Stephanus,
and, although Schneider denies that any classical writer uses quod
after cu7yi, it is followed by Madvig, who, taking as his guide the
inferior MSS. to which I have alluded, reads id ne metu quidem. If
Caesar wrote this, he meant that he had not thought it right, even
under the pressure of fear, to retreat to the Province if Long's :

reading is right, he meant that he had not thought it right to do so

even in that crisis. A. Klotz^ objects to Ciacconius's conjecture,
because Caesar nowhere else uses the expression non nemo, though
he often uses non nulli. True but Caesar uses numerous words
;

once and once only, Klotz then proceeds to build up a conjecture

of his own. Assuming that Caesar wrote Nam ne commutato, he
remarks that ne might easily have been changed into the MS. ut by
a careless copyist who fancied that the sentence was final, but that
whoever read carefully would have seen that ut was wrong, and
made the correction, ne, in the margin. In the next copy this ne
would have been taken as referring to the ut which in a preceded
metu, and so have got inserted before metu. need only make We
two very trivial corrections, and we get ut [ne] metu quidam existima-
ha{nyt, 'as some under the influence of fear thought' (that it would
be necessary to do), &c. For historical purposes this is identical
with Ciacconius's emendation, and I am inclined to believe that one
or the other hits the mark it does not seem likely that Caesar
:

would have made the observation which Madvig attributes to him.'*

It is amusing to read the romance which Mommsen founds upon
a piece of doubtful Latin. It was,' he says,^ a grave and decisive
' '

moment, when after the retreat from Gergovia and the loss of
Noviodunum a council of war was held in Caesar^s head-quarters
regarding the measures now to be adopted. Various voices expressed
themselves in favour of a retreat over the Cevennes into the old Roman
province, which now lay open on all sides to the insurrection and
certainly was in urgent need of the legions that had been sent from

'
Heller {Philologus, xlix, 1890, p. 685) also reads id ne turn quidem,, but
inserts et before cu7n infamia,
2 Diibner indeed [Caesar, i, 253) reads id ne metu quidem without supplying

quod after cum, which he translates by dans un temps ou

'
but if any one will
'
;

translate the passage to himself, he will see at once that this explanation is
hopelessly fiat and unprofitable. ^ Caesar studien,
pp. 259-60.
* In my recent translation of the Commentaries I inadvertently followed

Madvig's reading.
'
Hist, of Rome, v, 1894, pp. 85-6 (/?ym. Gesck., iii, 1889, p. 288).
774 A DISPUTED PASSAGE IN B. G., VII, 56

Rome primarily for its protection. But Caesar rejected this timid
strategy suggested not by the position of affairs hut hy government in-
structions andfear of responsibility.^ The words which I have italicized
are based solely upon the emendation ut nemo non (or non nemo)
tmn, &c., interpreted by a powerful imagination. Not even Dion
Cassius, not even that worthless compiler, Florus, says a word to
support them. Mommsen has, of course, a perfect right to adopt
the emendation. But supposing that it is right, what is there to
justify the statement that a council of war was held in Caesar's
'

head- quarters' ? Or that the 'timid strategy' which Caesar is sup-

posed to have rejected was suggested by government instructions ?
' '

No wonder that frivolous persons talk of that Mississippi of false-

'

hood called history '. Historical imagination is a great quality but ;

it should not be allowed to run riot. If Mr. Froude had written such
a passage, what thunderbolts would have been launched at his head
by the re^eXT^ycpeVa Zei's of Oxford and Somerleaze !

WHERE DID CAESAR FORD THE LOIRE WHEN

HE WAS MARCHING FROM GERGOVIA TO
REJOIN LABIENUS ?

Napoleon says that there has always been a ford at Bourbon

'

Lancy ^ but Desjardins remarks that les gues se deplacent '.-

'
;
'

This may be a reason for not accepting Napoleon's conjecture but ;

there is also a reason for definitely rejecting it. Caesar crossed the
Allier three days after he left Gergovia.^ According to Napoleon,'*
he crossed at Vichy
it and it is hardly credible that he crossed at
;

any point nearer Gergovia. After he had crossed, Eporedorix and

Viridomarus left him, hurried off to Noviodunum (Nevers), and
burned it. Nevers is more than 70 miles, in a straight line, from
Vichy and it is not to be supposed that Caesar was standing still
;

while Eporedorix and Viridomarus were riding thither. Not until

he had heard that they had burned the town did Caesar determine
to hurry and he did not reach the Loire until he had made forced
;
'

marches by day and night {admodum magnis diurnis nocturnisque

'

itinerihus confectis).^ Now Bourbon-Lancy is barely 40 miles in

a straight line, say 48 miles by road, from Vichy. Surely this
distance is too short to correspond with Caesar's narrative. The
conclusion is that we cannot tell where he forded the Loire but ;

probably the place was between Decize and Nevers.^

^ Hist, de Jules Cesar, ii, 284, note.
2 Geogr. de la Gaule rom., ii, 686, n. 1. ^ B. G., vii, 53, § 4.

4 Hist, de Jules Cesar, ii, 283. ^ g^

j^ q^ ^^y^ 54.0,
fi
M. Jullian {Rev. des etudes anc, xi, 1909, pp. 347-50, and Hist, de la Gaule,
iii, 481, n. 6) argues that when Caesar retreated from Gergovia he probably

marched in the direction of Noviodunum ; for this was the shortest route to
Agedincum (Sens), where he doubtless expected to rejoin Labienus, and it was
essential to safeguard Noviodunum. Probably, then, the ford was near
Nevers.
 775

LABTENUS'S CAMPAIGN AGAINST THE SENONES

AND PARISH
Whoever undertakes to study the problems presented by Caesar's
narrative of Labienus's campaign must bear in mind that, after
leaving Agedincum (Sens), Labienus marched, in the first instance,
up the left bank of the Seine. Various writers have tried to prove
that he marched up the right bank and consequently that the battle
was fought on that side of the river. I have refuted this theory on
page 353.
I. The first question is, what was the marsh or marshy stream
which baffled Labienus on his march from Sens to Lutecia (Paris) ?
Labienus, writes Caesar,^ marched for Lutecia
'
leaving the . . .

draft which had recently arrived from Italy at Agedincum. . . .

When the enemy became aware of his approach large forces assem-
bled from the neighbouring tribes. The chief command was con-
ferred upon Camulogenus. . Observing that there was a continuous
. .

marsh, which drained into the Seine and rendered the whole country
in its neighbourhood impassable, he took post behind it, and pre-
pared to stop our men from crossing. Labienus at first formed a line
of sheds, and attempted to fill up the marsh with fascines and other
material and thus to make a causeway across. Finding this scarcely
practicable, he silently quitted his camp in the third watch, and
made his way, by the route by which he had advanced, to Metlosedum,
a town belonging to the Senones, situated, like Lutecia, of which we
have just spoken, on an island in the Seine. After repairing the
. . .

bridge, which the enemy had recently broken down, he made the
army cross over, and marched on, following the course of the stream,
in the direction of Lutecia„ The enemy, informed of what he had
done by fugitives from Metlosedum, gave orders that Lutecia should
be burned and its bridges broken down then, moving away from
:

the marsh, they encamped on the banks of the Seine, opposite

Lutecia and over against the camp of Labienus (eo supplemento '

quod nuper ex Italia verier at relicto Agedinci. Luteciam profici- . . .

scitur . Cuius adventu ah hostibus cognito, magnae ex finitimis

. .

civitatihus copiae convenerunt. Summa imperii traditur Camulogeno

. . Is cum animadvertisset perpetuam esse paludem, quae influeret in
.

Sequanam atque ilium omnem locum magnopere impediret, hie con-

sedit, nostrosque transitu prohibere instituit. Labienus prima vineas
agere, cratibus atque agger e paludern explere, atque iter munire cona-
batur. Postquam id difficilius fieri animadvertit, silentio e castris
tertia vigilia egressus, eodem quo venerat itinere Metlosedum pervenit.
Id est oppidum Senonum in insula Sequanae positum, ut paullo
ante de Lutecia diximus. . . Uefecto ponte, quem superioribus diebus
.

hostes resciderant, exercitum traducit et secundo flumine ad Luteciam

iter facer e coepit. Hostes, re cognita ab iis qui Metlosedo profugerant,
Luteciam incendi pontesque eius oppidi rescindi iubent ; ipsi pro-
' B. G., vii, 57-8.
776 LABIENUS'S CAMPAIGN
feeti a "palude, in rifis Sequanae, e regione Luteciae, contra Lahieni
castra considunt).
The marsh has been identified with three streams, the Bievre, —
which once entered the Seine just above the cathedral of Notre-
Dame the Orge, which enters the Seine 15 kilometres, or about
;

9 miles, in a direct line, above the confluence of the Seine and the
Marne and the Essonne, which enters the Seine about 10 kilometres
;

above the point where it receives the Orge.

1. Before Napoleon's work appeared, most commentators were in

favour of the Bievre. Napoleon, however, affirms that it flows '

through a calcareous soil and can at no epoch have formed a

'
'

marsh capable of arresting an army '.^ It has also been argued that,
if the marsh had been the Bievre, the Gauls would have been as

near Lutecia as they could well be, and that what Caesar says about
their movement from the marsh to Lutecia would be not only super-
fluous but false.2
On the other hand, M. H. Houssaye^ maintains that Napoleon
was mistaken. The bed of the Bievre, he says, is of clay (argileux) ;

and in some places the depth of the mud in it is 6 feet. Lebeuf ^

says that the stream has always been subject to inundations. The
sixth Plan, says M. Houssaye, in Lamare's Traitc de la police, which
represents Paris as it was in the reign of Henri III, shows a small
marsh formed by the Bievre ^ and in 1579 the river overflowed its
;

banks, and the inundation destroyed several houses in the Faubourg

St. Marceau. Again, Caignart de Saulcy says, to leave a place does
'

not necessarily imply the necessity of making a long march in order

to get to some other place and he concludes that Caesar might
'
;

well have mentioned that the Gauls departed from the Bievre
' '

{fwfecti a palude^) even if they were only going to march to

^ Hist, de Jules Cesar, ii, 286, note.

- J. J.Quicherat, Melanges d'arch. et dliist., 1885, i, 215, 238-9.
^ Le premier siege de Paris, 1876,
pp. 75-82.
*
Recueil de divers ecrits pour servir d' eclair cissem ens a Vhist. de France, ii,
1738, p. 154.
" I am unable to discover the alleged marsh in this Plan.

® It should be mentioned that the reading of the a MSS., except S, is

not profecti a palude, but prospecta palude. This reading, however, if it is not
exactly nonsense, is absolutely without point ; for there was no reason why the
Gauls should have taken the trouble to reconnoitre or survey the marsh just
before they finally quitted it, or why, if they had done anything so silly, Caesar
should have taken the trouble to record the fact. Profecti a (or ab) palude,
though only found in the jS MSS., is read by all the best modern editors, except
Nipperdey and Diibner, who follows Hoffmann's unnecessary and absurd
emendation praesepti. Unnecessary, because profecti a (palude) makes perfect
sense and absurd, because the Gauls could not have been said to be protected
;

by a marsh which their enemy had abandoned the attempt to force, and which
they themselves were about to quit. Nipperdey's emendation proiecta (palude)
is even more absurd ; and the arguments by which he defends it {Caesar, p. 100)
are perhaps the very worst that learned and ingenious editor ever devised.
It is clear, he says, that the marsh was either on both banks of the Seine, or only
on that bank which was opposite Lutecia. In either case, he continues, as the
Gauls crossed from Lutecia to the opposite bank, it is clear that they did not
abandon the marsh (profecti a palude) for, if the marsh was only on the bank
;

opposite Lutecia, they marched to it, not away from it if it was on both banks,
;
 — ;

AGAINST THE PARISH 777

Lutecia.i M. Houssaye also asks why Camulogeniis should have taken

the trouble to march all the way from Lutecia to the Essouue when,
only one kilometre from Lutecia, the Bievre offered as formidable an
obstacle to the advance of Labienus. Moreover, he continues,
Camulogenus could not tell whether Labienus would march along
the bank of the Seine or along the line of heights on the west of it.
Therefore it would have been madness for him to take up his position
behind either the Essonne or the Orge for if the Komans had
;

marched along the heights, he might have been attacked in flank

and in rear.
On an impartial review of these conflicting arguments, it appears
to me that if Napoleon has not proved his case, that case is neverthe-
less impregnable. It is not likely that Labienus could have marched
along the heights, for they were probably covered with woods and ;

Camulogenus had the best of reasons for marching as far as the

Essonne, for he must have wished to prevent Labienus from devas-
tating the country of the Parisii. But be this as it may, there is one
argument which is absolutely fatal to the claims of the Bievre.
Caesar says, in the passage which I have just quoted, that when the
Gauls heard that Labienus had crossed the Seine at Metlosedum, or
Melun, they sent a message to Lutecia, ordering that the town should
be burned, and then marched thither themselves. This statement
surely implies that they were a considerable distance from Lutecia
and expected to find it burned on their arrival. If they had been
within a quarter of an hour's march from the town, why should they
have sent any message ? They could have kindled the flames them-
selves. But, as a matter of fact, all the evidence goes to show that
the Bievre was not even a quarter of an hour's march from Lutecia.

they clearly did not march away from it. Therefore we are to read proiecta
(palude) Now, as Schneider remarks, it seems doubtful whether Nipperdey
!

is writing seriously or only making a bad joke (omnis eius disputatio talis est,
ut dubium sit, utrum serio, an aliter loquatur. Caesar, ii, 513). First of all,
Caesar's narrative makes it perfectly clear that the marsh was only on one bank
of the Seine, namely the left ; for that was the bank by which Labienus was
advancing when he came to the marsh and, as the marsh, or rather marshy
;

stream, emptied itself into the Seine, and as it was continuous {perpetua), it
could only have been on one bank of the Seine. Secondly, Nipperdey forgets
that both banks of the Seine were opposite Lutecia. Thirdly, Caesar does
not say a word to show that the Gauls had crossed from Lutecia to the opposite
'

bank and what he does say shows that they had not done so. For the marsh
'
;

was on the left bank of the Seine ; Labienus, finding himself unable to cross
the marsh, returned to Metlosedum (Melun), crossed the Seine there, advanced
up the right bank, and encamped opposite the island in the Seine on which
stood Lutecia and the enemy encamped on the bank of the Seine opposite
;

Labienus, that is, on the left bank. Therefore the Gauls remained on this
bank throughout the campaign. Finally, even if Nipperdey' s imaginary and
absurd premisses were correct, he would still have failed to prove his point
for he ignores the fact that the marsh probably was, and certainly may have
been separated by a considerable distance from Lutecia. It is clear, then, that
profecti a palude yields perfectly good sense whereas all the other readings and
;

the various emendations are absurd.

^Cf. B. G.,i, 51, ^ 3, eo muUeres imposuerunt, quae ad proeliu?n proficiscentes
.. implorahant, &c., where Caesar relates that the Germans under Ariovistus
.

moved out of their encampment to meet the Romans, who were close at hand.
778 LABTENUS'S CAMPAIGN
Plans 1, 2, 3, and 4same TraiU de la police (tome i) to which
of the
M. Houssaye refers show Paris at various times from that of Caesar
to that of Philip Augustus, and represent the Bievre as entering the
Seine exactly opposite the eastern extremity of the island on which
Lutecia stood Plans 2, 3, and 4 are founded upon information
!

furnished by Gregory of Tours, Aimoin, and various archives.

Besides, a glance at the map will convince any unbiased reader
that if Camulogenus had been encamped behind the Bievre, he would
have made an effort to prevent Labienus from crossing the Marne ;

whereas it is clearly implied in Caesar's narrative that Labienus,

after he had abandoned the attempt to force the passage of the
marsh and recrossed the Seine at Melun, crossed the Marne without
opposition.
2. Quicherat identifies the marsh with the Orge, which has two
branches. According to Quicherat, the Gauls occupied a position on
the heights of Juvisy, about 600 yards north of the northern branch,
and were separated by at least a mile and a quarter from the camp
of Labienus, which would have been south of the southern branch of
the stream.i De quoi done,' naturally asks de Saulcy,^ veut-il se
' '

garantir a une demi-lieue de I'ennemi, avec ses mantelets ? ^ Besides, '

a marsh formed by the two branches of the Orge would have been so
broad that it would have been hopeless for Labienus to attempt to
make a causeway across it.
I conclude then that the marsh was the Essonne. Napoleon
points out that the ground on the banks of this stream is still cut '

up by innumerable peat mosses and that it was behind this

'
;

line that his uncle encamped in 1814, while the enemy occupied
Paris.4
II. J. J. Quicherat has written an ingenious paper ^ with the
object of proving that the accepted explanation, which I have

^ See map facing p. 234 (t. i) of Quicherat's Melanges d'arch. et cChist.

* Les campacjnes de Jules Cesar dans les Gaules, p. 50.
=«
See B. G., vii, 57, § 4 58, §§ 1-2.
;

* One of Napoleon's arguments, however, is questionable. The text of the

'

Commentaries,^ he writes, says clearly that Labienus

'
. . surprised the passage
.

of the Seine at Melun, and marclied upon Lutecia, where he arrived before
Camulogenus. To allow of the success of this nianoeuvre, the marsh . . must .

necessarily not have been far from Melun. The Essonne alone fulfils this con-
dition {Hist, de Jules Cesar, ii, 286, note).
'
But the Commentaries do not say
that Labienus arrived at Lutecia before Camulogenus. What they do say is
that Labienus, after crossing the Seine at Melun, secundo fiumine ad Lniteciam
iter facere coepil. Hostes, re cognita ah iis qui a Metlosedo profugerant, Lideciam
incendi pontesque eiu-s oppidi rescindi iubent : ipsi profecti a palude in ripa
Sequanae e regione Luteciae contra Labieni castra considunt. This does not prove
that Labienus was the first to arrive. (See Long's Decline of the JRoman
Republic, iv, 327-8.) Maissiat {Jules Cesar en Gaule, ii, 228) adduces an argu-
ment which, I think, has weight. Assuming, he says, that Camulogenus had
been encamped above Lutecia until the moment when the flotilla of Labienus,
coming down stream from Melun, appeared in sight, it is clear that he would
have promptly marched to the neighbourhood of Lutecia. Therefore, whether
he was encamped on the Essonne or on the Bievre, he must have reached
Lutecia before Labienus.
'^
Melanges d'arch. et d'hisL, i, 217-8, 220-1.

I
 ——

AGAINST THE PARISH 779

followed,^ of the campaign His principal arguments are

is incorrect.
as follows :
—
First, if Labienus encamped opposite Lntecia, he must
have crossed the Marne at its confluence with the Seine. But to cross
the Marne at its confluence with the Seine is as difficult as to cross the
Seine itself. Why then does Caesar not mention the passage ?
Secondly, Caesar thus describes the relative positions of the Gauls
and Labienus if si profecti a palude ad ripas Sequanae e regione
:

Luteciae contra Lahieni castra considunt. Now, Caesar uses the words
e regione pour indiquer
' " au droit de " '. Thirdly, Labienus
. . .

could not have had any object in encamping opposite Lutecia, since
it had been burned and its bridges destroyed. Fourthly, if Labienus's
camp was opposite Lutecia, he was extraordinarily imprudent to
send his fleet down stream, under the eye of the enemy, and to
march down himself, when he might have been taken in flank from
the hills of Chaillot and Passy, which were at that time covered with
woods. '
Tant de temerite,' says Quicherat, ne serait egalee que '

par la negligence du general gaulois, a qui I'agitation de I'ennemi

aurait completement echappe malgre la proximite,' &c. Fifthly,
according to the received view, there was no reason why, in the
early morning, just before the battle, Camulogenus should have been
informed of the movements of the various Roman divisions for he ;

could have observed those movements himself.

Accordingly Quicherat makes Labienus encamp in the angle
between the confluence of the Marne and the Seine, just south of
Creteil, and cross the Seine immediately above the confluence while ;

he places the camp of the Gauls about half a mile south-east of

Lutecia, in a line with it and the Roman camp.
The answer to Quicherat's first argument is obvious. Labienus
could easily have crossed, and doubtless did cross the Marne with
the barges which he had brought from Melun. Caesar omitted to
mention the passage for the same reason that he omitted to mention
—
the passage of the Yonne, because to mention it was unessential.
Quicherat's second argument has virtually to sustain the whole
weight of his case. If he is wrong in his interpretation of e regione,
his whole theory collapses. And assuredly he is wrong. If e regione
means what Quicherat says, it comes to this, that Caesar is prac-
tically giving us no information whatever as to the actual or the
relative positions of the Gauls and of Labienus. Any reader is free
to take his map and place them exactly where he pleases, provided
he places them in a straight line with Lutecia and with each other !

But if the words mean what everybody, except Quicherat and Creuly
believes, the sense is clear, with or without a map. Labienus is on
the north of the Seine, and the enemy are on the south, opposite
Lutecia and over against his camp. The meaning of the words is
fixed, beyond cavil, by two passages in B. G., vii, 35, where Caesar
describes the stratagem by which he transported his army across
the Allier :
2 Cum uterque ntrique esset exercitui in conspectu fereque
e regione castris castra ponerct (' The two armies were in full view of
1 See pp. lGl-4. 2 See p. 149.
780 LABIENUS'S CAMPAIGN
one another, and each encamped ahnost opposite the camp of the
other'); and again castris fositis e regione unius eorum fontium
quos Vercingetorix rescindendos curaverat ('encamping opposite one
of the bridges which Vercingetorix had broken down '). If e regione,
in these passages, does not mean opposite ', it means nothing.
'

Quicherat's interpretation of the phrase is overthrown by ForceUini,

who says^ 'absolute, adverbiorum more, e regione est recta linea . . .

. .Saepius vero e regione significat ex adverso

. and he illustrates '
;

this meaning by referring to the passages which I have just quoted.

So much for e regione. There is another word in Caesar's narrative,
which, by itself, overthrows the whole of Quicherat's case. Describ-
ing the measures which the Gauls took when they heard that the
Romans were trying to cross the Seine in three places, he says,
'
Leaving a force opposite the [Roman] camp, and sending a small
body in the direction of Metlosedum they led the rest of their
. . .

troops against Labienus {fraesidio e regione castrorum relict o et

'

farva manu Metlosedum versus missa reliquas copias contra

. . .

Lahienum duxerunt.^ ^ Now, on Quicherat's theory, the word relicto

is simply unintelligible. He assumes that a Gallic outpost had been
stationed exactly opposite the place where the bridge of Choisy now
stands ^ and this outpost was, on his theory, the praesidium that
;

was left to watch the Roman camp. But Caesar's words obviously
do not mean that the praesidium had been established from the out-
set of the campaign on the spot where it was left. Besides, according
to Quicherat, the Roman camp was about 4 kilometres away from
the praesidium and therefore the praesidium would have been
;

useless. Lastly, according to Quicherat, the Roman flotilla neces-

sarily waited, before starting on its voyage down the stream, under
the very eyes of the Gauls who composed the praesidium. To
wriggle out of this absurdity, Quicherat is compelled to maintain
that the Gauls would not have been able to tell why the barges were
about to go down the Seine. La descente de la flottille au con-
'

fluent (of the Marne and the Seine), he says, est une manoeuvre
' '

sur laquelle il n'y a rien a decider. Est-elle vraie ou feinte ? Est-ce

la Marne ou la Seine qu'on se propose de passer ? * This is flagrant '

special pleading. The manoeuvre, whatever its object, would have

been promptly reported to head- quarters. Again, de Saulcy ^ per-
tinently asks what would have been the use of Labienus's ordering
a noise to be made in the Roman camp, when, according to Quicherat,
the Gauls were two miles and a half away, and could not possibly
have heard it. On the accepted theory, the word relicto is perfectly
clear. The Gallic camp on the left bank of the Seine was opposite
the Roman camp on the right bank. Camulogenus, hearing that the
Romans left in camp were preparing to abandon it and cross the
river, left a force to hinder them, sent a detachment up the left bank
to follow the Romans who were rowing in that direction, and marched

^ Tolius latinitatis Lexicon, v, 1871, p. 142. ^ B. G., vii, 61, § 5.

^ See map facing p. 234 (t. i) of his Melanges. *
76., p. 221.
^ Les campagnes de Jides Cesar dans les Gaules, p. 51).
 AGAINST THE PARISH 781

illperson with the rest of his army down the same bank to encounter
Labienns.
Finally, it may be assumed that Labienus knew, or had ascer-
tained by reconnaissance, that there was no chance of his being
attacked in flank from the hills of Chaillot and Passy. Why should
C^amulogenus, who had no inkling of the stratagem which he medi-
tated, have posted troops on those hills, of all places ? Nor can
I see that Labienus showed any imprudence in sending his barges
dowji the stream. Quicherat cannot tell how near the stream the
Gauls were and they would have been just as likely to post troops
;

along that part of the river down which Quicherat believes that
Labienus sent the barges, as along the part between Lutecia and
Point-du- Jour. Besides, as de Saulcy ^ points out, in the latter case,
the barges would have been screened from observation, during the first
part of their course, by the islands in the Seine, whereas, on the theory
of Quicherat, they would have had no cover at all. Puis voila,' he
'

says, alluding to Labienus's passage of the Seine as described by

Quicherat, Puis voila que ce passage, qui prend au moins deux
'

heures, se fait forcement au nez et a la barbe de Camulogene, qui

met ses deux mains dans ses poches et attend que tout le monde
ennemi soit a son rang de bataille pour se permettre de le gener dans
ses mouvements Labienus knew that he must risk something.
!
'

But he ordered the captains of the barges to move as quietly as

possible and if the movement had been prematurely detected, he
;

would not have been any worse off than he had been before.
It is worth mentioning that the great Napoleon took the orthodox
view regarding the geography of this campaign, and that he did not
detect the blunders which, according to Quicherat, Labienus must,
on that theory, have committed.^
III. The camp of Labienus, it is sufficient to know, was close to
the right bank of the Seine, opposite Lutecia and over against the
camp of Camulogenus.^ But it certainly was not on the heights of
Romainville, where Duruy * places it for Romainville is a good
;

5 miles away from the Seine it would have been quite unnecessary
:

for Camulogenus, when he was about to march against Labienus, to

leave a force on the left bank to watch a camp which was 5 miles
north of the right bank and the whole tenor of Caesar's narrative
;

shows that Labienus encamped close to the river.

IV. One word with regard to the point where Labienus crossed the
Seine. Von Kampen ^ finds it opposite Auteuil, which is hardly
4 Roman miles from the site which he selects for the Roman camp ;

Napoleon^ a little further down the bank, near Point-du- Jour.

' Les campagnes de Jules Cesar dans les Gaides, p. 31.

Precis des guerres de Cesar, pp. 102-3.
-

* Napoleon {Hist, de Jules Cesar, ii, 287) thiiikj* that it was '
near the place
whore fSt. Germain -rAuxerrois now stands'. Various other attempts, which
I need not Jiiontion, have been niadc to dchno the site exactly.
* iSoc. de Vhist. de Paris el de rile de France,
1881, pp. 102-3.
^ Quindecim ad Gaesaris de h, 0. comm. tabulae, xi.

Hist, de Jules Cesar, ii, 288.

•*
782 LABIENUS'S CAMPAIGN
De Saulcyi places the passage opposite the island of Billancourt,
which not 4 miles from the assumed point of embarkation
is 6, but :

he argues that the three legions, marching in column, must have

covered a space of at least 3 kilometres and that, if the whole ;

column had passed the point which was 4 miles from the place of
departure, the head of the column would have reached a point
opposite the middle of Billancourt. Bronze swords had been found,
some years before de Saulcy wrote, in the bed of the river between
Billancourt and the left bank and he conjectured that Labienus
;

probably used this island and the islands of St. Germain and vSequin
to assist his passage. De Saulcy's arguments appear to me more
ingenious than convincing.^ Caesar does not say that Labienus
ordered the three legions to march 4 miles down the bank he says :

that Labienus ordered the knights who commanded the barges in

which the legions were to cross the Seine to move 4 miles down
the river and wait for him.^ Of course the barges did not all cross the
river at the same place but it is natural to suppose that the 4 miles
:

were reckoned from the point of departure to the point where the
foremost barge of the flotilla was moored and the discovery of ;

the bronze swords proves nothing.* M. Houssaye ^ sensibly asks

whether it is not reasonable to suppose that, as each company
reached the place of embarkation, it simply wheeled to the right so
as to form a little column at right angles to the stream, ready to
embark. Thus the three legions would have been able to embark
within a space of one kilometre at the outside.
V. There is another passage in the Latin text which, even among
those who are virtually agreed as to the battle-field and the sites of
the two camps, has given rise to discussion. After describing how
the Gauls were informed of the operations of Labienus, Caesar
writes :
—
On hearing this, they imagined that the legions were
'

crossing at three places, and that the Romans, in alarm at the defec-
tion of the Aedui, were all preparing for flight. Accordingly they
made a corresponding distribution of their own troops. Leaving
a force opposite the Roman camp, and sending a small body in the
direction of Metlosedum, with orders to advance as far as the boats
had gone, they led the rest of their troops against Labienus {Quibus '

rebus auditis, quod existimabant tribus locis transire legiones, at que

nines ferturbatos defectione Haeduorum, fugam parare, suas quoque
copias in tres partes distribuerunt : nam et praesidio e regione castrorum

relicto, et parva manu Metlosedum versus missa, quae tantum pro-

grederetur quantum naves processissent, reliquas copias contra Labienum
duxerunt).^ Desjardins identifies Metlosedum (or, according to the
^ Les campagnes de J ides Cesar dans les GaiUes, p. 41.
2 B. 0., vii, 60, § 1.
^ Cf. Rev. arch., 4® ser., vii, 1900, pp. 174-5.
* Such weapons may perhaps have been still used here and there anioug
backward tribes but lor centuries before Caesar's time (Gallic swords had been
;

made of iron. ^ Le premier siige de Paris,

pp. 90-1.
•*
B. G., vii, 01, §§ 4-5. Meusel {Jahresh. d. philol. Vcreins zu Berlin, xxxvi,
1910, p. 53) follows von Coler {Gall. Krieg, 1880, p. 295, n. 2) in deleting the
words atque onmes fugam parare
. . . for, he argues, how could the Gauls
;
 AGAINST THE PARISH 783

reading which he prefers, Metiosedum) with Meudon, which is heloiv

Paris, in the direction opposite to Melun. He considers that naves
must mean, not the boats (Untres), which Labienus sent up stream,
but the large barges, which he sent down and he adds, to strengthen
;

his argument, that, whereas, Melodunum is the only reading which

occurs where Melun is incontestably alluded to, the readings in this
passage are various.^ How misleading this assertion is, may be seen
by any one who will consult the apparatus criticus of Meusel's original
Moreover, as Quicherat rightly remarks,^ the sense of
edition.2
naves is generic :
—
il veut dire aussi bien un petit bateau qu'un
'

grand bateau.' Forcellini* supports this view. Navis,' he says, '

'
dicitur fere de navigio maiori, interdum etiam de cymba, lintre,
scapha, et huiusmodi.' The usage of our own language is similar.
It is a common thing to speak of an Atlantic liner as a boat '
'
;

and a large yacht is often described as a good sea -boat. ^ If, in the
passage ^ which I am discussing, Metiosedum meant Meudon, it
' '

would follow from Caesar's narrative that the Gauls took no heed
of the boats and the cohorts which went up stream in the direction
of Melun. Why should Camulogenus have taken the trouble to
send a small force (parva manu) to the point which the barges of
Labienus had reached, when he was marching with the rest of his
'^
forces, in the same direction, against Labienus himself ? Desjardins
tries to get over this difficulty by suggesting that Camulogenus had
expected to find Labienus posted above Lutecia, but actually found
him at the foot of Meudon and Issy, and awaited his attack near
Percy. De Saulcy ^ and C. Lenormant ^ agree with Desjardins. The
former says that Camulogenus, when he divided his army into three
parts, must have marched with the principal force in the direction
of what he was informed was the Roman ?nagnum agmen, that is, up
the stream, towards Melun and Lenormant says that Camulogenus
;

would not have sent his smallest force (parva 7nanu) in the direction
whence the greatest noise was heard. But the principal force of ' '

Camulogenus was the reliquae copiae which he led against Labienus ;

have believed that the Romans were preparing for flight when they heard
that they were crossing the Seine in order to attack them ? But nothing in
Caesar's narrative shows that the Gauls heard this and even if they did,
;

they must have seen that Labienus was preparing for flight in tlie sense
'
'

that he was abandoning his offensive campaign (cf. B. G., vii, 59, §§ 3-6) and
retreating to rejoin Caesar.
^ Geogr. de la Gaule ro7n., ii, 689.
^ Cf. Rev. arch., nouv. ser., v,
1862, p. 2, and p. 845, injra.
^ Melanges, &c., i, 238.

* Totius latiniiatis Lexicon, iv, 1868,

p. 230. Cf. B. C., i, 54, § 1.
^ Heller {Philologus, xvii,
1861, p. 284) remarks that any difficulty which may
be caused by the use of the words Untres and naves disappears if one reflects that
the passage Quibus rebus duxerunt {B. G., vii, 61, § 5) expresses the idea
. . .

which the Gauls had formed on hearing of the movements of the Romans and ;

that the word naves in the passage in question is to be understood as if Caesar

had written Untres quos esse naves Galli ex 7nagno remoruni sonitu saspkabantur.
This criticism seems over-subtle.
« B. G., vii, 61, § 5. ^ Geogr. de la Gaule rom., ii, 689.
^ Les campagnes de Jules Cesar dans les Gaults, p. 45.
^ Rev. arch., nouv. ser., iv, 1861, p. 281.
784 LABIENUS'S CAMPAIGN
and, in order to reach Labienus, Camulogenus had to march down
the stream, not uj). De Saulcy, however, Hke Desjardins, says that
the parva manus, which, according to him, was sent in the direction
of Meudon, must have returned to the GalHc camp, in order to tell
Camulogenus where Labienus was crossing the river.^ There is not
a M^ord about this in Caesar and his narrative clearly implies that
;

Camulogenus led his reliquas copias against Labienus directly after

he had sent the parva manus in the direction of Metlosedum. Lenor-
mant's remark is childish. Camulogenus sent his smaller force in
the direction whence the greatest noise w'as heard, because he had
reason to believe that the greatest noise was not made by the greatest
number of men.^ The fact is that Camulogenus was only partially
deceived by Labienus's stratagem. Moreover, as Schneider^ observes,
the niac/mwt agmen was so called because the greater part of it un-
doubtedly consisted of the baggage-train and if Caesar had meant
;

to convey that Camulogenus sent his smaller force in the dirertion

of the barges which had conveyed the legions of Labienus across the
river, he would certainly have expressed himself differently. The
barges were stationary the small boats were moving up the river.
:

Caesar would not have said that the parva manus was ordered to
advance as far as the naves should advance {parva manu quae . . .

tantum progrediatur quantum naves processissent), but that it was

ordered to proceed to the point where the barges were moored. And,
asks de Presle,*how can we believe that Caesar would have mentioned
'
Metiosedum for the first and only time without saying a single
'

word to indicate its position ? Whatever reading we are to adopt,

'
Metiosedum must have been above Lutecia ; and, as a matter of
'

fact, the right reading in all the four passages in which the town in
question is mentioned, is either Meclosedum or Metiosedum.^
But since the first edition of this book appeared, and when I fondly
imagined that the debate was closed. Dr. H. Sieglerschmidt has
intervened with an entirely novel theory and it cannot be denied
;

that in conjuring up difficulties which nobody else had perceived he

has shown wonderful ingenuity.^ He maintains that Labienus's
camp was at Boulogne-sur-Seine, 8 kilometres west of Lutecia that ;

the Gallic camp was on the heights of St. Cloud and that Labienus ;

crossed the river at Neuilly. Now^ the doctor's case rests upon the
groundless assumption that Melodunum and Metiosedum were
' ' ' '

two different places and when he appeals to the MSS.'^ he is dealing

;

with a matter on which he is imperfectly informed. He urges that

e regione Luteciae cannot mean opposite Lutecia ', because Labienus
'

and C^amulogenus would not have encamped near a town which,

^ Les campaynes de Jules Cesar dans les Gaules, p. 31.

^ I am glad to find that my argument has been anticipated by M. Houssaye,
^ Caesar, ii, 521-2. * Rev. arch., nouv. ser., v, 1862,
p. 93. p. 4.
'^
The true form is Metlosedutii, but probably Caesar wrote Meclosedum.
Sec p. 840.
« Rev. arch., A^ ser., vi, 1905,
pp. 257-71. I am sorry to sec that Sieglcr-
schmidt's view has been adopted by R. Oehler [Bilder-Atlas zu Cdsars Biicheni
de h. G., 1907, xxxviii. ^ /?gy_ arch., vi, 1905, pp. 206-8.
 s

AGAINST THE PARISH 785

having been burnt, had lost all military importance.^ Then, assuming
that the 'Metiosedum of B. G., vii, 61 was Meudon, and observing,
'

truly enough, that the two camps were below Metiosedum ', he '

triumphantly concludes that Lutecia could not have been between

them, but that the Gallic camp was on the prolongation of a straight
line drawn from Lutecia to the Roman camp.^ But he does not
understand the meaning of e regione Luteciae, which I have already ex-
plained,^ and which does not depend upon that of contra Labieni castra.
Moreover, although Lutecia had been burnt, it was still a strategical
position, commanding a network of great roads and neither Labienus
;

nor Camulogenus had anything to gain by moving away to the west.*

Sieglerschmidt argues, further, that Labienus only had two legions
in the battle and on this hypothesis he constructs an elaborate theory,^
;

which I need not examine,because the hypothesis is stultified by Caesar'

express statement that Labienus, when he was going into action, had
three .^ Mais,' asks Sieglerschmidt, who sees the difficulty, but des-
'

perately assumes that some meddlesome copyist wrote tribus instead

of duabus, ou est ensuite cette troisieme legion?
'
In the centre. Dr.
' '^

Sieglerschmidt! Caesar, in his description of the battle, only mentioned

—
two legions, the 12th, which formed the right wing, and the 7th, which
formed the left but he expected that his readers would have sense
;

enough to see that the one which he did not mention was in the centre.
VL Concerning the exact position of the battle-field there is
also divergence of opinion. According to Napoleon, the hill which
the Cauls occupied was that of Vaugirard, a little west of Mont
Parnasse. Von Kampen places the two armies in line of battle on
the plain of Crenelle, the Romans with their backs to the river, and
identifies the hill with Mont Parnasse itself. De Saulcy^ prefers
Montrouge. But this problem is insoluble.^

WHERE DID CAESAR, AETER HIS RETREAT FROM

GERGOVIA, REJOIN LABIENUS? HOW LONG
DID HE WAIT BEFORE SETTING OUT, AND
FROM WHAT POINT DID HE SET OUT ON HIS
MARCH TO SUCCOUR THE PROVINCE ?

There are not sufficient data to enable us to answer any of these

questions with certainty.
I. After leaving Cergovia, Caesar marched down the left or
western bank of the Allier recrossed it on the third day of his
;

1
Rev. arch., vi, 1905, p. 260. ^ /^^ p 268. ^ See pp. 779-80.
*
See M. Blanchet's sagacious remarks in Rev. arch., 4" ser., vii, 1906, pp. 173-4.
•^
Ih. vi, 1905, pp. 264, n. 1, 269-70.
« B. G., vii, 60, ' Rev. arch., vi, 1905, p. 270.
§ 4.
* Les campagnes de Jules Cesar dans les Gaules, p. 34.

» The theory of M. E. Toulouze {Rev. arch., 3« ser., xviii, 1891,

pp. 163-85)
that Labienus fought a naval battle with Camulogenus in the Seine, near Mor-
sang, rests upon nothing but assumptions, and is not worth discussing. If the
assumed battle had taken place, Caesar would certainly have mentioned it.
1093 3 E
786 WHERE DTD CAESAR
march pushed on to relieve Labieniis, who was in the neighbour-
;

hood of Lutecia (Paris) crossed the Loire by a deep ford into the
;

country of the Aedui and, having revictualled his army there,

;

marched on towards the country of the Senones. Meanwhile Labienus

defeated a rebel army on the left bank of the Seine, near Paris then ;

marched southward to Agedincum (Sens) and thence marched to ;

rejoin Caesar.^ For some weeks active hostilities between Caesar

and the rebels were suspended and within this interval Caesar sent
;

for, and was reinforced by, a body of German cavalry from beyond
the Rhine. At length he marched to succour the Province, which
was threatened by a rebel force and the route which he took led
;

'
through the furthest part of the country of the Lingones into the
country of the Sequani (in Sequanos per extremos Lingonum fines).
'

In relating the fact that Labienus marched from Sens to join him,
Caesar, according to the ordinary reading that of the y8 MSS.— — uses
these words mde cum omnibus copiis ad Caesarem pervenit.
: ABL have
^ M. Jullian {Hist, de la Gaule, iii, 463, n. 2) has raised the question whether

Labienus' s victory took place (as had always been inferred from Caesar's narra-
tive) after Caesar was defeated at Gergovia, or before. Remarking that Caesar
'
appears to say in B. G., vii, 59, § 1, that Labienus learned [immediately after
he reached Lutecia] that Caesar had quitted Gergovia', he concludes that, if
this is true, (1) Labienus did not leave Agedincum until long after the beginning
of Caesar's campaign in Auvergne ; (2) the transmission of the news of Caesar's
retreat from Gergovia, Labienus's victory and return from Lutecia to Agedin-
cum, and his march thence to rejoin Caesar cannot have occupied more than
or 7 days and (3) Caesar must have taken that time to march from Gergovia
;

to the point of junction. All this, he admits, is not impossible ; nevertheless he

believes that the news which reached Labienus was merely an exaggerated
version of the first defection of the Aedui (see B. 6'., vii, 37-43, and pp. ]ol-5).
I do not agree with M. Jullian for Caesar's statement in vii, 59, § 1, is
;

iinequi vocal. (1) If Labienus accompanied Caesar to Decetia (Decize), his

march from Decetia to Agedincum was longer than Caesar's march from Decetia
to Gergovia, and his march from Decetia to Lutecia was more than twice as
long it is quite probable that he was obliged on his way to Agedincum to
;

spend a day or two at Noviodunum (Nevers) in order to organize the depot

which Caesar established there (see p. 149) Caesar's campaign in Auvergne
;

evidently only lasted a few days ;and therefore the rumours which Caesar
describes would have had plenty of time to reach Labienus. (2) The distance
of Gergovia from Lutecia is about half as far again as the distance of
Gergovia from Cenabum (Orleans) ; news travelled from Cenabum to Ger-
govia in about 12 hours (vii, 3, § 3), and might therefore have travelled
ifrom Gergovia to Lutecia in less than a couple of days. On the day after
Labienus heard the news he quitted Lutecia and his march from Lutecia to
;

— —
Agedincum about 110 kilometres would have occupied 4 or 5 days. Thus
he would have reached Agedincum 7 or 8 days after Caesar's defeat. (3) Caesar
marched from Gergovia to the bridge by which he crossed the Allier in 3 days
(vii, 53, § 4), repaired the bridge, and then marched to the ford, near Nevers,
by which he crossed the Loire. This last march occupied at least two and pro-
bably three days (see p. 774 and cf. C. Jullian, Hist, de la Gaule, iii, 481, n. 5).

— —
After the troops had forded the Loire they spent some time probably a day
at least in collecting corn and cattle (-B. G., vii, 56, § 5). Then they marched
up the road from Noviodunum to Agedincum to rejoin Labienus. The distance
of Noviodunum from Agedincum is about 150 kilometres. If the junction took
place about two marches, say 50 kilometres, from Agedincum, the chronology
is satisfactorily explained. Why, then, should Caesar's statement of the
rumours which reached Labienus be distorted from its natural sense ?
B. G., vii, 53-66.
•^

I
 —

REJOIN LABIENUS ? 787

indiem (cum omnibus), which means nothing and MQ have indie.

;

Whitte conjectured that Caesar wrote inde die III} But even if
the emendation is correct, the data for fixing the place of junction
are insufficient, because we cannot tell how far Labienus marched
each day.
1. Napoleon, assuming, wrongly as I believe,^ that the ford by
which Caesar crossed the Loire was at Bourbon-Lancy, says that
his junction with Labienus must necessarily have taken place on
'

a point of the line from Bourbon-Lancy to Sens '. This point,' he '

adds, in our opinion is Joigny.' ^

'

2. Mr. W. C. Compton,^ who bases his opinion upon the doubtful

emendation die III., thinks that Joigny is too near Sens, and suggests
the confluence of the Arman9on and the Yonne.
3. General Creuly holds that Caesar crossed the Loire near Novio-
dunum (Nevers), and rejoined Labienus at some point between
Nevers and Sens, or more probably at Nevers itself for, he argues, ;

Caesar does not say that he actually penetrated into the country of
the Senones, but only that he intended to do so (iterfacere instituit).^
It is amusing to note that Creuly affirms elsewhere that, in Caesar,
'
le verbe instituere a toujours le sens d'une action et non pas d'une
simple resolution.' ^
4. C. Rossignol insists that Caesar crossed the Yonne at Auxerre,
arguing that if he had crossed it lower down, he would have found
himself at Agedincum, whither he had no intention of going and ;

that if he had crossed higher up, he would have struck off too soon
from the route, leading into the country of the Senones, which he
took after crossing the Loire. I may remark that it is not proved
that Caesar crossed the Yonne at all before he rejoined Labienus.
Rossignol goes on to argue that the junction must have taken place
at Eburobriga, near St. Florentin. He gives the following reasons :

first, at Eburobriga, Caesar would have been near enough to the

Aedui to be able to observe their movements secondly, he would ;

have been on the frontier of the friendly Lingones thirdly, he ;

would have been on the natural line of communication between the

basins of the Seine and the Saone, that is to say on a route which
would have led him into the country of the Sequani fourthly, in ;

the passage Labienus revertitur Agedincum inde cum omnibus

. . .

copiis ad Caesarem pervenit the word pervenit shows that Labienus,

after leaving Agedincum, did not change his direction and fifthly, ;

one day's march from Agedincum, in the same direction, would

have brought him to Eburobriga.'^ Good or bad, the first four of
^ Meusel's Lex. Caes., ii, 155.
2 See p. 774. 3 jji^t ^te Jules Cesar, ii, 292.
* Caesar's Seventh Campaign in Gaul,
p. 101.
•''

Rev. arch., nouv. ser., viii, 1863, pp. 502-3.

® This remark of mine, I find, has been anticipated in substance by Heller in

Philologus, xxii, 1865, pp. 169-70.

' Mem. de la Comm. des ant. du dep^ de la Cote-d'Or, v, 1857, pp. 11-4.
Reference to Meusel's Lex. Caes., ii, 95, 1075, and especially to B. G., vii, 58, § 2,
will show that Rossignol's remark about pervenit is nonsense. Rossignol, indeed,
argues that if Caesar had marched past Auxerre and had not crossed the Yonne,
3 E2
788 WHENCE DTD 0AE8AE MARCH
these arguments might be used to prove that Caesar rejoined Lahienus
at some other place, and the fifth is worthless because there is nothing
to show how far or how long Labienus marched after leaving Sens.
Moreover, there is nothing in Caesar's narrative to show that the
place where he rejoined Labienus was identical with the place where
he remained during the period of inaction that preceded his march
to succour the Province.
The foregoing theories agree in placing the junction on the road
between Nevers and Sens and, unless Caesar had some unaccount-
;

able motive for diverging from that route, they are so far right.
5. Von Goler,! however, believes that Caesar and Labienus met
at Troyes. Caesar, he says, had gone to Troyes in order that the
twelve cohorts which he had left in the preceding year to guard his
second bridge over the Rhine might rejoin him more easily. But
how does von Goler know that they had not rejoined him before ?
If they had not, how does he account for the fact that Caesar had
—
ten legions all that he is known to have had during this campaign
without them ? At Troyes, von Goler adds, Caesar would have
been near the friendly Remi, through whose territory the German
cavalry could have marched securely to join him. No doubt. But thej
same advantage could have been secured at other places nor was it' ;

necessary to secure it until after Caesar and his lieutenant had met.
Quot homines tot sententiae. But the sententiae will never lead to
anything. We
shall never know exactly where Caesar rejoined
Labienus. It is almost certain that the place was somewhere between
Sens and the point where Caesar crossed the Loire and we may be
;

sure that Caesar crossed the Loire at the nearest point that he could
safely find to the route that would lead him to Sens. I have shown
elsewhere ^ that the ford was not at Bourbon-Lancy, but somewhere
lower down the stream,
II. It would be futile to attempt to calculate the length of time
that elapsed between Caesar's junction with Labienus and his
departure for the Province. All that we can say is that it was
considerable, —
some weeks at least. For we are told that in the
interval Caesar had time to send across the Rhine for reinforcements
and to receive them that a Pan-Gallic council was convened and
;

met and that Vercingetorix, after he had been elected Commander-

;

in-Chief by this council, had time to raise new troops and send them
to attack the Province.^
III. L Napoleon considers that, during the period of inaction,
Caesar encamped not far from the confluence of the Armancon
'

and the Yonne ', that is to say, in the country of the Senones, and
near Joigny, where Napoleon places his junction with Labienus.

he would have entered a country the resources of which had already been drained
by the army of Labienus, and would have been too far from the faithful Lingones.
I cannot, however, see any force in either of these arguments for Caesar's army
;

liad just replenished their stores in the country of the Aedui, and in any case
they were about soon to enter the country of the I^ingones.
i'r?o//. Krieg, 1880, p. 290, n. 3,
2
See p. 774. ^ g q^ yij^ 63-5.
 TO SUCCOUR THE PROVINCE
n\
? 780

2. The Due d'Aumale^ assumes that the immediate object of

Caesar, after he rejoined Labienus, was to rest his soldiers and to
lay in a fresh stock of provisions. Although, continues the Due
d'Aumale, the Lingones were friendly, it would not have suited
Caesar to encamp near their chief town, Andematunnum (Langres) ;

for in that neighbourhood he would have been too far from the
Remi, who must have needed his protection against the Bellovaci,'^
and from whom he must have wished to obtain supplies his com- ;

munication with the Germans, from whom he was about to borrow

cavalry, would have been cut off by the hostile Mediomatrici and ;

he would have made it too evident to Vercingetorix that he intended,

sooner or later, to enter Sequania.^ Nor did he remain near Agedin-
cum ; why did Labienus abandon Agedincum before
for, if so,
rejoining him
Moreover, the Senones were hostile
? they had, in :

the preceding winter, been obliged to feed the Roman army of

occupation and what corn they had left, after supplying their own
;

wants, was probably, in obedience to the orders of Vercingetorix,

either destroyed or concealed.* The Due d'Aumale conchides that,
during the period of inaction which elapsed between his junction
with Labienus and the commencement of his march for the Province,
Caesar remained on the Aube, between Arcis and the river Voire, or
perhaps, as such a position might have been too close to hostile
territory, on the Marne, near Vitry.
3. General Creuly, believing that Caesar rejoined Labienus at or
near Noviodunum, argues that we may infer from his silence that
he remained there, because his silence would otherwise be inexcus-
able : he maintains that he had no intention of marching as far as
the Province in order to succour it {quofacilius suhsidium provinciae
ferre fosset), but only intended to go as far as some point in the
country of the Sequani, whence he could communicate equally with
the Aedui and the Province and he points out that Noviodunum
;

had the advantage of being comparatively near to the Province, and

that, situated as it was in the country of the Aedui, Caesar could
there exercise a decisive influence upon the political situation, and
might expect that the Aedui would return to their allegiance.^ But
it is not proved that Caesar rejoined Labienus at or near Novio-
dunum there is no evidence that, at that time, he had any intention
:

of succouring the Province at all he could easily have communi-

:

cated with the Aedui if he had remained comparatively near their

frontier and it is difficult to see why he should have thrown away
;

the alleged advantage of proximity to the Province by taking such

a roundabout way from Noviodunum to his destination as the route
through the country of the Lingones. Heller argues further that on
Creuly's theory it is incomprehensible that Vercingetorix should not
have opposed the junction of the two generals or that when Labienus
;

made his (assumed) march back to Noviodunum, Vercingetorix should

1 llcv. des Deux Mondcs, 2" per., xv, 1858, pp. 77-8.
2 B. G., vii, 90, § 5. ^ lb., 05,
§ 4 ; 00, § 1. * 10., 02, § 10 ; 04, § 3.
^ Eev. arch., nouv. sor., viii, 1803, pp. 502-5.
790 WHENCE DID CAESAR MARCH?
have divined Caesar's intention of marching past Alesia into tlie
country of the Sequani or, finally, that the German cavalry whom
;

Caesar induced to join him should have been able to make their
way unopposed to Noviodunum.^
The problem is insoluble. Most military men would, indeed,
agree with the Due d'Aumale that Caesar must have established
himself somewhere in the country of the Lingones. R. de Coynart,-
it is true, argues that he would not have taken up his quarters there
for fear of driving the Lingones to revolt by the requisitions of corn
which he would have been obliged to make and he concludes that
;

he remained in the country of the Senones. But Caesar had just

laid in a good stock of food, and was not likely to bear hardly upon
the Lingones for some time he did shortly afterwards march through
:

their country ^ if he thought it worth his while, he could pay for

:

what he took and the Lingones, having remained loyal so far,

;

doubtless saw that it would be to their interest to cleave to the

stronger side to the end. If Caesar had intended to remain in the
country of the Senones, he would not have voluntarily abandoned
their chief town, Agedincum. He would surely have taken up his
quarters, if he could have done so with due regard to strategy,
among a friendly people the Lingones were the only people, except
:

the Remi, upon whose friendship he could depend ^ and the country ;

of the Lingones was near that of the Remi, near that of the Aedui,
and conveniently situated for the reception of the expected rein-
forcements from Germany. We may also, perhaps, infer from
Caesar's silence that, after his junction with Labienus, he halted as
soon as he conveniently could and, as he tells us that, after he set
;

out on his march to succour the Province, he took a road leading

"
through the furthest part of the country of the Lingones {per '

extremos Lingonum fines), which must have been the southern or

the south-eastern part of their territory, it is clear that he must
have come from some point with regard to which that region was
most remote, that is to say, from some point either in the northern
or in the north-western part of their territory.^ But if we strive
after greater exactness, we have nothing certain to go upon.

WHY DID VERCINGETORIX ATTACK CAESAR WHEN

THE LATTER WAS MARCHING TO SUCCOUR
THE PROVINCE?
The Due d'Aumale argues that Vercingetorix's resolution to attack
Caesar, when the latter was marching to succour the Province, was
not inconsistent with his plan of campaign.^ Evidemment,' he '

says, Vercingetorix ignorait la presence des auxiliaires germains,

'

^
Philologus, xxii, 1805, pp. 169-70.
^
Spectateur militairc, 2^ hcy., xvii, 1856, pp. 220-1.
^
B. G., vii, 56, § 5 66, § 1-
;
* 1^-. 613, § 7.
'
Sec pp. 702-4. « Sec B. 0., vii, 14, 64, §§ 1-3.
 VERCINGETORIX ATTACKS 791

ou il se faisait illusion sur leur nombre

II ne comp- et leur valeur.
tait pour rien rinsignifiante cavalerie gauloise qui accompagnait
Cesar un mois plus tot. II croit n'avoir affaire qu'a rinfanterie des
legions. II sait combien elle est redoutable, mais, embarrassee qu'elle
est de bagages, il espere la condamner a une immobilite fatale ou
a une retraite qui aurait ressemble a celle de 1812, car lui aussi avait
ses Cosaques.' i
M. Jullian ^ is not wholly satisfied with this explanation. He offers
two alternatives. Perhaps, he suggests, Vercingetorix simply accepted
the inevitable. The unchallenged mastery which he had exercised
over his host in the earlier stages of the campaign was gone for the ;

leaders of his new

were jealous of his renown. Would the
levies
Aedui and the Sequani be unselfish enough to hearken to his counsel
and devastate their lands ? Sooner or later his cavalry would insist
on striking a blow would it not be best to let them have their way
:

while the circumstances were comparatively favourable ? Or again,

he may only have contemplated a flank attack upon Caesar's baggage-
train, but his officers may have got out of hand. Whichever of the
three hypotheses we adopt, says M. Jullian, we are confronted with
this fact :
—
la furie gauloise, contenue depuis six mois chez les
'

soldats et chez le chef, devait etre un jour plus forte que leur volonte
a tons.' M. Jullian knows the Gallic temperament but I think ;

that the chief was made of sterner stuff, as hard-headed, as much —

master of himself as when he warned the Bituriges to destroy Avari-
cum. Nor is Caesar's narrative consistent with the theory that he
allowed his judgement to be overruled.

WHAT WAS THE THE BATTLE BETWEEN

SITE OF
CAESAR AND VERCINGETORIX, WHICH IM-
MEDIATELY PRECEDED THE BLOCKADE OF
ALESIA ? 3
The data
for fixing the site of this battle are scanty but the search ;

is not hopeless. Caesar, marching northward from Gergovia to

succour Labienus, crossed the Loire and pushed on for the country
of the Senones Labienus, after his victory at Lutecia (Paris),
:

marched to Agedincum (Sens) and thence moved on to join Caesar.

Some time after their meeting, Caesar started from some point un-
known and marched through the country of the Lingones, intending
to get into the country of the Sequani, that he might be in a better
position for reinforcing the Province {quo facilius suhsidium pro-
'

vinciae ferre posset). Vercingetorix, who, after he had been con-

firmed in his office of commander-in-chief, had assembled his cavalry,
^ Rev. des Deux Mondes, 2" per., xv, 1858, p. 81.
2 Vercingetorix, pp. 248-53. Cf. C. Jullian, Hist, de la Gaule, iii, 497-9, and
p. supra.
1(57,
^ Sec Carle de V E tat- Major (1 : 80,000), Sheet 97, and Carte de France
(1 : 200,000), Sheet 34.
792 WHERE DID CAESAR DEFEAT
and apparently all his forces, at Bibractei (Mont Beuvray), marched to
intercept him but whether he started on this march from Bibracte
;

or from Alesia, is a disputed point. While Caesar was marching

through the furthest part of the country of the Lingones {cum
' '

. .per extrernos Lingonum fines iter facer et), Vercingetorix formed

.

three camps near the bank of a stream, 10 Roman miles from

Caesar's army, and next morning fought a cavalry battle with him
close to the stream. During the action the Gallic infantry were
drawn up on one of the banks of the stream, in front of their camps,
in order to produce a moral effect. On the right of the Roman army
there was a hill, down which Caesar's German cavalry charged. This
charge decided the battle and the Gallic cavalry, afraid of being
;

surrounded, fled. Vercingetorix forthwith retreated to Alesia (Mont

Auxois) Caesar pursued the beaten army until nightfall, and
:

arrived alter o die at Alesia .^

Altera die, as I have proved on pages 738-40, means, not on the '

second day (after),' but on the day (after) '. The words cum Caesar
'

in Sequanos per extremes Lingonum fines iter facer et have been much
discussed but to my mind they are tolerably clear. First of all,
;

they certainly mean that Caesar was still within the country of the
Lingones at the time when Vercingetorix made his three camps. It
has, indeed, been argued that when Caesar said that he was march-
ing in Sequanos through the furthest part of the territory of the
'

Lingones ', he meant that he was marching through the country of

the Sequani But I shall not notice this absurd interpretation here.
!

I have demolished it on page 356, because it was devised in order

to bolster up the theory that Alesia was not on Mont Auxois.
Heller, referring to a passage in B. G.,\, 1, § 6, argues that Caesar
used the word extremos from the point of view of the Province, and
therefore that he was describing the northern or north-eastern part
of the country of the Lingones.^ But Heller's premiss is wrong.
The passage to which he refers Belgae ah extremis Galliae finihus
oriuntur (' The territory of the Belgae, commencing from the most
—
distant frontier of Gaul,' &c.) is perhaps spurious^ and, anyhow,
has no analogy with the one which I am discussing. In the former
passage the writer is describing the geography of Gaul from the point
of view of an Italian and Caesar had already fixed the meaning of
;

ab extreinis Galliae finihus by saying that the Galli are separated

from the Belgae by the Marne and the Seine. In the other passage
Caesar is describing his march towards the Province from a northern
standpoint.
It has been argued that per extremos Lingonum fines might mean
'
through the southern part of the country of the Lingones, and close
to and parallel with their southern frontier '. De Coynart, indeed,
holds that the words must mean this. 'Faire voyage (iter facere),'

^ If hue {B. G., vii, 04, § 1) is right. Meiisol in the reissue of his school edition
deletes it. 2 B. G., vii, 5G,
§§ 3-5 ; G2, § 10 ; GG-8.
^ Philoloijus, xiii, 1858, p. 595. {See also xxii, 18G5, pp. 125-G.
* See p. 394.
 — —

VERCINGETORIX'S CAVALRY ? 793

he ne s'applique jamais qu'a une ligne suivie et iioii a uiie

insists, '

ligiie coupee.' ^ But de Coynart is over subtle. According to him,

Caesar was marching par la zone frontiere meridionale des Lin-
'

gons,' with the intention of crossing their south-eastern frontier.

Surely we see here une ligne coupee as well as une ligne suivie '.
' ' '

According to the view which de Coynart opposes, Caesar was march-

ing through the extreme south-eastern part of the country of the
Lingones, with the intention of crossing their southern or south-
eastern frontier. Surely we see here une ligne suivie as well as ' '

'
une ligne coupee '. Nobody would contend that Caesar meant
that, at the moment of which he spoke, he was actually crossing the
frontier for he rarely uses the word fines in the sense of frontier ',
;
'

and when he does so, the sense is unmistakable.^ Iterfacere simply

—
means to march ', in any direction, whether there are frontiers
'

to be crossed or not.^ If the words per extremos Lingonum fines

might mean through the southern part of the country of the Lin-
'

gones, and close to and parallel with their southern frontier ', they
might also mean through the eastern part of the country of the
'

Lingones (however far from their southern frontier), and close to

and parallel with their eastern frontier '. In other words, Caesar
would have made use of a phrase which anybody could interpret as
he pleased. His words can only, I believe, mean one thing,
'
through that part of the country of the Lingones which was furthest
from his starting-point in the direction in which he was going.'
G. Gouget* aptly quotes a passage from Caesar, which seems to
show that this was his meaning mandat ut crehros exploratores in
:

Suehos 7nittant, quaeque apud eos gerantur cognoscant. LIU imperata

faciunt et paucis diehus intermissis referunt Suehos omnes, postea
quam certiores nuntii de exercitu Romanorum venerint penitus ad . . .

extremos fines se recepisse.^ Now the direction in which Caesar

was going was towards the country of the Sequani, and their country
extended to the east and south-east of the country of the Lingones.
Entre Test et le midi,' writes the Due d'Aumaie, entre " le
' '

long de " ou " au travers de ", je reconnais que le texte des Com-
mentaires laisse toute liberte de choisir.' ^ But if Caesar was march-
ing from north to south through the eastern part of the country of
the Lingones and at a relatively considerable distance from their
southern frontier, he was not marching per extremos Lingonuyn fines.
'
For that part of the country could only have been called extreme '

in relation to the western or north-western part and if Caesar had ;

started from the western or north-western part, he would not have

^ Spectateur militaire, 2^ ser., xxx, 18C0, p. 426.
2 See p. 376. ^
^ gee Meusel's Lex, Caes., i, 1269.

* Mem. 'presentes par divers savnnfs a VArad. des wscr. et belles-lettres, vi, 1864,
pp. 208-12.
^ ' He instructed (the Ubii) to send
numerous scouts into the country of the
Suebi and to ascertain what they were about. The Ubii fulfilled their instruc-
tions, and, after the lapse of a few days, reported that all the fSuebi, on the
arrival of messengers with trustwortliy information about tlie Iloman army,
liad retreated to the furthest extremUi/ uf their country.'
. . . B. G., vi, 10, §§ 3-4.
* Rev. des Deux Mondes, 2" per., xv, 1858, p. 83.
794 WHERE DID CAESAK DEFEAT I

inarched eastward right across the country of the Lingones, and then
suddenly struck off towards the south. Moreover, according to all
the writers, except von Goler, who make him pass through the
southern, as distinguished from the south-eastern part of the country
of the Lingones, he was moving either up the valley of the Serein or
up the valley of the Arman§on or between the two rivers and in ;

any of these cases he would have been forced to pass through the
country of the Mandubii before entering Sequania. I do not think
that his words can bear this meaning and I find it difficult to
;

believe that he would have used a phrase which left his readers
'toute liberte de choisir between two widely different interpretations.
'

It will perhaps be objected that the battle-field may possibly

have been in the country of the Sequani because Caesar only says
;

that he was marching per extremos Lingonwn fines on the night

before the battle, and that the camp of Vercingetorix, near which
the battle took place, was 10 Roman miles from the point where the
Roman army encamped for the night. But it would have been
absolutely impossible for Caesar to reach Alesia on the day after
the battle from any point where the battle could be placed in the
country of the Sequani. I am aware that Dion Cassius ^ says that
the battle took place in their territory but thai part of Dion's
:

history which deals with the Gallic war is, as I have shown,^ full of
ridiculous blunders ; and in this case he either neglected to attend to
Caesar's narrative or misunderstood it.
It seems clear, then, that we must look for the battle-field some-
where in the south-eastern part of the country of the Lingones,
that is, in the neighbourhood of Dijon. But, as I do not anticipate
that every reader will agree with my interpretation of extremos
fines, I shall also examine the other sites that have been proposed.
I have discussed on pages 788-90 the question of the point from
which Caesar started on his march, and have arrived at the con-
clusion that it is impossible to fix it exactly, but that it was some-
where in the north or north-west of the country of the Lingones.
The question remains, from what point did Vercingetorix march to
intercept him ? He may possibly have started from Bibracte (Mont
Beuvray), where the council assembled by which he was elected
commander-in-chief where also, it should seem, the new levies
;

assembled which he ordered after his election ^ but it is much more

:

probable that he started from Alesia. We may infer, as I have

shown on page 165, from Caesar's narrative, that Vercingetorix
provisioned Alesia before the siege and the Due d'Aumale points
;

out that the country all round Alesia is intersected by the various
routes some one of which Caesar must have taken in order to reach
the country of the Sequani, and that Alesia itself was the best
position which Vercingetorix could have selected in order to rest in
security, to observe Caesar's movements, and to sally forth at the
right moment to harass his march.*
1 xl.
39, § 1. - See pp. 21G-7. B, G., vii, G3, §§ o-G
""
G4, § 1.
;

* Rev. dcs Deux Maudes, 2« per., xv, 1858, pp. 94-5. When the Due d'Aumale
infers (p. 82) from the worda in which Caesar describes the retreat of Vcrciu-
 —

VERCINGETORIX'S CAVALRY ? 795

1. According to Napoleon,^ Caesar started from Joigny, crossed

the rivers Arman9oii, Seine, and Aube struck off in a south-easterly ;

direction from Dancevoir up the right bank of the Aube and en- ;

camped on the night before the battle on the western bank of the
river Vingeanne, near Longeau. Vercingetorix, who had marched
from Bibracte by way of Dijon, to intercept him, encamped on the
heights of Sacquenay, overlooking the southern bank of the Badin,
a rivulet which flows into the Vingeanne on its western bank. In
this position, says Napoleon, he commanded the three roads, leading
respectively towards Gray, Pontailler, and Chalon, by one of which
Caesar must have intended to advance to the Saone. The battle
took place in the angle between the Vingeanne and the Badin and ;

the hill from which the Germans charged was Montsaugeon.

Napoleon argues first, that the alleged battle-field answers per-
fectly to Caesar's description secondly, that skeletons many of
;
'

which had bronze bracelets round the arms and legs ', as well as
bones of men and of horses, have been found in twnuli for some
distance along the line of retreat which Vercingetorix would have
taken to Alesia thirdly, that numbers of horseshoes
; evidently —
—
shoes of dead horses were found in 1860, at the dredging of the
Vingeanne and lastly, that Caesar could have reached Alesia on
;

the second day after the battle.

Von Kampen ^ agrees with Napoleon. He holds, it is true, that
altero die means on the next day
'
but he believes nevertheless '
;

that Caesar did not reach Alesia until the second day after the
battle. As regards the words altero die ad Alesiam castra fecit,
'

I think,' he says, that everything depends upon the point of time

'

from which the calculation is made.' Of course But it is absolutely !

clear from Caesar's words that that point of time is the day of the
battle. The actual move to Alesia,' von Kampen continues, did
' '

not take place till after the day of the battle for on the day of the ;

battle itself there was enough to do in attending to the baggage, and

also the return of the troops who pursued the rear of the enemy till
the evening had to be awaited.' But only two of the ten legions had
to attend to the baggage and Caesar does nofc say that the troops
;
'

who pursued the rear of the enemy did return. should they have
'
Why
returned ? Von Kampen's explanation cannot be got out of the Latin.^
getorix to Alesia copias suas, ut pro castris conlocaverat, reduxit protinusque
Alesiam, . iter facer e coepit
. . —
that le chef gaulois revint sur ses pas apres le
'

combat', he makes a mistake. The word reduxit simply means that after the
battle and before commencing his retreat, Vercingetorix withdrew his infantry
from the bank of the stream into camp. If reduxit meant led back (to Alesia), '
'

Alesiam iter facere coepit would simply mean the same as copias reduxit, and
protinus would mean nothing at all.
^ Hist, de Jules Cesar, ii, 294-9.

^ Quindecim ad Caesaris de b. G. comm. tabulae, xii.

^ Fugato omni equitatu, Vercingetorix copias, ut pro castris conlocaverat,

reduxit protinusque Alesiam iter facere coepit celeriterque impedimenta ex

. . .

castris educi et se subsequi iussit. Caesar impedimentis in proximum collcm

dcductis, duabus Icgionibus pracsidio rclictis sccutus hostes, quantum dici
tompus est passum, circitcr 111 milibus ex novissimo agminc intcrfectis altero
die ad Alesiam castra fecit.
796 WHEKE DID CAESAR DEFEAT
Napoleon's arguments are, one and all, worthless. The discovery
of bronze bracelets and of horseshoes proves nothing
indeed it is :

morally certain that neither Gallic nor Koman horses were shod
with nailed shoes, though they may occasionally have worn shoes
fastened with straps, which the French call hipposandales.^ Granted
that the alleged battle-field corresponds with Caesar's description,
other sites do the same. Napoleon's interpretation of altero die is
wrong.2 Caesar reached Alesia on the day after the battle and in ;

order to do so, he would have been obliged, if the battle had been
fought between the Badin and the Vingeanne, to march 80 kilo-
metres, or 50 miles, and to fight a battle, within two days. Again,
for no conceivable reason except to bring him to the mythical battle-
field on the Badin, Napoleon makes Caesar travel far away from his
natural route. Assuming, with Napoleon, that Caesar marched from
Joigny to Tonnerre, and from Tonnerre to Chatillon, he would surely
have moved from Chatillon up the valley of the Ource, and then
down the valley of the Tille, instead of turning away towards the
north-east and then, by a long detour, doubling back again towards
the south. Moreover, Napoleon admits that Vercingetorix's retreat
would have been cut off if he had returned to his camp, and accord-
ingly denies that he did so. But Caesar says that he did !

De Coynart ^ gives an additional reason for rejecting Napoleon's

site. He points out that it is 24 kilometres from the eastern, and
58 kilometres from the southern frontier of the Lingones, and
observes that such a position cannot be reconciled with Caesar's
statement that he was marching per extremos Lingonum fines.
2. D'Anville ^ thinks that the battle took place on the Armanyon,
somewhere between Tonnerre and Ravieres but he gives no reasons.
;

The Due d'Aumale and M. Gouget ^ have shown that he was wrong.
'^

First of all, the only site between Tonnerre and Ravieres that at all
answers to Caesar's description is nearly midway between the en-
virons of Agedincum, from which Caesar is supposed by d'Anville
to have started, and Bibracte, from which Vercingetorix is supposed
to have started to intercept him. Vercingetorix must then have left
Bibracte on the same day that Caesar left the environs of Agedincum,
and before he knew anything of Caesar's intended movements. Such
a coincidence is highly improbable. I do not see how this objection
can be answered, unless one may suppose that Vercingetorix started
a day later than Caesar, but marched much faster, or that he
had somehow got information beforehand of Caesar's intentions.
Secondly, the battle could hardly have taken place on the southern
bank of the stream for, as the Due d'Aumale puts it, il serait pen
;
'

^ See Bev. des Soc. savantes, x,

1869, p. 337, Daremberg and Sagho, Diet,
des ant. grecques et rom., iii, 2011-4, and Pro Alesia, 1908, pp. 449-51, 1909,
pp. 476-80. 2 See
pp. 738-40.
3 See Mem. de la Comm. des ant. du dep^ de la Cote-d'Or, viii, 1873, pp-
162-3, 168. * £!claircisscmens sur Vancienne Guide,
pp. 452-3.
^ Rev. des Deux Mondes, 2^ per., xv, 1858,
p. 97.
•^
Man. presentes par divers savants a V Acad, des inscr. ct bellcs'lettres, vi, 18G4,
pp. 216-21.
 —

VERCINGETORIX'S CAVALRY ? 797

vraisemblable que Cesar eut neglige de se oouvrir dii flanc dans sa

marche [up the valley of the Arman9on], et que Vercingetorix eut
pris position tournant le dos au territoire ennemi et faisant face a sa
base d'operations.' Moreover, if the battle had been fought on the
southern bank, the Gauls would have been obliged, in order to reach
Alesia, to cross the river immediately after the battle and this
;

would have been a serious operation for a beaten host to undertake.

The battle must then have been fought on the northern bank. But
if so, the Roman army, marching up the valley, must have had its
right nearest to the bank. Now it was on the right that the German
cavalry, having seized a hill, charged the Gallic cavalry and decided
the battle. Therefore the hill must have been close to the river.
Only one spot that answers to these conditions can be found,
between Stigny and Eavieres and there the ground is covered with
;

—
woods and deeply scored by ravines, in other words, unsuited to
a combat of horse.
3. Von Goler ^ finds the battle-field near Beneuvre. Arguing on
the fantastic assumption that Caesar started from Troyes,^ he holds
that he marched by way of Chatillon-sur-Seine with the intention
of passing Til-Chatel and crossing the Saone at Gray. I do not think
that Beneuvre is in extremis Lingonum finihus.
4. Rossignol does not attempt to point out the exact site of the
battle-field but he places the three camps of Vercingetorix at
;

Montbard, Nogent, and Courcelles, on the river Brenne.^ I shall not

examine his arguments for he supposes that Vercingetorix was
;

encamped only 8 kilometres, or 5 miles, in a direct line, say 6 miles

by road, from Alesia and this distance is obviously far too short.
;

5. De Coynart traces Caesar's route along the high ground between

the rivers Arman9on and Serein. He translates cum by pendant '

que and he takes trinis castris in the sense of three marches *

'
;
' !
'

He assumes that Vercingetorix started from Bibracte, which he

wrongly identifies with Autun, and that he marched about 27 kilo-
metres, or nearly 17 miles a day. This string of assumptions he
regards as data sufficient for determining the site of the battle and ;

he finds it near Moutier St. Jean, on the left bank of the Armancon,
about 12 miles by road from Alesia. He places the camp of Vercin-
getorix on a chain of hills on the right bank and he tells us that the
;

Gallic cavalry attacked the Roman column as it was marching along

the ridge between Vassy and the wooded eminence of Bar.
This theory is open to the same objection as that of d'Anville.
First, Vercingetorix, ex hypothesi, started from Bibracte to intercept
Caesar on the same day that Caesar started to relieve the Province,
Gall. Krieg, 1880, pp. 300-1.
1 2 g^e
p. 788.
Mem. de la Comm. des ant. du dep^ de la Cote-d'Or, v, 1857, pp. 16-7.
^

* It is true that Caesar . quintis castris Gergoviam pervenit {B. G., vii, 36, § 1)
. .

means ' Caesar reached G ergo via in five marches ' : but between that phrase
and circiter milia passuum X ah Romanis trinis castris Vercingetorix consedit there
isno analogy. According to de Coynart, these words mean that Caesar and
Vercingetorix marched exactly the same distance each day for three days, and
always encamped exactly 10 Roman miles apart and that all this time Caesar
;

was marching per extremos Lingonum fines !

798 WHERE DTD CAESAR DEFEAT
without knowing what route Caesar intended to take. Secondly,
the beaten Gallic cavalry must have crossed the Arman9on, with the
Romans at their heels, immediately after the battle. And lastly, the
site is too near Alesia.
6. I now come to the theory of the Due d'Aumale.^ He bases his
search upon the hypothesis that Caesar started on his march from
the neighbourhood of Vitry,^ and advanced up the left bank of the
Aube that Vercingetorix marched to intercept him from Alesia
; ;

and that (according to the true interpretation of altero die) the battle
must have been fought at a distance of not more than 30 or less than
15 miles from Alesia. He concludes that Caesar encamped, on the
night before the battle, about 7 miles south of Ferte-sur-Aube that ;

Vercingetorix encamped on a ridge rising above the right bank of the

Ource, near Presly and that the battle took place in an undulating
:

plain between Montigny and Louesme.

Now, on either of the two interpretations of the words fer extremos
Lingonum fines which the Due d'Aumale himself admits, he must be
wrong. Let the reader look at the map, and he will see that Caesar's
alleged position was almost in the centre of the country of the Lin-
gones. The Due d'Aumale endeavours, indeed, to reconcile his theory
with the words of Caesar by saying that Caesar's camp was pres de '

la frontiere lingonne {extremos fines Lingonum), decoupee de ce cote

par I'enclave mandubienne.' In reality, it was nearly as far from
the northern frontier of the Lingones as from the common frontier
of the Lingones and the Mandubii. But even if it had been quite
close to the latter frontier, theargument of the Due d'Aumale would
be of no avail. He forgets that Caesar was marching towards the
country of the Sequani, not that of the Mandubii and Dijon,
;

the point at which he conjectures that Caesar intended to cross the

Lingonian frontier, is very far from the assumed encampment.
Again, as Pistollet de St. Ferjeux points out,^ Vercingetorix, being
about 10 kilometres, or more than 6 miles from the battle-field,
could not have directed the movements of his cavalry or supported
them with his infantry.
The Due d'Aumale has also suggested that the battle may have
been fought in the neighbourhood of Is-sur-Tille. But this place is
14 miles north of Dijon. If the battle had been fought there, would
Caesar have said that on the previous night, when he was several
miles still further north, he was in extremis Lingonmn finibus ?
7. Pistollet de St. Ferjeux, assuming that Caesar started from
a camp the remains of which are visible on the hill of Ste-Germaine
near Bar-sur-Aube, and that he intended to cross the Saone near
Pontailler or St. Jean-de-Losne, places the Roman army, on the night
before the battle, near Arbot on the Aube, and the three camps
of Vercingetorix on three hills separated from one another by the
sources of that stream. The battle-field was a little to the north
of these hills Vercingetorix himself was between the Aube and the
:

1 Rev. des Deux Mondes, 2* per., xv, 1858, pp. 87, 94-5. - See p. 789.
^ Spectateur militaire, 2^ ser., xlii, 1863, pp. 00-1.
 VERCINGETORIX'S CAVALRY ? 700

hillon which stands the chapel of St. Remi and the hill from which
;

the German cavalry charged is identified with that of Charbonniere.i

But Arbot was not in extremis Lingonum finibus it was about :

30 miles from their eastern, and more than 50 miles from their
southern frontier.
8. F. Monnier ^ makes Vercingetorix encamp at the village of
Senailly on the right bank of the Arman9on. Now from Genay,
where Monnier tells us that the retreat of Vercingetorix began, to
Mont Auxois, the distance is not more than 10 miles and it is hard ;

to see why the Romans should have thrown away the immense
advantage of continuing the pursuit, on the day of the battle, right
up to and into Alesia.
9. Virtually identical with the view of Monnier is that of General
Creuly. He makes Caesar encamp before the battle at Montreal-
sur-Serein, and Vercingetorix near Visernay, which is about a mile
and a half south-east of Senailly. The distance from Visernay to
Alise is, he says, 18 kilometres, or 11 miles,
—
c'est qui convient
'

aux details de cette journee et de la suivante.' ^ I do not think so.

Every one of the foregoing theories, the reader will have perceived,
conflicts with the only sound interpretation which, in my judgement,
can be given of the words per extremos Lingonum fines.
The next view which I shall examine is that of Gouget.^
10.
He assumes that Caesar started from Langres (Andematunnum),
the chief town of the Lingones and he interprets per extremos
;

Lingonum fines in the same way that I do. Caesar's best route led,
he considers, down the valley of the Tille, and thence to the Saone,
which he would have crossed either near Auxonne or near St. Jean-
de-Losne. Pursuing this route, says Gouget, he must needs strike
the river Ouche, and cross it somewhere between Dijon and the
Saone. Evidently, then, Vercingetorix, who was preparing to inter-
cept him, encamped, the night before the battle, on the Ouche.
The left camp was on the southern bank of the Ouche, opposite
Dijon the central camp was on the northern bank, in the peninsula
:

between the Ouche and its affluent, the Suzon the right camp was
;

on the southern bank of the Ouche, just below its confluence with
the Suzon. The Romans encamped, the night before the battle, at
Arc-sur-Tille. The battle took place on the high ground which
extends along the northern bank of the Ouche from Dijon to Fau-
verney. The hill {summum iugum) from which the Germans made
their decisive charge was the culminating point of the gently sloping-
heights between St. Apollinaire and Mirande. The battle-field was
about 32 miles from Alesia, whither Vercingetorix retreated, up the
right bank of the Ouche.
Perhaps it is hypercritical to say that Gouget rather under-
estimated the distance. As the crow flies, it is about 48 kilometres ;

1
76., pp. 63-5, 67, 69, 74. 2 Vercingetorix, 1883, pp. 197-200.
^ Rev. arch., nouv. ser., viii, 1863, p. 509.
* Mem. presenter par divers savants a V Acad, des inscr. et belles-lettres,
vi, 1864,
pp. 208-12, 214, 228-33, 241, 255-6.
800 WHERE DTD CAESAR DEFEAT
and, allowing for the windings of the road, the actnal distance which
Vercingetorix and Caesar would have had to accomplish can hardly
have been less than 54 kilometres, or between 33 and 34 miles.i
Still, I have no doubt that such a march was within the bounds of
possibility. A considerable part of the distance would doubtless
have been covered in the pursuit on the day of the battle. Caesar
had every motive for pursuing his beaten enemy as hard as he could.
During the operations before Gergovia he made a forced march of
50 Roman miles in 28 hours.^ Is it incredible that, on the long
summer's day that followed this battle, he should have made a
forced march of 20 or 25 miles ?
It might, perhaps, also be objected to Gouget's view that the
slope of the ridge which he identifies with the hill from which the
Germans charged is too gentle. But it would be a great mistake to
suppose that that hill could have been really steep. It is practically
impossible, says Lord Wolseley,^ for cavalry to charge down a hill
of Vv^hich the gradation is as much as 10°.
Other reasons, however, based upon a careful examination of the
ground, compel me to discard Gouget's theory, which I was formerly
inclined to accept. The lie of the country seems to show that Caesar
would not have approached the Ouche by the line which Gouget
traces and, moreover, the ridge between St. Apollinaire and
;

Mirande, if not the whole battle-field, would have been invisible

from any of the supposed Gallic camps.
1 am glad that M. Jullian, who in his biography of Vercingetorix ^
adopted Gouget's theory in its entirety, still agrees with me in accept-
ing it in principle but in detail he has recently ^ somewhat modified
;

it, placing the battle-field nearer Alesia, about four miles north of

Dijon and just west of Bellefond. He suggests that Caesar encamped

the night before the battle on a hill north of Til-Chatel and over-
looking the Tille, and that he continued his march along the line
of the Roman road which traverses Bellefond. Vercingetorix, he
thinks, encamped on the heights of Hauteville, north-west of Dijon
and separated from Caesar by the river Suzon, thus ensuring his
retreat to Alesia. If so, the hill which the Germans occupied on the
right was the ridge just north of Asnieres. M. Jullian tells us that
what led him to discard Gouget's details was that he doubted
whether an ancient road had existed along the line which Gouget
supposed Caesar to have followed. Be this as it may, M. JuUian's
theory has the advantage of reducing the distance of the battle-
field from Alesia and the Suzon, which is now dry, has very high
;

banks, and would have been a sufficient obstacle. On the other

hand, Vercingetorix could not have seen any of the fighting except
on Caesar's right wing for the ground sinks rather rapidly from
;

^ I have measured the distance by the direct road with a map-measurer on

Sheet 34 of the Carte de France (1 200,000).

:

2 See
pp. 153-4 and 635.
^ Soldier s Pocket-BooJc, 5th ed.,
p. 359. * Vercingetorix, pp. 379-82.
^ Rev. des etudes
anc.y x, 1908, pp. 347-50. Cf. C. Jullian, Hist, de la Ganle,
iii, 495, n. 1.
 VERCINGETORIX'S CAVALRY ? 801

the Route Nationale, which passes Asnieres on the east, towards

Bellefond and Ruffey. Still, the site appears to me the most
satisfactory that has been proposed and I therefore provisionally
;

accept it.i

WHY DID VERCINGETORIX RETREAT TO ALESIA ?

'
Vercingetorix,' says Mommsen,^ had been prepared
'
for a struggle
under the walls, but not for being besieged in Alesia from that
;

point of view the accumulated stores, considerable as they were,

were yet far from sufficient for his army.'
Merivale thinks that even though the battle [which preceded
'

the blockade] was lost, the cause might have been maintained by
recurrence to the harassing system in which the Gauls had hitherto,
with one exception, so steadfastly persevered.^ If their vast forces
had been dispersed or drawn out of Caesar's immediate reach, and
the country wasted around him, he would not, we may presume,
have ventured to protract an indecisive warfare under pressure of
the circumstances which urged him to seek the Roman frontier.
The victory he had gained would in that case have been destitute of
any decisive result. But the fatal mistake of assembling the whole
Gaulish army in one spot, and there tying it, as it were, to the stake,
offered an opportunity to his energetic spirit which he was not the
man to forego.'^
General J. B. Renard ^ conjectures that Vercingetorix retreated to
Alesia in the hope of keeping Caesar chained to the spot while the
hosts of united Gaul were preparing to come and join him, and
annihilate the Roman army. But if so, why did he not send off his
cavalry with the fiery cross at once ?
Caesar does not tell us why Vercingetorix shut himself up in
Alesia for he did not care to enlighten those who could not see
;

that Vercingetorix had no alternative. Perhaps he hoped to be as

successful there as he had been at Gergovia. At all events, it is
certain that, beaten as he was and hotly pursued by Caesar, he was
bound to establish himself in the strong position which he had pre-
pared, lest he should be compelled to fight another and yet more
disastrous battle, or, as Sir Coleridge Grove has suggested to me,
lest his disheartened followers should fall away.

^ M. P, Perrenet {Rev. des etudes anc, xi, 1909, pp. 253-5) has proposed
a modification, which in my opinion is not an improvement, of M. Jullian's
theory. He locates the battle-field alternatively between Messigny and Asnieres,
or near Til-Chatel between the Tille and the Ignon,
2 Hist,
of Rome, v, 1894, p. 88 {Rom. Gesch., iii, 1889, p. 290).
^ This is not certain. See p. 741, n. 6.
* Hist, oj the Romans under the Empire, ii, 1850,
p. 30.
° Hist. pol. et mil. de la Belgiqiie,
1847, pp. 479-80.

1093 3 F
 802

ON THE NUMBER OF LEGIONS WHICH CAESAR

HAD AT ALESIA
From what Caesar himself says^ we should gather that, during
the blockade of Alesia, he had ten legions under his command. For
he sent ten legions into winter- quarters at the close of the sixth
campaign just before he started for Gergovia, he divided his army
:

into two parts, assigning four legions to Labienus and keeping six
himself and he nowhere says or implies that another legion joined
;

him. But Napoleon ^ argues that he had eleven. He points out that,
according to Hirtius ^, one of the two legions which, after the fall of
Alesia, Caesar sent into winter- quarters on the Saone, was the 6th ;
and, assuming that this was not one of the ten which Caesar's own
narrative accounts for, he concludes that, before the blockade oi
Alesia, it had remained in garrison among the Allobroges or in
'

Italy,' and implies that, although Caesar employed it in the blockade,

he, for some reason or other, omitted to mention it. The redistri])u- '

tion,' he proceeds, after the siege of Uxellodunum gives also the

'

same result, for in book viii, c. 46 the " Commentaries " give the
position of ten legions, without reckoning the 15th, which, according
to book viii, c. 24, had been sent to Cisalpine Gaul. These facts are
repeated again, book viii, c. 54.'
The facts would prove, not necessarily that Caesar had clever
legions at Alesia, but that he had eleven legions to dispose of wher
he distributed the troops in winter- quarters after the capture of the
—
town, if it were certain that the 6th was not one of the ten whicl
he had commanded since the beginning of 53 b. c. Napoleon takes
this for granted, because the legion which Caesar borrowed froii
Pompey was numbered I, and the remaining nine were undoubtedly
the 7th, 8th, 9th, 10th, 11th, 12th, 13th, 14th, and 15th.4 But P. Groebe^
has argued that the legion which, when it formed part of Pompey's
army, was numbered I was, so long as it remained under Caesar,
known as the 6th. For, he asks, if its number in Caesar's army was
I, why did he not say when and where the 6th was raised, as he did
in the case of every other legion ? And if the 6th was raised by him,
and not lent by Pompey, why was it not numbered consecutively,
as all the others were,^ and called the 16th ? It cannot, as some
writers have suggested, have been the legion named Alauda, which
Caesar raised in Transalpine Gaul for it, as we learn from various
;
'^

inscriptions,^ was numbered V. But why, it may be asked, was

Alauda numbered V and not XVI ? Because, says Groebe, whereas
Caesar numbered the legions which he himself created, in ascending

1 B. G., vi, 44, § 3 ; vii, 34, § 2 ; 57, § 1.

2 Hist, de Jules Cesar, ii, 292-3. ^ B. G., viii, 4, § 3.
* 76., ii, 23 ; vi, 32, § 5 ; vii, 51, § 2 ; viii, 54, § 3.
^
W. Drumann, Gesch. Roms, iii, 1906, pp. 706-7.
* Dion Cassius, xxxviii, 47, § 2. ' Suetonius, Divus hdius, 22.

* Gruter, Corpus inscr., 1707, vol. i, cccciii, 1 dxliv, 2 dxlix, 4; dlix, 7. ; ;

 THE NUMBER OF THE LEGIONS 803

order —
from XI to XV —
he did the opposite when he formed
existing cohorts into a legion. Since, then, Alauda was numbered
V, it is evident that a legion numbered VI already existed ; and
this can only have been the one which Caesar borrowed from
Pompey.
Groebe's argument, ingenious as it certainly is, seems open to one
objection :
—
Caesar did not say when or where the 5th legion was
raised ; why, then, should he not have been equally reticent about
the 6th ? But this objection only touches an unessential part of
Groebe's argument. Alauda was certainly not raised before the end
of 52 B.C., and may have been later.^ M. Jullian,^ indeed, although
he confesses that he is attracted by Groebe's hypothesis, doubts
whether Caesar would have ventured to alter the number of a legion
raised by the consul, and suggests that the 6th may have been in-
cluded among the 22 cohorts raised by Lucius Caesar in the latter
part of 52 b.c.^ But has not M. Jullian overlooked one fact by
which Groebe clinches his argument ? If, says Groebe, we compare
B. G., viii, 54, § 3 with B. C, iii, 88, § 2, we shall see (as Drumann
saw many years ago) that the legion which Caesar numbered XV,
and which he was obliged in 50 b. c. to hand over to Pompey, was
numbered III in the army of the latter.* We
are entitled, then, to
suppose that Caesar's 6th legion was Pompey's Ist.^
But, even supposing that an additional legion did join Caesar in 52
B. c. Napoleon is certainly wrong in asserting that it took part in
the blockade of Alesia. For Caesar expressly says that, before the

^ Alauda was raised, according to Prof. W. Ridgeway (Smith's Did.

of Greek
and Roman Ant., i, 96), about 55 B.C. He offers no proof of this assertion, and
no proof exists. Desjardins [Geogr. de la Gaule rom., iii, 48) argues that Alauda
was raised during the Civil War but Groebe gives good reasons for believing
;

that it was one of the unseasoned legions referred to in B. G., viii, 24, § 2, 26,
§ 2, and was raised in 51 b. c. Indeed, the 5th legion, as we learn from the author
of Bellum Africanum (1, § 5) was considered a veteran legion in 46 b. c.
2 Hist, de la Gaule, iii, 396, n. 2 494, n. 4 513, n. 6.
; ;

3 B. G., vii, 65, § 1.

* See also Neue Heidelberger Jahro., iv, 1894,
p. 187, n. 5.
''
A. von Domaszewski {ih., pp. 162-3), who holds that the 6th legion was
distinct from the 1st, remarks that, according to Plutarch {Pomp., 58, § 5) and
Suetonius {Divus lulius, 29), Caesar had ten legions immediately before the
Civil War, and, according to Cicero [Att., vii, 7, § 6), eleven: but, he says, the
difference is only apparent ; for Cicero evidently included Alauda in the eleven,
whereas Plutarch omitted it because until the troops who composed it were
invested with the Roman citizenship it was not officially reckoned as a legion
Now Cicero's letter was written in December, 50 B.C., and some weeks previously
Caesar, in obedience to the Senate, had sent back to Italy the legion which he
had borrowed from Pompey and the 15th, which he had himself raised {B. G., viii,
54). Therefore, if Cicero was correctly informed, Caesar must have had thirteen
legions before he gave up these two. In other words, he must have raised not
only Alauda, but at least one other legion as well. Von Domaszewski, who
apparently accepts this inference, argues that the other legion was numbered
XVI. But that Caesar had thirteen legions in 50 b. c. is utterly incredible.
I can only conclude that Cicero, who, when he wrote the letter to which I have
referred, was at Formiae, on his way back from Cilicia, was unaware or had
forgotten that Caesar had given up two of his legions. If so, Plutarch and
Suetonius were also mistaken.
3f2
804 THE NUMBER OF THE LEGIONS
blockade, it was impossible for reinforcements from Italy or the
Province to join him because the roads were blocked.^ How then
was the additional legion to march by itself all the way from Italy
or from the country of the Allobroges to Alesia ? If it did take part
in the blockade, either it must have arrived in Gaul before the roads
were blocked, and yet after Caesar started for Gergovia or it must ;

have begun its long march after the news of the defeat of Vercin-
getorix in the combat which preceded the blockade reached the
Province and these alternatives are equally improbable.
;

THE OPERATIONS AT ALESIA

L According to the great Napoleon's estimate,^ the entire force
which Caesar commanded at Alesia, including auxiliaries and cavalry,
amounted to 80,000 men. But there are really no data for forming
an estimate. We are not told, and it is useless to guess how many
cavalry he had or how many auxiliaries. We know that he had
10 legions and they must have suffered heavy losses in this and
;

previous campaigns but we do not know how far their losses had
:

been made good by the supplementum which Caesar had brought

into Gaul at the beginning of this year. All told and with the cavalry
and auxiliaries, they could not have numbered less than 40,000 ^ :

but they may have numbered more.

2. The writer of an article in the Neue Jahrhiicher fi'ir Philologie
und Paedagogik ^ complains that Napoleon and von Kampen ^ place
most of the twenty-three redoubts of which Caesar speaks, at the
foot of the hills surrounding Alesia, although Caesar himself says in
chapters 69 and 80 that they were on the hills. Five of the twenty-
three, he remarks, namely those distinguished in Napoleon's and
von Kampen's Plans by the numbers 10, 11, 15, 18, and 22,^have
been discovered (by Colonel Stoffel) ; and four of the five are on the

^ B. O., vii, 65, § 4. ^ Precis des guerres de Cesar, p. 109.

3 See pp. 559-63. « cxx, 1879, p. 107.

^ Quindecim ad Caesaris de h. G. comm. tabulae, xiii.

8 M. G. Fourier {Bull, monumental, 1902, pp. 391-4) refuses to believe that

No. 22, which is situated on a high point of Mont Rea, was really one of the
redoubts ; and even M. Pernet {Pro Alesia, 1907, p. 249), who assisted in
the imperial excavations, queries the identification. M. Fourier observes that the
hill on the north — —
that is to say, Mont Rea was not included within the cir-
cumvallation that the camp of Reginus and Rebilus was unfavourably situated
;

on the slope of the hill {B. G., vii, 83, § 2) ; and that Vercassivellaunus attacked
the camp after having occupied the narrow ridge which commanded the slope.
This last statement, I may remark, is not expressly warranted by the Commen-
taries. M. Fourier concludes that the entrenchment which Stoflfel identified
with a redoubt (No. 22) belonged to a prehistoric earthwork. The redoubt, ho
thinks, was situated on the south-eastern slope, and is probably represented by
'
une figure difficile a expliquer in pi. 28 of Napoleon's Atlas, en haut de la
'
'

lettre D'. But he apparently forgets that the redoubts were constructed at
the outset of the siege that Caesar had intended to include Mont Rea within
;

his circumvallation ; and that he was only prevented from doing so by want
of time.
 — ;

THE OPERATIONS AT ALESIA 805

hills. The inference is that the remaining eighteen were similarly

situated.
Now Caesar does not say, in either of the two chapters to which
the writer refers or anywhere else, that the redoubts were on the
hills. In chapter 69 he says that his camps were constructed in suit-
able positions, and that there (ihi), that is to say in their neigh-
' '

bourhood, twenty-three redoubts were made. In chapter 80 he

speaks of all the camps which covered the heights surrounding
'
'

Alesia. But we have no right to infer from this that the redoubts
were also on the heights. For either the word all is here, as in '
'

other passages, used loosely by Caesar, seeing that three of the camps
were not on the heights at all but in the plain of Les Laumes or ;

else he means all those camps which were on the heights, and not
the others. Besides, one of the redoubts was unquestionably on the
lower slopes of the heights several of them were unquestionably in
;

the plain of Les Laumes ^ and the word ihi is vague.^ The truth is
;

that it is impossible to determine the position of any one of the

redoubts, except the five which have left traces. All that Napoleon
and Stoffel could do or professed to do was to select those sites which
appeared to them the most convenient.
3. The author of the article in the Neue Jahrbilcher points
out^ that, according to von Kampen, the cavalry of Vercingetorix,
after the combat which Caesar describes in chapter 70, fled back
to the town in two divisions, one along the valley of the Oze,
the other along the valley of the Ozerain and he remarks that ;

for this view there is no evidence in Caesar. This is quite true

but unless all the beaten cavalry were nearer to one than to
the other of the two valleys, it is probable that von Kampen is
right.
The writer goes on to argue that the terminus of the flight must
have been the town itself, and not, as von Kampen holds, the Gallic
camp on the east of the town. Only those of the fugitives who could
not get into the town owing to the narrowness of its gates (angu-
stioribus portis) * sought refuge, he maintains, in the camp. I do not

^ Hist, de Jules Cesar, ii, 318 and Planche 25.

^ The MS. reading is castra oportunis locis posita ibique castella XXIII facta.
Commentators have remarked that the number of the camps must have been
originally given, because that of the redoubts {castella) is specified ; that the
fact that the camps were in suitable positions is self-evident; and that, since the
redoubts were not on the sites of the camps, ibi by itself is meaningless. Accord-
ingly, as excavation has proved that there were eight camps, Meusel has
adopted R. Menge's conjecture, castra posita VIII, castellague XXIII facta.
. . .

But A. Klotz {Caesarstudien, p. 260) objects that we should have expected to be

enlightened as to the position of the camps, and that Menge's conjecture leaves
ibique unexplained. He therefore offers an emendation of his own, castra . . .

posita {VIII, quae inter se munitionibus coniungebantur) ibique [that is, in mura-
tionibus] castella XXIII facta. I admit that ibi in the MSS. is extremely vague :

but the alleged superfluousness of the statement that camps were established '

in convenient positions is not removed by inserting their number

'
in v, 9, § 1, :

a similar self-evident fact is recorded in the words loco castris idoneo capto
'
'
;

and the munitio had only just been begun when the camps and redoubts were
made (vii, 69, § 6, with which cf. p. 807)
.^' p. 106. * B. 0., vii, 70, § 3.
806 THE OPERATIONS AT ALESIA
think it necessary to refute this singular argument ^ for to every ;

one except the writer it is clear that angustiorihus portis denotes the
gates of the camp itself. Moreover, there is one sentence in Caesar's
narrative which so unanswerably demonstrates the truth of von
Kampen's view that I am quite unable to conceive how the writer
could have overlooked it. Immediately after saying that the '

Germans hotly pursued them [the Gauls] right up to the entrench-

ments ', Caesar says that some of the fugitives dismounted and
'

tried to cross the ditch and climb over the wall '. Ditch and wall,
Caesar expressly says, were outside the camp. It follows that the
pursuing Germans were outside it too.
4. How did the cavalry of Vercingetorix succeed in getting away
from Alesia ? According to the Due d' Aumale,^ we must assume
either that the line of redoubts constructed by the Romans was
defective or that the vedettes did not do their duty with proper
vigilance. The line of redoubts was certainly defective, in the sense
that it was not completed for Caesar says that the cavalry went
;

out through a gap in the works .^ The only question is whether any-
thing more than redoubts had yet been constructed at all that is ;

to say, whether the line of contravallation, properly so called, had

even been begun. Caesar says that Vercingetorix sent out his cavalr}^
opere instituto,^ that is to say, after the Romans had begun to con-
struct their works, which would seem to imply that they had not
had time to do much indeed a later passage in his narrative would
:

leave on one's mind the impression that the work of constructing the
actual line of contravallation was not taken in hand until after the
departure of the cavalry; for he distinctly says that he undertook this
work after he had learned what Vercingetorix's purpose had been in
sending the cavalry out {quihus rebus cognitis ex perfugis et captivis,
Caesar haec genera munitionis instituit).^ But many commentators
have deceived themselves by arguing on the assumption that
Caesar's narrative, even in matters of chronology, was invariably
and rigidly precise. They seem to forget that he did not write
with the fear of German critics before his mind. It is possible
that he had begun the work of constructing the contravallation
before Vercingetorix sent out his cavalry and that, when he wrote
;

the sentence which I have quoted, he only meant that he modi-

fied and added to his plan, with a view to repelling the expected
army of relief. All that we can say with certainty, then, is that
the line of contravallation, if it had been begun, was so incom-
plete that there was nothing to prevent the departure of the GalHc
cavalry.
It is generally supposed that the gap through which they
escaped was in the western plain but the mere fact that the
;

first combat of cavalry took place there does not prove that

'
I confess that I did so in the first edition (pp. 784-5).
^ Rev. des Deux MoTides, 2^ per., xv, 1858, p. 110.
^ '
qua erat nostrum opus intermissum,' B. 0., vii, 71, § 5.
* lb., 70, § 1. ' lb., 72, § 1.
 THE OPERATIONS AT ALESIA 807

the works weremore incomplete than anywhere else. It seems to

me more probable that the cavalry moved uj) the valleys of
the Oze and the Ozerain first, because the bulk of Caesar's
;

cavalry were stationed in the plain, and, secondly, because the

map shows that if the Gauls had moved down the valleys, they
would have been exposed to attack for a far longer time than if
they had moved eastward.

M. Salomon Reinach, remarking that 500 of Vercingetorix's

cavalry w^ould have sufficed to summon reinforcements, and that
a modern general in like circumstances would have kept the horses
of the rest for food, argues that all were sent out because the Gauls
considered it impious to eat horse-flesh. La viande de cheval,' he
'

says, parait avoir ete tahou chez toutes les populations de la langue
'

aryenne.' ^ It was, however, eaten in Britain in the Bronze Age.^

Is it not likely that Vercingetorix sent out all his cavalry because,
while he could not afford to feed them, they would reinforce the
cavalry of the expected army of relief ? Besides, taboo is disregarded
by starving men.
5. It is necessary to inquire how far the results of Stoffel's ex-
cavations ^ harmonize with Caesar's statements. In the course of
the excavations certain trenches, were found which did not, at first
sight, seem to correspond to anything in Caesar's description.
M. V. Pernet, who assisted the excavators, and has very minutely
described the results of their labours, believes that the original
earthwork (ynunitio) which Caesar mentions in B. G., vii, 69, § 6 is
represented by four incomplete trenches in the plain of Les Laumes,
on the slope of the hill of Flavigny, on the slope of the hill of Bussy,
and along the foot of Mont Rea, and that these unfinished works
were abandoned after the departure of Vercingetorix's cavalry, when
Caesar took in hand the construction of the lines which he describes
in chapters 72-4.*
Caesar's narrative would, I think, leave upon the minds of most
readers the impression that all the works therein described the —
inner 20-foot trench, the two 15-foot trenches, and the rampart of
—
the contra vallation completely surrounded Alesia. He does not,
however, distinctly say this and the results of the excavations
;

show that it was not the case. Nor, indeed, was such an elaborate
system of works required for, as Napoleon remarks,^ the
;

onl}'" sections of the contravallation which required to be forti-

fied with' special care were those which crossed the plain of

^ Rice Holmes, Anc. Britain,

"
^ Rev. xxvii, 1906, pp. 1-15.
celt., p. 152.
* Some of the excavations liad been completed under the direction of members
of the Commission de la topographie des Gaules before Stoffel took charge of the
work, and the results were verified by him {Pro Alesia, 1907, pp. 122, 207).
Stoffel, however, if A, Bertrand is to be believed, directed the excavations d'un '

peu de loin, prcfcrant a la plaine des Laumes le sejour de Dijon ou de Scmur.'

See Rev. arch, 4** scr,, xiv, 1909, p. 141.
« Pro Alesia, 1907, pp. 158, 248-9, 279-80.
^ Hist, de Jules Cesar, ii, 319-21.
808 THE OPERATIONS AT ALESIA
Les Laumes and the valley of the Rabutin. The 20-foot trench ^
extended only across the plain of Les Laumes. Of the two 15-foot
trenches the inner surrounded Alesia, but the outer was confined
to the plain. The circumvallation had nowhere more than one
trench.
Achaintre ^ puts the question whether the 20-foot trench formed
a part of the completed system of works, or whether it was merely
a fosse perdu ', abandoned after it had served the temporary pur-
'

pose of protecting the legionaries while they were constructing the

proper works. The former view is defended by Lipsius, Folard, and
Turpin de Crisse. Guischard, however, argues that if this trench
had formed a part of the contravallation, properly so called, there
must have been another like it in the circumvallation, as we may
infer from Caesar's saying that the contravallation and the circum-
vallation were identical in plan.^ Moreover, as Achaintre justly
remarks, Caesar does not say one word to suggest that the
20-foot trench did form a part of the contravallation and the ;

passage in which he describes the attempt which the Gauls made

to cross it while the relieving army was making its first attack
upon the circumvallation in the plain,^ proves that the Romans had
abandoned it.

The distance between the 20-foot trench and the inner two of the
trenches in the line of contravallation was, according to the MSS.,
400 But,' says von Kampen,^ in place of reading fedes
feet.^
—
' '

CCCG, must be passus CCCC

it
'
nearly 2,000 feet — as a simple '

measurement of the space between the trenches traces of which —

still exist
—
gives this distance.' Napoleon contradicting the evidence
supplied by his own Atlas (Planche 25), says 400 feet ' and ' ;
'

Mr. W. C. Compton^ tries to explain the difficulty by saying that

'
at either end the 20-f eet ditch is no more than 400 feet in front of
the other fortifications '. But this will not do tested by the scale. :

Napoleon's Plan makes the distance afc the northern end much
more than 400 feet. The Greek paraphrast wrote rpia o-raSia, which
is equivalent to 375 passus;^ and perhaps we may infer that the
reading in his copy of Caesar was passus CCCC. Following Napoleon's
Plan, I have given approximately the distance as measured, accord-
ing to the scale, from the southern end of the innermost trench to
the contravallation.

^ The dimensions of this trench do not correspond precisely with those given
by Caesar. Pro Alesia, 1908, p. 348.
2Caesar, i, 375-6.
3pares eiusdem generis munitiones, diversas ab his, contra exteriorem hostem
perfecit. B. G., vii, 74, § 1.
* Ih., 79, § 4. ' lb.,
^
72, § 2.
^ Quindecim ad Caesaris de h. 0. Comm. tabulae, xiii.
' Hist, de Jules Cesar,
ii, 303.
Caesaris Seventh Campaiyn in Gaul, p. 105.
^

^ The Greek translation is said to have been made from Stephanus's edition of

1544 {Berl. phil. Woch., 1901, col. 41) ; but Stephanus reads pedes.
 THE OPERATIONS AT ALESIA 809

Excavation has shown that on the hills round Alesia the pits
which Caesar describes in B. G.^ vii, 73, §§ 5-8 (with which
(scrohes)

••••••
compare 74, § 1) were not in front but in rear of the circum valla tion,i
which was broken opposite each pair.
•

According to Caesar,^ the perimeter of the circumvallation was

14 Roman miles, or about 21 kilometres according to Stoffel's
;

interpretation of the results of the excavations,^ only 18 kilometres.

The tracing of the circumvallation in Napoleon's map* is, how-
ever, partly conjectural ; and Commandant Colin ^ gives reasons
for believing that the extent may actually have equalled Caesar's
estimate. On the whole, Caesar's description of the fortifications,
from which one cannot expect the rigorous accuracy of an en-
gineer writing in a professional journal, is subject only to trivial
corrections.
6. Guischard ^ holds the singular opinion that there was only one
trench in the contra vallation besides the 20-foot trench ; and that
the other of the two trenches which Caesar expressly mentions was
in the circumvallation. This view is refuted by the results of the
excavations but it may be well to show that they agree with Caesar's
;

narrative. Guischard argues that/os5a interior'^ means the trench

in the contra vallation, and /os5a exterior^ the trench in the circum-
vallation. But, as L. V. Berlinghieri ^ points out, after having men-
tioned these two trenches, Caesar goes on to say that behind them
he constructed a rampart and it is clear that this rampart belonged
;

to the contra vallation. 10 Guischard insists that when Caesar, after

describing the 20-foot trench, says that he traced all the other works
at a distance of 400 feet behind it {reliquas omnes munitiones ah ea
fossa pedes CCCC reduxit), he means by reliquas omnes munitiones
both the contra vallation and the circumvallation but this view is
;

refuted by the very fact that the reliquae omnes munitiones were
separated by the distance of 400 feet (or paces ') from the 20-foot
'

trench. If the reliquae omnes munitiones had comprised both the

contra vallation and the circumvallation, Caesar would not have said
that they were both separated by the same distance from the 20-foot
trench. It is clear, then, that reliquas omnes munitiones refers only
to the contravallation. Finally, after describing the construction of
the reliquae omnes munitiones, which included the two trenches
above mentioned, and of the subsidiary defences (lilia, cippi, and
stimuli), Caesar proceeds to describe the construction of the circum-

1 Pro Alesia, 1908, p. 418. 2 B. G., vii,

74, § 1.
^ Pro Alesia, 1908, p. 351. Napoleon's estimate {Hist, de Jules Cisar, ii, 322)
is '
about 20 kilometres '.
* PI. 25. 5 Pro Alesia,
1908, p. 351.
6 Mhn. hist, 1774, pp. 499-501.
crit. et ' B. G., vii, 72,
§ 3.
^ The term
Guischard' s own. Caesar does not mention the fossa exterior,
is
though of course he implies its existence.
* Examen des operations et des travaux de Cesar autour d'' Alesia, 1812,
pp. 77-84.
^° Hoc intermisso spatio duas
fossas XV
pedes latas, oadem altitudine perduxit,
(iuarum interiorem campestribus ac demissis locis aqua ex flumine derivata
complovit. Post eas aggerem ac vallum XII pedum extruxit. B. G., vii, 72, § 3.
810 THE OPERATIONS AT ALESIA
vallation. '
When
these defences were completed,' he says, Caesar '

constructed corresponding works of the same kind, facing the

. . .

opposite way, to repel the enemy from without (His rebus ferfectis '

. pares eiusdem generis munitiones, diversas ah his, contra exteriorem

. .

hostem perfecit.'^
Napoleon 2 represents the towers as erected on that part of
7.
the contra vallation which crossed the western plain, clearly imply- —
ing that they were not erected on any other part and he says ;

that the works which Caesar describes in chapters 72 and 73 were

'
peculiar to the plain of Les Laumes '. Both these statements are
incorrect and they are contradicted by Napoleon himself, by his
;

Plan (25), and by the Commentaries. Caesar tells us that there

were lilia and cippi, as he calls the subsidiary defences which
I have described on page 172, in front of the camp on the southern
slopes of Mont Pea ^ and he mentions the to-wers that defended
;

the loca praerupta, which are identified with the Montague de

Flavigny.*
8. holds that the agger or rampart in the line of
Guischard ^

contravallation, which Caesar describes in chapter 72, was protected

by a palisade (vallum), planted on the berme, as well as by the crenel-
lated parapet (lorica pinnaeque) which stood upon the rampart
itself. Caesar describes the rampart by the words aggerem ac vallum,
which show that there was a palisade somewhere but there is no :

evidence that it was planted on the berme and its proper place was
;

the top of the rampart.^

9. Guischard also holds that the towers which C^aesar erected at
intervals of 80 feet (?)'^ along the line of contravallation were mounds,
projecting in front of the curtain and protected by parapets ^ but ;

this absurd theory is refuted by the passages in which Caesar de-

scribes the construction of the wooden towers which Cicero erected
upon the rampart of his winter camp in 54 B. c.^
10. Describing the subsidiary defences which he calls cippi,

1
B.G.,
vii, 72-4. ^ jji^i ^g Jules Cesar, ii, 303, 312, 319-22.
The
writer of the article in the Neue Jahrhiicher (cxx, 1879, pp. 175-6) argues
'^

that Napoleon was wrong in identifying the hill on the north of Alesia, on the
southern slopes of which Reginus and Rebilus encamped, with Mont Rea. I
refuted his arguments in my first edition (pp. 792-4) ; but, considering that it
was at the spot which Napoleon identified with the scene of Vercassivellaunus's
attack that the whole of the Gallic coins and the great majority of the Gallic
weapons which the excavators unearthed were found, I think that it would be
superfluous to reprint the paragraph.
* B. G., vii, 73,
§§ 4-8 85, § 6 ; 86, § 5. ;

^ 31 em. mil.,
pp. 230-1.
« Gaesar, says A. E. Masquelez {Sped, mil., 2''- ser., xliii, 1863, pp. 354-5),

frequently uses the expression viminea lorica and we may therefore conclude
;

that the lorica was made of wattle-work. The conclusion may be right but ;

Caesar never uses the expression viminea lorica.

' A. Klotz {Caesar studien,
p. 219) thinks that the towers must have stood
much further apart than 80 feet, especially as artillery was used {B. G., vii, 81,
§ 5), and accordingly conjectures that for pedes we should read passus. If, as we
may suppose, the pieces of artillery were in the towers, he is probably right.
Cf. p. 808.
« Mem. mil, pp. 230-1. ' B. G., v. 40, § 6.
 —

THE OPERATIONS AT ALESIA 811

Caesar writes Itaque truncis arborum aut admodum jirmis ramis

:

ahscisis atque horum

delihratis ac praeacutis cacuminibus, perpetuae
fossae quinos pedes altae ducebantur. Hue illi stipites demissi et ab
infimo revincti, ne revelli possent, ab ramis eminebant. Quini erant
or dines coniuncti inter se atque implicati ; quo qui intra-
verant, se ipsi acutissimis vallis induebant^ ('Accordingly trees or
very stout branches were cut down and their ends stripped of their
bark, and sharpened to a point continuous trenches were then dug,
;

5 feet deep, in which the logs were planted and fastened down at
the bottom to prevent their being dragged out, while the boughs
projected above. There were five rows in each trench [?], connected
with one another and interlaced ; and all who stepped in would
impale themselves on the sharp stakes '). Does quini implicati . . .

mean that there were five parallel trenches, or five rows of boughs in
each trench ? Kraner,^ von Goler,^ and Long suggest the latter •*

interpretation and it seems the more probable, because the passage

;

immediately follows that in which Caesar describes how the boughs

were inseited in the trenches ^ if he had meant that there were five
:

per petuae fossae, he would naturally have said so when he first men-
tioned them, and by planting five rows of boughs in each trench
labour would have been saved. Napoleon ^ adopts the other inter-
pretation, but gives no reasons.
Berlinghieri,'^ however, does give a reason. He says that if each
of the trenches had been wide enough to contain five rows of boughs,
Caesar would have mentioned their breadth. This is no argument.
The fact that each trench contained five rows would of itself prove
that the breadth of the trenches was considerable. P. Bial ^ holds
with Berlinghieri that if Caesar had meant to convey that there
was only one trench, he would have written perpetua fossa, not
perpetuae fossae. I have not tried to prove that there was only
one trench, but that Caesar does not say that there were five.
But in any case Caesar would not have written perpetua fossa ;

for it would be ridiculous to imagine that he meant to de-

scribe an endless trench, surrounding Alesia like a ring. What
he meant was that, where the trenches were necessary, they were
continuous.^
11. Some commentators have maintained that the line of
circumvallation, as well as the line of contravallation, included
a fosse perdu ', that is to say a 20-foot trench similar to
'

the one which Caesar describes in his 72nd chapter. I suppose

^ B. 0., vii, 73, §§ 2-5.

2 Caesar, ed. Dittenberger-Kraiier, 1890, p. 326.
^ Gall Krieg, 1880, p. 311.
* Caesar, p. 390. ^ See Schneider, ii, 571.

« Hist, de Jules Cesar, ii, 303, and Planche 27. Cf. Pro Alesia, 1907, pp. 239-40.
' Examen, &c., p. 89.
^ Chemins, habitations et oppidum de la Gaule an temps de Cesar, 1864,
p. 210, n. 3.
" The writer of the article in the Neue Jahrbilcher (cxx, 1879,
pp. 117-8) gives
a fantastic explanation of the quini ordines, which I needlessly refuted in the
first edition (pp. 789-90).
812 THE OPERATIONS AT ALESIA
they thought that it was necessary in order to protect the soldiers
while they were at work from the attacks of the unarmed rustics
in the neighbourhood. Guischard ^ has taken the trouble to refute
this absurdity.
12. pages 175-6 of my narrative I have written, It was just
On '

after the expulsion of the Mandubii when the anxious watchers on

the hill saw, moving over the plain, a multitude of cavalry. The
infantry were on the heights of Mussy-la-Fosse behind.' Caesar
does not fix the date of the arrival of the relieving army at Alesia.
After describing the fate of the Mandubii, he writes,^ interea Com-
mius reliquique duces ad Alesiam perveniunt, &c. I should think
. . .

that interea is used loosely here, as meanwhile often is in Eng-

' '

lish. If the relieving army had already arrived, we may perhaps

presume that the Mandubii would not have been expelled from
the town. [Since I wrote these words, I have come across three
passages in which Virgil,^ as Nettleship remarks, uses interea in
a loose sense.]
13. Caesar says that the relieving army encamped not more
than one Koman mile from the circumvallation, on a hill outside
Alesia.* The hill is generally identified with Mussy-la-Fosse,
which is south-west of Mont Auxois, and beyond the plain of Les
Laumes.
Mr. Compton ^ gives reasons for placing the camp on
the slope of
this hill, facing Alesia, not, as von Kampen does, on the plateau.
He argues that first, the view towards the besieged town is no more
'

comprehensive from the top of the hill than from the side secondly, ;

a hill, so high and so far removed from the scene of action would be
quite unsuitable for military purposes, especially for cavalry ; thirdly,
Caesar expressly gives a mile from the Roman lines as the distance at
which the Gauls encamped, and i\iQ foot of the hill at the nearest
point is about a mile from the Roman works fourthly, the area
;

enclosed by a line of earthworks in von Kampen's map appears out

of proportion to the needs even of so large a force as that of the
text ; fifthly, the Roman cavalry pursue the Gauls " usque ad
castra "
'
—
right up to the camp
—
' which would be practically
impossible if the camp were on the top of so steep a hill.' There can
be no doubt that, if the Gauls encamped on the hill of Mussy at all,
they encamped, partly at all events, on its slope but any one may
;

convince himself, by merely comparing the area of the plateau of

Mussy with the area of the plateau of Mont Auxois and reflecting
that the relieving army was three times as numerous as the tightly
packed garrison of Alesia, that von Kampen's camp is not at all
too large.
Von Goler^ objects to the hill of Mussy-Ia-Fosse altogether, and
^ Mem. crit. et hist., 1774, p. 504. « B. G., vii, 79, § 1.

^ Aen., X, 1 ; xi, 1 ; xii, 842.

* Colle exteriore occupato non longius mille passibus ab nostris munitionibus

considunt. B. G., vii, 79, § 1.

^ Caesar s Seventh Campaign in Gaul,
pp. 107-8.
" Gall. Krieg, 1880,
p. 316.
 —

THE OPERATIONS AT ALESIA 813

places the encampment on the hill south of Pouillenay, which is

south-west of the Montague de Flavigny. He holds that, as this hill
abounds in springs and is situated near streamlets which flow into
the Brenne, the Gauls would have been able to fetch water with less
danger than they would have incurred if they had encamped on the
hill of Mussy. He also lays stress on Caesar's use of the singular colle,
arguing that, according to the received view, the Gauls would have
occupied not one hill but two.
Von Goler's arguments are not convincing. Streamlets, from
which the Gauls could have fetched water without risk, flow into the
Brenne near Mussy as well as near Pouillenay Caesar might, I think, :

have used the singular colle loosely ^ and Pouillenay is more than
;

a mile from the nearest point of the circumvallation. I do not

think that any one who had seen Mont Auxois would have any
hesitation in rejecting von Goler's theory.
14. The Romans, in repelling the first assault of the circumvalla-
tion, drove back the Gauls, fundis lihrilibus (sudibusque, quas in
opere disposuerant, ac glandibus),^ which I have translated by ' with
slings throwing large stones (and sharp stakes ', &c.). The meaning,
however, depends on the punctuation if we put a comma after
:

fundis, as Meusel does in the reissue of his school edition, it would

be ' with slings, large stones (and sharp stakes ', &c.). Forcellini^
holds that fundae lihriles were slings that threw stones weighing
a libra, or Roman pound (rather less than 12 ounces) each and ;

stones weighing a mina (about 15 ounces) each were slung by Balearic

slingers,* If lihrilibus is to be taken as a substantive, it may perhaps
be interpreted, after Festus,^ as ' stones of the thickness of a man's
arm ', which were thrown by hand.^
15. Describing the sortie which the besieged attempted while the
relieving army was making its first assault upon the circumvalla-
tion, Caesar writes At interiores, dum ea quae a Vercingetorige ad
:

erwptionem praeparata erant proferunt, priores fossas explent, diutius

in his rebus administrandis morati prius suos discessisse cognoverunt
quam munitionibus adpropinquarent'^ (' The besieged lost much time
in bringing out the implements which Vercingetorix had prepared for
the sortie, and in filling up the front [?] trenches and finding, before ;

they could approach the contra valla tion, that their comrades had
withdrawn,' &c.). The question is what Caesar meant by priores.
Most commentators make priores agree with fossas ; but while
Napoleon^ understands by priores fossas merely the 20-foot trench,
which was nearest to the besieged, Dittenberger beHeves that Caesar
was speaking both of it and of the nearer of the two parallel 15-foot
^ M. Jullian, I find, remarks Vercingetorix, p. 388, n. 2) that '
en r^alite ces
(

deux collines ne sont que deux branches d'un meme massif.

2 B. C, vii, 81, § 4.
^ Totius latinitatis lex., iii, 1865, p. 755. Cf. Daremberg and Saglio, Did. des
ant. grecques et rom., ii, 1364.
* Diodorus Siculus, xix, 109, § 2.
^ De verhorum significatione, ed. C. 0. Miiller, 1839, p. 116.
8 Cf. Vegetius, ii, 23, on libralia saxa. ' B. G., vii,
82, §§ 3-4.
^ Hist, de Jules Cesar, ii, 309.
814 THE OPERATIONS AT ALESTA
trenches. It has been objected to Napoleon's interpretation that
Caesar would not have used the plural/ossas to describe one trench ;

but I think that he might have used fossas in the sense of various
parts of one trench, just as he often uses ripas and ripis in the sense
of various parts of one bank.i Schneider,^ however, thinks it un-
likely that Caesar would have used the comparative when he meant
the superlative ; and it may be added that if he had meant the
trench nearest to the besieged, he would have written proximam
fossam, as he did in chapter 79. To Dittenberger's view there are
three objections. Is it likely that Caesar would have described two
out of three trenches, the nearest of which to the besieged was about
a quarter of a mile from the second, while the second was in juxta-
position with the third, as priores fossas ? What right has Ditten-
berger to assume that the besieged successfully crossed not only the
nearest trench but also the elaborate system of subsidiary defences,
which intervened between the nearest trench and the second, when
Caesar's narrative clearly implies that they did nothing of the kind.
And if they had reached the inner 15-foot trench, which was quite
close to the rampart of the contra vallation, how could Caesar have
said that they never got near the contra vallation ? Moreover, if
priores agrees with, fossas, the omission of et before priores is remark-
able. For these reasons I was once tempted to adopt the view of
Schneider ^ and von Goler ^ that priores does not agree with fossas,
but that it is in the nominative case and denotes the front ranks of
the besieged, who tried to fill up the nearest trench, while the rest
were engaged in bringing up the various implements that were
required for the proposed assault on the contravallation. But it
seems unlikely that Caesar should have drawn a distinction between
the front ranks of the besieged and those behind and there is a very
;

strong reason, besides the one which I have already given, for believ-
ing thiit fossas cannot mean the nearest trench ', or at all events
'

cannot mean it alone.^ The besieged, as we learn from chapter 79

(itaque productis copiis .proximam fossayii cratibus mtegunt atque
. .

aggere explent) had already filled up the nearest trench ^ and even ;

if the Komans removed the fascines, the besieged could soon have

filled it up again. For it must be remembered that there was no need

to fill it along its whole extent all that was necessary was to make
:

causeways here and there. But when the columns, after crossing
the causeways, had deployed into line, and were about to attack the
contravallation, they were of course obliged to fill up the whole of
the trenches. I believe therefore that priores fossas means the small
subsidiary trenches which the Romans called cippi. Guischard is
surely wrong in thinking that it also means the inner of the two
15-foot trenches;' for, I repeat, if the Gauls had begun to fill up

^ Cf. Meusel's Lex. Caes., ii, 1750, and his edition of the Civil War, p. 220.
2 Caesar, ii, 616. ^ lb., ii, 615-6.

* Gall. Krieg, 1880, p. 319, n. 3.

^ I failed to think of this reason before the publication of the first edition.
^ See Guischard, Metn. crit. et hist., 1774, pp. 497-8.
' Cf. C. Jullian, Hist, de la Gaule, iii, 522, n. 10, 523, n. 1.
 —

THE OPERATIONS AT ALESIA 815

this ditch, which formed part of the contravallation, properly

so called, how could Caesar have said that they did not approach
the contravallation ?

16. In the account of the sortie which Vercingetorix made

on the day of Vercassivellaunus's attack we read that Ver-
cingetorix marched out of the town and took out of his camp
the sappers' huts and other material which he required for en-
deavouring to force the Roman lines {Vercingetorix ex arce Alesiae
suos conspicatus ex oppido egreditur castris [a castris /5] longurios,
musculos, falces reliquaque quae eruptionis causa paraverat profert)}-
Now many commentators, assuming the identity of Mont Rea with
the northern hill, and assuming that Vercingetorix directed his
sortie against the works in the plain of Les Laumes, have urged that
he would not have marched out of Alesia through his camp, which
was on its eastern side, and have therefore altered castris (out of
camp) into crates (fascines). But, as Long ^ has pointed out with
his usual acumen, the alteration is unnecessary. Vercingetorix,' he
'

says, had a camp (c. 70), and though it was at the east end of the
'

hill, it was below it, and on such a level that it was much more easy

to carry such things as " musculi " from his camp to the lines on
the west side of the town .than to bring them from the high
. .

plateau of Alesia down its steep sides. Any man who has seen the
ground and read Caesar's text with care will reject the emendation
crates? There would be no use in having all these cumbrous things
on the top of the hill in the town.'
17. Maissiat* infers from a passage in Polyaenus ^ that some
Callic traitor informed Caesar of the intended attack of Vercassi-
vellaunus on the camp at Mont Rea. I do not believe this for ;

if Caesar had been warned, he would surely have reinforced the

threatened camp before the attack commenced.^

18. The statement on p. 178 of my narrative
—
The attack on'

the circumvallation in the plain was comparatively feeble for the ;

bulk of the relieving force was formidable only in numbers is a fair

'
—
inference from Caesar's narrative. 60,000 picked men were sent to
attack the camp on Mont Rea. According to the enumeration given

1 B.G., vii, 84, § 1. Cf. p. 608, n. 10. ^ Caesar,

pp. 402-3.
^ H. Schiller {tJher Entstehung. d. corpus Caes., 1899, p. 25) gives a reason
. .

for accepting crates which is in my opinion inconclusive. Remarking that the

reading of a is castris and oi $ a castris, he explains that in certain various
readings a is really a misunderstood abbreviation of aliter (otherwise) and he
;

—
shows that this abbreviation accounts for the at first sight inexplicable
reading auxilio dunum for Uxellodunum in viii, 32, § 2, auxilio dunum.
having been evolved from a. (= aliter) uxillo dunum. But a does not always
stand for aliter and to suppose that a castris grew out of a. cratis is surelj'
;

gratuitous.
* Jules Cesar en Gaule, iii, 135, n. 1. ^ viii,
23, § 11.
^ M. Jullian {Hist, de la Gaule, iii, 530, n. 5) wrongly, as I believe, supposes

that Polyaenus was thinking of the pursuit of the Gallic fugitives which began
in the middle of the night that followed the final struggle at Alesia {B. G., vii, 88,
§ 7). Polyaenus's statement is, in my opinion, so utterly worthless as evidence
that I do not even reproduce it ; but any one who reads it will see that it is
irreconcilable with Caesar's narrative. Cf. Pro Alesia, 1909, pp. 531-2.
816 THE OPERATIONS AT ALESIA
in chapter 76, 170,000 or 180,000, allowing for the numbers killed
and wounded on previous days, would have been available for
attacking the circumvallation in the plain. But Caesar does not
lay any stress upon this attack he merely notifies it while he does
: ;

lay great stress upon the efforts of the 60,000 picked men and of the
besieged.^ Indeed, although he implies that the circumvallation in
the plain was attacked, he does not expressly say so he merely says :

that the cavalry began to move towards the entrenchments in the

'

plain, while the rest of the host made a demonstration in front of

their camp '. We are perhaps justified in inferring that it was attacked,
because Caesar implies that both on the circumvallation and on the
contra vallation the Romans were distracted by the din of battle in
their rear ^ but possibly in speaking of the circumvallation, he
;

may only have been referring to the attack on the camp of Rebilus.^
A. Reville * and the Due d'Aumale ^ make some remarks upon the
matter, which perhaps infer too much, but are worth quoting.
Reville thinks that the efficiency of the relieving army was paralysed
by I'interet oligarchique et particulariste des nobles '. The nobles,
'

he goes on to say, feared the triumph of Vercingetorix as much as

the victory of the Romans. All this, indeed, is pure conjecture but ;

it is well-founded conjecture. The Due d'Aumale suggests that all

the Gauls of the relieving army, except the 60,000 picked men,
were a confused ill-armed mob and that the Aedui were prob-
;

ably half-hearted, and had perhaps, after the rebuff which they
received from the Pan-Gallic council, when they claimed the
right of directing the campaign, made secret overtures of sub-
mission to Caesar. Considering what Caesar says about their
disgust when they found that they were not to have the direction
of the campaign, and their regret at having flung away their
alliance with Rome,^ we may, I think, assume that they were
half-hearted.
19. Caesar says that on the day of the final attack he occupied
'
a suitable position ', which enabled him to note all the phases of
the action (Caesar idoneum locum nactus, quid quaque in parte geratur
cognoscit)? The most, indeed the only suitable position is on the ' '

Montague de Flavigny. (See Napoleon's Histoire de Jules Cesar,

ii, 311.) M. du Mesnil, chef d'escadron d'etat-major,' decides for
'

the hill near Gresigny, which is on the east of Mont Rea, because it '

was Caesar's interest to keep as close as possible to the point where

danger was imminent.' ^ But, as I shall show that the steep places '

^ B. G., vii, 83, § 8 ; 85-8 ; and Napoleon's Hist, de Jules Cesar, ii, 310, n. 2.
^Multum ad terrendos nostros valet clamor, qui post tergum pugnantibus
extitit, quod suum periculum in aliena vident salute constare. B. G., vii, 84, § 4.
3 M. V. Fernet (Pro Alesia, 1908,
pp. 419-20; 1909, pp. 580-1) conjectures
that the relieving army attacked not only in the plain and at Mont Rea, but
also on the Montague de Bussy, near the redoubt numbered 15, and on the
Montague de Flavigny, near camp A, where the finds lead him to believe that
there was fighting.
* Bev. des Deux Mondes, S*' per., xxiii, 1877, pp. 67-8.
5 /&., 2« per., XV, 1858, p. 116. " B. G., vii, 63, § 8.
W6., 85, § 1. « S'pect. mil, xxvii, 1839, p. 630.
 THE OPERATIONS AT ALESIA 817

{loca praerupta), to securewhich Caesar marched in person after he

had visited the works in the plain of Les Laumes, and from which
he went to the relief of Labienus, may probably be identified with
the Montagne de Flavigny, on du Mesnil's theory Caesar, before he
relieved Labienus, would have had to make a long march from the
north of Alesia, right across the plain of Les Laumes, to the south,
and back again to Mont Rea. See page 179 of my narrative.
20. Describing the final attempt of the besieged to break through
the contra vallation, Caesar says that, abandoning the hope of forcing
the formidable works in the plain, they attempted to storm the lines
at a point where the ascent was precipitous (interiores desperatis
campestribus locis propter magnitudinem munitioniim loca praerupta
\ex\ ascensu temptant, &c.).^ The Due d'Aumale, who wrote before
the appearance of Napoleon's book, considered that the loca prae-
rupta which the besieged attacked were Mont Pevenel and the plateau
of Sauvigny, in other words the Montagne de Bussy, and that, after
Caesar had beaten them, he rejoined Labienus by crossing the valley
of the Rabutin.2 Napoleon particularizes the loca praerupta as the '

works situated at the foot of the precipitous heights of the mountain

of Flavigny '.^ The description loca praerupta hardly applies to
Mont Pevenel or to the plateau of Sauvigny. Besides, the loca prae-
rupta were identical with or just above the steep inclines (declivia
et devexa) which Caesar mentions in chapter 88 the declivia et devexa
:

were visible from the camp on Mont Rea and the slopes of Mont
;

Pevenel were not. Heller, indeed, argues that Caesar must have
referred to Mont Pevenel, because there alone the line of contra-
'

vallation ' crossed the higher slopes.'* But, as Mr. Compton remarks,
'
the besieged might be said " temptare " these heights, if they
assailed the lines, the crossing of which would lead to them ^ '
;

and M. Jullian justly observes that if Vercingetorix had selected

^

any other point for attack, he would have lost precious time. The
plateau of Sauvigny was not crossed by the line of contra vallation ;
and therefore it is open to the same objection as the Montagne de
Flavigny. Mr. Froude describes the final attack of the besieged as
having taken place on the north side of Alesia.^ Caesar,' he says,
—
'

'
saw the peril '
—
of the camp on Mont Rea and sent Labienus '

with six cohorts to their help. Vercingetorix had seen it also, and
attacked the interior lines at the same spot. Decimus Brutus was
then despatched also, and then Caius Fabius.' No one who had
studied Caesar's narrative with close attention could have written
the last two sentences. The interior lines at the same spot could
'
'

not possibly be described as loca praerupta. Mr. Froude goes on to

say that Finally, when the fighting grew desperate, he [Caesar] left
'

his own station and he rode across the field, conspicuous ', &c.
. . .

But what field did he ride across ? If Mr. Froude is right in saying

1 B.G., vii, 86, § 4. Rev. des Deux Mondes, 2« per., xv, 1858, p. 139.
«
'^
Hist, de Jules Cesar, 311-2.
ii, * Philologies, xiii,
1858, p. 599.
^ Caesar s Seventh Campaign in Gaul, p. 111.
'Hist, de la Gaule, iii, 526, n. 7. ' Caesar : a Sketch, ed. 1886, p. 369.
1093 3 G
 818 THE OPERATIONS AT ALESIA
that Vercingetorix attacked the interior lines at the same spot ',
'

there was no field For the interior lines at the same

to ride across ! '

spot can only mean the interior lines nearest to the foot of Mont
'

Rea whatever the spot was against which the besieged directed
:
'
'

their final attack, Caesar, as he himself tells us, went to rescue its
defenders and, when he had done so, he rode to succour Labienus.
;

'
That is to say, according to Mr. Froude, he rode across the field '

from the interior lines at the same spot to the same spot
'
'
'
' !

Besides, how would Mr. Froude explain the words de locis superiori- '

hus (that is to say, from the slopes, occupied by the Gauls, above
Labienus's position) haec declivia et devexa cernehantur '. Haec declivia
et devexa cannot apply to what Mr. Froude speaks of as the interior '

lines at the same spot ', —

that is, be it remembered, close to the foot
of Mont Rea. But if they are understood as applying to the slopes
of Flavigny on the opposite, or southern, side of t^e plain, from which
Caesar could be seen descending, the narrative is perfectly in-
telligible, and every difficulty disappears.
21. Describing the final movement which he made to succour
Labienus, Caesar writes, equitum partem se sequi, partem circumire
exteriores munitiones et a tergo Jiostes adoriri iuhet.^ Napoleon holds
that the part which was to ride round the circumvallation (exteriores
munitiones) came, not from the Montague de Flavigny, where Caesar
was, but from Gresigny for the results of Stoffel's excavations
;

show that three of Caesar's cavalry camps were in the western plain,
and one in the valley of the Rabutin, near Gresigny .^ Apart from
this consideration, it seems very unlikely that the cavalry which was
to ride round the outer lines left the Montague de Flavigny and went
round to Mont Rea by the west for in that case they would have
;

had to encounter a large part of the relieving army, both cavalry

and infantry, on the way.^ Nor could they, as Long supposed,^ have
gone round Alesia from the Montague de Flavigny by the east
either, because in that case they would have reached the scene of
action too late, and they would have lost time by needlessly
riding all the way round the outer lines. Caesar's galloper, who,
if they started from Gresigny, must have ridden to order them to

advance, would have gone between the circumvallation and the

contravallation, and moreover could have ridden faster than a large
body of cavalry.
Just before Caesar came to his rescue, Labienus was re-
22.
inforced by a number of cohorts from the nearest redoubts (Labienus
. coactis U7ia XL cohortihus, quas ex proximis praesidiis deductas
. .

fors ohtulit,^ &c.). The MSS. differ regarding the number. BLM
have una XL AQ have una de XL, (39) and yS has de XL, which
; ;

is nonsense. Guischard ^ defends the reading una de XL, arguing

that the maction of Commius allowed this large number to leave the
1 B. Q., vii, 87, § 4. 2 jjist, de Jules Cesar, ii, 318.

3 M. Jullian, however {Hist, de la Gaide, iii, 529, nn. 2, 6), chooses this exposed
route.
" W. Smith's Did. of Greek and Roman Geoyr., ii, 259.
" B. G., vii, 87, § 5. « Mem. crit. et hist., 1774, pp. 507-8.

1
 — ;

THE OPERATIONS AT ALESIA 819

redoubts in the plain and join Labienus but 39 cohorts were almost
:

four legions, or two -fifths of the whole force and this number must
;

surely be in excess of the truth. I incline to accept Ciacconius's

conjecture, XI.
23. On I have stated that Keginus and Caninius sent
page 179
a galloper to inform Caesar, who, as we have seen, was probably on
the lower slope of the Montague de Flavigny, that they required
reinforcements, and that Caesar sent a galloper to order the cavalry
at Gresigny to make the charge v/hich proved decisive. There is
no direct evidence for either of these statements and it might be
;

contended that the messages were communicated by signals in the

manner described by Polybius.i But from Gresigny the signals could
not have been seen, because Mont Auxois would have interrupted
the view ^ and I am inclined to believe that Reginus could have
;

communicated with Caesar by sending a galloper, who would only

have had to ride about two miles and a quarter, more quickly than
by the elaborate system of signals which Polybius describes.
24. Eius adventu, writes Caesar, ex colore vestitus cognito, [quo
insigni in proeliis uti consuerat,] turmisque equitum et cohortibus
visis quas se sequi iusserat, ut de locis superioribus haec declivia
et devexa cernebantur, hostes 'proelium committunt? The MSS. all
have hostes ; but Nipperdey and A. Holder, followed by various
English editors, substitute nostri. If hostes is right, the meaning
is that the enemy, seeing Caesar approaching with reinforcements,
made a last desperate effort to storm the camp on Mont Rea before
he could arrive if nostri, that the Romans under Labienus were
:

encouraged to make a bold sortie. In defence of the emendation

nostri, which he borrowed from Jurinius, Nipperdey ^ says, First of'

all, it is incredible that the approach of Caesar and the reinforce-

ments which the Romans received should have induced the enemy
to engage. Secondly the enemy did by no means engage at that
moment they had done so long before, at the time when the attack
:

on the Roman lines commenced ; and since then they had not
relaxed the vigour of their attack.' I cannot see that these are
sufficient reasons for setting aside the authority of the MSS. Hostes
seems to me to make perfectly good sense ;and the appearance of
nostri in the same chapter, a couple of lines further on, militates,
I think, against the substitution of nostri for hostes. But, whichever
reading is right, I have no doubt that the effect of Caesar's approach
was to stimulate the Gauls to make a last desperate effort to storm
the camp before he could arrive, and to encourage the Romans to
make a bold sortie.
[Since writing the above paragraph, I have referred to Schneider's
edition (ii, 634). I find that he reads hostes, and that he defends it
by practically the same arguments as I have put forward myself
1 X, 43-7. Cf. Vegetius, iii, 5, and B. G., ii, 33, § 3.
^ The same objection holds good in the other case if, as Napoleon believes,
Labienus, whom Caesar sent to reinforce Reginus, was encamped on the Mon-
tague de Bussy.
^ B.Q., vii, 88, « Caesar, p. 110.
§ 1.

3a2
820 THE OPERATIONS AT ALESIA
and particularly that he expresses the same opinion as I have done
regarding the repetition of nostri.]
25. The reader must take the story of the surrender of Vercin-
getorix, which I have told in the text, for what it may be worth,
It only rests upon the authority of Plutarch and of Florus and ;

M. Salomon Reinach ^ regards it as simplement I'echo de quelquc '

pantomime historique representee a Rome but Mommsen ^ accepts '

:

it without question and Long ^ says, It is so lively and so natura'

;
'

a thing, so truly Gallic, that it has been got from some authority,'—
perhaps from some lost memoirs by one of Caesar's officers ? Florus
says, Ipse ille rex .swpplex cum in castra venisset, turn et phalerai
. .

et sua arma ante Caesaris genua proiecit ; and Plutarch,^ 'O Se to\

(rv/x7ravTo<s rjyc/xiov iroXe/xov Ovepyevropi^ avaXa/Sojv twv ottXoov to

KaWiara koI Koafxrjda'; rov 'iinrov e^tTTTracaro hia rcov ttuXwv, Kat KVKXii
Trept Tov Katitrapa KaOe^ofievov iXdaas, ctra d^aXo/xei/os tov Ittttov tyji
fi€V TTavoirXCav aireppuf/ev, auros Sk KaOto-as viro TroSas tov Katcrapo? ycrv-
Xtov rjyev, o.)(pL ov TrapeSoOr) (j)povpr](T6ix€VO<s CTrt tov Opcafx^ov. H. Flem^
ming^ holds that Caesar's statement (Vercingetorix) deditur'^— —
'
en son laconisme terrible, que le chef arverne fut con-
signifie,
duit a Cesar par ses propres compagnons d'armes.' Obviousl}
this is not inconsistent with the statements of either Florus o]
Plutarch.

THE PASSAGE TANTA TAMEN . . . INCUM-

BERENT (B. G., vii, 76, § 2)

(Huius opera Commii . atque utili superioribus annis eral

. . fideli
usus in Britannia Caesar pro quibus meritis civitatem eius im-
;

munem esse iusserat, iura legesque reddiderat atque ipsi Morinos

attribuerat.) Tanta tamen universae Galliae consensio fuit lihertatu
vindicandae et pristinae belli laudis recuperandae ut neque heneficiii
neque amicitiae memoria mover entur omnesque et animo et opihm
in id helium incumberent. Instead of ynoverentur the reading ol —
all the MSS. —
Meusel ^ adopts Kraffert's emendation, fnoveretur, the
subject of which he takes to be not Gallia but Coynmius. To begin
with, he argues, moverentur would have no definite subject. More-
over, Caesar never would have said that the Gauls were all indebted
to him for favours, beneficiis and amicitiae memoria are per-
whereas
fectly applicable to Commius. Further, if moverentur represented
Caesar's meaning, he would have written some such sentence as
(recuperandae) ut nemo aut beneficiis aut amicitiae memoria mover etur
omnesque (or sed omnes), &c. Finally, the word tamen shows that
what follows contains a special reference to Commius.
^ Pro Alesia, 1908, p. 376. See also Rev. des etudes anc, iii, 1901, pp. 131-9.
2 Bom. Oesch., iii, 1889, p. 291 {Hist, of Rome, v, 1894, p. 90).
3 Caesar, p. 406. « i, 45 (iii, 10). ^ Caesar, 27.
« Pro Alesia, 1909, p. 486. 7 B.C., vii, 89, § 4.
* Jahresb. d. philol. Vereins zu Berlin, xii, 1886, p. 290.
 — —

TANTA TAMEN . . . INCUMBEBENT 821

I am not convinced. Read this translation and see whether it

does not sound natural. Yet so intense was the unanimous deter-
'

mination of the entire Gallic people to establish their liberty and

recover their ancient military renown that no favours, no recollection
of former friendship, had any influence with them, but all devoted
their energies and resources to the prosecution of the war.' What
would be the effect of substituting him or Commius for them ?
' ' ' ' ' '

As the passage stands, it does contain a reference to Commius :

'
But so intense that ... all [including Commius], &c. The
. . . '

lack of a subject to moverentur is not surprising for in vii, 73, ;

§§ 5, 8, 9, appellabant is used twice and nominabant once without

a subject, because the subject milites is supplied by —
the common
sense of the reader. I can easily understand that Caesar would have
said rhetorically of the Gauls in general that he had shown them
favours and friendship, just as he said so of the Remi (v, 54, § 4)
and of the Aedui (ih. and vii, 54, §§ 3-4). Cf. i, 30, § 2. And, even
supposing that mover etur is right, may not its subject be Gallia ?
Cf. vii, 77, § 7.

THE EXECUTION OF VERCINGETORIX

The only ancient writer who mentions the execution of Vercin-
getorix is Dion Cassius. Stoffel says that it was in accordance with
Roman custom, and that Caesar could not have prevented it if he
had wished.^ But we cannot tell whether he did wish and for my ;

part I do not care to ask. Ihne,^ indeed, appears to doubt whether

Vercingetorix was executed at all. The silence of Plutarch, he argues,
who derived much of his information from the lost books of Livy,
throws suspicion on the statement of Dion, who often unjustly
charged Caesar with cruelty. I am no admirer of Dion but I doubt ;

whether he was impudent enough to invent a tale like this.

THE ATTITUDE OF THE AEDUI DURING THE

LATTER PART OF THE REBELLION OF VER-
CINGETORIX
I discuss this question principally because a theory has been con-
structed regarding it by F. Monnier, which, although it probably
contains a certain proportion of truth, has, I believe, been pushed
too far, and is, moreover, supported by arguments which will not
bear examination. His theory is that the Aedui were mainly respon-
sible for the ultimate failure of Vercingetorix.
L On page 175 of his Vercingetorix he says, A ce malheureux '

siege d'Alise, ou domina contre Vercingetorix la faction eduenne, la

^ Guerre civile, ii, 299.

* Rom. Gesch., vi, 523, n. 1. Cf. C. Jullian, Vercingetorix, pp 396-7.
822 ALLEGED TREACHERY
majorite des nobles gaulois s'opposa ... a la levee en masse que
voulait Vercingetorix, parce que, disent-ils, chaque chef ne pourrait
plus ni diriger, ni reconnaitre ses vassaux, nee moderari, nee dis-
cernere suos ; et non seulement les Eduens eurent alors la haute
main pour fixer les chiffres des contingents, mais ce fut a Bibracte,
chez les Eduens, que se forrna I'armee de secours, et le commande-
ment reel fut confie a des Eduens, Eporedorix et Viridomar, tous
deux ennemis de Vercingetorix.' And in the same spirit Desjardins
says,^ '
Ces chefs '
—
that is to say, the chiefs of Gaul generally
'
redevenus maitres de la situation par le blocus d^Alesia, craignaient,
en presence d'une multitude confuse, de ne pouvoir la contenir, de ne
pas reconnaitre les leurs, ni d'etre en mesure de pourvoir aux vivres :

— mauvaises excuses, mesquins faux-fuyants, vains pretextes, les

memes dans tous les temps, et colorant la defection.'
Now, with all deference to Desjardins, the decision of the chiefs
was wise, and their reasons were excellent. The experience of the
Belgic confederation which opposed Caesar in 57 B.C. had shown
how difficult it was to feed an overgrown army .2 The other reason
which Caesar assigns for the decision of the chiefs does not, as Monnier
imagines, imply that they were afraid that Vercingetorix would usurp
their authority over their own clients for why should he be able to
;

interfere less with a comparatively small than with a large number ?

It simply means that they were afraid that, if they sanctioned a levy
en 7nasse, their respective contingents would get out of hand. How
would Desjardins have proposed to feed a host consisting of all the
fighting men of Gaul ? And, if 250,000 were not enough, can any
one with an iota of common sense believe that a larger army, or
rather mob, would have succeeded ? It is curious that the Duke of
WelHngton and Sir Charles Napier, both no doubt unwittingly,
sanctioned by their authority the reasons given by the chiefs. All '

history,' wrote Napier,^ tells us that neither barbarous nor civiHzed

'

warriors of different tribes or nations long agree when compressed.'

Of the Duke, Sir William Napier tells this story ^ —
Speaking of
: '

Waterloo, the duke said to Mr. Kogers the poet, " Napoleon should
have waited for us at Paris." " Why ? " " Because 800,000 men
would then have gathered round him." " Is not that the reason
why he should not ? " " No why he should for when 800,000
! ;

men get together, there 's a damned deal of jostling." ^ '

Oeogr. de la Gaule rom., ii, 702.

^

B. G.j'u, 10, § 4. M. Jullian ( Vercingetorix, p. 278) also refers to this passage

2

'
mais,' he adds, il s'agissait alors d'une campagne longue et compHquee, et
'

non pas, comme pour le salut d'Alesia, d'une marche de quelques jours et d'un
assaut de quelques heures.' But has M. Jullian considered how the enormous
—
hosts which he demands a milHon instead of a quarter of a million were to —
be fed on their way to Alesia, many of them coming from the uttermost parts of
Gaul ? Has he thought of the transport which they would have needed, to say
nothing of organization ? Has he forgotten that the host which attacked
Sabinus in 56 b. c, when the campaign was neither long nor complicated, pre-
cipitated their attack for want of food {B. G., iii, 18, § 6) ?
^ Life and Opinions
of Charles James Napier, iii, 218. * lb.

^ M. Jullian {Vercingetorix,
pp. 274-5, 278) thinks that a million men nearly —
60 for each linear yard of the entire extent of the circumvallation would have —
 —

OF THE AEDUI 823

With regard to the other statements in the extract which I have

given from Monnier's work, there is no evidence that the Aeduan,
or any other faction, domina contre Vercingetorix
'
it is a pure'
:

invention to say that the Aedui fixed the numbers of the contingents
of the relieving army and to say that le commandement reel fut
;
'

confie a des Eduens Eporedorix et Viridomar ', is to be guilty not

only of a suggestio falsi but also of a swppressio veri. For Monnier
knows perfectly well that, besides Eporedorix and Viridomarus, two
other generals, namely Commius and Vercassivellaunus, were ap-
pointed ;and he also knows that the two latter figure much more
prominently in Caesar's narrative than the two former.^ Besides, how
is Monnier's theory to be reconciled with the fact that, only a few

weeks before, the Aedui had been severely snubbed in a Pan-Gallic

council, when they claimed the right of directing the campaign ? ^
Is it likely that, in the next council,^ they would have been allowed
to have matters all their own way ?
2. Monnier maintains that the fact that the army of relief was
assembled in Aeduan territory, proves that the baneful influence of
the Aedui was in the ascendant.^ Does he forget that the council
which so emphatically pronounced that not the Aedui, but Vercin-
getorix was to hold the chief command, assembled at Bibracte, the —
Aeduan capital ?

3. Monnier appeals, in support of his theory, to the detailed list

which Caesar furnishes ^ of the several contingents. Les con- '

tingents,' he says, sont d'autant plus nombreux qu'ils sont devoues

'

aux Eduens, et moins devoues a la cause nationale.' ^ If so, how

does he explain the fact that the contingents of the Arverni (Ver-
cingetorix's countrymen) and of the Aedui, with their respective
clients, were precisely the same ? But although he finds an ally
—
Albert Reville' whom I formerly answered^ from respect for
his reputation, I will not again try the reader's patience by
examining arguments which rest upon the uncritical use of worth-
less manuscripts and which no one who can construe easy Latin
would accept.
That the Aedui were treacherous, or at least half-hearted, it would
be absurd to deny we have Caesar's word for it and I have already
: ;

shown that they were partly responsible for the ruin of Vercin-
getorix.^ But it was not for lack of numbers that the relieving army

been able to fill up the trenches and rush ' the lines. I ask my friend to con-
'

sider whether they would not have got in each other's way. I cannot help
thinking that for once he has allowed the burning patriotism which with other
great qualities makes all that ho writes so fascinating to obscure his judgement.
^ B. 0., vii, 79, § 1 ; 83, § 6 ; 85, § 4 ; 88, § 4.
2 16.,
63, §§ 5, 7.
3 76., 75, § 1. Monnier requires us to believe that while the Aedui dared not
sever themselves from the other rebels {neque tamen suscepto hello suum consilium
ah reliquis separare audent [ih., 63, § 8]), those rebels were merely instruments in
their hands. * Vercingetorix^ p. 232.
^ B. O.y vii, 75. " Vercingetorix,
p. 235.
' Rev. des Deux Mondes, 3*^ per., xxiv, 1877, p. 476. Cf. ib., pp. 468-9.
8 On
pp. 802-3 of the first edition. » Sec
pp. 178, 182, 81.':-6.
824 WHO WROTE THE EIGHTH BOOK
failed it was lack of quality, lack of unity, above all lack of one
;

head that proved their ruin. Vercingetorix had no peer who could
impose his will on the relieving army as he imposed his on the
besieged.^

WHO WROTE THE EIGHTH BOOK OF THE COM-

MENTARIES ?

The orthodox view, that the last book of the Gallic War was
written by Aulus Hirtius, has been impugned by F. Vogel.^ The '

—
best MS.', as he admits apparently he means Amstelodamensis —
bears Hirtius's name at the head of the book but he observes that ;

in the same MS., at the head of the First Book, appears the superscrip-
tion, incipit liber Suetonii. Undeniably; but is the blunder of a scribe
to discredit the unhesitating statement of Suetonius ^ that the Eighth
Book was the work of Hirtius. Let us, however, consider Vogel's
remaining arguments. Admitting that Hirtius, like the writer of the
Eighth Commentary, was a friend of Balbus and that he took no part
in the Alexandrian or the African war,* he insists that the writer's in-
tention of bringing the narrative down to Caesar's death, but not to
'the end of the civil strife,'^ must have originated a considerable time
after the Ides of March, 44, and probably even after April 27, 43 B.C.,
— the date of the battle of Mutina, in w^hich Hirtius was killed for, ;

he asks, before the latter event what could have induced a military
historian to think of continuing his narrative beyond Caesar's
death ? He does not, indeed, deny that the preface of the Eighth
Commentary may have been written in 44 but if so, it must, he
;

maintains, belong to the second half of that year for not until ;

then did the political situation give occasion to fear a renewal of

the civil war. Now in April Hirtius was with Cicero at Puteoli,
practising oratory for his approaching consulship.^ In May Hirtius
and Balbus met at Aquinum and when Cicero appended to his
;

mention of this meeting the words sed quid egerint^ he certainly —

anticipated that they were going to discuss politics^ and not to
* M. Jullian's view {Vercingetorix,
p. 234) that Vercingetorix's power was
reduced after the council at Bibracte, and that he was hampered by his political
opponents even among the Arverni, is. I think, unsupported by evidence. The
council had acclaimed him unanimously he had all the troops that he wanted
:
;

and, so far as we know, the only opposition that he had to encounter was that
of the Aedui. I admit that the Arvernian chiefs whom he had banished {B. G.,
vii, 4, § 4) may have intrigued against him but why more after the council at
;

Bibracte than before ?

2 Neue Jahrh.f. Altertum, &c., v, 1900, p. 187.
d. klass.
^ Divus lulius, 56. —De
isdem comraentariis Hirtius ita praedicat Adeo :

prohantur omnium iudicio ut praerepta non praebita facultas scriptoribus

videatur, &c. Suetonius's quotation is taken from B. G., viii, Praef., § 5.
4 See B. G., viii, Praef., § 8.

••
lb., § 2. « Cicero, Att., xiv, 12, § 2.
' Fam., xvi, 24, § 2. See Tyrrell and Purser, The Correspondence of Cicero,
vi, 35, note.
^ Cicero's letter was written ajter the discussion, whatever it may have been
about. After the words sed quod (not quid)egerinf, he breaks ofl Probably he
meant ' —
But what they talked about, is no affair of mine.' See ib., p. 30, note.
 OF THE COMMENTARIES! 825

talk about Caesar's literary papers, which, moreover, at that time

were still in the hands of Antony .^ Soon afterwards Hirtius wrote
to Cicero ^ that he was leaving Eome for his Tusculan villa. On the
2nd of September Cicero ^ says that the Eoman people had offered
up prayers for the life of Hirtius, who had been very ill and his ;

recovery was slow.* Even on the 22nd of April, 43 b. c. a few days —

before Hirtius's death —
Cicero alluded to his weak health.
How, then, asks Vogel, could Hirtius have found time and strength
for the work which is attributed to him ? Would he not have
alluded to his weak health and his consulship in the preface in
which the writer excused himself to Balbus for his dilatory perform-
ance of his task ? Besides, we must infer from the preface that the
writer had taken part in the Spanish War. But Hirtius spent 46 b. c.
in Rome, and, as we may gather from one of Cicero's letters to
Narbo till the 18th of April, 45, a month
Atticus,^ did not reach —
after the battle Munda. So little, moreover, did he know of
of
the theatre of war that at that time he was ignorant of the
whereabouts of Gnaeus Pompeius. Finally, since the murder
of Caesar Hirtius and Antony had not been on good terms,
whereas the writer of the Eighth Commentary speaks of Antony
with enthusiasm.^
These are not arguments which can disturb a traditional belief.
Prefaces are generally written after the books to which they belong
are virtually completed why should the preface of the Eighth
:

Commentary be regarded as an exception ? Oratorical exercises

and discussion of politics with Balbus need not have consumed the
whole of the writer's time the papers of Caesar which were in
;

Antony's hands were not historical, but political and there is no ;

reason to suppose that Hirtius had not ample leisure for so light
a task as the completion of the Gallic War before his health gave
way."^ As to the argument which Vogel bases upon the date of
Hirtius's appearance at Narbo, H. Schiller ^ justly says that it might
deserve attention if Vogel could prove that Hirtius had not been at
Munda before. Moreover, the letter which Vogel quotes only proves
that Hirtius was at Narbo on the 18th of April, not that he arrived
there on that day he may have served in the Spanish campaign
:

without being present at Munda nor can I understand why know-

;

ledge of the movements of Gnaeus Pompeius should be deemed an

essential qualification of a writer who claimed to have taken part
in the war.^ How many officers who served in the Transvaal would
have been able at any given time to indicate the whereabouts of
General de Wet ?
Nobody will suggest that the writer of the Eighth Commentary
'
Cicero, Phil, i, 7, § 16. 2 j^tL, xv, 6, § 2.
^ Phil, i, 15, § 37. * Cicero, Fam., xii, 22, § 2.
•'
xii, 37, § 4. « 50, § 2.
1 See Berl. phil. Woch., 1903, col. 1420. « Ih.
* As a matter simply told Cicero that Sextus Pompeius had fled
of fact, Hirtius
to Hither Spain, and that Gnaeus had fled {Ait. xii, 37, § 4) and Cicero adds ;

nescio quo.
826 THE SECOND CAMPAIGN
was Oppius and whoever rejects the orthodox view must be pre-
;

pared to defend the paradox that, although he was an intimate

friend of Caesar and of Balbus, he was never mentioned in the corre-
spondence of Cicero .1

CAESAR'S SECOND CAMPAIGN AGAINST THE

BELLOVACI
I. The description which Hirtius^ gives of the geography of this
campaign is not sufficiently definite to enable us to determine the
various sites with absolute certainty. The essential part amounts to
this. The Bellovaci and their allies encamped on a hill, standing in
a wooded country and protected by a morass. This morass was at
the bottom of a deep and narrow valley, separating the hill from
another on which Caesar pitched his camp and the position of the
;

Gauls was so strong that it would have been impossible to storm it

without very heavy loss. The Roman camp had two ditches with
vertical sides (lateribus derectis). The hill upon which the Bellovaci
were encamped was separated by another small valley (mediocris
vallis) from a hill (iugum), the summit of which formed a plateau,
and the sides of which were steep and this hill was large enough
;

to admit of four legions being drawn up on it in line of battle. Caesar

reached it by bridging the morass and the bridge or causeway was
;

constructed within a single day. About 10 Roman miles from the

camp of the Bellovaci was a very strong place, upon which they
encamped after they had been compelled to abandon their original
position. Somewhere in the neighbourhood of this second encamp-
ment was a meadow not more than a Roman square mile in extent,
which was surrounded partly by woods, partly by a very deep river.*
This meadow, Hirtius tells us, was said to be not more than 8 miles
from the second camp of the Bellovaci.
It is generally assumed that the original camp of the Bellovaci
was situated somewhere within their own borders. But this is not
certain. When Caesar set out on his march against the Bellovaci,
they were preparing for a raid against the Suessiones. When he
encamped in their territory, he found that the bulk of the popula-
tion had left the country, and that only a few scouts had been sent
hack (remissi). After narrating this, Hirtius goes on to describe the
spot which the Bellovaci and their allies had chosen for their encamp-
ment but he does not tell us in whose territory it was. The natural
;

conclusion to be drawn from the word remissi is that it was not in

the country of the Bellovaci. Long, however, thinks differently.
'
The Bellovaci,' he says,^ a warlike people, would not quit their
'

See pp. 209-10 and A. Klotz, Caesarstudien, pp. 154-5.

^

See Carte de V Mat- Major (1 80,000), Sheets 32 and 33.

2 :

3 B.Q., viii, 6-20.

* flumine altissimo {B. G., viii, 18,

§ 1) ; or, if we adopt the reading of a
—
—
impeditissimo flumine a river very difficult to cross,
^ Caesar, p. 436.
 AGAINST THE BELLOVACI 827

country, but their allies would come to them.' Yes But just !

because they were a warlike people, they had quitted their country,
in order to invade that of the Suessiones and although the invasion
;

doubtless came to nothing, it is quite possible, and, as I think, it is

to be inferred from what Hirtius says, that they found a suitable
place for making their stand against Caesar before they got back to
their own country.
1. Napoleon claims to have discovered Caesar's camp. The site,
Mont St. Pierre, which is on the east of the Oise, and about 3 miles
south of its tributary the Aisne, was in the diocese of Soissons,i and
therefore probably in the territory of the Suessiones. A glance at
the map ^ will show that it corresponds exactly with Hirtius's
description but unhappily the sides of the ditches of the camp, as
:

revealed by Napoleon's excavations, are not vertical and M. Vau-

;

ville has recently proved that the camp was a Gallic fort.^
Desjardins * believes that the problem is insoluble. II est im- '

possible,' he says, avec des donnees aussi vagues qui se bornent

'

meme a des descriptions physiques convenant egalement a beau-

coup de lieux situes dans la meme region, sans que jamais un seul
nom geographique les accompagne, — de determiner exactement le
theatre des operations militaires,' and, after pointing to the un-
satisfactory results of the excavations, he adds, le ru de Berne
'

n'est pas un marais.' Certainly the rivulet itself is not a marsh :

but it flowed through marshy ground and when Desjardins says

;

that many places in the same district correspond with Hirtius's

description, he is, as I shall presently show, mistaken.
2. L. d'Allonville ^ labours to prove that Caesar's first camp was
close to Bellifontaine, which is in the commune of Liercourt, about
5 miles south-east of Abbeville. But all his labour is wasted for ;

Liercourt is in the territory of the Ambiani, only a short distance

from the mouth of the Somme and such a position is in complete
;

disaccord with Hirtius's narrative.

3. Peigne-Delacourt,^ an enthusiastic French antiquary, dis-
covered, before the publication of Napoleon's book, a wooden bridge,
apparently of ancient construction, which he took to be the bridge
mentioned by Hirtius,'^ near Clermont, in the Breche, which flows
into the Oise on its right bank. But Desjardins ^ objects that the
bridge must have taken a long time to make, and that the adjacent
—
country the high ground dominating Clermont does not suit —
Hirtius's narrative. Peigne-Delacourt ^ himself admits that his
bridge would have taken 60 days to make and, as it is clear from
;

Hirtius that the bridge was thrown across the marsh in a single

^
Seemap in Gallia Christiana, t. ix.
2
Carte de V Mat-Major (1 : 80,000), Sheet 33.
^
Mem. de la Soc. nat. des ant. de France, Ixviii, 1908 (1909), pp. 160-84.
* Geogr. de la Gaule rom., ii, 717-8.

^ Dissertation sur les camps rom. du dep^ de la Somme, 1828,

pp. 85-124.
* Etude nouvelle sur la campagne de J tiles Cesar contre les Bellovaques, 1869,

' B. G., viii, 14,

pp. 5-6. § 4.
^ Geogr. de la Gaule rom., ii, 719, note. ^ Etude, &o., p. 24.
828 THE SECOND CAMPAIGN
morning, this one admission shatters his whole theory. De Grattier i
points out that Caesar could have reached the hill (Mont de Cren)
which Peigne-Delacourt identifies with the iugum of Hirtius,^ with-
out making a bridge at all, and that the hill is not separated from the
plateau on which Peigne-Delacourt believes the Bellovaci to have
encamped by anything that can be properly called a mediocris
vallis?
According to von Groler,* Caesar encamped on the north-west
4.
of Morfontaine, on the right bank of the Vandy, which enters the
Aisne at Lamotte. This position is about 5 miles south-east of
the camp which Napoleon identifies with that of Caesar. The Gauls,
according to von Goler, occupied a plateau bounded on the south by
the rivulet Retheuil, on the west by that of Pierrefonds and by the
forest of Compiegne, and on the north by the woods of St. Etienne.
But, says General Creuly,^ the plateau was assailable on the side of
St. Etienne, and indeed on every side, except where it touches the
valley of the Vandy and a part of the valley of the Retheuil it :

would therefore have been unnecessary to enclose it, as Caesar in-

tended to do,^ by a line of contravallation and if such a line had
;

been made, it would have been 22 kilometres, or nearly 14 miles, in

extent
M. Litonnois'^ argues on the hypothesis that, according to
5.
Hirtius, the campaign took place in the country of the Bellovaci.
The river (altissimmn jiumen) which Hirtius mentions in chapter 18,
must then, says Litonnois, have been the Oise, because the only other
'
very deep river in the territory of the Bellovaci is the Aisne and
'
;

as Caesar had already mentioned the Aisne by its name Axona, he

would not have described it merely as an altissimum jiumen. But,
I may remark, Hirtius, not Caesar, was the author of the Eighth
Book of the Commentaries.
The Oise, then, continues Litonnois, was the furthest limit of
Caesar's operations in the campaign and his point of departure
;

was the common frontier of the Suessiones and the Bellovaci. There-
fore the camp of the Bellovaci must have been somewhere between
those limits. Now the only spot between them that answers to the
narrative of Hirtius is the hill of Gouvieux, near Chantilly.
There, protected on its rear by the Oise, was the camp of the
Bellovaci the valley of the Nonette, which flows into the Oise,
:

separated them from Caesar and the strong place to which they
;

ultimately retreated was somewhere in the hilly district of the

forest of Carnelle.^
Now Litonnois is mistaken in asserting that Caesar's point of

^
Com. arch, de Noyon, ii, 1868, pp. 168-9. B. G., viii, 14, § 4.
'

^ M. Jullian {Rev. des etudes anc, xi, 1909, pp. 359-60 Hist, de la Gaiile,
; iii,

584, note), who agrees with me in accepting Napoleon's site, nevertheless thinks
—
that a modification of Peigne-Delacourt' s theory Caesar encamping at Cler-
mont and the Bellovaci in the bois des Cotes
'
—
is not absolutely inadmissible,
'

* Gall. Krieg, 1880,

p. 339.
Rev. arch., nouv. ser., viii, 1863, pp. 513-4.
'"'
^ B. G., viii, 11, § 1.

' Comptes rendus et mem. du Com. arch, de Senlis, 1865, pp. 129-42.
» See Carte de V Mat- Major (1 80,000), Sheet 32.
:
 '
;

AGAINST THE BELLOVACI 829

departure must have been on the common frontier of the Suessiones

and the Bellovaci. Caesar came from the south-west, from Orleans — ;

and there is no evidence that he first marched into the country of

the Suessiones, and then back into that of the Bellovaci. It is true
that he had ordered Fabius to march into the country of the Sues-
siones, in order to repel the threatened invasion of the Bellovaci
but that does not prove that he marched into the country of the
Suessiones, to join Fabius, and thence into the country of the Bello-
vaci. This, however, is only a detail. I believe that the altissimum
flumen was the Aisne but even if it was the Oise, the camp of the
;

Bellovaci was not on the hill of Gouvieux. For, according to Liton-

nois, in their retreat they left the Oise on their right that is to say, ;

they moved southward and crossed the Nonette. But, if Caesar was
obliged to bridge the Nonette in order to reach the iugum which
Hirtius describes, how could the Bellovaci have crossed it with all
their wagons ?

6. abbe Devic ^ holds that the Bellovaci encamped on

Finally, the
Mont Cesar (which, as we have seen, M. Seymour de Ricci tentatively
identifies with Bratuspantium 2), and Caesar on Mont de Hermes,
a hill which is now covered by the Foret de Hez, on the further side
of a rivulet called the Try. But look at Sheet 32 of the Carte de
V Mat- Major. Where is the iugum separated by a mediocris vallis
from the hill on which the Bellovaci originally encamped ? The
abbe says that it is un espace de quelques centaines de metres de
'

largeur which is au pied proprement dit du Mont Cesar ', and

' '

which fait partie de la vallee [the

'
deep and narrow valley '

above mentioned], mais qui est un peu plus eleve que le marais '.
So, then, a space at the foot of Mont Cesar, which formed part of
' '

a valley, is to be regarded as a hill (iugum) The abbe identifies the !

'
very strong position (loco munitissimo) to which the Bellovaci
'

retreated with Mont Bourguillemont, which, he says, is 6 or 7 miles

from Mont Cesar but the map shows that it is not more than
;

5J kilometres, or little more than 3J Roman miles. And where is

Hirtius's very deep river ? The abbe points to the Therain
'
but '
;

he adds with a candour which stultifies his theory, nous avouons '

franchement que la qualification de tres-profonde ne lui est pas

applicable.'
In spite of Desjardins's objections, I believe that the site adopted
by Napoleon and selected, before the publication of his book, by
Caignart de Saulcy,^ is the most probable of any that has been
named. It corresponds in every particular with Hirtius's descrip-
tion and it is not correct to say, as Desjardins says, that others
;

correspond equally well. To quote de Saulcy, J'ai souvent par- '

couru, et dans tons les sens, ces magnifiques forets [Cuise, Compiegne,
and Laigne], et je n'y ai reconnu qu'un seul point qui concorde avec
la description d'Hirtius mais il est vrai qu'il presente une ressem-
;

^ Stude sur les ii^ et viii'^ livres des Comm. de Cesar, 18G5, pp. 111-2.
•^
See p. 402.
^ Les campaynes de Jules Cesar dans les Gaules, 1802, p. 401.
830 CAMPAIGN AGAINST THE BELLOVACI
blance si saisissante avec le terrain sur lequel tous les faits de cette
campagne se sont deroules, qu'il faudrait etre plus que difficile pour
ne pas y reconnaitre le lieu cherche.' General Creuly agrees with
de Saulcy and minute study of the Carte de V Etat-Major leads me
;

to the same conclusion. As M. JuUian remarks, the fact that the

trenches discovered by Napoleon's collaborators were not made by
Caesar does not prove that Caesar did not encamp on Mont
St. Pierre.i
II. Assuming the correctness of de Saulcy's and Napoleon's
identification of the camps of Caesar and the Bellovaci, the next
problem is to determine the geography of the second stage of the
campaign. The strong position (loco munitisswio),^ on which the
Bellovaci made their second camp, can only be Mont Ganelon.
Hirtius describes it as not more than 10 (Roman) miles from their
original position. The nearer part of Mont Ganelon is only about
6 miles off but the hill stretches so far to the north-west that his
;

vague statement is, in one sense, no exaggeration. Besides, as General

Creuly points out, the distance which the Gauls would have had
actually to travel in order to reach Mont Ganelon is greater than the
distance as the crow flies.^ Von Kampen^ makes the retreating
Bellovaci first cross the Aisne, and then cross the Oise but de Saulcy ;

holds ^ that they would not have attempted the passage of the Aisne
at all, in which they would have been liable to attack. Anyhow it
would have been unnecessary for them to cross both the Aisne
and the Oise all that they had to do, in order to reach Mont
:

Ganelon, was to move up the left bank of the Aisne and then
cross the Oise.
According to Hirtius,^ the plain in which the Bellovaci attacked
the Roman cavalry was said to be not more than 8 miles from
' '

their second camp. It must, as de Saulcy remarks,"^ have been on

the left bank of the Aisne, because Caesar would not have had time
to bridge the river on the day on which he marched against the
Bellovaci. De Saulcy and Napoleon^ place it opposite the village
of Choisy-au-Bac General Creuly ^ and von Kampen, about 2 miles
:

further up the Aisne, in the bend of the river between Choisy and
Rethondes. The former site is not more than a mile and a half
from the nearer part of Mont Ganelon. The latter, as Creuly observes,
agrees better with the statement of Hirtius.

^
Rev. des etudes anc, xi, 1909, pp. 359-60.
2
B.G., viii, 16, § 3.
^
Rev. arch., nouv. ser., viii, 1863, p. 513.
*
Quindecim ad Caesaris de b. G. comm. tabulae, xiv.
^
Les campagnes de Jules Cesar dans les Qaules, p. 417.
«
viii,20, §1.
i?. (?.,
''

Les campagnes, &c., p. 119. ^ Hist, dc Jules Cesar, ii, 332.

**
Rev. arch., nouv. ser., viii, 1863, p. 515.
 831

WHERE WAS THE BRIDGE BY WHICH DUMNACUS

CROSSED THE LOIRE ?

Napoleon ^ asserts that the bridge was at Saumur, but gives no

reason. L. Fallue^ believes, avec,' as he asserts, les plus savants
' '

geographes,' that the bridge was at Ce, on the road from Angers to
Poitiers, which, he argues, Dumnacus had probably followed, in
order to enter the country of the Pictones.^ But, even if Dumnacus
had taken this road, it does not follow that, in trying to escape
across the Loire,* he returned by it. M. Jullian,^ who thinks that
the choice lies between Ce and Saumur, prefers the latter because it
is nearer Lemonum (Poitiers) and is a very ancient place of passage.

WHO WAS GUTUATER ' '

?

The MSS. differ a great deal, the various readings being Gutrua-
tum, Gutuatrmn, Guttruatrum, Gutruatium, Gatriatrium, and Gutirua-
tum 6 but Gutuater was undoubtedly the same man as the
;
'
'

Cotuatus mentioned in B. G., vii, 3, § 1. Schneider indeed thinks

otherwise. Hirtius, he argues,' reminds us that Caesar has related,
in the preceding Commentary, that the great rebellion originated
among the Carnutes ^ but he tells us that Gutuater ' was the
;
' '

author of that signal crime without quoting Caesar as his authority.

'

This argument is more subtle than convincing. Again, says Schneider,

Hirtius speaks of Gutuater in such a way that we should take him
' '

for a man of great influence and exalted position, and not a desperate
outlaw like Cotuatus. Hirtius simply speaks of him as 'principem
sceleris illius et concitatorem belli. Sceleris illius means the massacre
at Cenabum or it means nothing. Cotuatus was the leader, or rather
one of the two leaders (the other being Conconnetodumnus) of the
Carnutes who perpetrated that massacre and the natural con- ;

clusion is that Gutuater was Cotuatus.^ How, then, are we to

'
'

account for the difference ? It is now recognized that Gutuatrum^

coming from a nominative Gutuater the Latinized form of a Gallic —
—
word, Gutuatros is the true reading in B. G., viii, 38. M. R. Mowat
^ Hist, de Jules Cesar, ii, 336.
2 Etudes arch, sur Vhist. de Jules Cesar 'par Vempereur Napoleon III, 1867,
p. 106. Cf. Eev. de VAnjou, nouv. ser., xiv, 1887, p. 215. In this article
M. A. Joubert says that popular tradition places the scene of the battle between
Dumnacus and Fabius in the neighbourhood of Ce. But arguments based on
tradition are worthless, unless it can be proved that the tradition is genuine,
and not the offspring of the brain of some local antiquary.
3 See B. G., viii, * lb., 27, § 2.
26, § 1.
^ Hist, de la Gaule, iii, « B. 0., viii, 38, § 3.
554, n. 4.
' Caesar, ii, 329. «
j^q^^ ^iii^ 38^ § 3,
^ I find that Heller very sensibly remarks, There is nothing absurd in the
'

idea that a man who had massacred Roman citizens, even though ho had on
that account obtained groat inUuenco among his own countrymen, should have
been called by Caosar a desperado {Philologus, xvii, 1861, p. 283).
'
832 WHO WAS GUTUATER
'
' ?

cites two inscriptions, in which GVTVATER and GVTVATR appear.

In the second inscription, as restored, the words gutuatri Martis
occur and M. Mowat beheves that Hirtius mistook the word, which
;

means a priest',^ for a proper name.^ So also does Otto Hirschfeld,

'

who, holding, as I do, that the Gutuater of Hirtius was the person
designated, in the MSS. of Caesar as Cotuatus, argues that while
Caesar gave the man's name, Hirtius only intended to point out
'

that the ringleader held the priestly office of gutuater '.^ Accordingly
he suggests that Hirtius may have written Cotuatum gutuatrum^
I confess that I doubt whether Hirtius knew what gutuater meant.

THE DURATION OF CAESAR'S PROCONSULSHIP

On page 191, describing Caesar's operations in 51 b. c, I have
written, Only one more summer had to pass, as the malcontents
'

had doubtless reckoned, and his government would be at an end.'

This sentence is based upon a passage in B. G., viii, 39, cum omnibus
Gallis notum esse sciret reliquam esse unam aestatem suae 'provinciae,
quam si sustinere potuissent, nullum ultra periculum vererentur. Long,
who holds that Caesar's term of office was not to expire until the
end of 49 B. c, remarks that Hirtius may mean that Caesar would '

have no time for a campaign in b. c. 49, for he would be com- . . .

municating with his friends in Rome about his election in that

year'.^ There would be no need to have recourse to this explana-
tion if we could accept Mommsen's view, that Caesar's term would
expire on March 1, 49. But Otto Hirschfeld^ has recently given
reasons for dissenting from this conclusion. If, he argues, the lex
Pompeia Licinia had provided that Caesar's tenure of office should
not expire before March 1, 49, Pompey could not have claimed that
the desire which was expressed in the Senate to recall him on
November 13, 50, was reasonable."^ It is now almost unanimously
admitted, on the evidence of a passage in Cicero's speech, De provin-
ciis consularibus,^ that the five years for which Caesar was originally
appointed terminated on March 1, 54 b. c. The extension of his
command, agreed upon at the conference of Luca and confirmed
(under the lex Pompeia Licinia) by Pompey and Crassus, as consuls,
in the following year, 55 b. c, was to last, according to Plutarch,^
Appian,i^Velleius,iiand Suetonius,^^ f or another period of five years;
but Dion Cassius ^^ affirmed, as the result of his inquiries, that it
1 See Rev. celt., xxviii, 1907, p. 120.
- Rev. de philologie, nouv. ser., i, 1877, pp. 274-5.
3 Sitzungsherichte d. kaiserl. Akad. d. Wissenschajten, Wien, cxiii, 1883,
p. 313, n. 7.
* Sitzungsherichte d. kdnigl. preuss. Akad. d. Wissenschaften, 1897, p. 1117.
^ Decline
of the Roman Republic, iv, 443.
« Klio, iv, 1904,
pp. 76-87 ; v, 1905, pp. 236-40.
' Cicero, Fam., viii, 11, § 3. « 15, § 37.

9 Pomp., 51-2 Cras., 15 ; Caesar, 21. '« B. C, ii, 17-8.

;

" ii, 46, § 2. ^2 ])ii^us lull us, 24.

^^ xxxix, 33, § 3 xliv, 43, § 2.
; It is worth noticing that, as L. Holzapfel
remarks {Klio, v, 1905, p. 108, n. 1), Dion himself (xliv, 43, § 1), when he puts
 —

DURATION OF CAESAR'S PROCONSULSHIP 833

had been limited to three, and Hirschfeld thinks that he was not far
wrong. Remarking that a passage in one of Cicero's letters ^ proves
that a date was fixed in the lex Pompeia Licinia for Caesar's recall,
he maintains that a clue to this period is to be found in a letter
written by Caelius to Cicero,^ which shows that a clause in the law
forbade the Senate to discuss the appointment of Caesar's successor
before March 1, 50. But, as the lex Sempronia had prescribed that
proconsuls should not take charge of their provinces until a year
and a half after the provinces had been assigned to them, Caesar's
tenure of his province was virtually assured, so long as that law
remained valid, for a year and a half after March 1, 50, in fact —
until January 1, 48. The lex Sempronia, however, was abrogated by
Pompey in 52 and (assuming that Hirschfeld is right in maintain-
;

ing that Caesar's command was not expressly secured until March 1,
49) the result was that the Senate would be entitled, without violat-
ing the letter of the lex Pompeia Licinia, not merely to discuss the
appointment of Caesar's successor, but to send that successor to
Gaul on March 1, 50. A plebiscite, passed in 52, had indeed authorized
Caesar to stand for the consulship without presenting himself in
Rome and, as he could not be elected consul before July, 49, this
;

plebiscite virtually authorized him to retain his command until that

date, if not until January 1, 48, when his year of office would begin ^ :

but the authorization was infringed by the lex de provinciis, which

Pompey carried later in the same year,^ although, yielding to pressure,
he declared, informally, that his law did not apply to Caesar.^ I will
only make one observation Cicero ^ implies unmistakably that the
:

lex Pompeia Licinia prolonged Caesar's command for five years.

Reams have been written upon the question but it properly ;
"^

belongs to the history of the civil war. What I am principally con-

cerned with here is the fact, recorded by Hirtius, that the Gauls, as
Caesar was aware, knew that the summer of 50 B. c. would be the
last in which he would be free to deal with them. Hirtius was
Caesar's intimate friend and it is incredible that on a point like
;

this he should have been mistaken. Hirschfeld, however, insists

that his words have hitherto been misunderstood. He tells us that
C. Bardt (a well-known Ciceronian scholar) pointed out to him that

into Antony's mouth the statement that Caesar was forced to return to Italy
before the lawful time, apparently confirms the tradition which assigned a legal
duration of ten years to Caesar's proconsulship. Still, Antony may only have
interpreted the law from Caesar's point of view.
^ AtL, vii, 7, § 6.
2 Fam., viii, 8, § 9. Cf. AtL, viii, 3, § 3 1 : and Suetonius,
; B. G., viii, 53, §
Diviis lulius, 28. See B. C, i, 9, § 2. ^

* Dion Cassius, xl, 56, 1. ^ Suetonius, Divus lulius, 28.

§
^ Att., vii,
7, § 6. annorum enim decern imperium et ita latum placet? See
Tyrrell and Purser, The Correspondence of Cicero, iii, 284.
' Mommsen, Die Rechtsfrage zwischen Caesar und dem Senat, 1857 ; F. Hofif-
mann, De origine belli civilis Caesariani, 1857 A. W. Zumpt, Studia Romana,
;

1859 P. Guiraud, Le dijferend entre Cesar et le Senat, 1878 Journal des Savants,
; ;

1879, pp. 437-49, &c. Hirschfeld's first article in Klio provoked a reply in the
same periodical (v, 1905, pp. 107-16) from L. Holzapfel and Hirschfeld wrote ;

a rejoinder (pp. 236-40).

1093 3 H
834 DURATION OF CAESAR'S PROCONSULSHTP
unam aesiatem must mean, not the summer of 50, but the summer
of 51 B. C, —
the year of the last Gallic campaign and this interpre-
;

tation is one of the points on which he relies to prove that Caesar's

command was terminable under the lex Pompeia Licinia on March 1,
50. How, he asks, could Caesar's personal intervention in the siege
of Uxellodunum have been explained by Hirtius on the ground that
his only remaining summer in command was that of 50 B. c. ? If
he could have looked forward to another entire summer, in which he
would be free to dispose of his army, why should he have been so
anxious to capture Uxellodunum at once ? But Hirschfeld's in-
terpretation does violence to the meaning of a plain Latin sentence.
It is certain that when Caesar started for Uxellodunum the summer
of 51 was already far advanced for immediately after the capture
;

of the fort, when he went to Aquitania, only the fag-end of the

summer remained.^ Now consider the earliest chronological indica-
tion which Caesar gives in regard to his first invasion of Britain :

'
only a small part of the summer remained (exigua parte aestatis
'

reliqua)? And only a small part of the summer of 51 b. c. remained

when Caesar started for Uxellodunum. Therefore, if Hirschfeld is
right, reliquam esse unam aestatem is equivalent to reliquam esse
exiguam partem aestatis If Hirtius had meant this, would he not
!

have said it ? ^ If he had written reliquos esse duos annos, Hirschfeld

will admit that he would have meant two whole years
'
why, '
:

then, should unam aestatem be distorted into meaning a small part '

of one summer ? I ask Hirschfeld to look through Meusel's Lexicon

'

Gaesarianum, iii, 1665 ff. He will not find there one solitary instance
in which reliquus or any of its cases, coupled with a noun, denotes
— —
a part only of the thing time or what not signified by that noun.
Nothing, then, can be more certain than that, if Hirtius had intended
to convey the meaning which Hirschfeld atributes to him, he would
have written, not (reliquam esse) unam aestatem, but (reliquam esse)
exiguam partem aestatis.
The one reason which the professor gives for accepting Bardt's
misinterpretation can be very easily disposed of. Caesar hastened
in person to crush the resistance of Uxellodunum because he knew
that if the garrison succeeded in defying his lieutenants, the mal-
contents throughout Gaul would be encouraged to prolong the
guerrilla war into the summer of 50 b. c. and he resolved that that
;

final summer should be free from all disturbance. He had been

obliged to spend eight years in coercing the Gauls he intended to:

spend the last in conciliating them.

^ in earn partem Galliae[s.c. Aquitaniam]est profectus, utibi extremum tempus

consumeret aestivorum {B. G., %dii, 46,§ 1). L. Holzapfe^(iL/^o, v, 1905, pp. 113-4)
has anticipated me in calling attention to this passage.
^ B. G., iv, 20, § 1. See also iv, 4, § 7 {reliquam partem Memis) ; v, 31, § 4
{reliqua pars noctis) ; vii, 10, § 1 {reliquam partem hiemis) ; vii, 25, § 1 {reliqua
parte noctis) ; B. G., iii, 28, § 6 {reliquam noctis partem)^ &c.
' See the passages quoted in the preceding note.
 — ^

S35

WHAT WAS THE HEIGHT OF THE TERRACE WHICH

CAESAR CONSTRUCTED AT UXELLODUNUM ?

Hirtius describes the terrace as follows extruitur agger in alti-

:

tudinem pedum LX, conlocatur in eo turris X

tabulatorum, non quidem.
quae moenihus adaequaret (id enim nullis operibus effici poterat), sed
quae super aret fontis fastigium.^
The reading LX
is doubtful. It is only found in S and in the
margin of h. The rest of the good MSS. have VI, and two inferior
— —
MSS. Andinus and Oxoniensis have IX.^ Long^ characterizes
sexaginta (LX) as a bad emendation
'
but, bad or good, it is not
'
;

an emendation at all. It is incredible that the terrace should have

been only 6 feet or even 9 feet high, for the words of Hirtius un-
mistakably imply that it was raised to the greatest possible height :

if Uxellodunum was on the Puy d'Issolu, it must have been very

much higher than 9 feet and if it had been only 6 or even 9 feet
;

high, it is to the last degree unlikely that Hirtius would have taken the
trouble to mention its height. But to say what its height really was,
is impossible."*

THE DATE OF THE ANNEXATION OF GAUL

Signor Ferrero,^ making one of those bold and startling conjec-
tures which have contributed so much to the popularity of his history
of Rome, asserts that at the end of 57 b. c. Caesar, taking advantage
of the impression produced by his victory over the Belgae, pro-
claimed, as an electioneering manoeuvre ', the annexation of Gaul.
'

In other words, with the consent of the Senate, which was of course
indispensable, he framed a lex provinciae, by which Gaul was formally
made a Roman province, and the lines upon which the administra-
tion was to be conducted were laid down. This,' says Signor
'

Ferrero, is proved by the great festivals [the thanksgiving service

'

(suppUcatio) mentioned in B. G., ii, 35, § 4] which were given at

that time, as contrasted with the indifference displayed by the
populace and all public bodies up to the close of 58, as well as by
the statements in Dion, xxxix, 5 and 25, Orosius, vi, 8, 6, Caesar,

B. G.y viii, 41, § 5. 2 Meusel's Caesar,

'
p. 242.
Decline of the Roman Republic, iv, 386, note.
^
* Achaintre {Caesar, iv, 454) remarks that, according to Orosius (vi,
11, § 25),
the combined height of the agger and the tower was 60 feet. If Orosius had said
this, it would look as if he had found the reading LX
in his copy of the Com-
mentaries, but had read Hirtius's narrative carelessly. But it is Achaintre
himself who was careless. What Orosius wrote was extruitur agger et turris
pedum LX, &c. Evidently pedum LX
belongs only to turris ; and, as Long
remarks (Caesar, p.452), Orosius 'must have calculated the height of the "turris"
from the number of stories '.
. . .

^ Grandezza e decadenza di Roma, ii, 47, note {The Greatness and

Decline of Rome,
ii, 38, n. *). .^^''"^"
3h2 .^<<rOF MEOM^i.^^
^^
ICHAEL'S
COLLEGE y/
o>J
 836 DATE OF THE ANNEXATION OF GAUL
B. G.^ ii, 35, and above
all T)e Prov. Cons., xiv, 34.' In the last-
. . .

named passage Cicero says sed tamen una atque altera aestas [the
campaigns of 58 and 57 b. c] vel metu vel spe vel praemiis vel armis
vel legibus potest totam Galliam sempiternis vinculis adstringere : —
he affirmed that Caesar had conquered Caul, and that the conquest
would be permanent whether he implied that Gaul had been
;

formally annexed the reader may judge for himself. If he will

examine the other proofs ', he will find them equally inconclusive
'
;

and in fact Signor Ferrero himself apparently lays little stress upon
them. Replying ^ to a criticism by M. Jullian,^ he urges that Caesar
had two strong motives for annexing Gaul in 57, first, the desire —
to strengthen his political position by a great coup ', and, secondly, '

the necessity of legalizing his position in Gaul, which was com-

promised by the illegality of his first two campaigns. But Signor
Ferrero relies principally upon another passage in the speech De
provinciis consularihus. Cicero, pleading Caesar's cause, sarcastically
asked whether he was remaining in his province in order to enjoy
the delightful climate and the beauties of the towns and to extend
the bounds of Roman dominion (amoenitas eum, credo, locorum,
urhium pulchritudo finiutn itnperii propagatio retinet)? How,
. . .

asks Signor Ferrero, could Cicero have implied that all the Gallic
territories were comprised in the fines imperii unless they had been
annexed ? But is Signor Ferrero quite sure that he knows the
"*

whole meaning of the word imperium ? If he will refer to three

passages in the Bellum Gallicum i, 18, § 9 45, § 3 —
ii, 1, § 4 he ; ; —
will find an answer to his question but he will also find that as
;

Rome, according to Caesar, exercised imperium over Gaul in the early

part of 57 and even in 58, he must either abandon his argument or
develop it, and maintain that Caesar proclaimed annexation immedi-
ately after he arrived in Gaul There are two facts which alone
!

dispose of Signor Ferrero's theory. Over and over again Caesar

opposes the condition of the Roman Province to that of the rest of
Gaul : it is sufficient to quote from the speech of Critognatus,
respicite finitimam Galliam, quae in provinciam redacta, iure et
legibus commutatis securihus suhiecta perpetua premitur servitute^ {^ look
at the Gaul on your border reduced to a province, its rights and laws
:

revolutionized, prostrate beneath the lictor's axe, it is crushed by

perpetual slavery '). Obviously these words prove that even in
52 B.C. Gaul had not yet been annexed. If further proof were
needed it would be enough to quote the words of Cicero, domitae
sunt a Caesare maximae nationes, sed nondum legibus, nondum iure
certo, nondum satis firma pace^ ('Caesar has subdued powerful
peoples but they are not yet bound by law, by definite rights, by
;

abiding peace '). It is useless for Signor Ferrero to try to explain

them away.
1 Bev. arch., 4« ser., xv, 1910, pp. 93-103.
Hist, de la Gaule, iii, 277, n. 4. ^ De prov. cons., 12,
2 § 29.
* I have selected the best of Signor Ferrero's arguments. The reader can
supply the omission.
^ B, G., vii, « J)e prov. cons., 8, § 19.
77, § 16.
 s

DATE OF THE ANNEXATION OF GAUL 837

Nevertheless he has reason on his side when he contends that the

annexation did not take place, as M. Jullian maintains, in the winter
of 51-50 B.C. M. Jullian, remarking that this is the orthodox view,
defends it by the following arguments.^ The constitutions of the
various tribes and their legal and financial position were, he says,
fixed, according to Hirtius,^ in 51-50, which seems to prove that tho
lex frovinciae was framed in the same year. Suetonius^ says that
at the time of the annexation Caesar fixed the legal and financial
position of the various tribes, which he did, according to Hirtius,
in that year. Sallust* specifies 51 as the date of the completion of
the conquest, and, like Suetonius, mentions the boundaries of the
conquered country therefore he was alluding to the annexation,
:

Signor Ferrero ^ objects, first, that Hirtius ^ says that nothing re-
markable occurred in the year 50, and secondly that annexation,
which, according to M. Jullian, would have needlessly exasperated
the Gauls in 56 B.C., would have done so just as much in 51-50.
M. Jullian, indeed, denies this en 51-50,' he says, les revoltes
:
' '

n'etaient plus a craindre elles avaient eu lieu.'

; But Hirtius '^

emphatically affirms that they were to be feared that was why

:

Caesar was so conciliatory. Caesar, he says, unum illud propositum

hahebat, continere in amicitia civitates, nulli spent aut causani
dare armorum ; nihil enim minus volebat quam sub decessum suum
necessitatem sihi aliquam imponi belli gerendi, ne, cum
exercitum deducturus esset, bellum aliquod relinqueretur, quod
omnis Gallia libenter sine praesenti periculo susciperet.^
In the face of these words M. Jullian must throw up his case
unless the texts to which he appeals definitively prove it. Do
they ? Kead Hirtius's statement. Immediately after the sentence
which I have just quoted, in which he says that Caesar's one
object in the winter of 51-50 was to avoid provoking the Gauls
to a fresh outbreak, he tells us that Accordingly he conferred
'

titles of honour upon states, bestowed handsome presents upon

^ Rev. arch., 4P ser., xv, 1910, p. 105.
^
B. G., viii, 49, § 1. M. JulHan refers also to Dion Cassius, xl, 43, § 3.
* Divus lulius, 25. * Fr., i, 11,
^ Rev. arch., xv, 1910, pp. 100-3. Signor Ferrero also makes a point when he
says (p. 100) that M. JuUian himself rightly maintains that the decern legati
who were assigned to Caesar in 56 b. c. were a commission appointed to organize
the conquered country. ' Mais,' says Signor Ferrero, cette commission suppose
'

r annexion.' Now most scholars hold that the legati were not commissioners,
but lieutenant-generals who were to serve under Caesar still Signor Ferrero'
:

argument is obviously legitimate. M. Jullian's opinion seems to be founded

on the statement of Suetonius {Divus lulius, 24) that the Senate, alarmed by
the military activity of Caesar after the conference at Luca, appointed legati
to investigate the condition of Gaul, though apparently, owing to his success,
they never got to work. But how can any one question the prevalent opinion
—
who reads Cicero's words {Pro Balho, 27, § 61), C. Caesarem senatus et genere
supplicationum amplissimo ornavit et numero dierum novo. Idem in angustiis
aerarii victor em exercitum, stipendio adfecit, impera tori decem I egatos deer evil,
lege Sempronia succedendum non censuitl See also De prov, cons., 11, § 28, and
G. Long, 31. T. Ciccronis oralioncs, iv, 1858, p. 102.
» B. a., viii, 48, § 10.

V
Rev. arch., xv, 1910, p. 104. « B. G'., viii,
49, §§ 1-2.
838 DATE OF THE ANNEXATION OF GAUL
the leading men, and imposed no new burdens thus he had no
;

difficulty in maintaining peace, for Gaul was wearied out by a

long succession of defeats, and therefore more disposed to submit
(it ague honorifice civitates appellando, principes maximis praemiis
adjiciendo, nulla oner a nova iniungendo, defessam tot adversis
proeliis Galliam condicione parendi meliore facile in pace continuit).
If M. Jullian is right in the meaning which he attaches to these
words, it is evident that Caesar's motive for formally annexing Gaul
and defining the constitution of the newly created province was to
avoid provoking a new rebellion. A homoeopathic remedy indeed !

But had not Caesar bestowed presents and titles before ? How can
any unbiased critic maintain that Hirtius was describing a formal
annexation ? If it had taken place, assuredly he w^ould have made
his meaning clear. As to Suetonius, he merely says at the outset of
the very brief paragraph which he devotes to Gaul^ that Caesar
'
reduced the whole of Gaul except the allied and deserving tribes
. . .

[the Remi and others] to the form of a province, and imposed upon
it under the head of tribute an annual contribution of [40,000,000
sesterces' ^] (omnem Galliam praeter socias ac bene meritas civitates,
. . .

in provinciae formam redegit, eique * * * in singulos annos stipendii

nomine imposuit). No conclusion as to date can be drawn from
these words nor is it possible to correlate them, as M. Jullian
;

tries to do, with those of Hirtius for Hirtius says nothing about
:

the formation of the province and is equally silent about tribute*

Indeed, if, as M. Jullian maintains, the arrangements which Suetonius
describes were made in 51-50, Suetonius contradicts Hirtius for ;

Hirtius says that Caesar imposed no new burdens upon Gaul, and
Suetonius says that he imposed a yearly tribute. To exact an
additional £400,000 a year from a country which had been remorse-
lessly plundered already would have been a Gilbertian way of con-
ciliating it.^ It follows that M. Jullian's theory derives no support
from Sallust, who merely defines the extent of the conquest and the
date of its completion.
Mr. Heitland^ agrees with me that in 51-50 Caesar only made
a provisional settlement. In the ordinary course,' he says, the
' '

report of the completed conquest should have led to the appointment

of a senatorial commission for drawing up ... a charter or organic
statute ... to regulate the administration. Or the recent precedent of
Pompey might be followed, and the scheme of settlement be entrusted
to the .proconsul alone. But Caesar well knew that the Senate
. .

would do neither their first object was to get him out of Gaul,' &c.
. . .

I conclude that the formal annexation of Gaul took place in

Caesar's dictatorship but the tribute may have been imposed
;

before the winter of 51-50, just as in 54 Caesar fixed the tribute '

which Britain was to pay annually to the Roman People',^ although

the country had not been formed into a province.

'
Dimes lalius, 25. The amount in ascertained from Eutropius, vi, 17.
"^

'
Possibly Caosar had collected tribute before, and in 50 B.C. fixed the sum.
* The liomcm Itepublk, iii, § 1100. ' B. 0., \, '2'2,
§ 4.
 . '

839

THE SPELLING OF CELTIC NAMES i

In bpelliiig Celtic proper have, throughout my own

names, I
narrative, generally followed Caesar and where the MSS. differ,
;

I have endeavoured by collating them to discover what Caesar

wrote or have been guided by other ancient writers or by the
'^

evidence of coins or inscriptions while, in default of positive testi-

;

mony, I have been guided simply by modern usage. But for mere
accuracy, where accuracy ran counter to usage, I have cared nothing.
That is to say, where the forms which Caesar adopted differed from
the true Celtic forms, I have generally preferred the former but ;

even in following Caesar I have sometimes been deliberately incon-

sistent. For instance, although Caesar wrote Haedui, I have written
'
Aedui ', not because h was wanting in old Celtic, but because Aedui '

is familiar and on the other hand, although h was wanting in old

;

Celtic, I have written 'Helvetii' and 'Helvii', partly because Caesar

used those forms, but also because they are established. Caesar dealt
with Celtic in the same way that Anglo-Indians dead and gone
dealt with Indian names. If his lot had been cast in India in the
days of John Company ', he would have written Cawnpore not
'

Kanhpur. As his lot was cast in Gaul in the last century B. c, he

wrote Orgetorix, not Orcetirix. I subjoin a list of doubtful names
and of others the orthography of which there is sufficient evidence
either to determine or with reasonable probability to conjecture.
My chief aim throughout has been to ascertain the forms which
were adopted by Caesar.^
Admagetobriga. See Magetobriga —
Aedui. See Haedui.—
Agedincum, which is sanctioned by the best MSS. of Caesar,*
and by the Itinerary o/Antonine,^ is right. An inscription discovered
at Sens, which stands upon the site of Agedincum, has agied.^
M. d'Arbois de Jubainville considers that the ie in this inscription
is perhaps only an altered form of the e in Caesar, comme on a dit '

"pied" pour pede.^ Gliick^ concludes that the first four letters
'^

of the word were certainly Aged and for the termination he relies
;

on the analogy of Alisincum, Vapincum, and Lemincum. It should

^ It is not certain that all the names to be mentioned in this note were Celtic
;

but into that question I need not enter.

2 I have been greatly helped by Meusel's articles in Jahresb. d. philol. Vereins

zu Berlin, xii, 1886, pp. 265-71, 277-9 xx, 1894, pp. 214-21.
;

* I need hardly point out that in the MSS. there is no distinction between

u and V, Suevi for instance being written Sueui^ or that in quoting from the MSS.
I give the nominative even when a name is only found there in an oblique case.
4 vi, 44, § 3 vii, 10, § 4
; 57, § 1 ; 59, § 4 ; 62, § 10.
;

=^
P. 383.
Rev. de philoloyie, ii, 1847, p. 356. The Gallic coin, bearing the inscription
'

ArHA, to which I referred in the Hrst edition (p. 813), has nothing to do with
Agedincum. See A. Blanchct, Tralle den monn. (jimL, pp. 10, 362.
' Dcsjardins's (Jeo(jr. de la (Jaide rom., ii, 469, n. 8.

^ Die bei Camr vorkonunenden kell. Namen, 1857,

pp. 15-8.
840 THE SPELLING OF CELTIC NAMES
be added that the genuine Gallic form was Agedincon, and that the
Gallic termination of all names which Caesar latinized into um
was on.^
—
Alesia. According to an inscription found on the plateau of
Mont Auxois, the Gallic form was Altsi/a. See Rev. arch., nouv.
ser., XV, 1867, p. 314, and p. 361, supra.
Ambivareti is, according to Gliick,^ the name of a people whom
Caesar mentions in B. G., vii, 75, § 2, 90, § 6. Gliick argues that
the word is formed from ambi ('around,' 'mutually') and var (in
Cymric gwared = freed '), so that Ambivareti would mean defend-
' '

ing each other '. The reading of the best MSS. in chapter 75, is
Ambluareti,^ which only differs in one letter from the reading for
which Gliick argues. Ambluareti, says Gliick, is a monstrosity '.'

Ambivareti, however, is not found in any MS.

Andebroglus is perhaps the true form of the name of one of the
two Reman envoys who waited upon Caesar in the spring of 57 b. c*
Andocotnborius has indeed more MS. support, being attested by ap,
while Andebrogius appears only in tt, on the margin of h, and (written
by a later hand) in B. Andocomborius, however, is rejected as
impossible by Holder,^ who, pointing to coins which bear the inscrip-
tion ANDECOMBO, coujecturcs that Caesar wrote Andecombogius.
Meusel ^ thinks it safer to follow tt.
Andes and Andi are found in the best MSS. of Caesar but ;
*'

Gliick ^ and A. Holder^ regard Andecavi as the true form. They find
support in Plinyio (iv, 18 [32], § 107), Tacitus {Ann., iii, 41), Ptolemy
(ii, 8, § 8), Orosius (vi, 8, § 7) and many other authors, as well as in

a coin mentioned by T. E. Mionnet,^^ w^hich is said to bear the legend

ANDECAV. No such name, however, is to be found either in the
catalogue of MM. Muret and Chabouillet or in M. Blanchet's authori-
tative Traite des mofinaies gauloises. M. d'Arbois de Jubainville ^'-
regards Andicavi, which is found in Pliny and Orosius, as the oldest
form. Holder derives the word from the particle ande and cav (in
Cymric caw), which means a band or tie '. ' ' '

Aremoricus, not Armoricus is the true form, though Armoricus

is found in all the MSS. of Caesar.^^ Pliny ,i* Ausonius,!^ Rutilus,^^
and Sidonius Apollinaris ^^ all write the e. The particle are, more-
over, is found in other Gallic names, for instance Arecotnici. See
also Holder, Alt-celtischer Sprachschatz, i, 202.
Atuatuca is found in the ^ MSS. of Caesar Aduatuca in the a.
;

^ A. Holder, Alt-celtischer Sprachschatz, 'passim. *

pp. 22-4.
3 In chapter 90 the best MSS. have Ambibareti. * B. G., ii, 3, § 1.
^ Alt-celtischer Sprachschatz, \, 143-4.
**
Jahresb. d. philoL Vereins zu Berlin, xx, 1894, p. 218.
^ ii, 35, § 3
; iii, 7, § 2
; vii, 4, § 6. «
p. 24.
^ Alt-celtischer Sprachschatz, 140-3.
i,
^" Pliny (ed. D. Detlefsen, 1866) has Andicavi, with v. 1. Andigavi.
^^ Descr. de medailles antiques, i. 80.
^2 See Desjardins, Geogr. de la Gaule rom., ii, 483, n. 3.
i«
'3 V, 53, § 6 vii, 75, § 4.
; Nat. Hist., iv, 17 (31), § 105.
^6 Ed. R. Peipor, p. 59, 1, 28 (v. 10, 25).
'6 Ilin., i, 213. 17 Curm., vii,
247, 369.
 THE SPELLING OF CELTIC NAMES 841

The former is undoubtedly correct for, though Aduaca, with the ;

variant Aduaga, is found in the Itinerary of Antonine} Ptolemy ^

writes krovarovKov, and Atuaca appears in the Table?
'

Atuatuci, which is found everywhere in y8 and in two passages

(ii, 29, § 1 vi, 2, § 3) in a as well, is certainly right.'*
; Orosius ^
has Atuatuci, and Dion Cassius ^ 'ArovartKot'.
Aulerei, Aulurci, and Auleurci are the forms found in the best
MSS7 of Caesar of a name which, on coins,^ appears as Aulirci.
The editors adopt the form Aulerei.
—
Caerosi. Caeroesi is found in the a MSS. of Caesar.^ Orosius ^^
has Caerosi and this, which is the reading of h, is adopted by
;

Meusel.^^ Holder reads Caeroesi in his edition,!^ but Caerosi in his

Alt-celtischer SfracJischatzP Gliick ^* says that the Gauls would have
written Cairoisi.
Caleti. —
Gliick 1^ prefers Caletes. Cadetes is found in B. G.,
vii, 75, §4, but Caletos in ii, 4, § 9 Pliny ^^ has Galeti; Orosius ^^ ;

Caleti Strabo
;
^^ KaXerot and Ptolemy ^^ KaXrJTaL.
;

Cavillonum is to be preferred to Cabillonum. Both forms occur

in other writers, and on coins but in the Commentaries there is
;

better MS. authority for the former. It is found (with the variant
Cavillonno) in all the MSS. in vii, 42, § 5,^^ ^nd in fi in vii, 90, § 7.
Cebenna is preferred by Gliick ^i to the usual reading Cevenna.
Cebenna is found in B. G., vii, 8, §§ 2-3, in L and (written by a second
hand) in M
in vii, 8, § 2 in/; and in vii, 56, § 2 in tt /. Gliick refers
;

to Mela,22 Pliny ,23 and Ausonius,^* and derives the word from ceb (in
Cymric kefyn, cefn), which means a ridge '. '

Cenabum is undoubtedly preferable to Genahum, although it is

not found in any of the MSS. of Caesar.^^ In the two passages in
which Hirtius^^ mentions the name, ax have Cenabum, and p Gena-
bum. That Cenabum is the true form is proved, says Desjardins,'^'^
by Strabo, Ptolemy, the Itinerary of Antonine, the Table, and the
inscription which was discovered in 1846 in the Faubourg St.Vincent
at Orleans. Gliick ^8 derives Cenabum from cena, which, he says, is
identical with the Irish cen, and from which Cenomani is derived.
So also M. d'Arbois de Jubainville.
—
Cenomani. Gliick ^9 prefers this to Ceno^nanni (which is the
^
p. 378. 2 Geogr., ii, 9, § 5. »
p. 12, col. I.
* Jahresb. d. philol. Vereins zu Berlin, xx, 1894, p. 216.
' vi, 7, § 14 ; 10, § 2. « xxxix, 4, § 1.

' ii, 34 ; iii, 29, § 3 ; vii, 4, § 6 ; 57, § 3 ; 75, §§ 2-3.

® E. Muret and M. A, Chabouillet, Cat. des monn. gaul. de la Bihl. Nat., 7046-9.
" ii, 4, § 10. '« vi, 7, § 14.
'^ Jahresb. d. philol. Vereins zu Berlin, xii, 1886, pp. 278-9.
pp. 43-4.
'^ 13 i, 677. 1^
p. 36. 1*
p. 40.
i«
Nat. Hist., iv, 18 (32), § 107.
1' vi, 7, § 14. 18 19
iv^ 3^ § 5. Geogr., ii, 8, § 5.
2" See Jahresb. ^i
Vereins zu Berlin, xx, 1894, p. 215.
d. philol. p 57^
22 23 ^at. Hist., iii, 4
ii, 5, §§ 74, 80. (5), § 31 ; 17 (31), § 105.
24 Opuscula, ed. R. Peiper, p. 150, 1. 102 (xi, 18, 5) p. 151, 1. 114 (xi, 19, 8). ;

2' The adjectival form Cciiabensi is found in it, in B. G., vii, 28, § 4.
2''
lb., viii, 5, § 2 ; 0, § 2. " (Jkogr, de la Oaule rom., ii, 477, n. 1.
-«
pp. 58-U. -»
p. 59 find n. 1.
842 THE SPELLING OF CELTIC NAMES
reading of the best MSS.),^ on the analogy of many other words of
the same termination. He also remarks that Polybius, Strabo, and
Ptolemy were wrong in writing Kevofxavot for Kiyvo/xavoi, as the
quantity of the e is fixed by the line Te Met agnatos visere Ceni-
manos? I do not question the conclusion but Latin poets occasionally ;

lengthened or shortened the quantities of proper names, for metrical

purposes. Thus Virgil writes Nee non Tarquinium eiectum Porsenna
iuhebat;^ while Horace writes Minacis aut Etrusca Porsenae manus^
—
Ceutrones. Centrones is, according to Gliick,^ the name which
Caesar gives to one of the client peoples of the Nervii.^ Ceutrones,
however, is found in all the good MSS. Gluck says that ceutr, from
which. Ceutrones would be derived, is a form foreign to Celtic. He
derives Centrones from centr (in Breton kentr, in Irish cinteir, a spur),
and infers that it means spur- wearing '. I have not much faith in
'

this derivation for it would be very strange that the clients of

;

a people who, as Caesar says,' had no cavalry, should have been

called spur- wearers '.
'

Gliick spells in the same way Centrones the name of an Alpine —

tribe, which Caesar mentions in B. G., i, 10, § 4. Ceutrones, however,
is found in all the best MSS., and in the famous inscription of For-

claz.^ Thus, unless Ceutrones was misspelt in the inscription, the

etymological argument of Gliick collapses.
Commius ought strictly to be spelled Commios. See Sir John
Rhys's Celtic Britain, 1904, p. 25, and Rev. celt., i, 294.
Coriosolites is preferred, as having the best MS. authority, by
Holder ^ and Meusel ^^ to the more familiar Curiosolites. Coriosolites
is found in X
in B. G., iii, 11, § 4 and vii, 75, § 4 also in AQa in ;

ii, 34 and in /5 in iii, 7, § 4 while Curiosolites is found in <^ in ii, 34,

;

and in a in iii, 7, § 4.
—
Diablintes. Diablintres, which is found in the a MSS. of Caesar,^^
is preferred by Gliick ^^ to the usual form Diablintes. The MSS. of
Pliny 13 have Diablintes, Diablinti, and Diablindi of Orosius ^^ ;

Diablintes. Ptolemy i^ has AtaySAtVai. Gliick decides for Dia-

blintres, out of deference to the a MSS., and because he thinks it more
likely that the rwas omitted by copyists than inserted. He derives
the word from the privative particle dia and blin, from which he
believes that blintres, connected with blinterus (' weary '), arose.
Meusel, giving due weight to the testimony of Orosius, who used
a manuscript of Caesar older than our archetype, and of Orosius,
decides for Diablintes^^

1 B. G., vii, 75, § 3. ^ inscr. Lat., v, p. 623, no. 15.

Corpus
« B. G., v, 39, § 1.
3 Aen., viii, 646. * Epod., xvi, pp. 62-3.
4. '

' E.G., ii, 17, § 4. 8 j)ict, arch, de la Gaule, i, 252.

» Alt-celtischer Sprachschatz, i, 1126-7. Holder refers to Laval, 1892, p. 9,

and Bull hist, et arch, de la Mayenne, 1892, p. 163. I have not been able to
verify these references.
i«
Jahresh:d. philol Vcreins zu Berlin, xii, 1886, p. 277. " B. G., iii, 9, § 10.
'^
pp. 93-5. '3 Nat. Hist., iv, 18 (32), § 107.
'* '" Geoyr.,
vi, 8, § 8. ii, 8, § 7.
'®
Jahresb. d. philol. Vereius zu Berlin, xii, 1886, p. 278.
 THE SPELLING OF CELTIC NAMES 843

Diviciacus, not Divitiacus, appears to be what Caesar wrote ;

and this form is also found in the oldest MSS. of Cicero, De Divina-
tione, i, 41, § 90, —
the only passage in which he mentions thename.^
Dumnorix is the popular form of the name of the Aeduan dema-
gogue whom Caesar put to death in 54 b. c, and is certainly what he
wrote. The coins exhibit three forms, Duhnoreix, Duhnorex, and
Dubnorx ^ but M. Blanchet ^ does not believe that they were
;

issued by Dumnorix.
Eporedorix, the Aeduan chieftain whom
Caesar mentions in
conjunction with Viridomarus, is generally so called in the MSS.*
Eporedirix is found in an inscription.^
Esuvii appears, as I have shown in my geographical note (Esuvii),
to be the name of the Aremorican people whose name occurs, in
various forms, in B. G., ii, 34, iii, 7, § 4, and v, 24, § 2. Gliick,^ who
knew nothing of the coin to which I refer in the note in question,
decided for Esuhii. From the point of view of spoken language, he
says, both Esuhii and Esuvii are right and it is difficult to decide
;

between them, because h and v were often confused. He thinks it

safer, however, to adhere to Esuhii, the reading of a in B. 6^,, iii, 7.
Geidumni is, according to Gliick the name of a people who are
'^

mentioned, among the clients of the Nervii, by Caesar ^ only. The

MSS. do not help us much for some of them have geidunos and
;

others geudunos and of these abbreviated forms the former might

;

mean either Geidunnos, which Nipperdey reads, or Geidumnos.

Schneider, however, prefers the latter, on the analogy of Garmnna and
Dumnorix.
Genava is, as Gliick^ shows, the true form, although the best
MSS. of Caesar have Genua}^ Genavensihus and Cenavensihus arc
found in inscriptions.^^
—
Gutuater. See p. 83L
Haedui. — Gliick ^^ prefers Aedui, remarking that there is no h in
old Celtic,and that where it is found in transcripts of Celtic words,
it is only a breath ', which the Romans prefixed to the pure Celtic
'

vowel. Thus they wrote Helvii and Helvetii for Elvii and Elvetii,
which forms are found in inscriptions.^^ Heller,^'* on the other hand,
referring to J. C. Orelli (Inscr. Lat. coUectio, No. 3432), points out
that, if some inscriptions have the form Aedui, others have Haedui.

' lb., pp. 269-70. AEIOTIGIIAGOS— an amusing jumble—is found on

coins attributed to Diviciacus, king of the Suessiones (E. Muret and M. A.
Chabouillet, Cat. des monn. gaul. de la Bibl. Nat., 7717-22, &c.), and A E V C A
I I I

{ib., 7737) on a coin attributed to Diviciacus, the Druid. Cf. A. Blanchet,

Traite des monn. gaul., p. 377.
^ See Diet. arch, de la Gaule, i., nos. 65-6, 163, and Rev. celt., i, 1870,
p. 295.
^ Traite des monn. gaul.,
p. 408, n. 2.
* As there is a consensus in five passages —
vii, 38, § 2 ; 39, §§ 1, 3 ; 55, § 4 ;
63, §9— the corrupt forms which occur in various M8S. in the others are negligible.
" Rev. celt., iii, 1876-8,
p. 167. "
pp. 95-102. ' p. 102.
« B. G., V, 39,
§ 1. »
pp. 104-7. '» i,
6, § 3 ; 7, §§ 1-2.
^^ Inscr.
Helv., ed. Mommscn, nos. 83-4 {Miilheil. d. antiq. Gesellschajl in
Z/rrich, x, 1854, p. 14). '2
p^, «)_i4,
•'
J. Gruter, Inscr. ant. (utius orbis Roin,, 1707, vol. i, dccxxviii, 9 ; J. W. C.
kSieinor, Codex inscr. Rom., 1851, 466. ^* Fhilologus, xvii, 1861,
p. 272.
844 THE SPELLING OF CELTIC NAMES
But this only proves that inscriptions, on questions of orthography,
are not necessarily conclusive. Meusel hits the nail on the head
when he replies to Gliick that the critic's business is simply to deter-
mine what Caesar wrote and his minute examination of the MSS.
;

(which is confirmed by my collation of L) proves that Caesar wrote

Haedui} Holder ^ derives the Celtic name from aidu-s (in Irish aid,
in Welsh aidd), which means zeal or rivalry '. edvis is found
'
'
'

on coins.
Latobrigi is found in the best MSS. in i, 28, § 3 and 29, § 2, though,
like Nitiohroges (q.v.), it was misspelt by some scribes in the first
passage (i, 5, § 4) in which it occurs, where A has Latovicis, B Lato-
bicis, M
Latocihis, and /5 Latocucis. A mutilated inscription has
ATOB * Ptolemy ^ has AaroySt/coi. Gliick ^ prefers Latovici, re-
:

marking that Pliny mentions a Pannonian people of that name,

'^

and referring to the forms, analogous as he believes, Eburo-vices,

Branno-vices and Lemo-vices. He identifies vici with the Cymric
guic and the Iiishftch, a district '. Zeuss ^ and Meusel hold to the
'

form Latobrigi, which, at all events, has good MS. authority.

Lemonum, the reading of /? and, in viii, 26, § 1, of L, is better
than Limo, which Frigell adopts on the authority of the rest of the
a MSS. Ptolemy, says Gliick,^ is wrong in writing Ai/xowv, just as
he is wrong in writing Klixovlkoi. Lemonum, according to Gliick,
is derived from lem (in Irish leamJi), an elm.' '

Litaviccus preferred by Gliick,^^ Holder, and Meusel. The coins

is

have Litavicos but Litaviccus is found in an inscription.^'^ Gliick

;
^^

considers that both forms are right, and remarks that in Celtic names
c is often doubled.
—
Lexovii. The true form of this name appears to be LixoviiP
Lucterius is the form found in the best MSS. of Caesar of a name
which is spelled on a coin i^ lvxtiirios. See Alesia.
Lutecia, not Lutetia, which appears in most editions, is almost
certainly what Caesar wrote. Lutetia has comparatively little MS.
authority .1^ Strabo ^^ writes AovKoroKLa, and Ptolemy ^'^ Aovkotckm.
Holder,^^ who
regards the latter as the true form, thinks that Lutecia
was abbreviated from it, just as Leucamulus was abbreviated from
Ijeucocamulus See, however, C. Jullian, Hist, de la Gaule, i, 177, n. 2.
.

—
Magetobriga. The form Admagetobriga is adopted by Holder .^^
AQ have (quod proelium factum sit) Admagetobrige, BMS Adinageto-
^ Jahresh. d. philol. Vereins zu Berlin, xii, 1886, pp. 265-9.
^ Alt-celtischer Sprachschatz, i, 65.
' E. Muret and M. A. Chabouillet, Cat. des monn. gaul. de la Bibl. Nat.,
4822-31. * Gliick, p. 113. ' Geogr., ii, 14, § 2.
«
p. 112. ' Nat. Hist, iii, 25 (28), § 148.
^ Die Deutschen und die Nachharstamme, 1837, p. 236. *
p. 117.
^"
p. 119. ^^ See Diet. arch, de la Oaule, i, no. 67.
^^ Holder, Alt-celtischer Sprachschatz, ii, 245.
'* See A. de Barthelemy, in Encyclopedie-Roret, p. 112.
^* Muret and Chabouillet, Cat. des monn. gaul. de la Bihl. Nat., 4367.
15
See B. G., vi, 3, § 4 ; vii, 57, § 1 ; 58, §§ 3, 5, 6. L has Lutetia every-
'^
where. iv, 3, § 5.
'^ Geogr., ii, 8, § 10. '" Alt-celtischer Sprachschatz, ii, 301-2.
'» lb., i, 42.
 THE SPELLING OF CELTIC NAMES 845

hriae, and L Admagetrohiae : but / has ad Magetohriam ; and

Schneider,! citing B. C, iii, 79, §4 {proelio ad Dyrrachium facto),
ih., (uno die
53, § 1 VI proeliis factis, trihus ad Dyrrachium), and
ib., 38, §
i, 4 (helium ad Ilerdam gerere), maintains that Caesar
. . .

would have used the preposition, not the locative.^ Gliick,^ who
rightly holds that Caesar wrote ad Magetohrigam, believes that
the name is derived from mag ('a field'). Holder, on the other
hand, interprets Admagetohriga as fortress of Admagetos '. But '

might not Magetohriga mean fortress of Magetos ? ' '

Mandubii is the form found in all the MSS. of Caesar, in B. G.,

vii, 68, § 1 and 78, § 3 but in 71, § 7, while the /? MSS. have
;

Mandubii, the a MSS. have Manduvii. M. d'Arbois de Jubainville*

believes that the true form is Manduhili, as an inscription with the
form Mandu-hilos has been found in the territory usually assigned
to the Mandubii.^ The MSS. of Strabo ^ (iv, 2, § 3) have Mav8i-
(BovXiiiV.
Meclosedum. —
Caesar mentions the town of which I provisionally
assume be the name, as written by him, four times. The various
this to
readings in B. G., vii, 58, § 2 are Metiosedum (tt^™), Mellodunum
(xp), Melledunum (BMS), and Meledunum (L) ; in 58, § 6
Metiosedo (S^), Melloduno (x), (a) Metclodone (BM), and (a)
Metlodone (L) in 60, § 1 Metiosedo (x^l^), ameclodone, that is a Meclo-
;

done (BM), and Mellodone (L) in 61, § 5 manu et losedum (BMS)

;

and manu. et tosedum (L). In all the passages Schneider adopts the
form Metiosedum, Nipperdey and Frigell Melodunum Mommsen,' :

after comparing the MS. forms with Mecleto, which occurs in the
Itinerary of Antonine,^ and Mecledonense castrum, which is mentioned
by Gregory of Tours,^ concludes that Caesar probably wrote Meclo-
dunum, and this form is now adopted by Meusel and Holder ^^ but ;

M. J. Vendryes has published an interesting article ^^ which will

convince any unbiased reader that the Caesarian spelling was either
Metiosedum or (as I would suggest) Meclosedum. Kemarking that
-sedum (a dwelling-place) is a Gallic word as well known as -dunum,
M. Vendryes observes that the name Melun is derived from -dunum,
just as Autun, Embrun, and Verdun are derived respectively from
Augustodunum, Eburodunum, and Verodunum, and that, on the
other hand, a mutilated Gallo-Koman inscription has been found in
Melun, containing the letters osedi.^^ Melun, therefore, has had two
Celtic names, ending respectively in -sedum and -dunum and since ;

Melun is derived from -dunum, the name which ended in -dunum

was evidently later than the other. As to the first part of the word,
1 Caesar, i, 66. ^ g^e
p. 446. »
pp. 121-3.
* Les noms gaulois chez Cesar, &c., p. 128.
^ R. Mowat, Inscr. de la cite des Lingons, V^ partie, p. 35, no. 67.
6 Ed. Muller and Diibner, p. 962.
7 Jahresb. d. philol. Vereins zu Berlin, xx, 1894, pp. 209-10.
8 Ed. Wesseling, p. 383. ^ Hist. Franc, vi, 31.

'" AU-celtischer Sprachschatz, ii, 491.

" Mem. de la Soc. de linguistique, xiii, 1905-6, pp. 225-30. Cf. O. Hirschfeld
in Corpus inscr. Lat., xiii, vol. i, pars i, pp. 443-4.
'3 Pull, de la Soc. d'arch. du dep^ de Seine-et-Marne, x, 1894, pp. 90-3.
846 THE SPELLING OF CELTIC NAMES
Gliiok ^ preferred the form Mello- on the analogy of Mellosutum ^
and Mellentum, and also because he believed the name to be derived
from meall, 'a hill'. But Heller^ demolished this argument by
pointing out that the town was on an island in the Seine, where
there is no hill and, as M. Vendryes says, Gliick's etymology has
;

been abandoned. M. Vendryes gives the following reasons for pre-

ferring Metlo- (for which Metio- was substituted in certain MSS. by
a natural copyist's error) to Meclo-. Phonetically a transition from
Meclo- to Metlo- would have been impossible whereas it would
;

have been quite natural for Caesar and his officers, when they heard
Gauls pronounce Metlo-, to transform it into Meclo-. M. Vendryes's
conclusion is that the primitive name of the town was Metlosedum, —
an abbreviated form of Metulos-sedum, the residence of Metulos
' '
;

and that this name was replaced by Metlodunum, which in its turn
became corrupted into a mediaeval Mellodunum, the etymological —
source of the modern Melun^
M. Vendryes's excellent article is perhaps open to one minute
criticism. He says that the Komans, incapables de prononcer le
'

groupe tl, en auraient fait Meclo-, et cette derniere forme, autorisee

par la prononciation romaine, serait devenue la forme ecrite ^ '
;

but he goes on to say that in Caesar's text 'Metlosedum est toutefois

^

la le^on la plus vraisemblable '. Yes, in the manuscripts but ;

surely not in the original work, written by Caesar's hand or dictated

by him ? ^ Did not he write Meclosedum ? Doubtless this is what
M. Vendryes intended to convey.
Nitiobroges is rejected by Gliick^in favour of Nitiohriges, which,
he argues, is formed from neith {' battle ') and hrig ('valour'). Strabo^
and Ptolemy 10 write Nirid^piyc?. On the analogy of Allohroges,
which form is found in Cicero, Caesar, Florus, Horace, Livy, Orosius,
Pliny, Pomponius Mela, Sallust, Suetonius, Tacitus, and Velleius
Paterculus, as well as in inscriptions, one might be inclined to
think that Nitiobroges was the true form. None of the MSS. of Caesar
have though 'AXXojSpLyc^ is found in some MSS. of
Allohriges,
Strabo and Ptolemy. At all events Caesar wrote Nitiobroges for ;

though the archetype has -iges in the place (vii, 7, § 2) where the
name first occurs, in the other three passages (vii, 31, § 5 ; 46, § 5 ;

1
pp. 138-9.
2 Query, MeUosedum ? Mellosutum is not in Holder's AU-celtischer Sprach-
schatz.
^ Philologus, xviii, 1861, p. 285. Heller suggests that Mellodunum may have
been written over MeUosedum by a copyist who wished to point out that
MeUosedum was situated on the island opposite the place where Mellodunum
(Melun) was afterwards built, on the hank of the river.
* Meclodunum would have produced not Melun, but Meillun.

' Op. cit., p. 227. « Ih., p. 228, n. 1.

' Unless, indeed, he dictated to a Celtic scribe who knew Latin, like his
interpreter, Troucillus. M. Vendryes asks us to bear in mind that the MSS. which
present the form -sedum, all have Metlo- [?] or Metio-. Certainly but may ;

not the scribes have known the Celtic form, which, as M. Vendryes says, Romans
mispronounced and miswrote ? I have adopted the form Metlosedum in quo-
tations (pp. 775, 778, 780, 782) to avoid puzzling the reader.
i»
«
p. 127, note. « iv, 2, § 2. Geogr., ii, 7, § 11.
 THE SPELLING OF CELTIC NAMES 847

75, § 3) the authority for -oges is overwhehning, and proper names

were occasionally misspelt in the passages of a manuscript where
they were first written.^
Orgetorix. De Saulcy —^ maintains that the correct form is Orge-

tirix or Orcetirix, found on coins of the Aedui.^ Mommsen ^

which is
says that none of the Orcitirixes whose names appear on coins can
be identified with the Helvetian chief of the same name and ;

M. dArbois de Jubainville ^ agrees with him but as the name is :

confessedly the same, I need not concern myself with Mommsen's

arguments.
Osismi, the reading which has the most MS. authority,^ is adopted
by Gluck.' Good MSS., he says, of Pliny ^ and several MSS. of
Orosius^ have the same. Mela^o writes Osismii, which is also found
in an inferior MS. of Caesar ^^ Strabo ^^ and Ptolemy ^^ 'Oo-tV/xioi.
;

—
Paemani. Caemam, which appears inySand in Orosius,!^ is adopted
by Meusel ^^ and tentatively by Holder.^^ But Famenne, the district
Avhich apparently reproduces the name of this people,^'^ might well
have been derived from Paemani}-^ Could it have come from Caemani ?
I suspect that this form is due to the preceding CaerososP
Petrocorii is found in all the good MSS. of Caesar ^o but ;

M. dArbois de Jubainville ^i regards Petrucorii as the true form.

Rauraci, the form found in the MSS.,^^ is rejected by Gliick,^^
Desjardins,24and Holder in favour of Raurici, which is found in an
inscription. The inscription may be right but who can tell ? ;

Inscriptions are not infallible, any more than coins. Some coins
show the form Massalia others Massilia.^^ Pliny ,^6 however, writes
;

Raurici, and Ptolemy ^7 'PavpLKot but Ammianus Marcellinus ^^ and

;

Orosius^s Rauraci.

^ Jahresb, d. Vereins zu Berlin, xx, 1894, p. 219.

philol. Cf. A. Holder,
Alt-celtischer Sprachschatz, i, 97-103.
^ Ann. de la Soc. num., 1867, p. 6.
^ Cat. des monn. gaul. de la Bihl. Nat., p. 110.

* Gesch. d. rdm. Milnzwesens, 1860,

p. 685, n. 71. M. Blanchet {Traite des monn.
gaul., pp. 85, 407) is inclined to agree with Mommsen ; but R. Forrer {Jahrb. d.
Gesellschaft f. lothring. Gesch., &c., xv, 1903, p. 130) defends the popular view.
^ Les noms gaulois chez Cesar, &c.,
p. 85.
® Ossismi is found in (p in B. G., iii,
9, § 10, and in a in vii, 75, § 4, where,
however, L has Ossisimi. '
p. 141.
* Nat. Hist., iv, 18
(32), § 107. The reading in Detlefsen's ed. is Ossismi,
with V. 1. Ossimi.
' ^» iii,
vi, 8, § 8. 2, § 23.
^^ Schneider's Caesar, i, 238.
'^ iv, 4, § 1. " Geogr., ii, 8, § 5. " vi, 7, § 14.
^^ Jahresb. d. philol. Vereins zu Berlin, xii, 1886, pp. 278-9.
'® Alt-celtischer Sprachschatz, i, 676. " See p. 404.
'^ See Miillenhoff, Deutsche Altertumshunde,
ii, 196-7.
^9 B. G., ii, 20 lb., vii,
4, § 10. 75, § 3.
21
Rev. celt., xi., 1890, p. 505.
22
In B. G., vi, 25, § 2, the a MSS. have Tauriaci ; but this is obviously wronff.
See Meusel's Lex. Caes., ii, 1626. 23
p 142.
'^*
Geogr. de la Gaule rom, ii, 463, n. 4.
2^ Walckenaer, Geogr.
des Gaules, i, 322, n. 3.
26 Nat. Hist., iv, 27 Qeogr., ii,
17 (31), § 106. 9, § 9.
28 XV, 11, 29
§ 11, &C. vi, 7, §5.
848 THE SPELLING OF CELTIC NAMES
Heller ^ defends the form Rauraci. He argues that if the name is
rightly derived from rauri (' lord '), the original form, on the analogy!
of Segontiac^. Dumnacus, and Diviciacws, must have been Rauriaci^
which would have been corrupted into Rauraci.
—
Redones. Like Rhenus, Rhodanus, raeda, and reno, the word is
spelled in some MSS. with, and in others without h,^ Mahn^ has
given reasons for believing that all five words should be aspirated.
But see Haedui.
Samarobriva. —
Desjardins* thinks that Samarahriva, which
appears on the famous milestone of Tongres, is the true form.
Santoni, not Santones, appears to have been the form adopted by
Caesar.^
—
Segusiavi. According to the best MSS., the people whom Caesar
mentions in B. G., i, 10, § 5, were the Sebusiani the people whom ;

he mentions in vii, 64, § 4, 75, § 2, were the Segusiavi. I have shown

elsewhere ^ that in all three passages Caesar was speaking of one and
the same tribe.
Nipperdey'' and Gliick,^ trusting to four inscriptions collected
by A. Bernard,^ maintain that Segusiavi is the true form and ;

Nipperdey adds that this form is supported by Strabo, Pliny,^^

and Ptolemy. The MSS. of Strabo ^^ exhibit a variety of forms,
among which are Seyocriavcov and '^rjyocnavwv, which agrees with
Nipperdey's reading. The MSS. of Ptolemy i^ likewise exhibit various
forms, among which are ^eyovcnavoL and ^iyovcnavot. Against
the authority of the inscriptions, says Schneider,^^ is to be set that of
a coin mentioned by Mionnet,^* which has the form Segusianus.
But when I refer to Mionnet's work, I find that the coin in question
bears the abbreviated legend segvsia only. On the other hand,
Cliick refers to a coin^^ which has segvsiavs (^ Segusiavus) and ;

the inscription on this coin is one of the four to which Bernard

refers. Schneider,^^ remarking that his business is to decide, not
what is the right form, but what Caesar wrote, adopts Segusiani,
'
because in B. G., vii, 64 and 75 the MSS. agree in exhibiting the
form Segusianos.^ But in both of these passages the MSS. have
Segusiavos and the inscriptions, coupled with the MSS. of Strabo,
;

Pliny, and Ptolemy, appear to prove that the true form is Segusiavi.
Sibusates, which is attested by p {B. G.,iu, 27, § 1), is to be pre-
—
ferred to Sibuzates the form adopted by most editors on the —
analogy of Cocosates, Elusates, and Tarusates.

1 Philologus, xvii, 18C1, p. 275. * B. G., ii, 34 ; vii, 75, § 4.

^ Etymologische Untersuchungen fiber geogr. Namen, 1859, pp. 27-32. Cf.
Philologus, xvii, 1861, p. 278.
* Geogr. de la Gaule ram., i, 137, n. 1.

^ Jahresb. d. philol. Vereins zu Berlin, xx, 1894,

p. 217.
» See p. 470. ' Caesar, p. 792. «
pp. 152-4.
^ Mem. de la Soc. des ant. de France, xviii, 1846,
pp. 343-8.
" Nat. Hist., iv, 18
(32), § 107. " Ed. Miiller and Diibner, p. 963.
2 Ed. Miiller, i, 217. '^ Caesar, ii, 532.

* Descr. des medailles antiques, i, 78.

^ A. Duchalais, De.srr.
des medailles gaul, 1846, p. 129, no. 377. Cf. A. Blanchet,
Traite des monn. gaul., p. 79. '« Caesar, i, 23 ii, 533.
;
 P

THE SPELLING OF CELTIC NAMES 840

Sotiatesis found in the/? MSS. in iii, 20, § 2, and in 20, § 3 in p ;

Sontiates in a throughout and in 20, § 3 in tt. In 21, § 1 Bfi have

Sociates, and in 21, § 2 /? has Sosates. Gliick^ pronounces for Son-
tiates. Pliny ,2 he says, has Sothiates various MSS. of Orosius ^
:

have Sotiates and Sociates of Dion Cassius * 'ATriaras. It is hard,

;

Gliick thinks, to decide between Sontiates and Sotiates, which, he

says, appears to be authorized by a coin ^ but he decides to follow
;

the a MSS., on the ground that there is no certainty that the name
on the coin is rightly deciphered. This is a mistake. Several coins
of the Sotiates exist, which are numbered 3604-3613 in Catalogue des
monnaies gauloises edited by MM. Muret and Chabouillet. In
Planche xi of the accompanying Atlas de monnaies gauloises, by
H. de la Tour, there is an illustration of one of these coins. On the
obverse it bears the legend rex adietvanvs, evidently the same —
name as that of Adiatunnus, the commander of the Sotiates, whom
Caesar mentions in B. G., iii, 22, §§1, 4 on the reverse sotiota.
:

Having regard to the coins as well as to the MSS., I think it probable

that Caesar wrote Sotiates.^
Suebi appears to have rather less MS. authority than Suevi ^ ;

but Strabo ^ writes '^6rj/3oL, and Ptolemy ^ '^vtj/Sol.

Sugambri, which has the most MS. authority, ^^ and is found in
Orosius,^^ is preferred by the modern editors to Sigamhri. The latter
form, however, and Sygambri occur in a MSS. Strabo ^^ writes
^ovyaixPpoi and Ptolemy 13 '^vya/Jc^poL,
Teuton! is to be preferred to the familiar Teutones}^
Treveri is preferred by Grliicki^ to Treviri, which, however, is
found, in some passages, in good MSS. Gliick appeals to the autho-
rity of Mela, Pliny, Ausonius, the Itinerary of Antonine, and many
inscriptions .1^ The MSS. of Tacitus also vary between Treveri and
Treviri. A. de Valois ^'^ argues in favour of Treviri, remarking that it
has the support of Strabo, of the Table, and of a coin belonging to
the time of Vespasian.
Turoni, not Turones, is the Caesarian form.^s
Venelli. —
Unelli is found in all the MSS. of the Commentaries
Pliny -0 has Venelli, and Ptolemy '^^ OueVeAAoi. Gliick ^^ prefers

1
pp. 154-5.
''
Nat. Hist., iv, 19 (33), § 108. Detlefsen reads Sottiates.
^ vi, 8, § 19. xxxix, 46, §2.
* ^ Duchalais, op. cit.,
p. 16, no. 30.
® See Meusel's remarks in Jahresb. d. philol. Vereins zu Berlin, xii,
1886, p. 279.
' See Meusel's Caesar, pp. 24, 37, 76-8, 80, 86-7, 139-40, 149.
« iv, 3, §4. 9 Geogr.,u, 11,
§ 11.
'« B. G., iv,
16, § 2 ; 18, §§ 2, 4 19, § 4 ; vi, 35, § 5.
;
''
vi, 9, § 1.
1- iv, 13 Geogr.,
3, §4. ii, 11, § 6.
'* Jahresb. d. philol.
Vereins zu Berlin, xx, 1894, p. 217. ^^
pp. 155-6.
'« Orelli,
6718, 5898, 7392, 6838, 7254, &c. J. Gruter, Inscr. ant., &c., 1707,
;

vol. i, Ixiv, 9 ;cccclxxxii, 5, &c. " Notitia Galliarum, pp. 560-1.

'8 Jahresb. d. philol.
Vereiiis zu Berlin, xx, 1894, p. 217.
'9 Meusel in his critical
edition says that, according to h\ Ramorino, in one
passage {B. G., iii, 11, § 4), in /, the original copyist wrote VencUos, for which
Unellos was substituted by a later hand.
-0 Nat. Hist., iv, 18
(32), § 107.
^' Geogr.,
ii, 8, § 2. " pp. 165-7.
1093 3 I
850 THE SPELLING OF CELTIC NAMES
Venelli, believing that the word comes from vena, which he regards
as the root of Veneti but, he admits, there is no evidence for explain-
;

ing either name. He also points out, in support of the form Venelli,
that Ubisci is found by mistake for Vivisci, Uridovix in one MS. for
Viridovix, and in Ptolemy (ii, 3, § 11) OvpoXdvcov for OvepoXdixtov,
which form is supported by the Itinerary of Antonine'^ and by
Tacitus.2
—
Veliocasses. Gliick,^ who cites an inscription in Orelli (No. 6991),
and the editors adopt this form. The readings in the MSS. of the
Commentaries are Velocasses (ii, 4, § 9), Belliocasses in a, and Vellio-
casses in jS (vii, 75, § 3). Gallic coins bear the legend veliocaoi.*
Vellavii. the reading of a, is adopted by Gliick ^ and the modern
editors. Gliick refers to an inscription ^ but in another one finds ;
"^

Vellavorum. The /? MSS. have Vellahiis.

Vercassivellaunus, the name of the chief who led the assault
on Mont Rea at Alesia, is preferred by Gliick ^ to Vergassillaunus
and other variants. He cites, in support of this reading, the name
Cassivellaunus, and compares with these two the similarly allied
Cingetorix and Vercingetorix.
Vercingetorix is the Latin form of Vercingetorixs.^
Viromandui is preferred by Gliick ^^ to Veromandui, which has
equal MS. authority .^^ The best MSS. of PHny i- have Viro?nandui,
and Ptolemy ^^ has Outpo/xai/Sue? while Orosius ^* and the Itinerary ;

of Antonine^^ have Veromandui. Gliick relies upon two inscriptions ^^

and upon the analogy of many compound names, for instance, Viro-
marus, Viromanus, Virovisca, Virovius, and Virovedrum and he ;

accounts for the form Veromandui by the fact that i and e were
often interchanged. M. d'Arbois de Jubainville adds that the name
is spelled with an i by Gregory of Tours ^^ and on a Merovingian
coin.i^ It is impossible, Gliick says, to decide whether the i is long
or short.

1 Ed. Wesseling, pp. 471, 476, 479. ^ ^^^„,^ xiv, 33.

^
pp. 101-2. " Desjardins, ii, 437, n. 2, 461, n. 4, 462.
^
p. 164. ® Hist, de V Acad, des inscr., xxv, 1759, p. 148.
' Orelli, no. 5221. 8
p. 174.
^"
^ Diet. arch, de la Gaule, i, no. 69. pp. 184-7.
" B. G., ii, 4, § 9 ; 16, § 2 ; 23, § 3.
'' Nat. Hist., iv, 17 (31), § 100.
13 Geogr., ii, 9, § 16. '* vi, 7, § 14. ''
Ed. Wesseling, p. 379.
^® Gruter, vol. i, ccclxxv, 3 ; J.de Wal, Mythologiae septentrionalis monumenta
epigr. Lot., 1847, p. 226, cccvii ; A. de Boisj^ieu, Ivscr. ant. de Lyon, 1840-54,
p. 200.
^^
In gloria martyrorum, p. 536, 1. 13, ed. B. Kruscli.
" Bibl. de VEcole des Charted, xxvi, 463, no. 690.
 —
:

ADDENDA
Paue 61, note 3. Meusel {Jahresb. d. philol. Vereind za Berlin, xxxvii,
1911, pp. 105-6) urges that the words vulgo tolls castris testamenta obsig-
nabantur {B. G., i, 39, § 4) may just as probably have been borrowed by the
'
Pseudo-Caesar' from Livy or Florus, as by Livy and Florus from Caesar.
1 have not argued that the relevant passage in Florus proves that the words
vidgo . .obsignabantur were written by Caesar ; but I find it impossible
.

to believe that they were a pure invention.

Pages 78-9. The standard-bearer
'
was lost.' Notwithstanding the . . .

arguments of Th. Steinwender {Rhein. Mus., N.F., Ixv, 1910, p. 134, n. 2),
I adhere with C. E. C. Schneider [Caesar, i, 181) to the opinion that the
standard mentioned in B. G., ii, 25, § 1, whether it was one of the three
manipular standards or not, was the standard of the cohort. Caesar's
meaning is unmistakable.
Page 91. '
The troops clambered short sword.' M. Jullian [Hist,
. . .

de la Gaule, iii, 298), referring to Dion Cassius (xxxix, 43, § 1), observes that
the Veneti had neglected to provide themselves with missile weapons.
But as we may be sure that the Romans had not been guilty of this neglect,
and as they probably had artillery as well (cf. B. G., iv, 25, § 1), I doubt
whether in any case the Veneti could have won the battle.
Page 112. All that night
'
plain of Hesbaye.' If Atuatuca was on
. . .

the site of Tongres (see pp. 371-83), this statement, in so far as it relates
to the plateau of Herve, is incorrect.
Page 206, note 4. If, as A. Klotz believes [Caesar studien, p. 145), the
description of Britain in B. G., v, 12-4, was derived from Posidonius,
the iron currency-bars
'
[taleae ferreae) were presumably mentioned by
'

Posidonius's informant but see Rice Holmes, Ancient Britain, p. 499,

;

n. 2, and Woch. f. Mass. Philologie, June 20, 1910.

Page 221. 'Signer Ferrero attributes ... to deal with Ariovistus.'
B. G., i, 36, § 4, from which we learn that Ariovistus complained that
'
Caesar was doing him a serious injury, for his coming depreciated the
tribute ', suggests, if it does not prove, that the Aedui, from the moment
of Caesar's arrival, relied upon his aiding them against Ariovistus. Other-
wise they would hardly have dared to withhold tribute which Ariovistus
had hitherto exacted from them.
Page 241. Now it is not true ... to furnish.' I am rather inclined to
'

suspect from B. G., ii, 15, § 3, 16, §§ 3-4, that the Nervii and their allies
the Atrebates, Viromandui, and Atuatuci had taken no part in the earlier —
stage of the campaign ; and I do not feel sure that the more distant tribes
sent their contingents.
Pages 243-4. Anyhow Veith '
heavy loss.' Some of the prisoners
. . .

of whom Veith speaks may have been taken from the relieving army in
the course of the pursuit.
Pages 270-1. Although the skulls
'
contrasted types.' M. Manouvrier
. . .

[Bull, et mem. de la Sac. d'anlhr., 5^ ser., v, 1904, p. 119), differing from

M. Verneau, believes that the negroid characteristics of the skeletons
found in theGrotte des Enfants were merely 'des caracteres individuels'
3 12
852 ADDENDA
and he adds that des correlations anatomiques peuvent concourir a
'

reaUser chez un blanc ce qui se produit d' ordinaire, et suivant des degres
tres variables aussi, chez les negres '.

Pages 274-5. ' like the Celtic . advanced guard.' M. Jullian {Rev.
. .

des etudes anc, 1910, pp. 302-3) finds fault with M. Dechelette for
xii,
maintaining that the Celts first invaded Gaul in the Hallstatt Period,
and thinks that he has no right to ' tirer une conclusion historique ou
ethnographique de premisses archeologiques '. The conclusion is merely
tentative ; but the physical resemblance of some of the invaders of whom
M. Dechelette was thinking, is remarkable.
Page 282, note 2. Sir John Rhys {Notes on the Coligny Calendar, 1910,
p. 34, n. 1 \Proc. Brit. Acad., vol. iv]) writes,
—
Hardly a neater piece of
'

evidence could have been produced than petru-decameto to prove that the
Coligny Calendar is in a Celtic language other than Gaulish, though Professor
Loth has brought it forward to prove the contrary. . Petru- is akin to . .

Welsh pedwar "four " . Latin qaattuor, that is to say, its initial p represents
. .

an earlier qit, as in Latin. How this could help " pour repousser la theorie
du maintien de 2? ou de g' dans la langue de Coligny " with its eqvos and
QVIMON does not appear dnd da to iSeqiiana being treated as representing
;

an older Seko-uanii, that theory is familiar enough to me, as I tried to

M ork it years ago I think that my learned friend will find it disappointing.
;

From his interpretation of the Gelignieux inscription he draws the following

conclusion :
—
" II semble bien certain que le testateur de Gelignieux faisait
usage du meme calendrier qu'a Coligny." But such certainty as M. Loth
found there has been converted into doubt by what has been explained
above as to the Coligny entries in point.'
Page 305, note 5. Fresh tables, prepared by M. A. Bertillon, but not
yet published, are referred to by M. P. Raymond {Bidl. de la Sac, d' etude
des sc. nat. de Nimes, xxx, 1902 [1903], pp. 45, 52) but my experience ;

leads me to doubt whether the results are seriously dififerent from those
of MM. Broca and Collignon.
Pages 334-5. In 1900 Dr. Beddoe wrote to me, I am much inclined '

to adhere to your view as to the difference of the hair in the Saxon and the
Gael.'
Page 417, line 10 from foot of text :
for Corchester read Corbridge.
Page 596. '
be objected
It might, indeed, have disappeared.' I ought
. . .

to have remarked that the German host itself fought in groups, tribal —
phalanxes.
Pages 629-30 (IX). I am not sure that the argument in this paragraph
is quite clear. The difference between StoffeFs view and that of M. Jullian
is simply that, according to Stoffel, the Boi and Tulingi were separated,
on the march, from the Helvetii by a considerable distance ; according to
]M. Jullian, they were not so separated, but were left to guard the wagons
when the Helvetii went into action. M. Jullian's view is not open to the
first objection which I have brought against Stoffel's.
Page 801, note 1. One more attempt has just been made to identify
the site of the combat that immediately preceded the blockade of Alesia.
Lieutenant- Colonel Frocard {Pro Alesia, Nos. 53-4, 1910, p. 766) places it
between Orvillc and Veronnes, about four miles north-east of Til-Chatel.
 ; ;

INDEX
Abrincatui, 469, 499. Caesar, 182 ; two legions winter in
Acco, 128-9, 198. their country (52-51 B.C.), 183;
Acies, 587-8. See Order of Batllo, Aeduan territory defined, 351-3;
Adiatunnus, 22. orthography, 843-4.
Adige, 37. Agedincum, six legions quartered at
Admagetobriga. See Magetobriga. (53-52 B.C.), 128, 134; Caesar
Adour, 93. concentrates legions near (52
Adra, 658. B. c), 136, 737-8 ; Caesar garrisons,
Aedui, their alliance with Rome, 3 ;
when marching to relieve Gorgo-
Vergobret of, forbidden to cross bina, 136 ; Labienus marches from,
frontier, 21 ; hegemony of, 24-5, against Senones and Parisii, 161 ;

517-9 ; rivalry with Arverni and returns to, and thence marches to
Sequani, defeated by Ariovistus, rejoin Caesar, 164,786-7; question of
37-8, 554 ; beg Caesar for aid its site, 353-4; orthography,839-40.
against Helvetii, 49 ; their cavalry Agger, built in siege of chief strong-
with Caesar beaten by Helvetii, 50 ; hold of Atuatuci, 81 in siege of ;

fail to supply Caesar with corn, 51 ; Avaricum, 140-1, 143-6, 745-6,

ask that Boi may be allowed to 748-51 ; in siege of Uxellodunum,
settle in their country, 57, 208-9 ; 192, 834 ; note on its construction,
Caesar negotiates on their behalf 599-607.
with Ariovistus, 58-9, 63 ; supply Aisne, Caesar's operations on (57 B.C.),
Caesar with corn during campaign 71-4, 659-68 in 51 B.C., 185, 187,
;

against Ariovistus, 40, 642, 644 826-30.

contingent of, under Diviciacus, Aix, 37.
ravage lands of Bellovaci, 71 ; Aix-la-Chapelle, 382.
Caesar treats with distinction, 75, Alaise, 355, 357-62.
102, 106 ; friendly to Caesar, 86, 95, Alauda, 653 n. 4, 802-3.
106, 119, 198; intercede for Albi, 135.
.Senones, 122 ; keep aloof at first Alene, 54.
from rebellion of Vercingetorix, Alesia, fortified and provisioned by
send troops to assist Bituriges, 133 ;
Vercingetorix, 165, 794 ; Vercinge-
Caesar demands supplies from, 136, torix marches from, to intercept
141 ; ask Caesar to settle dispute Caesar, 167, 794 ; Vercingetorix
between Cotus and Convictolitavis, retreats to, 169, 801 ; Vercingetorix
148 ; Caesar demands contingent blockaded in, by Caesar, 170-6
from, 148 ; signs of their impend- Caesar's earthworks at, 171-3, 807-
ing defection, 151-4 : Caesar inter- 12; final struggle at, 176-80;
cepts mutinous contingent, 153, credibility of Caesar's narrative of
767-8 ; contingent joins in attack operations at, 243-4 ; question of
on Gergovia, 156, 158, 247-8, 766-7; its site, 354-63 ; questions relating
Aedui definitely join rebellion of to Caesar's narrative of operations,
Vercingetorix, 159, 198 ; Caesar 804-20: orthography, 840. See
crosses Loire in spite of, 160-1, also 623, 851.
769-71, 774 ; contingent deserts Alise-Ste-Reine, 354. See Alesia.
Caesar, 160 ; Aedui claim direction Allia, battle of the, 1, 543.
of rebellion, but are snubbed, 164- Allier, 131 ; bridges over, destroyed
5 ; levy of, sent by Vercingetorix by Vercingetorix, 149 ; Caesar
against Allobroges, 166 ; army crosses, 149, 751-5 ; he marches
raised for relief of Alesia musters in down valley to intercept Aedui,
of,
their country, 175 ; probably trea- 153 ; recrosses, 159, 755 ; Aedui
cherous to Vercingetorix, 178, 815- try to hem him in between, and
6, 821-4 ; return to allegiance to Loire, 160, 769-71.
854 INDEX
Allobroges, aid Salyes against Romans, Arene Candide. 286.
3 ; rebel, 38 directed by Caesar to
; Ariovistus, invades Caul on invitation
feed remnant of Helvetii, 57 repel ; of Sequani, defeats Aedui, annexes
emissaries of Vercingetorix, 165-6 ;
a third of Sequanian territory,
their territory defined, 363-5, 501-2. defeats Aedui and Sequani and
'Alpine' race, 264, 276-7. their respective allies, 37-8 re- ;

Alps, 1-3 ; crossed by Caesar, 46, 48 ceives a title from the Senate, 40 ;

by what pass did he cross in 58 Celtae beg Caesar's aid against,

B.C.? 615-6. 57 Caesar's attempts to nego-
;

Alsace, 5, 38, 58, 637-8, 641-5, 648-51. tiate with, 58-9, 62-4 Caesar's ;

Alfero die, 738-40. campaign against, 60-7, 636-57 ;

Ambacti, 514-5. date of his invasion of Gaul, 553-

Ambarri, 365-6. 4 ; credibility of Caesar's notices
Amber, 8. of, 229-30. See also 594, 596-9,
Ambirmi, submit to Caesar, 75 ; their 851.
territory defined, 366. See also 396. Arman9on, 157, 794-7.
Ambibareti. See Ambivareti. Armecy, 54-5, 626-7.
Ambibarii, 366-7, 499. Armour, of Caesar's legionaries, 43.
Ambiliati, 367. 584-5.
Ambiorix. 22 attacks Atuatuca,
; Army, Caesar's, 42-4, 195, 559-99.
107 his interview with Roman
; Arretium, 2, 253.
deputies, 107-8 annihilates garri-
; Arroux, 53.
son of Atuatuca, 110-2; besieges Art, palaeolithic, 7 of Bronze Age,
;

Q. Cicero, 112-5; his guerrilla 10 of La Tene, 15.

;

warfare, 121-8, 188-9 struck pre- ; Artillery, Caesar's, 43, 44 n. 2, 582-3,

maturely, 198. See also 504. in siege of Atuatucan stronghold,
Ambivareti, 183 their territory, 367-
; 81 ; in siege of Avaricum, 140, 146 ;

8 orthography, 840.
; used by Fabius against Vercinge-
Ambivari ti, 368-70. See also 69 1 695, , torix, 154 in operations at Alesia,
;

697, 700. 177.

Ambleteuse, 435, 438. Artois, 94.
Amiens, 75. See Samarobriva. Arverni, help Salyes against Rome, 3 ;

Andebrogius, 70. their power broken, 4 their hege- ;

Andernach, 100, 706-8, 710. mony in Gaul, 24-5, 518-9 hire ;

Andes, 86 their territory defined,

; aid of Ariovistus, their hegemony
370 orthography, 840.
; usurped by Sequani, 37 inactivity ;

Angers, 84. of (58-53 B.C.), 131, 198; leading

Animal- worship, 30-1. men among, expel Vercingetorix
Anjou, 86. from Gergovia, expelled in turn by
Anthropology, methods of, 261-3. him, 132 Aedui jealous of, 133,
;

Anti-Roman party in Gaul, 25, 52, 57, 152 Caesar ravages their country,
;

106, 151, 520-3. 135 submit after fall of Alesia,

;

Antistius Reginus, C, defends camp 182-3 a renegade Arvernian be-

;

on Mont Rca, 177. 819. trays Lucterius, 183 their territory;

Apollo, 28-9. defined, 371. See Bituitus, Celtillus,

Appian, authority of, 215, 217. Gergovia, Vercingetorix.
Aquileia, 42. Arvii, 416-7.
Aquitani, 5, 12-3 ; campaign of asia, 295 n. 4.
Crassus against, 92-4, 688-9 ; credi- Asinius Pollio, 211-2.
bility of Caesar's statement of Asnieres, 800-1.
reasons for attacking, 227 hired ; Atrebates, defeated by Caesar at
cavalry of, assist Vercingetorix, 147; Neuf-Mesnil, 76, 78 join Belgic ;

ethnology, 5, 12-3, 288, 290-1, confederacy against Caesar (51 b. c),

297-8, 301-2. See also 194 n. 1, 198. 185; their territory defined, 371.
Archers, in Caesar's army, 42, 72-3 ;
See Commius.
at Atuatuca, 111 employed by ; Atuatuca, Sabinus and Cotta quar-
Vercingetorix, 144, 147, 176. tered at (54 B.C.), 105; camp
Ardeche, 135. attacked by Ambiorix, 107 Sab- ;

Ardennes, 124, 128, 734-5, inus' s virtually annihilated

force
Aremorica, 301. near, 110-2; Q. Cicero left in
Aremorican states, 370-1 ; ortho- command at (53 b. c. ), 124 ; at-
graphy, 840. tacked by Sugambri, 126-7 ; ques-
 ;;

INDEX 855

tion of its site, 371-83 ortho- ; Baculus. See Sextius.

graphy, 840. Baggage, 44 disposal of Caesar's, in
;

Atuattici, 76 ; their stronghold cap- battle with Helvetii, 54, 628 before ;

tured by Caesar, 80-2 credibility ; battle with Nervii, 76 in cavalry ;

of Caesar's narrative, 230-1, 242 combat before blockade of Alesia,

persuaded by Ambiorix to join in 167-8 in Sabinus's retreat from
;

attacking Q. Cicero, 1 12 defeated ; Atuatuca, 727-8.

by Caesar, 117-8 remain in arms, ; Baggage-drivers, 44, 78, 155.
121 ;did not join in relieving Baiocasses, 424, 444.
Alesia, 174 their territory defined,
; Balearic isles, 42.
384-7 question of site of their
; Ballista, 582.
chief stronghold, 387-93 ; ortho- Balventius, 578.
graphy, 841. Bar-sur-Aube, 56.
Atuatucorum oppidum. See Atu- Bards, 19, 131.
atuci. Basilus. See Minucius Basilus.
Augustodunum, 398. Basques, 11, 289-301, 336.
Augustus, 194 n. 1. Batavi, 453, 691-4.
Aulerci Branno vices, 31, 393; meaning Baumes-Chaudes, 8, 272, 275, 277,
of name, 31 ; orthography oi Aulerci, 298, 336.
841. Beaujolais, 50.
Aulerci Eburo vices, 33-4, 92, 96 n. 1, Beauvais, 71, 105, 401.
393-4. Belenus, 30.
Aulus Hirtius, author of the Eighth Belfort, 62, 639-45, 648.
Commentary, 184, 823-5 his re- ; Belgae, value of Caesar's grouping
marks on Caesar's Commentaries, of, 5, 12-3; ethnology of, 12,
203, 209 ; did he help Caesar to 257, 259, 304-5, 308-9, 311, 315,
write them ? 209-10 credibility ; 318-25 ; Caesar's first campaign
of his description of Uxellodunum, against, 69-85, 657-77 ; char-
492-3. acter of their resistance, 104, 195,
Auray, 90, 680, 684-5. 198 ; legions quartered in their
Aurunculeius Cotta, placed in joint country (54-53 B.C.), 105, 371-84;
command at Atuatuca, 105 urges ; Commius intrigues among, 129 ;
Sabinus to hold Atuatuca against hold aloof at first from rebellion of
Ambiorix, 108-9 his splendid con-
; Vercingetorix, 135 ; Caesar's final
duct in action, 110-1 ; killed, 111 ; campaign against, 185-8, 826-30 ;
was he subordinate to Sabinus ? 726. were Treveri a Belgic tribe ? 394-5 ;
Ausci, 394. motive of Belgae for conspiring in
Autessiodurum, 472-3. 57 B. c, 657 strength of their host,
;

Auvergne, ethnology of, 309, 314-5. 241-2 ; where did they muster ?

Auxerre, 426, 472-3, 787. 658-9. See also Atuatuci, Belgium,

Auxiliaries, in Caesar's army, 42 ; in Bellovaci, Eburones, Morini, Men-
battle with Helvetii, 54 ; in opera- apii. Nervii, Remi, &c.
tions against Ariovistus, 65-6, Belgium, 5, 395-7.
652-3 ; relieve Bibrax, 72 ; in Bellovaci, Caesar sends Aeduan con-
operations on Aisne, 73 in battle ; tingent to harry their country
with Nervii, 76 in final operations
; (57 B.C.), 71, 74; surrender Bratu-
against Bituriges, 185 ; and against spantium, 75 two legions quar-
;

Bellovaci, 187-8 ; their role in the tered among (54 b. c. ), 105 threaten ;

Roman army, 590, 594-7. See Labienus (52 send a

B.C.), 162;
Aedui, Archers, Cavalry, Germans, small contingent to join in relief of
Numidians, Slingers, Spanish. Vercingetorix, 174; Caesar's cam-
Auzon, 150-1, 155, 760-1. paign against (51 b. c), 185-8,
Auzon (or Aizon), 53. 826-30 credibility of Caesar's state-
;

Avantici, 405, 502. ment of their numbers, 241-2; their

Avaricum, 19 ; €aesar marches for, territory defined, 397-8.
138 ; Bituriges resolve to defend, Bellovesus, 542, 544.
against advice of Vercingetorix, 139 Berri, 133, 135, 139. See Bituriges.
siege and capture of, 140-6, 742-51 ; Berry-au-Bac, 71, 660-8.
losses at, repaired by Vercingetorix, Besan^on. See Vesontio.
147 ; occupied by Romans, 147 ; Beuvray, Mont. See Bibracte.
agger at, 599-()07, 745-6, 749-55. Bibracte, description of, 19-20 the ;

Avigliana, 430-1. goddess, Bibracte, 27 ; Caesar

856 INDEX
marches towards Bibracte, Helvetii Brian^on, 48, 405, 615-6.
try to cut him off from, 53 ; Hel- Bridges, of Gauls, 16 ; bridge at
vetii defeated near, 54-6, 621-8 Geneva destroyed by Caesar, 46 ;

Caesar's hostages sent to, by Caesar bridges Sa6ne,49 ; he crosses

Eporedorix and Viridomarus, 160 ; bridge over Aisne, 71, 659-68 ;

general assembly at, elect Vercinge- builds a bridge over Rhine, 100 ;

torix commander-in-chief, 164-5 ;

builds a second bridge, 123 sites ;

Caesar winters at (52-51 b. c), 183 ;

of these bridges. 706-10 how first ;

Caesar marches from, against Bitur- bridge over Rhine was built, 711-
iges, and returns, 184-5 ; question 24 ;Caesar bridges rivers in coun-
of its site, 398 ; did Vercingetorix try of Menapii, 122 bridge over ;

march from, against Caesar before Loire at Cenabum, 137 bridges ;

the blockade of Alesia ? 792, 794. over Allier destroyed by Vercinge-

Bibrax, attacked by Belgae, relieved torix, 149 Caesar repairs one of
;

by Caesar, 72 ;credibility of them, 149 he crosses Allier by,

;

Caesar's narrative of its relief, 159, 755 bridges at Lutetia de-

;

241 n. 6 ;
question of its site, 398- stroyed by Camulogenus, 162; how
400. Caesar generally built bridges. 612 ;

Bievre, river, 776-8. bridge by which Dumnacus crossed

Bigerriones, 400. Loire, 831.
Bituitus, 3, 24, 545. hriga, 282-3.
Biturlges, influence of Dumnorix with, Brigantio. See Briangon.
52 join rebellion of Vercingetorix,
;
Britain, trade of Veneti with, 18, 87 ;

134 ; Vercingetorix orders destruc- hegemony of Diviciacus in, 24

tion of villages in their country, Caesar said to be contemplating
they persuade him to spare Avari- invasion of, 87 his objects in in-
;

cura, 139 ; a legion quartered vading, 102 ; alleged inconsistency

among (52-51 B.C.), 183; their of some of his statements about,
rebellion in 51 B.C. crushed, 184-5 ; 206 Druidism derived from ? 32-
;

their territory defined, 400. See 3, 523-5. See also 851.

Avaricum, Noviodunum. Brittany, Crassus receives submission
Bituriges Vivisci, 400. 84 tribes rebel. 86-92
of tribes of, ; ;

Blanc-Nez, Cape, 103, 273, 432. they contemplate an attack on

Boar, on coins, 18 on standards,
; Roscius, 118; they rebel in 51
30-1. B.C., 189-90; ethnology of, 309-10,
Bodiontici, 502. 314.
Boduognatus, 78. Bronze Age, 9-11.
Boi (of Cisalpine Gaul), 2. Brutus, D., commands in sea-fight
Boi, join Helvetian emigration, 46 ;
against Veneti, 90-1 left in com- ;

in battle near Bibracte, 55, 233-4, mand by Caesar in country of

621, 626-7, 629-31; survivors Arverni (52 b. c), 136; his rank.
allowed by Caesar to settle in 564-5.
Aeduan territory, 57, 208-9 ; their Brythonic, 12. See Gallo-Brythonic.
stronghold, Gorgobina, besieged by Bussy. See Montague de Bussy.
Vercingetorix, 1 36 Caesar marches
;

to relieve, 137-8 ; send supplies to Cadurci, 165 n. 4, 402-3.

Caesar during siege of Avaricum, Caerosi, their nationality, 332, 338-
141 ; their territory conjecturally 40 ; their territory defined, 403 ;

defined, 425-6. See Gorgobina. orthography, 841.

Boia, 740-2. Caesar, Gains Julius, busts of, xix-
Bonn, 97, 706-9. xxiv ; date of birth, 557 his grouping ;

Borvo, 27, 29, 31. of Gallic peoples, 12-3, 257 consul, ;

Boulogne, 103, 432-8. appointed Governor of Gaul, 40 his ;

Bourbon-Lancy, 774, 787. person and character, 41-2 his ;

Boussieres, 76, 675. army, 42-4, 559-99 his intentions, ;

Braccae, 584. 44-5, 558 hastens to Geneva. 46

; ;

Brannovlces. See Aulerci. negotiates with Helvetii and pre-

Bratuspantium, 75 ; question of its vents them from crossing Rhone,
site, 400-2. 46-7, 614-5 goes back to Cisal-
;

Brenne, 169, 813. pine Gaul and returns with rein-

Brenner Pass, 37. forcements, 48-9, 615-6 defeats ;

Breteuil, 75. See Bratuspantium. Tigurini, 49, 617-9; Helvetii at^

 ; ;;

INDEX 857

tempt to negotiate with, but re- Gaul, 134 ; rescues Province, out-
ject his terms, 49-50 ; campaigns manoeuvres Vercingetorix, and re-
against and defeats Helvetii, 50-7, joins legions, 135-6, 736, 737-8;
016-34 ; his treatment of fugitive tnarches to relieve Gorgobina, cap-
Helvetii, 56-7 ; congratulated by tures Vellaunodunum, Cenabum,
deputies from Celtican Gaul, who and Noviodunum, 137-8, 740
solicit his aid against Ariovistus, besieges and captures Avaricum,
57-8, 634 ; attempts to negotiate 140-6, 742-51 outwitted by Ver-
;

with Ariovistus, 58-9, 62-4 ; seizes cingetorix, 141-2, 744-5 ; secures

Vesontio, 60 ; allays panic in his election of Convictolitavis as Vergo-
army at Vesontio, 60-2 campaign ; bret, 147-8, 154, 528 n. 2 ; sends
against Ariovistus, 60-7, 636-57 ;
Labienus against Parisii and Senones
resolves to conquer Gaul and re- and marches against Gergovia, 148,
turns to Italy, 67-8 ; results of his 751-5 establishes a magazine at
;

first campaign, 69 returns to Gaul

; Noviodunum (Nevers), 148-9 first ;

and receives submission of Remi, 70 operations at Gergovia, 149-51,

campaign of 57 b. c. against Belgae, 756-7 intercepts Aeduan contin-
;

69-85, 657-77 sends Galba into

; gent, 153-4, 767-8 ; attempts in
the Valais, 82 rejoicings at Rome
; vain to take Gergovia by coup de
over his victories, 85 goes on ;
main, 155-8, 245-9, 758-67 errone- ;

political tour to Illyricum, 86 pre- ; ously said to have lost his sword
pares for campaign against Veneti, in action, 767 ; marches to rejoin
87 ; conference at Luca, 88 cam- ; Labienus, 159-61, 769-74; Labi-
paign against Veneti, 88-91, 679- enus hears rumours that he has been
88 ; campaign against Morini, 94, forced to retreat to Province, 162 ;

689 ; returns from Cisalpine to rejoined by Labienus, 164, 785-

Transalpine Gaul, to deal with 90 enlists German cavalry, 166 ;
;

Usipetes and Tencteri, 95-6 ; marches to succour Province, de-

campaign against Usipetes and feats Vercingetorix in cavalry com-
Tencteri, 96-9, 689-706; bridges bat, and forces him to retreat to
Rhine, punishes Sugambri and Alesia, 167-9, 791-801 ; operations
returns to Gaul, 99-100, 706-24; in- at Alesia, 169-80, 804-20 ; receives
vasions of Britain, correspondence surrender of Vercingetorix, 180,
with Cicero, 101-2 has Dumnorix; 820 effects of victory at Alesia,
;

put to death, 102-3 quarters ; 184 disperses Bituriges and Car-

;

legions for winter of 54-53 b. c, 105, nutes (51 B.C.), 184-5; campaign
371-84 ;
promotes adherents to against Bellovaci, 185-8, 826-30;
power, sends Plancus to avenge as- ravages lands of Eburones, 188 ;

sassination of Tasgetius, 106, 521-3 executes Cotuatus, 191, 831-2 cap- ;

humbles Indutiomarus, 106-7 ; tures Uxellodunum and punishes

Ambiorix professes gratitude to- garrison, 191-3 ; conciliates con-
wards, 108 ; praises bravery of quered Gauls, 193 ; why he suc-
troops at Atuatuca, 111 relieves ; ceeded, 194-9 ; date of composition
Q. Cicero, 115-8 spends winter of
; and publication of Commentaries,
54-53 B. c, in Gaul, warns malcon- 201-9 authorship of Commewtone^s,
;

tents, 119 ; eulogizes generalship 209-10 ; credibility of Commen-

of Labienus, 120 ; borrows a legion taries, 211-56; want of precision
from Pompey and raises two others, in geographical statements, 344
120-1, 731-2 punishes Nervii
;
how many legions did he receive
and forces Senones and Carnutes to from Senate and Roman People ?
submit, 121-2 ; crushes Menapii, 557 his marches, 635 ;
; did he
122 ; crosses Rhine again, but confound Scheldt with Sambre ?
returns unsuccessful to Gaul, 123 ; 734-5 ; how many legions had he
campaign against Eburones, 124- at Alesia ? 802-4 duration of his
;

5 ; invites neighbouring tribes to proconsulship, 832-4.

harry them, 125 ; gently rebukes Caesar,. L., 166.
Q. Cicero for rashness at Atuatuca, Caleti, 185 ; their territory defined,
ravages lands of Eburones, 127-8 ; 404 ; orthography, 841.
distributes legions for winter of Calones (drivers and officers' servants),
53-52 B.C., executes Acco, 128; 44, 78-9, 109, 126-7, 155.
Gallic chiefs conspire asainst, 129- Calvados, 89.
30, 736; returns from Italy to Cambrai, 75.
858 INDEX
Camps, by what method some of Basilus, 124 against Eburones,
;

Caesar's were discovered by Colonel 128 ; ravage country of Arverni,

Stofifel,xxv~xxvii fortifications ; 135 ; in combat at Noviodunum,
of, 586-7 Caesar's camp on the
; 138; at Gergovia, 151-3, 155-6,
Aisne, 71-2, 659-68 Sabinus's in ; 158 ford Loire, 160-1
; in battle ;

country of Venelli, 92, 688 winter ; of Lutecia, 164 Caesar enlists ;

camps of 54-53 B.C., 105, 371-84; German, 137, 166 in combat ;

Caesar's at Gergovia, 149, 151, before blockade of Alesia, 168-9 ;

153-5, 756-7; at Alesia, 170-3, at Alesia, 170-1, 176, 179-80, 806,

805, 812-3, 815-6 ; in second cam- 812 ; disperse Carnutes, 185 in ;

paign against Bellovaci, 826-30. campaign against Bellovaci (51

Camulogenus, commands Parisii and B. c), 186-8. See also Aedui, Ger-
Senones in campaign against Labi- man, Spanish, Sugambri, Tencteri,
enus, 161-3, 775-85. Treveri, Usipetes.
Caninius Rebilus, C, defends camp on Cavares, 501-2.
Mont Rea, 177; transferred from Cavarinus, 119, 122, 504.
country of Ruteni to that of Pic- Cavillonum, 841. See Chalon.
tones, 188 forces Dumnacus to
; Cebenna, 841.
raise siege drives
of Lemonum, Celtae, Caesar's grouping of, 5, 12-3,
Drappes and Lucterius into Uxello- 257 ; culture of, 15-7 enfeebled ;

dunum, 189 ; blockades Uxello- by contact with Roman civilization,

dunum, 189-90. See also 253. 20 deputies from central tribes
;

Cannae, 595. congratulate Caesar on victory over

Canstadt race, 6. Helvetii, 57, 634 ; certain chiefs of,
Capdenac, 483-4. egg on Belgae to rebel, 69 ; mostly
Carcase, 93. support Vercingetorix, 132-3 in- ;

Carinthia, 36. effectual nature of their resistance

Carniites, assassinate Tasgetius, 106 ; to Caesar, 198 ethnology of, 303- ;

rebel against Caesar (53 B.C.), 121- 21, 336-7, 852.

2 ; Caesar investigates origin of Celtiberi, 294-5.
rebellion, 128 Carnutes strike first
; Celtillus, 24, 131, 504, 514.
blow in rebellion of 52 B.C., 130; Celts, 1 invade Gaul, 11-2;
; their
Caesar captures their chief town, language, 12, 281-3, 318-20, 328-
Cenabum, 137 attack Bituriges
; 31 ; their character, 23, 69, 335 n. 4,
(51 B.C.), punished by Caesar, 185; 338 ; their religion, 26-32 ; various
finally subdued, 190-1 their terri- ; senses of the word, 259-60 ; in
tory defined, 404 how news of ; Spain, 289, 294, 329 statements ;

their raid on Cenabum was spread, of ancient writers about, 303-4

736-7. Broca's theory about the Celts of '

Carthage, 3, 86. history ', 308-15 theories of Rhys, ;

Cassius, L., 49, 555. Jullian, and A. Bertrand about,

Castellon, 294. 312 n. 3 true ethnical significance
;

Casticus, 39, 504, 513. of the word, according to the

Catamantaloedis, 4, 504. author, 321 date of Celtic invasion
;

Cafnpulta, 582. of Italy, 542-6 were the Cimbri ;

Catiline, 37. a Celtic people ? 549-53 spelling ;

Cato, 99. of Celtic names, 839-50. See also

Caturlges, 48, 404-5, 501-2. Celtae, Gauls, Kymry.' '

Catuvellauni, 445, 468. Cemlbum, massacre of Romans at,

Catfivolcus, joins Ambiorix in attack- 130 captured by Caesar, 137-8
; ;

ing Atuatuca, 107 ; commits sui- legionaries resolved to avenge

cide, 125. See also 504. massacre temporarily
at, 141 ;

Cavalry, Caesar's, 42, 579-81 in ; garrisoned by two legions (51 B.C.),

campaign against Helvetii, 50. 53-4, 185 Caesar marches from, against
;

56 ; against Ariovistus, 63, 67, Uxellodunum, 190-1 question of ;

230, 646-7 ; in operations on its site, 405-15 orthography, 841. ;

Aisne, 72-3 ; in battle with Nervii, Cenomani, 2 their territory in Trans-

;

76-9 against Usipetes and Tenc-

; alpine Gaul defined, 393 ortho- ;

teri, 98-9 at Atuatuca, 107,

; graphy, 841-2.
126-7 ; in expedition for relief of Centurions, 43 panic among, at ;

Q. Cicero, 117-8 in operations ; Vesontio, 61-2 of the first rank, ;

against Indutiomarus, 120 ; under 108, 567-79 in battle with Nervii,

;
 ; — .

INDEX 859

76, 79 ; self-sacrifice of, near Atua- near(?), 98-9, 691-706; Caesar

tuca, 127. See Petronius, Lucanius, bridges Rhine between, and Ander-
Sextius Bacillus. nach (?), 100, 706-10.
Cephalic index, 261-3. Cocosates, 415.
Ceutrones (Alpine), 48, 415. Cohort, who made it tactical unit of
Ceutrones (Belgic), 457 ; orthography, Roman infantry ? 42, 563 ; number
842. of cohorts raised by Caesar in winter
Cevennes, 4 ; crossed by Cae.sar of 54-53 B.C., 731-2; depth of, in
(52 B.C.), 135. line of battle, 588 ; were there
Chablais, 82, 454. intervals between, in line of battle ?
Chalon, 183, 185, 351, 473, 618, 620, 588-97 ; how relieved in battle,
624-5. 597-9.
Chamblandes, 275-6. Coinage, Gallic, 17-9.
Chancelade, 7, 271, 336. Coins, found on Mont Rea (near
Chanonat, 156, 761. Alesia), 361.
Charente, 48. Col des Goules, 757-9.
Chariots, 15-6. Cologne, 706-8.
Chartres, 34, 189. Commentaries on the Civil War,
Cicero, Marcus Tullius, his contempt Caesar's, 212; credibility of, 251-4.
for Gallic culture, 19 ; intercourse Commentaries on the Gallic War,
with Diviciacus, 37 on restoration ; Caesar's, v-xii MSS. of, 201-2; ;

of tranquillity in Gaul (60 B.C.), when written and when published,

39-40 ; inclined to oppose Caesar 202-9; authorship of, 209-10;
(56 B.C.), 88 ; his correspondence authorship of the Eighth Book, 824-
with Caesar, 101 disgusted by
; 6; credibility of, 211-56; certain
riches of Labienus and Mamurra, passages in, to which articles or
183 ; his opinion of Caesar, 199 n. 1 ; paragraphs in Section vii are
his opinion of Caesar's Commen- specially dievoted : omnino
erant . . .

taries, 212 ; letter of Caesar to, possent (i, 6, §1), 613-4 a lacu ;

253. Lemanno . . . possent (i, 8, §1),

Cicero, Quintus, takes service under 614-5 ; muniri
ipse interim . . .

Caesar as a legatus, 101 ; commands iussit (i, 24, §§ 2-3), 628; GalH
a legion in country of Nervii, 105^ magno corpore pugnare (i, 25,
. . .

380, 383 ; defends his camp against §§ 3-4), 629 Boi et Tulingi ; . . .

Ambiorix, 728-30; Gauls

112-5, praesidio erant (i, 25, § 6), 629-30 ;

abandon siege of his camp, 117 ;

diutius cum se sustinere . se . .

joined by Caesar, 118; his legion contulerunt (i, 26, § 1), 630-1 ex ;

quartered near Samaiobriva, 119; eo proelio . pervenerunt (i, 26, § 5), . .

left in command at Atuatuca, 124 ;

631-4 ; bello Helvetiorum confecto . .

attacked by Sugambri, 126-7 convenerunt (i, 30, § 1), 634; tri-

Caesar's notice of his conduct, 249. duique viam . .processisse (i, 38,
.

Cimbri, 36-7, 61, 80, 175, 546-53. § 1), 637-8 ; itinere exquisito . . .

Cingetorix, supports Caesar, 106-7 ;

duceret 41, § 4), 638-9 ; alarios
(i,

declared a public enemy by In- omnes uferetur (i, 51, § 1), 652-3 ;
. . .

dutiomarus, 120 appointed chief

; ipse a dextro cormi proeliuin . . .

magistrate of Treveri on death of commisit (i, 52, § 2), 653-4 ; reperti

Indutiomarus, 123. sunt complures . . . vulnerarent
Cippi, 810-1, 814. (i, 52, § 5), 654 ; omnes hostes . . .

Ciron, 93. pervenerunt (i, 53, § 1), 655-7 ;

Civil war. See Commentaries. coniurandi studehant (ii, 1, § 2), . . .

Clanship, in Gaul, 20-1, 505-6. 657 ; postridie

contendit (ii, 12, . . .

Client tribes, territories of, 344-5, § 1), 668-70 saepihusque densis- ;

' 347, 502 ; status of, 517-9. simis impediretur (ii, 22, § 1),
. . .

Clientes (clients), 21-4, 508, 514-6. 675-6; Caesar cum VII. legionem . . .

Clodius, 129. inferrent (ii, 26, § 1), 676-7 ; quo in

Clothing, of Caesar's legionaries, 43, numero Venetos (iii, 7, § 4),
. . .

584-5. 678-9 ; si quando coeperant . . .

Clusium, 543. (iii, 12, § 3), 685-6; Crassus

Coast-line, of Gaul, impossibility of equitum . . . pervenerunt (iii, 26,
tracing correctly, 348. §§ 1-2), 688-9; Mosa profluit
Coblenz, Caesar marches towards (?), . . . in Ehenum infiuit (iv, 10,
96 ; Usipetes and Tencteri defeated §§ 1-2), 691-5 ; cognita Gallorum
860 INDEX
. . . existimahat (iv, 43, § 3), 690-1 ; warfare, 188 ; doomed to fail,
omnia excoqitantnr . . . angeatur 198 orthography, 842.
;

(v, 31, §5), 72G-7 ; prima luce . . . Conconnetodumnus, 130.

impedimentis (v, 31, § 6), 727-8 ; Conde-sur-Suippe, 660, 666-8.
milium passutim perfecerunt XV . . . Condrusi, Usipetes and Tencteri enter
(v, 42, § 4), 728-9 ; ferventes fusili their country, 96, 691, 695-7, 700,
. . . iacere coeperunt (v, 43, § 1), 702-3 their nationality, 332, 338-
;

729-30 Graecis conscriptam litfe'ris

; 40 their territory defined, 403-4.
;

(v, 48, § 4), 730-1 ; hoc more Gal- Considius. See Publius.
loru7n necatur (v, 56, §§ 1-2),
. . . Constitution, of Gallic states, 20-2,
731 ; trihus amiserat (vi, 1, . . . 504-9.
§ 4), 731-2; Scaldis . . Mosain . Convictolitavis, his election as Vergo-
(vi, 33, § 3), 734-5; cogunt . . . bret of Aedui confirmed by Caesar,
milia (vi, 35, § 5), 735-6 ; celeriter 148, 528 n. 2 ; intrigues against
. . . accidit (vii, 3, § 2), 736-7 ; Caesar, 152, 154 ; openly declares
altera die (vii, 11, § 1, 68, § 2), 738- for Vercingetorix, 159. See also
40 ; qui tum primum . . com-
. 527-8, 536.
parabant (vii, 11, §4), 740; miseri- Correus, heads a rebellion against
cordia vulgi (vii, 15, § 6), 742 Caesar (51 b.c), 185, 187; killed,
vicos . . . videantur (vii, 14, § 5), 188 doomed to fail, 198.
;

740-2 nostrarum turrium

; . . . Correze, 269-71.
adaequabant (vii, 22, § 4), 746 C6tes-du-Nord, 89, 309-10, 314.
aggerem latum extruxerunt . . . Cotuatus, 130, 191, 831-2.
(vii, 24, § 1), 748-9; legioyiesque Cotus, 148.
intra viyieas, &c. (vii, 27, § 2), 749- Councils, Gallic, 21, 164, 175; coun-
51 ; cum, uterque utrique . . . revo- cils of Gallic deputies summoned by
cavit (vii, 35), 751-4 ; castris prope Caesar, 57, 96, 106, 121, 128, 148,
. . . conlocaverat (vii, 36, § 2), 757 ; 623 n. 5, 634 ; councils of war, 108,
legionem unam . . . occultat (vii, 45, 138, 173. See also 533, 535-7.
§ 5), 760-1 ; ipsi ex finitimis, . . Crassus, Marcus, 88, 105.
transire videretur (vii, 55, §§ 9-10), Crassus, Marcus (a quaestor), in
769-71 ; nam ut commnfato . . . command of a legion near Samaro-
timehat (vii, 56, § 2), 771-4 ;
quibus briva, 105 ; placed in charge of
rebus auditis . . . duxerunt (vii, 61, Samarobriva, 115.
§§ 4-5), 782-4 ; cum Caesar . . . iter Crassus, Publius, strikes diecisive blow
faceret (vii, 66, § 2), 792-4 ; itaque in battle with Ariovistus, 67 ;
truncis vallis induebant (vii, 73,
. . . receives submission of maritime
§§ 2-5), 810-1 ; tanta tamen . . . tribes, 84, 86-7 arranges with ;

incumberent (vii, 76, § 2), 820-1 ; at Veneti and other tribes for supply
interiores . . adpropinquarent . of corn, 87 they demand that
;

(vii, 82, §§ 3-4), 813-5 ; Vercin- he should restore hostages, 87

geiorix ex arce profert (vii, 84, . . . marches for Aquitania, 89 ; his
§ 1), 815 ; interiores . temptant . . campaign in Aquitania, 91-4, 688-
(vii, 86, §4), 817-8; equitum partem 9 ; leaves Gaul, 105 his rank, ;

. . . adoriri iubet (vii, 87, § 4), 564-5.

818 ; coactis una XL . . . obtulit Crastinus, 578, 593.
(vii, 818 eius adventu
87, § 5), ; Credibility of Caesar's narrative of the
. . proelium committunt (vii, 88,
. Gallic war, 211-56; of the civil
§ 1), 819 ; extruitur agger . . . war, 251-4.
fastigium posset (viii, 41, § 5), Cremona, 2.
835. Crete, 42.
Commissariat, 195-6. See Rations. Critognatus, 175 credibility
; of
Commius, appointed king of theAtre- Caesar's report of his speech. 213.
bates,84-5, 197,504,724-5; deputed Cro-Magnon race, 7, 271-2, 336.
by Caesar to watch Menapii, 122 ; Coriosohtes, 84 n. 1, 87, 89; their
Labienus tries to procure assassina- territory defined, 415-8 ; ortho-
tion of, 129 appointed a general ; graphy, 842.
of army destined for relief of Ver-
cingetorix, 175 ; his operations at Dachsenbiiel, 275,
Alesia, 176 joint leader of Bello- ;
Danube, 8, 18, 36.
vaci and
rebels (51 B.C.), allied David, Felicien, 28.
185-6; obtains terms after guerrilla Debtors in Gaul, 21-2, 515-6.
 ;; ;;

INDEX 861

Docetia, 148, 755, 786 n. 1. besiege Q. Cicero's camp, 112-5 ;

Democracy in Gaul, alleged signs of, remain in arms (53 b. c), 121 their ;

529-41. country harried by Caesar, 124-6,

Denise, 269. 128, 188; credibility of Caesar's
Deva, 27. estimate of their numbers, 242
Diablintes, their territory defined, their nationality, 332, 338-40 ;

418-22 orthography, 842.

; their territory defined, 387. Set
Dijon, 167, 633, 794, 799-800. Atuatuca, Ambiorix, Catuvolcus,
Dioceses, 346-8. Sugambri.
Diodorus Siculus, 17, 19, 34; on Eburovices. See Aulerci Eburovices.
distinction between Celts and Gauls, Eleuteti, 402-3.
303, 311-2. Elisyces, 278.
Dion Cassiiis, authority of, 215-7. Elusates, 422.
Dis Pater, 30-1, 35. Embourg, 379-80.
Disentis, 275 n. 5. Embrun, 48, 404-5, 501, 615-6.
Diviciacus (Aeduan Druid), 25 begs ; Emmerich, 95, 690.
Senate for help against Ariovistus, Enamelling, 16, 20.
37 ;restored to power by Caesar, Engineers, 43, 91, 143, 579. Sec Fabri.
49, 52 begs Caesar not to punish
; Environment, influence of, 263-7.
Dumnorix, 52 guides Caesar from ;
Epona, 27, 31.
Vesontio to plain of Alsace, 62 Eporedorix, 152 reports treachery
; of
leads Aeduan levies against Bello- Litaviccus to Caesar, 153 ; seizes
vaci, 71 was he a Druid V 526 n. 1
; Noviodunum, 159-60 ; one of four
did he mislead Caesar about Druid- generals in command of army
ism ? 525-6 ; orthography, 843. destined for relief of Alesia, 175 ;
Diviciacus (king of the Suessiones), orthography, 843.
24, 504. Equites, 316, 512-3, 532, 537.
Divico, 49. Essonne, 161, 776-8.
Dolmens, 8-9, 13, 312 n. 3. Esus, 27, 29.
Domitius Ahenobarbus, 88. Esuvii, 84 n. 1 ;
join Veneti in resist-
Donnezat, 764-5. ing Caesar, 87, 92 ; Roscius's legion
Dora Riparia, 48, 431, 615. quartered among, 105 ; their terri-
Dordogne, 5-7, 13, 189, 270, 489-91. tory defined, 422-4 ; orthography,
Doubs, 60, 645, 647 n. 2. 843. See also 678-9.
Dranse, 82-3, 677-8. Ethnology, of Gaul, 5-8, 11-3, 257-
Drappes, 134, 189-90, 198. 338, 852. See also Anthropology,
Drubiaglio, 431. Belgae, Celtae, Iberians, 'Kymry,'
Druids, 26-7, 32-6, 523-9. Ligurians, Neolithic, Palaeolithic,
Dumnacus, 189 ; where did he cross Prehistoric, Round Barrow, Wal-
Loire ? 831. loons, &c.
Dumnorix, 25 forms compact with
; Etruria, 1-2.
Celtillus and Orgetorix, 39, 513-4 ; Etruscans, 542-4.
ready to help Helvetii, 40 ; induces Eure, 92.
tSequani to let Helvetii pass through Eutropius, authority of, 215-7.
their country, 48 ; commands Evocati, 166, 578.
Aeduan cavalry with Caesar's army, Evreux, 96, 161, 394.
50 ;intrigues against Caesar, 51-3 ;
intrigues again (54 B.C.), 102-3 ; Fabius, C, commands a legion in
killed, 103; was he Vergobret ? winter- quarters in country of Mor-
555-6 ; orthography, 843. ini (54 B.C.), 105; joins in relief of
Durance, 11, 48, 501. Q. Cicero, 115-6; sent back to
Durocortorum, 128 question of ; its his camp, 119; left in temporary
site, 354. command at Gergovia; 153-4
reinforces Caninius, 188 defeats ;

Ebrodunum. See Embrun. Dumnacus, 189 at Uxellodunum,

;

Eburobriga, 787. 190.

Eburones, meaning of name, 31 ; Fabius, L., 157.
Usipetes and Tencteri enter their Fahri, 43, 579.
country, 96, 695-7, 700-3 Sabinus ; Fulx 7tmralis, 611.
and Cotta encamp in their country, Fecht, 62, 648.
105 attack Atuatuca, 107; destroy
; Fibulae (of bridge), 711-21.
force of Sabiuus and Cotta, 1 10-2 Fines, 346.
862 INDEX
Finisteio, 8G. Sec Osisnii. 520-3 did Caesar originally iiitend
;

Flavigny. See Montague de Flavigny. to conquer ? 558 Gallic walls, ;

Floius, authority of, 215, 217. 746-8. See Celts, Gallia Cisalpina,
Forez, 50. Gallia Comata, Province.
Formans, 49, GIO. Geidumni, 457-8 orthography, 843.
;

Fort St. Andries, 693, G95-8, 702, 704. Geneva, 46 orthography, 843.
;

Fortune, Caesar's belief in, 41. Geography, Caesar's want of precision

Frontiers, of Gallic states, difficulty of in writing about, 344.
determining, 346-8. Gergovia, 131-2 Vercingetorix ban-
;

Fundae lihriles, 813. ished from, returns to, 132 Caesar ;

Furfooz, 274, 276. marches against, 148 Vercinge- ;

torix occupies, 149 first opera- ;

(jlabali, induced to join rebellion of tions at, 149-51, 756-7 Caesar ;

Vercingetorix, 135 hounded on by

; temporarily quits, to intercept
Vercingetorix to invade Province, Aeduan contingent, 153-4, 757-8 ;

165 ; their territory defined, 371. Caesar attempts to take, hy coup de

(jialba (king of the Suessiones), 71, 73, main, 155-8, 758-67 credibility ;

504, 658.' of his account of his attempt,

CJalba, Servius, his campaign in the 245-9 ;he abandons, in order
Valais, 82-4, 677-8; Caesar's notice to rejoin Labienus, 159. iS'ee also
of, 249 n. 8. 621-2.
CalHa Braccata, 15. Gergovie, 762-3, 766.
Callia Cisalpina, 1-2, 40, 48, 67, 70, German cavalry, 65 employed by ;

95, 134. Caesar, 137, 166, 169, 170, 176, 581,

Gallia Comata, 41. also light infantry, 185-6 cavalry ;

Gallo-Brythonic, 281, 320-1. hired by Commius, 186 aid of ;

Gap, 48, 404-5, 501, 615-6. (rcrman cavalry indispensable to

Garonne, boundary (roughly speaking) Caesar, 198.
between Celtae and Aquitani, 5, Germans, threaten Gaul, 23, 36
344 ;crossed by P. Crassus, 93 ;
Caesar resolved to prevent, from
tribes between, and Seine join Ver- conquering Gaul, 44 Labienus ;

cingetorix, 132. sent to prevent, from crossing

Garumni, 425. Rhine (56 B.C.), 89; Caesar's in-
Gates, 425. vasions of Germany, 99-100, 123
Gaul, invaded by Celts, 1, 5, 11-2; credibility of his account of in-
Gauls in Italy, 1-3, 542-5 Romans ; vasions, 249 ; Germans said to be
establish footing in Transalpine meditating attack on Romans (54
Gaul, 3-4 ; the country and its B.C.), 106; some tribes refuse,
inhabitants, 4-5 prehistory of,
; others promise to aid Indutiomai-us,
5-11; ethnology, 5-13, 257-338; 119, 121 ; Caesar prevents, from
character, civilization, political and aiding Ambiorix, 122-3 question ;

social organization of Gauls, 13-7, between Celts

of ethnical relations
20-5 ; unifying influences, 25-6 ;
and Germans, 326-35 were Belgae ;

religion, 26-32 ; invasion of, by of German 321, 332-3.

origin See ?

Teutoni and Cimbri, 36 by Ario- ; Ariovistus, Cimbri, Suebi, Sugam-

vistus, 37-8, 553-4 ; plan of bri, Teutoni, Ubii, Usipetes.
Orgetorix for conquest of, 38-9 Gesoriacum, 435 n. 1.

Caesar appointed Governor of, 40 ;

Gien, 405-14.
Caesar resolved to prevent Germans Gobannitio, 132.
from conquering, 44 conquest of,; Goidelic, 12, 281, 312 n. 3, 319-21.
by Caesar, 46-199 Caesar's re-
; Gorgobina, besieged by Vercingetorix,
marks on character of Gauls, 104, 136 ;he raises siege, 138 ; question
194 ethnology of, 257-338
; who ; of its site, 425-30.
were the true Gauls ? 325-35 Graioceli, 48, 430-2.
population of, 340-3 remarks on ; Grand-Pressigny, 8.
map of, 345-8 Gallic league, 350 ;
; Grannus, 29.
monarchy in, 21-2, 24, 504-5 ;
Great St. Bernard, 82.
clanship, senates, and law, 507-9 ;
Greaves, 584.
private landed property, 509-12 Greece, 3 ; commerce of Gaul with,
nobles of, 512-3 ; slavery, 517 ; 10, 16.
inter-tribal relations, 517-9 ; philo- Greek characters used in writing, 17-8,
Roman and anti-Roman parties, 730-1.
 INDEX 863

Greek studied by certain dlauls, 17. Caesar, 106-7 ; instigates Ambiorix

Urenelle, 273-7, 308, 323, 329, 785. to attack Atuatuca, 107 his ;

Gresigny, 816, 818-9. intended attack on Labienus pre-

Gris-Nez, 103. vented by Caesars victory over
Grudii, 458. Nervii and Eburones, 118 ; his plan
Guerande, 680-1, 683-4. of campaign, defeated by Labienus
Gutuater, 528 note, 831-2. and slain, 119-20 struck too soon,
;

198.
Hallstatt, culture in Gaul, 9-10, 15 ; Insubres, 2.
ethnology of period, 274, 336. Intra vineas, 749-51.
Hannibal, 2, 132, 182. Isere, 3, 48, 363-4, 501, 502 n. 6, 615.
Harudes, 59. Italy, Celtic invasion of, 1-2, 542-6 ;
Hastedon, 391-2. threatened by Cimbri and Teutoni,
Hautmont, 671-2, 675. 36 by Germans, 44
; endangered ;

Hegemony, of Arverni, 3, 24-5 ; of by presence of Ariovistus in Gaul,

Aedui, 24-5, 37, 517-8 ; of Sequani, 58 ; enthusiasm at Caesar's
in,
37. Gallic victories, 85 Caesar's cus-
;

Helvetii, plan invasion of Transalpine tom of wintering in, 67, 95, 119,
Gaul, 38-40 ; prepare to march 129.
through Roman Province, 46-7, Itiyierary of Anionine, 349.
613 negotiate with Caesar, pre-
; Itius, Portus, 103 ; its whereabouts
vented by him from crossing Rhone, discussed, 432-8.
47, 614-5 allowed by Sequani to
; Izernore, 355-6.
march through Pas de I'Ecluse, 48 ;

Aedui solicit Caesar's aid against, Javelin, 43, 55, 66, 78, 599, 629.
49 Caesar's campaign against, 49-
; Jovinus, 334.
56, 616-34 Caesar's treatment of
; Julius Caesar. See Caesar.
fugitives after battle near Bibracte, Jupiter, 28-31.
57 join in attempt to relieve
; Jura, 38, 47, 439, 441, 613-4.
Alesia, 174 ; military character,
195, 197 ; credibility of Caesar's Kings, in Gaul, 21, 504-5. 522. See
narrative of their emigration and of Ambiorix, Catamantaloedis, Catu-
his campaign, 217-27, 231-6, 237- volcus, Cavarinus, Commius, Divi-
41 ; ethnology of, 318. See also ciacus,Galba,Monarchy,Moritasgus,
594, 597. Tasgetius, Teutomatus, Vercinge-
Helvii, Caesar's levies concentrate in torix.
country of (52 B.C.), 135; attacked Knights, Gallic, 21-2, 24-5. See
by order of Vercingetorix and Equites.
defeated, 165-6 their territory ; 'Kymry,' 308-11, 318, 321 n. 4, 323.
defined, 432.
Herve, 80. Labienus, T., ordered by Caesar to
Hesbaye, 80, 112. guard lines on Rhone, 48 rejoins ;

Hirtius. See Aulus. Caesar near confluence of Saone and

Hoards, 89 n.l. Rhone, 49 ; co-operates with Caesar
Hohberg, 328. in attempt to surprise Helvetii, 53 ;
Holland, 5. left in command of Roman army for
Homme Mort, 1', 8, 11, 272-4, 277, 323, winter of 58-57 b. c, 68 informs ;

336. Caesar of conspiracy of Belgae, 69 ;

Homo Heidelbergensia, 268-9. pursues Belgae down valley of
Homo primigenms, 270. Aisne, 74 in battle with Nervii,
;

Houses, 16, 19. 79 in command of a legion during

;

Human sacrifice, 33-4, 132 n. 2, 526. winter of 54-53 B.C., 105, 383-4,
732-4 informed of disaster at
;

Iberian inscriptions, 11,289-90, 293- Atuatuca, 112; unable to join

5, 301. Caesar in relieving Q. Cicero, 116;
Iberians, 11 ; ethnology of, 287-302, informed of relief of Cicero, 118,
Iccius, 70. 733 defeats and slays Indutio-
;

Idols, 10. marus, 120 reinforced by Caesar,

;

Iliberris,290-3, 297. defeats Treveri, 122-3 ; attempts to

HI, 67, 650 n. 8, 656-7. assassinate Commius, 129; why did
lllyricum, 40, 86-8, 101. he not take the field at beginning
Indutiomarus, reluctantly submits to of 52 B. c. ? 134 ; charged by Caesar
 864 INDEX
duty of suppressing rebellion
wifcli Leuci, send supplies to Caesar, 65 n. I,
valley of Seine, 148 ;
in Caesar 642 n. 2 ; do not join in attempt to
anxious for his safety, 155, 160 ; relieve Alesia, 174 territory of, ;

his campaign against Camulogenus, defined, 444.

161-4, 775-85 ; rejoins Caesar, 164, Levaci, 458.
785-90 ; strikes decisive blow at Lex Pompeia Licinia, 832-4.
. Alesia, 179, 182, 817-9 his great ; Lex Sempronia, 833.
services, 196 ; did he or Caesar Lexovii, 89, 96 n. 1, 442, 444 ; ortho-
defeat Tigurini ? 231-2 fairness of ; graphy, 844.
Caesar's account of his exploits, 250. Liger, 280 n. 2. See Loire..

La Fere, 71, 659, 670 n. 3. Ligurians, 11 ethnology of, 277-87,

;

La Grande Briere, 681-3. 336 ; language, 281-3, 320-1.

La Guerche, 429-30. Lilia, 810.
La Tene, 15, 318, 548. Limagne, 130, 149, 159.
Lake dwellings, 275, 318. Limbourg, 380-2.
La Madeleine, 271. Lingones, Caesar overtakes Helvctii
Land, did the Gauls recognize private in their country, 56, 631-3 ; supply
property in ? 509-12. Caesar with xjorn for campaign
Langres, 56, 633, 789, 799. against Ariovistus, 60, 65 n. 1,
Latere aperto, 621-3, 625-7. 642 n. 2, 644 two legions winter
;

Latin, knowledge of, in Gaul, 17, among (53-52 b. c), 128, 134;
143 n. 1, 730. Caesar rejoins the legions, 136
Latium, 1. adhere to Caesar during rebellion of
Latobrigi, 46 their geographical
; Vercingetorix, 133, 164, 174 Caesar ;

position discussed, 438-42 ortho- ; rests his army in their country, 167,
graphy, 844. 789-90 Caesar marches through
;

Laugerie-Basse, 7, 271-2. their country with view of succour-

Law, in Gaul, 505-7. ing Province, 792-4 their territory ;

League, Gallic, 350. defined, 444-5 ; meaning of per

Legati, 43, 66, 77, 101, 156, 563-5; ser- extremos Lingonum fines, 356, 792-4.
vices of, during conquest of Gaul, Liscus, 49, 51, 505-6.
196; was Caesar just in his notice of Litaviccus, 23 ; tampers with Aeduan
their services ? 249-50. See Antis- contingent on march to join Caesar
tius Reginus, Aurunculeius Cotta, at Gergovia, 152-3, 767-8 recruits ;

Brutus, Caesar, L., Caninius Rebilus, for Vercingetorix, welcomed at

Cicero, Crassus, Fabius, Galba, Bibracte, 159 ; orthography, 844,
Labienus, Mark Antony, Munatius Loire, legions cantoned along valley of
Plancus, Roscius, Sextius, Titurius (57 B. c. ), 84 ; Caesar orders ships
Sabinus, Tiebonius. to be built in estuary, 87-8
Legionaries, 43-4 panic among, at
;
Brutus's fleet assembles in estuary,
Vesontio, 60-1 short stature of,
; 90 ; legions quartered between, and
81 conduct at Avaricum, 141
; ; Seine (56 b. c. ), 94 ; boundary
nationality, 195 n. 1 defensive ; between Aedui and Bituriges, 133,
armour and clothing, 584-5. 351-2, 426, 428-30 ; Caesar crosses
Legions, Caesar's, 42-4 ; raised by at Cenabum, 137-8 Caesar crosses ;

Caesar during Gallic war, 48, 70, in spite of Aedui, 160, 774 cam- ;

120, 731-2 how many did Caesar

; paign of Fabius in lower valley,
receive from Senate and Roman 189-90 ; where did Dumnacus
People ? 557 numerical strength
; cross ? 831.
of, 559-63 number of, at Alesia,
; Lorica, 584-5.
802-4; 1st legion, 802; 5th Luca, 88.
(Alauda), 653 n. 4, 802-3; 6th, Lucanius, Q., Ill, 569.
802-3; 7th, 78-9, 676-7, 802; Lucterius, threatens to iuA ade Pro-
8th, 78, 676, 802; 9th, 78, 675- vince, 133, 135 ; defends Uxello-
6; 10th, 62-3, 78-9, 156-8, 675- dunum, 189-90 goes out to fetch ;

6, 761-6, 802; 11th, 676, 802; supplies, escapes slaughter, 190

12th, 676-7, 802; 13th, 156, 764, delivered up to Caesar, 193
802; 14th, 802; 15th, 802-3. doomed to fail, 198 ; orthography,
LemOnum, 189 ; orthography, 844. 844.
l^emovices, 442-4. Luernius, 198.
Le Moustier, 269. Lutecia, 19 ; council at, 121 Labi- ;

Les Laumes, 181, 805, 807, 800-10, 815. enus marches for, 161 ; burned by
 ;; ;

INDEX 865

order of Camulogenus, 161-2 Caesar deters, from helping Am-

battle near, 163-4, 778-85 ; ortho- biorix, 122 ; do not join in attempt
graphy, 844. to relieve Alesia, 174 ; their terri-
Luzech, 485-9, 493. tory defined, 449-53.
Luzy, 53. Mentone, skeletons found at, 7, 270-2.
Mercury, 28-32.
Macon, 50, 183, 185, 351, 473, 617-21. Metellus, 40.
Magetobriga, site of battle of, 445-6 ;
Metius, M., 64.
date and circumstances of battle, Metlosedum, 161, 163, 775-85 ; ortho-
554-5 orthography, 844-5.
; graphy, 845-6.
Magius, 253-4. Meuse, skulls of Neanderthal type
Mamurra, 183, 711. found in basin of, 6 winds round
;

Mandubii, 165 ; expelled from Alesia, Mont Falhize, 80 Caesar crosses

;

their fate, 175, 812 ; their territory (55 B.C.), 96; Atuatuca situated
defined, 446-7 ; orthography, 845. east of (?), 105, 374-7, 383 were ;

Maniple, 43, 588-90, 594-7. See Usipetes and Tencteri routed near
Primi ordines. Order of Battle. confluence of, with Waal ? 691-706
Manlius, L., 92, 475. did Scheldt flow into ? 734-5. See
Mantinea, 596. also Atuatucorum oppidum, Am-
Manuscripts, of Caesar's Commentaries, bivariti.
201-2. Miette, 660-2, 664.
Map, of Gaul, remarks upon, 345-8. Milan, 2.
Marcellus, 319. Mile, Roman, 350.
Marches, average length of Caesar's, Milo, 129.
635 ; his forced march during Miners, of Aquitania, &c., 16.
operations at Gergovia, 153-4 ; to Minerva, 28, 30.
cross Loire, 160 ; from Agedin- Mines, ofRomans and Gauls in siege of
cum to Cenabum, 410, 494-5 ; to Avaricum, 144, 600-2.
Vesontio, 633-4. Minucius Basilus, sent with cavalry to
Maritime Alps, 3. pursue Ambiorix, 124 nearly ;

Marius, defeats the Teutoni, 36-7 ; catches him, 124-5 was he a ;

his military reforms, 42 ; made legatus ? 564-5.

cohort tactical unit of Roman Mommsen, 23.
infantry, 563. Monarchy, in Gaul, 21, 24, 504-5.
Mark Antony, 188, 190, 832 n. 13. Mont Auxois. See Alesia.
Marne, 5, 16, 29, 70, 162, 305, 310, Mont Falhize, 80, 388, 390-2.
778-80. Mont Ganelon, 187, 830.
Mars, 28-30. Mont Genevre, 48, 615.
Marseilles, 3. Mont Parnasse, 164, 785.
Martigny, 83-4. See Octodurus. Mont Pevenel, 169, 172, 817.
Massilia, 3-4, 10, 17-8, 26, 34, 542, Mont Rea, 169, 177-9, 361-2, 804 n. 6.
604-6. 807, 810, 815-9.
Matisco. See Macon. Mont Rognon, 759.
Matrona, 27. Mont St. Marc, 185.
Maubeuge, 76, 675. Mont St. Michel, 8.
Mauchamp, 71, 660, 662, 664, 666-7. Mont St. Pierre, 186, 827, 830.
Mediolanum, 2. Montague de Bussy, 169, 172, 807,
Mediomatrici, join in attempt to 816 n. 3.
relieve Alesia, 174 ; their territory Montague de Flavigny, 170, 172,
defined, 447. 178-9, 807, 810, 813, 816-9.
'
Mediterranean race,' 272. Montague de la Serre, 150, 155,
Medulli, 501. 760-1.
Meldi, 447-9, 472, 477. Montaigne, 256.
Melodunum. See Metlosedum. Montargis, 137. See Vellaunodunum.
Melpum, 543. Montbeliard, 62, 644-6, 647 n. 2.
Memini, 502. Morbihan, 86, 89-90, 680-2, 684.
Menapii, join Veneti, 89 Caesar ; Morini, join Veneti, 89 ; attack
marches against (56 B.C.), 94; Caesar's troops (55 B.C.), 102;
their country invaded by Usipetes Caesar's campaign against (56 B. c),
and Tencteri, 95 rebel after; 94, 689 ; Fabius's legion quartered
disaster at Atuatuca, 121 ; their among (54 B.C.), 105 Fabius sent ;

lands harried (55 b. c), 122 I back to, after reUef of Cicero, 119 ;
1093 3 K
866 INDEX
their territory defined, 450-3 ; was 786 n. 1 seized by Eporedorix and
;

Commius king of ? 724-5. Viridomarus, 159 question of its

;

Moritasgus, 504. site, 464.

Moselle, Usipetes and Tencteri de- Noviodunum (Suessionum), 74 ques- ;

feated near confluence of, with tion of its site, 464-6 did Belgae
;

Rhine (?), 99, 691-706. march from, against Caesar ? 658-

Mouzon, 105, 377, 382 n. 1, 384. 9 ;duration of Caesar's march to,
Musculi, 608-10. from his camp (?) on the Aisne,
Mussy-la-Fosse, 176, 812-3. 668-70.
Numidians, 42, 72-3.
Namnete-5, their territory defined, 453,
680 n. 3. Ohaerati, 515-6.
Namur, 105, 380, 383, 388-9. Ocelum, 430-2.
Nantes, 88. See Namnetes. Octodurus, 82-4, 677-8.
Nantuates, 82, 453-5. Oise, 71, 75, 187, 828-30.
Napoleon III, v, viii, xv^-xvd, xx v, xxvii. Ollovico, 4.
Narbo, 4 ;threatened by Lucterius, Opme, 150, 155.
rescued by Caesar, 135. Orbis, 110 n. 2, 728.
Narbonne. See Narbo. Orcet 150.
Neanderthal race, 6-7, 269-71, 336. Order of battle, Caesar's, 587-99, 852.
Negroid skeletons, 7, 271. 336, 851. Orgetorix, 38-40, 513-4; ortho-
Nemetes, 455-6. graphy, 847.
Nemetocenna, 456. Orleans, 84. See Cenabum.
Neolithic man, in Gaul, 8, 272-7, 336. Orne, 87.
Nervii, 75 ; defeated by Caesar (57 Orosius, authority of, 215, 217.
B.C.), 75-9, 671-7; survivors Osismi, 466-7 orthography, 847.
;

exaggerate their losses, 80, 205- Ouche, 799-800.

6 ; Caesar treats survivors with Oze, 169-76, 177. 805-6.
clemency, 80 ;Q. Cicero winters Ozerain, 169-71, 178-9, 805-6.
in country of, 105, 383 ; besiege
Q. Cicero's camp, 112-5, 729-30; 'P' Celts, 281, 319-20.
defeated by Caesar, 117-8 remain ; Paemani, nationality of, 332, 338-40
in arms, their lands ravaged by their territory defined, 404 ortho-
;

Caesar, 121 ; nationality, 332 ; their graphy, 847.

territory defined, 456-7 ; territory Palaeolithic man. in Gaul, 6-7, 269-72,
of their clients, 457-8. 336.
Neuf-Mesnil, 76, 79 battle with
; Parisian skulls, 309-10.
Nervii shown to have taken place Parisii, campaign of Labienus against,
at, 671, 674-5. 161-3 ; military character, 195 ;

Neuwied, 709. their territory defined, 467 ; ques-

Nevers, 148, 755, 774, 786 n. 1. Set tions relating to Caesar's narrative
Noviodunum (Haeduorum). of Labienus's campaign, 775-85.
Nievre, 136, 148. Pas de I'Ecluse, 47-8, 613-4.
Nimes, 135. Pentagram, 34.
Nitiobroges, 4 ; induced to join Perot te, 8.
rebellion of Vercingetorix, 133, 135, Petit Grosne, 50, 620.
147 ; their territory defined, 458-9 ; Petrocorii, 467 ; orthography, 847.
orthography, 846-7. Petronius, M., 157-8.
Nobiles (nobles), in Gaul, 21-2, 24-5, Petrosidius, L., 112.
512-3 ; Mommsen on power of Phalanx, 592-3, 595-7, 654, 852.
noble families, 513-4. Pharsalia, 593, 621-2.
Normandy, Crassus receives sub- Philip of Macedon, 17.
mission of tribes of, 84 tribes ; Philo-Roman party in Gaul, 25, 520-3.
rebel, 87, 91-2 they contemplate
; Pictones, lend ships to Caesar, 90
an attack on Roscius, 118. Caninius encamps among, 188
Notre-Dame, Lutecia built upon its their territory defined, 467.
site, 161. Pilum. See Javelin.
Noviodunum (Biturigum), surrenders Pisistratus, 21, 505.
to Caesar (52 B.C.), 138; question Piso, L., 98.
of its site, 459-64. Pithecanthropus, 268, 270.
Noviodunum (Haeduorum), used by Placentia, 2.
Caesar as a magazine, 148-9, 755, Plancus, L.. quartered near Samaro-
 ; ;

INDEX 867

briva (54 b. c), 105; sent to overawe Puy d'Issolu, 189, 489-93.
Carnutes, 106. Puy de Marmant, 764-6.
Pleumoxii, 458. Puy Giroux, 150.
Plutarch, authority of, 215, 217. Pyrenees, 7, 93.
Plutei, 610-1. Pythagoras, 34.
Po, 1-2.
PoUera, 286. 'Q' Celts, 281-2, 319-21.
Pommiers, 74.
Quariates, 283, 502.
Pompey, negotiates with Caesar at Quaternary Period, 6.
Luca, 88 lends Caesar a legion,
;
Quesnoy, 672.
120-1, 802-3 restores order at
;
Quiberon Bay, 90, 680.
Rome after murder of Clodius, 134 ;

Caesar's remarks about, 204. See

also 252-4. Rabutin, 170, 172, 807, 817-8.
Pontarcy, 660, 664-5. Randan, 768.
Pontarlier, 47, 613. Rations, 44, 585.
Pontavert, 71, 660, 665-7. Rauraci, 46 their territory defined,
;

Population, of Gaul, 340-3. 438-42 orthography, 847.

;

Posidonius, 194. Ravenna, 88.

Pottery, 16. Rebilus. See Caninius.
Praeconinus, L., 92. Red-hot (or white-hot) balls, 114,
Praefecti fahrum, 43. See Mamurra. 729-30.
Prehistoric races, of Gaul, 5-11, 13, Redones, 467-8 orthography, 848. ;

268-77, 336. Redoubts at Alesia, 804-5.

{castella),
Primi ordines, 567-79. Reginus. See Antistius Reginus.
Princeps, principatus, meaning of, 533 Reihen-Graber, 327.
n. 2. Reims, 70, 128. See Durocortorum.
Private property in land, did the Reinach, S., 31.
Gauls recognize ? 509-12. Religion of Gauls, 7-8, 10-1, 26-32,
Procillus. See Troucillus. 132, 180 n. 3.
Province, formation of Roman, in Remi, voluntarily submit to Caesar
Transalpine Gaul, 4 ; victory of and help him, 70-1 their territory ;

Cimbri and Teutoni in, 36 Caesar; threatened by other Belgic tribes,

appointed Governor of, 40 ex- ; 72-3 ; loyal to Caesar, 86, 133, 151,
posed to danger from Germans, 44 ; 164, 167, 174, 198 he treats them ;

Helvetii desire to march through, with distinction, 106 ; congratulate

46 ; Caesar refuses to allow Hel- Labienus on relief of Cicero's camp,
vetii to enter, 47 ; exposed to 118 Indutiomarus threatens, 120 ;
;

danger from Helvetii, 48, 225-6 ; intercede for Carnutes, 122 two ;

Caesar raises cavalry in, 50 Ario- ; legions protect them

detailed to
vistus complains that Caesar has (52-51 B.C.), 183; Suessiones
crossed frontier of, 63 Caesar
; placed in dependence on, 185 their ;

levies oarsmen from (56 B.C.), 88 ;

territory defined, 468-9 clients of, ;

threatened by Lucterius, 133, 135 ; 518 hegemony of, 518.

;

rescued by Caesar, 135 ; Aedui Rhine, 5 crossed by Celts, 12

; ; Ger-
intend to prevent Caesar from mans fight their way
to right bank
retreating to, 160, 769-71 Caesar
; of, 37 ; Ariovistus crosses, 37
said to be retreating to, 162; threat- Ariovistus and beaten host flee
ened by Vercingetorix, 165-6 to, 67, 655-7 ; some Transrhenane
roads leading to, from Further tribes offer submission to Caesar
Gaul, blocked, 166 Caesar marches
; (57 B.C.), 82 ; Labienus charged to
to succour, 167, 788-80, 791-4; prevent Germans from crossing, 89 ;
Caesar posts troops to guard (52-51 Usipetes and Tencteri cross, 95,
B.C.), 183. 689-90 Usipetes and Tencteri
;

Provincials support Caesar, 197. driven to confluence of, with

Ptianii, 467. Moselle (?), 99, 691-704; Caesar
Publius Considius, 53. crosses in 55 and 53 b. c, 99-100,
Punic War, second, 2. 123, 706-10 construction
; of
'
Putrid Plain,' 37. Caesar's bridges over, 711-24;
Puy de Dome, 131 ethnology of de-
; Sugambri cross, 125 ; German
partment of, 309, 314-5, 317 n. 5. cavalry cross, to reinforce Caesar,
Puy de Jussat, 758-9, 761. 166. ASee Triboci
868 INDEX
Rhodanus, 280, 282 n. 2, 455. See Round Barrow race, 273, 275, 325 n. 6.
Rhone. Ruteni, induced to join rebellion of
Rhone, Arverni and alHes defeated at Vercingetorix, 133, 135 hounded ;

confluence of, with Isere (121 B.C.), on by Vercingetorix to invade

3 ; Romans masters of lower Province, 165 ; a legion quartered
valley, 3-4 Romans defeated on
; in their country, 183 ; their terri-
banks of, by Cimbri and Teuton! tory defined, 469.
36 ; Helvetian marauders on right Rutilus, M. Sempronius, 564-5.
bank (60 B.C.), 38; Helvetii pre-
vented by Caesar from crossing, Sabinus. See Titurius.
46-7, 224-5, 614-5; Labienus Sabis, 673, 734. See Sambre.
holds Caesar's lines on, 48 Caesar ; Saint-Gildas, 90, 680, 684.
crosses, near Lyons, 49, 617-8 ;
Saint-Maurice, 82, 454.
Allobroges defend,
against Ver- Saint-Nectaire-le-Haut, 309, 315.
cingetorix's levies, 166 once ; Saint-Parize-le-Chatel, 57, 136. 427-8,
separated Ligurians from Iberians, 430.
277-8, 288 territory of Allobroges
; Salluvii, 279. See Salyes.
on northern bank, 364-5 Momm- ; Salyes, 3.
sen's amazing explanation of the Samarobrlva, 75 ; Trebonius's legion
passage lacu Lemanno, qui in quartered at (54 b. c), 105, 371 n. 7 ;
flumen Rhodanum influit, 455 ;
Caesar fixes his head-quarters there
Polybius's description of the Rhone, (54 B.C.), 106; Vertico carries a
544-5. dispatch to, 115 ; Caesar leaves
Risolles, 150, 155, 157, 757-9. Crassus in charge of, 115 ; three
Roads, 16. legions quartered near, 1 19 ; Caesar
Roanne, 50, 471. holds Gallic council at (54 B.C.), 121
Roche Blanche, 150 seized ;by question of its site, 469 ; ortho-
Caesar, 151, 756 camp on, held
; graphy, 848.
during Caesar's absence from Ger- Sambre, 75 ; battle on, 76-9 ; ques-
govia, 153, 757-8 held by Sextius,
; tion of site of battle, 671-7 ; did
157, 762. See also 765-6. Caesar confound Sambre with
Rodumna, 280, 471. See Roanne. Scheldt ? 734-5.
Rome, captured by Gauls, 1, 543 ;
Samnitae, 469-70, 680 n. 3.
Romans repel Gallic incursions and Samnite war, third, 1-2.
conquer Cisalpine Gaul, 1-3 estab- ; Sancerre, 428-9, 460.
lish themselves in Transalpine Gaul Santoni, lend ships to Caesar, 90
and form Province, 3-4 Roman ; Helvetii intend to settle in country
influence on Gallic coins, 18 ;
of,credibility of Caesar's remark
Roman army defeated by Tigurini, about their proximity to Province,
38,555 Roman interests menaced
; 221-3, 225-6, 470 their territory
;

by intended Helvetian emigration, defined, 470 orthography, 848.

;

39-40, 48 and by pressure of

; Saone, crossed by Helvetii and by
Germans upon Gaul, 44, 58 Caesar ; Caesar, 49, 616-9 Caesar marches
;

leaves Rome (58 b. c), 46 ; Dum- up valley of, in pursuit of Helvetii,

norix heads anti-Roman faction, 50-1, 620-1 ; Caesar marches up
52, 520 ; Ariovistus complains of valley (52 B.C.), 136 Caesar intends
;

Roman interference and bad faith, to cross, in order to succour

59, 63 ; Roman soldiers liable to Province, 167, 357, 359-60, 795,
panic, 60 ; rejoicings at Rome over 797-8 ; did it form boundary
Caesar's victories (57 B.C.), 85; between Aedui and Sequani ?

Gauls familiar with idea of Roman 352-3.

dominion, 86, 199 ; Caesar obliged Sappers' huts, 74, 114, 140, 144, 176,
to think of Roman politics during 178,608-11,815.
conquest of Gaul, 88, 121 ; Roman Sathonay, 49, 618-9.
supremacy galling to Gallic patriots, Savoy, 284.
104, 119, 129; riots in Rome (52 Scaeva, 568, 570-2, 574, 579.
B.C.), 129 ; Pompey restores order Scaldis, 673, 734-5. See Scheldt.
in, 134 ; Caesar desires to Romanize Scheldt, 5, 75 Nervian non-com-
;

Gauls, 193; no 'Little Roman' batants take refuge near estuary,

party, 228 how many legions did
; 80, 674-5 ; Caesar marches towards
Roman People grant Caesar ? 557. lower valley, Eburones take refuge
Roscius, L., 105, 118, 564-5. in marshes formed by estuary, 124 ;
 ; ;

INDEX 869
did Caesar confound Scheldt with Serbannes, 152, 768.
Sambre ? 734-5. Sertorius, 93.
Scorpio, 582-3. Sextius, T., his operations during
Scrohes, 809. attack on Gergovia, 157-8, 761-2,
Seduni, 82, 677-8 their territory ; 764-6.
defined, 453-4. Sextius Baculus, in battle with the
Segni, nationality of, 332, 338-40; Nervii, 79 ; at Octodurus, 84
their territory conjecturally defined, saves Cicero's camp at Atuatuca,
404. 126 ;was he an evocatus in 53 b. c. ?
Segusiavi, their territory defined, 470- 578.
1 ; in what part of their country Ships, 16, 86, 89-91.
did Caesar encamp (58 b. c.) ? 617- Sibusates, 474, 848.
9 ; orthography, 848. Silvanectes, 397, 477.
Seine, 4 legions winter between, and
; Sion, 82.
Loire (56-55 b. c), 94 campaign of ; Slavery, in Gaul, 22, 517.
Labienus in valley of, 148, 161-3, Slingers, in Caesar's army, 42, 72-3,
775-85. See Lexovii, Meldi. 177 ; at Atuatuca, 111 ; at Alesia,
Senate, Roman, support Massiliots 176, 813.
against Ligurians, 3 will not ; Soissons, 74- See Noviodunum (Sues-
definitely assist Aedui against Ario- sionum).
vistus, 37 try to guard by diplo-
; Somme, 94.
macy against threatened Helvetian Sos, 93, 474-7.
invasion, 39-40 grant title to ;
Sotiates, 93 ; their territory defined,
Ariovistus, 40 order a thanks- ; 474-7 ; orthography, 849.
giving service in honour of Caesar's Spain, 36 ; reinforcements from, join
victories, 85 induced to vote pay
; Aquitanians (56 b. c), 93 ; Iberians
for legions raised by Caesar on and Basques of, 287-301.
his own
responsibility, 88, 228 Spanish cavalry, employed by Caesar,
Caesar's treatment of Usipetes and 42, 107, 579, 581.
Tencteri condemned in, 99 num- ; Speeches, in Caesar's Comment aries^
ber of legions which they granted credibility of. 213.
Caesar, 557. Spy, 269-70.
Senates, of Gallic tribes, 21, 505-7 ;
Standards, 30, 79, 130.
senate of the Nervii, 80 ; senates of Statues, 31-2.
Eburovices and Lexovii massacred, Stature, 257 n. 2, 261 n. 1.
92 senate of the Bellovaci, 188.
; Steatopygous race, 272 n. 3.
Senones (of Cisalpine Gaul), 2. Stoffel, Colonel, letter from, to the
Senones, 24 ; rebel against Caesar writer regarding his methods of ex-
(54 B. c. ), 1 19-21 ; inquiry into their cavation, xxv-xxvii.
conduct, 128 rebel under Drappes
; Stradonic, 18, 20.
(52 B.C.), 134; Caesar captures Strasburg, 481, 650 n. 8.
their stronghold, Vellaunodunum, Sublicae, 721-4.
137 ; Labienus's campaign against, Sucellus, 31.
161-3, 775-85 military character,
; Suessiones, 24; Remi anxious to shake
195, 198 ; their territory defined, off their yoke, 70 ; join Belgic con-
471-3. See also 785-90. federacy against Caesar, 71 ; surren-
Sequana, 281-2, 320. der to Caesar, 75,670; threatened by
Sequani, 4 hire aid of Ariovistus
; Bellovaci, 185, 826 ; their territory
against Aedui, 37 ; subdued in turn defined, 477.
by Ariovistus, 38 ; allow Helvetii Suetonius, 40, 211.
to pass through their country, 48 ;
Suebi, threaten to reinforce Ario-
ask Caesar's aid against Ariovistus, vistus, 59 ; return home, 67 ; harry
57-8 Caesar occupies their strong-
; Usipetes and Tencteri, 95 ; their
hold, Vesontio, 60 send supplies ; superiority acknowledged by L^si-
to Caesar, 60, 65 n. 1, 642 and n. 2 ;
petes and Tencteri, 97 ; Ubii
Caesar quarters troops in their solicit Caesar's aid against, 100 ;
country (58-57 B.C.), 68 n. 1; he ready to fight Caesar, 100 ; send
intends to march through their reinforcements to aid Treveri
country, to succour Province, 167, against Labienus, 123 ; Caesar too
786-90, 791-4 ; he quarters troops wary to attack. 123 ; orthography,
in their country (52-51 B.C.), 183; 849.
their territory defined, 473-4. Sugambri, refuse to surrender cavalry
870 INDEX
of Usipetes and Tencteri to Caesar, 688 defeats them, 91-2
; placed ;

99 ;Caesar punishes, 100, 702, 707, in jointcommand at Atuatuca, 105,

710 ; harry land of Eburones, 125 ; 726 attacked by Ambiorix, 107 ;
;

attack Cicero's camp at Atuatuca, overrules his colleague and abandons

126-7, 735-6; credibility of Caesar's camp, 108-10 ; conduct in subse-
account of their raid into country quent disaster, 110-1 killed, 112 ; ;

of Eburones, 244-5 their territory,

; was Caesar wrong in employing him
697 n. 5 orthography, 849.
; at Atuatuca ? 118 n. 1 fairness of ;

Sulpicius Severus, 312 n. 3. Caesar's notices of, 249-50 how he ;

Sun-worship, 10, 26, 30. formed his column in marching

Supplem,entiim, 561, 804. from Atuatuca, 727-8.
Surus, 189. Tolls, 16.
Suzon, 799-800. Tolosa, 93.
Switzerland, 5, 38 ethnology of,
; Tolosates, 225, 480.
275-6, 318. ^eeHelvetii, Nantuates, Tongres, 105, 373-9, 382, 384, 706-8.
Seduni, Veragri. Tonnerre, 56, 631, 633, 796.
Toulon-sur-Arroux, .50-1, 53, 625,
Table of Peutinger, 349. 627.
Tabula, 450, 453. Tourmente, 189, 191, 489-91.
Tamahu, 273 n. 3. Towers (movable), in siege of Atua-
Taranis, 29-30. tucan stronghold, 81 used by ;

Tarbelli, 477-9. Gauls in siege of Q. Cicero's camp,

Tarn, 4 Lucterius threatens to cross,
; 113, 250-1 stationary tower built
;

135. on Caesar's second bridge over

Tarusates, 479. Rhine, 123 movable towers on
;

Tarvos Trigaranus, 27, 31. agger at Avaricum, 140, 144-6,

Tasgetius, 18, 106, 504, 520-2. 602-3, 606-7, 745 towers erected ;

Telamon, battle of, 2, 16. by Gauls at Avaricum, 143-4

Temples, 31-2. Caesar's towers at Alesia, 172, 177,
Tencteri, cross the Rhine, 95 ; 179, 810; tower at Uxellodunum,
Caesar's campaign against, 96-9 ; 192, 835.
credibiHty of Caesar's statement of Traders, 44, 60, 126.
their numbers, 233 effect of his
; Transalpine Gaul. See Gaul.
massacre of, in deterring Germans Transmigration of souls, 33-5.
from crossing Rhine, 99, 244 ques- ; Trebatius, 101.
tion of place where they crossed Trebonius, C, quartered at Samaro-
Rhine, 689-90 question of locality
; briva (54 B.C.), 105, 371 n. 7;
of their defeat, 691-706. Caesar marches with his legion to
Tenth legion. See Legions. relieve Q. Cicero, 116; campaigns
Terrasidius, T., 678-9. in south-western part of country of
Tertiary man, alleged traces of, in Eburones, 124 ; disarms Vellauno-
Gaul, 6, 268-9, 336. dunum, 137.
Testudines, 609-10. Treveri, auxiliary cavalry of, desert
Teutates, 29-30. Caesar in battle with Nervii, 78 ;

Teutomiitus, joins Vercingetorix, 147 ;

Labienus winters near western fron-
surprised in attack on Gergovia, tier of (54-53 B.C.), 105, 371, 373,
156. See also 504. 377, 380, 383-4, 732-4 ; disaffection
Teutoni, 36-7, 61, 81, 175, 546-53, of Treveri (54 B.C.), 106; Labienus
849. hard pressed by, 116; rebellion of
Themistocles, 25. (54-53 B.C.), 118-23; forced by
Thuringia, 123. attacks of Germans to hold aloof
Tiberius, 194 n. 1. from rebellion of Vercingetorix, 133,
Tigurini, defeat a Roman army (107 164, 174; finally subdued by
B.C.), 38, 555; defeated by Caesar, Labienus, 189 ; were they a Belgic
49, 231-3, 239-40, 617-9 their ; people ? 394-5 ; their territory de-
territory conjecturally defined, fined, 480-1 did Usipetes and Tenc-
;

480. teri enter their country ? 700, 702-

Tille, 167, 796, 798-801. 3 ; did Caesar build his first bridge
Timagenes, 34. over the Rhine in their country ':

Titurius Sabinus, holds bridge over 707-9 orthography, 849.

;

Aisne, 72-3, 666-8; sent to dis- Treves, 89 two legions quartered

;

perse northern allies of Veneti, 89, near (53-52 b. c. ), 128. ^ee Treveri.
 ;;

INDEX 871

Tribes, Gallic, mutual relations of, questions regarding Caesar's opera-

22-5, 104-5, 194, 197-8, 517-9. tions, 685-8. See also 851.
Triboci, 481-2, 638. Veragri, 82, 677-8 their territory de-
;

Tribunes, military, 43, 60, 79, 84, 112, fined, 453-4. See Martigny.
565-7. Verbigeni, 233, 235.
Tribute, 103 n. 2, 107, 193, 509, 518, Vercassivellaunus, 175 attacks Ro- ;

725, 838. man camp on Mont Rea, 177-8,

Tricasses, 445, 472. 815 ; captured, 180 orthography, ;

Tricorii, 501-2. 850.

Trigueres, 495-6. Vercellae, 37.
Troucillus, 52, 64, 652. Vercingetorix, coins of, 17 rebels ;

Trou du Frontal, 274. against Caesar, chosen king and

Troyes, 167, 788, 797. commander-in-chief, raises an army,
Tulingi, 46, 51 ; in battle near 131-2, 504 sends Lucterius to deal
;

Bibracte, 55, 233-4, 621, 627, 629- with Ruteni, enters country of
31 ; their territory conjecturally Bituriges, who join him, 133-4
defined, 438-42. forced by Caesars strategy to
Tumuli, 8, 10, 305, 312 n. 3. return to country of Arverni, 135 ;

Turoni, 483, 849. besieges Gorgobina, 136 raises ;

siege and attempts to recover

Ubii, Caesar invites Usipetes and Noviodunum, 138 persuades Bi- ;

Tencteri to settle in their country, turiges and other tribes to burn

97, 694-5, 697, 699, 701-3; beg towns and granaries, 138-9, 740-2 ;

Caesar to cross Rhine, 100, 702-3 ; obliged to consent to defence of

Caesar enters their country, 100 ; Avaricum, 139, 742 encamj)S near ;

give him information about move- Avaricum, 139, 742-3 ; harasses

ments of Suebi, 123 their territorv.
; Caesar, 141-2 moves nearer Avari-
;

697 n. 5. cum, 141, 743-4 outwits Caesar, ;

Uceni, 501-2. 141-2, 744-5 refutes charge of

;

Usipetes, cross the Rhine, 95; Caesar's treachery, 142-3 advises garrison ;

campaign against, 96-9 ; credibi- to evacuate Avaricum, 145 con- ;

lity of Caesar's statement of their soles troops for loss of Avaricum,

numbers, 233 question of place
;
146-7 ; raises fresh levies, 147 ;
where they crossed Rhine, 689-90 ; destroys bridges over Allier, 149 ;

question of locality of their defeat, plants himself on hill of Gergovia,

691-706. 149-50, 757 ; diligent in command,
Ussel, 485. 150-1 ; bribes Convictolitavis to
Uxellodunum, blockade and capture join rebellion, 152 fortifies western ;

of, 189-92 question of its site,

; approach to Gergovia, 155, 758-9 ;

483-93 height of Caesar's tower

; defeats Caesar at Gergovia, 157-8 ;
on terrace at, 835. unable to harass Caesar's retreat
Uzerche, 483 n. 5. from Gergovia, 159 joined by ;

Aedui, 159 resists their claim to

;

Vadimo, Boi defeated near lake of, 2. direct campaign, re-elected com-
Vangiones, 447, 493-4. mander-in-chief at Bibracte, 164-5 ;

Vapincum. See Gap. his plan of campaign, fortifies and

Vasates, 500. provisions Alesia, attempts to gain
Veliocasses, 185 their territory de-
; Roman Province, 165-6, 794 ; at-
fined, 494 ; orthography, 850. tacks Caesar with his cavalry near
Vellaunodunum, captured by Caesar Dijon, 167-8, 770-801 ; retreats,
(52 B.C.), 137; question of its site, beaten, to Alesia, 169, 801; failure
494-8. of his first sortie, 170-1, 805-6 ;
Vellavii, 499 orthography, 850.
; sends out cavalry to fetch succour,
Venelli, 84 n. 1, 89, 678-9; their 171, 806-7 ; economizes stores, 173
territory defined, 499 ; whereabouts army organized for his relief, 173-5,
did Sabinus encamp in their coun- 821-4; his final stand, 176-80,
try ? 688 ; orthography, 849-50. 812-20 surrender, imprisonment,
;

Veneti, their rebellion, 87-90 Caesar's

; and execution, 180, 820-1 place in ;

campaign against, 90-1 ; credibility history, 181-2 other chiefs jealous

;

of Caesar's narrative of their defeat, of him, 182, 197-8, 521, 816, 821-4 ;
236-7 their territory defined, 499
; ; credibility of Caesar's notices of,
question of theatre of war, 679-85 ; 213, 243-9 was his rebellion a
;
872 INDEX
democratic movement ? 529-41 ; Viridovix, 91-2, 532, 539.
orthography, 850. Viromandui, 76, 78 ; their territory
Vergobrets, 21 did two hold office
; defined, 499-500 ; orthography, 850.
simultaneously in one state ? 505-7. Vocates, 500-1.
See Convictolitavis, Cotus, Dum- Vocontii, 501-2, 615-6.
norix, Liscus. Volcae, 165, 502-3.
Verodunenses, 444, 447. Volunteers, 166. See Evocati.
Vertico, 115, 116 n. 2. Volusenus, C, at Octodurus, 84;
Vesontio, Ariovistus marches for, 60, attempts to assassinate Commius,
636-7 occupied by Caesar, 60,
; 129.
631-8 panic in Caesar's army at,
;
Vosges, 62, 64, 640-1, 648-51.
60-1 garrisoned by Caesar, 60, 62
; ; Vulgientes, 502.
legions probably quartered there
(58-57 B.C.), 69. See also 69-70, Waal, 691-704.
167, 359 n. 1. Walloons, 323-5, 334.
Vidal de la Blache, 25-6. Walls, Gallic, 80 n. 4, 143 ;
questions
Viducasses, 424, 444. relating to their construction, 746-8.
Vienna (Vienne), 136, 363, 617. Wine, 16.
Vieux-Laon, 72. See Bibrax. Wissant, question of its identity with
Vilaine, 90, 680. Portus Itius discussed, 432-8.
Vineae, 608, 749-50. Writing, 17.
Viridomarus, 152; seizes Noviodunum,
159 ; one of four generals in com- Yonne, 161, 167, 787-8.
mand of army destined for relief of
Alesia, 175. Zama, 591, 594-5.

Oxford : Printed at the Clarendon Press by Horace Hart, M.A.

 —

BY THE SAME AUTHOR

8vo, 2^p' ^^h ^^=^3 tvith Forty-four Illustrations in the Text and
Three Mwps. Price One Guinea net. Clarendon Press

ANCIENT BRITAIN
AND

THE INVASIONS OF
JULIUS CAESAR

Extracts from Notices by the Press

'Mr. Holmes brings book a power of lucid, vigorous

to his
narrative, a faculty for clearly and directly stating results, an
unflagging and devoted industry, and a full share of that in-
dispensable accuracy without which industry wastes its hours.
, , . The spirited story of Caesar's operations in Britain is in his
best style and is admirable in itself. To our thinking, it is the
ablest section in the book, and well worthy of the historian of
the Mutiny. . . . The volume before us is the work of a real
scholar. Times.

^
Dr. Rice Holmes has placed both antiquarians and^ historians
under a deep debt of gratitude by the production of his monu-
mental work on ''Ancient Britain". For the first time the vast
stores of material scattered through unnumbered monographs and
the proceedings of endless Societies have been systematically
examined and collated, and brought into an orderly narrative^
every step of which is supported by authority. The first real
inroad into British prehistoric times from the historical point of
view has now been made. It is impossible to deal adequately
. . .

in a brief review with a volume of such high qualities as that

before us. Every problem that arises is dealt with in an entirely
scientific spirit. He has a fine prose style, a pungent pen that
. . .

does not spare pretentious amateurs or hide-bound specialists,

a spring of humour not unlike Maitland s great gift, a historic
sense that enables him to see behind his material, and a certain
poetic instinct which gives to pre-history a reality that it has
never possessed before.' Contemporary Revieiv.

'Mr. Holmes . . . has redeemed English scholarship from the

reproach of half-heartedness. . . . His labours have indeed borne
much fruit ; and if we have selected for discussion a case in
which the seem to lack certainty, this only throws into
results
relief the fact that most of his conclusions compel our assent.'
Mr. H. Stuart Jones in English Historical Beview.
*
The hopes by the announcement that Mr. Holmes was
raised
to deal with Britain are more than realised by a careful reading,
and it is a pleasant duty to record our sincere admiration of his
book both in principle and detail.' Mr. Keginald A. Smith in —
The Classical Revieiv.
'
Upon most of the vexed questions connected wdth the two
expeditions of Caesar into Britain this book is a final and
authoritative publication . . . convincing logic is enlivened by
a fascinating literary style. . . . Dr. Holmes will have the gratitude
not only of scientific students of the subject but also of all

secondary school teachers, for whom books will be a

these
constant guide and inspiration.' — Professor Walter Dennison of
the University of Michigan, in Classical Philology.

Je crois que cette premiere partie constitue un des meilleurs

'

manuels de prehistoire qu'on ait ecrits en Angleterre, ou le

litterature classique compte quelques ouvrages tout a fait re-
marquables. Cette superiorite tient aux qualites d'ecrivain de
I'auteur, a son erudition, qui aime franchir les limites de son
pays, a son eclectisme eclaire, a la sagesse de ses conclusions.
. . . Les chapitres sur la Neolithique, et I'age du bronze et I'age
du fer, sont des plus agreables a lire . . . Ce livre fera bonne
figure dans la bibliotheque des archeologues franyais.' — Professor
Marcellin Boule in L'Anthropologie.

'
C'est un
bon livre que V Ancient Britain de M. T. Kice
bel et
Holmes forme un digne pendant a la Co7iquest of Gaul du
et qui
meme auteur. On y trouve le resultat de minutieuses recherches,
poursuivies avec patience pendant de longues annees on y sent ;

partout le noble enthousiasme du savant qui s'est passionne pour

son entreprise. Plein de faits, bien compose, d'une lecture
attrayante, I'ouvrage a une reelle valeur scientifique. Toutes . . .
[les dissertations] sont interessantes, documentees avec soin, con-
duites avec une critique judicieuse, et bien qu'elles s'adressent
surtout a des specialistes, historiens ou archeologues, le grand
public, qui aura lu la premiere partie de I'ouvrage, pourra encore
y trouver plaisir et profit.' — M. J. Vendkyes in Journal des
Savants.

'Vraiment, c'est I'oeuvre, cette fois, d'un historien, et je ne

serais pas etonne, si les circonstances sont favorables a M. Rice
Holmes, que I'Angleterre devienne un jour fiere de lui. . . .

Sur cet itineraire [Caesar's, as determined by R. H.] je me suis

explique deja avec M. Eice Holmes. Je I'adopte sans aucun
—
changement.' Professor Camille Jullian in Bevue des Etudes
anciennes.

'This work, like its predecessor, shows throughout the care,

thoroughness, and acuteness characteristic of the author ; indeed,
compared with the former work, it marks an advance, inasmuch
as the author estimates the work of philologists more justly than
before ; and with his own treatment of philological questions
one can almost always agree . . . his conclusions invariably show
sound judgement. The book as a whole is a thorough,
. . .

careful, and in every respect excellent work on the ancient

history of England.'

(*Auch dieses Werk zeigt iiberall die Sorgfalt, Griindlichkeit

und den Scharfsinn des Vf.s, ja es zeigt gegentiber dem alteren
Werke einen Fortschritt insofern, als der Vf. auch die Arbeit der
Philologen gerechter wiirdigt als friiher ; und mit seiner eigenen
Behandlung philologischer Fragen kann man fast stets einver-
standen sein . . . zeigt bei der Entscheidung stets gesundes
Urteil .Das Ganze ist eine griindliche, sorgfaltige und in
. .

jeder Beziehung vorziigliche Arbeit liber die alteste Geschichte

Englands.'-— Professor H. Meusel in Jaliresherichte des philologiscJien
Vereins zu Berlin.)

OXFORD
AT THE CLARENDON PRESS
LONDON HENRY FROWDE, AMEN CORNER, E.C.
:

EDINBURGH, CiLASGOW, NEW YORK; TORONTO

AND MELBOURNE
THE INSTITUTE OF MEDIAEVAL STUDIES
10 ELMSLEY PLACE
TORONTO 6, CANADA,