Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

The Historian's Craft

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 100

The istbrian's Craft Reflections on the Nature and Uses of History and the

Techniques and Methods of Those Who Write It.


800XS BT MARC BLOCH THE HISTORIAN'S GRAFT [Apdoge pour
VHistoire, ou Mdtier tfHistorien] STRANGE DEFEAT*. A STATEMENT OF
EVIDENCE WRITTSir IN I94O [UEtrange Difaite, Temoignage tcrit en 1940]
La SociStt jiodde: Les dosses et le Gouvemement des Hommes La SocUti
jiodde: La Formation de liens de Dtpendance Les Caract&res origmaux de
VHistoire mrde frangaise Les Rots thaumaturges: £tude dur le Caractire
mrnaturd attribui & la puissance royale, particuU&rement en France et en
VAngleterre Rots et Serfs: Un Chapitre de VHistoire capitienne Vile de France
(Les Pays autour de Paris)

THE HISTORIAN'S CRAFT


THE HISTORIAN'S CRAFT Toy Marc Bloch INTRODUCTION BY
JOSEPH R. STRAYER TRANSLATED FROM THE FRENCH BY PETER
PUTNAM A Caravelle Edition VINTAGE BOOKS A DIVISION OF
RANDOM HOUSE New York
Copyright, 1953, by Alfred A. Knopf, Inc. All rights reserved under
International and Pan-American Copyright Conventions. Distributed in Canada
by Random House of Canada, Limited, Toronto Manufactured in the United
States of America Reprinted by arrangement with Alfred A. Knopf, Inc. 47 46
45 44 43 42 41 40 39 38 37 36

TO
LUCIEN FEBVRE
BY WAY OF A DEDICATION

If this book should one day be published—if, begun as a simple antidote by


which, amid sorrows and anxieties both personal and collective, I seek a little
peace of mind, it should turn into a real book, intended to be read—you will
find, my friend, another name than yours inscribed upon its dedication page.
You can surmise the name this place requires; it is the one permissible allusion
to a tenderness too deep and sacred to be spoken. Yet how can I resign myself
to seeing you appear in no more than a few chance references? Long have we
worked together for a wider and more human history. Today our common task
is threatened. Not by our fault We are vanquished, for a moment, by an unjust
destiny. But the time will come, I feel sure, when our collaboration can again be
public, and again be free. Meanwhile, it is in these pages filled with your
presence that, for my part, our joint work goes on. It will keep what was always
its rhythm of fundamental agreement, enlivened, on the surface, by the
profitable interplay of our affectionate discussions. Certainly, more than one of
the ideas which I propose to uphold I have taken straight from you. With many
of the others, I cannot, in honesty, decide whether they are yours, mine, or both
of ours. I flatter myseli that you will often approve. And you will sometimes
rebuke me. In either case, there will be another bond between us.

Fougères, Creuse,
May 10, 1941

INTRODUCTION

Western man has always been historically minded, and this trait has been
accentuated during the last two centuries. Laymen are more aware than ever
before that they are living and making history—witness the care with which
great business organizations are preserving their archives, and the determination
of our military authorities to have the history of their commands written "while
it is hot." Certainly the number of historians, both professional and amateur, has
greatly increased in recent years, as has the quantity of historical writing—
quality is another matter. We have histories of games and histories of mail-order
houses, histories of diseases and histories of delusions, histories of
transportation and histories of highways, as well as the old standard mixtures of
political, economic, and social history.
Yet the more history we write the more we worry about the value and
nature of history. The increase in the number of books on historiography and
historical methodology is proportionally far greater than the increase in the
number of historians. Such books have been especially numerous in the last ten
or fifteen years, for obvious reasons. We are all asking, as the author of this
book asks in his first sentence: "What is the use of history?" What is the use of
history, when the values of the past are being ruthlessly discarded? What is the
use of history, when we repeat our old errors over and over again? And even if
we are sure that history has its uses, are we able to write the kind of history that
can be used?
These are the questions that troubled Marc Bloch, as they have troubled so
many of his fellow workers. They must have pressed on him with almost
unbearable weight in the dark days of 1941, when he began this book. A veteran
of the First War, called back to the colors at the age of fifty-three in 1939, he
had seen the collapse of France and of everything in which he believed. His
Jewish ancestry made it impossible for him to return to his professorship at the
Sorbonne; he took refuge first with the exiled University of Strasbourg at
Clermont-Ferrand, then with the University of Montpellier. He could have fled
to the United States, but he refused to leave France, even the France of Vichy.
As he said in his testament, he was so thoroughly French, so impregnated with
the spirit and tradition of France, that he did not think he could breathe freely in
another country. And if the book was begun under evil auspices it was
continued under worse. When the Germans crossed the line of demarcation,
after the landings in North Africa, Bloch was driven from academic life. He
became a member of the Resistance, a leader of the group centering in Lyons.
There he was captured by the Germans in the spring of 1944, imprisoned, and
cruelly mistreated. On June 16, as the Nazi hold on France began to weaken, he
was taken from his cell and shot in an open field near Lyons with twenty-six
other patriots.
The book was never finished. But, in the large fragment which was
completed, we find no bitterness, no discouragement. Bloch kept his serenity,
his faith in France, and his belief in the value of history. He used a few of his
war experiences, as he used other episodes in his life, to illustrate attitudes and
beliefs which he felt were common to many men. But this book is not a product
of the war; it is the fruit of the long years of peaceful study and reflection which
made him a master of his trade.
As such, it is worth careful study. Not that Bloch was the greatest French
historian of his generation, though he would certainly rank high in any list. Not
even that he was the most widely read—others excelled in that art of combining
exact knowledge with readability which has distinguished French scholarship
for many years. His real eminence lay in the fact that he actually put into
practice those recommendations for a new kind of history which the profession
has been endorsing—and ignoring—for the last fifty years. Others have talked
about the narrowness of purely political history, the evils of excessive
specialization, and the unreality of the conventional periodization of history —
without ever leaving their own limited fields. Bloch not only said that history
was a whole, that no period and no topic could be understood except in relation
to other periods and topics, but he constantly taught and wrote in accordance
with this belief. Though his most important work was in medieval history he
gave a course, in his last years of teaching, on the economic development of the
United States. Though the Sorbonne listed him as professor of economic history
he never made the mistake of assuming that economic factors explain all human
behavior. He knew that man is not entirely rational, that society is held together
as much by beliefs and customs as by economic interests. He worked constantly
for a "wider, more human history/' for a history which described how and why
people live and work together. He saw life as a whole, as a complicated
interplay of ideals and realities, of conscious innovation and unconscious
conservation. When he discussed institutions, they were not the petrified
fictions of the lawyer, but the changing patterns which emerge from human life.
When he discussed ideas, they were not the bloodless, literary abstractions of
doctoral dissertations, but the hidden forces which determine behavior and the
structure of society. He was capable of infinite attention to detail, but he never
forgot that the details had meaning only in the larger framework of the history
of human society. All this is illustrated in his fine work on La socUU ftodale, a
book which is far more than a description of feudal institutions. Throughout the
two volumes he sought, above everything else, to understand and explain the
state of mind and the habits of life which could produce and support feudal
organization. He was a great practitioner of the new approach to
Introduction 3d history in another way, in his ability to discover and use
new types of source material. Here again, historians have been preaching for
many years that written records are not enough, that we must learn to follow
other traces of man's activity in the past, but few historians have taken the
trouble to learn how to use these difficult materials. Bloch, in what was
probably his greatest book, Les caract&res originaux de Vhistoire rurale
frangaise, gave a perfect demonstration of how the job should be done. Old
maps, place-names, ancient tools, aerial surveys, folklore—all contributed to his
brilliant description of French society during the long centuries when
agriculture was the predominant occupation. These qualities were manifested
not only in his books, but in the review Annates d'histoire 4conomique et
sociale9 which he and Lucien Febvre founded in 1929. During the years before
the war no other historical periodical had as much influence on the rising
generation of scholars. In long review-articles Marc Bloch showed his
consuming interest in all aspects of history, his complete freedom from all
provinciality, and his generosity in recognizing or even overpraising the
contributions of others. In addition, he poured forth an almost inexhaustible
store of suggestions for further investigation. If all the books and articles which
he called for had been written we should be closer to that history of humanity of
which he dreamed. This book is his testament as a historian—a thoughtful,
honest statement by a great craftsman about the principles of his trade. Here he
expressed his aims,
za Introduction which were those of most historians of his own and
younger generations. Here he set forth his conviction of the unity of all history
and of the living connection between present and past which makes history
something more than a game for dilettantes. It is unfor* tunate that he could not
finish and polish his work— his style, never easy, is especially difficult in this
work and must have given his translator some bad half* hours. But, even
incomplete, Bloch has given us a noble statement of the historian's creed, a
guide to his fellow workers, and an explanation of the meaning of their work to
laymen.

Joseph R. Strayer

A NOTE ON THE MANUSCRIPTS


OF THE PRESENT BOOK
by Lucien Febvre

It is a delicate task, to be undertaken with many scruples, to prepare an


unfinished manuscript for publication, especially when even those parts that had
been given to the typist would certainly have received a last polishing from the
author before going to the printer. But such scruples are outweighed by the
satisfaction of making public, even in mutilated form, a notable book. Marc
Bloch long dreamed, as I have done, of putting down his ideas on history in an
organized way. I often think, with bitter regret, that while there was yet time we
should have collaborated to give our younger generation a kind of new Langlois
and Seignobos,1 to be the manifesto of another generation and the embodiment
of an entirely different spirit. It is too late. At any rate, Marc Bloch, when
events had deflected him from his path, attempted on his own to realize a plan
which we had often discussed together. I have elsewhere related how, serving as
a staff officer in Alsace, and restless under the idleness of the
"phony war/' he one day went to a storekeeper at Molsheim and bought a
schoolboy's notebook, no doubt just such a one as that in which Henri Pirenne,
interned in another village in the heart of Germany in the First War, wrote his
history of Europe. On the first page, Bloch wrote a title:
HISTORY OF FRENCH SOCIETY
IN THE STRUCTURE OF EUROPEAN CIVILIZATION.

Somewhat later, he composed the dedication: To the memory of Henri


Pirenne who, at the time his country was ftgjkting beside mine for justice and
civilizar Hon, wrote in captivity, a history of Europe. After which, according to
his custom, he drafted an introduction: Reflections for a Reader Interested in
Method. This is followed by a certain number of pages, still in manuscript,
constituting a first chapter, entitled: Birth of France and of Europe. The events


[Translator's note: This refers to the famous Introduction aux Etudes Historiques,
long used in courses in methodology both in France and in the United States.]
which Bloch has himself narrated in The Strange Defeat put an end to this work.
And when, returning to France after the tragic circuit from Dun- kerque to
London to Brittany, Bloch again set to work, it was to compose the present
book. Exactly when did he begin? I cannot say precisely. There is one early
date: at the bottom of the moving page which Bloch composed in my honor, we
read: "Foug&es, Creuse, May 10,1941." And on a loose sheet, inserted in one of
his files, one may also read:
A Note on the Manuscripts xv STATS OF WORK: MARCH 11, 1942. 1.
Write, in order to finish IV, generalities, civilizations, and read over. 2. Go on
to V (change, experience). After this latter date Bloch did in fact find the time to
finish Chapter IV and to begin Chapter V, to which he gave no final tide. That
was all. How would Bloch have finished his book? In the papers which have
been turned over to me, I have found no orderly plan for the projected work. Or,
rather, I have found one but it is anterior to the actual writing, and differs
considerably from the plan which he ultimately followed. He there anticipated
seven chapters. He entitled them, respectively: I. Historical Knowledge: Past
and Present II. Historical Observation. III. Historical Analysis. IV. Time and
History. V. Historical Experience. VI. Explanation in History. VII. The Problem
of Prevision. For a conclusion, Bloch intended to write a study on The Role of
History in Citizenship and Education* And he expected to devote an appendix
to the Teaching of History. The differences between this plan and the work as
executed need not be emphasized. If, in general, the
xvi A Note on the Manuscripts substance foreseen for the first five
chapters is to be found in the first four chapters of the present book, Bloch
would still have had to treat of the problem of chance, of the problem of the
individual, of the problem of "determinant" acts or facts; finally, of that problem
of prevision to which he would have had to devote an entire chapter. From these
indications we can see that we possess over two thirds of the work that he
conceived. It may be useful to transcribe here the latter part of the unfinished
plan: VI. EXPLANATION IN HISTORY. By way of introduction: the
generation of skeptics (and scientists). 1. The idea of cause. The destruction of
cause and of mo* tive (the unconscious). Romanticism and spontaneity. 2. The
idea of chance. 3. The problem of the individual and his differential value.
Supplementary, the epochs, documentarily without individuals. Is history only a
science of men in society? Mass history and the elite. 4. The problem of
"determinant" acts or facts. Vn. THE PROBLEM OF PREVISION. 1.
Prevision, a mental necessity. 2. The ordinary errors of prevision. Economic
fluctuations, military history. 3. The paradox of prevision in human affairs:
prevision which is destroyed by prevision; role of conscious awareness. 4.
Short-term prevision. 5. Regularities. 6. Hopes and uncertainties*
A Note on the Manuscripts xvii The absence of any more precise and
detailed notes by Bloch on these last parts of his book is profoundly to be
regretted. They would have been reckoned among the most original. Although I
well know his ideas— which are mine—on the questions raised by Chapter VII,
I believe that we never actually discussed this problem of prevision, which
Bloch promised to treat with so much judgment and originality at the end of the
work—and which, perhaps, would have been the most strictly original of the
whole* To establish the text of the present book, I have had before me three
large files, each a more or less complete copy of the text to be published. These
copies are in large part made up of typewritten sheets, amongst which are
interspersed manuscript sheets in Marc Bloch's hand, most frequently written
upon the back of a first draft which he had crossed out Essentially, my work as
editor has consisted in composing, from these three, one basic copy, complete in
all its pages and taking into account all the manuscript corrections which Bloch
himself made in the typewritten copy. No addition, no correction, even of mere
form, has been supplied to Bloch's text; it is this text, pure and entire, which is
to be found printed in this book. The work was to include references. We have
discovered only a few notes, written out in his hand. We have felt no obligation
to fill in this gap. A tremendous and not very profitable task, it would have
posed insoluble problems at every step.
xviii A Note on the Manuscripts Let me add that the three copies I
mentioned all end in the same way and with the same words: in history, as
elsewhere, the causes cannot be assumed. They are to be looked for. . . ."
Finally, and because it is a matter both of dedication and of solemn memory, I
cannot but say this: There was a person to whom Marc Bloch, before departing,
would have dedicated one of the great works that we still expected from him.
Those of us who knew and loved Marc Bloch were aware of the single-hearted
tenderness with which she enveloped him and his children—and of that
abnegation with which she had served him as secretary and helped in his labors.
I feel it as an obligation which nothing can prevent me from meeting—not even
that sense of sentimental reserve which was so strong with Marc Bloch—-I feel
it as a duty to set down here the name of Madame Marc Bloch, who died in the
same cause as her husband and in the same French faith.
TRANSLATOR'S NOTE I wish only to remark that, from the outset, the
translation of this book has been a co-operative venture, in which my friend and
teacher Robert R. Palmer, and my wife, Durinda, have divided the labor equally
with me. A further acknowledgment is due Joseph R. Strayer for the many
valuable suggestions made in his reading of the manuscript. PETER PUTNAM
[sxi] CONTENTS Introduction j Chapter I: History, Men, and Time 20 1.
The Choice of the Historian 20 2. History and Men 22 3. Historical Time 27 4.
The Idol of Origins 29 5. The Boundaries between Past and Present 35 6.
Understanding the Present by the Past 39 7. Understanding the Past by the
Present 43 Chapter II: Historical Observation 48 1. General Characteristics of
Historical Observation 48 2. Evidence 60 3. The Transmission of Evidence
Chapter III: Historical Criticism 79 1. An Outline of the History of the Critical
Method 79 2. In Pursuit of Fraud and Error 90 3. Toward a Logic of the Critical
Method 110 Chapter IV: Historical Analysis 138 1. Judging or Understanding?
138 2. From the Diversity of Human Functions to the Unity of Consciences 144
3. Nomenclature 156 Chapter V: Historical Causation 1QO
THE HISTORIAN'S CRAFT
[3] INTRODUCTION "Tell me, Daddy. What is the use of history?'' Thus,
a few years ago, a young lad in whom I had a very special interest questioned
his historian father. I wish I could say of this book that it is my answer. I can
conceive no higher praise for a writer than to be able to speak in the same tone
to savants and school- boys alike, but so noble a simplicity is the privilege of
the select few. At any rate, this question from a child, whose thirst for
knowledge I was not, perhaps, too well able to satisfy at the time, now serves
me well as a point of departure. Doubtless there are some who will consider this
a naive approach, but to me it seems entirely to the point.1 The problem which
it poses, with 1 In whioh I find myself, from the beginning, in an unlooked- for
opposition to the Introduction mix Etudes Historiques of Lan* glois and
Seignobos. The above passage had already been written* when, in the Foreword
of the latter (p. xii), I chanced to see a list of "idle questions." There, word for
word, appeared the following: "What is the use of history?" It is the same
doubtless with this problem as with any problem concerning the raison d*itre of
our thoughts and actions: those minds which by nature are indifferent to them—
or have intentionally determined to make themselves so— always find it
difficult to understand that other minds find them the subject of absorbing
reflections. Nevertheless, since the opportunity is thus offered me, I think it is
better immediately to establish my
4 The Historian's Craft the embarrassing forthrightness of that implacable
age, is no less than that of the legitimacy of history. Behold, then, the historian
called to render his accounts! He does so not without an inner tremor. What
craftsman, grown old in his trade, has not asked himself with a sudden qualm
whether he has spent his life wisely? The question far transcends the minor
scruples of a professional conscience. Indeed, our entire Western civilization is
concerned in it. For, unlike others, our civilization has always been extremely
attentive to its past. Everything has inclined it in this direction: both the
Christian and the classical heritage. Our first masters, the Greeks and the
Romans, were history-writing peoples. Christianity is a religion of historians.
Other religious systems have been able to found their beliefs and their rites on a
mythology nearly outside human time. For sacred books, the Christians have
books of history, and their liturgies commemorate, together with episodes from
the terrestrial life of a God, the annals of the church and the lives of the saints.
Christianity is historical in position, as regards a book which is justly famous
and which mine, arranged upon a different and, in certain of its parts, a much
less fully developed plan, does not by any means pretend to replace. I was the
pupil of both its authors, and particularly of M. Seignobos. Both showed me
valuable tokens of their good will. My education owed a great deal both to their
teaching and to their work. But both have not only taught us that the historian's
first duty is to be sincere; they fully appreciated that the very progress of our
studies is founded upon the inevitable opposition between generations of
scholars. Therefore, I shall be keeping faith with their teaching in criticizing
them most freely wherever I may deem it useful; just as I hope, some day, that
my pupils will criticize me in their turn.
Introduction 5 another and, perhaps, even deeper sense. The destiny of
humankind, placed between the Fall and the Judgment, appears to its eyes as a
long adventure, of which each life, each individual pilgrimage, is in its turn a
reflection. It is in time and, therefore, in history that the great drama of Sin and
Redemption, the central axis of all Christian thought, is unfolded. Our art, our
literary monuments, resound with echoes of the past. Our men of action have its
real or pretended lessons incessantly on their lips. Of course, differences of
group psychology can be noted. Cournot long ago observed that the French
people in the mass, everlastingly inclined to reconstruct the world on lines of
reason, live their collective memories much less intensely than the Germans, for
example.2 Without doubt, too, civilizations may change. It is not in itself
inconceivable that ours may, one day, turn away from history, and historians
would do well to reflect upon this possibility. If they do not take care, there is
danger that badly understood history could involve good history in its disrepute.
But should we come to this, it would be at the cost of a serious rupture with our
most unvarying intellectual traditions. For the present, our discussion has
reached only the * The antihistorical Frenchman: Cournot, Souvenirs, p. 43, on
the subject of the absence of any royalist sentiment at the end of the Empire,
remarks: ". . . for the explanation of the singular fact before us, I believe we
must also take into account the scant popularity of our history and the
underdeveloped consciousness of his* toxical tradition among our lower
classes, for reasons too lengthy fat analysis*"
6 The Historian's Craft stage of probing the conscience. And, indeed,
whenever our exacting Western society, in the continuing crisis of growth,
begins to doubt itself, it asks itself whether it has done well in trying to learn
from the past, and whether it has learned rightly. Read what was written before
the war, or, for that matter, what might be written today. Among the confused
murmur- ings of the present, you will almost certainly hear this complaint
mingling its voice with the others. I myself chanced to overhear its echo in the
very heart of the great drama. It was in June 1940—the very day, if I remember
aright, of the German entry into Paris. In a Norman garden, stripped of our
troops, we of the general staff consumed our idle hours in ruminating over the
causes of the disaster. "Are we to believe that history has betrayed us?" one of
us cried. So it was that the anguish of a mature man united its bitter accents with
the simple curiosity of the boy. Both demand an answer. "What is the use of
history?" What is here meant by "use"? But, before proceeding to this question,
let me insert one word of apology. The circumstances of my present life, the
impossibility of reaching any large library, and the loss of my own books have
made me dependent upon my notes and upon memory. Both the supplementary
reading and the research demanded by the very laws of the craft I here propose
to describe have been denied me. Will it, one day, be granted to me to fill in the
gaps? Never
Introduction 7 entirely, I fear. I can therefore only ask indulgence. I should
say: "I plead guilty," were it not that, by so doing, I might seem overly
presumptuous in assuming responsibility for the evils of destiny. Certainly,
even if history were judged incapable of other uses, its entertainment value
would remain in its favor. Or, to be more exact (for everyone seeks his own
pleasures), it is incontestable that it appears entertaining to a large number of
men. As far back as I can remember, it has been for me a constant source of
pleasure. As for all historians, I think. If not, why have they chosen this
occupation? To anyone who is not a blockhead, all the sciences are interesting;
yet each scholar finds but one that absorbs him. Finding it, in order further to
devote himself to it, he terms it his "vocation," his "calling." This
unquestionable fascination of history requires us to pause and reflect. Its role,
both as the germ and, later, as the spur to action, has been and remains
paramount Simple liking precedes the yearning for knowledge. Before the work
of science, fully conscious of its ends, comes the instinct which guides it Our
intellectual history abounds in examples of similar origins. Even physics began
with cabinets of curiosities, and the elves of antiquarianism have cut capers
about the cradle of more than one serious study. Such was the genesis of
archaeology and, more recently, of folklore. Readers
8 The Historian's Craft of Alexander Dumas may well be potential
historians who lack only training to find the purer and, to my way of thinking,
the keener pleasure of true research. Moreover, this charm will be far from
diminished once methodical inquiry, with all its necessary austerities, has
begun. On the contrary, all true historians will bear witness that the fascination
then gains in both scope and intensity. The same is true of any intellectual
discipline, but, of course, history has its peculiar aesthetic pleasures. The
spectacle of human activity which forms its particular object is, more than any
other, designed to seduce the imagination—above all when, thanks to its
remoteness in time or space, it is adorned with the subtle enchantment of the
unfamiliar. The great Leibniz himself admitted as much, and, when he turned
from abstract speculation on mathematics and theodicy to the deciphering of the
ancient charters and chronicles of Imperial Germany, he, like the rest of us,
experienced "the thrill of learning singular things/' Let us guard against
stripping our science of its share of poetry. Let us also beware of the inclination,
which I have detected in some, to be ashamed of this poetic quality. It would be
sheer folly to suppose that history, because it appeals strongly to the emotions,
is less capable of satisfying the intellect Nevertheless, were the nearly universal
fascination of history its only justification—if it were, in short, only a pleasant
pastime, like bridge or fishing—would it be worth all the trouble we take to
write it? To write
Introduction 9 it, I mean, with integrity, with truth, with the utmost
possible penetration into its hidden causes, and, hence, with difficulty? Mere
amusement, Andr£ Gide has written, is no longer permitted us in our day, even,
he added, when it is the amusement of the intelligence. That was said in 1938.
In 1942, as I write in my turn, how much graver a significance the remark
assumes! Surely, in a world which stands upon the threshold of the chemistry of
the atom, which is only beginning to fathom the mystery of interstellar space, in
this poor world of ours which, however justifiably proud of its science, has
created so little happiness for itself, the tedious minutiae of historical erudition,
easily capable of consuming a whole lifetime, would deserve condemnation as
an absurd waste of energy, bordering on the criminal, were they to end merely
by coating one of our diversions with a thin veneer of truth. Either all minds
capable of better employment must be dissuaded from the practice of history, or
history must prove its legitimacy as a form of knowledge. But here a new
question arises. What is it, exactly, that constitutes the legitimacy of an
intellectual endeavor? No one today, I believe, would dare to say, with the
orthodox positivists, that the value of a line of research is to be measured by its
ability to promote action. Experience has surely taught us that it is impossible to
decide in advance whether even the most abstract speculations may not
eventually prove extraordinarily helpful in practice. It would inflict a strange
mutilation
10 The Historian's Craft upon humanity to deny it a right to appease its
intellectual appetites apart from all consideration of its material welfare. Even
were history obliged to be eternally indifferent to homo faber or to homo poli-
ticus, it would be sufficiently justified by its necessity for the full flowering of
homo sapiens. Yet, even with this limitation, the question is not immediately
resolved The nature of our intelligence is such that it is stimulated far less by
the will to know than by the will to understand, and, from this, it results that the
only sciences which it admits to be authentic are those which succeed in
establishing explanatory relationships between phenomena. The rest is, as
Malebranche put it, mere "polymathy." Now, polymathy can well assume the
form either of recreation or of mania, but it cannot today, any more than in the
time of Malebranche, pass for one of the proper tasks of the intellect. Even apart
from any application to conduct, history will rightfully claim its place among
those sciences truly worthy of endeavor only in proportion as it promises us, not
simply a disjointed and, you might say, a nearly infinite enumeration, but a
rational classification and progressive intelligibility. However, it is undeniable
that a science will always seem to us somehow incomplete if it cannot, sooner
or later, in one way or another, aid us to live better. Moreover, should we not
feel this sentiment with particular force as regards history, so much the more
clearly destined to work for the profit of man, in that
Introduction 11 it has man himself and his actions for its theme? In fact, a
long-standing penchant prompts us, almost by instinct, to demand of it the
means to direct our actions and, therefore, as in the case of the conquered
soldier mentioned above, we become indignant if, perchance, it seems incapable
of giving us guidance. The question of the use of history, in the strict and
"prag* matic" sense of the word "use/" is not to be confounded with that of its
strictly intellectual legitimacy. Moreover, this question of use must always
come second in the order of things, for, to act reasonably, it is first necessary to
understand. Common sense dictates that we no longer avoid this problem.
Certain among our would-be counselors have already given answers to these
questions. They have sought to chide our optimism. Hie most indulgent have
said that history is both unprofitable and unsound; others, with a severity which
admits of no compromise, that it is pernicious. One of them, and not the least
celebrated, has declared it "the most dangerous compound yet contrived by the
chemistry of the intellect" These condemnations offer a terrible temptation, in
that they justify ignorance in advance. Fortunately for those of us who still
retain our intellectual curiosity, there is, perhaps, an appeal from their verdict.
But if the debate is to be revived, it is important that it be based upon more
trustworthy data. For there is one precaution which the ordinary detractors of
history seem not to have heeded. Their
12 The Historian's Craft words lack neither eloquence nor wit, but they
have, for the most part, neglected to ask themselves exactly what it is they are
discussing. The picture which they have formed for themselves of our studies
has not been drawn in the workshop. It savors rather of the debat- ing-platform
than of the study. Above all, it is out of date. Therefore, when all is said and
done, it may well be that all their energy has been expended only to conjure
away a phantom. Our effort here must be very different. The methods whose
value and certainty we shall attempt to assess are those actually used in
research, right down to the lowly and delicate technical details. Our problems
will be the same as those which the historian's material imposes upon him every
day. In a word, our primary objective is to explain how and why a historian
practices his trade. It will then be the business of the reader to decide whether
this trade is worth practicing. Let us take care, however. Even thus defined and
limited, the task is not so simple as it seems. It might be, were we dealing with
one of the practical arts which are sufficiently explained when time-tested
manual operations are enumerated one after another. But history is neither
watchmaking nor cabinet construction. It is an endeavor toward better
understanding and, consequently, a thing in movement To limit oneself to
describing a science just as it is will always be to betray it a little. It is still more
important to tell how it expects to improve itself in the course of time. Now,
such an undertaking inevitably involves a
Introduction 13 rather large dose of personal opinion. Indeed, every
science is continually beset at each stage of its development by diverging
tendencies, and it is scarcely possible to decide which is now dominant without
prophesying the future. We shall not shirk this obligation. The dread of
responsibility is as discreditable in intellectual matters as in any others. But it is
only honest to give the reader fair warning. The more so, as the difficulties
which every study of methodology encounters vary greatly according to the
point which the particular discipline has reached upon the always irregular
curve of its development For example, fifty years ago, when Newton still
reigned supreme, it was far easier than today to frame, with all the precision of a
blueprint, a thesis on mechanics. But history is still in that stage which is very
indulgent of statements of positive certainties. For history is not only a science
in movement. like all those which have the human spirit for their object, this
newcomer in the field of rational knowledge is also a science in its infancy. Or
to explain more fully, having grown old in embryo as mere narrative, for long
encumbered with legend, and for still longer preoccupied with only the most
obvious events, it is still very young as a rational attempt at analysis. Now, at
last, it struggles to penetrate beneath the mere surface of actions, rejecting not
only the temptations of legend and rhetoric, but the still more dangerous modern
poisons of routine learning and empiricism parading as common sense. In
several of the most essential
14 The Historian's Craft problems of method, it has not passed beyond the
first tentative gropings, and that is why Fustel de Coulanges and, even before
him, Bayle came very near the truth when they called it "the most difficult of all
the sciences." But is this merely an illusion? However uncertain our road at
many points, we are, it seems to me, at the present hour better placed than our
predecessors to see a little light on the path ahead. The generations just prior to
our own, in the last decades of the nineteenth century and even in the first years
of the twentieth, were as if mesmerized by the Comtian conception of physical
science. This hyp* notic schema, extending to every province of the intellect,
seemed to them to prove that no authentic discipline could exist which did not
lead, by immediate and irrefutable demonstrations, to the formulation of
absolute certainties in the form of sovereign and universal laws. Such was the
nearly unanimous opinion at the time, but when applied to historical studies it
gave birth, depending upon the temperament of the individual historian, to two
opposing schools. The first believed it really possible and tried their best to
establish a science of human evolution which would conform to a sort of pan-
scientific ideal. They were willing to abandon, as outside a true science of man,
a great many eminently human realities which appeared to them stubbornly
insusceptible to rational comprehension. This residue they scornfully called
Introduction 15 mere events or happenstance. It was also a good part of the
most intimate and individual side of life. Such was, in sum, the position of the
sociological school founded by Durkheim. (Of course the early rigidity of
principle was gradually softened in practice, though reluctantly, by men too
intelligent not to yield before the force of things as they are.) To this great
scientific effort our studies are vastly indebted. It has taught us to analyze more
profoundly, to grasp our problems more firmly, and even, I dare say, to think
less shod- dily. It will be spoken of here only with infinite gratitude and respect.
If it seems sterile now, that is only the price that all intellectual movements
must pay, sooner or later, for their moment of fertility. The other school of
inquirers took a quite different point of view. Unsuccessful in cramming the
stuff of history into the legalistic framework of physical science, and
particularly disturbed, because of their early training, by the difficulties, doubts,
and many fresh beginnings required by documentary criticism, they drew from
their inquiries the moral lesson of a disillusioned humility. In the final
reckoning, they felt that they were devoting their talents to a discipline which
promised neither very positive conclusions in the present, nor the hope of
progress in the future. They tended to view history less as truly scientific
knowledge than as a sort of aesthetic play, a hygienic exercise favorable to
health of mind. They have sometimes been called historiens historisants,
possessing the truly "historical" point of view; but such a judgment
16 The Historian's Craft does injury to our profession, for it seems to find
the essence of history in the very denial of its possibilities. For my part, I should
prefer to find a more expressive symbol for them in the moment of French
thought with which they are associated. The amiable and skeptical Sylvestre
Bonnard, if we accept the dates which Anatole France's book assigns his doings,
is an anachronism, quite like the old saints whom the writers of the middle ages
naively depicted in the colors of their own time. Sylvestre Bonnard (if we grant
that fictitious character a moment of existence in the flesh)—the "real"
Sylvestre Bonnard, born under the First Empire—would have belonged to the
generation of the romantic historians. He would have shared their stirring and
prolific enthusiasms, their ingenuous faith in the future of the "philosophy" of
history. Let us pass over the epoch in which he is supposed to have lived. Let us
restore him to the period in which his imaginary life was written by Anatole
France. He can then be regarded as the patron saint of a whole group of
historians, roughly the intellectual contemporaries of his biographer. They were
profoundly honest workmen, but a little short-winded. We may compare them
to the children of debauched fathers, their constitutions had been weakened by
wild historical orgies of romanticism. They felt rather small beside their
colleagues in the laboratory; they were more inclined to recommend caution
than daring. Even so vigorous an intelligence as that of my beloved teacher,
Charles Seignobos, once
Introduction 17 let fall a saying that may fairly stand as their slogan: "It is
useful to ask oneself questions, but very dangerous to answer them." Surely,
this is not the remark of a braggart, but where would physics be today if the
physicists had shown no greater daring? Our mental climate has changed. The
kinetic theory of gases, Einstein's mechanics, and the quantum theory have
profoundly altered that concept of science which, only yesterday, was
unanimously accepted. They have not weakened it; they have only made it more
flexible. For certainty, they have often substituted the infinitely probable; for
the strictly measurable, the notion of the eternal relativity of measurement.
Their influence has even affected the countless minds (and, alas, I must number
mine among them) which, thanks to defects in intelligence or early training,
have been able to follow the great metamorphosis only at a distance and as if by
a reflected light. Hence, we are much better prepared to admit that a scholarly
discipline may pretend to the dignity of a science without insisting upon
Euclidian demonstrations or immutable laws of repetition. We find it far easier
to regard certainty and universality as questions of degree. We no longer feel
obliged to impose upon every subject of knowledge a uniform intellectual
pattern, borrowed from natural science, since, even there, that pattern has
ceased to be entirely applicable. We do not yet know what the sciences of man
will some day be. We do know that in order to exist—and, it goes without
saying, to exist in accordance with the fundamental laws of reason—they
18 The Historian's Craft need neither disclaim nor feel ashamed of their
own distinctive character. I should like professional historians and, above all,
the younger ones to reflect upon these hesitancies, these incessant soul-
searchings, of our craft. It will be the surest way they can prepare themselves,
by a deliberate choice, to direct their efforts reasonably. I should desire above
all to see ever-increasing numbers of them arrive at that broadened and
deepened history which some of us—more every day—have begun to conceive.
If my book can help them, I shall feel that it was not in vain. I confess that that
is, in part, its aim. But I do not write exclusively, or even chiefly, for the private
use of the guild. The uncertainties of our science must not, I think, be hidden
from the curiosity of the world. They are our excuse for being. They bring
freshness to our studies. Surely we have the right to claim for history the
indulgence due to all new ventures. The incomplete, if it is perpetually straining
to realize itself, is quite as enticing as the most perfect success. To paraphrase
P£guy, the good husbandman takes as much pleasure in plowing and sowing as
in the harvest It is fitting that these few words of introduction be concluded with
a confession. Each science, taken by itself, represents but a fragment of the
universal march toward knowledge. I have given an example above; in order to
understand and appreciate one's own methods of investigation, however
specialized, it is indispensable
Introduction 19 to see their connection with all simultaneous tendencies in
other fields. Now this study of methods for their own sake is, in its turn, a
specialized trade, whose technicians are called philosophers. That is a title to
which I cannot pretend. Through this gap in my education this essay will
doubtless lose much in both precision of language and breadth of horizon. I
submit it for what it is and no more: the memorandum of a craftsman who has
always liked to reflect over his daily task, the notebook of a journeyman who
has long handled the ruler and the level, without imagining himself to be a
mathematician.
[20] CHAPTER I HISTORY, MEN, AND TIME J. The Choice of the
Historian The word "history" is very old—so old that men have sometimes
grown weary of it. It is true that they have rarely gone so far as to wish to erase
it from the vocabulary entirely. Even the sociologists of the Durk- heim school
make room for it They do so, to be sure, only in order to relegate it to one poor
corner of the sciences of man—a sort of secret dungeon in which, having first
reserved for sociology all that appears ta them susceptible of rational analysis,
they shut up the human facts which they condemn as the most superficial and
capricious of all. Here, on the contrary, we shall preserve the broadest
interpretation of the word "history." The word places no a priori prohibitions in
the path of inquiry, which may turn at will toward either the individual or the
social, toward momentary convulsions or the most lasting developments. It
comprises in itself no credo; it commits us, according to its original meaning, to
nothing other than "inquiry." Assuredly, since its first appearance on the lips of
men, more than two millenniums ago, its content has changed a great deal. Such
u the fate of all truly living terms in a language. If the
History, Men, and Time 21 sciences were obliged to find a new name each
time they made an advance—what a multitude of christenings! and what a
waste of time tor the academic realml In remaining quietly loyal to its glorious
Hellenic name, our history need be no more like that of Heca- taeus of Miletus
than the physics of Lord Kelvin or Langevin is like that of Aristotle. What, then,
is this history of ours? At the start, while focusing our attention upon the red
problems of investigation, it would be pointless to draw up a tedious and
inflexible definition. What serious workman has ever burdened himself with
such articles of faith? It is not only that their meticulous precision omits the best
in every intellectual creation —the half-formed impulse toward a knowledge
still undetermined but capable of extension. The worst danger of such careful
definitions is that they only bring further limitations. "This subject," declares
the Divine Lexicographer, "or that means of treating it, is, no doubt seductive,
but—take care, O young apprentice!—it is not history!" Are we then the rules
committee of an ancient guild, who codify the tasks permitted to the members
of the trade, and who, with a list once and for all complete, unhesitatingly
reserve their exercise to the licensed masters? * The physicists 1 [Translators
note: The following note is only a fragment on a loose sheet. The beginning is
lost] ... as Lucien Febvre has pointed out, history itself, when consulted as to the
path which the development of mankind has followed, is obliged to contradict
them most flagtantly. Not only does each science, taken separately, find its most
successful craftsman among the refugees from neighboring
22 The Historian's Craft and chemists are wiser—so far as I know, they
have never been seen to quarrel about the respective rights of physics, of
chemistry, of physical chemistry, or (assuming the existence of such a term) of
chemical physics. It is no less true that, faced with the vast chaos of reality, the
historian is necessarily led to carve out that particular area to which his tools
apply; hence, to make a selection—and, obviously, not the same as that of the
biologist, for example, but that which is the proper selection of the historian.
Here we have an authentic problem of action. It will pursue us throughout our
study. 2. History and Men It is sometimes said: "History is the science of the
past." To me, this is badly put. For, to begin with, the very idea that the past as
such can be the object of science is ridiculous. How, without preliminary
distillation, can one make of phenomena, having no other common character
than that of being not contemporary with us, the matter of rational knowledge?
On the reverse side of the medal, can one imagine a complete science of the
universe in its present state? areas. Pasteur, who renovated biology, was not a
biologist—and during his lifetime he was often made to feel it; just as
Durkheim, and Vidal de la Blache, the first a philosopher turned sociologist, the
second a geographer, were neither of them ranked among the licensed
historians, yet they left an incomparably deeper mark upon historical studies at
the beginning of the twentieth century than any specialists.
History, Men, and Time S3 Doubtless, in the origins of historiography, the
old annalists were scarcely embarrassed by these scruples. They narrated pell-
mell events whose only connection was that they had happened about the same
time: eclipses, hailstorms, and the sudden appearance of astonishing meteors
along with battles and the deaths of kings and heroes. But into these early
reminiscences of humanity, as garbled as the observations of a small child, a
sustained effort of analysis has gradually introduced the necessary
classification. It is true that our language, fundamentally conservative, freely
retains the name of history for any study of a change taking place in time. The
custom is harmless, for it deceives no one. In that sense, there is a history of the
solar system, because the stars which compose it have not always been as we
now see them. It belongs to the province of astronomy. There is a history of
volcanic eruptions which is, I am sure, of most lively interest as regards the
composition of the earth. It does not concern the history of historians. Or, at
least, it does so only in so far as its observations chance to coincide with the
specific preoccupations of our history. How, then, is the division of labor
determined in practice? To understand this, a single example will be worth more
than a thousand words. In the tenth century a.d., a deep gulf, the Zwin, indented
the Flemish coast. It was later blocked up with sand. To what department of
knowledge does the study of this phenomenon belong? At first sight, anyone
24 The Historian's Craft would suggest geology. The action of alluvial
deposit, the operation of ocean currents, or, perhaps, changes in sea level: was
not geology invented and put on earth to deal with just such as these? Of course.
But at close range, the matter is not quite so simple. Is there not first a question
of investigating the origin of the transformation? Immediately, the geologist is
forced to ask questions which are no longer strictly within his juris* diction. For
there is no doubt that the silting of the gulf was at least assisted by dyke
construction, changing the direction of the channel, and drainage—all activities
of man, founded in collective needs and made possible only by a certain social
structure. At the other end of the chain there is a new problem: the
consequences. At a little distance from the end of the gulf, and communicating
with it by a short river passage, rose a town. This was Bruges. By the waters of
the Zwin it imported or exported the greatest part of the mer* chandise which
made of it, relatively speaking, the London or New York of that day. Then
came, every day more apparent, the advance of the sand. As the water receded,
Bruges vainly extended its docks and harbor further toward the mouth of the
river. Little by little, its quays fell asleep. To be sure, this was not the sole cause
of its decline. (Does the physical ever affect the social, unless its operations
have been prepared, abetted, and given scope by other factors which themselves
have already derived from man?) But this was certainly at least one of the most
efficacious of the links in the causal chain.
History, Men, and Time 25 Now, the act of a society remodeling the soil
upon which it lives in accordance with its needs is, as any one recognizes
instinctively, an eminently '^historical" event It is the same with the vicissitudes
of a powerful seat of trade. Hence, in an example entirely characteristic of the
topography of learning, we see, on the one hand, an area of overlap, where the
union of two disciplines is shown to be indispensable to any attempt at
explanation; on the other, a point of transition, where when a phenomenon has
been described with the sole exception that its consequences remain
undetermined, it is, in some definitive way, yielded up by one discipline to
another. What is it that seems to dictate the intervention of history? It is the
appearance of the human element Long ago, indeed, our great forebears, such as
Mi- chelet or Fustel de Coulanges, taught us to recognize that the object of
history is, by nature, man.2 Let us say rather, men. Far more than the singular,
favoring ab* straction, the plural which is the grammatical form of 2 Fustel de
Coulanges, opening lecture of 1862, in Revue de Syntkese historique, t. II,
1901, p. 243; Michelet, course at the Ecole Normale, 1829, cited by G. Monod,
La Vie et la Pem&e de Jules Michelet, t I, p. 127: "We are concerned at the
same time with the study of the individual man, and that will be philosophy —
and with the study of the social man, and that will be history/* It is proper to
add that, much later, Fustel remarked in a more concise and fuller formula, of
which the foregoing exposition is hardly more than a commentary: "History is
not the accumulation of events of every kind which happened in the past. It is
the science of human societies/' But this is, perhaps, to curtail the role of the
individual too much in history; man in society, and societies, are not precisely
equivalent ideas.
26 The Historian's Craft relativity is fitting for the science of change.
Behind the features of landscape, behind tools or machinery, behind what
appear to be the most formalized written documents, and behind institutions,
which seem almost entirely detached from their founders, there are men, and it
is men that history seeks to grasp.8 Failing that, it will be at best but an exercise
in erudition. The good historian is like the giant of the fairy tale. He knows that
wherever he catches the scent of human flesh, there his quarry lies. From the
character of history as the knowledge of men derives its peculiar situation as
regards the problem of expression. Is it "science" or "art"? About 1800, our
great-grandfathers delighted in solemn debates on this question. Later, about
1890, saturated with the aura of a rather primitive positivism, the methodolo-
gists were indignant that the public should attach an excessive importance to
what they called "form" in historical works. Art versus science, form versus
matter: the history of scholarship abounds with such fine debates! There is no
less beauty in a precise equation than in a felicitous phrase, but each science has
its appropriate aesthetics of language. Human actions are essentially very
delicate phenomena, many aspects of which elude mathemathical measurement.
Properly to translate them into words and, hence, to fathom * "Not man, again,
never man. Human societies, organized groups." Lucien Febvre, La Terre et
revolution humaine, p. 201.
History, Men, and Time 27 them rightly (for can one perfectly understand
what he does not know how to express?), great delicacy of language and precise
shadings of verbal tone are necessary. Where calculation is impossible we are
obliged to employ suggestion. Between the expression of physical and of
human realities there is as much difference as between the task of a drill
operator and that of a lutemaker: both work down to the last millimeter, but the
driller uses precision tools, while the lutemaker is guided primarily by his
sensitivity to sound and touch. It would be unwise either for the driller to adopt
the empirical methods of the lutemaker or for the lutemaker to imitate the
driller. Will anyone deny that one may not feel with words as well as with
fingers? 3. Historical Time We have called history "the science of men." That is
still far too vague. It is necessary to add: "of men in time." The historian does
not think of the human in the abstract. His thoughts breathe freely the air of the
climate of time. To be sure, it is difficult to imagine that any of the sciences
could treat time as a mere abstraction. Yet, for a great number of those who, for
their own purposes, chop it up into arbitrarily homogeneous segments, time is
nothing more than a measurement. In contrast, historical time is a concrete and
living reality with an irreversible onward rush. It is the very plasma in which
events are immersed, and the field within
28 The Historian's Craft which they become intelligible. The number of
seconds, years, or centuries required for a radioactive substance to change into
other substances is a fundamental datum for the atomic scientist. But the idea
any particular one of these metamorphoses had occurred a thousand years ago,
or yesterday, or today, or that another such is bound to occur tomorrow—all of
which would unquestionably interest the geologist, because geology is, in its
way, a historical discipline—leaves the physicist perfectly unmoved. In his turn,
no historian would be satisfied to state that Caesar devoted eight years to the
conquest of Gaul, or that it took fifteen years for Luther to change from the
orthodox novice of Erfurt into the reformer of Wittenberg. It is of far greater
importance to him to assign the conquest of Gaul its exact chronological place
amid the vicissitudes of European societies; and, without in the least denying
the eternal aspect of such spiritual crises as Brother Martin's, he will feel that he
has given a true picture of it only when he has plotted its precise moment upon
the life charts of both the man who was its hero and the civilization which was
its climate. Now, this real time is, in essence, a continuum. It is also perpetual
change. The great problems of historical inquiry derive from the antitheses of
these two attributes. There is one problem especially, which raises the very
raison d'etre of our studies. Let us assume two consecutive periods taken out of
the uninterrupted sequence of the ages. To what extent does the connection
which the flow of time sets between them pre-
History, Men, and Time 29 dominate, or fail to predominate, over the
differences born out of that same flow? Should the knowledge of the earlier
period be considered indispensable or superfluous for the understanding of the
later? 4. The Idol of Origins It will never be amiss to begin with an
acknowledgment of our faults. The explanation of the very recent in terms of
the remotest past, naturally attractive to men who have made of this past their
chief subject of research, has sometimes dominated our studies to the point of a
hypnosis. In its mos^characteristic aspect, this idol of the historian tribe may be
called the obsession with origins. Moreover, in the development of historical
thought, it has enjoyed its moment of particular favor. It was Renan, I believe,
who once wrote (I quote from memory, therefore, I fear, inexactly) : "In all
human affairs, it is the origins which deserve study before everything else."
And, before him, Sainte-Beuve: "With curiosity, I scrutinize and make note of
all beginnings." The idea is entirely typical of their age. So also is the word
"origins." Shortly after The Origins of Christianity came The Origins of Con*
temporary France. Not to mention mere followers* However, the word "origins"
is disturbing, because it is ambiguous. Does it mean simply "beginnings"? That
would be relatively clear—except that for most historical realities the very
notion of a starting-point remains singularly elusive. It is doubtless a matter of
definition, but
80 The Historian's Craft of a definition which it is unfortunately all too
easy to forget to give. On the other hand, is "origins" taken to mean the causes?
In that case, there will be no difficulties other than those which are always
inherent in the nature of causal inquiry (and even more so, no doubt, in the
sciences of man.) But there is a frequent cross-contamination of the two
meanings, the more formidable in that it is seldom very clearly recognized. In
popular usage, an origin is a beginning which explains. Worse still, a beginning
which is a complete explanation. There lies the ambiguity, and there the danger!
Some most interesting researches might be undertaken on that embryogenic
obsession which is so marked among exegetes. "I do not understand your
agitation," Barr6s confessed to a priest who had lost faith. "What have the
arguments of a handful of savants about a few Hebrew words to do with my
feeling? The atmosphere of a church is quite enough." And Maurras, in his turn:
"How do the Gospels of four obscure Jews concern me?" ("Obscure," means, I
imagine, plebeian; for, as regards Matthew, Mark, Luke, and John, it would be
hard to ignore, at least, a certain literary notoriety). These pranksters are pulling
our leg. Neither Pascal nor Bossuet would speak so boldly. Doubtless, a
religious experience apart from history is conceivable. For the pure Deist, it is
enough to have the inner light to believe in God. But not to
History, Men, and Time 31 believe in the God of the Christians. For
Christianity, as I have already pointed out, is essentially a historical religion: a
religion, that is, whose prime dogmas are based on events. Read over your
creed: "I believe in Jesus Christ . . . who was crucified under Pontius Pilate . , .
and who rose from the dead on the third day/' Here the beginnings of the faith
are also its foundations. Now, this preoccupation with origins, justifiable in a
certain type of religious analysis, has spread in a doubtlessly inevitable
contagion into other fields of research where its legitimacy is far more
debatable. Moreover, history oriented towards origins was put to the service of
value judgments. What else did Taine intend, in tracing the "origins" of the
France of his day, but a denunciation of the political ill consequences of what he
considered a false philosophy of man? And whether the subject was the
Germanic invasions or the Norman conquest of England, the past was so
assiduously used as an explanation of the present only in order that the present
might be the better justified or condemned. So in many cases the demon of
origins has been, perhaps, only the incarnation of that other satanic enemy of
true history: the mania for making judgments. But let us return to our Christian
studies. It is one thing for a troubled and self-searching conscience to determine
its attitude toward the Catholic religion by some such code as is daily laid down
in our churches;
32 The Historian's Craft it is quite another for the historian to explain
present- day Catholicism as an observed fact. A knowledge of their beginnings
is indispensable to understand, but insufficient to account for, the actual
religious phenomena. To simplify our problem, we must postpone the question
as to how far the creed, identical in name, is the same in substance. Even
assuming our religious tradition entirely unchanging, we must find reasons for
its preservation. Human reasons, that is, for the assumption of divine
intervention would be unscientific. In a word, the question is no longer whether
Jesus was first crucified and then resurrected, but how it came to pass that so
many fellow humans today believe in the Crucifixion and Resurrection. Now,
wherever fidelity to a belief is to be found, all evidences agree that it is but one
aspect of the general life of a group. It is like a knot in which are intertwined a
host of divergent characteristics of the structure and mentality of a society. In
short, a religious creed involves the whole problem of the human environment.
Great oaks from little acorns grow. But only if they meet favorable conditions
of soil and climate, conditions which are entirely beyond the scope of
embryology. Religious history has here been cited only by way of example. In
any study, seeking the origins of a human activity, there lurks the same danger
of confusing ancestry with explanation. It is very like the illusion of certain old
etymologists
History, Men, and Time S3 who thought they had said all when they set
down the oldest known meaning of a word opposite its present sense, having
shown, for example, that bureau originally meant a coarse woolen cloth, or that
timbre 4 meant a drum. As if the main problem were not to understand how and
why the transition had taken place. As if, above all, the meaning of any word
were influenced more by its own past than by the contemporary state of the
vocabulary which, in its turn, is determined by the social conditions of the
moment Bureaux in bureaux de minist&re means a bureaucracy. When I ask for
timbres at my post-office window, I am able to use that term only because of
recent technical changes, such as the organization of the postal service itself,
and the substitution of a little gummed picture for the stamping of a postmark,
which have revolutionized human communications. It is because the different
acceptations of the old word, particularized according to profession, are today
so^idely different that there is no risk of confusion between the timbre which I
glue on my envelope and the purity of timbre which the music salesman praises
in his instruments. We speak of the "origins of the feudal system." Where are
we to seek them? Some say: "In Rome/' Others: "In Germany." The cause of
their confusion is obvious. Whether Roman or Germanic, certain practices, such
as clientele relations, companionship in arms, and the use of land tenure as
payment for serv^ * A postage stamp.
84 The Historian's Craft ice, were carried on by later generations in Europe
during the ages we call "feudal." But such practices were modified a great deal.
There were two words— "benefice" (beneficium) among the Latin, and "fief"
among the German-speaking peoples—which these later generations persisted
in using, while gradually and without realizing it, conferring upon them quite a
new significance. For, to the great despair of historians, men fail to change their
vocabulary every time they change their customs. All this is very interesting,
but it does not tell us the causes of feudalism. The characteristic institutions of
European feudalism were no mere patchwork of surviving scraps. At one stage
in our history, they arose from a total social situation. M. Seignobos has
somewhere remarked: "I believe that the revolutionary thought of the eighteenth
century . . . proceeds from the English thought of the seventeenth." Does he
mean by this that the French publicists of the Enlightenment, having read or
been indirectly influenced by certain English works of the preceding century,
adopted their political principles from them? We might accept this thesis if we
suppose that our phUosophes contributed nothing original in the way of
intellectual substance or atmospheric perspective to the foreign formulas. But
even arbitrarily reduced to a matter of borrowing, the history of this intellectual
movement is far from being clear. For the problem is still to know why the
transference of ideas took place when it did—no sooner and no later. A
contagion supposes two things: microbe multiplication
History, Men, and Time 35 and, at the moment when the disease strikes, a
favorable breeding-ground. In a word, a historical phenomenon can nevei be
understood apart from its moment in time. This is true of every evolutionary
stage, our own and all others. As the old Arab proverb has it: "Men resemble
their times more than they do their fathers/* Disregard of this Oriental wisdom
has sometimes brought discredit to the study of the past 5. The Boundaries
between Past and Present Must we believe, because the past does not entirely
account for the present, that it is utterly useless for its interpretation? The
curious thing is that we should be able to ask the question today. Not so very
long ago, the answer was almost unanimously predetermined. "He who would
confine his thought to present time will not understand present reality." So
Michelet expressed it at the beginning of his Peuple—a. fine book, but infected
with the fever of the age in which it was written. And Leibniz before him
ranked among those benefits which attend the study of history "the origins of
things present which are to be found in things past; for a reality is never het- ter
understood than through its causes/*6 But since Leibniz, and since Michelet, a
great 5 Preface to Accessions Historic^ (1700), Opera, ed. Dutens, t IV 2, p. 53:
"Tria sunt qua expetimus in historia: primum, volup* totem noscendi res
singulares; deinde, utUia in primis vitas pr&cepta; ac denique origjines
pr&sentium a prasteritus repetitas, cum omnia optime ex causis noscantur."
86 The Historian9s Craft change has taken place. Successive technological
revolutions have immeasurably widened the psychological gap between
generations. With some reason, perhaps, the man of the age of electricity and of
the airplane feels himself far removed from his ancestors. With less wisdom, he
has been disposed to conclude that they have ceased to influence him. There is
also a modernist twist inherent in the engineering mind. Is a mastery of old
Volta's ideas about galvanism necessary to run or repair a dynamo? By what is
unquestionably a lame analogy, but one which readily imposes itself upon more
than one machine-dominated mentality, it is easy to think that an analysis of
their antecedents is just as useless for the understanding and solving of the great
human problems of the moment. Without fully recognizing it, the historians,
too, are caught in this modernist climate. Why then should they not feel that,
within their province, there has also been a shift in the line which separates the
new from the old? What, for example, of the system of stabilized currency and
the gold standard which, only yesterday, would have figured as the very norm
of up-to-dateness in every manual of political economy? To the modern
economist, do they belong to the present, or to a history already reeking with
mold? Behind these confused impressions, it is possible to discover a number of
more consistent ideas, whose simplicity, at least on the surface, has captivated
certain minds.
History, Men, and Time 37 One short period seems somehow set apart
from the vast sweep of time. Its beginning was relatively recent, and its end
overlaps our own day. Nothing in it— neither its outstanding social and political
characteristics, nor its physical equipment, nor its cultural tone —presents any
important contrasts with our own world. It appears, in a word, to assume a very
marked degree of "contemporaneousness" with us. And, from this, it derives the
virtue or defect of being distinct from the rest of the past. A high-school
teacher, who was very old when I was very young, once told us: "Since 1830,
there has been no more history. It is all politics." One would no longer say
"since 1830"—the July Days have grown old in their turn. Nor would one say:
"It is all politics." Rather, with a respectful air: "It is all sociology." Or, with
less respect: "It is all journalism." Nevertheless, there are many who would
gladly repeat that since 1914, or since 1940, there has been no more history. Yet
they would not agree very well in other respects as to the reasons for this
ostracism. Some, who consider that the most recent events are unsuitable for all
really objective research just because they are recent, wish only to spare Clio's
chastity from the profanation of present controversy. Such, I believe, was the
thought of my old teacher. This is to rate our self-control rather low. It also
quite overlooks that, once an emotional chord has been struck, the line between
present and past is no longer strictly regulated by a mathematically measurable
chronology. In
88 The Historian's Craft the Languedoc high school where I served my first
term as a teacher, my good headmaster issued a warning in a voice befitting a
captain of education. "Here, with the nineteenth century, there is little danger;
but when you touch on the religious wars, you must take great care! " In truth,
whoever lacks the strength, while seated at his desk, to rid his mind of the virus
of the present may readily permit its poison to infiltrate even a commentary on
the Iliad or the Ramayana. There are other savants who consider, quite to the
contrary and with reason, that contemporary society is perfectly susceptible of
scientific investigation. But they admit this only to reserve its study for
branches of learning quite distinct from that which has the past for its object.
They analyze, and they claim, for example, to understand the contemporary
economic system on the basis of observations limited to a few decades. In a
word, they consider the epoch in which we live as separated from its
predecessors by contrasts so clear as to be self-explanatory. Such is also the
instinctive attitude of a great many of the merely curious. The history of the
remoter periods attracts them only as an innocuous intellectual luxury. On one
hand, a small group of antiquarians taking a ghoulish delight in unwrapping the
winding-sheets of the dead gods; on the other, sociologists, economists, and
publicists, the only explorers of the living.
History, Men, and Time 89 6. Understanding the Present by the Past Under
close scrutiny the prerogative of self-intelligibility thus attributed to present
time is found to be based upon a set of strange postulates. In the first place, it
supposes that, within a generation or two, human affairs have undergone a
change which is not merely rapid, but total, so that no institution of long
standing, no traditional form of conduct, could have escaped the revolutions of
the laboratory and the factory. It overlooks the force of inertia peculiar to so
many social creations. Man spends his time devising techniques of which he
afterwards remains a more or less willing prisoner. What traveler in northern
France has not been struck by the strange pattern of the fields? For centuries,
changes in ownership have modified the original design; yet, even today, the
sight of these inordinately long and narrow strips, dividing the arable land into a
prodigious number of pieces, is something which baffles the scientific
agriculturalist The waste of effort which such a disposition entails and the
problems which it imposes upon the cultivators are undeniable. How are we to
account for it? Certain impatient publicists have replied: "By the Civil Code and
its inevitable effects. Change the laws on inheritance and the evil will be
removed." Had they known history better, or had they further questioned a
peasant mentality shaped by centuries of experience, they would
40 The Historian's Craft not have thought the cure so simple. Indeed, this
pattern dates back to origins so distant that no scholar has yet succeeded in
accounting for it satisfactorily. The settlers in the era of the dolmens have more
to do with it than the lawyers of the First Empire. Perpetuating itself, as it were,
of necessity, for want of correction, this ignorance of the past not only confuses
contemporary science, but confounds contemporary action. A society that could
be completely molded by its immediately preceding period would have to have
a structure so malleable as to be virtually invertebrate. It would also have to be
a society in which communication between generations was conducted, so to
speak, in "Indian file"—the children having contact with their ancestors only
through the mediation of their parents. Now, this is not true. It is not true even
when the communication is purely oral. Take our villages, for example.
Because working conditions keep the mother and father away almost all day,
the young children are brought up chiefly by their grandparents. Consequently,
with the molding of each new mind, there is a backward step, joining the most
malleable to the most inflexible mentality, while skipping that generation which
is the sponsor of change. There is small room for doubt that this is the source of
that traditionalism inherent in so many peasant societies. The instance is
particularly clear, but it is far from unique. Because the natural antagonism
between age groups
History, Men, and Time 41 is always intensified between neighboring
generations, more than one youth has learned at least as much from the aged as
from those in their prime. Still more strongly, between even widely scattered
generations, the written word vastly facilitates those transfers of thought which
supply the true continuity of a civilization. Take Luther, Calvin, Loyola,
certainly men from another time—from the sixteenth century, in fact. The first
duty of the historian who would understand and explain them will be to return
them to their milieu, where they are immersed in the mental climate of their
time and faced by problems of conscience rather different from our own. But
who would dare to say that the understanding of the Protestant or the Catholic
Reformation, several centuries removed, is not far more important for a proper
grasp of the world today than a great many other movements of thought or
feeling, which are certainly more recent, yet more ephemeral? In a word, the
fallacy is clear, and it is only necessary to formulate it in order to destroy it. It
represents the course of human evolution as a series of short, violent jerks, no
one of which exceeds the space of a few lifetimes. Observation proves, on the
contrary, that the mighty convulsions of that vast, continuing development are
perfectly capable of extending from the beginning of time to the present. What
would we think of a geophysicist who, satisfied with having computed their
remoteness to a fraction of an inch, would
42 The Historian's Craft then conclude that the influence of the moon upon
the earth is far greater than that of the sun? Neither in outer space, nor in time,
can the potency of a force be measured by the single dimension of distance.
Finally, what of those things past which seem to have lost all authority over the
present—faiths which have vanished without a trace, social forms which have
miscarried, techniques which have perished? Would anyone think that, even
among these, there is nothing useful for his understanding? That would be to
forget that there is no true understanding without a certain range of comparison;
provided, of course, that that comparison is based upon differing and, at the
same time, related realities. One could scarcely deny that such is here the case.
Certainly, we no longer consider today, as Machia- velli wrote, and as Hume or
Bonald thought, that there is, in time, "at least something which is changeless:
that is man." We have learned that man, too, has changed a great deal in his
mind and, no less certainly, in the most delicate organs of his body. How should
it be otherwise? His mental climate has been greatly altered; and to no less an
extent, so, too, have his hygiene and his diet However, there must be a
permanent foundation in human nature and in human society, or the very names
of man or society become meaningless. How, then, are we to believe that we
understand these men, if we study them only in their reactions to circumstances
peculiar to a moment? It would be an inadequate test of them, even for that
History, Men, and Time 48 particular moment A great many potentialities,
which might at any instant emerge from concealment, a great many more or less
unconscious drives behind individual or collective attitudes, would remain in
the shadows. In a unique case the specific elements cannot be differentiated;
hence an interpretation cannot be made. 7. Understanding the Past by the
Present This solidarity of the ages is so effective that the lines of connection
work both ways. Misunderstanding of the present is the inevitable consequence
of ignorance of the past. But a man may wear himself out just as fruitlessly in
seeking to understand the past, if he is totally ignorant of the present. There is
an anecdote which I have already recounted elsewhere: I had gone with Henri
Pirenne to Stockholm; we had scarcely arrived, when he said to me: "What shall
we go to see first? It seems that there is a new city hall here. Let*s start there/'
Then, as if to ward off my surprise, he added: "If I were an antiquarian, I would
have eyes only for old stuff, but I am a historian. Therefore, I love life." This
faculty of understanding the living is, in very truth, the master quality of the
historian. Despite their occasional frigidity of style, the greatest of our number
have all possessed it. Fustel or Maitland, in their austere way, had it as much as
Michelet And, perhaps, it originates as a gift from the fairies, quite inaccessible
to anyone who has not found it in his cradle. That does not lessen the obligation
to
44 The Historian's Craft exercise and develop it constantly. How? How
better than by the example of Henri Pirenne—by keeping in constant touch with
the present day? For here, in the present, is immediately perceptible that
vibrance of human life which only a great effort of the imagination can restore
to the old texts. I have many times read, and I have often narrated, accounts of
wars and battles. Did I truly know, in the full sense of that word, did I know
from within, before I myself had suffered the terrible, sickening reality, what it
meant for an army to be encircled, what it meant for a people to meet defeat?
Before I myself had breathed the joy of victory in the summer and autumn of
191 & (and, although, alas! its perfume will not again be quite the same, I yearn
to fill my lungs with it a second time) did I truly know all that was inherent in
that beautiful word? In the last analysis, whether consciously or no, it is always
by borrowing from our daily experiences and by shading them, where
necessary, with new tints that we derive the elements which help us to restore
the past. The very names we use to describe ancient ideas or vanished forms of
social organization would be quite meaningless if we had not known living
men. The value of these merely instinctive impressions will be increased a
hundredfold if they are replaced by a ready and critical observation. A great
mathematician would not, I suppose, be less great because blind to the world in
which he liveSv But the scholar who has no inclination to observe the men, the
things, or the events around him will per*
History, Men, and Time 45 haps deserve the title, as Pirenne put it, of a
useful antiquarian. He would be wise to renounce all claims to that of a
historian. Moreover, the cultivation of historical sensitivity is not always all that
is involved, It may happen, in a given line, that the knowledge of the present
bears even more immediately upon the understanding of the past. It would be a
grievous error, indeed^ to think that the order which historians adopt for their
inquiries must necessarily correspond to the sequence of events, Even though
they restore its true direction afterwards, they have often benefited at the outset
by reading history, as Maitland said, "backwards." For the natural progression
of all research is from the best (or least badly) understood to the most obscure.
Certainly, it is far from true that the light of documentation grows ever brighter
as we pass down the corridor of the ages. For example, we are much less well-
informed on the tenth century of our era than on the epoch of Caesar or
Augustus. In the majority of cases however, the nearest periods correspond
better with the zones of relative clarity. We must add that, in proceeding
mechanically from early to late, there is always the risk of wasting time in
tracking down the beginning or causes of phenomena which, in the event, may
turn out to be somewhat imaginary. The most illustrious among us have
occasionally made strange mistakes through having neglected to pursue a
prudently retrogressive
46 The Historian's Craft method whenever and wherever it was indicated.
Fustel de Coulanges devoted himself to the "origins" of feudal institutions of
which he had formed, I fear, only a rather confused picture, and to the
beginnings of a serfdom which, misled by secondhand descriptions, he
conceived in entirely false colors. Now, more often than is generally supposed,
it happens that in order to find daylight, tibe historian may have to pursue his
subject right up to the present. In certain of its fundamental features, our rural
landscape, as has been previously mentioned, dates from a very remote epoch.
However, in order to interpret the rare documents which permit us to fathom its
misty beginnings, in order to ask the right questions, even in order to know
what we were talking about, it was necessary to fulfill a primary condition: that
of observing and analyzing our present landscape. For it alone furnished those
comprehensive vistas without which it was impossible to begin. Not, indeed,
that there could be any question of imposing this forever- static picture, just as it
is, at each stage of the journey upstream to the headwaters of the past. Here, as
elsewhere, it is change which the historian is seeking to grasp. But in the film
which he is examining, only the last picture remains quite clear. In order to
reconstruct the faded features of the others, it behooves him first to unwind the
spool in the opposite direction from that in which the pictures were taken.
History, Men, and Time 47 There is, then, just one science of men in time.
It requires us to join the study of the dead and of the living. What shall we call
it? I have already explained why the ancient name, "history," seemed to me the
best. It is the most comprehensive, the least exclusive, the most electric with
stirring reminders of a more than age-old endeavor. In proposing to extend
history right down to the present (contrary to certain prejudices which are not so
old as history itself), I have no desire to expand the claims of my own
profession. Life is too short, and science too vast, to permit even the greatest
genius a total experience of humanity. Some men will always specialize in the
present, as others do in the Stone Age or in Egyptology. We simply ask both to
bear in mind that historical research will tolerate no autarchy. Isolated, each will
understand only by halves, even within his own field of study; for the only true
history, which can advance only through mutual aid, is universal history. A
science, however, is not to be defined entirely in terms of its object. Its limits
can be fixed quite as well by the character of its appropriate methods. It remains
to be seen whether the techniques of inquiry should be considered
fundamentally different according as the investigation approaches or recedes
from the present. This, in its turn, suggests the entire problem of historical
observation.
[48] CHAPTER II HISTORICAL OBSERVATION J. General
Characteristics of Historical Observation To begin, let us place ourselves firmly
in the study of the past. The most obvious characteristics of historical
information, in the ordinary and restricted sense of the term, have been
frequently described. We are told that the historian is, by definition, absolutely
incapable of observing the facts which he examines. No Egyptologist has ever
seen Ramses. No expert on the Napoleonic Wars has ever heard the sound of
the cannon at Austerlitz. We can speak of earlier ages only through the accounts
of eye-witnesses. According to this view, we are in the predicament of a police
magistrate who strives to reconstruct a crime he has not seen; of a physicist
who, confined to his bed with grippe, hears the results of his experiments only
through the reports of his laboratory technician. In short, in contrast to the
knowledge of the present, that of the past is necessarily "indirect." No one
would dream of denying the element of truth in these remarks. Nevertheless,
they demand considerable modification.
Historical Observation 49 Let us suppose that a military commander has
just won a victory. That, immediately, he sets to work writing an account in his
own hand. That it was he who conceived the plan of the battle, and that it was
he who directed it. And finally that, thanks to the moderate size of the field (for
in order to sharpen the argument, we are imagining a battle of former times5
drawn up in a confined space), he has been able to see almost the entire conflict
develop before his eyes. Nevertheless, we cannot doubt that, in more than one
essential episode, he will be forced to refer to the reports of his lieutenants. In
acting thus as narrator, he would only be behaving as he had a few hours before
in the action. Then as commander, regulating the movements of his troops to the
swaying tide of battle^ what sort of information shall we think to have served
him best? Was it the rather confused scenes viewed through his binoculars, or
the reports brought in hot haste by the couriers and aides-de-camp? Seldom can
a leader of troops be his own observer. Meanwhile* even in so favorable a
hypothesis as this, what has become of that marvel of "direct" observation
which is claimed as the prerogative of the studies of the present? In truth, it is
scarcely ever anything but a delusion^ at least as soon as the observer has
expanded his horizon only slightly. A good half of all we see is seen through the
eyes of others. As an economist, I follow trading developments for this month
or for this week; f do so by means of statistics which I have not per-
50 The Historian's Craft sonally compiled. As a student of the present
instant, I apply myself to the task of sounding public opinion on the important
issues of the day. I ask questions. I note, compare, and compute the answers.
What do I then have but the rather awkwardly expressed ideas which my
communicants have formulated as to what they believe they believe, or what
they are willing to reveal. These are the subjects of my experiments, but,
whereas the physiologist who dissects a guinea pig sees with his own eyes the
lesion or abnormality which has been the object of his search, I know the mood
of my "man in the street" only through the chart of it which he himself agrees to
draw for me. Because the individual, narrowly restricted by his senses and
power of concentration, never perceives more than a tiny patch of the vast
tapestry of events, deeds, and words which form the destinies of a group, and
because, moreover, he possesses an immediate awareness of only his own
mental state, all knowledge of mankind, to whatever time it applies, will always
derive a large part of its evidence from others. In this respect, the student of the
present is scarcely any better off than the historian of the past. But there is
more. Is it certain that the observation of the past, even of the very remote past,
is always "indirect"? It is easy to see why this remoteness of the scholar from
the object of his knowledge makes so strong an impression upon many
historical theorists. It is be-
Historical Observation 51 cause they think of history primarily in terms of
events, even of episodes—of a history which, rightly or wrongly (and it is
immaterial at the moment) attaches an extreme importance to the exact
reconstruction of the actions, words, or attitudes of a few personages, brought
together for a relatively brief scene, in which, as in a classic tragedy, are
marshaled all the forces of the critical moment: the day of a revolution, a battle,
or a diplomatic interview. It is related that on September 2,1792, the head of the
Princess de Lamballe was paraded on the end of a pike under the windows of
the royal family. Is this true or false? M. Pierre Caron? who has written an
admirably honest book on the September Massacres, does not venture an
opinion. Had he been permitted to watch the ghastly cortege in person from a
tower in the Temple, he would have known what to think—at least if,
preserving his scholarly detachment in these circumstances (as might be
expected), and properly mistrustful of his own memory, he had further taken the
precaution of making a note of his observations on the spot. Unquestionably, in
such cases, the historian is mortified by comparing his position with that of a
reliable witness of a present event. He is as if at the rear of a column, in which
the news travels from the head back through the ranks. It is not a good vantage-
point from which to gather correct information. Not so very long ago, during a
relief march at night, I saw the word passed down the length of a column in this
manner: "Look out! Shell holes to the left!" The last man received it in the
52 The Historians Crap form, "To the left!" took a step in that direction,
and fell in. There are other contingencies, however. Archaeologists have
restored to the light of the present day a number of pottery urns, filled with the
bones of children, which had been securely sealed up in the walls of certain
Syrian fortresses built several thousand years before the birth of Christ. Since
we cannot reasonably suppose that these bones have strayed there by accident,
according to all the evidence, we are confronted with the remains of human
sacrifices, performed at the moment of original construction and somehow
connected with it. As to the beliefs which were expressed in these rites, we must
refer to contemporary testimony, or, if none exists, proceed by analogy with the
aid of other evidence. For how are we to understand a faith we do not share
except through the accounts of others? Such—and it bears repeating—is the
case with all phenomena of consciousness in so far as they are alien to us. As
for the mere fact of sacrifice, on the other hand, our situation is very different.
To be sure, we no more grasp the fact immediately, properly speaking, than the
geologist captures the living ammonite when he discovers its fossil, or than a
physicist sees the actual molecular motion when he studies its effects in a
suspension, as in Brownian movement. But the simple reasoning which, by
eliminating all other possible explanations, permits us to pass from the
authenticated object to the fact of which it is the proof, this act of rudimentary
interpretation—border-
Historical Observation 53 ing, indeed, upon those instinctive mental
reactions without which no sensation would become a perception—in no way
requires the intervention of another observer between the object and ourselves.
By "indirect knowledge" the methodologists have generally understood that
which arrives at the mind of the historian only by way of other human minds.
The phrase is not, perhaps, very carefully chosen; it confines itself to pointing
out the presence of an intermediary, without making it clear why this
intermediary should necessarily be human. But let us accept the common usage
without further quibbling. In that sense, there is certainly nothing indirect about
our knowledge of these immured sacrifices in ancient Syria. Now, a great many
other vestiges of the past are equally accessible. Such is the case not only with
almost all the vast bulk of the unwritten evidence, but also with a good part of
that which is written. If the best-known theorists of our methods had not shown
such an astonishing and arrogant indifference toward the techniques of
archaeology, if they had not been as obsessed with narrative in the category of
documents as they were with incident in the category of actions* they would
doubtless have been less ready to throw us back upon an eternally dependent
method of observation. In the royal tombs of Ur in Chaldea, there have been
found beads of necklaces made of amazon- ite. As the nearest deposits of this
mineral are located either in the heart of India or in the neighborhood of Lake
Baikal, it has seemed obvious to conclude that;
51 The Historian's Craft as far back as the third millennium before Christ,
the cities of the lower Euphrates maintained trading relations with some very
distant lands. The inference may be either true or false. However that may be, it
is undeniable that it is an induction of the most classic type; it is founded upon
the observation of a fact and the word of another person has absolutely nothing
to do with it. But physical objects are far from being the only ones which can be
thus readily apprehended at firsthand. A linguistic characteristic, a point of law
embodied in a text, a rite, as defined by a book of ceremonial or represented on
a stele, are realities just as much as the flint, hewn of yore by the artisan of the
stone age—realities which we ourselves apprehend and elaborate by a strictly
personal effort of the intelligence. There is no need to appeal to any other
human mind as an interpreter. To revert to our analogy of a moment ago, it is
not true that the historian can see what goes on in his laboratory only through
the eyes of another person. To be sure, he never arrives until after the
experiment has been concluded. But, under favorable circumstances, the
experiment leaves behind certain residues which he can see with his own eyes.
It is therefore advisable to define the indisputable peculiarities of historical
observation in terms which are both less ambiguous and more comprehensive.
Its primary characteristic is the fact that knowledge of all human activities in the
past, as well as of the
Historical Observation 55 greater part of those in the present, is, as
Frangois Simiand aptly phrased it, a knowledge of their tracks. Whether it is the
bones immured in the Syrian fortifications, a word whose form or use reveals a
custom, a narrative written by the witness of some scene, ancient or modern,
what do we really mean by document, if it is not a "track," as it were—the
mark, perceptible to the senses, which some phenomenon, in itself inaccessible,
has left behind? It matters little whether the original object is by its very nature
inaccessible to the senses, like an atom whose trajectory is rendered visible in a
Crookes tube, or whether through the effect of time it has only become so in the
present, like the fern, rotting for thousands of years, whose imprint is left upon a
lump of coal, or like those long-abandoned ceremonials which are painted and
explained upon the walls of Egyptian temples. In either case, the process of
reconstruction is the same, and every science offers a variety of examples of it.
However, the fact that many explorers in every field are able to understand
certain central phenomena only by means of other phenomena derived from
them in this manner by no means signifies that they all share a perfect equality
of methods. Like the physicists, they may themselves be able to produce the
appearance of these "tracks." On the other hand, they may be compelled to wait
upon the caprice of forces over which they have no influence whatsoever.
Depending on
56 The Historian's Craft these alternatives, their positions will vary widely.
What is the situation of the observers of human activity? Here the question of
chronology again arises. The fact that all moderately complex human acts elude
the possibility either of reproduction or of deliberate initiation seems to go
without saying, and, in any case, we shall return to the point later. To be sure,
there are psychological tests to measure the most elementary sensations as well
as the most delicate nuances of intelligence or emotivity. But these are
applicable only to the individual. They are almost entirely unsuited to group
psychology. One can not—even if one could, one would dare not—deliberately
produce a panic or a movement of religious fervor. However, if the phenomena
under study belong to the present or to the very recent past, the observer—
however impotent he may be to force their repetition or to shape their
development to his liking—is not equally helpless as regards their "tracks/'
Certain of these he is able literally to call into existence. These are the reports of
witnesses. It was as impossible to relive the experience of Austerlitz on
December 5, 1805, as it is today. However, suppose that the question was what
such and such a regiment had done in the battle? Had Napoleon wished to
inform himself on this point, several hours after the cease-fire, he had but to say
the word to have an officer give him an account. If no such report, public or
private, was ever made, or if those which
Historical Observation 57 were written were lost, then it would be vain for
us to ask Napoleon's question; it stands a good chance of never being answered,
along with many others far more important. What historian has not had
daydreams of being able, like Ulysses, to body forth the shades for questioning?
But it is no longer the season of the miracles of the Nekuia, and we have no
other device for returning through time except that which operates in our minds
with the materials provided by past generations. The advantages of the present
must not be exaggerated. Let us imagine that all the officers and men of the
hypothetical regiment at Austerlitz have perished, or, more simply, that among
the survivors there are no longer to be found witnesses whose memory and
powers of attention are trustworthy. Napoleon would then be no better off than
we are. Anyone who has taken even the humblest part in a great battle is very
well aware that it sometimes becomes impossible to be precise about a major
episode after only a few hours. We must add that not all "tracks" lend
themselves equally well to this evocation of the past for the edification of the
future. If the customs had neglected to register the imports and exports of
merchandise every day in November 1942, I would have in December
practically no means of determining the foreign commerce of a month before. In
a word, the difference between the study of the remote and that of the recent
past is, as previously stated, only one of degree. It does not extend to the
fundamental prob-
58 The Historians Craft lem of method. Still, the difference is important,
and it is only proper to examine the consequences. The past is, by definition, a
datum which nothing in the future will change. But the knowledge of the past is
something progressive which is constantly transforming and perfecting itself.
Anyone who doubts this need only recall how much it has improved under our
very eyes in little more than a century. Vast areas of mankind have emerged
from the shadows. Egypt and Chaldea have shaken off their shrouds. The lost
cities of central Asia have disclosed their now-unspoken languages and long-
extinct religions. A civilization, all unsuspected, has but lately risen from its
grave upon the banks of the Indus. That is not all, and the ingenuity of the
scholars in further ransacking the libraries or in opening new excavations on
ancient sites is neither the sole nor, perhaps, even the most effective means of
enriching our picture of the past. Hitherto- unknown techniques of investigation
have also come to light. We are more skillful than our predecessors in
examining languages for the evidence of customs and tools for the evidence of
techniques. Above all, we have learned how to probe more deeply in the
analysis of social developments. The study of popular rites and beliefs is barely
sketching its first outlines. Economic history, which, not so long ago, Cournot
did not even think to include in his listing of the various aspects of historical
research, is only beginning to establish itself.
Historical Observation 59 AH this is certain. All of it offers us the most
extensive hopes, but they are not unlimited hopes. This sense of virtually
unlimited progress, granted to a science like chemistry, which is capable of
creating even its own subject matter, is refused to us. Explorers of the past are
never quite free. The past is their tyrant It forbids them to know anything which
it has not itself, consciously or otherwise, yielded to them. We shall never
establish a statistical table of prices for the Merovingian epoch, for there are no
documents which record these prices in sufficient number. We shall never be
able to get inside the minds of the men of eleventh-century Europe, for
example, as well as we can those of the contemporaries of Pascal or Voltaire,
because, in place of their private letters or confessions, ive have only a few bad
biographies, written in a conventional style. Owing to this gap, one entire
segment of our history necessarily assumes the rather anemic aspect of a world
without individuals. But we must not grumble too much. We poor adepts of the
young sciences of man are often laughed at, but, in our strict submission to an
inflexible fate, we are no worse off than many of our confreres in the older and
safer disciplines. Such is the common lot of all studies calling for the
examination of past phenomena. The prehistorian who lacks written records is
no more incapable of reconstructing the rituals of the stone age than is the
paleontologist (I suppose) of reconstructing the glands of internal secretion of
the plesiosaurus whose skeleton alone still remains. It is always disagreeable to
60 The Historian's Craft say: "I do not know. I cannot know." It must not
be said except after an energetic, even a desperate search. But there are times
when the sternest duty of the savant, who has first tried every means, is to
resign himself to his ignorance and to admit it honestly. 2. Evidence "Herodotus
of Thurii here sets down his inquiries toward the end that the things done by
men should not be forgotten with the passage of time and that the great and
marvelous exploits, performed by both Greeks and barbarians, should not lose
their radiance." Thus begins what, with the exception of a few fragments, is the
oldest book of history which has come down to us in the Western world. For the
sake of illustration, let us compare it with one of the guides for the trip to the
beyond which the Egyptians put into their tombs in the time of the Pharaohs.
We shall then have face to face the archetypes of the two chief categ^'es into
which the innumerable varieties of documents at the dicnosal of the historian
are divided. The evidence of the first group is intentional; that of the second is
not. Indeed, when in seeking information we read Herodotus, Froissart, the
memoirs of Marshal Joffre, or the completely contradictory accounts, reported
these days in the British or German newspapers, of an attack on a
Mediterranean convoy, we are only doing exactly what the writers expected us
to do. On the other hand, the formulae of the Book of the Dead were destined
Historical Observation 61 only to be recited by the soul in peril and heard
by gods alone. The lake-dweller who threw his garbage into the near-by water,
where the archaeologists retrieve it today, wanted only to keep his hut clean.
The bull of papal immunity was so carefully preserved in the strongbox of the
monastery only in order, when the time arrived, that it might be brandished
under the nose of a meddling bishop. In none of these precautions was there the
least desire to influence the opinions either of contemporaries or of future
historians; and when the medievalist, in this year of our Lord 1942, leafs
through the archives of the commercial correspondence of the Cedame family
of Lucca, he is guilty of an indiscretion which would be loudly decried by the
financiers of our own day, if he took the same liberty with their files. Now, the
narrative sources—to use a rather baroque but hallowed phrase—that is, the
accounts which are consciously intended to inform their readers, still continue
to provide valuable assistance to the scholar. Among their other advantages,
they are ordinarily the only ones which furnish a chronological framework,
however inconsistent. What would not the prehis- torian or the historian of India
give to have a Herodotus at his disposal? Nevertheless, there can be no doubt
that, in the course of its development, historical research has gradually been led
to place more and more confidence in the second category of evidence, in the
evidence of witnesses in spite of themselves. We have only to compare the
Roman history of Rollin or
62 The Historian's Craft even that of Niebuhr with any of those short
summaries we read today. The former draw the heart of their matter from Livy,
Suetonius, or Floras. The latter are constructed in large measure out of
inscriptions, papyri, and coins. Only in this way could whole sections of the
past have been reconstructed. This is true of all prehistory, as well as of almost
all economic history and almost all history of social structures. Even in the
present, who among us would not prefer to get hold of a few secret chancellery
papers or some confidential military reports, to having all the newspapers of
1938 or 1939? It is not that this sort of document is any less subject to errors or
falsehoods than the others. There are plenty of fraudulent bulls, and neither all
ambassadorial accounts nor all business letters tell the truth. But this kind of
distortion, if it exists, at least, has not been especially designed to deceive
posterity. Moreover, these tracks which the past unwittingly leaves all along its
trail do more than simply permit us to fill in the narrative where it is missing
and to check it where its truthfulness is suspected. They protect our studies from
a peril more deadly then either ignorance or inaccuracy: that of an incurable
sclerosis. Indeed, without their aid, every time the historian turned his attention
to the generations gone by, he would become the inevitable prey of the same
prejudices, false inhibitions, and myopias which had plagued the vision of those
same generations. For example, the medieval-
Historical Observation m ists would accord but a trivial significance to
communal development, under the pretext that the writers of the Middle Ages
did not discuss it freely with their public, or would disregard the mighty force of
religious life for the good reason that it occupied a much less important place in
contemporary narrative literature than the wars of the barons. In a word, to
resort to a favorite figure of Michelet's, history would become less the ever-
daring explorer of the ages past than the eternally unmoving pupil of their
"chronicles." Moreover, even when most anxious to bear witness, that which the
text tells us expressly has ceased to be the primary object of our attention today.
Ordinarily, we prick up our ears far more eagerly when we are permitted to
overhear what was never intended to be said. What do we find most instructive
in the works of Saint-Simon? Is it their frequently fictitious news of the events
of the reign, or the remarkable light which the Memoirs throw upon the
mentality of a great noble at the court of the Sun King? At least three fourths of
the lives of the saints of the high Middle Ages ran teach us nothing concrete
about those pious personages whose careers they pretend to describe. If, on the
other hand, we consult them as to the way of life or thought peculiar to the
epoch in which they were written (all things which the biographer of the saint
had not the least intention of revealing), we shall find them invaluable. Despite
our inevitable subordination to the past, we have freed ourselves at least to the
extent that, eternally condemned to know
64 The Historian's Craft only by means of its "tracks/' we are nevertheless
successful in knowing far more of the past than the past itself had thought good
to tell us. Properly speaking, it is a glorious victory of mind over its material.
But from the moment when we are no longer resigned to purely and simply
recording the words of our witnesses, from the moment we decide to force them
to speak, even against their will, cross-examination becomes more necessary
than ever. Indeed, it is the prime necessity of well-conducted historical research.
Many people and, it appears, even some authors of manuals entertain an
extraordinarily simplified notion of our working procedure. First, as they are
only too eager to tell you, there are the documents. The historian collects them,
reads them, attempts to weigh their authenticity and truthfulness. Then, and
only then, he makes use of them. There is only one trouble with this idea: no
historian has ever worked in such a way, even when, by some caprice, he
fancied that he was doing so. For even those texts or archaeological documents
which seem the clearest and the most accommodating will speak only when
they are properly questioned. Before Boucher de Perthes, as in our own day,
there was plenty of flint artifacts in the alluvium of the Somme. However, there
was no one to ask questions, and there was therefore no prehistory. As an old
medie-
Historical Observation 05 valist, I know nothing which is better reading
than a cartulary. That is because I know just about what to ask it. A collection
of Roman inscriptions, on the other hand, would tell me little. I know more or
less how to read them, but not how to cross-question them* In other words,
every historical research supposes that the inquiry has a direction at the very
first step. In the beginning, there must be the guiding spirit. Mere passive
observation, even supposing such a thing were possible, has never contributed
anything productive to any science. Indeed, we must here make no mistake. It
may well be that the cross-examination remains purely instinctive. It is there,
nevertheless. Without the scholar's being aware of it, its dictates are etched into
his brain by the convictions and inhibitions of his former experiences, by means
of tradition and by means of common sense, which is too often to say by means
of vulgar prejudices. We are never quite so receptive as we should like to
believe. There is no worse advice for a beginner than that he should simply sit
patiently waiting for the inspiration of a document Such conduct has betrayed
more than one well-intended inquiry into either stalemate or checkmate.
Naturally, the method of cross-examination must be very elastic, so that it may
change its direction or improvise freely for any contingency, yet be able, from
the outset, to act as a magnet drawing findings out of the document. Even when
he has settled his itiner-
66 The Historian's Craft ary, the explorer is well aware that he will not
follow it exactly. Without it, however, he would risk wandering perpetually at
random. The variety of historical evidence is nearly infinite. Everything that
man says or writes, everything that he makes, everything he touches can and
ought to teach us about him. It is curious to note how many people,
unacquainted with our work, underestimate the true extent of its possibilities. It
is because they persist in an idea of our science which dates back to the time
when we scarcely knew how to read even the intentional evidence. In
reproaching "traditional history," Paul Val6ry has cited "the conquest of the
earth" by electricity, as an example of one of those "notable phenomena" which
it neglects, despite the fact that they have "more meaning and greater
possibilities of shaping our immediate future than all the political events
combined." For this, he deserves our heartiest applause. It is unfortunate, but all
too true that this vast subject has still received no serious treatment However,
apparently led astray by an excess of severity to excuse the very fault which he
has just condemned, Val&y adds that this phenomenon must of necessity elude
the historian because, he argues, there are no documents which refer to it
specifically. This time, shifting from the scholar to the science, he lodges his
complaint at the wrong door. Who believes that the electrical companies have
no archives, no records of consumption, no charts of the enlargement of their
Historical Observation 67 networks? The truth is that the historians until
now have simply neglected to question these documents. Certainly, they are
very much to blame, unless the fault lies with the custodians of the archives,
possibly too jealous of their precious treasures. Have patience. History is not yet
what it ought to be. That is no reason to make history as it can be the scapegoat
for the sins which belong to bad history alone. Marvelous as is the diversity of
our materials, it nevertheless creates a difficulty so serious as to rank among the
three or four outstanding paradoxes of the historical profession. It would be
sheer fantasy to imagine that for each historical problem there is a unique type
of document with a specific sort of use. On the contrary, the deeper the
research, the more the light of the evidence must converge from sources of
many different kinds. What religious historian would be satisfied by examining
a few theological tracts or hymnals? He knows full well that the painting and
sculpture of sanctuary walls and the arrangement and furnishings of tombs have
at least as much to tell him about dead beliefs and feelings as a thousand
contemporary manuscripts. Our knowledge of the Germanic invasions has
derived as much from the archaeology of tombs and place-names as it has from
the examination of charters and chronicles. As we approach our own times, the
requirements change without becoming less exacting. To understand modern
society, is it enough merely to plunge into reading parliamentary debates or
cabinet papers? Is it not also
68 The Historian's Craft necessary to be able to interpret a financial
statement, as unintelligible to the layman as so many hieroglyphics? In an age
when the machine is supreme, should a historian be allowed to ignore how
machines are designed and modified? Now, if almost any important human
problem thus demands the handling of diverse types of evidence, on the other
hand the types of evidence necessarily mark off the several branches of
technical scholarship. The apprenticeship of each is long, but full mastery
demands a still longer and almost constant practice. For example, very few
scholars can boast that they are equally well equipped to read critically a
medieval charter, to explain correctly the etymology of place- names, to date
unerringly the ruins of dwellings of the prehistoric, Celtic, or Gallo-Roman
periods, and to analyze the plant life proper to a pasture, a field, or a moor.
Without all these, however, how could one pretend to describe the history of
land use? Few sciences, I believe, are forced to use so many dissimilar tools at
the same time. However, man's actions are the most complex in the animal
kingdom, because man stands upon nature's summit. It is advisable and, in my
opinion, it is indispensable that the historian possess at least a smattering of all
the principal techniques of his trade, if only to learn the strength of his tools and
the difficulties of handling them. The list of "auxiliary disciplines" which we
expect our beginners to learn is much too short. What an absurd illogicality that
men who half the time can
Historical Observation m have access to their subject only through words,
are permitted, among other deficiencies, to be ignorant of the fundamental
attainments of linguistics! But whatever the variety of accomplishments we may
wish to ascribe to our best-equipped scholars, they will inevitably and,
ordinarily very quickly, discover their own limitations. We have no other
remedy than to substitute, in place of the multiple skills of a single man, the
pooling of the techniques, practiced by different scholars, but all tending to
throw light upon a specific subject. This method presupposes a spirit of
teamwork. It also demands the preliminary definition by common consent of the
several dominant problems. We are still all too distant from these goals.
Nevertheless, in large measure, they will unquestionably govern the future of
our science. 3 The Transmission of Evidence One of the most difficult tasks of
the historian is that of assembling those documents which he considers
necessary. He could hardly succeed without the help of various guides: archival
or library catalogues, museum indexes, and bibliographies of every kind. There
are people who express contemptuous amazement at the time sacrificed by
some scholars in composing such works and by all the rest in familiarizing
themselves with their existence and use. They behave as if the most frightful
waste of energy were not ultimately avoided, thanks to the hours thus devoted
to labors which, if not without a certain hidden charm,
70 The Historian's Craft are certainly lacking in romantic glamor. Let us
suppose that I have become duly interested in the history of the cult of the
saints, but that I am ignorant of the Bibliotheca Hagiographica Latina of the
BoUandist Fathers. It would be difficult for anyone who is not an expert to
imagine the amount of stupidly useless effort which this gap in my mental
equipment would inevitably cost me. What is truly regrettable is not that we
must stock our libraries with a considerable quantity of those tools, whose very
enumeration, subject by sub- )ect, belongs to special books of orientation. It is
rather that there are still not enough of them, especially for the most recent
periods; that their composition, particularly in France, has conformed only by
exception to a rational and comprehensive plan; and, finally, that the task of
keeping them up to date has been too often abandoned either to the caprice of
individuals or to the ill-advised parsimony of a few publishing houses. The first
volume of Emile Molinier's admirable Sources de tHistoire de France has not
been revised since its first appearance in 1901. That simple fact is in itself a
severe indictment. Granted that instruments do not create science, nevertheless
a society which pretends to respect the sciences ought not to neglect their
instruments. Nor would it be wise to rely entirely upon academic bodies for
these instruments, for their methods of recruiting, favoring seniority and
orthodox scholarship, do not particularly incline them to a spirit of enterprise.
Our War College and General Staff are not the only institutions in France which
have
Historical Observation 71 preserved the mentality of the oxcart in the age
of the automobile. Such guides, however well made, however abundant, would
be of little aid to the worker who had no preliminary idea of the terrain to be
explored. Despite what the beginners sometimes seem to imagine, documents
do not suddenly materialize, in one place or another, as if by some mysterious
decree of the gods. Their presence or absence in the depths of this archive or
that library are due to human causes which by no means elude analysis. The
problems posed by their transmission, far from having importance only for the
technical experts, are most intimately connected with the life of the past, for
what is here at stake is nothing less than the passing down of memory from one
generation to another. In historical works of a serious nature, the author
generally lists the files of archives he has examined and the printed collections
he has used. That is all very well, but it is not enough. Every historical book
worthy of the name ought to include a chapter, or if one prefers, a series of
paragraphs inserted at turning points in the development, which might almost be
entitled: "How can I know what I am about to say?" I am persuaded that even
the lay reader would experience an actual intellectual pleasure in examining
these "confessions." The sight of an investigation, with its successes and
reverses, is seldom boring. It is the ready-made article which is cold and dull.
72 The Historian's Craft Sometimes I receive visits from people who wish
to write the history of their village. Regularly, I give them the following
standard advice, which I shall here simplify only a little in order to avoid the
irrelevant scholarly details. "Except in recent times, peasant communities have
rarely had any archives. The seigneuries on the other hand, as relatively well-
organized and lasting enterprises, usually kept their papers from early times. For
all the period prior to 1789, therefore, and especially for the most ancient times,
the principal documents which you can hope to use will be of sei- gneurial
origin. The result is that the first question you will have to answer, and upon
which almost everything hinges, is this: Who was the seigneur of the town in
1789?" (Actually, it is not at all improbable that there were several seigneurs at
the same time, sharing the village between them, but we shall discard this
supposition in the interest of brevity.) "Three eventualities are conceivable. The
seigneury could have belonged to a church, to a layman who emigrated during
the Revolution, or, to a layman who did not emigrate. The first instance is by all
odds the most favorable. The chances are that the records are both older and
better kept. They were certainly confiscated in 1790, along with the land, under
the Civil Constitution of the Clergy. Since they were then carried to some
public place, we have reason to hope that they are still there today, more or less
intact, and at the disposal of scholars. The hypothesis of the emigr£ is also
promising. In this instance, too, the records sJtiouid
Historical Observation 73 have been seized and transferred, although the
outside chance of their willful destruction, as a vestige of a hatred regime, is
rather more to be feared. The last possibility remains. It might prove infinitely
troublesome. The former nobles, if they did not leave France, or in some other
way fall afoul the laws of public safety, were not at all disturbed in their
property. Of course, they lost their seigneurial rights, because these had been
universally abolished, but they kept all their personal property and,
consequently, their business papers. Since they were never confiscated by the
state, the papers we are seeking, in this third case, met the common fate of all
family papers. Even if they have not been lost, eaten by rats, or scattered by the
caprice of sale or inheritance through the attics of three or four houses on
different estates, there is nothing to oblige their present owners to let you see
them." I have cited this example, because it seems to me entirely typical of the
conditions which frequently determine and limit documentation. A closer
analysis will not be without interest. We have just witnessed the revolutionary
confiscations playing the role of a deity who often favors the scholar: the
goddess Catastrophe. Innumerable Roman municipia have been transformed
into banal little Italian villages, from which the archaeologist unearths some few
vestiges of antiquity with difficulty. Only the eruption of Vesuvius preserved
Pompeii. Certainly, the great disasters have not consistently
74 The Historian's Craft served history. The invaluable papers of the
imperial Roman bureaucracy, as well as heaps of literary and historiographical
manuscripts, were engulfed in the disorder of the Invasion. Before our very
eyes, two world wars have razed monuments and storehouses of archives from a
soil steeped in a glorious tradition. Nevermore shall we be able to leaf through
the letters of the merchants of ancient Ypres. During the rout, I saw the order
book of an army intentionally burnt Nevertheless, the peaceable continuity of
social existence is much less favorable to the transmission of memory than is
sometimes supposed. Revolutions force the doors of safes, and put ministers to
flight before they have had time to burn their secret papers. In early judicial
archives, the bankruptcy records yield up to us the papers of business concerns
which, had they been permitted to live out a profitable and honorable existence,
would inevitably have turned over the contents of their files to the pulp
machines. Thanks to the admirable permanence of monastic institutions, the
abbey of St. Denis still preserved in 1789 the charters which had been granted
to it by the Merovingian kings a thousand years before. Yet it is in the National
Archives that we read them today. Had the monks of St Denis survived the
revolution, is it certain that they would permit us to rummage through their
coffers? Not very certain, I fear, since the Company of Jesus does not permit the
profane an access to its collections, without which so many problems of modern
history will always remain hopelessly obscure, and the
Historical Observation 75 Bank of France does not invite experts on the
First Empire to examine even its dustiest records. Indeed, the spirit of the secret
society is inherent in all corporations. Here it is that the historian of the present
finds himself plainly at a disadvantage: he is almost totally deprived of these
unintentional confidences. For compensation, it is true, he has at his disposal the
indiscretions which his friends whisper in his ear. Their intelligence, alas, is
difficult to distingush from gossip. A good cataclysm suits our business better.
So it will be, at least, until society begins to organize a rational self-knowledge
by controlling its records, instead of depending on calamities for its
information. To do so, it must come to grips with the two principles responsible
for forgetfulness and ignorance: that negligence which loses documents; and,
even more dangerous, that passion for secrecy—diplomatic secrecy, business
secrecy, family secrecy—which hides or destroys them. It is natural that the
notary should be forbidden to reveal his client's transactions. But the laws which
permit him to shroud the contracts of his greatgrandfather's clients in the same
impenetrable mystery—whereas, nothing strictly hinders his letting their papers
turn to dust—are truly antediluvian. The motives which prompt the majority of
great corporations to refuse to make public statistics absolutely indispensable
for the sound conduct of the national economy are seldom respectable. Our
civilization will take an immense forward stride on the day when concealment,
raised to a rule of action and almost to a bourgeois
70 The Historian's Craft virtue, shall give way to the desire for
information, which is necessarily the desire to exchange information. But let us
get back to our village. The circumstances which, in this particular instance,
have determined the loss or the preservation, the accessibility or the
inaccessibility of the evidences have their origins in historical forces of a
general nature. They present no feature which is not perfectly intelligible, but
they are stripped of all logical connection with the object of the inquiry even
though the result of that inquiry is found to depend upon them. For it is not
immediately clear why, for example, the study of a little rural community in the
Middle Ages should be more or less informative, according as its owner, several
centuries later, should or should not have taken it into his head to join the forces
assembling at Coblenz. Nothing is more prevalent than this paradox. If we
know infinitely more about Roman Egypt than about Gaul in the same period, it
is not because we are more interested in the Egyptians than in the Gallo-
Romans: rather, it is that the dryness, the sand, and the rites of mummification
have there preserved writings which the climate and customs of the Occident
condemned to rapid destruction. The causes which make for success or failure
in the search for documents ordinarily have nothing in common with the
reasons which render these documents desirable: this inevitable element of the
irrational imparts to our research a tinge of that inner
Historical Observation n tragedy in which, perhaps, so many creations of
the mind discover not only their limitations but one of the secret reasons for
their failure. Again, in the example cited above, the fate of the documents,
village by village, once the decisive facts were known, became almost
predictable. Such is not always the case. Sometimes the result depends on the
final intertwining of so many independent lines of causation that all prediction
proves impossible. I know that four successive conflagrations, and then a
plunder- ing, devastated the archives of the ancient abbey oi Saint Benolt-sur-
Loire. How, on this basis, could 1 guess in advance what sort of papers these
ravages have chosen to spare? What has been called the migration of
manuscripts is, in itself, an extremely interesting subject of study. The progress
of a literary work through the libraries, the execution of copies, and the care or
negligence of librarians and copyists fully correspond to the vicissitudes and
interplay of the cultural main streams of real life. But could even the best-
informed scholar have predicted, prior to the discovery, that the sole manuscript
of Tacitus' Germania would come up high and dry in the sixteenth century in
the monastery of Herzfeld? In a word, at the bottom of nearly every search for
documents there is a residue of the unexpected and, hence, of the fortuitous. A
fellow worker, whom I knew well, once told me this story: On a shell-torn
beach at Dunkerque, he was awaiting a doubtful rescue without betraying too
much impatience when one of his comrades addressed him with a
78 The Historian's Craft look of amazement. "Extraordinary! You don't
even seem to mind this awful uncertainty!" My friend could have answered that,
despite the popular prejudice, the mental climate of research is not so
unsympathetic to ready acceptance of the lottery of fate. A while ago we asked
whether there is an antithesis of technique between knowledge of the past and
of the present. The answer has already been given. Certainly, the explorers of
the present and those of remoter times have each their particular way of
handling their tools. Moreover, both have their advantages, depending on the
particular case. The former have a more tangible grasp of life; but the latter in
their investigations command means which are often denied to the first. Thus,
the dissection of a cadaver discloses to the biologist many secrets which the
study of a living subject would fail to reveal, but is mute about many others
which are evident only in the living body. But, to whatever age of mankind the
scholar turns, the methods of observation remain almost uniformly dependent
upon "tracks," and are, therefore, fundamentally the same. So, too, as we shall
see, are those critical rules which observation must obey if it is to be fruitful.
[79] CHAPTER III HISTORICAL CRITICISM 1. An Outline of the
History of the Critical Method The most naive policeman knows that a witness
should not always be taken at his word, even if he does not always take fiill
advantage of this theoretical knowledge. Similarly, it has been many a day since
men first took it into their heads not to accept all historical evidence blindly. An
experience almost as old as mankind has taught us that more than one
manuscript has falsified its date or origin, that all the accounts are not true, and
that even the physical evidences can be faked. In the Middle Ages, in the face of
an abundance of forgeries, doubt was frequently a natural defensive reflex.
"With ink, anyone can write anything." Thus exclaimed an eleventh-century
country squire of Lorraine in reference to some monks who had armed
themselves in a lawsuit against him with documentary proofs. The Donation of
Constantine— that extraordinary literary concoction which a Roman cleric of
the eighth century ascribed to the first Christian emperor—was contested, three
centuries later, in the circle of the eminently pious Otto III. False relics have
been hunted down almost from the first. However, skepticism on principle is
neither a more
80 The Historian's Craft estimable nor a more productive intellectual
attitude than the credulity with which it is frequently blended in the simpler
minds. In the first war, I knew a worthy veterinarian who, with some
justification, refused categorically to believe anything in the newspapers. Yet
the fellow swallowed hook, line, and sinker the most nonsensical hocus-pocus
which any chance companion might pour in his eager ear. Similarly, the
criticism of ordinary common sense, for long the only one in use, and still
somehow seductive to certain minds, cannot lead very far. In reality, this
pretended common sense usually turns out to be nothing more than a compound
of irrational postulates and hastily generalized experiences. As regards the
physical world, it has denied the existence of the antipodes. It still denies the
Einsteinian universe. It treated as mere legend Herodotus* tale reporting that,
when turning the coast of Africa, the navigators saw the point from which the
sun rises pass from their left to their right. As it regards human actions, on the
other hand, the worst of common sense is that it exalts to the level of the eternal
observations necessarily borrowed from our own brief moment of time. This is
the principal vice of the Voltairian criticism, which is so penetrating in other
respects. Above and beyond the peculiarities of individuals of every age, there
are states of mind which were formerly common, yet svhich appear peculiar to
us because we no longer share them. "Common sense/' it seems, would refuse to
accept the idea that Emperor Otto I could have
Historical Criticism 81 signed, in favor of the Pope, grants of territories
which could never be made good, since they both belied his former actions and
were ignored in those that followed. Since his grant was incontestably authentic,
however, we are forced to believe that his mentality was different from ours
and, more particularly, that there was in his time a gap between words and
deeds which surprises us today. True progress began on the day when, as
Volney put it, doubt became an "examiner"; or, in other words, when there had
gradually been worked out objective rules which permitted the separation of
truth from falsehood. The Jesuit Papebroeck, in whom the reading of The Lives
of the Saints had instilled a profound mistrust of the entire heritage of the early
Middle Ages, considered all the Merovingian charters which had been
preserved in the monasteries as forgeries. No, replied Mabillon. There are
unquestionably *ome charters which have been retouched, some which have
been interpolated, and some which have been forged in their entirety. There are
also some which are authentic, and this is how it is possible to distinguish the
bad from the good. That year, 1681, the year of the publication of the De Re
Diplomatica, was truly a great one in the history of the human mind, for the
criticism of the documents of archives was definitely established. Moreover, it
was in every respect the decisive moment in the history of the critical method.
The hu-
82 The Historian's Craft manism of the preceding age had had its stray
impulses and its intuitions. It had gone no farther. Nothing is more
characteristic than a passage from the Essais in which Montaigne justifies
Tacitus for having reported miracles. It is, he says, the business of theologians
and philosophers to debate the "common body of belief/' Historians for their
part have only to "relate" it, as given by their sources, "so that they render
history rather as they receive it, than as they evaluate it" In other words, a
philosophical criticism, resting upon a certain conception of the natural or
divine order, is perfectly legitimate: and it is understood for the rest that
Montaigne assumes responsibility neither for the miracles of Vespasian nor for
a good many others. But he obviously does not really understand how it is
possible to conduct an examination, specifically a historical examination, of
evidence such as this. The doctrine of research was worked out in that
seventeenth century whose true glory, although it is sometimes misplaced,
belongs to its second half. The men of that time were themselves aware of it
Between 1680 and 1690, it was a commonplace to denounce the "Pyrrhonism of
history" as a momentary fad. "It is said," wrote Michel Levassor, commenting
upon this term, "that rectitude of mind consists in not being too ready to believe
and, on a number of occasions, in knowing how to doubt." The very word
"criticism," which up to that time had connoted little more than a judgment of
taste, is here passing to the almost new sense of a test of truth. At first, they
ven^
Historical Criticism S3 tured the term only apologetically, for "it is not
entirely good usage/' which is to say that the word still retained a technical
flavor. Nevertheless, it gained steadily. Bossuet shied away from it. When he
speaks of "our critical authors," we sense the shrugging of shoulders. Richard
Simon, on the other hand, incorporated it into the title of almost all his works.
The most circumspect were not misled by it. Indeed, what this word heralded
was the discovery of a method of nearly universal applicability. According to
Ellies du Pin, criticism was that "species of torch which lights our way down
the darkened corridors of antiquity, enabling us to distinguish the true from the
false/' Bayle was even clearer: "M. Simon, in his new Reponse, has laid down a
number of rules of criticism, which can be useful, not only for the
understanding of the Scriptures, but also for the profitable reading of many
other works/' Now, let us compare some birth dates: Papebroeck (who, if he was
mistaken about the Merovingian charters, nevertheless, has a place among the
founders of historical criticism), 1628; Mabillon, 1632; Richard Simon (whose
works dominate the beginnings of Biblical exegesis), 1638. Outside the
company of the scholars, properly speaking, let us add Spinoza—Spinoza of the
Tractatus Theologico-Politicus, that pure masterpiece of philological and
historical criticism— also 1632. In the strictest sense of the word, it is a
generation whose outlines take shape before us with amazing clarity. But we
must be more precise. This
84 The Historians Craft was the generation born at about the moment of
the appearance of the Discours de la Methode* We must not call it a generation
of Cartesians. Mabillon, for one, was a devout monk, naively orthodox, whose
last-written work was a tract upon La Mart Chretienne. It is doubtful that he had
any very direct knowledge of the new philosophy which was then regarded with
suspicion by so many pious folk, and it is still more doubtful that he would have
found very much to approve in it even if he had had any chance glimjnerings.
Then again—whatever the perhaps unduly celebrated pages of Claude Bernard
may seem to suggest—the truths of that mathematical sort of evidence for
which Descartes's systematic doubt sought to pave the way present few
common characteristics with those increasingly close approximations to truth
which both historical criticism and laboratory science are content to define. But
for a philosophy to impregnate an entire age, it is not necessary that it should
act precisely in accordance with a prescribed formula nor that the majority of
minds should come under its influence except by a sort of osmosis of which
they are often only half aware. Like Cartesian "science/' the criticism of
historical evidence makes a tabula rasa of its beliefs. Like Cartesian science,
too, it proceeds inexorably to knock out the props of antiquity only in order
thereby to arrive at new certainties (or very strong probabilities), which are
thenceforth duly proved. In other words, the idea
Historical Criticism 85 supposes an almost total overturn of older
conceptions of doubt. Up to that time, whether men considered that its sting
might seem painful, or found in it an infinitely high-minded sweetness, they
conceived of it only as a purely negative mental attitude, a mere vacuum.
Thenceforth, they thought that, rationally conducted, doubt could become an
instrument of knowledge. It is an idea which appeared at a very precise moment
in the history of thought. From that time, in short, the basic rules of the critical
method were fixed. Their general significance was so little overlooked that
among the subjects most frequently proposed for the competitive examinations
in philosophy by the University of Paris in the eighteenth century was one
having a curiously modern ring: "The testimony of men upon historical facts."
This is certainly not to imply that the later generations did not greatly improve
the tool of criticism. Above all, they widely generalized its use and extended its
applications. For a long time, critical techniques were practiced, at least with
any consistency, almost exclusively by a handful of scholars, exigetes, and
connoisseurs. Writers engaged in historical works of the high-flown sort
scarcely bothered to familiarize themselves with such laboratory exercises, far
too detailed for their taste, or even to take their results into account. Now, as
Humboldt put it, it is never good for chemists to be afraid
86 The Historian's Craft "of getting their hands wet/' For history, the
danger of a split between preparation and execution is double- edged. At the
outset, it cruelly vitiates the great attempts at interpretation. Because of it, these
not only fail in their primary duty of the patient quest for truth, but, deprived of
that perpetual renewal, that constantly reborn surprise, which only the struggle
with documents can supply, they inevitably lapse into a ceaseless oscillation
between stereotyped themes imposed by routine. But technical work suffers no
less. No longer guided from above, it risks being indefinitely marooned upon
insignificant or poorly propounded questions. There is no waste more criminal
than that of erudition running, as it were, in neutral gear, nor any pride more
vainly misplaced than that in a tool valued as an end in itself. The conscientious
effort of the nineteenth century struggled valiantly against these perils. The
German school, Renan, Fustel de Coulanges won for erudition its intellectual
stature. The historian has been brought back to the workbench. But has the
opposition been entirely won over? It would be optimistic to think so. Too
often, the work of research still wanders aimlessly with no rational decision
about where it is to be applied. Above all, criticism has not yet succeeded in
winning over that wholehearted approval of "good men and true" (in the former
sense of that phrase) whose backing, no doubt necessary to the moral health of
any science, is more particularly indispensable to our own. If men, who are the
object of our study, fail to under-
Historical Criticism 87 stand us, how can we feel that we have
accomplished more than half our mission? In truth, perhaps, we have not fully
accomplished it The grim esoterism, in which even the best of us sometimes
fall, the preponderance, in our current writing, of those dreary textbooks which
bad teaching- concepts have put in place of true synthesis, the curious modesty
which, as soon as we are outside the study, seems to forbid us to expose the
honest groping of our methods before a profane public—all these bad habits,
derived from an accumulation of contradictory prejudices, compromise the
essential nobility of our cause. They conspire to surrender the mass of
defenseless readers to the false brilliance of a bogus history, in which lack of
seriousness, picturesque rubbish, and political prejudices are supposed to be
redeemed by shameless self-assurance: thus, Maurras, Bainville, or Plekhanov
affirm that which Fustel de Coulanges or Pirenne would have doubted. A
misunderstanding between historical inquiry, such as it is or hopes to be, and
the reading public unquestionably does exist. The great debate about footnotes
is not the least significant giound upon which the two parties are engaged in
their absurd duel. For a great many scholars, the lower margin of the page
exerts a fascination bordering upon mania. It is surely absurd to overcrowd
these margins, as they do, with bibliographical references which might largely
have been spared by a list drawn up at the beginning
88 The Historian's Craft of the volume; and worse still, through sheer
laziness, to relegate to them long explanations whose proper place was
indicated in the main body of the text, so that the most useful part of these
works must be looked for in the cellar. But when certain readers complain that a
single note, strutting along by itself at the foot of the page, makes their heads
swim, or when certain publishers claim that their customers, doubtless less
hypersensitive in reality than they would have us believe, are tortured by the
mere sight of a page thus disfigured, these aesthetes merely prove their
impervious- ness to the most elementary maxims of an intellectual ethic. For,
apart from the free play of imagination, we have no right to make any assertion
which cannot be verified and a historian who in using a document indicates the
source as briefly as possible (that is, the means of Ending it again) is only
obeying a universal rule of honesty. Corrupted by dogma and myth, current
opinion, even when it is least hostile to enlightenment, has lost the very taste for
verification. On that day when, having first taken care not to discourage it with
useless pedantry, we shall succeed in persuading the public to measure the value
of a science in proportion to its willingness to make refutation easy, the forces
of reason will achieve one of their most smashing victories. Our humble notes,
our finicky little references, currently lampooned by many who do not
understand them, are working toward that day.
Historical Criticism 89 The documents most frequently dealt with by the
early scholars either represented themselves or were traditionally represented as
belonging to a given author or a given period, and deliberately narrated such
and such events. Did they speak the truth? Were the books ascribed to Moses
really his? Were the charters bearing the name of Clovis authentic? How valid
were the accounts of Exodus, or those of the Lives of the Saints? Such was the
problem. Because history has tended to make more and more frequent use of
unintentional evidence, it can no longer confine itself to weighing the explicit
assertions of the documents. It has been necessary to wring from them further
confessions which they had never intended to give. Now, the critical rules,
which proved themselves in the first instance, work equally well in the second. I
have before me a batch of medieval charters. Some are dated; others are not.
Wherever a date appears, it must be verified, for experience proves that it may
be false. Wherever it is missing, it is important to establish it. In either case, the
same course will be pursued. On the basis of the script (if it is an original), the
style of the Latin, the institutions alluded to, and the general aspect of the
enacting clause, I conjecture that a certain deed corresponds to the readily
recognizable practices of French notaries about the year a.d. 1000. If it claims to
belong to the Merovingian period, the fraud is exposed. If it has no date, one
has been approximately established. Similarly the archaeologist, whether he
intends to classify prehistoric tools accord-
90 The Historian's Craft ing to their age and civilization, or to track down
false antiquities, examines, compares, and defines types and techniques of
workmanship according to rules which are fundamentally the same in either
case. The historian is not—indeed, he is less and less— that rather grumpy
examining magistrate whose unflattering portrait is easily imposed upon the
unwary by certain introductory manuals. To be sure, he has not turned
credulous. He knows that his witnesses can lie or be mistaken. But he is
primarily interested in making them speak so that he may understand them. It is
not the least admirable feature of the critical method that, without the least
modification of its first principles, it has successfully continued to direct
research towards this larger goal. However, it would be wrongheaded to deny
that incorrect evidence was not only the stimulus to the first efforts for a
technique of truth, but continues to be the starting-point from which that
technique must necessarily proceed in order to develop its analyses. 2. In
Pursuit of Fraud and Error Of all the poisons capable of vitiating a piece of
evidence, the most virulent is deception. This can take two forms. First of all,
there is deceit as to the author and the date: forgery, in the legal sense of the
word. Not all the letters published with Marie Antoinette's signature were
written by her, there are some which were forged in the ninteenth century. The
so-called tiara of Saitphern&s, sold to the
Historical Criticism 91 Louvre as a Scythio-Greek antiquity of the third
century b.c., had been engraved at Odessa in 1895. Secondly, there is
misrepresentation of the facts. In the Commentaries, whose authorship is
incontestible, Cagsar has consciously distorted or omitted a great deal. The
statue displayed at St, Denis as representing Philip the Bold is, indeed, the
funerary figure of that king, as it was executed after his death; but everything
suggests that the sculptor simply reproduced a conventional model which is a
portrait in name only. Now, these two aspects of fraud raise entirely distinct
problems, with quite separate solutions. Certainly, the majority of those writings
which bear a forged name falsify their contents as well. The Protocols of the
Elders of Zion, apart from the fact that they were not written by the Elders of
Zion, made the widest possible departures from the substance of the truth. If an
alleged charter of Charlemagne should prove, upon examination, to have been
forged two or three centuries later, the chances are that the acts of generosity
which it attributes to the Emperor are equally fictitious. Nevertheless, even this
cannot be taken for granted, for some deeds have been forged with the sole
purpose of repeating the dispositions of entirely authentic papers which have
been lost. In exceptional cases, then, a forgery may speak the truth. Inversely, it
should be superfluous to recall that that evidence which is entirely above
suspicion as to its avowed origin is not, for that reason, necessarily truthful.
However, scholars take such pains to weigh a docu-
92 The Historian's Craft ment before accepting it as authentic that
afterwards they sometimes lack the stamina to criticize its contents. Moreover,
they are particularly loath to doubt those writings which are sanctioned by
impressive legal guarantees, such as actions of official authority or even private
contracts. Yet neither deserves much respect. On April 21,1834, prior to the
prosecution of the secret societies, Thiers wrote to the prefect of the lower
Rhine: "I advise you to take the greatest care to furnish your share of documents
for the great forthcoming investigations. The correspondence of all anarchists,
the intimate connections between events in Paris, Lyons, Strassburg, and, in a
word, the existence of a vast conspiracy embracing the whole of France—all
this must be made entirely clear." Unquestionably, here is a well- prepared
official documentation! As for duly sealed and dated charters, the least
experience of the present is enough to dispel all illusions about them.
Everybody knows that the most regularly established and notarized deeds teem
with intentional inaccuracies, and I remember recently having been ordered to
antedate my signature at the foot of a report demanded by one of the great
government services of the state. Our ancestors were not more fastidious in this
respect. "Promulgated upon such and such a day in such and such a place," they
say at the bottom of royal charters. But if you consult the accounts of the king's
travels, you will see more than once that, upon the specified date, he was
actually several leagues away. Innumerable acts of enfranchisement of serfs,
which no one who is not mad
Historical Criticism 93 would dream of arguing as forgeries, are asserted to
derive from pure charity, even though we can place beside them the actual bill
paid for emancipation. But to establish the fact of forgery is not enough. It is
further necessary to discover its motivations, if only as an aid to tracking it
down. So long as there is any doubt about its origins, there is something in it
which defies analysis and which is, therefore, only half proved. Above all, a
fraud is, in its way, a piece of evidence. Merely to prove that the famous charter
of Charlemagne to the church at Aix-la-Chapelle is not authentic is to avoid
error, but not to acquire knowledge. On the other hand, should we succeed in
proving that the forgery was committed by the followers of Frederick
Barbarossa, and that it was designed to implement dreams of imperial grandeur,
we open new vistas upon the vast perspectives of history. Here, then, we see
criticism seeking out the impostor behind the imposture. In other words, we see
it conforming to the basic motto of history, by seeking out man. It would be
puerile to pretend to enumerate the infinite variety of reasons which can lead to
lying. But the historian, who tends naturally to overintellectualize mankind,
would do well to remember that all of these reasons are not in fact reasonable.
With certain people, lying (although it is itself generally joined to a compound
of egotism and suppressed desire) is almost, as Andr6 Gide put it, a "gratuitous
act." The German savant who took such great pains to write, in
84 The Historian's Craft excellent Greek, an oriental history whose
authorship he attributed to the fictitious Sanchoniathon, might, with less effort,
have acquired a respectable reputation as a Hellenist Frangois Lenormant, the
son of a member of the Institute and himself later called to join that honorable
company, began his career at seventeen by deceiving his own father with the
fraudulent discovery at La Chapelle Saint-Eloi of inscriptions carved entirely by
his own hand; even when old and laden with honors, his last masterly coup was
said to have been the publicizing as Greek originals of a number of
commonplace prehistoric relics which he had simply gathered up from the
French countryside. Now, there have been mythomaniac epochs, as well as
individuals with a passion for lying. Such were the preromantic or romantic
generations toward the end of the eighteenth and beginning of the nineteenth
century. There were the pseudo-Celtic poems attributed to Ossian, the epic
poems and ballads written in what Chatterton believed was Old English, the
allegedly medieval poetry of Clothilde of Surville, the Breton songs devised by
Viiiemarqu4 songs said to have been translated from Croatian by M6rim£e, the
heroic Czech songs from the manuscript of Kravoli-Dvor, and others too
numerous to mention. During these decades, it was as if a vast symphony of
deception resounded from one end of Europe to the other. The Middle Ages,
especially from the eighth to the twelfth century, presents another example of
this mass epidemic. Certainly, the majority of the false charters,
Historical Criticism 95 capitularies, and pontifical decrees which were then
forged in such great numbers were designed to serve some selfish interest. To
secure disputed property to a church, to support the authority of the Roman See,
to defend the monks from the Bishop, the bishops from the metropolitans, the
Pope from the temporal rulers, or the Emperor from the Pope—such were the
sole objectives of the forgers. Their common characteristic is that persons of
incontestable piety and even of integrity did not hesitate to put their hands to
these forgeries. Obviously, such forgeries were hardly offensive to public
morality. As for plagiarism, it was at this time universally regarded as the most
innocent act in the world. Annalists and hagiographers shamelessly
appropriated entire passages from the writings of earlier authors. Although so
different in other respects, there is nothing less "futuristic" than these two
societies. The Middle Ages knew no other foundation for either its faith or its
laws than the teachings of its ancestors. Romanticism wished to steep itself in
the living spring of the primitive, as well as in that of the popular. So it was that
the periods which were the most bound by tradition were also those which took
the greatest liberties with their true heritage. It is as if, in a curious
compensation for an irresistible creative urge, they were naturally led, by the
sheer force of their veneration of the past, to invent it. In the month of July
1857, Michel Chasles, the mathematician, presented to the Academy of Science
96 The Historian's Craft an entire collection of unpublished letters of
Pascal, which had been sold to him by his regular dealer, the brilliant forger
Vrain-Lucas. From these it appeared that the author of the Provinciates had
formulated the principle of gravitation prior to Newton. This amazed a certain
English savant. How was it possible to explain, he asked, in effect, that these
writings took note of astronomical measurements made many years after
Pascal's death, and of which even Newton had no knowledge until after the
publication of the first articles of his work? Vrain-Lucas was not the man to be
troubled by such a mere trifle. He again sat down to his desk, and, before long,
fortified by his painstaking labors, Chasles was able to produce some new
"originals." This time, Galileo was the signer and Pascal the recipient. The
mystery was explained as follows: the illustrious astronomer had provided the
observations and Pascal the calculations. The whole correspondence had been
secret on both sides. It is true that Pascal was only eighteen at the death of
Galileo. But what of that? It was but another reason to admire the precocity of
his genius. Nevertheless, the indefatigable objector noted still another
peculiarity: in one of the letters, dated 1641, Galileo complains that he cannot
write without greatly tiring his eyes. Now, do we not know that he had actually
been stone blind since the close of 1637? "I beg your pardon," replied the good
Chasles after a little; "I admit that, until now, everyone had believed in that
blindness. But they were quite mistaken. For, it so
Historical Criticism 97 happens that I can now add to the discussion a
decisive paper which will refute this universal error. On December 2, 1641,
another Italian scholar informed Pascal that Galileo, whose sight had
unquestionably been failing for several years, had just lost it entirely/' Certainly,
not all forgers have displayed as much imagination as Vrain-Lucas; nor all
dupes so much gullibility as his pitiable victim; but the experience of life
teaches, and that of history confirms, that any or- fense against the truth is like a
net and that almost inevitably every lie drags in its train many others,
summoned to lend it a semblance of mutual support. That is why so many
famous forgeries occur in clusters. In the false privileges of the See of
Canterbury, in the false privileges of the Duchy of Austria—signed by so many
great sovereigns, from Julius Caesar to Frederick Barbarossa—and in the
forgeries of the Dreyfus affair which spread like a genealogical table (and I cite
but a few examples), we seem to perceive a growing gangrene. By its very
nature, one fraud begets another. Finally, there is a more insidious form of
deception. In place of blunt, forthright, and, I might almost say, honest untruths,
there are the sly alterations: interpolations in authentic charters or the
embroidering of imaginary details upon the roughly trustworthy scheme of a
narrative. Interpolations are generally founded on self-interest. Embroidery is
frequently for the purpose of embellishment. The havoc wrought upon ancient
or medieval historiography by a mistaken aesthetic sense
98 The Historian's Craft has often been exposed. Its role in our press is
perhaps no less important. The most unpretentious of our newspapermen
intentionally presents his characters, even at the cost of the truth, in accordance
with a rhetorical tradition whose glamor our age has not outworn, and our
editorial staffs include more disciples of Aristotle and Quintilian than is
generally believed. Certain technical conditions themselves seem to favor these
distortions. In 1917, when the spy, Bolo, was condemned, one daily is said to
have published the account of his execution on April 6. In reality, although it
was originally set for that date, it did not actually take place until eleven days
later. The journalist had set up his "copy" beforehand; convinced that the event
would take place upon the day anticipated, he deemed it useless to verify it. I do
not know what the anecdote may be worth. Certainly, such awkward mistakes
are the exception, but it is not at all an improbable assumption that, since time is
of the essence in delivering the copy, the reports of anticipated events are often
prepared in advance for the sake of expediency. Almost always, we may be
sure, the sketch will be changed on all important points, if that seems advisable
after observation; on the other hand, it may be doubted that very many
alterations are made in the supplementary details which are deemed necessary
for color and which nobody cares to verify. At least, that is how it seems to a
layman. It is to be wished that a professional man would supply us with genuine
enlightenment on the subject. Unfortunately, the press
Historical Criticism 99 has not yet found its Mabillon. Certainly,
submission to a rather antiquated code of literary propriety, deference to a
stereotyped psychology, and the rage for the picturesque are nowhere near to
losing their place as causes of fabrication. If only because of the ease with
which chance may transform the sincerest blunder into a lie, there are many
gradations from sham, pure and simple, to the entirely unintentional error. The
act of inventing a lie presupposes an effort which is distasteful to the mental
inertia common to the majority of men. It is much easier to accept with
complacency an illusion, at first spontaneous, which gratifies the interest of the
moment. Take the famous case of "the airplane of Nuremberg." Although the
fact has never been definitely established, it seems highly probable that a
French commercial plane flew over the city several days prior to the declaration
of war. It is probable that it was taken for a military plane. It is not improbable
that, in a population already a prey to the specter of the forthcoming conflict,
there were spread rumors of bombs dropped here and there. Nevertheless, it is
certain that none were dropped; that the officials of the German government
possessed every means of suppressing the false rumor; and hence that, in
welcoming it without verification, to use it as a motive for war, they clearly
lied. But they did so without inventing anything, and even, perhaps, without
being clearly conscious of their
100 The Historians Craft deception at the outset. The absurd rumor was
believed, because it was useful to believe it. Of all the types of deception, not
the least frequent is that which we impose upon ourselves, and the word
"sincerity" has so broad a meaning that it cannot be used without admitting a
great many shadings. It is no less true that many witnesses deceive themselves
in all good faith. The time is now ripe for the historian to profit from the
precious results of those observations from life which in the last few decades
have provided the tools of a nearly new discipline: the psychology of evidence.
In so far as they involve our studies, these discoveries seem to be essentially as
follows. If we are to believe William of St. Thierry, his friend and disciple, St.
Bernard, was very surprised to learn one day that the chapel where he had daily
attended divine service as a young monk opened onto the chevet by three
windows, although he had always imagined that it had had only one. The
hagiographer expresses astonishment and admiration at this trait, for does not
such a detachment from the things of this world betoken a perfect servant of
God? Unquestionably, Bernard appears really to have been uncommonly
absent-minded—at least if it is true, as is also related, that once later, he
traveled along the shores of Lake Geneva for an entire day without being aware
of it. Nevertheless, there is plenty of proof to show that it is not necessary to be
one of the foremost mystics to be
Historical Criticism 101 grossly mistaken about those realities which
should seemingly be the best known to us. In the course of famous experiments
at Geneva, the students of Professor Clapar&de proved themselves as incapable
of accurately describing the entrance hall of their university as did the doctor
"according to the honeyed word" the church of his monastery. The truth is that
the majority of minds are but mediocre recording- cameras of the surrounding
world. Add that, since evidence, strictly speaking, is no more than the
expression of remembrance, the first errors of perception run the constant risk
of being entangled with the errors of memory, that loose, that "slippery"
memory, denounced long ago by one of our old jurists. For certain minds,
inaccuracy possesses a truly pathological fascination—would it be too
irreverent to call this psychosis "Lamartine's disease"? At any rate, we all know
that its victims are not ordinarily the most reluctant to make positive statements.
But if this is so of more or less dubious witnesses, experience shows that there
are no witnesses whose statements are equally reliable on all subjects and under
all circumstances. There is no reliable witness in the absolute sense. There is
only more or less reliable testimony. Two principal sorts of circumstances
impair the accuracy of perception of even the most gifted person. The first
depends upon the condition of the observer at the time—such, for example, as
his fatigue or emotion—the second upon the degree of his attention. With few
exceptions, we see and really understand only that to
102 The Historian's Craft which we devote our particular concentration. If
a physician visits a sickbed, I am more willing to credit his report about the
appearance of the patient whom he has carefully examined, than about the
furniture of the room, to which he has probably given only a passing glance.
That is why, contrary to a common prejudice, the most familiar objects—in the
case of St. Bernard, the chapel at Citeaux—are usually those of which it is most
difficult to get an accurate description, for familiarity almost inevitably breeds
indifference. Now, a great many historical events can have been observed only
in moments of violent emotional confusion, or by witnesses whose attention,
whether attracted too late in the event of surprise, or preoccupied by the need
for immediate action, was incapable of sufficient concentration upon those
features in which historians have reason to be most interested today. Certain
examples are notorious. Whence came the first shot which precipitated the riot
in front of the Office of Foreign Affairs on February 25, 1848—and from which
in its turn the revolution was to result? Did it come from the troops or from the
crowd? In all likelihood we shall never know. How then can we now take
seriously the full-blown descriptive passages of the chroniclers, with their
detailed portrayals of dress, movements, ceremonies, and feats in battle? By
what stubborn habit of mind are we to preserve the least illusion of the accuracy
of all this stuff, the delight of the small fry of romantic historians, when we see
around us no one with the ability to remember
Historical Criticism 103 correctly in their entirety those details of the kind
which are naively sought for in the ancient authors? At best, these tableaux give
us the setting of the action, as contemporaries of the writer imagined it should
have been. That is extremely informative, but it is not the sort of information
which the lovers of the picturesque generally desire from their sources.
However, it is advisable to understand what conclusions about the nature of our
studies are implied by these remarks, which are pessimistic perhaps in
appearance only. They do not affect the fundamental structure of the past. The
words of Bayle remain forever true. "No valid objection will ever be raised to
the fact that Caesar defeated Pompey. On whatever grounds we may choose to
argue it, there is hardly anything more unshakeable than this proposition:
Caesar and Pompey existed. They were not merely in the minds of those who
wrote their lives." It is true that, if there were no certainty except for a few facts
of this type devoid of explanation, history would be reduced to a series of rough
notations without much intellectual value. Happily, such is not the case. It is
only the most immediately antecedent causes which are frequently rendered
uncertain by the psychology of evidence. A great event can be compared to an
explosion. Under exactly what conditions was produced the last molecular
shock, indispensable for the expansion of the gas? We must often resign
ourselves to ignorance. This is no doubt regrettable, but what of the chemists?
Their position is not always much bet-
104 The Historian's Craft ter. Nevertheless, the composition of the
explosive mixture remains perfectly susceptible of analysis. The revolution of
1848, despite the fact that, by a strange aberration, certain historians have
thought to make it the archetype of the fortuitous event, was a movement
clearly determined by a great number of extremely diverse and dynamic factors,
which had been preparing its way for a long time and which in fact enabled a
Tocqueville to foresee it. As for the fusillade on the Boulevard de Capucines,
what was that but the last little spark? Moreover, we shall see that all too often
the immediate causes escape our witnesses', and hence our own, observation.
They embody that privileged part of history which is unforeseeable,
"accidental." We may easily console ourselves if the flimsiness of evidence
conceals them from our subtlest instruments. Even if they were better known,
their connection with the great causal chains of evolution would represent that
residue of fallacy which our science will never eliminate, and which it has no
right even to pretend to eliminate. As for the hidden causes of human destiny,
the fluctuations in mental or emotional climate, the changes in techniques, and
the variations in the social or economic structure, our witnesses will hardly be
subject to the frailty of instantaneous perception. It is a happy coincidence, long
ago glimpsed by Voltaire, that what is most profound in history may also be the
most certain.
Historical Criticism 105 The faculty of observation is as variable among
societies as among individuals. Certain epochs show themselves more wanting
in this respect than others. For example, however feeble the understanding of
numbers among the majority of men today, it is no longer so universally
defective as among the medieval annalists. Our perception, like our civilization,
is saturated with mathematics. Were errors of testimony due solely to failings of
the senses or of attention, the historian would have little choice but to abandon
the subject to the psychologist. But, beyond common mental slips, there are
many errors that derive from a particular social climate. Such errors often
assume a documentary value in their turn. In September 1917, the infantry
regiment to which I belonged held the trenches of the Chemin des Dames to the
north of the little town of Braisne. In one attack, we took a prisoner. He was a
reservist, a wholesale merchant by trade, and originally from Bremen on the
Weser. Shortly thereafter, a curious tale came up from behind the lines.
"German espionage!" our well-informed comrades said, in effect "What a
marvel it is! We find one of their outposts in the heart of France. Astounding! A
merchant stationed in peacetime Braisne/'* We must beware of explaining the
story too Simply. To blame it all on an error of hearing is inexact. The error was
not merely in hearing but in understanding. Because the name x Note that
Br£me, the French for Bremen, sounds very much like Braisne in rapid speech.
106 The Historian's Craft was generally unfamiliar, it had not caught the
attention. By a natural mental quirk, a familiar name was substituted in its
place. Furthermore, a second and equally unconscious interpretation was at
work in the first The idea of German cunning, frequently all too true, had been
popularized by innumerable anecdotes, highly gratifying to the romantic
sensibilities of the masses. The substitution of Braisne for Bremen accorded too
well with this obsession not to have been spread, as it were, spontaneously.
Now, such is the case with a great number of the distortions of evidence. Nearly
always, the nature of the error is determined in advance. More particularly, it
does not spread, it does not take on life, unless it harmonizes with the prejudices
of public opinion. It then becomes as a minor in which the collective
consciousness surveys its own features. On the fronts of a great many Belgian
houses, there are narrow apertures designed to help the plasterers in setting up
their scaffolding; the German soldiers in 1914 would never have envisioned
these innocent contrivances of the masons as so many loopholes prepared for
snipers, if their imaginations had not long been deranged by the fear of
guerrillas. Clouds have not changed their shapes since the Middle Ages, yet we
no longer see in them either magical swords or miraculous crosses. The tail of
the comet sighted by the great Ambroise Par6 was probably very little different
from those which occasionally sweep across our skies. Yei; he thought he saw
in it a full suit of curious armor. Com-
Historical Criticism 107 pliance with universal prejudice had bested the
habitual accuracy of his gaze; and his testimony, like that of so many others,
tells us not what he actually saw but what his age thought it natural to see.
However, for the error of a single witness to become that of many men, for an
inaccurate observation to be transformed into a false rumor, social conditions
must be such as to favor its circulation. Certainly not all types of society are
equally auspicious. The extraordinary disturbances of collective life in our time
include many remarkable experiences in this regard. Those of the present
moment are too close to us to permit of exact analysis as yet. The war of 1914-
18 offers a greater perspective. Everyone knows how productive of false news
these four years proved to be, particularly among the combat troops. It is their
formation in the very extraordinary society of the trenches which is the most
interesting to study. The role of propaganda and censorship was considerable,
but in a way exactly the reverse of what the creators of these institutions
expected of them. As one witness very neatly remarked: "The prevailing
opinion in the trenches was that anything might be true, except what was
printed." The men put no faith in newspapers, and scarcely more in letters, for
these, besides arriving irregularly, were thought to be heavily censored. From
this there arose a prodigious renewal of oral tradition, the ancient mother of
myths and
108 The Historian's Craft legends. Wiping out bygone centuries by a
daring stroke, beyond the wildest dream of the boldest experimenters,
governments reduced the frontline soldier to the means of information and the
mental state of olden times before journals, before news sheets, before books.
Rumors did not ordinarily originate on the firing- line. There, the little groups
were too isolated from one another. The soldier could not move about except
under orders; most frequently, moreover, he did so only at the risk of his life.
Occasionally, intermittent travelers went the rounds: liaison officers, artillery
observers, telephone communications men repairing their lines. These notable
personages scarcely associated with the common soldier. But there were also
regular communications of far more importance. They were made necessary by
the demand for sustenance. The agom of this little world of dugouts and
observation posts were the field kitchens. There, once or twice a day, the
carriers from the various points of the sector came together, met again, and
chatted amongst themselves or with the cooks. The latter knew a great deal, for,
situated on the crossroads of all units, they had the additional and rare privilege
of exchanging a few words daily with the drivers of the regimental service
corps, fortunate men who were quartered in the vicinity of staff headquarters.
Thus, for a moment, around fires in the open air or the grates of the field
kitchens, there were momentary contacts among very dissimilar groups. Then
the fatigue parties, moving
Historical Criticism 109 off by the trails and communication trenches,
brought to the most forward parts of the front, along with their camp kettles,
this mass of intelligence true or false, almost always distorted in every
circumstance and ready for further elaboration there. On the tacti* cal maps, a
little behind those interlacing lines which designate the forward positions, it
would be possible to hatch in a continuous strip which would be the myth-
making zone. Now, history has known more than one society governed, on the
whole, by analogous conditions, but with this difference: that instead of being
the passing effect of an entirely exceptional crisis they here represent the
normal texture of life. Here also, almost the only effective means of
transmission is oral. Here too, the liaison between widely separated elements is
carried on almost exclusively by specialized intermediaries or at definite points
of juncture. Peddlars, jugglers, pilgrims, beggars take the place of the little
fellows wandering through the communication trenches. The regular meetings
occur at markets or on the occasion of religious holidays, as, for example,
during the high Middle Ages. Based on information gained from cross-
questioning passers-by, monastic chronicles greatly resemble the notebooks
which our supply corporals could have kept if they had had the desire. Such
societies have always been excellent culture media for false news. Frequent
contacts among men make it easy to compare divergent stories. They stimulate
the critical sense. On the other hand, we have faith in
110 The Historian's Craft that narrator who, at rare intervals, brings us
distant rumors over a difficult road. 3. Toward a Logic of the Critical Method
Criticism of testimony, since it deals with psychic realities, will always remain
a subtle art. There is no recipe for it. However, it is also a rational art, which
depends on methodical use of certain basic mental processes. In a word, it has a
dialectic of its own which we ought to try to define. Let us suppose that only
one object is left from a lost civilization, and moreover that the conditions of its
discovery forbid placing it even in a non-human context, such as geological
sedimentation. (For inanimate nature may play its part in such research.) It
would be entirely impossible to date this unique vestige, or even to render a
verdict as to its authenticity. In fact, we can never establish a date, we can never
verify, and, in short, we can never interpret a document except by inserting it
into a chronological series or a synchronous whole. It was by comparing
Merovingian charters, now with each other, now with other texts of a different
nature and time, that Mabil- Ion founded the science of diplomatics. It was the
collation of the stories of the Gospel which gave rise to Biblical exegesis. At the
bottom of nearly all criticism there is a problem of comparison. But the
outcome of this comparison has nothing of
Historical Criticism 111 the automatic about it. Of necessity, it ends by
revealing similarities and differences. Now, depending upon the circumstances,
agreement of one testimony with other testimonies may lead to opposite
conclusions. To begin with, we must first consider the elementary matter of
narrative. In the Memoirs which have thrilled so many young hearts, Marbot
relates, with a great multiplicity of detail, a feat of courage of which he makes
himself the hero: according to him, on the night of May 7, 1809, he crossed the
raging torrents of the Danube, which was then in full flood, in order to free
some prisoners of the Austrians on the other side. How are we to verify the
anecdote? We must summon the other evidences to the rescue. We possess the
orders, the records of march, and the reports of the opposing armies. They attest
that, upon the famous night, the Austrian corps whose bivouac Marbot claimed
to have located on the left bank was still on the opposite side. In another
connection, it is evident from the Correspondence of Napoleon himself that, on
May 8, the high waters had not begun. Finally, we have discovered a petition
for promotion drawn up by Marbot in person on June 30, 1809. Among the
claims he here set forth, he breathed not a word about his supposed exploit of
the preceding month, On one side, then, we have the Memoirs; on the other, a
whole batch of documents which belie them. We must now decide between
these conflicting witnesses. Which
112 The Historian's Craft alternative will be judged the most hkely: that the
general staff and the Emperor himself were simultaneously mistaken (unless,
God only knows why, they had knowingly falsified reality) and that the Marbot
of 1809, desperately eager for advancement, had erred through false modesty,
or that, much later, the old warrior, whose boasts are notorious in other
connections, had won another bout with the truth? Surely no one will hesitate.
The Memoirs have lied again. Here, then, the statement of a disagreement has
destroyed one of the conflicting testimonies. One of them had to yield. The
most universal of logical postulates demanded such an outcome. The principle
of contradiction pitilessly denies that an event can be and not be at the same
time. There are in the world scholars whose good nature has worn itself out in
seeking a middle ground between antagonistic statements. They are like the
little chap who, asked for the square of the number two, when one neighbor
whispered "four" and the other "eight," thought he had hit the mark in
answering "six." It then remained for us to choose which evidence should be
rejected and which retained. It was decided by a psychological analysis: the
supposed reasons for truthfulness, for deceit, and for error were weighed for all
the witnesses in turn. In this particular case, it was found that this appraisal
assumed the character of nearly absolute evidence. It would unfailingly show a
much larger coefficient of uncertainty under other circumstances. Conclusions
which are founded upon a
Historical Criticism 113 delicate imputation of motive fall upon a
graduated scale from the infinitely probable to the barely credible. But now here
are some examples of another sort. A charter which claims to be of the twelfth
century is written on paper, whereas all the originals of that period hitherto
recovered are on parchment; the shape of the letters here seems very different
from the forms which we see in other documents of the same time; the language
abounds in words and tricks of style entirely foreign to their practice. Or, again,
the edge of a tool, supposed to be paleolithic, betrays processes of workmanship
which, according to our knowledge, were used only in much more recent times.
We conclude that the charter and the tool are forgeries. As before, the
disagreement is damning, but for reasons of a very different nature. This time,
the line of argument is that within a single generation of the same society there
prevails a similarity of custom and technique too strong to permit any person to
deviate sensibly from the common practice. We take for granted that any
Frenchman of the time of Louis VII drew his downstrokes in pretty much the
same way as his contemporaries,* that he 2 In my youth, I heard a very
illustrious scholar, who was director of the Ecole des Chartes, tell us rather
haughtily: "I date the handwriting of a manuscript within twenty years without a
mistake." He overlooked only one thing. Many men and many scribes live more
than forty years—and if handwritings sometimes change with age, it is rare for
them to adapt themselves to the new handwritings
114 The Historian's Craft expressed himself in pretty much the same terms
and that he made use of the same materials. Likewise, we assume that if a
craftsman of the Magdalenian tribe had invented a mechanical saw to cut out his
arrowheads, his comrades would have used it as he did. In brief, the postulate in
this case is of a sociological order. Nevertheless, the resemblance ought not to
be too strong. It would then cease to support the case. On the contrary, it would
weaken it. Anyone who took part in the battle of Waterloo knew that Napoleon
was beaten. Any witness so singular as to deny the defeat we should regard as a
liar. Moreover, if we confine ourselves to the simple, blunt statement, there are
not very many different ways of saying that Napoleon was beaten at Waterloo.
But should two witnesses describe the battle in exactly the same language or,
despite a certain variation of phrasing, with exactly the same details, we should
unhesitatingly conclude that one of them had copied the other or that both had
copied a common model. In effect, our reason refuses to admit that two
observers, necessarily posted at different parts of the field and endowed with
unequal powers of attention, could have noted the same episodes, detail for
detail; that two writers, working independently of one another, should about
them. About 1200, there must have been some sextagenarian scribes who still
wrote as they had been taught about 1150. In fact, the history of handwriting is
curiously behind that of language. It is waiting for its Diez-or its Meillet
Historical Criticism 115 accidentally have chosen from among the
innumerable words of the French language the same terms, similarly arranged,
in order to narrate the same things. If the two accounts claim to have been taken
directly from reality, then at least one of them must be lying. Again, let us
consider two battle scenes sculptured in stone upon two ancient monuments.
They refer to different campaigns, yet they are depicted with nearly the same
details. The archaeologist would say: "Unless both artists were satisfied with
reproducing a traditional design, one has certainly plagiarized from the other/' It
matters little that the combats had been separated by but a short interval, that
perhaps they involved adversaries from the same peoples—Egyptians against
Hittites, Assur against Elam. We rebel at the thought that, considering the
immense variety of human positions, two distinct actions at different moments
could have exactly repeated the same gestures. As evidence of the military
annals which they pretend to record, at least one, if not both of these
representations, is, strictly speaking, a fraud. Thus, criticism oscillates between
two extremes: the similarity which vindicates and that which discredits. This is
because there is a limit to coincidence, and social unity is made up of links
which are, on the whole, rather weak. In other words, we estimate that the
universe and society possess sufficient uniformity to exclude the possibility of
overly pronounced deviations. But, as we picture it to ourselves, this uniformity
is confined to some very general characteristics. It includes,
116 The Historian's Craft we think upon delving further into reality, a
number of possible combinations so nearly infinite that their spontaneous
repetition is inconceivable: there must be a voluntary act of imitation. And so,
to add it all up, the criticism of evidence relies upon an instinctive metaphysics
of the similar and the dissimilar, of the one and the many. Once having set up
the hypothesis of a copy, we must establish the direction of the influence. In
each case, we must ask whether the two documents have borrowed from a
common source. If, on the other hand, we suppose one of them to be original, to
which shall we accord the honor of this title? Sometimes the answer is provided
by external criteria, such as the relative dates, if these can be established.
Failing these aids, psychological analysis, making use of the internal
characteristics of the objects or texts, again comes into its own. It goes without
saying that this conforms to no mechanical rules. For example, is it necessary,
as certain scholars seem to think, to lay down the principle that adapters are
continually adding new fantasies, so that the odds are always that the most
restrained and least improbable text is the oldest? Sometimes this is true. The
number of enemies fallen under the blows of an Assyrian king are seen to swell
enormously from inscription to inscription. But sometimes, too, common sense
protests. The most fabulous of the Passions of St. George is the first in date;
taking up the old ac-
Historical Criticism 117 count afterward, the successive biographers have
sac* rificed, one after the other, those features whose unrestrained fantasy
shocked them. There are many different ways of imitating. They vary with the
individual and, sometimes, with the vogue common to a generation. Like all
other mental attitudes, they cannot be taken for granted on the pretext that they
seem natural to us. Fortunately, the plagiarists are often betrayed by their own
blunders. When they do not understand their model, their misinterpretation
proclaims the imposture. If they attempt to disguise their borrowings, the
awkwardness of their stratagem betrays them. I knew a schoolboy, who,
riveting his gaze upon his neighbor's paper during a written examination,
painstakingly wrote down the sentences in reverse. With high cunning he
changed the subjects to predicates and the actives to passives. He succeeded
only in providing his professor with an excellent example of historical criticism.
To unmask an imitation is to reduce two or more witnesses to only one. Two of
Marbot's contemporaries, the Count de Segur and General Pelet, have given
accounts of the supposed crossing of the Danube which are analogous to his
own. But S£gur came after Pelet. He had read him. He did little more than to
copy him. As for Pelet, he did indeed write before Marbot; but he was his
friend, and, doubtless, he had often heard him recall his fictitious feats of valor
—for, in deceiving his friends, the indefatigable braggart
118 The Historians Craft was intentionally preparing to mystify posterity.
Since his seeming corroborators were only repeating his words, Marbot is then
left really as our sole authority. When Livy repeats Polybius, even with
embellishments, it is Polybius who is our sole authority. And when Einhard,
while pretending to describe Charlemagne for us, plagiarizes the portrait of
Augustus by Suetonius, we have really no witness left at all. Finally, there are
times when a prompter, who does not wish to identify himself, is hiding behind
the self- styled witness. In studying the trial of the Templars, H. C Lea observed
that whenever two defendants of two different houses were examined by the
same inquisitor they invariably confessed to the same atrocities and the same
blasphemies. On the other hand, if defendants from the same house were
questioned by different inquisitors, their confessions ceased to agree. The
obvious conclusion is that the judge dictated the answers. I imagine that the
annals of the judiciary might provide other examples of this peculiarity. Surely,
nowhere in the field of critical reasoning does the part played by what might be
termed limited similarity appear in a more curious light than in one of the
newest applications of method: statistical criticism. Let us suppose that I have
made a study of prices between two set dates in a tightly knit society in which
there is active exchange. Later a second worker, and then a third, undertake the
same research, but with
Historical Criticism 119 the help of data differing from both mine and each
other's: other account books, other market prices. Each of us separately draws
up his annual averages, graphs, and index numbers on a common base. The
three curves nearly coincide. We should conclude that each of them furnishes a
more or less exact picture of the trend. Why? The reason is not solely that in a
homogeneous economic milieu large-scale price fluctuations must necessarily
conform to a reasonably uniform rhythm. No doubt, this consideration would be
enough to throw suspicion upon drastically divergent curves, but it is not
enough to assure us that among all the possible plottings, that curve upon which
the three graphs agreed must necessarily be the true one, simply because they
agreed. Three scales whose balances are similarly false will give the same
reading, and that reading will be false. Here, all reasoning relies upon an
analysis of the mechanics of error. None of the three price lists could be held to
be free of errors of detail. In statistical matters, they are nearly inevitable. Even
if we disregard the personal mistakes of the investigator (and which of us has
not blundered frightfully in the appalling maze of ancient weights and
measures?), however miraculously attentive we imagine his research, there will
always remain the pitfalls within the documents themselves. Certain prices may
have been listed inaccurately, either through inadvertence or bad faith; others
may be exceptional—prices for friends or, conversely, prices for fools. All of
these are
120 The Historian's Craft very apt to upset the averages. Lists recording
average prices prevailing at the market were not always prepared with perfect
care. Over a vast number of prices, however, these errors compensate for each
other, for it would be extremely unlikely that they should all tend in the same
direction. If, then, results attained by the use of different data are confirmed by
their agreement, it is basically because agreement in the errors of the data—the
oversights, the petty details, and the petty favors—seems inconceivable to us.
The irreducible diversity in the evidence leads us to conclude that what there is
of final agreement must derive from a reality whose fundamental unity, in this
case, was beyond a doubt The reagents for the testing of evidence should not be
roughly handled. Nearly all the rational principles, nearly all the experiences
which guide the tests, if pushed far enough, reach their limits in contrary
principles or experiences. Like any self-respecting logic, historical criticism has
its contradictions or, at least, its paradoxes. We have seen that, for a piece of
evidence to be recognized as authentic, method demands that it show a certain
correspondence to the allied evidences. Were we to apply this precept literally
to the letter, however, what would become of discovery? For to speak of
discovery is also to speak of surprise and dissimilarity. A science which
restricted itself to stating that every-
Historical Criticism 121 thing invariably happens according to expectation
would hardly be either profitable or amusing. Up to now, there have been found
no charters drafted in French (instead of Latin, as previously) which are anterior
to 1204. Let us imagine that, tomorrow, a scholar should bring forward a French
charter dated 1180. Are we to conclude that the document is fraudulent, or that
our knowledge had been insufficient? A seeming contradiction between a new
piece of evidence and its surroundings may well have its source only in a
temporary defect in our knowledge. But there are sometimes genuine
disagreements in the ob* jects themselves. Social uniformity is not so powerful
as to be inescapable for certain individuals or small groups. Under the pretext
that Pascal did not write like Amauld, or that Cezanne did not paint like Bon-
guereau, are we to refuse to admit the recognized dates of the Provinciates or
the Montagne Ste. Victoire? Are we to infer that the oldest bronze tools are
forgeries, because most contemporary strata contain only stone tools? These
false conclusions are not at all imaginary, and one could make a long list of
facts which scholarly routine first denied because they were surprising—from
the Egyptian zoolatry with which Voltaire was so highly amused, down to the
Roman remains of the tertiary era. On closer inspection, however, the
methodological paradox is only on the surface. The principle of reasoning from
similarity loses none of its
122 The Historian's Craft force. It is only essential that a more exact
analysis should distinguish the range of possible divergence, while making clear
the necessary points of similitude. For all individual originality has its limits.
The style of Pascal belongs to him alone; but his grammar and the stock of his
vocabulary belong to his time. Though it employs an unusual language, our
hypothetical charter of 1180 is not impossibly different from other previously
known charters of the same date. It can be judged acceptable if its French
conforms in general to the state of the language as known from the literary texts
of that epoch, and if the institutions which it mentions correspond to those of its
time. Moreover, rightly understood, critical comparison is not content to collate
evidences from the same plane of time. A human phenomenon is always linked
to a chain which spans the ages. On the day when a new Vrain-Lucas, throwing
a handful of autographs upon the table of the Academy, shall pretend to prove
that Pascal invented general relativity before Einstein, we may assume in
advance that the papers will be forged. It is not that Pascal could not have
discovered what his contemporaries did not. But the theory of relativity took its
point of departure from a long anterior development of mathematical
speculation. However great he may be, no man can dispense with the labor of
generations by the sheer force of his genius. When, on the contrary, certain
scholars, confronted with the first discoveries of paleolithic painting, contested
their authenticity or date on the pretext that such an art
Historical Criticism 123 could not have flourished and then vanished, these
skeptics reasoned ill: chains may break, and civilizations may die. As Father
Delehaye writes, in substance, anyone reading that the church observes a
holiday for two of its servants both of whom died in Italy on the veiy same day,
that the conversion of each was brought about by the reading of the Lives of the
Saints, that each founded a religious order dedicated to the same patron, and
finally that both of these orders were suppressed by popes bearing the same
name—anyone reading all this would be tempted to assert that a single
individual, duplicated through error, had been entered in the martyrology under
two different names. Nevertheless, it is quite true that, similarly converted to the
religious life by the example of saintly biographies, St. John Colombini
established the Order of Jesuates and Ignatius Loyola that of the Jesuits; that
both of them died on July 31, the former near Siena in 1367, the latter at Rome
in 1556; that the Jesuates were dissolved by Pope Clement IX and the Jesuits by
Clement XIV. If the example is stimulating, it is certainly not unique. Should
some future cataclysm destroy all but the bare outlines of the philosophical
works of the past centuries, what a crowd of conscientious doubts would beset
the scholars of the future as to the existence of two thinkers, who, both
Englishmen, both bearing the name of Bacon, agreed in assigning an important
place in their doctrines to empirical knowledge? M. Pais has dismissed a great
many
124 The Historian's Craft ancient Roman traditions as mere legend, for
almost no other reason than that the same names, associated with tolerably
similar episodes, are in a like manner seen to repeat themselves. With all due
respect to the criticism of plagiarism, whose spirit denies the spontaneous
repetition of events or words, coincidence is one of those freaks which cannot
be eliminated from history. But it cannot be enough simply to acknowledge the
broad possibility of coincidence. Reduced to this simple statement, criticism
would waver eternally between pro and con. For doubt to become the tool of
knowledge it is necessary, in each particular case, that the degree of probability
of coincidence can be weighed with some exactitude. Here the path of historical
research, like that of so many other disciplines of the mind, intersects the royal
highway of the theory of probabilities. To evaluate the probability of an event is
to weigh its chances of taking place. That granted, is it legitimate to speak of the
possibility of a past event? Obviously not, in the absolute sense. Only the future
has contingency. The past is something already given which leaves no room for
possibility. Before the die was cast, the probability that any number might
appear was one to six. The problem vanishes as soon as the dice box is emptied.
Somewhat later we may not be sure whether, upon that day, a three or a five
actually turned up. The uncertainty, then, exists in us,
Historical Criticism 125 in our memory, or in that of our witnesses, and not
in the things themselves. In a correct analysis, however, the use which historical
research makes of the idea of probabilities is not at all contradictory. When the
historian asks himself about the probability of a past event, he actually attempts
to transport himself, by a bold exercise of the mind, to the time before the event
itself, in order to gauge its chances, as they appeared upon the eve of its
realization. Hence, probability remains properly in the future. But since the line
of the present has somehow been moved back in the imagination, it is a future
of bygone times built upon a fragment which, for us, is actually the past. If it is
incontestible that the event has taken place, these speculations have little more
value than that of a metaphysical game. What was the probability that Napoleon
would be born? That Adolf Hitler, the soldier of 1914, would escape French
bullets? We are not forbidden to amuse ourselves with these questions,
providing we understand them for what they really are: simple rhetorical
devices intended to illuminate the role of contingency and of the unforeseeable
in the progress of mankind. They have nothing to do with the criticism of
evidence. It is otherwise when the very existence of the fact seems uncertain.
For example, are we in doubt whether an author could spontaneously have
repeated many of the episodes, even many of the words, of another narrative,
without having copied it? Can we believe that chance alone, or some divinely
preordained harmony, suffices to explain
126 The Historian's Craft the very striking resemblance between The
Protocols of the Wise Men of Zion and the pamphlet of an obscure polemicist
of the Second Empire? According as such a coincidence should appear to have a
greater or lesser coefficient of probability prior to the composition of the
narrative, we shall either acknowledge or reject its likelihood today. For all that,
the mathematics of chance are based upon a fiction. From the outset, they
postulate impartial conditions in all possible cases: a specific cause favoring a
certain outcome in advance would be like a foreign body in the calculation. The
die of the theoretician is a perfectly balanced cube; were a single grain of lead
to be slipped under one of its surfaces, the chances of the players would cease to
be equal. But in the criticism of evidence, almost all the dice are loaded. For
extremely delicate human elements constantly intervene to tip the balance
toward a preferred possibility. One historical discipline, strictly speaking, is
excepted. It is linguistics, or, at least, that branch of it which is concerned with
establishing the relationships between languages. It is very different from
critical studies in the strict sense, yet it shares with these studies the common
characteristic of seeking to discover affiliations. Now, the terms upon which it
reasons are extraordinary close to the a priori convention of equality, as found
in the theory of chance. It owes this prerogative to the very peculiarities of
linguistic phenomena. Indeed, not only does the immense num-
Historical Criticism 127 ber of possible combinations of sounds reduce to
virtual insignificance the probability of any considerable fortuitous repetition in
different languages but, what is still more important, the meanings attributed to
these combinations are entirely arbitrary, aside from some few imitative
onomatopoeic words. No association of prior ideas dictates that the sound of tu,
as it is pronounced in either French or Latin, should serve to indicate the second
person. If, then, we find that this sound has this role in French, in Italian, in
Spanish, and in Rumanian—if we simultaneously observe a mass of other
equally inational similarities between these languages—the only sensible
explanation would be that French, Italian, Spanish, and Rumanian have a
common origin. Because the various possibilities were unaffected by human
interests, a practically pure mathematical calculation of the chances has carried
the decision. But this simplicity is far from the norm. Several of the charters of
a medieval sovereign, dealing with different matters, repeat the same words and
the same constructions. Therefore, the fanatics of "stylistic criticism" assert that
they were drawn up by the same notary. This might be granted, were chance the
sole consideration. But such is not the case. Each society and, still more, each
little professional group has its habits of language. As a result, it is not enough
to enumerate the points of similarity. It is further necessary to distinguish the
unusual from the commonplace among them. Only the really exceptional ex-
128 The Historian's Craft pression can identify an author; assuming, of
course, that its repetitions are sufficiently numerous. The error here lies in
attributing equal weight to all elements of speech: whereas in fact the variable
coefficients of social preference affecting each of them are like the grains of
lead which upset the equality of chance. Since the beginning of the nineteenth
century, a whole school of learned men has devoted itself to the study of the
transmission of literary texts. The principle is simple. We have three
manuscripts of the same work: B, C, and D. We may ascertain that all three give
the same, obviously erroneous, reading. (This is the method of errors, the oldest
method, that of Lachmann.) Or, more often, we find that all three show the
same readings, whether good or bad, but different from most of those of other
manuscripts. (This is the complete listing of the variants recommended by Don
Quentin.) In either case, we should decide that the three are "related/"
Depending upon the cir^ cumstances, it would be understood either that they
had been copied from each other in a sequence which remains to be determined,
or that they all derived by separate ancestries from a common model. In short, it
is quite certain that such a sustained coincidence could not be fortuitous.
Nevertheless, two observations of rather recent date have forced textual
criticism to abandon much of the quasimechanical rigor of its first conclusions.
Copyists have sometimes corrected their model. Even when they worked
independently from each
Historical Criticism 129 other, common mental habits must very often have
suggested similar conclusions. Terence somewhere uses the word raptio, which
is extremely rare. Not understanding it, two scribes substituted ratio, which
makes no sense but which was familiar to them. In so doing, was it necessary
for them to act either in concert or in imitation of each other? There you see one
type of error which is practically incapable of teaching us anything about the
"genealogy" of manuscripts. There is more. Why should the copyist use only a
single model? When he was able to compare several copies, he was not
forbidden ultimately to choose from among the variants to the best of his
ability. Certainly, this was very seldom the case in the Middle Ages, when
libraries were scanty; to all ap* pearance, it was much more frequent in
antiquity. What place are we to assign to those incestuous fruits of several
different traditions upon the beautiful trees of Jephthah which are customarily
drawn up at the beginning of critical editions? The will of the individual, like
the pressure of collective forces, cheats pure chance in the game of
coincidences. Thus, as the philosophy of the eighteenth century had already
seen with Volney, the majority of the problems of historical criticism are really
problems of probability, but such that the subtlest calculation must own itself
incapable of their solution. It is not only that its data are extraordinarily
complex. Most frequently, by their very nature they are unamenable to any
mathematical translation. For example, how are we to cal-
130 The Historian's Craft cukte the particular preference which a society
accords to a word or a custom? The science of a Fermat, a Laplace, and an
fimile Borel will not rid us of our difficulties. But at least, since it is placed at
the inaccessible extremity of our logic, we may call upon it to aid us, from on
high, toward a better analysis of our reasoning. He who has not lived among
scholars does not realize how loath they ordinarily are to admit the innocence of
a coincidence- Because two similar expressions are found in both the Salic law
and an edict of Clovis, a reputable German savant has declared that the law
must have derived from that prince. Let us pass over the banality of words used
here and there, A mere tinge of mathematical theory would be enough to
prevent such a false idea. When chance has a free hand, the order of probability
of a single coincidence or of a small number of coincidences is seldom
impossibly high. Even if they seem astonishing to us, the surprises of common
sense are rarely sensations of much value- It may be amusing to calculate the
probability that a stroke of chance should fix the deaths of two entirely separate
personages upon the same day of the same month of two different years. It is
equal to 1/3652* • This would be correct on the supposition that the chances of
dying were equal for each day of the year. This is not exact (there is an annual
curve of mortality), but it may be conveniently postulated here.
Historical Criticism 131 Let us now giant (despite the absurdity of the
postulate) that the suppression of the orders of both John Colombini and
Ignatius Loyola by the Roman Church was preordained. The examination of the
pontifical lists enables us to establish that the probability of the abolition by two
popes of the same name was as 11 to 13. The combined probability of the
deaths falling on the same day of the same month and of two homonymous
popes being the authors of the condemnation falls somewhere between 1/1,000
and 1/1,000,000.* Doubtless, no betting man would be satisfied with these
odds. But the natural sciences consider as next to impracticable, on the earthly
scale, even possibili- * From the death of John Colombini until today, 65 popes
have ruled the church (induding the dual and triple papacies of the period of the
Great Schism); since the death of Ignatius there have been 38. Hie first list gives
55 who are homonymous with the second, in which these same names are
repeated exactly 38 times (the popes, as we know, customarily taking names
already hallowed by tradition.) The probability that the Jesuates would be
suppressed by one of these homonymous popes was, therefore 55/65 or 11/13.
For the Jesuits, it rose to 38/38 or 1; hi other words, it became a certainty. The
combined probability is of 11/13 times 1, or 11/13. Finally, 1/3652 or 1/133,225
times 11/13 equals 11/1,731,925, which is equal to slightly more than
1/157,447. To be entirely accurate, it would be necessary to take into account
the respective durations of the pontificates. But the nature of this mathematical
diversion, whose sole object is to shed light upon the order of magnitude,
appeared to me to justify simplification of the calculations. [Trans&tfor's note:
The reader should here recall that Marc Block's manuscript was incomplete, and
thatT during the writing of it, he had no access to those reference books with
which he would certainly have checked and corrected the mathematical
inaccuracies contained in this passage. In any case, these inaccuracies do not
affect the main line of the reasoning, which is essentially correct.]
132 The Historian's Craft ties of the order of 10/15. Obviously, we are
wide of the mark—with good reason, as the reliably verified example of the two
saints bears witness. It is in the accumulation of coincidences that the
probability becomes practically negligible: for, by virtue of a well- known
theorem, the probabilities of simple facts must be multiplied by each other in
order to give the probability of the combination, and, since the probabilities are
fractions their product is, by definition, smaller than its components. In
linguistics, there is the famous example of the word bad, which means the same
thing in English and Persian, although the term has not the faintest common
origin in the two languages. Anyone who should pretend to found a connection
upon this isolated agreement would sin against the main law of all criticism of
coincidences: "Only large numbers are conclusive." The substantial agreements
or disagreements are composed of a multitude of specific circumstances. All
things considered, accidental influences cancel each other out. If, on the other
hand, we should consider each element separately from the others, the effect of
these variables can no longer be eliminated. Even if the dice have been loaded,
an isolated throw will always be more difficult to foresee than the outcome of
the game; consequently, once played, it will be subject to a much greater variety
of explanations. That is why the further criticism delves into detail the more its
probabilities tend to be blurred. Taken separately, there is scarcely a word in our
modern version of
Historical Criticism 133 Oresteia which we may be certain of reading as
iEschy- lus wrote it In its entirety, however, we need have no misgivings that
our Oresteia is really that of iEschy- lus. There is more certainty in the whole
than in its parts. To what extent, however, are we justified in mouthing this
glorious word "certainty*? Mabillon, long ago, admitted that the criticism of
charters could not attain "metaphysical" certainty. He was quite right. It is only
for the sake of simplification that we sometimes speak of evidence rather than
of probabilities. But we are more aware today than in Mabillon's time, that that
convention is not peculiar to us. It is not "impossible," in the absolute sense of
the term, that the Donation of Constantino is authentic, or—according to the
whim of some scholars—that the Germania of Tacitus is a forgery. Nor is it, in
the same sense, "impossible" that a monkey might accidentally reconstruct
either the Donation or the Germaniaf letter for letter, simply by striking the keys
of a typewriter at random. 'The impossible physical event," Coumot has said, "is
nothing but an event whose probability is infinitely small." So far as it finds
certainty only by estimating the probable and the improbable, historical
criticism is like most other sciences of reality, except that it undoubtedly deals
with a more subtle gradation of degrees. Do we always rightly appreciate the
immense advance embodied in the advent of a rational method of
134 The Historians Craft criticism applied to human testimony? Advance, I
mean, not only for historical knowledge but for knowledge as a whole. Not long
ago, unless there were very good reasons, in advance, for suspecting witnesses
or narrators of falsehood, three fourths of all facts stated were facts accepted.
Nor was it very long ago. Lucien Febvre has excellently demonstrated, for the
Renaissance, that men neither thought nor acted differently then than in periods
quite close to ours, which is why their masterpieces are still a living inspiration
for us. Nor should we say that such was, by nature, the attitude of that credulous
throng whose ponderous mass—misled, alas! by more than one pseudo-savant
—is constantly threatening even in our own day to sweep our fragile
civilizations into the abyss of ignorance and folly. The steadiest minds did not
and could not escape the common prejudices at the time. Was it told that a
shower of blood had fallen? Why, then, there are showers of blood. If
Montaigne read in his beloved ancients this or that nonsense about a land whose
people were born without heads or about the miraculous strength of the little
fish known as the remora, he set them down among his serious arguments
without raising an eyebrow. For all his ingenuity in dismantling the machinery
of a false rumor, he was far more suspicious of prevailing ideas than of so-
called attested facts. In this way, as in the Rabelaisian myth, old man Hearsay
ruled over the physical as well as the human world. Perhaps even more over the
physical
Historical Criticism 135 world than the human. For, having a more direct
experience, men sooner doubted a human event than a meteor or an alleged
inegularity of organic life. If your philosophy was repelled by miracles, or your
religion by the miracles of other religions, you had to force yourself painfully to
find in these strange manifestations some ostensibly intelligible causes.
Whether works of the devil, or occult influences, these "causes" still belonged
to a system of ideas or images completely foreign to what we should today call
scientific thought. The audacity of denying the manifestations themselves
scarcely entered anyone's head. Pompanozzi, shining star of that Paduan school
so opposed to Christian supernaturalism, did not believe that kings, simply
because they were kings even if anointed with oil from the sacred ampulla,
could cure sick persons by touching them with their hands. Nevertheless, he did
not dispute the cures. He explained them by a physiological peculiarity which
he conceived as hereditary: the glorious privilege of a sacred function was
reduced to the curative virtue of a dynastic spittle. Now, if today we have been
able to clear our picture of the universe of so many fictitious marvels,
seemingly confirmed by the agreement of generations, we are doubtless
primarily indebted to the gradual evolution of the idea of a natural order
governed by immutable laws. But this notion itself could not have been
established so solidly, the observations which seemed to contradict it could not
have been elimi-
136 The Historian's Craft nated, except by the patient labor of an
experiment performed upon man himself as a witness. We are enabled
henceforth both to expose and to explain the imperfections of evidence. We
have acquired the right of disbelief, because we understand, better than in the
past, when and why we ought to disbelieve. And it is by this means that science
has succeeded in throwing off the dead weight of a great many spurious
problems. But, here as elsewhere, pure knowledge is not divorced from action.
Richard Simon, whose name stands among the first rank of our founding
fathers, has not merely left us some admirable lessons in exegesis. He also used
his keen mind to rescue simple souls persecuted by the stupid accusation of
witchcraft. There is nothing arbitrary in the coincidence. In both roles, the need
for intellectual discipline is the same. The same installment served both needs.
Obliged always to be guided by the reports of others, legal action is no less
interested than pure research in weighing their accuracy. The tools at its
disposal are not different from those of scholarship. They are in fact those
which scholarship originally forged. In the useful employment of doubt, judicial
practice has only followed, rather lag- gingly, in the footsteps of the Bollandists
and the Benedictines. And the psychologists themselves did not think to seek a
scientific object in human testimony, as directly observed and elicited, until
long after the confused memory of the past had begun to be subjected to a
rational proof. It is a scandal that in
Historical Criticism 137 our own age, which is more than ever exposed to
the poisons of fraud and false rumor, the critical method is so completely absent
from our school programs. It has ceased to be the mere humble auxiliary to
exercises of the study. Henceforth, far wider horizons open before it, and
history may reckon among its most certain glories that, by this elaboration of its
technique, it has pioneered for mankind a new path to truth and, hence, to
justice.
[138] CHAPTER IV HISTORICAL ANALYSIS J. Judging or
Understanding? The formula of the venerable Ranke is famous: the historian
has no other aim than to describe things "as they happened, wie e$ eigentlich
gewesen" Herodotus had earlier expressed it: "to narrate what was, ton eonta" In
other words, the scholar, the historian, is urged to efface himself before the
facts. Perhaps, like many maxims, this one has owed its success only to its
ambiguity. We can plainly read in it a counsel of integrity: this, we cannot
doubt, was Ranke's meaning. But there was, besides, a counsel of passivity, so
that we here see two problems arising simultaneously: that of historical
impartiality, and that of history as an attempt at reproduction or as an attempt at
analysis. But if there is really a problem of impartiality, it derives solely from
the fact that the word is itself equivocal. There are two ways of being impartial:
that of the scholar and that of the judge. They have a common root in their
honest submission to the truth. The scholar records—better still, he invites—the
experience which may, perhaps, upset his most cherished theories.
Historical Analysts 139 The good judge, whatever his secret heart's desire,
questions witnesses with no other concern than to know the facts, whatever they
may be. For both this is an obligation of conscience which is never questioned.
However, there comes a moment when their paths divide. When the scholar has
observed and explained, his task is finished. It yet remains for the judge to pass
sentence. If, imposing silence on his personal inclination, he pronounces it
according to the law, he will be deemed impartial. And he will be impartial in a
judicial sense, not in a scientific sense. For we can neither condemn nor absolve
without accepting a table of values which no longer refers to any positive
science. That one man has killed another is a fact which is eminently
susceptible of proof. But to punish the murderer assumes that we consider
murder culpable: which is, after all, only an opinion about which not all
civilizations have agreed. Now, for a long time, the historian has passed for a
sort of judge in Hades, charged with meting out praise or blame to dead heroes.
We cannot but believe that this attitude satisfies a deep-rooted instinct For all
teachers who have had to correct their students' papers know how reluctantly
these youngsters are dissuaded from playing the role of Minos or Osiris from
behind their desks. The words of Pascal are more to the point than ever: "We all
play God in judging: this is good or this is evil/' Men forget that a value
judgment has a raison d'etre only as preparation for an action and a meaning
solely in relation to a system
140 The Historian's Craft of consciously accepted moral references. In
daily life, the demands of conduct compel us to use these ordinarily rather
summary labels. Where we can no longer act, where commonly accepted ideas
differ profoundly from our own, such labels become an embarrassment. Are we
so sure of ourselves and of our age as to divide the company of our forefathers
into the just and the damned? How absurd it is, by elevating the entirely relative
criteria of one individual, one party, or one generation to the absolute, to inflict
standards upon the way in which Sulla governed Rome, or Richelieu the States
of the Most Christian King! Moreover, since nothing is more variable than such
judgments, subject to all the fluctuations of collective opinion or personal
caprice, history, by all too frequently preferring the compilation of honor rolls
to that of notebooks, has gratuitously given itself the appearance of the most
uncertain of disciplines. Hollow indictments are followed by vain
rehabilitations. Robespierrists! Anti-Robespierrists! For pity's sake, simply tell
us what Robespierre was. If the judgment only followed the explanation, the
reader could simply skip it. Unfortunately the habit of passing judgments leads
to a loss of taste for explanations. When the passions of the past blend with the
prejudices of the present, human reality is reduced to a picture in black and
white. Montaigne has already warned us on this head: 'Whenever judgment
leans to one side we cannot help distorting and twisting the narrative in this
direction/' Moreover, to plumb the
Historical Analysis 141 consciousness of another person, separated from us
by the interval of generations, we must virtually lay aside our own ego,
whereas, to say what we think, we need only remain ourselves. This is a less
arduous endeavor. How much easier it is to write for or against Luther than to
fathom his soul; to believe Pope Gregory VII about Emperor Henry IV, or
Henry IV about Gregory VII, than to unravel the underlying causes of one of
the greatest dramas of Western civilization! To rise above questions of
personality, consider the problem of the land confiscated during the Revolution.
During the Terror, and reversing the earlier legislation, the government decided
to sell it off in small lots without competitive bidding. Unquestionably, this
seriously compromised the best interests of the Treasury. Certain modem
scholars have burst out angrily against this policy. What courage they would
have displayed, had they dared to say so while sitting in the Convention! Far
from the guillotine, this violence without danger may amuse us, but it would be
more worthwhile to investigate what it was that the men of the Year III really
wanted. Primarily, they hoped to favor the acquisition of the land by the little
people of the rural area; in preference to the balancing of the budget, they
sought the relief of the poor peasants, as a guarantee of their fidelity to the new
order. Were they right or wrong? What do I care for a historian's belated
decision on this point? We should only beg him not to be so hypnotized by his
own choice as to forget that at the time another was possible. Neverthe*
142 The Historian's Craft less, the lesson of the intellectual development of
mankind is clear: the sciences have shown themselves ever more fruitful and,
hence, in the long run more practical, in proportion as they deliberately abandon
the old anthropocentrism of good and eviL Today, we should laugh at a chemist
who separated the bad gases, like chlorine, from the good ones like oxygen.
But, had chemistry adopted this classification in its infancy, it would have run
the grave risk of getting stuck there, to the great detriment of the knowledge of
matter. However, let us beware of pushing the analogy too far. The
nomenclature of the science of man will always have its peculiar characteristics.
That of the science of the physical world excludes ideological doctrine. There
the words "success" or "failure," "incompetence" or "ability" could, at best, play
the role only of fictions, forever laden with dangers. On the other hand, they
belong to the normal vocabulary of history. For history has to do with beings
who are, by nature, capable of pursuing conscious ends. We may admit that an
army commander who is engaged in a battle usually strives to win it. If the
forces are approximately equal on either side, and he loses, it is perfectly
legitimate to say that he has maneuvered poorly. If such mishaps were habitual
with him, we should not deviate from the most scrupulous judgment of fact by
observing that he was, doubtless, not a very good strategist. Again, if we are
considering a
Historical Analysis 143 monetary alteration, whose aim, let us assume, was
to favor debtors at the expense of creditors, to term it either excellent or
deplorable would be to take sides with one of the two groups, and thereby
arbitrarily to transfer into the past an entirely subjective idea of public welfare.
But let us imagine that, by some chance, the operation intended to lighten the
burden of debt has ended in actual fact—such cases are known —in the contrary
result. "It failed," we say, without thereby doing anything more than honestly
stating a reality. As in psychology as a whole, the unsuccessful act is one of the
essential data of human evolution. There is something else. Has our general,
perchance, led his troops to defeat intentionally? We should not hesitate to
charge him with treason: because in plain language that is the proper word for
it. It would be a pedantic refinement for history to reject the aid of the simple
and direct vocabulary of common usage. Next, we must still try to understand
how common contemporary ethics regarded such an act. Treason can be a sort
of conformity, as with the condottieri of bygone Italy. When all is said and
done, a single word, "under* standing/' is the beacon light of our studies. Let us
not say that the true historian is a stranger to emotion: he has that, at all events.
"Understanding/' in all honesty, is a word pregnant with difficulties, but also
with hope. Moreover, it is a friendly word. Even in action, we are far too prone
to judge. It is so easy to denounce. We are never sufficiently understanding.
Whoever dif-
144 The Historian's Craft fers from us—a foreigner or a political adversary
—is almost inevitably considered evil. A little more understanding of people
would be necessary merely for guidance, in the conflicts which are unavoidable;
all the more to prevent them while there is yet time. If history would only
renounce its false archangelic airs, it would help us to cure this weakness. It
includes a vast experience of human diversities, a continuous contact with men.
Life, like science, has everything to gain from it, if only these contacts be
friendly. 2. From the Diversity of Human Functions to the Unity of Consciences
Understanding, however, has nothing of the attitude of passivity about it. Two
things will always be necessary for the practice of a science: a subject-matter,
but also a man. Human reality, like that of the physical world, is vast and
variegated. A mere photograph, even supposing that such a mechanically
complete reproduction had meaning, would be undecipherable. Is it suggested
that the documents have already inserted a preliminary screen between past and
present? Certainly, they often eliminate at random. On the other hand, they
almost never organize their subject- matter according to the demands of an
intelligence which seeks to know. Like any scholar, like any mind which
perceives at all, the historian selects and sorts. In short, he analyzes. And, to
begin with, he seeks out the similarities in order to compare them.
Historical Analysts 145 I have before me a Roman funerary inscription,
carved from a single block, made for a single purpose. Yet nothing could be
more variegated than the evidences which there await the probing of the
scholar's lancet. If we are especially interested in linguistic matters, the words
and the syntax express the state of Latin as men sought to write it in that time
and place, and through the transparency of its half-erudite language we may be
able to catch a glimpse of everyday speech. On the other hand, should our
predilection incline towards the study of beliefs, we are at the very core of
otherworldly aspirations. Or, if toward the political system, we are overjoyed at
the name of an emperor or the date of a magistracy. Toward the economy, the
epitaph may perhaps reveal an unknown trade. And I pass over the other
possibilities. Instead of an isolated document, let us now consider any given
moment in the evolution of a civilization which is known through a number and
variety of documents. There was not one of the men then alive who did not
participate almost at one and the same time in multiple manifestations of human
activity; who did not speak to and make himself understood by his neighbors;
who did not have his gods; who was not a producer, trader, or merely a
consumer; who, if he took no part in political events, did not at least experience
their consequences. Should we venture to recount all these activities without
selection and rearrangement, in the very same confusion in which each
document, each individual or
146 The Historian's Craft collective life presents them to us? That would
be to sacrifice clarity, not to the true order of reality—which is composed of
natural affinities and underlying connections—but to the purely superficial
order of contemporaneousness. A notebook of experiments is not to be confused
with the moment-by-moment diary of everything that has happened in the
laboratory. For when we think we see a kinship between certain phenomena in
the course of human evolution, what is it we mean except that each type of
institution, belief, practice, or event thus classified appears to us to express a
particular and, up to a certain point, a permanent tendency of the individual or
society? For example, should we deny that, despite all contrasts, all religious
emotions have something in common? It necessarily follows that we shall
always better understand any human fact if we already have an understanding of
other facts of the same sort. The use of money in the first period of feudalism,
more as a standard of values than as a means of payment, differed profoundly
from the established norms of Western economy about 1850; in turn there is
scarcely less contrast between the monetary systems of the mid- nineteenth
century and that of today. However, I submit that a scholar who knew only the
currency of about the year 1000 would find it difficult to grasp the peculiarities
of its use even at that date. It is this which justifies certain specializations which
are, in a sense, vertical: justifies them in the limited way in which
specializations can be Justified; that is, as correctives
Historical Analysis 147 to the lack of breadth of our minds, and the short
span of our lives. To neglect to organize rationally what comes to us as raw
material is in the long run only to deny time- hence, history itself. For can we
understand this or that period of Latin if we detach it from the earlier
development of the language? This form of ownership, or those beliefs were
not, of course, absolute beginnings. Inasmuch as their development proceeds
from the most ancient to the most recent times, human phenomena are governed
primarily by chains of similar phenomena. To classify them according to kind is
to lay bare the principal effective lines of force. But, some will object, the
distinctions which you establish in this way by cutting across life itself exist
only in your mind; they do not exist in reality, where everything is intermingled.
Moreover, you are making use of "abstraction." Granted. But why be afraid of
words? No science could dispense with "abstractions" any more than it could
dispense with imagination. Be it said in passing, it is significant that the same
thinkers who would banish the former generally display an equal ill-humor
towards the latter. It is the same badly understood positivism in both cases. The
sciences of man are no exception. In what way is the function of chlorophyll
more "real," in the absolute sense, than the economic function? Only those
classifications which depend upon false resemblances would be disastrous. It is
the business of the historian to be always testing his classifications in order to
justify their east-
148 The Historian's Craft ence and, if it seems advisable, to revise them.
Moreover, despite their common effort to encompass reality, they may start
from very different vantage points. For example, we have "the history of law."
The textbooks, always admirable tools of sclerosis, have popularized the term.
But, what does it mean? A legal rule is a social norm, explicitly imperative,
sanctioned by an authority capable of imposing respect by an exact system of
compulsions and penalties. In practice, such precepts can govern the most
diversified activities. They are never the sole means of controlling them: in our
daily conduct, we are constantly complying with moral, professional, or
fashionable codes which often make different demands from those of the code
of the law. Moreover, the frontiers of the latter are constantly fluctuating; and,
obviously, a socially recognized obligation does not change its nature simply by
being inserted in the law, even if it can acquire more or less force or clarity
thereby. Hence, law, in the strict sense of the word, is only the formal covering
of realities which are in themselves too diversified to furnish profitable subject-
matter for a single study. Moreover, it exhausts none of these realities. Take the
family— whether it be a question of the small matrimonial family of today in a
state of perpetual expansion and contraction or of the great medieval house, that
community consolidated by such a lasting network of feelings and interests—
for a true insight into its life, would it ever be sufficient simply to enumerate,
one after the other, the articles of any family law? Some men seem
Historical Analysis 149 to think so—how grievously they were deceived is
sufficiently demonstrated by our inability to retrace the inward evolution of the
French family even today. However, there is certainly something very real in
the notion of juridical fact as distinct from others. This is because, in many
societies at least, the application, and even in great measure the elaboration, of
rules of law have been the particular work of a group of relatively specialized
men. This group, whose members could of course combine their legal role with
other social functions, has been sufficiently autonomous to have its own
traditions and often even the practice of a particular method of reasoning. In
short, it may be that the history of law has no separate existence except as the
history of jurists; but this is not a bad sort of existence for a branch of the
sciences of man. Understood in this sense, the history of law sheds some
glimmers of light upon phenomena which are extremely diversified, yet subject
to a common human activity, and these glimmerings, if necessarily limited in
their scope, are very revealing. A different kind of subject is represented by
what we usually call "human geography/' Here the angle of sight does not come
from a professional mentality, as is the case, however unsuspected, with the
history of law. Nor does it, like religious or economic history, arise from the
specific nature of a human fact, either of the beliefs, emotions, outpourings of
the heart, hopes and fears inspired by the vision of forces transcending
humanity, or of the efforts to satisfy and or-
150 The Historians Craft ganize material needs. The inquiry is focused
upon a type of connection common to a great number of social phenomena.
"Anthropo-geography" studies societies in their relations to their physical
environments: relations which are obviously mutual, since man is constantly
acting upon things at the same time that they are acting upon him. In this case
again, therefore, we have nothing more nor less than a perspective whose
legitimacy is proved by its fruitfulness, but which must be supplemented by
other perspectives to be complete. Such, indeed, is the true function of analysis
in any category of research. Science dissects reality only in order to observe it
better by virtue of a play of converging searchlights whose beams continually
intermingle and interpenetrate each other. Danger threatens only when each
searchlight operator claims to see everything by himself, when each canton of
learning pretends to national sovereignty. Once more, however, we must beware
of postulating any false geometric parallels between the sciences of nature and a
science of man. From the view which I have from my window, each savant
selects his proper subject without troubling himself too much about the whole.
The physicist explains the blue of the sky; the chemist the water of the brook;
the botanist the plants. The task of reassembling the landscape as it appears to
me and excites my imagination, they leave to art, should any painter or poet
wish to undertake it The fact is that the landscape as a unity exists only in my
consciousness. Now, the peculiarity of the scientific
Historical Analysis 151 method, as practiced and justified by the success of
these branches of learning, is that it deliberately abandons the observer in order
to know more about the thing observed. To the natural sciences, the connections
which our mind weaves between tilings appear arbitrary; they deliberately break
them in order to re-establish a diversity which seems to them more authentic.
Even so, however, the organic world poses singularly delicate problems. For
greater convenience, the biologist may indeed study respiration, digestion, or
the motor functions separately; for all that, he is not unaware that there is a
whole person for which he must account. The difficulties of history are of still
another nature. For in the last analysis it is human coi*- sciousness which is the
subject-matter of history. The interrelations, confusions, and infections of
human consciousness are, for history, reality itself. As for homo religiosus,
homo caconomicus, homo politicus, and all that rigmarole of Latinized men, the
list of which we could string out indefinitely, there is grave danger of mistaking
them for something else than they really are: phantoms which are convenient
providing they do not become nuisances. The man of flesh and bone, reuniting
them all simultaneously, is the only real being. Certainly, our minds have
interior partitions which some have been peculiarly adept at raising. Gustave
Lendtre was constantly amazed to find so many excellent fathers of families
among the Terrorists. Even if our great revolutionaries had actually been the
blood-
152 The Historian's Craft thirsty monsters whose portraits so agreeably
titillate a middle-class public, such amazement would none the less betray a
rather limited psychology. How many men lead lives on three or four different
levels, which they wish and sometimes succeed in keeping apart? However, this
is far from denying the deep-seated unity of the ego and the constant
intermingling of its various attitudes. Were Pascal, the mathematician, and
Pascal, the Christian, strangers to each other? Did Francois Rabelais, the
learned physician, and Master Alcofribas of Pantagruelist memory, never cross
paths? Even though the roles played alternately by the same actor seem to
conflict as crudely as the stereotyped characters of a melodrama, it may be that
this antithesis, correctly considered, is only the mask of a deeper solidarity. Men
poked fun at the elegiac Florian, who, it appears, beat his mistresses. Perhaps he
lavished so much sweetness in his verses to console himself for his failure to
employ more of it in his conduct. When the medieval merchant, after spending
the day in violating church commandments on usury and just prices, went off to
kneel sanctimoniously before the image of Our Lady, or when in the evening of
his life he heaped up pious charitable endowments; when the great manufacturer
of a sterner age built hospitals with money saved out of the wages of ragged
children, were either of them seeking, as is usually said, only to obtain a rather
cheap insurance against heavenly wrath, or were they not rather, by these
outbursts of faith or phi- lanthropy, also satisfying, almost without conscious
Historical Analysis 153 recognition, those secret needs of the heart which
harsh daily routine had forced them to repress? There are contradictions which
closely resemble evasions. Let us pass from the individual to society. To say
that the latter is simply the sum of individual minds would be to say too little,
but, all things considered, it is at least their product; and we should not,
therefore, be surprised to discover the same play of continuous interaction. It is
an established fact that from the twelfth century until at least the Reformation
the communities of textile workers were one of the favorite breeding-grounds of
heresies. That is surely a worthwhile matter for a note on religious history. Let
us then file this little card carefully in its drawer. And let us throw some more
notes into the neighboring pigeonhole, the one marked "economic history/' Are
we to believe that, by this means, we have finished with these heretical
weavers? We have still to explain them, since one of their fundamental
characteristics was not merely that they made their religious and economic lives
coexist, but that they blended them together. Lucien Febvre is impressed by the
"certainty, or security, of moral tone" which the several generations just prior to
our own apparently enjoyed so fully. He finds two principal reasons for it: the
dominance over men's minds of the cosmological system of Laplace, and the
"abnormal stability" of money. No two human facts are apparently more unlike.
Nevertheless, they worked together to give a society its characteristic mental
attitude.
154 The Historian's Craft Certainly such relationships are just as complex
on the collective scale as in the individual mind. Today we should no longer
dare to say without qualification that literature is "the expression of society/'
That is by no means true—at least not in the sense that a mirror is the
"expression" of the object which it reflects. Literature may as easily express a
defense reaction to its society as an acceptance of it. Almost inevitably, it
carries along a great number of inherited themes, of formal devices learned in
the study, of outworn aesthetic conventions, which act as so many causes of
retardation. As H. Focillon sagely observes: "At any given date, the political,
the economic and the artistic do not occupy [I should have preferred "do not
necessarily occupy"] the same position on their respective curves." Because of
these disparities social life maintains a rhythm which is nearly always uneven.
In the same way, with the majority of individuals, the various psyches, to speak
the pluralist language of old-fashioned psychology, are seldom of an identical
age. How many mature men still preserve streaks of childishness! In 1837,
Michelet explained to Sainte-Beuve: "If I had introduced only political history
into my narrative, if I had taken no account of the diverse elements of history
(religion, law, geography, literature, art, etc.), my procedure would have been
quite different But a great vitd movement was needed, because all these diverse
elements gravitated together in the unity of the story." A generation later, Fustel
de Coulanges, in his turn, announced to his listeners in the Sorbonne:
Historical Analysis 155 "Supposing a hundred specialists had divided the
past of France according to lot, do you think that, in the end, they would have
written the history of France? I very much doubt it. At the very least, they
should miss the linkage of facts: now, this linkage is itself a historical truth'*
The contrast of the images is significant Michelet thought and felt in terms of
the organic; Fustel, son of an age for which the Newtonian universe seemed to
furnish the ultimate scientific pattern, took his metaphors from space. Their
fundamental agreement is all the more impressive. These two great historians
were too great to overlook the fact that a civilization, like a person, is no
mechanically arranged game of solitaire; the knowledge of fragments, studied
by turns, each for its own sake, will never produce the knowledge of the whole;
it will not even produce that of the fragments themselves. But the work of
reintegration can come only after analysis. Better still, it is only the continuation
of analysis, and its ultimate justification. In the scene as we first perceive it,
which we contemplate rather than observe, and in which therefore nothing is
distinct, we cannot discern the interrelationships. Their delicate network
becomes visible only after the elements have been classified. Moreover, in order
to remain true to life in its intertwining actions and reactions, we need not
pretend to seize it as a whole. Such an attempt is too vast for the powers of a
single scholar. Nothing is more legitimate, nothing is more salutary than to
center the study of a society upon one of its particular as-
156 The Historian's Craft p'ects, or, better still, upon one of the well-
defined problems underlying one or another of these aspects: the beliefs, the
economy, the class or group structure, the political crises. . . . By this systematic
selection, not only will the problems usually be more concretely stated, but even
the facts of connection and interchange will stand out with greater clarity.
Provided only that we want to discover them. Do you expect really to know the
great merchants of Renaissance Europe, vendors of cloth or spices, monopolists
in copper, mercury, or alum, bankers of kings and the Emperor, by knowing
their merchandise alone? Bear in mind that they were painted by Holbein, that
they read Erasmus or Luther. To understand the attitude of the medieval vassal
to his seigneur you must inform yourself about his attitude toward his God as
well. The historian never escapes from time. But, in an inevitable oscillation,
already treated above in the discussion of origins, he sometimes considers the
great waves of related phenomena which run over long periods, and sometimes
the specific moments in which these currents are channeled into the powerful
vortex of direct experience. 3. Nomenclature Nevertheless, it would be trivial to
confine ourselves to distinguishing the main aspects of the activities of either a
person or a society. Within each of these broad groups of facts, a new and more
delicate effort to analysis is necessary. We must distinguish the various
Historical Analysis 1ST institutions which compose a political system, the
various beliefs, practices, and emotions which make up a religion. Within these
fragments and in the whole itself, we must characterize those features in which
they resemble, or in which they differ from, other realities of the same
category . . . all problems of classification which are inseparable, in practice,
from the fundamental problem of nomenclature. For the first tool needed by any
analysis is an appropriate language; a language capable of describing the precise
outlines of the facts, while preserving the necessary flexibility to adapt itself to
further discoveries and, above all, a language which is neither vacillating nor
ambiguous. Now, there is where the shoe pinches us historians. One keen-
minded writer with no particular love for us has seen this clearly: "The great
day of definitions, of distinct and special terms, to replace those of confused or
merely statistical origin, has not yet arrived for history." So declares M. Paul
Val&y. But if this day of precision has not yet arrived, is it not possible that it
may yet come? And why, to begin with, is it so slow in coming? Chemistry has
fashioned its own supply of symbols, and even its own words. "Gas," if I am not
mistaken, is one of the few genuinely invented words belonging to the French
language. That is because chemistry had the great advantage of being applied to
realities which were, by their very nature, incapable of naming themselves. The
language of confused perception which it
158 The Historian's Craft has rejected is no less removed from its objects
and, in this sense, no less arbitrary than that of the classified and controlled
observation which it has substituted for it: whether we call it vitriol or sulfuric
acid, the substance itself has not influenced the choice. It is quite otherwise with
a science of humanity. Men gave names to their actions, their beliefs, and the
various aspects of their social life without waiting until they became objects of
disinterested research. Hence, history receives its vocubulary, for the most part,
from the very subject-matter of its study. It accepts it, already worn out and
deformed by long usage; frequently, moreover, ambiguous from the very
beginning, like any system of expression which has not derived from the
rigorously organized efforts of technical experts. The worst is that these
borrowings themselves lack unity. The documents tend to impose their own
nomenclature; if he harkens to them, the historian is taking dictation from an
epoch which is each time different. But of course he is thinking according to the
categories of his own time, consequently with its words. Should we speak of
patricians, a contemporary of the venerable Cato would understand us. On the
other hand, should an author speak of the role of "the bourgeoisie" in the crises
of the Roman Empire, how would he translate either the word or the idea into
Latin? Thus, two distinct orientations almost necessarily divide the language of
history. Let us examine each in its turn.
Historical Analysis 159 To reproduce or copy the terminology of the past
might, at first sight, seem a rather safe course. In application, however, it would
encounter manifold difficulties. In the first place, changes in things do not by
any means always entail similar changes in their names. Such is the natural
consequence of the traditionalist character of all language, and of the lack of
inventiveness common to most men. The observation is valid even for
utilitarian appliances, despite the fact that these are ordinarily subject to rather
clearcut changes in form and construction. If my neighbor tells me that he is
going out in his coup6, or his limousine, am I to understand that he is referring
to a horse-drawn carriage, or to an automobile? Only my prior knowledge of his
coach-house or his garage will enable me to answer. As a rule, aratrum meant a
plow without wheels; carruca, a plow with wheels. However, since the first
appeared before the second, can I be certain, if I find the old word in a text, that
it has not simply been carried over to a new implement? Inversely, Mathieu de
Dombasle called the implement which he had invented a charrue, although,
since it had no wheels, it was actually an araire. Yet this attachment to inherited
names appears much stronger as soon as we consider realities of a less material
order. That is because the transformations in such cases almost always take
place too slowly to be perceptible to the very men affected by them. They feel
no need to change the label, because the change of content escapes them. The
Latin word servus, which has
160 The Historian's Craft given the word serf to the French language, has
come down through the centuries. But it has done so at the expense of so many
successive alterations in the conditions so designated, that the differences
between the serms of ancient Rome and the serf of the France of St. Louis far
outnumber the similarities. Hence, historians have generally chosen to reserve
"serf" for the Middle Ages. For antiquity, they say "slave." In other words,
under the circumstances, they prefer the equivalent to the carbon copy. They do
so not without a sacrifice of propriety to exactitude; for the term which they
thereby transplant into a Roman environment did not come into existence until
about the year 1000, to describe the markets of human flesh where captive Slavs
seemed to provide the very model of a complete subjection which had become
entirely unknown to the indigenous serfs of the west. The device is useful, as
long as we confine ourselves to extremes. In the intervening gap, where must
the slave give way to the serf? It is the eternal sophism of the shock of corn. At
any rate, in order to do justice to the facts themselves, we are here forced to
substitute for the language of the past a nomenclature which, if it is not strictly
invented, is at least reshaped and shifted about. Conversely, moreover, the
names sometimes vary according to time or place, independently of any
variation in the things themselves. Sometimes, there are causes, peculiar to the
evolution of language, which may lead to the obliteration of the word without
the object or the action being in the
Historical Analysis 161 least affected. For linguistic facts have their own
coefficient of resistance or of malleability. By establishing the disappearance
from the Romance languages of the Latin verb emere (to buy) and its
replacement by other verbs of very different origins—deleter, comprar, etc.— a
scholar once thought himself able to prove the most extensive and ingenious
conclusions as to transformations in the commercial system of the societies heir
to Rome. If only he had asked himself whether this indisputable fact could be
treated in isolation! In the languages deriving from Latin, nothing was more
common than the dropping of short-sounding words; the anemia of unstressed
syllables gradually rendered them almost indistinct. It is a phenomenon of a
strictly phonetic nature, and it is a laughable error to have mistaken a vagary of
pronunciation for a feature of economic development. Elsewhere, there are
social conditions which resist the establishment or maintenance of a uniform
vocabulary. In extremely split-up societies, like those of our Middle Ages,
fundamentally identical institutions were frequently designated by very
different terms, depending upon the locality. Even in our own day, rural dialects
differ widely from each other, even in designating the most common objects
and the most universal customs. In the central region where I write these lines,
they say village for what, in the north, would be a hammeau. The village of the
north is here a bourg. These verbal deviations are, in themselves, facts worth
due consideration. However, should the his-
162 The Historian's Craft torian conform his own terminology to them, he
would not only compromise the intelligibility of his language but he would also
deny himself even the work of classification, which is his first duty. Unlike
mathematics or chemistry, our science has at its disposal no system of symbols
unconnected with national language. The historian speaks only with words;
hence, with those of his country. If he finds realities expressed in a foreign
language, he must translate. As for that, there is no serious obstacle as long as
the words refer to commonplace things or actions: that ready money oi the
vocabulary is easily exchanged at par. On the other hand, as soon as those
institutions, beliefs, and customs which play a profounder part in the peculiar
life of a society make their appearance, the translation into another language,
made after the likeness of a different society, becomes an enterprise fraught
with dangers. For to choose an equivalent is to postulate a resemblance. Shall
we, then, in despair of the case, resign ourselves to keeping the original term,
with appropriate explanations? Assuredly, that must sometimes be the proper
course. In 1919, when they saw that the Weimar constitution kept the ancient
name of Reich for the German state, certain of our publicists loudly protested:
"A curious republic which still calls itself an empirel" The truth is not only that
Reich does not, of itself, imply the idea of an emperor, but that, by its
association with a political history constantly oscillat-
Historical Analysis 163 ing between particularism and unity, the word has
a ring far too specifically German to suffer the least attempt at translation into a
language which reflects an entirely different national past But such mechanical
adoption of foreign words, obviously the easiest solution, cannot be made a
general rule. Even putting aside any concern for propriety of language, it would
still be annoying to see historians encumbering their observations with foreign
expressions like those authors of rustic novels who, by their use of
provincialisms, slip into a jargon which neither town nor country would
recognize as its own. To renounce any attempt at equivalence is often to do
injury to reality itself. A custom which derives, I believe, from the eighteenth
century permits the use of the French word serf, or of closely related words in
other Western languages, to designate the Krepostnoi of tsarist Russia. A more
unfortunate comparison would be difficult to imagine. In Russia there was a
system of attachment to the land gradually transformed into true slavery; with
us, a form of personal dependence which, despite its severity, was very far from
treating man as a thing deprived of all rights: the so-called Russian serfdom had
almost nothing in common with our medieval serfdom. Simply to say
"Krepostnoi," however, will hardly help our case. For, in Rumania, in Hungary,
in Poland, and even in eastern Germany, there have been types of peasant
subjection closely related to that established in Russia. Must we speak
Rumanian, Hungarian, Polish, German, and Russian by turns? Once
164 The Historian's Craft more, we should miss the essential point, which
is to map the underlying connections between the facts by expressing them with
an accurate nomenclature. The label has been badly chosen. But that does not
obviate the necessity of a common label, and one which is, therefore,
superimposed upon national words, instead of merely copying them. Here
again, passivity is forbidden. Numerous societies have practiced what may be
called a hierarchic bilingualism. Two languages are side by side, the one
popular, the other learned. What is generally thought and spoken in the first is
written, either exclusively or by preference, in the second. Thus, from the
eleventh to the seventeenth century the Abys- sinians wrote Gueze, but spoke
Amharic. Thus, the Evangelists reported in Greek, which was then the great
language of Eastern culture, conversations which we must assume to have been
originally exchanged in Aramaic. Thus, more recently, the Middle Ages for
long conducted its administration and wrote its narratives only in Latin.
Inherited from dead or borrowed from foreign civilizations, these languages of
the lettered, the priests, and the scribes came of necessity to express a great
many realities for which they had not originally been framed- They succeeded
in this only with the help of a whole system of transpositions, which were
inevitably rather awkward. Now—material evidences excepted—it is by means
Historical Analysis 165 of these writings that we come to know a society.
Hence, those societies in which such a dualism of languages prevails appear to
us in many of their principal characteristics only through a sort of veil.
Sometimes, moreover, a supplementary screen intervenes. The great land-
register instituted by William the Conqueror, the Doomsday Book, was the
work of clerks of Normandy or Maine. Not only did they describe in Latin
institutions which were peculiarly English, but they had first thought them over
in French. When the historian runs into this sort of nomenclature by
substitution, he has no other recourse than to do the work over in reverse. If the
corresponding terms were suitably chosen and, above all, consistently applied,
the task will be relatively easy. It will not be very difficult to recognize the
counts of real life behind the "consuls" of the chroniclers. Unfortunately, there
are less favorable cases to be met with. What was the colonus of our eleventh
and twelfth century charters? It is a meaningless question. With no derivative in
the language of the people, because it had ceased to apply to any living idea, the
word represented a mere trick of translation used by the notaries to describe, in
fine classic Latin, a whole series of very different judicial and economic
conditions. At any rate, this opposition of two necessarily different languages
actually typifies only an extreme instance of contrasts common to all societies.
Even within the most unified nations, such as ours, each little professional
community, each group distinguished by
166 The Historian's Craft virtue of its culture or wealth, has its own
characteristic form of expression. Now, not all groups write, or write as much or
have as much chance of passing their writings down to posterity. Everyone
knows that the official reports of a judicial examination seldom reproduce the
words just as they were spoken; almost spontaneously, the clerk of the court
orders, clarifies, restores the syntax, and weeds out the words which he has
judged too vulgar. The civilizations of the past have also had their clerks; it is
the voice of chroniclers and, especially, jurists which has come through to us
before all others. We must beware of forgetting that the words which they used,
and the classifications which they suggested by these words, were the result of a
learned elaboration often unduly influenced by tradition. What a shock it might
be if, instead of poring laboriously over the jumbled—and probably artificial—
terminology of the Carolingian manorial scrolls and capitularies, we were able
to take a walk through a village of that time, overhearing the peasants
discussing their status amongst themselves, or the seigneurs describing that of
their dependents. Doubtless this description of daily usage would fail of itself to
give us a total picture of life, for the attempts at expression and, hence, at
interpretation by scholars and men of the law also embody really effective
forces; but it would at least give us the underlying feeling. What an education it
would be—whether as to the God of yesterday or today—were we able to hear
the true prayers on the lips of the humble! Assuming, of course, that they them-
Historical Analysis 167 selves knew how to express the impulses of their
hearts without mutilating them. For there, in the last analysis, is the great
obstacle. Nothing is more difficult for us than self-expression. But we
experience scarcely less difficulty in finding names, free from both ambiguity
and false precision, to express the fluid social realities in which we have our
very being. The most usual terms are never more than approximations. These
include even the religious terms which one might easily imagine to have a
precise meaning. If you examine the religious situation of France, you will see
how many subtle distinctions a learned man like M. Le Bras is today forced to
substitute for that oversimplified label: "Catholic." This is food for thought for
those historians who, from atop the eminence of their belief (sometimes, and
perhaps more frequently, of their unbelief) are inflexibly dogmatic about the
Catholicism of an Erasmus. Certain other very vital realities have failed to find
suitable words. A laborer in our day finds it easy to speak of his class
consciousness, even though it may be rather deficient I doubt that this sentiment
of conscious and aroused solidarity has ever manifested itself with more force
or clarity than among the agricultural laborers of northern France at the end of
the Ancien Regime; various petitions, certain memorials of 1789 have preserved
its poignant echoes for us. Nevertheless, their sentiment could not then have
been named, because there was still no name for it.
168 The Historian's Craft To sum it all up in a word, the vocabulary of
documents is, in its way, only another form of evidence. It is, no doubt, an
extremely valuable one, but, like all evidences, imperfect and hence subject to
criticism. Each significant term, each characteristic turn of style becomes a true
component of knowledge—but not until it has been placed in its context, related
to the usage of the epoch, of the society or of the author; and above all, if it is a
survival of ancient date, secured from the ever-present danger of an
anachronistic misinterpretation. Royal unction in the twelfth century was treated
as a sacrament, and the term "sacrament*' was assuredly fraught with
significance, but it lacked, at that time, the far greater weight which theology
would assign to it today, having become more inflexible in its definitions and,
consequently, in its vocabulary. The advent of the name is always a great event
even though the thing named has preceded it; for it signifies the decisive
moment of conscious awareness. What a forward stride was taken the day the
initiates of a new faith first called themselves Christians! Certain of our elders,
like Fustel de Coulanges, have given us admirable examples of this study of
meanings, of this '"historical semantics." Since their time, the progress of
linguistics has further sharpened the tool. May young scholars never grow
weary of handling it and, especially, of extending its use into the most recent
times, which, in this regard, are much the least well explored. Certainly, in spite
of everything, the names, however
Historical Analysis 169 imperfect their over-all accuracy, have far too
strong a grip upon reality ever to permit us to describe a society without making
a considerable use of its words, duly explained and interpreted. We shall not
imitate those everlasting translators of the Middle Ages. We shall say "counts"
where it is a question of counts, and "consuls" where ancient Rome is the
setting. Great progress was made in the understanding of Hellenic religions as
soon as Zeus had definitely banished Jupiter from the lips of scholars. But this
practice is particularly applicable to institutional, technological, or religious
detail. To consider that the nomenclature of the documents was perfectly
capable of determining our own would, in short, be tantamount to admitting that
they had provided us with a ready-made analysis. Were that the case, history
would have little left to do. Happily, for our sake, it is not. That is why we are
forced to seek elsewhere for the broad framework of our classification. To
provide it, we already have at our disposal a whole lexicon which seeks to
transcend the connotations of any particular period. Elaborated without
predetermined plan by the successive modifications of several generations of
historians, it brings together elements of very diverse date and origin. "Feudal"
and "feudalism" were originally legal jargon, taken over from the courts of the
eighteenth century by Boulain- villiers, and then by Montesquieu, to become the
rather awkward labels for a type of social structure
170 The Historian's Craft which was itself rather ill-defined. "Capital" was
a usurer's and accountant's word whose meaning the early economists greatly
expanded. "Capitalist" is a remnant of the language of speculators in the earliest
European stock-exchanges. But "capitalism," which today occupies a far more
considerable place in our classics, is altogether new: its ending shows its origin
— Kapitalismus. "Revolution" has changed its former astrological associations
for a very human meaning; in the heavens, it was and still is a regular motion,
forever turning back upon itself; on earth, a sharp crisis always aimed straight
ahead. "Proletariat" is a word of antique style, like die men of '89 who first
made its fortune, following Rousseau: but Marx, following Babeuf, has set his
stamp on it forever. The very savages of America have given us "totem," and
those of Oceania, "taboo": ethnographic adaptations before which the classicism
of certain historians still hesitates. Neither this variety of origins, nor these
deviations of meaning are an inconvenience. A word is valued much less for its
etymology than for the use to which it is put. If "capitalism," even in its
broadest application, is far from embracing all those economic systems in which
the capital of moneylenders has played a role; if "feudal" currently serves to
characterize societies in which the fief was certainly not the most significant
feature, that in no way contradicts the universal practice of all sciences, which
as soon as they are no longer content with pure algebraic symbols are obliged to
draw upon the confused vocabulary of daily life. Are
Historical Analysis 171 we scandalized that the physicist persists in using
the term "atom," meaning indivisible, for what is actually the object of his most
daring dissections? Much more dangerous are those emotional overtones with
which so many of these words are charged as they come to us. Strong feeling is
seldom favorable to precision of language. Even among historians, custom
tends to confuse the two expressions, "feudal system" and "seigneurial system/'
in the most troublesome manner. This is arbitrarily to equate the complex of
dependent ties characteristic of a warrior aristocracy with a type of peasant
subjection which not only was very different by nature but had arisen very
much earlier, lasted much longer, and was far more widespread throughout the
world. The misunderstanding dates back to the eighteenth century. At that time,
vassalage and the fief still existed, but only as a legal form which had been
virtually devoid of meaning for several centuries. The seigneury, on the other
hand, although descended from the same past, remained very much alive. The
political writers had made no distinctions within this heritage. It was not only
that they misunderstood it. For the most part, they did not consider it
dispassionately. They detested it for its anachronisms and, even more, for the
oppressive forces which it persistently embodied. A common condemnation
enveloped the whole. Then, at the same time and under a single name, the
Revolution abolished the seigneury along with those institutions which were
properly feudal. All that remained was a
172 The Historian's Craft memory, but a persistent one, which descriptions
of the strife of the last days painted in vivid colors. Henceforth, the confusion
was established. Born out of passion, it continued ever ready to be spread under
the stimulus of new passions. Even today, are we entirely dispassionate when
we speak of industrial or financial "feudalisms"? In the background there is
always a reflection of the firing of chateaux during the burning summer of '89.
Now, unfortunately, such is the fate of a great many of our words. They
continue to live among us the unquiet life of public disputation. It is not the
historians who nowadays harangue us to consider capitalism and communism as
identical. Our symbols are variable according to time or place; they become
coefficients of emotivity leading to further equivocation. The reactionaries of
1815 hid their faces in horror at the very name of revolution. Those of 1940
used it to camouflage their coup d'etat However, let us assume that our
vocabulary had ultimately arrived at impassiveness. Even the most intellectual
of languages have their pitfalls. Certainly, we here feel not the slightest
temptation to republish those "nominalist pleasantries" of which Robert Simi-
and once observed, with good reason, that the sciences of man had a kind of
monopoly. What law denies us those facilities of language which are
indispensable for any rational knowledge? If, for example, we speak of the
factory system, we by no means create an entity
Historical Analysis 173 thereby. We merely group facts, as concrete as we
could wish, under an expressive name. The similitude of these facts, which the
name quite properly seeks to signify, is itself a reality. In themselves, therefore,
these terms are entirely legitimate. Their true danger derives from their very
convenience. If ill-chosen or too mechanically applied, the symbol (which was
there only to assist in the analysis) ends by dispensing with analysis. Thereby, it
promotes anachronism: the most unpardonable of sins in a time-science.
Medieval societies distinguished two principal human conditions: there were
men who were free, others who were not considered to be so. But the idea of
liberty is one which each epoch reshapes to its own liking. Therefore, certain
historians nowadays have judged that, in the allegedly normal, which is to say,
in their own acceptation of the word, the unfree men of the Middle Ages had
been badly designated. They were, they say, "half-free." As a word invented
without any textual authority, this intruder would be awkward in any case.
Unfortunately, it is worse than that. As a nearly inevitable consequence, such
false precision seems to have made superfluous any extensive research on the
line between liberty and servitude, as these civilizations understood the idea: a
line which was often uncertain and even variable according to the bias of the
time or of the class, but one of whose essential characteristics was precisely that
of having never allowed that marginal zone which the name of half- liberty
suggests with such tiresome persistence. A no-
174 The Historian's Craft menclature which is thrust upon the past will
always end by distorting it, whether by design or simply as a consequence of
equating its categories with our own, raised, for the moment, to the level of the
eternal. There is no reasonable attitude toward such labels except to eliminate
them. "Capitalism" has been a useful word. Doubtless, it will again become so
when we have succeeded in cleansing it of those ambiguities with which its
passage into common use has increasingly burdened it At the moment,
carelessly applied to the most diverse civilizations, it almost inevitably results
in concealing their original features. Was the economic system of the sixteenth
century "capitalist"? Perhaps. However, consider that sort of universal
discovery of the profit motive which was then trickling from the top of society
to the bottom, catching up the shopkeeper or village notary as well as the great
banker of Augsburg or Lyons; observe that the emphasis was placed upon loans
or commercial speculations far earlier than upon the organization of production:
how different then is this "capitalism" of the Renaissance, in its human
structure, from the far more hierarchically organized system of modem
manufacturing or the St. Simonian system of the era of the Industrial
Revolution. Which, in its turn . . . In any case, one very simple observation
should suffice to put us on our guard. If we think of capitalism, no longer as the
capitalism of one definite period, but capitalism in and of itself, Capitalism with
a capital
Historical Analysis 175 "C," what date shall we assign to its appearance?
The twelfth century in Italy? The thirteenth in Flanders? At the time of the
Fuggers and the exchange at Antwerp? The eighteenth century, or even the
nineteenth? There are as many birth-certificates as there are historians. Indeed,
they are almost as numerous as for that bourgeoisie whose rise to power is so
celebrated in our textbooks, and which is variously stated, for the edification of
schoolboys, to have occurred in the reign of Philip the Fair, or in that of Louis
XIV, unless it was in 1789, or in 1830. Could it be, after all, that it was not the
same bourgeoisie? Or not the same capitalism? And here, I believe, we strike
the root of the matter. We are reminded of the neat phrase of Fontenelle:
"Leibniz," he remarked, "laid down exact definitions, which deprived him of the
agreeable liberty to misuse his terms upon occasion." Agreeable perhaps;
certainly dangerous. It is a liberty with which we are all too familiar. The
historian seldom defines. He might well consider this an unnecessary
precaution, if he were borrowing from a usage which was itself strictly defined.
Since such is not the case, almost his only guide, even in the use of his key
words, is personal instinct He arbitrarily expands, restricts, distorts the
meanings —without warning his reader; without always fully realizing it
himself. What of the "feudalisms" throughout the world from China to the
Greece of the beautifully greaved Achaeans? For the most part, they bear
176 The Historian's Craft scarcely any resemblance to each other. That is
because nearly every historian understands the word as he pleases. However,
even if we do define, it is usually every man for himself. Nothing is more
significant than the case of an economic analyst as penetrating as John Maynard
Keynes. There is hardly one of his books in which he does not, from the
beginning, expropriate terms, usually pretty well established, in order to decree
entirely new meanings for them, meanings which sometimes vary from work to
work, but, in any case, intentionally depart from common usage. Curious whim
of the sciences of man, which, after long being ranked among the "Belles-
lettres" seem still to preserve something of the stubborn individualism of art!
Can we imagine a chemist saying: "Two elements are necessary to make a
molecule of water: two atoms of one, one of the other; in my vocabulary, the
former is to be called oxygen and the latter, hydrogen"? However well defined,
the private languages of historians will never, laid side by side, constitute the
language of history. To tell the truth, some better-planned efforts have been
attempted here and there, by groups of specialists (linguists, ethnographers,
geographers) whom the relative youth of their disciplines seems to have
protected from the worst corporate habits. The task has also been undertaken for
history as a whole by the Center of Synthesis, always alert to render services
and provide examples. We should expect a great deal from them, but even more
from a general diffusion of good will.
Historical Analysis 177 No doubt the day will come when a series of
understandings will permit us to clarify nomenclature, and then to define it
progressively. Even then, the individuality of the scholar will, as always, be
reflected in his choice of words, unless he is content to limp along from one
date to the next, like a mere writer of annals. The great epochs were marked out
by the dominations of conquering peoples who successively destroyed each
other. Thus, the collective memory of the Middle Ages subsisted almost entirely
upon the Biblical myth of the Four Empires: Assyrian, Persian, Greek, and
Roman. An awkward scheme, if ever there was one. Not only did it, by
compliance with the sacred text, prolong the illusion of a fictitious Roman unity
but, by a curious paradox for a Christian society—as it must appear to any
historian today—it made the Passion seem a less notable stage in the progress of
humanity than the victories of celebrated spoilers of provinces. Within each
nation, the succession of kings furnished the boundaries for the smaller
divisions. These habits have proven remarkably tenacious. Uhistoire de France,
a faithful mirror of the French school about 1900, still proceeds by stumbling
from reign to reign; at each prince's death, which it narrates with all the detail
which is reserved for great events, it calls a halt. When there are no longer any
kings, there are governmental regimes which are also, fortunately, mortal: hence
their revolutions serve as landmarks- More recently, there has been an
important collection of textbooks which segments the course of modern his-
178 The Historian's Craft tory according to national "preponderances"—the
sugar-coated equivalents of the "empires" of former times. It is hardly necessary
to say that the Spanish, French, or English "hegemonies" are diplomatic and
military by nature. All else one arrays as one can. Yet long ago the eighteenth
century made its protest heard. "It seems," wrote Voltaire, "that, for fourteen
hundred years, there have been none but kings, ministers, and generals in the
Gauls." As a consequence, there gradually appeared new divisions which, free
from the imperialist or monarchical obsession, could be ordered according to
profounder phenomena. As we have seen, "feudalism," as the name of a period
as well as of a social system, dates from this time. But the career of the term
"Middle Ages" is the most instructive of all. In its remote origin, it was itself
medieval. It belonged to the vocabulary of that semi-heretical prophecy which,
from the thirteenth century especially, had captivated so many,troubled spirits.
The Incarnation had put an end to the Old Law. It had not yet established the
Kingdom of God. Striving toward that blessed day, the present time was,
therefore, only an intermediate age: a medium cevum. Then, with the early
humanists, to whom that mystical language was apparently still familiar, the
idea was misappropriated for more profane realities. In a sense, the reign of the
Spirit had arrived. It was that "restoration" of thought and of letters which was
then making such a vivid im-
Historical Analysis 179 pression upon the best minds: such, for example,
as Rabelais and Ronsard. The "Middle Age" was now closed, having
represented only a prolonged waiting between a fruitful Antiquity and a new
Revelation. Thus understood, the expression had a shadowy existence for
several generations, confined, no doubt, to a few erudite circles. It was, it is
believed, just at the close of the seventeenth century that a German, an
unassuming writer of manuals, Christopher Keller, in a work of general history,
thought of labeling as the "Middle Age" the whole period of more than a
thousand years which extended from the Invasions to the Renaissance. By
whatever channels it was introduced, the usage became firmly established in
European and especially in French historiography about the time of Guizot and
Michelet. Voltaire had not known of it: "You wish ultimately to overcome the
disgust you feel at Modern history since the decline of the Roman Empire." We
here recognize the first sentence of the Essai sur les Mceurs. Nevertheless, there
can be no question that the true sense of the Essai, which was so influential
upon the succeeding generations, was responsible for the success of the term
"Middle Age," as well as that of its almost inevitable counterpart,
"Renaissance." Although the latter had been current in the vocabulary of the
history of the fine arts, as a common noun, with the necessary addition of a
complement (as in "the renaissance of the arts or letters under Leo X or Francis
I"), it was not much before Michelet that, along with the capital
180 The Historian's Craft letter, it won the distinction of being used by
itself to designate the entire period. For both periods the idea is the same.
Formerly, battles, court politics, the rise or fall of great dynasties had furnished
the general framework within which art, literature, and the sciences were fitted
more or less badly. Now it was to be the reverse. It is the most refined
manifestations of the human spirit which, by their varying progress, have set the
tone of the historical epochs. There is no idea which bears the Voltairian stamp
more clearly than this. But a serious weakness invalidated this classification of
Middle Ages and Renaissance: the distinguishing feature also implied a
judgment. "Europe, squeezed between sacerdotal tyranny and military
despotism, waits amidst blood and tears for the moment when the new
enlightenment will enable it to rise again to liberty, humanity, and virtue." Thus
Condorcet described the period to which unanimous consent was soon to assign
the name of Middle Age. But as soon as we no longer believe in that "night," as
soon as we no longer picture as a uniformly barren waste the centuries which
were so rich in the fields of technical invention, art, feeling, and religious
reflection, which saw the first expansion of the European economy and the
beginning of European nationalism, what reason can we have for confusing
under one fallacious generalization the Gaul of Clovis with the France of Philip
the Fair, Alcuin with St. Thomas or Ockham, the animalistic "barbarian"
jewelry with the statues
Historical Analysis 181 of Chartres, the cramped towns of Carolingian
times with the flourishing bourgeoisie of Genoa, Bruges, or Liibeck. In truth,
the term "Middle Age" has no more than a humble pedagogical function, as a
debatable convenience for school curriculums, or as a label for erudite
techniques whose scope is moreover ill-defined by the traditional dates. A
medievalist is a man who knows how to read old scripts, to criticize a charter, to
understand Old French. Unquestionably, that is something. It is certainly not
enough to satisfy a real science in its search for accurate periodization. Into the
confusion of our chronological classifications there has stolen a fashion which
is rather recent, I believe, but is all the more insidious because it has no rational
basis. We tend to count by centuries. For long disassociated with any exact
enumeration of years, this word also originally had its mystic overtones—its
accents of the Fourth Eclogue or of the Dies Irce.1 Perhaps these had not
completely died away at the time when history, with no great concern for
numerical precision, lingered complacently over the "century of Pericles/' or
that of "Louis XIV." But our language has become more rigorously
mathematical We no longer name ages after their heroes. We very prudently
number them in sequence every hundred 1 [Translator's note: It seems necessary
to point out that the French for "century" is stecle, which also means "age" or
"era** of "temporal world," in which sense the Latin saeculum occurs in
VergjFs Fourth Eclogue and in the medieval Dies lis.]
182 The Historian's Craft years, starting from a point fixed, once and for
all, at the year 1 of the Christian era. The art of the thirteenth century, the
philosophy of the eighteenth, the "stupid nineteenth"; these faces in arithmetical
masks haunt the pages of our books. Which of us will boast of having never
fallen prey to the lures of their apparent convenience? Unfortunately, no law of
histoiy enjoins that only those years whose dates end with the figures "01"
coincide with the critical points of human evolution. Whence there derive some
curious distortions of meaning. "It is well known that the eighteenth century
begins in 1715 and ends in 1789." Not long ago I read this sentence in a
student's paper. Whether it was naive or sly I do not know. In any case, it
exposed certain singularities of the usage rather neatly. But, were it a question
of the philosophical eighteenth century, we might certainly prefer to say that it
began well before 1701: the Histoire des Oracles appeared in 1687, and Bayle's
Dictionnaire in 1697. The worst of it is that since the name, as always, carries
the idea along with it, these false labels end by misrepresenting the
merchandise. The medievalists speak of the "Renaissance of the twelfth
century." Certainly, there was then a great intellectual movement. However, if
this is what we call it, we are too apt to forget that it actually began about 1060,
and hence to miss certain essential connections. In a word, we appear to assign
an arbitrarily chosen and strictly pendulumlike rhythm to realities to which such
regularity is entirely alien. It is
Historical Analysts 183 an impossible task. Naturally, we do very badly at
it We must look for something better. The long and the short of it is that, as long
as we confine ourselves to studying sequences or phenomena in time, the
problem is simple. We should look to the phenomena themselves for their
proper periods. A religious history of the reign of Philip Augustus? An
economic history of the reign of Louis XV? Why not: "Journal of what
happened in my laboratory during the second presidency of Gr6vy," by Louis
Pasteur? Or, inversely: "Diplomatic history of Europe from Newton to
Einstein"? Of course, we readily understand the potential charms of divisions
regularly arranged according to empires, kings, or political regimes. Not only
do they have the prestige which a long tradition associates with the exercise of
power, "with those deeds/' as Machia- velli puts it, "which have the air of
grandeur proper to acts of government or of the state'7; but an accession, a
revolution, has its chronological position determined to a year and even to a
day. Now, the scholar loves close dating. He finds it both an appeasement to his
instinctive horror of the vague and a great comfort to the conscience. He wants
to have read and to have checked everything which concerns his subject. How
much easier this will be if, standing before each file of archives with his
calendar in his hand, he can divide them into categories: before, during, and
after! However, let us beware of worshipping the idol of
184 The Historian's Craft false precision. The most precise measurement is
not necessarily the one which refers to the smallest unit of time—in which case,
we should have to prefer not only a year to a decade, but also a second to a day
—ii is the one which is best adapted to the nature of the events. Now, each type
of phenomenon has its own particular dimension of measurement and, so to
speak, its own specific decimal. Metamorphoses of social structure, economy,
beliefs, or mental attitude cannot conform to an overly precise chronology
without distortion. When I write that a very profound change, marked
simultaneously by the first large-scale imports of overseas wheat and by the
first great expansion of German and American industry, took place in the
economy of the West between about 1875 and 1885, I make use of the nearest
approximation permitted by this kind of fact. A date pretending to be more
exact would falsify the truth. Even in statistics a decennial average is, in itself,
no more crude than an annual or weekly average. It merely expresses another
aspect of reality. Moreover, it is by no means impossible a priori that phases of
phenomena of a seemingly very different order may overlap in experience. Is it
true that the advent of the Second Empire ushered in a new economic era in
France? Was Sombart right in identifying the rise of capitalism with that of the
Protestant spirit? Was M. Thierry-Maulnier right when he found in democracy
"the political expression" of this same capitalism (although not, I fear, quite the
same capi-
Historical Analysis 185 talism)? However dubious such coincidences may
seem to us, we have no right to reject them with closed minds. We simply must
not postulate such connections in advance. Certainly, the tides are related to the
successive phases of the moon. In order to know this, however, it was first
necessary to determine the periods of the tides and those of the moon, quite
apart from one another. When, on the other hand, we consider the evolution of
society as a whole, can we characterize its successive stages? The problem is to
find the dominant note. We can here only suggest the ways in which a
classification might be worked out. Let us not forget that history is a science
still in travail. Men who are born into the same social environment about the
same time necessarily come under analogous influences, particularly in their
formative years. Experience proves that, by comparison with either
considerably older or considerably younger groups, their behavior reveals
certain distinctive characteristics which are ordinarily very clear. This is true
even of their bitterest disagreements. To be excited by the same dispute even on
opposing sides, is still to be alike. This common stamp, deriving from common
age, is what makes a generation. Society, it is true, is not a single thing. It is
split up into different social classes, in which the generations do not always
overlap. Do the forces acting upon a young worker necessarily operate, at least
with equal
188 The Historian's Craft intensity, upon a young peasant? Add to this that,
even in the best-knit civilizations, the currents of dissemination are slow. My
father, born in Strassburg in 1848, used to say: "We were romantics in the
Provinces during my adolescence, when Paris had ceased to be so." Often,
however, as in this case, the contrast is no more than a lag. Consequently, when
we speak of such and such a generation in France, for example, we call forth a
complex and, sometimes, even a contradictory idea— but one in which it is
natural to retain the really dominant elements. Despite the Pythagorean dreams
of certain authors, it is obvious that the periodicity of the generations is by no
means regular. As the rhythm of social change is more or less rapid, the limits
contract or expand. There are, in history, some generations which are long and
some which are short. Only observation enables us to perceive the points at
which the curve changes its direction. At my university the dates of enrollment
made it easy to note one such turning-point. I early found that I was, in many
respects, closer to the classes that had graduated before me than to those which
came almost immediately after me. My classmates and I considered ourselves
as the last of the generation of the Dreyfus Affair. The experience of life has not
contradicted this impression. It is inevitable for the generations to permeate
each other, for individuals do not always react in the same way to the same
influences. Among our children, it is today fairly easy to distinguish, mainly
according to
Historical Analysis 187 ages, between the war and the postwar generations.
Always with this one reservation: in those who have not reached late
adolescence, but are past early childhood, the sensitivity to present events varies
a great deal according to personal temperament; the most precocious will be
truly "of the war"; the others will be left on the opposite banlc, Accordingly,
like any concept which seeks to express, without distortion, the affairs of man,
the notion of one generation is very elastic. It corresponds to realities which we
feel to be very concrete. It has long been used instinctively by those disciplines
whose nature is particularly hostile to the old divisions by reigns or
governments—such as the history of thought or of artistic forces. Increasingly,
it seems destined to provide a first step toward a rational analysis of human
change. But a generation represents only a relatively short phase. Longer phases
are called civilizations. Thanks to Lucien Febvre, we are well acquainted with
the history of the word, which is obviously inseparable from that of the idea.
Only slowly has the idea been disentangled—or, more precisely, disassociated
—from value judgment. We still speak (although, alas, with less assurance than
our elders) of civilization in itself, of civilization as an ideal, and of the difficult
ascent of mankind toward its noble tranquillity; but we speak also of
civilizations in the plural and merely as realities. From this point; we admit that
there may
188 The Historian's Craft be, if I may venture to say so, civilizations of
people who are not civilized. That is because we have come to recognize that
within any society, whatever its nature, everything is mutually controlled and
connected: the political and social structure, the economy, the beliefs, the most
rudimentary as Well as the subtlest manifestations of the mind. What shall we
call this complex, "in the heart of which," as Guizot once wrote, "all the forces
of its existence come together'? Created by the eighteenth century to express an
absolute good, the word "civilization," without losing its former meaning, has
conformed naturally to this new sense of fact, in proportion as the sciences of
man have become more relativist. What was formerly its sole meaning it now
preserves as an echo of human sympathy, whose value is not to be overlooked.
The antitheses of civilizations appeared clearly as soon as the contrasting
features of exotic lands were noted. Will any one deny that there is a Chinese
civilization today, or that it differs greatly from the European? But, even in the
same region, the major emphases of the social complex may be more or less
abruptly modified. When such a transformation has taken place, we say that one
civilization succeeds another. Sometimes there is an external shock, ordinarily
accompanied by the introduction of new human elements, such as between the
Roman Empire and the societies of the high Middle Ages. Sometimes, on the
other hand, there is simply internal change. Everyone will agree that the
civilization of the Renaissance is no
Historical Analysis 189 longer ours, despite the fact that we have derived
such a liberal inheritance from it. These varying tonalities are, no doubt,
difficult to express. They cannot be expressed by summary labels. The
convenience of "ism" words (Typismus, Konventionalismus) ruined even such
an intelligent attempt at evolutionary description as Karl Lamprecht's History of
Germany. The same error was earlier made by Taine, in whose writings we are
astonished nowadays by the almost personal reality of the "dominant
conception." Nevertheless, the fact that certain attempts may have miscarried
does not justify abandoning the effort. It is the business of research to introduce
more accuracy and exactness into these distinctions. To summarize, human time
will never conform to the implacable uniformity or fixed divisions of clock
time. Reality demands that its measurements be suited to the variability of its
rhythm, and that its boundaries have wide marginal zones. It is only by this
plasticity that history can hope to adapt its classifications, as Bergson put it, "to
the very contours of reality": which is properly the ultimate aim of any science.
i 190] CHAPTER V HISTORICAL GAUSATION In vain positivism
claimed to eliminate the idea of cause from science. Whether he likes it or not,
every physicist, every biologist thinks in temicS of "why" and "because." The
historian cannot escape this common law of the mind. Some, like Michelet,
connect matters in a great "living movement," rather than give an explanation in
logical form; others parade their apparatus of inductions and hypotheses; the
genetic link is present throughout. But from the fact that establishment of
relations of cause and effect constitutes an instinctive need of our
understanding, it does not follow that the search can be left to instinct. If the
metaphysics of causality is here beyond our horizon, the use of the causal
relationship as a tool of historical knowledge indisputably demands a conscious
critical treatment. Let us suppose that a man is walking along a mountain path.
He trips and falls off a precipice. For this accident to happen, the combination
of a great number of determining elements was necessary, such as, among
others, the existence of gravity; a terrain resulting from protracted geological
changes; the laying out of
Historical Causation 191 a path for the purpose, let us say, of connecting a
village with its summer pastures. It would, therefore, be perfectly legitimate to
say that, were the laws of celestial mechanics different, had the evolution of the
earth been otherwise, were alpine economy not founded upon the seasonal
migration of flocks, the fall would not have happened. Nevertheless, should we
inquire as to the cause, everyone would answer: "A misstep/' It is not that this
antecedent was most necessary to the occurrence of the event. Many others
were just as necessary. But it was distinguished from all the rest by several very
striking characteristics: it occurred last; it was the least permanent, the most
exceptional in the general order of things; finally, by virtue of this greater
particularity, it seems the antecedent which could have been most easily
avoided. For these reasons, it appears to have exerted a more direct influence
upon the result, and we scarcely can avoid the feeling that it was really the sole
cause of it. From the viewpoint of common sense, which has always been
reluctant to rid itself of a certain anthropomorphism in speaking of cause, this
last-minute component, this specific and unexpected component, is a little like
the artist who gives form to a plastic material which is already completely
prepared. Historical reasoning in contemporary practice does not differ in its
procedure. However necessary they may be, the most constant and general
antecedents remain merely implicit. What military historian would dream of
ranking among the causes of a victory that gravita-
192 The Historian's Craft tion which accounts for the trajectory of the
shells, or the physiological organization of the human body without which the
projectiles would have no fatal consequences? More specific antecedents, if
they have a certain permanence, form what is called, for convenience, "the
conditions." The most specific, the one which somehow represents the
differentiating element in the compound of generative influences, is accorded
the name of cause. We will say, for example, that the inflation of Law's time
was the cause of the over-all rise of prices. The existence in France of a
homogeneous and well-knit economic milieu would be only a condition. For
that ease of circulation which, by distributing notes on all sides, simply made
possible the rise, both preceded the inflation and outlasted it. There can be no
doubt that there is a faithful principle of research in this discrimination. What is
the use of dwelling upon nearly universal antecedents? They are common to too
many phenomena to deserve a special niche in the genealogy of any of them. I
am well aware, from the outset, that there would be no fire if the air contained
no oxygen: what interests me, what demands and justifies an attempt at
discovery, is to determine how the fire started. The laws of trajectories are as
valid for defeat as for victory: they explain both; therefore, they are useless as a
proper explanation for either. However, a graduated classification of causes,
which
Historical Causation 193 is really only an intellectual convenience, cannot
safely be elevated to an absolute. Reality offers us a nearly infinite number of
lines of force which all converge together upon the same phenomenon. The
choice we make among them may well be founded upon characteristics which,
in practice, fully merit our attention; but it is always a choice. Notably, there is
something extremely arbitrary in the idea of a cause par excellence, as opposed
to mere "conditions." Even Simiand, who was so possessed by the idea of
precision, and who had begun with an attempt (a vain one, I believe) for stricter
definitions, seems to have ended by recognizing the entirely relative character
of such distinctions. "For a doctor," he wrote, "the cause of an epidemic would
be the multiplication of a microbe and its conditions the dirt and ill health
occasioned by poverty; for the sociologist and the philanthropist, poverty would
be the cause, and the biological factors, the condition." This is in all honesty to
acknowledge the subordination of the perspective to the peculiar angle of the
inquiry. Moreover, let us take care: in history, the fetish of single cause is all
too often only the insidious form of search for the responsible person—hence a
value judgment. The judge expresses it as: "Who is right, and who is wrong?"
The scholar is content to ask: "Why?" and he accepts the fact that the answer
may not be simple. Whether as a prejudice of common sense, a postulate of
logicians, or a habit of prosecuting attorneys, the monism of cause can be, for
history, only
194 The Historian's Craft an impediment. History seeks for causal wave-
trains and is not afraid, since life shows them to be so, to find them multiple.
Historical facts are, in essence, psychological facts. Normally, therefore, they
find their antecedents in other psychological facts. To be sure, human destinies
are placed in the physical world and suffer the consequence thereof. Even where
the intrusion of these external forces seems most brutal, however, their action is
weakened or intensified by man and his mind. The virus of the Black Death was
the prime cause of the depopulation of Europe. But the epidemic spread so
rapidly only by virtue of certain social—and, therefore, in their underlying
nature, mental—conditions, and its moral effects are to be explained only by the
peculiar propensities of collective sensibility. However, there can be no
psychology which confines itself to pure consciousness. To read certain books
of history, one might think mankind made up entirely of logical wills whose
reasons for acting would never hold the slightest mystery for them. In view of
the actual state of investigation into the life of the mind and its obscure depths,
this is a further proof of the everlasting difficulty which the sciences experience
in trying to remain contemporaneous with each other. Moreover, it is to repeat
in exaggerated form the often denounced error of an obsolete economic theory.
Homo cecono- micus was an empty shadow, not only because he was
supposedly preoccupied by self-interest; the worst il-
Historical Causation 195 lusion consisted in imagining that he could form
so clear an idea of his interests. Napoleon once said: "There is nothing so rare as
a plan/' Does anyone con- sidei that the oppressive moral atmosphere in which
we are currently plunged comes only from the rational part of our minds? We
should seriously misrepresent the problem of causes in history if we always and
everywhere reduced them to a problem of motive. Moreover, what a curious
contradiction there is in the successive attitudes of so many historians: when it
is a question of ascertaining whether or not some human act has really taken
place, they cannot be sufficiently painstaking. If they proceed to the reasons for
that act, they are content with the merest appearance, ordinarily founded upon
one of those maxims of commonplace psychology which are neither more or
less true than their opposites* Two critics trained in philosophy, Georg Simmel
in Germany and Frangois Simiand in France, have amused themselves in
exposing several of these petitionee principi for us. The Hebertists, one German
historian writes, were at first in perfect accord with Robespierre because he
yielded to all their desires; then they broke with him because they considered
him too powerful. This, as Simmel observes in substance, is to imply the two
following propositions: a favor provokes gratitude; people do not like to be
dominated. Now, these two propositions are not necessarily false, to be sure.
But ndrtaer are they necessarily true. For,
196 The Historian's Craft could it not be held with equal likelihood that a
too- ready submission to the will of a party might excite its contempt for your
weakness rather than gratitude; and, on the other hand, have we never seen a
dictator who stifled even the slightest impulse to resistance by the fear which
his power inspired? A scholastic philosopher once remarked of authority that it
had "a nose of wax, which bends either to left or right indiscriminately." It is the
same with the pretended psychological truths of common sense. Basically, the
error is analogous to the one which inspired that pseudogeographical
determinism which is today once for all discredited. Whether confronted by a
phenomenon of the physical world or by a social fact, the movement of human
reactions is not like clockwork always going in the same direction. Renan to the
contrary notwithstanding, the desert is not necessarily "monotheistic," because
the people who inhabit it do not all bring the same spirit to its scenes. Scarcity
of watering-places would bring about the clustering of rural population, and
abundance of water would disperse it, only if it were true that people made
proximity to springs, wells, and ponds their supreme consideration. In reality
they sometimes prefer, for the sake of security or co-operation, or even through
mere gregariousness, to live in close groups even where every field has its
spring; or inversely, as in certain regions of Sardinia, where everyone builds his
dwelling in the middle of his little estate, they resign themselves to long walks
for the scarce water as the price of the isola-
Historical Causation 197 tion on which they have set their hearts. Is not
man himself the greatest variable in nature? Let us not here be misled, however.
In such a case, the fault is not in the explanation itself. The fault is only in
accepting any explanation a priori. Although up to now there have been
relatively few examples of it, it may well be that, under given social conditions,
the distribution of water sources determines place of habitation more than any
other factor. Certainly it does not determine it of necessity. It is by no means
impossible that the H6bertists really did respond to those motives which their
historian attributed to them. The error was in considering this hypothesis as
given at the outset. It needed to be proved. Then, once this proof—which we
have no right to consider as unfeasible out of prejudice—has been supplied, it
still remains for us by digging deeper into the analysis to ask why, out of all the
imaginable psychological attitudes, these particular ones should have imposed
themselves upon the group. For, as soon as we admit that a mental or emotional
reaction is not self-explanatory, we are forced in turn, whenever such a reaction
occurs, to make a real effort to discover the reasons for it. In a word, in history,
as elsewhere, the causes cannot be assumed. They are to be looked for. . . •
A NOTE ON THE TYPE This book was set on the Linotype in Electra,
designed by W. A. Dwiggins. The Electra face is a simple and readable type
suitable for printing books by present-day processes. It is not based on any
historical model, and hence does not echo any particular time or fashion. It is
without eccentricities to catch the eye and interfere with reading—in general, its
aim is to perform the function of a good book printing-type: to be read, and not
seen. Typography and binding designs are based on originals by W. A.
Dwiggtns.
HISTORY Marc Bloch was born at Lyon on July 6, 1886. Educated at the
Ecole Normale Superieure, from which he was graduated in 1908, he
specialized in history and geography. Later, he taught these subjects at the
lycees of Mont- pellier and Agen. In 1919 he was appointed Professor of
Medieval History at the University of Strasbourg, holding this post until 1937,
when he became Professor of Economic History at the University of Paris. Marc
Bloch's early career as a scholar and teacher was interrupted by service in the
French Army in World War I. At the age of fifty-three, and the father of six
children, he was again called in 1939, serving as a captain until his
demobilization in July 1940, shortly after the fall of France. His book Strange
Defeat is a remarkably penetrating eyewitness account of the state of moral and
physical prostration in which his country found itself at that time. In 1942, he
became active in the French Resistance; two years later he was captured by the
Germans, tortured, and executed. The effect that the news of his death had on
scholars everywhere is best told in the words of D. W. Brogan: "I remember
vividly the day on which the news of Marc Bloch's death reached us in
Cambridge, and how eagerly we pounced on the rumour—false, alas!—that he
had escaped. When we learned beyond doubt that he was dead, we felt that a
blow had been dealt to the whole world of learning." The author of many
distinguished works of scholarship, Marc Bloch is perhaps best known for Les
Rois thaumaturges, Les Caracteres originaux de Thistoire rurale fran- gaise, and
La Societe feodale, each one a triumph of modern historical science. COVER
DESIGN BY Lawrence Ratzkin ISBN 0-394-70512-2

You might also like