Review: A Survey of Tantric Hinduism for the Historian of Religions

Author(s): André Padoux

Reviewed work(s):
Hindu Tantrism by Sanjukta Gupta;Dirk Jan Hoens;Teun Goudriaan
Source: History of Religions, Vol. 20, No. 4 (May, 1981), pp. 345-360
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/1062460
Accessed: 01/09/2009 13:43

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to History
of Religions.

http://www.jstor.org
 REVIEW ARTICLE

A SURVEY OF TANTRIC HINDUISM FOR THE HISTORIAN OF RELIGIONS

Hindu Tantrism. By SANJUKTA
GUPTA,DIRK JANHOENS,and TEUN GOUD-
RIAAN.Handbuchder Orientalistik,2. Abteilung, Band 4, 2. Abschnitt.
Leiden:E. J. Brill, 1979.
As its title indicates, this volume, written by a team of scholars
from the University of Utrecht, is not an exhaustive treatment of all
of the Tantric forms taken by the religions of India over the course
of the ages, but rather a study of only one of them-Hindu Tantrism
(or should one not more correctly say, Tantric Hinduism?). Thus, it
is a limited endeavor. One could hardly do more in 185 pages and it
is indeed as an "introductory survey of Hindu Tantrism," a "general
but scholarly introduction to the field," that the authors present
their work, specifying that it is intended more for historians of
religions than for Indologists. It is intended even less for specialists
of Tantrism, who would find in it only few things that they do not
already know, and who would probably disagree on several points.
On the other hand, Hindu Tantrism would offer them the always
useful opportunity to reflect on a subject that is far from simple.
One should begin, however, by saying that this book deserves to
be marked favorably and recommended to the readers to whom it is
addressed-namely, nonspecialists who desire information on Tan-
trism or specialists who would like an aide-m6moire or a reference
book. It was not an easy task to offer in a few pages an objective,
sober, sufficiently documented, and generally exact presentation of
so complex a theme and controversial a subject as Tantrism. Proof of
this lies in the fact that such a study has never before been written
? 1981by The Universityof Chicago.0018-2710/81/2004-0003$01.00
346 Tantric Hinduism
in a truly satisfying manner either in the West or in India.1 Therefore,
let us be grateful to the authors for their often excellent work, even
if the specialist, who has his own biases and prejudices, might find
some faults in it, and disagree on the manner in which the field of
study has been defined as well as on some specific items and technical
details.
It must be noted that this work is based almost exclusively on the
Sanskrit tradition, with only a few references to sources, mostly
secondary, in Hindi and Bengali. It does not deal with Prakrit and
Apabhramsa literatures, nor does it mention sources in modern
Indo-Aryan languages which contain many interesting points for the
study of Tantrism in the Middle Ages and in modern times. Equally
excluded are sources in Dravidian languages, which contain a con-
siderable, and as yet little-known, literature.2 Doubtless, one can
admit that "the Sanskrit sources are by far the most important"
(p. 10). Certainly also, Tantric literature as we know it is very repeti-
tive, and the practices and ideas appear to be very similar throughout
India. A broader textual basis would therefore perhaps not appre-
ciably modify the picture. Yet, a question remains: Is not this
relative uniformity, this preeminence of Sanskrit, the result of our
ignorance of the totality of the Tantric domain? It goes without
saying that I am not taking the authors to task for their Sanskrit
bias, which can be justified by several reasons and which was
probably inevitable in such a book: I only want to outline the exact
scope of the work. Moreover, Medieval texts have been at least
mentioned, if not studied, in another volume of the Handbuch,3of
which still another is planned which, I believe, will treat Tamil and
other sources. This gives me an opportunity to say once more that
in view of the extent, the variety of sources, and the many aspects
of the subject, a serious and thorough study of Tantrism (a study
very much to be wished for) can only be carried out through the
efforts of an international collaboration of interested scholars. But
this is by the way. I would add that, considering the present state of
research, and the limits to which they were bound, the authors are
perfectly correct when they write that "in the meantime, any
survey of Tantrism, including the present one, can have only a
preliminary character" (p. 4).
1Several studies are cited in the bibliography of the book. T. Goudriaan's
observations in the introduction (pp. 4-5) on this subject are quite judicious.
2
There are a number of Tamil texts on the siddhasand an abundant literature
written in Kannada.
3 V. B. Mishra, Religious Beliefs and Practices of North India during the Early
MediaevalPeriod (Leiden: E. J. Brill, 1973). The Sanskrit texts, rather than the
practices and beliefs, are examined in the series A History of Indian Literature,
published by Otto Harrassowitzunder the direction of Jan Gonda. In this collec-
tion also: Medieval Religious Literature in Sanskrit, by Gonda (Wiesbaden, 1977).
A volume on the Puranas and one on "Tantrist literature," written by Sanjukta
Gupta and Teun Goudriaan,are announced.
 History of Religions 347
Having thus limited themselves to a Sanskrit base, the authors
specify how they want to define for the needs of their study the
terms "Tantrism" and "Tantric," and here it becomes rather difficult
to agree with them.
No doubt, we may agree with T. Goudriaan's opinion that "the
sakta texts are the Tantras par excellence" (p. 11) in the sense that
they extol the essential element of Tantra, namely, "power" or
"energy" (sakti). But to posit as separate entities on the one hand
Saktism, which would be "a world-view oriented towards the sakti,"
and, on the other, Tantrism, taken as "a conglomerate of ritual and
yogic practices and presuppositions" (p. 7), is, I believe, to establish
an unjustified distinction (or, at least, too marked a distinction, and
an illogical one at that if the sakta texts are to be held as the "Tantras
par excellence") between two elements, or two series of elements,
which appear usually in combination. I say this first and foremost
because, as far as I know, there are no daktatexts that are not Tantric.
One always find in them the particular traits which we consider as
characteristic of Tantrism. Might there be, on the other hand, a
non-sdkta Tantrism? One often adduces in this connection the case
of the Nathas or the Siddhas,4 who are predominantly yogic in char-
acter, but in whose texts (which are still little known) the role of
power or sakti seems to remain considerable, even essential. The
category of nonsakta Tantrism appears as preeminently applicable
to Vaisnava Tantrism, whether of the Paficaratra or of more recent
groups. But as a whole, sakta ideology is, in my opinion, funda-
mentally inherent to Tantrism, even if there exists sects for whom
the role of sakti is sufficiently nonprominent (although still notable
in doctrine and practice) for these to be considered nonsdkta.
The difficulty in defining the relationship between Tantrism and
Saktism is in fact due largely to the imprecision of these two notions
which we inherited from a time when the Tantric texts were as yet
little known, and when we had not yet realized the extent of the
Tantric phenomenon. Further knowledge and the resulting fuller
grasp of the two notions of Tantrism and Saktism will help clear the
confusion. It may even make the problem disappear. In any case, it
seems to me for the moment that we can refuse to subscribe to
Payne's formula, cited with approval in the book (p. 6), which
describes Saktism and Tantrism as "two intersecting but not coincid-
ing circles." These circles, I feel, do not intersect: They overlap
partially, the Tantric circle (at least as we currently understand it)
being the more extended and including the other-all this, of course,
according to the present state of our knowledge.

4 See
Kalyani Mallik (Siddha-Siddhdnta-Paddhati and Other Works of Nath
Yogis [Poona, 1954]) and A. K. Banerjea (Philosophy of Gorakhnath [Gorakhpur,
n.d.]); see also Gonda (Medieval Religious Literature in Sanskrit).
348 Tantric Hinduism

Regarding, however, the delimitation of the scope of this study,

there is another point to which I am more bound to voice my dis-
agreement. On page nine, after having enumerated eighteen con-
stituents of Tantrism (in the broad sense of the term)-a thoroughly
valid enumeration-T. Goudriaan adds: "As for Hinduism in
general, it might be proposed for Tantrism also that Tantrics are
those who are wont to call themselves by that name (or who, if
questioned, admit themselves to be Tantrics). In the sequel, we shall
concentrate upon the sakta variety of Tantrism and return only
shortly and intermittently to other varieties (Kashmir Saivism,
Saiva Siddhanta, Paficaratra, Jaina Tantrism)...."
Now, everyone knows that Tantric texts-those which we consider,
but with good reason I think, to be Tantric-do not in general,
define themselves as such. They do not explicitely claim to be
Tantric. At the very most they refer to the Tantras as containing
their doctrine, when they refer, for example, to other texts with the
formula tantrdntare ("in another tantra"), or when they include at
the beginning (notably at the beginning of the Saiva agamas) a
section called the tantrdvatdra,which describes the "descent" of the
text. Thus, one limits oneself too much when one excludes all texts
that do not declare themselves Tantric. But T. Goudriaan goes still
further. He tells us that tantrikas are "those [people] who are wont
to call themselves by that name." Who could this be, in fact, since
the dead cannot be made to speak? Are they the people in India
today who call themselves tantrikas? (Such is the criterion followed
by Gupta in her description of the puja.) But, as is well known, in
India today Tantrism has a bad reputation. In common speech, a
tantrika is a sort of magician. Let us also record all that has been
written in India in recent years on Anand Marg. Many practicing
Hindus who attend rites in the temples (rites which to an impartial
observer are of tantric character, or include tantric elements) would
refuse indignantly to admit the Tantric nature of this activity. Only
certain groups, notably those of the sdkta circles of northern India
and Bengal, consciously desire to be called Tantric. Thus, our authors
are led to give in their description a privileged position to these
small groups and to the texts of their tradition, while at the same
time excluding from their study the largest portion, one can say, of
Indian Tantric texts and traditions as they have been expressed for
more than a thousand years.
In limiting the sphere of investigation, the authors clearly have
the advantage of pursuing a more careful study. They avoid having
to embrace an immense and fluid subject matter which, within the
limits of this Handbuch, would have obliged them either to adopt too
general an approach, or to merely list a series of facts. From a
methodological point of view, however, this is very questionable.
First, because it reduces too much the field of study. Second, and
History of Religions 349

more seriously, because it makes the definition of that field of study

dependent on contemporary notions (or, rather, prejudices). Now,
that field rests on texts ranging from, say, the fifth to the eighteenth
centuries A.D.; it therefore belongs to "classical" Indology, not to the
religious sociology of present-day India. Contemporary criteria
should, thus, not be applied to it. Apparently, the authors were also
aware of this problem, as is made clear in the book's preface. And in
the body of the book, they do not keep consistently to that position.
The first part (which is the best), "Introduction, History, and
Philosophy," in fact appeals to all, or almost all, of the Sanskrit
sources. It even refers to some Dravidian sources (e.g., p. 23). The
second part makes reference to Kashmiri texts. Only the chapter on
puja holds to the northern and Bengali sdkta perspective, which is
regrettable, because this well-documented chapter would have gained
in interest if it had not been so limited, if it had not held the para-
doxical position, notably, of treating Tantric pujd without citing
even once the Somasambhupaddhati, a text which in H. Brunner's
edition is probably the richest and certainly the most accessible of
the sources on Saiva Tantric ritual.5 The fact that the three parts of
the book are not equally limited in their sources contributes to the
discomfort of the Indologist who, at times, being unaware of which
school or texts the author is writing, wonders whether he can agree
with an assertion that may be valid for one sect but not for another.
In any case, if a separate study of diverse traditions may prove
useful for clarity's sake (the same division is found in the History of
Indian Literature, cited above, n. 3), the general rule of not dealing
with everything at the same time should not prevent one from
underlining-very far from denying it-the common Tantric charac-
ter of all the schools.
There would be much to say about this very stimulating and
interesting work, both to record agreement and, occasionally, dis-
agreement. But one cannot enter here in all the details. Let us then
mention only some problems or a few specific points which seem
important to the subject of Tantrism.
The first and, no doubt, the main problem-the prejudicial ques-
tion, one might say-is only skimmed by Goudriaan in the first part
of the book. It is the question of the existence of Tantrism. Specialists
are more or less in agreement as to the constituting elements that
one finds in the majority of texts considered to be Tantric. Eighteen
of these elements are enumerated in pages seven through nine: This
enumeration is thoroughly valid, even if other elements might be
suggested, notably the nature of divinity, polarized into two aspects,
masculine and feminine, as this is fundamental to Hindu Tantrism.
But does the existence of such particular traits allow one to think
6 H. Brunner, ed.,
Somasambhupaddhati, 3 vols. (Pondichery: Institut frangais
d'indologie, 1963, 1968, 1977).
350 Tantric Hinduism
that something like Tantrism really exists? Is it not rather an
abstraction fabricated by orientalists on the basis of these traits?
Might Tantrism not be even, to cite H. V. Guenther, "probably one
of the haziest notions and misconceptions the Western mind has
evolved?"6
The word "Tantrism" assuredly is a Western creation. India tradi-
tionally knows only texts called Tantras. These texts, moreover, fall
far short of covering the entire Tantric literature; nor are only
Tantric texts called Tantras. India also knows the word tantrdsdstra,
"the teachings of the Tantras," as well as the adjective tdntrika,
"Tantric," which is opposed to vaidika, "Vedic," thereby placing a
new form of revelation and rites against Vedic tradition and rites.7
Avalon noted already in 1922 that "the adjective Tantric is largely
a Western term." Now, if at that time Tantrism could be conceived
as a phenomenon limited and specific enough to form only a particular
and even a slightly exceptional aspect of Hinduism (or of Buddhism),
today we know more. We realize that the practical and doctrinal
elements which we think we can call Tantric are evident in some
measure everywhere from a certain period (seventh to eighth cen-
turies, it seems) onward. So much so that from that time Tantrism
appears to be a "pan-Indian mode," according to Mircea Eliade's
expression, whom we must credit as the first to have realized this
fact. It becomes then exceedingly difficult to establish Tantrism as
distinct from what is not Tantrism.
Thus, one can be tempted to deny the existence of Tantrism as a
particular form of Hinduism (or Buddhism) and to see nothing in it
other than what these religions had become after a certain epoch.8
One also might circumscribe it very narrowly, as did J. Filliozat,9 by
arguing for its essentially ritualistic aspect: It would be, in reality,
"only the ritualistic technical aspect of (the) religion, be it Scaiva,
vaisnava, Buddhist or Jain-a treatise of religious architecture is
necessarily 'tantric.'"
I share neither of these points of view. But, I think, one cannot
rest content in merely ignoring them: the problem of the definition
of Tantrism, of the very existence of a Tantric specificity, is a very
real and important one. The fundamentally ritualistic, practical, and
applied character of Tantrism is undeniable, as is the presence of
Tantric elements nearly everywhere in Hinduism from the Middle
Ages to our own day. It seems possible, despite this, to believe in the
6 Herbert
V. Guenther, ed., The Life and Teaching of Naropa (New York:
Oxford University Press, 1971), p. 102.
7 Tdntrika and Vaidika traditions are
posed as differing, but not necessarily as
in opposition to each other.
8 It is
impossible at present to say when Tantric forms arose in Buddhism and
Hinduism. But the latter naturally preceeded the former, even though collections
of Buddhist dharanis are attested to in Chinese translations prior to any datable
similar Hindu element. The fifth century seems likely to me. The question is well
surveyed by Goudriaan (pp. 17-25).
9 J.
Filliozat, Journal asiatique 256, no. 2 (1968): 268.
History of Religions 351
proper existence of Tantrism-but on the condition that both the
doctrinal and the ritual aspects be retained. T. Goudriaan has pre-
cisely, and very aptly, devoted a whole chapter ("Tantrism in Hindu
Religious Speculations," pp. 47-67) to doctrine, stressing that "doc-
trine is, in Tantrism even more than in other Hindu religious move-
ments, inseparable from practice." The conjunction of these two
elements will permit us to define Tantrism-and to recognize its
followers.
Accordingly, we might posit Tantric Hinduism as different from
Hinduism in general (not only from smdrtal' Hinduism, which is
closely aligned with Vedism) by regarding it globally, using Madeleine
Biardeau' words, as an attempt "to place kdma-desire-in every
sense of the term, in the service of deliverance . . . not to sacrifice this
world to deliverance, but to reintegrate it in one way or another
within the perspective of salvation."" With such a goal, Tantrism
has recourse to a complex of ritualistic, psychic, and corporal prac-
tices that make use of elements of this world and of the body in
particular, practices and notions that correspond to a certain concep-
tion not only of the godhead, conceived as polarized into masculine/
feminine, but also of the universe and of man, both being immersed
in that divine power. As a result, the quest for liberation is funda-
mentally nothing but a tapping, a using, or even a manipulating of
that power. Tantrism, thus, closely associates a doctrine, an aim, a
"weltanschauung," and certain practices which are grounded in that
doctrine, explained and oriented by it: this, I believe, is essential.
Admittedly, some aspects of the doctrine and still more some such
practices can also be found in non-Tantric schools, as they are very
widespread. But, when the two are associated, when the practices
are oriented and organized by that doctrine and form a system within
it, one is inside Tantrism'2 or, to put it more cautiously, one is inside
one of the Tantric forms'3 of one or another of the Indian religions.
10"Ce n'est pas une secte a
proprementparler, mais bien plutot l'ensemble des
brahmanes repr6sentant l'orthodoxie an regard des sectes" (L. Renou, L'lnde
classique[Paris: Payot, 1947], 1:623.
11 Madeleine Biardeau, Clefs pour la pensee hindoue (Paris: Seghers, 1972), p.
209.
12 This
importanceof ideology must be stressed, as in the Hindu religiousworld
what is done is more important than what is believed. (As Frits Staal wrote,
Hinduism is orthopraxyrather than orthodoxy.) It is precisely the role of Tantric
ideology which permits the authors of this book to count as Tantric those people
only "who are wont to call themselves by that name" and who are not very
many nowadays, whereas the texts, all rich in ritual prescriptions but also
carriersof an explicit or implicit doctrine, are extremely numerous.One might be
tempted to think that Tantrism, despite its practical side and the importance of
the actual experience,is above all a matter of texts, only secondarilyof practicing
adepts (officiatingpriests in the temples or initiates), and very little of the faithful
community. This, at least, is what we see in India today, but has it ever been
otherwise?
13 I use the plural because of the
fundamentally sectarian, sometimes even
intolerant, character of tantric Hinduism. In this, Tantrism differs from bhakti
(which is universalist in character), despite common traits.
352 Tantric Hinduism
This being said, I should not like to appear more sure of myself
than I actually am. I am aware of the fact that this tentative defini-
tion of Tantrism is very largely theoretical, hypothetical even, but I
believe it, at present, to be justified. It seems to me, further, to offer
a good point of departure for a study of the question. This starting
point, however, despite the heuristic value which I believe it to
possess, may prove to be incorrect as our knowledge of the subject-
matter progressively increases. Perhaps Tantrism is in reality some-
thing other than what I have just written, or perhaps it does not
even exist as such. The debate is not closed.
Another debatable question presented by Tantrism (a question
that is examined on pp. 15-25 of the book) is that of its origin: What
led to the appearance in Hinduism of Tantric speculations and
practices-and how did this happen? It is impossible to answer con-
clusively. This is an additional reason for approaching Tantric
Hinduism with a structural working hypothesis, which would take it
as a system already formed (without ignoring, of course, the known
historical stages), rather than to embark upon a search for its origins,
which, in the present state of historical knowledge, would inevitably
be vain.
Goudriaan's examination of these problems (chap. 2, "Tantrism in
History") is brief, but accurate and sensible. He sets aside Marxist
schemas (p. 17) which almost never apply to India, but cautions also
against other Western prejudices (p. 29). He admits the "autochtho-
nous" hypothesis as a possible explanation of the appearance of some
notions or practices that are foreign to Brahmanism and whose
origins, if they are not locally founded or Dravidian, are not easy to
figure out. But he also stresses the common elements and the often
so striking continuities, in the rites as well as the ideas, which link
Tantrism to Vedism. This is probably all one can say at present
without yielding either to prejudice or to illusion. He just as wisely
does not give much credence to the various theories on "the original
home of Tantrism," which reflect above all the personal preferences
of their authors, and disagrees in particular with the hypothesis of
Chinese influences on Tantrism, an influence which is all the less
likely as Chinese Tantrism (which, moreover, is Buddhist) appears
very attenuated when compared to that of India.
The same reasonable-one might say agnostic-approach is used
for the obscure problem of traditional classifications of Tantric
schools and texts (pp. 40-46). One can hardly do more than take
note of the distribution of texts and practices into a varying number
of sampradaya (schools), amnaya ("remembrance"), mata (doctrine),
dcdra (method), etc., classifications at variance with each other,
perhaps even purely local, or even personal. To disentangle this
skein and to make it appear as a unified system-supposing that one
ever existed, which is not sure-is for the time being impossible, and
 History of Religions 353
to devote oneself to it is, I fear, a waste of time. We should first
develop further the study of certain schools,14and probably a unified
view will emerge eventually.
T. Goudriaan's study also brings out the fact that the "autoch-
thonous" (as opposed to Indo-Aryan, but also to drya and Brah-
manic) origins of some Tantric elements does not mean that Tantrism
is a form of popular religion. That it may have incorporated some
elements of popular origin is probable. That these elements may have
played a fecundating and important role is possible. But the synthe-
sis, the textual elaboration-the only one we can ascertain-is in no
way popular. The simple fact that these texts are in Sanskrit proves
that they are scholarly, the work of Brahmins, expressing the ideology
of the higher castes. They may even be the products of extremely
learned, intelligent and cultivated Brahmins:15Abhinavagupta comes
to mind, but a number of other names could be cited. Whereas the
popular origin remains the exception, or might come from the
Dravidian domain. But, even so, one must not confuse Tantrism
with popular religion. (Let us underline here that popular Indian
religion, today, ruled as it is by the same ideology as the religion of
the higher castes, cannot, generally speaking, properly be called
Tantric.) Doubtless, one can find in Tantrism elements which can be
considered popular but, in addition to the fact that they were taken
over in a scholarly textual elaboration, a characteristic of Hinduism
is precisely that it exists, it is practised, on different levels, from the
vulgar magical (so to speak) to the most subtle metaphysical, from
crude cult to the highest mysticism.
May I finally say on this subject that if the "renouncers," ascetics,
or "saints" of various sorts played an important role in the evolution
of Tantrism,16 their place on the edges of the Hindu social system
does not make them either popular elements or elements external to
the system, still less revolutionaries. As Goudriaan writes, "The
ritual egalitarianism of Tantrism in practice acted as a class-con-
14
Those most often studied until now are usually related, or belong, to Kashmir
Saivism, on which the basic study is still K. C. Pandey's monograph, Abhinava-
gupta, 2d ed. (Varanasi, 1963). P. C. Bagchi has done pioneering work on the
Kaula school. On the Krama school, we mav mention a recent study by N.
Rastogi, Krama Tantricism of Kashmir (Delhi: Motilal Banarsidass, 1979). See
also the eight volumes of L. Silburn on the subject, all published by the Institut
de civilisation indienne de l'Universite de Paris (Paris: Editions de Boccard,
1957-80).
15
Naturally, there are lower-caste practices in Hindu Tantrism, but they are
not the most important. Further, autochthonous elements did not contribute
necessarily to their appearance. There remains, finally, the case of Tantric sects
of popular origin and tradition which remained such; the Bauls would be a case
in point, and theirs is probably not an isolated one. All this is still very little
known.
16 L. Dumont considers the renouncer
as "the agent of development in Indian
religion and speculation, the 'creator of values.' " See his "World Renunciation
in Indian Religions" (Religion, Politics, and History in India [Paris: Mouton,
1970]).
354 Tantric Hinduism

firming force" (p. 32): The egalitarianism of Tantrism, like that of

bhakti,17has a religious and ritual value, but not a social one.
This leads me back to what I said earlier (n. 12). Tantrism as a
system practiced and experienced could hardly have ever concerned
many people. It could never have been a mass religion. It is very
likely that Tantrism as a system of religious practices existed only
"in limited circles all over India" (p. 28). "All over India" must be
stressed, and it should be added that these circles were no less
influential for being limited.18They produced an enormous doctrinal,
ritualistic, and technical literature, and created, or left their stamp
in various degrees of intensity on, the whole of nonsmarta Hindu
ritual, that is to say the whole of public ritual and a good part of
private ritual. The fact that those who attend the Tantric or tantri-
cised rites of Hindu temples, or those who perform themselves or
have performed domestic rites which carry the same Tantric imprint,
do not consider themselves-in any case today-as being tantrikas,
does not take away any of the Tantric character from these practices.
It has however as a result that we cannot consider these believers as
tantrikas,19 a condition that presupposes an adherence, at least in
principle, to Tantric doctrines. How was the situation in the past?
For example, from the tenth to the twelfth century, the time when
the temples, at Khajuraho or elsewhere in central India (which seem
to us the perfect illustration of the Tantric Weltanschauung) were
built? We cannot tell. But it is probable that, then as today, the
popular masses who visited the temples, who went on pilgrimage to
the pithas of the Goddess, who propitiated the sixty-four yoginis,
etc., thus worshipped these sometimes dreadful forms of the divine,
their refuge in a world filled with real or imaginary dangers, but that
Tantric ideology or practices were not an integral part of their
everyday life.20 One is thus tempted to believe that the presence of
17
Bhakti, the religion of devotion, is one of the fundamental currents in Hin-
duism. It is more widespreadthan Tantrism at all levels of Indian religious life,
and its motifs are to be found in most texts such as Puranas, Tantras, and
Samhitas. There is hardly a Tantric text that does not say that the rites it
prescribes must be accomplished with devotion (bhaktyd).The dualistic Saiva
Agamas have as their main goal the teaching of bhakti. On the relationship
between Tantrism and bhakti,see the excellent little book by M. Biardeau (n. 11).
18 This was noted by L. Renou: "Le phenomene sectaire n'a
jamais do, dans
l'Inde ancienne pas plus que dans l'Inde moderne, condenser toute l'activite
religieuse. Q'a du etre affaire de minorites agissantes" (L'Inde classique, 1:622).
19However, since their rites are Tantric, it is necessary to study them as such
together with the doctrines that accompany these rites. Hence my reservations
concerning the scope of this book on pp. 347-49 above.
20 It would
only be fair to add that the prescriptionsof orthodox, non-Tantric,
smarta Hinduism, also did not directly concern a very large number of people:
The worksof Brahmins,who were the only ones able to follow them, meant for the
higher "castes," these prescriptionswere and still remain limited to an elite. The
"greattradition" (as anthropologistssay), whetherTantric, Agamic, or otherwise,
would thus never have concerned what today we would call the masses, even
though popular Hinduism only exists in direct relationship with that tradition.
History of Religions 355

Tantrism was, in short, predominant in the texts, important in

temples and rites, but much more limited as a practice actually put
to work and truly lived and experienced.
The second part of Hindu Tantrism, "Transmission and Funda-
mental Constituents of the Practice," by D. J. Hoens, is the least
satisfying. Perhaps the author found himself particularly hampered
by the brevity imposed on him by the format of the work. In any
case, he gives the impression of being overwhelmed by the abundance
of a subject-matter over which he has little control. Above all,
perhaps in order not to return to what was said previously about
Tantric doctrines, he does not place the practices he describes in
some detail within the doctrinal frame which gives them their
significance. This causes a disagreeable impression of inconsistence
and arbitrary enumeration which can only baffle the nonspecialist
reader. Finally, there are many inaccuracies.
The subject, to be sure, is difficult and as yet little known. Not
that published texts are lacking; on the contrary, there are many of
them, but they vary in details of the prescriptions and have never
been systematically studied. A whole universe lies there, still to be
explored.21 Therefore, it is not easy to make a choice among those
texts, nor to draw from them a coherent, generally valid, "middle"
doctrine. But if one wishes to remain close to the sources-to several
sources which mutually disagree-one then either simply cannot give
a clear and systematic account of the facts, or else one risks presenting
as a general rule what is only a particular case. It would have been
better, therefore, I believe, to lay stress on the spirit of these prac-
tices, on their orientation, their place and their role in the Tantric
conception of the universe and of the position within it of the initiate,
rather than to heap a mass of often uninteresting or even useless
details (see p. 79), without even giving a complete picture of the
subject, for that matter.
The author first examines the "Tantric Transmission" (chap. 3).
He analyzes the practices of the diksa, the initiation, a necessary rite
of passage for every follower who wishes to escape from the fetters of
this world and to be free from transmigration. This rite is, in principle,
required for admission into a Tantric sect, hence is distinct from the
initiation received by every male member of any of the three classes
of the twice-born. There are several such initiations in all sects. They
vary according to the disciple's degree of advancement or, as Abhina-
vagupta teaches in his Tantraloka, according to the intensity of the
divine grace from which he benefits. Such initiations, by severing
some or all of the fetters to wordly life, will not so much directly
liberate the initiate as put him in a state where he is most able to
21
Generally speaking, Pauranic, Agamic or Tantric Hinduism-though
form the bulk of Hindu religion as it has been and still is practiced-have they
not
received from Indologists the attention they deserve.
356 Tantric Hinduism
receive the divine grace (anugraha). Since the Tantric perspective is
completely permeated by power (sakti), grace is also called "descent
of power" (saktipdta).22The further the candidate is from liberation,
the more ties he will have to cut, the more complex the initiation
will be, and the more it will have to be followed by a long and com-
plicated period of practice, discipline and worship (called sddhana).
All this varies considerably from one school to another.23 The author
has chosen to enumerate the stages of one of the initiations, the
kriyd-diksd, according to the Saradatilaka. He then briefly considers
a few other examples. The reader thus gets a general idea of the
initiation process.24
In the following chapter the author reviews the "Mantra and
Other Constituents of Tantric Practice." Of the chapter's twenty-
seven pages, twenty-one are devoted to Tantric speculations and
practices regarding sound, word, and mantra. This can be justified,
as Tantric teaching (tantrasdstra) and the teaching of mantras
(mantrasdstra)often are considered equivalent terms.
If I were not afraid of appearing inconsequent or irksome, I would
be tempted to charge Hoens with having followed too closely, in his
presentation of the subject, the plan of my thesis, to which he refers
several times.25 A different and more personal approach would
naturally have been of more interest to me. But that is a subjective
reaction that other readers may not share. However, they will have
trouble, I fear, finding the leading thread of Hoens's study, the general
idea which justifies the speculations and explains these practices, and
which is essential to this matter, as well as, more generally, to the
Tantric system. This central idea can be summarized by saying that
22 The role of
grace in Vaisnavism is well known. One is much less aware of it
in Saivism, especially in the Tantric Saiva schools, where it is also quite important.
In Kashmir Saivism, anugraha is one of the five basic actions-and the most
important-of Siva. The role of grace is also emphasized in the agamas. This is
important. It is one of the fields where Tantrism and bhakti meet.
23 In the introduction of vol. 3 of the
Somasambhupaddhati one will find a clear
exposition of the theory of dik~a according to southern Saivism. Several diksas
are described in the three volumes of the text. Similarly, chaps. 14-26 of the
Tantraloka would be of interest to students of Kashmir Saivism. Chap. 13 intro-
duces these chapters with a review of the different forms of the "descent of
power." The Tantraloka is available in the Sanskrit edition of the Kashmir
series and in the Italian translation by R. Gnoli, Luce delle sacre scritture (Torino:
Unione Tipografica-Editrice-Torino, 1972).
24 Here are a few of the
inaccuracies: On p. 81 the term dantaka.stha, which
Monier-Williams defines as "a small piece of wood used for cleaning the teeth," is
translated as "tooth-brush," which is slightly ridiculous, and not quite exact.
The reason this instrument was used so frequently for the purpose of divination is
not clear. On p. 85 there is a more serious mistake: Mantrasadhana is not
tion of mantra" but the set of practices that the disciple must perform after "perfec-
his
initiation in order to "embody in himself and activate the presence of the mantra"
that he received in the course of his initiation. (On this subject see H. Brunner's
"Le Sddhaka, personnage oublie du Sivaisme du sud," Journal Asiatique 263
[1975]: 423-35.) Also, ndddnta is translated on p. 87 as "having ndda as end."
It is just the opposite: "That which comes at the end of ndda."
25 Andr6
Padoux, Recherches sur la symbolique et l'energie de la parole dans
certains textes tantriques, 2d ed. (Paris: Editions de Boccard, 1975.)
History of Religions 357
the most efficacious and, at the same time, the most apparent and
most usable form of power-the form also which is present in all
human beings, creatures of the word, inescapably immersed in
universal sound and more specifically in language-is vdk, the Word;
that the Word, at its highest level, is identical to the omnipresent
divine energy which creates the universe, endows it with life, sustains
it, finally to reabsorb it within itself. And, lastly, that the cosmic
process and the human process of word, sound, or speech are parallel
and homologous. From this central idea ensues the entire impressive
development of the metaphysics of the Word, of the phonic and
phonetic cosmogonies, and of the practices of all sorts (initiation,
cult, yoga, meditation, etc.) which are achieved through the use of
speech or word, and more specifically through its most efficacious
and usable form, the mantras and bzjas.26This is essential to Tan-
trism. In fact, one recognizes the tantrization of Hinduism by the
presence of the very practices that make use of mantras and even
more of b-jas (that is, syllables or group of syllables devoid of obvious
meaning).
It is understandable that the author did not want to embark upon
a theory of mantragdstra,a complex subject, and one that has as yet
not been the subject of serious study.27 But, since he neglected that
theoretical frame too much, his presentation of the question looks
like a patchwork of oddly assorted pieces, badly joined, and not put
in a proper perspective. This chapter may, nevertheless, prove useful
to the nonspecialist reader insofar as it enumerates little-known facts.
Yet, it will no doubt leave such a reader with a complicated and
bizarre impression which would not have arisen if the author had
clearly stated the doctrines that justify these practices and specu-
lations.28
The same can be said of the few pages on the yantras and mandalas,
as well as of those on the mudras, all of which still await a serious
26
These syllables do not "always end in anusvara,m" (as stated on p. 105)-for
example, Phat and Sauh.
27 For an
early, incomplete, yet pioneering study on this, see A. Avalon's The
Garlandof Letters(Madras: Ganesh & Co., 1951). One will find an outline of this
problem in my Recherches,cited above. A more thorough study of the doctrines
and practices of the mantrasdstra remains to be done. Such a study would reveal
interesting aspects not only of the subject itself but also of religious behavior and
of the uses of sound and language in general.
28 The
expression "The Origin of the Alphabet" (p. 98) will surprise some people.
First, because it would be better not to use the word "alphabet" when referring
to the traditional order of phonemes in Sanskrit. But also because Tantric texts
do not describe the origin of phonemes, but, quite differently, the way in which
the universe is born from the phonemes, which are eternal. (The notion is Vedic.)
On the same page, "R" is not "stable," but, on the contrary, instantaneous. A
mantra is not exactly "any combination" of letters, but rather those combinations
of phonemes, not letters, which the tradition recognizes as such (p. 101). The
final paragraph (p. 110) on Hamsa is nearly incomprehensible. Why, for example,
should the sakti "SA" be "unutterable"? Because it is not "SA," but the sibilant
"S" without a vowel following (anacka), which, according to traditional grammar,
cannot be pronounced.
358 Tantric Hinduism
and thorough study, especially in the Hindu context. The mudrds, in
particular, raise a number of interesting problems, whether one
approaches them as a system of signs, or as gestures accompanying
and stressing an utterance (this is the problem of the relationship
between mantra and mudrd: In the beginning, was there Word or
gesture?); or whether one studies them as symbols of divine power,
seals, or imprints of that power and of the covenant between a
worshipper and a deity. Or else if one considers mudra in some of its
other meanings-as the feminine partner of Tantric yoga, for in-
stance, or as "mystical attitude" (it is then the "seal" of the Unut-
terable)-referring to the realization of the highest Reality.
The third part of the book, written by S. Gupta, is devoted to the
"Modes of Worship and Meditation." "Tantric sddhand [religious
practice]," she writes (p. 121), "consists of two parts: ritual worship
(pujd) and meditation (yoga). Both are of equal importance to every
Tantric. Even the siddha or avadhita, recognized to be so highly
spiritual that he can afford to disregard rules applicable to ordinary
Tantrics, continues to perform his daily pauj along with his yogic
practices." Hence the importance of chapter 5, which is devoted
to that subject.
Gupta recalls the traditional distinction of three types of cult:
Vedic, Tantric, and mixed. In fact, the high-caste "Vedic" (or
smdrta) ritual has, no doubt, hardly existed for a long time, except
in private rites. With this exception, one seldom finds today (and
probably it has been so for centuries) any pujd that is not Tantric
or tantricized. This pjad, like all rites, is of different types and varies
according to many elements; there is thus a great variety of pujds,
which, naturally, cannot be considered, much less studied, in one chap-
ter. The author has therefore chosen to describe only one in detail,
which is very justified, as ritual worship throughout all its variants
more or less always follows the same general pattern in all Hindu sects.
Her choice was made according to the criterion that I have already
criticized, "that a Tantric is one who firmly believes himself to be a
Tantric": indeed, one can be sure that the man who has such a belief
will perform a Tantric p-jd. But, as I said, the opinion of the tantrika
of today should not be taken account of in such a work as this one.
One must however, in all fairness to Gupta, underline that she
admits that there are "dangers" in her method (p. 121), and that
she states clearly her case and specifies (p. 129) which variety of the
contemporary rites she describes. Her presentation of the subject
gives therefore rise to no misunderstanding.
Gupta's description of the pajd done in some contemporary sdkta
circles (pp. 131-60) is complete, competent, clear, and gives a good
idea of this sort of practice. It is illustrated by eight photographic
plates. Her comments on the subject are interesting and at times
very illuminating. One regrets all the more that her restrictive choice
History of Religions 359

prevented her-if not from describing other types of pujd for which
space would have been lacking-from simply referring to rites used
by other Saiva or VaiSnava sects, which would have given her work
a greater and wider interest.29
The chapter ends with a brief passage on optional rites (kdmya-
puja), where she alludes, in particular, to the "six ritual acts" (sat
karmdni), performed in order to achieve some particular end, espe-
cially a material one, such as defeating or harming an enemy. Kdmya-
pujd's character, in such cases, borders on magic,30 or may be con-
sidered purely magic. But the problem of magic as different from
religion is a complex one, more so perhaps in India than elsewhere (if
only because the opposition magic/religion stems largely from Chris-
tian Europe, resting as it does on linguistic categories that do not
necessarily apply in other parts of the world). This appears especially
in Tantrism, where all or almost all rites and practices appear as
manipulations of a power existing and active at all levels which,
themselves, are indissociable. But this is not the place to discuss this
subject, nor that of the role and significance of "supernatural"
powers in Tantric Hinduism.
The sixth and final chapter of the book deals with yoga. This is a
particularly important and fascinating subject, as Tantric yoga, a
simultaneously corporal, mental, and spiritual process with micro-
macrocosmic dimensions and a simultaneously immanent and tran-
scendent aim, sets at work all of the Tantric practices and techniques
in order to lead the adept to liberation in the present life. This
ascetic discipline includes many possibilities, and virtually all of
them, even the most baroque, are put into play. Such practices
include attention to the "mystical physiology" of nddis and cakras;
the mastery of the breath; the use of mantras and bzjas; concentration
and fixing of one's consciousness on a single point; interiorization of
rites as well as of divinities and of the macrocosm; visualization; and
the mastery of supernatural powers.31 Tantric yoga thus appears as
the actualization of all the procedures and notions enumerated or
described in the course of the book. It is, therefore, specially fitting
that this book should conclude, or culminate as it were, with this
29 For
other descriptions in Western languages of pajd, see the English transla-
tion by Sanjukta Gupta of the Lak?mi Tantra (Leiden: E. J. Brill, 1972), which
depicts Vaisnava rites, or K. R. van Kooij's Worship of the Goddess according to
the Kdlikdpurana (Leiden: E. J. Brill, 1972). The most complete study (as I said
in n. 6, above) remains that of the Somasambhupaddhati in H. Brunner's edition
and translation.
30 On these rites, see an interesting
passage (pp. 251-413) in T. Goudriaan's
book Maya Divine and Human: A Study of Magic and its Religious Foundation
in Sanskrit Texts (Delhi: Motilal Banarsidass, 1978).
31 The use of yogic techniques to attain liberation in a context where devotion
and divine grace play an important role obviously presents some problems. These
were resolved in various ways by the different schools. In Kashmir Saivism, for
example, yogic practices remain subordinate to divine grace, their role increasing
as the intensity of grace weakens. For a study of the problem in a non-Tantric,
Vaisnava school, see R. Lester's Rdmdnuja on the Yoga (Paris: Adyar, 1976).
360 Tantric Hinduism

chapter. Gupta makes a good and sober presentation of the material.

It is too short, unfortunately, to bring out all the variety and breadth
of the subject, but, as it stands, it is already very instructive.
The author's textual references, while diverse, are principally from
sdkta literature. She refers most frequently to two works of srividyd
devoted to the goddess Tripurasundari, the Nitydaodasik&ar.avaand
the YoginThrdaya.The latter contains, in its last chapter, on pujd, a
curious description, not mentioned in this book, of a japa (an "enun-
ciation": One cannot call it a muttering which is the usual translation
of the word) of the srividyd, the mantra KA E I LA HRIM, HA SA
KA HA LA HRIM, SA KA LA HRIM. The three parts of this
mantra are visualized as being present in three of the cakras of the
subtle body (placed at different points along the spine of the material
body).32 They are then drawn upwards by the ascending movement
of their pronunciation (uccara)33which is linked to the cosmic and
human movement of "breath" (prdna). This ascending thrust of the
phonic energy of the mantra carries upwards the ku.?dalini and
culminates in its immobilization at the highest point of both verbal
pronunciation and the subtle body. The worshiper is then fused with
the Supreme Reality, pure bliss, the "ocean without waves" of the
transcendental divinity concealing within itself the totality of cosmic
appearances of which it is the source and the foundation. This whole
difficult and complex operation is described as being only one of the
moments of the pujd of the goddess, the diversity and extent of which
one can thus imagine.34
The nature of the Tantric vision of the world, its anthropocosmic
reach, is best revealed in such practices. Deity is thus achieved by a
corporal, mental, and spiritual process: A process by which the body
itself is made cosmic and divine and yet, at the same time, is tran-
scended. These practices express better than anything else the
extraordinary attempt of Tantrism to reintegrate man, made divine,
into a universe that is itself seen and experienced as divine. This
indeed is Tantrism.35
ANDRE PADOUX
Centre national de la recherchescientifique, Paris

32
According to the evocative formula of T. Goudriaan, the subtle body is
"intraposedwithin" the visible body (p. 57).
33 Here again, one may speak, with T.
Goudriaan, of the "vertical stages of
consciousness" (p. 61).
34 See A. Padoux, "Un japa tantrique: Yoginihrdaya,III, in Tantric
and TaoistStudiesin Honorof R. A. Stein,to be publishedin171-190,"
Louvain.
36 A last paragraphof 23 lines only (p. 183) deals with the "sexual practices," to
which passing references are made in that chap., in connection with pija. The
very limited space thus allotted to this aspect of Tantrism is, to my mind, quite
appropriate: The use of the sexual drive, and the practices arising from there,
correspondundoubtedly to something fundamental in Tantrism, but they are by
no means its main element.