Untitled

Reflections on
Resemblance,
Ritual,
and Religion
ee
at
co 5)
a
_—s eo Mesien ae 2
RL
19a6 1%
Sos Reflections on
BA$ Resemblance,
Ritual,
and Religion
BRIAN K. SMITH
New York Oxford

OXFORD UNIVERSITY PRESS
1989
| heol ogy Library
THEOLOGY
SCHOOL OF EM
AT CLAR ONT
California
Oxford University Press

Oxford New York Toronto
Delhi Bombay Calcutta Madras Karachi
Petaling Jaya Singapore Hong Kong Tokyo
Nairobi Dares Salaam Cape Town
Melbourne Auckland
and associated companies in
Berlin Ibadan
Copyright © 1989 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.,
200 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise,
without prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data

Smith, Brian K., 1953-
Reflections on resemblance, ritual, and religion / Brian K. Smith.
p. cm. Bibliography: p. Includes indexes.
ISBN 0-19-505545-4
1. Hinduism—Rituals. 2. Hinduism—Doctrines. I. Title.
BL1226.12.S65 1989
294.5'38—dc19 88-4009 CIP
Printing (last digit): 987654321

Printed in the United States of America
on acid-free paper
To the memory of Dr. V. V. Bhide, who rectified
the names: teacher, scholar, pandit, friend, host,
Brahmin, father, and man. I miss him terribly.
And for Wendy, who made it all more-ish.

an Pe -
ris
re
ear:
pene aR
_eaeel, ©
PM ey iyMo
, wes
nr ew neta Saige
= oni tag ee ignaw
om pea -
_—— me eS See m4
ra he Ot one tae 4 “Oper oh iae
aoe geekerasers oe
me 348, oe ; |
Le oo aoa ope
Justi chads’oaSay
is + aighy: ok
eee
’
eer hie oe ee
is e tne ara A
4 ie =4 tm
; cee
hy ; SS
i ie iraver & ay le _ :
: :
: a. i
eo
on
’ A at‘zx
~~ bare) a) Tiaets : aa ae!
lee io |W ee wh. 2 RAG ae os.
adn 7404s 2 ee? eee
” a% ur 28 sae :
ik; Li
— - 5 “> - a
+ii ce Aer
Preface
Scholars of the discipline alternately labeled Comparative Religion or

the History of Religions are notoriously, and sometimes dangerously,
schizophrenic. I speak here not of the more ordinary academic malaise
that splits the individual into two personae—teacher of the uninitiated
and hair-splitting author of learned works for other specialists, friends,
and fellow conventioneers. The particular bifurcation I am referring to is
peculiar to a discipline that requires its adherents to address, simulta-
neously, two different audiences in their research reports.
On the one hand, in order to be taken seriously at all in a highly
territorial academic world, one must engage one or more of the estab-
lished bureaus of academic knowledge, such as sociology, anthropology,
history (and all its branches), area studies, Sinology, Buddhology,
Indology, and ancient studies. On the other hand, one is expected as a
comparativist to uncover something of comparative consequence to
those in the field of religion in all its diversity. It is often a difficult
tightrope to walk.
I have attempted to address two audiences in this work, and I hope
that by doing so I have not lost both. Straight comparativists (if I might
be forgiven the oxymoron) and those simply interested in the problems
and methods of an approach to religion which is neither theological nor
“scientific” might find much of the Indological detail of the work tough
going. Philologically oriented Indologists and other specialists in Things
Indian may very well become exasperated with the introductory and
concluding chapters, as well as certain portions interspersed throughout
the book, in which I attempt to move out of the specifics of Vedism into
the larger realm of theoretical and methodological issues pertaining to
the study of religion as a whole. I will only say here that the author’s
Vili Preface
intention was to write a book whose parts cohered, and whose data and
approach to them were in a relation of mutual resemblance. But then
the author’s intention these days is far from being the only criterion of
either meaning or value.
Although this book is meant primarily as a contribution to Indian
studies, it is simultaneously an attempt to exercise one kind of interpreta-
tion of culturally and historically specific material from the vantage
point of the discipline of Comparative Religion or the History of Reli-
gions. This is not simply because issues of comparative interest—the
definition of religion, the nature and purpose of ritual, the dynamics of
sacrifice and substitution, change and continuity in religious traditions—
are directly addressed at various junctures. More critically, I have tried
to present the specifics of Vedism and Hinduism in such a way as to
exemplify theoretical and methodological issues of general import
within the larger study of religion.
I have not always been explicit in my intention to make Vedism and
Hinduism say something of interest to those whose professional or intel-
lectual attention is not usually focused on Indian culture and religion;
and I have not always spared the reader the particularistic details of
often very technical Sanskritic texts. I have, for all that, tried to repre-
sent ancient Indian religion in such a way that it might usefully serve as
an “e.g.” for comparative and theoretical problems within the academic
study of religion. In this I follow the programmatic trail blazed by recent
thinkers who are attempting to reconstitute a field whose past excesses
and errors still cling to it.
Parts of this book have been published before in other forms and in
other places. Chapter 1 previously appeared as part of “Exorcising the
Transcendent: Strategies for Defining Hinduism and Religion” in His-
tory of Religions. Portions of Chapter 2 were published as “Vedic Field-
work” in Religious Studies Review. Other sections of Chapters 2 and 8
appeared in “Ideals and Realities in Indian Religions,” also in Religious
Studies Review. Some of Chapters 3 and 4 were incorporated in “Gods
and Men in Vedic Ritualism: Toward a Hierarchy of Resemblance” in
History of Religions, “Sacrifice and Being: Prajapati’s Cosmic Emission
and Its Consequences” in Numen, and “Ritual, Knowledge, and Being:
Initiation and Veda Study in Ancient India,” also in Numen. Portions of
Chapters 6 and 7 comprise “The Unity of Ritual: The Place of the
Domestic Sacrifice in Vedic Ritualism” in Indo-Indian Journal, and “Sac-
rifice and Substitution: Ritual Mystification and Mythical Demystifica-
tion,” in Numen.
I would like to thank Barnard College and the American Institute
ws Preface iX
of Indian Studies for timely grants that helped considerably in bringing

this work to its conclusion. Special thanks are due to Wendy Doniger for
preparing the index.
I also take this opportunity to acknowledge gratefully the debt I
owe to those who, in one way or another, have taught me: Wendy
Doniger, V. V. Bhide, Holland Hendrix, Wayne Proudfoot, Bruce Lin-
coln, Ron Inden, Frank Reynolds, Arthur Droge, Alan Segal, Marilyn
Harran, G. U. Thite, Wade Wheelock, James Fitzgerald, Barbara
Stoler Miller, John Stratton Hawley, Ainslie Embree, William Harman,
David Carpenter, Frederick Smith, Karen Guberman, Craig, John,
Tim, Chris, Madhav, Celia, Ann, Judy and Steve, Gene (wherever you
are), Mom and Dad, Angela, Justin, Karen, and all my students. Sine
qua non.
New York and Chicago B.K.S.

June 1988
~~ ert he rie F
Oy Mr . Se &* ne)
Ao 20 Pgh Aye, .
Slee setsaulef
ae oo By reinie sree rk ony
Geel tne Wd rere as Sitar Heiaahi
PPL Wh et Sethe PG fame hain Ost inde”
eevee OY spe a, 0 wale meaty waite
hate ikateay Were, dkeer anes aaa
Heap. eee ent s-omrea hued Chee MS feyrenee

ok “eh Poy he b * vyrenes | oiey Ses Pee UALS
- . elk @a in pee AP aly ovale do-ens, ariel aa
: a 2 (4g5 ‘<Pat ape a -tdauiag Gv

> a +s om ENctette ia ePpoe” Wha
“ly ing ‘Stee To 5 Gee
in eee Fie
“4 a my , c+ © a of Cerri Sah g
’ i Pe ee
pa , ei pe hind haem Weng ons vt Ae
in Teta andes” Ay
7 be 7s. ( aprere
: = wate” cay ew fp
a le = W ae T27 thoeh og
a fs pea ke
ie oe el x reek ai
1 task % Sere eam iie
a = m ». iM pee ok pal eke

Contents
Abbreviations Xili
1 Making Connections: Hinduism and Vedism 3
2 Constructing Vedism 30
3 Ritual and Reality 50
4 The Ritual Construction of Being 82
> The Organization of Ritual Knowledge 120
6 The Organization of Ritual Practice 143
Zz Ritual Hierarchy, Substitution, and Equivalency 169
8 The Destiny of Vedism 200
Bibliography 226
Index and Glossary 253

:
er eae
miro wid
Rarer shee :
( =
ae ae
ity meng ay
Abbreviations
AgGS Agnivesya Grhya Sitra

AitA Aitareya Aranyaka
AitB Aitareya Brahmana
AitU Aitareya Upanisad
ApDhs Apastamba Dharma Sitra
ApGS Apastamba Grhya Sitra
ApSS Apastamba Srauta Sutra
AsvGS Aévalayana Grhya Sitra
AsvSS Asvalayana Srauta Sitra
AV Atharva Veda Samhita
BAU Brhadaranyaka Upanisad

BDhS Baudhayana Dharma Sutra
BGS Baudhayana Grhya Sutra
BGParis Baudhayana Grhya Paribhasa Sutra
BhGS Bharadvaja Grhya Sutra
BhSs Bharadvaja Srauta Sutra
BPS Baudhayana Pitrmedha Sutra
BSS Baudhayana Srauta Sutra
ChU Chandogya Upanisad
GautDhS Gautama Dharma Sutra

GB Gopatha Brahmana
GGS Gobhila Grhya Sutra
HGS Hiranyakesin Grhya Sutra

HSS Hiranyakesin Srauta Sutra
XIV Abbreviations
JB Jaiminiya Brahmana
JGS Jaiminiya Grhya Sutra
JSS Jaiminiya Srauta Sutra
JUB Jaiminiya Upanisad Brahmana
KB Kausitaki Brahmana
KGS Kathaka Grhya Sutra
KS Kathaka Samhita
KhGS Khadira Grhya Sutra
KSS Katyayana Srauta Sitra
LSS Latyayana Srauta Sutra
Manu Manu Snnrti

Mbh Mahabharata
MGS Manava Grhya Sutra
MS Maitrayani Samhita
MSS Manava Srauta Sitra
PB Pancavimsa Brahmana
PGS Paraskara Grhya Sutra
PMS Purva Mimamsa Sutra of Jaimini
RV Rg Veda Samhita
SadB Sadvimsa Brahmana

SanA Sankhayana Aranyaka
SB Satapatha Brahmana
SGS Sankhayana Grhya Sitra
SSs Sankhayana Srauta Sitra
TA Taittirlya Aranyaka
TB Taittiriya Brahmana
TS Taittiriya Samhita
TU Taittiriya Upanisad
VaikhGS Vaikhanasa Grhya Sutra

VaikhSS Vaikhanasa Srauta Sitra
VasDhS Vasistha Dharma Sttra
VGS Varaha Grhya Sutra
VSs Varaha Srauta Sitra
Yaj Yajnavalkya Smrti

Reflections on
Resemblance,
Ritual,
and Religion
What was the prototype, what was the counterpart, and what was
the connection between them?
Rig Veda 10.130.3
What is called a sincere work is one that is endowed with enough

strength to give reality to an illusion.
Max Jacos, Art poétique
iL
Making Connections:
Hinduism and Vedism
Working Definitions
There is considerable confusion about the nature of and need for defini-
tions in humanistic studies in general and in religious studies particu-
larly. Many seem to think that to offer a definition of an object is an
enormous—and audacious—act. To propose a definition is at best some-
thing to be deferred to the indefinite future; more often it is delegated to
others who have more accumulated intellectual resources than we.! De-
fining is too often imagined as finalizing—a statement of ultimate truth
about the object of study, a sacred moment in which the whole of the
knowledge of the subject has coalesced and is encapsulated.
Others regard those who presume to offer definitions not with awe
but with suspicion. For these leery scholars of the humanities, definition
is what lab-coated analysts in the “hard” sciences do, and the objects of
humanistic studies are held to be too precious to be subjected to the
same fate that natural objects have suffered. Rather, they are jealously
guarded from the tyranny of the defining will; the endangered objects
1. Even such an authoritative personage as Max Weber hesitated before the defini-
tional challenge, deferring it to the end of the study (and never delivering it): “To define
‘religion,’ to say what it is, is not possible at the start of a presentation such as this.
Definition can be attempted, if at all, only at the conclusion of the study.” The Sociology
of Religion, trans. by Ephraim Fischoff (Boston: Beacon Press, 1953), p. 1.
4 REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
are regarded as exempt from the shackles of defining strictures.* Here,

too, defining is imagined as finalizing—an imprisonment of a once free
object, a cruel pinning of an elusive butterfly.
Let us rethink the concept of definition in the humanities. To define
need not be imagined as to finalize. To define is not to finish, but to
start. To define is not to confine but to create something to refine—and
eventually redefine. To define, finally, is not to destroy but to construct
for the purpose of useful reflection.
It is somewhat delusory to imagine that we do not carry around all
sorts of definitions in our heads. In fact, we have definitions, hazy and
inarticulate as they might be, for every object about which we know
something. It is not defining that is so problematic for so many but,
rather, explicitly defining. On the other hand, those who do make the
effort to define objects of study in the humanities too often appear to
regard it as an end in itself: they work at definition instead of working
with definitions.
It seems to me that it is less than completely forthright not to define
the object in question; it is less than realistic to regard such definitions as
final explanations or canonical codifications; and it is less than maxi-
2. In the field of the study of religion, this protective attitude is sometimes mani-
fested through the argument that religion is irreducible and sui generis, i.e., undefinable.
See, e.g., Rudolph Otto, The Idea of the Holy, trans. by John W. Harvey (New York:
Oxford University Press, 1957). esp. p. 4. For the philosophical poverty of such stances
toward definition in the study of religion, see Hans Penner and Edward Yonan, “Is a
Science of Religion Possible?” Journal of Religion 52 (1972): 107-33; Wayne Proudfoot,
“Religion and Reduction,” Union Seminary Quarterly Review 37 (Fall-Winter 1981-82):
13-25; and esp. Proudfoot’s extended critique in his Religious Experience (Berkeley and
Los Angeles: University of California Press, 1986).
3. See, e.g., Clifford Geertz, Islam Observed: Religious Development in Morocco
and Indonesia (Chicago: University of Chicago Press, 1971). Although Geertz correctly
notes that “the comparative study of religion has always been plagued by this peculiar
embarrassment: the elusiveness of its subject matter,” he attributes this confusion not to a
poorly defined category but, rather, to the lack of sufficient empiricism: “The problem is
not one of constructing definitions of religion. We have had quite enough of those; their
very number is a symptom of our malaise. It is a matter of discovering just what sorts of
beliefs and practices support what sorts of faith under what sorts of conditions. Our
problem, and it grows worse by the day, is not to define religion but to find it” (p. 1).
Categories such as religion, however, are a sine qua non and initially make possible
“religious” facts, not vice versa. Furthermore, Geertz is disingenuous when he claims here
and elsewhere (see pp. 96-97) that he will eschew (explicit) definitions of religion when, in
fact, he proceeds to offer one. My thanks to Holland Hendrix for calling Geertz’s state-
ments to my attention.
Making Connections: Hinduism and Vedism 5
mally productive to envision definitions as the exclusive aim of one’s

scholarship. Let us, then, define our concept of definition as a tentative
classification of a phenomenon which allows us to begin an analysis of
the phenomenon so defined. Let us regard the tentative classification as
a “working definition’—one which permits us to begin work on our
conception of that which is thus defined, and one which therefore also
assumes the continual reworking of the working definition.
To that end, I propose to begin my study of an ancient Indian
religion, Vedism, with a working definition of the whole of which it is
usually regarded as a part: Hinduism. As we will see, defining Hinduism
(at least in the way I will suggest) does not place Vedism as one tradition
within this religion; it shifts the defining center from Hinduism to
Vedism itself. And because, among other reasons, historical and contem-
porary Hinduism defines itself in relation to its archaic (and, in many
respects, superseded) ancestor, our subsequent investigation of Vedic
religion cannot be simply an antiquarian exercise. The bulk of the pres-
ent work, which delves into the details of Vedic religion and ritualism, is
thus more than a study of an ancient religion per se: it is equally an
attempt to squeeze from the Veda the historically constitutive principles
of Hinduism. First, however, we must establish that Hinduism, as we
intend to define it, is best conceived as “Vedic.”
Creating Hinduism
Like most (and perhaps all) particular religions, “Hinduism” as a unitary

entity did not exist before it was created by outsiders. Hinduism is, one
might say, an illusion. It is an imaginary category emerging from the
minds of observers who felt an epistemological (and often political) need
to unify a diversity. “ ‘Hinduism,’ ” notes Wilfred Cantwell Smith, “is a
concept certainly [Hindus] did not have.”4
Just who invented “Hinduism” first is a matter of scholarly debate.
4. Smith, The Meaning and End of Religion: A Revolutionary Approach to the Great
Traditions (San Francisco: Harper and Row, 1962), p. 63. Whether Hindus had a concept
and word homologous to our “religion” is also dubious according to Smith. See also Louis
Renou, Hinduism (New York: George Braziller, 1961), p. 18: “In fact, there is no Hindu
term corresponding to what we call ‘religion.’ ” It is my view that this point deserves
further investigation. Smith’s arguments against dharma as such a correlate are not en-
tirely persuasive, and other terms may also be possible true analogies to religion in Hindu-
ism (€.g., marga.).
Almost everyone agrees that it was not the Hindus.* Derived from the
name of one of the principal rivers of the South Asian subcontinent (the
Indus), the label was used first by the ancient Persians, and somewhat
later by Central Asians, for the people and territory of Northwest India.
As a term designating a religion, still inclusive of what were later to be
differentiated as Hinduism, Buddhism, and Jainism, it may have been
first deployed by the Muslim invaders of the early part of the second
millennium.® As a discrete Indic religion among others, however, “Hin-
duism” was probably first imagined by the British in the early part of the
nineteenth century to describe (and create and control) an enormously
complex configuration of people and their traditions found in the South
Asian subcontinent.’ “Hinduism” made it possible for the British, and
for us all (including Hindus), to speak of a religion when before there
was none or, at best, many.
Although some have expressed their dissatisfaction with the fact
that “Hinduism” is a product of the imagination of foreigners, it is
nevertheless the case that Hinduism has come to take on a certain
reality—for non-Hindus and for Hindus—and it is now beside the point
to attempt to deconstruct it. It is more relevant to note what an odd
religion it is, at least as it has been explicitly defined by some. Take, for
example, J. A. B. van Buitenen’s description published in that popular
organ of knowledge, the Encyclopedia Britannica:
As a religion, Hinduism is an utterly diverse conglomerate of doctrines,
cults, and ways of life. . . . In principle, Hinduism incorporates all forms
of belief and worship without necessitating the selection or elimination of
any. The Hindu is inclined to revere the divine in every manifestation,
whatever it may be, and is doctrinally tolerant, leaving others—including
both Hindus and non-Hindus—whatever creed and worship practices suit
them best. A Hindu may embrace a non-Hindu religion without ceasing to
5. See, however, J. Laine, “The Notion of ‘Scripture’ in Modern Indian Thought,”

Annals of the Bhandarkar Oriental Research Institute 64 (1983): 165-79. Laine argues with
some persuasiveness that “Perhaps, then, ‘Hinduism’ qua ‘religion’ was the creation of
nineteenth century Indians”; that is, the formation of “Hinduism” was a native response
inspired by the confrontation with an already self-defined religion, namely the Christianity
of the British colonizers and rulers.
6. Smith, The Meaning and End of Religion, p. 64.
7. See J. A. B. van Buitenen’s article “Hinduism” in Encyclopedia Britannica, 11th
ed., p. 888, in which the author claims the term was “introduced c. 1830 by English
writers.” Smith, ibid., p. 61, concurs, finding the earliest mention of Hinduism in the
modern sense of the word in an English work published in 1829. See also P. J. Marshall,
ed., The British Discovery of Hinduism in the Eighteenth Century (Cambridge: Cambridge
University Press, 1970).
be a Hindu. . . . Few religious ideas are considered to be finally irreconcil-

able. The core of religion does not even depend on the existence or
nonexistence of God or on whether there is one god or many. Since
religious truth is said to transcend all verbal definition, it is not conceived
in dogmatic terms.§
I have quoted van Buitenen at length in order to present a represen-
tative type of definition, put forward by one of this century’s most
eminent Indologists, which one can only call inchoate. In the best possi-
ble light, van Buitenen seems to want to indicate the artificiality and
inadequacy of the general category of Hinduism (this, let us remember,
in an encyclopedia article entitled “Hinduism”). This religion, the au-
thor implies, is exceedingly diverse (what tradition of three thousand
years’ duration is not?), and to speak naively of a monolithic “Hindu-
ism” is to gloss over its manifold variability and complexity.
Still, van Buitenen also wants to provide the reader with a set of
general characteristics of the religion about which he is writing. As part
of its syncretistic essence (“an utterly diverse conglomerate of doctrines,
cults, and ways of life”), Hinduism is presented as so “tolerant” (a key
word in many such definitions) that virtually everything counts as
“Hindu.”? One can, apparently, believe and practice anything and still
8. Van Buitenen, “Hinduism,” pp. 888-89.

9. Arvind Sharma investigates the basis of Hindu tolerance in his “Some Misunder-
standings of the Hindu Approach to Religious Plurality,” Religion 8 (Autumn 1978): 133-
54, concluding that it is neither Vedantic monism nor caste pluralism that provides the
source of tolerance; it is rather, according to the author, the very essence of Hinduism as a
whole: “To try to be crystal-clear—the Hindu tolerance of plurality is not an iridescence
arising from the interplay of some of its facets; it is not an aspect but an ingredient of the
‘substance’ itself.” See also Kaisa Puhakka, “The Roots of Religious Tolerance in Hindu-
ism and Buddhism,” Temenos 12 (1976): 50-61, whose similar thesis is that “Religious
tolerance as practised in the Eastern traditions is neither a deliberately cultivated moral
virtue nor something forced upon them by socio-political conditions; rather, it is firmly
rooted in and hence arises spontaneously from the religious beliefs and practices them-
selves” (pp. 51-52). It is remarkable that such a stereotype can exist even in the face of
enormous historical evidence (compounded almost daily) to the contrary. Among the few
who seem to find it problematic to regard Hinduism as by definition tolerant in light of past
and continuing events is Nirad C. Chaudhuri, The Continent of Circe (Bombay: Jaico
Publishing House, 1965), esp. p. 33: “If the familiar words about tolerance and capacity
for synthesis of the Hindus were true, one would be hard put to explain why there are such
deep suspicions and enmities among the human groups of India, why there are endemic
outbursts of murderous ferocity.” Hindu tolerance, it would seem, is an idea that had and
has ideological and definitional functions—both for Western observers and Hindus
themselves—that override concrete realities, a scholarly examination of which remains a
desideratum.
be classified a Hindu under this definition, and, remarkably, “a Hindu

may embrace a non-Hindu religion without ceasing to be a Hindu.”
According to another well-known authority,
Hinduism has been linked to a vast sponge, which absorbs all that enters it
without ceasing to be itself. The simile is not quite exact, because Hindu-
ism has shown a remarkable power of assimilating as well as absorbing;
the water becomes part of the sponge. Like a sponge it has no very clear
outline on its borders and no apparent core at its centre.
And, the author concludes, “An approach to Hinduism provides a first
lesson to the ‘otherness’ of Hindu ideas from those of Europe.”!!
This kind of inchoate definition, and there are many others like it, is
far too inclusive to be meaningful, let alone useful.!* To argue that there
is no Hindu “orthodoxy” is also to deny that Hindus can have “heresy,”
which ignores abundant evidence to the contrary.’ To define ade-
quately (if tentatively) an object, in the humanities as elsewhere, is not
only to say what it is but also to indicate what it is not. 4
Among more coherent definitions scholars have offered of Hindu-
ism in the past, one school focuses on religious or philosophical themes
that seem to unite the religion—or, indeed, to constitute it as a religion
in the scholar’s mind. We might therefore label this the thematic type of
10. See J. D. M. Derrett, Religion, Law and the State in India (London: Faber and
Faber, 1958), p. 49. Derrett cites legal rulings whereby a Hindu might convert to Christian-
ity “without ceasing to be a Hindu in both social and spiritual terms.”
11. Percival Spear, India, Pakistan, and the West (London: Oxford University Press,
1949), p. 57.
12. Virtually all definitions of Hinduism include some reference to the extraordinary
social, doctrinal, historical, and practical “tolerance” of the religion. For another example
of the extremes to which one can take this nearly universal notion of Hindu catholicism,
witness the position of Govinda Das, who writes that Hinduism “rejects nothing. It is all-
comprehensive, all-compliant.” A Hindu is one who “does not repudiate that designation,
or better still, because more positive, . . . says he is a Hindu, and accepts any of the many
beliefs and follows any of the many practices that are anywhere regarded as Hindu.”
Quoted in Hervey DeWitt Griswold, Insights into Modern Hinduism (New York: Henry
Holt, 1934), p. 15. “Hinduism,” concludes Spear, “rests essentially on public opinion. Not
to be a Hindu means simply not being thought to be a Hindu.” India, Pakistan, and the
West, p. 58.
13. See Wendy Doniger O’Flaherty, “The Origins of Heresy in Hindu Mythology,”
History of Religions 10 (May 1971): 271-333.
14. Nor is it adequate for the scholar of Hinduism to throw up his or her hands and
declare, with Nehru, that the beast, by definition, defies definition: “Hinduism as a faith is
vague, amorphous, many-sided, all things to all men. It is hardly possible to define it, or
indeed to say precisely whether it is a religion or not, in the usual sense of the word.”
Jawaharlal Nehru, The Discovery of India (London: Meridian Books, 1960), p. 63.
definition. One principal form of this type of definition of Hinduism

identifies the concepts of transmigration (samsdara), karma, and libera-
tion from transmigration and karma (moksa, mukti, etc.) as definitive of
the religion.!5 It is not clear, however, how this approach differentiates
Hinduism from other Indic religions one usually does not regard as
Hindu (Buddhism and Jainism, for example) which also are preoccupied
with these themes.'¢ It would seem that this kind of formulation is also
too inclusive to serve as an adequate working definition of the object
under scrutiny.!”
In addition to the inchoate sort of definition (Hinduism as a vast
and unorganized warehouse of possibilities) and the usually more coher-
ent but still overly inclusive thematic definition (Hinduism defined by
“key concepts” that other Indic religions share), there is a third possibil-
ity in the history of scholarship on Hinduism.!* A definition of Hinduism
15. See, e.g., L. D. Barnett’s presentation of Hinduism identified by three concep-

tual headings under which characteristically Hindu “ideas” cluster: “works” (karma),
“wandering” (samsdara), and “release” (moksa). Hinduism (London: Constable, 1913), pp.
2-3. The best recent versions of the thematic approach to defining Hinduism may be found
in R. C. Zaehner, Hinduism (New York: Oxford University Press, 1966), in which he
explores brahman, dharma, moksa, samsdara, and karma, “the key concepts of classical
Hinduism”; Jan Gonda, Change and Continuity in Indian Religion (The Hague: Mouton,
1965), a sprawling and unfocused work on themes common to all Indian religions; and W.
Norman Brown, Man in the Universe: Some Cultural Continuities in Indian Thought
(Berkeley and Los Angeles: University of California Press, 1970), which is also less con-
cerned with Hinduism per se and more with what is typically and transhistorically Indic
religion. For an extraordinarily eccentric version of this thematic approach to defining
Hinduism, see Griswold, Insights into Modern Hinduism, pp. 17 ff.
16. See Brown, Man in the Universe, and David R. Kinsley, Hinduism: A Cultural
Perspective, ed. by Robert S. Ellwood, Jr. (Englewood Cliffs, N.J.: Prentice-Hall, 1982),
p- 53: “In many essential respects Buddhism and Jainism are very much like Hinduism.
Both movements presuppose karma and samsara and define man’s ultimate spiritual goal
in terms very similar to Hinduism.”
17. Another form this type of definition takes is the identification of three “paths”
(margas) or possibilities for the religious life that collectively constitute the Hindu religion:
the path of works (karma), the path of gnosis (jAdna), and the path of faith (bhakti). See
Charles Eliot, Hinduism and Buddhism: An Historical Sketch, 3 vols. (New York: Barnes
& Noble, 1954), I: xvi—xvii. On first sight this definition might appear to be more adequate
(although overemphasizing syncretistic formulations such as that in the Bhagavad Gita),
but upon reflection it is not clear that here, too, Indian Buddhism, e.g., could not fit within
the definition.
18. One should note, as Penner and Yonan point out in “Is a Science of Religion
Possible?” that recounting the history of definitions offered for the object does not in itself
constitute a definition of the object. At the same time, because it is, in fact, the Western
scholars of Hinduism who constituted that object in the first place, one cannot ignore or
bypass the semantic possibilities Hinduism has had for them.
that one might call the social and/or canonical identifies either or both of
two criteria as constitutive of Hinduism: (1) recognition of the authority
of the Brahmin class and (2) recognition of the authority of the Veda.
According to this position, those who accept either or both the authority
of the Brahmins and the authority of the Veda are Hindus; those who do
not accept one or both are not.
The authority of the Brahmin class as a definitional criterion of
Hinduism is often linked to the caste system (in which the Brahmins, of
course, have made for themselves a privileged position), and Hinduism
is regarded as either identical with the institution of caste or essentially
defined by it. In some works, Hinduism and the particular social system
of India are thoroughly conflated. From this point of view, as R. C.
Zaehner puts it, Hinduism is
as much a social system as a religion. . . . Its social framework has from

very early times been the caste system, and this has, until very recently,
become increasingly rigid, increasingly complicated, and increasingly iden-
tified with Hinduism as such. Indeed until a century or so ago the accep-
tance of the caste system was considered by the orthodox to be the sole
effective criterion of whether one was or was not a Hindu.!°
A variant of this position has both Hinduism and the caste system
pivoting on the place of the Brahmin in Indian religious and social life.
Hinduism is not quite equated with caste, but both Hinduism and caste
are defined by the authoritative position of the Brahmins. The Brah-
mins, writes Eliot,
are an interesting social phenomenon, without exact parallel elsewhere.
They are not, like the Catholic or Moslem clergy, a priesthood pledged to
support certain doctrines but an intellectual, hereditary aristocracy who
claim to direct the thought of India whatever forms it may take. All who
admit this claim and accord a nominal recognition to the authority of the
Veda are within the spacious fold or menagerie. . . . [Hinduism’s] unity
19. Zaehner, Hinduism, p. 8. See also Eliot, Hinduism and Buddhism, 1: xxxvii-
xxxiii: Hinduism is “a mode of life as much as a faith. To be a Hindu it is not sufficient to
hold the doctrine of the Upanishads or any other scriptures: it is necessary to be a member
of a Hindu caste and observe its regulations.” For Hinduism as a national culture, identi-
fied in large part by the caste system, see Sridhar V. Ketkar’s statement: “There is no
‘Hindu religion.’ Hinduism, which means Hindu society and its traditions, is not a religion,
but is akin to tribal or national culture.” Quoted in Griswold, Insights into Modern Hindu-
ism, p. 14. For more sophisticated arguments that Hinduism is a “way of life” of a
particular culture, and thus similar to Hellenism and Judaism, consult Zaehner, p. 1; A. C.
Bouquet, Hinduism (London: Hutchinson’s University Library, 1948), p. 11; and esp. S.
Radhakrishnan, The Hindu View of Life (New York: Macmillan, 1973).
and vitality are clear and depend chiefly on its association with the Brah-
man class.??
Hinduism, according to the logic of this position, may also be labeled
“Brahminism,” for it is “the spiritual prerogative of the Brahman caste
which is the cornerstone on which the whole Hindu social edifice was
built.”
Standing alone, however, the authoritative position of the Brah-
mins within the caste structure cannot adequately function as a criterion
for a working definition of Hinduism. First, the definition becomes less
one of the Hindu religion and more one of the Indic social system, and
from within the discipline of the academic and humanistic study of reli-
gion there is a difference between religion and the social structure it
legitimizes.*” The caste system is certainly grounded in and legitimized
by Hindu religious ideas and practices, but Hinduism clearly entails far
more than caste. Furthermore, it is well known in the West (and better
known in India) that the caste system harbors different religions within
it; there are and have been Buddhist, Jaina, Muslim, Sikh, Parsi, Chris-
tian, Jewish, tribal, and secular Anglo-Indian castes in addition to
Hindu castes in India.
Second, there have been major movements and entire traditions
normally regarded as Hindu that have rejected, at least at the level of
cultic belief and practice, both the authority of the Brahmins and the
legitimacy of caste hierarchy. Many of the tantric traditions (with their
reversals of bourgeois Hinduism) and the bhakti movements (emphasiz-
ing personalized and unmediated faith in and devotion to a gracious
deity) have been characterized doctrinally by a more or less radical
denial of the inherent religious prerogative of the Brahmin caste.73
20. Eliot, Hinduism and Buddhism, I: xvii, 34.

21. Zaehner, Hinduism, p. 5. Hinduism, as “Brahminism,” is thus also regarded as
inextricably linked to a national culture. See Bouquet, Hinduism, p. 70: “Hinduism,
dependent as it is upon the accepted dominance of the Brahmin caste as a hereditary
spiritual aristocracy, is tied to the country of its origin.”
22. Were there no difference posited between religion and social structure, the study
of religion would be entirely encompassed within the study of society. This is indeed the
position of many sociologists and social anthropologists, from Durkheim to the present. It
is not mine, and those who argue so from within the academic rubric of “religion” or
“religious studies” appear to me to have something of a death wish for the discipline. I
hope to deal with the many and complex issues of the relationship between religion and
society, in general, and Indian religion and the caste system, specifically, in another work.
23. For a “Hindu” tantric tradition and their attitudes toward caste, see Edward
Dimock, The Place of the Hidden Moon (Chicago: University of Chicago Press, 1966). For
one set of examples of anticaste, anti-Brahmin rhetoric from within an otherwise quite
Recognition of the authority of the Brahmins, then, is not a suffi-

cient criterion in itself to determine what counts as Hindu. It might,
however, begin to differentiate Hinduism from other Indic religions in
ways that alternative types of definitions do not (Buddhist, Jains, and
others might be excluded on the basis of this criterion) and thus avoids
the problem of overinclusiveness: too much that would appear to com-
mon sense as Hindu, such as many of the bhakti traditions, is ruled out.
Further, a large question is left hanging: on what is the authority of the
Brahmins established and legitimized?
To address this query, scholars call up the second criterion men-
tioned above. Brahmins are recognized as religious authorities because of
their intimate, special relationship to the Vedas, the authoritative (and in
this sense canonical) texts of Hinduism.?> The Brahmins are the tradi-
orthodox Hindu devotionalistic text, see Thomas Hopkins, “The Social Teachings of the
Bhagavata Purana,” in Krishna: Myths, Rites, and Attitudes, ed. by Milton Singer (Chi-
cago: University of Chicago Press, 1966), pp. 3-22. See also A. K. Ramanujan’s Speaking
of Siva (Baltimore: Penguin Books, 1973) for similar sentiments among the ViraSaivas;
and David N. Lorenzen’s paper on the followers of Kabir, “Traditions of Non-caste
Hinduism: The Kabir Panth,” an unpublished manuscript. I do not wish to overemphasize
the anti-Brahmin streak in these movements, however. It seems to me that what was often
at issue was not the concept of Brahminical authority but on what grounds that authority
was based. In other words, the critical point was usually not whether the Brahmins had
religious authority but how and why. There was as much discussion of who was the “true
Brahmin” in these movements as there was anti-Brahmin rhetoric.
24. “Another guideline for orthodoxy in Hinduism is reverence for the Brahmins
and the implicit acceptance of the social hierarchy known as the caste system. Here again
the Buddhist and Jains represent a dissenting view, declaring that Brahmins have no
special religious status.” Kinsley, Hinduism: A Cultural Perspective, p. 53. Again, one
must be careful here. As J. C. Heesterman has pointed out, the apparently anti-
Brahminical discourse of the “heterodox” religions was often not directed at the authority
of the Brahmins per se but at the sources on which they had claimed it. The issue was not
Brahmins versus others but on what basis one was a Brahmin: “The question that occupies
religious thought does not appear to . . . turn on brahmin superiority or its rejection, but
on the point of who is the true brahmin. On these points, both orthodox and heterodox
thinkers seem to agree to a great extent.” “Brahmin, Ritual, and Renouncer,” Wiener
Zeitschrift fiirdie Kunde Siid- und Ostasiens 8 (1964): 1-31, reprinted in a slightly different
form (from which this and all subsequent citations are taken) in The Inner Conflict of
Tradition: Essays in Indian Ritual, Kingship, and Society (Chicago: University of Chicago
Press), p. 42.
25. Not all Brahmins have had such a relationship with the Veda, of course, and
those who have not are ranked hierarchically lower. The point is that a certain knowledge
of the Veda is expected of the Brahmin whether or not individual Brahmins fulfill that
expectation. See Brian K. Smith, “Ritual, Knowledge, and Being: Initiation and Veda
Study in Ancient India,” Numen 33 (1986): 65-89.
tional bearers, purveyors, interpreters, and protectors of the Veda; the

authority of the Brahmin is dependent on the authority of the Veda, and
the Veda exists only because of the traditional function the Brahmin has
assumed for its preservation. Thus, Hinduism has also been defined by
appealing to both of these interdependent criteria: “The only essential
tenets of Hinduism are recognition of the Brahman caste and the divine
authority of the Vedas. Those who publicly deny these doctrines as the
Buddhists, Jains and Sikhs have done, put themselves outside the pale.”2’
Vedic Hinduism
Having reviewed the analytically separable (but in actuality usually con-

flated) types of definitions Indologists have constructed for the construct
called Hinduism—the inchoate, the thematic, and the social and/or
canonical—I now wish to offer my own working definition, locating
myself firmly within the camp of the canonical authority as constitutive
of the religion:?8 Hinduism is the religion of those humans who create,
26. They were also, from the point of view of the humanistic study of religion, in all
probability the creators or “authors” of the Veda, although Brahmins past and present
would deny that the Veda had human authors at all.
27. Eliot, Hinduism and Buddhism, I: 40. See also Barnett, Hinduism, p. 2, for the
religion as partly defined by “the conception of a social order, or caste-system, at the head
of which stand the Brahmans as completest [sic] incarnation of the Godhead and authorita-
tive exponents of Its revelation”; and Zaehner, Hinduism, pp. 8-12, who wants to say that
although Hinduism traditionally was defined by these two criteria, it has undergone many
changes in the past century or so. “Thus while it was once possible to define a Hindu as one
who performs his caste duties and accepts the Veda as revealed truth, this simple formula
can no longer satisfy, for Hinduism is today, more than any other religion, in the melting-
pot: what were once considered to be essentials are in the process of being discarded, but
the hard core remains” (p. 9). This “hard core” is, according to Zaehner, a thematic
configuration. See above, footnote 15.
28. A discussion of the canon in Hinduism, with varying degrees of comparative
awareness, may be found in the following works, some of which I use extensively below:
Thomas B. Coburn, “ ‘Scripture’ in India: Towards a Typology of the Word in Hindu
Life,” Journal of the American Academy of Religion 52 (September 1984): 435-59, and
“The Study of the Puranas and the Study of Religion,” Religious Studies 16 (1980): 341-
52; J. F. Staal, “The Concept of Scripture in the Indian Tradition,” in Mark Juer-
gensmeyer and N. Gerald Barrier, eds., Sikh Studies: Comparative Perspectives on a
Changing Tradition, (Berkeley: Graduate Theological Union, 1979), pp. 121-24; J. C.
Heesterman, “Veda and Dharma,” in Wendy Doniger O’Flaherty and J. Duncan M.
Derrett, eds., The Concept of Duty in South Asia (Columbia, Mo.: South Asia Books,
1978), pp. 80-95, and “Die Autoritat des Veda,” in G. Oberhammer, ed., Offenbarung
(Vienna: Indologisches Institut der Universitat Wien, 1974), pp. 29-40; Laine, “The
perpetuate, and transform traditions with legitimizing reference to the

authority of the Veda.
I do not claim that this definition (or any other presently available,
for that matter) is perfect. A perfect definition, fully explanatory and
logically unassailable, would finish the scholarly labor on Hinduism. My
definition allows for an interesting point of departure for further reflec-
tion along lines not fully explored in past scholarship. I would here like
to unpack its components and implications in order to argue for its basic
worth as a starting place.
First, the definition is fully consonant with a humanistic orientation
for the study of religion. Hinduism, like all other religions from this
point of view, centers on human beings who create, perpetuate, and
transform their traditions. Hinduism was created by human imagina-
tion;2? it was and is perpetuated by humans who regard their ancestors’
creation as valuable and worth preserving; and it is transformed by
humans who over time adapt and revise their ancestors’ religion. The
Hindus who create, perpetuate, and transform their traditions include
Brahmins, of course, but not to the exclusion of other classes and castes
of human creators, perpetuators, and transformers of traditions. The
Brahmins, insofar as they have for millennia monopolized the functions
of preserving and perpetuating the Veda, the authority of which all
Hindus call upon to legitimize their doctrines and practices, are critical
to Hindu orthodoxy but not definitive of it.
From this point of view, Indic religions might be classified into four
types: (1) those that accept the authority of both the Veda and the
Brahmin class (the mainline Hindu traditions); (2) those that accept the
authority of the Veda but reject the notion of inborn prerogatives for the
Brahmins (some of the more radically devotionalistic bhakti sects and
some strains of what is called Tantricism); (3) those that, although shar-
ing many important concepts with the other two (e.g., karma, samsara,
dharma, liberation), reject both the absolute authority of the Veda and
the Brahmins (Buddhism, Jainism, Sikhism, and other “heterodox”
movements); and (4) those religions imported to the Indian subconti-
Notion of ‘Scripture’ in Modern Indian Thought”; and Robert C. Lester, “Hinduism:

Veda and Sacred Texts,” in Frederick M. Denny and Rodney L. Taylor, eds., The Holy
Book in Comparative Perspective (Columbia, $.C.: University of South Carolina Press,
1985), pp. 126-47.
29. Here I agree with Wilfred Cantwell Smith’s oft-repeated thesis that there is no
(e.g.) Hinduism apart from Hindus. I have noted above the fact that there was no Hindu-
ism before scholars (also human) so constituted it as a religion.
nent that reject both the authority of the Veda and the Brahmin and the
aforementioned concepts (Islam, Christianity, Judaism, et al.). The first
two listed in this scheme, by virtue of their adherence to the authority of
the Veda, are, under the definition here set forth, equally “Hindu”; the
third and the fourth are not.
A second strength of such a definition of Hinduism might be that by
making the definition of Hinduism pivot on the legitimizing reference to
the Veda, the canon, Hinduism ceases to be utterly different from the
Western religions created, perpetuated, and transformed by “peoples of
the book.”>! Indeed, it is arguable that all religions, by definition, must
have a fixed canon (although not necessarily a written one) and a ruled
set of exegetical strategies for its interpretation.22 Although there are
critical distinctions between the concept of the Veda on the one hand
and those of (for instance) the Torah, the Bible, and the Qur’an on the
other, the points of similarity have too often been elided in academic
discourse—perhaps as part of the general Orientalist project to encode
Asian religions and cultures as exotically strange and foreign to the rest
of humankind.
30. My thanks to Bruce Lincoln for his help in organizing my thinking on this
matter.
31. For Hinduism as wholly other than Western religions, see the quotation of
Percival Spear cited above, p. 8, and Paul Younger’s statement in Jntroduction to Indian
Religious Thought (Philadelphia: Westminster Press, 1972): “Even a superficial acquain-
tance leaves one with the impression that the religious life of India is fascinating, complex,
and mysterious, but, above all, different from the religious traditions of the West” (p. 9;
emphasis added). For a general discussion of the constitution of the Oriental “other,”
consult Edward Said, Orientalism (New York: Pantheon Books, 1978).
32. This is an extremely important topic for the study of religion which has not
received sufficient theoretical and comparative treatment. For now, the best study (sugges-
tive rather than definitive, however) is Jonathan Z. Smith’s “Sacred Persistence: Toward a
Redescription of Canon,” in Imagining Religion: From Babylon to Jonestown (Chicago:
University of Chicago Press, 1982), pp. 36-52. See also the essays collected in Wendy
Doniger O’Flaherty, ed., The Critical Study of Sacred Texts (Berkeley: Graduate Theologi-
cal Union, 1979); and the surveys for particular traditions in Denny and Taylor, The Holy
Book in Comparative Perspective.
33. This is not to overrule the possibility that a unitary and fixed conception of “the
Veda” as a scripture on a par with the Torah, the Christian Bible, and the Qur’an was a
rather recent notion in India. Laine has argued just this: “Because ‘Hinduism’ is no one
‘religion,’ it has no one ‘scripture.’ The British were not so disposed to see this, however,
and consequently neither were a number of the Hindu intelligentsia with whom they were
in contact. Confronted by two religions of the book, Islam and (primarily Protestant)
Christianity, Hindu intellectuals sought to establish the ‘true’ scripture of ‘Hinduism.’ ”
“The Notion of ‘Scripture’ in Modern Indian Thought,” p. 167. See also the interesting
Third, this definition of Hinduism is neither too inclusive nor too

exclusive. It is not too inclusive, as some we have considered are, for it
clearly and unequivocally factors out Buddhism, Jainism, Sikhism, and
other Indic religions that do not create, perpetuate, and transform their
traditions with legitimizing reference to the authority of the Veda. They
have other canonical sources to whose authority they refer for legitimiza-
tion and certainly do not feel compelled to revert to the Veda, which
they do not regard as authoritative. It is on this very point, as we will
see, that Hindus themselves have regarded Buddhists, Jains, and others
as heretics.
The definition is also not too exclusive. The point deserves further
attention, but it does seem, for example, that even those bhakti tradi-
tions that have expressed cultic resistance to the authority of the
Brahmin caste privilege have not institutionalized resistance to the au-
thority of the Veda.** Some bhakti texts insist that the personalized
Hindu deity to which their devotion is directed is to be known through
knowledge of the Veda.*> Indeed, both the Tantric and devotionalistic
movements seem to have taken great pains to emphasize not only their
allegiance to the Veda but also, as we will see, their reproduction of the
Veda in their own sectarian texts.*°
Finally, although our working definitions in the academic and hu-
manistic study of religion need not fully correspond to the definitional
criteria used within religions themselves,”’ it is problematic if the two do
texts collected and analyzed in Richard Fox Young’s Resistance Hinduism: Sanskrit
Sources on Anti-Christian Apologetics in Early Nineteenth-Century India (Vienna: Institut
fiir Indologie der Universitat Wien, 1981).
34. This is not to say that within bhakti discourse there cannot be found disparaging
remarks about the value of the Veda, even within a very Hindu text such as the Bhagavad
Gita (e.g., 2.41-46). Indeed, the tendency to minimize the value of the Vedic textual
learning can be traced to the Veda itself: the much debated hymn at Rg Veda 7.103 may be
a satirization of Brahmin recitation of Vedic hymns, and one encounters many instances of
deprecating remarks on the worth of the “lower” Vedic knowledge in the Upanishads
(e.g., Mundaka Upanishad 1.1.4—S). In all cases, however, it does not appear that the
authority of the Veda is being challenged as much as built upon.
35. See, e.g., Varaha Purana 70.40.44 and Karma Purana 1.50.21-24.
36. For a general discussion of the Vedic allegiance of the Tantric and South Indian
bhakti movements, see Louis Renou, The Destiny of the Veda in India, trans. by Dev Raj
Chanana (Delhi: Motilal Banarsidass, 1965), pp. 5—6, esp. the notes. For particulars, see
N. Subbu Reddiar, Religion and Philosophy of the Nalayiram (Tirupati: Sri Venkateswara
University, 1977), esp. pp. 680-93 (“The Nalayiram as Dravida Veda”). I am grateful to
Francis X. Clooney for bringing this latter work to my attention.
37. I here disagree with one of the premises of Jonathan Z. Smith’s otherwise
admirable attempt at a “polythetic” definition of Judaism, “Fences and Neighbors: Some
not at least converge at certain points. Defining the Hindu religion in the
manner I am suggesting here certainly does not completely and slavishly
reproduce what seems to be the full array of criteria used by Hindus
themselves to speak of their religion (and these, of course, vary from
Hindu to Hindu). It does, however, pick up and highlight many of the
critical features of what appears to be the dominant native point of view.
That Hindus do indeed see themselves as defined, in part or in
whole, by their relationship to the Veda (a relationship usually mediated
by the Brahmin class) seems certain. The Veda, a massive collection of
ancient “texts” or “scripture” composed and orally preserved in San-
skrit,** has been regarded by many Indic people (those we call Hindus)
for several millennia as sacred—or to use the terminology of the religion
itself, the Veda was “revealed” to or “heard” (Sruti) by the foundational
ancestors (the rsis).
Acceptance of the absolute authority of the Veda has been the
criterion of orthodoxy among Hindus from at least the time of the early
Dharma Sutras and Smrtis (ca. third or second century B.c.£.). Manu,29
for example, declares (rather intolerantly, one might add): “All tradi-
tions (smrtis) and misperceived philosophies (kudrstis) which are outside
of [the authority of] the Veda produce no reward after death, for they
are founded on darkness, it is said.”4° Elsewhere in the text we read that
the sruti (i.e., the Veda) is amimamsa (“unquestionable”), and those
who persist in challenging its dictates are to be scorned and avoided;
their doctrines are anrta (“untrue” and “disordered”).4!
Contours of Early Judaism,” in Imagining Religion, pp. 1-18. For an example of such a
polythetic definition in the hands of Indianists, one might again cite Spear, India, Paki-
stan, and the West, pp. 59-60: “We have, then, a body of ideas, beliefs and values, which
together make up the mysterious amorphous entity which is called Hinduism. Each is
present in some one part of Hinduism and few in every part. Any one can be dispensed
with in any one section without forfeiting the title of Hinduism, and no item is absolutely
essential. But some of each class must always be there.”
38. It has often been noted that the use of the categories “text” and “scripture” for
the orally transmitted Veda is somewhat misleading and inappropriate. See, e.g., Staal,
“The Concept of Scripture.” The Veda is a collective term for four different Vedas (Rg,
Sama, Yajur, and Atharva), each comprised of four types of “text”: Samhita, Brahmana,
Aranyaka, and Upanisad. For a survey of the components, dates, and contexts of the
Veda, see Jan Gonda, Vedic Literature (Samhitas and Brahmanas), Vol. 1, fasc. 1 of Jan
Gonda, ed., A History of Indian Literature (Wiesbaden: Otto Harrassowitz, 1975).
39. For the full names of texts abbreviated in this work, see the list at the front of
this volume, and for full bibliographical citations, consult the Bibliography. All transla-
tions, unless otherwise noted, are my own.
40. Manu 12.95.
41. Manu 2.10-11, 12.96.
Hindu scholastics for the past several millennia have consistently

maintained that orthodoxy turned on acceptance of the foundational
authority of the Vedas. Kumarila, the teacher of Mimamsa (“orthodoxy
par excellence,” according to Renou), ranked all religions and philoso-
phies on the basis of their stated adherence to the Veda. Buddhists,
Jains, and others are heretical, “for they do not accept the fact that the
Vedas are the source [of their teachings], just as an evil son who hates
his parents is ashamed to admit his descent from them.”*? Sankara, the
great teacher of Advaita Vedanta, determined heresy on the basis of
veda-virodha (“contradicting the Veda”). Indeed, acceptance of the
authority of the Veda marks orthodoxy even in modern Hindu reforma-
tional movements,“ and most scholars have not wholly disregarded the
fact that “tolerant” Hinduism in general has its limits: those Indians who
did not and do not accept the sacrality of the Veda have been and are
regarded as non-Hindus by those who did and do.*
42. Tantravarttika of Kumarila Bhatta, ed. and trans. by G. Jha (Calcutta: Biblio-
theca Indica, 1924), p. 165. Kumarila’s statement, implying that Buddhist and Jaina doc-
trines are not new at all but are already found, in essence at least, in the Veda, is
reminiscent of the claim in the Mahabharata that whatever is can be found there, and what
is not there does not exist.
43. Commenting on Taittiriya Upanishad. For this and other similar citations, con-
sult O'Flaherty, “The Origins of Heresy,” pp. 272-73, 275.
44. For the place of the Veda in modern Hindu revival movements, see D. S.
Sharma, The Renaissance of Hinduism (Varanasi: Benares Hindu University, 1944); and
A. Bharati, “The Hindu Renaissance and Its Apologetic Patterns,” Journal of Asian
Studies 29 (February 1970): 267-88, esp. p. 276: “A radical statement of the tenets of the
Renaissance would be: /n nuce—India has forgotten her marvelous past; this past con-
tained not only material and cultural wealth, it also offered a complete solution of all
problems of the individual and of society. . . . It is all contained in the Vedas and the
Gita.” Gandhi’s Hindu credentials may also be gauged by statements such as the follow-
ing: “I claim myself to be sanatanist. . . For me, Sanatana Dharma is the vital faith
handed down from the generations belonging even to the prehistoric times and based upon
the Vedas and the writings that followed them.” Quoted in D. G. Tendulkar, Mahatma, 8
vols. (New Delhi: Publications Divisions, Ministry of Information and Broadcasting,
Govt. of India, 1961-69), III: 180. See also the citations from the writings of Rammohun
Roy, Debendranath Tagore, Vivekananda, and others in Laine, “The Notion of ‘Scrip-
ture’ in Modern Indian Thought.”
45. “Traditionally, Hinduism divides all philosophies into orthodox and nonortho-
dox. Of the latter, all of which deny the sacredness of the Vedas, the most famous are the
Buddhists and the Jains. . . . [BJoth religions are considered nonorthodox by Hindus
primarily because they deny the authority of the Vedas, considering them mere human
products.” Kinsley, Hinduism, p. 53. See also Lester, “Hinduism,” p. 126: “Hinduism
holds much in common with other religions having their roots in India: Buddhism, Jainism
and Sikhism; its distingusihing mark is reverence for Veda”; Heesterman, “Veda and
So it is that a definition such as the one proposed here that centers

on the relationship of people to the Veda does not radically depart from
similar discussions within the religion itself. At the same time, it does
not accept some of the assumptions about the Veda that Hindus appear
to think are crucial to their self-definition.
The orthodox claim that the Veda is a body of transcendent and
super- or extra-human knowledge“ is not included within our defini-
tional statement. From the standpoint of the academic and humanistic
study of religion, the Veda, like all other canonical literatures, was
entirely composed of human beings. This very statement, from the na-
tive point of view, defines us as not Hindu—and when we seriously play
the role of humanistic scholars of religion we are excluded from being
Hindu, or Christian, or Jewish, and so forth, so long as we are playing
the role. Refusal to follow Hindus on this point of authorship does not,
however, close off the possibility of following native definitional state-
ments in other regards.*’ The definition proposed here resembles, but
does not reduplicate, statements of what constitutes a Hindu issuing
forth from Hindus themselves.
Another juncture at which the definition we are considering here
and the self-definitions put forth by Hindus both converge and diverge
concerns the historical use of the Veda. Hindus by definition, we have
claimed, are those humans who create, perpetuate, and transform tradi-
Dharma,” p. 80: “The respect for the Vedas and the acknowledgment of its ultimate
authority are therefore quite logically given as the decisive criteria for Hindu orthodoxy”;
Renou, The Destiny of the Veda in India, p. 2: “We have just spoken of orthodoxy: the
Veda is precisely the sign, perhaps the only one, of Indian orthodoxy”; O’Flaherty, “The
Origin of Heresy,” p. 272: “The contradiction of the Vedas remains the basis of heresy in
the Hindu viewpoint”; and Gonda, Change and Continuity, p. 7; Hinduism “is defined as a
complex of socio-religious phenomena which are based on the authority of the ancient
corpora, called Veda.”
46. Veda derives from the Sanskrit root vid, “to know,” and means “true knowl-
edge” in the sense of timeless, absolute Truth. Transcendence is also implied in the
concept of Sruti, for it is said to be eternal and apauruseya, i.e., not created by human
beings, for which claims see, e.g., PMS 1.1.27—32. For the even more extravagant position
that the Veda is the creative source of all beings, see the Vedanta Sutras 1.3.27—29.
47. A suggestive possibility for rapprochement of the divergence on the question of
the authorship of the Veda is found in a Feuerbachian reading of Coburn’s “ ‘Scripture’ in
India.” In a discussion of Hindu concepts of revelation turning on the metaphors of
hearing and seeing, Coburn writes: “This identification of two senses of the rsi’s experi-
ence is no mixing of metaphor but an effort to convey the holistic and supremely compel-
ling nature of that experience. It engages one through, and yet transcends, the senses. It
seizes one with a unique and irresistible immediacy. It is in such experiences that the
human becomes contiguous, even identical, with the divine” (p. 442; emphasis added).
tions with legitimizing reference to the authority of the Veda. Hinduism

is thus, for non-Hindus, not timeless, static, or changeless (it is a series
of creations and transformations), although—and here is the point of
divergence—t is usually presented as such by its adherents (it is simply a
perpetuation).*® This disjuncture between Hinduism perceived and de-
fined by non-Hindus and Hinduism perceived and defined by Hindus
brings us to the heart of what it means to define Hinduism as suggested
here.
The Veda and Strategies for Orthodoxy
The great paradox of Hinduism, defined in this manner, is that although

the religion is inextricably tied to the legitimizing authority of the Veda,
in post-Vedic times the subject matter of the Veda was and is largely
unknown by those who define themselves in relation to it. Its contents
(almost entirely concerning the meaning and performance of sacrificial
rituals that Hindus do not perform) are at best reworked (being, for
example, reconstituted into ritual formulas or mantras for use in Hindu
ceremonies), and many cases appear to be totally irrelevant for Hindu
doctrine and practice. The paradox is stated succinctly by Paul Younger,
who notes, “In spite of the acknowledgment of its authority, the content
of the Veda does not seem to be used very directly in guiding the later
development of the Religious Tradition.”4? Or, as Heesterman puts it,
when post-Vedic Hinduism “appeals to the authority of the Vedas, it
more often than not refers to an unknown entity.”°° Although it appears
to be the case that Hindus do acknowledge the absolute authority of the
Veda for legitimizing post-Vedic Hindu beliefs and practices, the rela-
tionship to the Veda often seems to be, as Renou writes, like “a simple
‘raising of the hat’ in passing to an idol by which one no longer intends to
be encumbered later on.”5! The Hindus, it would seem, use only the
“outside” of the Veda, oblivious to its “inside.”%2
48. For an overview, and overgeneralization, of the Hindu devaluation of historical

change, see Louis Dumont, La Civilisation indienne et nous (Paris: Armand Colin, 1975),
chap. 2: “La Probléme de histoire”; and Madeleine Biardeau, “L’Inde et Vhistoire,”
Revue historique 234 (1965): 47-58.
49. Younger, Introduction to Indian Religious Thoughtp. 71.
50. Heesterman, “Veda and Dharma,” p. 80.
51. Renou, The Destiny of the Veda in India, p. 2.
52. See Wendy Doniger O’Flaherty, Other Peoples’ Myths (New York: Macmillan,
1988), for an exploration of this distinction.
a Making Connections: Hinduism and Vedism 21
If, then, the content of the Veda is by and large irrelevant to post-
Vedic Hindus, in what possible sense can Hinduism depend on Vedic
authority for its legitimacy and integrity as a religion? And how do
Hindus, from our point of view, transform their traditions over time
without, from their point of view, diverging from the simple repetition
of the eternal truth of the Veda?
One answer to this conundrum of a “Vedic Hinduism”—which ap-
pears in doctrine and practice to be mostly non-Vedic—is that the Veda
is not always treated as a closed canon. A clear example of the supple-
ness of the Veda qua canon is the fact that new Upanishads (one of the
components of the Veda or Sruti according to traditional classification)
have been composed throughout the history of Hinduism. New texts are
in this purely nominal way given the authority of the Veda, indicating
also that the Veda, as Hindus use it, is something of an “open book.”
Furthermore, religious literature in Hinduism is not confined to the
Veda or Sruti; it is also comprised of what is classed as smrti, “remem-
bered” or traditional wisdom. The smrti, in opposition to the Sruti, is
regarded by Hindus themselves as wholly composed by human beings
and passed on from generation to generation. The relation of smrti to
§ruti is complex, but two points are relevant for our purposes here. On
the one hand, virtually all the literature in which the actual doctrines and
practices of Hindu sects are codified is technically regarded as smrti and
therefore not the Veda. At the same time, these same admittedly
noncanonical texts are often said to be entirely in conformity with—f
not a mere restatement of—the (largely unknown and unexplored) ca-
nonical Veda and therefore are regarded as authoritative and, indeed,
canonical. Robert Lingat observes:
If one takes smrti in its etymological sense of human tradition founded on
memory, its authority cannot but be inferior to that of Sruti, which is direct
revelation of the rule. But in course of time its authority grew to the point
of equalling that of sruti.*
How does the non-Vedic, noncanonical, purely human smrti take
on all the authority of the canonical, transcendent, and eternal Veda?
How is transformation and change, embodied in smrti, made to appear
as perpetuation and continuity of the sruti? Several different kinds of
strategies for investing smrti texts with all the authority and timelessness
of the Veda might be enumerated.
One method is to assert that the smrti is the Veda, the Sruti. In
53. Robert Lingat, The Classical Law of India, trans. by J. Duncan M. Derrett
(Berkeley: University of California Press, 1973), p. 13.
22. REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
addition to texts called Upanishads, some post-Vedic texts designate

themselves, with some pointedness, “Samhitas.” Other texts created in
the post-Vedic age that represent actual (and usually non-Vedic) Hindu
doctrine and practice sometimes call themselves the “fifth Veda,” also to
claim “Vedic” status.*4 Still others, as Thomas Coburn has pointed
out for the principal scripture of Hindu goddess worship (the Devi-
Mahatmya)*®> and others have noted for the Ramayana,* are treated
exactly as if they were Veda—that is, they are chanted according to
schema of Vedic recitation. In this way, innovations are legitimized (and
denied as innovations) by simply expanding what counts as, or is treated
as, the eternal Veda.
Many other Hindu texts explicitly claim to be vedasammita (“equiva-
lent to the Veda”) or vedarthasammita (“equivalent in meaning to the
Veda”). The line between transcendent sruti and man-made smrti seems
here to evaporate as both become “Veda.” A most dramatic instance of
this rather common stratagem is the case of compositions of Nammal-
var, a South Indian Sidra, devotee, and poet whose four principal works
were equated in various ways to the four Vedas—and became known as
the “Dravidian” or “Tamil Veda.”’
Another approach is the apparently more modest argument that the
smrti is based on the Veda and has, as D. S. Sarma points out, only
derivative authority:
[T]he Sruti forms the supreme authority for Hinduism. All the other
scriptures which form the Smriti, are secondary, deriving their authority
54. Both of the great epics of Hinduism, the Ramayana and the Mahabharata, have
made this claim, and so, too, have many of the Puranas that represent the doctrines and
practices of certain forms of sectarian Hinduism in the first and early second millennia c.£.
(e.g., Skanda Purana 5.3.1.18; see also Chandogya Upanishad 7.1.2, providing a Vedic
charter for the “fifth Veda” phenomenon: itihasapuranam panicamam vedanam vedam). A
closely connected notion is that the great Puranas were emitted from the fifth mouth of the
god Brahma (Bhavisya Purana 1.2.56—57).
55. Thomas Coburn, Devi-Mahatmya: The Crystallization of the Goddess Tradition
(Columbia, Mo.: South Asia Books, 1985), pp. 63-67, concluding that “Considerations of
content indicate that the DM is clearly smrti. Considerations of function suggest an anal-
ogy with the function of sruti.”
56. Milton Singer, “The Cultural Pattern of Indian Civilization: A Preliminary Re-
port of a Methodological Field Study,” Far Eastern Quarterly 15 (November 1955): 33; and
esp. Philip Lutgendorf, The Life of a Text: Tulsidas’s Ramcaritamanasa in Performance,
Ph.D. dissertation, University of Chicago, 1986.
57. See N. Subbu Reddiar, “The Nalayiram as Dravida Veda,” pp. 680-93 of his
Religion and Philosophy of the Nalayiram. Reddiar also notes that the same status was
conferred on the works of others of the Alvars as well.
i Making Connections: Hinduism and Vedism 23
from the Vedas, and are accepted as authoritative only in so far as they
follow the teachings of the primary scriptures.°8
This is the position, worked out in great detail, of the mimamsaka
Sabara, who argues that the authority of the smrti is vouchsafed only by
demonstrating its ties to particular fruti passsages. This is not often
difficult to do, given a certain flexibility of interpretation.
But the basis in the Veda of the smrti teachings can be inferred by
other means. The original (human) teacher and “author” of the smrti
text can be regarded as omniscient, that is, the master of the Veda, of
Truth: “Whatever duty has been ordained for anyone by Manu has
already been fully declared in the Veda; for [Manu was] omniscient
(sarvajnanamaya).”°° Failure to find distinct precedents for smrti teach-
ings is thus a result of incomplete knowledge of the interpreter, and not
due to a discrepancy in the text; the smrti can be regarded as, inevitably,
yathasrutam (“in accord with the Veda”).
The Veda is here treated not so much as a set of texts but as a body
of knowledge that is incorporated in certain individuals who have memo-
rized and mastered it. Human teachers and exemplars, then, can serve
as the guarantors of the Vedic authority of post-Vedic Hinduism, just as
the rsis did for the ancestors in illo tempore.® In delineating the sources
of Hindu dharma, many texts®! list three such founts: the Veda itself; the
smrti (which is regarded as at least based on the Veda, if not a complete
replication of it); and “custom,” more fully descried as saddcara (“good
custom”) or Sistacadra (“the custom of learned men”). The Sistas are
those Brahmins “who are properly trained in the Veda and its supple-
ments and who know how to make discerning inferences from the
erin. ©
As noted above, the Brahmin class is, by virtue of its traditional
rights to knowledge and perpetuation of the Veda, critically important
for determining orthodoxy, for determining what is “Vedic.” Their au-
thority, however, is clearly derivative. The third source of dharma (the
58. D. S. Sarma, “The Nature and History of Hinduism,” in Kenneth W. Morgan,

ed., The Religion of the Hindus (New York: Ronald Press, 1953), p. 7. Sarma’s article,
which explicates Hinduism’s “underlying unity” as consisting of common scriptures, com-
mon deities, common ideals, common beliefs, and common practices, is among the best
short but comprehensive definitional statements of the religion currently available.
59. Manu 2.7.
60. The rsis, according to Nirukta 1.20, had “direct intuitive insight into dharma.”
61. These include ApDhS 1.1.1.1-2; GautDhS 1.1-2; VasDhS 1.4-7; Yaj 1.7;
Manu 1.6.
62. BDhS 1.1.1.6.
customs of learned Brahmins), then, no less than the second, is wholly

borrowed from and based on the Veda, as Lingat notes:
Sadacara is a religious life, exclusively orientated towards the acquisition
of spiritual merit. It amounts to the practices observed from generation to
generation by Sistas, or those who are at once instructed and virtuous.
They have not only plumbed the depths of the Veda and its related texts,
but they can deduce from them all the consequences which are implied,
and, what is more, they conform to these teachings in their conduct.®
Such a personalized conception of Vedic authority continues in recent

Hinduism. The modern saint Ramakrishna was said to be a living em-
bodiment of the “nectar” of the Vedas:
That nectar which has been obtained by churning the infinite ocean of the
Vedas, into which Brahma, Vishnu, Shiva, and the other gods have
poured their strength, which is charged with the life essence of the
Avataras—Gods Incarnate on earth—Shri Ramakrishna holds that nectar
in his person, in its fullest measure.™
Another way Hindus guarantee the Vedic authority of their post-

Vedic texts is to contend that in cases where the Vedic precedent is not
found, the smrti is based on a lost Veda. Already in the ritualistic Sutras
(ca. seventh through third centuries B.C.E.) smrti texts were claiming Sruti
basis: the ApGS (1.4.12.10) states that the rules for domestic rituals were
once given in Brahmanas but that these §ruti texts have been lost. This is
also the argument of last resort of Kumarila, the mimadmsaka who went
even farther than his colleague Sabara in claiming that the srti was
invariably based on Vedic authority. Lingat summarizes his position:
“When a rule of dharma has no source, we must conclude that it rests
upon a part of the Veda which is lost or somehow hidden from view.”®
Similarly, in the Puranas, one finds the claim that those texts were com-
posed when parts or the whole of the Veda vanished from the earth.©
This argument is tied in certain ways to the Hindu doctrine of the
degenerating nature of time, the present age, need it be said, envisioned
as the most degenerate of all.67 One result of this decline from better
63. Lingat, The Classical Law of India, pp. 14-15; see also pp. 178-80.
64. Vivekananda, quoted in Laine, “The Notion of ‘Scripture’ in Modern Indian
Thought,” p. 172.
65. Lingat, The Classical Law ofIndia, p. 8.
66. See, e.g., Siva Purana 7.1.1.35.
67. For the theory of declining yugas, resulting in degeneration of human capabili-
ties and necessitating changes in the dharma appropriate to each age, see Lingat, The
Classical Law of India, pp. 183-89, and his “Time and the Dharma,” Contributions to
Indian Sociology 6 (1962): 7-16.
times is that the Veda, once one hundred thousand verses in length, has
been reduced to its present, smaller size, with the lost portions only
incompletely known by untraceable smrti passages. Making a virtue out
of a potentially embarrassing necessity, Hindu texts with no visible
Vedic precedents can thus claim to be retrieving for Hindus parts of
their canon otherwise unavailable.
Closely related is the notion that Hindu texts are simplified forms of
the Veda. In these corrupt times, people are unable to understand the
Veda, and therefore simplified, “digest” versions have been created for
simple minds: “Those who have lost [the sense of] sruti must run back to
the Tantra to rediscover the path to the Vedas,”® and South Indian
bhakti texts written in the vernacular were also said to contain the
“cream,” “purport,” or “essential teachings” of the Veda. Many San-
skrit Puranas also present themselves as streamlined renditions of Vedic
wisdom now made available to women, Sidras, and others previously
excluded from learning the Veda.” With claims like this, the smrti texts
become “means of access to the Veda” and function as “intermediaries,
if not intercessors.”7! Indeed, some texts warn, without the smrti acting
as a buffer the Veda is useless or dangerous.”
The smrti is also regarded as enlarging the Veda in ways that comple-
ment rather than contradict the claim of simplification. As we have just
observed, many of the Puranas see their role as both the amplification
and clarification of the Veda.” The smrti in general is thought to serve a
popularizing function, bringing the message of the Veda (often in the
form of epic tales and mythological stories) to the masses.” In addition
to the myth of the shrinking Veda, one finds also the myth of the Veda
divided and multiplied into, finally, all the symrti texts of Hinduism.
Combining the themes of simplicity and enlargement (in the sense
68. Viramitrodaya, cited by R. C. Hazra, Studies in the Upapuranas (Calcutta:

Sanskrit College, 1963), p. 93.
69. Reddiar, Religion and Philosophy of the Nalayiram, pp. 682-85.
70. E.g., Bhagavata Purana 1.4.25; Devi Bhagavata Purana 1.3.21.
72. E.g., Mbh 1.1.204: “With both Epic and Purana one should support the Veda—
the Veda is afraid of one of little knowledge.” J. A. B. van Buitenen, trans., The
Mahabharata, p. 31.
73. See Naradiya Purana 2.24.15 ff.
74. See Sarma, “The Nature and History of Hinduism,” p. 8: “What the Vedas are
for the learned, the Ramayana, the Mahabharata, and the Bhagavad Purana are for the
common man. They are accepted as sacred books by all Hindus.” See also J. A. B. van
Buitenen, “Hindu Sacred Literature,” Encyclopedia Britannica, 15th ed., VIII: 933; and
C. Mackenzie Brown, “Purana as Scripture: From Sound to Image of the Holy Word in
the Hindu Tradition,” History of Religions 26 (August 1986): 68-86, esp. p. 76.
of explication) is the notion that Hindu religious texts condense the

essence of the Veda. The Puranas often state that they are sarvavedasara
or akhilagrutisara (“the quintessence of the whole Veda”)? or “the es-
sence of the meaning of the whole Veda” (sarvavedarthasdra),” or the
“soul” (dtman) of the Vedas.” This strategy is also deployed in texts of
the South Indian bhakti movement which claim to be the essence of the
Rg Veda or, indeed, of all four Vedas.”
Finally, we come full circle when we read that the smrti is not based
on the Veda but, rather, that the Veda is based on the smrti. According
to the Matsya Purana,’”? God thought of the Puranas before he spoke the
Vedas, and other texts®® clearly state that the Vedas are “established”
(pratistha) on the Puranas.
These are some of the ways (and there are undoubtedly more) in
which post-Vedic Hindus create new traditions and transform older ones
(i.e., change them, from the outsider’s perspective) while they are per-
petuating “the Vedic tradition” (i.e., do not change them, from the
orthodox Hindu view). By representing, through deployment of these
strategies for orthodoxy, new texts, doctrines, and practices as con-
nected in some fashion or another to the Veda, change is both legiti-
mized and denied, and continuity is both affirmed and stretched.
The working definition of Hinduism proposed here pivots on the
notion that the Veda functions as a touchstone for Hindu orthodoxy. It
functions, that is, as a kind of canon, a point of reference that is re-
garded as absolutely authoritative. It would seem, however, that the
Veda is a rather peculiar canon—at once irrelevant and never ignored,
open and unquestionable, endlessly reenvisioned and eternally un-
changed. Some, seizing upon the first terms in these pairs of opposites,
75. Bhagavata Purana 1.2.3, 1.3.42, 12.13.15; Naradiya Purana 1.1.36.

76. Naradiya Purana 1.9.97.
77. Skanda Purana 5.3.1.22.
78. See V. Rangacharya, “Sri Vaisnavism in South India,” in H. Bhattacharya, ed.,
The Cultural Heritage of India, 4 vols. (Calcutta: Ramakrishna Mission, 1953), I1V:167;
and Reddiar, Religion and Philosophy of the Nalayiram, p. 683: “The first twenty-one
verses of the Tiruvaymoli [of Nammalvar] convey clearly the purport of the Vedanta
system. They represent the twenty-one recensions of the Rg Veda. The thousand verses
which are set to music represent the Sama-veda teachings in a thousand recensions. The
hundred decads represent the Yajur-veda which has one hundred recensions. This composi-
tion abounds in the delineation of the eight sentiments and so represents the Atharva-veda
which has eight recensions. Hence the Tiruvaymoli shall be taken to represent all the
Vedas” (citing the Dramidopanisat-tatparyaratnavali 5).
79. Matsya Purana 53.3.20.
80. Skanda Purana 5.3.1.20; Naradiya Purana 2.24.16.
Making Connections: Hinduism and Vedism Zi]
have suggested that Hindus indulge in a pious but transparent fiction

with their claims of continuity with the Veda. The strategies for ortho-
doxy outlined above are dismissed as cynical rationalizations or down-
right deceptions.*! Others, perplexed by the apparent use only of the
“outside” of the Veda by Hindus (and the neglect of its “inside,” its
doctrines and practices), have attempted to uncover the “symbolic mean-
ing” of the canonical Veda for post-Vedic Hindus.*
The Veda is thought by some to signify merely a Hindu notion of
absolute knowledge or truth in a general sense. Lingat concludes that
“in reality, it seems that when a Hindu affirms that dharma rests entirely
upon the Veda, the word Veda does not mean in that connection the
Vedic texts, but rather the totality of knowledge, the sum of all under-
standing, of all religious and moral truths.”®? Paul Younger takes an
even more mystical view of the real meaning Veda has for Hindu ortho-
doxy. The Veda, Younger claims,
cannot be defined in terms of boundaries in the way Western concepts are
defined. There are no limits to where the Vedic ends and the non-Vedic
begins. Veda is like a center of light from which concentric circles of
truthfulness radiate outward.
Recently, Jan Heesterman has suggested that the defining attach-
ment Hindus have to the Veda is indicative of “the inescapable need for
ultimate authority.”® The Veda is preserved and sanctified in all its
doctrinal and practical irrelevance for post-Vedic Hinduism because its
irrelevance serves a purpose.** The canon can obtain its utter transcen-
dence and authority precisely because of its utter irrelevance, because
“it is unconcerned with and untouched by the vagaries of human life and
81. Vivekananda, e.g., pointedly complained that “in India. . . if I take certain
passages of the Vedas, and if I juggle with the text and give it the most impossible
meaning . . . all the imbeciles will follow me in a crowd.” Quoted by Renou, The Destiny
of the Veda in India, p. 61.
82. See Renou, The Destiny of the Veda in India. “One always believes oneself to be
in the wake of the Veda, while one turns one’s back on it. The term tends to serve as a
symbol” (p. 1); and Lingat, The Classical Law of India: “It is this hypothetical or symbolic
code, rather than the surviving Vedic texts, which the most ancient authors . . . have in
mind when they proclaim that the Veda is the primary source of dharma” (p. 8).
83. Lingat, The Classical Law of India, p. 8.
84. Younger, Introduction to Indian Religious Thought, p. 70.
85. Heesterman, “Veda and Dharma,” p. 84; see also “Die Autoritat des Veda.”
86. It is on this point that Heesterman’s argument can be distinguished from an
otherwise comparable theory put forward by Frits Staal on the “meaninglessness” (defined
as purposelessness) of Vedic ritual. See Chapter 2 herein.
society.”87 The Veda, then, symbolizes nothing short of transcendence

itself and maintains its authority by being intentionally divorced from
human reality. Rather than solving the paradox, Heesterman ends up
repositioning it as a conclusion: “The crux of the matter is that the
Vedas hold the key to ultimate legitimation. Therefore, even if the
Vedas are in no way related to the ways of human life and society, one is
still forced to come to terms with them.”*8
It is, indeed, entirely because Hindus are “forced to come to terms”
with the Veda that I have suggested my definition of Hinduism. A
reimagining of Hinduism along these lines, however, can remain content
neither with dismissing Hindu strategies for orthodoxy (connecting to
the Veda) as contrived fictions nor with the conclusion that the Veda is a
mere “symbol” for Hindus of more or less mystical notions of “Truth” or
“transcendence.” The first reduces Hindu orthodoxy to meaningless
form; the second depends on restating Hindu orthodoxy’s claims for the
superhuman essence of the (now “symbolic”) Veda.
Hinduism, I would argue, is to be defined as a process, not an
essence. It is not the essential nature of the Veda (or what it “symbol-
izes”) that is of defining import as much as it is the particular relation-
ships Hindus establish and maintain with it. And these relationships,
which we have called strategies for orthodoxy, might be regarded with
less skepticism and with more historical depth. It is possible that it is
here, with the connections themselves, that one will make and find the
connections, so to speak, between the ancient and seemingly “irrele-
vant” Veda and post-Vedic Hinduism.
Let us take our cue from Giorgio Bonazzoli and his remarks on the
nature of “canon” within religion:
A canon can be either a mere list of works that are accepted as normative
and therefore considered as authentic or it can be a complex of rules
through which one can establish whether a work should be accepted as
authentic and normative or not.®9
Compare Jonathan Z. Smith’s argument that a “canon cannot exist with-
out a tradition and an interpreter. . . . The process of arbitrary limita-
tion and of overcoming limitation through ingenuity recurs.”
It is clear from the foregoing that the principal rule for determining
87. Heesterman, “Veda and Dharma,” p. 84.

88. Ibid., p. 93.
89. Giorgio Bonazzoli, “The Dynamic Canon of the Puranas,” Purana 26 (July
1979): 127-28; emphasis added.
90. Smith, “Sacred Persistence,” pp. 49, 50.
“canon” (i.e., orthodox religious authority) in Hinduism is whether or

not it is “Vedic.” We have also observed the variety of strategies
(“subrules” for “overcoming limitation through ingenuity”) available for
constituting post-Vedic texts, doctrines, and practices as “Vedic”: reflec-
tion (this is the Veda); restatement (this is based on the Veda); reduc-
tion (this is the simplified Veda); reproduction (this enlarges the Veda);
recapitulation (this is the condensed essence of the Veda); and even
reversal (the Veda is based on this).
The common thread linking all these strategies is not only the neces-
sity to return to the Veda in order to progress within the boundaries of
Hinduism;*' it is also the notion that the particular post-Vedic text, doc-
trine, or practice resembles, in one way or another, the Vedic prototype.
The Veda is the prototypical canon that can lend its authority to resem-
bling counterparts. It is here, as we will see, that some interesting possibili-
ties open up for rethinking Hinduism as a set of “Vedic” traditions.
I have said that my intention in formulating a working definition of
Hinduism was not to finish but to start. The proposed definition—
Hinduism as the religion of those humans who create, perpetuate, and
transform traditions with legitimizing reference to the authority of the
Veda—has led to an investigation of the kinds of relationships post-
Vedic traditions establish with the Veda. It has also led to the ponderous
question of the relationship (or lack thereof) of Vedism, the religion of
the Veda from the “inside” and Hindu traditions which, it is said, use
only the “outside” of the Veda for their legitimization. Is it possible that
the Veda has bequeathed more than just its covers to its Hindu off-
spring? Put more pointedly, is it possible that the strategies of orthodoxy
themselves, connecting counterparts to prototype, are Vedic?
To answer these questions, we must begin another project, one
which will preoccupy us through most of the remainder of this book: the
attempt to delineate and demonstrate the underlying (which is not to say
unconscious or hidden) principles of the religion represented in the
Veda. We will then be in a better position to return to the definitional
question of what we think Hinduism is, and why.
91. This process recalls Mircea Eliade’s frequently stated notion that the “return to
origins” is a central feature of religion in general. See, e.g., The Myth of the Eternal
Return, trans. by Willard Trask (Princeton, N.J.: Princeton University Press, 1954); and
Myth and Reality, trans. by Willlard Trask (New York: Harper Colophon, 1963).
Ze
Constructing Vedism
Mistaken Identities
Vedism, like Hinduism, does not come to us in an unmediated purity,

untouched by the concerns of later ages. It too is the product of centu-
ries of definitional choices and strategies which have prefabricated what
we think we know about the oldest religion of the subcontinent. Al-
though no individual or era can entirely think themselves out of such
preconditions, it is worthwhile to attempt an understanding of how
Vedism has been imagined in the past in order to imagine it in ways that
might be more appropriate for the present.
As the history of scholarship on Vedism amply attests, the predomi-
nant feature of that religion! is an apparent preoccupation with the
“discovery” or constitution of linkages between what others regard as
wholly discrete objects, acts, or entities. Making connections has served
as the defining feature of Vedism as a whole. The fact that one sees in all
the multitude of ancient Vedic texts, produced by different authors at
different times and to some degree for different purposes, a common
concern to link together apparently unconnected things has itself made
possible a certain generalization on the part of the scholar of these texts.
1. In the study of Vedism, as elsewhere, there has recently been a trend to

deconstruct the object into historically, socially, geographically, and culturally discrete
elements; that is, there has been an increased recognition that what we call Vedism is a
general category comprised of originally differentiated strands. For the atomistic approach
to Vedic religion, see, e.g., Asko Parpola, “The Pre-Vedic Indian Background of the
Srauta Rituals,” in Frits Staal, ed., Agni: The Vedic Ritual of the Fire Altar, 2 vols.
(Berkeley, Calif.: Asian Humanities Press, 1983) II: 41-75.
Constructing Vedism Sil
We can speak of a Vedic tradition, making and finding connections

between texts from various sources and times, because all these texts are
characterized by their tendency to make and find connections.
“Vedic thought,” writes Renou, “as one catches glimpses of it in its
ritual form in the Brahmanas and in its speculative form in the Upanisads,
defines itself as a system of equations.”? The quintessentially Vedic sys-
tem of “equations,” “equivalences,” “homologies,” or “connections”—
all possible translations of the Sanskrit word bandhu3—unites the ritual
texts (Samhitas, Brahmanas, and Sutras)‘ with the more explicitly mysti-
cal, speculative texts (Aranyakas and Upanisads). There is an epistemo-
logical affinity between typically Brahmanical and ritualistic declaration
such as “The sacrifice/sacrificer is the year” and typically Upanisadic and
mystical statements such as “The true self (atman) is the universal princi-
ple (brahman).”>
An analysis of the very names by which Vedic texts are known
seems to indicate a common theme underlying differences in genre.
Brahmana, for example, designates texts “where the function is pre-
cisely to establish connections,” as Lilian Silburn writes.° And, as Renou
observes, “The word upanisad itself, as it is first used in the Satapatha
Brahmana, means only ‘equivalence,’ ” and “the aim of the whole of
2. Louis Renou, “ ‘Connection’ en védique, ‘cause’ en bouddhique,” in Dr. C.

Kunhan Raja Presentation Volume (Madras: Adyar Library, 1946), p. 55.
3. A partial bibliography on bandhus in the Vedic texts would include Jan Gonda,
“Bandhus in the Brahmanas,” Adyar Library Bulletin 29 (1965): 1-25; Armand Minard,
Trois enigmes sur les cent chemins: Recherches sur le Satapatha Bramana, 2 vols. (Paris:
Société d’edition les belles lettres, 1949), 1:50-51; Boris Oguibenine, “Bandhu et daksina.
Deux termes védique illustrant le rapport entre le signifiant et le signifié,” Journal
asiatigue 276 (1983): 263-75; Hermann Oldenberg, Vorwissenschaftliche Wissenschaft: Die
Weltanschauung der Brahmana-texte (GO6ttingen: Vandenhoeck and Ruprecht, 1919);
Asko Parpola, “On the Symbol Concept of the Vedic Ritualists” in Religious Symbols and
Their Functions, ed. by H. Biezais (Stockholm: Almquist and Wiksell, 1979), pp. 139-53;
Louis Renou, “ ‘Connexion’ en védique, ‘cause’ en bouddhique,” pp. 55—60; Stanislov
Schayer, “Die Struktur der magischen Weltanschauung nach dem Atharva-Veda und den
Brahmana-Texten,” Zeitschrift fiir Buddhismus 6 (1925): 259-99; and Michael Witzel,
“On Magical Thought in the Veda,” (Leiden: Universitaire Pers Leiden, 1979).
4. For the relationship between the Sutras, technically classified as outside the ca-
nonical Veda, and the Vedic texts per se, see Chapter 5 herein.
5. I will return to the question of the continuity or discontinuity between the ritualis-
tic texts and the Upanishads later in this work. For now, it is enough to point out that many
scholars view these two types of ancient texts as linked on the grounds of their program-
matic linkages.
6. Lilian Silburn, Instant et cause: Le Discontinu dans la pensée philosophique de
l’'
Inde (Paris: Libraire philosophique J. Vrin, 1955), p. 64.
32, REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
Vedic thought may be expressed as the attempt to formulate upani-

sads.”7 From Sylvain Lévi, who in 1898 argued that the priestly scholas-
tics who produced the ritualistic Brahmanas were the fathers of Indian
philosophy,’ to Jan Heesterman, who claims that Upanisadic doctrines
of world renunciation and monism were “already implied in classical
ritual thinking,” a large segment of scholars of Vedism have concluded
that the unity of the tradition derives from this fundamental, recurring
imperative to make connections.
At the same time, it is a remarkable fact about the Western recon-
struction of the Vedic religion that whereas the great identities of the
Vedantic Upanisads (linking the microcosmic true self to the macrocos-
mic One in an expression of mystical unity) have always seemed to
inspire admiration, statements of equivalence in the ritualistic Brah-
manas are often scorned and disdained as so much mumbo jumbo from
the imagination of the priests."
F. Max Miller, the doyen of nineteenth-century Indology and the
“editor” of the Rg Veda, despised the Brahmanas as examples of the
tragic corruption of the pure religiosity expressed in the earliest Vedic
compositions, the Samhitas. He denounced the later works as
a literature which for pedantry and downright absurdity can hardly be

matched anywhere. . . . The general character of these works is marked
by shallow and insipid grandiloquence, by priestly conceit, and antiquar-
ian pedantry. .. . These works deserve to be studied as the physician
studies the twaddle of idiots, and the raving of madmen.!!
Miller later repeated his opinion that the Brahanas were “twaddle, and
what is worse, theological twaddle,”!? and this view was shared by many
others.
7. Louis Renou, Religions of Ancient India (New York: Schocken Books, 1968), p.
18. See also Harry Falk, “Vedisch upanisad,” Zeitschrift der Deutschen Morgenlandischen
Gesellschaft 136 (1986): 80-97; and the older study by S. Schayer, “Uber die Bedeutung
des Wortes upanisad,” Rocznik Orientalistyczny 3 (1925): 57-67.
8. Sylvain Lévi, La Doctrine du sacrifice dans les Brahmanas (Paris: Ernest
Leroux, 1898), esp. pp. 10-11.
9. Jan Heesterman, “Brahmin, Ritual and Renouncer,” p. 41.
10. For a discussion of the history of Brahmana-bashing in the West, see Wendy
Doniger O'Flaherty, Tales of Sex and Violence: Folklore, Sacrifice, and Danger in the
Jaiminiya Brahmana (Chicago: University of Chicago Press, 1985), pp. 3-6.
11. F. Max Miller, A History of Ancient Sanskrit Literature, 2d ed. (London: Wil-
liams and Norgate, 1860), p. 389.
12. F. Max Miller, Chips from a German Workshop, 2 vols. (New York: Scribner,
Armstrong and Co., 1871), I: 113.
Constructing Vedism 33
Why were these works so regarded? As Arthur Macdonell would

write, they are “full of sacerdotal conceits, and fanciful, or even absurd,
identifications, such as is doubtless unparalleled anywhere else”; or, as
Julius Eggeling, the translator of the Satapatha Brahmana, reiterates,
“For wearisome prolixity of exposition, characterised by dogmatic asser-
tion and a flimsy symbolism rather than by serious reasoning, these
works are perhaps not equalled anywhere.”4
The system of equations, held to be characteristic—if not wholly
definitive—of Vedism, when viewed in the Brahmanas at least, was any-
thing but systematic. Regardless of how the philosophical identifications
in the Upanisads were received, for much of Indology the equivalences
drawn in the Brahmanas were “fanciful,” “absurd,” and “flimsy.” And
although it was not apparently perceived, such conclusions on the part of
nineteenth- and early-twentieth-century scholars of ancient Indian reli-
gion reduplicated the conclusions drawn by those at work on “primitive”
religion. The latter, no less than the former, found in such identifications
proof positive of the utter difference, and usually the utter inferiority, of
non-Western, nonmodern religions and “modes of thought.”
In the case of Vedism, the scorn generated by those who studied
and represented it was (and largely still is) not only because Vedic
Indians seemed to confound two or more distinct entities (a confusion of
modern and western categories) but also because they did so, it ap-
peared, randomly. “According to the Brahmana-texts,” G. U. Thite
could assert in 1975, “anything can be connected with anything else;
anything can result from anything; everything is all.”!© Even more re-
13. Arthur Macdonell, A History of Sanskrit Literature (New York: D. Appleton

and Co., 1900), p. 32.
14. Julius Eggleing, trans., The Satapatha Brahmana, according to the Text of the
Madhyandina School, 5 vols. (Oxford: Clarendon Press, 1882-1900; reprint ed. Delhi:
Motilal Banarsidass, 1963), I: ix.
15. The similarities between the Vedic system of connections and those definitive of
what once what called the primitive mentality are striking. It remains somewhat strange
that those who were concerned with identifications within primitive religions—from Tay-
lor, Frazer, and Lévy-Bruhl to Evans-Pritchard and Lévi-Strauss—were oblivious to paral-
lel (at least) instances in Vedic religion, and vice versa. For recent discussions of the
problem in “primitive” religion, see Raymond Firth, “Twins, Birds and Vegetables: Prob-
lems of Identification in Primitive Religious Thought,” Man (new series) 1 (1966): 1-17;
Audrey Hayley, “Symbolic Equations: The Ox and the Cucumber,” Man (new series) 3
(1968): 262-72; Robin Horton and Ruth Finnegan, eds., Modes of Thought: Essays on
Thinking in Western and Non-Western Societies (London: Faber and Faber, 1973); and
Bryan Wilson, ed., Rationality (Oxford: Basil Blackwell, 1977).
16. G. U. Thite, Sacrifice in the Brahmana-Texts (Poona: University of Poona,
1975), p. 5. Thite claims that the Brahmanas can be understood as being guided by a
cently, Frits Staal carries on the Indological tradition of denigrating the

Brahmanas by contending that the explanatory connections drawn in
those texts are inconsistent, contradictory, arbitrary, and, again, “fanci-
ful.”!7 One simply throws up one’s hands in despair in the face of this:
kind of prelogical (and postlogical?) thinking.
Thus, in our first foray into the nature of Vedism we have discov-
ered that, by general scholarly consensus, the religion is characterized
(or, it could equally be said, created) in large part by its recurring
concern with making connections. We have also learned, again on ex-
pert authority, that such connections are unintelligible; they are, at best,
mistaken identities and, at worst, the random whims of flimsy minds. It
is hard to imagine how texts whose subject matter is in this way repre-
sented could also be imagined as the definitional key to a world religion
such as Hinduism—unless, of course, Hindus are imagined in a manner
that makes such a connection possible, albeit neither respectable nor
respectful. Is it not more valuable (to say nothing of its plausibility) to
imagine that the mistake in the identities lies not with the Vedic Indians
but with those who have represented them in this way?
The Overactive Imagination
The distinctively Vedic proclivity for making connections is intricately

connected to another feature or characteristic of the religion. Vedism
also appears to be preoccupied with the theory and practice of ritual. As
a configuration of cosmogonies and cosmologies, of metaphysics and
epistemology, of social structure and sociology, of ethics and values, of
psychologies and anthropologies, wholly originating in and depending
on ritual activity, what we call Vedism can equally be labeled Vedic
ritualism.
The sacrifice, or yajna, was, in one way or another, the subject
matter of an enormous corpus of texts (the Veda and its ancillaries, the
principle he calls “henoism,” which considers “not only any deity... to be supreme
successively, but anything which is the subject matter at a particular moment becomes the
omnipotent, the highest, the only one identical with all etc., for the time being.” This is, of
course, an attempt to resurrect and rework what Max Miller labeled “henotheism” and is
as obfuscating and bizarre in its new form as it was in the original.
17. Frits Staal, “The Meaninglessness of Ritual,” Numen 26 (1979): 4; idem, The
Science of Ritual (Poona: Bhandarkar Oriental Research Institute, 1982), pp. 11-12, 36;
idem, Agni, I:3, 27-28, 67-68.
Vedanga), a literature that “consists almost entirely of sacred texts that

were in use at, or centered around, the specific Vedic religion of sacri-
fice and ritual.”!8 Although comparativists have largely ignored it, no
body of materials is better suited for the study of ritual than the Vedic.
As Staal argues, the Vedic ritual is not only distinctive of Vedism but is
crucial for general ritual studies because it is “the richest, most elaborate
and most complete among the rituals of mankind.”!9
But if the preoccupation with ritual provides us with a second princi-
pal component for describing Vedism as a religion, it also provides
another occasion for many to denigrate it as a religion. The supposedly
absurd and random nature of the equivalences in the Brahmanas is only
one reason why they (and those who proposed them) were, and are, so
demeaned by those who, in most cases, spent large portions of their
scholarly lives studying them. A second reason seems to be that these
equations were invariably produced within a ritual context.
Whereas the religion of the Rg Veda (ca. 1200 B.c.E.) was coded by
Indologists as one of guileless and creative poesy, later and more para-
digmatic Vedism was said to be thoroughly enmeshed in priestly machi-
nations. Hermann Oldenberg depicted this ritualistic Vedism as a
“vorwissenschaftliche Wissenschaft,” and Staal has recently spoken of a
Vedic “science of ritual.” And whereas the early Upanisads, from the
usual point of view, reflect a certain healthy realism (by which is meant
an antiritualism) and work out their magnificent philosophical monism
in the purely metaphysical realm, the Brahmanas are frankly, explicitly,
and hopelessly concerned with “equations” only insofar as they help to
explicate the hidden significances of the ritual.
The Upanisads breathe the air of the real world, or at least the real
world of metaphysicians; the Brahmanas operate in what Oldenberg
called “a sphere of strangest activity and the playground of a subtle
empty mummery”:
It has been fatal for all thought and poetry in India that a second world,
filled with strangely fantastic shapes, was established at an early day be-
side the real world. This was the place of the sacrifice with its three sacred
fires and the schools in which the virtuosos of the sacrificial art were
18. J. A. B. van Buitenen, “A Brief History of the Literatures of South Asia,” in

Edward C. Dimock et al., eds., The Literatures of India: An Introduction (Chicago:
University of Chicago Press, 1974), p. 15. See also Renou’s statement that “Vedic religion
is first and foremost a liturgy.” Religions of Ancient India, p. 29.
19. Fritz Staal, “Ritual Syntax,” in M. Nagatomi et al., eds., Sanskrit and Indian
Studies in Honor of Ingalls (Dordrecht, Neth.: D. Reidel, 1979), p. 122.
educated—a sphere of strangest activity and the playground of a subtle

empty mummery, whose enervating power over the spirit of an entire
nation we can scarcely comprehend in its full extent.7°
Not only, then, were homologies said to have been mistakenly pos-
tulated and haphazardly formulated; what is worse is that the equations
served as the basis of action. The identities were manipulated in order to
effect a change on the entities, actions, or objects that had been put into
such a relation. The bizarre imagination of the Vedic priests was quite
overactive, and the rituals in which such a “strange activity” occurred
are thus sometimes called magical.
Caught in a paradox equal to any we claim to see in “prelogical”
peoples, scholars can thus simultaneously argue that the equivalences
drawn in the Brahmanas were both random and ruled. For inasmuch as
magic means anything as a classificatory term, it generally creates a class
of rituals performed out of the supposed belief that action within the
ritual has an effect on the world because of certain “laws”—for example
the “law of similarity” whereby like produces or acts on like, or the “law
of contiguity” whereby action on the part produces an effect on the
whole.?! “Magic gives every outward appearance of being a gigantic
variation on the theme of the principle of causality,” wrote Marcel
Mauss; its “exclusive aim, apparently, is to produce results.”2? Magic,
then, is a type of ritual that presumes to cause an effect on y through the
ruled actions performed on x, under the basic assumption that x and y
are or can be essentially related because of certain universal “natural
laws.” For this reason, Vedic ritual can be called alternately magic or a
kind of science.
Rather than seriously address the intricacies of Vedic “magic,” how-
ever, most scholars have been content to envision it as the other side of
the coin of the absurdity of Vedic connections. Magic, as a label, has
rarely been applied to systems of belief and practice other than pejora-
tively, and this is at least in part because here, as with the Vedic identifi-
cations, the scholars is faced with what seems to be a confusion between
the symbol (x) and what is symbolized (y). As John Skorupski observes,
20. Hermann Oldenberg, Ancient India: Its Language and Religions, 2d ed., (Cal-
cutta: Punthi Pustak, 1962), pp. 21-22.
21. See Marcel Mauss, A General Theory of Magic, trans. by Robert Brain (New
York: W. W. Norton, 1972), esp. pp. 63-74. For another classical analysis of the logic and
coherence of magic, see E. E. Evans-Pritchard, Witchcraft, Oracles and Magic among the
Azande (Oxford: Clarendon Press, 1937).
22. Mauss, A General Theory of Magic, p. 63.
What is for us in the end most striking about magical practices is that they
require assumptions which in one way or another run counter to the
categorical framework within which we (at least officially) interpret the
world: as with the notion of a real identity between symbol and thing
symbolized. . . . In this lies their interest, and the strangeness which is
from our point of view their common characteristic.”3
To speak, then, of magical equations in Vedic ritual and philoso-

phy, is at the very least to indicate their utter foreignness and difference.
Even recent, sophisticated, and sympathetic treatments of Vedic ritual-
ism are mitigated in their value when they resort to these assumptions
and nomenciature. Thus, when Asko Parpola speaks of “magical equa-
tions, which are the most characteristic feature of the Brahmana texts”
and which are “exploited” or “manipulated” within the ritual to give
priests “control” over phenomena,” or when Michael Witzel asserts that
“magical identifications” are meant as “a full equation of both entities
and not to a mere similarity of both,” it is the problematic nature of
Vedic “magic” that is being emphasized.
Magic—as a categorical term—has also functioned in Indological
discourse to imply that one should distinguish Vedism from proper reli-
gion. Sylvain Lévi, whose book La Doctrine du sacrifice dans les
Brahmanas has remained the standard for nearly a century, referred to
the Vedic sacrifice as “a magical operation” complemented by a “brutal”
and “material” theology. “Morality,” wrote Lévi, “had no place in the
system: the sacrifice which ruled the relations between men and gods is a
mechanic operation which acts by its own energy.”6 By presenting
Vedic ritualism as without morality, without real theology, indepen-
23. John Skorupski, Symbol and Theory: A Philosophical Study of Theories of Reli-
gion in Social Anthropology (Cambridge: Cambridge University Press, 1976), p. 159.
24. Also Parpola, “On the Symbol Concept of the Vedic Ritualists,” in H. Biezais,
ed., Religious Symbols and Their Functions (Stockholm: Almquist and Widsell, 1979),
esp. p. 140.
25. Michael Witzel, “On Magical Thought in the Veda,” lecture published by
Universitaire Pers Leiden, 1979, esp. pp. 5, 10, 12. Witzel’s article, valuable as it is as an
analysis of Vedic connections, is at its best when it addresses the comparative questions,
e.g., p. 12: “This way of thinking, strange as it may seem, is by no means foreign to present
day Western man.”
26. Lévi, La Doctrine du sacrifice dans les Brahmanas, p. 9. See also p. 129: “The
sacrifice has thus all the characteristics of a magical operation. It is independent of deities,
effective through its own energy and susceptible to producing evil as well as good. It can be
only slightly distinguished from magic proper by its regular and obligatory character; it is
easy to direct it toward diverse goals, but it exists and imposes itself independently of
circumstances.”
dently or automatically operating apart from either gods or humans,

Lévi succeeded in persuading many of his heirs that Vedism was magi-
cal: both strange and not deserving of the name religion.
One way, then, that Vedic ritualism (as a kind of definitional ru-
bric) has been simultaneously depicted and (intentionally or not) depre-
cated was to use magic as a synonym for ritual—to highlight either its
epistemological eccentricity or its spiritual insolvency. Recently, a new
version (albeit with innovative twists) of this same project has been
envisioned or, rather, devised.
The Meaningless Theory
Over the past several years—in a series of articles and published lec-
tures, culminating in the mammoth Agni: The Vedic Ritual of the Fire
Altar?7—Frits Staal has put forth a representation of Vedic ritualism
which functions as the cornerstone of his new general theory of ritual.
Dismissing all previous and competing theories, Staal argues that ritual
“has no meaning, goal, or aim” and is best described as “pure activity.”8
Ritual is neither symbolic nor meaningful activity but rather is purpose-
less and meaningless: “To say that ritual is done for its own sake is to say
that it is meaningless, without function, aim or goal, or also that it
constitutes its own aim or goal.”?°
This remarkable thesis is supported solely by the Vedic exemplum;
here, apparently, Staal finds all the evidence he needs to reconstitute the
“meaning” of the general category of ritual, for no other examples are
anywhere given. He cannot make even the Vedic ritual “meaningless,”
however, without considerable methodological acrobatics.
At some points in his work, Staal asks us to imagine that meaning-
less (purposeless, useless) ritual is the historical result of the obsoles-
cence of once meaningful activities.*° In the case of Vedism, formerly
purposeful activities such as battles, cattle raids, and their supporting
27. My review of this work and an extended version of the critique that follows here
may be found in “Vedic Fieldwork,” Religious Studies Review 11 (April 1985): 136-45.
28. Staal, “The Meaninglessness of Ritual,” pp. 8-9.
29eibidyiprgs
30. “Typically human forms of ritualization seem in general to dissolve meaning, not
replace it.” Ibid., p. 13. For Staal’s speculations on the origin of rituals connected to the
maintenance of fire (“The carrying around of fire became a ritual activity as soon as it was
no longer necessary, viz., as soon as methods for kindling fire had been discovered”), see
Staal, Agni, I: 84, 102.
acts (songs, recitations, and rituals) became meaningless as the ancient

Vedic militaristic adventurers parked their chariots and made them-
selves at home in India:
When the nomads began to settle in the subcontinent and established
lasting relationships with the indigenous inhabitants, the expeditions re-
ceded to the background and the accompanying activities lost their accom-
panying functions. ... Thus ritual was the sole survivor and all other
activities were ritualized. . . . Rituals tend to absorb everything that has
gone before and has lost its original function.!
Staal does not allow for the possibility that “obsolete” activities acquire
new meanings over time by being ritualized, but he concludes that by
losing their original meanings the activities must carry no meaning for
those who have ritualized them.”
In the Vedic instance, however, there is a major obstacle to such an
explication: the Brahmanas, texts (late in Staal’s scheme of things) that
have as their very purpose the explanation of the meanings of the “pur-
poseless” activities of the ritual. Faced with these texts, Staal reverses
his quasi-historical argument. Instead of claiming that a meaningless
ritual arose from formerly meaningful activity, the theorist now (or in
addition) wants to say that originally pure and meaningless activity is
over time encrusted with meaning:
Much later, when ritual was contrasted with ordinary, everyday activity,
its meaninglessness became patent and various rationalizations and expla-
nations were constructed. Ritual became deeply involved with religion,
which always stands in need of the mysterious and unexplained. .. . In
the course of time rituals, instead of remaining useless and pure, became
useful and meritorious.*
Further, the “rationalizations” one finds in abundance in the Brahmanas
are explained away, in time-honored Indological terms, as arbitrary,
contradictory, and, once again, “fanciful.”** These subsequently canoni-
cal Vedic texts mostly provide “piecemeal interpretations,” and when
they attempt “large-scale interpretations of ritual . . . these can often be
shown to be failures.”35 Although the route taken is different, the end is
31. Staal, Agni, I: 102-3.

32. For further remarks on this methodological error, see Chapter 8 herein.
33. Staal, “The Meaninglessness of Ritual,” p. 14.
34. Ibid., p. 4; Staal, Agni, I: 27-28, 67-68; idem, The Science of Ritual,
pp. 11-12, 36.
35. Staal, Agni, I: 63; Idem, The Science of Ritual, p. 33.
the same: the Brahmanas are castigated and the Vedic ritual shown to
be, by normal standards, absurd.
Another version of the “meaninglessness of ritual” school of Vedic
studies has been somewhat more persuasively propounded by J. C.
Heesterman.** Heesterman posits a period before the Veda received its
final redaction; behind the received texts, in the shadows of classical
Vedism, is a “preclassical” age. The pattern of life established by the
earliest Indo-European invaders of Northwest India was one of alterna-
tion between settled agriculturalist existence and raids on the fields and
property of the peoples to the east in the lean season when the home
crops lay dormant. The preclassical Vedic Indians in this way ricocheted
between their settlements (the grama) and the eastern wilds (the
aranya), and between the roles of householder (Salina) and wanderer
(yayavara), the domesticated man-in-society and the raiding warrior.
This cycle, Heesterman argues, also describes the structure of the
preclassical sacrifice, as the munificent sacrificial patron—the yajamdana
who gives gifts to his guests and sacrificial oblations to the gods—
alternatively becomes a consecrated warrior—the diksita who recoups
his wealth in raids, contests, and battles. The sacrifice, the pivot of the
cycle, climaxes with the slaughter of an animal, a communal banquet,
and a distribution of gifts and prizes—an event equated with the rebirth
of the sacrificer as he sheds onto his guests and rivals the impurity
accrued by his engagement in violence and his responsibility for killing.
The sacrificer, however, is then left emptied of material goods, which
must be reacquired through violent confrontations and attendance at
sacrificial contests sponsored by others. The cycle of stifling and danger-
ous interdependence and alternation repeats itself endlessly.
The “inner conflict” of the preclassical system is finally reducible
to the problematics of killing and violence. There was, on the one
hand, the necessary nastiness of the warrior’s life—raiding and battling
in the wilds because, in a “world of scarcity,” the home-grown yield
had to be supplemented. On the other hand, there was the violence
and killing connected with the other half of the cycle, the offering of
the sacrifice by the engorged patron. The sacrifice was the site of
contests for the “goods of life”; Heesterman goes so far as to say that
the place of sacrifice was originally a battleground where one fought
for fire, food, and cattle: “Such fights may have been prearranged,
36. See esp. the essays collected in J. C. Heesterman, The Inner Conflict of Tradi-
tion. For a more developed description and critique of what follows, see my “Ideals and
Realities in Indian Religion.”
ceremonial, and conventional, but they were none the less violent be-
cause of it.”>” Furthermore, the preclassical sacrifice culminated in the
violent death of the sacrificial victim. Thus, together with the uncertain-
ties and dangers of reciprocity, cyclical exchanges, and violent encoun-
ters and contestations, the preclassical sacrifice held within it an insolu-
ble paradox: “The life-giving ritual is intimately connected with life’s
opposite, with death and destruction.”38
Having in this way reconstructed both the socioeconomic conditions
and the religious practices of a preclassical age, Heesterman then pre-
sumes a radical reformation, an “axial breakthrough.” The author occa-
sionally hints that the revolution may have been stimulated by changing
patterns of population distribution and economic advancement,?? but his
more fundamental assumption is that the transformation of the tradition
was “orthogenetic,” propelled by the tensions inherent in the preclassi-
cal pattern outlined above. In any event, Heesterman maintains that the
transformation was sudden and studied; it was not “a gradual and cumu-
lative process of erosion but a conscious reform.”
The reformation of the preclassical sacrifice was achieved by cutting
up the cyclical paradigm and reorganizing the ritual in a linear fashion.
In place of endlessly alternating phases, the ritual was made to have a
clearly delineated beginning and end. This simultaneously eliminated
the dependence on the participation of the rival/guest and the necessity
for exchange and reciprocity. The place of the rival/guest was taken over
by a fixed priesthood which was assimilated to the sacrificer in an overall
attempt to unify the ritual, and the system of exchange was replaced by
the belief in the automatic efficacy of the ritual itself.
The secret means the ritualistics developed to effect this revolution
was the system of connections (bandhus) which brought to an end the
dependence on the other, the uncertainties of the sacrifice, and the
violence and death that were at its center. What was once a sacrificial
battleground “has been turned into a serene and perfectly ordered ritual
emplacement.”*! Violence and death, the paradoxical centerpieces of
the preclassical sacrifice, were exorcised, and the problematics of killing
gave way to an obsessive concern for ritual exactitude and the fear of
37. J. C. Heesterman, “Other Folks’ Fire,” in Staal, Agni, II: 85. See also idem,
“The Ritualist’s Problem,” in S. D. Joshi, ed., Amrtadhara: Professor R. N. Dandekar
Felicitation Volume (Delhi: Ajanta Books, 1984), p. 172.
38. Heesterman, The Inner Conflict of Tradition, p. 68.
39. See ibid., pp. 106, 124-25.
40. Ibid., p. 91.
41. Ibid., p. 101.
error: “Uncertainty and catastrophe were replaced by the fail-safe cer-

tainty of ritualism.”4
But competition, conflict, violence, and the strains of mutual
dependence—that is, the realities of living in society—could not be eradi-
cated, and indeed the reformation did not aim at doing so. They could,
however, be stripped of religious meaning and legitimacy. The axial
breakthrough “denied conflict its ultimate legitimacy. Instead it posited
the ideal of an absolute transcendent order that is, as a matter of princi-
ple, incapable of worldly realization.”*# A transcendent ritual of the
Brahmin was created out of the deconstructed elements of the warrior’s
sacrifice.
This historical narrative of the transition from preclassical to classi-
cal India implies a general theory of sacrifice and ritual. And there is,
Heesterman claims, a critical difference between the two. Sacrifice re-
fers to rites in which offerings and exchanges are accompanied by violent
death in an atmosphere of uncertainty, risk, and “controlled catastro-
phe.” Ritual is, in many ways, the opposite of sacrifice.
The destruction of part or all of the oblation to the gods sets sacri-
fice apart from gift-giving transactions. Sacrifice is defined by Heester-
man by its violent center, by its concern with the harsh realities of life
and death: “Its primary material is what sustains life, that is: food. But
food—its acquisition, preparation and consumption—equally involves
death. Part of the food, therefore, even if it is only an infinitely small
part, has to be destroyed.”“ As an exercise in “controlled catastrophe”
with uncertain results—‘It ends either in success or in failure, in trium-
phantly enriched life or in loss and death”**—sacrifice is not orderly and
ritualistic. Because sacrifice “turns on destructive violence” it “over-
throws order.”*°
With this emphasis on the violence of sacrifice, Heesterman joins
recent theorists such as Walter Burkert and René Girard, who have also
pinpointed the act of killing as the meaningful core of the sacrificial
ritual.47 Heesterman stands apart from both these other theorists, how-
42. Ibid., p. 91.

43. Ibid., p. 157.
44. J. C. Heesterman, “Veda and Society: Some Remarks apropos of the Film ‘Altar
of Fire,’ ” in Asko Parpola, ed., Proceedings of the Nordic South Asia Conference, June
10-12, 1980 (Helsinki: Studia Orientalia, 1981), p. 55.
45. Heesterman, The Inner Conflict of Tradition, pp. 84-85.
46. Heesterman, “The Ritualist’s Problem,” p. 168.
47. Walter Burkert, Homo Necans: The Anthropology of Ancient Greek Sacrificial
Ritual and Myth, trans. by Peter Bing (Berkeley: University of California Press, 1983);
ever, in his refusal to burden the act of sacrificial killing with either
symbolic or emotional meaning. It is purely the actual and actualized
logical conundrum and social problem of the necessity of death for the
continuation of life that Heesterman posits as the mysterious heart of
sacrificial darkness.
The systematized, individualized, sanitized, and idealized ritual
codified in the ancient Vedic texts is radically different from sacrifice
defined in such a way. Whereas sacrifice is depicted as “popular or
organicist,” ritual is described as “systematized or mechanistic.” The
ritualism exemplified in the Vedic texts, in Heesterman’s opinion, is
characterized by its lack of risk and uncertainty—that is, by its inten-
tional and complete divergence from reality. Whereas sacrifice is “tied
up with the life of the community in a comprehensive, many-stranded
web of meaning,” the Vedic ritual “is simply separate, unrelated to
anything outside itself.”48 Ritual is an ersatz world of perfection, free
from contingencies, closed and controllable.
The invention of a transcendent ritual was intended to carve out a
place where human beings could become emancipated from the human
condition, from reality and its insoluble paradoxes. The price paid, how-
ever, for the perfection of the ritual order was its complete divorce from
the world. Perfection could “only be realized outside society and for the
limited duration of the ritual.”4? The movement from sacrifice to ritual
is, in sum, a movement from problematic reality to imaginary ideals. But
by virtue of its transcendence of reality—that is, by virtue of its irrele-
vance to the world—the ritual has also been rendered meaningless: “It is
a closed system that has meaning only in itself, in its own inner order—
an order that is as strict as it is artificial. In this sense one may agree with
Professor Staal’s thesis of the meaninglessness of ritual.”
It is quite clear that, for all of Heesterman’s argumentation and
marshalling of evidence, the preclassical age of Indian history is the
René Girard, Violence and the Sacred (Baltimore: Johns Hopkins University Press, 1977).
Briefly and oversimply, Girard argues that violence endemic to human society is displaced
finally onto the sacrificial victim through a complicated series of substitutions (see Chapter
7 herein). Burkert sees sacrificial killing as a dramatization of the hunt and, eventually, of
social survival in general. Consult also the recently published transcription of a symposium
on the subject of sacrifice and violence, Robert G. Hamerton-Kelly, ed., Violent Origins:
Walter Burkert, René Girard, and Jonathan Z. Smith on Ritual Killing and Cultural Forma-
tion (Stanford: Stanford University Press, 1987).
48. Heesterman, “Veda and Society,” p. 54.
50. Heesterman, “The Ritualist’s Problem,” p. 175.
scholar’s imaginative construct, as is the axial breakthrough on which

Heesterman pins so much. Why the ritualists who so carefully put to-
gether a “classical” Vedic ritual would “be concerned with the preserva-
tion of obsolete diverging traditions . . . even at the price of contradict-
ing the laboriously developed system”! is not at all obvious and in fact
runs counter to the author’s assumption that the ritualists consciously
and systematically obliterated the preclassical pattern in the process of
reinventing the sacrifice. Positing a preclassical age and an axial break-
through serves primarily to establish an absolute distinction between a
romanticized sacrifice and a meaningless ritual.
Sacrifice for Heesterman appears to be simultaneously realistic and
chaotic. Sacrifice simply enacts a real struggle for incompatible goals:
the goods of life, won by violence which entails impurity and evil; and
religious purity which is gained at the expense of disastrous material
dissipation. On the other hand, the success of the sacrifice is entirely
unpredictable and wholly dependent on the cooperation of others, espe-
cially on their willingness to reciprocate—just like action in the real
world. This vision of sacrifice, like that of the preclassical age as a
whole, is an idealized scenario of primordial reality before the invention
and implementation of ideals, before culture and its symbols (including
symbolic actions such as rituals) were overlaid upon—or set next to—
raw human life.
On the other side of the divide, Heesterman’s ritual—meaningless,
transcendent, ideal, and perfect—is also a creation beyond discussion,
beyond disputation, and beyond interpretation. Indeed, discussion, dis-
putation, and interpretation, according to Heesterman, are precisely
what the Vedic ritual seeks to eliminate. This may or may not be true for
the Vedic ritualists (and the point is arguable), but it is disingenuous for
historians of religion to reposition such notions as scholarly conclusions.
Moreover, Heesterman is not in fact simply transcribing native assump-
tions. It may not be too arcane to suggest that any thesis for the meaning-
lessness of any human creation is a meaningful interpretation of what is
regarded as a previously improperly understood phenomenon—and thus
is itself a paradox that cries for a solution.*?
Heesterman does avoid some of the problems of more doctrinaire
versions of the “meaninglessness of ritual” theory. Heesterman notes, as
others of the school do not, that the Vedic ritualists—with their “science”
of rules and controlled actions—could not succeed in their quest for per-
51. J. C. Heesterman, “Vratya and Sacrifice,” Indo-Iranian Journal 6 (1962): 3.

52. I return to this point in more detail in Chapter 8.
fection and exactitude.*3 Ritual, no less than human life outside its con-
fines, is a product of fallible beings who cannot escape their nature, who
cannot transcend the human condition, who cannot realize their ideals:
“We find that Vedic ritual in fact undermines its own claim to be the
absolute universal order. In the first place, one can, of course, never be
completely sure that one has not unwittingly committed an error in the
ritual proceedings, which, if unrepaired, will irretrievably impair the rit-
ual order. Thus there is always an element of uncertainty.”
Indeed there is, and contemporary theorists of ritual sometimes
ignore what the ritualists knew all too well. Ritual cannot be an ideal-
ized zone in which perfection is possible, for this sphere of activity, like
all others, cannot be totally regulated and controlled—as we shall have
occasion to observe in Chapter 4. The history of Vedic ritualism—a
history too often left untraced by those with other axes to grind—is
partly impelled by the fact that, even in ritual, perfection is impossible
in this life. Ritual manuals or prayogas, which attempt with ever in-
creasing specificity to account for all eventualities, for all the minute
detail of performing the ritual, were continuously produced through
the millennia right up to the present. The enterprise is doomed to
failure—there are too many possible contingencies to address, too
many details to bring under control—but the tragically heroic attempt
is one underexposed aspect of the destiny of the Vedic sacrifice in
Indian history.
The second major criterion for constituting a Vedism, then—and
really one that is complementary to the first—is its distinctive overriding
concern with its particular system of ritual practice and ideology. But,
again, as with the closely related first criterion (epistemological homolo-
gies), ritualism is made a definitional feature of Vedism in order to
evaluate it and, most often, to devaluate it. It is represented in Indologi-
cal discourse as the magical practices of different minds, the coercive
spells of those without true religion, the pure activity of those who can
do things without thought, or the realm of an artificial perfection created
by those who have left reality altogether. The overactive Vedic imagina-
53. One might contrast this observation with Jonathan Z. Smith’s notion that “ritual
represents the creation of a controlled environment where the variables (i.e., the acci-
dents) of ordinary life may be displaced precisely because they are felt to be so overwhelm-
ingly present and powerful. Ritual is a means of performing the way things ought to be in
conscious tension to the way things are in such a way that this ritualized perfection is
recollected in the ordinary, uncontrolled, course of things.” “The Bare Facts of Ritual” in
Imagining Religion, p. 63.
tion engenders, as Oldenberg wrote, “a second world” in which occurs

“a subtle empty mummery”; or, as Lévi insisted, an amoral machine; or,
as Staal and Heesterman would have it, a world of meaningless activity,
occasionally overlaid with “fanciful” native exegesis. Again, we must
ask, Whose imagination is it that is really overactive? Whose ritual—and
theory of ritual—is really meaningless to whom?
Rethinking Vedism
Any attempt to reimagine Vedism more valuably—and more realisti-

cally in light of the importance of the Vedas to the Hindu religion—must
begin with the reexamination of these two definitional data: epistemo-
logical connections and ritualism. The two are interlinked. Problematic
representations of Vedic ritualism often appear to be caused in large
part by assumptions or conclusions about Vedic homological thinking,
and vice versa.
In following chapters, I will show that Vedic “equations” are nei-
ther absurd nor random but are rather systematic expressions made
possible (and logical) by fundamental Vedic principles of metaphysics
and epistemology—of how the world is and how humans know and
represent their knowledge. I will also reevaluate the place of ritualism in
Vedic religion. I will argue that ritual action was presented in Vedic
texts not as symbolic or dramatic playacting, magical hocus-pocus, or
“pure,” transcendent, or meaningless activity. Rather, the sacrifice was
displayed as a constructive activity, creating the human being (ontol-
ogy), the afterlife (soteriology), and the cosmos as a whole (cosmology).
It was also, of course, a social instrument—constructing individuals as
part of a class and defining both the classified individual and the classes
themselves from within the universe of the ritual. In sum, I will maintain
that Vedism was a coherent and comprehensive system of doctrine and
practice.
Vedic philosophy turns on the assumption that it is possible to
correlate corresponding elements lying on three discrete planes of real-
ity: the macrocosmos (whose contents and forces are collectively called
adhidevata, “relating to the godly”) the ritual sphere (adhiyajna, “relat-
ing to the sacrifice”), and the microcosmos (adhydtman, “relating to the
self”). And, as Jean Filliozat, among others, has observed, connections
made and found among the cosmos, ritual, and self were not designed
solely to reduce these levels to any one, nor were they simple meta-
phors. Such homologies were to express metaphysical truth and to make

all three levels mutually explicative.°5
Furthermore, the credibility of the Indological notion that these
connections were random and whimsical has recently received a blow in
the form of a detailed study of one text by Klaus Mylius. Limiting
himself to a thorough analysis of the Kausitaki Brahmana, Mylius has
shown the homologies posed in that text to be consistent, principled,
and interlocking.*°
What remains largely undone is the full articulation of principles
that guided Vedic connections; the revelation of the scope, variety, and
nuances and practice that were made possible assuming those principles;
and the drawing out of the implications for our knowledge of both
Vedism and the later Indic religions that regard themselves as perpetuat-
ing it.
I would suggest here, and will be at pains to prove throughout the
remainder of this study, that there is a philosophical center around
which all Vedic thought resolves. That center I will call resemblance.
Although Vedic writers do not utilize a Sanskrit word that adequately
translates as “resemblance” until the time of the composition of the
ritual Sutras (s@mdnya; see Chapters 5 and 7 herein), the concept, I
believe, underlies Vedic religious and philosophical discourse in its entir-
ety. Vedic resemblance is, in sum, not unlike what Michel Foucault has
called an “episteme,” a central principle or rule that generates and
governs knowledge of all sorts.
Universal resemblance, whereby entities, things, forces, activities,
cosmic planes—indeed, all the components of the universe as a
whole—have essential affinities to related others, helps us to reform
our understanding of the Vedic preoccupation with making and finding
connections. I will maintain that to call these connections or bandhus
equations or equivalences is to mistake the relation of resemblance for
a relation of identity (which Vedic thought, no less than modern schol-
arly thought, usually found abhorrent, although perhaps for somewhat
different reasons).
Further, Vedic resemblance, as an epistemological, metaphysical,
55. Jean Filliozat, “La Force organique et la force cosmique dans la philosophie
médicale de |’Inde et dans le Véda,” Revue philosophique 116 (1933): 410-29.
56. Klaus Mylius, “Die vedischen Identifikationen am Beispiel des Kausitaki-
Brahmana,” Klio 58 (1976): 145-66; and idem, “Die Identifikationen im Kausitaki-
Brahmana,” Altorientalische Forschungen 5 (1977): 237-44.
and ontological principle, itself resembles “epistemes” well known in the

history of Western thought.5? Although the Vedic version of resem-
blance is distinctive in ways to be discussed shortly, it is not a bizarre,
incomprehensible bit of exotica; it is, rather, an ancient Indian example
of a well-known phenomenon in the history of forms of knowing, saying,
and acting.
This Vedic principle, again not unlike comparable systems center-
ing on resemblance elsewhere, operates within and is modified by hierar-
chy. The hierarchical nature of Vedism is only recently becoming clearly
visible and articulated in Indological discourse—one of the many wakes
created by Louis Dumont’s already classic work, Homo Hierarchicus.*
Dumont argues that the Hindu caste system is made possible by a body
of assumed ideas and values, an ideology of holism, interdependence,
and hierarchy. Furthermore, Dumont claims, a single principle governs
the hierarchically ordered caste structure and ideology: the principle of
relative purity based on the opposition between the pure (represented
by the Brahmins) and the impure (represented by the Untouchables).
Castes are ranked according to their relative purity and impurity—that
is, how closely they resemble the Brahmins on the one hand and the
Untouchables on the other. Finally, in this hierarchical system, the
higher castes are said to “encompass” those lower. Although Dumont
does not fully work this out, what seems to be implied here is an ontol-
ogy of “relative completeness,” the Brahmin being the “more complete”
instance of the human being while others, relative to the Brahmin, are
“less complete.”
Dumont did not apply his insights into the hierarchical nature of
Hindu caste—and Hinduism as a whole—to the pre-Hindu Vedic reli-
gion. It is clear that Dumont’s formulation must be adjusted to the rather
different circumstances of Vedism (“purity” and “impurity,” for exam-
ple, are certainly present but far less central), but it is also increasingly
obvious that Vedism, no less than Hinduism, is organized hierarchically.
One step toward reenvisioning Vedism as systematically ordered on
57. A partial bibliography of resemblance in Western epistemology would include

Michel Foucault, The Order of Things: An Archeology of the Human Sciences, trans. of
Les Mots et les choses, (New York: Vintage Books, 1973); Arthur Lovejoy, The Great
Chain of Being: A Study of the History of an Idea (Cambridge, Mass.: Harvard University
Press, 1936) Ralph Withington Church, An Analysis of Resemblance (London: George
Allen and Unwin, 1952); and Panayot Butchvarov, Resemblance and Identity: An Examina-
tion of the Problem of Universals (Bloomington: Indiana University Press, 1966).
58. Louis Dumont, Homo Hierarchicus: The Caste System and Its Implications,
trans. by Mark Sainsbury et al. (Chicago: University of Chicago Press, 1980).
2 Constructing Vedism 49
a hierarchical basis has been taken by Frits Staal, who has demonstrated
the hierarchical nature of the Vedic repertoire of rituals. Concentrating
on the “syntax” and denying any “semantics” (rituals, it will be recalled,
are for him meaningless), Staal rightly insists that “We must start with
the observation that the srauta rituals constitute a hierarchy.”5? The
ritual system forms a continuum or sequence, beginning with the most
simple and ending with the most complex. And, Staal hints (but does not
further explicate), there is a correlated hierarchical order of ritualists
based on relative competency and experience:
This sequence [of the ritual order] is not arbitrary. There is increasing
complexity. A person is in general only eligible to perform a later ritual in
the sequence, if he has already performed the earlier ones. Each later
ritual presupposes the former and incorporates one or more occurrences
of one or more of the former rituals. Sometimes these embedded rituals
are abbreviated. In general, they undergo modifications.
Staal argues here, in ways highly reminiscent of some of Dumont’s

points regarding the caste hierarchy, that the ritual system is ordered
on the basis of relative complexity of form and relative inclusivity of
structure—that is, on the basis of what I have called relative comple-
tion. Hierarchically superior rituals are those that incorporate within
themselves a number of inferior rituals, recapitulated in the ritual in
representative, albeit abbreviated and condensed, forms. The Vedic
ritual hierarchy, as in the caste hierarchy as explained by Dumont,
operates on the basis of the encompassment of the inferior (the “em-
bedded” rituals) by the superior and more complete on a scale of
mutually resembling elements.
This is an important beginning, but it is only a beginning. Hierarchi-
cal resemblance is a principle with far more hermeneutical possibilities
for the study of Vedism—and, I might add, for the study of Hinduism as
well—than have yet been realized. Furthermore, the importance of rit-
ual practice has not yet been integrated into this developing picture of
Vedism. The questions that will guide us in our rethinking of Vedism
are: how might the concept of hierarchical resemblance help us to
reimagine the religion in its entirety, and why did ritual activity so
preoccupy the authors of the Vedic texts?
59. Staal, “The Meaninglessness of Ritual,” p. 15.

60. Staal, “Ritual Syntax,” p. 125.
3
Ritual and Reality
Natural Chaos and Ritual Order
Reality, according to the Vedic savants, is not given but made. These
early philosophers of the sacrifice were ritualists through and through.
Their texts convey the sensibilities of those who endowed their intri-
cate ceremonial with enormous importance. Sacrifice, for them, was
not primarily on honorific gift-giving exchange between gods and hu-
mans. Neither was it regarded as a symbolic representation of an al-
ready concretized reality. The Brahmin ritualists themselves certainly
did not understand the sacrifice to be meaningless activity done for its
own sake or as a ritualized realm of antireality. Rather, the ritual was
the workshop in which all reality was forged.
This study of Vedic ritualism and the principle of resemblance that
guides its operation and organization is premised on the recognition of a
fundamental Vedic assumption: what is natural is inherently defective;
or, as Lilian Silburn has put it in her invaluable study, what is “immédi-
atement donné est discontinu et qu’il n’est d’autre continu que du struc-
turé.”! From this perspective, the natural is the chaotic, the disorga-
nized, the unformed. In cosmological terms, what is merely procreated
by the creator god is not a cosmos or a universal whole made up of
ordered parts. The origins of true cosmos are found not in this primary
generative act but rather in a secondary operation—a ritual act that
lends structure and order to a chaotic creation.
Mircea Eliade tirelessly argued that “every creation has a paradig-
1. Silburn, /nstant et cause, p. 1.

Ritual and Reality 51
matic model—the creation of the universe by the gods.”? but the Vedic
instance seems to elude the archetypical by posing a radical disjuncture
between divine creation and sacred cosmos. In the beginning, the cre-
ative act of “emission” by the Lord of Creatures, Prajapati, is not a
cosmogonic paradigm of sacred order but rather what Silburn rightly
calls a “profane act.” Put otherwise, cosmic procreation, in the imagina-
tions of the Brahmins, does not engender a ready-made universal order
but results in a problematic metaphysical excess. Similarly (as will be
detailed in the next chapter), at the level of individual human begin-
nings, birth and anthropogony are distinct and separate moments, the
first being only the necessary precondition for the second. As cosmic
creation is not cosmogony, biological reproduction is not the production
of a true human being.
It is characteristic—and perhaps also close to definitive—of Vedism
that between mere procreation one the one hand and true cosmogony
and anthropogony on the other is inserted a set of constructive rituals.
Between Prajapati’s creation and the origin of the cosmos are sacrificial
acts of the gods, giving form to formless nature. And between the procre-
ation of every person and the origin of true being are also rituals, mak-
ing a human out of the human in potens only. Cosmogony and
anthropogony in Vedic ritualism are actualized only within the sacrifice
and realized only by ritual labor or karman.
For the Vedic priests and metaphysicians, ritual activity does not
“symbolize” or “dramatize” reality; it constructs, integrates, and consti-
tutes the real. Ritual forms the naturally formless, it connects the
inherently disconnected, and it heals the ontological disease of unrecon-
structed nature, the state toward which all created things and beings
perpetually tend.
All viable forms, all properly structured beings, are found only in
the ontological space between what Jean-Marie Verpoorten has called
“two symmetrical excesses.”* The first is denoted in Vedic texts with
the word jdmi, describing the unproductive condition of homogeneity
or redundancy—an excess of resemblance.° The term is applied to
2. Mircea Eliade, The Sacred and Profane, trans. by Willard Trask (New York:
Harcourt, Brace and World, 1959), p. 31.
3. Silburn, Instant et cause, p. 54.
4. Jean-Marie Verpoorten, “Unité et distinction dans les spéculations rituelles
védique,” Archiv fiir Begriffsgeschichte 21 (1977): 59-85.
5. For a study of a similar, but not identical, Vedic concept—atirikta (“superfluous”
or “redundant”)—see Jan Gonda, “The Redundant and the Deficient in Vedic Ritual.”
Vishveshvaranand Indological Journal 21 (June—December 1983): 1-34.
various but comparable situations. In the ritual vocabulary, jami desig-

nates rites that are fruitlessly reduplicative within a ritual sequence:
“There is jami in the sacrifice when on the same day he employs the
same verses as offering verses. Even if there are many wives, as it were,
one husband is [necessary for forming] a proper pair with them.”¢ The
word also appears in Vedic kinship terminology to describe those who
are closely related to oneself to marry. A. C. Banerjea concludes a study
of the term by noting that jami “in all probability is a common designa-
tion for the members of a non-marriageable group, and as such non-
approachable by those who abided by the exogamous system.”’’ Here
again, jami connotes the excess resemblance or lack of sufficient differen-
tiation between two or more entities and the attending unproductive
nature of the relation. Yet another application of the term is found in
the Jaiminiya Brahmana, in which the undifferentiated and sterile qual-
ity of jami is likened to homosexual union, the barren joining of those
too alike:
An improper pairing and unproductive is jami, as is the fruitless coupling
of two men or two women. . . . That, on the other hand, which is not jami
is a productive and proper pairing. ... What is uniform [ekaripa] is ©
incapable of copulation and is unproductive.’
If fecundity requires a mitigation of excessive resemblance or unifor-

mity, it is also possible only if separate elements are capable of joining
with others. Complementary to, and the inverse of, the nonontological
state of jami is that of total distinction, isolation, and utter separation—
the excess of differentiation known as prthak. This and similar terms,
such as bahutva (“fragmented multiplicity”) and ndndatva (“unrelated
variety”), refer to a condition of overdiversification, unconnected disper-
sion, and atomism which is as inimical to true being as are homogeneity,
reduplication, and redundancy. Things and entities must be differenti-
ated in order to avoid the quality of jami, but they must also be con-
nected to escape the equally dangerous, and ultimately lifeless, condi-
tion of prthak.
6. AitB 3.47, 3.48. For other examples of the ritual employment of the term and
concept of jami, see KB 3.6, 8.8, 13.9, 28.5, 30.11; PB 7.2.5, 8.8.10, 10.4.7, 14.3.17,
16.5.1; SB.13.28, 1.81.25 223 11,223.77 4990 A724 ie os Bre ae
3.3.4.6, 3.4.19.9, 3.7.5.13; and Verpoorten, “Unité et distinction,” p. 68. At LSS 9.11.15,
it is called a “fault” (jamidosa).
7. A. C. Banerjea, Studies in the Brahmanas (Delhi: Motilal Barnarsidass, 1963),
p. 31.
8. JB 1.300, 1.330.
In Vedic ritualism, as Silburn observes, “nonbeing is that which is

devoid of connection,” whereas “being appears to be precisely that
which is welded together. . . .”? Both jami and prthak depict condi-
tions off the ontological scale—symmetrical and opposing states of
nonbeing—because they describe situations in which this ontology of
composition and connection is impossible. That which is jami precludes
production of true being because there are no sufficiently distinct com-
ponents to join together; that which is prthak precludes production
because no connections are possible between overly individualized com-
ponents. All true being locates itself (or, rather, is made and placed)
between these twin excesses of identity and isolation. Ontological for-
mation is the result neither of the reduplication of the same nor of the
radical differentiation of the dispersed many but is the consequence of
the construction of an integrated unity out of distinct but interrelated
parts.
This brief overview of the groundwork of Vedic metaphysics and
ontology already points to the real logic of the “homological” thinking
and the rationale behind the “obsession” with ritual activity so character-
istic of the religion. Vedic connections could not have been posed as
simple and confused identifications (for here would be the excessive
resemblance the Brahmins called jami), and I have intimated why con-
nections between otherwise unrelated elements must be made (to avoid
the excessive differentiation termed prthak). Further, the critical impor-
tance of ritual activity is inexorably bound up with both of these consider-
ations, for it is the ritual labor that effects both separation and unifica-
tion of parts into wholes.
These points will be further illustrated in the following pages, first in
an exposition of the Vedic cosmogonic myths which provide a paradig-
matic blueprint of the origins of present metaphysical conditions and the
raison d’étre for sacrificial action. Next, I will go into some detail on the
question of Vedic connections and the forces at work in the universe that
counteract disintegration and chaos and make possible linkages between
resembling forms and counterforms, prototypes and counterparts. The
metaphysics of resemblance, in turn, is the condition of possibility for the
efficacy of ritual work. Cosmic prototypes were thought to be manipu-
lated by ritual operations performed on accessible counterparts. Vedic
ritualism was directed toward activating the connections that bind the
ritual world to the world(s) at large; the ritual order lends its form to a
cosmic order, a universal structure emanating from the structured se-
9. Silburn, Instant et cause, p. 44.

quence of rites. Such grand results, however, depend on the perfect per-
formance of the ceremonial. In the construction of a ritual production,
each detail must be attended to and properly executed. The sacrificial
solution to the problem of chaos creates its own set of new problems,
turning on the impossible demand the ritualists made of themselves for
perfection in their sacrificial creations. For, as we will see in the next
chapter, the enterprise which the Vedic ritualists set out on is depicted in
their texts as a “dangerous ascension.”
Creation Is Not Cosmos
In the post-Rgvedic texts, the locus classicus of the developed ritualism

typifying Vedism, the function and pride of place of creator gods in the
Rg Veda Samhita (and, most especially, the god Purusa, the sacrificial
hero of RV 10.90)! is usurped by the deity Prajapati, Lord of Creatures.
Or, rather, the imagery, philosophical speculations, and implications of
the mythology of Purusa are extended to and developed in the mythol-
ogy of Prajapati. There can be no doubt that in the minds of the writers
of the later texts the two deities are the same; they are explicitly identi-
fied in several places,!! and the characteristics and functions of Prajapati
are obviously of a piece with those of Purusa as he appears in the famous
Rgvedic hymn.
In the later Samhitas and the Brahmanas, Prajapati is clearly the
overarching deity.'? Vedism traditionally counts thirty-three gods in its
10. For the cosmogonic myths of the RV in general, consult W. Norman Brown,
“Theories of Creation in the Rg Veda,” Journal of the American Oriental Society 85
(1965): 23-34; idem, “The Creation Myth of the Rg Veda,” Journal of the American
Oriental Society 62 (1942): 85-98; and F. B. J. Kuiper, Ancient Indian Cosmogony (New
Delhi: Vikas, 1983). For a sampling of the vast literature on RV 10.90, the primordial and
cosmogonic sacrificial dismemberment of Purusa, see Paul Mus, “Du Nouveau sur Rg
Veda 10.90?” in Ernest Bender, ed., Indological Studies in Honor of W. Norman Brown
(New Haven, Conn.: American Oriental Society, 1962), pp. 165-85; idem, “Ou finit
Purusa?” in Mélanges d’Indianisme a la mémoire de Louis Renou (Paris: E. de Boccard,
1968), pp. 539-63; and W. Norman Brown, “The Sources and Nature of Purusa in the
Purusa Sukta,” Journal of the American Oriental Society 51 (1931): 108-18.
11. That same Purusa became Prajapati.” SB 6.1.1.5. See also SB GAC 8 7.471715,
11.1.6.2; TB 2.2.5.3; and JB 2.47.
12. Recent works on Prajapati include Jan Gonda, “In the Beginning,” Annals of the
Bhandarkar Oriental Research Institute 63 (1982): 43-62; idem, “The Popular Prajapati,”
History of Religions 22 (November 1982): 129-49; idem, “Prajapati and Prayascitta,” Jour-
nal of the Royal Asiatic Society (1983): 32-54; idem, “The Creator and His Spirit (Manas and
cosmos; in these texts, Prajapati is said to be the thirty-fourth and the

summation of all the gods.'3 Like Purusa, who is “this all,” Prajapati is
also described as “all” or idam sarvam, the universal whole and total-
ity.'* He is the Cosmic One, the principle of unity and uniformity
(ekarupatva) who is, again like Purusa, all-encompassing.5
Prajapati is both space and time in their entirety, in their whole
forms. He is “these worlds,” and the worlds and the spatial directions
(the “quarters”) are, in turn, “this all.”!6 This deity’s completeness and
totality are also expressed by identifying him with the year,!’ for the year
is also said to be “all” or “this all”!8 and contains within it both the past
and the future!’—which is to say that the “year” is time itself. Or
Prajapati may be equated with the sum of the year’s parts: the six
seasons, the twelve months, the twenty-four new and full moon half-
months, the three hundred sixty days, or the seven hundred twenty days
Prajapati),” Wiener Zeitschrift fiir die Kunde Stidasiens 27 (1983): 5-42; idem, Prajapati’s
Rise to Higher Rank (Leiden: E. J. Brill, 1986); idem, Prajapati and the Year (Amster-
dam: North Holland, 1984); J. R. Joshi, “Prajapati in Vedic Mythology and Ritual,”
Annals of the Bhandarkar Oriental Research Institute 53 (1972): 101-25; R. T. Vyas,
“The Concept of Prajapati in Vedic Literature,” Bharatiya Vidya 38 (1978): 95-101;
Santi Banerjee, “Prajapati in the Brahmanas,” Vishveshvaranand Indological Journal 19
(June—December 1981): 14-19; and S. Bhattacharji, “Rise of Prajapati in the Brah-
manas,” Annals of the Bhandarkar Oriental Research Institute 64 (1983): 205-13. See also
the somewhat older work by A. W. Macdonald, “A propos de Prajapati,” Journal
asiatique 240 (1952): 323-38.
13. For Prajapati as the thirty-fourth god, see SB 4.5.7.2, 5.1.2.13; 5.3.4.23; PB
17.11.3, 22.7.5. For the deity as “all the gods,” see TB 3.3.7.3, 3.5.9.1; JB 1.342.
14. For Prajapati as “all” or “this all,” SB 1.3.5.10, 5.1.1.6, 5.1.3.11, 13.6.1.6; KB
6.15, 25.12; JUB 1.46.2. See also Jan Gonda, “Reflections on Sarva- in Vedic Texts,”
Indian Linguistics 16 (November 1955): 53-71; and idem, “All, Universe and Totality in
the Satapatha Brahmana,” Journal of the Oriental Institute (Baroda) 32, (September-
December 1982): 1-7.
15. “O Prajapati, no one but you encompasses all these creatures.” RV 10.121.10.
See also SB 10.4.2.2, 10.4.2.27, 10.4.2.31).
16. SB 6.3.1.11, 7.5.1.27, 13.6.1.3. See also SB 5.1.3.11.
17. E.g., SB 1.6.3.35, 3.2.2.4, 11.1.1.1; PB 16.4.12; JB 1.167. See also Gonda,
Prajapati and the Year.
18. SB 1.6.1.19, 2.6.3.1, 5.4.5.14, 6.6.4.3, 10.2.5.16, etc. See also the etymological
explanation for the year as “all” at SB 11.1.6.12: “Prajapati reflected, ‘All (sarva) I have
obtained by stealth, I who have emitted these deities.’ This became sarvatsara, for
sarvatsara is the same name as the year (samvatsara).” See also 11.1.6.13, where the
connection is explained in terms of the fact that Prajapati and samvatsara both have four
syllables.
OR Ves) NONE
and nights.” The frequent connection between Prajapati and the num-
ber seventeen?! is sometimes explained by adding the number of seasons
(here reckoned at five; see SB 3.9.4.11) and months: “There are twelve
months in a year and five seasons. This is the seventeen-fold Prajapati.
Truly Prajapati is all [sarva].””
And if Prajapati is both the spatial and temporal wholes, these two
may also be connected to each other in the logic of resemblance: the
year and “these worlds” are images of each other, for either may be
regarded as the “all.” Space and time are both seventeen-fold, and
thus “the year is indeed space” and “everything here, whatever is, is the
year.”*4 Prajapati may then also be regarded as twenty-one-fold, encom-
passing within him the twelve months, five seasons, three worlds, and
the sun.*
But just as Purusa was not restricted to his pantheistic immanence
(he is said to be one-quarter here, three-quarters beyond), so too is
Prajapati said to be yet more than the manifest. He is, as we have
seen, the thirty-fourth of the thirty-three gods; he is also said to be a
fourth “world” over and above the three worlds of heaven, middle
space, and earth.*° He is further imagined as “exorbitant” (atirikta),”’
“inexhaustible” (aydtayama),*8 and, by analogy, the seventeen syllables
of a certain ritual formula are said to be the inexhaustible portion of
the sacrifice.2? Elsewhere, the deity’s transcendence is indicated by
describing him as “unlimited” (aparimita)*° or the “unexpressed” or
20. Prajapati as the six seasons: $B 2.2.2.3, 5.2.1.3-4. Twelve months: SB 2.2.2.4,
4.6.1.11, 5.2.1.2, 5.4.5.20; JB 1.135. Twenty-four fortnights: SB DAP
deedBea Eh al WealWo oc in PA
see also KB 6.15. Three hundred sixty days and seven hundred twenty days and nights: SB
10.4.2.2.
21. E.g., SB 5.2.1.5; 5.3.4.22, 5.4.5.19; PB 2.10.5, 18.6.5, 19.7.6, 20.4.2
22. SB 1.3.5.10; see also SB 8.4.1.11.
23. SB 8.2.01719,6.4. 11,
24. SB 8.4.1.11, 12.8.2.36.
25. SB 6.2.2.3.
26. SB 4.6.1.4.
27. SB 11.1.2.5ff; KS 28.9; JB 2.192. See also Gonda, “The Redundant and the
Deficient in Vedic Ritual.”
205) Las 220.5:Be
29. SB 12.3.3.1. The formula is the series said by the adhvaryu priest at the time of
offering: “o Sravaya, astu Srausat, yaja, ye yajamahe, vausat.” The text claims, “This is the
seventeen-fold Prajapati.”
30. E.g., TS 1.7.3.2, 5.1.8.4; AitB 2.17; KB 11.7. See also SB 1.3.5.10, where the
“all” is also unlimited.
“undefined” (anirukta),*! and these two qualities are not essentially

different (“aparimitam vai anirutktam’”).*? Prajapati, then, like his el-
der alter ego Purusa, is both immanent and transcendent, the one
underlying the universe and extending beyond it, simultaneously finite
and infinite. He is both the gods and men, the divine and the human,
the mortal and the immortal;33 he is both the defined and the un-
defined (nirukta/anirukta), the limited and the unlimited (parimita/
aparimita)34
As his name suggests, the Lord of Creatures is chiefly charged with
the role of creation and as such is identified in the ritual texts not only
with Purusa but also with Visvakarman, the “All-Maker.”35 Prajapati,
alone and stricken with desire for offspring** or for company,3’ “emits”
(srj-) or emanates from himself the creatures.38 This cosmic emission
(srsti) is, or becomes, the manifest universe, the three worlds and four
quarters* and all things within them.*
Prajapati had a desire. “May I be propagated; may I be multiplied.” He »
heated up ascetic heat, and when he had done so, emitted these worlds—
earth, middle space, and heaven. He warmed up these worlds, and when
he had done so the luminaries [i.e., gods] were born. . . . He warmed up
the luminaries, and when he had done so the three Vedas were born.*!
31. E.g., PB 7.8.3, 18.6.8; SB 6.4.1.6, 14.2.2.21, 14.3.2.15. See also SB 1.4.1.21,
2.2.1.3, “sarvam vai aniruktam.” Consult also Louis Renou and Lilian Silburn, “Nirukta
and Anirukta in Vedic,” in J. N. Agrawal and B. D. Shastri, eds., Sartipa-Bharati or the
Homage of Indology: The Dr. Lakshaman Sarup Memorial Volume (Hoshiarpur: Vish-
veshvaranand Institute Publications, 1954), pp. 68-79. For the related ritual phenomenon
of offering to Prajapati silently, without mantra (upamsu or tusnim), because he is anirukta
and “all,” see, e.g., SB 7.2.2.14 and 12.4.2.1.
32. SB 5.4.4.13. Prajapati, in his guise as the year, is also described as aksayya
(“imperishable”). See SB 1.6.1.19 and 2.6.3.1; see also 11.1.2.12 and 12.3.4.11.
33. SB 6.8.1.4, 4.5.7.2. See also SB 10.4.2.2.
34. SB 14.1.2.18.
35. See $B 7.4.2.5, 8.2.1.10, 8.2.3.13, 9.4.1.12, etc.
36. E.g., KB 6.1; TS 3.1.1.1
37. “May I be more, may I be reproduced.” PB 6.5.1, 7.5.1; JUB 1.46.1.
38. For Prajapati’s creation, see Lévi, La Doctrine du sacrifice, pp. 18-21; and
Gonda, “The Creator and His Spirit.” For an analysis of the verbal root srj- and the
relationship between universal creation and the private “projection” of human dreamers,
see Wendy Doniger O’Flaherty, Dreams, Illusion and Other Realities (Chicago: University
of Chicago Press, 1983), pp. 16-17.
39. SB 6.3.1.11.
40. “Whatever there is.” SB 6.1.2.11.
41. AitB 5.32.
The creator and his creation are in the Veda ultimately the same being,
made of the same stuff.42 The universe, in Vedic speculation, is (or
should be) theomorphic. Prajapati is the vital life force (the “breath,”
prana) of his creation‘ or its soul (atman): “That very Prajapati, having
distributed [vidhdya)| himself, reentered it with his atman.”*
But despite, or perhaps because of, his principal characteristic of
cosmic fecundity (he is, it is said, productivity or prajanana itself) ,*
Prajapati often appears in the mythology as rather inept at his job. His
cosmic emission initially produces a creation that is faulty in one way
or another. The primordial procreative emanation engenders not a
cosmos—an ordered and orderly whole—but a cosmic problem de-
picted in terms of the sickly nature or horrendous living conditions of
the creatures: “Prajapati emitted the creatures; these, being emitted,
languished” ;*° they are created without breath,’ they suffer hunger for
lack or food,*® they are threatened by demons,*? and they find them-
selves without a “firm foundation.”*°
Many of these myths of Prajapati’s failed cosmogonic efforts can be
divided into two types: the cosmic emission is either insufficiently differ-
42. See also Jan Gonda’s comments: “In the case of the Vedic Prajapati creation is a
process of emission and exteriorization of some being or object that formed part of, or was
hidden in the creator himself, yet does not become completely independent of him, be-
cause Prajapati, being the totality (sarvam), embraces his creatures. . . . The creator god
is ‘identical’ with, that is immanent, inherent in, his creation.” “Vedic Gods and the
Sacrifice,” Numen 30 (July 1983): 18.
43. SB 6.3.1.9; see also PB 20.4.2. At SB 2.5.1.5, it is said that Prajapati emits
creatures in the same way that Vayu, the wind, enters all the quarters. Elsewhere (KB
19.2), Vayu is proclaimed the “manifest form (rapa) of Prajapati.”
44. TA 1.23.8.
AS 2B SB 5.1.3.10, 5.1.3.12; JB 2.175. This connection between the deity and
productivity leads the ritualists also to draw analogies between Prajapati and other particu-
larly fruitful entities such as goats and sheep, who are “most manifestly” like Prajapati
because “they bear young three times a year and produce two or three [offspring per
year]” (or possibly “two [offspring] three times [per year],” dvau trin iti). SB 4.5.5.6,
5.2.1.24; see also SB 3.3.3.8; TS 1.2.7.1; MSS 2.1.4.11; BhSS 10.17.4; ApSS 10.25.12.
etc., where the she-goat is said to be of the same “class” or varna as this god. The virile bull
is likewise connected, for he is “Prajapati among the animals.” SB 5.2.5.17. The most
potent (viryavattama) of the gods is also linked to the most potent of the animals, the horse
(SB 13.1.2.5). Prajapati is thus often invoked in the ritual for the increase of the sacri-
ficer’s fertility (e.g., PB 20.4.2).
46. PB 7.10.15; see also GB 2.3.9.
47. JB 1.111.
48. TB 1.1.3.5; PB 8.8.4, 6.7.19.
49. TB 1.7.1.4.
50. PB 24.1.2.
entiated or intemperately scattered into a chaos of unconnected frag-

ments. In other words, these two kinds of myths of defective creation
explore in narrative form the two metaphysical excesses discussed
above. When the story of Prajapati’s emission has the emitted creatures
indistinct or overly similar, the principle of jdmi or excessive resem-
blance is recalled. When, on the other hand, the creatures are said to be
dispersed and overly distinct one from another, the equally dangerous
metaphysical excess of extreme differentiation, prthak, is represented.
In either case, Prajapati’s procreative act results not in a cosmos but in a
metaphysical mess.
The jami type of creation myth is more infrequently found, but it
does appear often enough to suggest a category of similar stories. In
these myths, the deity emits creatures who are formless by virtue of
their cohesiveness or who are chaotically indistinguishable, being too
much alike: “Prajapati emitted the creatures. These emitted ones were
closely clasped together [samaslisyan],” indistinct and thus without indi-
viduality or “name and form [ndmaripa].”>! Or, as another text has it,
the emitted creatures are mistakenly created equal, left in a post-
creation tangle marked by discord, rivalry, and cannibalism: “Prajapati
emitted the creatures. They were undifferentiated (avidhrta), at odds
with one another (asamjandana), and ate each other.”* A creation re-
sulting in excessive resemblance is here depicted as a nightmare of
egalitarian chaos, an unstructured and homogeneous product. The
acosmic primordial state of absolute oneness (Prajapati before his emis-
sion) is thus reduplicated when the god engenders an anticosmic cre-
ation of uniformity and formlessness.
More regularly encountered are those stories in which Prajapati’s
procreative act has as its consequence a universe whose parts are scat-
tered, dispersed, separated, and utterly distinct—a prthak creation of
excessive differentiation. The very act of cosmic emission is most often
articulated with verbs having the dispersive prefix vi- (visrj-, “to emit,
discharge”; vikrs-, “to scatter”), emphasizing the transition from unity
51. TB 2.2.7.1. See also TB 3.10.9.1: “Prajapati emitted the gods. They were born
bound together (samdita) with evil.”
52. PB 24.11.2. This passage makes an interesting contrast to the later Hindu concep-
tion of anarchy, matsya-nydya (the “law of the fishes”), whereby bigger fish devour the
smaller with no other principle of order than brute size and strength. This later enunciation
of disorder is one of unrestrained hierarchy, whereas the Vedic text envisions a chaos in
which hierarchical order is completely absent. For another creation story that depicts
Prajapati’s creatures as undifferentiated cannibals, see JB 1.117: “Prajapati emitted the
creatures, who were emitted hungry. Being hungry, they ate each other.”
to diversity. But here, too, the product of the primordial creation is

defective for being metaphysically excessive. Indeed, in myths of this
type, creation results in a state of affairs that is equally problematic for
the creator and for his creatures.
When emitted, the shattered particles of reality lie in confusion and
disarray. In many texts, the creatures are said to “run away” from
Prajapati after being emanated, fearful and hungry, or they wander off
disoriented.*® In other variants, the dispersed creatures display an undisci-
plined independence, resisting the “superiority” (Sraisthya) of Prajapati,
their creator and principle of order and unification.» Still elsewhere, we
read that creatures flee from Prajapati out of fear that he will consume
them: “Prajapati emitted the creatures. These emitted ones went away
from him, fearing he would devour them.”*> Or, again, when Prajapati
had completed the creative work and was “emptied,” “The creatures
turned away from him. They did not remain with him as his fortune and
food.”°* In these passages, the terror the creatures have of being con-
sumed might be a trope for the reluctance of overly distinct elements to
allow themselves to be resumed within a unified whole.
One of Prajapati’s postcreative problems—that is, the metaphysical
problem of a universe that lacks a unifying principle—is nicely summa-
rized in the Satapatha Brahmana (10.4.2.3): “Having emitted all the
beings, he thought to himself, ‘How can I put these creatures back into
my self (atman)? How can I place them back into my self? How can I
again be the soul (dtman) of all these beings?’ ” All these passages
highlight the unstructured chaos of a creation in which the parts are
excessively differentiated and thus without the order that comes from
interconnection and unification.
Another of Prajapati’s quandaries is that he himself becomes totally
spent and broken down after his procreative emission. Texts depicts him
variously as “milked out” (dugdha), “drained” (riricana), “exhausted”
(vrtta), “diseased” (vydajvara), or “disjointed, fallen to pieces” (visrasta)5”
53. E.g., TS 2.1.2.1, 2.4.4.1; TB 1.1.5.4; PB 6.7.19; AitB 3.36; GB 2.5.9. See also
TB 21.2.1; 5529.9,
54. PB 6.3.9, 16.4.1—-3. See also PB 17.10.2.
59.3PB 21-221,
56. SB 3.9.1.1.
57. Dugdha: PB 9.6.7. Riricana: TS 1.7.3.2, 5.1.8.3, 6.6.5.1, 6.6.11.1; MS 1.6.12;
TB 1.1.10.1; SB 10.4.2.2; JB 2.149, 2.181, 3.282.; Vrtta: TB 1.2.6.1. Vyajvara: GB 2.4.12.
For visrasta and other forms of visrams- to describe Prajapati’s postcreative condition, see
TB 2.3.6.1; SB 1.6.3.35; 4.5.4.1, 6.1.2.12; Aita 3.2.6; SanA 8.11. See also $B 1.6.3.16,
where, in a version of the very ancient Vedic creation myth of Indra’s slaying of the
In this sad state, the creator fears death; his life force (prdna), essence
(rasa), and luminous power (tejas) leave him,°? as do the gods, the cattle,
and all his progeny.® The dispersive quality of the faulty creation is
mirrored in the dissipation of the creator.
The disarray and disconnection of Prajapati’s creatures and the
dissolution of the progenitor’s body are two sides of the same metaphysi-
cal coin. Prajapati, the aboriginal One and the potential unifying power,
has gone to seed, so to speak. He is broken down, emptied out, ineffec-
tual, lying pathetically in a heap. “And truly there was then no firm
foundation (pratistha) here at all.”®! The constituents of the universe,
Prajapati’s emitted parts, similarly are scattered in confusion and diver-
gence; they, too, in such a condition, “found no firm foundation.”
Moreover, time, like space, is also defectively engendered and in
much the same way. The disintegration of Prajapati’s cosmic body or
self caused by his procreative emission is also envisioned as a worrisome
discontinuity in the year, the temporal whole. When Prajapati is dis-
jointed, time is literally out of joint:
When Prajapati had emitted the creatures, his joints (parvans) became
disjointed. Now Prajapati is the year, and his joints are the two junctures of
day and night, of the waxing and waning lunar half-months, and of the
beginnings of the seasons. He was unable to rise with his joints disjointed.
Or, again, “That Prajapati who became disjointed is the year, and these
joints of his that were disjointed are the days and the nights.”
As the undifferentiated One and his undifferentiated creation are
seen as one form of anticosmos in one set of myths (of the jami variety),
so is the dissipation of the One and the unconnected diversity of his
emanation viewed as another kind of chaos in another cluster of
cosmogonies. In both cases, creation is not cosmos. The natural act of
serpent who has the universe pent up within itself, the latter “having been hit [by Indra] lay
contracted (samvlina) like a leather bag which has been emptied.” For some scattered
comparative speculations about the meaning of Prajapati’s postcreative fatigue, see
Gonda, Prajapati’s Rise to Higher Rank, pp. 52-54.
58. SB 10.4.2.2.
59 SB 6.1.92, 7.4.2.1, 74.2.4, 6.4.2.4, 10. 1.11, 13,1.1.4.
60. SB 9.1.1.6, 8.2.3.9; PB 6.7.19; JB 3.153, 3.230; TS 2.4.4.2. For the sacrifice
leaving Prajapati, see, e.g., TS 6.1.2.4; JB 3.155, 3.274.
oi: SB7:12.2.
62. PB 24.1.2.
63. SB 1.6.3.35.
64. SB 10.1.1.2.
procreation (albeit done by a supernatural entity) results in ontological

disaster, a nonbeing parallel to the nonbeing of precreation. In these
stories, the moral is that all life, all being, is not produced preformed—it
is, as we shall soon see, to be composed and constructed subsequent to
the merely procreative act. The creation of true cosmos, otherwise put,
is not simply a matter of the One becoming many but also and necessar-
ily of the reintegration of the many into a new whole. Cosmogony, the
production of an ordered universe out of a generated potential, is a
secondary act in the Prajapati myths as form is carved out of or con-
structed from the formless emission.
Let us take another angle on the same subject by returning for a
moment to the concept of sarva (“all”) in ancient Indian philosophy. Jan
Gonda has argued that many Vedic texts make a subtle but important
distinction between this word and another which is often regarded by
Indologists as synonymous. Sarva, according to Gonda, connotes an
undivided, perfectly complete, and seamless whole, as, for example, the
aboriginal Purusa or Prajapati. Visva, often also translated as “all,”
carries a meaning critically distinct from sarva, Gonda argues, and one
that might well apply to the postcreative but precosmic “all”:
The combinations of sarva- and visva- are therefore not necessarily tauto-
logical in character . . . vifva- pointing out the inability to proceed after a
certain total number has been counted, sarva- emphasizing the idea of
wholeness and completeness and the inability to discern defectiveness.®
The creation myths we have considered, then, may also be understood

as the tracing of the transition from sarva to visva, a transformation from
a perfect unity without parts to a defective totality. Another step in the
metamorphosis is required for true cosmos: the reintegration of the
totality (visva) into a constructed whole, a composed unity of parts the
texts call samdna. As Verpoorten has demonstrated, the move toward
cosmogony and ontology in the Vedic conception is one not of return to
primordial unity but rather of rebuilding a complete structure without
eliminating diversity: “Behind the word samana always lies a duality or a
multiplicity. The unity signified by it, far from excluding multiplicity,
supposes it, supports itself on it.”
Following the abortive attempt at cosmogony by natural means is
a mopping-up operation which fashions a cosmos out of the chaos that
is Prajapati’s emission. In the cosmogonic myths of the Vedas, this
65. Gonda, “Reflections on Sarva- in Vedic Texts,” p. 54.

66. Verpoorten, “Unité et distinction,” p. 81.
cosmos-building labor is signified by a shift in verbal forms. The procre-

ative dispersal, marked by words prefixed with vi-, is later counter-
acted by action of cosmic healing, repair, and construction described
with verbs beginning with sam-, conveying conjunction, concentration,
and assembly (samkr-, samtan-, samklp-, sampad-, samyuj-, etc.).97
Further, this constructive activity combating natural imperfection is
usually said to be ritual activity. The sacrifice is a cosmogonic instru-
ment, for the ritual process completes all the stages necessary for mak-
ing an ontologically viable universe—sarva, vifva, samdna. Ananda K.
Coomaraswamy summarizes the sacrificial structure and its connection
to cosmogony:
And what is essential in the Sacrifice? In the first place, to divide, and in
the second to reunite. He being One, becomes or is made into Many, and
being Many becomes again or is put together again as One.®
Let us now return to the creation narratives. In both of the two

types we have encountered—the jami and prthak varieties—we may
observe the same movement from emitted formlessness to ritually cre-
ated structure.
When Prajapati’s emission is depicted as insufficiently differenti-
ated, true cosmogony occurs when the created mass is divided into its
constituents. Excessive resemblance is thereby moderated by the imposi-
tion of difference and form (ripa). Creatures emitted “closely clasped
together,” without individuality, were made distinct when Prajapati “en-
tered them with form. That is why they say, ‘Prajapati is indeed form
itself.’ ”® In this instance, the progenitor is also the fashioner of distinc-
tive shapes. Elsewhere, however, Prajapati finds a tool, the sacrificial
ritual, which he then utilizes to create the essential distinctions among
indistinguishable creatures. When these latter were emitted “undifferen-
tiated, at odds with each other,” and engaged in mutual cannibalism,
This distressed Prajapati. He saw [i.e., discovered] the forty-nine-day sacri-
ficial session. Thereupon this [creation] became separated (vydavartata).
Cows became cows, horses [became] horses, men [became] men, and wild
animals [became] wild animals.”
67. For this point, see also Silburn, Instant et cause, p. 56.
68. Ananda K. Coomaraswamy, “Atmayajfia: Self-Sacrifice,” Harvard Journal of
Asiatic Studies 6 (1941): 396.
69. TB 2.2.7.1; see also TB 3.10.9.1, where Prajapati “divides” (or “releases,”
vyadyat) the gods whom he has created “bound together with evil.”
TOMER 24-1122:
Here it is pointedly the structuring effect of ritual that redeems Prajapati’s

creative error. In the jami sort of creation myth, the cosmic One repro-
duces himself in an undifferentiated totality which must be decomposed
before it can take shape as a cosmos, a constructed whole made up of
distinct but related parts.
When Prajapati’s procreative act results in an excessively differenti-
ated creation, the condition of prthak is also offset by the connective
power of the sacrifice. In this case, cosmos arises out of the ritual reinte-
gration of the splintered emitted parts. Creatures too independent and
unrelated, resisting Prajapati’s “superiority,” are brought back to the
fold by ritual means: Prajapati “saw that agnistoma [soma sacrifice] and
performed it. Thereupon the creatures yielded to him superiority.””!
Creatures who run away from their creator are reassimilated and ori-
ented with the help of a rite:
Prajapati emitted the creatures. These emitted ones ran away from him.
He saw this “undefined” (anirukta) morning soma pressing. With that, he
went into the middle of them. They turned toward him and circled him.”
Or, again, creatures who flee from Prajapati for fear of being consumed
by (or resumed in) him must be returned to the creator and integrator
and “eaten” in a manner that preserves distinction while interrelating
and subsuming the elements within an ordered whole:
He said, “Return to me, and I will devour you in such a way that, although
devoured, you will multiply.” He consumed them by means of [a certain]
ritual chant and caused them to multiply by means of [another] ritual
chant.”
The creation of cosmos requires both the fixing of Prajapati’s emis-
sion (the srsti, translated by Paul Mus as “la création brute”) and the
repair of Prajapati himself (atisrsti, “la surcréation” in Mus’s terms).74
71. PB 6.3.9. See also, SB 3.9.1.4; “He saw that set of eleven victims [for the soma
sacrifice]. Having sacrificed them, Prajapati fattened himself again, and the creatures
returned to him, and remained with him as his fortune and food. Having sacrificed in this
way, he became superior (vasiydn).” For other instances of using ritual to resolve a
disputed superiority in Vedic mythology, see TS 2.2.11.5—6 and PB 7.2.1.
Tee PB TIA02-
73. PB 21.2.1. For Prajapati as “eater,” see also PB 17.10.2 and TB 3.8.7.1,
3.9.10.1. Other comparable passages in which ritual rectifies Prajapati’s defective creation
include PB 7.10.15; 8.8.14; KS 29.9; GB 2.3.9.
74. “La création brute (srsti) et la surcreation (atisrsti), cette derniére décrite dans
les brahmana comme la reconstruction du corps de Prajapati-Purusa dispersé et a qui le
rite ramene symbolique ses membres (illatif), sont deux temps successifs et complé-
mentaires, mais en série. Ils n’adviennent pas, l’un et l’autre, qu’une seule fois.” Mus, “Du
Nouveau sur Rg Veda 10.90?” p. 182.
When Prajapati is in a postcreative state of fragmentation and disintegra-

tion, he is made whole again and healed by the gods. They restore him
when he has fallen down and put back into him the vital breath which
had left him.”?> Another passage relates that when Prajapati had fallen
into pieces, his breath and energy gone out from him, “truly there was
then no firm foundation here at all. The gods said, ‘There is no other
firm foundation apart from this one. Let us restore father Prajapati; he
will be a firm foundation for us.”” In yet another text, the revivification
of Prajapati is depicted as a cosmic healing, performed by collecting
together the luminous energy (tejas) and life essence (rasa) of all the
emitted creatures, thereby simultaneously reuniting the dispersed cre-
ation (the srsti) into a reconstituted unity and reinvigorating Prajapati
(the atisrsti) who is that unity.”
Several variants make explicit that this restoration or healing activ-
ity is essentially a ritual activity. The sacrificer in his ritual makes
Prajapati “whole and complete and raises him to stand upright, even as
the gods once raised him.”’8 Prajapati, disjointed after procreation, begs
the god of sacrificial fire, Agni, “Put me back together (tvam ma
samdhehi).” Agni does so by building him up piece by piece, just as the
altar of the agnicayana sacrifice is constructed brick by brick by the
sacrificer and his officiants:
It was five body parts of his that fell into pieces—hair, skin, flesh, bone,
and marrow—and these are the five layers [of the altar]. When he builds
up the five layers, it is also with those five body parts that he builds up
[Prajapati].”
And when Prajapati, conceived of as the year, is “out of joint” and
“unable to rise” because of the dissipating effects of the procreative act,
“the gods healed him by means the havis offerings,” the regularly per-
formed and obligatory sacrifices of the Vedic Srauta sacrificial repertoire:
With the agnihotra [the twice-daily sacrifice] they healed that joint [which
is] the two junctures of day and night, and joined it together. With the new
and full moon sacrifices, they healed that joint [which is] between thee
waxing and waning lunar half-months, and joined it together. And with
75. SB 7.4.2.11, 7.4.2.13.

76. SB 7.1.2.2. See also PB 24.1.2 for the story at the level of the srsti: “Prajapati
emitted the creatures. These did not have a firm foundation. With these [days of the thirty-
three-day sacrificial session] they had a firm foundation. These worlds did not have a firm
foundation. With these [days] they had a firm foundation.”
Teh .2.6-1,
78. $B 7.1.2.11.
79. SB 6.1.2.17.
the cdturmasyas [quarterly sacrifices] they healed that joint [which is] the
beginning of the seasons, and joined it together.®°
Time, space, and all within them are in Vedic cosmogonies an emana-
tion of Prajapati and participate in his being. However, in the words of
Silburn, for the ancient ritualists “neither things nor inert data exist, but
only the functions and activities of synthesis.”*! Prajapati is reconstructed
in a secondary cosmogonic act of ritual construction which also shapes
into form the discontinuous creatures of the cosmic emission. Unlike all
the king’s horses and all the king’s men, the gods and men, deploying the
formative and connective power of ritual, can put the shattered god and
his creation back together again—an operation of ritually produced reinte-
gration which Mus has cleverly called “information.”
The mythology of Prajapati also accounts for the origins of the
cosmos-making tool that is the Vedic ritual. Although sacrifice is utilized
to fix Prajapati’s abortive attempt at cosmogony, it is the creator god
himself (who else?) who first produced the ritual that will heal him and
his creatures. Prajapati is both the creator* and first practitioner of the
ritual, subsequently turning over the sacrifices to other deities.** The
creator generates the individual pieces of the universal puzzle and leaves
them in a chaotic jumble, but it is also he who produces the means for
interlocking those cosmic fragments.
In some texts, the structure of the creation of the sacrifice replicates
that of the creation of the universe as a whole. One encounters the same
formula: an emission of an unorganized totality and a subsequent reinte-
gration as the particular sacrifices are “measured out” (udma-), “as-
signed” (vyddis-), or “apportioned” (vibhaj-) to the appropriate deities,
thus forming a systematically organized whole.* In other instances, how-
ever, this same distribution of the sacrifices at creation is regarded as a
dangerous dispersal; it leaves Prajapati, once again, drained or emptied
until healed by yet another step in which the creator takes back into
himself, by ritual means, those particular sacrifices he had disseminated
among the gods.®
Indeed, the creation of the sacrifice is sometimes said to be as
80. SB 1.6.3.36; see also SB 1.3.5.16; 6.1.2.18, 8.7.1.1-3, 10.1.1.2-3, 10.4.3.20:

11.2.7.1—5, etc.
81. Silburn, /nstant et cause, p. 1.
82. Or “mouth,” e.g., TS 5.1.8.3; KB 13.1.
83. SB 2.3.1.22, 6.3.1.18; AitB 5.32.
84. TS 1.6.9.1; TB 1.3.2.5; AitB 3.13; SB 13.1.1.4.
85. TS 1.7.3.2, 6.6.11.1.
defective a production as the procreation of the creatures. When

Prajapati emits the sacrifice, his “greatness” (mahi) departs from him,
and he must go searching for it;8* or the sacrifice runs away with the
demons, leaving the gods in a big jam.’’ Here, too, simple creation is
insufficient and is to be followed up by an organizational effort that
rectifies chaos. The sacrifice must be “restrained” by and returned to
those to whom it properly belongs—who are, in the following story as
elsewhere, the Brahmins:
Prajapati emitted the sacrifice, and after the sacrifice the brahman power
and the ksatra power were emitted. After them were emitted those crea-
tures who eat sacrificial oblations and those who don’t. The Brahmins are
those creatures who eat sacrificial oblations, the Rajanyas [Ksatriyas],
Vaisyas, and Sidras those who don’t. The sacrifice departed from them.
The brahman power and ksatra power followed after it, each with their
own weapons. The weapons of the brahman are the horse chariot, armor,
and bow and arrow. The sacrifice escaped, recoiling, from the ksatra’s
weapons, and the ksatra did not catch it. The brahman followed it, caught
it, and restrained it, standing from above. Caught, restrained from above,
and recognizing its own weapons, [the sacrifice] returned to the brahman.
Therefore, even now the sacrifice finds support in the brahman and in the
Brahmins.*§
The order-producing sacrifice, when it is itself appropriate tamed
and in the hands of the priests (and not others), is not only regarded as
an instrument of cosmic healing and construction; it also is sometimes
said to have a procreative power supplementing that of the creator god.
In some myths, Prajapati emits the sacrifice first, and out of its parts are
further generated the components of the universe:
Prajapati emitted the sacrifice. With the agnyadheya he emitted seed
(retas); with the agnihotra, gods, men, and demons. With the new and full
moon sacrifice, Indra was emitted. He has emitted food and drink for
them from the sacrifices of vegetable oblations and the soma sacrifices.®
The fertility of the sacrifice at the cosmic level and its origin in the
being of Prajapati led to the frequent connection between the creator
god and the creative ritual: “That sacrifice which is now being per-
formed is Prajapati, from whom these creatures were born; and even so
86. SB 13.1.1.4.
87. TS 3.3.7.1; see also TS 6.1.2.4; JB 3.155, 3.274.
88. AitB 7.19.
89. KB 6.15; see also SB 2.5.1.17, 2.5.2.1, 2.5.2.7, 2.6.3.4; TS 3.5.7.3.
are they born today.” The repeated statement in the Brahmanas that
“the sacrifice is Prajapati”®! points to the productive power of the ritual
(“When he [the sacrificer] offers, he emits forth this all”)? and indicates
that Prajapati and the ritual share the same essence. The two are coun-
terforms of each other: “this all,” which is Prajapati, corresponds to (or
results from, anu) the sacrifice.°? Prajapati emits from himself or, as one
text says, “makes his dtman” the sacrifice,* and, as we have seen, the
sacrificial operation in turn recreates a unified Prajapati. Like Escher’s
“Drawing Hands,” Prajapati and the sacrifice bring each other into
existence.
And if Prajapati and the sacrifice are resembling forms, so too are
their individual components essentially related. Either the individual
sacrifices that make up the ritual repertoire are equated to Prajapati’s
corporal components (e.g., SB 12.1.4.1-3), or the parts of the “body,”
the rites, of one sacrifice are connected to correlative parts of Prajapati’s
body or self.°° Because of these connections based on resemblance, by
putting together a sacrificial performance, rite by rite, Prajapati (as well
as his creation) is also put together.
Vedic cosmogonies, like cosmogonies everywhere, depict the ori-
gins of the present. Stories of the primeval emanation of the One into
the many and of the secondary ritual act of reconstituting the creation
into a constructed whole provide Vedic ritualism with its metaphysical
problem (the defectiveness of the merely natural) and its ritual solution
(sacrificial reclamation). Life in the historical present, no less than in the
mythic time of origins, is regarded as intrinsically faulty without the
formative structure only ritual can provide.
And also not unlike other myths from other religions, these myths
seem to explore two remarkable—in that the study of religion too often
leaves them mute—and related doctrines. First, they radically devalue
the abilities of the creator god; he is portrayed as inept, the progenitor
not of perfection but of a defective product. Like other myths both
90. SB 4.2.4.16, 4.5.5.1, 4.5.6.1, 5.1.4.1, ete.

91. Or “Prajapati is the sacrifice.” See, e.g., SB 3.2.2.4, 5.2.1.2, 5.2.1.4, 5.4.5.21,
6.4.1.6, 14.1.2.18; TB 3.2.3.1, 3.7.2.1. In other instances, Prajapati is identified with the
year, and the year with the sacrifice (e.g., SB 5.2.1.2, 5.4.5.21, 11.1.1.1; see also $B
i Ba6 8
92. SB 3.6.3.2.
93. SB 3.6.3.1.
94. PB 7.2.1. See also SB 4.2.5.3, where Prajapati is also said to be the atman of the
sacrifice.
95. See KB 6.15; SB 10.3.1.1; AitB 2.18; PB 13.11.18; GB 2.1.26.
within this tradition and in other religions, the god is represented as

generating a world and its inhabitants which are disorganized, deficient,
and degenerate (elsewhere they are termed evil in stories analyzing the
problem of theodicy). Secondly, these myths legitimize the audacious
assumption of enormous power on the part of certain human beings—
here, the Brahmins, who are hereditary experts in the Vedic ritual and
whose stories depict themselves as charged with putting together coher-
ently what Prajapati has so badly messed up. Human beings of a certain
sort and with certain self-proclaimed privileges, together with the lesser
gods, must repair the damage done by the creator god’s action. It is no
wonder that, with these assumed responsibilities and supposed powers,
the Brahmins would (as we will see in the next chapter) call themselves
human gods.
True cosmos, true reality, and true being are fabricated, continually
reconstructed by sacrificial activity. This metaphysical power attributed
to ritual (and to those who can manipulate it) depends on the principle
of universal resemblance, to the intricacies of which we now turn.
The Metaphysics of Resemblance and Connection
Prajapati and Purusa are the anthropomorphically conceived expres-

sions of the creative source and principle of universal unification in
Vedic cosmology. The sacrifice, as the resembling dynamic counterform
of the Cosmic One, represents Prajapati/Purusa’s essence. It, too, is
generative and regenerative of being, encompassing within it the univer-
sal whole, and is thus regarded as another way of speaking about the
unifying force in the cosmos. When constructed properly—when rite is
made to follow rite in an unbroken chain, a unity of parts—the form of
sacrifice produces a formed, orderly cosmos in which all the parts are
interrelated.
In other sections of Vedic ritual and philosophical literature, this
unifying, constructive power is depicted as a neuter force, not unlike
mana, wakan, or the Holy Spirit in other religions.% Like Prajapati/
Purusa the brahman is said to create and pervade “this all.” It, too, is
regarded as sarva,” and it is this feature that Gonda emphasizes when
he writes, “Brahman is what is the whole, complete here, is what is
96. See Marcel Mauss, A General Theory of Magic, trans. by Robert Brain (New
York: W. W. Norton, 1972).
97. E.g., ChU 3.14.1.
entire, perfect, with no parts lacking, what is safe and well etc., i.e.
Completeness, Totality, the All seen as the Whole.” As such, the
brahman is not different from Prajapati himself? and has the same
creative and ontologically formative powers. In the Satapatha Brah-
mana (11.2.3.1-5), for example, the brahman “was in the beginning this
[all].” It created the gods and organized them into their proper spheres:
Then the brahman itself went to the transcendent side. Having gone there,
it wondered, “How can I again return to these [immanent] worlds?” It
returned again by means of name and form. . . . These are the two great
manifestations of the brahman.
The neuter force of brahman is here presented as the creator, formula-

tor, and integrative power made manifest in every one of the parts of the
universe, in every immanent “name and form.” The relationship be-
tween Prajapati and his creatures is simply represented in a different
idiom; for, as Maryla Falk has written,
namas and rupas [names and forms] are the negative, mortal, differenti-
ated condition of the one ndma, all-consciousness, consubstantial with the
one rapa, the Universe as the latter’s “own form.” While the nama is the
inner power of the individual being or thing, the ruépa is its sensuous
appearance.!
As in one conception of the universe the elements of the cosmos are
considered parts of a Cosmic Man, so in another but congruent notion
the brahman creates and enters multiple reproductions of itself, thus
providing their common origin and accounting for their mutually resem-
bling forms. This point is made also in the Satapatha Brahmana
(11.2.3.6), in which the gods gain immortality through tapping the brah-
man, and the sacrificer gains the “all” and an “imperishability” through
the ritual manipulation of name and form (brahman’s manifestation):
In the beginning the gods were subject to death. Only when they filled
themselves with the brahman did they become immortal. He who pours
out the libation to the mind thereby obtains form—form is mind, because
98. Gonda, “Reflections on Sarva-in Vedic Texts,” p. 64.

99. “Prajapati is the whole (sarva) brahman.” SB 7.3.1.42. See also SB 13.6.2.8:
“The brahman is Prajapati, for Prajapati has the nature of the brahman (brahma).”
100. Maryla Falk, Nama-Rupa and Dharma-Rapa (Calcutta: University of Calcutta,
1943), p. 19. See also Ananda K. Coomaraswamy’s definitions: “forms, ideas, similitudes,
or eternal reasons of things (ama, ‘name’ or ‘noumena’ = forma) and the things them-
selves in their accidental and contingent aspects (rupa, ‘phenomenon’ = figura).” “Vedic
Exemplarism,” Harvard Journal of Asiatic Studies 1 (1936): 44.
it is with the mind that one knows, “This is form.” And when he pours out
the libation to speech he thereby obtains name—name is speech, because
it is by speech that he grasps the name. As much as is name and form, that
much is this all. He obtains all, and the all is the imperishable.
Brahman is, however, something of a multivalent concept within
Vedism. Some shifts of the semantic range of the word can be identi-
fied as one passes from the Rg Veda (where the meaning tends to
center on the transcendental potency of the hymn or formula), through
the ritual texts (where it describes the power of sacrificial activity), to
the Upanishadic literature (where it is more generally conceived as the
ultimate principle of all being). Such a historical evolution should not
be overly stressed, as Jan Gonda, Louis Renou, and others have shown
in studies of this fundamental Vedic term.!°! The multivalence of brah-
man is explicable precisely because of its centrality; many different
objects, acts, and phenomena can be “designated by the same name,
because they all participate in or partake of that important and central
concept.” 10 }
Brahman is the basis or ground of a universe or mutually resem-
bling things and beings, a foundation in which the perpetual interplay of
resemblances find their source, condition of possibility, support, and
end. Brahman is to be understood in light of the Vedic preoccupation
with continuity and stability in the face of assumed natural discontinuity
and the instability of creation. Gonda rightly insists on the affinity this
central metaphysical term has with other important Vedic concepts such
as ayatana (“base,” “support,” “resort”),!° pratistha (“firm foundation”
in its spatial sense),!% and samstha (“end” or “temporal foundation”):
It would, therefore, be easy to understand if the ancient Indian searchers
for a firm ground or foundation for the universe, the human soul included,
had chosen a word derived from the root brh-, “to be firm, strong, etc.” to
designate that ultimate foundation of all that exists. Anyhow, it is a fact
that the concept of a support, that is a fundamental principle on which
everything rests, and the idea of firmness and immovability are often
expressed in connection with brahman or with God who is brahman. !°
101. Jan Gonda, Notes on Brahman (Utrecht: J. L. Beyers, 1950); Louis Renou,
“Sur la notion de brahman,” Journal asiatique 237 (1949): 7-46.
102. Gonda, Notes on Brahman, p. 14.
103. Jan Gonda, “Ayatana,” Adyar Library Bulletin 23 (1969): 1-79.
104. Jan Gonda, “Pratistha,” Samjnavyakaranam, Studia Indologica Internationalia
1 (1954): 1-37.
105. Gonda, Notes on Brahman, p. 47.
qe REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
What is missing from Gonda’s definition, however, is the active and

dynamic quality of brahman. This force is not merely a passive, ever-
present metaphysical foundation but is also a potency within all “names
and forms,” which, when activated, is their possibility for self-realization.
Indologists of the last century liked to compare brahman to electricity, a
latent energy that, when switched on by ritual means, produces cosmic
order (rta) and individual power. Louis Renou has more recently demon-
strated that it is the connective potency of the brahman that is at the heart
of the Vedic concept and which links together all the meanings of the
word.
Renou argues that the common semantic denominator is brahman
as “a connective energy condensed into enigmas.”!% The enigma or
mystery of the brahman force “is the equation between human behavior
and natural phenomena, the connection between rite and cosmos.”!”
The metaphysical, epistemological, and ritual are thus conjoined in a
concept that is both the ground of the universe and the possibility for its
establishment. “Brahman,” concludes Renou, “is nothing else but that
form of thinking in enigmas consisting of the positing of correlation, an
explicative identification: that very thing the Brahmanas designate by
the terms nidana or bandhu, and eventually upanisad.”'
Brahman is therefore the source and foundation of all that exists—
the nexus of all cosmic connections—and the connective force itself lying
behind all knowledge and action that constructs ontologically viable
forms. It is the fount and terminus of all potential and realized correspon-
dences, the condition of possibility for wholeness as well as the whole
itself. Brahman, in sum, is the connective energy that lies between appar-
ently (and naturally) disparate elements and makes efficacious the ritual
action that forges those elements into unity. Brahman is both the neuter
counterpart to the Cosmic Man and the analogue to the sacrificial opera-
tion: “What is related to ritual activity is related to the brahman.”
Whereas brahman is the ultimate nexus of and condition of possibil-
ity for all connections between the resembling parts of the universal
whole, the particular resemblances are often described in terms of proto-
types and counterparts, forms and images, or replicas. It is the intricate
texture of these individual relationships that make up, but are not the
sum of, brahman considered as the cosmic whole.
106. Renou, “Sur la notion de brahman,” p. 43.

107. Ibid., p. 12.
108. Ibid., p. 13.
109. BGS 3.13.9.
Vedic connections are of two sorts: what we might call vertical and
horizontal correspondences. The former connects an immanent form
and its transcendent correlative: “Deux plans, mais un étre,” as Mus
puts it.° This type of connection operates between the elements of the
same species located on different and hierarchically ranked cosmological
levels. Horizontal connections link resembling components of two differ-
ent species located within the same cosmological plane which share a
similar hierarchical position within their respective classes. Here I plan
to concentrate (but not exclusively) on the vertical connections within
Vedic metaphysics.
One statement of the epistemological program of Vedism (which I
have used as one of the epigrams to this book) is found in the Rg Veda
(10.130.3); “What was the prototype (prama), what was the counterpart
(pratima), and what was the connection (nidana) between them?”
Coomaraswamy has offered “exemplar” and “image” as translations for
prama and pratima, respectively, “which imply in strictness ‘model’ and
‘copy’. . . . The exemplary image, form, or idea is then a likeness in the
sense of imitable prototype.”!! On the grandest level, the entire uni-
verse is the pratima—liiterally the “countermeasure” or more freely the
image, counterpart, or projection—or Prajapati or Purusa, from whom
all was emitted. But it is especially those phenomena that are considered
“uniform” (ekaruipa) that are most manifestly representations of Praja-
pati’s nature. The year, the temporal unity, is singled out at SB 11.1.6.13
as the pratima of Prajapati’s very self (atman):
Prajapati reflected, “I have emitted the counterpart of my self, the year.”

And therefore they say, “Prajapati is the year,” for he emitted it as a
counterpart of himself.
In a passage that follows shortly thereafter (SB 11.1.8.3), the text de-
scribes in exactly the same terms another counterpart of Prajapati also
regarded as “all,” as a unified whole—the sacrifice. Elsewhere, Praja-
pati is said to have “given himself” to the gods in the form of the
110. Paul Mus, Esquisse d’une histoire du Bouddhism fondée sur la critique
archéologique destextes (Paris and Hanoi: Paul Geuthner, 1935) p. 121.
111. Coomaraswamy, “Vedic Exemplarism,” p. 50. For other discussions of prama/
pratima, consult S. Schayer, “Die Struktur de magischen Weltanschauung nach dem
Atharva-Veda und den Brahmana-Texten,” Zeitschrift fiir Wissenschaft Buddhismus 6
(1925): esp. 275-76; and Hermann Oldenberg, Vorwissenschaftliche Wissenschaft: Die
Weltanschauung der Brahmana Texte (Gottingen: Vandehoeck and Ruprecht, 1919), esp.
pp. 114-15.
sacrifice;!!2 Prajapati is equated with the sacrifice, and the gods to whom
he offers as well as the oblations with which he offers are declared to be
forms (rupas) of him.!!3
Sacrifice also, then, is a pratima of the Cosmic One in that the
composition of parts into wholes achieved within the ritual resembles
the body of Prajapati, a similarly constructed whole and the prototype
for all others. Because both the year and the sacrifice are counterparts of
the same prototype, they are themselves linked: the sacrifice has “the
same measure” (sammita) as the year.!!4
Some of the implications of regarding the sacrifice as a counterpart
of the cosmic whole are obvious. The construction of a sacrifice, an
ideally continuous and complete entity made out of the joining of dis-
crete parts (rites, performers, implements, offerings, etc.), is a recon-
struction of the universe itself in the sense that the one supposedly
reproduces—in a different form—the other. They are not identical but
resembling forms of unity, sharing the same essence but manifesting
themselves differently. The sacrifice is composed of the counterparts to
the cosmic prototypes (each element of the ritual being vertically con-
nected to transcendent correlatives), and the sacrifice as a whole is the
counterpart to the prototype that is Prajapati, the universe. The sacrifice
operates with “images,” whereas Prajapati’s body or self is comprised of
the “originals,” but both participate in the same ontological essence.
Other implications also follow from the prototype/counterpart
presupposition—these on the level of horizontal connections. The sacri-
ficer and his oblation are said to be interrelated in this way, such that
the sacrificer offers a form of himself when he offers clarified butter,
the sacrificial cake, or an animal; for, “by virtue of the counterpart
(pratimaya) it is the man.”!!5 There is no confusion here between the
two, any more than there is a confusion between the transcendent god
and the sacrificial ritual performed here by humans. The offering is not
identical to the sacrificer but is his projected representative; and the
sacrifice as a whole is, as Paul Mus puts it, “une counterpartie et un
substitute personnel pour homme qui l’offre.”"° “The man is the
sacrifice,” says the Brahmana. “The man is the sacrifice because it is
112. SB 5.1.1.2; see also PB 7.2.1.

113. SB 12.6.1.1.
114. SB 3.1.3.17, 3.1.4.5, 3.3.3.5, 3.4.1.14, 3.9.4.11, etc.
115. KB 10.3.
116. Paul Mus, “La Stance de la plénitude,” Bulletin de l’Ecole Francaise d’ Extréme
Orient44 (1947-S0): 598.
the man who offers it. And each time he offers it, the sacrifice takes
the shape of the man. Therefore the sacrifice is the man.”!!7 Elsewhere
the “sacrifice is the man” because it is “by means of him that this all is
measured,” the body and its parts of the particular sacrificer being the
standard of measuring various parts of the sacrifical arena.!!8
Counterparts are in this limited sense only to be regarded as “substi-
tutes.”!!° They “stand for” or “represent” their originals in that they are
more or less incomplete images or emanations of them. A series of
resembling forms, from the prototype to its least complete manifesta-
tion, can bring together elements within a set. The pratima of Prajapati
is the year, and the counterpart of the year is the sacrifice of twelve days’
duration, because “the year has twelve months, and this is the pratima of
the year.”!2° A sacrifice lasting a whole year would, of course, be more
fully the form of Prajapati, but one of twelve days can also serve the
purpose and participate in the form albeit less completely. The doctrine
of counterparts makes possible not only ritual efficacy—the manipula-
tion of ritual counterparts in order to influence cosmic prototypes—but
also ritual efficiency: “The gods said, ’Find the sacrifice that will be the
pratima for one of a thousand years, for what man is there who could get
through with a sacrifice lasting a thousand years?”!?! Man can partici-
pate in cosmic rituals through resembling sacrifices gauged to the human
condition—again, not with the completeness of the original but with the
efficacy and efficiency of the counterpart.
Substitutions of this type occur also between the various forms of
the sacrifices in the Vedic repertoire, as we will have ample opportunity
to observe in later chapters. For now, let us return to the case of the
sacrificer and his oblation, a substitution within a sacrifice. The sacri-
ficer himself, the human being, is said to be “nearest to” (nedistha)
Prajapati, the prototype of prototypes.!”? In relation to others, however,
the sacrificer himself functions as a prototype with counterparts. An
animal, being more nearly resembling the sacrificer than vegetable obla-
tions, in turn stands as the prototype in relation to the lesser forms
within the series. The baked cake (purodasa) is the counterpart of the
117. SB 1.3.2.1; 3.5.3.1; see also TS 5.2.5.1.

118. See, e.g., SB 10.2.1.2 and also cosmogonic myths such as RV 10.121.5, and RV
1.154.1 and 1.154.3, where the creator god “measures out” with his body the universe and
its parts.
119. For a fuller discussion of substitution in Vedic ritualism, see Chapter 7 herein.
120. KB 25.15; KS 7.15.
WSR S.
122: $B 5.1.3.8, 5.2.1.6.
sacrificial animal (pasu) and, following the chain, also therefore the
pratima of the sacrificer.!
Closely related to the pramd/pratima concept is that of riupa
(“form”) and its derivatives pratiripa (“image” or “counterform”) and
abhiriipa (“adaptation” or “appropriate form”). I began my treatment
of Vedic metaphysics with a passage in which the transcendent brahman
manifested itself in the immanent worlds by entering as the “name and
form” of individual things and beings. Another text notes the participa-
tion of all forms in the One Form by saying that Prajapati “became the
counterform (pratiriipa) to every form (rapa). This [all] is to be regarded
as a form of him.”!4
In a recent article, Asko Parpola claims that “the world riipa is used
in the Brahmana texts in a meaning close to our ‘symbol.’ ”!° Follow-
ing Hermann Oldenberg, Parpola argues, “The basic model of thought
underlying the use of the term rupa is... the distinction between a
Platonic sort of idea and its physical manifestations.”!76 Often indeed
(although not in the example given in the previous paragraph), the
distinction between rupa and pratirupa is akin to that between prototype
and counterpart. Ripa is the form and pratiriipa the image, or, accord-
ing to Coomaraswamy, the former is the imago imaginans and the latter
the imago imaginata, “reflections” or “projections” of prototypical
rupa.'?’7 The reflection of a man’s form in a mirror, for example, is in
several texts designated a pratiripa, as is his offspring.!78
It is not at all certain, however, that symbol adequately captures the
usage here. And as the following citation indicates, pratirupas or resem-
bling images are made as well as discovered, a phenomenon that tends
to distinguish this conception from the Platonic one.!2° In the soma
sacrifice, the sacrificer spreads out two black and white antelope skins
123. 6TB 3:2:8:8.

124. RV 6.47.18; see also BAU 2.5.19; JUB 1.27.4-5, 1.44.1.
125. Also Parpola, “On the Symbol Concept,” in H. Biezais, ed., Religious Symbols
and Their Functions (Stockholm: Almquist and Wiksell, 1979), p. 143.
126. Ibid., p. 148, citing Oldenberg, Vorwissenschaftliche Wissenchaft, pp. 106ff.
127. Coomaraswamy, “Vedic Exemplarism,” p. 51.
128. See SanA 6.2, 6.11; BAU 2.1.8; GB 2.3.22; see also TB 3.9.22.2.
129. See O’Flaherty’s statement on the relations between “making” and “finding” in
the cosmogony and epistemology of later Hinduism: “These two kinds of creation—
making and finding—are the same, for in both cases the mind—or the Godhead—imposes
its idea on the spirit/matter dough of reality, cutting it up as with a cookie-cutter, now into
stars, now into hearts, now into elephants, now into swans. It makes them, and finds them
already there, like a bricoleur, who makes new forms out of objets trouvés.” Dreams,
Illusion, and Other Realities, p. 212.
Ritual and Reality Wi
and touches them, saying, “You two are the works of art (silpa) of the
verse and the chant.” The attached explanation notes, “A work of art is
a pratirupa. Thus what he means [when he says the mantra] is ‘You two
are the pratiriipas of the verse and chant.’ ”130
Whereas‘the stress in the prototype/counterpart theory lies on the
knowledge that the two participate in the same essence (though even
here there is often an active placing of the elements on an ontological
continuum), the force of the notion of pratirtipa is that the image of the
form is constructed and made to conform to its model. It bears repeating
here that all form in Vedism is constructed rather than given; the model
and its image are equally “works of art” in this sense, though one may
have been made prior to or more fully than the other.
We now come to the crucially important term abhiripa (“appropri-
ate form”), which returns us to the fact that in Vedic resemblance things
can be neither exactly alike nor completely different. They are “appro-
priate” to their cosmological and ontological level, which distinguishes
them from resembling forms at other levels, and yet they are neverthe-
less related by their form to correlates elsewhere. The various counter-
parts and images are neither identical to nor wholly distinguished from
either their (vertical) prototypes or their (horizontal) formal siblings by
the concept of abhiripa; their particular appropriateness locates them.
One instance of the creation of appropriate form occurs when a
formula or mantra is moved from one ritual context to another. The
original must be modified to interlink with its new context—a formula
including the word soma, for example, is usually modified when it is
brought to a sacrifice in which there is no soma offered—and yet its form
is said to persevere through the modification. “A Mantra should express
the action concerned,” writes S. C. Chakrabarti. “If it does, the Mantra
is called abhirupa.”™!
Such appropriateness of form is indeed a general concern of Vedic
ritualism. Ritual action, like all cosmogonic (or anthropogonic) pro-
cesses, must avoid excess. Like Goldilocks’ porridge, a ritual should be
too extreme neither in one direction nor in the other, but “just right.”
There should be neither an “overdoing” (atirikta) nor a “deficiency”
(nyuna) in the ritual performance;!? the composition should be bal-
130! SB3.2.1.5:
131. S. C. Chakrabarti, The Paribhasas in the Srautastitras (Calcutta: Sanskrit
Pustak Bhandar, 1980), p. 135.
132. E.g., SB 11.2.3.9. See also Jan Gonda, “The Redundant and the Deficient in
Vedic Ritual,” Vishveshvaranand Indological Journal 21 (June—December 1983): 1-34.
anced and “well made” (sukrta). In some texts the following phrase is
repeated over and over again: “What is appropriate in form (abhiriipa)
within the sacrifice is perfect (samrddha).”'
This statement seems to be multivalent. It indicates that the sacri-
fice is “perfected,” that is, made to resemble closely its prototype (the
sacrifice of the gods, or the cosmos itself), through the inclusion within
it of things, beings, and activities that are appropriate in form to that
prototype; one deploys entities, objects, or acts thought to be sayoni
(“of the same origin”) as the prototype. It also might well be taken as a
statement of hierarchical distinction. Particular sacrifices may attain a
relative perfection by composing themselves of things, beings, and activi-
ties appropriate to the particular scope and purpose of the particular
sacrifice—and, most importantly, appropropiate to the particular sacri-
ficer, who is the “owner” (svamin) or “lord of the sacrifice” (yajnapati).
The sacrifice, then, is simultaneously a garment ready-made (bought off
the rack, as it were) with a predetermined set of relations, and a product
tailor-made for the individual sacrificer. Again, the abhiruipa concept
makes possible both connection (to the vertically oriented prototype and
the horizontally oriented elements of the composition) and distinction
(the appropriate form is not identical to the prototype or to other similar
structures, e.g., the rituals performed by other humans). Such a concept
brings us into the very heart of what I observe to be the central principle
of Vedism: hierarchical resemblance.
Having considered the nexus of all connections (brahman) and the
important terms for those entities, objects, or acts that are connected,
let us now turn to the Vedic connections in themselves. Of the many
different types of linkages, the two most general are designated by the
names bandhu and niddna.'** The first is a more or less generic term
meaning “connection” or “relation” both in the sense of “relative, kin”
and in the more literal sense of “bond.” Bandhus place different ele-
ments “in bondage” one to another. As in the theory of counterparts,
bandhus can be between components located on the same ontological
135." See AitB 1216) 1-177-1-19, ete kB 11, 3253.9) ete. GB 2:36. 250; 2445
etc. For a study concentrating on the appropriate form of the verses used in particular
sacrifices, consult V. C. Bhattacharya, “On the Justification of Ripasamrddha Rk- Verses
in the Aitareya Brahmana,” Our Heritage 4 (1956): 99-106, 227-37, and Our Heritage 5
(1957): 119-46.
134. Others include brahmana, upanisad, mithuna (“pairing”), sampad and
samkhyana (“numerical agreement”), samjnd (“agreement” in general), sammita (“com-
mensurate”), salomata (“congruent”), and sayoni (“cooriginated”).
level—horizontal connections—or between resembling items on differ-

ent planes altogether. !55
Some scholars have argued that niddna is used to specify that the
connection is transversing cosmologically or temporally separate spheres.
Louis Renou has thus defined nidana as “une connexion a base d’identité
entre deux choses situées sur plans differents.”!* The priest called the
agnidhra, for example, is said “to be” the fire god Agni “by virtue of the
nidana” that puts the two in the relation of cosmic prototype and ritual
counterpart.!’ In other, and comparable, cases, nidanas correlate other-
wise temporally distant actions. A particular offering (the parndahuta,
“full oblation”) in which the clarified butter is poured into the fire with the
exclamation “svaha!” is considered “the same” as the first offering
Prajapati made with that ritual utterance because of the niddna joining
the two.!58 This type of connection also appears at times to be what some
scholars would call magic: by flinging a bunch of grass outside the sacrifi-
cial ground (the rite called stambayajus-harana), the priest expels the
demons “through the nidana” of the two acts.
Despite Renou’s attempt to delimit the meaning of the term nidana
in this way, other sorts of connections not involving the linking of
“choses situées sur plans differents” are labeled with the same word. To
trot out again an example I have used in other contexts, the relation of
resemblance between the sacrificer and the animal victim within his
sacrifice is sometimes described as one in which there is a nidana.™ It
might be argued that the victim, upon death, becomes the sacrifier’s
“representative” on another existential plane, but within the sacrifice
where the connection is made it is not at all clear that the connection
links elements of discrete cosmological levels.
135. See Jan Gonda, “Bandhu- in the Brahmanas,” Adyar Library Bulletin 29
(1965): 1-29.
136. Louis Renou, “ ‘Connexion’ en védique, ‘cause’ en bouddhique,” in Dr. C.
Kunhan Raja Presentation Volume (Madras: Adyar Library, 1946) p. 57.
137. SB 1.2.4.13, 4.4.2.18, 5.5.1.8. See also statements that the sacrificial cow is, by
the nidana, the goddess of speech, Vac, at SB 3.2.4.10, 3.3.1.16, 4.5.8.3—4; and that a
certain pressing stone in the soma sacrifice “is” both the sun and the “out-breath of the
sacrifice” for the same reason (SB 3.9.4.7, 4.3.5.16).
138. SB 2.2.1.4. See also SB 3.4.2.8, where the “alliance” made between the sacri-
ficer and the priests in the soma sacrifice (the tandnaptra) is connected by a nidana to a
prototypical alliance made among the gods.
139. SB 1.2.4.12.
140. See, e.g., AitB 2.11. See also SB 3.7.1.11, where the sacrificial stake (yiapa) to
which the animal victim is tied is also connected, by the niddana, to the sacrificer himself.
Furthermore, the bonds joining components from distinct worlds

are depicted with words other than niddna. Bandhu, the most general
term for “connection,” is often applied in precisely this sense. Perhaps
the best example comes from the Rg Veda, where it is said that “the
ancient seers discovered in their hearts the bandhu of the manifest (sat)
in the unmanifest (asat)” (1.129.4; see also JUB 1.59.10). The word
bandhu is indeed often defined in such a way as to put this function of
relating the visible to the invisible, the immanent to the transcendent,
foremost:
In the Brahmana texts it denotes above all the mysterious connection or
relation between the entities of this world and the transcendental “ideal”
entities of the divine world, which are the foundation and origin of the
perceptible things.!4!
Regardless of the particular term used, it is this ability to link the
visible, manifest, and therefore limited counterpart to its invisible, tran-
scendent, and unlimited prototype that is the most spectacular of the
supposed effects of making connections. Placing a tangible, definite,
bounded thing, being, act, or form into connection with its prototype is,
as Louis Renou and Lilian Silburn write, “the effort to specify beyond
well-known things, beyond definite forms, a hidden zone where the
things and forms take on an inorganic aspect . . . which makes them
redoubtable.”!*2 The “unlimitedness” (aniruktatva) of the transcendent
is made accessible by those connections that reach out from the sacrifice,
and therefore the limited is ontologically “perfected”: “Applied to de-
fined and ‘structured’ things, anirukta is what completes and perfects
them. . . . According to a quite Vedic paradox, anirukta is that which
perfectly finishes by the very fact of its being unfinished.”!?
Connections thus bring together the immanent and the transcen-
dent in such a way that the inaccessible is made accessible by the play of
resemblances, and the manifest is fulfilled by participation in the tran-
scendent. The universal elements, emitted from the Cosmic One, attain
their full, actualized reality only when linked one to another and to their
point of origin. Such a composition based on connection rejoins the
Cosmic One into a unity of parts in which the simple and the limited—
while remaining simple and limited—participates in the whole, the un-
141. Parpola, “On the Symbol Concept,” p. 150. For other definitions of bandhu,
see Gonda, “Bandhu- in the Brahmanas.”
142. Renou and Silburn, “Nirukta and Anirukta in Vedic,” p. 76.
143. Ibid., p. 77. Mus addresses the same point in Barabadur, p. 51.
limited. Each particular “name and form” can realize its true nature only
by finding its place in this chain of resemblance—or, rather, by being
placed “in bondage” with all of its counterparts under the umbrella of
the prototype. Universal resemblance keeps separate while it unifies, its
specific economy regulated by avoiding the extremes of identity and
individuality.
4
The Ritual Construction of Being
Vedic Embryology and Other Conceptions of Human Deficiency
If, in Vedic cosmological theory, creation is not cosmos but only a

flawed precondition for the ritual task of construction, so, too, at the
level of human beginnings procreation is not the bringing into existence
of an ontologically sound human being. Vedic conceptions of conception
and reproduction reproduce the conceptions of cosmic creation exam-
ined in the last chapter: birth and anthropogony are, like creation and
cosmogony, regarded as two separate moments, distinguished by the
absence or presence of ritual reclamation. The bringing into existence of
a viable human being is, the ritualists proclaimed, a matter of ritual
rather than merely biological labor.
The procreative act of Prajapati is the prototype of all procreative
acts, including the sexual act which engenders new human life. Humans
are emanated parts of the creator god and are thus subject to the natural
defectiveness of all of Prajapati’s issue. And, as microcosmic versions of
the macrocosmic creator, humans replicate and recapitulate in their acts
of reproduction the creation of the universe as a whole. Human life is a
process of ritually constructing and refining a self, an adtman, out of the
raw materials of aboriginal creation, just as the atman of Prajapati is
reconstituted (literally “remembered”) by the postcreation rituals of the
gods.
Unlike the unified Prajapati of the totalistic cosmogonic process,
however, the anthropogonic constructs—individual humans—are shaped
into different types, made into a variety of products. Different kinds of
human selves are constructed with different ontological values assigned,
The Ritual Construction of Being 83
different degrees of the realization of completeness, different actualiza-

tions of the human ideal and potential. All selves are born defective and
incomplete, but particular selves were supposed to have differing poten-
tials for overcoming inherent defects. Birth as a male, for example, auto-
matically qualifies one for the possibility of realizing a degree of human
completeness—in the Vedic ritualistic view—not possible for females.
Birth as the son of a learned Brahmin presents opportunities not available
to the offspring of a Sidra. Birth into a wealthy family holds out the
potential for a self-realization different from that possible for a member
of a family of lesser means.
Humans, then, are the result of both their inborn potential and
their realization of it, and these two components come in unequal por-
tions. Intrinsic, inherent potential and the actualization of that potential
through a personal record of ritual performance combine to create dis-
tinctions in the “competence” (adhikara) of particular individuals. And
these distinctions are not regarded as merely social ones; they are rather
envisioned as ontological differences in the quality of the human self
produced in this life and of the “divine self” (daiva dtman) that the
human will assume in the next world.
Vedic theories of conception and embryology! can be divided into
two principal types, not coincidentally reduplicative of the two types of
cosmogonic patterns identified in Chapter 3. Creation, it will be remem-
bered, is a matter either of separating and distinguishing a formless unity
or of reuniting an excessively differentiated set of emitted parts into a
formed whole. In both cases it is an excess that must be moderated—an
excess of similarity (Jami) or an excess of difference (prthak)—in an
_effort to attain a structure of connected resembling parts.
The cosmic emission of Prajapati is replicated by the emission of
semen in the act of human procreation. Semen is considered the “es-
sence” or “life sap” (rasa) of the man, his condensed representative: “Of
created things here [in this world] earth is the essence; of earth, water;
of water, plants; of plants, flowers; of flowers, fruits; of fruits, man; of
man, semen.”? When this fluid, the concentrated identity of man, is
emitted into the womb of the woman, it is as if the very self of the
procreator has been reproduced in the form of the new embryo:
1. See George William Brown, The Human Body in the Upanishads (Jubbulpore:
Christian Mission Press, 1921); and Paul Deussen’s discussion of the topic in his The
Philosophy of the Upanishads, trans. by A. S. Geden (New York: Dover Publications,
1966), pp. 283-96.
2. BAU 6.4.1.
In a person, this one [the dtman] first becomes an embryo. That which is
semen is the luminous power (tejas) extracted from all the parts [of the
man]. In the self, truly, one bears another self. When he emits this into a
woman, he then begets it. This is one’s first birth.*
The male self is thus wholly, in condensed form, transmitted to and

perpetuated through the male offspring, and in some texts this redu-
plicative view of conception seems to be regarded favorably: “The father
enters the wife; having become an embryo, he [enters] the mother.
When he is renewed in her, he is born in the tenth [lunar] month.” It is
indeed generally the case that, as one text puts it, one desires a “counter-
part” (pratirupa) of oneself in the form of one’s child, and not a “counter-
feit” (apratirupa).5
But this vision of reproduction also may be viewed as problematic,
for the product of the generative act is redundant. The embryo here is an
exact replica of the father. Creation so conceived recalls the macrocos-
mic error of ja@mi, excessive resemblance. The fetus, a reduplication of
the father, is undifferentiated until it is recast and individually shaped in
the womb of the mother: “it comes into individual self-becoming (atma-
bhiya) with the woman, just as a limb of her own. Therefore it does not
injure her. She nourishes this self of his [the father] that has come to
her.”® Elsewhere the individualizing necessary to avoid the ontological
flaw of jdmi is said to be attained by the intermingling of both male
semen (the essence of the father) and female seed (the essence of the
mother). The dual emission produces an independent and sufficiently
distinct third entity:7
A male child is born when the seed of the man is greater [than that of the
woman], and a female child when the seed of the woman is greater [than
that of the man]. If both are equal, a hermaphrodite is born, or a boy and
a girl [i.e., fraternal twins]; and if the seed is weak or scanty, there will be
a miscarriage.®
In any case, the male semen that carries the self of the father and
creates a self which is the embryo must be tempered by, or combined
. AitU 2.4.1; see also AitB 7.13 and AitA 2.5.

. AitB 7.13; see also Manu 9.8.
. §A 6.11; see also BAU 2.1.8.
. AitU 2.4.2; see also AitA 2.5.
W
Hh
SHN. For the relationship between male and female fluids, see Wendy Doniger O’Fla-
herty, Women, Androgynes, and Other Mythical Beasts (Chicago: University of Chicago
Press, 1980), pp. 17-61.
8. Manu 3.49.
with, the female component—either in the womb itself, which provides

space for an organic “self-becoming,” or by means of the addition of a
commingled female seed. The excessively undifferentiated semen is thus
to be counteracted in some way in order to avoid the redundant sterility
of jami.
In this theory of human reproduction the first ontological task is to
distinguish the self of the fetus from the essence of the father’s self
represented by his semen. We are reminded here of the cosmogonic
myths in which the emitted creatures are “clasped together” in unproduc-
tive union, “at odds with one another” in rivalry stemming from exces-
sive similarity, and who are fearful of being “eaten” or resumed into the
body of their creator.
Another theory of human origins, equally stressing the defective-
ness of those origins, is also expounded in Vedic texts. This view, how-
ever, turns on the opposite ontological problem of excessive differentia-
tion or prthak. Just as Prajapati in the cosmic beginning emits creatures
who are sometimes described as dispersed and in disarray until restruc-
tured into forms and placed into a relation one to another, so too is the
human emission of semen sometimes envisioned as an act of radical
dispersion and discontinuity, an act of problematical dissemination. The
seed is described as “scattered” (sikta) into the womb in a microcosmic
version of the macrocosmic dispersal of Prajapati’s creatures, and both
microcosmic and macrocosmic fathers are depicted as “drained” after
their emissions.?
Procreative emission here, as in the macrocosmic case, is followed
by ontologically formative procedures. The semen ejaculated into the
—_—
womb is subsequently “transformed” or “developed” by ritual means, in
‘some cases with the aid of Tvastr, the divine builder and shaper of
forms.!° The seed is thus converted from its disseminated state and
metamorphosed into a ontological composition. Through ritual manipu-
lation, “he thereby transforms that scattered seed, and so the scattered
seed is transformed in the womb.”!! And, as that same text declares,
“whatever the seed is transformed into in the womb, as that is it born.”!?
The “transformation” of the unshaped (because it is either redun-
9. See, e.g., TS 6.6.5.1, where the cosmogonic act is conflated with both the human
act of reproduction and the sacrificial act of generation: “Prajapati emitted offspring. He
thought himself drained. ... He who sacrifices creates offspring. Then he [too] is
drained.”
10. See, e.g., KB 3.9; SB 1.9.2.10, 3.7.2.8, 4.4.2.16, 7.2.1.6.
fis (SB te 1s47:
12. $B-6:4-2.7.
dant or scattered) human potential into the reconstituted human being,

beginning at conception and continuing through birth, childhood, and
adolescence, is accomplished by the performances of rituals known as
samskaras. Often translated as “rites of passage,” the samskaras are
really rituals of ontological healing, construction, and perfection, as we
shall see in the next section. Those born into the Brahmin, Ksatriya, or
Vaisya classes are expected to progress out of the state of inborn incom-
pleteness and realize themselves, in different ways and to different de-
grees, through ritual action. A member of any one of these classes is
born defective but also possesses inherent proclivities that should be
actualized and exploited.
As Mircea Eliade has said in a passage that sums up much of his
insight into the nature of homo religiosus, “every creation has a para-
digmatic model—the creation of the universe by the gods.” Vedic
ontology—the creation of human being—is a process of ritual labor
and construction replicating at the anthropological level the ritual con-
struction of the universe at the cosmological level. The fabrication of
the self, like the fabrication of any or all of the “names and forms” that
comprise the ordered cosmos, is a matter of lending a degree of integ-
rity to an otherwise disfigured, unrealized, and unstable set of potenti-
alities. “In this dynamic world,” writes Verpoorten, “where all things
at every moment flirt with effervescence, stability and continuity are
the major preoccupations.” For, as we have also observed, the “natural
state” of all things and beings is “mere juxtaposition (abhidha-)” of the
parts.!4
The movement from a condition of ontological defectiveness to that
of integrated completeness which we have traced in cosmogonic myths is
recapitulated in the ritual processing of individual human beings. Ritual
formation and reformation of the unformed potential begins with those
rites called the samskaras. The term is derived from the Sanskrit root kr-
(“to do or make”), with the integrative prefix sam-, and connotes, as
Lilian Silburn writes, “the activity which creates forms, adapts them,
and makes them perfect.”!5 Or as Raj Bali Pandey puts it in his Hindu
Samskaras, all of these rituals aim not only at “the formal purification of
13. Mircea Eliade, The Sacred and the Profane, p. 31.

14. Verpoorten, “Unité et distinction,” p. 62.
15. Lilian Silburn, Jnstant et cause, p. 58. See also Jan Gonda’s definition, Vedic
Ritual: The Non-solemn Rites (Leiden: E. J. Brill, 1980), p. 364: “The untranslatable term
samskara is etymologically related to the verb samskaroti which expresses the general
meaning of ‘composing, making perfect, preparing properly and correctly with a view to a
definite purpose.’ ”
the body but at sanctifying, impressing, refining and perfecting the en-
tire individuality of the recipient.”
In order to understand the ontological importance of the samskdaras
in Vedic ritualism, it is instructive to look at the depictions of those who
have not yet undergone or do not ever (or did not when required)
undergo them. These unregenerates and, therefore, in the eyes of the
ritualists, degenerates are all regarded as less than really human, petri-
fied in a state of nonbeing. Every human, as we have seen, is cast into
this condition originally. Some persons and groups, however, are consid-
ered permanently faulty and incomplete. Still others are temporarily
returned to the aboriginal state through neglect or illness. All again are,
at death, placed into an ontological limbo complementary to that of the
newly procreated fetus. In all of these instances, the “firm foundation,”
the structured protection that ritual activity affords against the perpetual
tendency for all things and beings to deconstruct and return to their
elemental formlessness—of either the jami or prthak variety—is absent
or has been lost.
Although the rituals of healing and construction begin with the act
of intercourse and conception, up until the time of initiation into Veda
study and sacrificial ritualism (the samskdara known as the upanayana),
the boy is considered incomplete and not yet responsible. A Brahmin,
for example, who has not undergone initiation is called a “Brahmin by
birth only”’’ or a “Brahmin by relation only.”!* They are in a state of
nature and may therefore “act naturally,” but they are also prohibited
from the cultural act par excellence of sacrifice: “Before initiation [a
boy] may follow his inclination in action, speech, and eating [but he
shall] not partake of sacrificial offerings.”!9 Such natural humans do not
deserve the designation Brahmin at all; they are likened to wooden
elephants or leather antelopes which “have nothing but the name” and
are “equal to Stdras”2°—for the members of this latter class have not
been “reborn” into Aryan society through the initiation and therefore
are known as the “once born.”?!
The uninitiate, regardless of the class of his birth, is likened in many
texts to others who are also, and for the same reasons, defective, those
16. Raj Bali Pandey, Hindu Samskaras: Socio-Religious Study of the Hindu Sacra-
ments, 2d ed. (Delhi: Motilal Banarsidass, 1969), p. 17.
17. BGS 1.7.1; VaikhGS 1.1.
18. ChU 6.1.1.
19. GautDhsS 2.1.
20. VasDhS 3.1, 3.3, 3.11; see also Manu 2.157.
21. See Manu 10.4 and GautDhsS 10.50.
who are permanently excluded from Vedic rituals and thus are perma-
nently deficient, such as the Sudras:
They do not put any restrictions on the acts of [a child] before the initia-
tion, for he is on the level with a Sidra before his [second] birth through
the Veda.”
No religious rite can be performed by [a boy] before the initiation, be-
cause he is on the level with a Sidra before his birth through the Veda.”
So it is that, as the Atri Smrti declares, “By birth everyone is a Sudra;
by the performance of the upanayana one is called a twice-born.”*4 The
Siidras live in a state of inherited and perpetual irresponsibility—they
cannot perform an act serious enough to fall from position, being already
at the bottom of the heap*®—and are permanently stuck in the childlike
condition of natural deficiency that members of the other classes pass
through but can also pass out of by means of ritual renovation.
Whereas the defects of the Sidra are supposedly inherited and
permanent traits, others may lapse into an equivalent condition of onto-
logical defectiveness. If a Brahmin passes his sixteenth year without
undergoing initiation, if a Ksatriya passes his twenty-second or a Vaisya
his twenty-fourth, they become “fallen ones,” the patita-savitrikas, who
have lost their right to learn the Veda and to sacrifice.** These negligent
men, bereft of and now excluded from the perfecting and healing force
of ritual, degenerate instantly into a status identical to that of the Sidra:
“No one should initiate such men, nor teach them, nor perform sacri-
fices for them, nor have intercourse with them.”?” Elsewhere they are
called “slayers of the brahman” and “burial grounds.”8
In yet other texts, those who have “fallen” from the Veda and
Aryan society are equated with another group in ancient India, about
which little is known: the vrdatyas.2? Some believe these people were
22. BDhS 1.2.3.6.

23. VasDhS 2.6; see also ApDhS 2.6.15.18-25.
24. Atri Smrti 141-42.
25. Manu 10.126.
26. SGS 2.1.9; PGS 2.5.39.
27. PGS 2.5.40; see also JGS 1.12; AsvGS 1.19.9; SGS 2.1.10-13; VGS 5.3; GGS
2.10.6; Manu 2.40; VasDhS 11.75.
28. ApDhS 1.1.1.32, 1.1.2.5.
29. For the equation of the patita-savitrikas and the vratyas, see Manu 2.39 and Yaj
1.38. The older secondary literature on the vratyas is summarized in J. W. Hauer, Der
Vratya, 2 vols. (Stuttgart: W. Kohlhammer, 1927), I: 5-40. More recent studies include S.
Biswas, “Uber das Vratyaproblem in der vedischen Ritualliteratur,” Zeitschrift der
Deutschen Morgenlandischen Gesellschaft 105 (1955): 53-54; the iconoclastic study of J. C.
beyond the Aryan pale altogether, and others argue that they were a
renegade branch of Aryan society; in any case it is clear that the vratyas
were, in the eyes of the mainstream ritualists, like the Sidras and the
“fallen” ones, defective and incomplete because of their separation from
the Vedic ritual. They are depicted as strange, perverted, and lesser
forms of human beings, whose practices are topsy-turvy versions of
orthodox Vedism:
Those who lead the life of a vratya are defective (hina) and left behind.
For they neither practice the study of the Veda nor do they plough or
trade. . . . Swallowers of poison are those [vratyas] who eat foreign food
as if it were the food of a Brahmin; who speak improperly as if it were
proper; who strike the guiltless with a stick; and who, although not initi-
ated, speak the speech of the initiated.*°
The patita-sadvitrika degenerates into a condition similar to that of
the Sidra and the vratya through negligence. The physically afflicted
man decomposes into much the same state when, as a result of posses-
sion, “sin,” or other causes, he loses temporarily the continuity of being
that ritual activity does so much to foster and protect.*! “The vital
breaths pass through him who is ill,”>? just as they did when Prajapati
was in his postprocreative state. “Food goes forth from him,”*? escaping
from the body through the rupture that is sickness, just as the creatures
run away from the weakened and “drained” Prajapati. “He who is ill is
without a firm foundation,”* returned to the precariousness of impaired
origins, represented at the cosmic level by Prajapati, who is collapsed in
a heap awaiting reconstruction and healing.
One text, recalling the famous story of Indra’s slaying of the serpent
Vrtra, compares the depleted strength of Indra after the battle to the
Heesterman, “Vratya and Sacrifice,” Indo-Iranian Journal 6 (1962): 1-37; and Radha-
krishna Choudhary, Vratyas in Ancient India (Benares: Chowkhamba Sanskrit Series
Office, 1964).
S05 PBol7.1,2.
31. The various causes of different sicknesses according to Vedic and Hindu texts
are covered in G. U. Thite, Medicine: Its Magico-Religious Aspects according to the Vedic
and Later Literature (Poona: Continental Prakashan, 1982), pp. 4-38. See also Kenneth
Zysk, “Towards the Notion of Health in the Vedic Phase of Medicine,” Zeitschrift der
Deutschen Morganldndischen Gesellschaft 135 (1985): 312-18; idem, Religious Healing in
the Veda (Philadelphia: American Philosophical Society, 1986); and Johanna Narten,
“Ved. amayati und admayavin-,” Studien zur Indologie und Iranistik S—6 (1980): 153-66.
32. PB 18.5.11.
33: PBi5 13:3.
34. PB 16.13.4.
quintessential disease of depletion and rupture—diarrhea—in addition

to other comparable conditions: those of an expelled king, a childless
though fertile man, and the generically ill man:
Indra slew Vrtra. His strength went asunder in every direction. The gods
searched for an expiation for him, but none would please him. Only the
strong soma rite (tivra-soma) pleased him [and restored him strength]. He
should perform [this rite] for one who drinks soma only to have it pass
through him. Ruptured, as it were, is he whose soma passes through him.
The performance of the strong soma rite serves to block up, to rectify the
rupture. He should perform it for a king who is expelled [from his king-
dom]. The subjects pass through him who has been expelled from his
kingdom. The performance of the the strong soma rite serves to block up,
to rectify the rupture. ... It should be performed by one who wants
offspring. Offspring pass through him who, though fertile, gets no off-
spring. The performance of the strong soma rite serves to block up, to
rectify the rupture. . . . He should perform it for one who is ill in any way.
The vital breaths pass through him who is ill. The performance of the
strong soma rite serves to block up, to rectify the rupture.*
Finally, we may turn to the case of the recently deceased, who is

situated between a former life as a man and a future existence as an
ancestor. The transition is not immediately effected upon death; the
spirit (preta) must be ritually made over time into an ancestor (pitr), just
as the embryo and infant are constructed into a human being by ritual
means: “The dead man does not immediately after his death and without
more ado join the number of the ancestors who are worshipped; on the
contrary fixed ceremonies are necessary for elevating the deceased to
the rank of ancestors.”*° There is, as David Knipe puts it, a “series of
bodily constructions and dissolutions the deceased undergoes before
becoming established in the world beyond.”*’ The rituals of death, mir-
roring in many ways the samskaras of life, begin with the cremation of
the body. The once integrated constituents of the human body are dis-
solved into their parts and redistributed throughout the cosmos:
The voice of the dead man goes into fire, his breath into wind, his eye into
the sun, his mind into the moon, his hearing into the quarters of heaven,
35. PB 18.5.2—-6, 18.5.9, 18.5.11.

36. Dakshina Ranjan Shastri, Origin and Development of the Rituals of Ancestor
Worship in India (Calcutta: Bookland, 1963), p. 62.
37. David M. Knipe, “Sapindikarana: The Hindu Rite of Entry into Heaven,” in
Frank E. Reynolds and Earle H. Waugh, eds., Religious Encounters with Death: Insights
from the History and Anthropology of Religions (University Park: Pennsylvania State
University Press, 1977), p. 111.
his body into the earth, his self into space, the hairs of his body into trees,
and his bicod and semen are placed in water.*8
Man is decomposed into his natural and dispersed elements at death
precisely as the Cosmic One must first become multiple before a new
reformation can emerge when the parts are recomposed. And as
Prajapati, in his liminal state after procreative emission, but before
ritual reconstruction, is described as depleted or “fallen to pieces,” so
too is the preta, the “gone out” or “departed” human spirit, depicted as
dangerously detached before its reunification and transformation into an
ancestor: “They say of the preta, ‘He has been cut off.’ ”°9
The uninitiated child, the Sadra, the patita-savitrika, the vratya, the
sick man, and the newly dead—all these conditions are comparable regres-
sions to the primary and fundamental human condition of defectiveness.
Sacrifice is in Vedism the ontological medicine to fix all deficient cre-
ations, to repair all debilitated forms. Those precluded from, without
access to, or who voluntarily reject the healing balm of ritual are perma-
nently disabled, permanently incomplete human beings from the Vedic
point of view. And those who are eligible for and partake of the ontologi-
cal salve that ritual provides must perpetually take precautions. The ritu-
als must be regularly repeated to protect against the natural inclination of
structured entities to deconstruct into their atomistic constituents.
Constructing Selves and Statuses
Although it is not until the Sutras (ca. 700-300 B.c.£.) that the term
samskara is used to designate a specified set of particular rituals—
beginning with one performed at the conception and ending with the
death rites—other formations of sam- + kr- (e.g., samskrti, samskrta,
samskaroti, samskurute) describing the effects of rituals on humans do
appear regularly in the Veda, and with much the same meaning. As
early as RV 5.76.2, a form of the verb appears when a certain vessel, the
gharma, is to be made samskrta and therefore ready for ritual use; or,
again, at SB 1.1.4.10 an offering must be made samskrta before it may
be presented to the gods.”
Samskara rites, in all periods of Indian history, may be performed
either on things or on people. In all cases, according to the great philoso-
pher of ritual Sabara (commenting on PMS 3.1.3), the purpose of these
38. BAU 3.2.13; see also RV 10.16.3; SB 10.3.3.6.

39. SB 10.5.5.13.
40. For other examples of the use of the verb in early texts, see below, p. 101.
kinds of rites is to make the thing or being fit (yogya) for its function.
Fitness is said to be of two kinds: that which arises by the removal of taints
(especially those congenital faults stemming from biological existence)
and that which arises through the ritual generation of fresh inner quali-
ties.4! Still another definition highlights the creative role of these rites:
“Just as a work of painting gradually unfolds itself on account of the
several colors [with which it is drawn], so Brahminhood is similarly
brought out by samskaras performed according to the prescribed rites.”
Following the ritual philosophers, then, one might best consider
these rituals not as rites of passage—implying leaps from one ontologi-
cally stable condition to another—but rather both as rituals of healing
(eliminating the congenital defects of the self) and as rituals of construc-
tion (adding new qualities, actualizing latent tendencies, and perfectly
integrating them into an ontological composition). From conception to
birth, from birth to the “second birth” of initiation, and from marriage
to death, the life of the Vedic male* was punctuated by these rituals of
healing and formation, of transformation and transfiguration.
It is unnecessary to delve here into the analytical details of all of the
samskaras.“ A glimpse at one, and perhaps the most important, of the
series might suffice to give some indication of the social and, reputedly,
ontological function (what might be called the socio-ontological effect)
that these rituals had for those encompassed within Vedic ritualism.
Although the upanayana* is not specifically mentioned in the Rg
41. The Tantravartika on PMS 3.8.9, cited in P. V. Kane, History of DharmaSastra,

2d ed., 5 vols. (Poona: Bhandarkar Oriental Research Institute, 1968-75), II, 1: 190-91.
See also the definition by the later mimamsaka Mitramisra, who wrote that samskdaras
impart the necessary qualities for the performance of other rituals or eliminate congenital
defects and impurities (also cited in Kane, p. 191); and statements to similar effect in
Manu 2.26—28 and Yaj 1.13.
42. Parasara Smrti 8.19. The citation is also quoted in A. K. Ramanujan’s afterword
to his translation of U. R. Anantha Murthy’s novel Samskara: A Rite for a Dead Man
(Delhi: Oxford University Press, 1978), p. 141. See also Kane, History of Dharmasastra,
II, 1: p. 191: “The samskaras had been treated from very ancient times as necessary for
unfolding the latent capacities of man for development.”
43. Females did not undergo all the samskaras and were in particular prohibited
from the upanayana.
44. The interested reader might consult Pandey, Hindu Samskaras; Kane, History of
DharmaSastra, Vol. 11; Gonda, Vedic Ritual; and H. N. Chatterjee, Studies in Some
Aspects of Hindu Samskaras in Ancient India (Calcutta: Sanskrit Pustak Bhandar, 1965).
45. For this samskara, see Karl Glaser, “Der indische Student,” Zeitschrift der
Morgenlandischen Gesellschaft 66 (1912): 1-37; and Brian K. Smith, “Ritual, Knowledge,
_and Being: Initiation and Veda Study in Ancient India,” Numen 33 (1986): 65-89. For the
meaning of the term “upanayana,” consult Jan Gonda, “Upanayana,” Indologica Tauri-
nensia 7 (1979): 253-59.
Veda,* it does appear as a fixed ritual already in the Atharva Veda

Samhita’ and the Satapatha Brahmana,* and the details of the rite
given in those relatively early texts conform in most details to the
initiation ritual explicated in the Sitras. It appears that at a very early
date in Vedic history the upanayana and subsequent period of Veda
study were regarded as mandatory for the Aryan community.” By the
time of the Sutras, the upanayana was called simply the samskdara® to
indicate its paradigmatic significance, and according to Ram Gopal the
term samskdara “was gradually extended [from the upanayana] to all
domestic rites.”>!
Upanayana initiated a young boy into the study of the Veda under
the guidance of a teacher and into the performance of the sacrifice.*
But it was also an initiation into Aryan society; he became, by virtue of
the rite, “twice born” (dvija), a term that occurs as early as the
Atharva Veda Samhita.*3 In Manu, this “birth from the Veda” is explic-
itly contrasted with the merely biological birth from the mother and is
called the “real” (satya) birth**—a blatant move to devaluate the prod-
ucts of women and extol the cultural/ritual labor of men. This second
birth was thus claimed to be a socio-ontological birth standing in radi-
cal opposition to the defective natural birth, and was designed to rec-
tify biological faults and construct a higher ontological existence for the
young boy.
In this way, the natural birth of individual humans replicates the
natural birth of the cosmos as a whole, for both are regarded as equally
degenerate. And just as we understood the cosmogonic myths as state-
ments of the ineptitude of Prajapati, the creator god, so might we under-
stand the denigration of natural birth as statements of the ineptitude of
human mothers. The ritualists felt that only they were ultimately quali-
fied to produce proper offspring; both the cosmos and individual hu-
46. There are references to the period of Veda study under the tutelage of a teacher,
called brahmacarya, at RV 3.8.4 and 10.109.5, however.
47. AV 11.5.
48. SB 11.3.3, 11.5.4.
49. As early as the Taittiriya Samhita (6.3.10.5), a Brahmin was said to have been
born with three debts, one of which was to the primordial seers which was repaid through
brahmacarya.
50. PGS 2.5.42-43.
51. Ram Gopal, India of the Vedic Kalpasutras (Delhi: National Publishing House,
1959), p. 254.
52. For more details on these two aspects of the upanayana rite, see Smith, “Ritual,
Knowledge, and Being.” ‘
53. AV 11.5.3; for later usage, see ApDhS 1.1.1.15; GautDhS 1.8; Manu 2.68.
54. Manu 2.169-70, 2.148.
mans could be “truly born” only from a womb controlled by priests,

from the womb, that is, of the sacrifice.
The initiation or second birth was to occur at the age of eight for a
boy born of Brahmin parents, eleven for a Ksatriya, and twelve for a
Vaisya, the Sidras being excluded from the rite.55 And we have ob-
served above the consequences of failure to undergo the ritual. Because
of this “vehemence with which those who failed to undertake the
upanayana ceremony were condemned,” Jan Gonda concludes that the
ritual was “more or less a compulsory institution” in Vedic India.°®
Through it one was constituted an Aryan—a true human being. But
even within the order of the Aryan, the fully human, socio-ontological
distinctions were made between classes. The upanayana generated not a
generic “humanness” in the initiate but rather created humans who were
thought to be of different general types. As one text puts it, the socio-
ontological classes of Aryan society were distinguished not only by in-
born nature (prakrti-visista) but also by being constituted or made dis-
tinct through the ritual samskaras (samskara-visesa).°7
The special importance of the upanayana, then, was not only that it
initiated the individual into human society; it was also the means by
which the posited innate and distinctive class proclivities of ancient In-
dian groups were actualized. The class differentiations within the ritual
are a predominant feature, and this is no accident (see Table 1). In
hierarchical, as well as ritualistic, Vedic India, humans are ritually made
different, and differently.
I have already noted that the ideal times for undergoing the initia-
tion vary according to class—with the recommended age for the more
intellectually capable Brahmin lower than those for the other classes—
and so, too, is there a distinction among the ages after which initiation
may not take place.’ It is also stated that different seasons are most
55. For the ages for upanayana, reckoned either from conception or from birth, see
ASvGS 1.19.1-4; SGS 2.1.1-5; BGS 2.5.2; BhGS 1.1; HGS 1.1.1.2-3; ApGS 4.10.2-3;
VaikhGS 2.3; VGS 5.1-2; MGS 1.22.1; PGS 2.2.1-3; GGS 2.10.1-3; KhGS 2.4.1-6;
ApDhS 1.1.1.19; BDhS 1.2.3.7-9; Manu 2.36; VasDhS 11.49-51; GautDhS 1.5, 1.11. For
optional ages for the attainment of different qualities, see BGS 2.5.5; VGS 5.1-2; JGS
1.12; ApDhS 1.1.1.20-26; Manu 2.37; GautDhsS 1.6.
56. Jan Gonda, Change and Continuity, p. 391.
57. VasDhS 4.1.
58. Sixteen for the Brahmin, twenty-two for the Ksatriya, and twenty-four for the
Vaisya. See ASvGS 1.19.5-7; SGS 2.1.6-8; BGS 2.5.3-4; VaikhGS 2.3; VGS 5.3; PGS
2.5.36-38; GGS 2.10.4; KhGS 2.4.6; ApDhS 1.1.1.27; BDhS 1.2.3.12; Manu 2.38;
VasDhS 11.71-73; GautDhS 1.12-13.
Table 1. Class Differences in the Initiation Ritual

eS a Se kd AEE, SOA RE Re, Sar on ke ee
Brahmin Ksatriya Vaisya
ee ee eee ee ee ETS et Fee ee he
Age for initia- 8 11 ilps
tion
Last age for 16 22 24
initiation
Meters of gayatri tristubh Jagati
Savitri
verse
Seasons for spring hot season autumn
initiation
Upper gar- black antelope spotted deer goat or cow
ment
(uttariya,
ajina)
Lower gar- linen or hempen cotton or linen woolen
ment
(vasa)
Color of reddish brown red yellow
lower gar-
ment
Girdle from munija grass bowstring woolen or hempen
(mekhala)
Staff (danda) palasa or bilva nyagrodha badara or udumbara
Size of staff reaches top reaches forehead reaches nose
of head
appropriate for the performance of the ritual by different classes: Brah-

mins undergo the upanayana in the spring, Ksatriyas in the hot season,
and Vaisyas in the autumn.*? It is not accidental that these same correla-
tions between socio-ontological classes and the seasons of the year ap-
pear also in the Srauta Sutras in discussions of the auspicious times for
married householders to light the sacrificial fires. As Pandey notes,
59. See, e.g., ApGS 4.10.4; HGS 1.1.1.4; VaikhGS 2.3; BDhS 1.2.3.10; ApDhS
1.1.1.19. BGS 2.5.6 and BhGS 1.1 also allow for the initiation of the rathakara (“chariot-
maker“), the offspring of a Vaisya man and a Sidra woman. His initiation should occur in
the rainy season. For other stipulations on the time for upanayana, see HGS 1.1.1.5 and
BhGS 1.1.
60. See R. N. Dandekar, ed., Srautakosa: Encyclopedia of Vedic Sacrificial Ritual, 2
vols. (Poona: Vaidika Samsodhana Mandala, 1958-82) I, 1: 1-26.
These different seasons were symbolical of the temperament and occupa-

tion of different castes. The moderation of spring symbolized the moderate
life of a Brahmin; the heat of summer represented the fervour of a ksatriya;
autumn, when the commercial life of ancient India reopened after the rainy
season, suggested the wealth and prosperity of a vaisya. . . .©!
The individual items of the uniform of the student, bestowed during

the course of the initiation rites, also served to differentiate types of
students and to highlight and realize the particular qualities peculiar to
each. Every boy was given an upper garment, a lower garment, a belt, a
staff, and, according to only some of the ancient texts, the sacred thread
(yajnopavita) that would become so central to later Hindu conceptions
of the upanayana.
In most cases, the differences in uniform systematically draw on
ancient associations—that is, resemblances—between parts of nature
and the qualities of the different socio-ontological classes. The upper
garment, for example, is to be a skin from the black antelope if the
initiate is a Brahmin, from the spotted deer if a Ksatriya, and from a
goat or a cow if a Vaisya.? In the Brahmanas, the black antelope is
consistently connected with Brahmins,“ the brahman power,® the
“lustre of the brahman power” (brahmavarcasas),® and the sacrifice,*
the workplace of the Brahmin priest. Although the connection between
the Ksatriya and spotted deer is less straightforward, presumably the
latter resounds with physical and royal power. This conclusion is par-
tially confirmed by two facts. First, in a variant opinion on the proper
skin for a Ksatriya initiate, one text recommends that of the tiger.”
61. “. . . and the easy time of rains indicated facility for a chariot-maker.” Pandey,
Hindu Samskaras, p. 127.
62. It appears that the earlier practice was to use the upper garment to serve the
function later taken by the yajfopavita. See ApDhS 2.2.4.21-22 and Kane, History of
Dharmasastra, 1, 1: 289-91. Pandey, Hindu Samskaras, p. 132, is incorrect, however,
when he writes that “None of the Grhyasttras contains the prescription of wearing the
Sacred Thread.” See SGS 2.2.3; BGS 2.5.7; BhGS 1.1; VGS 5.8.
63. AsvGS 1.19.10; SGS 2.1.2-5; BGS 2.5.16; HGS 1.1.4.7; BhGS 1.1; VaikhGS
2.4; PGS 2.5.17-19; GGS 2.10.9; JGS 1.12; ApDhS 1.1.3.3-6; BDhS 1.2.3.14; VasDhS
11.61-63; GautDhS 1.16.
64. AitB 7.23.
65. KB 4.11; TB 2.7.1.4, 2.7.3.3.
66. PB 17.11.8; SB 9.3.4.14; see also ApDhS 1.1.3.9 and BhGS 1.1.
67. SB 1.1.4.3, 3.2.1.8, 3.2.1.28, etc.
68. PB 7.5.7—10 and 12.4.24—26 associate this animal with cattle, which would sug-
gest a connection with the Vaisyas, not the Ksatriyas.
69. KGS 41.13.
Second, a later commentary on a different text” equates the spotted

deer and the tiger. The tiger (and, by inference, the spotted deer) is
indeed an animal with resemblances to the Ksatriya. It is said to be the
embodiment of courage,”! the king of the wild animals,” and the repre-
sentative of the ksatra power within the animal kingdom.”3 As for the
relationship between the agriculturalist/merchant—the Vaisya—and the
productive and valuable cow or the prolific goat, little more need be
said. Both the cowhide and the skin of the goat are symbols of “food,””4
and the goat is said to be the very form (rupa) of “increase” or pusti.7
Such systematic identifications between the components of nature
and those of society are also present in the case of the belt wrapped
around the boy at initiation. In general, this item of the uniform is meant
to protect the boy who wears it’ and to give him strength,” and it may
have also been a symbolic umbilical cord of the newly reborn student.’
Most texts agree that the Brahmin’s should be made of munya grass
(Saccharum Sara), that of the ksatriya from the bowstring or the murva
hemp from which such strings were made (Sanseviera Roxburghiana),
and the Vaisya’s from wool or hemp.” Mufja grass, it would seem, brings
out the Brahmin’s connection to the sacrifice, for in the ritual it is used in
various capacities (for brooms, cords, seats, purifiers, and, interestingly
enough, the belt tied around the body of the sacrificer’s wife during the
ritual). Mu/ja is quite flammable and was used in the sacrifice to make the
fire blaze up,® thus restoring “vigor” or “life sap” (arj) to the sacrifice.*!
The resemblance between the Ksatriya and the bowstring is obvious:
“The bow, truly, is the manly strength of the Ksatriya.”*? As for the
Vaisya, hemp appears to suggest proliferation and reproduction,® and
70. Raghavananda on Manu 2.41.

71 SB-129 2:8:
Tz; SB127-.1.8;
73. AitB 8.6.
74. E.g., SB 7.5.2.6, 7.5.2.43, 6.3.2.4.
95! 0B. 13.9%
76. GGS 2.10.37; PGS 2.2.8; MGS 1.22.7.
77. TS 6A.3.3—4.
78. See, e.g., HGS 1.1.4.4.
79. AgvGS 1.19.12; SGS 2.1.15-17; BGS 2.5.13; HGS 1.1.1.17; BhGS 1.1; VaikhGS
2.4: VGS 5.7; BDhS 1.2.3.13; Manu 2.42; VasDhS 11.5.8-10; GautDhS 1.15. See also PGS
2.5.21-23; GGS 2.10.10; JGS 1.12.
80. See, e.g., BhSS 11.3.16; MSS 6.1.3.23; ApSS 15.4.2.
81. SB 14.1.3.15-16; TB 3.8.1.1; TA 5.4.3-4.
82. $B 5.3.5.30. ;
83. It is connected to the amnion at SB 3.2.1.11 and to the chorion at SB 6.6.1.24.
wool clearly bonds the pastoralist to his domesticated animals. According

to one text, the Vaisya’s belt should be made of rope used for yoking oxen
to the plow, emphasizing the agricultural function of this class.
The staff, too, serves to create and reinforce class differentiation
and hierarchy in the boys undergoing the initiation samskara. It is to be
of varying lengths and different woods according to class: palasa
(Butea Frondosa) or bilva (Aegle Marmelos) for the Brahmin, nyagro-
dha (Ficus Indica) for the Ksatriya, and badara (Zizyphus Jujuba) or
udumbara (Ficus Glomerata) for the Vaisya.%* Again, these prescrip-
tions are not haphazard. In the Brahmanas, palaga is frequently identi-
fied with the brahman power*’ or with soma® and is said to instill lumi-
nosity (tejas) and the luster of the brahman power.*® According to the
Baudhayana Grhya Sutra,” the staff of paldsa is received by the student
from the teacher with the words “You are soma. Make me a soma-
drinker,” and that of bilva wood with “You are the luster of the brah-
man power. You are for the luster of the brahman.” The nyagrodha or
banyan tree, with its bending branches which take root in the ground,
was regarded as a resembling form of the Ksatriya:
The nyagrodha is the ksatra power of trees, and the Ksatriya is the ksatra
power [among humans], for the Ksatriya here dwells fastened, as it were,
to the kingdom, and supported [by it]. And the nyagrodha is fastened
to the ground, as it were, by its downward growths, and supported [by
it].9!
The staff made of this wood is taken by the Ksatriya initiate with a
mantra imparting physical vitality or ojas. The Vaisya receives the staff
of badara with “You are increase. Put increase in me,” and that of
udumbara with “You are life sap (aj). Put life sap in me.”%3 The associa-
84. ApDhS 1.1.2.37.

85. See AsvGS 1.19.13; GautDhS 1.26; VasDhS 11.55-57; Manu 2.146; BDhS
1.2.3.15. For the inverse, however, consult SGS 2.1.21-23.
86. SGS 2.1.18-20; BGS 2.5.17; BhGS 1.1; ApGS 4.11.16; HGS 1.1.1.17; VaikhGS
2.4; JGS 1.12; ApDhS 1.1.1.38; VasDhS 11.5254. See also ASvGS 1.9.13; VGS 5.27;
PGS 2.5.25-27; GGS 2.10.11; Manu 2.45; GautDhS 1.22-23.
87. SB 1.3.3.19, 2.6.2.8, 5.2.4.18, etc.
88. SB 3.3.4.10, 6.6.3.7.
89. AitB 2.1.
90. BGS 2.5.18-19.
91. AitB 7.31; see also SB 5.3.5.13.
92. BGS 2.5.20.
93. BGS 2.5.22-23.
tion of udumbara with life sap and food is well attested in the Vedic
textget
Other rites within the upanayana besides those already mentioned
also serve much the same purpose: to make socio-ontological distinc-
tions, infuse appropriate qualities, and draw out intrinsic potentialities.
According to Jaiminiya Grhya Sutra,% the lower garment is put on the
boy with mantras adjusted “according to class” (yathavarna): Brahmins
are dressed with a prayer to Soma for great learning, Ksatriyas with one
to Indra for great dominion, and Vaisyas with a mantra requesting Posa,
a deity of prosperity, to deliver his goods. These varying desirable quali-
ties are also mentioned in another rite, the entrusting of the boy to the
deities, in which the mantras are again appropriately modified according
to the class of the boy, and yet other examples could be adduced of this
same phenomenon within the initiation ritual.”
Finally, we note a most important rite within the upanayana, the
imparting of the sdvitri verse (RV 3.62.10) to the newly initiated pupil.
According to some texts, this mantra (“We contemplate the excellent
glory of the divine Savitr; may he inspire our intellect!”) was to be
recited in different meters by members of the different classes. Brah-
mins were to learn the verse in the gdyatri meter, Ksatriyas in the
tristubh, and Vaisyas in the jagati.*® Here again, the rules displayed in
the Sutras draw upon ancient correlations. The gdyatri was directly
connected to Brahmins,” or to the brahman power or luster of the
brahman,' or to Agni, the priest of the gods.'! The tristubh was
systematically associated with Ksatriyas,! physical power and manly
94. E.g., AitB 5.24, 7.22; PB 5.5.2, 6.4.1; KB 25.15, 27.6; TB 1.2.6.5, 1.3.8.2; SB
Bao les 3e92554..2)5
OS GSiel2.
96. See HGS 1.1.4.8.
97. See, e.g., SGS 2.2.11-15 and VaikhGS 2.4. For different begging formulas used
by students of different classes, see BGS 2.5.48-51; VaikhGS 2.8; VGS 5.28; PGS 2.5.2-4;
BDhS 1.2.3.17; ApDhS 1.1.3.28-30; Manu 2.49; VasDhS 1.1.68-70; GautDhS 2.36. For
different salutes adopted by different types of brahmacarins, see ApDhS 1.2.5.16.
98. SGS 2.5.4-7; MGS 1.22.13; PGS 2.3.7-9; BDhS 1.2.3.11. According to other
texts (VGS 5.26; KGS 41.20), wholly different Vedic verses were to be imparted to the
different classes, for which consult also R. K. Mookerji, Ancient Indian Education (Brah-
manical and Buddhist) (London: Macmillan, 1947), p. 182; and Kane, History of
Dharmasastra, I, 1: 302ff.
99. PB 6.1.6; TS 7.1.1.4.
100. KB 17.2, 17.9; PB 6.9.25, 12.1.2; SB 4.1.1.14.
101. AitB 8.6; KB 9.2; PB 6.1.6; TS 7.1.1.4.
1020 tsi 15 PB 6.1.8;
strength,!° and with Indra, king of the gods.!% The jagati was in like
manner consistently connected to VaiSyas in the ancient texts,!° or to
cattle,!% prosperity in general,!” the earth,!® plants, and the All-
gods, the Vaisya analogue in the heavenly world.!!°
Another analogy was created between the syllabic composition of
the meters (the gdyatri is a triplet made up of eight syllables each, the
tristubh a quartet of eleven syllables per verse, and the jagati a quartet of
twelve syllables each) and both the respective ideal ages for initiation of
the classes (eight, eleven, and twelve) and the difference between these
ages and the last ages possible for performance of the upanayana for
each class (also eight, eleven, and twelve). Hermann Oldenberg sums up
this rather complicated set of resemblances:
The number of years given for the Upanayana of persons of the three
castes (Brahmanas 8-16, Ksatriyas 11-22, Vaisyas 12—24) is evidently
derived from the number of syllables of the three metres which are so very
frequently stated to correspond to the three castes, to the three gods or
categories of gods (Agni, Indra, Visve devas), etc., viz. the Gayatri, the
Trishtubh, and the Jagati.
“This is a very curious example,” Oldenberg goes on to opine, “showing

how in India phantastical speculations like those regarding the mystical
qualities of the metres, were strong enough to influence the customs and
institutions of real life.”!!! To put the case a bit less pejoratively, it was
indeed the production of a “real” life for the initiates that the ritualists
aimed for. And to do so, in the Vedic world, was to move the candidate
from inborn, biological imperfection to a ritually constructed—that is, a
ritually connected—identity.
The upanayana delivers a reborn Aryan, who was supposedly more
really human because more fully realized, activated through the activity
of ritual. The ritual of initiation was thus the first of many such rites
designed to draw hierarchical distinctions among ritualists of different
103. AitB 4.3; KB 11.2, 16.1.

104. AitB 8.6; KB 14.3, 30.2; PB 6.1.8; TS 7.1.1.5.
105.7S.7-1:1;553 PB 6:1.00.
106. AitB 3.18, 3.25; KB 16.2, 17.2; PB 6.1.10, 18.11.9; SB 8.3.3.3, 12.8.3.13;
tRoR ESESS
107. SB 3.4.1.13.
108. SB 1.8.2.11, 6.2.1.29, ete.
109. SB 1.8.2.11.
110. AitB 8.6; PB 6.1.10; TS 7.1.1.5.
111. Hermann Oldenberg, The Grihya Sutras, Vol. 1 (Oxford: Oxford University
Press, 1886; reprint ed. Delhi: Motilal Banarsidass, 1964), commenting on §GS 2.1.1.
social classes, and to constitute Vedic society as a whole made up of

unequal, ritually constituted groups.
The completion of the upanayana ritual, of the period of Veda study
and tutelage under the teacher which it inaugurates, together with the
performance of the equally crucial marriage samskara, qualifies the adult
householder to set his own ritual fire (or fires) and perform the Vedic
sacrifices. These more complex rituals continue the ontological develop-
ment and refinement of his being. If the function of the samskaras was, as
one text declares, to make fit the physical body,!? one of the principal
purposes of all Vedic rituals was to make fit the dtman, the self (inclusive
of the body, soul, and socio-ontological identity). The construction and
perfecting of the dtman in the sacrifice is often described in the
Brahmanas with the same combination of the verb kr- and the prefix sam-
that gives us the word samskara.
In the Aitareya Brahmana, for example, the soma sacrifice is de-
picted as a ritual of “self-perfection” (atma-samskrti) and is regarded as
a “work of art” (Silpa).! The sacrificer “fashions (samskurute) his own
self in a rhythmical way” through the ritual chants that are likened to
seeds being poured into the womb.!* Again, in another Brahmana, we
read that in the ritual the priests construct and perfect (samskurvanti) a
“heavenly self” (daiva adtman) for the sacrificer,'5 and in yet another
text the relationship between the constructive ritual activity and the
atman thus constructed is recognized by the sacrificer, who knows,
“This, my [new, heavenly] body is made perfect (samskriyate) by that
[sacrifice]; this body of mine is procured by that [ritual].”!"6
The body of Prajapati, the Cosmic Man, was joined together, reinte-
grated, and healed by the ritual acts of the gods. So also is the dtman of
the human made whole and perfected in his sacrificial activity. Each
sacrifice expresses and regenerates the sacrificer’s dtman, and the ritual
life of the individual becomes like a canvas on which is painted the
picture of his self. The portrait is complete in itself at any particular
112. VaikhGS 1.1.

113. See B. M. Barua’s comments on this passage in his “Art as Defined in the
Brahmanas,” Indian Culture 1 (July 1934—April 1935): 120: “The substance with potentiali-
ties or possible forms is given as a work of Art Divine and the methodical realization of
those possibilities is the achievement of human skill and intelligence. . . . Art consists in
the intelligent working up of a desired form on a natural material, making manifest what is
hidden or potential.”
114. AitB 6.27.
115. KB 3.8.
116. SB 11.2.6.13.
sacrificial performance and is filled out over time. It is at once an index

and a projection, a gauge of the relative realization of the self in this
world and in the other world beyond death. Sacrifice constitutes being,
on both the cosmic and human planes, through a process that is equally
one of construction and of discovery, of making and of finding. As J. F.
Staal writes,
The sacrifice can now be interpreted as one of the modes of human being
which constitutes being. This ontological interpretation enables us to see
how it was possible (ontically, as Heidegger would say) that such impor-
tance was attached to the ritual act. . . . The transformation or consecra-
tion which is effectuated through sacrifice, is not a transformation from
one being to another but the constitution of being itself.!!”
In many of the Brahmanas, the second birth of the individual is not
exclusively the upanayana but sacrifice in general.'!8 For “Unborn,
truly, is [the natural] man. He is really born through the sacrifice.”!9
Several texts state that “man is born three times”:
First he is born from his mother and father; then, when he to whom the
sacrifice inclines sacrifices, he is born a second time; and when he dies,
when they put him into the [cremation] fire and he arises again, he is born
a third time. Thus, they say, “A man is born three times.”!?°
Sacrifice, as has often been noted by scholars of the Vedic ritual, is
represented as a cycle, a series of births predicted on a series of prior
deaths.'*!
“Man depends only on his own ritual work, his own karman,” as Jan
Heesterman puts it.!?? Humans are responsible for creating an ontologi-
cally viable self for themselves and the “world” (loka) for that self to
117. J. F. Staal, Advaita and Neoplatonism: A Critical Study in Comparative Philoso-

phy (Madras: University of Madras, 1961), p. 67.
118. For the birth symbolism surrounding the consecration rite for a soma sacrifice,
the diksa (a rite which reduplicates at a higher level the initiation or upananyana), see
Kaelber, “The ‘Dramatic’ Element in Brahmanic Initiation: Symbols of Death, Danger,
and Difficult Passage,” History of Religions 18 (August 1978): 54-76”; Lévi, La Doctrine
du sacrifice, pp. 102—7; and esp. Gonda, Change and Continuity, pp. 315-462.
119. MS 3.6.7; see also JUB 3.14.8.
120. SB 11.2.1.1; see also AitA 2.5; JUB 3.11.3.
121. See Lévi, La Doctrine du sacrifice, p. 81; J. C. Heesterman, The Ancient Indian
Royal Consecration: The Rajastya according to the Yajus Texts and Annoted [sic] (‘s
Gravenhage: Mouton, 1957), p. 6; and Kaelber, “The ‘Dramatic’ Element in Brahmanic
Initiation.”
122. J. C. Heesterman, “Brahmin, Ritual, and Renouncer,” p. 34.
inhabit. And both self and world are ritual constructs. “Man is born into a
world made by himself,”!?3 a ritual world where new birth can occur. But
this homemade world is also a Joka in the sense of a status constituted by
the rituals one has performed. “Born out of the sacrifice, the sacrificer
frees himself from death. The sacrifice becomes his self,” 24 for man is the
sacrifice (puruso vai yajnah).'2> The dimensions of the sacrificer’s self or
being are correlative to the rituals he sponsors and participates in. Ritual
is thus regarded as the true mother and accurate measure of being.
The atman, in Vedic texts before the monistic Upanishads, repre-
sents the ritually constructed conjunction of innate and acquired poten-
tial, character, personality, and individuality: a social and ontological
persona. The daiva dtman, or “heavenly self,” to which we will return
shortly, refers to the ritually constructed extracorporeal self projected
into another ontological sphere, into a Joka other than the earthly one.
And loka here means “world” in the sense of an ontological condition or
status. A loka, etymologically related to the English locus, is where the
atman is placed. As Coomaraswamy observes, “The whole world or
universe . . . corresponds to the ensemble of all possibilities of manifes-
tation, whether informal, formal, or sensible; a world (loka = locus) is a
given ensemble of possibilities, a given modality.”!° Or as Gonda writes
in Loka: The World and Heaven in the Veda,
If “name-and-form” [ndma-ripa] means the psychical and physical compo-
nents of any personality, the term Joka may denote the “place” or “room”
occupied by these, wherever a personality may be found, the position (of
safety) or situation in which it may be.!?7 |
The sacrifice is the workshop in which the sacrificer’s worldly self

and status are forged. But it is also said to be generative of an impercepti-
ble “divine” self and world for the sacrificer. The svarga loka or “heav-
enly world” was, continues Gonda, “something or some state which
could be, or had to be, produced by the effects of the ritual acts of the
sacrificer.” And, further, “The condition which for convenience may be
calied ‘immortality’ belongs to the person concerned already in his
earthly existence, before his removal to the svarga loka.”'8 Heaven,
123. SB 6.2.2.7.
124.SB:11.2.2,5,
125. SB 1.3.2.1 and elsewhere.
126. Coomaraswamy, “Vedic Exemplarism,” p. 45.
127. Jan Gonda, Loka: The World and Heaven in the Veda (Amsterdam: N. V.
Noord-Hollandsch Uitgeuers Maatschappji, 1966), pp. 32-33.
128. Ibid., p. 97.
won by ritual action, is said to be “in this world”: “By that which is not
svarga,” that is to say, by the ritually created conditions in this world,
“one ascends to the svarga loka.”'”° It is to this ritually induced—and
ritually previewed—ascension to an ontological status beyond this world
that we now turn.
The Sacrificial Journey to Heaven
That the Vedic ritual should be thought to have a transformative effect

on those who undergo it is not entirely surprising. Participants in rituals
the world over claim to be changed by them; they are consecrated, made
sacred, by virtue of the power of ritual activity.
In Vedic India, however, this ritually infused sacrality is often ex-
pressed in terms of divinity, and the idiom can provoke confusion. What
did the ritualists mean when they claimed to be gods? Did they really
mean to imply an ontological identity between certain men and the gods
beyond?
There are two kinds of gods. For the gods [are gods], and the Brahmins
who have studied and teach are human gods (mdnusya devas). The sacri-
fice of these [sacrificers] is divided into two. Oblations [are sacrifices] to
the gods and sacrificial fees [are sacrifices] to the human gods, the Brah-
mins who have studied and teach. With oblations one pleases the gods;
with sacrificial fees one pleases the human gods, the Brahmins who have
studied and teach. Both kinds of gods, when pleased, place him in a
condition of well-being. !*°
Not only did the Brahmin officiants claim divinity, with its eco-
nomic advantages, among others; in the course of the ritual the sacrificer
(regardless of class) was also said to attain a godly state. He was trans-
formed in preliminary rites from an ordinary man into a divine entity.
“He passes from the world of men to the world of the gods,”"3! and, as
Henri Hubert and Marcel Mauss have written of the diksita (one who
has been consecrated for a soma sacrifice), “once his divine nature has
been proclaimed, it confers upon him the rights and imposes upon him
the duties of a god.”!?
129. AitB 7.10.

130. SB 2.2.2.6.
131. SB 1.1.1.4.
132. Henri Hubert and Marcel Mauss, Sacrifice: Its Nature and Function trans. by
W. D. Halls (Chicago: University of Chicago Press, 1964), p. 21. See also Lévi’s comment
that ‘homme se fait surhumain.” La Doctrine du sacrifice, p. 9.
Furthermore, the attainment of the divine self and heavenly world—

we have observed that it is one of the ontological purposes of the Vedic
ritual—was claimed to be realized not only after the death of the sacrificer
but also in the course of the ritual itself. In Vedic ritualism, it would seem,
man became a god and entered heaven within the confines of the sacrifice,
that is, in this very life.
The ritual creation of the daiva atman and svarga loka was part of a
transformative process that the texts liken to a journey, with the sacri-
fice itself the vehicle. The sacrificial journey corresponds to the ritual
procedure which Hubert and Mauss have described as a curve. It begins
with a series of consecration rites preparatory for, and instigating of, the
“ascent to heaven”; it reaches a climax coincident with the offering of
the principal oblation and equated with the “arrival” at the heavenly
world; and then begins a “descent,” a set of exit rites. This curve is
followed both by the sacrificer and by the object or objects of the sacri-
fice, the two being linked by connections or bandhus:
The religious condition of the sacrifier thus also describes a curve symmet-
rical to the one traced by the victim. He begins by rising progressively into
the religious sphere, and attains a culminating point, whence he descends
again into the profane. So each one of the creatures and objects that play a
part in the sacrifice is drawn along as if in a continuous movement which,
from entry to exit, proceeds along two opposing slopes.!*
In order to reach the world of heaven, the sacrificer is sometimes
said to take the form of a bird, the ritual chants serving as wings.'34 More
often, however, it is the ritual as a whole, the yajna, that acts as the
vehicle carrying the sacrificer to yonder world. In one text, it is the
sacrifice itself—and not the sacrificer—which is declared to be a bird.'5
The yajna is also likened to a cart,!*° the sacrificial spoons are said to be
“yvokemates,”!3’7 and the meters of the ritual utterances are equated to
harnessed cattle that draw the sacrifice to the world of the gods.18
Another image the Brahmanas frequently call upon is that of the
133. Hubert and Mauss, Sacrifice, p. 48. See also Lévi’s description of the sacrificial
journey, Le Doctrine du sacrifice, pp. 130-31: “Les deux temps de l’opération correspon-
dent aux deux movements du sacrifiant: ascension au ciel et retour sur la terre. Le
sacrifiant monte au ciel pour s’assurer un corps divin et immortel; en retour, il fait aban-
don aux dieux de son corps humain et périssable. Puis, sa place marquée et retenue au ciel,
il aspire a redescendre et rachete le corps qui il avait sacrifié.”
134. PB 5.1.10, 5.3.5; AitB 3.25.
135. SB 4.1.2.25.
136. SB 3.9.3.3.
137. SB 1.8.3.27.
138. SB 4.4.3.1.
sacrifice as a ship, transporting the sacrificer from the earthly to the

heavenly shores. The priests are its spars and oars, “the conveyances to
the heavenly world. If there is one blameworthy [priest], even that one
only would make it sink. And truly, every sacrifice is a heaven-bound
ship. Therefore one ought to keep a blameworthy [priest] away from
every sacrifice.”3° Other elements of the ritual are also assimilated to
the parts of the ship: “The Agnihotra is a ship bound for heaven. The
ahavaniya fire and the garhapatya fire are the two sides of that heaven-
bound ship, the Agnihotra. The captain of the ship is the priest who
offers the milk oblation.”!°
But perhaps the most frequently encountered is the comparison of
the sacrifice and the chariot, the vehicle of choice in Vedic India. The
introductory and concluding rites are likened to the two sides of the
chariot and should be symmetrical: “He who makes them equal to one
another safely reaches the world of heaven, just as one takes any desired
journey by driving a chariot with two sides.”!*! Similarly, another text
argues that the Agnihotra should be performed after sunrise so that it
will be like a chariot with both wheels: “Day and night are the wheels of
the year; truly, with them he goes through the year. If he offers before
sunrise, it is as if one were to go with [a chariot with] one wheel. But if
he offers after sunrise, it is as if one were swiftly to make a journey with
[a chariot having] both wheels.”'? In other texts, other components of
the ritual are connected to the parts of the chariot—the sacrificial fees
are the internal fastenings;!** the chants are the reins;!“ and the recita-
tions are said to be the “inner reins.”!45
The sacrifice must be a study, complete, and safely operated “vehi-
cle,” because the journey to heaven is no sure thing. It involves, the texts
assure us, great difficulty and danger. The trip to the heavenly world
within the sacrifice is designated a “difficult ascension” (durohana), and
the reasons for the difficulty begin to enlighten us about the recognized
differences standing behind the proclaimed identity between gods and
men.
139. SB 4.2.5.10.
140. SB 2.3.3.15; see also JB 1.166. In an interesting reversal, the Mundaka
Upanished (1.2.7), as part of a critique of ritualism in light of the new emphasis on
mystical knowledge alone, declares sacrifices to be “leaky vessels,” unfit for the true
voyage—the attainment of liberation from karma.
141. KB 7.7.
142. AitB 5.30.
143. PB 16.1.13.
144. PB 8.5.16.
145. AitB 2.37; see also JB 1.129-30 for the most detailed of these metaphors.
In one sense, the difficulty with the ritual ascension to heaven is

this: in order to ascend successfully, human ritualists are called upon to
follow the steps of the gods. These inhabitants of the heavenly world
also arrived at their destination, in the beginning, through ritual tech-
niques: “It was done thus by the gods. So it is done by men.”!46 Certain
rites and ritual formulas have such elevating power, guaranteed by their
prior ability to produce heaven for the gods: “By means of this the gods
went to the world of heaven; one who desires the world of heaven
should use it for reaching the world of heaven.”!47
The fundamental differences between gods and men, however,
make it difficult indeed for the latter to replicate the techniques of the
former. Indeed, the texts declare, the human condition is the exact
opposite of the divine. In contrast to the “perfection” (samrddha or
sampanna) characteristic of divine beings, the earthly and human is said
to be imperfect and “unsuccessful” (vyrddha).'48 “What is ‘no’ for the
gods is ‘yes’ for man,” says another Brahmana™? by way of emphasizing
the utter difference between the two kinds of beings. In several passages
we read that satya, here in the sense of ritual exactitude, is a quality of
the gods, whereas anrta (“error” or “disorder”) is the distinguishing
mark of things human.!°
Although in the consecration rites the human sacrificer is said
to acquire a godlike status within the ritual, the fact that the Brah-
manas call the apotheosis a “difficult ascension” implies that the au-
thors had some doubts that man could ever entirely shed his humanity
in this life and truly assume the ontological dimensions of his divine
prototypes. The perfection of the gods, acquired by and realized in
their ability to perform the ritual without error, was for humans an
ideal rather than a realistic goal. There were simply too many things
that could go wrong in the course of the performance. The sacrifice—
like all created things and beings—was always inclined to fall apart, be
cut, break, shatter, become defiled, inverted, or defective, as one list
enumerates the possibilities.!>!
Thus, one aspect of the difficulty of the “difficult ascension” is
precisely the demand for an inhuman exactitude and perfection. Walter
Kaelber writes in his study of the duérohana:
146. TB 1.5.9.4 and elsewhere.

147. PB 2.6.2; see also PB 2.12.2, 2.15.2, 3.2.2, etc.
148. SB 1.4.1.35.
149. AitB 3.5.
150. $B 1.1.2.17, 3.3.2.2, 3.9.4.1.
151. BSS 28.10.
In fact, the intricacies of the ritual constitute the greatest source of danger
for the sacrificer. No detail may be overlooked, no act incorrectly per-
formed. . . . Precisely because these ritual intricacies characterize the en-
tire sacrificial scenario, there can be little question that it constitutes a
difficult passage in every sense.!*?
Even more disturbing, perhaps, is the fact that one can never be certain
that ritual perfection—the timely and correct performance of every min-
ute detail—has ever been achieved. One text nicely sums up the angst
and uncertainties that the ritualists might have felt as they undertook
their heaven-bound project: “The world of the gods is concealed from
the world of men. ‘It is not easy to depart from this world,’ they say. ‘For
who knows if he is in the yonder world or not?’ ”153
The ascent to heaven is not only difficult; it is also perilous. One
must “fall away from this world” and “arrive” at the yonder world,
where the sacrifice “has its only [true] foundation, its one [true] end.”!4
But the road to heaven is, in Vedism, not always envisioned as paved
with gold. “Dangerous indeed are the paths that lie between heaven and
earth”;> for on either side of these roadways are eternally burning
flames which “scorch him who deserves to be scorched and allow him to
pass who deserves to pass.”'*¢ The sacrifice, the vehicle transporting one
to the heavenly world, is said to be “razor sharp,” and success or failure
is unequivocal: “Suddenly he becomes full of merit or perishes.”5” An-
other text refers to the “wilds and abysses of the sacrifice” and warns, “If
any venture into them not knowing the ropes, then hunger and thirst,
evil doers and fiends, harass them, even as fiends would harass foolish
men wandering in a wild forest.”158
Nor do the dangers cease when the journey ends. Although the
152. Kaelber, “The ‘Dramatic’ Element in Brahamanic Initiation,” p. 63. See also
Lévi, La Doctrine du sacrifice, pp. 123-24: “Dans ce dédale de prescriptions minutieuses
erreur est aisée et les conséquences en sont terribles. Le danger est partout qui guette le
sacrifiant.”
153. TS 6.1.1.1. In other texts, there is a rather specific, if variable, notion about
where heaven is, if not a certainty about when one knows one has arrived there. The world of
heaven takes about six months to get to (PB 4.6.17), or is some forty days’ journey on horse-
back upward from earth, or is at a distance equivalent to the length of the Sarasvati river (PB
25.10.16). Alternatively, “The world of heaven is as far away from this world as a thousand
cows standing on top of each other” (PB 16.8.6), or “the distance [may be] also a thousand
yojanas [about nine thousand miles], or a thousand days’ journey by horse” (PB 21.1.9).
154. SB 8.7.4.6.
155. SB 2.3.4.37.
156. SB 1.9.3.2.
157. TS 2.5.5.6; see also Katha Upanishad 3.14.
158. SB 1.9.3.2.
sacrificial trip to heaven is described as a difficult ascension, the perils to

the sacrificer do not evaporate when he has successfully been elevated to
that “world” in the ritual which coincides with the principal oblation.
Rites and mantras were designed not only to ensure the ascension to the
heavenly world but also to establish the sacrifice and the sacrificer in
that world in such a way that they do not “slip” or “fall” from it, for such
would be disastrous. 15°
Paradoxically, the very success of the sacrifice, the winning of the
heavenly world at the summit of the sacrificial journey, places the sacri-
ficer in jeopardy. For the svarga loka is a realm which man cannot, in his
present mortal state, survive. A myth of origins tells why this is so.
When the sacrifice had proved effective as a conduit to the other world,
“Death spoke to the gods. ‘Now surely all men will be immortal. What
will be my share?’ They said, ‘From now on, no one will become immor-
tal with the physical body. Only when you have taken that as your share
will he become immortal . . .’.”!© Because of this arrangement with
death, only the disembodied win immortality and heaven on a perma-
nent basis. Others merely rent them. The sacrificial journey places the
sacrificer in the other world only temporarily, just long enough to mark
out and reserve a space for the next life.
The ritual journey to heaven taken by human sacrificers, in other
words, must be a round trip. For if the sacrificer wholly repeats the
paradigmatic voyage of the gods and arrives at the heavenly world with-
out descending again from it, he would, the texts postulate, die. Death
would take its share, the mortal body of the sacrificer.
The gods “reached these worlds” through the sacrifice, and “having
reached these worlds by means of this, they finished the sacrificial ses-
sion” ;!*! that is to say, they did not descend again from the svarga loka in
the declining half of the sacrificial curve. The human sacrificer, however,
does not conclude the ritual at its highest point. Rather, the climax of the
ritual marks the start of the second portion of the procedure, a descent
from the world of heaven so laboriously won. “Having obtained the world
of heaven, the sacrificers establish themselves in this world.”!* Heaven,
having been attained, is quickly renounced.!
159) See, e.2., PB 2.6.2; KB 8.2.

160. SB 10.4.3.9.
161. PB 15.11.5.
162. AitB 4.21.
163. See Lévi’s comment in La Doctrine du sacrifice, p. 88: “Le mécanisme du
sacrifice est clairement répresenté par le rite du darohana, ‘lascension difficile.’ Il se
résume en deux periodes, I‘une ascendante, |’autre descendante. II s’agit d’élevar d’abord
le sacrifiant au monde céleste; mais la terre a ses charmes, et le sacrifiant ne demande pas a
The Brahmanas mince no words in their warnings to those who

spurn a two-way ticket to—and from—paradise, those who do not use
the return portion of the booking. The sacrificer must descend from the
world of heaven or “fall away from this world” permanently; he who
does not leave heaven would “go to the farthest distance.”! Those with
hubris, who, like the gods, perform the ritual “in the forward direction
only,” “may win the world of heaven, but they will not have long to live
in this world.”! They will “vanish from this world”!*? or “go to
Prajapati,”!©° a euphemism for death.
The dangers of remaining in heaven seemed to have been particu-
larly acute for the royal sacrificer, a king who has responsibilities for
ruling over portions of this world. In the Rajasuya sacrifice, the king
goes to the “world of heaven” but must return to earth, which is his
“firm foundation”:
In that he is consecrated by the Rajastiya, he ascends to the world of
heaven. If he did not descend to this world [however], he would either
depart to a region which lies beyond human beings, or he would go mad.
That ritual of shaving the hair (the kesavapaniya) with reversed chants is
for not leaving this world. Just as he would descend [from a tree], grabbing
branch after branch, so does he descend to this world with this [rite]. [The
rite is] for attaining a firm foundation.!®
The sacrificial prototype—the heaven-procuring ritual of the gods—
is thus not ordinarily to be exactly repeated by men any more than the
perfection of the gods’ rituals can be literally realized by the human
sacrificer. Just as perfection in the sense of ritual exactitude should be
understood both in its transcendent, prototypical form (that of the
gods) and in its relative manifestation or counterpart form (that of
humans), so, too, must the perfection represented as the heavenly
world be analyzed into its ideal and its humanly attainable forms. The
full realization of perfection, in both cases, is not possible for humans
in this life; hence Lévi’s comment that “the true voyage to heaven is
not accomplished until after death” and that “the only authentic sacri-
la quitter trop tot. Assuré de l’immortalité a venir par la premiére opération, il reprend par
la seconde opération sa place entre le vivants.”
164. KB 7.9.
LHS aeBe 5.7.28
166. AitB 4.2.
167. PB 4.3.5—6, 6.8.17-18.
168. PB 4.8.9.
169. PB 18.10.10; see also AitB 4.21; PB 5.5.4—-5.
fice would be suicide.”!”? And, as the Satapatha Brahmana cautions,

the premature death of one who is overeager to reside in the world of
heaven might prove to be counterproductive: “One ought not to yield
to his own desire and pass away before [he has attained] the full extent
of life, for such does not make for the heavenly world.”!7!
The svarga loka reached in the course of the ritual process is, then,
not only difficult and dangerous to get to; it is also only a model of the
prototypical world of heaven in which the real gods dwell. The ritually
created ersatz heaven is, in other words, but a counterpart to the tran-
scendent and paradigmatic heaven of the gods strictu sensu. The gods’
world, as Jean-Marie Verpoorten has observed, is “where reigns a unity
that has nothing of the constructed aggregate, but is that which is en-
tirely indivisible”; it is characterized by its unconstructed perfection or
samrddha. The ritually produced heaven of the human sacrificers, on
the other hand, is marked by samana, the “constructed unity of parts,”
which is “the configuration (chiffre) of man and his ritual activity.”1”
The gods live in the perfect, the whole, and the original, but man can
only fabricate—through ritual labor—an inferior substitute.
If the heavenly world of the ritual and the divine status acquired by
the sacrificer are not the true equivalents of those possessed by the gods,
perhaps it is the case that such are attained by humans after death.
Perhaps the daiva adtman the sacrificer makes for himself through ritual
action is indeed a full-fledged divinity for those who shed the body,
giving death its share, and take the final, one-way trip to the world of
170. Lévi, La Doctrine du sacrifice, pp. 93, 133. While, in general, Vedic texts
forbid suicide, there does seem to be at least one exception. As an example of a ritual
performed “in the forward direction only,” there is the sacrifice called the sarvasvara,
described in several Brahmanas and Srauta Sutras. It is apparently meant for an old man,
probably an accomplished ritualist, “who is desirous of having an end to his life” (KSS
22.6.1) or wishes to “go to yonder world without suffering from any disease” (PB 17.12.1;
JB 2.167). When various chants designed to make him “go to the endless, to yonder
world,” or “from this world to the world of heaven” (PB 17.12.3—4), are completed, the
sacrificer lies down between the fires, with his head covered and pointing to the south,
while another chant is recited over him. “And he dies at that time. . . . The rites of the
sarvasvara come to an end as soon as the sacrificer dies, as there is no purpose of the
sacrifice left to achieve” (KSS 22.6.6-8). Another text instructs regarding what one should
do if, perchance, death does not arrive at that time: “If he lives, he should perform the
final oblation of the soma sacrifice and thereupon seek his death by starvation” (LSS
8.8.40). For a treatment of this ritual and others, see J. C. Heesterman, “Self-Sacrifice in
Vedic Ritual,” in S. Shaked et al., eds., Gilgul: Essays on Transformation, Revolution and
Permanence in the History of Religions (Leiden: E. J. Brill, 1987), pp. 91-106.
171. SB 10.2.6.7.
172. Verpoorten, “Unité et distinction,” pp. 76, 84.
heaven, the world of the gods. Can men become gods after death, if not
truly in this life? And is the world of heaven awaiting the ritualist at the
end of his life the same world occupied by the Vedic deities?
Life after Death
The gods are in permanent possession of divinity and an immortal world

in which to exercise it. Man, on the other hand, must work for his
identity and status. He is dependent on his own ritual activity or karman
in its original sense; he is “born into a world made by himself.” Man’s
divinity, his daiva atman, is also forged by the ontological (and transcen-
dental) power of ritual work, and so, too, is his heavenly world—both
the heaven reached in every successful sacrificial performance and the
heaven to which the sacrificer ascends and in which he dwells after
death.
Furthermore, the conditions of this life—one’s ritually constructed
socio-ontological identity and status—are accurate previews of those of
the next. Both the self and the world that one builds for oneself here and
those that one constructs for the afterlife are products of the rituals one
performs. And they are, one might say, resembling counterparts of each
other.
“Yonder world is the corresponding form (anuriipa) of this world,”
declares one text, and “this world is the corresponding form of that.”!”3
Because these two /okas “correspond,” the heavenly world is presaged
in this one. In this world, as Paul Mus puts it, there is “the anticipated
image, or the real base, of the other world and of the immortality in that
beyond.”!”* Because the svarga loka that men enter with their daiva
atmans is, like more mundane /okas, ritually made, sacrificers with differ-
ent ritual résumés realize different heavenly worlds and different divine
selves.
In this universe of resemblances and interlocking forms and coun-
terforms, the complex of self/sacrifice/status (a@tman/yajna/loka) is pro-
jected simultaneously on different ontological planes. The self that is
constructed and represented in the sacrifice is a construction/representa-
tion of the self in this world (the sacrificer’s socially recognized per-
sona) and the self of the other world (the sacrificer’s daiva Gtman). In
like manner, the sacrificer’s world is at once the world composed of his
173. AitB 8.2.

174. Paul Mus, Barabadur, pp. 135-36.
sacrifices, the social status he has attained through those sacrifices, and
the individually designed heavenly world—reached temporarily at ev-
ery sacrifice and attained for good after death.
Both loka and atman, then, are particularized concepts in Vedic
thought. They are intimately linked to the particular sacrificer who fabri-
cates them in his ritual activity. Put otherwise, the “heavenly world” and
the “divine self” are not, in this framework of belief and action, unitary
concepts at all. They are, rather, tailored to individuals and hierarchi-
cally gauged.
{ The hymns of the Rg Veda often seem to reflect a belief in a future
‘life in the same heaven as that inhabited by the gods.'75 But certainly in
the classical texts of Vedic ritualism, the concept of svarga loka becomes
multiple and individually shaped. A cynical view of this innovation is
taken by A. B. Keith, who correctly notes that in post-Rgvedic texts there
are “diverse degrees of good acquired by different modes of sacrifice”:
It was obviously necessary to admit that every sacrificer would receive
reward by admission to the happiness of the world to come, but the
Brahmans had to consider the claims of the richer of the patrons, and had
to promise them more in the world to come than the poorer, who offered
and gave less. !76
A somewhat less jaded view of this shift from an egalitarian reward sys-
tem to a hierarchical one lies behind Paul Deussen’s summary of the way
heaven is calibrated to match the achievements of individual sacrificers:
The chief aim of the Brahmanas is to prescribe the acts of ritual, and to
offer for their accomplishment a manifold reward, and at the same time
sufferings and punishment for their omission. While they defer rewards as
well as punishments partly to the other world, in place of the ancient
Vedic conception of an indiscriminate felicity of the pious, the idea of
recompense is formulated, involving the necessity of setting before the
departed different degrees of compensation in the other world proportion-
ate to their knowledge and actions.!””
In Vedic ritualism, one obtains a “good” Joka through ritual activ-
ity, and a specifically contoured hell, a naraka loka, through misdeeds or
failure to perform good acts. Those who spit on a Brahmin, or flick the
175. The fullest depiction of the Rgvedic vision of heaven occurs at RV 9.113.7-11,
see also RV 10.14.10-12.
176. A. B. Keith, The Religion and Philosophy of the Veda and Upanishads, 2 vols.
(Cambridge, Mass.: Harvard University Press, 1925), p. 572.
177. Deussen, The Philosophy of the Upanishads, p. 324.
mucus of their nose on him, will spend their afterlife sitting in a stream
of blood, devouring their hair for food.! Those who consume food in
this world without first sacrificing some of it to the gods will enter one of
a variety of hells: “For whatever food a man eats in this world, by the
very same is he eaten again in the other.”!” A person who reviles,
strikes, and draws blood from a Brahmin “will not see the world of the
ancestors for as many years as are the grains of dust on which the
[Brahmin’s] blood falls.”!8° And in the myth of Bhrgu’s journey to vari-
ous hells,'*! the afterlife /okas are described in which the punishment
exactly fits the crime. Those who in this life cut down trees without
sacrificing are eaten by those trees after death in one Joka. In another,
those who cook animals for themselves without sacrificing are consumed
by those animals; and in yet another hell, unsacrificed rice and barley
feed on the unheedful.
At death, a kind of judgment is envisioned. Good deeds (by which
is nearly always meant ritual deeds) and bad ones are separated out and
weighed up:
For in yonder world they place him on the scale, and whichever of the two
rises that he will follow, whether it be the good or the evil. And whoever
knows this gets on the scale even in this world, and escapes being put on
the scale in yonder world. For his good action rises, not his bad action.!*?
From another text we learn that “the good that man does during his life
passes into his breaths, the bad into his body. When the one who knows
thus departs from this world, his good deeds rise up together with his
breath and his wrong deeds are left with his body.”!*} Elsewhere in that
Brahmana we read that the breath ascends to the heavenly world and
“announces to the gods the quantity: “So much good, so much evil has
been done by him,’ ”!* this evidently determining the kind of /oka the
departed soul will henceforth occupy.
In the early Upanishads, too, the /okas comprising the universe—
178. AV 5.19.3.
179. SB 12.9.1.1; see also KB 11.3.
180. TS 2.6.10.2.
181. JB 1.42-44 and SB 11.6.1.1-13. For a translation and analysis of the JB variant
of the myth, see Wendy Doniger O’Flaherty, Tales of Sex and Violence: Folklore, Sacri-
fice, and Danger in the Jaiminiya Brahmana (Chicago: University of Chicago Press, 1985),
pp. 32-40.
182. SB 11.2.7.33.
183. JB 1.15.
184. JB 1.18.
the possible ontological and soteriological spaces and situations—are

many. There is, first of all, a distinction made between the deva loka,
the world of the gods, and the pitr loka, or world of the ancestors.'85 But
we also learn of a wide variety of “worlds”: those of mothers, brothers,
sisters, friends, scents and garlands, food and drink, song and music,
and women.'* Eisewhere is provided a map of the ontological hierarchy
of the Jokas and their inhabitants:
If one is fortunate among men, wealthy, a lord over others, well provided
with all human enjoyments, that is the most perfect bliss of men. Now one
hundred times the bliss of men is one bliss of those who have won the
world of the ancestors. One hundred times the bliss of those who have
won the world of the ancestors is one bliss in the world of the demigods.
One hundred times the bliss in the world of the demigods is one bliss of the
gods created by ritual work (karma devas). One hundred times the bliss of
the gods created by ritual work is one bliss of the gods by birth (ajana
devas). . . .One hundred times the bliss of the gods by birth is one bliss in
the world of Prajapati. . . . One hundred times the bliss in the world of
Prajapati is one bliss in the world of Brahma. . . . This is truly the highest
world.!87
Among other points of interest in this text is the clear differentia-
tion between the “world” and “bliss” of the gods created by their own
ritual labors (pulled up by their own bootstraps, so to speak) and those
of the gods by birth. Here we have corroborating evidence for the posi-
tion argued above: there is an unequivocal distinction in Vedic ritualism
between divinity constructed and divinity eternally possessed. There is
no confusion between prototypical deities and the counterpart divinity
available to men who make their own divine self through sacrifice.
Man is indeed born into a world made by himself, in the afterlife as
well as in this one. The particular Joka or place a person inhabits in the
185. Already in the RV (e.g., 10.18.1, 10.2.7, 10.15.1—2), the path leading to the
world of the ancestors is distinguished from that leading to the deva loka. In the SB, the
gate to the world of the gods is said to be in the northeast (6.6.2.4), whereas that to the
world of the ancestors is located in the southwest (13.8.1.5). As opposed to the celestial
locale of the gods’ world, the ancestors supposedly live in the atmosphere or middle space
between heaven and earth (taking the form of birds, according to BDhS 2.14.9-10), or
under the earth (RV 10.16.3; SB 13.8.1.20). In the Upanishads, the pitr loka is associated
with the moon, darkness, sacrificial action, and rebirth (this world being a way station in
the recycling of souls); the deva loka is connected with the sun, light, mystical knowledge,
and eternal liberation (BAU 6.2.15-16; ChU 4.15.5—6, 5.10.1-3; Kausitaki Upanishad
12):
186. ChU 8.2.
187. BAU 4.3.33; see also TU 2.8.
next life is determined by the specifics of his actions in this life, and in
every case the world humans occupy after death is distinct from the
heaven of the gods by birth.
The daiva atman, like the heavenly world, is also a relative term,
with contours shaped by the individual. This transcendental self, how-
ever, is not to be mistaken for the biological self that emerges out of the
mother’s womb. The ritualists in this matter, as in the case of the second
birth effected by the initiation ritual, claimed that their ritual womb
brought forth superior fruit when compared to the merely natural womb
of women:
There are indeed two wombs: the divine womb is one, the human womb
the other. . . . The human womb is [a part of] the human world. . . . And
the dhavaniya fire is the divine womb, the divine world. . .. He who
knows thus has two adtmans and two wombs. One atman and one womb has
he who does not know this. 188
The divine self is “born out of the sacrifice”; that is, it is a ritual
construct. And, further, the sacrifice specifies the particular dimensions
of the daiva atman; the ritual history of the individual sacrificer trans-
lates into the special character of the divine self: “The sacrifice becomes
the sacrificer’s self in yonder world. And truly, the sacrificer who, know-
ing this, performs that [sacrifice] comes into existence [there] with a
whole body.”!8? Or, again, another text declares that the sacrificer “is
united in the other world with what he has sacrificed”;! his ritual
accomplishments on earth are the precise measure of his divine self in
heaven. The Rg Veda refers to the sacrificer’s “treasure,” to which he
will be joined at death—a kind of savings account in the next world
composed of sacrificial deposits made in ritual activity—and to a “splen-
did body” of the sacrificer’s own making: “Unite with the fathers, with
Yama, with the treasure of your sacrifices in the highest heaven. Aban-
doning defects, return home; unite with a splendid body.”!%!
Uniting with the heavenly self one has created in sacrifice is not
always easy. Just as there is danger involved in the passage to heaven
within the confines of the ritual, so is there a certain risk involved in the
188. JB 1.17; see also SB 7.4.2.20.

189. SB 11.1.8.6. For the emphasis on the wholeness of the daiva atman, see AV
11.3.32; SB 4.6.1.1, 11.1.8.6, 12.8.3.31; and Jan Gonda, “Reflections on Sarva- in Vedic
exts;~
190. TS 3.3.8.5.
191. RV 10.14.8.
transition from a human to a divine self after death. Self-knowledge and

the ability to recognize one’s “own world” when one sees it are critical:
One man, after having left this world, knows his atman, [saying,] “This am
I.” Another man does not recognize his own world. Bewildered by the
[cremation] fire choked with smoke, he does not recognize his own world.
But he who knows the savitra fire, he indeed, after having left this world,
knows the atman, [saying,] “This am I,” and he recognizes his own world.
And then the savitra fire carries him to the heavenly world.!%
Recognizing one’s own world and identifying one’s true divine self,
one realizes and adopts the tailor-made life after death that is the prod-
uct of the ritual craft. The collective oblation one offers over the course
of the years—that is, the quantity and quality of his sacrificial history—
determines the daiva atman as it is transformed from offering to transcen-
dent identity: “Whatever oblation he sacrifices here, that becomes his
atman in the other world. When he who knows thus leaves this world,
that offering which follows him calls out, ‘Come here. Here I am, your
atman.’ ”'3 Alternatively imagined, the divine self is the collective sacri-
ficial fee—which, it will be remembered, functions as the oblation to the
“human gods,” the Brahmins:
He who sacrifices, sacrifices with the desire that he may obtain a place in the
world of the gods. That sacrifice of his then goes forth toward the world of
the gods. The sacrificial fee he gives [to the Brahmin officiants] follows after
it, and holding on to the sacrificial fee [follows] the sacrificer.!4
Indeed, there seems to have been something of a nutritive value to
the various sacrifices of the Vedic ritual repertoire. Each sacrifice has a
different sustaining power for the divine self in the next world. And,
unremarkably, the hierarchically superior sacrifices (because they are
more complicated, powerful, and rare) provide the more enduring sus-
tenance:
And now for the powers of the sacrifices. He who performs the agnihotra
eats in the evening and in the morning in yonder world, for so much
sustenance is there in that sacrifice. And he who performs the new and full
moon sacrifice [eats] every half-month. He who [quarterly] performs the
caturmasya [eats] every four months. He who [every half-year] performs
the animal sacrifice eats every six months, and he who [annually] performs
4925 'TB:3.10.41.1-2:
193. SB 11.2.2.5.
194. SB 1.9.3.1.
the soma sacrifice [eats] every year. But the builder of the fire altar
optionally eats every hundred years, or not at all. For a hundred is as
much as immortality, unending and everlasting.!*
With this passage, we again encounter the theory that life after
death correlates with the ritual performances in this world—the classical
Hindu law of karma in the making. And also presaged here, and indeed
throughout the whole of Vedic literature, is the hierarchical thought
underlying, among many other classificatory schemes, the Hindu caste
system.
Like the law of karma, Vedic ritualism is premised on the assump-
tion that acts have consequences for those who act, both in this life and
in the future. They determine who you are (in the sight of the ritualists,
at least) and who you are becoming, which is what you will be. In
Vedism, however, the law of karma is in nuce; for at this point in the
orthodox tradition, the only acts that matter, that provoke ontological
reactions, are ritual ones.!%
And just as there is an ever-present hierarchical motif underlying
Hinduism in its entirety, Vedism is also hierarchically ordered at every
turn—indeed, perhaps more consistently so than in some forms of Hin-
duism. For, unlike the later Upanishadic and Hindu configurations, in
which asceticism could turn men into gods or in which gods took on
human form out of grace and love for their devotees,!” in Vedic ritual-
ism the gods remain gods, and men essentially remain men. While the
participants in the ritual might become godlike for the duration of the
ceremony, their fundamental mortality precludes anything more than a
temporary counterform of true divinity. And even after death, the di-
vine self and the heavenly world one adopts are but ritually constructed,
and hierarchically differentiated, replicas of the gods’ status.
195. SB 10.1.5.4.
196. Another critical difference between the Hindu doctrine of karma and the Vedic
ritualists’ theory of the consequences of actions is that the latter lacked the notion of
transmigration of the soul into new human births until the time of the early Upanishads.
Although in some of the Brahmanas there is a fear of “redeath” (punar-mrtyu) when the
ritually acquired merit that has sent the disembodied sacrificer to heaven runs out (see,
e.g., SB 2.3.3.9, 10.1.4.14, 10.4.3.10, 11.5.6.9; TB 3.10.11.2, 3.11.8.5-6; KB 25.1), the
notion “is not to be understood as transmigration but only as a resurrecton and repeated
death in the other world.” Deussen, The Philosophy of the Upanishads, p. 327.
197. For the mythological and theological puzzles engendered by such ideas in Hin-
duism, see Wendy Doniger O’Flaherty, The Origins of Evil in Hindu Mythology (Berke-
ley: University of California Press, 1976), pp. 57-93; and idem, Women, Androgynes, and
Other Mythical Beasts, pp. 65—76.
In this chapter and the preceding one, I have attempted to lay out
some of the ways in which ritual, resemblance, and hierarchical presup-
positions interlock to form a coherent and consistent cosmological, meta-
physical, socio-ontological, and soteriological system. Articulated most
fully and cogently in the Brahmana texts, Vedic philosophy results in a
system of practice in which there is a correlation—that is, a relationship
of mutual resemblance—between three hierarchically calibrated regis-
ters: (1) the scale of ritual performance (the relative size, complexity,
duration, and especially the inclusiveness of the sacrifice both as an
individual performance and as a collective term for a ritual résumé), (2)
the relative quality and realization of the sacrificer’s earthly self and
status (the ritually constituted nexus of innate and acquired socio-
ontological characteristics, comprised of inherent proclivities, ritual and
educational accomplishments, and the willingness and ability to realize
them), and (3) the hierarchical order of selves and worlds of the unseen
spheres (ranging from the gods by birth and their worlds downward
through the tiers of ritually constructed heavens of the gods created by
sacrificial work to the various hells of the damned and the negligent).
The principle of hierarchical resemblance has wider application
still. For when the Brahmins turned their attention to the organizing of
the vast system of ritual practices in the encyclopedic Sutras, their
method there remained consistent with the explanations of the universe
and its inhabitants found in the Brahmanas. In the Sutras, to which we
next turn, the philosophy of hierarchical resemblance is translated into a
theory of textual representation; the ritual repertoire is systematized,
categorized, and ordered in a way entirely in conformity with the princi-
ples guiding other systems of Vedic classification.
3
The Organization of Ritual
Knowledge
Resemblance and the Sitra Literature
Although the latest of the Brahmana texts and the earliest of the ritual
Sitras were probably composed more or less simultaneously,! the differ-
ence in genre is striking. The invention of the sutra format by ritualists of
the Black Yajur Veda inaugurated a mode of representing and organiz-
ing knowledge that was extraordinarily influential. Both for the ortho-
dox tradition and for emergent rivals, the Buddhists and Jains, the sutra
became the preferred literary form in India, most especially when tradi-
tions codified their more technical discourses.
The ritual Sutras of the Vedic heritage, the so-called Kalpa Sutras,
are the earliest instances of this type of literature. The oldest extant text
composed in sutras is the Srauta Sitra of the Baudhayana school, dated
to about the seventh century B.c.£. Other Srauta Sutras, representing
the ritual according to other schools, were composed over the course of
the next several centuries, together with appendages and related texts
on the domestic ritual (the Grhya Sutras) and, finally, on all aspects of
life—inside and outside the ritual—the Dharma Sutras.
The Kalpa Sutras, then, form the corpus of knowledge most inti-
1. The ritual Sutras, with a few very late exceptions, can be dated between 800 and
300 B.c.E., according to C. G. Kashikar, A Survey of the Srautasiitras (Bombay: University
of Bombay, 1968), p. 161. Ram Gopal’s dates of 800-500 B.c.£. end too early. See his
India of the Vedic Kalpasutras (Delhi: National Publishing House, 1959), p. 89.
: The Organization of Ritual Knowledge 121
mately connected to the practice and implications of the Vedic ritual.

Other sciences of the orthodox tradition, however, also codified into
Sutras, emerged out of the necessities related to correct ritual perfor-
mances—grammar, phonology, prosody, and etymology, on the one
hand, to ensure the proper preservation and recitation of Vedic man-
tras; and astrology, on the other hand, to guarantee precision in the
calculations for timing the occurrences of various rituals. Together with
the Kalpa Sutras, these six sciences are known collectively as the
Vedangas, the “subsidiaries to the Veda.” And while the interrelations
among these six ancient Indian subjects remain little studied—and
would handsomely repay investigation—tt is likely indeed that the dif-
ferent Sutras of the Vedanga share similarities not only in form but in
epistemological assumptions as well.” In all cases, the Sutras are testi-
monies to the systematizing instincts of the ritual science; they are far
better understood as taxonomies than as handbooks or manuals for
practice.
Another unprobed, and almost certainly significant, area of ancient
Indian history is the relationship between the Sutras of the orthodox
tradition and those of Buddhism and Jainism. Why were the earliest
scriptures of these dissenting groups written in the same genre as that of
the Vedic ritualistic literature? Otherwise put, what is the meaning of
the fact that the composition of the Kalpa Sutras tended to precede, and
then coincide with, the rise of the heterodox religions? How much
should be made of the fact that the new religions utilized the sutra
format to represent their founders’ sayings and aphorisms, whereas in
the Kalpa Sutras the aphorisms are those of the various teachers within a
ritual school? The answers to these questions are outside the scope of
this study, but it would seem possible that the Jains and Buddhists
consciously composed their texts as Sutras in order to place them in
relation to those of the ritualists—a relation of imitation, competition,
or, most likely, both.
2. For an example of partial and preliminary scholarship in this area, see Louis
Renou’s comparative study of the ritual and grammatical Sutras, “Les Connexions entre le
rituel et la grammaire en sanskrit,” Journal asiatique 233 (1941-42): 105—65, reprinted in
J. F. Staal, ed., A Reader on the Sanskrit Grammarians (Cambridge, Mass.: MIT Press,
1972), pp. 435-69.
3. In the case of the Kalpa Sutras, this point is attested to by the fact that a subse-
quent body of literature—comprised of both commentaries on the Sitras and derivative
texts known as prayogas—was produced to guide the actual performance of the rituals the
Sutras refer to. The rules for the ritual given in the Sutras, in other words, are far too
cursory and lacking in detail to function adequately as manuals of practice.
As for the composers of the orthodox ritual Sutras, they were con-
sciously in dialogue with the literature of the tradition out of which they
had come: the Veda. As opposed to the prolix style of the Samhitas,
Brahmanas, Aranyakas, and Upanishads, however, the Sutras are by
definition concise, sometimes to the point of being nearly unintelligible
to the outsider. Literally translated as “thread,” the sutra designates
both a brief proposition and the ensemble of propositions which com-
prise a work.
Brevity and concision are certainly characteristic of individual apho-
risms making up the Sutras, but the way the contents of the work as a
whole are interrelated is perhaps even more definitive of the genre.
Unlike the Brahmanas, for example, the Sutras organize their subject
matter not in a linear manner but rather into a peculiarly methodical,
systematic, and didactic structure.* Such a difference in style may indi-
cate a more general shift, as Louis Renou has speculated:
Different styles, due to different needs, may be ascertained: that of specu-

lation, formulaic and enveloped in mysterious symbols, with a clear bent
toward an archaic and emphatic prose style; and that of the description of
rites, with a realistic, objective prose.°
In any case, in the Sutras the “threads” are woven together into a whole
rather than tied sequentially into a linear chain. Indeed, Renou consid-
ers the succinctness of the individual propositions less important to the
definition of the Sutra genre than the unique way in which individual
propositions are related one to another:
The genre of sitra is defined by its relation more than by its content: a
sutra (in the sense of “rule” or “aphorism”) is primarily an element depen-
dent on context, even though it is grammatically autonomous; it is deter-
mined by the system and . . . [is] correlative to the group which surrounds
it:$
In both style and the form in which the contents interrelate, then,
the Sutra departs from its Vedic literary predecessors. Still, it would be a
mistake to overemphasize these differences. The Sutras wholly presup-
pose the philosophy of hierarchical resemblance enunciated—in a differ-
ent format—in the Brahmanas, and extend its application to the literary
form itself. The primary distinction between the earlier Vedic texts and
4. Jan Gonda suveys the stylistic and organizational features of the ritual Sutras in
his The Ritual Sdtras, pp. 629-47.
5. Louis Renou, Les Ecoles Védique, p. 211.
6. Louis Renou, “Sur le genre du sutra,” p. 166.
a The Organization of Ritual Knowledge 123
the ritual Sutras is merely one of idiom, not ideology. It is the principle
of resemblance that governs both.
Although, as Renou has noted, an individual satra derives much of
its meaning from its surrounding context, the Sitras also employ certain
general rules that permeate the entire text and supplement particular
aphorisms. Knowledge of the ritual as it is codified in these works de-
pends on prior knowledge of a set of “first principles,” “general rules,”
or “metarules.” These kinds of siitras, whose reach extends over all
other sutras, are called in Sanskrit paribhdsds.7? The development of a
series of metarules has been cited as one factor in dating the ritual
Sitras.8 The earliest, the Baudhayana Srauta Sitra, only occasionally
gives general rules, and scatters them throughout the text (the Paribhasa
Sitra which is appended to the BSS is almost certainly a late addition).
An apparently late work such as the Katyayana Srauta Sitra, on the
other hand, provides a set of paribhdsds at the very beginning of the
Sutra.?
For the novice—then and now—the paribhdsds are the key to un-
locking an otherwise often cryptic Sutra. Their general applicability!
clarifies the text as a whole, just as a light placed in a corner illuminates
the entire room.!! “A paribhasa of a Srautasitra,” writes S. C. Chakra-
barti, “means a general principle that facilitates the correct interpreta-
tion of the work.”!? Indeed, a set of metarules makes possible an individ-
ual sutra which is maximally succinct, a highly desirable goal for the
authors of the Sutras; a man was said to take as much delight in the
saving of a single syllable in his Sutra as he did in the birth of a son.
The metarule enables such economy in exposition because it states
once what thereafter can be assumed throughout the text unless specifi-
cally contradicted. In the ritual Sutras, for example, metarules state that
unless otherwise noted, the reward for a sacrifice is the attainment of
7. For the following discussion, I am greatly indebted to S. C. Chakrabarti’s pio-

neer study, The Paribhdsds. For the paribhasas in the Sutras of grammar, see K. V.
Abhyankar, “On Paribhdsa Works in Sanskrit Grammar,” Annals of the Bhandarkar
Oriental Research Institute 36 (1955): 157-62.
8. See, e.g., Kashikar, A Survey of the Srautasitras, pp. 153-57.
9. For some reservations on depending too heavily on this criterion for dating the
Sutras, consult Staal, The Science of Ritual, p. 21.
10. See, e.g., SSS 1.2.29.
11. The image is taken from the grammarian Patanjali. See S. D. Joshi, ed. and
trans., Patanjali’s Vyakaranamahabhdasya (Poona: University of Poona, 1968), on Panini
an a
12. Chakrabarti, The Paribhasas, p. 27.
13. Cited in Staal, The Science of Ritual, p. 22.
heaven, the fire will ordinarily refer to the dhavaniya fire (in the Srauta
Sitras), the usual oblation is that made of rice paddy or barley, and so
on. Once the metarule has been laid down, the persons and objects
subject to the ritual rule are often subsequently referred to only by
pronouns, and the reader must know the metarule in order to know
whom or what is meant by the pronoun at hand.
The principle of the metarule introduces the systematic mode of
presentation characteristic of the ritual Sutras and serves as the first
instance of the ways the Sutras assume and extend the general Vedic
principle of hierarchical resemblance. For the underlying force of the
metarule is the assumption of the resemblances of particulars to the
overarching generality. Put another way, the notion encountered here is
that knowledge of the particulars can be inferred from knowledge of
their archetype. To know the archetype—the metarule—is thus to know
all the manifestations of the archetype—the particular resembling in-
stances. Resemblance, however, is not identity. A metarule on meta-
rules states that the distinctiveness of the particular overrides the gen-
eral application of the metarule. Thus, “a specific injunction is stronger
than a general one.”!4
This interplay between the general and the specific is the heart of
the science of ritual as it is represented in the Sitras. And this science is
guided by much the same assumptions as those guiding the metaphysics
of the Brahmanas. For here, too, the discourse is delimited by the
excesses of identity and difference. The excess of resemblance (jami) is
found in the erroneous view that the particular is identical to the general;
hence the metarule cited above that overrides the general when particu-
lar distinctions are encountered. The excess of difference (prthak) finds
its transformed expression in the Sitras as well. There are cases when
there is no specific indication of to which genus or class a particular
species of rite or ritual belongs; it appears unique, unconnected, and
different from all others. As we will see shortly, the concept of resem-
blance (saémanya) developed in the Sitras makes it possible to avoid this
epistemological danger and to link any particular to its paradigm.
In Vedic philosophical discourse, the general and the particular
are expressed in terms of “prototypes” and “counterparts,” “forms”
and “counterforms.” In the organizational structure of the ritual Sa-
tras, a comparable contrast is set up between paradigms (prakrtis) and
variations of the paradigm (vikrtis). Entire rituals, individual rites, and
specific offerings within rites are thus constituted either as paradig-
14. Aév§S 1.1.22.

The Organization of Ritual Knowledge 125
matic guides to resembling variants or as modified variants of another

paradigm. Whereas in the Brahmanas prototypes are the hierarchically
superior and fuller forms and counterparts the inferior and less com-
plete versions thereof, in the Sutras the paradigm is the basic model
out of which variations are generated. Thus, for example, the agni-
stoma is the simplest form of the soma sacrifice, but also the ‘paradigm
for all more developed, more complicated, and hierarchically superior
soma sacrifices.
But although the variants are more fully developed versions of their
ritual paradigms, the economy of explanation in the Sutras reverses the
relationship between the two and replicates more exactly the hierarchi-
cal element of the prototype/counterpart relation in the Brahmanas. For
in the Sutras the paradigm is much more completely explicated—just as
the prototype is the more completely realized instance of its class. The
variation is only partially explicated in the text, for the previously given
specifics of the model are assumed and extended to the variant.
The explanatory form of the Sutras thus reverses the hierarchical
value of the sacrifices it is explaining, working from the most basic to the
most highly developed rather than from the superior to the inferior. The
relationship between the general and the particular—between the para-
digm and the variation—is described as one between what is “prior”
(purva) and what is “subsequent” (uttara) and therefore dependent:
“How should one know what is the precedent ceremony and what is
subsequent? That which is the paradigm (prakrti) is the precedent, and
that which one creates from it is subsequent.”!> The interdependence of
the agnistoma and its more complex variations is depicted in the same
manner: the latter are “subsequent” forms of the “prior” paradigm.'!°
One definition of the paradigm emphasizes this epistemological priority
in terms of the powers of extension it has; “Paradigm is the name of that
which is transferable.”!”
But the science is complex. For what functions as a variation of one
paradigm can simultaneously act as a paradigm for variations of another
class. The rite of animal sacrifice to Agni-Soma in the soma sacrifice is a
variation of the paradigmatic new and full moon sacrifice; and it is also
the paradigm for other animal sacrifices within soma rituals. In this way
and others, the Sutras fabricate a chain of intricate connections as they
organize the variety of rituals into a systematically interlinked whole.
15. BSS 24.5.

16. BSS 24.5: jyotistomah somanam pirvd tatih sarve soma uttara tatih.
17. SSS 1.16.2.
At the highest level of its operation, the principle of paradigms and

variations acts to relate all sacrifices one to another. The metarules
declare that the agnyadhana ritual, in which the sacrificer’s fires are first
lit, is the paradigm for the punarddhana or reestablishment of the fires
should they become extinguished. The new and full moon sacrifice is the
paradigm for all other sacrifices in which the oblation is made from
vegetable matter (the istis) and for the animal sacrifice dedicated to
Agni-Soma. The agnistoma serves as the paradigm for all soma sacrifices
up to those of twelve days’ duration; the dvddasaha soma sacrifice is
paradigmatic for soma sacrifices of twelve days or more. In this manner,
all rituals find their place in the explanatory chain.!8
The new and full moon sacrifice is perhaps the most important of all
the paradigmatic rituals. Knowledge of its structure is essential to knowl-
edge of nearly all other rituals—including, as we will see, the rituals at
the domestic level (the grhya sacrifices )—and the Sutras constantly refer
back to it or assume its details. The Srauta Sitras open either with a list
of metarules or, more frequently, with a detailed account of the new and
full moon sacrifice which serves much the same purpose as the set of
metarules—it provides transferable knowledge for the performance of
any sacrifice in the repertoire. The Sutras, then, in nearly every instance
follow an epistemological rather than a chronological order, the paradig-
matic new and full moon sacrifice preceding the ritual for lighting the
fires.19
The Agvalayana Srauta Sutra, to take but one example, opens with
some short statements of purpose (to demonstrate the ritual application
of the verses from the Rg Veda and to detail the procedure for the §rauta
rituals beginning with the agnyddhdana). It then provides an explanation
of why the following account will begin with the new and full moon
sacrifice, and not with the ritual of the lighting of the fires: “We will
explain the new and full moon sacrifice first because the teaching on the
basic ritual framework (tantra) is known from it.””° The inferential epis-
temological possibilities of the paradigmatic sacrifice take priority over
the chronological necessities of the ritual life in the Sutra organizational
scheme.
18. For a full list of the paradigmatic rituals and their variations as they are given in
the Sutras, consult Chakrabarti, The Paribhasas, p. 136.
19. Two exceptions are the late Vaikhanasa Srauta Sitra and the poorly transmitted
and fragmentary text of the Vadhula school, both of which begin with an account of the
Agnyadhana.
20. AsvSS 1.1.3.
fs The Organization of Ritual Knowledge 127
It is indeed the establishment of a system of inferences that is

fundamental to the Sutra literature. When the paradigm is known,
inferences to the variations are guided by what is transferable from the
former to the latter and what is not. But when there is no indication of
what is paradigmatic for a particular case, other metarules aid the
ritualist. The general rule for tracing an unknown paradigm from a
known variation invokes a principle that operates in many other situa-
tions in the Vedic mental universe. “In case of doubt, the rule is to be
inferred from resemblance (samanya)”;?! other texts lay down a similar
proposition, stating that “The modification (vikdra) is known from the
resemblance.”?? The resembling features of different sacrifices make
possible both the inferences between them and the discovery of the
appropriate paradigm from which to infer. Resemblance brings to-
gether rules and rites of a common type, linking paradigms and varia-
tions, in the ritual Sutras, just as resemblance relates prototypes and
counterparts in the metaphysics of the Brahmanas. The intricacies of
this web of ritual resemblances in the Sutras will engage our attention
in what follows.
The Organizational Economy of Resemblance
We have already encountered some of the technical terms employed in

the Sutras to analyze and structure the ritual system: “paradigm”
(prakrti), “variation” (vikrti), and the key concept of sémdanya, which I
have translated as “resemblance.” Before proceeding deeper into the
organizational fabric of the Sutras, it will be necessary to introduce a few
more of these critical terms.
All variations of a common paradigm utilize the same set of what
are known as “subsidiary rites” (anga), which are described fully only in
the Sutra’s treatment of the paradigm. Taken together, the subsidiary
rites constitute the sacrificial framework (tantra), and it is indeed by
virtue of the shared framework that rituals are classified together as
variations of a paradigm. In the procedure of any one of these varia-
tions, the framework is followed up to the point of the “principal rites”
(pradhana) of the individual sacrifice and is again adopted after those
principal rites are concluded. There is thus a “prior framework” (purva
tantra) and a “subsequent framework” (uftara tantra) common to related
21. SSS 1.16.18.

22. BhSS 6.15.9; HSS 3.8.
rituals, located on either side of a distinctive center comprised of the

principal rites.”
It is the “insertion” (avapa) of the set of principal rites in the middle
of otherwise resembling sacrifices that gives each variation of a common
paradigm its unique identity. The principal rites are the procedures
taken for offering oblations to particular deities who are invoked for the
success of the sacrifice. This is why it is said that the principal rites are
the critical part of the ritual; it is with them that the desired results or
“fruit” of the particular ritual are produced. Therefore, the subsidiary
rites are merely subservient adjuncts to the main event of any ritual
performance; their characteristic “mark” is their role in helping to
achieve the purpose of the principal rites.*4 In summary, as Chakrabarti
writes,
The principal offerings constitute the principal component part of a sacri-

fice, whereas the other acts that subserve them are treated as their subsid-
iaries. The subsidiary acts constitute the tantra, “the sacrificial frame-
work,” and the principal part is known by the term a@vdpa.”
With this arsenal of technical terms now in hand, we are ready to
progress further into the workings of the ritual science. The ritual is
organized in the Sutras on the basis of a series of judgments regarding
the basic similarity or difference between the various sacrifices.
One form of similarity is known as “extended application by anal-
ogy” (atidesa), and we have already seen one way in which it is de-
ployed: the ritual framework of the paradigm is extended by analogy to
all the variations of that paradigm. As a general principle, extended
application helps the Sutra writers to minimize words by tapping connec-
tions and resemblances:
The sacrifices some of the particulars of which are not specified borrow
them from other similar sacrifices that are enjoined with all the details.
This process of extended application of details by analogy is called atideSa,
by which the subsidiaries prescribed in connexion with one sacrifice are
made applicable beyond that sacrifice and transferred to similar sacrifices.
The sacrifice from which the subordinate rites are transferred is called an
23. For more details on the specific rites that constitute the parva tantra and the
uttara tantra and the exact location for the insertion of the principal rites, consult SSS
1.16.3-6.
24. SSS 1.16.6.
archetype (prakrti) and the sacrifice to which they are so transferred is

called an ectype (vikrti).6
Such extensions on the basis of similarity or analogy are further
analyzed into two kinds. The first is a kind of interior extension whereby
the purpose of a particular sacrifice extends from the center—from the
principal rites—to the periphery or subsidiary rites. The second type of
extension by similarity is the extension of details from one whole ritual
to another, from the paradigm to the variation.
The first of these two types involves the extended application of the
purpose (artha) of a particular ritual from the center (which is directly
concerned with the fulfillment of that purpose) to the subsidiary rites,
which, by themselves, produce no results. They derive their purpose, by
extended analogy, entirely from the principal rites they subserve. The
purpose of a sacrifice in this way radiates outward to the otherwise
purposeless subsidiary rites.
To take one text as a case in point of this phenomenon, the
Katyayana Srauta Sitra makes it quite clear that the principal rites are
the critical actions in the ritual procedure for the attainment of the
purpose and “fruit.” Rites which do not produce results are by definition
subsidiary to others;”’ and the principal rites, by contrast, may generate
results without their subsidiaries “because of the separation” between
the two.”8 Because the purpose of the ritual is independent of the subsid-
iary rites, these latter may in some cases be performed in any order at
all.?9
It is more often stated, however, that the principal and subsidiary
rites are interdependent. Multiple oblations to several deities occurring
during the principal rites of a single sacrifice are subserved by a common
set of subsidiary rites which are performed only once. This is because, it is
said, the subsidiaries “make possible the purpose” fulfilled by the princi-
pal rites, because both the subsidiary and principal rites share “one pur-
pose,” and because the Veda has made authoritative judgments regarding
their “union.”*° Subsidiaries, although unable to produce results inde-
26. Ibid., p. 131. See also S. G. Moghe, “The Evolution of the Mimamsa Technical
Term Atidesa,” Annals of the Bhandarkar Oriental Research Institute 58-59 (1977-78):
777-84.
27. KSS 1.2.4.
28. KSS 1.2.18-19.
29. KSS 1.5.1-2.
30. KSS 1.7.1-2.
pendently, are “attracted” to the purpose of the principals they serve;

they are called pararthatva, “subservient to the purpose of another.”*!
Another way in which the texts present the interdependence of the
subsidiaries and principals and the extension of purpose within the inte-
rior of the ritual from center to periphery is in terms of a “commonality
of context” (prakarana):
The subsidiary acts are linked with the principal ones enjoined in the same
context by prakarana, which means a mutual desire to be united with each
other. . . . The principal offerings of a sacrifice produce the result only
when they are duly accompanied by the subsidiary rites, and hence they
look for the latter. On the other hand, the subsidiary rites, which cannot
produce the main result independently, become fruitful by subserving the
purpose of the principal offerings. This mutual desire unites the principal
and the subsidiary components of a sacrifice, which, as a whole, leads to
the desired results.*?
The “commonality of context” thus provides a mutual attraction be-
tween subsidiary rites and principal ones; it also bonds together the
different principal oblations within the center of a single sacrifice. The
various principal oblations are said to “share a common procedure”
when they share a common context.*
In the interior of the ritual procedure, then, the extended applica-
tion on the basis of analogy or similarity works in several different ways.
It transfers the purpose of the ritual from the principal rites to the
subsidiaries, it makes possible a common subservient framework for
several different principal oblations, and it links those different principal
oblations together into a “common context.”
The second type of extended application by analogy is that by which
the features of the paradigmatic sacrifice are extended to the structure of
the dependent variant sacrifices. Whereas the first type of extended
application functions within one ritual, this second type operates be-
tween two or more rituals, and in the latter case as in the former, the
principle works in several different ways.
The extended application may transfer knowledge of the subsidiary
rites of the paradigmatic ritual to the variant rituals of the class. All
variations resemble each other in that they share the framework of
subsidiary rites found in their paradigm. Thus, for example, the ritual
aly See, e205 KSS 1.6.10. For the mimamsa analysis of this concept, see PMS 3.1.2-
6 and Madhva’s Jaiminiyanyadyamala (Poona: Anandasrama, 1892), 3.1.5.
32. Chakrabarti, The Paribhdasas, p. 126.
33. ApSS 24.2.26-27.
framework of the new and full moon sacrifice is extended by analogy to

all the variations (the istis and the animal sacrifice) “because of the
commonality of purpose” which that framework serves in all rituals of
the class.*4 Similarly, many features of the agnistoma extend to the soma
sacrifices of one to twelve days’ duration “because of the appearance of
the same auxiliary acts” in both the paradigm and the model.*5 An
alternative way of stating this same principle is to say that the distinctive
principal oblations of a variant ritual are performed “with the support”
of the subsidiary rites extended from the paradigmatic sacrifice of the
class.%¢
Another form that extended application between sacrifices may
take is to transfer the rules for the principal oblation themselves from
the paradigm to the variation. Thus, one text states that the rules for
offering the principal oblations of the new and full moon sacrifice extend
over all the “subsequent” sacrifices, with a few exceptions noted.”
Extended application on the basis of similarity or analogy—working
either within a sacrifice or between sacrifices—is a frequently encoun-
tered organizational principle in the ritual Sutras and is one way in which
the Sutra writers adapted the fundamental Vedic principle of resem-
blance. But resemblance, as we have had ample occasion to witness, not
only makes possible a set of connections based on similarities but also
requires differentiation. The principles by which the ritual scientists
distinguished sacrifices from one another comprise the flip side of the
same coin that made it possible for them to forge connections between
similar sacrifices.
Some of the ways in which rituals are distinguished have already been
noted. Different classes of sacrifices are so categorized on the basis of the
different subsidiary rites which form a definitive ritual framework for the
class. This distinctive framework is taken from the paradigmatic sacrifice
of each class and extended over the variants, constituting them all as
distinctive from other classes of rituals. Within each class, moreover, each
member maintains its individuality (and thus avoids the excess of resem-
blance the Brahmanas call jami). For, although every ritual within the set
shares the same subsidiary rites (the ritual framework), every one also has
distinctive and, again, definitive principal rites. As Chakrabarti puts it,
“The procedure cannot, however, be identical because the ritual proce-
34. KSS 4.3.2.

35. KSS 12.1.1; see also KSS 13.1.1.
36. KSS 1.5.4.
37. HSS 3.8.
dure varies with the principal offerings, and the difference in principal
offerings distinguishes one sacrifice from another.”%8
Finally, we have already seen that within the ritual organization of
the Sutra texts the general rule is overriden by the particular. Extended
application on the basis of the similarity between the paradigm and the
variation occurs only in cases where no specific injunction to the con-
trary is encountered with the variant. A specification in the variant
requires a cessation of the extended application, as, for example, in the
case of the different deities to which the principal oblations are offered:
In all cases where deities are specified, the gods of the paradigmatic
sacrifice should be withdrawn [from the procedure of the variant]. Divini-
ties between those of the two oblations of clarified butter and the offering
to Agni Svistakrt [i.e., the deities of the principal oblations] are then
replaced [by those specified for the variant sacrifice].*°
There are other grounds for the cessation of the extended applica-
tion from the paradigm to the variation; there are, in other words, other
methods for constituting differences between the rituals of a class. One
text rules that “the [extensions from] the paradigmatic sacrifice [to the
variations of it] are annulled on three grounds: by counterinjunction,
prohibition, or loss of purpose.”4? Examples of counterinjunction would
include not only the declaration of distinctive principal oblations to the
appropriate deities within the variant but also the specification of differ-
ent materials used for those or other oblations. A prohibition means that
in the variant it is explicitly required that certain of the otherwise ex-
tended rites from the paradigm be dropped. Loss of purpose (arthalopa)
refers to cases of superfluity. For example, when the variant ritual calls
for an oblation of rice gruel instead of the usual cake, the rites for
preparing the cake are dropped from the variant procedure because they
no longer serve any purpose.
Other Sutras provide other lists of occasions when the extended
application ceases. The Bharadvaja Srauta Sitra*! includes “accomplish-
ment” (siddhi) along with others already mentioned, which may very
38. Chakrabarti, The Paribh§as, p. 132; emphasis added. J. M. Verpoorten, in

“Unité et distinction,” p. 82, echoes the point: “Le début et la fin du rite doivent former
une unité. Ils seront semblables l’un a l’autre, se réspondront a tous égards par dela une
zone intermédiaire, un centre, qui, par sa consistance propre, les séparera, les distinguera,
et permettra ainsi de parer au reproche de jami, de répétition superflue.”
39. AévSS 2.1.22-23.
40. ApSS 24.4.2.
41. BhSS 6.15.6.
well be equivalent to the “change in purpose” listed in another text.

For, as yet another text explains, “The modification in the ritual frame-
work of sacrifices not mentioned here [in a section on the paradigmatic
sacrifices] should proceed out of consideration for the particular purpose
[of those sacrifices].”43 Sacrifices are distinguished by their principal
rites, and the individualized purpose of sacrifices is brought about by
those principal rites; therefore, a change in the center will sometimes
require a change in the subsidiary framework, to conform with a
“change in purpose” and to make possible the “accomplishment” of that
new purpose.
These, then, are some of the reasons for annulling the extension of
the procedure from the paradigm to the variant, some of the rules for
knowing where and how to distinguish the variant from the paradigm.
We come now to a somewhat different problem, the solution to which
brings us to the crucial concept in the Sutras of resemblance or saémdanya.
When there is no particular injunction specified in the variant, and when
there are competing claims of analogy within the paradigm, how does
one decide which of the similarities take priority for the procedure of the
variant sacrifice? What is at issue here is how to sort out competing
claims of resemblance which have caused a “contradiction” (virodha) or
confusion in a system based on inferences and transferences based on
resemblance.#
The solution to the problem requires an understanding of the way
the ritualists analyzed resemblance. The general rule for inferential
knowledge from the paradigm to the variant is that the “modification”
(vikadra) required is known from resemblance, that is, by finding the
most analogous instance in the paradigm. The general rule is further
refined, however, by an analysis of resemblance into three kinds: resem-
blance of the number of potsherds on which sacrificial cakes are baked,
42. Others included in SSS (9.1.3-7) are “appellation” (némadheya), prohibition,

“number” (parisamkhya), and “addition” (upajana). It is not clear what is meant by most
of these terms, however. See Caland’s notes to these sutras found in his translation of the
text. For other instances in which extended application is to be annulled, consult KSS
4.3.8, 4.3.22.
43. AsvSS 2.14.14.
44. In some text places (e.g., SSS 1.16.18; ApSS 24.3.46), this “contradiction” in
claims of analogy is designated by the term sdmdanya, which I have translated as “resem-
blance.” In those passages, it would seem as if one must translate the term in much the
same manner as I have suggested for the Brahmanic word jami—that is, as “excessive
resemblance” or, perhaps, “a conflict in resemblance.” Virodha is, in any case, the usual
designation for the problem at hand.
resemblance of divinity, and resemblance of oblation material.** These

three kinds of resemblance establish paradigmatic rites located in a vari-
ety of particular rituals. The extended application is thus from a paradig-
matic rite rather than from a portion of the paradigmatic ritual, and the
modification is based on that rite regardless of the class to which the
variant ritual belongs.
The first type of resemblance is fundamentally one of numerical
analogy. Thus, “when there is no specific injunction” regarding the
procedure required in the variant, the following guidelines come into
play: all cases in which there are cakes baked on eight potsherds are
modifications (vikaras) of the procedure followed in the rite of the prepa-
ration of the cake dedicated to Agni in the new and full moon sacrifice;
all instances of cakes baked on eleven potsherds are modifications of the
preparation of the cake dedicated to Agni-Soma in the full moon sacri-
fice; and so on.*6
The second sort of resemblance is that of the deity, and here, too,
the analogy is often purely numerical. Offerings in a variant sacrifice
for a single deity (for which there are no specific instructions to the
contrary) follow the rites accompanying the offering to Agni in the new
and full moon sacrifice. Offerings to two or more deities are modifica-
tions of either the cake offered to Agni-Soma in the full moon sacrifice
or that dedicated to Indra-Agni in the new moon sacrifice.4? Within
this category of resemblance, however, there are cases of analogy
which are not simply numerical. Thus, for example, if the variant sacri-
fice calls for an offering to Indra as part of the principal oblations, the
inferences for the procedure are not based on the paradigmatic offer-
ings to a single deity; rather, they are to be modifications of the offer-
ing to Indra-Agni (a dual deity) because of Indra’s appearance in both
cases.*8 Here we begin to see the possibility of conflicting resemblances
and the resolution of the conflict. Resemblance between the number of
divinities receiving offerings conflicts with the resemblance established
by virtue of the fact that the same deity appears in two cases. The
conflict in this case is solved by giving priority to the latter; it is the
“stronger” of the two resemblances.
The third type of resemblance the Sitras recognize is that of the
oblation material. The rite of preparing the offering of porridge for
45. BSS 27.14; see also HSS 3.8.

46. BSS 27.14; see also ApSS 24.3.38-39.
47. BhSS 6.15.9; ApSS 24.3.41-44; HSS 3.8.
48. BhSS 6.15.10; ApSS 24.3.45; HSS 3.8.
Soma in the cdturmasya ritual is paradigmatic for all offerings of por-

ridge in which there are no specific injunctions. Oblations of curds,
milk, the mixture of coagulated and fresh hot milk (amiksa), and offer-
ings of animal parts all are modifications of the rites for preparing and
offering the milk mixture for Indra at the new moon sacrifice.4?
It is now time to return to the question of conflicting claims of
resemblance. It should come as no surprise that in the Sutras, as in other
texts of the Vedic tradition, the hegemonic principle of resemblance is
modulated by the equally important principle of hierarchy. Conflicting
claims of resemblance are arbitrated by the establishment of a hierarchi-
cal order of precedence, centering in this case on the perceived import to
the effectiveness of the sacrifice. Thus, of the three types of resemblance
analyzed above, the resemblance of potsherds is determined to be the
least important and is overruled by either a competing resemblance of
divinity or one of oblation. And between deity and oblation, it is the
latter that prevails, for it is said to be “stronger.”5° Elsewhere, the
hierarchical superiority of the resemblance of oblation is justified “be-
cause of the intimate union of the material” from which the oblation is
made and the sacrifice as a whole.*! Other texts concur in stating that the
resemblance of oblation overrides competing claims to resemblance by
virtue of its “strength.”>2
Hierarchical ordering adjudicates other kinds of conflicting claims
in the system of resemblances, linkage, and transferences so elaborately
woven in the Sutras. In one text, a series of metarules provides just such
a hierarchy of resemblance for determining the means for inferential
knowledge within the ritual universe. In the event of a contradiction
between the instructions for performing the subsidiary rites of the para-
digmatic sacrifice and those for the variant sacrifice, the former prevail.
The “multiplicity” of the principal oblations in the variation may require
overruling the rules for performing the subsidiary rites transferred from
the paradigm, and the “position” of the principal oblations in the ritual
sequence overrides the demands stemming from their multiplicity. And
a direct injunction regarding the principal oblation takes precedence
over the inferences made on the basis of the considerations listed
49. BSS 27.14. The fact that the meatless new and full moon sacrifice is regarded as
the paradigmatic ritual for, among other variants, the animal sacrifice is explained by this
less than apparent analogy between the milk libation in the former and its modification, the
offering of the omentum of the animal, in the latter. See BhSS 7.6.9 and KSS 4.3.14-16.
50. BSS 27.14.
51. KSS 4.3.9.
52. See ApSS 24.3.46; HSS 3.8; SSS 1.16.17.
above.*3 The hierarchical order of the authoritative strength of the claim

thus moves from the most general to the most specific.
In all instances of such hierarchies within the Sutra’s organization of
the ritual, the underlying consideration is the most direct way to achieve
the “purpose” (artha) of the particular sacrifice. This concept of purpose
stands at the apex of considerations regarding the relations between rites
and rituals; it is the standard against which all others are gauged.
The resemblance of oblation takes precedence over other types of
resemblance because it is more directly and effectively connected with
the fulfillment of the purpose of the ritual. But even the resemblance of
oblation is overpowered by other and more pressing instances of “resem-
blance of purpose” (arthasGmdnya), that is, an even more powerful
analogy between the goals of two sacrifices: “When there is a contradic-
tion of resemblance between the oblation material and the purpose, it is
the resemblance of purpose that is more important because that [the
oblation] is for the sake of the other [i.e., the purpose].”*4 Other texts
note that the oblation “depends on” the ritual act®> and that the ritual act
is what brings about the accomplishment of the purpose of the obla-
tion.** Chakrabarti sums up the view of the Sitras by observing, “A
material is prescribed for serving some purpose, and not simply for the
sake of being introduced. Hence a material is less important than the
purpose it has to serve.”5’ As we will observe in a later chapter, the
overriding consideration of purpose makes possible any number of sub-
stitutions within a ritual. Here it is enough to observe the hierarchical
superiority of purpose within a scale of resemblances constituting a web
of connections linking rites and rituals. Indeed, one text makes this
principle the most important of all for guiding the ritualist: rules speci-
fied in the paradigm are to be modified in the variant rituals “according
to the purpose” of the individual sacrifices.*8
This brief survey of the inner workings of the ritual science as it is
represented in the Sutras does not pretend to be comprehensive.
Rather, this overview of the nuts and bolts of the metarules and their
53. HSS 3.8.

54. KSS 1.4.16; see also KSS 1.5.5; ApSS 24.3.48.
55. HSS 3.8.
56. BSS 24.2.
58. KSS 4.3.20.
59. The rules for applying and transferring mantras, for example, provide another
context for witnessing the ritual scientists at work. See Chakrabarti, The Paribhasas, pp.
132-36, 154-65.
applications was meant to underline two points. First is the fact that the
organization of ritual knowledge in these highly technical texts proceeds
in ways comparable to the philosophical discourse of the Brahmanas.
Although there are obvious differences in genre, age, authorship, and
goal distinguishing the earlier texts from the Sitras, there are more
fundamental continuities extending between them. Vedism—a religion
represented in both the Vedas per se and the ritual Satras—is in every
one of its textual productions characterized by the dominance of the
principle of hierarchical resemblance.
The second point, which will be developed in the next section,
concerns the implications of the metarules for evaluating the relations
between two of the components that comprise a Kalpa Sitra—the
Srauta Siitra, which covers the large-scale Srauta sacrifices, and the
Grhya Sutra, whose subject concerns the simpler rituals at the domestic
level. The principle of hierarchical resemblance, as it was recast into a
classificatory and literary tool in the ritual Sutras, links these two main
components of a Kalpa Sutra in much the same way as individual rites
and rituals are connected and differentiated. In other words, I will ex-
plore the way the epistemological connections the Sutras formulated
within a text were also extended between texts of the same ritual school.
The Srauta Sutra of the school appears first in the textual corpus
and functions in much the same way as the paradigmatic ritual, which is
also epistemologically prior. The Grhya Sutra, in which chronologically
prior rituals are described, is dependent on the Srauta Sutra in ways
redolent of the manner in which the depiction of the variant sacrifice
depends on the paradigm in the Sitra’s exposition.
Intertextual Connections
In the Srauta Sitras, metarules direct the ritualist in making necessary

inferences when there is insufficient information given in the text itself.
These metarules, helpful as they are, do not succeed in filling in all the
gaps. The Sitras are not adequate guides in themselves to ritual prac-
tice; they are not self-contained texts.
It is thus misleading to refer to these literary productions, as some
do, as handbooks or manuals of ritual practice. It is impossible, as
revivalists in modern India have discovered, to carry out any Srauta
sacrifice completely and properly when one has only the Sutra. There is
simply too much left out and too much taken for granted in the interest
of a succinct representation. Modern practitioners rely exclusively on
the prayogas for such supplementary information. These texts, many

quite recent, are true manuals of performance; their step-by-step instruc-
tions for the execution of the ritual go into exhaustive detail. Some
scholars speculate about the existence of similar kinds of ritual hand-
books, complementing and augmenting the Sutras, in ancient India,
although no instances of genuine antiquity have survived.
The ancient ritualists may have relied on prayogas to fill in the
lacunae of detail left by the Sutras, but they certainly depended on the
earlier ritual texts of their tradition, the Vedas. Like all types of Sutra,
the Srauta Satras are classified as smrti, “traditional” or “remembered”
knowledge originating in human teachers, and not Sruti, the “revealed”
and canonical knowledge springing from transcendent sources. The de-
pendence of the smrti, and especially the ritual Sutras, on the sruti is
profound. Renou considers the Vedanga as a whole as being “at the
same time without and within the Veda,” though he further notes that
“from the Indian point of view, it is a line of rigorous division” separat-
ing the two classes of texts.*! The line is at its fuzziest, however, when
the Srauta Sutras are considered. As C. G. Kashikar points out, “The
Srautasiitras deal with the Vedic sacrifice in a systematic and detailed
manner almost strictly following the Samhitas and the Brahmanas includ-
ing the Aranyakas.”* More than any other example of the Sitra litera-
ture, perhaps, the Srauta Sutras are thematically and epistemologically
dependent on, and a continuation of, the Veda.
One source, then, for supplementary knowledge for the Sitras—for
what I will call here intertextual inferences or connections—is the Veda.
Every Srauta Siitra assumes a certain knowledge of the Sruti texts of the
school. The portions of those earlier texts which present the rules of the
ritual (the vidhis) are most often either wholly repeated or simply as-
sumed in the Sutra’s representation of the ritual; and the philosophical
or explanatory aspects of the Brahmanas (the arthavddas) are entirely
presupposed. Furthermore, the Sutras also assume knowledge of the
Samhita of the school in which are located the verses which are utilized
60. For such speculations, see Kashikar, A Survey of the Srautasutras, p. 29; and Jan
Gonda, Vedic Literature, p. 339.
62. Kashikar, A Survey of the Srautasitras, p. 3; see also p. 15: “Broadly speaking,
therefore, there is no difference between the rituals represented by the Brahmanas and
those represented by the Srautasitras.” For the relations between these two genres, see
Naoshiro Tsuji, On the Relation between Brahmanas and Srautasutras (in Japanese with an
English summary) (Tokyo: Téyo Bunko, 1952). Tsuji’s work supersedes R. Lébbecke,
Uber das Verhdiltnis von Brahmana und Srautasitren (Leipzig; Kreysing, 1908).
in the ritual as mantras. In the Siitras, these mantras are almost always
given in what is known as the pratika form: only the first few words of
the complete verse are provided, the remainder to be supplied by the
already memorized text in the Samhita.
Another outside source for the details of ritual practice supplement-
ing the Sutra is the actual procedure adopted by authoritative ritual ex-
perts. This, too, might be regarded as a kind of intertextual connection—
the other “text” being the memories and practices of human beings to
whom the Sutras refer in cases of doubt or controversy or simply out of the
interest in conciseness of exposition.
Both the Veda and the ritual experts are included in a list of authori-
tative sources for supplementary knowledge of the ritual provided at the
beginning of one Sutra’s collection of metarules: “One should look to-
ward five authorities for help in knowledge of the ritual—the mantra,
the Brahmana, evidence, inference, and the interchangeability of as-
pects of rituals of the same class.”® The passage goes on to illustrate
each of these methods. By “mantra” the Sttra means that the order of
verses and formulas in the Samhita may serve as an indication of the
order of performance of the rites using those mantras. Brahmana refers
to the section of rules given in that text specifying the correlation with
particular mantras and ritual acts, as well as those rules which indicate
that certain rites are to be performed without the recitation of mantras.
The method called “evidence” involves the evidence of ritual procedure
as it is laid out in a Veda other than one’s own. Thus, in this case, the
ritual of the Yajur Veda is sometimes further supplemented by the rules
given only in the Sama or Rg Vedas. Inference, the text explains, means
this: “When, given the authoritative sacrificial framework, one cannot
make [the usual] authoritative inference, one should resort to [the ritual
experts] of the villages, [thinking] ‘This is the way of undertaking [the
ritual] taught by the knowledgeable ones here.’ ” Finally, the inter-
changeability of aspects of rituals of the same class is explained in terms
of legitimate substitutions based on resemblances between the elements
of variant rituals following the same paradigmatic sacrifice. Such are the
methods the Srauta Sutras themselves devised to complete an exposition
of ritual practice they left incomplete.
But what might be the ways in which the Grhya Sutras, equally
succinct representations of the domestic rituals of the Vedic repertoire,
supplement themselves by establishing linkages to authorities outside
the text? Unlike the Srauta Sitras, the Grhya texts can make no claim of
63. BSS 24.1.

close association with the ritual described in the Veda.“ Srauta, after all,
means “relating to the Sruti”; one of the alternative names for the grhya
ritual is smarta, “relating to the smrti,” indicating a rather different point
of reference.
Because the domestic sacrifice is not the ritual the sruti revolves
around, neither the order of the mantras in the Samhita nor the applica-
tion of the mantras to ritual action found in the rules of the Brahmanas
provides a useful method for augmenting a Gryha Sutra. That is, the
Grhya Sutra does not, by and large, explicitly cite the authority of the
Sruti to justify ritual rules, nor does it depend on the Vedic texts for
inferential knowledge concerning ritual procedure. When a Grhya Sutra
utilizes the verses of the Samhita as mantras in the domestic ritual, it
usually draws them out of their original context; the order of verses as
they are found in the Samhita is therefore not usually crucial—and is
often irrelevant—for supplementing knowledge of the grhya rites as it is
transmitted in the Grhya Sutras.
There are, however, more or less indirect claims of linkages to the
Veda issuing forth from the Grhya Sutras. One text of domestic ritual-
ism goes so far as to declare that the rules for the grhya sacrifices were
once also given in Brahmanas, but unfortunately those texts have disap-
peared. Still, the text maintains, a Sruti origin for the domestic ritual
may be inferred on the basis of “performance” (prayoga).® A related
Sitra states at its opening that “Here [are laid out] those rituals [i.e., the
domestic sacrifices] which are known from practice (dcara).”®
What is meant by these references to performance or practice?
There are several possibilities. The texts may be speaking of a localized
body of knowledge and traditional custom which remains outside the
exposition of the ritual in the Sutra. Such a practice would then be the
means for completing the practice outlined in the Grhya Sitra. Support
for this possibility comes from one text’s section on the marriage ritual in
which the Sutra notes that “the ways of different regions and of different
villages are various indeed and should be followed at the wedding. We
will explain here only those that are commonly prescribed.”67
64. This apart from the fact that the Grhya Sutras patterned the rituals they describe
after the Srauta ceremonial. See Chapters 6 and 7 herein; and also Ram Gopal, “Influence
of the Brahmanas on the Grhya Sutras,” Vishveshvaranand Indological Journal 1 (1963):
291-98.
65. ApDhS 1.4.12.10.
66. ApGS 1.1.1.
67. AsvGS 1.7.1-2.
Another possible explanation is the often cited authority of the

“practice of the learned,” that is, of the ritual experts who, in this case,
have mastered the performative details of the domestic ritual. These
experts, “who are properly trained in the Veda and its supplements and
who know how to make inferences from the gruti,”® are to be consulted,
to take one example, regarding the times not specified in the Grhya
Sutra for discontinuation of the study of the Veda.”
As seen in the definition above, these authorities on the domestic
ritual are learned in the Veda and the texts dependent on it, that is, the
entire corpus of instruction on the Vedic sacrifice. It is arguably the
connections and dependent relations that the Grhya Sitras maintain
with these other texts that led one writer to infer a lost fruti source
dealing with the domestic ritual per se.
It is certainly the case that the later commentators on the Grhya
Sutra passages at issue thought that the practice the text referred to as a
supplementary source of information on the domestic ritual was the
practice of the Srauta sacrifices. One commentator explicates the sutra
which reads, “Here [are explained] those rituals which are learned from
practice.” He glosses the text by writing, “So that the domestic rituals
might be intelligible in terms of the Veda, and inferable in terms of well-
known practice, and so that these [domestic rituals] might be properly
observed, there has been a prior explanation of the Srauta rituals [in the
Srauta Sitra of the school].””” The commentator goes on to observe that
the full disclosure of all the details for a complete discussion of the
domestic ritual would involve “too much repetition.” They have already
been set out in the Srauta Sitra and are to be extended to the Grhya
text: “The practice should extend from the parts to the whole.” The
other commentator on the text, Haradatta, speaks of the “continuity” of
the teachings on the §rauta and grhya rituals and of the “dependence” of
the latter on the former.”!
There is, then, a strong sense in which the Grhya Sutra and the
ritual represented in it function as a kind of “subsequent” component
dependent on and following the “precedent” which is the Srauta Sitra of
the school and the ritual with which it is concerned. This relationship is
68. BDhS 1.1.1.6.

69. GGS 3.3.29.
70. Sudargana on ApGS 1.1.1: yat eva dcaranumeyaveddvagamyani garhyani kar-
mani, ata eva tebhyah prathamamanustheyebhyo’pi purvam Srautanam vyakhyanam krtam.
71. Haradatta on ApGS 1.1.1: atrathaSabdena Srautopadesanantaram smartopa-
degam karisyamiti vadan tadapeksamasya darSayati.
highly reminiscent—indeed, I would argue, merely a reduplication—of

the relationship between the paradigmatic sacrifices and their variants as
constituted by the composers of the Sutras, the epistemological wing of
the Vedic ritualist world. Just as knowledge of “prior” and “subsequent”
rituals depicted within a Sttra is dependent on knowing how to extend
what is said about the former to the latter, so is there established a series
of intertextual connections and inferences moving from the Srauta Sitra
to the Grhya Sutra.
These two components of a Kalpa Sutra, taken together, form a
unified and multidimensional body of knowledge on the Vedic ritual
system as a whole. The two types of ritual Sutra divide up their interde-
pendent referents; the samskdras and other domestic rituals presup-
posed by the srauta ceremonial are described in the Grhya Sutras, but
the sacrifices on which many of the domestic rituals were modeled and
assume knowledge of are given only in the Srauta Sutra. The relation-
ship between the rituals (and those who perform them) at the srauta and
grhya levels is the subject of the next two chapters. I wish here only to
observe that the texts describing these two types of Vedic ritual are also
interrelated and interconnected—and in a manner entirely consistent
with the organizational scheme that structures the contents of the individ-
ual Sutra.
Chapters 6 and 7 are exercises in reapplying some of the organiza-
tional principles of the Sutras discussed in this chapter. I will demon-
strate how two major lines of inquiry into Vedic ritualism open up when
imagined in terms of the paradigm/variant model articulated in the St-
tras. In Chapter 7 we will investigate the relationship between the Srauta
and the grhya sacrifices by arguing that the latter are to the former as the
variations are to the paradigm. This will help to underscore the relations
of resemblance the ritualists established between the “subsequent” varia-
tion (the domestic ritual) and the “prior” paradigm (the srauta ceremo-
nial). In Chapter 6, however, I turn the model on its head in order to
take a different angle on the relationship between the two levels of
Vedic ritualism. I will argue that, just as the paradigmatic ritual is usu-
ally a simple and basic version of the more elaborated variations, the
grhya sacrifice is a reduced and elementary form of the more protracted
§rauta sacrifice.
And although the paradigm/variation model will be used in two
different ways in order to bring out two different kinds of features of the
relationship of resemblance between the domestic and the Srauta rituals
(and those who engage in them), in both cases we will not lose sight of
the hierarchical aspect of the relationship.
6
The Organization of Ritual Practice
The Unity of Ritual
A major assumption made in the ritual Sutras, and itself largely assumed
by me in Chapter 5, is that there is what one text calls a “unity of ritual”
(kalpaikatva) extending among the three great classes or levels of Vedic
ritual praxis: the domestic or grhya sacrifice, the Srauta sacrifice (prop-
erly called the haviryajna), and the rituals in which soma is offered and
consumed.! The Vedic ritual is thus conceived as a whole comprised of
three hierarchically ordered and formally analogous levels (within each
class there are seven subclasses).?
For our purposes, however, the three great classes might be co-
alesced into two by collapsing the soma sacrifices into the frauta ritual.
This move may be justified by the fact that the Vedic ritualists discuss
both (but not the domestic ritual) in the same texts, the Srauta Sutras. In
any event, the point I wish to reiterate and follow through new material
is this: the domestic ritual was included within a totalistic system in the
ritual Sutras, it participated in the web of interrelations linking the
components of that system, but it was also the lowest level possible
within the Vedic ritual universe—or, more precisely, within the Vedic
ritual universe as it was mapped out in the ritual Sutras.
1. SGS 1.1.13-15.
2. For this typology of the ritual into three classes, each with seven subclasses or
samsthas, see BSS 24.4; LSS 5.4.23-24; GB 1.5.7, 1.5.23, 1.5.25; VaikhGS 1.1; BDhS
2.2.4.23; GautDhS 8.19-20. BGPariS 1.6.22 concludes an extended metaphor in which the
Vedic ritual is likened to a tree by referring to the three great classes as the “roots”
(miulas), that is, the fundamental categories.
Before the Sutras, the last textual layer of Vedism per se, the domes-
tic ritual is largely undocumented. The Samhitas and Brahmanas ignore
it almost entirely (though not without some intriguing if somewhat am-
biguous references). Because the domestic ritual is not much treated in
these early texts, some have been led to speculate that in this period
the domestic cult was the province of a “popular piety” apart from the
“élite” religion of sacrificial Vedism. The Sutras, insofar as they for the
first time include domestic ritualism in their systematic treatment of the
Vedic ritual in its entirety, might thus represent a kind of “domestica-
tion” of the domestic cult itself.
In this scenario, the Sitras are evidence that the Brahmins envel-
oped the household sacrifices into their own well-established ritual cor-
pus, systematizing it to conform to their own Srauta sacrifice. A. B.
Keith, for example, writes:
The priests . . . appear to have aimed, as time went on, at absorbing en
masse the popular rites and decking them out with their own poetry and
their ritual elaboration. . . . So far from the texts hinting at distaste for the
popular ritual, they rather exhibit the priests determined to secure their
participation in it to the fullest extent, at the expense of the field of action
which first lay open to the head of the family as his own domestic priest.
This is, however, only a hypothesis, and not the only one possible.
As I will note below, the admittedly scanty references to something like
domestic ritualism in the earlier era of the Samhitas and Brahmanas may
indicate that a simplified form of Vedic ritualism, formally similar to
that of the later Grhya Sutras and formally distinguished from other
kinds of Vedic ritualism, was at least known and possibly already system-
atized before the composition of the Sutras.‘ It is, at any rate, just as
possible to posit an origin within Vedic ritualism for the domestic sacri-
fice as to imagine it “absorbed” by Vedic ritualism; but the question
remains unresolved and will not preoccupy us here.
On the other side of things—that is, after the time of the ritual
Sutras—the domestic rites were (again?) largely divorced from the in-
creasingly anachronistic and abandoned Srauta ceremonial. Domestic
ritualism of Vedic coinage survived, and continues to survive, in Hindu-
ism long after the Srauta rituals had been discontinued by all but the
3. A. B. Keith, The Religion and Philosophy of the Veda, 1: 56.

4. See below, pp. 161-67. Mention might also be made of AitB 8.10, which instructs
the purohita of the king to make certain oblations into a domestic fire (grhydgni). Also
note that at least one particular domestic rite of the grhya cycle, the upanayana, was
already outlined in some detail in AV 11.5 and SB 11.3.3 and 11.5.4.
The Organization of Ritual Practice 145
occasional king and pockets of Brahmins in peripheral and isolated areas

of the subcontinent.
When detached from their relations to the frauta sacrifice, domestic
rites could take on new functions and significances (a topic to which I
will return in the next chapter). But in the context of the “unity of ritual”
established in—and perhaps by—the Sitras, the domestic ritual cannot
be properly understood other than in situ.
In this chapter, I will explore the “situation” of the domestic sacri-
fice within Vedic ritualism. I will first present an analysis of the connota-
tions behind one of the principal generic names used in the Sitras for the
domestic ritual as one class of rites in relation to others: the grhyayajna
or “sacrifice of the household.” Next, I look at the domestic ritual, as it
appears in the Sutras, as a feminine counterpart to the masculine proto-
type. I will conclude with a discussion of the meanings behind a second
classificatory designation for this type of Vedic ritual: the pakayajna,
sometimes translated as the “cooked sacrifice” but more often meaning
a “small” or “feeble” sacrifice. Both of these terms (grhyayajna and
pakayajna) 1 will maintain, do not so much refer to a way of sacrificing in
itself but rather to one that has consciously been constituted in relation
to other levels of the Vedic ritual repertoire.
I begin by briefly outlining some of these relations. First and
foremost—and generative of others—was the fact that domestic ritual-
ism was basic to the ritualism of the ancient Vedic Indians. By this I
mean several things. Some of the rituals included under the heading of
the domestic class were prerequisites for the Srauta sacrifices. The
samskaras of childhood and adolescence, and most especially the initia-
tion and marriage rites, were one component of the necessary qualifica-
tions for those who would enter into the next level of Vedic ritualism,
the higher §rauta cult, and thus were chronologically prior.°
Others of the rituals included within the rubric of domestic ritualism
might be considered basic in the sense of being complementary to the
$Srauta ritual. House-building ceremonies, rites relating to childbirth,
domestic animals, and many others complemented rituals at the Srauta
level and would be performed by the Srauta sacrificer as occasions came
5. For a general discussion of the importance of these two samskaras for the Srauta
sacrifice, consult Madeleine Biardeau, Le Sacrifice dans I’Inde ancienne (Paris: Presses
Universitaires de France, 1976), pp. 36-38. See also P. V. Kane, History of Dharmasastra,
II: 190-95. The question of which of the samskaras were required, which were optional,
and when, historically, such decisions began to be made, is largely unexplored to my
knowledge. The Sutras themselves do not appear to address the topic.
up.° The domestic sacrifice was therefore basic also in the sense of being
fundamental to the necessity of properly maintaining the household.
A third way in which the grhya- or pakayajna might be considered
basic to Vedism is in a hierarchical sense. The domestic ritual had prepa-
ratory purposes and domestic functions to serve within the “unity” of
Vedic ritualism. It would appear that the grhya ritual was preeminently
suited not only for certain jobs but also for certain persons who were
also defined by their participation in it. And, we shall see, in form as
well as function, the domestic sacrifice was most often represented as a
condensed counterpart—a paka version—of larger, more complete, and
yet still resembling sacrifices. So, too, therefore, a hierarchical evalua-
tion of those relegated to the domain of domesticity was made in the
Vedic texts.
One might also see the relationship between the domestic and
Srauta rituals, as I suggested at the end of Chapter 5, as a replica-
tion of the relationship between the fundamental (and prior) paradig-
matic sacrifice and the often more complex (and subsequent) varia-
tion (grhya : Srauta :: prakrti : vikrti). And like most cases of the
paradigm/variant relation, the domestic ritual is elemental in the sense
of hierarchically inferior to the more developed ritual (in this case,
the Srauta sacrifice as a whole). The grhya ritual’s association with
home and family (and especially with the wife) served only to re-
inforce a hierarchical inferiority within the Vedic ritual, and within a
socio-ontological system.
Domestic Ritualism and Domesticated Ritualists
The domestic ritual or grhyayajna is, unsurprisingly, so designated be-

cause of its close ties to the domestic realm. It is the cycle of rites oriented
toward the needs and wants of the household; or, as one authority puts it,
the domestic sacrifice is “of the world.”’ The intimate connections among
the household fire, the sacrifices offered in it, the household in general,
and the householder, his wife, and his children in particular, are brought
out clearly in the commentators’ glosses on the Grhya Sutras and in other
6. See, e.g., BGPariS 1.16.1-2: “When he intends to set the [srauta] fires . . . he
extinguishes [the domestic fire] and thereafter prepares this [domestic fire] from ritual to
ritual [i.e., only when necessary].”
7. ApGS 1.2.9.
later writings. Devasvamin, for example, explicating ASvGS 1.1.1 (“Hav-

ing already expalined the Srauta rituals, we will now explain the grhya
rituals”) writes:
The grhya rituals are those done with regard to the home. The word
“home” here is used in three senses: [rituals] concerning the wife, those
concerning the domicile, and [rituals] concerning the stage in life [asrama;
i.e., householdership].8
Other later authorities of the tradition concur, emphasizing one or an-
other of these components of domestic ritualism. Gobhilaputra, who
composed a manual based on the Gobhila Grhya Sutra, explains the
purpose of the domestic ritual in terms of the household (grhya), which
he defines as the wife, sons, and daughters, together with presently
conceived unborn children and unborn children not yet conceived:
“These are considered the household, the kin relations of the sacri-
ficer. . . . Therefore the ritual law [pertaining to the domestic ritual] is
called gryha.”? Commenting on ApGS 2.5.14 (“The domestic fire is
perpetually maintained”), Haradatta Misra writes in a similar vein but
underlines especially the connection with the sacrificer’s wife (which will
be explored in more detail in the next section). “This fire,” writes that
commentator, “set at the time of marriage, is perpetually, that is, eter-
nally maintained for the purpose of rituals closely connected to the
wife.”10
In the ritual texts, the domestic fire (the sala- or grhydgni, the “fire
of the house”; also known as the aupdsana or aupasada fire, the “fire for
worship”; and vaivahikagni, “the marriage fire”) serves as something of
a metonym for domestic ritualism as a whole. It is distinguished from the
ordinary cooking fire of the home (the pacandgni) and is sacred in
relation to any other “worldly” fire (laukikagni). And the Sutras gener-
ally assume that all domestic rituals will be done with this sacred, set-
apart domestic fire.!!
8. grhe bhavani yani karmani tani grhyani. tatrayam grhasabdastrisvarthesu vartte

bharyayam salayamasrama iti.
9. M. Bloomfield, ed. and trans., Das Grhyasamgrahaparisishta des Gobhilaputra
(Leipzig: G. Kreysing, 1881), 1.35—36.
10. esa vaivahiko’gnirnityah sasvatiko dharyah patnisambandhanam karmanamar-
thaya.
11. In some texts, however, certain of the samskadras require separate fires. Many
schools require a so-called sutikagni for performance of the birth and naming ceremonies
(jatakarman and namakarman); see PGS 1.16.23; VGS 2.11; HGS 2.3.4-6; BhGS 1.26;
It is important to note that this fire used for the domestic ritual is to
be established inside the house itself; it is to be placed and then main-
tained on a square mound of earth (called the “home of the fire,”
agnisala) located in the eastern or northeastern section of the family’s
home." The details of this shrine within the home are, unfortunately,
only vaguely described in the Grhya Sutras, but what is noteworthy for
us is, first, its location within the house (reemphasizing the domesticity
of the ceremonial performed with it as the centerpiece) and, second, its
situation in the northeastern quarter. This latter detail is significant
because it, too, reiterates the strong bonds the domestic fire and sacri-
fice maintain with the home and household. It is in the northeastern
corner, according to one authority, where bali offerings are placed daily
for the “deities of the home” (grhadevatas), the “deities of the site of the
house” (dvasadnadevatas), and the “lord of the site of the house”
(avasanapati).'3 The domestic fire, in sum, is located where the invisible
beings most directly involved with the affairs of the house are concen-
trated. The association of the fire (and the ritual cycle it recapitulates)
and the home is in this way made manifest in part by the very location of
the domestic fire.
The time this fire is first established differs according to different
schools,'4 but in all cases it coincides with the attainment of a degree of
independence. The setting of one’s own domestic fire is preconditioned
on and signals a change from the status of a dependent member of
another’s household to an assumption of lordship over some domain of
one’s own, limited though it might be.'> A qualified male, one who has
been initiated and has learned some portion of the Veda, may set his
own domestic fire at the time when he separates from or takes control of
his patriarchal family. This time usually coincides with marriage. The
VaikhGS 3.15. Generally, this fire is extinguished and the domestic fire rekindled before
the conferring of the name on the child (HGS 2.4.8; BhGS 1.26), but see SGS 1.25.11 and
Oldenberg’s note in his translation for a school which expects the sutikagni to be kept for a
full year after the birth of the child.
12. See KhGS 1.2.1; ApGS 2.6.10; BGS 3.4.6, 3.4.9.
13. BGS 2.8.25.
14. For the texts on this subject, consult Kane, History of DharmaSastra, 11: 678-80;
and H. W. Bodewitz, The Daily Evening and Morning Offering (Agnihotra) according to
the Brahmanas (Leiden: E. J. Brill, 1976), pp. 194-95.
15. This point challenges J. C. Heesterman’s intriguing thesis that the domestic fire
symbolizes the householder’s dependence on and antagonistic relations with others in
society. See his “Other Folks’ Fire” in J. F. Staal, ed., Agni, II: 76-94.
fire and the cycle of rites inaugurated at the wedding subsequently act as
palpable signs of the ties between husband and wife and the family they
will produce and also signals new spheres of influence and power for
both man and woman.
For the groom, the lighting of the domestic fire marks a new inde-
pendent status as he separates from his father’s household and becomes
head of his own. Other recommended occasions for lighting a new do-
mestic fire are either when the patriarch dies and the assets of the
household are divided among the male members or when the death of
the patriarch occurs and the family stays together. In the latter case, the
eldest son moves into the vacated position of leadership within the fam-
ily, a move accompanied by the creation of a new domestic fire. He
thereby becomes “chief” of the domain (paramesthin), as one text puts
it,!° or “personally superior” (svayam jydyan) to other members of the
household, as it is phrased in another text.”
Another option, and the one that is apparently favored only by the
schools of Baudhayana and Bharadvaja, concerns the pupil who has
finished Veda study and is about to take the concluding bath of the
graduate and become a snataka (literally “one who has bathed”). He is
to take as his domestic fire the fire he has established at the initiation
and maintained throughout the period of scholarship:
In the fire that he establishes at the time of the upanayana, in that [fire he
offers oblations] at the time of the observance of the vows [for learning
different sections of the Veda]; in that [he offers] at the bath concluding
studentship; in that [he offers] at the marriage ceremony; and in that
[same fire he offers the oblations of] the domestic rituals.!8
In this case, the continuity of the domestic sacrifice stretches over
the entire ritual life of the sacrificer, beginning with the initiation that
16. GGS 1.1.12. See also Bloomfield, Das Grhyasamgrahaparisishta des Gobhila-
putra, 1.77.
7. SGS.1.1.5,
18. And, the text continues, “according to some,” this domestic fire is also the one
used for all optional domestic rituals (kamydni karmani) and the samskaras performed on
the children of the household (BGS 2.6.17—18; BGPariS 1.16.4; see also BGPariS 1.16.15—
16). For similar statements in the Bharadvaja literature, see C. G. Kashikar, “On the
Bharadvaja Grhyasitra and Its Commentary,” Bulletin of the Deccan College Research
Institute 35 (1975): esp. 68: “Bharadvaja prescribes that it is the fire established at the
Upanayana which is to be maintained and on which all subsequent rites including the
marriage ceremony are to be performed. The boy who has undergone the Upanayana-
ceremony should make offerings on his fire every morning and evening, and in the evening
150 REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION ~
constitutes the child into a responsible being and member of society.

Each stage in the ritual process, however, is carefully distinguished from
the following ones as the boy only gradually attains greater degrees of
independence.
The student, under the lordship of his teacher (who in some cases
was also his father), is allowed to sacrifice only with the fuel sticks and
with the minimal ritual utterances known as the vydhrtis (these are the
exclamations “om,” “bhiih,” “bhuvah,” and “svah,” which are said to
represent either the essences of the Vedas or the spatial worlds of the
cosmos). Beginning with the graduation bath and continuing up until the
time of marriage, the snataka is independent of the teacher/father but is
not yet a householder with a family of his own. During this period, the
daily oblations into the fire are only those of clarified butter, again with
the vydahrtis in place of full mantras. Only from the marriage onward,
when the sndataka has been reconstituted as a grhasthin or householder,
is the daily sacrifice made with food oblations, put into the fire by hand,
and accompanied by mantras of direct address to the gods: “To Agni,
svaha! To Prajapati, svaha!”9
In this way, the relative completion of the individual and the degree
of his command over a domain (a /oka or “world,” one might say) is
ritually calibrated by his differing prerogatives vis-a-vis the domestic
fire. Even the newly initiated boy may participate in the realm of domes-
tic sacrificial action, but it is only the married householder who fully
engages in the domestic cult in all its completeness.
The grhyayajna is, by way of summary, broadly characterized by its
particular association with the household and its inhabitants, by its loca-
tion within the home, and by its representational qualities as the sign of
the householder’s lordship over his own domestic domain. But, as I have
said above, the full meaning of domestic ritualism becomes apparent
only when placed in relation to the Srauta ceremonial. When compared
to the Srauta sacrificer, it becomes clear that the grhya ritualist is limited
to the domestic domain and the form of sacrifice appropriate to it, just
as the household fire is literally confined within the home.
he should consign that fire into himself and cause it to descend upon the actual fire the next
day. At his marriage he should cause the consigned fire to descend upon the actual fire,
and perform the rites on it. After the marriage, however, the householder cannot consign
the fire into himself; he must maintain it permanently.”
19. BGS 2.6.19-21 = BGPariS 1.16.5—9.
The Baudhayana Srauta Sitra2° defines the domestic sacrifice as

one which is offered in a fire other than those located in the vihdra, the
sacrificial grounds of the Srauta ritual.?! Vihdra means “separation” and
refers primarily to the separation of the three or five srauta fires that are
enkindled in that space. It is possible, however, that vihdra also con-
notes a domain of ritual action separate from the home, that is, set apart
from—and beyond—the arena of domestic ritualism. The vihdra is typi-
cally near, but not inside, the sacrificer’s house.”
Whereas the domestic sacrificer, who is also called an ekdgni (“hav-
ing one fire”), performs rituals with a single domestic fire located within
the domicile, the ahitagni or Srauta sacrificer performs “expanded” or
“extended” (vaitana) sacrifices with three or five fires located outside the
home.” If the grhyayajna is characterized by its domesticity, the frau-
tayajna is distinguished by this expansive, even imperialistic and coloniz-
ing, spirit. The vihdra is an extension, as well as a superseding, of the
household domain in which the ekdgni is confined for his ritual activities.
As we will see in more detail later, the Srauta ritual is an “extended”
version of the domestic sacrifice, and the domestic is a condensed coun-
terpart of the §rauta, in more ways than one. For now, in addition to the
number of fires and all they represent, mention might be made of the
fact that in the grhya rites the ritual labor done by a multitude of priests
at the Srauta level is handled by the domestic sacrificer himself, with the
optional help of the brahman priest.** The sacrificial gifts or fees
20. BSS 24.4.

21. yacca kim canyatra viharaddhiyate sarvasthas pakayajnasamstha iti.
22. See BhSS 5.2.11; ApSS 5.3.22; and HSS 3.2, where a settled householder (Sa-
lina) is instructed to set his fires at a place outside the house (to the east or north of it,
according to HSS). A frequent traveler (yayavara), however, should establish the fires in a
place inside the house. See Louis Renou’s comment in his Vedic India, trans. by P. Spratt
(Delhi: Indologoical Book House, 1971), p. 99: “The Vedic cult knows no temple. The
ceremonies take place either in the sacrificer’s house or on a plot of ground close by. . .
on which the three altars are arranged.”
23. For the frauta sacrifice as a whole depicted as vaitana, see, €.g., AS&vSS 1.1.1 and
ASvGS 1.1.1. The term especially denotes the “expansion” of the vihara as the Srauta fires
are established.
24. The grhya sacrificer always does the work of the adhvaryu, the priest charged
primarily with ritual manipulations, and according to some texts he is specifically said to
perform also the duties of the hotr (who recites many of the mantras). The brahman priest,
the overseer of the entire sacrifice, is not required at the domestic ceremonial (GGS
1.6.21; KhGS 1.1.23; BGPariS 1.5.5; SGS 1.8.6 with Oldenberg’s note; ASvGS 1.3.6); and
if he is not present, the sacrificer also “does the work both of the hotr and the brahman”
(see GGS 1.6.21, 1.9.8-9; and KhGS 1.1.17-18).
(daksina) owed to this latter officiant, if he is even present, are minimal

as opposed to the significant donations required at the Srauta rituals.*
A married man who has sired a son and who is not yet old (his hair
must still be dark),?° who is of the proper class (the servant class or
Sidras being excluded), who is learned in the Veda, physically fit, and
relatively affluent (he must own at least two cows for the performance of
even the simplest Srauta ritual, the daily agnihotra),”’ may establish the
three or five frauta fires and perform sacrifices in them. He thereby
becomes an Ghitadgni, “one who has laid [the srauta] fires.”
These Srauta fires are first established by expanding, although with
intermediaries, the domestic fire (grhydgni) into the “householder’s
fire” (garhapatydgni),*8 which is said to be five times more powerful.”
This new fire, like the old domestic fire from which it is kindled and
which it now supersedes, must be perpetually kept lit from this time
forward.*° From the gdarhapatya fire embers are carried to the places of
the other two Srauta fires, the Ghavaniya used for sacrifices to the gods
and the daksina or “southern” fire employed for offerings to the ances-
25. See GGS 1.9.10—11: “A full vessel is the lowest daksind at a domestic sacrifice.
The highest is unlimited.” See also MGS 2.2.28; KhGS 2.1.30; JGS 1.4; and GGS 1.9.6.
BGS 1.4.38 prescribes the gift of a cow; a horse or a cow is to be given according to VGS
1.38; and AsvGS 1.10.27 suggests only the remnant of the oblation material as daksind.
26. BDhS 1.2.3.5: “A Sruti passage says, ‘Let him light the Srauta fires while his hair
is [still] black.’ ” See Kane, History of DharmaSastra, I: 350-51.
27. Kane, History of Dharmasastra, p. 979: “The daily agnihotra required the main-
tenance of at least two cows, besides thousands of cow-dung cakes and fuel-sticks. For the
maintenance of agnihotra and the performance of darsapurnamasa (in which four priests
were employed) and the Caturmasyas (where five priests were required) the householder
was required to be well-to-do.” Some indication of the enormous expenses incurred by the
sponsor of the greater, and optional, Vedic rituals may be gathered by studying the
operating budget for the 1975 performance of the agnicayana in Kerala. Leaving aside
costs for the foreign observers, the sacrifice entailed expenditures of Rupees 160,810. See
Staal, “The Agnicayana Project,” in Agni, II: 457-63. A vajapeya was done in Poona in
1955 for approximately a quarter of the cost of the agnicayana, still a hefty price tag far out
of the reach of any but the very wealthiest and most powerful. For other qualifications
constituting the competency (adhikara) for one who would light the Srauta fires, see
Chakrabarti, The Paribhasas, pp. 142-47.
28. A succinct description and interpretation of the setting of the srauta fires may be
found in Heesterman’s “Other Folks’ Fire.”
29. See MSS 8.23.1-3, where the wife of a deceased ahitagni takes “a fifth portion”
of the garhapatya fire (before it, along with the other frauta fires, is used to cremate the
sacrificer and then is extinguished) as a new domestic fire. “She should tend this new
domestic fire according to the pakayajna procedure” (MSS 8.23.7).
30. One authority, KSS 4.13.5, requires the perpetual maintenance of all three fires
by the gatasri, “one who has become rich.” See also ApSS 6.2.12.
tors. In this way the continually maintained garhapatya fire is daily

“extended” or “spread out” (vi + tan) into the other two fires.3!
Notice here the continuity of the fire as one takes the single house-
hold fire, transforms and empowers it, and then reestablishes it as a
“householder’s fire,” the gdrhapatya. The old domestic fire may now be
extinguished and rekindled only when needed for special circumstances
(as noted above), or, according to some authorities, the srauta sarificer
may, if he wishes, continue to offer the regular domestic sacrifices in the
daksina fire.*2
As Heesterman points out, in many respects the “southern” fire of
the Srauta ritual “stands in opposition to the others. In contradistinc-
tion to the garhapatya and dhavaniya, which are situated on a line
running from west to east, the heaven-going direction, the daksinagni
is placed to the south. It marks the north-south direction that is con-
cerned with the fathers, i.e., predominantly with men.”33 In this way,
31. In addition to these three Srauta fires, some schools require two more of the
ahitagni. These are the sabhya (“the fire of the house of assembly”) and the dvasathya
(“the fire of the rest house”). See Alfred Hillebrandt, Vedic Mythology, 2 vols., trans. by
S. R. Sarma (Delhi: Motilala Banarsidass, 1981), II: 81-82; and Frederick Smith, “The
Avasathya Fire in the Vedic Ritual,” Adyar Library Bulletin 46 (1982): 73-92.
32. See VGS 1.6. According to HGS 2.6.16.2, the annual sacrifice to the serpents
(Sravana) should be done either with the domestic fire or on the daksinagni if the sacrificer
is an ahitagni. See KGS 47.1-2, where it is said that the domestic sacrifices are to be
performed either in the “wedding fire” (agni vaivahana), which can only be the domestic
fire, or in the aupdasada, which the commentators gloss as the Srauta daksinagni. The
relationship between the two fires—at two different levels of Vedic ritualism—is indeed a
close one, as Heesterman has pointed out in “Other Folks’ Fire”, p. 83.
33. Heesterman, “Other Folks’ Fire,” p. 83. On pp. 81-82 of that article, Heester-
man writes that “the fires for the Srauta ritual . . . form in a complicated way the extension
of the domestic fire.” He emphasizes, however, the connection of the domestic fire not
with the garhapatya but rather with the daksindgni, and highlights those texts that provide
the option of kindling those fires from embers taken from another’s fire. He thus sees the
fires as a symbolic link to social life in general in “a time of scarcity.” In a “preclassical” or
“agonistic” period, “the daksindgni is associated with both the domestic fire and with
others’ fires. The meaning is, quite simply, that for the business of life one needs the
others.” But this dependence involves conflict, and “thus the social aspect of the fire is
inextricably bound up with strife and violence” (p. 86). The continuity of the domestic and
§rauta fires and the ties they both maintain to social life thus presents a problem, according
to Heesterman, and one that was solved in the “classical” period in part by the radical
separation of the grhya and Srauta fires: “Now . . . the Srauta fire had to be permanently
attached to and identical with the sacrificer. This meant, in the first place, that the Srauta
fire had to be rigorously dissociated from the domestic, which inevitably remained bound
up with marriage and household—and hence with society” (p. 92). This interpretation
obviously runs counter to the one put forward here and tends to ignore the “domestic” and
the domestic ritual is effectively and symbolically encompassed within

the §rauta ceremonial.
But the continuities between the two realms of Vedic ritualism,
represented by the continuities between the fires with which they are
performed, should not cloud the hierarchical distinctions between them.
The garhapatya fire is often equated with the home,* with this world,*
and with things “related to the human” (mdnusya)**—just as is the do-
mestic fire. But as the superordinate and individualized version of the
comparatively weak grhyagni with its homey associations, the garha-
patya fire is also homologized to the earthly realm in its entirety (and not
merely to the householder’s domestic domain), while the other two
§rauta fires are connected to the cosmological worlds of atmosphere and
heaven.*” Extending the fires is thus an extension of the Srauta sacri-
ficer’s sphere of influence, an enlargement of the ahitagni’s lordship and
domain of ritual action. With the Srautayajna, one “conquers” or “wins”
all worlds; with the grhyayajna, one is circumscribed and domesticated.
The complementarity and continuity of the two does not obscure the
hierarchical relationship that constitutes them into inferior and superior
forms of Vedic ritualism.
The Feminization of the Domestic Ritual
With the exception of the two schools of Baudhayana and Bharadvaja,

whose position was delineated above, the Vedic ritualists held that it was
with marriage that domestic ritualism within Vedism was inaugurated. It
will be recalled that one of the names for the domestic fire was the
vaivahikagni, the fire relating to marriage, and many of the Grhya Sa-
tras underline the importance of the wedding by including within their
descriptions of that rite of passage the rules for the standard procedure
(tantra) common to all domestic sacrifices. The marriage ritual, in other
words, acts as a paradigm for all other sacrifices at the domestic level;
“social” aspects of the “classical” srauta cult. I have tried to keep these connotations intact
with an expandable conception of the terms household, domain, sphere of influence, and
lordship.
34. SB 1.1.1.18-19.
352) TSO 1285; 6245255 ete:
362°7S L.6:7 le
37. The brahmanas on this point are brought together by Hillebrandt in Vedic
Mythology, Il: 66-69.
and whereas the initiation transforms the child into an Aryan, it is the
wedding that establishes the male as something of a free agent, indepen-
dent of teacher and father, with a household domain under his umbrella.
The wedding ceremony also transforms the socio-ontological status
of the bride. Marriage, as it is conceptualized by the authors of the ritual
Sutras, is for women the counterpart of the initiation or upanayana;> it
draws her into a relationship with the Veda and with fire sacrifice
through the bond to the new husband. As the young boy gains access to,
and a modicum of control over, ritual activity only by first placing him-
self within the socio-ontological realm of the teacher, so does the
woman, by taking a place under the authority of her husband, both
subordinate herself to the lordship of another and connect herself to the
domestic fire and the rituals done in it.
The importance of the wife to the ritual is not unacknowledged by
the Sutra writers. Her presence is required for the proper performance
of both the domestic and Srauta sacrifices of her husband and, indeed, is
regarded as an essential part of the householder’s very being: he is
“completed” by her.*? She is integral to the householder’s lordship, and
together “they two become yoked like oxen in the sacrifice.”
The particularly close association of the wife and the domestic rit-
ual, touched on above, also serves to diminish its relative importance in
the eyes of the (male, need it be said) ritualistic writers. The “feminiza-
tion” of the domestic sacrifice, implemented through the connections
the ritualists drew between it and the wife, had as one of its purposes to
denigrate it in comparison to the Srauta ritual.
Let us first briefly review the ritual of lighting the domestic fire at
the wedding ceremony. The fire for this latter ritual—which will thereaf-
ter serve as the couple’s domestic fire—is kindled by rubbing together
two sticks of wood; or, alternatively, it is taken from the domestic fire of
a Vaisya, thus reemphasizing its connection to the earth, fecundity, and
38. See, e.g., Manu 2.67: “The rule relating to the wedding is stated to be the
samskara of the Veda [i.e., the initiation] for women; serving her husband is [the counter-
part of] the residency with the teacher; and maintaining the home is [the counterpart of]
the worship of the fire.”
39. See Manu 9.45 and BAU 1.4.17. See also Paul Mus, “The Problematic of the
Self, West and East, and the Mandala Pattern,” in C. A. Moore, ed., Philosophy and
Culture, East and West (Honolulu: University of Hawaii Press, 1960), esp. p. 601; and
Orlan Lee, “From Acts—To Non-Action—To Acts: The Dialectical Basis for Social With-
drawal or Commitment to This World in the Buddhist Reformation,” History of Religions
6 (May 1967): 289-92.
40. TB 3.7.5.11.
domesticity; or, as yet another option, it is first established from embers

borrowed from the fire of a Srotriya, a learned ritualist who maintains
the three Srauta fires and who is famous for his sacrificial achieve-
ments.*! The householder’s domestic fire is set up at this time either by
performing a simple rite as part of the wedding ceremonial or by the
performance of a condensed version of the ritual known in the Srauta
ritual as the Agnyadhana.*? When the wedding ritual is concluded, this
fire is encased in a vessel, and the newly wedded couple carry it with
them to their home.
From that time forward, the fire is perpetually maintained accord-
ing to all the Grhya Sutras. Although it is primarily the duty of the
householder himself to offer into this fire on a daily basis, thereby
fulfilling the obligation to “worship” (upasthdna) the fire continually for
as long as he lives, some texts also allow other members of the house-
hold to offer the daily sacrifices in this fire that is so closely bound to the
home and its inhabitants: “Beginning from marriage, he [the house-
holder] himself should worship the domestic fire, or his wife, his son, his
daughter, or his pupil.” If the fire should go out, the text continues, the
wife must fast as part of the expiation and preliminary to the reestablish-
ment of the domestic fire.
In another Sitra, the teacher first follows the more conservative
path and prohibits the wife and uninitiated sons from offering the obla-
tions, but then adds as an equally valid option a rule that allows the wife
to make the offerings (without recitation of the mantras).“4 Such is also
the ruling according to unnamed ritual teachers quoted by Gobhila
Grhya Sutra and Khadira Grhya Sutra; the wife may by herself perform
the twice-daily Agnihotra sacrifice at the grhya level: “For the wife is the
41. Fora survey of these and other alternatives for generating the wedding/domestic
fire, see Heesterman, “Other Folks’ Fire,” pp. 79-80. The author rightly notes that “these
alternatives all have one feature in common: the fire has to be procured from elsewhere, or
rather from someone else, whose willing or unwilling cooperation is therefore necessary.
And, strangely, it is nowhere said that one should continue maintaining one’s paternal or
ancestral fire. In fact, there does not seem to be such a thing as an ancestral fire, because
the fire ends with the life of the householder and is last used in his cremation. After that his
son has to set up his own fire.” In this instance, and with the reservations expressed above,
we may agree that “here, in ritual terms, we encounter the paradox we already noticed,
namely, that of permanence and instability, continuity and discontinuity” (pp. 80-81).
42. For such replicas at the grhya level of certain Srauta rituals, see Chapter 7 herein.
43. AsvGS 1.9.1; see also SGS 2.17.3. The issue of whether anyone other than the
householder was qualified to make oblations in the domestic fire was, however, controver-
sial. See Kane, History of Dharmasastra, U1: 683-84.
44. BhGS 3.12.
house (grha) and this fire is domestic (grhya).”45 In many texts, the
agnihotra is followed by a distribution of rice balls to various household
deities, spirits, and other beings. Some of the authorities specify that the
evening distribution is to be done by the wife—again, without the recita-
tion of Vedic mantras.“
To appreciate the significance of the fact that the relationship be-
tween the wife and the domestic ritual was regarded as so strong as to
persuade some teachers to allow the wife to perform the sacrifice on her
own, let us turn to the Srauta ritual. There are some rites that the wife
performs in that ceremonial under the guidance of the priests, but the
Srauta Sutras are adamantly opposed to the notion that she is capable of
any independent ritual action. In the Srauta sacrifices, the wife is a
necessary but dependent performer; she participates in them only con-
jointly with her husband.*’
There are, however, some unusual circumstances under which even
in the Srauta ceremonial the wife is empowered to offer the rituals on her
own. A look at these exceptions, however, only proves the rule: it is the
domestic ritual that is more properly the field of action of women, in
addition to domesticated men.
The occasions in which the wife may carry out sacrificial perfor-
mances ordinarily assumed by the a@hitagni are when he is away on a
journey, when he is physically incapacitated, or when he is dead. The
first two of these circumstances are covered in the following citation
from the Manava Srauta Sitra:
Together do these two [husband and wife] constitute a couple; together
do these two maintain the sacred fires; together do these two procreate
and obtain progeny. The eastern world pertains to the husband, the
western world to the wife. Since the wife takes upon herself the responsi-
bility of maintaining the vow of the husband who has gone out on a
journey or who has become incapacitated, therefore she is entitled to
half [of the ritually acquired merit]. Women are [in these instances]
qualified to perform sacrifices.
The text goes on to say that in the srauta sacrifice, “whatever one offers
on the dhavaniya fire relates to the sacrificer, and whatever is offered on
the garhapatya fire relates to the sacrificer’s wife.” In this, the tag line
justifying the woman’s qualifications to assume the sacrificial vow when
45. GGS 1.3.15; KhGS 1.5.17.

46. GGS 1.4.5; VaikhGS 6.17; Manu 3.121; see also §GS 2.17.3.
47. For this point, consult Chakrabarti, The Paribhasas, p. 142.
48. MSS 8.23.10-14.
her husband cannot (it is half hers), the portion of the sacrifice which is
specifically the wife’s is that which is offered in the superpowered Srauta
version of the domestic fire, the garhapatya.
The death of the ahitagni or his wife provides another situation in
which the characteristic linkages of both to certain of the fires which
represent their spheres of influence is made clear. Here, once again, the
texts present domestic ritualism as the special province of the woman.
Upon its demise, the body is cremated as a final oblation into the fire or
fires kept during one’s lifetime. In the case of the domestic sacrificer
(the ekdgni) who dies before his wife, the procedure is as follows: he is
burnt in the domestic fire he has faithfully maintained, and when that
fire has finished its last duty, it is allowed to go out, never again to be
rekindled. The texts do not make specific references to any subsequent
ritual responsibilities for the widow of a domestic sacrificer. Presumably
there is none. She takes her place within the household of a son or
another male relative and at the time of her own death is cremated in a
fire set especially—and exclusively—for this purpose, a fire regarded by
the authorities as “worldly,” that is, secular.
The case of the death of a Srauta sacrificer or his wife is more
complicated. The general rule is that whoever dies first is to be cremated
in the combined three fires. Both husband and wife, as we have seen,
are intimately tied to these fires, and both have claims to their efficacy.
So it is that should the wife of an adhitagni precede him in death, she is
cremated (together with the sacrificial implements) in the Srauta fires,
which are then extinguished.
And what does the surviving husband do when his fires have been
thus used up? Baudhayana Pitrmedha Sutra* gives several alternatives,
all of which emphasize the importance of the domestic fire for the conti-
nuity of the household even after it has suffered the loss of the female
partner. One option is to rekindle new Srauta fires out of the domestic
fire of the first marriage. If this domestic fire has not been maintained
after the time of establishing the three frauta fires, it is rekindled from a
portion of the old garhapatya fire—before, of course, the latter has been
used in the cremation ceremony for the wife. Having obtained a domes-
tic fire (either employing the one he has always kept or reigniting one he
had left behind), he may remarry, using the domestic fire at the second
wedding. With his new wife, he may then generate new Srauta fires out
of the old domestic one.
In this text, interestingly enough, there is no mention of the possibil-
49. BPS 2.4.2-17.

ity of the widowed ahitagni “downgrading” himself to the level of the

domestic ritualist. Having once set the Srauta fires, he can no longer go
back to the grhya ceremonial exclusively. He must either remarry and
reset the three fires or discontinue life as a householder altogether and
renounce the world and all sacrifices other than those done purely medi-
tatively. The domestic cycle of rituals, it would seem, is too intimately
associated with ‘a particular household, and a particular wife, to be
continued when that wife has passed away.
The text also covers the case of the wife whose ahitagni husband has
died. Unlike the widower, the widow is not allowed to remarry; there is
therefore no possibility of reestablishing the Srauta fires and continuing
sacrifices of that class. But also unlike the widower, the surviving wife of
an G@hitagni is permitted to continue, on her own, the sacrificial rituals at
the domestic level. She is to maintain the domestic fire she once shared
with her husband and perform the rituals so closely related to the house-
hold. Before having her husband cremated in his three fires, the woman
takes a fraction of the gdrhapatya, returns it back into a domestic fire,
and performs the regular cycle of grhya sacrifices in it until the end of
her life. This option is given in the Manava Srauta Sutra and is attributed
to a teacher known as Maruka:
Sakalya says that after the husband has died a woman becomes devoid of
sacred fires and is unfit to perform any sacrifice. But Maruka has ex-
plained how she should tend a fifth portion [of the garhapatya fire] with
proper service. If the wife [of the deceased] is observing the contemplated
vow [i.e., to maintain the fire sacrifice], the officiating priest should say to
her, “Attend upon the domestic fire.” . . . She should tend the domestic
fire according to the procedure of the domestic sacrifice.
Wearing soiled clothing and with other signs of her widowhood, the
woman is instructed henceforth to make daily oblations of clarified but-
ter (and not the oblations of grain that are usually prescribed) into the
domestic fire. At the end of her life, she is cremated in the domestic fire
she has in this way maintained.*
The “feminization” of the domestic ritual—that is, its bonds to the
wife, especially when compared with the more “virile” Srauta sacrifice—
highlights both its domesticity and, in a patriarchically oriented society,
its relative limitations. The ceremonial performed in the home and with
50. MSS 8.23.1-3, 8.23.7.

51. This is from BPS 2.4.3. The Manava text requires that she be cremated in a
special fire set for that purpose.
the domestic fire is so simple and so basic that even those, who in the
eyes of the ritual authorities, had minimal competency for Vedic ritual-
ism could carry it out. The ekdagni is put on a par with the wife of the
ahitagni. The domestic ritualist’s one fire is regarded as the equivalent of
but one-fifth of the “householder’s fire” at the Srauta sacrifice. And this
garhapatya fire is itself relegated to oblations “that pertain to the sacri-
ficer’s wife” and are thus encompassed within the larger ritual domain of
the sacrificer.
Basic Vedic Ritualism
Another generic term for the domestic sacrifice as a type, in addition to

the grhyayajna, was pakayajna. And this, too, was a designation de-
signed to indicate the relative inferiority of this class of rituals in relation
to other, more complete, and more realized types.
Whereas the “sacrifice of the home” was contrasted to the “ex-
tended sacrifice,” the pakayajna or “small sacrifice” was so called be-
cause there were mahdyajnas or “great sacrifices.”>? The nomenclature
highlights the elemental and basic nature of domestic ritualism within
Vedism; the padkayajna was conceived and represented as a diminutive,
condensed form of its larger model, the Srauta ritual.
The word paka in this context is not to be derived, as is sometimes
done, from the verb pac-, “to cook, bake, or roast.”*} Rather, the term
here is formed from the root pa-, “to suck,” with the diminutive -ka
suffix.*4 It is thus best translated as Manfred Mayrhofer has suggested:
“pakah: einfach, arglos, rechtschaffen, unwissend; sehr jung, m.
Tierjunges/simple, honest, ignorant; very young, m. the young of an
animal.”5
52. I use the term here only in its literal sense. As a technical term, it is used to
classify the aSvamedha (TS 2.2.7.5; KS 10.9; MS 2.2.9) or the soma sacrifice (TS 3.2.2.2.;
SB 2.4.4.14), among others of the major Srauta rituals. The mahdyajnas were reincarnated
in five simple domestic rites in later texts, as we shall see in Chapter 7 herein.
53. See, e.g., Hermann Oldenberg’s translation of “pakayajna” as “boiled offering.”
The Grihya Sitras, 1:xxiii. Pierre Rolland offers “sacrifices de maturité” in his translation
of the VGS, playing on the two possible meanings of the word paka.
54. V. V. Bhide, The Caturmasya Sacrifices (Poona: University of Poona, 1979), p.
210, citing Bhattabhaskara’s commentary on TB 1.6.6.
55. Manfred Mayrhofer, Kurzgefasstes etymologisches Worterbuch des Altindischen
(A Concise Etymological Sanskrit Dictionary, 4 vols. (Heidelberg: Carl Winter Universi-
tatsverlag, 1957), II: 243.
The word employed in this sense can be traced to the earliest texts
of the Vedic tradition.* It often appears as semantically in opposition to
various terms for “knowledgeable.” Different forms of pdka are con-
trasted with dhira (“intelligent, wise, expert”),’ grtsa (“clever, judi-
cious”),°* and vidvas (“knowing, learned”).°° Elsewhere, paka is glossed
by various terms meaning “ignorant.” At RV 10.7.6, for example, pdka
is juxtaposed to aprecetas (“foolish”), whereas at RV 1.164.5-6, it is
found together with avijanant (“ignorant”) and acikitvas (“unknowledge-
able”). Finally, an interesting context in which one finds paka in the
older texts, and one which foreshadows its later use as a technical term
within the ritual vocabulary, is RV 10.2.5, where the term is placed next
to dinadaksa (“a defective ability to act”).
Given this etymological and semantic history, the designation of
certain kinds of rites and rituals as pdaka is readily explicable. It is with
the connotations of “simple, small, uncomplicated, feeble, and weak”
that the ritualists composing the Sitras deployed the word to classify the
domestic sacrifice: “The word pdakayajna is used to describe these rituals
because of their smallness or feebleness (hrasvatva), for ‘paka’ indeed
means ‘small’ or ‘feeble’ (Arasva).”© Later commentators on various
ritual texts in which the term comes up concur in glossing pdka in these
contexts with the synonym alpa (“little, easy”).®
Although, as stated above, there is no systematic treatment of the
domestic ritual in the texts before the Sutras, there are a few references
to pakayajna. And in all such cases, the term is employed to label certain
simplified rites within the Srauta ritual which have direct connections to
domesticity. One example of this is when the grhamedhiya rite of the
caturmasya sacrifice (in which offerings are made to the Maruts who are
called “householders”) is referred to as a pakayajna,® in part because of
its associations with the household and domestic animals and in part
because of its abbreviated format.® In the grhamedhiya rite, which is a
56. For paka and its derivatives in the RV, consult Louis Renou, Etudes védique et
paninéennes, 17 vols. (Paris: Publication de ICI, 1955-69), IV: 129, VII: 90, 93, XII: 6.
57. RV 1.164.21, 2.27.11, 10.1.18, 10.86.19, 10.100.3; AV 4.19.3, 10.1.18; TS
rile ay
58. RV 4.5.2, 10.28.5.
59. RV 1.31.14.
60. VGS 1.2-3.
61. E.g., Sayana on TS 1.7.1; Haradatta on ApGS 1.2.9; Devasvamin on AsvGS
1.1.3; and Srinivasa on JGS 1.1.
62. TB 1.6.6 simply calls this rite pakatra (“concise”).
63. MS 1.10.15 and KS 36.9. See also Bhide, The Caturmasya Sacrifices, pp. 89-91.
pakayajna, certain rites from the usual paradigmatic procedure of the

§rauta ceremonial are dropped: the fore- and after-offerings which sur-
rounded the principal oblation, and the so-called kindling verses recited
by the hotr priest (the sémidhenika). In its discussion of the darvihoma,
which is not accidentally another name used by the Grhya Sutras for the
domestic ritual as a class,“ the Katyayana Srauta Sutra characterizes
that type of rite by noting the same adjustments, and adding others, to
the normal Srauta ritual procedure:
Darvihomas are to be done in a single fire. They do not have fore- and
after-offerings, and do not have the “kindling verses.” There are no calls
of instruction by the adhvaryu priest (the nigadas) in darvihomas. They
are to be offered by uttering the name of the deity [in the dative case] with
the exclamation “svaha!” [and not “vasat!” as in the usual srauta format].°
These references, in the texts concerned with the higher levels of

Vedic ritualism, to a condensed and simplified pakayajna or darvihoma
within the Srauta ceremonial itself foreshadow the descriptions given in
the Grhya Sutras of the domestic ritual. It is surely no coincidence that
the very same features of the Srauta pdkayajna and darvihoma are those
characteristic of the domestic ritual as it is systematized in the Grhya
texts. The domestic ritual is depicted there as also employing but one
fire, and the ritual framework (tantra) for such classes of Vedic ritual is
an adaptation and condensation of the §rauta new and full moon sacri-
fice (which supplies the tantra for all other Srauta sacrifices). The par-
ticular omissions specified in the Grhya Sutras are the exclusion of the
fore- and after-offerings and the dropping of the “kindling verses” and
“calls of instruction,” and some texts also specify the omission of the ida
rite in the domestic procedure.
This last detail is of some interest, for, oddly, in the earlier texts®
this ida rite (in which the oblation material is divided into portions that
are distributed among and subsequently consumed by the officiating
priests and sacrificer) is also labeled a pakayajna.® In this instance, as
64. E-g., BGS 1.4.39; HGS 1.1.3.1.

65. KSS 6.10.20-24.
66. See, e.g., ApGS 3.7.23: “By the ritual procedure relating to the new and full
moon sacrifice, other rituals are explained.”
67. E.g., AitB 3.40; TS 1.7.1.1; SB 1.7.4.19, 1.8.1. 1ff.
68. The specific exclusion of the ida from the domestic ritual (e.g., SGS 1.10.5)
might be understood in the same way as one understands the rule that the domestic fire is
to be extinguished when it has been reconstituted into the Srauta garhapatya fire. The ida,
like the garhapatya, is the representative of the pakayajna in the Srauta ritual, but it is a
supercharged representative. It therefore transcends domestic ritualism because it repre-
sents it in the higher srauta ceremonial.
with the grhamedhiya rite, the direct associations with food, cattle, fe-
cundity, and the home—that is, with the concerns of the domestic
realm—seem to lie behind the designation. And also connoted by the
label is the fact that the ida rite is simple, diminutive, feeble, and infe-
rior when compared to other rites of the §rauta sacrifice.
The great commentator on Vedic texts, Sayana, explains the link
between the ida and the pdkayajna by emphasizing the eating of and
connections to food in both.” The connection to food is, however, only
one of a series of associations between the ida in particular and the
pakayajna as a class—a network that includes as nodes food, cattle,
fecundity, the house, and all things human and earthly. Thus, for exam-
ple, one text explains the fact that the oblation is consumed rather than
offered in the ida rite by drawing on the equally important connection
between the rite and animals: “They [the priests and sacrificer] eat it and
do not offer it into the fire lest they throw the animals into the fire.””
Ida is not only a rite in the Vedic ritual; she is also a goddess whose
essence is projected into the rite which is her namesake.” And, follow-
ing the chain, the ida rite is, in its turn, the essence of the pdkayajna:
“He invokes the goddess Ida. The pdkayajnas have as their resembling
form the ida rite (idavidha). Through the ida rite whatever pdkayajnas
there are are included within the soma sacrifice.””? The mythology of
this goddess, then, may be instructive for our understanding of the
pakayajna in the Brahmanas.
This goddess’s creation is related as part of the story of the first
man, Manu, who is deluged by a great flood and is saved from drowning
by a fish he had previously cared for.” The fish pulls an ark with Manu
in it to a mountaintop, rescuing Manu (and the human race as a whole)
from a watery doom: “The flood swept away all creatures and Manu
alone remained here.”
Manu, reprising the role of the creator god Prajapati, desires off-
spring and deploys the usual Vedic mythological methods for unilateral
procreation: austerity and sacrifice. The latter, according to the text,
was a “pdkayajna consisting of an offering of ghee, curds, whey, and the
mixture of coagulated and fresh milk, placed into the waters.” After a
year passes, the solidified form of the oblations, the goddess Ida, arises,
69. Commenting on SB 1.7.4.19: pakayajnesu hutaseso bhaksyate, idahutaseso

bhaksyate; tena pakayajnarthetyucyate.
70. SB 1.8.1.38.
71. SB 1.8.1.11.
72. AitB 3.40.
73. SB 1.8.1.1 ff.
and the text depicts her as the daughter of Manu. She addresses her
progenitor by saying:
Those offerings of ghee, curds, whey, and the mixture of coagulated and
fresh milk—which you made in the waters—with you and them you have
engendered me. I am the blessing [procured from the sacrifice] and there-
fore am suitable for use in the [Srauta] sacrifice. If you use me in the
sacrifice, you will become rich in offspring and cattle.
Manu then invokes Ida at the sacrifice, incorporating the idd rite
within the procedure, and thus he “generated this human race, the race
of Manu.”
Ida, the goddess of nourishment; idd, the rite in which food from the
cooked oblations are consumed by humans; and the pdkayajna, the first
sacrifice of Manu and the womb of the goddess—all are in this text (and
elsewhere) associated with sustenance, fecundity, and cattle.” All are
also connected to humanity, to the sphere of Manu and his race—that is,
to the earthly realm in contradistinction to the heavenly worlds of the
gods. Ida is the daughter of Manu and is thus described as mdnavin,
“belonging to humans.” She is one of three goddesses ruling the three
worlds: Bharati oversees the highest realm of heaven, Sarasvati looks
after the intermediate region, but it is Ida who rules over the bhiloka, the
earthly realm, and exerts influence over offspring, cattle, rain, food, and,
in sum, all aspects of domestic life.”
In other situations within the srauta ritual, the texture of strands
connecting Ida to the pakayajna also displays both as inextricably bonded
to the domestic arena. In the agnihotra, for instance, Ida is explicitly
related to the house (the locus of the padkayajna from which she was
created) in addition to other connections to offspring, cattle, and so on.
One of the prayers included within the litany of the daily worship of the
fires” is to be recited while looking at the house.” The brahmana on this
74. For other texts on the goddess Ida, consult S. K. Lal, Female Divinities in Hindu
Mythology and Ritual (Pune: University of Poona, 1980), pp. 31-40; and J. R. Joshi, Some
Minor Divinities in Vedic Mythology and Ritual (Pune: Deccan College Postgraduate and
Research Institute, 1977), pp. 50-54.
75. TB 2.6.10.4, 3.6.13.1, etc. For the reincarnation of Ida as the later Hindu
goddess Annaptrna—a buxom deity of nourishment significantly depicted iconographi-
cally with a darvi ladle in her hand (the principal utensil for offering the pakayajnas)—see
Lal, Female Divinities, p. 40, with the plates on pp. 276-78.
76. TS 1.5.6.1: “I gaze on offspring, the offspring of Ida who belongs to Manu; may
they all be in our home.”
77. BSS 3.8; ApSS 6.16; KSS 4.12.1-4; SSS 2.11.2-5; MSS 1.6.2; see also BhSS
6.2.5:
passage reads, “ ‘I gaze on offspring,’ he says; truly he [thereby] wins all

the domesticated animals.”78 Another brahmana, this one on the ida rite
itself, identifies both the rite and the man with domestic animals: “[With
the ida] he firmly established domestic animals in domestic animals.”
Taken together, the constellation of connections indicates that the ida,
and therefore the pakayajna which is encapsulated in that rite, was
regarded as ritual activity most especially concerned with earthly and
domestic prosperity.
In yet another context, the series of interrelations among the
pakayajna, food, cattle, fecundity, the home, the human, and the this-
worldly is restated through a focus on another related element: the
VaiSsya class. As part of the consecration vow for a soma sacrifice, sacri-
ficers of the different classes consume different substances which corre-
spond to their socio-ontological natures. The Brahmin drinks milk, for it
is said to be full of luminous power; the Ksatriya consumes rice mash
(yavagu), “for rice mash is harsh, and is the form of the thunderbolt.”
The Vaisya, the agriculturalist and tiller of the land, ingests amiksd,
curds and milk, for this substance is declared the “form of the pakayajna
and is for obtaining earthly prosperity (pusti).”8°
Just as the Vaisya is the lowest of the three twice-born classes, the
pakayajna is at the bottom of the ritual chain. The hierarchical inferior-
ity of this class of Vedic sacrifices is also brought to the fore in texts
dealing with the ida, which represents the padkayajna as a whole.
As we have seen, the idd is linked to Manu, the first human, and is
thereby contrasted to the more efficacious and totalistic sacrifice of the
gods: “With a whole sacrifice (sarvena yajfiena) the gods went to
heaven. Manu labored [here on earth] with the pdkayajna and Ida went
to Manu.”®! The earthly and relatively incomplete ida rite within the
“divine” §rauta ceremony is thus a bit problematic: “Those who, in the
middle of the sacrifice, perform the ida rite, break it apart and injure it.”
The brahman priest, the physician of the sacrifice, is called upon to heal
and restore to full strength the ritual which has in this way been weak-
ened by the inclusion of the ida.8* Another passage from that Brahmana
uses the adjective pakayajniya (“like a pakayajna’) to depict the ida rite
as the weak link in the sacrificial chain: “This [part] of my sacrifice, this
78. TS 1.5.8.1.
79. TB 3.3.8.2.
80: TS 6:2.5.2-4.
SL LES 1:73:
82. SB 1.7.4.19.
ida rite which is like a pakayajna, is the very weakest (tanistha). The
demons must not injure my sacrifice at this time.”
Other early references to a pdkayajna in the Samhitas and Brah-
manas also tend to underwrite its connotations in the Sutras as the most
basic and hierarchically inferior ritual within the system. Yajnavalkya,
the iconoclastic teacher of the White Yajur Veda, is quoted in one text
as stating that the daily Agnihotra is a pdkayajna, and not a full-fledged
§rauta ritual, because of certain ritual features which are indeed reminis-
cent of the domestic sacrifice outlined in the Grhya Sutras.** These
features, the teacher claims, are “of the pdkayajna,” and the text goes
on to note that the agnihotra is the “form” (rapa) of domestic animals;
and furthermore, “the pakayajna is related to domestic animals.”®
Here again we encounter an early connection between the pdaka-
yajna type of ritual and domesticity. And what Yajfavalkya does here,
and what others do elsewhere, by calling rites and rituals of the Srauta
ceremonial by this term is also to point to a “small” sacrifice in compari-
son to other, “greater” forms of ritualism. In another passage from the
Satapatha Brahmana, we come across the phrase “the pdakayajrias and
the rest” to refer to the whole set of Vedic sacrifices beginning with the
lowest, the most basic, and moving “on up”: “Agni is indeed all sacri-
fices. All sacrifices are performed in him, the padkayajnas and the rest.”%
The Aitareya Brahmana*’ also considers the pdkayajna to be the
simplest type of ritual still within the pale. In a passage whose point is to
extol the virtues of the soma sacrifice, the ritual is said to contain the
essences of all Vedic sacrifices,*® both the “later” (higher) forms of the
soma sacrifice and the “earlier” (lower) forms of rituals which do not use
the soma juice as the oblation material. Among these “earlier” sacrifices
are the various Srauta rituals beginning with the agnihotra. But the list of
83. SB 1.8.1.16, 1.8.1.18.

84. E.g., the injunction juhoti rather than the usual Srauta injunction yajati, and the
eating of the remains of the oblation material (SB 2.3.1.21). See also KSS 6.10.27-28. For
a further comparison of the Srauta agnihotra and the domestic ritual, see $B 2.3.1.18 and
TB 2.1.4.5 on the one side; and BGS 1.4.40—42; BGPariS 2.1.39—44; and ApGSs 1.2.10-—11
on the other.
85. SB 2.3.1.21.
86. SB 4.5.1.13; see also SB 1.4.2.10.
87. AitB 3.40-41.
88. For this feature of Vedic ritualism, whereby a more complex sacrifice is said to
contain within it representative fragments of less complicated rituals, see Frits Staal,
“Ritual Syntax,” and G. U. Thite, Sacrifice in the Brahmana Texts, pp. 52-54: “A sacrifice
is shown to include (mystically) many other sacrifices and then the importance of that
particular sacrifice is established.”
rituals said to be encompassed within the soma sacrifice begins with the
pakayajna, placed at the very start of a hierarchical enumeration that
moves from the simplest to the most complex.
These early references are too few and too vague to allow us to
conclude that already in the Samhitas and Brahmanas a full-blown do-
mestic cult similar to that depicted in the later Satras existed side by side
with the srauta ritual. They do indicate, however, that some simplified
rites and rituals within the srauta cycle were from ancient times desig-
nated by a term that was later used for the domestic ritual as a whole,
and that these Srauta rites and rituals so labeled as pdkayajnas also were
associated with matters particularly within the sphere of domesticity.
The later use of the term paékayajna in the Grhya Sutras was, at the very
least, consistent with and a perpetuation of the older semantics.
The simplicity and domesticity of the domestic ritual—as it is de-
picted in texts of all periods in Vedism—is in harmony with the relatively
feeble ritual capacities and limited spheres of influence of those who
perform it. Principally, of course, I am speaking of the domestic (and
domesticated) sacrificer himself, but also the wife and the son or pupil,
all of whom for one reason or another are precluded from the sacrifices
at the Srauta level. The domestic ritualist, the ekagni, is, by virtue of lack
of sufficient wealth, knowledge, opportunity, or desire, relegated to a
delimited sphere of ritualism—a level of praxis associated in the texts
with the home (and not the cosmos), the earthly (and not the heavenly),
the human (and not the divine), the “womanly” (and not the “virile”).
The wife, because of the supposed liabilities of her gender, does not
have the competence to learn the Veda or perform sacrifices that require
the recitation of Vedic mantras. But because of her situation within a
chain of interconnected concerns and arenas of interest (the earth, fecun-
dity, prosperity, food, animals, and the home), she is allowed to partici-
pate independently in Vedism on the grhya plane. The young boy or
Veda student is not yet out from under the control and lordship of his
teacher and father, and is not yet fully adept in either the Vedic texts or
the ritual procedure. The domestic ritual for him functions as a kind of
ritual training ground, preparing him for future ritual activity at the
same or higher levels.
In the Sutras, pdkayajna describes a formally condensed and rela-
tively inferior class of Vedic rituals, the domestic sacrifices. Its simplicity
is in conformity with its basic goals and purposes, all of which revolve
around the prosperity and protection of the household. It is because of
its fundamental nature within the “unity” of Vedic ritualism that I have
called attention to the analogies it has to the conceptualization of the
paradigmatic (prakrti) sacrifices in the Sutras’ organization of ritual

knowledge. Formally simple, prior (either chronologically or epistemo-
logically), and the base on which more developed or “extended” rituals
are formed, both the domestic sacrifice and the paradigmatic sacrifice
function as “elementary forms” of the Vedic ritual system.
As we shall observe in the next chapter, the framework or tantra for
carrying out the pdkayajna was a contracted version of the more ex-
panded framework of the Srauta sacrifice. The domestic procedure was,
in other words, a counterpart to the prototypical srauta format, upon
which it was based. Turning around the comparison made in this chap-
ter, the grhya ritual was dependent on the Srauta ritual as the modified
sacrifice (vikrti) is dependent on the paradigm (prakrti).
Furthermore, with this compacted ritual framework, the domestic
sacrificer was obliged to perform regularly sacrifices paralleling those
obligatory to the ahitdgni. In this way, the same rituals could be per-
formed at either the Srauta or grhya levels, bringing the different orders
of ritual praxis into a relationship of resemblance and complementarity
which itself harmonizes with Vedic metaphysics, epistemology, and the
principles of ritual and textual organization and representation.
7
Ritual Hierarchy, Substitution, and
Equivalency
Is the Simple the Equivalent of the Complex?
Within the context of a “unity” subsuming several discrete, if interrelated

and resembling, levels of Vedic ritualism and ritualists, the domestic
ritual is linked to higher forms through the relations of complementarity
and correlativity discussed at the beginning of Chapter 6. As we have
seen, however, these relations are further specified by the oppositions of
domesticated/expanded, “female”/“male,” simple/complex, and, above
all, the hierarchical relation of inferior/superior. The Vedic ritual may be
a “unity” composed of mutually related Srauta and grhya cycles, but the
two are clearly differentiated and differently valued.
Later I will return to the issue of just how—and why—the ritual
experts constituted the domestic sacrifice as simultaneously a condensed
counterpart and an inferior resembling form of the Srauta ceremonial;
and how, ironically, such machinations made it possible in the course of
time for the grhya ritual to replace, for all intents and purposes, its
prototype and survive within a different “unity,” that of Hinduism.
For now, let us call up once again the model that was useful for
discussing the relationship between the two levels of Vedic ritualism, the
model, that is, of paradigm (prakrti) and variation (vikrti). When the
domestic ritual, within Vedism, is seen as a kind of modification or
variation of the Srauta ritual, the Srauta acts as the paradigm for the
domestic ritual. Just as in the metarules provided in the Sutras the vikrti
sacrifices are presented in a shorthand, condensed summary while the
prakrti rituals are described more fully and completely, so, too, is the
domestic sacrifice an abbreviated, condensed, and adapted version of
the more complete, expanded, and prototypical srauta sacrifice. This
type of interrelation between the two levels of Vedic ritualism is stressed
by Jan Gonda when he writes, “There is no denying that the domestic
cult as described in the grhya manuals was to a considerable extent
influenced by, and even modelled on, the Srauta rites which in some
cases run parallel.”! As we also will observe later in this chapter, domes-
tic rituals are systemized in the Sitras in large part by being represented
as miniature replicas of Srauta models.
Having come this far, however, in this explication of the Vedic
ritual system (a system which, I have claimed, was based on principles of
resemblance and hierarchical distinction), we must now confront an
apparent discrepancy. For there is another possible relation between the
Srauta and grhya levels of the Vedic “unity of ritual,” which, if left
unexplained, would seem to undercut the very basis of the system I have
been describing.
Because of the parallels and strong bonds connecting the domestic
sacrifice to the frauta ritual—the web of bandhus the ritualists fabricated
in the process of constructing their ritual “unity”—it may appear that the
relation between the two is really one of equivalency. As Madeleine
Biardeau puts it, “It seems that the grhya rites with their one fire may
be... to a very large degree substitutes for Srauta rites.”? Were the
domestic rituals within the Vedic ritual system (which were, as I argued
in Chapter 6, inferiorized in relation to the more complex §rauta sacri-
fices) also substitutes in the sense of being equivalent replacements for
the rituals on which they were modeled and to which they were linked?
The question cannot be easily dismissed. For in the ritual Sutras
themselves, the locus classicus of the Vedic ritual unity, passages are
found in which the case for equality between simple domestic ritualism
and the complex Srauta sacrifice is apparently made in unequivocal
terms. One rather late ritualistic text equates the domestic fire with the
three or five Srauta fires, seemingly overturning the careful distinctions
and qualitative differences drawn between them by other texts in Chap-
ter 6:
The [Srauta] fires are incorporated into the domestic fire, and the domestic
fire into the sacrificer. Therefore one should not swerve from [the regular
1. Jan Gonda, The Ritual Sutras, p. 549; emphasis added.

2. Biardeau, Le Sacrifice dans I’Inde ancienne, p. 39; emphasis added.
Ritual Hierarchy, Substitution, and Equivalency 171
performance of sacrifices into] the domestic fire. . . . There is no distinc-

tion (vivecana) between the domestic fires and the Srauta fires.3
Similar statements may be found in even the older Grhya Sutras. The
Bharadvaja Grhya Sutra, quoting a (now lost) Satyayani Brahmana,‘
notes that because the deities are the same in the agnihotra of both the
Srauta and grhya cycles, the domestic sacrificer reaps the same benefits
from his small-scale version of the ritual as does the ahitagni whose perfor-
mance is much more complicated, time-consuming, and expensive:
Because of the prescribed deities [in both versions of the ritual], and
because of the results promised in the Veda, the same deities as are found
in the [Srauta] agnihotra are worshipped in the ritual performed in the
domestic fire. The practice of the ahitagni is the practice [for the domestic
sacrificer of the agnihotra]; the world (Joka) of the ahitagni is [the same as]
that of him who offers in the domestic fire.>
Another claim to such equivalency appears in a different text, this one
also equating the domestic and dhitdgni sacrificers:
The fire [which is maintained] from the marriage is called the domestic
fire. In that are performed all rituals relating to the household. Through
his regular service [to that fire, accomplished through the daily perfor-
mance of the agnihotra, he attains] the state of the ahitagni. And by the
oblations sacrificed on the new and full moon days, [he attains] the state of
the [Srauta] sacrificer of the new and full moon sacrifice.®
How are we to understand these seemingly straightforward equa-
tions between what the ritualists have elsewhere hierarchically distin-
guished? Is the simple really the equivalent of the complex in these ritual
texts? Is the domestic sacrifice within the Vedic ritual system a true and
full substitute for paralleling Ssrauta rituals in these texts? And, if so, why
would anyone go to the considerable trouble and expense of performing
the complicated and costly Srauta rites? The very raising of questions
3. Bharadvaja Parisesa Sutra 188. The text and translation of this work are included
in C. G. Kashikar, trans. and ed., The Srauta, Paitramedhika and Parisesa Sutras of
Bharadvaja, 2 vols. (Poona: Vaidika Samshodana Mandala, 1964). Kashikar dates this text
to the first centuries C.E.
4. Remnants of this Brahmana have been published in B. K. Ghosh, ed., Collection
of the Fragments of Lost Brahmanas (Calcutta: Modern Publishing Syndicate, 1935).
5. BhGS 3.18.
6. HGS 1.7.26.1-3. See also BGPariS 1.16.14: “With this [domestic new and full
moon sacrifice, he attains] the state of the [Srauta] sacrificer of the new and full moon
sacrifice.” Similar claims for the equivalency of the domestic and Srauta sacrificers of the
animal sacrifice may be found at BGPariS 2.1.27.
such as these shake what we have assumed to be the systematically

formulated and implemented foundations of the Vedic religious, ritual,
and social hierarchy. Do texts such as the ones cited above represent
dissenting voices (perhaps of the economically dispossessed)? The
changed circumstances of a different historical era? Or can they be
comprehended in terms of the mainstream and ancient Vedic “unity of
ritual”?
To begin grappling with these serious queries, let us turn to the
subject of substitution, for the answers may turn on how we understand
the meaning of substitution and its operation. First, we will analyze the
nature of substitution within those rituals found the world over which we
call sacrifices; and, second, we will look specifically at substitution
within Vedic ritualism.
Sacrifice and Substitution
Substitution, the placing of a stand-in for an original which then repre-

sents it, is at the very heart of sacrifice. Such, in any case, is the common
conclusion drawn by any number of scholars of the subject who have
approached the topic from a wide variety of methodological angles.
Henri Hubert and Marcel Mauss, in their classic Sacrifice: Its Na-
ture and Function, contend that “the very nature of sacrifice” is “depen-
dent, in fact, on the presence of an intermediary, and we know that with
no intermediary there is no sacrifice.”’ This “intermediary” so crucial to,
even defining of, the sacrificial ritual is the ritual victim who represents
(and substitutes for) both the divine recipient of the offering and the
human being who is offering:
Sometimes there is a direct and natural representation: a father is repre-
sented by his son, whom he sacrifices, etc. In general, since a sacrifier
[sacrificer] is always obliged to undertake the expenses in person, there is,
by virtue of this very fact, a more or less complete representation. But in
other cases this association of the victim and the sacrifier is brought about
by a physical contact between the sacrifier (sometimes the priest) and the
victim. . . . Through this proximity the victim, who already represents the
gods, comes to represent the sacrifier also. Indeed, it is not enough to say
that it represents him: it is merged in him. The two personalities are fused
together.8
7. Henri Hubert and Marcel Mauss, Sacrifice: Its Nature and Function, p. 100.
8. Ibid., pp. 31-32.
It is, according to these authors, through the victim (who is the full
representative of and substitute for both the giver and the recipient of
the sacrifice) that the “communication” between the sacred and profane
realms is effected (and this is the “nature and function” of sacrifice
according to Hubert and Mauss). The purpose of sacrifice is, therefore,
to bridge the chasm between the sacred and profane worlds, and its
mechanism for doing so involves a kind of double substitution: the vic-
tim stands in for both the sacrificer and the deity and thereby draws
them together.
Many scholars have noted that all sacrifice is a kind of symbolic self-
sacrifice; as Ananda K. Coomaraswamy puts it, “To sacrifice and to be
sacrificed are essentially the same.”? The sacrifice is a symbolic transfor-
mation of the self as one immolates a “lower” self and gains a “higher”
identity:
In other words, the appetitive soul, the greedy mind, is the Sacrifice; we,
as we are in ourselves, seeking ends of our own, are the appropriate burnt
offering. . . . The sacrificer’s “death” is at the same time his salvation; for
the Self is his reward.!°
And insofar as the self-sacrifice which is sacrifice remains “symbolic,” a
substitute victim is obviously required. Sylvain Lévi has thus termed
sacrifice a “subterfuge,” and he states his position forcefully: “The only
authentic sacrifice would be suicide.”!! The substitution of another for
the self might thus serve to distinguish sacrifice from other related phe-
nomena: suicide, martyrdom, and all other literal forms of self-death.
Others have emphasized another function of sacrificial substitution:
displaced violence toward another. In this view, sacrifice is defined not
in relation to other forms of self-death but rather in relation to other
forms of murder—and in either case it is by virtue of the substitute that
sacrifice is distinguishable from other killings.
Sigmund Freud, in his brief but extraordinarily important series of
reflections on sacrifice at the end of Totem and Taboo, understood the
sacrificial victim as only consciously the substitute for the sacrificer (al-
though the element of the sacrificer’s “renunciation”—of the egoistic
and selfish will as well as of healthy independence—he found very signifi-
cant indeed, and characteristic of religion as a whole). He argued, how-
ever, that the victim was also an unconscious representation of or substi-
9, Ananda Coomaraswamy, “Atmayajna,” p. 359.

10. Ibid., pp. 375, 376.
11. Sylvain Lévi, La Doctrine du sacrifice, pp. 32-33.
tute for the deity, the figure of authority, the idealized, all-powerful
father figure. Sacrifice is thus simultaneously a conscious renunciation of
independence, performed out of guilt generated from repressed feelings
of hostility and resentment toward the authority figure, and an uncon-
scious display of those same feelings as the authority figure, in the guise
of the victim, is slaughtered and consumed:
We find that the ambivalence implicit in the father-complex persists in

totemism and in religions generally. Totemic religion not only comprised
expressions of remorse and attempts at atonement, it also served as a
remembrance of the triumph over the father. . . . Thus it became a duty
to repeat the crime of parricide again and again in the sacrifice of the
totem animal, whenever, as a result of the changing conditions of life, the
cherished fruit of the crime—appropriation of the paternal attributes—
threatened to disappear.
For Freud, in sum, sacrifice is shot through with ambivalence to-
ward authority, and most especially toward the father. It is stimulated by
guilt stemming from Oedipal wishes, and it alleviates that guilt by provid-
ing a means for renouncing such wishes; but it also perpetuates and
reproduces guilt by giving expression to the Oedipal fantasy of killing
the father by means of killing the substitute, the sacrificial victim.
A more recent work, indebted in many ways to Freud, on sacrifice
and substitution is René Girard’s Violence and the Sacred. Girard, like
Hubert and Mauss and like Freud, places the concept of substitution at
the very center of sacrifice and, again like the others mentioned, speaks
of a “double substitution” in the ritual. Following Freud, Girard sees
violence (repressed and expressed) as the key to sacrifice, and he writes
of a “fundamental truth about violence; if left unappeased violence will
accumulate until it overflows its confines and floods the surrounding
area. The role of sacrifice is to stem this rising tide of indiscriminate
substitution and redirect violence into ‘proper’ channels.”
The substitute or victim, who represents the community as a whole,
provides such a target for the acceptable expression of violence. But it is
here that a complicated “double substitution” comes into play. The
“original” or “real” victim is nothing other than the communal violence
inherent in society—and this must be expunged for the survival and
harmony of the group. A “surrogate victim”—a scapegoat, in other
words—substitutes for the community. But then, according to Girard, a
12. Sigmund Freud, Totem and Taboo, trans. by James Strachey (New York: W. W.
Norton, 1950), p. 145.
13. René Girard, Violence and the Sacred, trans. by Patrick Gregory (Baltimore:
Johns Hopkins University Press, 1977), p. 10.
“ritual victim” substitutes for the “surrogate victim” in order to preclude

revenge for the death of the surrogate. The cycle of violence is thus
short-circuited by the double substitution which at once conceals and
represents the real victim" as it transfers the violence from the real
victim to a surrogate and then again to a ritual substitute. This byzantine
chain of displacement and replacement is explained thus:
The ritual victim is never substituted for some particular member of the
community or even for the community as a whole: it is always substituted
for the surrogate victim. As this victim itself serves as a substitute for all
the members of the community, the sacrificial substitute does indeed play
the role that we have attributed to it, protecting all the members of the
community from their respective violence—but always through the inter-
mediary of the surrogate victim. ... Ritual sacrifice is founded on a
double substitution. The first, which passes unperceived, is the substitu-
tion of one member of the community for all, brought about through the
operation of the surrogate victim. The second, the only truly “ritualistic”
substitution, is superimposed on the first. It is the substitution of a victim
belonging to a predetermined sacrifical category for the original victim.!5
Whether conceived as a transaction leading to communication be-

tween the sacred and the profane (Hubert and Mauss), or as a self-death
leading to a transformation of the self, or as a violent expression and
displacement of hostility toward another (the father with Freud, the
community and its members with Girard), the general conclusions re-
garding sacrifice coincide. For all, sacrifice is defined by substitution. It
is, in the first place, a substitute for an impossible or prohibited real act,
such as the actual coalescence of the divine and the human, suicide, or
violent aggression and murder. All substitutions within a sacrificial
14. Girard’s position is that there is both a concealment and a necessary memory of
the original victim for the sacrifice to be efficacious: “Once we have focused attention on
the sacrificial victim, the object originally singled out for violence fades from view. Sacrifi-
cial substitution implies a degree of misunderstanding. Its vitality as an institution depends
on its ability to conceal the displacement upon which the rite is based. It must never lose
sight entirely, however, of the original object, or cease to be aware of the act of transfer-
ence from that object to the surrogate victim; without that awareness no substitution can
take place and the sacrifice loses all efficacy.” Ibid., p. 5.
15. Ibid., pp. 101-2.
16. See now also Jan Heesterman’s “Self-sacrifice in Vedic Ritual.” Heesterman
concludes, “What emerges from the ritual and from ritualist speculation is that self-
sacrifice as such is invalid. . . . Sacrifice, on the other hand, cannot be valid by immolating
just any victim that presents itself. The person, animal, or substance that is immolated
must be that part of the sacrificer that defines him as such, namely the goods of life he has
acquired by risking his own life... . Without this bond uniting the sacrificer and his
victim, sacrifice would be as invalid as self-sacrifice is per se” (p. 105).
ritual are therefore substitutions for a prior, and defining, substitution.

Secondly, all emphasize the critical importance of the victim to the
sacrificial process, and all see this victim as a multivalent substitute and
symbol for a pair of opposites: the sacred and the profane, god and man,
recipient and giver, higher and lower selves, father and son, society as a
whole and its individual members. The substitute victim is, as a symbol,
a representative of two (or more) different and even contradictory
things or beings.
And because it is a symbol, it is not critical to the efficacy of the
sacrificial process itself who or what is selected to act as the symbol, the
victim. It could be, and in the comparative view is, almost anything:
humans (of any number of culturally constituted types), animals, vegeta-
bles, nonsentient physical objects, physiological functions (as in the “sac-
rifice of breath” in the Upanishads), even ideas (“individuality,” “de-
sire,” “faith,” “truth,” etc.) can and do function as the symbolic victim.
This is not to say, however, that within particular cultures and
particular religious traditions the choice of the appropriate victim is left
to the whims of individuals. Far from it. The appropriate sacrificial
victims are, as Girard says, “predetermined.” They are also often hierar-
chically ordered and valued.!’? Such is certainly the case in Vedic
ritualism—or is it?
Vedic ritualism in its entirety, like sacrifice as a category within
religion, is based on substitution. The theology, metaphysics, and ontol-
ogy created by the Vedic ritualists presume the inaccessibility of tran-
scendent prototypes and the necessity, therefore, of ritual action using
counterparts or “symbols” for the “real thing.” Indeed, the sacrifice as a
whole is a counterpart of the transcendent Cosmic One, Prajapati or
Purusa, who has created the ritual as a pratima of himself.
I have pointed out elsewhere that in Vedism—as in sacrifice
generally—the victim simultaneously represents the deity and the sacri-
ficer, just as the ritual as a whole mediates between macrocosmos and
17. For a dicussion, inter alia, of the hierarchies of sacrificial victims in Indo-
European myth and ritual, see Bruce Lincoln, Myth, Cosmos, and Society: Indo-European
Themes of Creation and Destruction (Cambridge, Mass.: Harvard University Press, 1986),
esp. pp. 65-86. More specific treatments of the topic in the Indo-European contexts are E.
Mayrhofer-Passler, “Haustieropfer bei dem Indoiraniern und den andern indo-
germanischen Volkern,” Archiv Orientalini 21 (1953): 182-205; Gerard Capdeville, “Sub-
stitution de victimes dans les sacrifices d’animaus 4 Rome,” Melanges d’archéologie et
@ histoire de l’Ecole Francaise de Rome 83 (1971): 283-323; and Jean Puhvel, “Victimal
Hierarchies in Indo-European Animal Sacrifice,” American Journal of Philology 99
(1978): 354-62. My thanks to Bruce Lincoln for bringing these studies to my attention.
microcosmos. The often mentioned equation of Prajapati and the sacri-

fice (prajapatir vai yajnah), and that of the sacrificer (and/or the Cosmic
Man) and the sacrifice (puruso vai yajriah), may also be formulated with
“victim” (pasu) plugged in for “sacrifice.”
But all counterparts are not created equally, and so, too, are sacrifi-
cial victims not all equal in their ability to represent most fully the
prototype. The victim, mediating between and simultaneously represent-
ing both god and sacrificer, comes in different varieties with different
capacities to manifest the complete form. One list of the appropriate
victims for sacrifice is found in a passage whose purpose is to explain the
etymology of the term pasu:
Prajapati turned his attention to the forms (ripas) of Agni. He searched
for that boy who had entered into the [various] forms [of Agni]. Agni
became aware [of this and thought], “Father Prajapati is searching for me.
I will take a form unrecognizable to him.” He saw those five sacrificial
victims: man, horse, cow, ram, and goat. Because he saw (pasyat) [them],
they are therefore pasus. He entered these five sacrificial victims and
became these five sacrificial victims.!8
Later in that same text, another passage makes it clear that the order in
which the pasus are listed is not accidental but rather is calibrated to the
relative “excellence” of each:
He offers [alabhate, literally “seizes” or “possesses”] the man first, for
man is the first of the pasus; then [he offers] the horse, for the horse comes
after the man; then the cow, for the cow comes after the horse; then the
ram, for the ram comes after the cow; then the goat, for the goat comes
after the ram. In this way he offers them, in hierarchical succession
(yathapurvam) according to their relative excellence (yathasrestham).'9
Man (purusa) is generally proclaimed in the texts to be the highest
of all possible sacrificial victims. Not only is man listed first in various
enumerations of the pasus; he is also extolled by texts which place him
closest to Prajapati. In an account of a sacrifice sponsored and per-
formed by the gods, the list of daksinds given to the divine priests
includes the gift of a man to Prajapati?9—for, as we read elsewhere,
“Man is the nearest to (nedistham) Prajapati [of all creatures].”*! The
18. SB 6.2.1.1-3. The “boy” alluded to in the second sentence seems to refer to the
Rg Vedic myth in which the child of Agni is lost, searched for, and found (RV 5.2). My
thanks to Wendy Doniger O’Flaherty for this reference.
19. SB 6.2.1.18.
20. TB 2.2.5.3; see also PB 1.8.14.
21. SB:5.1.3:8; 5.2.1.6.
human being in another account is said to be superior because “man is

all pasus,” that is, encompasses them all by virtue of his hierarchically
superior rank (note also, again, the order in which the pasus are listed):
Prajapati at first was one. He desired, “May I emit food, may I be repro-
duced.” He measured out the pasus from his breaths. From his mind [he
measured out] the man; from his eye, the horse; from his breath, the cow;
from his ear the ram; from his voice, the goat. And since he measured
them out from his breaths, they say “The pasus are the breaths.” And
since he measured out the man from his mind, they say “The man is the
first, the most potent (viryavattama) of the pasus.” The mind is all the
breaths, for all the breaths are firmly established in the mind. And since
he measured out the man from his mind, they say, “The man is all pasus,”
for they all become the man’s.”
In yet another narrative concerning the types and relative ranking

of the sacrificial victims, this same order—from man, the highest form of
pasu, to the goat, the lowest on the scale—is reiterated. The enumera-
tion appears in the course of relating how the quality that constitutes all
of these animals as worthy of sacrifice (their medha) entered into and
then left them, creating in its wake new but defective forms of each. In
this text, we are also introduced to another component of the Vedic
theory of substitution. For although it is clear that the pasus are ranked
here, as they are elsewhere, the lowest on the hierarchical ladder is also
said to be the victim most often used and, in some ways, the best victim:
The gods offered man as sacrificial victim. Then the sacrificial quality
(medha) passed out of the offered man. It entered the horse. Then the
horse became fit for sacrifice and they dismissed him whose sacrificial
quality had passed out of him. He [the former man, now devoid of the
sacrificial quality] became a defective man (kimpurusa).” They offered
the horse, and the sacrificial quality passed out of the offered horse. It
entered the cow. . . . It [the former horse] became the white deer. They
offered the cow. . . . The sacrificial quality entered the ram. . . . It [the
22. SB 7.5.4.6.
23. The kimpurusa is synonymous with the kinnara (Manu 1.39), the purusa mrga
(“wild man,” TS 5.5.15.1; MS 3.14,16; VS 24.35) and the mayu (at SB 7.5.2.32 the
kimpurusa and the mayu are explicitly equated). Various translations have been offered
for these terms, ranging from “monkey” or “ape” to “dwarf,” “savage” or “mock-man.”
See, e.g., Eggeling’s note on SB 1.2.3.9. For the kimpurusa as a “horizontal androgyne,”
half equine and half human, consult Wendy Doniger O’Flaherty, Women, Androgynes and
Other Mythical Beasts (Chicago: University of Chicago Press, 1980), pp. 216, 309. It seems
probable that in Vedic texts it is the tribal peoples living outside of the Aryan settlements
to which all the terms, kinnara, kimpurusa, purusa mrga, and mayu, refer.
former cow] became a gayal. They offered the ram. . . . The sacrificial
quality entered the goat... . It [the former ram] became the camel. It
[the sacrificial quality] stayed the longest in the goat; therefore the goat is
the pasu most often used [as sacrificial victim].24
The least “excellent” of the pasus, the goat, is here presented as not
only the one “most often used” but also the one most saturated with the
quality that constitutes a sacrificial victim as such, the medha. This
peculiar virtue of the least desirable and hierarchically most inferior
victim is corroborated in other texts as well. One, for example, displays
the goat to advantage by claiming that it possesses the essences of all the
other (and hierarchically superior) pafus—thus turning on its head the
earlier cited text which claimed, in accord with the expected, that man,
the highest pasu, encompasses all other victims. In the text at hand,
however, the offering of a relatively lowly goat in sacrifice is “equiva-
lent” to the offering of all the possible victims:
In this pasu [the goat] is the form (riépa) of all pasus. The form of man
[incorporated in the goat] is what is hornless and bearded, for man is
hornless and bearded. The form of the horse is what is possessing a mane,
for the horse possesses a mane. The form of the cow is what is eight-
hoofed, for the cow has eight hoofs. The form of the ram is what possesses
ram-hoofs, for the ram possesses ram-hoofs. What is the goat, that is [the
form] of the goat. Thus, when he offers this [goat], all those pasus are
offered by him.*
The least worthy substitute is thus said to recapitualte all members of the
class of substitutes. Nor does this process of concentrating the sacrificial
quality into lower and more easily expendable objects in the category
stop with the goat.
When, according to the myth, the sacrificial quality finally left the
goat after its longish residency, “it entered this earth. They searched for
it by digging. They found it as those two, rice and barley (vrihi and
yava). Thus even now they find those two by digging.” And further,
these lowly victims, these vegetable pasus, are said to “possess as much
potency (viryavat) as all the sacrificed pasus would have.” This text is
echoed by another, which equates the vegetable oblation with the ani-
mal offering. “When they also offer a cake [made from rice or barley] in
the animal sacrifice [it is with the thought], ‘Let our sacrifice be with a
24. AitB 2.8; see also SB 1.2.3.6, 1.2.3.9.

25. SG 2 1S"
26. SB 1.2.3.7.
sacrificial victim possessed of the sacrificial quality; let our sacrifice be

with a fully constituted (kevala) victim.’ ”2’
Such texts from the Brahmanas seem not so much to solve the
problem we began with—whether inferior domestic sacrifices substitute
for, in the sense of acting as the full equivalents of, complex Srauta
rituals—as they appear merely to restate the quandary. The revaluation
of the lowest pasus as the recapitulation and apparent equal of the
highest sacrificial victims seems to indicate that here, as in the case of
the equations drawn between domestic and Srauta rituals, the hierarchi-
cal order of things has been overturned and rendered meaningless—
perhaps by rival groups within the Vedic ritual milieu who competed for
ritual status and social position by changing the economic rules of the
game.
But perhaps also there is a simpler answer to this puzzle of substitu-
tion and “equivalency,” one that emerges with consistency from within
the established Vedic system of ritual and resemblance itself. In order to
explicate the principles at work here, we must expand the discussion to
substitution of all sorts in Vedic ritualism, and we might again take our
cue from the paribhdsds or “general rules” on the topic as they are
prescribed in the Sutras.
Vedic Principles of Substitution
The passages from the Brahmanas on substitutions for the victim in

sacrifice are but one dimension of a more comprehensive theory of
substitution within the Vedic texts. And although such a theory is rooted
in the Brahmanas, it is fully and systematically developed only in the
ritual Sutras.*8
In addition to treating the various kinds of substitutes for animal
victims, the Brahmanas occasionally provide rules for substituting for
other oblation materials used in the ritual. To take only one example,
many of these older texts allow for the use of the patika plant and others
27. AitB 2.8.

28. The best study I know of to date on the intricacies of the Vedic ritual theory of
substitution, and one to which I am indebted for what follows here, is Frederick Marcus
Smith’s The Vedic Sacrifice in Transition: A Translation and Study of the Trikan-
damandana of Bhaskara Misra, Ph.D. dissertation, University of Pennsylvania, 1984. For
an interesting overview of the philosophy of substitution in Vedic ritual texts, see
Heesterman’s “Self-sacrifice in Vedic Ritual.”
in place of soma%—suggesting, perhaps, that the original soma was

unknown, or often unavailable, or possibly never existed at all.3? Soma,
one might say, had become a transcendent prototype early on; in this
case, as in others, both the “original” and the “substitute” were counter-
parts of an inaccessible prototype.*!
Such instances of rulings on substitution in the Brahmanas—
confirming general theories of “double substitution” surveyed above—
became authoritative precedents for the authors of the Sutras. Sruti
legitimization, in other words, provided one kind of authorization for
substitution.** Conversely, previous explicit statements in the Veda pro-
hibiting certain substitutions delimited those allowed by the Sutras.
Thus, nothing that is prohibited in the Sruti can be used as a substitute
“because of the unacceptability” of such innovation.*?
The Sruti was, however, not sufficiently explicit and comprehensive
in its rulings on substitutions to function as the sole guide for all the
instances in which substitutes might be necessary. Other standards were
also observed by the ritualists, such as the authority of learned Brah-
mins.*4 But the most important of these aids were certain general princi-
ples which directed the specifics of substitutions in all cases.
One of the chief reasons for substitution in sacrifice, according to
the Siitras, is to enable the “accomplishment” or “completion” of an
obligatory (nitya) ritual which otherwise would not be performed.* Re-
placing a specified element of the ritual with an acceptable substitute is
justified by the “purpose” or “end” (artha) of the ritual as a whole, and
in the case of the nitya or obligatory sacrifices, the very performance is
the ritual’s purpose. “Nitya rituals,” it is said in one Sutra, “are those
which, when left unperformed, bring damage (dosa) and worldly re-
29. E.g., SB 4.5.10.1ff; TB 1.4.7.5ff; JB. 1.354; PB 9.5.2ff.

30. The standard work on the question of the original soma remains R. Gordon
Wasson and Wendy Doniger O’Flaherty, Soma: Divine Mushroom of Immortality (New
York: Harcourt, Brace and World, 1968). The authors conclude that soma was originally a
mountain mushroom called the fly agaric which eventually became inaccessible to the
ritualists as they moved away from the Himalayas. That an “original” soma never existed
at all is a speculation that has not been put forward by Indologists and perhaps deserves
consideration.
31. See AitB 2.14 and TB 3.12.5.12 for the substitution of ghee for gold, “for both
are nectars of immortality.”
32. See, e.g., KSS 1.4.4-5, 1.6.11.
33. KSS 1.6.8; see also HSS 3.1; ApSS 24.4.1.
34. See, e.g., BGS 4.9.15.
35. See, e.g., SSS 3.19.2; HSS 3.1.
proach [to the sacrificer].”36 “One should be forever bound to the perfor-
mance of obligatory sacrifices,” states another text. “He whose obliga-
tory sacrifices are interrupted . . . takes an evil path. He does not go to
heaven, but rather falls.”37 And so, the Sutra continues, nitya sacrifices
may be done with any number of substitutions, as long as they are done:
“Thus, with a root, some fruit, honey or meat, the obligatory sacrifices
are to be performed continually. And [thereby] one does not interrupt
the obligatory sacrifices.” We may be here observing, by the way, one of
the early roots of the later Hindu theory of a@pad dharma, in which the
normal duties of the various castes are suspended in “times of emer-
gency.” Indeed, the laws of dharma and the laws of ritual are both
extremely complex, detailed, and demanding; perhaps for this very rea-
son the lawmakers provided various escape hatches to allow for their
human frailty and the vagaries of life.
Substitutions may be made before beginning an obligatory sacri-
fice—the reasons for doing so include, as in the case of dpad dharma,
various exigencies in times of distress or the unavailability of certain
required materials**—or after the sacrifice has already been started
(e.g., when certain of the necessary materials have become spoiled, lost,
or otherwise rendered unusable).*? The situation is different, however,
in the optional or kamya rituals of the Vedic ritual repertoire. There,
because the “purpose” or “end” of the sacrifice is the attainment of
“desired” or optional goals, there is much more emphasis placed on
doing the ritual with the required materials. “In the absence of a pre-
scribed [substance], one should not begin [an optional sacrifice], for its
proper accomplishment depends on that [prescribed substance].”4° Un-
like the obligatory sacrifices, an optional ritual should not be under-
taken if the materials for it are unavailable; substitution here is allowed
only if the required substances become unusable in the course of the
sacrifice. In those cases, a replacement may be brought in in order to
avoid “incompletion” and the “blemish” (dosa) the sacrificer would oth-
erwise acquire.*!
36. HSS 3.1.

37. BSS 28.13.
38. See, e.g., BSS 29.8, for the “times of distress” which may preclude the perfor-
mance of the agnihotra: a disturbance within the whole country; becoming afflicted with
disease; being away ona journey; living at a teacher’s house; unfavorable conditions regard-
ing place, time, and materials; and “circumstances in which other things are unavailable.”
39. HSS 3.1.
40. KSS 1.4.1.
41. KSS 1.4.3-4.
Substitutions are therefore prohibited in cases where there is a

direct Sruti injunction to that effect and in the case of an optional sacri-
fice not yet begun. Other prohibitions are generated, however, out of
the general principle of the importance of the purpose of the ritual and
the subservience of other concerns to it. One reads that no replacements
may be made for the sacrificial fire, the deity to whom the offerings are
made, the mantra which accompanies an oblation, or the ritual act it-
self.#7 All these components, it would seem, are too intimately con-
nected to the purpose of the ritual to be tampered with. As Chakrabarti
puts it, “Transcendental results are produced by the sacrificial fires,
deities, Mantras, and acts. If they are replaced by substitutes, the result
of the sacrifice cannot be achieved.” All the Sitras are agreed that
there can also be no substitute for the sacrificer himself. The “lord”
(svamin) of the sacrifice is also too closely connected to the “fruit” of the
ritual, that is, the result or “end” of the sacrifice, to be replaced by
another.
But if substitution is, in some cases at least, allowable on the
grounds of the fulfillment of the purpose of the sacrifice, how is one to
know what exactly may be substituted for an unavailable or unusable
original? “A substitute should be taken on the basis of resemblance
(samanya),” that is, resemblance to the original; and “this is the guid-
ing method (dharma) for all substitutions.”“ As in the epistemological
standards for linking paradigmatic rites and rituals to variations of them,
the fundamental principle for proper substitutions revolves around the
42. KSS 1.6.6-7; ApSS 24.4.1; HSS 3.1: Another list (HSS 3.1) also disallows
substitutes for the son of the sacrificer and the place and time of the ritual.
43. Chakrabarti, The Paribhasas, p. 177. See also KSS 1.6.6, where the reason for
the prohibition against substituting for these items is “because of their subservience to the
purpose of another (pararthavat).”
44. KSS 1.6.9; see also HSS 3.1; ApSS 24.4.1. Some exceptions to this rule are
allowed, however. A substitute for one of the sacrificers in a sattra, should he die before
the ritual is completed, is legitimate (KSS 1.6.11-12); and it would appear that the pres-
ence of the sacrificer is not required during the performance of those subsidiary rites not
directly concerned with the production of the “fruit” of the ritual. In this latter case, the
officiating priest also acts in the place of the sacrificer. See the oblique sutras at KSS 1.6.9-
10 and BGS 1.3.13. And, of course, in a very real sense, the victim serves as a substitute
for the sacrificer, thus obviating the need to substitute further.
45. KSS 1.4.2; MSS 3.1.3; ApSS 24.3.52; HSS 3.1; SSS 3.20.9; PMS 6.3.27.
46. SSS 3.20.10. See Frederick Smith’s translation of the “Trikandamandana of
Bhaskara Misra,” 2.26 (p. 453): “[Substitutes should be selected on the basis of] similarity
of effect (karya), form (rapa), color (varna), sap (ksira), flowers (puspa), fruit (phala),
smell (gandha), and taste (rasa). In the absence of the former [attribute], the next in the
list should be accepted.”
concept of resemblance. As Chakrabarti notes, “Similarity with the origi-

nal material is the determining factor in choosing the substitute, so that
the substitute may be capable of serving the purpose of the original
substance.”47 No substitute may “conflict” with the original, and, again
as in the discussion on paradigms and variations, an order of resem-
blances applies here, too, in cases of conflicting claims to resemblance.
It is the “resemblance of purpose” that is supreme and overruling when
it comes to otherwise equally resembling options.*
The question of which option to select for a substitute when there
are several available is addressed at KSS 1.4.14. The sitra states that
one of them should be chosen at the start of the sacrifice (picked on the
basis of its availability and for its “resemblance of purpose” to the origi-
nal) and should then be employed for the duration of the ritual. The
other options then cease to be options. If the substitute for an original
should for some reason become unusable in the course of the perfor-
mance, the secondary substitution should be made on the basis of its
resemblance to the original, and not on the basis of a similarity to the
substitute.4? “There can be no substitute for a substitute,” as Chak-
rabarti notes.°° On this point, we may be faced with a Vedic exception to
Girard’s general rule. In his theory of ritual substitution, it will be re-
called, the “ritual substitute” substitutes for a “surrogate victim” or
scapegoat which is itself a substitute for a “real victim.” The Vedic rule,
by prohibiting substitutes for substitutes, seems to disallow—at least at a
purely syllogistic level—Girard’s formula.
Also consistent with the principles of Vedism as they are manifested
in other situations is the fact that a substitute is considered a resembling,
but not identical, counterpart to its prototypical original. Thus, on the
one hand, the replacement is to be regarded and treated in the same way
as the original would have been. All the characteristic features, and all
the preparatory rites which make the substance fit for sacrifice, are
transferred to the substitute.*! The mantras recited for the original also
remain unchanged when there is a replacement: references in the pre-
scribed verse to the original (e.g., soma or goat) are unexpunged even
47. Chakrabarti, The Paribhdasas, p. 175.

48. HSS 3.1; KSS 1.4.16.
49. KSS 1.4.15.
50. Chakrabarti, The Paribhdasas, p. 176. See PMS 6.3.9, 6.3.35-36, and Sabara’s
commentary.
51. ApSS 24.3.53.
when a substitute is being used (e.g., piitika for soma, or a cake for a
goat).°2
On the other hand, however, it is clear that some substitutes more
nearly resemble the original than others, and it is here that we pick up
the threads of the earlier discussion on the hierarchical ranking of ele-
ments within a class. The Sutras, when listing the possible replacements
for any given original component of the sacrifice, do so with a clear
order of preference based on the strength of the resemblance of the
species to the genus.
Take, for example, the substitutes given for clarified butter (a sta-
ple of virtually any Vedic ritual). BSS 28.13 notes that the genus here is
the ghee made from cow’s milk, and the replacements for it (“if it is
unattainable”) are then listed: ghee made from buffalo’s milk, or from
that of a goat; oil of sesame seeds, of jartila seeds, of flax seed, of
safflower seeds, or of mustard seeds; resin from a tree; or flour of rice,
barley, or Syamaka mixed with water. Or, again, observe the progres-
sion in the list of substitutes for fire if it has gone out and cannot be
regenerated. One may, in that case, use an ordinary (lawkika) fire, or
offer in the right ear of a female goat (because, according to SB 7.5.2.6,
Prajapati created the goat from his ear?), in the right hand of a Brahmin
(where the “oblations” of sacrificial gifts, daksinds, are also placed), on
a cluster of darbha grass, or in water.™
It must be emphasized again that such lists of substitutes are not
randomly registered as equally valued alternatives. One Brahmana gives
as substitutes for the milk at the Agnihotra the following in descending
order of suitability: oblations of rice and barley (such as are standard in
other rites) or of other domesticated plants in general or of wild plants,
fruits from trees, water, or an oblation of “truth in faith” (satyam
sraddhayam).** The hierarchical nature of such lists is perhaps most
obviously exemplified in the discussion of substitutes for the daksinad or
“sacrificial fee” offered to the officiating priests by the sacrificer.*> In
52. SSS 3.20.11; AsvSS 3.2.19; KSS 1.4.9-10.

53. See ApSS 9.3.3-16; BhSS 9.4.5-9.5.5; HSS 15.1.51-64.
54. SB 11.3.1-4; see also MSS 8.10.4 for a substitute sacrifice of “truth” in the
agnihotra.
55. For the daksina, which is not really a “fee” as much as another kind of oblation
offered to the Brahmins, the “human gods” of the sacrifice, see Charles Malamoud,
“Terminer le. sacrifice: Remarques sur les honoraires rituels dans le brahmanisme,” in
Biardeau and Malamoud, Le sacrifice dans I’Inde ancienne, pp. 155-204; Klaus Mylius,
“Daksina: Eine Studie tiber den altindischen Priesterlohn,” Altorientalische Forschungen 6
Baudhayana Srauta Sitra,® it is said that the standard daksind for a

sacrifice is comprised of a cow, some gold, and a garment—obviously a
gift of some expense (and in some sacrifices the daksinds are far more
costly). If these items are “unavailable,” the text continues, a much
more modest (and inferior) substitute is prescribed: a gift of edible fruits
and roots. “If even this not [within reach of the sacrificer],” the teacher
concludes (perhaps with some exasperation), “he should not offer the
sacrifice.” Just as one could not offer a sacrifice to the gods without
providing some (indeed, almost any) oblation in the fire for them, so
one is prohibited from sacrificing without offering something to the
Brahmin priests, the “human gods.” And it may not be out of place here
to recall who the creators of these ritual rules were.
Condensation and Hierarchical Resemblance
Let us now turn again to the problem of certain claims for the domestic
ritual as an equivalent substitute for srauta sacrifices, having gained
some perspective on the issue through an overview of the Vedic theory
of substitution. As we have seen, substitutes in general in the Vedic
ritual are regarded as resembling, but not identical, counterforms of
their original prototype. As such, they are certainly connected to the
prototype (because of the shared participation within a class) but are
also clearly hierarchically distinguished as well (the species is not the
genus, the counterpart is not the prototype, and one counterpart is not
the same as any other). Such are the by now familiar workings of the
Vedic theory of resemblance, operating between the two poles of the
exactly the same (the equivalent, the identical, jami) and the completely
different (the unique, the disconnected, prthak).
Still, one might ask, how does all this make sense of statements
which seem to equate the lower with the higher, the lowly goat with the
“most excellent” pasu, the simple domestic ritual with the intricate
§rauta sacrifice? Is not hierarchical distinction eclipsed by these claims of
equivalent substitution?
Again, I think, a general overview informs the specific case. For, as
we have seen, substitution in the ritual texts entails not only a subse-
quent assumption of equivalency (the substitute is treated in the ritual as
(1979): 141-79; and Jan Heesterman, “Reflections on the Significance of the Daksina,”
Indo-Iranian Journal 3 (1959): 241-58.
56. BSS 28.13.
if it were the original) but also is in every case a simplification. In all the
instances treated in the Sutras, the chain of acceptable substitutes moves
from the most complex, highly valued, and rare to the simpler, less
costly, and more common. Such chains always end with the minimally
acceptable. And this is, I suggest, precisely what is assumed in and lies
behind those texts that speak of the “identification” between the grhya
and Srauta ceremonials within Vedic ritualism.
A larger perspective—that is, one that looks at the system as a
whole—puts the particular in its proper place. Minimalism presented as
equivalency is not unique to the situation perplexing us. The representa-
tion of the smaller, the less adequate, and the abbreviated as the “equal”
of the larger, the fully appropriate, and the unabridged is an integral
feature of Vedic ritualism. In one sense, the whole of Vedic ritualism is
founded on such a notion.
The “equation” of obviously inferior rituals to higher ones was not
invented by the authors of the Grhya Sutras. It was a relatively common
means of expression long before. Claims of equivalency between sacri-
fices of different values were made in the Brahmanas between rituals
within the Srauta cycle and between hypothetical and prototypical rituals
and their human counterparts. “Find the sacrificial ritual which is the
counterpart of that one of a thousand years’ duration,” say the gods in
one text. “For what man is there who could complete a thousand-year
ritual?”>’ The text continues with a string of equations between rituals of
greater length and those done more economically, concluding with the
Tapascita sacrifice which is said to be the counterpart of the thousand-
year sacrifice. The phenomenon, when it deals with sacrifices within the
realm of human possibility, most often takes the form of “equating”
lesser rituals to the soma sacrifice. This latter ritual is often said to be the
“most excellent” or “best” or “first” among all rituals; that is, it is the
highest and most nearly complete and perfect of the Vedic rituals.° It is,
therefore, the sacrifice other sacrifices replicate. The Sadvimsa Brah-
mana, for example, goes to great lengths to demonstrate how the rites
of the simplest of the Srauta sacrifices, the agnihotra, are “really” the
same as those comprising the soma sacrifice. The caturmasya ritual is in
like manner said to be the “form” (rapa) of the soma sacrifice, and the
animal sacrifice, one text claims, is not to be categorized among the
57. SB 12.3.3.5.
58. See SB 6.6.3.7; TS 7.1.1.4; ApSS 10.2.3.
59. SadB 4.1.6,9-11.
60. BhSS 8.25.13.
lesser rituals but rather as a “great sacrifice” or mahdayajna, that is, a

soma sacrifice.®! In similar fashion, another passage in that Brahmana
equates the new and full moon sacrifice with the soma sacrifice, conclud-
ing with a declaration that cannot help but recall the passages in ques-
tion in the later Sitras: “By performing the new and full moon sacrifice,
he wins as much as is won by performing a soma sacrifice. The [ritual]
thus becomes a great sacrifice (mahdayajna).”
Vedic resemblance thus allows for two different kinds of ritual con-
densations, both of which entail a kind of synecdochic reductionism
(whereby a part of the whole represents the whole) but differ regarding
what kind of part is made to represent the whole.
The first type of this synecdoche, which we have seen at work
before and which is readily understandable in hierarchical terms, is the
encompassment (or claim of encompassment) of the condensed essences
of lesser rituals within greater ones—a condensation upward, so to
speak. I observed above such a claim for the human victim, which is said
to encompass within it all other pasus, and for the soma sacrifice, which
contains within it all other sacrifices, both the lower and the higher.®
Elsewhere the soma sacrifice is said to be the “the most perfect” (sam-
pannatama) of all Vedic rituals on account of its all-encompassing na-
ture.“ Other rituals are similarly advertised. The rites within the
Agnicayana are assimilated into the rituals of the Vedic repertoire, and
thus “he obtains all the sacrifices with the Agnicayana.”® The Rajastya
has the same capacity: “He who offers the Rajastya envelopes all sacrifi-
cial rituals.”® Such great rituals make their claim to superiority in part
61. SB 11.7.2.2
62. SB 2.4.4.11-14. See TS 1.6.9.1-2, where one “who knows thus” is said to
obtain the results of the agnistoma, the ukthya, and the atiratra by offering, respectively,
the agnihotra, the full moon sacrifice, and the new moon sacrifice. See also G. U. Thite,
Sacrifice in the Braahmana-Texts, pp. 49-50. The author observes, “In order to elevate
any sacrificial rite efforts were made to show how the rite has some connection with
Soma.” For equations between the hierarchically lowest Srauta ritual, the agnihotra, and
the soma sacrifice, see H. G. Bodewitz, The Daily Evening and Morning Offering,
pp. 125-35. For claims of equivalency between various rites of the domestic upanayana
and the subsequent period of Veda study on the one hand and the Srauta rituals on the
other, see Brian K. Smith, “Ritual, Knowledge, and Being,” pp. 79-80, and Chapter 8
herein.
63. AitB 3.40-41.
64. AitA 2.3.3.
65. SB 10.1.5.1-3.
66. SB 5.2.3.9.
by virtue of the fact that they encompass within them the condensed
kernels of all other Vedic sacrifices.®
The second form of condensation within Vedic resemblance is less
obviously in harmony with hierarchical presuppositions. This is the type
we have been observing in this chapter: a condensation downward of the
essences of superior victims or superior sacrifices which are reprised
within inferior “equivalents.” Claims of equivalency between “great”
prototypes and their condensed counterparts, in the Sutras as well as in
the Brahmanas, can also be explicated in terms of hierarchical resem-
blance. They serve to unite Vedic ritualism from top to bottom, and vice
versa. The homologies are not intended to collapse distinctions but
rather to strengthen connections between interrelated phenomena. In
such a system, even the smallest and relatively weakest rite or ritual may
function as a resembling counterpart to the greatest and most complete
version of the rite or ritual. The inferior is not here regarded as an
equivalent replacement for but rather only as a condensed representa-
tive of its superior relative within a class. It is in this sense, and in this
sense only within the context of the “unity of ritual” that is Vedism, that
one may speak of the domestic rituals as “substitutes.”
There is another, more practical end served by drawing links be-
tween the great and the small, the expanded and the condensed. Recall-
ing that it is the “resemblance of purpose” (arthasadmanya) which is the
most important and overriding form of resemblance in Vedism, and that
substitution is most especially legitimate when it is necessary to bring
about the completion of the obligatory (nitya) sacrifices, we may now
turn again with new understanding to obligatory domestic rituals that
replicate obligatory frauta rituals. For it would seem that one important
function of domestic replicas is to act as substitutes for nitya rituals for
those ritualists at the lower levels of the Vedic ritual—and social and
economic—system; and the supreme resemblance, the resemblance of
purpose, is the one that links obligatory rituals at the two levels. Resem-
blance, it would seem, served as the glue for a certain kind of systemic
solidarity, binding together within a ritual, social, and economic whole
all the levels of the hierarchy.
By at least the time of the Grhya Sutras, if not before they were so
67. For the performance of lesser srauta rituals in the course of the performance of
the soma sacrifice, see Jan Gonda, The Haviryajnah Somah: The Interrelations of the Vedic
Solemn Sacrifices. Sankhayana Srauta Sutra 14, 1-13. Translation and Notes (Amsterdam:
North-Holland, 1982).
documented, certain of the nitya sacrifices of the Srauta cycle were per-
formed, in condensed forms, as part of the obligatory rituals of the domes-
tic cycle. Two of the most important of these, for both cycles, were the
daily agnihotra and the bimonthy new and full moon sacrifice. Called by
the same name, based on the same ritual procedures, and connected by
the “resemblance of purpose,” the agnihotra and the new and full moon
sacrifice were to be performed by both the ekdgni and the ahitagni.
The agnihotra ritual was correlated to the rising and setting of the
sun, and, among its other purposes, it was vital to the perpetual mainte-
nance of the fire(s) (the “fire worship” or agnyupasthana). With the
daily agnihotra, the ritualist nurtures the continuity of fire, of sacrifice,
and, in the minds of the ritualists, of life itself.6® H. W. Bodewitz, who
believes that the ritual was primarily meant as a transference of solar
light into the fire, also recognizes that “the fires are related to the
ahitagni himself, the hearth is connected with the head of the family, and
the fires are identified with the pranah [life breaths] of the sacrificer. As
a result the daily agnihotra offered in these fires may be explained as a
daily confirmation of this relationship.”
Equally essential for the maintenance of the sacrificial vow and for
the continuity of the sacrificer’s ritually constituted being was the new
and full moon sacrifice. It is required of the ahitagni for “as long as he
lives, or for thirty years, or until he becomes too old.””” Whereas the
agnihotra is synchronized to the daily rising and setting of the sun, the
new and full moon sacrifice, as its name suggests, is calibrated to the
lunar cycle. And the latter, like the former, “once begun should never
be discontinued.””! The agnihotra and the new and full moon sacrifices
are the most important of the obligatory frauta rituals and must be
continuously performed by ahitagnis of all ranks.”
68. P.-E. Dumont writes, “The daily obligation to offer the agnihotra is essential,
for it seems that one of the principal objects of the agnihotra is to perpetuate both the
continuity of the sacrifice and the continuity of the race of the sacrificer.” L’Agnihotra:
Description de l'agnihotra dans le rituel védique (Baltimore: Johns Hopkins University
Press, 1939), p. vii.
69. Bodewitz, The Daily Evening and Morning Offering, p. 2.
70. ApSS 3.14.10.
71. BhSS 3.13.11.
72. See, e.g., HSS 3.1: “(For the Brahmin, Ksatriya, and Vaisya] the agnihotra and
new and full moon sacrifices are obligatory. A soma sacrifice [is also obligatory] for the
Brahmin. From the time the fires are set up, the agnihotra and new and full moon
sacrifices are obligatory. For a non-Aryan chief (nisdda) and a chariot-maker (rathakara),
the agnihotra and new and full moon sacrifices become mandatory from the time the fires
These two obligatory Srauta sacrifices, among others, were repli-

cated in domestic ritualism with a formally simplified and relatively
inferior procedure. The framework (tantra) for the performance of all
domestic rituals was a contracted rendition of the more expanded and
full framework of the srauta sacrifice. With this condensed procedure,
the ekagni performs rituals that parallel those of his prototype, the
ahitagni, and that allow him to participate in the Vedic cult of fire
sacrifice and all that it entailed.
In one Grhya Sutra,” a reduced and domesticated version of the
establishment of the fires (agnyddheya or agnyadhana) is enjoined for
the ekagni. The procedure for its performance, the text notes, is known
from the srauta procedure: “All [rites here are performed as in the
ritual] that contains the eating of the cdtusprasya food,” by which is
clearly meant the agnyddheya depicted in the school’s Srauta Sitra.
Other domestic Sutras give instructions for reestablishing the fire (the
punaragnydadheya) should it go out or be neglected for twelve days or
more. For this ritual, too, the texts refer back to the frauta model,
noting also the adaptations necessary when applying the model to the
domestic situation.”
The domestic agnihotra and new and full moon sacrifices are simi-
larly adaptations—and condensations—of Srauta prototypes. Although
the agnihotra is performed at the same time and offerings are made to
the same deities at both levels of Vedic ritualism, the oblation in the
domestic sacrifice is not the milk obtained from the “agnihotra cow”
every ahitagni is required to own. Rather, here the prescribed oblation
consists only of uncooked rice and barley grains, offered with the hand
(and not with a ladle) by the sacrificer himself (and not by an officiating
priest).7> I have already noted (in Chapter 6) some of the rites elimi-
nated and abbreviated in the condensed performance of the domestic
version of the new and full moon sacrifice. Other modifications include a
simple offering of cooked food in a dish (sthalipaka) to Agni, replacing
are set up.” Most texts list seven nitya sacrifices for the ahitdgni, regardless of class: the
agnyadheya (establishing the fires), the agnihotra, the new and full moon sacrifices, the
seasonal dgrayana sacrifices of first harvests, the quarterly cdturmdsyas, the twice-yearly
animal sacrifice, and an annual soma sacrifice. For this last obligatory sacrifice as only so
mandated for Brahmins, see HSS above and KSS 1.2.12 (“according to some”).
73. PGS 1.2.1-13.
74. See, e.g., HGS 1.7.26.4-18.
75. For these and other differences between the Srauta and grhya agnihotras, con-
sult Bodewitz, The Daily Evening and Morning Offering, pp. 191-204.
the Srauta requirements of cakes to Agni, Agni-Soma, and Indra, and an

offering of clarified butter to Prajapati.”
In addition to these, the most fundamental and necessary of the
obligatory sacrifices, other nitya rituals are also replicated in domestic
miniatures. The seasonal sacrifices (4grayanas) are performed at the
same times, and offerings are made to the same deities, by both the
Ghitagni and the ekagni. But in this case, as in all others, the domestic
ritual is a much abridged variant of the more complete Srauta model.”
Animal sacrifices at the domestic level, including the yearly Stilagava (a
sacrifice of an ox or cow to Rudra) and the astakds (in which a cow is
sacrificed for the ancestors), are directly based on—and are collapsed
forms of—the twice-yearly Srauta animal sacrifice.’
In one remarkable section from a domestic ritual text of the
Baudhayana school,” instructions are given for domestic versions of the
full complement of obligatory Srauta rituals. Some general rules for
making the proper adjustments necessary for the domestic replication of
these sacrifices are noted:
Where there [in the srauta ritual] there are cakes, here [in the domestic
replica] there are oblations of cooked rice porridge (caru). The divinities,
the proper invitation verse (puronuvakya) and sacrificial verse (ydjya) are
the same [in both versions]. Where there there is the cry “vasat!” at the
time of offering, here there is the exclamation “svaha!”*°
Specific details are also provided in the text for the miniature perfor-
mance of the particular rituals of the Srauta cycle, including the method
for offering a vegetarian “animal sacrifice” (which makes the domestic
ritualist “the equal of the sacrificer of the animal sacrifice”) and the way
to offer a grhya “soma sacrifice” accomplished without the use of soma.
All these obligatory sacrifices were performed in their fullness only
by the ahitdgni utilizing the full array of three to five empowered fires,
76. Descriptions of the deities and offerings that comprise the principal oblations at
the Srauta new and full moon sacrifices may be found in Alfred Hillebrandt’s Das
Altindische Neu- und Vollmondsopfer (Jena: Gustav Fischer, 1879), pp. 107-16; and in
Urmila Rustag’s more recent work, Darsapurnamasa: A Comparative Ritualistic Study
(Delhi: Bharatiya Vidya Prakashan, 1981), pp. 235-46. For the domestic version of the
ritual, see Jan Gonda, Vedic Ritual, pp. 346-53.
77. The domestic Agrayana is performed with a sthdlipaka, and not a cake as in the
Srauta ritual, made from the newly harvested grains of each season. See Kane, History of
Dharmasastra, I1: 827-29; and Gonda, Vedic Ritual, pp. 430-32.
78. See AsvGS 1.11; MGS 2.4; KGS 51.
79. BGPariS 2.1.
80. BGPariS 2.1.37.
sacrificing the proper oblation material to the full number of relevant

divinities, and performing the entire ritual according to the more com-
plex and inclusive §rauta procedure. The domestic replicas, condensed
substitutes at a lower level, are but bleached-out simplifications of their
prototypical models.
As integral components of a unified and comprehensive Vedic
ritual system, the domestic rites were resembling “equivalents” or “sub-
stitutes” for those who could do no better; for those who, because of
lack of wealth, education, experience, or interest and desire, were
unqualified for participation in the Srauta cult. Within the “unity of
ritual,” that is, when seen in relation to Srauta ritualism, the domestic
sacrifice takes its place as the simple and basic form of Vedic ritualism,
a formally resembling but impoverished and incomplete manifestation
of superior ritual expressions. But as condensations of those higher
ritual forms, the domestic sacrifices (and those who engaged in them)
also found a place, humble though it was, within the Vedic world of
hierarchical resemblance.
From the Inferior to the Replacement: The Domestication of the

Vedic Sacrifice
It is not at all an easy thing to pinpoint when, historically, the Vedic

unity began to crumble and new systems of religious belief and practice
overshadowed and replaced it. As I have argued in Chapter 2 and as I
will maintain again in Chapter 8, there is a real sense in which Vedism
never did die but merely transmigrated into new bodies. Still, there is no
question that the subsequent incarnations of Vedism were significantly
different from the original (as counterparts differ from their proto-
types), and, by way of concluding this survey of that original form of
Vedism, it might be useful to reflect on the first signs of its death and its
earliest rebirths.
One possible site for locating the first shift from one system, one set
of religious assumptions and organizing principles, to another is the
forest of sixth-century B.c.E. India—the locale of the world renouncers
of both the “orthodox” and “heterodox” traditions. The latter, princi-
pally those that eventually emerged as what we call Buddhism and
Jainism, radically broke away from the Vedic ancestor in the course of
time, although they did not and could not ignore its legacy. The “ortho-
dox” traditions of world renunciation, most especially those based on
principles of monism (later to be designated Vedanta), are even more
complicated cases of continuity and change vis-a-vis the Vedic system

focused on here.
To some scholars it would appear that Vedic resemblance, govern-
ing a plethora of interlocking hierarchical registers, found its necessary
end in the monistic thought represented in the early Upanishads. For
these students of Indian history, the latter texts are continuous with, and
are the teleological conclusion of, the earlier ritualistic thought. In the
Upanishads, universal resemblance is brought to its logical terminus:
universal identity. The complex system of connections between resem-
bling phenomena, the web of bandhus integral to Vedic ritualism, and
hierarchical distinctions are collapsed in monistic thought into the ulti-
mate connection: the equation of self and cosmos (without the ritual
intermediary) formulated as the identity and full equality of dtman and
the brahman.
In this view, the ritualists are retrospectively seen as seeming “to
circle a final, long-sought identity, preferring to postpone recognition of
that ultimate correspondent in which all oppositions cease.”*! Jan Hees-
terman, in his seminal article “Brahmin, Ritual, and Renouncer,” takes
a somewhat different view of the continuity between the ritualistic texts
(the Brahmanas and Sutras) and the texts of the world renouncers.
Heesterman argues that ritualistic thought developed through three dis-
crete periods: a largely hypothetical “preclassical” period characterized
by agonism and reciprocity; a “classical” era in which the ritual is “indi-
vidualized” and reciprocal exchange eliminated; and, as the culmination
of the Vedic sacrificial ideology, a third period of “interiorization” and
world renunciation represented by the Upanishads. According to Hees-
terman, ritualistic thought
had to advance to its logical conclusion, that is the interiorization of the
ritual which makes the officiants’ services superfluous. . . . It would seem
to me that here we touch the principle of world renunciation, the emer-
gence of which has been of crucial importance in the development of
Indian religious thinking.*
Heesterman has made what is probably the best case to date for the
“orthogenetic, internal development of Vedic thought” extending to and
climaxing in the absolute independence of the sacrificer by means of
philosophical monism and renunciation of social existence. “The point I
81. David Knipe, In the Image of Fire: Vedic Experiences of Heat (Delhi: Motilal
Banarsidass, 1975), p. 33.
82. Jan Heesterman, “Brahmin, Ritual, and Renouncer,” p. 39.
want to stress,” he writes, “is that the institution of renunciation is

already implied in classical ritual thinking. The difference between classi-
cal ritualism and renunciation seem to be a matter of degree rather than
of principle.”83
Perhaps the case for a certain discontinuity, for the Upanishads as
emblematic not of an extension of Vedic ritualism but of its demise as
the dominant worldview of ancient India, can be made on the basis of
my work here. Ihave located the heart and soul of Vedic ritualism in a
principle at odds with that underlying monism. In the Upanishads, one
might be witnessing the conclusion of Vedism not in the sense of its
culmination but in the sense of its destruction. In the proto-Vedantic
view, the universe and ritual order based on resemblance has collapsed,
and a very different configuration based on identity (abhorred by the
Vedic ritualists as the “excess of resemblance,” jami) has emerged.
Upanishadic monism, one might say, blew the lid off a system con-
tained, as well as regulated, by hierarchical resemblance.
The formulation of a monistic philosophy of ultimate identity—
arguably one indication of Vedism dissipating and reforming into a new
systematic vision of the world and its fundamental principles—was born
outside of normative classification schema of Vedic social life and be-
came institutionalized as a counterpoint to life in the world. But another
locus of Vedism in transition and transformation might also be identi-
fied. It might be maintained that Vedic ritualism survived the Vedic
period by becoming a part of a larger whole, Hinduism, and did so not
by being excluded from social life but by being thoroughly integrated in
it, that is, by being domesticated. Perhaps Vedism culminated not in the
forest but in the home.
Beginning at approximately the same time as the first appearance of
the world renouncers and their new philosophies, and coinciding with
the production of the ritual Sutras, new forms of political power put the
older ideological hegemony of the Brahmin ritualists in jeopardy. The
Buddhist influence at the highest levels of polity, most dramatically
realized in but not ceasing with ASoka Maurya and his pan-Indian reign,
in all likelihood made the continued performance of large and ostenta-
tious Vedic sacrifices (with their overt statements of overlordship) im-
politic at best, if not simply impossible. The adherents of and successors
to the Vedic ritual system may have had to make some adjustments and
accommodations in light of these political and religious changes.
Through the time of the composition of all but the most recent
83. Ibid., p. 41.

Grhya Sitras, the domestic ritual was presented as in the shadow of the
$Srauta cult. Somewhat later, and perhaps under the very different circum-
stances suggested here, the domestic ritual is represented rather differ-
ently. In texts such as the Manu Smrti (ca. second century B.C.£. or
later), one of many literary productions of the classical Hinduism ori-
ented around caste duty and the obligations of the householder, the
domestic sacrifice or pakayajna appears not as an inferior version of
other and greater rituals but rather as the epitome of and effective
replacement for them; and the whole system of domestic ritualism itself
was concentrated into five daily activities of the householder.
The orthodox Hindu strategists inherited the means to rethink and
revalue domestic ritualism from Vedic ritualism: condensation, minimal-
ism, and the representation of the simple as the complex through
synecdochical reductionism. Set within a context in which the Srauta
rituals—and especially the greatest of these, the soma sacrifices and the
royal rituals—were largely anachronisms, these features of an older hier-
archical system took on new significances.
As early as the Satapatha Brahmana* and the Taittirlya Aranyaka,®
one reads of five “great sacrifices” (mahdayajnas) by which a householder
daily acquits himself of his ritual obligations to the gods, the ancestors,
the spirits, his fellow human beings, and the brahman. These “great
sacrifices” are also labeled “great extended sacrificial sessions”
(mahasattras), but only “by way of laudation,” as one text notes.8¢ For
within the hierarchical system, these “great” sacrifices, like other domes-
tic rituals also so designated, were in fact extremely simple forms of Vedic
ritualism:
He should daily offer [an oblation into the fire] with the exclamation
“svaha!” even if it is done only with a stick of wood. That fulfills the
sacrifice to the gods. He should daily offer [an oblation to the ancestors]
with the exclamation “svadha!” even if it is only a vessel of water. That
fulfills the sacrifice to the ancestors. Daily he should pay respect [to the
spirits], even if it is only [by offering them] some flowers. That fulfills the
sacrifice to the spirits. Daily he should give food to Brahmins, even if it is
only some roots, fruits, or vegetables. That fulfills the sacrifice to humans.
And every day he should recite [a portion of] the Veda, even if it is only
the syllable “om”. That fulfills the sacrifice to brahman.*?
84. SB 11.5.6.1 ff.

85. TA 2.10.
86. ApDhS 1.4.12.15: “In reference to these [five sacrifices] there is the laudatory
statement (samstuti) [that refers to them as] ‘mahdayajnas’ and ‘mahasattras.’ ”
87. BDhS 2.6.11.2-6; see also $B 11.5.6.2; TA 2.10.
As Charles Malamoud puts it, “The lesson of the five great sacrifices is
that there exists a minimum ritual necessary and sufficient so that the
daily religious obligation may be completely accomplished.”* The recipi-
ents of these five great sacrifices are the sum total of all eligible “eaters
of oblations” in Vedic ritualism; by concentrating sacrifices to them into
one daily sequence, the mahdyajnas in effect represent a minimalistic
form of Vedic ritualism as a whole for householders with limited means.
P. V. Kane writes, “Every man could not afford to celebrate the solemn
Srauta rites prescribed in the Brahmanas and Srauta Sutras. But every
one could offer a fuel-stick to fire that was deemed to be the mouth of
the great Gods of Heaven and thus show his reverence and devotion to
themi:”"?
And if the five mahdyajnas were the minimal condensed forms of
Vedic ritualism as a whole, they were also, even within the hierarchical
Vedic ritual system, the representations of the domestic cycle of sacri-
fices in its entirety. In one classification of the different types of grhya
rituals, the subclasses or samthas comprising the domestic level of sacri-
fice, the categories are suspiciously akin to four of the five daily “great
sacrifices”:
[A domestic ritual is designated] a huta by virtue of the daily agnihotra
oblation [to the gods]; an ahuta by virtue of the rice ball offerings [to the
spirits]; a prahuta by virtue of the rites to the ancestors; and a prdsita is the
oblation made in [the mouth of] a Brahmin.”
The only mahdyajna not potentially covered in this classification scheme,
this fourfold typology of the domestic ritual as a whole, is the “sacrifice of
speech” or the daily Vedic recitation, a sacrifice that may very well be
considered as included within other rituals in the form of the accompany-
ing mantras. Although the correlation here between the types of domestic
ritualism and the five daily great sacrifices is not explicit, it is, I think,
implied. The domestic ritual, the pdkayajna, itself the diminutive repre-
sentative of the Vedic ritual system, is represented in this text as further
88. Charles Malamoud, Le Svadhyaya: Recitation personnelle du Veda (Paris: Insti-

tut de Civilisation Indienne, 1977), p. 14.
89. Kane, History of Dharmasastra, Il 697-98.
90. SGS 1.10.7; see also SGS 1.5.1; PGS 1.4.1; JGS 1.1; KGS 13.2. The classifica-
tion at AgvGS 1.1.2 is similar. Another and competing method of classifying the types of
domestic rituals organizes them into seven categories, paralleling the seven subclasses of
the Srauta and soma ritual. See BSS 42.4; BGS 1.1; GB 1.5.23,25; VaikhGS 1.1; GautDhS
8.18; and Sudarsana’s commentary on ApGS 1.1; Sayana’s commentary on TS 1.7.1.1; and
Kane, History of Dharmasastra, I: 194n.
reduced into its quintessential kernel—five “great” sacrifices that may be

performed with a piece of wood, a glass of water, some flowers and fruits,
and by saying “om.”
The correlation of the divisions of the domestic sacrifice and those
of the “great sacrifice,” made only implicitly in the Sankhayana Grhya
Sutra cited above, is explicitly formulated in Manu Smrti.®! In that post-
Vedic and classical Hindu text, one reads:
Teaching the Veda is the sacrifice to brahman, the offerings of food and
water [the tarpana] is the sacrifice to the ancestors, the oblation in the fire
is the sacrifice to the gods, the rice ball offering is the sacrifice to the
spirits, and the honoring of guests is the sacrifice to humans. . . . These
five sacrifices are [also] known as the ahuta, the huta, the prahuta, the
brahmyahuta, and the prasita. Ahuta is the repetition of the Veda, huta
the oblation offered in fire, prahuta the rice ball offering to the spirits,
brahmyahuta the reception of twice-born [guests], and prasita the satiating
of the ancestors.”
In this text, the condensation of the whole domestic ritual, in all its
typological forms, into the five simple daily sacrifices of the householder
is complete. But what is at least as significant is the fact that the ratio-
nale for the performance of the mahdyajnas is different here. In the
Grhya Sutras and other texts of Vedic ritualism, the five great sacrifices
were a minimalistic method for the domestic sacrificer to pay back,
sacrificially, his “debts” to the various classes of beings in the universe.
In Manu, however, as Madeleine Biardeau observes, “The new idea—
which is added without replacing the old idea of ‘debts’—is that the life
of the householder implies an inevitable ‘violence’ with regard to living
beings.” The householder, we are informed, has five “slaughter-
houses” in the home: the hearth, the grinding stone, the broom, the
mortar and pestle, and the water vessel. These five daily “bind” him
with the sin of killing. It is, according to Manu, for the purpose of
91. It is likely that the similarity between the classification system in the SGS and
that in Manu is not just coincidental. Ram Gopal, in “Manu’s Indebtedness to Sank-
hayana,” Poona Orientalist 27 (1962): 39-44, analyzes a number of parallel passages in the
two texts and concludes that “the author of the Manu Smrti who drew upon the
Sankhayana Grhya Sitra . . . was probably a follower of the Sankhayana Sakha of the Rg
Veda and not a follower of Maitrayaniya Manava Sakha.” The question of Manu’s Vedic
affliation, if any, remains controversial, however.
92. Manu 3.70, 3.73-74.
93. Biardeau, Le Sacrifice dans I’'Inde ancienne, p. 42.
“redemption” (niskrtyartha) that the five great sacrifices, the essence of

the domestic ritual as a whole, are daily prescribed.™
“The main point of the religious activity of the Brahmins,” writes
Biardeau, “amounts to a system of expiations.”®5 The domestic sacrifice
is transformed—through its new guise as the five great sacrifices—from
a small-scale resembling form of ritual activity conceived of as construc-
tive activity to a defensive countermeasure against the “sin” of killing, of
himsa (perhaps under the influence of Buddhist and Jain valorizations of
ahimsa).
With this new vision of the purpose of sacrifice, and with this new
use of the old strategies of condensation and minimalism, we have per-
haps crossed the invisible boundary that sets Vedic ritualism apart from
its Hindu successors. Texts such as the Manu Smrti present the domestic
sacrifice as the quintessential sacrifice and the five great sacrifices as the
quintessence of the domestic sacrifice, and they do so in order to domes-
ticate the Vedic sacrifice and encompass it within a new religious configu-
ration. Furthermore, the daily life of the domesticated householder
takes on new meanings when it is no longer evaluated in relation to the
being and praxis of the ahitagni, the Srauta sacrificer. “The ideal image
of the yajfia is the superior yajna,” writes Charles Malamoud. “The
domestic yajna ordinarily is a reduced and impoverished copy.” But by
epitomizing the domestic sacrifice by five “great” sacrifices, texts such as
Manu’s “present the copy as superior to the model.” Resemblance
here, too, as in the proto-Vedantic Upanishads, gives way to identity as
the inferior substitute becomes, pears out of a certain necessity, the
effective replacement.
Biardeau has argued that with the development of the meaning and
practice of the five great sacrifices we are confronted with “un usage
dérivé, symbolique” of the term sacrifice in Indian history, “une pre-
miére extension de |’application du terme yajna.”*’ As will be seen in the
next chapter, this was indeed only the beginning of the Hindu appropria-
tion of the category of sacrifice, and it is in this manner, finally, that the
real destiny of Vedic ritualism may be traced.
94. Manu 3.68-69.

95. Biardeau, Le Sacrifice dans I’Inde ancienne, p. 42.
96. Malamoud, Le Svadhyaya, p. 21.
97. Biardeau, Le Sacrifice dans I’Inde ancienne, p. 41.
e)
The Destiny of Vedism
Vedism: Dead or Alive?
The Vedic sacrifice in post-Vedic India has been pronounced dead by

Indologists often and early, too early. The pioneers of Western knowl-
edge of the Vedic ritual—Hillebrandt, Lévi, Caland, and others—
assumed their subject matter was speaking to them from the grave. Those
experts regarded performances of Vedic sacrifices recorded in the nine-
teenth and early twentieth centuries as, at best, a kind of necrophiliac
theater performed by Brahmins reading from a script.
Scholars more recently—C. G. Kashikar, Frits Staal, Frederick
Smith, Asko Parpola, Michael Witzel, and David Knipe, to mention a
few—have demonstrated that the yajna was relegated to a historical
tomb a bit prematurely. To paraphrase Mark Twain’s reply to the Associ-
ated Press, evidence of enclaves of Vedic ritualists in present-day India
has proved that such reports of death are greatly exaggerated. Vedic
rituals are in no sense popular or widespread in contemporary South
Asia, but fieldwork data are now being accumulated which reveal a
surprisingly large number of living traditions of Vedic ritualists through-
out India.!
Too often, however, fieldwork on current performances and per-
formers of Vedic yajnas has tended merely to repaint the death mask
1. See, e.g., C. G. Kashikar and Asko Parpola, “Srauta Traditions in Recent

Times,” in Frits Staal, ed., Agni: The Vedic Ritual of the Fire Altar, 2 vols. (Berkeley:
Asian Humanities Press, 1983), II: 199-251. Other reports from the field should be forth-
coming shortly.
The Destiny of Vedism 201
rather than strip it off. Contemporary exempla of Vedic ritualism have

been too frequently presented as surviving fossils of the distant past,
wholly irrelevant to the religions of India now—and to those of the past
two millennia. The yajna, it is said with a certain commonsensical persua-
siveness, is at most an antique relic for post-Vedic religions; when it is
not simply ignored as a museum piece of no contemporary interest or
usefulness, it is sadly misunderstood. The now world-famous Agni-
cayana held in Kerala in 1976 attracted crowds of South Indian Hindus
who betrayed their ignorance of the ritual, at least in the eyes of the
congregated foreign scholars taping and filming the event, by mistakenly
perceiving the yajna as a strange kind of puja. They are reported to have
come to the sacrifice
in a spirit and mood of devotion. Among these outsiders [sic/] there was a
strong tendency to conceive of the ritual enclosure as a kind of temple or
place of pilgrimage, where devotees could receive “darshan” (a glimpse of
a presiding deity), make a donation, perform a circumambulation, and
return home with freshly acquired religious merit.*
The experts winced in embarrassment for the sad naiveté of Hindus who
in this way so blatantly manifested their misperception of the “true
meaning” of the Vedic sacrifice.
The yajna, it would seem, does not participate in what continuity
there is in the three-thousand-year-old Vedic religious heritage. When
Jan Gonda published his Change and Continuity in Indian Religion, he
included chapters on mayd, the “Isvara idea,” gift-giving, the guru, and
even the number sixteen as exempla of continuity—but no chapter on
sacrifice. Similarly, W. Norman Brown, in his quest for those cultural
entities “which in each successive periodic reincarnation of the civiliza-
tion [have] caused the new existence of the civilization,” did not enumer-
ate among them the ideology and practice of yajna.?
Sacrifice, as a fossilized archeological datum discrete from post-
Vedic Hinduism, is also not found in those definitional statements on
Hinduism examined in Chapter 1 which focus on certain pan-Hindu
religious themes. As observed above in the discussion of various ap-
proaches to defining Hinduism, scholars of one school are persuaded
that the unity of the religion is to be found in certain shared religious or
2. Frits-Staal, “The Agnicayana Project,” in Staal, Agni, II: 469.

3. W. Norman Brown, Man in the Universe. Similarly, Franklin Edgerton, in “Domi-
nant Ideas in the Formation of Indian Culture,” Journal of the American Oriental Society
62 (1942): 151-56, does not count yajria as such a “dominant idea.”
philosophical themes. Some forms of this strategy for defining Hinduism

identify the concepts of transmigration (samsdra), karma, and liberation
from transmigration and karma (moksa, mukti, etc.) as definitive of the
religion. Other versions of this type of the thematic approach to defining
Hinduism explore “key concepts” such as brahman and dharma in addi-
tion to samsara, karma, and moksa. In none of these sorts of attempts to
define Hinduism thematically, however, does yajna figure in the array of
definitional themes, and the way the recently discovered continued
existence of Vedic ritualism is being encoded (archaic, peripheralized,
and ultimately irrelevant to Hinduism) will not convince anyone of its
definitional importance to the configuration that is Hinduism. The
Vedic sacrifice, despite its newly discovered ongoing existence in certain
traditions of present-day India, is for all intents and purposes regarded
by observers today—as it was by the scholars of the past—as a relic of
the dead past.
In this concluding chapter, I present a short argument for the past
and continued importance of yajfa in India—and not just as a ritual
performed by a few offbeat Brahmins. I will begin with evidence that
yajna is indeed an entity that has transmigrated throughout the history
of post-Vedic Indian religious discourse, in a variety of historical settings
and divergent traditions. It has done so, in the first place, as a category
of explication and, subsequently, as a category of traditionalization. It
must be included, I will maintain, in any discussion of the possible
“continuities” stretching over the millennia in Indic civilization. Sec-
ondly, sacrifice often plays a crucial role in the self-definition of Hindu
religious institutions and practices as Hindu. It is perhaps one of only
two such categories that can perform such a canonical function in Hindu
religious life, the other being the category Veda.
My argument depends on viewing yajna in the history of post-Vedic
Indian religions not so much as a ritual act or set of acts (the purpose of
which might be constructive, expiatory, propitiary, purifying, cathartic,
etc.) but rather as a category that acts to provide explanatory power,
traditional legitimacy, and canonical authority.
It has been said that the essential function of sacrifice the world
4. One might partially exempt Madeleine Biardeau’s Le Sacrifice dans I’Inde, which
has as its thesis that sacrifice is “un tronc commun de l’hindouisme” (p. 13) and “que
l’action rituelle, sans cesse répétée, est le modéle méme de toute action, et que le sacrifice,
mode de communication de la terre avec le ciel, en est le centre. Tout peut devenir
sacrifice” (p. 153). Biardeau’s claims for the continuity of sacrifice in Indian religious
history differ from those presented here, however, and she is reluctant to go so far as to say
that yajria is a definitional element of post-Vedic Hinduism.
3 The Destiny of Vedism 203
over is to consecrate, to make sacred; and, indeed, the etymology of the

word sacrifice testifies on behalf of this position (sacer-facio, “to make
sacred”). Sacrifice, as a category in Indian religious thought, does indeed
make sacred: it projects traditional and canonical meaning onto a poten-
tially infinite set of human religious activities and doctrines, thereby con-
secrating and legitimizing them. One might say that the imaginary entity
called the Indian mind often thinks with a sacrificial consciousness.5
Hindu Uses of the Vedic Sacrifice
Let us begin with a survey of history of yajria as an explanatory category:

a mode of classification that renders phenomena explicable by connect-
ing them to an already assimilated precedent. For Indian writers and
thinkers from Vedic times to the present, yajfa is, contrary to some
Western opinions, in a real sense a datum or known quantity. It might
appear from what follows that sacrifice functions as a metaphor in In-
dian religious discourse, but it could equally be said that from the Indian
point of view other beliefs, actions, and phenomena are metaphors of
the sacrifice. That is to say, it is quite as reasonable to regard the Vedic
sacrifice as a model for understanding other phenomena as it is to as-
sume that it is only a model of other phenomena.®
The projection of the sacrifice from its ritual confines onto various
aspects of human life and of the world at large begins, of course, in the
Veda itself. The theory of ritual in Vedic texts, as we have seen in Chap-
ters 3 and 4, centers on the notion that sacrifice is not only the activities of
the priests (the adhiyajna level of meaning) but is also the means for
explaining the workings of the cosmos (the adhidevata level of meaning,
“relating to the macrocosmos”) and the human self (the adhydtman level
of meaning, “relating to the self, the microcosmos”). Sacrifice is thus the
primary category of knowledge in both Vedic cosmology and Vedic an-
thropology, as is indicated in the frequently encountered phrase puruso
vai yajnah, “man [cosmic and human] is the sacrifice.”
5. The phrase sacrificial consciousness I owe to my friend and colleague Holland

Hendrix, who employs it in reference to the development of early Christian doctrine and
practice.
6. Model of and model for are, of course, Clifford Geertz’s terms for what he
regards as the two principal functions of religion in cultural life: to provide both intellectu-
ally plausible and emotionally satisfying modes of living. Consult his “Religion as a Cul-
tural System,” in The Interpretation of Cultures: Selected Essays by Clifford Geertz (New
York: Basic Books, 1973), pp. 87-125.
The sacrifice as an explicatory model for understanding the cosmos

is dramatically exemplified in the well-known Purusa Sukta, RV 10.90,
in which the creation of the universe is represented as the primordial
sacrifice and dismemberment of the Cosmic Man. This myth—as well as
the related cosmogonic myths of Prajapati’s fragmentation of sacrificial
reconstruction discussed in Chapter 3—provides a blueprint etched by
the yajna for tracing the origins of the cosmos.
But it is the epistemological function of the Vedic sacrifice, in the
cosmological mode, that interests us at this juncture. This function of
the sacrifice serves to explain the cosmos in both its origins and its
ongoing operations. These latter are also comprehended through appli-
cation of the sacrificial paradigm, as the early Upanishad testifies:
The world [of heaven] over there, Gautama, is a fire. The sun is its fuel,
the rays of the sun are its smoke, the day is its flame, the moon its coals,
the stars its sparks. Into this fire the gods offer faith, and from that
oblation king Soma is born. The rain cloud, Gautama, is a fire. The wind
is its fuel, the mist its smoke, lightning its flame, the thunderbolt its coals,
and the roar of the thunder its sparks. Into this fire the gods offer king
Soma, and from that oblation rain is born. The earth, Gautama, is a fire.
The year is its fuel, space its smoke, night its flame, the four directions its
coals, the four intermediate directions its sparks. Into this fire the gods
offer of rain, and from that oblation food is born.’
Such a sacrificial comprehension of the workings of the cosmos often

assumes a reciprocal arrangement among the cosmic forces, one sacrific-
ing to another in order to keep the universe functioning in its regularity:
The fire here [on earth] offers itself in the rising sun; the setting sun offers
itself in the fire; the night also offers itself in day, and the day [offers itself]
in night; the in-breath [offers itself] in the expiration, and the expiration
[offers itself] in the in-breath. These six sacrifice themselves each in the
other.
Indeed, as the Satapatha Brahmana puts it, “this all (tat sarvam) partici-
pates in (@bhakta) the sacrifice,”® and “this all” is made explicable when
conceived in such a manner.
The sacrificial arena as the model both of and for the organization
of the cosmos persists long after the Vedic age has come to a close,
providing us with examples of the post-Vedic exercise of the “sacrificial
7. ChU 5.4.1-5.6.2.
8. KB 2.8.
9. SB 3.6.2.26.
consciousness.” The Buddhist stipa, a cosmological mapping centered

on a Cosmic Man of another sort, the Buddha, is made up of parts
whose names are borrowed from the terminology of the Vedic sacri-
fice.1° So, too, when Hindu temples appear in the middle of the first
millennium c.£., dedicated to gods propelled to glory by the bhakti
movement, they, too, patterned themselves on the plan of the Vedic
sacrificial arena.!! The Vedic sacrifice, as the Indian ur-model of the
cosmos, thus is superimposed on post-Vedic worship centers and micro-
cosms of both the Buddhist and Hindu varieties.
But it is at the level of adhyatman, making sense of the human self
and human life, that the Vedic sacrifice most obviously and continually
reappears throughout the history of Indian religions. The projection of
the sacrifice onto the human purusa and his or her existence, the mean-
ingful overlay of the concept of sacrifice onto human existence, also
begins in the Veda itself. The human body is understood as a reduplica-
tion of the sacrificial arena, for at the microcosmic level, too, puruso vai
yajnah: ,
Man, Gautama, is a fire. Speech is his fuel, breath his smoke, the tongue
his flame, the eye his coals, the ear his sparks.!*
Or again:
The sacrificial site (vedi) is indeed his breast, the sacrificial grass is his
hair. The garhapatya fire is his heart, the daksina fire is his mind, the
ahavaniya fire is his mouth.¥
The generation of human bodies depicted in sacrificial terms under
this paradigm is envisioned as a process of sacrificial oblations of male
seed into the “fire” that is woman. Into the “fire” that is man, the gods
make an “offering of food,” from which semen is produced. The semen,
in turn, is poured into the woman to produce a new life. For not only is
man depicted as a sacrifice, but “Woman, Gautama, is a sacrificial fire.
The vagina is her fuel, foreplay her smoke, the womb her flame, the
penetration her coals, and pleasure her sparks. Into this fire the gods
10. Consult L. A. Govinda, Psycho-Cosmic Symbolism of the Buddhist Stipa (Em-

eryville, Calif.: Dharma Publishing, 1976).
11. See L. A. Ravi Varma, “Rituals of Worship,” in Haridas Bhattacharya, ed., The
Cultural Heritage of India, 4 vols. (Calcutta: Ramakrishna Mission Institute of Culture,
1956), IV:448; and esp. Stella Kramrisch, The Hindu Temple, 2 vols. (reprint ed., Delhi:
Motilal Banarsidass, 1976), I: 1-17, 67-97, 139-44.
12°-Chwis.7-1.
13. ChU5:18.2,
make an offering of semen, and from that oblation the embryo is

born.”4
The body’s dissolution at death is likewise regarded in sacrificial
terms in Vedic texts, a final and complete offering of the corpse into the
cremation fire. Replicating the cosmogonic sacrificial dispersion of the
body of Purusa/Prajapati, the human body, too, is sacrificially disinte-
grated and reintegrated into the cosmos:
The voice of the deceased enters the fire, his breath [enters] the wind, his
eye the sun, his mind the moon, his ear the cardinal directions, his flesh
the earth, his dtman the atmosphere, his body hair the plants, the hair on
his head the trees, and his blood and semen are deposited in the waters.!°
As Bruce Lincoln writes by way of summarizing this and other related
Indo-European visions of death as sacrifice, “Like all sacrifice, death is
a repetitive, ritual act. Each death repeats every other death and every
other sacrifice: above all the first death, which was also the first sacri-
hee.
When the new doctrine of rebirth first appears in the early Upani-
shads, it is not surprisingly cast in the mold that is the Vedic sacrifice. The
earliest formulation of transmigration and the reincarnation of dtman
explains the process of rebirth as a series of oblations in five fires, culmi-
nating in the oblation of male seed into the womb fire of the female:
Thus in the fifth oblation, water comes to have a human voice. When the
embryo has lain inside there for ten months or nine months, or however
long, covered with the membrane, then he is born. When he is born, he
lives as long as his allotted life-span. When he has died, they carry him to
the appointed place and put him in the fire, for that is where he came
from, what he was born from. .. . Those who worship in the village,
concentrating on sacrifices and good works and charity, they are born into
the smoke, and from the smoke into the night, and from the night into the
other fortnight, and from the other fortnight into the six months when the
sun moves south. They do not reach the year. From these months they go
to the world of the fathers, and from the world of the fathers to space, and
from space to the moon. That is king Soma. That is the food of the gods.
The gods eat that. When they have dwelt there for as long as there is a
remnant [of their merit], then they return along that very same road that
they came along, back into space; but from space they go to wind, and
when one has become wind he becomes smoke, and when he has become
14. ChU 5.8.1-2.

15. BAU 3.2.13.
16. Bruce Lincoln, Myth, Cosmos, and Society, p. 127.
smoke he becomes mist; when he has become mist, he becomes a cloud,

and when he has become a cloud, he rains. These are then born here as
rice, barley, plants, trees, sesame plants, and beans. It is difficult to move
forth out of this condition; but if someone eats him as food and then emits
him as semen, he becomes that creature’s semen and is born.!7
And just as both the origins and the ongoing functions of the cos-
mos are represented in sacrificial tones in the Veda, so is the ongoing
process of human life—in addition to the origins and ends of that life—
depicted as a yajna. According to the Chandogya Upanisad, human
existence is to be understood as a kind of soma sacrifice, with its three
periods of youth, maturity, and old age corresponding to the three press-
ings in the morning, midday, and evening.'8
The sacrificial paradigm is also applied in discussions of the critical
moments of that sacrifice which is the life of humans. The “second birth”
Or initiation into Vedic society (upanayana), which was briefly analyzed
in Chapter 4, is one of the few rituals of Vedism that persists more or
less unchanged throughout the history of Vedism—and has always been
represented as a sacrifice. As Jan Gonda has demonstrated, the upana-
yana is both conceived as and modeled on the initiation and consecra-
tion of the sacrificer, the diksd, and initiation into the cult throughout
the history of Indian religions, orthodox and heterodox alike, preserves
and reduplicates this sacrificial origin.!9
Upanayana also inaugurates the study of the Veda (brahmacarya)
which is likewise understood as a kind of sacrifice. From the Satapatha
Brahmana we read that “He who enters upon brahmacarya enters into a
long sacrificial session (dirgha-sattra). The stick of fuel he places on the
fire is the opening offering; and that [which he places on the fire] when
he is about to bathe is the concluding offering; and what [sticks of fuel]
there are [placed on the fire] between these, are his [offerings] of the
sacrificial session.”2° A comparable passage has brahmacarya, through
etymological machinations, as mysteriously a sacrifice:
What people call “sacrifice” (yajfa) is really brahmacarya, for only
through brahmacarya does he who is knower (yo jnata) find that [knowl-
17. ChU 5.3.1-10, trans. in Textual Sources for the Study of Hinduism, ed. and
trans. by Wendy O’Flaherty with Daniel Gold, David Haberman, and David Shulman
(Manchester: Manchester University Press, 1989).
18. ChU 3.16-17.
19. Jan Gonda, Change and Continuity, pp. 315-462. See also Brian K. Smith,
“Ritual, Knowledge, and Being,” pp. 79-80.
20. SB 11.3.3.2.
edge]. And what people call a “sacrificial oblation” (ista) is really

brahmacarya, for only after having searched (istva) through brahmacarya
does one find the atman. And what people call the “protracted sacrificial
session” (sattrayana) is really brahmacarya, for only through brahmacarya
does one find the protection (trana) of the real (sat) atman.*!
The various individual components of the life of the student are
similarly comprehended in the language of the sacrifice. Alms which are
begged by the pupil and brought to the teacher are regarded as “obla-
tions” (havis) placed into the ahavaniya fire which is the mouth of the
preceptor,” and the portion of the alms consumed by the student is called
“the remnants of the sacrificial oblations” (havirucchistha).* Already in
the Satapatha Brahmana, the daily recitation of the Veda is regarded as a
“sacrifice to the brahman” (brahma-yajna), one of the five “great sacri-
fices” (mahdyajnas) or “great sacrificial sessions” (mahdasattras)*4 that in
the later Dharma Sastras become the centerpiece of the daily religious life
of post-Vedic Brahmanism. The svadhydaya or personal recitation of the
Veda is equated with the actual performance of the Vedic ritual:
Now for the sacrifice to the brahman. The sacrifice to the brahman is one’s
daily recitation of the Veda. Speech is the juhi-spoon of this sacrifice;
mind is the upabhrt. The eye is the dhruva, intelligence the sruva. Truth is
its concluding bath, the heavenly world its concluding rites.>
The conclusion of the period of Veda study is, as expected, homologized

with the conclusion of the Vedic sacrifice: the gift to the teacher pre-
sented by the departing student is specifically termed a daksind or sacrifi-
cial fee,?° and the ceremonial bath at the end of brahmacarya, the
samavartana, recalls the concluding bath of the sacrifice (the avabhrtha).
The notion that the accumulation of religious knowledge is a kind
of sacrifice together with the conceptualization of human physiology in
sacrificial terms combine to make possible the representation of world
renunciation (in the Upanishads and other texts) as yet another version
of the yajna. But the very act and subsequent life of renunciation—
which includes renunciation of sacrifice, of ritual action or karman—is
21. ChU 8.5.1-2.

Pas ApDhsS 1.1.3.43—44; see also Manu 2.231. For teacher as sacrificial fire, see
ApDhS 1.1.3.42; VasDhS 7.5—6; GautDhS 3.6-8.
23. ApDhS 1.1.4.2-4.
24. SB 11.5.6.
25. SB 11.5.6.3; see also $B 11.5.6.4-9; BDhS 2.6.11.7; ApDhS 1.4.12.3; ASvGS
113
26. ApDhS 1.2.7.19; JGS 1.12.
; The Destiny of Vedism 209
presented in the texts as a “higher” or “truer” sacrifice. The renouncer,

or “sacrificer,” “sacrifices only in himself (atmanneva yajati),”2” and
sucii a sacrifice is envisioned by the monistic Upanishadic thinkers as a
means to obtain mystical union in a pantheistic conception of that
oneness: “His offering is made in all worlds, in all beings, in all
selves.”28
This identification of the life and physiology of the renouncer—
who has resumed within himself the sacrificial fires which are then ex-
tinguished”—and the ritual of sacrifice creates homologies between all
the functions of life and the functions of ritual. The consumption of
daily meals becomes an dtmayajra or “sacrifice to and in the self.”3
Even the breathing of the renouncer is considered to be an “inner
sacrifice” (antara-agnihotra) or “sacrifice to and in the life-breaths”
(pranda-agnihotra):3!
After the brahmadhana [the ritual in which the sacrificial fires are “inte-
riorized” and life as a world renouncer is begun] the sacrificer himself [has
within him] the sacrificial fires. The inhalation (prana) is the garhapatya
fire, the exhalation (apana) is the anvaharyapacana [the daksina] fire, the
circulation in the body (vydna) is the aGhavaniya fire, and the cerebral
circulation (samdana) is the sabhya and avasathya fires. These five fires are
in the self (Gtman). He offers in the self alone.*2
In this way, every moment of the life of the world renouncer is regarded
as sacrifice, every breath an “oblation,” and life as a whole is considered
a “continuous” and “uninterrupted” yajna:
As long as a person is speaking, he is not able to breathe. Then he is
sacrificing breath in speech. As long as a person is breathing, he is not able
to speak. Then he is sacrificing speech in breath. These two are unending,
immortal oblations; whether waking or sleeping one is sacrificing continu-
ously, uninterruptedly.*°
27. Maitri Upanishad 6.9.

28. ChU 5.24.2.
29. For the rite of the “interiorization” of the sacrificial fires, see, e.g., BDhS
2.10.17.19.
30. For the dtmayajfia, see Ananda Coomaraswamy,“Atmayajna”; and H. W.
Bodewitz, Jaiminiya Brahmana I, 1-65 (Leiden: E. J. Brill, 1973), pp. 213-18.
31. For the prandagnihotra, see Bodewitz, Jaiminiya Brahmana; J. Varenne, ed. and
trans., La Mahdnarayana Upanisad (Paris: E. de Boccard, 1960), pp. 69ff.; and the follow-
ing texts: ChU 5.19-24; BAU 1.5.23; AitA 3.2.6; SanA 10; Kausitaki Upanishad 2.5;
Maitri Upanishad 6.3; VaikhGS 2.18; BDhS 2.10.18.
32. BDhS 2.10.18.8.
33. Kausitaki Upanishad 2.5.
The taking of sannydsa is, as Louis Dumont,*, Patrick Olivelle,*

and others have shown, a renunciation of the yajva itself; and, as
claimed in Chapter 7, the Upanishadic monistic philosophy is in dra-
matic opposition to the principles of Vedic resemblance. But in most
texts world renunciation is not presented as a renunciation of the con-
cept of sacrifice; the category has simply been appropriated in order to
extol the virtues of a very different set of beliefs and practices. The
atma-yajnin, the “sacrificer to the self,” is indeed said to be superior to
the deva-yajnin, “the sacrificer to the gods”**—but although his “sacri-
fice” may be better, it is still presented as a sacrifice.
In the Upanishads the yajfa was transformed and redefined—but
nevertheless preserved—as new religious practices—meditation, yoga,
and contemplation on the ultimate nature of the sel{—were equated to it.
Some scholars, seen in Chapter 7, argue that world renunciation repre-
sents an “orthogenetic, internal development of Vedic thought.”%” In the
Upanishads there was, however, “a redefining of the place of sacrifice
within the total religious scheme.”** Or, as Heesterman rightly observes,
“The question that occupies religious thought appears not to be: sacrifice
or rejection of sacrifice, but rather what is the true sacrifice.”%9
The Upanishadic redefinition of the “true sacrifice” might be best
seen not as the logical outcome of Vedic ritual thinking but rather as a
valuable objet trouvé useful to assimilate the foreign to the traditional.
Eliade has argued that the equating of practices such as tapas (ascetic
austerity) and yoga—both of which he regards as extra-Aryan in
origin—to the Vedic sacrifice allows for flexibility and change within the
Brahmanic tradition:
For tapas, too is a “sacrifice.” If, in Vedic sacrifice, the gods are offered
soma, melted butter, and the sacred fire, in the practice of asceticism they
are offered an “inner sacrifice,” in which physiological functions take the
place of libations and ritual objects. . . . The concept of this “inner sacri-
34. Louis Dumont, “World Renunciation in Indian Religions,” Contributions to

Indian Sociology 4 (1960): 33-62; reprinted in Religion/Politics and History in India:
Collected Papers in Indian Sociology (The Hague: Mouton, 1970), pp. 33-60.
35. Patrick Olivelle, “A Definition of World Renunciation,” Wiener Zeitschrift fiir
die Kunde Siidasiens 19 (1975): 75-83; idem, “Odes of Renunciation,” Wiener Zeitschrift
fiir die Kunde Stidasiens 20 (1976): 91-100.
36. Already in SB 11.2.6.13-14; see also BAU 1.4.10.
37. Heesterman, “Brahmin, Ritual and Renouncer,” p. 40.
38. Thomas B. Goman and Ronald S. Laura, “A Logical Treatment of Some
Upanisadic Puzzles and Changing Conceptions of Sacrifice,” Numen 19 (April 1972): 62.
39. Heesterman, “Brahmin, Ritual and Renouncer,” p. 42.
ce The Destiny of Vedism 211
fice” is a fertile one, which will permit even the most autonomous ascetics
and mystics to remain within the fold of Brahmanism and later of Hindu-
ism. ... The practical consequence of homologization is substitution
(which it justifies). Thus asceticism becomes equivalent to ritual, to Vedic
sacrifice. Hence it is easy to understand how other yogic practices made
their way into the Brahmanic tradition and were accepted by it.“
Regardless of whether the representation of the practices of world

renouncers in sacrificial terms is interpreted by its scholars as the organic
development of Vedic ritualism or as a strategy for incorporating within
the orthodox tradition practices of extraneous origin, the function is the
same: to conceptualize and articulate the new in terms of the old. The
category of yajna thus serves to explain the unknown (its explicative
function as a category); and, we might safely say, it also acts to tradi-
tionalize the innovative.
Sacrifice was, indeed, such a valuable category in the discourse of
ancient India that even the ostensibly anti-Vedic heterodox traditions
seemed compelled to argue that their religious practices, too, were sacri-
ficial. By doing so, these anti-Vedic movements could simultaneously
represent themselves as connected to the ancient Vedic practices and as
superseding them. In the early Jaina and Buddhist Sutras (which also
sometimes simply dismiss the Vedic sacrifice as worthless and heinous),
there are texts in which the issue is, as Heesterman has pointed out,
What is the true sacrifice? The edifying Jaina story of Harikesa,*! an
outcaste (canddla) and a Jaina monk, has him approach a Vedic sacrifi-
cial site. The monk scrupulously reproaches the Brahmin priests, who
are depicted as “stuck up by pride of birth, those killers of animals, who
did not subdue their senses, the unchaste sinners.” Harikesa informs
them that they are incorrectly performing the sacrifice: “You [use] kusa-
grass, sacrificial poles, straw and wood; you touch water in the evening
and in the morning; thereby you injure living beings, and in your igno-
rance you commit sins again and again.” The Vedic ritualists, portrayed
here as being remarkably quick to admit the error of their ways, inquire
of the Jain, “How should we sacrifice, o monk, and how avoid sinful
actions?”
The answer, of course, is to take up the practices of the Jain monk:
noninjury to living beings and renunciation of property, women, pride,
40. Mircea Eliade, Yoga: Immortality and Freedom, 2d ed., trans. by Willard R.
Trask (Princeton, N.J.: Princeton University Press, 1969), pp. 111-12, 113.
41. H. Jacobi, trans., Jaina Sutras, 2 vols. (Oxford: Clarendon Press, 1884—85;
reprint ed. New York: Dover, 1968), II: 50-56.
PDs REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
and so on. The Brahmins then pose another question and receive a reply
in which the “true” sacrifice is disclosed:
“Where is your fire, your fireplace, your sacrificial ladle? Where is the
dried cowdung [used as fuel]? Without these things, what kind of priests
can the monks be? What oblations do you offer to the fire?” “Penance is
my fire; life my fireplace; right exertion is my sacrificial ladle; the body the
dried cowdung; karman is my fuel; self-control, right exertion, and tran-
quility are the oblations, praised by the sages, which I offer.”
A similar tale is told in an early Buddhist text, the Kutadanta Su-

tra,‘2 in which the Buddha is cast as a Vedic priest attached to a king (a
purohita) in one of his previous lives. The Buddhist-Vedic priest in-
structs his patron in the proper mode of sacrifice, performed with appro-
priately Buddhist nonviolence (ahimsa):
At that sacrifice neither were any oxen slain, neither goats, nor fowls, nor
fatted pigs, nor were any kinds of living creatures put to death. No trees
were cut down to be used as posts, no Dabha grasses mown to strew
around the sacrificial spot. . . . With ghee, and oil, and butter, and milk,
and honey, and sugar only was that sacrifice accomplished.
What follows in the text is a hierarchy of “sacrifices” even better than

this bloodless one, each one “less troublesome, of greater fruit and of
greater advantage” than the last. At the very top of the hierarchy, the
highest “sacrifices” are the (very nonviolent) meditational practices of
the Buddhist monk.
In these texts, the heterodox traditions unveil their new doctrines
and religious activities in the guise of “higher” or “truer” forms of the
Vedic ritual. The latter is made to appear as relatively ineffectual in
comparison with newly revealed “sacrifices.” But what is important for
our purposes is the fact that even the heterodox religions found it expedi-
ent to call up the category of sacrifice as a hermeneutical device: new
doctrines and practices were made understandable by being encoded in
the already ancient and well-known vocabulary of sacrifice; and they
were made to appear old but lost up until now.
By the time of the Bhagavad Gita (ca. second century B.c.£. or
later), a text that manifests a syncretistic and emerging Hindu ortho-
doxy, any number of traditional as well as heterogeneous and non-Vedic
42. Found in T. W. Rhys Davids, trans., Dialogues of the Buddha, 3 vols. (London:
Pali Text Society, 1899; reprint ed. London: Routledge and Kegan Paul, 1973-77), I: 160-
85.
practices are synthesized into Hindu religious life by identifying them

with yajna:
Some yogins offer only sacrifice to the gods; others offer the sacrifice with
the sacrifice in the fire of brahman. Others offer the senses beginning with
hearing in the fires of self-restraint; and others offer the objects of sense,
beginning with sounds, in the fires of the senses. Others [offer] all acts of
the senses and acts of the breaths in the fire, kindled by wisdom, of the
yoga of self-restraint. Others [offer] material sacrifice, the sacrifice of
austerities, or the sacrifice of yoga; and [others offer] sacrifices of Veda-
recitation (svadhdaya) and mystical knowledge (jridna)—world renouncers
whose vows are keen. Some offer the inhalation of breath in the exhala-
tion of breath, and the exhalation of breath in the inhalation of breath,
controlling them both, intent on breath-control. Others restrict their
meals and offer the breaths in the breaths. All these are knowers of the
sacrifice, their defilements erased by sacrifice.
And with the Gita we also witness a new extension of the category
of sacrifice. Isolating and generalizing the component of the ritual called
the tyaga (the “giving up” or “renunciation” of the offering), the Gita
teaches the doctrine of karma-yoga as sacrificially oriented everyday
action whereby “all attachment and all the fruits [of action] are re-
nounced.” For, the text says, “this world is bound by the bonds of
karma except where that action is done sacrificially.”4
In other texts more or less contemporaneous with the Gita (as well
as in the Gita itself), this conception of sacrifice is also used to redeem
the function of kings and warriors whose professions otherwise contra-
vene the newly emergent Hindu doctrine of ahimsa:* “A king who
protects [all] beings through dharma, and who executes those sentenced,
daily sacrifices with sacrifices having thousands of sacrificial fees.”4” The
Mahabharata as a whole, as Alf Hiltebeitel among others has shown,
43. Bhagavad Gita 4.25-30.

44. Bhagavad Gita 18.6.
45. Bhagavad Gita 3.9.
46. A comparable example of this phenomenon—applying the category of “sacri-
fice” to a practice that might otherwise be judged abhorrent and in opposition to religious
principles—is the custom of widow-burning or sati. See, e.g., Siva Purana 2.2.7.4-6,
where Sandhya enters the sacrificial fire as an act of sati: “Her body itself became the
purodasa [the sacrificial cake] in that sacrifice.” My thanks to Paul Courtright for this
reference.
47. Manu 8.306; see also Manu 5.93, where a king is compared with those engaged
in an extended sacrificial session.
presents war as a “sacrifice of battle”:48 “He obtains the world of heroes

who offers up in battle his own body as an oblation.”*? Sacrifice, in cases
such as these, is called upon to reconcile conflicting Hindu doctrines
(ahimsa and the necessity for kings and warriors to fulfill their duties)
through reformulating the new problem as an old answer.
In yet another move that will have great consequences for the subse-
quent development of devotionalistic Hinduism, the Gita introduces the
new teaching of bhakti in the guise of the reinterpreted “true sacrifice”:
A leaf, or a flower, or a cup of water that a pious soul may offer Me with
bhakti, that do I accept, for it was bhakti that made the offering. Whatever
you do, whatever you eat, whatever you offer or give away, whatever
austerity you may perform, consign it to Me. You will be liberated from
those bonds which karma creates, whose fruits are good and bad. Your
atman yoked by renunciation and yoga, fully liberated, you will come to
Me.
In later texts of Hindu sectarian and devotionalistic traditions, sacrifice
continues to be a leitmotif long after it has ceased to be performed by
anyone other than the occasional king (seeking Vedic legitimacy) or
Brahmins stuck in the backwaters of South Asia. The ritual of puja to
the image of Visnu, Siva, the Goddess, or one of their avatars simulta-
neously replaces the yajra while it incorporates within itself many rites
from the Vedic sacrifice. The smarta Hindu puja ritual structure encom-
passes within it Vedic rites of the fire sacrifice (mostly at the grhya level)
but makes them subservient to more recent principal rites of service to
the image. Vedic ritualism, recapitulated in its most compact form now
functions as merely a subsidiary rite within a ritual oriented around a
different center and purpose.*!
Already in the Mahabharata, Krsna is said “to be” the sacrifice,
just as Visnu was even in Vedic texts. The image of Visnu, in a Hindu
Pancaratra Vaisnava text, is equated with the Vedic sacrifice—the parts
48. Ranayajna, e.g., Mbh 5.57.12; 5.154.4 See Alf Hiltebeitel, The Ritual of Battle:
Krishna in the Mahabharata (Ithaca, N.Y.: Cornell University Press, 1976). For one of the
books of the Mahabharata as “a coherent whole structured from beginning to end by the
entire Rajastya,” see pp. 95-101; and J. A. B. van Buitenen, “On the Structure of the
Sabhaparvan of the Mahabharata,” in J. Ensink and P. Gaeffke, eds., India Maior: Con-
gratulatory Volume Presented to J. Gonda (Leiden: E. J. Brill, 1972), pp. 68-84.
49. Mbh 18.1.14.
50. Bhagavad Gita 9.26-28.
51. See T. Goudriaan, “Vaikhanasa Daily Worship,” Indo-Iranian Journal 12
(1970): 161-215.
52. E.g., Mbh 3.13.44, 14.70.20.
of the body identified with the components of the ritual—and the wor-
ship of the image (puja) is said to procure for the devotee “the fruits of a
hundred sacrifices.”53 In a Saivite text, the Sivadharma, we read that
“the ritual greeting [to the image of Siva, namaskara] is superior to all
sacrifices, and having made the ritual greeting to Mahadeva he obtains
the fruit of a horse sacrifice.”>4
Certain Hindu texts do indeed “have as their sole object,” as wrote
A. Barth, “to teach a sort of cult at a discount, procuring the same fruits
as the great sacrifices.”>> Or, as Kamaleswar Bhattacharya observes,
“the rewards that the Vedic ritualists aimed for can now be obtained at a
smaller expense.”*® In these cases, the Vedic sacrifice provides later
Hinduism with a kind of standard of worth by which non-Vedic religious
practices might be gauged. In the Mahabharata,°’ to cite another exam-
ple of this frequently encountered phenomenon, the rewards of visiting
various Hindu pilgrimage spots are couched in terms of the old sacrificial
system. The mere appearance at one or another tirtha wins for the
pilgrim “the fruit of a soma sacrifice” or “the fruit of a horse sacrifice.”
Results which once entailed great expense and painstaking labor, and
which were in all likelihood available only to the religious, political, and
economic élite, are now easily obtainable by all thanks to the full equa-
tion of the later version of the original. In this way, the new and rela-
tively simple religious practices of Hindu worship are said to resume in
themselves the power of the most complex Vedic sacrifices, just as cer-
tain domestic rites came to contain the essence of the mahdyajnas. And
identification of the simple and the complex puts the former in a better
light: why bother to undertake massive and obsolete sacrificial endeav-
ors when their “fruits” can be efficiently (and economically) obtained
through new and improved techniques?
But the purpose of shrouding new Hindu practices in sacrificial
clothing is not simply to prove the superiority of the new to the old, but
first and foremost to present the new as the old. Sacrifice has functioned
53. Ahirbudhnya Samhita, chap. 37, cited in Kamaleswar Bhattacharya, “Le

‘Védisme’ de certains texts hindouistes,” Journal asiatique 225 (1967): 200-201.
54. Cited in ibid., p. 211. Compare a much earlier text, the ASvGS (1.1.5), in which,
after a discourse in which sraddha (“faith”) is extolled as the true sacrifice, the passage
ends by quoting an (unspecified) Brahmana by way of equating sacrifice and namaskara
(yajno vai namah iti hi brahmana).
55. A. Barth, Les Religions de I’Inde: Extrait de l’Encyclopédie des sciences
religieuses (Paris: G. Fischbacher, 1879), p. 60.
56. Bhattacharya, “Le ‘Védisme’ de certains texts hindouistes,’ p. 211.
57. Mbh 3.80-83.
throughout Indian religious history as a marker for traditionalism and as

a means for acceptable innovation. The very radical movement of
Tantricism, too, does not neglect to codify its practices in orthodox
sacrificial terminology. The partaking of the five “forbidden” (to the
orthodox, that is) substances is represented in Tantric texts as oblations
to various deities located within the body of the practitioner.*
Sacrifice thus serves as a stake for traditionalism and as a means for
acceptable innovation within the boundaries of orthodoxy, and the strat-
egy of reforming by appeal to the category of yajna is still at work in the
modern period in India. Tagore preached his secularized humanism in a
dramatic vehicle entitled “Sacrifice,”*? and Gandhi’s redefinition of
yajna in light of his own philosophy summarizes a process that has been
at work throughout Indian religious history—the search to find a
“deeper” and new meaning in the archaic Vedic concept:
[Y ]ajria has a deeper meaning than the offering of ghee and other things in
the sacrificial fire. Yajna is sacrifice of one’s all for the good of humanity,
and to me these offerings of ahutis have a symbolic meaning. We have to
offer up our weaknesses, our passions, our narrowness into the purifying
fire, so that we may be cleansed.
Yajna as a Canonical Category
Enough has been said here to indicate that the Vedic sacrifice is in-
deed a candidate for inclusion in the repertoire of entities constitutive
of the Indic cultural and religious continuity. In all times and in the
most diverse traditions comprising the Indic religious configuration, the
concept of sacrifice has been an old skin into which new wine is
poured.
But the deployment of the category of yajfa is also a means to make
Hinduism Vedic; and to make something Vedic in Hinduism, as argued
above in Chapter 1, is to make it canonical and orthodox. It is in this
sense that the Vedic sacrifice in Hinduism might be regarded as one
mode of defining Hindus; it is because of the integral connection be-
58. See Agehananda Bharati, The Tantric Tradition (New York: Samuel Weiser,
1975), pp. 260-62.
59. Rabindranath Tagore, “Sacrifice,” in A Tagore Reader, ed. by Amiya
Chakravarty (Boston: Beacon Press, 1966), pp. 125-48.
60. Mohandas Gandhi, Jn Search of the Supreme, comp. and ed. by V. B. Kher
(Ahmedabad: Navajivan Publishing House, 1961), p. 276.
The Destiny of Vedism mle
tween yajna and Veda that sacrifice might be called a canonical category
in Hinduism.
Hindus, also as already seen in Chapter 1, appear to use only the
“outside” of the Veda, oblivious to what is “inside.” And what is the
subject matter found inside the covers of the Veda? The mantras, rules
(vidhis), and explanations (arthavadas) of the Vedic sacrifice. The para-
dox of Hindu orthodoxy defined as adherence to the canonical authority
of the Veda which is not understood reduplicates the paradox of Hindu
practices presented in the guise of the Vedic sacrifice which is, by and
large, no longer performed.
The Hindu use of the Veda to claim orthodox authority for post-
Vedic texts and the Hindu representation of post-Vedic practices as
yajnas thus appear as two aspects of the same phenomenon. Sacrifice is
perhaps the only category in Hindu discourse that reduplicates the possi-
bilities of the category of Veda.*! When Hindus represent post-Vedic
practices as “sacrifices,” or equivalent to sacrifices, or as new forms of
sacrifice, they appear to be doing the same thing—and for the same
ends—as representing post-Vedic texts in various ways as “Veda.” Veda
as canon and sacrifice as a canonical category are touchstones to ortho-
doxy in the Hindu religion.
In both cases, however, the paradox that concerns scholars of Hin-
duism arises only when the Veda is regarded as a set of ancient texts and
the yajna as a set of ancient ritual acts. If, on the other hand, we envision
both Veda and yajna as categories for creating orthodoxy, the Hindu
preoccupation with them ceases to be so mysterious. Hinduism might
then be considered a process of innovation through traditionalization
and canonization, and with perhaps more than one “unchanging” point
of reference by which one returns in order to progress within the bound-
aries of Hinduism.
Both the Veda and the sacrifice have been represented by Indolo-
gists as irrelevant and misunderstood by Hindus—the ancient scripture
because it is unread and the ancient ritual because it is unperformed and
“meaningless.” One conclusion we might draw from this study is that
Hinduism could very well have inherited more from the ancient Vedic
religion than just the outside of an unread Veda and an empty shell of
the Vedic sacrifice. What I would like to suggest as we terminate this
projection into the ideology and practice of Vedic ritualism is what I
suggested in the opening chapter: what we have termed the “strategies
61. “Brahmin” is possibly the only other category with the same persistence and
centrality for the tradition. See J. C. Heesterman, “Brahmin, Ritual and Renouncer”; and
esp. Jan Gonda, Notes on Brahman (Utrecht: J. L. Beyers, 1950).
of orthodoxy” in post-Vedic Hinduism are indeed Vedic in origin. Put

succinctly, the necessary obsession to constitute as “derivative” Hindu
texts, practices, and institutions, and to connect them to Vedic originals
(this text is really the Veda, this practice is really a sacrifice, etc.) is itself
a Vedic mode of thought.
Inside the pages of the Veda, in the interior of the ideology and
practice of Vedic ritualism, one finds—in never-ending abundance—con-
nections, or bandhus, linking components of various planes of reality to
one another. The Vedic sacrifice is the workship for forging such ban-
dhus, and one principal type of connection links manifest counterparts to
their transcendent prototypes. The Vedic ritual device of connecting, for
example, the three fires with the three worlds is replicated in Hindu
strategies for orthodoxy—such as connecting a well-known Purana and an
unread Veda, or an oft-performed puja and a neglected fire sacrifice.
The metaphysics of the Veda are set forth in terms of forms (ripas)
and counterforms (pratirtipas), prototypes (pramdas) and counterparts
(pratimas), and the organization of the ritual structure itself reduplicates
this metaphysics. In the paribhasd siitras, the ritual system is depicted as
a set of paradigmatic models (prakrtis) and derivative variants (vikrtis).
And the epistemological program of Vedic thought, we recall, is nicely
summed up at RV 10.130.3: “What was the prototype, what was the
counterpart, and what was the connection between them?”
This Vedic quest for linkages between archetypes and resembling
manifestations of the archetypes seems to hold the key to understanding
the strategies for orthodoxy observed in Hinduism. The Veda and the
sacrifice function as categorical prototypes for all subsequent texts, prac-
tices, and institutions. These latter are represented as counterparts
through the forging of connections—bandhus or nidanas—back to the
Veda, back to the yajna, back to Vedic ritualism. The significance of the
Veda and the Vedic sacrifice for subsequent Hindu thought and praxis
might in this manner be traced to the Veda itself. Hindus think about the
Veda in ways that seem reduplicative of Vedic thought; and Hindus use
the canonical category of sacrifice in ways that might be interpreted as
“sacrificial” in the original Vedic sense: as a connective, constructive
activity. The principles and philosophy of Vedic ritualism, which I have
tried to represent in this book, seem to have survived Vedism and be-
come the instruments for creating and transforming orthodoxy in Hindu-
ism. Vedic connections are, in sum, the very means by which Hinduism
is connected to Vedism.”
62. In his conclusion to The Ritual of Battle, Hiltebeitel makes much the same
argument: “The epic poets would thus emerge not so much as programmers, transposing
Conclusion
“Analogies,” wrote Sigmund Freud, “it is true, decide nothing, but they
can make one feel more at home.”s Domesticating the chaos of the
unknown by categorizing it, classifying it, and making it conform alto-
gether to the familiar—this is, I think, an adequate description of the
work of religion, and of the work of those who study it. The method I
have called, in the Vedic context, resemblance (and what Freud and
others call analogy, or homology, or a system of ideal connections)
appears to be a very effective way by which both the religieux and their
observers make themselves at home.
Religion is an exercise of the creative power of the imagination.
Religion is one of the most persuasive attempts, with its claims to abso-
lute and eternal truth, to empower illusions with living reality; it is one
of the ways in which humans have made meaningful a world in which
meaning always must be imposed and overlaid. Human beings are gener-
ally encountered here taking responsibility for the entire cosmos about
them, just as the ancient Vedic Brahmins did within their ritual. They
are making sense of it—and this is what might be called the art of
transformation, the “making and finding” epistemological step common
to all who desire to live in a comprehensible world.
In Vedic ritualism, to a degree not always seen, the guidelines for
constructing categorical connections (analogies, homologies, etc.) were
explicitly formulated and, in the ritual itself, set into motion and acti-
vated. Here the art of transforming the unknown into an instance of the
already known, transforming the unique and disconnected into a mem-
ber of a class, followed the articulated rules of the Vedic brand of
hierarchical resemblance.
Locating all reality between the poles of unitary identity and wholly
differentiated individuality, the ritualists posited a series of models
which made possible knowable particulars: prototypes accumulated
their resembling counterparts; counterforms of forms were made and
found; paradigms organized the elements of the whole into categorized
one set of information into another form, but as rsis, in this case the rsis of the ‘Fifth Veda’
whose school is covered by the name of the elusive but ever-available rsi Vyasa. By calling
attention to this term for visionaries and poets, I refer in particular to the rsis faculty of
‘seeing connections,’ ‘equivalences,’ ‘homologies,’ and ‘correspondences.’ This faculty of
‘seeing connections’ would have involved the epic poet, not only with correlations between
myth and epic, but also between epic and ritual—especially that of the Brahmanic sacri-
fice. Thus the ‘mythic exegesis’ must coexist with a ‘ritual exegesis’ ” (pp. 359-60).
63. Sigmund Freud, New Introductory Lectures on Psychoanalysis, trans. and ed. by
James Strachey (New York: W. W. Norton, 1965), p. 72.
variants. Resemblance between a set of particulars and the archetypical,

and the hierarchical differentiation of the particulars into more and less
complete realizations of the archetype, provided an epistemological tool
that could be reapplied to any number of situations. The universe at
large, this world and its inhabitants, the other worlds and their denizens,
representations of ritual practice, and the ritual practice itself and those
who participated in it—all were systematically organized with this trans-
formative methodology.
Resemblance made possible a system of connections which, as we
have seen, was never confused within Vedic ritualism with a system of
identities. Connections were forged horizontally between analogues in
different hierarchical registers (between the various social classes and
particular animals or food, for example); and they were made between
higher and lower elements within the same hierarchical order (between
prototypes and counterparts and between the counterparts themselves).
In this latter type of connection, the vertically directed bondings
between superior and inferior forms, we have also observed two differ-
ent types of strategies, both utilizing synecdoche whereby a part stands
for the whole. First, there was what I have called an “upward condensa-
tion,” whereby the lower is encompassed within the higher (e.g., the
essences of the lesser sacrificial victims or rituals are subsumed within
the hierarchically superior forms). But, secondly, there was also the
possibility of making vertical connections between the superior and the
inferior in such a way that the essence of the higher is reprised within the
lower: a “downward condensation” or synecdochic reduction that could
easily be, and has been in the past, confused with true equivalency and
full replacement. Apparent homologies between the higher and the
lower were not meant, within Vedism, to overturn the hierarchical order
of things but rather to strengthen the relations among the elements
within the hierarchical order. As stated above, “equations” in this sys-
tem were not intended to collapse distinctions but rather to strengthen
connections among interrelated elements on a vertically oriented, and
hierarchically calibrated, scale.
Nor were these “mere ideas,” if there are such beasts. The over-
weening preoccupation of the ancient Indians with ritual led to belief
that these ideas could be and were put into effect. The Vedic ritualists
attempted to realize their epistemological constructs by ritually con-
structing the universe, heavens, this world, and society; a metaphysics
was produced through the ritual activation of an epistemology. Human
beings were likewise ritually constituted, in various forms and hierarchi-
cally gauged realizations of form, both in this life and in the projected
afterlife. An ontology was thus also a product of the ritual activation of

the epistemology of hierarchical resemblance.
But Vedic religion, far from being a mysterium tremendum created
by different minds a long time ago and far away, appears (and perhaps
only appears, perhaps must appear, and perhaps should appear) to us to
be a variation of a familiar human phenomenon. Their connections
between prototypes and counterparts, forms and counterforms, para-
digms and variants, are, at the very least, resembling versions of the
categories we deploy and the instances of them we then discover in the
course of the study of religion (among other disciplines within the mod-
ern academy). We, too, as an epistemological necessity, imagine ideal
(and, one might say, prototypical) categories (“religion,” “myth,” “rit-
ual,” “Vedism,” “Hinduism,” and so on) in order to organize a world of
otherwise chaotic particulars (or “counterparts”). The process of trans-
forming the disconnected and sui generis into instances of the general
categories (“the Hindu religion,” “the myth of the Cosmic Man,” “the
ritual or sacrifice which is yajna’’) is not at all dissimilar from the Vedic
system of connections or bandhus. To acknowledge these kinds of resem-
blances is to begin to engender a truly humanistic sense of commonality
(which is not, of course, simple identity), which could help overturn
previous assumptions—conscious and unconscious—of radical differ-
ence between the objects and subjects of study.
Moreover, the process of knowing also involves the process of /earn-
ing, of reimagining the categories and what is imagined to be contained
within them. This is what we might label the ongoing reformational work
which complements and perpetually corrects the transformational art of
discovery. Here, too, we have encountered interesting—and com-
parable—refiections of the study of religion in the religious data re-
flected upon in this work.
There is little doubt that over the course of the many centuries
during which the Vedic ritualists were composing the texts they have left
us, they, too, were adjusting the nature of their prototypes, forms, and
paradigms as they rethought the counterparts, counterforms, and vari-
ants that could be connected to them. But such historical manifestations
of reformational work are more clearly visible when we observe the
relationships established between later Hinduism and its Vedic fore-
bear. The orthodox offspring of Vedism, in their appropriation and
reworking of their ancestor, display quite plainly the activity of reform-
ing the old in light of the new.
Hindu doctrine and practice are presented as the already asserted,
the already believed, the already done. Hinduism was (and continues to
p61} REFLECTIONS ON RESEMBLANCE, RITUAL, AND RELIGION
be) presented, in other words, as Vedism or neo-Vedism—and in either

case as a faithful rendition of the established prototype. The authorita-
tive texts of the various Hindu sects are associated, in one way or an-
other, with the prototype of all authoritative texts of the tradition: the
Veda. While claiming to have discovered, in the process of exercising
the transformational art, a new (and supposedly better) version of the
old, Hindus have, from the outsider’s point of view, been engaged in the
reformational work of revising the category in light of new particulars in
need of connection. The particulars, in this case, are the ever-changing
detritus of history: new teachings on the nature of the universe, the
place of human beings in it, the praxis of a powerful and effective life,
the ideals of perfection. And they, by being assimilated into the “un-
changing” (“It is in the Veda”), provoke a revision in what the Veda was
in light of what it has become. The Veda, indeed, may be among the
most rethought and reworked items in the orthodox Hindu tradition.
Another Indian archetype that might, however, at least compete on
this score is the Vedic yajna, “sacrifice.” I noted above that yajna, like
the Veda which first describes it, persists in the imagination of Hindus
(and others), and for many of the same reasons—not because of what
yajna and Veda once were but because of what they continually can be.
Hindus have worked with these categories, they have engaged their
creativity through them, because they have not abandoned them. These
categories have been continually reinfused with the prestige of the ar-
chaic and the fundamental and are thus preserved. Sacrifice no less than
Veda has been a rubric which has made possible the creation of new
instances of it—because it has not ceased to be viable as a classificatory
tool. The work of reformation, it would seem, always presupposes a
certain illusion that some categories don’t change.
Such an illusion, necessary for maintaining the continuity and ortho-
doxy of religious traditions, is not one that is appropriately reduplicated
in scholarship about those traditions, however.
The theory that Vedic ritual—or any other—is without meaning (or
has “intrinsic meaning,” or is “transcendent” of historical and human
meaning and reality) simply reenlivens an essentially religious concep-
tion which is protected from discussion, disputation, and interpretation.
Indeed—so it is maintained, at any rate—discussion, disputation, and
interpretation are precisely what the Vedic ritual sought to eliminate as
it supposedly became purposeless and without meaning outside itself.
While this may or may not be true for the Vedic ritualists (and I have
tried in this work to present evidence that it is not), it is disingenuous for
historians of religion to reposition such notions as scholarly conclusions.
Any thesis for the “meaninglessness” of any human creation, as I

have stated above, is a meaningful interpretation of a previously improp-
erly understood phenomenon—and thus is itself a paradox that cries for
a solution. Ironically, what appears to some philologically trained
Indologists as meaningless—and therefore denied a purpose and a
history—cannot be other than meaningful, and for more reasons than
one.
In the first place, to designate a set of actions as yajna, sacrifice,
ritual, or anything else, is to give them meaning they otherwise do not
have. Of course, these actions have no intrinsic meaning—nothing has
intrinsic meaning. So much is obvious (and hardly cause for a new and
improved theory of ritual). Meaning is always imposed on “naturally”
meaningless phenomena—this is what human beings do, and must do, or
face chaos and insanity. Humans abhor meaninglessness in much the
same way as nature abhors a vacuum.
Secondly, once a set of actions has been transformed into a meaning-
ful entity by virtue of concepts such as yajna, sacrifice, ritual, and so on,
the set and the individual acts within the whole serve as signifiers for any
number of other signifieds. The yajna, for example, can signify the
cosmos, and individual activities within the ritual can signify various
cosmic phenomena.
That the meaning given to both the concept of yajna and the individ-
ual acts subsumed under the concept can and does change is also obvi-
ous. The “meaninglessness of ritual” school ignores the competing and
shifting meanings of yajia while making a mountain out of the molehill
which is the fact that there is no intrinsic and changeless meaning in the
term, in the actions the term makes meaningful, or in what those actions
then come to signify.
For these reasons, the Vedic ritual, yajia, was never meaningless
and is not meaningless for those who write (meaningful) scholarly arti-
cles and books about it. It was understood in different ways by different
people even in Vedic times and remained open to new interpretations
and new meanings in post-Vedic Hinduism (and in the study of Vedism
itself). The “transcendence” or privileged place of the Vedic sacrifice
did not, in fact, imply its eternal “meaninglessness” but rather its infinite
possibilities for meaning.
What cannot be ignored, however, is the fact that there is an un-
bridgeable discrepancy between the claims of the religious and the his-
torical sensibilities that the scholar cannot suspend. Religion, and not
just Vedic or Hindu religion, might be regarded as—among other
things, need it be said—a perpetual denial of the fact that the unchang-
ing ideals it imagines are unrealizable (that, e.g., Hindu texts, beliefs,
and practices are eternally faithful to Vedic texts, beliefs, and practices).
But, unless we are willing to imagine humans vastly different in their
commonsensical recognition of reality when they see it, religion must
also be conceived of as the perpetual reminder to the all-too-human
practitioners of it that their imagined ideals are unrealizable in a real
world of change and imperfection. It is this very fundamental—and
ultimately intractable—incongruity in religious belief and practice that
religions seek to overcome, in vain. Religion is a never-ending search for
reconciliation between the real and the ideal; it is a continual attempt to
coordinate the two.
In addition to the transformational art of religion (and now, I might
add, religion as both the academic subject and object of study) and the
reformational work of religion (again, in both senses of religion), one
might also speak of a coordinative value of religion. By way of conclu-
sion, we might now reflect on whether and in what ways religion as an
object—exemplified here by Vedism and Hinduism—might reveal some-
thing about this aspect of the subject of religion within the academy.
There is a certain sense in which what I have called Vedic resem-
blance might just as well be labeled coordination. The system of
bandhus—iinking macrocosm, sacrifice, and microcosm (adhidevata,
adhiyajna, and adhydtman)—was a strategy to co-order the planes of
reality (or, from our point of view, the planes of the real and the ideal)
and place them in a relation of mutual resemblance (Chapter 3). Other
reapplications of the philosophy of Vedic ritualism and resemblance had
much the same end. The organization of human beings (Chapter 4); the
organization of the ritual repertoire in the paribhasa sections of the
Srauta Sutras (Chapter 5); the organization of Vedic ritualists (Chapter
6) and ritual practice (Chapter 7); as well as the organization of ortho-
dox Hinduism (Chapters 1 and 8)—all these exempla of Vedism in ac-
tion reveal the same impulse to reflection, reduplication, or, if you will,
coordination.
But it was not, in any of these instances, an egalitarian set of interrela-
tions and transformations. Vedism (and Hinduism insofar as it inherited
certain definite and possibly defining features of Vedism) is a system
based not only on mutual resemblance but on hierarchically calibrated
distinctions. It is, in other words, a system based on judgments of better
and worse, more and less complete, higher and lower realization of form.
And the hierarchical yardstick was unfailingly applied to human beings—
the most obvious, and too often heinous, ramification being the Indian
caste system. In India, resemblance was posited not only on difference
The Destiny of Vedism a he
(which is the necessary precondition for making connections) but on

differences differently valued.
On the other hand, egalitarianism does have one ubiquitous incarna-
tion in India, but it is one that seems to destroy rather than to democra-
tize the system of resemblances. Vedantic philosophy, in nuce in certain
ancient Upanishads and in full bloom in the darsana that carries it
through Hinduism, pushes Vedic-Hindu resemblance, based on hierar-
chical distinctions, into identity based on egalitarianism. The Vedic-
Hindu world is reconstituted by Vedanta into a false world of maya, and
the truth of unity is attained only at the price of the denial of its applica-
bility in human life within society. Egalitarianism, in India, has almost
always entailed renunciation or samnydasa.“
I have attempted in this book as a whole to reimagine Vedism and
Hinduism in a way that highlights the similarities between the object
studied and the subject who studies; to underline the points of resem-
blance while not confusing them with identities; to rethink the assumed
differences without denying all difference; to make a product of the
human imagination there and then, and a product of the human imagina-
tion here and now, reflect each other. Whether one can do so in an
egalitarian mode, without ceding the right to make hierarchical value
judgments, is, one might say, the dominant and as yet unsolved problem
that plagues all humanistic studies in our age.
64. See Dumont’s important essay, “World Renunciation in Indian Religions.”

Bibliography
Sanskrit Texts and Translations

Ritual Literature
Agnivesya Grhya Satra. Ed. by L. A. Ravi Varma. Trivandrum Sanskrit Series,
no. 144. Trivandrum: University of Travancore, 1940.
Aitareya Aranyaka. Ed. and trans. by A. B. Keith. Oxford: Clarendon Press,
1909; reprint Oxford: Oxford University Press, 1969.
Aitareya Brahmana. 2 vols. Anandasrama-samskrta-granthavalih, granthankha
32. Poona: Anandasrama, 1931.
. 2 vols. Ed. and trans. by M. Haug. Bombay: Government Central Book
Depot, 1863.
. Trans. by A. B. Keith. In Rigveda Brahmanas, Harvard Oriental Series,
Vol. 25. Cambridge: Harvard University Press, 1920; reprint Delhi:
Motilal Banarsidass, 1971.
Aitareya Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Principal Upanisads. Poona: Vaidika Samsodhana Mandala, 1958.
. Trans. by Robert Hume. In The Thirteen Principal Upanishads, 2d ed.
Oxford: Oxford University Press, 1931.
Apastamba Dharma Sitra. Ed. by U. C. Pandeya. Kashi Sanskrit Series, no. 93.
Varanasi: Chowkhamba Sanskrit Series Office, 1969.
. Trans. by Georg Biihler. In The Sacred Books of the Aryas. Sacred
Books of the East, Vol. 2. Oxford: Clarendon Press, 1879; reprint Delhi:
Apastamba Grhya Sutra. Ed. by U. C. Pandey. Kashi Sanskrit Series, no. 59.
Varanasi: Chowkhamba Sanskrit Series Office, 1971.
. Trans. by Hermann Oldenberg. In The Grihya Sutras. Sacred Books of
the East, Vol. 30. Oxford: Oxford University Press, 1886; reprint Delhi:
Apastamba Srauta Sitra. 3 vols. Ed. by R. Garbe. Calcutta: Royal Asiatic
Society of Bengal, 1882-1902.
. 3 vols. Trans. into German by W. Caland. Calcutta: Vandenhoeck and
Ruprecht, 1921; reprint Wiesbaden: Dr. Martin Sandig, 1969.
Bibliography 227
ASvalayana Grhya Sutra. Ed. and trans. by N. N. Sharma. Delhi: Eastern Book
Linkers, 1976.
Asvalayana Srauta Sitra. Ed. by R. Vidyaratna. Calcutta: Asiatic Society of
Bengal, 1874.
. Incomplete trans. by H. G. Ranade. Poona: R. H. Ranade, 1981.
Atharva Veda Samhita. 4 vols. Ed. by V. Bandhu. Hoshiarpur: Vishvesh-
varanand Vedic Research Institute, 1960-62.
. 2 vols. Trans. by W. D. Whitney. Harvard Oriental Series, Vols. 7 and
8. Cambridge: Harvard University Press, 1905; reprint Delhi: Motilal
Banarsidass, 1962.
Atri Smrti. Ed. by Manmatha Nath Dutt. In The Dharma Sastra Texts. Calcutta:
M.N. Dutt, 1908.
Baudhayana Dharma Sutra. Ed. by U. C. Pandeya. Kashi Sanskrit Series, no.
104. Varanasi: Chowkhamba Sanskrit Series Office, 1972.
. Trans. by Georg Bihler. In The Sacred Books of the Aryas. Sacred
Books of the East, Vol. 14. Oxford: Clarendon Press, 1879; reprint
Delhi: Motilal Banarsidass, 1965. -
Baudhayana Grhya Sitra and Paribhasa Siitra. Ed. by R. Shama Sastri. Mysore:
Government Branch Press, 1920; reprint New Delhi: Meharchand Lachh-
mandas, 1982.
Baudhayana Srauta Sutra. 3 vols. Ed. by W. Caland. Calcutta: Asiatic Society of
Bengal, 1904—24; reprint New Delhi: Munshiram Manoharlal, 1982.
Bharadvaja Grhya Sutra. Ed. by H. J. W. Salomons. Leyden: E. J. Brill, 1913;
reprint Delhi: Meharchand Lachhmandas, 1981.
Bharadvaja Srauta Sutra. 2 vols. Ed. and trans. by C. G. Kashikar. Poona:
Vaidika Samsodhana Mandala, 1964.
Brhadaranyaka Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Chandogya Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Drahyayana Grhya Sutra. Ed. by T. U. N. Singh. Muzaffarpur: Shastra Publish-
ing House, 1934. :
Gautama Dharma Sitra. Ed. by Manmatha Nath Dutt. In The Dharma Sastra
Texts. Calcutta: M. N. Dutt, 1908.
. Trans. by Georg Biihler. In The Sacred Books of the Aryas. Sacred
Delhi: Motilal Banarsidass, 1965.
228 Bibliography
Gobhila Grhya Sutra. Ed. by Chintamani Bhattacharya. Calcutta: Metropolitan

Printing and Publishing House, 1936.
the East, Vol. 30. Oxford: Oxford University Press, 1886; reprint. Delhi:
Gobhilaputra Grhyasamgrahaparisista. Ed. and trans. into German by Maurice
Bloomfield. Leipzig: G. Kreysing, 1881.
Gopatha Brahmana. Ed. by R. Mitra and H. Vidyabhusana. Calcutta: Biblio-
theca Indica, 1872; reprint Delhi: Indological Book House, 1972.
Hiranyakesin Grhya Siitra. Trans. by Hermann Oldenberg. In The Grihya Sa-
tras. Sacred Books of the East, Vol. 30. Oxford: Oxford University Press,
1886; reprint Delhi: Motilal Banarsidass, 1964.
Hiranyakesin Srauta Sutra and Grhya Siitra. 10 vols. Ed. by K. Agase and
S. Marulakara. Poona: Anandasrama, 1907-32.
Jaiminiya Brahmana. Ed. by R. Vira and L. Chandra. Nagpur: Sarasvati Vihara
Series, 1954.
. Incomplete trans. by H. W. Bodewitz. In Jaiminiya Brahmana I: 1-65.
Leiden: E. J. Brill, 1973.
. Incomplete trans. into German by W. Caland. Amsterdam: Johannes
Miller, 1919; reprint Wiesbaden: Dr. Martin Sandig, 1970.
Jaiminiya Grhya Sutra. Ed. and trans. by W. Caland. Punjab Sanskrit Series,
no. 2. Lahore: Moti Lal Banarsi Dass, 1922; reprint Delhi: Motilal
Banarsidass, 1984.
Jaiminiya Srauta Satra. Included in Jaiminiya-Srauta-Siitra-Vrtti of Bhavatrata.
Ed. by Premnidhi Shastri. Sata-Pitaka Series, Vol. 40. New Delhi: Inter-
national Academy of Indian Culture, 1966.
Jaiminiya Upanisad Brahmana. Ed. and trans. by Hanns Oertel. Journal of the
American Oriental Society 16 (1896): 79-260.
Jaiminiyanyayamdala of Madhva. Anandasrama Sanskrit Series, Vol. 24. Poona:
Anandasrama, 1892.
Katha Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen Principal
Upanisads. Poona: Vaidika Samsodhana Mandala, 1958.
Kathaka Grhya Siitra. Ed. by W. Caland. Lahore: Research Depts., D.A.V.
College, 1925. {
Kathaka Samhita. Ed. by V. Santavalekar. Bombay: Bharatamudrandlayam,
1943.
Katyayana Srauta Siitra. Ed. by Albrecht Weber. Chowkhamba Sanskrit Series,
no. 104. Reprint Varanasi: Chowkhamba Sanskrit Series Office, 1972.
. Trans. by H. G. Ranade. Pune: Dr. H. G. Ranade and R. H. Ranade,
n.d.
Kausitaki Brahmana. Ed. by H. Bhattacharya. Calcutta Sanskrit College Re-
search Series, no. 73. Calcutta: Sanskrit College, 1970.
Bibliography 229
. Trans. by A. B. Keith. In Rigveda Brahmanas. Harvard Oriental Series,

Vol. 25. Cambridge: Harvard University Press, 1920; reprint Delhi:
Kausitaki Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Khadira Grhya Sitra. Ed. by A. M. Sastri and L. Srinivasacharya. Mysore:
Government Branch Press, 1913.
Latyayana Srauta Siitra. Ed. by Anandachandra Vedantavagisa. Calcutta: Asi-
atic Society of Bengal, 1972.
Maitrayani Samhita. 4 vols. Ed. by L. von Schroeder. Leipzig: F. A. Brockhaus,
1883-86; reprint Wiesbaden: Franz Steiner Verlag, 1970-72.
Maitrayani Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Maitri Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen Principal
Upanisads. Poona: Vaidika Samsodhana Mandala, 1958.
Manava Grhya Sutra. Ed. by R. H. Shastri. Reprint Delhi: Meharchand
Lacchmandas, 1982.
. Trans. by M. J. Dresden. Groningen: J. B. Wolters Uitgevers, 1941.
Manava Srauta Sutra. Ed. and trans. by J. M. van Geldner. Sata-Pitaka Series,
Vol. 27. New Delhi: International Academy of Indian Culture, 1963.
Manu Smrti. 5 vols. Ed. by J. H. Dave. Bharatiya Vidya Series. Bombay:
Bharatiya Vidya Bhavan, 1972-82.
. Trans. by Georg Buhler. Sacred Books of the East, Vol. 25. Oxford:
Clarendon Press, 1886; reprint New York: Dover Publications, 1969.
Mundaka Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Pancavimsa Brahmana. 2 vols. Ed. by P. A. Cinnaswami Sastri and P. Pattabhir-
ama Sastri. Kashi Sanskrit Series, no. 105. Benares: Sanskrit Series Of-
fice, 1935.
. Trans. by W. Caland. Bibliotheca Indica, no. 255. Calcutta: Asiatic
Society of Bengal, 1931.
Parasgara Smrti. Ed. by Manmatha Nath Dutt. In The Dharma Sastra Texts.
Calcutta: M. N. Dutt, 1908.
230 Bibliography
Paraskara Grhya Siatra. Ed. by M. G. Bakre. Bombay: Gujrati Printing Press,

1917; reprint Delhi: Meharchand Lachhmandas, 1982.
Purvamimamsa Siatras of Jaimini. 2 vols. Ed. and trans. by Mohan Lal Sandal.
In Mimamsa Sitras of Jaimini. Reprint Delhi: Motilal Banarsidass, 1980.
Rg Veda Samhita. 4 vols. Ed. by F. Max Miiller. Chowkhamba Sanskrit Series,
Vol. 99. Reprint Varanasi: Chowkhamba Sanskrit Series Office, 1966.
. 4vols. Trans. into German by Karl Friedrich Geldner. Harvard Oriental
Series, Vols. 33-36. Cambridge, Mass.: Harvard University Press, 1951—
EF
. Partial trans. by Wendy Doniger O’Flaherty. New York: Penguin
Books, 1981.
Sadvimsa Brahmana. Ed. by B. R. Sharma. Kendriya Sanskrit Vidyapeetha
Series, no. 9. Tirupati: Kendriya Sanskrit Vidyapeetha, 1967.
. Trans. by W. B. Bolle. Utrecht: Drukkerij A. Storm, 1956.
Sankhayana Grhya Sutra. Ed. and trans. by S. R. Sehgal. Delhi: Munshiram
Manoharlal, 1960.
Sankhayana Srauta Sutra. 2 vols. Ed. by A. Hillebrandt. Reprint New Delhi:
Meharchand Lachhmandas, 1981.
. Trans. by W. Caland. Nagpur: International Academy of Indian Litera-
ture, 1953; reprint Delhi: Motilal Banarsidass, 1980.
Satapatha Brahmana. 5 vols. Bombay: Laxmi Venkateshwar Steam Press, 1940.
. 5 vols. Trans. by Julius Eggeling. Sacred Books of the East, Vols. 12,
26, 41, 43, 44. Oxford: Clarendon Press, 1882-1900; reprint Delhi:
Taittiriva Aranyaka. 2 vols. Anandasrama-samskrta-granthavalih, granthankha
36. Poona: Anandaérama, 1981.
Taittiriya Brahmana. 3 vols. Anandasrama-samskrta-granthavalih, granthankha
. Partial trans. by P.-E. Dumont. In Proceedings of the American Philo-
sophical Society 92, 95, 98, 101, 107, 108, 109, 113.
Taittiriya Samhita. 8 vols. Anandasrama-samskrta-granthavalih, granthankha
. 2 vols. Trans. by A. B. Keith. Harvard Oriental Series, Vol. 18 and 19.
Cambridge, Mass.: Harvard University Press, 1914; reprint Delhi:
Taittiriya Upanisad. Ed. by V. P. Limaye and R. D. Vadekar. In Eighteen
Bibliography 231

Tantravartiika of Kumarila Bhatta. Ed. and trans. by G. Jha. Calcutta: Biblio-
theca Indica, 1924.
Vaikhanasa Grhya Siitra. Ed. by W. Caland. Bibliotheca Indica, no. 242. Cal-
cutta: Asiatic Society of Bengal, 1927.
. Trans. by W. Caland. Bibliotheca Indica, no. 251. Calcutta: Asiatic
Society of Bengal, 1929.
Vaikhanasa Srauta Sitra. Ed. by W. Caland. Bibliotheca Indica, no. 265. Cal-
cutta: Royal Asiatic Society of Bengal, 1941.
Varaha Grhya Sutra. Ed. and trans. into French by Pierre Rolland. Aix-en-
Provence: Publications universitaires de lettres et sciences humaines,
1971.
Varaha Srauta Sutra. Ed. by W. Caland and R. Vira. Reprint Delhi:
Meharchand Lachhmandas, 1971.
Vasistha Dharma Sitra. Ed. by Manmatha Nath Dutt. In The Dharma Sastra
Texts. Calcutta: M. N. Dutt, 1908.
. Trans. by Georg Bithler. In The Sacred Books of the Aryas. Sacred
Vedanta Sutras. With commentaries of Sankaracarya and Ramanuja. Trans. by
George Thibaut. Sacred Books of the East, Vols. 34, 38, 48. Oxford:
Oxford University Press, 1904; reprint Delhi: Motilal Banarsidass, 1962.
Yajnvalkya Smrti, 2d ed. Ed. by Mahamahopadhyaya T. Ganapati Sastri. New
Delhi: Munshiram Manoharlal, 1982.
Other Texts and Collections
Acarya, Rama Sarma, ed. Matsya Purana. Bareli: Samskrti Samsthana, 1970.
Biihler, Georg, trans. Sacred Laws of the Aryas. 2 vols. Sacred Books of the
East, Vols. 2 and 14. Oxford: Clarendon Press, 1879-82; reprint Delhi:
Dandekar, R. N., ed. Srautakoga: Encyclopedia of Vedic Sacrificial Ritual. 2
vols. Poona: Vaidika Samsodhana Mandala, 1958-82.
Datta, Bhagavad, ed. Nirukta Sastram. Amritsar: Ramalal Kapur Nikshepa
Trust, 1965.
Dutt, Manmatha Nath, ed. The Dharma Sastra Texts. Calcutta: M. N. Dutt,
1908.
Ghosh, B., ed. Collection of the Fragments of Lost Brahmanas. Calcutta: Mod-
ern Publishing Syndicate, 1935; reprint Delhi: Meharchand Lachhman-
das, 1982.
Gupta, Sri Anand Swarup, ed. The Kirma Purana. Varanasi: All-India Kashi
Raj Trust, 1972.
232 Bibliography
. The Vamana Purana. Varanasi: All-India Kashi Raj Trust, 1968.

Hume, Robert Ernest, trans. The Thirteen Principal Upanishads, 2d ed. Oxford:
Oxford University Press, 1931.
Jacobi, H., trans. Jaina Sutras. 2 vols. Sacred Books of the East, Vols. 22 and
45. Oxford: Clarendon Press, 1884-85; reprint New York: Dover Publica-
tions, 1968.
Limaye, V. P., and R. D. Vadekar, eds. Eighteen Principal Upanisads. Poona:
Vaidika Samsodhana Mandala, 1958.
Oldenberg, Hermann, trans. The Grihya Sutras. 2 vols. Sacred Books of the
East, Vols. 29 and 30. Oxford: Oxford University Press, 1886; reprint
Rhys Davids, T. W., trans. Dialogues of the Buddha. 3 vols. Sacred Books of the
Buddhists, Vols. 2-4. London: Pali Text Society, 1899; reprint London:
Routledge and Kegan Paul, 1973-77.
Sastri, Pandit Hrishikesa, ed. The Varaha Purana, 2d ed. Varanasi: Chaukhamba
Amarabharati Prakashan, 1982.
Shastri, J. L., ed. Brahmavaivarta Purana. 2 vols. Delhi: Motilal Banarsidass,
1984.
———. Linga Purana. Delhi: Motilal Banarsidass, 1980.
. Bhagavata Purana. Delhi: Motilal Banarsidass, 1983.
. Brahmanda Purana. Delhi: Motilal Banarsidass, 1973.
Sukthankar, Vishnu S., ed. The Mahabharata. 19 vols. Poona: Bhandarkar Ori-
ental Research Institute, 1933-60.
Van Buitenen, J. A. B., ed. and trans. The Mahabharata. 3 vols. Chicago:
University of Chicago Press, 1973-78.
Zaehner, R. C., trans. The Bhagavad Gita. New York: Oxford University Press,
1969.
Secondary Sources (Indological)
Abhyankar, K. V. “On paribhasa Works in Sanskrit Grammar.” Annals of the

Bhandarkar Oriental Research Institute 36 (1955): 157-62.
Apte, V. M. Social and Religious Life in the Grhyasiitras. Ahmedabad: V. M.
Apte, 1939.
Aquilar, H. The Sacrifice in the Rgveda. Delhi: Bharatiya Vidya Prakashan,
1976.
Banerjea, A. C. Studies in the Brahmanas. New Delhi: Motilal Banarsidass,
1963.
Banerjee, Santi. “Prajapati in the Brahmanas.” Vishveshvaranand Indological
Journal 19 (June—December 1981): 14-19.
Barnett, L. D. Hinduism. London: Constable, 1913.
Barth, Auguste. Les Religions de I’Inde: Extrait de l’Encyclopédie des sciences
religieuses. Paris: G. Fischbacher, 1879.
Bibliography 233
Barua, B. M. “Art as Defined in the Brahmanas.” Jndian Culture 1 (July 1934-

April 1935): 118-20.
Basu, J. India in the Age of the Brahmanas. Calcutta: Sanskrit Pustak Bhandar,
1969.
Bedekar, D. K. “The Revelatory Character of Indian Epistemology.” Annals of
the Bhandarkar Research Institute 29 (1948): 64-84.
Bergaigne, Abel. Recherches sur l’histoire de la liturgie védique. Paris:
Imprimerie nationale, 1889.
. La Religion védique d’apreés les hymnes du Rig-Veda. 3 vols. Paris: F.
Vieweg, 1878-83.
Bhankarkar, Ramakrishna Gopal. “The Veda in India.” Indian Antiquary 3
(May 1874): 132-35.
Bharati, Agehandana. “The Hindu Renaissance and Its Apologetic Patterns.”
Journal of Asian Studies 29 (February 1970): 267-88.
. The Tantric Tradition. New York: Samuel Weiser, 1975.
Bhargava, P. L. India in the Vedic Age: A History of Aryan Expansion in India,
2d ed. Aminabad: Upper India Publishing House, 1971.
Bhattacharji, S. “Rise of Prajapati in the Brahmanas.” Annals of the Bhandarkar
Bhattacharya, D. “Cosmogony and Rituo-Philosophical Integrity in the Athar-
vaveda.” Vishveshvaranand Indological Journal 15 (March 1977): 1-12.
Bhattacharya, Jogendra Nath. Hindu Castes and Sects. Calcutta: Editions In-
dian, 1968.
Bhattacharya, Kamaleswar. “Le ‘Védisme’ de certains texts hindouistes.” Jour-
nal asiatique 225 (1967): 199-222.
Bhattacharya, V. C. “On the Justification of rupasamrddha rk- Verses in the
Aitareya Brahmana.” Our Heritage 4 (1956): 99-106, 227-37; 5 (1957):
119-46.
Bhide, V. V. The Caturmasya Sacrifices. Pune: University of Poona, 1979.
Biardeau, Madeleine. “L’Inde et Vhistoire.” Revue historique 234 (1965): 47-S8.
Biardeau, Madeleine, and Charles Malamoud. Le Sacrifice dans I’Inde ancienne.
Bibliothéque de I’Ecole des Hautes Etudes, Sciences religieuses, vol. 79.
Paris: Presses Universitaires de France, 1976.
Biswas, S. “Uber das Vratyaproblem in der vedischen Ritual-literatur.” Zeit-
schrift der Deutschen Morgenlandischen Gesellschaft 105 (1955): 53-54.
Bloomfield, Maurice. The Atharva-Veda and the Gopatha Brahmana. Strass-
burg: Karl J. Triibner, 1899; reprint New Delhi: Asian Publication Ser-
vices, 1978.
. The Religion of the Veda. New York: G. P. Putnam’s Sons, 1908.
Bodewitz, H. W. The Daily Evening and Morning Offering (Agnihotra) accord-
ing to the Brahmanas. Leiden: E. J. Brill, 1976.
Bonazzoli, Giorgio. “The Dynamic Canon of the Puranas.” Purana 26 (July
1979): 116-66.
234 Bibliography
Bouquet, A. C. Hinduism. London: Hutchinson’s University Library, 1948.

Brown, Cheever Mackenzie. God as Mother: A Feminine Theology in India.
Hartford, Vt.: Claude Stark, 1974.
. “Purana as Scripture: From Sound to Image of the Holy Word in the
Hindu Tradition.” History of Religions 26 (1986): 68-86.
Brown, G. W. The Human Body in the Upanishads. Jubbulpore: Christian
Mission Press, 1921.
Brown, W. Norman. “The Creation Myth of the Rg Veda.” Journal of the
. Man in the Universe: Some Cultural Continuities in Indian Thought.
Berkeley: University of California Press, 1970.
. “The Sources and Nature of Purusa in the Purusa Sukta.” Journal of the
. “Theories of Creation in the Rg Veda.” Journal of the American Orien-
tal Society 85 (1965): 23-34.
Caland, Wilhelm. Die Altindischen Todten- und Bestattungs-gebrduche. Amster-
dam: Miller, 1896; reprint Wiesbaden: Martin Sandig, 1967.
. Altindischer Ahenenkult: Das Sraddha nach den verschiedene Schulen mit
benutzund handschriftliche quellen dargestellt. Leiden: E. J. Brill, 1893.
———. “On the Relative Chronology of Some Ritualistic Sutras.” Acta Orien-
talia 9 (1931): 69-76.
. Uber das rituelle Sitra des Baudhayana. Abhandlungen fiir die Kunde
des Morgenlandes herausgegeben von der Deutschen Morganlandischen
Gesellschaft, Vol. 12, no. 1. Leipzig: F. A. Brockhaus, 1903.
Caland, Wilhelm, and V. Henry. L’Agnistoma: Description complete de la forme
normale du sacrifice de Soma dans le cult védique. 2 vols. Paris: Ernest
Leroux, 1906.
Chakrabarti, S. C. The Paribhasas in the Srautasiitras. Calcutta: Sanskrit Pustak
Bhandar, 1980.
Chakrabarti, S. K. “On the Transition of the Vedic Sacrificial Lore.” Indo-
Iranian Journal21 (1979): 181-88.
Chatterjee, H. N. Studies in Some Aspects of Hindu Samskaras in Ancient India.
Calcutta: Sanskrit Pustak Bhandar, 1965.
Chauduri, Nirad C. The Continent of Circe. Bombay: Jaico Publishing House,
1965.
Choudhary, Radhakrishna. Vrdatyas in Ancient India. Chowkhamba Sanskrit
Studies, Vol. 38. Benares: Chowkhamba Sanskrit Series Office, 1964.
Coburn, Thomas B. Devi Mahatmya: The Crystallization of the Goddess Tradi-
tion. Columbia, Mo.: South Asia Books, 1985.
. “ ‘Scripture’ in India: Towards a Typology of the Word in Hindu Life.”
Journal of the American Academy of Religion 52 (September 1984): 435—
59.
. “The Study of the Puranas and the Study of Religion.” Religious Studies
16 (1980): 341-52.
Bibliography 235
Coomaraswamy, Ananda K. “Atmayajiia: Self-Sacrifice.” Harvard Journal of

Asiatic Studies 6 (1941): 358-88.
. “Vedic Exemplarism.” Harvard Journal of Asiatic Studies 1 (1936): 44-
64.
Das, Veena. “Language of Sacrifice.” Man (new series) 18 (September 1983):
445-62.
Derrett, J. D. M. Religion, Law and the State in India. London: Faber and
Faber, 1958.
Deussen, Paul. Philosophy of the Upanishads. Trans. by A. S. Geden. New
York: Dover Publications, 1966.
Devasthali, G. V. Religion and Mythology of the Brahmanas. Poona: University
of Poona, 1965.
Dimock, Edward C. The Place of the Hidden Moon. Chicago: University of
Chicago Press, 1966.
Dimock, Edward C., Edwin Gerow, C. M. Naim, A. K. Ramanujan, Gordon
Roadarmel, and J. A. B. van Buitenen. The Literatures of India: An
Introduction. Chicago: University of Chicago Press, 1974.
Drury, Naama. The Sacrificial Ritual in the Satapatha Brahmana. Delhi: Motilal
Banarsidass, 1981.
Dumont, Louis. La Civilisation indienne et nous. Paris: Armand Colin, 1975.
. Homo Hierarchicus: The Caste System and Its Implications, complete
rev. English ed. Trans. by Mark Saisbury, Louis Dumont, and Basia
Gulati. Chicago: University of Chicago Press, 1980.
. Religion/Politics and History in India: Collected Papers in Indian Sociol-
ogy. The Hague: Mouton Publishers, 1970.
. “World Renunciation in Indian Religions.” Contributions to Indian Soci-
ology 4 (1960): 33-62. Reprinted in Religion/Politics and History in In-
dia: Collected Papers in Indian Sociology. The Hague: Mouton Publish-
ers, 1970, pp. 33-60.
Dumont, P.-E. L’Agnihotra: Description de lagnihotra dans le rituel védique
d’apres les Srautasiitras. Baltimore: Johns Hopkins University Press,
1939.
—. “The Agnihotra (or Fire-God Oblation) in the Taittiriya-Brahmana:
The First Prapathaka of the Second Kanda of the Taittirilya-Brahmana
with Translation.” Proceedings of the American Philosophical Society 108
(1964): 337-53.
— . “The Animal Sacrifices in the Taittirlya-Brahmana: The Part of the
Hotar and the Part of the Maitravaruna in the Animal Sacrifice.” Proceed-
ings of the American Philosophical Society 107 (1962): 246-63.
. L’A§vamedha: Description du sacrifice solennel du cheval dans le culte
védique d’apres les textes du Yajurveda blanc. Paris: P. Geuthner, 1927.
. “The Full-moon and New-moon Sacrifices in the Taittirlya-Brahmana.”
Proceedings of the American Philosophical Society 101 (1957): 216-43;
103 (1959): 584-608; 104 (1960): 1-10.
236 Bibliography
. “The Horse-Sacrifice in the Taittirilya-Brahmana: The Eighth and Ninth

Prapathakas of the Third Kanda of the Taittiriya-Brahmana with Transla-
tion.” Proceedings of the American Philosophical Society 92 (1948): 447—
503.
. “The Human Sacrifice in the Taittiritya-Brahmana: The Fourth Prapa-
thaka of the Third Kanda of the Taittirilya-Brahmana with Translation.”
Proceedings of the American Philosophical Society 107 (1963): 177-82.
. “The Istis to the Naksatras (or Oblations to the Lunar Mansions) in the
Taittiriya-Brahmana with Translation.” Proceedings of the American
Philosophical Society 98 (1954): 204-23.
. “The Kamya Animal Sacrifices in the Taittiriya-Brahmana: The Eighth
Prapathaka of the Third Kanda of the Taittiriya-Brahmana with Transla-
tion.” Proceedings of the American Philosophical Society 113 (1969): 34—
66.
———. “The Kaukili-Sautramani in the Taittirftya-Brahmana: The Sixth Prapa-
thaka of the Second Kanda of the Taittiriya-Brahmana with Transla-
tion.” Proceedings of the American Philosophical Society 109 (1965):
309-41.
———.. “The Special Kinds of Agnicayana (or Special Methods of Building the
Fire-Altar) according to the Kathas in the Taittiriya Brahmana: The
Tenth, Eleventh, and Twelfth Prapathakas of the Third Kanda of the
Taittiriya Brahmana with Translation.” Proceedings of the American
Philosophical Society 95 (1951): 628-75.
. “Taittiriya-Brahmana 3.7.7—-10 and 3.7.12-14: Seven Anuvakas of the
Seventh Prapathaka of the Third Kanda of the Taittiriya-Brahmana with
Translation.” Proceedings of the American Philosophical Society 107
(1963): 446-60.
Edgerton, Franklin. “Dominant Ideas in the Formation of Indian Culture.”
Journal of the American Oriental Society 62 (1942): 151-56.
. “Philosophical Materials of the Atharvaveda.” In Studies in Honor of
Maurice Bloomfield. New Haven: Yale University Press, 1920, pp. 119-
a5:
. “Sources of the Filosophy [sic] of the Upanisads.” Journal of the Ameri-
can Oriental Society 36 (1917): 197-204.
. “The Upanisads: What Do They Seek and Why?” Journal of the Ameri-
can Oriental Society 19 (1929): 97-121.
Eliade, Mircea. Yoga: Immortality and Freedom, 2d ed. Trans. by Willard R.
Trask. New York: Harcourt, Brace and World, 1959.
Eliot, Charles. Hinduism and Buddhism: An Historical Sketch. 3 vols. New
York: Barnes and Noble, 1954.
Falk, Harry. “Vedisch upanisad.” Zeitschrift der Deutschen Morgenlandischen
Gesellschaft 136 (1986): 80-97.
Falk, Maryla. Nama-Ripa and Dharma-Riapa. Calcutta: University of Calcutta,
1943.
Bibliography 237
Filliozat, Jean. “La Force organique et la force cosmique dans la philosophie

médicale de I’Inde et dans le Véda.” Revue philosophique 116 (1933):
410-29.
Gampert, W. Die Stihnenzeremonien in der altindischen Rechtslitteratur. Prag:
Orientalisces Institut, 1939.
Gandhi, Mohandas. Jn Search of the Supreme. Comp. and ed. by V. B. Kher.
Ahmedabad: Navajivan Publishing, 1961.
Ghosh, Manomohan. “Pratisakhyas and Vedic Sakhas.” Indian Historical Quar-
terly 11 (1935): 761-68.
Glasser, K. “Der indische Student.” Zeitschrift der Deutschen Morgenldndischen
Gesellschaft 66 (1912): 1-37.
Goman, Thomas G., and Ronald S. Laura. “A Logical Treatment of Some
Upanisadic Puzzles and Changing Conceptions of Sacrifice.” Numen 19
(April 1972): 52-67.
Gonda, Jan. “Adhvara and Adhvaryu.” Vishveshvaranand Indological Journal 3
(September 1965): 163-77.
—.. “All, Universe and Totality in the Satapatha Brahmana.” Journal of the
Oriental Institute (Baroda) 32 (September—December 1982): 1-17.
—.. “Ayatana.” Adyar Library Bulletin 23 (1969): 1-79.
. “Bandhu in the Brahmanas.” Adyar Library Bulletin 29 (1965): 1-29.
. “The Baudhayana-Grhya-Paribhasa-Sutra.” In H. Hartel, ed., Beitriage
zur Indienforschuung: Ernest Waldschmidt zum 80. Geburtstag gewidmet.
Berlin: Museum fiir Indische Kunst, 1977, pp. 169-90.
. Change and Continuity in Indian Religion. Disputations Rheno-Trajec-
tinae, Vol. 9. The Hague: Mouton, 1965.
. “The Creator and His Spirit.” Wiener Zeitschrift fiir die Kunde Stida-
siens 27 (1983): 5—42.
. “Etymologies in the Ancient Indian Brahmanas.” Lingua 5 (1955): 61-
85.
. The Haviryajnah Somah: The Interrelations of the Vedic Solemn Sacri-
fices. Sankhayana Srauta Siitra 14, 1-13. Translation and Notes. Amster-
dam: North-Holland Publishing, 1982.
, gen. ed. A History of Indian Literature. Wiesbaden: Otto Harrassowitz,
1975. Vol. 1, fasc. 1: Vedic Literature (Samhitas and Brahmanas), by Jan
Gonda.
, gen. ed. A History of Indian Literature. Wiesbaden: Otto Harrassowitz,
1977. Vol. 1, fasc. 2: The Ritual Sutras, by Jan Gonda.
—. Hymns of the Rgveda Not Employed in the Solemn Ritual. Amsterdam:
North-Holland Publishing, 1978.
. “In the Beginning.” Annals of the Bhandarkar Oriental Research Insti-
tute 63 (1982): 43-62.
. “The Indian Mantra.” Oriens 16 (1963): 242-97.
. Loka. World and Heaven in the Veda. Verhandelingen der Koninklijke
Nederlandse Akademie van Wetenschappen, Afd. Letterkunde, Nieuwe
238 Bibliography
Reeks, Deel 73, no. 1. Amsterdam: N. V. Noord-Hollandsche Uitgevers

Maatschappij, 1966.
—.. Notes on Brahman. Utrecht: J. L. Beyers, 1950.
——. “The Popular Prajapati.” History of Religions 22 (November 1982):
129-49.
———. “Prajapati and Prayascitta.” Journal of the Royal Asiatic Society (1983):
32-54.
. Prajapati and the Year. Amsterdam: North Holland Publishing, 1984.
. Prajapati’s Rise to Higher Rank. Leiden: E. J. Brill, 1986.
——. “Pratistha.” Samjndvyakarana, Studia Indological Internationalia 1
(1954): 1-37.
. “The Redundant and the Deficient in Vedic Ritual.” Vishveshvaranand
Indological Journal 21 (June—December 1983): 1-34.
. “Reflections on Sarva- in Vedic Texts.” Indian Linguistics 16 (Novem-
ber 1955): 53-71.
. The Savayajnas (Kausikasttra 60-68). Amsterdam: North-Holland Pub-
lishing, 1965.
———.. “Upanayana.” /ndologica Taurinensia 7 (1979): 253-59.
—. “Vedic Gods and the Sacrifice.” Numen 30 (July 1983): 1-34.
. Vedic Ritual: The Non-Solemn Rites. Leiden: E. J. Brill, 1980.
Gopal, Ram. Jndia of the Vedic Kalpasttras. Delhi: National Publishing House,
1959,
. “Influence of the Brahmanas on the Grhyasutras.” Vishveshvaranand
Indological Journal | (1963): 291-98.
———. “Manu’s Indebtedness to Sankhayana.” Poona Orientalist 27 (1962):
39-44.
Goswami, Krishna Gopal. “Philosophy of the Pancayajnas.” Calcutta Review 65
(1937): 203-10.
Goudriaan, T. “Vaikhanasa Daily Worship.” Indo-Iranian Journal 12 (1970):
161-215.
Govinda, L. A. Psycho-Cosmic Symbolism of the Buddhist Stipa. Emeryville,
Calif.: Dharma Publishing, 1976.
Griswold, Hervey DeWitt. /nsights into Modern Hinduism. New York: Henry
Holt, 1934.
Gupta, S. K. “Ancient Schools of Vedic Interpretation.” Journal of the Ganga-
natha Jha Research Institute 16 (November 1958—February 1959): 143—
53:
Hacker, Paul. “Sraddha.” Wiener Zeitschrift fiir die Kunde Siid-und Ostasiens 7
(1963): 151-89.
Hastings, J., ed. Encyclopedia of Religion and Ethics. New York: Charles Scrib-
ner’s Sons, 1910. S.v. “Brahman,” by Alfred Hillebrandt.
Hauer, J. W. Der Vratya: Untersuchenungen tiber die nichtbrahmanische Reli-
gion altindiens. 2 vols. Stuttgart: W. Kohlhammer, 1927.
Bibliography 239
Haug, Martin, ed. and trans. The Aitareya Brahmanam of the Rigveda. 2 vols.
Bombay: Government Central Book Depot, 1863.
Hazra, R. C. Studies in the Upapuranas. Calcutta: Sanskrit College, 1963.
Heesterman, J. C. The Ancient Indian Royal Consecration: The Rajastiya De-
scribed according to the Yajus Texts and Annoted [sic]. Disputations
Rheno-Trajectinae, Vol. 2. s’;Gravenhage: Mouton, 1957.
. “Die Autoritat des Veda.” In Gerhard Oberhamme?, ed., Offenbarung,
geistige Realitat des Menschen. Vienna: Indologisches Institut der Uni-
versitat Wien, 1974, pp. 29-40.
. “Brahmin, Ritual and Renouncer.” In The Inner Conflict of Tradition:
Essays in Indian Ritual, Kingship, and Society. Chicago: University of
Chicago Press, 1985, pp. 26-44.
———.. “Householder and Wanderer.” In T. N. Madan, ed., Way of Life: King,
Householder, Renouncer. Essays in Honor of Louis Dumont. New Delhi:
Vikas, 1981, pp. 251-71.
. “Other Folks’ Fire.” In J. F. Staal, ed., Agni: The Vedic Ritual of the
Fire Altar. 2 vols. Berkeley, Calif.: Asian Humanities Press, 1983, II: 76-
94.
. “Power and Authority in Indian Tradition.” In R. J. Moore, ed., Tradi-
tion and Politics in South Asia. New Delhi: Vikas, 1979, pp. 60-85.
. “Priesthood and the Brahmin.” Contributions to Indian Sociology (new
series) 5 (1971): 43-47.
—. “Reflexions on the Significance of Daksinas.” Indo-Iranian Journal 3
(1959): 241-58.
. “The Return of the Veda Scholar (samdavartana).” In J. C. Heesterman,
ed., Pratidanam: Indian, Iranian and Indo-European Studies Presented to
F. B. J. Kuier on his Sixtieth Birthday. The Hague: Mouton, 1968, pp.
436-47.
. “The Ritualist’s Problem.” In S. D. Joshi, ed., Amrtadhara. Professor
R. N. Dandekar Felicitation Volume. Delhi: Ajanta Books, 1984, pp.
167-79.
. “Self-sacrifice in Vedic Ritual.” In S. Shaked et al., eds., Gilgul: Essays
on Transformation, Revolution and Permanence in the History of Reli-
gions. Leiden: E. J. Brill, 1987, pp. 91-106.
. “Tradition in Modern India.” Bijdragen tot de Taal-, Land-, en
Volkenkunde 119 (1963): 237-53.
———. “Veda and Dharma.” In Wendy Doniger O’Flaherty and J. D. M.
Derrett, eds., The Concept of Duty in South Asia. Columbia, Mo.: South
Asia Books, 1978, pp. 80-95.
. “Veda and Society: Some Remarks apropos of the Film ‘Altar of
Fire.’ ” In Asko Parpola, ed., Proceedings of the Nordic South Asia
Conference, Helsinki, June 10-12, 1980. Helsinki: Studia Orientalia,
1981.
240 Bibliography
. “Vratya and Sacrifice.” Indo-Iranian Journal 6 (1962): 1-37.

Heimann, B. “The Supra-Personal Process of Sacrifice.” Rivista degli studi orien-
tali (Rome) 32 (1938): 731-39.
Henry, V. La Magie dans I’Inde antique. Paris: Dujarric, 1904.
Hillebrandt, Alfred. Das altindische Neu- und Vollmondopfer in seiner ein-
fachsten Form. Jena: Gustav Fischer, 1879.
—. Rituallitteratur. Vedische Opfer und Zauber. Grundriss der Indo-
Arischen Philologie und Altertumskunde, Vol. 3. Strassburg: Triibner,
1897.
—. Vedic Mythology. 2 vols. Trans. by S. R. Sarma. Delhi: Motilal
Banarsidass, 1980.
Hiltebeitel, Alf. The Ritual of Battle: Krishna in the Mahabharata. Ithaca, N.Y.:
Cornell University Press, 1976,
Hopkins, Thomas. “The Social Teachings of the Bhagavata Purana.” In Milton
Singer, ed., Krishna: Myths, Rites, and Attitudes. Chicago: University of
Chicago Press, 1966, pp. 3-22.
Hubert, Henri, and Marcel Mauss. Sacrifice: Its Nature and Function. Trans. by
W. D. Halls. Chicago: University of Chicago Press, 1964.
Joshi, J. R. “Prajapati in Vedic Mythology and Ritual.” Annals of the Bhan-
darkar Oriental Research Institute 53 (1972): 101-25.
Joshi, S. D. Patanjali’s Vyakaranamahabhasya. Poona: University of Poona,
1968.
———. Some Minor Divinities in Vedic Mythology and Ritual. Pune: Deccan
College Postgraduate and Research Institute, 1977.
Kaelber, Walter O. “The ‘Dramatic’ Element in Brahmanic Initiation: Symbols
of Death, Danger, and Difficult Passage.” History of Religions 18 (Au-
gust 1978): 54-76.
Kane, P. V. History of Dharmasastra, 2d ed. 5 vols. Poona: Bhandarkar Orien-
tal Research Institute, 1968-75.
Karnik, H. R. “Morals in the Brahmanas.” Journal of Bombay University 27
(September 1958): 95-127.
Kashikar, C. G. “On the Bharadvaja Grhyasitra and Its Commentary.” Bulletin
of the Deccan College Research Institute 35 (1975): 65-75.
. “The Idea of Ultimate Reality and Meaning according to the Kalpa
Sutras.” Ultimate Reality and Meaning 2 (1979): 172-87.
. “The Pitrmedhasitra of Bharadvaja vis-a-vis Apastamba and Satya-
sadha Hiranyakesin.” Journal of Oriental Research (Madras) 28 (1958-
59): 1-10.
_ A Survey of the Srautasitras. Bombay: University of Bombay, 1968.
———.. “The Vedic Sacrificial Rituals through the Ages.” Indian Antiquary 1
(3d series), (April 1964): 77-89.
Kashikar, C. G., and Asko Parpola. “Srauta Traditions in Recent Times.” In
Frits Staal, ed., Agni: The Vedic Ritual of the Fire Altar. 2 vols. Berkeley,
Calif.: Asian Humanities Press, 1983. II: 199-251.
Bibliography 241
Keith, Arthur Berriedale. The Religion and Philosophy of the Veda and
Upanisads. 2 vols. Harvard Oriental Series, Vols. 31 and 32. Cambridge,
Mass.: Harvard University Press, 1925.
Kinsley, David R. Hinduism: A Cultural Perspective. Prentice-Hall Series in
World Religions, ed. by Robert S. Ellwood, Jr. Englewood Cliffs, N.J.:
Prentice-Hall, 1982.
Knipe, David M. In the Image of Fire: Vedic Experiences of Heat. Delhi: Motilal
Banarsidass, 1975.
———. “One Fire, Three Fires, Five Fires: Vedic Symbols in Transition.” His-
tory of Religions 12 (August 1972): 28-41.
——.. “Sapindakarana: The Hindu Rite of Entry into Heaven.” In Frank E.
Reynolds and Earle H. Waugh, eds., Religious Encounters with Death:
Insights from the History and Anthropology of Religions. University
Park: Pennsylvania State University Press, 1977, pp. 111-24.
Kramrisch, Stella. The Hindu Temple. 2 vols. Reprint Delhi: Motilal Banar-
sidass, 1976.
Krick, Hertha. Das Rituel der Feuer (Agnyddheya). Osterreichische Akademie
der Wissenschaften. Philosophisch historische Klasse. Sitzungsberichte,
bd. 399. Wien: Osterreichischen Akademie der Wissenschaften, 1982.
Kuiper, F. B. J. Ancient Indian Cosmogony. New Delhi: Vikas Publishing
House, 1983.
Laine, J. “The Notion of ‘Scripture’ in Modern Indian Thought.” Annals of the
Lal, S. K. Female Divinities in Hindu Mythology and Ritual. Pune: University of
Poona, 1980).
Lee, Orlan. “From Acts—to Non-Action—To Acts: The Dialectical Basis for
Social Withdrawal or Commitment to This World in the Buddhist Refor-
mation.” History of Religions 6 (May 1967): 273-302.
Lester, Robert C. “Hinduism: Veda and Sacred Texts.” In Frederick M. Denny
and Rodney L. Taylor, eds., The Holy Book in Comparative Perspective.
Columbia, $.C.: University of South Carolina Press, 1985, pp. 126-47.
Lévi, Sylvain. Le Doctrine du sacrifice dans les Brahmanas. Bibliotheque de
V’Ecole des Hautes Etudes, Sciences religieuses, Vol. 11. Paris: Ernest
Leroux, 1898.
Lincoln, Bruce. “Death and Resurrection in Indo-European Thought.” Journal
of Indo-European Studies 5 (Summer 1977): 247-64.
—. Myth, Cosmos, and Society: Indo-European Themes of Creation and
Destruction. Cambridge, Mass.: Harvard University Press, 1986.
Lingat, Robert. Classical Law of India. Trans. by J. Duncan M. Derrett. Berke-
ley: University of California Press, 1973.
_ “Time and the Dharma.” Contributions to Indian Sociology 6 (1962): 7-
16.
Lobbecke, R. Uber das Verhdltnis von Brahmanas und Srautasiitren. Leipzig:
Kreysing, 1908.
242 Bibliography
Lutgendorf, Philip. The Life of a Text: Tulsidas’s Ramacaritamanasa in Perfor-

mance. Ph.D. dissertation. University of Chicago, 1986.
Macdonald, A. W. “A propos de Prajapati.” Journal asiatique 240 (1952): 323—
38.
Macdonald, K. S. The Brahmanas of the Vedas. Reprint Delhi: Bharatiya Book
Corporation, 1979.
Macdonell, A. A. A History of Sanskrit Literature. New York: D. Appleton,
1900.
. The Vedic Mythology. Grundiss der Indo-Arische Philologie und Al-
terthumskunde, Vol. 3. Strassburg: K. J. Triibner, 1894; reprint Vara-
nasi: Indological Book House, 1963.
Malamoud, Charles. Le Svadhydaya: Recitation personnelle du Veda. Paris: Insti-
tut de Civilisation Indienne, 1977.
. “Terminer le sacrifice: Remarques sur les honoraires rituels dans le
brahmanisme.” In Madeleine Biardeau and Charles Malamoud, Le Sacri-
fice dans I’Inde ancienne. Paris: Presses universitaires de France, 1976,
pp. 155-204.
. “La Théologie de la dette dans le brahmanisme.” Purusartha 4 (1980):
39-62.
Marshall, P. J., ed. The British Discovery of Hinduism in the Eighteenth Cen-
tury. Cambridge: Cambridge University Press, 1970.
Mayrhofer, Manfred. Kurzgefasstes etymologisches Worterbuches Altindischen.
A Concise Etymological Sanskrit Dictionary. 4 vols. Heidelberg: Carl
Winter Universitats-verlag, 1956-80.
Mayrhofer-Passler, E. “Haustieropfer bei dem Indoiraniern und den andern
indo-germanischen Volkern.” Archiv Orientalni 21 (1953): 182-205.
Minard, Armand. Trois enigmas sur les cent chemins: Recherches sur le Sata-
patha Brahmana. 2 vols. Annales de l'Université de Lyon, troisime se-
ries, lettres, fascs. 17-18. Paris: Société d’édition les belles lettres, 1949.
Moghe, S. G. “The Evolution of the Mimamsa Technical Term atidega.” Annals
of the Bhandarkar Oriental Research Institute 58-59 (1977-78): 777-84.
Mookerji, Radha Kumud. Ancient Indian Education (Brahmanical and Bud-
dhist). London: Macmillan, 1947.
Miller, F. Max. Chips from a German Workshop. 2 vols. New York: Scribner,
Armstrong, 1871.
. A History of Ancient Sanskrit Literature, 2d ed. London: Williams and
Norgate, 1859.
Murthy, U. R. Anantha. Samskara: A Rite for a Dead Man. Trans. by A. K.
Ramanujan. Delhi: Oxford University Press, 1978.
Mus, Paul. Barabadur: Esquisse d'une histoire du Bouddhism fondée sur la
critique archéologique des textes. Paris and Hanoi: Paul Geuthner, 1935.
. “Du Nouveau sur Rg Veda 10.90?” In Ernest Bender, ed., Indological
Studies in Honor of W. Norman Brown. American Oriental Series, Vol.
47. New Haven, Conn.: American Oriental Society, 1962, pp. 165-85.
Bibliography 243
. “Ou finit Purusa?” Mélanges d’Indianisme a la mémoire de Louis

Renou. Paris: E. de Boccard, 1968, pp. 539-63.
. “The Problematic of the Self, West and East and the Mandala Pattern.”
In C. A. Moore, ed., Philosophy and Culture, East and West. Honolulu:
University of Hawaii Press, 1960, pp. 594-610.
———. “La Stance de la plénitude.” Bulletin de l’Ecole Francaise d’Extréme
Orient 44 (1947-50): 591-618.
Mylius, Klaus. “Daksina: Eine Studie tiber den altindischen Priesterlohn.”
Altorientalische Forschungen 6 (1979): 141-79.
. “Die Ideenwelt des Satapatha Brahmana.” Wissenschaftliche Zeitschrift
der Karl Marx Universitat 16 (1967): 47-55.
. “Die Identifikationen im Kausitaki-Brahmana,” Altorientalische For-
schungen 5 (1977): 237-44.
. “Die Identifikationen der Metren in der Literatur des Rgveda.” Wissen-
schaftliche Zeitschrift der Karl Marx Universitat 17 (1968): 267-73.
. “Die vedischen Identifikationen am Beispiel des Kausitaki-Brahmana.”
Klio 58 (1976): 145-66.
Nagaswamy, R. “Vedic Scholars in the Ancient Tamil Country as Gleaned from
the Sangam Classics and South Indian Inscriptions.” Vishveshvaranand
Indological Journal 3 (September 1965): 192-204.
Narahari, H. G. Atman in Pre-Upanisadic Vedic Literature. Adyar: Adyar Li-
brary, 1944.
Narten, Johanna. “Ved. amayati und dmaydavin-.” Studien zur Indologie und
Tranistik 5—6 (1980): 153-66.
Nehru, Jawaharlal. The Discovery of India. London: Meridian Books, 1960.
O’Flaherty, Wendy Doniger. Asceticism and Eroticism in the Mythology of Siva.
New York: Oxford University Press, 1973. Reissued as Siva: The Erotic
Ascetic, 1980.
. Dreams, Illusion and Other Realities. Chicago: University of Chicago
Press, 1983.
. The Origins of Evil in Hindu Mythology. Berkeley: University of Califor-
nia Press, 1976.
———. “The Origins of Heresy in Hindu Mythology.” History of Religions 10
(May 1971): 271-333.
. Other People’s Myths, New York: Macmillan, 1988.
. Tales of Sex and Violence: Folklore, Sacrifice, and Danger in the
Jaiminiya Brahmana. Chicago: University of Chicago Press, 1984.
. With Daniel Gold, David Haberman, and David Shulman. Textual
Sources for the Study of Hinduism. Manchester: Manchester University
Press, 1989.
. Women, Androgynes, and Other Mythical Beasts. Chicago: University
of Chicago Press, 1980.
O’Flaherty, Wendy Doniger, and J. D. M. Derrett, eds. The Concept of Duty in
South Asia. Columbia, Missouri: South Asia Books, 1978.
244 Bibliography
Oguibenine, B. L. Structure d’un mythe védique: Le Mythe cosmogonique dans

le Rg Veda. The Hague: Mouton, 1973.
—.. “Bandhu et daksinad: Deux termes védique illustrant le rapport entre le
signifiant et le signifié.” Journal asiatique 27 (1983): 263-75.
Oldenberg, Hermann. Ancient India: Its Language and Religion, ed. Calcutta:
Punthi Pustak, 1963.
. Vorwissenschaftliche Wissenschaft: Die Weltanschauung der Brahmana-
texte. GOttingen: Vandenhoeck and Ruprecht, 1919.
Olivelle, Patrick. “A Definition of World Renunciation.” Wiener Zeitschrift fiir
die Kunde Siidasiens 19 (1975): 75-83.
—.. “Odes of Renunciation.” Wiener Zeitschrift fiir die Kunde Stidasiens 20
(1976): 91-100.
Organ, Troy Wilson. The Hindu Quest for the Perfection of Man. Athens, Ohio:
Ohio University Press, 1970.
Pandey, Raj Bali. Hindu Samskaras: Socio-Religious Study of the Hindu Sacra-
ments, 2d ed. Delhi: Motilal Banarsidass, 1969.
Parpola, Asko. “The Pre-vedic Indian Background of the Srauta Rituals.” In
Frits Staal, ed., Agni: The Vedic Altar of Fire. Berkeley, Calif.: Asian
Humanities Press, 1983, II: 41-75.
———. “On the Symbol Concept of the Vedic Ritualists.” In H. Biezais, ed.,
Religious Symbols and Their Functions. Stockholm: Almquist and
Wiksell, 1979, pp. 139-53.
Pathak, V. S. “Vedic Rituals in Early Medieval Period.” Annals of the
Pillai, P. K. Narayana. Non-Rgvedic Mantras in the Marriage Ceremonies.
Trivandrum: Travancore Devaswom Board, 1958.
Puhakka, Kaisa. “The Roots of Religious Tolerance in Hinduism and Bud-
dhism.” Temenos 12 (1976): 50-61.
Raghavan, V. The Present Position of Vedic Recitation and Vedic Sakhas.
Kumbhakonam: Veda Dharma Paripalana Sabha, 1962.
Rai, Gangar Sagar. “Sakhas of the Atharvaveda.” Purana 14 (January 1972):
58-76.
. “Sakhas of the Rgveda as Mentioned in the Puranas.” Purana 6 (Janu-
ary 1964): 97-112.
. “Sakhas of the Samaveda in the Puranas.” Purana 8 (January 1966):
115-34.
———. “Sakhas of the Yajurveda in the Puranas.” Purana 7 (January—July
1965): 235-53.
. “Vedic Sakhas.” Purdna 15 (February 1973): 133-40.
Ramanujan, A. K. Speaking of Siva. Baltimore: Penguin Books, 1973.
Ranade, H. G. “Some Darsapirnamasa-Rites in the Satapatha Brahmana and in
the KatSS.” Bulletin of the Deccan College Research Institute 35 (March
1976): 121-26.
Rangacharya, V. “Sri-Vaisnavism in South India.” In H. Bhattacharya, ed., The
Bibliography 245
Cultural Heritage of India, 2d ed. 2 vols. Calcutta: Ramakrishna Mission

Institute of Culture, 1953, IV: 163-85.
Ravi Varma, L. A. “Rituals of Worship.” In H. Bhattacharya, ed., The Cultural
Heritage of India, 2d ed. 2 vols. Calcutta: Ramakrishna Mission Institute
of Culture, 1953, IV: 445-63.
Reddiar, N. Subbu. Religion and Philosophy of the Nalayiram. Tirupati: Sri
Venkateswara University Press, 1977.
Renou, Louis. “ ‘Connexion’ en védique, ‘cause’ en bouddhique.” In Dr. C.
Kunhan Raja Presentation Volume. Madras: Adyar Library, 1946, pp.
55-60.
. “Les Connexions entre le rituel et la grammaire en sanskrit.” Journal
asiatique 233 (1941-42): 105-65.
. “Les Débuts de la spéculation indienne.” Revue philosophique 143
(1953): 334-41.
. The Destiny of the Veda in India. Ed. and trans. by Dev Raj Chanana.
. Les Ecoles védiques et la formation de Veda. Cahiers de la Société
Asiatique, Vol. 9. Paris: Imprimerie nationale, 1947.
. Etudes védiques et paninéennes. 17 vols. Paris: Publication de V’ICI,
1955-69. j
. “Un Hymne a énigmes du Rgveda (1.152).” Journal de psychologie
normale et pathologique 46 (1949): 266-73.
. Hymnes spéculatifs de Veda. Paris: Gallimard, 1956.
. “La Maison védique.” Journal asiatique 231 (1939): 481-504.
———. “On the Word atman.” Vak 2 (1952): 151-57.
———. “Le Passage des Brahmana aux Upanisad.” Journal of the American
Oriental Society 73 (1953): 138-44.
. Religions of Ancient India. New York: Schocken Books, 1968.
. “Sur le genre du sitra dans la litterature sanskrite.” Journal asiatique
251 (1963): 165-216.
. “Sur la notion de brahman.” Journal asiatique 237 (1949): 7-46.
. Vedic India. Trans. by Philip Spratt. Delhi: Indological Book House,
197
———. “The Vedic Schools and the Epigraphy.” In V. Bandhu, ed., Siddha
Bharati IT. Hoshiarpur: Vishveshvaranand Research Institute, 1950, pp.
214-21.
—.. “Védique Nirrti.” Indian Linguistics 16 (November 1955): 11-15.
. Vocabulaire du rituel védique. Paris: Librairie C. Klincksieck, 1954.
Renou, Louis, and Lilian Silburn. “Nirukta and Anirukta in Vedic.” In J. N.
Agrawal and B. D. Shastri, eds., Sartipa-Bharati or the Homage of
Indology: The Dr. Lakshman Sarup Memorial Volume. Hoshiarpur:
Vishveshvaranand Institute Publications, 1954, pp. 68-79.
Rustagi, Urmila. Darsapirnamasa: A Comparative Ritualistic Study. Delhi:
Bharatiya Vidya Prakashan, 1981.
246 Bibliography
Sastri, V. A. R. “Arthavadas.” In J. N. Agrawal and B. D. Shastri, eds.,

Sartipa-Bharati or the Homage of Indology: The Dr. Lakshman Sarup
Memorial Volume. Hoshiarpur: Vishveshvaranand Institute Publications,
1954, pp. 165-70.
Schayer, Stanislov. “Die Struktur der magischen Weltanschauung nach dem
Atharva-Veda und den Brahmana-Texten.” Zeitschrift fiir Buddhismus 6
(1925): 259-99.
———. “Uber die Bedeutung des Wortes Upanisad.” Rocznik Orjentalistyczny
3 (1925): 57-67.
Schwab, J. Das altindische Thieropfer. Erlangen: Andreas Deichert, 1886.
Sen, Chitrabhanu. Dictionary of Vedic Rituals. Delhi: Concept Publishing, 1978.
Sharma, Arvind. “Some Misunderstandings of the Hindu Approach to Religious
Plurality.” Religion 8 (Autumn 1978): 133-54.
Sharma, B. R. “Symbolism of Fire Altar in Vedas.” Annals of the Bhandarkar
Sharma, D. S. “The Nature and History of Hinduism.” In Kenneth W. Morgan,
ed., The Religion of the Hindus. New York: Ronald Press, 1953, pp. 3—47.
. The Renaissance of Hinduism. Varanasi: Benares Hindu University,
1944.
Sharma, R. N. Culture and Civilization as Revealed in the Srautasiitras. Delhi:
Nag Publishers, 1977.
Shastri, Dakshina Ranjan. Origin and Development of the Rituals of Ancestor
Worship in India. Calcutta: Bookland, 1963.
Shende, N. J. Religion and Philosophy of the Atharva Veda. Poona: Bhandarkar
Oriental Research Institute, 1952.
Shivaganesha Murthy, R. S. A Study of the Important Brahmanas. Mysore:
Prasaranga, University of Mysore, 1974.
Silburn, Lilian. /nstant et cause: Le Discontinu dans le pensée philosophique de
l’Inde. Paris: Libraire philosophique J. Vrin, 1955.
Singer, Milton. “The Cultural Pattern of Indian Civilization: A Preliminary
Report of a Methodological Field Study.” Far Eastern Quarterly 15 (No-
vember 1955): 23-36.
Smith, Brian K. “Exorcising the Transcendent: Strategies for Defining Hindu-
ism and Religion.” History of Religions 27 (August 1987): 32-55.
———.. “Gods and Men in Vedic Ritualism: Toward a Hierarchy of Resem-
blance.” History of Religions 24 (May 1985): 291-307.
. “Ideals and Realities in Indian Religion.” Religious Studies Review 14
(January 1988): 1-9.
———.. “Myth, Religion, and the Real World.” The World & I (July 1987): 558-
67.
———. “Ritual, Knowledge, and Being: Initiation and Veda Study in Ancient
India.” Numen 33 (1986): 65-89.
——. “Sacrifice and Being: Prajapati’s Cosmic Emission and Its Conse-
quences.” Numen 32 (1985): 71-87.
Bibliography 247
——.. “The Unity of Ritual: The Place of the Domestic Sacrifice in Vedic
Ritualism.” Indo-Iranian Journal 28 (1985): 79-96.
. “Vedic Fieldwork” Religious Studies Review 11 (April 1985): 136-45.
Smith, Brian K., and Wendy Doniger O’Flaherty. “Sacrifice and Substitution:
Ritual Mystification and Mythical Demystification,” Numen (1989, forth-
coming).
Smith, Frederick Marcus. “The Avasathya Fire in the Vedic Ritual.” Adyar
Library Bulletin 46 (1982): 73-92.
. The Vedic Sacrifice in Transition: A Translation and Study of the
Trikandamandana of Bhaskara Misra. Ph.D. dissertation. University of
Pennsylvania, 1984.
Spear, Percival. India, Pakistan, and the West. London: Oxford University
Press, 1949.
Staal, J. F. (Frits). Advaita and Neoplatonism: A Critical Study in Comparative
Philosophy. Madras: University of Madras, 1961.
. Agni: The Vedic Ritual of the Fire Altar. 2 vols. Berkeley, Calif.: Asian
Humanities Press, 1983.
. “The Agnicayana Project.” In Frits Staal, ed., Agni: The Vedic Ritual of
the Fire Altar. 2 vols. Berkeley, Calif.: Asian Humanities Press, 1983, II:
456-75. .
—. “The Concept of Scripture in the Indian Tradition.” In Mark Juer-
gensmeyer and N. Gerald Barrier, eds., Sikh Studies: Comparative Per-
spectives on a Changing Tradition. Berkeley Religious Studies Series.
Berkeley, Calif.: Graduate Theological Union, 1979, pp. 121-24.
. “The Meaninglessness of Ritual.” Numen 26 (1979): 2-22.
—. “Ritual Syntax.” In M. Nagotomi et al., eds., Sanskrit and Indian Stud-
ies. Dordrecht: D. Reidel, 1979, pp. 119-42.
. The Science of Ritual. Poona: Bhandarkar Oriental Research Institute,
1982.
Tagore, Rabindranath. “Sacrifice.” In A Tagore Reader, edited by Amiya
Chakravarty. Boston: Beacon Press, 1966, pp. 125-48.
Tendulkar, D. G. Mahatma: Life of Mohandas Karamchand Gandhi, new ed. 8
vols. New Delhi: Publications Division, Ministry of Information and
Broadcasting, Govt. of India, 1961-69.
Thite, G. U. Medicine: Its Magico-Religious Aspects according to the Vedic and
Later Literature. Poona: Continental Prakashan, 1982.
. Sacrifice in the Brahmana-Texts. Poona: University of Poona, 1975.
Tsuji, Naoshiro. On the Relation between Brahmanas and Srautastitras. Toyo
Bunko Ronso seriesA, Vol. 33. Tokyo: Téyo Bunko, 1952.
Van Buitenen, J. A. B. “A Brief History of the Literatures of South Asia.” In
Edward C. Dimock et al., eds., Literatures of India: An Introduction.
Chicago: University of Chicago Press, 1974, pp. 14-20.
_____. “Hindu Sacred Literature.” Encyclopedia Britannica, 11th ed.
—__—. “Hinduism.” Encyclopedia Britannica, 11th ed.
248 Bibliography
. “On the Structure of the Sabhaparvan of the Mahabharata.” In J.

Ensink and P. Gaeffke, eds., India Maior: Congratulatory Volume Pre-
sented to J. Gonda. Leiden: E. J. Brill, 1972, pp. 68-84.
. The Pravargya: An Ancient Indian Iconic Ritual Described and Anno-
tated. Poona: Deccan College, 1968.
Verpoorten, Jean-Marie. “Unité et distinction dans les spéculations rituelles
védique.” Archiv fiir Begriffsgeschichte 21 (1977): 59-85.
Vira, Raghu. “Sakhas of the Yajurveda.” Journal of Vedic Studies 2 (April
1935): 61-77.
Vyas, R. T. “The Concept of Prajapati in Vedic Literature.” Bharatiya Vidya 38
(1978): 95-101.
Wasson, R. Gordon, and Wendy Doniger O’Flaherty. Soma: Divine Mushroom
of Immortality. New York: Harcourt, Brace and World, 1968.
Weber, Albrecht. The History of Indian Literature, 6th ed. Trans. from 2d
German ed. by J. Mann and T. Zachariae. Chowkhamba Sanskrit Series,
Studies, Vol. 8. Varanasi: Chowkhamba Sanskrit Series Office, 1961.
———. Uber die Konigsweihe, den Rajasuya. Berlin: Koniglichen Akademie der
Wissenschaften, 1893.
Willman-Grabowska, H. “Ekspiacja (prayascitti) w Brahmana (L’expiation
[prayascitti] dans les Brahmanas).” Bulletin international de l’Académie
Polonaise des Sciences et des Lettres 7-10 (July-December 1935): 237-42.
Winternitz, Maurice. A History of Indian Literature. 3 vols. Trans. by S. Ketkar.
New York: Russell and Russell, 1971.
—. “Notes on Sraddhas and Ancestral Worship among the Indo-European
Nations.” Wiener Zeitschrift fiir die Kunde morgenldndische 4 (1890):
199-212.
Witzel, Michael. “On Magical Thought in the Veda.” Leiden: Universitaire Pers
Leiden, 1979.
Young, Richard Fox. Resistant Hinduism: Sanskrit Sources on Anti-Christian
Apologetics in Early Nineteenth-Century India. Vienna: Institut fir
Indologie der Universitat Wien, 1981.
Younger, Paul. /ntroduction to Indian Religious Thought. Philadelphia: West-
minster Press, 1972.
Zaehner, R. C. Hinduism. New York: Oxford University Press, 1966.
Zysk, Kenneth. Religious Healing in the Veda. Philadelphia: American Philo-
sophical Society, 1986.
. “Towards the Notion of Health in the Vedic Phase of Indian Medicine.”
Zeitschrift der Deutschen Morganlandischen Gesellschaft 135 (1985): 312-
18.
Secondary Sources (General)
Burkert, Walter. Homo Necans: The Anthropology of Ancient Greek Sacrificial

Ritual and Myth. Trans. by Peter Bing. Berkeley: University of Califor-
nia Press, 1983.
Bibliography 249
Butchvarov, Panayot. Resemblance and Identity: An Examination of the Prob-

lem of Universals. Bloomington: Indiana University Press, 1966.
Capdeville, Gerard. “Substitution de victimes dans les sacrifices d’animaux a
Rome.” Mélanges d’archéologie et d'histoire de I’Ecole Francaise de Rome
83 (1971): 283-323.
Church, Ralph W. An Analysis of Resemblance. London: George Allen and
Unwin, 1952.
Denny, Frederick M., and Rodney L. Taylor, eds. The Holy Book in Compara-
tive Perspective. Columbia: University of South Carolina Press, 1985.
Eliade, Mircea. The Myth of the Eternal Return or, Cosmos and History. Trans.
by Willard Trask. Princeton, N.J.: Princeton University Press, 1954.
. Myth and Reality. Trans. by Willard Trask. New York: Harper Colo-
phon Books, 1963.
. The Sacred and the Profane. Trans. by Willard R. Trask. New York:
Harcourt, Brace and World, 1959.
Evans-Pritchard, E. E. “Lévy-Bruhl’s Theory of Primitive Mentality.” Bulletin
of the Faculty of Arts (Alexandria) 2 (1934): 1-16.
—. Nuer Religion. Oxford: Oxford University Press, 1956.
—.. Witchcraft, Oracles and Magic among the Azande. Oxford: Clarendon
Press, 1937.
Firth, Raymond. “Twins, Birds and Vegetables: Problems of Identification in
Primitive Religious Thought.” Man (new series) 1 (1966): 1-17.
Foucault, Michel. The Order of Things: An Archeology of the Human Sciences.
Trans. of Les Mots et les choses. New York: Vintage Books, 1973.
Freud, Sigmund. The Future of an Illusion. Ed. and trans. by James Strachey.
New York: W. W. Norton, 1961.
. “Obsessive Acts and Religious Practices.” In Character and Culture.
New York: Collier Books, 1963, pp. 17-26.
. Totem and Taboo. Trans. by James Strachey. New York: W. W. Nor-
ton, 1950.
Geertz, Clifford. Islam Observed: Religious Development in Morocco and Indo-
nesia. Chicago: University of Chicago Press, 1971.
——. “Religion as a Cultural System.” In The Interpretation of Cultures:
Selected Essays by Clifford Geertz. New York: Basic Books, 1973, pp.
87-125.
Girard, René. Violence and the Sacred. Trans. by P. Gregory. Baltimore: Johns
Hopkins University Press, 1977.
Hamerton-Kelly, Robert G., ed. Violent Origins: Walter Burkert, René Girard,
and Jonathan Z. Smith on Ritual Killing and Cultural Formation. Stan-
ford, Calif.: Stanford University Press, 1987.
Hayley, Audrey. “Symbolic Equations: The Ox and the Cucumber.” Man (new
series) 3 (1968): 262-72.
Horton, Robin, and Ruth Finnegan., eds. Modes of Thought: Essays on Think-
ing in Western and Non-Western Societies. London: Faber and Faber,
1973.
250 | Bibliography
Lévi-Strauss, Claude. The Savage Mind. Chicago: University of Chicago Press,

1966.
. Totemism. Trans. by Rodney Needham. Boston: Beacon Press, 1963.
Lévy-Bruhl, Lucien. How Natives Think. Trans. by Lilian A. Clare. Princeton,
N.J.: Princeton University Press, 1985.
Lovejoy, Arthur O. The Great Chain of Being: A Study of the History of an Idea.
Cambridge, Mass.: Harvard University Press, 1936.
Mauss, Marcel. A General Theory of Magic. Trans. by Robert Brain. New York:
W. W. Norton, 1972.
Montague, Ashley, ed. The Concept of the Primitive. New York: Free Press,
1968.
Moore, Sally F., Barbara G. and Meyerhoff, eds. Secular Ritual. Amsterdam:
Van Gorcum, 1977.
Neusner, Jacob. “Alike and Not Alike: A Grid for Comparison and Differentia-
tion.” In Jacob Neusner, ed., Take Judaism, for Example: Studies in the
Comparison of Religions. Chicago: University of Chicago Press, 1983,
pp. 227-35.
———.. “Judaism in the History of Religions.” In James S. Helfer, ed., On
Method in the History of Religions. Middleton, Conn.: Wesleyan Univer-
sity Press, 1968, pp. 31-45.
, ed. Take Judaism, for Example: Studies in the Comparison of Religions.
Chicago: University of Chicago Press, 1983.
O'Flaherty, Wendy Doniger, ed. The Critical Study of Sacred Texts. Berkeley
Religious Studies Series. Berkeley, Calif.: Graduate Theological Union,
1979.
Penner, Hans, and Edward Yonan. “Is a Science of Religion Possible?” Journal
of Religion 52 (1972): 107-33.
Proudfoot, Wayne. “Religion and Reduction.” Union Seminary Quarterly Re-
view 37 (Fall-Winter 1981-82): 13-25.
. Religious Experience. Berkeley: University of California Press, 1986.
Puhvel, Jean. “Victimal Hierarchies in Indo-European Animal Sacrifice.”
American Journal of Philology 99 (1978): 354-62.
Otto, Rudolph. The Idea of the Holy. Trans. by John W. Harvey. New York:
Oxford University Press, 1957.
Said, Edward. Orientalism. New York: Pantheon Books, 1978.
Sharpe, E. J. Comparative Religion: A History. London: Charles Scribner’s
Sons, 1975.
Skorupski, John. Symbol and Theory: A Philosophical Study of Theories of
Religion in Social Anthropology. Cambridge: Cambridge University
Press, 1976.
Smith, Jonathan Z. “Adde Parvum Parvo Magnus Acervus Erit.” History of
Religions 11 (August 1971): 67-90.
. “The Bare Facts of Ritual.” History of Religions 20 (Aug. and Nov.,
1980): 112-27; reprinted in Imagining Religion, pp. 53-65.
Bibliography 251
. “Fences and Neighbors: Some Contours of Early Judaism.” In Imagin-

ing Religion: From Babylon to Jonestown. Chicago: University of Chi-
cago Press, 1982, pp. 1-18.
. “I Am a Parrot (Red).” History of Religions 11 (May 1972): 391-413.
. Imagining Religion: From Babylon to Jonestown. Chicago: University of
Chicago Press, 1982.
. “No Need to Travel to the Indies: Judaism and the Study of Religion.”
In Jacob Neusner, ed., Take Judaism, for Example: Studies in the Com-
parison of Religions. Chicago: University of Chicago Press, 1983, pp.
215-26.
. “Sacred Persistence: Toward a Redescription of Canon.” In Imagining
Religion: From Babylon to Jonestown. Chicago: University of Chicago
Press, 1982, pp. 36-52.
Smith, Wilfred Cantwell. The Meaning and End of Religion: A Revolutionary
Approach to the Great Traditions. San Francisco: Harper and Row, 1962.
Tambiah, S. J. “Form and Meaning of Magical Acts: A Point of View.” In
Robin Horton and Ruth Finnegan, eds., Modes of Thought: Essays on
Thinking in Western and Non-Western Societies. London: Faber and Fa-
ber, 1973, pp. 199-229.
Van der Leeuw, Gerhard. Religion in Essence and Manifestation. 2 vols. Trans.
by J. E. Turner. London: Allen and Unwin, 1938; reprint Gloucester,
Mass.: Peter Smith, 1967.
Waardenburg, Jacques. Classical Approaches to the Study of Religion. 2 vols.
The Hague: Mouton, 1973.
Weber, Max. The Sociology of Religion. Trans. by Ephraim Fischoff. Boston:
Beacon Press, 1953.
Wilson, Bryan, ed. Rationality. Oxford: Basil Blackwell, 1977.
, oe
=
i tad Te)
1 es
hea A
‘toa. wh | é 7
vi ee a :
ote <A
ape Ht am ie veed, &
ee oewhe 1 eid oaiaialra ae wil "Fae
ees ora Catt setetn 3 a

PIR sei) ce
_ eps
-
et
ee:
=e D ol e
ai
aa
ieee cael
. :
P . Ph, he a Bey,» Ge beuys ae

wece
,
HN Cinta pay ett <aet! kent of snnlnln
te .
Ypx
i
epee
Se4 bm = . -
«
; >
el we (i
© e=4 vi
¥ a 250ethan
a7
tee 7
S"sot
7
troy ogee~ wa
eh new q x ’ ; 4 manele
7 ~ 4 = LG we ~-
iw
os y “i . sel a, ee
' A _. a
¥ 4 9 o
aw. ie)
oa a
j ye ~~. tive
rt “ “oe
‘
‘ .
Index and Glossary
abhirapa (appropriate form or adapta- all, 55-56, 58, 62, 68-69, 76, 204. See also
tion), 76-78 sarva; totality; visva.
acara (practice or custom), 140. See also altar, 38, 65, 118. See also Agnicayana
sadacara analogy, 129-30, 219. See also resem-
adhidevata (relating to the divine or macro- blance
cosmos), 46, 203, 224 numerical, 134
adhikara (competence), 83 ancestors (pitrs, fathers), 90, 116, 196-98,
adhiyajna (relating to the sacrifice), 46, 206
203, 224 anga (subsidiary rite), 127
adhyatman (relating to the self of micro- animals, 114, 163-66. See also cattle;
cosmos), 46, 203, 205, 224 pasu; sacrifice, animal; victim
adhvaryu (a type of priest), 162 anirukta vs. nirukta (undefined vs. de-
Agni (god of fire), 65, 79, 100, 132, 134, fined), 57, 64, 80
150, 177, 191-92. See also Agni-Soma; anrta (error or disorder), 17, 107
fire; Indra-Agni anthropogony, 51, 82, 86
Agnicayana (ritual of laying the bird- anthropology, 34, 86
shaped fire altar), 38, 65, 152, 188, anthropomorphism, 69
201. See also altar anurupa (corresponding form). See form
Agnihotra (daily morning and evening fire apad (emergency). See dharma
sacrifice), 65, 67, 106, 117, 152, 156- aparimita vs. parimita (unlimited vs. lim-
57, 166, 171, 185, 187, 190, 197 ited), 57, 80
Agni-Soma (dual gods of fire and soma), aphorism, 121-23
125-26, 134, 192 aranya (jungle or wilderness), 40
Agnistoma (the paradigmatic soma sacri- archetype vs. manifestation or particular,
fice), 64, 125-26, 131. See also soma 51, 124, 129, 218, 220. See also counter-
Agnyadhana/A gnyadheya (ritual of the part; prakrti; prama; prototype
first lighting of the fires), 67, 126, 156, art, works of (silpa), 77, 101
191 artha (purpose or meaning), 128-29, 131-
Ghavaniya (sacrificial fire principally used 33, 136, 173, 181-83, 189-90
for oblations to the gods). See fire arthavada (explanation), 138, 217
ahimsa (non-injury), 199, 211-14 Aryan, 87, 89, 93-94, 100, 154, 190n, 210.
Ghitagni (Srauta sacrificer), vs. ekagni, 151— See also society; twice-born
52, 157-60, 168, 171, 190-92, 199 ascension, difficult or dangerous, 54, 104-12
ascetic heat. See tapas body, 68, 75, 91, 101, 109, 114, 116, 205,
astrology, 121 2122213
Atharva Veda, 26n, 93 Bonazzoli, Giorgio, 28
atideSa (extended application by analogy), Brahma (the creator), 115
128, 134 brahmacarya (period of Vedic study),
atirikta vs. nyna (overdoing vs. defi- 207-8
ciency), 56, 77 brahman (the macrocosmic principle), 31,
atman (the microcosmic principle, self or 69-72, 76, 78, 194, 196, 198, 208, 213.
soul), 26, 31, 58, 60, 68, 82, 84, 101, See also atman
103, 117, 194, 206, 208, 214 power, 67, 96, 98
daiva (divine self), 83, 101, 103-5, 111- priest (overseer and healer of the sacri-
17. See also brahman; self fice), 151, 165
atmayajna (sacrifice to and in the self or Brahmanas (name of one class of Vedic
microcosmos). See self texts), 31, 47, 54, 68, 80, 96, 102, 110,
authority 113, 119-20, 137-38, 140, 144, 166,
of Brahmin. See Brahmin 180-81, 187, 194
of father, 174 absurd and random, 33, 35
of Veda, 10, 27, 217 Aitareya, 101, 166
avapa (insertion), 128 arbitrary, 33-34, 39
axial breakthrough, 41—42, 44 denigration of, 32, 39, 113
fanciful, 33-34, 39, 46
Jaiminiya, 52, 59n
bahutva (multiplicity), 52. See also many PancavimSa, 59n, 65n, 96n, 108n, 111n
vs. one; ndnatva vs. ekartipatva twaddle, theological, 32, 37-38
bali (rice ball) offerings, 148 SadvimSa, 187
bandhu (bond, connection), 31, 41, 72, 78, Satapatha, 31-33, 55n, 58n, 60, 64n, 70,
80, 105, 170, 194, 218, 224. See also 79n, 91, 93, 111, 115n, 166-67, 196,
connection 204, 207-8
Banerjea, A.C., 52 Taittiriya, 59n, 63n, 85n, 93n
banyan tree, 98 brahmavarcasas (luster or splendor of the
barley. See rice, and barley brahman power), 96, 98
Barth, A., 215 Brahmin (priestly class), 50-51, 67, 83,
Barua, B.M., 101n 87-88, 94-99, 104, 113-14, 117, 144,
bath, ritual, 149-50, 208. See also snataka 186, 211
battleground, sacrificial, 40-41, 214 and the Veda, 13
Baudhayana, 98, 120, 123, 149, 151, 154, authority of, 14, 69, 181, 195
158, 186, 192 authority of, as a component of Hindu-
bhakti, 11-12, 14, 16, 25-26, 205, 214 ism, 10-12, 23, 48, 217n
Bharadvaja, 132, 149, 154, 171 rejection of the authority of. See heresy;
Bharati, A., 18n heteroxdoxy
Bhattacharya, Kamaleswar, 215 breath, vital (prana), 58, 61, 65, 89-90,
Bhide, V.V., v 114, 176, 178, 190, 204-6, 209, 213
Biardeau, Madeleine, 170, 198-99, 202n British “creators” of Hinduism, 6
bird, 105 Brown, W. Norman, 9n, 201
birth, 51, 82-85, 111. See also Buddhism, 6, 9, 11-14, 16, 18, 120-21,
anthropogony; procreation 193, 195, 199, 205, 211-12
second, 102, 116. See also Aryan; Te- bull, 58n
incarnation; twice-born; upanayana Burkert, Walter, 42
Bodewitz, H. W., 190 butter, 74, 79, 150, 185, 192, 212
cake, sacrificial, 74-76, 132-34, 179-80, of Vedas, 26

192, 213n. See also purodasa condensed vs. expanded or extended, 151,
Caland, W., 133n, 200 169, 191
cannibalism, 59, 63. See also chaos; egali- conflict, 134-35, 183-84, 194. See also
tarianism competition
canon, 15, 28 confusion, 3, 33, 60-61, 133
Veda as, for Hinduism, 12-13, 21, 26 connection, 2, 26, 28, 30, 41, 46—47, 53,
caste, 10-11, 13-14, 16, 48, 118, 196, 224. 73, 78-79, 105, 218, 219n, 220. See
See also class; varna also bandhu; nidana
and Brahmins, 10-11, 14 consecration, 102, 105, 107
as definitional component of Hinduism, royal. See Rajasuya
10-11 construction, 30, 61-62, 76-77, 82-86,
non-Hindu, 11 220. See also deconstruction; healing;
rejection of the legitimacy of, 10-11 reconstruction; samdna
category, 33, 37-38 by rituals, 46, 64-65, 82, 92, 220-21
Hinduism, sacrifice and yajna as. See of self, 82, 86, 91, 101-2, 112
Hinduism; sacrifice; yajna of unity, 53, 111
cattle, 61, 163-64. See also cow continuity, 26, 201, 216, 222
Caturmasya (quarterly sacrifice), 66, 117, contradiction, 34, 39, 44, 123, 133, 136,
135, 152n, 161, 187, 191n 176
Chakrabarti, S.C., 77, 123, 126, 128-29, of the Vedas, 18, 25
131-32, 136, 183-84 controlled catastrophe, 41. See also
change and tradition. See tradi- Heesterman, Jan C.
tionalization and innovation Coomaraswamy, Ananda K., 63, 70n, 73,
chaos, 50, 54, 59-60, 66, 219. See also can- FOS 1035173
nibalism; confusion; disintegration; dis- coordination, 224
order; egalitarianism copulation. See procreation
chariot, 39, 67, 105—06, 190n correlation, 72, 190, 169, 198
Chaudhuri, Nirad C., 7n correspondence, 72, 219n. See also
children, 88, 90-91, 146-47 bandhu; connection; nidana
Christianity, 8, 11, 15 horizontal vs. vertical, 73-74, 77-78,
class, 46, 86, 94-99, 124-25, 127, 130-32, 220
134, 138-39. See also caste; society; Cosmic One or Cosmic Man. See One;
varna Purusa
Coburn, Thomas, 19n, 22 cosmogony, 34, 51, 53, 62-63, 68
comparison of religions, vii—vili, 35 cosmology, 34, 46, 50, 82, 204
competence. See adhikara cosmos, 50-51, 58, 82, 204. See also
competition, 40-42, 180. See also cosmogony; cosmology
Heesterman, Jan C. counterform (pratiriipa), 53, 68-69
complementarity, 145, 154, 168-69 counterpart. See pratima
completeness, 48, 55, 62, 70, 150, 154, vs. prototype. See prototype vs. counter-
160, 181, 187. See also all; totality part
complex vs. simple, 25, 169-70, 186-87, Courtright, Paul, 213n
215 cow, 63, 96, 152, 177-79, 185-86, 192. See
conception, 82-83, 86-87, 91. See also also cattle
imagination; procreation creation, 50—51, 54, 57, 82. See also
condensation, 83, 146, 151, 156, 169, 193, anthropogony; cosmogony
198-99 Creator, 54, 58. See also Brahma;
downward vs. upward, 188-90, 220 Prajapati
cremation, 90, 102, 117, 158-59 diksita (consecrated sacrificer), 40, 104
custom, 23. See also dcara; sadacara directions or quarters of space, 55, 90,
148, 153, 157, 204, 206
daksina (sacrificial fee or offering to the disintegration, 60-61, 65, 206. See also
priests). See fee chaos; differentiation; dissolution;
(southern fire). See fire prthak
darvihoma (one name for the domestic rit- disorder, 42, 107. See also chaos; confu-
ual), 162 sion
day and night, 55—S6, 61, 106, 204 dissolution, 61, 90-91. See also deconstruc-
death, 40-42, 70, 87, 90, 102, 111-16, 206. tion; disintegration
See also rebirth; sacrifice; violence divinity. See gods
of creator, 61 domestic ritual, 137-40, 143-60. See also
of father, 174 darvihoma; grhyayajna; pakayajna
incarnate as a god, 109 domestication
of patriarch, 149 of domestic ritual, 144
as sacrifice, 206 of Vedism, 195
of sacrificer or his wife, 157-59, 173 Doniger, Wendy, v, 8n, 19n, 20, 76n, 84n,
of Vedism, 200 118, 177n, 178n, 181n
debts, 198 dosa (blemish or fault), 181-82. See also
deconstruction, 6, 30n, 42, 87 error
deeds, good and evil. See evil; karma Dumont, Louis, 48, 210
deer, 96-97 Dumont, Paul-Emile, 190
defectiveness/deficiency, 50, 62, 67-69, durohana (difficult or dangerous ascen-
82-88, 91, 93, 107, 116, 161, 178-79. sion). See ascension, difficult
See also nyuina
definition, 3-5
of god, 57. earth vs. heaven, 91, 110, 154, 164, 167,
of Hinduism, 5—13, 201-2 204, 206
inchoate, 8—9 eating, 67. See also cannibalism; food; hun-
of religion, 3-4 ger
social and/or canonical, 9-10 of creatures by creator, 59-60, 64, 85
thematic, 8—9 ectype. See prakrti vs. vikrti
working, 4—5, 13, 26, 29 egalitarianism, 59, 224-25. See also canni-
deity. See gods balism; chaos
demon, 58, 67, 79, 166. See also evil Eggeling, Julius, 33
Derrett, J. D. M., 8 ekagni vs. ahitagni (having one fire, domes-
Deussen, Paul, 113, 118n tic sacrificer), 151, 158-59, 167, 190-
Devasvamin, 147 92
dharma, 23-24, 27, 213 ekarupa (uniform), 52, 55, 73
apad (emergency), 182 Eliade, Mircea, 29n, 50, 86, 210
as a definitional component of Hindu- Eliot, Charles, 9n, 10, 13
ism, 14 embryo, 84-85, 206
Dharma Sutra. See Sitra embryology, 82-83
differentiation emergency. See dharma, apad
excessive, 52, 59, 63, 85. See also ex- emission
cess; prthak cosmic, 51, 57-62, 66-67
insufficient, 59. See also jami; resem- of semen, 67, 83-84
blance encompassment, 48-49, 188
diksa (consecration of the sacrificer), 207 Encyclopedia Britannica, 6
Index and Glossary Pay
episteme, 47—48. See also Foucault, Mi- southern (daksina), 152-53, 205, 209
chel worldly (laukika), 147, 158, 185
epistemology, 34, 46, 204, 218-21 food, 42, 58, 60, 64n, 67, 89, 97, 99, 114,
equation, 31-32, 35, 37, 46—47, 170, 179- 163-64, 178, 204, 206-7
80, 186-89, 217, 219n, 220 form, 58n, 62-63, 74, 77, 97, 166, 177,
equivalence. See equation 179, 187. See also ruipa
error, 41-42, 45, 58, 107 and corresponding form (anurtipa), 112
essence, 61, 65,68—69, 74, 77, 83, 188-89 and counterform. See pratiriipa
of the Veda. See Veda Foucault, Michel, 47. See also episteme
evil, 69, 89, 114 foundation, firm, 58, 61, 65, 71, 87, 89,
excess, 51, 83-85, 124, 219 108, 110
of jami (nondifferentiation). See fragmentation, 59, 65-66
ekarupa; identity; jami; resemblance framework, ritual (tantra), 126-31, 133,
of prthak (differentiation). See differen- 139, 154, 162, 168, 191. See also prior
tiation; prthak vs. subsequent
exchange, 41-42, 194. See also reciprocity Freud, Sigmund, 173-74, 219
expiation, 90, 156, 199. See also healing fruit of the ritual, 212-15. See also artha
explanation. See arthavada
eye, as sun, 90, 206 Gandhi, Mohandas, 18n, 216
Geertz, Clifford, 4n, 203n
Falk, Maryla, 70 - general vs. specific or particular, 124, 132
“fallen,” 88-89, 91, 107-10. See also gift, 40-42, 50
patita-savitrikas Girard, René, 42, 174-76, 184. See also
father, 65, 84, 102, 150, 172-74, 177 violence
death of. See death; killing Gita, Bhagavad, 212-14
fecundity, 52, 58, 90, 156, 164—67 goat, 58n, 96, 177-79, 184-85, 212
fee, sacrificial (daksina), 104, 106, 117, Gobhila, 156
151-52, 177, 185-86, 208 Gobhilaputra, 147
female vs. male, 83, 92—93, 145, 155, 167, gods, 54-55, 65, 67, 70, 86, 90, 104, 107,
169 109, 118, 128, 134-35, 148, 150, 157,
feminization of ritual, 154—60 171, 176, 183, 187, 192, 196, 198, 204,
fetus. See embryo 206, 210
Filliozat, Jean, 46 human, 69, 104, 117, 154, 186
fire, 90, 206 world of, 115
cooking (pacanagni), 145, 147 Gonda, Jan, 19n, 58n, 62, 69, 71, 86n, 94,
domestic (grhyagni), 147, 152, 154-57, 103, 170, 201
170-71, 192 Gopal, Ram, 93, 120n
as mouth, 197, 205 grama (village), 40
for oblations to the gods (Ghavaniya), grhyayajria (one name for the domestic rit-
106, 116, 124, 152, 157, 205, 208-9 ual), 126, 140, 145-54, 197
householder (garhapatyagni), 106, 152— Grhya Sutra. See Sitra
54, 158-60, 205, 209. See also ahitagni;
ekagni Haradatta, Misra, 141, 147
marriage (vaivahika), 147, 154, 171 HarikeSa, a Jaina monk, 211
woman as, 205-7 havis (sacrificial offering), 65
of woman, 204 healing, 61, 63, 65-66, 87, 89-92, 101, 165
sacrificial (frauta), vs. domestic, 150- heaven (svarga), 103-13, 124, 154, 165,
54, 170-71. See also Srauta 204, 221
of Savitr (savitra), 117 heavenly self. See self
Heesterman, Jan C., 12n, 18n—19n, 20, 27- ideal, 42-45, 223-24. See also
28, 32, 39-46, 102, 111n, 148n, 152n, Heesterman, Jan C.; prama; transcen-
153, 156n, 175n, 180n, 194, 210-11 dence
hell, 113-14, 119 identity, 32, 58n, 72, 100, 194-95, 199, 219
heresy, 8, 18. See also heterodoxy mistaken, 30, 34
heterodoxy, 14, 18, 121, 193, 207, 211-12. vs. resemblance, 47, 53, 74, 81, 124,
See also Buddhism, Jainism 184, 186-87, 199, 220, 225. See also
hierarchy, 48—49, 78, 94-95, 100, 113, jami
117-19, 125, 146, 154, 165-67, 169, illness, 89-90. See also healing
177-78, 212, 224-25. See also inferior illusion (maya), 2, 5, 201, 219, 222, 225
vs. superior; resemblance, hierarchical image, 73-74, 77, 112, 199, 224
Hillebrandt, Alfred, 200 imago imaginans vs. imaginata (form vs.
Hiltebeitel, Alf, 213-14, 218n image), 76
Hinduism, 195-225 imagination, 5—6, 14, 28, 30, 32, 34, 36,
as category, 5, 7 43-46, 51, 203, 219, 222, 225
as defined by Hindus and non-Hindus, immanence. See transcendence
19-20 immortality, 57, 70, 103, 109, 112, 118
definition of, 5—13, 201-2, 216-17 impurity. See purity vs. impurity
as derived from Vedism, 5, 14—29, 49, Indra (warrior-king of the gods), 67, 89,
144-45, 195, 199, 216-17 99-100, 134-35, 192
historical origins of, 5 Indra-Agni, 134
as a “tolerant” religion, 6—8, 18 inference, 127, 133-35, 139
as a “vast sponge,” 8 inferior vs. superior, 60, 64, 149, 154, 166—
historians of religion, vii—viii, 44, 222-24 67, 169, 187, 210, 220. See also hierarchy
home vs. cosmos, 146-68. See also cos- information, 66
mos; householder initiation, 87, 92-101, 145, 149-50, 154,
homogeneity, 52 207. See also Aryan; twice-born;
homology, 31, 35, 45, 47, 53, 154, 211, upanayana
219n, 220. See also analogy; bandhu; injunction, 124, 132-34, 166n, 183. See
connection; correspondence; equation; also rule; vidhi
nidana innovation. See traditionalization and inno-
horizontal vs. vertical. See correspondence vation
horse, 58n, 63, 66-67, 177-79 inside vs. outside, 20, 27, 29, 217
sacrifice of, 215 intercourse, sexual. See procreation
householder, 40, 146-60, 196, 199. See interiorization, 194, 209-11
also grhyayajna. interpretation, 23, 39
Hubert, Henri. See Mauss Islam, 6, 15
humanism, 14, 19, 216, 225 isti (ritual in which the oblations are made
humans from vegetables), 67, 126, 133
sacrifice of. See sacrifice
vs. gods, 14, 21, 23, 45-46, 63, 67, 69, Jainism, 6, 9, 11-14, 16, 18, 120-21, 193,
82-87, 94, 107, 109, 164-65, 167, 187, 199, 211-12
196, 198, 205, 219-24 jami (excessive resemblance), 51-53, 59,
hunger, 58 63, 84, 124, 131, 186, 195. See also ex-
cess; identity; resemblance
Ida, goddess, 163-65 joint or juncture of day and night, or of
ida (rite in which the oblation material is the month, 61, 65. See also parvan
divided and consumed by the priests journey, 104-12. See also ascension, diffi-
and sacrificer), 162—65 cult
Kaelber, Walter, 107-8 mahayajna (“great” sacrifice), 160, 188,

Kalpa Sutra. See Sutra 196-99, 208, 215
kamya (optional ritual), 182 Malamoud, Charles, 197, 199
Kane, P. V., 92n, 152n, 197 male. See female vs. male; purusa
karma (law of), 118, 214 man. See potency; purusa
as a definitional component of Hindu- mantra (ritual utterance), 20, 77, 98-99,
ism, 9, 14 121, 139-40, 156, 167, 183-85, 197,
karman (ritual act), 51, 82, 86, 93, 102, 27
111-12, 115, 208, 212-13 Manu
Kashikar, C. G., 120n, 138, 149n, 200 as first man, survivor of the flood, 163—
Katyayana, 123, 129, 162 64
Keith, A. B., 113, 144 as mythical author of the Manu Smrti,
killing 17, 23, 93, 196, 198-99, 213n
of animals, 114, 211 many vs. one, 62-63, 68
of father, 174 marriage, 92, 101, 140, 145, 147-49, 154-
sin of, 40, 198—99, 211. See also ahimsa 55, vik
kimpurusa (defective man), 178 Mauss, Marcel, and Henri Hubert, 36,
king, 90, 110, 145, 204, 206, 212-14 104—5, 172-75
Kinsley, David, 12n, 18n maya. See illusion
Knipe, David, 90, 200 meaninglessness, 27n, 28. See also Staal,
ksatra (physical power or force), 67, 96-98 _ Frits
Ksatriya (class of warriors, rulers, and po- of meaninglessness, 38, 217, 222-23
liticos), 67, 88, 94-99, 165 of ritual, 38-40, 43-44, 46, 49, 201, 217
Kumarila, 18, 24 measurement, 75
medha (sacrificial quality), 178-79
metaphor, 46-47, 203
Laine, J., 6n, 15n
metaphysics, 34-35, 46, 51, 53, 59-61, 68-
legitimation, 11, 12, 14-16, 20-22, 26, 28-
69, 73
29, 42, 69, 181
metarule (paribhasa). See rule
Lévi, Sylvain, 32, 37, 46, 105n, 108n,
meters, 99
109n, 110, 173, 200
microcosmos. See macrocosmos vs. micro-
liberation (moksa), as a definitional com-
cosmos
ponent of Hinduism, 9, 14
milk, 106, 135, 165, 185, 191, 212
Lincoln, Bruce, 176n, 206
Mimamsa (the philosophy of the sacrifice)
Lingat, Robert, 21, 24, 27
as “orthodoxy par excellence,” 17-18
loka (world), 162-4, 112-14, 150. See also
Mimamsaka (philosopher of sacrifice) ,23—
heaven; hell; pitrloka; world
24, 91-92
luminous energy (tejas), 61, 65, 84, 98, 165
mind, 70-71, 90, 178, 206
minimalism, 187, 196-97, 199
Macdonell, Arthur, 33 model. See paradigm; prototype
machinations, priestly, 35 vs. copy, image, or variant, 73, 77, 199,
macrocosmos vs. microcosmos, 32, 46, 82, 203
85, 176-77, 224. See also adhidevata; monism, 32, 35, 103, 193-95. See also iden-
adhyatman; atman; brahman tity
madness, 33, 110 vs. resemblance, 210
magic, 79 moon, 90, 204, 206. See also New and Full
as pejorative term, 36-37 Moon
as synonym for ritual, 36, 45 mother, 93
Mahabharata (Sanskrit epic), 213-15 vs. father, 84, 102
Miller, F. Max, 32, 34n ontology, 46, 53, 82-86, 102, 221
multiplicity, 135. See also many vs. one; socio-ontological, 92-95, 99, 101, 112,
unity 119, 146, 154, 165
multivalence, 71, 78 order, 42, 50. See also chaos
Mus, Paul, 64, 66, 73-74, 80n, 112 orthodoxy
Mylius, Klaus, 47 Hindu, 17-18, 23, 211-12, 216-17
strategies for, 20-28, 216-18
namarupa. See name and form vs. heterodoxy. See heterodoxy
name. See form; name and form Otto, Rudolph, 4n
name and form, 59, 70, 76, 81, 86, 103 outcaste, 211
Nammialvar, 22, 26n
nandatva vs. ekarupatva (unrelated variety paka (“small,” condensed counterpart),
vs. uniformity), 52. See also bahutva; 146, 160-61, 169
many vs. one; multiplicity pakayajna (small sacrifice), 145-56, 160-
naraka (hell). See hell 68, 196-97
Nehru, Jawaharlal, 8n Pandey, Raj Bali, 86, 95—96
New and Full Moon nights and sacrifices paradigm, 50-51, 53, 86, 134
(darsapurnamasa), 55-56, 65, 67, 117, vs. variation, 124-27, 131-33, 135, 137,
125-26, 131, 134-35, 152n, 162, 171, 141-42, 146, 183-84, 219. See also
188, 190-91 model; prakrti vs. vikrti; prototype
nidana (connection), 73, 78-80, 218. See paradox, 20, 28, 36, 41, 43, 44, 80, 156n,
also analogy; bandhu; connection; 217. See also contradiction;
homology Heesterman, Jan C.
night. See day and night paribhasa (metarule). See rule
nitya (obligatory ritual), 168, 181-82, 189- Parpola, Asko, 37, 76, 200
92, 196 parvan (joint), 61. See also joint; moon
domestic vs. frauta, 192 pasu (sacrificial victim), 75-76, 177-79.
nonbeing vs. being, 53, 62 See also animals; cow; goat; horse;
noninjury/nonviolence. See ahimsd; vio- man; ram; sacrifice, animal; victim, sac-
lence rificial
nyagrodha. See banyan patita-savitrikas (those who have “fallen”
nytina (deficiency), 77. See also atirikta vs. from their right to learn the Veda and
nyuna; deficiency to sacrifice), 88-89, 91. See also “fallen”
patron, 40, 113. See also sacrificer;
oblation, 40, 42, 74, 104, 106, 117, 124, yajamana
134-36, 150, 156, 180-81, 183, 185, perfection, 43-45, 54, 70, 78, 80, 101, 107,
191, 193, 198, 206-9, 212-14, 216 110-11, 187-88. See also complete-
principal vs. subsidiary, 128-32 ness; defectiveness; error
obligatory vs. optional rituals. See kamya; performance, 45, 119, 138, 140, 171, 200-
nitya 201, 208. See also prayoga
obsession, v, 53 pitrloka (world of ancestors), 115. See also
O’Flaherty, Wendy Doniger. See Doniger ancestors; loka
ojas (physical vitality), 98 pitrs (ancestors). See ancestors
Oldenberg, Hermann, 35, 46, 76, 100 potency (virya), 178-79
Olivelle, Patrick, 210 Prajapati (Lord of Creatures), 51, 54-70,
One, Cosmic, or Cosmic Man, 32, 55, 61— 82, 93, 101, 110, 115, 150, 163, 176-77,
62, 64, 69, 72, 74, 80, 91, 101, 176-77, 185, 192, 204. See also brahman; One,
205. See also brahman; Prajapati; Cosmic; Purusa
Purusa prakrti vs. vikrti (archetype or paradigm
vs. variation of the paradigm), 124, quarters of space. See directions

127, 129, 146, 168-70, 218
prama (prototype), 73-4, 77, 218. See also Rajastiya (royal consecration), 110, 188.
counterpart; pratima; prototype See also consecration
prana (breath, life force). See breath ram, 177-79
pratima (counterpart), 73, 77, 176, 218. Ramayana (Sanskrit epic), 22
See also counterpart; prama; prototype Ramakrishna, 24
pratirupa (counterform), 76, 84, 186, 218. rasa (essence), 65, 83. See also essence
See also form rationalization, 27, 39
pratistha (firm foundation). See founda- reality, 50. See also ideal
tion, firm rebirth, 40, 87, 206. See also Aryan; initia-
prayoga (ritual manual), 45, 121n, 138, tion; transmigration; twice-born
140. See also performance reciprocity, 40-41, 44, 194
preclassical age, 40-44, 194. See also reconstruction, 41, 51, 66, 69, 74, 89, 91
Heesterman, Jan C. reduction, 3, 29
prelogical thinking, 34, 36. See also Staal, synecdochal or synecdochic, 188, 196,
Frits 220
preta (dead person, ghost or spirit), 90-91 redundancy, 51-52, 84-85. See also jami
“primitives,” 33 reduplication, 52, 205, 207, 217-18, 222,
principal vs. subsidiary, 127-35. See also 224
anga; principal rite excessive. See jami
principal rite (pradhana), 127 reflection, 29, 76, 224
prior vs. subsequent (parva vs. uttara), reform/reformation, 41—42, 47, 221-22,
125, 127, 141-42, 146, 166-68 224
procreation, 52, 62, 82-85 reincarnation. See transmigration
prototype vs. counterpart, 2, 29, 53, 72- reintegration, 66
74, 77, 80, 84, 107, 110, 115, 124-25, Renou, Louis, 5n, 18, 19n, 20, 27n, 31,
127, 155, 168-69, 176-77, 181, 186-89, 71-72, 79, 80, 122, 138, 151n
193, 218-22 See also prama; pratima renunciation, 32, 173-74, 194-95, 208-11,
Proudfoot, Wayne, 4n 213,225
prthak (excessively differentiated), 52-53, reproduction, 29. See also procreation
59, 63-64, 85, 124, 186. See also differ- resemblance, viii, 19, 29, 47, 68, 96, 112,
entiation; excess; jami 133-34, 219-20. See also saémanya
puja (ritual worship of Hindu gods), 201, excessive, 51-53, 63, 84. See also ex-
214-15, 218 cess, jami
Puranas (encyclopedic Sanskrit texts), 24— hierarchical, 48-49, 78, 119, 124-25,
26, 218 135-37, 146, 170, 184-85, 189, 193,
purity vs. impurity, 30, 40, 44, 48 220
purodasa (baked cake), 75. See also cake, universal, 47, 81, 194
sacrificial vs. identity. See identity; jami
purohita (type of priest), 212 restatement, 29
purpose (artha). See artha retas (semen), 67. See also seed
purusa (human man), 177-78, 203, 205. reversal, 29
See also patron; sacrificer; victim, sacri- Rg Veda, 26n, 35, 54, 71, 80, 91-93, 113,
ficial; yajamana 115n, 116, 177n
Purusa (Cosmic Man), 54-57, 62, 69-70, rice
176, 203-4. See also brahman; balls, 157, 197-98. See also bali
Prajapati; One, Cosmic and barley, 114, 124, 179, 185, 191, 207
purva. See prior vs. subsequent porridge, 165, 192
risk, 43-45 Sama Veda, 26n

rites of passage. See samskaras samana (a constructed unity of parts), 62—
ritual, 53, 63-65, 82 63, 80, 111
as distinct from sacrifice, 42—43 samanya (resemblance), 47, 124, 127, 133,
meaninglessness of. See meaningless- 136, 183. See also resemblance
ness; Staal, Frits of purpose. See artha
science of. See science of ritual Samhita (“collection,” a type of Vedic
ritualism, 34-38, 45—46, 195 text), 22, 31-32, 54, 138-39, 144, 166-
rgi (seer), 17, 23 67
rta (order or truth), 72. See also order; samsara (world of rebirth), as definitional
truth component of Hinduism, 9, 14
Rudra, 192 samskaras (transformative rituals that pu-
tule, 123, 217. See also injunction rify and form the body), 86, 90-92,
paribhasa (metarule), 123, 135—37, 180, 142, 155n. See also initiation;
218 upanayana; marriage
vidhi (ritual rule), 138 Sankara, 18
rapa (form), 58n, 63, 74, 76, 177. See also sannyasa. See renunciation
form; name and form Sarma, D. S., 22
sarva (all, perfect unity), 55-56, 62-63,
Sabara, 23-24, 91-92 69, 80, 165. See also all; vifva
sacrifice, 34, 63-65. See also mahdyajna; satya (the real, truth), 93, 107, 176, 185.
pakayajna; yajna See also rta; truth
animal, 74-75, 117, 125-26, 131, 135n, savitra (fire of Savitr). See fire
187-88, 191-92 savitri (RV verse 3.62.10, about Savitr, the
of battle, 214 sun), 99
as category, 202-3, 209-13, 216-19, 222 Sayana, 163
as counterpart of cosmos, 74 science of ritual, 34—36, 44, 125, 128, 131,
as distinguished from ritual, 42—43 136. See also Staal, Frits
identified with deity, 73 seasons, 55—56, 66, 94—96
identified with sacrificer. See sacrificer seed, 67. See also semen
obligatory. See nitya of a woman, 84
optional. See kamya self, 84, 91. See also adhyatman; dtman;
soma. See soma microcosmos
of self. See self; suicide construction of. See construction
to self, vs. to gods, 210 divine or heavenly. See adtman, daiva
as symbol, 173, 176, 199, 216. See also sacrifice as, in, and of, 103, 111n, 173,
symbol 209
vegetable. See isti semantics, 49
vegetarian version of animal, 179-80, semen, 83-84, 91, 206. See also emission;
192 retas; seed
sacrificer, 78, 167, 183, 210. See also pa- separation. See prthak
tron; yajamana sheep, 58n
death of. See death ship, sacrifice as, 106
as sacrifice, 74-75, 79, 103, 116, 173, Sikhs, 13-14, 16
177, 190 Silburn, Lilian, 31, 50-51, 53, 66, 80, 86
as victim, 74-75, 79, 105, 172-73 simple. See complex vs. simple
sadacara (custom of good men), 23-24. sin. See evil
See also aGcara; custom Sista (learned man), 23-24. See also
sam (prefix), 63, 86, 91, 101 sadacara
Siva, 214-15. See also Rudra substitution, 75, 111, 139, 170, 172-86,
Skorupski, John, 36 211
Smith, Frederick, 180n, 183n, 200 double, 172-75
Smith, Jonathan Z., 28, 46n Sidra (servant class), 22, 25, 67, 83, 87-
Smith, Wilfred Cantwell, Sn, 6n, 14n 89, 91, 94-99, 152
smrti (tradition originating in a human suicide, 11, 173, 175. See also self, sacri-
teacher), 17, 21-23, 137, 140 fice of
snataka (one who has had his graduation sun, 90, 204, 206. See also eye, as sun
bath), 149-50. See also bath superiority. See inferior vs. superior
social system, 10. See also Aryan; caste; Sitra (“thread,” name of one class of
society Vedic texts), 31, 47, 91, 93, 99, 119-27
society, 40, 42-43, 46, 87, 89, 93, 94, 97, Dharma, 17, 120
101. See also ontology, socio- Grhya, 98-99, 120, 137, 139-42
ontological Kalpa, 120, 137, 142
socio-ontology. See ontology Srauta, 49, 95, 120, 123, 137-39, 141ff.
soma (a sacred plant and the juice ex- See also Baudhayana; Bharadvaja;
tracted from it; also incarnate in a Gobhila; Katyayana
god), 77, 98, 135, 204, 206 svaha! (exclamation made while making
sacrifice, 64, 67, 76, 90, 118, 125, 143, oblation) (vs. vasat/), 79, 162, 196
165, 187, 204, 207 svarga (heaven), 103-13. See also heaven;
sacrifice, without, 180-81, 192, 215 loka
soteriology, 46 symbol, 27, 32, 36-37, 51, 76
soul (atman), 26. See also atman; self sacrifice as. See sacrifice
space, 55—S6, 91. See also directions synecdoche. See reduction, synecdochal
Spear, Percival, 8, 17n syntax, 49
specific. See general vs. specific
speech or voice, 71, 79n, 90, 178, 197, Tagore, Rabindranath, 216
205-6, 208-9. See also Vac Tamil
Srauta (of the Sruti, name of rituals not texts, 26n
classified as domestic). See fire; sacri- Veda, 22
fice; Sutra tantra (framework of the ritual). See frame-
vs. domestic, 150-54, 157 work, ritual
Srauta Sutra. See Sitra Tantra (esoteric text), 25
Srotriya (one who is learned in the Sruti or Tantricism, 11, 14, 16, 216
Veda), 156 tapas (ascetic heat), 57, 212
$ruti (revelation, Veda), 17, 21-22, 138, as sacrifice, 210
140-41, 181, 183 tejas (luminous energy). See luminous energy
srsti (creation, emission), 57, 64-65. See Thite, G. U., 33, 166n, 188n
also creation; emission thread, sacred. See Sutra; yajnopavita
Staal, Frits, 27n, 34-35, 38-40, 43, 46, 49, tiger, 96-97
102, 152n, 166n. See also meaningless- time, 55-6, 61. See also day and night;
ness joint; parvan; seasons; New and Full
status, 88, 91, 103, 112-13, 119, 180. See Moon; year
also loka degeneration of, 24. See also yuga
strategies for orthodoxy, 20-29, 211. See tolerance, as definitional component of
also orthodoxy Hinduism. See Hinduism
subsequent. See prior vs. subsequent totality, 55, 62, 64, 66, 70, 165. See also
subsidiary rite (anga). See anga; principal all; completeness; perfection; sarva;
rite visva
totemism, 173-74 van Buitenen, J. A. B., 6-8

traditionalization and innovation, 20-21, variation. See paradigm vs. variation;
26, 202, 211, 216-17 prakrti vs. vikrti
transcendence, 19, 21, 27-28, 42-43, 57, varna (class), 58n, 99. See also Brahmin;
70-71, 73-74, 76, 80, 110, 112, 116- caste; class, Ksatriya; Sidra; Vaisya
17, 176, 181, 222-23. See also ideal; Vayu (god of the wind), 58n
prama Veda, 137-38, 202. See also Atharva
transformation, 14, 19-20, 26, 85, 92, 102, Veda; Rg Veda; Sama Veda; Sruti;
104, 219, 221, 224 Yajur Veda
transmigration authority of, 14
as a definitional component of Hindu- authority of, as definitional component
ism, 9, 118n of Hinduism, 10, 14, 216-17
of the Vedic sacrifice, 193, 202 and Brahmins, 12-14
trees, 91, 98-99, 110, 114, 206-7 contradiction of. See contradiction
truth, 208. See also rta; satya essence of, 24-26. See also essence
Tvastr (the divine builder and shaper of etymology of, 19n
forms), 85 fifth, 22
twaddle, theological. See Brahmanas as Hindu canon, 12
twice-born, 88, 93. See also Aryan; initia- lost, 24
tion; upanayana shrinking, 25
tyaga (renunciation of the fruits of the of- simplification of, 25. See also complex
fering, ritual, or act), 213. See also re- vs. simple
nunciation as source of legitimation, 14. See also
legitimation
as symbol, 27—28. See also symbol
uncertainty. See risk three, 57
undifferentiation. See identity Vedanga (limb of the Veda, a subsidiary
unification, 53, 60-61, 68-69 branch of the Veda), 35, 121, 138
uniformity, 52, 55, 59, 73. See also Vedanta (“end of the Veda,” a philosophi-
ekarupa cal school), 26n, 193, 195, 199, 225
unity, 55, 69, 74, 80, 111, 201-2, 225 Advaita (non-dual Vedanta), 18. See
of ritual, 143, 145-46, 167, 169-70, 189, also monism
193. See also samana Vedism, 5, 30, 45, 49, 137, and passim
unlimitedness, 56, 80. See also anirukta vs. Verpoorten, Jean-Marie, 51, 62, 86, 132n
nirukta; aparimita va. parimita vertical. See correspondences
Untouchable, 48 vi (prefix), 59, 63
upanayana (ritual of initiation into Veda victim, sacrificial, 41, 64n, 172-86. See
study and sacrifice), 87, 92-101, 207. also animals; cattle; cow; goat
See also Aryan; consecration; initia- animal, 79n. See also pasu
tion; samskdras; twice-born human, 177-8, 188. See also pasu;
Upanishads (type of Vedic text), 21-22, purusa
31-32, 35, 71-72, 106n, 114-15, 118, identified with sacrificer. See sacrificer
176, 193-95, 199, 204, 206-10, 225 identified with deity, 172
uttara (subsequent). See prior vs. subse- surrogate, 174-75. See also substitution
quent vidhi (ritual rule). See injunction; rule
vihara (the sacrificial grounds of the Srauta
Vac (goddess of speech), 71, 79n ritual), 151
Vaisya (commoner class), 67, 88, 94-99, vikara (modification), 127, 133-34
155, 165 vikrti (modified sacrifice or ectype, varia-
tion of the paradigm). See prakrti vs. Witzel, Michael, 37, 200
vikrti womb, 83-86, 94, 101, 116, 205
violence, 49-42, 153n, 173-74, 198. See world, 55—56, 102-4, 154, 171. See also
also ahimsa earth vs. heaven; heaven; hell; Joka
virya (potency). See potency worlds, three, 57
virodha (contradiction), 133. See also con-
tradiction
yajamana (sacrificial patron), 40, 78. See
Visnu, 214
also patron; purusa; sacrificer
visva (all, defective totality), 62-63. See
yajna (sacrifice), 34, 154, 199-203, 209,
also sarva
213, 216
Visvakarman (architect of the gods), 57
as canonical category, 216-19, 222. See
Vivekananda, 27n
also sacrifice
vratya (people beyond the Aryan pale),
Yajnavalkya, 166
88-89, 91
yajnopavita (sacred thread), 96
Vrtra (demon slain by Indra), 89-90
Yajur Veda, 26n
Black, 120
warrior, 40—42
White, 166
waxing and waning lunar half-months. See
Yama (god of the dead), 116
New and Full Moon; parvan; moon
year, 31, 55-56, 61, 65, 73-75, 106, 204,
Weber, Max, 3n
206
whole. See all; samndana; sarva; totality;
yoga (discipline or technique), 210, 213-
visva
14
widow, 159
Younger, Paul, 15n, 20, 27
wife, 97, 146-60, 167. See also female vs.
yuga (age or aeon), 24
male; marriage; mother
wild game, 63
wind, 90. See also Vayu Zaehner, R. C., 10, 13n
ii £
4
Meu AY
THEOLOGY LIBRARY
CLAREMONT, CALIF.
epee:
oat ;
ears. a < * ; ae a J
7 uring jun Ae cle ae

a a> eer ns aiilien
e
aa
a
eae y
ae = A en
eit
ere" ppt a . fo ars rte
>. anos ’ Ditonto eee
, 5 en Bg a ‘ . A “~~ +
va neaate ; - i a WE: |,
VS peyGe mae Seer: : : a
mas sa pated a. é ;
are) ae Nl tu a lay
a ao > > 7 - By | , ~*~
ln Mi
el ocd sie apa 7-9 : 1G : ‘
1s tw oy a 7 ) at _
af :
=o4p- we oe - Aw ’ te i) ,
fae
mt 2 : an, v7;
We ne ee a eer? tendon, me
oe 1) RP ef "bs stat
wl? . and ae
po Avewe =the BG ."
a "9 4 wt
& =O of. © ~~ 7 sh Segal e_p 41

ea 7 ae ‘fhe |
’ = eke hg ees) , jreweS.
Te re we 4 a +ae, or ee
= ai ai eae 7)
Te ’ o<- | -
~~ oo, oe ee ee
ee ae aie
e wipe 9 q ae
= Phe mar") 42 « ie -
g 7 a ta, yet . a
7 J tee, 1% 7
» _ ge “ = . wy ’
ar ~~ oe, ©
: ~-'> 2" a: aa. c 7
25 j das ’ (i 7 SO jeg :
Z -_ ;
! it ") Ob dais Ps eyne Gael i
a ivt ety 3D 5
sa PSs 08) hes te = alley Be

lagen wD — =
Hee
GELEE.7,
oo lateg ne at
a (Coe F hae a _
’ Pastas 7
STaP Yr Sirs a
226012 Smith, Brian Key 1953-
S65 Reflections on resemblance, ritual
989 and religion / Brian Ke Smithe -—— Ne«
York : Oxford University Press, 1985
Xiv, 265 pe 3 CMe
Bibliography: pe [ 226]-251.
Includes indexe
ISBN 0-19-505545-4 (alke paper)
ie Hinduism--Ritvals. 2e Hinduiem
DoctrinesSe Ie Title
4444599 )
tides
M
rll
oie

Untitled

Uploaded by

Copyright:

Available Formats

Untitled

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Untitled

Uploaded by

Copyright:

Available Formats

Reflections on

New York Oxford

Oxford University Press

Copyright © 1989 by Oxford University Press, Inc.

Library of Congress Cataloging-in-Publication Data

Printing (last digit): 987654321

And for Wendy, who made it all more-ish.

Scholars of the discipline alternately labeled Comparative Religion or

of Indian Studies for timely grants that helped considerably in bringing

New York and Chicago B.K.S.

Heap. eee ent s-omrea hued Chee MS feyrenee

- . elk @a in pee AP aly ovale do-ens, ariel aa

: a 2 (4g5 ‘<Pat ape a -tdauiag Gv

a = m ». iM pee ok pal eke

1 Making Connections: Hinduism and Vedism 3

3 Ritual and Reality 50

4 The Ritual Construction of Being 82

> The Organization of Ritual Knowledge 120

6 The Organization of Ritual Practice 143

Zz Ritual Hierarchy, Substitution, and Equivalency 169

8 The Destiny of Vedism 200

Index and Glossary 253

AgGS Agnivesya Grhya Sitra

BAU Brhadaranyaka Upanisad

ChU Chandogya Upanisad

GautDhS Gautama Dharma Sutra

HGS Hiranyakesin Grhya Sutra

LSS Latyayana Srauta Sutra

Manu Manu Snnrti

SadB Sadvimsa Brahmana

VaikhGS Vaikhanasa Grhya Sutra

Yaj Yajnavalkya Smrti

What is called a sincere work is one that is endowed with enough

are regarded as exempt from the shackles of defining strictures.* Here,

mally productive to envision definitions as the exclusive aim of one’s

Like most (and perhaps all) particular religions, “Hinduism” as a unitary

5. See, however, J. Laine, “The Notion of ‘Scripture’ in Modern Indian Thought,”

be a Hindu. . . . Few religious ideas are considered to be finally irreconcil-

8. Van Buitenen, “Hinduism,” pp. 888-89.

be classified a Hindu under this definition, and, remarkably, “a Hindu

definition. One principal form of this type of definition of Hinduism

15. See, e.g., L. D. Barnett’s presentation of Hinduism identified by three concep-

as much a social system as a religion. . . . Its social framework has from

20. Eliot, Hinduism and Buddhism, I: xvii, 34.

Recognition of the authority of the Brahmins, then, is not a suffi-

tional bearers, purveyors, interpreters, and protectors of the Veda; the

Having reviewed the analytically separable (but in actuality usually con-

perpetuate, and transform traditions with legitimizing reference to the

Notion of ‘Scripture’ in Modern Indian Thought”; and Robert C. Lester, “Hinduism:

Third, this definition of Hinduism is neither too inclusive nor too

Hindu scholastics for the past several millennia have consistently

So it is that a definition such as the one proposed here that centers

tions with legitimizing reference to the authority of the Veda. Hinduism

The Veda and Strategies for Orthodoxy

The great paradox of Hinduism, defined in this manner, is that although

48. For an overview, and overgeneralization, of the Hindu devaluation of historical

addition to texts called Upanishads, some post-Vedic texts designate

58. D. S. Sarma, “The Nature and History of Hinduism,” in Kenneth W. Morgan,

customs of learned Brahmins), then, no less than the second, is wholly

Such a personalized conception of Vedic authority continues in recent