Religions of Ancient India

Religious Studies: Bloomsbury Academic Collections

This set on religions of the world contains nine facsimile editions of titles from our imprints T&T
Clark, Geoffrey Chapman, The Athlone Press and Cassell and discusses a variety of religions and
traditions outwith the traditional canon of Abrahamic religions, from ancient times to new religious
movements. With noted authors such as Jan Gonda, Louis Renou, Robert Charles Zaehner and
Raphael Werblowsky, the titles presented here offer a valuable resource for religious students and
scholars alike.

The collection is available both in e-book and print versions.

Titles in Religious Studies are available in the following subsets:

Religions of the World

Comparative Religion
Christianity and Society
Religion, Sexuality and Gender

Other titles available in Religions of the World include:

Beyond Tradition and Modernity: Changing Religions in a Changing World by R. J. Werblowsky

Freedom and Authority in Religions and Religious Education edited by Brian Gates
Hindu and Muslim Mysticism by R. C. Zaehner
Perspectives on New Religious Movements by John A. Saliba
The Community of Religions: Voices and Images of the Parliament of the World's Religions edited
by Wayne Teasdale and George Cairns
The Encounter with the Divine in Mesopotamia and Israel by H. W. F. Saggs
The World's Religious Traditions: Current Perspectives in Religious Studies edited by Frank
Whaling
Visnuism and Sivaism: A Comparison by Jan Gonda
Religions of Ancient India
Louis Renou

Religious Studies: Religions of the World

BLOOMSBURY ACADEMIC COLLECTIONS

Bloomsbury Academic
An imprint of Bloomsbury Publishing Plc
 Bloomsbury Academic
An imprint of Bloomsbury Publishing Plc

50 Bedford Square 1385 Broadway

London New York
WC1B 3DP NY 10018
UK USA

www.bloomsbury.com

BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc

First published in 1968 by The Athlone Press

This edition published by Bloomsbury Academic 2016

© Bloomsbury Academic 2016

Louis Renou has asserted his right under the Copyright, Designs and Patents Act, 1988, to
be identified as Author of this work.

All rights reserved. No part of this publication may be reproduced or transmitted in any form or
by any means, electronic or mechanical, including photocopying, recording, or any information
storage or retrieval system, without prior permission in writing from the publishers.

No responsibility for loss caused to any individual or organization acting on or refraining from
action as a result of the material in this publication can be accepted by Bloomsbury or the
author.

Every effort has been made to contact copyright holders of material reproduced in this volume.
If any copyright holder has not been properly acknowledged, please contact the publisher who
will be happy to rectify the omission in future editions.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

ISBN: HB: 978-1-4742-8818-7

ePDF: 978-1-4742-8819-4
Set: 978-1-4742-9198-9

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

Series: Bloomsbury Academic Collections, ISSN 2051-0012

Printed and bound in Great Britain

RELIGIONS OF
ANCIENT INDIA
RELIGIONS OF
ANCIENT INDIA

LOUIS RENOU

SCHOCKEN BOOKS • NEW YORK

 First SCHOCKEN edition 1968
Second Printing, 1970
Copyright 1953 by School of Oriental & African Studies,
The University of London
This edition published by arrangement with The Athlone Press
Library of Congress Catalog Card No. 68-16660

Manufactured in the United States of America

 T H E L O U I S H. J O R D A N B E Q U E S T

The will of the Rev. Louis H. Jordan provided that the greater
part of his estate should be paid over to the School of Oriental
and African Studies to be employed for the furtherance of
studies in comparative religion to which his life had been
devoted. Part of the funds which thus became available was to
be used for the endowment of a Louis H. Jordan Lectureship
in Comparative Religion. The lecturer, appointed annually, is
required to deliver a course of six or eight lectures on compara-
tive religion, for subsequent publication. The first series of lec-
tures was delivered in 1951.
 PREFACE

T H E S E lectures were delivered (in a slightly abridged form)

in May 1951, at the School of Oriental and African Studies
of the University of London, when I had the honour of
being invited by the Director and the Academic Board of
the School to give the first series of six lectures under the
Louis H . J o r d a n Bequest, on the Religions of India.
The field surveyed in the summary studies that follow is
so vast—with the exception of Buddhism, it embraces all
the religious manifestations of India, past and present—
that inevitably many aspects are only briefly treated, while
others receive no more than a cursory mention. The reader
desirous of more detailed information will find references in
the footnotes to the bibliography of the principal questions
dealt with.
My intention in these lectures was to give an account of
the present state of the main problems. The detailed
scholarly studies which constitute the standard works on
the subject, and which provide the basis and justification
of all contemporary work, do not always enable the reader
to see the facts with which they deal in their proper per-
spective and in their relationship to the general back-
ground. It is useful to take stock of our position from time
to time, so that we can form some estimate of the stage that
our researches have reached, whether they have made pro-
gress, or whether the position is merely stationary or even
perhaps less assured than it seemed before. The world of
Indology is constantly evolving; and while fresh views are
continually being advocated, the research work that is an
Vili PREFACE
essential preliminary to new advances, editions of texts,
studies of vocabulary, learned monographs and so on, fol-
lows more slowly and usually avoids the conclusions that
more impatient theorists wish to see adopted at all costs. In
this study I have tried to advance moderate, and, as I hope,
reasonable views.
There remains the pleasant duty of expressing my thanks
to Miss Sheila M. Fynn, Assistant Lecturer at the School of
Oriental and African Studies, for her clear and accurate
English translation of a text which, I am afraid, was not
particularly easy.
L. R.
 CONTENTS

First Lecture VEDISM: I i

Second Lecture VEDISM: II 23
Third Lecture HINDUISM: I 46
Fourth Lecture HINDUISM: II 68
Fifth Lecture HINDUISM: III 89
Sixth Lecture JAINISM HI

Index 135
RELIGIONS OF
ANCIENT INDIA
 I
Vedism: I

T HERE was a time, not very long ago, when every

Indianist (especially those of the German school)
considered that a monograph on Vedism, or an edition of
a Vedic text, was an ideal first piece of research to under-
take. The fact that Vedic studies no longer occupy such a
focal position is probably the outcome of this earlier con-
centration on them to the relative neglect of other fields.
The student is likely to feel that there is nothing left to dis-
cover (although in this he would be wrong), or at least that
what remains to be discovered would not repay the time
and trouble needed to attain a mastery of the subject.
Another reason for this decline of interest is the isolation
of the Veda. Nowadays our attention is centred on cultural
influences and points of contact between civilizations. The
Veda provides little of this sort of material, for it developed
in seclusion. Yet perhaps it is really more important to
begin by studying individual manifestations in and for
themselves, and to examine their own internal structure.
In this field of research we have an exceptionally good
collection of works of reference; there is no lack of diction-
aries, grammars, editions, translations and monographs on
various aspects of the religion. Though much remains still
to be done, the work already accomplished has been car-
ried out with greater thoroughness than is the case in many
other branches of Indian studies. Yet we cannot say that a
satisfactory picture of the subject as a whole has been
achieved, and a sense of frustration pervades the present
2 VEDISM: I

state of Vedic studies. The healthy scepticism of Whitney

and, more recently, of Keith, may have contributed to the
present uncertainty, no less than the over-enthusiasm of
various other research workers, both in Europe and in
India.
In the first place, no definite chronology can be estab-
lished, and this is an embarrassment to Western scholars.
The position is admittedly the same in many other fields of
Indian studies, for example in that of early Buddhism: but
in that case there is at least a basis for discussion. It is clear
that the oldest Vedic texts in their earliest redactions are
posterior to the Aryan invasion of India. But this gives us
only a very rough indication of date, especially in the case
of mantras where it is essential to distinguish between the
gradual process of the composition of texts, and their 'oral
transcription5 as we know it today. This transcription
developed late, under the impetus of a rapidly changing
language. The mantras, even those of the Rgveda, represent
a fairly late stage of phonetic development: their language
is not entirely homogeneous, and moreover diverges appre-
ciably from the forms which we can occasionally restore on
the evidence of the metre. We often hear it stated that the
textual tradition of the Veda has been handed down with
scrupulous exactitude; but this statement needs qualifica-
tion. From the time when it was established in its tradi-
tional form, the Rgveda has in fact been carefully preserved
from any alteration. We know this from present-day recita-
tions, which correspond exactly to the rules given in the
Pràtisàkhyas. But before this, throughout the long centuries
during which the hymns were composed and handed down
within individual families, and used at ceremonies, they
were exposed to change. The literary forms of the other
texts were established on widely varying principles. Many
treatises have been lost or damaged, entire schools have
fallen into oblivion; side by side with a rigorous conserva-
 VEDISM: I 3
tism such as that of the Satapatha-Brdhmana, we find a cor-
rupt text like the Paippaldda recension of the Atharvaveda.
One has only to consult the Vedic Variants1 to realize that
the changelessness of the Veda is a fiction.
In the late nineteenth century various erroneous specu-
lations on the chronology of the Veda were advanced, and
these did great disservice to the subject. It is regrettable
that a scholar of Hermann Jacobi's eminence should have
been associated with such misconceptions.2
It was at first hoped that the discoveries at Harappa and
Mohenjo-Daro might throw some light on the Veda; but
this hope was not fulfilled. What is known as the Indus
civilization appears to owe nothing to the Veda, for indeed,
in its origins at least, it is definitely of earlier date; nor does
it appear that the Veda owes anything to it. The Aryan
tribes may well have overrun it without in any way being
influenced by it, settling on the ruins of a decayed or decay-
ing empire. If the forms of religion revealed in the seals and
figurines of the Indus have any remote connection with
Indian forms, it is not so much with those of Vedism as
with those of Hinduism, a Hinduism which, though known
to us only by inference, must have already existed in Vedic
times, and probably considerably earlier. The Harappa in-
scriptions would no doubt tell us more on this subject, but
until they are deciphered it is idle to try to explain the war
of the Ten Kings as a clash between the people of the Indus
1
An unfinished work, begun by M. Bloomfield and F. Edgerton,
continued by M. B. Emeneau. Three volumes have appeared so far
(Philadelphia, 1930-34): i, The Verb; ii, Phonetics; iii, Noun and Pronoun
Inflection.
2
Cf. 't)ber das Alter des Rgveda', Festgruss R. Roth (1893), p. 68;
'Beitrage zur Kenntniss der Vedischen Chronologie', Getting. Nachr.
(1894), p. 105; 'Der Vedische Kalender und das Alter des Veda',
Z. dtsch. morgenland. Ges., 49, p. 218; 'Nochmals iiber das Alter des
Veda', ibid., 50, p. 69. Contra (among others) A. Barth, J. Asiat., 1894,
1 st part, p. 156 ( = Works of Barth, iv, p. 168); H. Oldenbèrg, Z- dtsch.
morgenland Ges., 48, p. 629; 49, p. 470; 50, p. 423.
4 VEDISM: I
and the Aryan invaders, or to connect the name 'Harappa*
with the river or region called Hariyùpiyà. In the very
primitive architecture which we can infer from descriptions
in the Vedic ritual texts there is nothing that can reason-
ably be compared with the buildings of Mohenjo-Daro,
unless we postulate a social and artistic retrogression in-
compatible with the high spiritual plane of the hymns. In
short, we are faced with complete defeat in this quarter.
It seemed possible at one time that a guiding date had
been found in an Anatolian text mentioning a treaty be-
tween a Hittite king and a prince of Mitanni, which
attested the existence of certain divine names in the four-
teenth century B.C. But this unexpected explicitness as to
date only opened up new problems: does the text really
refer to Vedic gods, or does it refer to Indo-Iranian gods?
Had they been brought from India by emigrants or colon-
ists, or were they at this date only moving towards India?
At the most, the treaty of Mitanni attests the existence in
the fourteenth century of a series of divine names of the
Vedic type; it gives no evidence of established texts or fixed
forms of religion at this time.
The geographical environment of the Veda is equally
uncertain. We have, of course, an approximate indication
of where the Vedic tribes were situated; it is possible to
pick out Iranian features, or rather, features of modern
Afghanistan, from a background of a more specifically
Indian type. This limited horizon gradually widens to-
wards the east to reach the Ganges, and certain scholars
take it for granted that a late text like the Satapatha-
Brdhmana can be located in Videha, the modern Bihar, on
the borders of Bengal. This is too bold a supposition. All
the Vedic texts, including the Satapatha, originate either in
the Upper Indus or in the region known as the Kurupan-
càlas which forms the natural eastward continuation of the
plains of the Punjab. Admittedly, tradition has it that there
 VEDISM: i 5
were schools, and therefore texts, in central and southern
India, and mediaeval inscriptions confirm this dispersal. But
the establishment of these schools was much later than the
composition of the texts. It is important to remember that
the source of Vedic inspiration is a single fountain-head,
from which many streams flowed out in course of time. We
are constantly brought back to this idea of a Veda existing
before the schools, incorporating in itself a mythology and
an agreed ritual, that had arisen in the original Vedic com-
munity of clans and families. We cannot explain the liter-
ary diversity of Vedism except by postulating an original
unity, and it is unfortunate that this 'Ur-Veda' cannot be
reconstructed with any certainty from the texts in our pos-
session. The scholar is confronted by similar situations in
many other branches of Indology.
We are hardly in a better position to describe the decline
of Vedism. We know that the official cult lapsed (to be
revived later, more or less artificially); the private cult
underwent transformations; the mythology, which was
probably becoming archaic at the time of the Rgveda, was
reconstituted. But there is no means of assigning a time or
place to these changes. We are reduced in the end to a
purely formal definition of Vedism: any text composed in a
certain style and following a certain pattern is Vedic. It
must, of course, be understood that creative literary acti-
vity in this field eventually gives place to the endless
elaboration of commentaries.
The texts are in fact our only means of defining this reli-
gion, which has left us no archaeological evidence, and
which possesses no dogma or founder, no church or history.
In spite of the profound imprint it has left on later Indian
culture, Vedism in its formative period had no far-reaching
influence. The Vedic clans, or Aryas, as they called them-
selves, were surrounded by the hostile mass of indigenous
dàsas, a dark-skinned pre-Aryan people who may have been
6 VEDISM: I

Dravidians. The clans themselves were divided: there were

the aris or 'strangers' who were sometimes allies or protec-
tors, but more often rivals to be watched. There were the
vrdtyas, whose religion Vedism tried to absorb into itself,
and whose texts may have been incorporated into the
Atharvaveda.
It is idle to look for a historical pattern underlying this
terminology: but none the less it is possible to see in the
descriptions of battles, alliances and quarrels, especially
those of the Indra cycle, a gradual transformation of chief-
tains and tribes into demi-gods, demons and divine cohorts.
The very ambiguity of this mythology reflects the vicissi-
tudes of a community living in constant danger.
Vedism was in the charge of a priestly élite who served a
military aristocracy; it may well be that the masses were
already Hindu. The private or domestic cult was only
Hindu practice with a veneer of hieratic formulae. T h e
ceremonial cult represented true Vedism, together with the
orthodox mythological traditions, the solemn religious
festivals and the poetic and other contests which were in-
cluded in them. It was, however, by no means a public cult.
If we look once again at the chronological side of the
subject, we shall see that the end of what is usually called
the Vedic period does not coincide with the beginning of
literary Hinduism nor with the beginning of the Buddhist
and Jaina movements. The links between Vedism and the
great epics, the earliest non-Vedic texts that we possess, are
extremely slender. The Kurus and their King Pariksit,
before whose son the Mahdbhdrata was first recited, are
mentioned as almost legendary figures in the Brhaddran-
yaka-Upanisad. According to the statements of the Purdnas
(which must be treated cautiously), the war described in
the Mahdbhdrata took place at a date corresponding to 1400
B.C. : this would then be the time of Pariksit, who is spoken
of as a living ruler in the Atharvaveda. It has also been
 VEDISM: I 7
pointed out that another Upanisad, the Chàndogya, mentions
the name of Krsna Devaklputra as a Brahman student, and
that this is evidently a reference to the hero who was an
ally of the Pàndavas. The links with the Ràmàyana are of
the same type, though even less conclusive : King Janaka of
Videha, father of the Princess Sita, is an important figure
in the latter part of the Satapatha. There is nothing in all
this on which a relative chronology could be based: but
one can sense a 'sentimental5 connection between the last
stirrings of the old theocracy and the secular society which
was to produce the Epic.
The connections with Buddhism are very deceptive.
Ever since Kern, 1 of course, analogies have been seen be-
tween Vedism and the Hìnayàna, and more particularly
between the domestic sutras and the Pràtimoksa; but the
explanation lies not in borrowing but in a basis of thought
common to both. The danger in considering Indian reli-
gions as separate groups of phenomena is that one tends to
forget that certain essentially Indian features characterize
them all. Other scholars, venturing into the realms of
speculation, have pointed out resemblances between the
Vedic mysteries centring on the fire-altar, and certain
features of the Buddhist cosmology which might be des-
cribed as architectural mysticism. Senart2 drew attention
to various episodes in the life of the Buddha which could
be interpreted in accordance with the old myths. These
comparisons can be multiplied indefinitely, but we must
not read too much into them. The Upanisads are a parti-
cularly delicate case; the problem, stated in simplified form,
has been whether the Upanisads were pre- or post-Buddh-
ist. Their subject-matter and method of presentation
1
Cf. J. C. H. Kern, Geschiedenis van het Buddhism in Indie (Harlem,
1881; French translation, Paris, 1901); and Manual of Indian Buddhism
(Strassburg, 1896).
2
Émile Senart, Essai sur la legende du Buddha, son caractère et ses
origines (Paris, 1875; 2nd edition, 1882).
8 VEDISM: I
have much in common with Buddhistic writings; the Pàli
style seems, indeed, to be a diluted imitation of the Upani-
sadic style. The secular approach of the Upanisads is char-
acteristic also of Buddhism and Jainism, those religions of
princes. If we work on the presupposition that in India
progress is from the simple to the complex, from brevity to
elaboration, then the Upanisads must be regarded as
earlier. This is my own view. But we must not be surprised
to see that in India parallel streams of thought may exist
side by side without any contact other than an unem-
phatic rivalry. If, on the other hand, we believe that the
Upanisads were only made possible through Buddhist in-
fluence, or, in other words, that 'it was Buddhism that
taught the Indians to philosophize', we are losing sight of
the fact that Vedic speculation is firmly established from
the Rgveda onwards, not only in the tenth book, which
summarizes a great mass of speculative material, but even
in what is known as the older Rgveda, for example, in iii.
54, 9 : 'I recognize from afar the ancient and immemorial
one. We are descended from him, the great Procreator, the
Father. The gods who do him homage, in their own vast,
separate domain, quickly took up their positions in the
intervening space . . .' Here we already have a full formu-
lation : the single original principle, and the realm of the
gods lying between Man and the Supreme Being. Religion
and speculation go hand in hand from the very outset.
If we cannot establish an absolute chronology, let us see
where there is any hope of setting up an internal relative
one. According to Max Muller, 1 the Vedic age extends
over four great periods, each one lasting for centuries. In
the first period the hymns were composed, in the second
they were established in set forms, in the third commen-
taries were composed on them, and in the final period the
1
C£ esp. A History of Ancient Sanskrit Literature (London, 1859; later
editions, i860 etc.).
 VEDISM: I 9
rites with which they are connected were described. In the
same way others have postulated an epic age, and a canon-
ical Jaina or Buddhist period. To cover any stretches of
time not thus accounted for, the decline and subsequent
renaissance of Hindu civilization has been invoked.
It is always dangerous to base a chronology on litera-
ture. It is clear that the Vedic hymns as a whole date back
to a period before the commentaries and descriptions; but
they are not evidence of a spontaneous aesthetic response
to natural phenomena. They presuppose an established
religion of which our knowledge is imperfect, but which
must have borne some resemblance to that described in the
later Kalpa. Only in the Yajur- Veda do we get a full idea of
it; and in the yajus the so-called mantra and bràhmana
periods are indistinguishable, according to Max Muller.
Some scholars have maintained that certain parts of the
Vàjasaneyi-Samhità were composed after the $atapatha\ and
Caland held that the sutras cannot be assigned to any speci-
fic period.1 We are led to the conclusion that some texts
attained their definitive form at an older period of the
language than others. Sylvain Levi held the paradoxical
theory that there was no Vedic chronological problem and
that all the Vedic works were compiled together at one and
the same time, relatively late.2 Similarly, Darmesteter
considered the gàthàs to be a late text, which differed from
the bulk of the Avesta in orthography only.3 There is some
truth underlying these theories : the establishment of the
texts from a combination of scattered sources must have
1
W. Caland in Getting, gelehr. Anz-, 1902, p. 122; edition of the
Satapatha-Bràhmana in the Kdnvlya Recension, i, 99.
2
This theory was put forward in his lectures, and is remembered by
those who attended them (cf. L. Renou, 'Sylvain Levi et son oeuvre
scientifique', J. Asiat., 1936, 1st part, p. 17); v. Memorial Sylvain Levi
(Paris, 1938), p. 8, and iflnde Civilisatrice (Paris, 1938), p. 35.
3
Cf. esp. the introduction to his English translation of the Zend-
Avesta in the Sacred Books of the East (French translation, Paris,
1892-93)-
IO VEDISM: I

taken place at one definite period, in response to a com-

mon demand; but the material used dated from widely
differing periods of antiquity.
The Rgveda is much more than an adjunct to ritual. It
might be called a literary anthology, drawn from family
traditions. The religious expressions found in it are poetic
exordia to the cult, and are not designed as the direct
accompaniment of ceremonies. We see something of the
same kind in the Pràtaranuvàka of the Agnistoma. I imagine
that the works which have survived are those which ful-
filled the requirements of a poetic competition. It has been
pointed out that the hymns suggest the atmosphere of a
contest in eloquence. The aim was to compose on a given
theme, or perhaps according to a given plan, not introdu-
cing direct accounts of the lives of the gods so much as veiled
allusions, occult correspondences between the sacred and
the profane, such as still form the foundation of Indian
speculative thought. A large part of Sanskrit literature is
esoteric. These correspondences, and the magic power they
emanate, are called brahman: this is the oldest sense of the
term. They are not intellectual conceptions but experi-
ences which have been lived through at the culmination of
a state of mystic exaltation conceived as revelation. The
soma is the catalyst of these latent forces. T h e designation
kavi is given to the poet who can seize and express these
correspondences, and to the god who sends him inspira-
tion. The term vipra, literally 'the quivering one', is also
used. This suggests the mystical quivering described by the
Kashmiri Spanda school. Traces of this mystical intoxica-
tion can often be found in cult practice. The kavi of the
classical period, the learned poet, transposes the old Vedic
ambiguity to the aesthetic plane by means of double mean-
ings and multiple senses; the classical vakrokti, 'tortuous
speech', calls to mind the epithet vankuh kavih used of
Rudra. This is the reason for the intricacies of Vedic style
 VEDISM: I ii
and vocabulary. A contributory factor, too, was the Indo-
Iranian tradition of verbal esotericism, evidenced by the
gdthas. Moreover, samsa (praise) is sacred in itself: the man
who has the gift of samsa receives it from a god in the same
manner as a material gift.
But on the whole the raison d'etre of the hymns lies in the
cult. Bergaigne showed that many of them were composed
with a definite liturgical purpose in view.1 He may have
pursued this line of thought a little too far, but it is a pity
that his researches, like those of so many others, were inter-
rupted. Some hymns were undoubtedly only accessory to
the liturgy, some were entirely secular in tone; others may
show a reaction against liturgical dominance. They were
all pressed into use by those who compiled the final ritual
forms, just as the Avestic gdthas were used in a sacrifice for
which they were not originally intended.
We cannot, however, reconstruct this early cult. Our sole
data would be the soma ceremonies, which are the only
rituals that the hymns treat in detail. In these rites alone
does the hotr play a part, and the Rgveda is primarily the
manual of the hotr and of the udgdtr (for whom a special
text, the Sdmaveda, was later compiled). But the Rgveda,
although it describes the preparation of the soma at great
length, hardly mentions the other operations which, to-
gether with the soma, make up the ritual of the great sacri-
fices. Animal sacrifices, for example, are barely touched
upon. In order to understand the nature of the early Vedic
cult we should need to possess the formulary of the adhvaryu,
the officiating priest. In the Tajurveda we have a much later
version of this, in the recensions of several distinct schools,
and full of borrowings from the Rgveda. We should also
need the paddhati, a guide which would enable us to follow
a ceremony contemporary with the hymns.
1
A. Bergaigne, 'Recherches sur Phistoire de la liturgie védique',
J. Asiat., 1889, l s t part, pp. 5 and 121.
12 VEDISM: I
At first sight it seems that we are better informed on the
subject of Vedic mythology. There are few hymns which
do not contain mythical allusions of some kind, and many
of them abound with references to the exploits and adven-
tures of the gods. But it is impossible to establish a history
of the gods from this material, obscured as it is by the con-
stant repetition of the same phrases in different contexts.
We are really dealing not so much with individual gods as
with mythological contexts, founded on what Max Muller
called 'henotheism'.1 The term was later ridiculed; but it
expressed his meaning, and represents a permanent feature
of Indian thought, which is especially noticeable in
Sàktism. It is the tendency of the worshipper to ascribe the
attributes of other gods to the particular deity whom he is
honouring. The mythology of the Rgveda is confused,
with no beginning or end; it is a mythology in the making,
constantly relating itself to the process of the Creation, and
to the theme of the early ages of the world.
We can, of course, invoke external evidence to throw
light on these allusions. The history of religion gives us
some guidance in classifying mythological types, and the
evidence of later India must not be overlooked. In the last
twenty years there has in fact been a rehabilitation in the
methods of comparative mythology, due to M. Dumézil.2
The reforms of Zoroaster obscure the surviving Indo-
1
In many of his works, and especially in the Lectures on the Origin and
Growth of Religion, as illustrated by the Religions of India (London, 1878;
later editions, 1882 etc.).
2
Cf. his early works Le Festin d'Immortaliti (Paris, 1924) ; Le Pro-
blème des Centaures (Paris, 1929). Others dealing more specifically with
Vedic material are: Ouranos-Varuna (Paris, 1934); Mitra-Varuna, 1940;
2nd edition, 1948; Jupiter, Mars, Quirinus, Essai sur la conception indo-
européenne de la société et sur les origines de Rome, 1941 ; Naissance de Rome
(=Jupiter, Mars, Quirinus, II), 1944; Naissance d'archanges (=Jupiter
Mars, Quirinus, III), 1945; Jupiter, Mars, Quirinus, IV, Explications de
textes indiens et latins, 1948; Servius et la Fortune, Essai sur la fonction
sociale de louange et de blame (etc.), 1943; Tarpeia, Cinq essais de philologie
comparative, 1947;/^ Troisième Souverain, 1949.
 VEDISM: I 13
Iranian features, but traces of them can be perceived in the
recent Avesta, especially in the Tafts, which have not been
fully explored from this point of view. The blurred outline
of a resemblance can be traced between Ahura Mazdàh
and the Vedic Varuna. The Amasa Spantas, formerly
thought to be connected with the Àdityas, now seem more
likely to be a transposition to the abstract plane of a series
beginning with Varuna and Mitra, and continuing through
Indra to the Asvins and others.1
Vedic mythology, like most Indian mythologies, con-
tains material which it is tempting to label as non-Aryan
or pre-Aryan, though no precise meaning can be attached
to these terms. M. Kuiper recently discerned a 'proto-
Munda' myth, as he called it, in the Veda: the story of an
archer-god who kills the boar Emusa with his bow drum-
bhùlL2 The fact remains that the representations of the
Divine in the Veda form an impressive and unified whole,
even though occasional elements are borrowed, just as a
great writer uses words from many sources. The synthesis
is a new systemic creation which owes little to what it has
inherited or borrowed.
Various planes are discernible in this mythology. On the
farthest plane, practically without mythology, are Dyaus
Pitar and Prthivi (Heaven and Earth). In the middle dis-
tance is the figure of Varuna, already fallen in some of the
hymns. With him we associate the age without rta, the
state of primeval anarchy which he brought to an end. In
the foreground stand Indra and his great cycle, aggressive
gods, who have absorbed the substance of many other
figures into themselves and now dominate mythical legend,
have the monopoly of relationships with the human race,
1
Gf. Dumézil, Naissance d'archanges, p. 56.
2
'An Austro-Asiatic Myth in the Rigveda', Mededeelingen d. koninkl.
NederL Akad. van Wetenschappen, Afd. Letterk., Nieuwe Reeks, Deel 13
(1950), no. 7.
 14 VEDISM: i
and are the chief protagonists in struggles against the
demons. Visnu and Rudra seem to be relatively new
figures. These various groups are linked together by the
agency of beings such as the Maruts and the Rudras. Lastly
the Fire in all its forms, the sacrificial Plant and the soma,
the liquor distilled from it, were given divine status and
incorporated into the legend of one of the gods; Bergaigne1
has clearly shown that the entire Vedic mythology was
reshaped or at any rate reorientated as a setting for Agni
and Soma, and that all the other divinities became counter-
parts or reflections of them.
The subject is still further complicated by the fact that
in the doings of the gods there are several levels of signifi-
cance. In part they are the transposition of natural pheno-
mena to the mythical plane. Vedic nature-worship, though
it was over-emphasized by Max Muller2 and perhaps also
by MacdonelP and Keith,4 is undeniable. We cannot, of
course, admit that lunar myth is as omnipresent as Hille-
brandt maintained,5 nor can we consent fully to Olden-
berg's thesis of planetary representations.6 But at all events
the Sun is an all-pervading, ever-present force, shown in
many forms, now directly, now symbolically. If it is true
that the rsis did all they could to obscure the lines of
approach that would have indicated a 'naturalist' inter-
pretation, it would account for much of the occultism of
the Veda, or the Vedic 'galimatias', as it used to be called.
1
La religion védique d'après les Hymnes du Rig-Veda (3 vols., Paris,
1878) ; cf. especially vol. i, p. xiii.
2
Op. cit.; also elsewhere, e.g. in Natural Religion (London, 1888;
later editions, 1892 etc.).
3
A. A. Macdonell, Vedic Mythology (Strassburg, 1897), passim.
4
A. B. Keith, The Religion and Philosophy of the Veda and Upanishads
(Cambridge, Mass., 1925), esp. p. 58.
5
A. Hillebrandt, Vedisene Mythologie (3 vols., Breslau, 1891-1902;
2nd edition, 2 vols., 1927-29).
6
H. Oldenberg, Die Religion des Veda (Berlin, 1894; 3rd edition,
J
923) ; 'Varuna und die Adityas', Z* dtsch. morgenland. Ges., 50, p. 43.
 VEDISM: i 15
1
M. Wikander was led by ethnographical analogies to
suppose that the Maruts represented a group of young men
formed with an initiation ceremony in view. But this only
amounts to saying that we must at all costs discover in the
Veda an institution that exists in other more or less primi-
tive cultures. M. Goossens recently produced an ingenious
theory: starting from the hypothesis of M. Grégoire, that
Asklepios, the son of Apollo, was originally the god of mole-
hills and the founder of a mole-cult, he then draws atten-
tion to a certain formula of the Tajus in which the mole is
called Rudra's animal. On this basis he builds up an am-
bitious mythological edifice with the object of identifying
Apollo and Rudra. 2 But in this case there is not even the
phonetic similarity which gives some support to the identi-
fication of Ouranos with Varuna, for example. The great
multivalency and indefiniteness of Vedic legend suggest all
kinds of associations to the mind which when formulated
prove insubstantial.
M. Dumézil has made a valuable contribution in bring-
ing to light the various social functions which underlie
Indo-European mythology; these functions are respectively
religious and juridical, military and temporal, and econo-
mic.3 M. Benveniste in fact found traces of a tripartite
social system in the Veda,4 and it is possible that certain
aspects of the representations of the divine and of the ritual
itself reflect a threefold arrangement. But tripartition
takes many forms in the Veda, and there is nothing to indi-
cate that it is an image of a social framework which is
hardly even mentioned in the Vedic hymns. In any case
1
Stig Wikander, Der arische Mannerbund (Lund, 1938).
2
H. Grégoire in collaboration with R. Goossens and M. Mathieu,
Asklépios, Apollon Smintheus et Rudra (Brussels, 1950).
3
Cf. works cited above, p. 12, n. 2, and chapter 1 of The Indo-
European Heritage in Rome (London, 1949).
4
Émile Benveniste, 'Traditions indo-iraniennes sur les classes
sociales', J. Asiat., 1938, 2nd part, p. 529.
i6 VEDISM: I

this explanation would cover only a very small proportion

of the total instances in the Veda.
There are many factors which might lead us to interpret
mythology in terms of ritual. Such interpretations have
been attempted for the Sautrdmanl and the Pravargya, for
example. But here again the difficulties are formidable. It
is often possible to discern certain correspondences be-
tween a particular formula and its accompanying rite, but
we cannot be sure whether we are dealing with an authen-
tic original parallelism or with a later adaptation, made
perhaps by playing on the words. In magical prayers the
correspondences are striking; they are remarkable too in
domestic rites, but in the official ceremonial they occur
more rarely. The most important mythical episodes, those
which reflect cosmogonie events, have no ritual equivalen-
cies; and in general they are not used in any way in the
classical ceremonies. We must be content with very
general theories if we are to avoid arbitrary explanations
such as those put forward in the old Brdhmanas, where we
find fabricated accounts of the origin of various details in
the liturgical ceremonial. In these stories there is much
that deserves attention, but the niddna or bandhu, the hidden
connection that they try to establish, cannot be accepted;
it is too visibly the product of the priestly mind. It is recog-
nized in the texts that comprehension must cease at a cer-
tain point : they declare 'paro'ksakdmd hi devdh\ 'the gods
love what is cryptic'.
The complexity of Vedic material is already recognized
in the JVirukta, which takes account of several methods of
exegesis. In company with the theories already surveyed,
especially the suggestion, difficult of proof, that historical
events were transmuted into myth, we must consider the
part played by 'ethnological' explanations : these are often
clearly justified, but if too much emphasis is laid on them
the individuality of particular variations tends to become
 VEDISM: i 17
absorbed into the undifferentiated substratum. We may
reject the psychical explanation, which occasionally ap-
pears in native exegesis, and which Aurobindo tried to
revive. * According to this theory the Veda is a vast piece of
symbolism representing the passions of the soul and its
striving after higher spiritual planes : thus the Veda, we are
told, ceases to be a barbarous and unintelligible hymnary.
I fear that it also ceases to be a document of prehistory and
becomes a manual of modern theosophy. Such an obvious
anachronism is not likely to convince any serious student.
If we had to choose one theory to work on, out of so
many passed in review and seen to be untenable, the pre-
ference would go to that of Bergaigne. His work, neglected
in his lifetime, has steadily gained in stature. We could not
adopt unaltered his general theory of interpretation,
according to which all mythological portrayals are vari-
ants of the sacred fire and the sacrificial liquor. But he has
shown us the right method, the method of seeking corres-
pondences between the world of men, the performers of the
sacrifice, the microcosm on the one hand, and the 'aeriaP
world of the gods, the macrocosm, on the other. The duty
of the rsis was to ensure the ordered functioning of the
world and of religious ceremonial by reproducing the suc-
cession of cosmic events, the ordo rerum, in their acts and in
the imagery they conceived. The term rta is a designation
of the cosmic order on which human order, ethics and
social behaviour depend. Analogous terms such as dharman,
dhaman, kratu and many others have a twofold application
according to whether they refer to man or the gods,
adhydtmam or adhidaivatam, as the Upanisads say. Seen in this
light, the Veda is a vast magical synthesis expressed in
symbolic terms. The images of the Veda have a ritual signi-
ficance in themselves; they bring about the ordered func-
tioning of a universe which is itself conceived as the scene
1
Cf. Hymns to the Mystic Fire (Pondicherry, 1946), p. xlviii.
 i8 VEDISM: I

of a vast sacrifice, the prototype of man-made sacrifices.

Thus Vedism is already Toga, a collective Toga in which
the composers of formulae, the early ancestors who in-
augurated the sacrifice, and the gods who are both wit-
nesses a n d participants, all play their part.
This, then, is the origin of Vedic esotericism, which, as
we have seen, is linked with the esotericism of later India,
as it appears in the Tantras, in learned poetry, in the
theories of aesthetics on which this poetry is based, and
even in legal tradition. The Indian mind is constantly
seeking hidden correspondences between things which
belong to entirely distinct conceptual systems.
In the Upanisads, all these correspondences are reduced
to the comprehensive equation atman/brahman, which ap-
peared to the new kavis as a résumé of the whole of Vedic
thought. The word upanisad itself, as it is first used in the
Satapatha Brdhmana, means only 'equivalence'. According
to S.B. X. 4 , 5 , i, the function of the upanisadis to formulate :
Agni is the Wind, Agni is the Sun, Agni is the Year. Hence
the aim of the whole of Vedic thought may be expressed as
the attempt to formulate upanisads. The texts thus defined,
the Chandogya and the Brhaddranyaka, which date from the
end of the Vedic period, are far from undermining the
speculations of the hymns and the Brdhmanas; in fact they
carry them to their logical conclusion. The Vedic and the
Upanisàdic texts both seek the same end, but they use
different means. Vedic thought is scientific in character, or
perhaps, more accurately, pre-scientific. Its chain of rea-
soning starts from the brahmodya, discussions of the cosmic
enigma or brahman, in the old sense of the word. T h e Vedic
riddles, as they used to be called, are not the intellectual
diversions of ingenious poets, but represent something of
far deeper significance.
It would never be my intention to try to find a single key
to the interpretation of the Veda. Mythological legend,
 x
VEDISM: I 9
considered by itself, expresses many widely differing truths.
The most expressive of these myths are those which deal in
some way with the creation of the universe, the establish-
ment of heaven and earth, the coming of light and the
release of the waters. The theme of the struggle with the
dragon is connected with this type of myth. The struggle is
described in many forms, but most commonly it is super-
imposed on an older, more abstract theme, that of the vic-
torious hero, usually Vrtrahan, who overcomes the enemy
resistance. The word vrtra used as a neuter noun meant the
defences of the enemy; later it came to be used as the name
of a demon in the shape of a dragon or a serpent, identified
with Ahi. We cannot account etymologically for the exis-
tence of a neuter noun vrtra, unless we trace it back to an
abstract idea, just as mitra was no doubt 'compact' per-
sonified, and varum was 'the act of covering' or perhaps
'the act of binding together'. These abstract ideas lie be-
hind many instances of hypostatization : the idea of evil is
never clearly personified as a major demon, but is repre-
sented under the multiple forms of 'hostility', 'violence',
'resistance'. To translate these words as personal names or
agent nouns is to do violence to Vedic terminology. Oppos-
ing the idea of evil or primitive anarchy there are numina
who represent the powers of order in their many forms,
both static like the rta or the dhdman, and dynamic like the
indriya, the tejas and the vdja. These forces, which regulate
relationships in the supernatural world rather as varndsrama
regulates human social relationships in classical times, are
endowed with life in the myths. We have already seen the
importance of rta; all that is and all that is to be depends
ultimately on it. It is the Vedic precursor of the idea later
called dharma. There is an opposition between rta and
anrta, disorder or falsehood, on the ethical plane, and be-
tween rta and nirrti, dissolution, on the cosmological plane.
The power of the gods is limited by the interplay of these
20 VEDISM: I

forces, just as later it is limited by karman or maya. Some

Vedic writers already feel that the stories told in the myths
belong to the realm of maya; the Satapatha says that Indra
never really fought; his very existence is sometimes called
into question. India has never believed unreservedly in her
own fictions.
These cosmic powers, precursors of the saktis, do not
constitute a system of clear-cut oppositions. In classical
times Siva, the terrible destroyer, could also be a kindly
protector; similarly in the Vedic system, vast spheres of
activity are controlled by ambivalent powers. A normally
well-disposed divinity may take on a ghord tanti, an awful
aspect; Varuna is alarmingly liable to assume the aspect of
Vrtra. Any being who is overshadowed, forced to yield his
position to a newer god, or who is relegated to the role of a
father, is apt to become baleful. It sometimes happens that
malevolent, demon-like beings, such as Pipru or Namuci,
have a well-disposed counterpart. Sometimes the ambiva-
lence is an integral feature of the divinity, as in the case of
Rudra, who is, in this respect only, the prefiguration of
Siva: the two figures have nothing else in common. Terms
like manyu, ari, mdyd,yaksa and so on have two sets of mean-
ings, according to whether they are used of good or evil
beings. Thus in the Avesta there are two separate series of
epithets, one applicable to the daeva world and one to the
dhura world; but in the Veda the series are no longer
separate. The background of the hymns is a troubled one,
a scene of passionate rivalries and internal struggles, where
great dangers have been faced and surmounted; the aban-
donment of the surd, the establishment of universal alle-
giance to Indra by gods and men alike, the eclipse of
Varuna, the acceptance of the Asvins, the advent of Rudra:
none of these events could have been accomplished with-
out great upheavals.
Another illustration of this ambivalence is the associa-
 VEDISM: i 21
tion of divinities in pairs. This custom did not survive into
subsequent stages of Indian religion : the later Harihara
is a subordinate figure, lacking consistent treatment. In the
Vedic period we find combinations of widely-varying
elements, Dyàvàprthivì, Mitràvarunau, IndràgnI, Indrà-
visnù and so on. These conjunctions do not always reflect the
liturgical arrangement, and it would be interesting to find
out by what principles they are governed. M. Dumézil has
rightly pointed out, 1 developing Bergaigne's theory,2 that
Varuna and Mitra represented two complementary aspects
of the sovereign power, one magical and terrible in char-
acter, the other juridical and benign. In the case of the
Asvins, the divine pair form a single entity in which it is
practically impossible to pick out the component elements.
The fact that Visnu and Siva later share in dominating the
Indian religious scene may possibly have some foundation
in memories of these old Vedic associations.
We have already noticed the remarkable predominance
of Indra, who soon eclipses most of the other divinities; the
voluntary or forced withdrawal of Varuna before his
youthful rival can be gathered from the text itself. Vedic
myth becomes 'Indraized'; the cult, too, as in the Ràjasùya,
begins to undergo the same transformation. In the other
figures of prime importance, such as Agni and Soma, the
ritual elements of which they are personifications can still
be clearly discerned. No authentic mythical episodes are
associated with them; the only legend that might be sug-
gested in this connection, that of the theft of the soma by
the Gandharva, is derived from an Indo-European story of
the abduction of a liquor that gave immortality, in which
soma has been substituted for the ambrosia of the legend.
The Veda would have been entirely different in form if
the battle-myths connected with Indra had not been intro-
1
In his Mitra-Varuna.
2
In the Religion védique, especially the beginning of vol. iii.
22 VEDISM: I

duced into it; these myths sometimes have a quite un-

Indian ring about them and contain tribal names which
cannot belong to the Aryan onomastic. The Indra of the
Vedas absorbed the substance of other divinities into him-
self, just as the Krsna of a later period absorbed Visnu, or
as the Goddess reduced Siva to the status of homunculus.
Indra is a hero of ancient times and retains the appearance
and characteristics of a hero. Apart from Krsna, he is the
only Indian god who ever had a childhood, and whose per-
sonality and actions betray human elements. Gods gain in
importance not by their virtues, but by the extent of the
mythology they inspire. If the influence of the Buddha and
Mahavlra, as founders of sects in ancient India, has been
too highly assessed, it is because they very early underwent
a kind of mythical apotheosis.
 II
Vedism: II

Ihaveling
N richness of mythological invention and assured hand-
of mythical themes, the Rgveda was destined to
no successor: Vedism is a mythology that is broken
off abruptly. The Atharvaveda, a collection intended for
domestic use and for the performance of magical rites, and
as a hymnary with an esoteric cosmogony, either mini-
mizes the importance of the gods or leaves them altogether
out of account. The Atharvan 'reform' is comparable in
some respects with that of Zarathustra in Iran. The
divinities have become merely decorative in function; the
activities they preside over are ill-defined; the part they
play is sometimes ludicrous. Indra is a shadowy figure of
magic; Varuna loses his virility. The only mention of most
of the gods is a more or less distorted version of something
that has come down from the Rgveda. Mythopoeic activity
ceases when the mind turns to magic, for magic establishes
a direct contact between the performer and the effect he
desires to produce. On the other hand, in the Atharvaveda
there is a renewed appeal to demoniac forces, whose power,
as we have seen, tended to be restricted in the Rgveda.
The Bràhmanas show a metamorphosis of a different
kind. They certainly create fresh myths, or rather the
beginnings of myths, for most of them terminate abruptly.
The story of Manu's surviving the deluge was designed to
enhance the efficacy of the milk offering; the offering is
Manu's daughter and through her Manu engendered the
human race. The significance of all this is far from clear.
24 VEDISM: II
The ancient rivalries, which are used to such rich effect in
the Veda, are now represented by monotonous struggles
between Devas and Asuras. Supreme power is now vested
in a new figure, namely Prajàpati, who appears as an un-
important tribal chief in a few passages of the Rgveda. But
like all these new gods, Prajàpati is a colourless figure,
devoid of legend. His creative or procreative functions
drain him of power. All the gods of the Brdhmanas are more
or less exhausted by their functions and tend to lose their
virility; their strength is spent, like that of a hunted-down
animal. But Prajàpati is something more, or rather, some-
thing less : the name is extended to include the anirukta, the
symbol of the non-defined, the non-determined; his real
name is Ka, 'Who?5, and the choice of the number that
expresses him, 17, has no rational ground. He represents
all that is undefined, whatever in the divine sphere is left
unexpressed by the series of the recognized divinities.
Under another aspect, he is the Sacrifice, and, again, the
function of the Sacrifice is to bring together all unco-
ordinated phenomena and build them up into an organic
whole (even though this structure may be only transitory),
and make of the sarvam a visvam.
It would be unjustifiable to assume that the period of the
Rgveda with its wealth of mythology was followed by a
mechanistic and virtually atheist period. It would be more
accurate to say that we are henceforth moving on a differ-
ent plane of thought; a similar contrast will be observable
later on, between the prolific legendary material of the
Visnupurdna and the abstract way in which Sankara or
Kumàrila represents the world of the gods.
In the Upanisads, the process of eliminating the gods is
complete. Mythology is now conceived as a setting for
apologues. Thus, in the Chdndogya, Indra is a Brahman
student committed to a novitiate lasting 101 years. As we
have already seen, the Upanisadic writers attach the
 VEDISM: II 25
greatest importance to the system of correspondences,
which are not conceived as capable of being externalized
in forceful imagery, as in the Veda, but rather as tending
to be reabsorbed into the impersonal, abstract principle
of brahman, the latent energy underlying the old enig-
matic formulae. The essence of the Upanisads is mimàmsà,
reflection, as opposed to the intuitive quality of the hymns
and the practical-minded elaborations of the Brdhmanas. In
this way the Veda comes full circle and epitomizes the
whole course of the evolution of Indian thought.
Some of the Vedic poets long to penetrate the mystery of
the ultimate origin of things, and to find out the nature of
the supreme reality that lies beyond the world of the gods,
for the gods do not help us to perceive it—indeed, they
conceal it from us. In its early stages, this kind of specula-
tion draws extensively upon mythology; it might perhaps
never have come into being without the stream of images
offered by the myths; but in the end it parts company with
mythopbeic thought. Cosmogonie themes are not easily
combined with legends of the gods. The Purusa, or primi-
tive giant, really only creates the social structure. Specu-
lative thought at this period does not envisage time and
space on a vast scale, as the classical period does. The word
Tuga, which later designates the cosmic eras, is used only
as a term in dice-playing.
The creation is an emanation, a procreation, the act of
an artificer or an artist, a sacrifice, a thought: all possi-
bilities are admitted, but none is finally confirmed. The
questions of who made the world and the human body are
constantly recurring. More and more new terms are sug-
gested to designate the hidden principle of all things, the
turlyam padam, the guhyà nàmdni. These speculations are
assiduously pursued in the Atharvaveda; it is as though
magical methods were thought to be more appropriate to
the subject than the usual ceremonies of praise; Time, the
 26 VEDISM: II
Breath, the Cow, the Cosmic Support (Skambka), the
Golden Embryo, the Viràj and many other things are put
forward as suggestions. Less ingenious minds are content to
propose the neuter One, or Ekam\ the plurality of appear-
ances, they say, is due to the action of visrsti, the principle
of individual creation.
Sometimes the full possibilities of a speculative theory of
the Veda become apparent only in post-Vedic times. M.
Filliozat1 has drawn attention to a Vedic theory that
identified breath and wind; the various physiological pro-
cesses were expressed in terms of the process of breathing,
conceived in the likeness of the wind passing through space.
In their cosmic aspect, these ideas are of Indo-Iranian
provenance : we know that the Iranian Vayu was in some
respects a Supreme Being. These speculations were echoed
in classical times, and influenced medical theory. There is
an underlying thread of ancient Indian pneumatology
running through Toga, with its prdndydma, and Tantrism,
with its upward surge of the kundalinl and its theory of the
bodily 'channels'. The equivalence between the dtman and
the universal soul was accepted the more easily because
originally, as etymology and Rgvedic usage show, the
term dtman connoted breath in its association with wind :
dtmd te vdtah, 'thy breath is the wind', says a well-known
passage of the Rgveda, referring to Varuna.
Vedic speculation was the work of a small group of
daring poets. Garbe,2 among others, believed that a spirit
of revolt against the priests, or at any rate against ritual-
ism, could be discerned in the Upanisads; but the Upanisads
1
Jean Filliozat, 'La force organique et la force cosmique dans la
philosophic medicale de l'Inde et dans le Veda', Revue Philosophique,
116 (1933), p. 410; La doctrine classique de la médecine indienne: ses origines
et sesparalièles grecs (Paris, 1949), p. 51.
2
R. Garbe, 'Die Weisheit des Brahmanen oder des Kriegers?' (an
article of 1893 reprinted in Beitrdge zur indischen Kulturgeschichte (Berlin,
1903), chap. 1.
 VEDISM: II 27
are in effect only supplements to the Brdhmanas, intended
for advanced pupils who wanted something beyond the
formal course of instruction. That is why the teaching of
the Upanisads starts at the point where exegesis stops: the
Chdndogya takes Bràhmanical meditations on melody as its
starting-point, the Brhaddranyaka develops the theme of the
mystic meaning ofthe Horse Sacrifice, which was dealt with
in the concluding paragraphs of the Satapatha-Brdhmana.
They answered the needs of the ascetics and anchorites,
for whom a religion of idols and cult-practice was not
enough. There is no more opposition between Upanisad and
Brdhmana than there is between thefirstand second Mlmdmsd
or between Sdmkhya and Toga.
But let us look once more at the problems of everyday
life. If we ask what were the desires of those people of the
Vedic age who were neither kavis nor vipras, the ordinary
cult-worshippers, and the patrons who commissioned the
sacrifices, both rich and poor, the answer is that they
wanted purely material blessings: money, cattle, sons,
good health, success in the arts of war and peace, and the
full Vedic span of life, which was a hundred years. This
utilitarian view of life is like that of primitive Rome: tubus
ut valeant, said Cato. There is a great contrast between this
materialistic outlook and the heights to which contem-
porary speculation sometimes rose. But in even the most
advanced Upanisads puns, magical formulae and instruc-
tion in eugenics are introduced without the slightest pre-
liminary. In short, the public for whom the hymns were
written is not at all preoccupied with the hereafter. There
can be no greater blessing than never to die; not to escape
from rebirth, which was to be the desire of classical India,
not even to escape from punarmrtyu, as the Brdhmanas say,
but simply to prolong the present life. Vedic ethics are
based on this naive aspiration, and are mainly concerned
with length of days.
28 VEDISM: II

Conceptions of the after-life are no less rudimentary.

Heaven, which lies at the highest point of the firmament,
consists of material pleasure; it is a paradise of light; it is
more beautiful than the earth because, we are told, there
are more nights there. It is sometimes spoken of with a kind
of intoxication, as in the blessed visions of Indra, when he
is exalted by the soma. Sometimes the other world was con-
ceived of as the realm of Yama, lying beneath the earth.
But this conception soon fell into disfavour because of the
natural aversion in which all things subterranean are held
and the fact that, although the Veda gives no clear idea of
the infernal regions, they are thought of as a black hole.
Hence Yama, formerly the ruler of paradise, is abruptly
transformed into the king of the infernal world. It is clear,
in short, that the Veda offers only the slightest of prece-
dents for the preoccupation with eschatology that charac-
terizes Hinduism, Buddhism and Jainism.
The same may be said of the theories of transmigration,
samsdra, and the future retribution of earthly actions, kar-
man. There is no clear trace of the former theory; and the
latter theory appears only in the primitive conception that
good deeds are rewarded by a life of bliss, the sukrta or the
sukrtasya lokah, evil deeds by a life of torment, duskrta or
duryona. Sylvain Levi, with some degree of exaggeration,
described the attitude of the Brdhmanas as amoral. 1 For
these writers, the essential is to perform the prescribed
action, the kriyd or karman, with scrupulous exactitude.
Sraddhd is in effect only the confidence that one has in the
efficacy of the action; the old term rta, with its wide range
of associations, has been replaced by satya, which means
exactitude. There are isolated passages in the Upanisads in
which the word karman is used in the sense of a good or bad
action on the moral plane; but it is never used for the pre-
sent effect of a past action or the foreseeable consequence
1
La doctrine du sacrifice dans les Brdhmanas (Paris, 1898), esp. p. 9.
 VEDISM: II 29
of an action performed in the present, conceptions which
constitute the essential meaning of the word in later usage.
Vedic determinism is confined to ritual. Even the fre-
quently cited passage, BAU. I l l , 2, 13, does not imply that
the writer was acquainted with the principle of systematic
retribution.
But Vedic religion is first and foremost a liturgy, and
only secondarily a mythological or speculative system; we
must therefore investigate it as a liturgy. In its principal
treatises we have a collection unparalleled in antiquity;
they describe every operation, every gesture of the Vedic
ritual appropriate to each ceremony, each officiant and
each school. I cannot here go into the interesting question
of the schools, for that would be to embark on problems
that still await solution. A philological comparison of all
the sources would probably enable us to reconstruct a
primitive form of ritual something like that prevailing in
the Rgvedic period; at present we only have the separate
recensions of groups that date from different periods and
that have long been widely dispersed.
The Vedic rites are made to conform to a systematic
arrangement; mythology may be lacking in system, but
ritual is overburdened with it. It appears that originally
separate rites were grouped together in vast systems in
response to new demands that had arisen in the course of
time, and under the influence of an advancing scholasti-
cism. There is a distinction between the great public rites,
called Srauta, and the domestic rites, called Grhya. The
former are carried out by professional officiants, and need
three fires; the formulary is taken from the Samhitd. The
domestic rites take place on the family hearth, and are per-
formed by the householder, using a formulary taken from
a special collection. The two series are entirely different in
character, in spite of the resemblances that arise from
borrowings. The Indians, with their taste for classification,
30 VEDISM: II
divide the solemn rites into seven samsthds or ordinary
celebrations, with vegetable and animal offerings, and
seven others, based on the soma oblations. But the soma
sacrifices necessitate ordinary vegetable and animal obla-
tions, and the Sautrdmanl involves milk, surd and a sacri-
ficial victim. The tendency to build up complex structures
from simpler elements is illustrated by the fact that some
sacrifices, all using a common tantra, are variations of a
single archetype. Some festivals, such as the sattras or
'sessions', are too complicated to be actually carried out,
and are intended rather as intellectual exercises. It is clear
that the texts contain a proportion of such exercises; we
must not regard them as consisting entirely of accounts of
actual religious practice.
I do not intend to engage in a theoretical consideration
of the nature of the ritual. Ritual has a strong attraction for
the Indian mind, which tends to see everything in terms of
formulae and methods of procedure, even when such
adjuncts no longer seem really necessary for its religious
experience. When Naciketas, the young Brahman, who
has visited the realm of the dead, tells the god Yama of his
desire to know whether or not man lives on after death,
the god's only resource is to advise him to set up a Tire'.
All the great mysteries are revealed to the man who knows
how to pile up the bricks of the Fire-Altar and make the
right offerings on it. There are no personal prayers, uttered
spontaneously, in either the Vedic or the Classical periods.
The ceremonies of religion take place on the days assigned
to them in the calendar, daily, fortnightly, or at certain
seasons. There is provision for votive rites, Kdmyestis, to
obtain special favours, but they are only the ordinary pro-
cedures with the addition of a votive formula. In the same
way an operation can be adapted, with only slight changes,
for use as an act of expiation.
The most imposing ceremonies, which may also be the
 VEDISM: II 3i
oldest, the Ràjasùya, the Asvamedha and the Vdjapeya, are
reserved for princes: they are rare and costly occasions,
which were a pretext for lavish celebrations, like pot-
latches in character, which are carefully recorded in
classical inscriptions. It is proposed to reconstruct one of
these great sacrifices, the Vdjapeya, at Poona in the near
future; this should be an event of great interest to Vedic
scholars.
There are various types of ritual, then, designed for
different purposes. The most solemn forms are preceded
by a characteristic ceremony, the consecration or dtksd;
this may be based on the private ceremonial that takes
place at the beginning of a Brahman student's career, for
it has many analogies with it. Both ceremonies have the
same object: to raise the participant from the sphere of the
profane to that of the sacred (a process ofdurohana, painful
ascent) by freeing him from worldly vices, as Mauss said.1
At the end of the ceremonial, the reverse process, ava-
bhrtha, or descent from the sacred sphere, takes place; and
the consecrated objects are carried away by flowing water.
This practice thus centres round the patron of the sacri-
fice, the layman who pays for the ceremony and receives
its spiritual benefits. It is interesting to note that the lay-
man himself, with his wife, takes his place among the
officiants, though his is not a very active role; he has no
hieratic formulary at his disposal, as the professionals have.
But his presence is essential; he recites some of the prayers,
and may even replace one of the officiants on occasion;
Yàjnavalkya was led to protest at his encroachments at one
stage. It must not be forgotten that at all periods the ser-
vice of the temples, or at any rate their administration,
could be entrusted to the laymen, and that the Smrti in-
cludes among the rights of the ksatriya and even of the
1
H. Hubert and M. Mauss, 'Essai sur la nature et la fonction du
sacrifice', Année Sociologique, 2 (1898), p. 52.
 32 VEDISM: II

vaisya that of 'sacrificing for his own benefit'. The wife

plays a very subordinate part, but all the same she is there,
a silent participant who occasionally has an action to per-
form, as when, in the Vajapeya, the husband and wife climb
on to the post that is surmounted by the solar wheel. This
may be a reflection of the low esteem in which woman was
held in Vedic India, as some of Yàjnavalkya's remarks
seem to suggest. Women are often introduced in licentious
roles, as in the Horse Sacrifice, when the queen has to lie
beside the slaughtered horse, and in the Varunapraghdsas,
when the priest asks her, 'How many lovers have you?' and
she has to answer by holding up as many blades of grass as
the number of lovers she admits to having. There is a fore-
shadowing of the drama in this scene, as there is elsewhere
in the Vedic ritual and hymns, especially the famous dia-
logues of the Rgveda which accompany or replace some
kind of passage in mime. There is no doubt but that the
cult contained an element of the drama : in classical times
pilgrimages were enlivened by portrayals of the deeds of
Rama and Krsna, and the Bengali jw/raj include plays and
farces.
The number of officiants varies: the main responsibility
rests on the adhvaryu, who is in charge of 'ways and means'
and who performs various actions, moves about and recites
at considerable length in a low tone. The hotr, who, as the
etymology of the word suggests, was originally the libation-
pourer, later becomes primarily a reciter; but his invoca-
tions, which are said aloud, impressive though they are,
play only a small part in the liturgy as a whole, rather like
the music of the chanters. The brahman, who, as his name
reminds us, is the repository of the unexpressed power of
the formula, is a silent spectator, whose duty it is to see that
the operation is carried out with accuracy; he is a pro-
fessional expert, like the Roman priest. His silence is just as
valuable as the speech and melodies of his colleagues.
 VEDISM: II 33
The complex system of sixteen or seventeen officiants
was a later development of the simpler form that appears
in the Rgveda; here eight names are mentioned in all, those
of seven officiants and theyajamàna, then called the 'house-
holder'. The functions assumed by these seven officiants,
headed by the hotr, connected each of them with one par-
ticular god; this was known as rtu, in the old sense of the
word, the connection between sacerdotal function and
tutelary deity: the kotr and the brahman were associated
with Indra, the adhvaryu (or both adhvaryus) with the
Asvins, the prasàstr with Varuna, the agnidh with Agni, the
nestr with Tvastr, the potr with the Maruts.
The Rgveda mentions a presser, who took part in the less
complex festivals. This function could be performed by the
layman, for at this early period the pressing of soma was
also carried out for domestic purposes, with a pestle and
mortar. The growth of the soma cult may bear some rela-
tionship to the expansion of Indra's role. In that case, the
oblations that involved bloodshed may have been part of
the cult of Varuna, the supreme and terrible god : they
were introduced at intervals in the somic ceremony. Fin-
ally, of the purely vegetable oblations, some were offered
to the gods who protected the harvest, like Pùsan, some to
those who watched over contracts, like Mitra, some to yet
other gods.
It must also be remembered that there were no temples
at the Vedic period: 'the sacrifice takes place within the
officiants themselves', says one of the Brahmanas. The term
ayatana, which later came to mean 'sanctuary', merely
designates the ordinary domestic hearth in Vedic times.
The temple cult of the classical period must have grown
out of the domestic cult. Sacrifices took place on a specially
prepared piece of ground, but the same spot was not neces-
sarily used again for subsequent ceremonies. There was no
building other than temporary huts, constructed on a
 34 VEDISM: II
framework of poles joined together at the top by transverse
beams, and roofed over with thatching. The instruments
used were also rather rudimentary, though their functions
were highly specialized : there was a whole set of spoons,
cauldrons and other receptacles, and for the kindling of
fire the ancient method of the tourniquet-wheel was still
used; there were peculiar shards, pieces of brick or earthen-
ware perhaps, arranged in the shape of a horse-shoe, on
which the paste was placed to cook over the embers. The
centre of the sacred ground is called the vedi: it was some-
times a raised mound, but more often a pit made to receive
the oblations and the various instruments, thus expressing
the invitation to the gods in concrete form.
It is impossible to describe the procedure in detail.
Several scholars have given general descriptions of it,
notably A. B. Keith. 1 The individual features of the ritual
could only be explained by invoking the same principle of
esotericism by which we are guided in any attempt at inter-
preting the mythical narratives of the Veda. In any case,
one cannot grasp even the outward meaning from reading
the text by itself unless one is gifted with the rare virtuosity
of a Caland. It would be necessary to follow several scores,
as it were, at the same time, and to bear in mind the vari-
ous formularies from which the recitations and chants are
drawn, seemingly at random. The best way to understand
the nature of these ceremonies is to be a spectator at one of
them as they are still enacted to this day throughout India
—in a spirit of historical reconstruction rather than from
religious motives, I should imagine, though one can never
be sure.
I recently had the opportunity of seeing one of these
1
Cf. The Religion and Philosophy of the Veda (op. cit.). Also A. Hille-
brandt, Ritual-Litteratur: Vedisene Opfer und Zauber (Strassburg, 1897).
A more recent account is that of P. V. Kane, History of Dharmasdstra
(Poona, 1930 etc.), esp. vol. 2, part 2 (1941).
 VEDISM: II 35
ceremonies myself: it was a fairly simple one, a DarsapUr-
namdsa performed according to the Hairanyakesa recension.
It took place at Poona, under the supervision of a Vedic
Institute. It was a convincing, and, as far as one could tell,
a faithful reproduction, although there was some inevit-
able simplification and several deviations, of which some
compared more favourably than others with the ancient
model. The age ofthe paddhatis and prayogas had obviously
intervened. One did, however, derive from it an under-
standing of the nature and the raison d'etre of the various
prescriptions, movements and gestures. This type of cere-
mony is much more spectacular than the classical cult with
its mumbled prayers and its noisy offering of oblations
round the idol. In those distant days India had a feeling
for liturgy comparable to that of the Roman Church.
The significance of certain movements prescribed in the
ritual depends on their connection with other actions.
Those that constitute the introduction to the oblation are
of this kind; a serpentine advance, or prasarpana, winding
in and out to the accompaniment of invocations, is a char-
acteristic feature of the processional entry to the sadas: the
officiants proceed in single file, each one holding the shoul-
ders of the man in front; they walk bent forward, sliding
their feet and licking their lips. This is because it is the
morning pressing that is taking place, and the sacrifice has
yet to increase. At the midday pressing, they walk upright,
but still with their heads bowed; only in the evening are
their heads raised.
The ritual is devised not only to dignify and strengthen
the corresponding secular procedures, but also as the sym-
bolic expression of speculative theory. Perhaps we shall
never know for certain whether speculative thought dic-
tated the form of the ritual, or whether, as* seems generally
more likely, speculation was a later development. But the
fact remains that in many cases an action is closely bound
 36 VEDISM: I I
u p with its esoteric significance. For the Cam oblation, the
officiant hollows out a depression in the middle of the paste,
pours liquid butter into it and looks for his reflection in it,
as in a mirror. If his reflection does not appear, he repeats
the process of pouring in butter. If the reflection still fails
to appear, it is a sign that his life is drawing to a close. This
appears to be a magical rite that has found its way into the
ceremony; it is sometimes used to find out whether a sick
man will recover or not. But as Vedic religion is con-
sciously optimist and maintains that no action is inexpi-
able, the danger can be averted by anointing the eyes with
the butter, and by beseeching King Soma to whom the
cake is dedicated: 'It is thou who touchest the heart (i.e.
who healest it by thy touch) ; it is thou who givest a healthy
colour; give me a healthy colour!' Such episodes are much
more interesting than the endless sequence of oblations,
punctuated by the harsh shout of 'svàhàP when the liquid
jet touches the flame and makes it leap up.
Other rites are really independent scenes which have
become associated with the ceremony, such as the dice-
playing scene at the beginning of the Royal Consecration,
the race of seventeen chariots in the Vdjapeya, the selling of
soma at the beginning of the Agnistoma, and the occasional
introduction of profane music. The profane is very close to
the sacred in many of these scenes, as even a very super-
ficial knowledge of the society and everyday life of the time
is enough to show us. The buying of soma is obviously an
imitation of a custom obtaining in commercial practice.
T h e adhvaryu arranges the plants on a red bullock-skin : he
then hands them to the make-believe merchant: 'Seller of
soma, have you any soma to sell?' 'Yes.' 'From Mount
Mujavant?' (From the Rgveda onwards, it was considered
that the best soma grew on Mount Mùjavant.) 'Yes.' 'I will
buy some from you for the upper part of this cow's foot—
for half its foot—for the whole foot. Well, then, for a quar-
 VEDISM: II 37
ter of the animal.' But the merchant is still not satisfied,
and says, 'What else will you give in exchange?' 'Here is
gold, here is a goat, here is a garment.' Needless to say,
each of these objects has a symbolic significance. The mer-
chant is punished for his tenacity: there is a pretence of
strangling him and of blinding him; his precious mer-
chandise is taken from him by force; he is driven away with
blows. This would not be surprising if the merchant were a
sudra or a member of the despised family of Kautsas. The
whole scene may have been intended to convey the
illegality of trading in soma.
Certain rites give us an insight into a god's nature. In
the Sdkamedhas there is a funeral ceremony which includes
an offering to Rudra; at the conclusion of the ceremony
the remains of the oblation are put into two baskets which
are hung one at each end of a beam, out of the reach of the
oxen : this is a viaticum offered to Rudra to persuade him
to go away without doing harm. M. Dumézil has drawn
attention to an episode in the Royal Consecration, * in which,
according to the commentaries at least, the new king loses
his virility at a decisive moment in the ceremony; he must
regain it by precipitating himself on a herd of cows belong-
ing to one of his near relations. In this way Varuna himself
once lost his virility when he was consecrated king. This
episode may be compared with the well-known story of
how Ouranos, who persecuted his own family, was cas-
trated by one of them. Admittedly, the analogy is not
exact, but there are many parallels between the two
episodes which provide some foundation for a comparison.
Other episodes are more closely connected with specu-
lative thought. Among these is the pravargya, with its
oblationary vessel called mahdvira, 'the great man', which
is worshipped, and considered to be the 'head of the
sacrifice'. It is associated with the demon Makha, the
1
Ouranos-Varuna (op. cit.), p. 72.
38 V E D I S M : 11
malevolent counterpart of Visnu, who was beheaded by
the gods. Here, as on other occasions, it is not fortuitous
that ritual emphasizes that the gods can be made to suffer:
the imagination looses divine forces, but human action can
reabsorb or control them. The Sautrdmanl is more elabor-
ate. Rònnow believed that the myth of the demon
Namuci could be traced in the general outlines of this
ceremony1 (and tradition supports the theory); the Rgveda
has not much to say of Namuci, but later texts tell us that
he concluded an imprudent pact with Indra, according to
which he should not suffer death under any foreseeable
circumstances, and that Indra beheads him in the half-
light with the foam of the waves. But it is not admissible
that the associated liturgy could have been similarly in-
spired, for it is on entirely different lines.
The Fire-Altar in particular, a brick construction in the
shape of a bird, has been the subject of a great deal of
speculation. It was ancillary to many other ceremonies;
the concluding stages of cult ceremonial centred round it;
it was the apogee of the religious drama. It is Prajàpati
who is sacrificed and dismembered, as was the primeval
man, the Purusa, in the Rgveda. The god is then restored in
the form of the altar, which is the representation in space
(as M. Mus observed2) of an abstract religious conception
of duration. As Prajàpati is in fact the sacrifice, the altar
also symbolizes the combining of scattered and anarchical
elements into the organic structural norm which will
henceforth ensure the ordered functioning of the universe.
As these examples show, the texture of Vedic ritual is
very rich. Elemental forces play as great a part in it as
abstraction and speculation. We can see in it the reflection,
1
*Zur Erklàrung des Pravargya, des Agnicayana und der Sautrà-
mam', Monde Orient., 23, p. 113.
2
'Le Barabudur', Bull. Éc.frang. Extr.-Orient, 32 ff. (1932 etc.), intro-
duction, passim.
 VEDISM: II 39
however distorted, of a society, with its customs and amuse-
ments, its court life and its humble callings, and its back-
ground of husbandry and stock-rearing following their
seasonal rhythm.
The domestic rites are often more closely related to
everyday life, since they are enacted inside the hpuse, with-
out any public audience or ceremonial setting. They are of
two kinds. Some are short daily practices, formulae accom-
panying some simplified form of oblation. The other rites
are more characteristic; taken together they constitute the
sarnskdras, that is, the dedications by which the individual
gradually approaches the state of the * twice-born5, reaches
it, and is confirmed in the privilege. In this way all the
most important phases of human existence are covered,
from birth and before until death and beyond. The obla-
tion plays only a secondary role, or is omitted altogether.
On the other hand, the idea of impurity, which is only
partially attested in the public ritual, appears very clearly;
this is another respect in which the domestic rite shows
itself to be the forerunner of the religious practices and
conceptions of classical India.
But, as has already been pointed out, the Vedic element
in these rites is very slight. The formulae alone give evi-
dence of hieratic inspiration, and even then, those bor-
rowed from the Rgveda stand out clearly as interpolations
at various points. This religion is really Hinduism, and
even at times an anticipation of Tantrism. Part of the
ceremonial dedicated to the Manes was incorporated into
the public ritual, among the four-monthly offerings or
among those of the new moon and the full moon; but some
of the practices can also be carried out at the 'domestic*
level. The fact that in the Rgveda, and to a greater extent
in the Atharva, there are wedding and funeral prayers that
have slipped in under cover of praise of the gods, shows
that those who composed the hymns were not indifferent to
40 VEDISM: II

the events of domestic life : it is not impossible that there

existed a small domestic Samhitd which may have been
later absorbed into the composite collection that consti-
tutes the tradition of the Atharvàngiras.
There are many forms of initiation : forms for the dedi-
cation of the young Brahman student, for marriage, and
even for funeral practices. They too, apparently, are
derived from the public ritual, which has influenced the
actions and the form of words: this is a process which
might be termed the 'Vedicization' of private ritual. O n
the other hand, all the agricultural rites seem to have been
private in origin. But we cannot come to any definite
decision on such problems when we are dealing with an
evolutionary process of which we know only the most
recent stages.
Expiatory practices play a considerable part. We come
across them scattered throughout the treatises on solemn
ceremonial practice as well as the others. They inspired a
new type of literature, also superficially Vedic in form, the
Dharmasiitras, in which the theory of expiation is expounded
together with a rudimentary form of law teaching. It was
in this way that religious conceptions achieved a position
among the rules which were being laboriously evolved to
define the legal status of the citizen and his duties in
society.
Lastly, there is the magical element. Ritual in general
abounds in operations based on magic, and it is well known
that there is no radical separation between the two realms.
But normal cult-practice follows the lines already laid
down for it by rigorous tradition. Magic seems to be in-
vented specially for each case as it arises. It is free of any
connection with the priesthood and with society and its
organization: it is a transaction that takes place between
the man who seeks its help for his own purposes and the
sorcerer, who has no official title, but is accorded the name
 VEDISM: II 41
ofpurohita when his client is a prince and he is acting in an
official capacity. The procedure of magic is based on more
or less arbitrary and sometimes esoteric prescriptions for
obtaining possession of something or for compelling some
result to happen. The formula fits the action more exactly
than in any other field; the action translates the formula
and gives it life. Imitative practices are very common. To
change the course of a river, reeds are planted and a frog
with green stripes is placed on the ground; as Oldenberg
said,1 the method is to make a simulacrum of the life of the
river-bed along the course that one wishes the river to
take, and the reality will follow.
It is curious that magic, like private cult-practice, finds
its way into the public ceremonial. Part of one of the
Vedas, the Atharvaveda, is devoted to it; it is true that this is
a late work that is not accepted as authentic in its entirety
and does not properly conform to ancient hieraticism.
Another Veda, the Sdmaveda, gave rise to magical prac-
tices by reason of the incantatory character that was early
attributed to the sdmans. Magic also left its mark in the
Brdhmanas, one of which, the Sdmavidhdnay is given up to it.
In this way, by a sort of agreed fiction, a large number of
magical practices are included in the description of the new
moon and full moon sacrifices. Sometimes a minute alter-
ation in the liturgy is all that is needed to transform an
ordinary action into abhicdra. The remains of the conse-
crated oblation can be used, for example, to anoint the
object to which the magical action is directed.
These acts, taken as a whole, are anarchical in charac-
ter. Stable order is entirely lacking. Like other forms of
religion in India, Vedism thus becomes debased to the
level of somewhat crude witchcraft. In the Upanisads,
ritualism goes to the opposite extreme, and is sublimated
into abstract speculation: but in the process there is the
1
Die Religion des Veda (op. cit.), p. 507.
42 VEDISM: II

risk that it will be destroyed or laughed to scorn, as in the

passage where the Mundaka waxes ironical on the subject
of those Trail craft5, the eighteen forms of the sacrifice. But
the Upanisads themselves, or at any rate the two earliest
ones, contain a certain amount of crude and childish
material, as we have already seen. Extremes meet in India.
It is difficult to sum up the general impression that the
Veda makes on us; in many respects it is a strange and
even monstrous testimony; the utmost caution must be
exercised in using it as material for any subject not directly
envisaged by it, whether it be linguistics, sociology, the
history of religion or ethnography. We must approach it
dispassionately, putting aside all the absurd and untenable
theses to which many researchers, in the West as well as in
the East, have subscribed. Again, we must always bear in
mind the pitfalls into which we should be led by blindly
following tradition, which, venerable though it may be,
has now, it must be admitted, largely lost its feeling for the
living reality represented by the old texts, and supplies the
deficiency by a kind of intuitional approach based on
memories of a much more recent period. We must not for-
get that there is a gap of twenty-five or thirty centuries
between Sàyana and the hymns, and we are conscious that
the tradition that connects them is an interrupted one.
The importance of the Veda to India is well known. Its
imprint on Hinduism is permanent and unmistakable; and
on Buddhism and Jainism, too, it has left a deep impres-
sion, if only in the reaction it produced in them. It seems
likely that many Indian literary disciplines would have
developed quite differently if there had not originally been
that striking sequence of hymns, commentaries, descrip-
tive aphorisms and philosophoumena, which were drawn
upon and imitated over so long a period. Vedism is a
religion, but it is even more a technique; a technique of
learned poets and erudite theologians, which has given rise
 VEDISM: II 43
to the most atheistic of the philosophical systems of Brah-
manical India, the Mimdmsd; the Mimdmsd only develops
the thought of the Brdhmanas; it is the jurisprudence of the
ritual act. The definitions, logical argument and interpre-
tative maxims elaborated by the Mimdmsd for students of
the old public ceremonial were so highly considered that
they were taken over into other fields^ notably the Dhar-
masdstra. Thus, throughout the Middle Ages, we see the
legal commentaries floundering on rules that had been
drawn up to meet the needs of a sacrifice long fallen into
desuetude.
Vedism even developed the secular disciplines, pho-
netics and grammar, astronomy, the rudiments of law,
even geometry, because its teaching made use of them. In
the Nighantus, probably the oldest lexicon in any Indo-Euro-
pean language, words are grouped as series of synonyms.
These synonyms are mostly secondary, metaphorical ac-
ceptations of the words, in accordance with the laws of
occult correlations and equivalences of which we have
already spoken; and it is this type of symbolical synony-
mies, as it might be called, that reappears in the lexica of
the classical period, in the commentaries and in the very
phraseology of many of the kdvyas.
Let us consider a few instances of Vedic survivals, chosen
more or less at random. It is claimed that the Pancardtra
system is founded on the Veda, that it belongs to a school
of the Tajurveda, the Ekdyana; the school is unknown to us,
but it is possible that there is some recollection of it in a
passage of the Chdndogya (VII, 1,2). The Vedic Vaikhdnasa
school was continued by a Vaisnavite Samhitd of the same
name, dating from a late period, which preserves a com-
plete fire ritual. The school of medicine evidenced by the
manual known as the Kdsyapa-Samhitd endeavours to repro-
duce the Brahmano, style, which seems to indicate a desire to
emphasize its connection with Vedism. Some of the later
44 VEDISM: II
sects, like the Nàtha Siddhas, which adopt a Tantric Toga,
have re-established a soma cult, though in symbolic rather
than concrete form. A late Buddhist text, the Tattvasam-
graha, gives evidence of a Vedic school, the Nimittasàkhà, of
which nothing has come down to us from any other source.
An earlier text, the VajrasUcl attributed to Asvaghosa, con-
tains quotations from the Veda. A text of Jaina inspiration,
the Tasastilaka, in polemizing against the Brahmanic cult,
gives some curious details about the Veda and the old
sacrifices.
We are quite uninformed about some aspects of Vedism.
Of religious feeling and community life in the Vedic
period we can know virtually nothing. The schools, as we
call them, are known to us only by their recensions. We are
constantly having to make inferences about religious
phenomena from philological evidence.
It would seem that whatever survived of Vedism has
become so integral a part of Indian thought that it is no
longer distinguishable as a separate element. The rest
died out, as did the conception of the god Varuna : today
we have to look to the extreme limit of Indian expansion,
the island of Bali, to find a temple dedicated to him. Yet
the recitation of the hymns is still practised, and I do not
think it is a case of purely formal adherence to an extinct
tradition. It is amazing to see that even to the present day
there are men living in the uttermost parts of India, who
have never met one another, and who yet, in their recita-
tion or chanting of the Veda, carry out the infinitely subtle
processes of a consummate analytical technique with com-
plete accuracy of memory. If the principal texts were lost,
they could be reconstituted, thanks to these men.
India in her exhaustion has often taken refuge in ahimsd
and the Vedàntic scale of values; but a new and more self-
assertive generation may be at hand, a generation imbued
with the spirit of Yàjnavalkya who, when he was asked if
 VEDISM: II 45
he permitted the eating of meat, replied, 'Yes, as long as it
is tender. . . .' The Veda may once again become a great
source of inspiration, as it was to the fiery Dayànanda
Sarasvati in the nineteenth century, who set out to estab-
lish a mystique of national and social import based on the
Vedas.
As a source of academic interest for the scholar, the Veda
is by no means exhausted: but let us in conclusion very
briefly consider its value as literature. In the nineteenth
century it was the fashion to deride the 'galimatias vedique\
as Bergaigne called it; 1 Max Muller said that one could
not read ten pages of Brdhmana without revulsion;2 and
Von Schroeder likened the formulae of the Tajurveda to the
ravings of mental delirium.3 The Upanisads escaped attack
because India's contributions to speculative thought are
always treated with respect. But the prose-style of the
Satapatha is a model of skilful articulation, and in its severe
purity reminds us of Plato. There is a strangeness of expres-
sion in the hymns which we may well find compelling now
that our standards of poetry are more flexible, and our
ideal is an art that combines the primitive with the
elaborate.
1
Esp. in his article 'Quelques observations sur les figures de
rhétorique dans le Rig-Veda', Mém. Soc. Lin., 4, p. 96.
2
Chips from a German Workshop (London, 1867, v o ^ *5 later editions
1868 etc.).
3
Leopold von Schroeder, Indiens Literatur und Cultur in historischer
Entwicklung (Leipzig, 1887), p. X I 3-
 Ill
Hinduism: I

T HE long succession of religious developments which

followed Vedism cannot be easily grouped. It is some-
times proposed to divide them into an older period, which
would be designated as 'brahmanism', during which the
main trend is towards uniformity rather than sectarianism,
and a later period, when sects abound, which would be
that of Hinduism proper. But the sects are certainly older
than literary evidence shows; the Tantric aspect, as it is
called, which after a certain period is characteristic of the
greater part of religious practices, also has origins far back
in the past. The same may be said of bhakti. If we confine
our search to the origins of religious phenomena, we tend
to overlook the fact that India is constantly contributing
new material, or revivifying the old; if, on the other hand,
we concentrate too closely on one particular stage of reli-
gious development, we may err in assuming that the facts
we are studying are without roots in the past. In India
everything is in one sense older, and in another sense of
more recent origin than is generally supposed.
If we wished to attempt a definition and classification of
the essentials of Indian religion we could take as our start-
ing-point religion as it is today, with its multiplicity of
local cults, beliefs and superstitions, and its many village
gods, and try to compare it with what we know of the
ancient religion from literary and archaeological evidence;
and we could then consider it in relationship to forms of
religion outside India. This method would inevitably
 HINDUISM: I 47
result in a collection of miscellaneous features which would
be conveniently termed cnon-Aryan', or perhaps, less
cautiously, 'Dravidian' or 'pre-Aryan', and which would
really be features common to primitive religions all over
the world. What would remain then as a basis for 'classical'
Hinduism? Nothing, apart from those elements emanating
from Vedism; and we must bear in mind that Vedism itself
contains elements of primitive religion, and therefore of
Hinduism (or, we might say, of pre-Hinduism), the exis-
tence of which at a period earlier than the Veda could be
verified by the evidence of the Mohenjo-Daro excavations.
The Vedic contribution to Hinduism, especially to
Hindu cult-practice and speculation, is not a large one;
Vedic influence on mythology is rather stronger, though
here also there has been a profound regeneration. Religious
terminology is almost completely transformed between the
Veda and the Epic or the Purdnas, a. fact which has not
been sufficiently emphasized; the old terms have disap-
peared or have so changed in meaning that they are hardly
recognizable; a new terminology comes into being. Even
in those cases where continuity has been suggested, as for
Rudra-Siva, the differences are really far more striking
than the similarities. It has been claimed, on grounds that
do not seem altogether unreasonable, that Siva is a figure
of Dravidian origin. The same claim has been made,
though less categorically, even for Visnu, who appears in
the hymns as a minor divinity. An Eurasian provenance
for the Great Goddess has been put forward as a possi-
bility (even if she already appears in the Veda under the
name of Aditi), and it has been suggested that Indra owes
his origin to Asia Minor. Similarly, Toga and even
Samkhya have been accounted for in terms of primitive
Asiatic features. It would have been quicker to enumerate
those elements that are demonstrably Aryan : they would
consist of perhaps a few functional gods (as it is the fashion
48 HINDUISM! I
to describe them), the soma cult and the rudiments of a
social system : little enough, in all conscience.
I believe that these theories are exaggerated, and that
they are based on superficial explanations. The empty
terms 'non-Aryan' and 'primitive5 are used too readily; in
seeking to prove too much, one runs the risk of finally
proving nothing but the obvious fact that Hinduism
possesses the morphological and typological features com-
mon to all forms of religion at a certain stage of develop-
ment. It must always be borne in mind that Hinduism is
the expression of a great civilization and is closely con-
nected with philosophical speculation and literary activity,
and that it is a product of the creative imagination and a
systematic construction.
Its sources are enormous, and consist of a great variety
of texts written in many dialects. There is hardly an Indian
language, from Sanskrit down to those of most recent
development, which has not been used to express religious
conceptions, usually before being used for many other pur-
poses. The slow and obscure beginnings of Sanskrit as a
secular language are well known; other dialects underwent
the same process. It has been constantly asserted that India
is obsessed with religion. It might be equally well main-
tained that India takes no cognizance of religion, at least
as an independent phenomenon. Religion is not conceived
as a duty, or as a problem facing every human being on
reaching maturity. It is a heritage and a tradition. The
only word which expresses it is dharma, which in the Veda
designated certain standards applying particularly to the
world of the gods. The term includes not only religion but
all the ethical, social and legal principles associated with
religion, and which together with it constitute the real
meaning of life for the Hindu. The word is so wide in
meaning that Ràdhàkrsnan can only define it as 'right
conduct'.
 HINDUISM: i 49
Hinduism is not built on any Canon or Gospel, and (I
suppose as a necessary corollary) it has no founder ór
dogma. The earliest available text, the Makdbhdrata, which
is rightly regarded as a summa of ancient Hinduism, devotes
no special attention to religion, even in its didactic pass-
ages. Its inspiration is fundamentally secular; its task is to
focus attention on a certain type of man, the Ksatriya. The
Laws of Manu provide a good illustration of the interlacing
of themes in Indian literature: here we have a legislative
text, or at any rate a book of legal maxims, which begins
with an account of the creation of the world, and ends with
an exposition of the future consequences of earthly actions,
the nature of the soul and the path of Liberation : nothing
could better illustrate the interdependence of all spheres
of human activity. If we trace the development of Smrti
from its origins, we can see how criminal law gradually
grew out of the theory of expiation, and how civil law itself
hewed out a path from ritual prescriptions.
If the Epic was intended for the edification of princes,
the Puranas seem to be specifically religious texts; they
claim to be divinely inspired, 'promulgated by Visnu, by
Siva or Brahman'; but this means little in India, where
every important text claims a divine origin. The Purdnas
are a store-house of myths and legends; together with the
Harivamsa, they are the principal source of Hindu myth-
ology. They contain descriptions of religious practices and
sanctuaries; they are the handbooks of the pilgrim. But
they are also what we should call works of popular science
dealing with cosmogony and genealogies of dynasties.
Their aim is to show how the reigning dynasties and the
dynasties to come (for there is a prophetic part, written, as
might be expected, a posteriori) are linked with the dynas-
ties of the past, back to the earliest days of myth. The
Purdnas even treat of grammar, poetics, and other secular
disciplines, as did the Vedic writings.
50 HINDUISM: I

It would have been easier to enumerate the texts devoid

of reference to religion: here there is a contrast with
Greece, where religious literature is practically non-exis-
tent. It is not by chance that no original treatise on
Lokdyata, the once flourishing materialistic school, has
come down to us. Everything which did not conform to
certain standards disappeared. Of the ancient theatre only
the noble genres have survived, with rare exceptions. The
Arthasdstra of Kautilya, an essentially secular politico-
economic treatise, passes for an inspired text: it begins
with a divine genealogy and ends with magical formulae
which are included, and not fortuitously, under the head-
ing upanisad.
Fundamentally, religious books can be defined as books
written for the use of a sect, like the Vaisnavite Samhitds,
the Saivite Agamas, and the Tantras proper. They proclaim
a new law, the tdntriko vidhih, as opposed to the vaidiko
vidhih; even the term tdntrikd srutih is found. They are
manuals, or if they were not so long I should say brevi-
aries, expressing definite conceptions. At this period, which
we may assume, as a hypothesis, to be the seventh century,
the great religious ideas have already been expressed
several times. It follows that religious literature in India is
a literature of reformation. We never see the first stages or
the foundation of a movement. It is a literature of an
anonymous and apocryphal nature : the first author to
whom a definite date can be assigned seems to be Jayadeva
in the twelfth century.
In these circumstances, it is not surprising that Hindu-
ism sought its inspiration in the Vedic Upanisads. Hence its
ceaseless efforts to compose new Upanisads, down to the
sixteenth century at least. These little treatises, however,
were scarcely suitable for their role : they look back to the
past; and they are the outcome of an academic preoccupa-
tion with a kind of meta-ritualism. They have scientific
 HINDUISM! I 51
pretensions, as we have already seen. The great classical
themes, Samsdra, Karman, the techniques of mysticism, the
problems of asceticism, are practically unknown to them.
In short, although we must acknowledge the great contri-
bution made by the Upanisads in directing Hindu doctrine
towards monism (for without their influence, India might
perhaps have gone over to dualism and have adopted the
doctrine of andtmatva, like the Buddhists) it must be ad-
mitted that they could have exerted no direct influence on
religious manifestations. The Veda is venerated from a dis-
tance, but it is hardly drawn upon at all.
There remains the Bhagavadgitd, which has been called
the Gospel of Krsna. This composite text, taken from the
Great Epic, expounds several teachings simultaneously; it
became the principal guide of many Indians just as the
Bhdgavatapurdna, the summa of Krsnaism, was later to be-
come, and as, later still, the Rdmdyana of Tulsldàs, the syn-
thesis of Ramaism, was to be in Northern India. Saivism
has nothing corresponding to these works. But they are all,
including the Gita, not so much new messages as sum-
maries of earlier teachings, and we can still discern in them
the ill-assorted elements from which the structure was
built up.
In order to write a detailed history of Hinduism, every
sort of evidence would have to be considered : the task has
not yet been accomplished even in India. We should have
little exact information on Vaisnavite ritual without the
Haribhaktivildsa, a sixteenth-century Caitanyan manual;
we should know little of religious feeling without certain
Bengali and Marathi poems, many of them of recent date.
Our best guide to the understanding of ancient mystical
movements is the attitude of Ràmakrsna in the nineteenth
century. In Indian studies ancient and modern evidence
alike must be taken into account; the testimony of folk-lore
and that of the "most erudite kdvyas must both make their
 52 HINDUISM: I

contribution. They are all documents to be considered;

and the most subtle writings are none the less authentic
evidence.
Literary evidence is not, however, a faithful mirror of
religious life. It emphasizes certain aspects and leaves
others out of account. The extent of the treatises on Buddh-
ism and Jainism is disproportionately large in relation to
the real importance of these movements. We should know
little of the position of women in India if we had nothing
but the idealized representations of Sita, DraupadI and
Sàvi tri. The idealization of woman is as great as her social
and religious status was low.
How, then, are we to approach Hinduism? In nearly
every religion there are two levels, that of popular belief,
and that of elaborated technique and speculation. In
Hinduism the distinction is very deep; yet at the same time
all stages of transition and coexistence are observable in it.
Its elementary level is as primitive as it is possible for a
religion to be, characterized by widespread idolatry (the
term should not be shirked), fetishism and animism. It has
been found that 80 per cent of the religious believers in
Southern India are devotees of the gràmadevatàs; and that a
great number of people are Hindus in name only, and can-
not even tell the Census agents whether they are Vaisna-
vites or Saivites. The higher levels are often very advanced.
But it is significant that the elementary beliefs are con-
stantly being endowed with new meanings. Malignant and
gruesome rites and narrowly materialistic conceptions may
be transformed into symbols. Veneration of the cow, that
rudimentary cult which is distinctive of Hinduism, may
take on a transcendent ethical value in the eyes of a
Gandhi. The myths, however worthless they may be, are
not excluded from any plane of thought. No philosopher
attacks orthodox mythology, as did Plato; much less the
common beliefs, as did Socrates. The Advaita distinguishes
 HINDUISM: i 53
between a lower mode of knowledge and a higher, tran-
scendent form; and there is a parallel distinction to be
observed in all religious manifestations, in the realm of
speculation such as that concerning pravrtti and nivrtti, and
in the division of society into ways of life or asramas, which
represents the contrast between the active life of the stu-
dent or householder, and the contemplative life of the
monk or the ascetic. Other religions select and eliminate;
Hinduism incorporates. Its every feature assumes a dis-
tinctive importance when studied by itself. It has been said
that Indian religion is essentially ritualistic; it has also been
said that Indian religion dispenses with ritual. India is
sometimes said to be completely Tantric, and sometimes
to be given over to bhakti. There is an element of truth in
all these contradictory statements : everything depends on
what aspect is selected for consideration and from what
standpoint it is regarded.
From Iran, there were probably Mazdean infiltrations
in Christian times, confined to a limited area. They have
left traces in the Mithra or Mihira of the Kusdna coins, in the
magas or priests described in some of the Purànas, in the
Surya of the Brhatsamhita, who is dressed in Iranian cos-
tume, and in a few other facts of a similar kind. But the
influence goes no deeper than this, and we cannot sub-
scribe to Przyluski's belief that the guna system, which is
clearly connected with the cosmic tripartition of the Veda,
was inspired by Iranian models.
Very little is known about the influence of Islam on
Hinduism, and the subject deserves close study. In the
course of a few centuries, a quarter of the Hindu popula-
tion abandoned the ancestral religion, either because they
were compelled to do so by force or because they were
attracted by the hope of material gains; such an event
cannot be without effects. There are, in fact, traces of a
few mixed sects, which we shall discuss later. But Indian
54 HINDUISM: i
literature shows hardly any Moslem influence, except, of
course, the Urdù. A sixteenth-century Hindi poem, Pad-
mavati, is sometimes quoted as an example; while its sub-
ject matter and style are essentially Indian, it is Sufi in
spirit; but it was written by a Moslem. It seems, in fact,
that in spite of undeniable resemblances, the mystical
movements developed independently. If Islam did influ-
ence Hinduism, it was by provoking a counter-reaction
which took the form either of an increased adherence to
the sects or of a stricter enforcing of the requirements of the
Hindu cult. It is hardly likely that it needed Islamic pro-
paganda to induce the various sects to abandon images,
when the whole of Hindu teaching, from time immemorial,
has emphasized the value of abstract truth and of wor-
shipping the intangible.
As regards the influence of China, if the admittedly
striking parallelism between the techniques of Toga and
those of Taoism lead us to assume that one borrowed from
the other it would seem that the initiative must be attri-
buted to India : for it is in India that the tradition is most
firmly established. If certain texts call Tantrism the
'Chinese practice', Cìnàcàra, the term refers to Tibet as the
region from which Vajrayana was diffused; it does not imply
that so profoundly Indian a movement had its roots out-
side.
In short, India has contributed much more to religion
than she has received from others. The expansion of
Hinduism from at least the second century onwards over
the whole of South-East Asia, from Burma to Java and
Bali, is well known. The facts have been partially obscured
by the predominance of Buddhism, which was usually
earlier in the field than Hinduism and appeared again
after it; for it never lost its hold over the masses, whereas
Vaisnavism and especially Saivism tended rather to be the
religions of the élite or of the State. Buddhism, free of
 HINDUISM! I 55
ritual and legal requirements, was a religion more readily
accepted in other countries than Hinduism, even a Hindu-
ism which had put aside the laws of caste. But Hindu
patterns have profoundly influenced iconography, the
ceremonies of kingship and the very conception of what a
king should be. Hindu imagery reached Tibet and the Far
East through the medium of Buddhism. It should be re-
membered that Buddhism played little part in developing
science and technology: for the diffusion of grammar or
poetics, for example, it made use of treatises of Hindu
inspiration, thinly disguised as Buddhist works.
The diffusion of Vaisnavite and Saivite ideas outside
India is enough to show that Hinduism, too, was a mission-
ary religion; at a very early date a Hinduist movement
took root in the Hellenistic world and penetrated as far as
Egypt. The decline of Hinduism after the Moslem period
must not be allowed to obscure this fact. The old law-
givers say that to be a Hindu, or, more exactly, to belong to
one of the three Aryan classes, means to have been born in
a certain area of Hindustan, the Àryàvarta (or homeland of
the Aryas); but this assertion need not be taken literally.
Hinduism long ago advanced beyond the limits assigned
to it by the laws of Manu, by means of conquest or peaceful
absorption, by marriage, and by adoption. Hinduism has
no word to express the process of conversion so frequently
referred to in Buddhist and Jaina apologetics, books
written by the converted for those to be converted; but
passages can be cited from the Mahàbhàrata which show
that people of low caste, enemies and foreigners were
received into the Hindu fold. Many people wanted to raise
their status and be admitted to Arya society; others fell
away from it through marriage outside its ranks and by
transgressions and misfortunes. A passage of Patanjali
attests that the Sakas and the Yavanas could perform sacri-
fices and accept food from an Arya without contaminating
56 HINDUISM: I
it. The fact is that Hinduism is a way of life, a mode of
thought, that becomes second nature. It is not so much its
practices that are important, for they can be dispensed
with; nor is it the Church, since it has no priesthood, or at
least no sacerdotal hierarchy. The important thing is to
accept certain fundamental conceptions, to acknowledge a
certain 'spirituality', a term much abused in current par-
lance. For many Hindus it would be quite legitimate to
take Jesus as istadevatd, without even regarding Him as an
avatara, so long as Indian tradition were acknowledged.
If we had a better knowledge of the social system in
which this religion arose, we could more easily decide
which features of it were public, and which were secret and
even esoteric. We tend to regard the various mdrgas, yogas
and darsanas as absolute truths in themselves, whereas they
are really aspects of the truth which exist side by side and
are not mutually exclusive. On the mythological plane, the
inconsistencies we sometimes discern are really non-exis-
tent. Despite the aggressive spirit of certain texts, indiffer-
ence to the identity of the gods is the dominating tendency:
'May Hari hear our prayer (says the Haristuti), Hari who
rules the three worlds, and whom the Saivites worship as
Siva, the Veddntins as Brahman, the Buddhists as Buddha,
the Naiyayikas as the Supreme Power, the Jainas as the
Liberated, the Mimdmsakas as karman* Even the heresies
are shown to pay homage to Hari. A much-quoted passage
from the Rgveda says: Ekam sad viprd bahudhd vadanti, 'That
supreme being (neuter) that is one, the holy poets say to be
many'. The same sentence is found in the popular songs of
the South. Hence the idea of maya, the principle of change,
and of lild, the divine 'play' described in mythology, by
which Indian thought disguises the mutual irreducibility
of the One and the Many. In religion, few ideas are found
to be finally irreconcilable.
The view that the outward ceremonial is intended for
 HINDUISM: i 57

the masses is in accordance with this line of thought. It is

worship of the mind which is important: meditation,
whether or not induced by external forms, by concrete
symbol or by mental image; and mental representation, it
is said, gives rise to experiences which are more concrete
and more dynamic than those evoked by representations in
wood and stone. Thus Hinduism prescribes a worship of
the absent, just as Vedism set up a hierarchy of formulae,
according to whether they were said aloud, whispered or
silently thought, just as the power of silence was reverenced
in ancient ritual, and as the Brahmanas prescribed as a
parallel to the Fire Oblation a Prànàgnihotra which was to
take place within the worshipper himself. The absent is
itself conceived in figurative terms: the imaginary pillar of
the Athawaveda supports the firmament; the absent Unga of
Chidambaram is alleged to be the most efficacious; the
Ganges is present in any stretch of water correctly invoked.
Religious speculation itself moves on two levels. Each
darsana, that is, each approach (for it would be wrong to
see them as so many systems), culminates in a way of
Liberation, or, as we should say, of salvation. This kind of
speculation, directed primarily or ultimately towards
soteriology, is in no way confined to professional philo-
sophers. Each Puràna examines the problems of brahman
and prakrti for itself, as if it were the first to deal with the
subject. Expositions of fundamental principles are found in
the Smrti. The Pancadasl popularizes abstruse metaphysics
in an agreeable verse-form. Johnston has shown that the
Sdmkhya of the Epic and the Purdnas is not altogether the
same as the Sdmkhya of the philosophers. 1
Here we see yet another level, which resulted in a fusion
of doctrines, the Sdmkhya-Veddnta compromise of many
ancient texts; such fusions are found at all periods: at the
1
E. H.Johnston, Early Sdmkhya, an Essay on its Historical Development
according to the Texts (London, 1937).
58 HINDUISM! I
present day we speak of the Toga-Veddnta, and at one time
teachers composed primers of Nydya-Vaisesika.
Hinduism has passed through many periods of time in
which adherence to various forms of practice ebbed and
flowed. In a given area, a period of idolatry would be fol-
lowed by a cult which repudiated images; and a whole
district would suddenly be converted to bhakti, or 'loving
devotion', against which the old non-dualist instinct would
subsequently revolt.
The part played by outside influences must be assessed
when considering facts of this kind. Leaving aside the
fundamental Asiatic element, the incalculable prehistoric
influences, we look first to Buddhism as a possible source of
influence. But in fact Buddhism does not seem to have had
any positive effect on Hinduism in early times. We may
suspect that the spread of Buddhism at first brought about
a crystallization of the spiritual values of Hinduism, and
later a relaxation of social and ritual intolerance; but
nothing can be proved. All the features common to ancient
Hinduism and the Pàli Canon spring from the Indian
background that they both share. Very much later, in the
field of philosophical speculation, we see in the Veddnta, in
Gaudapàda and Sankara, a tendency to remould the old
realism under the impetus of the idealist schools of
Buddhism; and in fact Sankara brahmanizes the Great
Vehicle, as many orthodox Hindus will agree. In Kumàrila
and the Logicians of the seventh and ninth centuries, there
arises a passionate controversy with the Buddhists, in
which neither side retreats at all from its position. There is
no question of influence. As for current religious practice,
it is difficult to see what impact Buddhism has made on it.
The foundation of monasticism in India by the Buddhists
and Jainas produced no answering movement outside their
own ranks until much later; and when it did come, it was
in circumstances that do not suggest any direct borrowing.
 HINDUISM: i 59
It has been suggested that Indian religious iconography
was inspired by that of Ancient Greece; but the material
for an iconography is already to be found in the Veda,
even to the description of many-armed divinities. It is
hardly likely that India had to await Greek impulsion to
transpose these indications into plastic form.
As we read India's literature and look at her works of
art, we are impressed by her fertility in creating divinities.
If, as Bergson said, 'le monde est une machine à faire des
dieux', then India has contributed more than her share.
Her mythology is almost as voluminous as at the Vedic
period; it has the same abundance of hymnology; but its
orientation is quite different. There is no longer any per-
ceptible connection (if there ever was one) between ritual
and legend. The Tantric form, a kind of dialogue between
Siva and the Goddess, is only a traditional framework. Of
course there is a basic correspondence between the form of
worship and the chosen divinity, and one's attitude to the
divinity; the details of the plastic representation are mean-
ingful, the arrangement of the hair, the dress, the orna-
ments, the facial expression and the gestures of hand and fin-
ger, the mudras, which have a language of their own. But
these are primarily static representations: ritual draws its
inspiration from a divine posture or dsana; it does not at-
tempt to reproduce a story. Already in the Veda, however,
one can see a tendency for ritual to become a dramatic
scenario, giving the impression of a myth in action. In the
Indramahotsava, the festival of Indra's standard, the king has
a tree felled, sets up the trunk in the middle of the city with
banners bearing pictures of Indra, and makes offerings to
it; finally it is rolled into the river to drift with the stream.
This ceremony is thought to be associated with a story
going back to the days of battles between gods and demons,
as in the Brdhmanas: the story went that the gods borrowed
Indra's standard in order to overcome their enemies. The
6o HINDUISM*. I
dìpàlì, a lamp-lighting festival still widely observed today,
though in a far less colourful form than formerly, was
originally a carnival representing the one-day reign upon
earth of the demon Bali, who, when vanquished by Indra,
was granted the privilege of returning to the earth once a
year to exercise his old power. The link between ritual and
image is very different from the primitive one, or rather
there is no longer any ritual, by which I mean liturgy in
the strict sense of the term. The public ceremonies of Vedic
times were not imitated even in the most spectacular festi-
vals of classical India; and private cult-practice contains
no mythiform elements. At the most there is an occasional
symbolic transposition, like the 'purification of the ele-
ments', which reflects the theme of kundalini. The officiant's
aim is to make the five elements, of which his body is com-
posed, merge into one another by means of formulae and
controlled breathing, until, little by little, consciousness
itself dissolves into the original prakrti. The rite of the
ascent of the kundalini is based on a more impressive piece
of imagery : that of the coiled serpent that sleeps within the
body and can be aroused by a certain technique, to pene-
trate one after the other the six superimposed circles until
it reaches the 'opening of Brahman' on the top of the head
and brings about the union of the being with Siva. But
there is no supporting myth underlying the imagery of this
technique. Moreover, when the mind by meditation
evolves the figure of a god, as in the case of Ganesa, his
characteristics really represent memories of certain sym-
bols translated into actions.
Let us consider the divinities again. Henceforward they
are quite distinct from natural phenomena, without, how-
ever, being any the less complex. Solar myth, for instance,
was relegated to the background, though it was later to
enjoy a brief revival under Iranian influence. Grierson put
forward the view that Visnu, the sun-god of the Veda,
 HINDUISM: I 61
1
represents a continuance of the solar theme; but the
theory is not convincing. It is true that there are Hindu
myths about the origin of the world which, if they do not
deal with the first beginnings, at least describe the later
stages of creation: the descent of the Ganges, the establish-
ment of the cosmic Unga, and the churning of the waters.
But after all, no mythology can ignore the act of creation.
These are not, moreover, the dominant themes of Hindu
mythology. Fertility myths are perhaps more prominent in
it, though J. J. Meyer has placed rather too much em-
phasis on them;2 too many disparate elements tend to be
grouped together under this convenient heading. But the
theme is certainly present, often appearing in the form of
sexual exploits. That teeming sexuality, which is already
traceable in the Veda, now appears in many of the gods.
It is the price paid for the asceticism and emasculation
imposed on India. In Siva it takes on a mystical aspect; in
Kama it is predominantly literary; in Ganesa it is popular
in expression (though at the same time mystical); in
Pradyumna and Krsna it is idyllic, in Skanda and the
southern Subrahmanya aggressive and indecent; in Indra
several aspects are combined. The typically Hindu figure
of Ganesa admits of various interpretations, according to
the level on which one sees him (for in the gods, too, there
are different levels of significance, ranging from the literal
to the transcendent plane) : this ancient elephant-genie of
the jungle, as M. Foucher calls him,3 when seen in his
popular aspect, is a libidinous figure who was to be used in
the symbolism of Tantrism. The eroticism which is some-
times considered to be characteristic of Hinduism, and
1
'The Monotheistic Religion of Ancient India', Asiatic Quart. Rev.,
28 (1909), p. 115.
2
Trilogie altindischer Màchie und Feste der Vegetation (Zurich und
Leipzig, 1937).
3
In his preface to A. Getty's book, Ganesa, a Monograph on the
Elephant-Faced God (Oxford, 1936).
62 HINDUISM: I
which appears most fully in the Tantric cult, and in the
extreme forms oibhakti, is already present in myth.
Battle-myths are frequent and are usually associated
with myths of the type just described. The struggles of
Siva against Arjuna and Sisupàla, his attack on Daksa and
his sacrifice, probably reflect the opposition that the Siva
cult met with before it was finally established. Myths of
expansion, like those of Rama and Agastya, are probably
based on memories of former conquests. The twofold aspect
of destroyer and protector, under which both the Goddess
and Siva appear, is the working of the Vedic legacy of
ambivalence; or rather, the actions of the gods are thought
of as encompassing many aspects of alternating activity
and rest. An interesting example is the battle cycle of the
Goddess, her fights with demons as in the Veda: it is
Candì who destroys the Asuras and their chief Mahisa,
who had in vain disguised himself as a buffalo. Or again,
the Goddess, having told the demon Sumbha who desired
her that no one could win her except by overcoming her in
battle (a well-known motif in legends of warrior-virgins),
routs the armies sent against her one after the other. But
there are other, very different portrayals, such as that of
the SatI who throws herself into the flames because of the
quarrel between Daksa, her father, and Siva, her husband
(a story that explains the origin of the practice known as
sail) ; commemoration of the spot where her bones fell to the
ground was the origin of setting up ptthas (sacred places).
Parvatl takes a vow of asceticism in order to tempt Siva;
thus mythology illustrates that great driving-force of reli-
gious life, the asceticism which subdues the passions. There
is the food-giving Goddess of Bengali legend, the Serpent-
Goddess Mànasà, the ogress Kali, who demands human
sacrifice; there are the ogresses or lesser spirits, Sitala or
Smallpox, SasthI who watches over the sixth day of the
new-born child's life; and lastly the divine Mother, the
 HINDUISM! I 63
supreme sakti, the energy of the gods. This was one of the
abstract forces of the Veda, now envisaged as an immensely
powerful woman. The patriotic exalting of India as the
mother-country that has arisen in recent times has helped
to keep this image alive. Even the male gods have been
drawn into this intensive feminization of mythical concep-
tions. Tukàràm calls upon Vitthala, that is, Visnu, as his
mother. The grdmadevatds are female. Another dominant
conception is that of the child-god, unknown in the Veda,
such as Krsna and Kumàra (especially as he appears in the
South, under the name of Subrahmanya).
The attributes of the gods share in the process of deifica-
tion: their animal mounts, their jewels, their weapons or
implements, the images associated with them. The worship
of particular aspects of the gods becomes predominant, a
tendency already discernible in the invocations of the
Yajurveda to Agni Grhapati, Agni Svistakrt, Agni Vaisvà-
nara, and so on. There are twenty-four nisthds or aspects of
Visnu, such as the bird Garuda, the Unga, and the disc
Sudarsana, each one having a symbolic meaning. The
function of iconography is to give concrete shape to some
of these conceptions, which might otherwise have been
indefinite in outline.
The legends of the gods seldom give us a consistent bio-
graphy of them. The two life-stories that do exist both
relate to deified heroes. Rama is the type of the victorious
prince, whose name is probably linked with the memory of
the Aryan invasion of Southern India and Ceylon. His
elevation to divine status was brought about by gradual
stages; Ramaist sects are late in appearing, although it is
alleged that there is mention of a god Rama in a passage
of Valmlki (which may be an interpolation).
Krsna's deification took place much more speedily: we
can see it happening in the theophany of the Gitd, and it is
at least possible that it provided the inspiration for Rama-
64 HINDUISM! I
worship. That Krsna was originally a mortal is very pro-
bable: he was a tribal chief, brought up in concealment
by shepherds so that he should be kept safe from his uncle,
the cruel king Kamsa; this is of course a folk-lore theme,
but it is not far removed from the events of everyday life*
Krsna founds cities, and settles near the mouths of the
Indus. In the war with the Pàndavas, he sides with the five
brothers who turn out to be his cousins : he now shows him-
self to be a cunning and unscrupulous counsellor, but his
advice brings about their victory. His last days are dark
and terrible, like the endings of so many heroes: he is fated
to bring about the destruction of his people and to look on
unmoved. Another strand is interwoven with this warrior-
cycle, that of the youthful Krsna, a pastoral demi-god who
sings and dances with the shepherdesses. Most of the
Indian themes, apart from that of asceticism, are found in
this complex legend, which alone provides a consistent
narrative of the hero's 6enfances\ like the mediaeval chansons
degeste. It was later to be imitated in the Adhyàtmaràmàyana,
which substitutes Rama as the hero. The conception of the
divine shepherd has been an important one in the history
of erotic mysticism.
There are no such complete mythical cycles associated
with the great gods. The conception of the Trimurti, whose
aspects are Brahman the creator, Visnu the preserver and
Siva the destroyer, is in accordance with ancient thought,
but it makes only a late appearance in the texts and never
gained much credence. While Brahman remains a some-
what insubstantial figure (though for some reason he plays
a part in the Buddhism of the Pàli Canon), a great accre-
tion of later detail was built round the other two. Visnu
absorbed local divinities into himself, such as the Vithobà
of the Marathi country, and Venkatesa and Tirupati in the
South. Nàràyana, who was probably a water-god, was
identified with him.
 HINDUISM: I 65
In comparison with Krsna or the Goddess, Visnu and
Siva seem to be dei otiosi; they have some of the character-
istics of the ancient Prajàpati. Sylvain Levi declared that
they were 'raisons societies'.1 Siva enters upon a long period
of inactivity as a Togin; in Tantric representations he
appears as a white and lifeless form at the side of a gigantic
Devi. The Satapatha reminds us that originally he had not
even a name (VI, 1, 3, 8). When he wants to take action,
he does so by proxy. In Bengal there is even a domestic
Siva, a kind of householder with a wife and sons. True,
there is also Siva the dancer, whose movements bring the
cosmic forces into being; but this is really a piece of ani-
mated symbolism rather than an action performed by
him. Visnu, too, very frequently appears as a reclining god,
vessel of the magical sleep in which the world is 'thought'.
He sends forth the vyùhas, emanations of power which are
diffused through the cosmos, and by whose agency he can
be present in the categories which create the forms. These
vyuhas are an artifice of the theologians, intended as a
demonstration of Visnu's part in the three fundamental
functions performed by the Trimurti.
The theory of the avatàras or incarnations of Visnu has a
wider appeal. This is a conception fundamental to myth-
ology, and, I would almost say, to the Indian mentality.
India's great men have so frequently in the past been
regarded as avatdras of the deity, and the tendency can
still occasionally be perceived today. It is the equivalent of
our doctrine of'the Word made flesh'; and, as Barth said,
it is the means by which striving after a higher monotheism
can be reconciled with the irradicable tendency to worship
multiple forms.2 In other words, the object was to safe-
1
UInde et le Monde (Paris, 1926), p. 83.
2
A. Barth, Les religions de l'Inde (Paris, 1882; new edition in the
Works of Barth, vol. 1, 1914, p. 153. English translation: The Religions
ofIndia, London, 1882).
66 HINDUISM: i

guard monism while at the same time acknowledging the

plurality of divine manifestations and the very principle of
divine intervention in the human sphere, or providence.
But this providence is not concerned with individual man,
but with humanity as a whole : its workings were accom-
plished in the remote past, and have little to do with
present-day humanity, which is perhaps unworthy of sal-
vation. The acknowledged purpose of the avatàras is to
save dharma when it is imperilled, the gods, the Brahmans
and the devout. The clearest formulation of this is found in
the Gita: 'Whenever the Law falters and chaos threatens, I
send forth my Self. I am born in succeeding ages to protect
the good and destroy the evil.5
The avatàras were originally independent legends which
came to centre on Visnu, perhaps because from the begin-
ning Visnu was the symbol of the propagation of the
divine : they are legends of prowess over demons, of the
usual type, which portray a hero, or a creature gifted with
superhuman powers, in animal or semi-animal form. The
last avatàr is still to come: the White Horse which is to be
its form is the symbol of a kind of Messiah, analogous to the
figure who in Buddhism is called Maitreya, and who has
other names in Iran.
Every Vaisnavite system has its own interpretation of
the avatàras. A distinction has arisen between partial ava-
tàras, both primary and secondary, which represent only
part of the Supreme Being, and complete avatàras, among
which R a m a and Krsna are often included. Eventually the
theory began to encroach upon Saivism. Visnu no longer
provided sufficient material for it, but his attributes played
their part in the larger picture.
This, then, is the Hindu conception of the world of the
gods; an overcrowded world, prolific of all kinds of per-
sonifications, some destined to live on, others only adven-
titious. There are very few elements capable of carrying
 HINDUISM: I 67
religious significance that escape divinization. The few
non-individuated groups of the Veda, like the Maruts and
Rudras, have given rise to many kinds of demons, who
intervene in human affairs on occasion, Yaksas, Vidyddharas,
Ndgas and many more. There is little sign of a hierarchical
order among all these beings, but their attributes are
usually very well-defined, as in the case of the Lokapdlas
who are connected with a cult of the cardinal points. For a
more highly-developed systematization, we have to look
outside Hinduism altogether, to the cosmological scholas-
ticism of the Jainas and the 'pure regions' of the Great
Vehicle.
There is no demoniac figure of any importance in Hindu-
ism, any more than there is in Vedism. It offers nothing
comparable to Mara, whose name the Hindu poets dis-
guise by making it a synonym of the God of Love. Demons
have a tendency to appear as devotees, like Ràvana, the
enemy of Rama, and Vrtra, the dragon killed by Indra.
Tulsldas says that Rama was incarnated to rescue Ràvana
from an undeserved fate. The new mythologists were pro-
bably deterred from creating fresh figures of the infernal
world by the ambivalence that still persisted in the sphere
of divine significances.
 IV
Hinduism: II

Wthan
E know much more about the behaviour of the gods
about what they meant to man. Each text has
its own attitude in this matter, the fact being that the gods
are really superfluous in Indian religion, and the essentials
could have been covered by the theory of karman and its
consequences, by the quest for Liberation. Yet the gods are
included in all doctrinal systems; not even the 'atheist'
schools deny the existence of divine beings. But many texts
portray them as beings of a limited power. They are gods
who have been 'brought forth' or who have 'come into
existence' as in the Sdmkhya; they are not only powerless to
intervene between man and karman, they are themselves
subject to this law. This is a position not unlike that of the
Buddhist cdevas\ According to the Saivasiddhdnta, they can-
not attain Liberation. They are aids to prayer and medi-
tation, rather like our saints.
The supreme deity is on quite a different plane, whether
he is known as Visnu—which may also mean Rama or
Krsna—or as Siva. The Supreme Being is no longer above
the world of myth, as in the Veda; he now plays a part in
mythology, though on an exalted level. This may perhaps
suggest that we are approaching a conception of God
rather than gods; and when we find that from the Svetds-
vMara-Upanisadonwards, the Isvara or Bhagavant is spoken
of as a being demonstrably beyond all contingency and
variation, this impression is strengthened. But it is a com-
plex question. In Sankara the idea of God belongs now to
 H I N D U I S M ! II 69
the sphere of higher knowledge, now to that of lower
knowledge, according to whether or not it is conceived of
as without qualities; in other words, whether or not it is the
repository of the impersonal neuter force called brahman.
It is admissible only by those who have not attained the
highest stage of perfection. Similarly dtman varies between
the old, semi-abstract idea of the Upanisadic atrnan, and the
new, concrete idea ofjlva. The idea of the neuter brahman,
which had been threatened by that of Isvara from the
middle Upanisadic period onwards, was re-established as a
principle of speculative thought by Sankara: brahman alone
is the Absolute, the Transcendent, but it is an impersonal,
neuter principle; Isvara is its restatement in terms of per-
son, but it is an imperfect restatement. To the Indian
mind, God is Thypostase deficiente de l'Absolu', as M.
Lacombe has said.1 The world is usually envisaged as an
attribute of the divine. In short, the divine is both tran-
scendent and immanent; traces of the original pantheism
are still apparent. There are, of course, endless gradations
of thought, but on the whole we may say that the negative
definition of the Brhadaranyaka-Upanisad gains the widest
acceptance. From early times it was said that the Supreme
Being is silence.
Let us look at the Arthapancaka, a thirteenth-century
South Indian text by Loka Pillai. The Isvara here is five-
fold; he is a personal god in the form of Visnu, dwelling in
the paradise Vaikuntha; his being is 'diffused5 among many
forms (in accordance with the conception of the vyUhas)
and he performs the trinity of cosmic functions; he is a
being capable of assuming special forms to appear on earth
(the conception of the avatdras); he is the inward guiding
power, antaryàmin, the equivalent ofatman; and finally he is
the area, the holy image made for cult-worship.
1
(Orally). Cf. Olivier Lacombe, UAbsolu selon le Vedànta (Paris,
1937)-
70 HINDUISM! I I
Of course, the poets of the people and the crowd-
preachers have simpler conceptions. A Southern poet,
probably of the eighth century, says: 'I look for bliss
neither to Indra, nor to Visnu, nor to Brahman; for me
only one god exists, Thou, our Supreme Being.' Ràmakrsna
in the nineteenth century approaches the idea of the
Christian God; but he, like many others, has a strange
mixture of beliefs : in moments of ecstasy he worships the
brahman without qualities; in ordinary devotion he includes
Krsna or Rama, and Siva or the Goddess in his worship.
The relationship between God and Man is, in large
measure, lacking: there is no reciprocity. God is not a
familiar despot like Jahveh; there is nothing of that some-
what cynical bargaining spirit towards him which we sense
when a poet of the Veda says, cDo thou give if thou wishest
me to give'. We have seen that the idea of Providence was
late in appearing, and did not impose itself strongly when
it did appear; the idea of a God who had pity on human
suffering made even slower progress. Grierson remarks that
it was Tulsldàs in the seventeenth century who initiated
the theme of divine compassion;1 yet it is foreshadowed
three millennia earlier, in the hymns to Varuna.
The transition from the neuter to the personal principle,
and the increasing predominance of representations in per-
sonal form, were brought about by the operation of a new
factor of great importance, known as bhakti. The term itself
is old, but the propagation of the idea it represents is rela-
tively recent, though we cannot say how or where it took
place. We should be cautious of attaching undue impor-
tance to the prosopopoeia of the Bhàgavatapuràna, in which
Bhakti personified is made to say, 'I was born in the
Dravidian country5. As affective 'participation' of the soul
in the divine, bhakti presupposes an object distinct from the
1
'Tulasi Dàsa, Poet and Religious Reformer', J . R. Asiat. Soc, 1903,
p. 447.
 HINDUISM! I I 71
subject. It would not therefore gain currency in a purely
monistic environment; it was better adjusted to the atheist
Sdmkhya than the Advaita Veddnta, for the Sdmkhya was
pluralist and relatively realist. Hence bhaktifirstemerges in
the Upanisads of Sdmkhya inspiration, especially in the
Svetdsvatara which Schrader called 'the gateway to Hindu-
ism'. 1 It was developed in the Bhagavadgitd, in which the
Sdmkhya-Toga concepts also dominate the speculative plane,
and much later in the Bhdgavatapurdna which is as much
Sdmkhya as Veddntin. Bhakti has always been better adapted
to a Vaisnavite background, probably because the emotive
imagery of Rama and Krsna was more fitted to sustain its
conceptions than the rather more abstract forms ofdoctrinal
Saivism. But it was an obscure impulse of the masses that
brought it into prominence, an impulse that probably origin-
ated in that centre of mystical outpourings which produced
the literature of the Southern 'saints'; this would indicate the
seventh century as the approximate date for the renais-
sance of bhakti. The Islamic hypothesis is unnecessary. In
brief the theory is that a God without attributes is inacces-
sible, and that there must be an intercessor. In the absence
of a founder or prophet, God 'incarnate' must be at once
the intermediary and the prophet of bhakti.
This religious conception was soon claimed or reclaimed
by speculative thought. Those powerful movements of
which Ràmànuja is the highest expression, freed themselves
from the impersonal brahman and adopted bhakti. Ràmànuja
defines it as 'an enduring recollection, having the nature of
an intuitive perception, which takes the form of abounding
love for the object of recollection, the supreme Self. Else-
where he calls it a calm, continuous movement, 'like a
stream of flowing oil'. Others say that it is a seva, the atti-
tude of a devoted servant to his master. Gradually the word
acquires many shades of associated meaning. Vallabha
1
F. Otto Schrader, Der Hinduismus (Tubingen, 1930), p. 1.
72 HINDUISM.* II
develops the idea of prapatti, an 'abandonment' of the
being to the divine; the initiative is then God's, and he
dispenses his grace; prasàda, and pusti or 'efflorescence', are
the terms designating this new aspect of belief. Theologians
argue whether man receives grace without any action of
his own, as a kitten is caught up by the scruff of its neck
and dragged away from danger by the mother-cat, or
whether, like a monkey that clings to its mother, he must
make a personal effort—a less Jansenist theory. In theolo-
gical terms this dispute is but another aspect of the old
conflict between daiva or fatality and purusakdra or the
value of human actions, between pravrtti, the doctrine of
action, and nivrtti, the doctrine of abstention.
The theory becomes increasingly subtle with the passage
of time. Formerly the easy path, the path of simple love, it
later tends to become the hard path to many people.
Already in the Bhdgavatapurdna nine degrees of bhakti are
distinguished: further subdivisions are recognized in the
Vaisnavite scholasticism of the Caitanya school. In India
to penetrate a subject more deeply is to classify it further.
The evolution of the word bhakti is comparable to that of
the Latin pietas, Bhakti originally meant participation in a
rite, just as pietas was the meticulous discharge of one's
duty to the gods; piaculum was the practice of religion
understood as the payment of a debt. Gradually the
Sanskrit term came to mean (as in Tulsìdàs) religion as a
whole. When the word has reached this stage of meaning,
then it is true to say that the whole of India is steeped in
bhakti.
This has unexpected consequences. Ramaite bhakti re-
mains relatively pure; it is the bhakti of a wife. Krsnaite
bhakti is erotic; it is the bhakti of a concubine; religious
beatitude comes to be conceived as a sort of carnal inter-
course with the god in the idyllic setting of Vrndavana. I
cannot say whether religious feeling profited greatly by
 H I N D U I S M : II 73
this development, but it was an undoubted gain for litera-
ture. Bhakti gave fresh impetus to religious inspiration; it
created a mixed genre, of which the Gitagovinda (a pastoral
poem that reminds us of the Song of Songs) was the proto-
type. The stimulus of an emotion, it is thought, arouses
religious feeling; generally, of course, it is the stimulus of
love, but it may also be that of hatred. Tulsidas says that
God's anger is as precious as his love; and we are told that
Prince Sisupàla gained Liberation because of his hatred
for Krsna. The mediaeval troubadours portrayed love as a
legal code, and drew up a set of rules for it; similarly,
Vaisnavite theologians conceive of bhakti in terms of
rhetoric. Rasas and bhdvas, moods and emotional states
repeatedly described by writers on poetics and the drama,
are given mystical significance. The final end is the highest
rasa, ujjvala, which represents the Eros of the god Krsna.
For the Sahajiyds worship remains platonic; it is directed
towards a woman deemed to be unattainable, for example
a chaste married woman.
Cosmogonie speculation, on both space and time, has
gained greatly in profundity since the Veda. In the realm
of space, we find extremely elaborate conceptions, such as
the cosmic Egg or brahmdnda, together with the various
celestial, earthly and subterranean spheres of which it is
composed. The Earth is in the central position : its pivotal
point is Mount Meru, above which stands the pole star.
Around Mount Meru lie the four island-continents, like
the four quarters of a circle, situated one at each cardinal
point. T h e early Buddhist conception is very similar. From
the Purdnas onwards it is replaced by a portrayal of con-
centric islands and seas, surrounding the principal conti-
nent, Jambudvlpa. This is a tangible representation of the
world. As an explanation of first principles, the theory is
sometimes advanced that there existed a material sub-
stratum or continuum, the prakrti, containing the quality-
74 H I N D U I S M : II
substances called gunas, a term which makes its appearance
in a passage of mystical symbolism in the Atharva. The
gunas, by mingling together, create the; material elements
that make up the world: this is a rudimentary atomism,
which formed the basis of the evolutive conception of the
cosmos put forward in Sdmkhya doctrine.
Indian speculation on time expanded the age of the
world to infinity; it conceived a series of kalpas, eras of
creation, each kalpa consisting of a thousand great ages or
mahdyugas, and each mahdyuga consisting of four yugas.
M. Filliozat has shown that there is agreement between
some of the Indian calculations and the figures arrived at
by Berosus and Heraclitus for the 'great years'. 1 The basis
of the computation is an astronomical fact, as Biot per-
ceived.2 The 'great year' of 4,320,000 years is only an
expansion of the Vedic yuga of five years, which was the
time taken by the sun and moon to accomplish a certain
number of complete revolutions simultaneously; 4,320,000
years is the period taken by the planets to accomplish the
same number of revolutions.
The present era, like every era, is made up of four
periods; we are now in the bad period or kaliyuga, which
began at a date corresponding to about 3000 B.C. This
period of retrogression and pessimism represents the after-
math following on the ideal of the Golden Age, an ideal
held in common by most ancient civilizations. The diffi-
culties and dangers prevailing at the time when the system
seems to have been laid down helped to gain credence for
it: it provided justification for inveighing against the
morals of the age and for denouncing those who are 'swans
outside and crows within', as Tulsldas was to say later.
1
Communication to the Société Asiatique de Paris, 9th April, 1948;
cf. J. Asiat., 1948-49 (Annuaire), p. 80, and 1950, p. 374.
2
J. B. Biot, Études sur l'astronomie indienne et sur Vastronomie chinoise
(Paris, 1862), p. 37.
 H I N D U I S M : II 75
The Purànas were written in a spirit of 'denial of history'
and Tear of history', as M. Eliade says.1 But there is a sort
of pre-scientific intuition at the root of this theory of the
progressive worsening of man's state; at all events the
system of four ages was retained in Indian astronomical
treatises, for it was considered to provide a satisfactory
principle of explanation. It should be added that each
kalpa is divided into fourteen equal cycles, called 'intervals
between the Manus'. The Manus are regents of the world,
combining in themselves the functions of legislator and
primitive king or father of the human race.
The end of the world is usually conceived as a confla-
gration followed by flood : it seems possible that this con-
ception is referred to in verses 39-40 of the Atharvan hymn
that we have already noticed as mentioning the gunas.
The end of the world, the pralaya, is not final, however.
There will be other worlds after it, as there were before it.
All these conceptions are dependent on the primordial
idea of 'perpetual recurrence'. The image often used to
illustrate this idea is that of a wheel that turns but does not
move forward.
On the subject of the next world, there is an inter-
mingling of widely-differing conceptions. There are widely-
accepted traditions about the pretas, departed spirits who
have not yet attained a definite status like the pitrs. There
is the more highly developed idea of a judgment of the
dead by Yama; and finally there is the idea of the direct
ascension of the soul to heaven. Heaven and hell belong to
the sphere of popular belief, but this did not deter some
theologians from adopting these conceptions, nor the late
Vedànta from reintroducing ideas of Paradise into its des-
cription of the nature of Liberation. Madhva puts forward
the theory of an everlasting hell.
1
Mircea Eliade, Le Mythe de VÉternel Retour (Paris, 1949), esp.
p. 207.
76 H I N D U I S M : II
The theory of transmigration is reconciled, as far as pos-
sible, with all these ideas. After the death of the body, it is
incumbent on the soul to assume the guardianship of
another body in one of the three realms in which the living
substance of the world is distributed. This is called
samsàra> a word that makes a hesitant appearance in some
of the old Upanisads, and means, properly, the universal
circulation of all creatures. The Gita calls it 'the great
fear'; it is more frightening than the idea of hell, because
man can see for himself its endless continuation. According
to the Pythagorean theory of reincarnation, it was attained
after a mystic initiation ceremony. But this is not so in
Indian samsara, which is a popular doctrine open to every-
one, and is expressed in crude forms, even in such a
weighty text as the Laws of Manu. Samsàra in India is a
necessary auxiliary to the theory of karman. The form of the
reborn body, in other words the status of the creature in its
new existence, depends on karman. The old cosmogonie
representations of superimposed heavens and hells and
states of being were modified or reconstructed in accord-
ance with this new doctrine. The universe becomes the
setting and the tool of a universal moral law.
The basis of karman is a scientific one, if we agree that it
is derived from the Vedic conception of rta: in this way,
karman is seen to be the natural order of things rooted in
the moral order and in causality. Action generates invis-
ible energy, unique of its kind (apurva), which influences
the soul and regulates the destiny of the individual. The
initial assumption is that actions are eternal : they follow
the spirit throughout its present and future lives, identi-
fying it unerringly 'as the calf finds its mother in a herd of
many cows'. Man is born as a debt, to use the vigorous
phrase of thetfatapathabrahmana;his existence is a 'long
sequence of borrowing and paying back' (Eliade). 1 Kar-
1
Op. cit., p. 145.
 HINDUISM*. II 77
man is both the future justification of past actions and the
explanation of the causes that have made our present life
as it is. Its logical result would be a complete determinism,
like that advocated in the early days of Buddhism by the
adepts of Gosàla, the Ajlvikas. But the principle was usually
modified in practice, for it tended to have an annihilating
effect on normal activity. Hinduism, moreover, has sel-
dom approached the problem systematically. The relation-
ship between karman and the divine presence in the world,
its effect on character, its legal, aesthetic and physical con-
sequences, were all considered, besides its specifically reli-
gious consequences. But on the other hand the difficulty of
reconciling the theory with human liberty was hardly
touched upon. Karman helped to make the dtman the sub-
ject of contemplation, which had for so long been turned
towards the infinite by the vistas of cosmogony.
From now on, the question of how the soul can be freed
from the cycle of rebirth constitutes the whole of religious
speculation. The principle can only be overthrown when
all action is spent, when the spirit is like a potter whose
wheel is no longer driven and ceases to turn. It is not
enough to refrain from action, not only because man finds
this impossible, but also because it would not annihilate
previous karman. The solution lies in following out to the
end one of the paths that lead to Liberation; among the
surest is Toga, a technique that aspires to confer mystic
powers. In this way the immortality of the Liberated is
attained. Bhakti, if carried out with sufficient thoroughness,
is another means to the goal; so is the apperception of
reality, as taught by the Veddnta; and so is the way of ritual,
which we shall discuss later. There is hardly any 'path5
which, in given conditions, India does not accept as valid.
If certain texts speak of the journey as very difficult, others,
of more popular inspiration, stress the value of more or less
everyday principles of morality in the effort. The Gita pre-
78 H I N D U I S M ! II

scribes disinterested action; desire alone corrupts. Accord-

ing to the Nydya, discussion and debate are among the
paths that lead to Liberation. Even alchemy, rasasdstra> can
lead to it, in that it teaches transubstantiation; Nàthism, as
it is called, envisages a profound change in the body which
would preserve it from all decay.
The nature of Liberation varies appreciably according
to whether the divine is represented as personal or imper-
sonal. Thus the Advaita, which sets up the impersonal
brahman as the supreme principle, conceives the union of
the liberated soul with the brahman as a dissolution of per-
sonality; it is compared to a river running into the sea. The
Togavdsistha says that it is like the condition of a stone.
Others compare it to that mysterious state that lies beyond
'deep sleep', the 'fourth' state, as it is called through
inability to define it: similarly, in Vedic times a fourth
pdda, transcending the three visible pddas of the cosmos,
was recognized. Yet even in the post-Sankaran Advaita,
there emerge activist conceptions of the state of the Liber-
ated, possibly under the influence of Mahàyàna Buddhism.
In an atheist system like ancient Sdmkhya, Liberation is a
state of illumination with no object to illuminate, a mirror
that reflects nothing; there is no consciousness; it is a state
verging on nothingness, a condition of absolute 'isolation'
(kaivalya). The theism of Toga, on the other hand, con-
ceives of union and integration. The philosophy of Vais-
navism shows a conception which is perhaps more highly
developed, according to which Liberation is union with the
personal god. Bhakti occupies a place of special impor-
tance; far from being thought of merely as a path, how-
ever, it replaces the idea of Liberation. Various stages of
union are recognized, but the personality is generally pre-
served, and sensual conceptions are sometimes associated
with it. Other doctrines, such as the Saivism of Kashmir,
preach the complete fusion of the spirit in the Paramani va.
 HINDUISM: I I 79
As M. de Glasenapp says, the functioning of the world then
becomes a process of objectifying the god's consciousness. 1
Other questions abound; for example whether all souls
are capable of Liberation, or whether there is predestina-
tion; or whether or not Liberation can be attained durinj
man's lifetime. The various systems give different answers.
Some doctrines conceive of a hitherto unknown human
category, the jwanmukta or 'liberated in life', whose spiritual
state should be carefully observed (so far as it is possible to
do so) : India has surely never lacked men of this kind,
from the time of the rsis to the present day.
A great deal has been said about Indian pessimism; and
its presence in karman and samsàra cannot be denied. The
Upanisads represent the end of a period of happiness that
derived from the Vedic atmosphere of well-being. The
Parable at the Well in the Mahabhdrata says that the world
is a fearsome jungle. But the evil is not beyond forgiveness;
there is no everlasting punishment; all suffering is justified,
by Indian standards. Here at last is a religion that gives an
objective account of the problem of evil. Moreover, one's
state may eventually improve, and this remote consolation
can itself help to free the spirit that is condemned to be
reborn. Lastly, and most important, it is within every
man's power to transcend the human state, if he has the
courage and if a guru shows him the path.
The karmamdrga, or way of ritual, has always been one
of the principal paths. Although it was opposed by some
systems and passed over by others, ritual remained a living
force, just as much as the mythical beliefs that it distantly
reflects. India without religious practice is inconceivable;
in fact, the sphere of this practice was notably extended
after the Vedic period.
It is true that the solemn rites of antiquity survive only
1
H. von Glasenapp, Die Philosophic der Inder (Stuttgart, 1949),
P- 279-
8o H I N D U I S M ! II
as show-pieces, scenes of simulated splendour immortalized
in inscriptions. Mentions of them are found from the
Sungas onwards, possibly indicating a reaction against the
State Buddhism instituted by Asoka. Animal sacrifices are
occasionally attested, despite the sects that urged ahimsa,
but they do not seem to stem directly from the Vedic
Pasu; they are particularly associated with the cult of
Kali, and more especially in Bengal. The function of the
old Srauta offices is supplied by the spectacular side of
Hinduism, pilgrimages, processions, visits to the tirthas, and
such festivals as still contain religious elements. We know
little enough of the festivals and pilgrimages of antiquity,
in spite of the màhàtmyas appended to Purànic literature,
which are impressive rather than detailed; we should like
to be able to draw up a calendar of festivals and a map of
pilgrimage routes, which would tell us a great deal about
the methods of transport both of men and goods. As well
as the regular festivals, there were ceremonies like the
Abhiseka, or royal enthronement, which is based on the
old Rajasuya, but introduces into it new procedures re-
flecting the growing prestige of the regional and imperial
monarchies. The voluntary self-immolation of the 'satis* is
in part a survival of the old purusamedha or human sacrifice,
and the action certainly has a religious character. In
private cult-practice, the Vedic elements remained more
or less unchanged, at least in the matter of the sixteen
samskàras>) which are accepted even by the Buddhists and
Jainas. They form part of dharma, and enable the young
Indian to enter Aryan society or guarantee him its privi-
leges.
New forms of worship arise that stand midway between
the public and private cults. Their characteristic is that
they are pùjàs, a word not yet fully explained, which indi-
cates worship of the holy image, preferably in the temple.
In principle this is an individual cult, although it is public,
 HINDUISM: II 81
and it may go back to an ancient hospitality rite: it re-
sembles the ceremonies depicting the reception of King
Soma, at the beginning of the Agnistoma. Its purpose is to
prepare a reception for the divinity present in the image,
who is clothed, adorned, fed and couched. Perfumes,
flowers and lights, all things of which little or no mention
is made in the ancient cult, now become essential to the
ceremony. If the rite is to be effective, the image must have
been consecrated by what is called the establishment of the
breath. The setting-up of the idol constitutes a separate
ceremony, a notable feature of which is the curious prac-
tice of opening its eyes.
The liturgy proper has quite fallen into desuetude. It is
true that we have little detailed knowledge of Hindu cult-
practice throughout the centuries, least of all of the Saivite
cult, of which no clear idea can be obtained from contem-
porary manuals. But whenever we can discern anything of
it, there is only a very approximate correspondence with
the ancient code ofclosely-packed injunctions and formulae,
and the skilful orchestration of the old siitras.
A formula is no longer linked to a definite action; it
develops independently. Often it is a murmur (the usual
name is japa, which originally referred to the recitation of
certainyajus) ; its content may be a long litany, such as the
1008 names of Visnu or Siva. The purascarana to Brahman
consists of 32,000 repetitions, according to the Mahdnir-
vanatantra. The effectiveness of the formula or mantra lies
not so much in the words as in the vital power with which
it is invested as a result of various mechanical devices. Very
often the actual acoustic qualities of the sound-units, and
still more their potentialities, are of greater importance
than the meaning. Words are formed arbitrarily, or altera-
tions are made in their normal structure. Disconnected
phrases are also found, as for example in the recitation of
the Pàsupatas, where such phrases are accompanied by
82 H I N D U I S M : II

hysterical manifestations. The discomfort caused by prdnd-

ydma may have been the partial cause of these abnormali-
ties, which remind us, on a larger scale, of the stobhas that
arose in Vedic chant. The 'germ' of the mantra is used by
itself; the concluding resonances of the sacred syllable 'orrC
become the subject of a whole series of speculations. Under-
lying all this is the idea that the recitation of the mantra,
even when inaudible, is the essential reality. For Tulsidas,
who is quite opposed to ritualistic religion, the Name of
R a m a is greater than Rama, and the divine Name is
greater than any divinity. The sacred word is an essential
implement of bhakti: it represents the impregnation of the
word by bhakti.
The formula was reinforced by gestures, mudrds, which
helped to keep its form intact: the technique of the mudrds
is still preserved even to the Far East. There are the
nyàsas, or placing of the hands on the different parts of the
body associated with various divinities. There are the dia-
grams oryantras, which express the visual element of medi-
tation, image or formula: they are geometrical figures,
composed of circles, triangles and lotus-petals, often with
letters inscribed on them. The yantra is conceived as a
miniature temple; the Sriyantra, the finest of the series, has
four openings, flights of steps, and a sanctuary where the
chosen divinity dwells. The disposition of the triangles
represents the male and female sexual organs, instruments
of the unio mystica; and the whole is framed by a wavy line,
sisirita, 'trembling as if with cold'.
But the yantra is after all only an image, and beyond it
lies the sphere of pure meditation. The Indians recognized
that the mental image can be richer and fuller than any
visual image, and that the reproach of idolatry could not
be levelled at it (though this was not a reproach that
greatly affected them). In certain forms of worship, cult-
objects, such as the flowers, are said to have spiritual
 HINDUISM*. II 83
equivalences; all the objects of external array become
symbols. In this way, any stretch of water can represent
the Ganges for the real believer. The initiate offers the
divinity the lotus of his heart as a seat, his thoughts as an
oblation, the ambrosia that flows from the lotus of the skull
as water to wash his feet; this ambrosia acts as ablution,
the breath as incense, the inward flame as a lamp. This is
a development of the Vedic prànàgnihotra, the Agnihotra
transformed into a sacrifice to the breath, in which the fire
was replaced by the officiant's mouth, and the ritual offer-
ings by homage to the five aspects of the breath and their
correlatives in the microcosm and in the macrocosm. Sub-
stitutions are practised everywhere; thus in the theory of
expiation, expiatory practices appropriate to each indi-
vidual sin are painstakingly established, and are then
replaced by general expiations, often purely theoretical in
character, which overthrow the whole system. The famous
vratasy religious 'observances', often fantastic in form and
setting, also contain substituted features that have a sym-
bolical basis.
The serpent that lies coiled inside the body, the kundalinl,
is another of these fantasies, a hallucination produced by
mystical ecstasy. The kundalinl myth is an aid to medita-
tion; it is a myth of intimate experience, representing an
experiment carried out by the subject on himself.
Meditation can also be induced without external aid, by
the exercises of Toga; its final state is that inward concen-
tration called samadhi, which would be translated inade-
quately by 'contemplation', and incorrectly by 'ecstasy'.
It is an ancient technique based on controlled breathing
and more or less acrobatic postures, adapted to the uses of
mysticism. Both the kundalinl and the more advanced
exercises of Toga have a representational function: they
unwittingly reproduce the orgasm that accompanies the
unto mystica> and the material manifestation of which is that
84 HINDUISM: II
outpouring of amria or soma> described to us in poetical
terms.
Many of the features we have discussed belong rather to
Tantrism than to normal Hinduism. They may be already
inherent in ancient forms of belief, but it is Tantric prac-
tice that systematizes them and brings out their full signi-
ficance. If we are to attempt to define Tantrism, we must
avoid facile judgments of the type that compares it to some
creeping disease gradually invading not only Hinduism,
but Buddhism as well. In the view of Avalon1 and others,
it represents the full flowering of the religious spirit of
India.
Tantrism is based on a code of esoteric practices. In the
extreme forms of it, at any rate, sexual representations,
elsewhere stifled or confined to mythology, are much in
evidence, as in the pancatattva, which consists in approach-
ing woman, newly dignified as iakti, by means of the five
'/7zY, madya, màmsa, matsya, mudrà, maithuna. Such doctrines
were no doubt of limited application, and their manifest
dangers were guarded against by careful discrimination in
the initiation of new adherents; the initiation process might
be either slow, semi-slow, rapid, or instantaneous, the
merits of instantaneous initiation being considered suffi-
cient to bring the neophyte to the verge of Liberation.
In the broader sense of the term Tantrism may be
defined as a technique designed to revitalize current prac-
tice and make it more expressive and more effective. By
bringing divergent tendencies to a head and by introducing
new trends of its own, the movement may be said to have
infused new life into Hinduism and to have saved its prac-
tices from becoming stereotyped. It is probably responsible
for the vigorous condition of many sects today, and to some
extent for the very survival of Hinduism.
1
Gf. Arthur Avalon, Principles of Tantra (London, 1914-16); Shakti
andShdkta (London and Madras, 2nd edition, 1920), etc.
 H I N D U I S M ! II 85
Tantrism, like all India's religious manifestations, has its
own philosophy, a combination oiSdmkhya and Veddnta; its
cosmogony is dominated by the concept of sakti, the divine
energy proceeding from the supreme Siva. From sakti pro-
ceeds bindu, the mystical drop which develops the com-
ponent elements of the universe; the system is still an
esoteric one, of the Sdmkhya type. According to the Gopdla-
tdpanl Upanisad, the universe proceeds from a mantra, the sun
from one sound-unit, the moon from another. Coexistent
with the tangible world is the world of sound, also sprung
from the original bindu. There is a return to the old Vedic
correlations, which are given dynamic form, and are con-
ceived not as the products of reflection but of intimate
experience. In Tantric mythology an important part is
played by female divinities, chief among whom is the
Goddess, appearing in many forms. The role of creator
belongs to her, the universal Wife and Mother, or to the
sakti that proceeds from her, in face of Siva's immobility.
Tantrism proposes its own path of Liberation. There is
an essential distinction between the 'path of the right' (or
right-hand path) and the 'path of the left' (or left-hand
path). The method of the path of the right is the usual one
of Toga and bhakti: Pott, who has recently made a study
of Tantric Toga, says : 'Here the union with the All-Highest
is aspired to in an emotional-dynamic sense: in an indi-
vidual effort to arrive at the elimination of the duality sub-
ject-object, this Toga knows of a self-abnegation, a blissful-
ness in which the soul—passive itself—is lifted up by the
divine grace, and in which form of Toga there is the possi-
bility of a union born and growing in a state of ecstasy.'1
This is Srwidyd, or 'sacred knowledge'; it is also called
Layayoga, the Toga of absorption, an expression that refers
to the rite or inward myth of the kundalinl, and the break-
1
P. H. Pott, Toga en Tantra in hunne Beteekenis voor de Indisene Archaeo-
logie (Leiden, 1946). Quoted from the resume in English, p. 157.
86 HINDUISM: II
ing through the six circles of the body, which is the charac-
teristic feature of this doctrine. It is only the imagery that
is Tantric; in other respects it is not so much Tantrism
proper as Tantric bkakti; the essential element is the pre-
sence and co-operation of the divine.
The 'path of the left' is very different, for there man
must rely on his own efforts. To become an avadhùta, a man
who has 'shaken off5 the passions, sense-impressions must
be, not temporarily suspended, as in Toga, but intensified,
so that man may realize the emptiness of the pleasure they
arouse. For the Toga of asceticism is substituted a Toga of
indulgence, Bhoga. It is a complete reversal of the usual
values: 'fair is foul, and foul is fair' might be its motto.
'A thorn can be plucked out with another thorn', as one of
the Tantric texts says. In a sense it is an easy path, for, says
the Guhyasamdja, 'perfection can be attained easily by satis-
fying all the desires'. But there are many dangers; man
cannot follow this path without preparation, which is not
made available to all. 'The enjoyment of wine, food and
women is the salvation of him who has understanding.'
Ta evam veda : already in the Brdhmanas this was the pass-
word that gave access to all that religion had to offer. To
quote Pott once more, 'when following the "left path"
the sddhaka strives after the destruction of the individual
ego by concentrating on destroying the elements out of
which the ego has been built up'. 1 The main purpose is to
disintegrate the personality, by a process of bringing com-
pelling influence to bear on the subject and of impressively
staged imitation, which is in fact magic. As in the macabre
forms of popular abhicdra, the favourite setting is the place
of cremation, the cemetery where the material body is
destroyed; demoniac conceptions are brought into play. The
term samddhi is interpreted as 'place of death' by the
Virasaivas, who profess a similar Tantrism. The central
iOp.cit.jp. 159.
 HINDUISM: II 87

figure of the cult is Bhairava, the terrible aspect of Siva,

with his assistants. Bhairava's seat is the eight-petalled
lotus; each petal supports one of the eight aspects of the
god, and is also connected with the human body and its
characteristic features; this is the lotus of the heart, the
Hrdayapundarika, that appears in Purànic literature. The
supreme liturgy is the sexual act, varied by all kinds of
practices which need not be described in detail.
Although these doctrines are esoteric, they are also
democratic, for they do not recognize distinctions of caste
or sect: they are a kind of freemasonry. Men are classified
according to their degree of qualification, the stage they
have reached, and the good opinion of their guru : the guru
has an important part in all these mystery cults for which
an exacting initiation is required. There are the pasus or
'beasts' who understand only the rites of the right-hand
path, or who, if they are admitted to those 'of the left', only
practise them in altered form. Then there are the 'heroes',
who follow the path of the left, either secretly, or, if they
have reached a very exalted stage, openly. Sometimes an
even higher category is recognized, the 'divine'; they are
those who have transcended ritual obligations and concern
themselves only with the symbols. They are the equivalent
of the 'liberated-in-life' of other doctrines.
We are not yet ready to seek for the origins of Tantrism.
The germ of the movement has been sought, and conse-
quently found, in the Veda, as is so often the case with
Indian problems which really have no beginning. Tantric
correlations are, in fact, clearly connected with the old
equivalences of the Vedic world; its magical and cosmo-
gonie background suggests that of the Atharvaveda, in which
arbitrary linguistic signs and 'veiled language' already
appear. But this only shows that Tantrism may have had a
shadowy prehistory and that it may contain very ancient
material. Its system is new, however, and we do not know
88 H I N D U I S M : II
what events led to its formulation, nor whether it is of
purely Indian origin or made up from miscellaneous
sources external to Hinduism. Even the genetical con-
nection between Hindu Tantrism and Buddhist Tantrism,
or Vajraydna, is far from being established. We can only
observe that it grew up, not within Hinduism as a whole,
but in the sects, in Saivite rather than in Vaisnavite circles,
and in Sàkta sects even more than in Saivite ones. The
general fictional setting is Saivite-Sàkta; it takes the form
of a dialogue between Siva and Parvati, in which the
Goddess alternately asks questions as a pupil, and answers
as a teacher the questions put by the god. These two forms
of exposition were called by the old terms Nigama and
Àgama. The dialogue form was inspired by the Purdnas,
where it is constantly used; it is also found in the Adhydt-
mardmdyana and in Tulsidàs.
Nowhere in the world is there a system of speculative
thought, or rather, a representation of mystical truth, that
is more radical or more paradoxical in form than Tan-
trism. India has shown that, as M. Eliade says, 'man can
transcend his mortal condition with equal success by the
way of Asceticism, or by its polar opposite, the way of
Eroticism. They both lead him to the same condition:
freedom from contingency, the state of the liberated in life,
the man-god.' 1
1
Techniques du Toga (Paris, 1948), p. 213.
 V
Hinduism : IH

H INDUISM, which we have so far studied as a single

entity, now appears as a mosaic of sects and indepen-
dent groups that are sometimes in rivalry; not that the
total number of Hindu believers has ever been represented
by the total membership of the sects, for this is far from
being the case. Though no statistics are available, even for
the present day, we have grounds for supposing that the
most active sects were themselves only isolated groups
within the great body of believers. This great mass is called
by a name that distinguishes them from the sectarians;
they are Smàrtas, people who follow the tradition of Smrti,
of the Puranas, in other words the Ancients; one might
almost say the orthodox, if that did not imply that the rest
were heterodox. The Smàrtas, too, seem to have gradually
taken over sectarian features : they adopted a special form
ofpùjà, for example.
Yet the sects are alive and pertinacious. They often give
evidence of a real effort to contest the laxness of existing
religion, and to revive its discipline; but occasionally their
object is to make religion more human and accessible.
Most of the great men of Hinduism have been founders of
sects and reformers. But for the sects, Hinduism would
have had no internal history; the subject will therefore
delay us a little.
Sects arise more readily in religions which have no
clearly defined dogma; for there is then no risk of heresy in
a personal preference for one special book or one special
90 HINDUISM.* I I I
form of worship, or in individual variations in points of
detail. Sects are sometimes just as radical as movements
that declare themselves as unequivocally outside Hindu-
ism: the Vlrasaivas are in some respects less Hindu than
the modern Jainas, for they have cast aside more elements
of the common heritage. The Mànbhàvs of the Marathi
country have retained practically nothing of Hinduism;
they minimize the importance of all the gods except
Krsna and Dattàtreya; they despise sruti and avoid the
temples; and yet despite the fear or scorn they inspire, they
do not feel themselves to be excluded from the community.
The fact seems to be that since Buddhism, which has long
since disappeared from India, and Jainism, which shows a
tendency to return to the fold of Hinduism, there have been
no heresies and no serious schism even, unless the rival
trends of Tengalai and Vadagalai within the Srivaisnavas
can be counted as such. The old heresies of the time of
the Buddha seem to have been reabsorbed into the general
background. The 'reformed sects', to use Farquhar's
phrase, 1 that is the Kablrpanthis, the Sikhs and a few
others, are on the borderline of Hinduism, rather than
outside it. It is notable that all these movements sprang
from Vaisnavism; Saivism, although apparently more ex-
tremist, has always been on the whole more conservative
of tradition. Kablr rejects ritual and Brahmanical specu-
lation, but retains certain basic beliefs, such as karman and
samsdra; he teaches of maya, reserves the name Rama or
Hari for the Supreme Being, recognizes the divinization of
the guru and the efficacy of repeating the divine Name.
T h e Sikhs, who took over some of the features of Kabir's
'religion', consider themselves as a separate sect on political
rather than religious grounds. Nevertheless, they have
established a Bible of their own, the Granth, composed on
1
J. N. Farquhar, An Outline of the Religious Literature of India (Lon-
don, 1920), p. 330.
 HINDUISM'. III 91
the model of the Rgveda; in inspiration it resembles the
Gita, and its general principles are bhakti and divine grace.
Both Kablr and the Sikhs are the spiritual descendants of
Ràmànanda, an ascetic whose reforms seem to have been
motivated as much by social as by religious considerations,
for he made the vernacular the medium of preaching (as
the Buddha had previously done), and received low castes
and outcastes into his community, as well as women. It
has been suggested also that there may be some connection
between these movements and the Nàths, who practise a
Saivism based on Toga, possibly the Hathayoga. Tradition
has it that there was a personal contact between Gora-
khnàth and Kabìr or Nànak.
How, then, can an Indian sect be defined? Some fea-
tures, such as the wearing of certain emblems, are merely
superficial. Adherence to a particular sacred book and to a
particular divinity is more decisive. This does not mean
that there is a sect attached to each member of the pan-
theon; the Puranas which express allegiance to Visnu or
Siva are not for that reason the manifesto of any particular
sect. In effect, the sects of India are divided into two
groups, Vaisnavite and Sai vite; the Sàkta movements can
be regarded as an independent development with a Saivite
background. We must also take into account the evidence
that a community giving allegiance to the Sun-god, the
Sauras or Saurapatas, existed in ancient times; the move-
ment may well have been associated with the rise of a solar
cult brought from Iran, or at any rate promoted by Iranian
influence. There are also traces of a group paying devotion
to Ganesa, which is thought to have flourished between
the sixth and eighth centuries; but this may have been an
esoteric tendency superimposed on other beliefs.
Adherence to a particular sacred book does not neces-
sarily imply innovation; it is more common to draw upon
the past than to invent anew. But it often happens that the
 92 HINDUISM: H I

poems or sermons of the founder are regarded as a new

Canon, as with the Granth of the Sikhs, which was vener-
ated like an eleventh guru. The Caitanya devotees have an
oral literature, attributed to their leader; the written works,
the six Samdarbhas, are little more than a theological com-
mentary. Some books are common to many sects, among
them several Upanisads, the Bhagavadgitd and the Bhdgava-
tapurdna, which few Indians reject. The Sivanàràyanls, a
movement of the eighteenth century, assert that their
scriptures are of great antiquity, and that they were incom-
prehensible until an inspired ascetic translated them.
Every group makes similar efforts to prove that it is of
ancient origin; this is a typically Indian tendency, shown
in the efforts made by the composers of the Epic and all
the great texts of antiquity to build up a conjectural struc-
ture, starting with a revealed Ur-text, existing from time
immemorial, which was lost, recovered later, and sum-
marized, this summary in its turn being lost and re-
summarized, and so on. In other texts, a mythical sage is
claimed as author, so that a whole literature takes on an
apocryphal aspect.
Each sect also adopts a philosophical standpoint of its
own; that is to say, it takes up a position in relation to the
classical Veddnta and Sdmkkya; in addition, it has something
to contribute to a theory of cosmology. It must be admitted
that we know nothing of the speculative thought of certain
groups; it is possible that those who developed none dis-
appeared the more easily for that very reason. In this con-
nection, it should be noted that the Sankaran Veddnta is not
connected with any sect and that none of them directly
adopted it. It is true that the Smàrtas pass for Sankarans,
but I think it more likely that this is because of the reli-
gious foundations attributed to Sankara than because of his
philosophical activities. Neither is the Sdmkhya in its pure
state the subject of sectarian speculation. No doubt these
 HINDUISM: ni 93
systems were too rigid and had too little hold over the
masses. On the other hand, the theistic Sdmkhya and the
non-Sankaran Veddnta provided the doctrinal basis of a
series of religious movements, a fact which emphasizes the
important part played by bhakti in the development of the
sects. It was by means oi bhakti that they were able to sway
the tendencies of philosophical doctrine; there is no sect
without some element of bhakti.
There is not necessarily any organized religious system
within these groups. Hinduism has no sense of hierarchy;
the monastic communities, the essence of Buddhism and
Jainism, were not widely imitated in this respect. India
early established the categories of mendicant monk, an-
chorite, and ascetic, and on this foundation built up the
asramas, in the old sense of the term, which designates the
divisions of human existence. But it is not until the ninth
century that the first traces of a Hindu order appear,
allegedly founded by Sankara, with its ten brotherhoods,
its regional directors, and its supreme head who bears the
reverential title of jagadguru. Later there is mention of
monasteries founded by the Virasaivas, by several Vais-
navite sects (notably by Ràmànuja at Srirangam and
Melkote), and even by the Kablrpanthis. The inscriptions
of the South give details of the practical organization of the
mathas and sanctuaries from the thirteenth century on-
wards; under the Pàndyas there are references to the
Ekadandins, elsewhere to the Tridandins. But all this
amounts to very little, and the history of Hindu mona-
chism, if it can be said to have a history, has yet to be
written.
Finally, a sect cannot be founded without personal
initiative. Of course, there are movements that do not
claim one man as their founder, and others, like the
Pàsupatas and the Kànphatins, that trace their origin to
legendary figures. But there must have been an historical
94 H I N D U I S M : III
basis on which the legendary element was superimposed;
we must bear in mind the recognized Indian tendency to
confuse the facts as time goes on. There are many simi-
larities in the historical origins of the various sects, and the
biographies of their holy men present many examples of
reduplication. The correspondences between the Buddha
and Mahàvlra have caused much surprise, but there
are other similar cases. The usual pattern is as follows : at a
certain point in his life a man dits himself off from his
normal surroundings, though nothing in his former ways
or in his heredity predisposed him to this action; he receives
illumination; he goes forth to preach a new doctrine, and
introduces new practices; he meets with many difficulties
and sometimes with persecution. He finally succeeds in
forming a group of disciples, laymen as well as religious
devotees, and he designates one of them to be his suc-
cessor. This is not a universal legend, however, for there
were no founders of cults in Greece.
The Master either commits his doctrine to writing him-
self, or his words are recorded by someone else. Caitanya
and Ràmànanda, who left virtually no literary record, are
isolated cases; and their disciples have largely made good
the deficiency. Cakradhara, who gathered the Mànbhàvs
together, is a semi-legendary founder; the son of a minister,
he led the life of a libertine and died at twenty-five. Then,
just as Safikara entered into the body of King Amaru to
learn the ars amandi (the only discipline of which he was
ignorant), so the soul of Sri Gàngadeva passes into Cak-
radhara's body, and he returns to life; then, after a per-
sonal sorrow, the death of his wife, he becomes an ascetic,
founds a community, and ends his days in the Himalayas.
The lives of Lakulisa and Gorakhnàth contain features
that are even more fanciful. But the legends are usually
built up round some definite locality, and have some
factual basis that can be corroborated.
 H I N D U I S M : III 95
The founder and his successors may become the objects
of a personal cult, which is sometimes carried to great
lengths. This fact can be explained without reference to
extra-Indian influences, for it has always been a feature of
Tantrism. Perhaps the first historical evidence of it is the
divinization of the Tamil 'saints' under the Colas. The
followers of Caitanya eventually exalted their master to
such a degree that he almost eclipses Krsna. In the case of
the Vallabhàcàryas, whose hereditary leader is a guru
directly descended from the founder, worship of this type
caused considerable scandal in the eighteenth century. The
conception oiavatdras facilitated the process of deification.
For many Hindus nothing could be more dangerous than
the abolition of images : they put the living guru in their
place.
Sectarian teaching, by definition, aims at reform. Gn the
one hand, a sect preaches the purification of religion,
renunciation of images and of animal sacrifice; in this way
it provides a check to the natural drift of religious practice.
Some groups emphasize the need for asceticism, because
their dominant element is composed of samnydsins. On the
other hand, a sect may aim at making a wider appeal to
the masses, may protest against esotericism and caste dis-
tinctions, and demand that the texts shall be made more
easily understandable and that preaching shall be in the
vernacular; in the eleventh century, for example, the
Lingàyats, before the time of Basava, composed sermons in
Kanarese, the vaccinas. It may be surmised that the sects
played an important part in the spread of Indian dialects.
But it may also happen, as in the case of the Mànbhàvs,
that a sect favours a greater degree of esotericism, and
adopts secret writings. Finally it must be noted that cer-
tain little-known movements came into being under stimu-
lus from outside, though as yet we have little information
on this subject; it is certain, however, that the Vaisnavite
 96 H I N D U I S M : III
Sahajiyas of Bengal were influenced by Buddhism; the
Bengali cult of Dharma was influenced by Buddhist Tan-
trism; and there is a current of crypto-Buddhism among
the Saivite or Sàkta Nàths. It is more than probable, too,
that there is an Islamic element in the Dharma cult and in
the Bengali Bàuls (Bengal was certainly the region most
exposed to outside influence), as well as in Kablr and the
Sants. But the basis of all these movements is indisputably
Hindu.
The attitude of the sects to the caste-system must also be
considered. There is no evidence of any serious revolt
against the system as such; but there is insistence that all
members have equal religious rights; otherwise no true
community would be possible. The disintegration of caste-
barriers is brought about indirectly, by the. establishment
of a new scale of values. Gaitanya, who is supposed to have
preached universal brotherhood, displays a guarded atti-
tude to caste; the initiation of a man belonging to one of
the three highest classes is not undertaken by a iUdra
master. Certain groups draw their membership from one
professional class; the Ràidàsls, for example, are tanners,
and the Sadhanpanthis are butchers. The tanner caste of
Càmars tended to form a sect on their own. It is said that
there is no intermarriage between the two Srivaisnava
schools, the Tengalai of the south and the Vadagalai of the
north, although it is only a theological dispute that separ-
ates them. The Virasaivas have a body of hereditary
priests, the jangamas or Hingas in motion5 : and hereditary
privilege leads eventually to the caste-system.
We are also faced with the problem of deciding when the
sects arose. It would be tempting to suppose that the Upani-
sads were the products of schools that were in the process
of developing into sects: it was the speculations of these
groups of adherents to a particular form of ritual that came
to be known as the Vedic sàkhàs. But this theory, though
 HINDUISM*. III 97
not improbable in itself, would be somewhat arbitrary. On
the other hand, the only example of a school developing
into a sect is provided by one of these sàkhàs, that of the
Vaikhànasas of the Tajurveda; but it would be possible to
see this as a case of the reverse procedure of a community
of ascetics styling themselves a sdkhd; however this may be,
the Vaisnavite Samhitd of the Vaikhànasas is a direct con-
tinuation of the Kalpasutra of the same name. But let us
consider the subject in more general terms: we know that
there were devotees of Visnu and Siva at a fairly early
date; inscriptions attest dynastic nomenclatures that are
sometimes Vaisnavite and sometimes Saivite in style. This
does not necessarily imply the existence of clearly defined
groups. It is true that in the Epics and later works specific
names are found, such as the Pàsupatas, and the Bhàga-
vatas referred to in the second century B.C. by Heliodorus,
who was the ambassador of the Greek King Antalikhita
(Antialkidas), to the ruler of Besnagar, Bhàgabhadra of the
Sunga dynasty. But we cannot be certain that the refer-
ences are to sects; the texts are not explicit on the point.
Ramaism emerges as an independent movement only at a
late period, but the cult of Rama is an ancient one; there
are surely grounds for the supposition that Ràmaist groups
existed in ancient times. Similarly, the Sàkta cult must also
be ancient, although the Sàkta sects are of comparatively
recent formation. But nothing is really known of the history
of the sects before the end of the tenth century, or in the
South perhaps a little earlier; the history of Saivism can be
traced back slightly further. The essential doctrines are
well established at this period, and the sects have reached
the stage of revivifying speculations that have become out-
worn.
It is clear that Saivism is related on the one hand to
Nyqya-Vaisesika, and on the other hand to Toga. Siva is
the central figure in the mythological background of the
98 HINDUISM: in

Toga mystery and a Saivite atmosphere was favourable to

its development; it is only at a late period that there is any
Saivite adherence to the Vedànta, and then it was confined
to Srlkantha's movement. Saivism attaches great impor-
tance to the practices, especially to asceticism (the majority
of ascetics are Saivites), but it values bhakti less highly; it
leans towards esotericism. It is interesting to note that it
was Saivism that gained the firmest foothold in Eastern
Asia, and produced a vast body of speculative literature
principally in Old Javanese; according to Zieseniss,1 this
literature represents a stage earlier than that of the
Saivasiddhanta of Southern India. It was Saivism that pro-
fited most by the expulsion of Buddhism and the decline
of the Jainas. The Vaisnavite movements do not definitely
take the lead until the eleventh century, when bhakti comes
into prominence, and under the personal impetus of
Ràmànuja, which seems to have been a decisive factor. At
this period, too, Vaisnavism predominates in Cambodia
and in other parts of South-East Asia; there is a remark-
able degree of coincidence between the ebb and flow of
these movements on the Indian mainland and in the
countries to which Hinduism has spread. At the present
time, Vaisnavism is generally predominant, except in
Bengal where the Sàktas have retained their hold. Ritual
is less in evidence in Vaisnavism than in Saivism (though
it is not altogether lacking, as has been supposed); and
Vaisnavism is in general milder and purer in form; there
is no trace of the crude and semi-barbaric features of the
Siva cult. The extreme aspects of Tantrism are less appar-
ent in it, and its contribution to literature is greater; nearly
all the great religious works are Vaisnavite, and there is on
the whole a close correspondence between doctrine and cult.
1
Studien zur Geschichte des Sivaismus: die Sivaitischen Systeme
in der altjavanischen Literatur', /. Bijdr. Taal-, Land-, Volkenk.
Ned-Ind., 98 (1939), p. 74.
 HINDUISM! III 99
The existence of these sects side by side has, of course,
given rise to controversy and to a kind of endemic rivalry;
and all the sects present a common front against Sankara.
But there have been few violent clashes, and little or no
persecution; if the Mànbhàvs and the Virasaivas found
themselves in trouble, it was because of the intractability
of their behaviour in society rather than because of their
beliefs. Tolerance prevails; or perhaps it should be called
indifference. Gaitanya was an adept of bhakti, and an
advaitin at the same time, like Sridhara; Sankara himself
was a devotee of Siva; Appayadiksita, a Smàrta who was
converted to Sàktism, composed commentaries on both
Saivite and Vaisnavite works. The administration of the
temples may have given rise to local clashes, but these
were not directly connected with sectarian adherence; the
inscriptions of the Pàndyas mention quarrels of precedence
and so on in the thirteenth century.
I shall now attempt a brief historical survey of Hindu-
ism, along the lines I have indicated.
As we have seen, there are grounds for supposing that
some form of Hinduism existed at the Vedic period, and
probably even earlier. This is the prehistory of Hinduism;
its history begins with the emergence of the great texts of
Hindu dharma, and the first appearance of the Epic and of
Smrti. At this point Hindu belief achieves a status of its
own; there is an effort to define its position in relation to
Vedic practice, just as there was a definition of the status of
the Pàninean bhàsà in relation to the chandas. The same
period saw the origins of Buddhism and Jainism; a survey
of the allusions to Hinduism in the heretical Canons would
be of great interest. The Buddha attacks caste prejudices
and the proud attitude of the Brahmans, but he scarcely
touches on the fundamentals of belief; in the TevijjaSutta
he asks how the god Brahman whom no one has ever seen
can be an object of worship; but the cult of Brahman was
100 HINDUISM.' III
one of those practically unknown to Hinduism. There is
virtually no general attack on Hindu practices, except on
bloody sacrifices, which are condemned in the name of
ahimsà. The Jaina Canon refers to a large body of ascetics;
the Pàli Canon alludes to intensive speculative activity, and
disputes between sophists, agnostics and determinists; it
implies a background not unlike that of the oldest Upani-
sads. But the evidence is not sufficient to give us any clear
idea of the state of contemporary Hinduism; we are led to
the conclusion that it was in large measure tacitly accepted.
The Jaina critique was perhaps more emphatic, and cer-
tainly continued longer; we find examples of it in Sid-
dharsi, in Devasena, and as late as the tenth century in
Somadeva, who criticizes the sects and schools. But these
attacks had become in some measure a literary convention.
After the decline of Vedism, the official religion was at
first Buddhism, under the Mauryas. Nothing is known of
the Hinduism of this period; it is only when an ancient
Indian religion is adopted as the State religion that we
have any real knowledge of it, for from then on there is
evidence of it in public ceremonial, in inscriptions and on
coins, in monuments, and in court poetry and panegyrics.
It is said that a reaction took place under the Sungas, and
again a little later under the Andhras, because certain of
these kings restored the Vedic sacrifices. But this is a very
slight piece of evidence. My view is that the people of
India as a whole must always have been Hindu; they did
not have to be won back from Buddhism, which, in spite
of royal patronage, was from the start more or less confined
to monasteries and schools. It is true that royal protection
was accorded to Buddhism at several subsequent periods,
under the Indo-Greeks at the time of Menander, and
under the Kusànas, or at any rate under Kaniska in the
first or second century A.D.; in other words, under foreign
rulers. It is a pity that the Hindu dynasties of the pre-
 H I N D U I S M ! III 101
Gupta period did not pay such explicit testimony to their
faith. We have only the barest indications to guide us : the
inscription of Ghusundi, dating from about 150 B.C., men-
tions the construction of a building to the glory of Sam-
karsana and Vasudeva; there are the Saivite coins of
Gondophares a century and a half later; soon after this, the
title of mahesvara is assumed by Kadphises and there is a
Saivite coinage under Vàsudeva. But we cannot gather a
great deal from this, even when taken in conjunction with
the inscription of Heliodorus already cited, and the general
observations of Megasthenes on Indian beliefs of the fourth
century B.C. There are no grounds for assuming, as some
writers do, that Bhàgavatism flourished in Northern India
between the fourth and second centuries B.C., and was then
superseded by Saivism under members of the Kusàna
dynasty. We have slightly fuller inscriptional evidence of
the spread of Krsnaism in Central India from the first or
perhaps the second century B.C.
With the Gupta dynasty, in the early fourth century, the
position changes. This great indigenous dynasty was the
first to support the whole body of Indian orthodoxy; the
first Guptas called themselves paramabhdgavatas; they were
the patrons of the Brahmans and of Hindu communities;
mentions of schools and sects begin to be found. Moreover,
we know that under the Guptas Sanskrit, which had pro-
bably been for some time the language of administration,
came to maturity as a literary medium. This may have
helped to strengthen the Hindu faith, together with the
rise of an architectural style of non-Buddhist inspiration at
the same period. But all the same we must beware of talk-
ing of this as a Hindu renaissance, in the manner of scholars
in the past. No renaissance of Hinduism was possible be-
cause it never died out, or even diminished in strength; the
fact is simply that at this period conditions were so favour-
able to it that the traces it left behind are unusually con-
102 H I N D U I S M : III
spicuous. The development of Tantrism and the Sàkta
doctrines may be tentatively ascribed to the end of the
Gupta period. The testimony of Fa-Hi en and others seems
to indicate that the attitude to Buddhism was friendly; this
attitude is epitomized in the eclecticism of Harsa in the
first half of the seventh century. Nevertheless, the decline
of Buddhism, as attested, for example, by Hiuen-Tsang,
must have involved some violent feelings: the official ur-
banity of the great monarchies was one thing; the uncom-
promising attitude of the sects and local groups quite
another. But whatever the circumstances of the change
were, from this time onwards most of the dynasties are
Hindu; Vaisnavism predominates under the ancient Càl-
ukyas; then, from the seventh century onwards, Saivism
takes the lead, even under foreign rulers like the Hephta-
lites; their adherence to Siva has been explained by M.
Ghirshman as the result of a secondary identification of
Siva with Mithra. 1
The period from 700 to 1200, which Vincent Smith terms
the Ràjput period,2 also saw the Moslem conquest. In the
field of speculation, it is the time of the great commenta-
tors, chief among them Kumàrila and Sankara. To them is
due the final defeat of Buddhism, in a campaign inaugur-
ated in the South by the zeal of the Tamil 'saints', the
Vaisnavite Àlvàrs and the Saivite Nàyanàrs. At this period
Dravidian religion begins to assume characteristic form:
the conception of'saints' is developed; a certain religious
syncretism appears; and perhaps now also the first sects
emerge in the form of independent, clearly-defined com-
munities. In the South, too, the first religious persecutions
took place from the seventh century onwards, that of the
Vlrasaivas in the Telugu country and of the Srivaisnavas
1
R. Ghirshman, Les Chionites-Hephtalites (Cairo, 1948), p. 57.
2
Cf. The Early History of India, etc. (Oxford, 1904); The Oxford
History of India (Oxford, 1919).
 H I N D U I S M : III 103
in the Tamil country. This outbreak of fanaticism is
directed not only against the Buddhist and Jaina heresies
but on occasion against fellow Hindus : the Cojas, who
were Saivites, persecuted Ràmànuja, who had to take re-
fuge in flight. There are isolated cases where a man like
Vemana constitutes himself the violent opponent of ritual
and ofthe external show of the cult; Saivism at this time pro-
duced the influential doctrines of Siddhdnta in Southern
India, and Trika in Kashmir, though it is impossible to say
whether these systems were associated with active groups.
Saivism is also manifested in crude forms, such as the
Kàpàlikas, Kàlàmukhas, and so on, and the Vlrasaivas and
Lingàyats round about the twelfth century; Saivism pre-
dominates in the Deccan : the Jaina Somadeva, in the tenth
century, knows nothing of the Vaisnavite sects. Towards
the twelfth century bhakti is developed under the influence
of Ràmànuja, and in a more popular, or at any rate a more
picturesque form in the Bhdgavatapurdna, the date of which
is uncertain. The great Vedic sacrifices were rarely per-
formed, as Al-Biruni attests, but the authority of the Veda
was constantly reaffirmed, by Kumàrila and the com-
posers of the Nydya-Vaisesika on the speculative level, and
on the practical level by the Smàrtas, though there is no
agreement as to when and where the latter came to the
fore. Religious activity was centred round the great shrines,
at any rate in the South, and, as has already been pointed
out, the ninth century saw the establishment of the first
monasteries, which were to find the Saivite regions of the
South particularly favourable to their development.
The following period may be said to continue until the
end of the eighteenth century without spectacular change;
during this time religious commentaries followed close
upon one another in all branches of knowledge. The
Sankaran Veddnta in particular makes enormous strides,
and moves by imperceptible stages towards a doctrine of
104 HINDUISM.* III
theism; other forms of the Veddnta, said to be Vaisnavite,
emerge under Madhva, Nimbàrka, and Vallabha, among
others. The development of Tantrism continues, especially
in Bengal where it tends to merge with the Sàkta cult,
which was firmly established there from at least the four-
teenth century onwards. Bhakti is most commonly associ-
ated with Vaisnavism (Krsnaism), as it was from the
beginning, but from the fifteenth century at least it is also
associated with Ràmaism; wave after wave of masters and
poets contribute to its growth. The sects reached the zenith
of their development between the thirteenth and mid-
sixteenth centuries; after this period devotional smòrta be-
comes dominant or regains its power in many regions.
From the fifteenth century onwards, an important feature
is the syncretism of Kabir, and the rise of the 'reformed
sects', such as the Sikhs, which seems to have been due to
his influence. Religious literatures are developed, the
Marathi from the end of the thirteenth century, the
Bengali from the fourteenth century, and the Hindi from
the fifteenth century; the Dravidian literatures, the Tamil
hymns and panegyrics at least, go back much further. On
the question of Islamic influence, we have already men-
tioned the existence of mixed sects and of sects arising from
Sufi impulsion, but these have only a very limited following;
reference has also been made to the Islamic persecutions,
which resulted on the one hand in the defection of a con-
siderable number of Hindu adherents (especially those who
did not belong to active sects) from their ancestral faith, and
on the other hand in a strengthening of Hindu solidarity
and in a more rigorous enforcement of religious practices
and precepts. There were two main periods of persecution,
under Flruz in the fourteenth century and under Sikander
II and the Lodis in the fifteenth; the reign of Akbar in the
sixteenth century was marked by a period of tolerance, but
under Aurangzeb in the seventeenth century the attack
 HINDUISM*. III IO5
was renewed; it led to a peasant revolt, and the rise and
subsequent destruction of a semi-military sect, the Sat-
nàmis; Aurangzeb forbade the tonsure, closed down the
ghats and destroyed the shrines of Mathurà. The figures of
Sùrdàs and Tulsìdàs in the sixteenth and seventeenth cen-
turies epitomize the yearning for a consolidation of the
living forces of Hinduism in Northern India to meet the
threat of Moslem oppression.
The period from the eighteenth century onwards is often
characterized as one of decline, because of the falling off of
inspiration both in literature and doctrine, and the deca-
dence of the Sanskrit substructure and of traditional learn-
ing. Of course, this decline, if such it was, barely touched
the Hinduism of the people; it did not preclude local
revivals such as the Sàktist revival in Bengal, and the
upsurge of religious patriotism in the Marathi country.
But at this time it seemed likely that the influence of
Western civilization and the increased activity of Christian
missions, especially in the South, would eventually alter
the face of Indian religion.
But in the nineteenth century an important change took
place. The awakening of nationalistic feeling gave rise to,
or at any rate encouraged certain movements that favoured
a return to tradition and were in reaction against the
liberal ideas brought in from the West. Another factor of
still greater importance was the succession of men of strong
personality who came to the fore in various parts of India
and gradually attracted the attention of cultured circles in
all countries to what came to be known as the 'spirituality'
of India. In fact, the West, on discovering these men and
their achievements, endowed Indian spirituality with an
importance that was disproportionate, not to its intrinsic
value in the history of ideas, but to the extent of its influ-
ence in India and the numbers it affected.
The earliest of these men was a Bengali Brahman, Ram
io6 H I N D U I S M : III
Mohun Roy; he set himself the task of making his co-
religionists give up crude practices such as idolatry and
satz. Many previous teachers had made similar attempts,
but his plan of campaign was different: he sent out a
stream of skilful propaganda, tracts, newspaper articles
and reviews, for he had a great talent for public relations.
He deliberately adopted the Western scale of values, and
cited Christian theology to prove the necessity of mono-
theism, which he declared to have existed also in ancient
India. The group that he founded, the Bràhmasamàj, was a
partly social and partly religious organization; in some
respects it was not unlike the clubs of Western society.
Ram Mohun emphasizes the necessity of far-reaching
social reform; he seems to bring religion down to earth, as
do also his successors and imitators: Keshab Chunder Sen,
who, however, is too insistent in his desire to bring Hindu-
ism closer to Christianity; Debendranàth Tagore, who, by
contrast, intensified the traditionalist tendencies within
Bràhmoism, as it came to be called; finally Mahàdev
Govind Rànade, who emphasized social problems. The
succession of movements for religious reform, which has
continued right up to the present day, may be said to
derive from Ràm Mohun Roy. Gandhi and Tagore are
agreed on the necessity for bringing religion down to earth,
however much they may differ on other points; and the
'Ràmakrsna Mission', which we in Europe and America
know chiefly in its aspect as an organization for neo-
Vedàntin propaganda, is in India itself principally con-
cerned with educational and philanthropic work.
In the field of speculation, the writings and teaching of
Ràdhàkrsnan and Aurobindo Ghose may be considered as
attempts to modernize Hinduism and to utilize its living
power by adapting it to the needs of minds accustomed to
Western theology and philosophy. Ràdhàkrsnan eradicates
from Indian forms all features that are peculiar to them,
 H I N D U I S M ! III 107
and builds them up into a symbolism that can be adapted
to any form of religion. Aurobindo preaches a new syn-
cretism based on a reinterpretation of the religious mani-
festations of India, from the Veda to Tantrism; the key to
the future progress of humanity, he says, lies in India's past.
In pursuance of this theory, he propounds a new interpre-
tation of the great Sanskrit texts of antiquity, and at the
same time develops new methods of Toga in conformity
with modern trends, in his dsrama at Pondicherry. Before
him, Vivekànanda at the end of the last century had
already attempted a re-interpretation of the Veddnta; using
public oratory as his medium, he inculcated the theme of
the unity of all religions, and spoke on this subject in per-
son at the Congress of Religions in Chicago.
The example of Ràmakrsna may perhaps seem more
striking to Europeans; he was a professed ascetic, with ex-
ceptional powers of mysticism; he was not a highly edu-
cated man, but he was firmly resolved to prove by per-
sonal experience that it was possible to attain mystical
power through the medium of any form of the divine; and
thus he prepared the way for a universal form of mysti-
cism.
Some of these movements have been carried to extreme
lengths. Keshab Chunder Sen, as we have seen, tried, to
christianize his sect; mention must also be made of Bài
Gangadhàr Tilak, the Marathi politician and philologist,
whom his admirers call the father of Indian nationalism
(he has also been called the father of Indian unrest) ; his
passionate devotion to Hindu traditions led him to indulge
in some of the most fantastic speculations that have ever
appeared on the origins of Vedic civilization. There was
also Dayànanda SarasvatI, a Gujarati ascetic, who founded
the Àryasamàj : he was in favour of returning to an un-
qualified adherence to the Veda, and claimed that explicit
principles of pure monotheism and of social and moral
 io8 H I N D U I S M ! III
reform could be found in the hymns. As well as his work of
religious reform and his eminent achievements as an exe-
gete, he was active in urging that Hinduism must establish
within itself a Church militant that should work unceas-
ingly for political and social reform.
Rabindranath Tagore is probably the man whose work
is best known in the West, and whose reputation seems
most likely to last there. He too was a profoundly religious
man, but he was without fanaticism; his humanistic cul-
ture made his religion enlightened and gentle. The syn-
thesis he advocates, though today it is temporarily neg-
lected, is still the one most likely to succeed, for it does not
exact adherence prior to personal conviction. In Tagore's
nature humaneness and poetic feeling are harmoniously
united : in him the poet is at one with the religious thinker,
the educationalist and the politician. Although he is a
representative figure of the new India, he also exemplifies
tendencies common to the most enlightened members of all
communities. It is through him that great poets through-
out the world are conscious of an intimate affinity with
India.
I have neither the knowledge nor the inclination to pro-
phesy what the future of Hinduism will be. Unless there is
a radical social upheaval (which is within the bounds of
possibility), it seems unlikely that the Hinduism of the
people will greatly change; we have seen it to be inherent
in India's evolution and in all the principles that have
grown up around religion from time immemorial to form
an indissoluble unity. People all too easily accuse religion
of having contributed to economic and social depression.
In many minds religion is automatically associated with
privilege of class and caste, a fact that also tends to pre-
judice the survival of Sanskrit culture. If Hinduism can
avoid this aspersion, it may be able to gain new vitality by
enlisting the more enlightened classes in its service; for this
 H I N D U I S M ! III 109
it will need the leadership of a few men of strong person-
ality to organize a movement of supporters and sympa-
thizers. This will be an opportunity to see whether the
machinery of a sectarian organization, which was so fre-
quently and easily called into service in ancient India, is
still capable of effective action, whether it is sufficiently
flexible to avoid heightening the 'communalistic' ten-
dencies which many Indians deplore.
Some people think that Hinduism should cease to be
ethnical in character (assuming that it ever has been so),
and become once more a missionary religion. There are
already several organizations for spreading a knowledge of
Hinduism in the West, but very often their propaganda
does not reach the right circles. When Hinduism is
'exported', it tends to be regarded as a kind of theosophy
—after all, the basic doctrinal principles of theosophy are
rooted in Hinduism—or as a brand of Christian Science,
tinged with pseudo-Vedàntism. It can only become a force
for good in the world when it emerges in India itself as a
purified form of religion, free from primi tivism and the cult
of images. Extreme practices, such as Hathayoga and Tan-
trism 'of the left5, which often make such a deep impression
on Europeans, never constitute the main strength of a
religion; they are special features that should not be imi-
tated outside the land of their origin.
As many Indians admit, Hinduism is greatly in need of a
priesthood. The fact that there are no seminaries and, with
a few exceptions, no monastic orders, has often been
deplored. Indians observe, not without envy, that in
Britain, for instance, many able men become ministers of
religion on leaving the University, whereas in India there
are no official religious leaders, but only solitary devotees
working out their own salvation and preoccupied with an
inward mystic revelation which they cannot communicate
to others. If one wishes to get in touch with someone who
 no H I N D U I S M : III
officially represents Hinduism, it is very difficult to know
whom to approach. The need for a Church is perhaps felt
more acutely today than it has been in the past.
The troubles of the present age, which are rightly or
wrongly attributed to Western materialism, have helped to
increase the prestige of Hinduism. Some people see it as
the authentic survival of a tradition, or rather, of the one
Tradition, and make it the basis of their philosophia
perennis. Others try to incorporate it in a universal religious
syncretism. Whether these attempts will succeed must be
left for the future to decide. The fact remains that Hindu-
ism provides an incomparable field of study for the his-
torian of religion: its aberrations are many, but there is in
it a great stream of mystical power; it manifests all the
conceptions of religion, and its speculation is continually
revealing them in a new light. It combines powers of con-
stant renewal with a firm conservancy of fundamental tradi-
tion. In bhakti, and still more in Toga, it has perfected un-
rivalled techniques of mystical initiation, that contrast
strongly with the frequently haphazard methods of spirit-
ual training in the West. Above all, in the interpenetration
of religion and dharma in general and the reciprocal stimu-
lus of abstract thought and religious experiment, there is an
underlying principle that, given favourable conditions,
may well lead to a new integration of the human person-
ality.
 VI
Jainism

J AIN ISM has by no means aroused as much interest as

Buddhism and Hinduism. It is practically unknown to
the general public, despite the scholarly works that have
been devoted to it, notably in Germany, but also in Great
Britain and Italy. But if in the West there is little research
at present being carried out in this field, in India, by con-
trast, where Buddhistic studies have never been very popu-
lar, the study of Jainism, especially of the post-Canonic
period, has been widely pursued. Numerous societies have
been formed to further Jainist studies, and collected edi-
tions of the texts have been issued by various bodies.
It must be admitted that Jainism does not at first sight
seem to present the same general interest as Buddhism; the
personality of its Founder is not so compelling; the texts
are difficult, puzzling, and, frankly, tedious, as far as the
non-specialist can judge them from translations, which are
comparatively few in number. It is a religion of austere
aspect, that might be described as Buddhism's darker
reflection.
It was, in fact, its resemblance to Buddhism that most
interested those who first investigated it. Several of these
scholars held the theory that the Jainas were a sect that had
separated from the early Buddhists. Others, basing their
opinion on the fact that the Jainas could claim a slight
chronological priority and that they even had some form of
prehistory, considered Buddhism as an offshoot of Jainism.
Both of these movements are reformations directed
112 JAINISM
against Brahmanism, especially against the ritualistic
aspect of it which predominated at the period, i.e. the
sixth and fifth centuries B.C., when Vedism was on the
wane. Their attack is, however, confined to the religious
aspects of dharma, the Hindu Law; they accept the social
order, or, at any rate, do not openly revolt against it. Both
draw largely on the Hindu substratum for their teaching
and the general framework of their systems.
One of the most important divergences between the two
movements concerns the Master who originated the doc-
trine constituting the Jaina Canon, a doctrine that was at
first oral in form, and was later written down. He is called
the Great Hero or Mahàvìra, or sometimes the Victorious
One or Jina (a name that is also occasionally bestowed on
the Buddha), and he is described as the last of a sequence
of prophets or patriarchs, stretching back to the remote
past. Of course, ancient Buddhism also recognized earlier
Buddhas, the obscure precursors of Sàkyamuni, and the
conception is more clearly formulated in the Great Vehicle
with its doctrine of innumerable past and future Buddhas.
But no definite dates are assigned to them, and there is no
attempt at building up their lifé-histories.
In the Jaina system, then, we find the Tlrthamkaras, or
'forders' (a metaphor analogous to that of the Latin ponti-
fex: the Romans use the image of a bridge, the Indians that
of a ford). They are a coherent group of twenty-four men,
who have attained Omniscience, and devote themselves to
guiding humanity towards the true path. It should be
noted that Buddhism also recognizes a group of twenty-
five Buddhas, which could be taken as implying that
Sàkyamuni was the culmination of the sequence of twenty-
four forerunners described in Jaina tradition.
The conception of the Tlrthamkaras is an ancient one;
the iconography of Mathurà attests that it was current in
the first century A.D. Of course, these legendary lives con-
 JAINISM "3
form to a set pattern. The general outline is in accordance
with a tradition that is automatically followed in descrip-
tions of founders of sects in India: princely birth, sudden
renunciation of the world as the result of 'illumination',
and the founding of a community or body of disciples, one
of whom becomes the successor. Of the symbolic details,
some are identical with Buddhist forms (e.g. the tree sacred
to each of the patriarchs), and some with Hindu forms,
such as the associated animal, which is comparable with
the 'mounts' of the Hindu divinities. The latter feature is
post-Canonic, and so also is the conception of auxiliary
divinities, frequently referred to in Buddhistic tradition,
the Yaksas and Yaksinis, who act as intercessors. An
element of fantasy, fully consonant with the Jaina passion
for numbers, appears in the matter of the great ages to
which the Tirthamkaras live: each one has a slightly
shorter life than his predecessor. This feature was sug-
gested by the Hindu avatdras. The Cakravartins or Univer-
sal Rulers, who are twelve in number, each one allegedly
contemporary with two Tirthamkaras (and who are also
found in Buddhism with different associations), must surely
have some connection with the Manus of Brahmanical
theory, who are fourteen in number. Finally, the lives of
the Tirthamkaras are but elaborated versions of the life of
Mahàvìra, the last of the sequence: not only is the general
pattern the same, but even the details concerning the birth
are identical, such as the fourteen dreams that come to the
mother of each Tlrthamkara at the beginning of her
pregnancy.
A slight variation in tone is perceptible, however, when
we come to the twenty-second in the series, Nemi. Not that
we approach historical fact (things do not progress as
quickly as that in India) ; but his legend is more circum-
stantial than those of his predecessors; as in the case of the
Buddha, there is a description of previous births. Even
114 JAINISM
more remarkable is the traditional contemporaneousness
of Nemi and the divine hero Krsna. It has been pointed
out that of the sixty-three 'great men' ofJaina prehistory,
no less than twenty-seven had associations with Krsna;
Krsnaism seems to have left its mark on Jaina legend, a
Krsnaism which we must assume (as we are frequently led
to assume in the case of other Indian manifestations) to be
an earlier form than that described in the Brahmanical
texts. It has been suggested, somewhat daringly, that the
legend of the Buddha is a reflection of that of Krsna; if the
theory of a Krsnaite association is to be retained, I should
be more inclined to transfer it to the field ofJainism.
With the twenty-third Tlrthamkara, Pàrsva, we are
getting near to the historical period, if any meaning can be
attached to the expression when dealing with this subject.
It is a significant fact that the figures quoted in connection
with him have been reduced to almost normal proportions.
We are told that Pàrsva died 250 years before Mahàvlra,
for it is suspected that he founded a doctrine from which
his successor drew his inspiration and which he 'reformed'.
Mahàvlra's own parents are said to have been lay followers
of Pàrsva. It must be admitted that a multitude of fanciful
suppositions have grown up around Pàrsva. But all the
same most scholars recognize him as the real founder.
Mahàvlra was a great preacher who drew the crowds, but
his contribution to doctrine seems to have been small. We
are the more readily disposed to give credence to this
theory because we are dealing with Indian material, and
in India originality is far less highly valued than fidelity to
tradition; but none the less it is possible that future research
will show Pàrsva's role to have been less significant. How-
ever this may be, Mahàvlra seems to have developed the
ethical aspect of Jainism by introducing a fifth axiom which
brought about a modification in the import of the fourth.
He is said to have instituted the practice of confession.
 JAINISM 115
Finally, it was he who required his monks to dispense with
clothing, setting the example himself, whereas Parsva's
monks were clothed. The point may seem unimportant,
but external details of this sort have always greatly exer-
cised Indian sects; the problem of nakedness has in fact
persisted in Jainism, and, as we shall see, has been the
cause of a great schism that continues to the present day.
We now come to Mahavlra, who is still called Var-
dhamàna, 'the growing one', an epithet of Vedic origin.
Like Krsna, he has a twofold birth, one in a Brahman
family in Bihar—Bihar is the classic homeland of reli-
gious movements in ancient India—the other in a noble
family, that of a clan chief in Magadha. In this way the
sympathies of the two highest social classes are enlisted,
albeit at the expense of physiological vraisemblance. Pelliot
thought that there were too many kings' sons involved in
these doctrines of renunciation.1 True, but after all who,
according to the Indian way of thinking, should renounce
the world, if not those who have tasted all it has to offer?
The first coincidence emerges from a comparison of
Jaina and Buddhist sources : the brother of the princess of
Magadha, who was the mother of Mahàvìra, has a son-in-
law, Seniya, who is none other than the Bimbisara of
Buddhist tradition, the father of Ajàtasatru, whom the
Jainas call Kuniya.
Taken as a whole, Mahavlra's life closely resembles the
Buddha's. The general postulates, the pious aims of their
biographers, the times and places are the same. Miraculous
elements are not wanting, and are especially in evidence in
accounts of the Master's birth; but even in recently com-
piled sources there is nothing comparable to the prodi-
gality of Buddhism. The Lalitavistara, a poetical amplifica-
tion of the theme of the Buddha's diversions, has no Jaina
1
Cf. his preface to the French translation of H. v. Glasenapp's
Brahma und Buddha (Berlin, 1926; French version, 1937).
 u6 JAINISM
counterpart. The typical ancient biography, the Life of the
Jina by Bhadrabàhu, would find its closest parallel in
Aévaghosa's Life of the Buddha. The predestined child is
brought up in princely splendour; he marries a girl of
noble birth, a marriage not recognized by the more puri-
tan of the two later branches of Jainism. By her he has a
daughter who later marries. He leaves his family when his
parents die (here there is a contrast with the Buddha) ; he
enters upon a life of meditation and spends two years seek-
ing the path; then he becomes a naked ascetic, vowed to
the wandering life which he leads for twelve years. The
element of asceticism is much more strongly stressed than
in the case of the Buddha, who gave up the wandering life
after six years.
In the thirteenth year he attains Omniscience; the reve-
lation comes to him at the foot of a sàia tree, which is com-
memorated in the texts as a sacred tree (caitya). From that
time onwards, the Master preaches the Law, tirelessly
travelling round the Magadha-Anga-Videha region, train-
ing disciples, but also facing many demonstrations of hos-
tility. He dies at the age of seventy-two, in the reign of
Ajàtasatru, in a place called Pàvà, that the Jainas are pro-
bably wrong in supposing to be different from the Pàvà
through which the Buddha was to pass soon afterwards in
the course of his last journey.
The correspondences between Mahàvlra's life and that of
the Buddha must spring from a common, pre-existing
legendary source. Some of the common elements might
just possibly have been derived from a third sect, that of
the Ajivikas, which seems to have exerted a noticeable in-
fluence on the other two. The leader of this sect, Gosàla
Mankhaliputra, is said to have been a disciple of Mahàvìra
for six years, after which he left him, attained the state of
Jina and died sixteen years before Mahàvìra. He enjoined
the practice of severe asceticism, especially in the matter of
 JAINISM 117
food, and developed the theory of karman (which he may
have originated) ; this theory he carried to almost impos-
sible lengths, denying free will and man's responsibility.
Western scholars date Mahàvira's death in 467 or 477
B.C.; tradition puts it in 527. In any case it is after the
Buddha's death, whether one accepts the date traditional
in the South, 543, or the amended date, 483. In spite of
this, a Pàli text states that the Jina died first; it probably
suited the writer's purpose that the Buddha should survive
longer.
Buddhist sources are scarcely more informative about
the Jainas than Jainist sources are about Buddhism. They
do mention the community of the Nirgranthas, 'those who
have cast off their bonds', and their leader, whose name in
Pàli is Nàtaputta, and who has been plausibly identified
with Mahàvira. But they say virtually nothing about doc-
trine. It is difficult to realize how completely contem-
porary movements in India could be unaware of each
other.
In spite of the difficulties and contradictions in the texts
we are perhaps better informed about the early days of the
Jaina community than about those of the Buddhist Church.
We know that there were nine groups among the first
disciples, with eleven group-leaders or heads; we know the
names of those who recorded or transmitted the Master's
words, the Theras or Ancients. The Jainas were acutely
aware of the progressive attenuation of this knowledge;
only the first two disciples are kevalins, full possessors of
knowledge; and the history of Jainism has been charac-
terized by the effort to retrace the path and rediscover the
original sources of knowledge.
The most remarkable of these ancient religious leaders
is Bhadrabàhu, the sixth Thera. He is the second most
important figure in Jainism. He was contemporary with
the first Maurya, and lived, therefore, in the fourth cen-
n8 JAINISM
tury B.C.; it fell to him to take the initiative in the famous
migration to the South which was a flight from imminent
famine. The story may have been invented subsequently
to explain the Jaina colony in Mysore, and to give the sup-
port of orthodoxy to the doctrinal developments brought
about by the exodus. At all events, it is suspicious that two
versions of the story are found; according to the second
version, the event took place under the principate of
Sthulabhadra, a disciple of Bhadrabàhu.
It is recounted that when these emigrants, or some of
them at least, returned to the North, they found that there
had been a relaxation of religious observances. In India,
the people of the South have always been rather strict in
such matters. In this way there arose a state of affairs con-
ducive to schism, which, however, did not occur until
much later, in 79 B.C. This seems to be the last of the seven
heresies described in Northern tradition, the earliest of
which was originated by Mahàvìra's own son-in-law, and
arose in his lifetime. But the schism of 79 B.C. was a pro-
found one, in no way comparable to the imperceptible im-
pregnation of primitive Buddhism by the Great Vehicle.
The Southern group consists of the Digambaras : as the
name indicates, they are those who remain faithful to
Mahavira by wearing no clothing, while the other branch,
the Svetàmbaras, resume, or perhaps never abandon, the
white garment, more suited to the Northern climate. It is
perhaps going a little too far to identify these Svetàmbaras
with the adherents of the old Pàrsva sect, on the presump-
tion that the movement had survived through the cen-
turies.
The austerity of their habits matches their doctrine. In
Europe (and in India too, I fear) little is known of the
ancient Digambaras. We do know that they repudiated
the Canon as we have it today, holding that after the death
of Jina the old texts had disappeared stage by stage and
 JAINISM II9
had been replaced by new ones. They therefore set up a
kind of substitute Canon, a collection of Tour Vedas',
which they themselves, presumably, do not claim as
authentic.
Let us now consider the very important question of the
Jaina Canon. The first Council, that of Pàtaliputra, was
held under Bhadrabàhu, in the fourth century B.C. Like
most Councils, its chief aim was to collect the texts into one
body and to define the extent of the authentic scriptures.
In this it resembled the Buddhist Councils, especially the
second Council, at Vaisall, which also took place in the
fourth century; its object, according to Southern tradition,
was the establishment of the Canon, but it is far more pro-
bable that its real purpose was to combat the Mahàsàm-
ghika schism, just as the Jaina Council was really directed
against the Digambaras.
However this may be, the J a i n a Scriptures were begin-
ning to fall into a corrupt state; it was already too late to
save the twelfth and last of the Angas, or basic texts, in its
entirety. The Jainas, in fact, have a tradition that the
Canon does not wholly derive from Mahàvìra's teaching.
Part of it is thought to have been drawn up by Bhadrabàhu.
As late as the fifth century A.D., the Council of Valabhl
made a final compilation, their object being to establish
the form of the texts after the secession of the Digambaras:
the version adopted at Valabhl is the one we have today, a
version established nine centuries after the Master's life-
time. But the composition of the collections themselves
must date back much further; the tradition that the main
work of elaborating the Canonical Scriptures took place in
a short period of two centuries, between the death of Vira
and the time of Bhadrabàhu, may not be far from the
truth. The fact that the Digambaras reject this tradition
must not be allowed to prejudice the facts; and it will be
the task of future scholars to pursue this problem by means
120 JAINISM
of textual criticism of the Canon itself, according to the
method so brilliantly demonstrated by Schubring. 1
We cannot hope to know what the very earliest form of
the Canon was. It may have been found in its entirety in
those pUrvas, 'prior' or 'primordial' texts, which were lost
after their substance had been incorporated into the twelfth
Anga, a work that was also lost at a later date.
In spite of all these vicissitudes, it must be admitted that
the J a i n a Canon gives an impression of greater antiquity
than the Buddhist Canon. The work as a whole, which is
also called the pitaka or 'basket', is arranged less system-
atically; it contains some independent texts; the interpola-
tions are more easily recognizable as such. Its 'philo-
sophical' portion and the Pàli Abhidhamma cover no com-
mon ground. Finally, there is only one single tradition,
nothing of the diversity of the Buddhist schools. O n the
other hand, we note that the Jainas, being more concerned
with technology than the Buddhists, have included treatises
on cosmography, mathematics and other semi-secular
matters, after the fashion of the Brahmanical Veddngas.
Despite internal dissensions, the influence of the com-
munity seems to have spread fairly rapidly. The migration
to the South, if it really took place, must have been the
origin of a vast group that still exists today, but which is
not so numerous as the Northern groups. The latter first
spread to Orissà (where the Canon seems to have attracted
the interest of King Khàravela), and then to Bengal; and
they early reached the North-West, where the extraordin-
arily rich finds at Mathurà are evidence that a community
flourished there at a very ancient date.
The periods when Jainism enjoyed royal protection
1
In nearly all his works, from the earliest, Das Kalpasiitra (Leipzig,
1905), down to the most recent, Studien zum Mahdnisiha, by F.-R.
Hamm and W. Schubring (Hamburg, 1951). Cf. also his comprehen-
sive study, Die Lehre der Jainas, nach den alten Quellen dargestellt (Berlin
and Leipzig, 1935).
 JAINISM 121
were those of its greatest activity; when such favour was
withdrawn, its influence was greatly restricted. This is true
of Buddhism also. In the South, the most flourishing period
was under the Ràstrakùtas in the ninth century; in the
twelfth century a decline set in, and from then on we find
evidence of persecutions, first under the Southern Gur-
jaras, and later under the Pàndyas. As in the case of
Buddhism, the influence of Ràmànuja and the Srivaisna-
vas brought about wholesale conversions to Vaisnavism.
Under the Còlas carne the more violent attack of the
Virasaivas, who destroyed temples and archives and har-
rassed the commercial classes, the mainstay of the whole
community. The Jainas have been more severely affected
by economic vicissitudes than any other sect. Finally, we
must bear in mind the effects of Moslem oppression, both
in the North and in the South, though they were, of course,
more pronounced in the North.
In the North, the Gujarati country became the principal
centre. A decline set in in other areas, especially in those
Eastern regions where Jainism had originated. Indian reli-
gious movements tend to become established far from their
original homeland, where they leave little trace. More-
over, some of the Gujarati Kings were interested in J a i n a
art, as was Kumàrapàla in the thirteenth century, for
instance, the patron of the famous Hemacandra who set
himself to transform the kingdom into a model Jaina
state.
In the last analysis, the progress made balanced the set-
backs, and the community has always been quite firmly
established. It forms a strange contrast with Buddhism,
which was far more spectacularly successful in its origins
and which early enjoyed imperial support. I do not believe
that one movement suffered more than the other from
persecution; but it may be that Jainism adopted a more
compliant attitude towards the civil power, and that its
122 JAINISM
opposition to Hinduism was less pronounced. The chief
cause, however, is to be sought elsewhere. It was the
strength, and yet at the same time the weakness of
Buddhism that it drew its adherents largely from among
the poor, whereas Jainism turned to the rich and influen-
tial. Again, the monasteries are an essential part of
Buddhist life; if the monasteries are destroyed, the blow is
mortal. In Jainism, the lay community has a far greater
relative importance; it plays its part in the administration
of its religion and in the cult. Jainism is in fact just what it
claims to be, a fourfold Church, a fourfold ttrtha, composed,
that is to say, of monks, nuns, and male and female lay
followers. Another advantage is the great richness of the
extra-Canonical literature, which greatly surpasses that of
the Buddhists, in its variety of genre, at any rate; these
works, moreover, are to be found in all the local languages,
Tamil, Kanarese, Gujarati, to say nothing of Sanskrit and
Middle Indian : all these factors contribute to the strength
of the Jaina position.
Finally, the very stability of the doctrine may have con-
tributed to its survival. It never underwent any process
comparable to the general re-assessment of values atten-
dant on the transition from the Hinaydna to the Mahd-
yana. As we have seen, it suffered a violent schism that
resulted in a wide gulf separating two groups; and if we are
inclined to think the cause trifling, we must remember that
nothing is trifling where religious matters are concerned.
But nevertheless, on both sides of the gulf the same doc-
trine, ethics and philosophy were retained, and this scission
proved to be the only one.
Of course, as always happens everywhere in India,
many sects arose; among the Northern Jainas there are
said to be eighty-four sects, called gacchas. Tradition main-
tains that several of them go back to Pàrsva. It is also said,
more credibly, that they arose in the tenth century. Among
 JAINISM 123
the Digambaras, the divisions seem to have taken place
earlier, but they are fewer in number : only four groups or
ganas are found. Each group possesses genealogical tables
of its own, recorded in special texts which are not infre-
quently confirmed by inscriptional evidence; like the
Buddhists, and in contrast to the Hindus, the Jainas have
a strong historical sense. But the sectarian divisions do not
indicate the slightest doctrinal divergence. These Jaina
groups are not so much sects in the Hindu sense (for these
are often aggressively insistent in their philosophical claims)
as brotherhoods, rather like our monastic communities.
The Sthànakvàsls of the eighteenth century are not in
quite the same position : they are indeed a reformed sect,
who, rejecting image-worship and temple services, seek to
return to the ancient form of Jainism. Movements advo-
cating a return to the past, whether they make a lasting
appeal or not, are recurrent in Hinduism; there is no need
• to invoke Islamic influence to explain the occurrence of
such a slogan. But the Sthànakvàsls go further, by also
rejecting part of the Canon. Yet it is not certain that in
India even this would be sufficient to constitute a heresy.
Moreover, there had been a sixteenth-century precedent
in an analogous movement, that of the Lumpàkas, which
seems to have come to nothing.
Reforming zeal constantly manifests itself; and founders
of new sects quickly pass into oblivion in India. Four years
ago, I paid a visit to the sect of the Teràpanthls, who were
holding their annual assembly in a country town in Ràjpu-
tànà. This group, whose name means 'those who follow the
path of the Thirteen', was founded in 1760 by a Sthànakvàsi
layman called Bhikanji. Bhikanji was motivated by the
conviction that his co-religionists had fallen away from the
primitive customs of the movement. He wanted to return
to these early ways, and founded a group of which he
became the first pontiff; the present pontiff is the ninth of
 124 JAINISM
the line. To be present at a gathering of the Teràpanthls
is to have experience of the life of a sect which makes every
effort to return to the precepts and customs of the time of
Mahavlra. The scene is pleasantly picturesque, with its
rustic setting, and its central dais decorated with banners
of naive design; in the assembly the monks and nuns are
carefully segregated from each other, and from the crowds
who have come out of curiosity, though they too are atten-
tive and devout in appearance. The alternating chant of
the religious, with the crowd taking up the refrain in chorus,
provides an interlude in the sermon, preached in the
Master's vibrant tones; as he preaches, his face has the
fervency of an apostle; he is like a Mahavlra redivivus. The
rule of the Teràpanthìs is extremely severe; for these wan-
dering ascetics must renounce almost everything and beg
their scanty food all their lives. It is not unusual to see one
of them (as I have) freely choose to die in the way charac-
teristic of the Jainas, ending a life of austerities by abstain-
ing from food altogether. Nevertheless, it is a way of life
that many aspire after eagerly; postulants beseech the
Master to admit them to it; relations and friends add their
entreaties on the applicant's behalf. Such is religious
fanaticism in India, where the secular life is accounted of
little value in comparison with the rewards won by follow-
ing the path of mysticism.
In order to trace in detail the history of the Jaina Church
it would first be necessary to make a critical examination
of the enormous mass of literature, partly narrative, partly
moral in intention, that the Jainas, the most prolix of the
Hindus, have poured out in the course of centuries. In
some cases they have falsified historical fact in their eager-
ness to convert princes and high dignitaries; thus, they
allege that Candragupta Maurya became a Jaina and
ended his life as an ascetic in Mysore, among the commu-
nity that had migrated there. Some sources claim as patrons
 JAINISM 125
of Jainism the Emperor Asoka, the legendary Vikramà-
ditya, the Hephtalite king Toramàna, and many more.
We have already mentioned the conservative tendencies
of Jaina thought. It is true that the later literature carries
us a long way from the sources in the luxuriant imagina-
tion it displays. The developments in iconography, the
establishment of the great shrines and the public worship
that grew up with them, all these factors seemed to com-
bine to deflect mediaeval Jainism into new paths; but
despite all this, its great moral and philosophical principles
have remained stable, and religious feeling has probably
not altered greatly. If we turn to Hinduism for a parallel,
we have only to ask ourselves how much real effect the
frenzied Sivas and grimacing Kalis of common representa-
tion produced on Sankara and the brotherhood he
founded.
The fact that there are Brahman priests or pùjàris em-
ployed in the service of the Jaina temples emphasizes the
lack of a secular clergy in religions of a monastic type like
Buddhism and Jainism. Any features taken over from
Hinduism have taken on a Jainist aspect; one has only to
consider the way in which Indian heroes are treated in
Jaina legend. Tantrism, which corrupted other systems of
thought, Buddhism included, produced only a limited
effect on Jainism, manifested by the sudden appearance of
many female divinities, long-delayed descendants of Kali
and Tara.
The monastic rule was the Jainas' greatest creation. It
is a severe rule, dominated by the conception of non-
violence, ahimsd, a conception which the Jainas may per-
haps have originated, in their attitude of reaction against
Vedic Violence'. The monks and nuns are called bhiksu
and bhiksuni, as in Buddhism. The five monastic vows,
designed for the attainment of ataraxy, are negative ones,
as are the five yamas or abstentions of Toga, with which
126 JAINISM
they are identical: not to do violence, not to lie, not to
steal, not to have sexual intercourse, not to have posses-
sions. The first four correspond to the sllas or the siksdpadas
of Buddhism, which are usually expressed in expanded
form in ten precepts. But their real provenance is probably
an ancient code which both the Brahmanical samnydsins
and the bhiksus drew upon. Jaina asceticism grew up out of
a background of pan-Indian, or perhaps pre-Indian,
asceticism, which can be traced also in Buddhism, though
Buddhism early repudiated it. What are the dhutdngas,
those practices of the Pàli Canon appropriate to the dhuta
(called the avadhiita in Brahmanical asceticism) but sur-
vivals of an ancient code of asceticism which must have
been abolished or suppressed in common usage? Primitive
Buddhism was probably closer to Jainism than the form
revealed in the Canon. Not only did it lay more emphasis
on asceticism, but it must have prescribed that regular
alternation between the wandering and the sedentary life
that preceded the institution of permanent vihdras; this
alternation is clearly recognized in the Jaina Canon, but
the Pàli Canon regards it as a practice borrowed from what
it calls the titthiyas or heretics. This system, again, was pro-
bably modelled on the cycle that governs the life of the
brahmacdrin, who enters once more upon a period of study
when the rainy season sets in.
The precept of non-violence manifests itself in forms that
remained peculiar to Jainism, and which may perhaps
appear somewhat naive in our eyes—the broom made of
white wool with which the devout sweep clear the path
they are about to tread, and the mask placed in front of the
mouth so that no living matter shall be breathed in.
The Jaina rule is characterized by severe and exacting
prescriptions. The day is divided into four periods, and the
way in which the time is spent is strictly regulated. Rules
are laid down for every detail, even to the amount of food
 JAINISM 127
that the monk may take daily (thirty-two mouthfuls of the
size of an egg), and the conditions under which he may
accept lodging for the night, for he must take his food in a
different place from that in which he spends the night. The
practice of austerities is carried to great lengths : physical
asceticism is practised by keeping the body in unnatural
positions, and especially by fasting: total fasting for rela-
tively short periods and partial fasting, which assumes
various forms, and may last for as long as 522 days. More
importance was probably attached to fasting in the primi-
tive form of Buddhism; but the uposatha (the term that
describes it, and which is itself a borrowed word) is really
only a day of abstention from work and an official festival,
like the Vedic upavasatha that took place on the eve of some
of the great ceremonies.
Mental asceticism consists of progressive exercises in
concentration, by which a higher state of consciousness,
that of kevalin, may be attained in fourteen modes. The
same word kevalin is used to denote Liberation according
to the Sdmkhya, and the technique bears an obvious rela-
tionship to Toga. The fourteen modes, strangely enough,
are not successive stages following one another in time;
they are like a keyboard over which the spirit ranges,
moving up or down as the nature of its actions fluctuates.
An extreme form of asceticism takes the form of committing
suicide by abstaining from food. This is called samlekhand,
a term that suggests the idea of self-inflicted suffering, and
which, according to the texts, properly denotes the reced-
ing of the sensible world and of sensation that is the pre-
liminary to a death-fast. The term is known in Buddhist
tradition, in which it designates certain severe macera-
tions. It is also to be compared with the samlikhitam of the
Atharvaveda, which is applied to the ruined gambler, a man
who has been 'completely cleaned out', as we say in fami-
liar speech. One's impression is that suicide by fasting,
128 JAINISM
which is certainly not unknown in Buddhist tradition and
in Hinduism in general as far back as Vedic times, has
been promoted by the Jainas to the status of a religious
institution; the inauguration of the fast is regarded as a
sacrament, and is accompanied by a re-affirmation of
vows, solemn renunciation, and so on.
Another characteristic practice is that of public con-
fession. This is a feature that Jainism shares with Buddhism,
and the Jaina term pratikramana, which means 'return',
'way back', is not far removed from the Pàli term pdtide-
saniya, denoting the sins to be confessed, or literally 'to be
related by turning back'. In Jainism, confession of a sin
committed at night takes place in the morning, and that of
a sin committed during the day, in the evening; a more
solemn confession takes place at the end of a fortnight, as
with the Buddhists. The man making the confession
acknowledges his faults, becomes spiritually purified, and
expresses his desire for improvement. It seems likely that in
the mechanical way in which it functions, and in the
power of completely wiping out karman that is inherent in
it, the J a i n a confession represents an earlier evaluation
than the more refined Buddhist conception, in which
repentance plays an important part. This distinction is in
accordance with the general opposition between the auto-
matism, or perhaps it might be called the realism of the
Jainas, and the Buddhist preoccupation with ethical values,
which represents a further stage of reflection. The last day
of the period of wandering life is marked by a general con-
fession, which thus terminates the active part of the reli-
gious life and also coincides with the end of the Jaina year;
this confession is the counterpart of the Buddhist confession
that marks the end of the rainy season, and constitutes a
feature of what is known as the pravdrand or 'closing'.
Ten degrees of expiation are prescribed in Jainism;
firstly, there are corporal punishments, which resemble the
 JAINISM 129
voluntary practices of asceticism; then comes a 'cutting
down' of the monk's seniority, which entails beginning
again from the 'root', i.e. from the initiation; and lastly,
expulsion from the community. Parihàra is a kind of
segregation of a monk who is being punished or arraigned,
and is comparable to the Buddhist parivdsa, which is a pro-
bationary period of supervision. Transgression is known by
the Vedàntic term maya, a veil obscuring true knowledge.
So much for the monks. As for the laymen, it is amazing
to see how detailed and rigorous were the rules laid down
for them in ancient times. Confession is in principle obli-
gatory for them, and they are allowed to undertake the
death-fast. They have twelve vows to observe, whereas the
monks have only five; this is no doubt because secular life
is more varied than monastic life. They too have a mystical
scale, comprising eleven stages (pratimd), by which they
can attain to a state as meritorious as that of-the monks,
and which is perhaps more difficult of attainment by
reason of the worldly temptations that beset them; on
reaching the eleventh stage, the layman is, in fact, a monk.
Here there is a noticeable contrast between the Jaina and
the Buddhist viewpoints, for Buddhism, at any rate in Pàli
tradition, devotes relatively little attention to the updsakas.
But admittedly, it is on this point that the Jaina attitude
has most relaxed its severity between early times and the
present day.
Other aspects of this religion are no less highly deve-
loped. Its cosmography, for example, is founded on the
same principles as that of Buddhism and Brahmanism, but
the terminology has been revised. The world, seen in cross-
section horizontally, has as its axis a disc, a conception
deriving from the Cosmic Support of the Atharvaveda. At
the centre of the disc stands Mount Meru; around Meru
lies Jambudvlpa, the island of the rose-apple tree; six
mountains lie across it, and a sea encompasses it. The other
130 JAINISM
continents are also envisaged as concentric islands, indefin-
ite in number, each one encircled by a sea; a vast ocean
embraces the whole. In the Buddhist conception, the four
principal islands surrounding Meru are situated at the four
cardinal points. This is an ancient scheme found also in the
Mahàbhàrata. Concentric islands and seas are unknown
to the Mahàbhàrata, though the idea is adopted in later
Hinduism, and they are established as seven in number.
The Jaina conception appears less archaic in this matter
than in others.
As a vertical cross-section of the world, instead of the
cosmic Egg of other systems, the Jainas envisage a complex
figure, in the form of a triple pyramid or of a figure of
eight. In the post-Canonical texts, this portrayal evolves
into a human figure, the Lokapurusa, a reminiscence of the
Vedic Purusa. The successive series of heavens form the
head and the breast; the earth is represented by the waist,
where the figure narrows; the lower half of the body repre-
sents the layers of subterranean worlds. The different
spheres are each inhabited by beings appropriate to them,
from the great gods and the siddhas or 'perfect' beings,
down to the demons. This construction is linked with a
mythology in which, side by side with the lesser gods,
whose power is limited, and who arise from karman and are
dependent upon it, we recognize more powerful gods,
Sakra, Isàna, Sanatkumàra, who are barely recognizable
representatives of the Brahmanical gods Indra, Siva and
Brahman. One curious feature is the division of the divine
beings into castes; the Tràyastrimsas, for example, are
administrators in high authority; the Lokapàlas are guar-
dians of order, and so on.
The representations of the infernal regions are among
the most elaborate ever devised by the Indians. The seven
hells are a reflection of the seven pàtàlas of the Brahmans;
as in the Brahmanical conception, both the tormentors and
 JAINISM I31
the tormented are human beings, and there are no demons;
torture by cold, a conception characteristic of Buddhism,
is unknown, at least in ancient Jainism.
Finally, it should be noticed that the conception of a
Supreme Being, to which Hinduism attained by such slow
and painful stages, is just as foreign to Jainism as it pro-
bably was to Buddhism.
As a parallel to their cosmology, the Jainas' conception
of time is one of indefinite extension, and this, as we know,
is quite opposed to Vedic representations. Like post-Vedic
Brahmanism, Jainism asserts that humanity has experi-
enced a progressive decline, the avasarpinu The bad era in
which we are living, the fifth, began soon after the death of
Mahavira (and therefore well after the beginning of the
kali age of Hinduism). It will be followed by an even worse
era before humanity enters upon the utsarpim, the period
of ascent. These six ages are obviously an expanded form
of the four ages so frequently described in other systems;
the pendulum-like course through decline to ascent is
reminiscent of the day and night of Brahman.
All Indian sects have their speculative side; they con-
sider problems relating to what we should call meta-
physics, and develop literatures of varying extent on the
subject. Although this aspect of the thought of a sect can-
not really be separated from its general religious practices,
it will be expedient to give a very brief account of the
subject here.
In the first place, Jainism is an dtmavdda; it recognizes
the Self or jtva as a stable, immaterial and eternal prin-
ciple, endowed with consciousness and initiating action.
This doctrine is, of course, diametrically opposed to that of
Buddhism, but it is in accordance with the spirit of the
Upanisads and the Veddnta.
The theory of karman, the keystone of the system, is con-
ceived in principle much as it is in other systems, but, in
132 JAINISM
conformity with the general trend of Jaina thought, it
takes on a classificatory and encyclopaedic character.
Karman is a real substance, a sort of poison that infects the
soul and renders it liable to be invaded by the other sub-
stances, space and time. The procedure is to destroy former
karman and ward off the approach of new karman) this is
accomplished by asceticism and the other methods of puri-
fication, both ritual and mental. Karman not only deter-
mines the destiny of the soul; it imparts a permanent
quality to it, called lesyd, a word as yet unexplained. Lesyd
is a kind of reflection cast on the soul by matter. Six kinds
are recognized, each one having its own particular colour
(and this is the most important distinction), its own texture
perceptible to the touch, its own taste and duration; they
correspond to the ethical state of the creature in relation to
his own plane, whether it be human, divine, demoniac or
animal.
As in current Hindu teaching, samsdra, the indefinitely
prolonged transmigration of souls, is linked with the theory
of karman. Thus the Jainas have taken over en masse the con-
ceptions that were dimly perceptible amid the confusion of
Indian thought of the post-Vedic period, dtman, karman,
sarnsdra. The state of liberation is conceived of as being
enjoyed in the highest part of heaven, a mountain-peak.
There the siddha, or liberated one, dwells, freed from the
body, yet occupying a position in space, a two-dimensional
being, in fact. He possesses full consciousness and infinite
power, but he makes no use of it (a conception of super-
erogatory endowment that is typically Indian), for his
state is one of absolute repose. This conception is, if I am
not mistaken, close to that of nirvana according to Hlnaydna
or Sdmkhya. In later representations, either activist concep-
tions are introduced, or else there appears the idea of the
unto mystica, which profoundly modified not only the nature
of Liberation, but also the means by which it is attained.
 JAINISM 133
I think these observations have sufficed to show that the
Jaina movement presents evidence that is of great interest,
both for the historical and comparative study of religion
in ancient India and for the history of religion in general.
Based on profoundly Indian elements, it is at the same
time a highly original creation, containing very ancient
material, more ancient than that of Buddhism, and yet
more highly refined and elaborated.
Few religions have made less effort to spread their doc-
trines. Jainism has never been in any way a missionary
religion, as Buddhism has always been and as Hinduism
itself has been from time to time. The archaeological
evidences of Jainism discovered in one region of Central
Asia are a purely chance phenomenon.
At the present day, certain Jaina groups are at last
spreading propaganda outside their own territory, by
issuing pamphlets and establishing centres. The time is
certainly opportune for them to expound the merits of
their fundamental guiding principle, that of non-violence.
Only the future can decide whether an attempt of this
nature has any hope of success. Jainism does not lack
influential and devoted lay-followers, who would be ready
to give their patronage and support, nor is there any dearth
of scholars interested in its origins and literary tradition;
there are in fact today proportionally more scholars work-
ing in this field than in any other branch of Indian studies.
But its contemporary thinkers have not succeeded in ad-
vancing beyond the domain of commentary; as for the
ascetics, their knowledge is by definition incommunicable,
if not undemonstrable. The chief need of the Jainas is, in
fact, for great spiritual leaders, leaders such as Hinduism
has produced more than once, even in recent times.
 INDEX
i.
Abhidhamma, 120 Bhagavadgìtd, 51, 63, 66, 71, 76, 77,
Adhyàtmaràmdyana, 64,88 9i,92
Aditi, 47 Bhdgavatapurdna, 51, 70, 71, 72, 92,
Àdityas, 13, 1411 103
Advaita, 52, 71, 78 Bhàgavatas, 97, 101
Agastya, 62 Bhikanji, 123
Agni, 14, 18,21,33,63 Bimbisàra, 115
Agnistoma, io, 36,81 Brahman, 49, 56,60, 64, 70, 81, 130,
Ahi, 19
Ahura Mazdàh, 13 Brdhmanas, 16, 18, 23, 24, 25, 27, 28,
Ajàtasatru, 115, 116 33,41,43,45,57,59,86
Àjlvikas, 77,116 Bràhmasamàj, 106
Akbar, 104 Brhadàranyaka-Upanisad, 6, 18, 27, 29,
Al-Bìrùm, 103 69
Àlvàrs, 102 Brhatsamhità, 53
Amaru, King, 94 Buddha, 22, 56
AmaSa Spantas, 13
Andhras, 100 Caitanya, 51, 72, 92, 94, 95, 96, 99
Antalikhita, King, 97 Cakradhara, 94
Appayadlksita, 99 Càlukyas, 102
Apollo, 15 Càmars, 96
Arjuna, 62 Candì, 62
Arthapancaka, 69 Candragupta Maurya, 124
Arthasàstra of Kautilya, 50 Cato, 27
Àryasamàj, 107 Chàndogya-Upanisady 7, 18, 24, 27, 43
Asklepios, 15 Còlas,95, 103, 121
Asoka,80,125
Asuras, 62 Daksa, 62
Asvaghosa, 44,116 Dattàtreya, 90
Asvamedha, 31 Dayànanda Sarasvatl, 45, 107
Asvins, 13,20,21,33 Debendranàth Tagore, 106
Atharvaveda, 3, 6, 23, 25, 39, 41, f9 Devasena, 100
74,75,87,127,129 Dharmasàslra, 43
Aurangzeb, 104, 105 Dharmasùtras, 40
Aurobindo, 17, 106, 107 Digambaras, 118, 119, 123
Aveste, 9, 13, 20 D'raupadi, 52

Bài Gangadhàr Tilak, 107 Ekadandins, 93

Basava, 95 Ekàyana school, 43
Berosus, 74
Bhadrabàhu, 116, 117,118, 119 Fa-Hien, 102
Bhàgabhadra, King, 97 Firùz, 104
136 INDEX
Gandhi, 52, 106 Khàravela, King, 120
Ganesa, 60, 61,91 Krsna, 22, 32, 51, 61, 63, 64, 65, 66,
Garuda, 63 68,70,71,73,90,95, 114, 115
Gaudapàda, 58 Krsna Devaklputra, 7
Ghusundi inscription, 101 Kumàra, 63
Gitagovinda, 73 Kumàrapàla, King, 121
Gondophares, 101 Kumàrila, 24, 58, 102, 103
Gopàlatà~pani-Upani$ad, 85 Kundalinìrite, 26, 60, 83, 85
Gorakhnàth, 91, 94 Kùniya, 115
Gosàla Mankhaliputra, 77, 116 Kusànas, 53, 100, 101
Granth, 90, 92
Guhyasamdja, 86 Lakulisa, 94
Gupta dynasty, 101 Lalitavistara, 115
Lilà, 56
Harappa, 3 Lingàyats, 95, 103
Hari, 56,90 Lodis, 104
Harìbhaktivildsa, 51 Lokapàlas, 67, 130
Harivamsa, 49 Loka Pillai, 69
Hariyùpiyà, 4 Lokàyata, 50
Harsa, 102
Hathayoga, 91, 109 Madhva, 75, 104
Heliodorus, 97, 101 Mahàbhàrata, 6, 49, 55, 79, 130
Hemacandra, 121 Mahàdev Govind Rànade, 106
'Henotheism', 12 Mahànirvànatantra, 81
Hephtalites, 102, 125 Mahisa, 62
Heraclitus, 74 Makha, 37
Hiuen-Tsang, 102 Mànasà, 62
Hfdayapundarika, 87 Mànbhàvs, 90, 94, 95, 99
Manu, 23, 49, 55, 76
Indra, 13, 20, 22, 23, 24, 28, 33, 38, Manus, the, 75, 113
_ 59,61,67,70,130 Mara, 67
léàna, 130 Maruts, 14, 15,33,67
Isvara, 68, 69 Maurya dynasty, 100, 117
Maya, 56
Jambudvipa, 73, 129 Megasthenes, 101
Janaka, King of Videha, 7 Menander, 100
Jayadeva, 50 Meru, Mount, 73, 129, 130
Mtmdmsd, 43
Kabir,90,9i,96,104 Mitanni, 4
Kabìrpanthls, 90, 93 Mithra, 53, 102
Kadphises, 101 Mitra, 13, 19,21,33
Kàlàmukhas, 103 Mohenjo-Daro, 3, 47
Kàlì, 62, 80, 125 Mùjavant, Mount, 36
Kalpa, 9 Mundaka Upanisad, 42
Kàma, 61
Kamsa, 64 Naciketas, 30
Kanìska, 100 Nàgas, 67
Kànphatin sect, 93 Nànak, 91
Kàpàlikas, 103 Nàràyana, 64
Kdsyapa-Samhità, 43 Nàtaputta, 117
Kautilya, 50 Nàths, 91
Keshab Chunder Sen, 106, 107 Nàtha Siddhas, 44
 INDEX 137
Namuci, 38 Sdkamedhas, 37
Nàyanàrs, 102 &akra, 130
Nemi, 113, 114 éàktism, 12
Nighantus, 43 Sdmaveda, 11, 41
Nimbàrka, 104 Sdmavidhdna, 41
Nimittasdkha school, 44 Samkarsana, 101
Nirgranthas, 117 Sdmkhya, 27, 47, 57, 68, 71, 74, 78,
Nirukta, 16 85,92,93,127,132
Nydya, 78 Sanatkumàra, 130
Nydya-Vaisenka^ 58, 97, 103 éankara, 24, 58, 68, 69, 92, 93, 94,
99,102, 125
Ouranos, 15,37 Sasthl, 62
Satapatha-Brdhmana, 3, 4, 7, 9, 18, 20,
Padmavati, 54 27, 45,65, 76
Pancadasi, 57 Sautrdmanì, 30, 38
Pancardtra, 43 SàvitrI, 52
Pàndavas, 64 Sàyana, 42
Pàndya dynasty, 93,99, 121 Seniya, 115
Panini, 99 Siddharsi, 100
Pariksit, King, 6 SikanderlI, 104
Pàrsva, 114, 118, 122 Sisupàla, 62, 73
Pàrvati, 62, 88 Sita, 52
Pàsupata sect, 93, 97 Sitala, 62
Patafijali, 55 Siva, 20, 21, 22, 47, 49, 56, 59, 60,
Pradyumna, 61 61, 62, 64, 65, 68, 70, 81, 85, 88,
Prajàpati, 24, 38, 65 9 1 , 97, I O 2 , !25, 130
Prdtimokfa, 7 Sivanàràyanls, 92
Prdtisdkhyas, 2 Skanda, 61
Purdnas, 6, 47, 49, 53, 57, 73, 75, 88, Somadeva, 100, 103
89,91 Spanda school, of Kashmir, io
Pùsan, 33 érldhara, 99
Srikantha, 98
Rabindranath Tagore, 108 Sthànakvàsis, 123
Ràdhàkrsnan, 48, 106 Sthùlabhadra, 118
Ràidàsis, 96 Subrahmanya, 61, 63
Rdjasùya, 31,80 Sudarsana, 63
Ràm Mohun Roy, 106 Kumbha, 62
Rama, 32, 62, 63, 64, 66, 67, 68, 70, Sunga dynasty, 80, 97, 100
71,82,83,97 Sùrdàs, 105
Ràmakrsna, 51, 70, 107 Svetàmbaras, 118
Ràmànanda, 91, 94 $vetdsvatara-Upani$ad, 68, 71
Ràmànuja, 71, 93, 98, 103, 121
Rdmdyana, 7, 51 Tantrism, 26, 39, 54, 61, 84, 85, 87,
Rdstrakùtas, 121 88, 95, 96, 98, 102, 104, 107, 109,
Ràvana, 67 125
Rgveda, 2, 8, io, 11, 12, i4n, 23, Tara, 125
24, 26, 32, 33, 36, 38, 39, 45n, 56 Tattvasamgraha, 44
Rudra, io, 14, 15, 20, 37, 47 Teràpanthis, 123, 124
Rudras, the, 67 Tevijja-Sutta, 99
Theras, 117
Sadhanpanthis, 96 Tirthamkaras, 112, 113, 114
Saivasiddhdnta, 68, 98, 103 Tirupati, 64
i38 INDEX
Toramàna, King, 125 Vemana, 103
Tràyastrimsas, 130 Verikatesa, 64
Tridandins, 93 Vidyddharas, 67
Trika, 103 Vikramàditya, 125
Trimurti, 64, 65 Virasaivas, 86, 90, 93, 96, 99, 103
Tukàràm, 63 Visnu, 14, 21, 22, 38, 47, 49, 60,
Tulsidàs, 51, €7, 70, 73, 74, 82, 88, 63, 64, 65, 66, 68, 69, 70, 81, 91,
105 97
Tvastr, 33 Visnupurdna, 24
Vithobà, 64
Upanisads, 7, 8, 17, 18, 24, 25, 26, 27, Vitthala (Visnu), 63
28, 41, 42, 45, 50, 51, 71, 76, 79, Vivekananda, 107
92,96,100,131 Vrnddvana, 72
Vrtra, 20, 67
Vaikhànasa school, 43, 97 Vrtrahan, 19
Vaikuntha, 69
Vdjapeya,3i,32,36 Yàjfiavalkya, 31, 32, 44
Vdjasaneyi-Samhitd, 9 Tajur-Veda, 9, 11, 15, 43, 45, 63,
Vajrasùci, 44 97
Vajraydna, 54, 88 TaksaSy 67
Valabhl, Council of, 119 Yama, 28, 30, 75
Vallabha, 71, 104 Tasastilaka, 44
Vallabhàcàryas, 95 Talts, 13
Valmlki, 63 Toga, 18, 26, 27, 44, 47, 54, 5^, 77,
Varuna, 13, I4n, 15, 19, 20, 21, 23, 78, 83, 85, 86, 97, 98, n o , 125,
26, 33, 37, 44, 7o 127
Varunapraghdsas, 32 Togavdsistha, 78
Vasudeva, 101
Vedànta, 58, 69n, 75, 77, 85, 92, Zarathustra, 23
93,98,103,104,107,129,131 Zoroaster, 12

ii. BIBLIOGRAPHICAL
Avalon, A. 84 Farquhar,J. N. 90
Barth, A. 311,65 FilliozatjJ. 26,74
Benveniste, E. 15 Foucher, A. 61
Bergaigne, A. 11,14,17: 2 i , 4 5
Bergson, H. 59 Garbe, R. 26
Biot,J.B. 74 Getty, A. 6in
Bloomfield, M. 3n Ghirshman, R. 102
Glasenapp, H. von 79
Caland, W. 9,34 Goossens, R. 15
Grégoire, H. 15
Grierson, G. A. 60, 70
Darmesteter,J. 9 Hamm, F. R. i2on
Dumézil, R. 12, 15, 2 i , 3 7 Hillebrandt, A. 14
kubert, H. 3m
Edgerton, F. 3n
Eliade, M. 75, 76,88 Jacobi, Hermann 3
Emeneau, M. B. 3n Johnston, E. H. 57
 INDEX 139
Keith, A. B. H,34 Peliiot, P. 115
KernJ.G.H. 7 Pott, P. H. 85
Kuiper, F.B.J. 13 Przyluski,J. 53
Lacombe, O. 69 Rònnow, K. 38
Levi, Sylvain 9, 28, 65
Schrader, F. O. 7i
Macdonell, A. A. H Schroeder, L. von 45
Mathieu, M. i5n Schubring, W. 120
Mauss, M. 3i Senart, Émile 7
Meyer, J. J. 61 Smith, Vincent 102
Miiller, Max 8,9, 12, 14,45
Mus, P. 38 Wikander, S. 15

Oldenberg, H. 3n, 14,41 Zieseniss, A. 98