.

J, KUIPER

Essays selected and introduced by

JOHN IRWIN

I
VIKAS PUBLISHING HOUSE PVT LTD
V I K A S P U B L I S H I N G H O U S E P V T L T D
Regd. Ofice: 5 Ansari Road, New Delhi 110002
H. 0. Vikas House, 2014 Industrial Area, Sahibabad 201010
Distt. Ghaziabad, U.P. (India)

COPYRIGHT @ Vikas Publishing House Pvt Ltd, 1983

1V210816
ISBN 0-7069-1370-1

Printed at Nu Tech Photolithographers, Jhilmii, Delhi 110032 (India)

Contents

Editor's It1 troduc tion

The Basic Concept of Vedic Religion

The Golden Germ

The Three Strides of V i p u

The Bliss of AZa

Cosmogony and Conception-A Query

The Heavenly Bucket

The Ancient Aryan Verbal Contest

An Indian Prometheus?

The Worship of the Jarjara on the Stage

1ride.u
Thi
spag
eint
ent
ional
lyl
eftbl
ank
EDITOR'S INTRODUCTION

F.B.J. Kuiper was until 1972 professor of Sanskrit at the University of

Leiden, where his main work had been in philology.' However, he has been
heard to say that from an early stage in his career he was conscious of the
competing demands of two prime interests: comparative linguistics, on the
one hand, and early religion, on the other. It is rare indeed to find in the
same person the talents required to make a first-class linguist and to explore
in depth the nature of religious myth; yet, as the reader will quickly discover,
Kuiper commands both fields with impressive mastery. This very fact places
him in a category rather apart from most historians of religion, who are
primarily concerned with the evolution of religious ideas. Kuiper is more
especially interested in the meaning of ancient religion as revealed by the
internal structure of its myths, to which language is an important key.
The myths with which Kuiper is especially concerned are those relating to
the origin of the universe. 'Cosmogony'- the term applied in academic circles
to this class of myth -derives its terminal from Greek,gonia, implying 'birth'
or 'generation' of the cosmos. This distinguishes it from the better-known
term 'cosmology' which is concerned with theories of the universe as an
already existing, ordered whole. I n the earlier stages of religion, it was
'cosmogony' rather than 'cosmology' that dominated man's ideas about the
world: I t is perhaps symptomatic of the relative backwardness of Indian
studies in this particular field that, although much has been written about
Indian cosmology, this appears as the first book devoted specifically to the
ancient Indian cosmogony.
For those new to the subject, it should perhaps be stressed that the ancient
world's obsessive concern withshow the world began had nothing to do
with curiosity or the love of story-telling. It was the expression of a religious
Mystery, based on an urgently felt need to get into right relationship with the
sacred world as source of cosmic order, upon which early man felt his exis-
tence and survival to depend. Since it was also based on a conception of Time
as cyclic, the birth of the cosmos was felt to be the key to the perpetuation of

1 For a bibliography o f Kuiper's publications up to 1967, see PratidZlzam, studiespresented toF.B.J.

Kuiper on his 60th birthday, edited b y J.C. Heesterman and others, T h e Hague, 1968.
life expressed in the return of the sun, the seasons, and the germination of
seed.
In short, it provided the sacred model of all re-generation, and repetition
and renewal were of its essence. This in turn called for ritual re-enactment of
the cosmogony at each moment of crisis if the world was not to slip back from
Order into Chaos. I n India, as in many other traditions, all Time was felt to
be encon~passedwithin the Year; hence the supreme moment of crisis was the
changeover from the old to the new year. However, celebration of the New
Year was not always or necessarily calendric; it was commonly associated
with the harvest-cycle, so that there could be several New Year celebrations
within one year, varying from region to region. The only feature common to
every New Year festival was that it was in some form a ritual re-enactment of
the cosmic start when heaven and earth were separated, and our organized
universe with its countless other dualisms (gods and demons, fire and water,
male and female, light and darkness, and so on) were brought into being.
There is no reason to suppose that there was any difference among the archaic
cultures as far as the basic pattern of the cosmogony was concerned: indeed,
as more evidence becomes available, it becomes increasingly clear that it was
the common property of the whole of the known ancient world.
T o understand the place and the nature of cosmogonic myth in archaic
cultures, we have to counter the influence of the Greeks. By secularising
mythology and turning it into literature, the Greeks did much to obscure its
original function and meaning. The description 'Creation Story' should not
be allowed to detract from its original nature as religious Mystery, communi-
cated in metaphor. Moreover, in most cultures, the creation of the world was
thought of as too mysterious a business to be explained in any one way. In so
far as the cosmogonies do get transformed into stories or legends (and one
thinks here, too, of the countless transformations in the Hindu epics and
PurZnas) they have already lost something of their character as Mysteries. By
the same logic, the Westerner has to counter the influence of the Bible-
especially the Book of Genesis as translated by those who could have had
little knowledge of the original connotation of the words. For instance, we
can be certain that the original author could not have meant 'create' in the
sense of creating something out of nothing; rather, he would have meant the
shaping or moulding of what was already in existence.
In the Vedic context, Kuiper describes the cosmogony as an evolution in
the strictly etymological (non-Darwinian) sense of that word, from Latin
e -holvere, which means a 'rolling out' or 'unfolding'.Indra's rcile in the Rgveda
was not, he says, to create the world out of nothing but to act as 'a kind of
magnetic force' over the primordial world of chaos. By this means, Indra
caused the undifferentiated powers and entities of that world to form into two
poles of existence, constituting the dualities described above.
Like all new thinking, Kuiper's essays demand effort from the reader; yet,
as the first essay shows, he also reveals himself as a master of clear exposition,
which should encourage the reader to persevere. What may be helpful at this
point is something about the genesis of his thinking in relation to the Dutch
background and what is sometimes called the 'Leiden school' (a term permissi-
ble only if we use a small 'S', since it never had any statutory existence).
There is no doubt that to anybody interested in the religions of the ancient
world, the intellectual climate at Leiden University in the 1920s and 30s was
highly creative and stimulating. First and foremost was the figure of William
Brede Kristensen (1867-1953) whom Kuiper has described as 'the Nestor of
Dutch historians of religion'. A Norwegian by birth, Kristensen succeeded
C.P. Tiele as professor of Comparative Religion -a term Kristensen himself
preferred to avoid, because in current usage it often implied division between
'higher' and 'lower' religions which he regarded as false. Likewise, Kristensen
disassociated himself from the evolutionary approach then fashionable under
the leadership of Rudolf Otto's Idea of the Holy. In the latter case, he thought,
a purely philosophical notion of what constituted the 'holy' was being forced
upon historical reality. Instead, he set out to understand ancient religions
from the angle of the believer, by rediscovering the believer's own terms of
reference. This involved him in close analysis of cult as well as doctrine. His
special fields were Egypt, Western Asia, and Greece and in each of these
areas his work was based on philological knowledge of texts and some acquain-
tance with archaeology. As far as method was concerned, his closest parallel
was the work of his slightly senior English contemporary, Jane Harrison
(1850-1934).Kristensen was especially interested in the 'netherworld' aspect
of ancient religion, and ~nostof all its relationship to the concept of Resur-
rection:This may have been one-sided; nevertheless it was a self-chosen
limitation and it gave him firm grip of the material.?
Within the context of the Leiden school, the balance was restored by the
much younger cultural anthropologist, .P&. dc Josselin deJong (1886-1964)
who about 1926 adopted the structural approach to the study of culture. T o
him, and to other cultural anthropologists from the time of Hubert and Mauss
at the beginning of the century, religion was an aspect of culture. Although
de Josselin de Jong published little himself, his thinking was widely dissemi-
nated through the work of his students-especially G.J. Held in The Maha-
bharata: an ethnological study, Amsterdam and London, 1935,and G.W. Locher,
The Serpent in Ku~akiutlReligion, Leiden, 1932.3
A third major contribution to religious studies-and especially to cos-

2Kuiper says that he only once attended a lecture by Kristensen ("jt~stfor the pleastire of seeing
and hearing the grand old man") since he was too occupied with Latin, Greek and comparative
linguistics. Not until he was already working in Indonesia (1934-39)did he study Kristensen's
classic Leven uit Dood in association with the Rgveda. This, and all other references to Kuiper's
v i e w where not attributed to a source, are based on personal discussion o r correspondence.
3 For further discussion of the 'Lciden school' of anthropologists, see Rodney Needham in his
introduction to Emilc Durckheirn and Marcel Mauss, Prilr~iiiveC[assification,University of Chicago
Press. 1963, pp. XXX f f .
 mogonic myth -came from A.J. Wensinck, who was the University's professor
of Semitic Philology. His important works included The Ideas of the Western
Semites Concerning the Navel ofthe Earth, 1916; The Ocean in the Literature ofthe
Western Semites, 1918,and The Tree and Bird as CosmologicalSymbols in Western
Asia, 1921-all published in the Transactions of the Dutch Royal Acaderny of
Science (Koninklijke Akademie van Wetenschappen), Amsterdam.4 Wen-
1 sinck's pioneer study of the 'Navel of the Earth' no doubt stimulated Adriaan
de Buck in his in~portantdoctoral thesis on the Primordial Mou~ld(oerheuvel)
l in ancient Egyptian r e l i g i ~ nLater,
. ~ de Buck was appointed first Leiden pro-
1 fessor of Egyptology. Kuiper says he became acquainted with de Buck's ideas
\
through the latter's student, B.H. Stricker, who was only a few years his
1I junior but whom he did not meet until World War I1 or shortly after. Stricker
then gave Kuiper an off-print of one of his articles which (he says) impressed
him 'as a revelation'. Soon after this, Stricker won international recognition
Il as an Egyptologist.
Having taken his doctor's degree in 1934 on a subject of con~parative
linguistics, Kuiper then went to Indonesia to teach Latin and Greek. "It was
from those years (1934-39),"he writes, "that I started working as an Indologist,
and as far as my duties allowed I concentrated on study of the l$gveda. Olden-
berg's Relip'on des Veda left me with the conviction that that book-although
the work of a master of Vedic philology -clearly failed to grasp the essence of
i Vedic religion. I n view of the limitations of the 19th-century attitude to
\ religion, one could hardly have expected otherwise. Although beginning my
I own studies from scratch, I took as my first guides the dissertations of G.W.
l
Locher and G.J. Held[see para. 2, p. 31, both of whom had been my fellow
3 students. Locher took his Ph.D shortly before I went to Indonesia; Held, one
1
i or two years later. Both of them I had known personally, and with Held in
i
! particular I had had several talks on his work before I left; and in Jakarta we
I met again. I missed the training of an anthropologist, and theories on kinship
were beyond me: I regret that I have never been able or energetic enough to
;I
1 make u p for it. On the other hand, I had a strong feeling that the claim of
j anthropologists that religion is a projection of the social order, is unfounded.
I felt intuitively that what was needed was a combination oi Kristensen's
method and example with the rigid structural approach of anthropologists.
During the war years the University of Leiden was closed by the Germans,
but after the war when I came to know the academic world better, I was
dismayed at seeing how the younger generation of historians of religion, on

*Special mention should perhaps also be made of Wensinck's article, "The Semitic New Year
and the Origin of Escliatolo~y,"published in Acia Oi<eitialia,I, 1923, pp. 158-99. In this article he
demonstrated links between Babylonian cos~nogonictexts and celebrations of the New Year.
Later, Kuiper was first to suggest that the Ipp-veda was in essence "a textbook for the ceremonies of
the New Year festival".
5Adriaan de Buck, De Egyptische voorstellingei~ beirefleitde den oerheuvel, thesis published in
facsimile by Eduard Ijdo, i e i d c n , 1922.
the one hand, and the cultural anthropologists, on the other, lived in closed
compartments. I t seemed to me that each was working with his own methods
and presuppositions while ignoring what was being done in neighbouring
fields."
Kuiper's review of F.D.K. Bosch's The Golden Germ (published here in
English translation for the first time, pp. 23)shows thatby 1951he had accepted
the structural approach to the study of religion while at the same time disasso-
ciating himself from the anthropologists' claim of 'the primacy of society in
classification'."
The book to which Kuiper owed most of all in the crystallization of his
thinking was Hans Scharer's Die Gottesidee der Ngadju Dajak i n Sud-Borneo,
first published in German at Leiden in 1946, and re-issued in English transla-
tion by Rodney Needham under the title Ngaju Religion: The Conception of
God among the North Borneo People (The Hague, 1963).This work had been
written as a doctoral thesis under de Josselin de Jong, although kinship
analysis played no part in it: it was wholly concerned with reconstruction of
the Dyak's 'theological system'.
This sums u p the Leiden background against which Kuiper's new interpre-
tation of the R p e d a slowly took shape. Addressing himself now to the basic
question of what constituted the special character of the R p e d a , Kuiper was
especially conscious of what he calls "the curious one-sidedness of the hymns".
Prior to this, scholars had evaded this problem. Looking at the history of
religion from an evolutionary (Darwinian)point of view, which presupposed
an almosi obligatory progression from 'primitive' to 'advanced', they based
their analyses of the R p e d a on the fact of its being chronologically the oldest
document; and since religion in its 'primitive' stages was thought to have
been concerned with worship of 'natural forces', Indra's fight with the demon
Vpra was interpreted as a nature myth concerned with thunderclouds and
rain. Against this, Kuiper arrived at his view that Indra's fight was a Creation
myth, and that the dualism of Deva and Asura was fundamental to its struc-
ture.
I t is not necessary to give here an outline of the Vedic cosmogony as it
eventually took shape in Kuiper's thinking, because this difficult task has
been achieved with masterly clarity in the first essay entitled "The Basic
Concept of Vedic Religion". Although it is among the last to have been
written, we have given it first place because it is the logical starting-point for
readers coming afresh to the subject.
Looking back in retrospect, it is interesting to see that although Kuiper
carried the subject much further than anyone else, a number of other scholars
had been niovihg independently towards cosmogonic identification ofthe
Indra-V~tramyth at the same period-yet unknown to one another because
of isolation forced upon them by World War 11. Especially notable for their

@Wordsquoted from Kodney Needham(see fn. 3).

work in this direction were W. Norman Brown in America and Heinrich
Liiders in Germany. As it happened, neither of these scholars had influenced
Kuiper in the genesis of his own thinking. Although Norman Brown was
first in print on the subject in an article he contributed to the Journal of the
American Oriental Society in 1942,7this journal was not available in wartime
Holland and did not come to Kuiper's notice until several years after the
war. I n the case of Liiders, there was likewise a long delay; and when in 1952
the first part of his life-work, V a ~ n awas
, ~ posthumously published, it became
clear that his ideas on Indra had passed through several stages. Although he
had finally arrived at recognition of some basic elements of the cosmogony,
one is left with the impression that his theory of the celestial ocean had for
long been a barrier in the way of correct interpretation. His final insights
can only be guessed from the fragments that Alsdorf as editor of the posthu-
mous volumes has so painstakingly put together.
Kuiper has remarked that 1949was a key stage in the development of his own
thinking, because it was then that he had first understood the purport of the
difficult hymn X. 124 (although his interpretation was not explicitly presented
until thirty yearslater, in his niost recent major work, Varuna and Vicl<saka,
Amsterdam, 1979).However, others may claim that his most important single
advance had been made in 1946 with his recognition that V~tra's'hill' or
'rock' was none other than the oerheuvel or 'Primordial Moundl,gwhich some
of us have now come to regard as the most important of all images in the early
history of religion and art. This identification appeared in a book review
written in Dutch, and in view of its importance as a landmark in Indian
cosmogonic research, the relevant passage is here given for the first time in
English.

I n my opinion, any attempt to explain V p a should start from the equi-

valence of V p a and Vala and place them in the total mythological concept
of the world.Then,I think, it will be clear (what can only be pointed at
here) that the myth takes us back to that stage of the creation when there
was not yet a heaven,and the arnrta tree (Rpeda I. 164.20ff.) did not yet
grow u p to heaven, and the gods consequently did not yet have the amytal
soma at their disposal, as it was still guarded, together with all the goods of
life (such as water=rasa, the essence of life; and sun=the light of life and
the worlds) in the primordial hill by the serpent. . ..The fact that Indra wins
them and makes the sun rise is his great creative act, which is repeated
7 W. Norman Brown, "Tlle Crration Myth of the Rig-veda," Joun~ulof the American Or-ic~~tal
Society, vol. 62, 194% p]'. 85-98. Recently re-pul~lishcdin It~dioa~tdIr~clology.Selected Articles by W.
Normalt Brown, edited by Kosanc liocher. Banaras, 1978, pp. 20ff.
8 Heinrich Liidcrs, Vuruga, aus dent Nachlass herausgegeben uon Ludwig Alsdorf; two volumes,
Gottingen, 1951, and 1959.
9 Kuiper uses the traditional Er~glisllrendering, 'Primordial Hill: He does not exclude 'Mound'

as an appropriate alternative. but he does not share the editor's preference for the latter, leaving
open the choice between 'hill' and 'mound'
 every morning and causes a new crisis every year at the winter solstice.
Therefore, Hillebrandt may be right when he puts Indra's act (Ved. Myth.,
11.2.182 ff.) at the beginning of the devayana and also connects Usas with it.
{n that light a ritual fight during the wi: &ersolstice between an Arya and a
Siidra, as representatives of the two cosmic moieties and ending with the
sacrificial death of the ~ i i d r awould
, indeed be well imaginable as a re-
iteration of the mythical act."lO

It is in the nature of all creative scholarship that it extends under-

standing in areas far beyond those with which the author is immediately
concerned. I n this case one might cite the value these essays will have to
students of early art and architecture where cosmogony is now generally
recognized as having played a fundamental r6le. Indeed, there is a sense in
which the very act of artistic creation-in other words, of fashioning, structu-
ring or giving form to (literally, in-forming) inanimate materials -is a 'crea-
tion' in the ardhaic cosmogonic sense we have been discussing, that is, the
bringing to 'life' of materi~lsalready existing. It is no surprise, therefore, that
the cosmogony was conceived as the sacred model of all artistic creation, and
that the building of a temple and the fashioning of a divine image, no less
than the poet's struggle for the right word," were occasions for ritual re-
enactment of the separation of heaven and earth.
The first steps towards recognition of this were already being taken in the
1890s by W.R. Lethaby -an Englishman of quiet genius whose works left no
impression on the academic world of his day, but which are now seen as
prophetic.12 Lethaby realized that the building of monuments in thc ancient
world was first and foremost a rite whereby man sought to identify himself
with the source of cosmic order by placing himself at the 'Centre of the
Universe', otherwise known as the Navel of the Earth (Sanskrit p~rthiuz~a~bhz,
Greek omnphalos, Hebrew,tabbZ~eres,and so on) where the universe was 'born'.
Since every spot where man was in contact with the divine shared this mythic
location, every shrine stood at the Centre of the Universe. T o the rationalistic
thinking of the Evolutionists, restricting themselves to principles of empiri-
cal science, the very notion that every sanctuary shoulcl be conceived as the
centre of the world seemed absurd, and under the influence of Darwinian
ideas of biological evolution it was all too ea5y to imagine that the history of
religion, too, had involved progression from 'primitive' to 'advanced'.
W r o m Kuiper's review ot Leo Buschardt, \+a, Det rituelle Duen~ondrabi den Vediske Somakulf,
Copenhagen, 1945. The review was published in Museum, Maandbhd voorPhiloiogie etGeschiedenis,
no. 52, 1947, c o l ~ m n s198-200.
"Besides the several examples given in these essays, see also Jar1 Gonda, "A Note on the Vedic
Student's Staff,"Jounral of the OricnlalInstitute, University of Baroda, vol, 14,1964-65,pp. 262-72.
1eW.R. Lcthaby?Arclrifecfure,iMysficisnrandMyth, London, 1891 (reissued by the ~;chitectural
Press Ltd., London, 1974). Of this remarkably lucid book, the Times reviewer wrnte when it came
out that "the author'a insight is so esoteric that to plain people its deliverances are simply
unintelligible" (The Times, Y1 Decembel. 1891).
Hence, the notion of every sanctuary existing at the Centre of the Ulliverse
was seen as another instance of the 'childishness' of early religions. Against
this, Lethaby recognized that even more absurd in the situation was applica-
tion of the reductive logic of empirical science where it had no rele\rance. In
other words, it was obvious to Lethaby that the ancient notion of sacred
space and location had nothing to do with the profane science of geometry
but answered to an entirely different set of ideas. With this recognition was
linked the discovery that in the ancient world the shri~lewas a point of passage
between the human and divine worlds.
Already by the 1920s and 30s there were a number of archaeologists and art
historians thinking on these lines, and it is significant that they were all ori-
entalists. Especially notable among them were A.M. Hocart (who for a short
time had been Archaeological Commissioner in Sri Lanka); Paul Mus, author
of a monumental work on Borobudur, but whose main experience had been
in Indo-China; the German archaeologist Walter Andrae, specializing in
Western Asia, and the Belgian Car1 Hentze who was a Sinologist. The first to
apply these ideas to the study of Indian monuments was A.K. Coomaraswamy.
After World War 11, fresh impetus was given by Stella Kramrisch, in The
Hindu Temple (1946), and F . D . K . Bosch, in The Golden Germ (originally
published in Dutch in 1948, and re-issued in English translation in 1960).
Special mention might also be made of two remarkable papers which appeared
within the first post-war decade: one was by A.K. Coomaraswamy, "The
Hindu Temple," published in the same year as his death;ls and the other by
Mircea Eliade, "Centre d u Monde, Temple, Maison," issued in 1957.14During
the interval of eight years between the publication of these two seminal papers,
we can see now that there had been an important theoretical advance. Whereas
previously the temple as "architectural microcosm" had been thought of as
an "image of the universe," this definition was seen to have embodied only
part of the truth. Only with the publication of Eliade's essay was it made clear
for everybody that a microcosm was not so much an "image of the universe"
as an image of the creation of the universe. I n other words, it must be seen as
a dynamic image, not a static one. It implies not so much a plan or a formula,
but a process. It is precisely this dynamic aspect that Kuiper's studies so
bx-illiantly illustrate. Many of its features recur in other cultures, and since
in the total picture the Indian cosmogony turns out to be one of the earliest
comprehensively understood, its importance to the student of the origins of
human culture can hardly he overestimated.
1980
]o11,11 n 0 1 1 t , Ashford Chace, Petersfield, Harzts., U.K.

'l'Recently republished in A.K. Coomamswamy, SelectedPapers, edited by Roger Lipsey, vol. I.

l4 Originally delivered
at an international conference arranged under the title, Le Svmbolisme
connig&&s Monuments relzpew, held in 1955 at the Istituto ltaliano Per il Medio eh Estremo
Oriente, Rome, and published in the I!roceedings,Serie Onentab Roma,vol. XIV, 1957, pp. 57-82.
 DIG REEIGIO

The basic concept of Vedic religion is no doubt an unusual object

of study, as it implies a t least three assumptions: first, that the
time has come for a more general survey of this religion; second,
that there actually does exist a basic concept; and third, (;hat
mething sensible can be said about it. Each of these assumptions
rhaps be questioned. The following study will show how I
think it has become possible in the present state of Vedic studies
to discuss the underlying idea of Vedic theo1ogy.l

The key to an insight into this religion is, I think, to be found in

its cosmogony, that is, the myth which tells us how, in primordial
time, this world came into existence. This myth owed its funda-
mental importance to the fact that every decisive moment in life
was considered a repetition of the primeval process. Therefore the
myth was not merely a tale of things that had happened long ago,
This is the text of a lecture given to an audience with no previous
knowledge of Vedic religion. This explains why the concentration on the main
lines haa sometimes led to some simpli6cation in the presentation of the facts.
Tbus, for the &e of dearness only V m q a is mentioned because the other
Adityas were not relevant in this context. Oversimplificationwill, I think, only be
found a t the end, where the status of the gods of totality could best be illustrated
ere reproduced without material changes beoauee it
h to summarize what I think Vedic religion waa basi-
a study whioh will be published elsewhere. Therefore,
only a few refemnws have been added in footnotes.
nor was i t a rational explanation of how this world had become
what it is now. The origin of the world constituted the sacred
prototype of how, in an endlessly repeated process, life and this
world renewed themselves again and again.
To this aspect of the myth, however, we shall have to return
later on. Let us, for the present moment, see what the myth tells
us about the origin of the world.
In the beginning there was only water, but these so-called
primeval waters bore in themselves the germ of life.2 From the
bottom a small clod of earth rose to the surface, where it floated
about. The clod spread on the surface and became a mountain,
the beginning of the earth, but it continued to float on the water^.^
There is a variant version in which the highest god, the World
Father, drifted about, but the presence or absence of this god ia
only of marginal importance. The primordial world itself was
sacred, and for the process of this genesis to take place there was
no need of a creator. Things were considered to exist, somehow,
in their own right. In this first stage, however, as represented by
the mountain, the world was still an undifferentiated unity. The
poets sometimes speak of a darkness as the initial state, but this
is clearly a mere attempt to express what could not properly be
expressed in words. None of the contrasts which constitute our
phenomenal world yet existed. There was no heaven or earth, no
day or night, no light or, properly speaking, darkness.
Nor did that contrast yet exist in which for archaic man the
cosmic dualism manifested itself most clearly in human life,
namely, the all-pervading contrast between man and woman, male
and female. Hence i t is that the myth, a t this stage of the genesis,
sometimes refers to bisexual primeval beings to whom others owe
their existence. In mixing up the roles of father and mother the
myth is, indeed, consistent and, in its own way, logical. A special
group of gods, the Asuras, was connected with this first stage.
Their great importance will become clear later on.
It is obvious that this first stage of the cosmogony is not a
creation myth a t all. There were in the beginning already certain
things the existence of which was taken for granted, as a last
irreducible fact. The first Vedic thinkers who have left it trace in
the Rigveda were well aware of these bounds set to their specula-
qP&. See Sylvain LBvi, La docttdne dzr sacraw dans lea b r i i h m ~(Paris,
1898), pp. 13, n. 3; 159, n. 3. Compare, e.g., I n d k Maim, Congratulatory Volume
Gonda (Laden, 1972), p. 145, n. 1; F. B. J. Kuiper, "Cosmogony and Conception:
A Query," History of Religkona 10 (1970):99.
For references m, e.g., Kuiper, pp. 100 ff., 109-10.
tions. The last words of a philosophical hymn are: "Whence this
creation came into existence . . . that only the Supervisor of this
world~knows-or perhaps not even He."4

This state of undifferentiated unity came to a sudden end by the

second stage of the cosmogony. It started with the birth of the god
Indra, outside the primeval world. It is not said whence he came,
nor could the texts have been more specific on this point as it was
apparently part of his character that he came "from n ~ w h e r e . " ~
Indra's mythical function is limited to promoting the emergence
of a dual world of individualized forms from the undifferentiated
chaos. Again, Indra is not a creator in the Old Testamentary sense
of the word, and there are even versions which describe the process
as due to some internal forces, without the intervention of the god
a t all. His exploit, which we shall have to consider more closely,
can best be defined as a demiurgic act. He starts a process in the
primeval world of unformed matter, a process owing to which a
world of mere potentiality became the world of reality, in which
light has arisen and forms a contrast with darkness, in which life
exists along with death, and in which good is counterbalanced by
evil. Indra does not create anything but rather acts as a kind of
magnetic force which, as one text says, causes all powers, all
entities in the world, to side with one of the two poles of e x i ~ t e n c e . ~
One might perhaps call the process an evolution, in the strictly
etymological sense of the word. Anyway, if one refers to this myth
as a creation myth (as I prefer t o do), i t should be borne in mind
that the word is here used in a very indirect way.
As for the Rigveda, it is for special reasons only concerned with
this second stage of the genesis of the world. From the stray ref-
erences in the hymns the following picture can be reconstructed.
Indra's demiurgic act appears to consist of two different parts,
which concern the primeval hill and the tree of life, respectively.
Let us first look a t the former. The hill, which is still floating on
the primeval waters, has to be riveted to the bottom, and t o be
opened. There is, however, a strong force of resistance in this

Rigveda X . 129. 7.
-- On -the
- uncertainty about Indra's origin, see Rigveda 11. 12. 5 (kziha &a),
A. 'is, 1u.
BawEhclyana 6'rautaaiitra XVIII. 46 (p. 401, line 11): "When the Devas and
Asuras were waging the Great War, all these beings split into two groups; some
went to the Devas, others to the Asuras."
 mountain, and so Indra's heroic fight, although sometimes de-
scribed as being directed against the mountain, is more ofbn
directed against that force, which is denoted by the word vrtra.
V?.tra means "obstruction, resistance." I n the myth the power of
resistance is personified by a dragon, and Indra must accordingly,
like Saint George and other mythological heroes, slay the dragon.
It should not be forgotten, however, that this dragon, which itself
came t o be called Vrtra, only represents a special aspect of the
primeval mountain, a resistance which Indra had to overcome in
order to split open the hill. I n this fight Indra is victorious. He
slays the dragon, and from the hill, opened by force, life bursts
forth under its two aspects of water and fire. I n the creation myth
the water is represented by four rivers, which stream from the top
of the mountain in four different directions, and the fire by the sun
which rises from Ghe mountain or the waters. At the same time
the mountain is no longer floating about. It has now found a
support (as the texts say)7 and starts growing on all sides, until it
has the expanse of the earth. Still, the primeval mountain remains
the cosmic center and the nail which keeps the earth in its place.
As for the second part of Indra's act, which concerns the tree of
i
life, Indra here functions as a pillar in propping up the sky, which
1 until then had been lying upon the earth. I n so doing he creates
R

the duality of heaven and earth. From a mythological point of view

this is not a separate event, because the contrast between heaven
and earthisonly one particular aspect of the all-pervading dualism.
Thus, with the sun rising to the sky, the contrast between light
and darkness is born, which is parallel to that between life and
death.
In&ra's mythical role remains limited to this single exploit.
Again and again the poets say that he slew the dragon, extended
the earth, and lifted up the sky, but that is about all they can tell
us about him. There is, however, one particularly interesting detail
in his creation act, and here we return to his connection with the
world tree. This tree belonged t o the dual cosmos, since it was
identical with the cosmic pillar which, in the center of the world,
kept heaven and earth apart. It must accordingly have arisen
when the sky was separated from the earth. The obvious conclusion
is that Indra, at the moment when he "propped up" the sky, must
have been identical with the tree. On the other hand, there are
sufficient indications t o show that in general Indra has nothing to

7 On prati@G, me, in gener~l,Kuiper, pp. 109-10.

 do with the cosmic tree and the world center. His identity vcith
the pillar a t the moment of creation, when he himself literally was
the world axis, must accordingly have had a momentary character.
This inference is confirmed by data about the Indra festival of
much later times. From these we learn that i t was then still cus-
tomary to erect every year, during the New Year's festival,* a pole
in honor of Indra. Its most interesting feature is that during the
few days that it stood erected and was worshipped, it was consid-
ered to be identical with god Indra and was sometimes denoted by
his name. This gives a special significance to the fact that after
some seven days the pole was pulled down, taken away, and thrown
into a river, which would not have been possible unless the function
l of the god himself, whose name it bore, had for the time being
. come to an end. This, again, confirms the conclusion drawn from
the Vedic evidence that Indra was a seasonal god, whose myth-
ological act consisted in creating and renewing the world and
B inaugurating a new year.
I It does not mean that the god was entirely absent during the
rest of the year. Just as it was said that Indra immediately after
his birth slew the dragon, so he was present to aid his devotees,
whenever they had to overcome inimical forces in various shapes,
such as aborigines sheltered in their fortresses or demons of disease.
Even the poet craving for inspiration considered his mind a micro-
cosmic replica of the earth resting upon the subterranean ocean,
and he prayed to the god to break his inner resistance so that the
inspiration could stream forth from the ocean of his hearkg

So much for the description of Indra's part in creating this world.

It is, however, only one side of the genesis. It seems never to have
been recognized that there is a different aspect, which constitutes
the very core of the Vedic conception of the world. This is the
problem of the Asuras.
As we have seen, the Asuras had been the gods of the primordial
world. Since Indra broke its obstructive power, the question nat-
urally arises, What became of them and their world?
As far as their world is concerned, the answer is clear: after the
SW, in general, J. J. Meyer, Trilogk altdndCcher MaChte und &ate der Vegeta-
I . 67. 18, g&
t b n (Ziirich and Leipzig, 1937), 3 : 4 , 113; and MalicZbhiirda aorpvd-
aare, "at the end of the year."
B For the breaking of the inner "resistanoe" (uytrd),see Indo-Iraniun
(1964):126.
powers of life had been set free, the primordial world became the
sacred earth, which together with heaven formed a pair of cosmic
moieties. But while on the surface of the earth life manifested
itself, there was also the subterranean world, with the primeval
waters on which the earth was believed to rest. This, too, had
come to form part of the organized world after Indra's creation act.
As we shall see, however, it had not quite lost its original character
and remained an ambiguous element in the creation.
As for the Asuras themselves, they constitute the central prob-
lem of Vedic religion. After Indra had created the dual cosmos, the
Asuras were no longer the only gods, since the dualism also
extended to the world of gods. Along with Indra a new group of
gods made their entrance. Their name, Devas, was the old Indo-
European word for celestial gods. As such, the Devas were opposed
to the gods of the primordial world, and the fight of Indra, the
chief and protagonist of the Devas, against the dragon must have
been considered to be also directed against the Asuras. Indeed, later
Vedic texts no longer refer t o Indra's fight but instead always
speak of the cosmogonical fight between Devas and Asuras. Their
enmity had a tragical character because the Devas were the
younger brothers of the Asuras. That the younger group proved
superior to the older is a pattern not unknown in systems of social
organization.
After Indra had slain the dragon, which signals the defeat of the
Asuras, the two parties of Devas and Asuras had to come to terms
with each other. What then happens is the most momentous event
in the whole cosmogony. It appears impossible to incorporate all
the Asuras in the ordered cosmos. Only some of the chief Asuras,
such as their king Varuna, go over to the other party and side
with the Devas. The rest of them, however, are driven away from
the earth and take refuge in the nether world. (See fig. 1.) We owe
it t o the very archaic character of the Rigvedu that a direct and
clear trace of this split within the group of Asuras has been pre-
served. I n this oldest text a distinction was still made between
devhv kcisurci;, "Asuras who have become Devas," and, on the other
hand, &urG cidev@, "Asuras who are not Devas." This fact had
already attracted the attention of a discerning scholar in the nine-
teenth century,1° but since the true nature of the Asuras, as far
as I can see, was and has ever since been misunderstood, it was
impossible to appreciate the implications of this distinction. Only
l0 See P. von Bradke, Dyriua Asura, Ahura Mazdd und die. Asuras (Halle, 1885),
passim.
 Stage I
primordial World Asuras ( V a y , etc.)

Chaos Ordered cosmos

Stage I1
Dual World

I
in the light of the cosmogony its full relevance becomes clear.
I Although the original nature of the Asuras had already been recog-
nized without the heip of these Rigvedic data, they arc, as a
confirmation of the correctness of this reconstruction, welcome and
valuable.
I used the word "reconstruction" because in later Vedic texts
it is hard to find any direct trace of the situation which 1 have
sketched here. Never again is Vsruna called an Asura. This term
is henceforth reserved for the banished demons, whose cosmo-
gonical fight with the Devas is constantly referred to in this lit-
erature. Varuna, however, had become a Deva, no less respectable
than Indra and the others, although still marked by certain inaus-
picious features. Only a more profound study discloses that Varuna,
although his title of Asura had long since been tabooed, continued
to be much more ambiguous than is usually realized. If, however,
Varuna's character has to a large extent been misinterpreted, this
is not due to a lack of interest on the part of students of Vedic,
religion. Far more studies have in the last few decades been devoted
to him than, for instance, to Indra, and he has recently been
characterized-and rightly so-as the neuralgic point in Vedic
studies.ll

See Louis Renou, in Pestgabe fur Herman Lommel (Wiesbaden, 1960), p. 122:
"le point nevralgique des Qtudesvediques."
 For that reason the figure of Varuna must, even in the frame-
work of these sketchy outlines, occupy us somewhat longer than
Indra.
First, it must be remarked that Varuna's transition to the Devas
is not an isolated phenomenon. It rather reflects a stereotyped
pattern. In this connection it may not be superfluous to warn
against the current misconception that the Asura~as so-called
aemons impersonated evil. It should not be forgotten that their
world is one of unformed, potential life-the material out of which
the cosmos is shaped. The Asuras are not fallen angels but potential
gods. Sometimes it is related that an Asura of his own accord
leaves his world and sides with the Devas. In other cases he is
"called forth" (as the texts say) by the Devas, who are unable to
achieve their aim without the assistance of a certain Asura. The
myth here clearly points to bounds which are set to the powers of
the ordered world.
Second, when Varuna is willing to become a Deva, Indra offers
him a rulership.12 Seemingly-and this is the current interpre-
tation-Varuna is here enticed by the promiseof sovereignty.What
in fact is meant is that Varuqa becomes "lord of the waters," his
traditional title until much later times. Then, however, the true
meaning of this function was no longer known, as the term was
mostly interpreted as "lord of the ocean." There can be little doubt
that the original meaning was quite different. In the Veda the
term "the waters" denoted, first and foremost, the primeval waters
upon which the earth rested, and it can be proved that it was of
these waters that Varuna became the ruler. Even after his incor-
poration in the ordered cosmos, accordingly,Varuna had a function
which corresponded to his origin as god of the primordial world.
From 'now on he resided in the netherworld, at the roots of the
world tree and near to (or in) the subterranean cosmic waters.13
Third, the difficulties which Varuna has presented to modern
research are, i t seems, largely due to vain attempts to describe in
rational, noncontradictory terms a god whose very characteristic
is his ambiguity. His inner contradictions, indeed, defy any
attempt a t a strictly logical definition.
An illustration of his basically ambiguous character can be found
in his relation to the banished Asuras, his former brothers. As we
have seen, all the Asuras who were not integrated in the organized
world fled ko the netherworld, accordingly to the same world
la Rkgeteda X. 124. 5.
l3 See, e.g., Indo-Iranian,J o u d 8 (1964),:107 R.; and India Maim, pp. 150-51.
where Varuna resided. The Vedic texts omit, purposely I think,
to mention this fact, but the question inevitably arises of what
kind of relation there was between Varuna and the rlsuras.
The answer to this question is given in the much later
bhErata. In some passages i t describes in an ingenuous way Varuna
as seated in his subterranean palace, surrounded by Asuras as
attendants.14 What exactly the authors of these passages may have
thought when composing these lines, is hard to say; but it is
obvious that others must have felt shocked by this intimacy
between a respected Deva and the Asuras, and since the latter
were there and could not be done away with, they made Varuna a
jailer, who had to watch the fettered Asuras.15
The truth of the story clearly was, although this is never overtly
stated, that Varuna, even after he had become a Deva, continued
to entertain secret relations with his banished brothers. It reminds
us of a Vedic tale about a priest who publicly officiated for the
Devas but secretly was the priest of the Asuras.16

I now come to the mam point of this mythology. The Asuras had
been driven away but not annihilated. They were not part of the
cosmos but continued to exist beyond the pale, as a constant
menace to the existence and coherence of the ordered world.
One need not be a psychologist to presume that their banishment
may have meant, in Freudian terminology, a repression and that, if
repression %herewas, the inevitable consequence must have been
anguish on the part of the cosmos. This conclusion would, however,
ignore the fact that their banishment was only temporary. At certain
intervals-I think the Rigvedic evidence allows us to say: a t the
beginning of every new year-the war between Asuras and Devas
was renewed. On the social level i t was reenacted by contests,
which may be interpreted as reiterations of the cosmic strife and
as a ritualization of human aggressiveness.
l4 Also in the MahiibTWta, Varuna is located in the netherworld; see 111.
160. 10; V. 96. 5 ; and V. 106. 12. For the Asuras attending him, see 11. 9. 15-17.
Varuna rules over them and protects them; cf. VIII. 45.32, Bombay ed., pcllayann
asuriin (against 30. 77, critical ed.), and XII. 4497, Calcutta ed., apdm rdjye
.
'suriiniim ca . .prabhum (against XII. 122. 29, critical ed.). Although it cannot
be proved that these are the authentic readings, the occurrence of .variant readings
may in these cases be significant.
l6 Varuna's world is a refuge for V~tra'sallies; see 111. 98. 3, 99. 21, 100. 1,
101. 7 ff. (cf. I. 17.28). V a r w a is the jailer of the fettercd Asuras; see V. 120.44 ff.;
111. 42. 6, 27-28. Cf. I. 19. 6, asurdndm ca bandhanam (said of the ocean).
l6 For the tale of ViBvarBpa, see Taittiriya Sariahitli 11. 5. 1; and Jaiminiya
B r i i h m a ~11. 153-57.
 However, I will limit myself to the theological problems and
leave a discussion of the social aspects of these potlatch-like cere-
monies to those who are better qualified. I n this context I must
insert a remark on a point which I will not stress but which cannot
be ignored, because i t seems obviously to follow as a conclusion
from the preceding discussion. This point is Varuna's position
during the annual period of crisis, which apparently formed a trans-
ition to the new year. If a t this time the Asuras returned on earth
and renewed their war with the Devas and if, a t the end of this
period, the world order was to be restored again, just as i t had been
established in the beginning of time, one conclusion would seem
inevitable: in that case Varuna was during these days once more
the adversary of Indra and had to be reconciled again. I n other
words, his secret conspiracy with the Asuras, traces of which we
have found in later literature, must for a short while have turned
into an open alliance. The reason why i t is no use dwelling on this
point is that the Rigveda provides no evidence in support of this
conclusion. It has been suggested that this is simply due to the
circumstance that the majority of the Rigvedic hymns had been
composed for New Year ceremonies, during which Varuna was
particularly inauspicious and dangerous. This would explain why
all that concerned Varuna's darker aspects was tabooed in these
hymns. Although I Selieve that this is materially true, i t is clear
that intentional reticence can seldom be proved and that there is
no point in arguing about things which are not explicitly said.
I now return to the problem of the banished Asuras. Rather
than indulging in psychological speculations of our own making,
we can state the basic problems of Vedic religion in terms which
the poets themselves used.
They had, indeed, two words which perfectly well expressed
their reflections on existence. One word is sdt, literally "being, that
which is." It is used with reference to the phenomenal world, the
ordered cosmos. Its opposite is h a t "the nonbeing," which denotes
the world of unformed matter, the undifferentiated state. The
cosmogonic myth describes their relation as one of successive
states in such phrases as "In the first period of the Devas sdt was
born of cisat,"17 or, "The seers, searching with insight in their
hearts, found the origin of sdt in hat."18

l7 Rigveda X. 72. 2. The creation and every renewal of the world must have
been regarded as a process of shaping the unfonned, which made the assistance
of the Asurns indispensable.
Ibid., X. 129. 4, satd bkndhum drsati nir azdndan.
 In the light of the interpretation here given, the question might
be raised to what extent the Vedic poets were thinking in terms
of successive states and how far they were referring to a logical
relation, the latter state necessarily presupposing the former. In
fact, the word which I translated as "origin" has various connota-
tions and might also be interpreted as "relationship."
Whatever a Vedic poet might have thought of our problems of
interpretation, the question would be a legitimate one, for the
myth did not merely imply a succession of two different states. As
a matter of fact, the emergence of an ordered cosmos did not put
an end to the existence of the world of h a t . In fact, the problem
of the Asuras here recurs clothed in different terms. Asat, the
primordial world of chaos, was not entirely replaced by the cosmos
but continued to exist on the fringes of this world and as a per-
petual menace to the latter's existence.
When I ~ e n t u r eto translate this philosophy in more modern
terms of my own, I should say that the world order, as the Vedic
Indians saw it, was a precarious balance between the powers of
cosmos and chaos, and that this world was only part of a much
wider universe, which also comprised the nonindividualized world
of unformed matter.

v1
The Indian Genesis started with a state described in such phrases
as "In the beginning all this was nothing but the waters" and
"There was no h a t nor s$t."lg The next stage was the primordial
world of h a t . This finally became our world, described as a state
of tension and struggle, with the annual intrusion of the repressed
h a t and the Asuras into the established order, which is the world
of sdt and the Devas.
Still, conflic-ii is not the last word that the Vedic myth has to say
about the nature of the universe. I think the following outline is
sufficiently well founded to be presented here.
I n a philosophical hymn of the Rigveda we read the words " h a t
and s$t in the highest heaven.''20 The term "in the highest heaven''
is well known in these hymns. It sometimes clearly refers to a
place which transcends the dualism of this world, for instance,
when Indra is said to hold heaven and earth in the highest Leaven.
It is no doubt identical with Vi~nu'sthird or highest step, which
l* See the laiia&ya-hymn, R@w& X . 129. 1, 3.
a0 R i g v h X. 5.7, &amca a& ca param'vyuydman ddksagla jdnmann &d&rupi-sthe.
is hidden from the mortal eye. Many years ago I tried to demon-
strate that in the Veda Visnu, far from being a subordinate
assistant of Indra, must have been a central figure, of greater
importance than Indra himself.21 While Visnu's first two steps
express his relationship with the two opposed parts of the cosmos,
his third step corresponds to a transcendental world in whieh the
two conflicting parties are united in an all-encompassing totality.
In this light we can view the poet's words about "Asat aqd scit in
the highest heaven." They point to a world whieh transcends the
cosmic antithesis and in whieh scit and cisat have been reconciled in
the synthesis of an all-embracing unity. (See fig. 2.) In this respect
Visnu must have been, since the earliest time, a higher god than
both V a r u ~ aand Indra, as he transcended the dualism which
they i m p e r ~ o n a t e d . ~ ~
The "highest heaven" is a specific term of the poetical language
of the Rigveda. Later theological textBsrefer to a "third heaven."
Since they never mention a first or a seebnd heaven and since the
number three traditionally expresses the concept of totality, there
can be no doubt that the terms "highest heaven" and "third
heaven" denote the same idea. The concept of a transcendental
world explains some difficult problems of Vedic mythology, whieh
are, however, too technical to detain us here.
Rather than dwelling on Viynu's highest heaven in the Rigveda,
I will conclude with a few words about the development of Indian
See Indological Studies i n Honor of W . Norman Brown (New Haven, Corn,
1962), pp. 137-51.
As is apparent from fig. 2, there is a double dichotomy: the contrast between
&at and scit and, within the ordered cosmos, that between netherworld and upper
world. I n Vedic texts there is evidence of Vignu representing the totality of the
cosmos. If the reconstruction and especially the interpretation of the nether-
world as part of the cosmos but somehow related to the world of &at is correct,
the question arises of whether Vignu from the outset comprised sdt and &at. As
for the primordial waters, they were apparently incorporated in the cosmos a t
the moment of Varwa's transition to the Devas, but in the annual periods of
crisis, with the return of the powers of chaos, they must again have become the
world of &at. At that moment Vignu transcended the dualism of Asuras and Devas
( h a t and S&),just as he had transcended it in the beginning, when the dual cosmos
arose (see n. 21 above). I n the later eosmogonical myth of the churning of the ocean,
as told in the MahEbhiirata, Vignu also stands above the two parties of the Asuras
and Devas. About the tenth century A.D., when this myth was retold in the
Bluigauata Puriina, the gods were represented as adoring Vignu in the words "Thou
alone a r t both sat and asat, the dual world and that whieh transcends dualism"
(VIII. 12. 8, ekas tvam eua sadasad dvayam aduayarfa ca). Although these quotations
refer to the cosmogony, during whichvignu's transcendental aspect must have been
particularly manifest, it was a n essential part of his nature. As for the parallelism
between the "third heaven" and the primordial world in the briihmanas (see, e.g.,
Asiatische S t d e n 25 (1971): 94), this is due to the fact that in the latter the
dualism did not yet exist, while the former had transcended it. Hence it is that
the third heaven could be substituted for the primordial world as the place from
whioh Soma was fetched.
 Totality, V i y , etc.

chaos Ordered cosmos

-
sit, Devas

religion in post-Vedic times. No doubt, foreign, non-.Aryan

religions had a great influence in this period. However, when con-
sidered from a genexal point of view, their influence appears to
have been a marginal phenomenon. The basic line of development
can best be illustrated by starting from a belief which, although
only attested in later sources, may well date from an early period.
According to this belief, Viqtp is for eight months of every year in
the upper world. In this period he takes part, as a Deva, in the
processes of the cosmos. I n the remaining four months, however,
he is in the ne~herworld, where, reclining on the world serpent,
he sleeps on the surface of the subterranean waters. Incidentally,
there is a close connection between sleep and the netherworld.
One has to know this belief, which has survived into modern
times, in order to understand how in the classical period this
annual process has come to be projected, in gigantic proportions,
onto the scale of a world year. Just as in Vedic belief the world
was a t the end of every year menaced by the intrusion of the
powers of chaos, so in this later belief, a t the end of a world period,
this world is doomed to be annihilated and to return to its primeval
state of chaos.
What remains after the world with all its gods has passed away
is the cosmic waters and on them, sleeping on his world serpent,
god Viqnu, who comprises within himself a potential new world
with reborn gods.23
a3 See, e.g., Mkkandeya's vision, MahZi6hZirata 111. 186. 92 ff. The idea that as
a result of the return to the undifferentiated state only the waters remain is in
conformance with older conceptions. Cf. also, e.g., BhGratiya N@yd&tra 22 (20). 2
"When Lord Acyuta, after destroying the worlds by his supernatural power and
.. ."
changing the universe into one ocean, lay sleeping on the serpent aa his couch.
 In this vision the eternal war between chaos and cosmos has
not vanished, but the Vedic notion of periodical fights between
Asuras and Devas-and the concomitant ritual of annual con-
tesbs-has here ceded to the concept of a transcendental harmony
-a harmony in which the conflicts and antagonisms of our
existence have been reconciled in the figure of one God who sur-
vives all vicissitudes of a transient world.
2. THE GOLDEN GERM

In his magnum opus The Golden Germ, Alz Introduction to Indian Symbolism,"
F.D.K. Bosch has presented us with the ripe fruit of a long life devoted to the
art and culture of ancient India and Indonesia. Like a veritable kalpataru he
here pours out the results of a research that is as original as it is fascinating.
Any attempt to summarize its contents must necessarily confine itself to some
outlines and cannot do full justice to the book.
I n the introduction the author points out that the task so far fulfilled by
Indian archaeology was one of describing and classifying the data. Thereby
archaeology is on a par with ethnography and should rather be styled
'archaeography', which would leave the task of a more profound interpre-
tation of the data to a new type of 'archaeology'.
Remarkable transitional forms between the decorative motifs of the parvan
(the joint in the so-called 'stalk' of the lotus) and the makara-head (see figs. i &
ii) induced Bosch to identify the arch (toralta), which often connects two
makara-heads with the lotus "stalk", and the top-piece in the shape of a kiila-
head with the lotus "roe?"' Starting from these forms he then formulates his
basic hypothesis that, in contrast to Brandes's idea that the vegetal motifs
are a stylized development of animal motifs, the lotus vegetation has rather
been the origin of this decorative art with all its wealth of variations. Bosch
shows, indeed, that the stalk-motif is not only a decorative element, as has
generally been supposed, but rather (e.g.,in the reliefs ~f Bharhut and Sanchi)
the central symbol, the source of Life which nourishes all that exists, men and
animals being represented as connected with it. Bosch then associates this
mystical source of ~ i f with
e the Vedic myth of the golden germ (hirapyagar-
bha), which is born in the waters. Elsewhere, the One to which all beings are
attached (accordingly the Cosmic pillar, stambha)is said to be erected in this
primordial germ. From this Bosch concludes that the golden germ is the

*(Editox's footnote:The English edition of this work appears in Indo-Irantan Monographs, Vol.
11, published by Mouton, T h e Hague, 1960. This essay appears here in English for the first time.
I t is a translation, with only a few minor changes, of a review of the original Dutch edition
published under the title De Gouden Ktem, Inletdtng tn de Indtsche Symboliek, Edition N.V. Uitg,
Mij Elsevier, Amsterdam-Brussels, 1948. Page references in this translation refer to the English
edition of the book.)
 k
Fig. l. l m

Fig. a and O Bharhut; c chal?+ Kalasan; d Barabudur; e Sanchi; f Barabudur;

g and h chandi Kalasan; i Barabudur; j Bharhut; k Siinchi; l and m Bharhut.
 germ of that tree, and with this notion he connects the myths (occurring in
much later times) of a lotus which rises from a navel. I n this way he finally
equates the lotus "root" with hirapyaga~bha.
On the other hand, however, the myth of the golden germ is unmistakably
connected with the mysterious birth of Agni (fire) in the primeval water,
whereas the lotus is more directly related to the water of Life. I n this germ,
accordingly, both elements (which may be called Agni-Soma in accordance
with ritual terminology) are contained. This fact Bosch explains by assumillg
that a male element, a fiery breath of life, has entered the female waters.
T h e equivalent of this breath of life he finds in the ancient Vedic idea that a
fig-tree (duattha) represents the cosmic tree. I t is sometimes referred to as an
inverted tree. This he explains, mythically, as representing the descent of
the breath of life, and naturalistically, as symbolizing the dvattha growing as
a parasite on the stems of other trees.
The (non-inverted)fig-tree turns out to be also an ever-recurring motif in
this art. Its root is represented as an antefix (which Bosch terms bra.nlzmatniila),
and shows a close resemblance to the lotus-root (padnmmiila.).This antefis,
however, is usually itself rooted in the stem of another tree with a lotus-like
vegetation. I n the decorative element the two trees, although mergecl into a
unity, can still be clearly clistinguished as the components.
The main part of the Imok deals with the different shapes in which the
"stalks", the stem and the nziifus can be found in art and in other domains of
Indian civilization. Thus the system of classification is derived from the
1'4 stem of the tree, which as the centre is the fixed landmark of the whole

l
i
system. On the microcosmic level Bosch finds this basic orientation in the
mythic anatomy of the yoga, while macrocosmic;tlly, Mount Mcru, as a cosmic
pivot, is explained as a replica of the pnclmcmiila. Many other objects are
explained as symbolizing the lotus "root", such as the flower-vase. the "pot-
bellied" persons, the conch, the jewel, etc., whereas the kris and other objects
are interpreted as standing for the stalk or the bi-anch.
Sin~ilarly, various solar figures are taken as symbols of the root of the fig
tree, while special attention is given to thcgr~~tz~:~gun(kekqon)as standing for
the stem of the tree. After a study of the stem as represented in the ht~nlan
figure (including the origin of the statue of the Buddha!) a separate chapter is
devoted to the cult image as a form of the wishing-tree.1 I n the concluding
chapter the author discusses some basic questions, summarising his v i e w on
the religious values of tree-symbolism, and offering some surprising new
vistas on the literature. For instance, he explains how terms like palann and
k@@aof the epics, and the typically Indian frame story. the srazla!tapltaln, are
Here the author puts an end to his argumentation, because somewhere there lntlst be a limit.
One would have liked to find here some other manifestations of the tree-for instance, the juyaru
of the Indian theatre, which was brought on the stage and worshipped, thereby marking the spot
as the sacred world (see otherwise Gonda, Actu On'entaliu, XIX, p. 367 ff.); the ketu of the war
chariot (cf. Rgvedu IV. 24. 10),and the pillar of the Vedic house.
rooted in the syn~bolismof the tree.
AS ~ o s c hhas convincingly clemonstratecl, the influence of the tree and the
lotus in Indian culture has been profouncler than generally aclinowlctlge~l
up to now. Although he has not succeeded in proving all points 5t.llicll h~
himself considers to be of centrnl importance. ancl although in some details
his exposition is, quite understandably, open to criticism, it must be
emphasized at once that the importance of the fresh insights gained and of
the new problems stated far outweigh the critical comments that might be
made. I n many respects, indeed, the later have to start from the new findings
this book offers.
The predominant impression that remains is that Bosch has not only
thrown fresh light on many aspects of Indian culture but has led the reader to
tlfle very heart of the Old Inclian conception of life and the woi-Id. It is not
surl1risii~g.therefore, that from this centre such an illuminating light shoulcl
radiate in a11 directions.
I f ?in spite of these precloininating feelings of aclmiratiort, one is to voice a
reservation on certain points, this is mainly due to the fact that the aiithor
has apparently not sttcceedecl in entircly emcling the danger that threatens
especially the explorer of virgin soil: namely that of not keeping the necessary
distance to his subject. as a result of which he is no longer able to view it in the
right perspective. (But then, does not a subjective one-sidedness often seem
to be a necessary preconclition for arriving at new ideas? And shoulcl we not
acquiesce in this as a necessary part of the in<yci^i11which we live?)
This is the only possible explanation for the fact that Bosch has come to
consicler as the main object of his work the defence of a disputable and one-
sided theory, thereby consiclcrably weakening the book as a whole. In the
very beginning he confronts his reader with the clilemma: is the vegetal
decorative motif a '"stylization" of the aninlal 01- did a reverse development
take place?--thereby entirely ignoring the third possible solution. viz. that
both the plant and the animal are old motifs~anci that their mutual penetration
was meant to intensify their effect as religious symbols. Here. at a decisive
point in the line of his argmlmlt the author !'ails to recognize the real
problem.
A second point that is not touched upon altl-tor~gllit is of vital interest for
Eosch's argument is the Veclic myth. H e was right in associating the vegetal
symbolism with the Rgvedic myth of hirn?~~agarb&z, which gave its name to
the book. T h e lotus, however, is practically unknown in the $ p e d n , and the
Vedic Aryans certainly had nolotus-myth.This need not inean that this myth
was of later origin: it obviously originatecl in ;~tttochtl~onous circles ancl only
a later stage of the process of Inclianization can account for its being grafted
on the Vedic myth. This, however, is not relevant in this context. The essential
thing is that Bosch's ahistoric approach (which in other parts of his work was
motivated ancl justified by the nature of his material) here blinclecl hi111 to the
fact that the Vedic myth of the golden germ, accordingly, must have existed
indq~endcntlyfrom the lotus-symbol. It cannot. therefore. be derived from
the vegetal motif. Similar objections may be raised to the interpretation of
tlie at~zrta-bowl,in which Eoscl~.with the visual approach of the archaeologist,
sees a developinent froin the lotus-root, and of many other symbols. These
points .ir ill be cliscussecl in the lollowing pages.
Here it may only he asLeet if the tnakara-symbol can really be consiclered
aciecjuately explaineel by the supposition that the Indian "saw" in the node of
the lotus-plant a monster-head with gaping mouth (p. 34). Leaving aside the
basic question whether religious symhols are actually derived from visual
i~npressionsof nature in the way Bosch solnetimes seems to assume2-as for
edic seer, he receives his revelations in a different way, cf. R S X . 129-
one may still call to millcl the following fact. While vegetal synlbols for the
bowl with the elixir of life (soma-anzyta)are, as far as I can see, unknown in the
&vecla, the animal syinbol co~tlclbe connected with the snake o r demon
(Vrtra, Vala, Srtsi!a), who here guards the fire and the sonta on (or, i n ) the
n~ountain,and who must first be killecl before Agni and Soina can be liberated
to spread their blessings in this world.
I11 later priestly speculations it is soruetimes not the mountain but Vrtra
himself in w h o ~ nthese two are said to be ~ o n f i n e dSoinetiines
.~ Soma is even
iclentified with V p r a . q t is not suggested here that the nzakara-motif is a
straight-line development from the mythological figure Vrtra. This ii; a
problem quite apart, which it is up to the archaeologist to decide. Still, the
preceding considerations justify, I think, the conclusion that the Aryan reli-
gious tradition leaves the possibility open that animal figures may have been
mythological equivalents of the lotus root.5
AGNI-SOMA AND T H E SYSTEM OF CLASSIFICATION
Ode of the points which Bosch, it seems to me, has not been able fully to clari-
fy is the place which Agni and Soma occupy in the classification. H e rightly
2The ways in which the author expresses himself on pages 34f and 46 reflect different ideas and
give the i~npressionthat his opinion 011 this point has not yet fully crystalized.
3 E.g. Katc;?fakiBrZhnza?ta 111. 6, XV. 2, Sht. Br. 1.6.3.13ff. Cf.girir vai vytro,Mairr. Sarich. IV. 5.1.
T h i s has often been discussed. See e.g., Hubert et Maus,Lc sacrifice, p. 129, S. L& ,I'. La doctrine
du sacrifce dans les brahtnanas, p. 168, n . 5. Baschardt, Vrtra, Det Rituelle daernondrab i den vediskc
Sornakulf,pp. 113ff.
If it is true (as Bosch has argued with great plausibility) that the pilasters which in Hindu-
Javanese art often support the two makars at the end5 of the toraga-posts must be explained as the
mountains of sunrise and sunset (p.146),the parallelism is strongly marked here between makara
as a demon of the u ~ ~ d e r w o r from
ld wltich the sun rises in the morning, sinking into it again at
night, and the cosmogonic Vp-tra figure. Compare also the all devouring heads of demons of K8k
p6 (plate 49c), and the Javanese representations of a small lion o r bird (both of them sun symbols)
in the open makara-mouth, and, on the other hand, e.g. the Vedic myth of Indra, who steals the
amrta of the Asuras from the mouth of S u p i . That the ntakara is based on old Vedic religion may
also be proved by the kidang motif of east Java (p.lSl.), which occurs as a variant of the makara and
is also an underworld symbol. T h e kidang has also a Vedic forefather in S u p p , the demoniac
gazelle of the Ilgveda (Hillebrandt, Vedische ~Mythologie112,p. 260, n. l.) [but see now Indo-Irattian
Journal, 20, p. 1001.
attributes a great importance to the opposition be~weenAgni and Soina. IIe
inclines, however, to represent this as a cosmic contrast, as on p. 6 1, rasa-
Soina as the essence of the Waters is contrasted with gni as the fiery essence
of the creative breath, and on p. 65, where there i\ ieference to "Agni
generated by the creative breath" (who is said to impregnate the female
element of the waters). This is, I am afraid, a 111isrel)resentation of the ancient
co~xxptionof the world. T h e reader is here facecl with the difficulty that the
of the myth evoked by the author is oniy partly supported by
at is incidentally said on p. 86 about the sketch of the classifica-
tion system, to the effect that this is partly l~ypothetical,is also true oi the
reconstruction of the myth. Since the personal interpolatiom by the authoi
are not indicated, it is incumbent on this reviewer briefly to formulate the
outlines of the Vedic myth of Agni-Sorna as he sees it.
I n the primeval water (which, in essence, is identical with Arnyta-Soma)
Agni is born. T h e mystery of Agni's birth is unquestionably the central
motif of the Indo-Iranian mythology. As Agni in the shape of thosun ascenci5
the sky, the cosinos arises in its dualism of upperworld and uilderworld and
with the cosmic tree as the centre and supporting pillar. Agni's ascension
can be viewed either in a monistic way, as the ascension of the Sun horse
(Dadhikrsan, L7ccaz~tSravas), o r in a ciualistic vision, the primordial hill in
which Agni and Soma were hidden being split open from the outside (by
Indra). Agni is called the offspring of I-Ieaven ancl Earth,although the two
constituted an undifferentiated primeval world before his arising and, conse-
quently, there was no separate Heaven. ythologically this is on a par with
the fact that Zndra is born as a celestial god prematurely (that is, before
heaven and earth have been separated): it is not without reason that the tale
of his birth re~nainsvague and the figure of his mother caunot be grasped.
I n fact, not until the sun is liberated from the powers of the snake of the
underworld, can Indra "prop u p the sky (from the earth)". I n other word5
only then does he erect the cosmic tree whose cult name therefore i\ mdm-
dhvaja. I n a monistic view the sun itself is the simrrbha ($gvecln 11'. 23.5).
If one now looks at the picture of the myth as offered b) Bosch, the main
point6 that strikes the reader is that nowhere, as far as one can see, is there
reference to a descending fiery essence of creative breath. Agni's birth is a
mystery but since he, after being born in the underworld, ascend5 to heaven
in the shape of the sun, and, in his function of sacrificial fire, carries the
(soma)offering u p to the gods, he takes part in both worlds. In a sense he is a
"mediator". like the Persian Mithra whose character has some relationship
to his.
The whole cosmogony is, in this view, characterized by the notion of an
V am not entering upon minor points, e.g. that Prajapati (in spite of certain myths in the
Brahmanas, in which the authors are struggling with the problem of how themu1tiplic;ty has arisen
from unity) is nevertheless very clearly a totality god comprising both worlda (=KLSpap). The
figure of Vac also seems misrepreseilted to me (cf. the identification o f \'$c with \'iiiaia in ;ilnttr. S.
111. 7 . 3 ) .
emergence from an undifferentiated primeval world. It is not quite clear
what may have given Bosch the idea of a male creative breath, consisting of
pure lightand intelligence (e.g.. pp. 51,59).7It 1112y be conjectured that this is
mainly based on the two words &2:1 nv;ta'~iz in Rgvecln X.129.2, which he
translate5 "(the One) breathed breathless" (p. 51).The meaning of the second
word is unccrtain.8 Anyway, the undifferentiatecl primeval Unity has
obviously no connection with the rGle that is here assigned to a creative breath
that descends from h e a v e i ~As
. ~ it seems to me, it is completely at variance
with the general gist of X. 129. which consistently speaks in unmistakable
words of clam "the One", to postulate nevertheless a duality, and a "union of
these two elements" (p. 52).This is substituting a rationalistic explanation for
what to the poet's mind (whose ideas are as co~isistentlymonistic as those of
the poet of X.82!) was a mystery. ~rhichhe deliberately abstained from
explaining.
If, however, the idea that the descenclitig creative breath was an element of
the ancient myth can no longer be maintained, the basis on which Bosch (p.
65) founds his interpretation of the ielestial fig tree is taken away. On the
other hand, as the reality of the motil ol the double tree cannot be doubted,
an explanation for it will have to be sought in a different direction. A
suggestion will be found at the end of this review-article.
There is a second point which is closely connected with the preceding
one. As was noted above, Bosch finally comes to identify the contrast between
Agni and Soma with that of upperworld (with a male character) and under-
world (conceived as a female being); cf., e.g., p. 65. Quite apart from its
many contradictions,lO this explanation can hardly be correct. It is sufficient
to point out that both Soma and Agni were liberated by Indra's creative act
and appear in the upperworld, and that Soma, too, is thought of as ascending
to heaven (ritually by means of the sacrificial fire, mythically in all likelihood
through the stem of the cosmic tree). A confirmation of our conclusion that
both Soma and Agni belong to the upper world is bound in the Vedic system
of classification, in which Soma is located in the North, and Agni in the
East. Both, accordingly, are connected with the upper moiety of the cosmos.
Besides, "King Soma" is a typically male figure, who contrasts with the
female surci (a beverage connected with the underworld).

CLASSlPICATION A N D T H E MAFIABHARATA

Incidentally, a reference was made above to the system of classification.

Owing to the premature death of Kasten Rilnnow, the book he was writing bout Vjyu and the
p a p a doctrine never appeared. As very little research has been done in this field, it is impossible
to give a more definitive judgment on the possible r61e of VByu in the cosmogony.
Cf. Neisser, Ztrm Worferb. des RV. I , p. 138.
9There is n o proof that Bosch was thinking of passages such asJaim. Br. 111,pp. 359-360adltar1at
pr@atn akuruta. I n my opinion they cannot simply be used as mythological material, but it is not
possible to enter here upon this problem.
'O Bosch, too, discusses some complications, p. 143.
Since, however, this is not only expressly connected with and derived fsom
the tree motif (pp. 83-92), which is considered primary, it is necessary to go
somewhat more deeply into this problem in order to decide whether the
claim that the whole system of cosmic classification is derived from the tree
motif is justified. Similar claims of the priority of the tribal organization, to
which Bosch raises objections (p. 87) will here be left out of consideration.
On the other hand Bosch's comments on the MahabhZrata, which illustrate
his view, are an occasion to discuss this important issue. It may be useful to
preface some brief remarks on the way the classification works. For this
purpose we shall confine ourselves to Varuna-Mitra.
For Vedic man the cosmos is primarily divided into upperworld and
underworld, which are projected on the horizontal plane of the classification
in North and East and in West and South respectively (Maitr.Samh. IV. 1.10).
Within this dichotomy, however, the process of dividing is continued, as is
manifest in the contrast between Soma-North and Agni-East. This contrast
is also expressed in the association of Agni with right-hand and of soma with
left-hand, which must be an inheritance from Indo-Iranian prehistory. I n
the Avesta, indeed, Haoma is associated with the left eye (Yasna X1. 4 and 5).
A similar contrast must have existed between West and South. From the
'Jeda to late Hindu sources, the West, where the sun in the evening sinks
back into the primeval waters, has been assigned with remarkable consistency
to the Adityas, the "older" gods of the primeval world. Of these Varuva and
Mjtra are the two leaders. However, here again the dichotomy is continued
(in the third degree). While Varuna stands for night, death, the setting sun,
Mitra is associated with day, victory over death, the rising sun. It may be
said, therefore, even though this is not expressly stated in the texts, that the
contrariety between Varuna and Mitra, which is the expression of a demonic
ambiguity that belongs to the essence of the deity of the underworld, causes
Varuna and Mitra, when considered by themselves, to be opposed to each other
like West to East. Here, in his connection with the rising sun (Agni), is the
point of contact between itra and Indra (who in the classification is assign-
ed to the East). This also accounts for the Inch-features that characterize the
Persian Mithra.
This process of repeated partitions results, accordingly, not in a single
system of classification, but in a series of systems, one including the other. At
every level two groups or entities crystallize. In the main fourfold division
(N.E.S.W.) the Adityas (in this case : Varuna and Mitra) are one solidary
group, but when taken apart they are each other's absolute opposite. What is
Mitra's is not Varuna's (Sat. Br. 111.2.4.18). It therefore depends on the delimi-
tation of the "field" which the two opposite entities will be."
It may be noted in passing that what we read here from the Vedic data agrees more than
superficially with the principles of structural linguistics. (According to recent information the
structural linguist Roman Jakobson has been working for some time on a "structural" explanation
of myths). One may shudder at the thought of a future in which mythology would have become a
 It is in this conteat that the dramatic fight of the Great Epic will first be
cor\sidered more closely. osch's .c ery important comments on the P+davas
anci Karlravas confirin throughout the main thesis of Held's doctoral thesis
Tlie AhhZbhZrata Particularly striking is the $wallelism of the epic with the
olil myth of IiadrG and VinatB, which is situated at the beginning of the
creation (devnyuga).XaSyapa (tortoise), a typical representative of the totality,
has two spouses. One of these, Kadrii, has a hundred nQas for sons, the
inata, has only two sons, Garuda and Aruna.
The myth clearly represent5 the strife between the two cosmic moieties as
tile result of a contest bet\+eenthe two inythic primordial mothers. It is further
striking feature of Kadr5 (Aditi),that in the SuparpZkhyZna she is said to
be one-eyed (ka!~n). This characteristic, which is the very point in the contest
in sl~alp-sighteclnessbetween her and VinatH, is no boubt an old trait. A
parallel in the epic is that not only the father of the hundred Ka~lravas(who
bears the name of a snake-god, DhrtarZ~irZ)is blind but also the mother
~ a n d h a 6symbolically blinds herself at her marriage by bandaging her
eyes. This deed, hardly motivated at all in the epic (cf. Mhbh. I. 103. 12)can
on15 be rmderstood if G5ndhar? is considered equivalent to Kadrii.12 In that
case her blindness was a mythological necessity which was no longer uncler-
stood bv later poets.
ar is fought in Kuruksetra, near the S'amunS and the Ganges.
osch's bright ideas that these t~7orivers represent the cosmic
orientation right versus left (p. 89). However, his supposition [in Dutch edi-
tion only-Ed.] that the two rivers synlbolize the orbit of the sun between
sun-rise ancl sun-set seems less forunate. Since in late texts the "right-hand"
YamunZ is called the daughter of the sun (tapa~znduhztal;etc.) it can hardly
be clue to mere coincidence that the "left-hand" Ganges is said to flow down
from the moon on diva's forehead. Now it is striking that the older city of
the Kauravas lies on the Ganges (and thus is associated with the notions
"moon" and "left") and that it is called the city of Elephants (Hktinapura,
HastinHpura), the elephant being a typical aninlal of the underworld in
Indian mythology. Nor can it be coincidental that the younger city of the
Pandavas, viz. Indraprastha, is situated on the bank of the Yamun'i, and that
it IS named after the protagonist of the "celestial phratry", viz. Indraprastha.
From this, Bosch convincingly concludes that Kuruksetra, and in a wider
sense Madhyadeia, "the middle region" (which presupposes a five-fold

structuralistic toy in the hands of systen~atists.On the other l ~ a n done should seriously consider
wlleiher a structure-imposing activityof the human mind has not put its mark on the form of the
myth as well as on that of the language. That such a structure in itself only provides a skeleton, a
frame of thought, and does not tell anything about the religious life within this frame is as certain
as it is probable that a knowledge of this structure can save us from unnecessary errings of our
plmutasy in the jungle of mythology.
12 Tile ~ h m o m e n o nof underworld characters having one eye or being blind is anything but
1940,
clear. I>uni@zil'sremarks on this point do not get to the root, it seems to me. (cf.Mitra-Vaia,.u~y,
1YW '
division of &-y&arta)was looked upon as the cosmic centre. It is interesting
to note in this connection that here lay the holy places Er~ha-t&ha (111.
81.15) and ~okoddhzra-&ha (111. 81.37). Tl11s ' means that here the cosmic
boar raised the first earth from the primeval waters and that, accordingly, it
was thought to be the earth's navel, the primorclial hill.
It is in the cosmic centre that the battle between the two parties is fought.
By this very situation it can be recognized as a replica of the fight between
Gods and Asuras on the occasion of the churning of the urn?-ita.13Just as the
P@@avasare here the right-hand party, and the Kauravas the left-hand one,
so PrajLpati creates the Gods on his right hand and the Asuras on his
left.14 In this connection the role of V i ~ p y - K r s ~isaparticularly interesting.
In the Vedic classification Visnu is situated in the centre ancl in accordance
with the cyclic character of the year (both normal and cosmic) he is either in
the underworld (Visnu's sleep) or in the ~ p ~ e r w o r l cEle l . thus belongs to
both parts, which is expressed in later mythology by his two viihanus, viz,
&$a, the serpent of the underworld, and Garuda, the celestial bird. As for
Visnu's position during the amrtamanthana, the epic gives only s o h e vague
indications (Mhbh. I.16.14),but the reliefs of Angkor Vat provide a valuable
illustration. Here Visnu is represented in the very centre, standing between
the two parties and giving both his hands.15 It is a well-known fact that he
finally helps the gods to gain the victory by fraud. In the epic ICrpi's role
is similarly ambiguous. He, too, is related to both parties, but it is due to his
intervention that the Pzhdavas end by gaining the victory.16 The correctness
of Held's conclusion that in the epic Krsna must from the beginning have
had the function of V i s ~ ucannot,
, therefore, be doubted. (As for the relation
of this Krsna to the divine child Krsna, the god of the cow-herds of Mathur&
this is an entirely different matter. It can only be interpreted in terms of
mythological equivalence).
We may take a further step. Caland, in his very original lecture on "De
" ~ ~argued very plausibly that the relation
Incarnaties van den god W i s l ~ p uhas
of Arjuna to Krsna can be considered a replica of Indra's to Vi~r?u.l8Just
as it is said in the epic that that party will be victorious with which K r ~ q a

13 For urnyta "not dying = living", cf. Thieme, Untcrsuchungen zur Wortkundc und Auslegung des

Rigveda, p. 64, no.1.

14KKa;th.Samh. IX. 11.
15 Cf. the ritualistic interpretation in Sat. B , 111. 4.4.15: Agni is day, Soma is night, Vi?nu the

connecting link.
16The Bhagavadgita is spoken by KT-na "between the two armies" (1.21) and thus, "between the
parties".
l 7 Paper read before the Provinciaal Utrechtsch Genootscnap van Kunsten en Wetenschappen,
31 May 1927, p. 27.
18 Against this backgromld it may be possible to explain the fact that gold is mentioned with

remarhable frequency in connection with Arjul?a and his monkey-banner (e.g. Mhbh. I. 216.3. &
8).
sides fyatah fisttas tatojoyah 1.97.15 etc.), so we shall have to attribute a far
more fundamental importance to Vi~pu'sseemingly insignificant r6le in
Indra's V~tra-fightthan earlier Vedic scholars like Hillebrandt and Olden-
berg were ready to admit: the two-sidedness of Vis~!u's nature is apparently
the cleterininant factor which alone corild incline the balance in the combat of
the cosmic moieties.
Also here, however, the dichotomy continues. The number of five
PZndavas, corresponding to the two sons of Vinatii, already points to a further
partition within this group. The total conception can be visualized as astarnbha
on the top of which there is a horizontal wheel with four spokes (pointing to
the four points of the compass), whereas at the bottom there are the Kauravas
whose number is legion. Bosch has been aware of this problem of further
partition but has not arrived at a conclusion (p. 88). The question there asked
can be answered by pointing to the digvijaya ( M b h . II.23.9), where it is said
that A r j u ~ ~marches
a towards the North, ~ h T m atowards the East, Sahadeva
to the South and Nakda to the West, while Yudhighira remains in the town
"in good fortune". 1x1 other words, the three sons of the first wife Kunti are
found in the centre and in the North and East (the "divine" quarters), .cvhereas
the sons of M B ~ ; are in the half connected with the underworld.
Corresponding to the two sons of Vinatg, Garuda and Aruna, we here find
with a further partition the pairs ~117ma-Arju?aand Nakula-Sahadeva, who
have their counterparts in the social structure in respectively the kptnja
and uai9a.19 This division has a stereotyped character. In a modern litho-
graphic picture in the Bombay edition of the MahZbhiirata,20Yudhis~hirais
seen seated in the middle under a parasol and having a lotus in his hand,
w h i l e - ~ h h and
a A r j q a are seated at his right hand, and the twins at his left.
The connection with the social structure has particularly been pointed out
by Wikander (following the leacl of Dumkzil) who stressed the typically
passive role of Yudhisthira as dharnzamCja.21 The poets of the epic were still
fully conscious of this connection, as is shown by Mhbh. I. 1.80f., where the
specific qualities of the various PGdavas are mentioned which characterize
them as the prototypes of the various social classes. So Yudhighira is charac-
terized by purity, B h h t by firmness, Arjuqa by aggressive heroism, and the
twins by obedience and modesty.22 Such an inquiry may certainly be
important for Indo-Iranian p r e l ~ i s t o r y Still
. ~ ~ it is a secondary dichotomy,
19Therefore the idea offive branches (or phalli) resting in the stem (or vulva) (p. 89) cannot be
right. One shoulci note the memorial verse (1.1.66 and V.29.46) in which Yudhisthira is called
the tree, A r j u p the stem, ~ h h the
a branches and Nakula-Sahadeva the leaves and flowers. I
cannot see any more in it than the tree motif growing rank at a late date, especially because
something parallel is said of the Kauravas.
20 Frontispiece of the Sabhii-parvan (Ml~bh. 11).
21 Religion oclt Bibel V1 (1947),pp. 27-39. Cf. also G. Ihm&il, Jupiter-Mars-Quirinus IV, p. 37ff.

22That the heavenly fathers of the twins, the ASvins, were considered to be vaiiyas in the world of
the gods seems to follow from Mhbh. I. 1.81.
23 Cf. especial1 Wikander, "Sus le fonds indoiranien des &pop&es de la Perse et de 1 I'Inde", La
Y .
Nouvelle Clio no. 7 (ju~llet1950),pp. 310-329.
and as such it is of minor interest for our understanding of the epic than the
primary contrast between PanGavas and Kauravas, which corresponds to the
social dichoton~yi i ~ y a Szidra
: ($guedzc G y a viir-nu: dGa rtcitxa).
A justification for the prececling digression may be found, first, in the
unclainental importance of the problems coilcerning the Mahiibhlarata and,
,econcl, in the need to place the various views (like those of Wikander and
Held) in their relative positions. On the other hand, this more detailed picture
p t t c ; us in a position to ask where a1q7 indicatioil can be fouilcl in favour of the
theory that the tree motif was primary ancl determinant. As far as I can see,
the answer must be that it is possible to parallel the classification consistently
with the world-tree. This, however, is rather due to the fact that this cosmic
synlbol is a styli~eclform of the natural tree, and that only in this for111 it
corresponcls to the classificntion based upon a progressive partitioning. No
wonclei, then. that this classification can also be found in the symbol. 2 V t is
clear. h o w e \ ~ rthat
, the functional meaning of an opposition like North and
East (=divine)versus 'IVest and South (=demonic) callnot be explained fiom
the branching of a tree But presupposes a pre-existent complex of notions
that is fully inclepenclent of it.
Thi4. thcn. i\ sufficient to show that the symbol, ~chichis a meaningful
sign, cannot be prior to what it stands for. It may be added, in greater detail,
that the basic scheme of the AlalGbhZrata was found to be the representative
of the cosmic centre (Kttruk>etia)with the P g ~ d a v a sstanding on the right
sick and the Kauravas on the left. This was found to be an exact replica of the
situation during the atll!tnt7za~ztlza~ta,where Mount blanclara functioned as
the centre. There is nothing which co~llclcompel us to start from the tree as
the primary motif. Bosch. it is true, derives the cosmic mountain itself from
the lotus "root" (p. 96) but the idea of a primorclial hill must certainly have
been an Aryan notion (although not exclusively an Aryan one) whereas the
lotus-root, as pointed out above, seems to have entered Inclo-Aryan culture
as an indigenous influence. Any combinations which these equivalent symbols25
would show in art must be secondary. There is no reason, therefore, to consi-
der the possibility that one may have arisen from the other.

T H E INVERTED FIG-TREE

As was pointedout above, the fig-tree plays a prominent part in the explana-
2"lle same holds good for Bosch's analogous remarks about the world-order (p. 238): the
classification makes use of the tree-symbol.
2There seems to have been as close a relation between mountain and jar as between lotus plant
and lotus root. Just as in the branching of a tree, there seem to have been in the mythical world
conception, after Indra "split open" the mountain, two or four rivers streamingfrom the top of the
mountain (cf. GaiigZ-YamuG and the four world-oceans, Ath. V., Mhbh.). The "jar with four
openings" ( k d l a b ca'turbila) of Ath. V. XVIII. 4.30 can therefore stand as pars pro toto for the
mountain f the arnyta cask. Cf. also the "young maiden with four braids, who wraps herself in
veils" (KS. X. 114.3.) ?
iion which Bosch gives of the cloublc-treo motif. In order to give a clear idea
of the conlplications offered by the tests. the'trees mentioned in them should
first be listed :
(i)the common a s k t t h a , Ficus religiosa L, especially honoured as represent-
ing the world-tree and as such undoubtedly considered as rooted in the
primordial hill (although this cannot be proved from the old texts);
(ii) the inverted fig-tree, occasionally mentioned as a kind of divine
mystery;
(iii)the aivatthn, growing as an epiphyte on the stem of the hmT(or another
kind of tree), its roots hanging down from it;
(iu) the qagrodlm (banian tree) with aerial roots. This offers a pn'ori a
wide scope to various interpretations. Since publication of the original Dutch
edition of The Goldelt Germ, Eineneau has given us better information about
the epiphyte.26
Bosch starts his reasoning with the fig-tree rooted in heaven, ecjuating it to
the descending creative breath. H e then explains (iii)and (iv)as variants of
(ii),resulting from a supposed ambiguity of the texts (p.'75ff.). 1will confine
myself here to the starting-point, viz. the inverted fig-tree.
I n Katlz. Up. VI.1 (imitated in Bhag. G. 12. 1-2) an afvntthn is spoken of in
rather mysterious terms. Its root points upwards, its branches hang down, all
worlds rest in it, and it is brnhnzm. A similar description is found inMaitr. Up.
VI.1, but here the clirection of the branches is not mentioned. The same
image ancl the same terminology occur in the presumably somewhat older
Taitt. KT.I. II.5,27 though here the tree is not specifecl as an aivattha. 1lei-e
again a mystery is connected with it: whoever knows this tree is beyond the
reach of death (cf. esp. Rpedn 1.164.21). Finally, the earliest instance in
RBpeda 1.24.7: "In the unfathomable space King Varuna, of purified
intelligence, upholds the tree's stzcpa; they ( = t h e branches) stand directed
downwards. May their rays be fixed in us."
So far no satisfactory explanation has been give11.~8As Emeneau has rightly
pointed out, it is out of the question that these Indian poets should simply
have confused the as'vattha and the banian tree.2gIf an attempt is to be made to
continue the endless discussion about this subject, three points must be kept
in mind:
(i)I n all passages it is clearly the tree of life to which man has o r should
have a certain relation. From the beginnii8g this tree was probably considered
to be an duattha (cf.Rpeda. I. 164.22),but even if an afvatth~~was not identified
M.B. Eincneau, "The Strangling Figs in Sanskrit Literature:' Urtiv. of Califoniia Publication
iir Clnssica~P/rilok~g~S (1049), pp. 345-370.
27 Ibid., p . 367.

29 See summary in ibid., p. 366. The recent, thorough philological treatment by Thieme, Untemr-
ckungetr aw nhtikzozdc uttd Auslezung des Iiigreda (Halle, 10491, pp. 55-73, untoi-tuhately offers
little that is new. In my opinion Thieme did not penetrate to the core of the matter and has
misjudged that character of the tree, as well as of the two suparnas.
Thus Garbe. R. Otto, Geldner. and also Bosch, p. 70.
with it until the Kath. Up.90, this is not essential for the argumentation.
( 2 2 ) Everywhere it is a mystery that is hinted at. As this tree has so clearly

the character of an (esoteric) religious symbol, it is a p i o n doubtful that a

naturalistic explanation, wl~etheras a banian tree (Geldner. Bosch) or as a??
epiphyte (Eineneau, p. 367) is justified.
( m )In the R p d a it is Varuna who holds the tree, ancl this important fact
has generally been ignored.
V a r u ~ aand the other Asuras are the "older" race of gods, i.e. the gods of
that primordial world which precedes the partition into upper and nether
world.31 As such V a r u ~ adwells in the primeval water from which the earth
has mergecl and on ~vhichit rests. Though he withdraws as a god already in
the tenth book of the Ijgt~eda,traces of 111s former character must have been
preserved in the popular cult. As late as about the 11th century A D . Varu?in-
deva was the common term for a slab on a well in ChainbZ (in N.W. India),32
for this well was the entrance to the subterranean waters. As, however, am@
is the essence ( ~ a s nof
) these waters, the (subterranean)amrta jar, in which the
tree of life had supposedly been planted (as is testified by decorative art),
stood in Varuna's realm: rooted in the netherworld, it reachecl with its crown
into the highest heaven. In the daytime this realm was invisible to hunlan
eyes, but as in the system of classification the upper world was related to day
ancl the ~ietherworldto night, in India (just as elsewhere) the night sky as
heavenly ocean was equated to the primordial water.33 Thus it is said of
Varrtna that, while proceeding amidst the waters, he lookr dorun on the justice
and injustice of men (VI.49.3).Var~tna'srealm, in so far as it was located in
the sky. was actually ~onsicleretlan invertecl world. This is proved by the
word nzcma "pointed downwarcls" which is used for the branches of the tree
but also in connection with the cask, cf. &gveda V.85.3: "Varuqa has poured
out the cask with its rirn34 turned clownwards, over heaven and earth and
the intermediate space. Thereby the king sprinkles the barley." This
inexhaustible (VII. 72.10) cask is elsewhere called the "well-cask abounding
in water" (zitsam kdvandhnm ud&m, VIII. 7.10)35, and is undoubtedly the
same as the "golden well" (IX.107.4)in the sky, which can only refer to the

30 Eineneau, p. 367.
31 It is, naturally, not possible to give here any documentation nor to enter upon deiails in what
follows.
32 1. Ph. Vogel, Arltiquities of Chantba State, I . p. 177f. and passim.
33 Consequently there is no longer any sense in the old controversy about whether V a r u p
was originally the god of the water or a celestial god.
34 For this word see Wrackernagel-Debr~lnner, KZ. 67 (19421,p. 171 f .
35 It might be objected that the rain, which V a r u ~ p and the hlaruts pour from this cask, does
not only fall during the night. However, the Ganges does not only stream at night either, but the
sadlru who imitated its descent from heaven (Bosch,p. 187, no. 126)made it take place in the latter
part of the night until dawn (for mythologically the Ganges rises on the moon). Only at certain
moments does earthly reality answer to the ideal norm of myth.
n ~ o o n . ~ T limage
~ i s still lives on in the epic tale of the amyta-churning (the
day aspect ), in which the ocean that,must be chl~rneclis denoted as "a jar".$'
The veclic svena-myth can, in my opinion. be esplainecl in a similar way. In
this myth an eagle (later Garupa) steals the amt-tn from the sky, which has the
character of a netherworld because of the snake-nature of the Soma guardian
I(rscilzu ("the footless arcl1er").~8I11 the mnc. way the tall mast with a golden
fish or yellow bowl on top (Bosc1-1,p. 155f.)is the zndrndhzlalu, but in its nightly
aspect.
Therefore it seems likely that an explanation of the inverted fig-tree must
be sought in the night aspect of the cosmos. Rooted in the heavenly anz!?a-jar,
the it sends its beneficial vital force into the hearts of the faithful,
who are believecl to be connected with the branches of the tree.40
If this conclusion is correct, the symbol of the inverted fig-tree is based on
a conception, essentially clifferent from that of an as'vattha ancl a !a& I n that
case Gosch's reasoning (p. 75 f.), which tries to combine the two on the ground
of a supposecl ambiguity of the literary texts, cannot be right. Indeed one
sometimes finds the day and night aspect combined into one image (which
gives the image its maximum synlbolic power). This is, for instance, the case
when a pillar rests with its base on a trunrbha and bears an inverted ("songsang")
kzrnzbha on its top." The inverted form is then, however, an indispensable
requisite. As it is lacking in the clouble-tree motif, the inverted fig-tree,
together with the desceilcling creative breath, cannot be maintained, in my
opinion, as the basis of Bosch's reasoning.@
It shoulcl be noted in passing that the equivalence of the Vedic am!.ta-jar
with the non-Vedic lotus root does not justify deriving of the first syinbol
from the second ( p 1 2 1 ff.). When, further, Bosch derives the conception of

36 Cf. for this Ilillebr.andt, Ved. Myllro.2 I, p. 321f. mid p. 326 (about the golden cask in thepur
a,vodhi'Atlr. V.X. 2.31.). Thc identity of Am+ and Xgni, as stated by Bosch, only holds good if it
i, limited to Agni in its nightly aspect as moon.
37 Mlzbh. I. 15.12. Bombay edn., ~~tatlt>~ut&ir kalaSodudilt. bharus>wt! U I I I ~ ~ U latra
I I ~ and 16.32.. ksobtr!a-
&it knlas'al~sarvair (cf. IiS. VI. 6O.(i). The kalstZhas become the eniblem par excellence of I-Iinduism.
comparable to the cross of Christianity and the half-moon of Islam.
98 In the epic Garucja must creep through a revaluing wheel in the sky to find the kumbha (jar).I
(Kuiperjdifferfroin Bosch (p. 149) in seeing in this only a doublingof the kumbha-moon motif (cf.
p. 159,. 75: the golden wheel as representing the jar of the underworld). T o interpret the wheel
as the sun, seems mythologically impossible to me.
39 A striking confirmation of this is found in Indonesian traditions (Bosch, p. 246).
40 K.S. If. 5.-l:'Who k n o w his (viz.Agni's)fixed rules,grows (with thesejas with the twigs," (vayB
iviinu roha~e)may perhaps be considered a literary expression of that notion, which Bosch was the
first to recognize on the reliefs of Bharhut and Sanchi.
41 Cf. Bosch, p. 157, whose explanation seems hardly acceptable to me.

42 In case one wants to compare the aerial roots of the banyan tree (Bosch, p. 69f.), it should be

noted that this does not seem to be parallel to the aSvattlta but rather opposed to it and that, as is
shown by its connection with Rudra-Siva, it symbolizes the cosmic ttiuytti at the end of the sacrificial
year. On the other hand the as'vattha is connected with the cosmic expansion (pravytti) during the
sacrificial year, and with Visi>u. Only once is the banyan tree called the tree of V a r u p , in which
case it must represent in earthly reality the inverted aSvattha of myth.
"pot-bellied" beings from the jar, he is dealing with a problem to which a
purely archaeological argumentation cannot do justice.
If we may assume as correct that several names of "pot-bellied" beings (e.g.
udunzbaln-, Hzclz~~~ba-,Herambu-, kumbhE?lda)are ol l'roto-Munda origin, and
that especially some non-Aryan word groups denote both "jar" and "belly"
e.g. kdbiin&ta, bha?zda, bhanka "pot, jar"; kdbandha, phana+, phZnda "belly", cf.
ku-mbh@$z).-which is again connected with the stronger sensorial character
of these languages as compared with Aryan-then there is every reason to
consider the "pot-bellied" beings as having been adopted from a foreign
culture. In that case an explanation will have to be given which takes other
associations into acc0unt.~3

THE DOUBLE-TREE MOTIF

If we are right in concluding that an essential part of Bosch's explanation

for the double-tree motii must be abandonccl, the question crops up how this
symbol can have arisen. One suggestion may be given here in conclusion. Let
us start with the Vedic conception of the world: together with Agni's ascension
to the sky, the tree rises u p from the primordial and supports the roof of
heaven as the world-pillar. Its roots, however, reach through the mountain
into the primeval water,45on which the earth rests, and thence they bring
upwards the amrta (the essence of these waters).
I n the cosmogonies of the later Yajurvedic texts46 we find, however, by the
side of the old motif of the primordial the beginning of which is brought
u p on the snout of a boar from the bottom of the water, the typically native
symbol of the lotus plant, on the leaf of which the boar puts the first earth.
The first reference to this occurs as early as Rpedn VI. 16.13.The two symbols
are henceforth usecl side by side as equivalents. The conclusion, then, seems
justified that in principle the lotus + fig-tree motif is a substitute for the
priinorclial hill + tree n1otif.'~8 It is possible that at a later date the notion of
the aitlattha parasitizing on a iatn;(these two standing already at an early date
for the upper-and nether-world) was connected with the image of tree+
sun (brn/matnrda) and mountain +
jar (padnzn~?zukz)as representations of
Agni and Somn
I have preferred an ample discussion of some fundamental points to losing
myself in a detailed analysis of the infinitely varied phenomena dealt with

43Theremarkable fact that most of the words for "jar" already in the I?p-t~fo give the impression
of being of foreign origin raises problems in this connection which can only be hinted at here.
'*It has been remarked ahose that the conception of the Vedic primordial hill cannot be derived
from the lotus root (Bosch, p. 96).
45 Cornpare with this the Jamnest. p ~ i u i i g u t ~ .
46 Taitt.Bruhm. and T'rinii~iillru/~m (sec 1-iertc.l. Die F1i111111elsloye
im Veda zoidin Au'estu, p. 24f.).
47 T h e m;ilrkZof the Iutlr-a-festival (Bosch, p. 153).

I n Bosch's view the relation is, of course, the other way round (p. 96).
in this book. It was impossible to accept that the vegetal motif is always the
primary one. In this connection it is necessary to take a firm stand against
Bosch's idea that the "macrocosn~ic"collception of the world must necessarily
by a "mechanical" one (p.231 f.). In my opinion this is a ~nisconception.The
question may be left open whether there is any sense in contrasting in this
connection macro-and microcosn~ic.I t may suffice to observe that a concep-
tion of the world that has its root in the old Vedic myth (and therefore in the
Agni-mysticism) autolnatically cannot have been "mechanical".
It may be said that The Goldetz Germ is a book of genius, written by a
scholar with a n inspired viGon. T h e task the author set himself was, however,
eztre~nelyheavy, owing to the be~vilderingdiversity of the material that had
to be dealt with and to the rtnrtsual obscurity and clilficulty ol the problems of
the s~inbolisin.Even for a nmn with Bosch's originality of vision ancl great
eruclition it war impossible t o srrcceecl completely in the first instance. It is to
he hoped that in a future cleiinitive edition Bosch will organize the material
in a ciiilerent way, ancl that alter some revision of the basic points of his
reasoning. he will also decide to give up the idea that the plant motif was the
nlain 5ource of mythological conceptions. For then this book will become
what it is already in essence: one of the classics in the field of the history of
Indian culture.
 1. The theories suggested in explanation of the character of God Vignu-
are remarkably divergent. The prevailing view that he was a solar deity1
has often been questioned and criticized in recent times.2 Besides he has been
1 Thus already &kaptini, if his interpretation of the three strides as referring

to ppthivyci;m, antarikse, divi (Yska, Nir. 12.19) implies the trias Agni-Vf~yu-Siirya
(cf. Durga's commentary and B~haddevatLi1.90, but on the other hand €B. 1.9.3.9)
and Aurnavshha (ibid., see Macdonell, J R A S 1895, p. 170), probably Tiiska, Nir. 5.17:
iipivistah] pratipannaraimih, B~haddevata(c. 400 B.c.?) 11.69: Vigvur nirucyale sziryah
sarvam sarvdntarai ca yah, Skandasvfimin (c. 500 A.D.)ad 1.22.17: Visnur evQ 'dityamii-
peva l o h n prakiidayati (etc.), Comm. ad PB. XVII1.7.13 bradhnasya] dditgasya V i s f ~ u -
rcpasya, Mahidhara (c. 1590 A.D.) ad VS. 5.15 (= Agni-Vsyu-Siirya), and further, e.g.,
Whitney, J A O S 3 (1853), p. 325, A. Weber, Zwei vedische Texte uber Omina und Portenta
), p. 338, A. IZuhn, Die Herabkunft des Fevers und des Gattertranks (1859), p. 66 f.,
p. 251, Edw.Moor, The Hindu Pantheon (1810,1864,1897),p. 13 (sun, earth, water, space),
Laseen, Ind. Altertumsk. 12(1867), p. 919, Max Muller, Hymns to the Maruts or Stormgods
(1869), p. 116 f . (= SBE. 32, p. 133 f.), Eggeling, SBE. 12 (1882), p. 73 n.2 ("? the all
pervading Sun"), 26 (1895), p. 62 n.2, E. Hardy, Die Vedisch-brahmanische Periode der
Religion des alten Indiens (1893), p. 33 E. (Sun and Moon), L. von Schroeder; Indiens
Literatur und Cultur (1887), p. 324 B., Mysterium und Mimus (1908), p. 56 (Sonnengott
und Fruchtbarkeitsgott), Arische Religion (1916-1923) 11, pp. 64, 669 (earlier a moon-
god), Barth, Religions of India (1891), p. 165 f., Maedonell, J R A S , 1895, p. 175 ("ori
nally a sun-god"), Vedic Iclythology (1897), p. 38, E. W. Hopkins, The Religions of India
(1895), p. 41, Epic Mythology (1915), p. 85, P. , Sechzig Upanishad's des V e d a
(1897),p. 277 n., Th. Bloeh, Worter und Sachen I , Oppert, Zeitschrift ffir Ethnolo-
gie 37 (1905), p. 331 f .,D. R. Bhandarkar, Lectur Ancient History of India (1919),
p. 128, Havell, Handbook of Indian Art (1920,"927), p. 164 ("the sun at noon supporting
the heavens [Vishnu-Sarya], or the sun a t midnight under the earth reposing on the
coils of the serpent of eternity [Ananta or S&shal,the Milky way"), H. D. Griswold,
The Religion of the Rigveda (1923), p. 284, A. B. Keith, Religion and Philosophy of the
Veda (1925), p. 109, A. Hillebrandt, Vedische Mythologie II"1929), p. 319 f., A. Nohen-
berger, Die indhche Flutsage und das Hatsyapuriipa (1930)
allindischer Mdchte und Feste der Vegetation (1937), 11, p.
24 (1938), p. 89 ff ., R. N. Dandekar, Kane-Volume (1941),
jumdar and Pusalker, T h e Vedic Age (1951), p. 367 f., e
of early T ' i ~ v u i s m(1954), p. 172: "the eternal phenomenon of the pervading and omni-
present, mighty and blessing stream of celestial light, warmth, and energy."
P Cf. ;.g., Hopkins, J A O S 16 (1896), Proceedings p. cltlv B., S. Konow, Visvabharati
Qz~arte?;yI11 (1925), p. 216, S. Konow-P. Tuxen, Religions of India (1949), p. 61 f.,
Neisser, J A O S 45 (1925), p. ndt, Ved. Myth. 1 1 2 , p. 319, K. Oldenberg,
Religion des Veda (1923), p. 23 y Xrappe, Mythologie univcrselle (1930), p.
141 f.
 resent the or the Fire-god,4 S ~ r n a or , ~ a rnoun-tain god
with ~ e g e t a t i o n ,a~ god of f e r t i l i t ~ or
, ~ a deity connected with
He has been stated to be an awakener to life,g or the sacrifice,1°
ar personification of the brahman- or atman-," or a god of veneration
iation.12 It has further h e n suggested that "the original character
was a non-Aryan13 or a proto-Indian14 religious concept. He has
en considered a philosopher's,15 as well as a "late popular" god,'$ a stridi
ant,lT no less than an anima, the essence of the pitciras and, a t the sa
d of evolution,lgof rnovernent,2O or of irnrnanenm21
ondgott" (Mylhologische Bibliothek V I I I nr. 4, Leipzig,
1916), pp. 5-17, von Schroeder, Arische Religion 11, p. 669. Cf. Nardy, Die Vedisch-
brahmanische Periodc, p. 33 f .
vkdique 11, pp. 416, 41 ("l'identitt! primitive de Vishnu avec
ma, prototypes de tom les sacrificateurs"), Sarkar, The Folkelements oj Hindu

ta Aptya ( I ) ,p. 93 f.: "der vedische Soma-gott par pr6f6rence."

of Aryan Rule in. India (1918), pp. 28, 111, 182 f . (but a t the

2 H. Giintert, Der arische WeltkGnig und eiland (1923), p. 292 (an ithyphallic
hologie universelle, 1930, p. 141 B.). Cf. Neisser, J A O S

l'histoire des religions CXVII (1938), p. 167.

.
de Pest, p. 22: "dou6 . . du pouvoir d'6veiller la vie dans le
rois pas mythiques. "
Gods and Heroes (1922), p. 37 E.,J. Charpentier, Pe
cording to E. Oldenberg, Die Lehre der U

Deussen, Der Gesang des Weiligen, Introduction p. iv.

3% B. Paddegon, in: Van der Leeuw (e.a.), De godsdiensten der wereld I1 (1940), pp.

,p. 22 (concerning Krqna), W. Ruben, Eisen-

thor, ~'ffberdie aftindische Gottin Dhiganl und Verwandtes" (1917), pp. 47 n.2, 49, J.
 The very existence of such a diversity of views might provoke some questions
\vith regard to the methods applied in the field of "Comparative Religion."
For we are here concerned with one of the prominent gods in a religion that
we know from the very beginnings of history up to the present day. The mass
of evidence no doubt constitutes a special difficulty, but on the other hand
it is unquestionable that the manifest inability of modern science to under-
stand this god is not due to a lack of data. The main problem is rather, how
to find an adequate interpretation for the evidence contained in the ancient
texts. One is reminded of Andrew Lang's words: "Sothing in a11 mythology
is more difficult than the attempt to get a clear view of the gods of Vedic
Indeed, we are here confronted with the fundamental difficulty of
Vedic myt,hology, viz., the impossibility of understanding a single rnythologi-
cal figure isolated from the context of the mythological system. Monographs
on single deities are indispensable as a first step, because no interpretation
can be attempted before all data have been gathered and arranged. The final
interpretation, however, will have to account for a god's function within the
P
total system.
2. In the oldest Vedic text Visnu's function seems to be restricted to his
taking three strides through the Universe. Kothing suggests that this text is
particularly reticent about the real nature of the god's divine act. So a correct
interpretation of the Rigvedic evidence is of vital importance. At the end
of the nineteenth century Macdonell suninlarized the contemporary research
in these words: "The opinion that Visnu's three steps refer to the course of
the sun is almost u n a n i m ~ u s . "They
~ ~ are equally true today. There is only a
difference of opinion as to the identity of each of the steps, which some in-
terpret as referring to the rising, culmination, and setting of the sun, while
others hold them to mean the three divisions of the Universe. I+om R h -
donell's classification of the Rigvedic we learn that Vignu:
1 ) took three steps (1.22 18, VI11.12.27),
2 ) strode with three steps through this morld (tddm 1 2 2 I T ) , or traversed n i t h
three steps this bide extended sphere (sadhitsthan~1.154.3),
3) traversed the earthly regions (pc%thzoanz rdjdmsz), and fiaed the upper sphere
(tittaram sadhdsthawt) while stepping thrice (1.154.1),
All-god, viewed from the side of life, Siva the same, but viewed from the side of death,"
p. 224: "8iva especially is the god of involution, Vishvu of evolution" (brit see also pp.
145, 194!). Cf. in this connection Hopkins, Religions of I n d i a , p. 388: "Vibhnu and Civa
are different gods. But each in turn represents the All-god, and con'sequently each repre-
sents the other."
Hopkins, J d O S 36 (1916), p. 264.
21 R. Otto Gottheit u n d Gottheiten der Arier (1932), p. S3 R., Zeitschrift fzir ;Illsszo~ts-

kunde u n d Religionswissenschaft 49 (1936), p. 296 f f . Cf. Gefulcl cles Uberweltl~chen,p.

111: "Einschlupfer, a haunting something."
M y t h , Ritual and Religion 11, p. 148, qu ted b y Hillebrandt, Veil. illyth. I?, p. 1.
Vedic Mythology, p. 38. Cf. Keith, Religion and Philosophy of the V e d a , 'p. 109:
"the solar nature of the deity is reason:tbly plain."
J R A S 1895, p. 171.
 4) stepped widely over the earthly scil. regions (phrihiv&ni) with three strides
(1.155.4); thrice traversed the earthly regions ircijL5msi phrthivdni V1.49.13),
5) thrice traversed this earth (etBm p r t h i v f m V11.100.3,4),
6 ) strode t h e e (steps) (thither) >vilere t h e gods rejoice (VIII.29.7).

e other passages ma e it clear that the third step is the highest a

mortal men can only see the two lower ones (1.155.5, VTI.9(3.1). This
, which is sometimes comected with the third and highest place
.1.3),25is likened to an eye fixed in heaven which singers light up
.20 f.) and is said to shine down greatly (1.134.6). The poet prays that
he may attain to that dear abode (1.154.5).
This is the general picture that can be derived from the scattered allusions.
For further details the reader may be referred to the manuals. The facts
n~entionedallow some general conclusions. I11 the first place it is apparent
that to the Rigvedic poets the exact place where these strides were t a
TV% of minor importance: the words pdrlhiuib~i rcijiintsz (or even p & i
alone) were apparentlg sufficient to indicate the place of Vi~nu'sact.26
also AthS. XII.l.10: "(the earth) on wllich Visnu strode out."
place the number three, though traditional, is not the only o
for sorneti~nesVisnu is said to have stridden over the
rth." Now, three and seven are both numbers which
e question arises, if there is sufficient reason for assurni
explanation for the numbcr three. Only later, especially in the Yajurveda,
the three steps equated to p!.thzvi, antdriksarn and dyazih (e.g., VS. 2.25;
, TS. 1I.4.12.3ff ., V.2.1.1, TB. 111.1.2.6; cf. clivd v& vip.~a(v)utd v& plthivy6
) v6 'ntdrzkpfid VS T S
mnh6 v& v i s ~ a ( u utci , etc., and see also AthB. VI1.26.8,
the other hand, there occur also quite different interpretations
VI.15.1 l ) , and it remains doubtful, if much weight should be laid on
these ritualistic specuiatioas. till, the correctness of tbis interpretation,
which has found general acceptance in later Binduism (Mahabhiirata, IGli-
dssa, etc.), has seldom been queitioned. Oldcnberg, however, rightly objected
that this seemingly simple explanation is at variance with the Rigvedic texts
which stress the fact "dass cler dritte Schritt ein eigenartiger, erhabenster ist,
dass er in geheinmisvolle hochste IIohe fiihrt."28 Cf. asyci rcijasa?~paralc6
(VI1.100.5), ta'd vis?zo?h purama'rn pad&?t sddd paAynrzci siircip~!t / d i v h a cci1;sur
i t a t a m (I.22.20), ydtra deviso mddccnti (V111.29.7), na'ro y d b a deuaydvo
nzddanti (I.154.5), vL?<10?1 pad6 param6 m i d k m litsa[~ (I.l54.5), ydtra g6vo
R ~ r g n i g n e ,Iielzg?on vtklzqr~e11, p . 46.
m Sec also Oldenherg, Relzgzon des V e d , p. 320. Ilillebmndl's theory of "verschie-
dene, bisweilen in dnssdbe Lied aufgenodnzene Traditionen" (T'cd. X@. 112, p. 317)
does not carry conviction.
%7 1.22.16 ?&to v i ~ ~ vut cra k ~ a m CpTfltio~yhh snptQ dhhnrabhih ("la terre avec scs instal-
lations m~tonomes"Itenou, Etudes vtrliques IV, p. %), cf. VIII.69.7 sncez:ahi t r i ? ~saptd
sakhy.ri?~pad& "so wollen wir . . . dreilnal in den sieben Schritten cles Freundes zusam-
mengehen," Geldner. Otllerwise VI11.59.5 t r i ? ~saptd.
2 Y R e l z g i m ~des T'eda, p. 229 f .
bh&rihyrigci:ay&sa?~(G), visco deva tvcim pnramcisya vitse (VII.99.1), and 1.155.5.
Other facts corroborate the inference that Vis~u'sstrides are not connec-
ted with any natural phenomenon. the words gci 2~ etridhbtu prthivim utci
dy&m bko ddhhrn bhzivanani vihvb (1.154.4) do not suggest that the poet
had the trZm1 lok& of the later Veda in mind. Hillebrandt rendered it "der
allein Eirnmel und Erde, je zu dreien, und alle Wesen gehalten hat."29
We might prefer: "who alone supports in a threefold manner Heaven and
Earth, all creatures."
For these reasons one can hardly evade the conclusion that in the Rigveda
Visnu has no particular connections with the sun. I t is possible that the sun,
when in the zenith, was associated with Visnu just as the rising sun was a
manifestation of Mitra and the setting sun one of Varuna (AthS.
If so7 this could account for AthS. 3311.2.31, but the text is not explicit.
As for RS. 1.155.6, this passage points to a connection with the solar year of
360 days30 but does by no means imply that this year, set in motion like a
revolving wheel (cakr&z nci vrttcim), is directly connected with the sun: in
the Briihmanas the year always represents the totality (like Prajiipati), for
when time is conceived as a cyclic process, all is contained within the com-
pass of a year. The technical term for setting in motion this cyclic process is
cakrcim prd vartayati ever since RS. 1.164.11-14 and the calcrci- which soon
came to be associated with Visnu (cakrap&.~i- &B. V.S.2) is probably the
symbol for this progression of the Universe.
On the other hand, it is hard to accept Oldenberg's suggestion that the
number three is nothing but a "Wiederspiegelung einer allgemeinen Neigung
der Phantasie. "31
3. With regard to the aim of Visnu's strides the Rigveda is as explicit as
one may desire: it may be summarized in the one word "life." Visnu is said
to stride out jZvcise (VI.69.5), urugbyliya jzvdse (I. l55.4), mdnave bddhithga
(VI.49.13). Cf. VI.100.4. Some scholars assign the same act to Indra on the
strength of VIII.63.9, bu6 Geldner is apparently right in reading asyd @sno
vy ddanci ur4 kramista jZvdse (for vycidana): "Er schritt weit aus nach dem
Reisbrei, fur diesen Bullen zum Leben," so that it becomes an additional
testimony for the aboriginal odana-myth that has been grafted on the Vedic
29 Lieder des Rgveda, p. 89. Not 'Idas Dreifache, d.h., die dreitheilige Welt" (Roth,

P W . 111, col. 933, Ludwig, Der Rigveda I, p. 158, Geldner, Der Rigveda I, p. 212). It
should be noted that tridhhtu is never used t o denote a tripartition of the world: in
IV.42.4 tridhhtu prathayad v i bhdma (as compared with, e.g., VII.36.1, V.87.7, X.62.3,
VI.72.2, X.82.1; 149.2, 11.10.7, V.58.7, 1.62.5, VIII.89.5, 1.103.2 = 11.15.2) i t must be a n
adverb, as i t is in VI.44.21 tridhdtu divi, VII. 4 t&va tridhdtu prthivt uih dya* Gaidvd-
nara vraMm agne sacanta, 1.34.7 phri tridhhtu pythivgm aidyatam (:8 tisrQh prthiv&-).
For tridhhtu "thrice" see Itenou, Etudes vkdiques et pcS@n&ennes IV, pp. 54, 70. It may
be equated t o tridhd (cf. 111.56.6) and tredhh (tredhh ni dadhe paddm 1.22.17).
30 Macdonell, J R A S 1895, p. 174, f., cf. Oldenberg, Rgveda, Textkritische und exege-
tische Noten I , p. 151.
31 Religion des V e d a , p. 231. These words reflect his "peculiar mental attitude towards

myth and ritual" (Held, T h e M a h a h d r a t a , p. 117.)

Vytra-n~yth.~* Cf. further urzcsyd nu urujmann dprayuchan (AthS. VI.
hbtau pyvasva bahzibhir vasdvyai!~ (AthS. VII.26.8). It is not clear,
M a ~ d o n e l holds
l ~ ~ this motive to be ;'a secondary trait." The frequency
which it is mentioned suggests that it is an authentic feature.
A further characteristic of Visnu is "his friendship for Indra, with W
he is frequently allied in the fight with Vrtra. This is indicated by the
that one whole hymn (VI.69) is dedicated to the two deities conjoi
a n d . . . by the fact that in hymns extolling Visnu, Indra is the only
deity incidentally associated with him either explicitly (VII.99.5-6; I.li
or implicitly (VII.99.4; 1.154.6, 155.1; cp. 1.6'7.T)."34Visnu's assistant
the fight with Vrtra has created some surprise. Oldenberg held this to c
"nur in einer ganz auf der Oberfliiche liegenden Anahnlichung an Indi
Others tried to explain this alliance between the two gods from their supj
natural substratum. Thus Gondn, who accepts the traditional explan:
of the three strides as referring to the diurnal ascent of the sun througl
atmosphere to the zenith (and more particularly, to the sun's energy
"pervading, omnipresent, and fecundating stream of light and ener
suggests that "Visnu's acting as Indra's companion or assistant in fig1
Vrtra may be interpreted as reflecting ancient ideas with regard to
given by the power inhering in sunlight to the weather-god when the 1
is about to destroy the demon who precludes the waters from fertilizing
soil."36 Similarly Griswold tried to connect the functions of lightning
sun, which he supposed to be personified by Indra and V i ~ n u . ~A?third t l
explains the mythical association of Visnu and Indra as the result of a
torical union of the Visnu-worshippers-and the Indra-w0rshippers.3~How
is there any reason to suppose that Visnu's r81e as Indra's "assistant"
different from his general salutary activity in behalf of Mankind anc
Universe? The answer will to a large extent depend on one's views a
Indra's character. Without entering into this problem I should like to
attention to a passage which seems never to have been discussed in er
studies on Visnu. RS. 1.156.4cd reads as follows:
dc7dhhra &&am uttamitm aharvidam
vraj4m ca viapuh st~-khiv& apoorputb

"Er besitzt hochsten Verstand, der den richtigen Tag ausfindig macht,
Visnu in Begleitung des Freundes den Rinderpferch aufschlie~st."~~
32iiIededelingen der Koninklijke Nederlandse Akademie van Wetenschappen,
Letterkunde, Nieuwe Reeks, Deel 13, No. 7 (1950).
33 Vedic Mythology, p. 39.
34 Macdonell, Vedic Mythology, p. 39.

36 Religion des Veda 231.

36 Aspects of early Vigpuism, p. 60. For Indra as a weather-god see also p. 31, as
senting power in nature, see p. 172.
The Religion of the Rigveda, p. 285 n.1.
38 Willebrandt, VedGche Mythologie 112, p. 313.
a* Transl. Geldner. For the accent of uporput6 see Oldenberg, ZDMG 60, p. 733.
mythological significance of the phrase vrajdrn apa-zzr- becomes clear by a
conzparison with IV.51.2 vy .;7, vrajcisya tdmaso n'v&ochci?zt~~crwaii chzicaya[l
pdvakhh (cf. VI.G2.11).@If Visnu was credited with the act of opening the
nether v.-orld along with his friend, he must accordingly ha\-e taken part in
the process of creation. To this same conclusion points the \vord aharvtd-.
No matter, whether it is translated "knowing the (right) day"-which
meaning nearly all authorities assign to it-, or rather "finding the day"
(cf. svarvld- "finding the sun," gdtuz~id-"finding a way [out of a critical situa-
tion]"), the ~ o r din any case seems to point to n definite day. In VIII.5.21
it also occurs associated with the idea of the opening of doors: utci n o divyb
isa utd s i n d h ~ % raharvidd / cipa d z h w a var.satha?r "und schliesset uns die
himmlischen Labungen und die Strome auf s-ie die Tore, ihr Zeitenkenner!"
(Geldner). I t should be noted that Grassmann's rendering by "der Tage
kundig" has been accepted by Bohtlingk and Geldner." But his further ex-
planation "fiir seine Handlungen die rechten Tage oder Zeiten wissend,"
which calls to mind the classical expression deiakdlajiia-, hardly defines the
meaning intended by the Vedic poets. If it means "knowing the right day"
we may rather connect it with such phrases as 1.123.9 janaty dhnah prathn-
mdsya n h n a . But since in VIII.S.7 the ASvins are also addressed as svaruidci:
"finding the sun [of the new year]," it would seem more natural to render
aharvidd in V111.5.9, 21 by "finding the [first] day [of the new year].42If this
conclusion is correct, the words dciksam aharvidam indicate that V i s ~ utakes
a part in the cosmogonical act of finding the sun in the darkness of the nether
world and furnish an indication as to the time at which Visnu was thought
to take his three strides.43
4. Before discussing the number three itself, it is necessary first to pay
some attention to Visnu's place in the cosnlic classificatory system. As Held
rightly remarks: "The gods cannot be explained, it is true, from the classifi-
cation-system, but we can understand thern through it."44 In three Atharva
vedic hymns (111.27, 231.3, XV.14) nearly the same classification is met
with, viz.:

X.
Soma
W. E.
Varupa Agni (XV.14.1 : Maruts)
S.
Indra

4Q See I I J IV, p. 226.

41 Translation of VIII.5.9: "die Tage kennend"; Glossnr: "die Opfertage (oder die
Tageszeiten) kennend, opferkundig."
*"Cf. Lud;vig: "Tagfinder."
4 3 Unfortunately the passage VIII.66.10 is too obscure t o allow an inference as to

ahardfi- (parallel to suardfi-?).

4%. J. Held, The illahdbhdrata, p. 222.
Indra's connection with the southern quarter, though found also elsewhere,
is remarkable. I n addition to these four quarters the ~rdhvddCk is always
imparted to Brhaspati, and the dhruvd dlk to Visnu (see AthS. 111.27.5,
XII.3.59, XV.14.5). What exactly is meant by the dhruvd dik becomes clear
from such parallel phrases as &vdcZ dik MS. 11.13.21 (p. 167) and iydm dik
TS. V.5.10.2, TB. 111.11.5.3 (= adhard dzk, Comm.). Cf. also adhcistdt
AthS. IV.40.5, ihd TS. V.5.10.4 (corresponding to dhruvdydn? diSC AthS.
III.26.5)45and asydTz dhruvdydm madhyamdydm pratisthdyd~ndiki AB. VII1.-
14.3. I t is accordingly the cosmic center and the nadir with which Visnu is
associated. The parallel texts of the Yajurveda show some variations. The
northern quarter is here allotted to Varuna (resp. Mitr6ociru~az~) and the
western quarter to S ~ m a The . ~ ~reason for this most abnormal division,
which apparently is a later corruption of the normal system as represented
in the Atharvaveda, is unknown. As for the dhruvh dik, XIS. agrees with AthS.
in associating it with Visnu, while TS. and TB. have respectively Yama47
and Aditi instead.
Now, the mythological meaning of the center has long been known: it
represents the totality of the parts distributed over the four quarters. So we
are driven to the conclueion that a t an early date Visnu occupied a more
central position than either Indra or Varuna, who are the protagonists of the
opposed groups of Devas and Asuras and thus stand each for one of the moie-
ties only. In contrast with them, Visnu must consequently represent the
unity of the two antagonistic parties, upper world and nether world. He
stands for, and is, each of the two worlds (just as later he is, in a way, the
heavenly bird Garuda and the serpent of the subterranean waters Sesa),
but under the aspect of their unity, like Prajiipati. I n AthS. X.10.30 the
cosmic cow is said to be Heaven and Earth (i.e., the totality of the Universe),
Visnu and Prajiipati. I n XIX.17.9; 18.9 the dhruvh dCk is associated with
Prajiipati and pratisthd, in XlVIII.3.25 with a god Dhartr.
Being related to both worlds, Visnu also belongs to the gods of the nether
world. In AthS. XI.6.2 he is invoked along with Varuna, Mitra, Bhaga,
Am6a and Vivasvant for deliverance from distress (d~nhas-).I t may have some
importance, therefore, that Visnu and Varuna are sometimes addressed
~onjointly*~ and that the mythical bird Suparna, the prototype of Garuda,
is said to be Vciru~asyadzitw, Yamcisya y6nau (RS. X.123.6). This explains
why visnu"s position in the cosmogonical Vrtra-fight was of the utmost
importance and a t the same time fundamentally ambiguous: his was to
some extent the position and the rBle which in other mythologies is attributed
to the divine trickster.49He could not fight the powers of the nether world,
46 Hrtuer's doubts (Pestschrift Winternitz, p. 144, n.1) do not seem justified.
46TS.
V.5.10.2, MS. 11.13.21 (p. 167, 3, 5 f.), TB. 111.11.5.3.
47 I n the Mahiibhiirata Visnu is Yama, see Hopkins, Epic Mythology, p. 207.
Cf. AthS. VII.25.1,2 (cf. IX.2.6), KS. XI11.4 (p. 184, 7, 10), MS. IV.14.6 (p. 223,
1, 5, 9, 12), TS. 11.1.4.4, TB. 11.8.4.4 E. See Hillebrandt, V e d . Myth. 1 1 2 , pp. 69, 325.
See J. P. B. de Josselin de Jong, De Oorsprong van den goddelijken bedrieger (Mede-
 a did, because these too were part of his essence. His natural place
ajgpati's) 50 was between the two parties, as indeed his position is
Churning of the O ~ e a n . ~Visnu
' is the typical mudhyasthnh,
link between the two cosmic moieties, reflexes of which cone
ized in such ritualistic speculations as SB. III.4.4.15:
the night, and Visnu "what is between the two" (y& rin
(p. 102, 13 B.)agn& i ~ d g a ? n ,s o m a k S a l y a v , visr.m?p lejananz
t of the arrow). The factual importance of such identifications may be
t such passages show that the authors were quite aware of t
function of Visnu as the connecting link. In passing it may
called to mind that in later times Yili;~um-as (and still is) believed to s t w for
four months in the nether ~ o r l d where
, he sleeps on $esa in ihe subterranean
waters and thak in contrast with other gods, he is endoiwd with two zc7hanas
which, while being a t variance t?ith each other, a t e same time stress by
their relationship the totalitmian character of the go From a purely mytho-
logical point of view Visnu, who by his position in e center must also in
Vedie belief have been immediately associated xith the co
have ascended along this pillar a t the beginni
in the second half of iV2 but, owing perhaps t
texts are primarily concerned with the beginning of the
not contain any reference to such a belief.
5. That the nlythological concept of the nether world was associated wit
e earth appears from the fact that . 111.1.5.3 substitutes Aditi for V i y p
the divinity presiding over the na The schelion to this passage has the
ollowing comment: i y a v dig] i t y a n e n d 'dhnrd dilc pradnriyute; aditir] b h a m i s
may accourli for Visnu's connection
the gods won from the Asuras t
horse through Ind runa, the goat and the sheep t rough Indra-Brhas-
deelingen der Koninklijke Akademie van etenschappen, Afd. Letterkunde, Deel
Serie B, N o 1 [1929]).
See I I J V, p. 58 and cf. JB. 1.19 trln . . . anlara vikraml~ci'&-that.
According t o the Mah&bhiIrata1.16.13 f. (crit. ed.) the Asuras held the head
serpent, and the Revas its tail, but ananto bhagavan devo yato, iVhrciyanas tatah,
"bignu, who is mythically identical with &$a, his nether world aspect, is the con
link between the two parties. We stands between them, e.g., on the reliefs of
Vat, but also in modern representations (cf. e.g., the frontispiece in M u ~ b a
c a r a n u p panmimga, sarpvat 1978). Since the Asuras are the eider brothers of the Devas
hbh. XII.34.13 crit. ed.: asura bhrhlaro j y e ~ t h ddevciS cdpz yavfyasah), it is interestin
t o note t h a t among the Papuas of Waropen (New Chinea) the cooperation and rivalry
between two no.1-exogamous groups is seen as the head and the tail of a triton shell, the
head being the clan of the "eIder brother" and the tail the clan of t h
See G. J. Held, Papoea's van W-aropcn, Leiden 1947, p. 49 f . (as su
s s o ~ xxx r r , p. 180).
62 Cf. in the Edda the squirrel on the trunk of Yggdrasil, who "
words from above and say them to Nt8h~ggef[the dragon] below"
For Aditi and the Earth, see MO--lonell, Ved. Myth., p. 121, Agrawala, I n d i a n
Culture 4 (1938), p. 407.
pati, but rice and barley (vrihi??~ cca ycava??z ca) through Indra-Visnu.s4 With
the nadir are associated earth and fire, herbs (6sadhi-), forest trees, and
plants (virzidh-), AthS. XV.G.1; herbs, 111.26.3, SII.3.59 (but Paippalgda:
food and plants respectively), Visnu and plants, 111.2'7.5. I t may he added
that the earth is mhlciran-r 6sacll~;incinzdhruvhrn bhkmim p?hivrni, XII.1.17.
Visnu and the nadir are connecttd with Virgj, XV.13.5, who again is asso-
ciated with food,55and with Aditi (cc'clilim virbjam VS. 13.43, cf. AthS. X I . -
3.11). To the same relation points AthS. 11.12.1 li&msya pdtny urzigiig6
'dbhuta?~(cf. RS. VIT.35.10d lcs4trasyn pdti!~ beside Visnu in 9c?). I t is not
surprising, therefore, that in the Yajurveda, which particularly stresses the
chthonic character of Aditi, she is called V i s t ~ u p a t n i as
, ~ ~are also VirBj5'
and dhruvi dik." A discussion of Aditi lies outside the scope of this article,5g
but it may be pointed out that she is sometinles equated also with Heaven
and Earth.6o I-Ieesterman holds her to be identical with A i ~ u r n a t i ,who~ ~ is
opposed to Niryti, the "personified representation of the cover of the em-
bryo." This may be one of the sources of Vis-u's special connection with
women. 62
6. As far as I see, Professor . Norman Brown has been the first to state
that Indra's fight with Vrtra is a creation myth.63 Since the Iligveda con-
nects Visnu's three strides with Indra's slaying of the serpent demon, we are
driven to the conclusion that Visnu's act must somehow belong to the same
mythical context.64Several well-known authorities, however, have felt un-
able to accept this conclusion. I'Iillebrandt held the association of Visnu
with Indra to be quite irrelevant: "Die drei Schritte, die er LB. IV.18.11;
VIII.12.27; 52.3 fur Indra tut, sind eine ganz iiberfliissige Sache und hier
5 4 For these three groups of gods see below, p. l50 and cf. AB. 111. 50, &v$. VI. 1,

Eggeling, S B E . 26, p. 429, n. 1.

a n n a m vzrdt, e.g., J B . 11.15810. See M . Mauss, Mblanges S . Lbvi (1911), p. 333,
Hauer, Pestschrift Winternitz, p. 144, n.1, Hopkins, E p i c Mythology 78, Gonda, Aspects
of early V i s n z ~ i s m p.
, 67 (vircij = totality).
6BVS.29.60, TS. IV.4.12.5, VII.5.14.1, TB. 111.1.2.6, AB&. IV.12.2. For Viggu and
Aditi in the ritual see Caland, illtindische Zauberei, p. 112. In the epic the earth is
Vaigvavi (Kopliins, E p i c Mythology, p. 80).
57 KS. XXII.14 (p. 70, 20).
K8MS. 111.16.4 (p. 189, 16), TS. 1V.4.12.5 dhruvh di&n vijqzupatny Ctghord 'sy&Sdnii
sdhaso y& manbtd.
Sfi See, e.g., Bergaigne, Religion vbdique 111, p. 88 ff., Max Muller, S B E . 32, p. 241 ff.,
Vodskov, Sjaledyrkelsen og Naturdyrkelsen, pp. 321-73, ICeith, I n d i a n Culture, 111,
pp. 721-30, V. S. Agrawala, I n d i a n Culture IV, pp. 401-409, IZuiper, De goddelijke Moeder
in de Voor-Indische religie (1939).
Cf. Macdonell, V e d . Myth., p. 121, TS. 1.5.11.5, IV.4.12.5, IZS. XXIII.8 (p. 83, 19),
and Weesterman (see nest note), p. 202.
81 J. C. Heesterman, T h e rlncient I n d i a n Royal Consecration (1957), p. 18.
See Gonda, Aspects of early V i s n u i s m , p. 19 E.
63 J A O S 61 (1941), p. 79; vol. 62 (1942), p. 98. For further references see ITJ IV (1960),
p. 218, n.6.
For the ancient Indian concept of creation as a process of arranging the primordial
matter see, e.g., Held, T h e Mahdblarata, p. 118.
auf den ~ ~ t r a i a r n paus f dem visguitischen Sagenlireise iibenlomrnen, um
beide aus irgendeinem fur die Verehrer wichtigen Grunde miteinander zu
~ e r k n i i p f e n . "Oldenberg
~~ expressed a similar opinion, "Aber wie die vedi-
schen Dichter es lieben, die Taten, ~velchedie Ordnung der Welt und das
gliickliche Dasein der Nenschheit begriinden, mit den1 Vrtrasieg in Ver-
bindung zu setxen, so muss hier %?nu zugleich mit der Vrtratotung das
vollbringen, was die ihm eigene Tat ist: er muss weit ausschreiten und da-
durch Indra das grosse Schlachtfeld fur seinen Sieg schaffen."66 I t may be
objected that Visnu's act cannot be said to be "superfluous" as long as its
real character is not understood, and that the existence of separate Visnuite
sagas is an unproved theory.67 Others have rightly explained the connection
between both acts from the character of the t ~ gods o (see above, p. 142). We
therefore start from the following assumptions which seem to us sufficiently
warranted :
l ) The myth of Indra's fight with Vrtra refers to the creation of the organized
cosmos, consisting of upper and nether world.
2) Vignu's mythical act, accomplished like Indra's for the welfare of the world,
and sometimes directly connected with i t , is also likely t o form part of this creation
act.
3) The poets do not lay much stress on the places where the steps were taken,
the most important trait of the myth apparently being their number.
4) I n the Rigveda the third step was conceived as quite mysterious and invisible
t o the human eye. Later theological speculations often try t o find some correlate
in the phenonlenal world for such mythological concepts (as in the case of the in-
verted tree) . 6 a
5) In spite of such assonances as sdkhe viggo vitardm vZ kramasva@gthe prevailing
idea with respect t o Visnu's three strides seems t o have been t h a t of an ascension.
There is no doubt as t o the third step being the highest. Reference t o this ascension
is met with in VIII.69.7 dd ydd bradhndsya vigfdpam indrai ca gdnvahi "wenn ich
und Indra, wir beide, nach Hause zur Hohe der Sonne hinaufsteigen."?o Geldner is
probably right in taking st. 7 and 16 (ddha dyukgitm sacevahi!) as spoken by Vignu:
in t h a t case the idea of Indra's ascension must We due t o his association with Visnu.71
The same idea of ascension associated with Vignu also accounts for his being an
unnetr-; cf. JB.II.6S1 unnetar un md naye ' t y ciha, v i g ~ u rv& unnetd, yajiio vai vigguh,
66 Vedische Mythologie 112,p. 313.
Religion des V e d a , p. 231.
@7 The fundamental weakness of such theories has rightly been censured by Norman
0. Brown, A m . J. Arch. 53 (1949), p. 218, in the words: "The history of religion is con-
ceived as a blind diffusion and collision of tribal traditions, with little or no regard for
the functional interrelationship between religious institutions and the total culture of
which they form a part."
6a See Bijdragen Koninklijk lnstituut voor Taal-, Land-en Volkenkzcnde 107, p. S0 ff.
@* Oldenberg, Xachrichten Giitt. Akad. W i s s . 1915, p. 374. I n view of the "magical"
use of repetition and assonance in prayers and exhortations t o the gods (e.g., VS. 9.9
vajzno v6jajito vhja?n sari$ydntah, RS. 1.124.13 dsto&vanz stonzyd(h), 1.113.18 uiattr
L..

u@sah, 19 jdne janaya) no argument can be derived from the use of vi for the etymology
of V$$?$U-.
Cf., e.g., comm. on PB. XVIII.7.13: bradhnasya] cidityasya vig~ur.rlpasya.
St. 14 refers t o Indra and Vignu. Cf. also Oldenberg's note on st. 15 and see below
p. 151.
 yajiia evai ' n a m tat sarvasntdt p&pmano vimucyo 'nnayati72 and MSinGS. 1.11.18
Vigqus tvcim ~ n n a y a t uand
, ~ ~especially for the identification of V i ~ n uand the sacri-
fice.
7. The main problem with which one is confronted when trying to under-
stand the real character of Visnu's strides has never been clearly stated.
On the one hand there has been a tendency, ever since the oldest Yajurvedic
texts and the pre-Y8ska interpreters of the Rigveda, to connect them with
the triple division of the universe. On the other hand it cannot be doubted
that, whatever may be the origin of this threefold division, the cosmological
concept upon which the T'edic (and particularly the Rigvedic) mythology
is mainly based is that of a cosmic dichotomy. The Vedic conception of the
universe centers about the oppositions of Heaven-Earth, Day-Night, Devas-
Asuras: "twofold indeed is this (universe), there is no third" ($B. III.3.2.2).74
So, quite apart from other considerations, the dual division is likely to be
earlier than the tripartite one,?5 and the ritualistic interpretation of the
Yajurveda (e.g., VS.2.25 divi v i s ~ u rvydkamista . . . antdrilcge . . . prthivykm)
may not reflect the mythical meaning. The same is true of the Iranian in-
terpretation in the Dcnkart.7fi From the survey of (he Rigvedic passages
(above p. 139) it emerges that the poets never refer to this criple division:
they rather had the dual conception in mind, e.g., 1.154.4 yd U tridhktu prthi-
v i m utd dy6m 6ko dhdhkra bhfwandni vihd, where Visnu is said to hold in a
threefold way Earth and Heaven, all that exists.
For a correct interpretation we must start from the mythical significance
of the number three in Vedic I t has long been observed that the
predon~inantrole of the number five in the Veda (cf. pcirica jdndh, pdiica
k&h, pdfica c a ~ s a d h pciiica
, ksitih) is due to the fact that the five points
of the compass ( p d ~ c apradiAah 1 x 3 6 2 9 ) "represent in the Vedic scriptures
the entire ~ o r l d . " 7When
~ five points are occasionally mentioned, that in the
middle, where the speaker stands (madhyatcih RS. X.42.11), denotes the fifth.79
72 But in MS. 1.3.39 (p. 46, 4) dnnetar vdsfyo nci dnnayci 'bhi, KS. IV.13 (p. 38,9) zinne-
tar v(tsyo 'bhyzinnayci nah, KKS. 111.11 (p. 35, 9) unnetar vasyo 'dhyztnnayd n a h there is
no reference t o V i ~ n u .
T 3 But this formula is a "correction" of V i s n u s tu6nv etu TB. &S. (cf. TS. 111.2.6.1
Viq.yzus tv6nu v i cakrame). Note the use of u n - n i - as the technical term for the erection
of the yGpa (RS. 111.8.4, 9).
74 Otherwise 1.2.1.12; 2.4.21 (S. LBvi, La Doctrine d u sacri$ce, p. 92).
7 6 Held, The Mahcibhcirata, p. 141.

7 6 See above, p. 140. For the three strides of the priest in the Zoroastrian religion see
DumBzil, Orientalia Suecana V (1956), p. 14, Haggerty Krappe, kfythologie unzverselle,
p. 165 f . (cf. Darmesteter, Le Zend-Avesta I, p. 401, Hillebrandt, h'eu-vnd Vollmondsop-
jer, p. 171 f.).
77 No mention will be made of such general studies as, e.g., Usener, Dreiheit, Rhei-
nisches Museum 58 (1903), p. 12 ff ., W. B. IZristensen, Kringloop en Totcsliteit (Verza.
melde Bijdrngen, p. 231, esp. p. 281 ff .).
7 s Held, The fifah8bhcirata, p. 120.
Macdonell, Vedic Mythology, p. 9.
 Indeed, as Held remarks, "the number of the whole is obtained by continu-
ally adding one to the whole already obtained.7780 The correctness of this
observation is evident from such instances as pointed out by Heesterman:*l
the "sixteenth round" (sotjabin- graha-) added as an extra element to a total
made up of fifteen parts ' h o t only exceeds but also encompasses the preceding
fifteen-partite totality." Cf. AB. IV.1.4 "with the godabin as a vajra he en-
compasses (goes round, surrounds: parigachati) cattle." Prajiipati and the
year (which is the all-5ncompassing totality) are often connected with the
, number seventeen; in SB. X.4.1.16 this is explained as being 16 plus one,
Prajiipati himself being the seventeenth. (Similarly JB. II.1046:trayastrzmbam
cva pitarag Prajcipatim). In ICB. XIX.2 the thirteenth month is said to stand
for the whole year: ctci;v&nvai sa~pvatsaroyad e.ya trayodabo nzdsas, tad atraiva
sarvah sapvatsara @to bhavati.
Kow it has long been clear that Visnu's three strides are somehow con-
nected with the totality of the universe, but it has never been expressly
stated, what exactly is the mythical significance of the third step. Its explana-
tion must be sought in the cosmogony, i.e., in the creation myth. I n the be-
ginning there was the undifferentiated primeval world consisting of the wa-
ter@ and the beginning of the primordial hill, which the cosmogonical boar
had dug up out of the waters.83Heaven still lay on the earth.84By slaying
Vrtra, Indra rivets the hill, opens it, and "props up" (stabh-) the sky :85 thereby
the dual organization of the cosmos is created. But a t the same moment Visnu
"strides out": his first step corresponds to the nether world (which includes
the earth), his second step to the upper world, but his third step is a mystery,
not perceptible to the human eye, for it corresponds to the totality of the
opposed moieties, just as the thirteenth month stands for the totality of the
preceding twelve months. All that exists, is in the three steps, or in the third
that represents them. Hence it may be asked whether tSgu (scil. tri:szi paddsu)
vZbua??z bhzivanam 6 viveid3 (VS.23.49). As compared with the thesis of the
primordial world, and the antithesis of Indra's creation, Visnu's third step
is the synthesis. In later ritualistic speculations this idea can sometimes still
be traced, e.g., Indra got two parts of Vrtra's indriyam vbyam, but Visnu
the third (JB. 11.243). When the Asuras had stolen the rasa and vBya of the
sacrifice, Indra regained one-third with the hclp of Varuna, one third with
Brhaspati, but the last third with Visnu's assistance: "thus they excluded
80 Held, op. cit., p. 123.
The ancient I n d i a n royal Consecration, p. 13 f.
82 salilbnz apraketdm, e.g., RS. X.129.3, AthS. X11.1.8, TS. V.6.4.2, VII.1.5.1, TB.
1.1.3.5, 11.2.9.3, KS. VIII.2 (p. 84, 14), XXII.9 (p. 65, 13), J B . III.3609, SB. X I . 1.6.1,
etc.
83 I n Yajurveda, e.g., KS. VIII.2 (p. 84, 14), cf. MS. 1.6.3 (p. 90, 4), TS. VII.1.5.1,
T B . 1.1.3.6, BB. XIV.1.2.11; otherwise JB. 111,36011, AitUp. 1.3, SB. VI. 1.1.12, etc.
84AB. IV.27.5, JB. III.3614, PB. VII.lO.l, TS. 111.4.3.1, BB. 1.4.1.21 f., 111.2.1.2.
See Geldner, Der Rigveda in Auswahl I1 (Iiommentar), p. 113.
8s Cf., e.g., Geldner's note on his translation of VII.86.1.
 em from all" (16% sarvasmdd evci. '~~lardynn, JB. 1.1 0 ) . Here, as in the
passage from 1.181 mentioned above, Varuna apparently stands for the
Asuras and the nether world, Brhaspati for the Devns and the upper world,
and Visnu for the t0tality.~6
8. The difference between the current view of the Vedic god and the one
re advocated is apparent. According to the former view Visnu was in the
igveda a mere assistant of the great Indra, who gradually gained in im-
portance, finally to rise to the paramount position of mankind's Savior.87
In our opinion Visnu, far from being a mere assistant, must have been con-
ceived rnythologically as standing between the two parties in the Vrtra-fight,
just as he stood in an ambiguous position between Aeuras and Devas in the
amytamanthana, and to some extent also as Krsna stood in the battle of the
MahEibhiirata.88I may be permitted to quote here some words written many
years ago: "Just as it is said in the epic that that party will be victorious
with which Krsna sides (yatah K r s m tat0 jayah, Mhbh. VI.21.12, crit. ed.),
so me shall have to attribute a far more fundamental importance to Visnu's
seemingly insignificant rijle in Indm's Vrtra-fight than earlier Vedic scholars
like Hillebrandt and Oldenberg were ready to admit: the two-sidedness of
Visnu's nature is apparently the determinant factor which alone could incline
the balance in the combat of the cosmic moieties."sg Visnu, no less than
Indra, was considered a victor (RS. VI.69.8, JB.II.242 f.). Unlike Indra,
however, who apparently came "from nowhere," he origiilally belonged to
the nether world, though representing (like Aditi and Anunmti) its auspicious
aspect, which was opposed to dmhas-. He rose up from the center a t the very
moment when the dual world was being created, and so he is connected with
the pillar which now supports the sky. Just as this pillar connects Heaven and
Earth "like an axle two ~lvheels,"~~ so Visnu is the connecting link, which
forms part of both worlds (see p. 145). As the sacrifice strode forth from (or
through?) the skambhd-," so Visnu as the sacrifice ascends to the sky and
transmits the powers of the earth to the heavenly gods. Rorinow character-
88 Of course no argument can be derived from these speculations, as in many others
the meaning of the number three is no longer known: in VS. IX.31-32, TS. 1.7.11.1
Vignu is associated with three syllables, but the highest number is here seventeen sylla-
bles, connected with Prajiipr~ti.In JB. 1.156 the gods are said to have overcome the
Asuras with two savanas, after which Indra alone accomplishes the third savana (which
in 1.180 is connected with Indm-Visnu!).
E.g., G o n d ~Aspects
, of early Vignuism, p. l63 (cf. p. 30 f.). But on p. 32 he is stated
t o be equal in rank t o Indra (see also IZeith, Religion and Philosophy, p. 109).
88On Kygna as the divine trickster see Held, T h e Mahdbhdrata, p. 299, Hopkins,
Epic Mythology, p. 215. In the Veda only Vignu, as representing the total Universe,
could induce the sky t o "make room" for Indra's vajra: Brhaddevats VI.123 udyatasyaiva
vajrasya dyaur dackitu naamd 'ntaram.
89 Bijdragen lioninklijk Instituut 107 (1951), p. 77.

RS. X.69.4 y6 bkgeneva cakriyd SCLcZbhir vigvak tastbmbha pythiv6m uuld dyhm (said
of Indra).
AthS. X.7.16 yajii6 yhtra pbrdkrdntah (differently, Lindepau, ZII. 111, p. 236).
 him as the Soma-god par ezcel1en~e:~"he words with which the
esses V i s y and Indra in VI.69.6 "ye two are the [primeval] ocean
bowl which contains the S ~ m a "must ~ ~ primarily refer to Vispu (like those
the preceding stanza scimasya mdda urzi cakramcithe, see Hillebrandt, Ved.
yth. 11, p. 316, and those in VIII.69.7, see above, p. 147). Indeed, V$nu
presses the Soma for Indra (1.22.1), and er texts state that the,Soma be-
longs to Visnu (scimo vaiswvi g ~XIII.4.
. ). His is apparently the mythical
Soma bowl (kala'Sa-), like the mddhva zitsah (1.154.5). Visnu might also be
called the god of pravrtti- or cosmic progressiong4 but while Held rightly
remarks that nivytti- (regression) was also the movement of Visnu-NLtrByana
in later Hinduism, "the sleeping Visnu, symbol of life that has ceased from
activity and become merged in death,"g5 references to this belief are lacking
in the Veda, On the other hand there is clear evidence of Visnu's connection
with the mountains: he is "dwelling" or "standing" on the mountain(s)
and "regent of the mountains."" The specific mythological significmce of
the mountains (or mountain!) in the Vrtra-myth is the primeval hillIs7and
if the concept of Usas dwelling on the surface of the mountain (ndris&zu-)
has rightly been associated with her cosmogonical appearance, Visnu's epi-
thets will also have to be referred to the god's epiphany: he stands on the
summit of the mountain (I.155.l) . g S Particularly remarkable is 1.154.2:
"Because of this his mighty deed is Visnu lauded, like some fierce beast that
is much dreaded, That wanders as it lists, that haunts the m0untains."~9
Since he assunled different forms (VII.100.6) it may be asked if the mrg6
bhzmcih may perhaps be his serpent form (cf. VI11.93.14, V.32.3; 34.2). Later
art represerks him standing on Mount Mandara, and arising from it as the
cosmic pillar. I n the middle of the seventh century, A.D., an artist a t MLtmal-
lapuram portrayed the god, while taking his three strides, as being the sup-
porting pillar of the Universe.loO
O2 Trita Xptya (I), p. 93.
sanzz~dr6sthall kaldiah somadhhnah.
94 See Held, op. cit., p. 128 and see above, note 19.

Q 5 Held, p. 145. See especially Hopkins, Epic Nythology, p. 207 on Vi,~nu as sa~phartr-.
P E giriksit- 1.154.3, g i r i ~ t h 6 - 2, w i ~ p u hpdroatdnam (scil. Q d h i p a t i l ~ )TS.
, 111.4.5.1.
Doubtful is VS.*16.29 ndmo girilayhya ca l i p i v i ~ f & aca, which Rfahidhsra refers t o Siva.
See IIJ IV, pp. 219, 222.
Q 8 (Indra-Visnu) yh s6nuni pdruatd?xi?n . . . tasth6tur; cf. V.87.4 ddhi snzibhir and see
IIJ IV, p. 226 on a d ~ i s d n u -epithet
, of Usas.
9 9 An explanation of Visnu's relation t o the mountsins has been suggested by Olden-
berg, Gott. A'achrichten 1915, p. 375, and (for ICrsna) by R. Otto, Zeitschrift f u r Missions-
kunde ~ n Rcligionszui~senschaft,
d 49/10 (1934), p. 293. Keither seems plausible.
looE.g., Stella Ilranlrisch, T h e Art of India, D. 206. Cf. AthS. X.7.32; 35; 38; 41.

(Editor's Footnote : First published in Indological Studies in Honour of W. Nonnan Brown, American
Oriental Society, New Haven, 1962.)
1, The first stanza of Zarathustra's Song of the Choice (Y. 30.1), in
spite of countless translations and interpretations, still continues to pre-
sent considerable difficulties to modern interpreters. The vulgate text
reads :
a] td vaxiyd is'antd yd mazdded Izyajcij viduE
staotdcd ahursi yesnydcd va~h3uimanaghd
humqzdrd a3d yecs yd raodbii darasati rtrvrfzdl
A normalized Old Iranian reconstruction of it would be much as follows:
[dt td vaxiyd iiantah yd mazdded yat cit visuiai
stautd ca ahurahya yasniyli ca vahaui manalzah
humanzdrd ........ yd raucahpis' darsatd vrdzd]
It goes without saying that such a reconstruction is only approximate and
cannot replace the vulgate text. It is not certain, for instance, if (and if so,
in which positions) Zarathustra pronounced voiced fricatives for [b, d, g]."
On the other hand [vrdzd] instead of [rvdzdI3has only tentatively been
assigned to this dialect on account of the early date which the present
writer is inclined to ascribe to Zarathustra.* In justification of this pro-
The reading of the MSS. is indeed raoc5biS, not raocabiS(Schlerath, OLZ, 57 [1962],
col. 574).
a For references see Duchesne-Guillemin's account of Old Iranian studies in Kratylos,
7 (1962), p. 7, Benveniste, Etudes sur la langue ossste, p. 18.
S The metathesis vr->rv- must be of comparatively early date, see Tedesco, Zll, II
(1923), p. 53 f. (evidence from Syrian), Karl Hoffmann, Altiranisch, p. 12 (Median and
Elamite evidence), and Benveniste, Etudes sur la Iangue ossste, p. 35. Parallel develop-
ments are found in, e.g., Cyprian bu~tvafor *buqva = &pva (Hesychius) and in Old
Frisian ruald, rwald for wrald "world". According to a French grammar from 1595
English written "se prononce comme si r &toitdevant W". See K. Fokkema, Uit het
spel der klanken, Enige beschouwingen uit de friese klankleer (1952), p. 20.
"ee ZZJ, V (1961), p. 43. The same early date that was assigned to Zarathustra by
D'jakonov and Oranskij is now assumed also by K. Rudolph, Numen, 8 (1961), p. 81
ff., and Eilers, Religion in Geschichte und Gegenwart3, S.V. "Zarathustra", to whom
Schlerath, OLZ, 57 (1962), col. 583, refers.
 nstruction it may, how
retations would not ffered if the translators had
t to explain what, i

tual problems of Y. 30.1 is the form ahurdi in the second

holomae, Reichelt, Nyberg, Duchesne-Guillemin, Gahl,
Tavadia and Humbach have taken to be a dativc5 The first hemistich of
the Ahunavaiti lines (Y. 2 consists almost without exception of seven
syllables. Eight syllables occur in 3 .8a (perhaps also in 30.3b, where
howeter, ca may have been seconda inserted, and 32. l b, where rnandf
m = mndi) but there is little evidence for hemistichs of six syllables:
in a_tmay stand for disyllabic [aat], see Oldenberg, Note;? ad RS. I.
6.4; as for 28.5 and 33.2~see below; for 32.11b the reading aghavascd
(Humbach) is hardly correct.
It would be tempting, therefore, to assume a disyllabic ending
, ~ more so as the accent of Greek Q@ and the intonati
a h ~ r d ithe
Lithuanianpaskui, vilkui point to a prehistoric contraction "-0+ei > *-W.
In the Vedic language, however, there are, out of a total number of 157
occurrences of asmai (dsmai, asmai, asmai-asmai) only four (according to
Olde~~berg)~ where the ending may possibly be scanned as a disyllable.
The very few instances of asmai which Oldenberg considers certain (only
$11.13.1 and VIII. 31.2) form a remarkable contrast with the 150 odd pas-
sages where the ending is monosyllabic. Besides, no parallel case of -ai is
stated to occur among the 84 Rigvedic occurrences of tdsmai (56), ydsrnai
(22), kdsmai (5) and arnQmai (1). In the Gathic dialect ahmcli, yahmdi
and kahrndi have a monosyllabic termination -di (cf., e.g.,
the nominal dative in -di or -di ri seems to be generally considered as
'.
Cf. Bartholomae, Altiran. Worterb., col. 1161 (= "des Ah."), Reichelt, Awestisches
Elementarbuch, pp. 244,498, Avesta Reader, p. 189, Nyberg, Irans forntida Religioner,
p. 180 f. (accepted by Duchesne-Guillemin, Zoroastre, p. 238, [against JAs, 1938, I,
p. 2461, and by GaBl, Acta Orienralia Acad. scient. hungar., I1 [1952], p. 177), Tavadia,
indo-Iranian Studies, I1 (1952), p. 88, Humbach, IF, 62, p. 305, Die Gathas, I1 (1959),
p. 19.
su" Andreas, Gott. Nachr., 1909, p. 44, Humbach, Die Gathas des Zarathustra,
I1 (1959), p. 19. Similarly Taraporewala, BDCRI, X (1950), p. 43, who assumed ahuriif.
See Wackernagel, Altindische Grammatik, I, p. 51f.
W 4 i e Hymnen des ljigveda, Bd. I , Metrische und textgeschichtliche Prolegomena
(1888), p. 188; but see his &gveda, Textkritische und exegetische Noten, ad V . 33. 1,
VIII. 2.41 ; 62.1 ; 103.6.
The same is true of ahmiii in the later dialect, see Geldner, uber die Metrik des
jihgeren Avesta (1877), p. 9. The argument is based on the assumptionthat the element
added to *to- is *-smo-,cf. tdt : trfsmcit. Hence tcismai stands for PIE. *tdsmti(< *to-
smo-ei), just as virdy-a stands for P E . *wirZi.
 c di, it is true, there are

metre invites a disyllabic s of -di, viz. 3 3 . 2 ~tdi v&& rddanti an

yllable occurs in 28.5b

n the other hand, no
which should be read

efioitely be ruled out, although the isolated

, when considered in the light of the entire

"dativus pro genitivo", though plainly unacceptable

Gathic dative forms in -di see Karl Lichterbk, KZ, 33 (1893), p. 196.
Not included are ahurdi and afdi some passages they cannot be
and 27.14 vahiStdi, 43.9e manydi, rdi (but 2 8 . 3 ~33.13a and 46.1
included). The only instance of -di in the dative that is mentioned by Bar-
Volkes(1879), p. 7, is mazcldi,
l1 Bartholomae, Die Gded's (1879), pp. 16,35 and 17,61, followed by Maria Willcins
'es in the Syntax of the Gathas of Zarathushtra (19291, pp. 88, 152.
IV (1960), p. 246.
Forschungen, I11 (1887), p. 14.
,Gott. Nachr., 1911, p. 25, Hurnbach, Die Gathas des Zara-
 calfunction of dahdkriiand ahurdicannot reasonably be doubted. Similarly
Yt. 10.89 zaota ahurdi mazddi, zaota amaSanqm spantanqm must obviously
be read [zautd ahurahya mazddh]. In these passages the purely graphic
origin of the spelling -di for -ahe, due to a confusion of aleph and cheth
in the Sassanian script, has long been recognized.15 The same graphical
explanation eliminates (as Lommel has been the first to perceive) the syn-
tactical difficulty of having to take ahurdi in 30.lb as a dative.
The assumption of a genitive form in -di for [-ahya] in the Gathas is
supported by a parallel instance that has been pointed out in 32.6~
Bwahmi v3 mazdci xJaBrdi aJdicd sanghd vidqm.
Humbach here assumes 7 + 8 syllables by reading vidqm "sol1 ausge-
breitet werden". Bartholomae's analysis as vi-d(fi)-cim - which Humbach
accepts - was, however, based on the idea that imperative forms in -dm
(instead of -tarn) could be derived from any verbal root. Despite Gathic
4cqm in 48.9~(as against 48.7a -dycitqm),lBboth the Vedic evidence1?and
the general rule that -am occurs with verbs that have r in the 3rd pers.
plur.ls plead in favour of Thieme's proposal19 to take vidqm as identical
with Vedic vidcim (AthS. V. 30.13)' which belongs to vid- "to find". Jn
that case, however, vidqm cannot be trisyllabic and since second hemistichs
of less than eight syllables are unknown in the Gatha Ahunavaiti, aidicd
must necessarily be read as [rtahya C~].~O
More doubtful is 28.5b
gdtzimcd ahurcii saviJtdi saraoSant mazddi,
where Humbach proposes to scan gifttimcd, which would yield 7+8 syl-
lables. However, not only is a disyllabic a" unlikely in *gaHtu- and has the
Veda no instance of a trisyllabic *ga"tu-,but since -c&connects gdtrim with
the preceding word vat?damnd, it is also impossible to ignore such Vedic
l6 See Lommel, Festschrift Andreas (1916), p. 104, B. Geiger, Die Am& Spantas
(1919, p. 68, Meillet, JAs, 1920, I, p. 198.
l 8 See Bartholomae, Grundriss. Iran. Philologie, I, Reichelt, Awest. Elementar-
1 buch, pp. 129, 139. It should be noted that Vedic -( s monosyllabic (for cistam X.
85.11 see Oldenberg, Noren, a.& so that a trisyllabic vidqm would necessarily have to be
taken as *vi- d a m , an innovationfor proto-Indo-Iranian *vi- &H-dm.
l7 Macdonell, Vedic Grammar, p. 339.
l8 See Wackemagel, Kleine Schrvten, p. 31 1.
lS Thieme apud Altheim, Paideuma, III, 6-7 (1949), p. 273.
Lommel, Worter w?d Sachen, 19 (1938), p. 241 ("wahrscheinlich zu Unrecht",
Schlerath, OLZ, 57 119621, w l . 575). Altheirn su ted to read *aSdyacd. Bartholo-
rnae's radical emendations (Die GdBd's, 1879, p. 33), accepted by M. Wilkim Smith,
Studies in the Syntax of the Gathas (1929), p. 84, were based on an incorrect analysis of
afdicd.
 ndation of the last word as in the UAv. passages quoted
cision must depend on the value one attaches to the metri-
d on one's general ideas about the reliability of the text

for a metrically abnormal line of 7 10 +

would be deleted as an
Jahyn &tahytra]. From a purely
azdd d a m c d 28.9a,
first word is also likely to be an
e assumption of synaloephe in
sandhi phenomenon in the Ga-

a Cf. Wumbach, IF, 63 11957), p. 102, n. 4; dt, IIJ, X (1957), p.

167 ff.
l e the last two cases are admittedly doubtful, ddicci in 32.6~would
fairly certain instance of the spelling -di for -ahya in the Gathas,
is parallel to those occurring in the Younger Avesta. For those
scholars who hold Andreas's theory to be no Ion tenable, it may be
useful to stress the fact that, while Lommel's and llet's graphical in-
terpretations of the Gathic passages just mentioned are based on that
theory, the procedure of explaining h a 1 -& as a misreading for HY
[= ahyaj in the older script is much older than Andreas's theory. The
mere comparison of Yt. 5.4; 8.31 zrayd (v.1. zraycii) vouru.kaJayct with
Yt. 5.3; 8.30 aoi (avi) zrayd vouru.kaziarn is sufficient to show that zrayd(i)
must be a locative, parallel to vouru.kaVay-a,and consequently stands for
[zrayahi]. As far back as 1889 Bartholomae, ZDMG, 43, p. 668 n. 3,
pointed out that zraydi owes its origin to a misreading of an older manu-
script (with ci standing for h) and a few years later he and Mills drew at-
tention to the additional evidence of the ending of the 2nd pers. sing. in
-c2for [-ahi]." In the Altirunisches Worterbuch, col. 1702, Bartholomae
still maintains this graphic explanation, which Reichelt, Awestisches
Elementarbuch, p. 34, omits to mention. That Bartholomae's explanation
of these YAv. readings agrees in principle with Lommel's and Meillet's
is not surprising, as the conviction that our Avesta text is a transcription
of an older text, written in some sort of Pahlavi script, can be traced back
to the middle of the 19th century.26 The same interpretation, based on
the identity of the characters for aleph and h in the Pahlavi script, was
also proposed for Aog. 28 agrcli = [ahrahya] by Duchesne-Guillemin,
JAs, 1936 I, p. 246.

2. These readings, which unequivocally prove "the defective nature of

the archetype of all the manuscripts" (as Meillet put it),26point to the
conclusion that at the time when the older text (written in Pahlavi
characters) was transcribed in the newly created Avestan script, the exact
forq and the meaning of the text was, in some Gathic passages at least,
no longer precisely known. This means that the authentic text of the
prophet's own words had not been entirely preserved by an uninterrupted

Asiens im griechischen Zeitalter, I, p. 91 n. 75. Most modern translators, however, as-

sume a correlation between ya0d and ifhiin 33.la. As for 28.9a, both Andreas-Wacker-
nagel and Lentz, Yam 28, pp. 31,73 f., are inclined to delete nzazdii.
a%holomae, ZDMG, 48 (1894), p. 149, and Mills, ibid., 49 (1895), p. 483.
references in IZJ, V (1961),p. 40 n. 4 and cf., e.g., Roth, Yapa31(1876), p. 17,
Bartholomae, Die Gcieci's (1879),p. 2. Lommel's explanation of zraycii was accepted by
L. H. Gray, JAOS, 61 (1941), p. 104n. 26.
JAs, 1920, I, p. 195, MSL, 21, p. 189.
character of this protraction of hal vowels is particularly clear in Z m -
Bu&rahd 53.la, 3b, the only Gathic instance of a genitive form in -ahe for
final vowel. -e must be due to a misinterpretation of the
t date from a relatively late peri
i, which occurs beside (~Jahyd,
sho
ve" spelling'in th
that it preserves

e as an introduction to a discussion of the text

yecd yd raodbig darasatd urvdz6, to which I
shall here confine myself.
That the meaning of this line was no longer own t6 the author of the
,who held it to refer to the urviixm i amahraspandcin "'the bliss of
Sa Spantas", appears from Schaeder's discussion of this passage
h his Iranische Beitriige, I, p. 90. Bartholomae suggested that yecd here
stands (quite irregularly!) for ycica'?' but his various attempts to translate
,e.g., Woffmann's well-baland judgement in Iranistik, I, p. 9 f. There is, ac-
cordingly, a core of truth in Andreas' theory, despite Schaeder's denial in ZDMG, 95
(1941), p. 298. The existence of a W Avesta during the reign of the last
Arsacid is now an established fact. assumption of a phonetic develop-
ment -ahya > -ahe > -di (BB, 17, artholomae, Grundr. iran. Phil., I ,
p. 232.
Cf., e.g. Bartholomae, Die Gd6d's und heiligen Gebete (1879), p. 10 f., M
stierne, NTS, 12 (1942), p. 32 n. 6, Humbach, Die Gat&, I, p. 23.
Essays on the sacred Language, Writings, and Religion of the Parsis (2nd
ed., 1884), p. 73, Bartholornae, Die Gdod's (1879), p. 68, Reichelt, Awest.
Elmentarb. (1909), p. 34, Meillet, JAs, 1920, I, p. 198, Schaeder, Iran. Beitrige, (1930),
p. 261 n. 5.
L o m e l W6rter und Sachen, 19 (1938), p. 241.
:sumatdprylyena y& ca, y& cdntar roc&idlokanenadnaruiit4.
m e in taking aSd yecd as standing for aSd *y&d (Gat. Nachr.,
these words into Latin only show (what is obvious at first sight) t
construction with two relatives, one of which follows immediately after ~

the other, is impossible. Indeed, "ea praedicabo ... quae (sunt) record-
anda ...' a bene recordante una cum A3d quaeque, quae una cum luce
videatur voluptas" is not intelligible in Latin either, for quae is here re-
dundant, and quaeque voluptas quae can hardly be paraphrased as et
volecptatem (as Bartholomae does).32 The difference between the con-
struction here imputed to Zarathustra and, e.g., Plaut. Amph. 779 tu qui
quaefacta i ~ ~ f i t e a rthe
e , ~meaning
~ of which is clear at a glance, is obvious.
Apart from these syntactical objections there is the diEculty that yd is
never written as ye, except before m and n.34 On the other hand, there
are in the Gathas eleven occurrences of ydcd without any variant reading;

1909, p. 44). Only in the Altiran. Wiirterb.,col. 1217 is it written +ye (NSf.) 56. Spie
Commentar iiber das Avesta, 11, p. 220, left it untranslated.
a2 Altiran. Worterb., col. 566. Cf. col. 1834: "quae-que laetitia", "und von der
Wonne, die von dem, der es [naml. mazd&?] sich gut merkt, (zugleich) mit dem Afa,
die (zugleich) mit dem Lichtraum erschaut werden kann"(thus also in Die Gatha's des
Awesta, 1905, p. 13); col. 1217: "et (voluptatem) quae una cum ASo quae una cum luce
videatur voluptas". Much clearer is at any rate Andreas' translation (Gott. Nachr.,
1909, p. 48): "die Lobpreisungen und Gebete) die sich der Wissende insgesamt merken
muss, sowie die mit dem sehr weisen Recht verbundenen Freuden, die mit ihrem Licht
herrlich anzuschauen sind". In the essential points nearly all modern translators agree
with Bartholomae and Andreas, e.g., Moulton, Eurly Zoroastrianism (1926), p. 172,
Christensen, Avesta, Zarathushhiernes hellige Skrifer i Udtog (1923), p. 4, Schaeder,
ZDMG, 94 (1940), p. 403, Barr, Avesta (1954, p. 91, GaAl, Acta Orientalia acad. scienr.
hungar.,I1 (1952), p. 178, who renders "quaeque quae splendoribusconspicietur laetitia
(= quaeque beatitude sit, quae una cum sapientissima Veritate in splendoribus cae-
lestibus conspicietur), "und was fiir eine Freude (= himmlische Seligkeit) die ist, die,
im Verein mit der sehr weisen Wahrheit, durch das (himmkhe) Licht (oder: in der
himmlischen Herrlichkeit) sichtbar wird". Duchesne-Guillemin, Zoroastre (19
238,296a has : "Et la joie que verra, par la justice, dans 1a lumihre celui qui les aura bien
retenues". Only Humbach, Die Gathas, I, p. 84, gives an entirely different rendering:
"Ihr Wohlachtsamen! In Wahrhaftigkeit niihere ich mich mit der Erhebung, die durch
die Lichter sichtbar ist", the meaning of which he characterizes as "recht unklar" (vol.
11, p. 19). Earlier translations like those by Haug, Essays on the sacred language (etc.),
2nd ed. (1878), 3rd ed. (1884), p. 149, Geldner, Studien zum Avesta (1882), p. 47, KZ,
27, p. 286, 28, p. 409, Hiibschmam, Ein zoroastrisches Lied (1872), p. 13,
ZDMG, 42 (1888), p. 82, Bartholomae, Arische Forschmgen, I1 (1886), p. 11
passed by in silence.
Discussed by Benveniste, BSL, 53 (1958), p. 52.
Gathic ayeni 34.6c, 46.la, 50.9a, YAv. a&i V, 3.27, uzayeni Yt. 15.32, V. 22.1, etc.
Cf. Grundr. iran. Phil., I (1895), p. 174 (§298,5), where, beside zbayemi, ayeni and
hdcayene, as an isolated instance of e before c is mentioned yeco "ich will bitten". This
must refer to Y. 51.2a, see ~ r i s i h eForschungen, 11 (1886), p. 118 (in Die GdBd's, 1879,
in 30.1~ is taken as *yZcZ). As for riyesl53.6c, this does not stand for *dy&l
omae, Die GdBii's, 1879, p. 75, Altiran. Wort 289, W. P. Schmidt,
,but is equivalent to Skt. dyacche; see the
 8,lOd, 50.10ab. For this
reason alone yecd cannot mean "and which", nor "I approach, I ~ o m e " . ~
The correct interpretation of d d yecci emerges from ajdic& 32.k beside
dahydcd 30.10~~ 31.21b, as in the above discussion (p. 99). Since the
character for h in the older script was occasionally misread as an aleph
and incorrectly transcribed in the archetype which was written in the new
Avestan script, we may expect to find in the Gatha text -dya/-cii beside
-ahyd/-a&. At the time when the final vowels came to be writ
&@a ca was no longer reco
the scribes to split up WO
they obtained pronominal formsS6*a$dyaca was written as *aSd yaca
and in the last "word", which regularly became yecd in the ortho
of the later manuscripts, the short final vowel of -ahya was p
(It must be admitted, though,
taken to stand for varazanah-
readings of the manuscripts
ted a priori: beside d d
ydcd Lm,aJciycicd J,,
dcii yecd J,, a&i yaecd S,.a7 Since the corrupt readings a3d
must date back to the time when the old text was tram
script and since they point to the conclusion that their real meaning was
r h o r n at that time, the variant readings must apparently be
traced back to the archetype in Avestan script. They raise no doubt some
problems of their own, but they do not teach us anything about the orig-
inal form of the word, which Zarathustra must have pronounced as
[rtahya]. Cf. 53.6d, 9b and YAv. anarata-, Astvainata- (etc.).
As a result the following translation may be su sted for the whole
W I will tell those things, 0 you who are coming (?) to

xnb8rOyd 43.2b, yd hi for yiihi 46.14c, 49.9d, mdlrtri for [nui it tu] Yt. 10.69, vwazi nd
which should be noted whatsoever (?)by the initiate, th
the prayers to Vohu anah, O very wise men,3oa
ch manifests itself40t ther with the lights".41

3. That urvcZzti in Y. 3 0 . 1 ~demands a genitive appears from Y. 36.2

urvriziitahyd urvrizyri "and with the bliss of the mos
rectly from Y. 32. (ahydcli ...) dhurahyd urvizizam
bliss of this Ahura azdsi [shall the nobility beg]".
This bliss, which Ahura d;i grants the soul during or after life-time
(urvdsisma ... vahiStam ah 38) can hardly have been different from
"the most blissful union with ASa".42 0th notions are inde
lated. Thus the intoxication caused b Baoma is said to be a
ASa and bliss (or, as it is mostly translated, "by blissful ASa"
Asa, who is the friend (32.2b) and "prudent counsellor'"
azda, and who perhaps is said to dwell in his abode (30.10bc),

lia acad. scient. hung., I1 (1952), pp. 175, 178, Tavadia, Indo-lranian Studies, I1 (1952),
p. 88, Numbach, WZKSO, I, p. 84 n. 13. Lommel gave the translation "herbeistrebt"
as an alternative in Gott. Nachr., 1934, p. 96 (thus also Humbach,
as Cf. mqzdra- (Andreas, Gott. Nachr., 1909,p. ,Duchesne-Guillemin, Les eomposks
de I'Avesta, 1936, pp. 101, 131, G d , Acta Orient., 11 [1952], p
Altind. G r a m . , I, Nachtrage, ad p. 82,35 f., Humbach, MuSS, 9,
83 n. 9). However, there is no direct comexion with Ved. mddhiq- which stands
*#W-dhH-ro- and takes the place of 'medhdh (cf. also Wackernagel-Debrunner, A&&.
Gramm., 111, p. 362). The form humqzdr8 has sometimes been construed with &ci
(Andreas, G&, Acta Or., 11, p. 178: "sapientissima Veritate", "im Veaein mit
weisen Wahrheit"), which is implausible from a semantic point of view and,
excluded if dci yecci stands for a genetive. So humqzdrd most likely refers to the
addressed by the prophet. Since an instrumental case cannot easily be construed with
darasatd, the only possibility that remains is to take
Arische Forschungen, 11, p. 118, Wumbach, Die Gathas, I, p.
darasatd :Ved. dardatii- is the typical word for the e
Agni, and the sun, and is further used in comexion with vcipup, all of which belong to
the mystery of the nether world (see below). d . should not be taken, therefore, as a ver-
bal form.
" No attempt will be made here to explain 51.2ab tci paourvim ahurci aSdi
yecci / taibydcci drmaitz &i&i mcii LftciiS xSa8ram, althou tive afahyacd would
be possible if taibyd xSa8ram is a stylistic peculiarity of as for tavd xSaBram
53.9d, 8wd xSa8rci 43.14~.It is taken in this sense by, e.g., Geiger,
p. 207. Since dcii yecci and faibycicci seem in any case to be par

Y. 10.8 h6 yci h a o d e m& aSd hacaite wvamana.

is also closely associated with the latter's "dominion" (xkOra-). The last
* must either be translated as "in thy dominion, 0 Wse, and
in Ala's shall your (plur.) praise (or, andment) be proclaimed (or,
found)", or else as "in thy dominion, ,shall your (plur.) and ASa's
praise be proclaimed (or, found)". In both cases the implication is clear.
Thus in 49.8 the prophet prays that Ahura MazdZi may "grant the most
blissful union with ASa" to Faralaoltra and himself: "in thy dominion,
for all time, let us be (thy) beloved ones".45
. Geiger has shown, 46 this association of the word xJaf3i-a- with
God and ASa is one of the traces of Old Aryan cosmological ideas survi-
ving in Zarathustra's theology. In the Rigveda Mitra and Varuna are
kgatriyyd (dual), the lords of pta and light, and the "luminous dominion"
is theirs.47 The question as to where this domain was situated will, there-
fore, have to be studied more closely.
Varuna is the god of the primeval waters which belong to the p
world of Chaos. He is an Asura, and just as the Asuras are called
brothers of the celestial gods ( d e ~ c i s )and
, ~ ~just as Varuqa's an
reign (asuryd-) is said to be "the oldest", 49 SO mention is sometimes made
of the "first gods".60 Hence, when V a r q a is said to have "measured the
first creation" or "the first place"5l this is likely to refer to the pre-Indric
world governed by Varuna and the Asuras.
After the primordial hill has arisen from the wateis, these are thought
"M 32.6~Owahmi v5 rnazcki xSa0rcii &died s5ngh6 vidqm.
a 49.8cd vaghdu Owahmi d xSa0rcii / yavcii vispdi fraCStdgha dghrimd.
Die AmaSa Spantas, pp. 204, 232, esp. p. 207.
k,mtriyri VII. 64.2, VIII. 25.8; rtdsya jydtimpdtiI.23.5; jydti~matkjatrdm I. 136.3.
RS. X . 124.4,151.3 (with Geldner's notes on his translation), AS. VI. 100.3 (P. von
Bradke, Dyau~Asura, p. 89), MS. IV. 2.1, SB. 1.2.4.8, XIV.4.1.1 (BAU. I.3.1), TB.
11.3.8. l f., 11.2.9.5 and, e.g., Bergaigne, Lnr Religion vkdique, 111, p. 1 E.,Hillebrandt,
Ved. Mythologie, IIa, p. 427 31.2, Neisser, Zum Worterb. des Rigveda, I, p. 142 f., J. J .
Meyer, Trilogie der Vegetationslniichte, 11, pp. 3, 257, 111, p. 211 (but &o Index,
p. 288!). As for KS. XXVII. 9 (p. 148, 17 f.) te 'surd bhiiydrhsa.4kreydrhsa b a n , kani-
.
ycirPls4 pdpiydrfisa dnujdvaratard iva &v& . . the reading with dnu- is of wuse quite
correct (against my suggestion, IZJ, TV, p. 219). The different age of the older and
the younger gods, which has often wrongly been explained from a supposed
change in the cult, was still known to the authors of the Mahdbhdrata, cf. WI. 34.13
(crit.4.) asurd bhrc?taro jye~fhi&v& cdpi yaviyasab. See Indological Studies in honor
of W. Norman Brown (1962), p. 145 n. 51.
VI1.65.1~yciyor asurydm &itam jyk~fham,cf.IV.42.2b ah&&rdjd vdrwo mrihyam
t h y asury&iprutharmf &&ayantu, and see Geiger, Die AmaSa Spantas, p. 205 n.
VII.21.7 &v& cit te asurydya pbrvk 'nu k$atrdya marnire s#uinisi, X.72.3 devdnrim
0 yk td &an &a jdtd devil devkbhyd purd, Macdonell, Vedic

" lTII.41.10~ S& piirvydm mame, and 4c sd d t i i pkvydm p d n ~ .Cf. also

Liiders, V a r w , p. 712 n. 11, and Renou, Festgabe Lommel, p. 125 on AS. III.13.2.
of as being covered by, and enclosed in this mountain. The earth is sai
not only to lie upon the waterss2 but also to envelop the waters like
"recepta~le".~~ By delivering the waters from this receptacle Indra in-
augurates the "second creation". In other words, since the hill is the
sacred image of (and, as such, identical with) the earth, the primeval wa-
ked upon as a "hidden ocean" under the earth - a notion also
bylonian mythology.64
Varuna, accordingly, is the god of the nether world. Here he resides
and watches over the Cosmic Order (Ptci), which is said to be '"fixed and
hidden where they unharness the horses of the sun".56 There can hardly
be any doubt as to what is meant here. According to the Atharva-Veda
Agni becones Varuna in the evening, and the Kaujltaki-Brahihmaga states
that the sun, "having entered the waters, becomes V a r u ~ a " . So,
~ ~ while
setting in the western ocean (where according to $23. X. 6.4.1 the night is
born) the sun assumes the character of the god of the nether world, who
in the classificatory system is always associated with the western quarter.
Indeed, the words rtdsya ydnih "womb of the Pta" are said to be equiva-
lent to "water" (Naighagtuka 1.12), while the parallelism betweenparavdt-
and the "seat of &tawin IV. 21.3 suggests the inference that that "seat"
was situated in the nether world. Thus the horses of the Sun can be said
to come "from the seat of Pta" (see note 122),just as "the sun arises from
the earth" (I157.1).
.
AB. III.6.4pphivy apsu (scil. pratiqfkitd). See also note 125.
MS. 1.4.10 (P. 593) yLirhy apdgyhpiycldimdm tdrhi m h d dhydyed,iycim v6 erasdrh
pdtram. Similarly KS. XXXII.7 (p. 26, 14 f.).
64 Cf., e.g., W. B. Kristensen, Symbool en Werkelgkheid, p. 286. For the Indian ideas
see Hopkins, Epic Mythology, p. 26: "But it must be remembered that 'under earth' is
water, a part of Varuga's domain", and Liiders, Varum, pp. 315-322 "Der Samudra
irn Felsen" (but on p. 34 he wrongly states that "das unterirdische Totenwasser" is
"ganz vergessen" in India); cf. also Renou-Silburn, JAs, 1949, p. 13 on the atmospheric
water-basin "semblable aux ffots de l'oct5an, cette tournoyante qui, semble-t-il, soutient
les astres".
RS. V.62.1 rt&a ytdm cipihitarh dhruvdrit vcilh shryasya, ycitra vimucdnty &v&. 'I
fail to understand how Geiger, Die AmaSa Spantas, p. 168 (cf. pp. 156 n., 176) can
speak of "des in den Naturvorghgen wirkenden Pta, das am Hirnmel (B V. IV, 42,4;
V, 45,7f.), dort, WO man die Sonnenrosse ablost (V. 62,l) seinen Sitz hat", and how
Liiders, Varuw, p. 318, can state that only the highest Heaven can be the resting-point
of the sun, which he finds indicated in the words mdr ycid&man (seealso pp. 321,328).
For the use of dhruvd- with special reference to the nether world see n. 68.
AS. XIII.3.13 sd vdrupab sdydm agnir bhavati, KaqBr. 18.9 sa vd e$o 'pab prav3ya
varupo bhavati, perhaps also RS. 11.38.8 y&irddhy&h vciru~oy h i m &yam cinifitath
nimisi jdrbhurdwb, VII.87.6ab civa sindhrm vrirugo dyadr ivu sthdd drupsd nrf Svetd
m r g b t&i~mrZn. Thieme, Mitra and Aryaman, p. 71, and Renou, Festgabeyur Herman
Lommel, p. 125, hold AS. XIII.3.13 to be due to a secondary development.
mark in ZZJ, 111, p. 210 and cf. K. Hoffmann, OLZ, 49 (1954), col. 394,
yet bear that name (presumably of foreign origin), as this text is less con-
cerned with primitive geography and much more with mythical cosmolo-
the sacred image of the earth, is gener-
i-, etc. On the other hand, the axis is
the cosmic tree, which in the ancient mytholo-
gical conception of the world arose from the primordial hill. Therefore,
arupa, since he supports earth and heaven by means of the central pil-
lar of the Universe (skambhd-), may be supposed also to hold the roots of
that cosmic tree (see below, p. 116).
Varuga's "lofty dwelling", his "house with a thousand doors"61 is also
called a "stone house" (harmyri-). In the early morning the goddess
g from the nether world, is said to come "from the h
,just as Agni is born in this stone house before beco
navel of the radiant The notion of darkness appears to
intimately associated with this "stone house". It was, indeed,
place of the dead, just as Varuna was the god of death,
Yama was supposed to dwell in it. The same association with d
is also found in the story of Indra bringing the bellicose SuSna "into the
darkness, into the stone house"85, and indirectly in the words "blinded
in the stone houseY'whichthe Rigveda uses with reference to Kanva. As
the Jaiminiya Brcihmma tells us, Kanva, though warned, inadvertently
sat down on a seat, which immediately became a stone and enveloped him.
This stone, the text says, "was darkness, as it were".@@
What is said of Yama's abode must also be true of Varuna's, for the
dead who follow the paths along which the blessed fathers have gone
"will see both kings, Yama and the god Varuna, revelling in their partic-
ular ways".67 Varuna's nether world is called a "stone house" because
he dwells in the depth of the cosmic mountain. (Cf. IIJ, IV, p. 249, V, p.
46 f., and J. J, Meyer, Trilogie der Vegetationsmachte, p. 205 n. 2, p. 230
E.). It is no mere coincidence, indeed, that an epithet of the mountain,
viz. "whose dwelling is firmly fixed", is also applied to Mitra and Varuna,
and that the Adityas, while "having a well-founded abode" (in the earth),
cause the sun to rise.8s
In this connexion mention must be made of an old epithet of the Adit-
yas that occurs in RS. 11. 27.3: "Those Adityas with many eyes, who are
wide (U&-) and deep (gabhird-) and undeceived, although they wish to

ky$&& pttvam is explained as follows: tamo vai k y ~ p d hmrtycis

, tdmo, ... Caland's
suggestion that the world of the dead originally was in the west should be noticed in
this wnnexion.
s"S. X.114.10, AS. XVI11.4.55 (see below, n. 67 and n. 73).
V.32.5d yliyutsantarh tdmasi harmyt a%&, cf. tamogdm in 4.
VIII.5.23 yuvdm kd~vciyanrisatyd 'piriptciya harmyt i&vad iitfr &iasyatJ&. Cf.
1.118.7 yuvd& kh~vciyd'piriptciya c&& prdty adhattam s@utim juju$i&, JB. 111.73
(line 5) sd (scil. dsandi) hainam tad eva iilci bhgtvci 'bhisamvive~fi,111.74 (line 4) tama
ivdrit. In view of Bild it should be noted that in the Rigveda harmya- is equivalent to
&man- (Geldner, Kommentar, p. 114).
@".14.7 ubhd rod& madhdyd msirlantci / yamdm pdycisi vdruwm ca devrinl. Cf.
masti te Varupo rcijci Y a d ca samitimjayd Mhbh. III. 140, 13 crit. ed.
V.72.5 dhruvdk$enui(Mitra and Varu~a),cf. IV.13.3 (Adiiyas), III.54.20 (the moun-
tains). Varu~adwells in his dhruvdsu k$iti$u VII.88.7, Mitra and Varupa are sittin
dhruvt sddasy uttamt sahhrasthtipe 11.41.5 (cf. VIII.41.9d vdrupsya dhruvtirh S@),
which cannot be different from the sahrisradvciram .
uttamci- refers to the night aspect of the cosmos.
dhruvd dik "the nadir" is equivalent to dvdci dik MS. 11.13.21 (p.
IV.40.5, adhard dik, cornm. on TB. 111.11.5.3. The name Varu~a
have been an appellative for this subterranean domain, cf. VII.
v c i r w bbuvdni. But even if the name Varupa should not be a ta
the absence of a linguistic equivalent in the Avesta hardly allows
sions (Thieme, JAOS, 80, p. 308) iis there is every reason to cons
taboo substitute (IIJ, V, p. 55), whatever the older n
cf. X.124.4.
both far and deep. A decisive argument in favour of this interpretation
is the fact that the blessed dead are also called "deep".71
Therefore, when the sun is called an eagle (supared-) who overlooks the
i), but is at the same time the Asura "with the deep excite-
vepas-)," it should be reme red that the suparpb is

the fact that it was apparently in the nether world that Varuna made
Vasi$$ha an Finally, when the two A4vins are Galled "seers of d
should not be dissociated known fact that the
initiate, who had become e Varuqa himself.77
g to the depth of the ne
mythology; it has often
that the term "deep Ahurayywhich the Avesta use
i t h ~ - ais, ~fully
~ pardel to the "wide and deep"

.
79 Older t r a m
lations used to render this as "the sun in the sky" but since there is not
the slightest evidence in the Rigveda, apart from this passage, for the
meaning "heaven", Oldenberg and Geldner have rightly rejected this
interpretati~n.~~ Indeed, what the.seer aspires to see is the mystery of
Agni's presence in the darkness of the "stone house",s1 just as it had been
seen by those gods and mythical seers who (probably at the beginning of
the new year) descended into the nether world as "sun-finders" (svarvid-,
svardfi-).
The presence of the sun in Varuna's dwelling cannot surprise us, as the
sun is only one of the forms of t%e god Agni. As was seen above, the sun
sets there where e t a is hidden (i.e., in Varuna's domain), the sun when
setting becomes Varuna, and Agni becomes Varuna in the evening (p. 107).
Additional evidence might be RS. X. 8.5b (to Agni): "Thou becomest
V a r u ~ awhen thou comest (?) to Rta". Indeed, although Agni is said to
have left the Asuras as a result of Indra's creation act (X. 124.1, 3, 5),
there is a form of Agni in the nether world into which the setting sun
"enters" and from which it arises in the morning (AB. VIII. 28.9 and 13).
Hence Stirya is called an asurydh purdhito (VITI. 101.12, cf. Eliade, Eranos
Jahrbuch, XXVII [1958], p. 207). From this nether world Agni is every
day born anew. He is, indeed, born "from the waters, from the stone",a3

(Humbach denies any connexion between the Vedic hymn and, e.g., Y. 43; see Die
Gathas, I, p. 14, 11, p. 48). The real character of svard;S- "sun-seer" (ZZJ, IV, p. 220)
appears from this passage, which also explains the original meaning of hvara.darasa-:
this was certainly not "der den Blick der Sonne hat, sonnengleich blickend" (Bartholo-
mae, Altiran. Worterb., col. 1849). Curiously enough, Liiders holds the contemplation
of the "wonder" to be of minor importance (nebensachlich). Cf. especially AS. V.1.8cd
drirhn nu td varutza y h te vi,sfhd dvdrvytatab kgavo vdpiW~i.
80 Geldner, Der Rigveda in Auswahl, 11, Kommentar (1909), p. 114, Oldenberg, Noten,
11 (1912), p. 61, Liiders, Varupa, p. 318. The meaning "heaven" is still defended by
Bartholomae, Zzim altiran. Worterb., p. 173 n. 1, Neisser, Zum Worterbuchdes $gveda,
I (1924), p. 134; cf. also Hillebrandt, Asia Major, I (1924), p. 790: "die einzige Stelle,
wo diman "Himmel" bedeuten k a n n".
X.46.3 jdta d harmyksu. Cf. the horse (= Agni) that Uttalika sees in the nagaloka:
Mhbh. (crit. ed.) 1.3.153, 157, 173. For harmyd- = &man- see Geldner, Kommentar,
p. 114.
ZZJ, IV, pp. 220,242,271 (etc.), V, p. 177. In this sense onz has also to take the
words X.154.5 kavdyo ykgopdydntishryam "the seers who watch over the course of the
sun". Cf. X.67.4 tdtnasi jydtir ichdn (Byhaspati), etc.
1I.l.lbc tvdm adbhyds tvdm dhanas p h i , tvarh vdnebhyas tvcim d~adhibhyas(. ..
jdyase Sicih). The last three words might be interpreted naturalistically, as referring
to the fire generated from the flint and the wood, but the first word ("from the waters")
shows that the mythical birth is at the same time meant. In the Yajurvedic verse (MS.
11.13.13: p. 162, 10; KS. 40.3: p. 136, 5ff.; &S. 16.35) yd apsv dntdr agnir yd vrtrk ydb
puru,se yd liSmani (with a variant in AS. 111.21.1 yk agndyo apsv dntdr yd vytrk yCprirzie
yd dirnaszr) the reference to Vytra would seem to point to mythical, rather than natural-
or "born from the waters, from the cow, from Rta, from the he
is "the son of the and Indra "generated [at the creation] the fire
between two stones".86 Therefore it is for Agni "even in the rock [as it is]
at home".87 Indra found "the treasure of Heaven" hidden in the rock
like a bird's youig [in an egg].@ The sun itself is once called a variegated
stone placed in the midst of the sky,89and the dawns, thought of as cows,
have a rock for their pen.90
In connexion with the fact that the vrajci- or valci-, in which the cows
are penned up, is sometimes called a "sfone" (diman-), or is said to be
closed witha stone,91attention may be drawn in passing to the remarkable
circumstance that the Vedic description of Varuna's subterranean palace
as "supported by a thousand columns" (sahcisrasthzjnac 11. 41.5) or "the
lofty building with a thousand doors"(byhdntant mcfnatn sahdsradvrfram
VII. 88.5), and that of the nether world as "secured by a hundred doors"
(iatcid~ra-),~~recurs almost verbatim in the description which the Mahci-
bhrfrata gives of the sabhd in which the dicing between the PB~davasand
the Kauravas is to take place (11. 45.46 crit.ed.):
stlzzjnnsahasmir byhatim iatadvdrdm sabhdi+zmama
ntanoramdm darianiycim diu kurvantu ii[pinah
Let the artisans quickly build for me a sabhii which is supported by a thousand
columns, lofty and secured by a hundred doors, pleasant and beautiful.
To this passage, quoted by Hillebrandt, Asia Major, I (1924), p. 787,
others might be added, e.g., 11. 5 1.17 crit. ed. :

istic, associations. For the latter see, e.g., Oldenberg, Noten I, p. 7. It should also be
noted that Agni is at night the mkrdhk bhiivo, whence in the early morning the sun is
born. Cf. tcita.fr sdryo jrifate prdtcir udydn X.88.6.
84 1V.40.5d abjd gojd rtajd adrod rtdm.
X.20.7 n'dre.frslinrim dyrim dh14.fr.
11.12.3 yd ciSmanor antar agniriz jajdna.
1.70.4~ddrau cid asmd antar duront.
1,130.3ab avindad divd nihitam grihd nidhim vtr nu garbham pcirivitanz aimany
anantt antar dimani.
V.47.3cd ( a t q d ? ~supar&) madhye divd nihita?~p&ir riSmd vi cakratne rajasas
pdty dntau. See on this stanza A. Kuhn, Abh. kon. Preuss. Akad. Wiss., 1873, p. 144 f.,
Hillebrandt, Asia Major, 1 (1924), p. 790.
IV.l .l 3cd iiimavrajd[z sud~ighdvavri antar lid usrd djanii u ~ a s ohuvdndb.
IV.1.13; 16.6; VI.43.3, 1X.108.6, X.68.4 (cf. Hillebrandt,AsiaMajor,I(1924),p.
789 f.); 11. 24.4 ciJ.mdsyam avatdm; PB. XIX.7.1 asurdndm vai valas tamasd prdvyto
'imdpidhcinai cdsit, iasmin gavyam vasv antar &it, tam devd n&aknuvan bhettum, etc. Cf.
Varunadeva as the name of the stone slab of a well in Chamba State (IIJ, IV, p. 249, V,
P. 52)
1.51.3b utdtraye iatridure~ugdtuvit, which must refer to the nether world on account
of a) tvam gotram ciigirobhyo 'vmor apa and the following stanza.
 tambhdk hemavaidaryacitrdk
SatadvcZrcsli2 torapasphd{iSrligdm
sabhGm agrydk kroiamdtrdyatdm me
tadvistGra'm diu kurvantu yukt*

It does not seem possible to draw any other conclusion from this striking
parallel than that the sabha' was meant as a replica of the nether world.
he problem whether the sdmiti- may then be equated to the upper world
remains unsolved.93 Some confirmation of our conclusion may be found
in the rule that a sabhd should be erected to the south of the town, which
suggests some connexion with the realm of the dead.94 I further refer to
Weld's ample discussion of the sabhd as the sacred initiation hall (op. cif.,
p. 202 ff.), a discussion in which the religious importance of tbe door is
also rightly stressed (p. 212). It is hardly a mere coincidence that Dvfirakfi,
the "Town of Doors" (p. 215) is situated in the western-most part of In-
dia, in Kathiawar on the shore of the "western ocean" (like Pylos in
Greece!). It may be suggested that the town was originally considered
the sacred gate to Varupa's world (which is also situated in the west in the
classificatory system).s5
It was pointed out above that the mythical cows are sometimes sai
be penned up in a rock (diman-),or in a vald- that was closed by an &man-,
As it is generally agreed that the cows stand for dawn, and as Dawn is said

ZZJ, IV, p. 264. See also Held, The Mahabharata (1935), p. 237, Zimmer, Alf-
indisches Leben (1879), p. 172 ff. Note however also AS. XVIII.2.56 yamdsya sd&na&
sdmitij cdva gachatdt.
B"pastanlba Dharma Sritra 11.25.5 dakjinam pura* sabhd.
See Held, op. cif.,pp. 162,215 and W. B. Kristensen "De Symboliek van de Poort
(etc.)" in Symbool en Werkelijkheid, pp. 215-224 (on Pylos, etc.). For Dv%rakl cf.
Skdn& Purdqa VII.4.24.5 svargdrohananiheni vahate yatra Gomati. Only in passing
can attention be drawn to the name of Sopara in Konkan (north of Bombay), which
is perhaps a parallel case. Skt. S~rpdraka-,Srirpdraka- (Mhbh. Rdm. MdrkPur. VarBS.)
Pali Suppdraka-, Soprlraka- is generally derived from Brirpa-, n. "winnowing basket";
see, e.g., 0.Stein, ZZI, 111 (1925), p. 304 n. 2, Charpentier, JRAS, 1927, pp. 111-115.
However, the Jdtakamrild, p. 88 1.12, has the form sripdraga-: tadadhyu$itarlr ca pat-
ianam supdragam i f y evdkhyrltam asit, yad etarhi siipdragam iti jiicyate. Cf. supdraga-
in 88,15, 93,8 etc. According to the Indian tradition Konkan is the kfatra- won by
' ParaSuriima (Charpentier, p. 113). In view of the geographical position of Sopara it

would be tempting to connect its name with Varupa's epithet suprirdkfatra- VII.87.6
[supdraka- being the region which gives an "easy crossing" to the setting sun) but Pa.
Suppciraka- must then contain Prakritic gemination (also in Sodppara, Periplus 52,
Soupprirn Ptolemy 1.6) and Siirpdraka- must then be explained as a secondary Sans-
kritization, which it is difficult to accept. For Sopara see McCrindle, Ancient India
as described by Ptolemy, p. 40f., B. Ch. Law, Historical Geography of Ancient India
(1954?), p. 299, D. C. Sircar, Studies in the Geography of Ancient and Medieval India
(l!?@),p. 31, etc.
to come "from the harmnydniin the East" (p. 108) it may be concluded that
the cows' shed (vaki-) was mythologically equivalent to the "stone house"
(harmyd-) of the nether world, where the Vedic poets apparently also lo-
cated the "sun in the rock" (svdr ydd diman). On the other hand, the fact
that some Vedic texts contain references to cows in the sabhh- is particu-
larly interesting in the light of the theory that at that time the sabhd- was
still a sacred hall where social contests took place.96 If the presence of
these cows was in some way connected with those contests (which, it is
true, cannot be proved), the question may be raised if these contests also
comprised the re-enactment of the mythical "driving out" of the cows
from the vala-. Apart from all other considerations, these contests, when
considered separately, point to the inference that they were looked upon
as re-enaqtments of the primordial fight between Devas and Asuras for
the creation (or, renewal) of life. In the absence of decisive evidence, how-
ever, this question cannot be answered.
It may be stated in conclusion that it is quite natural that the gods
Mitra and Varuqa, who reside in this subterranean palace where the sun
is hidden in the rock, are themselves also "sun-seers".Q7

5. During the night, however, this picture alters fundamentally. Varu-

na's nether world then extends over the earth as the night-sky. The cos-
mic waters are at that time a celestial ocean (~arnudrri-).~~
In a brahmodya
that takes place between the adhvaryu and the hotr during the ASvamedha
ceremony, a cosmic riddle refers to it in the question "What is the pond
that is equal to an ocean?', to which the answer is "The sky is the pond
that is equal to the ocean".sg
When we consider the fact that the moon, which travels at night and
illuminates Varuna's works (1.24.10) is said to run amidst the waters in the
See the references in IZJ, IV, pp. 246 f., 266,277 f. For cows in the sabhd see Rau,
Staat und GeseNschaft irn alten Zndien (1957), p. 76. The interpretationof the Yajurvedic
prayer given in ZZJ, IV, p. 267, is incorrect.
@ V . 6 3 . 2 bvidathe svardijci.
Max Muller, SBE, 32, p. 58. Cf., e.g., 111.22.3 agne divd achi jigLisi, V111.26.17,
VS. 13.31, Renou-Silburn, JAS., 1949, p. 13, and Liiders, Varuna, p. 321 : " ... ergibt
sich, dass man sich die Sonne in der Nacht in einem Felsen, zugleich aber auch im
Meere im hochsten Himmel ruhend dachte, wobei es unklar bleibt, ob der Felsen in
dew Meere liegt oder etwa das Meer umgibt. Aus diesem Meere erhebt sich die Sonne
."
am Morgen .. The same idea of the identity of the night-sky and the primeval waters
occurs also in the Near East, see W. B. Kristensen, Leven uit den Dood, 1st ed. (1926),
p. 119, Symbool en Werkelijkheid (1954), pp. 180, 271.
** VS. 23.47 (etc.) kih samudr&ama& sdrah. 48 dyauh samudrdsamarii srirah. SW
further ~iiders,'~ a & a , p. 11l f. and the critical remarks by K. Hoffmann, O ~ Z49 ,
(1954), col. 393, and JB. I. 165 1.7 ayam vdva samudro 'ndrambhaeoyad idam antarik~am.
sky like Suparna (the celestial bird),lOOand that the waters "amidst which
king Varuna moves forward looking down upon men's truth and false-
hood"lo1 must have been conceived as a celestial ocean, the inference that
"the pond that is equal to an ocean" must be particularly the night-sky
would seem cogent. Accordingly, the interpretation of Varuna's "spies"
(spdJah) as stars in the night-sky (e.g., L o m e l , Oriens, 6, p. 330) is likely
to be correct. Indeed, Varuna "clasps the nights in his arms" (VIII. 41.3).
It should also be noted in this connexion that the briihmanas associate the
Asuras with the night, and the Devas with the day (SB. XI. 1.6.1, MS. IV.
6.7, KS. VII.6). Cf. MS. IV. 7.8 (Varuna = ocean).
One general conclusion that may be drawn from these considerations
is of sufficient importance to be stated here. The question whether Varuna
was "primarily" a god of the waters, a chthonic, or a celestial god, has
been the subject of fierce disputes. It should be stressed, therefore, that
in the light of the preceding structural interpretation such disputes are
senseless. I need not discuss here, e.g., the studies of Liiders, who held
Varuna to be primarily the god of the waters, or those of J. J. Meyer, who
stressed the chthonic aspects. Varuna is both the god of the primeval
waters under the earth (and in the western ocean), of the "stone house"
in the nether world, and of the night-sky. There is nothing in the Vedic
material to suggest an evolution from "primary" notions about the god
to "secondary" developments.lo2
In the archaic cosmology of the Indo-Iranian people the observation
that the sun after entering at evening the nether world in the west again
rises the next morning in the east could only be explained by the theory
that during the night the sun returns through the nether world from west
to east. This idea, well-known from other mythologies, is attested in In-
dia in a variant form, which substitutes for the nether world its equivalent,
viz. the night-sky. The nightly journey of the sun through the sky, how-
ever, necessitates the rather intricate theory of the "Nachtweg der Sonne"
to explain how the sun could return, unseen, to the eastern horizon.lo3
loo RS. 1.105.1 candrdmd apsv dntcir d supargd dhdvate divi.
lo1 VII.49.3 ydsiim rdjii vdrugo y&i mddhye satydnrtk avap&yaA jandndm. Cf. VIII.
47.1 1 ddityii dva hi khyrltddhi kdliid iva spdiah.
l o 2 BBijagen Koninklijk Instituut voor Taal-, Land- en Volkenkunde, 107 (1951), p. 81

n. 33. Renou, Festgabe Lommel(1960), p. 124, denies any comexion in the Rigveda
between Varupa and the night except for VIII. 41.3. Similarly Thieme, Mitra and Arya-
man (1957), p. 71. See on this problem also ZIJ, I11 (1%9), p. 210 f.
Ioa Weber, Ind. Stud., 9 (1865), p. 278, Bergaigne, Religion VPdique, I. p. 7 , Speyer,

JRAS, 1906, pp. 723-727, Caland, WZKM, 26 (1912), p. 119 f., E. Sieg, "Der Nacht-
weg der Sonne nach der vedischen Anschauung", Gott. Nachr., 1923, p. 1 ff., E. Benda,
Der vedische Ursprung des symbolischen Buddhabildes (1940), p. 3 ff.
 e are not here concerned with the theory itself, but with its origin, which,
as far as I am aware, has never been correctly explained. It will be clear
that the mythological equivalence of nether world and night-sky could
easily lead to such an explanation.
Of far greater importance, however, is the notion of the cosmic moun-
tain which, as was seen above, is closely associated with that of Varuna's
domain. When Varuna is said to support earth and heaven by means of
the cosnlic axis, it may safely be inferred that he must have been conceived
as holding the foundations of the cosmic hill and the roots of the cosmic
tree .while sitting in his "stone house". Since this mountain lays upon the
waters and is their receptacle (see p. 106f.) it must also be in the celestial
ocean of the nocturnal sky.
It is there, indeed, but several details show unequivocally that during the
night the nether world was thought of as hanging over the earth in an in-
vertedposition. The characteristic word to denote this, is nicfna- "down-
ward, hanging down". Thus Varuna turns the mythic cask (kdvandha-)
upside down (nicfnabdra-) to moisten the earth.lol The same god holds
the roots (lit. "tuft", stdpa-) of the cosmic tree, while its branches are
hanging down (nicfna-).lo5 The "inverted tree" is well-known from many
mythologies but it is desirable from a methodological point of view that
comparative studies should be based on preliminary inquiries into the
place of this tree in the context of each separate mythological system.
There are numerous valuable studies on the inverted tree in Indian
mythology.100 The main facts are the following: the identity of the tree
as a Ficus religiosa (as'vatthd-) is first stated expressis verbis in the Katha
Upanijad: "With the root above and branches below is tliis everlasting
ASvattha". This verse is imitated in the Bhagavadgitd.lo7 In the Tait-
tiriya Arapyaka the same words occur with reference to a "tree" which is
lo4 V.85.3 nicfnabdram varunah kdvandhanz pra sasarja rddasi antdrikjam / tina viiva-
sya bhlivanasya rdjri yavam nu vptir vy linatti bhrinza. For the kavandha-, which is ap-
parently identical with diva4 kdSam V.53.6, the zitsa- (1.64.6,11.24.4, VIII.7.16, X.30.9)
see, e.g., Hillebrandt, Asia Major, I, p. 791, Vedische Mythologie, I*, pp. 321 f., 325 f.
and my note in Bijdragen Kon. Inst., 107, p. 82.
Io5 1.24.7 abudhnP rcfjci varuno vdnasyordhvdm sthpaih dadate pritadakjah / nicinri sthur
updri budhrza e $ h .. . .
loO Cf.,e.g., E.Kuhn, Festgruss Bohtlingk (1888), p. 68 ff., L. v. Schroeder, Festgruss
E. Kuhn (1916), p. 59 ff., and especially Coomaraswamy, Qu. J. Myth. Soc., 29 (1938),
p p 111-149, M. B. Emeneau, "The Strangling Figs in Sanskrit Literature", Univ. of
Calif: Pub!. in Class. Philol., vol. 13, no.10 (1949), pp. 3%5-370,F. D. K. Bosch, The
Golden Germ (1960), p. 65 ff. (L. Rocher, Dialoog, IV/2 (1963-1964), p. 91, fails to
notice the difference between the aivattha and the nyagrodha.)
lo7 KU. VI. 1 iirdhvamrilo 'vcikicikha ejo 'Svatthah sancitanab, Gitd 15.1 rirdhvam~ilam
adha~Sdkhamdvattham prcihur avyayam.
not specified,lo8and the same is true of the Rigvedic passage: "In the
unfathomable space king Varuna, he of purified intelligence, upholds the
tree's stdpa-; they [= the branches] stand directed downwards.
rays be fixed in us". In the speculations of the Maitri-Upani~adon the
mystical brahman-tree, the "three-footed brahman" is said to have its
roots above.lo9
Whether or not the advatthd- is meant also in the earliest passages, is of
minor importance. It cannot be questioned, at any rate, that everywhere
the reference is to the inverted cosmic tree. Although the notion of such
a tree must already have been familiar tp the Aryans before they pene-
trated into the Indian subcontinent, here they must at some time have
adopted the native habit of equating the mythic concept of the world tree
with the Ficus religiosa instead of the oak or ash-tree. This habit can be
traced back to the prehistoric civilizations of India in the chalcolithic
age.ll0 Since the riddle-hymn RS. I. 164 calls the cosmic tree an dvatthh-
(st. 22), it is possible that also the tree held by Varuna in I. 24.7 was al-
ready identified with the fig-tree.
Many of the modem attempts at an interpretation fail to stress suffi-
ciently the fact that all the Vedic passages clearly refer to a mystery, to an
esoteric religious symbol: Whoever knows this tree is delivered from
death, says the Taittiriya Aratzyaka. Therefore naturalistic explanations
which equate the inverted tree to the banyan-tree (Ficus indica) with its
aerial r o o t ~ ~ ~ ~seemo u lparticularly
d inadequate.l12 Thieme, who ex-
plains the tree as an image for the night-sky, holds that the notion origi-
nated in the concept that the cosmic tree has no "foundations" and, there-
fore, must be held from above.l13
Curiously enough, the fact that it is Varuna who holds the tree has sel-

lo8 TA.1.11.5 firhvamrilam avcikhikham vykqath yo veda sathprati ...

log MU. VI.4 Crdhvamhlam triprid brahma, Jtikhd rikaa-vdyv-agny-udaka-bhrimy-
cidaya, eko 'fvatthancimai,'tad brahmai, 'tasyai 'tat tejo yad as6 ridityd.
11° Marshall, Mohenjo-Daro and the Indus Civilization, I (1939), p. 63 ff. (quoted by
Bosch, op. d., p. 68).
ll1 E.g., Geldner, Ved. Stud., I (1888), p. 113 ff., Emeneau, h., p. 347, Bosch, op. cit.,
p. 69, n. 12 (with references), Renou, Et. v&d.et p&, VII, p. 72 f. Emeneau rightly
points out that the "upside-down" tree is a cosmological imagining of great antiquity
in the culture centers of Eurasia, and one which came to be identified with local flora
and thus was reinterpreted. He wisely refrains from attaching too much value to the
rale of the dvattha as an epiphyte.
11' L.von Schroeder, I.e., p. 67, rightly observes: "Aus dem Leben ist das kaum ge-
griffen, und ein wirMich befriedigend Wares Bild verbindet sich nicht damit", Cf. also
Deussen, Allgem. Gesch. d. Philos., Ijl, p. 182, Fr. Weller, Versuch einer Kritik der
, 174 f. n. 7.
K ~ h o p . p.
Untersuchungen zur Wortkunde und Auslegung des Rigveda, p. 67 n. 2.
dam (if ever) been given due consideration. From the circumstance that
he supports earth and heaven "in the seat of ta", it was inferred above
that he must consequently support the world a S "from below" (adhustdf,
as the epic says of the serpent Sew). In other words, the inverted tree of
Vedic mythology must be connected with the idea that in the nocturnal
aspect of the cosmos the sky is the nether world (that is, the cosmic waters
and the mountain with the tree) turned upside down. To some extent
enda was right in assuming the identity of earth (kjcirn-) and night-sky,
but his conclusion that the earth was thought of as extending upwards at
night as far as the shining sky (dyli-) and as covering and concealing this,
cannot be correct. The inversion of the position of nether and upper
world is something quite different from an upward expansion of the earth.
In conclusion the fact should be stressed that if the sun returns at night
to the eastern horizon through the night-sky = nether world, the "sun in
the rock", accordingly, is during the night in the sky.

6. This conclusion furnishes the clue to a correct understanding of Old

Iranian cosmology. It can hardly be questioned, indeed, that the Old
Iranian use of the word asman- in the sense of "heaven" owes its origin
to the ancient Aryan concept of the inverted mountain in the night-sky.l14
The Avestan phrase asmanam xUanvantam"sunny sky" (orig. "rock with
sun"),115 although probably re-interpreted, still reflects the ancient idea
of the svhr ycid diman. Its religious significance appears from its use as a
personal name Asrnd.xUanvant-. ost instructive, however, is the fact
that not only has usnvant- "rocky"llE become the name of a certain moun-
tain (which is only natural), but so has xvaiivant- "possessing the sun".
This indirectly indicates that in earlier (prehistoric) times the phrase
asman- xvanvant- still referred to the primordial hill and to the religious
concept of the svhr ycid &man. The epithet of the primordial hill was only
secondarily transferred to a real mountain, which accordingly must have
regarded as a replica of the mythical mountain.
hen Zarathustra says that the holy Spirit "is clothed in the hardest
heavens",l17 this remarkable epithet still preserves a reminiscence of the

114 Against Reichelt's theory of "Der steinerne Himmel" in IF, 32 (1913), pp. 25-57
and Hertel's elaboration in Die Himmelstore im Veda und Awesta (= Indoir. Quellen U.
Forsch., 11) (1924), see Hillebrandt, Asia Major, I, pp. 785-792, Benveniste-Renou,
Vrtra et Vr@ragna,p. 191 n. 3, Charpentier, Brahman, p. 12 n. 2.
Visp. 7.4, Vend. 19.35, Sir. 2.27.
llB In amvantarn gairim, see Altiran. F'iirterb., col. 220.
l'' Y. 30.5b y6 xraoMift5ng assnd vastd; cf. YAv. xriifdisrna- "hard earth". According
to Nyberg, Irans forntida Religioner, p. 493, this is "a much more primitive idea" than
"rock" in the night-sky : for it is, indeed, the noctu
wears as a mantle studded with stars",11s and this
turnal aspect of the "fixed abodes" (dhruvbh kait
man-) in which Varuna dwells. The special use
Avestan to denote "heaven" (although it contin
asayga-, in the sense of "stone") is also found in
a vazrka Auramazdd haya imdm biimim
add 'X great god is AhuramazdSi, who created this earth, who create
yonder sky". This lexical fact can hardly be dissoci
which point to the conclusion that the greater part o
a few exceptions such as the Sakas119and perhaps t
by Herodotus) laid special stress on that aspect o f t
that was more particularly concerned with the neth
ingly, professed some form of azdeism. As we know from the Veda,
this aspect comprised initiation and apparently was (like many religions
concerned with the gods of the nether world) a mystery religion. Iran
gave the cult of the mystery god(s) a marked preference to that of the
heavenly gods. The specific Iranian religion of Ahura
understood as the result of a development which laid such a heavy stress
on the personal relation to the mystery god that this completely ousted the
worship of the other group of gods.
Some serious misinterpretations of Aryan mythology are the Inevitable
result of the failure to recognize the mythological equival
world and the night-sky. It must be regretted that even
construction of Vedic cosmology has failed to draw this ultim
sion from the materials which he has so judiciously studied. K.
has rightly pointed out, how weak the foundations were upon which his
theory of the celestial ocean was based.120
At first sight, it is true, the statements of the Veda may seem fully con-
tradictory. Thus, when Usas, who is "born in eta" and is a kinswoman
of Varuna,lZ1is said to awake from "the seat of taw(which is also calle
"the stone house", "the rock of Heaven", "the cave", "

Rcichelt's "stone sky". It has undoubtedly been inherited from the myth010
proto-Indo-Iranian people but is not particular!^ "primitive".
Yt.13.3 yim Mazdd vaste vaghmm stahrpaZsagham mainyri.tcZStam. Cf. also
Bartholomae, Arische For~ch~IZgen, 11, p. 123, Wikander, Vayu, p. 30
Abaev, "Skifskij byt i refonna Zoroastra", Archiv Orienfdlni,
Benveniste, sur la langue ossdte, p. 129.
lao OLZ, 49 (1954), col. 391.
lel 1.118.12ytej& 1.123.5 vdruqasya jdmir ("sister", Renou, Etudes sur le vocabulair~
du R V., p. 49). With yfejd- cf. W. B. Kristensen, Leven uit den Dood, p. 131.
mountain" or simply "the mountain")122it would seem natural, in the
light of the preceding remarks about the "stone house", to take all these
statements as referring to the nether world, in spite of the word divb "of
Heaven", which has once been added. In another passage, however, she
is said to come "even from the light of Heaven".lZ3 This light, which (if
the preceding interpretation is correct) must be situated in the darkness
of the nether world, leads us back to our starting-point, the "bliss of ASa
which manifests itself together with the lights".
7. The preceding observations may have served to show that the image
which Zarathustra formed of Ahura Mazda's celestial world, though per-
haps re-interpreted in details, was still essentially based upon the ancient
Aryan concept of the inverted mountain in the night-sky. Just as Mitra
and Varuna, as lords of the cosmic mystery, knew the secret of "the sun in
the rock" and were, therefore, "sun-seers", so, with a transfer of the epi-
thet from the lord to his domain, Ahura Mazd5's xJuOra- is called "sun-
seeing" in the Gathas.lZP
Rta is hidden (dpihitum V. 62.1) in the nether world, as the sun is in the
"rock" (VII. 88.2). If, however, this nether world and this rock were e-
quated to (or, localized in) the cosmic mountain, which in its turn is myth-
ologically equivalent to the whole earth, it must be taken into considera-
tion that "the earth lies spread on the ~aters".~~"o there must be some
connexion between Rta, the Sun, and the cosmic waters under the earth.
That "Agni hidden in the waters7' was one of the central religious con-
cepts of the Rigveda need hardly be mentioned, nor need the importance
of the myth of the mysterious birth of the Golden Germ (Hirapyugarbhd-)
be stressed. This connexion explains why the waters are said to be "know-
le2 ..
IV.51 .S ytcisya . sddaso, VII.76.2 harmydni (cf. Geldner, Kommentar, p. 114 and
see IIJ, IV, p. 226, where the accent of harmyd- should be corrected), V.45.1 divd
.
.. ddrim, 2 ~lrvdd,3 pdrvatasya gdrbho, pdrvato, see Liiders, Varuoa, p. 325 R. The
horses of the Sun are said to come, like U$as, from the seat of Rta: 1.164.47 td dvavytran
scidancid yrdsya.
"a 1.49.1 divdS cid rocandd ddhi.
Y.43.16d xV5ng.darascii xSaordi, which hardly means "den Anblick der Sonne
gewahrend, so~engleichanzuschauen" (Bartholomae, Altiran. Worterb., col. 1880,
Geiger, Die AmaSa Spantas, p. 210). According to Humbach, Die Gathas, 11, p. 36
(ad 32.13a) xSa8ra- sometimes denotes the ruler himself. His rendering in vol. I, p. 15
"im sonnengleichen Machtbereich" differs from that in MiiSS, I, p. 28 n. 1 (xV5ng
darasdi "damit die Sonne geschaut werde"). See also vol. 11, p. 52. It must be admitted
that also in the compound the genitive xV5ng- is hard to explain. Cf. hvara.darasa-,
hvara.darasd and hvara-darasya-.
SB. VII.4.1.8 cipo vaipu.ykaram rcisdm iyam parpam, yathd ha vd idari,pu.ykarapar-
@amapsv adhydhitam, evarn iyam apm adhydhitd. Cf. AB. III.6.4pythivy apsu Cpratighi-
td (note 52), and Liiders, Varuna,p. 121 ff.
ing Bta" and "possessing the sun", an

Somewhat parallel statements are found in Zarathustra's S

the Rigveda says that the sun is the face or appearance (
Rta,12' Zarathustra calls ASa "possessing the sun" (xV3nvaj' 32.2b)IZ8and
in later texts ASa's dwellings are described as xvanvaiti3.lZ9On the other
hand, it was stated above that the idea of darkness was closely associated
with the nether world (see p. 108). So the rocand- orjydtip and the raocah-
which according to the texts were present in this world require some eluci-
dation.
RS. IX. 113, a hymn to Soma, ends with a vision of celestial beatitu
which the poet hopes to attain to. The last stanzas 7-1 1 run as follows :ls0

7. Where there js the eternal light, in which world the sun has been
placed, in that immortal, imperishable world put me, O pavamEna!
8. Where dwells king Yama, where there is the descent from heaven
(avarddhanak divdh), where are the flowing waters, there make me
immortal.
9. Where one can freely wander about in the threefold firmament, in the
laa AS. IV.2.6 arnitd rtaj8& (cf. N . J. Shende, BDCRI, IX [1949],p. 271 f.), RS.
1.10.8, V.2.11, VIII.40.10f. svdrvatir up&; 1.136.2pdnthdrtdsya sdm ayali2sta raimibhib
(cf. Yt. 13.57 yd strqm mdghb hiirb anayranqm raocaghqm pat% dac'sayan daonif?), JB.
111.359 rtam jyoti~mad(ex conjectura).
VI.5 1 .l lid u tydc cdkpir mdhi mitrayor d& k f ipriydriz vdruwyor
Slici dariatdm dnikam rukmd nd divd liditd vy ddyaut. See Griswol
Rigveda (1923),p. 135, Geiger, Die A m & Spantas (1916),p. 176.
Not "the glorious light", Moulton, Early Zoroastrianism, p. 355.
lZe xuanvaitif aSahe varazb Y.16.7 (etc.). See Geiger, Die A m f a Spantas, pp. 176,201.
Humbach, IF, 63, p. 48 compares RS. vrjdne mdrvati. For ASa and fire see also
Duchesne-Guillemin, East and West, 13 (1962), p. 201. That Ahura Mazdii's name
has later come to denote the sun (Khot. urmaysde, Yidya, Rlunji ormiszd, see Bailey,
Zranistik I, p. 134) is hardly ~ 0 ~ e c t with
e d these ancient Aryan conceptions.
laO 7 ydtra jydtir djasram yrismifi lokk mdr hitrim
trismin mdm dhehi pavamdmi 'm$e lokci dkfita
indrdyendo pdri srava
8 ydtra rdjd vaivasvatd ydtrdvarddhana& divab
ydtrdmhr yahvdtir dpas tdtra mdm am~tumkrdhi, etc.
9 ydtrdnukdmcim cdrapam trinrikk tridivk div&
lokd ydtra jydti~mantastdtra mdm amitam kydhi, etc.
10 ydtra k h d n i k d d ca ydtra bradhmisya viSfcfpam
s v a m ca ydtra t t p t i ~ca tdtra d m amitarh k@i, etc.
11 ydtr- ca mddd ca mlidab prarmida hate
kdmasya ydtrxipt& kdmiis tdtra d m amitam kydhi, etc.
The refrain has been omitted in the translation. For Vedic ideas about the wor
the dead, cf., e.g., Zirnmer, Altindisches Leben (1879),p. 410 ff., N. J. Shende, BD
IX (1949), pp. 251, 254, etc. (data from the Atharvaveda). Otherwise BAU. IV.3.10.
 threefold heaven, where are the luminous worlds, there make me im-
mortal.
10. Where wishes and desires are (realized), where there is the highest
point of the sun, where are the food offered to the deceased ancestors
and satisfaction, there make me immortal.
I I. Where joys, pleasures, various forms of gladness and delight are seat-
ed, where the highest wishes are obtained, there make me immortal.
Irrespective of whether, in a visionary state of mind, the poet here as-
pires to see the bliss of the blessed dead or rather prays for a place in the
"immortal world" in afterlife131,this much is clear that this is the tradi-
tional picture of the blissful life in Yama's realm. With vaivasvatd~z,the
patronymic which denotes Yama in st. 8, cf. X. 14.1 "Vivasvant's son, who
brings together the men, king Y a n ~ a " . ~ ~ ~
As stated above (p. log), Yama's world is at the same time Varuna's.
This world, which Yama gives as a resting-place to the deceased and
which is "adorned with days, waters, and rays"133 is Yama's "stone
That the water (salild-) in the world of the Fathers which is
mentioned in AS. XVIII. 3.8 need not refer to a celestial ocean was rightly
stressed by K. Hoffmann, OLZ, 1954, col. 392.
Here is the avarddha~alizdivdh (st. 8), which words, though generally
rendered by "closed place of Heaven",13%re more likely to mean "the
descent from heaven" and to refer to the place "where they unharness the
horses of the Sun" (V. 62.1). Just as Ait. Br. IV. 14.5 opposes samvat-
sarasyd 'varodhanam "descent of the year" to udrodhananz "ascent", and
just as K a d . S. 98 uses avarodha- "downward movement" as the opposite
of rodha-, the Rigvedic expression avarddhanam d i v 4 must be taken
as contrasting with drddhanalit divdh "ascent to heaven" (RS. I. 105 .l 1,
IV. 7.8; 8.2, 4). The dh has been longer preserved in the nominal deriva-
tive than in the verb: the Ait. Br. uses praty-avarohati IV. 21.3 beside
avarodhanam IV. 14.5 (but has also dtlrohanam). Cf. Ai. Gramm., I,
pp. 250,252.
lX1 X.17.4 ydtrdsate suktto ydtm t i yayris tdfra tvd devdb savitd dadhdfu, AS. VI.120.3.
VS. XV.50, etc.
laa vaivasvatdrit samgdmana~n jandncim yarnrim rdjdnam.
la3 X.14.9 dhobhir adbhir aktribhir vycikfarit yamd dadcity ava&unr asmai. Cf. the

yahvdtir dpab in IX.ll3.8.

lab X.114.10 yadd yamd bhdvati harrnyi hit&, AS, XVIII.4.55 ydthci y a m ~ y a harmydm
avapan p&ca indnavbb (see Shende, BDCRT, IX, p. 249).
lS6 E.g., PW,I, col. 486, Monier-Williams ("a closed or private place, the innermost

part of anything"), and the discussion in Neisser, Zunz Worterbuchdes &gveda, I (1924),
p. 129. Hertel's interpretation in Die Himmelstore im Veda und Awesta (1924), p. 47,
is unacceptable.
 If this interpretation is correct, the words "where the sun has been
placed" in st. 7 must necessarily refer to the "sun in the rock". Indeed,
the deceased man after reaching Yama's world gets back his eyes and
sees the sun (AS. XVIII. 2.46). The "threefold firmament" mentioned in
st. 91a6may again refer to the night-sky but the words "the highest point
of the sun" (bradhndsya viytdpam) in st. 10 present some difficulties. Only
the notion of the nightly journey of the sun through the sky from the west-
ern to the eastern horizon can account for the idea that the sun reaches a
summit (viytdpa-) at night too, just as Varuna is said to be in the highest
heaven (X. 14. 7-8). Although it is nowhere stated in the material
collected by Sieg (see n. 103), the sun when turning in the evening towards
the east (cf., e.g., I. 115.5 yadid dyukta haritah sadhdsthdd, Sieg, p. 5
"umschirren"), must again ascend the vault of heaven. Incidentally, an
indication that the journey back led the sun through the world of the
dead is perhaps contained in the words of I. 164.38ab dp&i prdri eti
svadhdyd grbhitd 'martyo mdrtyend sdyonih, which Sieg (p. 8) renders as
follows: "Abgewandt geht er nach Osten, geleitet von seiner Gotthaftig-
keit, unsterblich teilt er mit den Sterblichen den Wohnsitz".
Thus we are entitled to state that according to the Rigveda Yama's
and Varuna's world contains the eternal light and is luminous.

8. It is not necessary for our purpose to give a more detailed account of

the occurrence of light in the Vedic descriptions of the blissful life of the'
dead. What can be done is to reconstruct, on the basis of the texts, the
cosmological ideas about light in the darkness of the nether world, which
obviously was considered a mystery, no less than was the presence of
potential life in death. The parallelism of light and life in ancient Aryan
thought is indeed too well-known to need special discussion. What can-
not be reconstructed, however, is the correlate of this cosmological notion
in the religious experience of the individuals. Of the ecstasy which accom-
panied this experience a precious document has been preserved in RS. VI.
9.6. One gets the impression that the seer, by meditating upon the sacri-
ficial fire (st. 5), attained a vision of the cosmic mystery, the "sun in the
rock" (or, Agni in the nether world, into which the sun enters in the even-
ing and from which it arises in the morning, AB. VIII. 28.9 and 13). Some
other passages, too, may perhaps be explained as referring to the seer's
entering the nether world,la7but all details inevitably escape us. How-
'*$ AS. XVIII.4.3 trtiye ndke cidhi, etc., see Zirnmer, Altindisches lkben, p. 410.
Liiders, Varuqa, p. 61, etc.
Cf. perhaps 1.22.14 viprri ... gandharvcisya dhruve' padd.
ever, such general aspects as are still sufficiently clear entitle us to state
that Agni's birth, or the vision of the sun in the darkness, was the central
theme of what can be denoted by no other term than Aryan t n y ~ t i c i s r n . ~ ~ ~
This conclusion may prove of material interest for the study of later
forms of Indian mysticism but a discussion of this problem lies outside
the scope of this article. Only in passing, therefore, attention may be
drawn to the fact that the words "in which world the sun is placed"
(yrismih lokb svdr hitdm IX. 113.7), when compared with "the sun in the
rock" (svdr ydd diman VII. 88.2), suggests the idea that the later specu-
lation of the Upanisads about the dtrnarl "placed in the cavity" is a direct
continuation of the older mystical speculation of the Veda.
In the Kattza Upani~adI. 14-19 Yanla instructs Naciketas in the cult of
the fire: "I will declare to thee - mark me well - the fire that leads to
heaven, which I know well. Know that this (fire), which is a means of
acquiring the infinite world and which is the foundation (of this world)
is placed in the cavity".13pAfter Friedrich Weller's detailed discussion of
the last word it can no longer be questioned that guhdydm must mean%
der Hohle des Herzens", and that the following verses, which deal with the
arranging of the Ngciketa-fire, must be a later interp01ation.l~~ The sec.
ond chapter, indeed, which is devoted to the dtman doctrine, has the fob
lowing verses: (12) "The wise who by means of meditation on his Sell
recognizes the ancient whom it is difficult to behold, who has entered into
the hidden, who is placed in the cavity, who dwells in the abyss and who
exists from times of old, as God, he leaves joy and sorrow behind", and
(20) "The Self, which is subtler than what is subtle, greater than what is
great, is placed in the cavity of that creature. He, who is free from desire
and from grief, beholds the majesty of the Self by the grace of the Crea-
tor".141 The last verse also occurs with some variant readings in Taitt.h.
X. 10.1, Mahdndr. Up. VIII.3 and Svet. up.111.20.
That the "cavity of this living being" is identical with the heart cannot

l" It should be noted that in the Atharva Veda Varuqa is the great teacher (cicdryci-)
See Renou, Festgabe Lommel, p. 127 and cf. RS. VII.87.4 uvdca me vdrupo mkdhirciya
trib saptd nhdghnyd bibharti/vidvdrr paddsya gzihyci nd vocad, VIII.41.5 yd usrdndm
apicyd vkda ndmdni ghhyci, etc.
la9 1.14 arrantalokciptim at110pratijfhdm viddhi tvam enam nihitariz guhdydm.
"O
Friedrich Weller, Versuch einer Kritik der Kafhopani~ad(Berlin 1953), p. 6 ff.
11.12 tam durdariadz girdham anupravi~famguhcihitaril gafzvare$ham purtipam /
adhyrltntayogddhigamena devam nzatvd'dhiro har~aiokaujahdti; 20 aaor apiycin mahato
mahiycin dtnrdsya jantor nihito gufrdydriz tam.akratu[zpaiyati vitaioko dhdtn(r)prasddcirr
mahimcinam cimzanah. For girdham anupraviflam cf. IU.1 guhciriz pravi~fau, IV.6,7
guhdm pmviiya, and NySsa Upan. 2 guhciri~prave~~utn icchrlnri (quoted by Jacobs, Con-
cordance, s . ~guhci).
.
 85
indeed be doubted in view of, e.g., Chdnd. Up. VII. 3.3 "this dtman is in
the heart".142 Regnaud and Hillebrandt have already pointed to the paral-
lelism between the Vedic mythological concept of Agni hidden in a secret
place and agni/dtman which here is nihito guh~iydm,l~~ but they did not
recognize that what we find in the Katha Up. is not merely "ein der alteren
Mythologie gelaufiger, aber hier philosophisch umgewerteter Begriff"
(as Hillebrandt put it) but a direct continuation of an old Vedic tradition
of meditative practices.
It would require a special study to demonstrate the parallelism which
for the Vedic poets exists between the macrocosmic opening of the pri-
mordial hill and the microcosmic opening of the mind, as the result of
Indra's vytrahcitya-. I must confine myself to the statement that the Rig-
vedic seer gets his vision with or in his heart (hydd or hydi).144 This heart
is equated to the cosmic mountain and its subterranean ocean. Hence it
is that the Rigveda refers to the "ocean of the heart" (TV. 58.5 hidydt
, samudrdt, cf. 11 ant* samudrk hrdy dntdr) and divine inspiration is looked
upon as an "opening of the doors of the mind" that is parallel to the open-
ing of the cosmic "enclosure" (~rajCi-).l*~ Just as the Rigvedic poet prays
that Agni may open his "thought" like an aperture (kh6-,viz. in the cosmic
hill of the primeval world),14@ so the Mahdndrdyava Upani~adsays that
PrajGpati, who is "hidden in the cavity" of the heart, pierces from within
the kh6ni (of the mind) to "enjoy the objects".147 If it may be assumed
that guhdydm in the Upanisads is a substitute of the later language for
such Rigvedic terms as vrajk, firvk, diman, which all denote the nether
world viewed as an enclosure, the parallelism between the revelation of
"the sun in the rock" and the Upanisadic vision of the dtrnan "placed in the
14¶ sa vci eja citmd hydi, VI. 17 hydaye sarivrivijfab. Cf. Mahcincir. Upan. 11.6 sa . ..
abhyantaram prciviiat sa vci e,w paiicadhci 'tmcinam vibhajya nihito guhciydm sa v6
eJo 'smacldliydantarddakytcirtho 'manyata, TB. I, 2.1.3 guhci iarirasya madhye.
l" Paul Regnaud, La Katha Upanisad (Paris-Lyon, 1898), p. 117, Hillebrandt, Aus
Brdhmagas und Upani&n, n. 142 (both quoted by Weller, op. cit., p. 6, n. 6 and n. 8).
lU Cf., e.g., RS. 1.24.12 tddaydm kdto hydd d vi cajfe, VII.33.9 td in ninydm h$ayasya

praketaib saluisravaliam abhi scim caranti ("sie dringen nach den Ahnungen ihres
Herzens in das tausendfach vemeigte Geheimnis ein" Geldner), X.129.4 sat6 biindhum
dsati nir avindan hrdi pratfjya kavdyo manijcf, 177. 1 patamgrim aktdm dsurasya
mciydyd hydd pdyanti mdnasci vipdcitab / samudrd ant& kavciyo vi cakjate,
etc., and Mahrlndr. Up. 1.12 na cakpgci pdyati kaS canainarh hydd manQci
manasd 'bhik/ptab. Cf. also Geldner's note ad 1.171.2 on hydd ...rncinasci, Renou,
Etudes sur le vocabuiaire du RV., I, p. 60 f. and AV.zaradticci manar~hdcciY.31.12.
See in general ZIJ, IV, p. 249.
IV.11.2 vi $cihy agne g m t b manisdm khdm vbpasci tuvijdta stdviinab.
14' 11.6 sa vci ejo 'smaddhydantarcidakytcirtho 'manyatci 'rthcin dciniti. atab khdninuini
6hittvoditaFpaiicabhiraSmibhir vi,wycin attiti. Similarly Kafha Up. IV.1. For the macro-
cosmic parallel cf. e.g., RS. VII.82.3 dnv apdm kMny atyntam djaai.
cavity" is apparent. The circumstance that not only the fire (Kathn Up. I.
14) but also Prajgpati (Maitr. Up. IT. 6), Purusa (Mund. Up. 11. 1.10) and
the udgitha (Maitr. Up. V I . 4) are said to be hidden in the cavity148points
to the conclusion that the notions associated with the "cavity" were
rooted in mythology. As was stated above, it is, however, not possible to
go more deeply into the problems of later mystical speculations.
This Old Aryan mysticism is also directly reflected in Zarathustra's
phraseology. When he speaks of managlz6 vagh;ius' sv&zvaj haghuS 53.4c,
which Lommel renders as "'lichtvollen Gewinn des guten Denkens' als
himmlischer Lohn"l49he mere mention of a "reward that contains the
sun" shows that there are still traces of Old Aryan mysticism in Zara-
thustra's preaching. Schaeder's characterization of this religion as "ar-
chaische Mystik"lSOwould seem to be fully justified.
In Zarathustra's hymns references were found both to "ASa with the
sun" (32.2b) and to "the bliss of ASa, which manifests itself together with
the lights" (30.1c), whilst analogous phrases occur in later texts. Thus
Ahura Mazdl promises to lead the soul of the pious thrice across the
bridge "to the best existence, to the best ASa, to the best lights".151 The
pious man, in his turn, worships Ahura Mazd3 "Whose is the cow, Whose
is ASa, Whose are the lights, with Whose light the paradise is filled".153
The yazata RaSnu is invoked in the words: "Whether thou, 0 holy RaSnu!
art in the bright, all-happy, blissful abode of the holy Ones, we invoke,
we bless RaSnu, the strong".153
Especially significant is the close connexion that exists between the
blissful life and ASa. First of all, this concerns the beatitude of the dead :
according to Xerxes the pious man "becomes blessed (rtdvd) when

lP8 A mythical background of the concept of grihd is also suggested by the fact that
some texts say that the "cavity" is in the highest heaven, cf. ~aitt.,dr.VIII.2, Taitt. Up.
11.1.1 (on the brahman) yd vPda nihitam gkhdydni param6 vybman, sb 'Snute sarvdn
kdmcin, saha brdhmapd vipaSciteti (see Sdyapa ad TA.), Mahdndr. Up. X.5 parepa
ncikam nihitant guhiiyd* vibhrdjate yad yatayo viianti, MupQ. Up. 111.1.17 drirdt sudrire
tad ihcintike capaSyatsv ihaiva nihitam guhdydm, KafhaUp. 111.l guhdrizpravi~tauparatne
parrirdhe. See n. 142 and cf. ChU. 111.13.7 for the macrocosmic parallelism.
Id$ Worfer und Sachen, 19 (1938), p. 243; cf. also Die Religion Zarathustras (1930),
p. 203.
150 See Lentz, ZDIMC, 103 (1953), p. 431 f. with references. K. Rudolph's criticisms
("etwas ungliicklich als archaische Mystik bezeichnet") and the contrast with "ein
klarer Kopf" which he supposes to be inherent in mysticism (Numen, S [1961],p. 107 f.)
can only be explained from a not uncommon but nonetheless erroneous notion of
mysticism.
lK1 Y.19.6 d vahiSfdj ayhaof, ci vahiitdj a S f , d vahiitaCiby6 raoca'by8.
lb2 Y.12.1 ye$hCgduS, ye$hC&m, yetjh?raocd, yerjhCraodbiirdi6wan xvd6rd [= 31.7al.
Yt.12.36 upa vahiSfam ahrim afaonqrn raocaghan~visp8.x~ci6ram[= Y.68.131.
dead",15a and in the Vasna Haptaghditi the faithful pray to Ahura
to grant a reward for this life and for after-life "so that we may obtain
that, viz. the union with Thee and with ASa to all eternity".155 Perhaps
the adjective a$c.ayhan- Yt. 13.151, used with reference to "previous
teachers", was meant in this sense, if it may be taken as "der sich das ASa
erworben hat" (Bartholomae).
In the Rigveda, however, the knowledge of the cosmic order is not only
the prerogative of the gods and the blessed dead (e.g., RS. X. 15.1), but
also of the initiated seers (kavi-, X. 64.161, to whom such epithets as
rtajn"&-"knowing the cosmic order", rtdvan- etc., are almost exclusively
attributed.158 One of these seers was Vasisfha, whom "the skilful [Varuna]
made an fii by his power"157 and in the hymn of his vision (VII. 88) it is
suggested that the contemplation during life-time of "the sun in the rock"
was part of the initiation (see n. 79). In exactly the same way Zara-
thustra speaks of "the most blissful union with ASa" as something to be
attained during life-time,15sthat is, he prays to become an aiavan- in this
life already.
It would also be possible to demonstrate the existence of an old tradi-
tion of mystical contemplation from the technical terms (such as cit- and
its derivatives) which Indians and Iranians have in common. It is hoped,
however, that the preceding remarks are sufficient for proving that, when
Zarathustra professes that he will speak of "the bliss of ASa which mani-
fests itself together with the lights" he is using the traditional terminology of
Aryanmysticism.However, thereis this characteristicdifferencebetween the
Rigveda and the Avesta that the former lays more stress on the wonder
of the sun in the rock, while the latter emphasizes the contemplation of
ASa. It is significant that in the Veda there does not occur a word *rtadti-
"seeing the Rta" by the side of svardti- "sun-seer", but that only rtacit-,
rtajEci- "knowing the Rta" are found. The Vedic poet tells us that there
is a golden swing (the sun) in Varuna's palace159but he was apparently

XPh 55 utd mrta rttivd bavatiy.

Y.41.6 yd tag upd.jamydmi tavacd saram dah.ydcd vispdi yavZ.
lK6 E.g.,IZJ,IV,p. 185f.
16' VII,88.4 $m cakdra svdpd mrihobhih. Cf. Luders, Varu~la,p. 321.
See above, note 42. B. Geiger's conelusion that 49.8 refers to "die Vereinigung
mit dem gottlichen ASa im Paradiese" (Die AmaSa Spantas, p. 197 n.) is, as far as I can
see, not supported by the context, nor by 41.6, where the aSavans pray for the union
with ASa "for this life and for the spiritual existence" (cf. 35.8 and see Baunack, Studien
auf dem Gebiete des Griechischen undder arischen Sprachen, 1/2,1888, p. 399 f.).
AS. VII.83.1 aps&te rdjan varuga grhd hiraaydyo mitdh (like the gdrta-, RS. V.62.7
and S), RS. VII.87.5 gttso rdjd vdru& cakra etam diviprerikhdm hiranydyam Subhi kdm,
88.3 prdprerikhd irikhaytivahai Subhi kdm, where the sun must be meant (Luders, Varuga
unable to visualize such an abstract concept as the Cosmic Order. Only
the sun as its "pure m a n i f e s t a t i ~ n "could
~ ~ ~ serve this purpose. In marked
contrast with the Vedic poets Zarathustra prays "Then indeed, show me
ASa, which, verily, I invoke".161 This may be compared with Y. 60.12
"0best ASa, 0 most beautiful ASa, we will see thee, may we reach thee,
we will be entirely thy companions"162and perhaps also with 28.5a, if this
may be translated as "0 ASa, when shall I see thee?'.ls3 The words "in-
timate of ASa which is accompanied by the sun"lGPdo not mean, therefore,
that the sun was here a manifestation of ASa in the same way as in the
Veda.
Despite this different emphasis, Zarathustra no doubt considered him-
self an "initiate" (vidvah-, vidu-, a k z ~ a n - )like , ~ ~Vasi@ha,
~ and the Gathas
show that he maintains with his God a spiritual intercourse of the same
kind as VasighaleBwith Varuna. This parallelism has not escaped Zim-
mer, who characterized Vasistha as "a potential Zarathustra of his
race" 167
However, the way in which the Vedic seer reacted to his personal ex-
perience was different from Zarathustra's. If a general statement may be
hazarded on the basis of a single Vedic hymn, it would seem that the
Vasistha of RS. VII. 88 was not concerned with anything else beyond the
contemplative life and, in so doing, he conformed to the general cultural
pattern of Vedic India. Zarathustra, on the other hand, became a prophet
who tried to convert others to the life of "initiates". The prophetical
trait which he introduced into what must essentially have been a mystery

p. 321). The use of dridye in VII.88.2 vapur driaye is significant in connexion with
svar-d$- (see note 79).
lso VI.51.1 rtdsya i k i dariatam dnikam.
Y.43.10a at tii m6i ddiJ d a m hyai md zaozaomi; cf. Vyt. 42 at'Sqm rat& zaraOuStra
dahe da&ay@ ( K , daisayaf)pantqm. For lexical traces of "sun-seeing" in Iran see the
notes 79 and 124.
d a vahiSta, aSa sraC.fta,darasiima Owd, pairi Owd jamydma, hamam Owa haxma. Cf.
Baunack, Studien, p. 400, Bartholomae, Altiran. Worterb., col. 1739, Geiger, Die
AmSa Spantas, p. 187 n., Bailey, Zoroastrian Problems in the ninth-century Books, p. 1 1
n. 4.
lB8 Lentz, Yasna 28, p. 23 f., takes a& as an instrumental.
Ie4 Y.32.2b d d huS.haud xv&mitd.
lE6 AS for Y.43, see however Humbach, Die Gathas, 11, p. 48.
lee Cf. also the dialogue between the Athawan of AS. V.ll and Varu~a, discussed by
N. J. Shende, BDCRI, IX (1949), p. 284 f., and Renou, Festgabe fiir Herman Lommel,
p. 126 f.
Zimmer, Altindisches Leben (1879), p. 412: "weniger geistiger Grossen, unter
denen z.B. ein Vasi~fiadas Zeug hatte ein Zarathustra seines Volkes zu werden".
reli@on,188of the same kind as the Varuna cult in India, manifests itse
in his efforts to found a "Civitas Dei" of aiavans on earth. Still, the
traces of mystic conten~plationthat must have been characteristic of the
zdB religion, remain noticeable in Zarathustra's preach-

here are, indeed, other aspects in Zarathustra's religion than the some-
what rationalistic philosophy and the ethical appeal to the will of each
individual which have been over-emphasized in later theology. It is hoped
that the preceding pages, devoted to the elucidation of the Old Aryan
background of the phraseology of a single line from his Songs, have
that there is some reason for a revaluation of the original character
hustra's preaching.
This does not necessarily mean that Zarathustra's place in the evolution
of Iranian religion was of minor interest. Although Ahura Mazd5 must
have been worshipped in Iran long before Zarathustra made his appear-
ance, the latter's strong personality has left a lasting mark on the older
religion. It was, however, not the object of this study to stress what was
novel and original in the prophet's reform but rather to show, in the light
of a single passage, to what extent the mystical background of the older
religion had left its traces in Zarathustra's mind.
The ancient Aryan belief that RtB resides in the realm of the dead
still survives in the doctrine of the Younger Avesta that the souls of the
deceased dwell in the "sun-possessing" abode of ASa (Y. 16.7), which
abode must originally have been identical with the Aryan concept of the
"sun-possessing" rock of the nether world (see p. 118). Therefore, the
"most blissful union with ASa", in which Zarathustra and his initiates
(aSavan-) experienced the "bliss of ASa", must have comprised a con-
templation of ASa, as a mystery, hidden like the dpihitam rtdm of the
Rigveda. It has been argued that this contemplation and the accom-
panying mental "distinction" (vici0a-) were prerequisites for the choice
of ASa, as prefigured.@ Y. 30.5.165
la8 Cf. MoE, RHR, 157 (1960), p. 181, Numen, 8 (1961), p. 63. With respect to my
remark on the reinterpretation of the word aSavan- by Zarathustra it should
that I recently found the same observation in an earlier article by J. H.
Jaarboek Ex Oriente Lux, 6 (1939), p. 227, Godsdiensten der Wereld, 11,p. 402. Any
mystical trait in Zarathustra's preaching is mostly denied, cf., e.g., von Wesendonk,
a'er Iranier, p. 69.
: X. Internationaler Kongress fur Religionsgeschichte (Marburg, 1961), p. 1

(Editor's Footnote :First published in Indo-ImnionJouml, Mouton, The Hague, vol. 8, no. 2,
1964.)
 INTRODUCTORY REMARKS

1. The immense importance of Cosmogony in many so-called

primitive religions is too well kne,,wnto need a lengthy argument.
Although the cosmogonic myth in its essence constitutes a "sacred
history," which describes the beginnings of the world, the role
which this primordial stage plays in man's religious conception of
his world differs widely. Mircea Eliade in his study "Cosmogonic
Myth and 'Sacred History'"1 states that in most mythologies this
primordial stage of the world ends a t a certain moment, and that
the supernatural beings, after having shaped the cosmos, abandon
the earth and disappear: "We can also say that any mythology that
is still accessible in an appropriate form contains not only a be-
ginning but also an end, dete~minedby the last manifestation of
the supernatural beings, the cultural heroes or the ancestors"
(p. 174). He stresses the importance of "the myth of the estrange-
ment of the creator, and his progressive transformation in a dew
otiosus." Man focuses his attention more and more upon those pri-
mordial creative events "which are of consequence for human life.
I n other words, the coherent series of events which constitute the
sacred history is incessantly remembered and extolled, while the
previous stage, everything which existed before that sacred history
-first and above all, the majestic and solitary presence of the
1 R e l i & ~ S t d h 2 (1967):117-83.
 gh God is still remembere
d man, but this is almo
have ended his role by achieving

old good for most reli

some recent descriptions of Indonesian religions t
occurs in a role which can hardly be called that of
One of those primitive religions in which the beginnings of the
cosmos still play a considerable part is that of the Ngaju Dayaks,
the main outlines of which are briefly sketched by Eliade. It is
true, Schiirer in his a le description of the religion2 warns
us, that "the Creati does not give an account of the
creation, but tells us tal community is ordered, and how
the whole cosmos and all its phenomena are related to the com-
munity and its different groups. It-explains the social structures of
the tribe, shows the functions of the various groups, and is the
basis of the system of cosmic classification" (p. 158).
Although in this cosmogony different stages can be distinguished
in the complicated process of creation, the following fundamental
pattern seems to emerge from the picture given by Schgrer : The
contrast between the undivided unity that preceded the creation
of the cosmos-the time "when everything was still in the mouth
of the coiled Watersnake" (the primeval waters)
which is represented as the development, in
a dual organization of the cosmos. A common
however, is that in all of them the opposition between Upperworld
and Underworld, created as a totality, is the central motif.
When viewed from this angle, it does not make a great differ-
ence whether we are concerned with the two mountains which first
arose from the waters and which, clashing together, produced seven
different things, or with the supreme deities whose seats these
mountains were, and who were in a way identical with them,
namely, the god Mahatala, representing the Upperworld, and the
goddess Jata, who stands for the Underworld. Jata creates the
earth and the hills, ahatala the Tree of Life, and together they
create the first human beings. More obscure is the relation between
Mahatala and his sister or wife, Putir Selong Langit, who were
identified with the male and the female hornbill, respectively. As
a result of the extraordinary fight between these two birds the
2 H. Sohirer, Ngaju Religion, the Conception of God among a South Borneo People
(The Hague, 1963). (Translation of "Die Gottesidee der Ngadju Dajak in Siid-
Borneo" [Ph.D thesis, Leiden University, 19461.)
Live par excellence of t h
the contrast between

Eliade rightly stresses the preeminent importance of this notion

of totality, and of the godhead's manifestations as such (p. 176).
I n this connection i t may be worthwhile to study in more detail
some aspects of the Vedic cosmogonic myth. Such a study, how-
ever, involves some methodolo@cal .problems which need some

fact that our oldest source for Vedic

V&. It cannot reasonably be doubted that
from the other Vedic texts by a rather
wide chronological distance. At the time when the other texts
came into existence the Rig-V& must already have been con-
sidered the canonical text par excellence of Vedic reiigion.3 It is
g, therefore, that many descriptions of Vedic mytholo-
cdonell's excellent work, are almost exclusively based
f this collection of 1,028 hymns. No objection can be
 t, however, when
bgy in its totality. I n that
ositions which

will only embrace what is within the scope of a limited horizon,

and will haxe authoritative weight only in regard to things in a
more or less special reference, and that the negative conclusions
especially which may be deduced from such documents must be
.
eived with not a little reservation. . . I am therefore far from
ieving that the Veda has taught u erything on the ancient
social and religious condition of even India, or that every-
thing there can be accounted for by reference to it." Oldenberg
held a similar view : "Es liegt aber in der Natur dieser
sie die verschiedenen Gebiete des religi6sen
enbestandes htichst ungleichmassig beriihrt. Alles Licht
, die beim Soma-opfer
hen voranstehen. . . .
Grund genug, bei Sc
Rgveda nicht e m
acdonell oonfines himself to pointing to the "defective
infornation" of these hymns, Lommel defends his completing the
eager data of the Rig-Veda by utilizing the later BrEhmcc~cctexts
in the following words: "Der Umstand, dass sie spiiter als der
ihre sprachliche Festlegung erfahren hat, ist wie gesagt,
anglicher G m d , ihre Zeugniskraft fur d t e mythische und
herabzusetzen. Man weiss ja, dass das
ten festhalt ; und die Ritudtexte
vieIf&ch Anteil an der gberlieferungstreue der rituellen
Es hat zwar Fortbildung und Entwicklung stattgefmden ;
es ist hinzugetreten, es ist aber vielfach moglich und manch-
nicht schwer, Urspriingliches und Sekundares zu unter-
scheiden. Vielfach herrscht
was von diesem abweicht, muss durchaus nicht etw
sein, sondern kann ofimals die aben des Rigveda in bedeut-
samer Weise erganzen."*
y quoting these scholars I do not mean to say that the te
in which they tried to determine the character of the Rig- V& can
still be accepted without any modification. On the CO
think that the particular character of the majority of the Rigvedic
hymns, and especially the remarkable predominance of the Indra-
V ~ r myth
a that struck the earlier students of Vedic mythology,
is primarily due to the fact that ;the nucleus of the Rig-Veda, the
so-called family books (11-VII), was a collection of hymns that
had been composed for the celebration of the New Year ceremony.5
owever that may be, the fundamental point a t issue is
cannot easily use the data of the Rigvedic hymns without
Emself what exactly the nature of this poetry was.
3, A single instance may illustrate the methodological dilemma
that faces every student of Rigvedic mythology. Only once do
h d in this text the statement that ga "clasps the night in
arms" (V111 later Vedic texts the night is
constantly id .a, Renou denied any c
of V a m a with the night except for this single pass
cordingly assumed, going exclusively by what the texts say, that
the association of Varma and night is the result of a later develop-
ment and does not belong to the earlier stage of Vedic religion as
found in the Rig- V&. It is characteristic that, the real problem that
is obviously a t the bottom of this approach has never been stated
explicitly. 'This is the question as to how such equations
"Varqa is the night," which occur everywhere in the BrGhmayaa,
should be interpreted, It is clear that on the basis of the naive
realism which characterized the attitude of the older generation
of philologists and resulted in such d i s p u k as "is V a m p
the water or the e the equations of the Brdhmaw could
be taken seriously. This was the more difficult because all
ations were found to occur side by side. Neverthe-
consistency with which they occur in the speculations of
should have warned these scholars against passing
too lightly over them. at these texts actually give is not, indeed,
/Berlin, l923),pp. 8,lO;
H. Lommel, Symbolon,
l/Stuttgart, 1964),4:156.
):169-83 (hereafter IIJ).
OS. 4-6 [1960]),p. 124.
(1966):6; and Abel Ber-
definitions which express a ect identity of two gi
but equations in terns of a cosmic classificatory
this framework it is equally possible to say that t h
night and the Devas the day,7 or that one of the Asuras
is the night, and another ( E t r a ) is the day. S
contested that the Rigvedic mythology was alre
such a system of contrasting entities, the question that has to be
answered in every individual case is whether an opposition not
found in the Rig-Veda but only in more recent texts (such as the
Briihmnm) is the result of a later systematization of theolog'
or rather part of the oldest priestly theory about the Univers
the latter case the real problem is whether the Rig-P
about it.8 The possibility that a taboo may have compelled
Rigvedic poets not to mention certain inauspicious aspects of a,
like V a r q a has been considered by several scholars, such
Sylvain U v i and Hillebrandt.9
Renou, on the other hand, sticks strictly to the philological facts
a%teeted,and if certain facts are not mentioned in the Rig- T/'
apparently feels compelled to conclude that they were unkno
not only among the Rigvedic priests but in the entire corpus o
religious notions a t that time. Although it must be a
from a strictly philological point of view this is a S
the di%culty is that we are not here concerned with purely pMo-
logical data alone but with a coherent system of religious notions.
It might be argued that Renou's approach is, the
tious than that of Barth, Oldenberg, or Ellebran
does not decide the problem as to whether in this specific case the
isolated reference to the night reflects a general mythological
concept about which the Rigvedic poets were purposely reticent,
or if this is the first indication of a later development of t
held about the god V a r q a . Only a structural approach
us to understand the function of this god in the total system o
mythological concepts.
A second point that illustrates the methodological dilemma is
the place where Rta, the Cosmic Order, was considered to reside.
From Rksalhhitii (hereafter &S.) V.62.1, where Varuna and
are addressed in the words: "In accordance with [?l @a your
fixed Rta is hidden where they unharness the horses of the sun,'"
7 For referenoee, see IIJ 8 (1964):116.
8 Sea ibid., 3 (1969):211; 6 (1961):63, 67.
9 Cf. 8ylvain Uvi, .La d o c t r k du a&@ dam h brlihmapcw (1898), pp. 167 E.;
Alfred Hillebrmdt, Veddoche hobgie, 2d ed. (Bmlau, 1929), 2:127; F. B. S.
,IIJ 6 (1961):63.
inn. 66,
ttin~pnl
 w e of the anthills (Atharva-V&
a reference to the subtenanean
1s is said to attain
be king of the waters
sdcration des puits,

This digression was untwoidable because the methodological

problem of how to evaluate the Rigvedic evidence in its relation Lo
ature is one of the moot points of Vedic
hological studies. A strictly pMologica1 approach, such as that
of Renou, results in an evolutionistic view which ascribes all
s not directly expressed in the Big-V& to a later develop-
I n the light of the preceding considerations, however, it
seem a justifiable procedure to make--.mith due reserve-
an attempt to reconstruct the Vedic cosmogonic myth by making
use both of the Rigvedic evidence and of that of the later Vedic
hxts, such as the Briihma~as; in other words, by supplementing
he very deficient data of the oldest text by those of later sources,
hich apparently are still rooted in essentially the same system
of mythological conceptions.

4. Owing to a neglect of the data found in more recent texts, the

handbooks on Vedic mythology omit to state that
tinguish two main stages in the process of creation. Not before the
last few decades have scholars come to recognize the fact that the
central myth of the Big-V&, the fight of In& with the dragon
(@rci/Vrtrd), is a creation myth.l"his, however, is only the
aecond stage of the cosmogony. The concentration of the Rigvedio
poets on thia part of tho cosmogony exclusively is probably due to

11 E. W. Hopkins, E ~ M ~ h o l o (Straasbwg,
gy IQlB),p. 26. (Vamyz, by Liiders,
was published in Gott
1% V i p m f l i , 41.2 allmann, Lea enseiggnementa i m -
graphiquea de Z'Agni-Purliw (P&, 1963), p. 132.
18 For references, see IIJ 4 (1980):218, n. 6. In some respects, L. Busohrtrdt,
Vrtra, Det rit& DcemMtdrab i den &, Det Kgl. Dsnslie Viden-
,vol. 30, no. 3 (Copenhagen,
\

of preeminent i
eremonies of the

from this myth of the V@a slaying, that of

d had no direct relevance for the ritual.
till, the frequent references ritual texts, where i t
ccurs in the fixed formula "
beginning," shows that this
speculations on the creation of the
had its own slightly divergent varia
ing is on the whole identic
oldest stock of
ha BrGhmav (VI.
is explained,
(jdmyah) verily ,are the
is born (jclyate)." The e
feminine plural word @a$) to wives god is also found in the
Rig-Veda. Clear references to the p a1 waters, however, are
only found in the last and of the Big-Veda in the
e the references to
y waters which con-
ated the Fire god

u are the most motherly

stands and goes." I n the

odd], which the waters their f i s t embryo, in

sages are found in the

%heprimeval waters ently belong to the old inheritance of

myth of the origin of the world from the waters is the product of
late Vedic poets as Liiders assumes in V a r u (p.~ 121).
aters are accepted as a given fact about whose
origin it would be useless to speculate. At most, the Vedic thinkers
ay have asked themselves the question
.l293 puts about the origin of the cosmo
iti has come into existence?" Accor to the BrtZhmawa, these
ater8 are the "fundament" (p of "these worlds" or of
 other hand, the a re-
hich encloses the 5
The emergence of the earth in these waters is the first stage of the
cosmogony, and different theories existed about the way in which
the earth came into being. One of them was that the earth was the
result of a process of coagulation, an idea also found in Semitic
mythology.16 This can be compared wit nzingya BrLihmav
111.360, line 1 1, BThad-AranYah Upani .2., and Aitareya
Upanisad 1.3.17 The best-known variant of this theory is the
"Churning of the Ocean," which may be called a Creation myth
insofar as the coagulation is here the result of the joined efforts of
the Devas and the Asuras. It
form in texts anterior to the
in the Rig- Veda
others' hands ( any historical connection with the
myth of the primeval waters is notoriously a moot point.18
The second idea, that of the mundane egg,l@is either connected
with that of a coagulation of the waters or with that of a Creator.
The first variant is met with in the version of the Jaiminiya
Brtihmapa (111.360,line 7), according to which there were originally
only the waters, the billows of which collided with each other. As a
result of this, a golden egg came into existence and lay there for a
hundred years of the gods. I n a different version, that of the
&&upatha Brtihmana (XI.1.6.1-2), a golden egg arose from the
heat produced by the waters as they wanted to propagate them-
selves. That egg floated about for a year and in that time a man,
PrajBpati, the god of the primordial undivided world, came into
being. He then broke the egg open but i t had no foundation,
and he floated about on that egg for another year after which
14 Cf. &&patha B~tihnaapaVI.7.17, VI.8.2.2, VII.4.1.8; Aitareya Brtihmapa
111.6.4; Aitareya Upan+ad 1.2.
15 See Ma.itr&Jay;iSahhitLi 1.4.10 (p. 59, line 5) iycim v&
'Gcigrahu, K$haka Sahhitii XXXII.7 (p. 26, line 15
16 Cf. A. J. Wensmoe, The I d m of the Weatem Semites coneeming the Navel oj
the Earth, in Vterhandelingen der Koninklijke Akademie van Wetemchappen te
Amsterdam, Afdeelkng Letterkunde, n.s., 17, no. 1 (Amsterdam, 1916): 5:
"coagulated billows of Tehom." The same idea occurs in the Orphic cosmogony:
"H6d5r Bn ex srkhC kal h~%,ex h& ep&g6h6 g&" (cf. B. H. Stricker, De Geboode
van H m [Leiden, 1968],2:142, and also A. H. Krappe, La ge&e des m@e-~[Pm&,
19381, p. 256).
17 SW &so mzirtirtaram, prdmCrcm X.5.3.3). Van Buitenen's divergent
interpretation of the Byh. AT. Up. p JAOS 77 (1957):Qldoes not m m
preferable.
1s See, e.g., Geldner, Pe8tgws Rdh (1893), p. 192; Kuiper, T u m r Ju
Volumc I (1958), p. 360, n. 21.
l9 H. Lommel, "Der Welt-ei-Mythos im Ri -V&&," in M b n g e s Bally (Geneva,
1939), pp. 21440.
-PIJOfieq? J0J P W !'g 992 'dd 'Bdd'8- !'g 628 ' L ~ z :
' P m W $3 9E'I "WmPDH !(.P "W)) E'66Z'IIX WJ
'91I 'd '(IZBI '?zd!e?) FJ P& 'uapmls?.~~n Wws 'n-ea .a ws W
"Emti~a,"but this is

Athwa-Veda?5 may even be historioally identical with the

arZixa of the Avesta, and thus have its origin in the common
ndo-Iranian mythology (see Charpenkier).
There seems, also, to have existed a speoifically Indian myth in
23 Maitr. Sa~hh. 1.6.3; Ktith. Sa&h. VIII.2; Kaefhaln-Ka#haSavih. (VI.7; Taitt.
Saliah. VII.1.5.1. Cf. Taitt. Brahm. 1.2.1.3 (see also Apaatamba &auta Sfitra V.1.7)
with commentary; and Taitt. Ara?yaba X.1.8.
24 Cf. gat. Br. XIV.I.2.11. See Kuiper, Mdedelingender Ko?inklijkeNederlandse
AI%adem-ievan Wetenschappen, n.s., 13, no. 7 (Amsterdam, 1950): 18. This
view was criticized by J. Gonda in Aspects of Early Vi+uism (Utrecht, 1954),
pp. 139 ff. (see also Eliade, Revue de l'histoire des ~eligiom[l
disregarded the fact that in ell the older Pajurvedic texts
exclusively given to the boar which occurs in a non-Aryan myth (whose central
motif was the boiled rice, o d u d ! ) , and which had a clearly demoniacal Vfira
character (see also Benveniste in E. Benveniste-L. Renou, Vftra d VrBragw
[Paris, 19341, p. 195). It is charwteristic that this myth, as a parallel of *he Vfira
myth, is ~ o c i a t e not
d with the World-Father Prajtipati but with Indra. Both on
I
l

n reads as follows: '

nest of the waters, on it he piled the fire, that became this (earth),
then indeed did he find support." What must have been meant by
the expression "nest of the waters" appears from those passages
where the moist lairs (Grdrd ybnayah) of the f i e god are contrasted
with those which "have a nest" (hE&ih).26 The word "nest,"
accordingly, seems to refer to a more solid state of aggregation (in
the midst of the waters?).
This passage had3 here been quoted in full27 not so much because
of the interesting initial manifestation of the Creator god as the
Wind, as because i t describes so chmacteristically the lack of
'csupport" (prati~th), of a 6xed point in this primary stage of the
for the second, contaminated, version, it can here
be disregarded as it occurs only in a single text which dates from
the end of the Brghmana tradition and, being composite in more
than one respect,28 is of no interest for our purpose.
From the preceding passages it may be concluded that the first
stage of the cosmogony was an undivided unity, a rudis indigm-
t ~ u e-- ~ in which
l ~ ,the instabilis tellzcs deserves partii%&i~notice.
;1 some ritual speculations, it is true, Prajgpati, the Father of the
3 Universe, finds a t last aprati&hd, a support, by piling the sacrificial
I
fire on the "nest of the ~ a t e r s . " ~The
' J most prominent character-
istic of this primordial world remains, nevertheless, that the m m -

2% Cf. Ha&. Samh. 11.7.15 (p. 98, line l l ) , III.4.7 (p. 63, line 14); K@h.Samh.
XXXIX.3 (p. 121, line 4).
27 Translation by Keith of Taktt. Samh. V.6.4.2-3. The parallel texts K@h.
Sahh. XXII.9 (p. 64, line 13) and Ka&.thala-Ka#mSamh. XXXV.3 (p. 179, line 16)
are similar.
28 In Tailt. Brcihm. 1.1.3.5-6, Prajgputi sees a lotus leaf standkg in the waters.
He then t h i i : "Verily that [eazth] exists on which this [lotus leaf] can be
founded." He then becomes a boar, dives for some earth on the bottom of the
waters and spreads this out (aprathayat)on the lotus leaf. Hence the name qzjhivt
for the earth. I n this version the lotus leaf standing in the waters would seem to
be a transformation of an older motif, that of the golden reed: "He who knows the
golden reed standing in the seu-he verily is in secret Praj6puti" (translation by
Whitney of Atham-V& X.7.41).
29 K@h. Samh. XXII.9 (p. 65, line 14); Ka&.tluda-Kafh Samh. XXXV.3
(p. 179, line 16); Taitt. Sahh. V.6.4.2.
 on the waters and that the
d point, a "support."
e little doubt that this lack of a settling point is of
essential importance in the initial stage of the cosmogony. As we
shall see below, the appearance of a male figure in this primo
world is needed to create such a fixed point from which'the e
can develop. As a mere parallel, just to underline the contrast
between male and female in this context, the following episode
may be quoted from the Creation myth of the Ngaju D a y a h
(Schlirer, pp. 28 ff.). As a result of an extraordinary fight between
the two birds, there arise two boats, a golden and a jewel boat. At
the same time a maiden and a young man come into existence. The
maiden "gets into the golden boat and sets forth on the primeval
waters. . . . The young man gets in the jewel boat and sets out on
the primeval waters. Suddenly the two boats run together. The
young man sees the maiden and his heart is seized with love for
her. Ke wishes to marry her. She agrees to his request, but on
certain conditions. They are still floating about on the primeval
waters and have no place where they can settle. She longs to have
such a place before she will become the young man's wife. The
young man is sad. Mahatala sees his distress. Ke makes an island
in the primeval waters. . . . On the island rises the mountain
Kangantong Gandang." The interesting point is that the island is
made as a result of the male and female principles meeting each
other while floating on the surface of the primeval waters. I n the
Vedic myth the corresponding process has a more aggressive
character but, as I hope to demonstrate, here, too, the appearance
of a male god is the decisive factor.

THE CREATION OF THE DUAL COSMOS

5. Just as the way in which the earth rose in the primeval waters
is described either as an autonomous process or as the result of an
intervention by a Creator god, just so the second stage is presented
in two different ways.
On the one hand we f h d in the Brtihmnas the fixed formula,
"These two worlds were together; they separated."30 Sometimes
the idea occurs in a more elaborated form, as in the $atapatha
Briihmana (1.4.1.22), where it is said that the heaven was so near to
the earth that it could be touched by the hand. It is quite in line
80 Rig-V& and S6ma-Veda: imau vai lokau aahG 'atclm, tau vyaitiim; Yajur-
L Veda: inte v d aahE 'skim, te vyaitiim.
 whole society seems to

in the Rigvedic hymns.

K M . S U M . XIII.12 (p. 193, line 12) iwq vui aahii ? s h , te v6gur vyavcit;

@S. VLII.94.11.
 tbook for the cere-
m to provide a satis-

en refer to the "powers of resistance"

ated in the mythio
to such an extent t h

this myth the poets (with two exceptions) always refer to a single
mountain.37 Indra wins the salutary goods for this world either
by slaying with his v4jra the dragon who tholds them from the
ods (devcis), or directly by piercing the mountain. Liiders (Varuna
,pp. 170 E.), rightly stresses this ~it~uationin the following w
So wird denn auch in den Schilderungen des V$rakampfes von
I n h immer wieder gesagt, dass er seine AngriEe gegen den
richtee, gerade als oh dieser der eigentliche Peind sei."
out that the poet of @S. 1.54.10 seems to conceive this
m a kind of stone case, which lay in the belly of the dragon (p. 172),
d in a separate no on page 174 he writes: "'Das himmlische
eer, in dem sich die asser, Soma und die Gestirne befinden, ist
sen, einem Steinbehiilter, eingeschlossen. Daher
im Vfirakampf im r gegen den 'Berg.' Er ist der
Steinverschluss der himmlischen sser, den Vflra verschlungen
hat. Demelbe Berg ist der Vala, aus dem die orgenrtiten befreit
werden" (a similar passage occurs on p. 332).
Although Liiders's assumption of a celestial ocean is open to
dispute, a critical discussion of this detail would here be out of
order. The main point is that Liiders has rightly recognized She
situation in the fight of Indra with the dragon, whom the poets
describe as lying either on the waters (Liidew,p. 167, n. 6) or on the
mountain (Luders, pp. 170 E.). The real character of this mountain,
however, remains unclear. illebrandt contemplated
bility of identifying it with the imalaya or the
(Vedische Mythobgie, 2:161), and Luders did not live to elaborate

a* Cf. Hillebrandt (n. 9 above), p. 182; Kuiper, Museum 59 (1954), col. 84; IIJ
4 (1960):222, and 5 (1961-62):169 ff.
85 See Benveniste-Renou (n. 24 above),
35 See n. 9 above, p. 234, p
--
. pp. 9.5-97.100.
37 RS.I.32.2(butapluralin1),51.4,54.10,57.6;IV.21.8;V.32.1~d2;~.30.5;
VIII.64.5; VII.79.4; IV.17.2. Cf. &lsoVI.24.6;1.73.6,etc. Forthe plural in IV.19.4,
see below.
his idea of the stone case.38 Neither
inclined to take the equation of V$ra an
IV.5.l. ([pp. 62,151) seriously.

6. Some years before the publication of Luders's life work, i t had

been pointed out that the mountain against which Indra directed
his assaults obviously was the primordial mountain, and that his
fight against the dragon describes a stage in the process of
Creation when "the gods, accordingly, had not yet the disposal of
the amrkq'amna, since this, along with all the essentials of life (such
as water =rwa, essence of life, and the sun =the light of life and
the world) was still guarded in the primordial hill by the serpent
representing the nether world. The fact that Indra wins it and
causes the sun to rise constitutes his great Creation act, which is
repeated every morning and which leads up every year, a t the
time of the winter solstice, to a new crisis."sg It
t here the netherworld, which forms part of the dual cosmos,
been too rashly identified with the undivided primordial
world.
Taken by itself, the idea that some of the Vedia mythologerns
may have some connection with the winter solstice was no novel
one. Ludwig had been the first to explain the Rigvedic hymns to
Dawn (Ugas) as referring to this period of the year, and Hille-
brandt had adopted his idea.40 Although this theory has univer-
sally been rejected, it still deserves serious consideration.41
Hillebrandt took a further step in that he also associated the
V$ra fight with the winter solstice (Vedische Mythologie, 2:182),
but he failed to recognize the true character of the myth. Liiders
on the other hand, in criticizing Bllebrandt's view and stressing
the character of a Creation myth, disregarded the fact that a myth
of the creation of a primordial world tends to be perpetually
actualized in ritual.
When the lines quoted above were written I did not yet know
W. Norman Brown's article on "The Creation Myth of the Rig
38 See Alsdorf's remark in H. Ludeig, V a m ? (Cottingen,
~ 1951), 1:332, n. 6.
39 Museum 52 (1947), col. 200, in a review of Buschardt, V e r a (see n. 13 above).
See also Bijdragen van het Koninklijk Inatituut vow Taal-, Land- en Volkenkunde
107 (1951):72, and Mededelingen Kondnklijke Nederlanclae Akademie van Weten-
achuppen, n.s., 14 (Amsterdam, 1951): 219.
40 Ludwig, RigzIeda 4 (1881):xi, and 6 (1888):173a; Hillebrandt, Vedische
Mvtholnoie. lst'ed; (1889). 2:26 ff.. and 2d ed. (Breslau, 1927). 1:28 ff.
Liiders's earlier interpretatio
Weltschopfungsmythus,"
1951.42 The circumstance
were, independently from each other, published about the same
time is a noteworthy coincidence. That the Vrka myth is a
Creation myth (whatever modification one m
for this term itself) seems now to be general1
The preceding brief remarks m
provide a background for the fol
Lain." I n the nineteenth century
interpreted as clouds which contained the rain.
was still inclined to believe that
the myth (Religion des Veda, p.
I can see, had the "mountain"
hill, whose presence in the Rigvedic hymns had not, indeed, been
seriously considered at all. It may, therefore, not be out of order
to add a few words in justification of this interpretation.
first be observed, then, in view of such te
e slew the snake lying on the mountain" (
riyEnhm), that the concept of a snake
the primordial hill is also a well-known motif in other religions, for
example, among the Semites.43 Second, it should be noted that in
the Vedic tradition traces are found of an ancient belief that the
earth and the mountain(s) originally lacke a firm foundation an
moved about until Indra them. Thus &X.
Macdonell's translation) : ' made the wides
quaking steadfast, Who set a t rest the agitate
measured out air's middle sp
support: he, men, is Indra."
11.7-8): "Evenly the earth has exp
which wanted to runzhas subsided
carefully" (cf. VI.30.3 : "The mountains
Thou hast fixed [dT@hni] the spaces,
"The shaking mountains and the plains he fixed"). I n two

42 See Liidem, 'Iraru?2a,pp. 183 ff.

43 See Wensinck, T h e Navel of the Earth (n. 16 above), pp. 59 ff.; Tree a ~ Bird
d
aa Coslnological Symbols WesternAaia (ibid.[192l]), p. 13. For Greece, see Harrison
(n. 21 above), pp. 320 ff.: "when we remember that the omphalos is the very seat
and symbol of the Earth-Mother, that hers was the oracle and hers the hoIy
oracular snake that Apollo slew." Cf. also E l i d e , TraitB d'histoire dea religions
(Paris, 1953), pp. 250 ff.
 : now it came near the

t is not necessary for the purpose of this study to enter into a

mare detailed discussion of the primordial hill and its symbolic
representation on the sacrificial place by the fire altar ( v a i ) . It is
d, however, that the preceding observations have made it
iently clear that, in spite of the differencebetween the various
versions in which this mythic concept is represented, there is a
fundamental notion that is common to both forms of the myth,
namely to that of the irmlabilis Gerra (conceived either as the cosmic
egg or as the mud which the boar brought up on his snout from
the bottom of the waters) and to that of the primordial hill which
lacked a "foundation" and had to be fixed. By combining the
'ous data me get the following picture: I n the beginning there
only s small hill, floating on the surface of the waters. From
44 Itiaitr. Satiah. 1.10.13 (p. 152, line 12); K&h. Safih. XXXVI.7 (p. 74, line 5).
it the earth expanded in all directions (aprathata). By piercing this
"mountain," in which the germs of all life were contained, Indra
a t the same time made i t firmly rooted in the bottom of the waters.
Since this mountain was the cosmic center, the central point of the
earth, the whole earth became thereby "iirm and steady" (dhruvd,
ci4ithiE). Thus the cosmic mountain not only was the origin of the
earth but also came to function as the "peg" which secured the
earth a firm support (prati~thci).This idea still survives in the later
literature, where Mount Mandara as the cosmic pivot is called
"Indra's peg" (Idrakila)and the concept of a "mountain function-
ing as a peg" is expressed by the t e r n kZl6dri.45 The word "earth-
peg" (ku-kila) for "mountain" has its parallel in Java where the
name paku buwana, "peg of the world," is for the same reason
given to the hill Tidar, north of the Borobudur.

COSMOOONY AND CONCEPTION

7. It need hardly be mentioned that such notions about the origin
of the world as described above on the basis of the Old Indian data
are far from rare. I n the very instructive material collected by
A. J. Wensinck from the literature of the Western Semites one
meets with such passages as the following (from an Arabic source):
"Forty years before Allgh created the heavens and earth the Ka'ba
was a dry spot floating on the water and from it the world has
been spread out." Although the purpose of the present study ex-
cludes a comparison with other mythologies, an exception must be
made for the words in which the Syrian, Jacob of Edessa, described
the primeval waters and the earth before the cosmogony: "Scrip-
ture says: 'The earth was invisible' because of Tehom which was
stagnant and surrounded it on six sides like the embryo which is
swkounded by the membrane in its mother's womb." Similarly a
Jewish source likened the beginning of the world to an embryo:
"The holy one created the world like an embryo. Like the embryo
proceeds from the navel onwards, so God began to create the
world-proceeding from its navel onwards and from there it was
spread out in different directions."*B As these passages show,
people sometimes interpreted the cosmogonic myth of their sacred
tradition in terms of a cosmic conception, as we have seen on the
preceding pages.
45 See Johnston's note on Buddhucarita 1.21 in The Buddhucarita, ed. E. H .
Johnston (Calcutta, 1936), pt. 2, p. 6, n. 21.
46 Wensinck, The Navel of the Earth (n. 16 above), pp. 18, 3, 19.
"' (see XI. 1) with respect
read this into (or, recognized it in) the cosmogonic my%hof their
sacred tradition. n the other hand, in the case of the young
Aranda as interpr ed by Eliade, it might be better to speak of an
indirect anamnesis, which enables the novice to identify himself
with the hero of the prim'ordial world. If it can be shown that an
anamnesis of the first type is possible and does actually occur, this
might throw some light on the psychic mechanism by which that
of the second type, the indirect one, can take place. I n that case,
indeed, it might be argued that the novice can identify himself with
the primordial hero owing to the fact that the myth functions as a
generalized picture of everyone's prenatal state, and that via the
myth, every novice is thus brought into contact with his own pre-
natal existence, wtthout, however, being aware of it.
It follows, in the third place, that i t d l be necessary to distin-
guish between two problems: that of a possible function of the
cosmogonic myth of helping an individual to reexperience his
conception and thus, in a way, start his life again, and
naturally very complex ques as to whether (and if so, how) a
myth, which forms part of ole culture and, as such, is con-
ventionalized, can be expl in the personal
anamnesis of a single individual, in short, the problem of the

culty which besets a student of Indian religion

when he hazards to put these questions is not so much that he
ventures upon entering a field in which he is a layman, as that
experts are themselves still far from having arrived a t an agree-
ment. This is the main reason that has withheld me, for nearly
twenty years, from overtly formulating this as a problem worthy
of consideration.
The moot point with which we are here confronted is that of
Exactly twenty years ago Nandor Podor
Search for the Behed47 in which, elaborat-
ing Rank's ideas about Das Trauma der Ceburt,48 he pointed out
that events which have taken place during the prenatal period
apparently have been recorded by the embryo and can be repro-
duced in dreams in his postnatal adult life. The idea that man
carries with him throughout his Wetime the recordings of non-
perception from his prenatal life obviously conflicts

47 Nandor Fodor, Tlre Search for tlre Beloved: A Clinacal Inveatigath of ths
e w York, 1949), viii+.ioo pp.
Bedeutung fCir d& Pdpjdmam-
01.14 (Leipzig, 1924), 207 pp.
seriously with the current assu sychologists that e
sciousness only begins to develop sfter birth, along
velopment of the organs of sense. It is not surprising, $herefore,
that psychologists have expressed their disbelief for this very
possibili-ty of prenatal nonsensory
astic change in one's general out1
sciousness. Not only the psycho1
ever, but also the large majority of psychiatrists
unable to accept the conclusio

possibility of reexperiencing one's own conception might be of the

eatest importance for our interpreta"con of religious
and particularly of cosmogonic myths. To illustrate t
sufficient to quote here Ghe words in which a Dutch
pist, after stating that "the faculty of recollection is c
the ovum and not with the spermatozoan
recollections of the ovum as they are repro
adults. The ovum. "has come out of the ov
for things to hap
state of oceanic
about, as it were, on
of being past of this
sciousness. There o
ovum being part
depicted in dreams
such as a group, a c
aponding ovarial expe xperience of the time
before ovulation. No about spermatozoa"
at this stage.49
It is, of course, quite in line with the best tr
science that in such a controversial problem,
assumptions of a discipline are being questioned, the dispute m

49 M. LieLaart Pesrbolte, P~enatalDynamim (Leiden, 1964), p. 176; and The

Orgastical Experience oj' SF and Metapsychologic Psychagogy (Leiden, 1955),
pp. 18 ff.
go on for many years to come, and that it will be a long time before
an agreement is reached. n the other hand, in some forms of
painting and poetry, elements occur which
reflecting prenatal experiences, and some art
ts appear to be quite aware of this aspect and
e of the psychiatric theory. The main trouble for
at the psychologists do not seem to be very much
ion of the psychiatric evidence, as a result of
ch an agreement on a scientific level may indefinitely be post-
ed. Therefore, the only thing that can be done pending the
question is to formulate provisionally the problem which naturally
dent of archaic cosmogonic myths.
may poink out that in the Indian cultural tradi-
tion, in which some indivi uals believe, and are believe
re~ollectionof a former existence, a remembrance of one's own
prenatal life is not a t variance with the general cultural pattern.
though this does not in the least decide the scientific problem
referred to, the descriptions in literary works of what their authors
considered to be a direct anamnesis of their prenatal existence show
that if one tries to inter me data in the light of the theory
Peerbolte and others, one is not
in the context of the culture con-
gh it must be conceded that in India the cultural
former existences rather than on prenatal life, an
lection is found in a
viicakar, who is sup-
.D. The lines 11-25 of
the fourth hymn of the T i r u v h h m (which Reverend Pope, for
not to translate in his well-
elephant to ant, in the defect-
the effects of karma, I n the b a h
having gone through the
onth having gone through
nty whether or not the embryo
having gone through [perhaps,
having gone through wanton-
one through great difliculties,
om the springing lymph, I n
gh the lowering of the earth
onth having gone through
S which arose in the ninth
gone through the
 sea of miseries, expesence
is passage, which pretends

of secondary importance.51
Instances of a possible indirect anamnesis are of lesser interest
because in these cases the author describes his experience in terms
of the traditional concepts of his religi t is true that what he
describes within this conventional CO
aonal experience, but a possible interp
of prenatal life finds no support in this con
of course, of such well-known passages
where Vasi&ha, addressing his V a r u p , records how he has
entered the god's subterr ce and has been initiated by
the god: "When V a r q a the boat and propel it to the
midst of the ocean, when we move across the waters'
waver in the swing." Here the seer i
of the undivided primordial world.52
ot receive his final vision until the god has placed him in the
(cf. verse 4: "Varu~aplaced Vasigtha in the boat; the well-
working [god] has made him an fsi [seer] by his mighty powers").
n the light of the description of the oceanic feelings given in the
modern psychiatric literature, i t would seem interesting thdt the
poet Vasigtha here refers to his "rocliing" or "swinging"
(inLhayZvahai) on the waves. I n contradistinction from t
mystic, however, who probably lived so
later, this Vedic seer was apparently
connection between his mystic vision
The modern student of Vedic religion may consider the possibility

presents many serious dficulties, which may have withheld

om translating it. For the translation given here I p r o m much
from the expert knowledge of Mr. A. Govindankutty, who is worbing at the Kern
Institute. The translations of individua hich Pope gives in the "Lexicon
and Concordmce" to his edition mostly entirely different understand'
of what the poet meant to say in the ines. The translation by R.
Schomerus, Die Nymnen des McZpikka-V S a g a (Tiruv3aga),Religiose Stimmen der
Volker (Jena, 1923), p. 17, cagain differs in many respects.
61 See, e.g., the commentaries on Vasubandhu's Abhidhamkoda 3 in Louis de le
Tall& Poussin, tram., L'Abhidhumakoh de Vambandhu, troisidw chapitre (Park/
Louvain, 1926), pp. 64 ff., where a number of references to parallel sources
given. I owe this information to my colleague, Prof. D. Seyfort Ruegg. Also,
Alex Wayman, 1TJ 3 (1969):64, 70, on the ten "statw of womb.'
68 For this interpretation of the character of V w a , we, Q.
I10 ff.
that the mythological figure of Varuna as the god of the pri-
mordial world could psychologically function for the devotee as a
means of getting into contact with the earliest layers of his per-
sonality and of reexperiencing his own prenatal state, but this is
bound to remain an interpretation that admits of no objective
verification.
I n the light of the current parallelism between myth and ritual,
however, it should be noted that in the Vedic ritual of the initiation
(dik~li),the diksita must again become an embryo (&at. Br. III.
1.3.28) in order to be reborn. The dangerous and inauspicious
character of the diksita while being tied (which is expressly stated
in Kii& Xa~hh. r. Sarizh.III.6.7: p. 69,
11)must proba g in Varuna's realm.53
The same notion of rebirth also underlies the statement of the
ritual texts that the sacrificer by sacrificing regenerates his own
self (&at.Br.VII.2.1.6). There is, however, no myth which directly
refers to this previous stage in the womb, nor is there any associa-
ogonic myth with the birth of an individual, such
awaii. Here the Hawaiian creation chant, Kumu-
lipo, is sung when the princess is with child until the child is born.54
9. I n the fourth and fifth sections of this study an attempt was
made to demonstrate that the Vedic cosmogonical myth comprises
two different stages: fist, that of an undivided primordial world
which consisted of "the waters" and the beginning of the earth
floating on the surface of them; second, that inaugurated by Indra's
throwing his weapon, the vdjra, against the dragon or the mountain
itself, which thereby became firmly founded.
If the interpretation of this act as a second stage of creation is
accepted, the strongly aggressive character of Indra's deed, and
the necessity of overcoming a strong "power of resistanceJJ
(vytrd),impersonated by the dragon as guardian of that mountain,
remains remarkable. It has further been inferred from the Rigvedic
texts that it was this second stage that played a predominant role
in the cult, inasmuch as the annual ceremonies which aimed a t a
renewal of life were a ritual reenactment of this primordial Creation
act. For Vedic society, accordingly, this "breaking of the resist-
) was the main mythic event, which conditioned
a renewal of the world and the individuals, and in the (presumably
annual) ceremonies this renovation was brought about by con-
tests a;nd fights which imitated the primordial combat of the god.
 According to the psychiatrists quoted above, some dreams pro-
duced during the psychoanalytic treatment of patients point to
oceanic feelings of "bobbing on the waves." hile in the earlier
psychoanalytic literature this had been explained as a recording of
the embryo being surrounded by the amniotic water, the study of
prenatal dream material has led to the conclusion that these
oceanic feelings must be traced back to the state of the ovum be-
tween ovulation and fertilization. Attention has further been
drawn to a parallelism between these feelings and the psychic state
which the mystic Eckehardt described in the words: "one is as
empty as one was before one existed."55
If we assume, then, by way of a working hypothesis, that the
first stage of the cosmogonic myth is "somehow related"56 to Ghe
oceanic feelings which the ovum seems to have recorded in its state
between ovulation and fertilization, the question naturally arises
2 Indra's fight against the "mountain" may correspond to the
fertilizing activity of the spermatozoon. It would not be difEicult
to interpret in this light the vcijra, which penetrates the mountain
and opens it, whereas there is hardly any detail in the Indian
cosmogony that points to the Oedipus complex of the orthodox
Freudian psychoanalysis. Some passages might be quoted that are
particularly interesting in this respect (e.g. Taitt. Sahh. 11.5.3.2.):
"When Indra had slain Vytra, his power and strength [indriycih
viryitm] went into the earth; this became the plants and roots."
I n this connection, such specific details as the riveting of the
mountain or earth, which floated about on the waters, may be
significant. I n case one looks for an equivalent in prenatal ex-
perience, the parallelism with the embedment of the fertilized ovum
is obvious. That the fertilization is expressed in the mythological
symbolism by Indra's fight against the mountain is not surprising.
It is, .@deed, a well-known fact that the spermatozoon has to
overcome a strong resistance by dissolving the hormonal liquids
that surround the ovum.57 I n conception-dreams the Yvum ap-
pears sometimes to have recorded this process as an aggressiveness
on the part of the spermatozoon, which may result in a shocklike
experience of the conception.58
65 Lietaart Peerbolte, Prenatal Dynamim, pp. 176 ff.; also Nandor Fodor, New
Appoaches &o Dream Interpretation (New York, 1951), p. 94.
86 For these words, see below.
b7 Recent experiments with fertilizationof ova extra Utemm (see R. G . Edwards,
B. D. Bavister, and P. C. Steptoe in Nature, February 16,1969) provide a striking
illustration.
88 For the conception-shock,me Lietaert Peerbolh, Prenatal Dy.nasnica, p. 79
("there must be some shook, some thrill in conception") and pp. 166 ff., and Orgm-
to consider more dosely a ritual in which the fertilization of the
ovum would seem to be symbolically reenacted. This is the dance
which is still performed among the atak when the god
ane na Bolon has completed his circular course and a new year
ill begin.
10. The main features of this "magic-religious dramatization of
the 'total' cosmic renewal" are here briefly summarized from an
eyewitness account of it.59
open space a double square (one crosswise in the other)
in such a way that the eight corne
o the eight points of the compass.
mbol of the Tree of Life is drawn.
ents the middle world, shows the
) that are symbolic of the three worlds. I n the
center of this cosmic symbol there is a drawing of a small tortoise
a hurma di Gano), which represents the underworld deity.
n the center of this naga, again, there is a circle, which accordingly
is the mosL central point of this representation of the cosmos, and
in this circle an egg is placed. Nearby, in a basket Bled with rice,

tional attack"), p. 38 (the spermatozoan ag

Fodor, New Approaches, pp. 95 ff.
m");
T h e StruGture of the Toba-Batab Belief i n the Eligh God
 e tree of life dies, a new one de-
velops from the same trunk, nourished by the fruit fallen from the
old tree." Tobing points to "the essential oneness of myth, rite and
magic." There is a "symbolical destruction of the cosmos, the
community included," and renewal by the planting of the same
tanggal panaluan, which has been in the possession of the commu-
nity from generation to generation, but which is now decorated
with young leaves.
All this may be perfectly correct but it leaves the meaning of the
egg in the very center of this symbolical Universe unexplained.
Tobing only remarks that i t "no doubt symbolizes the fruit of the
old tree of life and consequently it is a representation of the new
cosmic life" (p. 173). That it symbolizes new life can be hardly
doubted (cf. e.g., Eliade, Trail&,p. 354), but if it represents the fruit
of the tree of life why is an egg chosen as a symbol, and why has
the datu, after having moved the staff in a way that might be a
symbolic representation of the coitus, to pierce i t in the last and
most dramatic moment of the ceremony? This final climax, the
planting of the Tree of Life in the egg, rather suggests the idea of a
cosmic conception, a reenacting of the primordial fertilization of
the world-egg.
h that case the "rebirth" of the cosmos and the community
would have to be induced by a new conception, as the first stage
of a new world and new life. This notion, it is true, is not reported
to be current among the Toba-Batak. On the other hand, if this
supposition is correct (and it is, I think, the best way in which the
central role of the egg can be explained) we have to face the conse-
quence that such a symbolical fertilization of the world-egg is
inconceivable unless there was in the past an intuitive knowledge
of how the human conception comes about, that is, a direct
anarnnesis. Anyway, the element of aggressiveness in this planting
of the "new" Tree of Life, and in general, the close connection
between destruction and renewal should be noted. The ritual is
instructive in connection with Indra's Creation act.
11. If, by way of hypothesis, we relate Indra's fight with the
dragon '?€tesistance" to the struggle of the spermatozoan to con-
quer the "resistance" of the ovum, some further questions arise.
The analysis of conception dreams has also brought to light the
existence of a strong libidinous connection between the ovum and
the maternal psyche before the conception.60 I n psychopathologi-
cal cases there are indications of a fixation of the embryo's psyche
60 L i e t a r t Peerbolte, Prenatal Dynamics,pp. 62 ff.
2
mother takes revenge, he
of Grendel, a monai;rous

61 Cf. e s p i a l l y in AV&: A

7: '"repo~&~.tavec Iw D h u ' qui p u t

DEtnu': toute la 16gende do la mere de
tenca a mother D&nufor V$%raY'
(which
 n it is noted that in the Old
similar monster is said to have
about the yz~letide (accordingly in midwinter),
1, until in the third winter it is slain by the hero
etail is particularly interesting because in West
Cornwall the period between Christmas Day and Twelfth Night is
guise dancing, the streets of Saint Ives and the
nzance being nightly paraded by parties of young
people attired in strange costumes. Down to a hundred years ago,
however, the guise dancers in Cornwall, W also in Scotland, used
to enter the kitchens of the larger houses and to act there the old
play of Saint George and the Dragon, hundreds of versions of which
are found all over Great Britain.6"uch data from the Germanic
world may lend some support to the theory that in Vedic India
Vrtra had to be slain a t the beginning of every new year, as a re-
enactment of the Creation.
As for the mythological significance of the dragon's mother,
however, it must be admitted that in the Vedic myth we are moving
on insecure ground since the evidence is very scarce. The myth of
Indra's killing DcZnu, indeed, admits of two possible interpreta-
tions. On the one hand, it might be argued that the phrase "son of
the primeval stream" has given rise to the notion of a mother
that bears the name Ddnu, in much the same way as in later times
the Rigvedic word $de.tp6ti ("Lord of the power,") mostly used as
m epithet of Indra's, was reinterpreted as "Husband of $aciWand
thus generated the figure Saci, wife of Indra. er all, the terms
ytrciputrG ("having V ~ t r afor a son") and shh ("procreating
mother") only occur in a single stanza of a relatively late hymn and
may have been peculiar to the phraseology of a single late poet,
who took the relation of the dragon to the waters that begot him
more literally than others did. But then there still remains the
irely different question as to why this mother had also to be
ed, and this is not so easily answered. If, on the other hand,
we assume, in spite of the very meager textual evidence, that the
killing of the mother of the dragon is a very old element of the
myth, which can typologically be paralleled to the killing of
Grendel's mother, it would be possible to point to the findings of
the conception analysis, namely that the ovum as a "recording
apparatus" preserves the memory of the ovary and of the mother's
psyche, which, in repsychotic states, may be experienced in
134 See, e.g., A. K. Hamilton Jenkins, Thestory of C m w e l l (London, 1948), pp.
131 ff.
 In the process of individualizaLion
rm of the ovum by the spermatozmn,

to exist as a separate indi

some fundamental problems of interpretation of myths. For the

Vedic Indian the Indm myth existed in its own right, and no
modem theory which tries to find an equivalent to it in the
subconscious mind is relevant as far as his system of concepts ia
concerned. To him the myth was a means of viewing the world and
his life in the context of a given set of values. Myth in general
enabled him to experience the brief compw of his individual
existence cls meaningful, both within his society and in a wider
cosmic context. On the one hand, therefore, it is unacceptable to
treat myths as mere fantasies or literary motifs, a procedure still
not uncommon among philologists. On the other hmd, there is no
guarantee that every motif is equally meaningful. Besides, even if,
in principle, the possibility is conceded that the main pattern of
the cosmogonical myth can be correlated with an individual's
prenatal life, we lack the criteria to determine just which elements
should be viewed in this light. It would obviously be childish to
expect that the whole cosmogony could aimply be read as a re-
cording of an individual's prenatal experiences.
For that reason no attempt will be made here to explain further
details, although some of them, such as Indraysovercoming his
father Tva& (RS.III.48.4, IV.18.12), might prove meaningful
from this point of view. The object of this study, however, can in
the present s t a h of things only be to draw attention to the poasi-
bility that the cosmogonical myth may have some aspects that
have so far escaped notice.

TREE OB LIFE

13. The only exception that must be made concerns the Tree of
Life, because this is one of the most important motifs in many
cosmogonies. This tree, rising from the primordial hill, separates
and at the same time connects heaven and earth (or, Upperworld
and Underworld), and is as such the mogt striking symbol of the
dual organization of the world and its inherent unity. From the
 renatal psychology it has been suggested that
the umbilical cord.05 If the cosmogonical myth
is actually "somehow connected" with a recollection about the
ovum, both in its unfertilized state and after the fertilization, it
is, indeed, fully legitimate to ask what the prenatal correlate of the
Although any attempt to answer this
ound to remain a provisional guess,
one proposed by Lietaert Peerbolte

ics of the dual cosmos in Vedic

of Heaven are (1) "later born,"
the gods of the Underworld; (2) more
Inteilrgeut than their elder brothers. The latter are nearer to the
primeval wodd of undifferentiated unity, which world was
onaracter,zec. by n and sleep (cf. Vrtra sleeping upon the
waters), wi~ere&s t e v m are "on this side of the Creation"
.129.aj)s Now, one of the first things to develop in the
, afxer life has been "awakened" in it by fertilization, is
ow and the brain. It is a well-known fact that the
uch faster in the first months than the rest
ot imply, it is true, that consciousness a t
alized in the brain. On the contrary, if
pi"e~*ataldream analysis which point to the exist-
a .hculty of (necessrcrily extrasensory)
ite of their admittedly revolutionary
tted &hatthe "viewpoint" of the per-
ousiiess ie, %& first remarkably floating.
em oa' the spermletozoon approaching
zhe orum q3pt.m~to have been recorded as seen from a point out-
s h e both th, ovum and the spermatozooiz, the nonindividualized
~olzsoio~~slzess stul hoveriag, as it were, above the scene. On the
ome2 nai,cb, tmm recordings, if accepted as such, must, although
~ r s s e r vm
e ~the dtibco~~scious mind, be considered to form part of
a pmson\ caonsoloszliiebs ir, s b widest sense of the word, which in
being ne tmd to the brain.
mere working h-ypothesis,
 that this myth "somehow" reflects the embryonic
human being, the inevitable conclusion would
be t h t the development of the newly conceived individual's
psyche, which development runs parallel to that of the embryo,
is characterized by a fundamental dualism, the brain functioning
"seat" of the individual's consciousness, whereas at the
nd of the axis psychic components of a lower, say phylo-
enetic, stratum might be localized.
Whether or not indications for such an early dual organization
n actually be found is a question that can only be answered by
the study of dreams concerning the conception and the first stage
of embryonic development. However, entirely speculative though
this idea admittedly is, it would seem to 6nd some support in the
curious mystic anatomy on which the theory of the Yoga practice
based. The doctrine of the six caKa85fJ an serpent
n$dinj: which, sleeping at the lowest point (the m),has
to be awakened and to rise along the spine for the yogin to find
final deliverance, can only be understood as the product of some
intuitive knowledge and the experience of countless generations.
This mystic physiology, although obviously at variance with the
scientific reality, was (and is) apparently effective for the purposes
of the yogin. Eliade observes57 that the yogins "maitrisaient une
zone infiniment plus vaste que la zone psychique 'normale,' qu'ils
p6n6traient dam les profondeurs de l'inconscient, et savaient
'r6veiller' les couches archdques de la conscience primordiale,
fossilis6es chez les autres humains." What he here, from the point
of view of the Yoga, characterizes as a fossilized psychic layer
corresponds pretty well with the primitive, possibly phylogenetic,
8tratum whose existmce was assumed above on the basis of the
cosmogonic myth.
If, then, on the lines of the working hypothesis here proposed,
one considers the possibility that some elements of the physiology
of the Yoga originated in images of the subconscious mind which
reflected the earliest stages of the embryonic development, some
striking parallels can be pointed out in other religions. Prom
Taoism the "embryonic breathing" as a means of returning to the
embryonic state and getting one's life renewed is well known,58
and Eliade rightly equated the three "breath-exercises"
For this doctrine, me 0.M. Rinze, "Die sieben Lotusblumen des Kundalini-
Yoga als Darstellung der archabehen Gesta].tastronomie,"in Symbolon, Jahrbuch
fa? S ~ ~ ~ O I ~ 5:180-219.
O T ~ C ~ U ~ ,
the context of this theory must, however, be passed over because
they are too complex to be considered within the compass of this
study. Thus it might be asked if there is any typological connection
between the shaman's mounting the world axis to attain to Heaven
and the rising of Kundalini along the spine. Attention may also be
drawn to representations in Indian sculptural art in which a tree
is seen rising from the upright body of a god or a human being.70
I n connection with the Rigvedic riddle of the two birds, one of
which eats the berry while the other looks on without eating, it
may be pointed out that in India (Pailcatantra V. 14) and else-
where the symbol of one bird with two heads is also met with. I n
the mythology of the Indians of Guiana (South America) the chief
king-vulture-spirit has, according to the Makuchis and Taulipangs,
two heads and one of these is the "devourer." See C. W. de Goeje,77
who suggests that this might allude to the two-sided character of
the bird: "heavenly spirit and earthly devourer."
It was assumed above that the "eating" bird of the Vedic myth
was the moon and on account of the equations "Agd =Sun" and
"Soma =Moon" on the one hand, and "Haoma =left" in Old Iran
on the other hand (the left eye of an animal being sacrificed to
Waoma, Yasna 11.4), the position of the moon on the left hand of
the axis could be accounted for. If, however, we start from the
wider set of equations "Agni =Sun =day =Upperworld" versus
"Soma =Moon = night =Underworld," it may be expected that
Upperworld and Underworld are not only situated a t the top and
the bottom of the world axis but also, on a horizontal plane, to the
right and the left of it. This is, indeed, what is found in the Indian
myth which relates how Prajiipati, the impersonation of the un-
divided cosmic totality, created the heavenly gods (Devas) with
(or on) his right hand, and the Underworld gods (Asuras) with (or
on) his left one.78 The same pattern recurs elsewhere, for example
in the Javanese shadow-play (wajang), where the performer
(&lang) shows the shadow of the World-tree (kekajon) in the
center of the screen, while the noble heroes are always shown on
his right hand but the demoniacal characters on his left. For other
instances I refer you to Bosch.79
76 See F. D. K. Bosch, Th4 Golden G e m (The Hague, 1960), pp. 201-3 with
plate 70; and cf. Eliadc, Traitb, p. 243.
77 Internationales Archiv far Ethnographic 44 (1943):55.
78 Cf. MaUr. Samh. 1.9.3 ( W . 132. line 16):Kcith. Samh. IX.11 h. 112. lines 18 ff.1.
7 9 The @olden @ern,p. 230. he same motik two heads f w h g each other a i d
two eyes which occur twice on either side of a tree symbol, are found in a painting
of Mrs. Evans, "an authentic naif, ran undisturbed and breath-takingly gifted
primitive" (Newmeek, August 4,1969, p. 37, where the painting is reproduced).In
may perhaps be underatood rts a regression toward the point
where the dualism arose, that is, toward the conception.

15. The ideas sketched in the preceding pages h& been developed
as a working hypothesis long before the first volume of Stricker's
work, De Geboorte van Horus, wrts published.82 The more striking,
therefore, is the partial parallelism between his approach and the
one here advocated. As Stricker remarks (pp. 11, 139, 155), it is
common knowledge that in the ancient Egyptian religion embry-
ogony and cosmogony are one and the same thing. The main
object of his amply documented work (which is to be comple
by two more volumes) is to demonstrate the constant parallelism
in all religions of the ancient Near East and Greece between
embryologicd and cosmogonicaJ ideas. It need not be stressed that
his material strongly supports the h t of the two suppositions on
which this paper is based, that is, that the cosmogony was con-
ceived as a macrocosmic conception, a fertilization of the World-
egg. Since Stricker also includes Vedic material, it may incidentally
be noted that with regard to the V$ra myth, which he explains as
a birth myth (p. 44), the Vedic evidence, when taken in its entirety,
would rather seem to point to a different interpretation (see above,
Section 9).
On the other hand, there is a fundamental difference between
his work and the present paper with regard to the explanation
proposed to account for the parallelism between embryogony and
cosmogony. Stricker quite understandably starts from a more or
less objective knowledge about the semen and the female organs
of procreation. Real knowledge about the organs of the human
female, however, was, even among the Greek physicians, ad-
mittedly very slight, as seen in Stricker (pp. 39, 57). On the other
hand, the working hypothesis of an anamnesis advocated in the
present paper presupposes not so much a knowledge about the
embryonic state in general as rather the possibility of reliving one's
own prenatal life. I n this respect there is an unmistakable paral-
lelism between the ancient religious practices which led a person
back to his embryonic state and induced him consciously to start
again, and on the other hand, a modern psychoanalytic treatment.
Both aim a t a catharsis which may result in a religious experience.
8% Stricker, 1:9-86 (we also MededeZingen en Verhandelinqen uan l& VOM-
aziathch-EgyHch Ueenootsohap "Ex Orknte Lux," vol. 14 [Leiden, 1963],2: 87-207
(we alao Mededelingen, vol. 17 [Leiden, 19681).
his fellows. Thie is
definite cultural patterns, so pre-eminently among the Indians."85
De Goeje, who was intimately acquainted with the medicine
men of the Amerindians of Surinam, on the one hand states that
"Indian ecstasy is certainly genuine," but then adds that "it
is doubtful, whether what is then revealed to them as dreams,
is always real first-hand knowledge, pure revelation of the world-
spirit. Clearly those d by the ideas current with
the respective tribe. probable that in the inter-
ams and phenomena in the outer world, phantasy
may play too active a part and often a medicine-man overrates
his psychic powers. Imposture may occur, but it is an exception."86
Anyway, it is the psychic mechanism which enables a seer to get
his personal "genuine" vision that is of prime interest in this con-
nection. Could its genuine character possibly be due to the fact
%hatthe seer "sees" his own conception. reflected in the mirror of
his culture? However that may be, this much is clear-the Vedio
poet, when speaking of the origin of the world, refers not to an
a1 knowledge but to a mystische Xchau.
dic poet's vision of Love as the primeval agent re-
minds us of a similar Orphic myth about the world-egg in the
Birds of rlristophanes, which 3ane Harrison renders as follows:87
I n the beginning of Things, black-winged Night
Into the bosom of E r e b o ~dark and deep
Laid a wind.-born egg, a d , as the seasons rolled,
Forth sprang Love, earning with wings of gold,
Like to the whirlings of wind, Love the Delight-
&c?. Love with Chaos in T&aros laid him to sleep;
And we, his children, nestled, fluttering there,
Till he led m forth to the light of the upper air!

Love is no* a notion of embry ,and the religious importance

of the cosmogonio myth can be expl~inedfrom a rational
man being's conception and prenatal life, even if
could be supposed to have existed in the remote
en these glyLhs must have arisen. In the ease of
is little evidence for su.ch a knowledge. There-
importance attached to the myth of the cosmic
unmistakable parallelism with the development
of Lhe embryo (Stricker) call a different explanation.

86 Ruth F. Benedict, "The Conmpt of the Gvardian-8pirit in North America,"

XemoiP8 of tL Americapt Anthrop Asso&t$on (1923),p. 41.
86 De Goeje (n. 71 above), p. 61.
87 Pmk~te92~( (n. 21 above), p. 626.
 ns from ths ea

in the circumstance t
vidual revelation8 are
to birth. The ensuing
 agree with Stricker that in the
macrocosmic projection of the
ard to the question as to how this pro-
about, however, we hold different views.
would seem incompatible with the
umption that it is on some primitive science, that is, on
nly allernative is that i t is based on
personal "revelations" of seers, which, from a psychological point
of view, must be considered to have consisted of images from their
subconscious minds which reproduced recordings of their prenatal
stab. This, again, presupposes the acceptability of the psychiatric
theory.
I n conclusion a rem a1 character may be added.
Ever since his early a Eliade has stressed the
importance in archaic ons of '1'6ternel retow," of .
man's desire periodical1 o the primeval origins, to live
again in illo tempme in order to bring about a total renewal of the
Cosmos.92 That the archaic cuh.wes regarded life more as a
cyclical than as a linear process and that the beginning of a new
year consequently h the character of a new creation had, it is
true, been known 10 beibre. The notion of a periodical renewal
and reinlegration o ciety and the Cosmos has also been shown
to exist in the Indonesian religions of the Ngaju Dayak and Tobe-
Batak. That in the Taoist embryonic breathing this regression
could, on the level of the individual, take the form of a return to
the embryonic state was also known. What Eliade particularly
stressed, however, was the negati haracter of life when viewed
as a linear process, the "Terror o
The modern "secularized" life of the Western world has no
counterpart to the archaic ceremonies which enabled man a t cer-
tain intervals to aboliah time and to experience his world as re-
newed. I n the modern world, as Eliade sees it, on the one hand
man lives in endless time. On the other hand there are indications
to show that in this same world the quest for a widened conscious-
ness is becoming of prime irkrest. It i s clear that.in both respects
the rediscovery of the prenatal world by modern psychiatry, if its
findings turn oaf;to be correct, will meen an immense widening of
man's inner world. If modem science confirms that it is possible to
recollect and reexperience the beginnings of one's existence, the
92 See, e.g., l'echniquea du Yoga, pp. 83, 234 8.
98 See, e.g., C o r n 8 and Hiato9y, the Mgth of the Itwernal Return (New York,
13/39), pp. 146, 163, p
(Editor's Footnote :First published in Hktory of Religionr, journal of the Department of the
History of Religion, University of Chicago, vol. 10, no. 2,1970.)
 I. Although the Vedic notions of a celestial bucket have more
an once been discussed, it ould seem that the real difficulties of
the problem have not been sufficiently realized. Equating it to
"den steinernen Himmelsbrunnen" l raises questions which remain
hile Luders in his Varuna interpreted the pail in the
ory of the celestial ocean, it cannot be overlooked
t his impressive life-work, as far as Alsdorf's devoted editorship
managed to restore it, obviously represents different stages in
the development of Luders's views on Vedic m y t h ~ l o g yOn
. ~ pp. 174
and 332, where Alsdorf gives Luders's last ideas on the "mountain",
one is bound to conclude that Liiders had here almost come to
realize t h a t t h e Vala was the primordial hill, even though he did
not apparently realize the full cosmogonical implications of his
cosmological views. As Alsdorf (p. 174 n. 2) remarks, it was obviously
impossible for him to conform the chapter on the V~tra-fightto this
new insight, which would have required a complete re-writing of
it. Anyway, it looks as if Liiders, at the end of his life, stood on the
threshold of a quite different conception of Vedic mythology. One
oi the most striking passages in this respect is found on p. 387 n. 5,
where he clearly recognizes the consequences of his new interpre-
on of the Vala myth. I t is not, however, the object of this brief
er critically to examine Luders's theory, including the silent
presuppositions of his approach. Lommel has rightly pointed to
its positivistic or rationalistic character; see DLZ 74 (1953)~cols. 402,
04, 405; Oriens 6 (1953)~p. 333. But then, a "naIve realism" was
characteristic of the older generations of philologists in general. I t
has, no doubt, much hampered the progress in this field.
2. The following observations are based upon-the assumption
that the speculations about the waters being a pratisthd of the
earth, and about the earth as a receptacle of these (subterraneous)
V i l l e b r a n d t , Yedische Mythologie Ie, p. 437 n. I and
- p. 3 2 2 "Brunnen
am Himmel". Cf. also Asia Major I ( 1 9 q ) ,p. 791.
Cf. Alsdorf, Yorwort des Nerazrsgebers, p. X V I I I : "zvir haben zwanzig,
dreiBig und rnehr Jahre alte Ausfiihrungen vor uns."
waters 1 reflect old cosmological ideas. I n later Indian and Indone-
sian art the Tree of Life is represented as rooted in this pdtm under
the earth.2 Indra's demiurgic act consisted in his opening the
primordial hill, which was considered either identical with, or
enveloping this re~eptacle.~ In the Rigveda the latter is sometimes
likened to a pail full of goods, e.g. IV.zo.6 "The frightful one pierces
[the Vala] as a solid enclosure crammed with goods like a pail (kdSu)
(full of) water".* From other religions the notion that the under-
world is such a receptacle of goods is well known.5

1 Cf. $B. VI.7. I .17 tcisyd ' p a evd pratisthd / apszi h'lme lokdh prdtisthitdh.
VI.8.2.2., XI1.5.2.14 dpo vd a s y a sdrvasya pratisthd, VI1.4.1.8 ydthd ha
vd i d d a puskaraparyim apsv ddhydhitam evdm iydrn apsv cidhydhitd, cf.
SRK. 1.1.1.6 (: SBM. 11.1.1.8) dtha h e y d p tdrhi pythivy [dyalte 'vdsa. ydthd
puskaraparndvz leldyid e v d A ha sma leldyati. tddz ha sma v d h h s d m vahatz,
(etc.), AB. 11.6.4, GB. 11. 3.2 dyaur antarikse pratisthitd, 'ntariksam pythi-
uydtn, pythivy apsv, dpah satye . . . See, e.g., S. LBvi, Doctrine d u sacrifice,
p. 159 n 5, Liiders, p. 120 n. 5, Kuiper, IIJ, 8, p. 107. For the earth as a
receptacle cf. MS. 1.4.10 (p. 59,5), KS. XXXII.7 (p. 26, 14) i y d h vd etds&n
pdtram. As for Liiders's objection that this theory is later than the main
part of the Rigveda ( V a r u n a , p. 121f.), see below, sect. 6 and in general,
Hist. Rel., 10 (1970), pp. 93-98.
2 See F. D. K. Bosch, T h e Golden Germ, A n Introduction to I n d i a n S y m -
bolism (Indo-Iranian Monographs, vol. 11), 1960, plate 83 (painting a t the
Joganmohan Palace, Mysore) and plates 27a-d, 28a-c, joa-d (the pzlrytaka-
laia), 66-67 (the gunungans of the Javanese wayang). Bosch explained the
"bowl of Plenty" as a secondary development of what he called the pad-
mamiila (pp. 110-113, 156f.). Otherwise, e.g., E. B. Havell, A Handbook of
I n d i a n Art (1920/1927), p. 43: "The vase forming the base of the pillar
stood for the cosmic waters", G. J . Held, T h e Mahabhar&ta (1935)~p. 209:
"this kalaqa might be the representation of the celestial ocean".
3 Cf. ddrim: bhid 1.62.3 = X.68.11, 1V.3.11, V.52.9, VI11.60.16, X.28.9,
45.6; d.r 1V.1.14; 16.8. Cf. adribhid V1.73.1. ds'mhnam: bhid 1V.16.6. Crvdm:
tyd v1.17.1 (G. gdvyam, cf. V.2g.12 gdvyarn cid C. . . . dpa vran), X.74.4
(ii. gdmantam, cf. IV. I. I 5 ; I 6.6 vrajdm gdmantam . . . v i vavruh). vrajdm :
dy VI.66.8, X.gg.1 I (IV.2o.6 ex coniectura). valdm: bhid 11.11.20, 15.3, 15.8,
111.34.10, VIII.14.7, X.62.2 [cf. 1.52.5 uulasya paridhz'n, X.68.10 valdsya
jdsum] ; dy 1.62.4, x.138.1. vytrdm: ava-bhid 11.1I .IS [cf. vytrdsya Sirah.
1.52.10, VIII.6.6, 76.21 girim: bhid Iv.17.3, X.89.7 (ndvam in nd kumbhdm).
pdrvatam: vi-bhid 1.85.10 [vi-v? V.32.11. Cf. khdni : t ~ d11.15.3 (nadindnz),
VII.82.3 ( a p d m ) . d t s a m : tyd IX.11o.g; dy V.32.1. avatdm: tyd 11.24.4.
* ddartd v d j r a y sthdviravz nd bhimd udniva kdSa?n vdsund ny@.tam. The
emendation vrajdm is generally accepted, cf. VI.66.8 sd vrajdm ddrtd,
X.gg.1I v r a j d q darayad.
5 Cf., e.g., W. B. Kristensen, De rijkdom der aarde in mythe elz cultus, in Vev-
zanzelde Bijdragen tot kennis der antieke Godsdiensten (1947)~pp. 291-314
[= MNAW N.R. 5 No. 12,19421. Kristensen discusses the origin of the symbol
of the pithos and refers to Gruppe, Griechische Mythologie, p. 816, Blinkenberg,
Hades' munding, in Det Kongelige Danske Videnskabernes Selskab Medde-
 The Wgvedic references to a orld crammed with go
or vciszlbhir nyfs%F) are rather equent. ~ h u X.108.7
s
"treasure-house, founded on a rock, full of cattle, horses (and)
goods", which is guarded by the Panis.1 These words (nidhdF, . . .
a'dribudhno), which obviously denote the Vala, remind us of the
koSo bh2nzibzldho in ChUp. 111.15.1, although here the term is
couched in cosmological speculation^.^ I t has, indeed, been argued
elsewhere, that in the Rigveda a'dri- is one of the terms for the
nether world and the primordial hill.8 At first sight, this might seem
at variance with divd . . . nidhinz in 1.130.3 "He found the treasure
of Heaven, laid down in a secret place like a bird's young, enveloped
in the rock, inside the endless rockH,4but Hillebrandt was probably
right in taking this "treasure of Heaven" as denoting the sun.6
Cf. 11.24.6 nidhim $an& m paramd& gziha' hita'm (in a reference to
the Vala-myth), VII.88.z sv&r ya'd &$man"the sun in the rock"
and X.68.7 (quoted by Hillebrandt, n. z), which I translate as
follows: "Re [viz. Byhaspati] himself drove upwards the (young) of
the mountain, the reddish (cows), as if (he drove out) the young of
a bird after having broken (their) egg-shells".' The verb zid ajrrti,

lelser, 2, 5 ( r g ~ g ) From
. the more recent literature I only quote E.A.S.
Butterworth, The Tree at the Navel of the Earth (Berlin 1970)~p. 101ff.:
"'The Lunar Crescent and the Bowl". '
I ay& nidhitr sarame ddribudhno gdbhir dSvebhir vdsubhir n y $ a h , rdksanti
tdm pa@yo ycl szcgopd(F) . . . See Hanns-Peter Schmidt, Byhaspati U . Indra,
pp. 186, 188, 240, who has contributed much to a better insight into the
nature of xhe Panis. nidhlh is not ,,der geraubte Schatz" (Liiders, p. 231,
in accordance with the traditional interpretation. He also holds the Panis
to dwell along a celestial river, p. 385 n. 2 ) .
"utarik;odara?~ koio bhiimibzcdhno jiryati, diSo h y asya sraktayo dyaur
a s y o t t a r h bilam. See Liiders, p. 81.
See IIJ. 8, p. 108 and Hist. Rel. 10 (1g70), p. 106ff.
dvindad divd nihitag gdhd lzidhig v& nd gdrbham pdrivitam dSmany
anante' antdv drimani. Eeldner's reference to dyzibhakta is obviously due to
Slyana's influence (ad IV. I . 18: diptiyuktav, devaih satnbhaktam vd), but
while "himmolgeschenkt" (rdtnam, IV. I. I 8) is possible, "von den Himmli-
schen geschenkt" (dhelzdva!~,1.73.6) is quite impossible. Liiders, p. 164,
renders "die vom Himmel zugeteilten" but p. 616 "die dem Himmel geho-
renden".
Asia &jor I (1924), p. 789f.
6 See IIJ 8 (1964))p. 11off. Hillebrandt, who rendered diman as "eine
Stein-, eine Eischale", has overlooked the cosmogonical implications of the
word.
@n&va bhittvd Sahundsya gdr6ham zid usriydh pdrvatasya tmdnd 'jat.
It would seem impossible to take as the subject of bhittvd either the gdrbha
(Geldner, H.-P. Schrnidt, .Byhaspati 74. Indva, p. 220) or their mother (Renou,
though commonly translated as "to drive out, expel", expresses
that the U&as are driven from the underworld: as their pen to the
upperworld (cf. Liiders, p. 125 on zid vahati). The words "treasure
of Heaven" in 1.13013, accordingly, may denote the sun while im-
prisoned in the primordial hill before it is delivered by Indra. I t
must be admitted, however, that in V.45.1 divd visylinn a'driy
zlkthaiF "opening the rock of Heaven with his verses" the addition
of divd cannot easily be exp1ained.l
3. If such words for "rock" like ddri and cis'man could denote
the underworld, a comparison of X.1o8.7 nidhih . . . cidribudh~zo
with 1.130.3 divd ~zihitam. . nidhim . . ant& ds'mani may give the
clue to the interpretation of IX.86.3 kdSam divd cidrinzdtarawt. Soma
is here addressed in the words: "Run thou, like a courser that is
urged on, to the prize, as a sun-finder to the pail of Heaven that
has the rock for its mother". The last epithet has often been dis-
c u s ~ e dI.t~would seem that, just as the Vala is "founded upon the
rock", so the pail is enveloped by it like a gdrbha by its m ~ t h e r . ~
As will be argued below, the term kdSa refers, rather than to Hille-
brandt's "Himmelsbrunnen" (see p. 144, n. I) or Luders's "Urquell am
Himmel",4 to a receptacle in the underworld. In that case, however,
the question will have to be answered how this could come to be
called a "pail of heaven".
P-

EVP. 15, p. 73). The correct translation was given by Liiders, p. 522. For
the gdrbha of the mountain, see 1V.1g.j and Liiders, pp. 171, 325ff.
"Herbei (kam) der den Fels des Hirnmels durch Wissen, durch Spriiche
auseinandergehen lassende (Dichter)", H.-P. Schmidt, Byhaspati U . Indra,
p. 175, who, however, does not discuss the problem of dzvd. In IIJ. 8 , p.
I r ~ f it
. was pointed out that, apart from 'this single passage, U ~ a is
s every-
where said to come from the nether world (as might be expected a prior;!).
Othenvise Liiders, p. 325, whose reconstruction of the cosmology resulted
in his statement "Die Morgenroten aber sind bei Nacht im Himmelsquell"
(P. 385 n. 2).
2 See Oldenberg, Noten on this passage and cf. Ludwig "des himels kufe,
die vom stein geborene", Grassmann "mm felsentsprossnen Himmelskrug",
Hillebrandt, Ved. Myth. P, p. 437 n.1 "den himmlischen Eimer, den stei-k
nernen Himmelsbrunnen". Liiders, p. 332 n. 2 took it to mean "da8 der
himmlische KoSa aus Stein gemacht ist", but on p. 380 n. 3 he followed
Geldner and interpreted ddri as "PreBsteine", which meaning he had reject-
ed in the preceding passage.
a "Vermischung der himmlischen und der irdischen Gegenstande" (Liiders,
p. 380), the k d b standing for the dronakalalia of the later ritual. Otherwise
Oldenberg, ZDMG. 62, p. 465 (Hdhavaniya) and Hillebrandt, Ved. Myth. 18,
P P 437, 478.
4 Varuna, p. 376 n. 12, cf. 380 n. 3 .'die Kufe im Hirnrnel", p. 383 (kdia,
= "Wasser, das am Himmel flieBt"), p. 285, etc.
 In the cosmogonical myth, after ndra had split open the primor-
dial hill (seep. 145, n. 3), the waters, in the shape of four rivers, flo
through the holes (khdni) in four directions over the earth1 Hence
the epithet d$rnavraja of the rivers in X.139.6 nadin.lzdm a'$Gvnzod
ddro a'$mavraj&Gm. hen an +ri is mentioned in this connection,
as in 1.73.6 pnrivdtah . . . vi sindhavalz samdyE ssasrz4r a'drinz "From
afar the rivers, together, flowed through the rock", there is (against
Liiders, pp. 164, 616) no reason to take it here in a sense different
from the current one. A reference to the primordial hill of the
cosmogony has long been recognized in AS. XV
milk from the pail, the jar with four orifices, as milch-cow, refresh-
men t rich in 'honey' (ma'dhu),for well-being". The kala'da cdtzcrbila
suggests a comparison with the Rigvedic "maiden who has four
braids" on which two birds have sat down.3 Here the via%is meant,
which is a well-known symbol for the earth.

1 See Liiders, p. 284ff.: "Die vier Himmelsstrome im Veda", and further,

e.g., B. H. Kapadia, Thefourfold Divisio~zofthe heavenly R i w r in the Pztrdpzs,
Pursnam 4 (1962), pp, 146-153, and for non-Indian parallels, e.g., Iars-
Ivar Ringborn, Paradisus terrestris, M y t , Bild och Verkligitet (1958), E. A. S.
Butterworth, The Tree at the Navel ofthe Earth (1970),pp. 8, 53, 116, 118.
"k6.ian-t dzcha~ztikalds'am cdtzwbilam i d d y dhinzqz mddhu
See my note in Rijdr. TLV 107 ( r g g ~ )p. , 79 n. 25. For cdtiwbila, cf. 11.15.3
vdjveyta khdny a t m a ~ .izadindm,
z etc. (above, p. 145, note 3). Earlier interpreta-
tions arenot known to me. Liiders, Varuna I ( I Q ~ I ) ,p. 285 (cf. 11, 1959, p. 381)
gave a somewhat different explanation of this "Vorstellung von den vier aus
dem Wasserbehiilter im hochsten Himmel kommenden Stromen" and
referred to IX.74.6 critasro ndbho nlhitd avd divd h a v h bharanty a m f t a p
ghytas'chtah, where he associates .~zikit&"(dort) angebrachte" with tyttye . . .
rdjasi in psda b. It would seem very doubtful, however, if avd divd, havir
bharanti can mean "tragen vom Himmel herab die ambrosische Spende"
(p. 285). In view of eV.40.6 (nadydh) av6 divd vrirtamdnh(h) "unter dem
Himmel" (Geldner),VIII.40.8 yd 1zz2 ivetdv av6 divd, uccdrdta(h) "unterhalb
des Himmels" (idem) and 1.163.6 avd divd "unterhalb des Himmels" it is
certainly more natural to construe nihith avd divd "sind unterhalb des Him-
mels angebracht" (Geldner), whether or not this means the same thing as
in 1X.89.5 samdnh antdr dharzine ni~attdh.
a RS. X.114.3 cdtuskaPardd yuvat@ . . . tdsydqn supar?zd v.+sann ni sedatur
ydtra devd dadhirb bhGgadhe'yam (YV. cdtuhiikhandd)
1.c. Since it is beyond doubt that the verse refers to
5.1 with Caland's note, cf. Ludwig, Der R-tgveda V, p. 304, Hillebrandt,
Ved. Myth. 1" pp. 429, 431 n. 2, Geldner, ubersetzung d . ) , the epithet
is mostly explained as referring to the catugkond vedih (Ludwig, Geld-
ner). The use of -ka$arda- in the epithet would, however, be better under-
standable if the maiden is explaiued in a mythical, rather than ritual,
way, the braided hair standing for the four rivers that stream down from
the top of the mountain. The vedi mythically represents the whole earth,
cf., e.g., MS. 111.8.3 (96, 5 ) , ICS. X V.'$ (106, 13))XXV.6 (110, z), XXVIII.1
 A similar notion of a subterranean Soma-vessel accounts for
RS. VI.69.6, where the poet addresses Indra and V i ~ n uin words
which in fact only apply to V i ~ n ualone, viz. samudrh sthah kala'iab
s o n z a d h d m ~(cf. kaldid somadhdlzd in 2). They express the idea that
i11 the cosmic waters under the earth 1 there was a receptacle for the
a , the elixir of the watem2 Hence the term Kala'Sa could
also refer to the primeval waters themselves in the tale of the
Amytamanlhalza, cf. Mhbh. 1.15.2 crit.ed. mathyatdm knlaiodudhi~
"churn ye the ocean, which is the ( a m ~ t a -jar".3
) Since the Tree of
Life was rooted in this very somadhdlza the conception of the
"Soma-yielding ASvattha" (nhatf/zasontasuvana) in ChUp. VEII.5.3
may have been based upon the idea that through the stem of the
world-tree the Soma rose up from its subterranean receptacle to
the third heaven, where the gods r e ~ e l l e dThieme,
.~ Untersuchzl?zgen
zzlr Worikz.de and Auslegtmg des Rigveda, p. 68, rightly concludes
that only in post-Rigvedic times this tree came to be thought of as
a celestial one. Originally the ASvattha was identical with the
cosmic axis and as such rooted in the earth.
Like the kdia, the dtsa ("well") from which the waters sprang
forth (Liiders, p. 384ff.) was located under the earth. Cf., e.g.,
V.32.1 "Thou splitst the well open, thou sentst forth the (water
from the) orifices, thou setst to rest the flood (arnavd) that was
hemmed in. When thou, 0 Indra, uncoveredst the big mountain,
(152, g), XXXI.10 (12, 8), XXIX.8 (177, 8), XxXV11.16 (97, 3),TS. VI.2.4.5,
SB. 1.2.5.7, 1.3.3.9, 111.7.2.1. If
the yuvati, accordingly, here stands for the
primordial hill, the two birds can be equated to the wellknown two birds on
the cosmic tree in RS. 1.164.20 (lastly discussed in Hist. Rel. 10, 1970,
p 127).
For samudrd = dpah see p. 145, note I and cf. IQJ. 8, p. 125.
Macdonell, Vedic Mythology, p. 107 and cf. MS. 1V.1.g (11, I) awzitarfi
vd dpo, GB. 11.1.3 amytam dpap, TS. 1.7.5.3, SBM. XI.5.4.5 dpo 'mjtam.
Geldner took kaldia in the sense of "belly" (cf. kdva?tdha-) but this is im-
possible. For Vi~nu'sconnection with Soma cf. IL22.1 and VIII.3.8, 12.16.
Cf. G. J. Held, The Mahabharata, p. 209. Later poets did no longer
understand this compound and replaced it by such terms as ksiravdridhih,
249.14 mathyat&vz ksiravdridhi?~,20 ~ z a t h i t uksi-
~
ksirascigarah, 57 mathyatdtp wathyatdp sindhz~h.
See above, p. 145, note 2.
Cf. AS. V.4.3 divattho devasddanas tyttyasyciwt itd d i d , RS. X.135.1
ydsmin vrksG supalds'e' devaih sampibate yamd?~and cf. Paul Thieme, Unter-
suchungen zt4r Wortkunde und Auslegung des Rigveda, p. 67. Perhaps the
Ilya vrksa of Kaus. Up. 1.5 should be connected with these passages. See
Thieme, Wissenschaftliche Zeitschr, der Martin-Luther Universitat Halle-
Wittenberg, Jahrg. I (1951-52), Heft 3 (Cesellschafts- und sprachwissen-
schaftliche Reihe Nr. I), p. ~ z f f .
thou sentst forth the streams (dhdrd) and slewest the son of the
D?inuW.lAs was pointed out above, also Liiders ended by consider-
ing the possible identity of "der Himmelsquell im Felsen" with the
Vala (p. 387, nn. 4 and 5).
4. The same generative and life-promoting water, however, was
also believed to fall down from heaven as rain. This raises the
fundamental question as to how this water was thought to ascend
to heaven, a difficulty that t o my knowledge has never been stated.
This is particularly manifest in stanzas like 11.24.4 "The well that
has a rock for orifice and streams of "9nad3tzd', (the well) which
Brahmanaspati split open with force, that all "sun-seers" have
drunk; together they have poured down abundantly the spring of
at er".^ There can be no doubt that this n'Swz8sya avatd is identical
with the Vala (e.g., Liiders, p. 387 n. 5).
The answer to this question is twofold: I) In the nocturnal aspect
of the cosmos the cosmic waters form the night-time sky and are,
accordingly, automatically above the earth. 2 ) In the day-time
aspect the water is drawn from the spring in the earth up to the
sky and then poured down. There was perhaps an additional idea
that rain could fall down from the branches of the ASvattha sonza-
savana, which reached into the third heaven (AS. V.4.3) and sup-
ported the heaven like a pillar. This possibility will not be considered
here.
As for the night aspect-the only case for which the notion of a
celestial ocean would seem firmly established-, it is not necessary
to enter here into an ample discussion after my remarks in IIJ. 8
(1964)~pp. 107ff. I t may only be observed tha't the reconstruction
there given of Varuna dwelling a t the lower end of the world axis
and supporting the Tree of Life is fully confirmed by later sources
like Mhbh. V.96.6 crit. ed., where N5rada and MBtali, on their way
to Varuna, descend into the underworld (avagzhya tat0 bhGmim),
cidardar Gtsam dsfjo v( khdwi tvdm arnavdn badbadhZnd& avam@,
mahdntam indra pdrvatam vi ydd vdh syj6 v i dhdrd dva ddnuvd?p han. Cf.,
:.g., 1X.11o.5 abh y-dbhi h6 Svdvasd tatdvdithd 'tsa?p nd k d w cij j a n a p L a m
aksitam.
a dSmdsyam avatdm brdhma?zas pdtir mddhudhciram nbhi ydm djasd 'tmat,
t d m evd viSve papire svardfio bahzi sdkdl;t sisicw &am udrd1;~.m.Cf. Hille-
brandt, Asia Major I (1924),p. 791, Liiders, p. 387. Geldner's translation
"Sie schopften allesamt reichlich von dem wasserspendenden Quell" is
obviously wrong.
Cf. PB. XIX.7.1 asuvdndq vai valas tarnas6 pv6vyto 'Sm6pidhdnai cdsit.
For aSmdpidh8na cf. RS. IV.28.5 ddardvtam dpihitciny dind.
where they find in the very centre (V.97.1 nbbhisthbne) the nbguloha
and its capital PBtBla. The same location under the ;SIeru, that is,
at the lower end of the world axis, is also found in Kuvalnyamdi,
par. 139 (p. 71, line 15, I1J I, p. 237 n. 6) and in SByana's commen-
tary on SB. 1.4.1.37 (Weber's ed., p. 111). While S e ~ asupports the
world axis from below (V.ro1.2, TT11.6g.q8), he and Taksaka
surround Varuna (11.9.8). In the same way S e ~ aand VBsuki are
located at the lower end of the jarjara during the piirvaranga of a.
dramatic performance (Bhar. K$ 1.94, 3.80).
All this is pretty well in agreement with the Rigveda, where
Varuna is said to dwell "at the origin of the rivers, with seven sis-
ters, (himself) amidst (them)", to be "the creator of the primeval
world" where "his seven" are. This world is referred to as "Varuna's
fixed dwelling-place [= in the centre], he governs the sevenJ' and "
in the most characteristic passage it is said: "He has created the
first creation, who with his pillar kept both worlds asunder, as Aja
(supports) the sky".l
In the light of this evidence it can hardly be doubted that in the
much-disputed stanza RS. 1.24.7, in spite of the use of the word
~ t h p aVaruna
,~ holds the roots of the inverted cosmic tree upwards,
in accordance with firdlmavdn of the later tradition (TA. 1.11.5,
Kath.Up. VI.1, Maitri U p VI.4, BhagavadgitB 15.1). The idea that
a nyagrodha could have been meant3, although still maintained as

1 Cf. RS. V111.41.~ yd& siwUu%ldrn zipodnyk sdfitasvasd sci ~ n a d l ~ y n 4 ~ t d

(see Liiclers, p. 712)) 41.4 sd m d t ~piiruydwz padci~p tdd ?~dlW?zasyasdptyal-?z
sci hi go#& ivh,yo, 41.9 ucirutznsya dhrmh?z scida!~ S$ sapt&zdm irujyati, 41 . ~ o
sd dhdma pz2rvyciwz m a m e yd skambhe'na vi ~ d d a s ?ajd nci d y & m cidhDaynt.
nbudhne' r&jE udrzt~zoz ' d ~ a s y o'rdhocipt stzjpavz dadnte pCtcidaksnt2, +zicbzd
s t h u ~zipdri bz4dhmi esdm asme' a~ztdrd h i t d h keicilia!~ syutz. For s t k z ~ has
"genereller Injunktiv" see in general Karl Hoffmann, Der Injzc+zktiv inz
Vedn, p. 113ff. For abzcdhne' Renou, EVP. 7, p. 72, points to VIJI.77.5
abztdhne'su rdjasszt. As for st&a, both the later traditio~land the words
updri bztdhnd e ~ i i m(which must refer to the roots) point to the conclusion
that the traditional renderings (Ludwig: "stamm", Grassmann: "Schopf",
Hillebrandt and Geldner: "Krone", Renou, EVP. 5, p. 94: "aigrette")
cannot be correct. Only the circun~stancethat the poet here expressed a
mystery can account for the use of this word instead of mrjla-, which had
already found acceptance : III.30.17 sahdmzlla, and further X.87.10 ?-m%,
X.87.19 sahdmCrn, VI1.104.24, X.87.2, 14 mdmdooa.
Geldner, Vedische Studiew I (1889), pp. 113-115 and cbersctzung I,
p. 25 n., L. von Schroeder, Festgrziss E. K w h ~(1916), pp. ggff., A. Cooma.
raswamy, Quarterly Journal of the Mythic Society 29 (193S), pp. "1-149,
F. D. K. Bosch, T h e Golden Germ ( I Q ~ o )p., 65ff., R. Otto, e.g., I'a~u?zn-
Hyvnnelz des Rigveda ( I G ~ Sp. ) , 25 n. 82.
late as 1960 by Renou, EVP. 8, p. 72f., is no longer discussib1e.l
The idea of an inverted tree is well known from many rnythol~gies.~ So
is also the identification of nocturnal sky and cosmic waters.Vf
this sky represents the underworld in an upside-down position, the
bowl a t the foot of the world-tree (see,p. 145, n. 2) must then be
inverted and hang over its roots.
Leaving aside the old problem of the possible identity of this
howl and the moon * (which is, in any case, not a problem of naive
realism but one of mythical equivalence) I can here confine myself
to a reference to RS. V.85.3 "Yaruna has poured out the cask,
with its rim turned downwards, over heaven and earth and the
intermediate space. Thereby the king sprinkles the soil of the whole
world, as the rain (sprinkles) the b a ~ l e y " That
. ~ the word nicc%wabZ~a
here refers to the inverted cask on top of the world axis in the
nocturnal sky may be argued on general mythological grounds.
Whether, however, -4s. 1V.15.12 dva &ir a$& syja, words with
which Varur~ais addressed, must in an analogous way he interpreted
as "let flow the waters that are hanging downwards" is questionable:
RS. V.83.7 does not plead in favour of this translation. On the other
hand, our interpretation of V.85.3 may furnish the definitive ex-
planation of the so-called "bell-shaped" or "lotus" capital of Indian
art as a cosmic symb01.~
5. More important in the Rigveda is the idea that one or more
See in particular Paul Thieme, Untevsitch?~ngenzzw Il'o~tkztnde zmd
A mlegzmg des Rigueda (Hallische Monographien 7, 1919),p. 68ff. and further
my remarks in Bijdr. TLV 107 ( ~ g j ~ pp.
) ,70-83.
Cf. e.g. Uno Holmberg (Harva), Der B a z m des L e b e m (Annales Acade-
miae Scientiarurn Fennicae, series B, vol. S V I ) , Helsinki 1922-23, pp. 17,
59, etc., H. Bergema, De Root>z des Levens i n Schrift e n Histolie (Hilversum,
1938), p. 275 n. 116, E. Iiagarow, Der zmgekehrte Scl~a1?za?zeo7,ba1rnz, in
Archiv fiir Religions\vissenschaft 1929, p. 183f., M. Eliade, Tvaite' d'ltistoire
des Religions (1953). p. 239ff. and p. 281 (references), L e chawzanisoze (1951)~
p. 244ff. (references on p. 245 n. I ) , E. A. S. Butternorth, T h e Tgpee of L i f e
at the Navel of the Earth, pp. 16, 66.
Cf., e.g., W. B. Kristensen, Leuen ziit d e ~ zDood, p. 14f., Synzbool e n
Wevkelijkheid,p. 180, Verzamelde R i j d m g e n tot kennis der antieke Godsdimsten,
p "79.
I. Cf., e.g., Oldenberg, Kleine Schviften, p. 703, Hillebrandt, V e d . M y t h .
12,pp. 326, 386f., Butterworth, T h e Tree of L i f e , p. 101ff.
?zicinnbiirayl ucir?i?znF kriuandhnwt p ~ dsnsarja vddasi nntliviksam, te'nn
ais'z~asyaBhliz.aozasyn r 2 j ~yci?la*iz 91d vy$ir v y z'lnatti bhdma.
For the origin of the bell-shaped capital see E. Havell, A Handbook
o f I n d i a n A r t (1920/1927),p. 41, G. J. Held, T h e Mahabharata (1935),p. 208,
F. D. K. Bosch, T h e Goldevt Gernz (1960), p. 156f., Butterworth, T h e Tree
of L i f e , pp. 104 (Sumer), 109.
gods draw water from a souree and then tilt the bucket. This is
clearly expressed in the Parjanya-hymn V.83.8 c d "Draw thou
(zid acZ!) the large bucket, pour it out. Let the rivulets, set free,
run forwardH.l The two stages are most clearly distinguished in
1.85.10 and 11: QrdZiv&z nunztdre 'vat& td djasZ dZdyh@ttn'r(t cid
bibhidur v i $n'rvatam "They [the Maruts: pushed the source upward
with force: they pierced the mountain although it was strong" and
jihvz&z nzmudre 'vatn'vz tdyd diid 'siGcann l i t s a y gdtanzZya t?s?zdje
"They tilted the well in that direction ( ? ) (and) poured out the
spring for the thirsty Gotama". Liiders, p. 386 n. I , who took the two
expressions as synonyms, misinterpreted zZrdltv&n m d - as "nach
oben stol3en", "umdrehen".
In 1.88.4 it is the Gotamas who, apparently in a rain magic, push
up the receptacle: "For days they [the Maruts!, greedy, have been
circling round about this rain-making divine poem. The Gotamas,
formulating their magic incantation, pushed up the pail $7 means
of their songs, for drinki~lg".~ The only object of the priests was to
cause the water to ascend to heaven. Thereafter, it was the task of
the Maruts to pour down the rain "for drinlting". The word jihmd
in jilimdbdra was apparently a stereotyped term for the tilting of
the pail, cf. 1.116.9 ptir2 ' v a t h ~~ttisatyii'~ludetlidnzz~ccdbzdmagz
cakmtZmr jihmdbdmm. "You, 0 Siisatyas, have pushed away the
source, you made its bottom turned upv.ards and its brim tilted".4
There is, it seems, 1-10 appreciable ~liffer~ncr~ in meaning with nTcina-
bdra in VIII.72.10 sidcn'nti nhrznstl '7~irtiliz~ccdcnkrampcirijm&zam,
l nzahdnta;gz kds'awz lid acd ?zi siiicn s)..d/lilailthi!~izzdyk vi;itd!l plwristdt.
The relation of the preceding stanza 7c dftiiil szi kni*:ca visita;z nycificnvz
t o the one quoted is not clear. Liiders, p. 381, is probably right in taking
diti ancl kdia as two different objects. See further, e.g., Kaegi, Hlttldert
Lieder, p. 97, Hillebrandt, Liedev azts dem Rgveda, p. 70, ICarl Hoffmann,
Dev Injzmktiv ~ P Z Veda, p. 261, etc.
dkdni gtdhrd!~$dry k zw dgzw i d p d h i y a y ~ ~ Z v k d ~ y ca
d p dt e ~ h n Irr(ih~mz
,
k,~yvd?ztogdtamiso aukaiv ~1rdlzvdyn z ~ n z i d mz<tsndhimpibadhyai.
The meaning "brim" for -bdra according to Gelclner; see further ITaclcer-
nagel-Debrunner, K Z . 67 (1942), p. 171 c=IVaclcernagel, Xleine Schriften,
P 3871.
There remains a crux in YIII.40. j y& snptdbudli?znw~a v n a v d y jihmn'Ddrmn
apousz~td(!z)"(Indra and hgni) who open the stream \vith seven bottoms
and with tilted brim", where, apart from the mythical speculations under-
lying sapdbridhna,the notion of opening the source in the mountain would
seem incompatible with that of tilting a bucket (which, moreover, is not
Indra's task, cf. VIII.771.S). The predilection for compounds with snpt$-
in this group of hymns should be noted, cf. 39.8 saptdnzdnzqa, 41.2 saptdsvasd,
g sd saptdnhnz irajyati.
nicinabdmm dksitani "They [the Zzotw] pour out with obeisance the
inexhaustible source that goes round ( ? ) with its bottom upwards
(and) its rim downwards".
Like Parjanya, the Maruts are said to draw the rain-water, cf.
RS. V.55.5 zid hayathd ??zarztta?zsamudratd yzkya'?h v.r$i& varsayathd
pzwi;i?zah, AS. W.1g. j zid irayata mzrutatz samudmtn's, IV.zy.4
ap& samz&& divam zid vahanti, KauS.S.III.3 indrapras'i$H varu.li?a-
praszktd dpah sanzztdrHd divanz ad valzanti (see Liiders, p. 104f.).
The term samudra' was used in Vedic times both for the oceans that
surrounded the earth in the mythical cosmology and for the cosmic
waters under the earth, as is shown by the phrase lzfdya samudrd
(IIJ 8, p. 125). I n its last connotation it was synonymous with
avatn'. The cosmological conception on which these passages are
based is, accordingly, quite the same as that underlying the Varuna-
hymn V.85.3, quoted above, and Luders (p. 719) was wrong in
assigning them to "der spaterm, mehr naturalistisch gerichteten
Anschauung iiber die Regenentstehung", however understandable
this conclusion was on the basis of his premises. The principal
difference between the Marut-passages and the Varuna-hymn is that
in the latter we are concerned with the nocturnal sky, which was
identical with the cosmic waters. That is why 7'arui>a did not need
to draw the water before causing the rain to fall down on earth:
the kn'va~tdhawas, indeed, immediately at his disposal. This idea is
espressed with all desirable clearness in 1.161.14 "Through the sky
go the &ruts, on earth goes Agni, the JYind goes here through the
air, 17aruna goes through the waters, the satrzz4dras (adblzir yHfi
vdru !l all s a m z ~ d m i ? ~ ) " .
Geldner has been the first to recognize that the verb d cydvayati,
apart from its being used in the general sense "to draw towards
oneself", was also a technical term for the drawing of the (subterra-
neous) water. ,4s such it is more currently used than zid acati, and
is more unequivocal than t e r m like dzilz (e.g., VIII. 1.10 kdvandham
ztdri?zaln dzdz~hrepjs'nayo), which will, therefore, here be left out of
account. The term d cyiivayati has been curiously misunderstood:
Luclwig rendered it by "beschleunigen, sturzen, her(an)eilen
machen", Grassmann: "ausgiessen, ergiessen, betvegen, herziehen"
( Worterbzdz : "ausschutteln, ausgiessen") , BIas Muller, SBE
l cczkvn "den untern rand cles gefaszes", Ludwig, v01 IT-, p 3 8 7 I t seems
here to be a synonym of b w l l i d . although in the late hymn X.1o1.7 avatriw
d i m n c a k m m looks rather l~lcea variation of I1 24 4 ri&zEsyanz aliatcim
XXXII: "to shake, shake down", Liiders (p. 380): "herabfallen
machen", Renou, EVP. 10, pp. 29, 36: "amener, attirer vers-nous-
le-mettant-en-branle" (but cf. p. 93 at the bottom!). Cf. also
Hillebrandt, Asia Major I (1924)~pp. 789-791, who curiously dis-
regarded the specific meaning of d in d cyhvayati, which differen-
tiates it from cyhvayati and prn' cyEvayati. Since, as a result, the
stanzas concerned have mostly been misunderstood, they are here
cited with a translation. For considerations of space all comments
have been left aside.
VIII.72.8 d dadn'blzir vivdsvata i ~ z d m hkddanz acucyavit, klz6dayZ
trivftii divdh "With the ten (fingers ? Geldner) of Vivasvat hdra has
pulled up the heavenly bucket, with a threefold hammer (?)".
V.53.6 d yd& ndrah suddltavo dadZdDse divn'h kddam d c ~ c y a v u h v, i
parjn'nyak s ~ j a n trddasi
i dnu dhn'lzva.izh yanti vfitn'yah "The heavenly
bucket which the heroes, rich in water,l have pulled up for their
worshipper, t h a " c r ~ a r j a n y a " they pour out over the two worlds;
the rains are going over the dry soil."
V.59.8 dczlcyavztr divyn'm kddanz et& j p r ~ d r n ' s y amarDto gpzdndfz
"These Maruts, (sons) of Rudra, have pulled up, 0 seer, the heavenly
bucket while they were praised."
X.42.2 kddanz wd piir?zdq vdszlnli n y f & a m d cyzvaya maghade'yliya
Shram 'Draw the hero j= Indra] like a full bucket, crammed with
goods,2 for the giving of presents."
IV.17.16 gavydnta indram sakhydya viprZ advliydnto v i ; a $ a ~
viijdyanta&,ja.rzCydnto janiddm n'ksitotiwz d cyHvaydnzo 'vate' nd kddawz
"Desiring cattle we, the priests, draw Indra towards us for an
alliance, desiring horses and the prize (him), the bull ; desiring wives
(him), the giver of wives, as (we pull up) a bucket in the well."
In the last two stanzas it is quite clear that "to pull up (a bucket)"
is only a specialized use of "to draw towards oneself" in, e.g.,
.92.7 d cylivayasy &!dye. Cf. IV.32.18

6. In conclusion it may be slated that when the Maruts are said

to milk the DCsa (1.64.6, VIII.7.16), or when Indra causes the udder
of the mountain to flow (V.p.z), the Dtsa and hdhar refer primarily
to the cosmic water in Varupa's realm under the earth, where, a t

1 For ddnzt.see the references in Hist. Rel. 10, p. 122 n. 61. Cf. 1.79.3
ddlzur usmd &par&pi?zvnte divdh (Liiders, p. 121 n. 2).
Geldner's translation "den mit Gut beladenen Ilelden" is obviously
wrong. Cf. IV.28.6, X.108.7, 11.16.7, IX.g7.44, AS. IX.4.3.
the lower end of the world axis, the amyta is located as late as
Mahabharata V.97. crked. The circumstance that these waters
are only mentioned the most recent parts of the Rigveda (Luders,
p. I Z I ~ . ) is due to the fact that the family collections
concerned with this aspect. When this htsa is calle
dering" (I.64.6), its water is meant after it h been drawn up to
heaven. For the same reason the hdhar
referred to as &dhar divya'm (IX.1o7.5,
(e.g., 1.64.5). See also Luders, p. 38gff.
If the preceding interpretation is correct, it must be concluded
that (apart from the identity of Varuna's nocturnal iky with the
cosmic waters) these passages do not confirm the theory of a celestial
ocean.% However, so long as the fundamental problem of the
ythological (not cosmological!) meaning of the "third heaven",
and the background of its partial identity with the primeval or
nether world is not entirely clarified, no final judgment on the
"celestial oceanJ' in the Rigveda would seem possible.
Leiden.

Cf., e.g., Bergaigne, La religion vidique, passim (see Index, 111, p. 348))
Macdonell, Vedic Mythology, p. 85, W . Neisser, Z u m Worte~buchdes Rgueda
I (1gz4), pp. 57-59, H. Lommel, Asiatics, p. 408f., Liiders, p. r11ff. "Der
Himmelsozean" (but see Karl Hoffmann, OLZ. 1954, col. 391).

-
(Editor's Footnote : First Published in India Maior: Congratuhtoty Volume presented to I
Gonda, Leiden, 1972.)
 l. There is a considerable difference of opinion nowadays about the
social and cultural background of the Rigvedic poetry. In Geldner's
gveda-translation and in Renou's recent studies which are based on it1
there is a tendency to overemphasize the importance of literary contests,
for which the poems are thought to be designed. Thieme, in a funda-
mental discussion of this trend in modern Vedic studies, not only noted
"an unmistakable tendency to secularize the RV", but also stated to
"hear in some renderings of Geldner's overtones of their own that call
to mind unfortunate associations with the emberg master-singers
and the minnesingers' tournament of song on the
following studies, devoted to a social as well as religious phenomenon,
may contribute to correcting the perspective and to eliminating some
views about the Vedic society that are still materially based on t
theories of the Vedische Studien. The influence of these theories is indee
still perceptible in Geldner's latest interpretation of the Rigveda.
In point of fact, the duels between poets may rather be regarded as a
special instance of a more general type of contest, which included un-
poetical verbal contests as well as chariot races, combats, etc. Here
however a serious methodological difficulty faces us. The question
naturally arises on which occasions these contests may have taken place.
Thieme, who is disinclined to accept the theory of such contests at all,
stresses the necessity of looking for a serious, genuinely religious content
in the Rigvedic hymns.8 I quite agree with him but, while the contests

l Karl Friedrich Geldner, Der Rig-Veda aus &m Sanskrit ins Deutsche r&rsetzt
(Harvard Oriental Series, vols. 33-36, 1951-1957); L.
pdqiniennes, I (Paris 1955),p. 1 ff.
a JAOS, 77 (1957),pp. 53 and 56. Cf. also H. P.Skhmidt
a Ibid., p. 53.
from Earth, by which act a cosmic dualism of upper world and nether
world (represented by Devas and Asuras-Diinavas respectively) is con-
stituted. To Varupa, the ancient god of the waters of Chaos, a ne
function is now assigned as guardian of the cosmic 1
remains hidden in the nether WO
this process of differentiation, the
itself as a nether world in oppo
continues to be the "older" worl
the younger dynasty of the heav
(p, 148,16 ff.) calls Cnujdvaratara- "younger brothers" (read am-?).
this process of creation (as we
standing its predominantly demiurgical character) the goods of life,
symbolized by Agni and Soma, were for the first time brought to light
ther world. According to Vedic mythology, accordingly,
er have been released from the power of Inertia, the cosmic
regressive force, that lay on the hill (girt-,pcirvata-) and kept the
of the nether world closed (see p. 249). This myth is appar
inheritance from the primitive Indo-Iranian religion. A different version
is found in the archaic myth of the Churning of the Ocean, according to
which the goods have been won directly from the depths of the Ocean

pion of time as a cyclical process is also met

with in India. This implies that the beginning of every new year
conceived as a new cosmic start and, consequently, that at the end of
every year the cosmos returned to its point of departure, the undxeren-
tiated state of the Chaos, to be reborn. Direct indications for this last
conception, it is true, are not found in Vedic literature: from the South
Indian Pongal festival, during which the undifferentiated state of the
cosmos is imitated and re-enacted in social life by a temporary abolish-
ment of all social difference^,^ and from parallels that could be adduce

See the references in J. J. Trilogie altindischer Machte und Feste der Vegeta-
tion, 111, p. 122. The time at the new year was celebrated seems to have varied
in later periods. See Macdonell-Keith, Vedic Index, 11, p. 157; Hillebrandt, Ved.
Mythologie, I%,p. 31, lIe, p. 177; Caland's note on his translation of PB.V.9.1, H. P.
Schmidt, Vrata, p. 36 n. 46; Heesterman, The ancient Indian Royal Consecreation, p. 7f.
or further references see ially C. d'onofrio. Studi e
 ans, but this is a hypothesis.

g of the Sun.8 He is there-

the sun", svardti-

the sun". On the

other hand, the priests are said to be longing for the sun (svarydvab.. .
viprrtb.. kuiikdsab 111. 30.20), men invoke Indra's aid and are helped in
races '6which have the sun as their prize'7, e.g. I. 130.8 indrab sanuitsu
ydjam-nam &yam prdvad.. .sviirmilhegv rfji,w VIII. 68.5 svdrmilhe+vuydm
ndrab / nhc2 hdvanta Btdye, I. 63.6 ( t v h . . .) sv&mi/he ndra rtjd havante.
Now we know from the Vgjapeya ritual that chariot races can have a
ritual characterQnd in the passages quoted we have no reason to take
with Geldner the word cSji- in sense "Kampf" instead of "race",
which is its normal meaning. a1 races are also suggested by such
phrases as IX. 65.11 svardtiam / hinvt v u vrtjiizam: just as the chariot
of the gods is called "sun-finding" (VII. svarvidrt vcfsumatrt rdthena),
so the horse that is the winner in the race can be considered a "sun-
finder". In the Avesta the same (or nearly the same) epithet is also
attributed to Yima, who was hvara.daras6 maJyrtnqm (Y. 9.4). The many
mafzgalavaciritsi for "race-horse" in the Rigvedic language should be
noted in this comection.l0
What was the religious significance of these "races that had the sun as
their prize", of this "acquiring of the sun"? The poets tell us that Indra's
primordial act was continuously repeated. Cf. e.g. VI. 17.8 ddha
viive purd indra devd tkariz tavhai.6 &&ire bhdrrtya / &vo ydd a
azShi+vtadevh svdr,sdtG ta indrarn dtra "Da stellten dich allein, den
Starken, alle Glitter an die Spitze, urn (den Sieg) zu gewinnen, o Indra;
als der Ungott sich uber die Gotter uberhob, da erwahlen sie im Kampf

Religioni, 24/25 (1953/1954), p. 141 f. (cf. Lornmel, Der arische Kriegsgott, pp. 51-70),
for the data about ancient Iran cf. e.g Religionens Varld, pp. 201-209,
Numen, I, p. 37 ff. (etc.), Tavadia, jEin tus bei den Zoroastriern in Iran,
AR W, 36 (1939), pp. 256-276.
Cf. Hillebrandt, Ved. Myth., 11, p. 182, n. 1.
See Weber, Sitzungsber. Berl. Akad., 39 (1892), p. 787 ff.; Hillebrandt, Ved. Myth.,
I, p. 484; Hwterman, op. C., p. 127 ff.
Auspicious terms for the race horse are &ri-, vdjin- (cf. vdjino vijajito v&a&
also cftya- (lit. "by-passing", cf. tdd dhdvato 'nyclrr cfty
.II. 12g6,city asti any& RS W . 1.14), and vycfti- (cf.
vy-oti-gm-, etc.).
 the first two pddas relate a nlythicaf
fourth refer to a recent assault of the
demon (aorist) and the new decision of the gods (present tense), no
that the sun must be won again. This, as well as the races for sun-
winning, cannot refer but to one definite period of the year, viz. the winter
solstice. Cf. p. 269 and n. 7. If this is true, the question must be raised
whether also other expressions of the Vedic language may perhaps have
a far more concrete meaning than Geldner's translation attributes to
them. The poets frequently pray for deliverance from anxiety (cirizhas-).
Now arhhasaspati- is the name of an intercalary month in VS. 7.30,21.31,
which allows us to infer that cin3has- could refer to the end
The association of cirizhas- and t as- "darkness" (see p. 23
the same conclusion. Only in passing can attention be drawn to other
words that possibly refer to the same period, e.g. p a r i v ~ t s a r d ,rdtri
~~
pdritakmyii,12pdrya-,l3 etc.
If from the facts mentioned we draw the general conclusion that some
at least of the hymns to Indra concern the critical period of transition
from the old to the new year, and some at least of the chariot races (e.g.
the svdrmtlha- ciji-) must have taken place at that time, we must assume
that men tried to assist Indra in his fight against th and Darkness by
their ritual. Now similar rites are known from th ahiivrata-ceremony,
at which an Arya (VaiSya) and a Siidra had to fight a ritual combat over
a white piece of leather representing the sun, and from the ceremonial
buying of the Soma (Somakrci-yap) as a preliminary to the
sacrifice. In such circumstances the powers of the nether worl.
personified by a certain social group, the ~iidras,while the dryas acted
as the representatives of Indra and the heavenly Gods.
Now one of the technical words for Indra's divine gifts is maghci-, e.g.
IV. 17.8 hbntii yd vytrbriz sdnitotci vdjapiz ddtd maghd~imaghbvd surddh@
"der Toter des Vrtra und der Gewinner der Beute ist, der Gaben schenkt,
der Freigebige, Belohnende". On the other hand maghcivan-, Indra'
current epithet, is also a title given to liberal patrons, e.g. X. 81.6 ih
'smdkam maglzdvii stirir astu. This suggests an equation of Indra and
these lords, who in the social sphere may have personified him in his
function of present-giving (maghcitti-). Hence, when a lord gives a portion

Hillebrandt, op. C.,I, p. 30, n. 3.

lZ Hillebrandt, op. C.,11, p. 253 f.; Renou, Etudes sur le vocabulaire da RV, 1, p.
42 f. ; etc.
ISSee e.g. Mayrhofer, Kurzgef. etym. Wiirterb., U[, p. 258; Abaev, ktinzologiCeskij

slovar' osetinskogojazyka, I, p. 422.

An ethnological Study (thesis Leiden, 1935), p.
if it was too dangerous-in this
Death, then the ominous God, and
victory (RS. X. 124.5), could only be spoken of in
terns during this Lime.la Cf.

3. U ~ a and
s the
fundamental questions for our understanding of the charac-
gveda is, how we must conceive the nature of the Goddess
Dawn (USUS). Following Ludwig, Hill
that the twenty hymns to Usas do not
day but especially the first dawn of the new year.17 A similar view, based
on a comparison with the Old Ostara and the Lettish Uhsing,
was taken by von Schroeder.18 olars however have rejected it.
eith objects that Hillebrandt holly unsupported by any
evidence, and depends on a t
is arbitrary".lB Oldenberg points to the lack of clear references to the
beginning of the new year and controverts Hillebrandt's arguments
based on the use of U~as-hymnsto in
and on a possible connection of U ~ a s
night, out of which both Indra and are said to be born;
II. 10.12).20 Foy combated Willebrandt's view that R
ujahrslied" and his arguments ere accepted by Olde
Renou.S1 Also Renou does not see sufficient indications for H
theory: "ceci reste une hyp~thkse".~
ow, every theory about Usas must needs be a hypothesis.
we are confronted with the real methodological difficulty, which
has never been stated explicitly. en reading Foy's important criticisms
one soon recognizes that the basic difference of opinion is ultimately
rooted in a different approach to the Vedic hy
en we try to defend again Hillebrandt's theory, this is mainly because
the current view, though seemingly prompted by common sense, fails to
lB See IZJ, IIT, p. 211.
l7 See Ludwig, Ilgveda, IV, p. , M, p. 173 a; ISiUebrandt, Ved. Myihologie, Is,
p. 28 ff.
l8 L. von Schroeder, Arische Religion, 11, pp. 16, 58 f, 205.
l0 A. B. Keith, Religion and Philosophy of the Veda, p. 121 f. NiUebrandt's theory
was only accepted by G . Montesi, Studi e Materiali di Storia delle Religioni,
(1957), p. 26 (to which M. M016 kindly drew my attention after this study had
written).
Oldenberg, Religion des Veda, p. 243 n. 1.
Oldenberg, Noten; Renou, Etudes vkdiques et pcinindennes,
Renou, Et. vkd. etpcip., LU, pp. 6, 52, 101.
 en not as a piece of
resuppositions from which
illustrated by the words
te", which once slipped from Oldenberg's
y did not overlook the diitlculties; rather

ost prominent feature is the prayer for wealth and

d seem far from self- t that the dawn of a
lored to bestow wealth fact that, apart from
ds given in the early m~rning,"~ no serious attempt
for this remarkable association of the dawn
ably due to the implicit conviction that
,which induced the Vedic poets
alth and progeny. As a result of
this conviction any attempt to understand the poets' religious attitude
towards their gods must seem useless and runs the risk of being denounced
as "rec~nstructionsal6at0ires".~ On the other hand, it cannot be denied
that the light of Usas "n7est autre que la reprise de la lumibre originelle".26
it is our conviction that the true nature of Usas cannot be really
ood unless we take the cosmogonical aspect into consideration.
n the following pages we shall point out some aspects of Uws which
seem to us not sufficiently explained by the hypothesis that the dawn of
be addressed in the way the gvedic poets address Usas.
Our investigation on the principle that the hymns as religious
poetry deserve to seriously and that, whatever may have been
the role of the poe nique, it is our task to understand a
those associations the Vedic poet were fundamental.
nical problems as the connection of the
tardyapla. Cf. VII. 76.2 prd me phinth8
dnthah y h te svd itaro devaydndt, AB. IV. 14.5

. 107.1, Thieme, Der Fremdling im RV., p. 27 (but see now

255, n. 36 and cf. e.g. JB. E. 130).
 yo vai sarftvatsarasyd 'varodhanad codrodhanain ca veda, sa vai svasti
sahvatsarasya pdram as'nute, etc.z8

A. Cosmic aspects
The importance of Usas in the cosmic process is apparent from the
fact that the two representatives of the opposite powers of light and
darkness (or upper and nether world) are u ~ h and - ntikta- (U~dsdndkilf,
Ndkto~dsd). Still the role of Usas in the cosmogony is hardly ever
mentioned. There is a reference to a (first?) Ups, through whom the
gods arranged the works(?) of the Sbhus (IV. 51.6 kvri svid &lfk katam '
puriipi y6yB vi&&d vidadhzir rbhags'm), and to the early awakening of
Usas prior to the whole creation in I. 123.2 pdrvti vis'vasmdd bhlivanlfd
abodhi jdyaizti vdjam bfiatcf sdnutrl 'Triiher als die ganze
Hohe erwacht, siegend, den Preis erringend". Renou compares tiplirvyci
I. 46.1. Her role as a jhyantf deserves notice (see below).
Usas is born from the darkness of the nether world.
taken in themselves, might seem simply to refer to the
but the frequent references in others to the "firm rock" (i.e. the primordial
hill opened by Indra) suggest that also in the first group the birth of
Usas was conceived as a reiteration of the cosrnogonical process. Cf. I.
123.1 k ~ d lid d asthcrB aryd vihdyzih, 9 s'ukrd kr~pddajani~tas'vitlccf
(otherwise 1. 113.1 citrdh praketd aja fa vibhvd), IV. 51.1 thmaso
vayrindvad asthdt. While she arises, the is must keep sleeping in this
darknessz7, cf. IV. 51.3 acitrt ant& ah sasantv dbudhyamdn8s
tcimaso vimadhye, I. 124.10 dbudhyamlf~zdfzpaphyah sasantu. This is not
contradicted by the passages which state Usas to come from the sky, e.g.
I. 49.1 Qo bhadrtbhir i diva cid rocancfd a'dhi), or to be the daughter
of Heaven (dulzits' div ssim, VII. 75.1 divijch, VI. 65.1 divojdh). The
most fundamental misunderstandings about Vedic mythology have
arisen from the fact that the mythical identity of the nocturnal sky with
- the nether world has scarcely ever been clearly stat .28 The parallelism
between, e.g., I. 48.15 &so ydd adyd bhBntind v i v gtivo rdivdh and
IV. 51.2 vy jr vrajdsya tdmaso dvdro 'chdntlr avra ayah pavakcfh (cf.
111. 5.1), IV. 52.6 vy &varjydti8d tdmah, VII. 75.1 hpa drtihas trima dvar
djq?am, VII. 79.4 v i dr&i.sya dliro ddrer aurgob is not accidental: it
rests on the identity of dvhd divdh and vrajdsya dvdrii, dtiro cidreh.
a V l l e b r a n d t , Ved. Myth., Ia, p. 33. Rigvedic references to the winter solstice are
according to Hillebrandt I. 61.15, V. 29.5, X. 171.4; 179.2 ("But none of these passages
are conclusive", Vedic Index, U[, p.
a7 "t&&bres&temeUes",Renou, op. C.,m, p. 73.
P"lee Bijdagen Kon. Instituut, 107 (1951), p. 81 f.
 orld (that is, the subterranean waters and the nocturnal
s realm and here is the seat of @h, which is "hidden
hey unharness the horses of the
of Varupa (I. 123.5), is
and is said to awa e from Pta's abode, cf. iV. 51.8 @hya devfh s
budhdnd gdvcirfz n6 u j h o jarante. This ne
primordial hill, the firm rock which Indra has opened. So Indra
to have engendered U ~ a and s the Sun, cf. 11. 12.7 y6h sibyah yci
indrah soryajiid wcisah svdr janat,
etc.29 As in the parallel cases of Agni and Soma, however, the victorious
ercomes that of Darkness is also assigned to Usas herself,
...scinutriI. 123.2. She, who dwells on the surface of the
nu- VI. 65.5), has opened the doors of the firm rock:
VII. 79.4 vi drfhcisya ddro 6drer awqob, 5.7 rujdd drfhdni
rj,dgdm. Cf. also IV. 51.2 cisthur u c wcisah purhtcin mi
svciravo 'dhvark,su / vy & 'chcintir avran" chdcayah
pdvakdb. A different WO n is i i r ~ d - .Hence
~ ~ the
fathers are said to have found the hidden light and to
together at the tirvd-; cf. VII. 76.

pelling of darkness
t d m h . Therefore
is said to open' in I. 113.4 bhhvati nedrf sGny'tcin13m
vah must be the dziro bdrer (VII. 79.3, the vrajhya
var jydtisci t d m b ) and
cf. I. 121.4 (Indra) h. Only once U ~ a sis
ove, I. 48.15), which
reference to the nocturnal sky also accounts for Usas being called duhit
d i v e . This interpretation, which can here only be indicated in passing
8.7, where U p s is said to have
s denotes the nether
parcvcitah shyasyo 'dbyancid lidhi. A similar interpretation applies to
the stone house (harmyd-) from which the light of U ~ a scomes to this
 orld: VII. 76.2

Great Goddess (mahf111.

will be discussed below.

. Darkness and the Goods of t $ e

The appearance of Usas

goods of Life from the b

associated her appeara
ultimately due to her

addressed in the words

treasures, thou keeps
priraritdhir h u vratcini
the doors of the fi
 e cosmic struggle of light and life against

Srayante, V.SSt4 ydd u,sa azichab

ince this trait is SO frequently mentioned, it must have been regarded

o we are driven to the con-
e U ~ a here
s addressed can scarcely have been the dawn of
eject Nillebrandt7stheory as
" have dismissed this argu-
s succession of days can be
called the first: the conclusion that the U g ~ saddressed must be the first
of a new period. seems unavoidable. Indeed, the poets state explicitly
that this Uws is the beginning not only of a single day, but of a long
series: U ~ a sbrings the days (plural!). Cf. VII. 77.2 gciv&n m-td netry
a hmyd-, properly "related to the waves of the subterranean waters", hence
"darkness"? See Neisser, Zum Worterb. des RV, I, p. 182; Renou, Et. vkd.,m,
p. 86 f.
 , 76.6 grivdni netrf, TS. V . 3.1 1p rttlndm pdtniprathame''yyrim dgdd
nezrf janitri prajdndm. As the dawn of the first day Usas is also
know its name (I. 123.9 jdnaty dhnah prathamdsya ndma iukrd
k r ~ n d dajani~taSviticf, where the aorist shows that the dawn of this first
day has just appeared). Whether a sirnjlar idea prevails in VII. 80.1,
where the Vasisthas are said to have awakened first at the appearance of
Usas is doubtful.
The cosmogonical prototype of Usas as the first of many is referred
to in TS. IV. 3.1 1a (etc.) iydm evd sd yd prathamd vyatichad, (m) yrhya
gdrbhah prathamd vy@ti~yapdm kka mahimdnam bibharti, (n) yd prathamd
vyatichat sd dhenzir abhavad yamk (cf. AthS. 111. 10.1). In the Rigveda
she is also called agriyd (see below). I'ostly Usas is regarded as re-
presenting the whole series which she inaugurates, and sometimes, when
there is a general reference to the u,&ah, she is obviously included in
their number. Inversely, the plural seems to denote especially the first one
in I. 92.1 etd u tyd u~cisahketlim akrata ptirve drdhe rdjaso bhdntim
aiijate. Note also TS. IV. 3.1 1f trihidt svdsdra tipa yanti ni~kytcidi
samdndih kettim pratinlun"cdmdniih, where the dawns of a month seem to
be taken as one group. Sometimes, however, Usas is opposed to the
great mass of ordinary days, e.g. I. 123.11 bhadd tvdm u ~ vitardmo vy
dcha nd tdt te any&zqdso naianta. Renou's remark "L'U. actuelle comme
privilkgike, selon la norrne bien connue"32 fails to appreciate her im-
portance as a netry dhndm fully.
A new succession of happy days is apparently inaugurated in I. 124.9
dsdm pdrvdsdm cihasu svds!@dm dpard pdrvdm abhy kti paicdt / tdh
pratnavdn ndvyasir nandm asmk revdd uchantu sudind u~dsah. The same
hymn ends with the following words (13) dstodhvariz stomyd brcihmagci
me' 'vivydhadhvam uiatl'r ujdsah / yupndkam devir dvasd sanema saha-
srigam ca Satinam ca vdjam. In contrast with the other hymns to Usas,
RS. IV. 51 is entirely addressed to the Usgsah, which only that
what is said about the first Dawn is also essentially true, though to a
less degree, of all the following dawns. Cf. also I. 92.1 prdti gdvd 'rwir
yanti mdtdrah "es kehren die roten Kiihe wieder, die Miitter".
We are drawn to the same conclusion by the frequent use of the word
iigra-, which also suggests that the beginning of a new period was
celebrated. Usas, who is the foremost (agriyd X. 95.2), distributes the
first goods (I. 123.4 dgram-agram id bhajate vciszkdm; but Geldner: "das
Allerbeste von allen guten Dingen", similarly Renou), she goes in front
of the other dawns (VII. 80.2 dgra eti yuvatir dhraycigd prdcikitat sdryah
" Etudes vkdiques, DI,p. 60.
 agnim), she inaugurates the sac&ce (VI. 65.2 dgrariz yajfihya
brhatd ndyantir). It is natural but significant that as a netry d h d m she
t the beginning of the days, cf. V. 80.2
ath& kpvatf ydty dgre / brhadrathd
jydtir yachaty dgre dhnrlm. The same phrase recurs in V. 1.4 ydd im
sdvcite a,dsrS viriipe hetd vciji jdyate dgre dhmZm, 5 jcinigta hi jhnyo dgre
'
2, X. 110.4). Equal1 cant is the fact that A
is said to wait for the beginning of the D to shine at that begin
IV. 13.1 prdty agnir ugdsii &gramakhyad, VII. 8.1 dgnir dgre
9.3 citrdbhcinur u&sdrit bhaty dgre VII. 68.9 dgre budha
cinmd, V. 1.1 dgre byhdnn q h i m zirdhva' asthcln nirjaganv
5 vi bhdty dgra upiscim idham@.
at these references to the beginning of a new
oo explicit and too frequent to be ascribed to the wbims
of certain poets. If this is true, only Willebrandt's theory can account
for them. Oldenberg's objection that this interpretation of the Uws-
hymns is "vielmehr hineingetragen als aus ihnen herauslesbar" (Rel. des
Veda, p. 243 n. 1) disregards these facts. Deluded by "diese so durch-
sichtigen Texte" he has been blind to their real problems.

D. This particular day

th equal frequency the hymns stress the importance of the present
day, at which U ~ a sis invoked to appear. A period of darkness has
obviously come to a close: in I. 92.6 the poet says dtdrigma
prirdm asya "we have attained the end of this darkness". Is thi
to be the end of the last night? Elsewhere we read: "Arise! The living
spirit has come to us. Darkness went away, the light is coming. (The
ess) left its path for the sun to go. have gone (to the point)
men prolong their life-time" (I. 11 lid irdhvam jivd lfsur na
jydtir eti / cfraik pdntham ydtave S
. Again we must ask: is it likely from a psycho-
hat Vedic poets should have welcomed every new
day as a point where they prolonged their existence? Renou, Et. vdd. 111,
p. 52 reniarks on this stanza: "Surrection soudaine de la phrasCologie
c o m e 92,6 . . . qui pourtant ne doit pas nous obliger
allusion au dkbut de l'ann6e . . .: ceci reste une hypoth
objected that any day could be regarded as a special day
onne connue". This is true, but was every day actually
 9.2 tknci sus'rdvasani jdnamprtPvddyd duhitar divah, I. 92.14 zbo
adykhd gomaty ds'vcivati vibhcivari / revdd asmk vy dcha sCwtcivati, 15
yuksvd hi vcijinivaty ris'vcidi adydrupddi usah, I. 113.7 @o adykhd subhage
vy ircha, 12 ihddydjah s'rkjfhatamci vy dcha, 13 s'divat purdsd vy livcisa devy
dtho adykda'riz vy dvo maghdni, 17 adyd tdd ucha gmatk maghoni, I. 123.3
ycid adyd bhcigdk vibhdjdsi ntbhyah, 8 sadfs'ir adyd sadfs'ir id u s'vd dirghdk
sacante vdrupasya dhdma (see above, p. 226), 13 t@o no adyd suhdvci vy
itcha, IV. 51.4 kuvit sd devih sandy0 ndvo vci ydmo bablziiydd u,saso vo
adyd, V. 79.1 maht. no adyd bodhaydso rciyk divitmati, VI. 65.3 civo dhcita
vidhatt rdtnam adya', VIJ. 75.2 mahk no adyd suvitdya bodhi, 78.5 prdti
tvcidyd sumcinaso budhanta.
Other words that stress the present day are nzindm, nd. Cf. I. 12
ratnavdn ndvyasir niindm asmk revdd uchantu sudinci usdsah (c
G o adykhd . . . revdd asmk vy licha), I . 129.11 vi niindm uchcid
prd ketlir, I. 48.3 uvbsosd uchdc ca nli devf jird rdthcincim, IV. 51.1
nihdni divd duhitdro vibhcitb gcitzirit krpavann u ~ d s jdnciya,
o VII. 75.8 nd
no gdmad virdvad dhehi rdtnam. Especial emphasis is expressed in V I .
65.4-5, where the words kvo dhcita vidhatt. rdtnam adyd are varied with a
etition of the word idd "now": id& h i vo
dcis't@a ~ c i s a h/ idd v@rciyajdrate ydd ukth
vahathci purd cif. 5. h i ta u ~ adriscino
o gotrd gdvcim dngiraso grvdnti.
Renou, op. c. 111, p. 87 remarks : "Instant priviltgit soulignt par le pronom
dtictique". Cf. also p. 52: "L'idte gtnCrale est la m&me:exalter l'U.
d'aujourd'hui parmi l'ensemble des U., comme la divinitt invoqute est
exaltbe parmi I'ensemble des divinitts invocables". I do not think this
does full justice to the particular character of the passages cited.

E. As in former days
Sometimes the poet's prayer for the present day is accompanied by a
reference to former days: pnrvdthci I. 92.2, V . $0.6, pratnavdt I. 124.9,
VI. 65.6 (cf. V. 79.1). Though far from conclusive in themselves, these
words find their most natural explanation if they are taken to refer to
the beginning of former periods, rather than to the preceding days.

F. The new life

The poets pray for increase of their vital strength (VI.
vdyah krnuhi Sdcibhir), for prolongation of their earthly existence (X. 18.3
drbghiya hyuh pratardk dddhcinGh). Psychologically it is hardly con-
ceivable that such prayers should have been repeated at the beginning of
every new day. The fact that such prayers are constantly directed to
Ups, and that they are accompanied by those for progeneration is,
therefore, incompatible th the assumption that the everyday dawn is
meant. Renewal of life is no common everyday experi
very renewal that U ~ a is s expected to bring about, cf.
ndvyam cfyur drkhdna. Other passages are I. 48.10 viivasya hz
jfvanah tvk v i ydd uchdsi, I. 113.8 vyuchdntij k i m udirdyanty
kdn? can&bodhdyanti, 16 Eid irdhvah jivol &W no dgdd dpa
jybtir eti / haik pdnth ve sibydyd 'ganma ydtra
(d = VIII. 48.1 Id, Ath . 2.3&), 17 asmk dyur ni
vis'vqvdre, VII. 77-5 (asmt . . . vibhdhy)
irdvah stlribhyo amy'tak vasutvandk
s the parallel phrase lid irdhva~ jivd.risur na dgdd in st. 16
jivdm udirdyanti in I. 113.8 should not be confounded
with VII. 77.1 viiva?h jivdm prasuvdnti i: Geldner's translation of
I. 113.8 "im Aufieuchten alles, was lebt, auftreibend (aber) keinen Toten
mebr erweckend" is not quite correct. The erect position represents life,
and the victory of life over death. In Egypt the dead man is addressed
with the words: "Arise, for thou art living; arise, for thou art not dead!"s8
n quite the same manner U~as,as the personification of Vitality (S-
. 123.5 etc., cf. slinari I. 48.10), is said to stand erect: 111. 61.3 iirdhv
amttasya ketlih "thou standest rect as the symbol of Life" (see
low). Since Lommel has tried to emonstrate that the expression
tirdhvd- sthd- means "'mit fe bereit stehen" (221. 8, 1931, p. 270 B.) it
may be useful to dwell on this point. Several passages are not conclusive,
but the evidence as a whole allows us to state 1) that the notion of
standing erect was of special importance to the poets. Cf. IV. 6.1 iirdhvd
zi sli 00 adhvarasya hotar dgne ti;r#ha, 2 Grdhvdm bhdnlim savitkviiired, 3
4 Grdhvb adhvaryzir jujwdgb asthdt. Note the
ajd akrdb. 2) that the standing position was the
manifestation par exce f life. Cf. I. 172.3 ijrdhvdn nah karta
jivhe, I. 36.14 krdhf na cardthdya jivcise, I. 140.8 zirdhvds tasthur
mamr&ih prdydve phah. indirect indication that siiny'td- means
"vital strength" and nothing else is found in I. 134.1 zirdhvrl te cinu
siiny'td m h s tighatu (where 2rdh Id be connected with tislhatu.
Cf. with this stanza VIII. 45.12 divk-dive sahcisrd s22ny'td iatd /
jaritJbhyo vimdmhate, and fur tid iratdk siinJtd lit pliralitdhir
asthub, I. 48.2 Eid iraya prdti mE szTmy'td q a 4 col&
h these passages may further be compared 111.
pratici bhlivamini viivo 'rdhvd ti~lhasyamttasya
.B. Kristensen, Net ieven uit den dood, 1st ed., p. 103 ff.
 ketzih, where Usas as the symbol of life (amjta-) is said to stand erect,
and V. 1.2 tirdhvd agnih sumdndh prdtdr asthdt . . . m a h h devds tdmaso
nir amoci and X . 1.1 dgre byhdnn u&hBn tirdhvd asthdn nirjaganvdn
tdmaso jydti~dgdt,where Agni's arising from the realm of darkness, if
connected with the beginning of the new year, must have implied his
victory over death (cf. IV. 5 1.2, VI. 64.1). 3) that in the other passages
referring to Usas and Agni the same idea of a rebirth (I. 123.9 ajani~fa!)
may have been present in the mind of the poet, although this is not
expressly stated. Cf. 111. 55.14 (Usas) pddyd vaste pururkpd vdpzirhsy
iirdhvd tasthau trydvirh rtrihdpd, V . 80.5 e Subhrd nd tanvd viddnd
'rdhvCva sndtf drs'dye no asthdt, VI. 63.4 (A dhvd vdm agnir adhvartsv
asthdt prd rdtir etijtirpinighytdci, 11.35.9 dd d hy dsthdd upcistham
jihmdndm tirdhvd vidytitalli vcisdnah, IV. 6.1 (see above), X . 20.5 jusdd
mdnwasyo 'rdhvds tasthdv fbhvd yajEt, I. 164.10 tisrd mdtfs t r h
pitin bibhrad &ka tirdhvds tasthau n h dva gldpayanti. In VII. 39.1
iirdhvd- is combined with Sri-: tirdhvd agnih sumatini vcisvo as'ret praticf
jtirpir devdtdtim eti. Cf. finally VIII. 27.12, where Savitr is said to stand
erect: zid u syd vab savitd suprapitayd 'sthcid zirdhvd vdrepyah. 4) that our
previous conclusion that in the Rigvedic hymns, as referring to the New
Year festival, the idea of a renewal of life is closely connected with that of
present-giving and victory over opponents may account also for 111. 8.1,
where the tree etected as a flpa is addressed with the words: afijcinti tv
adhvart devaydnto vdnaspate mddhund daivyena / ydd iirdhvds ti##hd
drdvipehd dhattdd ydd vd kjdyo matt2 asyd uphthe and for the following
passages where the god is invoked for aid in contests: I. 30.6 zirdhvcis
tisfhd na zitdye 'smin v&e Satakrato, I. 36.13 tirdhvci ii p2 pa titdye tijfhci
devd nd savitd / Zzrdhvd vdjasya shitd yddaiijibhir vdghddbhir vihvdydmahe,
V I . 24.9 pr&d yandhi sutapdvan vdjdn sthd ii ,~ri iirdhvci zitf drisagyann
aktdr vykyfau pdritakmytiydm, VIII. 19.10 ydsya tvdm tirdhvd adhvardya
tisthasi ksayddvirah sd sddhate / sd lirvaehih shitti sd vipanyribhih sd
Adraih sdnitci kytdm. 5) that iirdhvd- sthd- is only twice explained by a
special place, viz. 11. 30.3 t-rrdhvd hy hthdd ddhy antdrikst 'dhd vytrdya
prri vadhdm jabhdra (the sole passage where Indra is said to stand erect;
it should be noted that Indra has slain V@raimmediately after his birth),
and IX. 85.12, X. 123.7 iirdhvd gandharvd ddhi ndke asthdd. In the fist
passage iirdhvd hy hthdd may be taken in the usual sense, which may be
said to characterize Life in its victory over Death.= No inference can be
drawn from AV. ara8wd hiStanta Yt. 13.76.

Cf. also Renou, Et. vid., III, p. 12, n. 2.

G. Life and Security
Numerous are the passages where Usas is said to have overcome the
darkness and its danger: in close association with tcimas- the hymns
,drlrrih-, dwitcf-, e.g. V. 80.5 cfpa
jydti$(f'giit, VII. 75.1 cfpa dmlh
nii viivii
da in jux
so is bhcfya-. It must be due to mere accident that we do not here meet
with the expression tamo bhayam (cf. e.g. ChU. I. 3.1 zrdyar;ts ta
i). But the opposite amy abhayam "life and security"
was present in the minds of the poets ;compare VII. 81.6
ftani vasutvancim (IV. 5.13 amy'tasya pcftnfb, 111. 61.3
I. 75.3 bhiinrivo . . .

rather point to a perio

new year put an end?

N. Progeneration
The clearest manifestation of vital strength to the Vedic poet was
progeneration, as well as material wealth. This is what he longs for as
the clearest proof of his vitality, and since U e s was considered a janitrf
she is implored to bestow
ni didihi prajdvat (cf. I. 13
the phrase rayim
suvfram), which is mostly translated by "
parallelism between material wealth and
of one's blessed state, it may be useful to point out that rayi- does not
exactly denote "wealth" in our sense of the word. In 11. 11.13 asmc!
rayiriz riisi vfrcfvantam VI. 65.6 suv rayim g.mt# rirfhi, etc., we have
clear instances of a etymologica rayim rci- (Indo-Ir. 'raHfm raH-),
.8. The Indian notion of wealth
blessing, a gift (cf. Lat. dfves:divus, Gr. ~ t 8 a i p v ) .
same association of vital strength with wealth (and social prestige)
e.g. ChU. 11. 11.2 (cf. 12.2-20.2) sa ya
veah, prcigf bhavati, sarvam d p r eti,
bhavati, mahan kirtya.
'si prajayii ca dhane
 rayfn? zzdate
/ s'is'ihf nab S
S: asya kule viro jiiyate (ChU. 111. 13.6).
one of the Vedic deities that are specially implored to
stow progeneration. A few quotations may suEce: I. 92.8 z&as td
as'ydn? yas'cisani suviram . . . rayim, 13 &as tdc citrdm d
vrfjinivati / ykna tokdriz ca trfnayariz ca dhrlmahe, I. 123.1
, IV. 51.10 ray& divo duhitaro vibhdtfh prajhvantariz yachatii
vih . . . suvfryasya piitayah . Once it is admitted that
Uws has very close connections with is not surprising that she
should be invoked to secure progeny. e religious importance of
the goddess Usas lies no doubt in her close association with the origin of
life in the cosmogonical process, and every reappearance of the dawn is
to some extent a re-enactment of this event, a rebirth of life itself. But
again we must put the question: Is it likely that every d a m should have
been invoked for life and offspring? Even texts which might suggest an
exuberant adoration of every new day still make a significant distinction
between the sun's udaya- and pratydyana-, cf. ChU. 111. 19.3 (in a
cosmogonical myth) atha yat tad ajdyata so 'sciv Kdityas, tan?j d y a m d m ~
gho~ciulfilavo 'niidati~thantsarvdpi ca bhtatiini sarve ca kdmds, tasmcit
tasyodayam prati, pratydyanaliz prati, gho~dulfilavo 'nfitti~{hanti sarvdgi ca
bhtltdnLsarve ca kdmdh (cf. however I. 11.7 sarvdni ha vcl imdni bhatdny
ddityam uccaih santani geyanti). The words pratyriycnariz prati might be
taken as an explanation pointing to the fact that every new sunrise is a
return (cf. RS. I. 123.12), but modern translations give the following
rendering: "A son lever et A chacun de ses retours" (Senart), "at its
rising and at its every return" (Hume), "a1 suo sorgere, a (ogni) suo
ritornare" (Papesso). If so, the udaya- must be different from both the
primordial birth and the everyday sunrise and can, it seems, only denote
the new year's day. Although the little use that is made of technical
terns for New Year remains unexplained, the prayers for progeneration
(including cattle, e.g. prajayd ca pakubhis' ca prajdyate KS.
p. 189,19) can only be accounted for, if Usas inaugurates a new year.
From RS. VI. 3.6 sd iliz rebhd nd pra'ti vastrr wrdh s'oci~iiriirapiti mitrci-
mahdh "We ein Barde ruft er bei jedem Aufgang der Morgenrote laut
mit seiner Fla~nrne''~~ it does not follow that the rebhd- (see p. 276)
praised the Dawn every morning.

For prdti vcisto see Oldenberg, ZDMG, 55, p. 284.

 dently closely connected with vital strength
aning of szlnjtfi- these concepts seem to
have fused to such an extent that it m y seem difficult to separate the
various aspects. A few quotations may illustrate the importance of
is besought to bestow: I. 92.8 &as t h
avargalit rayim risvabudhyam, I. 113.5 db
C 'him? pdrthivasya vcfsva @o adyCh&sub
ur . . . y h v IJfina iaJami
10 rayim divo duhitaro

colouring: it denotes the good

otions must have had their

orted to drink the Soma with

toter in der Anhaufung von Schatzen, o Held" the ritual act of invigorating
the god is closely associated with a reference to a present-giving, which
may be the god's annual renewal of life, but which may equally well be
an actual war or a ceremonial contest of the potlatch-type, in which
ealth was won (or lost). The mythological concept (X. 139.3 indro nd
tasthau samarC dha'nclirufm)may also cover its replica in social life, and the
vcfstZni m y be the goods won from the enemies that are the earthly
representatives of the gods of the nether world.

J. Uws as the goddess of contests (Potlatch?)

At this point we are faced with the most important aspect of the
worship of Ugas: she bestows wealth, but this wealth is also won in
strife, in contests which are fought with the traditional weapons of the
ns, viz. with races and word duels.
.I. 48, which starts with a praise of U p s as rdycf devi dhvatiin st. 1,
calls her jird rdthdncfm in st. 3, and depicts her in st. 6 as vi yd srjrfti
lit vy ctrthinah "who lets loose the contest, the competitors" (see
below). The prototype of the competing hero, god Indra, was at the
same time victorious by his speech and his horses:
vljpro cfrvadbhir hdntz vytrdm ntbhih $drab "er gewinnt
mit den Streitrossen; der Held erschlagt den Vora mit
ut although the ideal type of the sabhtya- - was not yet quite
rgotten in later times (see below, p. 265), th edic society shows L
further grade of specialization: the maghdvan- or s k i - here mostly secures
the assistance of one or more priestly poets, who fight this part of the
contest for him: VIII. 19.10 sd drvadbhih scinitlf sd vipanydbhih sd iikaih
sdnitd kytdm "der hat als annerbeherrscher Erfolg, der gewinnt mit
Rennpferden, der mit seinen Barden, der gewinnt den ersten Preis mit
seinen Tapferen". They form a party with common interests: if the
sakhyd- discussed below (p. 250) denotes this association of a siiri- with
his helpers, the co-operation of these sdkhdyah . 71.2,6) might even be
considered the ritual counterpart of the myt co-operation between
Indra and his ndrah, who are also called sdkhlfyah (I. 165.11). Anyway,
in many passages the word vaydm does not refer to the poets alone (and
thus rules out the idea of merely literary tournamen ut to the different
members of their party, e.g. when the poet says " win the vdjas
with our poems and race-horses" (dhibhir rirvadbhih VI. 45.12). As a
single instance we may quote IV. 51.11 vaydh sydma yaidso jdnegu.
Often, however, the poets make a clear distinction between themselves
and their patrons, e.g. VII. 78.5 maghdvcino vaydriz ca, I. 141.13 amf ca
y t maghdvdno vaydm ca, VI. 46.9 maghdvadbhyai ca mdhya~zca. Cf.
particularly VII. 81.6 irdvah skibhyo amftah vasutvandrit vdjam asmd-
bhyam gdmatah / codQyatrf maghdnah siiny'tlfvaty u,sd uchad dpa sridhah
"den Opferherren (bringe sie) unsterblichen Ruhm und
rinderreiche Gewinne. Die Lohnherren anspornend moge die freigebige
Usas die Unfalle wegleuchten", V. 86.6 (Indra-Agni) td stirip irdvo
brhdd rayim grndtsu didhrtam i ~ a hg.rpdtsu didhytam, 1. 124.1
revdd uchu maghdvadbhyo maghoni rev& stotrt siinrte jlfrdyanti, V. 79.
abki y t tvd vibhlfvari stdmair g~ndntivdhayah / maghair maghoni suiriyo
ddmanvantah surlftdyah stijlfte LiSvastin.rte, 6 a i ~ udha virdvad ydia &o
maghoni stirigu / y t no rddhdlizsy dhraylf maghdvcino drdsata slijate diva-
sllnrte, 7 tkbhyo dyuntndm byhdd ydia @o maghony d vaha / yd no rddhcirizsy
cis'vyd gavyd bhdjanta siirdyah s.a., VI. 65.6 uchd divo duhitah pratnavcin
no bharadvlfjavdd vidhatk maghoni / suvfralit rayik grpatt ririhy urugayirit
ddhi dhehi irdvo nab. These stanzas suggest the conclusion that the stress
formerly laid on the importance of the dakging as the poets' fee was
rather excessive. The sole fact that the patrons are denoted by the same
term maghdvan- which is characteristic of Indra (and Uvas) allows only
dependency of the poets, viz. on the dvans, and on the
divine maghdni), her task is to awake the earthly gvers (I. 124.10 prri
bodhayo 'jab pypatd maghoni, which admits of different interpretations;
similarly I. 48.2 cbda rddho maghdndm, IV. 51.3 rddhodciydyo '~ciro
maghdnih). Most important would be the consistent references to vdja-
in connection with U$asySB if only the exact meaning were beyond doubt.
Although originally a quaiity attributed to a horse, a mule, a ram, the
waters, etc.,"' its meaning in the Rigveda seems to be a more restricted
and technical one. It is on the one hand the victorious power which
bestows the gifts and incites the gods themselves: Usas is vdjaprasata-
(I. 92.8) and vzjini-, ti-, the god VSja incites the gifts (VS. 18.33
vdjo no adyd p , cf. 9.2. vdjasya p r a s a ~ i ) . On
~ the other
hand she wins th . I. 123.2 jdyanti vdjam brhati sdnutri, I.
48.1 1 ijo v&ariz hi vdmsva yh.4 citrd m wejdne, 16 (sdm . . . mimikpd . . .)
sdfi vdjair vtijinivati. As in the case t7ny'td-, however, Usas also bestows
these v&as on men, e.g. I. 48.12 sd 'smdsu dhd . . . @o vdjariz suvfryam, I.
92.7 prajdvato nyvrito cis'vabudhycin $0 gdagrcidz &a mcisi vdjrfn. The same
idea is expressed in the next stanza by the word rayi-: +as tdm ajytirh
yaidsah suvfrarh ddscipravargam rayim h.4vabudhyam / suddrizsasti Srdvasd
yd vibhdsi vdjapras~tdsubhage bficintam "Aurore, puisd-je atteindre la
richesse qui porte distinction, qui consiste en h o m e s utiles, cornmenqtnt
en esclaves, s'achevant en chevaux! Toi qui brilles d'un renom fait de
rCussite, 6 bienheureuse, toi qui es rnue par les prix de victoire, . . .
(cette) haute (richesse)!" (Renou). Renou and Gonda (op. C., p. 49 n. 68)
rightly reject the translation "booty". If so, what can have been this
vdja, which on the one hand was a quality, a sort of power, and on the
other hand consisted not only of horses an cows but also, apparently,
of slaves and renown (cf. VIII. 96.20 sd vdjasya Sravasyhsya ddtd, VI.
45.12 etc.)? Renou holds the central meaning to be "prix"; besides the
word may perhaps have denoted the "tournoi". If however a.manYsvdja
was a quality, which manifested itself' through wealth and which was
won through the acquirement of this wealth (cf. I. 124.13 yupn
devir civasd sanema sahasrinarh ca Satinariz ca vhjam) the fundamental
question is: Where was this wealth won, if ordinary war is ruled out?
The terminology of the Rigveda does not allow us to determine, to what
extent mock battles and ceremonial combats may have played a role in

For references see Renou, Et. &d., m[, p. 21.

Gonda, Aspects of early Yi$@uism,p.
Geldner, note ad I. 77.4.
l
l

VI. 28.6, see p. 27

the winner of such

seem more appropriate than "rendez-vous" (Renou),

eda are strongly reminiscent of the winter ritual of the North

1
"i
1
l
l
l

i
i
1

won. Chariot races were equally important and that is, why not only
the horses but also the chariots themselves were said to be Sravasyli-
ory". Cf. IX. 96.16 abhi vdjaliz sciptir iva iravasyd 'bhi
a, V . 37.3 dsya Sravasytid rcitha cf ca gho~at,V .
vayciliz Sravasydm d huvlimahe. Perhaps this
light on the very obscure stanza I. 48.3 uvho 'pd uchrfc
thtincim / ye' asyli dc&rage&fdadhrire samudrt nd Srava-
syhvah. Geldner translates: "die Wagen in Bewegung setzend, die auf
1 ihr Kommen gewartet haben wie die auf Ruhrnestaten ausgehenden
(Seefahrer) auf die Meeresflut", Renou (-Oldenberg): "dkesse animatrice
l des chars. Ceux qui, A ses approches, se tiennent prets, tels des gens
l avides de renorn (qui s'appretent pour un voyage) en mer, . . .". The
I chief difEculty is, indeed, samudre' nci since the interpretations put forward
are unacceptable: Srdvas- is a notion closely connected with society,
n this society by the means which it 'pro
eda suggestive of an overseas trade, or o
of gaining renown by wealth acquired by such a trade. On the other
hand there are several passages which suggest the inference that the
cosmogonical streams released by Indra were as much desifous of glory
as their ritual counterpart, the Soma: Soma is likened to horses in IX.
66.10 pdvamdnasya te kave vdjin sdrgd a s f i a t a / drvanto nd Sravasydvab,
I X . 87.5 ett sdmd abhi gavyti' sailzdsrci nzahk v&iiy
pavitrebhih pdvamEna asygraA chravasydvo nd py
prd snindso rdthd ivd 'rvanto nd Sravasydvah /
The association with the cosmogonical waters
I. 125.4 tipa k~arantisindhavo mayobhtiva ijdn
dhendvah / pygdntariz ca pdpurik ca Sravasydvo ghytcisya dh&d tipa yanti
vis'vbtab "Erquickend stromen dem die Fliisse, dem die
der geopfert hat und opfern wird. Dem, der reichlich s
schenkt, Sessen von allen Seiten die Schmalzstrijme wett
ow, just as Soma is rnythologically conceive
om the subterranean samudrd-, thought of as
and as such is likened to horses sirous of renown", so we
interpret I. 48.3 as "the goddess wh nhates the chariots that ar
in readiness at her approach and are desirous of renown like (the waters)
in the subterranean ocean". If the arrival of Usas coincided with the
annual vytrailzdtya- and with the renewed release of the waters and the
goods of life, this association would be quite understandable. In any
case, the frequent references to Usas arriving with her own chariot(s) do
not prevent us from taking rdthdn&n here as denoting the chariots ready
for the fight for "glory" and for the hitdni dhdnam. A few stanzas from
a single hymn to Indra may be quoted in illustration of this interpretation:
VI. 45.11 tcim u tvd ydh purhitha yd vd ntindni hit6 dhdne / hdvyah sci
Smdhi hdvam, 12 dhlbir drvadbhir drvato v@d& indra iravdyydn / tvdyd
j q m a hit&+ dhdnam, 14 ya' ta ntir amitrahan mak&javastama' 'safi / tdyd
no hinuhi rdtham.

K. conclusion.^
We have tried to demonstrate, through the discussion of a single
deity, that there is more evidence in the Rigveda suggestive of some

ss Against Liiders, Varutla, pp. 268-271 ;cf. also K. Hoffmann, OLZ, 1954, col. 391 f.
 of the new year than to
nstration, however, rY
to discuss the entire evidence, including the hymns to Indra and Agni,
and such very significant terns as svarvid- and svdr~dti-,whose true
character as referring to the finding and winning of the su6 of the new
is mostly misunderstood. Cf. also parivatsark,
deed, no Vedic god can be fully understood
outside the context of the whole mythology. That of the
us always back to the cosmogony and its annual reiterati
Year festival.
There are certainly a few passages in the U~as-hymnswhich suggest a
reference to everyday events, such as those depicting birds flying out in
the early morning and men going to their work (I. 48.5, 6; I. 124.12,
IV. 51.5) and the phrase divCdive in the very special context of I. 123.4
(see further p. 227). twithstanding these, and in spite of the fact that
in some other religi he daily return of the sun is indeed celebrated,
it may be stated in general that the hymns to U ~ a are s unaccountable as
documents of religious thought, unless we take U ~ a sto be in the first
place the Dawn of
On the other hand, the preceding observations by no means claim to
have fully explained all aspects of Usas. The sole fact that the Indo-
Iranians worshipped, besides the Sun, which is the most radiant manifes-
tation of the creation of the Cosmos, a goddess Dawn, indicates that
they recognized in her an essential aspect, different from that of the Sun.
It is not yet possible for us to understand, just what aspect induced them
to worship such a separate goddess beside Agni-Siirya. However, the
fact that Usas seems to have been opposed both to Night and to Day-
break, shows how intricate the theological ideas may have been. Cf. e.g.
TB. 111. 8.16.4 u&se svshd v&fyai svdhk 'ty Gha, rdtrir vd W&, dhar
vy&fib, ahordtrd evd 'vamdhe, dtho ahordtrdyor evci prcititififhati, etc.
(Comm. : u~abiabdenardtryabhim-nini devato 'cyate). Indeed, Uws, as a
kinswoman of Varuqa (I. 123.5) and a sister of Night, is both closely
to Night and the nether world (cf. also G. Montesi,
28 (1957), p. 11 ff.). The relation Usas: Night may
have been parallel to that of Mitra: Varuna. Anyway, there remains a
particularly obscure aspect of the mythological concept of U~as,,which
it is impossible to explain, viz. her erotic character. Since this is, however,
clearly an essential feature of hers, we can be sure that we are a long way
yet from a complete insight into the nature of the Indo-Iranian goddess
Dawn (cf. 111. 61.7).
these, it is true, s
 In addition to the readings of the .the evidence of the metre is of
some importance for dete ct Old Iranian form of the word
(cf. Geldner, gber die en Avesta, pp. 31 and 39). Out
of the nine passages whose metrical structure is clear enough to allow
some conclusions," three demand a four-syllabic form [viycixana-1,viz.
Yt. 10.25 diiro'.saokam vyiixanam
Yt. 10.61 ara6wd.zangam zaZnaghuntam
spasam t a x m m vyiixanam
W. 13.85 yqmca ii8ro' urvcizijstahe
spantahe vydxanahe
In the remaining passages the word is treated as a trisyllable, viz.
Yt. 5.73 (avaj dyaptam dazdi-nd)

&nova t e a vycixa

Ut. 13.52 buy@ nci stdhyo' vycixmB

y&nb bd8a friiyazciite
y. 3.10 strThyanqm vycixaniutqm
Ut. 13.16 dyhqn? raya xvaranayhaca
us ncS zayeiti vycixan6
vyaxmiihu g&"aya_t.uxdd
At first sight, it might seem most natural to take these readings with
Geldner as denoting [vycixana-1, resp. [viylixma-1, the more so, as the
cognate word vydxino'hu in our last quotation seems to represent a
trisyllabic [vycixmahu], loc. plur, of vycixmn-. Since however Indo-
Iranian verbal roots ending in a guttural always have the corresponding
palatal before the morpheme -an~-~Z (cf. AV.raocana-, aojana-, ana5Zana-,
aiana-&), Waclcernagel suggested to take vycixana- as [v(i)ycixna-1;see
Festgabe Jacobi (l926), p. 5. The validity of his argument may be doubted,
since proto-Indo-Iran. kh - at least in those cases where it represented
an earlier group of two phonemes kH - seems never to have been

The metre of Yt. 13.108 is not clear. Henning's discrediting of the!whole theory
of Geldner's Metrik shoots far beyond the mark. Though less correct than the Gathic
prosody (which Henning leaves entirely out of consideration), it sometimes allows
conclusions which are confirmed by historical linguistics. That m a d can be read
both as a disyllabic and as a trisyllabic word (TPS,1942, p. 42, n. 2) is not so abstruse
as Henning apparently held it to
,Altind. Gramm., l,p. 151 ;Festgabe Jacobi, p. 5 (Kleine Schriften,
p. 421).
a Unless this is a thematic derivative from +&an-. Probably connected with SM.
aghh-, see Acta Or., 17, p. 35.
palatalized before front vowels. Still his conclusion must be correct, as
a word [vyaxana-] would have to be traced back to a root "vyakh-
*ydkh-, or *cikh-. Now, surd aspirates have hardly constituted a separate
class of phonemes in pr~to-IE.,*~ whereas Indo-Iranian roots ending in
kh are very rare: the sole Rigvedic instances are, indeed, ilikh-, Wkh- and
rikh-.
On the other hand, there is one word which seems decidedly to demand
the assumption of a root (v)ycix-, viz. vyiixa- in Giih 2.8 vycixamca
hanjamanamca ya_t as amaJanqm spantanqm. Bartholomae rightly pre-
ferred this reading vycixamca (Mf, E, and two other MSS.) to the lectio
facilior vycixmmca (Pt, and five MSS.) of Westergaard's and Geldner's
editions. Indeed, the context requires a substantive, and vycixnam is
obviously a clumsy emendation of a copyist, who substituted the well-
known adjective for the obscure and isolated vyiixam. This does not
imply, however, that the better reading is also the correct one. The whole
passage in GZh 2.8 is apparently an imitation of Y. 57.12 (sraobm), yd
vispaEbyd haca arazaZibyd vavanvd paiti.jasaiti vydxma amaSanqm span-
tanqm "(SraoSa), who glorious comes back from all battles to the vycixman
of the Ama3a Spantas". Since the stem vycixa- is not attested else~here,4~
there can be no doubt that vycixamca is a corrupt reading for *vyclx(a)maca,
owing to an incorrect vocalization of wy'hmc. In contrast with the disyl-
labic stem vyclxman- in Yt. 13.16, the end of Y. 57.12 may be read as
follows :
[patijasati viydxma

Nothing, accordingly, prevents us from taking vidxna- (v.1. in Y. 62.5,

Yt. 5.73; 10.25, cf. vayclxna- v. 1.) as a more exact rendering of the Old
Iranian [viydxna-1. Its frequent spelling as vyclxana- and its use as a
four-syllabic word in three metrical passages is paralleled by the word
JyaoOana-, n. "action" which, though standing for [JyauOna-] (cf. Ved.
cyautrui-), is only written as a disyllable in the YaSts, the Vendidad and
some late texts, whereas the Yasna text always has iyaoOana- instead.

U See, e.g., Wackernagel, Altind. G r a m , I, Nachtr. ad p. 140, 11; Notes on Vedic

Noun-Infixion, p. 28.
Ved. farikh&- "shell" (AthS.): Gr. x6y~ocprobably not from IE. +konkho-, since
x6~hoc(and xdrhm?)rather point to a foreign origin, which might also account for kh.
The correspondence B: x would have to be explained as in B&-, m. "hemp" @S.+),
Oss. sun: Gr. x&waprc(note the Rigvedic p!). Forpythuka- see Lingua, S, p. 320, n. 1.
The reading vy&anqm for v y d u a ~ n q mNy. 3.10 in F, Pt, etc. is of course due to
In the Gathas it is actually used 8.5 (which consists
+
of lines of 5 7 syllables) :
huxs'aerci xi3ntqm md.n3 duSa.xJaBrd xJ3ntH

@pi.zqBam vahis'ta
n3 xvara8@i
fSuy6

[huxia6rd xs'yan t , mH nah ddxs'aerci xiyanta

vahviycih cistais' s'yauB.ntii3, aramatai
apizan$am vahis'td
gavai vrzyatHm. ifPn nah huar8di fhyah]

"Good rulers must rule us, may not evil ruler

good insight bless, O Aramati,
the Cow her activity. Fatten her for our food
in U. 31.15 might be read as [du3iiyaudndi]
(Andreas-Wackernagel: dus'Jyau8n i, Humbach: duJjrauOanHi), here no
other interpretation seems possible (despite Humbach, who assumes
5 + 6 in b and d). Similarly zstnd- is trisyllabic in RS. VI. 47.2~
purdei ycis' cyautnd Jdmbarasya. n quite the same manner the four-
syllabic vydxana- will have to be taken as [viy@x.na-]. The current
spelling vydxana- (which in Y. 62.5 has the weighty support of such
MSS. as J, K, Pt,., Mf, JpJ and iyao8ana- suggests the influence of a
normalizing school tradition, although in other cases, such as GAv.
varana- Y. 45.1, 2 against varana- U. 31.11, 48.4, 49.3 (Yt.) no such
k e d tradition is met with.
A derivative of vycixana- is the hapax legomenon vycixanya- in Yt.
14.46 a8taZca tZ vHc6 y6i uyra ds darazra cis, uyra ds vydxaine Hs, uyra cis
varaerayne cis, uyra cis baZs'azya cis "These are the words that are strong
and fh, strong and viyHxniya-, strong and victorious, strong and
healing." Here we must apparently assume a stem [viycixn(i)ya-] parallel
to [vyeragnya-]and [baigazya-1, but while the last two words are derivatives
from a substantive, vyHxaine is more likely to be the common adjective,
which has here incidentally adopted the ending of the other words.
Benveniste, Yrtra et VyOragna, p. 44 n. 3 reads vycixmanya-, which, though

Many details remain uncertain. See, e.g., Lommel, Goft. Nachr., 1935, p. 130.
For the meaning of yaoZdd- see W. B. Kristensen, Leven uif den Doodl., p. 116; Ver,
zamelde Bijdragen, p. 239; for the durative present [xSynntbm] see ZZZ, 8, p. 248 f.
for [huaredi] cf. Y. 34.11 [huarB& (I amrtrisca]. -4 quite different translation is now
proposed by Humbach.
8 Seeold, Vedic Metre, p. 22, and Oldenberg, a. I.
probably an error, would be a more regular formation. The
however have only the readings vycixaine, vicixni (vjcixana, vcii.dxaine).

meaning assigned to vycixana- in the Altir. . (see p. 243) has

ly been questioned by Benveniste, 1.c. He poi out that the word,
when used as an epithet of the Turanian DSinus (Yt. 5.73) must refer
not to their eloquence but rather to "quelque qualitC guerrihe". It is,
indeed, everywhere used in a context referring to "la fermett, la force
ou l'autoritt", and in juxtaposition with such adjectives as taxma-
"valiantyy,stclhya- "firm", karJd.rriza- "regulating the furrow, or boun-
dary-line" (cf. Thieme, Mitra and Aryaman, p. 29, Gershevitch, The
Avestan Hymn to Mithra, p. 210). Cf. Yt. 13.16 vycixand vydxmdhu
g&"ayaj.ux86 ,,qui dans les tournois fait triompher sa parole", from
which words Benveniste rightly concludes: "I1 s'agit en &et d'une joute
oratoire d'oh cet homme sort victorieux". This led Gershevitch, op. C.,
p. '77, to translate it as "challenging". Since however Benveniste, in the
absence of a convincing etymological explanation, refrained from deciding
whether the word originally referred to the debate or the combat, his
definition "'oiyov~crri]~,propre A la rencontre, apte au dtbat ou au combat,
champion" has given rise to some misapprehensions. Thus Nyberg,
Irans forntida Religioner, p. 68 (cf. Die Religionen des alten Irans, p. 438)
paraphrased this in his own words by "connected with the battle or the
battle-field, god of the battle-field", and Wikander, Vdyu, I, p. 200 n. 1,
considered the possibility of deriving arJya- Yt. 13.108 from araJa-
"bear" because "apte A la lutte" would then be a fitting epithet. We
must therefore consider its use more closely. Yt. 13, after stating that
through the brightness and glory of the Fravagis the embryo is maintained
in the womb (11), and that the womb in females, the milk in females'
breasts is purified (S), the females conceive offspring and become blessed
with children (19, has the following lines (16) :
hghqm raya xvaranaghaca
us nd zayeiti vyclxand
vydxmdhu g~2fayaj.ux8d
yb bavaiti xratu.kdtd
yb nclidyhghd gaotamahe
pard.yh parStdi_tavditi . . .
"Through their brightness and glory a man is born who is victorious in
debates, whose authoritative words are listened to in the verbal contests,
 ho is esteemed for his quick o comes off from the
triwnphing over the weaker Gautama". As for -uxda-, Thieme is probably
right in rendering it by "der feierliche und massgebende Spruch",dB
while by3ti-1, the dispute in which only one's quick wit can lead to
victory, may be compared to Ved. p l ~ t h d"Ratselfragen" RS. IV. 2.1 1,
X. 89.360and to pr&- "(assertion in) a dispute".
Quick wit was of immense importance in ancient
iti prativaktavya~sabhcisu vijigi~ubhib!). It may be
fore, to consider more closely the religious background of this notion.
That inspiration is suddenly born from Man's unconscious mind, could
only be due to divine assistance. Indraysbreaking down of the resistance
(vytratdrya-, vytrahdtya-) not only brings to light the goods of life by
ting open the primordial hill, but also, and at the same time, breaks
an's inner resistance and gives inspiration. Our formulation may seem
modem, the ideas certainly are not. The Vedic poet juxtaposes in one
verse vc@asdti-"the winning of the prize" and medhhdti- "the winning
of'the (religious) insight" as manifestations of the same salutary activity
of Indra, e.g. VIII. 40.2 sd nab icadd cid rfrvatd gdmad d vdjasdtaye,
g h d d medhhdtaye. The god of inspiration par excellence, however,
is Agni. Born in the nether world as a child of the Cosmic Waters, he
manifests his victoriousness by his arrival as a guest (dtithi-) among
men, and at the same time he brings with h, as the gurupar excellence,
knowledge of the cosmic Truth, wisdom, insight, and inspiration. He is
indeed the dhhd rthya (I. 67.7). The well-known hymn RS. *VI. 9
depicts how the poet, by devout concentration upon the god, experiences
this inspiration as an ecstasy. Hence Agni is called "the inventor of
resplendent speech" (#v& iukrhya vdcaso mandta 11. 9.4), "bringing
the light of inspired speech" (vipdrit jydtilit~ibibhrat 111. 10.5). Most
instructive is RS. IV. 11.2-3, which Geldner renders as follows: "2.
Schliesse, o Agni, dern Lobsanger durch Beredsamkeit den (dichterischen)
Gedanken (wie) einen Kanal auf, wenn du, von starker Art, gepriesec
wirst. Gib uns den reichen. Gedanken ein, du Herrlicher, den du mit
d e n Gottern gut aufnehmen wirst, du Hellfarbiger. 3. Von dir, Agni,
stammen die Sehergaben, von dir die Gedanken, von dir die trefichen
Gedichte. Von dir kommt Reichtum, dessen Zierde die Sohne sind, dem
danach trachtenden opferwilligen Sterblichen". Agni, like Indra, bestows
at the same time material riches, sons, and inspiration (mani~d-,bhhi
m Thieme ap. Altheim, Weltgeschichte Asiens, I, p. 98 f.
50 Oeme, Der Fremdlng im Rigveda, p. 65; Renou-Silbiurn, JA, 1949, p. 43, n. 2
(equivalent to brahmodya-).
 cinma). The most interesting detail, however, is that Agni is implored
to open the kM-, which is (like zitsa-) the specitic term for the aperture
in the cosmic hill,the well which Indra has opened with his vcija. Cf. IV.
28.1 dhann dhim drindt saptd sindhiin, Lipclvrnod dpihiteva khdni, 11. 15.3
vcijrena khdny atman nadhclm, V. 32.1 cidardar zitsam dsrjo v i khdni, VII.
82.3 (to Indra and Varuna) dnv apdtn khdny atrntam djascl. Now inspi-
ration was to the Vedic poet an "opening of the doors of the mind",
which was on a level with the "opening of the doors of the primordial
hill", cf. e.g. IX. 10.6 cipa dvdrd matinch pratnd rpvanti kiircivah with IV.
51.2 vy i?i vrajdsya trimaso dvdro 'chdntir avran" chzicayah pdvakdh (see p.
225). The same mythological concept is expressed in IV. 11.2 by the
word khd- "aperture". This aperture is the well which Agni must open
to give free course to Inspiration from the hfdya- samudrd-, the primordial
waters in Man's heart (IV. 58.5, cf. p. 241). Now it has long been
observed that this notion of a cosmic well was inherited from the
common Indo-Iranian religion; the older athematic form khd- occurs
both in RS. 11. 28.5 rdhydma te varuga khdm rthya and in the Avesta,
cf. Y. 10.4 haieimca aiahe xd ahi "Verily thou art the well of the cosmic
Truth". In explanation of this mythological concept a few words may
here be inserted. Varuna is the god of the cosmic waters beneath the
earth. A striking confirmation of this interpretation, which alone
enables us to understand the place of Varuna in Vedic cosmology, may
be seen in the curious fountain slabs dating from the 11th and 12th
centuries A.D. which Vogel has found in Chamba State and which the
inscriptions call Varugadevah (3. Ph. Vogel, Antiquities of Chamba State,
1911, p. 29 R.). Varuna was also the god of stagnant water (TS. VI.
4.2.3, SE. IV. 4.5.10, cf. MS. IV. 8.5: p. 112,3 ff.; KS. XXIX. 3: p. 171,3;
KapKS. 45.4: p. 270,22; JB. 11. 67), because this represents the sub-
terranean waters of Varuna's realm of inertia. Now the well-spring is
the place where Man comes in contact with the nether world and, since
e t a resided in these waters (RS. 1. 62. l), also with l$ta itself. (The other
'
aspect of the nether world explains the belief that the svdqtam irigam
and the pradardh are nirytigrhita-, TS. 111. 4.8.5). The occurrence of
similar ideas in Delphi and Rome may be noted in passing. It is not
surprising, therefore, that just as Agni is implored to bestow inspiration
in the Veda, so the Fire-god Atar is in the Avesta. Cf. Y. 62.4 day6
me' . . . mastim spiind xfviwram hizvqm urune' ufi xratiim pascae'ta masitarn
mazdntam apairi.ci8ram "Gib mir . . . Wissen, Heiligkeit, eine flinke
Zunge, der Seele Auffassungskraft, alsdann Gedachtniskraft, grosse
umfassende unvergangliche" (Bartholome-Wolff). The notion that the
 ,to the Fire-god must therefore
do-Iranian religion.

Reverting to the passage Yt. 13.16 we must still consider more closely
the most interesting word, viz. xramkcita-. This has

olomae), "von dem man

comes clear from a CO
to Speech (X. 71), which is one of our mo
ation concerning the ancient an verbal contests.
(sakhyd-) has among its members (sdkhliyah) some who have become
"both stiff and stout" (5a sthirdpita-) and who they, therefore, do not
send to the contests (5b naina~hinvanty dpi ew). These are '"eft
hind in knowledge" (?c 'ha tva& v i jahur vedyhbhir), they are
useless ( 9 c 4 cf. IV. 5.14!). the most prominent among those who
"have in their heart gained mental quickness" (8a
idvep) they send to the contest and (10) "Allfrien
companion famous Returns from the assembly ha11 a victor, He drives
away reproach and wins them nurture: He is sent out as. fitted for the
a sabhcisdhd- may rightly be called "esteemed
. Cf. Geldner, ubers., I, p. 289, n. 1.
We is also called par& "triumphant" (c.gen.). Bartholomae explains
this word as "voran nd, siegreich" and compares Ved. puroydvan-
"going in front, leading". o d d accordingly be the nominative of a
stem par&yd-, cf. Ved. eva 'gokg quickly" etc. Since however this
explanation is not quite plausible on account of the meaning of par&yd,
we may consider the possibility that it is the nominative -ylih (from
*-yaccih)of a stem -ycih- (from *-ya'~h-);~~ cf. e.g. GAv. hudd [hudh.'lih],
U. 48.3, nominative oi'hurzlch- [h&'&-] . main reason for doing so
Our
is the passage Yt. 13.108

adyehe vylixamhe
yiiskarastamhe mazabyasnanqm
worship the FravaSi of the holy and good ArSya; of ArSya, who is
victorious in debates, who is the best achiever of yah among the wor-
For the analysis
place du ton, p. 21, a
 as ydh- originally the technical term for
atological meaning it has arathustra's theology due to
aPized use of the word? it has preserved a different
meaning in the Ua5ts from that which it has in the Yasna, may also be
inferred from Yt. 11.3 arSux66 vdxS ydhi varaOrajqstam6, which Lornmel
translates "dieser pruch, recht gesprochen, ist bei der Entscheidung (?)
der siegreichste9'. is doubt as to the eschatological meaning, which
artholomae assumed also for this passage, seems justified. Two details,
indeed, speak against this interpretation. Firstly, the ydh- achieved by
the ydskaraf- is a decision brought about by a word, accordingly a
victory in a verbal contest. Secondly, this word is said to be varaorajan-.
Now, vytrd- is in the Veda the force of resistance which Indra and the
other powers of Creation, such as Agni and Soma, have to break down
before the creation of the dual world of the Cosmos can be realized (see
above, p. 218). The vytrahdtya- accomplished by men is clearly a re-
enactment, a reiteration of the Creation itself. Since it is quite incon-
ceivable that this unequivocal reference to the Creation could be due to
a secondary specialization, we must conclude that it was already a
conception of Indo-Iranian religion.62 The fact that Haoma is called
varaorajan- in the Avesta, just as Soma is a vytrahdn- in the Veda, shows
that in the Later Avesta some traces of the older usage of the word have
been preserved. Perhaps such occasional speculations as agdsdm~bhYalrz
vai viry2gt 'ndro vytrdm ahan MS. IT. 1.3 (p. 5,1) also allude to this
ancient mythological idea. In any case the Vedic texts leave no doubt
that the verbal contest was also a reiteration of the primordial fight with
the cosmic forces of resistance (vpdni); this is, indeed, true of all sorts
of combat: "He verily slays Vrtra who is a victor in the battle" (vyrrdriz
khdlu vd esd hanti ydh sarizgrdmdm jdyati S. 11. 1.3: p. 4,17). In AthS.
11.27 thepdtd-plant is invoked to "overcome the debate of the opponent";
cf. the refrain prdiam prdtiprdio jahi. With the aid of the same plant,
however, one hopes to conquer his enemies, cf. st. 5 tdyd 'hdriz Sdtriint
siiksa indrab sdliivykdm iva "by means of it I shall conquer the enemies,
as Indra (conquered) the SiilTiv~kas". The parallelism which the poet
states to exist between his fight and Indra's should be noted. Now in st.
2 this plant is said to have been found by an eagle -just as an eagle (we
may add) is known to have brought Soma (the plant whose juice in-
vigorated Indra in his cosmic combat) to the god -, and to have been
dug out by a boar with his snout -just as a boar has dug out the earth
from the bottom of the cosmic waters, before Indra could expand it
6a Otherwise Ugo Bianchi, Zamdn i ohrmazd (1958), p. 35.
(2ab suparga's tvd' 'nvavindat sakards tvd 'khanan nasd). If however such
importance was attached to the debate, it must have been considered a
creative act itself, a renovation of life. The fact that the battle-drum is
exhorted to "carry forth its sound as a speaker his speech" (AthS. V.
20.11 vdgvi 'va mintram prd bharusva vdcam) points to the same asso-
ciation of ideas. In st. 2 the battle-drum is said to "own Indra's foe-
subduing fire" (aindrds te s'z2smo abhimdtisdha'?~)and in V. 21.8 its sound
is apparently considered to reproduce that of Indra's feet during his
creative dance.53 On the other hand the man who triumphs in a verbal
contest represents Indra hin~self. Thus the poet of RS. X. 166, who in
st. 3 refers to such a contest in the words vd'cas pate ni ~edhe'mdn ya'thd
ma'd a'dharam va'ddn, compares himself to Indra in st. 2: ahdm asmi
sapatnaht 'ndra ivd 'risto a'k~ata??"I am a slayer of my rivals, unhurt
and uninjured like Indra". Therefore the battle-drum is exhorted to
sound as victoriously as the powerful speech of such a conqueror of his
opponents: indeed, its v&- (AthS. V. 20.11) is considered to possess the
creative power of Indra, the vytraha'n- par excellence. Only these parallels
enable us to appreciate the original full meaning of the words ars'uxdd
vdxs' ydhi vara&ajqstarnd in Yt. 11.3. Indirectly this gives a clue to
the meaning of ydskarat- (which will be discussed below, p. 256) and
probably also of pard.y&, which may be taken as the nominative of
parb.ydh- "victorious in the contest with (gen.)". Pard- is then equivalent
to Skt. parcitz "higher, more than" (see Altiran. Worterbuclz, col. 857).

Although this interpretation, based upon the evidence of the Vedic and
Avestan texts, has to be judged on its own merits, a comparison with,an
archaic culture of our days may lend some support to it. The following
quotations have been taken from Hans Scharer's admirable study on the
religion of the Ngaju Dyaks in South-Borneo.54 T o n einem Glied der
hohen Gruppe wird erwartet, dass es ausgeglichen, physisch und psychisch
harmonisch, selbstbeherrscht, zuverlassig, arbeitsam, ruhig, tapfer,
redegewandt und tugendhaft sei" (p. 46). The gods "geben ihm Gesund-
heit und Mannesmut und erfullen sein Herz mit reichen und verstandigen
Gedanken und offnen seinen Mund, dass er ein guter Redner wird, der
in den Formen der Urzeit spricht und Rat erteilt, so dass sein Wort wie
das der fernen Ahnen klingt, die in ihm und aus ihm sprechen. Sie . . .
stehen ihm bei in der Rechtssprechung, damit er als Richter weise urteile
und bei jeder Anklage als Sieger hervorgehe" (p. 89). "Der wahre und
Turner Jubilee Volrme, p. 360.
54 H . Scharer, Die Gottesidee der Nga&u Dajak in Siid-Borneo (thesis Leiden, 1946).
vollkommene Mensch ist basewut = beruhmt. 'Er steht in der Gemein-
.
schaft da wie der hohe Baum auf dem Dorfplatz . .'. Durch ihn
hindurch sehen und sprechen die Gottheiten auch die Gemeinschaft und
den ganzen Kosmos an und segnen sie mit den Heilsgaben, die sie ihm
verleihen" (p. 90). On p. 114 Scharer points out "dass Rechtssitzung und
Rechtssprechung . . . ein kosmisches Geschehen und eine Sache der
totalen Gemeinschaft sein mussen". "Der ganze Rechtshandel ist ein
Streit zwischen zwei Gruppen, der mit Worten gefuhrt wird. Wer am
besten sprechen kann oder fur seine Verteidigung den besten Sprecher
besitzt, gewinnt den Streit. Der Sieg verlieht Ansehen, nicht weil die
Unschuld an den Tag gekomen, sondern weil der Streit gewonnen
worden ist. Aus diesem Grunde wird, wenn ein Rechtsfall geschlichtet
.werden muss, so vie1 gesprochen und gelogen. Man kann nicht sagen,
dass der Dajak ein grosserer Lugner sei als jeder andere Mensch . . .
wenn er beim Rechtshandel lugt, dann geschieht es nicht einfach deshalb,
weil er sich aus einer unangenehmen Situation retten will, sondern weil
er den Wortstreit gewinnen muss. Wir konnen, wie wir noch sehen
werden, diesen Streit nicht einfach als eine profane Angelegenheit inter-
pretieren, er hangt zusamrnen mit dem heiligen Streit der beiden Vogel
auf dem Lebensbaum".
"Bei der Rechtssitzung ist aber auch die Djata [one of the two highest
gods] vertreten, denn wenn die Ordnungen iibertreten werden schenkt
sie kein Gedeihen und Leben mehr bis die Tat gesuhnt, die Schopfung
wiederholt und der Kosmos erneuert ist" (p. 115). "Es handelt sich
auch hier um die Wiederholung des urzeitlichen Heilsgeschehens, denn
die Wiederherstellung der verbrochenen Ordnungen ist nur moglich
durch die Wiederholung der Schopfung und die Erneuerung des ganzen
Kosmos. In den beiden Gruppen, die sich feindlich gegenuberstehen,
tritt die totale Gemeinschaft auf und die Richter sind die Vertreter der
totalen Gottheit. Zwischen den beiden Parteien werden im balai
[correspondingto the sabhd] die Gong aufgestellt und die heiligen Lanzen
aufgerichtet. Sie sind . . . identisch mit dem Lebensbaum. Die beiden
Gruppen, die sich bekampfen, sind identisch mit den beiden Vogeln auf
dem, Lebensbaum und ihr Streit ist eine Wiederholung des Streites
zwischen den beiden Vogeln. . . . Den wichtigsten Platz nimmt der
Wortstreit ein, der leid schaftlich gefuhrt wird. Es fehlt dabei nich an den
grobsten Beleidigungen und Beschuldigungen. Man macht den Gegner
lacherlich und beschimpft ihn auf alle Weise. Der Streit wird aber auch
mit den Waffen gefuhrt und wenn man sich gegenseitig nicht mit Kugeln
und Pfeilen beschiesst, so doch mit heiligen Mitteln, die uber den Feind
 .
Unheil und Verderben bringen sollen . . Darnit ist der heilige Streit
eendet, aber damit ist auch die Schopfung wiederholt, denn aus dem
eit und aus der Selbstvernichtung entsteht nicht das Nichts oder [dasj
aos, sondern dcr Kosmos. Mit dem Abschluss des Streites sind die
nungen wieder hergestellt und man kann wieder als ein neuer Mensch
inem neuen Kosmos und einer erneuerten Gemeinschaft ein neues
en beginnen. Die Rechtssprechung lasst sich nur verstehen im
usammenhang mit dem urzeitlichen Heilsgeschehen und dem schopferi-
schen Handeln der totalen Gottheit, denn sie ist deren Wiederholung und
Dramatisierung" (p. 133).

rom a somewhat different point of view W. B. K r i ~ t e n s e nhas, ~ ~ in

onnection with the very word vydxana-, characterized the role of
quence as follows: "Eloquence had not quite the same meaning for
ancient peoples which it has for us. They did not associate the word
h verbal art only, or with an artistic skill. The main thing to them was
authority of the spoken word, its wisdom and power, the success it
as attended with. The eloquent word of the popular leader gave his
audience the impression of absolute validity; it was authoritative, because
one felt that it revealed a law of life. It was as irresistible and valid as
e law of life itself. It was no mere beautiful sound, which existed only
r a moment. Once pronounced it maintained itself: it created a new
situation, it turned itself into reality. Eloquence, therefore, was nothing
short of a creative force, a vital energy. Its essence was the mystery of
reation and Life". Kristensen was no doubt right in explaining in this
ay the use of vyaxana- as an epithet of Miera, SraoSa and Nairybsagha,
who "overcome the enemies of Mazdii in this world" (op. C., p. 130).
bus Miera is called ars'.vaEagham vydxanam "speaking correctly and
eloquent" (Yt. 10.7), which reminds us of the words arS.ux66 vdxS
iscussed above. Still more instructive is Yt. 10.25
ahztram gufram amavantam
ddtdsaokam vydxanam
vahmd.sandagham barazantam
as'ahunaram tanumq0ram. . .
"the Lord, who is profound and strong, weal-giving and eloquent,
gratifies of prayers and high, whose potency is great and who is the
ID De goddelijke Heraut ett het Woord van God (Verzamelde Bijdragen tot kennis
der antiekegodsdiensten, Amsterdam, 1947, p. 129; first published in the Mededeelingen
der Kon. Akad. v. Wetensch., 1930).
incarnate Word" (cf. also Kristensen's remarks, p. 130, and see now
Gershevitch, Avestan Hymn, pp. 85, 179 E.). iera is also celebrated as
spasam taxmam vydxanam '"he watcher valiant and eloquent" Yt. 10.61.
But Yt. 13.85 mentions along with the fravaSis of SraoSa and Nairyb.sarjha
also that of Atar (the Fire) in the words yqmca d6rd urvdzi$tahe, spantahe
vydxanahe. In this connection it may be called to mind that Atar is the
god who, like the Vedic Agni, bestows inspiration and xs'viwram hizvqm
(see above, p. 249).
Benveniste has already drawn attention to the passages where the gods
are invoked to bestow sons who have authority among the people; cf.
Y. 62.5 frazantim kar.?d.rdzqm vydxanqm . . .qzd.bOjim, Yt. 13.52 buy@
ahmi nmdne . . .buy51 nd stdhyd vydxand, and . 3.10, where it is said
that the Moon
daste narqm pourutdtam
stdhyanqm vydxananqm
vanatqm avanamnanqm
ha6ravanatqrn hamara63
"gives a great number of male children, firm and eloquent, victorious
and undefeated, who smite at one stroke their opponents".

From the preceding discussion of the Avestan evidence, the specific

meaning of the technical terms will have become sufficiently clear. The
word for the "ceremonial meeting" is vydxman-, while the quality of the
winner in the contest is denoted by vydxana- "eloquent". Cf. vydxand
vydxmiihu gOfaya&ux% Yt. 13.16. The verb is vydxmanyeiti "to speak
in a contest"; the words which the demon SniiviGka speaks in defiance
of Karastispa before he is slain by the hero are introduced by the
formula h6 avaea vyrIxmanyata. Close parallels in the Homeric idiom
are, e.g., B 597 G T E ~ ~yhp
G E & X ~ ~ JVGX?&L&V,
G ~ A 388 $lt~ih~C&V@80v.
Less clear are the three passages in the TiStr YaSt (Yt. 8.15, 17, 19):
hd i6a vydxmanyeiti, hd i6a parasanyeiti: kd mqm nOrqm frdyazdite . ..
nfirqm ahmi yesnyasca vahmyasca aguhe astvaite a$@ haca yaj vahiita.
Bartholomae translates "Der hier spricht in der Versammlung, der hier
stellt die Frage"; cf. Lornmel: '"er geht dort mit sich zu Rate,
sich dort", Darmesteter : "Et 18 il convoque les h o m e s , il leur demande".
Now in 10-11 TiStrya has promised Ahura Mazdii to assist those men
who will worship him. In 13 he makes his first appearance (niirqm!) and
claims libations with haoma and milk from them. We openly declares to
be one who ought to receive sacrifice and prayer, but men do not fuliil
his wish and, being consequently too weak, he is defeated by ApaoSa
(22-24). Although it is hardly possible to ascertain the exact shade of
meaning, it may be conjectured that vydxmanyeiti here means "boasting
he claims a new status"56 (with reference to the words ntirqm ahmi
yesnyasca etc.). We may again compare the Homeric verb byopciopa~,
also derived from the word for "meeting", which means "to speak with
authority" (byop~cra~o xai C L ~ ~ k sometimes
~ ~ ~ ~ ~ implying
) , 5 7 the utterance
of boastful words, cf. @ 229-230 4 Epav ~ & x w A a i. . .&G . . .XEVE~UXLE~
$YO~&EO~E.
As pointed out above (p. 251 f.), also ycih- must be reckoned among the
technical terms of the ceremonial contest. It apparently denotes the
contest itself (like paraNi-). Cf. Yt. 11.3 ahund vairyd vacqm vara0ra-
jqstamd, arSux66 vdxi ydhi varaBrajqstam6 "The Ahuna Vairya is the
most victorious (resistance-breaking) of words; when pronounced in the
right way, it is the most victorious word in the contest", and Yt. 13.108
vydxanahe . . .ydskarastamahe (see above). The word ydskarat- is in all
other passages of the Fravardin YaSt an epithet of the FravaSis (see 64,
75, 76). Bartholomae translates "der beim Schlusswerk . . . tatig ist";
Lommel mostly refrains from rendering it, but suggests "was (die Ent-
scheidung?) bewirkt". The word applies to persons (Yt. 13.108 and the
FravaSis) and to words, cf. Yt. 1.1 kaj asti mq0rahe spantahe amavas-
tamam, kaj vara Bravastamam, kaj xvaranaguhastamam, kaj ydskarastamam?
"What of the Holy Word has the greatest offensive power, what the most
defensive? What is the most glorious? What is the most ydskarat-?" and
in 1.5 frdi_t m2 taj nqma framriiisi. . . yaj tZ asti maziSramca vahiitamca
sraPJtamca ydskarastamamca. . . "reveal unto me that name of thine. . .
that is the greatest, the best, the fairest, the most ydskarat-. . .". In Yt.
13.64 the FravaSis are said to be greater (masyehii), stronger (aojyehi;),
more valiant (tqiyehii), having a greater offensive and defensive power
(amavastard, varaoravastard), more healing (baPiazydtard) and more

In this connection attention may be drawn to the description of the owasa (a

potlatch-like ceremony) on Nias (Indonesia): "The person sponsoring the owasa
exploits the situation to the fullest in showing the guests what he has accomplished,
by strutting about the village square laden with jewelry, making speeches in praise of
himself and/of relatives. . . while the guests look on with awe and admiration, and the
kinsmen of the sponsor beam with pride and bask in glory. By showing off in this
manner, the sponsor of the owasa, as in the potlatch, claims a new status and title, the
.
validity of which . . must come from those attending through approval of the spon-
sor's feast" (P. Suzuki, The religious Sy~tenzand Culture of Nias, Zndonesia, thesis
Leiden 1959, p. 105).
A classical description of &yop?tJ~ as a source of social prestige occurs in Horn.
0 168 ff. For similar testimonies from the Latin literature (Tac. Dial. 8 ipsa eloquentia,
..
cuius numen et caelestis vis . !) see, e.g., Wagenvoort, Zmperium, p. 106.
yaskarat- than can be expressed by words. No more can be inferred from
these passages but that yrlskarat- denotes some power which is also present
in certain words or formulas. While yciskarat- must mean "bringing
about the decision in the yrlh-", the last word itself seems to denote the
(dangerous or critical situation of the) cosmic or social contest. Besides
Yt. 11.3 it is met with in Yt. 13.41
kahmliici_tydyhqm jasd
kahmdicij qzaghqm biwivd
These lines seem grammatically corrupt (dativus finalis? Cf. U. 36.2 nd
maziStdi ydghqm paiti.jamyd "komm uns zu der grossten der Entschei-
dungen entgegen" Bartholomae, cf. Humbach, MiiSS., 9, p. 77), only the
parallelism with qzah- "distress" is instructive (see p, 221). The use of
the same word in Zarathustra's theological terminology will not be
discussed here. As in analogous instances, Eater Avestan may be expected
to have preserved the older and common meaning of the word.
the Later Avestan passages are not conclusive, but those pointing to
some connection with speech are significant. They allow us to reject
most conjectures of its meaning that are merely based on etymologies,
as "kriegerisches, gefahrliches Unternehmen" (Baunack, Studien, p. 362),
"kochen, Kochung" (Skt. yas-, Hertel, Indo-Iran. Quellen U. Forsch., VI,
p. 153, cf. Schwyzer, IF., 47, 1927, p. 237), "der Gang zum Feuerordal,
in dem Gott die kiinftige Entscheidung vollziehen wird" (Schaeder,
ZDMG., 94, 1940, p. 40!, n. 2), "Opferumzug", "Bittgang" (Humbach,
MiiSS., 11, p. 13, n. 17, resp. VIII, p. 8 l , IX, p. 77). The semantic develop-
ment remains obscure, but there can be no doubt as to its being an ah-
noun from yrl-.

C . THE ETYMOLOGY

The formal analysis of the first section leads us to posit two Indo-Iranian
words, *vi-yak-na- and vi-yak-man-. Since Nyberg has withdrawn his
interpretation of Phl. vyaxrnan in GrBd.68, and since the v. 1. vyaxma
Y. 57.12 is the only reading with a short vowel, there is no reason to take
yZ- as merely graphical for ya-.
Still Iran. yrlk- has to be connected with IE. *yek- "to declare
solemnly". . This meaning is still preserved in OHG. Jeharz, gehan
"sprechen, aussagen, erklaren, behaupten, bejahen, eingestehen, beichten"
(cf. jiht, bijiht f. "Aussage, Bekenntnis", NHG. Beiclzte). In Old Saxon
gehan means "to declare publicly" (e.g. gihid that he god s i "he declares
HiIfsbuch &S Pahlavi, p. 67,17; see, however, p. 265.
that he is God" NE1. 5104), while begehan has the connotation of a
presumptuous declaration (e.g. quidit that he Krist si, kuning obar thit
riki, begihit ina sdgrdtes "he says that he is Christ, king over this kingdom,
he presumes so much" WE1. 5194). Since this is the very meaning which
we have to presuppose for AV. vydxana-, this connection seems to stand
firm. With the Germanic words (and with Welsh iaith, Breton iez
"language", from *yekti-) M. van Blankenstein has further connected
Umbr. iuku, iuka "preces" (ntr. plur. of ioko-), Lat. iocus "joke" and
Lith. ju6kas, id.59 The meaning of the Umbrian word can easily be
explained from "solemn declaration". Indeed, Devoto renders the
words estu iuku habetu, which are used as an introduction to a prayer,
by "has orationes habeton.*O The Latin word, however, which has long
been taken as a cognate of it,61 has an aberrant meaning, which has
induced some scholars to separate it from this w~rd-family.~zWe are
here confronted with the first instance of the curious semantic develop-
ments which technical terms of the boasting contest may undergo.
Bragging and mockery are the two requisites of the verbal contest all
over the world. With Scharer's description of its function among the
Dyaks we may compare such a classical instance of the Old Germanic
gilpcwide as occurs in BCowulf 499 E., where BCowulf, ridiculing Unferb,
has the laugh on his side (61 1 f.). Lith. ju6kas "joke, mockery, laughter",
which points to a similar semantic development, is besides interesting
because it shows the long root vowel which we had to assume for AV.
yak-. As one of the many semantic parallels we may quote Goth.
swaraiz "to swear", Old Engl. andswaru, Old Saxon antswdr "answer":
OChS1. svariti "bppi<~tv,AOLGO~CTV, p k X ~ ~ 8 asvara
~ " , "rixa", svara "pciXq".
The only word that presents serious difFiculties is Skt. ydcate. In the
Rigveda, where its use is confined to the last three books (VI113IX2X3)
it means "to ask, solicit, entreat" and refers to prayers to the gods, with
the sole exception of VIII. 2.10 sdmdh . . .s'ukrhdiiram ydcante "verlangen
nach Milchmischung". This meaning can of course easily be explained
from "taking a solemn vow to the gods with the simultaneous declaration
of one's desires". Cf. Ved. d-s'liste "he asks" (tvd I. 30.10), "he prays for"
(tdd I. 24.11), Lat. voveo and Greek sij~opar "I pray", which have
IF, 23, p. 131ff. Generally accepted, cf. Pokorny, Indogerm. etym. Worterb., p. 503.
Uhlenbeck controverted it in Beitr. z. Gesch. d. deutschen Spr. U. Lit., 35 (1909),
p. 168 f., on account of Lit. juokirios, ju6ktis "to laugh" and connected jehan with
Skt. ybias-.
Tabulae Iguvirzae (2nd ed.), p. 366 f.
R. von Planta, Gramrrzatik der oskisch-umbrischen Dialekte, I , pp. 167, 411 f.
Uhlenbeck, p. 169; Devoto, p. 367.
developed similar meanings from that of "declaring solemn1
Atharvaveda downwards the word is also used in the se
There is however the syntactical difficulty that even in the
is construed with a double accusative, just like its syno
+
h a s - yiic: tydn nnzi . . . h a dditydn ydcipTrnahe VIII. 67.1,
ta't tvd ydcdmahi 'vah X . 22.7.
+
dvas- yd-: rcci ydmi . . . devci& a'vo VIII. 27.1.
+
rayi ydc-: rayim . . .itzhv indram asma'bhyam ydcatcit IX. 8
+
rayi- yd- : ydt tvd rayim ydmi VIII. 3.1 1.
For this reason Grassmann and other scholars explaine
consisting of yii- and a root determinative -k-.B3 Hu
assumes an etymological connection with AV. y~?is-,~*
time identifying Gathic yecH (Y. 30.1~.51.2a) with ycl
evidence, however, shows the root to have a long vo
can hardly stand for [ y i i ~ d the
] , ~old
~ idea must defin
In other respects too, Humbach's interpretation o
(based on AthS. XV. 13.8) seems incorrect: as I hope to discuss else
.
aid yecii . . urvdzd in Y . 30.1 is a corrupt reading, which mus
crept in during the early Sassanian period, for as'ahyaca . . . urvfizlc7
['tahya ca . . . vrdzd] "and the bliss of gta". However, this
remove the difficulty that Ved. ydc-, from its earliest occurrenc
wards, shows a grammatical construction which must be due to the
analogy of yd- (and ptcch-). Traces of an older meaning than that of
asking are indeed remarkably rare.

Most interesting, therefore, is the use of abhiycicanci- in the

Skt. compound satyclbhiyclcand-, a synonym of satyavacana-, satyd-
dhighiina- "appeal to truth". Edgerton records one occurrence in Divy.
154,5 etam ciryiitz Kiilam rdjakurndram satydbhiycicanayii yathdpaur&tzar;;l
ktirudhvam iti. He seems to have overlooked, however, that safyopayd-

Grassmann, Worterb. zum RV, col. 1107; W . P. Schmid, IF, 62 (195

(from yd- "to beg", secondarily fused with yd- "to go", cf. Gr. -+m).
64 AV. y&-, an "inchoative" present of yd- (cf. cipd ydnarn v6 ydscim
Humbach refers to MiiSS, 8, p. 83. The same explanation has already been propose
by de Harlez, Per Persson (cf. BB, 19, p. 279), Meillet, BSL, 24, p. 114;
stierne, NTS, 12, p. 112.
Resuscitating an idea of Bartholomae's, GrIrPhil., I, 1, p. 174.
As for ciyest?, this does not stand for "dydsd (W. P. Schmidt, IF, 62, p. 235
belongs to d-yam- (Aliir. Worterb., col. 1263), cf. OP. ayasatd and Nyberg,
fornt. Religioner, p. 499. Hurnbach's explanation of yecd Y. 30.1 had already been
suggested by Kanga (rejected by Taraporewala, Journ. K. R. C a m Or. Institute, 12
(1928), p. 46).
Bombay ed. Grantha rec. v. Schlegel N.W. recension N.E. recension
11. 55.6 11. 59.4b-5a IT. 55.4b-5a
nil nil = N. E. rec. tasyci nadydh pare pdre tasyci naa'ycifzpare pare
izcitidz2re malzddrumab niifidr7rematzlidrmnza(z
+satydpi +pdvitah irim&z satydbIziycicanah irimLTiz
tzyagroa'lzo haritaccliadaft nyagrodho haritacchaclah
st. l 8 st. 18
nil nil nil yaycice tam tat0 'blzyetya yaycice tam tato 'blzyetya
nyagroa'lzani satyaycicanam iycimam satyopaydcarzam
11. 68.16 IT. 68.16 11. 68.16 11. 74.13 11. 70.14b-15a
,
nikiilavrk~amcisddya samiilacaityarn ciscidya samiilatn caityam iiscidya
divyarit satyopaycicanam vrk~aihsatyopaycicanatn vfi~arnsatyopaydcarzam
abhigamyci 'bhivcidyam tani abhigamya pm<zamnyaitzariz abhigamya pranamyaitmm
Kulingcim pnividan purim Trililigcin? viviiu!z purim Bh~Xligdtnviviiu!~purim
 cana- is used in the same manner in Avad. 48,6 tatah saptame divase . . .
gomayamapdalake khte sarvagandhamdlye~ti 'pahyte~uparvatarah tirtha-
kopdsakena satyopaydcanarn kytam: yena satyena . . . anena satyena . . .
upagacchantv iti. Similarly in 49,2.@' In these words the original meaning
of ydcana-, viz. "solemn declaration" seems to have been preserved. If
this is so, however, it is hard to account for the fact that both compounds
also occur as variant readings in the Riimgyana, where they are
epithets of holy trees. As their meaning cannot have been "die Bitten
wahr machend, - gewahrend",(Bohtlingk),their application as adjectives
to trees remain unexplained. They are only attested in two passages.
The relevant stanzas in the various recensions are quoted on page 260.
Bharadv3ja9sadvice that sit31 when passing by the holy fig tree, should
pay homage to it (55.6 tasmai sitd 'fijaliliz kytvd prayufijitd '&ah s'iv*,
with w.11.; a different text in NW. 59.6 and NE. 55.6) seems to be an old
motif of this story. All versions use the verb yacate more than once, cf.
55.21 ydcamdnd, 26 avalokya tatah sitdm cfydcantim aninditciim, and it
recurs as upa-ycfc- in KBlidSlsa's reference to this passage in Raghuv.
13.53 tvayci purastdd upaydcito yab so 'yah valah iyama iti pratitah. In
the different texts of the NW. and NE. recensions it also occurs thrice,
cf. 59(55).6 abhiydceta, 16 (alydcate 'dam kytcffijalih, 18 yaydce tam tat0
'bhyetya nyagrodham satyaydcanam (sic NW. !). In this passage, however,
the epithet of the tree is lacking in the Bombay and Grantha editions,
while von Schlegel (6 satydbkiydcanalz), the NW. recension (5 satycfpi
pdvitah, 18 satyaydcanam) and the NE. recension (5 satydbhiydcanah,
18 satyopaydcanam) have it in three different forms and in different
places of the episode. It would seem, accordingly, to be a later intrusion
in this passage. On the other hand, in 11. 68.16 (episode of the voyage of
the messengers) satyopay2icanam is found in all recensions. The later
alterations in the text of 11.55 may be due to the wish to add the specifica-
tion of the (Nikiila) tree found in the later episode of 11. 68.
However, satyopaycfcana- cannot originally have conveyed the meaning
which must in later times have been attached to it, viz. "fulfilling the
wishes". As an adjective, it could only mean "whose wishes are true",
which was no fitting epithet for a tree. On the other hand, satyopaydcana-
is quite clear in the sense of "Act of Truth", if taken as "solemn de-
claration of truth". The question must be raised, therefore, if the
original reading of the second p3da of 11. 68.16 may perhaps have been
krtvcf satyopaydca~zam(cf. Avad., p. 48,6!) instead of vrkjam (divyam)
satyopaydcanam. The original reading, when it was no longer understood
4
$7 -upavdcana- in Speyer's index is an error.
i
\ because of the obsolete word upaydcana- "declaration", may have been
!
altered so as to turn the substantive into an adjectival epithet of the tree.
b The analogous use in Mahgv. 111, p. 402, lines 1-568 may perhaps be
1f taken as an indication that this RgmFiyana-passage of 11. 55 had come
I to be associated, as a well-known literary reminiscence, with prayers
i addressed to a fig-tree in general. The text of Mahgv. reads as follows:
tatra nyagrodhe yadycchayd purujena zcpaydcitam, tasya ca upaydcanan;l
sanzyddhani. tena tasya nyagrodhasya mahdp17jasatkdralit kytaliz, tasya
t

i
ddni purupzsya dy~tvd'nukytim dpadyantd apardpare ca striyo ca tatra
nyagrodhe sa ty op a y d can a m ti kytvd upaydcanti, sarvdye vdrd[zasiye
tani nyagrodhak satyopaydcanato abhijn"hto, yo ca tatra nyagrodhe
upaydcati, rani tasya upaydcanam samydhyati. It is obvious that a line of
argument resulting in an accumulation of hypotheses, none of which
admits of verification, is not very attractive. On the other hand, it is
very unlikely that the word satyopaydcana- originally had the sense
here assigned to it, and the whole explanation looks like a product of
popular etymology. If so, the original meaning of the word, and the
way in which it has come to be used in this context, can only be explained
by means of some hypothesis.

The rare traces of ydc- "to declare solemnly" in Buddh. Skt., isolated
though they are, justify further investigations. The sole passages for
which Roth records an aberrant meaning occur in the AthS. They fall
into two groups, one with the meaning "to offer", and perhaps a second
with the meaning "to promise". Of the second group two of the three
passages are fully inconclusive. InlAthS. VI. 118.3 ycismd yncini ycisya
jdydm upaimi ydtiz ya'camarzo abhyaimi devdh and in 119.3 dmijdnan
mdnasd ybcamdno ydt tdtraino cipa t6t suvdmi Ludwig renders the participle
by "flehend, bettelnd", and Whitney by "begging", but the context
remains obscure, Cf. 1 dddsyan . . . san'zgyndmi. In VII. 57.1 Sarasvati
is implored to fill up with ghee ycid ds'dsd vddato me vicuksubh6, ycid
ycfcamdnasya ccirato jdndm cinu "what has gone wrong on the part of me
speaking with expectation, what of [me] going about among people
begging" (Whitney; "bettelnd" Ludwig). Since the second (and last)
stanza of this short hymn has been taken from RS. X. 113, the context
of the first stanza is, again, not clear. It was later used in a rite which
shows ydc- to have been taken in the sense of asking, but this may be
secondary. The stanza is only mentioned because it may originally have
referred to a failure to increase one's social prestige.
" Prof. J. W. de Jong kindly drew my attention to this passage.
 Less questionable is the meaning "to offer" in AthS. IX. 6, a hymn
dealing with the entertainment of a Vrgtya. It is met with in st. 4 ydd
abhivddati, dikpfm zipaiti, ydd udakdm yhcaty, ap@ prd nayati "when he
[the lord of guests] greets them, he enters upon consecration; when he
offers water, fie brings forward the [sacrificial] waters" (Whitney) and in
st. 48 dtithin prdti paiyati, hinkmoty; abhi vadati, prd stauty; udakdm
ydcaty, zid g@ati "[when] he meets the guests with his eyes; he utters
hing; [when] he greets [them], he preludes; [when] he offers water, he
sings the udgitlzd". Whitney's reasons for preferring the translation
"I ask" in XV. 13.8 asyai devhtciyd udakdm ydc&ni are not clear.
The most natural explanation of this meaning "to offer", which ob-
viously cannot have secondarily developed from the meaning "to ask", is
to take it as an independent semantic development of the older meaning
"to declare solemnly". If so, this usage alone is sufficient to show that
the meaning "to ask" cannot be the original one and that the etymology
which connects ydc- with yd- must be rejected. From a formal point of
view attention may be drawn to ydcEd-, or ydcEU-, f., which has
adopted the palatal of the present stem, while the original verbal noun
*ydk-nd- with a guttural survives in AV. [vi-ydxna-1.
A few words remain to be said about AV.vycixana-. While middle verb
forms with sam- denote in Vedic a reciprocal relation, those with vi-
express a separateness in space or time. This may simply imply an alter-
nate action, e.g. I. 185.1 v i vartete dhani cakrtveva "day and night turn
round one after the other like two wheels", VI. 9.1 hhas' ca kr~pdmdhar
drjunam ca v i vartete rhjasi vedydbhih. Cf. the active verb vi-car-, which
means "auf verschiedenen Seiten getrennt gehen" (Liiders, Philologica
Indica, p. 764) rather than "abwechselnd gehen" (Ludwig, translation of
VI. 49.3). However, this separateness may imply a hostile attitude, e.g.
VIII. 1.4 v i tartiiryante maghavan vipaicito "Es eilen in verschiedener
Weise (i.e. wechselseitig sich bekampfend) vorwarts . . ." (Thieme, Der
Fremdling im RV., p. 51), vikurute "acts in a hostile way" (Mhbh., see
S. K. De, BDCRI, VIII, p. 27). In this way we must apparently also
interpret KS. VlII. 7 (p. 90,ll) gdm ghnanti, tdm vidivyante, tdk sabhd-
sadbhya upaharanti. The middle form with vi- (cf. MS IV. 4.6, p. 57,10,
as against the parallel version gkm asya tdd dhah sabh~ydmdivyeyuh

KS.KKS. (see Ckrtel, Zur Kapi~fhala-Kafha-Samhitd, p. 62) and AthS. Also

written ydckyri-, cf. Wackernagef-Debrunner, Altind. Gramm., II/2, p. 732 f., Nachtriige
ad I , pp. 153,19 and 186,12; Wackernagel, Kleine Schriffen,p. 289; Renou, Grammuire
vidique, pp. 58, 61.
MS. I. 6.1 l , p. 104,6)'O stresses both the reciprocality and the com-
petition of this game. The variant vidivyanti KapKS. VII. 4 (p. 75,2),
Pataajali on P3n. 11. 3.60, probably reflects a later linguistic stage. Now
the same meaning is met with in several combinations of vi- with a
verbum dice~di,~l e.g. mitho vivadamdnayoh Manu 8.109, 9.250. The
Indo-Iranian word *vi-yak-na- accordingly presupposes a verb *vi
yacantai "they declare solemnly in defiance of each other". An analogous
noun formation in +a- of the Vedic language is, e.g., vi-drd-pa- (KS.),
which does not mean "awakened" (PW., etc.) but "having run in different
directions, dispersed" (see K. Hoffmann, IIJ, IV, p. 21 n. 12); cf. vidruta-.
After roots ending in lk-, however, derivatives with the morpheme -na-
are extremely rare in Vedic; see Wackernagel-Debrunner, Altirrd. Gramm.,
1112, p. 729. In conclusion it should be observed that Tokh. B. yak
"demander, mendier" (Duchesne-Guillemin, BSL, 41, p. 145) does not
prove the existence of an IE. root *yek- "to beg" since the Tokharian
word is non-existent according to Couvreur, Revue beige dephil. et d'hist.,
23 (1944), p. 233.

.Vedic viviic- and vigiidC

1. The Indian institution that can be paralleled with the Iranian
vycixman- is, as pointed out above, the sabfzd-. Besides the sabhd there
existed also the ~cimiti-.'~The circumstance that both are called daughters
of Praj5pati (AthS. VII. 12.1) shows that each of them had a distinct
place in the whole mythical conception of the world. Indeed, whenever
mention is made of two groups of sons (the Devas and the A s u r a ~ )or ,~~
of two wives (Kadrii and Vinatii, YV. Mhbh.) of Prajiipati, the number
two always reflects the fundamental cosmic dualism of upper world
*
(Heaven) and nether world. The conclusion that sabhd and s h i t i , as
the two daughters of Prajiipati, also represent the two cosmic moieties,
which again must have found its expression in their different social roles,

See however Oertel, Zur Kapi$hala-Kalha-Samhitci, p. 66, who suggests (like

Caland, WZKM, 23, p. 53) that the authentic reading may have been vidivyeyub.
See also Delbriick, Altind. Synt., p. 178.
l' See Delbriick, Altind. Syntax, p. 465 f.
See Ludwig, Der Rig Veda, III, p. 253 R.; W. Rau, Staat und Gesellschaft im alten
Indien, p. 82.
7a PB.XVII1. 1.2, SBM.II. 1.1.8; 1.2.13,XIV.4.1.1,etc.,BAU. 1.3.1, ChU. 1.2.1,

KS. IX. 11 (p. 112,8), MS. I. 9.3 (p. 132,16).

 seems inevitable. Ludwig conjectures that the sdmiti was especially
reserved for the VG-,which could be understood if the sdmiti was con-
nected with the nether world.74 As for the sabhd, most details are still
obscure. It would seem rather certain that the place where the social
contests were fought was a'sacred one,'&if the social contest here fought
was a replica of the primordial fight between the Devas (Indra) and the
Asuras. It must be admitted, however, that the actual evidence of the
briihmana-texts does not allow us to prove this. See Rau, op. C., p. 75 ff.,
but cf. Held, The Mahabharata, p. 202 ff.
Apart from all conjectures, however, we may state that just as the
Iranians prayed for a son who was valiant and vyrixana-, so the Indians
longed for a son who was ~ a b h i y a - . The
~ ~ well-known prayer of the
AS~arnedha~~ runs as follows: d brdhman brdhmapd brahmavarcasl'
jdyatcim, d 'smin r a t r e rdjanyd i~avydhidro mahdrathd jdyatdm, ddgdhri
dhenh, vddhd 'na(?'Ydn, ds'zih sdptih, pdrakdhir ydsa, jispd rathe~fhdh,
sabhkyo ytivd, 'syd ydjamdnasya vird jdyatam "In this kingdom may a
prince be born, an archer, a hero, and a great car-fighter; a milk cow, a
draught ox; a swift racer, a pzira~dfii,a victorious warrior, a sabhiya,
may a hero be born to this sacrificer". The true meaning of sabhkya-
was no longer fully known to the authors of the brghmanas, which
suggests that the term belonged to a form of society that had ceased to
exist. Cf. SBM. XIII. 1.9.8 e ~ dvai sabhiyo ylivci ydh prathamavayasf,
tcismdt prathamavayasf striqdm priyd bhdvukah, TB. 111. 8.13.1 yd vai
pzjrvavayasf sd sabhkyo ytivd, tdsmcid ylivd ptimdn priyd bhdvukah and
such commentaries as Mgdhava on RS. I. 91.20: vidathydm sabhkyam,
viz. yajiiasabhayos' ca yah svakirti, S5y. ibid.: sabhdydni sddhum, sakala-
idstrdbhijiiam ity arthah, idem ad TB. 11. 8.1.3: sabhdydk sddhuh,
Mahidhara ad VS. : sabhciydm yogyah (Keith: "a youth fit for the assem-
bly").
It seems to me that Zimmer's rendering by "schlagfertig" is still the
nearest approach to the meaning which must have been attached to the
word. Rau, in his important discussion, reverts to "gesellschaftsfahig,
gesittet" (PW.). In the first place, however, it should be observed that
TB. and SBM. differ in so far as only the last text adds stri~dmto priyd
bhdvukah, which induced Eggeling, SBE. XLIV, p. 295, to translate

74 See however Rau, op. C., p. 82 f. [Note Ap. Dh. SO. 11. 25.5 dak$i@empuram
sabhd! 1
See Held, The Mahabharata, D. 233.
See Ludwig, op. C.,III, p. 254 f.; Zimmer, Altindisches Leben, p. 174; Rau, op. c,.
p. 77.
sabhtyo ytivd as "a blitheful (or sociable) youth". Evidently the word
sabhdya- must have meant "member of (and probably, successful speaker
in) the sabhd". Now, what a speaker in the sabhd wished most of all
appears from AthS. VI. 12.2 t t me santu sa'vdcasah: it was the assent of
all members. This is also what the S8nkh. Ar. 8.9 expresses in the words
sa ya evam etdk daivdm vipdljZ veda Srutavadanatamo bhavati, bhlimiprd
'sya kirtir bhavati, SuSrGjante hd 'sya parjatsu bhdj[y]amcipasye 'dam astu
yad ayam ihate, yatrd 'ryd vdg vadati vidur enam tatra (cf. Ait. Ar. 111.
2.5). In JB. I. 27118 the various aspects of social prestige are expressed
by the words priyah, Srej$zah, grdmanilz and kirtih (cf. jyejfhai ca ha vai
besthas' ca svdndk bhavati ~ 8 n k h .Ar. 10.2). Such a speaker may be
said to be priyb bhdvukah "disposed to be popular". As it seems certain
that these words had no erotic connotation at all, the question may be
raised if the curious addition strinhm in SB. is perhaps due to the pre-
ceding gloss on the words ptiramdhir ybqd, viz. yojity evci rapa'm dadkiti,
thsmdd rapi?li yuvatih priyd bhdvuka "whence the beautiful maiden is apt
to become dear (to men)".
It is not possible to enter into a discussion of Rau's arguments, but we
should like to point out that in the formula of the Yajurveda ya'd grhme
ycid a'rapye ya't sabhdydm ya'd indriyt / ya'd PnaS cakymd vaya'in the word
sabhdydm hardly means "in public" (in der Offentlichkeit) but rather
denotes something apart both from the public life in the village and from
the forest. Since the opposition cannot be one between "public" and
"private", it may.be suggested that the reference is rather to the sacred
character of the sabhd and all that is performed in it, in contradistinction
to the profane life. Nor am I convinced that TS. 111. 4.8.5f. proves the
sabhd to have been a "cow-shed", although it is certainly difficult to
account for the tfncni in the sabh~i."~ From I. 7.6.7 paicivah khcilu vai
brdhmapa'sya sabhd "Vieh ist ja fiirwahr die sabhd eines brdhmana" we
can only infer that wealth of cattle and horses affords the right to build
one's own sabhd: a'va sabhdriz runddhe, prci sahcisram p a S h dpnoty, d 'sya
prajdydk vdjf jdyate (vdjin-: "winner of a potlatch"? see p. 240). Our
study may further have shown that "Gewandtheit in der Versarnmlung"
can indeed have been an ideal of the ancient Indian education (Rau, p.
77). These remarks do not detract from the value of Rau's very in-
structive book.
In addition to the preceding remark on sabhdya- in the brghmanas it
may be observed that in the Yajurvedic prayer the son is hoped to be a

For COWS in the sabhd see p. 278.

 201

ieIding cow (dbgdhri dhenzih), a draught horse (vddhd

a race horse (dStih s i i ~ t i h ) . ~The
~ b first and third of these
e young rdjanyd- as a lord who gives rich dak;cin& and who
ng a race was indeed a means of deciding a dispute.78
o win the goods of his opponents as a good racing
horse. The same idea is then expressed in a more direct way: while
j i ~ ~ rathe~thcfh
zi and sabhkyo ylivri obviously refer to his victoriousness
in chariot races and verbal contests, pliralizdhir yd~richaracterizes him as
a munificent young lord. Pziralizdhi-, indeed, the opposite of arriti-,
means "present-giving" : attempts to find other meanings or connotations
such as "poetical inspiration" will prove misleading. The young
lord is conceived of as personifying some ancient mythological figures,
the cosmic Cow and the "Woman Present-giving" (just as Usas is said
to have become a dhenzi- for Yama, TS. IV. 3.1 1). In this connection it
may be useful to compare the so-called diidk kl'ptayah (AthS. X X .
128.1-5, GB II. 6.12: p. 263,8), the first stanza of which runs as follows:
ydb sabhiyo vidathydh szitvd ydjvri ca pdru,~ah, while st. 5 reads y t ca
devbm dyajanta 'tho yk ca parddadzib / sdryo divam iva gatvdya maghdvdno
virapianz'e. Munificence and victoriousness (in battles as well as in word-
duels) are the characteristic traits of the ideal young hero in many
archaic civilizations. In the Anglo-Saxon gnomic verses79we meet with
such stanzas as
geongne ejeling sceolan gdde gesidas
byldan td beaduwe and td bkahgife
"good comrades should encourage the young lord to battle and to ring-
giving" and the BCowulf resumes the Old Germanic ideal (which was that
of the Homeric heroes) in the words (287 E.):
dgh wadres sceal
scearp scyldwiga gescdd witan
worda ond worca, s i je wel jenced
In quite the same way Bhartrhari 14 (ed. Kosambi 1948) still opposes
sadasi vdkpatutd to yudhi vikramah. Since similar phenomena are met
with in non-Indo-European societies," the concordance between the
testimonies of Old Indian, Greek and Old Germanic texts does not in

See however the quite different translation of this passage by Oldenberg, Rel. d.
Veda, p. 370.
Oertel, T r a m Conn. Acad., 15 (1909), p. 174; cf. JB. I. 213&,AB. IV. 7.4, KB.18.1.
H. Sweet, An Anglo-Saxon Readers, p. 168 f.
See H. Scharer, Die Gottesidee der Ngadju-Dajak in Siid-Borneo, pp. 46, 89.
 itself prove the proto-Indo-European character of these phenomena,
although they are likely to reflect an aspect of the proto-Indo-European
civilization. Cf. M. Mauss, $ociologie et Anthropologie, p. 242.
The interpretation of sabhtya- as "a successful speaker" may find a
support in JB. I. 261-262, where it is stated that the son of an udgdtr-
who knows the identity of manas- and retasyi- is mahdmand manasvi, the
son of him who knows the identity of cakw- and tr@ubh-, or of irotra-
and jagati-, is ahritamukhi paiyo a'r~iyii.. .darSaniyah, resp. iuirzip~i
Srotriyepa. . .iravapiyab, while it concludes with the words atha yo '%g
a n ~ t u b "iti vidviin udgdyati, iasto 'dgdtd vdco 'rCirddhy(?) asmdd djiiyate
sabheyah "To him who knowing that VHc is identical with the anugpbh
sings as an udgdtr, a son is born who is a reciter (read SamstZ?), a chanter,
who accomplishes with his word (read updrdhy?), and who is a sabheya-."
With this interpretation is also in harmony RS. 11. 24.13 sabhkyo v@ro
bharate matf dhrinii "Als Wortfiihrer in der Versammlung beliebt tragt er
[viz. Byhaspati] durch Einsicht die Gewinne davon", where Zimmer's
rendering "ein schlagfertiger Sanger" might be amended into "ein schlag-
fertiger Redner" (for vkra- cf. vipdliz jydtim~i,etc.). The same meaning
suits I. 91.20 sdmo dhenlim sdmo cirvantam didm sdmo virdh karmanydliz
daddti / sddanydm vidathydliz sabhdyam pitrircivapam yd dciddiad asmai
(see Ludwig, Der RV., 111, 254). For the juxtaposition of vidathyd- cf.
AthS. XX. 128.1 quoted above.

2. Vedic vivdc-.
The Vedic verb most closely corresponding to AV. vi-yiic- is vi-vac-,
which means in the middle forms "to dispute with one another about
(loc.)". Cf. RS. VI. 31.1 v i tokt apsd tcinaye ca sdrt 'vocanta car~apciyo
viviicah "Die Stamme sind um den Samen, um Wasser und leibliche
Nachkommenschaft, um die Sonne streitend in Streit geraten". Like the
middle of vi-vac- the verbal noun vivdc- is (apart from three occurrences
in AthS. XX) exclusively attested in the Rigveda. Since this is also the
case with all other terms of the potlatch-ceremonies, we are driven to the
conclusion that the Rigvedic society differed considerably from the later
forms of social life as reflected by the other Vedas. Such terms as
sabheya-, which survive in later Vedic texts may have come to refer to
different situations in which a man could manifest the power of his
words. The meaning of vivdc- was according to Grassmann "Streit,
Schlacht, Wettkampf", which was more correct than "widerstreitender
Ruf, Streit" (Roth, Wackernagel, Altind. Gramm. 1111, p. 261). Since
vac- does not mean "cry, scream, yell", the suggested meaning "opposing
 ams) must be ruled out and only "contest, fight"
l

l The question may be raised, why the archai dic forms of social
l
l life have left so few traces in the later texts. This ssibly be connected
1 ndamental changes in the Vedic religion. If our conclusion that
1 vedic religion was centred about the rites of annual renewal at
i the winter solstice is correct, this form of religion inherited from pre-
i Indian times must soon have been replaced by another ritual more in
harmony with "1'Inde des moussons"'. On Indian soil the ancient myth
1
i
l
of Indra slaying V ~ t r a and
, the winter ritual celebrating this event, were
bound to lose their meaning. This, however, is a hypothesis which only
i
profound investigatio into the development of the Vedic cult might
prove to be correct. e may here confine ourselves to the statement
1
. TB.), vivdcana- ( BK., AA. JUpB.),
. JB) "decision, authority", vivalcty- "Verbesserer" are
l
I
all based on the active form v i vivakti "he declares, corrects", while the
middle 3 Du. v i *vivacdteao5"they dispute with one another" and its
l verbal noun vivdc- are confined to the oldest texts.
l
In the first place vivdc- refers to Indra's cosmic fight. Cf. 111. 34.10
indra dsadlzir asanod &dni vhaspcititi52r asatzod antcirik~am/ bibiztda valciliz
!
! nunudt vivdcd 'thdbhavad damitd 'bhikratfiildm. The whole hymn extols
i Indra's cosmic function: he is a "Finder des Gutes" (st. 1 vidcidvasulz), a
1 winner of the sun, who found the light (st. 4 svarsa'h, civindajj-vdtitz) and
gave inspiration to the invoker (st. 5 cicerayad dhiya irnd jaritrt), who
l
has won Earth and Heaven (st. 8), the race horses, the Sun and a wealth
i of gold, while slaying the dasyus and assisting the Aryas (st. 9). Now
l the frequent juxtaposition of the words vytra'tzi and dcisyavah, Scitravah,

l l
amitroh in the Rigveda points to the conclusion that the Aryans con-
ceived their battles against the Dasyus as a reiteration, a replica, of
Indra's mythic fight. Although it is advisable in principle to omit all
1 reconstructions of possibly underlying ideas which are not explicitly
1
1 stated in our texts, the fact that the Rigveda often associates the ideas
I
of war and contest allows us to conjecture that war, no less than present-
! giving, contests and chariot-races, was thought of as stimulating the
creative forces of the Cosmos during the winter ritual. In this connection
L

I
80a The same semantic relation exists between vi-braviti "he declares, explains" and
vi bricte, which functions as the Rigvedic present of vi vac- but is no longer used in
later texts (sole occurrence in VI. 25.4 toke' vd gdju i'a'naye ycid apsri vi krandasi urva'rc?sisu
bra'vaite, cf. VJ. 31.1 vi toke' apsri fa'naye ca she' 'vocnnta car~~a~zdyo vivc?cah!).
 on "The Serpent in Kwakiutl
n", p. 8 1 f. may be t if the giving away of copper
pieces is a sacrificial act, it follows that the whole exchange traffic of the
potlatch is, really, nothing but one great sacrificial rite. . . But then
this act was also a rebirth rite, a passing on to a new phase of life.
.. . This notion of passing on to a new phase of life we also found in
the winter ritual. As a matter of fact, the very principle of the potlatch
is intimately bound up with the root idea of the winter ritual. Just as
light and life, so also honour and wealth are acquired through destruction
and ruin. . . Besides the conception of sacrifice, also the idea of combat
belongs to the most essential elements of the potlatch complex. Both
otlatch and war are under the sway of the serpent and may even replace
each other."
This explains why the poet of 111. 34, after mentioning the battles
between Aryans and Dasyus in st. 9, immediately passes on to their
mythic prototype in the next stanza. The masculine verbal noun vivcic-
must have denoted the opponent in a word duel but since Indra's fight
was the prototype of the ceremonial contest, he, as a darnitd'bhikratGncim,
could be said to have dispelled his vivcics and thereby to have won the
day-light (dhdni, cf. svarss'h in st. 4) out of the darkness of the primordial
Chaos. The Old Aryan winter ritual was indeed a mere replica of the
primordial vivdc- between the cosmic powers of nether world and upper
world, and that is why the prize of the battle between the carsapriyo
vivcicah in VI. 31.1 (see above) could -be said to be virility, progeny,
water and S ~ n . ~ o b
ithout entering into a discussion of the word carsapi-, which would
equire a separate study, we proceed to consider VI. 33, which invokes
ndra's assistance against Dgsa as well as Aryan opponents (st. 3 ubhriycihi
arnitrcin, ddsd vytrdpy s'ryci ca). Stanza 2 runs as follows tvdli? hindrd
'vase vivdco hdvante carqapdyah iu'rascitau / tvdm v9rebhir v i papfhir
ahiyas tvtita it sdnita vdjarn drvci "Denn dich, Indra, rufen die streitenden
Volker m m Beistand (im Kampf) um den Heldenpreis. Du bist durch
die Redekundigen mit den Pani's (Knausern) fertig geworden. Von dir
begiinstigt gewinnt der Renner den Siegespreis". Instead of "Helden-
preis" we may rather interpret s'u'rasdti- as "obtainment of valiant sons",
cf. VI. 31.1 tdnaye, V1I. 23.6 virdvad, etc. While here the fight against
the Papis is apparently put on a level with chariot races, the cosmic
'Ob A classical instance of this cosmic vivcic- which founded Indra's new status is RS.
XV.42. For a different interpretation of the "assauts de vanteries ou d'insolences"
between Indra and Varum see Dum6zi1, Les dieux des indo-europkem, p. 83.
importance of such fights stands out more clearly in VII. 30.2 hdvanta u
tvci hdvyam vivdci tand~uSi&db sihyasya sdtali / tvdm viivesu sinyo
jdnep tvdm qtrhni randhayci sufza'ntu "Dich rufen sie, den Rufenswerten
im Wettstreit, die Helden (im Kampf) um ihre Leiber, um die Sonne zu
gewinnen. Du bist bei allen Volkern der Heeresgott. Gib die Feinde in
(unsere) Gewalt, dass sie leicht zu erschlagen seien." If we are ready to
admit that the Vedic poets were speaking about something definite and
did not merely repeat ancient poetic formulas devoid of real sense, these
words can only be taken to refer to a winter ritual, a ceremonial contest
inaugurating the new year. Similar allusions to contests associated with
thk winning of the sun are frequent, cf. e.g. VI. 46.4 asrndkam bodlzy
avid mahddhani ta&v apsk su'rve "Sei uns Schutzer im grossen Kampf
urn die eigenen Leiber, um Wasser und Sonne!". The same situation is
also suggested by the context of VII. 23.2 dydrni g h d ~ aindra deva'jdmir
irajydnla ydc chwlidho vivdci / nafzisvdm dyus' cikiti jdnep t&nr'ddnihdrnsy
dti parFy asmdn "Das den Gottern vertraute Geschrei hat sich erhoben,
o Indra, da iiber die Gewinne im Wettstreit entschieden wird; denn
unter den Menschen wird die eigene Lebensdauer nicht gekannt. So
hilf du uber diese Angste hinweg!" The exact meaning of Surkdha
irajyante; obviously a technical expression of the potlatch terminology,
is unknown. The active form irajydtz X. 140.4 is translated by "waltend",
"anfuhrend", "freischaltend", and the adjective irajyti- by "gebietend".sl
Therefore Oldenberg's translation "Gelenkt worden ist das Geton, d e n
die Gotter verschwistert sind, o Indra, als sich die Guter (Speisen?) zur
Verfugung stellten beim Streit der Reden" seems more correct. In any
case some inferences seem sufficiently certain. There is a verbal contest
(vivdc-), this time between priestly poets (Vasi~iha,st. l), which is likened
to a real battle and therefore called a samaryd- (st. 1); the brdhmdni have
arisen in a strife for social prominence (Sravasyd-, st. l), the prizes
(Srtrkdhah)have been put up, and at this very moment the poet's thoughts
are struck by the idea of Death, of a crisis which has to be overcome.
Since there must be a natural association between this contest of priests
and the crisis (cimhas-), and since the notions of dmhas-, duritd- are
closely linked up with that of tdmas- (see p. 221), we are again driven to
the conclusion that this vivzc- took place at the inauguration of the new
year. The Vasi~fhasimplore the assistance of Indra (6ab evid indrariz
vrj.anam vdjrabdhu~zvdsi~flzdsoabhy drcanty arkaih), who is expected to
bestow the prizes on them in reward of their praises, cf. 4d tvdrn hi
Cf. Grassmann, Worterb. zum Rig-Veda, resp. Geldner, Glossar, S.V. and fiber-
setzlmng, a.1.; Neisser, Zunz Worterb. des ljgveda, I , p. 163.
dhibhir ddyase vi v an, 6c sd nu stutd virdvad dhdt
ib) in st. 2 is more likely to b
o by being sumdnasah cont
"das Geschsei um Wilfe7' (
ortance of gaiety (see p.
52.5 vis'vaddtziiiz S

dno vuydh ca. Attention may be drawn to

sizes (vhjiin, gdmad), which apparently were
referred to in st. 2, mention is also made of a
vircivad (sc. dhdnnarn), which may be equated to td~zayeVI. 3 1.1, s'drasdtau
VI. 33.2.
mentioned in VI. 45. p~rtit2
hir vcijayat&m "Zu ihm, ter vie
anger, die urn die spreise wetteifern".
hough being a vivdc- of priestly poets, this was not a mere poetical
contest. In the preceding hymn of the same poet there is again a reference
to the deliverance from rir32has- an
vy dsmdd dv&o yuydvad vy riniha
l'cntzyasya je& indra siirfn
rei'oahn und gute Fahrt, . . .
en zu gewinnen, so nimm fiir unsere Gonner
45 the poets' contest and the chariot races are
apparently two forms of one and the same ceremonial strife. Cf. 12
dhibir rirvadbhir drvato v@d& iizdra s'ravdyy&i / tvriyd jejma hitcini dhdnam,
14 tdyi (scil. zitf) pzo hinuhi rdtham, 15 sci mrjthenu r a t h b m o 'smtfkend
'bhiy~igvand/ jisi j i g 0 hitririz dhdt~arn. ost clearly this parallelism is
expressed in st. 2 avipri cid vdyo dddha nds'dnd cid rirvatii / indro jitii
hitd~zdhcilzam "Auch dem Unberedten verleiht er hwung, auch mit
dem langsamen Rempferd gewinnt Indra den ausge en Preis". As a
matter of fact, Indra grants his s ort in both forms of the contest
because he presides over it and de s it (st. 20 sd h i vis'vdni pdrthivd6
e'ko vhszini prjtyate). This is the r why both forms of support are
sometimes mentioned in one and the same stanza, e.g. I. 178.3 jitd
ny'bhir indrab prtsli s'u'ruh Srdtii hdva?h ndd/~amEtzusyakdrdh / prrjbhartd
dyantii giro yddi ca trndnB bhdt "
eger in den Schlachten; er erhor
oten befindlichen Dichters, Er bringt den agen des freigebigen
Opferers voran, ihm dicht zur Seite (bleibend), und er erhoht die Lob-
Bdhava: stutiSaDtiaE; Sgyaua: stufiriipafz Subdab.
reden, wenn er in eigner erson erscheint".
"Qpferconcurrenz . . . in das ttrennens gekleidet" (Geldner,
Ved. Stznlieii, 11, p. 163) but a reflex of the different aspects' of lndra as
the god presiding over the winter rituai. As ointed out above (p. 237),
the poets do not act on their own account but on behalf of their patrons
(sari-).As the contest was between parties, the poet who was victorious in
the verbal contest in the sabfzd was a kilbi~aspr'tfor his party (X. 71.10).
The sank idea recurs in 1. 178.4 et]$ ntbhir i~zdrahsuiravasya'praklzddd&
pyk~dabhi nzitrino bhGt / vivdci satrdkard ydjamdnasya
Sci~ksah"'Also iibertsifft annen im Drang nach Ruhm
als starker (Feinde)vertilger alle, die Freundschaft haltcn.
ettstreit um den Speisegenuss (u'liksipii!) wird er gepriesen als der, der
ganze Arbeit macht, er des Opferers Preis!" Although several details
remain obscure (e.g. rnitrivo), the parallelism between ndra's sui~'avasyd-
and the i ~ dvivdci seems obvious.
In connection with our suggestion that vivdc- ceased to be used in
later Vedic texts because the potlatch-like type of contest had become
obsolete, it might of some interest that the only occurrence in the
tenth book of the gveda could be interpreted in a different way: if
so, its use would have to be regar as a reminiscence of the older
literature, whose true meaning was no longer known. In X. 23.5 yd vdcd
vivdco mydhrcivdcah p u d sahbsis?.d'iivd jaglzdna / tcit-fad id asya pali&syariz
grnitnasi the possibility that the word has been taken in the sense of
vivdcas- (AthS. XII. 1.45 Jrinam . . .bahudhd viviicasam "people of
different speec;,") cannot be entirely ruled out. Probably. however, it is
also here used in its ccwmon meaning. Geldner translates: "Der mit
ort die Wi8-rredenden, die Missredenden, die vielen
tausend Feinde erschlug. Dies:: und jene Mannestat von ihm besingen
wir . . .". The words vdcd. . .jaglzdtia are an interesting formula charac-
terizing the cosmic fight as a word du4. The Atharvavedic occurrences
of viviic- are literal quotations from the Rigveda, viz. XX. 11.10; 12.2;
73.6 corresponding to RS. IH. 34.10, VII. 23.2 and X. 23.5.

3. Instead of Rigvedic vi vac- later texts (since T

in the sense "to be at variance, contest, litigate, dispute with someone
about (loc.)". Thz derivative vivdda-, m. (since SvB. V. 3.2) means "a
dispute, quarrel, contest", later particularly "a contest at law, legal
dispute, litigation, lawsuit". The word vivdc-, as a technical potlatch-
term, had died out with the institution it denoted.
On the other hand the hupax legomerzoiz vigadd-, which is a word of the
Rigvedic language, seems still to reflect the older form of social life. It
occurs in X. 116.5 ugrdya te sdho bdlam dadclmi pratityci Sdtrzin vigadkp
v&a. Grassmann rendered it "etwa Schlacht" Bohtlingk, pw, VI, p. 81,
"Geschrei, ein Durcheinander von Rufen", while Geldner in a note on
his translation remarks: "wenn von gad 'sprechen', eigentlich Wortstreit
= viviida, vivlic". He accordingly translates the stanza: "lch gebe dir
[viz. Indra], dem Gewaltigen, Uberlegenheit und Kraft ; tritt den Feinden
entgegen und zerhaue sie im Streit". Renou, Et. vkd. I, p. 19, who renders
the last words by "dichire-les dans les disputes" concludes : "l'accent
portait donc bien sur la joute oratoire, non sur le combat sanglant".
This interpretation, however, involves some difficulties from a linguistic
point of view. The verb gadati, a rhyme word of vadati and presumably
originally peculiar to colloquial forms of speech, is not attested in Vedic
texts before &VS. SFinkhS. K&. (also in P i i n i ~ ~ i Now
) . ~ ~ it is certainly
possible that in lower social strata a new word vigadd- for vivcic- has been
created, since we know from AthS. VII. 12 that debates in the sabltii
were still practised at that time. On the other hand Thieme was quite
right in pointing out that the only word gada- that is actually known
from Vedic and Avestan means "disease": RS agadd- "without disease",
AthS. *vigada- "surrounded by diseases" (later vigada- "without disease"
&S. 19.90); see JAOS, 77 (1957), p. 53 n. 1. Hence a determinative
vigadi~u"in diverse diseases" might seem "at least possible". It must be
objected, however, that in such compounds vi- means "trennend", "sich
ausdehnend" or "abweichend" (Altind. Gramm. 1111, p. 261). Thieme is no
doubt right in rejecting the idea of a "joute omtoire" in this stanza, but
when he concludes that "All cues point to a battle of Indra, strengthened
by the poet's words (c), against the [evil spells of] sorcerers or demons"
this is in full harmony with the assumption that vigadd- is a synonym of
vivrlc-, which also denotes Indra's fight with demons. Its occurrence'as a
hapax legomenon in the tenth book might be explained from its colloquial
character, its absence from the later Vedic literature from the circum-
stance that this kind of ceremonial word-duels soon went out of use in
the post-figvedic period.

The word narmd- is only attested in VS. 30.6; 20. The later language
from the Mah5bhBrata downwards uses narman- instead. None of the
MS. IV. 9.12 (p. 133,lO) jagdda is a corruption of jagdra RS. X . 55.5 (etc.).
etymologies proposed for it carries conviction. Whitney posited a root
ny- "to sport" (Roots, p. 92), but its meaning has been deduced from the
very words which this root was intended to explain. However, his
suggestion that nyt- "to dance" is a root-extension of ny- was accepted
by Persson, Uhlenbeck, Walde-Pokorny, and kernagel-Debr~nner.~
On the other hand, Persson compared the G nic word nar with it,s5
and in Pokorny's Idg. vergleich. Worterbuch, p. 975, narrnh- and ny'tyati
"dances" are derived from a root *sner- "drehen, winden" (see Uhlen-
beck, etc.).
Now this last explanation cannot be correct. The primary meaning
of nyt- seems to have been "to manifest one's strength" (cf. nr~ntzhzica
n y t h d n o bmartah V . 33.6) and in corrobo ion of the assumption of an
IE, root *H,ner- "vital strength" both rgenstierne (in a personal
communication) and Bailey, JRAS, 1953, p. 105 f., pointed to the existence
of an Indo-Iranian verbal root n p , surviving in PariiCi nar- "to be able"
and perhaps also in Osset. n&sun "schwellen, zunehmen, aufgedunsen
werden". A derivative from this root is the Rigvedic verbal noun nyti-,
which seems to denote a manifestation of strength or vitality. 16X. 29.2,
it is true, it remains doubtful whether the word refers to the victorious
reappearance of Usas (as a manifestation of her sC-nftri- "vital ~ t r e n g t h " ) , ~ ~
or rather, on account of the assonance nytazi sydma nftamasya nmhm, to
Indra's exploit which renders her reappearance possible. But its general
meaning is not uncertain in X. 18.3, where the relatives of the dead go
back after the burial ceremony to reinforce the vitality of the living:
pr&co agdma nytdye hdsdya drdghiya dyu[zpratardrn dhdhdndh "'Wr sind
dem Tanz und Lachen entgegen gegangen, unser Leben noch weiter
verlangernd" (Geldner). The usual translation "dance" is based on the
secondary meaning of nyt-. Now it is possible that the root ny- has also
secondarily expressed the same notion but the context does not provide
any clue as to the exact meaning. Reinforcement of thC vital strength of
the community after the death of one of its members could be accom-
plished in different ways. Among the Anglo-Saxons, for instance,
boastful talk (gilp) was one of the most effective means of re-establishing
the vitality of the group (V. Grmbech, Menneskelivet og Guderne, p. 69).
84 See P. Persson, Studien zur Lehre der Wurzelerweiterung und Wurzelvariation
(1891), p. 63 f., Uhlenbeck, Xicrzgefasstes etym. Worterb. der altind. Sprache, p. 1
Walde-Pokorny, Vergl. Worterb. der idg. Sprachen, 11, p. 333 ;Wackernagel-Debmmer,
Altind. Gramm., 1112, p. 750.
See Falk-Torp-Davidsen,Norwegisch-Danisches etym. Worterb., p. 1521.
BB Mededelingen der Kon. Nea'erl. Akad. v. Wetenschappen, Afd. Lett., NR. 14.15
(1951), p. 218.
Hence "confirmation of one's vitality" (against the threatening power of
Death) is likely to have been the general meaning of nrti-, which also
suits the context of X. 29.2. The word may have had more concrete
connotations but the available evidence does not allow us to go beyond
this general statement.
Ved. narmci-, obviously a derivative from the same root, proves that
the meaning "to dance" was not inherent in the root. It also allows us
to discard the proposed connection with IE. "sner- "drehen,
It is likely, indeed, that this word refers to s$a?ctt. In the te
Puru~amed/zawe read (V . 30.6, TB. III. 4.1.2) nrttdya sat&& gitdya
bikya'riz, dhcirmliya sabhd rd*, nariqfhwai bhimakiliz, nar~nkyarebhdh,
hdsLi_vakdrim, etc. It is the rebtzd-, accordingly, that is particularly
ted with narmci-. Since the use of rebhd- is confined to R§., A
GB.) and VS. (= TB.) and the word does not survive into the later
language (where Rebhila-, proper name of a singer and a sdrthavdha,
~cch.111. 2.5, 3.8, . 25.6) it is not surprising that the
commentaries did not know it. hidhara explains iubdakartiirarn
viicd#atn, -SZiyana has medtzdvinak tikuialam, and Roth followed
them in taking " lauderer" (prattler, chatterer) to be the
meaning here. It is not obvious, however, that we should here depart
from the common meaning "panegyrist" which the word has in the Vedic
texts. In RS. I. 113.17 the rebtzd- is likened to a charioteer and is said
to arouse or stimulate the dawns with the reins of his word (syhanti
vlicd zid iyarti vcihni stdvlino rebhd wdso vibhc7te. Cf. also I. 127.10,
VI. 3.6, VII. 63.3). His name is explained by VII. 76.7 e$d netrf rdd~zaiah
st2ny'tdndm U J ~uclzcinti ribhyate v d s i ~ ~ l ~ a Since
i h . ~ ~the rebhd- as such
manifests Indra's nrti-, the narmci- "manifestation of strength" associated
with him is likely to refer to his vac- and to denote his "powerful speech".
Not always, however, must the notion of speech have been necessarily
implied. In VS. 30.20 narmdyu pu&icalz%z, where the context allows no
inference as to the exact meaning, the possibility of a different application
cannot be ruled out. On the other hand, the abusive talk between a
pumicali- and a brahmaecirin- required by the ritual of the Soma sacrifice
(cf. Dr2hy. XI. 3.9-10 and see Caland's note on A ~ s .XXI. 19.5) may
have been important enough to justify the association of the puthScal6-
with narmd- in the sense of "abusive language". Only in the verbal
contests, however, could this word, whose original meaning cannot have
differed very much from that of npnpb, n. "act of strength, heroic force"
For ribh- "strahlen, glanzen" see Oertel, Trans. Conn. Acad., 15, p. 176 n. 1,
p. 62.
Zur Ilhpi~rjzala-Katha-Sakhita,
have acquired this connotation. A similar, though not identical, semantic
development must be assumed for narman-, n. "joke" of the later
language,87"whose origin must also be sought in the bragging contests of
ancient Aryan society, comparable to the gyipcwide, of the Anglo-
as well as to the ~ h ~ o X aofi the Homeric heroes. Cf. also Zimmer,
Aliindisches teben, p. 173. The manifestation and confirmation of one's
power and social prestige comprised the ridiculing of one's opponents
(p. 253, 2581, which accounts for the sema c difference between izarrnan-
and nymnci-, as it does for that betwee at. ioczcs "joke" and Umbr.
iuko- "prayer7'.

A confirmation of this explanation is perhaps provided by another

derivative of the same root, viz. naif$@-, which occurs twice in th
One of these passages, XI. 8.2 has6 nari~fdnyttbni SLir
(which reminds us of nythye hhsciya quoted above)
"laughter, sport, dances entered into his body after
. 12.2 vidma' te sabhe r2dma, nari&n nii'nfa vci asi, has
know thy name, O assembly; 'mirth' verily is thy
" (Bloomfield, SBE, 42, p. 138). However the last hymn, which
field defines as a "Charm to procure influence in the assembly"
has very little to do with mirth. The poet invokes the joined aid of sabh
and stimiti, the two daughters of Prajiipati, he hopes to speak agreeably
and prays that all assembled may utter speech in harmony with him
( t C me saiztu sdvdcasa!~, cf. saj6&, RS. IV. 5.1, Renou, Et. vkd., 11, p. 5 9 ,
he takes the power (or energy, vcircas-) and discernment (vijn"bna-) from
all people present. This again has a close parallel in Old Germanic
literature, see Grsnbech, op. C., p. 64 f. *Fromthe context we may infer
that nari@7-was a special, perhaps even a secret, name of the subhd as
the place where the contest was decided. Perhaps his knowledge of this
name secured the contestant the effective assistance of the sabhd itself in
his contest (which assistance had been invoked in the preceding stanza).
In any case, "Geplauder, Scherz", "sport" or "mirth" is not the meaning
we should expect in this connection. Nor does loomfield's note (S
42, p. 544) to the effect that the sabhd was also place for gaming an
social intercourse seem an adequate definition of the purpose to which
this place was put. The function of gambling in the combat for social
prestige has been amply discussed by Held, The Muhabharata, p. 243 ff.
This function corroborates our impression that the "social intercourse7'
Already in narmakiimycZ JB. D. 259, Dr. 270 "urn eines Scherzes willen" (K. HOB-
mann, MiiSS, 10, 1957, p. 64 f.)?
 at times rather the character of an embittered combat.
The most eloquent witness in this connection is the sapatnaghnak sGktam,
RS. X. 166, a prayer for divine assistance to overcome the opponent in
the word duel (cf. 3 vdcas pate ni sedhemdn ydihd mdd ddharak vdddn,
4. d vas' cittdnt d vo vrafcim d vo 'h& sdmitini dade). Bloomfield refers
to RS. VI. 28.6 br/zdd vo vdya ucyate sabhdsu, but also these words,
addressed to the dak~iqd-cows,may rather mean that the prestige of the
yajamlina is enhanced when his cows are praised in the sabhd. See in
general Bloom6eld, The Atharvaveda, p. 72, where he discusses the
Atharvavedic hymns aiming "at success, prominence or superiority in
social and political life, favor among men, influence in the assembly,
success in debate and subordination to one's will".88
What one hopes to obtain in the sablzd is power. Indra himself enters
the sabhd attended by the same vdyas- that the dak~ipci-cowswere praised
for (RS. VIII. 4.9 Svdtrabhc$d vdyasd sacate sddd candrd ydti sabhdrn zipa),
and the victorious man boasts of his being unassailable like Indra
himself (X. 166.2 ahdm asmi sapatnahindra ivdrijto dkyatah / adhdh
sapcitnd me pad& imi scirve abhi~fhitiih).It would seem a reasonable
conjecture, therefore, that the sabixi was conceived as the deity that
presided over the contest (AthS. VII. 12.1 sabhd ca md sdmitis' cdvatcim)
and as such was endowed with, and personified, the nariytii- "manifes-
tation of one's social prestige". Whoever knew this (secret?) name of
the sabhd could be sure that the goddess would grant him victory. Hence,
when in AthS. XI. 8.24, along with such positive feelings as "delights,
joys, enjoyments" (dnandd mddah pramzido) also "laughter, nari&d and
dances" are said to enter the body of man, the word nar&ci- may here
well mean his :'manifestation of superiority". Indeed, laughter and dance
were not merely expressions of one's high spirits: that they were also
productive of a new vital strength is not a conjecture based on general
ideas about "primitive mentality" but a certain inference to be drawn
from RS. X. 18.3 quoted above.
From a morphological point of view nari~fci-is not clear. It is a
formation like ndvi~ti- "praise", pdni~ti- "admiration", but nari;cst-
cannot be a verbal noun.S9 In view of AV.amaratcit- (with haplology for
*amyta-tat-) the question may be raised whether naristd- may stand for

It should be noted that cows were also the stake of dicing in the sabhti, cf. MS.
I. 6.11, KS. VIII. 7, KapKS. VII. 7, ApSS. V. 19.2.
See Lindner, Altind. Nominalbildung, p. 133; Wackernagel-Debrunner, AItind.
Gramm., 1112, pp. 365, 617; Renou, Monographies sanskrites, 11, p. 11; Burrow,
Skt. Language, p. 163.
 narigizd-td- "the state of being *nciri&a-", the last word then being a
derivative from the verbal root like civi~fiza-,tcipi~iha-,Creek ?p: I C ~ S O ~
etc. (but the ethnic name Naristi, Tac. Germ. 42, must be kept apart on
account of its aberrant vocalism). For the accent of the supposed older
form 'rzariglzci-t6- see Wackernagel-Dcbrunner, Altind. Gramm. 1112,
p. 619. This explanation could also account for the variant nari~ihd-in
the formula nari+@6yai bhimalcim (VS. 30.6, TB. 111. 4.1.2).90 Both the
obvious derivation of the last 'word from bliimci- and its association with
a word for "manifestation of prestige" suggest a meaning "awful,
tremendous, formidable" (Mahidhara: bhayamkaram), rather than
"fearful" (Sllyana bhirurn, capaldk~am,Roth, Monies-Williams). As for
its formation with -1a-, see Altind. Gramm. 1112, p. 863.

Conclusion

"All these feasts bear the character of secret or open warfare, not for
property or possessions, but for social influence, prestige; sometimes
between individuals, mostly between groups or both simultaneously. The
.
weapon is the gift.. Considered in this light these feasts present them-
selves as purely social phenomena. en, however, we consider that
these antagonistic groups, whose peculiar relation also manifests itself
in ceremonial matches or even in fights, at the same time stand for
certain parts of the cosmos, and that consequently their strife demon-
strates the antagonism of primeval cosmic forces, we recognize in these
feasts not only the ceremonial sediment of social rivalry, but also a
religious rite in the shape of a cosmologic drama." These words, written
by J. P. B. de Josselin de Jong ("De Oorsprong van den goddelijken
Bedrieger", Meded. Kon. Akad. Wet., Afd. Lett. 68 B/l, 1929, p. 26f.)
with reference to the kolekole-festivals in Melanesia, the kula-system
in the Trobriands and the potlatch of N.W. America, would seem to
apply fairly well also to the Rigvedic society. The ceremonial contests
of poets, who act as the representatives of their patrons (or their parties
in general), and whose strife is obviously the counterpart of chariot races
and warfare, were at the same time a ritual that aimed at a renewal of
life and the winning of the sun. In connection with our conjecture that
these ceremonies originally took place during the winter solstice, attention
may be drawn to the extinction of the great BahrSim. fire in Iran at the
For a different explanation of the variant nari~fhd-see Bloomfield, SBE, 42, p. 544.
The meaning of nririsfIra- ApSS. 11. 20.6 is not clear, see Caland's note a.1.
 dian soil this festival and the verbal
have fallen into desuetude, and the
later potlatch ritual, with gambling as its most
came to be celebrated in the rainy season acco
bltarata, p. 293. On the other hand, the difficulties inherent in any
attempt to reconstruct the social phenomena of the ancient Indo-Iranian
civilization are apparent from this study. Verbal contests could occur
in principle on any occasion where the nd vydxand or sabhkyo yrivd
appeared as a speaker in a meeting. ow, it may certainly be true that,
just as in the Dyak society (see p. 253), also with the ancient Aryans any
verbal contest could have an importance of its own which transcended
the private interests of the persons concerned, but the evidenc
admits of a demonstration of such speculations. In any case, the
presents a picture of highly specialized functions. The patron apparently
does not act as a sabhkyo yzivd during the great annual festival but, while
ra as maghdvd by his munificence, he obliges the poets
ist him in return by defending his party in the word
duels in rhe sabhd. That in such duels, where the contestants had to rely
on the "mental quickness in their heart" (X.71.8), elaborate hymns
could have been of much importance for the ultimate decision is hardly
probable. It is possible that, owing to the specialization which reserved
this task for the priestly poets, these duels had already assumed the
purely priestly brahmodyas. However that may be, in
gv'eda, despite its vague terminology, would seem to
point to the existence of a second function of the poets besides that of
composing and reciting the sacrificial hymns, even though it is impossible
for us in most cases clearly to distinguish between both. This much
seems clear, anyway, that'the main tasks reserved for the patron were
the chariot races and the present-giving, which both were also weapons
in the competition of this potlatch-like festival (cf. e.g. Held, op. cit.,
p. 245).
References to analogous phenomena in the civilization of other "Indo-
uropean" peoples have purposely been omitted in the preceding study.
mention may be made of some interesting parallels (to which
. de Jong drew my attention) to be found in the New Year's
festival in Lhasa, which, though amalgamated with the
mony since the reorganization by the fifth Dalai Lama, still preserves
 e characteristic traits of the archaic festival, including horse

references).

(Editor's Footnote : First published in Indo-IranianJ o u m l , Mouton, The Hague, vol. IV,
no. 4,1960.)
8. A N INDIAN PRO

I. The idea that Vedic mythology had a counterpart to the Greek

myth of the Titan Prometheus, who stole the fire from Zeus for the
benefit of men, is still current among Vedic scholars. It dates back to
I 5 5 2 , when Rudolph Roth in J&ka's Nlrukta san~mtden Nighanfuvas,

p. I I 2 , discussed Yzska's interpretation of MZtari6van in the following

words : 'Die Deutung von Miitariqvan auf Viiju lasst sich ail? den Texten
nicht rechtfertigen und beruht wohl nur auf der E t p o l o g i e von W .
?van. Die zahlreichen Erwahnungen im Veda zeigen clas Wort in zwei
Deutungen. Einmal bezeichnet es Agni selbst . .. sodann aber auch
denjenigen, der ein anderer Prometheus das von der Erde verscliwun-
dene Feuer vom Himn:el, v011 den Gottern herabholt und zu den Men-
sclien, z11 den Bhrgu bringt .. . Wie Prometheus der iibei-rnenschlichen
Ordnuilg der Titanen angehort und nur c h u m den Funken iin Himmel
holen konnte, so ist Miitarigvan zu jenen halbgottlichen Geschlechtern
zu rechnen, welche die tedische Sage bald in Gemeinschaft der Gotter,
bald auf Erden wohnen lasst .. . Von diesen zwei Bedeutungen des
Wortes Miitaripan scheint mir die erste, wonach es das Feuer selbst
bezeiclinet, die ursprii~iglichezu sein. ' Whatever the value may be of
the et>mological explanation 'in der Mutter schwellend', on which
his last conclusion is based, it should in any case be noted that Roth
himself seeins to have been well aware of the weak foundation on
which his interpretation of the q t l l rested. To the words 'von den
Gottern herabgeholt' he added a special foot-note in which he refer-
red to RS. 111.9.5 and 111.5. I o as the 'Hauptstellen'.
Seven years later a book was published which was t o become one of
the classics in the field of coinparative mythology,' viz. Die Herabkunft
I. Ct. ' 4 . K u h n ' s e p o c h - m a k ~ n gessay' (Eggel~ng,S B E . XII, p. 294 n. 3). T h ~ sis s t ~ l tl l u e

7
 217
des Feuers und des Giirtertrankes (Berlin I 859) by Adalbert Kuhn. In the
beginning of the first part (pp. 1-1 I 8) Kuhn rather perfunctorily dealt
with 'die herabfuhrung des Agni zu den nilenschen', for which he
simply referred to Roth's 'ample discussion' (p. 3). He accepted
without further comment Roth's equation of M~tariivanand Prome-
theus, whose name he explained from Skt. pranliitha-'theft' and pra-
mantha- 'twirler' (pp. I 6-1 8 , earlier in KZ. 4 , p. I 24). On p. I 8 he
concludes : 'Nach diesen vergleichungen bedarf es denn wohl kaum
noch der ausdrucklichen erklarung, dass wir in dem feuerraub cles
Prometheus eineri mythos anzuerkennen haben, der sich dem von
M2tariqvan klar zur seite stellt, wie iclil denn auch bereits oben ange-
oeben habe, dass auch Roth in diesem einen zweiten Prometheus sehe.
b
Dass er aber mit ihm identisch sei, hoffe ich in der vorangehenden
ausfiihrung uber seinen namen klar gemacht zu haben . .. ' The second
part of his book (pp. I I 8-233) was devoted to 'Die herabholung des
gottertranks ' .
Kuhn was unquestionably right in treating the winning of Agni and
Soma as two parallel myths. In various ways the Vedic myths express
1 the idea that in the beginning Agni and Soina were in a world whose
power of resistance (vy&-) was impersonated by a dragon (a'hi-) . The
following reflexions on the Vedic myth are based upon the assumption
that this world was an undifferentiated primeval unity, comparable to
the Greek Chaos.'
Some myths imply that Agni and Soma were released from this
primordial world even before Indra slew Vytra. In the Rigveda this is
expressed in I. 9 3 . 6 : 'One (of you) MiZtariivan fetched from heaven,
the other the eagle has stolen from the rock' dnyam divd mataris'va
jabhiird, 'mathniid anydm pdri +end a'dreh) .3 On the other hand it is said
in spite of all justified criticism of later generations, e.g. ,Votlskov, Sjdedyrkelse og Katurdyrkclse
I (1 890-1 897), p. I 27.
2 . See further, e.g., I I / . IV ( I 960), p. 2 I 9, 270, VIII (1964), p. 107.
3. See Johanna Narten, [I/. IV ( 1 960), p. I 2 3 . Keith, Religion and Philosophy of the LiJu,
p. 2 2 I , denies that this connection of Agni and Soma was 'primitive'.
in one of the Saphit% that when Indra was about to slay the dragon,
Agni and Soma warned him not to do so because they were in Vrtra7s
womb. Cf. TS. II. 5.2.2-3 nld prd hcir, civdm antdh sva iti and RS. 111. 29.
14, where Agni is said t o have been born from the womb of the Asura
(ydd a's~irayajatha'rcid djciyata). The mythological implication of this ver-
sion of the Vedic Creation myth is, accordingly, that Agni and Soma had to
be liberated from the primordial world before Indra could slay the dragon
to found the dualistic cosmos. Many Vedic texts state, indeed, that Indra
conquered Vrtra with the help of Agni and Soma, cf. MS. 11. I . 3 (p. 5, I )
ajnisdmcibhycil;nvai viY&! 'ndro vrtrdm ahan and KS. XXIV.7 (p. 97, I 8),
KKS. XXXVII.8 (p.202, 20), TS. 1.6.11.6 =V1.1.11.6 (andKeith's
translation,p.~oo~vithn.~),~~.II.~.~.1~,V.2.~.~,AB.II.~.12.
In one of the most interesting cosmogonical hymns it is said that Agni,
Soma and Varul>a left the world of the ancient Father Asura, who here
represents the primeval worldofundifferentiatedunity (RS. X. I q . 4 , cf.
verse a). Then Indra4 invited Soma to come outside so that they could
conjointly slay Vytra (verse 6 : hdnijva vrtrdm, nire'hi soma), a procedure
which is stronglyreminiscent ofthe Romanevocatio deorum ex urbibusobsessis
(as Macrobius has it). A nlythological parallel is the story ofhow Indra had
to persuade UianH KHvyah t o come over from the Asuras to the party of the
Devas before the latter could conquer their foes : Jairn. Br. I. I 26' aslniin
abhyupcivartasve 'ti, ~audh.SS.XVIII.46 (p. 403,3 F.) sa hci 'jiiapto 'surebhyo
'dhi devcin upasamiyqa. tato ha vcietad deviiasurijn mahcisamjrcin~am
jigyuh. It
is clear that God Soma, who strengthens Indra before the combat with
Vytra, can be said to have assisted him. Cf. the Soma-hymn IX.6 I . 2 2
sd pavasva yd dvithe' 'ndram vrtrciya ha'ntave. The notion of Sdnia-
vTtrahdn- was, indeed, an inheritance from the Proto-
cf. Haon~a-vara6rajan- Y.9. I 6, Yt. 14.57.
These few details may be sufficient t o show how ~roblematical
Roth's and Kuhn's equation of ~tariivan and Prometheus was.

4. Vodskov, Sjceledyrkelse og Naturdyrkelse, p. 2I 2, took Agni to be the one who bade Soma
to appear.
Kuhn's interpretation basically rested on two assumptions, first, that
Agni and Soma originally were in heaven ('in den Wolken entstehend',
p. 253) and, second, that the fire was stolen from the gods for the bene-
fit of men. It should be noted that when Kuhn wrote (p. 6) 'es heisst
namlich.. , dass M8tariqvan den Agni von den gottern hergebracht habe'
he omitted the cautious proviso made by Roth. It will be clear that
from a mythological point of view the correctness of these words is
open to serious doubts. In the Vedic Syena-myth the eagle (or whatever
other bird may have been denoted by the word) steals Soma for Indra,
the protagonist of the Devas, to give him the force necessary for slaying
the dragon. This is not the place to discuss from what world Soma
was stolen. The Rigveda specifies it as 'from afar' (pardvdtah) , 'from
the rock', 'from the iron strongholds', 'from the sdnu' (either of
heaven or of a mountain), 'from the sky', whereas the brzhmanas
simply refer in fixed phrases to 'yonder world', 'the third heaven' or
simply 'the heaven' (only SB.). Whatever the explanation of these
terms may be (which I hope to discuss elsewhere), they do not contain
the slightest indication that Soma was stolen from the gods. On the
contrary, it is sometimes expressly said that the gods tried to win Soma
from yonder world (AB., SB., see below, p. 95). Not until the
Suparnzkhyzna and the Mdh~bh~rata was the fundamental character of
this myth so much forgotten that the Soma could be said to have been
stolen from Indra. It goes without saying that any attempt to interpret
the meaning of the Syena-myth on the basis of the distorted data of the
Suparnzkhyzna (e.g., Jar1 Charpentier, Die Suparcasage, I 9 2 0 , pp. 149,
2 8 7 , J. von Negelein, GGA. I 924, pp. 66 f., 1 I 7 !) is doomed to failure.
2 . In view of what was said above about the older version of the
Syena-myth, a brief discussion of a single exception found in a brzh-
mana must here be inserted. In the Yajurvedic texts the Syena-myth is
presented in a more or less ritualized way. Here it is the three metres
which fly up to the sky, the third of which, viz. the GPyatri (sometimes
in the shape of an eagle), steals Soma from one or more Soma-guar-
dians. The resistance offered by these guardians is no doubt an old
ieature of the myth. Thesc guardians, although sometimes confused
with h e Gandharvas, were doubtless serpent^.^ As such, they are
t-iiaracteristic of the parallelism that exists in many points between the
prrnieval world which was undivided anct-the nether world of the later
d~raiisticcosmos.
Now, while the Taittiriya Sal2lhitii refers t o these guardians in a
rather neutral way, in the words (TS. VI. I . 10.5) etk vd amrismrm lokk
c6mum araksan, tkbhyd 'dhl sdman? dharan 'they indeed in yonder world
warded the Soma; from them they grasped6 the Soma' (Keith), the
b
Kiiil-iaka Sai21hitti confuses them with the Gandharvas, who properly
bciong to a different episode of the myth. Cf. KS. XXIV. 6 (p. 96,6)
c?-c.I i3 ctad (read etam?) gandharvci a g o p q a n n ainusmiml loke. The parallel
19.1,sigein the Kapisthala-Kafha Samhitii has a curious but unmistakable
b l i p m that it here replaces gandharvci by devci: KKS. XXXVII.7 (p. 2 0 1 ,
I 0 ) CC i d cram derci a g o p q ~ a n n an~usmiml loke. The scribe who here

anadvcrtcnt!y wrote deva but left the reference t o the gandharvah in

line i 3 intact, may have had in mind a passage like Suparniikhy~na
2 . 2 , where it is the deva!~that watch over the Soma.
While this is obviously a mere slip, the same cannot be said of two
passages in the Maitr2)-a~>iSamhitZ : MS. 111.7.7 (p. 84, I 6 f.), 8 . I o
(p. ioq, r o) reads etC vai der&cinl somarciksaya : etkbhyo v6 a'dhi chundciq~n
~clrnm 6hnrun. 7he internal contradiction between the demoniacal

L)L~ ~ 2 !i.,,m
s whom the Soma is stolen and the notion of 'guardians of
iii, p i s ' is ol>\rous. There must accordingly have been a shift in the
~ ' 3 t,&oiiatedl: it11 the 'guardians', and this shift was probably due to
A e :,,uaI ;itt of 'i.iiclicating7 the objects that served as the price of
\OC)IZ. -0 i-iesc 50tna-~~arclims. The ritual of the buying of the Soma
isit L!!r-r 0!1a 71, ~ d considered
, a re-enactment of the mythi-

a 31' I lj n i 1 11i . , ~. \cli;~pln 1 Indien ( I 9 l o), p. 6 5 and J . Charpentier, Dic Su-

cal theft of the Soma from the serpents who were its guardians. There-
fore, the guardians might again fee1 offended by this ritual and to ward
off their evil influence they had to be appeased by objects called
somakrriyan+, cf. KS. XXIV.6 (p. 96'5) svcinnabhrad iti ~omakraya~cin
anudis'ati. Now the original meaning of this act of 'indicating' the
~omakra'ya~cih had come to be reinterpreted in the course of' time. The
Taittiriya S a ~ ~ h i texplains
z the rite as follows (TS.VI. I . I 0.5) : ,vdd
ete'bhyah somakr+ancin n6 'nudije'd, rikrito ' y a sdmah yiin, nB 'yai 'tk
'mdsmim loke' sdmam rakseyur; ya'd ete'bhyah somakrLijVancin anudis'a'ti
kritd 'ya sdmo bhavay, et2 'yZ 'mdsmim loke' sdmam raksanti 'If he were
not to indicate t o them those that serve as the price of Soma, he would
not have purchased the Soma, and they rvo~~ld not guard the Somafor him
inyonder world. In that he indicates to them those that serve as the price
of Soma, he really purchases the Soma, and they guard the Soma for him
inyonder world' (Keith). Owing to the reinterpretation of the function
of the Soma-guardians that is expressed in these additional remarks
the author of the passage in the Maitrzyani Sawhits could take a further
step and declare them to be guardians who guarded Soma 'in yonder
world' for later use by the gods. It is evident, however, that his phrase
devdncim son~ara'k~ayah, isolated as it is in Vedic literature, does not
prove that the Soma was originally stolen from the gods. Nor can such
a conclusion be based upon Jaim.Br. 1.287 atha he 'ndras)/a tridive Soma
cisa 'Now Soma was in the third heaven of Indra', which merely
foreshadows the later notions of the Supaqzkhyiina and the epic. Cf.
also Mhbh. V.97.4 crit. ed. atra 'mpam suraih pitva nihitam niha-
taribhih (viz. in the nZgaloka, which is an old trait!).
An equally curious shift in the ~ a t a ~ a t h a - ~ r ~ h(111.
m a 6.2.8
~ a and
r 5) is not relevant for the Syena-myth.
3. Since, then, Soma was stolen for Indra and the Devas and the
same must be assumed to hold good for Agni, the contrast with the
Greek myth of Prometheus is obvious. Because Prometheus had decei-
ved Zeus (Hesiod, Theogonia 537-545, Erga 47-48), the latter hid the
 fire and withheld it from men. Prometheus, however, managed to
steal it from him. See Theogonia 565-567 and especially Erga 50-53 :

It is curious that this fundamental contrast between this myth and

that of MStariSvan has not withheld the large majority of scholars, up
to the present day, to accept Roth's interpretation of MZtariSvan as a
second Prometheus. Only Bergaigne, in accordance with his structural,
1
!
non-comparative approach, does not n~entionPrometheus by name but
r he, too, accepts Roth's interpretation. Cf. La religion ve'diqtre I ( I S J S ) ,
1
i p. 54: 'Mais au vers 111, 9, 5, il est dit ell propres termes que le feu
i cac14, et produit par la friction, a dti: tiri: par Mstariqvan 'd'un lieu
I 6loigni-', tird 'des dieux.' Cf. pp. 17, 52, 55 on the descent of the
fire. Of the other authors the following may be quoted in the order of
appearance of the first editions of their works. H.S.Vodskov, Sjale-
4rrkelse og Natlrrd~rkelse ( I S go- I 8 97), p. I 24 : 'baade Manu og alle de
andre kunde hentet den fra Himlen som Prometheus gjorde det'.
Hillebrandt, Yedische hktholoyie ( I S9 1-1 902), 1' (192 J ) , p. I 55 : 'Die
Verse, welche von M[Stariqvan] als Prometheus Indiens sprechen, sind
zbvar zahlreich genug, urn ihn als F e U e r b r i n g e r zu kennzeichnen,
aber den naturlichen Ausgangspunkt seines Wesens hellen sie nicht
auf' (,cf.n. 5 , without comment on 111.9.5, and 11" ( I 929), p. 352). H.
Oldenberg, Die Religion des Veda (1 S94), 3rd and 4th ed. (1923),
p. I 2 2 : 'sein Bote bringt ihm und damit der Menschheit vom Himmel
das Feuer, dessen vornehmste Tugend fur den vedischen Dichter seine
Wirksamkeit heim Opfer ist. Dies die indisch diirftige Form der Vor-
stellungei~,die cler Tiefsinn griechischen Geistes zur weltumfassenden
Tragik des Pronletheusmythus erhoben hat. ' Cf. p. I 08 n. 3. A. A . Mac-
donell, Vedic Mythology ( I 897), p. 7 2 : 'MZtarihan would thus appear
to be a personification of a celestial form of Agni, who at the same time
is thought of as having like Proinetheus brought down the hidden fire
from heaven to earth.' E.W. Hopkins, The Religions of India ( I 902),
p. I 09 f. : 'Aryan, as Kuhn has shown, is at least the conception if not
the particular form of the legend alluded to in this hymn, of fire
brought from the sky to earth, which Promethean act is attributed else-
where to the fire-priest.' Cf. p. I 68 : 'no detailed myth was current in
primitive times'. Jar1 Charpentier, Kleine Beitrage zur indoiranischen M,~tho-
lbgie (Uppsala Universitets Arsskrift I 9 I I ) , pp. 7 3,74 (no comment on
HI. 9.51, pp. 76-77 ('dass MZtariSvan in den alteren Teilen des RV.
wirklich eine Art Prometheus, ein halb-gottliches Wesen ist, das das
Feuer voin Himmel bringt oder hier auf der Erde erzeugt hat'), p. 8 r
('der indische Pron~etheus'),p.83. A.B.Keith, jR.45. 1916, p.555:
'It is clear, however, that the Vedic myth already regards the action of
the descent of fire in the form of lightning and the fall of rain therewith
as a species of theft, and the development of a myth like that of Prome-
theus is not very difficult. ' Leopold von Schroeder, Arische Religion
(1916/ 1923), 11, p.485: 'Haufiger wird die wunderbare Tat dem
M h r i ~ v a nzugeschrieben, einem indischen Prometheus ... ' Id .D.
Griswold, The Religion o f t h e eigveda (192 3), p. I 63 : 'We have here in
general the Vedic equivalent of the Greek myth of Prometheus.'
A.B. Keith, Religion and Philosophy of the Vedu (1925), p. 13S : 'he
brings Agni from the sky, from afar for men, or from the gods' ...
'The only alternative view which has any plausibility is that of Olden-
berg who sees in him a Pronletheus only, without any divine nature
other than the bringing down of fire. ' P. I 62 n. I 'The legend of the
theft of fire (Prometheus, Loki) is only faintly seen in the figure of
MZtariyvan. ' Johanna Narten, Ig. IV (19601, p. x 34: 'DaB in
rigvan der indische Prometheus vorliegt, wurde auch bis
bezweifelt ... Doch wurde in der indischen Version das Rau
griechischen Mythos vermiBt ... Die Feststellung der Wur
l '
entreiBen, rauben" und ihres Vorkommens in Verbindu
Mltarihan innerhalb des altesten indischen Literaturwerks zeigt, daB
der M~tai-isvcLil-Prollletbeus-Mythos ursprunglich also auch in dfesem
Punkt iibereinstinm~te- in Indien ging dieser Zug allerdings nach-
vedisch verloren. '
4. Just as in the case of Soma, the place from which Agni is brought
to men is described in the Rigveda in various terms. He is said to have
been brought 'from heaven', 'from afar', from 'the womb of the
waters' o r to have freed himself 'from d a r k n e ~ s ' Owing .~ to the strict
parallelisn~between the mythical origin of Agni and the ritual act of
kindling the fire, references to the first often intermingle in these
verses with those to the latter. Cf., e.g., VI.s.4 apdm upa'sthe mahisd
agrbhnata, vijo r i j a n a m upa tasthur ~gmlj.an1 / d diitd agnirn abharad
v~va'svato,vais'vcinarani mataris'va parciva'tah 'In the womb of the waters
the buffaloes took hold of him, the clans worshipped the king who is
to be The messenger of Vivasvant, MZtariSvan, brought Agni
Vaiivanara hither from afar', V. I . 2 mahdn deva's ta'maso nir amoci 'The
p a t god freed himself from darkness', VI. I 5 . I 7 imdm U tya'm atharvava'd
L

a g ~ i i m~ m n t h a n tvedha'sah
~ / y d m adkiiya'ntam dnayann a'miiram Syciyydbhyah
'This Agni the arrangers (of the ritual) produce by whirling as (did)
Atharvan, him, the unerring one, who moves tortuously, whom they
have brought from the dark (places). '
As for the term parava't-, it has long been observed that it often
denotes the underworld, e.g. 1.48.7 (Usas) esd 'yukta parciva'tah
s h y a y o 'da'yancid a'dhi 'she just now harnessed (her horses) from afar,
from the point where the sun rises', IV. 2 I . 3 d p t v indro diva' d prthivyd
mak$ samudrdd uta' vd puriscit/&;lrqarcid a'vase n o marritvcin parciva'to vci
sa'danad p d y a 'Indra should soon come from heaven, from the earth,
from the sea or from the firm ground, from Svarl~ara,accompanied by
the Maruts, to assist us, o r from afar, from the seat of Rta'. Cf. also
.
Renou, Ilj.4 ( I g 6 0 ) , P. I og The last words parciva'to vci sa'dancid Tta'ya,
which Geldner took to mean 'den hochsten Himmel' must rather refer
7 . C f .also X . 45.6 vijcim cid ddrim abhinat porqdiijdndydd agnim dyajanta pdiica 'Er spaltete
sogar den festen Fels in die Ferne ziehend, als die funf Volker den Agni anbeteten' (Geldner).
to the nether world, for the sun rises in the morning 'from the seat o!
@a' (rtdya sadanat 1.164.47) and Usas, who come5 'from afar'
(paravdtah 1.92.3) awakens 'from the sn'das of Rta' (W. 51.S). Sec
further V.62.1 and in general llJ. IV (1960)) p. 226, V111 (196q), p. 1 0 7
and cf. the references to the 'stone house' o r 'rock' where Agni is
born and from which he rises in the morning (lv.VIII., pp. r I I n. 8 i ,
I o 8 , I 2 o n. r 2 2). AS such parava't- is used as an euphemistic termf o p -

Nrrti- 'destruction', as Kenou, Ind. Ling. I 6 ( I 955), p. r 24 n. q has

pointed out.
It remains one of the unsolved problems of Vedic r - n ~ t h o l oW ~h~j
terms for 'nether world' and 'primeval world' can alternate with
'heaven', 'the highest heaven', 'the third heaven'. Pos~iblvthese
terms denoted the mysterious world of the totality of t h e
dualistic cosmos, Visl>uls third step, and were then also used in
application to the undivided primeval world. Be that as it may, thic
much is obvious that the paravdt-, which could be identified ~ i t h
Ni'rytl-, was not a dwelling-place of the Devas.
The same equivalence of the distant region (parZvdt-) and heavcn is
found in the myth of MPtariSvan. O n the one hand MZtariSvan is said
to have brought Agni for Manu from afar : I. I 2 8.2 ,vim lnZtariiv5 nldnace
parovo'to, deva'nt bhdh parova'tah, while on the other hand M ~ t a r i i v a n
brought Agni from heaven (did), in contrast with Soma who is said to
have come from the rock (1.9 3.6, see above p. 8 6).
5. In the Rigveda there is only one version of the myth or the
carrying off of Soma. When Indrawas about to slay Vrtra - at amoment,
accordingly, when heaven and earth were not yet separated ancl the
world was still the rudis indigestaque moles, the dsat, from which the
cosmos was to arise - the eagle brought the Soma)r hlm (and for
Manu, IV. 2 6.4). Since, however, Indra stood for the Devas in genera',
it could also be said that the Devas, in their strife with the Asuras,
longed for the beverage of immortality. This version is found in the
Yajurveda, side by side with the well-known myth of
 Suparni. Cf., e.g., MS. 111.7.8 (p.85, 13), 8.10 (p.109, 10) sdmo vai
arnritrci 'sit, te' devd gciyatrim prdhinvann: amtim sdmarn dhark 'ti 'Soma
was yonder. The Devas sent Glyatri, saying "fetch yonder Soma".'
Similarly Ait. Br. In. 2 5. I and 26.1-3.
According t o this version Soma was brought for the gods, cf. SBK.
11.6.3. I deve'bhyas tdsyci cihdrantyz avdd dsta 'bhyLiydtya paqcim prcicjccheda
(quoted from Eggeling, SBE. XII, p. I 83 n. 2) 'while she was bringing
him for the gods, a (footless) archer aimed at her and severed one of the
feathers ', SBM. 111.2.4.2 tkbhyo gqatrf sdmarn dchci 'pptat 'Ghyatri
flew up t o Soma for them', 111.6.2.8 sd ha kadrdr uvcica / Ztmdnam vai
tvci 'jaisam, divy &ad sdrnas, tdm devkbhya dhara ... 'Then said KadrE
'Verily I have won thine own self; yonder is Soma in the heaven:
fetch him hither for the gods ... ' (Egg.), AB. 111.26. I te devci abruvan
gqatrim: tvam nu imam somam rqznam cihare 'ti 'The gods said to
Glyatri "Do thou bring king Soma for us" '. The Kiifhaka formulates
this in a more general way, cf. KS. XXVI. 2 (p. I 2 3, 2-3), KKS. XL. 5
(p. 2 29, 2-3) : sarvebhyo vciesa cihri,yatemanuvebhyah pitrbhyo devebhyas . ..
'Verily, it is for all that this (Soma) is brought: for men, for the pitaras,
for the gods. '
Since the myth of Miitarihan runs entirely parallel to the Syena-
myth, it would be quite in line with the general mythological con-
text if MltariSvan brought Agni from the same primeval world not
only 'for Manu' (mdnave 1. I 2 8.2) but also ffor the gods' (devkbhyah) .
It should be noted in this connection that nowhere in the Rigveda is
Mntarihan said to be an enemy of the gods. He rather cooperates with
them to fetch the fire for men, cf. X.46.y cd ilknyam prathamdm
mcitarihci devds tatakpr mdnave ydjatram 'Thee, 0 Agni, who must be
invoked first, thee, who deservest adoration, Mltarihan and the gods
have shaped for Manu. '
The preceding conclusion, based upon a purely mythological inter-
pretation of the data, that Mhtarihan must have brought the firefor the
gods, is in flat contradiction with the view, universally held ever since
I 852, that he has stolen i t f r o m the go&. Wow, it seems never to have

been sufficiently realized (although Roth had implicitly warned his

readers !) that the whole theory of an Indian Prometheus is based excln-
sively8 upbn Roth's interpretation of the form dcvdbhyah as an ablative
in 111.9.5:
sasyvd~samiva tmdmi
'gnim itthd tirdhitaml
ainaIp nayan mataris'vii parzvdto
deve'bhyo mathidm pdri

which Geldner translates as follows: 'Der gleichsam von selbst weg-

gelaufen war, den Agni, der dort verborgen war, den fiihrte
aus der Ferne her, von den Gottern weg, nachdem er (aus dem Holze)
gerieben war'. Roth inferred from this passage that Mztariivan 'das
von der Erde verschwundene Feuer, vom Himmel, veil den Gottern her-
abholt' and Kuhn (p, 6) 'dass Mftari~vanden Agni von den gottern
hergebracht habe ' (see p. 8 8). Sin~ilarlyOldenberg, Religion des Yeda 3-4,
p. I 2 2 : 'von den Gottern her'. All translators have followed Roth in
this respect. Cf.
I 8 76 : Ludwig I, p. 3 3 6 'ihn fiirte MBtari~vanausz der ferne, den

durch reiben erzeugten, von den gottern her' ; Grassmann I, p. 64 'Ihn

fiihrte von den Gottern Mztariqvan her, von Ferne den erriebenen'.
I 897 : Oldenberg, SBE. 46, p. 256 'Him MftariSvan brought hither

from afar, from the gods, when he had been produced by attrition (of

8. As for the Bhrgus, mentioned in Roth's second 'Hauptstelle' (see above, p. 85), viz.
111. 5. I o y d d i bhrlgubhyah pdri mitariiv8g6h8 sdntam hayavd'har77 samidhe' 'wenn MStariSvan ihn,
der sich vor den Bhrgu's verborgen hielt, als den Opferfahrer entziindet hat' (Geldner), it is
not quite clear how Roth interpreted this passage but his general statement 'Da e r [viz.
.
MBtariSvan] das Feuer zu den Bhrgu bringt' and 'der ... das Feuer .. zu den Menschen, zu den
Bhrgu bringt' agrees with most modern translations of this particular passage, e.g., Oldenberg,
SBE. 46, p. 241 'for the sake of the Bhrgus', Renou, E VP. XII, p. 55 'pour les Bhrgu'. See also
Johanna Narten, I l j . IV, p. 133 f. Anyway, i t cannot be used in support of the theory of an
Indian Prometheus, in spite of Kuhn, p. 6 ('von den Bhrgu her'). [See now also Hanns-Peter
Schmidt, Brhaspati und Indra (1968), p. 69.1
228

the wood) ' ; Macdonell, Ved. Myth., p. 7 I paraphrases 'MZtariSvan

brought from afar the hidden Agni, produced by friction, from the
gods '.
I 95 I : Geldner I, p. 347 (see above)

I 960 : Johanna Narten, IIJ. IV ( I 960), p. I 3 3, who established the

correct meaning of math-: 'den fiihrte MZtariSvan aus der Ferne her,
den von den Gottern weg geraubten'.
1964: Renou, EVP. XII, p. 57 : 'C'est MZtariivan qui l'amena du
fond de l'espace, de chez les dieux, (cet Agni par lui) dkrobk' (p. I I 8 :
Cquivoque entre 'barattk ' et 'dCrob6 ').
As for the formal interpretation of the verse, it should be noted that
inathZydti 'he steals' is sometimes construed with the adverb pdri
'from', which takes various positions with regard to the ablative. It
can stand after it (cf. IX.77.2 ydm divds pdri Syend mathZydt 'whom the
eagle stole from heaven') or independently, as in 1.93.6 cimathncid
anydm pdri +end ddreh 'the other the eagle stole from the rock'. No-
thing prevents us, therefore, from construing mathita'm pdri with
parcivdto (cf. I. I 2 8 . 2 devdm bhcih parzvdtah, p. 94 and VI. 8. 4, p. 93).
Now, it was assumed above that just as Soma had to be stolen before
Indra's vytrahdya-, so MZtariSvan's theft of the fire took place in the
undifferentiated 'primeval world. If this is correct, the myth of
MZtarihan may be compared with Agni's leaving the 'Father Asura'
at the moment of the creation of the dudlistic cosmos. In RS. X. I 24.2
Agni goes stealthily away (grihZpdn .. . emi) and in verse 4 he, Soma
and Varuna choose Indra and leave the 'Father' : indram vmZndh
pitdram jahcimi / agnih sdmo vdrunas te' cyavante.
In quite the same way it is said in I. 141.3-4 that MZtariSvan steals
Agni, who is hidden (griha sdntam), and that Agni is led away from the
Father. This father, who can hardly be any one else but the pit!-Asura- of
X. I 24.3, is here denoted as mddhva cidhavdh, which Johanna Nai-ten, I l j .
IV, p. I 33 interprets as 'Herschiittler des SiiBtranks'. Cf., however,
Iienou, E VP. I 2, p. I o 2 . The relevant lines are the following :
 ( 3 ~ d )ydd im dnu pradivo ma'dhva 5dhar.e'
$liha sa'ntam mcitaris'vZ mathzyiiti
(qa) prd ydt piti2h paramdn nya'te pdri
'when MZtariSvan steals him who for a long time past has been hidden
with the one who stirs the sweet drink'. (4) 'When he is carried away
from the highest Father .. ' .
This Father, who as an Asura apparently stands for the primeval
world, cannot possibly be localized in the world of the Devas. For that
reason the correct translation of 111.9. scd can only be : 'MStariSvan
brought him, who had been stolen from afar,for the gods. ' This correct
syntactical interpretation of a single word disposes of the 'Indian
Prometheus ' .g
9. For typographical reasons the distinction between ardhacandra and m had to be ignored,
while for editorial reasons m has been substitued for ni of the manuscript.

(Iklitor~alFootnote : First published in Asinlische Sizrdien, Zeitschrifl der Schuwzerischen Gesellschaftfur

Asiettktotdc, Bern, vol. 25, 1971.)
 TAG

1. It would have been impossible for us to form an adequate idea of the

fundamental importance of the piirvaraliga of the ancient Sanskrit drama
as a religious drcirnenon, if we had not had the good luck of the Bhdratfya
Nd{yas'dstra having been handed down to us.l This work describes in
great detail the ritual which preceded the actual dramatic performance
and which was apparently intended to consecrate the stage. In the context
of this study, which deals with one articular'part of that ritual only,
no attempt can be made to substantiate this general statement, which
involves that such translations as "preIiminaries"%r "Vorspie1"s fail to
do full justice to the original character of the ptlrvaraliga.
In any case, although late theoreticians of about the fourteenth century

References to the Nd~yaiistraare to chapters and verses of the Calcutta edition

(siglum C only added where necessary) by Manomohan Ghosh, The Nci.tyasascistra
ascribed to Bharata-Muni, vol. I (chapters I-XXVII) [2nd ed.] Calcutta, 1967, vol. I1
(books XXVIII-XXXVI), (bibl. Indica no 272A) Calcutta, 1956. Sometimes better
readings are given. Where necessary, variant readings are quoted from the edition of
Baroda (siglum B), of which only the first volume (chapters I-VII) was accessible to me
in the revised 2nd ed. : NdfyaSdstra of Bharatamuni with the commentary Abhinavabhii-
rati by AbhinavaguptBcBrya, ed. by M. Ramakrishna Kavi, revised and crit. ed. by
K. S. Ramaswami Sastri, Siromaai (Baroda, r the chapters XIX vol.
I11 of the 1st ed. (Gaekwad's Oriental Series, V) Baroda, 1954, W dis-
posal. The Kcivyamdlrl-edition [KM] is quoted in the 2nd ed. by Pandit KedSunath
(Klvyamiilii No. 42, Bombay, 1943). The Sanskrit text of RaghuvhSa's edition (vol. f ,
chapters I-VII, Delhi-VBrBqasi-Patna 1964) is worthless for critical purposes.
a E.g., Manomohan Ghosh, The NcityaSristra (Translation), Vol. I , revised second ed.
(Calcutta, 1967), p. 76 and passim; A. i3. Keith, The Sanskrit Drama (Oxford, 6924),
p. 292: "the preliminary scene or Ptimamiga, which is practically non-existent in the
classical drama."
See P. Thieme, "Das Indische Theater" (in Kindermann, Fernostliches Theater,
Stuttgart, 1966), pp. 37,53 and p. 39: "Vorspiel" being the fist introduction as against
prastcivami, the second introduction; H.-0. Feistel, Das Vorspiel auf dem Theater
(dissertation Tiibingen, 1969).
 AD.* simply reckon the piirvara~igaamong the four preliminaries of a
j lay,5 people were even at that time still aware of its religious f u n c t i ~ n , ~
i
! he Ncifyas'dstra itself refers to it, in one recension, as a secret act,? which
aims at worshipping the gods.s Konow is right in saying that thept7rvaraiiga
is entirely distinct from the drama p r ~ p e r Roughly
.~ speaking it may
be said that, to a certain extent at least, the pu'rvaranga is a double of the
consecration which took place after the building of a new and
i the rites of which are described in the third chapter. From its character as
a consecration of the stage (rarigapiijanuj it may be inferred that the
drama itself must originally have had a sacred character. In point of fact,
the parvaranga was a sacred act which, as the Nd~yas'cistmstresses, was
equal to a sacrifice (1.126~3.96,5.112, cf. 3.93). So the consecration of
the stage reminds us of the words which the Vedic teacher Siitayajiii once
spoke in regard of the sacrificial ground: "Verily, this whole earth is
divine: on whatever part thereof one may sacrifice (for any one), after
%
enclosing (and consecrating) it with a sacrificial formula, there is a place
of worship" (SB. 111.1.1.4, transl. Eggelingj.
This means that, although pzirvaranga and drama were certainly
different things, both of them must originally have had a ritual character.
1 In the classical drama the pzirvaraiiga must to a lar& extent have lost its
I . original function and the words alah vistarena, often spoken by the
szitradhcira immediately after the ncindi, obviously express the feeling of
actors and audience that the "preliminaries" were a cumbersome neces-
i
The date of the Scihityadarpana may have been between 1300 and 1350 (? S. K. De,
Sanskrit Poetics I, p. 214) or between 1300 and 1384 (P. V. Kane, History of Sanskrit
Poetics, p. 302).
Scihityadarpapa VI.21 tatraptkvalh piirvararigab sabhcipiijd tatab paralh, kathanarh
kavisamjficider nci[akasyii 'py athd 'mukham. The first three words are quoted in the
commentary on the Vikramorvaii as a verse of MSrtygupta, see Sylvain Uvi, Le
th&tre indien 11, p. 64.
Zbid, VI.22 yan n@ynvastunab pl2rvam rarigavighnopaicintaye, kuiilav* prakurvanti
piirvararigalt sa ucyate (Also quoted from the BhcivaprakrtSn, Uvi, 1.c.).
" NS. 36.11 KM kathaydmi kathdm guhydm yan mcim pprcchatha suvratcih, ptirva-
rarigavidhcinasya tdm ca me sariuribodhata. Otherwise 36.13 C bravimi vo hy ayarh
viprci yan mcim (etc.)
B Cf. 36.29 (36.24 KM) evanipijjddhikdrdrtharitptirvararigaE krto mayci ncindstofrakytair
mantrair devatcibhyarcanam prati and 32.483 piirvararigavidhcine tu kartavyo g6najo
vidhib, devap0jcidhikdras'ca tatra samparikirtitah. Even the late Ndfakalak$aqaratnakoia
by Siigaranandin (circa 1200-1250 A.D.? P. V. Kane, op. C., p. 423) describes the
piirvaraliga as being (v. 1121.) "in the form of praise of the gods with instrumental and
vocal music and dance"; or (v. 2159) as "the worship of the gods".
S. Konow, Das indische Drama (= Grundriss der Indo-drischen Philologie und
Altertumskunde, 11. Band, 2. Weft D) 24: "[er] steht ausser-
halb des eigentlichen Dramas", p. 25 "
verschiedene Sachen".
sity that had to be finished as fast as possible (5.163 f.). What had survived
of the circumstantial ritual at the time of, e.g., KBlidBsa, cannot be
ascertained.1°

2. This study will be limited to the role of the jarjara in the piirvaranga.
The latter consists of a long series of episodes. The NGlyaiGstra enumerates
twenty "members" (angas), later theoreticians, such as Sggaranandin, even
twenty-two. From the description in NS. 5.8-15 the following main
elements can be distinguished:
1. Nirgita (5.33-44), which is performed to appease the Daityas and
DBnavasl1 and to honour the gods. It consists of seven sub-members.12Its
dangerous character is apparent from the fact that it is performed behind
the scenes (nepathye) instead of on the stage and from its second name
bahirgita, which is said to be a euphemistic term used to please the gods.ls
Cf. N!,? 5.11 "These bahirgitas should be performed by the performers,
while standing behind the curtain, by means of string and other instru-
ment~."~*. According to Abhinavagupta it consists of nine different
parts715as against the ten parts of the piirvaranga, beginning with song
and ending with theprarocand, which are performed "outside the curtain",
that is, on the stage.lBIn this connection it may be noted that the term
bahih "outside" naturally denoted different places depending on one's
viewpoint. Mostly it denotes the place outside the stage, as in the defini-
tion which Sylvain LCvi17 quotes from Bharata: rangabhiimer bahib

l0 On the word (5.17) avatarapam Abhinavagupta (ed. Baroda, vol. I*, p. 213 line 1)
has the following comment (avatarayakofau chandaso 'k~arasamam)vcimabahirgitci-
nuvarti vcidyam stribdlamrlrkhcidikuttihalcid~ananam"the instrumental music after the
bahirgita is meant to awake the interest of women, children, fools and the like". This
should not be understood as a depreciatory remark on theptirvarakga, since it is merely
a quotation from 33.226+ C (p. 173 line 6 from the bottom), 34.195+ KM (p. 641
line 1) tatraiva ccinte prlrvarangavidhcinam anuprcipya chandabsamerui 'k~arasamenaca
vcidyena bahirgitavidhcinam tu trilayam vartaniyam ... yatra trisaptapratycihdrci[dyal-
vatiryakofis tatra vcidyalh pravartate ... stribri[amrlrkhcivakirye range kutihalajanana-
samartham viidyam samutpannam bhavati. Cf. also, in a different context, 27.61 C
bdlci mrlrkhci[h]striyai caiva hcisyanepathyayob sadci (viz. tu~yanti).Cf. Sagaran. 1157.
l1 Cf. 5.40-41 and 58 daityadcinavatu~@artham, which can only refer to the nirgita.
l4 Cf. 5.44 nirgitam yan mayci proktam saptartipasamanvitam.
Cf. 5.44 aszlyayc ca devcincim bahirgitam idam smyfam, 41-42 etan nirgitam evam
tu daitycinCm spardhayci dvgcib, devcincim bahumcinena bahirgitam idam smytam.
l".ll etcini tu bahirgitciny antaryavanikcigataibprayoktrbhib prayojyiini tantribhdn-
dakytcini tu.
l6 Vol. Is, p. 210 line 3: antaryavanikcikgirti nava prayojycini.
...
ycjni ca yavanikciyci bahir gitakaprayogcidiniprarocana-nfcini trlni
" Le thPGtre indien 11, p. 62.
sthdnaliP yat, tan nepathyam iry ucyate. Thus the term bahirgita no doubt
denotes the music that was performed behind the scenes.lDWhen viewed
from the dressing-room, "outside the curtain" (yavanikdyd bahih) of
course refers to the stage, and "inside" to the nepatl~ya.~~
The difference between the inauspicious first part of the pzlrvaranga
and the following dance and recitation, which took place after the per-
formers "having pushed aside the curtainW2lhad entered the stage, must
originally have been considered essential. This is apparent from the fact
that in one of the late chapters of the Nd$yaSdstrathe bahirgita is referred
to as a separate ritual, distinct from the pllrvarariga p r ~ p e r , ~ % uist also
the logical consequence of the etymologizing definition in NS. 5.7 yasmLSd
range prayogo 'yam pzlrvam eva prayujyate, tasmdd ayaliz pCrvara~igo
vij2eyo dvijasattam& which excludes the nirgita from the pzlrvaraliga.
As late as the last quarter of the tenth century A. D. Dhanika defined the
piirvaranga as beginning with the Uttl~cipana.~~ When, however, some late
authors state that part of thepiirvarariga was performed behind the scenes
(javanikdnta~~),~~ this may reflect the later practice of omitting the ritual
on the stage and beginning immediately with the introduction (dmukha)
of the play itself. Thus the Ndtyadarpana IV and Sagaranandin, v. 1125

This is preferable to balri~rsthdtiam(Ldvi).

lS Otherwise H.-0. Feistel, Das Vorspiel auf dern Theater, p. 18 "ausserhalb der
[eigentlichen] Musikstucke [horbar werdende Tone]", p. 40 "das ausserhalb [der
eigentlichen] Musikstucke [wie Utthiipana usvv stattfindende] Musikstuck [As~ritai".
For yavanikd see, e.g., Raghavan, The Theatre of the Hindus, p. 161 (who quotes the
Sarizgitacicddmapi) and Mankad, ilncient Indian Theatre (1950), p. 20.
Cf. arztaryavanikdm c7hw crnepathyam, quoted from Sfgara (LBvi 11, p. 64). S%ga-
ranandin, 2186-7 has antaryamanikd but javanikd.
5.12 vighafya vai yavanikdtn.
as See 32.460 (410 KM) prayujya ca bahirgitam picrvaranga~izprayojayet, piirvaratige
pravpe tu ndfyadvdrarh samiiirayet. In the abridged TryaSra form of the piirvarariga
the bahirgita seems to have been skipped, if at least 5.146 (140 B) kartavya&picrvarangas
tu tryaJro 'py utthdpanddika?~may be interpreted in this way. Ghosh's translation "(With
this kind of measurement ...) should be performed the Tryasra Preliminaries which
include the Utthiipana and such other items" is hardly correct. Feistel's statement to the
effect that the causative prayojayati was used in the sense "to perform" (op. C., p. 19)
is correct for the present of the active, where it takes the place of prayupakti. Hence
prayojayet but prayuiijita, ptc. prayukta, adj, necessitatis prayoktavya, prayojya,
gerund prayujya, inf. prayoktum, noun of the agent prayoktr. The ptc. samprayojita
(5.46) is, as far as I can see, rare.
" Ad DaSarlipa 3.2: picrvari2 rajyate 'sminn iti piirvaranga!~,utthiipancidiprayogah.
a4 Lkvi 11, p. 64 quotes from a commentary on the. Sarirgitakalpatarrc:sictradt~drdk/~yo
bharato nafa eva dikpdlastictgarjarapicjddindndyant picrvarangam javanikiintar
vidadhita. The text itself, as quoted by Uvi, p. 27, reads: dikpdlastut~rjarap~ijdritarn
eva piirvararigam ncindivyatiriktam k p d slitradhdre vinirgate tatsamagunatvcit sthdpuka
eva satradhriro nrindim javanikrintah pafhitvd rarigabhiimdv Ggatya kdvyam dsthdpayet.
 state that only the ndndiis important but that the other angas, such as the
Utthdpana, can be skipped.25Cf. Sggaranandin 1093 f., 1157.
2. Utthcipana (NS. 5.60-89, cf. 14.44ff.1, which consists of four
"metrical periods".26 The gitakas which form an introduction to it seem
to be part of i t z 7
3. Parivartana (5.90-101, cf. 5.23), salutation of the Dikpglas.
4. Caturthakcira, a p27jd to the jarjara, etc. (5.101-104).
5. Ncindi, benediction of the gods, the brahmins and the king (5.105-
113).
6. Jarjaras'loka, worship of the jarjara (5.118)
7. Vindmana, the inclining of the jarjara (5.1 19)
8. Cdri, a dance (5.120-127)
9. Mahdciri (5.128-133).
10. Trigata (5.136-140), cf. Trika (5.16).
11. Prarocand "laudation" and exit of the sEtradhdra (5.141-142).
The author of the Ndfyas'istra clearly considered the utthdpana and
parivartana the two parts which immediately follow the bahirgita.28The
exact meaning of the first of these words was apparently no longer known
to him. His explanation is the first of a series of etymologizing definitions
(5.227):
yasmdd utthdpayanty cidau prayogam ndndi$d?hakrfh
parvam eva tu rarige 'smims tasmdd utthdpanam smrtam
"Because the reciters of the benediction first (v.1. here) "produce" a
.
performance for the first time on this stage, this is known as an uttlxipana."
It is at once clear that this definition is as irrelevant as, e.g., that of cdri
in 5.27: s'rrigcirasya pracarandc cdri sqlizparikirtitd "The cciri is so called
because it consists of movements depicting the Erotic Sentiment" (transl.
Ghosh). The sense "to excite, to produce", which the author here appar-
ently assigns to utthdpayati, is very rare indeed.29 Elsewhere in the
Nd?yas'dstra the verb is used in its current sense "to erect", e.g., 2.55f.
See K. H. Trivedi, The Ndfya Darpapa, p. 176.
According to Abhinavagupta (I" p. 227 line 7) Parivartas are "repetitions in
singing": parivarrci gr?nakriyribhyHvrttayab,te catvdrah, That they were sung is also
apparent from 4.325,32.462,472. Feistel, p. 50, rightly rejects the translation ".uvalking-
round" and proposes "metrische Periode". One parivarta consists of four samnipdtas,
each of which corresponds to a pada of a stanza, cf. 5.65 catvdrah samnipdtfii ca
parivartab sa ucyate.
See however Feistel, p. 27f.
" Cf. 5.14 tatai cotthGpanam kkryam parivartanam eva ca, 5.156 vrtte hy utthipane
b&b k p e ca parivartane, 5 Add. 2 dddv utthdpani kdryci parivarta~stathd bhavet,
ibid. 4 utthdpamyci '$!akalam parivartanasya ~afkalarn.
The P W ; quotes repum utthcTpayati from Raghuvarizia 7.36 for the meaning "erregen,
heworbringen". For vmtdtt&pamm see below, section 3.
 stambham utthcipayet "he should erect a pillar".30 But even apart from
the curious expression utthcipayatiprayogam the verse of the Nd?yaScistra
does not make sense, with the redundance of ddau (a reading also occur-
ring in the Bhdvaprakdia) and pzirvam, or of the variant readings atra
and asmin.
On account of the common meaning of utthdpayati the general opinion
has been that the term Utthcipana referred to the erecting of the jarjara
on the stage." This has, however, been contested by F e i ~ t e l who , ~ ~ is
certainly right in maintaining that in the detailed prescriptions of NS.
5.59ff. (=60ff. C ) there is no explicit statement about the erection of the
jarjara. He observes that it constantly remains, now in the hand of one
of the assistants, then in that of the siitradhcira himself, who makes certain
movements with it, lowers it and lifts it again (5.81, 113, 121 = 83, 119,
127 C).33The last thing that is said about the jarjara is that the siitradhdro
returns it to his assistant. In the absence of any evidence to the contrary
Feistel assumes that the latter still has it in his hand when, at the end of the
pzlrvaranga, he goes off the stage. In modern performances, it is true, it
seems that the jarjara is often34left standing on the stage, which must then
be explained as a deviation from the ancient practice. Feistel, therefore,
concludes, that the only possible object of utthdpayati isprayogam, as it is
in the definition quoted above. This he translates (p. 29) "Weil die
Rezitatoren der Niindi hier (d.h. an dieser Stelle des Vorspiels) eine Auf-
fiihrung [and zwar nicht irgendeine sondern] eben die erste, zur Ent-
stehung kommen lassen, deshalb ist [dieses Glied] autoritaliv als 'Auf-
stellung' gelehrt." Although he fully admits the artificial character of the
etymologies given in the definition^^^, he feels compelled to take this one
The reading dr~fvcicotthcipanam hrsfvci in 31, p. 488 line 27 KM is c o m p t for
dy~fvcicotthciya hp{&igi,22.249 KM (258 B), 24.256 C. The expression utthcipanav&iyam
in 33.232+ (p. 176 line 12) is not clear to me in the context.
E.g., S. Konow, Das indische Drama, p. 23 (see below, fn. 47), J. Gonda, "Ursprun
und Wesen des indischen Dramas", Actta Or. 19, p. 367.
as Das Vorspiel auf dem Theater, Ein Beitrag zur Friihgeschichte des klassischen
indischen-ScfzauspieIs (Dissertation Tiibingen, 1969), pp. 29 and 116f.
as This, however, is based on an unacceptable alteration of the text, see below p. 2
and fn. 108. .
"z(um) T(eil)", Feistel, p. 117 n.1, who refers to Gargi, Folk Theatre of India, pp.
48, 168, 191 (not accessible to me). Konow, Das indische Drama, p. 37, also says:
"Deshalb heisst sein Banner jarjara und es wird als Zeichen des Schutzes Indra's bei
allen B-iihnenvorstellungen aufgestellt (Bh. 1,8f.)", but Bharata is, as far as I can see,
absolutely silent about it. Cf. also Konow, p. 24: "Der stitradhfira streut Blumen aus,
bringt das Banner des Indm an und reinigt sich aus dem Wasscrkrug" and Cilappati-
kLam 3.128f. mlaikk6lai etir mukamciko vaitta pig.
P. 29: "Dieses und die folgenden Vorspielgliedei werden mittels 'Etymologien'
(nirukta: cf. NS. 6.12-13) definiert."
seriously and to accept the possibility of an idiomatic expression prayo-
gain utthdpayati "eine Auffiihrung zur Entstehung kommen lassen" as
the origin of the noun utthdpana in the technical sense it has here.

3. Some technical terms of Indian dramaturgy are, indeed, derived from

ut-thdpayati. There is, first, btthdpaka "~hallenge".~~ What is more
interesting is that also utthdpana occurs in a technical sense in the
compound vast6tthdpana. It will here be briefly examined to show that it
cannot contribute anything to our understanding of the name of the
second "member" of the p6rvarariga.
Various meanings or shades of meaning seem to have been attached
to the term vastiitthdpana in the course of time. The Ncifyahistra seems to
take it in the sense of "Action (event?) which is fancied as based upon
[a situation which may be] agitated or not, and which consists of a com-
bination of all sentiment^".^^ If this interpretation is more or less correct,
utthdpana must have meant something like "ev~cation"~~ of certain
events which were not acted on the stage. Abhinavagupta, however,
illustrates it with a reference to a scene in the Kytydrdvapa which is full
of emotions.39Later theoreticiansare more explicit and on them LCvi based
his definition "la crdation d'objets par des moyens rnagiques ou des
p r o d d 6 analogiq~es."~~ A dramatist may be said to "evoke" the image
of certain events which cannot be acted, for instance because they are of a
supernatural character, but he cannot "evoke" the image of a per-
formance (prayoga).
The reason why Feistel, in spite of this difficulty, sticks to the explana-
tion of the NZilyaSBstra is, as we have seen, that the jarjaraprayoga, the
"performance with thejarjara", in his opinion did not justify the idea that
the jarjara was erected on the stage. Lienhard has since accepted Feistel's
"dkfY, S. Uvi, I, p. 91.
N$ 22.60 C, 20.59 KM,20.70 B sarvarasasarncisakrtam savidravcividravciirayam
vcipi, mifyam vibh6vyate yat tad vastiitthcipanam srnrtarn (v.1. jileyam B KM). Here
savidravam vidravciirayarn in C does not make sense. Abhinava, who reads vidravaih
saha, tair vihinath (ca), apparently had the reading of B and KM before him.The term
vastiitthcipana he explains by vastiimirn: bahii&[m arthZnaiIrn, utthcipanath: prasari-
g6gatanibandhanarn "a composition of things that have accidentally arrived (W-
curred?)." Cf. Srigaranandin 1372ff. mimirasayuktam bandhincith ce$fitarn.
M. M. Ghosh's translation "Elevation of the Plot" is no great help.
yathi tatraiva [that is, in the K.] Aigadcid abhidriiyam2pciyci Mandodaryci bhayath,
A@dasyo 'tsiihd, RLivqth ddyStvii tasyaiva hi "eteIrCTpi surcijitti" ityZdi vadato m a b ,
Rcivapasyci 'tikrodhd ...
Le thddtre indien I, p. 92. Cf. P W VI, col. 856 "das Erfinden von Dingen, das Vor-
ftihren unwirklicher Dinge." This is mainly based upon W a r i p u 2.59 mciyddyutth6-
gitarir vastu vastiitthcipa~rnQyate (copied in Siihityadarpm 6.134).
explanation, although he rightly objects to Feistel's translation "Auf-
ste1lung"for utthdpana : "Eine Auffiihrung kann schwerlich 'aufgestellt' .. .
erd den."^^
The solution proposed by Feistel therefore raises several questions. In
order to answer them it will be necessary to touch upon some aspects of
the Indian drama which will be dealt with at greater length elsewhere.
Here the following two questions will be discussed:
a. What was the original meaning of utthdpana?
b. What were the successive movements made with the jarjaja during
the so-called jarjaraprayoga?
They are so intertwined that it will not always be possible to keep the
one separated from the other.

4. According to the old tradition which has been preserved in the first
chapter of the NdiyaSdstra the legendary first performance of a drama
took place before a gathering of the gods while they were celebrating
Indra's Banner Festival. From this tradition some interesting conclusions
can be drawn about the origin of the Sanskrit drama, which will be
discussed elsewhere. The only thing that is of immediate importance in
the context of the present study is the fact that Indra is said to have given,
after the performance, his dhvaja to the actors. In so far as this dhvaja is
used in connection with dramatic performances it is called jarjara (NS.
1.69). Since there is no reason to question this tradition about the
fundamental identity of jarjara and i n d r ~ d h v a j athe
, ~ ~answer to the purely
fomal question whether the word utthdpana was used with respect to the
indradhvaja is bound to have some consequences for the jarjara. Besides,
the gods that are supposed to reside in the jarjard3 characterize it as a
replica of the cosmic tree.
Ever since the Kaus'ikaszitra and Garga, one of the oldest authorities
on the Indra festival,44the verb ut-thd- has been used with reference to the
l See IIJ, 15, p. 57f. Lienhard therefore suggests a different meaning for d t t h ~ ~ a n a ,
viz. "Antrieb" or "Beiebung" and refers (p. 58 n.3) to the meaning "antreiben, in Gang
bringen, beleben" of utthripayati.
LP In fact, the jarjara as described in the first chapter also has the function of Indra's
vajra, but this is ignored in the Indian tradition, apart from the fact that the vajra is
among the powers and deities that are said to reside in the jarjara (NS. 1.91,3.78), just
as the yiipa, the sacrificial stake, is said to be a vajra, e.g. $B. IV.4.2 vajro vai yiipo,
IV.4.10, etc.
a See fn. 54. For a similar cosmic symbolism connected with the sacrificial stake cf.,
e.g., $B. IV.4.13.
Garga is quoted as an authority by VarHhamihira in his BrhatsarhhitG and by the
V&trudharrnottarapurd~,see J. J . Meyer, Trilogie altindischer Machte und Feste der
Vegetation (Ziirich-Leipzig, 1937) 111, p. 17.
"rising" of Indra's pole.45The special religious importance attached to
the upright position of the indradhvaja (at the inauguration of the new
year, as some texts add) accounts for the stereotyped use of utthita
indradlzvaja in similes, such as R2m. 11.71.24 crit. ed. utthitau ca nara-
vydghrau ... varjdtapapariklinnau prthag indradlzvajdv iva. It is natural to
suppose that on the stage a similar position of the jarjara was equally
important. On the other hand it is clear that this does not necessarily
mean that the jarjara (for which preferably a bamboo staff was chosen)
had to be erected on the stage.

5. This leads us to a consideration of the second question: what exactly

46 Cf. iKariS.S. 140.3 Srava!retio 'tthdpanam, 140.7 athe'ndram utthdpayanti, 140.9
adbhutarh hi vimcinotthitam upatis[hante (cf. Gonda, JAOS. 87 (1967), pp. 419, 423),
VarBS. 42.37 uttblhati Sakraketau (loc. absol.), 38 Sravanena dvdda4ydm utthcipyo
'nyatra vd Sravandt, 58 utthcipayel Iakrma sahasracak~usa?r,59 ketum utthdpayec ca, 61
utthdnam i g a m aiubhatn, 67 utthitam, Utpala ad VarBS. 42.61 itrdradhvajasamutttid~~a~n
(by the side of ucchriyam&~a),Vi~nudharrnottarapur.11.155.18 yantreno 'tthdpanam
kurydc chakrahetoh samdhifa!~,23 utthdrze ca praveie ca, 24 (kdlavit sapurohitah)
prijayed utfhitati~keturn, 11.157.4 (in a mantra) u t t i ~ f h aKdlik~ipur&~a
, 90.1 Sakrotthdna-,
90.43 Sakrotthdpatia- (Meyer 111, p. 101), Bhavi~yapur.11.2.8.82 *Sakram utthdpayet
(Meyer, p. 48 n.l.), Hemiidra, Caturvargacitrtrinza~~i I11,2, p. 908 Sakradlivajotfhdpa~~a~n,
sthdpayec Chakram, Sakrotthdpalram, p. 910 (Brahtmip&purd!~a)Sakrarn rctthdpayed
rdjd, p. 91 1 Sakram utthdpayet, na cotthdp.va!i Pura~irdhara[l,kdle ca sthdpite Sakre,
$akram notthdpayed yadi, p. 912 (Bl~avi~yottara) p. 980
i~idradlivajasanzuftl~d~~an~,
utthdpayet, Mhbh. V.58.15 indraketirr ivo 'tthdjja sarvdbharanabhfi~itnh(v.1. indradhvaja
ivo 'tthitah)1.162.2,IX.16.52, Rdm. 11.71.9 crit.ed. ~itthdpyamdna!z~akrasyayantradl~vaja
iva c y u f a ) ~Agnipurdpa
, 102.26 samutthdpya sumantraii ca. Cf. Vis?ilrd~zarmotra,.apur.
11.1552 tatati Sakradl~vajasthdnam rnadhye samsthdpya yatnatah. Other idiomatic
expressions are ut-syj-, especially in the Mahdbhdrata: VI.114.84, VII.68.65 indradhvaja
ivo 'tsyjlah, VII.48.11 indradhvajdv ivo 'tsy.$au, often also uc-chri-, e.g. 1.504* line 5
ucchriyate, 1.64.14 ucchritadhvajasari~t~ibhan~, 1.162.2 Sakradhvajam ivo 'cchritam (v.1.
utthitam, see above), V11.63.7 parighai?~ ... ucchrifendradhvajopamaih, IX.16.52
indradhvaja ivo 'cchritah, 1.57.20 kriyate ('ty)ucchrayo nypai!~,VarBS. 42.96 ucchrayena
praveie, Utpala ad VarBS. ucchrite tasmin, Garga (ibidem) ucchriyamdna, V i ~ p u -
dharmottarapurd~ia11.154.13 sauvarnam ucchritam divya~nSakradhvajam iti Srutam,
11.157.1 and 2 $akrocchrdya-, Yiijiiavalkya 1.147 Sakrapdte fatho 'cchraye, HemBdri,
Caturvargacintdma?ri III,2, p. 401 atha Sakradhvajocchrriyavidhih, p. 402 ketwtd ...
ucchritena, ucchrdyo, nypatis tricchrayi~yati,p. 403 ucchrdyapnn, ketoh samucchraye
(Meyer 111, p. 28), III,2, p. 911 (Devipurdna) samucchrayet, ucchrdyo. In a different
context (with reference to the pillars of the playhouse) Abhinavagupta (I2, p. 58 line 6)
paraphrases utthdpanam with ucchrayanam. In the Rdniaya!la there is, by the side of
utthdpyarnrtrta (see above) also an occurrence of ud-dhri- in IV.16.37 (346' crit.ed.)
indradhvaja ivo 'ddhcitah. For the epics see J. J. Meyer, Das Weib i m alfindischen Epos,
p. 210 n.3, whose material has here been used. In Prakrit the verb ubbhei (*Grdhvayati)
is used in the story of Domuha, see Jacobi, Ausgewiihlte Erzal~hngenin Mdhdrds/~[ri,
p. 40, lines 11, 15, 18. In the Agnipur&ia 102.8 the equally rare brop- is used: dhvajci-
ropana, dropyamdna. In the Vedic language the verbs used for the raising of the
sacrificial stake are uc-chri-, ud-yam- and (un-)mi-. The exact reference in the simile
&S'. 1.10.1 lid varilS4m iva yemire is not clear.
happened during the jarjarapr~yoga?~~ Since opinion
detailed analysis of the text of the Nd?yaicistra is neces
When the seven parts of the nirgita have been performed behind the
scenes, the first thing to take place on the stage is the gita(ka), music in
praise of the gods. Thereupon a song is started which is called the
utthcipani dhruvda7 and serves as an introduction to the act of carrying
the jarjara onto the stage. The utthcipana itself consists of four "metrical
periods". At the beginning of the second parivarta the dtradhdra enters
(5.67 stitradhcirapraveiana), accompanied by his two assistants (p&&
One of them carries the jarjara (5.70), the other the golden
~cirirs'vika).~~
pitcher (bhyfgcira), while the stltradhdra, who has flowers in his hands,
goes between It is a very important moment. All three of them
must be ritually pure and cheerful ( s u ~ n a n a s )They
. ~ ~ all perform some
Cf., e.g., NS. 5.53 jarjarasyaprayoge tu tuffd vighnavindyakdb.
47 NS. 5.60 gitakdnte tatai cdpi kdryd hy utthcipani dhruvci. Konow, Dm indische
Drama, p. 23, defines utthcipana as 'kin Lied zur Aufstellung des Banners des Indra
(jarjara)." This is of course a slip of the pen. The dhruvcis serve as an introduction to a
new ariga ("member") of the piirvarariga, cf., e.g., the dhruvci parivartani (5.91) with
which the parivartana begins. For this reason alone it is clear that utthripani as the
name of a dhruvci is quite different from utthcipani "concluding verse" in KauiS. 82.31,
83.23 ; see the commentary on 83.21. See in general on the dhruvcis N$. 32. lff. and cf.,
e.g., 32.444 pr&idiki, nai~krdmiki, prdveiiki, and for the last (prciveiiki dhruvci 32.365),
P. V. Kane, History of Sanskrit Poetics, p. 43. Only once utthcipana is used in the par-
ticular sense of dhruvci utthcipani, viz. in the Appendix to chapter 5, v. 4. In this later
addition a more detailed description of the utthcipana is given (v. 15), which ends with
the words (v. 18) evam utthcipani kciryciplirvarangaprayoktybhib.
For the derivation of pdripcirivika from the Siitra word paripdriva "being at or by
one's side" (Wackernagel, Altindische Grammatik 1111, p. 313) see (Wackernagel-)De-
brunner, op. c. 1112, p. 309ff. It is a well-known fact that some editions (KM, C ) read
pciripcirivaka, but others (B with Abhinavagupta) pdripdrivika, and that both readings
are also found in the manuscripts of the Mahcibhcirata (XII.129.9, where in the preceding
verse the word is explained aspcirSve sthdpyafi).pciripcirivikamust be the older form as it
occurs not only in the Bhlsa manuscripts (e.g., Abhi~ekancifaka1.1.6) but also in the
ASvaghow fragments and in Kslidlsa, Mrilav. 1.1.2. (v.l.), Slgaranandin 1118.
48 5.70 bhpigdrajarjaradharaubhavetdm pdripririvikau, madhye tu slitradhrk tcibhyim
vrtab paiicapadim vrajet. Abhinavagupta here explains the two objects ritually:
bhyrigcirab : iaucdya, jarjaro : vighnaicintyai although in his note ad 1.60 B he identifies
this pitcher with Varuqa's present, cf. Is, p. 28 line 1 bhyrigdrah : pdripcirivikopayogi.
For an explanation of the mythological background of the three figures see the Proceed-
ings of the XXVIZ International Congress of Orientalists, A m Arbor, Michigan, 13th-19th
August 1967 (Wiesbaden 1971).
5.69 dik,dcib iucayai caiva praviieyub samam trayab. For the ritual importance of
being sumanas see, e.g., the references in IIJ, 4, pp. 272, 278 and especially such
descriptions of the Indra festival as Mhbh. 1.507*, lines 4-8 kri4itvci nrpaidsandt,
sabhcijayitvd rcijcinam krtvd narmciiritdb kathdb ramante ndgarcib sarve tathd jrina-
padaib saha, siitdi ca mdgadhcii caiva nafante nafanartakaih, prityci ca naraidrdda
surve cakrur mahotsavam and Hemldri, Caturvargacintimapi II,2, p. 409 line 6 ydvac
co 'cchrita dste tdvat paurdb sad5 hr~pib.
240
complicated steps and the stltradhdra then strews the flowers in the centre
of the stage (5.74). In accordance with the cosmic meaning of the various
parts of the stage after its consecration (3.20ff.), this part represents the
cosmic centre and BrahmH's seat (brdhma mandala). With a ceremonious
salutation to Brahma ends this second parivarta.
The third "period" (5.77-78) consists of three acts, viz. the pradak~ina
of the centre, the sipping of water (dcamana) and the taking up of the
iarjara (jarjaragrahaga). They are described as follows: the stltradhc?ra
rises quickly from the tnandala (in the centre).61 This can only be under-
stood in connection with his preceding salutation of BrahmH: "with the
hand three salutations should be made on the floor."52 Whatever may
have been the exact way in which these salutations were performed, it is
clear that the sctradlxira must have been kneeling or prostrate during the
"salutations". His next acts have a somewhat enigmatic character. The
special direction that the s6tradhdra should rise quickly is in line with what
follows. After going round the centre with very specific dance steps
(pradak~ina)he calls the assistant who carries the golden pitcher, performs
the ablution (Sauca) and sips water (dcamana) in the right way. "He
should sprinkle his body in due order with water and after having carefully
purified himself he should exactly (yatnatah) at the moment when the
[fourth and last] samnipdta [of this third metrical period] begins, grab the
iarjara which destroys (all) obstacles."53 Here the third parivarta ends.

6. A few observations may here be inserted. It is clear that the utthdpana,

although traditionally regarded as one separate "member" of the
pcrvarariga, in fact consists of two basically different elements. The
stltradhdra enters the stage "with the wish to worship Brahmii" (5.71
Brahnzano pajanecchayd). At the end of the salutation, however, he must
again purify himself very carefully, in spite of the fact that he was already
dikjita when entering the stage. This shows that his next act must have
been particularly important.
" 5.78 utthdya rna?1dalCt trirnarn.
5.75-76 abhivddandizi kdryd?ti tripi Imstena bhritale, kdlaprakarjahetoi ca pdddndm
pravibhdgata!z. M . M . Ghosh translates "and to measure the length of time during the
salutation he shoi~ldthrice touch the ground with his hand, and his steps should be
[suitably] divided." From a text-critical point of view the correctness of abhivdddni
(5.73 B), although metrically correct, is open to doubt. In 5.88 the verse recurs with
vandandny aiha kdry&i.
53 5.81 bh,rrigdrad/~drarndhiya Saucatiz cdpi samdcaret, yathdnydyaiiz trc kartvayd tena
liy dcarnanakriyd (82)dtrnaprokjaparn evd 'dbhijz kartavyar2 tu yaihdkrarnarn,prayatna-
kytaiarvena siiiradl~drerlayatt~atait(83) sarilnipdtasarnari~grdhyo jarjaro vighnajarjarah.
Cf. Cilappatikdranz 3.121-122.
 A second point can here only touched upon in passing. So far it
seems to have escaped notice that the relation between the jarjara and the
gods was much more complex than might be inferred from the name
indradhvaja. Although it is ideally identical with Indra's banner, the gods
who reside in its five joints are (from the top to the bottom) BrahmH,
Sa~kara,Visnu, Skanda and the ahHnHgas Sesa, VBsuki and T a k ~ a k a . ~ ~
Its cosmic character is, accordin beyond doubt. The deities mentioned
personify in different ways the totality of the Cosmos, which is symbolized
by the centre.66The jarjara represents the world tree, which is rooted in
the nether world (the NF~gas)~ and reaches into heaven (Brahmr?),
i
whereas in its very middle joint Visnu is located as the connecting link
l between the upper and nether worlds.=' Indra, however, is absent from the
1 very upakarana which was his present to the actors (1.59) and which
represents his dhvaja and vajra. As will be argued elsewhere the explana-
tion is to be found in Indra's character of a seasonal god. There are,
i indeed, if I am not mistaken, traits of a more complex nature in the figure
of Indra than are mostly recognized. At the end of the yeaIdS a new pole
I was erected (utthdpita), as a reiteration of Indra's cosmogonical act of
I
I
"propping up" (ut-tambh-) the world pillar to which the Vedic poets
often refer. As long as the banner festival lasted, the dhvaja was actually
I believed to be identical with IndraS9but after the removal of the pole the
I
god's role had temporarily ended and he reassumed his more modest task
i
of dikpda of the East (5.94). During the rest of the year the pillar belonged
to the gods of the totality, in particular to Brahmii, Visnu and the NBgas.
It was necessary to point in passing to these facts because they are of
importance for the interpretation of the siitradhdra's strange behaviour
and for the connection that exists between the two parts of the Utthdpana.
In spite of Indra's prominence in the first chapter of the Ndtyaidstra, it is
BrahmH who, as the grandfather of the world (lokapitdmaha), is actually
the highest god. He is the creator of dramatic art and to him, as a god of
the totality, the rna~dalain the centre of the stage is dedicated. This
cosmic centre, however, is also the place where the world tree was believed
to be standing. The siitradhdra, accordingly, has just been worshipping
lI the place where mythologically the tree either is standing (after Indra's
1.92-93, 3.78-79. The jarjara is paiicaparv5 caturgranthib (23.174).
For Skanda this is probable but it still has to be proved.
&"or the migas at the foot of the cosmic tree see India Maior (Congratulatory Volrrme
J. Gonda), p. 15 1 .
67 See IIJ, 13, p. 2 83.
Mhbh. 1.57.18 gate salhvatsare.
See J . J . Meyer, Trilogie 111, passim.
demiurgic act, the "second creation")60 or should be standing (before that
act). What then follows is the end of the ariga which bears the name utthd-
pana. This cannot mean that "die Auffiihrung des Schauspiels endlich in
Gang kommt" because it is only one of a series of religious ceremonies
which have nothing at all to do with the drama proper. With each of
these ceremonies special gods are pleased or appeased and in this con-
nection it is worthy of notice that the god who is specifically pleased at
the uttlzbpana is B ~ a h r n l not
, ~ ~Indra (whose name is not mentioned at
all). At first sight this might perhaps seem an argument which proves that
the utthupatza cannot have had any connection with the jarjara but this
conclusion, as has been argued above, would be wrong. First, utthdpana
was exactly the technical term for the erection of the indradhvaja at the
banner festival and can, therefore, be expected to have also been used
with reference to the jarjara, and, second, there was also a close connec-
tion between the jarjara and Brahmii, who resided on the one hand in its
upper joint and, on the other, in the mandda in the centre of the stage.'j2
Just as Indra's banner was supposed to be standing in the centre,G3
so the only appropriate place for the jarjara would have been in the
brdlzina nlandala. The chief difficulty lies in the words grdhyo jarjarah.
The verb is the characteristic term used in connection with the jarjara,
cf. 5.83, 86 jarjaragrahapu, 84 gyhitvd jarjaram. It is, however, also uzed
when one is about to fashion a jarjara from a stalk of bamboo, cf. 23.172
"One should carefully seize in accordance with the ritual rules the
bamboo in order to make a jarjara of it",693.176 "after having worship-
ped the bamboo according to the rules one should seize (grab?) if to
make a jarjara out of it".G5
The act of seizing or grabbing which forms the end of the thirdpari-
vartaGFis obviously a ritual of high importance: the preceding ablution
proves this sufficiently. Why then the element of hurry in it? Not only
must the stltradlzdra rise quickly, but the third and fourth "metrical
periods" (in the latter of which he o ~ l makes
y movements with his hands
See History of Religions 10, p. 104ff.
See 5.50 tathd cotthdpane yukte Brahmri tu$o bhaved iha "and also, when the
utfhdpat~ahas been performed, Brahrnk is pleased". The last word is here, as often, a
mere stop-gap.
See 1.94 rarigapi(hasya madhye tu svayam Brahmri pratijfhitah, i.gyart/zatn rariga-
madhye tu kriyate prr~pamok~anam and 3.23 (at the consecration of the playhouse)
padmopavigam Brahmrinam tasya madhye niveiayet.
Vi,r!zud/zartnottarapurrina11.155.2 madhye.
@"3. 173 sarizgrdhyo vidhitzri venur jarjarrirtham prayatnatatz.
G5 23.176 uprisya vidhivad veputn ggyhpiycij jarjaram prati.
5.83-84 pradak~ipridyovijiieyo jarjaragrahanantakah.
 d feet) are also to be perfo
The second assistant is nowhere referred to, but since it is not S
that he puts the jarjara down, it must be as
. . in his hand. the stltradhdra, after salutations to god
8, calls the ass who carries the golden pitcher, the other
assistant probably also comes nearer. Immediately after the ablution the
szitradhira seizes the jarjara, apparently from his assistant's hand.
hy, then, does this "member" bear the name utthcipana? The only
one who is "saluted" rahmii (5.75 abhivandyo Brahmd), whereas the
worship of the jarjara s place much later (5.102 piijanaliz jarjarasya).
It seems that at the time of Abhinavagupta (circa 1000 A.D.) there was no
longer a living tradition and that he, too, groped in the dark: "For nothin
whatsoever has been erected by [the ritual which merely consists of] the
salutation of Brahma and the addressing of the jarjara. Also in the
beginning of the parivartana [its?] function of utthipaka is due to its
serving to appease the '~bstacles'".~~
During the next ariga, viz, the parivartana, the st-itradhdra"salutes" the
deities of the various quarters (5.94-96). No mention is made of the
jarjara in this passage (5.90-101).
What follows I consider a separate member of thepiirvaranga, although
the Indian theorists regarded it as part of the p a r i ~ a r t a n a A . ~ ~fourth
person, the caturthakdra, enters the stage with flowers in his hands (5.102
pu,spipi pragrhya), in the prescribed way.69It is a meaningful moment,
which is elsewhere referred to as "the entrance of the fourth man".70
The flowers are offered aspzijd to the jarjara, to the ~ r c h e s t r aand
, ~ ~to the

See Abhinavagupta (vol. I*,p. 234 line 2) nu hi brahmcibhiviidamjarjariibhimantra-

?amiitre(m kim cid apy utthcipitam bhavati vighna?inmanopayogitveno 'tthcipakatvalit
parivartanridclv api syiit. For the vighnas cf. 1.65ff., 5.53 jarjarasya prayoge tu tu@Z
vighnaviniiyakiih.
But see below, fn. 72.
Abhinavagupta here refers to the sthcinarh tu vai,ytavam, with which the sritradhZra
and his assistants had entered the stage (5.69).
See 33.227+ (p. 174 line 15 C) = 34.198+ (p. 641 line 10 KM)caturthakcirapraveie
(-praveie~uKM).
Or the instruments! Ghosh renders 5.103 kutapasya sarvasya by "all the musical
instruments", although elsewhere he takes kutapa as meaning "the members of the
orchestra", see Introduction to his translation, p. LII, Text P, Introduction p. XLVII,
where he points to contradicting (or different) explanations by Abhinavagupta. Cf.,
e.g., P, p. 64 line 5f. (ad 2.82 B) saliphe~akagciyam-vc3daka-samriha~, but otherwise,
e.g., Is,p. 72 line 6. Also Subba Rao in Appendix 6 to the the first volume (2nd ed.) of
the Baroda edition (Ia, p. 445) refers to "the group of musicians"; cf. Feistel, p. 66.
That aprijci to the instruments is by no means excluded is proved by modem practices,
see below, p. 258.
~Qtradhclra.~~
During his pzijana the singing is interrupted, only stobhas
being allowed (5.104). After offering his p2jd the "fourth man" makes his
exit.

7. Although this study only has a limited aim, viz. to examine what the
Nci{yaiclstra can teach us about the nature of the utthcfpana, it may be
useful to interrupt the line of argument for a while and to insert, as a
parergon, an attempt at an interpretation, in the light of mythology, of
the data thus far gathered. These considerations, which will be only
briefly sketched, will be discussed in a wider context in Varu~aand
Vidzi~aka.
It has been argued elsewhere73that, since thejarjara is said to be Indra's
plesent to the actors, and the golden pitcher Varuna's (1.59-60), and since
the pitcher is known as a symbol of the nether world,74the two assistants
who enter the stage with these two paraphernalia can be explained as
impersonating Indra and Varupa themselves and as standing for upper
world and nether world. It is hard not to recognize the importance of
these symbols in the light of what the first chapter says about them. It has
further been observed that if this conclusion is correct, the sdtradhdra
who goes between them in all likelihood stands for the totality of the
dualistic world, that is, for BrahmX In that case the normal symbolism of
right versus left demands that the jarjaradhara went to the right of him,
and the bhyiigdradhdra on his left. If so, the vandana (not pzija) of
Brahms was a confirmation of the s6tradhdraYsfunctional identity
with the god of the centre. The number of salutations is also meaningful
in this connection, because the number three symbolizes the idea of
totality. The sdtradh8ra then hastily rises and carefully performs an
ablution, after which he seizes the jarjara.
If this is correct, the sctradhdra now entirely personifies the sacred

'' See 5.101f. parivartanam evaliz sycit tasyci 'nte [!l praviiet tatah (102) caturthakcirah
pu~pciqipragrhyavidhipiirvakam, yathdvat tena kartavyarh plijanam jarjarasya tu (103)
kutapasya ca sarvasya siitradf~drasyacaiva hi. The first line is particularly interesting,
because its author seems to have considered the parivartana ended at the moment of
the "fourth man's" entrance. This would conflict with 5.156 C (150 B), where the
parivartana apparently is referred to as caturtiiakdradattcibhi!~ sutnanobhir alanikyte
"adorned with flowers given by the fourth man". Feistel, p. 66f., considers the possibility
that this "Zwischenglied" is a later intrusion. For reasons given sub 7 it would rather
seem a very old and essential part of the ritual, which however got out of use at an early
date because its meaning was no longer understood.
See the summary in the Proceedings referred to above, fn. 49.
For references see "The heavenly Bucket", India Maior (Congratulatory Volume
J. Gonda), p. 144ff.
 world. We is the god of d while holding the
upright in his hand he represents the cosmic centre. That the jarjara had
to be erected follows from its function as a cosmic symbol. If so, the use of
the term utthdpana for the "member" which ends with the seizing of the
jarjara must be due to the fact that at this moment the erection
world pillar in the centre of the universe is re-enacted by the szi
with the jarjara.
The dtradhtira then ta S five steps in the direction of the orchestra
(5.85), for reasons which still remain to be explained, and greets the gods
of the four quarters. Through this act the relation between the high god
in the centre and the other gods is rec~nfirmed.~~The description is interest-
ing (5.95E.): he first greets Indra in the east, Yama in the south, Varuna in
the west and Kubera in the north, which is the normal group of Dikpmilas.
Then, however, he greets with three strides (5.99 tripadaih, v.1. tripadya)
the three gods in the centre Rudra (Siva), Brahmmi and Visnu, who are
characterized by a "masculine", a "neuter" and a "feminine" step. All
three are gods of the centre and it was seen above that as such they also
reside in the three upper joints of the bamboo staff7% which functions as
the jarjara. The gods of the all-embracing totality always come last in the
enumeration of the system of classification. It is not stated in the text
just how the siitradhdra pays honour to them: whether in the brtihma
mag&la, which properly speaking belongs to Brahmmi alone, or in the
jarjara, which he holds in his hand and in which they all three reside.
This is the end of the parivartana (5.101), which apparently owes its
name to the salutations to the four quarters. In the light of the inter-
pretation here proposed it is more likely that the szitradhdra remains
,standing in the centre of the stage, which he personifies, while greeting
the gods of the four points of the compass, than that he goes round the
stage. The meaning of parivartana must then have been "turning-round"
(Feistel), not "walking-round" (Ghosh).
So far all acts have concerned the world of the gods only. The "fourth
man" who now enters from "outside" without having purified himself
obviously belongs to a different world. His only task is to strew flowers
as a piijd to the jarjara, which has just been erected in the hand of
Brahmmilsiitradhdra, to the instruments of the orchestra, and lastly to the
slltradhtira. It deserves notice that of the two paraphernalia which the
See 5.92 calitaib pddavinycisair vandydd devdn [probably corrupt for vandyd devd,
so 5.90 B] yathddijam, 95 tato 'bhivddanam kurydd devatdmim yathddiJ'am, although
5.92, 93, 94 are lacking in some editions. See, e.g. 5.91ff. KM.
See above fn. 54 and for the vepu (bamboo) from which the jarjara is fashioned the
notes 64 and 65.
 assistants have brought with them on the stage, only the jarjara is an
object ofptjd. This is not, because the golden pitcher as an emblem of the
nether world (better known under the name kalaSa or kunzbha) was less
sacred,77 but because in this particular ritual all attention was concen-
trated upon the newly erected dltvaja. That Abhinavagupta in the eleventh
century A.D. no longer understood the meaning of this episode of the
"fourth man" may be inferred from the purely etymological explanation
with which he contents himself.7s
The man who enters this divine world apparently represents the world
of human beings. His first act is to worship the erected jnrjara, just as
men every year worshipped Indra's banner when it was erected again.
It seems to me that we here touch upon the basic meaning of the utthipana
in the pikvarariga. According to the mythical tale told in the first chapter
the first dramatic performance took place in the world of the gods on the
occasion of Indra's banner festival. The importance of this "legend" is
that it preserves not a historical but a religious tradition. It forn~ulates
the ideal norm for a dramatic performance. The Ndtyakistra tells us that
there was an old tradition about the first drama having been the Amlta-
manthana, accordingly a re-enactment of the cosmogony. On the other
hand, Indra's banner was erected "at the end of the year" (according to
the Mahdbhdrata) and it is not surprising that at that time a representation
of the cosmogony should have been performed. Later on, dramas came
to be performed at other festivals, too, but the ideal norm was observed
by inserting a miniature banner festival into theptrvarariga. To that end a
man was needed to confirm the sacred function of the jarjara and to
demonstrate by his ptjd that the dlzvajnrnaha was celebrated on the stage.
Besides, the jarjara, as a vajra, kept off the demons.
It can also be understood that his'ptjri to the siitradlzrira meant the
decisive confirmation that the latter impersonated god Brahmii. The
stobhas, which at this moment replaced the normal song, underlined by
their "magical" character the religious importance of the fourth man's act.
It is not surprising that immediately after this "confirmation" the sttra-
dhira, as the All-Father himself, pronounces the benediction by which he
blesses the earth, the king, the brahmins and cows, etc. The benediction
is accompanied by "amen" (evam astu) pronounced by the two represen-
tatives of heaven and earth, who confirm its effectiveness. A comparison

" AS has been remarked above, the identity of this golden pitcher with the pitcher
wbich was Varuna's present'to the actors is stressed by Abhinavagupta. Seen. 49.
Cf. 12,p. 235 line 8: slitradh&asyapdripcfrSvikayos' ca tisrah kriyah, tadvyatirikfam
ca cafurtham karma karoti 'ti caturthakCrah.
with the prayer pronounced after the piijd to the jarjara in 3.11-13 is
instructive. There are similarities but in the latter passage a man is
praying, whereas in the benediction it is the god who is thought of as
speaking and blessing the world (35.96 sarvam stauti hi lokam).
This, it seems to me, was the meaning of the utthdpana. The explanation
leaves at least two questions open. First, the five steps of the siitradhdra
in the direction of the orchestra, and the "fourth man's" piijd to it, remain
unexplained. If the structural approach on which the preceding inter-
pretation is based has any merit, it is certainly this that it shows the gaps
in the theory which need being filled. That divine honours were paid to
the instruments (or perhaps the musicians) of the orchestra is unquestion-
able but the meaning of this piijd is not yet clear. In view, however, of the
great parallelism between the consecration of a new playhouse as
described in Chapter Three and the ritual of the piirvaranga (which in
point of fact is a consecration of the stage) attention may be drawn to the
following stanza:

3.1 1 C sampaya devatc* sarv* kutapam *sa&prapiijya ca

jarjarciya prayuiijita piijdm ndfyaprasiddhaye
3.12 B, 11 KM snmpiijya sarwin ekatra kutapam samprayujya ca
jarjardya prayuiijita piijdm nirfyaprasiddhaye

Of the two readings devatdh and ekatra, one seems to be a corruption of

the other. If devatdh is correct, samprapijya has a fairly good chance of
also being the correct reading. Ghosh gives it as a conjectural emenda-
tion in his edition of the text (C2), but it is in fact the reading of two
manuscripts used for the second edition of Baroda and it had been
adopted in KM1. Although samprapiijayati is rare (the PW records only
one occurrence from the Harivamia), prapiijayati is not uncommon and
prefixation of sam- for metrical purposes is well known. In the light of the
pijd offered by the caturthakdra it would be attractive to translate "After
offering- pijd to all deities and to the orchestra he should offer it to the
-

jarjara". Modern practices can no doubt deepen our insight into what the
old text actually means. Jones's recent observation about the worship of
the copper drum (Tarnil rnulavu, Sanskrit mydanga) in Kerala, which the
CBkyBr consider a brahrnacdrin and which they decorate ac~ordingly,'~
shows what enlightenment can be expected from this side. Abhinavagupta,
See Clifford R. Jones, "Source Materials for the Construction of the N@yamao#apa
in the Silparatna and the ~antrasamuccaya Silpa Bhaam, JAOS, 93 (1973), pp. 286296,
especially p. 295 n. 29. Raghavan's study "The Multifaced Drum" on the paiica-
mukhavddya, there referred to, was inaccessible to me. Cf. 1.84 bh6& same divaukasab
"In the rn~rsicalinstnunentsare all gods" and G. L. Hart, JAOS, 94 (1974), p. 159.
248
however, read ekatra, which is the sole condition on which satizprayujya
could be maintained, as kzttapam sampruyujya would not make sense.
From the lapidary style of his commentary the exact text of the manuscript
he had in hand cannot be r e c o n s t r u ~ t e d . ~ ~
The second point that must be left an open question is the terminology
of "the seizing of the jarjara" (jarjaragral~ana).The term may suggest that
that this had a brutal, aggressive character. The crucial moment in the
cosmogony, when Indra "propped up" the sky and, by separating heaven
and earth, created the dualistic cosmos, was an episode in his strife with
the Asuras. The szitradhcfra seizes the jarjara from his assistant who, as
the bearer of Indra's emblem, has been explained as standing for Indra
himself. If the szi'tradlzdra actually grabs it, this might represent a re-
enactment of the element of strife that accompanied the first utthcfpana
of the cosmogonical myth. In any case, however, the consecration of the
jarjara on the stage is performed by the szitradhcfra as representing the
highest god.
8. After this digression, which was only meant to show how the many
details can fit into a meaningful pattern, the following "members" of the
piirvarailga will be briefly examined.
After the ptljd offered by the "fourth man" a new d/zruvd introduces the
ncindf "benediction" (5.105-1 15) already referred to. The next izqk&
vaky~tcfdlzrzivd forms the transition to another important episode. This
d/zruvd is called jarjaradokadars'ikcf, which name, although not entirely
clear,81 obviously refers to the following ailga. It is followed by the praise
The manuscripts apparently read: ekatre 'ti: sthanflabhicbhcige. kutapam iti:
caturvidlicitodyabhdndcitri.ekatra niveianatir jarjarasya picjcirrharn avastlrcipanam dryena
nivasanniyam iti bhdve kytyalr. In the second edition of Baroda the last words are
emended to avasthcipanam. nivasafy upcisyam ili blrcive kyfya(i. Instead of it, the end of
the passage should rather be read as follows: ekatra nivehnam, jarjarasya picjcirtham
avasthcipanam, sritradlicirena "nivasanti 'mum" iti bhdve, kytyam, the translation of the
whole passage then being "ekatra: on an open, bare piece of ground; krrfapam: the four
kinds of musical instruments [viz. tata, avanaddira, ghana and siisira]. The sritradhdra
must put [all gods] in one place, that is, array them [in the jarjara] for the purpose
of thepicjci to the jarjara, because they can be considered to reside in it." For the act of
niveiaiia cf., e.g., 3.20 kciryatir devafcincim niveianam (thus 3.21 B, but daivatcindm C).
In no case should this be taken as an argument that the jarjara was put down during the
Utthlpana of the prirvarafiga. On the whole Abhinavagupta's interpretation of ekatra
... samprayujya seems rather forced, but from a text-critical point of view the question
of the original reading of this verse must be left open. As for the niveiana, if actually
referring to the jarjara, a parallel is met with in the belief that the sacrifical stake be-
longed to all deities, who resided in various parts of it: TS. VI.3.4.7 sci v& esri sarvade-
vatyb ycid yLPo, yad yhPam mirzdti sdrvd evci devcirdh princiti. Cf. $B. IV.4.10 and n. 43.
Abhinavagupta 12, p. 238 lines 2-3 explains it as follows: jarjarastutiiloko yarab
sritrditcirega pafhyate, 'to jayater dariikd tatpurassari ' t y arthah "that is, because the
(stotra, stava) of a god who is just then specially worshippede2 or of
whom the king is a special devotee, or else of the class of b r a h m i n ~ . ~ ~
Then the szitradhcira recites with a deep voice the jarjarailoka (5.1 17-118),
which is known as rangadvdra This is immediately followed by
another s'loka which accompanies the "inclining" of the jarjara. Here it is
necessary to quote the text (5.118b-120):

gaditvri jarjaraSlokam rafigadvcire ca yat smrtam

(1 19) pafhed anyam Slokam jarjarasya vinCmnam
jarjaram namayitvd tu tatad cririm prayojayet
(1 20) priripcirSvikayoS ca syrit paicimenC 'pasarpqam

"After having pronouncede6the s'loka [in praise] of the jarjara and what
is taught [as to be pronounced] in the raitgadvcira (?),86 [the szitradhcira]
recites another Sloka, which accompanies the inclining of the jarjara.
Having inclined tbe jarjara he thereupon performs a Cdri and the two
assistants retreat backwards."
Since Haraprasad Shastria7already rightly interpreted this as "at the
end of this another verse is to be recited for dismissing the jarjara and
laying it down", it is curious that Konow should have misunderstood
these words as meaning "Der sMradhFira ... vetbeugt sich vor dem Banner
des Indra."8e The word vindmana only means "inclining"89 and the

slitradhdra utters the Sloka which contains a praise of the jarjara, therefore it is called
'pointing ahead to the praise of the jarjara', that is, preceding it." For jayater, which
makes no sense, I read ja[rjarastu]ter, cf. 5.26 KM jarjaraflokadariikd (-td C ) and see
for jarjarastuti SBgaranandin 1130-31 ; dariikd is "introducing" ("einfiihrt" Feistel, p.
70).
See 5.117-118. It is interesting that Abhinavagupta (IP,p. 238 line 6) mentions as a
possible case that the drama is performed at the beginning of a festival for a special
god: ydli? devatLim uddiiyo 'tsavMau ndfyam kytam sd tatra stotavyd.
m For 5.118 C brahmapastavah (=brahmapas stavah), 5.114 B brahmapas stavam
(also Feistel 5.111) the manuscript a of the Baroda edition reads brdhmawtavam, cf.
5.116 KM brdhmapastavam. This reading is apparently corrupt, although it occurs in
Abhinavagupta's commentary: sa [viz. prek.yGpatiS] ced uddsinas tarhi brdhmapa[E]
ayam asau "If the sponsor of the play has no preference, this or that brahrnin."
If the reading rangadvdram iti smrtam (two manuscripts, followed by Feistel) is
correct. For the rarigadvrirn cf. 5.14, 27, 53 and below, p. 265. All editions read
rarigadvdre (B C KM Raghuvda).
For this word cf. 33.227+ (p. 175 line 3 C )jarjara[iloka]niga[da]nam.The corre-
sponding words of this extremely corrupt part of this chapter are lacking in KM, p. 641.
"in what is calied the Rangadv&ra9', Ghosh (but ca?). See above. n. 84.
See JRASB, 5 (l%), p. 357.
Das indische Drama (192O), p. 24.
Cf., e.g., NS. 25.109 B Siraso 'tha vincimanam, 25.111 KM Sirasas' ca vincimandt
(26.113 C has a different reading), 13.174 C dak~ipahvinametpdr4vam and Abhinava-
gupta Is, p. 101 line 9, who quotes from Kohala vinamanum "moving down".
 correctness of this reading cannot reasonably be doubted.80 Up to this
.moment the siitradhdra must have held the jarjara upright in his hand.
The same is prescribed for the Indra pole, viz. that it should stand upright
which is, indeed, symbolic of life.s1 There can be no doubt, therefore, that
after the jarjaradoka the jarjara should incline. The double worship of the
staff, first by the "fourth man" and now by the szitradhdra himself, has
given rise to wrong conclu~ions.~~ It is, however, characteristic of the
indradhaja that the highest honour is paid to it immediately before it is
pulled down.s3The importance of this second jarjarailoka is indicated by
the name of the dhruvd which introduces it and marks it as a separate
"member" of thepgrvarafiga. The only difference between the indradhaja
and the jarjara is that the first, which is actually pulled down, is said to

5.1 15 B, 117 KM jarjarasya vincZranam,jarjaram namayitvd tu. The correct reading

occurs in a manuscript of the Palace Library, Trivandrum.
*l In IIJ, 4, p. 232, an attempt has been made to prove that the expression iirdhvd
sthd- in the Rigveda symbolizes life versus death (for which parallels can be found in
Ancient Egypt). For the indradhvaja the vertical position is also prescribed, with the
exception of VarBS. 42.60 arinagare natigra~nkdrayed dvi&~adltdya"one should make
the top incline towards the town of one's enemy, to kill the adversary", In spite
of this magical practice the indradhvaja is worshipped when it "stands up" (utthita,
VarBS. 42.67, etc.) and in a mantra it is addressed with the word utti~fha(VQnu-
dharnrottara II.157.4), which may be compared with BS. 111.61.3 iirdhvb ti~fhasy
amttasya ketli)~"thou [ U ~ a sstandest
] erect as the symbol of life", IV.6.1 iirdhv6 ... agne
tis!ha "stand upright, 0 Agni". Quite different from the magic of the VarBS. is the
Vedic practice of making the yiipa incline towards the East, e.g., $B. IV.4.17 prdcim
nalnaty [so to be read for sainnamaty Eelsingh] etad vai Vi,ytohparamam padam "he
makes it incline towards the East" (not "he turns," Gonda, Aspects of early Vi$?u&m,
p. 94, BollCe, translation, p. 99). This is parallel to the prescription that the sacrificial
ground should be sloping towards the East or North(-East), cf. MS. 111.8.4 (94,4),
$B. 11.10.20, ,$B. 111.1.1.2, Caland, ad ApSS. X.20.1. However, also with the ytipa
magic may be performed, cf. KS. XXVI.6 (129.8) agner ardhid zipanatam yiipasya
kurycid, bahijfdn nir!zatam; agner vd ardhdd upanatena yajamhiya lokarit karoti,
bahisfdrt nirnatena bhrftrvyariz yajfifn nirnudate (otherwise MS. 111.9.4: 118,lO).
* V e i s t e l , p. 112, writes "Diese Preisung lasst die in 5.100 sq. erwahnte Jarjaraver-
ehrung uberflussig erscheinen" and suggests expunging the whole passage of the
caturthakdra (5.101b-104b C) as a later interpolation.
O3 Cf. Vi$?~udharmottara 11.155.25-26 sampiijayed Rdma tadd dinacatujfayam,paricame
divase prdpfe .$a]krakeftrri~visarjayet (26) piijayitvd rnahdbhdga balena caturariginci
"nachdem er sie zusammen mit seinem vierteiligen Heer (noch besonders) verehrt hat"
(J. J. Meyer), VarBS. 42.67 dinacatu~fayarnutthitam arcitam samabhipiijya nrpo 'hani
paricame, prakytibhi?~saha lak,sma visarjayed balabhidah svabalfbhivyddhaye, Hemiidri,
Caturvargacintdmani II,2, p. 402 ghdte pdte kurydd ucchrdyane ycidrii piijd, Domuha-
cariyam (Jacobi, Ausgewahlte Erzahlungen in Mdhdrdshyri, p. 40 line 22f.) cigay6
punpimd. piio malzdviccha~ena kusumavatthdihim Dornuhardipci indakeic. mahci-
ticraravega annammi dipe pa(iio meigie.
 "loosened from its props9', to '5"fall",9Vobe "dismi~sed"~~ or
whereas the movements of the jarjara in the hand of the
8ra only symbolically imitate its erection and pulling do
The inclining of the jarjara is obviously parallel to the visarga of the
indradhvaja. Visarjana is the eomnon word for the desecration of images
of gods. Consequently, from this ent onwards the jarjara must not
ven for a single moment, be erected again. It has ceased to be a sacre
object.

standing, their faces

arpapam have caused

on seems to imply that he n

from the moment of dese

referred to as a reiteration of the

Rdm. IV.17.2 crit.

ajciv iva mahendrasya

niva&o, 29 p a h n t a h . Cf. Rcim. IV.16.27 crit. ed. vibhra~'itendradhv&vat, 11.71.9

crit. d.utihcipyamcinab Sakrasya yantradhvaja iva cyutab.
VarBS. 42.67, Yijnudharmottara II.155.25 visarjayet, VarBS. 42.56 visarga-,
ram kunyciycim tu visarjayet, p. 911
niii supte visarjayet, p. 912 latpcitas' ca
a tu, arcayitvci bhagavatim tatab sadyo visarjayet.
the lines u t t i ~ f h t evividhamandiravmndabandho,

m tu vdmnrh cci'dhomukhak bhujain, na

hould be observed that the
has a stereotyped character, cf. 13.171f. (12.133f. KM) krtvci '
o&ukham karam ncibhipradeie vinyasya
sfhsinaritk~ivci'vahitthaih ca v d m m ccd
a savyuri2 ca kh@akcim&hnm.
See Abhinavagupta Is, p.
 ,10° it is natural that Um2 should especially be pleased at the
performance of this dance.lo1
As for the direction "near the navel" (mibhipradeie), it is reminiscent
of the Vedic Aivamedha sacrifice, in which the sacrificer, seated on a
"throne-seat" after his consecration, is lifted up knee-high, then navel-
high, then as high as the mouth.lo2Jeannine Auboyer rightly explains this
as "une prise de possession de 1'univers".lo3 If the inclining (vindmana) of
the jarjara has been interpreted correctly as a symbolical imitation of the
pulling down of the indradhvaja, then the horizontal positionlo4 must
indicate its lying on the earth and the navel can be symbolic of the earth,
intermediate between nether world and heaven. Abhinavagupta has, if
am not mistaken, been misled by purely auditive associations and see
to have misinterpreted the passage.lo5
After the Cdri the sdtrad/z&a performs aparivartana "turning-round'"
(cf. 5.101), with the jarjara in his (left?) hand, recites two Slokas and then
also retreats "backwards" (v.1.: "'with his face towards the audience").
He thus joins the assistants and hands the jarjara [no doubt still in a
horizontal position, although this is not stated] over to one assistant
hat is, thej a r j a r a d h ~ r a ]Were
. ~ ~ ~a in a few words must be said about the
xt. All editions (5.127 read haste (tu) nyasya
jarjaram uttamrrm. The la het of the indradhvaja in
m uttamam) and in Vi~~udharnzottara 11.154.13. Two manu-
however, have the obviously corrupt reading unnatam. The
circumstance that one of these rnanusc pts is from North India, whereas
the other composite source is based on uth Indian manuscripts presents
l* Cf. 5.122-123 yathb parvai MaheSvarah, saho 'mayd kriditavzn. Abhinavagupta
interprets this as follows (ad 5.119 B): sahornaye 'ti: Srrigdrasya pracarapdt ["with
movements depicting the erotic sentiment", cf. 5.271, "Umd tu$d" iti coktatvrid
[quotation from 5.541 iti bhbvah. kriditava: strice~titene' f i bhdvall.
l" See 5.54 tafhli cdryd&prayrtkfdybnz Umd tlc~fdbhaved iha.
lo9 See P.-E. Dumont, L'Apamedha (1927), p. 236.
lo3 See Jeannine Auboyer, Le tr6ne et son symbolisme dam Z'lnde ancienne (1949),

p. 140. Cf. also the comment of 2%. XII.8.3.20-21.

l 0 V h e reading trrldd/i.rtarn "kept in ba1ance"is perfectly clear. There is no reason to
read talddlzytarn instead of it (see Ghosh, Translation, p. 92 n. 2).
lo6 The formal similarity of this verse with 13.172 and 183 (see above, n. 98) led
Abhinavagupta to explain ndbhideie as a reference to the khafakdrnukha gait of women.
Cf. P, p. 240 line 9 khafakdmltkizamadhye strigatau hi vak~yate"nbbhipradeie vinyasya
am ca klzafakdrnukham" iti. I doubt if there is any connection, but cf. Feistd, p. 75
gen Stockes reprasentiert". The umbilical region is also men-
in different contexts, e.g., 13.174 nyasen ndbbhitate tatah.
ikahaste tu nyasya jarjaram uttanzam ( B KM
manuscripts da and rnapdrip6rSvakayor haste (
ries), which is impossible.
a curious palaeographicaf since tta can be misread as nna in so
forms of Devanagari as alayalam script. A
"lifted up, raised" is ritually impossible. Feistel, p. 77,
"wieder erhoben" and diffidently suggests reading ut
streckt". It is not quite clear why there should be any need for emending
unnatam which, taken by itself, is impeccable in the sense of ucchrita,
am is improbable in classical Sanskrit for purely philological
e that as it may, this emendation had to be mentioned
because Feistel, although rightly transl ing on p. 77 "den ausgezeich-
neten Jarjara", on p. 116f, silently accepts his conjecture as the basis for
his conclusion: "doch bleibt er ... stets in der Nand, zeitweilig des einen
Begleiters, zeitweilig in der des Sgtradfrdra selbst, der mit ihm
gungen ausfiihrt, ihn senkt und wieder aufrichtet" [italics mine].
incorrect.
Since at the moment when the szlitradhcira returned the jarjara to his
assistant he must symbolically have represented the visarga ("
of the indradhvaja it is interesting that the NSfyas'&tra else
term jarjaramok~a.As molqa can be used in the sense of m
etc.", this was apparently the technical term for this act of
It occurs in a description of the professional evduation o
the performance. See 27.40

jarjaramok~asyci'nte ndlikasiddhid ca lekhyasiddfzis' ca

kartavyci tv iha satatad ndiye 'smin prdknikaih samyakIw

Ghosh, although reading antar for ante in his text, still transl
"After the putting down of the Jarjara [by the Director] in a dramatic
production the Assessors (prds'nika)should always achieve in due manner
the accuracy of timing and of recording [of
points]." From this verse Abhinavagupta i
the parvarakga was also
" j ~ d g e s " .'This
~ ~ ~can only
lo7 According to the PW it is only attested in the Atharvaveda ( ~ ~ S a u II.7.3, it.
.29.3). The meaning assigned to it is in classical Sanskrit expr
unnata, utk~ipta,uddhrta, etc.
'08 Other editions have (27.41 B, 39 K
niilikciycis tu, krrrtavyas tv iha satatak
K M , footnote) jarjaramokgzsyd 'ntan
satatarh mitye 'smi<n>pr&'nikaih sanzyak. C reads nrllfka.
Og Abhinavagupta ID1, p. 317 line 7 '~arjaramok~asyd 'nta" Sti: p
'pipariksya iti dariayati.
 eferring to the "dismissing" of the bamboo sta
, accordingly, excluded all acting up to this point (the rai?gadv&a,
.M), as belonging to religion and lying outside the sphere of dramatic
rformance, from being judged by aesthetic standards.

hat occurs after this episode, interesting though it is for our

into the nature of the pCrvarariga, is of minor importance for th
under discussion. The sdtradhdra comes again to the front of the
rariga and calls his assistants. It has been argued elsewhere that the ensuin
Trigata between the two assistants, one of whom is now called vidiisaka,
ica of the Vedic verbal contest between Indra and Varuna (RS.
2) and that the bhpigdradhEra, as the personification of Varuna, here
e vid@akal10 This identity is also expresssed by his traditional
which largely corresponds to that of the Vedic jurnbakd. The
sistant, however, is the rational participant in this sariZjaIpa

ever that may be, the only (and last) point which is of interest in
ext of this study is that the assistants, when coming back to the
the stage, can no longer have carried their paraphernalia with
. It is generally assumed, and rightly so, that they have all the time
remained on the stage because the text does not make mention of their exit
because they are present at the back when the sdtradhdra joins them.
n the other hand, when one of the assistants comes back as the vidbaka,
e must have been recognizable as such. In view of our complete ignorance
about Indian stage conventions in the period when this passage W
written, it is not advisable to draw definite conclusions from this.
deserves notice, however, that immediately after the i
jarjara the two assistants had receded to the back of
cimend 'pasarpanam may possibly have had a threefold function. First,
it may have marked a caesura between the preceding worship of the
jarjara, in which the assistants had a very subordinate part, and the verbal
emtest, following the desecration of the jarjara, in which contest they
represent the two cosmic moieties under the patronage of the
est god. Second, it may have allowed them somehow to get rid of
e ~paraphernalia
t for which there was no longer any use: one of them
ad lost its sacred character, the other was no longer needed for ablutions
nd their task of marking the different nature of the two assistants was
en over by the make-up of the vidti:~aka,who must
 ehow have been distinct from the other assistant.ll1 Third,
aven given the representative of Varuna (that is, the one who had
the golden pitcher) some opportunity to make up as the vid@aka.

11. Conclusions. This study leads us to the conclusion that form

scholars were right in referring the term utthdpana to the erection of the
jarjara, and that Feistel was right, too, in denying that the jarjara was
actually placed on the stage. What in fact happened was that the sctradhdra
rformed the utthdpana, at the moment when he seized (or took over)
e jarjara from his assistant, simply by holding it upright in his hand. At
is moment he and the jarjara formed mythologically the cosmic centre
and were worshipped accordingly by a representative of mankind, the
"fourth man". Consequently the following drama could be considered to
be performed, according to the ideal norm of the myth, on the day of
ndra's banner festival, conform to the legendary first performance.
As a result of its inclining, however, the jarjara was desecrated and
could no longer remain on the stage. The fact that the text does not say
word about its removal, although it was necessary to get rid of it, ma
perhaps have been due to a religious taboo. There are also some tex
which forbade the king to see the removal of the indradhvaja after it h
been pulled down. In ase, the circumstance that the assistants we
e them the opportunity to remove the

EXCURSUS on pdcirnend 'pasarpapam

Ghosh translates 5.120 pdrfripdrs'vakayod ca sydt pdcimend 'pas
by "the two Assistants should step backwards." Feistel, p. 72,
different interpretation: "Und [unmittelbar davor] sollen di
Begleiter sich nach Westen zuriickziehen", to which he adds the follo
explanation: "Auf der Biihne also in Richtung auf den Zuschaue
bzw. die ihm zugewandte Biihnenrampe." Since he further rightly pa
phrases apasarpanarn by "sich in ehrerbietiger Weise von jemandern
ehen, und zwar riickwiirtsgehend", he assumes (p. 79)
assistants recede to the front of the stage and there remain
with their backs turned to the audience.
The text itself gives no indication as to which interpretation is co

111 Cf. 27.8 iilpakytam, 35.25 KM. chedavibh&itavadana ("the face m

stripes"?), Abhinavagupta Ia p. 219 line 8 vid@akave,sabhd,scicdra, comm
.l05 vidri$akave,sadhiiri in general G . K. Bhat, The
d, 1959), p. 48ff.and SB din 2169 f.
as paicimena is used in the Ndtyaidstra both in the sense "backwards'-
and So the question arises in which direction the Indian
playhouse was oriented. Feistel implicitly assumes that the stage was
orientated to~vardsthe East, Konow, p. 4, also without further comment,
assumed an orientation towards the South and Subba Rao towards the
West.l13 I have not found an explicit discussion of this point in the modern
studies of the Indian theatre that were accessible to me.l14
The key to the problem is the arrangement of the orchestra. A general
rule is given in NS. 14.11 "What direction the drum and the entry to the
dressing-room are facing, that is in the dramatical performance to be
considered the East".l16 A more detailed description'is given in the very
corrupt text of 34.297+ (p. 640 line 10) KM2, which is complemented and
corrected by 33.221 4- (p. 172 line 13) Cl and by Abhinavagupta's quota-
tions ad 5.17 kutapasya tu vinycisatz (ed. Baroda 12, p. 212 line 1). The
orchestra was facing three different directions, cf. Abhinavagupta (p. 214
8) rangapiihasya tisrp dikqu kutapavinydsdt. It formed a square that
open towards the audience. The drummers (avanaddhakutapa) were
sitting between the two doors that gave entry to the dressing-room,l16
facing the audience and the East. To the South of the open square the
singer was sitting with the vegu- and the vamia-players. They were facing
North where, opposite them, the female singers were sitting.l17
refore, even if one would translate paicimena by "westwards", this
Cf.on the one hand 2.35 paicimepunar bhcige (paicime ca vibhcige B) "the rearmost
division" Subba Rao, Baroda ed. 12,p. 434 (cf. Ghosh), which is equivalent to 2.34
pr$fhatab (Mankad's interpretation, Ancient Indian Theatre, p. 31 is excluded). Cf.
paScimena "behind" (Lii!yiiyana) and Mhbh. IV.61.4 paicimatah "from behind". On
the other hand the word is in current use in the Ncifyaicistra as a designation of the
West, cf. 2.40 paicimena, 2.74 paicime, 2.48 paicimottare.
See D. Subba Rao's reconstruction of the plan of the theatre in the Baroda edition
14 (G.O.S. X m V I , 1950, Appendix, pp. 434,439. The same is implied in Clifford R.
Jones's recent paper in JAOS, 93 (1973), p. 295.
Cf., e.g., V . Raghavan, "Theatre Architecture in Ancient India" in The Theatre
of the Hindus (1955), p. 156ff.(reprinted from Triveni 1931, pp. 69-77), D. R. Mankad,
Ancient Indian Theatre (1950).
l" yatomukham bhaved bhcigbm dviiram nepathyakasya ca, d mantavyci tu dik
pav6 niiryayogena nityaiah.
11* Cf. 14.2 (13.2 KM) ye nepathyagrhadviire mayiip~irvamprakirtite,tayor bM&sya
vinyiiso madlye kciryab prayoktybhih. This does not refer to the whole orchestra
(kutapa) but only to the drums. In its general sense (Siigaranandin 2177) bh&fa
sthcipite bhcindavinycise triscimni parikirtite, ciircivagakalfr
P
itaprayogakam, W ere Ghosh indeed translates "instruments
of the rnrabiga in the modern theatre in Kerala is still the same,

give the text as I think it should be corrected, without

further comment: tatro 'pavi~feprcinmukhe rcuige kutapnvinyEsd kartavyd. tatra
would involve "into the rear part of the stage". That, however, the
implication is "to step backwards" (as Ghosh translates) is probable, also
on account of the variant reading in 5.127 C fair eva ca padaih kdryafi
prEkrnukhend 'pasarpaqarn (for paicirnend 'pasarpaqam) "he should with
his face towards the front (that is, the audience) withdraw with the steps
described before". Abhinavagupta (I2,p. 239 line 5) ex
folIows: prdkmukhEv eva pr~thanivyttibhihpadaih p
petcim ity anena kevalast2tradhdraprayojyarn&di cdrirn
that the assistants should withdr

e, vol.
Thi
spag
eint
ent
ional
lyl
eftbl
ank
 INDE

Abaev, Vasilij Ivanovich: 7%. 1 19, 5911.19 and n.20, 60n.23.

15511.13. amyta: 6 , 33, 37, 38, 39, 126, 127. ampa
Abhidharmakoia: o f Vasubandhu: bowl (jar) 28, 37, 38, 143. Amyta-
ll5n.51. Soma: 29, 106.
Abhinavagupta: 230n. l , 232-257 ~ m & 48.:
passim. Andreas, Friedrich Carl: 5711.6, 58n.14,
Aditi: 48, 54. Presides over the 60n.23, 6211.31, 63n.32, 651.39.
nadir: 49, 50. A. and the Earth: Angkor Vat: 33, 49n.51.
49n.53. Associated with V i ~ g u50,
: Anthill: 97.
50n.56. World Mother: 100. Gave Anumati: 50, 54.
birth to "mortal egg': 100. Apasramba $aura Siitra [gp$.]:52n.73,
Adityas: 9, 31, 69,70. 71n.83, 101n.23. 142n.3, 210,
Agni: 26, 28, 3811.36.and n.40,40,44,47, 212n.88, 213n.90, 250n.91.
65n.40, 85, 148, 161, 164, 185. Arjuna: 33, 3311.18,34.
Agni in the nether world: 71, 83. Aruna: 32, 34.
A. leaves the Asuras: 71. &U: Ritualfight with a h d r a : 7 ,
A. born in the stone house: 68,71,225. 155. Day-Night associated with
A. born every day anew: 71-72,83-84. ~rya-&dra: 104. (See also: Aryans).
A represents the Day: 3311.15,49. A. Aryan(s): 27, 77. A. religious
becomes Varuga in the evening: 67.7 1 . tradition/thought: 28, 35, 89. A.
A. hidden in the waters: 80. A. opens mythology: 79. A. mysticism: 84,
rhe stream: 147n.4. Victory over Death: 86ff. A. concepr of inverred
167. Source of inspiration in Man: mountain: 78, 80. A. represents Indra
183-l84.Hymns to A.: 156, 176. and the Gods: 155-156, 182. A.
Agni-Soma: 26, 28-30, 39, 160, 185. verbal conrests: 184, 214 and 151-215
Agni-Soma in VflraS womb: 218. passim. A. battles against Dasyus:
Agni-Siirya: 153, 176. 203-204.
Agnipiir&a: 23811.45. Arysvarta: 33.
Agni~toma:157. Axa: 63, 87, 89. Bliss of A.:56-89 passim.
Agrawala, V.S.: 49n.53, 5011.59. Friend and counsellor of Ahura
Ahura MazdE: 65-66 79, 80, 86, 87, 89, Mazdz 65-66. Manifest in light: 80,
184-185, 188-189. Name possibly a 86, 87. Possessing the Sun: 8 1.
taboo-subtitute: 6911.68. Name later Accompanied by the Sun: 88.
denotes the Sun: 8 l n. 129. Interpretations of the name AS;:
Aitareya-Arapyaka [Ait. Ar.] [A.A.]: 63n.32.
200, 203. asat: 18-20, 2011.22.
Aitareya Briihmaqa [Ait. Br.] [AB]:44, Asuras: 10, 13--15, 19, 20n.22, 48, 52,
48, 50n.54, 53, 53n.84, 6711.52, 98. Gods of the prin~ordialworld 13,
8011.125, 82, 83, 138n.I. 158-159, 37. Inhabit (signiSyl the nether world:
20111.78, 203, 219, 226. 14, 16, 17, 54, 104. A. who become
Aitare.ya Upani~ad[Ait. Up.]: 5311.83, Devcrs: 14. Varuqa an
99, 99n. 14. Agni leaves the A.: 71. Fight between
Alsdorf, Ludwig: 6 , 138. (See also: Devas and A. : 1
Liiders, Heinrich).
Altheim, Franz: (Thisme apud
 Index
nedict, Ruth F.: 131-132, 13211.85,
134n.91. ,
nveniste, Emile: 56n. 2, 57n.3, 6311.23,
cow from A.:49. A. kept the 78n.114, 7911.119, 101n.24, 105n.35,
121n.61 and n.62, 180-181, 189.
Beowulf: 121, 192, 201.
Bergaigne, Abel: 4211.4, 5011.59, 6611.48,
751.103, 9411.6, 9611.10, 121n.61,
& m t a s ~ ~ r[A&.
a S.]: 50n.54 150n.l, 222
Bhaga: 48
~hagavadg?ig33n. 16, 36, 76, 145.
BhZgavata Puriina: 20n.22.
Bhandarkar, D.R.:4111. l.
Bhiiratlja Ni$ya&stra [Bhir. ~ d ] :
21n.23, 145, 230, 230n.1, 23111.7, 232-
259 passim.
Bharhut: 23, 38n.40.
Bhartlhari: 201.
BhZsa: 239n.47.
Bhat, G.K.: 255n.111.
~hyma:34.
71n.83, 81n.126, 82n.131, 83n.136, Bhygus: 216, 227n.8.
bhyhgiira: 239-240. Emblem of nether
world: 244, 246, 246n.77.
bhytigSlradhZra: 239-240. Impersonation
of Varuqa: 244. Acts as vidiisaka:
254.
Bianchi, Ugo: 18511.52.
101, 154, 167, 183, 185, 189, 191, Bloch, Theodor: 41 n. l.
193. Aveslan: 177-198 passim. Bloomfield, Maurice: 10111.25, 21 1-213.
Boar: see Cosmic Boar.
Bosch, F.D.K.: 5, 23-40 passim, 76n. 106,
Bahram Fire : 213-214. 12811.76 and 11.79, 129n.81, 139n.2,
14511.3, 14611.4.
Brahma: Born from the golden egg:
tt, L.D.: 42n.10. 100. Resides in tfie jarjara: 240-242.
Salutations to B. by siitradhiira: 243.
Siitradhrira 's identify with: 244-246.
rtholomae, Christian: 50n.10 and n. l l, had-haqyaka Upanisad E~yh.Ar.Up.1
57, 5811.13, 5911.16, 6011.22 and n.23, [BAu]: 99, 99n.17, 19811.73.
BrhaddevarZ 41n. l , 5411.88.
Byhaspati: 48, 53-54, 7111.82, 140.
Blhat Samhit; of VarTthamihira [Var.BS.]:
23811.45, 25011.91 and 11.93.
Brown, W. Norman: 6,50,50n.63,51n.67,
106-107, 12111.62, 152n.6.
Buddha: 26.
Buddhacarita: 109n.45.
unack, Johannes F. and Theodor: Buddruss, Georg: 68.
87n.159, 8811.162, 184n.51, 191. Burrow, Thomas: 212n.89.
nda, E.: 75n.103. 78. Buschardt, Leo: 711.10, 28n.4,97n.13,
 106n.39. dak~inii-cows:73-74, 174. 2 12. As stake
utterworth, E.A.S.: 139, 14011.1, 1 in ritual gambling contests: 174,
212n.88. The sabhii a 'cow-shed'
(val6) 200.
Dandekar, R.N.: 41n.l.
d. Willem: 33, 7511.103, 153n.7, DZnu: 121- 122 and 121n.61, 144. ~2inava/:
n.70, 210, 213n.90. 121. Asuras-Diinavas: 153.
Darmesteter, James: 52n.76, 189.
Dasyus: Battles o f Aryans against:
mankind: 245-247, 255. 203-204.
CarurvargacintZmaqi [Caturvargac.]: De, Sushi1 Kumar: 23111.4.
23811.45,239n.50, 250n.93, 25 111.95. Debrunner, Albert: 197n.69, 198,20911.84,
Ciurndogya Upani$ad [Chznd. Up.]: 100, 2 1211.89, 2 13, 23911.48. Wackernagel-
140, 140n.2, 169, 198. Debrunner: see Wackernagel.
Chariot Races (ritual): 104, 151, 154-155, dc Buck, Adriaan: 4.
172, 174, 203, 213. de Goeje, C.H.: 128, 129n.80, 132, 13311.89.
Charpentier, Jarl: 4211.10 and 11.18, de Jong, J.W.: 196n.68.
73n.95, 78n.114, 101, 10111.24, 219, de Josselin de Jong, J.P.B.: 3,5,
220n.5, 223. 4811.49, 213.
Ghristensen, Arthur: 63n.32. de la Vallie Poussin, Louis: 115n.51.
Churning of the Cosmic Ocean: See: Delbrueck, Berthold: 19811.70 and 11.71
Ocean, Churning o f the de Mallmann, ~ a r i e - ~ h k ; e s e79n.
: 12.
Coomaraswamy, A.K.: 8,7611.106, 145n.3. d' Onofrio, C.: 153n.7.
Cosmic Boar: 33, 39, 53, 100-101, 108, Delphi: 183.
185- 186. See also Emusa. DEnkart:62.
Deussen, Paul: 41n. I , 4211.l l , 7711.112,
Cosmic Mountain: 55, 68, 76, 78, 80, 85, 100n.20.
106, 107-109. Power of resistance
Devas: 5, 14, 52, 98, 104,,153. Asuras
(vytra) in: 1 16. Dragon as guardian
who become D.: 14. Varupa becomes
o f 116. Cosmic m . equivalent t o the
a D.: 15- 17. Brhaspati stands for t
whole earth: 80. Inverted mountain
D.: 54. D. sons of Prajiipari: 198.
in the night sky: 78, 80. See also:
War between D. and Asuras: 1111.6,
Primordial Mountain.
14-15, 17- 18, 22-23, 74, 199. The D.
Cosmic Pillar (skambhb, stambha): at Churning of the Ocean: 49n.5 I,
12,23,49,54,68, 145,241. 99. The D. longed,for Soma: 225.
Primordial tree as central world Soma broughr for: 226. The D. win
pillar: 124. V i y u as cosmic p. arising the cow from rhe Asuras: 49. D.
from Mount Mandara: 55. associated with Day (Asuras with
Cosmic Tree. See Primordial Tree. Night): 52, 75, 95, 104. Vip.zu
Tree o f Life. rranscends dualism of D. and As
aters: 16, 80, 82, 98, 138-139. 20-21, 33, 48, 54.
rpent in: 21. Agni born as child ,oJ ~ e v & u i i p a : 238n.45.
29. V i ~ n sleeps
u in : 21, 33, 49. Rth Dhanika: 233, 233n.23.
n in: 96. Cosmic waters Dhartr: 48.
'possessing tht sun ': 80-8 1 . Nocturnal Dhytar7istra: 32.
ocean: 37, 74-75, 76, Divyiivadc7na [Divy]: 193, 195.
Duchesne-Guillemin, Jacques: 5611.2,57,
63, 6511.39,81n.129, 198.
Dume'zil, Georges: 32n. 12, 34, 3411.21,
5211.76, 207.
1M-107, 116, 152-153. Dumont. P.-E.: 252n.102.
 Index
Durckheirn, 6mile: 3n.3 8011.124, 8111.129, 87n. 158, 88n. 162.
DvarakI: 73. Geldner, Karl Friedrich: 36n.29,
37, 44, 45, 4511.49, 46, 46n.39, 47n.41,
51, 5311.84 and 11.85, 5711.9, 6011.23,
Eiida: 49n.52. 63n.32, 6411.37, 69n.66, 70, 71, 71n.80
Edgerton, Franklin: 193-194. and n.81, 7711.111, 80n.122, 84n.144,
Eggeling, Julius: 4 l n. l , 5011.54, 183n.2, 96n.10, 99n.18, 127n.74, 14211.2,
199-20, 23 1. 14411.2, 14511.2, and n.3, 147n.3, 148,
Ehrenreich, Paul: 12911.80. 149, 151-152, 154-176, 177-179, 182,
Eilers, Wilhelrn: 5611.4. 184, 20511.8 1, 206-208 passim, 22411.7,
Eliade, Mircea: 8-9, 7!, 90-92, 101n.24, 228.
107n.43, 111, I16n.54, 125n.67, Gershevitch, Ilya: 18 1, 189.
126n.69, n.70, n.72 and 11.73, Ghose, J.C.: 4111. l.
13311.88, 134n.90 and n.91, 136, Ghosh, Manomohan: 230n. l, 23 1-256
136n.92 and 11.93, 142n.2. passim.
Emeneau, M.B.: 36-37, 76n. 106,77n. 1 11. Gonda, J.: 2611.1, 4111.1, 46,
Emusa (Erniisa): 101, 101n.24. See also: 50n.55 and n.62, 5411.87, 101n.24,
Cosmic Boar. 131x84, 173, 235n.3 1, 23811.45.
Gopatha ~ r s h m a p a[GB]: 13911. l ,
143n.2, 201, 210.
Faddegon, B.: 43n. 12. Gotarnas: 147.
Feistel, H.O.: 230n.3, 23311.19, 234n.26, Gray, L.H.: 104n.26.
and n.27, 235-237, 235n.32, 11.33, Griswold, H.D.: 41n.1, 46,
n.34 and n.35, 24311.71, 244n.72, 245, 8111.126, 223.
249n.81, 250n.92, 253, 254n.110, 255, Grpmbech, Vilhelrn: 209, 21 1.
256. Giintert, H.: 42n.7
Fig Tree (sacred, celestial) : see gylpcwide (gilp): 192, 209, 2 1 1.
akartha.
Fodor, Nandor: 112n.47,
117n.55. 118n.58. Hamilton Jenkins, A.K.: 14211.64.
Fravardin YaSr: 190. Haoma: 65, 128, 185, 218.
FravaHis: 181, 184, 190. Hardy, E.: 4111.1, 4211.3.
Harivarhs'a: 100n. 2 1.
harmy;: 68, 74, 79, 80, 82
GaBI, J.J.B.: 57, 5711.5, 8311.32, Light of U p s comesfrorn: 160- 161.
64n.38, 6511.39. Agni born from: 225.
Gambling Contests (ritual): 174, 21 1 Harrison, Jane: 100n.21, 10711.43, 132.
212n.88. Hart, G.L.: 237n.79.
Gindharl: 32. Hastinapura: 32.
Gandharvas. 22. Haug, Martin: 62n.29, 63n.32,
Gahgl (Ganges) 32, 371x35. Ganga- 6411.38.
Yamunii: 3511.25. Havell, E.B.: 41n.l. 42n.6,
Garga: 237, 25 1 n.97. 13911.2, 146n.6.
Garuda: 32-34, l 10. Steals the amyta: Heesterman, J.C.: 50, 53, 11611.53,
38. Suparva the prototype oJ 48. 153n.7, 154n.9, 15811.23.
GcithrS ahunavait5 59. He!d,G.J.: 3,4,42n.19,45n.31,
Githss: 59-60, 62-64, 6Sn.41,88, 180, 47, 49n.51, 50n.64, 5211.75 and 11.78,
184. 53, 54n.88, 5511.94 and n.95, 73n.95,
Glthic: 57-58, 6On. 23, 62. 139n.2, 143n.3, 146n.6, 156, 172, 214.
GZyatri: 219, 226. em5dri: 23811.45, 23911.50,
ilhelm: 6511.41, 66, 67n.55, 25011.93, 25111.95, 11.96 and n.97.
Index
Mountain: 105, 106, 117, Kiils DLjnu:
121, 122. fills Tvagr: 123. 1. Sorrrce
of inspiration in Man: 182-183. I.
Hertel, Johannes: 39n.46, 78n.114, prototype of the competing hero:
82n.135, 191. 170- 171. 1,'s assistance invoked in
Hillehrandt, Alfred: 7, 2811.5, battle: 204-205. 1.k creative art: 71,
34, 38n.36, 4111.1 and n.2, 4411.26, 45, 104, 106, 110, 118, 185. I.-Vrtra'
451-1.29,4611.38, 4811.48, 50-5 1, 5 ln.65, fight a Creation myth: 5, 50-5 l , 97,
52n.76, 54, 55, 6611.48, 6811.63, 70, 105, 118, 152-153, 185. Rivets Pxes,
7ln.80, 7611.104, 7811.1 14, 85, 85n.143, pegs) the hill (mountain): 53, 107,
95, 95n.9, 9611.10, 10511. 34 and n.36, 109. Props up the sky: 12, 53, 104,
106, 13811.1, 140-141, 14111.2, n.3, 248. Splits the primordial bilk 29,
and n.5, 14411.2, 1456.2, 146n.4, 3511.25, 85,- 118, 142, 182. Breaks the
14711.1, 149, 153n.7, 15411.8, and n.9, force-~Fresistance (VS&): 12, 116,
151n.ll and 11.12, 157-158, 15711.17 152, 185, 217. Releases the waters:
and n. 19, 159n.26, 160n.29, 162-164, 67, 142, 143-144, 147n.4, 175.
222. Liberates Agni and Soma from the
Hinze, O.M.: 12511.66. primordial hill: 29, 30, 72. Engenders
hirapyagarbha: 23, 26, 27, U p s and the Sun: 160. "Winner of
80, 100. the sun ": 154- 155. I.S creative dance:
Hocart, A.M.: 8. 186. I.- Vyana'fight associated with
Hoffmann, Karl: 5611.3, 6711.56, the New Year (winter solstice). 203,
7411.99, 79, 79n. 120, 82, 100n.22, 204, 206-207. I.- Vytra' myth a New
145n.2, 150n.l. 175n.39, 198, 21 1. Year myth: 94, 98, 104-105, 106-107.
Wohenberger, A.: 4111. I. I. and Visnu: 19-20, 45, 48, 50, 53,
55, 143. Role of V i g u in l.-Vytra'
Holmberg, Uno: 14611.2.
fight: 34, 46, 54.. V i ~ n presses
u Soma
Hopkins, E.W.: 41n.I. n.2, and
for 1.: 55. 1. and Varuva: 16-18,49,
n.5, 43n.19, and n.20, 4811.47, 5011.55,
254. I. and Byhaspati: 49. 1. and
and 11.56, 54n.88, 5511.95, 67n.54,
,$uSr?a:69. Soma pressed for I. in
97n. l l, 223.
Vedic prayer and sacrifice: 170, 172.
Horn, Paul: 62n.27.
Indra fesrival: 13, 3911.47, 237, 246.
Hr6lf.s Saga Kraka: 122.
Indra-kila ("lndra's peg'?: 109. I. S
Hrozn?, Bediich: 42n. 14.
banner: see indradhvaja: 1,'s vajra
Hubert, Henri and Mauss, Marcel: 3,
5411.88, 105, 116, 23711. 48.
28n.4, 156n.14.
Hubschmann, Heinrich: 63n.32, 6411.38. indra-dhvaja (Indray banner): 29, 38
Humbach, Helmut: 57, 5711.5 and n.6, 237, 238, 241-242, 244, 245, 250-251,
5811.14, 59, 6011.21 and n.23, 63n.32, 25 1n.9 1. Jarjara identified with indra-
64n.35 and 11.36, 65n.38 and 11.39, 41, dhvaja: 237, 241-242, 255.
71n.79,80n.124,81n.l29,88n.165, 180, Comparison of jarjara ritual with
191, 193. erection and pulling-donn of indra-
dhvaja: 250-251 253. Inclining of the
jarjara a symbolic imitaiion of
Indonesian Religions: See: Ngaju pulling-down of the indra-dhvaja:
Dayaks; Papuans; Toba Bataks. See 252.
also: Jata; Mahatala. Indraprastha: 32.
Indra: 5, 6-7, 11-15, 2811.5, 31, 10111.24,
176, 208, 212, 217, 218, 221, 224, 245.
Fight with Vytra': 12, 34, 50-5 I, 85, Jacobi, Hermanh: 238n. 45, 250n.93,
97, 138. Slaying of Vytra': 50, 5 1, 53, 251n.95.
98, 117, 152-154, 203. Fights the ~aimin$a Briihmana [Jairn.
 l7n. 16, 3011.9, 49, 49n.50, 50n.55, 5 1, ~ ~ a u s h a Upanifad
ki [Kaus Up.]: 1
53, 53n. 82, 11.83, and 11.84, 54, amakrishna, N2yas'cisfra
54n.86, 69, 7411.99, $In. 126, 96, 99, of Bharatamuni (edition 1 2 , 1956):
100, 104, 15411.10, 15811.33, 183, 200, 230n.1, 23211.10, 233n.22, 239n.48,
201n.78, 202, 203, 211, 218, 221. 240n.52, 249n.83 and n.84, 252, 256.
a aim in& Upani;ad BrZhmana [JUpBr.]: awaguchi, Ekai: 215.
100, 203. nsth, Pandit, Bhiirntiya N&ya&isfra
Jakobson, Roman: 3 In. I I. avyam;ilZ, edn. 1943): 23011. l .
jarjara: 26n. l, 145. 232, 234, 235-257. 231n.7 and n.8, 23211.10, 23311.22,
jarjaramok~a:253. 235n.30, 239n.48, 24311.70 and n.7 1,
jarjarapra~~oga: 236-257. 247, 249n.81, n.83, n.84, 11.85 and
Jata : 9 1, 92. 11.89, 250n.90, 25 111.98,252, 256n. 1 16.
Johansson. Karl Frrdinand: 42n. 18. eith, Arthur Berriedale: 4111. I ., 43n.23,
68n.63, 220n.5. 50x59, 5411.87, l59n.19, 162, 176,
Jones. Clif'ford R.: 24711.79, 25611. 113 199, 217, 221, 223, 23011.2.
and n. 1 16. onow, Sten: 4111.2, 23111.9, 23511.31 and
11.34, 239n.47, 249, 254. Sten Konow-
Poul Tuxen: 4111.2.
ba: 109 ktijke Akademie van Wetenschap
andha: 39 sterdam: 4.
Kadrk 32, 198. 225 226. 139, 141, 143, 147.
aegi. Adolph: 14711. I: risch, Stella: 55n.100.
agarow. E. : 14611.2. rappe, Alexandre Haggerty:
d a b : 3811. 37, 246. 42n.7, 5211.76, 99n. 16.
KHlidHsa: 44, 101, 195, 232, 239n.48 Kristensen, W. Brede: 3,4,52n.77,67n.5
KiilikZpurZya: 23811.45. 7411.98, 79n. 131, 13911.5, 14611.3,
Kane, P.V.: 231x4 and n.8, 23911.47. l66n.33, 18011.47, 188- 189, l88n.55.
va: 69. K&hu: 38.
ighala-Kqha Sari~hitrl[KapKS.] Kmna: 33, 54, 5411.88.
[KKS]: 52n.72, 10 113.23, 10211.27 and k~atriya:34.
11.29, 183, 19711.69, 198, 212n.88, 220, Kubera: 245.
226. Kuhn, Adalbert, 4111. l, 216-217, 218-219.
Kas'yapa: 32. Kuhn, Ernst: 76n. 106.
K@haka Sulizhitcl [K?i(h. Samh.] [KS]: Kuiper, F.B.J.: 1-9. The references in
33n.14, 44, 48n.48, 49, 50, 5011.57 and this volume (passim) are to the
following publications (in
chronological order):
De goddeiijke Moeder in de Voor-
Indische religie (Groningen-Batavia
1939);
Review of Leo Buschardt, Vytra,
220, 221, 226, 25011.91. Det rituelle Daemondrab i den
!ha Upani~adEKath.Up.1 [KU]: Vediske Somnkult (Copenhagen
36, 37, 76, 76n.107, 84, 85, 85n.147, 1945), in Museum 52 (1947), pp. 198-
86, 86n.1?8, 145. 200 [see p.61;
-tyZyana Srautastrtrc [K&.]: 208 Review of K.F. Geldner, Der Rig-
Veda aus dem Sanskrit ins Deursche
ICriaasika Siitra [KausS.]: 82, 148, 23811.45, uebersetzl, 3 Baende.(Cambrid
Mass. 1951). ibid. 59 (1954). pp.
u$taki-~riihana~a[KausBr.] [ $1 -8S;
53, 67, 6711.56, 201n. 78. A n Ausiro-Asiatic Myth in the
Index
Rigveda, Mededelingen der Maior, Congratulatory Volume
Koninklijke Nederlandse Akademie presented to J. Gonda (Leiden 19721,
van Wetenschappen, Afd. pp. 144-156 [re-published in this
Letterkunde 1317 (1950), pp. 163-182; Volume, pp. 138- 1501;
N6ropi Khalkiji, ibid. 1415 (195 l), 'The worship of the jarjara on the
pp. 201-227; stage', IIJ. 16 (1975), pp. 241-268 [re-
'Naar aanleiding van de Gouden published in this Volume, pp. 230-
Kiem', Bijdragen tor de Taal-, Land- 2571;
en Volkenkunde I07 (1951), pp. 67- Varuna and VidG;aka, Verhande-
85: [republished in English lingen der Kon. Ned. Akad. van
translation in this Volume, pp Wetenschappen, Afd. Letterkunde, dl
23-40]; 100 (Amsterdam, etc. 1979).
Review of Paul Thieme, Mirra and kumbha: 38n.38, 246. See also: 'kala6a'
Aryaman (New Haven, 1957) in: kumbhiicda: apparent foreign origin of
Indo-Iranian Journal 3 ( 1959). pp. the word: 3911.43.
207-2 12; kunpalin;: 125, 126, 128.
Review of Ugo Bianchi, ZahxZn i Kunike, H.: 42n.3
ahrmazd (Torino, 1958), ibid., pp. Kuruksetra: 32, 35.
212-216; Kurylowicz, Jerzy: 184n.51.
Revpw of J. Duchesne-Guillemin, Kuvalayam5li: 145.
The ?Western Response to Zoroaster
(Oxford, 1958), in IIJ. 4 (1960), pp.
182-189; Lang, Andrew: 43.
'The Ancient Aryan Verbal Lassen, Christian: 41n. l , 4211.15 and n. 16.
Contest', ibid., pp. 217-281 [re- Law, B. Ch.: 73n.95.
published in this-volume, pp. 151- Leiden, Univefsity of: 1, 3, 4, 5.
2 151: Lethaby, W.R.: 7
'Remarks on i'%e A ~ e s t a nHymn Lentz, Wolfgang: 61n.23, 86n. 150,
to Mithra', IIJ. 5 ((1961),p. 36-60; 8811.163.
'The three Strides of Viszp', Uvi, Sylvain: IOn.2, 2811.4, 52n.74, 95,
Indological Studies in Honor of W. 139%1, .23111.5, 232-233, 236.
Norman Brown (New Haven, Conn. Lhasa: 214-215.
1962), pp. 137- 151 [re-published in Lichterbeck, Karl: 58n. 10.
this Volume, pp. 41-55]; Lienhard, Siegfried: 236-237.
'The Bliss of A h ' , IIJ. 8 (1964). Lietaert Peerbolte, M: 1 1311.49, 117,
pp. 96-120 [re-published in this 117n.58, 12411.65,131n.83.
Volume, pp. 56-89]; Lindner, Bruno: 212n. 89.
'On Yasna 30.7~: Dr. J.M.Unvala Locher, G.W. : 3, 4, 204.
Me~morialVolume (Bombay, 1964). Loki: 223.
pp. 80-88; ~okoddh7m-i-&ha: 33.
'Cosmogony and Conception: A Lommel, Nerman: 59, 59n. 1, n. 15 and n.
Query', History of Religions 10 20, 60, 60n. 23, 6111. 25, 62n. 30,
(1970). pp. 9 1 - 138 [re-published in 75, 86, 93-94, 99n. 19, 127n. 75,
this Volume, pp. 90-1371; 138, 15011. 1, 154n. 7, 166, 180n. 47,
'The Origin of the Sanskrit Drama' 185, 189, 190.
(summary), Proceedings of the Lueders, Heinrich: 6, 66n. 51, 6711. 54
XXVII International Congress of and n. 55, 68n. 57, 7011. 79, 7111.
Orientalists, Ann Arbor, Michigan, 80, 74n. 98, and n. 99. 75, 79. 80n.
13th-19th August 1967 (Wiesbaden, 122 and n. 125, 83n. 136, 8711. 157
1971), pp. 298-299; and n. 159, 96, 97n.
105-107, 138, 139n. 1 , 1
 Index
141, 143-144, 147n.2, 148, 149, Mauss, Marcel: 3, 28x1. 4, 156n. 14,
150n. 1, 152, 15211. 5, and n. 6, "1'72, 202.
17511. 39; 197. Mayrhofer, Manfred: 155n. 13.
Ludwig, Alfred: 45n. 29, 47n. 42, 106, MazdZ: see Ahura MazdZ.
10611. 40, 141n. 2, 14311. 3, 145n. 2, McCrindle. John Watson: 73n. 95.
148, 157n. 17, 196, 198, 202, 227. Meillet, Antoine: 59n. 15, 60n. 23,
61, 61n. 26. 62n. 29.
Macdonell, Arthur A.: 41x14, 43, 43x1. 23 Meru, Mount: 26, 68, 109. Niigaloka
and n. 24,, 45n. 30, 4611. 33, 49n. 53, under the centre OS: 145. See also:
50n.60.52n. 7 9 . 5 9 ~1 7 . 6 6 ~60.92.93, Mandara, Mount.
9411. 4, 107, 12111. 61, 14311. 2, 150n.'1, merudanda: 126.
153n. 7, 160n. 29, 184, 222-223, 228. Meyer, Johann Jakob: 13n. 8 , 4 l n . l , 6 6 n .
magh;: 155- 156, 172. 48, 6811. 63, 69, 75, 153n. 7, 23711. 44,
Mahlbhiirata [Mhbh.]: 1311. 8, 17.20n.22, 238n. 45, 24111. 59, 250n. 93.
21n. 23.30-35,38n.37,44.4811.47,4911.51, Mills, L.H. : 6111. 24.
66n. 48, 96, 99, 100n. 21, 121, 143, 144, Minard, Armand: 100n. 21, 121x1. 61.
148, 197,208,219,221,238n. 45,23911. Mithra: 29, 70, 188-189. Indra-features
48 and n. 50, 24 1 n. 58, 246, 25 1 n. 94 in: 31.
and n. 95. 256n. l 12. Mitra: 31. 48, 6511. 40, 95. Associared
Mahiiniiriiyana Upanijad [MahZnZr. with rhe rising sun: 3 1, 45, 80. Mirra
Up.]: 84, 85, 8611. 148. and Varupa a dichotomy: 3 1. M. and
Mahatala: 91, 92, 103. V. lords of Rta: 66. Hold both earth
&lahijvariisa [Mahav.]: 196. and heaven: 68. M. dwells in.the 'stone
Mah5vrata-ceremony: 155. house: 69, 74, 80.
MahTdhara: 210. Moon: 74. 126-128, 189. Soma equated
rnakara: 23, 28. 28n. 5. wifh the M. : 128. Ganges rises in the
Mairr+at~i Sariihitii [Maitr. Samh.]: M. :,37n. 35. Gangesflowsfrom the M.
MS] 48.48n. 48.50n. 58,52n. 72,5311.83, on Siva S forehead: 32.
66n. 48, 6711. 53. 71n. 83, 75, 99x1. 15, Moor. Edward: 4111. 1.
101, I0ln.-23, 102n. 26, 108n. 44, 116, Morgenstierne. Georg: 209.
12811.78, 13111. 1, 142n.2and n. 3; 183, Moulton. Jarnes Hope: 63n. 32, 81n. 128.
185, 197-198. 19811. 73,208n.83,212n. Mountain: See Cosmic Mountain.
88, 218, 220, 221, 250n. 91. Mrcchaka(ikii [M ycch.]: 2 10.
~ a i r r Upani~ad
r [Maitr. Up.], [MU]:'46, Mueller. Max: 4111. 1. 50n. 59, 7411. 98,
77, 77n. 109, 86, 145.. 148-149.
Mlmallapurarn: 55. M u ~ d a k aUpani;ad[Mupd. Up.]: 86,8611.
MZnavo Gyhyo-Siitra [MiinCS]: 52. 148.
Mandara, Mount 35, 55. See also Mus, Paul: 8. 4211. 9 and n. 13.
Meru. Mount.
ME$kkavIcakar: 114-1 15. Naciketas: 84. NZcikera,fire: 84.
Mankad, D. R. : 233n. 19, 25611. 1 12. Nr?ga(s): 24 1. N.s ar ,/,or of Cosmic Tree:
Manu: I O l n. 21, 198. Agni brought for: 24111. 56.
225. Sorna brought for 226. Nggaloka: 145.
MErkapdeya: 2111. 23. n6kru: 159.
Maruts: 37n. 35, 4111. 1 , 47. 147-148, Nakula: 34.
149, 224. Naigha!7!uka: 67.
Marshall. Sir John: 7711. 110. Nairy6. sanha: 188. 189.
Maspe'ro: 125n. 68. narnii, narman: 208-1 13,
MZtali: 144. nark!;: 208, 2 1 1-2 13.
MBtari;ann: 2! 6-229 passim. NBrada: 144.
dl~sisyaPcrrBna: i43n. 3. Sarrcn, Johiinna: 217n. 3. 223-224, 227n.
 8, 228. 51.
~ii!akulak~anaratnakoia: 23 In. 8 See also ~ a r a & i m a : 73n. 95.
Sagaranandin. ~iiri~iirs'vika: 239-277 passim.
Nii!ya Darpana: 233-234. 234n. 25. Pa janya: 148.
~Z!~a&istra:See Bhijrafija ~d!ya&stra. PataZjali: 198.
Navel of the Earth: 7. p'olZ-plant: 185.
Needham, Rodney: 3n. 3. Periplus: 73n. 95.
Neisser, Walter: 30n.8, 4111. 2, 42n. 7, Persson, Per: 209.
66n. 48, 71x1. 80, 82n. 135, 15011. 1, Petech, Luigi: 215.
162n. 31, 205n. 81. Pokorny, Julius: 209.
New Year (Ceremonies): 2, 4n. 4, 13, Pongal Festival (South India): 153.
153n. 7. Vedic Hymns for: 18, 94, 104- Potlatch: 18, 156, 170,200. In Indo-Aryan
105, 167. The 'First Creation'and: 97- civilization: 156, 172,207,2 14. Usasas
98. N. Y. conceived as a new cosmic goddess oJ 170-175 passim. In North
start: 153. Renewal of thejire at: 213. America: 174,2 13. In Nias, Indonesia:
214. In Lhasa: 214,215. See also: lndra 1%. 56.
(lndra- v I t d myth). Prajiipati: 2911. 6,45,85. Creates the Gods
Ngaju Dayaks (religion of): 91, 103, 136, and Demons: 33, 108, 128. Represents
186-188, 201. totality: 45, 48. His natural place
~ i b h h o ~49n.
~ r :52. between (above) upper and nether
Nikiila-tree: 195. worlds: 49. The number 'seventeen'
Nirrti: 50, 225. associated with: 53, 54n. 86. PrajZpati
nrtE 209. and the Creation: 99-100, 101, 102.
nyagrodha (Banian tree): 36, 145-146. Mountains the jirst offspring oJ 108.
NyZsa Upani~ad:8411. 141. subhi and shmiti as daughters of: 198,
Nyberg, H.S. :57n. 5, 78x1. 117, 181, 191. 211.
pratighii: 12, 98, 102, 138.
Ocean, Churning of the: 49, 4911. 51, 99, Prenatal extrasensory perception:
153. See also: Primeval Waters, 112-1 15, 117, 120-121, 124-125, 130-
Cosmic Waters. 131.
Oertel, Hanns: 197, 198n. 70, 20111. 78. Primeval Twins: 129.
21011. 87. Primeval (Primordial) Waters: 11-12,
Oldenberg, Hermann: 4, 34, 4111. 2, 42n. 14, 16, 19,55,67,98-100, 100-102, 142.
10, n. 15 and n. 17,44,44n. 28.45n.30 Emergence of Earth from: 39,53, 100-
andn.31,46n.35andn.39,51,51n.66, 101. Primordial Hill arises from: 66.
n. 69 and n. 71, 54, 55n. 99, 5 7 , 5 7 1 ~8, Varuna dwells in: 16,37. V a w a ruler
59n. 16.71, 71n.80,72n. 83.92n. 3,93, (god) 08 16. 66, 75, 97, 183. Birth of
95, 107, 141n.2and n. 3, 146n.4,157n. Agni in: 26, 29, 71-72, 224. Sun sinks
20 and n. 21, 164, 16%. 35, 18011. 48, into Primeval Waters in evening: 3 1.96
20111. 77b, 222, 227, 22711. 8. (see also: Sun, nightly journey of).
Oppert, Gustav: 4111. 1. Night Sky equated to: 37 (see also:
Otto, Rudolph: 3,36n.29,42n. 13,4311.21, Cosmic Waters, nocturnal sky as
5511. 99. 145n. 3. heavenly ocean). Golden Egg arises in:
99,103. Hiranyagarbha born in: 23,26.
PaippalCida: 50. Arabic source on: 109. See also:
PaiTcatantra: 128 'Cosmic Waters'and'Ocean, Churning
Paiicavimb Brijhmana [PB]: 51n. 70, 0 0
14411. 3, 198n. 73. Primordial Hill: 11, 29, 33, 53, 55,
Piindavas: 32, 33, 34, 35, 72. 66-67, 68, 78, 106-109, 138, 140, 160.
Piinini: 208. Brought upfront the Waters by Cosmic
Papuas (of Waropen, New Guinea): 49n. Boar: 33, 39, 53, 100-101. Tree of Life
 Index
arising from: 39, 143. Power of 157n. 21 and. n. 22, 158n. 24 and n. 25,
ronce ( v p d ) within: 11-12, 152. 159n. 27, 16011.30, 162n. 3 1, 163, 16311.
imprisoned in: 141 . Indra splits the 2, 164- 165, l67n. 34, 173, 17311.36, 174,
Hill: 29, 109, 142, 152-153, 160. U ~ S 197n. 69, 208, 2 1 1, 2 12n. 89, 224-225,
born ofl 159. Opening of P. Hill 227n. 8, 228. Renou-Silburn: 67n. 54,
equated with opening of the mind: 85. 74n. 98.
Symbolic representation of P. Hill at Rgveda [RV] or Rgvedal Rk-SamhitB
the sacrificial place: 108. See also [RS]:2,3,4,5,10,Il,Iln.4andn.5,
'Primordial M o u n t a i n , 'Cosmic 14, 16n. 12.17.18; 18n. 17andn. 18.19,
ountain'. i9n. l9 and n. 20,20,26n. l, 27,28,29,
mordial Mound: 4, 6. 30, 35, 36, 37, 38n. 40, 39, 39n. 43,43-
mordial Mountain: 67, 76, 80. Covers 47. 50. 5 1. 5 In. 69.52.53n. 82,54,54n.
and encloses the primeval waters: 67. 90.57-58.66n. 48.67n. 55 and n. 56.68,
Power of resistance (vrtrd)m: 116. h g o n 6811. 61 and n. 62, 69. 69n. 64, 70, 71,
as guardian ofl 116. Zndra's jighr 7 In. 8 l, and n. 82,75n. 100 and n. 102,
directed against: 12, 106. See also 7611. 104 and n. 105, 77. 79n. 121. 80,
'Primordial Hill', 'Cosmic Mountain. 80n. 122and n. 123,8l,8ln. 126,n. l27
rimordial Tree (World Tree, and n. 130.82, 82n. 131, n. 132, n. 133
Cosmic Tree): 26, 29, 35, 36-37. and n. 134, 83, 84, 84n. 138, 85, 85n.
Inverted tree: 76, 77, 78, 145-146. 144, n. 146and n. 147,87,88,88~.160,
Kekajon (World tree of Java): 128. 92-95. 96, 97, 97-98,99, 100, 103n. 30,
Jarjara represents World Tree: 241 - 104-105, I04n. 32 and n. 33, 105n. 37,
242, 245. See also: 'Tree of Life'. 106-107, 108, 115, 116, 118, 121, 123,
Prometheus: 2 16-229 bassiin). 126-127, 128, 131, 132.139-141,139n.3
Przyluski, Jean: 42n. 13. and n.4, 14011. l,n.2,n.4andn.7,142-
Ptolemy: 73n. 95. 144, 143n. 5, 145, 145n. 1 and n. 2, 146,
firamdhi: 161, 199, 201. ,147n. l.n.2and n.4, 148-150, 151, 152,
P u ~ u s a 86
: 153, 154-157, 15411. 10, 158, 159-161,
pirrvarairga: 145, 230-236, 253, 254. 161-162, 162-164, 164-165. 165-167,
168-169, 170-175, 175-176, 179, 182-
183, 184, 186. 192-193. 196, 197, 199,
Radin, Paul: 13411. 91 202-210, 211, 212, 213, 214, 216, 218,
Raghavan, V. : 233n. 19, 256n. 114. 219, 224-225, 226-229, 238n. 45.
Raghuvam6a [Raghuv.]: 195 MZdhava :S commentar.r on RS: 199.
&mZya~a 195,238,238n.45,25 In. 94and See also: Celdner, Keith, Ludwig,
n. 95. Macdonell. Oldenberg, Renou. Roth
rangadviira: 249, 254. (2 16-2 19 and 227).
Rank, Otto: 112n. 48 Rome: 183.
Rao, D. Subba: 256,256n. 112, and n. 1 13. Ronnow. Kasten: 30n. 7. 42n. 5, 54-55,
Raznu: 86. 55n. 92.
Rau, Wilhelm: 74n. 96, 198, 199, 200. Roth, Rudolph: 6111. 25i67n. 31. 196. 202,
rebh;, 221. 213, 216-217. 218-219, 227.
Rebhila: 210 Rra': 66. 67, 68. 71. 72. 78, 79-81. 87,
Regnaud, Paul: 85. 95-96. R. hidden in the nether \vorld:
Reichelt, Hans: 57, 62b. 29, 78n. 114.79n. 80. 96-97. 183. 224. V a r u ~ assigned
a as
117. guardian qf: 153. R. resides in realm of
Renou, Louis 15n. I I, 44n. 27, 66% 51, the dead: 89. Bliss of' R. : 193. U p s
67n. 56, 75n. 102, 77n. 1 1 1, 7911. 131, horn from: 79, 160.
84% 138,85n. 144.88n. 166,94,94n. 6, Ruben, W.: 42n. 13.
95, 96n. 10,97, 12111. 61, 145n. 2, 146, Rudolph, K . : 86n. 150.
149, 151, 151x1. I, 152n. 4, 155n. 12, Rudra: 3811. 42, 245.
Index
sabh;. 72, 200, 212. The S. a replica of and Earth: 68, 78, 145. Resides in rhe
nether world: 73. A sacred initiation Jarjara: 24 1.
hall: 73. A sacred place: 198-199. Shamasastry, R. 127n. 74.
Sacred cows in: 200. Social contests in: Shastri, Haraprasad: 249.
74. Verbal contesp in: 174, 208, 2 1 1- Shende, N. J.: 88n. 166.
2 12,2 14. Gambling in: 174.2 1 1. S. and Sieg, Emil: 75n. 103, 83.
iti the two daughters of PrajZpati: ?ircar, D.C.: 73n. 95.
Siva: 32, 43n. 19: 55x1. 96. ~ u d r a - & a
spb2gya: 199-202, 214. 38n. 42, 245.Sivd-Maheh~ara:25 1-252.
Saci: 122. skambhci, stambha: See 'Cosmic Pillar'
: 239n. 48, 255n. 1 1 1, 256n. Skanda: 241.
116. See also: NC[akala~apzratnakosh. Skandasvzmin: 4111. I
Soma: 6, 2011. 22, 28. 30, 39, 47, 48, 141,
rpapa: 23111. 4, n. 5 and n. 6, 160, l85,2 17-229passint. Soma-amyta:
236n. 40. 28. Soma bowl (kalas'a): 55. S. in the
$&apiiai: 4111. 1. Cosmic Waters: 143, 175. Apah-~orna
(primordial water): 153. S. represents
Ssmaveda: 103n. 30 the Night: 3311. 15,49. S. equated with
s k i i i : 2 1 1. the Moon: 128. Agni-Soma: 26,2
samudrd 74, 148. 39. 160. Agni-Soma in yrtra's WO-mb:
SjHnchi: 23, 3811. 40. 2 18. Soma- Guardian: 38. Soma-
Sahkara: 241. yielding aivattha: 143, 144. V i z u
bikhiiyana <rapyaka [Ss?kh. A?]: 200 represents Soma: 42. V i ~ presses
w S
hikhiiy,ana Srautasiitra [SLlikh. S.]: 208 for Indra: 55. S. pressed for Indra in
Sarasvati: 196. Vedic prayer and sacrifice: 170, 172.
Sarkar, Benoy Kumar: 42n. 4. Sopara: 73n. 95.
S$ 18, 19, 20. See also: isat. Speyer, J.S. : 75n. 103.
Satapatha Brzhmana [ ~ B I31,41n.: 1,49, Spiegel, Friedrich: 62n. 3 1.
52,53,53n. S2and n. 83.66n. 48,67,75, Sraob: 188, 189.
80n. 125, 96, 98, 99, 99n. 17, IOln. 24, Stein, Otto: 73n. 95.
103, 104, 108, 116, 121, 127, 131n. I, Steptoe, P.C. See Edwards, R.G.
142n. 3, 183, 200, 207, 218, 219, 221, Stricker, B.H. : 99n. 16, 130-135 passim,
7n. 42 and n. 43,248n.80.250n. 91. 130n. 82.
Zdhyarkdina recension [SBM.l: iiidra: Ritual combat with an Aryf 9,
139n. I, 143n. 2,19811. 73, 199,203,226. 155. Day-night associated with Arya-
Kayv'jla recension [SBK.]: 139n. I, Sijdra: 104. ~~draspersonifyporvers of
203. nether world: 155.
61ta~aji~i: 23 1. Sun: 31, 7211. 83, 75, 80-81, 87-88, 96,
S'iyaaa: 145, 210, 213. 126-128, 176. Sun in the rock: 70-7 1,
Schaeder, Hans Heinrich: 62, 62n. 27, 83, 84, 85, 87. S. imprisoned in
63n. 27, 63n. 32, 86. primordial hill: 14 l. I n q a causes the S,
Schaefer, Ernst: 2 15. to rise: 106. Rises from the seat of Rta:
Schaerer, Hans: 5, 91, 110, 127, 186-188, 225. Rising S. a nzanifestarion of Agni
192, 20111. 80. and Mitra: 45. Setting S. sinks into
Schlerath, Bernfried: 56n. 4, 59n. 20. primeval waters: 3 1, 96. Nightly
Schmidt, W.P.: 6311. 34, 193n. 63. journey of the: S. 75-76, 78, 83-84.
Schomerus, H.W.: 1 1511.50. Selling S. a manifestation of Varuya:
Schwyzer, Eduard: 191. 45.67. S. called an eagle (suparq;): 70.
&$a: a Ghana of Vi.yip:33,48,110. V i y u Horses of the S. :29,67,82,95. V i p p ' s
sleeps on: 49. At Churning of the three steps refer to the S.: 43, 46. in
Ocean: 49n. 51. Supports both Heaven R V . , h o w e v e r , V Q e u has n o
 Index
connection with the S. :45. Winning of
the S. : 154-155, 203, 213.
Suparqa: 48, 75, 226. Messenger of
Varupa: 70 Uhlenbeck, Christianus Cornelius: 209.
Supar@khyiina: 2 19, 220, 22 1. Um5: 252.
Sjlrya: 7 1 Usas: 7; 55, 65n. 40, 106, 14111. 1, 156,
Susna: 28, 2811. 5, 69. 201, 209, 224. Born in Rta: 79, 160.
Suzuki, P. : 1YOn. 56. Kinswoman of Varuqa: 79, 160, 176.
Siitradhiira: 239-257 passim. Cosmic aspects: 159- 16 l. Dawn of t h i ~
kvetiifvatara Upanijad [Svet. Up.]: 84. (present) day: 164- 1 65. Overcomes
Syena: 219-220, 226. darkness (and insecurify): 168.
Connection with the New Year (winter
solstice): 106, 152, 157-159, 175-176.
Taittir& Krapyaka [Taitt. Ar.]: 36, Prayers to, for a new life: 165-167. U.
76, 77, 7711. 108, 84, 8611. 148, 145. a n d p r o g e n e r a t i o n : 168-169.
~ a i t t r G a Briihmana [TB.]: 3911. 46, Indentzfied with the Puramdhi: 161-
48n. 45, 49, 50n. 56, 52n. 73, 53n. 82 162. Uand wealth: 170. Clasgoddessof
and n. 83, 66n. 48, 101n. 23, 102n. 28, contests: 1 70-1 75.
199, 203, 207, 210. Usener, Hermann: 52n. 77.
Taittir?'ya SamhitZ [TS]: 17n. 16, 44, &sa (primordial well, cosmic well):
48, 4811. 46, 50n. 56, n. 58 and n. 60, 37, 143, 144, 183. Milked by the
53n. 83, 54n. 86, 5511. 96, 101n. 23, Maruts: 149. Drawn up to heaven: 150.
102n. 27and n. 29, 1041.31, 117,142n. Inherited from common Indo-Iranian
3, 143n. 2, 161, 163, 168, 183,218,220, religion: 183.
221, 24811. 80. u t t h i i p a n a : 2 3 3 - 2 3 7 , 2 3 9 , 242,
TaittirGa Upanisad [Taitt. Up.]: 255, 257.
8611. 148. Uttanka: 7111. 81.
Taksaka: 145. Resides in the jarjara:
241.
Taoism (embryonic breathing in): 125- Vic: 29n. 6
126, 136. Vairya: 190.
Taraporevala, lrach J. S. : 57n.6.64n. 37. ~aisnavT:5011. 56.
Tavadia, Jahangir C. : 27, 6511. 38, 214. vai$ya: 34.
Thieme, Paul: 33n. 13, 36n. 28, 59n. 19, Vc7jasaneyi Sumhit; [VS]: 44, 50n. 56,
60n. 23, 67n. 56, 69n. 68,75n. 102, 77, 51n.69,52,53,54n. 86,55n. 96.74n. 99,
143, 146x1. 1, 151, 15811. 33, 181, 197, 82n. 131, 108, 15411. 10, 208, 210.
2301-1. 3. Vala: 6, 28. Identz3ed with primordial
TiJtr Y a k 189. hiN:138. Identijied with'kob: 139. A
TiStrya: 189-190. treasure-house: 140. Founded on a
Toba-Batak: Ritual dance of: 118-120. rock: 140-141. Source of waterfrom the
Religion of: 136. rock: 144.
Tobing, Ph. L. : 92, 11th. 59. van Blankenstein, M. : 192, 19211. 59.
Tree of Life: 11, 36, 91-92, 118-119, ~ a r i h a - & h a : 33.
120, 123-130. Planting of the T.in the Varuna: 9, 15-17, 18, 20, 3811. 42, 47,
egg: 120. R m e d in piitra of primeval 48,53,65n. 40,68-69,75,94,95-96,97,
waters: 139. Supported by Varuna: 116, 156-157. V. king of the Asuras: 14.
144. n o birds of: 126-127. 128. See V. stands for the Asuras: 54. V.
also: 'Primordial Tree'. becomes a Deva: 15, 16, 2011. 22.
Tkirijtra Briihmana: 39n. 36. Resides in the subterranean cosmic
Trivedi, K.H.: 23411. 25. waters: 16,37. God of stagnant waters:
Tuxen, Poul: 4111. 2. 183. V. S subrerranean palace: 72, 74,
1 15. V. at roots of !he world free: 16,68.
Upholds the world tree: 36, 37,7677,
 145. Upholds "inverted tree'': 145. V. Vorulja: 48. V+hqu- Krgra: 33. V i g u -
drvells at lotr~erend o f world axis: 68, Niiriyapa: 55. jeSa the viihana of V. :
144. V. supports Heaven and Earth by 33, 49, 110. Garuda the viihana of V.:
cosmic pillar (axis, tree): 68, 76, 77, 78. 33, 110. As'vattha connected with V. :
V. guardian ofcosmic law (Rta): 153. 3811. 42. V. resides in the jarjara: 241,
Presence o f Sun in V. S dwelling: 7 1. 245.
Setting Sun a manijestation of V. 45, Visnudharmortarapurijna: 23811. 45,
67, 73n. 95, 96. V.? realm also locared 242n. 63, 250n. 93, 25111. 98, 252.
in the sk.v: 37. In the nocturnalsky: 79. Visnusmyti: 9711. 12.
148, 150. Suparna a messenger of: 70. Vikakarman: 98.
Usas a kinsuoman of: 79, 160, 176. ~,idvarC~a: 1711. 16.
U ~ a born
s ,from r e a h of: 160. V. and vivIc: 198-208 passim.
Mitra: 31. 66, 68, 69n. 68, 74, 80. Vivasvant: 48. 82. ~ Z t a r i h a n rhe
YamaS realm the same as V. S: 82-83. messenger of: 224.
V. as iicZrya. 84n. 138. Vasigha made a Vodskov, H.S. : 50n. 59, 21711. I , 218n.
r ~ iby V. : 70, 87. Intercourse with 4, 222.
Vasigha: 88, 1 15. Varuna-cult in India: Vogel, Jean Philippe:37n. 32, 182.
89. The vid+aka a personification of: Vohu Manah: 65.
254-255. Greeted by Siitradhiiru:245. von Bradke, P. : 14n. 10, 66n. 48.
Varu~adeva (in ChambZ State): j2n. von Eickstedt, Egon: 42n. 13.
91, 183. von Negelein, Julius: 219.
Vasigha: 70, 87, 88, 205. von Planta, R. : 192n. 61.
Visuki: resides in the jarjara: 241. von Schroeder, Leopold: 4 In. 1,42n. 3 and
VEyu: 30n. 7 n. 15,7711. 112, 145n. 3, 157n. 18, 160n.
Vedas: Vedic belie& 21, 22, 3 1, 55. 29, 223.
Vedic conception of the Universe: 52. Vftri: 5, 6, 12, 17n. 15, 28, 711-1.
Vedic myth: 19, 20, 23, 27, 38, 39, 40, 83, IOln. 24. Indra ?fight against V. :
46, 52, 77, 83, 84, 85, 92, 97, 153, 225. 12, 34, 50-5 1, 97, 13.8. V i ~ n u Srole in
Vedic religion (Vedic theology): 9, 14, Indra- ~ r t r ajight:
' 45-46. 48, 50, 54.
18. .Vedic poets: 19, 47, 85, 164, 165, IndraS sla.ving of V. :50, 51, 53, 98,
182-183. Vedic texts: 15, 16, l7.20n. 22. 117, 152-153, 203. Significance of
Vedic language: 57, 59-60. Later Vedic primordial hill in Vytrcf-myth: 53, 55,
literature: 97. Vedic literary contests: 152-153. ~ p d - m y t ha Creation myth:
15lff. Vedic studies: 9, 15. 107, 115. Indro- Vrtrdfight associated
Vendidad: 177. 179. with New Year: 94, 98, 104-105, 106-
Vidiisaka: 254-255. 107. Mother of V. : 121-122.
vlg~dri:207-208. vrtra' (force of resistance): 12, 104,
VinatZ: 29n. 6, 32, 34, 198. 185. 2 17. vrtrrfhatya: 1 16, 152.
VirZj: 50. vyijxana: 1 77- l98 passim. 2 14.
Visnu: 9, 41-42, 1 10-1 1 1, 143. V. usuully
considered a solar deity: 41, 43. Vedic
V. had no connection with the Sun: 45. Wackernagel, Jacob: 5911. 18, 651-1. 39,
Vedic V. higher than Indra:20,54. V. in 147n. 3, 178, 202, 208, 239n. 48.
the upper world: 2 1, 33. V. sleeping on Wackernagel-Debrunner: 3711. 34,
the cosmic waters: 21, 33, 49. V. 14711. 3, 19711. 69, 198,209n. 84.2 12n.
situated at the centre: 33-34, 48, 49. 89, 213, 239n. 48. Andreas-
R e p r e s e n t s t h e U n i v e r s e : 54. Wackernagel: 5811. 14, 60n. 23.
Associated with the cosmic pillar: 49, Waters: See 'Primeval Waters',
54, 55. Associated with mountains: 55. 'Cosmic Waters'.
V. ? thirdstep: 19-20,41-55,225. V. and Weber, Albrecht: 4111. I, 7511. 103, 145,
Indra: 46-47.48-49. 50-52, 104. V. and 154n. 9.
Weller, Friedrich: 77n. 112, 84, 140. Dwelling in the harmya': 69. Ins
Wensinck, A.J. : 4, 99n. 16, 107n. 43, Naciketas in thefire cult:
109. identified with Varuna's :82-83.
Westergaard, N.L. : 60, 179. Yarnunz: 32.
Whitney, Williarn Dwight: 4111. 1, 196, YSiska: 4111. 1, 52, 116.
197, 209. Yasna: 86n. 151, and n. 152, 8711. 155, f

Widengren, Geo: 15411. 7. 88x1. I61,n. 164andn. l65,89,128.177,

Wikander, Stig:34, 34n. 23, 35, 12111. 179, 180, 183, 184, 185, 191, 193, 218.
61 and n. 63, 181. Yasna Haptanhiiiti: 87.
Wilkins Smith, Maria: 5811. I I, 59% YaSt(s):86n. 153, 167, 177, 178, 179,
20, 60x1. 23. 181, 184, 185, 186, 188, 189, 190, 191,
Williams Jackson, A. V. : 60n. 22 218.
Wuest, Walther: 42x1. 15. Yggdrasil: 4911. 52.
Yoga: 125-127.
Yudhisthira: 34.
Xerxes: 86-87, 87n. 154.
Zarpthustra: 56, 5611. 4, 58,
Y2jllavalkya: 238n. 45, 251n. 95. 80, 81, 86, 88, 8811. 161, 89, 185.
Yajurveda:39, 44,48, 50, 52, 5311. 83,7111. Zeus: 216, 221-222.
83, 98, 101, 103, 108, 200, 225-226. Zimmer, Heinrich: 83n. 136,
Yama: 48, 81, 241, 245. V i ~ n uidenr19ed 16On. 29, 21 1. Re: s a b h g y a :
with in Mahiibhiirata: 481-1. 47. 199.

line 12-for p. 142 read p. 46; p. 51 11.71 line 2- for

line 19- for p. 139 read p. 4 .52 n.76-for p. 140 read p. 44, p