Tantric Mantras

Providing a systematic and complete overview of the highest scholarly quality

on Tantric mantras in Hinduism, this book presents a summary on the nature of
Tantric mantras, their phonetic aspect, structure and classifications. Additionally,
it explains the metaphysical–theological nature of Tantric mantras and gives an
introduction to their beliefs and practices. In individual chapters, André Padoux
discusses the extraction and examination of mantras, certain characteristics such
as their ‘perfect nature’ and their imperfections, and describes certain mantrics
practices.
For the first time, André Padoux’s work on Tantric mantras is made accessible
to an English-speaking readership. This book will be of great interest to scholars of
Religious Studies, Theology, Indology, South Asian Studies and Asian Religion.

André Padoux has had a distinguished career as a diplomat and then as a scholar
of Sanskrit, notably of ‘Kashmir’ Shaivism and Shakta Tantrism. He was Director
of Research at the Centre National de la Recherche Scientifique, Paris, for many
years. He is a renowned expert on the ‘science of mantra’ (mantraśāstra) and has
published important articles in this field.
Routledge studies in Tantric Traditions
Series Editor: Professor Gavin Flood
University of Stirling

The Routledge Studies in Tantric Traditions series is a major new monograph

series which has been established to publish scholarship on South, East and South-
east Asian tantric traditions. The series aims to promote the serious study of both
Hindu and Buddhist tantric traditions through the publication of anthropological
and textual studies and will not be limited to any one method. Indeed, the series
would hope to promote the view that anthropological studies can be informed
by texts and textual studies informed by anthropology. The series will therefore
publish contemporary ethnographies from different regions, philological studies,
philosophical studies, and historical studies of different periods which contribute
to the academic endeavour to understand the role of tantric texts and their meaning
in particular cultural contexts. In this way, the series will hope to establish what
the continuities and divergencies are between Buddhist and Hindu tantric tradi-
tions and between different regions. The series will be a major contribution to the
fields of Indology, Sinology, History of Religions, and Anthropology.

Identity, Ritual and State in Tibetan Buddhism

Martin A. Mills

The Khecar⎜Vidy⎩ Of ⎩Din⎩Tha

A critical edition and annotated translation of an early
text of hahayoga
James Mallinson

An introduction to Tantric philosophy

The Paramarthasara of Abhinavagupta and its commentary
by Yogaraja
Lyne Bansat-Boudon and Kamalesha Datta Tripathi

Tantric Mantras
Studies on mantrasastra
André Padoux
Tantric Mantras
Studies on mantrasastra

André Padoux
First published 2011
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2011 André Padoux
The right of André Padoux to be identified as author of this work has been
asserted by him in accordance with sections 77 and 78 of the Copyright,
Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Padoux, André.
Tantric mantras : studies on mantrasastra / André Padoux.
p. cm.—(Routledge studies in Tantric traditions)
Includes bibliographical references and index.
1. Mantras. 2. Tantrism—Rituals—Texts—History and criticism.
I. Title
BL1283.855.P33 2011
294.5'435—dc22
2010050162

ISBN: 978–0–415–42386–1 (hbk)

ISBN: 978–0–203–81449–9 (ebk)

Typeset in Times New Roman

by Swales & Willis Ltd, Exeter, Devon
To the memory of Harvey P. Alper who conceived
and edited Understanding Mantras, the only
overview of the subject of mantras.
brahma vai vāc.
(Aitareya Brahmana)
OM. , that syllable truly is Brahman.
That syllable indeed is the Supreme. …
Knowing that syllable, indeed whatever one desires, is his.
(Katha Upanis.ad 2.16)
The vast majority of those mantras are nonsense syllables such as Hring, Hung,
Hum, Phat, sparks from the blazing furnace of aboriginal superstition whence the
system arose, or from the equally superstitious stores laid up in the Atharvaveda.
(Farquhar, An Outline of the Religious Literature of India)
Contents

Foreword ix

1 Introduction 1

2 The extraction and examination of mantras: mantroddhāra

and mantravicāra 12

3 Japa 24

4 Nyāsa: the ritual placing of mantras 54

5 A Hindu rosary ritual (Jayākhyasam

. hitā, chapter 14) 81

6 On the defects and the perfecting of mantras: mantrados.a

and mantrasam . skāra 89

7 Mantric practices and the nature of mantric utterance 95

8 Body and mantra: mantras in the human body 103

9 The oral and the written: mantra and mantraśāstra 111

Notes 123
Index of Sanskrit terms 160
Authors and works cited 165
Foreword
Gavin Flood

In the following pages the reader will find a systematic exposition and explanation
of Tantric mantras as they have existed broadly speaking within ‘Hindu’ Tantric
traditions. It could be argued that mantras are at the heart of Indian religious prac-
tice and although they have been studied both from the perspective of indigenous
philosophy and through western academic disciplines, there is still much that can
be said about them. Some twenty years ago Professor Padoux reminded us of
the vastness of the subject of mantras and the need for us to understand Tantric
theories of mantra, such as those that developed in the medieval period in Kash-
mir and elsewhere, and also contemporary practices and practitioners of mantra.1
The present book goes some way to filling that gap and addresses questions of
fundamental concern to Indology, South Asian Studies, and the study of religions,
describing what mantras are, showing how they are used in a ritual context, and
presenting what the Tantric traditions themselves regard mantras to be. The pur-
pose of this foreword is not so much to summarise the following chapters but
rather to introduce the idea of mantra and to put the material Padoux presents into
a wider historical perspective.
But what are mantras? This entire volume might be seen as an answer to that
question within the remit of the Tantric traditions. Among the definitions listed in
Monier-Williams’ Sanskrit Dictionary are ‘sacred text or speech, a prayer or song
of praise . . . a Vedic hymn of sacrificial formula . . . a sacred formula addressed to
any individual deity . . . a mystical verse of magical formula (sometimes personi-
fied), incantation, charm, spell. . . .’2 This is a beginning but these definitions do
not tell us much about the nature of mantra, its scope or precise use. As Padoux
has observed, mantras are in Sanskrit, they are conventional formulas transmitted
through tradition, and are not made up.3 In the Tantric traditions mantras must
be seen in a cosmological context and different mantras correspond to different
levels of the hierarchical cosmos understood as levels of consciousness or speech,
as Padoux has demonstrated. This association of mantra with cosmology is argu-
ably the case throughout the history of Indian religions. The cosmos is under-
stood as the flow of the energy of the word (vāc) emanating from its transcendent
source and returning there. Mantras are both expressions of this word and means
of returning to that source. Mantras, Padoux writes:
x Foreword
appear as privileged instruments for the return to the source of the energy of
the word. More than that, a mantra is often itself a symbol or, rather, a form
of this primal energy. It retains this pre-eminently in a most effective and
practical form. But it is also alive with an inner force tending intensely toward
the primal source of all speech, toward the Power which is the Word. Mantra,
therefore, brings together both the practically effective and creative, and the
transcendental and liberating, powers of the Word.4

Mantras express or embody principles, powers, or deities that are thought by the
tradition to exist at higher cosmic levels. Thus to repeat a mantra is to directly
engage with a deity and become located within the cosmical hierarchy. We
might say that even if mantras are non-linguistic, as with the ‘seed syllables’ or
bīja mantras, they are meaningful in so far as they serve to locate a person within
the structure of the universe and they are drawn from the Sanskrit alphabet. As
Padoux has shown, one of the characteristics of Tantric mantras is that the man-
tra is the sonic form of a god. Jan Gonda also makes the point that a mantra that
contains the name of a god ‘is indeed regarded as embodying the energy of the
god which is activated by pronouncing the formula.’5 This sound form of the deity
is generally imparted to the disciple at initiation at which time the master (guru)
empowers the mantra with divine energy (mantravīrya). The master illuminates
the energy of mantra, says the Mālinīvijayottara-tantra,6 thereby bringing it to life
and making it efficacious.
While Tantric traditions entail different metaphysical systems, they all share
the basic assumptions about mantra and the practices associated with it. Mantra
pervades Tantric practice from esoteric meditation on the letters of the alphabet
to repetition of mantras that accompanies the general ritual sequence of purifica-
tion of the elements of the body (bhūtaśuddhi), the construction of a divine body
through imposing mantras upon it (nyāsa), inner or mental worship involving the
visualisation of the deity, and outer worship where the deity is offered external
substances such as flowers and incense. By repeating mantras the practitioner is
attempting to gain both power and liberation in the belief that repetition will make
his mind conform to the mantra and so imbibe its power. Mantras can thus be used
for magical or daily purposes, as Alper has observed, such as locating lost cattle,
and also for redemptive purposes such as ‘escape from sam . sāra, the diminution
of the effect of bad karma, transportation to the realm of the god to whom one is
devoted.’7 Mantras have been used for magical purposes such as attempting to
kill enemies or attract women, and soteriological purposes such as uniting with
the pure consciousness of Śiva in non-dualistic Śaivism or become equal to Śiva
(śivatulya) in the dualistic Śaiva Siddhānta.
Mantras have been used not only in ‘external’ ritual but also in yoga and med-
itation where they are linked to the breath (prān.a). This would seem to be an
ancient practice and is found in Tantric Upanis.ads such as the Dhyānabindu and
the Yogaśikhā, where the breath is linked to the mantra ham . sa. This comprises two
syllables HA and SA which are thought to be produced naturally in the process
of breathing and when combined can make the phrase aham sah., ‘I am he’ and
 Foreword xi
so’ham, ‘he is me’, indicating the practitioner’s identification with Śiva. Further-
more, the natural respiration is thought to be automatically repeating this mantra
in a ‘recitation of the non-recited’ (ajapājapa).8
Behind these practices, even within traditions that have divergent metaphysics,
a model of the mind is entailed that consciousness conforms to its objects: the
nature of consciousness is such that it is purified by focussing on a pure object or
conversely made impure through focussing on an impure object (driven by anger
or lust). Perception, usually driven by desire, reaches out into the world to grasp
objects and then conforms or is formed by them. This is not dissimilar to the medi-
eval Christian idea of extramission in which perception grasps its objects through
the eye actively reaching out and emitting a ray into the world.9 Through repeat-
ing a mantra which has been brought to life by the master and filled with divine
power, the disciple’s mind is grasping a pure object, thereby becoming pure itself.
This is an ancient idea attested in the Yoga-sūtras where yoga is understood as the
process of redirecting the mind away from this habitual conformity to objects of
consciousness that it calls ‘the fluctuations of consciousness’ (cittavr.tti).10 This
fundamental idea seems to be unchanged in the Tantric period and helps us under-
stand the way in which mantras are believed to be efficacious.
One of the major issues is whether mantras are, in fact, language. On the one
hand we have the view that mantras have illocutionary force in that to utter a man-
tra is to perform a speech act, while on the other we have the view that mantras
are not speech acts at all for they are not meaningful. Frits Staal has presented a
systematic argument that vedic mantras are not language in the sense of conveying
meaning and therefore cannot be understood in terms of semantics.11 Mantras are
connected to ritual which certainly has a structure or syntax but has no semantics,
being a kind of evolutionary leftover from a pre-linguistic stage of human evo-
lution. On this view, what is important about mantras is their rhythm and their
location within a ritual structure, not their meaning. While this is not the place
to outline Staal’s argument or to systematically present any counter-argument,
the material presented in this book lends weight to the idea that mantras must be
understood in the context of human meanings and the worldviews they inhabit.
While many mantras in themselves do not have direct linguistic meaning, they are
used within a meaningful human context of intention and purpose, the achieving
of particular human goals and the expressing of hope for power or liberation in this
or some other life.

The history of mantra

Mantras have been used for a very long time. They are attested from some of the
earliest documents of religious history, namely the hymns of the R.g-veda, and
may go back millennia before that. Mantras are attested in mainstream brahmani-
cal tradition and are central to brahmanical ritual, they are pervasive in Tantric
traditions (the subject of the present volume), and exist in the other indic religions
of Buddhism and Jainism. Mantras travelled to Indonesia, China, and Japan and
now to the West and related practices are found in other religions – dhikr in Islam
xii Foreword
as Eliade observed,12 and prayer, such as the Jesus Prayer, in Christianity. But it
is in the pre-medieval and medieval Tantric traditions that mantras became par-
ticularly important in religious practice, with systematic teachings about mantras,
the mantraśāstra, being developed. Indeed a name for the Buddhist Vajrayāna is
mantrayāna, a term that could equally be applied to Hindu Tantric traditions.
The history of Tantrism has been well documented by Alexis Sanderson in a
number of publications13 and André Padoux himself has written a masterful sur-
vey.14 A general picture emerges of traditions developing in the medieval period
that held to a different source of religious authority other than the primary revela-
tion of the Veda and the secondary revelation of the epics and Purān.as, namely
the Tantras. A distinction was therefore made between a follower of the Vedas,
a vaidika, and a follower of the Tantras, a tāntrika, a distinction that operated
across sectarian divides of Śaivism, Vais.n.avism and Śāktism. Within these emer-
gent Tantric traditions were varying degrees of conformity to vedic values and
traditions with extreme Kaula and Kāpālika groups, on the one hand, rejecting
vedic, brahmanical values as mere social convention and inhibition that had to be
transcended for a higher enlightenment, to more conformist traditions such as the
Śaiva Siddhānta, on the other, maintaining the compatibility between vedic and
Tantric revelation. Indeed, intellectuals in Kashmir such as the Nyāyika Jayantha
Bhat. t. a, writing apparently from his prison cell, towards the end of the ninth cen-
tury, defended the Tantric revelation but within the boundaries of reason and vedic
values.15
Within Śaivism, by the time of Abhinavagupta (c. 975–1025 C.E.) as Sander-
son has outlined in his publications, alongside orthoprax and orthodox vaidika
tradition whose followers were known as Maheśvaras or Rudrabhāktas and who
worshipped deities according to the Purān.as, we have, technically, the teachings of
Śiva (śivaśāsana) followed by the Śaivas who have undergone an initiation (dīks.
ā). The Śaivas themselves can be distinguished between the Ati Mārga, the higher
or outer path of the Pāśupata ascetics, and the Mantra Mārga, the path of mantras
that can be divided into the Śaiva Siddhānta and non-Saiddhāntika groups, par-
ticularly the Trika, Krama and Kaula groups. Both the Śaiva Siddhānta and non-
Saiddhāntika groups revered different Tantras as revelation and differed as regards
metaphysics. The Śaiva Siddhānta were dualists or pluralists who maintained an
absolute ontological distinction between Lord (pati), self (paśu lit. ‘beast’, ‘cow’)
and world (pāśa, lit. ‘bond’, ‘noose’) whereas the non-Saiddhāntika groups tended
to be non-dualists, maintaining that self, world, and Lord are ultimately identical,
the world being identical with consciousness (sam . vit, cit, caitanya) or an emana-
tion, a vibration (spanda) of consciousness. These non-Saiddhāntika groups tended
to be non-dualistic also in their ritual practice, denying a distinction between the
pure and the impure and using impurity in a ritual context such as the consump-
tion of alcohol, the ritualised use of sex outside of caste restrictions, and offering
meat to often ferocious female deities. A similar pattern can be found among Vais.
n.avas with the Pāñcarātra being the Tantric wing of the Vais.n.ava tradition rever-
ing a body of Tantric texts generally called Sam . hitās, although here the element
of impurity is lacking. The Śākta traditions similarly develop in the early Tantric
 Foreword xiii
period with the Śrī Vidyā focussed on the Goddess Tripurasundarī and generally
conforming to vedic values, especially with the adoption of the tradition by the
Vedānta and the Śan.karācārya of Śr. n.gerī.16
All of these groups use mantras which are a defining feature of different Tantric
traditions. There is, for example, a classification of Śaiva traditions flowing from
the five mouths of Śiva in his form as Sadāśiva, namely Īśāna, Tatpurus.a, Aghora,
Vāmadeva and Sadyojāta, each of these facing one of the directions and from
which a Tantric tradition flows.17 Similarly, as Padoux describes, Kaula traditions
or transmissions (āmnāya) are based on the directions, each transmission associ-
ated with different deities and different mantras. Thus, for example, in the eastern
transmission (pūrvāmnāya), Kuleśvara and Kuleśvarī are worshipped surrounded
by the eight Mothers, Brahmī and so on.18 While these classifications are idealised,
what is notable in all this is the centrality of mantras and the identification of the
deities of the traditions with mantras. Indeed, as Padoux shows in the present vol-
ume, great emphasis is placed on mantras in all of our texts, with details of how
to form them, how to use them, and their fundamental nature as identical with or
emanating from a deity. Thus Abhinavgupta in the Tantrāloka identifies speech
(vāc) as the essence of a deity and as power (śakti) expressed through the Sanskrit
language. Thus the alphabet from the letter a to ks.a expresses the inner nature of
the Lord, outside of time.19 While these traditions somewhat declined during the
colonial period, they have not completely disappeared and the old traditions of
mantra are still current in modern India along with the revived or more recent uses
of mantra by modern gurus, particularly as they have come to the West. In Kerala,
for example, mantras are used by the Tantris, orthodox Nambudri Brahmins, in
temple worship and they are also used at a popular level in the mantravādam to
cure disease and snake bites.
André Padoux presents us here with important studies in mantraśāstra which
are brought together for the first time in this book. The importance for Indol-
ogy and the study of religions of his meticulous work cannot be underestimated.
Through focussing on the mantraśāstra, Padoux has demonstrated in detail how a
concept and practice has gripped the imagination of a civilisation. In the follow-
ing pages the reader will discover the Hindu Tantric understanding of mantra and
the implications this work has for future research. We cannot understand Indian
religions without understanding the place of mantra and the treatment of the Hindu
Tantric systems presented here also needs to be done with Buddhist and Jain tradi-
tions as well. There are promising avenues of inquiry for comparative religion,
for in understanding human spiritual practices we need to understand the place of
verbal utterance (mantra, prayer, dhikr) in their ritual and meditational contexts
across religious traditions and thereby deepen our understanding of ourselves and
the institutions we have created and inhabit. But that is a task for another day.

Notes
1 Padoux, A. ‘Mantras – What Are They?’ p. 295. In Harvey Alper (ed.) Understanding
Mantras (Albany: SUNY Press, 1989), pp. 295–318.
xiv Foreword
2 Monier-Williams, M. Sanskrit-English Dictionary (Oxford: Clarendon Press, 1899),
pp. 785–86.
3 Padoux, A. Comprendre le tantrisme: les sources hindoues (Paris: Albin Michel, 2010),
p. 176.
4 Padoux, A. Vāc: the Concept of the Word in Selected Hindu Tantras, trs by Jaques Gon-
tier (Albany: SUNY Press, 1990), p. 85.
5 Gonda, Jan. Vishnuism and Shivaism: A Comparison (London: Athlone Press, 1970), p.
67.
6 MVT 2.10.
7 Alper, H. ‘Introduction’, p. 7.
8 Padoux, A. Comprendre le Tantrisme, p. 183.
9 Tachau, Katherine ‘Seeing as Action and Passion in the Thirteenth and Fourteenth
Centuries,’ p. 337. In Hamburger, J.F. and A.-M. Bouché (eds) The Mind’s Eye: Art
and Theological Argument in the Middle Ages (Princeton University Press, 2006), pp.
336–59.
10 Patañjali Yoga-sūtras 1.4 trs with commentary by Hariharānada Āran.ya. The Yoga Phi-
losophy of Patañjali, trs by P.N. Mukerji (Albany: SUNY Press, 1988 (1963)).
11 Staal, Frits. Discovering the Vedas: Origins, Mantras, Rituals, Insights (New Delhi:
Penguin, 2008), p. 191. ‘Mantras are not mysterious statements with deep meanings.
They have no meanings because they are not language.’ For some discussion of the
relation between mantras, language, and meaning, see Alper, H. (ed.) Understanding
Mantras (Albany: SUNY Press, 1989); Yelle, Robert A. Explaining Mantras: Ritual,
Rhetoric, and the Dream of Natural Language in Hindu Tantra (Routledge, 2003)
12 Eliade, M. Yoga: Immortality and Freedom, trs W. Trask (Princeton University Press,
1970), pp. 216–19.
13 Sanderson, A. ‘Śaivism and the Tantric Traditions’, in Sutherland, S. et al. (eds) The
World’s Religions (London: Routledge, 1988), pp. 660–704; ‘The Śaiva Age – The Rise
and Dominance of Śaivism during the Early Medieval Period’, in Shingo Einoo (ed.),
Genesis and Development of Tantrism (Tokyo: University of Tokyo, Institute of Orien-
tal Culture, 2009), pp. 41–349.
14 Padoux, A. Comprendre le tantrisme: les sources hindoues.
15 Jayantha Bhat. t. a. Nyāyamañjari. English translation by V.N. Jha (Delhi: Śrī Satguru
Publications, 1995), p. 562.
16 On the development of the Śrī Vidyā, see Padoux, A. Le Coeur de la Yoginī: Yoginīhr.
daya avec le commentaire Dīpikā d’Amr.tānanda (Paris: de Boccard, 1994), pp. 29–34.
17 Hanneder, J. Abhinavagupta’s Philosophy of Revelation: Mālinīślokavārttika I, I-
399 (Groningen: Egbert Forsten, 1998), p. 16 quoting Abhinavagupta’s Tantrāloka
15.203cd–206ab.
18 See Padoux, A. Le Coeur de la Yoginī, pp. 35–40.
19 Padoux, A. Comprendre le tantrisme, pp. 166–67.
1 Introduction

The medium is the message.

(Marshall McLuhan)
donner un sens plus pur aux mots de la tribu.
(Stéphane Mallarmé)

Why study Tantric mantras? One may well ask. Why Tantra? Why mantras? The
answer is that Tantra, the Tantric phenomenon, is not, as was first believed, a
limited and bizarre form of Hinduism (or Buddhism) but a fundamental aspect
of the Indian religious world. It has pervaded the near totality of Hinduism (and
of part of Mahayāna Buddhism) since, perhaps, the fifth or sixth centuries: reli-
gion, ritual, theology and metaphysics, iconography and temple building, even
the structure of the state, in India and in part of Asia, would not have been the
same without Tantra. The first ‘non-human’, purely verbal, theoretically eternal,
self-proclaimed revelation of the Veda, the śruti (‘what was heard’, the Vedic rev-
elation being oral), which goes from the Veda to the Upanis.ads and so forth, and
is followed by Veda-based texts (the smrti), has in some respect always had the
pride of place: it forms the basic orthodox˚ socio-religious teaching of Hindu soci-
ety. Another, scriptural, revelation, that of the Tantras (texts deemed to have been
revealed by deities), came later on and progressively pervaded most of the Hindu
world. It did not so much reject the other revelation as depreciate it, considering
it to be of merely social value and unable to offer the same access to liberation
or other rewards as the esoteric Tantric teaching.1 And it progressively permeated
and often transformed the orthodox Hindu world. One may say without exaggera-
tion that since at least a thousand years Hinduism has been very largely Tantric or
‘tantricised’, not Vedic. The Tantric revelation cannot therefore be treated as sec-
ondary. Even though many of its texts, scriptures and exegetical works are arcane
and obscure, the presence of the notions and practices it brought forth, together
with its vision of the world, is to be felt everywhere in Hinduism – possibly also
because most of these notions and practices have deep roots in the Indian soil.
As for mantras, they play, as we shall see, an essential role in Tantra. There
is no possible study or understanding of the Tantric phenomenon without an ex-
amination and understanding of the peculiar ritual use of forms of speech that are
2 Introduction
mantras. Studying them is therefore entirely justified. But before taking up the
subject of particular mantric practices, some precision on Tantric mantras in gen-
eral, and an attempt at defining their nature and role as they appear in Tantric
texts, is appropriate. Over the course of time, the Sanskrit term mantra was used
for different sorts of verbal or phonic ritual utterances. (I use ‘utterances’ here
because mantras are by nature oral, aural – not written.2) The earliest mantras are
the Vedic hymns, poems or chants. They differ greatly from Tantric (and even
Pauranic) ones – even though some are occasionally used in Tantric rites (in those
of the Pāñcarātra mainly). But are also to be found in the Vedic ritual meaningless
syllables, the stobhas, which are akin to Tantric mantras; among these is OM . , an
utterance of paramount importance in Tantric practice and theologico-metaphysi-
cal speculations. The sort of mantras developed in śrauta texts (that is, those based
on the Vedic revelation, the śruti) or those used in later Veda-based texts, the smrti
(those of the smārta or paurān.ic form of Hinduism), are also used in Tantric Hin- ˚
duism.3 Tantric mantras deserve a special study because, however important ritual
formulas may have been in Vedism and in later Veda-based Hinduism, the role
there of mantras was never as great as in Tantric Hinduism, where they came to
pervade all types of religious or ritual action and gave birth to an enormous mass of
literature concerning their nature and uses and extolling their power – an ensemble
usually called mantraśāstra, the science or doctrine of mantras. The term is not
of very ancient origin, but is a current and convenient one. It is an important one,
too, since it sometimes came to be taken as synonymous with tantraśāstra, also a
modern term, often used to refer to the Tantric ensemble of works and doctrines: a
‘rapprochement’ which is one more proof of the fundamental place of mantras in
the Tantric domain. The importance of these formulas in the Tantric world is also
shown by the fact that an early form of Śaivism was named mantramarga, and
that a traditional division of Śaiva tantras is between those of the mantrapīt.ha
and those of the vidyāpīt.ha (vidyās being female mantras, whereas a mantra is that
of a male deity4). In fact, all Tantric texts deal either entirely or in part with man-
tras, a literature whose first elements date probably from the fifth or sixth century
AD, and which went on for centuries and is still productive today. Tantric mantras
are also to be found in a number of non-Tantric texts, in several Purān.as, for in-
stance, and even in some treatises of dharmaśāstra, that is, works concerning the
laws, rules of conduct and customary observances of the Hindus.
Mantras also play an important role in Buddhism, the Tantric form of which is
often called mantrayāna. Mantras as used in the Buddhist context in India, Tibet,
South East Asia and the Far East are a vast and very interesting subject, very well
worth careful and comprehensive study. We will, however, not deal with it here,
where only the Hindu aspect of the Tantric phenomenon is considered.
A characteristic feature of Hindu Tantric mantras, which differentiates them
from non-Tantric ones, is the prevailing use of non-linguistic elements, that is,
syllables or group of syllables devoid of meaning but deemed to be imbued with
supernatural power and efficacy. These are usually added at the beginning of the
mantra (after the initial OM . , which, except – though not always – in the case of
bīja/bījamantras, is always there) or before the final ritual exclamation (the jāti 5)
 Introduction 3
of the mantra. A mantra may also be a sound sequence entirely made up of one
or several such syllables; for example, the case of SAUH., the mantra of Parā, the
supreme Goddess of the Trika, or the Navātmamantra HSKS.MLRVYŪM . , or the
fifteen-syllable mantra of the goddess Tripurasundarī, which, in its more usual
form, runs HA SA KA LA HRĪM . HA SA KA HA LA HRĪM . SA KA LA HRĪM . . The
whole Sanskrit alphabet, as well as each of its constituent phonemes, either in the
‘normal’ traditional order (the varn.asamāmnāya, called in that case śabdarāśi or
mātrkā), or in the nādiphānta order of the Mālinī,6 are also considered and used
˚
as mantras. 7

The presence of such syllables as constituent parts of mantras can be explained

in many ways. But it has also been theoretically grounded in the idea that all
syllables being Sanskrit phonemes are forms of vāc, the Word, the primordial cos-
mic power, extolled in the Veda in Sanskrit. They are therefore all imbued with the
infinite power of vāc, which in the Veda was identified with the brahman: brahma
vai vāc (Aitareyabrāhman.a 4.21,1), ‘as much as brahman spreads out, as large is
the Word’ (Rgveda 10.114,8), a notion that has survived in Tantric traditions. In
Tantric texts,˚ Sanskrit phonemes are often called mātrkā, a term translated as ‘little
˚
mothers’. But these are presided over by, and are discrete forms of, the Mātrkā, the
divine Mother in her aspect as vāc, the supreme power governing the world,˚ hence
their force and efficacy.8 However, the power of mantras is not confined in any
one of these phonemes: it pervades the whole formula. A traditional Tantric saying
is mantrāh. varn.ātmakāh. sarve, varn.āh. sarve śivātmakāh., ‘All mantras are made
of [Sanskrit] phonemes, the nature of all phonemes is that of Śiva.’ This principle
does not prevent speculation on the particular functions and forms of efficacy of
certain parts of the mantra, as we shall see later. One may add here that, contrary
to the notion we have just seen that mantras are born from the Sanskrit phonemes,
several tantras consider the bīja OM . or bindu (the sonic dot, M. ) as the source of
the alphabet (and hence of the mantras and of the cosmos).9
Curiously, considering the absolute privilege of Sanskrit in mantras – which
are Sanskrit ritual formulas10 – one finds in some Tantric works the notion that
inarticulate sounds, or ungrammatical Sanskrit, or Prākrit syllables (whether used
in mantras or otherwise) may be considered nearer to the Absolute than Sanskrit
phonemes because they are farther from the norm of articulate rule-governed
speech, a speech that can be used for mundane, inferior, purposes, which is not
the case of mantras since these are linguistic utterances, but not speech. This view
was put forward by Maheśvarānanda (a Krama author of the twelfth century)
in the commentary of his Mahārthamañjarī. The notion was upheld earlier by
Abhinavagupta, the great Tāntrika master of Kaśmir (fl. c.975–1025), who wrote
in the Parātrīśikavivaran.a – PTV – ‘The supreme Lord himself has explained
that an unmanifest sound is generally somewhat like a mantra’ (p. 189).11 He also
said that the power of mantras rests essentially on the unutterable and inaudible
‘parts’ (kalās) of the subtle phonic vibration that follow the bindu in the uccāra12
of OM . (or of other bījas ending with M. ) and through which the subtle sound, the
nāda, of the mantra progressively dissolves into the absolute. Such statements
may seem to contradict the privileged role of sanskrit phonemes in mantras; but
4 Introduction
all phonemes, whether Sanskrit or not, are deemed to be born from the primordial
Word whose original, highest, aspect is voiceless, a voiceless Word pregnant with
all the possibilities of speech, but silent. Silence, in fact, was always held in India
to be superior to uttered word or audible sound, a notion that goes back to vedism:
in the Brāhman.as, the anirukta, the indistinct speech, was related to the Limitless
and its power. The superiority of the unuttered over the uttered is an ancient and
pervading notion in Indian thought. This extolling of silence explains the tradi-
tionally acknowledged superiority of the inaudible, secret (upām . śu), and still more
of the silent, mental (mānasa) mantra enunciation or repetition (japa) over the
audible ‘voiced’ (vācika) one.13
But the very nature of mantras, and not merely their utterance, is often pro-
claimed, extolled even, as being not only mental but also spiritual, immaterial both
in their essence and in their modus operandi. This appears in the saying quoted
on p. 3, but also, for instance, in a traditional semantic analysis (nirvacana) of
the term mantra often quoted in Tantric texts. The Sanskrit term mantra, accord-
ingly, is said to be made up of the verbal root MAN (to think) and the suffix tra
(used to make words denoting instruments or objects); a mantra would thus be an
instrument of thought – not of ordinary discursive thought, of course, but of an
intense, effective, non-discursive one.14 A traditional analysis of the term more
often found in Tantric texts relates, however, the suffix tra to the root TRAI (to
save or protect), an explanation which understands therefore mantras as forms of
protective or salvific thought. The two interpretations of mantra coalesce some-
how in the oft-quoted formula: mananam . sarvaveditvam . trān.am. sam. sārasagarāt/
mananatrān.adharmatvān mantra ity abhidhīyate (‘A mantra is so-called because
it is in the nature of thought and protection, thought being omniscience, and pro-
tection [the release] from the ocean of transmigration’). This would not apply to all
mantras, for many are mere ritual formulas: powerful, effective, aimed at specific
ends, but not specifically protective or salvific. It is, however, valid insofar as it
expresses an important aspect of the nature of mantras: their efficacy being sym-
bolic is mental, whilst the power they are believed to possess is deemed to be of an
extra-mundane origin, and is always – if sometimes indirectly – transcendentally
oriented.
Mantras in general, but especially in the Tantric domain, have different forms or
aspects: they are considered as existing and acting on different planes, and are put
to various uses, both transcendental and wordly. They thus appear to us as having
– and they do have – an ambiguous nature. This, however, appears all the more
because of the diversity of mantraśāstra, which includes texts and speculations of
different traditions or schools, extending over fifteen centuries; a variety which
we tend to ignore, taking as a homogeneous body of doctrine what is in fact a
conglomerate of often very differing views. This point is to be clearly underlined:
there is in India no systematic non-contradictory theoretical or practical view of
Tantric mantras. A synthetic presentation of the subject such as this one cannot
therefore be more than an overview of different conceptions and practices – and
even so it cannot altogether escape the risk of being at least partly an intellectual
construction of mine.
 Introduction 5
This being so, we can nevertheless usefully, if not construe a theory of mantras,
at least try to describe a number of characteristic traits of Tantric mantras.
The first point to be underlined is that mantras, basically (and this is a generally
held view), have a twofold nature: transcendent and empirical. On the one hand,
they are conceived either as being in essence forms of vāc, the Word, the supreme
unique cosmic Power, or as being individualised powers, discrete supernatural
entities,15 being in that case the highest aspect, the essence of deities or of specific
aspects of deities; being even by themselves, intrinsically, deities – as is some-
times said, Tantric pantheons are pantheons of mantras rather than of divinities.
Mantras also represent, symbolise, for ritual use, cosmic entities. They are also
conceived as discrete supernatural phonic powers, ritually or meditatively utilis-
able to various ritual or magical ends, their power being that of their phonetic
content: they are words of power. But, on the other hand, mantras in so far as
they are ritual formulas invoked, held in their minds or uttered by human beings,
acting on (and through) their thought and body, exist mentally and phonetically
on the level of the empirical world; they are sometimes made up of words or
meaningful sentences (they may even sometimes be written). There are, too,
many yogic somato-psychic practices where mantras are imagined, sometimes
visualised, as present in the body of the performer, acquiring thus a quasi-
physical or physiological presence whilst still being considered as transcendent
supernatural entities.16 Whilst they are perceived as human, they are at the same
time divine, and deemed to be effective on different planes of the cosmos. This,
one could say contradictory, nature of Tantric mantras appears clearly in many
of their ritual, meditative, mystical or magical uses. The double nature or aspect
of mantras was (in the non-dualist Śaiva perspective of the Kashmirian Trika)
expressed in Abhinavagupta’s PTV: ‘Mantras, it says, are the Venerable [Power]
of the phonemes. They are in the nature of the world as well as of the supreme
Lord, and other [deities].They are thought and liberation. They consist of discur-
sive thought as much as of pure consciousness.’17 This view, in spite of its theoreti-
cal general validity, is, however, not equally shared by all Tantric traditions. Views
also evolve in the course of time, and thus this non-dualistic Śaiva conception, if
valid in the non-dualistic Śaiva domain or in later Saiddhāntika works influenced
by non-dualistic thought, is not the conception of earlier fifth to eleventh century
āgamas.18
Here, another more general point may be raised; namely, if Tantric mantras
are not only groups of (often meaningless) phonemes, but consist sometimes of
words or sentences, how does one distinguish the sentences, words or syllables
that are mantras from those that are not? The answer is that there are no for-
mal criteria to define mantras: mantras are those ritual formulas or utterances that
are pronounced as such by the tradition to which they pertain. There are indeed
formal traits proper to mantras, notably the initial OM . and the final exclamations
– the so-called (originally Vedic) jātis: namah., svāhā, vas.at., vaus.at, hum . and
phat. – but these are not always there. The lists of tantric mantras or bījas,
Daks.in.āmūrti’s Uddhārakośa, for instance, enumerate these ritual formulas but do
not give any theoretical definition. When mantras are described (as we shall see
6 Introduction
in chapter 6), the aim is usually not to define them, but mainly to enumerate and
extol mantras of a particular tradition and to disparage or condemn those of other
schools or traditions. There exist, too, classifications of mantras according to their
length, to their nature, or to how and where they can be used – of these we shall
see a few instances in chapter 6. Tantric mantras are also deemed to have six
characteristic elements: a seer (a rs.i) believed to have seeen it first, a particu-
lar meter (chandas), a presiding deity˚ (devatā), a seed (bīja), one of its syllables
(sometimes the first one) and an energy (śakti) (usually one of its phonemes).19 To
which is sometimes added a ‘wedge’ (kīlaka), which is considered as making it
more efficient.
Let us come back to the different uses of Tantric mantras, and to the conceptions
associated with such uses. Mantras are ritual formulas; their main use is in ritual.
There is no ritual without mantra. All Tantric ritual actions are accompanied by, or
consist mostly or entirely in, the (vocal or mental) uttering of mantras. Sometimes,
when a mantra is uttered, a ritual gesture, a mudrā, is to be displayed, accompany-
ing but not replacing the action of the mantra.20 Such, as we shall see, is the case
in nyāsa, but elsewhere also. However, and this is important, mantras can be used
without any outside visible ritual action being performed. This happens, evidently,
in japa21 and in all meditative mantric practices as well as in the mental worship,
the āntara or mānasa pūjā performed before the outer, ‘gross’ pūjā, which con-
sists solely in the uttering of the mantras corresponding to the different parts of the
ritual worship of the deity. So important is the role of mantras in the Tantric pūjā
that one may well say that the pūjā is nothing but a play or display of mantras (and
visualisations). The mantras used in this way in ritual are supposed to be directly
effective. They do what they say: they purify, they protect,22 they cut, they trans-
form (water into amrta, for instance, or the body of the performer into a divine
one, or into one made ˚ of mantras, a mantradeha), they bring about the presence of
a deity in an icon23 or on a man.d.ala, etc.24 In this capacity, mantras are used in ritu-
.
al to build up, constitute, an icon – the linga used in Śiva’s pūjā, for instance, is to
be imagined by the officiant as the throne (āsana) of Sadāśiva: a cosmic imaginary
structure is thus built up, all elements of which are to be mentally placed, using the
.
mantras embodying these cosmic elements, on the linga which, in this respect, is a
mere vertical piling up of mantras, but of mantras as cosmic entities.
However, when used in the course of a ritual process, mantras cannot all be
considered as having the same status or nature. In most cases they are used as
supernatural forces of different power able to effect specific tasks. This statute
is reflected in the Śaivasiddhānta terminology where they are usually called an.u,
‘atoms’, that is, limited spiritual entities. Mantras are also spiritual supernatural
entities in the system of the seven subjects/agents of cognition, the pramātr
(already quoted above), which extends from Śiva, the supreme one, to the ‘great˚
lords of mantras’, the mantramaheśvaras and the ‘lords of mantras’, mantreśvaras,
powers on a somewhat lower plane, then to the mantras numbering 70 million, all
of which are deemed to help humans on their way to salvation.25
When mantras bring about the presence of a deity – in such rites as āvahana or
prān.apratis..thā – when they embody a deity in their phonetic structure, they do not
 Introduction 7
merely participate in the nature of that deity: they are the very essence, by nature
supreme and unlimited, of that deity. Tantric traditions express this by saying that
they are the vācaka, that is, what expresses, enunciates, and therefore causes the
existence of the deity; the latter being as such (in its nature and form) ‘what is
to be expressed’, vācya, by the mantra. In this respect mantras are higher than
deities.26 But they are also, as already said, themselves deities. We see here two
different visions of mantras, but visions which, though different, cannot, I believe, be
entirely separated, for they are two inseparable faces of the same, ambiguous, real-
ity.27 During the same ritual, a pūjā for instance, mantras are in some cases mere
effective words of power, in other cases, the phonic essential form of a deity or of
a deity’s bodily form or parts.28 Such is also the case in the Śaiva initiation: some
mantras purify the ritual area, others act on the initiand’s body; for instance, by
‘cosmicising’ it by placing on it the different parts or ‘worlds’ (kalās or bhuvanas)
constituting the cosmos (which, ritually, all have/are mantras). But other mantras,
in the same ritual, appear and ‘function’ as divinities: in the initiation ritual of the
sādhaka, the nirvān.adīks.ā, for instance, the man.d.ala in front of which the blind-
folded initiand is brought is said to be blazing with the light of the mantras placed
on it – these mantras are luminous because they are deities. In the ritual worship,
the pūjā, the five brahmamantras – Sadyojāta, Aghora, Tatpurus.a, Vāmadeva
.
and Īśāna – placed on the linga are forms of Sadāśiva (visibly represented by a
.
human face when the icon is a mukhalinga). What is more, the same bījamantra
may be the mantra of a deity and be invoked as such, but it can also be used during
a ritual or mental process as the support of mental representations: this happens,
for instance, in some uses of OM . or of SAUH..
29

We must note here that in so far as Tantric mantras are deities, they, like all dei-
ties, have a visible form that the adept is to imagine mentally, which he does in
such cases as when he is to ‘see’ a mantra placed on or in his body, on an icon or on
a man.d.ala. This explains why the compendiums of Tantric iconography are works
of mantraśāstra. Their aim is to give the descriptions of the deities to be visualised
(dhyāna30) in ritual, each description beginning always with the uddhāra:31 the
description of how to ‘extract’ – compose, that is – the mantra of that deity.32
A particular and important case is that of the mantra given to an initiate who
wishes to become a sādhaka, a reward- or power-seeking (bubhuks.u) initiate. This
mantra is the means whereby he will reach such aims as he wishes to attain. He
must therefore be able to master it so as to use it for his own purpose. This is done
through a complex and sometimes long ritual process during which the mantra is wor-
shipped daily as a god or a goddess: a mantrasādhana also named puraścaran.a,33
‘preliminary ritual’. The mantra is thus a means to a particular end (mundane or
otherwordly),34 but it is also an autonomous supernatural power to be mastered (a
sādhyamantra). Such mantras must evidently be perfectly adapted to their user
and to the end he wishes to attain: hence the practices of selection (mantravicāra)
described here in chapter 2.
Mantras, we must not forget, are also used in magic. The well-known six Tantric
‘magic acts’ – the s.at.karmān.i, killing, subjugating, causing dissensions, attracting,
etc. – are all done with mantras.35 These are also used in amulets or written on the
8 Introduction
body to various magical ends: inferior uses, of course, but ancient ones, and still
present nowadays – superstition dies hard.
More interesting than such ritual uses of mantras (some of which we shall see in
the chapters in this book) and perhaps more worth mentioning in this introduction
are some speculations on the nature of mantras, and on how they function, that are
to be found in Śaiva texts of Kashmir. These are put forward especially in works
of two particular non-dualistic schools – the Trika/Pratyabhijñā and the Spanda
– and are therefore proper to these schools and must not be taken as generally
valid. They, however go deeper than many others in the analysis of how, within
Tantric traditions, mantras may be conceived and are put into use, of the source
of their power, and of how they can be imagined to function. They shed thus an
interesting light on some aspects of the subject and deserve therefore to be, if
briefly, considered here.36
The basic metaphysical tenet of these non-dualistic schools is that the ultimate
supreme Reality is the divine absolute and omnipotent Consciousness which merely
appears as separate individual consciousnesses. In truth, in these ‘sam . vidadvaya’
systems, nothing exists but Consciousness. The essential nature of mantras being
divine – that of vāc – is therefore Consciousness: it is spiritual power, and it can
only be used effectively by an adept if he/she gets hold, masters, this spiritual
energy, this power of mantras, mantravīrya as it is sometimes called. The nature
of the mantra, in this perspective, is summed up in sūtra 2.1 of the Śiva-Sūtras
(ŚS), the basic text of non-dualistic Śaivism, which runs: cittam . mantrah., ‘The
mantra is consciousness.’ Consciousness, Ks.emarāja explains in his commentary
(Vimarśinī), is that through which one becomes aware of the supreme Reality, that
by which one realises one’s identity with the Lord. A mantra, he adds, ‘is not a
mere conglomerate of different syllables. It is the very mind of the devotee who,
through intense awareness of the deity of the mantra37 (mantradevatāvimarśapa-
ratvena), is fused, identified with that deity.’ Ks.emarāja then quotes two tantras
which affirm clearly the consciousness nature of mantras and the fact that they can
be grasped, and put to use, only through the consciousness of the user, the mantrin,
whose ultimate nature is also that of the divine omnipresent and eternal Power, the
supreme śakti.
The next sūtra, 2.3, of the ŚS is: vidyāśarīrasattā mantrarahasyam, ‘The
Being whose body is [pure] Science or knowledge, such is the secret of man-
tras.’ Ks.emarāja explains that vidyāśarīrasattā refers to the essence of the su-
preme non-duality as it expands, its ‘body’ being then made of the totality of
sounds (śabdarāśi) – the whole sanskrit alphabet (or Bhairava as alphabet), that
is – ‘whose essence is the flashing vibration (sphurattā), the awareness of the
fullness of the absolute I, the undivided totality of the universe. Such is the se-
cret of the mantras’.38 This identifies the essential inner nature of mantras with
the Absolute as the ever-expanding power of the supreme godhead. The same
conception is upheld by Abhinavagupta who, in his commentary (Vimarśinī) on
Utpaladeva’s Īśvarapratyabhijñākarikā (ĪPK), 1.5.14, says that the nature of
mantras ‘is the free activity of consciousness which is the energy of the supreme
Word’. This, he considers, is especially embodied in the so-called supreme great
 Introduction 9
mantra (paramahāmantra) AHAM . , ‘I’, which, embodying the self-affirmation of the
Absolute, brings together in its phonetic structure (A + HA+ M . ) all the phonemes
of the Sanskrit alphabet of which all mantras are constituted39 – a conception,
incidentally, whose roots are Vedic.
Interesting too as to how mantras function is a passage of the Spandakārikā
(SpK), 26–27, which runs: ‘Having seized that strength [of pure consciousness],
mantras, endowed with the power of omniscience, perform their functions as do
the senses of the embodied. There and nowhere else, they dissolve, quiescent and
spotless, along with the adept’s mind and thus achieve śivahood’.40 Mantras, thus,
though pure consciousness, can be used to attain different ends, even mundane
ones, while still retaining in essence their transcendental nature. This, however,
they are said to do ‘along with the adept’s mind’: not alone, but only insofar as
they are present in the adept’s consciousness – an important point. As is often said
‘the mantra alone, the mantrin alone are powerless’: they can only act together.
Hence the essentiality of the mental aspect of mantric practices – an aspect we
shall see several times in this volume. The inner nature of consciousness of man-
tras thus posited by these Śaiva schools explains the role they give to mental prac-
tices necessary to activate and make use of mantras. This, of course, takes place
in all mantrasādhana of whatever persuasion, but it is specially insisted upon in
those non-dualistic schools.41 Thus the ŚS 1.22 says that ‘by one-pointed attention
(anusam . dhānāt) on the Great Lake one experiences the power of mantras’; this
42

Lake is the supreme Consciousness on which the adept is to concentrate mentally.

The role of memory or rather of recalling to mind (smaran.a) is also insisted upon
by some authors, Abhinavagupta notably; a view founded on the doctrine of non-
duality of consciousness (sam . vidadvaya) for which, as is said for instance in the
ĪPK 1.4.1, ‘he who remembers is none other than the supreme Lord’. Such meta-
physical conceptions do not prevent these traditions from prescribing a number
of ritual practices for the efficient use of mantras, practices where, as we shall
see in this volume, concentrated mental activity, one-pointed attention, is present
and important.43 Mantric practices have necessarily a mental aspect. Mantras are
phonic entities, but their uses belong to the realm of representation and creative
imagination, their efficacy is symbolic, imagined, not real: one is here entirely in
the mental field. This, of course, being said without forgetting that the efficacy as
well as the definition and the uses of mantras, Tantric or non-Tantric, are socially
determined: mantras have no other legitimacy and power than those given them by
the social group where they are being used.
It would be fair to add that the question of the nature and efficacy of mantras
– especially of Tantric mantras – has also exercised the minds of Western scholars.
Frits Staal’s study, Ritual and Mantras – Rules without Meaning,44 reviews many
theories, comparing, for instance, mantras with the babbling of babies and with
bird songs. Mantras have also been compared with poetry since in both cases the
phonic substance of the utterance, of words, is essential. But poetry is language,
discourse (even if of a particular sort), which Tantric mantras are not. In mantras
‘la teneur prévaut sur le sens’, to quote Louis Renou: there can be no meaning,
merely the phonic content, whereas, except in the very limited case of ‘concrete
10 Introduction
poetry’, a poem always means, or suggests, or points towards something, which
the mantra does not do, or at any rate not in the same way – it may invoke a
deity or suggest an action, but the mantra does this through its form, its phonic
structure, not by ‘saying something’.45 Mantras do not function ‘nominatively’, to
designate something or describe an action, but ‘invocatively’, to invoke a deity,
or performatively, to bring about a situation, to ‘do’ something. Tantric mantras
are also sometimes seen as instances of the transcending of language towards a
higher, perhaps silent, level of speech, a view that can be reconciled with some
Indian conceptions.46 The term japa, the ritual utterance of a mantra, is sometimes
translated as prayer, which, in fact, japa actually is in some cases. I shall come
back to this point in chapter 3.
Tantric mantras being what I have more or less aptly tried to say, the selection
translated here of papers I wrote or read in the course of years on the subject deal
mainly with mantric practices, which is the main aspect of mantras: how they were
and are being made use of. Theory is secondary – im Anfang war der Tat – even if
it explains and often guides or orientates. Ideology is never a mere superstructure:
if mantras, too, ‘marchent à l’idéologie’, ideology never comes first.
Chapter 2 deals with the ritual practices for the creation and choice of mantras.
Then are seen two important mantric practices: nyāsa, the placing of mantras on the
body or on various supports (chapter 4), and japa, their ritual repetition (chapter 3).
Chapter 5 is on a form of japa: the repetition of a mantra using a rosary (aks.amālā)
as it is described in a Vais.n.ava text of the Pāñcarātra, the Jayākhyāsam
. hitā (JS). Then
are seen the defects or flaws (dos.a) mantras may suffer from and the ‘perfectioning’
rituals (sam . skāra) used to eradicate these defects and to make mantras more effec-
tive or more powerful (chapter 6). This will be completed by a review of some uses
of mantras and of practices meant to make them more effective (chapter 7), by a few
notes on how mantras are deemed to circulate in the body of the practitioner (chap-
ter 8) and, finally, on the relationship between the oral and the written in the case of
mantras (chapter 9); chapters where we will come back briefly on the question of the
nature of Tantric mantras as phonic and mental entities or representations. Other as-
pects of mantras or of mantric practices, some of which are mentioned in passing in
these eight chapters, could also have been tackled: it simply happened that I did not
study them carefully enough to deal specifically with them. One may add, however,
that for an exhaustive study of Tantric mantras, of their manifold uses and of all that
has been written on the subject, or done with them, in the field of Tantra, a much
larger volume than this one would have been necessary. It would quote more Sanskrit
texts, describe a larger number of facts and refer to other speculations. But mantric
practices and notions, like many, if not most, Tantric ones, even when interesting, are
often repetitive; a longer study, though more informative, would therefore probably
not add very much to the understanding of the particular and very typically Indian
phenomenon of mantra. A phenomenon which, in fact, is but an aspect of the ancient
and universal belief in the power of speech, or the Word, and especially of the near
universal superstitious belief in the power of ritual formulas and nonsense syllables:
a popular credence also sometimes philosophically refined (as we have just seen,
and will see here) in metaphysical schools of thought. But we too, when we use a
 Introduction 11
number of conventional formulas, believe in the performative or illocutionary force
of words. We should not therefore be too surprised by Tantric mantras for they are
merely a particular, specially intense, sometimes bizarre, case of a universal human
faith in the power of speech – speech, this remarkable activity uniquely proper to
humans – êtres de parole.47
12 The extraction and examination of mantras

2 The extraction and

examination of mantras1
Mantroddhāra and mantravicāra

The use of mantras being prescribed in different circumstances (in ritual, magic,
yoga, or spiritual and meditative practices) are quoted, revealed or prescribed in
different ways. In ritual, and especially in the course of the mandatory daily wor-
ship of a deity, but also in many other rites, such as dīks.ā, pratis..thā, pavitrārohana
and so forth, their utterance at various moments of the ritual performance is pre-
scribed by Tantras and other Tantric texts, or in compilations (nibandhas) or man-
uals (paddhatis) of ritual. The only condition for their use is that they are uttered
exactly and as prescribed, by a socially and ritually authorised person. Such has
been the case from Vedic times down to the present. It is, one may say, the normal,
ordinary form in which the utterance of these constituent elements of all rites is
prescribed, the form most often met with. It is, however, not the most characteristi-
cally Tantric and most interesting one. More interesting, and typically Tantric, is
another way, first of quoting or revealing a mantra, then of examining and select-
ing it so as to be sure it is well adapted to its user and to the end pursued. This
method is used whenever the mantra – which is the phonic, efficacious form of a
deity and is therefore not to be revealed to inapt or uninitiated persons – is to be
communicated to an adept or devotee. This form of revelation or secret (ritual or
non-ritual) communication of a mantra is what is called mantroddhāra, whereas
the method for examining and adapting a mantra is called mantravicāra, two
practices some forms of which we shall see presently.
A mantra being the phonic form of a deity, replete with all the deity’s power
and effectiveness, cannot be freely revealed to just any person. It is to be revealed
in a particular, secret (or at least indirect) way. Though being a divine entity, a
mantra is also in practice not merely a phonic entity, or a sound, but a verbal form,
made up of letters or phonemes, that is, made up of the constituent elements of
the Sanskrit alphabet, the varnas. Its revelation will consist therefore in extracting
the phonemes of which it is made from the ensemble of phonemes of the alphabet,
the varn.asamāmnāya, which, in this context, is not a mere collection of letters,
but the phonetic form of the absolute, of vāc, the Word, which, as is said in the
Rgveda, is the mother of all the gods.2
˚ One may say in a general way that the root or fundamental mantra (mūlamantra
or mūlavidyā 3) of a deity to be worshipped, or the mantra given to an adept for
a specific reason, to achieve a particular end, has to be selected in a particular
 The extraction and examination of mantras 13
way. The principle underlying this selection is that mantras are basic, effica-
cious forms or aspects of the Word, of vāc, which is the original, fundamental
ever-active form of the supreme godhead, its energy or power, śakti. To be made
use of, mantras, which are made of phonemes and/or words, must therefore be
taken out of, extracted from, this Word totality – hence the term and practice of
mantroddhāra.
The term mantroddhāra (or mantroddhāran.a) is made up of mantra and
uddhāra/uddhāran.a, from the verbal form ud-DHR (to draw out, extract or select).
Mantroddhāra is thus the extraction of the mantra ˚ out of the totality of sound or,
more precisely, out of the totality of the Sanskrit phonemes, the varn.asamāmnāya
(the alphabetical form of vāc4). It is at the same time the selection and composi-
tion of the mantra of a particular deity, or of the mantra to be given to a particular
person – the notion of extraction being, however, always essential.5 The various
forms of mantroddhāra are to be found in a large number of texts, mainly in the
‘digests of mantraśāstra’, to use T. Goudriaan’s words,6 such as the Prapañcasāra
(eighth century), or Laks.manadeśika’s twelfth century. Śāradātilaka, in manuals
or compendiums of Tantric ritual such, for instance, as Mahīdhara’s Mantrama-
hodadhi (sixteenth century), Krs.n.ānanda Āgamavāgīśa’s Tantrasāra (seventeenth
century), or the nineteenth-century˚ Mantramahārn.ava. There are also, as we shall
see, mantrodhāra(n.a) chapters in most Śaiva āgamas or tantras, in the Pāñcarātra
sam. hitās, and, of course, in various Tantric works, all these too numerous to be
quoted here.
We shall review rapidly in these pages some typical ritual processes, either
simple or complex, which are all practical applications of the general organising
principle of mantroddhāra.
The simpler and certainly the most frequent form of mantroddhāra consists
in merely giving the syllables in the order which constitutes the mantra. Mantras
being the energy aspect, the active form of deities, and being therefore powerful,
their syllables are not simply enumerated for everybody to know them, but usu-
ally given in such a way that only initiates can identify them. There are, for such
purposes, conventional names for the Sanskrit phonemes, which are not the same
in all traditions: in practice, the guru will tell his disciple how to understand the
secret uddhāra of his initiatory tradition. There are also books which list the con-
ventional names of the Sanskrit letters:7 they can be made use of to decipher an
uddhāra – but, as is well known, a mantra learned of in a book or deciphered with
the aid of a book is powerless and therefore useless.
Here are a few instances of this simpler form of mantroddhāra:
First, that of the bījā HRĪM . , of the goddess Tārikā, which I take from a
Pāñcarātra text, the Laks.mītantra (LT), chapter 44, 7a. It is as follows: “The
concentrated [sound of] Sūrya (H) and anala (R) together, joined with Vis.n.u,
(Ī ) with vyomeśa (M . ) at the end.” To take another instance, in the Netra (NT),
2.21, for the Netramantra OM . JUM . SAH., the uddhāra of the first syllable is
given like this: “The first of all, that which comes after the [phonic] form of
the universe, the one who kills the universe”. “The first of all”, says the com-
mentary, “is the first phoneme, A; that which comes after the form of māyā
14 The extraction and examination of mantras
(māya being identified with the universe8) which is Ī, is U; what kills this uni-
verse is time, expressed by the letter M.” We have thus A+U+M= OM.
A more cryptic uddhāra is that of the ‘heart-mantra’, SAUH. given in a Śaiva
Trika text, the Parātrīśikā. It runs as follows: ‘United with the fourteenth, O Fair
One, associated with the end of the Master of the tithis, the third brahman, O fair-
hipped Woman, is the heart of the self of Bhairava’:9 the fourteenth is the four-
teenth Sanskrit phoneme, AU, the tithis are the sixteen ‘vowels’10 corresponding
to Śiva, who is their master, the last of these being H.; the third brahman is to be
understood as the letter SA or, more precisely S. One has thus SAUH. – provided of
course one knows the proper order of the phonemes, and also that ‘third brahman’
means SA.11
Those are uddhāras based on what we may call play of words. Though sometimes
very cryptic, they are simple processes. But there are also ritual mantroddhāras,
complex processes where the extraction of a mantra consists of several ritual acts
to be performed by a master (guru, ācārya or deśika)12 or by a qualified, that is, an
initiated adept (a sādhaka). These masters and adepts, being initiated, have already
a mantra which they received when initiated and which they have mastered by the
process of puraścaran.a;13 they may however need to use a particular mantra for
a specific purpose – in kāmya rituals especially – and would then need to have
access to it either in a simple, ‘literary’ fashion, as shown above, or they would
have to ‘extract’ it ritually, which is what we shall see now.
The general structure of the process to be used is always the same, referring
also to the same basic principle. The performer, in such cases, is first to invoke
and to display ritually, according to prescribed rules, the fifty phonemes of the
Sanskrit alphabet,14 and pay homage to them; then he will extract, that is, select,
from among this phonetic ensemble the phonemes of the mantra. Then, when
this totality of the divine power or energy (śakti) of vāc in her phonematic form
has been duly, ritually, made present and honoured, the adept picks out of it the
elements which, when assembled, will make up the mantra. Being derived from
a totality, the mantra may thus appear theoretically as a somewhat less exalted
form of phonic power than the whole alphabet (though there are mantras which
are deemed to embody this very totality – OM . notably ), but it is a more efficient,
15

and especially a more usable one – one, also, which can be adapted to particular
uses or circumstances.
One might say that the mantra thus made to appear out of the totality of the
alphabet by a selection of its constituent elements brings together in a more practi-
cally usable and more effective form the individually distinct phonemic entities
of the alphabet. To quote Arthur Avalon, ‘In the mātrkās, the mantra lies scat-
tered. Mantroddhāra is the formation of the mantra by ˚selection of the mātrkās’.16
Analogously, and infinitely more authoritatively, Abhinavagupta’s disciple, ˚ the
Kaśmirian Ks.emarāja (11th c.), said in his Vimarśinī on Vasugupta’s Śivasūtra,
2.3: ‘the secret essence of the mantras is none other than the blissful [Mātrkā],
“whose nature”, as has been shown, “is that of cognition”. And this is the˚rea-
son why in every āgama the presentation of the extraction of mantras is preceded
by the spreading out (prastāra17) of the mātrkā or of the mālinī’.18 Mātrkā is the
˚ ˚
 The extraction and examination of mantras 15
‘normal’ order of the Sanskrit phonemes, from A to KS.A, mālinī is an apparently
haphazard disposition of the phonemes, the first being NA and the last one PHA:
nādiphāntam, as is sometimes said. Without considering these two terms here,
we shall note that mātrkā (in the singular when applied to the whole collection of
phonemes, or in the ˚plural if these are considered separately, and in that case
the word can very appropriately be translated by ‘little mother’) is used for the
varn.as to underline their character as mothers (or as a Great Mother) who bring
into existence the whole cosmos, and, more specifically, the mantras. As is often
said in Tantric works, she is the unknown mother of all the mantras, or, to quote
Ks.emarāja’s commentary on the Svacchandatantra,19 2.30–31, ‘the universal
mother, unknown, of all bound souls, begetter of all mantras and tantras, going
from A to KS.A’.20 She is the original ground or substrate (pūrvabhitti) of all words
and objects.
Many quotations from various sources could be adduced as proof of the validity
of Ks.emarāja’s formula as well as of the general frequency of the mantroddhāra
practice. For example, we may look at a passage of the first chapter, on the extraction
of mantras (mantroddhāraprakaran.am), of the kriyāpāda of the Mrgendrāgama
(Mr g), with its commentary by Nārāyan.akan.t. ha. The latter, commenting˚ on sūtra 2
˚
of this chapter, repeats (after the āgama) that the Power of the Lord is kun.d.alinī as
the supreme word (parāvāc). In her, a first form of subtle sound, a supreme reso-
nance (paradhvani), the nāda, appears, which then condenses as a drop (bindu) of
phonic energy (described by Nārāyan.akan.t. ha as an inner murmuring – antarsam .
jalpa). From it comes the aks.ara, the Imperishable, the syllable OM . , which begets
the mātrkā, the mother of all words, consisting of the Sanskrit alphabet from A to
˚ which all the mantras are born. The chapter then goes on enumerating
KS.A from
the various mantras of this Śaiva tradition, which are displayed in form of a dia-
gram. The mantras are thus deemed to be ‘extracted’ in that they derive ontologi-
cally from the phonemes of the alphabet. No actual method for ‘extracting’ them
is described, but the principle is clearly there.
This being so, and the principle being always the same, in actual practice the
ritual of mantroddhāra is performed in different ways, details varying from text
to text, but the overall pattern remaining always the same. The adept is always to
spread out21 the mātrkā in front of himself, to pay homage to or worship it, then to
˚
extract from it the letters which make up the mantra.
Thus, the SvT (1.30ff.) says that:

[Is to be prepared an area of] white, red, yellow or black earth, purified by
sprinkling [water on it], without any twigs, endowed with all [possible favour-
able] characteristics and able to bring about all that is to be wished, perfumed
with pleasant-smelling ointments, adorned with flowers and beautified above
by a canopy. [There,] the master, purified, smeared with sandal- and aloe-
paste and made fragrant with incense, his mind calm and pleasant, facing the
east or the north, with one-pointed22 and concentrated mind, is to spread out
the mātrkā in the order from A to KS.A.
˚
The tantra goes on to say that the sixteen vowels are Bhairava and the consonants
16 The extraction and examination of mantras
are Bhairavī: the supreme god and his energy whose conjunction engenders the
universe. The officiating master is to worship (prapūjayet) the god Mātrkābhairava
with the vowels (avargen.a), and his consort Bhairavī with the groups ˚ of conso-
nants (kādinā): the tantra (1.34–36) enumerates eight goddesses, the eight Mothers,
the as.tamātrkā,23 that are to be worshipped each with each of the eight groups of
phonemes (the ˚ vargas). The pūjā once performed, the tantra adds, the master is to
extract the mantra in the proper order as prescribed. Ks.marāja explains this injunc-
tion as follows: ‘At the end of this mātrkāpūjā [the mantras are to be extracted] in
˚
the proper order, that is successively beginning with the āsanamantra, [then] the
mūrtimantra,24 etc. This is to be done as prescribed, fixing one’s attention very care-
fully on the deity expressed by each mantra.’25 This last prescription refers to the
rule that the mantras of deities to be worshipped are to be enunciated beginning with
the mantra of the throne (the āsana) on which the deity is to be placed (or, before
it, the root-mantra, the mūlamantra), then the mantra of the form or body of the
deity (mūrtimantra), then the other ones. These others are the mantras of the deity’s
. .
‘limbs’ (anga) and ‘ancillary limbs’ (upānga) and, finally, those of other deities.26
The NT, a work from the same tradition as the SvT though probably slightly
later, explains in its second chapter (2.17–21) that on a levelled, pure area,
anointed with sandal and aloe and perfumed by different substances, the mas-
ter, anointed, perfumed and bejewelled, with peaceful mind, must first draw the
outline of a lotus whose eight petals point towards the four cardinal and the four
intermediate directions of space. Having placed (nyasya) in the centre of the lotus
the mantra OM . , he writes or draws out (likhet) each of the eight groups (varga)
of letters of the mātrkā on each of the eight petals, starting from the one pointing
east. Then, ‘having˚ worshipped with utmost devotion (pūjayet parayā bhaktyā)
the goddess mother of all mantras, and having offered her flowers, incense, etc.,
he is to extract the mantradevatā’ the phonemes of which are then enumerated in
a covert fashion.
Let us now look at a work from a different Tantric (Vais.n.ava) tradition, the
Jayākhyasam . hitā (JayS) of the Pañcarātra, whose sixth chapter describes the
extraction of the main mantras (mukhyamantroddhāra) of that system. The descrip-
tion is very detailed. The ritual practice is more complex than the one we have just
seen; it enters into more particulars, but the organising principle is the same. An
area is first to be prepared – that is, flattened, cleaned, smeared with cow dung,
sprinkled with the five products of the cow (pañcagavya27), anointed with sandal,
perfumed with incense, etc. Then, having paid homage to this area with white
flowers and a mantra, the yogin28 is to draw with clay a round or square throne for
the mātrkā (mātrkāpīt.ha) on which he spreads out (prastāra) the phonemes of the
˚ While˚ doing so he is to realise that the mātrkā is nothing else than the
alphabet.
multiple aspects of the supreme deity as it manifests˚the cosmos. The phonemes
he has written out are therefore no mere letters but forms of the divine energy.
He is then to place by nyāsa29 the mātrkā on his body which is thus pervaded by
her power as well as inhabited by the˚fifty mātrkās as deities. Now he is to draw
a large eight-pronged figure (mahācakra) on ˚whose centre he writes OM . , and
in the eight directions the eight vargas. There follows a ritual worship (pūjā),
 The extraction and examination of mantras 17
to be done with devotion (bhaktyā30) and with the usual ritual offerings, of the
mātrkā written on the consecrated area and of the one placed on his body, a wor-
ship˚ during which the officiant is to invoke by their name31 and worship sepa-
rately all the fifty phonemes, ‘visualizing them as shining and blazing like the mid-
day sun’. ‘These sounds (śabda)’, the sam . hitā (6.58b–60a) adds, ‘are fragments
(am. śā), shining aspects (bhāsvaravigrahā) of the Lord. Increased by the power
of the Lord, they are the cause of all the mantras, whose birth (utpatti) they bring
about by their mutual conjunctions. In this moving and unmoving world, there is
nothing that is not brought about by them.’32 Then, the main mantras, beginning
with the mūlamantra of Vis.n.u followed by the mūrtimantra, are to be extracted,
these mantras being, of course, not spelled out directly: their constituent phonemes
are designated by their names and their place in the varn.asamāmnāya order (śl.
61–69). This whole section (śl. 5 to 69 of the chapter), which I have very much
summarized,33 is interesting in that, while following the general organizing rit-
ual pattern previously outlined, it underlines the meaning of the process, namely
the passage, through a process of progressive condensation and precision, from the
undifferenciated energy of the deity, first to a more concentrated but still diffuse
aspect, that of the mātrkā, then finally to the discrete form of the mantras. It also
underlines the necessary ˚ identification of the Tantric performing adept with the
power he both manipulates and worships.
A different process of uddhāra is shown in another Pāñcarātra text, the
Sanatkumārasam . hitā (Indrarātra, 2.55ff.). There, the mantra Om . namo nārāyan.āya
is given, but its eight syllables are to be placed in eight of the sixty-four compart-
ments of a diagram where fifty-one letters34 and twice the mantra OM . are displayed,
the extraction of this mantra being followed by its resorption, together with that of
all mantras, into the bosom of Vis.n.u.
A passage of the LT (chapter 24) is also worth quoting, not so much for its ritual
details as for the way in which the uddhāra is presented which, putting it in its
divine-cosmic context, underscores remarkably the meaning of the process. Here
is, briefly, the content of the first thirty-two stanzas of the chapter:35 the supreme
brahman, the absolute I, the omnipresent supreme soul, Laks.mī-Nārāyan.a, takes
on, so as to help all living beings, the nature of the mātrkā made of śabdabrahman,
and more specifically the nature of mantras. The master ˚ is to invoke the Supreme
as the saviour mantra (tārakamantra), which here is OM . . He is to take the three
phonemes of OM . , A, U, M, conjoin them, adding (‘beautifying them with’) bindu
and then with the phonic resonance, the nāda, which dissolves eventually in the
origin of sound. He is now to meditate on the Absolute as made up of these three
letters which are identified with Aniruddha, Pradyumna and Sam . kars.an.a – the
Vyūhas of Vis.n.u. Meditating this mantra, identifying with it, the adept unites with
the eternal brahman and with the energies that emanate from it and which sustain
and animate the universe. Mantra and the universe are then identified. Later on, the
. .
master will place ritually (by nyāsa) the mantra with its twelve angas and upāngas
on his body, which identifies him with the deity as master of the cosmos, then,
continuing, he fuses mentally with the divine transcendent formless Absolute
whence however he will come back to an ordinary state of consciousness so as
18 The extraction and examination of mantras
to be able to transmit the tāramantra to his disciple. We have here therefore not
merely the extraction of a mantra, but a fusion of the master with the Absolute
which enables him to transmit this reality to his disciple. This explains the com-
plexity and intensity of the meditative activity of the master who is both to identify
with a mantra and then to transfer its power.
A number of other texts prescribing or describing different forms of man-
tro-ddhāra could still be cited: the Kiran.āgama (Kir), 12, for instance, or the
Rauravāgama (RAU), kriyāpāda 2b–4, or the Pūrvakāmika, 2, or the Īśanaśivaguru-
devapaddhati (Īśgp) II, kriyāpāda 6, etc. But this is not necessary since many texts
are not very explicit, therefore not worth quoting, while in other texts we would
find the same general method, or very similar ones, for extracting mantras.36
As regards the diagrammatic display of phonemes for the mantroddhāra, a
practice found in some texts, notably in some Śaiva Kaula traditions, is worth
mentioning. The diagrams used in such cases are called prastāra or gahvara, that
is, display or cavern. The letters to be chosen are in effect spread out (pra√STR)
in these diagrams, but this usually in such a way that only initiates can find them: ˚
they can therefore be considered as being hidden in the diagram as it were, as in a
cave or a cavern (guha or gahvara).
We find such a prastāra, for instance, in chapter 1.83–85 of the Vāmakeśvarīmata
(VM), a tantra of the Tripurā (or Śrīvidyā) tradition, in a passage which gives
the uddhāra of the vidyā of the goddess Karaśuddhikarī, AIM . KLĪM . SAUH.. The
tantra quotes the letters of the vidyā by referring to their place in a prastāra of a
triangular shape divided into forty-nine triangular compartments, each with one of
the forty-nine Sanskrit phonemes inscribed in it in their order from A to HA (it is
printed on p. 45 of the Kashmir Series edition of that text). The AIM . , in that case,
is described in the tantra as ‘the bīja which is between E and O, the one which is
alone and related to the word’: the phoneme AI is in effect placed between E and
O on the first line of the prastāra, the commentary explaining that ‘the one which
is alone’ is bindu. The same method is used for the seven other letters of the vidyā.
Jayaratha, in his comment on this passage, quotes two texts, the Nityākulatantra
and the Rasamahodadhi, which describe the same sort of prastāra. Those, as well
as the gahvaras, can be of varied shapes with phonemes disposed in a variety of
orders. The phonemes in the diagrams, too, can be those of the ‘normal’ alphabet,
the mātrkā (or the śabdaraśi which is another name for the same alphabetical order
when it˚ is considered as the phonemic form of Bhairava), or that of the mālinī.37
The prastāra-gahvara system is likely to be found in different Tantric traditions
since all insist on the secret nature of their teaching.

To be made use of, mantras are not only to be conferred on somebody or to be

constituted by being extracted from the Sanskrit alphabet. It is also necessary for
mantras to be adapted to the person to whom they are given and to the use they are
to be put. This adaptation is checked through a process of examination or inves-
tigation (vicāra or parīks.a) of the mantra. The idea on which such an adaptation
is based is that for the mantra to be effectively used there should exist between
it and its user a specific link or relationship: this is expressed by the composite
 The extraction and examination of mantras 19
term mantrām . śa, am
. śa being taken in the sense of share or portion, the user being
considered as a portion of, or as participating in, the mantra, as having a particular
affinity with it. This commonality between the mantra and the mantrin permits
the latter to use fully and effectively the former. This control of mantrin–mantra
commonality is effectuated by diverse divinatory methods which vary according
to traditions or circumstances, their diversity corresponding also to the wish to be
able to use different methods in case the one first resorted to proves unsatisfactory,
or to the possibility of chosing the method one prefers.38
Here too we shall limit ourselves to a few instances, quoting a few Śaiva and
Vaisn.ava texts, so as to show the general structure or the spirit of these am . śa
checking processes, rather than describing many particular practices. We shall
first refer to such texts as the SvT and NT, adding a few other references. On the
Vais.n.ava side, instances will be taken from the LT and the Kramadīpikā.39 These
processes are in fact to be found in all ritual manuals, ancient or modern, where
mantric practices are expounded, and they are always very similar.
One of the best Śaiva description of the amśa practice is to be found in the
eighth chapter of the SvT.40 It expounds the means with which to ascertain the
participation, relationship or affinity (am . śa) of an adept with a particular deity or
supernatural entity and therefore with the mantra of that deity which that adept will
be given by his master. Six different cases are enumerated.
There are, first, two simpler ones, where the disciple spontaneously feels
devotion for a particular deity. This is called bhāvām . śa if it is due to his natural
(sahajam) feeling or emotion (bhāva) for a deity on which his mind dwells con-
stantly (anusmaran.a). If this is due, in a more or less similar fashion, to his innate
disposition or impulse (svabhāva), if for instance he feels constantly impelled to
.
worship a linga,41 the case will be that of the svabhāvām . śa (śl. 2 and 12). If there
42

is no such natural affinity, the disciple and his master will have to resort to divina-
tory means. The most usual one is for the disciple, blindfolded, to throw a flower
(pus.pa) onto a man.d.ala on which are drawn or symbolically placed Bhairava or
other forms of Śiva and of his retinue of deities: the mantra chosen will be that of
the deity on which the flower falls. This is the pus.papātām . śa, it is mentioned in śl.
13, chapter 8, of the SvT.43 ‘This mantra’, the tantra adds, ‘will be effective if it is
worshipped according to the prescriptions of the śāstras.’ The throwing of a flower
on a ritual diagram is a fairly common ritual practice. It is described in several
texts, ancient or modern. It is also used in other contexts, for instance in initiation,
to decide the name of an initiand (see SP, vol. 3, p. 103).
It may happen that the above methods for choosing a mantra (or those we
shall see later) do not satisfy the adept, who wishes to receive a particular
mantra. He may then try to obtain the desired am . śa by offerings of human flesh
in the sacrificial fire, offerings to which the tantra (14b–15a) alludes indirectly,
but which the commentary names clearly: naramām . sa. Ks.emarāja uses also (p. 9)
another, more frequent, conventional term, vīradravya, the heroic [ritual] ingre-
dient,44 while explaining how this offering is to be made by the sādhaka or the
ācārya.
If the preceeding methods to secure the am . śa remain unsuccessful, the disciple
20 The extraction and examination of mantras
can still obtain the desired am. śa by worshipping the mantra and making oblations
in the fire (yāgam. purā krtvā agnau homam . kārayet – 16a), this having been done
˚ rules, the disciple ‘being united, thanks to a full obla-
according to the prescribed
tion, with the plane of the Eternal (pūrn.āhutiprayogena yojayec cchāśvate pade ),
setting his heart upon that plane, will succeed in what he wishes’ (16b–17a).

Now remain to be seen the more current ways for the adept to get or determine the
mantrām . śaka that are to be found in a number of Tantric texts as well as in ancient
or modern manuals.
The SvT, chapter eight (śl. 20ff.) and the NT, mention only one method, con-
sisting of writing on two lines, one under the other, the mantra to be checked and
the name of the adept who is to receive it, then to count in the traditional Indian
way, on the phalanxes of the right hand, the number of letters in the normal order
of the Sanskrit alphabet (never in reverse) which separate the initial letter of the
name of the disciple from the first one of the mantra. Four am . śas may thus appear:
siddha (which one may translate as ‘obtained’, or ‘effective’), sādhya (‘attain-
able’), susiddha (‘very efficacious’ or ‘completely atttained’), and ari, śatru, or
ripu (‘hostile’, ‘enemy’), according to whether the counting stops on the three pha-
lanxes of the ring-finger or on one of the four that follow, etc. Only those stopping
on the first and third are favourable (siddha and susiddha) and allow the mantra to
bear fruit – they are even (ibid. śl. 24b) bhuktimuktiphalaprada: their fruit is lib-
eration and rewards. In the two other cases (second phalanx, or the fourth – on the
little finger), the mantra is unfavourable or dangerous and one must strictly abstain
from giving it to the adept or initiand. Ks.emarāja, whose commentary on śl. 20–24
explains how to proceed, adds (perhaps needlessly) that this am . śakaparīks.a does
not apply to monosyllabic mantras nor to mālāmantras.45 This being done, the
main deity (Bhairava Kapālīśa in this case) is to be worshipped with his retinue.
There are still other methods both to make sure that the mantra is perfectly
adapted to its potential user and to distinguish more precisely (notably by deter-
mining sub-categories of the four am . śas) between the effects of the mantra and the
methods for using it. As a result, the adept is sure to get a mantra which will be per-
fectly adapted to him and which will respond to his needs, whilst he will also know
beforehand what effects he can expect from its use, and on what conditions.
One will thus check if the adept and the mantra belong to the same ‘family’
(kula) by comparing the first letter of the mantra with the initial of the mantrin’s
name, the phonemes of the alphabet being, for this purpose, displayed on a plate
with four columns (a kulākulacakra) considered as being mutually either compat-
ible or incompatible: for the kula to exist, the two initials must be in two compat-
ible columns.
One can also, using a square or a round diagram, compare the rāśi, the lunar
mansion or zodiacal sign conventionally attributed to the adept and to the mantra:
according to the rāśi where they will be, one will know if the mantra would or
would not be favourable, or what result one may expect from its use. The same
process is made use of to compare the naks.atra, the constellation or asterism, of
the adept with the naks.atra assigned to the first letter of the mantra, these being in
 The extraction and examination of mantras 21
two different sections of the diagram: some coincidences are deemed favourable
(siddha, susiddha), others are bad (sādhya or ari).46
Another predictive method consists in using a diagram called akad.amacakra
because the four phonemes A, KA, D.A and MA are placed in its upper central
division. The letters of the alphabet are placed in it by groups of four forming
twelve radiating columns in a circle or twelve lines along the sides of a rectangle.47
According to the distance between the first phoneme of the mantra and the ini-
tial letter of the adept’s name, the mantra will be considered as siddha, sādhya
susiddha, or ari. There is also the rn.idhānicakra, the diagram of the debtor and
the creditor, with which – by a rather ˚ complicated calculation – one will see if the
mantra is favourable (for which it must be the adept’s debtor), or else holds a debt
claim against him and is thus unfavourable.48
The method most frequently used, I believe, consists in tracing an oriented
square surface, divided into four quarters, each one being divided also into quar-
ters: there are thus sixteen small squares in groups of four where the phonemes
of the Sanskrit alphabet are written either in the normal alphabetical order (the
aks.arasamāmnāya) or in a different order corresponding to particular symbolical
values given them (in this case, it is often called akathahacakra, from the names of
the letters placed in the first little square (on the left). The four squares divided into
four are assigned, going clockwise, to each of the am . śa: siddha on top left (north-
east), sādhya on top right (south-east), susiddha down left (north-west), and ari
down right (south-west). The am . śa is determined, not by the distance separating
the square containing the first letter of the name of the disciple from the square
containing the first letter of the mantra, but by their juxtaposition, and it will be
expressed by associating the categories assigned to the two squares. The am . śa
would thus be, say, siddha-ari or sādhya-sādhya, or siddha-susiddha, etc., there-
fore dangerous, useless or favourable, etc.: there are sixteen possibilities instead
of four only.49 The validity of a mantra for an adept can also be checked using the
kūrmacakra, a diagram in the shape of a tortoise (kūrma) which is to be drawn by
the adept who is then to write the letters of the Sanskrit alphabet, by groups, in
nine different sections of the body of the tortoise (the vowels in the central part, the
consonants grouped by vargas, on top, bottom, sides, ‘hands’ and paws. Accord-
ing to the sections where the initial letters of the mantra and of the adept’s name
will be, and by some calculations, the mantra will be found to be siddha, susiddha,
sādhya or ari.50
One sees from the above that all possible guaranties are sought for the mantra to
be perfectly adapted to its user. The need for such an adaptation is self-evident in
the initiation – dīks.ā or abhis.eka – of the sādhaka, since he is a bubhuks.u, a seeker
of bhukti, of supernatural accomplishments, to be attained through the mantra
which has been given him and which is to be mastered – it is a sādhyamantra – this
mastery being the first thing he is to do, through the ritual of mantrasādhana,51
after been initiated. A mantra adapted to its user is also a necessity in the larger
context of desire-oriented (kāmya) rites. These kāmyakarmān.i, as they are called,
are rites practised so as to obtain particular results, notably – but not necessarily –
through the cult of particular deities (especially of power-yielding goddesses such
22 The extraction and examination of mantras
as the Yoginīs52) who will bestow on their worshipper boons of different sorts.53
Among those cults are the so-called ‘six [magical] acts’, s.at.karmān.i, all of which
are performed with mantras54 (there are also the, less frequent because condemned,
practices of sorcery, abhicāra). But the field of kāmyakarmān.i is infinitely larger
than magical actions: it includes everything in religious life that is done to satisfy
a human wish, be it for one’s mundane satisfaction or for one’s salvation – which
is more than anything else something devoutly to be wished.
It is to be noted, too, that however important it may be in some contexts,
mantravicāra or mantraparīks.a is not always necessary. It is even sometimes pro-
hibited. As already mentioned, monosyllabic mantras55 and the longer ones, the
so-called mālāmantras, cannot be examined in such a way. There are also man-
tras which are not submitted to am . śavicāra or to diagrammatic parīks.a because
of their high status: these mantras act directly everywhere and without limita-
tion, such as the divine śakti whose phonic aspect they are. This is the case of
the great Śaiva mantras such as the para- or hrdaya-mantra SAUH., or of HAUM .,
the prāsādamantra, which many texts consider ˚ as the mūlamantra of Śiva; or
of the mūlamantra of the goddess, or of that of Vis.n.u, for Vais.n.ava traditions,
and so forth.56 Such mantras are said to be common to all, universal (sāmānya
or sādhāran.a). There is, finally, what appears to be, as already mentioned, the
most frequent case, that is, the cases where the uddhāra of a mantra is given
without any ritual practice being mentioned, its constituent letters being simply
given in a more or less covert manner or even overtly quoted. There are also,
mainly, all the innumerable cases where mantras are to be uttered during a ritual,
where, evidently, the performer cannot examine their adequacy before using them,
which does not prevent them from being perfectly effective: they effect what they
say, or make present the deities they ‘express’ (being their vācaka). There are,
as is well known, ritual situations or processes which do not imply any outward
action, where merely mantras are uttered (for instance the ātma- or āntara-pūja
which, in Tantric practice, always precedes the visible, material, ‘external’, bāhya,
worship). There is no rite without mantras, and these are effective without any
previous ritual fuss being necessary.
To conclude, we can, I believe, confirm the principle mentioned at the begin-
ning of this study, namely that the mantroddhāra or mantravicāra process is linked
to the nature of mantras as it is described in Tantric texts, or, more generally, to the
Tantric conception of the power of the word. Mantras being the efficient and utilis-
able form of this word-power, they must be adapted both to the aim pursued and
to their user. This user, being himself animated by a power that is fundamentally
of the same nature as that of the mantra, can put the mantra in effective use only
by fusing his own energy with that of the mantra, which is posssible only if there
is adequacy between them.
It appears therefore in all that has been seen here (as in what is studied in other
articles in this volume) that one cannot examine particular aspects of mantras
without touching directly or indirectly upon the problem of their nature, which
is ambiguous. Though they are forms of the supreme Word, vāc, though they are
deities, devatās, in their own right, though existing therefore on a transcendent
 The extraction and examination of mantras 23
plane, mantras exist also empirically when – as (in principle at least) utterable
phonetic units – they are used by human practitioners. They are both divine in
essence and forms of human speech. They can, or must, therefore, when being
made use of on the human plane, be submitted to ritual control procedures. These
technical mantric rites meant to ensure their adequacy are inseparable from the
sacred or magical Word, vāc, while being the external social acts that, by guar-
anteeing the proper use in context of the forms of vāc that mantras are, ground
and ensure their authority on the social plane. The efficacy of mantras, whether
one believes it to be real or symbolic, cannot be dissociated from its ritual, social
context.
3 Japa1

Before studying japa, one must first see how this term is to be translated. In other
words, what does it mean for us? An answer – if an answer is possible – must take
into account the descriptions and definitions of japa given in Sanskrit texts as well
as the nature of the ritual or religious uses of japa. But the uses and the definitions
of japa have varied according to traditions, and, still more, have evolved in the
course of centuries. A unique and generally valid translation of the term would be
possible only if there were a unique and generally admitted conception and prac-
tice of japa, which is not the case. The only solution to our problem is thus not to
translate the word whilst trying to show the various conceptions and practices of
japa which have existed (and still exist now) in India.
Japa, is a masculine substantive, from the verbal root JAP which, according to
Monier-Williams’ Sanskrit–English Dictionary, means ‛to utter in a low voice,
whisper, mutter (especially prayers & incantations), to pray to anyone in a low
voice; to invoke or call upon in a low voice’; japa itself being defined as ‛mutter-
ing, whispering … muttering prayers, repeating in a murmuring tone passages from
scripture or charms or names of a deity, etc.; muttered prayer or spell’. Böhtlingk
and Roth’s Sanskrit Wörterbuch gives for JAP: ‛halblaut, flüsternd hersagen, her-
murmeln’ and for japa: ‛das flüsternde Aussagen eines Gebetes, Liedes, usw.; ein
auf diese Weise hersagtes Gebet’. Hence the translation of japa by Murmelmedi-
tation sometimes used in German (by J.W. Hauer, in Der Yoga, for instance), a
translation which evokes rather well the double aspect – muttered recitation and
mental concentration – of japa. The Sanskrit encyclopaedia Śabdakalpadruma
describes japa as vidhānena mantroccaran.am, ‛uttering a mantra according to rule
or precept’, and then quotes several texts, especially from the Purān.as, which set
out the rules governing the ritual recitation of mantras.
These few indications – many others could be adduced – suffice to show
japa as being a muttering or whispering of a text or of words of a religious or
magical character, and particularly of mantras; a muttering or whispering (a
prayer perhaps) to be accomplished according to certain rules – it is a ritual
utterance to be delivered in a low voice. The main meaning of the word japa
is indeed to mutter. (We shall see that japa can also be done aloud as well as
silently or mentally.) Japa, finally, is never an isolated utterance. It is the repeated
utterance of a formula which is, to quote the Sanskrit expression, repeated ‛again
 Japa 25
and again’ (bhūyo bhūyah.), even sometimes, as we shall see, an enormous number
of times.
The term japa is ancient. One finds it in the Śrautasūtra and Grhyasūtra where
it refers to a muttered prayer or, in a more technical sense, to ˚the recitation of
the vyāhrtis bhur bhuvah. svar om . . In that case japa is the muttered recitation of
stanzas ˚or ritual formulas by the officiating priests or by the yajamāna at spe-
cific moments of the Vedic cult. The most important form of japa in the Vedic
domain is in the svādhyāya, the personal recitation of the Vedic text, which is, it
is said, a japayajña, a sacrifice consisting of a reciting, an offering of the word.
The svādhyāya is the brahmayajña, the sacrifice to the brahman, the highest of
the five Great Sacrifices, the mahāyajñas, prescribed by the tradition.2 As japa is
an utterance consisting of speech, it is performed with what is the most high, since
the transcendent brahman is speech: brahma vai vāk (AitBr 4.21,1). We see here
proclaimed an idea that we shall find later on, repeated in many ways all of which
tend to underline the very high capacity for salvation of japa and its fundamental
function in many rites. The paramount importance traditionally ascribed to japa
appears also in the Bhagavad Gītā when Kr◦ s.n.a says (BhG 10.25) yajñānām . japa-
yajño ‘smi, ‛Among the sacrifices, I am the sacrifice of the muttered word’. Manu
also says (2.85) that ‛the sacrifice consisting in japa is ten times higher than the
one consisting in performing a prescribed ritual action’, vidhiyajñāj japayajño
visis..to daśabhir gun.aih.. Manu adds: ‛the japa done in a low voice is a hundred times
higher, and the one done mentally a thousand times’ (upāmśuh. syācchatagun.ah.
sāhasro mānasah. smrtah.). This hierarchy is ever present and will be especially
made use of in various ˚ Tantric forms of japa, of which those considered as the
highest are not vocal utterances but purely mental or spiritual practices. The
higher value given to word over action and, among speech acts, to silent over
audible utterance is, from the earliest period down to modern times, a generally
admitted notion. The word, vāc, originally, insofar as identified with the brahman,
with the (non)utterance of the Brahmin Vedic priest, is silence.3 The Absolute is
silent. The word is manifested by issuing forth from this primal Silence – into
which it is eventually to dissolve – a dissolution which is sometimes accom-
plished by some forms of japa. Of those, the general rule is that the lowest form
is the vocal (vācika) one, which can be heard by somebody else. Then there is the
upām . śu japa where the performer articulates the words or sounds in such a way
that they cannot be heard. This is the middling sort of japa. The highest and best
one is mental, mānasa (which does not exclude the articulation of the mantra,
provided it is done mentally).4
We are not concerned here by the history of japa, but only by its Tantric prac-
tice. We may, however, mention, among the older references to japa, a passage of
the Moks.adharmaparvan of the Mahābhārata (12.189–193) where japa is defined
as the recitation of a useful Vedic text (japet vai sam. hitam. hitam), without explain-
ing what sort of text. Japa appears there as a practice derived from the svādhyāya
and thus considered like the latter as being of an exalted nature since it is an
offering of the all-powerful and divine Vedic word. (This reminds one of BhG
10.25.) This practice (japavidhi) is meant for the man in the world as well as for
26 Japa
the renunciant. It includes such observances as control of the senses, truthful-
ness and so forth, and either a sacrifice, or, on the contrary, the renunciation to all
sacrifices and to the senses. This japa bestows supernatural powers to the man in
the world, and it will lead the renunciant to dhyāna, then to samādhi and further to
fusion with the ātman. The jāpaka is thus equal to the yogin (and even superior).
Japa appears therefore as a way to salvation parallel to that of the Sām . khya and
Yoga. It is possible that in this passage the MBh wanted to affirm the superior-
ity of japa, as a tradition going back directly to the Veda, over that of the Yoga-
Sām . khya, considered as somewhat inferior: two currents which will converge
5

later in Tantric japa. Note that this japa can confer supernatural powers to the per-
former. (We may note too in passing that to prove that the jāpaka is equal – even
superior – to the yogin, the MBh tells the story of the jāpaka Kauśika Paippalada,
that is, of a sage from an Avestan tradition which can be related to Kashmir.)
We may note also, among the notions either explicit or underlying the Tantric
japa practices or speculations, two ancient notions, namely the identification
of the emission of words in japa with that of the semen and – connected with
the notion of the superiority of silence over uttered word – the pre-eminence of
the brief condensed utterance over the explicit, long one. AitBr 2.38 says thus: ‛the
hotar mutters the japa: he emits there the semen. The japa is inaudibly uttered;
inaudible too is the emission of semen.’6 The enunciation of the Vedic word is thus
homologous with the emission of semen, and just as semen is not to be squan-
dered, so the Vedic word must be kept, stored as it were, for retention generates
strength, whilst squandering weakens.7 This persisting ideology which extols the
brief over the long, the concentrated over the diffuse, the retained over the emitted,
the silent over the audible and the unexpressed over the expressed, results inevi-
tably in giving pride of place to the techniques for preserving (notably sexual)
energy, but it also results in placing short or monosyllabic mantras (the bījas,
those concentrations of phonic energy) over long ones and, for japa, in placing the
mental, externally unexpressed one over the vocal audible one.
Let us underline, to conclude these few remarks, the extreme importance always
given in japa to its phonetic aspect: all traditions underline the absolute necessity
of an absolutely exact and precise utterance. This trait constantly insisted upon in
mantraśāstra is probably to be traced back to the importance given in Vedic times
to the exact pronunciation of the Veda and in particular of the svādhyāya which
later became the japa, where what is important is the phonetic content, not the
meaning: mantravid evāsmi nātmavit, says Nārada in CHU 7.1, 3).

One naturally cannot review in an article all the cases where a japa is prescribed,
nor can one even merely mention all categories of uses of japa that have existed.
One can, however, consider the main cases where this meditated ritual, repetitive
enunciation, is prescribed in tantras, āgamas, ritual manuals or other Tantric texts,
and describe the forms it appears to have most often taken. The instances quoted
here will come mainly from the Śaiva-śākta field, for it is the larger and richer
Tantric domain. Some will come too from the Vais.n.aiva domain of the Pāñcarātra,
but Pāñcarātra was very largely influenced by śaivism: the Tantric period was in
 Japa 27
fact a ‘Śaiva age’.8 A few Buddhist practices may be alluded to in passing. As has
already been noted in chapter 2 and will be in chapter 4, practices and notions vary
according to sects or traditions, and even within the same tradition. But within this
diversity there still exists a unity of vision and theory and even in the pattern of
the practices,9 so that such a general approach as our one is, I believe, possible.
We shall thus see first the ritual practices, the technique as it were, of japa. Then
will be seen the mental, or somato-psychic and spiritual aspect of japa as a way
towards liberation as well as a devotional practice. We shall end by looking at a
few popular forms of japa.
Admittedly, this threefold division of the subject made for clarity’s sake is arti-
ficial since, in actual practice, ritual technique and mental effort are both always
present in japa (it is a ‘Murmelmeditation’ to use the very evocative German
expression), being also often associated with devotion and the quest for liberation,
as we shall see.

Ritual practices, the japa techniques: japavidhi

Japa is currently used by Hindus as a common religious practice,10 a time of con-
centration, of devotion, of mental or vocal prayer. It is also – and, if one refers
to texts, one could say that it is statistically mainly – an essential element of a
large number of ritual practices. It is one of the domains where there appears
most evidently the ritual minutia and superabundance or redundancy characteris-
tic of Tantric Hinduism. It is, of course, a domain where bhakti plays a prominent
role, but where it seems, for one who relies on normative works (but perhaps we
misread them?), that the ritual manipulation of the power of the word – mantras
being the efficiently usable form of the deity – has there the pride of place; devotion
and ritualism being of course not incompatible – many rites are prescribed as hav-
ing to be done ‘with devotion’, bhaktyā. There are, too, Tantric texts which extol
the transcending of rites. Such is the case of the Kaulas, as appears for instance
in Abhinavagupta’s TĀ, or in the KT, which says: ‛silence is the supreme japa’
(maunam eva paro japah. – KT 9.38). But in most cases japa is a ritual action.
A japa, the ritual formalised repetition of a mantra, is thus prescribed in the
mantrasādhana where it is part of the puraścaran.a; in the cult of deities which
always include one or several japas; or in kāmya rites, especially for those of a
‘magical’ sort such as the so-called s.at.karmān.i, etc.; finally in daily rites, whether
the private sandhya or the pūjā, a field where the prescription of japa is very
ancient (see Kane, History of Dharmaśāstra, vol. 5, p. 685), in the initiation
(dīks.ā) rituals, in those of atonement and reparation (prāyaścitta), for the installa-
tion of temples and images, etc., all occasions where a deity is invoked or where
some formula of praise or homage is repeated so as to benefit from the deity’s
grace or favour, or in view of some mundane or other-wordly benefit.11
Japa may consist in different utterances. We have seen that of bhur bhuvah. svar,
which is Vedic. In Tantric practice, what is recited is usually the mantra of the
deity (or of an aspect of the deity) that is aimed at by the rite, and, in such cases,
all sorts of mantras may be used, going from a short or long formula to mere ritual
28 Japa
syllables – bījas – either isolated (H.RĪM . , HSAUH., KLĪM . , etc.) or in a group, as
for instance the fifteen syllables of the śrīvidyā (HA SA KA LA HRĪM . , etc.). Or,
more often it is a (often much longer) enunciation where bījas and words are
associated.12 Considering the extent of the Tantric pantheons and the quasi-infinite
number of supernatural beings, powers or elements that can be invoked for a vari-
ety of ends, the number of ritual formulas that can be uttered in japa is enormous,
this in spite of a great monotony of forms and pattern.
Japa may consist, too, of the mere repetition of the phonemes of the Sanskrit
alphabet. Many texts prescribe the japa of the ‛garland of letters’ (varn.amālā) from
A to KS.A, which in that case appears more or less as a phonetic variant of the rosary
(aks.amālā).13 Its fifty phonemes are sometimes to be enunciated with the bindu, that
is from AM . ĀM . , etc. to KS.AM
. . I have not seen any case of japa of the other order
of the alphabet, the n.ādiphāntakrama of the Mālinī. But the forty-two phonemes
of the bhūtalipi (the ‛writing of the elements’), from A to ŚA, are sometimes used
for japa, notably as an ‛encapsulation’ (sam . put.a) of a mantra so as to increase its
power.14 In all these cases the aim is, for the jāpaka, to master or at least to invoke
the power, the śakti, of these phonetic groupings. He also invokes this power when
doing the japa of a mantra since all mantras are made up of phonemes.15
Whatever its phonetic substance, the japa must always be done following the
very strict rules governing the utterance of mantras, these rules being usually spe-
cially stringent for kāmya rites. Such rules, however, vary according to traditions
as well as to circumstances. They are, as is usually the case in India, numerous and
very detailed – details into which we cannot enter here. We shall limit ourselves to
those which are more frequent and usual, and to those that are more characteristic
both of the usual form of japa and of the meaning and scope generally attributed
to this practice in the Hindu world.
TIME. The prescriptions concerning the time when a japa is to be done will
not detain us much since this is usually the time for the performance of the rit-
ual ensemble of which japa is a part. The rules on time for japa are therefore
those governing the daily, occasional or optional Hindu rites (sandhya, pūjā,
puraścaran.a, etc.). We may, however, note that some Tantric rites are to be performed
during the night (see, for instance, Gandharvatantra (GT) 29 on the kulayāga of
the vidyā of Tripurasundarī). On the matter of time, the observing Hindu will fol-
low the rules of his social group.16 The proper time is also the one prescribed for the
deity of the mantra to be particularly worshipped.17 The rules may vary according
to the result one is looking for. The length of the japa is also prescribed. It depends
upon the number of repetitions – we will come back to the subject later (pp. 31–32).
There are also rules of a general sort to be followed in all cases where special rules
do not apply; this happens in particular when japa is used as a prayer or a spiri-
tual exercise, not as a portion of a particular rite – see on this, for instance, LT 28
where the daily obligations of the adept are prescribed: he is, the tantra says, to
unite spiritually (yoga) with the supreme deity, then, ‛when tired with yoga, he
must do a japa, when tired of this, he will again practice yoga’ (ibid. p. 48). Japa,
as we shall see further on, is among the practices prescribed in the various forms
of yoga.18
 Japa 29
PLACE. The place to practise japa is the subject of meticulously detailed pre-
scriptions, which take into account so many different possibilities that following
the rules is made easier. Though varied, these rules always aim at determining a
favourable place, that is, a peaceful one, pure, far from all natural and (still more)
supernatural danger: the peace and mental concentration of the person who repeats
a mantra must not be troubled, for it is only when at peace that he/she can hope
to gain the merit, or obtain the favourable mundane or spiritual advantage he/she
expects to result from the process. Such a place is often called a pun.yaks.etra, an
auspicious place, a term that is also used to refer to a sacred spot or a pilgrim-
age site, generally to all places where one performs a pious action or where one
acquires merit: all elements denoted by the word pun.ya.
The choice of the pun.yaks.etra is specially important in the mantrasādhana,
the ascetic practice which gives the sādhaka the mastery over his mantra (a rite
also called puraścaran.a19), for it is a long and complex process which can only
be pursued in an appropriate and carefully chosen place.20 Once the mantra is
mastered, its japa can be performed whenever the ritual during which it occurs
is to take place. Here, for instance, is a quotation from the KT (15.22–24) on
the subject: ‛The auspicious places [can be] a riverside, a cavern, a pilgrimage
place, the meeting of two rivers, a sacred grove, solitary parks, the foot of a bilva
tree, a mountainside, a temple, the seaside, one’s own house. Such are the places
prescribed for the sādhana of mantras.’ The KT then adds: ‛Or [the adept] will
stay where his spirit rejoices’: a remarkably generous prescription! Thereafter the
KT quotes some other favourable spots, as well as those to be avoided, among
which are those where one may meet kings and their retinue.21 Rules evidently
vary according to texts and traditions. For some Tantric systems, and especially
for some rites where the officiant is possessed by a wrathful deity or where power
results from impurity and transgression,22 the most impure and inauspicious places
are, on the contrary, preferably chosen, these being usually cremation grounds
(śmaśāna), where dark powers lurk and where the horror and impurity of death is
often associated with the transgression of orthodox rules of conduct, those notably
concerning sexual purity.23
Chosen according to the result wished for, the pun.yaks.etra must also
(except in such cases as just mentioned) be cleaned and purified. Sometimes, a
man.d.ala is to be drawn. It is, in particular, necessary to protect it from all pos-
sible bad and dangerous powers by mantras and mudrās, and/or by invoking and
worshipping the guardians of the quarters of the sky (dikpāla) and by offering
them a bali.24
In this context of choosing a stable, peaceful and auspicious spot, the texts
also say on what the jāpaka is to sit (the āsana), what bodily attitude (karan.a) he
is to assume,25 and what direction he must face. The clothes to be worn are also
sometimes mentioned, especially their colour, which is often to be the same as
that of the deity being invoked.26 If a rosary is used, the texts say how to hold the
hand, which is often wearing a pavitra27 (apavitrakaro … na japet); they say too
in which direction one must look, etc.28 In addition to these external observances,
mental composure, peace, silence, self-control, inner purity, and so forth are also
30 Japa
prescribed, since they are necessary: there is no japa without mental concentra-
tion on the mantra and on the deity. All these rules evidently do not apply in the
case of transgressive rites; the Brahmayāmalatantra, chapter 21, for instance, in
a passage describing the vidyāvrata observance, says that the yogin, covered with
ashes, contracting his brow (bhrūkut.ī), is to do japa while laughing boisterously
(at..tahāsa).29 The Siddhayogeśvarīmata,30 27.23ff. contains analogous prescrip-
tions for the japa of the Heart of Bhairava mantra: in a cremation ground, at night,
among vampires, Yoginīs, etc.
Even when japa is merely a portion of a larger ritual ensemble – and some
texts say that japa must necessarily be part of a cult: sam . pūjyaiva japam . kuryān
na mantra kevalam . japet (TBhS p. 323, in a passage on puraścaran. a) – it should
nevertheless begin in principle by a preliminary rite consisting mainly of nyāsas
on the body. First the nyāsa of the rs.i, etc. and of the main mantra (mūlamantra)
which is to be recited, then of its ˚an.gas (hrdaya, śiras, śikhā, etc.) on the hand
and on the body (karān.ganyāsa).31 Next, the˚ reciter will place on his head a vidyā
called kullūkā, which varies according to the mantra recited, then, on his heart and
neck he places two other mantras, called mahāsetu and setu.32 The japa itself is
then to be performed (sometimes after other preliminary rites33) according to pre-
scription. For the daily domestic worship, the ritual can be very limited, but even
then some ritual action is always to take place.34
Texts also describe cases when no japa is to be performed; or what is to be
done in case it is interrupted or disturbed by some external element (seeing certain
animals, for instance), or by a bodily incident, such as sneezing, yawning, etc.,
and then how to take it up again so that it does not lose its efficacy. Atonement
rites (prāyaścitta) are prescribed for the purpose.35 There is no need to go into
the details of such prescriptions. We mention some of them here because they
underline the technical or the ritual manipulation aspect of japa – one might say
its aspect of magical use of the power of the word – and also because such details
underline the typically Hindu constant attention given to the rules protecting a
ritual action from dangers and impurities.
All these preliminary conditions and prescribed rules being respected, the japa
itself can now be performed.

We have already seen that, following an ancient rule, a japa can be of three
sorts: it can be voiced (vācika), done audibly (nigadena) uttered in a very low
voice (upām . śu); or done mentally (mānasa). These different forms of japa are
used for different ends: see thus LT 29.35, Mrg Kp 4, or Cp 1.67 and their com-
mentary, or JS 14.3–4, etc. – vācikam ˚
. ks.udrakarmabhya upāmśum . siddhikarmān.ī
mānasam . moks. akarmārtham. For the GT. 18.434, mental japa is prescribed only
for Tripurā’s mantra, while other mantras can be repeated otherwise.36 Other works,
especially modern ones, refine these distinctions by distinguishing between up to
twelve (and even more) different sorts of japa, notably according to the way the
utterance is associated to the movement of breath.37
Of a general and traditional sort is the rule, too, that the mantra recited is to be
clearly uttered, neither too slowly nor too fast: na drutam . nāpi vilambitam . japen
 Japa 31
mantram.38 This prescription also applies to the silent and the mental japas since,
even when silently uttered or mentally evoked, words or syllables are to be artic-
ulated. The prescribed number of repetitions is also carefully to be taken care
of: na nyūnam . nātiriktam
. va japam . huryāt dine dine (TBhS, p. 326 quoting the
Mun.d.amālātantra). This is an essential prescription in kāmya rites where the result
wished for depends on the exact observance of the prescribed rules.
It goes without saying, finally, that japa presupposes not only the repetition of
the mantra but also a careful attention paid to this recitation: on the phonic content
and on the import of the formula being recited, which is to say to the deity that is
thus invoked or worshipped. This implies, to various degrees, singleness of mind,
mental concentration or spiritual effort (dhyāna, dhāran.a, bhāvanā) – something
already prescribed in Yogasūtra 28.39 Japa also presupposes devotion (bhakti). As
we shall see later on, this mental/spiritual aspect of japa can in some cases be the
most important one. It may even be, in the most complex and subtle japa practices,
those in which the whole japa consists. But even in ‘ordinary’ cases it is always some-
how there (this bears out the German translation of japa by ‛Murmelmeditation’):
the mere mechanical repetition of a mantra without paying attention to it does not
suffice. This agrees with the conception of mantras found in many Tantric works,
which is that mantras, though deemed to act by their own power (mantravīrya),
function only together with the mind of the mantrin (or worshipper, as the
SpK 2.11 says: sahārādhakacittena). Japa is thus associated with dhyāna in the
two meanings of that word: meditation and visualisation of the deity. Many texts
clearly say that while doing japa one is to meditate and mentally evoke the aspect
of the devatā as described in dhyānaślokas: dhyānam . kurvan japet vidvān, says
the Śes.am . hitā 21 (p. 111). Bhat t
.. anārāyan. akant
.. ha’s commentary on MR g Cp 7.70
˚
on the calling to mind (smrti) of the deity describes this act as a mental representa-
˚
tion (anusmaran.a), which is twofold: dhyāna, the visualisation of the aspect of the
god, and japa (ibid. p. 106, transl. p. 178). The mantra being the phonic form of the
deity, its recitation is a means for fastening on the deity the performer’s attention,
a phonetic way to bring about a sensuously perceptible awareness of a real divine
presence.40 This is what we shall see when describing complex forms of japa. It
is implicit, for instance, in the prescriptions of LT 28 (see above p. 28) where the
japa prolongs the yogic meditation. This is also what one finds – in a very different
way – in the devotional, ‛popular’, ancient or modern forms of repetition of the
deity’s Name (bhajan, nāmasam . kīrtan), which are also ways not only to worship
the deity, but also to strive towards a mystical union (see infra, p. 45).
Though being always a mental or spiritual activity, japa nevertheless remains,
from another point of view, essentially a technique for using or manipulating to
certain ends the power enclosed in the phonic substance of mantras. It is often
performed in ways which remind us of the vikrti of Vedic recitation where the
order of words and even that of syllables is modifi ˚ ed, inverted, so that the sound
aspect only, the combination of words and sound, not the meaning, plays a
role. (This insofar, of course, as there is no human utterance without a mean-
ing or value; insofar too that all ritual actions since the Veda imply an intention,
sam . kalpa, explicit or implicit, which orientates and gives a meaning to that ritual
32 Japa
act, which cannot be purely automatic and without meaning for the performer.41)
These technical manipulations of the phonetic substance of mantras are named
yoga, pallava, sam . put.a, and so forth: we need not expatiate on them here since
they are described in this volume in chapter 7. They are also to be found in
Buddhism, in the Sādhanamālā, for instance.
Since it consists in repeating a formula, japa always lasts for some time, the
length of which varies greatly. When it is a mantrayoga practice associated with
the ascent of kun.d.alinī, there may be no repetition at all. The act may even be,
at least in theory, of the most extreme brevity, less than a second, and still be
considered as a japa.42 It can also be limited, in the course of a ritual, to very few
utterances, or to one utterance only: the case is frequent. When repetitions are
prescribed, their number is often 108, a number of religious symbolic value, or,
still better, 1008. But tens or hundreds of thousands, even millions of repetitions
are also prescribed. Such prescriptions seem, in fact, often to be rather theoreti-
cal, their raison d’être seeming to be more to set a correspondence between, on
the one hand, the number of repetitions and, on the other, the cosmic and human
context of the utterance and the result aimed at, this without much regard for prac-
tical feasibility. Of course, the texts, prescribing these japas being revealed and
considered as instructions given by a deity to another deity, express themselves
sub specie aeternitatis, leaving it to others (the human master who teaches and
the ritual manuals which comment the texts) to see to it that divine principles
agree somehow with practical human necessities. Hinduism, one well knows, is
as precise in its principles as it is flexible in their application. In temple rituals,
on important festival days, the number of japa repetitions voiced by a number of
officiants, may, adding all those who recite, reach incredibly large numbers.43 In
private practice, a complete japa may last months and even years.
A relationship is sometimes postulated between the length of the japa and the
cosmic cycles (yuga), or the four varn.as (or the different types of adepts). Thus
the Tārārn.avatantra quoted in the TBhS (pp. 315–316), says that the japa is to last
twice longer in the Tretayuga than in the Satyayuga, thrice more in the Dvāpara,
and four times more in the Kaliyuga; and in those four cases (three of which are
purely theoretical), japa will go from one lakh for the vīras to four for the paśus,
in Satyayuga; then from four for the first category and sixteen for the second one
in the present cosmic age. It appears too – this is a more useful point – that the
number of repetitions is smaller for obligatory (nitya) rites than for kāmya ones,
which is reasonable since the efficacy (or the belief in the efficacy, which comes
to the same) of a japa increases with the number of repetitions, and since kāmya
rites are those done for a specific and specially wished for end. The harder the aim
is to reach, the greater the number of repetitions will be. In the puraścaran.a of a
mantra, the japa will also be very long since the aim is to master the power of the
mantra. In the prāyaścittas, the duration of the japa will naturally depend on the
seriousness of the fault, mistake or omission to be atoned for (or on the importance
of the event or circumstance to be prevented or ‛appeased’). We may mention
here the case of the ajapājapa, the japa of ham . sa. The number of repetitions,
in this case, cannot to be counted; but since it is associated with breathing it is
 Japa 33
supposed to be repeated 21,600 times in twenty-four hours; if it is performed dur-
ing several months (as can well be the case), repetitions can easily run into mil-
lions. We will come back to this practice later (see pp. 43–44). It seems also that
japa can sometimes be done without counting. The texts I have seen seem, how-
ever, to condemn this practice. The Brhatkalottarāgama quoted in the Īśp, vol.
3, p. 122, says that an uncounted (asam ˚ khyātam) japa or homa will be seized by
.
demons (grnantyāsurarāks.asāh. – information H. Brunner).
˚ of course many prescriptions concerning the bodily postures, the
There are
clothes to be worn, the mental attitude, etc., necessary during japa. Normally it
is to be done seated in a particular posture (padmāsana, siddhāsana, etc.). Some-
times one must remain motionless, without even moving the head or neck.44 Japa
can also be practised while walking, lying or during some activity, provided it
is pure (except in transgressive rites). Cases where rules are relaxed are to be
found more in the context of devotion (bhakti), where feeling and emotion is more
important than ritual. On the contrary, the minute ritual precisions aim at ensuring
the rigorous respect of the bodily or mental attitudes deemed necessary for secur-
ing the result aimed at.45 Beyond (or rather below) the, one could say, technical
necessity of the scrupulous respect of ritual rules, all the prescriptions we have
seen are there to make sure that all conditions exist that permit the reciter to con-
centrate on the japa without any distraction or disturbance.
It also happens, especially when japa is part of a pūjā, that mudrās are to be
shown whilst reciting, the gesture underlining or reinforcing in some way the
inner attitude of the performer, or expressing respect to the deity. The main hand
gesture in japa is, first of all, that of the hand counting the number of repetitions.
Here, too, prescriptions are meticulous. We cannot describe them all. Here are
merely a few indications.
To count the number of repetitions of the mantra one uses either one’s fin-
gers or a rosary (aks.amālā). In the first case, the hand is considered as a rosary
whose grains would be made up by the phalanxes, hence the term karamālā. We
know that Indians currently count on their fingers, touching each phalanx with the
thumb. For japa one does this going from the ring-finger to the little finger, end-
ing on the index, thus counting ten; this of course on the right hand. (The left hand
may also be used in the same way to count each ten repetition, cowries, etc, being
used for counting hundreds.46)
More important than the karamālā is the aks.amālā or aksasūtra, the rosary.
Its use and all related ritual details, from its preparation to its destruction, as they
are described in the fourteenth chapter of the Jayākhyasam . hitā of the Pāñcarātra,
are the subject of chapter 5 of this book. We therefore need not expatiate on the
subject here. Suffice it to say that the rules set down by the JayS apply generally
in all Tantric traditions, with naturally a number of variants. Rules concerning
the material of the beads, the number of their ‛faces’, how to use the rosary, and
so forth, are to be found in several āgamas or tantras, such as the SvT, the GT,
the Kir, the MatP, etc., or in the TBhS, the Īśp, etc. The rules all aim at giving
the greatest possible efficiency to the japa when done with a rosary – which is
never considered as a mere material help but as a ritual instrument used to invoke
34 Japa
or coerce supernatural forces. Thus, the rosary can be used only after it has been
purified, consecrated, ritually impregnated with divine power (śakti or prān.a)
which has been placed in it by a particular ritual process. Most rosaries have 108
beads. They include one more additional bead placed on the knot joining the ends
of the string, named Meru, which is not only used for counting the repetitions but
also must never be ‛overstepped’.47 The rosary therefore is used by counting the
grains, not round, but back and forth. This is done holding it in the right hand on
which it rests on the middle, the ring and the little fingers,48 the thumb being used
for moving it. The general rule is that no one is to see the japa being thus practised,
the hand is usually hidden under a cloth, or, better, is kept in a small bag called
gaumukhi.49 Sometimes, the place where to hold one’s hand whilst doing japa is
also prescribed; this varies according to the aim for which, or to the time when,
the japa is performed.
Except for the longest japas, the repetition must be done continuously, without
any interruption. If it is interrupted, it must be taken up again, or an atonement rite
(prāyaścitta) is to be performed. One must be careful not to release, or, worse, to
let drop the rosary during japa – nor, in fact, at any other time: ‛a rosary one has
let fall must be washed with perfumed water and [purified] by reciting a hundred
gāyatrīs’, says the SP, prāyaścittavidhih., śl. 49 (SP2, p. 262).
The japa must end ritually by paying homage to the rosary, to which one will pay
respect by joining hands or by holding it to one’s head. It will then be carefully put
away while uttering Om . or some other mantra. When the japa is to be done in the
course of a pūjā, it must, when finished, be offered to the deity that is being wor-
shipped.50 There are also rules concerning how to dispose of a worn rosary (which
must not be used since it could break during japa, which would be very inauspi-
cious):51 one must not leave it aside, but throw it ritually in water after having knot-
ted it together with a stone while uttering the Viśvaks.ena mantra,52 doing a libation
and invoking Hari by reciting seven times the mūlamantra (so the JS 14.34–35).
One can also (this is better) cut it and unwind the string which is then thrown in deep
water, the beads being used again for the new rosary that is to replace the old one.
The rules we have just seen are those generally to be followed when japa is per-
formed as a devotional act, or for some other reason, or as part of a ritual action:
pūjā, puraścaran.a, and so forth. It is in those last cases that the prescriptions are
most complex and are to be most carefully followed since these are the cases when
japa is most clearly a manipulation of the power of the word, this being more
specially the case in rites aiming at mastering the power of a mantra or in kāmya
rites, in initiation, etc. But even in the daily mandatory (nitya) cult, japa is part of
the ritual and must be performed according to rules.

Puraścaran.a. Japa is a part of the puraścaran.a ritual (also called pūrvasevā), a

preliminary ritual action (puraśkriyā) to be performed before the cult of every deity
so as to master this deity’s mantra, a ritual specially important in kāmya rites. It is
generally considered as being made up of ‛five members’ (pañcān.ga) usually men-
tioned in the order: japa, tarpana, abhis.eka, homa, brahmabhukti: recitation of the
mantra, ‘satisfaction’ of the deity, consecration, oblation in the fire, offerings to the
 Japa 35
Brahmins,53 japa being naturally mentioned first since the aim of the rite is to
master a form of the power of the word. The puraścaran.a is sometimes described
somewhat differently. Thus, for the KT, it comprises pūjā, japa, tarpana, homa
and brahmabhukti;54 but it always includes a japa, and may even consist merely
of it, either alone, or accompanied by oblations in the fire (homa). Thus Mrg
Cp. 76: ‛[Every day,] after having ritually worshipped Śiva, he will worship the ˚
mantra to be propitiated and perform the japa. Then, he shall perform a homa.’55
In such cases the number of repetitions is usually very great,56 the japa then lasts a
long time. Thus, for the TĀ 23.31ff., after his consecration (abhis.eka), the ācārya
is to meditate and recite (dhyāyej japet) during six months all the mantras that
have been revealed to him, and is to identify himself with them and thus partici-
pate in their power (yadaiva tanmayībhūtas tadā vīryam upāgatah.). That a japa
can bestow such mastery is explained by the pervasion of the consciousness of the
reciter resulting from the very duration of the repetition. This is an opportunity to
say again here that, for Tantric texts, nothing is higher than mantras, and therefore
nothing surpasses the value, the merit and the efficacy of japa. The KT says thus
that in comparison with the sacrifice of japa, no other rite can be greater, nor supe-
rior, for by japa one can attain all the four aims of mankind:

japayajñāt paro yajño ‘stīha kaścana /

tasmāj japena dharmārthakāmamohs.ām . śca sādhayet //

The japa done during puraścaran.a is among those where the prescriptions we
have seen (place, time, clothing, etc.) are especially important since this is a very
strictly organised ritual aiming at harnessing the powers of the word, a harness-
ing which may have all kinds of results, saving as well as ‘magical’, since in the
Tantric context it is possible for any ritual action to aim at conferring supernatural
powers as well as liberation.57
Normally the japa is followed by oblations in the fire (homa) which appear
thus as a complement or achievement of the japa since the fire carries upward the
whole sacrifice as an offering to the deity.58 The number of oblations is always less
than the number of repetitions and proportionate to it, the usual ratio is one homa
to ten japas, but it can be different; this rule applies notably in the case of the pūjā.
Like all rites, the puraścaran.a ritual varies according to traditions. It can be part of
the most extreme Tantric practices.59 Its role remains, however, always the same,
and it is an essential one for, in theory, it is the ritual which gives the adept the
mastery over a mantra which will be used in all the rites he will accomplish to all
possible ends – rewards (mundane or celestial) as well as liberation.

Sandhyā pūjā. Japa is also to be performed as part of the daily cult of the deity, be
it private or public (in a temple), done for oneself or for others, obligatory (nitya)
or optional (kāmya).
In the daily brahmanical sandhyā, to be performed theoretically three times a
day (morning, midday, evening), the essential element of the ritual is the recita-
tion of the gāyatrī: ‛This is the absolutely necessary portion of a Hindu prayer and
36 Japa
this silent muttering is called japa’, ‛japa or prayer has been said to be the princi-
pal part of gāyatrī’, says the author of the classical Daily Practice of the Hindus
(pp. 33, 39). This gāyatrī, with OM . and the vyāhrtis, is to be repeated ten, twenty-
eight or 108 times, say the texts, which underline ˚ the social and cosmic import of
this act, which not only confirms the caste status of the performer but also ensures
the cosmic movement of the sun.
This is for Brahmanism. But in the ritual Hindu worship of deities, the pūjā japa
is equally mandatory. It is usually one of the last items of the pūjā. In the daily
domestic Tantric cult as described, for instance, in the SP (SP 1, sect. III, 93–100,
pp. 216–225), Śiva’s cult ends with the japa of the god’s mūlamantra, repeated
108 times (śl. 93) using a rosary. According to the SP, the japa is ‛consecrated by
the [mantra] HRD, enveloped by the cuirass [mantra] and protected by the sword
˚
[mantra]’ (hrdabhimantritam . varma ves.itam . khad.garaks.itam), then it is offered
˚
to Śiva with a mudrā.60 This means, as H. Brunner explains, that one recites the
mūlamantra 108 times with a rosary on which one has recited the mantra HRD
while holding in the other hand a flower which one ‛envelopes and protects’ with ˚
the two other mantras. These are then considered to be the receptacle for the japa
which along with the flower and consecrated water (arghya) will be offered to
Śiva. For this offering, the officiant puts it in the hand of the god asking him to
accept it and to grant him full success (siddhir me bhavatu – śl. 95), that is, obtain-
ing the spiritual or mundane advantages he is looking for. This is an interesting
ritual whether one takes it as being essentially a symbolic (self-interested) gift
to the deity of the prayer (considered as something one can theoretically either
give or keep after it has been uttered), or as an inflow of power transmitted by
an object61 (with the transfer of the japa to the image of the deity by means of
a flower). These are two conceptions of the word and its powers of which other
instances are found in the mantraśāstra.
On the same rite, this is what the Mrg, Kp. 3.29–30, says:
˚
Who wishes liberation must offer his japa while uttering the Heart [mantra]
at its beginning and at the end; this he does bowing down without wishing
for anything [in counterpart] and by using the dative case: “To You I offer
[this japa].” The sādhaka offers it while keeping in mind what he wishes [to
obtain], and declaring the number [of japas he has done]: “O Bhagavan ! keep
[this japa for me.]” This is to be done everyday.

Bhat..tanārāyan.akan..tha, in his Vrtti, adds that the mumuks.u offers the japa ‛bow-
ing down without any desire’ and ˚ the bubhuksu ‛keeping in mind what he desires’
.
(ibid. p. 55, note 2). Here again are two uses of japa.
Other ritual treatises, in a non-dualistic context, underline the unity of the man-
tra and of the deity (of which it is the vācaka, ‛what expresses’ it, its phonic aspect)
and the fact that the repeated utterance of the mantra identifies the officiant with
the deity (this in spite of the fact that this identification is already acquired from
the start according to the rule nādevo devam arcayet).62 This identification will
be reinforced before the japa begins by the placing (nyāsa) on the body of the
 Japa 37
ritual actor of the mantra, of its kullūkā, setu, mahāsetu, of its rs.i, chandas, etc.,
of its an.gas and other parts (avayava).63 In the same non-dualistic
˚ perspective,
the giving over of the japa in the hand of the deity can also be felt as symbolising
the fusion in the deity. The GT (18. 51), for instance, explains that by offering
his japa, the officiant (of the puraścaran.a, in this case) is to consider that the
‛luminous fruit’ resulting from it is absorbed by the god: tejorūpam . japaphalam.
śivayā svīktam . smaret. In a similar fashion, Āmr tānanda, in his commentary on
YH 3.192 which prescribes this offering in the ˚left hand of the goddess64 (this
being done symbolically since the cult is made on a śrīcakra without any icon),
interprets this mental rite as a symbolic fusion of the mind of the officiant with the
divine Consciousness.65 We may note that the offering of the mantra to the deity
(japanivedana or japasamarpana) is not specially Tantric: it is a prescribed item
of the daily obligatory cult, the sandhyā, the process of which remains very largely
Brahmanic-Vedic. The Hindu domestic sandhyā, however, is more or less Tantric
according to traditions: more Tantric when Śaiva, less when Vaiśnava.66 The cult
in the temples, when described in the āgamas or tantras, however, is Tantric. It
includes japa, since it does not differ in principle from the domestic ritual; one
nevertheless feels, when reading the texts, that japa is less important in temple
than in domestic cults.
The way according to which japa is done when it is a part of the pūjā
varies according to each tradition, the rules being generally those we have seen
before.67

S.at.karmān.i Japa plays evidently a fundamental role in all magical rites, for in
all such rites the expected mundane effect is always to be achieved by mantras,
which are usually to be repeated. The magical act consists in a manipulation of the
powers of the word. If kāmya rites are performed in view of a mundane or supra-
mundane (but interested) result, this aspect can also be found in Tantric obligatory
(nitya) rites. For instance, in the śrīcakra cult as described in the YH, which is
mandatory for all adepts, the pūjā accomplished in each of the constituent parts
of this diagram gives a particular supernatural ‘great power’ (mahāsiddhi).68 One
knows, of course, that the sādhaka who performs this ritual is more often a bub-
huks.u than a mumuks.u, and that in a Tantric context liberation-oriented practices
also give supernatural powers. More specifically magical appear to be the ritual
actions done exclusively for mundane ends, and especially those aiming at sending
evil harmful spells (abhicāra), rites sometimes called ‘cruel’ (krūra). Such is the case
of the so-called ‛six [magical] actions’ (s.at.karmān.i) described in a number of Tantric
texts. These are generally given as being: śānti, vaśikaran.a, stambhana, vidves.an.a,
uccāt.ana and māran.a, that is, appeasing, dominating, enchanting, immobilisa-
tion, hostility and killing.69 To these are sometimes added fascination (mohana),
attraction (ākars.an.a) and pus..ti, which gives prosperity, thus a total of nine
actions.
During these rituals (which we will not describe here), there is always a moment
when the japa of a mantra is prescribed, be it to invoke a deity, to ask for help,
or for some other reason, the mantra being usually made up of several bījas and
38 Japa
some other words, together with the name of the person or of the entity aimed at
by the rite. It is mostly in such cases that the interpolations or ‘encapsulations’ of
syllables named yoga, pallava, sam . put.a (which we shall see in chapter 7) usually
occur, each being used for aiming at a particular result. Thus the TBhS (p. 369)
prescribes the use of pallava for maran.a, uccāt.ana and vidveśana, for expelling
poisons and fighting against evil spirits and bhūtas, the yoga practice being used
for pus..ti, śānti and vaśyakaran.a rites, and so forth. The strictest respect of all rules
of utterance is evidently necessary in such cases since they are manipulations of
the power of the word.70 Particular mantras are to be used in such rites, where the
same text prescribes the audible form of utterance, this being explained by the
lower nature of such practices (see above, p. 25). The inflection (jāti) placed at
the end of the mantra varies according to the rite.71 Sometimes where to look is
also prescribed: in different directions for different rites; or different substances
for the beads of the rosary (rudrāks.a for vaśya, shell for śānti, etc.). The number
of repetitions is also prescribed, as well as how to hold the hand or which fin-
gers to use to count the beads – for instance, thumb and index for vidves.a and
uccāt.ana etc.72 All such rules do not apply merely to the s.at.karmān.i: in all kāmya
rites, the japa is long and plays an essential role.73

Prāyaścitta. In the rites of appeasement, protection and atonement (śānti and

prāyaścitta) – rites analogous in some respects to those we have just seen – japa
also often plays an important role. The rites of appeasement and atonement for
faults in personal conduct or in the performance of domestic private rites con-
sist mostly in a japa, sometimes brief (108 gāyatrīs, for instance), sometimes
very long – thus, for instance, 100,000 Aghoramantras (according to the SP2,
pp. 240–241) for the Śaiva initiate whose personal lin.ga has been lost, burnt or
taken away. Such recitations are sometimes accompanied by other acts or rites of
atonement. In temple rituals, on the contrary, japa does not appear essential: homa
and snapana are more important.
It is, however, present, and necessary if only because there is no rite without
the utterance of mantras. In prāyaścitta as in other rites, the phonic content and
the accompanying rites are strictly codified (in SP, for instance, 117 ślokas are on
this subject: SP2 pp. 224–329). They are to be most carefully executed. The spirit
of these rites – which aim at repairing/compensating the dire effects of errors or
omissions – is in effect not one of devotion, regret or penance: they are techni-
cal actions74 – this even though creative mental concentration (bhāvanā) plays a
noticeable role, especially when deities are invoked (but bhāvana too is a tech-
nique). Japa intervenes in prāyaścitta because this ritual, like most others, brings
into play the efficacy of the ritual word.

Dīks.ā. Japa intervenes also in initiation – dīks.ā – the rite through which the
disciple receives a mantra from his guru. It also takes place in the consecration
(abhis.eka) of the ācārya and of the sādhaka. However, it appears there more as
part of the cult that accompanies the dīks.ā than of the initiatory rite itself. It may
thus take place only towards the end of the ritual: the mantra, once bestowed to the
 Japa 39
initiand and thereafter ‘satisfied’ by oblations in the fire, is then to be repeated by
the newly initiated in a first japa which prefaces, as it were, the longer one he will
have to perform during his mantrasādhana.75 The initiating master, too, according
to the SP (SP3, p. 496–497), must everyday do some japa (svalpam ahni japam .
kuryāt). The TĀ (23. 31–40) says that the ācārya who has been consecrated must,
during six months, identify himself by bhāvanā with the mantra. But there are
many different initiations, their ritual varying considerably.
The japa may sometimes take place as a preliminary purifying or perfect-
ing action for the candidate to initiation. Thus, the S.āt.S, chapter 16, prescribes
400 recitations of the Vasudevamantra during four days so as to be purified (see
too JRY, 2.10 – ref. Sanderson). In chapter 17 of this sam . hitā, a mantra is to be
repeated 700,000 times as a preliminary to the nārasim . hadīks . ā, but this has prob-
76

ably more to do with appeasement or atonement rites than properly speaking with
initiation. During the initiation ritual mantras are continuously to be uttered and
especially placed by nyāsa on the body of the initiand. They may be recited to him
by the initiating master,77 who is also to do japas for other reasons.78 Often the
repetition is accompanied by oblations in the fire, so that the ritual is more a homa
than a japa, which appears here as of limited importance in the initiation process
since this repetition is nothing more than making use of a mantra already given,
taking place therefore after the dīks.ā rather than being part of it.
There are many other rites where a japa takes place, but which we shall not
see here.79 We mentioned earlier the role of japa in yoga, where it is a means
for focusing the attention on the pran.ava OM . . It can also be prescribed in other
very different contexts where the repeated word can be considered as effective or
where it may have a psychological usefulness. This is the case when japa is used
in medicine or in alchemy for preparing drugs or for curing a disease. Caraka, for
instance, prescribes japa for a rejuvenation cure where it is supposed to have the
highest purificatory efficiency (japasaucaparam). As we have seen, japa may also
have curative effects in non-medical contexts of all sorts (puras.caran.a, etc.).

Mental, somato-psychic or spiritual practices

All the ritual meticulousness we have just seen (albeit not exhaustively) is essen-
tial to japa. It is not merely its outer aspect. It expresses its nature and spirit.
Japa, indeed, is a practice both vocal (or at least linked to a particular plane of
the word) and ritual. It is a highly ritualised way of putting into action the powers
of the word. But, as we have already noted in passing, this vocal practice is also
a mental one. It always implies and requires the full attention of the practitioner.
This mental aspect is essential. Moreover, in japa as in all rites, the mere respect of
formal rules, the mere mental and bodily conduct of the practitioner does not cover
all the possibilities of this ritual action – and still less its meaning. This meaning,
too, is a constitutive element of japa as it is of all word-action, for there is no word
or speech, ritual or otherwise, without an intention to say – vivaks.a – without
intentionality, without an orientation towards a goal one wishes to reach through
what is being uttered, murmured or mentally formulated. This is the aspect of japa
40 Japa
we will now address while looking at practices that are very largely interiorised,
where japa is not really the repetition of a mantra, nor even a murmured medita-
tion – ‘Murmelmeditation’ – but a spiritual and yogic practice.
Japa, of course, is always in some respect a religious action and may even
be a spiritual exercise. Insofar, too, as japa can be a request or an invocation, it
necessarily implies an attention focused on a deity: the creative form of medita-
tion called bhāvanā, or the loving, devotional, participation in the deity of bhakti.
This is seen in many texts. The religious aspect of japa is banal, and can be seen
everyday in the life of Hindus. In such cases, japa is a prayer and may justly be so
called. It is very likely that this devotional aspect increased in the course of centu-
ries. Perhaps was it always there. Nowadays it is certainly the most visible aspect,
its formal ritual aspect diminishing, without disappearing.80
What, on the other hand, has always been there, much before the first traces of
bhakti, is yoga, with all the somato-psychic practices it implies, that have been
especially developed in Tantric Hinduism and Buddhism. In the Tantric world, the
utterance or recitation of mantras is nearly always associated with such practices.
Japa becomes thus not a mere ritual repetition, but a mantrayoga practice where
the utterance (uccāra) of the mantra is associated with the movement of prān.a,
that is to say with the bodily and cosmic energy, kun.d.alinī, and becomes thus
a corporal–mental, even spiritual, exercise, often very intense and complex. We
shall look at such practices now. For clarity’s sake, I will distinguish (somewhat
artificially) between japas of a more yogic type, where the techniques of men-
tal concentration, or the visualisations play a greater role, and others, which one
may call more spiritual or metaphysical, where mystical experience, fusion in the
Absolute, the experience of brahman are in the foreground, with often a minimal
technical aspect; japas, therefore, where sometimes metaphysical gnosis (jñāna)
dominate, sometimes bhakti. Of these different sorts of japas we shall see only a
very few typical instances, briefly described without entering into details.
One of the most complex forms of japa and furthest from the mere mur-
mured repetition of a mantra is probably the one described in the third pat.ala,
śl. 171–192, of the YH. One can find it (with the Sanskrit text) described and
explained in its French translation with Amrtānanda’s commentary.81 I need not
therefore describe it here. Suffice it to say that˚ it comprises four different practices:
the japa of the three parts of the śrīvidyā, then of six ‘voids’ (śūnyas.at.kam), of five
conditions or states of consciousness (avasthā), wakefulness (jāgrat), etc., then of
seven ‘equalisations’ (vis.uvat).82 In this japa, one is not to repeat (even mentally)
a mantra or a bīja, but to create and to follow mentally a complex ensemble of
imagined inner-bodily representations through which the constituent elements of
the śrīvidyā are perceived as abiding in certain points of the yogic body seen as
present in the physical body, and to move therein upwards together with the ascent
of kund.alinī. This ascending visionary intra-corporal thrust of the mantra is at
the same time an ascent towards ever higher planes of consciousness: the phonic
plane of the utterance of the mantra being ever more subtle as it goes from the
constituting phonemes of the vidyā to the also ever subtler phonic subdivisions
of the utterance, the kalās, from nāda to unmanā, the ‘non-mental’ phonic energy
 Japa 41
identical with the pure divine Consciousness, the Brahman, with which the yogin
is finally identified.
It is to be noted that in such a japa, the adept is to visualise mentally as present
in his body not only the pattern of centres and nād.īs of his yogic body, in all
their details, but also, in the centres, he is to ‘see’ the phonemes of the vidyā and
especially the kalās from bindu to unmanā, which crown each of the three HRĪM .,
visualising also the ascending movement that animates them. He also visualises
the nine constituent sections of the śrīcakra which he is to perceive mentally as
being tiered in his body, and which the kun.d.akinī, raising together with the sub-
tle vibration of the three HRĪM . , will pierce, going from the outer square (in the
mūlādhāra) to the central bindu where Śiva and Śakti unite (in the dvādaśānta
cakra, twelve fingerbreadths over his head).
For the adept who follows with his mind’s eye the ascent of the subtle vibra-
tion of the nāda which is, as it were, the ‘fine pointe’ – the subtle summit – of the
mantra, but which is also its essence and substrate, this practice is a japa in that it
presupposes the continuous and pregnant presence, in the mind of the performer
of this nāda, of this quintessence of the mantra which continuously vibrates inside
his mind–body. He is to recite (japet) it, not in repeating and uttering it, but by
fastening and focusing on it (and of course on the whole imagined inner pattern
of the yogic body, which is the stage where the whole inner play takes place)
an intense and unwavering attention. This really is an extreme form of mental
practice, but still a japa. Indeed as we shall see,83 some Tantric texts say that an
attention intensely focused inwardly on the nāda is the real and most eminent
form of japa.
Such a japa is, in fact, a bhāvanā,84 an intense meditation which creates in the
mind of the adept and superimposes on his body image the imaginary structure of
the yogic body with the ascent of kun.d.alinī which takes place in it, with the kalās
of the mantra and/or the coalescence of these kalās and of the śūnyas and vis.uvats,
which are also levels or aspects of the yogic body. The power of the bhāvanā both
creates this mental representation and identifies the adept with it and with the
nāda. Thus, following the nāda till it disappears in the silence of the ‘transmental’,
the adept fuses also with the supreme level of the word, which is Śiva. All such
mental practices in their highest aspect always bring the performer to transcend
all forms of word and fuse into the silence the Absolute – we shall come back to
this later.85
Instances of japas consisting in the upward movement of a mantra together
with kun.d.alinī are found in a number of Śaiva texts. Thus, in JRY 2.10, where the
japa of the Sim . havāhanakālasam . kars.in.ī mantra, preliminary to the dīks.ā, consists
in having the mantra passing the sixteen ādhāras.86 There are similar practices in
Tantric Buddhism. For instance, in the japa made ‛while dwelling on the shape
of syllables’ as described by Mkas Grub Rje87 when the adept creates mentally
in front of him the form of a deity and, while reciting the mantra, imagines that
its syllables, seen as a garland of letters, are on a moon-shaped disk placed in
the heart of the deity. With his inhaling breath, he then transfers the disk and the
syllables into his own heart where he keeps, repeats and contemplates them whilst
42 Japa
retaining his breath. Then, breathing out, he places the disk and the mantra in the
heart of the deity. The process is to be repeated several times. We see thus that
visualisations of divine entities felt as present in the body, and the association of
the recitation with the movement of breath (that is, with prān.āyāma, therefore
with the mastering of the cosmic energy of prān.a) are Buddhist as well as Hindu
Tantric practices.
The link between breath (prān.a) as both respiratory breath and the bodily
aspect of cosmic energy is underlined in many Tantric works, especially Śaiva
ones. Breathing and speech are, of course, associated by their very nature and, in
India, as is well known, the subject has been abundantly tackled.88 Japa is shown
in many cases as linked to prān.a as the power circulating in the human body and
linking it to the cosmos and to the godhead. Thus, when SvT 2.139–140 says:
japah. prān.asamah. kuryāt, ‛japa is to be done in harmony with breath’, Ks.emarāja
explains in his commentary (vol. 1(2), p. 78) that this is ‛the breath moving in the
centre’ (madhyavāhi prān.ah.), that is to say the sus.umnā, the japa being performed
by conjoining one by one the elements of the mantra with the energy of breath
(prān.aśaktau), this going on together with a peaceful fusion of the conscious-
ness of the adept into that energy of the mantra (mantraparāmarśa). Similarly,
Bhat..ta Nārāyan.akan..tha, commenting Mrg CP. 7.70, describes japa as a ‛repetition
˚
directed by the energy of breath’ (prān.aśaktyanusāren . āvartanam).
In the seventh āhnika of the TĀ, which deals notably with the devouring of
time (kālagrāsa), that is, its domination by the yogin, Abhinavagupta says that
the rule of japa is that it is to be performed in harmony with breath (prān.asāmya).
Referring to two ancient tantras, the Yogiśvarīmata and the Yoginīkaula,89 Jaya-
ratha, explains in his commentary that this only ensures success,90 and it is so
because mantras are external manifestations of the energy of the supreme goddess
Varn.akun.d.alī, that is, of the energy of the phonemes constituting the mantra, as it
goes up along the sus.umnā with the uccāra, being thus none other than kun.d.alinī.91
This energy, awakened by the harmony of japa and prān.a, rises up and, on the
transmental plane of unmanā, unites, blazing, with Śiva. One finds here what we
have already seen for the japa of YH: the same mental visions are active there, iden-
tifying word and energy, showing that the utterance of the mantra and the ascent of
kun.d.alinī are the same, leading the yogin to mystically fuse with the Supreme.
In Vais.n.ava texts, we also see japa being used to fuse with the divine energy or
with some aspects of the deity. Thus, in a passage of the Sātvatasam . hitā (6.204ff.)
where the yogin, meditating in his heart through the japa the four Vyūha, Anirud-
dha, etc. issuing from Vis.n.u, fuses them mentally into one another (and at the
same time fuses together the states of consciousness, waking, etc.) and thus attains
samādhi and unites with the brahman (ibid., pp. 119ff.).
This spiritual aspect of japa and its use to realise intuitively the supreme Real-
ity is underlined in the passage quoted previously of Ks.emarāja’s commentary
on SvT. In it he explains japam ārabhet by bhūyo bhūyah. mantram . vimrs.et, ‛one
becomes conscious repeatedly of the [essential nature] of the mantra’, and ˚ quotes
śl. 145 of the VBh which describes japa as a bhāvanā, an identifying meditation
of the Supreme. I shall come back later to this last aspect of japa. With regard to
 Japa 43
the link between japa and prān.a, I would first like to mention here the very often
prescribed practice of the ajapājapa, ‛the recitation of the not-recited’ (if it may
be translated thus), and describe it briefly, for it is an interesting form of japa, spe-
cially in that it underlines one of its main aspects – that of the respiratory breath.
In this practice, in effect, the breath is what is actually ‘recited’ or rather ‛not
recited’ since there is no word to be uttered. This does not mean that there is no
mantra, for if there were none, there would be no japa. One considers in fact that,
with the breath, the ham . sa (ha when inhaling, sa when breathing out, or con-
versely, according to some texts, the Dhyānabindu Up. 61 for instance) is being
enunciated; there is therefore actually a mantric utterance.
Ham . sa is the swan, the symbol since Vedic times of the supreme Reality as
92

well as of the transmigrating soul identical in essence with this Reality. Practising
the japa of ham . sa is therefore uttering the Name of the Supreme and, due to the
cosmic nature of the word, it is also identifying oneself with the world play of the
Deity and thus tending towards a fusion with its divine Origin. Ajapājapa being in
fact the movement of breath, it is, like it, uninterrupted: it is deemed to be repeated
21,000 times in twenty-four hours, this being traditionally the number of breath
cycles during that time.93 Despite its physiological basis, ajapājapa is not devoid
of the usual ritual strappings of all japas: it is in fact this very ritual, together with
the will to perform and with the utterance of ham and of sa, that makes it a japa.
Mere breathing is not a japa – except (if one is to believe the ŚS 3.27 and Ksema-
raja) in the particular case of an advanced yogin fused with the deity, whose every
act is a worship of, or a participation in, the Supreme (see below, pp. 46–47).
A number of texts, notably Śaiva ones, prescribe ajapājapa.94 Here is how it is
described in the Daks.in.amūrtisam . hitā, a Śākta work on the cult of goddesses:

I will tell you the ajapājapa the knowledge of which suffices for the adept to
be identified with the supreme brahman. Who performs everyday the pūjā of
the word ham . sa will be freed of illusion and will gain liberation. Having learnt
it through the grace of a master, he will then recite it and, by this movement of
in and out-breathing, he will be freed from all fetters. The two syllables of ham .
sa are [respectively] in the in-breathing and in the out-breathing. This is why
breath is called ham . sa [the Swan], o Goddess! that moves in the body as the
self 21,600 times in a day and night. This is how this supreme [not-recited],
made of the bliss of the vibration of prān.a, is to be recited every day. Birth is
the beginning of the japa, death is its offering. The recitation is thus there in
the mantrin, o Goddess, though there is nothing to be recited.

The DMS seems here to identify the natural movement of breath and japa. This,
however, is not so. It prescribes in effect how (śl. 8) to evoke (sam . smaret) the
visualisation of phonemes which the adept is to see mentally with their different
colours in each of the cakras (mūlādhāra, etc.) of his body. This japa is therefore
a yogic practice. In addition, nyāsas are to be done aiming at perfecting the body
(dehasya siddhaye); then, the mantrin is to meditate the Lord and to utter the
ham. sagāyatrī. Thereafter, the DMS adds:
95
44 Japa
the mantrin will do a prān.āyāma carefully and completely, as prescribed, with
the mūlamantra [the hamsa, that is], o Goddess! filling his body [with air]
by the left nostril, keeping it during three times [the time of aspiration], and
rejecting it by the right nostril, closing alternatively each nostril with his little
finger and his ring finger, without ever using the index nor the middle fin-
ger. The two wings of this hamsa, o Goddess, are the twofold movement [of
breath]. Agni and Soma are these two wings, the Tāra [mantra OM . ] is its head,
the threefold bindu his crest and his eyes; in its mouth is the nāda. Śiva and
Śakti are its two legs, Kāla and Agni its two sides. This supreme swan, omni-
present, dazzling like ten million suns, shines by itself. It is cosmic resorption,
this ham. sa, which gives discrimination (viveka): whoever practises cease-
lessly the ajapājapa will not know any new birth.
(DMS 7.24–30)

Concerning ajapājapa, we may quote also a stanza from another Śaiva text, the
Śivayogaratna of Jayaprakāśa96 according to which among the elements which
can stop the movement of the mind during ajapājapa is ‛the immobilisation of the
breath due to the bindu, which is in the centre of the ajapāmantra’ (ajapāmantra-
madhyasthabinduna krtakumbhaka). This bindu is the m . of the syllable ham
. of the
ham sa. It is the japa’s ˚ central point in the sense that it is the last phoneme of the
.
ham . (of the in-breathing or out-breathing, that is), after which, with sa, begins the
other respiratory movement. The bindu is thus, as it were, in the interstitial time
between these two moments of the respiration: at the junction of the breaths, thus
metaphysically at their central point (madhye) in the interstitial void where breath
stops imperceptibly and where the Absolute may be experienced. Such a notion
can be found in other texts, notably those of the Kashmirian non-dualist Śaiva
systems. Reaching this central point would thus be the ultimate highest goal of the
utterance of the ham . sa in the ajapājapa. This is underlined by Ks.emarāja in his
Uddyota on SvT 56–59, where he quotes VBh 155–156.97 But, in the present case,
ham . sa is understood as the essence of the phonemes, as a symbol of the Absolute,
as the sound ‛that nobody emits and to which there is no obstacle, the Swan which
arises spontaneously, present in the breast of all living creatures’, to quote the
well-known formula applied to the ‛unstruck’ (anāhata) sound which is the very
essence of the word, the animating power of the mantras, and therefore a divine
cosmic entity, not a mere respiratory breath, even if controlled by yoga. The case
is therefore that of a mystical experience of the Word-absolute, very far from the
usual ritual context of japa. But this is also one of the possible aspects of this ritual
practice, as we shall see later.
Let us note too, concerning the ritual recitation of mantras, that ajapājapa is
also mentioned in the works of (or attributed to) Goraks.anātha, a practice still
found nowadays among Nāthas and Gorakhnāthi yogins.98
In Vajrāyāna Buddhism, such practices are also to be found, linked with breath,
not done with ham . sa but with the word evam, the first word of all the Sermons
of the Buddha: ‛So I have heard’ (evam . mayā śrutam). These two syllables,
sometimes taken as symbols of prajñā and upāya, are generally considered as
 Japa 45
associated with the ascending (prān.a) and the descending (apāna) breaths which
circulate in the id.a and pin.gala (or lalana and rasan.a) canals, the process being
deemed to result, with the suspension of breath, in the junction and immobilisation
of these two breaths in the sus.umnā (or avadhūti), and thus, in the realisation of the
highest bliss, mahāsukha. Such a practice, though not an ajapājapa, is very near to
it since it brings into play the same mental representations.99
With or without such yogic aspects, ajapājapa appears in other cases as being
chiefly a spiritual and mystical practice aiming at the union of the practitioner (or
rather the devotee) with the deity he/she worships. Of course, all japas aim to a
certain extent at some fusion with a deity. This appears in the passage from the
TRT quoted above, apropos of the Yoginī cult. But there are also more particular
cases where japa is not a japa in the normal sense of the word since there is no
mantra repetition. It even happens, as we shall see, that the very mental or spiritual
state of the adept, without any mantric practice, is considered as being a japa.
The YH 3.6–7, about what it calls the supreme worship (parā pūjā), pre-
scribes the adept to concentrate in his heart on the Supreme, turning away from
differentiated murmuring connected with dualistic notions (vikalparūpasam . jal-
pavimukhah.). Amr tānanda explains in his commentary (Dī, p. 195–196) that this
is not an ‛external’˚ japa (bāhyajapa) when one whispers (actually or mentally)
the syllables of the mantra with all the particularities inherent to the linguistic
activity of speech and therefore of discursive thought, but an ‛inner’ (āntaram)
japa. This consists in fixing one’s attention on the nāda, the subtle phonic vibra-
tion (āntaram . nādānusam . dhānalaks.anam. japam) which is perceived in the heart,
the mystical centre of the human creature. Amrtānanda then quotes two stanzas,
˚
the first of which describes this nāda as the aspect assumed by the deity on the
inner plane of the ‛unstruck’ (anāhata) sound. The other stanza says ‛One must,
mastering the movement of the senses, enunciate the inner nāda. This indeed is
what is called japa: the external japa is not a [real] japa.’100 In this case, therefore,
japa is not in anyway the repetition of a mantra, but the mystical realisation, in the
‛heart’, of the immaterial phonic vibration of nāda.101 So as to make his meaning
clearer, Amrtānanda ends by quoting sūtra 145 of the VBh which can be translated
approximately˚ as follows: ‛This creative meditation which is meditated again and
again on the supreme Reality, this, here, is a japa. In this way, what is recited is
the spontaneous nāda consisting of a mantra.’102 For the VBh, what is really a japa
(japah. so’tra), what is deemed to be ‘recited’ and which in this case is considered
as being a mantra (mantrātmā), is the spontaneous subtle phonic (svayam . nādah.)
aspect of the deity.
In his short commentary of this śloka, Śivopādhyāya quotes the SvT 4.399,
aham eva paro śivah., which would make this practice close to ajapājapa. Aham,
however, in the Kashmirian Śaiva schools, is the absolute I, the supreme tran-
scendent Śiva as holding in Himself the whole cosmos.103 As Ksemarāja says in
a well-known formula: ‛What is called the condition of the absolute I is the rest-
ing of the Light’ – prakāśasyātmaviśrāntir ahambhāvo hi kīrtitah.. For the VBh,
japa appears as consisting of repeated (bhuyo bhūhah.) plunges or immersions
in this luminous phonic essence of the Self. According to Śivopadhyāya, these
46 Japa
immersions could be related to the play of ham. sa understood as the subtle ‘breath’,
not different from nāda, of the absolute, which the yogin can mystically feel within
himself as an infinitely subtle resonance perceived in the heart.
The same interpretation of japa as a spiritual practice is also to be found in
Ks.emarāja (whose interpretation of the Pratyabhijñā has certainly influenced
Amrtānanda) and in Abhinavagupta. Commenting on sūtra 1.27 of ŚS, Ks.emarāja
also˚quotes SvT 4. 399 as well as sūtras 145 and 155–156 of VBh. Abhinavagupta,
in TĀ 1.89–90, has the same position, quoting the Triśirobhairavatantra104 which,
describing the behaviour of the yogin who has attained fusion (aikātmya) with
Śiva, says:

One whose spirit is free from all impurity, and who, stopping his memory,
meditates the Supremely Meditable, present in all that moves or does not
move, reaches also by means of japa the supreme Śiva named Bhairava. One
says indeed that japa is the very nature of this [God] which is above the planes
of being and non-being.105

For Abhinavagupta, thus, it is not merely meditation of Śiva, the ‛supremely

Meditable’, that leads to fusion with him, but also the japa, if it is practised in
complete purity of spirit, forgetting everything that has to do with ordinary life.
This japa, he adds, is an intense and repeated synthetic awareness (bhūyo bhūyah.
parāmarśanam . japah.) of the essential nature of Śiva, of the Self which is the
106

supreme Word (tasya śivasya svarūpam . paravāksvabhāvam . ātmarūpam). In

essence, this japa is nothing else than the direct intense total awareness of the con-
sciousness that shines in the very centre of being and non-being, or, to render madh-
ya otherwise, in the interstitial [void] between being and non-being (tanmadh-
yasphuratsam . vitparāmarśamātrasārah.). The direct intuitive apprehension of the
absolute that underlies the manifested world in the interval between two moments
of consciousness is, as one knows, an important notion of that school. The con-
ception of japa as a way towards union with the deity is, however, not proper
to Abhinavagupta’s Trika, or to the Pratyabhijñā. Thus Maheśvarānanda in the
Parimala on his Mahārthamañjarī, a Krama work, writes that what one calls japa
is nothing else than the intense awareness of the essence of the mantra given by the
spiritual master (kevalam upapāditamantrasvarūpa-parāmarśo japa ityucyate).
This notion is much earlier than these texts since it is found in such tantras as the
Triśirobhairava, or in the VBh.
A cosmic and metaphysical dimension is given to japa in the Paramārthasāra,
78 and its commentary by Abhinavagupta: the japa, associated with the vibration
of the prān.a, is an inner vision by the yogin of the expansion of the cosmos, the
constituent elements of which are seen by him as revolving in eternal succession
like the beads of a rosary used by a devotee. An analogous conception is in the
PTV, p. 75, commenting sūtra 35 of the PT.
A most mystic and metaphysical conception of japa is found in the nineth
(perhaps even eighth) century ŚS of Vasugupta, a basic text of the non-dualistic
śaivism of Kashmir. Describing the condition of the liberated in life who lives
 Japa 47
in constant union with Śiva since he is identified with Him. Sūtra 3.27 of the
ŚS states: kathā japah., ‛[His] conversation is a japa.’ The whole being of such a
yogin, Ks.emarāja explains in his Vimarśinī, is nothing else than an intense iden-
tifying meditation of the supreme I (aham). Referring to a passage of the SvT
(4.399) already mentioned (p. 46), Ks.emarāja quotes the Kālīkrama:107 ‛The con-
sciousness that this God – who transcends all the gods and whose nature is the
highest awakening – has of Himself is the supreme Śakti, totally omniscient.’
He adds: ‛whoever, therefore, remains in the awareness of the absolute unborn
I, whose essence is the great Mantra, his conversation, etc. becomes a japa – a
ceaseless repetition of the intuitive consciousness he has of the godhead who is
his own self.’108
In this perspective, japa not only ceases completely to be a ritual repetition: it is
not even the utterance of a mantra, insofar at least as a mantra is a particular formal
ritual utterance. The texts we have seen speak of mantras in such a context – which
permits them to say that it is a japa – merely because, in the metaphysical concep-
tion of the word, of vāc, developed in non-dualist śaivism, the four planes of the
word – parā, paśyantī, madhyamā and vaikharī – are always present in the cosmos
as in the human being. But also because, in this metaphysical perspective, the
supreme word, insofar as it enunciates the cosmos, as it is the godhead ever present
in the cosmos, is the absolute I, which is the great Mantra (paramahāmantra). This
is the mantra that the liberated in life, identified with the supreme Śiva, necessarily
enunciates as does the godhead. Whatever he says, being on the level of paravāc,
is the great Mantra. This means that in practice his speech will always say the
truth, will be completely efficacious and entirely liberating – not enslaving as the
ordinary speech of human beings.
It is worth remarking that Ks.emarāja concludes his commentary on ŚS 3.27
by quoting sūtras 155b–156 of VBh on the japa of ham . sa which may appear
109

as being somehow a transposition of ŚS 3.27 onto a more ordinary plane. In fact,

these two conceptions are complementary ones in the sense that just as all that
the liberated in this life says is a mantra and therefore a japa, all breathing, being
identified with the godhead, is ham . sa, which is the affirmation of the oneness of
the yogin’s self with the absolute – so’ham – or with the absolute I – aham. Such
a yogin, therefore performs a japa even when he does not speak, which is all the
more true since the japa, here, is a direct and repeated awareness of the divine, a
fusion which, for this yogin, is always present.110
One finds the concept of japa of the ŚS in the fourth āhnika of the TĀ which
deals with the ‛means of power’, śāktopāya, that is, the way to liberation by mys-
tical fusion with the divine Energy. ‛Who is immersed in this uncreated Heart,
Abhinavagupta says, whatever he does, breathes, or thinks, one considers that
it is for him a japa.’111 Indeed, Jayaratha explains, in all mundane activity, such
as talking and so forth, the spirit of the yogin never stops being immersed in the
intense consciousness of the absolute uncreated I (akrtakāham . parāmarśaviśrānto
hi yogi), and therefore all this for him is japa (yat kim ˚ cid bāhyavyavahārayogam
. .
vyāharet so’sya sarvo japah.); which means that for this yogin everything shines
(= appears) as mantra: as what is by nature a return to the awareness of the
48 Japa
godhead which is his own self (svātmadevatā). As Jayaratha notes, this can be
related to Abhinavagupta’s conception of the efficacy of mantras (mantravīrya),
which he considers to be the efficacy of the supreme Word. Jayaratha notes too
that others, of different traditions, have also said that ‛conversation is japa’
(bāhyair api yo jalpah. sa japah. ity uktam).112
In a different, mystical perspective, Utpaladeva, the author of the ĪPK, says in
his Śivastrotrāvalī 3.17: ‘Nothing is higher than Me, yet I practice japa. This is
why, verily, japa [is being done] in unity [with myself] whichever rosary [is being
used]. This is what You teach.’113 Here it is because the yogin conforms with a
divine model, with what Śiva does (and perhaps we may here refer to the notion
of the absolute I – aham – that Śiva enounces and that is shown to the devotee as
a model for identification with the divine plenitude), that his japa is taken in its
highest dimension: identity with the deity, a mystical union which the devotee
who recites his rosary with the usual ritual would (or try to) experience.114 This
stanza is especially interesting in that it gives japa a divine model. It is, however,
not unique in doing so, for, in an emanative vision of the cosmos such as that of
the Pratyabhijñā or of other systems where the whole universe appears through
the descent of a divine archetype, or by the luminous appearance (ābhāsa) and
reflection (pratibimba) of an archetype, everything that exists empirically must
need have its original model, its archetype in the godhead.115

The devotional japa: popular aspects of japa

We have given until now the pride of place to the yogic, metaphysical or mystical
aspects of japa because one finds there some of its most interesting aspects. They
show indeed the extension that Tantric yoga can give to this ritual action, all the
philosophical developments that accompany it and, related to it, the interpreta-
tions, meanings and range that can be given to it within the metaphysics of the
word developed mainly in the Śaiva/S.ākta non-dualistic traditions of Kashmirian
origin. But such cases, which are interesting because they are very subtle or imagi-
native, are, for this very reason, comparatively rare. Like many things Tantric,
they were probably always more present in texts than in actual practice – this is
merely because many practices – the most complex ones especially – were meant
for a small group of initiates, not for a larger social group. In the majority of cases,
japa has probably always been, and it continues to be, the ritual recitation of man-
tras that we have seen in the first part of this study, a practice whose spirit is quite
different from the other one, without, however, there being a clear cut distinction:
we have seen thus that the Śivastotrāvalī, a mystical work if there ever was one,
mentions the use of the rosary. A ritual recitation, too, can lead the practitioner to
a mystical experience.
Japa, indeed, the repeated invocation of a deity, when not used for a specific,
often a magical end,116 is a prayer, invocation, call for supernatural help; it can,
like any prayer, lead to contemplation. We must note that in these last cases
the metaphysical perspective usually differs from that of the non-dualistic
Kashmirian Śaiva systems, their context being usually a dualistic or semi-
 Japa 49
dualistic (viśis..tadvaita) one, that of Rāmānuja, for instance, and of other śrīvais.n.ava
masters, or that of the Pañcarātra. One can hardly wish to take refuge in a deity,
ask for its help, without there existing at least a measure of difference between the
devotee and the deity being worshipped.117 In fact, the japa practices we shall see
now are mostly Vais.n.ava ones. They may have developed especially from a time
– the thirteenth century perhaps – when bhakti, often in its most emotional forms,
permeated more and more Hinduism. They went on developing down to our days
where they are part of the everyday religious life of Hindus. Such practices are
not those of the (so-called) mantraśāstra (especially when it is identified with
tantraśāstra), but they often appeared first in Tantric milieus, or milieus influ-
enced by tantricism, and they have kept, even in their present forms, Tantric traits.
This is why one may mention them as a last part of this study. More than before, I
shall limit myself to a few instances.
Going back first to what we saw previously about the japa of gāyatrī dur-
ing sandhyā, where I referred to Ś. Ch. Vidyārn.ava’s The Daily Practice of the
Hindus, I will quote here K.C. Pandey’s definition of japa as ‘the constant rep-
etition of a word or sentence’, to which he adds that this repetition ‘produces a
state of consciousness bordering upon ecstasy’ through which ‘one enters in the
fourth kośa or sheath called vijñānakośa or sheath of Buddhi or intuition’. He then
illustrates his point by quoting the case of Tennyson, the poet, who could enter
in a manner of hypnotic trance by merely repeating for a long time his Christian
name.118 One may indeed consider that japa, like all repeated actions, lulls the
attention and creates a manner of altered state of consciousness where thought
looses contact, as it were, with recitation and shifts to something else. But we
must not go too far in this direction: my point here concerns only the repetition of
the name.119
For indeed japa, as a devotional repetition, as a form of contemplative or
repetitive prayer, appears often – especially from a certain period onward – as
the repetition of the name of the deity: nāmajapa (or its chanted repetition,
nāmasam . kīrtana), practices which, with the bhajans of our days, are the most fre-
quent expressions of Hindu devotion. In these cases, even more than in those we
saw previously, the origins of the practice is very ancient: its conceptual basis is
the ever-present Indian belief in the power of the word. This goes back to the Veda,
where the name is considered as embodying, comprising the essence (and the
power) of what is being named: to utter it is to act upon it, make it present, favour-
able.120 Hence all the invocations, so frequent in ancient texts; hence notably the
nāmastotras, the praises made up of names and attributes of a deity that are recited
as a litany in that deity’s honour, or the very long repetitions of a divine Name.
The most ancient nāmastotra is probably the Śatarudriya of the Vājasaneyi
Sam . hitā, a hymn still being recited by Śiva’s devotees notably during the Mahā-
śivarātri night, a hymn that the Mānavaśrautasūtra names rudrajapa. ‘By means of
which japa does one gain immortality?” the novices ask in the Jābālopanis.ad (3), ‘By
that of the Śatarudriya,’ Yājñāvalkya answers.121 The MBh (Moks.adharmaparvan)
mentions lists of nouns – ‘hymns of thousand names’ (sahasranāmastotra/stava)
recited in praise of Śiva, Vis.n.u or Durgā so as to worship them and make them
50 Japa
favourable. It is mainly in the Purān.as that such stotras are to be found – for
instance, the Devīmāhātmya (or Durgāsaptaśatī). Those, however, are all series of
different names or attributes of the same deity, whose mere recitation is therefore
not strictly speaking a japa, but it becomes one if it is repeated, the repetition
numerous times of the same name being a nāmajapa. This last – the repetition
of the divine Name to devotional ends – seems to be found for the first time
in the Bhāgavatapurān.a (6.2,7–12), the great text of Vais.n.ava bhakti, the
Āgastyasam . hitā (7–71) of the Skanda Purān.a doing the same for the Śaivas.
122

These two are not ancient evidences of the practice: did it exist earlier? We do not
know. It is certain, however, that it developed from that time (twelfth to thirteenth
century) onwards in Śaiva and Vais.n.ava milieus among devotees of various forms
of the goddess;123 more recent periods show cases where the japa of a deity over-
steps the limits of its ‘sect’. Thus the Nāthas, who are Śaivas and who practice
mainly the ajapājapa, perform also the japa of Rāma’s name.124 One can also
hear nowadays, in South India, bhajans of Radha-Krishna organised by Smārta
Brahmins, who are Vedānta non-dualists, disciples of the Śan.karācāryas.125
In fact, even though the name of any deity may, in principle, be repeated in japa,
in actual practice the name of Rāma is the one mostly used in this way, Rāma (or
rather Rām), the name of one of Vis.n.u’s avatāras, coming to be considered, espe-
cially in the northern part of India, as the name of God. For Kabir, for instance, and
for the Sants, rāmsumiran (rāmasmaran.a) is synonymous with nāmsumiran, this
being a consequence of the popularity in India of the Ramāyāna epic. The solitary
or group, murmured or chanted, repetition of Rāma’s name is thus recommended
as one of the best devotional practices by the Adhyātmarāmāyana (fourteenth to
fifteenth century), the sacred Book of the Rāmānandins, highly valued in Rāmaist
sects whose influence in north India has been very important. Those, this text
says, who chant everyday ‘Rāma Rāma’ do not fear death; they will be saved even
if they kill a Brahmin; the gods themselves seek happiness by chanting Rāma’s
name.126 It may even suffice simply to remember this Name to be saved. Tulsidas,
in the Ramcaritmanas, the Hindi Rāmāyāna, goes even further: for him, Rama’s
Name is infinite (ananta), without equal (apara), revealed in the kali Age, it is in
reality ever present, eternal; revealing Rāma under his two aspects, with and with-
out qualities (sagun.a and nirgun.a), it transcends both. But the Name is also near,
for it only transmits to the devotee the essence of the god he worships. To repeat
the Name is therefore one of the ways to salvation.127
Another very popular practice, developed from the sixteenth century, is that
of the nāmasam . kīrtan, the repetition of the name of Kr◦ s.n.a associated with either
Rādhā’s or with Hari’s (Vis.n.u’s) name. Akin to japa, it nevertheless differs from
it since being a collective and sung, sometimes danced, repetition; its sequence
and atmosphere are quite alien to that of japa which, in essence, is a solitary
whispered repetition. It is mentioned here because there are cases – that of the
Haripāt.h, for instance – where Harināmasam . kīrtan and Harismaran.a (with the
ajapājapa) meet in some respects. Namasam . kīrtan is practiced mainly in Krishna
religious groups and especially in the Bengali gaud.īya tradition which goes back
to Caitanya (1486–1532), for whom śravan.a and kīrtana, the remembering and
 Japa 51
the chanting of the divine Name, are the two primary elements of bhakti. For
Caitanya, the mere (even involuntary) utterance of the name of Kr◦ s.n.a can bestow
liberation.128 Of all possible ways of worshipping Kr◦ s.n.a, he used to say, the best
is nāmasam . kīrtan, a perfect worship which fosters divine love. It is a collective
129

very emotional action, very far from the ritual practices and speculations we have
seen previously. Such devotional behaviour is to be found with Surdas and most
of the Sants, the saint-poets of the Indian Middle Age. The Sikhs, too, recommend
and practise the ‘proximity of the saints’, the satsang, which is often a form of
sam . kīrtan, of common devotional singing of the name or names of the Lord, Rāma
usually.
An interesting form of japa – a popular but also refined one – is that of the
Haripāth, the ‘Invocation of Hari’, of Jñāndev/Dñyandev, the saint-poet of Mahar-
ashtra (end of thirteenth century). A devoted Vais.n.ava, he nevertheless had links,
through his master Nivr ttināth (or Nivr ttidev), with the Tantric tradition of the
˚
Nāthas, theoretically founded ˚
by Gorakhnāth, where, as we have seen (above,
p. 44) ajapājapa was practised.130 The Haripāth is a short text which is to be rhyth-
mically chanted with intervals and followed by the singing of devotional hymns.
It is therefore not a japa properly so called. But the Haripāth prescribes several
times to repeat ceaselessly the Name of Hari: ‘Repeat the Name of Hari, and all
fetters are destroyed.’ The very name Haripāth designates such an invocation,
pāth being understood in the traditional sense of recitation or study of a sacred
text and having therefore the meaning of harināmakīrtana. Haripāth and harijapa
appear thus as synonymous.131 In another, analogous, case, that of Jñāneśvar’s
Jñāneśvarī, the Name is not to be merely uttered, it must be interiorised, for the
japa must not be a mere mechanical repetition; it must become an ‘inner audition’
(śravana – hariśravana) of Hari’s Name, that is, a devotional meditation which
unites the devotee with the deity. On this subject, the Jñāneśvarī says: ‘All that
my devotee does is none other than the worship [of me]; all that he thinks is my
japa; all that he is, is but absorption in me’.132 Such expressions of the greatest
devotion, of a ‘participation’ in the deity, remind one of the Śivasūtras (3.27) and
of Ks.emarāja’s Vimarśinī which we have seen before. It is likely that Śaiva non-
dualistic notions transmitted by the Nāthas have played a role here – the yoga tra-
dition of the Nāthas (and Tantric notions) being taken over in a bhakti context.
A similar influence of the Nāthayogins might probably also explain the pres-
ence of the ajapājapa in Kabir, who mixes or confuses it with the remembrance
of the divine Name (smaran, sumiran), and the fact that he mentions the so’ham
as well as the two letters Ra and Ma of Rāma.133 Probably another larger and more
ancient tradition plays a role, insofar as for Kabir (as well as for other Sants of
northern India) only the ‘sound’ (śabda), the word received from the mouth of the
Master, or revealed by the satguru to the pious disciple, can bring to the latter a
divine revelation;134 the satguru being the Supreme Being ‘without characteristics’
(alaks.an.a), ‘without colour’ (nirañjana). For them, the name of Rāma (rāma rāma
– Rām Rām) is the basic mantra, the mūlamantra of the godhead, which holds
within itself the supreme Reality, the two syllables of Rāma (or the monosyllabic
Rām) being sometimes conceived as equal to OM . , the symbol of the Absolute.
135
52 Japa
The infinite power of a revealed word or of the name of a deity, the harnessing for
reaching salvation of this power by the utterance and repetition, are ancient Indian
notions pervading mantraśāstra: they evolve, take on different aspects in the
course of time, but, perhaps underlying, they are always there, solidly rooted.136
As most of the practices mentioned developed after the eleventh century (and
especially since the fifteenth to sixteenth centuries), it is possible to speculate
whether their development was influenced by the presence of Islam in India. One
may indeed consider that such interiorised devotions were more easily practised
out of the temples (which, in addition, were often destroyed by Muslim rulers) than
those involving visible ritual action. On the other hand, it has also sometimes been
thought that the nāmajapa, where the name of the deity is often to be repeated,
associated with the control of the respiratory breath, and often in a group, which
is contrary to the traditional conception of japa, may be due to an Islamic influ-
ence. In effect, the japa of the Name, insofar as it is a remembrance (smarana,
sumiran) of the deity, and follows the rhythm of the breath, it is very reminiscent
of the dhikr (a term that Henri Corbin used precisely to translate by ‘mémoration’
– remembrance – and which, like sumiran, is both remembering God and men-
tioning this remembrance). In both cases, starting with a repeated utterance, the
pronunciation tends to fade, leaving only a presence felt in the heart.137 But like-
ness or similarity does not prove influence, and if there was actually influence, it
could well have been that of japa on dhikr,138 if only because the existence of japa
is proved long before that of dhikr. Contacts between Hindu and Muslim mys-
tique in India were numerous and important. Many circumstances, translations
and a number of texts testify to the fact. It seems, in fact, that Hinduism has more
frequently marked Islam than the other way round. One could easily quote in this
respect the names of many ‘saints’, mystics, or poets of different sects. The case
of the Baùls is well known.139 The Nāthas, too, whom we have mentioned previ-
ously, seem to have played (especially in the case of japa) an important role in this
exchange of ideas and practices.140 On the subject of the devotional repetition of
the name of God, instances from the Western world could also be quoted. But this
is not our subject here.141
To come back to the Indian domain or to neighbouring fields, we can mention
before ending some Buddhist practices akin to the nāmajapa such as those found
in Mantrāyāna texts, especially in the ones concerning the cult of the Amitābha
and Bhais.ajyaguru Buddhas. Amitābha’s cult spread in China where the invoca-
tion of his name (Amita) was recommended by the Pure Land School as the surest
means to gain salvation. Repeating ‘Homage to Amita Buddha’ sufficed to ensure
a good rebirth, by the sole effect of this buddha’s grace, without any necessity of
any personal effort. Like the Hindu nāmajapa, this prayer could be practised alone
or in common, be mentally repeated, chanted or sung. The method became quickly
popular. In Japan can also be found Amidism which prescribes the repetition of
the nembutsu ‘namo amida butsu’, a copy of the Sanskrit corresponding formula.
This doctrine and practice appeared in Japan in the seventh century. They were
popularised in the twelfth century by Honen. ‘So as to gain birth in the Pure Land,
nothing is as effective as the repetition of the Name of the Buddha’, he used to say.
 Japa 53
(In that sect there is even a danced nembutsu, adori nembutsu, which is reminis-
cent of the Krishna sam . kīrtan.)
We will conclude these few observations on those trivialised, popular forms
of the Buddhist japa, by referring to what is probably its most well-known form,
namely the repetition of Om . mani padme hum . , a mantra in devotional homage to
Avalokiteśvara, which one meets everywhere in Tibetan Himalayan Buddhism,
the human repetition of mantras being taken over by writing or engraving: it suf-
fices then to look at such formulas piled up so as to form ‘man.i walls’, or those
which, written on banners, scatter their beneficial messages in space, or else those,
inscribed by thousands on strips of some material, who are multiplied by the whirl-
ing of the prayer-wheels. All this is very far from the metaphysical speculations on
the power of the word, or from the difficult practice of bhāvanā, which we have
seen here. But in all these cases, popular or erudite, one remains in the domain of
the belief in the power of the word and in the efficacy of ritual, both to be found in
India from the most ancient times but also ever present elsewhere, since humans
believe everywhere in this efficacy. This belief underlies all the mental construc-
tions and actions we have seen in this chapter, even those where devotion domi-
nates. This is why japa is not strictly speaking a prayer:142 it is also and mainly
something else; this is why one cannot translate the term, why, too, japa is such a
diverse, rich and interesting practice.
4 Nyāsa
The ritual placing of mantras1

In practice, mantras are nearly always used in the context of ritual, indispensa-
ble, as we have seen, to their extraction or selection, to their transmission, or to
their efficacy in action. We intend here to inquire into one ritual practice, that of
nyāsa.
The word nyāsa is formed from the prefix ni (‘below’, ‘under’) and the verbal
root AS (‘to throw’, ‘to project’) – from which the verb NYAS, nyāsati, to throw, to
project, is derived along with the masculine substantive nyāsa, translated in Mon-
ier-Williams Sanskrit–English Dictionary as ‘putting down or in, placing, fixing,
inserting, applying … drawing, painting, writing down … depositing, intrusting,
delivering … mental appropriation or assignment of various parts of the body
to tutelary deities’ – a diversity of definitions one also finds in the St Petersburg
Dictionary: ‘niedersetzen, hinsetzen, aufsetzen … das Auftragen mysticher
Zeichen auf verschiedene Teile des Körpers’. P.V. Kane (History of Dharmaśāstra,
vol. V.2, p. 1120) writes: ‘nyāsa … means mentally invoking a god or gods, man-
tras and holy texts to come to occupy certain parts of the body in order to render
the body a pure and fit receptacle for worship and meditation’. He adds a little fur-
ther on: ‘The Kulārn.ava explains it as follows: “nyāsa is so called because therein
riches that are acquired in a righteous way are deposited or placed with persons,
whereby all-round protection is got”’ (nyayoparjitavittanām an.ges.u viniveśanāt/
sarvaraks.akarād devi nyāsa ityabhidhīyate) (KT 18.56, p. 352).
We can also cite Bhāskararāya who, in his commentary on the Lalitasahasranāma
(śl. 4) defines nyāsa as the placing of divinities on different parts of the body,
this placing being made through mental concentration (bhāvanā): nyāso nāma
tattaddevatānām . tattadavayaves.vavasthāpanam, avasthitvena bhāvaneti yāvat.
2

We can see emerging in these definitions – apart from the ordinary, banal sense
of the word ‘to place’, ‘to deposit’ – different meanings or nuances which appear
in the mantraśāstra and which we shall meet here later, meanings that are all
around the notion of placing or depositing on the body or on an object a mantra
or some other sign that brings the presence of a deity, the transmission of a subtle
or ritual entity, of an energy or spiritual power. This transference is accomplished
by placing – generally but not necessarily – the fingers, hand, or hands on the part
of the body, or on the object or substance, where the entity must penetrate and by
which placing it is transformed. The operation is thus both mental and corporeal.
 Nyāsa: the ritual placing of mantras 55
Nyāsa is usually considered to be a rite typical of Tantric Hinduism and Bud-
dhism, since it is used for the manipulations and assimilations into the body of the
sacred that are characteristically tantric. We find indeed a considerable variety of
such practices in all ritual texts, in the tantras, āgamas and so on, as in the treatises
on yoga and spiritual practices (see the Tantrāloka (TĀ) by Abhinavagupta for
example). The word itself in this technical sense, however, hardly ever occurs
before the āgamic and purānic works (which are already ‘tantricised’), nor does it
occur in ancient Buddhist texts. Some placings are prescribed in relatively ancient
ritual works whose spirit is not Tantric. One thinks here of the Gr. hyasūtras and
above all their pariśis..ta (thus, for example, the Baudhayānagr.hyasūtra II, 18 or
IV, 7 which does not use the word nyāsa but sam-spr.ś), or a number of stotras.
Moreover, as is the case for other practices and words, we can hardly base an
argument about their origins on the silence of the texts: these practices could well
have existed before being textually attested. We are therefore permitted to think
that these are elements from very ancient times that certainly existed in one form
or another in a non-Tantric context well before the tantras and āgamas.3
The practice of transmitting a spiritual or magical force by touching not only
exists in Hinduism and Buddhism (Tantric or not) but also in other Indian religions.
It is almost universal.4 We find a number of examples in the Judeo-Christian tradi-
tion where we have the transference of the sins of Israel to a scapegoat (Lv., XVI,
21), the transmitting of the spirit of wisdom (Deut., XXXIV, 18), or the bestowing of
the holy spirit on the apostles through the laying on of hands (Acts, VIII, 18–19), etc.
We find analogous acts in other traditions along with speculation on magical power
or the symbolism of the hands and the symbolic value of gesture.5 Nyāsa is one case
among others, however with the particular association of belief in the power of ges-
ture and in the efficacy of speech such as developed in India.6
The nyāsa, then, can appear as a gesture in support of speech, as a gestural and
corporeal participation in speech. The act of placing carries and posits on a point
of the body – or on an object, although this is less frequent and the object is usually
an image of the corporeal aspect of a deity – a mantra which will impregnate this
body or object with its energy thereby transforming or divinising it.
We have already noted in this regard, and we will return to this later, first that
the hand which performs the nyāsa is itself often the object of a preliminary nyāsa
(karanyāsa) which purifies it and charges it with a force that it can transmit.7
This hand must make a particular gesture – a mudrā – when it imposes a mantra,
‘sealing’ it thus as it were on the proper place. We could here raise the question of
the origin of the mudrās which, like mantra and nyāsa, are both typical of Tantric
ritual and probably as old as the religions of India. But there is also the problem
of the deep connection between mantra and mudrā, between gesture and speech.8
Finally, if one is interested in the archaeology of signs, there is the problem of the
first origins of the link and that of the precedence of one element over the other: in
the beginning was there word or action ...? Many questions we cannot tackle here;
we simply note that they exist.
But the problem of nyāsa – we repeat – must not be limited to ritual touch-
ing, which is an important aspect, certainly, but not the only one. In so far as
56 Nyāsa: the ritual placing of mantras
nyāsa consists in placing divine entities on (or in) the body, in divinising or cos-
mising it, and, by transforming it, in giving it a role in the cosmic and/or9 divine
dimension of such acts as the pūjā, yoga or Tantric meditation, we find ourselves
in the complex Indian problematic of the relationship or connections of body–
universe–divinities–God. We know that this universal, total vision of the whole
mass of cosmic manifestation, is very ancient. Tantrism insofar as it underlines
and organises – or re-organises – these correspondences, refers back to a vedic
conception: ‘All the gods reside in the human body as the cows in a cowshed’ says
the Atharvaveda (AV XI, 8, 32). We can add that in this ancient and always per-
sistent Indian conception, if the body reproduces the structure of the cosmos, the
cosmos in turn is modelled on the human body,10 both being governed by a princi-
ple, at the human (ātman) or cosmic (purus.a) level, which is, in essence, identified
with the Brahman to which one is eventually to achieve identification through
asceticism without leaping over the intermediate stages, notably that of the gods.
Hence the usefulness of nyāsa11 as a means to achieve this identification.
The nyāsa, as all the practices of Tantric yoga focused on the body, underlines
its importance in achieving liberation. In a system of thought in which there is
no total distinction between the human condition and other states of existence
(animal, notably, but also divine), it is often underlined that only in the human
body can liberation be attained: this is one more justification of corporeal ritual
practices – practices which must be understood in the context of another basic
Indian conception of humankind: the absence, namely, of a division between the
bodily and the mental, or between gross (sthūla) and subtle (sūks.ma); the differ-
ence between them being one of level rather than one of nature. Man is thus not
separated from the cosmos, which itself emanates from a divinity that operates in
the fashion of a yogin, the whole being founded on an absolute without form, at
the same time transcendent and immanent. Such a man, too, from the highest level
of his individual consciousness to his physical body, is constituted by more or less
subtle ‘bodies’, namely causal, gross, subtle bodies (kārana, sūks.ma and sthūla
śarīra) whose structures correspond. Thus, if nyāsa is made by human hands on
the body’s surface, the effect of this action which is mainly mental or spiritual
bears as much, or even more, on the subtle body than on the gross body. This
appears clearly in the case of placings (antarmātr.kanyāsa, for example) done on
the cakras, which do not have a physical reality but are ‘localisations’ in the gross
body of elements of the (so-called) ‘subtle body’,12 as we shall see later.
We shall also see nyāsas associated with Tantric yoga, which, while using phys-
ical methods, rests theoretically as in its practical application on the ‘mystical
physiology’ of the cakras and nād.is where circulate the various currents of prān.a,
the ‘breaths’, whose nature is both bodily and subtle, and also cosmic insofar as
prān.a is a form of divine energy (śakti).13 We could say that we are here in a psy-
chosomatic domain, were it not that this term implies a dichotomy between psyche
and soma which India ignores and always adds a cosmic dimension.
But if the nyāsa is in practice, and most often in the mantraśāstra, the placing
of phonic elements on the body, we must not lose sight of the fact that in ritual,
where their usage is so frequent, nyāsas serve also to place divinities or entities
 Nyāsa: the ritual placing of mantras 57
on the instruments of worship or on various points of the ritual area. That is, they
also appear as a way of placing materially (if one can say this in spite of the fact
that the mental and linguistic aspect is present) something, without reference to
the body of the officiant.
Finally, an essential element of nyāsa is the form of speech which it serves to
place on a body or an object which it penetrates and impregnates by its energy.
One must therefore recall the importance of speech (vāc) in Tantra and how it is
conceived. Speech, which is the divine energy itself or the energy aspect of the
deity which is supreme Being and supreme Consciousness, emits the universe,
animates it and reabsorbs it. It is difficult to be precise about the essential and
ultimate nature of this speech – is it sound or consciousness? Explanations vary
according to the different schools. Perhaps it is possible to say – if one may adopt
a non-dualist Śaiva perspective – that there is on the highest level an ultimate
Reality, pure consciousness, where speech is in seed, being co-essential with it and
being its aspect as energy. This speech, which is at the same time consciousness
and energy, evolves becoming more explicitly sound, then discourse, manifesting
thus the universe, bringing it into existence by saying or naming it. Speech is ‘that
which says or expresses’ (vācaka) all things, which are ‘that which is to be said,
or expressed’ (vācya). Such is the correspondence between mantras, or phonic
seeds (bīja) and the divinities which they represent (or rather of which they are the
subtle and essential form).14 In nyāsa, it is this phonic or consciousness element
– we can say this spiritual force – having a sonorous form for support (itself the
essential form of the deity), that is transported and deposited by the hand with the
ritual gesture of placing (or through the mental concentration which accompanies,
or replaces, this gesture).
If there is a gesture in nyāsa, how is one to understand the link between this
purely physical act and the transfer of spiritual energy? Of course, seen objec-
tively, this is merely one form among others of the manipulations of symbols or
of organs which many cultures use to transform the body (or to modify its image),
or to act upon objects, practices which are cases of symbolic effectiveness. But
if this is for us a clear and simple explanation of the functioning of nyāsa, it was
felt as an important problem within Indian thought for the practitioners of nyāsa,
in terms of their problematic: there was for them an effective – one could say
material – transfer of the phonic and spiritual force of the mantra and a ‘sealing’
of it through a gesture on the chosen place; not simply an accompanying ges-
ture and still less a symbolic operation. The question, in fact, is hardly discussed,
the practitioners not being accustomed to putting in doubt the received tradition.
Some of them, however, did ask themselves what is really the nature of nyāsa: is
it a mental or spiritual operation, or a material one? Or rather, what is the part of
each of these two elements? This is done, for instance, at the beginning of chapter
4 of the Pārameśvarasam . hitā, entitled ‘Rules concerning the placing of mantras’
(mantranyāsavidhi)15 where we read:

The mantrin, his body purified, is to perform the placing of mantras which
identifies him with the god of gods (devadevasamo bhavet) and makes him
58 Nyāsa: the ritual placing of mantras
qualified to accomplish all ritual action such as pūjā and so on, [placing] thanks
to which he will appear [in or for himsef] all perfections or supernatul pow-
ers (siddhi). Having accomplished this, he will be without fear among wicked
people and will be able to triumph over all dangers. Although what is called
nyāsa must be considered a mental operation, it does not create a [new] situa-
tion immediately and outside of the [material] action [which constitutes it].16

This last sentence is very explicit. The intellectual, ‘mental’ (mānasa) component
of nyāsa is what the tradition considers essential. It is the concentration of the
practitioner’s spirit that is effective: it is the concentration of his mind on what
he is doing, that is, his one-pointed mental concentration on the phonic and spir-
itual assemblage comprising the mantra (or the bīja) that he places, or the mental
representation, the visualisation (dhyāna or bhāvanā) of the entity to be imposed,
or the inner experience or feeling in the self of the transformation effected by the
nyāsa.17 Since the placing acts on the plane of the so-called ‘subtle body’,18 or that
of consciousness, and due to the element of speech (the imposed mantra) whose
nature is divine energy, there is a natural pre-eminence of the mental or spiritual
element. Parallel to this, there are actions (kriyā) described in the manuals, that is,
the gestures of the placing of which there can be many (for example, if one must
impose all the mātr.kā), gestures which are equally indispensable as nothing is
effective without their being performed.19 But these actions are subordinated to
the most important element which is spiritual or mental, and without which they
would remain without effect: what is mental is always superior, in India, to that
which is material or perceptible to the senses. This, however, without losing sight
of the fact that there is no difference in nature between the bodily and the mental,
but rather a difference of level. Without forgetting too that the nyāsa is generally
carried out by the adept after a ritual or spiritual exercises which purify his body20
or which purify and divinise the hand which performs the nyāsa so as to render it
capable of serving this spiritual manipulation.21
This nature of nyāsa, without being clearly explicated, is clearly implicit in
many texts. All, for example, which prescribe before the nyāsa to concentrate
on the divine or otherwise entity to be placed, or which say that one must first
reabsorb and assimilate the mantras in oneself (that is to say, what they symbol-
ise) before imposing them. Thus in the ŚT (chapter 5, śl. 118–120, in relation to
the varn.amāyī dīks.ā), the master first identifies with the divinity, uniting himself
with the supreme self, and then imposes the phonemes of the Sanskrit alphabet on
his disciple, impregnating him with their spiritual force and identifying him with
the divinity: he transmits through nyāsa the condition which he had previously
brought about in himself.22 In all cases, too, where what is imposed is not a mantra
or bījā but a figure (the śrīcakra for example – thus in YH, chapter 3), or when a
mantra is made of a group of unpronounceable letters such as the navātmamantra
RHRKSMLVYUM (TĀ 15, 239f. see also here, p. 60), the essentially mental or
spiritual character of the operation of nyāsa is evident.
There exists, finally, complex nyāsas which are more tantric meditations than
placing properly so called, where the adept must first visualise (and feel) that all or
 Nyāsa: the ritual placing of mantras 59
part of his body is transformed by the imposed mantras and becomes identified with
energies, cosmic divisions, etc. There is also the case where the nyāsa of various
energies is done in the ‘breath’ (prān.a) of a yogi whose body has first been divinised
by other nyāsas,23 etc. In cases of this sort we can ask whether we are still in nyāsa,
but this would be to forget that nyāsa always has a mental or spiritual aspect which
is often (notably in more complex cases) an operation almost uniquely intellectual.
This explains why the texts prescribing the placings, in place of nyaset or vinyaset,
use the verbal forms smaret, cintayet, dhyayet, bhāvayet or kalpayet.24
Let us finally note that alongside such terms as underline the mental aspect of
nyāsa, we also encounter in ritual texts, so as to underline the act of placing, other
words for nyāsa. Thus the verb KS.IP (to project, to place, to pour) can serve to
indicate the placing of certain entities on the body or on an object (for example,
TĀ, 16, 231, for the imposition on previously imposed tattvas; or TĀ 17, 2–3, the
projection of the three malas on the body of the disciple after the placing on him
of the adhvans): in effect these are nyāsas. The Indian lexicographers often gloss
nyāsa with ks.epa or niks.epa: thus Ks.irasvāmin in Amarakośa II, 9, 81, nyasyate
niks.epyate nyāsah.. One also finds niveśa and viniveśa, viniyoga, samarpan.a (or
YUJ: yojayet, or ni-DHĀ: nidhāpayet etc.). All these terms have near meanings:
the sense of to deposit, to place, to install, to apply, to put, or to consign to a place,
or else, to join, to conjoin – the words differ but refer to the same thing.
There are also texts using the verb LIKH, to write, when prescribing or describ-
ing nyāsa. This is very natural when it refers to placing mantras or bījas on a dia-
gram: one places and traces them there according to precise rules.25 Or, in certain
rites where one must draw a yantra and bījas on a liquid while reciting the bījas
in order to consecrate the liquid, thus for the preparation of the arghya for the
pūjā. But there are cases where the texts are more ambiguous. For example, in
chapter 293 (mantra-paribhās.a) of the AgPur (śl. 39–47), the ‘Lords of the letters’
(lipīśvara) and the energies of Rudra (represented by phonemes) must be written
then imposed (likhyā … vinyaset), without the text indicating how or where the
phonemes are to be written before being imposed. We do not have, it is true, an
entirely reliable text of this purān.a: the passage in question is perhaps corrupt or
incomplete: the uncertainty remains.26

To gain a complete sense of what nyāsas are, their nature and role, we must exam-
ine several types and different usages (their range and diverse effects). In effect,
as in nearly all Tantric ritual, Hindu or Buddhist, the nyāsas are used in a wide
variety of circumstances and conditions, from the highest spiritual practice to the
most vulgar magic (including an important zone where it is difficult to know if
one is in the religious or the magical, the two domains being difficult to separate,
especially in India: ‘There is no religion without magic and no magic that does not
contain a grain of religion’).
As the more complex nyāsas are those which best show the sense and scope
of this practice, we will first consider this type of procedure. We will then exam-
ine more rapidly diverse sorts of simpler nyāsas which we find in ritual worship
(pūjā), in spiritual practices or magic, as well as in the ritual acts of daily life. Not
60 Nyāsa: the ritual placing of mantras
that the sense of these diverse types of nyāsa differs: it is always the same, the
impregnation of the body or of an object touched by a divine or cosmic entity or a
spiritual force. But this ‘metaphysical’ dimension is more or less noticeable and in
particular more or less explained in the prescriptive texts: it is only for this reason
that I distinguish them.
Let us now look at the texts rich in ritual developments and in religious and cos-
mic implications. I will resort mainly to two works which are easy to have access
to, both of them Śaivite: the one Kashmirian, Abhinavagupta’s TĀ, the other also
connected to Kashmir and commented upon in the spirit of the Pratyabhijñâ by Amr.
tānanda, the Yoginīhr.daya (YH).27 For the TĀ we will look at passages from chap-
ters 15 and 16, the former dealing with the so-called regular initiation (samayadīks.
ā), the latter with the putradīks.ā, the initiation of a ‘spiritual son’.28 It may be useful
to recall in this connection that, in the TĀ, initiation is not limited to the transmission
of a mantra by a master to his disciple, a type of rite of passage giving access to a
particular form of spiritual life, but is a procedure enabling the initiated disciple to
realise, from the particular point where he finds himself (hence the diversity of dīks.
ās29), his identity with Śiva. These rites being all the more complex as the disciple
has a longer road to travel and more elements to transform and purify. Moreover,
the goal aimed at being usually liberation in life (jīvanmukti), being in all cases the
liberation of an incarnate being, the body (or rather, the bodies, gross and subtle) of
the initiand, as well as his thought must be transformed and divinized. Hence the
important role, in this process of that fundamental means of corporeal transforma-
tion, and of interiorised action, that is nyāsa.
The ritual of the samayadīks.ā as described in chapter 15 of the TĀ is particu-
larly long and complex. Nyāsas are used there in nearly all the rites and there are
moments when they are the essential part of the rite. To begin with, we find a
general placing (sāmānya nyāsa) of the fifty phonemes of the mātr.kā and of the
mālinī,30 after which the disciple to be initiated purifies his body and his ‘breath’
(prān.a) (that is, destroys in himself all that binds him to empirical life and duality)
through meditating on the flaming image of the astramantra PHAT. 31 (śl. 232–36).
He thus finds himself in the pure, divine Self (tis..thecchuddhātmani), immobile
and without waves, but where there nevertheless appears as a primordial wave the
image (mūrti) of Śiva formed by the mūrtimantra OM . HAM . . It is in this state,
32

the adept being already very purified and detached from empirical life, that the
‘special placings’ (viśes.a nyāsa) that interest us here will be performed.
These comprise, first, a ‘great sixfold nyāsa’33 of the ‘god Navātman’,34 that is
to say of the navātmamantra formed from nine letters RHRKS.MLVYUM . . These are
imposed on nine points of the body; followed by nyāsa on fifty points of the body
of the mātr.kā; then of the śiva-, vidyā- and ātma-tattvas (which cover the whole
manifestation) on the tuft, the heart, and the feet; then of the eight gods or mantras
Aghora etc. on the head, face, neck, heart, navel, generative organs, thighs and feet;
of the mantra god Bhairavasadbhāva, with his ‘limbs’ (an.ga) on the usual places
for nyāsa of the an.gamantras (heart, head, tuft of hair, chest, eyes – sometimes also
the ‘weapon’); and finally, on the same places, of the mantra god Ratikśekhara,
RYLVUM . , which must be specially worshipped at this moment in the rite.
 Nyāsa: the ritual placing of mantras 61
This first mahānyāsa is that of the various aspects of Śiva. It is followed by a
second series of six placings: of aspects of the Energy (śāktam . nyāsam śl. 248)
which must be made ‘on the preceding ones’, that is, on the same places. This is
important: it is necessary to exactly superimpose the two series of nyāsas in order
to make the two categories (male and female) of cosmic or divine entities coincide
and so to unite Śiva and Śakti (and thus have the divinity in its totality) in the body
of the adept. One therefore imposes on the navātmamantra the energy Parāparā
with her ‘limbs’ and ‘faces’ (sān.gavaktra); on the mātr.kā, the mālinī,35 on the
three tattvas of Śiva etc., the three aspects of divine energy (or, more exactly,
the three goddesses) Parā, Parāparā and Aparā; on Aghora etc., the eight energies
Aghorī etc.; on Bhairavasadbhāva and his an.ga, the five Vidyān.ga; finally one
must place on the same spots the Energy as Mātr. sadbhāva, Mistress of Yoga, in
her plenitude, supreme and unwavering, destroyer of time, the initiating master
having at the same time (as for Ratiśekhara) to worship her specially, mentally
evoking her ‘limbs’ and ‘faces’ and her retinue of twelve energies. For this is,
in fact, the energy at her most high and pure level, the supreme Consciousness,
present without any division in all other deities or entities previously imposed. It
is this supreme Śakti united with Śiva that the double mahânyāsa infuses in the
body and spirit of the disciple.
Abhinavagupta (śl 259–61) makes clear the scope of this operation in putting in
parallel the six parts of the double nyāsa and the five modalities of consciousness
(from jāgrat to turyātīta) to which he adds anuttara, the first principle, ‘without
second’, each of these six states being considered as being formed by itself along
with the five others. This makes 6 × 6, that is 36: the number of tattvas from Śiva
to earth. The nyāsa therefore has the effect of imposing on the body of the adept
all the modalities of consciousness (from the ordinary waking state to supreme
consciousness) along with the group of thirty six principles that constitute the
cosmos. He thus is identified with the totality of consciousness and the pure, cos-
mic emanation paradigmatically present in Śiva. The adept is thus identified with
the Supreme Śiva united with Śakti, pure,36 total Consciousness, dazzling, where
appears in all its diversity; but archetypally and without duality, the totality of the
cosmic manifestation (śl. 262–68).
It is interesting to note that at this point Abhinavagupta thought it useful to
approach the problematic of nyāsa. He notes in effect (śl. 268–269) that it is
paradoxical that an empirical method such as the placings can produce a result of
a transcendent order: the transformation of a man into a god. But this is, he says,
because nyāsas are operations of an interior, spiritual order: that which one thinks,
for good or bad, that one becomes. Whoever therefore thinks intensely (bhāvayati)
‘I am Śiva and nothing other’ becomes Him (śl. 269–270). ‘He creates for himself
there an unwavering certainty, an awareness in the depth of the heart, associated
with a thought without duality which, itself, generates a flood of thoughts bound
to duality [but all] orientated toward the identity [of the empirical subject] with
Śiva’.37 One thus comes to destroy all belief of another (mundane) kind and to
have only the unshakeable conviction that our nature is pure, eternal and divine.
The spiritual character of the activity of nyāsa cannot be better underlined.
62 Nyāsa: the ritual placing of mantras
Having by this sacrifice thus transformed his body, full of bliss, entirely away
from the world, the adept should meditate on the body as identical with Śiva.
Any connection with the world once destroyed and the limited condition
being dissolved, what remains in his body other than the essence of the bliss
of Śiva? Worshipping his body day and night and carrying in himself the
thirty six tattvas, the adept is identified with Śiva. Perfectly content he lives
in peace in this cosmic body.
(śl. 238–236)

Thus Abhinavagupta sums up the condition of the person who has received the
double mahānyāsa. What else, indeed, could he want?
This is not, however, the end of the rite of initiation, but only a necessary con-
dition for its complete accomplishment. The process therefore continues, first,
with a new group of nyāsas which is also worth summarising.38 There again the
initiand is to identify himself with Śiva. One could believe that this condition
has already been attained through the previous nyāsa. But it is completed here
through an ‘interiorisation’ by the adept of a group of aspects of the divinity or,
more precisely, of an ascending movement of energy which he will integrate with
his ‘breath’ (prān.a). He imposes, in effect, on his ‘breath’ a series of entities rep-
resenting the rise of energy from its lower base to the supreme level, a series visu-
alised as a trident extending from the point of departure of the breath in the area
of the navel, to a point twelve fingers above the head. This ‘breath’, conceived as
present in the physical body, is an element of the yogic structure of centres and
veins imagined as present (‘intraposed’ to use T. Goudriaan’s term) in the physical
body. The nyāsa thus, even if accompanied by hand gestures which impose the
entities, is essentially internal, mentally visualised. It completes the preparation of
the adept to the mental and material acts of worship that are to follow by making
his ‘breath’ and his consciousness – and not only his body – the seat (āsana) of
the deity.39
He first places the Power that supports (ādhāraśakti) four fingers below the
navel, then on top of that the four elements: earth, water, fire and air together with
the fifth one, ether, the whole occupying four fingers and being imagined as form-
ing the swelling (āmalaka) near the base of the trident. The latter, called ananta,
rises above the swelling through twenty-four finger-spaces from his navel to the
uvula (the cakra of the palate), being made up of the twenty-four tattvas from
tanmātra to kalā. Then, on the knot formed by the māyā tattva, one finds the
eight qualities of consciousness (dharma, etc.) placed on a square man.d.ala in the
four directions of space and their intermediaries. Just above this knot (between
the uvula and the brahmarandhra), the initiand is to meditate on (dhyāyet – not
nyaset) the śuddhavidyā tattva in the form of a lotus with eight petals: on these
and in the calyx one places and worships the nine energies Vāma, and so forth,
from right to left (pradaks.ina), then the nine energies Vibhvī, etc. in the reverse
order. After this the adept is to evoke (smaret) the sun, moon and fire with the
gods Brahmā, Vis.n.u, and Hara placed respectively on the petals, the stamen and
the calyx of the lotus. Above these he places Īśvara, then above him (at the level of
 Nyāsa: the ritual placing of mantras 63
the top of his skull) Sadāśiva, the mahāpreta, lying as a blazing corpse, emaciated
but filled by the laugh of destruction.40 He should then evoke (smaret) three shafts
of light which, rising up from the navel of Sadāśiva, go up from the brahmarand-
hra – these are the three energies śakti, vyāpinī and samanā – forming the trident
(trisūla), they spread out and up to the plane of the dvādaśānta, twelve finger-
spaces above the head, reaching the level of unmanā.41 where the three supreme
goddesses of the Trika are seated each on a Bhairava lying on, a white lotus. The
initiation ritual goes on, a cult being then to be performed on the lotuses to this
supreme plane of the deity.42
We have here, then, a nyāsa where cosmic and divine entities are imposed
(placed, seen, and felt) and the image to be visualised of a trident. This implies a
double identification with the one and with the others and implies that the visu-
alisations and meditation occupy the main place, with physical touching, insofar
as there is any, having only a secondary role. This passage well illustrates the
role that nyāsa has in a spiritual practice that rests on mental representations; and
how the hand gestures on the body (when there are such gestures) bring about,
through the representations and mental effort which accompany them, a transfor-
mation of the body image, the ‘cosmicisation’ of the performer and his assimila-
tion to the divine.
In chapter 16 of the TĀ, on putrakadīks.ā, there are interesting passages (śl.
77–163 and 207–247) where the cosmic sixfold division of the ‘ways’ (adhvan)
and the mantras which express and are used to purify them are imposed on the
body of the adept. There is here again a complex nyāsa in which mental repre-
sentations play a crucial role. It would merit being examined, but to do this here
would give too much space to this kind of operation.43 On the subject of complex
nyāsas, it will be better to look at those which are more directly linked with the
cult and use of a man.d.ala (or cakra).
There is a good example of this in chapter 3 of the YH (śl. 8 to 92), which describes
the cult of the goddess Tripurasundarī which, as with the rites we have seen, has as a
goal not only her worship but also the identification of the performer with the deity.
The ‘external’ cult allows – thanks to the ritual and above all thanks to the nyāsas
– the adept to attain this identification that those more talented or more favoured by
divine grace would attain merely by contemplation (bhāvanā: śl. 5–6).
The adept must first make a sixfold nyāsa (s.od.hānyāsa)44 of: 1) the fifty Gan.
eśa, placed where one normally imposes the mātr.kā; 2) the nine planets, imposed
on various points of the body with each of the eight series of phonemes plus ks.a;
then 3) the naks.atra, associated with phonemes; 4) the six groups of Yoginīs on
the six cakras from mūlādhāra to ājñā, with syllabic bījas; 5) the twelve signs of
the zodiac with groups of phonemes; 6) finally the fifty pīt.hās associated with the
fifty mātr.kās.45 One thereby achieves a first ‘cosmicisation’ of the body of the offi-
ciant since one finds there the stars, places of sacred geography, together with such
forms of divine energy as are the mātr.kā and the bījas. The officiant thus already
transcends his human condition.
He must, however, still identify with the Goddess and, to this end, he will
become assimilated to her as she resides, surrounded by secondary deities, in the
64 Nyāsa: the ritual placing of mantras
śrīcakra used to worship her (since she abides at its centre), but which also, and
mainly, symbolises her cosmic dynamism as she eternally emits and reabsorbs
the universe. The adept therefore imposes the cakra on his body, the nyāsa bring-
ing him to merge with the deity in total non-duality (therefore going back to the
origin). The ritual process unfolds going from the outer part of the cakra to its
centre.
First, the process requires the outside line of the square surrounding the cakra
to be placed on ten points of the body with the mantra ‘homage to the outside line
of bhūpura’. Then, invoking in the same way the two other lines of the square, he
imposes on the same points of the body these lines as well as ten siddhis, then the
ten mudrās which also reside there. In the same way he imposes the other parts of
the cakra (the two lotuses with sixteen and eight petals, then the four concentric
series of small triangles formed by the intersections of the nine triangles that are
within the cakra) on various points of the body, with a mantra of homage to each
of these parts. He then imposes on the same points the divine entities residing
in each part of the cakra. This is thus a double operation, identifying the body
with the cakra by the placing of its nine parts, then achieving the divinisation by
placing all the aspects of the goddess associated with the cakra.
Then the adept imposes in his heart, with the mantra ‘homage to the tri-
angle’,46 the central triangle of the cakra on the outside of which are the three
goddesses Kāmeśvarī, Vajreśvarī, and Bhagamālinī and in the centre the goddess
Tripurasundarī.
But the process does not stop there; another series of impositions must still be
made on the body, this time going from the centre to the exterior, from the mūladevī
to the secondary divinities and to the siddhis, from the central bindu to the outer
square. This being done, the adept comes to realise that the goddess, Mistress of
the gods – says the commentator – and his own self are identified, inseparably
united, the supreme reality, dazzling and vibrant, ‘the unifying fusion of light and
consciousness’ (prakāśavimarśasāmarasyarūpa). Being the cause and foundation
of the process of nyāsa, the goddess brings the adept to unity with the divine
self (śl. 83); it is therefore with her that the nyāsa reaches its ultimate aim.47 The
process, in fact, does not end there, but it is not necessary to follow it further.48
If nyāsa is essentially a mental or spiritual practice – especially in cases where
its scope and aim are the most profound – it is nonetheless, as a general rule,
accompanied by a gesture and consists, in the most common practice, in the man-
ual act of placing a mantra – hence the importance, already mentioned above (pp.
55–56), of everything relating to the hand; though, probably, as we have already
seen with the theory of nyāsa, the role of the hand appears as subordinate to the
mental operation. One can say that above all it is a symbolic act (which would
not be disparaging, for anything related to mantra is precisely of symbolic effec-
tiveness). It is, on the other hand, in more elaborate philosophical texts (not very
frequent in a Tantric milieu – except, of course, in the Saiva exegesis) that the spir-
itual foundation of nyāsa is expressed. A less intellectual thinking will naturally
give the hand a directly effective role. It goes without saying that, for most adepts
of yore or of today – be they devotees, mantrins, or magicians (very numerous,
 Nyāsa: the ritual placing of mantras 65
those!) – the hand is really charged with the influx that it places. But in Tantric
texts too (the TĀ, for instance) one often finds side by side these two levels of
practice or of religious discourse. These works thus prescribe nearly all the rites
for purifying the hand which performs the nyāsa and for charging it with the effec-
tive force it must transmit. This is in particular what justifies the rite of karanyāsa
or of karān.ganyāsa (placing on the hands, or placing of the an.gamantra on the
hands) often prescribed first when a series of placings is required.49 One thus finds
at the beginning of the Śaiva samdhya the sakalīkaran.a50 rite,whose goal, accord-
ing to Aghoraśiva, is first to penetrate the hand with divine energy (sakalahastavy
āptaśaktitvena), then with this transformed hand, to identify different parts of the
body of the officiant with the ‘parts’ (kalā51) of Śiva.
The fourth chapter of the Pārameśvara-sam . hitā (after the fourth śloka, which we
have seen given a definition of nyāsa as a mental process) describes a karanyāsa
(or hastanyāsa) characteristic enough to merit being reproduced here:

Given the role of the hands in the placings, one should begin with their nyāsa.
Thus ‘imposed’, the hands are charged with a luminous force. One places there,
indeed, twelve phonemes52 from wrists to fingertips, that are thereby as full of
light and heat as the sun (ādityatapavat). The rule is that nyāsa is done three
times: according to emanation, conservation and resorption. In the sr.s..tinyāsa
(emanation), one imposes the tāra (mantra) on the palm of the right hand, then
on the thumb, etc.: the ten fingers, in order, up to the left little finger, thus
receive ten phonemes, the twelfth being placed on the left palm. In the nyāsa
of conservation (sthiti), one begins with the left thumb, then the four other
fingers, and ends with the palm, imposing six phonemes beginning with the
tāramantra. Then follows in due order the six other aks.ara on the left hand,
from the thumb to the palm. (The sam . hāra – resorption – nyāsa is made from
the left palm to the right.) After this, one must impose the rest of the phonemes
ending with the pran.ava. The nyāsa must be made with the index finger on the
thumb of the same hand, then with the thumb on the index and the other fingers,
going from the right thumb to the left little finger, then the two palms, right
and left. One must then impose the mantras hr.daya etc., the twelve luminous
members (an.ga) (of Vis.n.u) right in the middle, the lotus with shining rays, the
club shining with its own brilliance on the left palm, in the right the luminous
disc, the conch in the left palm, the kirit.a on the right hand, the śrīvatsa in the
middle of the left, the kaustubha on the right palm and the rosary on the left, Śrī
on the right hand and Pus..ti on the palm. He must then impose the garud.amantra
on the ten fingers, going from the right thumb to the left little finger. This is the
rule of the hastanyāsa. By this the supreme energy of the Lord will penetrate to
the centre of the lotus of the heart where it is transformed into vital breath; then,
dividing itself into ten, it extends up to the hands (pān.imārgena nirgatā), where
it fills the ten channels (nād.ī) which reach to the fingers.53

In the Śaiva context one finds, as well as the karanyāsa already mentioned, a
particular rite of the ‘hand of Śiva’ (śivahasta) used, above all, for initiation (and
66 Nyāsa: the ritual placing of mantras
sometimes in worship): it transforms the right hand of the master into a divine
hand. Abhinavagupta mentions it in relation to the samayadīks.ā in the fifteenth
chapter of the TĀ:

The master should, with the left hand, worship on the right hand the luminous
wheel of mantras, containing all the ‘ways’ (adhvan) and destroying all the
bonds. He places it on the head of the disciple, on which the ‘ways’ have been
first imposed, then touching his whole body, this hand shall destroy all bonds
which bind him, and [the disciple] will then be an initiated samayin, shining,
united in his lifetime to the Supreme Lord.54

In the offering of the pavitras, described in the SP, the master transforms his hand
by means of various anointments and imposes there the an.gamantras.55 This ritual
of śivahasta can appear as outside the domain of nyāsa, but it is not. One is still
within the problematic of placings, mental or manual, and of the transmission
through thought or hand of a transforming power.
The hand, as we have seen, is generally the right hand but it can be the
left or both. A nyāsa can also be made in a different way, with a flower for
example. Being a beneficial ritual, the nyāsa should normally be made with the
right hand. The texts sometimes make it clear. They especially mention the cases,
rather rare, in which the left hand is to be used. Thus the fourteenth chapter of the
TĀ, in a passage previously summarised, indicates the practice expounded being
‘of the left’ (vāmācāra),56 all the rites prescribed there are to be performed with
the left hand (except where both hands are prescribed) using the thumb and ring
finger.57 The left hand, hand of impure activities, is also to serve in harmful magical
practices (abhicāra). Finally, there are several cases of using both hands together
or separately. Such is the case (cf. p. 76) of the vyāpakanyāsa. There are others
cases, those notably of nyāsa on symmetrical parts of the body, or when (in the
karanyāsa) the placings are made separately on each hand.
Whatever hand is used, the nyāsa is done with a particular gesture of the fingers,
a mudrā: usually (as said above) the thumb and ring finger joined, which is often
called mahāmudrā. The TBhS says:

The sage must perform the placing with flowers, with the ring finger, or
mentally, says another text on mantra. [‘ring finger’ being understood as:]
with the ring finger united to the thumb. The Padyavāhinī says in effect:
it is everywhere understood that nyāsa is made with the thumb and ring
finger joined. It explains further: with flowers on the image,58 with the ring
finger and thumb on his own body and mentally on the mūlādhāra and
other cakras. This is excepting the nyāsa of the .rs.i etc., of the
karān.ganyāsa and of the external nyāsa of the mātr.kā for which other
mudrās are prescribed.

These mudrās are enumerated in the same work (p. 163): they vary not only with
regard to the phonemes imposed but according to the place where the placings are
 Nyāsa: the ritual placing of mantras 67
made: ring finger and middle finger on the forehead and the mouth, index, mid-
dle finger and ring finger on the eyes, etc. Other texts give different prescriptions
– and there are evidently other prescriptions in Buddhist texts (or in Jain ones).59
Mudrās are continually used in Tantric ritual; they have coded meanings and
accompany or support speech – above all mantras – uttered in pūjā as well as
during other ritual or magical actions. There are lists or descriptions in a number of
works.60 As noted above, mudrās like mantras surely go back a long way: they are
archaic quasi-universal forms of behaviour that India has particularly developed
and codified. As previously mentioned, the relationship existing between mantra
and mudrā does in several respects pose problems that go beyond the scope of
this brief study. Remaining in its frame, we have already seen that the Indians did
ask themselves the question of the relationship between the effect of nyāsa and
its material performance. On the subject of mantras and mudrās in general, the
following formulas, usually ascribed to the Mahānayaprakāśa,61 are sometimes
quoted: svarūpajñānapanarūpam . mudrā-samsthānam and svavimarśātmanā
trānam īs.yate mantralaks.anam; the mantra would thus bring liberation by reflex-
ive awareness of the self, whereas the role of the mudrā would make known the
nature of reality. We may consider these formulas as largely arbitrary since they
are based in part on a play of words; we must, however, retain the distinction
they set between, on the one hand, the awareness and, on the other, the visible
expression or sign of the essential nature, for this is certainly the manner in which
mantras and mudrās are understood and experienced by those who use them. We
may note too that among other meanings the term mudrā means ‘seal’ (or ‘imprint’
made with a seal) and that, in the case of nyāsa, the hand gesture appears precisely
to seal in the prescribed place the entity or mantra being imposed (mudrā is also
translated into Chinese as yin which designates an official seal).62
As a gesture, mudrā is an essential element in the bodily participation of the
adept in the transformation of his body that nyāsa is to bring about: the move-
ment reinforcing and concretising in a visible manner the action accomplished by
mantra and mental concentration.63
But although corporeal by nature, the mudrā, too, can be done internally. Thus
the LT 35.74, in a chapter describing the rites to be performed for the internal
worship (antaryāga) of the deity, says: ‘The sādhaka should meditate on the
mudrās in spirit’ (manasā bhāvayen mudrāh.). The mudrās, in this case, are men-
tally performed or, more exactly, intensely meditated upon and ‘realised’.64 Cases
of this sort, of mental or interiorised ritual worship, where no gesture is made, are
not exceptional and they are always considered as of a higher nature than materi-
ally performed rites. The mudrā, too, in such cases, acquires a higher value.
Mudrās, in fact, necessarily have a mental aspect since they are not merely
performed, but also thought, held in mind, mentally imagined by the adept at
the same time as he performs them bodily.65 But there may also be something
more: they can be ‘lived’, experienced, by him on a metaphysical plane. When,
for example, the YH 1.57–71 describes the divine energy as taking the form of
nine mudrās, the sādhaka must at that time make the prescribed, symbolic ges-
tures, the mudrās, with his hands, and realise them as being forms of the Energy
68 Nyāsa: the ritual placing of mantras
(kriyāśakti) – deities also – thus closely associating the bodily, theological and
metaphysical planes.66
We can cite on the same subject another work of the Krama tradition of
Kashmir, but by an author of South India, Maheśvarānanda (tenth century), who
put forward an interesting metaphysical interpretation of nyāsa in stanza 45 of his
Mahārthamañjarī. Here is an approximate rendering of this text: ‘The network of
gestures pertaining to the domain of dualistic thought brings about a contact with
non-duality. The water of the offering is the play of the knowable; the flowers are
the nourishing states of our own essence.’67 Maheśvarānanda, in his own com-
mentary, the Parimala, explains that this ‘network of gestures’ is the an.ganyāsa
effected by the adept who, when making these placings on his body, becomes
aware of the Self, which destroys all duality. In effect he unites the awareness of
the absolute with the parts of his body touched by the nyāsa which are thus under-
stood in spirit through a synthetic, undifferentiated intuition whose nature is the
realisation of the total plenitude of the absolute ‘I’. Maheśvarānanda cites then a
verse from his paramaguru Śivānanda:

I accomplish the purification – the total purification of the two hands – which
is pure consciousness expanding to the fingers and where the action being
done coincides exactly with its Cause. I realize and make present by the prac-
tice of the s.ad.an.ga [nyāsa] the surging of the noble energies, the Omniscient
and the others, which abide in the Self of the Lord.

Neither these stanzas, nor their gloss, are easy to grasp; the terminology, the meta-
physics, are those of the Krama of Kashmirian Śaivism. But the interpretation of
nyāsa as an operation of a metaphysical order where gesture is subordinated to
consciousness, itself dominated by the non-dual awareness of divinity, is clear:
the scope and field of action of nyāsa is transcendent, its nature is spiritual. The
formula of the Pārameśvara-sam . hitā – vyāparo mānaso hyes.a nyāsah. – appears
in comparison to be modest; but all the formulations encountered – and I think
I have made an impartial choice – put the placings on the same side, that of
consciousness.
It remains, to complete this study, to review briefly some examples of nyāsa
prescribed in the texts and the manuals (examples, of course, that we have not
already seen).

R.s.yādinyāsa – the placing of .rs.i etc.

When a text prescribes the use of a mantra for certain ends, the announcement
is almost always followed by an indication of the sage (r.s.i) to whom it was
revealed, by its meter (chandas), the devatā it ‘expresses’, the syllable or word
which is its seed (bīja), that is, its abbreviated form or essence. Sometimes, too, its
energy (śakti) is indicated along with its pointed end (kīlaka) and, finally, its use
(viniyoga).68 Now, the first thing to do, if one wants to use the mantra, is to
impose on oneself those of these elements that are quoted: an obligatory rule to be
 Nyāsa: the ritual placing of mantras 69
followed even when it is not expressly prescribed. This may take place at the
beginning of the ritual, if only one mantra is concerned (as in most cases cited in
the Mantramahārn.ava); but if a longer rite requires the use of several mantras, the
.rs.yādinyāsa can be prescribed each time – thus the placing of the mātr.kā, which
very often must follow the .rs.yādinyāsa and the s.ad.an.ganyāsa of the principal
mantra, begins with the placing of the .rs.i etc. (see below p. 79). (This is so much
so that even the hams.a mantra of the ajapājapa, which in many regards is only
the natural movement of the breath accompanied by the two syllables ham and
sa, is supposed to have its .rs.i etc. that it is necessary to place initially: at most it
is sometimes admitted that there is a spontaneous placing–sahajam nyāsam, cf.
Daks.inamūrtisam . hitā 7.17–20). The raison d’être of this rite is that the practitioner
thus begins his assimilation of the mantra by effecting a first general identification
with it through taking into himself its principal elements, after which comes the rest
of the ritual (of which we have seen examples), at the end of which the adept will
attain a perfect identification with the supernatural entity of which the mantra is the
phonic form and through which the sought after goal is reached. But to be able to
effect a complete transformation, a veritable ‘change of ontological status’ (to speak
like Mircea Eliade), since it is a question of passing from ordinary life to a state of
liberation in life, of identification with the absolute69 – or even, more modestly and
more usually, when the aim of the rite is merely to become able to use the mantra
for utilitarian ends – it is necessary for the adept to have from the outset a body and
mind prepared and already somewhat modified and purified by assimilating some-
thing of the mantra. This is why this preliminary nyāsa is useful.70 ‘Without the
placing of the .rs.i, the metre and the devatā, the recitation of a mantra, even indefi-
nitely multiplied, would remain fruitless’ (a tantra cited in the TBhS, p. 189).
This nyāsa is always done in the order given above and on the same points of the
body, successively: the head, the mouth, the heart (which are the three obligatory
points corresponding to .rs.i, chandas and devatā), then genitals (or mūlādhāra),
feet and/or navel and on the entire body (sarvān.ga). The formula used is of the
type: (Om) + the utterance of the .rs.i etc. in the dative + namah. (the part of the
body touched being sometimes also stated in the locative). Thus, for the mantra
Om namah. śivāya, the .rs.i etc. are presented as follows:
.
.rs.ī Vāmadeva, panktichandas, Īśānadevatā, om. bījam, namah. śakti, śivāya
kīlakam, caturvidhapurus.ārthasiddhyarthe nyāse viniyogah..

We have thus:

om vāmadevārs.aye namah., on the head

pan.ktichandase namah., on the mouth
Īśānadevatayai namah., on the heart
om bījāya namah. on the genitals
namah. śaktaye namah. on the feet
śivāyeti kīlakāya namah. on the navel
caturvidha … viniyogāya namah. on the body.71
70 Nyāsa: the ritual placing of mantras
One might ask what motivates the choice of one or another part of the body for
the nyāsa. Perhaps comparing nyāsas in several different traditions would make
it possible to discover permanent elements and to reach conclusions: such an
enquiry (if it is worth the trouble) remains to be carried out. For the present case,
the selection of the places where to impose mantras seems to be that of important
points of the image of the body as it is mentally experienced in ritual worship, in
yoga, and in mantric practice.72
Sometimes this series of placings is to take place after the karanyāsa, but it
normally precedes another series, also a general occurrence: the an.ganyāsa or
karān.ganyāsa.

. .
Anganyāsa or karānganyāsa
The placing on the body – or on the hands, then on the body – of the an.gamantras
is also one of those somehow preliminary operations which, taking place at the
beginning of a rite, serve to impregnate the body and the mind of the officiant with
the energy and powers of the mantra or of the deity of whom the an.gas represent
qualities or fundamental aspects. Necessary at the beginning of all mantric prac-
tice, included therefore in the obligatory rites, it is part – as the nyāsa of the ris.yādi
and that of the mātr.kā – of the everyday practices of Hinduism.
This rite is also called sakalīkaran.a. The SP defines it thus: ‘One must under-
stand by sakalīkaran.a the placing on the body, beginning with the heart and
ending with the hands, and on the fingers beginning with the little finger, of the
mantras of which the first is HR.D’.73
The six an.gamantras (with which are associated, in Śaivism, the five vaktras,
faces or mouths of Śiva, to form the eleven samhitāmantra) are perhaps originally
a Śaiva conception, but one finds it in all of purānic or tantric Hinduism and also
with the Jains. In spite of their name, the an.gas are not limbs or parts of the body,
but elements with symbolic value, corresponding to parts of the mental (visual-
ised) image of a divinity or a mantra – or else to the powers which emanate from
deity or mantra, such as the rays from the sun74 – and which are placed by nyāsa on
different points of the human body which is thus put in correspondence with them.
These elements are ‘lived’ by the adept and experienced as such and as pervaded
by divine powers.75 ‘One must perform the s.ad.an.ganyāsa to acquire perfection of
body (dehasya siddhaye),’ says the Pārameśvara-sam . hitā (7.21). (This is apropos
the ajapājapa of the hamsamantra which has no limbs nor body, since it is nothing
else than the breath assimilated to the divine energy: besides, the placing of the
an.gas of this mantra is in fact done with luminous entities: the sun, moon etc. – the
Hamsopanis.ad, 12, gives a similar prescription). The an.gas are the heart (hr.daya),
the head (śiras), the tuft of hair (śikhā), the ‘cuirass’ (kavaca), the eye (netra)76
and the weapon (astra). They are placed respectively on the heart, the head, the
place of the tuft, the upper part of the arms close to the shoulders, the eyes or the
forehead, and the hands; or, in the case of placings upon the hands, on the fingers
and the palm. They are always enumerated and placed in the same order and the
way they are formed obeys precise rules, the same, apparently, in all traditions.
 Nyāsa: the ritual placing of mantras 71
These mantras are uttered in the following manner: one first utters (although not
always) Om which is followed by one or several monosyllabic bījas77 and some-
times by another element78 – these varying according to the principal mantra. After
which comes the enunciation of the an.ga in the dative, then the final exclamation,
which is, according to the an.ga and in order: namah., svāhā, vas.at., hum
. , vaus.at. and
phat..79 This order being fixed, the placings are made sometimes with only this part
of the mantra – called jāti, that is, species or category – which suffices to represent
and identify the an.ga.80
Here is, by way of example, the an.gamantra of the triple vidyā Parā, Parāparā
and Aparā according to the MNT, a modern fabrication, to be sure, but propound-
ing many traditional notions:

Om
. sām
. hr.dayāya namah., om . sīm. śirase svāhā,
Om
. sūm
. śikayai vaus
. at, om. saim. kavacāya hum,
Om
. saum. netrebhyah. . . . sah. astrāya phat.
vas at , om
(ibid., III, 60–61)

According to certain texts (as in a tantra cited in the TBhS p. 170), the hand forms
different mudrā according to the an.gamantra being placed.
This sixfold placing is sometimes made on the hands before being made on the
body: the hand, naturally, receives the an.gas before it places them elsewhere. One
has then a variant of the karanyāsa mentioned previously – where sometimes (in
the sakalīkaran.a for example) the an.gas are also placed.
The formula used is a little different from that of the an.ganyāsa. It is either
a short formula, of the type angus..thabhyām . namah., tarjanibhyām svāhā, etc.,
81

or a more complex one, of the pattern: bīja + divinity and an.ga with the dative,
+ finger or palm with the locative, + final exclamation, as, for instance, hrīm .
śrīmadekajat.ayai hr.dayāya an.gus..thabhyām . namah.. The placing is normally made
from the thumb to the little finger, then on the two palms. As with other nyāsas,
it must sometimes be done three times: according to emanation, from the thumb
to little finger; according to resorption in reverse; or according to maintenance,
going from the base of the middle finger to the index finger.82 Certain texts pre-
scribe different mudrās for different placings (TBhS, p. 170). The formulas use the
locative dual for the fingers and palms because the nyāsa must be made simultane-
ously and separately on the two hands: with the thumb on the fingers of the same
hand, with the index finger on the thumb, then with the four fingers of the right
hand on the left palm and vice versa.83

Mātr.kānyāsa
We remain, with the mātr.kānyāsa, in the placings which ensure the preliminary
divinisation of the officiant, essential for effectively worshipping a deity or a
mantra. One thus finds it almost always in the first part of the pūjā, although
in principle not in the worship of a mantra where, on the contrary, the nyāsa of
the .rs.i etc. and of the an.gas is always present. This last, however, is usually
72 Nyāsa: the ritual placing of mantras
associated with the placing of all letters of the mantra (mantramātr.kānyāsa), an
operation which one can be all the more tempted to parallel with the nyāsa of the
mātr.kā since the latter, when imposed, is treated as a mantra (one sometimes says
mātr.kāmantranyāsa) with its .rs.i, an.ga etc: all these procedures are repetitive and
tend to interpenetrate. The TBhS underlines this when it begins the section of the
mātr.ikānyāsa by citing the Phet. kārin.ītantra: ‘without the placing of their letters,
the mantras are dumb. To ensure the success of all the mantras, one must therefore
first place the letters’.84 In many cults, we find series of mātr.kānyāsas added to the
antar- and bahir-nyāsa which form the mātr.kānyāsa in the strict sense, which we
describe here, the imposition of the same fifty or fifty-one phonemes associated
to cosmic elements (for example, the kalās – see the kalāmātr.kānyāsa cited below
p. 77) or the nyāsa of the letters of a mantra itself sometimes associated with that
of the mātr.kā. One obtains in this way series of mātr.kānyāsas: nine are enumer-
ated later (p. 78) and many other examples could be given.
The mātr.kā – or the mātr.kās: the ‘little mothers’ – are the phonemes of the
Sanskrit alphabet regarded as discrete aspects of divine energy or, taken all
together, as the totality of this energy under the form of speech (the ŚT 6.2, calls
this the phonetic body, varn.atanu, of the goddess) which ‘expresses’ the cosmic
manifestation, bringing it into existence by enunciating it. They are the basic
phonic elements being used to form the mantras as well as to create, support,
animate (and to reabsorb) the universe.85 Their placing can therefore appear as an
essential factor in the divinisation of the officiant, this they may be in theory if
not in practice. The mātr.kānyāsa may seem to have a more important role in the
traditions where – as in the Śaiva non-dualistic systems of Kashmir – speculations
on the cosmic role of the phonemes were particularly developed. The importance
and role of the mātr.kā is, however, part of a shared Tantric fund, it is therefore not
surprising that this nyāsa is also found in one form or another in all traditions. It
is, however, especially in the Śaiva and Śākta texts that one finds the mātr.kānyāsa
in the complete form that we will describe now.
The placing of the mātr.kā usually follows the .rs.yādi- and karān.ga-nyāsa. It is
normally divided into three parts. First, since the mātr.kā is considered as a mantra,
the .rs.i etc., then the an.ga, will have first to be imposed. This is arbitrary and is due
only to the idea that the mātr.kā is, in this context (insofar as one imposes it?), to be
treated exactly as a mantra. Of course, .rs.i, metre, etc. vary according to the texts
and the schools. The nyāsa is done in the usual way. It is followed by karanyāsa
and an.ga- or karān.ganyāsa on the hands and/or on the body.86
Then the internal placing, antarmātr.kānyāsa, takes place in general a process
entirely mental, consisting of concentrating mentally on the six cakras (or ādhāras,
of the yogic body) which are tiered from the base of the spinal column to the sum-
mit of the head and which must be visualized as lotuses, each with a different
colour and a particular number of petals from two to sixteen, on each of which one
places mentally one of the fifty phonemes: ‘That which is called the internal nyāsa
consists of placing and [mentally] uttering (uccārya) each phoneme followed by
the word namah., going from the mūlādhāra to the brahmarandhra’.87 The texts in
general underline the importance of mental concentration, of the image mentally
 Nyāsa: the ritual placing of mantras 73
evoked, of the centres of the yogic body and of the phonemes. The adept must
– at least in theory – clearly visualise the cakras of his body and ‘see’ there the
phonemes he has placed as brilliant letters forming drops of energy whose pres-
ence transforms him by awakening these centres of vital and cosmic energy. This
mental representation can be accompanied by the recitation, audible or murmured,
of the phonemes placed. To this is sometimes added a visualisation of a divinity
also mentally perceived in the body.88 ‘This nyāsa must be made in meditation, the
mind well concentrated. One must visualize Vis.n.u, pure Consciousness, extending
from the mūlādhāra to the brahmarandhra and place the mātr.kā with the essence
of nectar flowing from the supreme bindu, while separately uttering each of the
letters’.89 Some texts prescribe, to accompany the placing (or to precede it), a
prān.ayāma where the breath is imagined as associated with different categories of
phonemes, then as circulating with them in the entire body which they impregnate
with their energy.90 The antarmātr.kānyāsa reflects, in this regard, the same spirit
as the complex nyāsas that we have seen above (p. 59f.) where the image of the
body mentally evoked plays an essential part.
Finally comes the external placing (bahirmātr.kānyāsa) of the fifty phonemes in
grammatical order from A to KS.A, on fifty points of the ‘gross’ body, made with
the hand – generally with a different mudrā according to the part of the body and
by enunciating either only the phoneme ‘decorated’ with the bindu and followed
by namah., or these elements preceded by a bīja variable according to different
schools.91 The external nyāsa is often made three times in a different order, accord-
ing to whether the phonemes are placed emanationwise, from A to KS.A (since that
is the order which manifests the universe), or according to conservation, or from
KS.A to A, and in this case it is the process of phonetic resorption which is imposed.
We have already seen the triple sr.s..ti-stithi and sam. hāra-nyāsa and explained its
meaning (see above p. 65). The rules of external nyāsa vary also according to the
texts: the Tārābhaktisuddhārnava quotes several.92
There is another placing of the fifty phonemes which is less frequent: the
mālinīnyāsa, that we have already seen (above p. 60) as part of a complex nyāsa
described in the TĀ. The mālinī is a particular order of phonemes, the first being NA
and the last PHA (it is therefore called nādiphāntakrama): it is the alphabetical form
of the goddess Mâlinî, an ‘alphabet deity’ (see above p. 60 and note 30). It is found
especially in the non-dualist Kashmirian Śaiva system of the Trika (being described
in particular in the third chapter of the MVT, quoted and paraphrased by Abhinav-
agupta in his PTV), and also in the Kubjikā tradition. The mālinī is considered as an
especially effective and energetic form of the phonemes in, the passage that I have
just recalled (TĀ 15.121–142), its nyāsa is prescribed to charge the body with energy
(nyasecchāktaśarīrārtham . bhinnayonim . tu mālinīm), which is a citation from MVT
3.36. We also meet it in chapter 145 of the Agni-Purān.a (which is a sign, among
others, of the influence of the kulāmnaya on this text). In these cases, the nyāsa
is not described as part of a complete rite of mātr.kānyāsa that we have just seen,
and does not take place necessarily at the beginning of the rite. A particular aspect
of those nyāsas is that the letters of this alphabet are conceived as being disposed
vertically from head to feet in the shape of a human body – that of the goddess
74 Nyāsa: the ritual placing of mantras
Mâlinî, and thus, if imposed on the body of an adept, the phonemes are placed in the
same way on his body. The operation is therefore somewhat different, but its mean-
ing and effect – a transfer of power embodied in an alphabet – remains the same.93
The three nyāsas that we have seen are probably the most frequently and most
generally used in Hinduism (they even appear to some measure in Jainism). They
help us to underline some traits of the theory or practice of the placings. There
are, however, other nyāsas, probably less common but not uninteresting. We will
therefore mention some of them chosen among the most characteristic.
We have already encountered the pīt.hanyāsa when dealing with the s.od.hānyāsa,
of which it usually forms a sixth part, having the effect of placing in the body of
the adept the fifty-one sanctuaries of the goddess, thereby identifying the body
with a sacred world where, on pilgrimage, one goes from one sanctuary to another.
One could say in this regard that the pilgrimage is internalised together with the
fifty or fifty-one divine places, or centres of divine energy, distributed at different
points of Indian territory. This nyāsa is both Śaiva and Śākta since it refers to the
myth of the dismemberment of the body of Satī.94 There are variants correspond-
ing with variants in the numbers of pīt.ha being invoked which goes from four to
fifty-one (and according to a later tradition up to 108). It may be found also in
other traditions.95 In fact, the expression pīt.hanyāsa, in its most current sense,
designates the placing of entities serving as seat or throne (pīt.ha) of the divinity
on the lower part of the icon being used for worship. This is well attested in ritual
texts (for example, for the Pāñcarātra in the JayS, chapter 20, pp. 223–227).96
We have previously seen other placings which divinise or ‘cosmicise’ the
adept, as with the placing of the tattvas or the adhvans. There is also the nyāsa
of the kalās, etc. We can also mention, among such notions, the mūrtinyāsa pre-
scribed in the SvT (1.59–60, vol. 1, p. 50), where the adept must place on his
entire body (sarvan.ges.u) the nine tattvas from kalā to śiva,97 then the tritattva
(ātma, vidyā and śiva-tattva) covering the thirty-six tattvas. He thus places in
himself the divinity as it manifests the universe: this is a process of deification.
The nyāsas examined so far are essentially those that the adept makes on his
own body to purify or transform it. But placings are very often made on objects to
which they transmit the power of a deity or mantra; a power that the adept must
in theory first assimilate by an initial and more or less complex nyāsa.98 This
is continually met in the ritual: the placing of a mantra on water used for wor-
ship (arghya) and which, when thus ‘sacralised’, becomes the nectar of Śiva, etc.
But there are several more curious ways of proceeding. One could thus mention
the ritual for the invocation of Śiva (śivāvāhana) of the SP1 (III, 61–64, vol. 1,
pp. 184–91), where the officiant, meditating on the supreme Śiva, utters inter-
nally the bīja Haum . , making it go up from the mūlādhara to the brahmarandhra
and, from his face where it shines like the moon, he visualises it descending in a
flower kept in the cup of his hands (pus.pāñjaligatam . dhyatvā); he then unites it
to the icon (mūrti) which is used as support for the worship, where the divinity
itself will consequently reside: a process which is made entirely mentally (dhyāna
or bhāvanā), then by mudrās, without there being, strictly speaking, nyāsa. We
can ask, here, again if we are still in the domain of the placings. In fact, the ritual
 Nyāsa: the ritual placing of mantras 75
seems to provide instances where mantras produce an effect without any nyāsa
being really performed: sometimes a mudrā or a gesture accompanies them, some-
times touching with a flower.99 There are also some cases where the mudrās seem
to be active by themselves.100 It would probably be too much to consider these
practices as being strictly speaking placings. They are, however, very near, be it
only because they fall within the same problematic: that of the transfer of energy
by mental concentration and/or gesture. Or, from a more general point of view,
partake of the problem of the efficacy of gesture which in fact lies at the centre (or
in the background) of the practice of nyāsa.
It should be noted also that, in a ritual close in spirit and in its effect, the jīvanyāsa
(a term that could be translated as ‘the placing of life’), such as described in the
AgPur (chapter 96, śl 90–93), the master enunciates the mantras, making them
go up from the navel cakra to the dvādaśānta where they dissolve in the efful-
gence of the supreme Śiva, who is carried by mental concentration and placed
on the lin.ga being used for the worship, where from now on the god will abide.
This purely mental ritual process (manasā … dhyatvā) is called nyāsa (jīvanyāso
bhaved evam). It is followed by other placings on the base of the lin.ga. Chapter
59 (this one Vais.n.ava) of the same purān.a, describing the rite of the adhivāsana of
Vis.n.u, says in a similar way that the deity is installed (sthapana) in the image
which is vitalised (sajīvakaran.a) by placing on it the mūrtimantra.
The jīvanyāsa is normally used for infusing the icon used in worship with life. It
can, however, also be used for the adept after the rite of bhūtaśuddhi or dehaśuddhi
(purification of the elements of the body), by which the adept reabsorbs the ele-
ments of which his body is composed one in the other (a resorption that makes
the gross body disappear). His vital principle (jīva) finds itself thus outside of him
because it resides, for a moment, in the dvādaśānta or in the bindu. To this ancient
body it is necessary to substitute a new body, pure, subtle, into which one will
bring back the principle of life previously displaced. This is done through the rite
of jīvana, or sakalīkaran.a, where some nyāsas are used.101
When a pūjā is made to a deity using a diagram, it is by nyāsas – normally done
using blades of darbha grass – that the deity, then his/her an.gas, attributes, and
the secondary deities who surround him/her, and so on, are placed in the cakra or
man.d.ala (see for example, LT chapter 38 or SP3). It is the same for any rite where
a man.d.ala is used in which deities are placed (cf. SP2, vol. 2 p. 320 with regard to
prayaścitta; or various passages of SP3, etc.).
In more current use, mantras are put on objects to make them sacred, to purify
and protect them etc. Thus the astramantra is placed on the doorframe of the door
to the sanctuary or on the threshold to protect the place of worship (SP2, vol. 2,
pp. 48 and 356); or nine deities are placed on the blades of grass used to make a
pavitra (ibid. p. 102); or the mantras astra, kavaca and hr.daya are deposited on a
receptacle where they constitute a phonic and spiritual ‘casket’ so as to protect these
same pavitras (ibid., pp. 130–131). Examples could be multiplied indefinitely.

But rather than pursue this enumeration, it would now be better, after hav-
ing seen separately different nyāsas, to try to follow – briefly – a whole ritual
76 Nyāsa: the ritual placing of mantras
where placings play a role, whether the ritual is centred on a mantra or is the
pūjā of a deity.
In the cult of a mantra, the mantrasādhana or puraścaran.a, the adept, who has
received a mantra from his master through initiation, is to perform rites and follow
observances which will give him, in the end, control of the mantra from which he
will be able to draw all possible spiritual (or material) benefit. Hélène Brunner,
in her study already cited (above, note 29) on the sādhaka, has summarised this
ascesis according to the Mr.gendrāgama and several other āgamas. There are very
similar prescriptions in other Śaiva, Śākta and Vais.n.ava texts. The schema is gen-
erally the same, comprising a certain number of nyāsas placed especially at the
beginning of the practice in order to communicate to the adept, by several series of
placings, the energy of the mantra which will serve him as instrument of action and/
or of salvation. As it is usually necessary, during the sādhana, to perform a worship
to the mantra, this is done using a diagram, on which nyāsas will also be made.
The placings are in general in the following order. The .rs.yādinyāsa, a
vyāpaka-nyāsa102 on the entire body, the kara- or karān.ganyāsa on the hands, the
hr.dayādis.ad.an.ganyāsa (sometimes also the placing of the five brahma- or
vaktra-mantra, sadyojāta etc.); then the placing on the body of the letters and/or
the words of the mūla-mantra (mantravarn.a and mantrapada-nyāsa) follow from
those of the an.ga, upān.ga and secondary elements.
After this comes the pūjā of the mantra, normally done with a man.d.ala
where the mūrtimantra must be placed in the centre by nyāsa, which installs
there the image of the mantradevatā, then, of course, around it, the an.ga, upān.ga,
āvaran.a-, or parivara-mantras etc.103 The pūjā itself and all that follows it then
remain to be performed, during which rites other nyāsas will take place and where,
in particular, it might happen that the sādhaka must place the cakra on his own
body (cf. YH 3 cited above p. 64): the succession of the nyāsas is considerable,
and even quasi-infinite if one thinks that the operations are repeated hundreds or
thousands of times.
It is, however, in the pūjā of the divinities that the series of nyāsas are longest
and where they take place most often. This is quite natural considering the number
of elements playing a role in the worship at the same time which it is necessary
to purify or consecrate, and all the more in the context of manipulation of divine
energy and of internalisation or ritual identification which is that of the Tantric
pūjā. Evidently, we do not meet here as complex nyāsas those which took place in
initiations: those which are supposed to involve a total spiritual and bodily trans-
formation of the disciple and which, quite naturally, require the most complex and
effective procedure for the transfer of energy. The case of the pūjā is different and,
one could say, simpler. But even here the number and the diversity of the placings
are considerable because the worship is often a long process, sometimes involving
a number of deities. Now, rites as well as divinities vary according to traditions
and sects and even according to worshippers; it would therefore be vain to try to
draw up a complete list. It will suffice to end this overview of nyāsa by giving a
few brief elements concerning the usual succession and function of some of the
placings made during the pūjā.
 Nyāsa: the ritual placing of mantras 77
There is a brief exposition of nyāsa in the cult of Kr. s.n.a, pp. 90–92 of the thesis
of R.V. Joshi, Le rituel de la dévotion kr.s.n.aïte.104 They are part of the worship of
the already consecrated image (without nyāsa apparently) and are done on the
image itself. They are limited to very few elements: antarmātr.kanyāsa on six
parts of the image which correspond to the six cakras of the body; keśavanyāsa:
the placing of fifty-one phonemes with fifty-one different names of Kr. s.n.a and his
Śakti (am Keśavāya Kīrtyai namah. etc.); and lastly a tattvanyāsa considered to be
particularly important. Mantras accompanied by mudrās are used during the rest
of the ritual but they are not strictly speaking nyāsas, which appear to be remark-
ably few in number.
Appreciably longer are the following lists which seem much more character-
istic. These are drawn from manuals of different epochs. The Pūjāvidhinirūpan.a
of Trimalla105 first enumerates only eleven placings, performed at the beginning
of the pūjā:

1. .rs.yādinyāsa;
2. s.ad.an.ganyāsa;
3. antarmātr.kānyāsa;
4. bahirmātr.kā- (sam. hāra-sr.s..ti and sthiti);
5. mūlavidyāsam . . itamātr.kānyāsa;
put
6. nyāsa of six goddesses;
7. nyāsa of five vaktras of Śiva, followed by:
8. s.ad.an.ganyāsa;
9. yoni-nyāsa: two placings of nine divinities;
10. nyāsa of the eight mothers Brahmī etc.
11. lastly, we have the vyāpakanyāsa of the mūlamantra of Bhuvaneśvarī;
followed by various offerings, the dvārapūjā, the pīt.hapūjā, then the cult of
the divinity.

If we look at the Tārin.īpārijāta (which quotes several Tantras), we find, in the

pūjā, after the preliminary libations, the construction and offering of the throne,
prān.āyāma and bhūtaśuddhi, a series of eleven nyāsas:

1. jīvanyāsa comprising several elements (see above note 101);

2. antarmātr.kānyāsa, comprising first several placings on the body, then
.
.rs.yādi- and anganyāsa, dhyāna of the goddess, ending with placings of
fifty phonemes in the cakra;
3. a triple bahirmātr.kānyāsa (sr.s..ti, sthiti and sam
. hāra), each comprising
r.s.yādinyāsa and dhyāna of a devatā with the placing of the mātr.kā;
4. kalāmātr.kānyāsa: the placing of the mātr.kā associated with fifty-one kalās,
preceded by the .rs.yādi and an.ganyāsa and a dhyāna;
5–10. s.od.ānyāsa (described according to the Tantracūd.aman.i) comprising the
placing of fifty Rudras, nine graha, eight lokapāla, Śiva and Śakti on the
six cakras, Tārā (nyāsa of fifty phonemes on the six cakra), and lastly ten
pīt.has with groups of phonemes on ten points of the body;
78 Nyāsa: the ritual placing of mantras
11. nyāsa on the body of the .rs.i etc., then of the an.ga (karān.ganyāsa) mantra
of Tārin.ī on the hands. After this the adept must trace the yantra used for
the worship, followed by nyāsa of fifty pīt.ha (seen previously, p. 74, lastly
the rest of the pūjā.

The two series do not differ much except by the presence in the second one of the
jīvanyāsa and of the sixfold s.od.aśanyāsa. These are simple models. An example
of a longer series can be found in the Balārcanapaddhati, a Śākta ritual manual
published in 1933 in Surat.106 This work is taken simply as an example among oth-
ers of contemporary prescriptions where the placings are multiplied one could say
for pleasure’s sake, without this abundance adding anything to the significance or
interest of the operation; it is typical of the Tantric tendency to ritual multiplication
and redundancy.
The part of the text concerning nyāsas begins with the bhūtaśuddhi which is
made by placing the fifty phonemes from A to KS.A on different parts of the body
(ending with the thousand petal lotus at the top of the cranium). This purification
is followed (as already seen, p. 75: jīvanyāsa) by the ‘infusion of breath’
prān.apratis..thā, realised by .rs.yādi-, karān.ga-, then vyāpakanyāsa, followed by a
dhyāna of the devatā, after which the placing of the vital breath itself is made on
the body of the adept with the prān.apratis..thāmantra. Then come the following
nyāsas:

1. antarmātr.kānyāsa: first a prān.āyāma – inhaling with the vowels, retain-

ing the breath with the pavarga (the occlusives from pa to ma) and expi-
ration with ya etc.; .rs.yādinyāsa, then karanyāsa with the vowels and the
vargas of the consonants followed by their hr.dayādis.ad.an.ganyāsa.
Dhyāna of the kun.d.alinī coiled up in the mūlādhāra with all the pho-
nemes, then mounting with them along the sus.umnā – one must visualise
the six cakras and mentally place there the vowels and the groups (varga)
of consonants.
2. bahirmātr.kānyāsa: first, as previously, the .rs.yādi-, kara, and s.ad.
an.ganyāsa, then placings of the sixteen vowels (om . am . namah. to om .
ah. namah.) and of the eight groups of consonants to which one adds two
supplementary ‘groups’: ham-h.am and ks.am placed on the two hands.
3. śuddhamātr.kānyāsa: .rs.yādi-, kara and an.ganyāsa. Dhyāna of the goddess
associated with fifty phonemes, then nyāsa of them on fifty points of the
body.
4–5–6. mātr.kānyāsa according to sam . hāra, sr.s..ti and sthiti.
First, the .rs.yādikaras.ad.an.ganyāsa. Then a dhyāna, different in each
of the three cases. Then nyāsa of the phonemes ‘decorated’ with bindu
(om . ks.am . namah. to om am . namah.) on different points of the body for the
sam . hāranyāsa; or the phonemes with the visarga (om . ah. namah. etc.) for
sr.s..ti, or, for sthitinyāsa, the phonemes with bindu and visarga, going from
d.a to .tha (the three placings are to be done in a different order according
to whether one is a brahmacārin, a householder or a renouncer).
 Nyāsa: the ritual placing of mantras 79
7. śaktimātr.kānyāsa: .rs.yādinyāsa, s.ad.an.ga- then karanyāsa. Dhyāna of the
Śakti, then mātr.kānyāsa (om . am . hrīm . namah. on the head, om . ām . hrīm .
namah. on the mouth etc.).
8. śrīpran.avakalāmātr.kānyāsa: .rs.yādi, then hr.dayādinyāsa. Dhyāna of
Pran.avakalāmātr. kāsarasvatī, to whom homage is to be paid; then invoca-
tion of the fifty kalās and their placing with fifty phonemes from A to KS.A.
9. Śrīkan..thādimātr.kānyāsa: .rs.ādi, then hr.dayādinyāsa of a rather long
formula: ‘om . hrīm . śrīm . namah. śivāya am . brahmān.i mām . raks.a raks.a
ham . savahane mām raks. a raks. a padmahaste mām . raks . araks. hām
a . hrīm. hrūm .
am. kam am am
. . . . . am am hrām . hrīm . hrūm . hsauh sam
. . jīvāni hr. dayāya namah .’
etc. Next dhyāna, then placing of the fifty phonemes according to the pat-
tern: om . hsauh. am . śrīkan..theśa pūrnendarībhyām . namah. (on the head) etc.
10. Keśavādinyāsa: .rs.yādinyāsa, dhyāna, then the placing, after having
made mentally the offerings, of the pūjā (mānasopacaraih. sam . pūjya
nyaset). Nyāsa of the fifty phonemes as: om . klīm. .am etc.; next a
prapañcayāgamātr.kānyāsa with .rs.yādi and karanyāsa, and then an
‘oblation’ (homa) of phonemes (varn.amayam homam) with the formula
om. hrīm . ks.am hamasah. so’ham . svāhā (for KS.A) up to om . hrīm . am . ham . sah.
so’ham . svāhā (for A) on different parts of the body.
11. mūlamantranyāsa, that is, the main, basic (mûla) mantra of
Bālātripurasundarī: .rs.yādinyāsa, then a triple placing of aim . , klīm. and
sauh. (which are the bīja, the śakti and the kīlaka of the mantra). Then a
triple nyāsa of fifty phonemes by encasing (sam . put.a) each one of them
successively between one of the three monosyllables (e.g. aim . am . aim .,
klīm. .am klīm
. etc.).
12. navayoninyāsa: nine placings of aim . , klīm. and sauh. on the body.
13. Rityādinyāsa: ‘aim . rityai namah . on the mūlādhāra, klīm
. rityai namah. on
the heart and sauh. mano bhavayai namah. between the eyebrows, such is
the nyāsa’.
14. amr.teśvaryādinyāsa: sauh. amr.teśvaryai namah. between the eyebrows,
klīm. yogeśvaryai namah. on the heart and aim . viśvayonyai namah. on the
mūlādhāra.
15. mūrtinyāsa: placing on the head, mouth, heart, genitals and feet of the
five brahma- or mūrti-mantras: shrauh. īśanāya manobhavāya namah., shrem .
tatpurus.āya makaradhvapāya namah., shrum aghorāya kandapāryāya
namah., shrīm . vāmadevāya manmathāya namah. and shrīm . sadyojātāya
kāmadevāya namah..
16. cakranyāsa: same placing of the five mantras īśāna to sadyojāta on the
five faces (pañcavaktāni) of, probably, a lin.ga.
17. [nyāsa given as uncertain: mūlamantras.ad.an.ganyāsa: am . klīm . sauh., two
times on the heart, tuft etc.].
18. bān.anyāsa: five placings on the head, neck, face, genitals and heart:
dram . dravan.abhānāya namah., drim . ks.obhanabān.āya namāh, klīm .
vaśīkaran.abān.āya namah., blūm . ākarśan . abān.āya namah. and sah.
sammohanabān.āya namah..
80 Nyāsa: the ritual placing of mantras
19. devatānyāsa: eightfold placing with the bījas aim . , klīm
. , blūm
. , strīm
. of
eight female divinities (Subhaga, Bhagā, Bhagamālinī, etc.) on different
points of the head, heart, navel and mūlādhāra.
20. pīt.hanyāsa: placings on different points of the body of entities forming
the seat or base (pīt.ha) of the divinity. Then dhyāna of Bālātripurasundarī,
which completes and ends the whole series of nyāsas.

It would have been useful to complete this exposition, which rests on Hindu
examples only, with an examination of nyāsas such as those practised in Jainism
and Buddhism. We have already quoted several examples of practices similar for
Hindus, Jains or Buddhists, but it would be interesting to do more and notably to
mark in what ways the two last traditions, especially the Buddhist, differ from the
first. I have, however, not been able to gather enough documentation to enable me,
as for Hinduism, to give a kind of general outline of usual uses in the Jain of Bud-
dhist milieus. What would be especially worth examining are the practices (along
with the discourses that accompany them), rather than the general problem of the
nyāsa which, it seems to me, is in many regards the same – mutatis mutandis – in
all traditions: it is always a matter of making the body or an object sacred by the
placing of entities which carry a divine or spiritual force. But this not a topic I can
enter into now.
Limited thus to Hinduism, this brief examination of nyāsa will, I hope, contrib-
ute to show the importance of this ritual practice in the Tantric or ‘tantricised’ forms
of the Hindu religion. In the developed and organised form we have described
(and which is still used in India), the practice is perhaps comparatively recent. But
it is very likely that its origin goes back to an ancient past, and, above all, that it
is founded on an ancient way of conceiving, apprehending and ‘living’ the body
in religion and asceticism. India, indeed, has always invested the religious and
the cosmic in the body. It is the place where the sacred geography, the levels and
structure of the cosmos, the gods (‘like cows in a cattle shed’) are reflected and
incarnated, where is played out again on a human scale – but in connection with
the divine – the cosmic process of creation, manifestation and resorption. The role
and raison d’être of the nyāsa is precisely to actualize the cosmic potentialities
of the body, to structure it in cosmic correspondences or equipollences, an actu-
alization that is the very goal aimed at by the rites and the religious or spiritual
practices. Hence the importance of nyāsa as one of the essential components of
the practice of the Indian religious man who seeks release not only in spirit but ‘in
this life and this body.’
5 A Hindu rosary ritual1
(Jayākhyasam. hitā,
chapter 14)

The use of a rosary or string of beads (aks.amālā, aks.asūtra) as a means for keep-
ing count of a repeated incantation or prayer is very usual in Hinduism as well as
in Buddhism. Most of the texts concerning the ritual recitation of a mantra, japa,
that is – be they purān.as, tantras or manuals of ritual – include passages on how
to use a rosary, and how to assemble or to dispose of it. As is to be expected, such
instructions are specially numerous and detailed in Tantric works.
One such text is the Jayākhyasam . hitā (JS) of the Vais.n.ava Pāñcarātra, which is
one of the ‘Three Jewels’, ratnatraya (with the Paus.kāra- and the Sātvata-sam . hitā),
of the Pāñcarātra: one of the most authoritative of such works. The JS is also one
of the earliest samhitās since some of its passages are quoted by Utpalavais.n.ava,
in Kashmir, in the tenth century. As far as one can know or imagine, it originates,
if not from Kashmir, at least most probably from the north-western part of India.
Like all Pāñcarātra sam . hitās, it deals with a great variety of topics, several of its
chapters concerning mantras: chapter 11 is on mantranyāsa, 14 on japa, 19 on the
signs showing the mastery over mantras, chapters 26 to 29 on the rites to be per-
formed after initiation by the initiate so as to master and make use of the mantra he
has received (mantrasādhana). The ninety-five ślokas of chapter 14 are on japa,
the ritual recitation of mantras (japavidhānam). Of these we shall leave aside those
concerning such common rules as, for instance, the three sorts of japa, or the times
and places when and where to perform it, and concentrate on the ślokas dealing with
the rosary itself, insisting especially on such prescriptions as appear to be proper to
this sam. hitā. As we shall see, these prescriptions bring together, in a very typically
Indian Tantric way, ritual rules, yoga practices and metaphysical notions.
The chapter begins (śl. 5b-27) by saying how to prepare and put together a
rosary: how to make the string, how to chose and assemble the beads. This is to
be done by the user of the rosary (who is in principle an initiate2). It is not a mere
technical operation, but a ritual process implicating personally the owner-user of
this object. It is the initial, inaugurating, portion of the japa. The point in this case
is not merely to acquire a ritual object for religious use, but to create and fashion
it with reference to the use it will be put to, both as an object made up of certain
materials and, more importantly, as an instrument charged with divine power, an
image as it were of the deity whose japa the user will accomplish with it and which
will thus help him to gain the fruits he is looking for.3
82 A Hindu rosary ritual
The beads of the rosary (which the JS calls man.i, jewels, even when they are
made of some ordinary material) are first, as is always the case in such texts, clas-
sified as being of three sorts: superior, middling and inferior. A classification, of a
general order it seems, distinguishes between three sorts of vegetal seeds (śl. 5–6),
the best being aks.a (rudrāks.a, that is), the middling ones are made of myrobalan
fruits, dhātrī, the less good are jujube stones (badara).4 Another distinction is
made by reference to the uses the rosary will be put to, in which case one distin-
guishes between different metals: gold to obtain riches, prosperity or beauty; cop-
per for power of thought or sovereignty; lead or brass if to master several sorts of
beings or demons, or to operate in hells. The lowest sort of rosary is made of iron.
It is to be used for so-called inferior or bad rites (ks.udrakarmān.i), that is, black
magic, or sorcery. We may note that liberation (moks.a) is not mentioned here
among the aims of japa. It is quoted further on (śl. 13–46) in connection with the
material of the beads: those in precious stones and of coral give long life, health,
riches, happiness (and can be used in all rites); those made of crystal give libera-
tion – which is therefore not associated with the most precious material. Crystal,
however, usually symbolises the clarity of a pure being or that of a peaceful mind.
Here as in many other cases in Tantra (see note 2) liberation appears as being not
quite on the same level as the other, interested, aims that an adept may pursue – a
fact that is not entirely surprising insofar as the normal addressee of a tantra is a
sādhaka, that is, a bhubuks.u, not a mumuks.u.5
The JS does not merely mention the different sorts of beads (a matter on which
other texts are more explicit). It describes how the rosary is to be ritually put
together, then the rites which make it ready for use and cause the user to become
apt and worthy to use it fruitfully.
The beads must all be of the same sort and the same size. They must first be
washed free of all impurity (ks.alayet) with scented water and the Astra mantra,
OM . HAH. Astrāya PHAT., the mantra having as much (or even more than) a puri-
fying effect as water.6 The string (sūtra) is made by carefully twisting together
three cotton or hemp threads, and it is then ‘washed’ in the same way as the beads
(śl. 19b-20). All the constitutive parts of the rosary are thus submitted to the puri-
fying action of water and of mantras.
The beads once threaded (not freely, but following strict rules), the two ends of
the string are knotted together an additional bead being included in the knot. This
is a very important point, for this bead, called Meru – thus evoking Mount Meru,
the axis mundi – marks the limit of all recitations of the rosary, being thus both its
two extreme and its central point. The recitation abuts (being then briefly inter-
rupted) against this bead which must never be overstepped (see infra. p. 86).
The rosary thus constituted must be anointed with sandal paste, placed in a
pure vessel and honoured with ‘flowers, incense, and so forth’, then a purifica-
tion (śuddhi) is to be performed: ‘One must conceive it as being burned by the
Astra[mantra], fanned by the Varma[mantra], then bathed by the mūlamantra
as made of supreme ambrosia.’7 Here too the ritual operations are deemed to be
effected by mantras: mantras are the main element of Tantric Hindu ritual, which
is a universe where one ‘does things with words’ (to use J.L. Austin’s phrase).
 A Hindu rosary ritual 83
‘The rosary once thus purified’, the JS adds, ‘it is to be conceived as a body’
(śam. śodhyaivam . purā sūtram . dehavac cintayet tatah. – śl. 30a). This is to say that
the adept must imagine (smaret) it as ‘having four arms, being free from impuri-
ties, beyond compare like Nārāyan.a, one hand making the gesture of granting
wishes, the other that of fearlessness, [the two other hands] being joined in saluta-
tion’; after which the adept is to imagine this rosary as being ‘like a flame on the
door of Brahmā (that is, on the top of the head) of this [divine body]’.8
The JS then discloses the mantra of the rosary (aks.asūtramantra), which is not
the mantra to be recited with the help of the rosary but the mantra of the rosary, the
formula which embodies its essence and is used to worship it. The process here
is indeed the same as that of the mental worship of a deity, whose aspect is to be
imagined by the worshipper, who has a mantra, and is invoked, made present in
the cult icon and then ritually worshipped: these successive parts of the Tantric
pūjā we shall see here being acted out for the rosary.9
Śl. 32b-34a give the uddhāra,10 the extraction, the word content, that is, of the
aks.asūtramantra, which runs: OM . kaustubha trailokeśvaryada gopanāks.asūtrāya-
namah., a formula in praise of the rosary, seen as a divine jewel (kaustubha), which
protects and gives sovereignty over the three worlds.11 The rosary being considered
as a deity has anthropomorphic features, hence the ritual process of sakalīkaran.a
now prescribed, which consists in placing on the different parts of the rosary’s
‘body’ and ‘hands’ ancillary mantras of the main mantra representing its different
.
‘parts’ (its angas: hrdaya, śiras, śikha, varman, netra and astra12), its powers, that
is. Then a pūjā is to˚ be performed: a brief worship, apparently, immediately fol-
lowed by another, mental, worship, which is a complex yogico-spiritual practice
where the mentally produced image of the divine power present in the rosary is to
be seen and felt by the officiating adept as present in his body, as illuminating the
place where he stays, and is finally imagined as being transferred from the adept
to the rosary, all this being produced by that particular form of intense identifying
mental concentration called bhāvanā.13 The passage is as follows:

This supreme Power of Vis.n.u, inseparable from the supreme Self,

shining like the full moon moving in the sky, surrounded like the sun by
rays, increased in power by the energy of the fire that destroys the cos-
mic cycles, must be invoked as before together with the inspired breath
(pūrakena) in the order of emanation14 as situated in the lotus of the heart.
Then, the breath being held (kumbhakena), O Best of Brahmins, it is to
be kept [there]. One must then invoke her, staying in the lotus of the heart
as before, as [going] upwards up to the brahmarandhra, whence, progres-
sively, by the central channel, she will reach the dvādaśānta, this being
done whilst breathing out (recakena yogena). The rosary is to be meditated
with this [Power].15

The adept is now to imagine ‘the place of the sacrifice [that is, the place where he
practises the japa] as illuminated by the rosary as the sky by the full moon’. This
imaginary vision is to be pursued as follows:
84 A Hindu rosary ritual
He must then think that the King of mantras [which is probably Visn.u’s
mūlamantra OM . KS.ĪM. KS.ĪH. namah. nārāyan.āya vis.vātmane HRĪM . svāhā],
which is partless (nis.kalā) though inseparable from all the [other] mantras, as
being immersed in this [rosary], and meditate the Power of the Lord whose
nature is the same as that of the mantra, as well as the phonemes of the man-
tra of the rosary as identical with it, for they are [inseparably mixed up] like
water and milk. Then, taking hold of the rosary by its ‘back’16 with his two
hands joined and holding flowers (pus.pāñjalau), the twice born17 must imag-
ine that the [mūla]mantra – which is both one and made up of its constituent
parts (sakalanis.kala) – falls, following the same process as the creation of the
cosmos (srs..tikramen.a), on the ‘back’ of the rosary made perfect (susam . skrta)
˚
[by the preceding rites], and that, reaching it, [the mantra] on the rosary ˚will
loose its strength like a shower of sparks [flying from a fire].18

This mental visualisation where Vis.n.u’s energy, which is mantra (mantrātmā) in

the sense both of the basic mantra (mūlamantra) of the deity and of the original
Word (source and essence of all mantras, therefore both one and infinitely diverse)
is imagined falling like a shower of sparks on the rosary so as, as it were, to per-
vade it with its power and reinforce the ‘mantrisation’ of the rosary, identifying it
with the Word, the Energy (śakti) of Vis.n.u. The adept is now to address the God,
saying: ‘O Acyuta! [You who are here] in the shape of the rosary, give [it] to me.’
He imagines thus that the rosary is given to him by Acyuta so as to perform the
japa of the mantra. He takes it again by its ‘head’, saying ‘Be me favourable’.19
The adept is thus deemed to have received from the deity the rosary he has in fact
made up himself, but which, being henceforth pervaded by mantras and by the
energy of the deity, is an image, an aspect or form of that deity. For a bhākta, a true
devotee, everything he receives comes from his god.
The deity, however, has until now only been invoked, which is necessary but
not sufficient. It is also necessary that, being present, Vis.n.u/Acyuta should also be
favourably inclined towards his devotee and therefore ready to listen to him and
answer his prayer. The preliminary rites of the Tantric pūjā (āvahana, sthāpana,
etc.) are now to be performed. That such rites should now to be performed is quite
normal since the rosary is treated as an icon.
The JS says:

The rosary having been received, it must be installed. Then, having wor-
shipped it with the water from the arghyapātra, with flowers and with incense,
one must place on it [the mantra] with all it contains and all that accompanies
it, [considering it as] surrounded by all the powers, [and saying] ‘O Acyuta!
Master of all gods! Be thou present and remain in this ensemble of beads
named rosary until moon and sun pass away.20

The adept is now to show the so-called gesture of the rosary (aks.asūtramudrā)
which consists in touching with the thumb, in a particular order, the phalanxes
of the right hand. This mudrā, which is in fact the gesture by which one usually
 A Hindu rosary ritual 85
keeps count of the recited japa formula – the so-called karamālā, ‘hand-rosary’
– can be seen as a kind of outward, visible, confirmation of the ritual process just
performed. The JS underlines the significance of this gesture by saying that this
mudrā is nityasannidhikārin.ī: making the deity constantly present. Sannidhi or
sannidhāna is, as we have seen, one of the four rites always associated with the
creation of a cult icon – which is precisely the case of the rosary.
‘Then, one must begin the japa, which must not be seen by other people’
(japam. samārabhet paścād adrs.ya itarair janaih.), says JS śl. 52a. The JS, how-
ever, does not immediately start ˚ describing the japa. Two other mental acts are
to take place first. One consists of concentrating one’s attention (anusamdhāna)
on the five states of consciousness (avasthā). The other is an intense identifying
meditation (bhāvanā) of the mantra, felt as present in the adept as well as in the
rosary. The aim of these two actions is to give to the japa a deeper significance
and intent: not merely that of a devotional concentration on the deity, but, more
intensely, that of an intuitive intellectual apprehension of the meaning and import
of the speech act of the japa in an universe born from the divine Consciousness
whose highest plane is vāc, the Word, and which is entirely pervaded by vāc.
There is a correspondence between the five planes of individual consciousness
(avasthā) (from waking (jāgrat) to the fifth plane (turyātīta), the plane of fusion
in the absolute), the five levels of cosmic consciousness and the levels or planes
of the Word. Grasping the essence of the mantra to be recited in japa is thus grasp-
ing the transcendent and immanent deity of which the mantra is the vācaka: this
occurs on the fifth and supreme level which pervades and sustains the four others.
Below is this passage of the JS:

One must first realise by meditation (bhāvayet) that the senses, the word and
the mind are caused to be by Vis.n.u, the supreme Self, in the form of energy
(śaktirūpen.a), through mantra (mantren.a). What is in thought, where the
object is present, goes to speech, consciousness being thus associated with
the object, speech coming third, O Narada! This evolutionary process reaches
progressively the plane of action. This is how the mantra evolves (vivartate)
progressively from the fourth plane of consciousness (turya) to that of wak-
ing (jāgrat21), after which, O Brahmin! it goes through the same stages in the
reverse order. The nature (svarūpa) of the senses is that of vāc whose nature
is that of consciousness (cidrūpin.ī). What is thought (citta) becomes man-
tra, which becomes the Unborn, Hari. This supreme Brahman is omnipres-
ent (vyāpaka) and the śakti is that of Nārāyan.a. She indeed, by transforming
herself (parin.amena), reaches the plane called fourth (turya), which becomes
deep sleep (sus.upti), which becomes dream (svapna), which becomes waking
(jāgrat), which in turn will be the first of the four planes. [All these planes]
must be mentally evoked (smaret)22 [from jāgrat] up to the [plane of] the
energy of the Lord as being all together and forming one whole. O Brah-
min! all that one sees in this world, ephemeral [variegated but unsubstantial],
like a painting, this external phenomenal field, is what is called waking. The
[equally] ephemeral but always superior to the preceding one, transitory, like a
86 A Hindu rosary ritual
dream, is the state of dream which is the cause of [the state of] waking. Know
that the perfectly peaceful state of deep sleep is superior to that of dream. To
attain the energy form of Vis.n.u is to enter the fourth state. When one identifies
oneself with this, this is called the state beyond the fourth (turyātītam). One
must thus realise with one-pointed attention (anusam . dhayet) that the supreme
mantra which is above all mantras and that the two [conditions] subtle and
gross, as well as the fifth state, of the ātman, are but one. This one-pointed
attention once attained, the japa is then to begin.23

Then a bhāvanā is to take place through which the adept, thanks to the intensity
of his mental effort, identifies himself with the images he causes to appear and
perceives mentally as present in his body:

The supreme Lord, resplendent, of crystal purity, abiding in the centre of

the lotus of the heart, shines forth by his own luminous power (svatejasā).
His great Energy made of sound (śabdamayīśakti) flows like fire, and the
pure succession of the phonemes which is hers and of the same nature as Her
comes forth engendering the mantras. From Her is born the King of mantras,
sprouting out of Her like the flower of a creeper, O Best of Munis! This is how
one must realise (smaret) how the mantra appears in the rosary. Then, it must
be dissolved again in the heart. Such is [the bhāvanā] of the only Lord of the
mantra appearing inside and outside.24

This meditative practice strengthens the link between the adept and the deity he
will worship whilst reciting his rosary since he experiments visually the deity and
its mantra as present in his heart, from where he instils them in the rosary and then
takes them back into himself. This process is parallel to that of the ritual worship
(pūjā) of a deity performed with an icon (mūrti, bimba) where the deity is first
to be imagined and felt as present in the heart of the officiating person, who will
transfer it mentally, with the proper rites, into the cult icon or on a man.d.ala, which,
vitalised, animated by the divine presence thus instilled in it, will be ready to be
used for the ritual worship25 at the end of which the officiant will take back the
deity into himself. Such a practice has the effect of bringing about an intensely
experienced participation of the officiating person in the rite he performs. The JS
underlines the union thus brought about between the reciter of the mantra and the
deity: it is in a state of mental union or identification with Vis.n.u that the beads
are to be told one after the other: vis.n.uvad yogam āsrtya aks.am aks.am . samāharet
(śl. 71b). ˚
The japa itself can now be performed. It consists of 100,000 repetitions of the
mantra (śl. 69) – which implies that it lasts several days, or weeks. The beads are
to be counted carefully (prayatnāt), for it is by concentrating on this action that
the prescribed aim of union with Vis.n.u is attained. Each repetition of the mantra,
on each bead, is to be done together with the movement of the breath (prān.odaye-
naiva). An important point, too, is that the count of the beads should not overstep
the central ‘Meru’ bead: one must go round it (pradaks.in.am . kuryāt) because this
 A Hindu rosary ritual 87
bead is the ‘body’ (mūrti) of the formless mantra (nis.kalasya mantrasya) – it is
‘the bead-form of the pure omnipresent supreme Soul’.26 A different aspect of the
ritual performance of the japa is then mentioned: the JS, śl. 76b–78a, says now
that the adept is to have different imaginary perceptions of his rosary according
to the different results aimed at when he performs the japa. If the japa is made to
appease (śāntika), the rosary is to be imagined as translucent like a pure crystal;
if the aim is to gain riches (paus..tika), it is imagined as yellow; if to dominate oth-
ers (vaśya), it is to be seen as red like the kim . śuka flower; for attracting others
(ākars.an.a) it will be orange like the nrpaśaila; black in the case of māran.a, that
is to kill or destroy; blue like the wing˚of a jay when practised for causing discord
(vidveśa); and dark grey when used for driving away (uccāt.ana). These actions for
which a japa can be performed are a variant of the group of magical acts described
in many tantras as the ‘six acts’ (s.at.karmān.i) of magic, their number being usu-
ally six. They are among the usual aims of the so-called optional, kāmya, rites,
that is, rites undertaken for the attainment of some benefit.27 The JS, 78b-84, then
goes on to mention other such acts. It says first that there are two sorts of japa:
one luminous (jyotirmaya), which is inferior (apara), and the other made of word
or sound (śabdākhya, lit. ‘called word’) which is superior (para). The first one is
used for such magical aims as entering in another body (paradehapraveśa) or suc-
ceeding in doing other ‘cruel’ ritual acts (krūrakarmān.i); for this the pure sound of
the recited mantra is to be imagined as dissolved in light, this being done also to
gain rewards or liberation (bhuktimukti) and to destroy defects and suffering. The
other form of japa, said to be ‘with sound’ (saśabda), is used for the appeasing
rites (śāntau), and to succeed in all one does, to obtain happiness and pleasure as
well as for such actions as kindling the ritual fire (ādhāna), etc. ‘It is to be men-
tally perceived as Janārdhana shining like drops of water in the night. One must
imagine that with one’s mantra the eternally manifest sound issues without any
[articulation] effort from the lotus of the heart … The mantra thus recited gives all
desired fruits.’28
One may well find it surprising that the first uses quoted here of japa, usu-
ally considered a devotional action, preceded and accompanied here by ritual and
meditative practices which place it on a high and subtle metaphysical–religious
plane, should be acts of magic, of black magic even – magical powers being men-
tioned before liberation. But Tantric texts, whether Vais.n.ava or Śaiva, are meant
first, as mentioned before (note 2), for the initiated adept, the sādhaka, who is a
bubhuks.u, looking for rewards (bhukti) or powers (siddhi) to be obtained by mas-
tering a mantra, not a mumuks.u, looking for liberation (moks.a).
The JS 84ff. goes on to say that the mantra is to be recited whilst being care-
fully intent on the wished for end. This must be done going regularly from syllable
to syllable, without any fault, neither too fast nor too slow, clearly uttering all
words, with a concentrated mind intent on the highest thought (atyutkrs..tadhiyā).
The japa, it is also said, is to be performed in the morning, at midday˚ and in the
evening, that is, at the usual times of the daily obligatory (nitya) cult.
Finally, the JS (90–95b) describes how the adept is to replace the string of the
rosary when it is worn out. This is to be done according to precise ritual rules, for
88 A Hindu rosary ritual
the rosary having been divinised, as we have seen, cannot be disposed of by sim-
ply throwing it or storing it away. Thus:

having cut the string at its ‘point of junction’ [that is, where the knot which
holds the rosary is] whilst uttering the visarjana (dismissal) mantra,29 accom-
panied by the appropriate mudrā, O Narada! then having made a new string
and having consecrated it as said [before], one is to place on it the grains as
is proper. Having knotted it, the mantrin is to bathe it according to rule, then
to impose on it [the mantra and so forth. This is to be done] for fear that the
string of the first rosary breaks. A cult is therefore to be performed, then an
offering in the fire (homa), for [the breaking of the string] causes the destruc-
tion of the [good] karman. Having taken hold of the old string, he must divide
its three constitutive threads, and, having knotted them around a stone he
must throw it in deep water uttering Vis.vaks.en.a’s30 mantra, and, having made
an offering, he is to invoke Hari whilst [reciting] seven times the mūlamantra,
O Best of Brahmins! with a fully concentrated mind.31

It thus appears that from the time one assembles the constituent parts of the rosary
to the time one disposes of them, the rosary is the constant object of complex ritual
actions. This is because, for its user, who for such texts as the JS is a sādhaka,
the aks.amālā is less an instrument of devotion than a means to put into action the
power of the Word and use it for specific ends. Such ends can be reached only
through a thorough and minutely organised ‘mise en scène’ with an intense bodily
and mental participation of the mantrin in the ritual process. The mere devotional
use of the rosary does of course also exist, and that especially in the Pāñcarātra
Vais.n.ava context, and in such cases all ritual is transcended. But this devotional
use does not interest the text we have seen, which, on the contrary, shows a char-
acteristic case of the ritual proliferation typical of Tantric Hinduism.
6 On the defects and the
perfecting of mantras1
Mantrados.a and mantrasam
. skāra

As we have seen, Tantric mantras are the phonic form of deities – deities as man-
tras or mantras as deities. They are in this respect divine entities existing eter-
nally as aspects of the supreme Word (vāc), or at least existing on a pure, perfect
level of speech, free of all imperfections. But they are also conceived as limited
entities, either as ‘souls’ (an.u), of different degrees of perfection, or as spiritual
powers existing on different levels of the cosmos. Finally, in ritual practice, they
are empirically existing forms of phonic utterance: they are quoted in ritual manu-
als, ritually, often audibly, uttered and heard, sometimes even written, partaking
thus of the imperfection of the created world.2 This explains why they can suf-
fer from flaws or defects, dos.as, of which they must be freed so as to recover
their pristine perfection and be able to be perfectly effective (or harmless for
their user). Hence the existence of such ritual practices as mantraśuddhi and
mantrasam . skāra.
We may note in passing that, whether pristine or retrieved, the divine, tran-
scendent nature of mantras is always present in practice on the empirical, human
plane, since mantras are to be used in this world as empirically uttered formulas
of a divine nature. They would be powerless if they were not divine: the immate-
rial absolute essence and the physically uttered sound are thus brought together in
action: the nature of mantras is contradictory, paradoxical, even.3

Mantrados.a
A number of Tantric works deal with mantrados.as. I shall refer here to a few
of them only, mostly from the Śaiva traditions. The longest list of dos.as I know
of is that of the Kulārn.avatantra (KT), one of the main Kaula tantras (dated
c. eleventh to fourteenth century). The KT 15.65–69 quotes sixty dos.as. The
eleventh-century Śāradātilaka (ŚT), of Laks.man.adeśika, one of the most impor-
tant digests of mantraśāstra, enumerates (2.64–110) then describes fifty such
defects, its descriptions being expatiated upon in Rāghavabhat.t. a’s commentary,
where other works are quoted – notably the Pin.galāmata, an important ancient
Yāmalatantra, and the Mantramuktāvali, a voluminous, more recent, work.4
The Tantrarājatantra (TRT), a Śrīvidyā work (1.75–48 or 1.73–815), mentions
twenty-five dos.as, the Kaulāvalīnirn.aya (7.10–12) a dozen. The earliest (eighth
90 On the defects and the perfecting of mantras
century) of these texts, the Netratantra (NT), in a passage on the efficacy of the
netramantra (8.59–63), quotes nine dos.as as instances of such defects as can be
cured by a particular use of that mantra. Many other works mention mantrados.as
without expatiating on the subject: the Gandharvatantra (GT), chapter 9, for
instance, the Somaśambhupaddhati (SP), the Īśanaśivaguudevapaddhati (ĪŚGP),
Appayadīks.ita’s Śivārcanacandrikā, and so forth.
The lists of mantrados.as I have seen do not appear to enumerate them accord-
ing to any logical order, nor do these listings follow the same order: they look like
random collections. If one compares the two longest lists, those of the KT (sixty
dos.as) and of the ŚT (fifty), only twenty dos.as are quoted in both texts and ten
only are among the twenty-five ones of the TRT. This results in a total of nearly
a hundred: a rather large number. The dos.as enumerated are not always clearly
described. The ŚT (2.72–110) describes all the dos.as (with Rāghavabhat.t. a’s added
comments). Ks.emarāja’s Uddyota on NT 8.59–63) gives some explanations and
so does, too, briefly, Prān.amañjarī on TRT. If the names of some dos.as are self-
explanatory, others are not, having sometimes apparently little or nothing to do
with their actual nature.
The dos.as enumerated in the ŚT 2.64–110 are all described as being of a purely
linguistic nature, that is, as resulting either from the presence of particular letters
or bījas or from the fact that these letters or bījas occur in certain portions of the
mantra, or else because the mantra has a particular, inauspicious, number of syl-
lables. Thus, if we look at this list, we find that a mantra of one syllable (aks.ara)
is niram. śa; one of two letters is sattvahīna, devoid of essence or reality; a mantra
6

of four varn.as is kekara (squint-eyed); one with two bhūbījas (including twice the
bīja LAM . of the earth, that is) is ruddha, ‘obstructed’, bringing neither bhukti nor
mukti. A mantra of three letters and without ham . sa (HAM . ) is said to be in deep sleep
(sus.upta) and therefore ineffective;7 and so forth. One may well ask why defects
considered as resulting from purely phonetic traits should often be called by names
evoking human bodily or mental traits (squint-eyed, sluggish, frightened, and so
forth). This may be nothing else than a picturesque way of expressing the alleged
inability of such mantras to effect properly what they are supposed to do. The ver-
bal identification of mantras with human beings, however, is perhaps not entirely
whimsical: it may reflect somehow the conception of mantras as particular types
of beings, as individual souls (an.u), which is indeed how they are conceived of in
the Shaivasiddhānta, which is also consonant with the Śaiva system of the seven
categories of experiencers (pramātr), one of which is that of mantras.
The fact that negative effects on ˚ the efficacy of mantras should be attached
to certain traits of their phonetic pattern should not surprise us, for mantras are
utterances whose power is grounded in their phonetic substance, in their form,
which is therefore essential. As is well known, the phonetic content and pattern of
mantras is far more important that their meaning – supposing they have a meaning,
which is often not the case. This has been mentioned before.8 But mantras are also
never spontaneous, improvised utterances. They are conventional rule-governed
utterances transmitted by tradition. They are the essential, phonic, form of deities
(expounded, for the adept in dhyānaślokas), or words of power transmitted by the
 On the defects and the perfecting of mantras 91
ritual process of uddhāra, their phonetic content being always carefully described
in ritual texts so as to ensure their faultless transmission, their original perfection.
How then can such imperfections as those listed in the ŚT or other such works
happen? Also, why should some syllable or word order be deemed faulty when
such a large number of very different mantras exist? To the outside observer, this
looks very arbitrary – which it probably largely is. A possible explanation could be
that patterns listed as faulty are phonetic patterns that should not result from the
rules of mantroddhāra laid down by a particular tradition and which would appear
if these rules were not followed. It is also possible that these lists are in fact meant
to exclude phonic patterns proper to other traditions. The lists of dos.as may also
prove useful in such mantric practices as vidarbha, pallava, grathana,9 etc. where
the transpositions, interpolations, etc. of the syllables or words of a mantra can
result in one of the forbidden patterns. The question, however, remains as to why
a number of patterns that do not look prima facie abnormal should be set apart,
being considered as faulty or inefficacious.
There are, however, other dos.as which (more reasonably, we would be tempted
to say) result from the way in which a mantra is transmitted, uttered, or made
use of. Such is the case of several of the twenty-five dos.as listed in the TRT
(1.75–80). Thus a mantra which has been heard by a third person is ‘burnt’
(dagdha), it is trasta (frightened) when not recited the prescribed number of times,
garvita (haughty) when not transmitted according to the rules, chinna or khan.d.ita
(cut or torn) when incomplete, hīnavīrya (without force) when given by a master
who does not possess the adhikāra, the authority to transmit it, and so forth. The
TRT also mentions the case of mantras not used at the proper time, not recited
as they should be (they are rugna, destroyed) when not uttered clearly, klis..ta (in
bad condition) when recited too slowly, avamanita (despised) when recited with-
out paying attention or without faith, etc.). The imperfections in those cases lie
more with the user than with the mantra, but they are nevertheless classified as
mantrados.as, defects to be carefully avoided since mantras are rule-governed
utterances and are therefore effective only insofar as these rules are carefully
respected.
Mantrados.as can also result from the fact that they are uttered in cases
when they are not to be mentally formulated. The mantra then issues from
the mouth of the mantrin, hence, for some texts, the existence of dos.as due to
the contact of the mantra with the organs of speech. One finds this sort of
dos.a in various texts, in the Śivārcan.acandrikā 10 for instance, or in a passage
of Aghoraśiva’s commentary on ŚP 1.3,43 (vol. 1, p. 145) where he quotes an
unidentified text which runs: dantādhāros..t.tasam . care mantrasyāśaucanirhanāt/
nādāntoccāren.aivamantraśuddhir udāhrtā, mentioning thus both cause of impu-
˚
rity and its cure. There is also the TRT 1.80 which says that mantras suffer from a
dos.a when the teeth of the mantrin are not clean ...
Among the dos.as quoted in various texts there is also the case where a mantra is
said to be asleep (supta, susupta, prasupta, svāpaga). According to the ŚT (2.84),
for instance, a mantra is susupta when it has three varn.as and no ham . sa, this
being apparently a defective phonetic pattern. For other texts, a supta or svāpaga
92 On the defects and the perfecting of mantras
mantra is one used when it is ‘asleep’ or not ‘awakened’ at the proper time. This
we find, for instance, in the TRT 1.77. Other texts, however, do not consider the
supta (or svāpaga, etc.) state of the mantra as a dos.a, but as a state in which a non-
defective mantra happens to be during certain periods of time, as are human
beings, or because of a particular movement of prān.a. Supta in this case, is taken
as opposed to prabuddha:11 it is not a case of dos.a.
The KT 15.57–63, as mentioned previously, enumerates sixty dosas. It begins,
however, by mentioning some preliminary conditions for a mantra to be effective.
The first is that it should not suffer from the impurity of birth at its beginning or
of death at its end, which would occur, it seems, if it were not to begin and end
with OM . . The ignorance of its aim (artha) or of its ‘consciousness’ (caitanya) – by
which is meant, we may assume, the lack of consciousness of the mantrin – would
rob the mantra of all efficiency; as would also the ignorance of the yonimudrā.
Some other preliminary conditions are also mentioned. The same defect, with the
same sort of śuddhi, is also to be found in the GT. The dos.as listed there (65–69)
are only partly the same as those of the ŚT, and include such traits as ripu usually
considered not as a mantrados.a but as a bad am . śa.
We may note, finally, on the problem of the faultless state of mantras, that though
they must be pure to be effective, they must also, according to some, not avoid, but
on the contrary bear necessarily a particular form of impurity, the aiśvaryamala,
the impurity connected with power, which exists even in the purest beings inso-
far as they act. In the system of the seven pramātr, this impurity as well as the
‘impurity of authority’ (adhikāramala), permitting ˚an activity, is considered to be
present in mantras since they are considered as a category of conscious subjects
(pramātr), and this gives them the capacity to act in this impure world.
˚
The general rule is, however, that mantras must be without defects or impurity
so as to be effective. We shall therefore see now how this necessary purity is given
or restored to them.

Mantraśuddhi mantrasam
. skāra
A mantra is to be made free of all defects so as not only to be more effective but also
to protect its user from the dire effects which could result from a flawed formula.
This is done through a ritual practice usually called mantraśuddhi or mantraśodhana
(purification of the mantra), which is mentioned or described in several texts. Of
these the older and more authoritative ones (among those I have access to) are the
NT (with Ks.emarāja’s commentary), which enumerates nine sam . skāras, and the
ŚT, listing ten. Ks.emarāja quotes the Ucchus.matantra12 (recording seven sam . skāras
only); whilst Rāghavabhat.t.a’s commentary on the ŚT cites abundantly the ancient
yāmalatantra Pin.galamata. Mantrasam . skāras are also enumerated in more recent
works such as Krs.n.ānanda’s Tantrasāra (sixteenth century), which is quoted in
˚
the TBhS (seventeenth century). The practice has not disappeared: the texts
quoted here are, I believe, still made use of in India by practitioners of mantras.
We will limit ourselves to quoting and comparing the mantrasam . skārās of
the NT and the ŚT which, interestingly, are practically the same in name and
 On the defects and the perfecting of mantras 93
aim whilst having to be performed in completely different ways. The nine NT
(18.6–8) sam . skāras are: dīpana, bodhana, tād.ana, abhis.ecana, vimalīkaran.a,
indhananiveśana, sam . tarpan.a, guptibhāva and āpyāyana, that is illuminating,
awakening, thrumming, sprinkling, purifying, using as fuel, satiating, hiding and
strengthening. ‘Thanks to these nine sam . skāras’, says the NT (18.8), ‘one who
masters the science of the mantras (mantravāda) will be master of the efficiency
of the mantras.’ According to Ksemarāja’s commentary (and to the Ucchus.matan-
tra), most of these sam . skāras are simply linguistic devices, consisting as they do
in adding before the mantra and/or at its end such elements as OM . , or as one of
the jātis (the ritual exclamations at the end of mantras), or in interpolating them in
the mantra. Dīpana is thus done by adding two OM . , bodhana by adding namah.,
sam. tarpana is done with two LAM. (the bija of water). Tādana, however, as its
name implies, consists in thrumming13 the letters of the mantra (which is thus
to be written) whilst uttering twice the jāti phat.. Guptibhāva is more complex
since it consists in ‘enclosing’ (sam . put.īkaran.a ) the mantra between two Netra-
14

mantras and then reciting it ten thousand times. Sam . put.īkaran.a is also used in
indhananiveśana, ‘using as fuel’, the mantra being able in that case to burn all
fetters or poisons afflicting a person.
The ŚT 2.112–123 quotes, and then describes, ten mantrasam . skāras which
include eight of those of the NT list excepting indhananiveśana, to which are
added at the outset janana (birth) and jīvana (vivifying), which are not in the NT.
These two first sam . kāras are, in fact, preliminary rites since janana is simply the
extraction of the letters, the aks.aras, of the mantra from among the fifty phonemes,
the mātrkās,15 of the Sanskrit alphabet ritually written on a lotus-shaped man.d.ala,
which is˚ nothing else than a variant of the rite of mantroddhāra described in chap-
ter 2 of this book. It is the preliminary rite necessary for all mantras to exist as
words of power. It is here quite logically followed by jīvana, since once ritually
‘born’, the mantra, to be effective, is to be animated by this rite, which, according
to the ŚT, is done by reciting (japa) the mantra a hundred times while interpolat-
ing the pran.ava OM . between each of its letters: an oral rite, therefore. The eight
other sam . skāras, however, though being the same in name as those prescribed
in the NT, differ from these in that they are not purely linguistic, oral, perform-
ances, but rites performed on a written mantra. Thus bodhana is done by hitting
the written mantra with laurel flowers while uttering the fire-bīja RAM . . Abhis.eka
is done by touching the mantra written on birch-bark with aśvattha leaves (this to
be done 108 times according to the Pin.galamata, or as many times as there are let-
ters of the mantra according to ‘another tantra’, both quoted in the commentary),
and so forth for other sam . skāras. The purification, vimalīkaran.a, is to be done,
according to S.T 2.119, by ‘burning’ the mantra with a fiery mantra (jyotirman-
tra), but it may also be done, according to the commentary, by a yogic practice,
the raising of the mantrin’s kun.d.alinī. For dīpana and guptībhava, mantras and
japa are prescribed. We may be tempted to consider this variety of procedures
– purely linguistic or ritual, written or oral, or yogic, as illogical, or even whimsi-
cal, which they perhaps are. The reason why a particular procedure is to be used
for a particular sam . skāra surely does not appear logical. However, whether logical
˚
94 On the defects and the perfecting of mantras
or whimsical, this variety can (ought?) be seen as reflecting the complex nature
of mantras which are speech acts, thus oral, but are put into action by ritual; they
are endowed with a divine power of which kun.d.alinī is the bodily aspect; and in
actual practice they can be written. The KT 15.71–72 enumerates the same ten
sam. skāras: ‘As a sword rubbed on a whetstone gets sharpened, concludes the tan-
tra, in the same way the mantras bloom and radiate thanks to the ten sam . skāras.’
7 Mantric practices and the
nature of mantric utterance1

Most ancient Tantric texts, and some more recent ones too, contain a mixture
of sometimes very elaborate theological and metaphysical notions with a mysti-
cal and soteriological aim, together with a variety of magical speculations and
practices, these two categories of notions and actions forming two very largely
overlapping areas rather than two separate ones. A coincidence that should not
surprise us since rites are governed by rules, that is, by theoretical notions which
give them their meaning. The Tantric initiate for whose benefit such texts were
composed, the sādhaka, was normally not a mumuks.u, a seeker of liberation, but
a bubhuks.u, a religious virtuoso (to use Weber’s terminology) intent on obtaining
supernatural powers or rewards and for whom moks.a was the Tantric liberation in
life, jīvanmukti, with its accompanying siddhis. These texts thus never left aside
entirely the quest for liberation, as appears for instance in the Trika system of
upāyas expounded mainly in Abhinavagupta’s Tantrāloka (TĀ) on the basis of
earlier Tantras such as the Mālinivijayottara (MVT) and the Siddhayogeśvarīmata.
Considering the fundamental role of mantras both in the quest for liberation and
in the acquisition and practice of powers, it is not surprising that in all these works
mantras are described as effective in those two fields. The mantras used for these
various purposes are sometimes different; but they may also be the same ones. I
would like to consider briefly here a group of mantric practices used when magical
effect are aimed at but which can also be used for other reasons – these two differ-
ent sorts of purposes being admittedly not easy to distinguish.
I shall consider here a passage from the Netratantra (NT),2 a Śaiva text (undat-
able, but surely not later than the tenth century AD), which was commented upon
by Ks.emarāja (fl. c.1000–1050), a disciple of Abhinavagupta, and was interpreted
by him in the spirit of the non-dualistic Śaiva system of the Kashmirian exegesis
of the Trika-Pratyabhijñā. The NT (often called by Ks.emarāja Mrtyujitbhat. t. āraka)
extols the ‘Eye’ (netra) of Śiva (sometimes also called Amrteśvara ˚ or Amrteśa)
˚
and especially the mantra of that deity, the netramantra (also named mrtyujit ˚ or
amrteśvaramantra), the basic form of which is Om ˚
. jum
. sah.. The text describes at
˚ the mystical, yogic and ritual or magical uses of that sacred formula.
length
Chapter 18 of NT describes a ritual called śrīyāga thanks to which the ‘good
sādhaka’ initiated in the netramantra will be able to oppose those who try to coun-
teract the action of his mantra or to make it turn against him. This ‘countermagic’
96 Mantric practices and the nature of mantric utterance
practice is technically called pratyan.girā, pratyan.girasa or pratyan.giratva3. To
make his own mantra irresistible, the sādhaka first submits it to the action of some
ritual treatments, of which nine are enumerated.4 One may, however, the tantra
adds (śl. 9), give the mantra eleven different forms ‘thanks to which the sādhakas
will succeed’, that is, they will obtain what they are seeking. śl. 10b–12a list these
forms:

samputam . grathitam
. grastam . samastam
. ca vidarbhitam /
ākrantam. ca tathādyantam . garbhastham. sarvatovrtam /
tathā yuktividarbham . ca vidarbhagrathitam tathā˚ //
ity ekadaśadhā mantrā niyuktāh. siddhidāh. smrtāh. /
˚
Thus listed are eleven different ways of interlocking the netramantra with the
name (or designation) of the person aimed at, or the action to be effected: the
nāma, abhidheya or sādhya.
The grammar of the above list, given here as printed in the KSTS edition of the
NT, is not satisfactory. It is clear, however, that it enumerates some things that are
to be done to the mantra, and this is indeed how Ks.emarāja explains these terms in
his Uddyota (vol. 2, pp. 77–79), confirming his point, in each case, with a quota-
tion from an unspecified text, probably the Ucchus.matantra. His definitions coin-
cide with those given for several of these terms in other Tantric works (of which I
have checked only a few), the practices listed here being fairly common in magic,
especially for the so-called s.at.karmān.i, the ‘six [magical] actions’ described in a
number of Tantric texts whether Hindu or Buddhist.
Here are these eleven terms – and the corresponding practices:

Sam . put.a: the word means a bowl or a rounded casket, or the space enclosed inside
these objects, or between two covers, etc. The mantra (or any part thereof, or
any element with which it is associated) must be symbolically ‘encased’ inside
such a space or casket, covered as with a lid or within two covers.5 Ks.emarāja
defines it here as the placing of the mantra before and after the other ele-
ment which is thus encased, as it were, within it: ādyantayor mantranyāsah. sa
sam . put.avat. The pattern is thus B A B (where A is the mantra and B the nāma/
abhidheya / sādhya). The same practice is prescribed in NT 6.18, which adds (19)
that one may thus ‘encase’ any mantra within the amrteśvaramantra and, if it is
thus recited, this mantra, whether or not intrinsically ˚powerful, will immediately
become efficacious (NT, vol. 1, p. 131). The technique of sam . put.a (sometimes
called put.a) may also be applied to parts of a mantra, an aks.ara of which being
thus encased between another one repeated twice (cf. ŚT 7.52), the sequence is
then b1 a1 b1 b2 a2 b2, etc.
One may note that in NT 6.18–19 the technique of sam . put.a is prescribed for a
japa, that is, is used orally; but in the Uddyota on NT 6.18 Ks.emarāja describes
it as written: mantram ādau likhet. In the Agnipurān.a (AgP), chapter 138, where
sam . put.a is prescribed for the magical action of vaśīkaran.a and ākars.an.a, it is
described as the placing of the mantra around the sādhya – above, under and to
 Mantric practices and the nature of mantric utterance 97
its right and left: a spatial pattern, not an oral operation. The mantra, in this case,
though theoretically a sound pattern, not a written series of letters, is mentally
conceived as disposed in that way whilst it is uttered.

grathita: this practice is also known as grathana – from the verbal root GRATH
(GRANTH), to bind. It consists in binding together, as it were, the mantra and
the sādhya, as in a knot, by alternating their syllables, ‘each syllable of the
abhidheya being enclosed within those of the mantra’, according to the quotation
made by Ks.emarāja: abhidheyārn.am ekekaikam . mantravarn.aih. sam . put.īkrtam.
These words are also quoted by Jayaratha in his Vivaran.a on the VM 2.35–6 ˚
(or NS.V, 2.36–37, p. 157),6 which prescribes the enclosing of each letter of the
sādhya within the bīja of Kāmakalā (ĪM . ), this being done (in writing) on a dia-
gram, around the border of which all the letters of the alphabet are to be inscribed.7
The AgP 138.6, the Tantrarājatantra (TRT)1.72, or the ĪŚGP 1.5 give the same
definition of grathana as Ks.emarāja.

grasta: the term means swallowed or eclipsed, but also surrounded. Its use, for
a practice consisting (according to Ks.emarāja) in putting the name or sādhya
in the middle, with the mantra written on all four sides (madhyasthasya nāmno
dikcatus..taye mantraniveśah.), appears thus wholly justified.8 This practice is in
fact what the AgP 138 calls sam . put.īkaran.a.
In its present sense grasta is not found in the AgP nor in the TRT, the ĪŚGP or
the Phet.karin.ītantra (Phet. K). This can be said of the following term also.

samasta: the nāman is uttered or written before the mantra, this being done twice:
B A B A. Samasta meaning ‘put together’, ‘united’, the term could apply to any
other combination as well. Samasta like grasta seems proper to the NT.

vidarbhitam: ‘first the name, then the mantra is vidarbha’, says Ks.emarāja, and
he quotes: abhidheyam . bhavet pūrvam . tato mantra sakrd bhavet / vidarbhitam.
˚ the letters of the mantra
The Phet. K 3.148–149 defines it differently as alternating
and those of the abhidheya, the pattern being not B A, but b1 a2 b2 a3 b3, etc. The
AgP 138.7 says that one should alternate two syllables of the mantra with one of
the sādhya (mantrāks.aram . dvayam. likhyā sādhyāks.aram punah.): a1 a2 b1 a3 a4
b2 a5, and so forth.

The VMT 2.33–34 (or NS.A 2.34–35) prescribes this practice for an amulet
where the letters of the mantra and sādhya are inscribed in a śrīcakra, a practice
described differently by Śivānanda in his R juvimarśinī (as: a1 b1 a2 b2 a3 b3 etc.)
and by Vidyānanda in the Artharatnāvalī˚(a1 a2 b1 a3 a4 b2, etc.). The same text
(śl. 21–22) prescribes a nyāsa of the mūlamantra on the image of the goddess
entwined with the an.kuśabīja (KROM . ). There are thus different forms of vidar-
9

bha in japa or in writing.10 Whatever its form, however, vidarbha always consists
of associating sādhya and mantra while cutting apart somehow either these two
elements, or the whole formula, or their constituting syllables.
98 Mantric practices and the nature of mantric utterance
But why is this called vidarbha? In this technical mantric sense, the term seems
not to be found in Sanskrit dictionaries or encyclopaedias. Edgerton’s Buddhist
Hybrid Sanskrit Dictionary, however, defines vidarbhayati (with vidarbhana
and pravidarbha) as ‘intertwines (letters), i.e. writes them between letters of a
mantra’, and it refers to several passages of the Sādhanamālā. As I have shown
elsewhere,11 the term can be explained by reference to darbha, the grass of that
name, which has a cutting edge. The Śabdakalpadruma says: ‘darbhah. – drn.ati
vidārayati, dradalibhyām ˚
. bhah.’, Unādisūtra 3.151, ‘iti bhas’, that is, it refers
darbha, in the sense of the verbal root DR, ‘to cut’, and since Unā 3.151 says that
after DR and DAL one must have bha, this ˚ results in darbha. The prefix vi would
˚
then add the notion of separation, mantra and sādhya being cut off and separated.
darbha can also be linked with the root DRBH, to string or tie together, the practice
of vidarbha being then understood as first˚ separating (vi), then stringing together
mantra and sādhya.

The following six mantric practices – together with grasta and samasta – are, I
believe, found in the NT only.

ākrantam: madhyasthasya nāmno mantro yadi ves..tanayā nyāsta ākrāntam, says

Ks.emarāja: ‘when the mantra is placed so as to surround the name which is in the
centre, [this is called] seized or invaded’. This looks like a variant of what NT calls
grasta and AgP sam . put.īkaran.a. (Note that the past participle ākrānta applies to
the nāman, not to the mantra.)

ādyanta: mantrād anantaram . nāma tatas trimantra ity ādyantam, ‘the name after
the mantra, then thrice the mantra, this is “before and after’’’. The sequence is thus
A B A A A .The term is self-explanatory.

garbhastha: madhyasthasya mantrasya caturdikkam . sādhyanāmanyāso garbhasth-

atvam, says Ks.emarāja. This is in fact AgP’s definition of sam . put.a, except that
the Uddyota on NT mentions the four directions of space instead of the four sides
of the written mantra. It is also the reversal of what is defined as grasta. Like the
AgP’s sam. put.a, this can be done only in writing.

sarvatovrta: mantrasya ādyantayoh. sādhyanāmaniveśah. sarvatovrtam, ‘the plac-

˚ sādhya or name before and after the mantra is [having˚ it] completely
ing of the
surrounded’. This is the reversal of what is called sam
. put.a by the NT. The pattern
is B A B.

yuktividarbha: the text quoted by Ks.emarāja defines this as placing the abhidheya
before the mantra and three times after it: B A B B B. It is the inversion of ādyanta.
The reason why this practice is so called is not clear.

vidarbhagrathita: it is repeating three times the mantra after the nāman: nāmnah.
paścāt trimantranāsyah.: B A A A. The reason why it is so called is also unclear
since neither knotting (grath) nor cutting (vidarbha) occur.
 Mantric practices and the nature of mantric utterance 99
This list of eleven mantric practices of the NT is an unusual one both in that it
is longer than the usual one of six12 linked each with one of the s.at.karmān.i (which
are not mentioned here), and in that it does not include three of these more usual
ones, namely yoga, rodha and pallava, about which a few words must now be
said.

yoga: is defined in the TRT, Phet. K and ĪŚGP as placing or uttering the
sādhya before the mantra: mantrādau nāmasam . sthānam
. yoga ityabhidhīyate
(Phet. K 3.142). The AgP’s technique is different: adau mantra tatah. sādhyomadhye
sādhyah. punar manuh.. We have thus either B A or A B B A. In both cases there
occurs a conjunction or union (yoga) of the mantra and of whatever word or
formula is associated with it, such union being deemed to increase the power of
the mantra.

rodha: consists (according to AgP 138.3–4 and Phet. K 2.144) in placing the
mantra before, in the middle of, and after the sādhya – A B (A) B A: nāmnā
ādyantamadhyes.u mantrah. syād rodhah.. rodha, from the verbal root RUDH, is
probably understood as the surrounding, investing, or blockading of the sādhya by
the mantra. It is mentioned in the ĪŚGP, but not in the TRT.

pallava consists, according to the TRT or Phet. K, in putting first the mantra, then
the sādhya (A B). The sādhya appears thus as a shoot or extension (pallava), as
it were, of the mantra. The AgP has a different description – it seems to add the
mantra both before and after: ādau sādhyam. likhet pūrvam . cānte mantrasamanvi-
tam. The pattern would then be A B A.

I have described the foregoing mantric practices not so much because I find
them intrinsically interesting, but because I believe they contribute to throw some
light on the nature of mantras or of mantric utterance or, more generally, on the
Tantric conceptions of the relationship between sound or word and image or
writing.
These practices, indeed, concern mantras, that is, oral/aural formulas transmitted
by word of mouth from master to disciple and put into practice through utterance
or recitation (uccāra, japa) only. Mantras can also be enunciated silently (tūs.n.īm);
such silent utterance (where the utterer ‘restrains his voice’, vācam. yacchati), and
still more the inner and purely mental one (mānasa), being in fact considered as
the highest and most meritorious ones: the mental utterance is the highest because
the Absolute, the supreme godhead is silent. Silence is higher than the highest
plane of the word (vāc) because it is its primal Source. The mantrin, in such case,
evokes (or perhaps fuses with) a transcendent plane of vāc which, imperceptible,
is present as its animating power in all uttered mantras.
But how is it, then, that this power can also be evoked, made present, when a
mantra is written? First, the writing or the placing of a written formula is not mere
writing but a ritual performance which can be fruitful insofar only as it uses, as it
is permeated by, the power of vāc, of the Word which inhabits the, by nature oral,
100 Mantric practices and the nature of mantric utterance
mantra thus written down. The written formula is thus (deemed to be) infused with
the energy of the oral formula. This explains how and why written mantric prac-
tices can be prescribed in a work such as the NT whose mūlamantra, the mrtyujit
OM ˚
. JUM . SAH., is in principle effective only when uttered. It explains too why such
practices are endorsed by Ks.emarāja, that is, by one of the main exponents of the
non-dualistic Kaśmirian brand of Śaivism, whose concept of mantra is directly
issued from that of the Śivasūtra (ŚS 2.1: cittam . mantrah.), the main commentary
on which is Ks.emarāja’s Vimarśinī, a work where he states that the supreme Word
(parā vāc) or Consciousness (sam . vit) is the true nature of mantras.
Another explanation, and one that fits with the doctrine of the Trika-based
pratyabhijñā, is that mantras exist and function on different levels. Abhina-
vagupta thus said that by nature they ‘consist of discursive thought as well as
pure consciousness’ (vikalpasam . vidmayah.). In their functioning, too, they may
13

well appear differently. They may be felt and used as pure consciousness (sam .-
vit) by the liberation seeker (the mumuks.u) – and still more by the liberated in
life (jīvanmukta), and be also (even simultaneously) used audibly on the plane of
discursive thought and of the gross form of the word (vaikharī); they can even be
written by a mantrin, a sādhaka, performing kāmya rites (which are in fact those
which are the main concern of the NT).
These uses may differ widely; they are, however, grounded in the same, funda-
mentally oral, conception of mantras. Just as magic is never entirely separate from
religion, of whose concepts it is merely a profane use, so mantras, even in their
lowest, magical uses, cannot be separated from their highest, purely oral ones. In
fact, such lower uses of mantras are effective only insofar as all mantras are in
essence the supreme Word, pure divine energy.14
Thus a text such as the NT may expound some written (and in some respects
magical) uses of mantras whilst still upholding the essentially spiritual nature of
these ritual formulas. Let us take also, for instance, the case of mantras or bījas
placed in ritual diagrams (cakra, yantra or man.d.ala). These may be placed in
the diagram by the rite of nyāsa, which is at the same time a bodily and oral
and mental action.15 They can also be inscribed, written, there. Some texts
assert explicitly the power, value or meaning of these written mantras. A typical
instance is that of the kāmakalā as described in the Yoginīhrdaya (YH), one of the
basic texts of the Śrīvidyā tradition.16 A passage of this work ˚ (YH 2.21, Dīpikā,
pp. 130–131) describes – rather cryptically – the so-called kāmakalācakra as made
up of two intersecting triangles with the bīja ĪM . inscribed in the central part of the
cakra.17 This bīja is to be meditated both as the vācaka of the goddess and as the
kund.alinī (coiled like the letter Ī in nāgarī script) in her ascent from mūlādhāra to
brahmarandhra: the ritual process involved here is both visual and mental, resting
as it does on the meditation of a ‘written sound’. The Ī of the kāmakalā can also be
meditated as made up of three dots, that is, as it is written in the Brāhmī script: this
use of an ancient script is perhaps a case of deliberate archaism. It can be taken,
too, as suggesting an (unlikely!) ancient origin of this practice.
In the ritual of nyāsa, too, the mental visualisation of the written shape of the
bīja that is being placed is sometimes prescribed, or a written bīja or mantra is
 Mantric practices and the nature of mantric utterance 101
to be assigned somewhere. In the Mantraparibhās.ā (chapter 293) of the AgP, for
instance, a lipinyāsa is prescribed, for which the śaktis of Rudra (that is, the fifty
aks.aras from A to KS.A) are to be written, then ritually placed on the body of
the sādhaka (vilikhet … vinyāset, śl. 147).18 More important, perhaps as regards
the role of written letters or mantras (it is well known that any aks.ara can be
considered as a bījamantra, from AM . to AH. and to KS.AM . ) is the conception of
Lipidevī, the (written) Alphabet Goddess19 mentioned, for instance, in AgP 293, or
that of the ‘Tree of letters’, lipitaru, of the ŚT 7.08–14, under which the Goddess
is seated. Among alphabet deities is also the goddess Mālinī of the Trika (and of
some Śaiva Kaśmirian systems) which consists in a particular order of the Sanskrit
alphabet, going from N.A to PHA (the n.ādiphāntakrama).20 One finds it written in a
triangular ritual diagram, a [mālinī]gahvara shown in the Tantrasadbhāvatantra.
But, according to some texts, it can also be written in a gupta script, the letters
being correlated with body parts. Since a mantra is always fundamentally an oral
principle, when this representation of the Mālinī mantra-deity is ritually used,
that is, mentally worshipped, the practice would combine oral, visual and mental
elements.21 The written form of the letter A in the devanāgarī script is also some-
times the basis of a meditation, the different parts of the letter being meditated as
embodying the movements of the supreme energy, kund.alinī.22
We must also refer here to the ritual procedures used for the uddhāra, or for
the research of the am . śa of mantras, where letters of the Sanskrit alphabet are,
for the mantroddhāra, written (often in a particular secret order) on a diagram,
called prastāra or ghavara,23 or, for the research of the am . śa, on cakras of vari-
ous shapes, all procedures that are used in all Tantric traditions, whether Śaiva or
Vais.n.ava.24 Incidentally, we may remark that the prescription in tantras of such
written procedures shows that these texts, though supposedly revealed, were origi-
nally written. We must also not forget the numerous uses of written mantras on
amulets (kavaca) which, extremely frequent in modern times and nowadays, are
in fact a very ancient use.
Coming back to the oral nature of mantras and to the practices we have seen
of sam . put.a, etc., one must say that, to a large extent, these notions and practices
can be traced back to Vedic origins. The Vedas were composed several centuries
before writing began to be used in India. This explains perhaps the bias against the
writing of revealed texts that was to appear later on, when writing began to be used
(seventh to eighth century BC?). Writing was thus confined to practical, mundane
uses and excluded from religious/ritual ones. However, since India was from early
times a highly literate society, producing one of, if not the largest, literature in the
world, there inevitably occurred interactions between the written and the unwrit-
ten traditions, a situation which resulted – to take a particular instance – in the
(apparent) discrepancy between the in principle oral–aural nature of mantras and
their written uses.25 We may note, too, that such mantric procedures as vidarbha
and so forth, where the constituting parts of a mantra are separated, interpolated,
duplicated, inverted, etc., have as their forerunners such Vedic usages as viharan.
a and vikrti (intertwining or transposing). In the same way, the use of ‘meaning-
˚
less’ syllables or bījas added to, or interpolated in the sequence of a mantra can be
102 Mantric practices and the nature of mantric utterance
traced back to the vyāhrtis of the Veda or to the syllables chanted by the Udgāta
priest of the Sāmaveda.˚
Developed in a Sanskrit tradition going back to the Vedas, the Tantric mantraśāstra
could not but participate in the main traits of some of the linguistic aspects of that
tradition (to which, from a certain time onward, it contributed largely). Seen in
this light, that is, as part of a stable and continuous but ever-changing culture, such
seemingly out of the norm uses as those we have seen here appear as merely an
aspect of a culture where the oral and the written have interacted during more than
two millennia, and thus can never be entirely separated.26
8 Body and mantra1
Mantras in the human body

An interesting aspect of Tantric mantric practices and speculations concerns the

(imaginary, of course) presence and circulation of mantras in the human body,
or, more exactly, in the structure of channels (nād.ī) and centres or nodal points
(granthi, cakra or padma) imagined in the hat.hayoga tradition as being present
in the physical human body. This imaginary pattern is often called ‘subtle body’.
But this is a misnomer since the subtle body properly so-called, sūks.maśarīra or
lin.gaśarīra, in Sanskrit, is made up of an ensemble of tattvas, forming the element
that transmigrates from body to body, and has by nature no form or aspect: it can
be conceived, but not visualised; whereas the yogic structure is to be imagined,
visualised, and it does not survive the death of the body. We shall therefore not call
it subtle body, but rather yogic or imaginary body. In this structure, with which the
Tantric adept lives and which he imagines (and sometimes visualises mentally)
as being within his body, mantras are deemed to be present in the nodal points
and to circulate in the nād.īs like a fluid, as a flow of power. To look briefly at this
bodily presence of mantras may be of some interest from the point of view of the
nature of mantras, but also as regards the way in which Tantric adepts perceive
and experience their body. For the yogic body is not only ‘intraposed’ (to use T.
Goudriaan’s felicitous expression) within the physical body, but also it oversteps
it (the dvādaśānta cakra is twelve fingerbreadths over the head). Furthermore, it
includes cosmic visionary elements which transform it (or rather transform the
way in which it is experienced), elements which also connect it, sometimes iden-
tify it, with aspects of the cosmos. The Tantrika’s experience of his body is thus of
a very specific sort, being both human and cosmic or divine.
Mantras are evidently not present by nature in the human body. They are
imagined as being there through the meditative effort of the yogin who not only
imagines abstractly their presence in his body, but also comes to ‘visualise’, to see
them mentally there, abiding in the cakras or circulating in the nādīs. Many texts,
ancient or modern, prescribe or sometimes describe this mental practice.
It is most often by nyāsa (on which see here the chapter on the subject) that man-
tras are placed on or in the body of the adept or devotee, and are thus imagined by
him to be present in him or to help him to identify himself with the deity he invokes
or worships. In this chapter, the aspects of the subject I would rather underline
are some peculiar forms of the presence of the word in the body (brought about
104 Body and mantra: mantras in the human body
notably by such ritual practices as nyāsa, but also otherwise) and the transforming
effect of this presence on the image of the body2 and therefore on the way the body
is perceived and experienced by the adept or devotee. The latter’s body appears
thus as permeated by the power of the word which may simply purify it (for the
performance of a ritual, for instance), but which can also inhabit it to such an
extent that it becomes the receptacle of the deity and even is identified with it. It is
sometimes said in such cases that the adept has (at least for the time the rite lasts)
a ‘mantra-body’ (mantraśarīra or mantradeha),3 a problematic term, for what can
such a body actually be? The problem, at least for us, is a double one. For, first,
of what kind are the mantras that are present in the body? Are they aspects of the
word, subtle forms of power, or deities? Then, how can one imagine and, still
more, how can one have the actual, ‘lived’, experience of a mantra-body? What
sort of experience and mastery the yogin has of a body thus transformed? Then,
the body of the adept fully inhabited by mantras is often described as a divine
body (dīvyadeha), or it is said that the nature of the adept becomes that of a mantra
(mantratva), which are other complex notions. By their presence in it, mantras
do not merely purify or divinise the body: being supernatural entities with a cos-
mic dimension and role, they ‘cosmicise’ it. Embodying as they do the traditional
Indian homology of the human body and the cosmos, they give a cosmic dimen-
sion to the image the Tantric adept has of his body: he transcends his limits.
A case in point is the description given in the Jayākhyasam . hitā (JS) of the cult
4

of Vis.n.u. In this text, as is the rule in Tantric ritual worship, the pūjā, the so-called
‘external’ (bāhya) cult, that is, the cult materially, visibly, performed, is preceded
by an ‘inner sacrifice’ (antaryajana), which is entirely mental, all the parts of the
cult having to be done in imagination, using as a support of these imagined actions
the image of the body of the performer. During this mental ritual process, the
officiating person resorbs, in ‘the ether of his heart’ (hrdākāśe), his consciousness
(caitanya) and his vital force (jīva) into a ‘mantra body’ ˚ (mantraśarīra), the mantra
being the mantra of the deity, therefore the god himself. The adept is then deemed
to have no other body than this divine mantra-body. This moment of imaginary
fusion with the deity is followed by several mental representations which will
achieve the complete transformation of the officiant, so that, with his purified and
transformed physical body – of which he somehow again takes possession – he
will be able to enter the next, physically performed, part of the worship.5
To come back to nyāsa, it is normally accomplished using a particular position
or gesture of the fingers (a mudrā) which are to touch the part of the body where
the mantra is to be placed, the gesture being done together with the vocal or men-
tal enunciation of that mantra, which is also to be visualised.6 What is thus to be
mentally evoked (by dhyāna) is the aspect as described in ritual texts or manuals
of iconography of the deity whose phonic essence and highest aspect is the mantra.
The mantra itself is therefore what the adept sees mentally while placing it.7 By
nyāsa one may also place abstract entities, difficult to imagine visually but which
are also symbolised by a mantra. Such is the case of such cosmic elements as the
kalās or the tattvas. These cosmic elements can be placed by nyāsa of their man-
tras on the body (of the initiand notably) which is thus imagined as ‘cosmicised’.8
 Body and mantra: mantras in the human body 105
According to some texts, what is to be imagined is the written form of the
letters, to be visualised as shining since mantras are divine and therefore lumi-
nous. The forces thus placed on the body are therefore deemed not only to inhabit
and transform it, but also somehow to illuminate it. One sees thus that the mental,
imaginal aspect of nyāsa is essential to it, the image being not merely mental, but
also experienced, felt as being present in the inner structure of the yogic body of
the performer.
This mental visual aspect of nyāsa is all the more important when mantras are to
be placed, as is sometimes prescribed, on the cakras or other centres of the yogic
body. The practice is then necessarily entirely mental, even if a gesture towards
the body is sometimes to be done: while uttering the mantra one is to imagine it
as being placed on the cakra which is thus permeated with its power. The small
figures of deities one sees in the traditional Indian images of hat.hayoga9 are the
mantras which the yogin is mentally to place there. It is always mantras that
populate (if one may use this word) the centres and channels of the yogic body
described in Tantric texts.
Another form of the presence of mantras in the body is found in such ritual
practices as the cult or invocation of deities. As we have just seen, the inner, men-
tal pūjā which precedes the external concrete ritual worship is a mental worship,
done entirely with mantras, which transforms the nature, the ‘ontological statute’,
of the body of the officiating person. The deity, in this case, is to be invoked and
imagined as present in the heart of the adept. Since this deity dominates the world
which is as it were his/her throne (āsana), all the constitutive elements of the cos-
mos, the tattvas, from that of the earth to the one that precedes the deity, are to be
imagined as tiered below the deity, being placed mentally, using mantras, by the
officiant in his body. Thus, in the mental cult (mānasayāga) of Vis.n.u described
in chapter 12 of the JS, the officiant who is already identified with the god
(vis.n.umāyā) will build up his body as the āsana of the deity by visualising medi-
tatively (by dhyāna) both the visible (anthropomorphic) and the phonetic (uttered
and/or written) aspect of the mantras of all the elements constituting the cosmos,
which will be thus placed in imagination in his body. He must first place, with
their mantras, the supernatural powers and deities present in the cosmos together
with the corresponding cosmic levels. He tiers them vertically along the axis of the
sus.umnā and the ascending path of the kun.d.alinī, beginning with the ādhāraśakti,
the supporting power which is the pedestal of the universe, imagined as being in
the mūlādhāra, continuing with the earth, the directions of space, the Vedas, the
three Luminaries (Sūrya, Soma, Agni), Sun, Moon and Fire, and other divine enti-
ties, up to the god Gārūd.a (Vis.n.u’s mount) with Varāha. The throne is to be made
up of the mantras of the twenty-seven tattvas from earth to īśvara. We may note
here that this interiorised throne thus built up with the mantras of all the universe
reflect two visions of the cosmic process: one mythical, theological, the other that
of the structure of the cosmos inherited from the Sām.khya.10
In the Śaiva domain, we find an analogous process described in the SvT 1.88ff.
for the cult of Svacchandabhairava: the throne of the god is first installed in the
body of the adept, then the mantras which constitute phonetically the cosmos, then
106 Body and mantra: mantras in the human body
the god which is to be mentally worshipped.11 Similar but more complex is the
mental installation of the three supreme goddesses of the Trika – Parā, Parāparā
and Aparā – for the purpose of initiation, shown in chapter 15 of the TĀ. The
worshipper, in that case, must first divinise his body by imposing on it mantras
of somatisation (mūrtimantra) as a preliminary to the mental creation in his yogic
body of several divine entities. Then he is to install in himself, by their mantras
tiered along the sus.umnā, the divisions of the cosmos, the tattvas, with their pre-
siding deities. This will transform him into the throne of the goddesses who are
to be imagined as above his head, each seated on a Bhairava lying on a lotus,
each placed on the tip of the three prongs of a trident issuing from the top of his
head.12
Mantras, however, are not merely present in the imaginary body of the Tantric
adept. They move inside it, spreading out their power, divinising the adept, and/or
uniting him with the supreme godhead, leading him towards liberation. Several
Śaiva or Vais.n.ava examples could be given of this intra-bodily, liberating circula-
tion of mantras which, more than the nyāsas, put into play the creative power of
imagination when associated with the existential perception of the ‘lived’ body.13
The very term uccāra, used for an action which both enunciates a mantra and
activates its power, underlines the bodily basis of this process. For uccāra denotes
the ascending movement (uc-√CAR) of the vibrating subtle phonic substance of
the mantra, the nāda (it is the nādoccāra), the inner ascent which goes together
with the movement of prān.a in the yogic body. It is an inner, intra-corporeal
movement of the vital breath, together with a movement of consciousness towards
the deity.14 The fact that this breath is ascending, associated with the ascent of
kun.d.alinī, shows that it is not a respiratory breath. It does take place in the adept’s
body, but on the level of the yogic body ‘intraposed within the visible body’, to
quote again T. Goudriaan.15
There is no dearth of cases of this intra-corporeal circulation of mantras. A first
case in point is the mantric practice of the ajapājapa, the ‘recitation of the non-
recited’ where the bodily transit of the mantra is identified with the respiratory
breath. I have described it in chapter 3 and therefore leave it aside now. I would
like, however, to draw attention to the fact that in the ajapājapa the practitioner’s
inner consciousness is to be eventually focused, not on the respiratory breath, but
on the ‘central breath’ (madhyamaprān.a) which is an aspect of the divine energy.
The practitioner’s attention shifts from concentration on the movement of breath
going on in his physical body to concentration on the vital, subtle breath which
takes place in his yogic body: there is thus a coalescence of two visions and expe-
riences of ‘breath’. Tantric texts often insist on the necessary connection between
mantra and prān.a in the sense of subtle vital breath. This is done for instance by
Abhinavagupta in the TĀ (7.39ff.) where, referring to the Siddhayogeśvarīmata,16
he says that it is in harmonious connection with the vital breath (prān.asamā) that
the kun.d.alinī reaches the transmental unmanā plane of the mantric utterance,
adding that mantras, even repeated thousand of times, would be without effect
(na siddhyanti) if not conjoined with the eternal ever-surging divine power by
being enunciated in conjunction (yukta) with the movement of prān.a.
 Body and mantra: mantras in the human body 107
We shall see now a few cases where mantras placed or deemed to be present
in the yogic body move in it without involving the physical body, which is none-
theless not entirely non-implicated, since it is the milieu where the yogic body is
imagined as present. It is therefore with his whole ‘lived body’ – his ‘Leib’ – that
the Tantric adept lives, experiences himself, even though all that happens in such
cases is, as we shall see, entirely imaginary.
One of the best-known cases of such intra-bodily mantric practices is the utter-
ance, the uccāra, of OM . , the pran.ava, whose nasal ending, M . , is considered as
persisting through eight phonic stages (kalā), from bindu to samanā. The three
phonemes, A, U, M, constitutive of OM . , which are the three first kalās, are on
the level of the (audibly or mentally) spoken word; the eight other kalās (bindu,
ardhacandra, nirodhini, nāda, nādānta, śakti, vyāpini, samanā), then unmanā,
form thereafter eight progressively more and more subtle planes of the enuncia-
tion which, in unmanā, eventually dissolves in the (silence of the) Absolute. It is
a phonic ascent going from this world of distinct speech to fusion in the silence
of the supreme godhead. Since it is an ascending movement, it naturally follows
the movement of kund.alinī, the divine power as present in the human body. The
utterance of the mantra (uccāra) and the ascent of kun.d.alinī are, from this point of
view, identical. This ascending flow of OM . is in fact sometimes explicitly named
ham . soccāra (in SvT 4.262, for instance).
There are, in Śaiva tantras, different descriptions of the uccāra of OM . : the
number of kalās may vary as well as the powers or cosmic divisions to which they
correspond. This last point is of no importance in the present context. I mention
it here to emphasise the cosmic and transcendent aspect of most mantric utter-
ances, and especially of this one. For though the uccāra of OM . is imagined as
taking place in the yogic body of the adept, this body, if only because mantras are
placed on it, has a supra-human dimension. It is patterned along the structure of
the anatomical body, but it transcends it both because it is pervaded by the divine
power and because it is the means through which the human creature can overstep
its limits and tend towards the Absolute.
To come back to the bodily presence and ascent of OM . , its ascent, associated
with the movement of prān.a and ham . sa, begins, according to SvT 4.263,17 with
A, on the level of the heart centre (hrd). It continues with U in the throat
˚
(kan..tha), M in the palate (tālu), bindu between the eyebrows (bhrūmadhya), and
nāda on the forehead (lalāta). Then, higher in the head, appear and dissolve the
next ever-more subtle planes of the phonic vibration (with the corresponding
cosmic planes), up to unmanā, which is not mentioned in the SvT but is prob-
ably above the head. The description of this practice by the NT 22.25ff. is largely
similar, OM . being, however, divided into five kalās only – A U M, bindu and kalā
– conceived as pervading (vyāpti) the body, that is, being generally present in it,
from the feet to the heart (A), from the heart to the palate (U), from the palate to
the forehead (M), and then, for nāda, from forehead to dvādaśānta, the subtle
centre twelve fingerbreadths above the head.18 This practice results in pervading
the whole body of the adept with the power of the mantra which he will thus be
able to master.
108 Body and mantra: mantras in the human body
One of the most elaborate description of the ascending utterance – the uccāra
– of a mantra is, I believe, the japa of the śrīvidyā, the mantra of the goddess
Tripurasundarī, which is to be performed at the end of her pūjā, and which is
described in the third chapter (śl. 169–174) of the Yoginīhrdaya (YH), a Śaiva text,
probably from Kashmir, dating from the tenth or eleventh ˚ century. The four stan-
zas which describe it are far from clear, but they are explained in Amrtānanda’s
(thirteenth to fourteenth century) commentary. ˚
This japa, the YH 3.169 says, consists in ‘the conjunction [carried out] in the
three parts [of the śrīvidyā] and in the threefold kun.d.alinī of the vibration of the
cakras tiered one upon the other’.19 This abstruse formula underlines the link of
the mantra and the kun.d.alinī during this ascending process. What is actually to
happen is as follows.
The mantra that the officiant is to raise from the lowest cakra, the mūlādhāra,
to the highest one, the dvādaśānta, is the śrividyā, made up of the following fif-
teen syllables: HA SA KA LA HRĪM . HA SA KA HA LA HRĪM . SA KA LA HRĪM ..
The utterance of the three HRĪM . , the hr lekhā, the flame of HRĪ, which crowns
˚
as it were each of the three groups of syllables is deemed to prolong itself, as in
the case of OM . , after the three letters HRĪ, through the nine kalās, from bindu
to unmanā we have already seen, this movement going from the lowest cakra,
the mūlādhāra, to the summit, the dvādaśānta. This is how the nāda, the phonic
vibration of the mantra associates in the same ascending movement (the same
upwards going thrust of energy) the three parts of the mantra and the three sec-
tions into which the kun.d.alinī is considered as being divided.20 This being so,
the officiant is to imagine that in the mūlādhāra are the phonemes of the first
group, HA SA KA LA crowned by HRĪM . (HRĪ + bindu), over which, going upward
along the sus.umnā (therefore together with the ascent of kun.d.alinī, prān.a and
nāda), the six kalās from ardhacandra to samanā, reaching and overstepping the
svādhis..thāna and man.ipūra cakras, the ascent going thus up to the heart (hrd
or anāhata) cakra. There, in the hrd, is now be imagined the second group ˚of
phonemes, HA SA KA HA LA followed ˚ by HRĪM . , with over them, again, the
kalās ardhacandra and so forth, which will go on ascending in the same way, over-
stepping the cakras viśuddha and tālu (neck and palate), up to the bhrūmadhya
cakra (between the eyebrows). There, the third group, SA KA LA followed by
HRĪM . , will be mentally placed, with over them the kalās which, in this case,
include unmanā, and which also carry with them in their ascent (according to
Amrtānanda’s commentary) the kalās of the two first groups. Thus the uccāra of
the ˚mantra reaches with the totality of its subtle part its highest point, where all
phonic vibration dissolves into the Absolute. There, too, it oversteps the limits
of the physical body of the adept since the plane of unmanā is reached in the
dvādaśānta cakra twelve fingerbreadth above the head. The śrīvidyā thus trav-
els through the whole yogic body of the adept, from the lowest to the highest
cakra, pervading it with the divine power of the śrīvidyā, which is the phonic
power form of the goddess, thus bringing him to the plane of the Absolute. Such
a mantric practice needs an intense mental imaginative concentration on the
part of the practitioner. If it is effectively realised and experienced, it cannot but
 Body and mantra: mantras in the human body 109
influence the officiating adept’s consciousness of himself, all the more so if the
process is to be repeated at least once every day during the daily regular (nitya)
worship of Tripurasundarī. It goes without saying, that in the everyday ritual prac-
tice of śrīvidyā adepts, such an intense identification is probably a rare case; it
ought nevertheless normally to take place since, in principle, all Tantric cults aim
at bringing about an identification of the officiant with the deity he ritually wor-
ships.21
Another interesting case of the presence and circulation of mantras in the chan-
nels and centres of the yogic body is that of the ‘subtle meditation’ (sūks.madhyāna)
of the Eye-Mantra (the Netramantra) of Śiva, god and mantra, both also named
Mrtyujit (‘Conqueror of Death’), described in chapter 7 of the NT, a meditation of
˚ there are two forms.
which
The tantra first (śl. 1–4) mentions a particularly large number of channels and
centres of the yogic body: there are six cakras (plus the dvādaśānta), sixteen ‘sup-
ports’ (ādhāra), three ‘objects’ (laks.ya), five voids (śūnya), twelve knots (granthi),
then three ‘celestial bodies’ or ‘glories’ (dhāman), these centres being connected
with each other by innumerable veins or channels, the nād.īs. This being so, for
the first method (śl. 5–15), the adept is to imagine Śiva’s energy as present in the
central ascending breath, and to place mentally on this ascending force the man-
tra which then vibrates. This subtle phonic vibration, following the course of the
sus.umnā, will pierce through the cakras, the sixteen ādhāras and the twelve
granthis till it reaches the dvādaśānta where it will fuse with Śiva. Thence it flows
down by the same way till it reaches the heart cakra which the adept is to imagine
as filling up with ambrosia (amrta) which then spreads in the whole body through
˚
the numberless nād.īs, which results in making the adept immortal: the physical
body of the yogin is thus transformed, made divine, by this mental imaginative
action on the (equally imaginary) yogic body.
The NT then describes (śl. 16–52) another practice where the meditation begins
in the mūlādhāra (here called janmasthāna, place of birth), whence the adept’s
consciousness, together with the mantra, will first go down to his toes.22 Then, it
will go upwards again, reaching in succession the centres we have seen before,
which, being as it were obstacles on the way,23 must be overstepped or pierced
through. To this effect, the yogin uses the ‘Sword of Gnosis’ (jñānaśūla) which
is, the commentary says, ‘the flashing of consciousness transformed into mantric
energy’, the power of the mantra fused with the consciousness of the adept, that is.
This is achieved by a particular yogic technique causing the ascent of kun.d.alinī.
His consciousness having thus reached, together with the mantra, the highest
point, the dvādaśānta, the adept acquires henceforth the power to move into the
‘ether of consciousness’: he is khecara.24 Pervaded by this Energy he fuses into
the supreme godhead. The NT, very Tantric in that, adds that this point once
reached, the Energy engenders amrta which, as in the first case, will spread in the
body of the adept who becomes ˚immortal: he is mrtyujit, ‘conqueror of death’.
Here too mental representations act on the physical ˚ body. Not entirely surpris-
ingly, one might say, since the human creature is inseparably body and mind and
since he ‘s’existe’ en s’imaginant’ (he exists imagining himself). This squares,
110 Body and mantra: mantras in the human body
too, with the Tantric conception of liberation as being less a renunciation to the
world than a supernatural (‘magical’) control and domination of oneself and of
the universe. That such a goal can be reached by the presence and circulation of
mantras in the body is also very Tantric – and even, from many points of view,
typically Indian.
9 The oral and the written
Mantra and mantraśāstra1

In the theoretical approach as well as in the practice of mantras in Hinduism,

the premium has always been (and still is) placed on the power of the oral word:
Revelation, there, is not Scripture, but śruti, ‘what has been heard’ (even though
the first mantras were not heard, but ‘seen’ by the Vedic rśis), a premium overem-
phasised in Tantra. Writing, however, is also sometimes ˚ used, and the literature
on the subject, the so-called mantraśāstra, the ‘teaching of the mantras’, is made
up of a vast quantity of written texts. It thus appears that one writes on what is
oral – which itself is sometimes transposed in writing. This written as well as oral
ensemble is, of course, in Sanskrit. The oral literature we shall consider here is a
part of the learned, lettered, not of the popular, domain, even though the learned
(oral or written) sometimes ‘debases’ into the popular. The question of the rela-
tionship of the oral with the written – secondarily of the learned with the popular
– insofar as it appears in the case of mantras, is what we shall try to see here.

A purely oral, spoken, ‘speaking’, word …

By their very nature, which is the nature of the Word, of vāc, mantras are word,
utterance, voice. Vāc, however, since Vedic times, is also and foremost the Abso-
lute, the all-powerful, the brahman: brahma vai vāc, says the Aitareya Brahmana,
6.21.1. Hence their power, their efficacy. Hence, too, their immateriality. Oral,
uttered or unuttered, the word is ‘speaking’ or ‘spoken’ (vāc vadantī). Being made
up in essence of the very highest plane of the word, mantras are therefore on the
side of the Absolute: they are both word and the Absolute. There is an often-quoted
saying in Tantric Śaiva texts: all mantras are made up of phonemes (mantrāh.
varn.ātmakāh. sarve), but the phonemes are of the nature of Śiva (śivātmakāh.).
The mantras are thus divine, and are therefore nothing to do with writing, which
is purely human.
Being the vācakas of deities (or of other supernatural entities), that is, what
‘expresses’, or ‘says’ them, bringing them thus to existence, being the effica-
cious essence of these entities, mantras are powers.2 They have their own power
or energy (mantravīrya or mantraśakti). They are efficacious utterances, able to
effect various actions, to produce results, provided they are used according to pre-
scription and by authorized persons: this is what they do when used in ritual.
112 The oral and the written
All this is always done by word of mouth, by the spoken word, audibly or
mentally uttered. The uttering of a mantra can be so low as to be inaudible, or else
purely mental – the mental utterance being in fact the highest one, as was already
stated in the Laws of Manu, 2.85. This utterance is a necessity: differing in this
from the written word, a mantra always implies the existence of a speaker. It is
never cut off from its production, from its origin, as is the case of the written word.
A mantra may, of course, be deemed to act by itself without any human interven-
tion, but this it does only insofar as it has been first uttered, thrust towards its goal
by somebody being present. It cannot, like a piece of writing, become active after
having remained dormant between the pages of a book, or the folios of a manu-
script. It is necessarily associated not only with an enunciation but also (in spite of
Frits Staal!) with an intention to enunciate. True, there are cases, as we shall see,
when the mantra, cut off from its utterer, is supposed to act in writing; but these
cases are either exceptional or are part of popular, ‘magical’ practices.
Written words, in fact, are elements of language. But mantras, as is everywhere
theoretically admitted, are not part of language. They are made up of phonemes or
syllables (varn.a, aks.ara), sometimes of words – of the phonetic material of speech,
that is; but they are not meant for communication. They do not depend from, nor
.
are they conditioned by, any outside factor (they are nirākanks.a, ‘autonomous’).
Above all, they are not ruled by the conventions (sam . keta) which govern the world
of speech as a means of communication. Being asām . ketika, ‘non-conventional’,
the mantra is a form of word existing on a primary, divine plane, prior to and above
language. It is not turned towards the outside world, but towards the divine to
which it may lead the user, whilst being able also (and mainly) to act in the world.
This non-linguistic character of mantras is not due to their being often made up of
meaningless syllables for, even when they are made up of ‘normal’ Sanskrit words
or sentences, mantras are not speech. This of course is a metaphysical or theologi-
cal principle. It is, however, justified if one considers how mantras work: their
efficacy is not due to what they ‘say’, but to how they say – to their actual word-
ing. It rests on their phonetic tenor (hence the necessity of an absolutely perfect
enunciation). This efficacy rests on the substance of the expression (to speak like
Hjelmslev), on the phonetic structure, on the sounds being somehow emitted, but
not on any literal meaning (supposing there is one).
The inseparability of the mantra and of its utterer is underlined in many theo-
retical works and by a number of mantric practices. It appears also in the often-
proclaimed link between mantra and prān.a, the vital breath linked to the Self,
which is also one of the aspects of the cosmic energy (according to L. Renou,
prān.a is ‘the vitalist aspect of the ātman’). In the tantras and other Tantric texts,
the utterance of a mantra is usually named uccāra/uccaran.a: an enunciation which
is an upward movement (ut-√CAR, to go upwards, to ascend) of the vital breath
together with the mantra. Uccāra is in fact the ascending thrust of the phonic
energy of the mantra. It is evidently not a respiratory breath. The term notes espe-
cially the ascent of the kun.d.alinī caused by the upward moving conjunction, in
the sus.umnā, of the prāna and apāna vital breaths: what happens there is, above
all, the movement of the energy which goes upwards in the yogin’s ‘body’, but
 The oral and the written 113
even in such a case a mantra is deemed to be uttered. Even when there is no vocal
utterance, the mantra is nevertheless present and active in the consciousness and in
the body (or more exactly in the image of the body, in the ‘lived’ or ‘experienced’
body) of the adept. One might describe this by saying that the mantra is ‘uttering
itself’ in the consciousness of the adept whilst spreading out like a fluid in his
body.3 The uccāra of a mantra, Abhinavagupta says when describing this proc-
ess, is prān.acidātma (TĀ 5.128): it is both ‘breath’ and consciousness. There is a
conjunction of the vital and spiritual energy of the yogin and of the energy of the
mantra. All this, evidently, cannot in any way be connected with writing.
We may remark here that all this is in complete conformity with the Indian
traditional view according to which things of importance – especially when con-
nected with religion – must not be dealt with in writing. The rule is as old as the
Veda, which, as is well known, was transmitted orally during two millennia and
is still being taught and used in ritual orally and with the most minute phonetic
precision. T.he superiority of the silent over the audible utterance of mantras, also,
goes back to Vedic origins.
In the same way, mantras – those at least received in initiation to be used for
spiritual or magical practices – can only be transmitted orally, by word of mouth,
by the initiating master, the instruction going, according to the Sanskrit expres-
sion, ‘from ear to ear’ (karn.āt karn.opadeśena), for this only guarantees the secrecy
of the transmission. Mantras are of divine origin, they were originally revealed to
a first master (ācārya, guru) by the deity and then transmitted, always by word of
mouth, through a succession of initiated masters down to the present times. Thus,
only those who possess the needed qualification and authority (adhikāra) may
transmit or receive, orally and according to precise rules, these secret and power-
ful sacred formulas. Many texts say that mantras that have not been received regu-
larly (or, still worse, found in a written text!) have no value. They are ‘fruitless’,
ineffective. They may even be dangerous.
It is interesting to note that these strict mandatory rules on the orality of mantras
have survived for centuries in a social milieu in which a literature in Sanskrit was
being produced (and there are many other languages in India) which is one of the
largest in the world.

of which writing can only be a reflection or a comment …

Writing, in principle, appears as always devalued. It is merely a practical tool, an
artefact without any value of its own. For Indians, as for Rousseau, ‘writing is a
mere supplement of the spoken word (‘l’écriture ne sert que de supplément à la
parole’), a ‘dangerous supplement’ (un dangereux supplément) even, since by its
availability, by the fact that it extends the field of communication by spoken word
or gesture, it wrecks the circle of direct interpersonal communication. By writing,
one moves (down) from the qualitative, where the spoken word goes from the
deity to the spiritual master, then to the disciple, to eventually turn back towards
the divine through yogic practices or ritual action using mantras, to the quantita-
tive where the word circulates freely in a homogeneous milieu where everybody
114 The oral and the written
can receive it, but where it has by this very fact lost all its power: when uttered it is
mere noise, flatus vocis. Worse: once written, the spoken word becomes merely a
stilted transfer, the visible, concrete, inferior form of a living, immaterial, supreme
reality (as unauthorised copy, it may even get the piratical user into trouble4).
In spite of the, always valid, principle of the strictly phonetic, oral nature of
mantras, their users or inventors (in the two ancient meanings of the word: invent
as well as discover) did not entirely escape the rule of writing since they lived in a
milieu where writing was being used. We will indeed see several mantric practices
where writing is not simply a practical device, but is sometimes linked to the very
nature of mantras: there are therefore legitimate, ‘respectable’ uses of writing not
as a substitute, but as an additive to the oral.5 To this can be added cases where it
can be said that the oral debases into writing, cases we will merely allude to later
on in spite of their important place in practice, the frontier between these two uses
being in fact not always very visible, the ‘inferior’ uses being not merely modern
or contemporary ones (nor purely written ones).
As we noted earlier, there is first the mantraśāstra, that is, all the texts, of all
sorts, which describe, prescribe, or deal in some way with mantras. Thousands
of pages have thus been written during some two millenaries on the purely oral
mantras. What is more, these texts, while ceaselessly proclaiming that mantras are
secret formulas, sometimes give explicitly, in writing, their phonic tenor – which
is perhaps to be explained by the conviction that a mantra discovered in a book
has no power or efficacy.

but which it can nevertheless help to reveal

Very often, nevertheless, such texts reveal only in veiled terms the phonemes that
make up the mantra: using periphrases or conventional terms which only qualified
masters or initiates can understand. They give, not the mantra, but its uddhāra:6
how to produce it or ‘extract’ it, how to make it arise out of the Word. This can
safeguard its secrecy very effectively. In effect, if some conventional terms are
common ones (as mayā for Ī, indu for bindu, etc.), and if there are manuals which
give such meanings – bījakośa, bījanighan..tu, uddhārakośa, etc. – and also com-
pendiums, listings of ‘secret’ terms, these riddles are sometimes very difficult
to solve. Every tradition or initiatory lineage has its own system, carefully kept
secret. The written, in such cases, assists the oral: it guards its secret against the
ignorant, whilst disclosing it to those permitted to know it. Functioning as an
aide-mémoire, it helps to keep up the oral tradition – a role it certainly had from
ancient times.
The uddhāras do not merely use abstract notions or mythico-theological ele-
ments so as to indicate/refer to the letters of a mantra. They sometimes refer to
their written form. Writing has in such cases a directly practical role. The uddhāra
also often implies a spreading out (prastāra) of the phonemes of the Sanskrit
alphabet so as to ‘extract’ (ud-DHR) from among them those which are to consti-
˚
tute the mantra. This prastāra consists in spreading out (pra-STR) the forty-nine
to fifty letters of the alphabet, writing them on an appropriate˚ surface, either
 The oral and the written 115
grouping them down by phonetic categories (varga), or distributing them accord-
ing to specific rules on a diagram. The letters once spread out, one ‘extracts,’
chooses that is, those which are to make up the mantra, these letters being often
designated by conventional (secret) names. This use of prastaras underlines the
fact that mantras, as aspects of the word, are born from the mātrkā, the phonic
‘mother’ of the universe of which the Sanskrit alphabet is part, from ˚ which they
are extracted. It shows both the subordination of the mantras – made up of letters
– to the totality of the Word, the Sanskrit alphabet (or the supreme deity insofar
as it is made up of Word, alphabetical), and the fact that mantras draw their power
and efficacy from that alphabetical totality.
The spreading out of the mātrkā is also made use of differently, by using secret,
˚
initiatory prastāras, named gahvaras, a term which means cavern, hiding place,
secret. These are theoretically secret diagrams, of different shapes, where the let-
ters of the Sanskrit alphabet are placed in a conventional order known only by ini-
tiates. They are thus ‘caverns’ where the secret formulas, the mantras are hidden.7
The user of a tantra that prescribes the use of a certain gahvara, knowing how to
draw it and the places where the letters are to be written, it suffices that the places
where they are written and the order in which the letters are mentioned to enable
him to construct the mantra. The design of such diagrams, the placing of the letters
and their ‘extraction’, are all codified, sometimes complex ritual operations. This
is not surprising since what is being done there is not merely drawing a diagram
and writing letters on it: it is invoking and manipulating the divine power of the
word so as to gather it up in a highly effective formula, a mantra, the power of
which the adept will then master through the ensemble of rites and conducts of the
puraścaran.a or mantrasādhanā,8 and will then use – in principle orally, perhaps
also by placing it on a diagram, a man.d.ala – to some spiritual or mundane end. The
mantra thus produced can also be the mantra given to the disciple in the initiation
(dīks.ā) ritual. It is interesting to see here written letters being used to make appear
ritually a formula which in principle must remain purely oral. There are other
ritual procedures of that type used to verify if the mantra is well adapted to its user
and to the result he aims at. This checking, named mantravicāra or mantraparīks.
ā, is done using diagrams of different sorts.9

or even sometimes to show

One may perhaps be tempted to consider such procedures as rather external to
mantras: they are preliminary to their use, with no influence on their nature or
functioning. There are, however, also cases where it is in their very use, even in
some measure in their essential nature, that mantras appear in a written form. Or,
more, they may appear in a form that the user must imagine as being in writing.
There are finally cases when mantras can be made use of only when not simply
placed by nyāsa, but inscribed in a diagram.
In this domain, the oldest proven instances seem to be mainly Buddhist ones.
They are numerous in Mahāyāna Buddhism, especially in Tibet where Sanskrit
mantras are generally used. We may wonder whether this is not due in part to the
116 The oral and the written
influence of China, where the written sign, not the word, is essential. In effect, the
best known and most characteristic Buddhist use of bījas, Sanskrit syllables, in a
written form, is the Sino-Japanese (mostly Japanese, in fact, in the Shingon sect)
of the siddham writing. This lies outside the Indian domain; the case is neverthe-
less worth mentioning here since, in some of the Tibetan traditions, the recitation
and visualisation of the mantra are inseparable: there is a coalescence of the seen,
meditated form and of the sound uttered and perceived inside, not merely the use
of a phonic element in an outer visual frame.
Of all this, Hinduism also shows many instances, some being very ancient.
One of the most evident is the visualization of bījas or of phonemes prescribed
in the kun.d.alinīyoga, where the yogin is to see with his mind’s eye, as present in
the cakras of his yogic body,10 Sanskrit syllables or bījamantras written on the
leaves or on the pericarp of the ‘lotuses’ which constitute the cakras. The sound,
the oral utterance, of these aks.aras or bījas, is to be imagined, felt even, by the
yogin as present in these centres. The kun.d.alinī, too, as we have seen, when going
upwards, may be identified with the upward thrust (uccāra) of a mantra. In some
cases (for instance, in the japa of the śrīvidyā described in the third chapter of the
Yoginīhrdaya), the adept imagines the subtle phonic vibration (the nāda) of the
mantra˚ as ascending from the mūlādhāra to the dvādaśānta, the centre twelve
finger-breadths above the head. In such a case, what is experienced by the adept
is the phonic, oral/aural aspect of the process. It seems, however, that in some
texts, the bījas or aks.aras/varn.as imagined as present in the visual structure of the
kund.alinī are to be visualised in their written form, on which it is probably easier
to concentrate one’s attention than on a sound, by nature more evanescent.
There is, on the other hand, another mantric practice, where mantras or syllables
are necessarily to be written: that of nyāsa. This consists in placing ritually (with a
mudrā) on the body or on a support, a mantra, bīja or phoneme which is simulta-
neously to be mentally visualised (dhyayet, say the texts, dhyāna being here used
in the Tantric sense of visualisation11), such visualisations being usually of their
written form. (Admittedly it can also be the form of the deity of the mantra, but
this is not suggested by the texts I have seen.) Some ritual treatises, sometimes
explicitly, say that one is to visualise clearly the intensely shining letters of the
mantra. Chapter 293 of the Agni-Purān.a prescribes the lipinyāsa – the placing of
written letters (lipi) – of the fifty letters of the Sanskrit alphabet, followed by the
nyāsa of eight Lipīśvaras, ‘Lords of letters’, which are to be written (likhyate). The
Purān.a does not say how these deities are to be drawn (written?).

One has also speculated on the written aspect of these oral letters
Having said that in principle mantras cannot be but oral since their nature is that
of the word, of the primal vāc, identical with the formless Absolute, the brahman-
sound (śabdabrahman) as it is sometimes called, we must now look at cases where
writing comes in, not merely in the ritual uses or practices of mantras, but at their
highest level, on the level of the deity. In such cases writing appears as linked to
the very nature of mantras.
 The oral and the written 117
Let us note first that in some texts (the Agnipurān.a, 293.51, or the Brahmān.d.a-
purān.a, for instance) there exists a Lipidevī or Lipibhairavī: a ‘writing-goddess’;
in the Śāradātilaka, 7.8–14, the goddess’s body is made up of letters (lipitanu) and
she is seated under a ‘tree of letters’ (lipitaru). This alphabet-deity is the Mistress
of the Word, Vāgīśvarī (or Devī Vāgīśī). It is in fact an aspect of Sarasvatī whose
attributes she bears (notably the rosary and the book). Her plane, however, is not
the supreme one: she is the goddess on the level where she gives birth to the world.
In the non-dualist system of the Trika, on the other hand, it is the supreme goddess,
Parā, who is paravāk, the supreme Word, a variant of Sarasvatī-Vāgīśī, who bears
these attributes, the alphabet being present here, most subtly, as mātrkā, śabdarāśī
or mālinī, as the supreme Energy or totality of the Word, that is, and ˚ therefore on
a plane very much above the plane of writing. There is, however, a continuity or a
proximity – underlined by the presence in both cases of Vāgīśvarī holding a book
(pustaka) – between the totality of the word and the written Sanskrit alphabet. One
could thus consider that there is here, as it were, a primary presence of writing: a
‘primordial trace’, or a ‘primordial writing’ (une écriture première), to use Der-
rida’s words; it being, however, understood that, in India as elsewhere, one may
postulate the precedence of the oral over the written. It is only insofar as, in the
course of time, one has speculated on the word and that the mantraśāstra has suc-
ceeded the mantra, that graphical elements appeared to reinforce the empire and
power of the oral, to sustain, explain and describe its supremacy.
If we turn now towards the mantraśāstra, we will find striking instances – in
fact, very well-known ones – of the intervention of the graphical in mantras or, to
say it differently, of the interaction on all planes of the visual and the phonetic.
The most obvious example is probably that essential element of all bījamantras,
the anusvāra, the nasalisation and prolongation of a syllable which one names
bindu (which is to say, drop, dot) because of its written form as a dot placed
over the nasalised phoneme. (It is sometimes called indu, ‘moon’: it is a luminous
circle.) In mantras, it ends such bījas as OM . , HRĪM
. , LAM
. , etc. Now, this written
form is the basis of all the speculations on the bindu conceived as a ‘drop’ of con-
centrated phonic energy, as the point where the energy of the word gathers upon
itself before dividing itself and spreading out to manifest outwardly its power and
to manifest the universe.
Phonetically, the anusvāra, as a resonance which progressively fades out or
which diminishes before a consonant, does not offer much food to the imagina-
tion. It is clearly and evidently its visual aspect which plays a role. Only the visual
perception based on the written form can explain that the bindu at the end of a
mantra could be generally described as the part which concentrates its dynamism,
or (to use Ks.emarāja’s words in his Śivasūtravimarśinī) which ‘flies like the arrow
thrust forward by the tautened bow’. Among the elements usually enumerated that
characterise a mantra (the rs.i, etc.11), the bindu represents usually the kīlaka, the
˚
‘point’ or wedge that will pierce the target aimed at by the mantra.
In the traditional order of the letters of the Sanskrit alphabet (the varn.a-
samāmnāya), the anusvāra is followed by the visarga, the sixteenth ‘vowel’,
which, in Indian grammar, designates a release, an escape of breath after a vowel
118 The oral and the written
at the end of a word, its written form being two dots (:). It is thus graphically a
division in two of the bindu, whereas – owing to its name and its place – it emits or
produces (vi-SRJ) the thirty-four consonants which follow it and, by this very fact,
˚ totality of the cosmic manifestation. This written form, a divi-
gives birth to the
sion in two of the bindu, also gives rise to speculations on the birth of the word.
We may note here that very few mantras end with a visarga. I need not expatiate
here on bindu and visarga, but refer readers to studies on the subject (for instance,
Padoux 1990, chap. 5, pp. 272–286).
We may, on the other hand, apropos of the oral/written interaction as seen in the
mantraśāstra, mention two other instances of speculations based on the written
form of letters in the devanāgarī script.
First, on the drawing of the letter A, the first vowel, which is, in the Tantric
metaphysical speculations on the word, the Absolute, the Unsurpassed (anuttara),
the original phonic energy underlying all the other phonemes – therefore above
all limits and all form. But, in emanative systems (as are Tantric traditions), it is
necessary for the supreme deity, the Absolute, to hold in itself the seed or para-
digm of all that it will emit. The energies that will create and sustain the cosmos
are therefore all inchoately in the Absolute. Abhinavagupta elaborates upon this
idea in the third chapter of his Tantrāloka (where speculations on the written form
of phonemes are in Jayaratha’s commentary on TĀ 3.67). Such ideas are also to
be found in another, older, text, the Vātūlanāthasūtra. The lines of the written
form of the letter A are described as representing the four basic energies of Śiva:
Raudrī is the head (-), Jyes.t.hā, the weapon (3), Ambikā, the arm (/), and Vāmā
(who ‘vomits’, that is manifests, the world) is the mouth (|). It is indeed not in its
written form that A is the Unsurpassed with all its energies. But we see here the
symbolic value given to the written form of a letter being used in support of the
metaphysical interpretation of a phoneme. Another Sanskrit phoneme whose writ-
ten form has given rise to the same speculations is E, whose written form is more
or less triangular.13

Then there are also diagrams and amulets

One finds nearly always, in the centre of the yantras or cakras of Tantric female
deities, a triangle point downwards, or two inverted superposed triangles sym-
bolising the (sexual) union of Śiva and Śakti. There are also sometimes, in these
diagrams, several inverted superimposed triangles (as, for instance, in the srīcakra
of goddess Tripurasundarī) replicating and reflecting the first couple, symbolising
thus the creative dynamism of the goddess united with the god. In the centre of the
central triangle of the śrīcakra, there is also a bindu which is the goddess herself,
a spot where both the oral and the written focus since the symbolism of the bindu
is both geometrical-visual and phonetic-oral.
Admittedly, such representations are, strictly speaking, outside the domain of
the oral/written relationship in mantras. It is, however, a neighbouring domain,
which intersects and oversteps this relationship, and is an important one; for what
is in question here is not only the relation of the phonetic to the written, but also,
 The oral and the written 119
more largely, the relation of the oral to the visual. For this, indeed, is the context
within which is set the more general problem of the relationship of mantras to
ritual diagrams (man.d.ala, yantra, cakra), which is the problem of the relationship
between the phonetic, temporal (and fugacious, evanescent) and the spatial (and
lasting). There are, of course, mantras, bījas, or phonemes in all ritual diagrams
which often are no more than a means to display the phonetic elements for putting
them into action: the word, there, becomes active as being spatially placed, or, if
one prefers, space is ritually made use of only when qualified, infused with power
by the word. In such cases, the functioning of the yantra and of the mantra appears
as being totally linked. There are also cases when they function in full parallel,
cakra and mantra appearing to the yogin as the two faces of the same divine real-
ity, as two interchangeable modes of access to this reality. This is what takes place
with the śrīvidyā and the śrīcakra in the Tripurā tradition. It also happens (for
instance, for the śrīcakra) that one goes on directly from one regime to the other:
from the spatial to the phonetic, this being all the more easy when, as sometimes
happens, the man.d.ala is not concretely marked out, being only visualised in imag-
ination. This is the case when the yogin is to visualise diagrams in his body, in the
cakras, for instance, visualised with their design and the deities that abide there
together with mantras or bījas and phonemes.
But there are also what we may call diagrammatic mantras whose utterance
and functioning – their use, that is, in mantrayoga – is constituted partly of visual,
spatial elements (drawings and written letters), these being either concrete or men-
tally visualised.
There exists therefore a common zone where correspondences, isomorphisms,
between visual and audible are played out. This is in fact all the less surprising
in a civilisation where the Word, the oral, is placed at the origin of everything
that exists, its supremacy being always extolled; but a civilisation for which, at
the same time, the direct visual (saks.āt) perception is the best possible proof of
the existence of something; where, too, in the common religious practice, the
view, the vision of the image of the deity by the devotee – the darśan – together
with the deity’s gaze going towards the devotee, are of the utmost importance.
We would be tempted to say here that we have, in the Indian case, the (apparent)
paradox of the omnipotence of the Word, which is invisible, being proclaimed,
extolled by ‘visual seers’. This Word was ‘seen’, in the beginning, by the
Vedic poet-seers, the rs.is: ‘the sacred Formula (brahman) that was first
born in the East, the seer˚ discovered it from the shining edge [of the world]’
(Atharvaveda 4.1).
Many instances of the presence of visual elements in the structure and uses of
mantras – in a, by nature, phonetic (oral/aural) process – could easily be adduced.
Typical is thus the fact that when enouncing the subtle phonic levels (kalās) which
prolong the utterance (uccāra) of OM . and are above all actual physical utter-
ances, since they are infinitely subtle and eventually disappear completely in utter
silence, the practitioner is to ‘see’ these levels as symbolic geometrical figures.
One could also quote the case of the kāmakalā, a symbolic representation of the
sexual union of Śiva and Śakti, which is something both graphically visible and
120 The oral and the written
phonetically invisible since it is a diagram with an inscribed letter (ĪM . ), but also
a bīja to be uttered.14
But we cannot dilate here on this sort of learned speculations and practices,
intermixing oral and written, which are found in the Sanskrit texts of mantraśāstra.
It would be interesting to know their origins and when they began, but on this we
cannot do more than suggest possibilities. The earliest documents on the subject
are in fact Buddhist and Chinese.15 Speculations on the Word, as well as magic,
go back to Vedic times. It is therefore not surprising that even the magical uses
of the word should have developed in a literate milieu. We may be permitted to
believe that as soon as Indians began to write – at any rate very early – they used
mantras as, shall we say, magic formulas, using for this purpose writing or lines
of some sort, or by inscribing them in ritually drawn diagrams. Therefore, rather
than a passage from the oral to the written attesting to a modern evolution, or a
passage from the learned Sanskrit to the popular domain, there would have been
from an early time, beside a strictly oral domain, the highest and most respect-
able one, a lower domain, nearer to magic (by which I mean the manipulation
of sacred power to notably mundane ends), where oral and written co-exist but
where the written plays often an autonomous or quasi-autonomous role. One must,
of course, especially in modern times, make a distinction among the written ele-
ments: a devotedly preserved manuscript has not the same status as a printed text.
The intrinsic value of the book matters little, in fact; what is important is to find in
it the efficient formula and the way to use it.
If in India, today, one obtains mantras or amulets from specialists: tāntrikas,
mantiravatis,16 pūjāris sometimes, or other magicians, these make them (suppos-
edly) ritually, using rites where mantras are necessarily uttered. The oral mantra is
somehow embodied in the written formula, thus investing it with power. Even for
amulets, charms or yantras described in books, it is often specified (see, Chawdri,
1984) that, to put these magical contrivances into action, rites including the utter-
ance of mantras and sometimes long japas are necessary, which shows the con-
tinuous power of the oral utterance. Nothing, in fact, is able to replace the enuncia-
tion, with the mental and bodily participation it implies: the presence in the body,
one might say, of the word one knows ‘by heart’, or and perhaps better, in Sanskrit,
which is kan..thasthā :‘staying in the throat’, ready to be vocally uttered.

And the circle of users

But who are those who use these oral or written mantras? In what context? Of
what kind is the relationship between the user and the mantra he uses? Being in
Sanskrit, mantras, in principle, may be used only by Brahmins or at most by other
twice-born groups (provided they are not Vedic mantras). This circle of users,
however, became larger as a consequence of the ‘mantrisation’ of Hindu traditions
in the course of centuries: any initiated person, of whatever caste or sex, can, if the
prescribed rules are respected, receive a mantra he/she will be able to make use
of if he/she conforms to the rules. The ācārya, sole person permitted to transmit
a mantra, is still usually a Brahmin; but the receiver who uses the mantra is not
 The oral and the written 121
necessarily a Brahmin. More, as bhakti developed at the expense of ritual rules,
any devotee could worship his chosen deity using the deity’s mantra. It is not only
Brahmins, especially nowadays, who confer mantras on adepts or devotees – and
their mantras are not less effective for that! The circle of users permitted to utter
mantras tends thus to extend indefinitely till the most extreme (the lowest?) level,
reached with the uses of mantras written on amulets or found in books.
We may note, too, concerning the relationship of the user to the mantra, that
Sanskrit having ceased for more than a millennium to be the mother tongue of
any high-caste Indian, to use a mantra, which is in Sanskrit, is for the user both
to manipulate the power (intensified in the mantra) proper to that strange divine
tongue. and to participate in the particular, learned, world of the Sanskrit cul-
ture. It is a peculiar relationship: to an unusual, both learned and timeless, form
of speech. Sanskrit is the liturgical language of Hinduism. It is not the common
language of the Hindu religion, still less that of the believers, who are alien to the
Sanskrit world. This relationship of the user of a (Sanskrit) mantra remains the
same whether or not he understands the formula he utters. This does not matter, in
fact, since a mantra acts by its mere utterance, by its phonetic content, its elocu-
tionary value, not by its meaning – supposing it has one.17
Whether or not the use of a mantra resorts to writing, it remains an aspect of
the word whose power can only be put into action by oral utterance. The power of
the oral is always what underlies and shapes all uses of mantras, be it a concrete
(audible or inaudible) utterance, the inner utterance in the form of the ascending
movement of prān.a, or a purely mental vision of the sounds: one of these forms of
orality is always present (even if momentarily) in all mantric practices, even when
apparently written. This, I believe, is what appears through any careful study of
mantraśāstra.

Bibliography
Agnipurān.a (1966) A prose English translation by M. Nāth Dutt Shastri, Varanasi,
Chowkhamba Sanskrit Series Office.
Alper, H.P. (ed.) (1986) Understanding Mantras, Albany: State University of New York
Press.
Atharva Veda Sam . hitā (1962) trans. by W.D. Whitney, Delhi: Motilal Banarsidass (reprint
of the 1905 edition).
Avalon, A. (trad.) (1953) Kāmakalāvilāsa by Punyānanda, Madras: Ganesh & Co.
Brown, C.M. (1986) ‘Purān.a as scripture: from sound to image of the holy word in the
Hindu tradition’, History of Religion, 26 (1), pp. 68–86.
Brunner, H. (1974a) ‘Le sādhaka, personnage oublié du śivaïsme du sud’, Journal Asiati-
que, 263, pp. 411–443.
Brunner, H. (1974b) ‘Un tantra du nord: le Netra Tantra’, BEFEO, 62, pp. 125–197.
Bühnemann, G. (1992) ‘On puraścaran.a: Kulārn.avatantra’, chap. 15 in T. Goudriaan, ed.,
Ritual and Speculation in Early Tantrism, Albany: State University of New York Press,
pp. 61–106.
Chawdri, L.R. (1991) Practicals of Yantras, New Delhi: Sagar Publications.
Coburn, T.B. (1991) Encountering the Goddess. A Translation of the Devīmāhatmya and a
Study of its Interpretation, Albany: State University of New York Press.
122 The oral and the written
Diehl, C.G. (1956) Instrument and Purpose. Studies on Rites and Ritual in South India,
Lund: Gleerup.
Dusenbery, V.E. (1992) ‘The Word as guru. Sikh scripture and the translation controversy’,
History of Religion, 31 (4), pp. 385–402.
Goudriaan, T. (ed.) (1992) Ritual and Speculation in Early Tantrism, Albany: State Univer-
sity of New York Press.
Graham, W.A. (1986) ‘Scripture’, in M. Eliade, ed., The Encyclopedia of Religion, vol. 13,
New York: Macmillan, pp. 133–145.
Malamoud, Ch. (1981) Cuire le monde. Rite et pensée dans l’Inde ancienne, Paris: La
Découverte.
Malamoud, Ch. (2002) ‘Noirceur de l’écriture’ in Le jumeau solaire, Paris: Seuil,
pp. 127–146.
Ong, W.J. (1982) Orality and Literacy. The Technologizing of the Word, London:
Methuen.
Padoux, A. (ed.) (1986) Mantras et diagrammes rituels dans l’hindouisme, Paris: Editions
du CNRS.
Padoux, A. (1990) Vāc, The Concept of the Word in Selected Hindu Tantras, Albany: State
University of New York Press.
Padoux, A. (1993) Le cœur de la Yoginī – Yoginīh daya, avec le commentaire Dīpikā
d’Amrtānanda, texte sanskrit traduit et annoté, Paris: Collège de France.
Renou, ˚L. (1960) ‘Le destin du Veda dans l’Inde’, Etudes Védiques et Paninéennes, VI,
Paris: E. de Boccard.
Notes

1 Introduction
1 Tantra is esoteric since its adepts, tāntrikas, are supposed to have been initiated. The
sectarian initiation, dīks.ā, being held as a first step towards liberation. All Tantric
adepts or believers, however, are not initiated, But with or without initiation Tantric
teaching is always described as secret. On the role of secrecy, see studies by Hugh
Urban, notably The Economics of Ecstasy. Tantra, Secrecy and Power in Colonial
Bengal (New York: Oxford University Press, 2001).
2 In some cases, as we shall see (for mantroddhāra, for instance – see chapter 1,
pp. 12ff.), mantras are to be written, or to be visualised mentally in their written form,
but these are exceptions to the general rule of orality.
3 A most common form of those mantras – a form current in Tantric practice too – is for
instance OM . HAUM . śivāya namah..
For an overview of mantras in Hinduism, see chapter 22 in G. Flood, ed., The Black-
well Companion to Hinduism (Oxford: Blackwell, 2003). On mantras generally, the
only comprehensive study remains to this day H. Alper, ed. Understanding Mantras
(Albany: SUNY Press, 1989).
4 The word mantra is of the masculine grammatical gender.
5 On these exclamations – or inflections, as they are technically called – see the entry
jāti of the Tāntrikābhidhānakośa (TAK), vol. 2.
6 On these, see A. Padoux, Vāc. The Concept of the Word in Selected Hindu Tantras
(Albany, SUNY Press, 1990, repr. Delhi, Sri Satguru Publications, 1992); this, how-
ever, being a translation of a 1964/1975 French thesis, reflects the state of knowledge
of the subject at a time when many tantras now being studied were unknown or not
available.
7 These alphabets are also forms of Bhairava or of the goddess: they are alphabet-
deities.
8 ‘The mātrkā, the unknown universal Mother who gives birth to the cosmos through
˚
the [Sanskrit] phonemes from A to KS.A’, to quote from Ks.emarāja’s commentary on
Śiva Sutrā 1.4. See here, chapter 4, on nyāsa: the ritual placing of the alphabet is called
mātrkānyāsa. The conception of the goddess Mātrkā and of her role is not the same in
˚
all tantras. ˚
This subject is being studied by Judit Törzsök: paper read in Pondicherry in
2009.
9 For instance, Ajtāgama 1.13–14, the tantrāvatāra chapter, on the ‘descent’ of the
tantra, not a chapter on mantras.
10 Not so surprising, however, if one refers to the Vedic stobhas, such as ī hā bu hvā.
11 pārameśvare’pi avyaktadhvaner mukhyatayaiva prāyaśo mantratvam nirūpitam.
This work has been edited with an Italian translation by R. Gnoli (Roma, IsMEO,
1985). There also exists a (less scientific, and difficult to use) English translation by
124 Notes
Jaidev Singh (Delhi, Motilal Banarsidass, 1988) – a new edition is being prepared by
Bettina Bäumer).
12 The utterance of the mantra, that is. On uccāra, see A. Padoux, Vāc The Concept of the
Word in Selected Hindu Tantras, chapter 7; or the entry uccāra of the TĀK, vol. 1. See
too, here, chapter 3 on japa. In Tantric usage, the uccāra of a mantra is often a yogic
practice where the mantra moves upward together with the kun.d.alinī – such is the case
of the japa of the śrīvidyā described here, pp. 40–41. On this subject, see A. Padoux,
‘Corps et mantra. la présence des mantras dans le corps’, in O. Botto, et al., eds., Du
corps humain au carrefour de plusieurs savoirs en Inde (Bucarest-Paris, de Boccard,
2004, Studia Asiatica vols. IV–V), pp. 563–578.
13 This was already stated in the Laws of Manu (2.85) and remains valid to this day. See
chapter 3, on japa, p. 25.
14 It may be interesting to quote here the Tamil poet Tirumular’s (c. seventh century)
definition of mantra as ‘perfect concentration of the mind on anything’.
15 In the dualistic Śaivasiddhānta, mantras are often called an.u, that is, atom or rather
individual soul. Mantras appear also as discrete, different, entities in the system of
pramātr.
16 Such is˚ the case, for instance, of the japa of the śrīvidyā which takes place at the
end of the cult of Tripurasundarī described in chapter 3 of the Yoginīhrdaya (YH)
where the śrīvidyā, which is the mantric aspect of the goddess, ascends ˚ along the
sus.umnā in the body of the worshipper, or of the ‘subtle’ meditation of the
Netramantra described in chapter 7 (pp. 15–51), both briefly described here,
chapter 3.
17 mantrā varn.abhattārakā laukikapārameśvarādirūpā mananatrān.arūpā vikalpasam .
vinmayyāh. (PTV, p. 243).
18 Nor were they the same in the śaivāgamas and in the earlier Pāñcarātra sam . hitās, and
still less in the later ones.
19 These elements were taken over from the indexes of Vedic hymns (anukraman.ikā).
They are completely arbitrary.
20 It is fair to say, however, that mudrās, as parts of a ritual or used otherwise, are some-
times held to be by themselves effective. This is the case in many rituals where, for
instance, a deity is invoked and placed on a ritual support with a mudrā. Such is also
the case of the mudrās described in chapter 1 of the Yoginīhrdaya (1.56–71). These
mudrās are also aspects of the power of the goddess, and deities ˚ worshipped during
the pūjā. On this, see chapter 4.
21 Though japa, especially when made using the rosary, includes rites – see here
chapter 5.
22 This is done by the astra (weapon or arrow) mantra PHAT. or the kavaca (cuirass)
mantra.
23 In the prān.apratis..thā ritual.
24 They may therefore be described as performative utterances or as illocutionary forces,
in the sense of J.L. Austin.
25 Then come, in decreasing importance and value, the vijñānākalas, the pralayākalas
and finally the sakalas. On this system, see H. Brunner’s ‘Mantras et mantras dans les
tantras śivaïtes’, in R. Torella, ed., Le parole e i marmi, Studi in onore di Raniero Gnoli
(Roma, IsIAO, 2001), pp. 185ff., or Vāc, p. 104 n.
26 Which was the case of vāc in the Veda: ‘the Word that speaks and is the harmonious
ruler of the gods’ (RV 8.100,10).
27 My position differs˚ in this from Hélène Brunner’s in her 2001 study referred to in
note 25, where she distinguishes clearly betwen ‘Mantras’ as powers and ‘mantras’
as sound units, this distinction corresponding to two different uses of mantras. The
distinction of uses, I believe, does not necessarily imply a distinction of nature.
28 The mūlamantra of a deity is the deity itself. The mūrtimantra evokes the deity’s form;
the an.gamantras of a deity symbolise and evoke the an.gas, the ‘limbs’, that is, the
 Notes 125
constitutive parts or qualities (gun.a) of that deity. The ‘weapons’ (āyudha) of a deity
have also mantras used to evoke them or to place them on the deity’s image.
29 See the uccāra of OM . , SAUH. or of the sam
. hārabīja KHPREM . /HSHPHREM . described
in Vāc, pp. 401–426.
30 dhyāna in Tantric context is the word for the mental visualisation of a deity – in addi-
tion, of course, to its usual meaning of meditation.
31 On this subject, see chapter 2.
32 Two such well-known manuals of iconography are Mahīdhara’s Mantramahodadhi
(sixteenth century) and the somewhat later Mantramahārn.ava. The former has been
translated with introduction and notes by G. Bühnemann, The Iconography of Hindu
Tantric Deities, vol. 1 (Groningen: Egbert Forsten, 2000). The second volume deals
with the pantheons of the Prapañcasāra and of the Śāradātilaka (ibid., 2001).
33 This ritual is described in several texts, notably in chapter 15 of the Kulārn.avatantra
(KT), on which see G. Bühnemann’s study in T. Goudriaan, ed., Ritual and Specula-
tion in Early Tantrism (Albany: SUNY Press, 1992), pp. 61–106.
34 The rituals thus performed are therefore classified among the kāmya, ‘desiderative’
rites.
35 On these see T. Goudriaan, Māyā Divine and Human (Delhi: Motilal Banarsidass,
1978), pp. 251–412.
36 This subject is approached more fully in chapter 7, on mantras, of Vāc, quoted in note
12.
37 Or should we say ‘mantra-deity’? Or ‘mantra qua deity’?
38 tasya yā sattā aśes.aviśvābhedamayapūrn.āham . vimarśatmā sphurattā sā mantrān.ām .
rahasyam upanis.at (ŚSV 2.3, comm., p. 50 of the KSTS edition, Srinagar, 1911).
39 In this perspective, mantras, made of phonemes ‘extracted’ from the Sanskrit alphabet,
are also (or at least one of them can be) at the source of the alphabet.
40 tadākramya balam . mantrāh. sarvajñabalaśalināh./
pravartante ‘dhikārāya karan.ām īva dehinām //26//
tatraiva sam. pralīyante śāntarūpā nirañjanāh. /
sahārādhakacittena tenaite śivadharminah. // 27//
41 This, the Saiddhāntika authors would see differently since for them the mantras are
powers that act by themselves. Their position is explained by H. Brunner in the article
quoted in note 25.
42 mahāhradānusam . dhānān mantravīryānubhavah. /.
43 See infra, p. 58 for nyāsa or p. 46 for japa.
44 New York: Peter Lang, 1990/93; Indian edition, Delhi, Motilal Banarsidass,
1996.
45 Vedic recitation, too, was governed by the rule of the absolute respect of the phonetic
content of the recited mantra. The Vedic reciter concentrates on the sounds he emits,
not on the meaning of the words he utters: this is still the case today. The Tantrika is
heir to the Vedic rs.i.
46 ˚
Such views are briefl y reviewed in A. Padoux, L’énergie de la parole, pp. 178–190
(Paris, Le Soleil Noir, 1980, reprint Fata Morgana, 1994).
47 On mantras in general I would refer the reader to Understanding Mantras, quoted
in note 3 and, perhaps, more specially, to the article I wrote as a (far from definitive)
conclusion to that book: ‘Mantras, what are they?’ (pp. 295–318).

2 The extraction and examination of mantras

1 This is a new version of a paper ‘La sélection des mantras’ published in the BEFEO,
LXV, 1978, pp. 65–85.
2 As is often said: mantrāh. varn.ātmakāh. sarve varnāh. sarve śivātmakāh., ‘all mantras are
made of phonemes which are all made up of Śiva’.
126 Notes
3 vidyās are feminine mantras (or female mantradevatās), or mantras of female deities.
On the nature of mantras, see the Introduction.
4 See Padoux, Vāc, chapter 1 on vāc, and chapter 7 on mantras.
5 On the term uddhāra, see the entry on this word in vol. 1 of the Tāntrikābhidhānakośa
(henceforth referred to as TAK), p. 231.
6 In chapter VIII of Teun Goudriaan and Sanjukta Gupta, Hindu Tantric and Śākta
Literature (Leiden, Brill, 1981).
7 A well-known one is Daks.in.āmūrti’s Uddhārakośa, edited by Raghu Vira and Shodo
Taki, Delhi, 1978. Or the Tantrābhidhāna with Vīja-nighantu, edited by Tāranātha
Vidyāratna (‘Tantrik Texts’ vol.1, Calcutta 1913).
8 The reason for this identification, as well as that for M and time, is not clear for me.
9 caturdaśayutam . bhadre tithīśāntasamanvitam//
tritīyam
. brahma suśroni hrdayam . bhairatmanah.:
10 In the varn.asamāmnāya ˚(notably in the case of the phonematic emanation, the varn.
aparāmarśa as expounded in the Tantrāloka (TĀ)), the additional phonetic signs
bindu and visarga are included among the vowels which are thus considered as being
sixteen in number. (I follow here the traditional practice of considering the diphtongs
as vowels. See Padoux, Vāc, pp. 223ff.).
11 tritīyam
. brahma sakārah., says Jayaratha ad TĀ 30–.167 – this because of Bhāgavadgīta
17.23: om . tat sad iti nirdeśo brahman.as trividhah. smrtah., where sad (therefore, here,
SA or S) is quoted as the third of the three phonemes (om ˚ . tat sat) that are the brahman.
The esoteric play on words goes here quite far, but the more obscure, the better! (The
use of words in more than one sense (śles.a) is a current Indian traditional interpreta-
tive practice – on the subject, see for instance Sanderson’s remarkable study, ‘A com-
mentary on the opening verses of the Tantrasāra of Abhinavagupta’, in S. Das and E.
Fürlinger, eds. Sāmarasya, Studies in Indian Arts (New Delhi, D.K. Printworld, 2005)
pp. 89–148.
12 See the articles ācārya, guru and deśika in vols. 1 and 2 of TAK.
13 On this process, see Hélène Brunner’s article, ‘Le sādhaka, personnage oublié du
śivaïsme du sud’ (JAs,1975, pp. 411–443, and Gudrun Buhnemann, ‘On Puraścaran.a:
Kulārn.avatantra, Chapter 15’ in T. Goudriaan ed., Ritual and Speculation in Early
Tantrism (Albany, SUNY Press, 1992), pp. 61–106.
14 Admittedly, alphabet is not the best term to name the ensemble of the 49/50 Sanskrit
phonemes, varn.as. I use it however for simplicity’s sake since it is familiar to us. As
is well known, the Sanskrit alphabet includes 14 vowels (svara) and two additional
phonetic units, 4 semivowels (antah.stha), 25 consonants (vyañjana), and 4 sibilants, a
total of 49 letters to which, in Tantric traditions, is added the consonant group KS.A.
15 The whole alphabet is also sometimes considered as a mantra. There are, too, alpha-
bet-goddesses.
16 Note to his edition of the Mahānirvānatantra, III, 11–12 (p. 33), a passage where the
mantroddhāra is very much simplified, but Avalon’s formula is nevertheless gener-
ally valid. Arthur Avalon was no expert in Sanskrit, but the pandits he worked with
were. His contribution to the Tantric field was important – and remains useful. The
Mahānirvān.atantra is a late (18th c.) fabrication. It includes elements showing British
influences, but it is still worth quoting on some specific points.
17 We shall see later on the technical meaning of this term.
18 varn.asam . ghat.tanāśarīrān.ām . mantrān.ām. saiva bhagavatī vyākhyātarūpā
vidyāśarīrasattā rahasyam . iti pradarśitam/pratyāgamam . ca mātrkāmālinīprastāra-
pūrvakam . mantroddhārakathanasya ayam eva āśayah.. ˚
On mātrkā and mālinī, see Padoux, Vāc, pp. 312–327.
19 A Śaiva˚ text; one of the early (perhaps eighth century) Bhairavatantras. It was com-
mented upon, according to the non-dualist Trika system, by Abhinavagupta’s disciple
Ks.emarāja (the Uddyota). Text and commentary are published in the ‘Kashmir Series
of Texts and Studies’ (Bombay, 1921–1933).
 Notes 127
20 mātrkām . paśunām ajñātām . viśvamātaram . sarvamantratantrajananīm . ādiks.ābtām ...
˚
(SvT-Uddyota, vol. 1, p. 26). See Śivasūtravimarśinī 1.4 (p. 16): tasya ādiksāntarūpā
ajñātā mātā mātrkā viśvajananī: ‘its unknown mother, mātrkā, begetter of the uni-
verse, formed of˚[the phonemes] from A to KS.A’. ˚
21 prastaret, says SvT 1.31b, verbal form explained by Ks.emarāja (ibid., p. 277, l.1) as
an ārs.a form for prastrn.īyāt.
22 ekacittah. is explained ˚in the Uddyota (vol. 1, p. 26) as mantrodayanibhālanapravan.ah.,
intent upon the perception of the production (creation or coming up) of the mantra.
23 The Mahāmātrkā or Mahāmātarah. are in fact usually seven: the so-called Saptamātrkā
– and this is ˚the number mentioned in this passage of the SvT and in Ks.emarāja’s ˚
Uddyota, in spite of the fact that the text actually enumerates not seven but eight god-
desses.
24 The mūrtimantra is the mantra which, in the ritual, serves to invoke the subtle bodily
aspect (mūrti) of the deity being worshipped – or which is, or represents, on a subtle
plane, the body of the deity .
25 ante’syā uddharen mantrān yathākramaniyogatah. // 37//
Comm.: asyā – mātrkāpūjāyāh., yathakramam – āsanamantramūrtimantrādikrame-
˚
na yo niyogah. – tattanmantramūrtimantrādikramena yo niyogah. – tattanmantrāvācya
devatānusam . dhānaparatvam . tena.
26 See for instance Rauravāgama, kriyāpada, 1.1–21.
On the notion of an.ga, I refer to Hélène Brunner’s study, ‘Les membres de Śiva’,
Asiatische Studien/Etudes Asiatiques, XL-2, 1986, pp. 90–132.
27 This is composed usually of milk, curds, ghee, cow urine and dung. Another group of
five offering substances is that called pañcāmrta, which, in ordinary rituals, consists
of five ‘innocuous’ substances such as molasses, ˚ honey, ghee, curds and milk, but,
in Kaula rites, is made up of human fluids and excreta mixed with alcoholic liquor
(vāmāmrta). On these, see A. Sanderson, op. cit., note 63, p. 113.
28 The offi˚ciating person is, if not an ācārya, at least an initiated putraka or sādhaka.
Note that here as in the case of the NT, the adept is to act ‘with devotion’: the manipu-
lation of the sacred does not exclude faith, devotion, nor does it exclude its correlate,
divine grace.
29 In the Pāñcarātra tradition (as in other ones), each of the fifty phonemes – as a pho-
neme and as a deity – has a name or names. A may thus be called aprameya, prathama,
vis.n.u, etc. Śl. 32 to 58 of this chapter enumerates these names. Such lists of names
are given as appendices in the Sanskrit editions of the Jayākhyasam . hitā or of the LT
(see also app. XIII of the English translation of that tantra). See too Daks.in.amūrti’s
Uddhārakośa, or chapter 347 of the Agnipurān.a (ekāks.arābhidhānam.).
30 The LT, chapter 25, śl. 30–37 (p. 86, p. 136–137 of the English translation) expresses
itself in a similar fashion: ‘The phonemes made powerful by the śakti of Laks.mī are
the source of all mantras. They must be intensely worshipped, praised and meditated
upon using their names ..., they participate mutually in each other when used in man-
tras. There is nothing in this movable and immovable world that is not produced by
them. Although the forms of mantras are supreme, divine and eternal, they are yet
conceived of as produced by the phonemes ... Just as one imagines divisions in space
which is one, so, for convenience’s sake, one imagines that there are divisions between
phonemes [though they are indivisible sound-units] ... Having thus clearly realized in
his mind this presence of the phonemes in the mantras, and having worshipped them
on the ground, in the lotus [the diagram] and in the body of the goddess, he should
extract them (mantrān samuddharet)’. The LT, chapters 23–24, describes in details
how to display the mātrkā, the mother of the mantras (mantramātrkā), the divine power
creating and pervading˚ the cosmos, and how to worship her before ˚ extracting the man-
tras. The tantra adds that once the lotus-shaped cakra has been drawn, the adept is to
visualize the mantramātrkā, who is the Self of Vis.n.u as seated on the pericarp of
the lotus, her body being ˚ made of the fifty phonemes. He is to worship her before
128 Notes
extracting the mantras. The Sanatkumārasam . hitā (Indrarātra, chapter 2), gives a simi-
lar description.
31 A good summary of this passage is given by Marion Rastelli, Philosophisch-theolo-
gische Grundanschauungen der Jayākhyāsam . hitā (Vienne, 1999), pp. 126–128.
32 There are fifty-one letters when the vedic l is added after KS.A.
33 LT 24, 1–34, pp. 80–82; pp. 127–130 of the ˚ English translation.
34 In the LT, chapters 23 and 24, there is a very complete and interesting description of
how to display the mātrkā, the mother of mantras (mantramātrkā), the divine energy
pervading the cosmos˚she has begotten, and how to worship˚ her before extracting
the mantras. The tantra adds that once the diagram has been drawn, one must visu-
alise the mantramātrkā who is the very Self of Vis.n.u as being on the pericarp of the
˚
lotus, her body adorned by and made of the fifty phonemes. She is to be worshipped
by the master who must identify himself with her before extracting the mantras (LT,
pp. 77–81). The Sanatkumārasam . hitāa (Indrarātra, chapter 2: mantroddhāra, pp. 228ff.
of the Adyar, 1969 edition) gives a largely similar description.
35 On mātrkā and śabdarāśi as on mātrkā and mālinī, see Padoux, Vāc, chapter 5,
˚
pp. 305–329. ˚
36 On the prastāras and gahvaras, see Appendix I of The S.at.sāhasra Sam . hitā,
chapters 1–5 (Leiden, 1982), edited and translated by J.A. Schoterman: it is a very
complete and interesting study of the prastāra and gahvara system, based on an as
yet unedited text of the Kubjikāmata Kaula tradition, the Sam . vartārthaprekāśa of
Mukundarāja, where a large number of such diagrams are reconstructed.
Appendix II of Schoterman’s study (pp. 210–221) is on Mālinī, both as an order of
letters and as a goddess, and on Śabdarāśi, both as the Sanskrit alphabet and as the god
Śabdarāśibhairava.
The fourth chapter of the Kubjikiāmatatantra – KMT (75–106) describes the extrac-
tion of mantras from the mālinīgahvara. mālinī being feminine, it is the energy aspect
of phonemes that prevails here. The goddess Mālinī is said in the tantra to be the
supreme matrix (parā yoni) of the Rudras and Yoginīs.
There is a Yonigahvaratantra where a gahvara is said to be the matrix (yoni) of all
mantras. Gahvaras are also described in the Goraks.asam . hitā.
37 One finds frequently in India prescribed processes that are as theoretically or appar-
ently constraining as they are in actual practice flexible and adaptable. This is met with
continually in Āgamic ritual prescriptions. But it corresponds, in fact, to a very general
attitude in Hinduism, where, very often, the general prescribed ruling is accompanied
by so many circumstantial precisions or exceptions that it can eventually be adapted
to any issue. In the same way, in purān.ic hinduism, to quote Madeleine Biardeau
(‘Etudes de mythologie hindoue’ BEFEO, LVIII, pp. 63–65), ‘while the official aim is
to be freed of rebirth, all functions so as to allow the world to go on existing eternally’.
This seems to be a characteristic trait of the Hindu religious-social consciousness, one
of the ways in which Hinduism integrates and resolves its contradictions.
38 Two texts that I happen to have at my disposal: others could as well have been
quoted.
39 Pp. 1–26, vol. 4 of the KSTS edition of the SvT.
40 This is Ksemarāja’s gloss.
41 śl. 3 to 10 enumerate various deities or supernatural beings of whom the adept might
be an amśa and the efffect this will have. Thus the adept who has brahmāms.a will be
devoted to the Veda, with rudrām . śa he will be rudrabhakta; the adept with amśa of the
Gandharvas will sing constantly, that of Kāma will love courtesans (gan.ikapriya), and
so forth. In all these cases, explains Ks.emarāja, the deity will give the adept the fruit
corresponding to its power after being meditated upon, its mantra being recited and
being offered an oblation (svasiddheh. phaladāh. sarve svadhyānajapahomatah. – śl.
11a).
42 Pp. 8–9, vol. 4; the process is not described in the Tantra but in the commentary.
 Notes 129
43 There are many periphrases for designating human flesh used as ritual offering.
Used here are the two rather unusual terms vaihāyasa and dhvaja, which, according
to Ks.emarāja’s explanation, are used as meaning [the flesh] of a man who has been
hanged (udbaddha), and of a man who has been executed by impalement (śūlāropita).
However, as Professor Alexis Sanderson (whose help in this matter I gratefully
acknowledge) suggests Ks.emarāja may have interchanged the meanings of these two
terms which are sometimes to be met with, the two sorts of human flesh being paired
for option, in Hindu and Buddhist tantras.
Common terms for sacrificial human flesh are mahāmām . sa (great flesh), mahādhūpa
(great incense), mahāks.mā, etc.
44 The same practice is described briefly (and more clearly) in Ks.emarāja’s commentary
of NT 8. 59b–63, vol.1, pp. 208–209.
45 On this see for instance the ŚT, 2. 127–128 and its commentary (vol. 1, p. 102). This
cakra is shown and explained in G. Bühnemann’s study of puraścaran.a in T. Gou-
driaan, ed., Ritual and Speculation in early Tantrism. Albany: SUNY Press, 1992,
p. 100.
46 This cakra is mentioned in the KT 15, 86. It is reproduced on p. 97 of G. Bühnemann’s
study quoted above.
47 It is shown and explained in Bühnemann, op. cit., p. 101.
48 See Bühnemann, op. cit., p. 96.
On all these methods, see the commentary of ŚT, 2.129–131 (vol. 1, pp. 102–112)
which refers and quotes several tantras: MVT, Mantramukāvali, and (frequently) Pin.-
galāmata. The Tārabhaktisuddhārn.ava (TBhS), a seventeenth century manual, in its
section on am . śa, quotes in its tenth chapter (pp. 359–366) many such texts.
Krsn.a-vais.n.ava practices of this sort (very similar to the Śaiva ones) are described,
in French, with reference mainly to the Rāmārcan.candrikā, in R.V. Joshi, Rituel de
la dévotion krs.n.aïte (Pondichéry, 1959), pp. 20–22. The diagram with four squares
is shown there.˚ This diagram and those used in other am . śa/ka practices (often called
śuddhi in the sense of discharge) are to be found in most contemporary ritual manuals,
in all Indian tongues. For example, in Hindi: Mantrasiddhi kā upāya, by Bhadraśīla
Śarma (Allahabad, 1968).
49 The kūrmacakra is described in the ŚT, 2.133–137 and in pp. 114–117 (vol. 2) of the
commentary, without, however, the way to use it being clearly explained. This is an
important diagram used also in other ritual circumstances, in puraścaran.a notably. It
is also described in the short chapter 20 of the Phet.kārin.ītantra, where it is explained
how to use it to determine the am . śa; the tantra also mentions the mantras whose valid-
ity has not been checked.
50 It is sometimes also called mantrārādhana. The term puraścaran.a can also be applied
to the worship of the mantra – sādhanam . mūlamantrasya puraścaran.am ucyate, says
a text quoted in the TBhS, p. 315 – since it is the preliminary act (puraskriyā) of the
mastery of the mantra. On puraścaran.a, see G. Bühnemann’s study ‘On Puraścaran.a:
Kulārn.avatantra, chapter 15’, in T. Goudriaan, ed., Ritual and Speculation in Early
Tantrism.
51 See for instance the cult of two Yoginīs prescribed in YH 3 166.
52 The SvT, in chapter 8, śl. 3–10, enumerates the different powers the adept will possess
if he has the am . śa of the mantra of various deities: see above, note 36.
53 On these magical acts, see T. Goudriaan, Māyā Divine and Human (Delhi: Motilal
Banarsidass, 1978), which is according to its sub-title ‘a study of magic in its religious
foundations in Sanskrit texts, with a particular attention to a fragment on Vis.n.u’s Māyā
preserved in Bali’ – a very interesting study.
The TBhS, chapter 10 (pp. 359–366) gives the rules for mantravicāra to be prac-
tised with different diagrams for the ‘six [magical] acts’.
54 The KT 15.97 excludes thus kūt.amantras, these being however, according to the
tantra’s commentary, mantras consisting of a group of consonants. The same tantra
130 Notes
(15.97–99) excludes also different mantras, such as those given by a woman, received
in a dream, or those originating from the four Kula traditions (caturāmnāyaja) – which
would exclude at least the mūlamantras of all the Kaula deities – and so forth. In this
case, as in other ones, exceptions are invoked to blunt the edge of the general rule.
55 These mantras are among those excluded by the KT, as said in the preceding note.
56 For instance in the Sanatkumārasam . hitā, Indrarātra, 2.

3 Japa
1 This is an English version, corrected, emended or when necessary updated of the paper
japa, published in the BEFEO, LXXVI, 1967, pp. 117–164.
2 The Amarakośa 2.7,47 says: svādhyāyah. syaj japah., underlining thus either the role of
japa in the svādhyāya or the fact that all japa is a sacrifice of the word, and thus, like
the svādhyāya a mahāyajña, a very important ritual action.
3 The Brāhman.as relate silence (tus.n.īm), mumbling (jalpa), speaking in a low voice
(upāmśu) to the indefinite or illimited (anirukta): ŚBr.5.4.4,13). The mumbling or
murmur (jalpa) is associated with life (AiTBr. 2.39,1. See L. Silburn, Instant et cause,
Paris, 1995.
4 According to the LT (39.35), the vācikajapa is to be used for minor rituals; the
upāms.u for rituals which bring success or power (siddhikarmāni), the mental
(mānasa) one, for rites which bring both success and liberation. The japa done by
dhyāna (meditation or visualisation of a deity?) is for achieving success in all cases
(sarvasiddhi).
The three sorts of japa are sometimes related to the three levels of vāc: paśyantī,
madhyamā and vaikharī – thus Ks.emarāja ad SvT 2.146–7. Or the commentary of
Ks.emarāja on SvT 2. 146–7:
ātma na srn.uteyam . tu mānaso ‘sau prakīrtitah.
˚
ātmanā śruyate yas tu tampām. śum
. vijānate/146/
pare s.rn.vanti yam
. devi saśabdah. sa udāhrtah./
˚
where he explains: ˚
manaso madhyamāyam vāci/upāmśusaśabdau tu
sūks.masthūlaprayatnayām . vaikharyam.
5 These aspects of japa in the MBh are studied by V.M. Bedekar, ‘The place of japa
in the Moks.adharmaparvan (MBh XII. 189–193) and the Yogasūtras’, ABORI, vol.
XLIV, 1963 (pp. 63–74). On the relationship of japa with Yoga, see J.W. Hauer, Der
Yoga (Stuttgart, 1958), pp. 198–199.
6 hotr japam . japati retas tat siñcatyupāmśu. japaty upām . śviva vai retasah. siktih.. The
˚ text (2. 31–32), regarding the recitation of the silent praise, tūsnīm śamsa, which
same .. . .
consists in the japa of bhur bhuvah. svar, considered as the essence of the sacrifice,
parallels it, too, with the emission of semen. See L. de La Vallée Poussin, Etudes et
matériaux (London, 1898), pp. 120–123.
7 On Hindu conceptions concerning sexual energy, its spending or its storing, see Sudhir
Kakar, The Inner World. A Psycho-analytic Study of Childhood and Society in India
(Delhi, 1981). See too, A. Padoux, ‘Le monde hindou et le sexe’, Cahiers Internatio-
naux de Sociologie, vol. LXXVI, 1984.
8 I refer to Alexis Sanderson, ‘The Śaiva Age – The Rise and Dominance of Śaivism
during the Early Medieval Period’, in Shingo Einoo, ed., Genesis and Development of
Tantrism (Tokyo: Institute of Oriental Culture, University of Tokyo, 2009, pp. 41–350).
9 A unity that reflects the unity of the Hindu world. Buddhist .tantric practices, too, are
marked by the same representations (on the creative efficacy of consciousness or on
the image of the body notably).
10 ‘It has come to designate, in classical India, all types of prayer’, see L. Renou, ‘La
valeur du silence dans le culte védique’, JAOS, 1979, pp. 11–18.
 Notes 131
11 Hélène Brunner, who read a first draft of this chapter and made a number of useful
suggestions, remarked that it would have probably been more rational, and clearer, for
me to distinguish between the cases where the worship is the main portion and raison
d’être of a rite, japa being accessory, and those where the japa is the essential part, the
worship being accessory. She also drew my attention to the fact that it is in kāmya rites
– especially those performed by the sādhaka – that japa takes the larger place, which
is true, see note 16. .
12 As, for instance, for Gan.eśa: OM . HRĪM . ŚRĪM . NLAUM . GAM. Gan.apataye vara varda
sarvagan.am . me vaśa ānaya THAH. THAH..
13 akārādi ks.akārāntam aksaramālā vidur budhāh. (Japalaks.an.am – ms. no. 545, Gov.
Orl. Library, Madras). The Nityotsava (NU), p. 127, enumerates thus the rosaries made
of the varn.as of the mātrkā ‘from the letter A to the letter KS.A’ and those made of
rudrāks.a, pearls, rubies, ˚crystal and so forth. The japa of the varn.amālā is prescribed
in a number of texts. A Śaiva Upanis.ad, the Aks.amālikopanis.ad, describes this japa,
which is done with a rosary and includes various mental visualisations.
14 See Phet.kārin.ī (Phet.kT) 18.1–8:
bhūtalipyā pūt.īkrtyā yo mantram bhajate narah. /
˚
kramotkramācchatāvr ttyā tasya siddho bhaven mantrah. //1//
˚
The following ślokas describe this action and its effects. This technique is also pre-
scribed by the KT 15.16. On sam . put.īkaran.a, see here, chapter 7, note 5.
15 mātrkājapamātren.a mantrān.ām . kot.ikot.ayah./
˚ syur na sandaho yatah sarvam tadudbhavam //18// – KT 15.18).
japitāh. . .
16 See, for instance, the japa of the gāyatrī among Brahmins of Gujerat described by L.
Sinclair Stevenson, The Rites of the Twice-Born (New Delhi: Orl. Books Reprint Cor-
poration, 1971), p. 223. Or, more generally, R.B.Ś. Chandra Vidyarnava, The Daily
Practice of the Hindus, chapter IV, Sandhyā,, especially pp. 39–42.
17 Various prescriptions on favourable periods for japa are to be found in TBhS, pp.
316–317, where several texts are quoted. The JayS 14.188–189 prescribes different
periods according to the nature (sātvika, etc.) of the adept. See also AgPur 292.31ff.:
mantrasya devatā tāvat tithivāres.u vai japet:/ etc.
There is also a rule in several texts (AgPur, ŚT, TRT, etc.) saying that japa is fruit-
less when done while the mantra ‘sleeps’: svāpakāle tu mantrasya japo na phalapra-
dah., says the Rudrayāmala quoted in TBhS, p. 324. But the times when a mantra
is ‘asleep’, or ‘awake’ are not to do with the time of clocks. For the AgP (292.8)
.
they are linked with the flow of prān.a in the nād.is idā and pingalā (see Padoux ‘On
Mantras and Mantric Practices in the Agni-Purān.a’, Purān.a, vol. XX/1, 1978, p. 59).
Other texts say that the ‘sleeping’ state of the mantra is the effect of a dos.a due to
its phonetic structure – so ŚT 2.84: trivarn.o ham . sahīno yah. suśuptah. sa udīritah. –
this defect is cured, the ŚT says, by showing the yonimudrā together with the udāna
prān.a. The commentary on this śloka (ŚT 2.110, vol. 1, p. 94), however, says that the
correction of the defect is necessary only in kāmya rites. This point would tend to jus-
tify Hélène Brunner’s suggestion I mentioned previously (note 10) that insofar as japa
is a manipulation of the power of the word for particular ends, japa would especially
be used by sādhakas rather than by liberation-seekers (mumuks.u).
18 Infra, p. 42.
19 This is also called pūrvaseva, which designates more generally the preliminary propi-
tiation of a deity by the sādhaka to obtain supernatural powers – see the entry on that
word in TAK3.
20 I will come back on this later (see pp. 34–35). On the mantrasādhana, see Hélène
Brunner’s study, ‘Le sādhaka, personnage oublié du śivaïsme du sud’, J.As., 1975, pp.
411–443, where, for the choice of the place, the author refers notably to the Mrg Cp.
1.95ff. ˚
21 KT 15.22–30. Similar prescriptions are to be found in the PRKS 3.2.35–3/1 (p. 544),
132 Notes
or in the TBhS, p. 317, which quotes three different texts; or in the Makut.āgama
2.766–768 (information H. Brunner).
22 See A. Sanderson, ‘Power and Purity among the Brahmins of Kashmir’, in M. Carrith-
ers, et al., eds., The Category of the Person (Cambridge: Cambridge University Press,
1985), pp. 200–202.
23 See, for instance, the prescriptions of the JRY 2.11 for the sādhana of the goddess
.
Mahājhankarin.ī: the japa is done at night, in a cremation ground, a temple to the Moth-
ers or a place for ‘Heroes’ (vīrasthāna), while wearing black clothes, etc.
24 Before a ritual is performed it is necessary to enclose the ritual area by means of a
rite named digbandhana, which both delimitates and protects it from dangers coming
from outside. On such a rite, see, for instance, TBhS pp. 317–348.
More generally, on the question of the protection of the ritual area against dangers
and impurity felt as always menacing – on the Hindu preoccupation with boundaries,
see D. Shulman, ‘The enemy within: idealism & dissent in South Indian Hinduism’, in
S.N. Eisenstadt et al., Orthodoxy, Heterodoxy and Dissent in India (Berlin: Mouton,
1984), pp. 11ff.
25 In principle, one must not be lying nor standing – see KT15.032–33, or the MatP Kp.
11.12b–13 where other references are given (information H. Brunner).
26 See Mrg Cp. 75b: sādhakah.. sādhyaves.adharah.. The same text, Kp.3.1–45b–47, men-
˚ colours of the different aspects of Siva (inf. Brunner). All worship (as well
tions the .
as many other rites) are to be performed wearing the proper sort of clothing. Usually
it must be pure and clean; but in Tantric context it may, on the contrary, be chosen so
as to mark the transgression: rags, black material for the Kāpalikās, transvestism in
goddess cults, and so forth.
27 Such pavitras are usually rings of sacred herbs bound around the finger. On the sub-
ject, see the SP, vol. 2.
28 On all these points, see Rāghavabhat..ta’s commentary on ŚT 16.54–56 (vol. 2,
pp. 656ff., which is on puraścaran.a, and quotes a number of texts. Detailed prescrip-
tions are also found in recent works, which shows that these instructions are still valid:
see, for instance, Karanidan Sethiya, Mantra-vidyā, Calcutta, sam. 2033 (in Hindi),
pp. 10–16 – the directions given concern japa in general. For the author, the colour
of the clothes to be worn must be in accordance with the bhāva of the deity. Sewed
garments are often prohibited – which is a general rule among observant Brahmins,
for all rituals.
29 The Brahmayāmalatantra is as yet (2009) unpublished, It has been studied and partly
edited by Shaman Hatley (thesis, 2007).
Bhrūkut.i, contracted brow, is a facial expression typical of wrathful deities –
Bhairava, here. At..tahāsa is the laugh of Bhairava or other wrathful deities (such, for
instance, as the corpse-like Sadāśiva imagined by the yogin as lying at the base of the
triśūla on top of which are the three supreme goddesses of the Trika – Parā, Parāparā
and Aparā.
30 This still unpublished tantra is being studied and is in the course of edition by Judit
Törzsök.
31 On these nyāsas see the next section.
32 NU, p. 93; Prān.atos.in.ī, p. 421, quoting the Sarrasvatītantra, which prescribes a pre-
liminary purification of the mouth, mukhaśodhana. The GT prescribes the utterance of
the setu OM . before and after the japa. Other prescriptions in the JS, chapter 14.
33 NU pp. 93 and 94.
34 See for instance the LT 28.47–50 already quoted which prescribes alternate japas and
meditation.
35 ŚT 16.56, comm. (pp. 655–6). For Krs.n.a rites, see R.V. Joshi, Le rituel de la dévo-
tion krs.n.aïte (Pondichéry, 1959). Many ˚ informations given there come from the
˚
Haribhaktivilāsa, an important ritual manual. See too the Nāradapañcarātra – S.rī
Nārada Pañcaratanam (Sacred Books of the Hindus, 23, reprint, New York 1974).
 Notes 133
36 The comm. on Mrg, Cp 1.76 prescribes audible, murmured or mental japa depending
as to whether the ˚mantra to be mastered by mantrasādhana is of an inferior, middling,
or superior sort (without saying what are these three sorts of mantras).
37 Thus NU, pp. 129–130, quoting the Svachandatantrasāra: japas tu s.ad.vidhah.
.
vācika mānasa yogavācika yogamānasika vanmānasikayaugikah. – the four last sorts
corresponding to the association of the utterance of the mantra with yogic practices.
On japa, prān.a and yoga, see further down. The Makut.āgama 2.760 lists japas in
descending order as follows: mānasa, kan..thos.tha (?), upām . śu and vācika.
TĀ 29. 82–88, quoting the Yogasam . cāratantra, describes different sorts of japa
which are yogic practices involving the movement of prān.a.
38 ‘Hasty japa brings about sickness and slow utterance destroys wealth’ writes a con-
temporary author: introduction to an edition of the Mantramahodadhi (Mud) (Delhi,
1981)
39 YS 1.28: tajjapastadarthabhāvanam ‘Its [sil. of the mantra] repetition and concentra-
tion on its meaning [should be made]’.
40 One may well ask oneself, in such cases, whether the mind of the practitioner is fas-
tened on this mental image, or on the sound of the mantra he recites. Perhaps, the
sound forms, as it were, a basis whence the visual image arises.
41 On this subject, see the studies of J.F. Staal (for whom rites are without meaning),
The Science of Ritual (Poona, 1982), and Ritual and Mantras: Rules without Meaning
(Indian edition, Delhi, 1996).
42 See YH 3.191, and Dī, p. 332. The japakāla may last a fraction of a second: there is
no real duration. But such japas are in fact a mental attitude, a bhāvanā, of the prac-
titioner. Many such instances are to be found – see, for instance, the Kulamūlaratna-
pañcakāvatāra of the Kubjikā tradition, 3.17.
43 In the Mahādevayajña ritual of Tanjore, for instance (in 1951, as reported by C.G.
Diehl in Instrument and Purpose, Lund, 1956), were performed a hundred thousand
repetitions of the thousand Names of the Goddess, a hundred million recitations of the
śrīvidyā, etc.
44 PRKS, p. 550. The form or the material of the seat, the clothes, the bodily posture, etc.,
vary according to the context and circumstances of the japa. See, for instance, MatP
Kp 11.12ff. (vol. 2, p. 198). In all cases one is to follow the prescriptions given for the
ritual which includes the japa.
45 One could say here that insofar as we may consider mantras to be performative utter-
ances in the sense of Austin and Searle, the circumstances mentioned previously in
which the japa is to be made to produce the desired effect are merely an intensely
meticulous case of the necessity of the appropriate social conditions of utterance for it
to be effective as defined by these two authors. On performative utterances, one may
refer to J.L. Austin, How to Do Things with Words (Oxford, 1975), and to J.R. Searle,
Speech Acts (Cambridge, 1980).
46 Śākta texts sometimes prescribe counting on eight phalanxes only, which results in
sixty-four (8 × 8), an important number in Tantra: there are sixty-four Bhairava tantras,
sixty-four Yoginīs, and so forth. See for instance Yoginītantra 2.30, Another method
(on nine phalanxes), GT 18.11–12.
.
47 The JayS 14.72–76 (meror tanghananis.edhah.) explains how to do this. For japa of the
varn.amālā,, KS.A is considered as the Meru, but it can be included in the count: see
Mantrakaumudī of Devanātha Thakura (Darbhanga, 1960), 14: anulomavilomais tair
binduvān.mātrkāks.araih./ks.amerukaih. sās..tvargaih. kliptayā varn.amālayā //13//.
If one uses˚ the hand for counting (karamālā), the ˚ second of the middle finger (at
least according to some texts) is to play the role of the Meru. But the counting is not
done going forward and backward (see Yoginītantra 2.40: anulomavilomabhyām .
sarvamālāsu sam . japet/kevalam . cānulomena prajapet karamālayā). For the JayS
14.72ff., the rosary is always to be counted in the same way, a particular gesture avoid-
ing the overstepping of the Meru.
134 Notes
When one does japa of the phonemes (varn.amālā), from A to KS.A, one recites it twice,
anuloma then vilomena, the first phonemes of each of the eight varga (A, KA, TA, etc.)
so as to reach a total of 108. According to the JayS, the aks.amālā has 108, 54 or 27
beads.
48 This is a case of the rule prevalent in the Hindu world that the index is never to be
used except in destructive or malignant rites. See GT 18.27–30 which prescribes
the use of different fingers for different aims. Prescriptions of the same sort in the
Śivārcan.acandrikā quoted in SP1, pp. 216–217: one must use different fingers depend-
ing on the the japa being audible, whispered, or mental.
49 PRKS, p. 551: vastren.ācchādya tu karam . daks.in.am
. yah. sadā japet.
50 See YH 3.189b: ‘the japa once thus accomplished, it must be offered in the left hand
of the Goddess’, evam . krtvā japam . devyā vāmahaste nivedayet.
51 If a new rosary breaks, one˚ is to repeat a thousand times the Aghoramantra, but merely
a hundred times if the rosary is an old one (see prāyaścittavidhih., 48, SP2, pp. 260–
261). The breaking of the rosary is karmachidrakaran.am: it causes the destruction of
the [good] karman (JS 14.93).
52 OM . HŪM . VAUM . Viśvaks.en.āya namah..
53 Pp. 325–329 of the TBhS are on puraścaran.a., enumerating or describing its rites. See
too Rāgahvabhat..ta’s commentary on ŚT 15.54 (vol. 2, pp. 656ff.) which quotes many
texts; or GT. 18. For the vais.n.ava rite, see the Nāradapañcarāttra, III, chapters 10 to
14. Kane, History of Dharmaśāstra, vol. 5/2, pp. 1008–1112 are on puraścaran.a, with
many textual references. See also H. Brunner’s study, ‘Le sādhaka’ quoted in note 13
of chapter 2.
54 KT 15.18, the pūjā of this list is the threefold daily sandhya – a nitya ritual.
55 See Hélène Brunner’s (French) translation, and her notes, pp. 282–284 of the study
quoted in notes 16 and 44.
56 The ŚT 18.14 (vol. 2, p. 716) prescribes 24 million (tattvalaks.am . japan mantram)
repetitions of the mūlamantra of Śiva, followed by 24,000 oblations, which implies
that the ritual lasts several days or months. Manuals of ritual discuss such details and
the time necessary for such practices.
57 See thus the Nāradapañcarātra, III. chapter 14, śl. 15ff., where the japa of two man-
tras of Krs.n.a protects children against demons (15), keeps away harmful astral influ-
˚ gives long life (28), brings rain (45), protects from illnesses and from
ences (19),
sudden death (52), etc. This japa is to be accompanied by offerings in the fire whilst
meditating and visualising the child Krs.n.ā in his ‘games’ (līlā).
58 Cf. GT 18.68: agnir eva mukham ˚
. devyās tasmād agnau hutet sadā. Those are ancient
notions.
59 Kane (op. cit., p. 1109) says thus that, according to the Kaulāvalinirn.aya, a rite of japa
is to be done during the night in a cremation ground with a corpse. The TBhS (p. 334),
quoting the Svātantratantra, describes another nocturnal rite where the japa accom-
panies sexual ritual practices.
60 The same prescription (but without the mudrā) is found in the Mrg. Kp. 4, 5 (p. 42,
where Bhat..tanārān.ayakan..tha’s commentary is very explicit). ˚
61 This is at least how the rite appears to me. H. Brunner, who knows the subject much
better than I do, feels that I stretch unduly the meaning of this action. For her, the
flower symbolises the whole japa, this symbol being what is given to the deity (see
Mrg’s French translation, p. 55, note 2).
62 But˚ a Tantric pūjā is always seen as a process of identification with the deity that is
worshipped, an identification that is ritually achieved (or played out) several times
during the cult.
63 On these elements, see here chapter 4 on nyāsa.
64 devīnām . vāmahaste tu devānām . daks.in.e ‘rpayet, says the Rudrayāmala quoted in the
Tārin.īpārijata, p. 94: this eighteenth-century work gives a number of details on the
japa (japamālā), of the phonemes of the alphabet, pp. 87–89.
 Notes 135
65 The LT 40.18 says (the goddess is speaking): ‘having completed the japa accord-
ing to the prescribed method, he should give it over to Me and imagine the śakti in
the form of japa as being in my mouth’. Here too the deity absorbs the power of the
word when the utterance of the japa ends. The officiant is now to ring a bell identified
with Sarasvatī, the goddess of speech, which will cause the mantras to come and thus
contribute to the efficacy of the recitation. There follows a ritual worship which will
ensure the mantrin’s mastery of the mantra or the fruitful result of its use (mantrasid-
dhim. nigacchati – śl. 29).
66 See for instance Daily Practice of the Hindus, chapter 9, ‘Sandhyā of the Yajurvedins’,
section 23, japanivedana: having recited the gāyatrī 108, 28 or 10 times, the officiant,
seated in front of the image of the deity, pours some water in the palm of his right hand
while saying anena (daśa, as..tavim . śati, etc.) sam
. khyayā japam . bhagavan (Brahmā
Visn.u, etc.) svarūpi sevitā priyatām . namah., and pours that water in the hand of the
devatā (ibid., p. 109).
On the other hand, the private cult that the temple officiant performs at home (or in
the temple) for himself, where the japanivedana we have seen takes place, is entirely
Tantric.
A mental nivedana linked to the movement of prān.a, at the end of the japa, is pre-
scribed in SvT 2.136–143.
We may, apropos of the practices described above, underline once more the fact that
Tantric practices, far from contradicting the Brahmanic tradition (even if it often runs
counter to it), takes place within this tradition: Tantra, I believe, brings to light the
inner virtualities of Brahmanism.
67 Here is, as an example, how a small modern manual (Śrīkāīīnityārcan.a, Prayāg,
Kalyan Mandir, sam . 2030, p. 89) describes (I have shortened a little) the japa at the
end of Kālī’s pūjā: after the preliminary rites, pay homage (tarpan.a and pūjā) to the
rosary with the mantra Śrīdaks.in.akālikā-aks.amālām . tarpayāmi pūjāyāmi namah.; then
pray to the rosary to give success (me siddhidā bhava). One holds it then with the right
hand and says another prayer to remove all obstacles. Having saluted the rosary, one
will place it on the middle phalanx of the middle finger and touch it with the thumb
saying OM . , which is done while ‘meditating’ (dhyāna) the form of the goddess. Then
one will pray to the rosary, asking it to grant happiness and success. Having held it
to his head while reciting as before the setu and mahāsetu, the adept or devotee will
.
practise a prān.āyāma. He then does a rs.yādinyāsa and a karānganyāsa. Then the japa
˚
itself is to be done. Once finished, it is to be offered, together with a special arghya
and amrta, in the left hand of the goddess, while saying OM . guhyātiguhyagoptri
tvam ˚
. grhanas matkrtam . japam. . siddhir bhavatu me devi tvatprasādānan maheśvari.
idam ˚ ˚
. japam . śrīdaks.in.akālikādevyā samarpayāmi svāhā. Then one salutes her making
the yonimudrā and saying HRĪM . siddhayai namah.. After which the rosary is to be kept
in a secret place. The japa is to be followed by a homa.
The TĀ 29, describing a pūja according to the Mādhavakula and quoting the
Yogasam . cāratantra, prescribes a japa of a 100,000 to 1,200,000 repetitions, linked to
the ascent of kun.d.alinī; it is said to be a murmuring (sam . jalpa) whose essential nature
is an awareness of the subtle phonic vibration (nādāmarśasvarūpin.ī).
68 In the case of the japa (as part of the mantrasādhana) prescribed in the NS.A 5,
according to the number of repetitions (from 100,000 – one lakh – to 900,000), the
mantrin would be freed from the sins of his previous lives, would be granted the
eight ‘great powers’ (mahāsiddhi), or, for nine lakhs, the fusion with the deity (ibid.,
śl.5–6).
69 The fact that śānti, a favourable rite, is among the six magical actions (and is even
often quoted first), as well as the fact that these acts are usually described in mythi-
cal and religious texts, show that the distinction religion/magic which I make here is
problematic and, in fact, quite difficult to apply in the Indian context. This would also
raise the more technical problem of the distinction between kāmya rites (that of śānti
136 Notes
notably) and nityā rites – on which see H. Brunner’s observations in her seminars at
the Ecole Pratique des Hautes Etudes, Sciences Religieuses, in 1984.
70 See also Vīn.aśikhatantra (ed. Goudriaan, Delhi, 1985), śl.224ff.
71 śāntipus..tivaśākrtyucāt.ana māran.e /
svadhāsvahāvas ˚. at. hūm
. ca vaus.at.phat. yojayet kramāt //
See too the explanatory note 62 in SP3, p. 166, or ibid., pp. 532–533, on astrābhis.e-
ka (ritual sprinkling). In the māran.a, etc. rites, one utters the mantra without bindu or
nāda.
72 One finds prescriptions of the same sort in the Śes.asam . hitā (14.58–64,a,d 15, pp.
79–83) in a passage concerning the s.at.karmān.i. Concerning the use of fingers, that of
the index for destructive rites is normal (see above). In fact, prescriptions vary largely
according to texts. The relationship between some physical aspects of japa and the
nature of the rite does not appear to us prima facie to be logically grounded – though,
admittedly, a more precise research on the subject may uncover symbolic reasons for
this situation.
On the magical uses of mantras in contemporary India, one may refer to C.G.
Diehl’s study, Instrument and Purpose, previously quoted.
73 See the observations of Hélène Brunner in SP3.
74 The omissions or errors repaired by prāyaścitta are, so to speak, technical accidents
during the utterance. They are not the procedural errors or omissions – the ‘infelici-
ties’ – which, in Austin’s then in Searle’s theory, make the speech acts ineffective.
(On mantras as illocutionary or perlocutionary acts, see A.C.S. McDermott, ‘Towards
a pragmatics of mantra recitation’, Journal of Indian Philosophy, vol. 3, 1975, pp.
283–298.)
75 See H. Brunner’s study ‘Le sādhaka.’, JAs, 1975, previously quoted.
76 Sāttvatasam . hitā, with a commentary by Alasinga Bhatta, edited by V.V. Dvivedi (Var-
anasi, 1982).
77 Thus the brahmavidyā recited in the ear of a dying person, the dīks.ā being in that case
limited to the repetition of a mantra, according to the TĀ 19. 31–33 – the recitation, in
that case, is a mere ‘reading’ ((pat.ha) of the mantra.
78 See, for instance, SP3, note 54a, p. 221, where is quoted the Mrg, Kp7, 98–110, pre-
scribing a japa of the Svapnamān.avakamantra, an invocation ˚of the god of dreams
(necessary to be known for the dīks.ā).
79 In the funerals, antyes..ti, for instance (see SP3, pp. 372–3 and 610), but it is in fact a
prāyas.citta; or, in the cult of ancestors, the s.rāddha (ibid, pp. 656–657. The MatP Cp
10.73 prescribes 108 repetitions of the Bahurūpamantra to purify those who bear the
corpse of an ascetic (yātin) to the place where he will be buried.
80 On this point, see Milton Singer, ‘The Radhâ-Krishna bhajanas of Madras city’ in M.
Singer, ed., Krishna: Myths, Rites and Attitudes (Honolulu: East-West Press Center,
1966). See too the observations of Hélène Brunner in the Supplément of vol. 3, ‘Rites
et Fêtes’, Fr. L’Hernault & M-L. Reiniche, eds. of the study carried out and published
by The Ecole Française d’Extrême Orient, Paris, 1990–1999, Tiruvannamalai, Un lieu
saint sivaïte du Sud de l’Inde.
81 See pp. 377–392 of that translation; a chart is given on p. 378 of the ascent of the
śrīvidyā along the sus.umnā.
82 On these, see the very detailed notes of H. Brunner in SP3, pp. 358–396.
83 Infra, p. 46.
84 Commenting on the passage of the YH 3 on the japa of the avasthās and on the
śūnyas, Amrtānanda uses the term bhāvanā to describe this process. On bhāvanā,
see F. Chenet,˚ ‘Bhāvanā et la créativité de la conscience’, Numen, vol. XXXIV, 1987,
pp. 45–96.
In an analogous though somewhat different perspective, the TRT 38.18, apropos of
the cult of the Yoginīs in the śrīcakra, mentions three different ways for identifying
 Notes 137
with the deity: nyāsa, which makes the body of the adept similar to that of the deity,
japa, which creates this similarity (japas tanmayatārūpabhāvanam), and homa, as
a symbolical offering of the universe which dissolves the multiplicity of the created
world into the unity of the Self.
85 This sort of japa is sometimes described as a remembrance (smaran.a). Thus YH 3.184,
on the japa of the avasthās, or the Varivasyārahasya of Bhāskararāya, 52, on the japa
of the avasthās, śūnyas and vis.uvas. This is due to the influence on these authors of the
Pratyabhijñā system for which smaran.a is a means for the realisation of the Absolute
since, by linking the past to the present, it focuses the attention on the supreme reality
underlying all phenomena.
86 This information I owe to Alexis Sanderson.
87 F.D. Lessing and A. Wayman, op. cit., p. 187.
On another analogous Buddhist practice, see Stephen Beyer, The Cult of Tārā
(Berkeley, 1973), p. 450–452, ‘Reciting the mantra’. Simpler is the japa of chapter
3 of the Mahāvairocanasūtra (ed. Tajima, pp. 117ff.) where a meditative recitation
(bunsho manju) is done by contemplating in one’s heart the shape of the letters of the
mantra: one sees there the importance of the written sign in Sino-Japanese culture.
88 See Padoux, Vāc.
89 The Yogīśvarīmata, also called Siddhayogeśvarīmata, Siddhamata, Siddhatantra, etc.,
is often quoted in the TĀ. Its doctrine is said to be expounded in the MVT. A critical
edition of selected chapters of this tantra, with an annotated translation, was done by
Judit Törzsök in an unpublished habilitation thesis (Paris, 2010). However, few of
the quotations of this tantra given in the TĀ are to be found in this edition, which is
perhaps that of a shorter version.
The Yoginīkaula is quoted only once in the TĀ: is it the same text as the Yoginītantra
mentioned in the JRY?
90 He quotes the following śloka:
japet tu prānasāmyena tatah. siddhir bhaved dhruvam /
nānyathā siddhim āpnoti hāsyam āpnoti sundari // (vol. 4/2, p. 30).
91 Here is Jayaratha’s commentary: tattadekapin.d.ādyātmakamantrarūpatayā bahir
ullasantī varn.akun.d.alinyākhyā parameśvarī śaktir yadi nāma prān.asamā prān.a-
sāmyenodayam iyāt tadunmanā śvaikātmyena prasphured ityarthah. (p. 32).
92 This at least is how one usually translates ham . sa, this bird being in fact a duck of
sorts.
93 See for instance, SvT 7.54–55 or VBh 155: ‘21.600 time by day and night this japa is
prescribed as that of the Goddess’. See too TĀ 7.47–50, where various divisions of the
mantra and of its utterance all result in 21,600. This number of breaths per twenty-four
hours is in fact perfectly real.
94 Thus SvT, VBh, TĀ, ŚSV, such Upanis.ads as Dhyānabindu, Mahāvākya, Yogaśikhā,
and so forth. In the Brahmavidyā Up., the japa of ham . sa is combined with that of OM
.
and is associated to the ascent of kun.d.alinī.
95 Gāyatrīs can be composed on the model of the savitrī for all divinities.
96 A Tamil author who may have lived in the sixteenth century. This work was edited
and translated in French and commented on by Tara Michael (Pondichéry: Institut
Français d’Indologie, 1975).
97 Here is SvT 7.59:
nāsyoccarāyitā kaścit pratihanta na vidyate /
svayam uccārate ham . sah. prān.inām urasi sthitah. //
The ślokas of VBh run as follows:
sakāren.a bahir yātihakāren.a vis.et punah. /
ham . saham. sety amum
. mantram . jīvo japah. nityaśah. //
s.at.śatāni divā rātrau sahasrān.y ekavim. śatih. /
japo devyāh. samuddis..tah. sulabho durlabho jad.aih. //
138 Notes
They can be translated as follows:
With the letter sa [breath] flows out; with the letter ha it flows back.
Thus the living being recites this mantra ham . saham . sa.
Twenty-one thousand times, night and day, is this japa of the supreme goddess,
it is very easy [but] difficult for the ignorant.
98 See, for instance, the Goraks.aśataka quoted by Briggs, in Goraknāth and the Kan-
phata Yogis (Delhi: Motilal Banarsidass, 1982), pp. 292–293; or the Yogavis.aya, 29,
p. 47 of Kalyan Mallik’s Siddhasiddhāntapaddhati & Other Works of Nāth Yogis
(Poona: Oriental Bookhouse, 1954; or the Yogamārtanda, 33ff. (ibid., pp. 58–59). The
ajapājapa is still in current use even in non-yogic or ascetic milieus. An analogous
practice is to be found (with so’ham. instead of ham . sa) in the Ham . sopanis.ad, 10–13,
the Brahvidyopanis.ad, 78–79, among the Sants, etc.
99 See the texts quoted by S.B. Dasgupta in An Introduction to Tantric Buddhism (Uni-
versity of Calcutta, 1950), pp. 122–125, and in Obscure Religious Cults (already
quoted here), pp. 96–99.
100 samyendriyasañcaram . proccaren nādam āntaram /
es.aiva japah. prokto na tu bāhyajapo japah. //
.
This stanza comes perhaps from the Sanketapaddhati, an important work of the
Śrīvidyā known only by quotations.
101 Except probably insofar as the inner perception of the nāda can be actually perceptible
as some sort of subtle inner phonic vibration. Several works describe different sorts
of nāda accessible to yogins, some of which appear as perceptible. There is thus a
division into sixty-seven nādas, eleven of which are ‘unstruck’ (anāhata) sound: see
Vidyānanda, Artharatnāvalīūī ad NS.A 1.12. The subtle form of nāda is often com-
pared to the end of the vibration of a bell which becomes ever more fine untill it disap-
pears altogether in utter silence.
102 bhūyo bhūyah. pare bhāve bhāvanā bhāvyate hi yā /
japah. so’tra svayam . nādo mantrātmā japya īdrśah. //
103 The I is aham, which written a+ h +m ˚
. symbolises the totality of the cosmic manifesta-
tion which, going from the first (A) to the last (HA) letter of the Sanskrit alphabet, is
brought together into a unique dot, the origin of everything, the bindu M ..
A and Ha are sometimes taken as symbolising the manifestation and the resorption
of the cosmos; thus Ksemarāja ad SvT7, 29.
104 This tantra, also named Triśirastantra, Trśiromata, etc., is often quoted by Abhinav-
agupta. It is known only by quotations. It apparently expounds the conception of the
supreme godhead as made of the three goddesses Parā, Parāparā and Aparā.
105 kalmas.aks.īmanasā smrtimātranirodhanāt /
dhyāyate paramam ˚
. dhyam . gamāgapada sthitam /89/
param . śivam . tu vrajati bhairavākhyam . japad. api /
tatsvarūpam . japah. prokto bhāvābhāvapadācyutah. //90//
106 Parāmarśa, this essential notion of the Trika, is very difficult to translate; see the entry
on that word in TAK, Vol. 3.
107 This text seems to be known only through quotations by Ks.emarāja and a few other
authors. It is likely to be a Krama text.
108 mahāmantrātmakākrtakāham . vimarśārūd.hasya yad yad ālāpādi tat tad asya
˚
savātmadevatāvimarśānavaratāvartanātmā japo jāyate.
109 For Bhāskara (ŚSVārttika), japa appears in this case as an enunciation resting on
energy (vīrya) and leading to identification with the supreme Being. According to him,
it can be of four sorts: ‘of energy’ (śākta), identifying the soul with the Self, ‘without
parts’ (nis.kalā), which seems to be the utterance of OM . , that of the ham. sa ‘made of
the klaās of nāda – nādātmika – and finally the japa of the living being (jīva), or of
the body (paudgala) which is probably the movement of respiratory breath. This com-
mentary is earlier than Ksemarāja’s, and probably nearer to the spirit of the ŚS.
 Notes 139
110 We might say of ham . sa with Lilian Silburn: ‘All human beings recite it perpetually,
without their knowing it, whether awake or asleep. But the jñānin, different in that
from the ignorant, is aware of this ham . sa; he enjoys the presence of Śiva united with
his energies, powerful, omniscient’ (my translation from p. 178 of L. Silburn’s French
translation of the Śivasūtras with Ks.emarāja’s Vimarśinī).
111 akrtrimataddhrdayārūd.ho yat kim . cid ācaret /
˚
prānyād vā mr˚śate vāpi sa sarvo’sya japo matah. //
˚
112 In the fourth chapter of the Tantrasāra, which gives a shortened version of TĀ 4, Abhi-
navagupta says that pure intuitive reasoning (sattarka), the pure gnosis (śuddhavidyā)
which is the way towards the Absolute can be attained by sacrifice, by oblation in the
fire (homa), by japa or by yoga, all of which he defines as an immersion in divine
Consciousness. Of japa, he says: ‘it is an inner intense awareness of the fact that
the supreme Reality exists by its own nature, independently from the existence of
the diversity of the cognisable, for all that is then to appear is a consciousness which
grasps and transcends all this dichotomy’ (ubhayātmakaparāmarśodaya).
113 matparam . nāsti tatrāpi jāpako’smi tadaikyatah. /
tattvena japa ityaksamālayā diśasi kvacit //
114 When commenting on this stanza, Ks.emarāja adds: ‘You teach that this japa which
in essence is an awareness of the Unsurpassed … is to be accompanied by the plac-
ing [on the body] of the letters of the alphabet’ (japah. … anuttaravimarśasāro …
varn.alipinyāsena yuktyā śiks.ayasi).
In fact, all japas are oriented towards an identification with the deity. This is obvious
for the devotional japa, as we shall see. But even in the puraścaran.a preliminary to
kāmya rites, this identification is described as necessary: without it the japa is useless.
115 See, for instance, the YH 3.3 where Bhairava justifies the daily practice of the pūjā
of the śrīcakra by saying: ‘It is performed perpetually by me’– cakrapūjā ca sadā nis.
padyate mayā.
116 On this, one may refer to C.G. Diehl’s study, Instrument and Purpose, quoted here,
note 43. See notably, pp. 267–324, on ‘Mantirikam’, where many prophylactic heal-
ing rites, rites for success and so forth are described, of which japa is an important
element.
117 A difference which of course does exist up to a point in non-dualistic context, where
the essential unity of the human being and the godhead, of the ātman and of the brah-
man, is effectively experienced at the end of the ascesis by the liberated in life. The
Śivastotrāvalī is an instance of non-dualistic devotional literature.
118 Elsewhere he writes about japa in ān.avopāya: ‘shall we say that the repetition of
a religious formula plays the same part in bringing about the liberation as does a
lullaby in putting a child to sleep, a state of physical quiescence’. Further on, he com-
pares the effect of the two other upāyas to autosuggestion (śāktopāya) or deep sleep
(śām . bhavopāya) where the loss of consciousness is complete – but sleep, in this case,
is taken in the sense of the second avasthā, therefore as a state superior to wakefulness,
which is not our perspective. The case is nevertheless worth considering.
119 See RV 10.63, 2: ‘For all your names are worthy of homage, worthy of praise, Gods,
[your names] are also worthy of sacrifice’, viśvā hi vo namaśyāni vandyā devā utà
yajn.iyāni vah..
120 The same sort of japa is prescribed in Śaiva Upanis.ads (Atharvaśiras, Kaivalya,
Jābali).
121 kim. japyenāmrtatvam . brūhīti/sa hovāca yajn.āvalkyah. śatarudriyen.eti /
122 Cf. SkPur 1.1.14:˚
śiveti dvyaks.am . nāma vyāharis.yanti ye janah. /
tes.ām. svargaś ca moks.aś ca bhavis.yati na cānyathā //
123 Such practices seem to have existed earlier in Buddhism for the worship of Amitābhā.
But both the ‘spirit’ and the actual practice of such a japa differ very much from those
of the Hindu japa.
140 Notes
124 This according to Ch. Vaudeville, Kabir (Oxford: Clarendon Press, 1974). Curiously,
the Haripāt.h of Jñāndev (8.2) says that the japa of Rama’s name is the ‘breath’ of Śiva:
ātmā jo śivācā/rāmajapa: Ch. Vaudeville, L’Invocation. Le Haripāth de Dyāndev
(Paris: EFEO, 1969), p. 64. See note 120.
125 See Milton Singer, ed. op. cit., note 80, ‘The Radha-Krishna bhajanas of Madras
City’, especially pp. 118–121.
126 Adh.R, I.6.69; II.5.25; IV.1.78 – this reference is found in F. Whaling, The Rise of the
Religious Significance of Rāma (Delhi: Motilal Banarsidass, 1988), p. 160.
127 Tulsidas, while affirming that no other name than Rama’s saves, admits however the
efficacy of other names, notably those of Sītā and Bharata ‘for many are the names of
the Lord’.
We may note that for Tulsidas Rāma is the mantra of Śiva who, at Kaśi (Varanasi),
saves all beings by giving them this mantra. The Haripāth says (8.2) that the japa of
Rāma’s name is the breath of Śiva – is this an allusion to the ajapājapa? ‘Hari, Hari,
such is the mantra of Śiva’ says also the Haripātth quoted by Ch. Vaudeville, op.cit.,
p. 64–65.
128 Caitanyacaritāmrta of Krs.n.adāsa, ādilīlā, 8.22, quoted in E.C. Dimock, The Place of
the Hidden Moon˚ (Chicago ˚ University Press, 1966), p. 226.
129 See A.K. Majumdar, Caitanya. His Life and Doctrine (Bombay: Bharatiya Vidya Bha-
van, 1969), p. 226. Rūpagosvamin, in his Bhaktirasāmrtasindhu, 1.2,48, defines kīrtan
as follows: nāmalīlāgun.ādinām uccair bhās.ā tu kīrtanam,˚ ‘the fact of saying or sing-
ing the name, the play and the virtues or qualities of the Lord’ (ibid., p. 143).
130 On the Naths as they exist now in India, the most recent study is that of Véronique
Bouillier, Itinérance et vie monastique. Les ascètes Nāth Yogīs en Inde contemporaine
(Paris: Editions de la Maison des Sciences de l’Homme, 2008).
131 The Haripath recommends also the silent japa (26.4): ‘Dñyandev has taken the vow
of silence. He wears his rosary in his heart. This is where he invokes the Name of
Hari’ (ibid., p. 146). This, however, does not prevent the Haripāth from recommend-
ing mainly the collective singing of the Name: ‘the [only] means of salvation is the
company of the saints’ (5.4).
132 to karī tetulī pūjā/to kalpi to japi majhā /… /samādhi majhī // This quotation, like all
cited here on the subject is taken from Ch. Vaudeville’s study of the Haripāth, quoted
in note 124.
133 See Ch. Vaudeville, Kabir Granthāvali (Doha), (Pondichéry: Institut Français d’Indo-
logie, 1957), pp. XVIII and 36.
134 Ibid., pp. XVII–XVIII. See also passages from the Gorakhbāni, the sayings of
Goraknāth quoted by Ch. Vaudeville in the introduction to her Kabir (Oxford, 1974),
p. 138: ‘Get hold of the śabda, O Avadhūt ! Get hold of the Śabda! The ‘stages’
(sthāna) are useless obstacles.’
135 This was, for instance, the belief of the Rāmanandins, whose mantra is OM . Rāmāya
namah., or merely rāmāya namah., for OM . is considered as the equivalent of Rāma. On
this and on Tantric elements in Kabir, see Ch. Vaudeville, op.cit., especially section V
of the Introduction, ‘Tantric concepts in Kabir’s verses’.
136 One may mention here the persisting use of sumiran, mental japa (mānasajapa), in the
contemporary sect of the Radhasoamis: a modern variant of a very ancient tradition.
On the Radhasoamis, see H. H. Sahabji Maharaj Sir Anand Sarup, Yathārtha Prakāśa
(Dayalbagh, Agra, Radhasoami Satsang Sabha, 1984).
137 On dhikr, see the interesting study of G.C. Anawati and L. Gardet, Mystique musul-
mane, aspects et tendances, expériences et techniques (Paris: Vrin, 1970), pp. 187–
258, which notes analogies between dhikr and japa.
138 According to Anawati and Gardet, op cit. p. 197, the existence of the practice of dhikr
made in common is not attested before the twelfth century, and could have appeared
under the influence of japa.
139 There are in Bengal Muslim Baùls, the Fakirs.
 Notes 141
140 Their case, but also the whole question of the rapports between Islam and Hinduism,
was studied by Sayid A.A. Rizvi, A History of Sufism in India (New Delhi: Munshiram
Manoharlal, 2 vol., 1978–1983): vol. 1, ch. 6, ‘Interactions between Medieval Hindu
Mystic traditions and Sufism’, vol. 2, ch. 8, ‘The Sufi Response to Hinduism’. Other
more recent studies on the subject could be quoted; for instance Thomas Dahnhardt,
Change and Continuity in Indian Sūfism (New Delhi: D.K. Printworld, 2002).
An interesting case of syncretism is that of the Ismaelis, see T. Kassam and Fr.
Mallison, Ginans. Texts and Contexts. Essays on Ismaeli Hymns from South Asia in
Honor of Zawahir Moir (New Delhi: Matrix Publishing, 2007); or Dominique Sila
Khan, Conversions and Shifting Identities. Ramdev Pir and the Ismaelis of Rajasthan
(New Delhi: Manohar, 1997).
141 A paragraph on the subject in the original French version of this article is not translated
here.
142 But what is a prayer strictly speaking, if not what one uses to consider as such in
the traditionally Christian Western world? Not a generally valid definition, therefore.
Using the word prayer for japa cannot but lead to misunderstandings.

4 Nyāsa: the ritual placing of mantras

1 This is an English version by Gavin Flood of the article ‘nyāsa: l’imposition des man-
tras’, published in the BEFEO, vol. LXVII, 1980, pp. 62–102. The original text was,
where necessary, corrected and updated by the author.
2 We also find the terms vi-ny-AS: ‘to put or place down in different places, spread out,
distribute, arrange …; vinyāsa (masc.), putting or placing down … arrangement, dis-
position, order …, scattering, spreading out, putting together’ (Monier-Williams Dict.
– we also need to retain the nuances of these senses).
3 The Brāhman.a and the Kalpasūtra testify to the ancient use of ritual touching: we meet
them quite often in the Śatapathabrāhman.a, accompanied with mantra recitation (e.g.
ŚB I.1,2, 10–11; I.2, 2, 4–11; III.2, 1, 4–6 etc.). The word used is abhi-MR.Ś, that is, to
touch.
We note that it is the more recent smr.ti texts which prescribe nyāsa, although certain
authors condemn their use as non-vedic (avaidika): cf. Kane History of Dharmaśastra
vol. 2, p. 319. From a certain period the use is general.
4 The nyāsas have not, to my knowledge, been the object of study in any European
language except for Aghehananda Bharati ‘Symbolik der Berührung in der hinduis-
tisch-buddhistischen Vorstellungswelt’, in Studium Generale, vol XVII, no. 10, 1964,
Heidelberg, pp. 609–620. This article which places the practice in a general anthropo-
logical perspective is somewhat brief about the nyāsas themselves.
5 See, for example, ‘hand’ in Hasting’s Encyclopaedia of Religion and Ethics. For India
we can observe that hand gestures have a protective value and are used for consecra-
tion or homage, as witnessed on ancient Buddhist monuments down to our own day.
On this point see the study by Ph. Vogel The Sign of the Spread Hand or ‘five finger
.
token’ (pañcāngulika) in Pali literature (Amsterdam, 1919). See also J. Auboyer Le
Trône et son Symbolisme dans l’Inde ancienne, pp. 11–15.
6 Cf. my Vâc. The Concept of the Word in Selected Hindu Tantras (Albany: SUNY
Press, 1990).
.
7 On the karanyāsa (or karānganyāsa), see later p. 70.
8 This is a natural connection insofar as mantras are a variety of speech-act, and that
speech is bound to gesture. If this form of speech which is mantra can be regarded
as relevant to language, it is naturally also bound to gesture since in many respects
language is in gesture as much as in speech. Mantra and mudrā are systems of coor-
dinated signs. It could be interesting to make a semiotic study of them, without losing
sight of the fact that both are grounded in a deeper level of meaning to which they
merely give an expression.
142 Notes
9 It is a fact that tantric pūjā, yoga, etc. are complex acts whose sense and scope are at
the same time human, cosmic and divine, but which one can also consider in some
cases or to a certain extent as taking place entirely on the divine plane, the gestures and
operations accomplished by the officiant or adept being then only figures. (Unless it
is the sacrificial act which supports the worlds – an ancient brahmanical idea that pūjā
has not entirely abandoned.)
10 We find this in a number of passages of śruti (from Veda to Upanis.ads). Tara Michaël
has underlined the fact in the introduction to her translation of the Hat.hayogapradīpikā
(Fayard, 1974), pp. 65f.
11 We will further see how nyāsa can be used to impose a mantra which symbolises the
supreme brahman: in this case one goes (or one tries to go) directly from the empirical
plane to pure transcendence.
12 Tara Michaël, op cit. p. 73.
13 On the practices associated with prān.a and nyāsa, see below. Prān.a forms somehow
a bond between the subtle and gross body of man.
14 On the nature of mantras, see the Introduction. See too Vāc, quoted above, note 6.
15 Pārameśvarasam . hitā, Kriyākhan.d.a (ed. U.V. Govindacarya, Śri Vilasam Press, Sriran-
gam, 1953). The text is essentially on ritual.
16 Ibid. śl. 4: vyāparo mānaso hyes.a nyāsākhyo yadyapi smr.tah. /
na badhnati sthitim . samyak tathā ‘pi kriyayā vinā //
17 See, for example, SP1, II, śl. 2–3 on the Sūryapūjā. ‘One must imagine oneself trans-
formed into the sun’ etc. ātmānam . ravirūpen.a sam
. cintyā (vol. 1, pp. 72–73). These
operations are described more clearly in chapter 15 of Abhinavagupta’s Tantrāloka,
summarised below (pp. 60–63).
.
18 The level of manas and prān.a in the sense of vital breath is that of the lingaśarīra,
composed of three ‘envelopes’: the prān.amaya, manomaya, and ānandamaya-kośas.
Cf. below, note 35.
19 TĀ 15, 159: nyāsakramen.a śivatāladātmyam adhiśerate. The sacrificer, the sacrifice
etc. come to identify with the condition of Śiva by the process of nyāsa, that is, by the
execution of the ritual procedures.
20 Or at least after a ritual activity which presupposes a certain concentration.
21 Cf. above p. 56.
22 We see here that if India does not make a clear break between body and spirit, the
nyāsa is a means to integrate these two elements.
23 For example, TĀ 15, 295f. summarised below pp. 60–63. One can also refer to the
third chapter of the YH, where śl. 83 indicates that the adept, who has performed the
nyāsa of the śrīcakra on his body, must consider the goddess as inseparably united to
himself (svābhedena vicintayet); see below p. 64).
24 There are cases, however, where meditation, evocation, or mental representation,
appear distinct from placing itself. The texts can, besides, designate or not as nyāsa
the operations they prescribe.
We should recall that the nyāsa, as with japa, can be considered as also having
in certain cases (sometimes above all) a psychological aspect: to turn the attention
towards the divinity rather than towards the actions one is performing – such is the
case, for instance, for the nyāsas performed during the daily ritual bath, etc.
25 Notably by using as ink particular substances such as orpiment, sandal and so on, or by
using a particular script. This, we especially meet in Chinese and Japanese Buddhism
where the bījas are drawn in so-called siddha characters according to a peculiar technique
which links the writing of the letters to mental concentration and control of the breath. Cf.
Shuji-shu, a collection of siddham bījas, the original work of Chozen Bikshu, published
by Matsumo Etasu, Koyasan, 1978. Here it is the user of the bījas who draws them.
But the adept can also use a yantra made by an other who has already written the
bījas or the whole mātr.kā, so that it is enough to touch them during the ritual so as to
perform the prescribed placing.
 Notes 143
26 In his translation of these ślokas, M.N. Dutt writes that these entities ‘should be
contemplated as written in fire inside the different parts of the body,’ which has no
connection with the Sanskrit texts but gives a possible solution to the problem. One
practitioner of nyāsa I consulted on this point told me that in the mātr.kānyāsa ‘one
touches the concerned part of the body with the mahāmudrā (joining the ring finger
with the thumb). While touching, one has to meditate that the letter is there in blazing
form.’
27 The Tantrāloka of Abhinavagupta, with the commentary of Rajanaka Jayaratha,
Kashmir Series of Texts and Studies, Bombay, 1918–1938, 12 vols. of Sanskrit text.
Yoginīhr.dayam Amrtanandayogikrtadîpikâ bhâvânuvâdena ca sahitam,Varanasi,
Motilal Banarsidass, 1989. This edition by Vrajavallabha Dviveda is the only reliable
one of this text.
28 Abhinavagupta, in these two chapters, follows a more ancient text, the Mālinīvijayottara
Tantra (MVT), a work held in special esteem by Kashmiri Śaivism, in which chapters
6 and 8 explain, briefly, the same practices intended for initiation to attain supernatural
powers and salvation.
29 The samayadīks.ā is the less elevated of the initiations. After this comes the putraka-
(or nirvān.a-) dīks.ā, then that of the sādhaka and finally that of the ācārya. On the
subject, refer to the entries on dīks.ā of vol. 3 of the TAK. See also the article by H.
Brunner ‘Le sādhaka, personnage oublié de l’Inde du sud’, Journal Asiatique, 1975.
30 Note added 2011: on this alphabetical order, see Somdev Vasudeva, ‘Synaesthetic
Iconography: the Nâdiphântakrama’ in D. Goodall and A. Padoux, eds., Tantric Stud-
ies in Memory of Hélène Brunner (Pondicherry, 2007), pp. 517–550.
31 One sees here that a mantra, the sound aspect or form of the deity (and in this respect
more essential than the deity) can also be considered – meditated upon – as a visible,
extant, form: that of the deity itself.
The astra is the weapon of Śiva (his cult is described in the same chapter 15 of the
TĀ, to śl. 378f.) – cf. infra p. 39, n. 77.
We note that a mantra can be meditated, conceived visibly in spirit, not only in the
more or less anthropomorphic form of the deity or one of his attributes, but also in a
written form as luminous letters, or else only as a coloured light or a luminous disc
(this is especially the case in so-called Tantric Buddhism).
32 He will activate here a non-visual, phonic image which will be the subtle form of Śiva.
All the gods have in principle a mūrtimantra. This is not confined to Śaivism.
33 See below p. 63.
34 Here it is the mantra itself which is the god. This is a phonic form of Śiva – but
Navātman has also an anthropomorphic aspect, as a beautiful young man. It is the
same for the mantra gods Bhairavasadbhāva and Ratiśekhara as, accompanied by their
energy, for Mātr. sadhbhāva etc; for all deities, in fact; tantric pantheons are pantheons
of mantras rather than of deities.
35 Situated precisely at the level of the energy Parāparā and forming the energy (śākta)
aspect of the phonemes. See Vâc, pp. 320ff. and above, note 30.
36 Pure, says Abhinavagupta, because born from the mūrtimantra which appears (ibid.,
śl. 238 and 261–262) when consciousness is immersed in the pure Reality.
On the emanation within Siva, see chapter 3 of the TÂ and Vâc, pp. 306ff.
A purely spiritual and metaphysical interpretation of a nyāsa is given in the PTV,
p. 20, a passage commenting śl. 27 of this text attributed to Abhinavagupta.
37 Tathā śivo’ham . nānyo’ smītyevam . bhāvayatas tathā / etad evocyate dard.hyam
vimarśahr.dayān.gamam // 270 // śivaikātmyavikalpaughadvārikā nirvikalpatā /
38 This nyāsa is described in śl. 295–312 of chapter 15, which develops the information
of the eighth chapter, śl. 54f. of the MVT. They are part of a section on the ‘mental
sacrifice’ (manoyāga) which is to take place before the ‘external rites’.
39 The SP 1 (vol 1. p. 154f.) describes a cult of the throne of Śiva where the same series of
energies and entities are invoked to make the whole cosmic manifestation the throne
144 Notes
of Śiva. Here, the cult being mental, the adept is at the same time the seat of the divine,
the divine itself, and the worshipper.
40 But laughing, conscious, his body shining like ‘ten million fires of time’ (kālāgni):
MVT 8, 68, SvT 2.81.
41 Unmanā, for the Śaiva traditions, is the transcendental Void, śunyatiśunya, the supreme
principle without a second (nirābhāsam . param. tattvam anuttamam, NT 12, 23).
42 On this installation of the trident in the body for the initiation, see A. Sanderson, ‘Man-
dala and Agamic identity in the Trika’, in A. Padoux, ed., Mantras et diagrammes
rituels dans l’hindouisme (Paris: Editions du CNRS, 1986), pp. 169–205.
43 A procedure analogous to the purification of the adhvans is described in chapters 84
and 85 of the Agni Purān.a dealing with the nirvān.adiks.ā, chapters which are Śaivite.
As described there, these rites are related to the process, in fact general in Hinduism,
of the bhūta- (or deha-)śuddhi, where the elements constituting the body and cosmos
are reabsorbed the one in the other up to where only pure divinity exists (for exam-
ple, the LT. chapters 30 and 54: bhūtaśuddhi; Jayākhya Sam . hitā chapter 10; Kālika-
Purān.a chapters 54–55, etc.). The operation is essentially mental: it is through a medi-
tation based on the image of the body that the elements are reabsorbed; but insofar as
the process is accompanied by mantras placed on different parts of the body, nyāsas
can play a role. On the process of bhūtaśuddhi, see the article by Gavin Flood, ‘The
purification of the body in Tantric ritual representation’, Indo-Iranian Journal, 45,
2002, pp. 25–43.
The complex nyāsas are well represented in the texts, thus there is the mahā-
s.od.hānyāsa of the fourth chapter of the Kulārn.avatantra, or the nyāsa described in
the second chapter of SvT or in the eleventh chapter of the GT. The chapter 25 of the
AgPur also provides an example.
The Parameśvara Sam . hitā (4, śl. 53, and chapter 54) describe a purification (śuddhi)
of the body by nyāsas. On the system of the six adhvan, see Vâc, pp. 330–371.
44 One may note the frequent occurrence of the sixfold nyāsa in the Śaiva and Śākta
texts. The s.od.hānyāsa (gan.eśa, graha, naks.atra, yoginī, rāśi, pīt.ha) is described for
example in the Vīra-tantra, the GT (chapter 10) and the Mantramahodadhi (11.48). It
involves imposing six forms or aspects (rūpa) of the goddess. Cf. VMT I.1.
The TBhS 5, p. 163f. gives variants. The KT (chapter 4) has a different s.od.hānyāsa,
the same being given by the Prapañcasārasārasan.graha (chapter 8).
But why the number six? The kara-, an.ga- and karān.ganyāsa which are the most
current, are encountered in all traditions (see below p. 70) are also six.
45 The GT chapter 10 describes a sixfold nyāsa of the śrīvidyā composed of the six same
same series of rūpa of the goddess. But this is a modern work that borrows entire
passages from other texts, notably the YH. The introduction of this text published in
Srinagar in 1934 is, however, interesting because it gives a table of different nyāsas
with their mantra, bīja and phonemes, the points of placing and so on. This edition has
run out and is difficult to find.
The Nityotsava (G.O.S. 1923, vol. 23) describes in a detailed way a laghus.od.hanyāsa
(p. 98) which is made on the fingers, the palms and the back of the hands. It follows the
complete s.od.hānyāsa (pp. 98–109) made on the entire body and where the placings
are always preceded by dhyāna, that is, a mental visualisation of the visible aspect,
described by the texts, of the imposed entities. The same text furthermore expounds
(pp. 109–117), under the name mahās.od.hānyāsa, a placing on the body of the officiant
of the śrīcakra.
On the placing of the pīt.hās see below p. 74.
46 In all this ritual, it is in effect the cakra itself that is imposed (in chapter 2, śl. 6–12,
pp. 93–97, of the same text, it is the nyāsa of nine divinities presiding over nine parts
of the cakra – the cakreśvarī – which is prescribed). These mantras therefore invoke
parts of the cakra, and not the deities which reside there. Sometimes, however, in
operations of this kind, the mantras used refer both to the cakra imposed and to the
 Notes 145
deity concerned. It also often quotes the part of the body on which a placing has been
made.
47 This is only effective in the case of an adept very advanced in spiritual life: for the
majority of adepts the identification remains only in the realm of representation – an
observation that applies to all rites of identification described here or elsewhere.
48 There are many descriptions of the imposition of a diagram. For the śrīcakra we may
quote the VM 1.112, or the Nityotsava, pp. 109–117 of the GOS edition, where the
adept imposes successively the deities of the cakra on his body meditatively per-
ceived as identical with the śricakra, as having the same shape: śarīram . cintayed adau
nijam . śrīcakrarūpam. Jayaratha says similarly in his commentary on the VM, p. 60:
evam tripurīkr.tavigrahah. sādhakah..
As an instance of contemporary practices, we can quote (among many others) the
ritual booklet Śrīvidyānityārcana (ed. Kalyan Mandir, Prayag, 1967). It prescribes
two impositions of the cakra during the cult: first external (bahiścakranyāsa) on the
body from feet to head; then internal (antaścakranyāsa) on the cakras which is purely
mental.
49 In the GT, chapter 9, which is the first of the section on nyāsa of this tantra, begins with
a purification of the hands (karaśuddhi) made with the mahāvidyā AM . ĀM . SAUH. (śl
3–4). After this the adept purifies his fingertips with the same mantra and a red flower
which he then throws away. The .rs.yādinyāsa of the mantra (see below p. 68f.) and all
the rest of the ritual are described afterwards.
In the Mantramahārn.ava, a vast collection of Śaiva and Vais.n.ava mantras which
describes worship and practice, the .rs.yādinyāsa – which places as it were the prelimi-
nary mantras on the body of the celebrant – comes first, but it is immediately followed
by the karān.ganyāsa preceding the worship itself.
50 That is, makes full or complete (sakala), or which provides the body of the officiant
with parts (kalā). This is another name for the karān.ganyāsa. This can be found in
a Sanskrit text quoted by Aghoraśiva, with a French translation, in appendix 1 (pp.
322–325) of the first volume of the edition of the SP by Hélène Brunner.
In a general manner, the rite of the sakalīkaran.a consists of imposing the an.gas of
the deity on the body of the disciple to divinise it and to create, in the course of the
pūjā, the body with its diverse parts (sān.gam) as the deity to be worshipped.
51 On the different meanings of this term, and especially on the notion of portions or parts
of a deity, see the long entry on kalâ of the TAK, vol. 2.
52 Twelve is, in fact, a solar figure. It is the number of kalās of the sun, whereas sixteen
is the number of kalās or digits of the moon.
53 Pārameśvara-sam . hitā 4.5–22 (p. 18): I have translated the passage, simplifying it
a little. The JayS 11.10–22 (pp. 94–95) prescribes practically the same nyāsa: from
the mūla and mūrtimantra, then the an.gamantras hr.daya etc. and the attributes and
aspects of Vis.n.u. The three final śloka are analogous to those which end the passage
translated here. The LT 25.61–63 (pp. 135–136) prescribes nyāsa first on the fingers
with five śakti, then with the an.gamantras, and following with the placing of the an.gas
of Vis.n.u on the palms. See also Agastyasam . hita 2, chapter 12 etc. In the Śaiva domain,
one can cite the Tantrasāra of Kr. s.n.ānanda, p. 74 etc, A karaśuddhi made with the
mūlavidyā of the Goddess is also prescribed before the pūjā in the VMT 1.109 (p. 57).
A mūrtinyāsa, distinct from the karanyāsa, is prescribed in the JayS. 20.180–181 (p.
212).
The placings can be of deities, of aspects of a god or, in the case of a mantra, of
different words of the mantra etc. One can also place serially groups of phonemes
of the mātr.kā, this is one of the forms of the mātr.kānyāsa. The procedures differ
according to the schools, traditions etc; it does not seem, at first sight, that they can
be reduced to a general rule. There is even in the Mantramahārn.ava 1.6 (folio 95f.) a
mantra of Tryambaka of a vedic type: Om . tryambakam yajāmahe sugandhim .
pus..tivardanam urvarukamiva bandhanānmr.tyor kuks.īya māmr.tāt, of which the
146 Notes
.
karanyāsa is made in the following way: om . tryambakam aśngus..thabhyām. namah.,
yajamahe tarjanibhyām . namah., sugandhim . , pus..tivardhanam . madhyamabhyām
. namah.
etc. up to: māmr.tāt karatalabhyām . namah..
54 TĀ 15.456–459 (vol. 9 pp. 230–231).
55 SP2, pp. 52–53. The text is brief, a single śloka, but Hélène Brunner explains it in a
note, referring to the Purusam. hitā itself cited in the commentary of the Mrg, kriyāpāda
7.14, which mentions the rite. One washes the right hand, anoints it, imposes the brah-
mamantra on it, then one draws, with the thumb of the left hand on the palm of the
right, a man.d.ala where one places Śiva, to which one pays homage. See also SP3,
p. 289. or the Suprabhedāgama, chapter 4 analysed by Brunner (Journal Asiatique,
1967).
The Buddhist hastapūjā (Finot, ‘Manuscrits sanskrits de Sādhana,’ Journal Asia-
tique, 1934), includes the nyāsa of bījas, mentally placing them on the hand together
with a lotus and divinities; this is a rite akin but different. The hastābhimantran.a
mentioned in the Adikarmapradīpika, edited by L. de La Vallée Poussin (Études et
Matériaux, pp. 192 and 217) appears, on the contrary, as a hastanyāsa, although the
word nyāsa is not used: hastam abhimantrya [mantram . ] vibhāvya.
56 In the sense, it is explained, of being opposed (vāma) to transmigration (śl. 278) and
of secrecy (śl. 279) as is normally the case of the practices of the so-called left hand,
in particular in the Kaula schools.
57 A gesture one sees often made by Indians outside of a ritual context. Generally, the
positions or movements of the hands of Indians differ from those observable in West-
ern societies.
58 According to the commentary of the SP1 – III.41 (vol. 1, pp. 142–143 note), the
āsanamantra involves putting a flower on the throne which is thus consecrated. This
resembles the Nityotsava (nyāsaprakaranam, p. 95) where a triple āsana must be
imposed with flowers on the śrīcakra.
The GT as cited in Principles of Tantra, p. 158f., seems to have a rite of installing
a devatā in a yantra by means of flowers to which the adept has transmitted divine
energy with his breath and the vāyubīja YAM ..
59 See, for example, Dale Saunders Mudrā (New York: Pantheon Books, 1960), or
Stephen Beyer The Cult of Tārā (Los Angeles: University of California Press,
1973).
60 The SP, for example, prescribes a number of mudrās, and in the first volume of
Hélène Brunner’s edition there is an appendix of descriptions and photographs. One
can also cite āgamas and sam . hitās etc. In volume V of the History of Dharmaśāstra
(pp. 1123–1131), P.V. Kane presents a brief study with an abundant – but still far from
complete – Sanskrit bibliography. For Buddhism, one must add to India, Tibet, China
and Japan.
61 I have not checked their provenance. There are several texts, from the non-dualist
Śaiva Krama tradition, of that name.
62 As the royal seal is a symbol of authority, the mudrā symbolises the power of the
divinity; or as the seal authenticates a contract, the mudrā seals a pact between the
worshipper and the worshipped. Cf. Dale Saunders, ‘Symbolic Gestures in Buddhism’
Artibus Asiae vol. 21, 1957.
63 In the context of Buddhism, a justification of the mantra-mudrā association has some-
times been based on the theory of the three bodies, vāk-kaya-citta: these being three
coordinated and complementary aspects of an ensemble.
64 Bhāvayati, causative of BHU, means to make become, give birth, but also to conse-
crate, to control and to have in mind, to consider, and to know. The word bhāvanā (f.),
derived from BHU, designates intense, mental concentration, the creative power and
efficacy of imagination, and mystical realisation (in the ‘way of energy’, the śāktopāya
according to the TA). On bhāvanā, François Chenet, ‘Bhāvanā et créativité de la
conscience’ (Numen, vol. xxxiv, 1987, pp. 45–96).
 Notes 147
65 In this I entirely disagree with Frits Staal’s position: see his Ritual and Mantras. Rules
without Meaning (Delhi: Motilal Banarsidass, 1996).
66 We know that the word mudrā has several senses, one of which is that of a mysti-
cal attitude; mudrā is thus the seal of the ineffable, purely spiritual; in this sense see
Mahārthamañjarī (MM), cited in the next note; see also TĀ 32.1–3 – on which A.
Padoux, ‘The body in tantric ritual: the case of the mudrās’, in Teun Goudriaan, ed.,
The Sanskrit Tradition and Tantrism (Leiden: Brill, 1990).
67 Avikalpatayā marśa vikalpavargasyān.gasannāhah. / Arghyam . vedyavilāsah. pus.pān.i
svabhāvapos.akā bhāvāh.. The reference is: Maheśvarānandapran.itā mahārthamañjarī
svopajñānaparimalākhyavyākhyopetā (Varanasi: Vārān.aseya, Sanskrit Viśvavidyālaya,
1972), pp. 112–113.
68 These elements are not always mentioned. The .rs.i, the chandas, and the devatā are
indispensable and their mention is ancient (along with the viniyoga), dating back to the
anukraman.i of the Veda. The other elements are Tantric and they do not always appear;
they cannot therefore be considered as necessary elements of all Tantric mantras.
The .rs.i is, of course, entirely fictitious; it is in fact sometimes a deity. As for the
chandas, it can be real if the mantra is of Vedic origin or if the Tantric mantra’s struc-
ture corresponds to a certain meter. But the non-Vedic mantras are often formed from
a group of short syllables (often unpronounceable) or even consist in one bīja, such
as HRĪM . or SAUH., or the fifty varn.as of the mātr.kā, which do not, strictly speaking,
have a meter. There are, however, lists of .rs.i and chandas for each of the fifty varn.as.
In other cases, several meters are given. This mention is therefore preserved in many
cases only as a kind of reference to the Vedic tradition. The reference is real when
elements of a Tantric mantra are derived from a Vedic text, in which case these will
be given the same meter as the Vedic passages from where they come. Otherwise, the
reference seems to be purely theoretical; although, for contemporary followers, the
chandas appears to expresses, if not really a meter, at least a number, an inner rhythm
particular to the mantra: for such adepts, this is nothing theoretical.
It sometimes happens that the bīja is formed from one or from several words, which
is contradictory. The śakti is, in theory, the central part of the mantra and is believed
to concentrate in itself its energy; its place in the mantra is in fact variable. As for the
kīlaka, a pin, bolt or wedge, it should normally be at the end of the mantra as in an
arrow so as to pierce its target, but its place and form also varies.
Another element, the kavaca (in the sense of amulet or protective formula) is some-
times added.
69 This is the principle, valid even if this transformation is generally only theoretical and
imagined.
70 We see here the usual dialectic of the tantric pūjā. deva eva yajed devam, nādevo
devam arcayet, that one finds constantly and which also explains the purifications
during the pūjā as well as the succession of deifying nyāsas (those which we have pre-
viously seen for example). One must be already deified to be permitted to ritually wor-
ship a deity by a cult which is deemed to lead to deification, and one must be already
‘mantricised’ in order to use and assimilate more completely a mantra. It is necessary
to have, in some measure, at the beginning that towards which one strives.
Should we not see there a sign of what is, it seems to me, a well-pronounced intel-
lectual tendency in India: a tendency not to admit (or a difficulty to conceive) a first
beginning or a progress which would be a new advent: before starting, things are as it
were already there. This is the view, so widespread, of the satkāryavāda, for which the
effect pre-exists in the cause or, for cosmology, of all cyclical cosmologies.
Of course, insofar as the pūjā continues – and preserves certain elements of – the
old, brahmanical sacrifice, the preliminary divinisation of the tantric officiant can be
regarded as continuing, under a different form, the rites by which the Vedic sacrificer
and the victim were preliminarily consecrated in order to enter into communication
with the gods in the sacrificial area. Perhaps we should add that, in the course of the
148 Notes
ritual, the nyāsas seem to have only a temporary role and effectiveness: in each part of
the rite, placings are done that are cancelled by those which follow, or by the ācamana
which separates them (an observation that I owe to Hélène Brunner). The cancellation
of a rite by the following one(s) is also one of the reasons of the redundancy so char-
acteristic of Hindu rites.
71 This according to the Mantramahārn.ava 1.6 (folio 91).
Some texts specify the mudrā to be used for these placings. Thus in the
Śārasan.graha cited in the TBhS, 5, p. 70: the placings of the .rs.i, chandas and devatā
are done with four fingers excluding the thumb – the same way for the bīja and the
śakti. But the kīlaka, according to another author, is imposed with both hands on all the
body: it is the same if, as is generally the case, the viniyoga finishes the series.
72 For a better view of the mūdras shown during a pūjā, see for instance G. Bühnemann,
The Worship of Mahāganapati according to the Nityotsava (Witrach: Institut für Ind-
ologie, 1988); or Richard H. Davis, Ritual in an Oscillating Universe. Worshipping
Siva in Medieval India, Princeton: Princeton University Press, 1991).
A small modern treatise, the Yoga-mantra-sam . hitā, explains: the .rs.i who has ‘seen’
the mantra is for the adept as his master: that mantra is placed on the head with respect;
the meter being made of words and of syllables is placed on the mouth; the deity,
residing in the heart of all creatures, must be placed in the heart; and the seed on the
genitals.
73 See the translation by Brunner (SP vol. 1, p. 132). Here is the text:
Hr.dayādikarāntes.u kanis..thādyangulīs.u ca / hr.dādimantravinyāsah. sakalīkaran.am .
matam //
This nyāsa is also called s.ad.an.ganyāsa, hr.dayādinyāsa, or hr.dayādis.ad.an.ganyāsa
and, for the hands, angus..thādis.ad.an.ganyāsa.
The adept must make the an.gamantras according to the rules at the same time as he
places them, thanks to which he deifies his hands and body. Sometimes netra does not
appear among the an.ga, which are then only five: this is the pañcān.ganyāsa prescribed
in various texts – thus the Nāradapañcaratra 3.3,10–11.
74 There are also secondary aspects called upān.ga: accessory limbs which are placed
after the first ones.
75 They are also sometimes represented as separate deities that one ritually places on or
around the image being used for worship – thus SP1 (vol. 1 pp. 208–210). On the an.ga
and their symbolism in Śaivism see SP1, introduction, p. xxxiv and pp. 194–197), also
vol. 3, pp. 400–404 (note 440), and Hélène Brunner, ‘Les membres de Śiva’, Asia-
tische Studien/Etudes Asiatiques, XL.2 (1986), pp. 89–132.
76 In the Śaiva tradition, this is the third eye of Śiva, that of consciousness, situated on the
forehead between the two others eyes: the word netra in the an.gamantra netra is in the
singular (for the third eye) or in the plural (for the three eyes), but never in the dual.
77 The consonants which serve to form these bījas vary according to the deity, the school
etc. But the vocalic element is stereotyped – sad.dhīrghair ādyādyabījena karān.-
ganyāsam says the Phetkārin.itantra, as cited in the TBhS (p. 170). One forms the
bījas on the basis of the bīja or of the principal vidyā (ādyabījena) of the mantra while
associating it, in due order, with the six long vowels Ā etc. (s.ad.dīrghair ādi…), in fact
with ā, ī, ū, ai, au, and ah.. One will thus have, starting from the vidyā SAUH . : sām
. , sīm
.,
sūm. , saim, sah.; starting from HR.ĪM . : hrām etc.
The vaktra mantras, on the other hand, are done with short vowels. The same
instructions are in the Agni Purān.a 25, 5 fl. (Vais.n.ava chapter); the same (for only the
an.gamantra) in the JayS 6.105ff. etc.
78 In the Pāñcarātra tradition, the LT (ch. 33) gives the an.gas of the tārikāvidyā of Tārā
used in the cult of Laks.mī: one adds to the an.ga a term (jñāna, aiśvarya, śakti, bala,
tejas) whose invocation carries to the adept, through nyāsa, each of the qualities enu-
merated – the an.gas are of the model Om . hrām. jñānāya hr.dayāya namah. (we note
that the an.gamantra for netra is netrabhyām, the two eyes, instead of the three eyes of
 Notes 149
Śaivism). The commentary on the Śaiva ST 18.6 (p. 713) associates in the same way
to the six an.gas the enunciation of the six qualities (gun.a) or infinite glories of Śiva,
omniscience, etc. The SP1 quoted above associates the an.gamantras with five other
attributes of Śiva (the netra is sometimes omitted – cf. above note 68.
79 Phat. is usually called the astra (or heti) mantra – it is the weapon, the missile, espe-
cially the arm of Agni. It shines, says the TĀ (15.232), as the flame of the fire of time,
destroying and purifying the gross and subtle body of the adept. The same designation
of phat. occurs in Buddhism – cf. Finot, ‘Manuscrits sanskrits de sādhanas’, JAs. 1934,
p. 60: phat. phat. sarvān.ges.vastram … arpayet.
80 The rule often allows exceptions in certain traditions. See, for example, the notes of
N.R. Bhatt in his edition of the Raurāvāgama (I.1.4, vol. 1, p. 18); see also the com-
mentary of the Ham . sopanis.ad by Upanis.adbrahmayogin (ed. Adyar, p. 566). There
may also exist variants in the case of optional rites (kāmyakarmān.i).
81 Preceded sometimes by a bīja formed as indicated above.
82 This is according to the TBhS (p. 170). But there are other ways of doing it: thus the
Nāradapañcarātra III.3.24. This triple nyāsa has the main sense of symbolising the
threefold activity of the divinity and therefore of placing it with all its cosmic dyna-
mism. But each of the three nyāsas can also be supposed to have a particular effect (cf.
Nāradapañcarātra loc. cit.) or else can be prescribed for different categories of adepts
(cf. Tārin.īpārjāta, p. 23).
83 One finds the an.ga- or karān.ganyāsa described or prescribed as well in texts other than
those presented above or in the preceding notes. For example, the TĀ (chapter 30), the
SvT 1.71–72, the NT 12.28–33, the Kālīkāpurān.a, the JayS 6.105ff. etc.
84 Mantrā mūkatvam āyānti vinyāsena vinā lipeh./
Sarvamantrasiddhyartham . tasmād adau lipim . nyaset// (TBhS p. 159).
A small modern manual (Śrīvidyānityāh.nika) describes the mātr.kā as a great mantra
serving to place the goddess Mātr. kāsarasvatī: śrīmātr.ikāsarasvatīnyāsamahāmantra.
85 ‘The mātr.kā, mother of the gods, origin of all mantras’, says the GT 9.14, in the expo-
sition of nyāsa. And Abhinavagupta in the TĀ 15.130–31: ‘This inner energy born
from Bhairava under the aspect of the mass of sound (śabdarāśi), one calls mātr.kā,
because it is the mother of all that appears’ (TĀ vol. 9, p. 58).
86 The formula used is generally of the type: a group of consonants (varga) framed by
two vowels (one short, one long), these and the others with the bindu: am . , kam
. etc.
+ part of the hand or an.ga + the appropriate jāti (namah. etc.); this allows the plac-
ing of twelve vowels (6 × 2) and all the other phonemes. Cf, for example, Nityotsava
(nyāsaprakaran.am), p. 73.
87 Vais.n.ava Gautamīya cited in TBhS p. 160. The placings on the lotuses are not neces-
sarily made in the order from bottom to top, as in the Daks.inamūrti-sam . hitā cited by
TBhS p. 160. This sam . hitā, which is a text of the Śrīvidyā and not of the Pāñcarātra,
contains in its eighth pat.ala the description of a mātr.ikāpūjā including s.ad.an.ga- and
bahirmātr.kānyāsa, thanks to which, it tells us (śl. 16), the mantrin will become varn.a-
svarūpa, thus identified in his essence with the divine energy of the phonemes. This
pūjā continues with the dhyāna of the mātr.kādevī, then by the drawing of a yantra
where worship will be performed to her.
88 TBhS p. 160. The same in the Nityotsava p. 94: iti dhyatvā manasā pus.pāñjalim .
dattvā mātr.kāh. … svān.ges.u nyaset.
89 TBhS citing the Vais.n.ava-Gautamīya. This description, as the name of the text would
suggest, is a Vais.n.ava work. I have not, however, found in the Vais.n.ava texts I have
consulted a description of the mātr.kānyāsa being exactly the same as that of the Śaiva
texts I quote here, which is more complex.
90 Citations of another tantra (tantrāntara) in the TBhS 5, p. 159.
91 The prescriptions are varied. But often the vowels are placed on the face and head,
the consonants on the body, and the semi-vowels, spirants and KS.A on the heart,
associating them there with the elements that constitute the body, the breath, and the
150 Notes
self. Thus the GT 9.25–31; AgPur, chapter 293, 37–39; lipinyāsa; Tārin.īpārijāta,
pp. 18–19 etc.
92 TBhS, pp. 160–162.
93 Cf Vāc pp. 320–327, based on TĀ 15.120–142 – vol. 9, pp. 62–74. The mālinī is
also mentioned in other, earlier tantric works, such as the Siddhayogesvarimata, the
Tantrasadbhâva, the Satsāhasrasamhitā, etc.
On the subject I refer to Somdev Vasudeva’s interesting and novel study, ‘Synaes-
thetic Iconography: 1. the Nâdiphântakrama’, in D. Goodall and A. Padoux, eds.,
Mélanges tantriques à la mémoire d’Hélène Brunner, Pondichéry: Institut Français
de Pondichéry et Ecole Française d’Extrême-Orient, 2007, pp. 517–550. Note that the
above paragraph on mālinīnyāsa was re-drafted in 2009.
94 Cf. the study of D.C. Sirkar, The Śākta Pīt.has (JRASB letters, XIV, 1948, pp.
1–108 and MLBD). It is a presentation and edition of a text, the Pīt.hanirn.aya, or
Mahāpīt.hanirūpan.a.
A special number of the Hindi periodical Kalyān dedicated to pilgrimage (‘Tīrtān.k’,
Gorakhpur n.d.) gives (pp. 513–528) a list of 108 places of the cult of Śakti, then of
the fifty-one pīt.ha of the goddess, with a map and descriptions of the places, followed
by a study of the secret pīt.ha of the goddess: śaktipīt.harahasya.
The TĀ (15.81–98) enumerates the pīt.ha ‘external’ or ‘internal’, that is, situated in
the world or interiorised in different points of the physical or ‘subtle’ body.
95 The system of 108 pīt.ha has in effect been taken over and adapted in the Vais.n.ava
tradition.
We may note, apropos of pīt.has, that some texts (thus the ŚT 2.134 comm; vol. 1,
p. 115) call the sixteen vowels pīt.ha, as contrasted with the consonants then called
ks.etra. We see here the connection between phonemes and space. Jayaratha, in fact,
commenting on the first śloka of the VM where the goddess is described as assuming
six aspects (Ganeśa, rāśi etc.) which are placed in the s.od.hanyāsa, stresses the point
that the pīt.has correspond to extension since the goddess also appears in this form
(deśrūpatayā): ‘The Lord,’ says Abhinavagupta in the ĪPV 2.1,5, ‘manifests spatial
succession by the diversity of forms (mūrtivaicitrya)’; this corresponds to the path of
space (deśādhvan) in the theory of the six ways (TĀ 6.34–36).
The din.nyāsa, the placing of the directions of space, which is a part of several rites,
does not seem to be associated with the same notion since it consists in reciting the
formulas of the nyāsa and showing a mudrā while turning to the eight directions of
space in order to win the favour of the devatās of the directions; this is not really a
nyāsa.
96 We also find a brief description in the Mantramahodadhi which concludes: iti dehamaye
pīt.he dhyayet svābhīs..ta-devatām: ‘he should meditate his own deity on the seat which
has become his body’. Also see Tārin.ipārijāta pp. 34–37, which enumerates the enti-
ties and mantras to be placed; Tārarahasya 3, p. 119; Siddhasiddhāntapaddhati 2.1–9;
or a recent manual Balārcanapaddhati and so on.
97 That is to say, from the twenty-sixth to the thirty-fourth tattva in the system of thirty-
six tattvas of Tantric Hinduism.
98 It is perhaps interesting, from the point of view of the range or psychology of the
nyāsa, to mention here a poem cited in the Principles of Tantra (p. 1142), where the
sixth verse, recalling the unity of the worshipper with the divinity he worships, adds:
‘otherwise, how does Nyāsa on Thy limbs affect Me or Nyāsa on my limbs affect
Thee?’ ‘They are one and the same,’ comments A. Avalon, ‘and therefore when the
sādhaka does Nyāsa on himself, he does Nyāsa on the Devatā also’ (ibid. note 4).
Another point of view, contemporary but undoubtedly exactly traditional in senti-
ment, is that of R.K. Ray Encyclopedia of Yoga (Varanasi, 1975) which, in relation to
pīt.hanyāsa, writes: ‘Nyāsa is useful in effecting the proper distribution of the Śaktis of
the human frame in their proper positions so as to avoid the production of discord and
distractions in worship. The actions of Nyāsa on the body, finger and palms stimulate
 Notes 151
the nerve centres and nerves therein. Thus, for the attainment of that state in which a
Sādhaka feels that the Bhāva (nature, disposition) of the Devatā has come upon him,
Nyāsa is a great auxiliary. It is, as it were, the wearing of jewels on different parts of
the body. The Bījas of the Devatā are the jewels which the Sādhaka places on dif-
ferent parts of his body. By Nyāsa he places his Abhis.t.adevatā (Is.t.a-devatā) in such
parts, and by Vyāpaka-nyāsa (comprehensive Nyāsa) he spreads His presence through
himself and he becomes permeated by Him, losing himself in the divine Self.’ (See
too MNT 5.105 and passim; ŚT 6.1ff; Prapañcasāra Tantra p. 138 and passim; Yoginī
Tantra 2.1.1 and passim).
99 The nyāsa can be made with a flower or blades of grass as said before – see p. 71.
100 SP1 (vol. 1, pp. 124–125): darśayitvāmr.tīmudrām ‘and one shows the mudrā
which transforms into nectar’ (Brunner examines this problem in the introduction,
p. xxxvi).
101 The Tārin.īpārijāta, pp. 18–19, describes, after expounding the bhūtaśuddhi, a
jīvanyāsa made with the prān.apratis..thāmantra of the goddess Tārin.ī along with a
whole series of placings (r.s.yādi, kara, s.ad.an.ga etc.) with dhyāna, japa etc. The struc-
ture of the rites is the same as that of the prān.apratis..thā such as those found in diverse
modern manuals (Śrīkapadruma or some other).
The jīvanyāsa is mentioned in the same way in the MNT (5.105). The adept places
by this rite the divine breaths in his own body: taddehe devyāh. prān.ān nidhāpayet (the
commentary glosses taddehe by tasmin navīne dehe: ‘in this new body’, in this body
renewed, transformed by the bhūtaśuddhi.
The same occurs in SP 1 (vol. 1 pp. 128–133).
102 The vyāpakanyāsa consists usually in passing the two hands several times on the
entire body, from feet to head, while repeating the mūlamantra (see TBhS 5, p. 171).
We have seen that this can also be done differently. But it always consists in the plac-
ing of a mūlamantra, or of a fundamental entity or diagram, on the entire body, which
must be entirely impregnated and penetrated (vyāpta).
Cf. Principles of Tantra p. 1140: ‘and then by means of Vyāpaka, or comprehensive
Nyāsa, to feel the presence of the devatā as one undivided entity whose substance is
Mantra all over one’s body from the feet to the crown of the head’.
103 See, for example, the LT chapter 46: pūjā of the laksmīmantra. Three other chapters
of this text (47, 48, 49) describe the cult of other mantras: the nyāsas are identical.
For the tārikā-mantra, the principal mantra of the Tantra (chapters 42–43), different
nyāsas are prescribed, but this is because there is no ‘external’ worship here but a
sādhana where mental elements, visualisations, japa, and mantrayoga, play an essen-
tial role.
Other Vais.n.ava examples include: the pūjā of Vasudeva-mantra with eleven
aks.ara in the Varśakriyākaumudī pp. 182f.; or the placings of the mūlamantra of
eight and twelve aks.ara in the Sanatkumārasam . hitā I, adhyaya 9 (p. 72f. ed. Adyar,
1969).
104 R.V. Joshi Le rituel de la dévotion kr.s.n.aïte. Pondichéry: Institut Français d’Indologie,
1959 (publication of the IFI, no 17).
105 A short text published, translated and annotated by F. Nowotny in the Indo-Iranian
Journal I. 2, 1957, pp. 109–154.
106 Śrībalārcanapaddhatih … śrī paśupateśvarānandanāthaviracitā, Surat, sam. 1991
(small text of ninety folios).

5 A Hindu rosary ritual

1 The French original version of this paper, ‘Un rituel hindou du rosaire (Jayākhyasam
. hitā,
chapitre 14)’ was published in the Journal Asiatique, CCLXXV, 1987, pp. 115–129.
152 Notes
2 Tantric texts are normally meant for the instruction of initiates, these being gener-
ally sādhakas, adepts looking for rewards or powers (bubhuks.u). On the sādhaka, H.
Brunner’s study, ‘Le sādhaka, personnage oublié du śivaïsme du sud’ (J.As, 1975, pp.
411–443), is still very much worth consulting. The Śaiva sādhaka- or nirvān.a-dīks.
ā is described in vol. 3 of H. Brunner’s edition of the Somaśambhupaddhati – SP
(Pondichéry, 1977).
3 One may of course acquire a fully constituted rosary: this is what all non-initiates – the
majority of devotees – do. But such a rosary must be consecrated before it is used.
4 Perhaps surprisingly in a Vais.n.ava context, the rudrāks.a, the berries of Śiva’s tree, are
considered the best ones.
5 Tantras usually count women and children among those seeking moks.a.
6 Astra, the Weapon mantra, is used in many ritual contexts, being placed (nyāsta) or
‘sprinkled’ (sam . proks.ya).
7 sam . pūjya pus.padhūpādyais tasya śuddhim athā caret //28// dagdham astren.a
sam . cintya varman.ā māruteritam / āplavyamūlamantren.a paramāmrtarūpin.ā //29//.
Astra and Varman are two of the six ‘limbs’ (an.ga) of Vis.n.u (or ˚Śiva). Astra, the
Weapon, is a divine destructive arrow. Varman, or Kavaca, the Cuirass, is protecting:
it safeguards the purity given by the fire of the Astra deemed to burn away all impuri-
ties. Bathing and fanning are two of the usual services (upacāra) offered to a deity. We
may note here the important role of mental representations in the Hindu cult, the pūjā,
which is to a large extent a play of imagination.
8 Caturbhujam . tu virajo nārāyanam ivāparam //30// varadābhayahastam . ca
baddhāñjalidharam . smaret/brahmadvārasthitam . tac ca sūtram. dhyayecchikhopamam
//31//. The mentally conceived divine anthropomorphic figure of the rosary is imag-
ined as bearing it also on its head as is the case of the āyudhapurus.as who represent
symbolically the ‘arms’ of a deity and bear them on their heads.
9 On the ritual for the Tantric worship of a deity, see SP, vol. 1, pp. 90ff.
10 On the uddhāra of mantras, see the first paper, above,
11 In the introduction, p. 37, the editor of the JS gives a shorter form of the mantra:
OM . THAM . aksasutrāya namah..
12 On the an.gas, see H. Brunner’s study: ‘Les membres de Śiva’ in Asiatische Studien/
Etudes Asiatiques, XL. 2. 1986, pp. 89–132.
13 On bhāvanā, see F. Chenet, ‘Bhāvanā et créativité de la conscience’, in Numen, vol.
XXXIV, 1987, pp. 45–96.
14 srs..tikramen.a, since the power of the deity is entering the image of the rosary, a process
˚
analogous to that of the cosmic process of emanation (srs..ti).
15 Yā parā vais.n.avī śaktir abhinnā paramātmanah. //35// ˚ pracalatpūrn.acandrābhā
sūryavatkiran.āvrttā / yugaks.ayograhutabhuktejasā tīvrabrm . hitā //36// prāgvat
srs..tikramen.aiva˚ smared hrtpadmamadhyagām / pūraken.a˚ tu viprendra kumbh-
˚
akena nirodhitām //37// ˚ hrtpadmadmād utthitām . bhūyo brahmarandhrāvadīm .
smaret/brahmarandhrāt tato˚ vipra prollasantīm . śanaiś śanaih. //38// pūrvavat
sandhimārgena cintayed dvādaśāntagām . / recakākhyena yogena tayā sūtram . tu
bhāvayet //39//
16 Prs..the: that is, by the part opposite to the Meru.
˚ that the JS teachings are only meant for members of the three higher varnas.
17 Note
18 nis.kalo mantrarāt tatah. //40// mantravrndasamāyukto magnas tatreti cintayet/
mantrātmā bhagavacchaktis sūtramantrāks ˚ arāni vai //41// sarvam ekīkrtam dhyāyed
. . .
yathā ks.īren.a sodakam / tatra pus.pāñjalau prs.he grhītvā cintayed imam ˚ //42// susam .
srtāks.asūtrasya patantam . ˚ ˚
prs..thato dvija / srst.ikramāt samāyātam . mantram. sakalanis.
˚
kalam //43// sphulin.gan.asan ˚.kāśam ādheyatanutām
˚ . gatam /
19 idam . vijñāpya deveśam aksasūtramayācyuta //44// prayaccha mantrajāpārtham .
smared dattam . ca tena tat / grhītvā śirasā paścāt prasīda iti coccāret //45//.
20 prāptam ˚
. sam. srāpayet paścād arghyapātrāc ca vārin.ā / sam . pūjya pus.padhūpādyair
mantram . tac ca vinyaset //46// sādhāram . sādhanam . caiva śaktipūrvais samāvrttam /
˚
 Notes 153
sannidhau bhava deveśa sanniruddho bhavācyuta //47// sūtrākhye man.ijāle’smin
yāvac candrārkatārakam /.
21 The four stages from thought to action thus correspond to the planes of consciousness,
from turya to jāgrat. The inverse movement is that of the return from discursive wak-
ing state to pure inner consciousness.
22 smaran.a is not mere remembrance or memory. It is a recalling to mind and a mental
concentration on a thought or on a mantra. In the Trika, smaran.a is used to designate
the intuitive consciousness of the highest Reality. In the PTV (p. 245), Abhinavagupta
glosses smarati with anusam . dhatte, that is, one-pointed attention. This sense given to
smarati goes back to the ancient meaning of SMR: on this, see L. Silburn, Instant et
Cause (Paris: Vrin, 1955, pp. 374–346). ˚
.
23 śl. 52b–64: karan.am . vānmanaścaiva mantren.a paramātmanā //52// vis.n.unā śaktirūpen.a
bhāvitam . bhāvayet purā / cetasā yad uparūd.ham . tad yuktam . vastu vāggatam //53// vas-
tuyuktam . ca caitanyam . vāktrtīyā ca nārada / samārohet karmapade vikāratve śanaiś
śanaih. //54// evam ˚
. vivartate mantras turyāj jāgrāvadhi kramāt / prātilomyena vai vipra
punar eva nivartate //55// karan.am . vāksvarūpam . syād vāk ca cidrūpin.ī bhavet / yac
cittam. sa bhaven mantro yo mantras sa tvajo harih. //56// brahmādya jāgratparyantam .
punar atraiva samnayet / vyāpakam . yat param . brahma śaktir nārāyan.ī ca yā //57//
sā hy eva parin.āmena turyākhyam . bhajate padam / turyam . sus.uptatām eti sus.uptam .
svapnatām . vrajet //58// jāgratvam . svapnam āyāti evam . jāgrādaditah. punah. / bhaga-
vacchaktiparyantam . sandhānam . caikatam . smaret //59// yad anityam idam . vipra cit-
ravat parid śyate / bāhyam . vis.ayajālam . jāgrad etad udāhrtam //60// anityapratipattir
yā asminn upari sarvadā / bhan.gure svapnatulyo yas sa svapno ˚ jāgrakāran.am // 61//
sus.uptam . śāntatām . viddhi svapnavrtteh. param . tu yat / śaktirūpasya vai vis.n.oh. prāptas
turyatvam eti sah. //62// tayā saha˚ samatvam . ca turyātītam . tad ucyate / mantrātītam .
param . mantram . sthūlasūks.madvayam . tathā //63// ātmānam . pañcamam . vipra
ekatvenānusam . dhayet / krtvaivam . anusam . dhānam . prārabheta japam . tathā //64//
24 hrtpun.d.arīkamadhyastho˚ bhārūpah. parameśvarah. / nirmalasphat.ikaprakhyah. pras-
˚
phuran yah. svatejasā //65// tasya śabdamayī śaktī (? ?) jvālāvan nismrtam . mahat
(?)/taddharmadharmin.ī śuddhā tasyā vai varn.asam ˚
. tatih. //66// nismrtā mantrajananī
tadagrāc- caiva mantrarāt./nismrtam ˚
. (tu) yathā pus.pam . latāgrān munisattama //67//
evam evāks.asūtre tu uditam ˚
. ca kramam . smaret/hrllayam . tallayīkuryāddhrllayasya ca
tat thatā //68// ekasya mantranāthasya antarbāhyoditasya ˚ ca /. ˚
25 Such a rite for the so-called ‘outer worship’ (bāhyayāga) of the deity is described in
chapter 13 (śl. 100ff.) of the JS. A similar ritual is to be found in the Śaiva context
– (see for instance the SP1 – similitude due to the fact that the Pāñcarātra was strongly
influenced by shaivism.
26 This is carefully described in śl. 73–76a. It appears to be done by a particular move-
ment of the thumb which moves the rosary along the finger (the index, normally) on
which it rests, the counting going on in the same sense. To avoid crossing the Meru,
other texts prescribe counting the beads alternatively in one sense and then in the re-
verse.
27 There are variant lists of these acts, some including nine acts. On the s.at.karmān.I, see
T. Goudriaan’s study, Māyā Divine and Human, already quoted, pp. 251–412.
28 The passage I allude to is as follows: « paradehepraveśe ca tattvānām . preran.e dvija/
krūrakarman.ī siddhyartham . japo jyotirmayas smrtah. // 79// tatrāntarlīnam amalam .
śabdam ˚ . dos.aduhkhaks.ayan.karam //80//
. tu paribhāvayet / bhuktimuktiprasiddhyartham
śāntau tu sarvakāryān.ām . siddhyartham avicāratah. / sukhasaubhāgyasiddhyartham .
tatah. ‘py ādhānakarman.i //81// japo bhavati śabdākhyas tatrāntaras tu vibhāvayet /
niśāmbukan.asan.kāśaprakāśātma janārdana //82// hrtpadmāt tu svamantren.a
bhāvayeccchabdam utthitam / nityoditam ˚
. tu tenaiva prayatnarahiten . a tu //83// pran.ā-
pannapadasthena bhāvanāmis.ritena / japyamānas tu vai mantras sarvakāmaphala
pradah. //84//.
29 In all ritual worship, the deity first installed in the icon by the rite of āvahana so
154 Notes
as to be worshipped is removed from the ritual icon by the visarjana rite, once the
worship is over. The same process is used for the rosary. The mantra being used is:
OM. bhagavanmantramūrte svapadam āsādya ks.amasva ks.amasva OM . namo namah..
30 This mantra is: OM . H.RŪM . VAUM . vis.vaks.enāya namah..
31 sandhānam upasam . hrtya pūrvam . yac cāks.asūtrakam / visarjanākhyamantren.a
mudrāyuktena nārada˚ //90// sūtram . cābhinavam . krtva prāgvat sam. skrtya ves..tayet /
. cācaret punah. ˚//91// sam
yathāsthitam ˚
. mān.inām . ca yojanam . sthāpya vidhivan mantrī
bhūyas tasmim . s tu vinyaset/purā yad āhrtam . caiva sūtracchedabhayāt tatah. //92//
pūjām ˚
. krtvā tathā homam . tatkarmachīdrakāran . am / purān.asūtram āpādya vib-
˚
hajet tantujam. ca tat //93// badhvā pās.ān.akhan.de ca āgādhe‘mbasi niks.ipet / vis-
vaks.en.īyamantren.a ks.iptvā ‘camya smared dharim //94// mūlamanttren.a viprendra
saptavārān samāhitah..

6 On the defects and the perfecting of mantras

1 A paper on the same subject was published as pp. 499–510 of the Pandit N.R. Bhatt
Felicitation Volume, P-S. Filliozat, S.P. Narang and G.P. Bhatta, eds., Delhi, 1994.
2 On this dual aspect of mantras, see Hélène Brunner’s study ‘Mantras et mantras dans
les Tantras śivaïtes’ in Le parole e i marmi. Studi in onore di Raniero Gnoli nel suo 70°
compleano, Raffaele Torella et al., Rome: IIAO, 2001, pp. 183–212.
3 See Introduction, p. 95, and the next chapter.
4 On this tantra of the śākta Yāmala tradition (existing only in manuscript form), and on
its commentaries, see A. Sanderson ‘The Sulva Exegesis of Kashmir’ in D. Goodall and
A. Padoux, Mélanges tantriques à la mémoire d’Hélène Brunner, Pondichéry, 2007,
pp. 231–424. On this tantra, and on the (perhaps thirteenth century) Mantramuktāvali,
as well as on the ŚT, see also T. Goudriaan and S. Gupta, Hindu Tantric and Śākta
Literature, Wiesbaden, 1981, pp. 49, 134–137,
5 1.75–78 refers to the Tantrik Texts edition of the TRT (reprint Delhi, 1981); 1.73–81 to
the edition of its first chapter with the Sudarśana commentary of Prān.amañjarī (Calcutta,
1940). This Tantrarāja is a different text from the tantra quoted by Abhinavagupta in
the Tantrāloka as Tantrarājabhat..tāraka, which is the 24.000 śloka Jayadrathayāmala.
6 The absence of am . śa is a serious defect since, as we have seen previously (see chapter
2, pp. 19–21, on mantravicāra, a mantra can only be made use of by a sādhaka if there
is an am. śa.
7 This is a different case from that of a mantra which is ‘asleep’ without being defective,
and is to be ‘awakened’ before being made use of.
8 See Introduction, pp. 3–4.
9 On these practices, see the next chapter.
10 According to the Śivārcanacandrikā, the mantra in such a case is purified by repeating
it thrice whilst raising the kun.d.alinī and then doing a pūjā to the devatā.
11 See Agnipurān.a, chapter 292 or Tārābhaktisudhārn.ava (p. 324) quoting the
Rudrayāmala.
12 This tantra is perhaps the same as the Mahocchus.ma quoted by Jayaratha in his Viveka
on TĀ 1.18 as one of the eight Brahmayāmalatantras (ibid. vol. 1, p. 42), a group
of texts where Kapālīśabhairava has Mahocchus.mā among the four goddesses of his
retinue (see A. Sanderson, ‘Saivism and the Tantric traditions’, in C. Carrithers et al.
eds, The Category of the Person, London, 1988, pp. 660–704).
13 It is unclear how this is to be done.
14 On this term, see chapter 7, p. 96.
15 The term mātrkā for the letters of the alphabet is used when their nature of powers
(śakti) is to be˚ underlined – for instance, when they are ritually placed on the body by
nyāsa.
 Notes 155
7 Mantric practices and the nature of mantric utterance
1 The author is grateful to the authorities of the Kuppuswami Research Institute in
Chennai for permission to reprint this article which first appeared in vols. LVI–LXII,
1986–92, of the Journal of Oriental Research, Madras, 1992 (pp. 65–76). This version
includes, when necessary, a few corrections and emendations to the original text; the
references given in the notes have, when necessary, been updated.
2 The NT, with Ks.emarāja’s Uddyota, was published as vols. 16 and 61 of the Kashmir
Series of Texts and Studies (KSTS), Bombay, 1926–1939. This edition is reprinted in
one volume edited with an introduction by V.V. Dvivedi, Delhi: Parimal Publications,
1985.
.
3 On this practice, see the entry on pratyangira of the Tāntrikābhidhānakośa (TAK),
vol. 3 (forthcoming).
4 These nine practices are as follows: dīpana, bodhana, tād.ana, abhis.ecana, vimalīkaran.a,
indhananiveśana, santarpan.a, guptībhāva and āpyāya. This list is partly the same as
that of the usual ten mantrasam . skāras used to counteract the mantrados.as quoted, for
instance, in the Śāradātilaka 2.112–123, or in the Kuārn.avatantra: on the subject, see
here chapter 6.
5 The action of encasing is samput.īkaran.a/kriti, a term used more often, I believe, met-
˚
aphorically, especially in the Trika, to describe the mutual encasement (that is, the
conjunction or fusion) of two metaphysical entities, usually Śiva and Śakti. It is also
used to express the mutual inherence of Śiva and the cosmos (see MVT 11.07; TĀ
3.205–206; Parātrim . śikāvivaran.a, pp. 264–265, etc.). The term is used in the NS.A
(1.29–30) to describe the coincidence of the first two (inner) triangles of the śrīcakra:
one identified with Śiva, the other with Śakti.
6 These two works are the same text under two different names: Vāmakeśvarīmataa,
with the commentary of Rājanaka Jayaratha, KSTS, LXVI, Srinagar, 1945. Nityā-
s.od.aśikārn.ava, with two commentaries, Rijuvimaśinī of Śivānanda and Artharatnāvalī
˚
of Vidyānanda, edited by V.V. Dvidedi, Varanasi, 1968.
7 This is a magical practice: cakra, bīja and letters are to be inscribed on gold leaf to be
worn as a protective amulet (raks.ā).
8 The meaning of grasta in this context is not to be linked with its meaning in grammar/
phonetics, where it is a fault of pronunciation due to the utterance of a letter hindered
or held back at the throat – cf. R kprātiśakya XIV,3: jihvāmūlavigrahe grastam etat.
9 In their commentaries on NS ˚. A or VMT, Śivānanda and Vidyānanda explain:
.
ankuśah. kromkārah..
10 This is clearly stated in Prān.amañjarī’s commentary on TRT 1.72 where she describes
several practices and says that there are still other ones. Cf. TRT with the commentary
Sudarśana of Prān.amañjarī critically edited for the first time by Jatindrabimal Chaud-
hury, Calcutta, 1940. Prān.amañjarī lived in Varanasi in the eighteenth century.
11 See ‘Un terme technique de mantraśāstra, vidarbha’, JAs, CCLXV (1977), pp. 345–
349.
12 These are usually pallava, yoga, rodha, grathana, samput.a and vidarbha (order of the
Phet.K). The longer list of the NT originates perhaps – like the one previously given of
nine mantric practices – in the Ucchus.matantra, that is, in an earlier stratum of Tantric
texts.
13 In the PTV, p. 243 – see above, Introduction, p. 8 and note 17.
14 On the nature of mantras as well as on the non-dualistic Śaiva conceptions of the word,
refer to my Vāc. The Concept of the Word in Selected Hindu Tantras (Albany: SUNY
Press, 1990; Indian edition, Delhi: Satguru Publications, 1992), and to H.P. Alper, ed.,
Understanding Mantras (Albany: SUNY Press, 1988), which includes an – at that date
– exhaustive bibliography on the subject.
15 On nyāsa, see here chapter 4.
16 The YH with Amrtānanda’s commentary, the Dīpikā, was edited by V.V. Dvivedi,
˚
156 Notes
Delhi: Motilal Banarsidass, 1998. I refer here to this edition which is the only reli-
able Indian edition of this work. French translation with introduction and notes by
André Padoux: Le coeur de la Yoginī, Yoginīhrdaya, avec le commentaire Dīpikā
d’Amrtānanda, Paris: Collège de France, 1994. ˚
17 ˚
An explanatory note with a drawing of the kāmakalā is given on p. 202 of my transla-
tion of the YH. On the kāmakalā, and especially its sexual connotations, see David
G. White, Kiss of the Yoginī. ‘Tantric Sex’ in its South Asian Context, Chicago: The
University of Chicago Press, 2000.
18 We saw above (p. 97) a nyāsa, on the image of the goddess Tripurasundarī, of her
mūlavidyā, an.kuśena vidarbhitā: intersected and bound up with the an.kuśabīja
KROM ..
19 This is Sarasvatī, also called Vāgīs.ī. The name Lipidevī, however, implies that this is
a written alphabet goddess. Sarasvatī, in fact, holds a book in one of her hands; the
written word is thus associated with her, on the divine plane.
20 It is described in several tantras, the Siddhayogeśvarīm . ata, Mālinīvijayottara,
Kubjikāmata, in the PTV, as well as in the AgP 145.
21 On the Mālinī, see the n.ādiphānta entry of the TAK, vol. 3. An interpretive study
of this letter-deity is that of Somdev Vasudeva, ‘Synaesthetic Iconography: 1. The
N.ādiphāntakrama’, which quotes several texts, published in D. Goodall and A. Padoux,
eds., Mélanges tantriques à la mémoire d’Hélène Brunner/Tantric Studies in Memory of
Hélène Brunner, Pondichéry: Institut Français de Pondichéry, 2007, pp. 517–550.
22 This is described in Jayaratha’s commentary on TĀ 3.67. On this, see my Vāc (quoted
above, note 14), pp. 241–243.
23 On these diagrams, see the appendix of J.A. Schoterman’s edition of the
S.at.sāhasrasam. hitā, op cit.
24 On mantroddhāra and the quest for am . śa, I wrote an article published in the BEFEO
65, 1978, an English (revised) version of which is published here as chapter 2.
25 On should perhaps underline in this respect that the same Sanskrit term, pāt.ha, is used
for reading and for reciting. The coalescence of these two meanings is interesting, and
is surely not accidental.
26 Transposed in another civilisation where the written sign is fundamental, mantras have
evolved differently, as instanced by the siddham script of the Japanese Shingon sect,
where Sanskrit bījas are used in meditation very much like Chinese characters. See on
this R.H. van Gulik, Siddham. An Essay on the History of Sanskrit Studies in China
and Japan, New Delhi: Sharada Rani, 1980.
ADDENDA 2009: On the nature of mantras and especially on the oral/written rela-
tionship, see A. Padoux, ‘L’oral et l’écrit. mantra et mantraśāstra’, translated here as
chapter 9.
On the role of writing in India, see Charles Malamoud, ‘Noirceur de l’écriture. Re-
marques sur un thème littéraire de l’Inde ancienne’, in Paroles à dire, paroles à écrire,
Paris: Editions de l’EHESS, 1997, pp. 85–114.

8 Body and mantra: mantras in the human body

1 In this paper I take up again, but more briefly, a subject on which I wrote an article
‘Corps et mantra. De la présence des mantras dans le corps’ published in Oscar Botto,
et al, eds., Du corps humain, au carrefour de plusieurs savoirs en Inde. Mélanges
offerts à Arion Ros.u par ses collègues et ses amis à l’occasion de son 80e anniversaire,
Bucarest et Paris, 2004, pp.563–578.
2 On this important notion, see Paul Schilder’s classical work on the subject: The Image
and Appearance of the Human Body, first published in 1935.
3 One says, too, that the yogin attains a state of mantra: mantratvam.
4 This important text of the Pāñcarātra contains many ritual prescriptions. It was edited
 Notes 157
in the Gaekwad Oriental Series (vol. 54), Baroda, 1967. See on this Marion Rastelli,
Philosophisch-theologische Grundanschauungen des Jayākhyasam . hitā. mit einer
Darstellung des Täglichen Rituals, Vienna, Verlag der Österreichischen Akademie der
Wissenschaften, 1999. Though very interesting, this text has not been translated.
5 This mental worship is studied and carefully commented upon in chapter 12 of the
thesis of M. Rastelli, quoted in the preceding note.
6 There are, as far as I know, no serious systematic studies of Hindu mudrās used in
ritual. On a particular Tantric case, see A. Padoux, ‘The body in Tantric ritual: the case
of the mudrās’, in T. Goudriaan, ed., The Sanskrit Tradition and Tantrism, Leiden,
1990, pp. 60–75.
7 This assertion is to be qualified, for a deity can very well be conceived as being ‘made
of’ letters only. The mantra-god Navātman, for instance, is made up of the nine pho-
nemes HSKŚMLVRYŪM . , A Tantric deity having always one or several ‘faces’ (vaktra)
and ‘members’ (an.ga), its mantric form will also include theses additions, which will
also be made up of phonemes – see TĀ 30.15–16a. Are we faced, then, with mental
images or with mentally evoked phonetic enunciations? As for Navātman, this god is
also described as a beautiful young man, associated with different goddesses, notably
the hunchbacked goddess Kubjikā: what is evoked in his case? The anthropomorphic
figure, or the syllabic enunciation? On the navātman mantra, see the navātman entry
of the TĀK, vol. 3.
8 On this, see the Somaśambupaddhati, vol. 3, edited and translated in French by Hélène
Brunner (Pondicherry, 1977), on the Śaiva initiations. Plates in the appendix to that
book show how such cosmic elements are placed on the body of the initiand.
9 Such, for instance, as those illustrating the S.at.cakranirūpan.a edited by A. Avalon as
The Serpent Power, being the Shat-Chakra- Nirūpana and Pādukāpañcaka, Madras:
Ganesh & Co., 1953.
10 The process is described in full detail by M. Rastelli, in pp. 246–270 of her study Phi-
losophisch-theologische Grundanschauungen der Jayākhyāsam . hitā (Vienna, 1999).
11 This mental rite is studied by A. Sanderson: ‘Meaning in Tantric ritual’, in A.M. Blon-
deau and Fr. Schipper, eds., Essais sur le rituel III, Paris: Bibliothèque des Hautes
Etudes, Sciences Religieuses, vol. CII, pp. 15–95.
12 This worship of the goddesses is described and commented by A. Sanderson, ‘Man.-
d.ala and Āgamic Identity in the Trika of Kashmir’, in A. Padoux, ed., Mantras et
diagrammes rituels dans l’hindouisme (Paris 1986), with a drawing showing the body
of the adept drawn by Stefanie Sanderson – printed again on p. 308 of the 1999 reprint
of R. Gnoli’s translation of the TĀ (Milano: Adelphi, 1999).
13 I am not sure ‘lived body’ is good English – probably not. What I wish to refer to here
is the distinction put forward by Max Scheler (and taken over by existential psychoa-
nalysis, the Daseinanalyse) between the Leib, the body as it is experienced, lived, and
the Körper, which is the perceived, physical, anatomic body; both of course are never
separated – one does not ‘live’ merely with one’s physical body.
14 On uccāra, see the entry on this word in vol.1 of the TĀK. Prān.a is not the respira-
tory breath, but a ‘subtle’ inner vital ‘breath’. L. Renou defined it as ‘l’aspect vital de
l’ātman’. On this subject, see the entry prān.a of the TĀK.
15 In S. Gupta, D.J Hoens, T. Goudriaan, Hindu Tantrism (Leiden, 1979). Though purely
imaginary, the yogic body, being ‘intraposed’ within the anatomic body, is both very
largely delineated on it and linked to it when its aspect and functioning are visualised:
all this ‘happens in the body’. This authorises some – Mircea Eliade, for instance – to
speak in such a case of ‘physiologie mystique’: the expression is debatable, but it
underlines well the link, the interaction of those two ‘bodies’.
16 This important tantra is often quoted or referred to by Abhinavagupta. It deals mostly
with the cult of mantra-goddesses possessing magical powers. A shorter version of
this work was edited and translated in English by Judit Törzsök for an Oxford (unpub-
lished) thesis. A fuller edition is now expected.
158 Notes
17 The SvT describes this in the cosmic and soteriological perspective of the six ‘ways’
(adhvan).
18 The NT completes the series of twelve kalās with those from nādānta to unmanā as
corresponding to different, ever higher, planes of the deity. A precise and very use-
fully commented description of this passage of the NT is given by Hélène Brunner
in her study of that tantra: ‘Un tantra du nord: le Netratantra’, BEFEO, LXI (1974),
pp. 125–197.
19 kūt.atraye mahādevi kun.d.alinītraye ‘pi ca/
cakranām . pūrvapūrves.ām . nādarūpen.a yojanam//
20 For the Yoginīhrdaīya, as for many Śaiva texts, the kun.d.alinī is considered as compris-
˚ igneous, solar and lunar. The first, igneous, section starting from the
ing three sections:
lowest cakra, the second part, solar, extends from the heart-cakra to the bhrūmadhya,
where begins the third one. To this threefold division corresponds that of the śrīvidyā;
it is also present in the nine cakras constituting the śrīcakra which are also distributed
into three groups. During the japa, all these three groups are brought together by the
ascent of the nāda which prolongs the uccāra of the three HRĪM ..
21 For the Śaiva cultic practice as described in the āgamas (and as it is still, in some
cases, performed in India today), see R.H. Davis, Ritual in an Oscillating Universe.
Worshipping Śiva in Medieval India, Princeton: Princeton University Press, 1991. We,
of course, do not know if a real mystical union with the deity in ritual was current
among Śaiva adepts in the tenth to fourteenth centuries. We know, on the other hand,
that officiants often limit themselves to the mere enunciation of the prescribed mantras
– mantraprayoga – without trying to do anything else than follow formally the pre-
scriptions of the ritual handbooks: this is indeed very far from the world of the ancient
texts – but we do not know much about that world.
22 The big toe of the right foot is the spot where Kālāgni, the (destructive) Fire of Time
which burns the cosmos to cinders, comes forth. We see here that the destruction of the
body of the adept is foreseen .
23 The cakras or granthis, spots, ‘nods’ where the power, the prān.a, concentrates are
in a way obstacles for they are thresholds to be passed. This is why, to progress, they
are to be cut or pierced: granthiccheda. See the entries on these terms in the TĀK,
vol. 2.
24 On khecara and the state of being khecara (khecaratā/khecaratva), see the
corresponding entries of the TĀK, vol. 2,

9 The oral and the written

1 Translation of ‘L’oral et l’ecrit. Mantra et mantraśāstra’, in C. Champion, ed., Tradi-
tions orales dans le monde indien, Paris: Editions de l’EHESS, 1996, pp. 133–148.
2 They are, in fact, deities: Tantric pantheons are pantheons of mantras.
3 On the ‘circulation’ of a mantra in the nād.īs, see, for instance, chapter 7 of the NT
which describes the ‘subtle meditation’ (sūks.madhyāna) of the Netramantra OM . JUM.
SAH.. This mantric practice is briefly described in Brunner, 1974, and Padoux, 1990,
pp. 411–406.
4 The recordings on cassettes (or otherwise) of the mantras of the pūjā that one finds
nowadays in India show another aspect of the ‘vulgarisation’ of mantras. The ritual
formulas of the usual daily Hindu worship, however, are not secret – differing in this
from those for the worship of esoteric Tantric deities.
5 We must not forget, on this subject, that silent, purely visual, reading is a compara-
tively modern usage. Earlier, reading was pronouncing, thus giving back to the oral its
rightful place. See Ong, 1982.
6 uddhāra: from ud-DHR, take out, raise up, select, extract, See, here, chapter 2 on the
extraction and selection˚ of mantras.
 Notes 159
7 On prastāras and gahvaras, see J.G. Schoterman, The Satsahasrasamhita, Brill,
1982.
8 On the rites through which an adept – a sādhaka – masters the mantra he has received,
see Brunner 1974a, pp. 411–443.
9 On these rites, see here chapter 2, pp. 18–22.
10 That is, the imaginary structure of centres (cakra) and channels (nād.ī) conceived as
present in the physical body. It is usually, but wrongly, called ‘subtle body’; the subtle
body (sūks.ma- or lin.gaśarīra) properly so-called being something else.
11 On this sense of dhyāna, see the entry on that word in the TAK, vol. 3 which refers to
G. Bühneman, The Iconography of Hindu Tantric Deities (two vol., Groningen. Egbert
Forsten, 2000–2001). By nyāsa one may also place divine entities, cosmic divisions,
yantras or cakras, etc.
12 When tantras, etc., name a mantra, they generally quote the rs.i who has first ‘seen’ it,
˚
its metre (chandas) – which gives it a ‘vedic’ air – then the deity (devatā) it expresses,
and the syllable which is its seed (bīja). To these are sometimes added the power
(śakti), the point (kīlaka) and the use (viniyoga) of the mantra.
13 A triangle (point down) is the symbol of the goddess, of the feminine. The phoneme
E is thus in essence the locus of creation – and of bliss. It is, Jayaratha says (ad, TĀ
3.94), the ‘mouth of the Yoginī’ (yoginīvaktra) where one receives initiation and expe-
riences the bliss of liberation. But vaktra, in this case, is not the mouth, but the sexual
organ of the Yoginī, where the yogin in sexual union with his ritual partner attains bliss
and fusion with the Absolute. The deity, in this case, is above all form, whilst being
also (at least metaphysically) triangular.
This sort of practice is described in chapter 29 of the Tantrāloka – on which see
Silburn, Kundalinī. The Energy of the Depths, Albany, 1988.
14 On the uccāra of OM . , see Padoux, 1990, chap. 7, pp. 402–411. On the kāmakalā, see
Punyānanda’s Kāmakalāvilāsa, edited and translated by A. Avalon, or, for a study and
interpretation of this diagram, David G. White, ‘Transformation in the Art of Love.
Kāmakalā Practices in Hindu Tantric and Kaula Traditions,’ History of Religions,
18/2, 1998, pp. 172–198.
15 Buddhist dhāran.is were translated in Chinese, written therefore since the fourth cen-
tury. But the cult and homage paid to the sūtras as holy ‘Scripture’ seem to have
existed earlier in the Mahāyāna traditions. The most marked cult of the Book is that of
the Sikh Adigranth, but it is by reciting it that it is honoured and that a beneficial result
is produced. See Dusenbery 1992.
In the Hindu field, a purān.a, as a book or, better, in manuscript, may be honoured.
Such is the case notably of the Devīmāhatmya: see Coburn, 1991.
16 On the mantiravatis, see Diehl, 1956.
17 On the anārthakya (lack of meaning) of mantras, see Renou, 1960, specially,
pp. 70ff.
Index of Sanskrit terms

am. śa 21, 92, 101, 129 n. 48–9, 129 n. 52, astra (an.ga) 70, 75, 82, 83, 124 n. 22, 143
154 n. 6, 156 n. 24 n. 31, 149 n. 79, 152 n. 6, 152 n. 7
akathahacakra 21 astramantra (PHAT. ) 75, 82
akad.amacakra 21 aham . 45, 47, 48, 138 n. 103
aks.amālā 10, 28, 33, 81, 88 ākars.an.a 37, 87, 96
aks.ara 15, 65, 90, 93, 96, 101, 112, 116 ākranta 98
aks.amālā Syn. aks.asūtra 10, 28, 33, 81, ācārya 14, 19, 35, 38–9, 113, 120, 127 n.
88 28, 143 n. 29
aks.asūtra 33 ādyanta 98
aks.asūtramantra 83 ādhāra 41, 72, 109
an.ga 16, 17, 30, 37, 60, 61, 65, 70–1, 72, ādhāraśakti 62, 105
75, 76, 78, 83, 124 n. 28, 127 n. 26, āpyāyana 93
145 n. 50, 145 n. 53, 148 n. 73, 148 n. āyudhapurus.a 152 n. 8
75, 148 n. 78, 149 n. 86, 152 n. 7, āvahana 6, 84, 153 n. 29
157 n. 7 āsana 6, 16, 29, 62, 105, 146 n. 58
an.ganyāsa 68, 70–1, 77, 78, 144 n. 44 āsanamantra 16, 146 n. 58
an.gamantra 60, 65, 66, 70, 71, 124 n. 28, id.ā 45
145 n. 53, 148 n. 73, 148 n. 76, 148 n. indhananives.ana 93, 155 n. 4
77–8 uccāt.ana 37, 38, 87
ajapājapa 32, 43, 44, 45, 50, 51, 69, 70, uccāra 3, 40, 42, 99, 106–8, 112–13, 116,
106, 138 n. 98, 140 n. 127 119, 124 n. 11–12, 125 n. 29, 157 n. 14,
adhikāra 91, 113 158 n. 20, 159 n. 14
adhivāsana 75 uddhāra 7, 13–14, 17, 18, 22, 83, 91, 101,
adhvan 59, 63, 66, 74, 144 n. 43, 158 n. 17 114, 126 n. 5, 159 n. 6
adhvanyāsa 63, 144 n. 43 unmanā 40, 41, 42, 63, 106, 107, 108, 143
anāhata 44, 45, 108, 138 n. 101 n. 41, 158 n. 18
anuttara 61, 118 upān.gamantra 24, 25, 76
anusam . dhāna 9, 85 upām . śu 4, 25, 30
anusmaran.a 19, 31 upāya 44, 139 n. 118
anusvāra 117 rnidhanicakra 21
apāna 45, 112 ˚rs.i 66, 68–70, 72, 78, 147 n. 68,
abhicāra 22, 37, 66 ˚ 148 n. 72
abhis.eka 21, 34, 35, 38, 93 rśyādinyāsa 68–70, 76, 77, 79, 135 n. 67,
abhis.ecana 93 ˚ 145 n. 49
ari 20, 21 Om . 34
arghya 36, 59, 74, 135 n. 67 OM . JUM . SAH. 13, 82, 100, 158 n. 3
ardhacandra 107, 108
avadhūti 45 kan..thasthā 120
avasthā 40, 85, 137 n. 84–5, 139 n. 118 karamālā 33, 85, 133 n. 47
 Index of Sanskrit terms 161
karaśuddhi 145 n. 49, 145 n. 53 n.ādiphāntakrama see mālinī
karān.ganyāsa 30, 65, 66, 70–1, 76, 78, tattva 59, 61, 62, 74, 103, 104, 105–6, 150
79, 135 n. 67, 141 n. 7, 144 n. 44, 145 n. n. 97
49–50, 149 n. 83 tattvanyāsa 77
karanyāsa p. 55, 65, 66, 70, 71, 76, 78, 79, tālu 107, 108
145 n. 53 tritattvanyāsa 74
kalā 3, 7, 40–1, 62, 65, 72, 74, 77, 79, tād.ana 93
104, 107, 108, 119, 145 n. 50, 145 n. turya 85, 153 n. 21
51–2, 158 n. 18 turyātīta 61, 85, 86
kavaca 70, 75, 101, 124 n. 22, 147 n. 68 tus.n.īm 99, 130 n. 3
kāmakalā 100, 119, 156 n. 17, 159 n. 14 Trika 3, 8, 14, 46, 63, 73, 95, 100, 101,
kāmyakarman 21, 22, 149 n. 80 106, 117
kīrtana 50–1 dīks.ā 12, 21, 27, 38–9, 41, 60, 115, 123 n.
kīlaka 6, 68, 79, 117, 147 n. 68, 148 n. 71, 1, 136 n. 77–8, 143 n. 29
159 n. 12 dīpana 93, 155 n. 4
kun.d.alinī 15, 40, 42, 78, 106, 108, 112, devatānyāsa 69, 80, 150 n. 95
116 dvārapūjā 77
kula – kaula 20 dvādaśānta 41, 63, 75, 83, 107, 108–9,
kulākulacakra 20 116
kullukā 30, 37 dhāman 109
kūrmacakra 21, 129 n. 49 dhāran.ī 31, 159 n. 15
KS.A 28, 73, 78, 79, 101 dhyāna 7, 18, 23, 31, 52, 74, 77, 78–80,
ks.udrakarman 82 104, 105, 116, 125 n. 30, 130 n. 4, 135 n.
gan.eśanyāsa 63, 144 n. 44 67, 144 n. 45, 145 n. 87, 151 n. 101
garbhastha 98
gāyatrī 34, 35–6, 38, 49, 131 n. 16, 135 n. naks.atra 20, 63, 144 n. 44
66, 137 n. 95 naks.atranyāsa 63, 144 n. 44
gahvara 18, 101, 115, 128 n. 36, 159 n. 7. navayoninyāsa 79
guptibhāva / gopana 93, 155 n. 4 navātmamantra 58, 60–1
grathita 97 nād.i 103
granthi 103, 109, 158 n. 23 nāda 3, 15, 17, 40, 41, 44, 45, 46, 106,
grasta 97, 98, 155 n. 8 107, 108, 116, 136 n. 71, 138 n. 101, 138
grahanyāsa 77, 144 n. 44 n. 109, 158 n. 20
cakra (see also man.d.ala) 41, 56, 62, 63–4, nādānta 107, 158 n. 18
66, 72–3, 75–8, 100, 101, 103, 105, 108, nādoccāra 106
109, 116, 118, 119, 129 n. 45–6, 144 n. nāman 97, 98
46, 145 n. 48, 155 n. 7, 158 n. 20, 158 n. nāmajapa 49, 50, 52
23, 159 n. 10, 159 n. 11 nāmastotra 49
cakranyāsa 66, 76, 77, 100, 105, 144 n. 46 nam-sumiran 50
chandas 6, 37, 68, 69, 147 n. 68, 148 n. netramantra 90, 93, 95, 96
71, 159 n. 12 nyāsa 6, 10, 16, 17, 30, 36, 39, 43, 54–80,
janana 93 97, 100, 103–6, 115, 116, 137 n. 84, 141
japa 24–53, 81, 82, 83, 84, 85, 86, 87, 93, n. 1–106, 151 n. 99, 151 n. 103, 156 n.
108, 116, 120, 124 n. 12, 124 n. 16, 124 18, 159 n. 11
n. 21, 130 n. 1–142, 142 n. 24, 151 n. parā(vāc) 15, 47, 87, 100
101, 151 n. 103, 151 n. 1, 158 n. 20 parāmarśa 138 n. 106
janmasthāna 109 parivāramantra 76
jalpa 130 n. 3 pallava 32, 38, 91, 99, 155 n. 12
jāti 2, 5, 38, 71, 93, 149 n. 86 pin.galā 45, 131 n. 17
jāgrat 40, 61, 85, 153 n. 21 pin.d.anātha
jīvana 75, 93 pīt.ha 63, 74, 77, 78, 80
jīvanmukta/ī 60, 95, 100 pīt.hanyāsa 74, 80, 149
jīvanyāsa 75, 77, 78, 151 n. 101 pun.yaks.etra 29
jñānaśūla 109 putrakadīks.ā 63
162 Index of Sanskrit terms
puraścaran.a 7, 14, 27, 28, 29, 30, 32, mantrados.a 89–94, 155 n. 4
34–5, 37, 76, 115, 129 n. 49–50, 132 n. mantraparīks.ā
281, 134 n. 53, 139 n. 114 mantravicāra 7, 12–23, 129 n. 53
pustaka 117 mantravīrya 8, 31, 40, 111
pūjā 6–7, 8, 17, 27, 28, 33–7, 43, 48, 50, mantraśarīra 104
59, 67, 71, 75–9, 83, 84, 86, 104–5, 108, mantraśāstra 13, 26, 36, 49, 52, 54, 56,
124 n. 20, 134 n. 54, 134 n. 62, 135 n. 67, 89, 102, 111–21
139 n. 115, 142 n. 9, 145 n. 50, 145 n. 53, mantraśuddhi 89, 92–4
147 n. 70, 149 n. 87, 151 n. 103, 152 n. 7 mantrasam . skāra 89–94
pran.ava 39, 65, 93, 107 mantrasādhana 7, 9, 21, 29, 39, 76, 81,
pratyabhijñā 100 115, 133 n. 36, 135 n. 68
prastāra 15, 16, 18, 101, 114–15, 128 n. 36 mahānyāsa 61, 62
prān.a 112 mahāpreta 63
prān.apratis..thā 6, 78, 124 n. 23. 151 n. 101 mahāmudrā 66, 143 n. 26
prān.āyāma 42, 44, 73, 77, 78, 135 n. 67 mahāsetu 30, 37, 135 n. 67
prāyascitta 27, 30, 32, 34, 38, 75, 136 n. mātrkā 3, 15, 16–17, 18, 58, 60, 61, 63,
73, 136 n. 79 ˚ 69, 70, 72–3, 93, 115, 117, 123 n. 8,
66,
PHAT. 60, 82 127 n. 30, 128 n. 34, 131 n. 13, 142 n.
bāhya(japa) 18, 28, 41, 44, 64, 73, 75, 78, 25, 145 n. 53, 147 n. 68, 149 n. 84, 149
107, 108, 114, 117, 118, 126 n. 10, 136 n. 85, 154 n. 15
n. 71, 138 n. 103, 149 n. 86 mātrkānyāsa 60, 63, 71–80, 123 n. 8, 135
bindu 3, 15, 17, n.˚ 26, 145 n. 53, 149 n. 89
bīja 2, 3, 5, 6, 13, 18, 26, 27, 28, 37, 40, antarmātrkanyāsa 72, 73, 77, 78
57, 58, 59, 63, 68, 71, 73, 79, 80, 93, 97, bahirmātr ˚ kānyāsa 72, 73, 77, 78, 149
100, 101, 116, 117, 119–20, 142 n. 25, n. 87 ˚
144 n. 45 146 n. 55, 157 n. 68, 148 n. mātrkāpūjā 149 n. 87
71, 148 n. 77, 149 n. 81, 155 n. 7 ˚
mānasa (japa) 4, 25, 30, 58, 99, 130 n. 4,
bījamantra 2, 7, 101, 116, 117 133 n. 37 140 n. 136
bubhuks.u 7, 21, 36, 37, 87, 95, 152 n. 2 mānasayāga / pūjā 105
bodhana 93, 155 n. 4 māran.a 38, 87, 136 n. 71
brahman 3, 14, 17, 25, 40–1, 42, 43, 111, mālāmantra 20, 22
116, 119, 126 n. 10, 139 n. 117, 142 n. mālinī 15, 18, 60, 73, 117, 128 n. 36, 150
11 156 n. 26, 159 n. 12 n. 93
brahmamantra 7, 146 n. 55 mukti / moks.a 82, 87, 90, 95, 152 n. 5
brahmayajña 25 mudrā 6, 33, 36, 55, 64, 66–7, 71, 73,
brahmarandhra 62, 63, 72, 73, 74, 83, 100 74–5, 77, 84–5, 104, 116, 124 n. 20, 141
Brahmī, etc. 77 n. 8, 146 n. 60, 146 n. 62, 147 n. 66, 148
bhakti 27, 31, 33, 40, 49, 50, 51, 121 n. 71, 150 n. 95, 151 n. 100, 157 n. 6
bhajana 31, 49, 50 mumuks.u 36, 37, 82, 87, 95, 100, 131 n. 17
bhāvanā 31, 38, 39, 40, 41, 42, 53, 54, 58, mūrtinyāsa 74, 79, 145 n. 53
63, 74, 83, 85, 86, 133 n. 42, 136 n. 84, mūrtimantra 16, 17, 60, 75, 76, 106, 124
146 n. 64 n. 28, 127 n. 24, 143 n. 32, 143 n. 36,
bhāvām . śa 19 145 n. 53
bhukti 21, 87, 90 mūlamantra 12, 16, 17, 22, 30
bhūtalipi 28 mūlavidyā 12, 145 n. 53
bhūtaśuddhi 75, 77, 78, 144 n. 43, 151 n. mūlādhāra 41, 43, 63, 66, 69, 72, 73, 74,
101 78, 79, 80, 100, 105, 108, 109, 116
man.i 82 mrtyujit 95, 100, 109
man.d.ala 6, 7, 75, 93, 146 n. 55 ˚
meru 34, 82, 86
madhya 46 moks.a 82, 87, 95, 152 n. 5
madhyamā 47, 130 n. 4
mantiravati 120 yantra, cf. vidarbha 59, 78, 100,
mantradevatā 16, 76 118–20, 142 n. 25, 146 n. 58, 149 n. 87,
mantradeha 6, 104 159 n. 11
 Index of Sanskrit terms 163
yuktividarbha 98 śabdabrahman 17, 126
yoga (mantrasam . skāra) 99, 155 n. 12 śarīra (lin.ga, sūks.ma, sthūla) 56, 103, 142
yoginīnyās.a 105 n. 18, 159 n. 10
yoni 128 n. 36 śāktopāya 47, 139 n. 118, 146 n. 64
yoninyāsa 77 śānti śāntika 37, 38, 87, 135 n. 69
yonimudrā 92, 131 n. 17, 135 n. 67 śikha 30, 70, 83
Rāmsmaran.a, Rāmsam . kirtan, śiras 30, 70, 83
Rāmsumiran 50 Śiva 74
rāśi 20, 144 n. 44, 150 n. 95 śivahasta 65, 66
ripu 20, 92 śuddhi śodhana 82, 92, 129 n. 48, 144 n.
rudrāks.a 38, 82, 131 n. 13, 152 n. 4 43
rodha rodhana 99, 155 n. 12 śūnya 41, 109, 136 n. 84, 137 n. 85
laks.ya 109 śmaśāna 29
lipitaru 101, 117 śraddhā 136 n. 79
lipinyāsa 101, 116, 150 n. 91 śrīcakra 37, 41, 58, 64, 97, 118, 119, 136
lipīśvara 59 n. 84, 139 n. 115, 142 n. 23, 144 n. 45,
vaktramantra 76 145 n. 48, 146 n. 58, 155 n. 5,
varga 16, 17, 21, 78, 115, 134 n. 47, 149 158 n. 20
n. 86 śrīvidyā 28, 40, 108–9, 116, 119,
varn.a (class) 15, 32, 90, 91, 131 n. 14 124 n. 12, 124 n. 16, 133 n. 43, 144 n. 45,
varn.a (phoneme) 112, 116, 126 n. 13, 147 158 n. 20
n. 68 s.at.karman.i 22, 37, 38, 87, 99, 136 n. 72
varn.atanu 72 s.ad.an.ganyāsa 68, 69, 70, 77, 78, 79, 48 n.
varn.amāla 28, 131 n. 13, 133 n. 47, 134 73, 49 n. 87
n. 47 sam . kīrtan 51, 53
varman = kavaca 83 sannidhi/sam . nidhāna 85
vaśikaran.a 37 sam . put.a 28, 32, 38, 79, 96–7, 98, 101
vāc 3, 5, 8, 12–13, 14, 22–3, 25, 47, 85, sam . vit 100
89, 111, 116, 124 n. 26, 130 n. 4 sam . skāra cf. mantrasam . skāra 10
vācaka 7, 22, 36, 57, 85, 100, 111 sam . hāranyāsa 65, 73, 78
vācika(japa) 4, 25, 30, 133 n. 37 sam . hitamantra
vācya 7, 57 sakalīkaran.a 65, 70, 71, 75, 83, 145 n. 50
vāmācāra 66 sattarka 139 n. 112
vidarbha vidarbhita 91, 97, 98, 101 satsan.g 51
vidyā 2, 18, 28, 30, 40–1, 71, 74, 126 n. 3, sandhya 27, 28, 35–6, 37, 49, 134 n. 54
148 n. 77 sannidhi 85
viniyoga 59, 68, 147 n. 68, 148 n. 71, 159 samayadīks.ā 60, 66, 143 n. 29
n. 12 samanā 63, 107, 108
vimalīkaran.a 93, 155 n. 4 samasta 97, 98
vis.uva 40, 41, 137 n. 85 sarvatovrta 98
visarjana 88, 154 n. 29 sādhaka˚ 7, 14, 19, 20, 29, 36, 37, 38, 67,
vīradravya 19 76, 82, 87, 88, 95–6, 100–1, 127 n. 28,
vaikharī 47, 100, 130 n. 4 131 n. 11, 131 n. 17, 131 n. 19, 143 n.
vyāpakanyāsa 66, 76, 77, 78, 29, 152 n. 2, 154 n. 6, 159 n. 8
151 n. 102 sādhya 20, 21, 96–8, 99
vyāpta 151 n. 102 siddha 20, 21, 142 n. 25
vyāhrti 36, 102 siddhi 58, 64, 87, 95
˚ supta, susupta, svāpagā 91
śakti 6, 8, 13, 14, 28, 34, 56, 63, 68, 79, sumiran 51, 52, 140 n. 136
84, 85, 101, 107, 127 n. 30, 135 n. 65, sus.upti 85
145 n. 53, 147 n. 68, 148 n. 71, 148 n. sus.umnā 42, 45, 78, 105, 106, 108, 109,
78, 155 n. 16, 159 n. 12 112, 124 n. 16, 136 n. 81
śatarudriya 49 susiddha 20, 21
śabda 17, 51 sūksma 56
164 Index of Sanskrit terms
sūks.maśarīra 103, 159 n. 10 n. 92, 138 n. 109, 139 n. 110
sūtra 8, 45–7, 82, 159 n. 15 haripath 51, 140 n. 127
setu 30, 37, 135 n. 67 hariśravan.a 51
SAUH. 3, 7 hastanyāsa 65, 146 n. 55
sthāpana 75, 84 hastapūjā 146 n. 55
smaran.a/smrti 52 hrdaya 30, 65, 70, 75, 83
svabhāvām ˚ 19 ˚ 149 n. 79
. śa heti
smaran.a/smrti 52 homa 33, 34, 35, 38, 39, 79, 88, 135 n. 67,
svādhyāya ˚25, 26, 130 n. 2 137 n. 84, 139 n. 112
ham. sa 32, 43, 44, 46, 47, 90, 91, 107, 137 HRĪM . 41, 48, 108, 117, 158 n. 20
Authors and works cited

BEFEO Bulletin de l’Ecole Française d’Extrême-Orient 125 n. 1, 128 n. 37,

130 n. 1, 141 n. 1, 156 n. 24, 158 n. 18
JAs Journal Asiatique 143 n. 29, 146 n. 55, 151 n. 2
KSTS Kashmir Series of Texts and Studies 128 n. 39, 155 n. 2, 155 n. 6

AgPur Agnipurān.a 59, 75

AJ Ajtāgama 123 n. 9
Atharvaveda 56, 119
Abhinavagupta 3, 5, 8, 9, 14, 27, 42, 46–8, 55, 60, 61–2, 66, 73, 95,
100, 106, 113, 118, 138 n. 104
AhS Ahirbudhnyasam . hitā
Aitareyabrāhman.a 3, 111
Īśsp Īśānaśivagurudevapaddhati 18, 90, 97, 99
Amrtānanda 37, 40, 45, 46, 60, 108, 136 n. 84, 156 n. 16
ĪPK ˚
Īśvarapratyabhijñākārikā 8, 9, 48
ĪPV Īśvarapratyabhijñāvimarśinī 150 n. 95
Utpaladeva 8, 48
Uddhārakos.a of Daks.in.amūrti 5, 126 n. 7, 127 n. 29
Rgveda 3, 12
˚
Kāmakalāvilāsa by Pun.yānanda 159 n. 14
Kir Kiran.āgama 12
KT Kulārn.avatantra 27, 29, 35, 54, 89, 125 n. 33, 144 n. 43, 155 n. 4
KMT Kubjikāmatatantra 128 n. 36
GT Gandharvatantra 28, 30, 33, 37, 90, 92, 132 n. 32, 133 n. 46, 134 n.
48, 134 n. 53, 134 n. 58, 144 n. 43–5, 145 n. 49, 146 n. 58, 149 n. 85,
150 n. 91
Gorakhnāth 51, 140 n. 134
Ks.emarāja 8, 14–15, 19, 20, 42, 43, 44, 45, 46, 47, 51, 90, 92, 93, 95,
96, 97, 98, 100, 117, 123 n. 8, 127 n. 21, 127 n. 23, 128 n. 40, 128 n. 41,
129 n. 43, 129 n. 44, 130 n. 4, 138 n. 103, 138 n. 107, 138 n. 109, 139
n. 110, 139 n. 114
166 Authors and works cited
Jabalopanis.ad 49
Caitanya 50–1
JRY Jayadrathayāmala 39, 40, 132 n. 23, 137 n. 89, 154 n. 5
JayS Jayākhyasam . hitā 33, 74, 81–8, 104, 127 n. 30, 131 n. 17, 133 n. 47,
145 n. 53, 148 n. 77, 149 n. 83, 151 n. 2
Jayaratha 8, 16, 18, 42, 47–8, 97, 118, 126 n. 10, 137 n. 91, 145 n.
48, 150 n. 95, 154 n. 12, 159 n. 13
Jñāneśvarī
TRT Tantrarājatantra 89, 97
Tantrasāra (of Krs.n.ānanda Āgamavagīśa) 145 n. 53
TS ˚
Tantrasāra (of Abhinavagupta) 13, 92, 139 n. 112, 126 n. 10
TĀ Tantrāloka 55, 95, 118, 126 n. 10, 142 n. 17, 154 n. 5, 159 n. 13
TAK Tāntrikābhidhānakośa 123 n. 5, 126 n. 5, 155 n. 3, 156 n. 21, 157 n.
7, 157 n. 14, 158 n. 23, 158 n. 24, 159 n. 11
TBhS Tārābhaktisudhārn.ava 30, 31, 32, 33, 38, 66, 69, 71, 72, 92, 129 n.
48, 129 n. 50, 129 n. 53, 131 n. 17, 132 n. 21, 132 n. 24, 134 n. 53,
134 n. 59, 144 n. 44, 148 n. 71, 148 n. 77, 149 n. 82, 149 n. 84, 149 n.
87–90, 150 n. 92, 151 n. 102
Tirumular 124 n. 14
Triśiromata/Triśirobhairavatantra 46
DMS Daks.inamūrtisam . hitā 43–4
Devībhāgavatapurān.a
Devīmāhātmya 50, 159 n. 15
Dhyānabindūpanis.ad 43, 137 n. 94
Nātha 44, 50, 51, 52
Nārāyanakan.t.ha 15
NT Netratantra 90, 95, 124 n. 16
NS.A Nityās.od.aśikārn.ava 155 n. 6
Nityotsava 131 n. 13, 144 n. 45, 145 n. 48, 146 n. 58, 149 n. 86, 149
n. 88
PS Paramārthasāra 46
Parameśvara-sam . hitā 57, 65, 68, 70, 142 n. 15, 144 n. 43, 145 n. 53
PRKS Paraśurāmakalpasūtra 133 n. 21, 133 n. 44, 134 n. 49
PTV Parātrīśikavivaran.a 3, 5, 46, 73, 124 n. 17, 143 n. 36, 153 n. 22, 155
n. 5, 155 n. 13, 156 n. 20
PTlv Parātrīśakalaghuvrtti
. ˚ 93
Pingalamata 89, 92,
Phet.K Phet.kārinītantra 72, 97, 99, 131 n. 14, 148 n. 77, 155 n. 12
BhG Bhagavadgīta 25, 126 n. 10
Bhāgavatapurān.a 50
Bhāskararāya 54, 137 n. 85
.
MatP Matangapāramēśvarāgama 33, 132 n. 25, 133 n. 44, 136 n. 79
Manu Manusmrti 25, 112, 124 n. 13
˚
 Authors and works cited 167
MM Mahārthamañjari by Maheśvarānanda 3, 46, 68
Mantramahārn.ava 13, 69, 125 n. 32, 145 n. 49, 145 n. 53, 148 n. 71
MMUd Mantramahodadhi 13, 125 n. 32, 133 n. 38, 144 n. 44, 150 n. 96
MNT Mahānirvān.atantra 71, 126 n. 16, 151 n. 98, 151 n. 101
MBh Mahābhārata 26, 49, 130 n. 5
MVT Mālinīvijayottaratantra 73, 129 n. 48, 137 n. 89, 143 n. 28, 143 n. 38,
144 n. 40, 155 n. 5
Maheśvarānanda 3, 46, 68
Mrg Mrgendrāgama 15, 76
˚
Rau ˚
Rauravāgama 18, 127 n. 26, 149 n. 80
LT Laks.mītantra 13, 151 n. 103
Lalitāsahasranāma 54
VMT Vāmakeśvarīmata 18, 97, 144 n. 44, 145 n. 48, 145 n. 53, 150 n. 95,
155 n. 8
Varivasyārahasya of Bhāskararāya 137 n. 85
Vīn.aśikhatantra 136 n. 70
Veda 1–3, 26, 31, 49, 100, 102, 105, 124 n. 26, 128 n. 41, 142 n. 10,
147 n. 68
Vājasaneyi sam . hitā 49
VBh Vijñānabhairava 42, 44, 45–7, 137 n. 93–4, 137 n. 97
S.at.sāhasrasam . hitā 150 n. 93, 159 n. 7
Sātvatasam . hitā 42, 136 n. 76
ŚT Śāradātilaka 13, 89, 117, 125 n. 32, 155 n. 4
Siddhayogeśvarīmata/Siddhamata 30, 95, 106, 137 n. 89, 150 n. 93,
156 n. 20
Śivānanda 68, 97, 155 n. 6, 155 n. 9
Śivayogaratna of Jayaprakāśa 44
ŚS Śivasūtra 14, 51, 100, 139 n. 110
ŚSV Śivasūtravimarśinī 127 n. 20
Śivastotrāvali of Utpaladeva 48, 139 n. 117
S.at.cakranirūpana 157 n. 8
Sanatkumārasam . hitā 17, 128 n. 30, 128 n. 34, 130 n. 56, 151 n. 103
SP Somaśambhupaddhati 90, 152 n. 2, 157 n. 8
SpK Spandakārikā 9, 31
SvT Svacchandatantra 15, 16, 19, 20, 33, 42, 44–7, 74, 105, 107, 127 n.
20–21, 127 n. 23, 128 n. 39, 129 n. 52, 130 n. 4, 135 n. 66, 137 n.
93–4, 137 n. 97, 138 n. 103, 144 n. 40, 144 n. 43, 149 n. 83, 158 n. 17
Haripāth of Jñāndev 50, 51, 140 n. 124, 140 n. 127, 140 n. 131–2

*
 Taylor & Francis Group

Want to stay one step

ahead of your colleagues?
Sign up today to receive
free up-to-date
information on books,
journals, conferences and
other news within your
chosen subject areas.
Visit
www.tandf.co.uk/eupdates
and register your email
address, indicating your
subject areas of interest.

You will be able to amend your details or unsubscribe

at any time. We respect your privacy and will not
disclose, sell or rent your email address to any
outside company. If you have questions or concerns
with any aspect of the eUpdates service, please
email eupdates@tandf.co.uk or write to: eUpdates,
Routledge, 2/4 Park Square, Milton Park, Abingdon,
Oxfordshire OX14 4RN, UK.
 R O U T L E D G E PA P E R B A C K S D I R E C T
Routledge
Paperbacks Direct
Bringing you the cream of our hardback
publishing at paperback prices
This exciting new initiative makes the best of our
hardback publishing available in paperback format for
authors and individual customers.
Routledge Paperbacks Direct is an ever-evolving
programme with new titles being added regularly.
To take a look at the titles available, visit our website.

www.routledgepaperbacksdirect.com
 ROUTLEDGE

Are there some elusive titles you’ve

been searching for but thought
you’d never be able to find?
Well this may be the end of your quest. We now offer a
fantastic opportunity to discover past brilliance and purchase
previously out of print and unavailable titles by some of the
greatest academic scholars of the last 120 years.

Routledge Revivals is an exciting new programme whereby

key titles from the distinguished and extensive backlists
of the many acclaimed imprints associated with Routledge
are re-issued.

The programme draws upon the backlists of Kegan Paul,

Trench & Trubner, Routledge & Kegan Paul, Methuen, Allen
& Unwin and Routledge itself.

Routledge Revivals spans the whole of the Humanities and

Social Sciences, and includes works by scholars such as Emile
Durkheim, Max Weber, Simone Weil and Martin Buber.

FOR MORE INFORMATION

Please email us at reference@routledge.com or visit:
www.routledge.com/books/series/Routledge_Revivals

www.routledge.com