Indian Insights - Buddhism, Brahmanism and Bhakti

INDIAN INSIGHTS:
BUDDHISM, BRAHMANISM AND BHAKTI
Papers from the Annual Spalding Symposium on Indian Religions
LUZAC ORIENTAL
46 Great Russell Street, London 'VI/Cl
I997
Luzac Oriental I997
All rights reserved: no part oj this publication may be reproduced, stored in a

retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise without prior permission oj the Publishers.
British Library Cataloguing-in-Publication Data.

A catalogue record of this book is available
from the British Library.
Printed and bound in Great Britain by

Bookcraft (Bath) Limited,
1vlidsomer Norton, Somerset
Contents
Preface
Vll
Dermot Killingley
The Paths of the Dead and the Five Fires
Peter Connolly
21
The Vitalistic Antecedents of the

Atman-Brahman Concept
Mark Allon
39
The Oral Composition and Transmission

of Early Buddhist Texts
Sally Mellick Cutler
63
Still Suffering Mter All These Aeons:

the Continuing Effects of the Buddha's
Bad Karma
Lynn Thomas
83
The Nature of the Repetition

in the Indian Idea of Cyclical Time
David Smith
91
Vi~nu and the Dance of Siva
Anthony Tribe
Mafijusri and 'The Chanting of Names'
(Niimasal'{lgitz): Wisdom and its Embodiment
in an Indian Mahayana Buddhist text
109
VI
Indian Insights: Buddhism, Brahmanism and Bhakti
Robert Mayer
137
Caskets of Treasures and Visions of Buddhas

Indic antecedents qf the Tibetans gTerm-rna Tradition
Theodore Gabriel
Freeing the Tiyyas: Narayar:ta Guru and
Religious Ideology in Kerala
Gavin Flood
Ritual Dance in Kerala: Performance,
Possession, and the Formation of Culture
L. S. Cousins
Aspect of Esoteric Southern Buddhism
169
185
Hiroko Kawanami
Buddhist Nuns in Transition:
the Case of Burmese thilri-shin
Notes on Contributors
225
Preface
The essays in this volume were all originally presented at the Spalding
Symposium on Indian Religions (formerly the Symposium on Indian
Religions) held annually in spring in Oxford, England. This volume,
which consists of papers given between 1989 and I994, is the fifth
generated by the annual Symposium. The previous volumes have been:
Perspectives on Indian Religion: papers in honour of Karel Werner, edited by Peter
Connolly and published in I986 by Sri Satguru Publications, Delhi; The
Yogi and the Mystic (1990), Symbols in Art and Religion (1990), and Love Divine:
Studies in Bhakti and Devotional Mysticism (1993), all edited by Karel Werner
and published by Curzon Press, London, as part of their Durham
Indological Series.
The Symposium on Indian Religions was convened for the first time in
1975 by Dr. Karel Werner, who was then Spalding Lecturer in Indian
Religions at the University of Durham. The original intention was to
provide an annual forum for scholars working and teaching in a subject the
nature of which which often placed them in institutions with no close
colleagues with whom to exchange ideas. From the beginning the
Symposium answered a real need. And though the subject has expanded
over recent years, and conferences and seminars have proliferated and
provided other opportunities for interacting with colleagues within the
discipline, the annual Symposium remains the only regular forum. Its ongoing success in combining amenable informality with a useful opportunity
to share scholarly work (completed or in progress) with others in the field
has long been acknowledged and taken advantage of Over the years, it has
attracted scholars and graduate students with an interest in Indian and
Indian-derived religions from all over Britain and many other countries,
including Australia, Germany, India, the Netherlands, Sri Lanka and the
United States of America. 'While some have attended only occasionally,
others are committed regulars.
Aspects of Hinduism - from Brahmanical orthodoxy to localised
sectarian practices, Buddhism - in many of its different forms, Jainism and
Sikhism have all been explored from the perspectives of disciplines as
diverse as anthropology, philosophy, psychology, religious education and
VIII
textual analysis'. The structure of the symposium is such that each paper
can be discussed at length by participants, a process which offers the author
an opportunity to benefit from informed comment before proceeding to
publication or further presentations.
Reflecting the diversity of topics addressed in the symposia themselves,
the present collection focuses on the Hindu-Brahmanical and Buddhist
traditions in a variety of manifestations. The volume is organised thematically, essays having been arranged in pairs where each contribution can be
regarded as complementary to its partner. The studies by Dermot
Killingley and Peter Connolly are both concerned with Upani~adic ideas
and their relation to other aspects of Vedic material. Dr. Killingley
discusses the possible continuity between Vedic eschatology and the
doctrine of rebirth as presented in the Upani!ads, whereas Dr Connolly
suggests that on the subject of praT}a Vedantic exegetes have misinterpreted
their primary sources. Mark Allon and Sally Mellick-Cutler focus primarily
on the textual tradition of Pali Buddhism. The former concentrates on the
implications of patterns and structures in key sections of the earliest stratum
of canonical texts, relating them to their oral transmission, and the latter
draws on a wide range of canonical and non".canonical material to discuss
the theme of the Buddha's bad karma. Lynn Thomas and David Smith
both explore'mythological dimensions of the Hindu tradition through
literature and art. The contributions of Anthony Tribe and Rob Mayer
deal with Mahayana Buddhist material: the former through the analysis of
a text that was influential in both India and Tibet; the latter through
investigating Indian antecedents for a practice that is often regarded as
distinctive of Tibetan Buddhism.
The final pair of essays on Hindu-Brahmanical themes focus on the
south Indian state of Kerala. That by Theodore Gabriel explores the life
and teachings of Narayar,ta Guru, one of its most outstanding religious
figures, while Gavin Flood vividly describes and offers an interpretation of
'the dancing of the teyyams', one ofKerala's most dramatic religious rituals.
The collection concludes with two essays on aspects of Buddhism in southeast Asia about which little is generally known. In exploring the features
and origins of 'esoteric' Buddhism in the Theravada countries of the
region, L. S. Cousins' study demonstrates that there is far more diversity
within Theravada Buddhism than is usually acknowledged. Hiroko
Kawanami's essay addresses the position of nuns in contemporary
Burmese Buddhism. She suggests that their position is one of 'transition'
between tpe historical ambiguities surrounding the status of nuns in the
past and their defining a new more respectable and stable religiosity for
themselves in the future.
Overall this collection offers a wide-ranging yet thematically coherent
exploration of many facets of the Indian religious heritage. And in drawing
on sources as diverse as classical texts and contemporary fieldwork, the
Priface
IX
essays demonstrate both the perennial interest the ancient material

commands and also the way the religious traditions they engendered are
adapting to the modern world. As such the volume should be of value to all
scholars and students with an interest in Indian and Indian-derived
religious thought and practice.
All the contributions are published here for the first time and thus
constitute a substantial body of original research. Many of the
contributors are regular participants at the annual Spalding Symposium,
so readers attending future meetings may be able to explore issues raised
in this volume with the authors themselves. Details of future symposia can
be obtained from Peter Connolly through Luzac Oriental.
Sue Hamilton and
Peter Connolly
October 1996
Dermot Killingley, University of Newcastle upon Tyne
Introduction
The passage known as the doctrine of the five fires (paficiigni-vidyilJ, which
occurs in two versions in the Brhad-Arar;yaka Upan~ad and the Chiindogya
Upani~ad, is often cited as an early statement of the contrasting ideas of
rebirth according to one's actions, and salvation. 1 It does indeed refer to
rebirth, and in the ChUp version, though not in the BrhUp version, this is
related to a person's previous actions; both versions also contrast rebirth
with the world of Brahman. These features of the passage fit well with the
ideas of karman, rebirth and salvation; but they should not blind us to
other features which link it to older ideas, and which may help us partially
to trace the origins of the passage itself
It is commonly asserted that belief in rebirth is not of Aryan origin, and
it is often attributed to non-Aryan, or specifically Dravidian sources with
which the Vedic Aryans were in contact. The evidence for this, when it is
offered at all, seems to be that rebirth appears suddenly and without
precedent in the history of Vedic literature. A typical, and perhaps seminal,
expression of this view is Richard Garbe's 192I encyclopedia article
"Transmigration (Indian),,:
suddenly and without any transitional stages that we can perceive, the Indian
people was seized by the oppressive belief in transmigration ... the theory, as
it meets us for the first time in the literature, appears already fully formed in
the shape of belief in a permanently continued but ever-changing existence. 2
A more recent writer, who sees the doctrine of the five fires as one of
"several very ancient folk explanations of the mechanism of rebirth",3
argues:
The Upani!ads and Buddhism have basically identical ideas on rebirth ...
This similarity, coupled with the absence of a Vedic karma-rebirth doctrine,

suggests that the UpaniJadic and Buddhist doctrines may be diverging interpretations of a common, non- Vedic rebirth tradition. 4
Vedic to this writer seems to refer only to the hymns of the ~V; he does not
count the Upani~ads as part of the Veda, and he seems unaware of the
BrahmaTJas. This attitude is all too common, and makes it possible to
exaggerate the revolutionary nature of the Upani!ads.
The view which this quotation represents also oversimplifies what is said
about life after death in the Vedic hymns. The same writer says:
The afterlife belief in the J:(g Veda is simply that after death, the soul leaves
the body and enters heaven or hell or eternity.s
This statement ignores the great variety of Vedic thought on life after
death, both in the hymns and in later Vedic literature. The main way in
which Vedic thought on the subject differs from later Hindu thought is that
it usually regards life after death as something to be achieved, rather than
as something to be escaped from. But this achievement is thought of in
several ways.
Firstly, one can become immortal through one's offspring: "0 Agni,
may I reach immortality with offspring" (~V 5, 4, ro; cf. ~V 6, 7 0 , 3).
Immortality through offspring appears in one of the many accounts of how
the creator god Prajapati, who is also the primordial sacrificer and father,
reproduced himself by creating the cosmos. This particular account occurs
in the context of the Agnihotra, the offering at sunrise and sunset. After
desiring to reproduce himself, Prajapati first produces Agni (fire), but is
afraid that Agni will devour him. He then produces milk and butter, plants,
and the ritual sound svahii; with these he performs offerings 'and turns away
Agni, who is also Death.
And il1deed, anyone who, knowing this, offers the Agnihotra, reproduces
himself with offspring even as Prajapati reproduced himself, and saves
himself from Agni, Death, when he is about to devour him (SBT 2,2,4, 7).
The idea that a man survives after death in his offspring is used in AiUp 4,
3-4 (to be discussed below). It is also mentioned in BrhUp 3, 9, 28, though
only to be rejected.
A second idea is survival through dispersal of the person into the
corresponding parts of the universe (cf. ~V IO, 58):
lVIay your eye go to the sun, your breath to the wind; go to the sky and to the
earth in due order. Or go to the waters, if that has been ordained for you;
take your stand in plants, with your body (J:(V 10, 16,3)'
Dispersal is not thought of as destruction here; it is prayed for as something

of benefit to the deceased. It appears as a reward of knowledge in SBr IO, 3,
3,8:
The Paths qf the Dead and the Five Fires
And when he who knows this passes away from this world, he passes into fire
with his speech, into the sun with his eye, into the moon with his mind, into
. the directions with his ear, and into the wind with his breath. Being made of
them, he becomes whichever of these deities he chooses, and is at rest.
The dialogue ofYajfiavalkya and Artabhaga (BrhUp 3, 2, I3), which is well

lmown as an early instance of the idea of deeds determining the fate of a
person after death, begins with a similar account of dispersal, in which the
number of items is expanded, and dispersal seems to be the common lot
rather than a particular achievement.
Thirdly, there is the idea of survival in one's deeds, particularly ritual
deeds (i;tiipiirta), which prepare a place for the deceased in the next world:
Join with the Fathers, with Yama, with your sacrifices and good works
(iJtfipiirta) in the highest heaven. Leaving faults behind, return home; join
your body with strength (!?-V!O, l4, 8).
The phrase "world of good deeds" (sukrtasya loka- AV6, IIg, r; AV 6, 120, r;
121, I), though it need not have originally referred to a state after
death, can be mentioned here. It appears in MuUp T, 2, I, and in KaUp 3,
r; a similar idea is expressed in MuUp I, 2, 6, where a man's offerings
carry him to the world of Brahman which consists of his good deeds (sukrto
brahma-lokaM. 6 Yajfiavalkya uses the idea of survival after death in or
through one's deeds, in a passage that is often cited as a landmark in the
history ofkarman (BrhUp 3, 2, 13); he is drawing on an ancient idea.
Let us now turn back to the passage we started with, and examine its
Vedic antecedents. The passage exists in two versions, in BrhUp 6, 2, and
ChUp 5, 3-5, IO. The ChUp version is longer, and contains ideas about
karman as the determinant of rebirth which are completely absent Jrom
the BrhUp version. In some respects, therefore, the BrhUp version seems
more primitive; but it does not appear that the ChUp is simply a reworking
of it. Since each contains some material that is not in the other, it is more
probable that each is a reworking of a lost common source. In examining
these two versions, we can take BrhUp 6, 2 as our starting point and note
where ChUp 5, 3-IO is different.
AV 6,
Story
Like many doctrinal passages in the early prose Upani;ads, the passage in
both versions is introduced by a story which relates the occasion on which
the doctrine was taught. This frame story is one in which Gautama Arul)i,
the father of Svetaketu, is instructed by a k;atriya, Pravaha!.laJaibali, after
the latter has asked the son a series of five questions which neither the son
nor the father is able to answer. It appears in two versions: BrhUp 6,2,1-8;
and ChUp 5, 3. A similar story occurs as KauUp I, I; here, however, the

teacher's name is different, and there is no indication that he is a k~atriya.
Further, this story does not mention the five questions. Renate Sohnen has
argued, through a detailed internal examination of all three stories, that
KauUp I, I is the earliest version of the frame story, and not a reworking of
the other two as has often been supposed. 7 However, it is possible that th~
history of the doctrine is partly independent of the history of the story, so
that even if KauUp I, I is the oldest extant version of the story, the doctrine
taught in the remainder of KauUp I need not for that reason be older than
the doctrine in BrhUp 6, 2, 9-I5 and ChUp 5, 4-IO. In fact, KauUp I differs
from the other two in its doctrine as well as in its story: the five fires are not
mentioned, and the paths of the dead are substantially different. Since it is
with the history of doctrine that we are concerned here, we shall
concentrate on BrhUp 6, 2 and ChUp 5, 3-IO, treating KauUp I as a related
text rather than as a version of the same text.
Some modern interpreters attach great significance to the motif of the
~atriya teacher.s Not only did Vivekananda claim that all the important
ideas in the Upan~ads came from ~atriyas - and he claimed to be a ~atriya
himself - but others have tried to show from such passages that the doctrine
of rebirth came from ~atriyas, who had it from non-Aryans, with whom it is
assumed that the k~atriyas were in more intimate communication than the
Brahmins were. Those who speculate about k~atriya teachers tend to
exaggerate their frequency in the Upani~ads, and to underestimate the great
variety of views, some of which could be of non-Aryan origin, which are
recorded as being taught by Brahmins; they also ignore the fact that
k~atriyas playa part in the theological discussions (brahmodya) in the
Briihmar;.as. 9 It has been suggested that the motif of the k~atriya teacher in
this and other passages points to the novelty of its ideas in Brahmin circles,
rather than to a specifically k~atriya origin.1O Rather than speculate on this
point, we shall look here at the texts and their antecedents, taking the
k~atriya teacher as a literary motif rather than a matter of historical record.
Another point to note is that the learner in this story appears as the
teacher in another well-known Upan~adic story, ChUp 6, where he instructs
his son Svetaketu, who has a role also in the present story. This is not the
only occasion in which the teacher in one story is the learner in another:
the great Yajnavalkya, the teacher ofJana~a king ofVideha in BrhUp 3-4
and an authority on ritual in much of the SBr, appears as learner in SBr II,
2, I, and in the same passage Janaka, a ~atriya, is the teacher. But the
attribution of a teaching to a particular teacher need not be historical. We
will return to SBr II, 2, I later.
The general outline of the frame story which introduces the doctrine of
the five fires is a recurrent one, in which received learning is found to be
inadequate,~and new teaching is sought. Svetaketu has been taught by his
father - whether in the way recorded in ChUp 6 we need not inquire.
The Patizs ifthe Dead and the Five Fires
Despite his father's teaching, he is unable to answer five questions put to

him by a ~atriya, PravahaI,laJaibali of the Paficalas. The questions, in the
BrhUp version, ar.e:
Do you know how these creatures go different ways.when they
depart? ..
Do YQu know how they retl\rn to this world? ...
Do you know how the other world is not filled up by the many who keep
on going there? ...
Do you know at the offering of which oblation does water acquire a
human voice, rise up and speak? "
Do you know the way to the path of the gods or of the ancestors?
(BrhUp 6, 2, 2).
The first three questions are clearly about the dead; the fourth is a
riddle whose meaning only becomes clear when it is answered later in the
story. The last question need not be about the dead, but the answer shows
that it is. The ChUp version has similar questions, but in a different order,
the riddling one being the last.
Svetaketu knows none of the answers. He goes back to his father, and
complains that his father's instruction has not enabled him to answer these
questions put by a ~atriya. Gautama goes to PravahaI,la and asks for an
answer to the questions. At this point the BrhUp is more elaborate than the
ChUp, but both versions include the motif of the questioner refusing the
. offer of material goods and insisting on his question - a motif that recurs in
KaUp I, 2I-29, and whose message is that material goods are finite, but
knowledge is, as Gautama puts it, "abundant, infinite, boundless".
PravahaI,la agrees to give him the knowledge which, he says, has not dwelt
before in any Brahmin.
The teaching which follows is in two parts, which I shall call "the Five
Fires pericope" and "the Two Paths pericope". 11 The frame story, as in
several other Upani!ad passages, disappears from view once the teaching
starts. It is not alluded to again in the BrhUp version; but the ChUp version
alludes to two of the five questions: how water speaks (Ch Up 5, ra, I) and
how the other world is not filled up (ChUp 5, ra, 8).
The Five Fires pericope

The Five Fires pericope follows a pattern, which is repeated five times in
the ChUp version and six in the BrhUp version. In each occurrence of the
pattern, a part of the cosmos is called a fire, something is offered in it, and
something else is produced. The product of each fire becomes the offering
in the next:

Offering
Fire
Product
faith (fraddhii;
other world
soma
soma
rain-cloud (pariaT!Ja)
ram
ram
earth
food
food
man
semen
semen
woman
man (BrhUp)/embryo (ChUp)
Each fire is analysed into five parts: fuel, smoke, flame, embers, sparks.
Each of these parts is identified with some part or aspect of whatever is
represented by the fire; the details vary somewhat between the two
versions. The pattern is thus fivefold, and is five times repeated; the BrhUp
version adds a sixth fire, which will be discussed below.
The motif of analysing a phenomenon by identifYing parts or aspects of
it with parts of the ritual is a common one: for instance, Ch Up 2, 13 analyses
the sexual act in this way, and BrhUp 6,4, 3 similarly analyses the sexual
parts of a woman; both passages are similar to the description of woman as
a fire in the Five Fires pericope. BrhUp 6, 4, 3, like oilr pasage, uses the
fivefold motif, being one of a series of such analyses based on the five
sections of the Siimavedic chant. The fivefold analysis, with reference to ritual,
indicates completeness: there are five sections to the chant, five layers of
bricks and five layers of stones in the great fireplace described in the SBr.
There are five kinds of bloodless offering (TS 6,5, II), and five species of
sacrificial beast: man, horse, cow, sheep, goat (SBr 1, 2, 3, 6; AV II, 2, 9).
Outside the purely ritual sphere, there are five seasons (SBr II, 7, 4, 4), and,
let us not forget, five fingers (SBr 1, I, 2, 16), which may be the natural basis
for the whole system of counting in fives. 12 The number five features also in
many cosmological and anthropological formulae - of which the Sarpkhya
system is a relatively late example whose origins can be traced in the Veda.
Perhaps the five khandhas are another. The significance of the number five
in the Veda is summed up at the end of a well-known cosmogonic passage:
The sacrifice is fivefold. The beast is fivefold. Man is fivefold. This whole
world, whatever there is, is fivefold. So he who knows this obtains the whole
world (BrhUp I, 4, 17).
The BrhUp version, again introducing a feature which is not in the ChUp
version, adds a sixth fire to the five: the actual fire in which man is
cremated. In this fire, the offering is the deceased man, and the product is "a
man of the colour of light" (puru~o bhiisvara-va77Jil~); that is, the man in a postmortem state achieved through the ritual process of cremation. The addition of a sixth fire does not seriously break the pattern of five, since the sixth
fire is fire itself Similarly there are fifteen digits of the moon, and the moon
is the sixteenth. 13 The passage quoted above, BrhUp I, 4, 17, demonstrates
the fivefoldness of man by first listing five parts: mind, voice, breath, eye,
ear, and then adding a sixth part, the self(atman).
To return to the five fires of our passage, they clearly represent a series
of processes by whIch life comes to birth. The first process, being heavenly,
is a little obscure, but it seems to mean that life originates, as it well might
in Vedic thought, with an act of the gods; the input of this process, represented as the offering, is faith (iraddha, an essential element of any Vedic
ritual, comprising confidence in its efficacity and a determination to
perform it properly). The output is the nourishing fluid of the universe,
soma. This becomes the input of the second process, which produces rain
from the cloud. The third process is the production of food grains from the
earth when nourished by rain. The fourth is the eating of food by man,
leading to the production of semen which is the concentrated essence of
food; the fifth is sexual intercourse leading to conception. Frauwallner 14
sees this as a doctrine arising from a basic fact, "the living force ofvVater ...
everything being clothed in the secret language ofthe sacrificial mystique."
This attempt to treat Vedic thought as prescientific science ignores the fact
that the starting-point of the process is not water but faith. Such an attempt
may help us initially to understand Vedic thought, but is at best a partial
way of interpreting it. To understand it fully, we must appreciate that ritual
is not just a clothing for the ideas, but their very basis.
The Two Paths
peri~ope
The Two Paths pericope is structured round the well-known opposition

between village and forest. The village is the ordered space in which
agriculture is practised, social relations operate, families are raised and
rituals can take place; while the forest is the natural, unordered space in
which plants and animals grow wild, social relations are renounced and
rituals are performed only mentally. The general sense of the Two Paths
pericope is that the life of the forest leads to Brahman, through the path of
the gods; the life of the village leads back to this world, through the path of
the ancestors.
The BrhUp version uses the theme of the sixth fire as a bridge passage,
linking the Five Fires pericope to the Two Paths pericope by introducing
the subject of death. Both versions introduce the first of the two paths by
making it the reward of those who know the five fires (and, in the BrhUp
version, the sixth fire). The theme of the reward of a particular piece of
knowledge is of course common in the Brahmar;as and Upani;ads, but it is
treated here at greater length than usual. The reward belongs both to those
who know the fires and to those whp perform upasana of sraddha ('faith') as
sarya ('truth') (BrhUp version) or as tapas ('heat; asceticism') (ChUp version).
Upasana and the verb upas are both often translated 'worship', but this is
inadequate. Upasana is a mental operation in which something is identified

with something else, often more remote and less easy to know; for instance:
The Man in the sun - I meditate on (upiis) him as Brahman (BrhUp
2, 1,2).15
The verb upiis regularly takes two accusatives (a direct object and an object
complement), so in our text fraddhii7J7. satyam upiisate (BrhUp 6, 2, IS) may be'
translated as "meditate on faith by identifying it with truth" or "meditate
on faith as truth". 16 Satya ('truth') and sraddhii ('faith') make a pair in Vedic
ritual thought: truth, which means that no untruth must be uttered by
those who take part in ritual, and faith, the confidence that the ritual will
have the effect promised by the Veda. As essential though non-material
elements of the ritual, they can take the place of the yqjamiina and his wife if
he has no wife (AiBr 7, ro), or the offering and the fire if no materials for
offerings are available (SBr II, 3, 1,4).17 They thus enable a man living in
the forest to wntinue his ritual life without performing actual rituals.
In the ChUpversion, the forest-dwellers meditate on faith not as truth
but as asceticism (tapas). 18
In the BrhUp version, the path of the ancestors is won by those who win
worlds by means of sacrifice, giving and tapas. In the ChUp version, which
has already used tapas as part of the entrance to the path of the gods, the
path of the ancestors is won by those who upiis sacrifices and good works
(i!tapiirta) as giving. This may be only a way of saying that they practise both
i!tiipiirta and giving, the reference to upiisana being only a device to create a
superficial parallel of wording with the entrance to the path of the gods.
Each of the two ways is described in a series of stages. The two versions
agree, except for some differences in the middle stages of each of the ways.
Path
of the gods:
BrhUp:
Harne day
bright fortnight l9 uttarayal).a20 c1eva-Ioka 21 sun
ChUp:
Hame day
bright fc)rtnight
Path
uttarayar.lil
year
~un
moon
lightning
Brahman
lightning
Brahman
of the ancestors:
BrlzUp:
,.;moke
night
dark fortnight
dak~]r:Iayana
pilr-Iob22
ClzUp:
smoke
night
dark f()rtnight
dak~iDayana
pitr-Ioka
moon
space
moon
The path of the ancestors leads to the moon, which has ancient associations with immortality, the ancestors, and soma. 23 The inclusion of the
moon in the path of the gods in the ChUp version is therefore anomalous.
However, the path of the ancestors continues beyond the moon:
BrhUp: space
ChUp:
wind
'pace wind
rain
smoke mi,t
cloud rain
earth
(man
plant' food
semen
woman - Kal).va text only)
The differences are not really significant; both versions agree in

describing a series leading from cremation to the eating of food and the
conception of a new persono Here, the ChUp version raises a problem
which is ignored by the BrhUp version: what ensures that the plant is eaten
by a man? This problem is introduced as follows:
It is difficult to escape from [the plant stage]. For whoever eats the food,
and whoever emits semen, that is what he becomes again (ChUp 5, IO, 6).24
It is at this point that the ChUp introduces karman as the deciding factor,
which is absent from the BrhUp version.
Mter describing the path of the ancestors, both versions add a third
path, not mentioned before, leading to low forms of life; we are not told at
which point this path separates from the others. It is mentioned only very
briefly, and without any reference to ritual or other traditional concepts,
so that its inclusion looks like an afterthought. The ChUp also adds some
details on specific forms of high and low rebirth to its account of the path
of the ancestors (ChUp 5, 10, 7), and a verse listing four specific sins (ChUp
5, 10, 9)
The questions in the frame story have now been answered, although
they are not mentioned again except when ChUp 5, 10, 9 says that this is
how water gets a human voice at the fifth offering, and why the other
world is not filled up.
Parallels to the Five Fires pericope

Each of the two pericopes has parallels elsewhere in Vedic literature, which
may throw light on its origin, and therefore on the origin of the whole
passage as it stands in BrhUp and ChUp. First, the Five Fires pericope has a
parallel in SBr II, 6, '2, the story already mentioned in whichJanaka king of
Videha instructs Yajnavalkya. Again we have the theme of a kJatriya
teacher who answers questions which a Brahmin is unable to answer.
Incidentally, at the end ofthe story Yajfiavalkya grants Janaka a boon, and
Janaka chooses the privilege of asking Yajfiavalkya questions (to which, it
appears, Yajfiavalkya is obliged to give answers); this is the boon
mentioned in BrhUp 4,3, 1. The story also says thatJanaka was thenceforth
a Brahmin.
Janaka challenges a group of travelling Brahmins by asking each ofthem
to explain the agnihotra (the daily ritual at sunrise and sunset). Yajfiavalkya
gives the explanation which most pleases the king. The king, after giving
him the usual thousand cows (here perhaps only a verbal compliment), says
that even Yajfiavalkya does not know the uprising (utkriintz), destination
(gah), foundation (pratiJ!ha;, satisfaction (trptl), return (punar-iivrttl), and world
that rises again (lokaTfl pratyutthqyinam) of the agnihotra ritual. Yajfiavalkya,
10
after advisin.g his companions not to engage in debate withJanaka in case

it comes to be known that a k~atriya has out-talked Brahmins, ignores his
own advice by leaving them and driving after Janaka to ask him for his
explanation of the agnihotra. The king says that the two agnihotras rise to the
atmosphere, rise thence to the sky, return to the earth, rise up and enter
man, rise up and enter woman. The dual forms of the verbs make it clear .
that the subject throughout is the two agnihotras. In each of these stages,
Janaka names a fire, fuel, and a pure offering (fukram iihutim). The pattern is
similar to that of the two passages we have examined; a series of processes,
described in terms of the sacrificial fire, leads to the production of food
from the earth, its ingestion by man, and the production .of semen. .
fire:
atmosphere
wind
offering: sun-motes
fuel:
sky
sun
moon
earth
fire
plants
man's mouth
man's tongue
food
woman's genitals
womb
semen
(SBr II, 6, 2).
The points promised by Janaka are not all explicitly answered; but the
agnihotras are four times said to 'rise up', five times to 'satisfy' (the object in
each instance being the 'fire'), and once to 'return'. As in the BrhUp and
ChUp passages, the first three fires are related to the three parts ofthe Vedic
cosmos, heaven, atmosphere and earth, though this time the position of the
first two is reversed; the fourth fire is related to the process of eating, and
the fifth to sexual reproduction. The son who is born is said to be the
'world that rises again' (loka1[l pratyutthiiyinam) of the agnihotras, and it is
added that for him who eats or has sexual intercourse knowing this, the
agnilzotra is offered. This seems to mean that the yajamiina is reborn in his
son, and that the full benefits of the agnihotra accrue to those who have the
knowledge of this doctrine of five fires, perhaps even if they do not perform
the agnilzotra ritual- a similar idea to the better-known priiTJiignihotra. 25
A similar passage, again in the context of the agnihotra, is in the
Jaiminrya-Briihmar;a. Here too we have a series of fires; in this case, each of
them is identified with Agni Vaisvanara.
offering:
fire:
product:
amrta, water
sun
soma
ram
soma
thunder earth
ram
food
food
man
semen
semen
man
fire
woman
man
man
UaiBr I, 45-6).26
vVe are much closer here to BrhUp 6, 2 and ChUp 5, 3-10. As in those
passages, each fire is given a product, which (except for the last) becomes
the offering in the next fire. Though the JaiBr passage is concerned with
the agnihotra, there is no reference to an actual agnihotra or a human
yajamiina; the sacrifice is a heavenly one, performed by the gods. We are told
that "at the fifth creation the divine waters speak with a human voice"; and
here this is more clearly apposite than in the BrhUp and ChUp passages,
II
because the initial offering is here not 'faith' but 'immortality (amrta), water
(iipa!;.) ,, apparently in appositionY A sixth fIre, the cremation fIre, is
int~oduced, and ,the passage culminates in an account of what happens
after death. This account even has something in common with the Two
Paths pericope, in that from the smoke of the cremation fire the dead man
goes to night, then to day, then to the dark fortnight, then to the bright
fortnight; here, however, night and day, and the two fortnights, are stages in
a single path, not in two separate paths. The man is asked a series of
questions, the answers to which act as passwords to successive stages of the
path; this motif appears in KauUp I, which we will look at later.
The theme of the 'world that rises again' in SBr II, 6, 2 is echoed here,
in the phrase "world for rising again" (laka/:! punar-utthqyaz); but this refers to
the goal reached by the man in his post-mortem state, who is the product
of the sixth or cremation fire, whereas in the SBr passage the "world that
rises again" (laka1!! pratyutthayinam) is the product of the fIfth fIre. The
concern of the JaiBr passage, therefore, is not with gaining the full benefits
of the agnihatra, but explicitly with gaining a place in the other world after
death. 28
We now know something of the history of the Five Fires pericope before
it came to the ears of the redactor of the common source from which BrhUp
6, 2 and ChUp 5, 3-IO are derived. This ,Pericope existed in at least two
versions, which have come down to us as SBr II, 6, 2 andJaiBr I, 45-6. Our
redactor has developed it further by taking from the SBr version the motif
of the k!atriya's questions, which he uses in his frame story, and from the
JaiBr version the concern with life after death. He has taken from JaiBr I,
45-6 the role of the gods as performers of the fIve sacrifIces, and the
location of the process in the other world (asau laka!;.); whereas in SBr II, 6, 2
the subject is the agnihatra which takes place in this world and is the work of
a human yo;jamiina. He has also taken the motif of the sL'{th fire from the
JaiBr version; although this sixth fire is mentioned explicitly only in' the
BrhUp version, the ChUp version implies it by linking the cremation fire to
the five fires in the following statement:
When he dies at his appointed time, they take him to the same fire from
which he came, from which he came into being (ChUp 5,9,2).
That is, after coming into being from five symbolic fires, he returns to an
actual fire.
This linking of cremation with birth recalls passages in which cremation
is said to be a birth, e.g.:
Having performed an offering, Prajapati procreated, and rescued himself
from Agni, Death, who was about to eat him. And indeed, whoever knowing
this offers the agnihotra, procreates the same offspring as Prajapati
procreated,29 and rescues himself from Agni, Death, who is about to eat him.
I2

And when he dies, and when they place him on the fire, then he is born out
of the fire, and the fire burns only his body. So, just as he is born from his
father or mother, so is he born from the fire (SBr 2, 2, 4, 7-8).
Thus, those who perform agnihotra with this knowledge escape death in
two ways: through the agnihotra, and through cremation.
Counting these two as births in addition to their natural birth, the performer of the agnihotra could be said to have three births. Though this
point is not made in the above passage, a similar point is made in another
passage of the same Briihmmy.a. Here, three births are enumerated: natural
birth, the sacrifice, and cremation:
A man is born three times. Thus: first he is born from his mother and father;
then when one who has a sacrifice coming has the sacrifice performed, he is
born a second time; and when he dies and they place him on the fire, and
when he comes into existence from that, he is born again. That is why they
say a man is born three times (SBr II, 2, I, I).
The topic of three births appears again in the Aitari!Ya UpaniJad. Here, as
throughout this UpaniJad, the subject is the iitman. The account starts with
the formation of semen and its emission in a woman (AiUp 2, I). In
accordance with the male-dominated view of procreation which permeates
the Vedas, this is the iitman's first birth. The second is the actual birth
(AiUp 2, 3). The iitman of the offspring (assumed to be male) is spoken of in
relation to the father, who in prospering it is prospering his own self
(AiUp 2, 3). Underlying the passage is the idea that a son is a continuation
of his father; which implies that while both are alive the father has two
selves.
This self of his is appointed as a deputy for good actions. Now his other self,
having done what he had to do, comes to the end of his lifespan and passes
away. Even as he is passing away from hence, he is born again. This is his
third birth (AiUp 2, 4).
This third birth stands for death: specifically, the death of a man who has a
son. The passage ends by saying that the sage Vamadeva, by knowing this,
became immortal at his death, and obtained all desires in heaven (AiUp 2,
5-6). The implication is that if a man has a son to continue his good actions
(Pur;ya karman) - ritual actions are no doubt meant - his death is like a birth
into the other world.
This idea of death as a birth, or cremation as a birth, is taken up by the
redactor as a link between the five fires and the topic of rebirth.
Before leaving the topic of the five fires, we can look at other passages
which share with it the idea of a series of processes leading to the formation
of life. For the sake of comparison with the Five Fires pericope, they are
tabulated here alongside BrhUp 6, 2.
The Paths qf the Dead and the Five Fires

BrhUp 6,2:
TaiUp 2, I:
Self
space wind fire
iyfuUp 2, 1,5:
Man
fire
ChUp 6, 2-5:
Being
heat
faith
soma
rain
soma
water earth:m plant' fc.lOd
water' earth
water
fond
plant' fimd
food
13
semen man
man
semen nff.<pring
man
These texts are less closely related to our passage than are those which
mention five fires. However, the correspondences between them and the
Five Fires pericope show that we are dealing with a recurrent pattern of
ideas which appears in many different contexts. The most persistent
features of it are that rain produces food, and that food, through semen,
produces offspring; the stages preceding rain are more varied. The MuUp
version is. particularly striking, since at first sight it looks like a haphazard
juxtaposition. Comparison reveals that the old series, well attested in the
BrahmaT}as and early prose Upan~ads, has survived in a verse Upanirad. The
BrhUp and ChUp Five Fires passages are unique in starting with faith
(fraddha;, but they are not unique in including soma, since this appears in
the MuUp series. Its inclusion must be related to its position as king of
plants, and its association, or in some cases identification, with the moon,
which in turn is associated with plants and with immortality. 31
In all these texts we are presented with a linear series, not a cycle. They
were worked into a cycle by our redactor, who clearly did not get all his
ideas ready-made from some hypothetical non-Aryan circle, but operated
with materials derived from the Vedic tradition of ritual and cosmological
thought.
Parallels to the Two Paths pericope

While the BrhUp and ChUp passages are the only ones in the classical
Upanirads to combine the Five Fires and the Paths of the Dead, there are
several Upan~adic passages which take up the topic of the Paths of the Dead.
iV!UUp I, 2, IO-II says that those who know only ~!aPi1rta ('sacrifices and
good works') enjoy heaven and then return to this world or a lower one;
those who practise tapas and fraddha reach the atman through the door of the
sun. Here, an Upaniradic author working on the theme of the contrast
between ritual and transcendent knowledge has taken up and condensed
our passage, letting the path of the ancestors stand appropriately for ritual,
and the path of the gods for knowledge. He probably used the ChUp
version, since he mentions low births, which appear in the ChUp version of
the path of the ancestors but not in the BrhUp version. It is also the ChUp
version, not the BrhUp version, that uses the term ~!iipi1rta.
KauUp I, 2 gives a different version of the paths of the dead. The path of
the gods is named (deva-yana, KauUp I, 3), but its landmarks are quite
14
different, and do not include the sun. The path of the ancestors is not
named, but its features are recognisable in the description of the jOl,lrney of
the dead to the moon, which is called the door of the heavenly world; the
moon feeds on the pra:r}as of the dead in the bright fortnight (pilrva-pa~e) and
causes them to be born in the dark Eortnight (apara-pa~er;a). However, the
parting of the paths is not at the cremation fire, which is not mentioned; all .
the dead go to the moon, and are asked who they are. Those who fail to give
the true answer are sent down in the form of rain, and are reborn; those
who answer correctly proceed on the path of the' gods. This path thus represents a further progress beyond the path of the ancestors, rather than an
alternative to it as in our two passages. KauUp I, 2 describes rebirth according to deeds and knowledge, and lists some examples of species that one
can be born in (the list is longer than the one in ChUp 5,10,7, and contains
none of the same items). The passage includes some material taken from
JaiBr I, 46, including the motif of a password to the world beyond death; it
also gives an elaborate geographical guide to the route followed by those
who have full knowledge, leading eventually to an interview with Brahma.
There are other Upani~ad passages that we may take as cognates rather
than reworkings of the Two Paths pericope, since they lack features that are
significant in the BrhUp and ChUp passages.
PrUp I, 9-10 is closely parallel to our two passages. After identifYing the
year with Prajapati, it describes two paths of the year, the southward
(da~rJiiyana, i.e. the half-year from the summer solstice to the winter solstice)
and the northward (uttariiyar;a, from winter solstice to summer solstice). The
account is very brief, but the context implies that these paths are travelled by
the dead. The southward path/is the path of the ancestors, which leads the
dead to the moon from which they return. This path is reached by those
who perform upiisana of sacrifice and good works (~tiipi1rta) as action (krta),
and who desire offSpring. The northward path leads to the sun, from which
there is no return; it is cessation (nirodha). It is reached by those who seek the
Self through tapas, brahmacarya, sraddhii and knowledge. The term 'path of the
gods' is not used here, but 'path of the ancestors' is.
'The passage is part of the answer to the first of the six questions in the
PrUp: "Whence are these creatures born?" (PrUp I, 3). The answer begins
with Prajapati's well-known desire for offspring, and his production of a
pair (mithuna, a word with sexual connotations), breath (priir;a, masculine)
and wealth (rOyi, feminine); these are identified with the sun and the moon,
the bright fortnight and the dark fortnight, the day and the night. Wealth is
identified with the path of the ancestors, and by implication priir;a is the
path of the gods. We can tabulate these identifications:
Breath:
~un
JiVealth: moon
northward path
hright fortnight day
[path of gods]
non~return
southward path
dark fortnight
path of ancestor~
return
night
(prUp 1,4-13).
15
One curious feature of this passage is its use of the term upasana in its
account of the path of the ancestors: ye ha vai tad-i~taPi1rte krtam it>' upasate
"those who upiis the sacrifices and good works of it [i.e. of a year] as deeds"
(PrUp I, 9).32 This 'is closely parallel to the ChUp version ofthe path of the
ancestors: ya ime grama i~taPi1rte dattam it>' upiisate "those who, in the village,
upas sacrifices and good works as giving" (this portion does not appear in
the BrhUp version), The PrUp makes better sense than the ChUp here;
using the notion of the year as the totality of time, it teaches an upiisana
which identifies the ritual deeds of a year with those of a whole lifetime.
This PrUp passage may therefore be close to our redactor's source.
We need not take the statement that those who take the path of the
ancestors return (punar avartante) as referring to rebirth in the classical sense.
This path is taken by those who desire offspring, so the passage may be
about two types of immortality: immortality through sons in this world,
and immortality in a post-mortem state in the other world. We can
compare it with SBr 2,2,4,7-8, SBr II, 2, I, I, and JaiBr I, 46.
BrhUp 5, !O describes just one path of the dead: wind, sun, moon, a
world without heat and without snow. There is no mention of any
qualification for entering it; it seems to be open to all who die. But ChUp 4,
15, 5 gives a path which is open only to those who know the Man in the
eye:
flame
day bright fortnight northward path year
sun
moon lightning Brahman
(ChUp 4, IS,S)
The stages on this path are exactly the same as those on the Path of the
Gods in ChUp 5, !O, 1.
Looking into the history of the topic of the Paths of the Dead before it
came to be associated with the Five Fires, we find that it is very -ancient,
though its content has not always been the same. ~V!O, 14, 2, in a context
which clearly concerns the dead, speaks of the way (gantu-, singular) found
by Yama, by which the ancestors have gone. In the same hymn, verse 7
tells the deceased to go by the paths (path-, plural) by which the ancestors
have gone. However, there is no classification of different kinds of path.
~V!O, 88, 15 mentions two paths:
There are two paths, I have heard, of the ancestors: of the gods, and of
mortals. By them comes together all that moves in the world, all that is
between the Father and the Mothe (.{tVro, 88,15).33
The context is a hymn to Agni Vaisvanara, describing his activity in the
two worlds of the sky and the earth, and also in the atmosphere; there is
nothing about the dead in the hymn. The Father and the Mother are no
doubt sky and earth, and the two paths seem to be related to them. The
syntax of the first half of the verse is ambiguous. As translated above, it
means that the ancestors have two paths, that of the gods and that of
16
mortals; but it could also be translated "There are two paths, I have heard
from the ancestors: of gods, and ofmortals."34
The verse is repeated many times in the Veda. 35 It is quoted in BrhUp 6,
2, 2,36 in the frame story of our passage; but it is understood there as
"There are two paths, I have heard - of the ancestors and of the gods - of
mortals." This construction fits the topic of the Two Paths, but it strains the
word-orde~ considerably. The verse is understood in the same way when it
occurs in SBr 12, 8, I, 21 (quoted from VS 19, 47), There, the two paths are
explained as those by which "he [the yajamanaJ leads the ancestors to
heaven"; but we are not told what is the difference between them.
Apart from this verse, there is some precedent in the hymns for the pair
'path of the ancestors, path of the gods':
VYe have come to the path of the gods, to carry out whatever we can (~V 10,
2, gab) ...
Agni, you who foreknow, and know by experience, the path that leads to the
ancestors, shine forth brightly when kindled (~V 10, 2, 7).
Verse 3, like most of the hymn, refers to Agni's role as regulator of the
ritual; verse 7 refers to another of his roles, as conveyor of the dead. But the
two paths are mentioned so far apart that they hardly constitute a pair.
The pair becomes clearer in the Atharva- Veda. Agni Kravyad - 'flesh-eating
fire', the cremation fire - is told to go by way of the path of the ancestors,
and not to come back by the path of the gods (A V 12, 2, IO); this is an
example of the desire to separate the dead from the living which we find in
the funeral hymns of .flV IO and in later death rituals. There are several
references to these two paths elsewhere in the AV (e.g. AV 18, 4, 1-2);
apparently the path of the gods is the one by which Agni carries offerings to
the world of the gods, and the path of the ancestors is the one by which he
carries the dead to the world ofYama. We thus begin to see why a pair of
paths should be associated with Agni; but at this stage only one of these
paths belongs to the dead. These A V hymns also show us how the two
paths in .flV IO, 88 lsab came to be understood as those of the gods and of
the ancestors, rather than as those of the gods and of mortals; but again it is .
only the path of the ancestors that is travelled by the dead.
The association of the moon with the path of the dead is ancient. The
addition of an alternative path that leads through the sun may have
arisen from a set of ideas in the SBr. The sun measures days and years,
and so is identified with time and with deathY To pass beyond death is
to pass beyond the sun, and the way to do so is through the sun itself.
The sun thus becomes the door of immortality, as it is for instance in
[sUp 16. It thus appears as a better alternative to the moon, which is the
door to the. world of the dead. The inclusion of the moon in the path of
the gods, as in ChUp 5, IO, I, is a survival of the older notion of a single
path of the dead.
The Paths if the Dead and the Five Fires
17
Conclusion
The two well-known passages on the Five Fires and the Paths of the Dead
do not represent a sudden intrusion of the ideas of rebirth, and of karman
as determining rebirth, into Vedic thought. Belief in karman itself has a
long history, if by karman we mean the capacity of an action to alect the
destiny of th.e agent. This belief is clear in the BriihmalJas as well as the
Upani~ads, particularly in relation to actions in the ritual sphere with which
the Veda is mainly concerned. One of our two passages, BrhUp 6, 2, does
not even mention karman as a determinant of rebirth. Indeed, it is not
clear from this passage whether what is born is a moral agent or even a
conscious being; the passage can be read as describing a biological process
in which life descends in the form of rain and is eventually born as a living
beu"lg.
The theory of karman and rebirth, together with mok~a, did not appear
as one package at the time of the Upani~ads, however logical and cohesive
that package may be in its classical form. It is made up of several distinct
elements, which we have discussed above: karman as having inevitable
results, good or bad; karman as an asset which the dead enjoy in the other
world; rebirth of a man in his son; rebirth through cremation; the path to
the world of the dead through the moon; and the path to immortality
through the sun.
The topics of the Five Fires and the Two Paths occur together in only
two passages of the classical Upani~ads. These two passages appear to be
based on the work of a redactor who used older Vedic material. This
material can be divided into two main groups: the Five Fires material,
which in the course of its previous history has had added to it the idea of
birth from the fifth or sixth fire; and the Two Paths material. The redactor's
principal innovation, besides bringing these two groups of material
together, is to transform. each of them from a linear series into a cycle.
Notes
1.
2.
E.g. Keith, 1925, vol. 2, 1'. 575; Basham, 1954,1' 242; Reat, 1977; Collins, 1982, p. 51; 'ltdl,
1989, p. 32 ; Olivelle, 1992, Ill' 441:
.
Garhe (Pl'. 434-5) rei"rs simply to the SET, withoutfunher indication. He dOl" not attrihute
the filily formed theory itself to non-Aly,m orig'in; instead, he postulates "the primitive helief'
- whether it he that or the Indian ailorig'imll trihes or that or the lower strata of't1w Aryan
pl:nple Lhallhe }iOU):; or men pass alter d(~aLh into plants and Clnimaix, and conjectures that
this l1e!ier"gave merely the !-Irst impuJ:.ie to the formation nrlhl~ d()(:lrinl: oftransmigralion".
1
3.
4.
5.
6.
'
Reat, oj,. cit., p. 164.

Ibid., p. 163.
Ibid., p. 163.
For sukruLrya loka- and related phrases such as sukrtiiT(lloka-, sec Gonda, 1966, esp. Pl'. 13off.
The idea that good deeds prepare a place in the world heyond death is found in the Avesta
d ai., 1961, p. 159.
(HiidoxtNtLrk, eh. 2); the l<:xl is translated hy lVI. I'doj(: in Yoyotte
IS
7
8,
9
10.
I I.

Siihn"n, 1979. Siihn"n (p. 180 with not" 7) cites Deussen, Keith, Ruben and Renou as datingth" ,K"auUp lat"c than th" other two. I am g-ratdul to Dr Sohnen-Thieme.lor g-iving- me a copy
of"this artick.
E,g-. Deussen, 1906, pp. 17-21; I'nlUwallnel; 1973, vol. I, p, 34- The theory is criticized in
Kcith, 1925, vol. 2, pp. 493,,6; Tilll, 1981, pp. 24C
Bodewitz, 1976, p, 183.
Olivdk, 1992,1']1. 37C
The term pecicope (Ii'om Greek perikop'- 'section'; t1w final e is pronounced) is borrowed here
li'nrn the terminology of"biblieal criticism. Its oldest use in connection with the Bible refers to
a portion ol'll~xt selected f(Jr liturg-ical reading; hut the commonest modern use refers to a
portion which the textual critic idcntifil" as havin!'; exi.sted separately be/on' being- placed in
its canonical contc:xt. :f\,iIuch twentieth-century biblical scholarship has heen concerned with
thl: S()llrC(~S) mainly cnnjecturally n~conslrucLed, {i'om whic:h particular pericopes appear La
hmw heen taken ('source criticistn 1)J the original literary form of' each pericope (,Corm
criticism!), and lh(~ ways in which pericopes have heen arranged and adapted in th('~ir
canonical context (,redaction critir.ism '). This article is meant to show that :-;imilar methods
12.
can be <tpplied to Upan~l'adic texts.

Eg[1;ding, Salapatha-brdh.mar;'Ll, vol. I, p. 16n. 'gives further Vt?dic rdi.;rcnce:-;.
13
Gonda, 1965, pp. I~22-5'
14.
h'llIwalinel; 1973, vol.
IS.
SomelilTIeS such an identiFication is referred to a:-; an llPan~l'Gd, a word orsimilar etymolog"ical

meaning" This, rather than the usual conjcr.ture 'sitting" down near [the: teacher]" may b(~
the origin 0[' the use of'the word u.pani:rad to mean a text in which such identifications are
tClug'hl. E.g-.:
I,
p. 36.
Now the teach in!'; ,or UPan~rad.l'. The Sakayanins upiil' Ag-ni as Vayu; some people "ay
;\iini is the SlIn ... (SBr 10, 4, 5,1-2).
I6.
HUlTIe:'s translatioll 'truly worship I~l-ith' is hased on Elilurc to n~r.og'nise the meaning
nrupa.r
wit h two accusatives.
17"
Ill.
Ig.
20,
21.
22,
2324.
25.
26,
27
28.
29,
30,
31.
32.
33.
CJ: B()d(~\,vil7" op. cit" p. I21 1110,

I'"l' this pail; cl: tda eva .iraddhii 'beat is truth' (SBT II, 3, I, I).
The ha][' of" the lumlr month in which thl' moon is waxing; the dark flll'tnight is the half" in
\vhich it is waning-,
Tht'; northern course of'lhe sun, the hall' or the year I'rom the winter solstice to the summer
solstiel'; da"l"i~.ii)'ana is the othl,r half" or the ye<tr.
The world of"the g-ocls.
The wodd ol'the alwl:stors'.
C;ollda, 19 65, PI'. 38 70.
ThlTe arc limhl'l' prohlems whieh arc nol ml,ntiOlwd: what happens if"the plant is eaten by a
1(~ll1ak, and hc)w (\() lloll-h(:r!)ivorotls animals reproduce?
The priilJiignifwtra is a twic<,-d<lily ritual in which the of!(,ring' i, the Il)od eaten by the sacrificer.
SCI' Bodewitz, ]aiminrya Briihma1Jf1 J, 1-~5, PI'. 213-338. The term is lIsed in C'lzUp 5, 19-24,
which g'jves insLIuctions [i)!' a lilT'm or agnilwtra in which ()ffi~rinp;s arc made to the faculties,
and lhcn:hy satisfy their cosmic c()ut1te:rparls,
Tran,hted in Ilodewitz, ]aiminrya Brahma~a 1, 1-~5, pp. [[4-6.
Ibid., pp. [[7, n3
Cl: Ilodewitz, 1976, p. 162.
etai[r haiva priijati~n prajiiyate yii~n pngiipatib. priijiiyata. E!,;iidin!'; translalI:S ,lightly diffl'l'l,ntly:
Hreproducc." himselr hy off'ipring- even as pJ'(~jarati reproduced himself''', The qur:stinn
underlying the: diffcn:ncc or lranslation is whether the rdt:rence is to actual off'ipring-, horn as
a n:ward f()1' peri(wming the agnilwtra, or lo the agnilwtra itsdC In the case ni'Pr~japatiJ who is
110lh lhe primordiallhthcr and the primnrclial)lajamana, tht: lwo an: the same:; he procreates
Ily s(lc:riiicing.
The: text says IlplanL'i on earth", which allow!') us to plac.(~ the carth before the plants in the
;;(~I'ics, dl~SpilC the \Vord-()rde~r,
S(~C aiKJVc) noll: 23,
H ume~'s and Radhakrlshnan's translaLions ig"norc lht: word tad,
dve STUll rLI'T1JL1va(n PitfIJiilll ahalfJ. deviiniim uta martyiillam / tabhyam idam vil'vam Iffat sameti yad anlara
ca.
f)ilo.ra~n miitara~n

34.
35.
36.
37.
19
The ambiguity arises lrom the three genitive plurals. I\.. E Geldner (vol. 3, p. 2[\2) construes
in the second way. Renou (1965, p. 93) construes it in the first way, on the grounds that the
use of .fru with the genitive in the sense of 'hear (i'om, "learn lrom' is not otherwise found in
the lJ.gvedic hymns. It is, however, lound in later *dic texts, e.g. flUp 10, repeated IfUp '3: iti
.fufruma d1zzriit;iim ye nlt.l tad mcaca/qire "Thus have we heard lrom the wise who have exp!,1ined it
to us"; similarly KeUp I, 3. Oldenberg (1912), p. 295, cited hy S6hnen, pp. 207f., note 43,
construes it in a third way: "There are two paths of the ancestors, gods and mortals", so that
both paths belong to all three classes ofheings; this i, the construction followed by Saya(la in
his commentary on the lJ.v.
Bloomfield, 1906, p. S'S, lists twelve occurrenC~$, and lour citations by the opening words
(praffka).
With the insignjficant change o~ sruff to s.m.
.
See especially SEr 10, 4, 3; also SEr I, 9, 3, '5; 10,2, 4, 3; 10,5, 2, 4; II, 2, 6, II; t{, I, 3, 4.
Abbreviations
AiBr
Ai Up
AV
BrhUp
ChUp
!fUp
JaiBr
KaUp
MuUp
PrUp
SBr
TS
~V
VS
Aitareya Brahma7J.a
Aitareya Upan~ad
Atharva Veda
Brhad-.iIrargaka Upani1ad
Chandogya Upani1ad
Iso. Upan~ad
Jaiminfya Brahma7J.a
Katha Upan~ad
MU7J.{iaka Upan~ad
Prafna Upan~ad
Satapatha Brahma7J.(l
Taittirfya Sa1'[!hita
~g- Veda Sa7]7.hita
Viijasaneyi-Sa1'[!hita
Bibliography
Basham, A. L., The Wonder that was India. London: Sidgwick &
Jackson, 1954
Bloomfield, M., Vedic Concordance. Cambridge, Mass.: Harvard
University Press, 1906.
Bodewitz, H. W, The Daily Evening and Morning Offering (Agnihotra)
According to the Brahma7J.as. Leiden: Brill, 1976.
___ ., Jaiminfya Brahma7J.a 1,1-65: Tramslation and Commentary: With a
Study: Agnihotra and Pra7J.agnihotra. Leiden: Brill, 1973.
Collins,.Steven, Se1jless Persons. Cambridge: Cambridge University
Press, 198'2.
Deussen, Paul, Philosophy of the Upanishads, tr. A. S. Geden. Edinburgh:
T & T Clark, 1906.
20

Eggeling, j., The Satapatha-brahmaTja. 5 vols., Sacred Books of the East
series. Oxford: Clarendon Press, 1882-1900.
Frauwallner, E., History qfIndian Philosophy. 2 vols. Delhi: Motil;{l
Banarsidass, 1973.
Garbe, R., "Transmigration (Indian)". Encyclopaedia qfReligion and
Ethics, ed.j. Hastings, 13 vols. Edinburgh: T &. T Clark,
1908- 26,vol. 12,PP434f
Geldner, K. F, Der Rig- Veda: aus dem Sanskrit iibersetzt und mit einem
lazgenden Kommentar versehen. 4 vols .. Cambridge, Mass.: Harvard
University Press, 1951-7'
Gonda,Jan, Change and Continuity in Indian Religion. The Hague:
Mouton, 1965.
___ ., Loka: World and Heaven in the Veda. Amsterdam: NoordHollandsche Uitgevers Maatschappij, 1966.
Hume, R. E., The Thirteen Principal Upanishads: Translatedftom the
Sanskrit, 2nd edn. Oxford: Oxford University Press, 1931.
Keith, A B., Religion and Philosoplry qf the Veda and Upanishads. 2 vols.,
Cambridge, Mass.: Harvard University Press, 1925.
Olivelle, Patrick, SaT(lnyiisa UpaniJads: Hindu Scriptures on Asceticism and
Renunciation. New York: Oxford University Press, 1992.
Radhakrishnan, S., The Principal UpaniJads. London: Allen & Unwin,
1953
Reat, Noble R., "Karma and rebirth in the UpaniJads and early
Buddhism". Numen 24 (1977), pp. 163-85.
Renou, Louis, Etudes vediques et paTjineennes XIV (Hymnes a Agni).
Paris: E. de Boccard, 1965.
Sohnen, Renate, "Die Einleitungsgeschichte der Belehrung des
Uddalaka Arul).i: ein Vergleich derdrei Fassungen Kau~U LI,
ChU 5.3 und BrU 6.2.1-8". Studien zur Indologie und Iranistik, vol.
7 (1979), pp. 177-213.
Tull, Herman W, The Vedic Origins qfKarma: Cosmos as Man in Indian
Myth and Ritual. Alpany: State University of New York Press,
I9 8 9
Yoyotte,j., et al., Le Jugement des Morts. Paris: Editions du Seuil, 1961.
The Vitalistic Antecedents of the

Atman-Brahman Concept
Peter Connolly, Chichester Institute of Higher Education
The classical literature of the Vedanta darfana employs the terms sat, cit,
and iinanda to characterize the nature of ultimate reality (iitman-Brahman),
though such descriptions, as Deutsch points out, " ... are not so much
qualifying attributes of Brahman as they are the terms that express the
apprehension of Brahman by man."! The classical Vedanta teachers such
as Sankara and Ramanuja are also insistent that the vitalistic principle
(prar;a) is merely a phenomenal rather than an ultimate reality.2 At the same
time, these teachers maintain that their views are nothing more than
interpretations of the Vedantic scriptures, primarily the Upani!ads. In what
follows I shall seek to demonstrate that the views of both Sankara and
Ramanuja are, in most cases, misinterpretations of the relevant Upani!ads
by (a) showing that in many Upani~ads, priir;a is regarded as an ultimate
reality and (b) indicating how Upani~adic conceptions of iitman and brahman
frequently incorporated features that were originally employed to
characterize prar;a.
Pre- Upani!adic Concepts of Prii'l}f1

The recognition of priir;a as an ultimate principle actually pre-dates the
Upani~ads. In the Atharva Veda (I I -4) priir;a is described as the ultimate source,
ground and controller of all. This hymn, according to A. H. Ewing, presents us with 'the highest meaning of priir;a,' with priir;a as the 'primeval cosmic principle. '3 The passages where this primeval status is most clearly
established are:
Vs
Homage to priir;a in whose control is this all, who hath been

lord of all, in whom all stand firm.
22
Vs
10
Breath !prar;.a) clothes (anu-vas) human beings !praja) as a father

clothes a dear son; breath is lord of all, both what breathes al;td
what does not.
Vs
12
Breath is Viraj, breath is the directress, breath all worship,

breath is the sun, the moon; breath they call Prajapati.
Vs IS
Breath they call Matarisvan; breath is called the wind; in breath

what has been and what will be, in breath is all established
!prati~thita) ..
That the conception of priir;a set out above was held by a number of
Vedic' teachers, is evident from an examination of later texts such as the
AitaTl!)!a Arar;:yaka and the Satapatha Briihmar;a. The first of these contains five
component Arar;:yakas, the second of which divides naturally into two parts.
Part one (adhyayhas r-3) deals with the uktha (the high chant of the lfgueda)
whilst part two constitutes what is more generally known as the Aitareya
Upani~ad. The material dealing with priir;a is found almost entirely in part
one.
To understand the views about priir;a that are held by the author of this
text one needs to begin with the story of the faculties trying to determine
which of them is the hymn (uktha), this being employed as a synonym for the
supreme principle (2.1.4). Speech, sight, hearing and mind all quit the body
but it only falls when the priir;a departs. Similarly, each in turn returns to the
body bu t only on the return ofpriir;a is it re-animated. At this demonstration
of the priir;a's supremacy the other faculties proclaim that:
... breath only is the hymn. Let men know that breath is the hymn. The
gods said to breath, 'Thou art the hymn, thou art all this, we are thine, thou
art ours.'5
The identity of priir;a and uktha can then, presumably, be read back into
where it is said that all existence springs from the uktha. 6 Certainly,
the verses following 2.1.3 simply reiterate this view in a variety of ways, a
reiteration that is continued throughout adhyayas 2 and 3.2.1.5 commences
with the statement that, "The gods carried him forward."7 The most logical identity of 'him' is the priir;a-who-is-the-uktha and this is confirmed by
SayaI}.a, the only traditional commentator to comment on the entire
Arar;yaka. 8 Also, in 2.1.5 there is a distinction made between priir;a and
apiina and each is placed on a level with other faculties (speech, etc). So we
read:
2.1.2,
Day is breathing forth !prar;.a), night is breathing down (apana). Speech is fire,
sight yonder sun, mind the moon, hearing the quarters ... 9
Then we are told that:

... this is the union of those sent forth. 10
The Vitalistic Antecedents of the Atman-Brahman Concept
'23
'Those sent forth' are obviously the faculties, speech, etc, which are
regarded as the forms of deities residing in the body. "This," as Keith
points out," "is obscure" but a reasonable interpretation, given the context,
would be that it is the prii7:za-which-is-the-uktha. Prii'lya and apiina would then
simply be derivatives of this in the same way that speech etc, are.
In '2.1.6 we are again reminded of the ultimate nature of priilJa when we
are told that:
... all of this is covered by praT}l. This ether is supported by praT}l ... 12
Such a power is obviously a creative force and the remainder of the
adhyaya and the whole of the next are devoted to establishing a link
between, if not the identity of, priilJa, the cosmic puru~a and Prajapati. The
powers of priilJa (which is called 'this person' in '2.1.7 and 'he who shines' in
'2.'2. II) are that he creates earth, fire, sky, heaven, the sun, the quarters, the
moon, the waters and VarU!;ta (lord of the moral order (rta) and the deep
water). Furthermore, he is, "all these verses, all these Vedas, all sounds
(gho~a) . .. "13 a list which, presumably includes the thirty-six syllabled brhati
metre, which is 'the whole self'. 14 Finally, we may note that in '2.'2.3 Indra
tells Visvamitra that he (Indra) is j)rii1}a, Visvamitra is prii7.la and all creatures
are prii7.la.
We can thus conclude that in this part of the Aitareya .Arargaka, the only
part dealing at all comprehensively with the topic of prii7.la, the vital force is
considered to be the source of all and the ground or being of all.
In A.A. '2.1.8 we are informed that it is knowledge of priilJa, of how it
functions and how it exists in the human body, 'the hiding place of
brahman', that brinS"s immortality. Why this should be so can be inferred
from a study of the Satapatha Briihma7.la. In 1.4.3.8 there is a reference to the
'antasthii prii7.la' (the middle or central prii7.la) which Ewing regards as
synonymous with the 'madhyama prii1}a' of the Upani~ads.15 He also suggests
that here the clear intention of the verse is to present prii1}a as 'the controlling
influence', the 'Inner Ruler' which is 'an active, conscious, even Divine
Force which dominates the entire organism'. 16 Eggeling translates thus:
... what central breath [antasthii priiT}l] there is (in the body) .. , that one
indeed is the internal motive force of the breathings ... And whosoever
knows that internal motive force of the hreathings, him they regard as the
internal motive force. 17
If Ewing is correct then we have here a concept which is almost identical
to the Upani!adic antaryiimin (inner controller). The Satapatha Briihma7.la also
equates prii7.la with the immortal element in man (S.B. '2.'2.'2.8-15; IO.'2.6.18)
and states that the priilJas, which are 'the highest thing of all this universe'
(8-7+'21), are immortality (9.}-'2.3'2). Mythically, this is presented in terms of
the division of Prajapati in S.B. IO.1.4.1:
24

Now at the beginning,Prajapati was (composed of) both these, the mortal
and immortal- his vital airs alone were immortal, his body mortal ...
Such a division is also reflected in the human being; hence S.B. 6.7.1.II
informs us that,
... that part of the vital air which is immortal is above the navel and
streams out by upward breathing; but that which is mortal passes by and
away ftom the navel ...
A crucial part of this knowledge of pra1Ja is, as was noted above in
connection with the Aitarrya Ara7JYaka, that it exists in two principal modes, a
unitary one, when it is the foundation of all existence and the inner
controller of the individual, and diversified one, when it is the various
cosmic forces and the breaths and faculties which exist in the body. Hence,
8.7.3.21 informs us that,
... this vital air [pra(lll] whilst being one only, extends over all the limbs, over
the whole body/a
and 11.1.2.3 explains that:
... this sacrifice is the blowing (wind) [vqyu = pra(lll]: he blows, as it were, as a
single one, but when he has entered into man he is divided into ten parts. 19
Most statements about the number of pra1Jas in the S.B. suggest that
there are ten of them. However, some texts increase the number by varying
amounts; hence we find references to eleven (8-4-3.8; 11.2.1.2), twelve
(12.3.2.2) and thirteen (3.8-4-1).
.
The unified pra1Ja enters the body by way of the head 20 and then spreads
throughout, infusing every limb. 21 In doing so it nourishes and vitalizes the
body.22 Those parts not reached by the pra1Ja dry up and wither away.23
The distribution of pra1Ja appears to be effected by means of definite
pathways, though the text is not clear on this.
The vitalisation of the body appears to be the result of the activities of
the five praIJas: pra1Ja, apana, 1!)!ana, udana and samana. However, as was noted
in connection with the Aitarrya Arargaka, the one pralJa also manifests as the
different sensory faculties. There are various lists of these in the S.B. The
'pralJas in the head' or 'the pra7Jas of Prajapati' (7.5.2.6, 9.2.2.5 and 9.3.3.8)
are mind (manas), eye (calqus), breath (praIJa), ear (srotra) and voice (vac). The
eyes, ears, nostrils and mouth are the seven pra1Ja5 mentioned in 7,5.2.8-12
and the same list, with speech substituted for mouth, is found in
9.3.1.IO-12. Hence, when pralJa diversifies in the body it manifests as both
vital forces (the five pra1Jas) and perceptual faculties.
This is not all, for, as we might expect from the inclusion of manas
among the list of pra1Jas, pra1Ja is also the source, if not the substance of mentation. Thus, " ... SavitT is the mind and the thoughts are the vital airs ... ",24
and " ... die divine inspirers doubtless are the vital airs, for these inspire all
thoughts ... ".25
The Vitalistic Antecedents qf the Atman-Brahman Concept
25
A similar sentiment is found in A.A. 2.3.5, where we are told that:

.. ' the self that is speech is imperfect, since a man understands if driven to
thought by bn;ath (priir;.a), not if driven by speech,26
Thus, in many parts of pre- Upani~adic Vedic literature the vitalistic principle, prar;a, is presented as the self-existent source of everything, the
ground or being of ephemeral mortal forms, It is diversified as prar;a,
apana, etc" the perceptual faculties and, possibly, thoughts. One who
knows this, who knows the deity, knows the supreme prar;a, knows the
immortal (which is the true essence of a person) and attains immortality.
We have here, then, the same salvific scheme that dominates the
Upani~ads. At the heart of this scheme, however, resides the vital principle,
the prar;a, not atman or Brahman.
Pra:r},a as an ultimate principle in the Upani}ads

An examination of all references to prar;.a in the UPani~ads reveals that the
prominent view is quite similar to that already outlined with regard to preUPani~adic literature: prar;.a is the primeval source of all and the immortal
inner essence of individuals which manifests in the body as the various
breaths (apana, etc.) and faculties, In a number of instances this is presented
in an unambiguous manner, in others, however, close analysis is required to
demonstrate their espousal of such a view.
However, this is not the only view of prar;a to be found in the Upani~ads.
In a number of places it is presented as what I shall call a cosmic principle,
i.e. one which is derived from the fundamental principle but which itself is
the source of further manifestations at the level of phenomenal or
individualised existence. In yet other places we find accounts of prar;.a only
as it appears at the phenomenal leveL In some cases these could be taken to
be statements about the phenomenal manifestation of a higher principle
which is assumed but not referred to. In others, however, it is clear that the
author(s) are operating with a different model of existence than that
assumed in the passages referred to above and that, in these instances, prar;.a
does not have the high status ascribed to it by the other accounts.
Clearly the characterisation of prar;a as a cosmic or phenomenal
principle by Vedantins such as Sankara and Ramanuja can be supported
by reference to the above-mentioned passages, e.g, B.A, U. 1.2; C. U. I.I,
3.r2-r8 and 6; TU. r-3; Kafha Upani~ad 4 and 5; pu. 3; Mait. U. 2,6-;.
N evertheltss, by far the most frequent way of presenting prar;a in the
Upani~ads is as an ultimate principle. Many of these accounts are picked out
for consideration by Badarayar;a because of their 'problematic' nature (i.e.
they present prar;a, not atman or Brahman, as the ultimate principle). His
hermeneutical strategy in the Brahmasutra is essentially designed to show
26
that Upanifadic teachers who presented priir;a as the ultimate did not really
mean what they said. In their own ways Sarikara and Ramanuja ~cho this
sentiment, My aim in this section is to demonstrate that Badarayal).a,
Sarikara, Ramanuja and other Vedantins who followed them have, in fact,
distorted the message of the above-mentioned teachers.
Passages in the Upani~ads where priir;a is presented as an ultimate principle
can be divided into two groups: straightforward and metaphorical. The
straightforward passages are B.A.U. I.6.1-3, 2.3.1-6, 39, 4-4-7 and 4.4.22;
c.u. 3.1S-4, 4.3. 1-4, 4.1O-4-S and chapteq; Pu. 2; Mur;{i. U. 3.1.4 and Kauf
U. 4.20. The metaphorical passages are those which present the various
faculties arguing about their respective status, particularly about which of
them is supreme: B.A. U. I.3, 1.S.'2'2 and 6.I.I-14; c. u. 1.2.1-14 and S.LI-lS;
Kau~. U. 2.14 and 3.3, and P U. 2.1-13 Of these B.A. U. 1.3 and C. U. 1.2.1-14
are versions of the same account. The same is true for B.A.U. 6.I.1-14 and
C. U. S. r.r-1S
Some ofthe straightforward passages offer simple statements about the
. ultimacy of priir;a, e.g. priir;a is everything (c. U. 3.1S-4, Kath. U. '2.3.2); priir;a is
Brahman (c. u. 4.1O-4-S); priir;a is the iitman (Mur;(i. U. 3.14). Others develop
related themes from different perspectives and these warrant more detailed
discussion.
One theme is that of the two aspects of the ultimate: an unmanifest and a
manifest; an immortal and a mortal; a mobile and an immobile; a
transcendent and an existent; the one god and the many gods (B.A. U.
2.3.1-6, 3.9). In every case priir;a is identified with the first item in each pair.
Another theme is that of the single source and the many manifestations.
Thus, in B.A. U. 3.9, where it is stated that priir;a is the one god (eka deva), we
are told that the 33 gods are but manifestations of that one god and that the
eleven Rudras are the ten priir;as (priir;a, apiina, vyiina, udiina, samiina, viic, cak}us,
srotra, niisii; manas?) and iitman (here meaning body or, possibly, ego
(ego/personality). Atman (self) is declared to be none of these priir;as (neti, netz).
This latter iitman is obviously to be distinguished from the eleventh of the
prii1Jas and would thus seem to equate with the one god, the priir;a, which has
the prii1Jas as its manifestations. Similarly, B.A. U. 4.3-4 equates iitman,
Brahman and priir;a. This iitman/Brahman/priir;a is the light within the heart,
abides in the space within the heart, is immortal, is the lord, controller and
ruler of all. and is the one made of consciousness among the priir;as
(uijniinamaya priir;e~u). This iitman/Brahman/prii1Ja has, as its manifestations,
vijniina, manas, priir;a, cak}us, srotra, prthivT, apas, viiyu, iikiisa, tejas, atl[jas, kiima,
akiima, krodha, akrodha, dharma, adharma and everything (sarvamaya). The gods
worship it as the light oflights !jyot~iimjyotiM and as life immortal (iiyuramrta).
The equation of iitman, Brahman and priir;a can also be deduced from a
consideration of verses which connect with B.A. U. 2.3.6, where it states
that" ... the priir;as are truth and this (e~a) is the truth of those (te~iim)." This
priir;a which is the truth of the truth. is the person in the right eye, the
27
essence of the truth. In B.A. U. 2.1.20 iitman is described as the truth of

truth (saryalja saryam) and in C. U. 4.15. I as the person seen in the eye, who
is without fear, immortal, Brahman.
The final passage presenting priilJa as an ultimate principle that I will
comment on is C. U. 7. This chapter does not, on the surface, present priilJa
as an ultimate principle for it distinguishes it from iitman. However, 'in my
opinion this view has been redacted into an original account which
presented priilJa as the ultimate principle. The chapter deals with
Sanatkumara's instructions to Narada on the progressive worship of
Brahman. The teaching begins with the statement that worshipping
Brahman as name (niima) is quite legitimate. However, we are told that
speech (viie) is greater than name; in turn, mind (manas) is greater than
speech; will (saT[1kalpa) is greater than mind; thought (citta) is greater than
will; contemplation (dhyiina) is greater than thought; understanding (vijiiiina)
is greater than contemplation; strength (bala) is greater than understanding;
food (anna) is greater than strength; water (ap) is greater than food; heat
(tejas) is greater than water; ether (iikiiia) is greater than heat; memory
(smara) is greater than ether; hope (iiia) is greater than memory; breath/life
(priilJa) is greater than hope.
At this point the narrative changes. Whereas each of the preceding sections ended with the words "Venerable Sir, is there anything greater
than ... ? Yes, there is something greater than ... Do, Venerable Sir, tell me
that ... " section 15, where it is stated that priilJa is greater than iiSii, ends with:
Pra7J.a is all this. Verity, he who sees this, thinks this, understands this,
becomes an excellent speaker [ativadin]. Even if people should say to him,
'you are an excellent speaker', he should not deny it'>?
The following section introduces a different kind of treatment, where
concepts such as truth (sarya), understanding (vijiiiina) and faith (Sraddha) are
described. It is clear that this set of verses forms part of a different analysis
than that offered by the first set. Instead of commenting on the relative
merits of the different concepts Sanatkumara introduces them as being
desirable to understand. Hence, each of the sections from sixteen to twentythree, which introduces the concept of the infinite (bhilma) , ends with the
words "Venerable Sir, I desire to understand ... ". The infinite is described
in the following two verses, after which we read in 25.1, "Now next, the
instruction with regard to the self-sense (ahaT[1kiira)."2B The following verse
(25.2) introduces "the instruction in regard to the self (iitman)", which is
described as being "this all". Then, in 26. I, we are told that priilJa springs
from the iitman (iitmata~ priilJa~), hope (iiia) springs from the self, memory
springs from the self and so on back down the list in 7- I.3-15.
What is peculiar about 26. I is that none of the qualities listed between
16. I and 25. I are mentioned as springing from the self. This creates a
distinct sense of discontinuity. The continuity can be restored, however, by
28
taking 7-25.2 and 7-26.1 andplacing them immediately after 7.15+ In fact,
this is what William Beidler does in his interpretation of this ,chapter
though he does not indicate that he is doing so. The only problem with
such a move is that 7-15.4 and 7.25.2 seem to contradict each other. The
former states that priir;a is everything (priir;a hy evaitiini sarvii[!i bhavatz); the
latter makes an almost identical claim about iitman (iitmaivedam sarvam itt)
whilst at the same time, in the next verse, stating that priir;a is derived from
the iitman. It is this incompatiliblty which, I would suggest, explains why
7-16-25 was interpolated at this point. If the redactor wanted to
subordinate priir;a to iitman it would be foolish for him simply to add 7.25.2
and 7.26.1 onto 7-15-4 for the incompatibility just discussed would be readily
apparent. What better way to disguise it than by creating a break between
the statement exhalting priir;a and that exhalting iitman? If the interpolation
could have a style which created the impression of continuity all the better.
This, I would suggest, is exactly what we have here: an original text
presenting priir;a as the ultimate principle which has been modified by two
additions. One of these begins either halfway through 7-25. I, where
ahaf(lkiira is introduced, or at 7.25.2, where iitman is introduced. The purpose
of this addition is to subordinate all the principles mentioned in 7-1-15 to
the iitman. The second addition comprises 7-16 to 7.24.2 or 7.16 to 7.25. The
purpose of this addition is to obscure the incompatibility of the statements
in 7.15-4 and 7.25.3. C.U 7.1-15 thus constitutes a complete unit in its own
right and presents priir;a as the ultimate principle.
The view of priir;a presented above finds its clearest Upani~adie expression
in the Kaup:taki Upani!ad, the only early Upani~ad not commented on by
Sankara. In Kau~. U 4-20 Ajatasatru teaches Drpta Balaki of the Gargya
clan, as he did in B.A. U 2.1.I7- The teaching is that during sleep speech (viie)
together with all names, eye (cak!us) together with all forms, ear (frotra)
together with all sounds and mind (manas) together with all thoughts enter
the priir;a. On waking, the priir;as proceed from the self (atman priir;a), the
gods (deva) from the priir;as and the worlds (loka) from the gods. VVe are then
told that the priir;a, the intelligence-self (prajiiiitman) enters the bodily self
(sanriitman) up to the hairs and nails like a razor in a razor case: "on that self
these other selves depend as upon a chief his own men". 29
It is clear that there is much in common between this passage and
B.A. U 1.4.7. It seems likely that one borrowed from the other or that they
both drew on a common srouce. The Kau~Itaki account is more detailed
and makes it clear that the iitman/Brahman of B.A.U 1.4.7. and 10 is the
priir;a/prajrT,iitman, this last term being used to indicate the supreme self in
A. U 3. Similarly, the context in which these equations occur is also found
at B.A. U 2. I7 where the term vy'iiiina-maya-puru!a is employed to refer to the
iitman. It would appear, then, that the terms priir;a, iitman, prajiiiitman, vy"fiiinamaya-pU1wa and Brahman were regarded as being synonymous in meaning
by a number of Upani!adic teachers. This Kau~itaki account, because it
The Vitalistic Antecedentf oj the Atman-Brah:man Concept
29
draws the ~ious elements from other places together, thus gives substance
to what elsewhere was simply inference.
The other passages where prar;.a is presented as an ultimate principle I
have dubbed metaphorical. They deal in different ways with a contest
between the various faculties for supreme status. By and large, these
accounts reiterate what has been said above about the natUre ofprar;.a in preUpani~adic and Upanadic literature. They can thus be seen to support my
interpretation of that material. In these stories priir;.a is presented as a major
cosmic principle or as the one ultimate principle from which all else derives.
At the material level of creation it diversifies itself to produce both the
physical aspects of existence and the more subtle aspects which animate the
physical ones. All these accounts describe the subtle manifestations in terms
of the sensory faculties plus mind (manas), speech (vac) and breath (praTJa), the
manifestation most directly derived from and closest in nature to the
original cosmic praTJa. Almost every account makes priiTJa an immortal
principle and the Kau~ltaki accounts make it the intelligence self (prajnatman)
and the self (atman). Hence, as in the pre- Upanadic accounts, prar;.a gives rise
to mentation - through the manas - and, perhaps expressed more clearly
here than in the earlier material, it is that reality which can be described as
consciousness or intelligence (pr:aJnit; or as self (atman).
The Vitalistic Blueprint

It is clear from the foregoing that within the vedic tradition a considerable
number of teachers regarded praTJa as the ultimate principle of existence,
the immortal source and foundation of everything else, the inner controller
of all living beings, unitary in itself but diversified within beings in a variety
of ways. Knowledge of this immortal, unitary,praTJa constitutes the goal of
the religious life. Yet praTJa is not the only term employed to designate this
ultimate principle. In the Upanads descriptions that are virtually identical
to those of praTJa mentioned previously can be found associated with the
terms atman and Brahman. The main passages offering such descriptions
are B.A.U. 1.4.7,2.1.17, and 4.1.2; A.U. 1.2; Mait. U. 6.1-3, and 8, and
Kena U 1.
B.A.U. 1+7
The Brahmar;.a of which this verse is a part opens with the words atmaivedam
'agra iisit pu~avidhah: "In the beginning this (world) was only the self(atman),
in the shape of a person."30 The following verses then describe the process
of creation through the bifurcation of the self into man and woman. These
transform successfully into all living fonns. Verse five then informs us that
"he became the creation" (tatal;. sr.sf,ir abhavat) and verse seven that at the
time 'this' was unmanifest (taheda'l'{! tarhy avyiilqtam iisit), it became manifest
30
through name and form (niimarilp~). He (the iitman) entered that which had
become manifest,
even to the tips of the nails, as a razor is (hidden) in the razor case, or as fire
in the fire source. Him they see not for (as seen) he is incomplete, when
breathing he is called breath (prar;a), when speaking voice (vac), when seeing.
eye (cak!us), when hearing ear (frotra) when thinking mind (manas). These are
merely the names of his acts ... The self is to be meditated upon for in it all
those become one. 3l
In verse ten, echoing verse one, we are told that in the beginning this
(self? world?) was Brahman (brahma va idam agra asit). This Brahman, like
. iitman in verse one, became the whole creation (tasmiit tat sarvam abhavat).
These verses thus serve to equate atman and Brahman. The whole section,
however, is highly reminiscent of statements aboutpraTJa who, as seen, is
incomplete, who manifests as praTJa, vae, cai9us, frotra and manas and who is
ultimately the unitary immortal source of all. In other words, the pattern
employed in this text to establish iitmar;lBrahman as the fundamental
principle is one which is clearly modelled on earlier accounts of the nature
of priiTJa.
B.A.U.
'2.1.I7
This verse describes how thevijfiiina-maya-puru!a (the person made of

consciousness) takes the consciousness (uiJfianam) ofthe praTJas, here referring
to the various faculties, into itself and resides in the space within the heart
(antar-hrdaya-iikiiSa) during sleep. The verse ends with the statement "when
the breath (priiTJa) is restrained, speech (vae) is restrained, the eye (eai9us) is
restrained, the ear (fratra) is restrained, the mind (manas) is restrained."
Verse twenty indicates that this viJfiana-maya-puru!a is, in fact, the iitman,
from which come all priiTJas, all worlds (laka), all deities (deva) and all beings
(bhilta). The verse ends with the statement "priir;as are the truth (satya) and
their truth is this (iitman)",'2
Here again we have the connection between the priiTJas (faculties/vital
breaths) and the self (iitman). The relationship between the two is identical
to that which in other contexts operates between the unitary praTJa and the
diversified praTJas.
B.A. U. 4. I-'2
There are seven verses in the first BrahmaTJa of this chapter, for which
Radhakrishnan provides the sub-heading "Inadequate definitions of
Brahman".33 It is part of Yajiiavalkya's teaching. Here he asks King
Janaka of Videha what other teachers have said about the highest
Brahman (parama brahman). Upon receiving this requestJanaka states that
Jitvan Sailini says "speech (vae) is the highest Brahman", U dailka
Saulbayana says "breath (priiTJa) is the highest Brahman", Barka VaqI,la
3r
says "the eye (cak!us) is the highest Brahman", Gardhabhivipita

Bharadvaja says "the ear (sratra) is the highest Brahman", Satyakama
J abala says "the .heart (hrdaya) is [the highest] Brahman". To all these
claims Yajnavalkya has just one reply: "This Brahman is only one-footed,
Your Majesty", i.e. it is incomplete or inadequate.
However, Yajiiavalkya himself makes no statements about the nature of
the highest Brahman. Rather, in 4.'2 he asks Janaka where he will go after
death. Janaka admits his ignorance and asks Yajiiavalka to enlighten him.
Yajiiavalka points out the person in the right eye is Indra (Indha) and the
person in the left eye is his wife (ViriiJ). These two are united in the space
within. the heart and move in that channel (nii{lf'J which goes upward from
the heart. Nourishment flows to them through the channels (hitii~) of the
heart. "Therefore that (self composed of Indha and ViriiJ) is, as it were, an
eater offiner food than the bodily self (sarira atman)."34 The next verse (v-4)
then states that his (the self's) western side are the western breaths, the
eastern side, the eastern breaths, etc. but the self (atman) is not this, not this
(neti, netz); (see C. U. 3.13. r-6 for a correlation between individual breaths
and each of the five directions). The atman is then described in exactly the
same words that are found in the description ofthe atman in B.A. U. 3.9.'26:
it is incomprehensible, indestructible, unattached, unfettered, free from
suffering and injury.
These two Brahmar;as constitute what is obviously a version of the
competition of the faculties. Vae, prar;a, eak!us, srotra, manas and, here, hrdaya
are all deemed to be incomplete expressions of Brahman. The complete
expression appears to be iitman, who is said to be none of the priir;as (here
linked with the various directions; in 3.9 stated to be prar;a, apiina etc.) and
to be incomprehensible etc. Again, there can be little doubt that the
manner of introducing and describing the iitman is modelled on other
accounts where the unitary prar;a occupies the place of the atman. Here,
however, instead of the atman which is prar;a occupying the body completely
like a razor in a razor case, as in Kau! U. 4.'20, it abides in the heart.
A.U 1.'2
This is complex text which is difficult to interpret but, given the foregoing
analysis of other Upani!adie passages and the argument set out below, its
status as an account of the nature of iitman which is 'modelled' on similar
accounts of the nature of prar;a is not difficult to appreciate. The story line is
that of the iitman creating the universe. First he creates the worlds (loka) ,
which are water (ambhas), light space or light rays (maria), death (mara) and
water (apas). Water (ambhas) is above heaven or sky (cfyaus) , which is its
support. The light space (maria) is the atmosphere; death (mara) is the earth
(PrthivD and beneath that are the waters (apas). Many translators take the
following verses to teach that puru!a was then created out of the waters by
the self The text reads" so 'dbhya eva puru~am samu{fhrlyamilrehayat",35 and this
32
could legitimately be rendered as something like, "from out of these

(udbhya-ablative plural) he took and gave form to the puru~a". In other
words, the puru~a was derived from all the worlds created by the self and not
just from the waters. Then the iitman broods over (abhyatapa) the puruJa and
'hatches' (nirabhidyata) the world guardians (lokapiila), which are the various
faculties and phenomena, from the parts of the puruJa's body. The pattern .
of derivation can be set out as follows:
mouth (mukha)
-l>-
speech (vac)
-l>-
fire (agm)
nostrils (niisike)
breath (praTjl)
wind (vayu)
eyes (akjim)
sight (cak!us)
the sun (aditya)
ears (karTjU)
hearing (frotra)
directions (diJitj
skin (tvac)
hairs (lama)
plants and trees
heart (hrdJ
mind (manas)
moon (candra)
navel (nabht)
digestive faculty (apana)
death (mrtya)
phallus (fifna)
semen (reta)
waters (apa)
(o~adhivanaspatya)
Once created, these divinities (devata): speech, fire, breath, wind, sight,
sun, hearing, directions, hairs, plants and trees, mind, moon, digestive
faculty, death, semen and waters, all fall into (prapatan) the great ocean
(mahatya11Java) where they experience hunger and thirst. They then ask the
iitman to find them somewhere to become established (pratiJ!hita) and eat
food. They are offered a cow and then a horse but both are deemed to be
inadequate. Then the/a puruJa is offered and this is found to be an
acceptable home. Each of the principles/deities in the right hand column
enters the one from which it was derived and these, in turn, enter into the
parts of puruJa's body from whence they came. Hunger and thirst, however,
are then left without an abode, so the iitman allows them to enter the puruJa
along with the other principles/ deities. Finally, the iitman produces material
form (milrtz) from the water(s) (a/iipa) and this acts as food.
We are then presented with a version of the contest of the faculties. Food
tries to escape being consumed by the puruJa, who tries to seize it with each
of his faculties in turn: speech, breath, sight, hearing, skin, mind, generative
organ and digestive faculty {apiina).36 It is the digestive faculty alone that is
able to seize the food. Then comes a peculiar statement: "Vayu is the
grasper of food, Vayu is the one who lives on food." Vayu, of course, is
derived from priil}a in the first list so why it is introduced at this point is
unclear.
So constituted, the individual puru!a seems to be complete. It is a microcosm of the macrocosm and has the ability, indicated by the list of derivations, to apprehend phenomena and satisfy the needs of hunger and thirst.
The iitman wonders how (or whether) the puruJa, as it is constituted, can live
The Vitalistic Antecedents Wthe Atman-Brahman Concept
33
without him. He then considers the means by which he could enter this
being and dismisses each of the routes listed in columns one and two since
he does not identify himself (solely?) with any faculty or part of the body.
He thus decides to' enter through the door called 'the cleft' (vidrtz), located
at the crown of the head (siman). Once inside, he perceives 'this very person
- 'etam eva puru~a' (i.e. himself?) as brahma tataman' Gust that supreme one).
The following verse tells us that this perc~iver (the iitman) is Indra, whose
abode is often stated to be in the right eye.
The purpose of the whole chapter seems to be the presentation of a view
which not only makes the self the source of all existence but also makes
external phenomena derive from the puru~a. The facts that the iitman is the
ultimate identity of all the worlds, faculties and phenomena, that he enters
the body through the top of head and that he is identified with Indra all
remind one of priir;.a, which in other contexts does all these things and has
all these characteristics.
The final chapter of this Upani~ad, just four verses long, supports this
connection. It provides information about the nature ofthe iitman. The
sanskrit is not clear, however, and verse one could be either a series of questions and answers or just a series of questions.
Hume points out that all the published texts of this Upani~ad open with
the words ka 'yam though Muller, Bohtlingk and Deussen amend it to ka yam.
Given the context, I would favour the amendation for the issue would then
be cast in familiar mould: that of the faculties competing. Hence, verse one
would read:
Who is he whom we worship as the iitman? Which one is the iitman? [Is it he]
by whom one sees, or [he] by whom one hears, or [he] by whom one smells
odours, or [he] by whom one speaks speech, or [he] by whom one distinguishes between the sweet and the unsweet?
The impled answer is 'None of these', and this would certainly fit with
the conclusions of all other similar competitions. However, instead of
straightforward answers, verse two provides a list of mental phenomena, all
of which are stated to be names of intelligence (prajfiiina). Verse three then
informs us that this prqjfiiina is everything. It is Brahman, Indra, Prajapati
and all the gods. It is the five elements, the foundation (prati~thita) of all
things, the guide or eye (netra) of all things. The final verse tells us that
'he',37 by means of the intelligence-self (praJfiiitman) , left this world and,
having obtained all his desires in the world of heaven, became immortal
(svarge taka sarviin kiimiin aPtviimTta~ samahhavat).38
The prqjfiiitman thus seems to be the self referred to in verse one. Such a
conclusion would put this text in agreement with the other UPani~ad of the
I;?gveda, the Kau~ftaki, in employing the term praJfiiitman to refer to the
supreme self. There, of course, the prqjfiiitman is explicitly equated with priir;.a
and such an equation would not be out of character here. If so, we see,
34
once again, that the concept of prii7J.a has been employed as a kind of blueprint for the presentation of iitman as the supreme principle.
Mait. U 6.1-3 and 8

He (the self) bears himself in two ways, as he who is breath (pra7}l) and he
who is the sun (iiditya) , Yonder sun, verily is the outer self; the inner self is
breath
There are, assuredly, two forms of Brahman, the formed and the
formless. Now that which is formed is unreal (asatya); that which is formless is
the real (satya); that is the Brahman, that is the light. That which is the light is
the sun
The self (iitman) is the lord (uiina), the beneficent (famMu), the real
(bhava), the terrible (rudra), the lord of creation (prajiipat!), the creator of all
(viivasrk), the golden germ (hira7!Jagarbha), truth (satya), life (prii7}l), spirit (haTJ1sa),
the ordainer (iiistii) , the pervader (viP-lU), Narayal).a [abode of man], the
shining (arka), [the] vivifier (savita), the upholder (dhiitii), the maker (vidhiitCi) ,
sovereign (samrii]) Indra, the moon (indur) ,., He who has all forms, the
golden one, who is all-knowing, the final goal, the only light, who gives heat,
the thousand rayed, abiding in a hundred places, the life (priir;a) of creatures,
the yonder sun, rises,39
0"
Once again, the concept of self (iitman) as it is presented in this material

appears to be employed as a synonym for the unitary prii7J.a. In the Satapatha
Briihma7J.a Agni is often put in the same role as the unitary prii7J.a. Hence, in
SoB. IO.3.3 r-8 we are told that Agni manifests in the body as speech, eye
(sight), mind (mentation), ear (hearing) and 'the agni who is everything
here': prii7J.a. This is exactly the kind of statement which, elsewhere, describes
the manifestation of the unitary priir;a in the body. The first four pass into
prii7Ja during sleep and emerge .again on waking. Cosmically, fire
corresponds to speech, the sun to the eye, the moon to the mind, the ear to
the quarters and Vayu to prii7Ja. Similar correspondences are found
throughout the Satapatha Briihma7J.a. Numerous references make Agni, Vayu,
(or prii7Ja) and Aditya the three principal manifestations of Agni40 and in
other places there are statements about Agni which exactly parallel those
made about the iitman in the above-quoted Upani}adic text, namely that Agni
is the sun (iiditya) in the cosmos and breath (prii7Ja) in the individual.
In this Upani~adic passage the sun (iiditya) is the formless (amilrta) and real
or true (satya) aspect (rilpa) of Brahman, In B.A. U 2.3,1-6 we have a similar
account. Here the formless (amilrta) and real or true (sat) aspects (rilpa) of
Brahman are prii7Ja. Taking all this information together it is clear that
although it is not explicit in the text the author is drawing on a range of
established associations and equivalances where the unitary Agni, the
unitary prii7J.a and the iitman can all be equated. This Agni/prii7Ja/ iitman is the
real, unformed Brahman, the supreme principle which manifests as a
variety of cosmic and individualized phenomena. In: the final analysis,
however, all these manifestations are unreal (asat or asatya).
The Vitalistic Antecedents oj the Atman-Brahman Concept

Kena Up.
35
Brahman is described as the ear of the ear, the mind of the mind, the
speech of the speech, the eye of the eye and the pralJa of the pral]l (1.2).
Brahman is not expressed through speech but is that by which speech is
expressed; is not thought by the mind but is that by which the mind thinks;
is not seen by the eye but is that by which the eye sees; is not heard by the
ear but is that by which the ear hears; is not breathed by the breath (pralJa)
but is that by which the breath breathes. A later verse of the same UPani~ad
(3.1) tells us that Brahman once conquered for the devas and they gloried in
his conquest (a reference to B.A. U I.3 and C. U I.2. 1-14 where pralJa defeats
the asuras?). It would appear, therefore, that here again we have an example
of the pralJa concept being employed as a 'model' for the description of
Brahman..
Conclusion
Overall, I think the foregoing analyses demonstrate that, in some Upani~adic
circles at least, the concepts of atman and Brahman were developed on the
basis of already existing conceptions ofpralJa. I am not claiming, however,
that Upani!adic accounts of atman and Brahman are always to be understood
in this way. Some Upani~ads, such as the ifa and the MalJrfilkya, do not
mention pralJa at all and it could hardly be argued that their presentation of
atman and Brahman are derived from descriptions of pralJa. The same
applies to the Svetasvatara Upani~ad which mentionspriilJa only once (2. g).
Despite these qualifications it is obvious, in the light of the foregoing that
the characterization of pralJa as found in the writi~gs of Badarayal).a,
Sarikara, Ramanuja and other classical Vedantins constitutes a far from
accurate interpretation of the Upani~ads.
Just why these influential Vedanta teachers were so rigorous in their
denial of any equation between atman/Brahman and pralJa is a complex
issue in itself but one explanation which links with later developments is
that the pralJa concept is not philosophically user-friendly. It has already
been noted how the meaning of the term pralJa changes according to the
level of description being employed, and this makes it difficult for
philosophers to use it in a precise way. It should be mentioned that the
same was true of iitman in the pre- Upani~adic literature, where it had a range
of meanings from trunk, through body to self, breath and spirit. By the
time of the Upani~ads though, it had lost its more physical meanings.
PralJa is also a dynamic reality, constantly moving, constantly changing.
For the later Vedantins such an entity could not be truly real. For them,
only that which did not change could be real. Linked with this is the fact
that the Vedanta tradition came increasingly under the influence of the
Sarpkhya philosophy. Although the author of the Brahmasiitra and
theologians such as Sarikara frequently criticize the Sarpkhya school for
36
being unorthodox, a general pattern that can be discerned is that the later
the Vedantin text the more Sarpkhya concepts are assimilated.
This process can be traced back at least as far as the Kafha UpaniJad, and
there is plenty of evidence from the great epic, the Mahabharata, that Vedic
versions of Sarpkhya existed alongside non- Vedic ones. 'The former were
promulgated by brahmins; subscribed to the view that pUTUJa was single and"
unitary; incorporated the Vedic gods into their systems; and reckoned that a
householder could gain release as well as a renouncer. The latter criticized
the brahmins for conducting animal sacrifices and thus breaking the code
of ahirrzsa; subscribed to the view that there were many PUTUJas; made no reference to gods in their descriptions of the world; and emphasized the
importance of renunciation.
The Sarpkhya cosmology is very orderly. One tattva (principle) emerges
from another in a pre-determined order; the faculties (indriya) are neatly
divided into two sets of five: the faculties of perception (buddhzndriyas) on the
one hand, and the faculties of action (karmendriyas) on the other; the pUTUJa
of Sarpkhya is immutable, not subject to change.
Sarpkhya thus appeals to an orderly philosophical mind and its highest
principle, puruJa, is immutable. In fact, later Vedantins tend to take
UpaniJadic references to pra7Jas as references to the indriyas of Sarpkhya.
However, scrutiny of UpaniJadic passages referring to pra7Ja reveals that the
lists of pra7Jas in the UpaniJads never include the karmendriyas of Sarpkhya.
This shows just how far the later Vedantins were prepared to go in the
direction of misrepresenting UpaniJadic teachings in the service of their
own ideas.
A study of references to pra7Ja in the Brahmasiltra and in the writings of
Sankara and Ramam0a, the two most influential Vedanta teachers, reveals
much the same thing. As mentioned earlier, the primary task of the author
or authors of the Brahmasiltra was to deal with problematic passages from
the UpaniJads, that is, passages that were difficult to interpret within the
framework of emerging Vedanta philosophy.
Most of these problematic passages dealing with pra7Ja are, in fact, those
presenting pra7Ja as the ultimate principle or where pra7Ja seems to be the
source of mentation or cognition. In all cases, the aim of BadarayaI,la,
Sankara and Ramanuja is the same: to claim that the UpaniJadic teachers
did not mean what they said when they described pra7Ja as the source of
everything or identified it with atman or brahman. Consequently, the
literature of the Vedanta school (except the UpaniJads themselves) has little
to say on the concept of pra7Ja. "When it is mentioned it is usually to
comment that pra7Ja is a purely phenomenal principle which has nothing to
do with sentience or cognition.
With these considerations in mind, a rather bold and provocative conclusion seems appropriate. It is this. On the subject of pra7Ja the great
Vedanta commentators wilfully misrepresented the teachings of the
37
Upani!ads. By doing so they rendered a great disservice to those who turn

to them for guidance when seeking to understand the message of those
ancient scriptures.
Notes
I.
2.
g.
4.
5.
6.
7.
8.
9.
lO.
I!.
12.
Ig.
,+
'S.
16.
'7.
18.
Ig.
20.
21.
22.
2g.
24
25.
26.
27.
28.
29.
30.
g1.
32.
33
34.
35.
g6.
Deut<ch, 1969, p. 9.
See, for example, their commentaries on Bralzmasiltra 1.1.2g(24); 1.1.28-gI (29-g2); I.g.g9
(40-4) and 1.4.16-18. Numbers in parentheses refer to Ramanuja's arrangement
Ewing, IgOg, p. g.
Whitney, Ig05, pp. 6g2-g.
Keith, I969 [1925], p. 205.
op. cit. p. 20r.
op. cit. p. 206.
See loco cit.
Loc. cit.
Loc. cit.
Loc. citin.
op. cit. p. 208:
op. cit. p. 2Ig.
op. cit. p. 220.
Ewing, IgOg, p. 7
Loc. cit.
Eggeling, 1882, (I) p. 121.
Idem, I8g5, (4) p .. '44
Idem, '900, (5) p. g cf 5.2-4-IO and II.1.2.g.
7.5.1.22.
1.3.2.g, 8.3.4.5, IO.2.6.I5 and IO.3.1.5.
8.2.2.8.
8.1.4.1,8.7.2.14,8.7.3.6.
63l.I:3 and '5
7.1.1.24.
Keith, 1969 [1925]. p. 220.
Radhakrishnan, '953, pp. 468- 83.
op. cit. p. 487.
Radhakrishnan, '953, p. 791. This passage makes the existence of the worlds (Ioka) dependent on the gods (deva) which, in turn, are dependent on the prat;aJ (pralJLl, apana, etc.) which,
in turn, are dependent on priilJLl/prcifiiiitman. Such an arrangement reverses the common
Upaniiadic pattern of emphasizing cosmology over psychology by making the cosmos (the
adlzidaivata realm) dependent on the individual (the adhyiitma realm).
op. cit. p. 163.
op. ciL p. 166 (slightly amended).
My translation.
Radhakrishnan, '953, p. 246.
op. cit. p. 25g
op. cit. p. S'S.
'1Ve may note here that whilst speech, breath, etc. aU come from column two, skin and
generative organ come iom column one. Furthermore, the order of digestive faculty and
generative organ are reversed in the 'seizing' list. Exactly why skin replaces hairs here is not
clear. Radhakrishnan suggests that 'touch' is what is implied. The rea.<;on why touch, which
would have been the most logical entry in column two, was initially displaced by hairs was
probably that the author (or redactor) wanted to get 'oiadhivanaspatya' into column three and
this would seem to be a peculiar derivation irom touch. The reason for the reversal of J'iSna
and apiina is more obvious. The writer wanted to end the list with the only successful faculty.
The reason for the original order would seem to be the simple physical progression irom
higher to lower: heart, navel, and phallus.
38
37.
38.
39.
40.

PU7U:'ia or iitman, not; as many commentators suggest, Vamadeva.
Radhakrishnan, 1953, p. 524.
Op. cit. pp. 816, and 821-2.
e.g. 6.3.3.16; 6.7-4-4; 7-1.1.22-23; 8-5-2.8; 9.1.123; 9.2.1.21 and IO.6.2.I-II.
Ab breviations
A.A.
A.o.
B.A.U
co.
Kafh. 0.
KauJ.o.
Mait. 0.
MU7J{i. U
po.
S.B.
TO.
Aitarl!)la Ara7!Jaka
Aitarl!)la UpaniJad
Brhad-Ara7!Jaka UpaniJad
Chiindogya UpaniJad
Kafha UpaniJad
Kaujitaki UpaniJad
Maitri UpaniJad
MU7J{iaka UpaniJad
Prafna UpaniJad
Satapatha Briihmana
Taittirfya UpaniJad
Bibliography
Beidler, W The V1Sion of Self in Early Vedanta. Motilal Banarsidass, 1975.
Deussen, P Sixty UpaniJads of the Veda (2 vols) (trans. Bedekar V M.
and Palsule G B.). Motilal Banarsidass, 1980.
Deutsch, E. Aduaita Vediinta: a Philosophical Reconstruction. University of
Hawaii Press, 1969.
Eggeling,J. The Satapatha Briihmana (5 vols). Motilal Banarsidass,
(reprint of 1882 [IJ, 1885 [2J, 1894 [3J, 1895 [4J and 1900 [sJ
eds), 1963.
Ewing, A. H. The Hindu Conception of The Functions ofBreath (part 2).
Allahabad: Liddell's Printing Works, 1903.
Hurne, R. E. The Thirteen Principal UpaniJads (2nd ed.). Oxford
Keith, A. B. The Aitarl!)la Ara7!Jaka. Oxford University Press (reprint of
the 1906 edition), 1969.
Radhakrishnan, S. The Principal Upanishads. George Allen and
Unwin, 1953.
Whitney, W D. The Atharua Veda SaTflhita (2 vols). Motilal Banarsida;s
(reprint ofthe 1905 edition), 1971.
The Oral Composition and Transmission of

Early Buddhist Texts
Mark Allan, Stjohn's College, Cambridge
It is generally agreed that early Buddhist literature, of which the Pali texts
of the Theravada canon are the most numerous and best preserved
examples, was composed and transmitted orally. 1 This is considered to be
the case for the following reasons:
There is no reference to writing or writing materials in the principal
Pali nikiiyas, 2 though there are many references to learning and
reciting discourses (see below).3
2. Although there are a few passages in the Pali Vinayapitaka which
indicate that the art of writing was known at the time when these
Vinqya texts were put into their present form, these do not refer to
texts and their preservation. 4
3. Despite detailed rules governing the use of all items used by monks
and nuns, the Vinqya has no rules governing the use of writing
materials. 5
4. There is no archaeological evidence for the use of writing in India
during the early phase of Buddhism, that is, before the time of ASoka6
- although this view may have to be revised in the light of recent
finds in Sri Lanka of BrahrnJ. characters on potsherds dating from
this period. 7
5. Finally, many of the stylistic features of these texts indicate an oral

8
ongm.
I.
As just. noted, there are many passages in Pali canonical texts depicting
monks and nuns learning and reciting the Buddha's teachings and discourses, which seem to indicate that during the Buddha's lifetime material
was formulated so that it could be remembered and recited. In a passage
occurring in the Vinaya and Udiina,' for example, it is reported that the
Buddha asked the monk SOI.la to expound the Dhamma. In response SOl).a
40
Indian
Ins~ghts:
Buddhzsm, Brahmanism and Bhakti
recited the Atthakavagga (sabbiin' eva attlzakavaggikiini sarena abhiisz), the name
now given to a group of verse suttas in the Suttanipiita. 10 In the Vinaya Tnention is made of monks who are expert in the suttas chanting a sutta (suttantikehi suttanta1'[! sa1'[!gayantehz). 11 In the Smigftisutta of the Dfghanikaya a distinction is made between the Buddha teaching the Dhamma, a monk teaching
the Dhamma to others as he has heard and learnt it, and a monk reciting
the Dhamma as he has heard and learnt it (sajjlziiya1'[! karoti'J.12 In the
Suttavibhmiga of the Vinaya there is a particularly interesting piicittiya rule
which prohibits those who have not taken the higher ordination from being
taught the Dhamma by being made to recite it word by word (or "line by
line", padaso dhamma1'[! viicf:Yya).13 The formulation of this rule arose because
certain monks were teaching some laymen in this manner. The old commentary takes this as a particular form of recitation, almost in the manner
of Vedic chanting. The commentator Buddhaghosa (Sp 741) interprets this
passage as referring to a particular manner of reciting verse. Although it is
somewhat obscure, it certainly seems to imply that students were made to
learn fixed texts by heart. Again, there are many references to reciting the
Piijimokklza. On one occasion, for example, the Buddha refused to recite the
Piijimokklza because the assembly of monks was not pure. 14 And finally, there
are many passages which refer to monks being learned, having heard
much, grasping and remembering the Dhamma, and so on. 15
Despite these references, we do not know what material was actually
composed during this period, nor the form and manner in which it was
composed. Nor do we understand the relationship of this material to the
original discourses, or the relationship of these initial compositions to
Buddhist texts as we have them today 16
All schools of Buddhism agree that soon after the death of the Buddha a
council, or sangi:ti, was held to confirm and rehearse the Buddha's
teaching. 17 Some schools also maintain that sangftis were held at other times
in the history of the Buddhist community. Also, the Theravada tradition,
for its part, considers that after the first sangfti a tradition of specialisation
arose whereby groups of monks, called bhiir;akas, began to specialise in the
knowledge and recitation of particular collections of texts. 1B
What material was rehearsed at the first smigi:ti (and at those which
followed) and whether, in fact, fixed texts were "recited" on these occasions
is likewise uncertain. vVe do not yet understand the way in which the
bhiir;aka system worked, nor its impact on the material being transmitted. 19
The Theravada tradition maintains that its texts were first written down
in the 1st century BeE in Sri Lanka, while information about the use of
writing in the other Buddhist schools is generally lacking. 20 The Theravada
account, occurring first in the DijJava1'[!sa, is extremely brief, consisting of
two verses only.21 vVe do not know whether writing was utilised as an aid to
composition or transmission before this time; but it has been suggested that
there is some evidence for a manuscript tradition in the case of certain
The Oral Composition and 'Hansmission ofEarly Buddhist Texts
41
texts before this date. 22 Again, we do not yet fully understand what impact
writing, or the writing down of the canon,had on the material and its
transmission. 23
The period of oral composition and transmission can probably be measured in centuries. R. Gombrich has suggested "three to four centuries".24
But as S. Collins has argued, the Buddhist tradition also remained in
various ways an oral/aural one, despite the introduction of writing; that is,
the monks and nuns recited and listened to oral as well as written textS.25
With early Buddhist texts being composed and transmitted orally it is
not surprising that they exhibit so many striking features which appear
alien to the modern reader and which, as stated earlier, are generally taken
to be indicative of the oral status of this material. Stylistic features alone do
not prove that a given text was originally oral, for written texts can, for
various reasons, deliberately mimic the style of texts belonging to an earlier,
oral phase of the tradition. Besides, the impact of the new medium on the
style of the texts being composed would not have been immediate. 26 But as
there are other reasons for taking this to be an oral literature, we can
regard the stylistic features of these texts as being, at least in part, a product
of their oral origins.
For some decades now a field of study has developed in the West which
has attempted to understand the way in which oral literature is composed
and identifY its peculiar characteristics. Particularly important to the foundation of this field were Milman Parry's studies of Homeric epic verse. 27
Parry argued that many of the stylistic features of these texts indicated that
this literature had its origins in an oral tradition, and he developed the
theory that in an oral epic tradition the poet creates his poems as he
performs with the aid of what he referred to as formulas and themes, which
are the building blocks of the performance. 28 In consequence, every performance of the poem was a new creation, although each version may have
been very similar. In an attempt to confirm these ideas Parry and A. B.
Lord conducted field work in what was then Yugoslavia where a living tradition of oral epic verse survived. Mter Parry's death, Lord continued the~e
studies and further developed this theory. Most importantly, he emphasised
the improvisatory nature of oral performance, regarding "oral" to be "formulaically improvised". 29 He therefore argued against the conception of
fixed, memorised texts in oral traditions, stating, for example, that "sacred
texts which must be preserved word for word, if there be such, could not be
oral in any except the most literal sense". 30 In response to this, some have
considered that the rote learning of a lengthy text and its verbatim repetition is the product of a culture which knows writing for, they argue, it is
only through a fixed, written text that we can have the notion of word-forword fixity.31 But Lord's tendency to see his model as universally valid and
his attributing of particularly restricted meanings to certain terms have
been criticised by a number of scholars.32 The Parry-Lord model may
42
describe what occurs in the Homeric or Yugoslav traditions, or even be

appropriate to oral epic traditions in general, but oral tradition;; are
diverse, and what holds true for one may not be appropriate for another.
l\IIany factors can influence the character of an oral literature and its
method of composition and transmission: the nature of the information
being relayed; the attitude towards this material and the extent to which
accuracy is required; the character of the performers or composers, their
status in society, the type of training they have undergone and the
circumstances under which they perform; the nature of the audience and
its expectations and therefore its demands on the performer or performers;
the medium used (verse or prose) and whether the performance requires
musical accompaniment.
The Buddhist and Yugoslav-Homeric traditions differ in virtually all of
these factors. In epic verse traditions the medium is verse, and for the most
part, epics portray the lives and activities of heroes. They are often
performed to musical accompaniment, and in the Yugoslav case at least,
they are primarily performed for entertainment. Also, performances are
very much public events and the audience has an important influence on
what is performed, or at least, on what episodes are performed and the
degree to which each is elaborated. The status of the performers is also
particular: they are bards or poets who perform individually, not communally, and they have usually acquired their performance skills through a
long period of training. In contrast, in the early Buddhist tradition prose is
by far the most dominant medium. The function of the literature is to
preserve the teaching of a religious leader and the rules deemed necessary
to guide the conduct of the members of that religious community. The
information being transmitted is often complex, consisting of descriptions
of practices and detailed analyses of concepts and psychological processes,
all of which require a high degree of accuracy.33 The "performers" were
monks and nuns, and increasingly they were members of monastic institutions. They came from diverse social backgrounds: some were brahmans
who presumably had undergone their traditional training, others would
have had no formal training in literary/performance skills. And finally,
material was performed communally, as well as individually and privately.
This last factor seems to be one of the most overlooked. Yet it is
particularly important, for communal or group recitation or performance
requires fixed wording. It is not possible for more than one individual to
perform at the same time in the manner described by Parry and Lord
without producing utter chaos, for in that method each individual creates
his compositions anew each time he performs.
The first application of the ideas of Parry and Lord to early Buddhist
literature was L. S. Cousins' article "Pali oral literature" (1983). Cousins
argued that in the earliest phase in the production of Buddhist literature the
monks performed accounts of the Buddha's discourses and presented his
The Oral Composition and Transmission ofEarly Buddhist Texts
43
teaching in the manner proposed by the Parry-Lord model, that is, with "a
strong improvisatory element" (p. 9). With time this material then came to
be fixed due to its religious authority (p. 6). The differences between
accounts of the same event or teaching found in different collections within
the Pali canon and between the parallel material belonging to different
schools are evidence for an initially improvisatory stage, for such variations,
he states, "are too fi"equent to arise from the natural variation of a manuscript tradition or even from a rigidly memorised oral tradition" (pp. 5-6).34
R. Gombrich, in a paper entitled "How Mahayana began" (199ob)/5
argued against the improvisatory stage proposed by Cousins, seeing early
Buddhist texts as "deliberate compositions which were then committed to
memory, and later systematically transmitted to pupils" (p. 24), because, he
states, "the whole purpose of the enterprise ... was to preserve the
Buddha's words" (p: 22). Further:
The early Buddhists wished to preserve the words of their great teacher, texts
very clifferent in character from the general run of oral literature, for they
presented logical and sometimes complex arguments. The precise wording
mattered" (p. 21).36
An investigation of the stylistic features of early Buddhist texts can, I

think, make an important contribution to this debate. 37 AB part of my Ph.D.
research a number of the most prominent stylistic features of the prose
portions of Pali canonical sutta texts, and more specifically, of the prose of
the IYlghanikiiya, the first book of the Suttapifaka, were investigated. 38 The
following discussion is based on this research. 39
One of the most dominant characteristics of the prose portions of Pali
canonical sutta texts is the use of standardised phrases or passages to express
or depict a given concept, action or event. These standardised phrases have
been variously called "formulas", "cliches", "stock expressions", "stock
phrases" and "stereotyped phrases". For the most part I will. use the term
"formula" .40
The narrative portions of Pali sutta texts contain numerous passages
which depict someone approaching another person, and the phrases used
to depict these approaches are formulaic. The material encountered is
extensive and diverse, and I will therefore restrict myself here to a
discussion of those approach-formulas which are based on the unit
yena .. . ten' upasarrzkami, upasarrzkamitvii, "x approachedy, having approached
(he did such and such),'.
The material in the Dfghanikiiya can be divided into two groups. The
formulas of the first group depict someone approaching the Buddha, a
monk or another person, and a monk approaching the Buddha or another
monk. The simplest formula within this group depicts the approach of the
visitor, then his or her interaction, usually verbal, with the person
approached. In the more complicated formulas the visitor approaches,
44
shows some form of respect; adopts a particular posture (standing or

sitting), then speaks with the person approached. A particular range of
fixed units of meaning is employed within each division of this overall
structure to construct distinct formulas. Which units of meaning are
employed, and hence which formula type and specific formula is used,
depends on the narrator, the classification of the person approaching and
the person approached, their attitude towards each other, and the purpose
of the visit.
So, for example, when a brahman is depicted approaching a king, the
following combination of units will be used:
(Then)41 + the brahman approached the king. Having approached
he said this to the king. 42
In contrast, the formula used to depict a brahman approaching the

Buddha in order to question him will be:
(Then) + the brahman approached the Bhagavat. Having approached, +
he exchanged greetings with the Bhagavat, and having exchanged agreeable and courteous talk (with him), + he sat down to one side. + Seated to
one side, the brahman + said this to the Bhagavat. 43
This is characterised by respect being shown and the showing of this
particular form of respect, by the brahman sitting down rather than
standing, and by the brahman speaking first. There are also certain forms
of address associated with this 'interaction.
Again, the following combination of units will be used to depict a monk
approaching the Buddha when he has been summoned by him:
(Then) + the monk approached the Bhagavat. Having approached, + having
paid homage to the Bhagavat, + he sat down to one side. + To the monk
who was seated to one side + the Bhagavat said this. 44
In contrast to the previous formula, this is characterised by the monk
showing this particular form of respect and by the Buddha speaking first.
Again, there are certain forms of address encountered in such an approach.
The second group of formulas depict the Buddha approaching someone
and a monk approaching someone other than the Buddha or another
monk. Here the situation is quite different. Not only do the structures of
these formulas differ from those of the previous group, but these formulas
also utilise a completely different range of fixed units of meaning. Although
the factors which determine the use of particular fixed units of meaning,
and hence the overall formula, are the same, their relative importance
differs markedly from the first group. Here the purpose of the approach is
the fundamental determinant of the formula used, with the wording of
approaches of different purposes differing greatly from each other. We have
in effect "purpose built" formulas.
The Oral Composition and Transmission
if Earfy Buddhist Texts
45
So, for example, the formula used to depict the Buddha visiting an
ascetic is:
Then the Bhagavat approached the ascetic. Then the ascetic said this to the
Bhagavat: 'May the Bhagavat come, venerable sir. Welcome to the Bhagavat,
venerable sir. It is long, venerable sir, since the Bhagavat took the opportunity
to come here. NIay the Bhagavat, venerable sir, be seated. This seat has been
prepared.' The Bhagavat sat down on the prepared seat. Having taken a
lower seat, the ascetic sat down to one side. The Bhagavat said this to the
ascetic who was seated to one side. 45
But the formula used to depict the Buddha attending a donor's meal is:
Then the Bhagavat, dressing in the morning and taking his bowl and robe,
approached the house of the brahman together with the community of
monks. Having approached, he sat down on a prepared seat. Then the
brahman personally satisfied and served the community of monks headed by
the Buddha with the finest hard and soft food. Then, when the Bhagavat
had finished his meal and had washed his bowl and hands, the brahman
took a lower seat and sat down to one side. The Bhagavat said this to the
brahman who was seated to one side.'6
The formulas of this category are generally characterised by the
following features: (r) the Buddha or monk is depicted getting dressed and
taking his bowl and robe when the visit is a public one; (2) the approach is
to the place of the person who is approached rather than to the actual
person; (3) the Buddha or monk does not show respect to the person
visited; rather, some gesture of respect or subordination is shown by the
person approached; (4) the Buddha or monk sits down on a prepared seat;
and finally, (5) such approaches usually occur in sutta-narrator passages.
Features 2, 3 and 4 tend to subordinate the person being approached to the
Buddha or monk who is approaching.
The study of the material in the D'ighanikiiya shows that the wording of
passages which depict the common event of someone approaching another
person has been standardised in this text to the extent that only a limited
range of stock phrases or formulas is exhibitedY These formulas have set
structures and are composed of a variety of possible fixed units of meaning.
As mentioned, which units are employed, and hence which formula type
and specific formula is used, depends on certain factors. Given a knowledge of these determining factors, the wording of a particular approach is,
in the majority of cases, predictable. 48 At minimum, this indicates that there
is an overall homogeneity to the narrative portions of this collection of
suttas. 49 Whether there was a tendency to use a standardised diction from
the beginning, or whether standardisation was undertaken at the great
smigltis, or councils, or later by the bhiilJilka tradition or when the canon was
written down, is yet to be determined.
46
There are two principal, alternativ'e methods for the composition of oral
literature and therefore for early Buddhist texts. The first is that proposed
by Parry and Lord for oral epic verse and taken by Lord as the only
method possible in oral cultures. In such an improvisatory method no two
performances are exactly alike. If the early phase of Buddhist literature was
one of composition-in-performance, then those texts which we consider to
be representative of this period must be seen to be 'frozen' versions of a
particular performance. 50 The second method entails the composition of a
fixed text which is then memorised and transmitted verbatim. 51
The standardised diction outlined here can be seen as an aid to
composition within both of these methods: whenever a particular approach
needed to be portrayed, the wording was already available. In other words,
these formulas acted as prefabricated building-blocks. 52 In addition to this,
the use of a standardised and predictable diction would also have aided the
learning by heart and recitation of a large body of fixed material; that is,
within a tradition of the composition and transmission of fixed texts this
feature would have a mnemonic function. 53
This research on the formulaic diction of these texts becomes
particularly interesting when the wording of passages which depict similar
concepts, actions or events found in different Suttapitaka and Vinayapitaka
texts are compared. For example, the event of l\IIara approaching the
Buddha towards the end of the Buddha's life is found in the Dfghanik5ya,
Udana, SaTJ1yuttanikaya, and Anguttaranikiiya. 54 In the Dfghanikiiya and Udana
occurrences we have the fullest formula with Mara approaching the
Buddha, standing to one side, then speaking:
Then, not long after the venerable Ananda had departed, Mara the evil one
approached the Bhagavat. Having approached, he stood to one side.
Standing to one side, Mara the evil one said this to the Bhagavat. 55
In the SaTJ1yuttanikaya we have a briefer formula with no mention of Mara
standing to one side:
Then, not long after the venerable Ananda had departed, Mara the
evil one approached the Bhagavat. Having approached, he said this. 56
But in the Anguttaranikaya the passage is so brief as not even to mention
the approach, merely reading:
Then, not long after A.nanda had departed, Mara the evil one said
this to the Bhagavat. 57
Here, as in other instances, it is seen that the Dfghanikiiya and Udana are
the most wordy texts. This means that, although the suttas of the Dfghanikiiya
are longer than those to the A7iguttaranikiiya primarily because of differences
in their structure, the use of a more elaborate and detailed diction by the
former is certainly a contributing factor.
The Oral Composition and Transmission ofEary Buddhist Texts
47
We saw earlier that the formula used in the Dzghanikiiya to depict the
Buddha or a monk approaching an ascetic is characterised by the ascetic
showing respect to the Buddha or monk and not vice versa, by the ascetic
addressing the Buddha or monk in a reverential manner, and by the ascetic
taking a lower seat; all of which tends to subordinate the ascetic to the
Buddha or mon1e Further research shows that this formula is particular to
the Dzgha- and Majjhima-nikiiyas. In contrast, the Sarrzyuttanikiiya,
Anguttaranikiiya and Vinayapztaka use a simpler formula which depicts the
Buddha or monk greeting the ascetic, sitting down to one side, then
speaking to him. 58 When depicting such encounters with ascetics, the
authors of the Sarrzyutta, Anguttara and Vinaya seemed to have considered it
unnecessary to portray the Buddha or monk being honoured in such an
exaggerated manner.
Again, differences exist between the various canonical texts in their
wording of the "going to an invited meal" approach-formula mentioned
earlier. In contrast to the previous example, the same basic formula is used
in each text, but the syntax of the fixed units of meaning and presence of
particular units differs from text to text. For example, in the Dfghanikiiya
and Udiina we have the Buddha approaching the donor's house "together
with the community of monks" and sitting down, while in the
Majjhimanikiiya, Anguttaranikiiya, Suttanipiita and Vinayapitaka we have the
Buddha approaching the donor's house, then sitting down "together with
the community of monks"; that is, these latter texts associate the unit
"together with the community of monks" with the verb depicting the
action of sitting down rather than with the verb portraying the approach. 59
Also, in the Dfghanikiiya, Majjhimanikiiya, Udiina andSuttanipiita we have the
donor "taking a lower seat and sitting down to one side" after the meal,
while the Sarrzyuttanikiiya, Anguttaranikiiya and Vinayapitaka fail to include this
"taking of a lower seat" phrase. 6o
The situation seems to be quite complex, with a text such as the Udiina,
for example, following the diction of the Dzghanikiiya with regard to some
formulas, but not others. It is possible that such differences may have resulted from the bhii7'}aka tradition, or period of specialisation. Alternatively,
differences in diction may have resulted from the way in which each text
was used by the Buddhist community. In other words, it is possible that
different texts were intended for different audiences and had different functions and that their wording was modified accordingly. Or again, in some
cases these differences may be due to the different manuscript traditions of
these texts. Further research certainly needs to be undertaken to properly
identify and understand such differences.
Another common feature of the prose portions of Pali canonical sutta
texts is the tendency to proliferate similar word elements and units of
meaning to form sequences or "strings". We frequently encounter sequences
of two, three or more adjectives or adjectival units qualifYing the same noun,
48
a number of nouns ill acting as the 'subject of the same sentence or as the
object of the 'same verb. We encounter sequences of adverbs modifyiI}g the
same verb, or a number of parallel verbs occurring together in the same
sentence, and so on. Wherever such sequences of parallel word elements or
units of meaning occur, they are arranged according to what will be referred
to here as the Waxing Syllable Principfe; that is, in sequences which consist
of similar word elements or units of meaning of an unequal number of
syllables, the words or units of meaning of fewer syllables must precede (to
use an expanded form ofPa.J;rini's phraseology, via Caland).61 For example,
in the UdumbarikaSihanadasutta we find an ascetic telling the Buddha that he
had challenged him "as he was foolish, confused, and unskilled": yathabalena yatha-mu/hena yatha-akusalena. 62 This consists of a string of three
adverbial expressions. The first has 5 syllables, the second 5 syllables and
the third 7 syllables; that is, the pattern is 5+5+7. This arranging of
elements according to an increasing syllable length tends to produce a
crescendo effect in these sections of the text, and to a certain extent
parallels enumeration, another important stylistic feature of this literature.
The exception to this general principle is where a sequence, and
especially a long sequence, can or must be divided into groups on the basis
of associations in meaning or grammatical or morphological form, in
which case the Waxing Syllable Principle only works within each group,
restarting again with the next group. For example, the stock description of
the lowly-talk engaged in by ascetics,63 which also occurs in the
Udumbarikasfhanadasutta, consists of a long list of topics of conversation. This
list can be divided into groups on the basis of associations in meaning: riijakatha'f[L cora-katha'f[L mahamatta-katha'f[L, "talk of kings, thieves and ministers".
This has a 4+4+6 syllable pattern. This group is then followed by senakatha'f[L bhaya-katha'f[Lyuddha-katha'f[L, "talk of armies, fear and battle" (4+4+4
syllables); anna-katha'f[L pana-katha'f[L vattha-katha'f[L sayana-katha'f[L, "talk offood,
drink, clothing and bedding" (4+4+4+5 syllables); mala-katha'f[L gandhakatha'f[L fiati-katha'f[L yana-katha'f[L, "talk of garlands, scents, relatives and
vehicles" (4+4+4+4 syllables); gama-katha'f[L nigama-katha'f[L nagara-katha'f[L
janapada-katha'f[L, "talk of villages, towns, cities and districts" (4+5+5+6).
And so on. The pattern of this list so far is thus 4+4+6, 4+4+4, 4+4+4+5,
4+4+4+4,4+5+5+6.
This ordering principle is not only apparent in the more obvious
sequences of adjectives and adjectival units, nouns and noun phrases,
adverbs and verbs, but it also seems to be operational in the ordering of
parallel units of meaning which occur in different, but nonetheless closely
associated, c1auses, sentences and paragraphs, and in the ordering of
sequences of parallel sentences or semi-independent units of meaning, as
well as a number of other structures, such as those involving saddhi'f[L. A
number of examples encountered in the material studied have problematic
patterns. Solutions to these can often be found if certain amendments are
The Oral Composition and Transmission qfEarfy Buddhist Texts
49
accepted - a svarabhakti vowel not scanned or a word thought to bea later

insertion omitted - or if only the iIILrnediately parallel units are compared.
It also seems that conceptual considerations or the desire to produce a
particular word play may occasionally override the Waxing Syllable
Principle. A few examples await plausible solutions.
It is particularly common in these sequences for the comporient
elements, and especially the initial members of the sequence, to share
sound and metrical similarities. 54 In the above example ofyatha-balenayathamu?hena yatha-akusalena, it is seen that, apart from the obvious sound
similarities due to yat/zii- being the first member of each compound, the
endings of all three are virtually identical: -lena/ -fhena/ -lena. Again, the first
two compounds differ only in their core syllables: -ba- and -mu-, which are
both labial consonants in conjunction with long vowels. The two initial
compounds of this sequence therefore share the same metrical patterns and
are virtually identical in sound. The -u- of the second member (-mufhena) is
also echoed in -akusalena of the third. Hence, there is a tendency in these
texts to proliferate similar word elements and units of meaning, that is to
expand the wording, while at the same time there is a tendency to bring
this expanded wording closer together by choosing words which share
sound and metrical similarities.
This phenomenon of ordering similar word elements according to their
syllable lengths has been known for some time, but an analysis of the
nature and extent of its application within Pali texts has riot been
undertaken before. 55
The proliferation of similar word elements and units of meaning and the
ordering of the member elements of such sequences according to the
Waxing Syllable Principle, which thus prodll ces an overall crescendo effect,
tends to give a rhythm and homogeneity to this material. This rhythm and
homogeneity is then greatly enhanced whert, as is frequently the case, the
member elements also share sound and metrical similarities. The presence
of rhythmical patterns in prose, and especially in long prose texts, must
have been extremely important to those who performed or recited this
material, and may be functionally parallel to the rhythm produced in verse
by metre.
As with the use of formulas, the structures briefly discussed here would,
by acting as an organisational principle, function as aids to composition
within both of the compositional methods outlined above. However, within
a tradition of the composition of fixed texts, which are designed to be
memorised, this combination of stylistic features would also have functioned
as a mnemonic aid, for it is surely easier to remember a sequence of words
arranged in this manner according to syllable length. Similarly, it is easier to
remember two different words when they share sound similarities and have
the same metrical patterns. And again, the presence of some form of
rhythm would also have facilitated the memorisation and recitation of this
50
material. But whichever method we consider to be that utilised for the

composition of these texts, it is seen that the choice of words and, their
arrangement was heavily influenced by the fact that these texts were
composed and transmitted orally.
The final stylistic characteristic that will be discussed here is repetition. By
repetition I mean the repetition of sentences, passages or whole sections of
the text, and the repetition of set structures. This discussion is based on the
study of repetition in one sutta of the lJi.ghanikiiya, the Udumbarikaszhaniidasutta,
the 25th sutta of this collection.
In order to establish the degree to which this particular sutta is repetitive,
the text of the PTS edition was scanned into a word processor and all
abbreviated passages were reconstructed. It was then possible to establish
the word count for the complete sutta and for those sections which were
being repeated, and hence to calculate what percentage of the text was
repetitive. 66
In order to quantify repetition, the level at which the repetition is
occurring within the text and the type of repetition involved must be
established. Repetition can occur at a number of levels. A passage is
repetitive at a primary level when it does not form part of a passage which
is itself repeated within the text. If it does, then it is repetition at a
secondary level. Sometimes repetition at a tertiary level is discernible.
In this study five categories or types of quantifiable repetition were
established. They are: Verbatim Repetition (VR), Repetition with Minor
Modifications (RMM) , Repetition with Important Modifications (RIM),
Repetition of Structure Type-r (RS-r) and Repetition of Structure Type-2
(RS-2).
In Verbatim Repetition a passage is repeated word for word with no
modifications needing to be made by the one who recites or performs this
material. For example, the stock description of the lowly-talk engaged in by
ascetics mentioned earlier is repeated verbatim four times in this sutta,
representing about 5 % (4.5 %) of the text. Or a long passage which
describes three stages of what the Buddha considers to be true ascetic
practice, and which represents about 6 % of the text, is repeated verbatim
three times, making up about n% of the sutta. In total 30% of the
Udumbarikaszhaniidasutta involves Verbatim Repetition at a primary level.
Passages which are repeated with alteration to only a small proportion
of their wording were classified as Repetition with Minor Modifications.
For example, it is not uncommon to form the opposite of a passage
expressing a positive or negative state by merely repeating that passage and
adding or omitting certain prefixes or particles. 67 Just under 35 % (34.5 %) of
this sutta involves this kind of repetition on a primary level.
The third category, Repetition with Important Modifications, involves
repetition of a passage, but with important changes to the wording,
whether in syntax, grammatical number, tense or person, or enlargement
5I
or contraction. of the wording. Material of this category represents about

4 % (3. 8 %) of the sutta studied.
Two types of repetition of structure were also established. In the first
type a structure is repeated along with virtually all of its wording, but with
key elements replaced to produce differences in meaning. So, for example,
in the common passage which depicts the practice of the four brahmauiharas,
or divine abidings, the same structure and wording is repeated four times,
first for loving kindness (mettii) , then for compassion (karur;ii) , then
sympathetic joy (muditii) , and finally, for equanimity (upekhii).6B In this way
each repetition differs by only one word.
In many passages of this class, the elements which differ in each
repetition share morphological, sound or metrical similarities, or
similarities in structure (or some combination of these), thereby minimising
the impact of the changes being made. For example, in this sutta, after
defeating the ascetic Nigrodha in debate, the Buddha criticises him for not
having the following thought:
Enlightened (buddha) the Bhagavat teaches the Dhamma for enlightenment
(badhiiya); tamed, or controlled (danta), the Bhagavat teaches the Dhamma for
control (damathiiya); calmed (santo), the Bhagavat teaches the Dhamma for
calm (samathiiya); crossed over (tirJ.1JD), the Bhagavat teaches the Dhamma for
crossing over (tara7Jiiya); extinguished (parinibbuto), the Bhagavat teaches the
Dhamma for extinguishing (parinibbiiniiya).69
The initial element of each parallel sentence (buddha, danta, santo, tirp}D,
parinibbuta) is a past participle. The first four have the same number of
syllables and equal metrical patterns, and sound similarities are evident at
least in danto and santo. In the second group of elements which differ in
each repetition, damathaJa and samathiiya are morphologically parallel, share
the same number of syllables and have the same metrical pattern. They
differ, in fact, only in their initial letter. Similarly, the last two elements in
this group, tarar;iiya and parinibbiiniiya, are morphologically similar. In this
way, the elements which differ in meaning within each repeated structure
appear similar in outward form. The effort involved in making the required
modifications is thereby minimised for the reciter.
A total of 16% of this sutta is composed of material of this RS- I category.
In those passages which were classified as Repetition of Structure
Type-2, a basic structure is repeated, but with far less repetition of the
wording, or in some cases, with modification to the structure of the
wording which is replaced. Material of this category represents nearly 3 %
(2.5%) of this sutta.
In total almost 87% (86.8%ya of the Udumbarikasi:haniidasutta involves
quantifiable repetition of one kind or another at a primary leveL This is
surely a significantly high proportion of the text. It must also be noted that
the verbatim end of the scale is particularly well represented.
52
Repetition is undoubtedly a mnemonic device. This is based on the

simple observation that the more frequently a passage, unit of mea~ng or
word is repeated the more likely it is to be remembered. Or as a verse in
the Dhammapada states: "non-recitation is the rust of incantation". 71 The
repetition encountered in Buddhist texts has frequently been taken to have
a mnemonic function, but few have elaborated on their statements or
investigated repetition in any systematic manner. 72
I have so far argued that the first rwo stylistic features discussed in this
paper could have functioned as aids to composition both within a tradition
of composing material during the performance in an improvisatory
manner and in a tradition of composing fixed texts which were to be
transmitted verbatim. In addition to this, it was proposed that these
features would also have had a mnemonic function within the latter
tradition. In contrast to this, it is difficult to see the gross forms of repetition
just discussed - the repetition of whole passages, with or without
modification, and the repetition of structures with the replacement of
various proportions of their wording - and the scale. on which this is
pursued, that is the proportion of the text involved, as anything other than
proof, or at least as a very strong indication, that these texts 'were designed
to be memorised and transmitted verbatim. 73 In contrast, material such as
the contemporary Yugoslav epics studied by Lord (1960) or the
contemporary Indian epics studied by Smith (1991), Beck (1982) and
Roghair (1982), which is composed "during the performance", although
exhibiting many forms of repetition, does not exhibit the form of gross
repetition encountered in Pali sutta texts.
The five categories of repetition established in the study upon which the
above discussion of repetition in the Udumbarikasfhaniidasutta is based can be
graded according to the degree to which they each facilitate the learning
and retention of this material. Verbatim Repetition obviously represents
the greatest aid to memory. The greater the percentage of a text that is
verbatim repetitive the easier it is to learn and remember. At a primary
level, 30% of the Udumbarikasfhaniidasutta consists of repeated passages of
this classification.
The remaining four types of repetition each encompass a range of
differences. In terms of the modifications to be made by the reciter, and
hence the effort involved in making such changes, the Repetition with
Minor Modifications and Repetition of Structure Type-1 categories on the
one hand, and the Repetition with Important Modifications and Repetition
of Structure Type-2 categories on the other, are seen to be parallel and to
encompass a similar range of differences.
As mentioned, almost 35 % of the sutta studied involves Repetition with
Minor Modifications at a primary level. Another 16% involves Repetition of
Structure Type-I. Together these two categories, which are similar in terms
of their mnemonic significance, represent about 51 % (50.5 %) of this sutta.
The Oral Composition and Transmission of Early Buddhist Texts
53
The study also showed that almost 4 % of this sutta involves Repetition
with Important Modifications at a primary level and that approximately
3% involves Repetition of Structure Type-2. Together these two parallel
categories represent about 6% (6.3%) of this sutta.
As 87% of the Udumbarikasfhaniidasutta involves some form of quantifiable
repetition on a primary level, I3 % of this text is therefore only encountered
once. Much ofthis consists oftheopening and closing sections of the sutta.
Although not occurring again, the passages and elements which make up
this I3% commonly involve non-quantifiable forms of repetition (as do
those which are repeated again) and may be found in other suttas of the
Dfghanikaya.
Many of the passages which are repetitive at a primary level in the text
are themselves composed of or incorporate quantifiable repetitive elements,
which is repetition at a secondary level. For example, a passage which is
repeated verbatim may itselfbe composed of a passage repeated verbatim
twice. This secondary passage therefore occurs four times in the text. This
secondary repetition would further increase the familiarity of the material
being learnt and facilitate recitation. 74
This study has focused on one sutta in the Dfghanikaya. But much of the
material found in both the repetitive and non-repetitive passages of this sutta
is also encountered elsewhere in the Dfghanikiiya, which of course is significant
if a body of suttas such as are contained in the present Dfghanikiiya was learnt
and transmitted by a particular group of monks or nuns. This repetition
decreases the uniqueness of the material which is not repeated again within
this sutta, and increases the familiarity of those passages which are.
Further, various forms of non-quantifiable repetition are an integral part
of all passages, whether these passages are repeated again within this
particular sutta or not, whether they are found in other suttas or are unique
to this sutta, whether classified as being repetitive at a primary or secondary
level. Passages are built up through the proliferation of similar word
elements, units of meaning and structures. Many elements share sound and
metrical similarities. Vocatives of address and particles such as atha kho and
kho are continually used as markers throughout the sutta. Certain verbs are
repeated in their non-finite forms to resume the following clause. 75 The
wording used to express or depict a given concept, action or event is
standardised, and diversity of vocabulary is avoided. And so on. In this
way, although we have been able to quantify gross repetition of certain
classifications, there are many forms of repetition employed by this class of
Pali text which cannot be quantified, yet which must also be considered to
facilitate greatly the learning and recitation of this material. Repetition thus
thoroughly permeates every dimension of this class of Buddhist literature.
The characteristics of the prose portions of Pali canonical sutta texts
discussed in this paper show that the authors of this material attempted to
minimise differences and maximise similarities. They did this by using a
54
Indian Insights: Buddhism) Brahmanism and Bhakti
standardised diction (which we have referred to as formulas), by

proliferating similar word elements often chosen for their soul)d and
metrical similarities, and by pursuing repetition on a truly large scale, to
mention but a few. Of these stylistic features, it is gross repetition which
provides the greatest evidence that these texts were composed as fixed texts
which were to be memorised and transmitted verbatim. As previously
mentioned, these stylistic features do not prove that this literature was
essentially an oral one, for written texts can utilise or mimic characteristics
of an earlier oral tradition. Nor do they prove that these texts were
conceived as fixed texts. But when combined with such historical factors as
accounts of communal recitation, events which required a fixed text, then
we are surely on firmer ground.
.
Although I have attempted to show that the early Buddhist sutta texts
were, in the words of R. Gombrich, "deliberate compositions which were
then committed to memory", 76 I would certainly agree that accounts of
what the Buddha is supposed to have said and discourses on his teaching
would have been given by the monks and nuns after the Buddha's death in
an improvisatory manner, at times drawing heavily on memorised material,
or as R. Gethin (1992) has argued, by using lists as a foundation. Such
discourses may then have become the basis oflater fixed texts. But these
accounts and discourses were fundamentally different from the essentially
fixed, memorised texts transmitted by the community, however imperfectly.
Finally, the Parry-Lord model does not exhaust the oral or literary / performance dimension of oral cultures. In ancient, pre-literate India there was a
strong tradition of composing fixed, religious texts which were designed to
be memorised and transmitted verbatim.
Notes
1.
2.
3.
4.
5.
6.
7.
8.
9.
ro.
O. von HimJ.ber, 1990, chap. V (esp. p. 22), p. 30, chap. XIV; K. R. Norman, 1993a, p. 280;
R. Gombrich, 1990a & 199ob; L.S. Cousins, 1983, esp. p. I; S. Collins, 1992.
R. Gomhrich, 1990h, p. 27; O. von Hinuber, 1990, esp. p. 30.
S; Collins, 1992, esp. pp. 124-25; R. Gombrich, 1990b, p. 26.
T. W. Rhys Davids & H. Oldenberg, 1881, pp. XXXII-XXXV; cf: R. Gombrich, 199ob,
PP27-8.
T. W. Rhys Davids & H. Oldenberg, 1831, pp. xX;'Cu-XXX!!!; R. Gombrich, 199ob, p. 28.
R. Gombrich, 1990b, p. 27. For the most recent views on writing in India, see O. von
Hinuber, '990, esp. pp. 54, 72 and K. R. Norman, 1993b, esp. pp. 243, 245-47.
See R. A. E:. Coningham, 1993. S. U. Deraniyagala dates these finds to 600-500 Be; R.
Allchin/R. A. E. Coningham tentatively date them to 400-450 BG.
O. von Hinuber, '990, pp. 22-3; R. Gombrich, 1990a, pp. 7-8; 1990b, pp. 21-2; L. S.
Cousins, 1983; G. von Simson, 1965; 1977, p. 479.
References to Pali texts are to the Pali Text Society's editions. Abbreviations of titles of
works and of terms and signs follow the Epilegomena to Vol. I of the Critical Pali
Dictionary.
Vin I 196/' Ud 59 reads so/a.m aitJzakavaggikani sabban' eva sarena abha1}i; c[ S. Collins, 1992,
P 125

II.
12.
13.
14.
15.
55
Vin I 169.
DIll 241f: cf: S. Collins, 1992, pp. 126-7'
Vin IV 14-5; cf: PED s.v. pada (padaso).
Vin rr 236-41 = Ud 51-6.
E.g. balzu.,,uto iwti sutadlzaro mtasannicayo (M I 356); so ca blzikklzu bahussuto hoti agatiigamo
dlzammadharo vinayadharo matikiidlzaro pa~ito vyatto medlziim lajiz kukkuccako sikklziikamo (Vin I 337).
C[ S. Collins, 1992 (for a brief description of tbe teacbing and learning process, see esp.
~~
16.
17.
18.
C[ O. von HinUber, 1990, pp. 26, 28.

O. von HinUber, 1990, chap. VI; K. R. Norman, 1989, p. 29.
E. W Adikaram (1946, chap 3) remains the standard reference for the blza~ka tradition. C[
K. R. Norman, 1989, pp. 32-4; R. Gombrich, 199oa, p. 7; 199ob, pp. 25-6; S. Collins, 1992,
pp. 124-5; L. S. Cousins, 1989, PP-4-5'
19. For examples of differences which may be due to the bhii~ka tradition, see below. See also G.
von Simson, 1977, p. 486; O. von HinUber, 1990, chap. X; K. R. Norman, 1989, pp. 34,50.
20. A. K. Warder (1980, p. 294) mentions Bu-ston's account which states that all schools
committed their texts to writing in the 1st century CE or earlier. Warder (pp. 345-6) also
refers to the Sarvastivadin account which states that they wrote theirs down c. 100 CEo
pitakattayapaliii ca tillsa att/zakatlzaT[! pi ca
2 I.
muklzapat/zena aneJ'UT[! pubbe bhikkhu mahaman
Iziinirrz diwiina .I'attiinarrz tadij blzikklzu samagatii
ciratllzitattlzarrz dizammllll'a potthakesu liklziipayurrz (DIp XX 20-1 = Mhv XXXIII 100-101).
Most scholars seem to uphold the validity of' this account. Cf H. Bechert, 1991, pp. 9-10;
1992, esp. p. 52; K. R. Norman, 1983, p. 5; 1989, p. 36-8; R. Gombrich, 1992, p. 160; 1. S.
Cousins, 1991, p. 55; S. Collins, 1990, p. 96ff. 1992, p. 128; T W Rhy.1 Davids & H.
Olden berg, 1881, pp. XXXV-=XVI.
22. .J. Brough, 1962, pp. 28-9, 218-19' The Pali version ofthe verse to which Brough refers has va
sayati, where the other versions support an original sa vayati.Brough (ibid., p. 218) dismisses
the possibility that such a transposition of syllables could have occurred in the course of
purely oral transmission as, he states, "such a supposition would indeed imply an
unbelievably slipshod parampara. But in manuscript copying this is a common and readily
understandable error." However, such a transposition seems equally possible in oral
transmission. C[ S. Collins, 1990, fnt. 25.
23. See K. R. Norman, 1999b.
24- R. Gombrich, 1990b, p. 21; cf: 1992, p. 160. T. W. Rhys Davids & H. Oldenberg, (1881, p.
XX.'<VI) suggest "about three hundred years". 1. B. Horner (1992, p. 186), speaking of the
Vinaya rules, states that the "rules were formulated orally and transmitted orally for probably
more than
25.
26.
27.
28.
29.
200
years"
S. Collins, 1992, esp. p. 121; cf: O. von HinUber, 1990, p. 9, chaps. XIV & Xv.
G. Bonazzoli (1983, esp. p. 267) and H. Bakker (1989, pp. 330-32) both argue that, in the case
of the purar;a.l', stylistic features do not necessarily indicate oral composition. They both see
the puriilJ1J.l' as resulting li'om an interplay between oral and written transmLlSion.
For examples of the critici.lm of'the use of stylistic features or formulaic density as an
indicator of oral or written origins which are encountered in the wider field of oral literature
research, see for example, A. B. Lord, 1975, pp. 12-20; 1986, p. 4781; 1987;]. M. Foley, 1985,
pp. 26" 42, 50,56.
For a brief overview of Parry's work, see H. Lloyd-Jones (1992) and the introductory essay to
]. M. Foley (1985).
Parry defined the formula as "a group of words which is regularly employed under the same
metrical conditions to express a given essential idea", a definition which has remained
prominent in the discussion of oral literature. This has obvious limitations for our purposes
for we are dealing with prose and the metrical dimension or requirement is therefore
inappropriate. A number of scholars have argued for the exclusion of the metrical
component ii'om the definition of a formula. See, for example, P. Kiparsky, 1976, pp. 84, 87;
M. O'Connel; 1980, pp. I04-106; G. H. Rog'hair, 1982, pp. 60-6; O. M. Davidson, 1988; cf:J
D. Smith, 1987, esp. pp. 596" 602.
In the introduction to his important and influential publication The Singer q/Tales (1960, p. 4),
Lord set out his definitions: "stated briefly, oral epic song is narrative poetry composed in a
manner evolved over many generations by singers ortales who did not know how to write; it
56
30 .
3!.
32 .
33
34
35
36.
37
38.
39
40 .
43
44
45

consists of the building of metrical lines and half lines by means of formulas and lormulaic
expressions and of the building of songs by the use of themes, This is the technical sense in
which I shall use the word 'oral' and 'oral epic' in this book."
A. B, Lord, 1960, p, 280,
I. lvL L. Hunter, 1985, But lengthy verbatim recall could, in lact, be verilied by a group of
specialist reciters, As will be shown in the.10llowing studies, there are also many stylistic
features which can aid the oral tran.<mi,<sion of a fixed text.
See, lor example, J. D. Smith, 1977. Lord's response to Smith's criticism is lound in A. B.
Lord, 1987, p. 6Sff.
Althoug'h not thought to be a memorised text, the contemporary oral Indian epic of Pabaji
exhibits a high degree of stability, which.J. D. Smith (1987, pp. 600-602; 1989, p. 33)
attributes to the religious status of the hero. Thus a religio-philosophical dimension may
result in an even greater degree of stability, or even fixity.
There is, in fact J much scope for such changes to occur within a "rigidly memorised
tradition". This is discussed in greater detail in the conclusion to my Ph.D. thesis (see note
below).
This article first appeared in 1988 in the Journal if Piili and Budd/zi"t StudieJ.
The only other work which has attempted to address the question.ofthe method used by the
early Buddhists lor the composition of their texts is R. Gethin's"The MiitikaJ: memorization,
mindfulness and the list" (z992). Gethin, like Cousins, considers the delivery of discourses and
accounts of what the Buddha said to have been an improvisatory affair, and the differences
between parallel versions or the same text as resulting Irom such a method. The various
versions of the DaJuttaraJUtta (I Da.iottarasutra), for example, differ in their inclusion or omission
of certain lists becau,se they represent different per/ormances or what is essentially the same
list-giving discourse (ibid., pp. 157-8).
Other publication.< which have discussed the orality 01' the early Buddhist tradition are: O.
von Hinilber, Der Beginn der Sclzrijl undftilhe Schrjfilichkeit in Indien, 1990 (especially Chap. V "Die
Milndlichkeit del' altesten buddhistischen Texte" and Chap. XIV "Die milndliche
Textweitergabe bei den Buddhisten") and Untersuchungen:cur iWilndliclzkeit.friilzer mitteiindi"cfzer
Texte der Budd/listen, 1994; S. Collins, "Notes on some oral aspects ofPali literature", 1992; and
R. Gombrich, "Recovering the Buddha's message", 199oa.
The most important studies of the style of Buddhist or Brahmanical literature are H.
Oldenberg, Zur Gesclzic!zte deT aitindischen Prosa, 1917, esp. pp. 39-52; G. von Simson, Zur
Diktion einiger Lehrtexte des buddlzistisc!zen Sanskriikanon.J, 1965; J. Gonda, Stylistic repetitions in tlze
Veda, '959. Of less importance are S. Weeratunge, "Some significant stylistic traits of
Buddhist Sanskrit prose", '992 and P. Kwella, "Some remarks on the style of some Buddhist
Sanskrit texts", 1978.
The thesis title is Some stylirtic fiatures of the prose portions if Fiili canonical sutta textr and tlzeir
mnemonicfonction. It was submitted to the University o['Cambridge in September 1994.
These studies were restricted to an analysis of prose, because, as stated, this is by far the
most dominant medium used by the early Buddhists. It is also particularly interesting as
most oral literature is verse. I chose to work with Pali canonical J'Uttatexts, not because they
are the oldest, but because they represent the most complete and best preserved body of
texts representative of the early phase of Buddhist literature.
This diversity of terminology in part reHects a general uncertainty as to what actually constitutes -a "formula", "stock phrase", and so on, and whether, say, a "[armula" is different
from a "stock phrase" or "stock expression". A detailed discussion of the terminology and
definitions is presented in my Ph.D. thesis.
The inclusion of the unit "then", atlza kho, depends on the context.
E.g. D II 237 atha kilO blzo Mahagovindo briihmal}1l yena Re1JU riliii ten' upasaTJlkami, upasaTJlkamitvii
Rer;uTJl riijiinaTJl etad avoca.
E.g. D I 236 atha kilO Vase!i/laBlziiradviijii ma1}1Jva yena Bizagavii ttn' upllsaTJlkamir!,su, up".JaTJlkamitvii
Blwgavatii saddhiTJl sammodiTJlsu, sammodanijaTJl kathaTJl siiriiTJiyaTJl vftiriiretvii ekamantaTJl nirfdiTJlsu.
ekamantaTJl nil'inno klzo Viise!!11O mii1}1Jvo BlzagavantaTJl etad avoca.
E.g. D II 144 iiyasmii Anando ... yena Bhagavii ten' upa.raTJlkami, upa.JaTJlkamitvii BlzagavantaTJl
ablziviidetvii ekamantaTJl nilTdi. ekamantaTJl nirinnaTJl!d1O iiyasmantaTJl AnllndaTJl Bizagavii etad avoca.
E.g. D I 178-9 atlza !d1O Bizagavii yena sam'!Jappaviidako tindukiicfro eka.riilako Maiiikiiya iiriimo ten'
upa.raTJlkami. . ,. atl," kilO Blzagavii yena Po!tizapiido paribbiijako ten' upa.JaTJlkami. atlza !dw Pot.tlzapiido
The Oral Composition and Transmission qfEarly Buddhist Texts
46.
4748.
paribbiijako BhagavantaTfl etad avoca: 'etu kilO Mante Blwgava, sagataTfl Mant, Bhagavato, cira.(laTfl kilO
bhante Bhagava imaTfl panjiiyam akasi yadidaTfl idh' iigamaniiya, nisfdatu Mante Bhagava, idaTfl amnaTfl
pannattan' ti. nirfdi Bizagava pannatte asane. Po!!hapado pi kllO paribbiijako annataraTfl nfcaTfl awnaTfl
galzetva ekamantaTfl nisfdi. ekamantaTfl ni.sinnaTfl kho Po!.tJzapiidaTfl paribbiijakaTfl Bhagavii etad avoca.
Note that the ascetic addresses the Buddha as 'bhante'.
E.g. D I 226-7 atha kho Bhagavii pubba~hasamayaTfl nivasetva pattacfvaraTfl adaya saddhiTfl
bhikklzusaTflghena yena Siilavatikii ten' upasa7Jlkami. ... atlza kilO Bizagava yena Lolzicca.rra briilzma1JClJJ'a
niv,sana7Jl ten' upa.raTflkami, upasa7Jlkamitva pannatte asane nisfdi. atlza kho Lohicco Eralzma~
Buddhapamuklza7Jl blzikJdzu.raTflglzaTfl pa~ftena klziidaniyena blzqjaniyena sahattlzii santappesi sampavareri.
atJza !rho Lohicco brahma~ BhagavantaTfl bhuttiim7Jl onftapattapa~iTfl anfiatara7Jl nlCaTfl iir'ana7Jl galzetva
ekamanta7Jl nilldi. ekamanta7Jl nisinnaTfl kllO Lohicca7Jl brahma~m Bhagava etad avoca.
The situation is generally the same in other canonical texts, though the formulas used may
differ (see below).
The examples not mentioned in this paper of complicated and particularly detailed approaches, and
49.
50.
51.
52.
53.
5+
56.
57.
58.
59.
60.
6r.
62.
63.
or tho.')e which do not quite conforrn to the norm, show that these structures were not
blindly imposed upon the material. The authors of this material were tully capable of breaking with the norm where necessary. Meaning was still the ultimate determinant of diction.
Cf. B.]. Manne, '992, p. ,63,
See L. S. Cousins, '983 and R. Gethin, '992.
See R. Gombrich, '990b.
Cf]. Gonda, '959, pp. 4'-3
Cf G. von Simson, '965, pp. '42-3; '9n pp. 479-80.
But as mnemonic means C(aidingmemor/', it therefore not only includes '(aiding the
remembering of what is memorised", but also "aiding the remembering of non-memorised
elements", [ar example, the course of events or the appropriate, or approximate, wording. We
thereiare find [armula, and other elements of composition in an improvisatory setting being
referred to as "mnemonic aids" (B. A. Rosenberg, '987, pp. 82-3) and "mnemonic elements"
D. Smith, '989, p. 40). It is used in this paper in the sense of "aiding the learning and recall
of a memorised text" (A. B. Lord, '987, p. 67, makes the distinction between remembering
and memorising; cCJ. D. Smith, '989, pp. 36-7).
The stylistic [eatures discu.ssed in this paper may have had other functions hesides aiding
composition. However, space does not permit a discussion of'these here.
D II 104; Ud 63; S V 260; A IV 310.
atlra kho Maro po.pimo. acirapakkante iiyasmante Anand, yena Bhagavo. ten' upll.la7Jlkami, uposaTflkamitvo.
ekamanlaTfl attlriiri. ,kamantaTfl tizito klro j\1iiro papima Blzagavanla7Jl etad avoca (D II 104).
atlra kllO Maro papimo. acirapakkante iiyll.lmante Anande yena Blragavo. ten' upll.laTflkami, upa.raTflkamitvo.
etad avoca (S V 260).
atlra IdlO Maro papima acirapakkante iiyll.J'mante Anande BlzagavantaTfl etad avoca (A N 3IO).
E.g. S II 32-3 atlra kllO iiyll.lmo. Sariputtoyena afifiatittJziyana7Jl paribbiijakana7Jl aramo ten' upll.laTflkami,
upll.la7Jlkamitvo. tehi afinatittlziyehi paribbiijakehi saddhi7Jl sammodi, sammodanfya7Jl katJza7Jl siiro.~fyaTfl
vltil'aretvo. ekamantaTfl nirtdi. ekamantaTfl nilinna7Jl kllO iiyll.lmantaTfl SariputtaTfl te afinatittJziyii paribbiijaka
etad avocu7Jl.
E.g. Ud 82 & 89 read atha kho Bhagavo. pubba~zll.lamayaTfl niviir'etvo. pattaclvaraTfl adiiya saddlzim
bhikkhu.wwgll.ena yena [house of' host] ten' upll.la7Jlkami, upll.la7Jlkamitvii pafifiatt, iir'an, nirldi, while M
II '46, Sn p. ",, A IV ,87 and Yin I 2'7-8 read atJza kilO Blragava pubba~za.ramaya7Jl nivayetvo.
patlacfvara7Jl adiiyayena [house of host] ten' upa.ra'f!lkami, upasa7Jlkamitva pannatte amne nilidi saddlzim
blrikklzusamglzena.
E.g. D II 97 reads atlza kho Ambapo.li ga~ika BuddhapamukhaTfl blzikklzusa'!'glw7Jl pa~tena klziidaniyena
bllOjaniyena sahattho. santappesi sampavaresi. atha kilO Ambapo.l! ga~ika BhagavantaTfl blzuttavi7Jl
Onttapattapa~i7Jl afifiatara7Jl nfca7Jl iir'anaTfl galzetva ekamantaTfl nilidi, while its parallel at Yin I 233
reads atha kho Ambapo.ll ga~iko. Buddlrapamuklw7Jl blzikkIzUJ'a7JlglwTfl pa~tena klziidaniyena bizqjaniyena
sa/zattJza santapp,tva sampavaretva Bhagavanta7Jl bhuttii.vi7Jl onttapattapalJ-i7Jl ekamanta7Jl nisMi.
W. Caland, 1931, p. 59; cf.]. Gonda, '959, p. 61.
D III 54 saccaTfl Manle blziir'ita me esa vo.ca yatM-balena yatha-rniipzena yatlzii.-akusaiena ti.
D III 36-7 seyyathfdaTfl riijakatha7Jl corakatlwTfl malziirnattakatha7Jl senakatha7Jl bhayakatlzaTfl
yuddhakatha7Jl annakatlza7Jl panakathaTfl vattlzakatha7Jl sayanakatlw7Jl maliikatlza7Jl gandhakatlza7Jl
fio.tikatlzaTfl yanakatlra7Jl gamakatlraTfl nigamakatlwTfl nagarakathaTfl janapadakathaTfl ittlzikatlwTfl
[purisakatlra7JlJ siirakatlw7Jl visikhiikatlw7Jl kumbattlzii.nakatha7Jl pubbap,takathalJl nanattakatha7Jl
lokakklziiyikaTfl [katha7JlJ samuddakkhiiyikaTfl [katha7JlJ itibizaviibhavakatha7Jl iti va.
a.
55.
57
58
64,
65
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.

Most other occurre'nces of this formula omit purisakatlza'f[l. All occurrences read 10kakJr.lzayika'f[l
samuddakklzayika'f[l itiblzaviiblzavakatlza,!,. C[ D 17-8, 66, '78-9; III 54; M I 5'3; II I, 23, 30; S V
4'9ff; A V 128-9; Yin I 188; IV 164.
The identification of specific classes of sound similarities, such as alliteration, assonance,
homoioteleuton, etc., are not important here, for, as noted byJ Gonda (1959, pp. 376--7),
such distinctions were probably not made. What mattered was the repetition and similarity
of sounds.
See W. Caland, 1931;J Gonda, 1959, e.g. pp. 60-4, 125!: G. von Simson, 1965, e.g. 2.3, 2.7,
8-4-6; H. Smith, Epilegomena to the Critical Pali Dictionary, p. 35* wax. comp.; O. von
Hintiber, 1990, chap, VII; I993, pp, I04-II3. O. von Hintiber (1994) is the most detailed
study so far published. Unfortunately, it arrived on my desk alter my own studies were
completed and the current paper wa' delivered.
'
The complete text is 34 % larger than the abbreviated PTS version.
For example, the pa<sage puna ca para'f[l Nzgrodlza tapani tapa,!, samiidiYati, bllOjanesu vodiisa'f[l iipqjjati
"ida'!' me klzamati, ida,!, me na-kklzamali" Ii. so ya'f[llzi kilO 'JJa na-kklzamali ta'f[l sapekilO pajaizali,y(1)1 pan'
assa klzamati ta'f[l gatlzito muccllito qjJMpanno aniidinavadassavi anllraralJllpaiiiio pariblzuf[jali (D III 43) is
later repeated as puna ca para,!, Nzgrodlza tapilili tapa'f[l samiidiYati, bllOjanesu na vadasa,!, iipqjjati "ida,!,
me klzamati, ida'f[l me na-kklzamali" ti. so ya1)1lzi kilO \ra na-kklzamati ta'f[l anapeklzo pqjaizati,ya1)1 pan' aIra
klzamati ta'!' agathita amucelzita anqjJlziipanno iidfnavadasriivf nissaralJllpaiiiio paribizunJali (D III 46).
E.g', D III 49-50.
'tao'a te Ni,grodlza miiiiUJJa sato mahallakas]'a na etad ailOsi: "buddha so Bizagavii bodhaya d/zamma,!, deseti,
dan to so Bizagavii damathaya dhamma,!, deseli, santo Ja Bizagavii samathaya dhamma1)1 deseti, ti'~", so
Bhagava tara~aya dlzamma1)1 deseti, parinibbuto so Bhagavii parinibbiinaya dlzamma1)1 desetf" , ti
(D III 54-5)'
The complete JUtta has a word count of 5,871. The word count for the passages which have
been classed as VR is ',761. The word count, lor the other four categories are: RMM 2,028;
RIM 222; RS-, 937; RS-2 149. The five categories have a total word count of 5,097, which
repre'ents 86.8% of the text.
Dhp 24' asafJlzayamalii manto. (Narada Thera's translation).
G. von Sim,on (1965, esp. p. 142ff.) is the exception. cr T .w. Rhys Davids, 1881, pp,
xxn-=m; R. Gombrich, 1990b, p. 24; W. B. Bollee, '970, p, 172; L. S. Cousins, 1983, p, 9; G.
von Simson, 1977, p. 480; A. Syrkin, '983, p. 160;J Gonda, 1959, pp. 78, 351.
C!: T. W. Rhy' Davids, ,88" p. xxn; R. Gombrich, 1990a, p. 7; '990b, p. 24.
For example, of the passages which are VR at a primary level, 78.9% consists oC quantifiable
repetition of one kind or another. In total 53.4% of the material which is repetitive on a
primary level involves quantifiable repetition on a secondary level.
In the above discus,ion of approach-formulas, for example, upasa'f[lkami, upara'f(lkamilva (". ,. he
approached. Having approached") and ekamanta1)1 nisfdi, ekamanta1)1 ni.rinno (" ... he sat down to
one side. Seated to one side") were encountered.
76.
R. Gombrich, 199ob, p. 24.
Bibliography
Adikaram, E. W Early history ofBuddhism in Ceylon. Colombo: 1946.
Bakker, H. "Some methodological considerations with respect to the
critical edition ofPuranic literature", XXIII Deutscher
Orientalistentag 16-20 Sept. 1985 in Wurzburg, (ZDlvIG, supplement
7), (ed.) Eo von Schuler. Stuttgart: 1989, pp. 329-41.
Bechert, H. "Methodological considerations concerning the language of the earliest Buddhist tradition", BSR, vol. 8, no. 1-2,
199 1, pp. 3-1 9.
- - - . "The writing down of the Tripitaka in Pari", WZKS, 36,
199 2 , pp. 45-53
. The Oral Composition and Transmission qfEarfy Buddhist Texts
59
Beck, B. E. F. The Three Twins: The telling qfa South Indianfolk epic.
Bloomington: 1982 .
. Boilee, W B. (ed. & tr.) Kur;.o.lqjo.taka. London: 1970.
Bonazzoli, G: "Composition, transmission and recitation of the
.
Puro.r;.a.(, Puro.r;.a, 25, no. 2, 1983, pp. 254-80 ..
Brough,]. The Co.ndho.rfDharrnapada. London: 1962.
Caland, W ''A rhythmic law in language", Acta Orientalia, IX, 1931,
pp. 59-68.
Collins, S. "On the very idea of the Pali Canon", ]PTS, Xv, 1990,
pp.89-I26.
- - - . "Notes on some oral aspects ofPali literature", 11],35, no.
2/3, 1992, pp. 121-35
Coninghain, R. A. E. ''Anuradhapura Citadel Archaeological
Project: Preliminary results of the excavation of the south
rampart 1992" South Asian Studies, 9,1993, pp. III-22.
Cousins, L. S. "Pali Oral Literature", Buddhist Studies Ancient and
Modern, (eds.) P. Denwood and A. Piatigorsky. London: 1983,
pp. I-II.
- - - . "The 'Five Points' and the origins of the Buddhist schools",
The Buddhist Forum, vol. II, (ed.) T. Skorupski. London: 1991,
PP27-60 .
Davidson, O. M. ''A formulaic analysis of samples taken from the
Shilhnama of Firdowsi" , Oral Traditions, 3ir-2, 1988,
pp.88-I05
Foley,]. M. Oralformulaic theory and research. An introduction and annotated
bibliography. New York: 1985.
Gethin, R. "The Mo.tikas: memorization, mindfulness and the list",
In the Mirror qfMemory. Riflections on mincffUlne9.r and remembrance in
Indian and Tibetan Buddhism, (ed.)]. Gyatso. Albany: 1992,
PPI49/ 2 .
Gombrich, R. "Recovering the Buddha's message", The Buddhist
Forum, vol. I, (ed.) T. Skorupski. London: I990a, pp. 5-20.
- - - . "How Mahayana began", The Buddhist Forum, vol. I, (ed.) T.
Skorupski. London:I990b, pp. 21-30.
- - - . "The Buddha's Book of Genesis", 11], 35, no. 2/3, 1992,
PPI59/ 8 .
Gonda,]. Srylistic repetitions in the veda. Amsterdam: 1959.
Hintiber, O. von Der Beginn der Schrifl undftilhe Schrifllichkeit in Indien,
(Akademie der Wissenschaften und der Literatur, Mainz.
Abhandlungen der Geistes- und Sozialwissenschaftlichen
Klasse,Jahrgang 1989, Nr. II). Stuttgart: 1990.
- - - . "From colloquial to standard language. The oral phase in
the development of Pali", Premier Colloque Etienne Lamotte
(Bruxelles et Liege 24-27 septembre 1989). Louvain-la-Neuve:
1993, pp. IOI-II3
60

Hinuber, 0. von Untersuchungen zur lvIundlichkeitjruher mittelindischer
Texte der Buddhisten, (A~ademie der Wissenschaften und d.er
Literatur, Mainz. Abhandlungen der Geistes- und
Sozialwissenschaftlichen Klasse,Jahrgang 1994, Nr. 5).
Stuttgart: I994.
Horner, 1. B. "The monks: Buddhist and Christian. Gotama's rules
compared with the rules of St Benedict", Buddhist Essays: a
Miscellany (Volume in Honour of the Ven. H. Saddhatissa).
London: 1992, pp. 185-97.
Hunter, 1. M. L "Lengthy verbatim recall: The role of the text",
Progress in the psychology oflanguage, vol. 1, (ed.) A. W Ellis,
London: 1985, pp. 207-235.
Kiparsky, P "Oral poetry: some linguistic and typological
considerations", Oral literature and the formula, (eds.) B. A. Stolz
and R. S. Shannon, Ann Arbor: 1976, pp. 73-ro6.
Kwella, P "Some remarks on the style of some Buddhist Sanskrit
texts", IT, Vl, 1978, pp. 169--;5.
Lloyd-Jones, H. "Becoming Homer", New York Review ofBooks,
XXXIX, no. 5, March 5,1992, pp. 52--;.
Lord, A. B. The Singer of Tales. Cambridge, Mass.:1960.
- - - . "Perspectives on recent work on oral literature" , Oral
literature; seven essays, (ed.)JJ Duggan, Edinburgh: 1975,
pp. 1-24 (reprint from Forum for Modern Language Studies, voL X,
no. 3, I974)
- - - . "Perspectives on recent work on the oral traditional
formula", Oral Tradition, 1/3, I986, pp. 467-503.
- - - . "Characteristics of orality", Oral Tradition, 2ir, I987,
PP54--;2.
Manne, B. J Debates and case histories in the Fiili Canon, (unpublished
Ph.D. thesis) Rijksuniversiteit te Utrecht: 1992.
Narada Thera (tr.) Dhammapada, New Delhi: I972.
Norman, K. R. Fiili Literature, including the canonical literature in Prakrit
and Sanskrit if all the Hfnayiina schools ofBuddhism, Wies baden:
I9 83
- - - . "The Pali language and scriptures", The Buddhist Heritage
(Buddhica Britannica, Series Continua I), (ed.) T Skorupski. Tring:
I989, pp. 29-53 (= Collected Papers, voL IV, I993, pp. 92-I23)'
- - - . Review of O. von Hinuber, Der Beginn der Schrift und jruhe
Schriftlichkeit in Indien, inJRAS, (3rd series) voL 3, part 2, 1993a,
PP277-8 1.
- - - . "The development of writing in India and its effect upon
the Pali Canon", WZKS, 36, [I992J, I993b, PP.239-49.
O'Conner, M. Hebrew verse structure, Winona Lake:I980 (referred to
by R. G Culley "Oral tradition and biblical studies", Oral
Tradition, IiI, 1986, pp. 46-7).
The Oral Composition and Transmission ojEarly Buddhist Texts
61
Oldenberg, H. Zur Geschichte der altindischen Prosa) Mit besonderer

Beriicksichtigung der prosaisch-poetischen Erziihlung, (Abhandlungen
der Keniglichen Gesellschaft der Wissenschaften zu
Getting-en. Philologisch-Historische Klasse. Neue Folge Band
XVI Nr. 6). Berliri: 1917.
.
Rhys Davids, TW & Oldenberg, H. (trs.) Vintrya texts, vol. I, (SBE
XIII) Oxford: 1881.
Roghair, G. H. The Epicoj Palnii{iu. A study and translation ojPalnii{iu
Virula Kath a Telugu oral tradition from Andra Pradesh, India. Oxford:
1982.
Rosenberg, B. A. "The complexity of oral tradition", Oral Tradition,
211, 1987, pp. 73-90.
.
."
. Simson, G. von Zur Diktion einiger Lehrtexte des buddhistischen
Sanskritkanons (ph.D thesis). Munchen: 1965:
- - - . "Zur Phrase yena ... tenopajagiima / upet:Ya und ihren
Varianten im buddhistischen Sanskrit'~, Beitriige zur
Indieriforschung (Festschrift Waldschmidt), (ed.) H. Bechert, Berlin:
1977> pp. 47g-88.
Smith, J. D. "The Singer or the song? A reassessment of Lord's
'Oral Theory' ", Man (N.S.), 12, 1977, pp. 141-53.
- - - . "Formulillc language in the epics ofIndia", The Heroic
Process. Form,jUnction andfantacy nfolk epic, (eds.) B. Almquist, S.
O. Cathiin, P. O. Healai. Dublin: 1987, pp. 591-6II.
- - - . "How to sing a tale. Epic performance in the Pabuji tradition", Traditions ofheroic and epic poetry, Dol. II Characteristics and
techniques, (ed.)J.B. Hainsworth. London: 1989, pp. 2g-4I.
- - - . The Epic ojPiibUjz. A study, transcription and translation.

Cambridge: 1991.
Syrkin, A. "On the first work of the Sutta Pitaka: The BrahmajiilaSutta", Buddhist studies ancient and modern, (eds.) P. Denwood & A.
Piatigorsky. London: 1983, pp. 153-66.
Warder, A. K. Indian Buddhism. Delhi: 1980.
Weeratunge, S. "Some significant stylistic traits of Buddhist Sanskrit
prose", Buddhist Esstrys: a Miscellany (Volume in Honour
of the Ven. H. Saddhatissa). London: 1992, pp. 26-37.
Still Suffering Mter All These Aeons:

the Continuing Effects of the Buddha's Bad Karma
Sally Mellick Cutler
When we consider the nature of Gotama, the man who is said to have
achieved the status of an Awakened One or Buddha, we generally think of
him as a spiritually supreme individual whose experience of enlightenment
(bodhi, lit. 'awakening') places him apart from all other men. The
biographical details concerning Gotama Buddha preserved in .the
canonical texts of the Theravadin school of Buddhism, the Pali Tipifaka,
generally reveal a life of triumph and perfection. There is no coherent,
connected biography of the Buddha in what are regarded as the earliest
Buddhist texts, but references to him in those texts reveal a man whose
achievement of Buddhahood implied that he could no longer be compared
with any other human, or indeed divine, being. The conception of the
Buddha as a kind of 'super-man', which was greatly developed in the
centuries after the Buddha's death by schools such as the Maha-sanghikaLokottaravadins\ is expressed in unambiguous terms in a number of the
suttas, 'discourses', of the Pali Canon. In the Sor;.adar;.ria Sutta of the Dfgha
Nikaya, the brahmin SOJ:.1adaI;l(;la says of the Buddha: "the ascetic Gotama's
fame is based on his achievement of unsurpassed wisdom and conduct". 2
In the Gopakamoggalliina Sutta of the Majjhima Nikiiya, the elder monk Ananda
says:
There is no one monk entirely and completely endowed with those qualities
with which the Lord, the arahat, the all-enlightened was endowed. For the
Lord was the producer of the unproduced Path, the preacher of the Path
that had not been originated, the knower, cognizer, perceiver of the Path. 3
The special feature which originally set the Buddha, apart was his
realisation of the Dhamma, the true nature of existence, which was the
core of his teaching; this understanding led to his release from sa7J7.Siira, the
64
cycle of existence, and his attainment of nirvibJa, the state in which the basis
for continuing existence has been extinguished. However, many other
special features quickly accrued to him as his renown as a teacher spread,
and emphasis began to be placed on extraordinary aspects of his nature,
and on his separateness from the mass of unawakened beings to whom he
preached. The Lakkha7}a Sutta of the D'lgha Nikiiya is a comparatively late .
discourse, revealing the development of doctrinal beliefs concerning the
nature of a Buddha, and referred to by K. R. Norman as 'an elaborate
piece of Buddhology'" It describes thirty-two special physical attributes
which were said to be the marks of a great man, indicating a being destined
to be either a universal emperor (cakkavattin) or a Buddha. These bodily
marks, which were attributed to Gotama and said to be the result of his
virtuous actions in former lives, included having wheels on the soles of his
feet, a golden complexion, a tuft of hair between his eyes and 'net-like'
hands and feet. In later texts, such as the Pali commentary on the
Vimiinavatthu and the Sanskrit Liilita Vistara, a further list of eighty marks was
added, including physical characteristics such as copper coloured nails and
hidden sinews. In a eulogistic description of the Buddha .in the S07}ada7}(ia
Sutta, the following is included:
The ascetic Gotama is well"born on both sides ... he is handsome, goodlooking, pleasing, of the most beautiful complexion, in form and countenance
like Brahma, of no mean appearance ... He bears the thirty-two marks of a
Great Man. He is welcoming, kindly of speech, courteous, genial, clear and
ready of speech. He is attended by four assemblies, revered; honoured,
esteemed and worshipped by them. 5
In a thirteenth century Sinhalese work, the PVjiivalrya, it is said:

She [YasodharaJ saw the Lord Buddha, who was ever radiant, like the disc
of the rising sun, because of the thirty-two principal marks and the eighty
subordinate marks of a Great One. 6
The possession by Gotama Buddha of such marks is regarded as the

physical indication of his transcendental nature:
An ancient Indian idea that a great man has thirty-two special physical
features was also applied to the idealized image of Buddha; here they are
symbols of qualities attained by cultivation of good practices. 7
This is explained in canonical and commentarial works such as the

Cariyiipitaka and the Jiitakatthava7}7}anii as being the result of his previous
good karma. These texts describe how, during a vast numb.er of previous
lives, the bodhisatta, the Future Buddha, had performed skilful, meritorious
actions in order to fulfill the perfections essential to the aim on which he
was resolved, and reveal that those actions had produced results beneficial
to his attainment of Buddhahood. At the beginning of the Sibi Jiitaka, the
The Continuing Effect of the Buddha>s Bad Karma
65
second story in the (possibly fourth century) SanskritJatakamala, it is said:

. According to. tradition, when the Lord was still a bodhisattva he was at one
time king of the Sibis. This was because his store of good deeds had, by
dint of constant accumulation over an immense period of time, become
part ofhim.B
There is in exisfence a large body of jataka stories in which good deeds
performed by the Buddha during previous births in animal, human and
divine forms are described in order to explain incidents in his final life.
Such stories, over five hundred of which comprise the Pali collection
referred to above, provide an extended karmic biography for the Buddha,
and underline a view of the universe as being characterised by karmic
interconnection and progression. Many of the texts of the Khuddaka Nikiiya>
the division of the Pali Canon which contains some of the latest works to be
accepted as canonical by the Theravadins, are concerned with explaining
the working of karma. These texts typically describe the benefits resulting
from the performance of skilful and meritorious actions, the past career of
the Buddha providing the supreme exemplar. The Vimanavatthu is one such
text; it describes the comfortable circumstances enjoyed by a variety of
deities and explains them as the result of meritorious actions performed by
the celestial beings during former human births. The complementary text,
the Petavatthu, on the other hand, provides karmic reasons for the suffering
being undergone by fifty-one departed ones, petas. 9 Another such text is the
apadana/o a collection of karmic stories in verse, the majority of which deal
with the past lives of Buddhist elder monks and nuns who were the direct
disciples of Gotama Buddha and who arrived at arahat-ship (liberation)
after hearing his instruction. The poems of the Thera- and Thefi-apadana
sections of the collection explain the positions and attainments of the
monks and nuns as being the result of resolves made and actions
performed by them in their former lives. While the past actions are
generally good, some of the poems in the Therapadana also explain certain
undesirable aspects of the final births of the elder monks as being the result
of former bad actions performed by them.
While most of the references to the Buddha in the Pali Canon reinforce
the image of him as an extraordinary man, there are a few stories which
deal with unpleasant events in his life and enable us to consider him in
human, rather than super-human, terms. Of the episodes in the canonical
literature in which the Buddha is said to have been afflicted with physical
suffering, the best-known is that contained in the Mahiiparinibbiina Sutta of
the Digha Nikliya which describes the sickness which affected Gotama after
eating a meal provided by the smith Cunda. This is actually the second
reference in this sutta to the Buddha's being struck by illness, but it is the
one which is most often referred to as it was closely followed by the
Buddha's death:
66

And after having eaten the meal provided by [the smith] Cunda, the Lord
was attacked by a severe sickness with bloody diarrhoea and witl1 sharp
pains as ifhe were about to dieY
Further examples of canonical references to the Buddha being physically

afflicted include two occasions on which he is reported to be suffering
from back-ache l2 and consequently unable to preach. liVe can also include
in this category the physical effects of the severe austerities performed by
the bodhisatta prior to his realisation of the efficacy of the middle path
between austerity and excess; these are referred to in several places in the
Pali. Canon, and are depicted in Buddhist art. There are also several references in canonical literature to occasions on which the Buddha suffered
from verbal or physical attack from people jealous of his reputation as a
teacher and of the popularity of his teachings. These include the three
attempts made by the monk Devadatta to kill the Buddha or to have him
killed, which are recorded in the Vincrya Pitaka,13 and that made by nonBuddhist ascetics in order to have the Buddha and a group of his monks
blamed for the murder of the woman Sundari, which is recorded in the
Udana. 14
These separate incidents are collected together with several others, and
explained as the remaining karmic result of former bad deeds performed
by the Buddha, in one of the poems contained in the Apadana collection. As
noted above, most of the poems in the Pali Apadana reveal the achievements
of the great elder monks and nuns closely associated' with Gotama Buddha
to be the fruit of actions of piety and faith performed by them in former
births. Although the Pali term apada,na, like its Sanskrit equivalent avadana,
has often been translated as 'noble or glorious deed, heroic exploit', a
number of the poems in the Apadana collection reveal this translation to be
inaccurate. Apadanas such as those attributed to the elder monks
Mahamoggallana and Upali contain stories about good and bad deeds
performed in previous lives, and describe how both types of action
continue to bear fruit in the final lives of the protagonists. In the
Upalittherapadana, the elder monk Upali relates two stories of the past. The
first past story describes his performance of an act of generosity towards the
former Buddha Padumuttara, and his voicing of an aspiration for the
future in the presence of that Buddha; these have as their ultimate reward
the attainment of a place in Gotama Buddha's monastic order (sangha), and
liberation from the cycle of existence. The second past story, however,
concerns his harassment of a Buddha, as a result of which he was born as a
lowly barber in his final existence, although this was not a barrier to his
attaining an exalted monastic position and ultimate release. Upali. says to
the Buddha: "Because of that [evil] action, 0 Constant One, I was born
into a low rank. Transcending that [inferior] birth, I have entered the safe
citadel [nirvava]." 15 The version of the Mahamoggallanattherapadana preserved
The Continuing Effect if the Buddha's Bad Karma
67
in the canonical collection is much shorter and less developed than that of
the elder monk Up ali, despite dealing with one of the most celebrated
disciples of Gotama Buddha. The elder monk Mahamoggallana also
initially relates a story of the past in which he describes his honouring of a
former Buddha with musical entertainment and a meal. After he has eaten,
the Buddha prophesies that eventually, "after falling from one of the
hells",'6 the donor will be born as a brahmin, become the second of the
chief disciples of the Gotama Buddha, and attain nirviilJa, "instigated by the
basis of good karma". 17 This reference to a future birth in one of the hells is
apparently connected with the brief admission by the elder monk, later in
his apadiina, that in another former life he had killed both his parents. He
states that, as a result of this, he has always died from an act of violence and
will do so even in his final existence. IS This apadiina, therefore, reveals that
the karmic traces of a former evil action are not cancelled even by an
accumulation of meritorious actions, and demonstrates how they may
remain effective and bear fruit even after liberation has been achieved.
In the commentary on the verses attributed to the elder monk
Mahamoggallana in the Pali Theragiithii collection, the term kammapilotikii is
used to describe the remaining effects of the former bad action and to
explain why the elder monk, despite his attainments, was killed in an attack
by assassins. 19 The Pali word piloti(kii) is defined as 'a thread, small piece of
cloth or rag' and is generally used in the compound patapiloti(kii), 'garment
of rags, rag cloak or robe'. 20 Kammapiloti(kii), therefore, may be rendered in
English as 'strand or remnant of karma, karmic remainder'.21 This term is
only ever used in Pali in the context of bad actions and their results and is
not found in either the apadiina or its commentary in connection with the
elder monk Mahamoggallana. It is, however, used in both texts in
connection with the Buddha himself. The apadiina in which unpleasant
episodes in the final life of Gotama are explained as the remaining result of
bad deeds performed by him in the past is called the Pubbakammapiloti
Apadiina if the Buddha (pubbakammapiloti niima buddhiipadiinaT(t). In two Thai
versions of the Pubbakammapiloti Apadiina there is an apparent attempt by a
late editor to make sense of the words pilotikassa kammassa which occur in the
third verse of the poem, that in which the Buddha explains what he is
about to relate. The Thai editor adds the lines: "Seeing a forest-dwelling
monk, I gave him a rag robe (pilotika). I then made my first resolve for
buddhahood", which creates the following ending for the verse: "[The
fruit] of my rag robe [-giving] action ripens even while I am Buddha."
Without the interpolated segment, the verse reads: "Listen to me 0 monks.
[Hear about] the action done by me [and how the fruit] of the remainder
of that action ripens even while I am a Buddha." The reference to a good
action at the beginning of this apadiina is, however, misleading and
inappropriate; all other versions of this text introduce it as a description by
the Buddha of his own past bad actions as a means of providing karmic
68
explanations for unpleasant events which affected him in his final life. In
the commentary on the Udiina, it is stated that the remaining frllits of
former actions which caused pain and other kinds of unpleasantness to the
Buddha are also known as kammiini pilotikiini. 22
Although the Pubbakarnmapiloti Apadiina is attributed to the Buddha, it
is not included in the Buddhiipadiina section of the collection which in
its- current form contains only a single eponymous poem describing
an otherwise unknown story from a past life of the Buddha. The
Pubbakammapiloti Apadiina is actually contained within the Theriipadiina
section,23 a placement which reflects the problematic nature of the text. It is,
however, comprehensively discussed in the exegesis on the-last verse of the
Buddhiipadiina in the commentary on the first three sections of the apadiina,
the V1SuddhqjanaviliisinP4 The first two verses of the Pubbakammapiloti Apadiina
are, however, discussed in the appropriate part of the commentary on the
1heriipadiina, indicating that this poem had been placed in its probleqlatic
position by the time the V1SUddhajanaviliisinfwas composed. The commentary
does not give any explanation for the placement of this apadiina of the
Buddha in a section devoted to karmic biographies of elder monks. It is
possible that it reflects a deliberate attempt by an early redactor to obscure
the connection of this particular poem with the Buddha because of its
problematic doctrinal implications. This solution is not, however, entirely
satisfactory, as the structure of the apadiina collection is such that the
complete removal of the poem, rather than its relocation, would have been
possible as well as doctrinally justifiable. In its commentary on the
1heriipadiina, the Visuddhajanaviliisinfprovides glosses on 56r apadiinas (the last
two of which are duplicates of poems included earlier in the collection).
However, currently available versions of the 1heriipadiina generally contain a
total of 550 apadiinas, in an apparent attempt to emphasise the connection
between that collection and the JiitakatthavaTJr;.anii. It seems likely that the
nine additional poems were deliberately removed from the collection by a
redactor who wished to emphasise the correspondence between the apadiina
stories of the great elder monks and the jiitaka stories of the, Buddha. 25 If this
is so, it implies that the Pubbakammapiloti Apadiina could have been just as
easily excised by a redactor who did not accept the position that the
Buddha's sufferings were the result of his own bad karma. That it was not is
one of several interesting questions raised by the presence of this poem in
the apadiina collection.
A version of this poem may have existed before the first sectarian
splits in the Buddhist monastic order, as avadiinas corresponding to the
PubbakarnmapilotiApadiina are preserved in Sanskrit, Tibetan and Chinese,26
and it is possible that a solution to the problem of the poem's placement in
the Pilli collection may be found through a comparative study of these
texts. 27 The structure of the Pubbakammapiloti Apadiina does not conform to
that of the standard apadiina and, like its subject matter, places it apart from
The Continuing Effect ojthe Buddha's Bad Karma
69
the other poems in the Theriipadana section of the Pali collection. In twentyseven verses it provides succint descriptions of twelve past misdeeds and
their results, thereby drawing clear causal connections between the past
and the present within a narrative framework which differs significantly,
not only from that of even the shortest of the other apadiinas in the
Theriipadiina, but also from that of the Buddhiipadiina with which it could be
expected to correspond. The PubbakammapilatiApadiina is introduced by two
verses which locate the Buddha amongst a group of monks (the number is
not specified) gathered near Lake Anotatta, which is said to have been a
favourite resting place of the Buddha and a popular place for assemblies. 28
Such introductory verses are not part of the standard apadiina pattern and
they are not found in any other poem in the Theriipadiina, although they are
found in the Buddhiipadiina and the Paccekabuddhiipadiina. 29 The significance
of the possession of this feature by such a limited number of poems in the
Pali collection can only be properly evaluated by studying them in a wider
Buddhist context and by comparing them with corresponding texts in the
Theravadin and other Buddhist traditions.
Although the current form of this poem describes twelve former
misdeeds of the Buddha, it should be noted that the version of this apadiina
preserved in two manuscripts from Thailand and in the most recent Thai
printed edition of the apadiina contain verses describing a thirteenth
example of former misconduct and its fruit. While there is no reference to
this episode in the V1Suddhajanaviliisinf, it appears to suggest that an enlarged
version of the Pali text once existed and circulated as an independent text,
presumably postdating the composition of the commentary and possibly
confined to mainland South East Asia. It does seem that, despite its
apparent concealment deep within the Theriipadiina, the Pubbakammapiloti
Apadiina continued to be developed and discussed as an individual poem,
separate from its canonical context. This process may have occurred
alongside the general discussions in Pali commentarial and post-canonical
literature about whether Buddhas were, like all other beings; subject to the
effects of previous bad karma, or whether they were karmically free. The
concentration on 'problem action' (panhakamma) and its results in the
Pubbakammapiloti Apadiina not only conflicts with any view of the apadiina
genre being one comprising only stories of glorious and noble deeds, but
also distances it from all the other poems in the Theriipadiina. As we have
seen, the other poems containing descriptions of bad action and its result
place them alongside descriptions of good action and its results. This.may
indicate that an apadiina of the Buddha dealing with good karma was
originally linked with the Pubbakammapiloti Apadiina so that a complete
karmic biography was revealed through the contrasting texts. The
Buddhiipadiina which comprises the first section' of the Pali collection does
not, however, appear to fit the requirements for such a text in either form
or content. In this context, a comparative study of this apadiina and the
70
Cariyiipifaka, which IS largely based on the Jataka collection and which calls
itself an apadiina of the Buddha (buddhiipadiiniya) , could be productive. As
Jonathan 'Walters says: "the stories about bad karma and bad effects are
part of the same story which tells of good karma and good effects". 30
In the Visuddhajanaviliisinf, the cOrnrrlentator has not followed the order of
the verses as they occur in the apadiina. The verses have been discussed in
an order chosen to present the past actions according to the chronology of
the Buddha's final life and to reflect the 'historical' order of the unpleasant
events which they explain. The order of events in the poem itself, however,
appears to be based on a classification of those events which are
experienced as a result of the former actions into three general categories:
slander from enemies, assaults by Devadatta, and physical illness or
deprivation. 31 In the Pubbakammapiloti Apadiina the Buddha begins his
explanation of his former actions by saying:
Listen to me, 0 monks. [Hear about] the action done by me [and how the
fruit] of the remnants of [that] action ripens even while [I am] a Buddha. 32
The story which comes first in the Thai version of this apadiina deals with
an occasion on which, as a cowherd, the bodhisatta prevented a cow from
drinking water. He says: "As the fruit of that action (tena kammavipiikena),
here in my final birth I am not able to drink as I wish when I am thirsty."
This is apparently a reference to the Buddha's request, reported in the
Mahiiparinibbiina Sutta, that Ananda bring him water from a muddy stream
after he has fallen ill,33 a request with which Ananda is unwilling to comply.
In the first story in the majority of extant versions of the Pubbakammapiloti
Apadiina, the Buddha states that he was once a scoundrel named Munah
who slandered a paccekabuddha. The next two verses read:
As the fruit of that action, I transmigrated in hell for a very long time,
experiencing sensory suffering for many thousands of years. As the
remaining effect of that action (tena kammiivasesena), here in my final birth I
was slanderously accused over Sundart. 34
The story of Sundart is found in both canonical and commentarial sources;
it describes how a woman was persuaded by a group of ascetics hostile to
the Buddha to damage his reputation by claiming that she spent her nights
with him in his perfumed hut. She was then murdered and responsibility
for her death was falsely attributed to him.35 The next episode in the
apadiina also deals with slander, and it is apparent that the Buddha's
concern with the outward appearance and public conduct of the monks in
his sangha, frequently expressed in both vinO:Ja and sutta texts, is based on an
awareness of the strength of public concern about the morality of
wandering ascetic teachers and their followers. The Buddha continues his
catalogue of past misdeeds by stating that, in a former birth, he slandered a
follower of a buddha named Sabbabhibhii who is not otherwise known in
The Continuing Effect qf the Buddha's Bad Karma
71
the Pari tradition. He says: "For that slanderous accusation, I transmigrated

in a hell for a very long time",36 and reveals that the remaining fruit of his
action was that he was the object of much false accusation when he
achieved birth as' a human. The culmination of this was the claim by the
non-Buddhist ascetic woman CificamaI).avika that she had spent a night
with Gotama and had become pregnant by him. Cificamai;tavika: is also
linked with attempts to harm the Buddha in a number ofjiitaka stories, and
there appears to have been some conflation of the episodes concerning
these two women37 in the Buddhist literary tradition. In the third story in
the Pubbakammapiloti Apadiina, the fruit of the action is not linked directly
with the Buddha and it is also used,like the first story, to provide a karmic
background to the Sundari story. In this episode, the Buddha relates how
he once falsely described a powerful seer as a sensualist to his students.
The students then spread the li~ from household to household as they went
on their alms rounds, as a result of which they were all included in the
accusations over the death of Sundarl. 38
The next group of stories is related in order to provide karmic
explanations for the assassination attempts instigated by the Buddha's
jealous cousin, Devadatta; these are collected together in the Cullavagga of
the Vincrya Pitaka, where the great power of the Buddha rather than the
unpleasantness of the events is emphasised. The first of the three stories in
this category tells how, in a former birth, the Buddha killed his half-brother
for his inheritance by throwing him over a cliff:
Formerly [in another birth], greedy for wealth, I killed my own half-brother
by throwing him off a mountain path and crushing him with a rock. 39
As a result of that action, Devadatta attempted to kill the Buddha by

throwing a huge rock at him. However, only a splinter of rock actually
struck the Buddha, piercing his toe and making it bleed. The next story
describes how, as a child in a former life, the Buddha threw a piece of
pottery (presumably intentionally) at a passing paccekabuddha. He says: "As
the fruit of that action, here in my final birth Devadatta hired thugs to killme."40 This is the only story among those in the first two groups in which
the result does not seem to parallel the former bad action. Interestingly, the
corresponding verse quoted here in the apadiina commentary gives a
completely different, and more appropriate, result for this action. It says:
''A.s the fruit of that action, the discontented Devadatta threw a stone at me
after I had become a Buddha."41 The third story in this group explains an
event which was very popular in canonical and post-canonical Buddhist
literature and in Buddhist art, namely the attack on the Buddha by the
elephant Nalagiri. The Buddha describes how, in a former birth, he was an
elephant-trainer who caused his elephant to attack a paccekabuddha who was
going on an alms-round. He says:
72
Indian Insights: Buddhism; Brahmanism and Bhakti

As the result of that action, the fierce elephant Nalagiri was let loose and
charged me in the excellent town of Giribajja!2
Although it is not explicitly stated in the apadiina, this attack is generally

said to have been instigated by Devadatta.
The final group of stories in the Pubbakammapiloti Apadana provides a
series of karmic explanations for a variety of physical discomforts said to
have afflicted the Buddha. In the apadana, the result of the final unskilful
action which is described affects Gotama prior to his attainment of
enlightenment leading to its being the first to be commented on in the
Visuddhajanavilasinz. The first of this group of six stories deals with a former
time in which the Buddha was a king and, in a fight, killed a man with a
sword. He says:
As the result of that action, I was greatly tormented in a helL As the
remaining effect of that action, now the skin on my foot is completely
ulcerated, for the effect of action never perishes [without result] .43
The commentary on this verse relates this affliction to the incident in
which Devadatta threw a rock at the Buddha, and states that an infection
or abscess developed from the splinter wound on the Buddha's toe. The
next story is most unusual, and it is the only one to refer to an unskilful
action of thought rather than of word or deed. It is also very well attested,
being found in every version of the text, although the result is not identical
in each instance. The Pali version reads: "[In another former birth] I was a
fisherman's son living in a village of fishermen. Seeing fish killed made rne
happy. "44 The result of this pleasure in the suffering of other beings is said
to have been that he was afflicted with headaches in his final life. In this
story, a more unpleasant karmic reward is said to have resulted for the
fishermen who actually killed the fish: they were born as Sakyans and were
all slaughtered by Vi<;lu<;labha, the king of Kosala, whose father was a
contemporary of the Buddha. 4s The next story deals with a former birth in
which the bodhisatta insulted the followers of the former Buddha Phussa,
. telling them to eat barley, an inferior grain, rather than rice, and thereby
implying that they were not worthy of superior food. 46 This resulted in the
Buddha being forced himself to eat barley throughout the three months
Rains Retreat which he spent in Verafija during a time of famine; an
account of this episode is contained in the opening section of the Vincrya
Pitaka. The next story relates to a birth in which, as a wrestler, the Buddha
caused the death of another wrestler during a competition. Because of this
action, in his final life he was afflicted by the back pain which was referred
to above. 47 The penultimate story describes how, as a doctor in a former
birth, he unnecessarily administered a purge to a rich patient, as a result of
which he suffered from diarrhoea at the end of his life. 4B The final story in
the Pali apadiina describes how in a former life, as the youth J otipala, he
The Continuing Effect qf the BuddhaJs Bad Karma
73
insulted the Buddha Kassapa by saying: "How then could enlightenment

come to a baldy? Enlightenment is very difficult to obtain."49 As a result of
this, the bodhisatta had to spend six years practising severe austerities at
Uruvela before realising that he was searching for the truth on the wrong
path. Gotama was only able to attain enlightenment and reach his goal
after this lengthy period of striving, which was far greater then that
undergone by any of the former Buddhas whose biographies are set out in
the Buddhavarrzsa.
The most accessible of the non-Pali versions of this text is that preserved
in the Bha~ajyavastu division of the Mi1lasarvastivadin Vinl9la, in the section
which is usually referred to as the Anavataptaglitha,50 and which is now
available in a French translation. 51 This translation is based on Tibetan and
Chinese sources, supplemented by Sanskrit fragments from Gilgit. In the
Anavataptaglithli, a prose description of eleven examples of former unskilful
actions performed by the Buddha precedes a versified account in which
both the order of events and their number differ from those of the prose
text. 52 There is also a Sanskrit version of this fext preserved in the
Dafakarmaplutyavadlina section of the IIth century Bodhisattvlivadlinakalpalatli
(Avadlina-kalpalatit) which corresponds to the verses of the Anavataptaglithli in
form and in the number of events which it describes.
vVe have seen that the standard version of the Pubbakammapiloti Apadlina
contains descriptions of twelve episodes from former lives of the Buddha,
two of which apparently relate to a single result, although they are treated
separately in both the text and the commentary. The results can be
summarised as follows: two apparent episodes of slander by Sundart and
one of slander by Cificamru;tavika; three attacks carried out or instigated by
Devadatta, although only one .of them actually causes injury to the
Buddha; four instances of physical illness: an infected foot, headaches,
backache and diarrhoea; and two instances of physical discomfort: being
fed on inferior food for three months and undergoing austerities for six
years. The verse section of the Anavataptaglithli and the Avadlinakalpalatli both
contain descriptions of ten former unskilful actions and their fruit. All ten
of the stories related in the Anavataptaglithli are also found in the Pali apadlina
text, although they do not completely agree in three particular instances,
and the order of events is not identical in the two texts. Only one of the
episodes related to Devadatta's attempts to destroy the Buddha is described .
in the Anavataptaglithli, namely that concerning the bodhisatta's murder of his
half-brother which resulted in the only successful physical attack by
Devadatta on the Buddha. The most significant difference is found in the
Mulasarvastivadin story used to explain the injury to the Buddha's foot
which the Pali text states to be the result of one nobleman killing another
man in a fight. In the Anavataptaglithli account, the Buddha describes how, in
a former life as a merchant, he killed a fellow merchant on his ship with a
sword; he explains that the remaining result (karmlivafesena) of this was that
74
his foot was pierced by a splinter of wood. Differences between the two
versions of the remaining stories are slight, as can be seen in the follpwing
examples. In the Anavataptagatha account of the second of the former actions
connected with the Buddha being falsely accused by Sundart, it is actually
stated that the fruit of that action was suffered by the Buddha as well as by
those who were his pupils at the time. Again, in the Anavataptagathii version of
the unusual story connected to the Buddha's previous birth as a fisherman's
son, the headache said to have been suffered by the Buddha as a result of his
evil action is specifically linked to the occasion on which his former
companions were killed by Vi<;lu<;labha. In the Mulasarvastivadin version of
these episodes, the descriptions of the former actions and their results
appear to have been constructed more logically and therefore express the
karmic conections more clearly than those in the Pali apadana.
In the Anavataptagathii, the Buddha is said to have described his former
actions and their continuing effects in response to the questions of a group
of 500 monks who had accompanied him to Lake Anavatapta, the SanskTit
equivalent of Anotatta. According to the Mulasarvastivadins, this was one
of the ten acts which must be performed by a Buddha before he can attain
nirva1Ja. 53 The Buddha introduces his description by requesting the
assembled monks to listen to an account o( the evil action (papakarman)
performed by him in the past. 54 The majqr part of the Anavataptagathii is
comprised of the stories of 36 of those monks, in which they reveal "the
fabric of their karma"55 through the recitation of episodes from former lives
and the results of the actions performed during them. Hofinger uses a
variety of translations including the one above for the Tibetan term las 19i
rgyu ba, which he equates to the Sanskrit term karmaploti,56 in which ploti,
'thread, cord', is used in the sense of 'connection, interweaving'.57 However,
a more detailed study of this term would be helpful, as Tibetan translations
of other similar texts use the slightly different term, las 19i rgyud, which is
actually equivalent to Sanskrit karmasaTfltana or karmasaTfltati, 'continuity of
karma'. As Paul Harrison points out in his discussion of one such text:
'Continuity of karma,' Tib. las k,yi rgyud, stands here doubtless for Skt. karmaploti or karma-pluti. 58
.
.
It is apparent from the Anavataptagathii that this term was used in Sanskrit in
the context of both good and bad actions although, as we have noted above,
the Pali term kammapiloti is only ever used to refer to bad actions producing
unpleasant results.
The section of the Anavataptagatha concerned with the karmic history of
selected elder monks is followed by a verse eulogy of the Buddha, after
which are placed the prose and verse accounts of the Buddha's bad karma.
It is possible that it is this arrangement which is reflected in the apparently
anomalous placement of the Pubbakammapiloti Apadana in the Pali collection.
Further support for this suggestion may be drawn from the fact that the
The Continuing Effict of the Buddha's Bad Karma
75
apadiina immediately preceding the PubbakammapilotiApadiina, the

So r;akojiufsa ttheriipa diin a, is almost identical with the poem attributed to
Kotivirpsa in the Anavataptagiithii, and ends with the words:
The elder monk SOI;a Kotivisa, at the head of the company of monks,
answered the question [on karma] asked at the great lake Anotatta. 59
If there was originally some connection between the Anavatptagiithii and the
apadiina, it is possible that the number of stories attributed to elder monks
in the Pali text originally may have been 500, corresponding to the group
of 500 monks at Lake Anotatta, rather than 550 corresponding to the
number ofjiitaka stories associated with the Buddha. This could also suggest that the earliest version of the Theriipadiina may have contained a general introduction which was lost early in the text's transmission, or that
each apadiina attributed to an elder monk had its own introductory verses
corresponding to those in the Buddhiipadiina and the Pubbakammapiloti
Apadiina and to the brief introductory statements in the prose stories of the
Anavataptagiithii.
In the Avadiinakalpalatii version of this text, the Buddha relates the stories
of his former bad actions to a group of monks, in an unspecified setting,
during an account of the former lives of his two chief disciples in order to
illustrate the ma.-xim that" every creature must suffer from the effects of his
works".5o Here again we have a connection between the past stories of elder
monks and those of the Buddha himself In the Avadiinakalpalatii account, the
term karmapluti is used and is taken, I am sure correctly, to be a variant form
of karmaploti. Prof. Gombrich has, however, pointed out that it could also
make sense to see pluti, 'leap',61 being used here in the same sense as in the
Sanskrit technical term mar;p.ukapluti, 'a skipping of sutras' .52 Karmapluti would
then have the slightly more specific meaning of 'karma submerged and
skipped during the development of the perfections and re-emerging to be
worked out in the final existence'. Eight stories in the Avadiinakalpalatii
correspond to those in the Pubbakammapiloti Apadiina and the Anavataptagiithii,
but it contains only one story of a former action connected with slander, the
result of which is said to have been "slander by women". Furthermore, it
does not contain any reference to the attacks on the Buddha said to have
been instigated by Devadatta. It follows the Anavataptagiithii in setting the
story about killing a man with a sword in the context of a dispute between
two merchants. In the Avadiinakalpalatii, however, the result of this is not
connected with any action performed by Devadatta against the Buddha in
his final existence; the Buddha is simply said to have suffered from a
particular type of ulcerous sore (khadiravrar;a) on his foot. 63 In the episode
concerning the Buddha's former birth as a wrestler, the Avadiinakalpalatii
states that the bodhisatta killed his opponent by breaking his back, but says
that the result of this was that he suffered from fatigue rather than
backache. 54 Neither the Avadiinakalpalatii nor the Anavataptagiithii contain any
76
reference to the extra episode concerning the coyvherd found in the three
Thai versions of the Pubbakammapiloti Apadiina. Further work is still required
on the Dasakarmmapluti chapter of the Avadiinakalpalatii, a slightly daunting
task as it is both garbled and rambling and therefore less accessible than the
succinct versions of the text contained in the Pubbakammapiloti Apadiina and
the Anavataptagiithii.
The Avadiinakalpalatii also contains descriptions of two episodes from the
former life of the Buddha which do not occur in the Pali or Mulasarvastivadin versions of the text, but which do overlap with episodes related in
three other corresponding texts. The first of these episodes relates to a
former life in which the bodhisatta upset the bowl of a paccekabuddha (Sanskrit
pratyekabuddha), the result of which was that he did not receive any alms
food on one occasion in his final life. The second episode refers to his
attempt in a former life to blame an innocent paccekabuddha for a fault of his
own. The text states that, although he later confessed to his fault, the
remnant effect of that action was that he was insulted by brahrnins even
after he had achieved Buddhahood. Both these events are dealt with in
Part Three of the UpayakauSa[ya sutra, a version of which has recently been
translated into English by Mark Tatz, who regards it as a sutra of the early
Mahayana school and dates it tentatively to theIst Century BeE. 65 'While
the six years of austerities are referred to in the second section of the sutra, it
is the third section of the text, that which Tatz entitles The Ten Karmic
Connections (reading karmasa1'(!tatz), which corresponds to the Pubbakammapiloti
Apadiina. It describes the unpleasant events which afflicted the Buddha in
his final life in order to demonstrate that the apparent sufferings of the
Buddha are inspired by his skill in means (upayakauSa[ya) and to support a
view of the Buddha as supramundane (lokottara). Tatz says:
For the most part, the Upifya addresses incidents belonging to the common
tradition that are of special concern because they seem to contradict the
transcendentalist conception of Sakyamuni. 66
The Buddha introduces the section on karmasa1'(!tati with the statement that
the ten karmic connections should be regarded as "having a hidden
meaning" as "there is no possiblity at all of a fault stemming from an
obstacle caused by past deeds".67 The first episode dealt with in this chapter
is that in which his foot is (apparently) pierced by an acacia (khadira) thorn.
He tells Ananda that he killed a merchant with a spear in a previous
existence, and that the injury to his foot is "the residue of the fruitition of
that deed". 68 This is, however, only a stratagem to make people understand
the functioning of karma and realise gnosis V"iiiina), and he further states that
the thorn is a skill in means and not "an obstacle caused by past deeds".69
The two results which correspond to those in the Avadiinakalpalatii are
similarly said to be examples of skill in means. When the Buddha says that
he returned from an alms round with an empty alms bowl, it is in order to
The Continuing Effect of the Buddha's Bad Karma
77
console other monks who receive no alms, and to inspire householders and
gods to turn to his teaching. 70 This text provides an example of a specific
instance of abuse rather than saying generally that the Buddha was insulted
by brahmins; it states that, when he was scolded by the brahmin
Bharadvaja, it was in order that the example of his forbearance in the face
of such provocation should generate the idea of enlightenment iri four
thousand sentient beings.
These episodes are also found in the section known as the UpayakaufafyaViniscaya, 'the analysis of skilful means', in the fourth part of the
SaTflskrtasa7J7.Skrta- Viniscaya. This work, which is available only in Tibetan,
and which has been briefly studied by Peter Skilling,71 is:
a wide-ranging treatment of the universe ... according to a number of
Buddhist schools and teachers, both of the Sravakayana a'nd the
Mahayana. 72
The placement of the Upayakaufafya- Viniscaya within this compendium is
problematic, reinforcing its connection with the Pubbakammapiloti Apadana
and reflecting the nature of its response to the concept that the Buddha
continued to be affected by the results ofms previous bad karma, Although
the list of incidents in the Up ayakausafy a- Viniscaya is attributed to the
Sammitiya school, a Sthavira sect, it actually occurs in that part of the
compendium which deals with Mahayana tenets and texts. The author,
Dasabalasrlmitra, introduces the list by stating that ':A.ccording to the Arya
Sammitiya school, the Lord [committed] sixteen karmic lapses". 73 In this
version of the text, the Buddha is said to have abused a paccekabuddha in a
former life and, as a result, to have been himself abused by the brahmin
Bharadvaja in his final life. In another former life, he hid the alms bowl of a
paccekabuddha, as a result of which he did not receive alms on one occasion
and had to return with an empty bowL This version of the text contains all
the episodes which are found in the Pali and Mulasarvastivadin versions,
although the episode in which he is said to have given the wrong medicine
to a patient is related twice, the second time the result is said to have been
that the Buddha suffered from "the illness of langour". There are three
additional episodes which have no counterpart in any of the texts previously
considered, and they have been summarised as follows:
(1) Because of the karma he created by disobeying the instructions of his

teacher. '. the group of five monks rejected [the bodhisattva] as their teacher,
(2) Because of the karma he created by causing dissension within the group
of disciple of the sage Gangiyar, his sangha became divided at the city of
Kosapara. (3) Because of the karma he created when, as a great doctor, the
Lord estranged his rival from the King, his sangha was divided by Devadatta.
The citation ends with the verse:
78

As a result of this karma much ~uffering was undergone in the hells; the
related fruit of this karma is called the remainder to be experienced 0l)ly.74
Not one of those additional episodes occurs in the last text I wish to deal
with, which also does not appear to contain any account relating to abuse
from a specific brahmin or from brabrrllns inL general. This work, the De Tzs
Karmaya, is a Sinhalese prose text tentatively dated by its original cataloguer,
Hugh Nevill, to before the 14th century;75 it is available only in manuscript
form. It calls itself an akusalapadiina sannaya, 'a paraphrase of the apadiina
(about) unskilfu1 (actions)" which presumably refers to a version of the
Pubbakammapiloti Apadiina although, as its name suggests, it is supposed to
deal with a total of 32 former actions,76 considerably more than are
contained in the Pali text. All the espisodes from the Pubbakammapiloti
Apadiina are described in this work but, apart from an apparent reference to
the occasion on which the Buddha was unable to obtain alms food, the
additional episodes do not seem to overlap with those in the Sanskrit
versions. Nevill paraphrases this as follows: "Owing to a fault in a former
life, Gautama Buddha once went a day without food. The cause is not
explained."77 It does, however, contain, as its first example of a former
misdeed, the story of a herdsman who would not allow his cattle to drink
muddy water. This is said to have resulted in the Buddha suffering from
thirst at the time of his parim'rviir;.a and is equivalent to .the extra episode
which is found at the beginning of the Thai versions of the Pubbakammapiloti
Apadiina. Nevill attributes this text to the 'heretical' Dhammarucika sect of
the Abhayagiri monastery, but it may simply represent the development in
Sri Lanka of a text originally brought from Thailand or Cambodia. It also
includcr three extra episodes which are not attested elsewhere. In the fITst
one it is said that the Buddha had formerly accused an ascetic of being a
thief, as a result of which he was abused by a drunkard. It is also said that, in
a former life, he blocked the path of a paccekabuddha with rubbish as a result
of which his path was blocked on one occasion when he set out on an alms
round. Finally there is a complicated episode which Nevill summarises thus:
In the Kusa birth, he took back some cakes allotted to a Pase-Buddha
[paccekabuddha]. In retribution a woman gave him a plate of rice, thinking he
was [the elder monk] Maha-Kassapa. Seeing Maha-Kassapa approach, she
demanded it back. 78
There is a further Sanskrit text, the Karmaplotika sutra, which occurs as

the 13th legend in the Bodhisattvajiitak-iivridiinamiilii and the 56th legend in
the ]iitakamiiliivadiinasutra,79 but I have not yet been able to do any work on
the copy of it which I have obtained. Study of this sutra may prove to be
particularly rewarding as, in both of the collections in which it is found, it
is placed after a text called the Siilryasi:rrthajiitaka, which deals with the life of
the Buddha up to his attainment of enlightenment. Further work on this
text, and on the De Tis Karmaya, is a priority.
TIe Continuing Effect oj the Buddha's Bad Karma
79
It is apparent that there were a number of stories .of this type from which
the various versions of the pubbakammapiloti text were compiled and that the
idea of a collection 'of stories explaining unpleasant incidents seen to have
affected the Buddha' during his final life must belong to a common, and
ancient, source. Peter Skilling says of these individual stories about the
Buddha's bad karma:
.
there is sufficient agreement between the available sources to suggest that
either the basic accounts predated the fundamental schisms, or that the
schools borrowed from each other.80
Study of the various related texts would seem to show that the basic
accounts are certainly older than any divisions, the use made of the stories
reflecting the attitude of the various schools to the relationship between the
Buddha and the law of karma.
Notes
I.
2.
3.
4.
5.
6.
7.
8.
g.
ID.
II.
12.
'3.
'4.
'5.
16.
'7.
18.
Ig.
20.
21.
For a discussion on the place of thi, school's view of the Buddha as supramundane (Iokottara)
with regard to the origins of Mahayana Buddhism, see Williams, Ig8g, pp. 16-20.
.
Translation Walshe, Ig87, p. 128. Text DN I p. n6, 20--:22. Abbreviations of Pali texts follow
those of the Epilegomena of the Critical Piili Dictionary, Ig48. The abbreviation ap. is for
apadiina.
Translation Thomas, Ig49, p. 212. Text at MN III p. 8,8-14; ;cSN III p. 66,15-18, where the
statement is attributed to the Buddha.
Norman, Ig83, p. 42.
Translation Walshe, Ig87, p. 127. Text DN I p. n6.
Translation by H. Peiris and L.C. van Geyzel at Reynolds, 1970, p. 182.
Cleary, 1993, p. 1641.
Translation Khoroche, Ig8g, p. 10.
Fragments of a Turfan manuscript of the Sarvastivadin K.sudrakiigama indicate that it
contained two texts identified as the Vimiiniivadiina and Pretiivadiina which Bechert reveals to
have been based on the same tradition a, the Pali vaitlzu collecions. See Bechert, Ig74.
Quotations- !i'om the apadiina are taken from the critical edition with translation of portions
of that text contained in my D.Phil. thesi" Oxford University, 1993.
Translation Walshe, Ig87, p. 257. Text DN II p. 127,34-36.
These occur at MN I p. 354, 25-26 and SN IV p. 184, 8.
See Horner, Ig52, pp. 266-71.
Udp.28 ..
Upiilittlzeriip. y. 133: lena kammena alUl1Jl dlzlra ilinattll1Jl qjjlzupiigatoisamatikkamma laTfljiitbfl PiiviJiTrz
ablz'!)lll1Jl puraTfl
Malziimoggalliinattizeriip. v. '4: nir'!)lii so cavitviina.
Ibid. y. 15: kusalamulena codito.
Ibid. vv. 18-lg:yaTflyaTflyon'upapqjjiimi nir'!)laTfl atlza miinusaTfllpiipakammasamatigittii blzinnaslso
mariim' aizaTfLI Iidll1Jl pacclzimakaTfl m'!)llzaTfl carimo vattate blzavolidlziipi ediraTfl m'!)llzaTfl mara(lllkiile
bizaviJ'Sati.
Th-a III p. 180,n-13.
See PED p. 39', S.Y. pata. The term patapilotiJcakantiziJcii, 'wearers of patched rag robes', is also
found in Hofinger, '982, p. '95.
Peter Masefield suggests "karmic fluff" in his tranlation of the section of the Udiina
Commentary in which thi., apadiina is quoted in full. See Masefield, 1994, translation
pp. 632-35 and extensive notes pp. 713-22.
-
80
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39
40.
41.
42.
43.
44.

Ud-a p. 263,12.26.
It is number 390 in all editions oC the Therapadiina except that published by the Palj Text
Society where it is number 387.
The Visuddlzajanauiliisinf does not contain any commentary on the Tllen-apadiina, although its
anonymous author reCers to the number of apadiinas in that section at the beginning of his
commentary e.g. Ap-a p. !OI, 20. For the date of this commentary (post-eleventh century)
see Cousins, 1972.
These nine poems are, however, included in the most recent editions of the apadana in
Burmese, Devanagarj and Sinhala scripts. This is probably due to attempts by the modern
editors to reconstruct the very corrupt text of the collection with the help of the commentary,
although this is not indicated in the notes on the text.
See Norman, 1983, p. 92; Walters, 1990, pp. 77-9; and Skilling, 1992, pp. 137-43.
This possibility was first raised in Bechert, 1953, pp. 10-12.
See Malalasekera, 1974, pp. 96-9.
They are also found in several apadana-s attributed to elder nuns e.g. the Yasodlzarii thenapadiina but and their presence in poems of the Tllen-apadiina requires further study.
Walters, 1990, p. 88.
This categorisation is given in Walters, 1990, p. 71.
Pubbakammapiloli ap. v. 3: su~iitha MikMzavo mayhaT(l yaT(l kammaT(l pakataT(l maya/pilotikassa kamma.lsa buddhatte pi vipaccali.
DN II pp. 128-9. Translation in Walshe, 1987, pp. 257-8.
Pubbakammapiloli ap. vv. 5 and 6.
See Malalasekera, 1974, pp. 1216/.
Pubbakammapiloli ap. v.7cd: taT(l abbhakM,va niraye ciraT(l saT(lmritaT(l maya.
See Malalaqekera, 1974, p. 864.
Pubbakammapiloli ap. vv. 10-14.
Pubbakammapiloti al'. v. 15: vematubizataraT(l pubbe dhanahetu haniT(l ahaT(l/pakkllipiT(l giriduggasmiT(l
siliiya fa apiT(lSayiT(l. Padas ab are quoted in the lvfilinda-!ikii p. 26,20.
Pubbakammapiloli ap. v. 18: tena kammavipiikena idlza pacellimake blzave/vadlzattlzaT(l maT(l devadatto
aMimiire pay,!jayi.
Ap-a p. 123,5-6: tena kammavipiik~na buddlzablziita.rsa me sato/paviddlzesi silaT(l tattlza Devadatto
viglziitava.
Pubbakammapiloli ap. v. 20: tena kammauipakena blzanto nalagiri gajo/giribbaje puravare diiru~ maT(l
upagami. Bhanto, 'uncontrolled', appears to be intended aq a causative here.
Pubbakammapiloli ap. w. 2ICd-22: tena kammavipiikena niraye paccilYlT(l bhusaT(l/ /kammuno tassa sesena
idani sakalaT(l mama/pad, elzavi pakopesi na hi kammaT(l vinassali.
Pubbakammapiloli ap. v. 23' alza7[l keva!!agiima.rmiT(l ahuT(l keva!!adiirako/macchake ghiilite dil"ViijanayiT(l
J'omanassakaTJl
45.
46.
47.
48.
49.
50.
51.
52.
53.
5455
56.
57.
58.
59.
See Malalaqekera, 1974, pp. 171-2 and 876j.

Pubbakammapiloli ap. w. 25-26.
Pubbsakammapiloli ap. v. 27.
Pubbakammapiloli ap. v. 28.
Pubbakammapiloli ap. v. 29cd: kuto nu bodlzi mu~[lassa bodhi paramadullablza.
Lamotte refers to it aq the Paiicaiatastlzauiriivadiina and says that it was translated into Chinese
in 303 GE (Lamotte, 1988, p. 692).
See Holinger, 1990, pp. 123-8. See also Bechert, 1961.
Holinger caUs this poetic version the Buddlziivadiina although, in his review of Holinger's
translation, Peter Skilling points out that there is no evidence lor such a title in the text itself
and states that it "seems to be an invention oC the author."
Holinger, 1932, pp. 175/.
Holinger, 1990, p. 123: "Ce que .l'ai commis, moi aussi, comme peches (papakarman),
ecoutez-en ce recit de rna bouche".
See Hofinger, 1982, e.g. pp. 199, 202, 205 etc.
Holinger, 1990, p. 85, note 2.
See MW p. 715 and BHSD p. 395.
Hanison, 1995, p. II.
See Bechert, 1961, p. 21 and Tllerapadana 389: thero kO/ivlW so~ bhikkl,uJ"anglll1.rsa aggato/paiilzaT(l
pu!!lzo viyaka.li anotalle mal,a.)"ar, Ii.
The Continuing Effect of the Buddha's Bad Kama

60.
61.
62.
63.
64.
65.
66.
6].
68.
69.
70 .
71.
72 .
n
7475.
76.
77.
78.
79.
80.
8r
Mitra, 1882, pp. 57-8.

MW p. 715.
MW p. 776.
.
Av~diinakalpalatii 50V. 57 Tucci (1949) says here that the Buddha's foot was "pierced by a
khadira thorn". This is similar to Hofinger's interpretation and may indicate either a
different term being used in Tibetan versions of this text, or a different understanding of the
term khadiravra1)a.
Avadanakalpalatii 50 vv. 138-39.
See also Harrison, 1995, p. g.
Tatz, 1994, p. 9.
Ibid., p. 71.
Ibid., p. 76.
Ibid.
Ibid., p. 79.
Skilling, in his 1987 article, tentatively dates this work to the 12th or 13th Century CEo
Skilling, 1987, p. 3.
Skilling unpublished.
Ibid.
Somadasa, 1989, p. 123.
Neither of the manuscripts wbich I consulted, albeit superficially, in the British Library
contained references to as many.as 32 misdeeds, and the manuscripts did not agree on the
number of episodes or on their content.
Somadasa, 1989, p. 123.
Ibid.
See Hahn, 1992.
Skilling unpublished.
Bibliography
All references to Pali texts except the apadana are to the editions
published by the Pali Text Society.
Bechert, Heinz "Uber das Apadanabuch", in Wiener Zeitschrift for die
Kunde Siid-und Ostasiens Band II. Vienna: 1958.
Bechert, Heinz Bruchstilcke buddhistischer Verssammlungen 1. Die
Anavataptagatha und die Sthaviragathii. Berlin: 196I.
- - - , "On a fragment of Vim anava dana, a Canonical Buddhist
Sanskrit Work", in Buddhist Studies in Honour of1. B. Horner, ed.
1. Cousins, A. Kunst and K. R. Norman. Dordrecht: 1974.
Cleary, T The Flower Ornament Scripture. Boston: 1993.
Cousins, 1. S. "Dhammapala and the ,+lka Literature: A Review of
D'igha-atthakatha-p:ka Linatthava:r;u.mna", in Religion Vol. 2/2.
Newcastle upon Tyne: 1972.
Hahn, Michael Haribhatta and Gopadatta: Two Authors in the Succession of
Aryasura on the Rediscovery ofParts of Their }atakamalas. Tokyo:
1992.
Harrison, Paul "Reflections on the Personality of the Buddha", in
Otani GakuM, Vol. 74, NO4 Osaka: 1995
82

Hofinger, Marcel Le Congres du lac Anavatapta (Vie des Saints
Bouddhiques), Extraits du Vincrya des Mulasarviistiviidin BhaiJajyiwastu
I: ligendes des Anciens (Sthaviriivadiina). Louvain-la-Neuve: 1982.
Hofinger, Marcel Le Congres du Lac Anavatapta (Vies des Saints
Bouddhiques), Extraits du Vincrya des NIulasarviistiviidin BhaiJajyavastu
II: ligendes du Bouddha (Buddhiivadiina). Louvain-la-Neuve: 1990.
Horner, I. B. Book qf the Discipline Part 5. London: 1963.
Khoroche, Peter Once the Buddha was a Monkey. Chicago: 1989.
Lamotte, Etienne History qf Indian Buddhism. tran. Sara Webb-Boin.
Louvain-la-Neuve: 1988.
Malalasekera, G. P. Dictionary if Piili Proper Names Vols I and II.
London: 1974Masefield, Peter Udiina Commentary Vol. II. Oxford: 1995.
Mitra, R. The Sanskrit Buddhist Literature qfNepal. Calcutta: 1882.
Norman, K. R. Piili Literature. A History ofIndian Literature Vol.
VII. Weisbaden: 1983.
Reynolds, Christopher An Anthology qf Sinhalese Literature up to 1815.
London: 1970.
Skilling, Peter "The Sarp.slqi:asarp.slqi:a-Viniscaya of
Dasabalasrtmitra" in Buddhist Studies Review 4,10 London: 1987.
- - - , "Review of Hofinger, Le Congres du lac Anavatapta II" in
Journal qf the Siam Sociery Vol. 80 Pt.!. Bangkok: 1992.
- - - , "Notes on the Buddhology of the Samminyas". Draft of
article for future publication.
Smith, H. Critical Piili Dictionary, vol. !. Copenhagen: 1948.
Somadasa, K. D. Catalogue if the Hugh Nevill Collection if Sinhalese
Manuscripts in the British Library Vol. II. London: 1989.
Tatz, Mark The Skill in Means (UpiiyakaufaTya) Sutra. Delhi: 1994.
Thomas, Edward]. The Life ifBuddha as Legend and History. London:
1949.
Tucci, Guiseppe Tibetan Painted Scrolls. Rome: 1949.
Walshe, Maurice O'C. Thus Have I Heard (Translation ofthe Dfgha
Nikiiya). London: 198].
Walters,Jonathan S. "The Buddha's Bad Karma: A Problem in the
History of Theravada Buddhism" in Numen Vol. XXXVII, 1.
Leiden: 1990.
Williams, Paul j\!1ahiiyiina Buddhism. London: 1989.
The Nature of the Repetition

in the Indian Idea of Cyclical Time*
Lynn Thomas
Although the fact that the Indian model of time is cyclical is never seriously
at issue, the precise nature of the repetition implied by this pattern does
pose some problems, a point which is examined by A N. Balslev in A Study of
Time in Indian Philosophy. 1 Balslev draws attention to the fact that significantly
different ideas can be designated 'cyclical' and argues that it is important to
clarify the nature of the Indian theory. It becomes clear why this is so
important when she presents the criticisms of the Greek concept of 'circular'
time offered by St Augustine in The City of God. Faced with this model of
time from the standpoint of Christian salvation doctrine St Augustine reacts
with horror:
... as those others think, the same measures of time and the same eVents in
time are repeated in circular fashion: on the basis of this cyclical theory, it is
argued, for example, that, just as in a certain age the philosopher Plato
taught his students in the city of Athens ... , so during countless past ages, at
very prolonged yet definite intervals, the same Plato, the same city, and the
same school has existed again and again, and during countless ages to come
will exist again and again. Heaven forbid, I repeat, that we should believe
that. For Christ died once for our sins ... 2
Although this passage relates to cultural and religious milieux foreign to

India the problem raised by such a view of time is one inherent for any
soteriological system. An exactly repeating cycle must, by definition,
preclude any meaningful soteriology for eternal salvation, in any form,
necessarily demands an escape route from a cycle of time. In St Augustine's
terms, if Christ dies in an infinite repetition any salvation resulting from his
death must only be temporary; in Indian terms, if the individual's struggle
I would like to thank John Brockington ior hi, comments on this paper and Nick Allen
and Jim Benson for their comments on an earlier version.
84

,
from ignorance to emancipation must be repeated in every cycle mo~a also

becomes simply another passing and contingent state, rather than a
definitive release from such states. That this idea would be as repugnant to
Hindu religious thought as it was to St Augustine is shown by Sarp.kara's
reaction to such an idea when he argues that if assumptions are allowed as a
form of reasomng, there is nothing to stop "a malicious man from assuming
that even the released souls are to enter on a new cycle of existences". 3
Balslev argues that despite some scholars' assumptions that the Indian
idea follows the Greek one in this particular, Indian texts do not, in fact,
present such a "completely deterministic, fatalistic" model;4
. .. the world-cycles can be compared to one another in terms of generic
similarity just as Qne day resembles another, but the idea of exact repetition
involving the return of the particulars does not occur,s
Balslev's study is mainly concerned with the discussion of time in Indian
philosophical works and has little to say about its treatment in the broader
literature of the epics and Pura:r;as. As these popular texts will have done
more to inform general Hindu thinking on the subject than any others, it is
worth examining the nature of the repetitition they present to see how far it
concurs with the more sophisticated understanding of the philosophers.
The exact nature of the repetition in the cycle and its potential problems
seem to have been of relatively little concern to the epic and Purii1Jic
composers. There are very few references to the i:opic in the many passages
on time and creation and nowhere does it seem to have been discussed or
defined in the detail explored by the Greeks and their commentators. In
the Mahiibhiirata although the most common image for the endlessly
repeating cycles of time and the world is that of a wheel perpetually turning
- kiilacakraT(! jagaccakraT(! JugacakraT(! ca 6 - only a few passages touch on the
nature of the repetition involved in this. In the first of these, which occurs at
the end of the account of creation at the beginning of the epic, the cycle of
creation and destruction is likened to the turning of the seasons:
yad idaT{! driyate k.irrzcid bhiitaT{! sthavarajmigamam
puna!; saT{!/qipyatesarvaT{! jagat prapte yugak!aye (36)
yathartav rtulingani nanariipaTji paryaye
drfyante tani ta'f!)' eva tatka bhavayugadi~ (37)
eva"! etad anadyantaT{! bhiitasaT{!harakarakam
aniidinidhanaT{! loke cakraT{! saT{!parivfl,rtate (38)
Whatever is seen to exist here, standing and moving, this whole world is
destroyed again when the end oftheyuga arrives. (36)
As the manifold seasonal signs successively appear when the season
changes, so too do existences at the beginnings oftheyugas. (37)
The Nature qf the Repetition in the Indian Idea qf Cyclical Time
85
Thus is this wheel, beginningless and endless, turning in the world,

endlessly causing creation and destruction. (38) 7
This model of se~sonal change is again taken up by NilakaIl-tha's
commentary on a variant reading of 12.211.38:
riul; sa1'{lvatsaras tithya/.t [variant: t:0aM iftop-Je ca priyiipriye
yathiit'itiini paiyanti tiidrfa/.t sattvasa"f!l.k!ayab
Nrlaka7J!ha - t:0a/.t kalir itiyugiiniim upala~aT}lm
Seasons, years, tyas [taking variant], heat and cold, likes and dislikes as these reappear when they have passed, such is the case with the
destruction of entities.
Nllaka.r;.tha - tisya is ckali': by this is implied the yugas:
Another passage, however,' not included in the Critical Edition, seems
to offer a more rigid interpretation of the nature of the repetition:
[Siva speaking]
da~a dakfa na kartavyo maTI:Jur vighnam ima1'{l prati.
yajaha1'{l naharas tubhyarrz drHam etat puriitanam
Dak~a, D~a,
do not be angry about this obstruction [to your sacrifice].

In former times it was seen that I was the destroyer of your sacrifice. 8
However, although at first sight this could be taken as presenting the idea of
identical cycles, the relevant phrase - dr~tam etat puriitanam - remains
ambiguous and could simply signify that the action was ordained: it was
previously (fore)seen.
In the Puro'r;as the subject is treated in scarcely more detail but there are a
few references which again could suggest the idea of exact repetition. In the
Kurma Puro'r;a the following verse occurs twice, firstly after a passage on the
computations of time 9 and then following a description of the four yugas:10
manvantareT}l caikeT}l sarvii7!:)' eViintariiTJ.i vai vyakhyiitiini na sa1'{ldeha/.t kalpab
kalpena cazva hi .
By one manvantara all antaras are explained, have no doubt, and by one
kalpa, each kalpa.
The verb used in this verse is ...Jkhyii, with the preverbs vi and 0" meaning
'explain, fully detail, relate, tell'. It does not seem to have the sense of
'describe' and could simply refer to the durations of these periods rather
than their characteristic natures. This is certainly the more natural reading
of the verse following the computations of time.
A set of verses which are obviously related to each other occurs in the
Manusmrti, Miirkar;eya Puro'r;a andVimu PurO,r;a, all following accounts of
Brahma creating the inhabitants of the world. 11 The following is the Vt,l7JU
Puriir;a account:
86

tat sasarja tado. brahmo. bhagavo.n o.dikrd vibhu~
te~o.'l'fl ye yo.ni karmo.1J.i pro.k sr~tyo.'l'fl pratipedire (s9)
tiiTlj eva te prapadyante sriYamo.no.~ puna/;! pun~
hi7J1STo.hi'l'flsre mrdukriire dharmo.dharmo.v rto.nrte
tadbhiivito.~ prapadyante tasmiit tat tasya rocate '" (60)
yatharto.v rtulingo.ni no.niiriipo.1J.i paryaye
drfyante to.ni to.ny eva tathii bhiivo.yugiidU (64)
karoty eva'l'flvidhii'l'fl sr~ti'l'fl kalpo.dau sa pun~ puna~
sisrk~o.saktiyukto 'sau srjyasaktiPracodita~ (65) 12
Then the blessed creator, Lord Brahma, created it all. Creatures who
performed the actions/received back the karmas which were theirs in'the
previous creation, (59)
Being createdagain and again, return to precisely those same
actions/karmas. Cruelty or non-injury, gentleness or fierceness, dharma or
adharma, right or wrong - produced from that, they return to it; therefore they delight in it ... (60)
Just as the manifold seasonal signs are seen when the season changes, so
too are these conditions seen at the beginnings oftheyugas.(64) 13
Wlison, Buhler and Pargiter all translate the crucial V.59 and its equivalents in such a way that the idea of identical repetition is emphasised:
Wilson [Vi;>ryu Y.59]: ''And these things ... discharged the same functions as
they had fulfilled in a previous creation ... " 14
Buhler [Manu v.28]: "But to whatever course' of action the Lord at first
appointed each (kind of beings), that alone it has spontaneously adopted in
each succeeding creation."15
Pargiter [Miirk v.39]: "Whatever actions they were severally endowed with
originally at their creation, those very actions they are endowed with when
they are created again and again." 16
However, once again a closer examination shows that the ambiguous
use of the word karma, which can designate either action in general or,
more specifically, those actions which form and condition future states,
means that an exact repetition of previous actions in future kalpas is not the
only possible interpretation. Certainly, it is not an interpretation that
occurs to Medhatithi in his Manubhii~a. Not only are these verses not seen
to espouse identical repetition, this is not even one of the possible erroneous
interpretations he considers. Instead he takes karma in the sense of the
accumulation of actions and their results, so that beings are allocated their
place at the start of the present creation in accordance with their actions in
the last, a quite different emphasis and one which presents no problems for
Hindu soteriology. For Medhatithi, in these verses Brahma creates beings:
The Nature if the Repetition in the Indian Idea
if Cyclical Time
87
,., not without reference to the actions done by them during the preceding
cycles; he makes the creature born in that family of creatures which is
indicated by the act done by it during the previous cycle ". if the creature
has, in the past, done a good act, it is led to be born in a family in which it
would be enabled to experience the good results of that act ,,' 'What
happens is that at the beginning of each new creation, the acts done by
creatures in the previous cycle come out, after having, during Dissolution,
lain latent within their source,. , 17
An earlier verse in the Mark.a(lrleya Pura(la account reinforces Medhatithi's

interpretation and makes it clear that karma is being used in the specific
sense of past actions conditioning future states;
Creation is impregnated with the good and bad actions of previous existence
- pi1rDakarmabhi~ -, 0 brahman; and because of this well-known law, created
beings, though they are destroyed in the dissolution, are not delivered from
the consequences of their actions. IS
While Medhatithi's jl1anubhiiva is obviously a later interpretation of the
text, the fact that the tradition has interpreted these potentially ambigious
verses in a way that does not necessitate the idea of an exact repetition,
serves to reinforce Balslev's argument that this is not the Indian model.
Certainly, if exact repetition were being presented in these passages it could
net co-exist with any system of Indian soteriology. That this is not the
intention can be supported most obviously by the fact that a soteriological
escape from the cycle is deliberately written into the theory of time and,
according to Biardeau, largely makes up its purpose in the fate of liberated
beings at the time of the pralaya who, having attained the highest state, are
finally released to union with brahman. '9 Other aspects of the yuga theory
also directly contradict the idea of an identical repetition. For example,
although Brahma is born at the beginning of each cycle and performs the
same functions at this time, the details of his birth are different in each
kalpa. Similarly, while the fourteen manvantara periods which subdivide the
cycle always occur, they are presided over by different Manus, saptarris and
gods each time. 20
Such a co-existence of the same events with different particulars suggests
that it is most likely that the repetition was understood on the model of the
image most abidingly used, namely, the turning of the seasons, an image
found in the lVlahiibhiirata, HarivaT(lsa, and Manusmrti, as well as at least the
ViJ(lu and MiirkaTJrleya PuraTJasY Such an image of time, while making the
units correspond in the overall pattern of occurrences, does not necessitate
an exact repetition of events: just as every spring plants come to life and
every autumn they die, but not the same plants, so too at the beginning of
every cycle everits are set in motion and clifierent categories of beings are
created to play them out, but not the same beings. As the manvantara theory
88
and many dialogues in the }'1ahabharata show, even the lesser gods are not
presented as individuals but rather as roles to be filled by differen(players
at different times. 22 This would con.firm the model of 'generic similarity'
that Balslev argues for and which seems much more likely to lie behind the
Indian model than an idea of circular repetition.
Notes
I.
2,
3,
4,
5
6,
7
8,
9,
!D,
II,
12.
'3,
'4
'5,
16.
'7,
18.
'9,
20,
21.
22,
Balsley, 1983, pp, '40-50.

Augustine, City of God, Book XII, chapter XIv; cited Balsley, Ibid" pp, '4'-2,
Sarpkara, Vediinta Siltra (1.2,17), cited Balsley, Ibid" p. '47.
Balsley, Ibid., p, '49,
Ibid" p. '46,
5,66,12, This and all Mahiibhiirata references are to Sukthankar, er. al. (eds,), 1933 [1970].
I.1.36-g 8,
12,App,L28,397-8.
KUnnaPurii1J1l 1.5.14, This and all Kiinna Purii1J1l references are to Gupta, (ed,), '972.
KUnna Purii1J1l 1.28.52,
Manusmrti 1.28-30; Miirka~rirya Puriir;a 48'39-44; Vi,rr.ru Purii1J1l 1.5,59-60,64-65, Mahiibharata
12,224.47,48 has the same wording as lWiirkar;r!rya Purii1J1l 48.39,40 but does not go on to give
'
the analogy with the seasons.
Verse 64 is the same as lWahiiblziirata 1.1.37 discussed above.
Vi,rr.ru Purii1J1l 1.5,59-60,64, This and allV!J(lU Purii1J1l references are to Vidyasagara, 1882.
Wilson, 1840, p, 43
BUhler, 1967, p. 13, translating:ya1)l tu kannaniyasmin sa nyayuilkta prathama1)l praMu!l/ sa tad eva
svaya1)l blliBe srjyamiina& puna& punail.
Pargiter, '904, p. 236, translating: te,iiT[Z ye yiini karmii,!i PTiik S{{!e& pratipedire/ tany eva
pratipadyante srjyamiina& puna& puna&.
Jha, '9 20, p, 67
Miirkar;r!rya Purii1J1l48,2, translated by Pargiter (1904) p, 235,
See ViJ1!U Purii1J1l 6,3; Garua Puriir;a 233, Biardeau discusses the significance of the fate of
beings during the pralaya in great detail, and it plays an important part in her interpretation
of the structuring of time in the Puriir;as: see IvL Biardeau, 1971.
See, for example, MiiTkar;rirya Purii1J1l46,32,33
Mahiibhiirata 1.1.37; Manusmrti 1.30; V!J(lU Puriir;a 1.5,64; Markar;r!rya Puriir;a48,44 (all'discussed
above); HarivarrzJa 7.54.142* lines '3,14.
See, for example, the story of the five Indras in Mahablziirata r.r89,
Bibliography
Balslev, A, A Stucfy of Time in Indian Philosophy. Wiesbaden: 1983
Biardeau, M., "Etudes de Mythologie hindoue: 3", Bulletin de l'Ecole
d'Extreme-Orient, LVIII, 1971
Buhler, G" (trans.), The Laws ofManu. Delhi: 1967.
Gupta, AS., (ed.), Kurma PuriilJa. Varanasi: 197'2.
Jha, G., (ed.), Manu-smrti: The Laws ofManu with the Bhii~ya of
Medhiitithi. Calcutta: 19'20.
Pargiter, F., (trans.), MiirkalJp.rya PuriilJa .Calcutta: 1904.
The Nature qf the Repetition in the Indian Idea qf Cyclical Time

Sukthankar, V. S., et. al., (eds.), Mahiibharata: Critical Edition. Poona:
I933 [I97 0 J.
Vidyasagara,j., (ed.), Virr!-upurii7Ja. Calcutta: 1882.
Wilson, H. H.) (trans.), Vi~7Ju Purii7Ja. London: I840.
89
Vi~l)U and the Dance of Siva
David Smith, Lancaster University

A stri1,:ing feature of Nataraja's temple in Cidambaram is the right-angled
conjunction of the shrines of Siva (Nataraja) and Vi~I.lu (Govindaraja). At
one point, marked as a lotus on the stone floor, the worshipper can - with a
flickering of the gaze to and fro - view both the south-facing King of
Dancers and the east-facing King Cowherd. This joint presence of the two
major deities has often been remarked upon. B. Natarajan, who in his
recent Tillai and Nataraja devotes a chapter to the Govindaraja shrine, is no
doubt correct when he says that Cidambaram is "the most conspicuous
example of a religious centre where Siva and Vi~I.lu co-exist". 1 This paper
considers the presence ofVi~l}u near Dancing Siva and examines the Saiva
attitude to Vi~I.lu in Cidambaram during the twelfth and thirteenth
centuries.
The conjunction of the two shrines has given rise to various theological
speculations, as is only natural. Thus T G. Aravamuthan sets out to show:
... that, in its origin as a great religious centre, it was marked by the conviction of the Oneness of God-head. Chidambaram was, in fact, a place
where the Unity of God-head is averred and is sustained. 2
The Chidambaram Temple, with both Narayal).a and Nata-raja coalesced
into One God, though shown separated as two, is fine proof of the continuance to these days of the spirit of integration of Gods into a Tad Ekam,
'That One', which is the distinguishing mark of the J.?-g- Veda. Man's desires
kept him appealing to this God or that, and kept him making offerings as it
pleased him for the moment. None-the-Iess, deep in him lay, and lies the
conviction that all the Gods appealed to are but facets of one God-hood.
And as we find at Chidambaram, the devotee was ever taught to exhibit his
abiding belief in that Unity.'
S. Sankaranarayanan argues that:
The Vedic scholars of Chidambaram seem to have felt that this symbol,
N ateSa, is based more on the Rgvedic theme of the creation of the universe
92
Indian Insights: Buddhism, Brahmanisrn and Bhakti

from the raised foot of the Supreme and that it does not fully answer to the
Upani~adic specification of the reclining Akasa (Jete) as the cause of the
Universe. So those Vedists seem to have strongly felt the need for an
additional symbol to represent this reclining Akiifa Atman ... the Akasa could
be represented only in an indirect way by means of what is indicated by the .
expression virr;upada. This expression, being a homonym, can denote ... also
the very Viwu Himself, because He has the name (pada, i.e., nama-pada)
Vzrr;u. Further, among all gods He alone is known to have reclining posture
as Se~asayi. So the Vedists decided to have a Se~asayi Vi~r;tu figure for a
symbol they required. 4
These theological musings must, however, be reviewed in the light of the

historical assessment which I shall attempt after having first given an
account of the shrine as it exists today. The Cidambaram temple is huge,
covering some 33 acres, including the gardens that are contained within its
outermost walls. The first and innermost enclosure contains the supreme
shrine of Dancing Siva, the Hall of Consciousness, the Cit Sabha, facing
south, with its distinctive curved roof covered in gold leaf. lminediately in
front of this is an almost identical structure, wherein worship ofNataraja
and of a small, portable, crystallinga is performed. A courtyard with minor
shrines surrounds the twin halls to the west, north, and east. The south-west
part of the enclosure is taken up by the shrine ofVi~IJ.u. Vi~IJ.u lies on Se~a,
and faces north, towards Nataraja. A separate entrance for the Vi~IJ.u shrine
is provided in the east wall of the enclosure, with a flagpole just outside. The
rectangular garbhagrha' of the shrine has a barrel roof, and has an
ardhamar;{lapa and a prada~r;a patha (cir~umambulation passage). The sanctity of the site and the legal wrangles over the centuries make investigation
difficult, but the present buildings of the Vi~IJ.u shrine are plainly not old. 5
Natarajan gives the following description of the contents of the shrine:
Within the Govindaraja shrine, we have sub-shrines dedicated to
Hirar;tyanarasi.rp.ha, Sudarsana, Ver;tugopala, Yoganarasi.rp.ha, Hanuman,
Vi~vaksena, the Ramapadas, Ramanuja and Kurattalvan. In a huge,
impressive, solemn stone sculpture, the presiding deity, Govi~daraja, is
shown reclining on His serpent-couch, the seven-headed A.dise~a, in
yoga-nidra, head to the south and with Brahma seated on a lotus emanating
from ;His navel, and looking northwards. The utsava icons made of metal, of
Srlnivasa ~th SridevI and BhudevI, are located in front of Govindaraja.
Around them are located images of Parthasarathi, A.r;ttal, Rukmir;tI,
Satyabhama, and the utsava vigrahas of Nammalvar, Kulacekaranalvar,
Tirumailkai Alvar, Ramanuja and Mar;tavala rna Muni. 6
As already noted, this shrine faces east, and is at a right angle to the
Nataraja shrine. Soma Sethu Dikshithar in his Guide to the temple refers
to the spot on the floor marked by the lotus:
Vi~u and the Dance of Siva
93
This is the place where one can see both the deities at the same time .... It
is said by elders that one should see Lord Govindaraja through the left eye
and Lord Nataraja through the right. The significance of this is that the left
side of Lord Siva is the part ofVi~I;lU who is embodiment ofSivasakti. 7
Here again there is a theological gloss on the co-presence of the two
divinities. But discussion of the innermost courtyard of Nataraj a's temple is
not complete without remarking on what is absent. This enclosure does not
contain the 'Original Shrine' of Cidambaram, the linga shrine, the
Mulanatha. This is situated in the second enclosure. Also in the second
enclosure is the oldest structure in the temple, in terms of architectural
style, namely the Nrtta Sabha (though the plinths of the twin Sabhas might
be as old). The Nrtta Sabha, built no later than the reign of Kulottunga III
(II78-1216), is a pavilion on a high plinth that contains a shrine to
Drdhvatal).Q.ava Siva, and also a shrine to Sarabhda Siva. Both shrines
seem to be additions to the original structure, the arrangement of the
pillars of which seems to be designed to allow space for a single dancer in
the central area. The structure is provided with a wheel and a horse on its
east and west sides at the south end, and is thus a chariot, being drawn
towards the Cit Sabha. This detail of the structure is easily forgotten today,
since almost all of horse and wheel is now covered over with concrete. The
supposed movement of this shrine is all the less likely because of the
obstruction of the wall of the first enclosure. But the essential point to be
made is that both these shrines are intimately connected with Nataraja but
are not in the first enclosure. Vi~l).u, as Govindaraja, is in the first
enclosure; and we have seen three brief theological interpretations of this.
Let us now consider the textual history of the Vi~l).u shrine.
The earliest reference to Vi~l).u in Cidambaram is found in Tirumankai
Alvar (second half of eighth century AD); both he and Kulacekaral]. Alvar
(ninth century) refer to the "three thousand" priests, which would seem to
mean that the Vi~l).u shrine was under the control of the Saiva Dik~itas,
who have always been known as the "three thousand". Tirumankai Alvar
encourages people to go Citrakuta in Tillai (Periyatirumoli III. 2) and deeds
of various avatiiras are alluded to; and the next song refers to the childhood
adventures of Kql).a, who is said to be "inside Citrakutam of Tillai
(Periyatirumoli III. 3).8 Kulacekaral]. dedicates Perumii!-Tirumoli X to Rama
in "Citrakuta of the town of Tillai", where Rama is seated on a throne
with beautiful jewels (X. 2).9
Next in point of time is the Tirukkovaiyiir, a complex and difficult text,
attributed to Mal).ikkavacakar (ninth or tenth century):
To seek the feet of Him who destroyed the three fortresses
(as a boar) bored the earth and when stymied
Surfaced up and prayed thus: '0 my Father, grant me grace.'
He that endowed me with twin hands to adore His feet twain,
Vi~I;lU
94

Thereupon relented a little and showed him one ofRis feet.
(Thus blessed) he continues to beseech Rim at Tillai's court
For the boon of revelation of the other foot also. 10
R Natarajan takes the standard view when he says of this verse that it is
"the poetic imagery of Saivite poet reflecting the fact that these two Lords.
of the Trinity dwelt together in the Hall of Tillai" (p. 27). However, none of
the passages so far considered explicitly refers to a Vi~l).u shrine beside
Nataraja. Vi~l).u is described as lying prostrate before Siva's shrine. This
follows very naturally from the reference to Vi~l).u as Boar seeking the base
of the Ziriga, a common motif on the outer western wall of the garbhagrha in
Siva temples.
The next reference to Vi~l).u comes a century or two later, in the long and
informative inscription that details the works done in Cidambaram by
Naralokavira, in the reign of Kulottunga I (I070-II20). Naralokavira is
usually said to have been an "officer of Kulottunga I", but is perhaps more
likely to have been a semi-autonomous war-lord. Verse 23 of 120 of 1888
says that Naralokavira consecrated an image of Nandisa, Nandisa whose
permission had to be sought by Vi~l).u (Vaikur:ttha) and the other gods each
day as they stood outside before they could come in and bow down before
Siva. The absence of any reference to a shrine of Vi~l).u in this detailed
account of the temple is striking.
Onakkuttan's poems in praise of Kulottunga II (II33-II50), and the
same poet's Takkayiikaparar.zi, all refer to that monarch's removal of Vi~l).u
from Tillai and dropping the deity in the ocean, his original home. This is
generally assumed to have been a hostile act; and yet immersion of images
in the ocean is a regular form of devotion. N aralokavira built a ma7Jrlapa on
the sea-shore for that very purpose for Nataraja (verse 6, 120 of 1888).11
Also during the reign of Kulottunga II was written the sthalapuriir.za of
Cidambaram, the Cidambara Miihiitmya. 12 Here Vi~l).u plays a prominent
part, though completely subordinate, in events leading to the first
performance of Siva's Dance of Bliss, the iinanda-tiir.zrlava. In a sophisticated
narrative device, Vi~l).u in the Milk Ocean first calls out to Siva while in a
state of trance; and Se~a, the snake who forms his couch, feels his master
become very heavy. It transpires that this is because Vi~l).u is remembering
the events earlier in the day. He had gone as usual to Kailasa to pay his
respects to Siva, and was told by Siva to take on the form of a woman, for
they were going to the Devadaru Forest to assess the sages there. Siva
himself would go as the Divine Beggar, Bhik~atana. The sthaZapuriir.za refers to
standard Purar:tic myth when Siva arouses the desire of the sages' wives, but
is decidedly innovatory in having him accompanied by Vi~l).u as Mohim.
When the sages attack Siva with creatures from their sacrificial fire, he
defeats these new opponents and assimilates them into his iconography. He
then dances the Dance of Bliss, with the sages and Vi~l).u as his audience.
Vz:,(Zu and the Dance of Siva
95
Vi 0l}u also features in an earlier part of the Cidambara Miilziitmya, when

he is said to have brought the Milk Ocean to Cidambaram. VVhen all the
gods come to Cidambaram to see Siva's Dance, Vi~I).u brings his Milk
Ocean, and making it his couch, witnesses the Dance of Bliss (7. 3I-2). But
this Vai~l}ava presence is equalled by Siva supplying an ocean ofmiL\ for
Upamanyu, Vyaghrapada's son, in a tank to the north of the temple (a
tank recently filled in to make a school playground).
I pass over here later legends that the Vi~l}u image was, after its removal
by Kulottunga II, taken by Vai~l}avas to Lower Tirupati a.TJ.d there consecrated by Ramanuja, and consider next the explicit reference to the Vi~l}u
image made by Umapati in his Kuiicitiinghristava, around I300.13
Even today to His right
in his jewelled shrine
Vi~I).u
is lying on the couch of the coils

of the five-headed Lord of Snakes
with his feet to the south and his face to the north.
Although he is always asleep,
he constantly meditates
on His praiseworthy foot.
(Kuficitiirighristava 42)14
Here at last is an explicit reference to the image of Vi~l}u in the

Nataraja temple and to the existence of the shrine (sadana); Umapati's
poem gives a very full acount of the temple. The verse clearly shows that
worship of Vigm did not, as Kulke and many others have claimed, cease
after Kulottunga for a long period. A century and a half after Kulottunga
it was alive and well. A minor difference is that Se~a is said to have five
heads, whereas today it has seven heads. VVhat is particularly noteworthy
is that Umapati says that the image has its face to the north and its feet to
the south. This is the reverse of the orientation of the image today; it also
corresponds, incidentally, in orientation at least, with the reference in the
Tirukkovaiyiir to Vi~I).u lying prostrate before Siva.
Umapati makes one other explicit reference to Vi~l}u's shrine (geha) ,
along with four other independent shrines, these others being in the third
enclosure:
the shrines
of the Blessed Mnlasthana,
Devl, Vi~I).u, Elephant-faced Vinayaka, and Skanda.
(Kuficitiirighristava 76)
In modern times 'the five shrines' are frequently spoken of by the

priests, but the Vi~l}u shrine is no concern of the Saivas, and the Cit
Sabha is included to make up the five. 's
96
Indian Insights: Buddhism, Brahmanilm and Bhakti
The first inscriptional evidence of a Vi~l).u shrine in the N ataraja temple

does not come until 1539. Inscription '27'2 of 1913 tells that Acyutadevaraya
had ordered the consecration of the Tillai Govindaraja in Cidambaram,
according to the rituals of the Vaikhiinasa Sutra, with the income of four
villages for the upkeep of daily worship. It is after this consecration that
Appayya Dik~ita is popularly supposed to have written his hymn to
Harihara, the Hariharastuti. These eight or ten verses are sometimes said to
be called Harihariibhedastuti, but in fact this is a series of epithets, one to
Vi~l).u followed in each case by one to Siva, with the weight of emphasis
falling on the second and Saiva epithet. In the eight verse version, the only
one available to me/ 6 there is no mention of Cidambaram or Nataraja,
and it may well be that later tradition invented the connection with the two
shrines oftoday.
Evidence of very serious disagreement between Saivas and Vai~l).avas in
Cidambaram is furnished in 1599 by Pimenta, a Jesuit travelling through
South India:
In our way wee saw Cidambaram the mother Citie of their Superstitions,
furnished with gorgeous Temples. Their Brachmanes have thirties thousand
Duckets revenue, whereof but twelve thousand are now payed.
The N aichus of Gingi was come thither, in whose Dominion it standeth ...
The Next day wee went away, which we had not done, had any told us of a
strange Spectacle that day there to bee seene, which wee after came. certainly
to know.
There were twentie Priests which they caliJogues, which threw themselves
from the highest pinnacle of the Temple for this cause. There is a Temple of
Perimal, in which is worshipped an Ape called Hanimant, whom they report
to have been a God ... Now at this time was a great Controversie amongst
these Gentiles, whether it were lawfuli to place the Signe ofPerimal (which is
nothing but a Mast or Pole gilded, with an Ape at the foot) in the Temple of
Cidambaram. Some refused, others by their Legats importunately urged,
and the Naichus of Gingi Decreed to erect it in the Temple, the Priests of the
Temple which were the Treasurers, withstanding, and threatning if it were
done to caste downe themselves from the top. The Brachmanes of the
Temple sware to doe the like after they had buried the former, which yet
after better advise they performed not. About twentie had perished in that
precipitation on that day of our departure; whereat the Naichus angrie,
caused his Gunners to shoot at the rest, which killed two of them, the rest
wandering in uncertaine places. A Woman also was so hote in this zealous
quarrell that shee cut her owne throat. The Mast with the Ape was
nevertheless erected. 17
Today, the subsidiary status of the Govindaraja shrine is revealed by
the fact that it is not allowed to raise a flag on its flagpole. I shall return to
Pimenta's account in my conclusion.
Vi,s(zu and the Dance of Siva
97
After this review of the textual evidence relating directly to ViglU in

Cidambaram up to the end of the sixteenth century, which shows Umapati
(c. 13(0) to have been the first to make unmistakable mention of the shrine,
I turn now to the general iconography of the temple, in relation to ViglU.
A very clear statement of the relative importance of deities is made by the
gopuras of the temple gateway, three of which were completed in the
thirteenth century. All four share a common sculptural programme. The
first of the two stories in stone presents a variety of figures, one of whom is
ViglU; the second storey is composed of a variety of forms of Siva. Each
gopura has on its inner facade a sculpture of Vi~r;tu mounted on a human
two-armed Garuc;la. Vi~r;tu's rear hands hold the discus and conch, the front
hands are in afijali and plainly manifest his devotion as he looks inwards to
Nataraja in the centre of the temple. Each of the gopuras also has Siva
appearing from within the liriga with Brahma and Vi~r;tu searching for its
limits. Moreover, when Siva is shown as Urdhvatar;.<;lava, the figures on the
Western and Eastern gopuras have an accompanying drummer who wears a
crown and might therefore be Vi~r;tu (the Kiiflczmiih1itmya story ofVi~r;tu
becoming Nataraja's drummer is discussed below). Even if the drummer is
not ViglU, the gopuras could hardly state the subordinate status of Vi~r;tu
more plainly. Vi~r;tu is "dearly shown as drummer at the right hand side of
the Urdhvatar;.<;lavamurti in the Nrtta Sabha, with Kau at the left. These are
the only Vi~r;tu images outside the V~r;tU shrine, and are clearly older than
that shrine.
The final evidence I wish to present in this consideration of Vi~r;tu in
Cidambaram is the general attitude to Vi~r;tu expressed by Umapati in the
KuflcitiiiLghristava. This poem is a personal statement by a notable priest of
the temple, and may fairly be said to express the world-view of
Cidambaram Saivas around 1300. We have already considered Umapati's
description of the Govindaraja shrine. Shortly before this, in his poem
Umapati refers to the principal event of the sthala-purii1J-a:
In the beginning Srup.bhu the Mild along with Mohinl
was strolling in the dense wood called Daruka.
Deluding by His beauty the wives of the Brahmans,
and that throng of silent sages by Vilir;tu's illusion,
nullifying the magic attack the Brahmans staged,
He took pity on the Brahmans, and displayed His Dance.
Him, the Dweller on the mountain,
Whose foot is curved,
I worship.
(Kuficitiiilghristava 40)18
That Siva should go as Bhik~atana is no illusion, for that is one of his

major forms and, it should be mentioned, one of the two most prominent
representations of Siva on the gopuras. Vi~r;tu's form as voluptuous woman
is, however, not merely an instance of that God's power of illusion,
98
Indian Insights: Buddhism, Brahmanism and Bhakt:i
but also an instance of the literary powers of the creator of the sthala-pura1J1l.
The utter passivity of Vi~r:Iu's role allows his presence to be emphasized
without giving any scope for claims ofhis importance in Cidambaram.
Vi9r:IU's femininity is again referred to later in the poem:
His face all the more like a night lotus come alive
from the stream of clear nectar from the moon on His crown,
delightful with His unrestrained song.
In the forest He held immobile
the throng of the sages' wives,
He, the complete embodiment of the erotic,
Seeing Him, V~1:lU'S illusory form
determined to marry Him.
(Kuncitanghristava
130)19
The preceding verse subordinates Vi9r:IU to an external female: alluding

to the Lalitopakhyana, Umapati posits the Goddess's fingernails as the
source ofVi9r:Iu's avataras:
Creating the avatara portions ofV~I).u
on the battlefield with her ten nails
With His missile she destroyed the Daitya BhaI).<;ia
along with his city and the host ofhis sons.
Sh.e it is who resides in the Sricakra.
Him who performs auspicious deeds,
Whose foot is curved,
(Kuncitanghristava
I worship.
129)20
Both these verses are from a stunning sequence of verses about the
forms ofVi9r:Iu, in most of which Vi9r:IU is put down by Siva:
When V~I).U had assumed the form of a fish
and slain the demon who'd taken the Vedas
down into the world of snakes,
the gods were terrified to see him,
big as a mighty mountain,
put out the flames
of the .fire in the ocean
with a blow of his tail.
The chief gods praised Him,
and~as the Skull-bearer, Kapalin,
He struck him down
a~d put out his eyes.
(Kuncitiinghristava
122)21
Vimu and the Dance
of Siva
When ViglU took on the excellent body of a turtle

he became terrifying, he swallowed the ocean;
his mouth was ablaze with flames like the kiilakilfa poison,
he shook the whole world in what he did;
he could not be conquered by Brahma
nor any other god.
It was He who dragged him ashore
and made his limbs His necklace.
(Kuficitiiitghristava
I23)22
vVhen the earth was thrown in the sea

Vaikul}.j:ha Vi~l}.u assumed the form of a great pig,
quickly raised up the earth on his tusk
and slew the demon Hiral}.ya in battle.
But when he carried on roaming about
and acting in a way offensive to all
Bhairava with His trident's sharp prongs
sending forth flames and ash
extracted his tusk
and wore it on His chest.
(Kuficitiiitghristava 124)23
All the gods trembled in fear ofVi~l}.u
when he took on the body of the Man-lion,
so Brahma sang His praises
and He took on the excellent body of Sarabha, the king of birds.
Speedily He cut him up with the tips of His claws
and adorned Himself with the skin,
illuminating the world with His fangs.
He who grants the wishes of all,
vVhose foot is curved,
(Kuficitiiitghristava 125)24
I worship.
Vi~I).U
as the Dwarf, shrank his body

to make his petition in Bali's hall of sacrifice,
and made his body grow from earth to sky
and placed his foot - itself the triple world on the demon's crown.
But then he terrified the whole world with his roaring.

The hosts of gods praised Him as He slew the Dwarf
and ripped out his skeleton and tore offhis skin.
(Kuficitiiitghristava 12 6j2S
99
100
hdian Insights: Buddhism, Brahmanism and Bhakti

Rama, lord of ascetics, son ofJamadagni,
desiring to clear the sin of his mother's fault
and needing a mighty axe, excessively sharp,
to cut off the wings of the mountains the warriors who slew his father,
meditated constantly on this Dance of His
that is witnessed in every individual self
Rama gained his excellent wishes, long life and the others.
(Kuiicitiiilghristava 131)26
Balarama, Rohini's son,
the fan-palm his insignia,
had his aged wife turned into a young woman
by the sacred ash obtained from His worship;
by much pilgrimage to His holy places
the misfortunes arising from the sin
of murdering the excellent ascetic
were allayed;
and by His grace
he cut up the wicked Pralamba
with his excellent mighty plough.
(Kuiicitiiilghristava
132)27
At the end of every Kali age Vi~l).u Mukunda

in the form of Kalkin mounts his horse
and rides on earth,
slaying, with the sword he'd got from His pure mercy,
the host of heretics no bigger than a thumb,
so evil-minded and sinful are they by this time,
reforming the whole world to its proper condition.
(Kuiicitiiilghristava 133)28
This list of avatiiras either eliminated by Siva or owing everything to him

is followed up by a Tamilnadu horror story, supposed to have happened in
Tiruvaiyaru, about ten miles from Tanjavur:
One of His devotees, an excellent Brahman,
in the middle of his pilgrimage
was eaten by multi-headed Madhava Vi~l).usuper-parrot-headed,
lion -, tiger-, horse-headed,
elephant-, bear-, eagle-headed,
monkey-headed,
man-headed-
VipJu and the Dance if Siva
WI
and in this fury Vi~I).u went on,

threatening time and again all the worlds.
But in vain, for soon He slew him
and saved His devotee
and the whole universe too.
(Kuficitiiizghristava 134)29
The Pan-Indian tradition is surpassed by this cumulative nightmare
form of Vi:;;r;m. Such power as Vi~DU has, is, to put it mildly, unstable.
Even when considered in his most permanent form, his most distinctive
emblem, his discus, has its origin in Siva:
In the beginning VaikuI).j:ha Vi~I).u
was constantly worshipping Pasupati
with a thousand flawless lotuses.
Once the last lotus could not be seen.
Then plucking out his own lotus eye
he placed it at His pure feet.
From Him he received his discus
and became splendid.
That Lord of Dancers, Whose foot is curved,
(Kuficitiiizghristava 56)"0
I worship.
Vi~DU'S discus is also mentioned when Umapati refers to the Madurai
cycle of myths .. The PaD9yas with their Madurai connections succeeded
the Colas, and this is marked in Cidambaram around the time of
Umapati by the construction of the small shrine dedicated to Minak~i and
Sundaresvara, south of the Sivakamasundari shrine on the western side of
the temple site. Umapatirefers to the Madurai story of how Siva assuaged
the enormous hunger and thirst of the leader of his gaJ)as. To allay
KUD90dara's thirst Siva brings not only the milk ocean but all seven
oceans. Umapati also refers to KUD90dara's theft ofVi~DU'S discus, so that
Vi~Du'S weapon is marginalised as well as his milk ocean!
The king of the circle of Millas

Whose strength is famous
and who stole Vi~I).u's discus,
had the torment of his hunger and thirst absolved by Him
-a deed which made motionless the daughter of the PaI).<;iyan
-when He brought the seven oceans
joyfully in His own fingers
to accomplish His mother-in-law's wish;
(KwTcitiiizghristava 249)31
Another local myth, from TiruvarD.r, the next most eminent Saiva
temple in Cola times, refers to Vi~DU and the Milk Ocean, and Vi~Du'S
102
recitation of the Nlucukundasahasraniima:

To destroy the demons and protect the whole world
Mura's foe, in the ocean of miL~
along with Lak~mi
performed difficult penance
and with devotion worshipped in his heart
the Lord with Ambika beside Him.
Vi~l).u,
He obtained the excellent mantra

that is the hymn called His Thousand Names
and by reciting it
became the Lord of the World.
(Kuficitailghristava 136)32
The tank at Tiruvarur, considered to be biggest tank in South India, is

called Kamalalaya, and Kamala (Lak~rni) is said to have been born from
it, and there to have married Vi~T,1u. Another important site, and where
the Lalitopiikhyiina mentioned above seems to have originated, is
Kancipuram. When discussing the gopuras, I mentioned above the possibility that the crowned drummer figure beside DrdhvataT,1<;lava might be
Vi~T,1u. The following verse, and the Kiificzmiihiitmya passage on which it is
based, justifY that assertion.
o Lotus Lal<l?mi, don't let's play today

I don't want my couch, the lord of snakes,
nor Garuc;ia.
o Earth, do you come at once,
for I am going to beat the drum
in Tillavana with great skill
to accompany the dance of the Foe of the Cities.
So saying, in days of yore Hari went to His stage,
Him who is unborn, 'Whose foot is curved,
I worship.
(Kuficitailghristava 230)33
The account of the Kiificzmahiitmya is as follows. One day as Vi~T,1u was

practising yoga, he noticed in his heart a flame in the middle of which Siva
was dancing. Overcome with wonder, he plunged into trance. Noticing this
inert condition of Vi~T,1u, his wife Lak~rni, Garu<;la and AnantCl-, who were
beside him, became anxious, without knowing exactly what had happened
to their lord. Mter a few moments he awoke and sang of the magnificence
of Siva. Questioned by his wife, he told her that he had seen the Supreme
Being (Siva) dance in his heart and that his dance had fascinated him
(151-66). His questioners begged him to allow them to see this dance. Vi~T,1u
brought them to Cidambaram and gave himself up to penance in respect of
Siva. Siva appeared and said to him, "If your people desire to see my dance,
first go with them to Kanci, where you will install and adore a linga.
Vi,l(lu and the Dance qf Siva
103
Afterwards return here, then we will fulfill your vow"(167-175). Siva

thereupon went to KanCI, accompanied by all his entourage and installed a
lingam in the south of Ekampam '" . Then he returned to Tillai
(Cidambaram) (176-178). Siva then began to dance.Lak~mI, the king of
birds (Garu<;1a) and the king of serpents (Adise~a) saw him and meditated on
him (179). The Gar;tas made use of musical instruments while he danced.
Vi~r;tu wished to do so too. He played the kettledrum. Afterwards he asked
Siva to grant him the boon of playing this instrument in all his dances
(180-182).34
In the list of avatiiras discussed above, Balarama is present rather than
Kr~r;ta. What does Umapati have to say ofKr~r;ta? Kr~r;ta is initiated in Saiva
worship by Upamanyu (Kuncitiinghristava 53); at the kalpa's end Kr~r;ta,
floating on a peepulleaf, thinks that there is no one but himself, but is then
confronted by Bhairava with his trident (Kuncitiinghristava 136). The
distinctive name ofVi~r;tu in Cidambaram, Govindaraja, is a problem that is
not discussed in this paper; but the name Govinda is used twice by Umapati
in his poem. In Kuncitiinghristava 36 it is said of Siva that his foot was sought
by Govinda (govindiinvefitiinghn), alluding both to Vi~l).u as Boar seeking the
base of the linga and to Vi~r;tu as Govindaraja as foremost devotee of
Nataraja. hI the other instance, earlier incarnations ofVi~r;tu, waiting upon
. Siva on Mount Kailasa, see Kr~r;ta coming to seek a boon for the gods:
On the best of mountains,
the lotus-eyed Vi~I).us,
who had each in turn
relinquished their authority,
greeted Govinda,
who had come to obtain the happiness of the gods,
and after embracing him with their hands in delight,
and asking how he was, .
they enlightened him as to Him:
."Dear one, it's because you don't
comprehend the Essence of Brahma in your heart,
that you ask for action action is the cause of suffering.!"
(Kuficitii:izghristava 215)35
We have now reviewed the principal references to Vi~r;tu in Umapati's

poem. Clearly for this priest there was no question of Vi~r;tu being equal to
Siva, any more than there was in the Cola iconography of the temple.
Vi~r;tu was the foremost of devotees, who was happy to act as drummer for
the Dance. On this internal evidence it is hard to imagine in those days a
Vi~r;tu shrine as substantial as we find today. With regard to the testimony
of the two Nvars, the mention of the "three thousand" priests, does, it is
true, suggest a connection with the Nataraja temple, but it is no more
104
than a suggestion. The earliest Cola art was fervently Saiva, and it is
unlikely that Aditya I countenanced a Vaiglava presence in the centre of
the holiest of Saiva temples. The tvvelfth and thirteenth centurie~ afford
strong evidence of Siva's perceived superiority over Vi~l).u, and Pimenta's
account at the end of the si.xteenth century provides a spotlight on fervent
opposition to the Vi~l).u shrine. There is no internal evidence that
Appayya Dlk9ita's Hariharastuti applied specifically to Cidambaram. It
seems likely that any theology built on the co-presence of the two deities
in the first enclosure at Cidambaram and their equality or near-equality is
only of very recent origin.
Notes
I.
2.
3.
4
5.
6.
7.
Natarajan, 1994, p. 165. The Govindaraja shrine has been the subject of considerable discussion. In addition to the work of Natarajan, Aruvamuthan and Sankaranarayanan (notes
2 and 4), see KrishniL,wami Aiyangar, 191I; Mabalingam 1940; Ramakrishna Aiyer, 1946;
Swamy, 1979.
Aruvamuthan, 1969, p. 81.
Ibid., p.102.
Sankaranarayanan, 19 88 , p. 49
For a brief general discussion of the Cidambaram temple, see my chapter "Chidambaram"
in G. lvIichell (ed.), 1993, pp. 58,S.
Tillai and NatariiJa, p. 179.
Part I Legends, History and Architecture by T Satyamurthy; Part II Descriptions of Shrines
by Somasekharendra Dikshitendra (translated by R. Sasikala) in Somasethu Dikshitar (ed.),
1987).
8.
9.
10.
I!.
12.
13.
14.
15.
]6.
J7.
18.
I am indebted to Friedheim Hardy lor this information.

I am grateful to Friedheim Hardy for showing me his draft translation of this hymn.
Tirukkovaiyiir, trans. T N. Ramacandram, 1989.
B. Natarajan refers to an inscription of Kulottunga II in Kanchi, which records provision
made for the feeding of Slivaisl)avas by an officer of the king, and remarks that "this seems
to take something of the sting off Kulottunga II". He continue." however, "this does not
alter the position that there is a blot on the escutcheon" (Tillai and Natariija, pp. 62f.).
Cidambara lvIahatmya, ed. Soma.,ekhara Dikshitar, 1971. It was written in the reign of
Kulottun.ga I, according to Kulke, 1970; but see my comments on his dating in my The
Dance ofSiva, Chapter 2.
.
Kuficitiiizglzristava of Umapati Sivacarya, ed. with a Tamil translation and notes by K. M.
R,\jagal)eSa Dlksita, 1958. I am preparing a new edition of this text with translation and
commentary.
yaddak]e'dyiipi vipyur mapimayasadane dakeipe sViilighriyugmaTfl
IcrtviilyaTfl cottaraJyaTfl s'aramuklzaphapirabhogataipe fayana~1
nityaTfl nidraTfl prakurvann api hrdi satataTfl yat padaTfl dhyayafi,dyaTfl
deayii taTfl citl'abhefaTfl sutag'ijavadanaTfl kuiicitiiizghriTfl bhaje'haml /4211
Somasetu Dlksita in the Sanskrit introduction to his edition of the CitsabizefotsavaJRtra lists the
five princip~l shrines as I. Nataraja, 2. Sivakamyumambika, 3. Srlmulanatha,
4. Mahagal)apati, 5. Subrahmal)ya, called the SrlPal)r;lyanayaka (p. 3 of unpaginated
sequence).
Published in Ramesan, 1972, p. 160.
Putchas, 1905. pp. 208[
molzinya jambhur adav atinivi~auane darukaklrye caranya!z
saundaryiid uipradaran muniga1'J.llm api ta/'fl mayayii mohayitva!
uyarthlkrtyabizicaraTfl duy'akulavvzitaTfl saTflpradarfyatmanrttaTfl
uiprii7[Lf canuagrahldya.r tam apigiris'aya(n kuiicitiilighri/'fl bhaje'haml 14011
V'l;S('!U
Ig.
20_
2 I.
22.
23-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
and the Dance qf Siva
dViitrirpfadriigabodhacyutamakufaJ'afisvacclzaPiYil!adlziiriisOJ'{!giiviiptasvqjiviimb,gablzavavadanoddiimagftabhiriimam I
stabdliibhutar,iniiiig!11JlLTn adhivipinOJ'{! sarva.frrigiirariipa1"{l ya1"{l dT/!Vii V"ipJumiiyii pal.imatim atanol. kufiuil.iitighri1"{l bhqje'ham I Ir 30 I I
viPlDr OJ'{!s'iivaliiriin du.fakaranaklzarairyuddharatige s[janti
kiime.fiilokanotfyadgajavadanakaiadhvastaviglzneS'!JIantriil
yasyiistrCIJib'u bhaW-a1"{l sasutaga1Jllpura1"{l niif'!JIiim iisa daiIYam
devfJ'ricakrasaT(lSthii tam api J'ivakara1"{l kuiicil.iiitghri1"{l blzqje'lzaml I 12g1 I
piit/tfniikiirablzqiOJ'{! plzarJibhuvanasamiinfl.avediisuraghna1"{l
pucchiiglziitapra.!'iinliimbunidhiJ'ikhiiikha1"{l parvatendriinukalpaml
v0Ju.Tfl du!Viitibhftyii vibudlzapari/rrhaiIJ stiiyamiina{z kapiilf
yas ta1"{l nighnanjahviira dvinayanam api ta1"{l kuiicil.iitighri1"{l bhaje'haml 1r221 I
bhfmOJ'{! pftamburiifirrz kamathavaratanU1"{l kiilakiitopamiigni
jViiliisandigdlzavaktra1"{l jagad api niklzila1"{l ksoblzayanta1"{l svafr:r!Yail.z1
brahmiidyair apy qj'!JIya1"{llzarim adlzivasudlza1"{l kar~'!JIan yas- tadfyiin
atigiin katjhe dadlzau ta1"{l niklzilasuranuta1"{l kuncil.iiitghri1"{l bhaje'lzaml 1r231 I
potrfriipairz mahiinl:rJZjaladhigatadharii1"{l dawtr'!JIoddlzrlYa vegiil.
yuddhe Izatvii hirawOJ'{! suratipum akhiliiniftakrtyaii carantaml
vai.ku&,lza1"{l bhairavo yas' Sita.fikJzavigaladvaimikilii.lu.fiilaprokl.OJ'{! krtvii l.adfyOJ'{! radam urasi dadhau kuncil.iitighri1"{l bluge'lzam/ 1r241 I
blzftyii santriirmniine naralzarivap~o devatiinii1"{l samiihe
dhiitrii sa1"{lsl.i9Iamiina(z farablzavaratanu/.z siiluva{z pak,irqja!.zl
vegiit ta1"{l ched'!JIitvii svapadanaklzamuklzo:is tattvaciilankrto 'bhud
da1"{lWiisa1"{ldfpl.alokas tam akhilavarada1"{l kuncitiitiglui1"{l bhaJe'haml 1r2S1 I
yiicfiiifdzarvf!qtiifzgOJ'j1 balimaklzasadasi svii1"{l tanu1"{l vardh'!JIitvii
tfyiiil iiblzumisuriirer makutam api padiim iinayitvii trilokfml Izufzkiiriid blzf,r'!JIanta1"{l suraga1JtlVinuto viimanOJ'{! yo vinighnan
kafzkiila1"{ll.asYa delzatvacam api hrtaviin kuncitiifzglzri1"{l bhqje'lzaml 1r27 I I
miitrdrolziighafiintzi!z svapitrnidhanakrtktatra.iailendrapakeadvandvacchediitifikrrJaprabalapara..-um ajJy iipl.ukiimo munfndra{zl
nrtl.a1"{l yasyedam iitmany anugatam a.rakrd bhiiv'!JIan jiimadagnyo
riimo !eblze ciri!JlU/zpramukhavaravaran kuncil.iifzghri1"{l bhllje'lzaml I 13 I I I
yatpVJiilabdlzablzutiprablzavitayuvatzohiivavrddlzasvqjiiyo
yat.!qetriineka.revib'amitamunivaradlwOJ'j1raniigovikiira{z1
tiiliifzko rauhi1JJaiz prabalalzalavariikr~tad~tapralmnbo
jato yasya prasiidiid amrtakaradharOJ'{! kuncitiitiglui1"{l bhqje'haml 1r321 I
iirul!yiifvOJ'{! mukundaiz pratikalivigmnOJ'{! kalkiriiprprt/zi1rJii1"{l
gacclzan pa~arJavargOJ'{! kumatimagh'!JIutOJ'{! tatra tatraiva sOJ'j1st.ha1"{l1
ati~thii.kiirablziiJa1"{lyadamalakaruTJiiviiptaklzagena nighnan
sarvii1"{l blziimi1"{l yal.hiival. kalayati tam qiOJ'f! kuncitiiitglzri1"{l bhqje'lzaml !I331 I
madl!ye miirgOJ'{! svablzakta1"{l dviiavarqjanu,ra1"{l gat;u;1abherU1Jasi1"{liza'!Yiiglzrib'veMar~atiirksyaPlavaganaramukhe miidlzave blza~'!)IitviJ.1
krodlziid bralzmiiTJavarga1"{l mulzur api Cll mudlzii taV'!JIaIY iil'u ta1"{l yaiz
Izatvii rak~at svabhakta1"{ljagad api sakalOJ'{! kuncitiifzghri1"{l bhaje'haml 1r341 I
vai.ku1f!ho nitymn iidau pa.!'upatim Ilnaghail,z puTJarfkai.r salzasrer:ziircii1"{l kurvan kadiicin na carmnakamala1"{l dutmn iirft tad eval
utpiitya sv~padmaTfl yadamalapad'!JIor arp'!JIitviitizayasmiit
cakra1"{l sa1"{lpriipya rde tam api natapati1"{l kuncitiifzglzri1"{l bhaje'haml lSi I
vi.pJ;i cakriipahartu/.z pratizitabalqju,ro blziitacakrii.dlzipa![)ia
~ttrtdu/zkhapras'iinti pradaniiacarita.rtabdlzapii1Jyiitmqjiitaml
J'va.!'ruviinclziiblzipiiriJai niiadlzuri ca mudii"nita.raptiimburib'irrz
Maktas-trfpakvapit{ii.l'anarucimudita1"{l kuncil.iifzglzri1"{l blzqje'haml/249/ I
daityiinii1"{l dlwaT(lSaniirtiza1"{l sakalam aPijagadr~aTJiirtiza1"{l muriiril.z
lak,rmyii ~riimburiifau krta.kat/tinatapiis siimbamiirti1"{l prabhu1"{lyaml
bhaktya sa1"{lpiijya citte manuvarmn apiyanniimasiihasras'a1"{ljna1"{l
stotrOJ'j1labdlzviijapans latprabhur ablzavad qjOJ'j1 kuficitiifzglzri1"{lblzqje'lzaml 1r361 I
padme miistv atfya lilii. plzar:zipaiiJ'!JIa71a1"{l vainat'!)lo 'pi niilaTfl
bhiimi tva1"{l s'fghrmn iiyiihyalzam api natane tillavanyiiTfl puriire!.zl
105
ro6 l"ldian Insights: Buddhis'iTL, Brahmanism and Bhakti
34
35.
Iwstiibhyii7fl viidayirJiimyahaha catutara7fl maddala7fl [read mardala7flJ cetyudlrya

priiyiidyadrmigam adau harir api tam aja7fl /cuficitiinglzri7fl blzaje'lzaml 12301 1
Dessigane, Pattabiramin and Filliozat, 1964. pp. 86J:
tyaktasvaJ"viidlzikiirii!1 sarasijanayaniiJ cagata7fl devasau!dryapriiptyai govindam uklva ku.falam api kariiJle,apurva7fl pramodiitl
vatsa Iva'!'. brahmatatlva7fllzrdi na vikalayan prart/ryase du!Jdzalzetu7fl
karmetyudbodlzayanto yam api girivare kuficitiingiIri7fl bhaje'lzaml 12151/
Bibliography
Aruvamuthan, T G., "The Narayana-Nataraja Complex at
Chidambaram" in Transactions if the Archaeological Society if South
India, Transactions for the Years 1962-65, 1969.
Dessigane, R., Pattabiramin, PZ. and Filliozat,Jean, Les Iigendes
yivailes de Kiiiicipuram. Pondicherry: Institut Franc,:ais
d'Indologie, 1964.
Dikshitar, Somasekhara (ed.) Cidambara A1iihiitmya. Kadavasal: Sri
Meenakshi Press, 197I.
Dikshitar, Somasethu (ed.) The History if Chidambaram: Sri Nataraja's
Temple (A Guide to Sri Nataraja Temple). Chidambaram: S.
Somaraja Dikshitar, 198r
Dik~ita, Rajaga.[,lesa (K. M"i. Rajaka.[,leca Tik~itar) (ed. and Tamil
trans. and notes) Kuiicitiinghristava if Umiipati Siviiciirya.
Cidambaram: Kirasvetu Piras, 1958.
Krishnaswami Aiyangar, S., Ancient India. Madras: 19II.
Kulke, H. Cidambaramahiitmya: Eine Untersuchung der religionsgeschichtlichen und historischenHintergrunde for die Entstehung deT
Tradition einer sudindischen Tempelstadt, Freiburger Beitrage zur
Indiologie vol 3. Wiesbaden: Otto Harrassowitz, 1990,
Mahalingam, TV, Administration and Social Life under Vijayanagara.
Madras: University of Madras, 1940.
Michell, G. (ed.), Temple Towns if Tamil Nadu. Bombay: Marg
Publications, 1993.
Natarajan, B., Tillai and Nataraja. Madras: Mudgala Trust, 1994.
Purchas, Samuel, Haklll:Jtas Posthumus or Purchas His Pilgrimes Vol x.
Glasgow: James MacLehose, 1905.
Ramacandram, T N. (trans.) TirukkOvaiyiir. Tanjavur: Tanjavur
University, 1989.
Ramadrishna Aiyer, VG. The Economy of a South India Temple.
Annamalainagar: Annamalai University, 1946.
Ramesan, N. Sri Appaya Dikshita. Hyderabad: Srimad Appayya
Dikshitendra Granthavali, 1972.
Vz,s(lU
and the Dance qj Siva
!O7
Sankaranarayanan, S., "Foundation of South Indian Temple

Culture - Some observations with special reference to the
Tirumala and Chidambaram temples", Sankara Parvathy
Endowment Lectures, 1983-4, in Journal qjthe Madras University
Section A: Humanities, Vol LX, no. I,January 1988.
Smith, n, The Dance qjSiva. Cambridge: University Press, 1996.
Swamy, B. G. L. Chidambaram and Natariija. Mysore: Geetha Book
House, 1979.
Mafijusri and 'The Chanting of Names'
(Namasar[lgZtz): Wisdom and its Embodiment in an

Indian Mahayana Buddhist text
Anthony Tribe
Introduction
Among the bodhisattvas who play such a central role in the literature of
Mahayana Buddhism the figure ofMafijusrt is one of the most prominent.
The bodhisattva pre-eminently associated with wisdom, and functioning~s
the major interlocutor in many sutras, he is also depicted as a spiritual
friend (kalyiiTJflmitra), a convertor of beings to the Buddhist Dharma, and an
object of devotion and meditation. I His origi:p.s are obscure: unlike
Vajrapar:ri, for example, he does not appear in non-Mahayana literature,
though Marcelle Lalou has suggested the gandharva paficasikha as a possible
antecedent. 2
Mafijusri usually has the status of a tenth stage (bhilmz) bodhisattva.
However, in some texts he is said to be a fully enlightened Buddha. 3 Of
these, perhaps the most important and influential is a short work called the
NiimasaT[lgfti, 'The Chanting of Names' (hereafter NS). Also known as the
Maiijusriniimasa7[!g'iti, it has been translated into English twice, OI).ce by
Ronald Davidson and once by Alex Wayman. 4 The first modern scholarly
edition of the Sanskrit text was published by 1. P Minaev in 1887 along
with his edition of the Mahiil!Jlutpatti, the Sanskrit-Tibetan dictionary. 5
The NS appears to depict Mafijusri not only as a Buddha but also as the
Primordial Buddha (iidibuddha) and to portray him not simply in terms of
an embodiment of wisdom - whether that of a bodhisattva, Buddha or the
Adibuddha - but as that wisdom itself This description of Mafijusri's
depiction in the NS is couched in tentative terms. since there are a number
IIO
of puzzling features about the NS's structure and contents that raise
questions about whether the 'Names' of the NS's title are indeed those of
rvIafijusri, one of these being the fact that in the main text he is only once
referred to directly.
The primary aim of this paper is to' give some account of the nature of
MafijuSri's depiction with respect to wisdom and its embodiment in the
NS. A secondary aim is to look at how this depiction is translated into ritual
terms in Vilasavajra's commentary on the NS, the Niimamantriirthiiualokinf,
"An Explanation of the Meaning of the N ame-mantr-as" (hereafter
NMAA). Vilasavajra was one of the first Indian commentators on the NS,
probably writing in the latter part of the 8th century GE,6 and his
commentary interprets the NS within a sophisticated and complex ritual
framework. A further reason for considering Vilasavajra's work is that it
can cast light on the mutual i.."lteraction betvveen ritual and doctrine: on the
one hand ritual structures can encode or reflect doctrinal stances; on the
other, to accomodate them into particular ritual structures, doctrinal
categories may be modified. Satisfactory pursuit of these aims requires the
consideration of some of the difficulties concerning the NS's structure and
contents and in general I have tried to come to conclusions about these by
looking at the NS itself rather than by approaching them though
Vilasavajra's commentary. Often, however, Vilasavajra confirms a
conclusion implied by independent examination of the NS. In order to
avoid confusion I will refer to the 'names' of the NS as 'Names', i.e.
capitalising the initial letter. This is particularly appropriate since most of
the Names are not names as generally understood and also allows
statements such as "most of the Names are not nantes" to make sense.
Since the context of the NS is essentially tantric and since it was
classified by the Tibetan tradition as a tantra it could be considered
provocative, if not perverse, to describe it as a Mahayana text, as in this
paper's title. Would it not be better described as a Vajrayana work given
that 'Vajrayana' is the term most commonly used for tantric Buddhism?
This is not the place for a detailed discussion of the relationship between
the expressions Mahayana and Vajrayana, but a general reason for
adopting the former in relation to the NS is that the Vajrayana is often
perceived as being subsumed within the Mahayana rather than being
opposed to or subsuming it. 7 More speci..+ically, the NS does not contain a
single instance of the term Vajrayana, employing instead the expression
mantranaya, "the vVay of Mantras" . Thus, it refers to Mafijusri ~et us assume
for the present) as, "born from the great Way of Mantras, having the great
vVay of Mantras as his nature".8 An alternative and synonymous term,
mantramukha, is also found so that, for instance, the NS describes itself as,
"the quick [means of] success for boclhisat\:vas observing the practices of
the vVay of Mantras (mantramukha)".9 This quotation suggests that the NS
sees the Way of Mantras as a particularly effective method of practice
lYIafijufri and 'The Chanting ofNames' (Niimasa7[lglti)
II1
employed within the Mahayana rather than as something that confronts or

supersedes it. The NS also contains a description of the results of such
practice, given in traditional Mahayanist terms:
o VajrapaJ;li; 0 Vajradhara, before very long, the best of men, the one
who preserves the Niimasarrz;tfti of ultimate meaning and who has properly
collected the Provisions of Merit and Knowledge, will soon acquire [all]
Buddha-qualities and fully awaken to unsurpassed perfect enlightenment. lo
The NamasaT(lg'lti
Structure
The NS is a short work. It consists of 162 verses in anu~tubh metre followed
by a prose section praising the NS and describing th<;: benefits of reciting it
(anusa7[lsCi). There follows a short section containing mantras (mantravinyiisa),
and five concluding verses (upasa7[lhiira). The first 25 verses act as an
introduction to the main part of the text. They give it the appearance of a
tantra: Vajradhara accompanied by a retinue of fierce Vajrapar,ti-s asks the
Buddha Sakyamuni to teach the NS. Sakyamuni replies and, being pleased
with Vajradhara's words, agrees to his request. The core verses (NS
26-162) consist primarily of a series of predicates in the nominative case
and are a list of figures, attributes, qualities, actions and categories
understandable as embodying wisdom. These are the Names of the NS.
The concluding verses depict Vajradhara and his retinue rejoicing in what
they have heard and further praising the NS." (Table 1, overleaf,
summarises the NS's structure.)
Ronald Davidson argues that the opening and closing verses along with
the anu.sa7[lsii, mantravinyiisa and colophon probably represent a later stratum
in the development of the text. In other words, the core text (NS 26-162)
constitutes the initial form of the NS. Davidson sees this as representing "a
basic meditative form complete with devotional homage in the final five
verses" and as being "the instructions of a vajriiciirya" Y Both these
statements are rather puzzling given the nature of the core verses. Being
largely composed of a list of predicates, they contain no instructions as
such. The manner in which these verses are "a basic meditative form" is
also not immediately obvious, unless Davidson has in mind their potential
for recitation, which is made explicit by the frame text. The early
commentators on the NS certainly developed visualisation texts that
employed the verses in various ways,13 but these are secondary and
exegetical works and do not indicate anything characteristic of the core
text itself. Nonetheless, the core text does possess a unity of style and
organisation of content that separate it from the rest of the work.
I 1'2
TheNS:
Structure
Frame Text:
(1-25)
Divisions in some NS Manuscripts
Opening Verses:
a) Vajradhara's request (1-16)
Adhye~aI)a
b) Sakyamuni's reply (17-22)
Prativacana (17-22)
c) Introduction to core text (23-25)
(1-16)
:;la\kulavalokana (23~4)
Mayajalabhisarpbodhikrama (25-27)
a) Initial gatha & homage (26-27)
Core Text:
VajradhatumahamaI)r;lala (28-41)
(26-162)
Suvisuddhadharmadhatujnana (42~66c)
b) The Names (28-157)
Adarsajnana (66di6)
Pratyavek~aI)ajnana
(77-II8)
Samatajnalla (II9-142)
Krtyanu~thallajnana
c) Homage (158-162)
Frame Text:
(143-157)
Pancatathagatajnanastuti (158-162)
Concluding Sections:
a) Praise & Benefits ofthe NS
Anusarpsa (in 6 parts)
b) Mantras
Mantravinyasa
c) Concluding verses (163-167)
Upasarphara (163-167)
d) Colophon
Table
1.
-.
The Structure ofthe Namasarj7gtt:!
(verse nos.
ill
parentheses).
Indian NS commentators tended to see verses 1-16'2 as the basic text.

Vilasavajra gives 16'2 as the figure for the total number of verses, stating
that the remaining text is equivalentto 150 verses. 14 This view that the later
parts of the NS (anuiaT(lSii, mantravinyiisa and upasaTj7.hiira) are different from
what goes before is also found in Mafijusrimitra's Upadesa for the recitation
of the NS.15 VVhen the NS is chanted in present-day Nepal, the whole text,
including the colophon, but excepting the anusaT(lSii, is used. 16 Despite the
existence of these text divisions, caution is needed before concluding that
they represent different strata of textual composition. 17 Evidence from the
analysis of the content of the NS should be taken into account before
drawing any definite conclusions.
Influence ofthe Niimasa1'(lgfti.
If the testimomy of the very large number ofNS related works preserved in
the Tibetan Tanjur - I'29 in total- is anything to go by, the NS exerted a
particularly strong influence on the imagination of the Indian Buddhist tradition, especially in its tantric form, for a period of three hundred years or
more from the middle of the 8th century onwards. The works in Tibetan
Marf/ufn: and 'TIe Chanting ojNames' (NiimasaTJ'lgfti)
II
translation range from lengthy commentaries to shorter siidhana texts and

ritual works covering many aspects of tantric -and non-tantric Mahayana
practice.
One reason for its enduring influence may lie in the fact that the NS
does not explicitly promote anyone philosophical position, whether
Madhyamaka, Yogacara or Tathagatagarbha. Thus the series of descriptions of the qualities, attributes and embodiments of the enlightened state
of which the verses are composed are amenable to a wide range of interpretation, as well as being eminently suitable for liturgical and devotional
use. The tantric content of the NS most ob~ously belongs to the phase of
Buddhist Tantra known as Yogatantra, IS where Mahavairocana is
envisaged as the cenqal embodiment of enlightenment. However, there is
no systematic development of this tantric material and as a result the NS
could be placed within different tantric contexts without too much strain.
Among the earliest commentators on the NS are Mafijusrlmitra,
Vilasavajra and Vimalamitra, who can with varying degrees of caution be
placed in the middle to late eighth century, giving a terminus ad quem for the
NS. An estimate of the terminus a quo for such works is always a difficult
matter: Davidson suggests a date of composition of perhaps fiftY years
before the earliest commentary.19 Vilasavajra's commentary, the NMAA,
survives in Sanskrit in a number of Nepalese manuscripts. It is a fairly long
work - equivalent to approximately 3,000 an~tubh stanzas - which sets the
NS within a primarily Yogatantra context. 20 It is an important text, not
only because it contains a wealth of citations from both tantric and nontantric sources but also for the reason that it is probably the earliest
Buddhist tantric exegetical work to survive in its original language. The
NMAA was translated into Tibetan (Toh. 2533) in the early IIth century by
Smrtijfianaklrti who himself wrote a NS commentary (Toh. 2538),
following the tradition of interpretation initiated by Vilasavajni.
Whereas the early commentators interpreted the NS as a Yogatantra
later writers tended to link it either to the Kiilacakratantra or to. the
Yogimtantras. 21 Of the later commentators, Mafijusrlklrti, in the early 10th
century, wrote the last long Yogatantra commentary (Toh. 2534)22 and
Kalacakrapada, the traditional founder of the Indian Kiilacakra cycle,
commented on the anufaTJ'lsii (Toh. 1399). Naropa, the latter's pupil and the
teacher of Marpa, related the NS to the Kiilacakra system in his
commentary on the consecration (abhi1eka) chapter of the Kiilacakratantra,
the SekoddeSaflkii. 23 An important commentary written from the Kiilacakra
perspective is the Amrtaka1',likii of Ravisrtjfiana (Toh. 1395). This work, its
sub commentary, the Amrtaka1',likoddyotana ofVibhuticandra, the NMAA and
a commentary of unknown authorship, the GUhapadii, are the only NS
works known to survive in the original Sanskrit. 24
One of the last Indian writers of significance on the NS was Advayavajra,
who probably lived in the late 10th to early IIth centuries. The Tanjur
II4
contains nine works by him which place the NS within the framework of the
Yogimtantras. Advayavajra also promoted the NS in lay-Buddhist circles. In
the Kudntinirghiitanam, "The refutation of Wrong Views", recital ofthe NS is
enumerated as part of daily ritual for householder bodhisattvas. 25 A number
of other well known figures also wrote' commentaries on the NS inclucling
I)ombiheruka (Toh. 2542), Anupamarak:jita (Toh. I396), and the tantric
Candragomin (Toh. 2090).
An episode from the ~ege!ldary) life of Candragomin that involves the
NS is related by the Tibetan historians Eu-ston and Taranatha. The episode
is part of the story of Candragomin's visit to Nalanda and his challenging
the famous Madhyamika philosopher Candrak'irti to debate. Accorcling to
Taranatha's account, on his arrival at Nalanda, Candragomin sees
Candrak'irti preaching outside the boundary walls of the university and
stands nearby listening to him. Candrakirti, wondering whether
Candragomin is an opponent wanting to debate, asks him where he comes
from and what subjects he knows. To the latter question Candragomin
replies, "I know the grammar of Pal}ini, the hymn of praise in 150 verses,
and the Niimasarrzgfti." The account continues,
Thus, though in words he did not expres.s pride insofar as he said that he
knew nothing beyond these three treatises, by implication he claimed that
he knew all about grammar, sutra, and tantra. 26
Despite its conflation of Candragomin the grammarian with the tantric
Candragomin the story indicates the special status given to the NS.
The NS was translated into Tibetan by the great IIth century Tibetan
translator Rin-chen bzang-po and extensively revised in the late 13th
century by bLo-gros brtan-pa. However, an earlier translation than Rinchen bzang-po's must have been made since the NS is mentioned in the
lDan-dkar catalogue of Buddhist texts, which probably reached its presesnt
form in the early 9th century CE,z7 and there are also NS manuscripts from
Dun-huang. 28
In Nepal, where Manjusri is the legendary creator of the Kathmandu
valley, and where the Dharmadhatuvagisvaramar;u;iala of Manjugho~a is
inscribed in metal on a raised plinth in front of many temples and stupas,
the NS is still chanted in Sanskrit as part of the daily morning ritual in
some of the baha/:ls (Skt. vihiira).
Mafijusri in the NamasaT(lgiti

Manjusr! as j\1afijusrijfiiinasattva
Looking at how Manjusri is described in the frame text provides a point of
departure for examining his portrayal in the NS. In the opening verses,
Mafi.jufri and 'The Chanting ofNames' (NiimasaT[lglti)
IIS
Vajradhara refers to "the NiimasaT[lglti ... of the Knowledge-Being Mafijusri

(mafijufrfjiiiinasattva)"29 and characterises this figure as "the KnowledgeBody"Uiiiinakaya), "the Lord of Speech" (g~spatz), "the Embodiment of
Knowledge" 0iiiinamiirtz), and "the Self-existing One" (svayaT[lbhif).30
Significantly, the figure so described is not Mafijusri as such, but "the
Knowledge-Being Mafijusn". This same expression is used on four other
occasions: three times in the prose anufaT[lsii section and once in the
colophon. In three of these the context is a similar description of the NS as
being that of the Knowledge-Being Mafijusri,'1 The remaining instance
(one of those in the anufaT[lsii section) is that of a recommendation to
meditate on his form. 32
The word 'Mafijusn' is used just once more in the frame text, again in a
compounded form. In the opening verses Sakyamuni, replying to
Vajradhara, refers to Maiijusri as "the Knowledge-Body Mafijusri"
(maiijufrfjiiiinakaya).33 The connection between the two titles, "the
Knowledge-Being Mafijusri" and "the Knowledge-Body Mafijusri" is
clarified in the following phrase, found on two occasions in the anufa1J1Sii:
" ... the NiimasaT[lgiti of the Knowledge-Being Mafijusri who is the
Knowledge-Body 0iiiinakaya) of all the Tathagatas."34 lfthe termjiiiinakaya is
construed as a karmadhiiraya compound, ie. if it means '.'thebody [or
collection] that is Knowledge", then to say that Mafijusri is thejiiiinakaya of
all the Tathagatasis to say that he is the Knowledge that underlies or lies
within all Tathagatas. 3S
This description of Mafijusrl as the Knowledge-Body of all the
Tathagatas parallels that of the Perfection of Wisdom (prajiiiipiiramitftJ as the
mother of the Buddhas as seen, for example, in chapter 12 of the A!tasiihasrikii
Prajfiiipiiramitii. Here, the metaphor of 'mother' is used in the sense of the
'source' of the enlightenment that makes one a Buddha. In a similar vein
the NS describes Mafijusn as "the progenitor of all the Buddhas". 36 As a
bodhisattva, Mafijusn is often portrayed as a son of the Buddhas - which is
how the very next pada describes him: as "the best, supreme son of the
Buddhas"37 - but as the Knowledge-Body he is their progenitor. This use of
the metaphors of 'mother' and 'progenitor', despite differences of nuance,
conveys the sense that Buddhahood has a prior condition, whether this is
formulated as prajiiii or jfiiina. Under the influence of the Tantra, the
Perfection of Wisdom becomes Prajfiaparamita, a deity embodying
Wisdom; whereas Mafijusri, the bodhisattva of Wisdom, undergoes a
contrary transformation to become jiiiina, the Knowledge he embodies.
The Mafijusri who is the Knowledge-Body of all the Tathagatas is not
simply Mafijusri but Mafijusrijfianasattva. Like the term jiiiinakaya,
jiiiinasattva is best understood as a karmadhiiraya; ie. "Knowledge-Being"
should be analysed as meaning "a Being that is Knowledge". This analysis .
of jiiiinasattva as referring to something that is both "knowledge' and 'a
being' makes it easier to see how the term sometimes denotes pure wisdom
lI6
(e.g. as "the Knowledge-Body") and sometimes its embodiment (e.g. as "the

Embodiment of Knowledge" and, indeed, the NS Names).
The discussion of the word 'Mafijusrljfianasattva' in the NMAA is also
pertinent:
He [ie. MafijusrlJ is called a 'Knowledge-Being' since he dwells in the
heart of all the Tathagatas. The Knowledge-Being Mafijusrl is not the
bodhisattva who is the master of the Ten Stages (bhuml). Rather, he is Nondual Awareness (aduaya;jfiiina), the Perfection of Wisdom itself For this very
reason Dignaga says, "The Tathagata is the Perfection oflNisdom, that is
to say, Non-dual Awareness".38
Vilasavajra here spells out that the Mafijusrl who is the KnowledgeBody of the Tathagatas (as described in the anufaT[lsii) cannot be the same
as the bodhisattva MafijusrL 'Whether his further identification of the
Knowledge-Being MafijuSrI with advayajfiiina is supported by the text of
the NS is a question that will have to be deferred for the present.
Within the literature of tantric visualisation (siidhana) the expression
'Knowledge-Being' (jfiiinasattva) is familiarly paired with the term 'PledgeBeing' (samayasattua). This is a more technical usage, jfiiinasattua here
representing the real nature of the deity or deities visualised by the
practitioner. In the mechanics of the siidhana the Knowledge-Being and
Pledge-Being are usually imagined as separate deities (or sets of deities), the
latter being the designation for the form that the mantrin initially evokes
and identifies with. At the moment the Knowledge-Being is visualised as
merging with the practitioner-as-Pledge-Being the practitioner is, in the
terms of the ritual, fully enlightened. 39
Neither the NS nor Vilasavajra's commentary refer to this distinction
between jfiiinasattua and samayasattva. Nonetheless, their use of the term
jfiiinasattua is not unconnected with its function when paired with
samayasattva. Just as, in the latter context, the jfiiinasattva is depicted as the
defming and enlightened aspect of a deity, Mafijusrljfianasattva is
portrayed as the jfiiina that makes Buddhas what they are and without
which they would be mere (unenlightened) forms. He is, as the NS anufaT[lsii
repeats twice, in the hearts of all Tathagatas.
There is, however, an important difference between how jfiiinasattua
functions in the two contexts. For the meditator who has invoked and
merged thejfiiinasattva with the samayasattua, the termjfiiinasattva can denote
a wide variety of deities, depending on the particular siidhana involved. In
the NS the implication is that Mafijusrl is the Knowledge-Being, thereby
underlying all deities and form of wisdom. Whether the distinction
between jfiiinasattua and samayasattua had evolved at .the time of NS or
Vilasavajra's commentary is uncertain. If it had not, it may be that the use
of the termjfiiinasattva in the NS and its early Yogatantra commentaries
contributed to its development. Vilasavajra translates the notion of
NlaifjuJri and 'The Chanting qfNames' (Niimasa1[lglti)
117
Mafijusrljfianasattva being in the heart of the Tathagatas into visual and

ritual terms in his NS siidhana in a manner that could easily pave the way
for such an evolution. 40
The 'Names' in the "Chanting of Names"
Having examined MafijuSrl'S depiction in the frame text, there remains the
issue of whether and, if so, how Mafijusrt is portrayed in the core text. The
bulk of the core text (130 of 137 verses) consist of the Names of the NS, in the
form of a series of predicates whose subject is never clearly stated. An
important question, therefore, is whether Mafijusrl is the unstated subject.
Do the Names name Mafijusrl or some other figure? The depiction of the
NS in the frame text as being that of, or associated with, Mafijusrtjfianasattva
leaves open the possibility that although the text is in a sense his - e.g. it
could have been composed or spoken by him - the Names could be those of
some other figure or object. Finally, if the Names do name Mafijusrl, do
they name him as a bodhisattva or as MafljusrIjfianasattva? The question of
whom (or what) the Names name needs to be addressed before the nature of
Mafijusrl's depiction in the core text can be pursued any further.
The title of the NS appears to make it obvious that it is Mafijusri's
names that are being recited. Both Davidson and Wayman give
Maf[juJriniimasaT[lglti as the title of the work they are translating. However,
this title is not given to it anywhere in the text itself or in its colophon.
Moreover, the compound maiijusriniimasaT[lgzti is ambiguous. It can be
analysed either as 'The Name-Chanting of Mafijusrl' or, rather differently,
as 'The Chanting of Mafijusrl's Names'. Davidson, whether intentionally
or not, retained that ambiguity when he translated it as "The Litany of
Names of Mafijusrt" .41 Wayman opted for the unambiguous "Chanting the
Names ofMafijusrl".
Maffjusriniimasa1[lglti is in fact given as a title for the NS in a number of
(mainly later) Indian commentaries and other secondary NS works,42 so its
use as a title is not without precedent. An examination of the titles used in
the NS itself may clarify how the title MafijufrzniimasaT[lgzti should be
understood. As has been noted, the NS calls itself the Niimasa1[lglti in the
frame text. The longer title, Maf[juJrfjiinasattvasya Paramiirthii Niimasa1[lglti~,
"The Supreme Name-Chanting of Mafijusrljfianasattva" occurs in the
colophon. This description of the text is confirmed. in both the opening
verses (NS ro-II) and in the anusaT[lsii. Thus, though the Name-Chanting,
ie. the text as a whole, is said to be Mafijusrt's, at no point are the actual
N ames said to be his. On this basis the title Nlaf[jusriniimasa1[lglti is best read
as 'The Name-Chanting of Mafijusrf, not as 'The Chanting ofMafijusrl's
Names'. It is, nonetheless, understandable how the title Maf[juJrfjiinasattvasya
Paramiirthii NiimasaT[lglti~ could be shortened to MafijuSriniimasa1[lglti and used,
perhaps in preference to Niimasa1[lgZti, in order to preserve the connection
with MafijusrL 43
118
This analysis of the title is the one followed in the NMAA. To describe
the genitive relation between the NS and Mafijusnjfianasattva Vila,savajra
uses the term sa1'{lbandha, meaning 'connection' or 'association', so that 'the
NS of Mafijusnjfianasattva' is to be understood as 'the NS associated
with, or related to, Mafijusnjfianasattva'. 44
The question of whom or what the Names of the NS name thus
remains unanswered. There may be some help in the two somewhat
obscure but crucial verses that open the core text:
A A I i U D E AI 0 AU AM A1:I. I, the Awakened One, the Embodiment
of Knowledge, am in the heart of the Buddhas of the three times. 01\1 Obeisance to you, Diamond-Sharp, Destroyer of Suffering, Embodiment of
the Knowledge of Wisdom, Knowledge-Body, Lord of Speech, Arapacana.
(NS 26-2 7)45
Of present concern - leaving aside for the time being consideration of

what these verses might mean - is the question of the identity of the speaker
who, after the enumeration of twelve vowels of the Sanskrit alphabet,
declares "1, the Awakened One, the Embodiment of Knowledge ... ". An
answer is provided by the three verses that precede these two. The three
form one sentence and immediately follow Vajradhara's request to hear the
NS and Sakyamuni's subsequent assent. The verses, in abbreviated form,
read as follows:
Then Sakyamuni, the Fortunate One, having surveyed the great entire family of mantras ... spoke this verse connected with the Lord of Speech, which
is endowed with six kingly mantras, which has the Non-dual as its source,
[and] which has the characteristic of non-origination. (NS 23-5)46
So, the verse starting with the 12 vowels is that of the 'Lord of
Speech' (girii1'{l pateM, a common epithet for Mafijusrl. As a result,
although it is Sakyamuni who is speaking, it is the verse of Mafijusrl
that he speaks. In other words the 'I' of "I, the Awakened One ... am in
the heart of the Buddhas of the three times" is Mafijusrt The NMAA also
takes this view, specifically identifying the 'Lord of Speech' of NS 25 with
Mafijusrijfianasattva. 47
Following Mafijusrl's declaration is the verse of homage, "OM Obeisance to you, Diamond-Sharp ... ". Usually taken to be the "six
kingly mantras", as described in NS 25, it could be interpreted as a
homage to six different forms of Mafijusrt If so, it is a homage to the
subject of the previous verse. However, it is not clear who is uttering this
verse. Is Sakyamuni continuing to report a self-homage spoken by
Mafijusn in, say, the form ofMafijusnjfianasattva? Is Sakyamuni speaking
his own words? Or is there some other alternative?48 Mter these two
verses, the remaining 135 verses ofthe core text follow (NS 28-162). With
the exception of the final five verses of homage (NS 158-162) they can all
MaftjuJri and 'The Chanting ofNames' (NiimasarrzgJti)
II9
be read as a series of predicates (the Names), often a separate one for each
quarter verse. 49
The relation of these remaining 135 verses to the two that precede them
is also not completely clear. The first, NS 28, starts with the phrase tadyathii
(,accordingly', 'that is to say'), normally used before an expansion or
elaboration of a preceding item or statement. If it is taken to govern the' rest
of the verses of the core text,50 then they can be understood as an
elaboration of NS 26-7, and therefore the subject of the predicates or
Names is the'!, of "I, the Awakened One ... am in the heart of the Buddhas
of the three times", namely the Lord of Speech, MafijuSrl.
This means that the Names of the NS are, despite the previous analysis
of the title, names of Mafijusri. Arriving at this conclusion has involved a
number of arguments and assumptions concerning the relationship between
various parts of the text. IfNS 26-162 did start off as an independent text
then the identity of the unstated subject of the predicates or Names cannot
be ascertained. What can be said is that the subject has to be masculine in
gender and function as something or someone that underlies the whole
range of Buddhist exemplars of realised wisdom in much the way as
MafijuSrijfianasattva is envisaged as doing in the frame text.
The earliest commentators knew the text as we have it, with the
introductory verses and the concluding sections. So far as is known, it was
never commented on other than as a whole. 51 For the present, I will treat
the text as of a piece and allow the predicates of the core text to count as
evidence in the inyestigation ofMafijuSri's portrayal in the NS. In any case,
the portrayal ofMafijusri in the frame text is consonant with what could be
said of the putative subject of the core verses taken in isolation.
Nonetheless, there is also a sense in which the Names are not names of
Mafijusrl, and this has to do with the question of what sort of thing the
Names are. They are not a straightforward list of epithets and attributes
that could be ascribed to Mafijusri as a bodhisattva or even as a Buddha.
For instance, they include a number of pr9per names of different figures,
such as Samantabhadra (NS II5), Vairocana (NS 62) and Vajrasattva
(NS 71). Samantabhadra or Vajrasattva are not, and cannot be, alternative
names of Mafijusri; he has to be seen, rather, as their identity, as what
ultimately underlies their nature. From this perspective, the Names are
names of the forms in which Mafijusn asXnowledge-Being, i.e. asjfiiina, can
appear. As such they could, and do, range from Buddhas to doctrinal
categories. They could be described as being MafijuSrijfianasattva as he
appears at the level of conventional truth (saTflvrtisatya). In summary, the Lord
of Speech who declares himself to be in the heart of the Buddhas of the
three times and who can be taken as the subject of the Names is Mafijusn,
but Mafijusri the Knowledge-Being rather than the bodhisattva Mafijusri.
The NMAA broadly confirms this picture. It states that the NS Names
are both supramundane and mundane, and denote two things, firstly, "the
120
Yoga, Kriya and Carya Tantras, the different categories of sacred utterance
(pravacana), the sutras, the Abhidharma and the Vinaya"; and second).y, "all
things moving and unmoving".52 Given that these two cover all objects of
experience, Vilasavajra has perhaps gone beyond what the text justifies. The
inclusion of all mundane objects almost certainly reflects Vilasavajra's
Y ogacara perspective, namely that all objects of experience result from an
erroneous partition of Non-dual Awareness (advayajfiiina) which, we should
remember, Vilasavajra has identified with Manjusnjnanasattva. So, if everything is, ultimately, Non-dual Awareness then everything is, ultimately, the
Knowledge-Being Mafijusn. But the Names are chanted, adds Vilasavajra,
at the level of conventional truth. This, he says, is, "in accordance with the
principle that all this is mere names until [one has reached] the upper limit
of existence". 53 It is not until his commentary on the fifth section of the NS,
however, that Vilasavajra clearly states that the Names are to be understood
as qualifYing Manjusri the Knowledge-Being. 54
The fact that the termjfiiinasattva is not found as one of the Names
further suggests that the NS takes Manjusrijnanasattva as the figure that is
'Named'. However, as noted earlier, 'Manjusn' is found once as a Name.
The only uncompounded. occurrence of the word in the whole text, the
name (and Name) is found in NS 157, the very last verse before the five
verses of homage that complete the core text of the NS. Its final two piidas
read, "The glorious one, producing every success; Maiijusri, the best of the
glorious".55 This Naming of Manjusri as Manjusri makes sense if it is the
Knowledge-Being Manjusri, as Non-dual Awareness, that is being Named,
or embodied, as the bodhisattva ManjusrL 56
ManjusrI in the Core Text of the NS
This understanding of the Names, however, means that any and every
Name is relevant to an account of Manjusri's portrayal in the NS. Given
the possible scope of discussion, it is only possible in the present article to
sketch a few of the more important emphases found in the NS in relation
to how Manjusri is depicted as wisdom and its embodiment. 57 It should be
borne in mind in what follows that when it is stated that Manjusri is called
or identified as something or other in a verse containing the Names the
identification is never explicit and that the name Manjusn should be
understood to refer to the Knowledge-Being Manjusri rather than the
bodhisattva Manjusri, unless otherwise stated.
Earlier it was noted that Vilasavajra identified Manjusnjnanasattva as
Non-dual Awareness (advaycrJfiiina) and the question was raised as to whether
this could be supported by the text of the NS. Such an identification is made
in NS 99b, where Manjusri is said to be "Awareness that has Non-duality as
its nature".58 Elsewhere he is simply described as 'Non-dual' (advaya).59
Although advaya is not a word that is used extensively in the NS it has a
significant role. NS 26, the verse that opens the core text by enumerating
lvIaffjufri: and 'The Chanting ifNames' (Niimasarrzgiti)
121
the twelve vowels, is described in the previous verse as having its source in
the Non-dual (aduayodayiim),60 and the anusarrzsii describes the NS as having
Non-duality as itsultimate meaning. 51 Use of the term does slant the NS
towards the Y ogacara, though other more familiar or technical Y ogacara
terms are not present. Thus, Manjusri is not described, as one might
expect, as free from the duality of object and subject (griihyagriihaka) and
there is no mention of the three natures or aspects (trisvabhiiva).
The presence and significance of the term 'Knowledge-Body' 0liiinakiiya)
as an epithet for Manjusnjnanasattva in the frame text has already been
discussed. In the core text it is used three times, in NS 27, 162 and Il4. The
term is in the vocative case in the first two instances; firstly, in the verse of
the "six kingly mantras" (NS 27) and, secondly, in the last of the five
homage verses closing the core text (NS 162).52 In NS 1I4 it occurs, as part
of a Name, within a compound word that describes Manjusn as "The
holder of all Knowledge-Bodies (nibfe~(]jiiiinakDyadhrk)". The commentator
Narendrak"irti glosses 'all' as 'of all Tathagatas', makingjiiiinakiiya's meaning
close to that found in the frame passages. 63
The initial series of epithets given by Vajradhara to Manjusrijnanasattva
(in NS 10) and noted earlier included "the Embodiment of Knowledge"
(jiiiinamilrtz). As well as occuring twice in the anufarrzsii this description is found
three times in the core text. The only occurrence of the word as a Name is
found in NS 100, where Manjusri is said to be "The Embodiment of
Knowledge, the Tathagata".64 It also occurs, as has been seen, in the
declaration of the opening verse of the core text (NS 26), "I, the Awakened
One, the Embodiment of Knowledge Uiiiinamilrtz) , am in the heart of the
Buddhas of the three times". Its third appearance, this time in a
compounded form, is in NS 27 (of the "six kingly mantras"): "Ol'yI Obeisance to you ... Embodiment of the Knowledge of Wisdom
(prajiiajiiiinamilrtaye) " .55
As an epithet jiiiinamilrti emphasises the point that Manjusri as
Manjusnjnanasattva is not only wisdom (jiiiina) but also its embodiment
(sattva). He can be the "Tathagata" and "the supreme son of the Buddhas"
as well as "the progenitor of the Buddhas". The NS contains a wide range
of figures embodying wisdom from the whole spectrum of the Buddhist
tradition. The identification with Samantabhadra, Vairocana and Vajrasattva has been noted. Elsewhere Manjusrijnanasattva is, variously,
Mahavairocana (NS 42), K~itigarbha (NS Il5), Amoghapasa (NS 60);
Vajrabhairava (NS 66), Yamantaka (NS 68) and Vajraraja (NS 71). He is
also an Arhat and a bhi~u of controlled senses (NS 52).
Aspects of Mafijusrfs Portrayal in the Namasarp.gIti
Three more aspects of Mafijusri's portrayal in the NS are relevant to the
present discussion: his identification with the .Adibuddha, with the five
Awarenesses (jiiiiniim), and his relation to the letter A
I22
hLdian Insights: Buddhism, Brahmanism and Bhakti
The word Adibuddha is used just once in the NS, in the one hundredth
verse where Manjusri is described as "Buddha, without beginning or end;
Adibuddha, without [causal] association" (NS 1Ooab).66 This single
reference to the Adibuddha bestowed considerable status on the NS for
followers of the Kiilacakra cycle. However, it appears that the NS does not
place any particular emphasis on the notion of Manjusri as Adibuddha. It
is not, for example, found among the epithets of NS 10 in the opening
verses as is the term svayarrzbhil, "Self-Arisen One", which along with
iidibuddha can be suggestive of a non-Buddhist substantialist doctrine. 57 In
any case, the NS is not a systematic work in which the meaning of terms is
developed or elaborated. Also, the identifications (i.e. the Names) are
generally not repeated; terms such as jiiiinakiiya, jiiiinamilrti, suayarrzbhil,
iidibuddha appear just once.
In NS 59 Manjusri is described as "The Lord who has the Five
Awarenesses as his nature" .68 The concept of the Five Awarenesses Uiiiina) ,
sometimes known as the Five Realisations (abhisarrzbodhz), is an important
structuring notion and hermeneutical device in Buddhist Tantra from the
period of the Yogatantras onwards. In Yogatantra cycles the Awarenesses
are generally assigned to the five Tathagatas of the mal).<;lala: one to the
central figure - the lord of the mar;t<;lala (cakrefa) - and one each to the
Tathagatas of the cardinal directions. Thus, Ak~obhya typically has the
Mirror-like Awareness (iidarsajiiiina); Ratnasarpbhava, the Awareness of
Sameness (samati[jiiiina); Amitabha, the Discriminating Awareness
(pratyavekJar;i[jiiiina); Amoghasiddhi, the Praxis Awareness (krtyiinu!thiinajiiiina)
and NIahavairocana, as the central deity, the Awareness of the Perfectly
Pure Dharma-Sphere (suuifuddhadharmadhiitujiiiina).69 The first four
Awarenesses have their origin ill the Y ogacara tradition as aspects of 'Great
Awakening' (mahiibodhz),1 but the fifth, the suuifuddhadharmadhiitujiiiina,
appears to be a creation of tantric Buddhism. This addition may well
exemplify how Mahayana doctrinal categories can be modified to meet
tantric requirements, in this case for a fivefold structure (due to the
maI,l<;lala having a centre and four directions). The identification of
Manjusri with the Five Awarenesses both confirms the view that in the NS
Manjusri becomes the wisdom which underlies Buddhahood and also
indicates strong Yogatantra influence on the work. 71
The Five Awarenesses were also adopted as a way of structurillg the NS.
From at least the time of Mafijusrimitra and Vilasavajra the NS was
divided into I4 sections. 72 The verses in the core text that constitute the
Names (NS28-157) are divided into six sections. The verses of the first
section (NS28-4I) are said to be on the VajradhatumahamaI,l<;lala, the
Yogatantra maI,l<;lala that has Mahavairocana at its centre, as just
described; those of the remaining five sections are said to be devoted to
each of the Five Awarenesses in turn. In the extant manuscripts of the NS
the text is sometimes given these divisions, sometimes not.
lvIaffiuJri: and 'The Chanting qfNames' (.Niimasa1'[lgfti)
123
The letter A also possesses a significant role in the Yogatantras. It is the

first letter of the alphabet, and thus can be seen as the letter or sound from
which all the others emerge. Its use as a negative prefix also made it the
perfect symbol for the essence of the Perfection of Wisdom. 73 For the NS the
letter A appears to be more fundamental even than Mfu"1jus~fianasattva.
Thus NS 28, which follows the two key verses containing the twelve vowels
and the "six kingly mantras", reads:
That is to say, [the Knowledge-Being Mafijusri is] the Fortunate One, the
Awakened One, the Fully Awa..l(ened One, born from the letter A. The letter
A is the head of all letters, the great good, the supreme syllable. 74
If the hypothesis that this verse and the 129 that follow are meant to
elaborate and expand on the previous two is correct, the intention here
could be to point back to the opening sequence of vowels which precedes
the statement, "I, the Awakened One, the Embodiment of Knowledge, am
in the heart of the Buddhas of the three times". The vowels can be seen as
representing an articulation of wisdom that is closer to it than words can
be. They are the source of, and therefore should precede, Mafijusri's
declaration of being in the heart of the Buddhas. For the NS, the sound of
the letter A stands between Vimalakirti's silence and MafijusrI's speech in
the Vimalakzrtinirdeia.
Further elucidation of the significance or role of the twelve vowels is
inappropriate here, except for two points. Firstly, Vilasavajra identifies
them with twelve Tathagata stages (bhilmz).75 Secondly, their important
position in the NS is indicated by a story in the Svaya1'[lbhil PuriilJa. This late
Nepalese Sanskrit account of the appearance ofSvayarpbhu in Nepal and
of the origins the Kathmandu valley contains a story of an Indian monk
Dharmasrimitra who, in past times, is described as wanting to learn the
meaning of the twelve vowels of the NS. In his quest for the answer he
had started on the long and dangerous journey to China, presumably to
Mafijusrl's residence at Wu' Tai Shan, to ask Mafijusrl himself their
meaning. Passing through Nepal on his journey, Dharmasrlmitra is saved
the trouble of going all the way to China by meeting Mafijusri in the
Kathmandu valley instead. Mafijusrl initiates the monk and explains the
NS to him. Unfortunately, the Svaya1'[lbhil Purii7Ja is not very forthcoming
on the contents of the teaching that Dharmasrlmitra receives from
Mafijusrl concerning the meaning of the vowels, saying only that "by
means of these letters, all siistras, and even the Vedas, are produced". 76
Vilasavajra's Ritual Elaboration of the Niimasarpgfti in the NMAA

In Vilasavajra's commentary on the NS the root text is interpreted
within a complex ritual structure. The main element of this structure is a
I24
long siidhana centred on NS 26" the two verses that open the core text by
enumerating the twelve vowels, "A A I IuD ... ".77 In what follows I will
attempt to show how, by this means, Vilasavajra transposes the NS
depiction ofMaiijusri into a ritual framework.
The siidhana can be seen as elaborating a statement in the anufaT{lsii that, .
He who ... recites [from memory] this crest jewel called the NiimasaT{lgiti
three times each day, or who recites it from a book [and] who, taking the
form of the Fortunate One, Mafijusrijfianasattva, reflects and meditates on
that form ... will before very long see him [ie. MafijuSri] in his Form Body
(riipakqya), and he will see all the Buddhas and Bodhisattvas in the vault of
the sky in their Transformation Form Bodies (nirmii1J.ariipakiiya).78
That Vilasavajra had this passage in mind is indicated by his own
citation of the latter part of it in an account of the benefits to be gained
from practising the NMAA siidhana. 79 The anusaT{lsii passage, which
continues by detailing how such a practitioner will be protected from a bad
rebirth as a result of his actions, has a parallel in the MaffjufnparinirviiTj.O.Si1tra,
an early Mahayana sutra devoted to Mafijusrt. 80 (It should be noted that
Mafijusrt's 'Parinirvat;ta' is not real but performed by him out of concern
for the welfare ofliving beings.) Near its conclusion the text declares:
Every being who hears even the name of Mafijusri being pronounced will
be separated from the miseries of rebirth for twelve hundred thousand
aeons. Whosoever bows down to and venerates him will be reborn in the
family of the Buddhas and will be .protected by the power of Mafijusri.
Those who meditate attentively on the form of Mafijusrl, [and] on his
teaching ... will individually see hi.rrl and they will all obtain Insight. Those
who are not able to see him should recite the Silra7!lgamasiitra and say the
name ofMafijuSri. In a period of between one and seven days Mafijusri will
come to them. 81
In many respects Vilasavajra's siidhana is not far removed from its.roots
in the buddhiinusmrf;i type of practice thus described. Mantras replace saying
the name of Maiijusri - though mantras often consist of, or include, the
name of the figure or object they denote- and instead of imagining the
form of Maiijusri before one, it is visualised as ones own.
Vilasavajra starts with preliminaries that create the non-tantric
Mahayana ethical context for the visualisation. Appositely called "the
ritualization of moral attitudes" by Stephan Beyer,82 in the present context
this stage consists in the generation of the bodhicitta in its two forms, firstly
through the vows of the five Buddha-families and secondly, through the
development of the Five Awarenesses. In the next stage the meditator
imagines his or her consciousness to be naturally radiant and located in
empty space. This is a preparation for the visualisation proper, which
begins with the mental creation of a mat;tc;lala with two enclosures. The
Maf[jufrf and 'The Chanting qfNames' (Niimasar[lgfti)
125
outer enclosure is square 'with an arched doorway in each of its walls. The
archways and walls are decorated with "with bells, ornamental strips of
cloth, banners, garlands and Yak-tail plumes". In the circular inner
enclosure thrones 'for the central figure and the surrounding subsidiary
figures are imagined.
The practitioner next visualises bm- or herself as Mahavairoc;'ma at
the centre of the maI).<;1ala, white in colour and four-faced. 83 In the four
cardinal directions the Tathagatas Alqobhya, Ratnasarpbhava, Arnitabha
and Amoghasiddhi are to appear, as are the remaining deities of
VajradhatumaI).<;1ala upon which the siidhana is based.
What is particular about the siidhana is the development of the
visualisation of the central figure, NIahavairocana. In his heart, ona moondisc, transformed out of the syllable DHI!;I, sits the Adibuddha, five-faced
and eight-armed. The Adibuddha's four right hands grasp four swords,
one in each hand, and bs left hands each hold a volume of the Satasiihasrikii
Prajfiiipiiramitii. 84 If the logic of maI).<;1ala-symbolism is that the closer one is
to the centre, the closer one approaches ultimate truth, then what is being
said here is that the Adibuddha in some sense underlies, or is the source of,
the Five Tathagatas. Whether this underpinning has an ontological as well
as a ritual dimension is a further issue; Vilasavajra, at least, appears not to
treat the figure of the Adibuddha in ontological terms.
The Adibuddha, however, is not the final element in the visualisation of
the central figure and the movement inwards continues. At the heart of the
Adibuddha is a wisdom wheel (prajfiiicakra) on whose spokes and circular
bands are placed mantras taken from the NS. At the centre of this wheel
and transformed out of the letter A is the figure of Mafijusrljfianasattva,
seated on a variagated lotus throne, six-faced and two armed, holding the
stem of a lotus in each hand. On each of the lotus flowers rests a volume of
the Perfection ofWisdom. 85
The tunneling into the heart of the maI).<;1ala is still not quite complete.
On a moon-disc in the heart of Mafijusrljfianasattva the practitioner
imagines the letter A, "the essence ofthe Perfection of Wisdom, the cause of
the arising of the Awareness of the Omniscient Ones, the origin of all
Sravakas and Pratyekabuddhas, the accumulation of the merit and wisdom
of all Mahabodhisattvas, the letter of ultimate reality, the cause of all
letters",86 the letter or sound whose enunciation precedes NIafijuSrI's
declaration in NS 27 that he is in the heart of the Buddhas of the three times.
The NS description of MafijuSrIjfianasattva as the Knowledge-Body of
all the Tathagatas, and as himself "born from the letter JY.', is hereby
translated into a ritually visualised structure. The structure also implies that
Mafijusrljfianasattva is a more fundamental figure than the Adibuddha
since he is in the Adibuddha's heart. Given the Adibuddha is one of the
Names rather than the figure being Named, this again is consistent with
the text of the NS.
.
126
Indian Insights: Buddhism, Brahmanism. and Bhakti
With Mafijusrijfianasattva at its heart the mar;u;lala embodies the

understanding that the Names of the NS are those ofMafijusrtjfianasattva.
Vilasavajra treats the Names as mantras: they become the "Namemantras" of the title of his commentary. Some of the Names become the
mantras of particular deities, so that the Names of the fifth section of the
NS, for example, become the Name-mantras of the deities of the
Vajradhatumar;H;lala, which take their places on the seats visualised for
them. Indeed, all the Names, construed as mantras of different
'embodiments' of Non-dual Awareness, are assigned a place in or around
the Vajradhatumar;l<;lala, thereby creating a single unified structure for the
whole NS.,
Conclusion
I have tried show that the depiction ofManjusrt in the NS is not primarily
one of him as a Buddha or even as the Adibuddha. Though he is so
portrayed, such depictions are only possible insofar as he is identified as the
jfiiina that they embody. Thus, the Names of the NS are not
straightforwardly the names of Mafijusri. I have suggested that this is
indicated in the full title of the work, which should be read as "The NameChanting associated with the Knowledge-being Mafijusri". It is this
portrayal that is translated into the sphere of method by Vilasavajra.
Abbreviations
Cecil Bendall, Catalogue of the Buddhist Sanskrit Manuscripts in the
University Library, Cambridge (Cambridge, 1883).
NlVIAA Aryaniimasa7{lgftiflkii Niimamantriirthiivalokinf ofVilasavajra.
NMAA text references are to Tribe, 1994. 'NMAA 4.123'
means 'NMAA chapter 4, line 123'. In addition, a folio
reference is usually given to Bendall Add. 1708 (identified by
the siglum 'N.), a 15th century palm-leaf manuscript of the
NMAA.
NS
Niimasa7{lgfti. 'NS 20' means 'NS, verse 20'. Unless otherwise
specified, text cited follows NS.Dav.
NS.Dav. Text of the NS as published in Davidson, 1981. NS.Dav. 67, 3
means NS.Dav., page 67, line 3.
Toh.
Hakuju Ui et al., A Complete Catalogue oj the Tibetan Buddhist
Canons (Sendai: Tohoku Imperial Univers~ty, 1934).
Bendall
Mafijufri and The Chanting qfNames' (NiimasaT[lgfti)
127
Phoneticizations
Derge
Kanjur
Tanjur
sDe dge
bKa' 'gyur
bsTan 'gyur
Notes
Note. This article is a revised version of a paper originally given at The Eighteenth
Indian Religions, Oxford, 1992.
1.
2.
3
4
6.
7
8.
9
~ymposium
on
The most comprehensive study of Maiijusri in (non-tantric) Mahayana literature is Etienne

Lamotte's monograph, 'Maiijusri', Ig60. See also,.Williams, 1989, pp. 238-41; Akira, 199o,
PP29 1-3
.
Maiijusri's role as Sakyamuni's interlocutor (e.g. in chapters 12 and 14 of the Lotus sutra)
often shades into that of being his chief spokesman (e.g. in the V't1TlalaJcUti.nirdeia, the SaptaJatilca
PrWiiaparamita and the Maf{juJ7fbudd/zak,et:ragUTJlluyulzasutra). An important source for the role of
kafyii1J.amitra is the Ga1J.r/auyUhasutra. Other sources include the Ajiitaiatruriijasutra - whiCh
describes Mafijusri as the spiritual friend of the bodhisattvas - and the Drumaki",narariijapariprcclzii, in which AjataSatru is told he has the great advantage of having Maiijusri as a
spiritual friend (Lamotte, op. cit., p. 95). Chapter 12 of the Lotus sutra provides an example of
Maiijusrl's ability in converting beings to the Dharma, where he is said to have converted
innu~erable numbers of them after visiting the underwater palace of Sagara, king of the
niiga.l'. The Ratnakiira1J.r!auywzasutra describes how Manjusri converts some of the followers of
the Jain teacher Satyaka Nirgranthaputra (Lamotte, 1960, p. 39). Two text, that promote
Maiijusri as an object of devotion and meditation are the Mafijufribuddlzakjetragu1J.Quywwsutra
and the Maf{juJ'rfparinirviiJJl1siltra.
See Lalou, 1930, pp. 66,0. For a discussion of Lalou's suggestions regarding the origins of
the figure ofMaiijusri see Tribe, forthcoming.
MaiijuSri is said to be a Buddha of the present in the Aizgulimiilfyasiltra, though not of our
universe, but one named Nityapramudita, ~ways Happy' (Lamotte, 1960., p. 29).
See Davidson, Ig81; Wayman, 1985. Davidson includes a collation (rather than an edition,
since he does not consult any new manuscripts) of the Sanskrit text. He collates the three
editions ofMinaev, Vira and Mukherji (see note following), noting their variant readings. For
the annotated translation Davidson consulted four early commentaries in their Tibetan
translation, including the NMAA ofVilasavajra. There i, also a very useful introduction. It is
by far the best piece of work published on the NS.
Wayman provides a Sanskrit text but it is not really a separate edition since he reprints
Minaev's text with a very few emendations. The translation is accompanied by annotations
that draw principally on four commentaries (different from those consulted by Davidson),
again in their Tibetan translations.
See Minaev, 1887, pp. 13,59; Davidson (lg81, p. 16) report, that Minaev had intended a
translation of the NS but that hi, death in 1890 prevented it, appearance. In the 1960s two
further editions were publi'hed, neither of which, however, made use ofMinaev's edition: see
Vira, 1962; Mukhe~ji, 1963. The Sanskrit text of Vira's edition, which is of the NS verses
only, wa, reprinted by Vira and Lokesh Candra, (see bibliography under 'Vira'). Mukherji
consults neither Minaev's nor Vira's edition, and since he only had one manuscript
containing part of the anuiaT!'Sii he retranslates the missing part into Sanskrit from the
Tibetan.
On the author and date of the NMAA, see Tribe, 1994, pp. 9-23.
For a discussion of the relationship between the terms Vajrayana, Mahayana (and
Sravakayana) in contemporary Newar Buddhism see Gellner, 1992, pp. 108-14.
maiziimantra1U!)!odblzilto malziimantranuyiitmaIcaJ.z (NS 42cd).
lqiprasidd/zir mantramuldzacaryaciiriTJ.iim bod/zisatlviinii,!, (NS.Dav. 63, 4).
128
10.
II.
12.
'3.
'f.
'5.
,6.
17.
18.
Ig.
20.
2 I.

acirad eva vajrapiilJii vajradhara paramiirthaniimasalJ'lgztisa/'(ldhiirakapuru,apulJ'lgava(z susalJ'lbhrtapU1J:)iqjiiiinasa7[Zbhiirc0 kP-prataraT[l buddhagu~iin samudiinfyiinuttariilJ'l samyaksalJ'lbodhim abhisa7[Zbhotsyate
(NS.Dav. 68,4-6). Citing this passage the NMAA (5.287-9) has the variants sarvabuddhagu~iin,
"all Buddha-qualities", and paramiirthaniimasa7[ZgZtisa7[Zdhiiraka1:z puru,apu7[Zgava(z. Both of these
seem better readings.
Despite possessing a tantra-like frame the.NS does not describe itself as a tantra but as a
SalJ'lgZti, a chanting text. However, the colophon to the NS describes it as "ocurring in the
Samadhijala chapter of the 16,000 verse Mahayogatantra called the Aryamiiyqjiiia". Also, in
the opening verses, Vajradhara says he wishes to hear the NS that was "proclaimed in the
great tantra [called) the Miiyiijiiia by immeasurable joyful great mantra-bearing Vajradharas"
(NS '3) so that he can obtain "The [method) of Awakening According to the Miiyqjiiia" (NS
7). There does survive a kliiyqjiilatantra in Tibetan translation (Tah. 466) but not in 16,000
verses, and its Samadhi chapter does not contain the NS. The relationship, if any, between
the two works awaits investigation. References to the kliiyiijiiiatantra in the NS only occur in
the frame text. If these sections are indeed a later addition then the references could have
been included as a way of legitimating the text and giving it a canonical status. Bu-ston
classified the NS as a tantra placing it at the head of the tantra section of the Kanjur,
immediately preceding the Kiilacakratantra with which it had come to be associated. However,
some manuscript editions of the Kanjur, e.g. the sTog Palace, place the NS before the
Miiyiijiiiatantra (See Skorupski, Ig85, p. 22g). Also, in line with the statement in its colophon,
the Tibetan rNying rna canon classifies the NS as a Mahayogatantra. See Kaneko, Ig82, no.
Ig6. Aside from questions concerning its classification, however, the NS is clearly a tantric
work and was treated as such by its many Indian commentators.
Davidson, Ig81, p. 3.
E.g. see Mafijusnmitra's MafijuJ'riniimasalJ'lgZryupadefii (Jam dpal gyi mtshan gdon pa'i man ngag:
Tah. 2555), which Davidson himself translates (lg81, pp. 45-7) from the Tibetan.
Vilasavajra states that the NS consists of 162 verses plus '50 verse equivalents for the
anuialJ'lSii (NMAA 1.43-44; A.2v5--fi). Nonetheless, he is aware that the mantravirryiisa and
upasalJ'lhiira follow the anuialJ'lSii. In the immediately preceding passage they are enumerated
as the final two sections of the NS, and the commentary has a seperate chapter devoted to
each. It seems clear that when he assigns '50 verses to the anufalJ'lSii he is referring to the
whole of what follows verses 1-162.
See Davidson, '98" p. 47, note 149
Bajracarya Dibyabajra informed me (Jan. 1992) that the anufa7[Zsii used to be chanted though it was not clear whether this was within his lifetime - but that more recently it was felt
that there was no need to include the anufalJ'lSii as it described the benefits of the NS and was
thus not strictly parl of it. He argued that the full benefits could still be obtained without
chanting the anUJ'alJ'lSii, and, indeed, modern Nepalese editions of the NS omit it.
Davidson argues (1981, p. 47, note 149) that Mafijusnmitra and Vilasavajra's treatment of
the anuialJ'lSii, mantravi1!Jiiisa and upasalJ'l/zara sections as being different from the remainder of
the NS suggests that they arose as a unit, later than the earlier verses, and that this supports
his theory concerning the stratification and development of the text. However, his hypothesis is that NS 26-162 form the earliest stratum, not NS 1-162.
Elements in the NS, such as the presence of wrathful forms embodying wisdom (see NS
66-76), suggest it is not a straightforwardly Yogatantric work. This is supported by the
colophon's statement that it is part of a Mahayogatantra. The Tibetan rNying rna pa canon
treats Mahayogatantras as a distinct class and includes the NS within it (see above, note II).
Ivlahayogatantras share characteristics with works such as the Gu/zyasamqjatantra, elsewhere
classified as Yogottaratantras, and in general, they may be seen as transitional between
Yogatantras and the later Yoginltantras that employ the full kiipiilika symbolism of the
cremation ground.
Davidson, 1981 , p. 5
As with the NS (see above, note 18), the NIvIAA contains a number of elements that
characterise it as part of the transitional period between the Yoga- and Mahayogatantras. It
also contains citations of the proto-Yoginltantra, the Sarvabuddhasamiiyogaiikinyiiiasamvara (Tah
366-7).
The difference of interpretation is reflected in the placing of N.S-related works in the Tanjur.
In the Derge edition six are assigned to the Kalacakra division (Tah. 1395-1400), thirty-two
MafijuJrf and 'The Chanting qfNames' (JViimasaTflgjti)
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
129
to that of the Anuttaratantras (Tah. 2090-2121), and ninety-one to the Yogatantra section
(Tah. 2S32-2622). As Davidson notes some of these placings are soinewhat arbitrary and do
not accurately reflect the texts' actual perspectives (1981, p.IS).
This work, the AryamaiijufnniimasmpgZtipkii, is 187 folios in the Derge edition of the Tanjur.
See M. E. Carelli, ea., Sekoddefafikii, Gaekwad's Oriental Series no. 90 (Baroda: 1941). Naropa
also used the NS extensively in his Hevojratantra commentary, the Vojrapadasiirasa1[lgrahapaiijikii.
Wayman (198S, pp. IO-22) has traced S3 citations or incorporations of the NS in this work
An edition of the Am[laka7Jikii and the Am[laka7Jikoddyotana by Dr. Lal BaJ1arsi has heen recently
published (date unlmown) in the Bibliotheca Indo-Tibetica Series oftl1e Central Institute of
Higher Tibetan Studies, Sarnatl1. For an edition and English translation of chapters I-S of
the NJ'vL<\.<\, see Tribe, '994. Chapters 3 and 4 of the NMAA have also been edited by
Sakurai (1988). The Curflwpadii survives, as far as I know, in a single palm-leaf manuscript (see
Cowell and Eggeling, 1876, pp. 2S-6).
See Shastri, 1927, p. 4, line 5.
Translation taken fr"Jm Chattopadhyaya, 1990, p. 203. The "hymn of praise in 150 verses"
must be Matrceta's Satapanciiiatkastotra. Bu ston's more abbreviated account can be found in
Obermiller, 1932 (see pp. 132-3 in 1986 edition).
See Lalou, 1953, p. 329, note 438. Giuseppe Tucci has argued that the !Dan dkar catalogue
was subject to revisions and additions with the result that it must be dated to later than the
reign ofKhri srong Ide btsan. He suggests a figure ofBI2 CE (see Tucci, 1958, p. 46, note 1).
See de La Vallee Poussin, 1962, pp. 43, 126.
I use the word 'Knowledge' to render jniina here. This term is famously difficult to translate:
on the whole I prefer 'Awareness' in that it emphasisesjnana's experiential or subjective pole,
but 'Awareness-Being' seemed even less satisfactory that 'Knowledge-Being'. However, for
advayojniina and the panea jniiniini I have consistently used 'Awareness' (thus, 'Non-dual
Awareness' and 'the Five Awarenesses'). Elsewhere, and outside of these more technical uses
ofjniina, I have also relied on 'wisdom' as a translation.
These epithets all occur in NS IO, which the NMAA reads as, bhagavan jiiiinakiiyasya
mallOPMaJya sNpate(z / maiijufTf/niinasattvasya jiiiinamurte~ svaya,!,blzuva(z / /, treating bhagavan as a
vocative that opens a new sentence. NS.Dav. takes the first two words as compounded, "the
Knowledge-Body of the Fortunate One (blzagavojjiiiinakiiyasya)".
sarvatathiigat'ijniinakiiyasya mafijufi[jiiiinasattv(]JyiivB7Jikapari:Iuddhii niim(]Jmp,,",~ (NS.Dav. 61, 21-22
[anusa'!'sii]); Magavato mafijufrfjniinasattvasya sarvatathiigatajniinakayasya jiiiinamurter advayaparamiirthii1[l niimasa1[lgZti,!, (NS.Dav. 65, 3-4 [anufa1psii]); mafijufrf}fiiinasativasya paramiirtizii
niimasaTfltitiJ.z (NS.Dav. 69, 9-10 [colophonJ).
blzagavato mafijufrfjiiiinasattvasya riipam iilambayann (NS.Dav. 66, 23 [anufa1JlSii]).
malliirtizii1[l niimasa,!,giti,!, pavitriim aglzaniiJanz,!, /
maf[jufrijiiiinakayas)'a matta~ ,'rolU'!' samudyataf,z / I (NS 21)
See note 31 for the two anufa1[lsii citations.
This accords with Vilasavajra's analysis of the compound: ki,!,viJi,!asYajnanakiiyasyelijfiiinam eva
kiiyojiiiinakiiyas tasya (NJ'v~ 1.203-4; A.7rS).
janakab sarvabuddhiinii,!, (NS 60a)
buddlzaputrab para vara~ / (NS 60b). Again this (provi,ionally) assumes that the predicates of the
NS core text qualify Maiijusrtjiianasattva as their subject. For discussion of this issue see the
section following.
jniinaJ'attva iii / sarvatatlziigaialzTdayavihiiritviit / maf!jufTiJ' casau jiiiinasattvai ati mafijufrfjniinaJ'atlva!z /
naya,!, daJablziimi:fvaro bodhisatlva~ ki1[l larhy advOY'ijniina,!, prqjniipiiramitii saiva mafijufrij'fiiinasattva~
/ / ala eviilza digniigapiidafz / praJ'fiiipiiramitiijfiiinam advoya,!, sii tathiigata iii 1/ (NMAA 1.215-220;
A.7v4-6). The citation is the opening verse of Dignaga's Yogacara Perfection of Wisdom
commentary, the Projniipiiramitiipi7Jrfiirtha. This reflects Vilasavajra's own largely Yogacara
perspective, which in my view mirrors that of the NS more accurately than that of the
Madhyamaka.
In distinction to thejniinasattva, the samoyasattva is regarded as the purified receptacle or basis
for the arising ofliberating wisdom. One of the self-characterisations of the Mantrayana is in
terms of a contrast between itself as the "The 'Nay of the Result" (phalayiina) as against "The
Way of the Cause" (jzelUyiina), typified by the slura-based Way of the Perfections (piiramitiinoya).
'Nhat this means is that the mantrin brings the result (or 'effect') of the path, namely,
Buddhal100d in the form of a visualised deity, into the path as a factor in its development
130
40.
41.

through a magical identification with it, thereby accelerating progress towards the goal. The
non-tiintrika, on the other hand, progresses in a more slow, graded, step by step manner,
maturing the causes that will result in his or her attainment of enlightenme,nt.'Having
created the sam'9'asatlIJa, therefore, the mantrin effects a ritual identification with the goal by
invoking thejfiiinasatlIJa who descends and merges with the satrU!YasatlIJa. Tsong kha pa discusses
the notion of lzetu- and plzal'9'iina in his sNags rim chem mo, citing Sraddhakaravarman's
YogiinuttaratantriirtlziivatiirasaT/Walza (Toh. 3713) as a source. For an English translation, see
Hopkins, 19n pp. 105-ra7.
I do not know of any research on the development or role of the pairing ofjfiiinasatlIJa and
Jam'9'asatlIJa. mKhas grub rje, in his rGyud sde spyi'i mam par gzhag pa rgyas par brjod (Fundamentals
if Ilze Buddhist Tantras), has the following to say in the chapter on Yogatantra:
One generates the Symbolic Being (sam'9'as'atlIJa) and draws in the Knowledge Being
IjfiiinasatlIJa), then applies the seals of the four Seals, but not if there LI only the Symbolic
Being or only the Knowledge Being. The purpose of executing the seals of the four Seals
is to merge and unify the Body, Speech, Mind and Acts of the Knowledge Being with the
body, speech, mind and actl of the Symbolic Being. There would be no foundation for
merger if either were present by itself. This is comparable to having both Self Generation
and Generation in Front. (Quoted from Lessing and Wayman, 1980, p. 235).
ThL, suggests that the samayasatlIJa-;jiiiinasatlIJa distinction was in use in the Yogatantras, though
mKhas grub rje cites the explanatory Tantras, the Paramiidya and the Vqjras'e/dzara, rather
than the root Yogatantra, the Sarvatat/ziigatatatlIJasarrzgraha, a, confirming the picture he gives of
the relation between the seals (mudrit) and the satlIJas. In the NMAA Vilasavajra cites both the
Paramiidyatantra and VqjraSeklzaratantra, which suggests he wa, familiar with their content.
However, mKhas grub rje's citation of these works is not any indicator of how full a
treatment they give to the subject. The question of whether Vilasavajra wa, aware of' the
notion of drawing the jiiiinasatlIJa into the sam'9'asatlIJa, hut choose to elahorate his siidhana
differently, or whether his approach contributed to the development of that notion requires
further investigation.
Davidson (1981) neither di,cusses, the title of the NS nor the L,sue of whom the NS Names
name.
42.
43.
44.
45.
46.
47.
48.
ThL, title is found within the titles of the secondary works, e.g. the Atyamaf!jUJ7iniimas'arrzgftifi/cii
of Manjusnkirti (foh. 2534). These texts survive only in Tibetan translation, however, the
original Sanskrit versions being lost. As a result, the titles appear at the beginning of the
translation~ (and sometimes in colophons or index volumes), either transliterated into
Tibetan script or in translation (or both). See al,o Toh. 2537, 2098, 1396, 2536, 1399, 2539,
2093, 1398, 2535, 2090.
I have discounted the possibility that, in the title phrase maiijUJ'rijiinasatlIJa.rya paramiirthii
niimasl11[lgfti/.~ niimasa1'{lgiti/.z L, a siipelqasamiisa, i.e. that the genitive mafijuJrfjiinas'atlIJa.rya qualifies
niima rather than niimasa1'{lgiti/.L Given that the term niimasarrzgfti/.z often appears on its own in
the NS and that maffjuffiniimas"QTrzgfti/.z never appears such an interpretation seems unlikely.
laJya maftiu.l7fjfiiinasatlIJa.rya paiicqjfiiiniitmakflJ)'a sarrzbandlzi.niyii niimasa1'{lgitis tiiT[Z ciiharrz dlziir'9'wiimiti
(NMAA 1.221-222; A.7v6).
a iii fu il e ai 0 au I11[l a{zstlzito hrdi /
jfiiinamilrtir a/zarrz buddho buddlziiniirrz t!jadlzvavartiniim / /
orrz vqjrafi1cpJ.a du{rkhaccheda prqjiiijjfiiinamiirtq)le /
jiiiinakiiya va"uiJ'vara arapacaniiya te nama{z / / (NS 26-27).
atlza s'ii!gamunir bhagaviin sakalarrz mantrakularrz mahal /
mantravidyiidharakularrz vyavalo!ga kulatr'9'am / /
lokalokottarakulam lokiilokakulam malzat /
malziimudriikula~ ciigryarrz malz~PMakularrz malzat / /
imiirrz :ra1JlTlantr~riijiinasaTflyuktiim adv'9'od'9'iim /
anutpiidadlzarmir.Urrz giitlziirrz blzii1ate sma gi:riirrz pate& / / (NS 23-5).
giriiT[! pater maf[jufrfjiiiinasatlIJasya sarrzbandhinf yiinutpiidadharmi7]i giithii tiim imiiTfl bhiitate (NMAA
3.5,; A.rrr2-3)
ThL, verse remains obscure to me. Vilasavajra divides it into six mant~as, which he takes a,
those of six primary manifestations of Mafijusqjfili.nalattva. For a series of visualisations
centred on these figures, see NMAA 4.201-253. See also, Davidson, 1981, p. 22, note 63.
Maf[jufri and 'The Chanting qfNames' (NiimasarJ7gfti)

49.
50.
.5!.
52.
53.
54.
55.
56.
131
NS 28, however, can be read otherwise. See the section on Mafijusn as the letter A, below.
tadyalM. could be interpreted as governing just NS 28, or NS 28-g, but this makes the status
of the remaining verses very problematic.
Although Mafijusnmitra's NamasaTJ1gitivrtti (Toh. 2532) deals with just NS 1-162, his
Maffju.l'nniimasa7J1g'ff!) lupadefa (Jam dpalgyi mtshangdon pa'i man ngag, Toh. 2555) refers also to the
anu.l'aTflSii and mantravinyiisa.
niimiiniyogakriyiicaryiitantrapravacanasfltrantabhidhmmavinCIJalauk.ikalokottarii.~i sarvastl,avar,gaitgamCini
ca / tc{iirrz niimnaTJ1 saTJ1g'flir iti (NMAA !.58-60; A.3r4-5): Vilasavajra is exploiting the
Abhidharma sense of nama as a linguistic unit of greater length than a word here, and aL~o
the sense of 'things' as nameables. See Pruden, 1988, vol. I, p. 250ff.
gaunyii sll'i)iii na paramiirt/,atal., s!!TJ11J!1Yii tu / niimamiitram idaTJ1 sarvam a bhaviigrapa:ricchedam iti 1!Yi9'iiI
(NMAA 160-61; A.3r5---{)).
Wayman (1985, p. 9) takes the view that the Names are expressive of the ultimate truth
(paramiirllzasa)ia) about Mafijusn, and that the glosses and further identifications made by
commentators represent the level of conventional truth. Some of the verses of the NS might
suggest this interpretation, but for many it is not really tenable. Is a purified Bhilqu or a
Pratyekabuddha Mafijusn's ultimate nature? Surely it is the .other way round. Wayman's
analysis is based on his reading the full title of the NS as containing the single word
paramiirtlzaniimasa7J1!fitily. in which he takes paramiirtha as qualifying niima. As a result Wayman is
committed to explaining the Names as ultimate. In fact the title, as given in the colophon of
all the MSS that have been edited, contains the two words paramarthii niimasa7J1!fitily.. Here it is
clearly the NS, rather than niima, that is qualified as paramiirtha. The single word
paramiirll,aniimasa7J1g'fti/.' does occur in the anu.l'arrzsii, where, on the basis of the reading in the
colophon, as well as on the content of the text, it makes most sense to take paramiirtha as
qualifYing niimasa7J1g'fli, rather than niima.
tadyall'ii bhagavCin budd/,aJ., s!!TJ1buddho 'kiiras!!TJ1bhava i)i evamiidibhir niimamantriilqarapadair . .. vife,yo
JiiiinasattvaJ., sarvalathiigatahrdCIJavihiir'f maffju.l'rf/.z frimatiirrz vara iii dra{fa'!YaIJ (NMAA 5.26g---<!72;
A.29v4-6): "It should be understood that 'Mafijusn, the best of the glorious', the KnowledgeBeing who dwells in the heart of all the Tathagatas, is qualified [in the Namasa7J1g'ftz] by
phrases, whose words are Name-mantras ... , the first of which is 'That is to say, [he is] the
Fortunate One, tJ:ie Awakened One, the Fully Awakened One, born from the letter N.."
sarvasYlTJ1patlcaraJ., inmiin maffjuJ7i{' ir'fmatiirrz vara/, / / (NS 157cd). The second piida of this line
could be rendered as, "Mafijusn is the best of the glorious". Although this might appear
more appropriate, it would treat the piida in a different way from what is apparently the
general rule lor most of the core verses of the NS, where each piida is itself aN arne.
Lokesh Candra's article, 'Nama-sailgiti is a Hymn of Advaya Names', (1987), came to my
attention after I had formed these conclusions concerning the interpretation of the titles and
Names of the NS.
Candra notices the ambiguity of the expression Maffju.l'r'fniimasa7J1g'fti and similarly argues that
it should be read as 'the NS of Mafijusn'. His conclusion, however, is that Maiijusn is no
more than a figure in the transmission of the NS, ie. that Mafijusn recites the NS, and argues
that the Names are advCIJa names ofVairocana (or 'Mahavairocana', Candra seems not to
settle for one in particular) and the Awarenesses Uiiiiniinz) of the Tathagatas of the
. Vajradhatumru;u;lala. His argument largely rests on interpreting the term advCIJaparamiirllzii found qualifYing the NS in the colophon of a few Sanskrit manuscripts (e.g. Bendall, Add.
1323) - as meaning 'the essence of the Yoga tantras'. The NS colophons generally read
paramiirt/zii (NS.Dav. 69 records advayaparamiirthiijust once) though advqyaparamiirtha could have
been imported under the influence of its appearance, unnoticed by Candra, in the anu.l'arrzsa
(see NS.Dav. 65, 4). In any case, advCIJaparamiirthii is more straightforwardly understood in the
sense of "having the Non-dual as its ultimate meaning" . I remain unconvinced that the term
adv'!ya has a technical usage. whereby it refers to the Yogatantras, and to the
SarvatallziigatatattoasaTfWaI,a in particular, as Candra maintains. In general, Candra fails to take
account of the content of the NS, particularly NS 25" and relies, rather uncritically, on
colophons and titles of the NS and its commentaries. For instance, alter noting the existence
of the NMAA he comments (p. .t81) that "Mafijusn is not an integral part of the title and the
commentary explains the Meaning of the names and mantras". It may be true that the word
'Maiijusn' is not part of the Nl\tIMs title, but this does not prevent Mafijusnjfianasattva
being the central figure of the NMAA who is qualified by the Name-mantras - not 'names

and mantras' - (niimamantrii-) whose meaning (-iirthii-) is to be explained (-iivalokini), according
to the commentary's title. Nonetheless, Candra's conclusions are not so distant from mine or,
indeed, Vilasavajra's. The Names of the NS appear to predicate a figure that underlies all
forms and embodiments of wisdom and whereas I believe there are good reasons for
identifying this figure as Mafijusrljfianasattva, Candra takes the figure to be
(Maha)vairocana. Although he takes the frame text to be a later stratum Candra has to refer
to it in order to argue his position. If this is the ca,e, he then needs to account for the other
frame text material that points to MafijusrUfianasattva as being the figure Named.
Furthermore, if it is (Maha)vairocana who i, Named, the inclusion of both Mahavairocana
(NS 42) and Vairocana (NS 62) as Names requires explanation. Both Candra and Vilasavajra
agree that, in effect, it isjiiiina that is Named, though Candra appears to. follow Wayman in
taking the Names themselves to be on the level of ultimate truth (po:ramiirtlzasatya).
57 An important dimension to the 'personality' of Mafijusli, which is beyond the scope of the
present discussion, is his association with speech and language, as seen in many of his other
names and epithets, e.g. Mafijugho~a, VligISvara, Mafijusvara.
58. jiiiinam advCl)iarilpadlzrk (NS 99b).
59 advCl)io dVCl)iaviidi ca blziltakopi1!)lavasthilal; (NS 47ab): "Non-dual; teacher of duality; established at
the limit of reality". The reading dVCl)iaviidi follows the gloss of the NMAA; NS.Dav. ha,
'dvCl)iaviidi, "teacher of non-duality".
60. For the Skt. of NS 26, see note 45. The interpretation of advCl)iodCl)iiim[... giilhiimJ is not
problem-free. The NMAA appears to construe it as a locative bahuvr'ihi compound, giving "in
which there is the source of the Non-dual" (NMAA4-123-4).
advCl)iaparamiirthiiT[l niimasa1JltiT[l (NS.Dav. 65, 4)
mf9iiijiila namas tubh,yaT[! namas Ie buddlzanii{aka /

namas Ie J"o:rva so:rvebh,yojiiiinakf9ia namo 'J"tu Ie / / (NS 162).
67
For the Ski:. ofNS 27 see note 45.

Wayman (1985, p. 97) cites this comment from Narendrak'irti's Maf{jufrinamasar,lffiti1!)liikhyana
(Toh. 1397).
jiiiinamiirtis tathagataIJ. (NS 100d).
For the text of NS 26-27 see note 45.
aniidinidlzano buddha iidibuddha niranvCl)ia~ / (NS 100ab). The predicate 'without association'
(niranvCl)ia) appears to indicate that the term iidibuddha is understood as refering to something
.
beyond the sphere of causation.
sVCl)iaT[lblzil appears a, a Name in NS 61Cd: "Born from the sky, Self-Arisen" (gaganodbhava~
sVCl)iaT[lbh~).
68.
69
72 .
73
paiicqjiiiiniitmako vibh~ (NS 59b).

These are the five Tathagatas of the VajradhatumaQc;lala of the Sarvalath1igalatattvasarpgraha
[tantra], the fundamental work of the Yogatantra class. The correspondences with the Five
Awarenesses are not made in the tantra. They are present in the commentaries of
Sakyamitra (KosaliilaT[lkiira, Toh 2503) and Anandagarbha (Taltviiloka, Toh. 2510), though
these authors may be placed in the 10th century CE, being, therefore, considerably later than
Vilasavajra who also makes the correlations (see NMAA i.1g-16; A.1V5-6). On the
development of the five Tathagatas, see Yoritomi, 1990, pp. 693/16 (English text); for two
diagrams of the VajradhatumaQc;lala see Snellgrove, 1987, pp. 209-213.
For a discussion of the role of the four Awarenesses in the Y ogacara, see de La Vallee Poussin,
1928-9, pp. 681-g2.
The importance of a five/old structure is emphasised by all the Names ofNS 59: '~wakened
One, with five body nature; Lord who has the Five Awarenesses as his nature; with a crown
that has the five Buddhas as its nature; with five eyes, unattached" (paiicakf9iiitmako buddha~
paiicqjn1iniitmako vib!z~ / pancabuddhiitmamakufa/.z pancacak,ur aJ"angadhrk / /). Note the proximity
of the Five Buddha, and Five Awarenesses. However, no identification between them is made
explicit.
See Table I for a summary ofthi, information.
Hence, the shortest of Perfection of Wisdom sUtras, "The Perfection of Wisdom in One
Letter", which, though lost in Sanskrit, is preserved in Tibetan translation (beom ldan 'das ma
J"hes rab ~'yi phil. rol tu pl!vin pa de b;:lzingshegs pa tlzam.r cad Igi]um]igegcig ma ;jzes qya ba, Toh. 23).
For an English translation, see Conze, 1973, p. 201. Another tantric Perfection of Wisdom
sUtra, the Perfection of Wisdom in 150 lines (Adl!yardlza.fatika prqjii1ipiiramitii), associates Mafijusli
Maffjufrf and 'The Chanting qfNames' (NiimasalJ1.gr:ti)
I33
with the teaching of the letter A as the essence of the Perfection of Wisdom (see ibid., p. 188) (I
am grateful to Dr. Paul vVilliams for drawing this to my attention).
74. tadyatlzii blzagaviin buddlzal,z J'aT[lbuddlzo 'kiira.raT[lbizaval,z / likiiral,l J'arvavarr;4gryo mailartJzal,l paramiik,rarail
/ / (NS 28),
The translation I have given follows Vilasavajra (see NMAA 5.269-272; A.2gv4-6, partially
cited in note 54). Other.1 are clearly possible, for example, "That is to say, the Fortunate One,
the Awakened One, the Fully Awakened One, is born fi'om the letter PI', taking akiira.raT[lbizavaJ.1
as the predicate, being a more natural reading of the first half~verse. Thus, Wayman (op. cit.,
p, 68): '1'.ccordingly, is the Buddha, Bhagavat, the Sambuddha arisen from ;';.'. Davidson (op.
cit., p. 22) takes the two half verses together with akiiraJ.! as the predicate: '1'.nd in this way, the
blessed one, the Buddha [11aiijusli], the completely awakened, born from the syllable A, is the
syllable A, the foremost of all phonemes, of great meaning, the supreme syllable".
Vilasavajra's interpretation, however, has the advantage of making the whole of the first half~
verse into a series of predicates (and therefore Names) of the unstated subject,
Maiijusrjjiianasattva, in line with the notion that NS 28-IS7 consists of a series of predicate
Names. The second half-verse has to be understood, in thil view, as parenthetical.
Compare, also, the mantra, OT[l akfjro muklwTf! J'arvadlwrmii~am atfyanutpannatuat, "The sound 'No is
the first of all dharmas since it is unarisen from the beginning". In the NMAA it is cited in the
context of instructions for developing the Discriminating Awareness (pratyavek,ra1Jajiiiina)
(Nl'vlAA 446; A. 131'4). A useful discussion of the letter A and its role in Yogatantra literature is
found under the entry 'No in Demieville, Ig2g. Allied to the idea that all sounds emerge out of
the vowel A is the view, probably first elaborated by the Mahasa1J1ghikas and found in a
number of Mahayana sUlras, that the Buddha taught by expressing just one sound that was
interpreted by each hearer according to their individual spiritual needs. On this see Lamotte,
1962, chap. I, note !}'2.
7S. See Nl'vlAA 4.130-146 (A. Isv4-16r3), which contains a citation from the V,!jrama1'){iiiiaT[lkiira-tantra.
76. l'vEtra, Ig81, p. 2S5. Neither this account nor the NMAA say anything about why there are
twelve, rather than sixteen or four, vowels, Such as it is, the explanation given in the SvayaT[lbhii
Purii~ does not conflict with that of the NMAA, though by implication Vilasavajra goes
further so as to include all mundane objects aB product, of the vowels.
n The siidhana evolves through chapters 4-10 of Vilasavajra's commentary, with the whole of
chapter 4 being devoted to the establishment of the basic ma"c;Jala and central deity. See
NMAA 4.1-258 (A.gr6-IST7, though the latter folios of chapter 4 are lost in Bendall Add,
1708). The following ac'count is necessarily highly condensed.
78. yo. imiiTf! nama.raTfl!fitim nama cii#ma~iT[l pratyalzam ... tril,zfr.rtua ka1J!hagatiim iivartayi,yati / PlL'itakagataT[l
va pathamiinai! pravartayi,yati / Magavato maiijufrijiiiinasattuasya rnpam iiiambayann anuvicintayaT[ls
tatriipam anudlryZiyan / tam eva rnpakiiyeniicirad eva . .. drak[yati / gaganataiagatiil(!f ca sarvabuddhabodhisattuan nananirmii~rnpakiiyaif,! salwgatiin dralqyati / (NS.Dav. 66, 21-6)
79. See Nl'vlAA 4lg3-198 (A.I7VS-18r2).
80. The lvlaiijuJnparinirvii~iitra, the original Sanskrit of which has been lost, was translated into
Chinese by Nie Tao-tchen, the collaborator ofDharmarak,a, at the end of the third century
CEo It is a short work, less than four pages in Lamotte's French translation (Lamotte, '960,
PP3 6-9)
81.
82.
83.
Translated from Lamotte, Ig60, p. 38.

Beyer, Ig73, p. 2g.
The NMAA text depicts Mal1avairocana a little more fully, giving' an explanation (viiuddlz!) of
his attributes: '1'.nd then, on the principal seat, [he should visualise] Mahavairocana,
generated by means of the syllable AI:L [Why has he Jour face.I?] Because consciousness which is of the nature of the Dharma-Sphere since, by its nature, it lacks such forms as the
grasped - i.1 [our-faced, since the four faces that are the freedoms (vimokra) of emptiness and the
rest, are the avenues through which all samiidizi-s arise since their [that L" the freedoms'] object
is the Dharma-Sphere. He is white in colour because he has the Dharma-Sphere as his
nature; he ha, braids of hair [stacked up on his head] as a crown and is unadorned, because
he is one whose mind is tranquil; [and] he makes the bodlryagrf gesture because he has both
wisdom and means as his nature." (tata.1 ca pradhiiniiJ'alla iiizkiirer;a parini,pannaT[l mahiivairocanam /
yato yad evacittaT[l prakrtya griilzakatfyakiiravirahiid dlzarmadhiitusvablziivaT[l tad eva caturmuklzaT[l
iiiayatiidicaturoimok,ramukiliinaT[l dlzarmadlziitor iilambanatuena sarvasamiidhipra.riitiJzetutuiit / fuklavarr;aT[l
I34
84.
85.
86.

dharmadhiitu.rvablzavai:vat / jatamakutopetalJ1 'nirablzara~am ca fantacittaluat / bodhyagnmudrayuktalJ1
pr'!iiiopayatmakaluat / / NMAA 4.100-105; Ar4V5-r5f2).
This is the longest of the Perfection of Wisdom sutras and as such came to be regarded as the
original fi'om which all shorter sutras were derived as condensations, The text provides a few
more details concerning the visualised Jorm of the Adibuddha: "[The five faces are] endowed
with five colours: dark blue in the east; yellow in the south; red in the west; green in the
north; on the top, he has a white face, theface of the most excellent horse. He is tranquil,
with the ornaments of a youth, possessing the erotic [sentiment, and] wearing about himself
a cloth of many colours." (paficavarr;opetalJ1 / piirver;a nflalJ1 dalqi7}na pltalJ1 pascimena raktam uttarer;a
lzaritalJ1 / miirdirni parama.roamuklzaoad aoaJtiritalJ1 muklzalJ1 foetalJ1 ~aJya talJ1> / santalJ1
kumarablzara~petalJ1 JaJ'rJigiiralJ1 oicitravaJ'traparidhiinam ... NMAA 4.109-II3; A '51'4-5).
The full instruction for the visualisation of Maiijusnjiianasattva runs as lollows: "[Next,] he
should visuali.se himself as the Fortunate One, the Knowledge-Being [1vIaiijUSli], born lrom
the letter A situated in the middle of that [wisdom-]wheel, with six laces, radiant like the
autumn moon, with the best of sapphires in his beautiful hail; with a halo that has the
brilliance of the orb of the newly risen sun, with all the Tathagatas as [head-] ornaments,
immersed in Jamadlzi, seated on a variagated lotus throne, with two books of the
Prajiiaparamita above blue lotuses held in his two hands, [and] as in the S'entiment of
Tranquillity." (ta.rya cakra.rya madiryaoartinam akarqjalJ1 blzagavantalJ1 !alJ17lukiralJ1jiiiinaJattvalJ1
.raracc/w.rankaprabiram indranflagraJaccfralJ1 baliirkama~r1alaccirayaprablzama~r1alalJ1 Jarvatatiragatamayabhara~1J1 JamiidhiramapannalJ1 vicitrapadmiisanopavietam ubhqyakariisaktanfiotpalordhvaJtJzitaprajiiaparamitiipustakadvayalJ1 fantaraJ'opetam iitmiinalJ1 vicintya NtvrAA 4.172-6; AI7r2-4).
prajiiaparamitiiJ'vablzii,oalJ1 JarvlljiiaJiianodayakiira~1J1 .I'arvasravakllpratyekabuddhanam utjJattibhiitalJ1
.I'arvamalziibodhil'atluanarp. pW!y,!jiiiinaJ'alJ1bhiirabhiitalJ1 paramarthCi.k]aralJ1 .I'arvak!arii~alJ1 kiira~bhiitam
[akaralJ1 oinyas~t] (NMAA 4.17]-180; Ar7r4-6).
Bibliography
Akira, Hirakawa. A History oj Indian Buddhismfrom Siikyamuni to Early
MahiiJiina, tr. and ed. Paul Groner. Hawaii: University of
Hawaii Press, I990.
Beyer, Stephan. The Cult oj Tiirii: Magic and Ritual in Tibet. Berkeley:
University of California Press, 1973.
Candra, Lokesh. 'Nama-sangHi is a Hymn of Advaya Names',
Annals of the Bhandarkar Oriental Research Institute, vol. 68, 1987,
pp. 177-194.
Carelli, M. E. (ed.) Sekoddefa{ikii. Gaekwad's Oriental Series, no. 90.
Baroda: Oriental Institute, 1941.
Chattopadhyaya, A. Tiiraniitha's History ofBuddhism in India. Simla,
1970 (repr. Delhi, I990).
Conze, Edward. The Short Prqjfiiipiiramitii Texts. London: Luzac, 1973.
Cowell, E. B. and Eggeling,J. 'Catalogue of Buddhist Sanskrit
Manuscripts in the possession of the Royal Asiatic Society
(Hodgson collection),' Journal ofRoyal Asiatic Society I876,
PP5-5 0 .
Davidson, Ronald M. 'The Litany of Names ofMafijusrl', in Tantrie
& Taoist Studies in Honour ofProfissor R. A. Stein, Vol. I. Melanges
Chinois et Bouddhiques, no. 20. Brussels, 1981, pp. 1-69.
Mafijufri and 'The Chanting ofNames' (NrimasaT(lglti)
135
Dernieville, PauL (ed.)H8biJgirin, Dictionnaire Encyclopedique du

Bouddhisme d'apres les sources Chin6ises etJaponaises, voL I. Tokyo:
Maison Franco:Japonaise, 1929.
Gellner, David. jlvionk, Householder and Taniric Priest: Newar Buddhism
and its Hierarchy if Ritual. Cambridge: Cambridge University
Press, 1992.
Hopkins, Jeffrey. Tantra in Tibet, The Great Exposition if Secret lvlantraVolume I. London: George Allen & Um0..n, 1977.
Kaneko, Eiichi. Kotantora zenshu kai mokuroku [A complete catalogue
ofthe rJYying ma rgyud 'bum]. Tokyo: Kokusho Kangyokai, 1982.
La Vallee Poussin, Louis de. Vijfiaptimritratrisiddhi: la siddhi de HiuanTsang, 2 vols. Paris: Libraire Orientaliste Paul Geuthner, 1928-9.
- - - . Catalogue of the Tibetan Manuscripts from Tun-Huang in the India
Office Library. London: Oxford University Press, 1962.
Lalou, M. Iconographie des etoifes peintes dans le Mafijufr;:mulakalpa. Paris:
Libraire Orientaliste Paul Geultiner, 1930.
- - - . 'Les textes bouddhiques au temps du Roi Khri-sron-Idebcan',JournalAsiatique CCXLI, 1953.
Lamotte, Etienne. 'Mafijusri', T'oung Pao, vol. 48, 1960, pp. 1-96.
- - - . (tr.) L'enseignement de Vimalaklrti (Vimalaklrtinirde.fa). Louvainla-Neuve: Institut Orientaliste de Louvain, 1962.
Lessing, F. D. and 'Wayman, A. (tr.) mKkas grub rje's Fundamentals of the
Buddhist Tantras. Indo-Iranian Monographs 8. The Hague and
Paris: Mouton, 1968. 2nd ed., Delhi, 1980.
Minaev, 1. P Buddhizm. Izledovaniya i lvlatmalui, vol. 2. St. Petersburg,
1887Mitra, R. The Sanskrit Buddhist Literature ofNepal. 1882; repro New
Delhi: Cosmo Publications, 1981.
Mukherji, Durga Das. AryamaffjufrinrimasaT(lglti: Sanskrit and Tibetan
Texts. Calcutta: Calcutta University Press, 1963.
Obermiller, E. "History ofBuddhism", being an English translation of
Bu-stan's Chos-'byung, 2 parts. Heidelberg, 1932. 2nd part
republished as The History ofBuddhism in India and Tibet. Delhi,
1986.
Pruden, Leo M. (tr.) Abhidharmakosabhrivam by Louis de La Vallie
Poussin. Berkeley: Asian Humanities Press, 1988.
Sakurai, Munenobu. 'A Study on the NamamantrarthavalokinJ. (1)A Critical Text ofIts Chapters III & IV'. Bunka (Culture) 51,
No. 3/4, 19 88, pp. 323-354.
Shastri, M. H. (ed.) AdvayavajrasaT(lgraha, Gaekwad's Oriental Series,
no. 60. Baroda: Oriental Institute, 1927Skorupski, Tadeusz. A Catalogue qf the sTag Palace Kanjur.
Bibliographica Philologica Buddhica, Series Major IV. Tokyo:
The International Institute for Buddt'list Studies, 1985.
136

Snellgrove, David. hdo- Tibetan Buddhism. Indian Buddhists and their
Tibetan Successors. London: Serinclia Publications, 1987.
Tribe, Anthony. The Names oj Wisdom. A Critical Edition and Annotated
Translation oj Chapters 1-5 oj Viliisavcgra's Commentary on the
Niimasarrzgiti, with Introduction and Textual Notes. Unpublished
doctoral thesis: University of Oxford, ]994.
- - - . 'Manjusrt Origins, Role and Significance.' Parts I and II.
Western Buddhist Review 2, forthcoming.
Tucci, Giuseppe. Minor Buddhist Texts, part II. Rome, 1958.
Vira, Raghu. Maf[jusn---Niima-Sangiti, edited in Sanskrit, Tibetan,
A10ngolian, and Chinese, Satapitaka Series 18. New Delhi, 1962
(reprinted in Raghu Vir a and Lokesh Candra, eds., KiilacakraTantra and Other Texts, Part 1. Satapitaka Series 69. New Delhi,
1966).
Wayman, Alex. Chanting the Names ojjvfafijufn: the j1;faffjufn-niimasarrzgiti, Sanskrit & Tibetan texts, translated with Annotation and Introduction. Boston and London: Shambhala, 1985.
Williams, Paul. Nlahiiyiina Buddhism: The Doctrinal Foundations. London:
Routledge, 1989.
Yoritomi, Motihiro. j1;fikkyo butsu no ken9u [Studies on Esoteric
Buddhas]. Kyoto, 1990.
Caskets of Treasures and Visions of Buddhas 1

Indic antecedents qf the Tibetan gTer-rna Tradition
Robert Mayer, University of Kent at Canterbury
Although much has been written about gter-ma in recent decades, its
interesting continuities with earlier Indic systems of scriptural revelation
seem so far to remain understated. In this paper I therefore intend to argue
that the gter-ma tradition is primarily a Tibetan elaboration of Buddhist
systems already well attested in Indian literature many centuries before the
introduction of Buddhism to Tibet - rather than a syncretic development
derivative of indigenous Tibetan religion, or a Buddhist invention entirely
unique to Tibet, as some scholars have suggested. Thus gter-ma would
appear very much in line with the rest of Tibetan Buddhism, a salient
feature of which is the reception of given Indian Buddhist ideas, followed
by their subsequent development within Tibet. To illustrate: Tibetans, not
Indians, developed the bodhisattva doctrine into the sprul-sku or incarnate
lama system; concretised the division of the Madhyamaka into its
Svatantrika and Prasangika branches; and developed the tathiigatagarbha
doctrine into the comprehensive "Great Madhyamaka" (dbu-ma chen-po).
What characterises such Tibetan developments is that they are not
syncretic; in other words, they do not involve the admixture of indigenous
Tibetan beliefs with imported Buddhist beliefs. Nor are they fresh
inventions entirely unique to Tibet. On the contrary, they are more
fruitfully understood as Buddhist developments of Buddhist ideas, albeit
worked out on Tibetan soil. What I wish to argue is that this seems as true
of the gter-ma tradition as of, say, the systematic division of Madhyamaka
into Svatantrika and Prasangika.
What then are the Indian antecedents from which the gter-ma tradition
evolved? Here, I shall be looking at only two of several such antecedents:
firstly, the systems of revelation and transmission described in the early
Mahayana sutra, the Pratyutpanna-buddha-saTJ1-mukhiivasthita-samiidhi-sutra
(henceforth PraS); and secondly, the tantric cult, both Hindu and Buddhist,
of discovering hidden treasures (nidhz). Other relevant topics I cannot
I38
discuss here include the revelatory mystical journeys of such sages as

Nagarjuna and Asanga; other Mahayana sutra texts that describe
scriptural revelation; and the methods of revealing tantric scripture~. All
these features of Indian Buddhism have parallels, survivals, or revivals, in
the Tibetan revelatory traditions. Nevertheless this short paper must
necessarily be limited in scope.
The PraS
The PraS, or The Samiidhi qf Direct Encounter with the Buddhas qf the Present,
received increased western scholarly attention after Paul Harrison's critical
edition of the Tibetan text was published in I978,2 followed by his
annotated English translation published in I990. It was in I983, after
reading Harrison's unpublished PhD thesis, that Paul Williams first drew
my attention to the PraS as a precursor of the Tibetan gter-ma tradition.
Likewise Harrison himselfhas also commented that chapter I3 of the PraS
"is an interesting adumbration of the later Tibetan gter-ma tradition". 3
According to Harrison, the PraS is, along with the AJtasiihasrikii-prajfiiipiiramitii, the oldest datable Mahayana sutra. The Chinese translations of
both scriptures were presented on the same day in I79 CE by Lokak~ema,
the Indo-scythian translator then active in the Chinese capital ofLuoyang. 4
The PraS was also one of the first scriptures to be translated into Tibetan,
since it is listed in the lDan-kar-ma Catalogue. 5 The PraS became very
important in the Far East for the Pure Land school, and it might have been
important in Central Asia,6 from where one Sanskrit fragment has been
recovered. In India, although the type of doctrine the PraS teaches was
widely accepted, there is little hard evidence of the PraS itself being very
popular, since it'is mentioned only obliquely in a few extant Sanskrit
sources. 7 Nor is it usually mentioned in the surviving standard Mahayana
compendia of sutra materials, except for one interesting exception - the
third Bhiivaniikrama of Kamalaslla. 8 This may be historically significant,
because Kamalaslla wrote his Bhiivaniikrama especially for the people of
Tibet at the time of their first conversion to Buddhism, and he was also
associated with Padmasambhava, with whom his master Santarak~ita
formed a close partnership;9 and it is the figure of Padmasambhava, of
course, who lies at the heart of the Tibetan gter-ma tradition. Thus, we have
some historical evidence that the gter-ma-like doctrines of the PraS were
known among the very circles of people that Tibetan sources maintain
were responsible for creating their gter-ma tradition.
According to traditional sources, there are three systems of scriptural
production and transmission counted by the rNying-ma pa: the bka)-ma (or
Oral Transmission), the gter-ma (or Treasure Tradition), and the dag-snang
(or Pure Vision Tradition). The bka)-ma is uncontroversial, comprising
Caskets qfTreasures and Visions qfBuddhas
139
simply the lineal transmission of scripture from master to pupil, in many

cases reputed to have begun with the historical Buddha. The other two
systems of transmission (which people sometimes tend to conilate, despite
the major conceptual differences between them) are both controversial and
are both dealt with at length in the PraS, from where we can see how the
Treasure and Pure Vision systems aroused the same kind of criticism two
thousand years ago that they still arouse today. For example, in PraS
Chapter 13, the future treasure discoverers lament the difficulty they will
face in the future in propagating scriptures which will never have been
heard of before. Likewise, the whole of Chapter 6 of the PraS is devoted to
a defence of the Pure Vision teachings. lb Nevertheless, subtextual nuances
suggest that the PraS and Tibetan revelatory traditions alike attempt to
derive a degree of inverted legitimation precisely from the indignation they
arouse in those they characterise as insufficiently broad-minded.
The samiidhi after which the PraS is named - the samiidhi of Direct
Encounter with the Buddhas of the Present - is the main teaching of the
PraS, largely contained in Chapter 3; it describes the deliberate cultivation
of a Pure Vision type of transmission. But in Chapter 13, the PraS
prophesies the history of its own transmission as a text, and this portrays a
Treasure type of transmission. Although this paper is largely concerned
with the Treasure tradition, I shall nevertheless very briefly first touch upon
the Pure Vision system.
The Pure Vision system

Chapter 3 of the PraS describes how meditators should systematically
cultivate visionary encounters with celestial Buddhas (e.g. Amitayus) by
means of specific contemplations. Thus they can receive teachings directly
from the celestial Buddhas, and subsequently propagate these as newly
revealed scriptures. II Paul Williams believes that this teaching of visionary
encounter with Buddhas, as expressed so vividly in the PraS but included
in other sutra texts as well, "provides a convincing basis for understanding
the origins of at least some of the Mahayana sutras" .12 But in the PraS,
there is an unexpected quality to this samiidhi: it can be practised by
ordinary persons without advanced supernormal powers. 13 This compares
very closely with the rNying-ma-pa Pure Vision (dag-snang) tradition. Tulku
Thondup writes:
Pure Vision teachings are not Terma, They are merely teachings given by
Buddhas, deities and teachers in visions. For this discovery the discoverer
does not need to be such a highly realised person, 14
However, Dudjom Rinpoche (I904-Ig87) seems to imply that the Pure
Vision tradition does in practice, even if not in theory, often depend upon
140
Indian Insights: Buddhism, Brahmani>ril and Bhakti
highly realised persons as the visionaries. Perhaps the visions of realised

beings inspire greater confidence; or perhaps different scriptural sources
vary a little. Dudjom Rinpoche cites the Arya-sarva-pwga-samuccaya-samiidhisutra, describing how realised bodhisattv.as perpetually dwell in a purified
perceptual sphere, which allows them to be in continuous dialogue with
celestial Buddhas and deities. Thus they receive innumerable visionary
teachings, which they sometimes propagate amongst their fortunate
disciples. These, writes Dudjom Rinpoche, are the Pure Vision teachings. l5
In either case, it would appear that identical or similar methods to those by
which Mahayana scriptures were produced in the early centuries CE still
continue to be used by present-day Tibetan visionaries in producing
contemporary Pure Vision teachings.
The Treasure system

The Treasure system has a more complex and distinctive structure than the
Pure Vision system, with diachronic as well as synchronic elements.
Although it seems to be described in passing in several sutras, for example
lll. The Manifistation qfLights Sutra from the lvlahiiratnakuja collection (T 3IO),!6
as far as I know the Treasure system is most fully described in the PraS's
Chapter I3, which describes a complex Treasure system which shares a
basic structure and key technical terms with the Tibetan gter-ma systemY
The correspondences are so precise that it seems quite reasonable to
conclude that the Tibetan system is at least in part derived from the system
described in the PraS.
Chapter I3 of the PraS opens with the layman Bhadrapala, here the
Buddha's chief interlocutor, asking the Buddha a very significant question.
Bhadrapala, famous in many Mahayana sutra texts as the leader of the
"Sixteen good Day] men" /8 appears here as the first among a group of
eight great lay bodhisattvas, the principal recipients of the teachings that the
Buddha gives in the PraS. They are accompanied by an important second
grouping of five hundred monks, nuns, laymen and laywomen, who are
also attending the teachings as secondary recipients. Bhadrapala's question
is: what will happen to the teachings the Buddha has just given after his
parinirvii1Ja?
The Buddha replies to Bhadrapala's question with a major prophecy.
He predicts that the teachings of the PraS which he has just given will
disappear forty years!9 after his parinirvii1Ja. Just before that time, however,
the eight great lay bodhisattvas led by Bhadrapala will make written copies of
the teachings and seal them in caskets (sgrom-bu). These caskets will then be
hidden in stupas, in the earth, under rocks, and in the mountains; and will
be placed under the guardianship of deva and niiga spirits. Then Bhadrapala
and his friends will die and be reborn in the deva realms, where they shall
Caskets if Treasures and Visions ifBuddhas
141
remain for a long while. But at a terrible time in the future, the "last 500
years" when true Dharma is all but lost to mankind, Bhadrapala and his
friends will be reborn on earth. There they will once more rediscover the
teachings "entrusted" (glad) to them by the Buddha at the time when he
first taught the PraS. Having searched for and recovered from the guardian
spirits the teachings they had in a past life sealed in caskets and hidden in
stilpas and rocks etc., they shall practise them once more, and eventually
propagate them among the beings of that "final epoch", for whom they
had all along been specially intended by the all-seeing compassionate
Buddha. The secondary recipients of the PraS described above will also be
reborn into that same time and place, both to serve the rediscoverers, and
also to be the principal holders and guardians of their teachings. Finally, a
cryptic verse, found only in the Tibetan version, states that eight monks, as
well as many "in the North" who rejoice in Dharma, will appear to receive
these rediscovered teachings. 20
The similarity of the Treasure system of the PraS to the Tibetan gter-ma
tradition is unmistakable. By comparing the various parts of the two
systems individually, we cim see these similarities more clearly.
a. As we have seen, the PraS has two clearly demarcated groups of
disciples with distinct functions. Similarly, the Tibetan gter-ma system
has an exact parallel in the two categories of disciples who attend the
initial teachings given by Padmasambhava: the primary recipients who
will become the future gter-stons, and the secondary recipients who will
be reborn along with them as their chos-bdag or "Doctrine-holders", to
serve them and become the indispensable principle holders and
guardians of the rediscovered teachings. Both gter:{ton and chos-bdag
alike should be prophesied in great detail by Padmasambhava. 2!
b. Just like Bhadrapala and his seven friends in the PraS, the Tibetan
gter-ston is nearly always a lay bodhisattva. Except for a tiny minority
who have been monks, a gter-ston is normally a householder with
consort, children, and possessions. 22 In fact, a female consort is
thought to be very important for the gter-ston, if he is to function
properly as a treasure rediscoverer.
c. Llkewise, there exists a similar parallel between the 500 secondary
recipients of the PraS, and their rNying-ma-pa counterparts, the chosbdag. As far as I am aware, in both cases their religious status is
comparatively immaterial. They can be either lay men or women, or
monks or nuns. What counts is their relationship to the Treasure
teaching and its discoverer. 23
d. The rNying-ma-pa system is also extremely similar to the PraS in the
manner of the treasure's concealment. As in the PraS,
Padmasambhava's students supposedly committed his treasure
teachings to writing,24 and then sealed them in caskets. These caskets
142
Indian Insights: Buddhism} Brahmanism alJd Bhakti

are invariably called sgrom-bu, precisely the same term used by the
PraS. As in the PraS, the sgrom-bu were then supposedly hidden in
stupas, in the earth, under rocks, in the mountains, and so on, where
deva and niiga spirits were appointec;l. to guard them.
e. A key technical term in the rNying-ma-pa gter-ma system is the word

gtad-rgya, translated by Tulku Thondup as "mind mandate Transmission". He explains it as follows:
(Gtad-rgya) is the main instrument for the concealing of gTer. The transmission and concealment take place in the essential nature of the mind of
a disciple by his power of concentration. The Guru integrates his enlightened mind with the awareness state of the mind of his disciple, and that
integration is the absolute transmission and concealment of the teaching.
Concealment of symbolic script etc. are supports of transmission. 25
In the PraS this same technical term occurs, although its meaning is
not spelled out: the five hundred secondary recipients urge the Buddha to
"entrust" (gtad) the PraS to the eight great lay bodhisattvas, so they can
rediscover it in future lives.
.
Of course, there is one major difference between the PraS and the
Tibetan gter-ma tradition: whereas in the PraS the historical Buddha is the
originator of the Treasure teachings, in the tantric rNying-ma-pa system it
is the historical Padmasambhava and, less frequently, a few other gurus of
his time, who are the originators of the Treasure teachings. But the
Tibetans justify the substitution of Padmasambhava for the Buddha
because of their notion that Padmasambhava, as a realised tantric guru, is
himself a fully e~lightened Buddha, a nirmiir;akiiya of Amitabha, Sakyamuni,
and others.26 Thus they maintain a formal equivalence to the system
described in the PraS.
A paradoxical consequence of the existence of such scriptures as the
PraS, is that a potential for tension is created between a form of religious
orthodoxy and a form of textual orthodoxy. Textually, the PraS, with its
decidedly Madhyamaka slant, is without question considered valid
scripture by all Mahayana schools; yet religiously, its teachings on how
systematically to reveal new scriptures are potentially subversive of another
strand in Buddhist thinking, which understands orthodoxy as the strict
preservation ofa received scriptural canon. Within Tibetan Buddhism, two
characteristic responses to this dilemma are instantiated by the dGe-Iugs-pa
and rNying-ma-pa schools respectively. The dGe-Iugs-pa inspiration is to
value consolidation of existing scriptures and doctrinal stability more highly
than fresh scriptural production. As I understand it, they therefore take the
view that meditational standards have declined so radically in recent
centuries, that no new important scriptural production could reasonably be
expected anymore; hence the teachings of the PraS are considered mainly
Caskets of Treasures and Vi.<ions ofBuddhas
143
to apply to a previous age, rather than our own. Thus they favour what one
might loosely call a "closed canon". The rNying-ma-pa, on the other hand,
value fresh scriptural production so highly that they feel quite prepared to
undertake its risks. However, to reduce the dangers of destabilisation, they
have developed an elaborate system of checks and balances. They apply,
for example, the category "False Treasure" (gter-rdzun) for purported
treasure productions that are not adequately in accord with the established
tradition. These can thus be easily rejected, usually by being identified as
the effects of specific demonic forces attempting to mislead people by
masquerading as genuine revelations. 27
It is very well known that differing attitudes to ongoing revelation
became an important cause of conflict throughout most of Tibetan
Buddhist history, with some authorities rejecting fresh revelations as
fraudulent, and others accepting them as valid. But it should not be
forgotten that this kind of tension seems to be inherent to Mahayana
Buddhism itself, and not by any means restricted to Tibet. In China, for
example, an important text for Far-Eastern Buddhism called the Kuan-ting
ching (Book of Consecration; T 1331), was produced by means of the
treasure system in fifth-century Chiang-nan, during the Six Dynasties
period, that is, seven centuries before the first appearence of gter-ma in
Tibet. The Consecration sutra is quite explicit about its claimed origins as
a treasure, describing at length its initial teaching by the historical Buddha,
its subsequent concealment, and its eventual recovery from a grotto, where
it had been hidden in ajeweled casket, written in letters of purple and gold
upon sandalwood tablets. But this process was clearly controversial, since
the Consecration sutra also has a lot to say about the fierce opposition its
revelation will arouse among conservative monks. As it happened, the
revealer of the Consecration sutra, probably a monk called Hui-chien, was
able to see his revelation gain the canonical status which it still enjoys,
because in his day the political and religious climate was favourable to fresh
scriptural revelation. But in other historical periods less fortunate revealers
of scripture could find themselves severely condemned as charlatans and
rebels, and even the Consecration sutra only managed to retain its
canonical status by rapidly acquiring a false identity in the traditional
bibliographies as a text translated from Sanskrit. 28 Clearly, Chinese
Buddhism suffered from a similar type of ambivalence and conflict
regarding its, apocrypha as did Tibetan Buddhism. There is also, of course,
plentiful eveidence that Indian Buddhism suffered from tremendous
tensions concerning the revelation of fresh scripture. Given that nearly all
Mahayana and Tantric Buddhist scripture, and even the Abhidharma, can
be seen as apocryphal from the early Buddhist point of view, often claiming
to be the utterances of the historical Buddha even when this clearly was not
the case, such tensions were a fundamental feature of Indian Buddhism
from an early period.
144
Nidhi
The second topic to be examined in this paper is the tantric cult, both
Hindu and Buddhist, of finding hidden treasures, usually material, called
nidhi in Sanskrit. Within Buddhism, this cult was more prominent among
the earlier kriya-tantra texts than among the later tantras, a fact which lends
support to the traditional claim that Padmasambhava was concerned with
nidhi. It seems that elements of the nidhi tradition combined with the
Mahayana traditions described above, in the formation of the Tibetan gterma tradition; for not only is the Tibetan word gter-ma a direct translation of
the Sanskrit word nidhi, but the Tibetan gter-ma tradition also resembles the
Indian nidhi tradition in several important respects. For example, it includes
under the rubric of gter-ma material treasures as well as sacred texts.
Likewise, it shares a central characteristic of the Indian tantric cult in
placing a heavy emphasis on the role of fierce treasure protectors such as
naga and yak~a spirits. Furthermore, it also incorporates the Indian tantric
idea of treasures being repeatedly recoverable from the same special
magically-endowed sites (fri-mukha or gter-kha).
In the Hindu tantric and magical traditions discovery of nidhi has its
own distinctive lore. It comprises a special set of rituals in its own right, and
is not simply subsumed under the rubric of such standard categories as
akar!aTJa, the attracting of women or wealth. On the contrary, nidhidarsana
or nidhlinalabha, "finding a hidden treasure", is listed in the Hindu tantric
and magical compositions as a distinct and independent category.29 Hence
at least one entire text devoted to the subject survives in Sanskrit, the
Nidhidarsana, or "Discovery of Treasure", by Rama Vajapeyin. 30 Finding
treasure is also of course referred to in a host of other texts, ranging from
the Mahlibhlirata to the Jayadrathayamala and including the Samavidhlina
BrahmaTJa, the Atharvaveda-parifi!taTJi, the Saktisarrzgama Tantra, the
Siddhanaga7J'unakak~aputa and the Kamaratna. 31 Many such texts give entire
sadhana rites for the finding of treasure; for example, the third ~atka of the
Jayadrathayamala has a patalasiddhi chapter in the section on the sadhana of
the goddess Ghoraghoratara Kali, which deals with sadhana at caves (bila-),
the best of which are called sri-mukha, and by means of which one can
attain patala, the subterranean paradise of nagas and asuras, where treasure
could be obtained. 32
In such Hindu texts, the treasures were usually said to be guarded by
fierce naga spirits, but sometimes a yakJinf or some other very dangerous
spirit is mentioned instead. Thus it was believed that only an accomplished
siddha, or those with the support of an accomplished siddha, could ever
procure such treasures, since the treasure guardians would harm or even
kill a merely ordinary human being who had the temerity to attempt to
take out the treasure they guarded. Hence the rituals to extract the treasure
Caskets qfTreasures and Vzsions qfBuddhas
145
were quite complex. They included rites to find out where the treasure
was; for example, a siddha might propitiate the Saiva godess NidhIsvarI,
"Mistress of Treasure", said to be the wealth-god Kubera's mother, because
it was believed that she would grant a vision of where the treasure was
hidden. Then the treasure seeker also needed very powerful rites to protect
himself from the treasure protectors, and various ointments to make the
treasure become visible. 33
The treasure was often believed to be located at a special site sometimes
known as a frt-mukha, or "treasure face", which, it was thought, could be
visited repeatedly by many different siddha treasure-seekers over a long
period of time and still yield up treasure for all of them; in other words, the
source was considered supernatural and thus not exhausted as a
commonplace supply of treasure would be. The treasure itself was usually
said to consist of magical eli'cirs and gold, and it was usually stated that a
proportion at least of money thus discovered had to be used for directly
religious purposes. 34
Early Buddhist tantras, and, as we shall see, the Tibetan gter-ma tradition,
include much of the same type of belief as the Hindu system. The
Maffjufnmillakalpa (henceforth MMK) is the Buddhist kriya-tantra best known
to Western scholars because a Sanskrit version of it has survived. In this
text (and many other kriya-tantra texts like it) we find a good deal of material
on the recovery of nidhi, including such topics as the summoning ofniiga or
yak~a spirits, or Gar;tapati, to allow one to see the treasure; and frequent
refences to "asura's caves", a favoured Buddhist term equivalent to the
Hindu patala. Like their Hindu~counterparts, in the Buddhist tradition
these asura caves are the sites where yo gins can obtain every kind of
desirable goal, over and above treasures, ranging from complete
immortality and occult knowledge to sensual pleasures. We also find in the
MMK, as in the Hindu texts, an injunction that half of the wealth
recovered as nidhi must be spent on the ThreeJewels. Sometimes the MMK
insists that the treasure seeker has to begin his search during eclipses, with
the recitation ofmantras. 35 But good astrological moments are not always
necessary; in another early Indian Buddhist kriya-tantra, the Kani-krodhavqjrakumara-bodhisattva-sadhana-vidhi (now preserved only in Chinese), we find
the following description of a ritual to find nidhi:
There is also a rite for those who desire to acquire treasure (nidhl). Do not
select a particular [astrological] season, day or hour, and it is not necessary
to maintain the discipline. In the vicinity of the treasure, the mantrin should
raise one foot and recite the mantra. Turning around to the right, he should
gaze towards all of the four directions and take possession of the area (simabandha). Carefully raising one foot, he should recite the mantra one hundred
and eight times. If the guardians of the treasure obstruct him, then they will
be burnt in a mass of fire. They will come screaming to the mantrin and bow
146 Indian Insights: Buddhism) Brahmanism and Bhakti

before him vanquished. The mantriil should say to them, "Open this
treasure store and give all that is herein to me!". They will then open it and
give everything to the mantrin. If they are mean-spirited and do not give it to
him, then he should say, "Brahma,. NaraYaI,la, MaheSvara, the warrior
goddesses and **** will come and crush your treasure store. You give it to me
quickly! If you do not do so, the wrathful VaJrakumara will destroy all of
your family!". When they have heard what he said, they will all obey and say
to him, "Noble One! Come and take what you will, we shall not hinder
you!". Then he should say to them, "You may open the store yourselves and
give it to me!". They will then immediately open the treasure-store and give him [the treasure] respectfully.36
The following is also from the same text:

There is another' rite if you desire to locate hidden treasure. Get some
yoghurt fro~ a yellow cow; a snake skin, and some shark-oil, as well as some
arka wood' and cotton. Make a lamp with these things. Recite mantras to
empower it, and then light it at night near the place where there is treasure.
You will know the amount of treasure that is there by the size of the flame.
If you need to expel the gods who guard the treasure and other obstructors,
take a slab of rock or a pebble or some mustard seeds or some empowered
water and cast it at the treasure. The obstructor on the treasure will
withdraw. If you suspect that there is a large niiga there, this will also leave. 37
It seems that in the Hindu and Buddhist tantric traditions alike, nidhi
was predominantly seen as material wealth, a category in which magical
elixirs are included. But in some of the more exoteric strata of Vedic or
Hindu literature, the words nidhi (treasure) andnidhipa (treasure protector)
carried a slightly different nuance. From citations in the KauSikasi1tra and the
Grhyasiltras, Jan Gonda believes that nidhi here refers to something of
spiritual value deposited in a spiritual realm. This was guarded by
protectors called nidhipa, whose function was to ensure that only those who
had deposited the nidhi in the first place, or those for whom it was intended, .
could eventually reclaim it, when they eventually arrived in the spiritual
realm in person. Agni, Prajapati, or Brhaspati would often act as "keepers
of the deposits". 38 This classical use of nidhi to mean spiritual treasure does
not seem to be entirely lost in the tantric tradition, since the material
treasures discovered there retain at least a spiritual connection - some or all
of the wealth q1Ust go to spiritual purposes, and consumption of the sacred
elixirs is inherently spiritual. But despite a reasonable search, I have been
unable to find any Indian tantric reference to the use of the term nidhi to
mean a scriptural or textual discovery, as it so frequently does in the
rNying-ma~pa system.
Nevertheless the similarities between the lndic material mentioned
above and the Tibetan gter-ma tradition are clear.
Caskets rif Treasures and Vzsions rifBuddhas
I47
a. Like the Indian tantric cult, the Tibetan tradition includes under the
rubric of gter-ma the discovery of wealth, sacred elixirs, and valuable
objects, just as much as the recovery of religious texts and scriptures.
For example, Padma Gling-pa was offered a skuUful of gold by the
treasure-protector (gter-srung) Khari, who also promised to gradually
give him all the wealth of the local rulers of Tibet. 39 Likewise,
Dudjom Rinpoche lists many longevity pills, jewels, flasks of the
"waters of life", images, relics, and other valuable objects that were
discovered by various gter-stons. 40 Dodrupchen III echoes the Indian
tantric tradition's emphasis on elixirs when he writes that, "according
to some interpretations, the amrta rendering liberation by tasting is
praised as the best among the Terma substances". 41 All these various
types of material treasures seem to be quite as widespread as the
scriptural treasures. Both equally go by the name of gter-ma.
b. Like their Indian tantric antecedents, Tibetan gter-stons constantly
revisited the same treasure sites, from which many generations of
treasure-finders spanning many centuries could recover treasures ..
These were called gter-gnas, "treasure places", or gter-kha, literally
"treasure faces", i. e. "treasure sites", and often were situated at very
dangerous or inaccessible places. Within the gter-kha would be a gtersgo or "treasure door", a miraculous door in the rock which only the
appointed gter-ston could open and within which he would find the
casket (sgrom-bu) containing the treasure. Mter the treasure was
removed, the door would miraculously be resealed, leaving only a
mark on the rock. This complex of gter-kha' and gter-sgo seems very
similar indeed to the Sanskrit notion of sn-mukha; indeed, gter-kha is
probably a direct Tibetan translation of the Sanskrit technical term.
c. The ambivalent nature of the guardian spirits of the treasures is
again common to both traditions. In the Indian nidhi tradition and
in the Tibetan gter-ma system alike, these guardians are extremely
dangerous, despite the useful function they carry out. In his study of
Padma Gling-pa, Michael Aris graphically describes the dire calamities thought to have befallen both humans and animals as a result of
offending the gter-ma keepers (gter-bdag), whether unwittingly or not;
all of which underscores the importance of constantly placating
them.42 Dodrupchen III states this ambivalent nature of the
treasure-guardians very clearly. He points out that many of the
treasure-guardians were deliberately selected and appointed by
Padmasambhava from among the chiefs of the various classes of evil
demons opposed to Buddhism. The idea is the recurring tantric
theme of controlling evil spirits: "by appointing. their chiefs as
[gter-ma] protectors, the subjects won't be able to transgress their
orders".43
148
We can see from the above that some of the aspects of the rNyiIlg-ma-pa
gter-ma tradition which are not derivable from the teachings of ,such
Mahayana sutra scriptures as the PraS, are derivable from early Buddhist
tantras of the kriyii class. (Of course, there are also other elements, such as
the role of r/iikinz.s, taken from mahiiyoga tantric sources, that are not analysed
at all in this paper). Yet on closer analysis, much of the kriyii tantric material is essentially an expansion of aspects of the sUtra material, for the PraS
already includes such topics as niiga and deva spirits who guard the treasures,
and rocks and mountains in which they guard them. Thus it is clear that
. the most important basic structuring concepts of the gter-ma tradition come
from the sUtra rather than the tantra tradition; it is only in the area of certain practical details of concealment and retrieval, and in the idea of material treasures and elixirs being discovered, that elements are derived from
the early tantric sources.
In conclusion, we can see that it might well be mistaken to regard the
gter-ma and dag-snang systems as syncretic, or essentially indigenous to Tibet.
On the contrary, it s.eems that these traditions constitute a predominantly
Buddhist development of Indian Buddhist ideas, albeit carried out on
Tibetan soil. Even the visionary journeys to receive teachings (for example
directly from Padmasambhava in his paradise, or from the deities at
Bodnath in Nepal), experienced so often by gter-stons such as Padma-Glingpa and others, and seen by some Western scholars as strong evidence of a
non-Buddhist shamanism,44 to my mind more probably carry a quite
different connotation. Although, admittedly, we do have evidence of
shamanistic journeys in non-Buddhist Tibetan religion/ 5 such journeys are
also central to the Pure Vision tradition as described in classic Mahayana
sources. Thus it would seem more likely that the visionary journeys of
Padma Gling-pa and others are simply an emulation of the magical journeys of scriptural revelation made by exemplary Indian Buddhists such as
N agarjuna, revealer of the Prcg'iiiipiiramitii scriptures, and Asariga, revealer of
the famous teachings attributed to the Buddha Maitreya.
The prevailing Western academic view of gter-ma has so far tried to
understand it predominantly ill terms of the historical conditions influencing its first appearence in 11th century Tibet. Hence it is seen largely in
terms of a response by the followers of the Old Tantras, to the challenge
posed by the arrival in Tibet and translation of the New Tantras. 46 'While
not intending to take issue with this view, I think that such a sociological
perspective can fruitfully be broadened by a textual consideration of the
degree of fidelity that the actual methods of gter-ma production bear to the
much older mainstream scriptural traditions. Such data tends to confirm
the views of anthropologists such as Maurice Bloch that the nature of ritual
is extremely slow-changing; and of Stanley Tanbiah that the possession of
a detailed written scriptural corpus can confer on a tradition the power of
constant or repeated regeneration to a very precise template over very long
Caskets qf Treasures and Vzsions qfBuddhas 149

periods of time. Indeed, it seems possible that in observing the highly
systematic workings of contemporary Tibetan scripture-revealers, we might
in fact be observing.a unique survival, or at least a close replica, of the
workings of the revealers of many of the most famous Indian and Chinese
Buddhist scriptures of the last two millennia. As a grow"ing number of
scholars have remarked in recent years, a close study of both Chinese and
Tibetan Buddhist apocrypha might well yield a useful contribution towards
an un'derstanding of the development of Mahayana and Vajrayana
Buddhism in India.
Notes
1.
2.
3.
4.
5.
6.
7.
8.
g.
IO.
II.
12.
'3.
'4.
'5.
16.
'7.
18.
19.
20.
21.
22.
23.
24.
25.
26.
An earlier less complete version of this research wa., presented at the 6th IATS Conference,
Fagernes, '992. Thanks to Prot: Per Kvaerne lor permission to publi'h the updated version
here. I would also like to acknowledge the considerable help of Alexis Sanderson,Janet
Gyatso and Stephen Hodge in giving me valuable references and other forms of help, and of
Cathy Cantwell, l'vlike Hookham and Dzongsar Khyentse Rinpoche in critically reading the
early draft,.
See the careful attention paid to this text in Williams, 1989.
Harrison, '9go, p. xvii.
Ibid., p. vii.
Harrison, Ig78, p. xi.
Harrison, Igg0, p. xxv.
Ibid., p. xxiii.
Ibid.
Snellgrove, Ig87, p. 430.
Harrison, 1990, pp. xxi, 54-60, IOO, I04.
Ibid., pp. 3'-44.
Williams, Ig8g, p. 30.
Ibid., p. 221.
Thondup; Igg0, p. '57. See also Thondup, 1986, pp. go-gI.
Duqjom Rinpoche, Igg1, pp. 747-8.
Chang, 1983, pp. 200~20I, 216-8. The teachings of this sutra seem to adumbrate the rDzogschen practice of thod-rgal.
The\f'ibetan defenders of the gter-ma tradition regularly cite a substantial number of
scriptures from which they derive legitimation. Although the PraS is olten mentioned, for
example in Sog-bzlog-pa's bKa'-tlzang ]id-lfJi mun-sel, or in Guru Bkra-shis' history, or in
Thondup, Ig86, p. 109, such traditional apologists seem to give no special weight to the PraS.
It would be an interesting exercise to study the entire range of these citations in detail; several
of them seem to describe the same kind of scenario found in the PraS. Thanks to Dan Martin
lor these references.
$o{la.l'a satpurueiil.l. See Harrison, Igg0, p. 6.
See Harrison, 1990, p. g6, n2. He shows good rea.,ons why the text must originally have
given 40 years, despite the fact that the surviving Tibetan version gives 4,000 years.
See Harrison, Iggo, p. I04, nI4. The Tibetan text reads b,yang-phyogs-su, possibly from an
unusual Sanskrit formation with a different meaning.
For a discussion of the cllos-bdag, see the explanation of Dodrupchen III, in Thondup, 1986,
p. 162. This author was the third reincarnation of the clzos-bdag of jigs-med gling-pa.
Thondup,lg86, p. 82.
Ibid., p. 88.
Often, the writing is in an extremely condensed symbolic or code form, designed to awaken
within the gter-ston a memory of the teaching imprinted o'n his mind in his past life.
Thondup, Ig86, p. 236
See Dudjom Rinpoche, Iggl, p. 746.
150
27,
28,
29,
30,
31,
32,
33'
34,
35,
36,
3}
38,
39.
40,
41,
42,
43.
44
454 6,

Evil ministers oC Padmasambhava's time made evil aspirations to confound the true gier-ma
and mislead future generations with faJse treasure-teachings, See Thondup, 1986, pp, 154-6,
Strickmann, 1990, pp, 7S-IIS, Thanks to Janet Gyatso for drawing my attention to thi'; valuable source.
Goudriaan, 1978, pp, 2g4,30}
Goudriaan and Gupta, Ig81, p, 124,
Goudriaan, 1978, p, 30}
Thanks to Alexis Sanderson for this information,
Thanks to Alexis Sanderson for much of this information,
This summary of the Hindu cult of nidlzi is based on Goudriaan and Gupta 1981 (op,ci!,),
Goudriaan 1978 (op,ci!,), and additional information very kindly given to me by Alexis
Sanderson'from his own notes,
Thanks to Alexis Sanderson for much of this information,
From an unpublished partial translation very generously made lor me by Stephen Hodge,
Stephen Hodge ibid,
,
Gonda, 1965, pp, 183-193'
Aris, 1989, pp, 57-8.
Dudjom Rinpoche, 1991, (voL II), pp. 319-327.
Thondup, 1986, p, 152.
Aris, 1989, p. 44Thondup, 1986, p, II4
Aris, 1989, pp. 53-63,
Of course, it remains highly debatable as to what extent the so-called Tibetan "folk" traditions (such as the dpa'-bo etc) are really non-Buddhist,
Snellgrove, 1987, p. 397,
Bibliography
Aris, M. V Hidden Treasures and Secret Lives. London and New York:
19 89.
Chang, Garma C.C. (ed). A Treasury qfNfahilyiina Siitras. University
Park and London: 1983.
Dudjom Rinpoche, Jikdrel Yeshe Dorje The Nyingma School qfTibetan
Buddhism: Its Fundamentals and History, trans. Gyurme Dorje and
Matthew Kapstein. Boston: 1991.
Gonda,]. The Savayajnas. Amsterdam, 1965.
Goudriaan, T Nfilyii Divine and Human. Delhi: 1978.
- - - and Gupta, S. Hindu Tantnc and Siikta Literature. Wiesbaden:
19 8 1.
Harrison, P. M. The Tibetan Text qf the Pratyutpanna-BuddhaSaT[lmukhiivasthita-Samiidhi-Siitra. Tokyo: 1978.
- - - . The Pratyutpanna-Buddha-SaT[lmukhiivasthita-Samiidhi-Siitra
(unpublished doctoral thesis, Australian National University,
Canberra), 1979.
- - - . The Samiidhi qfDirect Encounter with the Buddhas qfthe Present.
Tokyo: 1990.
Snellgrove, D. L. Indo- Tibetan Buddhism vol.II. Boston: 1987.
Strickmann, M. "The Consecration Sutra: A Buddhist Book of
Spells", in Chinese Buddhist Apocrypha, ed. C. Buswell. Honolulu:
199 0 .
Caskets qf Treasures and Visions qfBuddhas
151
Thondup, Tulku, Hidden Teachings qfTibet. London: 1986.

- - - . "The Terma Tradition of the Nyingmapa School", in The
TibetJournal, vol. Xv, nO.4, Winter 1990.
Williams, P Mahayana Buddhism. London: 1989.
Freeing the Tiyyas: NarayaI).a Guru and Religious

Ideology in Kerala
Theodore Gabriel
"One caste, one religion, one God for humanity". With this striking
aphorism as his motto Narayal).a Guru led a movement which had farreaching consequences in the religious and social spheres, not only of
Kerala but of the whole of India. But first to set the background more
dearly. Kerala is one of the smaller states in the South-Western region of
the Republic ofIndia. It has a population of about 29 million, half of whom
are Hindus, the remaining half being equally divided between the
Christian and the Muslim communities. In earlier times Kerala had been,
according to historians, a much larger region, extending from the 15th
parallel (roughly the location of Goa) to the cape and ruled over by a
Perumat (emperor) from the town of Mus iris (KoduilIiallu:r). But gradually
invaders appropriated the northern regions until, in more recent times,
Kerala came to be distinctly divided into four principalities, commencing
from the northern river Netravati. to .Cape Kanyakumari. The northernmost regions, .which were ruled by the emperor known as the Northern
Kolattiri and the Samiltiri (the Zamorin), were, in 1766, invaded and
subjugated by Mysorean Muslim rulers Haidar Ali and Tippu Sultan,
while Cochin (Kochi) and Travancore (Tiruvitankur) retained their
independence in spite of attempts by Tippu Sultan to invade these
territories also. Later the northern regions of Kerala were ceded to the
British by Tippu Sultan under the treaty of Srlrangapattal).am in 1792.
Mter the British took over the administration of the northern regions they
came to be called Malabar, a district of the Madras Presidency, and the
Cochin and Travancore principalities continued to be ruled by Maharajas
under the aegis of the British empire. In 1957 when the Indian states were
reorganised on a linguistic basis, the three regions consolidated to form the
Malayalam speaking state of Kerala. Kerala is bordered on the west by the
Arabian Sea and on the East by the mountain ranges known as SahyaQ.ri
I54
(the Western Gha:ts), which rise to an altitude of 9,000 feet. Keraja is

perhaps well-known to the Western world for three things: its ancieIlt trade
with the Midclle East in spices, the first century church of the St Thomas
Christians, and Sankara, the great Hindu theologian.
Keraja has a highly diverse and divisive caste system. Vivekananda
called the province the "the mad house of caste" in India. Especially severe
were the rules of pollution. Specific distances were prescribed between
various castes. A breach of these rules would result in the higher caste
individual being deemed as polluted by the lower caste one, who would
then be chastised in draconian fashion. For example, a Pulaya (one of the
lowest of the untouchables) should keep a distance of 96 feet from a
Brahmin. Otherwise he would be put to death. A TIyya (a higher
untouchable) should stand 36 feet away from a Brahmin, and I2 feet away
from a sudra, failing which he would be severely beaten. Amazingly the
concept of ritual pollution extended even to different categories of
untouchables. A Pulaya, for instance, should keep away from a TIyya to a
distance of at least 90 feet. It was believed that salvation was assured for
those who meted out instant punishment - death or beating - to offenders.
On major roads and temple roads notices were posted banning the aua17J.as
from using them. One could pollute a caste Hindu merely by passing near
a tank where he was bathing or by plucking coconuts from a tree under
which he passed. When the Nampiltiris (high caste Brahmins) travelled,
Nayars used to run in front shouting to warn low caste Hindus and
untouchables of the approaching Brahmin, enabling them to leave the road
and hide behind trees. Untouchables were also required to pay various
special taxes including a head tax and a breast tax. Altogether about one
hundred special taxes were levied from the untouchables. The lower
categories of untouchables were virtually slaves and worked on the land
with no remuneration except some rice for subsistence. They could also be
sold and exchanged among their high caste masters. It was only in I854
that the British Presidency in Madras ordered the Rajahs of Travancore
and Cochin to end this kind of slavery and free all the untouchables. About
I30,000 in Travancore and 40,000 in Cochin were freed. l Still other
discriminatory practices continued unabated.
Narayal)a Guru was born into the Tiyya caste in I856 in the town of
Cempazanthi, about 8 miles north of Trivandrum (Tiruvanantapuram),
the capital of Keraja. The Tiyyas are also known as E;r:avas and in some
places as Cogas (Cevakan Or warrior). They are believed to be migrants
from the northern part of Sri Lanka known as E;r:am from which the ethnic
name E;r:ava is derived. The term Tiyya, by which the caste is mainly
known in Malabar, is believed to be a corruption of the Sanskritic term dmp
meaning Island. Both names testify to the fact that the E;r:avas and Tiyyas
are migrants from Sri Lanka. Some scholars, however, opine that while
there was certainly intermixing with the Sri Lankan migrants, there is also
Freeing the T[yyas: Niiriiyar;a Guru and Religious Ideology in Kera{a
155
an indigenous element in the TIyya community, and actually state that the
Eravas are the earliest inhabitants of Keral-a. They are also believed to
have been Buddhists, and Buddhism might have been the factor which led
to the linking of the Sri Lankan and the indigenous Eravas.
There is considerable archaeological and numismatic evidence for the
widespread prevalence of Buddhism in Kerala. In the Buddhist chronicle
Mahiivarrzsa it is mentioned that king Asoka sent the monk RaksitadhIra to
Kerala for preaching Buddhism. 2 He is believed to have converted 60,000
in Kerala and established 500 vihiiras. It is opined by scholars that most of
the major temples of Kerala had been Buddhist vihiiras up to the 7th
century CEo The architecture and the structure of temples testify to this fact.
The 7th century Chinese traveller Hsuan Tsang comments on the
numerous vihiiras in Malakutta (Malabar).3 The avilokite~vara bodhisattva
dharma sasta is believed to be the God Cattan who became a popular deity
of the Tiyyas. The Sasta of Sabarimala, the most popular pilgrimage
centre in Kerala, may also have been originally the same bodhisattva.
According to C. V. Kunhiraman the Tiyya festivals of Kutirakettu,
Talappoli and A.rat are modelled very closely on Buddhist festivals. 4 The
temple of the most popular goddess ofKerala, the Kodunnallur Bhagavati,
is said to have its origins in a convent established by a famous Buddhist
nun, Karnaki, in CE 120. Professor K. G. Narayanan believes that the transition of Malayalam into a distinct language from DraInila was due to the
infusion of Pali words into the language of Kerala. 5 The most striking
evidence comes from the pilgrimage to Sabarimala. The observance of
vows of non-violence, vegetarianism and sexual abstinence, and particularly
the cry of 'farar;.am' (refuge) which the pilgrims call out continually are
evocative of Buddhism. Untouchability was always totally absent in this
great temple. It is believed that Buddhism reigned supreme in K~rala for a
thousand years from the third century BCE. The combined onslaught by the
Keralan emperor Kulasekhara Perumal and the great Hindu theologian
Sankara is believed to have led to the decline of the Buddhist faith in
Kerala. Legend states that Sankara challenged the Buddhist monks to a
theological debate which they lost, and the tongues of the defeated debaters
were cut off. However, the authenticity of this eventis doubtful. As a matter
of fact, the rise of Brahminism in Kerala led to the total destruction of
previous historical records pertaining to Buddhism, making it very difficult
for scholars'to reconstruct the history of the penod. At this time there was a
great influx of Aryans from the north via Tulunad who gradually came to
dominate the religious scene and gained political power among the rulers
by performingyagas believed by the kings to be beneficial to themselves.
The Buddhists who adopted Hinduism became Nayars, a fildra caste, who
allowed the Brahmins to t;ike their girls in concubinage. The Buddhists
who remained independent both religiously and socially were categorised
as untouchables and became the TIyyas. The continued tradition of great
I56
Indian Insights: Budlihism, Brahmanism and Bhakti
Sanskrit scholars and Ayurvedic physicians among the Tiyyas in spite of

the pollution concepts and discriminatory practices to which they were
subjected is evidence of the Tiyyas having been Buddhists and a prominent
group in no way inferior to the rest of society in those more egalitarian
days. The Brahmins composed works such as the Smikarasmrti and
JetininJaya in order to strengthen the nascent caste system. By the 9th
century the very complex caste system of Kerata was well entrenched.
The avar7Jas such as the Tiyyas were not allowed to worship the major
gods of Hinduism, for example VigLU and Siva. They had their own
pantheon of gods who are considered inferior to the Hindu gods and virtually categorised as devils. Gods such as Cattan, Camundi, Karinkali, and
Kalamatan are now considered to be evil beings, forms of Asura rather than
Sura. Some of these names even degenerate into terms of abuse in the
l\IIalayalam language. The TIyyas and lower castes were not allowed to offer
bhasma (sacred ash), milk or ghee. Their offerings were mainly palm toddy
and blood of chicken. They were not allowed to use Sanskrit in their rituals.
According to K. Vivekanandan there was a saying, no doubt among the
savar7JGS, that madya (liquor), miiT[lsa (meat), matsya (fish), mudrii (female organ)
and maithuna (sexual intercourse) were their ways to salvation. 6
NarayaDa Guru was born when discriminatory practices were at their
zenith in Kerata. Right from his childhood he reflected on and protested
against the oppression of the untouchables. It is said that he used to play
with boys of the Pulaya caste, the lowest of the lowest, and even touch
them. When his mother and other elders scolded him about this he used to
argue with them. He neither observed superiority over castes lower than
the Tiyyas nor acted with deference before castes higher than them. He
was said to be honest, devout and of a thoughtful disposition. He showed
considerable brilliance at an early age and mastered Sanskrit and Tamil.
He also displayed talent as a poet and writer. He has written about 60 compositions, in Malayalam,Tamil, and Sanskrit, mostly in verse. He worked as
a teacher for some time, teaching not only Tiyya children but also those of
lower castes. Since they were not admitted to the Tiyya school he used to
visit them in their homes and give them personal coaching, much to the
dismay of his caste peers as well as the Brahmins and sildras.
Gradually Narayar:ta Guru felt discontent with his life and decided upon
total renunciation. He went to Marutvamala, a thickly wooded hill near
the town ofNagerkovil and spent a number of days in total isolation and
meditation. However, his fame as a yogi spread, and he attracted a number
of visitors and disciples. He established an iifram in Aruvippuram where he
also built a temple and consecrated a Siva linga. The Brahmins were greatly agitated at this intrusion into their privileges and questioned him on
what authority he had consecrated a Siva milrti. The Guru replied "I have
installed an Erava Siva not a Brahmin Siva." The Aruvippu:ram sanctuary
became a famous centre for the dissemination of NarayaDa Guru's social
Freeing the Tfyyas: Niiriiym;.a Guru and Religious Ideology in Keraia
IS7
and religious ideology. It was visited by many distinguished visitors

including Tagore and Gandhi. At Sivagiri in I903 a sangha echoing the
egalitarianism of the Buddhist term was formed for the welfare of the
untouchables. This is the Sri NarayaI.1a Dharma Paripalana Sangham (The
association for the preservation of Sri NarayaI.1a Guru's dharma, popularly
knovvn as S.N.D.P) which has become a major force in Keratan Society
today.
In spite of his obvious mystical nature and attachment to Advaitic theology, NarayaI.1a Guru often seemed pragmatic and rationalistic in his
pronouncements. For instance, when his devotees desired to make his
Sivagiri iifram a centre of pilgrimage in the model of the great pilgrimage
centres of India, the Guru was initially quite against the proposal. He was
not in favour of the ostentation and needless expenditure associated with
most Hindu pilgrimages. and festivals. Finally he relented but made some
strict stipulations regarding the conduct of the pilgrimage. He chose the
Christian new year's day, first ofJanuary, as the pilgrimage day. Only ten
days' vows were to be observed by the pilgrims so as not to hamper their
normal work. These were based on Buddha's pafica fuddi - cleanliness in
person, food, mind, speech and action - from the eightfold Arya path. Any
extravagance was totally banned. Even the garments prescribed for
wearing on the pilgrimage were to be dyed yellow in turmeric so that they
could be whitened by washing in water and used again in normal life.
Conferences were to be organised as part of the pilgrimage celebrations led
by experts on various secular subjects such as education, hygiene, cooperation, agriculture, commerce, handicraft and other themes useful for
the community's upliftment. He said that the main objective of the
pilgrimage should be the advancement of the people and the nation. He
knew that the Tiyyas and other avarr;as were extremely backward educationally and economically. He always exhorted them to be thrifty, and not
to waste even a single pie. Once a disciple asked him" Is it better to
cremate the dead or bury them?" The Guru replied "grind them in a mill
and use as fertiliser". Seeing the disciple's shocked expression the Guru
asked, "Why? Will it pain them?" Smiling, the Guru said, "observe ten
days' prayer for the dead. Do not spend more than ten rupees during the
ten days." The Guru was thinking of the high expenditure usually involved
in Hindu cyclical rituals fDr the dead. Once his favourite disciple Kumaran
Asan, the famous poet, was about to conduct a tilaka havan, a funerary rite.
The Guru asked him, "Do you believe in reincarnation?" The poet replied
in the affirmative. The Guru said, "The soul is reborn as soon as it dies.
Then for whose benefit are you conducting this ceremony?". 7
NarayaI.1a Guru's ideology will now be examined under three headings,
namely caste, religion and God. I have already given some indication of his
world view and his stance on these categories. The following will look into
more detail at his views under these three headings.
158
Caste
NaraYaI,la Guru's objectives regardil1.g the caste problem extended beyond
discontinuing discriminatory practices. He wanted a total eradication of the
concept of caste from human minds. In this he differed from Mahatma Gandhi who was not opposed to the caste system as such, and believed in
varTJiiframadharma, though he was opposed to caste-based discrimination and
worked hard for the upliftment of the untouchables. NarayaI.J.a Guru said:
"Do not ask about caste, think about caste, speak about caste". In India
people are very conscious, even inquisitive about a person's caste and
religious identity. The Guru worked towards an ideal which would eradicate
the very concept of caste from society. In his poem Nirvrf,ipancakam he states:
Ko nama desa: kajati:
Pravritti: ka kiya dwaya
Ityadi vadoparatir~
ra.rya ta.ryaiva niTV.riti:
mo~a is attainable only by one who transcends divisive

contentions such as caste, nationality, and occupation.)
(Nirvriti i.e.
In his poemJiitinirr;.ayam he wrote:

j\!Jan~yar;.am
manuD'atvam
]atir,'gotvam gavam yadha

-Na Bra/lmar;.iidira.ryaivam
Ha, tatvam vetf:(i kOpi na
fJatinirr;.ayam verse I) 9
Human beings belong to the human caste,ju.st as all cows belong to the
bovine caste. Brahmins etc. are not the-caste of human beings. Alas! No
one realises this truth.
According to him all human beings belonged to one caste - the human
species. In Madras he told some visitors:
When one dog sees another it immediately recognises it as a dog. All animals
have this faculty of recognising their own species. Only human beings have
doubts. They cannot recognise their own species. They are worse than
animals! 10
Caste, however, was a difficult issue to tackle. As we know, it is underwritten by many of the teachings and scriptures of Hinduism. The mythology of the Puru~a Silkta, and Dharma Siistras such as the Manusmrti support
caste. Manu stipulated that to carry out the duties of another caste is
adharma, sin. In the Riimayar;a, Rama beheads Sambhuka, a fildra who dared
Freeing the Tryyas: Nariiyar;a Gum and Religious Ideology in Kera/a
I59
to perform tapas and was said to have caused the death of Brahmins by this
blasphemous act. The karma doctrine justifies the low status of si1dras and
untouchables. The Guru quoted from the Asri Samhita:
Janmanii Jayate fiidra
Sa7!lSkiira ryr dwija ucyate
II
All are born as fiidras. It is through sa7[lskiira (self improvement) that one
attains Brahminhood.
The Guru was well aware of what he was up against. He used to cite the
instance ofViswamitra. Though through intense tapas he acquired enough
spiritual power to create a new heaven, Vasista refused to recognise him as
a Brahmin. The Guru remarked that if it was so difficult for a kfatriya, who
is only one inch away, figuratively speaking, from a Brahmin, how much
more arduous would it be for an untouchable or a si1dra who is so far
removed, to be recognised as equal to a Brahmin! In Kera~a Brahmins
were considered to be the visible gods. The respect and deference meted
out to them was unparalleled elsewhere in India. K. G. Narayanan states
that to legitimise their hegemony they created the legend of Parasurama
raising Kera~a from the sea by throwing his axe from Sahyadri (the
Western Ghats) and then distributing the land reclaimed to 64 Brahmin
families. Until recent times most of the landowners in Kera)a were
Nampiltiris, and the si1dras and Tiyyas their tenants. The Brahmins
instituted the rigid and severe rules of social intercourse which were heavily
weighed on the side of their social and economic advantage.
NarayaI,la Guru's Vedantic icieo1ogy was helpful for the eradication of
caste. He taught his followers that he could see the primal soul form, the
Brahman in everyone, and not caste differences. In his poem .lItmopadesa
Satakam he comments on this ontological identity:
Avanivarennariyunnatokke arthal
AvaniyilJidimamiJyoriitma rilpam
(Atmopadefa Satakam 24)12
That we know as me and he are, ifwe reflect, the primal soul form of
the world.
According to him there was only one caste - the human caste. To refute
that Hindu mythology invariably supported the caste system, he pointed to
the instance of the sage Parasara muni who was born to an untouchable
mother, and Pakkanar and the Peruntaccan and ten others of Kera)a
mythology, who were avatars ofViglU born 'toa Pariah woman. According
to the lvlahiibhiirata the illustrious Bharata dynasty descends from a woman
of the fisherman caste. On the first temple he consecrated the Guru
inscribed in his own hand: "This is a place where everyone lives in
brotherhood without caste differences or religious animosity."
I60
Indianlnsights: Buddhism} Brahmanism and Bhakti
One visitor argUed:

People gain more skill if there is occupational specialisation over the
generations from father to son. The caste system reduces competition and
ensures employment for all.
The Guru replied:
The person who is preoccupied with just one kind of work, without interest
in any other will not be able to do anything properly. Moreover the caste
system does not take into account individual talents and preferences. The
son may not like the father's trade. This is not good for any profession. 13
Rajagopalachari, the first Governor General of independent India and
a Brahmin scholar, during a visit to the Guru's iifram remarked that caste
was a natural phenomenon and that he could identify anyone's caste merely
by looking at them and listening to their accent. The Guru paraded some
inmates of the iifram before the distinguished visitor and asked them to
recite some verses from the scriptures. He then asked Rajaji to identify
their caste. Achari made so many glaring mistakes that he felt greatly
embarrassed and retracted his proposition. 14 Naraya;r;ta Guru admitted
many members of the lowest of the untouchables such as Pulayas and
Pariahs to his iiframs and integrated them into the iifram community. They
often cooked the iifram meals and sat down to eat with members of higher
castes. Some of them were trained as priests and officiated at ceremonies in'
the temples that the Guru established. He exhorted the untouchables to
give up their caste occupation and engage in other professions. Toddytapping was the traditional occupation of the Tlyya caste, a profession
which inevitably led to excessive drinking and consequent .evils in the
community. The profession made others look down upon the Tlyyas.
NarayaI,la Guru said:
Liquor is poison. You should not manufacture it, supply it, or drink it. The
tapper's body stinks, his dress stinks, his house stinks, everything that he
touches stinks.
NarayaI,la Guru admitted every one irrespective of caste distinctions to
worship in the temples that he founded, even the lowest categories. This
was, in most instances, against the protest and disapproval of many nyya
leaders. He advised Tlyya parents to give Sanskritic names to their
children, hitherto the privilege of Brahmins. He asked them to give up
worshipping the deities relegated to them by the Brahmins and worship
Hindu deities. This Sanskritisation-process improved their self-image and
confidence, though it incensed most of the sava17Ja castes.
During their meeting at Sivagiri in I920 Gandhi had a long discussion
with the Guru on his' caste and religious ideology. Gandhi, who was a
believer in va17Jiiframadharma, opined that the caste system followed nature.
Freeing the Tfyyas: Niiriiyar;a Guru and Religious Ideology in Kera,la
16r
He cited the analogy of the leaves on a tree which are of different sizes.
The Guru replied that the differences were superficial and extending the
analogy said that though the leaves were not of equal size their juice tasted
alike. 15 He continued that although individuals seemed different they were
all manifestations of the same essence, basing his argument on Advaitic
theology. Gandhi said that the Guru's reply was very convincing. Gandhi
was highly impressed at the way that the Pulaya and other untouchables in
the iifram recited prayers and Upani~adic verses. Nitya Chaitanya Yati states
that it was the Guru's emissary T. K. 1Vladhavan who impressed upon
Gandhi the need to include untouchability in the action programme of the
Indian National Congress. I6
Religion
Though by upbringing NarayaJ:.la Guru was a Hindu, he was essentially a
pluralist as far as his approach to religion was concerned. The influence of
Buddhism on his thinking is quite apparent from what has been said so far.
But he respected all religions. In the wanderings of his early years, when to
some extent he suffered ostracisation from his community due to his radical
ideas on caste, it was Christians and Muslims who befriended him. He had
enjoyed commensal meals with Muslims eating from the same plate, as
Muslims are wont to do, and listening to Qur'an readings and taking part
in religious discussion with them. He used to state that the purpose of all
religions is the same. He said of this fundamental unity in diversity that
when rivers merge into the ocean one cannot distinguish between their
different waters. In the poem Sri NariiyaT}a Smriti he wrote
Aneka gin sambhiita
nimna geva mahOdhadh'im
Ekam saniitanam vastu
Priipya viframyiiti swayam I7
Just as rivers flowing down from several mountains merge into the ocean
all religions merge and rest in the one Eternal Being.
As with caste this personal hermeneutic of religions was based on his
advaitic beliefs. He stated that religions have authority only to set the
jlvatma in the right direction. His poem Mata lvIzmiir[lsa begins with the
words "Pala matavum ifka siiram". Religions are many but their essence is the
same. He continued:
Not seeing this like the blind men and the elephant
The unenlightened argue ceaselessly
Which observing remain steadfast undisturbed.
(AtmopadeJa Satakam 44)IB
162
He used to quote the passage trom the altii in which Knr.la stated that
he would appear to the devout in whatever form they meditated upon him.
Another of his oft-quoted sayings is "iV1atametiiyiilum manuDan nanniiyiil mati
(It is enough if a person is good whatever may be his religion)". This
reflects his belief in the moral and ethical efficacy of all religions. Knowing
that God is one he seemed to stress the humanist aspects of religion rather
than creeds and practices. It is noteworthy that he made the Tlyyas give up
many rituals such as animal sacrifice, the ecstatic and self-mortifYing sword
wielding oracles (KOmaram), many of whom were fraudulent, and unnecessary rites of passage such as the tiilikettil kalyiiTJam (symbolic pre-puberty
marriage).19 To him they were inessential, needlessly expensive and even
cruel. The TIyya and other untouchables had gained notoriety in what was
termed black magic spells for charming girls, visiting sickness on enemies,
lyncanthrophy etc. The Guru asked them to abjure these practices.
He held an all-religion conference in the Advaita iiSramam in Alwaye
(Aluva) during 1924- The motto of the conference emblazoned on banners
at the conference venue was:
Not to argue and win, but to inform and be informed.
In his inaugural speech the Guru confessed his pluralistic approach and
that he was instructing his disciples to adopt the same attitude to religions.
NarayaI).a Guru realised that religion, like caste, was a powerfully divisive
force in India. He exhorted the study of all religions and believed that the
scriptures of an religions, like philosophical, scientific and artistic works, are
the common heritage of humankind. He opined that the Buddha and
Christ had no use for dogma and creeds. They sought for ways of salvation,
found them and preached them to others. The saying "Vedam
aupaurusheyam" is usually interpreted as implying that the authors of the
Vedas are unknown. The Guru had a different interpretation which
postulated that the Vedas expressed ideas that were impersonal i.e. not
confined to any particular religion or person. 20 In 1916 he went further.
He wrote that he did not want to be associated with any particular religion.
He proclaimed that he transcended both religious and caste identity. He
admitted that he had established some temples at the behest of Hindus, but
was ready to build places of worship for Christians and Muslims. 21
Speaking at a SNDP conference he stated that the term Erava did not indicate any particular religious or caste group. He said that anyone should be
permitted to join the organisation irrespective of his caste or religious identity. Similarly in his discussions with C. V Kunhiraman, who to some
extent was a critic of the Guru's policies, he said that he was ready to admit
Christians and Muslims into his iisrams.
The Guru pointed out that there is striking commonality in mystical
experience in all religions. He also argued that if the Vedic, Dvaitic,
Advaitic, Visi~tadvaitic, Vaisqika, Mim8.Ip.sa, etc. can all be designated as
Freeing the Tryyas: NiiriiJar;a Guru and Religious Ideology in Keraf,a
r63
Hinduism, the different ways, which in conformity with-particularities of

time and place various spiritual leaders have preached for achieving mo~a,
could be brought under the umbrella term ekamata (one religion) or
manuvamata (human religion). 22 This approach seems similar to
Radhakrishnan's concept of a universal tree of faith of which the historical
religions are different branches. In his Eastern Religions and Western Thought
Radhakrishnan opined that unless the different religions of the world transformed themselves into the Universal Faith all of them would eventually
fade away.23 The idea underlying what Radhakrishnan termed the
'Universal Faith' and Narayal).a Guru's ekamata seems to be the same. Both
believed that religions being in conflict with each other and not unifying
would eventually spell their extinction. Following the same line of thinking
in his debate with C. V Kunhiraman, one of the founders of the Rationalist
Society of Kerala, the Guru referred to the diversity of Hinduism, stating
that the 280 million Hindus of that time in reality could be said to follow
280 million religions. However, since the beliefs of all of them had some
common characteristics they are all considered to be followers of one religion, viz. Hinduism. He therefore concluded.that since all religions in the
world had common features all of us could very well be considered to be
followers of one religion which he would call man~yamata - Human
Religion. The Guru seems to be implying that the disparities between them
were not so significant that they could not be subsumed under one conceptual system. The Guru said that in order to combat the rampant violence
of Buddha's times the Buddha emphasised non-violence as the main principle oflife. Muhammad stressed brotherhood, which was a need in the
divided Arabia of his time. In India today the need was to contain caste
and religious conflict. Thus the idea of ekamata or manuvamata is relevant.
This seems to be pragmatism pure and simple, but gave Narayal).a Guru a
rationale for the plurality of religions. As an Advaitic and a mystic he anyway did not care much about doctrinal and ritualistic disparities. He said
that when we study all religions impartially and respectfully we will realise
that the conflict is not due to mata (religion) but mada (arrogance).
A keen European missionary visited the Guru to try and convert him.
The missionary told him, "you must join the Christian faith". The Guru
asked him "how old are you?". The missionary replied" I am thirty". The
Guru said "then I have been a Christian long before you were born". 24
However, in another context Narayal).a Guru did extol the determination
and strong faith of the Christian missionaries. 25
Some leaders in the untouchable and low caste communities advocated
conversion to Buddhism, Islam or Christianity to escape caste-based
discrimination. It may be recalled that Dr Ambedkar led many untouchables to Buddhism. Narayal).a Guru felt that such ~onversions were
superficial. He was not in favour of religious conversion that was not
prompted by genuine conviction and change of belief. Narayal).a Guru
164
pointed out that f~llowing the Nyaya tradition he felt that each individual
has his own understanding of religion. However, he often stated Jhat he
believed in the freedom to change religion. The son may not like the
father's religiOri. Though he believed that religious freedom is essential, he
felt that one's religious affiliation itself could not guarantee one's salvation.
Though he admired and respected Buddhism, Christianity and Islam he .
felt that there were bad Buddhists, Christians or Muslims just as there were
bad Hindus. He asked his disciples why some TIyyas wanted to abandon
the Hindu faith. They replied that they considered the Hindu scriptures
vitiated by animal sacrifice, polytheism and caste differences. The Guru
replied that he could still find some useful truths in the Hindu scriptures. 26
However, whatever may be the religious literature it was more important
that people behaved properly. They have to be pure in action, word and
thought, a sentiment which echoes the noble eightfold path of Buddhism.
He said that it is no use if the scriptures are good but its followers are evil.
His often quoted slogan was matametiiyiilum manuvan nanniiyiil mati (It is
enough that one is good whatever may be one's religion). He was quite
pragmatic and liberal on this as on many other issues.
One God
The Guru's theism was firmly based on Advaita (monism; non-dualism).
He was a self-confessed follower of Sankara's theology, with the proviso
that he did not concur with varTJii.framadharma. It is strange that Sailkara
compromised his rigorous monistic philosophy to the extent of supporting
the caste system. Kesavan Vaidyar opines that Sailkara accepted it as a
lower reality caused by avidya, the innate ignorance clouding our
perception. 27 The Guru saw God in all individuals including himself In
accordance with the concept aham brahmiismi (I am Brahman) he installed a
mirror in place of an idol in the temple he set up in Certala. He was quite
radical, though fully in consonance with his Advaitic beliefs, in installing
images in temples. In one temple he consecrated a burning lamp as the
deity. In another the words saryam, daya and dharma (truth, compassion and
righteousness). At the .AIuva Advaita temple he preferred to have no image
at all. He sought to raise the TIyya's theism to a high intellectual and
philosophical level. He asked them to abandon the gods to which the caste
Hindus had relegated them, such as Cattan, Caml1ndi and Bhmattan, and
which gave them a feeling of inferiority. In his poem Deiva Dasakam he
states that the creator, the created, and the material for creation are one
and the same (Deiva Dasakam 5). In one temple he had a ploughshare
installed as the deity. A disciple removed it and placed the Guru's sandals
in its place. "When the Guru questioned this action the disciple said that
they had vowed to worship only the Guru and nothing else. The Guru
asked, "If my sandal is myself, why not the ploughshare?". 28 The
Freeing the Tfyyas: Niirf!ya1JD. Guru and Religious Ideology in Kerafa
r65
fundamental unity of Advaitic ideology helped him to be compassionate to

others. He used to say that if the iitman in everyone is the same then the joy
~f another is our joy and their pain is our pain. In his poem Atmopadesa
Satakam he reiterates this concept:
Him, Me, all these that we think
Is really the primal soul form in the world.
What we do for our joy should work for the joy of others.
(Atmopadefa Satakam 24)29
Though he was avowedly a follower of Sar'J.kara's philosophy it is a fact
that Narayal).a Guru promoted the bhakti miirga among his followers. An
examination of his literary output reveals that a number of his compositions were praises to God, and that not all of them were directed to Siva.
Poems such as ViisudevaSatakam and Sf[ Krsna KZrtanam are directed to Visnu
rather than to Siva. In this he seems t~ be following Ramanuja, ~~t
Sankara. However, his loyalty to Vedanta was unflinching and provided a
strong foundation for his belief in the identity of all religions and deities.
The poem EsavaD'opanijad bhii~a is a rare hymn of prayer and praise to
Brahman, the impersonal being, and in addition an exposition of Advaitic
theology. His longer and more fervent works of praise such as Sivastuti and
$ar;mukha stotram though reveal his Saivite slant. However, it is clear that
Narayal).a Guru was not very much concerned about what people
worshipped. He believed that God was one and was impersonal. He was
happy for Christians to worship Jesus, Jews to worship Jehovah, Muslims
Allah, and Hindus any of the various deities of Hinduism. He does not
seem to have been particularly concerned about the ontological particularities of the Hindu deities. K. K. Pannikkar quotes that Narayal).a Guru
had opined that some of the Hindu gods may have been human beings
who became deified in course of time. Siva might have been a forest
dweller who with his great deeds of strength and valour became
prominent in people's consciousness. Similarly Rama might have been a
leader who with his good nature and benevolence earned the praise of his
people. 30 The Guru also pointed out that K~emendran in his
Dasiivatiiracarita had made Buddha an avatar of Vi~l).u. The objective
reality of these gods was not important to him. Once he told his disciples
that it would be enough if they placed around the temple lamp portraits of
illustrious individuals as deities. He saw God in all of them.
Conclusion
Narayal).a Guru was a remarkable individual of our times. He was not just
another of the many godmen of India, some of whom have been
discredited as seekers of wealth and fame. Narayal).a Guru was only
166 Indian Insights: Buddhism, Brahmanism and Bhakti

interested in the welfare of his country and community. The sphere of
influence of some of the great Indian l'eformers such as Ram Mohan Roy
and Vivekananda was among the elite and the intellectuals of Indian
society. NarayaIfa Guru came down from the sylvan surroundings of his
mountain abode to serve the poor and downtrodden sections of society.
His principal objective had been to try and ameliorate the divisive
nature of Indian society. For this he did not assume a confrontational
approach. He, sought to educate society through his teachings, jokes and
repartees and through his writings. His principle: in reform was dvugdhvam
jfiiiniigniniim sarvam (In the fire of knowledge everything is destroyed). To him
God was formless and unknowable except through self-realisation, in his
own words the transcending of the triputi - the known, the knower and the
knowledge. This is attainable in various ways. This theology enabled him
to transcend religious distinctions and the philosophical problems of a
pluralistic approach to religions. He remained firmly rooted in an
undogmatic stand undaunted by powerful orthodoxy and centuries-old
traditions. He was a firm and uncompromising believer in the unity of
humankind. In the AtrnopadeSa Satakam he states:
Everything exists, it is true. However the philosopher understands that all is
one. If we do ~ot look inward Maya plays many tricks on us.
(iItmopadesa Satakam 88)31
Notes
For the transliterarion of Malayalam. words I have employed the
scheme given in Gaur, Albertine, Catalogue qf JIIlalayalam Books in the
British Museum, London, London, Trusteee of the British Museum, 1971.
1.
2.
3
4.
5.
6.
7.
8.
9.
!D.
II.
12.
13.
14.
IS.
16.
17.
Vivekanandan, 1986, p. 27.

Narayar;an, 1984, p. 247.
Vivekanandan,1986, p. 15
Kunhirarmi.n, KBrafa Ganl:[am, pp. 971-2, quoted in Naraya'!an, 1984, p. 22.
Naraya'!an, 1984, p. 266.
Vivekanandan, 1986, p. 37.
Vaidyar, Kesavan, Sfi MirqyCl1Jl!. Gintaka? (Thoughts of Sri NarayaI;1a), Kottayam: D C Books,
1990 , p. 93
Quoted in Sanu, 1986, p. 507.
Quoted in Bhaskaran, 1991, p. 41.
Vaidyar, Kesavan, Sri MirqyCl1Jl!. Guru Swantam Vacanan1JfJ.fi/iite (Sri Naraya,!a Guru through His
Own Words); p. 21.
Quoted in Vivekanandan, 1986, p. 266.
Guru, Sli Naraya'!a, 1991, p. iI6.
Vaidyar, Kesavan, Sfi Mirqya1JfJ. Guru Swantam VacanalJ.1'!llfi/iite, 1990, p. 24.
Yati, Ntya Caitanya, SfiMirqya1JfJ. Guru, p. 19.
Sanu, 1986, p. 441.
Yati, op. cit., p. 20.
Bhaskaran, 1991, p. 41.
Freeing the Tfyyas: Naraya7jil Gum and Religious Ideology in Kera!a

18.
19.
20.
21.
22.
23.
2425.
26.
27.
28.
29.
30.
31.
167
Guru, Sri NarayaI)a, Iggl, p. IIg.

MOrkot, Kunhappa, 1988, p. 69.
Sanu, op. cii, p. 4)1.
Vivekanandan, op. cit., p. 183.
PaI)ikkar, 1989, p. '281.
Radhakrishnan, 1940.
Vaidyar, Kesavan, Sri NiiriiyalUl Gum Swantam fiacana~lUl!ifiite, p. 59.
Ramachandran, 1981, p. 68.
Vaidyar, Kesavan, SnNiiriiyalUl Gum Swantam VacanarylUl!ifiite, p. 58.
Vaidyar, Kesavan, SriNiiriiyalUl Cintaka!, pp. 31-2.
Vaidyar, Kesavan, SriNiiriiyalJ!l GumSwanlam Vacanary~a!ifiite, p. 71.
Guru, Sri NarayaI)a, 1991, p. II6.
PaI)ikkar, 1989, p. 270,
Guru, Sri NarayaQa, 1991, p. 124.
Bibliography
Bhaskaran, T, Sri NarayaT}a Gumvinte Kavya SaraT}i (The Poetic Path of
Sri NaraYaI,la Guru). Kottayam: D C Books, 1991.
Guru, SrI Narayal),a, Atmi5padesa Satakam, Deiva Dasakam, Sarvamata
Saram. Aleppey: Vidyarambham Publishers, 1991.
Karunakaraswamy, Sri Guru Bhagavatam. Aleppey: Vidyarambham
Publishers, 1991
MurkiSt, Kumaran, Sri NarayaT}a Guru Swiimika?ute Jzvacaritram
(Biography ofSrt Narayal),a Guru Swarnikal). Calicut: P
Brothers, 1930.
MurkiSt, Kunhappa, KuttikaJute Sri NiirayaT}a Gum (Children's Sri
Narayal),a Guru). Calicut: Mathrubhumi Printing and
Publishing Co., 1988.
Narayal),a, K. G., Erava- Tfyya Caritra Pathanam (A Study of EravaTIyya History). Kayamkulam: Anaswara Publishers, 1984.
Pal),ikkar, K. K., Sri NiirayaT}a Paramahamsan (in Malayalam). Aleppey:
Vidyarambham Publishers, 1989.
Parakkattil,jose, Samilhya Rangathe Vip/avakiiri (Revolutionary in the
Social Scene). Kottayam: Auroville Publishers, 1991.
Radhakrisnan, S., Eastern Religions and Western Thought. London: 1939.
Ramachandran, K. M., SrrNiirayaT}a Guru (in Malayalam).
Kottayam: Kairali Children's Book Trust, 1981.
Sanu, M. K., NiirayaT}a Gum Swami (In Malayalam). Kottayam:
. Sahitya Pravarthaka Coop. Society Ltd, 1986.
Vaidyar, C. R. Kesavan, Sri NiirayaT}a CintakaJ (Thoughts of Sri
NaraY3.1;a). Kottayam: DC Books, 1990.
- - - , Sri NiirayaT}a Guru Swan tam VacanaT}T}aJililte (Sri Narayal),a
Guru through His Own Words). Kottayam: DC Books, 1990.
Vivekanandan, K., Sri NiirayaT}a Darsanavum Manava iVIunnettavum (Sri
Narayal),a Guru's Philosophy and Human Progress).
168
Indian Inszg;hts: Buddhism, Brahmanism and Bhakti

Eranakulam: The Author, 1986.
Yati, Nitya Caitanya, SriNarayaJJa G/1ru. Sreenivasapuram: The EastINest University ofBrahmavidya, undated.
Ritual Dance in Kerala: Performance,

Possession, and the Formation of Culture
Gavin Flood
Kerala is a land rich in natural resources - water, trees, fish, fruit - and rich
in cultural and religious traditions, with Christians, Jews, Moslems, and
Hindus sharing a comparatively small area with a high density of
population. Each year throughout Kerala the festival season begins around
November and carries on through until about March. Part of the festival
scene is theteyiit/am, 'the dancing of the tl!Jiyams', the dance-possession rituals
which occur at innumerable shrines throughout the land. In this paper I
wish firstly to describe one of these festivals, to convey an impression and
underline the vibrancy and startling impact of the teyyam culture, and
secondly to suggest some ways of understanding it. The paper will focus on
three areas, performance, possession and the formation of culture and will
provide an example of the ways in which the 'religious' cannot be separated
from the 'cultural'. The paper arose out of a preliminary research trip in
I993 to investigate Tantric traditions in Kerala, where I met Dr Rich
Freeman who has been worIGn:g on the teyyam festivals for some years and
who first introduced me to the Kerala traditions. I refer the interested
reader to his extensive work for a full account of the teyyam phenomenon. I
The Deities
There are a variety of social groups in Kerala, indigenously classified as
either savarr;a, within the 'class' system, or avarr;a, outside of that system.
The Nambudri Brahmans are savarr;a, while all other castes are classified as
avarr;a or fildras, who traditionally have kept a flXed distance from the
savarr;as. 2 As there are a variety of social groups, so there are a variety of
deities. People of all castes revere the high, pan-south Indian deities such as
Aiyappan and MurukaJ}, and their various manifestations and incarnations.
I70
They revere low status deities such as the various local goddesses,
particularly Mariyammal}, the 'smallpox' goddess, and they revere deities
associated with family lineages and locations, such as goddesses linked to
particular Nambudri family homesteads or illams. The shrines which are
the focus of the teyyam cults belong to the lower, ava0a castes and the deities
celebrated in the festivals are low caste deities.
The teyyam deities, whose Malayalam name is possibly derived from the
Sanskrit deva, are generally ferocious, and while being sometimes associated
with the high gods of the Hindu pantheon, they tend not to lose their fierce
natures and particular identities. They are also associated with the class of
deities known as vzrans, heroic protectors and guardians of the temple. In
the teyyam rituals, a performer, dressed and adorned as the god, becomes
possessed by the deity and 'dances' or struts around the shrine in a
possessed state, giving the 'sight' or 'vision' (darsanam) of the deity to the
onlookers. He will perform a sequence of actions which include singing the
songs of the teyyam and, if there is one, reciting the narrative associated
with the teyyam. Most towns and villages have a shrine or collection of
shrines, which house a number of teyyams, perhaps the most famous of
whom is Mutupan, a composite image of two hunting figures and a dog.
The teyyam performance described here, which Dr Freeman and I
witnessed, occurred at N'ilesvaram in the north of Kerala in October I993.
Here within a sacred enclosure are three shrines housing SL,{ teyyams - two in
one shrine, three in the central shrine and one in the remaining shrine. The
shrine is the property of the Caliya or weaver community, though other low
caste groups perform the teyyams: the Nlalayan caste of professional sorcerers
and teyyam performers, and the Vawan caste of washers, of lower status
than the Malayans, who also practise ayurvedic medicine. The shrine
complex itself is called the vzrankavu, the 'sacred grove of the heroes'. Indeed,
this shrine complex, as many others in Kerala, had a sacred grove attached
to it; an area which is kept wild and which might contain a shrine to the
deity of the grove within it. Within the shrine area enclosed by a low wall
there are the three shrines (see plan opposite), two buildings in which the
two castes performing the teyyam change, and a building in which the elders
of the cornmunitysit to watch the proceedings, where people are weighed
against fruit which they then offer to the shrine deities. To the north and in
front of the the teyyam Gujikan's shrine is the kalalatara, the altar or mound
(tara) for the toddy pot (kalala). Toddy is an alcoholic beverage made from
palm juice, offered to the deities and drunk by the ritual participants. The
pot is an elaborate, ornate structure which is cicumabulated around the
shrine before being offered to the teyyam.
The teyyam festival lasts for about two days, with each deity being
performed in turn by a single dancer from each of the performing castes
honoured to perform on that occasion. The particular dancer is chosen by
the elders of the Caliya community. It is an honour to be chosen, and the
NI
kala~atara
(1)
~
Pl
Shrine 3
Shrine 2
Shrine
Z~
o Pl
8~
en
Pataviran
GuJikan
n
or:
g-:S
Culiyarbhagavati
MuvalamkujicamuI).c;iI
~
g..~
'
".9,
::; n
'"
CJq
ft
"
GROVE (Kavu)
Pl
rl-
~z
--
1--'1
>-'
(1)1
Patakulannarabhagavati
tl fJ
CJq
VisI).umurti
.
::; '
(l
~'
172
dancer will perform all the teyyams permitted by his caste. This will iIlVolve
being awake for long periods of time and dancing for several hours as. one
deity follows the next in sequence. The teyyam dancer is decorated and
adorned in the image of the deity and moves through a sequence of
prescribed actions. The events begin with the rapid drumming (tutaiuia0 by
a small group of musicians. The performer then begins his dance by the
kala!atara altar, singing praises to the deity, telling the teyyam's story, and
performing invocatory rites. Upon seeing his reflection in a mirror which
has been given to him along with a sword, the manifestation of the teyyam in
the shrine, the performer becomes possessed by the deity. He walks or
dances to the shrine of the teyyam, which is now his own shrine, sometimes
supported by attendants, to the accompaniment ofthe rapid, rhythmic beat
of the drums. Once before the shrine he sings or chants in l\IIalayalam a
series of laudations to the deity, praising the deity first in the third person,
then in the second person and finally in the first person and the possession
is complete. The performer enacts out the teyyam mythology, tells of the
teyyam's deeds and parades around the compound taunted by a crowd of
young men to whom he responds with varying degrees of aggression. 3
The teyyams of each shrine are enacted in turn. The first to be performed
are a male deity Visr;l.Umurti and a goddess MuvajamkujicamUl].9L
Visl).umurti is a local deity who is associated with Narasirp.ha. In the myth, a
landlord of the Nayar caste mistreated and forbade his servant to worship
Visl].u, so Visl].umurti was empowered by Visl].u to exact vengeance upon
the Nayar landlord, which he did by tearing out his entrails. This is, of
course, directly analogous the the story of Narasirp.ha, with the wicked
Nayar landlord replacing Hiral].yakasipu and the servant replacing
Prahlada. This myth is repeated by the teyyam and the killing of the Nayar
landlord enacted before Visl].umurti's shrine.
MuvajamkujicamuI).9i is even more ferocious. She was the goddess
invoked to protect a Nambudri against another who was magically attacking
him. The attacking Brahman then ensnared the goddess through the power
of his mantra, put her in a copper vessel, sealed the lid, and buried her in a
hole. Furious, the goddess burst out of the vessel and chased the Brahman
who ran, terrified, to the Trikanyalapan Siva temple for protection. Here
the goddess agreed to settle down if she could be installed there beside Siva,
which duly happened! As with the story ofVisl].umurti, this story is told by
the teyyam dancer in front of the shrine before he becomes possessed by her.
MuvajamkulicamuI).Qi is perhaps the most ferocious, with a massive
headdress she parades around the compound, taking sacred weapons from
the shrine, such as a sacred bow, swords and shield, given to her by the
priest, and, taunted by a crowd of young men, she symbolically (and less
symbolically) attacks them. This symbolic, and sometimes actual, violence is
performed in a state of possession in which the dancer trembles and
breathes rapidly, while still retaining enough control to perform the
Ritual Dance in Kerala
173
required sequence of events. Sometimes, however, the teyyam becomes very

ecstatic and injuries, rnainly to the teyyam, have occurred.
The middle shrine at NIlesvaram contains a god, Patavlran, and two
goddesses, Pata.kulari.ri.arabhagavan and CuliyarbhagavatL This teyyam is
performed by the Val).l).ans. There is no surviving mythology about
Pataviran and we know nothing about him except that he is connected with
a war between two families (gotras). Patakulari.ri.arabhagavan (whose name
refers to the place of her origin and that she is fond of singing) was a Nayar
household goddess who migrated with families as they moved and ended up
in the teyyam shrine at NIlesvaram. Her traditional role in a number of
Nayar temples dedicated to pan-Indian gods has been to take food offerings
to the deity and to the Nayar guardian Agampati. She is sometimes associated with Bhagavan or Bhadrakali, the Nayar goddess of war, epidemics,
land and fertility who is worshipped in smaller, private Nayar temples. The
other goddess of the shrine, Culiyarbhagavan, is a war goddess associated
with the royal Nayarhousehold and with the Nambudri tantris (Brahmans
who install images in temples) but has no mythology associated with her - ,
or at least any mythology she might have had is forgotten.
The last teyyam to be performed is Gulikan, a local deity associated in
mythology with a second death which emerges after Siva has destroyed
Yama. Gulikan is also the name of the invisible planet causing calamities
(and is again associated with Yama in this way, who is associated with
Pluto). All these deities are 'hot' and impure in contrast to the pure, 'cool',
high gods of the Brahmanical Hindu pantheon. They are local deities with
local mythologies, sometimes associated with hunting, whose status is
elevated by being associated with deities of the pan-Hindu pantheon.
One of the most interesting occurrences during the performance is
when the deity moves out ofthe compound and dances through the streets
of the town to a local temple whose central deity is a form of Siva. Here the
teyyam knocks at the door to demand entrance but is not allowed to enter,
although there may be verbal exchanges between the teyyam and the
Brahmans inside. She then returns to the teyyam shrine giving her darsanam
to people on the way and entering houses and so blessing them. Once back
at the shrine, she again parades around the compound and then is seated
upon a stool by the toddy-pot altar (kala!atara). The toddy pot, which has
cicumabulated the shrine, is offered to the teyyam and a chicken is sacrificed
by beheading. Its blood is put into a bowl, which is then tipped out as a bali
offering to the teyyam. With the sacrifice cifthe chicken (or several chickens)
the rite is more or less at an end and the teyyam comes back to a condition
of normality, eventually retiring to the changing rooms. 'With the end of
one performance there are a few hours to wait for the next, when the same
pattern of possession, dancing around the compound, visiting and being
rejected from the temple, the toddy offering and chicken sacrifice is
repeated.
174.
The tf!Yyam rituals ~re open to ma:O:y levels of understanding and we can
observe here important structural and semantic processes concerning the
transmission of values and the formation of culture. I wish here to highlight
three levels at which we can understand the tf!Yyam's dance: a level of
performance, a level of possession and alevel of culture formation.
Performance
A distinction can be drawn between ideology, which is primarily cognitive,
and performance which is primarily active and affective. It might be
argued that the cognitive dimension to performance is minimal and that
performance does not reflect a reality external to it. I would, however,
argue that the tf!Yyam contains both cognitive (and therefore ideological)
and also performative dimensions. That is, the performance, among other
things, expresses an ideology, by which I mean a system of belief, its
expression in power, and a legitimation of power relations. While there
may be other aspects of life and personality expressed in the tf!Yyam
performance, it is the expression in performance of a cognitive, ideological
dimension which illumines the nature of the tf!Yyam phenomenon.
The tf!Yyam is totally embedded within the cultural and religious matrix
of the Malayali communities of Kerala, and loses all meaning outside of
those contexts. While there are obvious iconographic connections with
Kathakali theatre, unlike the Kathakali, the teyyam could not be exported.
Indeed, I would guess that Kathakali is a: development from the dance
possession of the tiyattam. But whereas in Kathakali the audience observes a
performance which is removed from the gods it depicts, in the tf!Yyam the
audience is totally involved in the performance whose gods are not '
removed, but immediately present. In the tf!Yattam the audience and the
gods are all on the stage.
In The Future qfRitual, the performance theorist Richard Scheckner lists
six templates for the analysis of play. These six are: structure, the
relationships among events constituting play acts; process, the way play acts
are generated and the:ir phases of development; experience, the emotions of
the participants;. function, the purposes served by the play act; ideology, the
values communicated and criticized, knovvingly and unknovvingly; and the
frame, the way in which the performers and audience know when the play
act begins, is taking place, and is over.5 Though Scheckner does not place
these categories in any hierarchy, some are arguably more significant than
others and the tf!Yyam can be examined in the:ir light.
Taking the first two together, we can see that the structure and process
of the tf!Yyam expresses social relationships and the self-perception or
ideology of the community vvithin the vvider Hindu context. A number of
factors have to be considered here, particularly the spacial arrangement of
175
the shrines and the relation of the performers to the onlookers, what might
be called the social geography of the teyiipam. The rectangular shrine
compound itself, the vlrankavu, is enclosed by a low wall, a sacred space set
off from the everyday world of work. Within this sacred space, having
removed their footwear, people move about reverentially, except when the
teyJiam is dancing and the crowd follow the deity in a boisterous mood.
Within the sacred space of the mrankavu the three shrines are located next to
each other. These shrines are only approached by those in authority: the
oil-lamp lighter who keeps all the lamps burning during the festival, and
the priest who administers to the god, handing the teyyam dancer the deity's
weapons and shield. Castes lower than the Caliyas, other than the teyyam
performers, are not permitted within the shrine area, although on the
occasion here described, Dr Freeman and myself were exceptionally
allowed to observe events from a privileged position upon a teyyam shrine,
thanks to a local dignitary.
The relationships between events which constitute the teyyam are
significant in making statements about social relationships to do with group
status and gender. The women stay back from the teyyam, while it is the
young men who let off steam through chasing after the deity and risking a
few bruises from the goddesses' wrath. The teyyam can assault the crowd,
but is calm when he goes before the elders of the community who receive
his darsanam. But one of the most significant sequences in the festival
process is the marching of the teyyam outside of the compound into the
streets, into the world of work and people's homes, and marching to the
temple administered by the Nambudri Brahmans. Here the teyJiam does not
enter the temple. The low caste deity does not sully the purity of the high
caste god, as the low caste Caliyas would not pollute the Nambudris. As the
vlrankavu is to the temple, so is the teyyam to the high god Siva, and as the
teyyam is to Siva, so the Caliyas are to the Nambudris. Thus the teyyam
performance demonstrates a power relation between deities and between
social groups which is acted out in the social geography of the teyyam shrine
and its relation to the temple.
The teyyam legitimates power relationships between social groups within
the wider society. The social geography of the festival marks out the Caliya
caste from other groups and marks out their deities from the high caste
deities. The social geography of the vlrankavu and the implicit rules about
who can and cannot attend the festival, makes a statement about, and
legitimizes, a system of values; the performance expresses an implicit
ideology. The meaning of the teyyam serves to sustain implicit relations of
domination. 6 The teyatfam is significant, therefore, not only in terms of who
is included in the rite, but, as Bourdieu has brought to our attention with
rites of passage, who is excluded. 7 Within the teyyam an ideology of social
restriction and the containment of groups within defined boundaries is
promoted and reinforced. Lower social groups are excluded from the
176
festival, higher groups tend not to c~me, and the teyyam is rejected by the
high caste deity. Although immensely powerful, the teyyam's power still
cannot meet the power of the high Hindu gods, but nevertheless needs to
be accommodated and absorbed by them, as can be seen in the mythology
of the goddess MUvalarr1kulicarnUIf<;ii:, who is installed beside Siva and so
contained and controlled in his temple.
The values of the Caliya community and, indeed, wider society are
reflected in the festival. The performers, and the deities they embody,
accept the order encoded in the rites. If value is 'intrinsic to ritual', as
Rappaport claims,s then the teyyam embodies and expresses that value
system. The teyyam performance functions to convey the values of the
community and expresses, if not clearly articulates, an ideology.
The ideology expressed in the teyiitfam is not monolithic, but operates by a
number of unspoken rules. It expresses power within the Caliya community:
between the Caliyas and lower social groups, making a statement about
their distinction from those groups, and between the Caliyas and higher
social groups, particularly the Nambudri Brahmans. The relation of the
Caliyas to lower: caste groups is, however, not merely one of superiority, but
also one of reciprocity and a recognition of the power of those groups who
perform the teyyams. The teyyam dancers are structurally lower than the
Caliyas in a caste hierarchy, yet during the festival, the teyyam, who as a
human is a low caste dancer, is the Caliya deity; the consciousness or
caita1!)lam of the teyyam has entered the dancer and he has become that god.
The ideology contained within the teyyam festival therefore speaks from a
number of perspectives, or, in Terry Eagleton'S words, 'from a multiplicity of
sites'.9 This ideology is not 'pure' or unitary, but expresses a number of
conflicting interests and asserts itself through the voice of the teyyams against
the dominant, brahm~ical power, while at the same time endorsing that
order, or power-structure, and establishing its place within it. The teyyam
presents itself against the 'other' of the Brahmanical temple deity, while at
the same time allowing itself to be appropriated within the structural
hierarchy of a Brahmanical, pan-Hindu scheme. The teyyam thus tacitly
endorses the power structure of the Caliya community within the wider
'Hindu' society and legitimates that power. The ritual violence of the~ teyyam
reproduces tacit norms and values, reinforcing the dispositions to behave in
particular ways; reinforcing what Bourdieu has called the habitus, the
dispositions which generate particular practices. lO
Possession and Violence

Eagleton has pointed out that a successful ideology engages significantly
with people's wants and needs l l and the teyyam certainly seems to provide
emotional expression for those involved, and emotion - Scheckner's third
177
template - is very significant in understanding the festival. There is an

obvious emotional exuberance of many participants and the teyyam himself
is in a state of possession. There has been much written on ti"te cathartic
nature of ritual and on role reversals and inversions which occur during
those times; I am thinking here particularly of the work of Rene Girard and
Victor Turner. Girard has argued that the purpose of a festival is the
protection of society against horrifYing violence; that without the safetyvalve of ritual festivals, violence would erupt within and disrupt
communities. Particularly, he argues, festivals cuhninate in sacrifice which
channels violence into the sacrificial victim and a purification of social
tension ensues. 12 The ferocity of the deities of the teyyam and the sacrifice
which is the culmination of each rite, can be seen 1...'"1 this light. The violence
of the teyyam is a violence which, through its expression, renews and purifies
the community. That is, on the Girardian hypothesis, the potential of
violence exists within the community, and when channelled into the
cultural form of the teyyam and the sacrifice of fowls, diminishes the threat
of violence.
However, while this theory. is attractive - the teyyam certainly
demonstrates a controlled violence as do the crowd - the teyiitfam strongly
resists merely one kind of interpretation and the teyiittam does not easily fit
Girard's characterisation of the festival as:
a deliberate violation of established laws ... celebrations in which sexual
promiscuity is not only tolerated but prescribed or in which incest becomes
the required practice. 13
Certainly none of these features are present during the teyyam festivities:
there is no apparent sexual promiscuity, there is no apparent incest or symbolic incest, roles are not reversed and the leaders of the community are not
abused, but on the contrary, retain a dignified distance. N everthdess, there
is a stro"ng element of ritualised violence by the deity towards the crowd and
towards the high caste deity, and some degree of violence from the crowd
towards the. teyyam - at the very least the crowd lets off steam through
taunting the teyyam and lighting innumerable, deafening fire crackers.
The emotion evoked in the crowd is therefore integral to the structure
and process of the teyiittam. The violence of the goddess separates her from
the cool high caste deity in the temple, and thereby excludes that deity,
erases that deity's possibility. Similarly the violence of the crowd might be
said to express the exclusion of other social groups and the exclusion of
women from male social expressions. The ritualized violence underlines
the potential violence within the community and, indeed, beyond the
community in the power of the goddess or god.
vVe are still left with the interesting question concerning the inner
emotional state of the teyyam dancer. The teyyam is possessed by the deity.
This is publicly observable and occurs at a specific juncture of the rite
178
Indian Insights: Buddhism., Brahmanism and Bhakti
when the dancer looks into a mirror and then is entered by the god. The
question which arises to the western observer, so trained in reificatiqn, is
what is 'really going on' in the dancer's mind? In his study of teyyam, Rich
Freeman asked this question to a number of dancers and received a
number of replies, ranging from the dancer's merely performing what is
expected of them, to dancer's entering a trance state in which they are not
in conscious control of what is happening to them. 14
The question about the inner emotional states of the dancers is
interesting. It is really a question about the wayan individual appropriates
cultural symbols and imbues them with personal significance and meaning.
In his work on possession in Sri Lanka, Gananath Obeyesekere has shown
how cultural forms can have private meanings whose genesis must be
sought in the deep, unconscious motivations of those involved. ls This
uncovering of deep motivations is, of course, the realm of Psychoanalysis.
With regard to the teyyam dancers, a similar exercise could theoretically be
performed. Each dancer, while fulfilling a social role, experiences a variety
of emotional states and appropriates the cultural meaning of the teyyam to
his own private realm, in a number of ways.
Yet while this is interesting and relevant to the particular individual
involved, it is not so relevant to the study of teyyam as a social fact. As
Freeman has observed, possession is a 'performative act'.16 Whatever the
inner, emotional state of the dancer, whatever the private state of
consciousness, whatever the personal biography which has led him to
perform that role, all this is irrelevant to the social and cultural expression,
to the performance. 17 The private world of the individual is completely
subsumed beneath the cultural fact of the teyyam, who performs in a
culturally determined and anticipated way. The mask becomes the face;
the dancer becomes the deity at the level of performance. Whatever is the
inner state of the dancer beneath the level of performance - while of
interest to those, such as ourselves, acculturated to the importance of a
private, inner world - it is an irrelevance to understanding the teyyam as a
cultural event. This is not to say that cultural events cannot have deep
motivations, I think that they can, but these deep motivations are
individual appropriations of cultural symbols; the imbuing of culturally
available events with private, and perhaps unconscious, meaning. Nor is
this to say that the origins of cultural meanings cannot in some sense be
unconsc;:ious, but rather, that in the public sphere of cultural meanings, the
unconscious origin is lost and can only be appropriated or re-appropriated
by a particular individual. At the level of performance, deep, personal
motivation does not illumine what is happening.
So far I have described the teyiittam and suggested a number of
interpretations based on Scheckner's templates for understanding
performance. Of particular importance here is the idea of implicit powerrelations within the value system or ideology of the teyyam and the
179
reinforcing of norms or behavioural habits (the habitus). I have also

suggested that while unconscious or deep motivation may be of interest in
understanding the particular teyyam dancer, it is not significant in
understanding teyyam as a cultural event, as a performance. In other words,
there is a clear distinction between deep motivation and cultural event or
performance.
While the teyyam as the personal expression of an individual psychology
is not open to general interpretation - there would be the need for many
interviews with a qualified analyst - the teyyam as a cultural event or
performance is open to analysis and interpretation. So far we have
examined the teyyam as performance at the level of social structures and
power relations, or in terms of ideology and value. I lastly wish to examine
the teyyam as a performance which generates cultural meaning. While this
overlaps with the kinds of thing I have been speaking about so far, this next
level in, as it were, potentially deepens and sharpens the analysis. I take the
generation of cultural meaning to specifically refer to the generation of
signs, an idea which will become clear presently.
The Teyyam and Cultural Meaning

Within Linguistics a distinction is made between competence and
performance. Competence is the potential language available to speakers
and implies language structure, its grammar, phonology and even
semantics, free from any particular context. Performance, on the other
hand, is the manifestation oflanguage use in contexts, which is open to the
kind of analysis known as Pragmatics, the discipline concerned with
understanding of linguistic performance. Pragmatics is the study of the
relation of signs to interpreters of those signs.18 Indeed, it could be argued
that all semantics is in fact pragmatics, for meaning is always contextual.
The development of a pragmatic analysis of the teyyam would
demonstrate the formal connections between linguistic utterances during
the teyiittam and the relation ofthose utterances to the actors and audience
in the performance. This would involve a study of the Malayalam linguistic
utterances during the festival which is beyond the scope of the present
paper. But one example where a pragmatic analysis might be developed is
when the dancer, having looked into the mirror and been possessed by the
teyyam, proceeds to recite the teyyam's praises before his or her shrine, first in
the third person, then in the second person, addressing his remarks directly
into the shrine, and lastly in the first person, indicating that the possession
is complete and that he has become the deity. Here the relationship
between the dancer and the deity, indeed, between the deity and all human
beings, is encoded in the structure of the language; the language and the
context of that particular incident reveal a lot about the nature of the
180
teyiiptam; a meaning which cart only be perceived in the relation of the

language used to the situation. The praise of the teyyam in the three pe:r,:sons
is a speech-event which can only be understood with reference to the
context of the teyyam; conversely the speech-event illumines the nature. of
the teyyam as well.
Although a pragmatic analysis of the language used in the teyyam is a
desideratum~ we can nevertheless go some way towards understanding the
generation of cultural meanings through examining the non-linguistic
aspects of the teyyam. If we extend the notion of performance from the
linguistic arena, it-becomes open to interpretation in terms of non-linguistic
signs and through understanding the teyyam as performance, we can
understand to some extent the way in which culture is generated~ The
following remarks have been strongly influenced by Valentine Daniel's
book Fluid Signs, 19 in which he develops a analysis of the culture of a Tamil
village, based on the understanding of the sign in the work of Charles
Peirce, the American philosopher and founder of Semiotics.
This is not the place to develop a description of Peirce's complex
classification of the sign - he has nine types of sign and ten classes - but put
briefly, his interpretation of the sign is based on the relationship between
what he calls the signans or, more familiarly, the signifier, and the signatum,
the signified. The:: sign, whi~h he calls the representamen, stands for, or
represents, something: Furthermore it represents something to someone,
whom he calls the interpretant, in some respect, which he calls the ground.
The terms representamen, .object, interpretant and ground are the way in
which a sign signifies. With regard to the idea of performance, Peirce
distinguishes between an icon, something which functions as a sign whose
features resemble its object; the index in which there is a causal connection
between it and its object; and [mally the symbol in which there is an
arbitrary, conventional, or habitual association with its object. The symbol
therefore requires the interpretant to make the connection between the
representamen and its object. In the sign, be it an index, icon or symbol,
there is an irreducibly triadic relation between the representamen~ the
object and the interpretant. 20
The teyyam we can see as both an icon and a symbol in Peirce's terms.
That is, the actor of the teyyam represents the deity to the audience who
perceive him in 'the context, the ground, of the festival within the vfrankavu.
The deity in the shrine is aniconic, present there only in the form of
weapons, the swords, shields, bow, or a trident in the case of Gulikan, and
so there is an arbitrary connection between the representation of the deity
in the shrine and its representation in the teyyam dancer. At this level the
teyyam is a symbol of the deity; the relationship between the deity and the
dancer is purely arbitrary and conventional, determined by the ground or
context of the tryiiffam festival. At another level the teyyam is the dancer and
the teyyam is a representation of itself; the representamen is its object for the
181
interpretant or observer. In this case the teyyam is an icon, sharing the

identical qualities and form of what it represents, namely itself. The
possessed dancer becomes the icon of the teyyam and the dancer's
personality or individuality is lost beneath this.
In this process we can see the development of cultural meanings related
to value and social structure which we discussed earlier. The teyyam dancer
is the representamen of the object which is the deity, for the interpretant
who varies in different contexts. That is, there are a number of
interpretants within the grand-context of the teyyam festival: the young men
who taunt the deity, the women who watch at the side, the elders of the
community, the low caste man who waits in the street for the deity's
darfanam, the Nambudri priests who do not give entry to the teyyam into the
Siva temple, the local politician fishing for votes, and the western
'observers'.
In each of these cases the particular context in which the teyyam
functions varies. For the western observer, the teyyam is rarely seen as the
deity in the way in which many of the interpretants present would perceive
the teyyam. For the elders, before whom the teyyam displays himself, giving
them his darfanam, the teyyam is their deity who blesses them for another
year by his walking presence among them. The community elders, as
interpretants, in turn become representamens themselves for others;
symbols of the power relations within the Caliya community. For the
interpretant of the N ambudri priest in the temple, the teyyam is a low caste
deity who would pollute the shrine of the high caste god. The N ambudri
priest in turn becomes the representamen of the high caste, Brahmanical
value system which is the dominant ideology. The Caliya woman as the
interpretant of the teyyam depends upon the wider context of her age,
education, status and so on. She may, perhaps, perceive her potential
power or representation in the t0lyam Goddess, or may not.
The point is that social and cultural meanings are generated through the
complex relationship between representamen (the teyyaml dancer), the
object (the teyyaml deity), the interpretant (at one level all the individuals
present, at another level the different social groups and gender
differentiations) and the ground (the context in which the teyyam occurs).
This ground will, of course, vary with different individuals and groups, but
most would share the common ground and assumptions of the Hindu
social world and practices of deity propitiation, making vegetarian and
non-vegetarian offerings to the gods, and receiving their blessings in the
form of turmeric powder on their foreheads or some good fortune, such as
a child becoming well again.
The teyattam is a deep and complex phenomenon, as well as a colourful
and, partly, enjoyable one (except, perhaps, for the chickens). Both purity
and violence meet in the teyyam who becomes the focus of awe, hope and,
indeed, play. But there is a point at which analysis has to stop, beyond
182
which meanings become imprecise, and we have to simply accept it as a

representation of itself As the dancer Isodore Duncan once said in a
different context, 'ifI knew what it meant I wouldn't have to dance' it'.
Notes
1.
2.
3.
4
5
6.
7.
8.
9
ro.
II.
12.
13.
14.
15.
16.
17
18.
19.
20.
Freeman, 1991.
Ibid., p. 95.
For a description and analysis of a variety of t~JI)Iam performances, see Freeman, 1991,
especially chs. 3 and 4.
Ibid., p. 155.
Scheckner, 1995, P25
See Thompson, 1984, p. 4.
Bourdieu, 1991, pp. II7-26.
Rappaport, 1992, pp. 5-30.
Eag-leton, 1991, p. 45
Bourdieu, 1977, p. 78.
Eag-leton, op. cit., 'p. 45.
Girard 1977, pp. II9-42.
Ibid., p. II9.
Freeman, I99I;PP. 319-20.
Obeyesekere, 1981, pp. 169-82.
Freeman, 1993, p. II6.
See Freeman, 1991, p. 319.
Levinson, 1983, p .1.
Daniels, 1984.
Peirce, 1977, p. 81. For Peirce's theory ofsig-ns see Hookway, I98~, pp. n8-44.
Bibliography
Bourdieu, P. Language and Symbolic POW(ff. Cambridge: Polity Press,
I99I.
---~ Outline qf a Theory qfPractice. Cambridge: Cambridge
University Press,' 1977
Daniels, E. V. Fluid Signs: Being a Pmon the Tamil Wcry. Berkeley:
University of California Press, 1984.
Eagleton, T. Ideology, An Introduction. London: Verso Press, I99I.
Freeman, R. Purity and Violence: Sacred power in the Teyyam Worship qf
Malabar, (unpublished Ph.D., University of Pennsylvania, 1991).
- - - . 'Performing Possession: Ritual and Consciousness in the
Teyyam Complex of Northern Kerala' in Heidrun Bruckner,
Lothar Lutze and Aditya Malik (eds.) Flags qfFame; Studies in
South Asian Folk Culture. New Delhi: Manohar, 1993, pp. I09-I38.
Girarq, R. Violence and the Sacred. Baltimore: The Johns Hopkins
Hookway, C. Peirce. London: Routledge, 1985.
Levinson, S. C. Pragmatics. Cambridge: Cambridge University Press,
1983.
183
Obeyesekere, G. lvledusa's Hair; An Essay on Personal Symbols and

Religious Experience. Chicago and London: Chicago University
Press, 198!.
Peirce, C. S. Semiotics and Signifies edited by C.Hardvvick.
Bloomington: Indiana University Press, I977Rappaport, R. 'Ritual, Time and Eternity', :(ygon, Vo1.27 no. 1,1992,
PP5-3
Scheckner, R. The Future ofRitual. London: Routledge, 1995.
Thompson,].B. Studies in the Tlzeory ofIdeology. Cambridge: Polity
Press, I984.
Aspects of Esoteric Southern Buddhism
L. S. Cousins
This paper is concerned with a relatively little known form of Buddhism,

now found mainly in South-East Asia. There is a surprisingly widespread
notion that Theravada Buddhism is, at least doctrinally, a rather uniform, if
not monolithic, type of Buddhism. This is certainly a mistaken impression;
so in the first part of this paper I shall begin by attempting to outline briefly
some aspects of the historical diversity of this type of Buddhism and then go
on to survey the various trends of twentieth century and contemporary
Theravada Buddhism as different scholars have described them. Having
established something of the nature of this in fact rather varied history, I
shall then discuss the historical origins of the specific tradition which I shall
call Esoteric Buddhism and which some other scholars refer to as Tantric
Theravada. In the second part of the paper I survey certain of the particular
ideas of this type of Buddhism, giving passages from texts which adopt this
approach.
Part One
Historical roots
The name Theravada strictly refers to a branch of the Buddhist sangha
which adheres to a version of the monastic rules deriving from the second
century after the Buddha's parinibbiina and claiming to be the original
teaching (theraviida), later understood as 'doctrine of the Elders'. Since the
alternative (more conservative) version of the Mahasanghikas ceased to be
used in the late media:val period (probably in Nepal), the monastic
traditions deriving from Tibet and China are properly speaking also
Theravadin, or at least belong to traditions which were once Theravadin.
This usage is quite comparable to such terms as 'Catholic' or 'Orthodox' in
the history of Christianity. (Both parties would of course claim to be both
r86
Indian Insights: Buddhism Brahmanism and Bhakti

J
orthodox and catholic.) So for the m~st part I shall use the more satisfactory
(and neutral) geographical designation of Southern Buddhism.
This reminder of the historical origins of Southern Buddhism serves to
recall the considerable extension of this tradition in both space and time. It
is the religious tradition of the majorIty in the present-day countries of
Burma (Myanmar), Cambodia, Ceylon (Sri Lanka), Laos and Thailand
(Siam) with smaller populations in geographically contiguous areas of
Bangladesh, China, India, Malaysia and Vietnam. In total numbers the
Buddhist population amounts to over I'20 million people in this homeland
area, not counting a few millions in the recent missionary outreach, mainly
in India and Indonesia and among Chinese diaspora populations but also
in smaller numbers in Europe, the Americas and Australasia.
In much of this homeland area Buddhism has quite ancient roots. This is
most obviously the case with Ceylon which arch::eological data suggests was
colonized from northern India around the sixth century BG, probably as
part of a more general process in which urban settlements were extended
southwards from the Gangetic area by colonisation. I The recent discovery
at Anuradhapura of Brahmi inscriptions which appear to date from a
century or two before the reign of the Emperor Asoka in the third century
BG indicates close contact with cultural developments in the north. 2 vVe may
then suspect that Buddhism had already reached the island and made some
headway before its official introduction in the third century. Normally this
kind of royal acceptance is likely to follow after a degree of prior penetration
among the general population has already taken place. At a slightly later
date the widespread nature of the Buddhist presence is well-evidenced by
the arch::eological discovery of some hundreds of cave sites for monastic
practice, scattered over a large part of the island.
Buddhism also has ancient roots in other areas of its present
geographical outreach. Tradition in fact attributes to the Emperor Asoka
missions to SuvaI,lI,labhumi 'the Gold Country' - this must be either a
particular location in Southeast Asia or a general term for the whole area.
Scholarship has generally rejected the historicity of these accounts, perhaps
rightly; however, given the level of sea-born trade and other contacts they
cannot be wholly ruled out. Future arch::eological research may yet provide
evidence of them. Indeed it is clear from the arch::eological discoveries at
Beikthano and elsewhere that the up-country Pyu people had obtained
technical innovations directly or indirectly from Mauryan India. At present
there is little firm evidence of cultural importations in the period from
around the third century Be to the second century AD. Caution must be
exercised here, however, since it is possible that Mauryan-style Buddhism
would in any case have left little in the way of material remains.
In any case it is probable that forms of Buddhism using some kind of
Middle Indian (pali or closely related) were present among the Pyu and
Mon peoples of present-day Burma and Thailand by the second century
c
Aspects if Esoteric Southern Buddhism
187
and even perhaps somewhat earlier.3 Most probably Buddhist traditions

were introduced there from Ceylon or South India together with a form of
Brahmi writing in the early centuries AD. It is likely that this would have
been a form or forms of the Vibhajjavadin tradition current in Ceylon and
the Tamil country. This would be a pre-Buddhaghosa form and not
necessarily exclusively of the orthodox Mahavihara school.
Indeed it is clear from both literature and arch<eology that throughout
the first millennium AD. other schools and traditions of Buddhism were
present in Ceylon itself. In particular the important and at times
numerically stronger' Abhayagiri school, while preserving a slightly
divergent recension of the Pali scriptures, seems to have also studied North
Indian Sanskrit literature, especially that of the Mahayana and later also
that of the Mantrayana. Arch<eological discoveries of Sanskrit texts, statues
of bodhisattuas and various cult objects make this reasonably certain. Indeed
some J\;1ahayanist practices continued until relatively recent times with a
few remnants still current today.5
A similar picture emerges for Lower Burma in the mid-first millennium.
Hence the presence of Sanskrit materials and Mahayana cult objects does
not necessarily indicate the presence of Sarvastivadin or other schools. It
could equally be accounted for by the presence of a branch of the
Abhayagiri tradition or some similar Pali-based school open to
developments on the Indian subcontinent In either case we could expect
traditions and practices from diverse sources to be handed down to
posterity. In fact even the orthodox Mahavihara school incorporates a
good deal of material of North Indian origin in its little-studied later
commentarialliterature. 6 However, at a later period, as the evidence from
Pagan clearly shows, distinct influences from the predominantly tantric and
Sarvastivadin Pala Buddhism of Eastern India were certainly present in
Burma. 7
Looking more widely, there is no doubt ofthe presence of Sanskrit-based
schools of Buddhism in Indo-China, and in some parts of present-day
Thailand, Malaysia and Indonesia. We can then be sure that South-East
Asia received a rich and varied heritage of Buddhist traditions from its own
ancient beginnings, from later developments in India and equally from a
plural situation in Ceylon.
AD
Varieties of Southern Buddhism

There has in fact been considerable discussion among scholars of differing
trends and tendencies in the Southern Buddhist countries. We cannot
summarize the whole literature here, but some key points must be
mentioned. First of all, the distinction between traditional and innovative,
modernizing forms of Buddhism was already known to scholarship ip the
late nineteenth century. In 1962 Heinz Bechert adopted the term 'Buddhist
188 Indian Insights: Buddhism) Brahmanism and Bhakti

modernism' to describe the latter, offering (in 1964) a (non-exhaustive) list
of twelve main characteristics. In I970 Gananath Obeyesekere coined the
memorable expression 'Protestant Buddhism'. This has been preferred by
some scholars, but I must confess to finding it slightly tacky.
It is of course clear that there is no such thing as a monolithic movement
of Buddhist modernism. Some of the tendencies classified under this
heading have clearly been endemic in Buddhist history. For this reason,
writing in I984, I preferred to speak of three important trends, variously
combined with traditionalism: modernism, reformism and ultimatism. By
modernism I meant to indicate both institutional change to accommodate
new social and economic circumstances and the adoption of concepts and
practices influential in the modern world. Reformism on the other hand is
the attempt rather to restore an earlier, ideal state of Buddhism in general
and the sangha in particular. This is hardly something new - indeed it is
probably as old as the formation of the Theravada itself. A new element is,
however, introduced through awareness of modern historical criticism,
which differentiates more sharply the various strata of the tradition. The
third trend, ultimatism, is the tendency to focus on the higher aspects of
Buddhist practice and neglect traditional preliminary practices and
outward forms. It too is a perennial in the history of Buddhism.
More recently still George Bond has distinguished three new developments alongside and partly stemming from Protestant Buddhism. The first
of these and certainly the most important is N eotraditionalism, understood
as a return to traditional ways with a minimal accommodation to the needs
of the modern situation. The other two are the Insight meditation movement and moves towards an emphasis on social action.
The pace of change has, however, accelerated greatly in the last few
decades with the result that scholars are now beginning to speak of PostProtestant Buddhism (and perhaps even Post-modernist Buddhism?). The
former term has been adopted by Obeyesekere and Gombrich to describe
what they see as radical changes in recent decades in Sri Lanka. They
characterise these as an infiltration of devotional religiosity and magical
practices. 8
In a recent paper Bechert has surveyed ten major new trends In
contemporary Buddhism. I list them briefly in slightly amended form:
I.
2.
3.
4.
5.
6.
7.
The Buddhist contribution to the world-wide so-called 'green'

movement;
A reaction against Buddhist nationalism;
Reassertion of the teachings and values of traditional Buddhism;
Radical return to the roots of Buddhism (e.g. Santi Asoke);
Renewal of the ideals of the 'forest-dwelling' monks;
Revival of samatha meditation;
Revival of ritualism;
Aspects ofEsoteric Southern Buddhism

8.
9.
10.
r89
Syncretism with various other Buddhist traditions;

A tendency towards 'remythologization';
Reassertion of women's rights. 9
The fourth of these is a form of what I call ultimatism. The majority of

the others could be classified as either Neotraditionalism (3;7;9) or as a
further development of tendencies long present in Buddhist modernism
(r;8;ro). Some contain elements of both these (5;6).
One other important cause of change is the movement of ideas and
practices around the different Southern Buddhist countries. This is clearly
very old and can sometimes be circular. Note the way in which the monastic
ordination was restored to Ceylon from Burma under Vijayabahu at the
end of the eleventh century, only to be exported to both Burma and
Thailand on various subsequent occasions, restored again from Thailand in
the eighteenth century and imported several times from Burma in the
nineteenth century There could easily have been other occasions which we
do not know about. Similar movements have clearly occurred both with
texts and with ritual and meditation practices. We must then bear in mind
that the study of individual Buddhist countries in isolation could be
misleading.
Nor is this the only problem that could arise from the process of study
itself. Some of the perception of change is clearly a result of the greatly
increased numbers of scholars studying various aspects of Southern
Buddhism. Many of these tendencies have probably been present,
especially in South-East Asia, for a very long time, simply coming into
fashion and going out of fashion at intervals. Ceylon on the other hand
may be something of a special case, as Buddhism on the island appears to
have passed through a kind of genetic bottleneck under Hindu and
Christian rule in the late medi::eval and early modern period, particularly
as regards the Buddhism of the elite.
The origins of Esotericism

The above discussion has directed our attention away from the background
practices of traditional Buddhism. In fact even this shows a great deal of
variety, incorporating and preserving elements from many different
sources. Given the long historical background mentioned above this should
not be surprising. It is clear too that an elaborate and varied interface with
popular deity cults has been the norm for almost all forms of Buddhism,
since at least the time of Asoka and quite probably earlier.
Moreover, in traditional Southern Buddhism a rich ceremonial and
elaborate ritualism has also been the norm for a very long time. Much of it
goes back to the last centuries of the Anuradhapura kingdom in Ceylon
and no doubt to a similar date elsewhere. Some at least is pre-Buddhaghosa,
190
with apotropaic elements deriving from the Canon itself 10 This ritualism is
sometimes attributed to Mahayana influences, but may equally be derived
from those elements of pre-Mahayana Buddhism which underlIe the
development of Mahayana. Similar problems arise if it is attributed to
Hindu influences - it may equally have its source in the cult of the deities
which was certainly a normal part of the life of lay supporters of the
Buddhist saJigha since the very beginnings of Buddhism. Of course, this is
not to dispute the existence of many elements imported from Mahayana
and Hinduism at a later date. Rather, it is to suggest that they could be
introduced precisely because they fitted well with practices already current.
What I am concerned with here is a specific form of Southern Buddhism
which I call Esoteric Buddhism. Let me first make clear that this is not the
same as the more general practices of a magical kind which are as endemic
in the Southern Buddhist countries as they were in Europe until quite recent
times - in their local form involving the use ofyantra or mantra and/ or ritual
for purposes of protection, healing, harming, empowering and general
assistance. Nor is it the same as the equally endemic trance-based ecstatic
and mediumistic practices used for similar purposes.
These are phenomena which are widespread in agricultural societies the
world over. Gombrich and Obeyesekere draw attention to the way in which
they have recently invaded the urban milieu in Sri Lanka. Similar
phenomena can be seen in Bangkok. It should be noted, however, that this
can itself be seen as an aspect of Buddhist modernism, a tendency whicn
often in practice amounts to Europeanisation. I have myself seen European
spiritualist literature displayed at centres for spirit-mediurnship in Bangkok.
More generally the last thirty years has seen produced in both Europe and
North America a large body of occult literature of various kinds. Inevitably
such.works have an influence in Asia; it is precisely the urban middle class
which is most exposed to them. Equally, young people studYing abroad
have met such ideas and practices.
Returning to. esoteric Buddhism, what I am referring to is a type of
Southern Buddhism which links magical and, ritual practices to a
theoretical systematisation of the Buddhist path itself. Of course the
distinction I have made is in part an artificial one. Popular magic and ritual
is on the one hand the raw material of esoteric Buddhism, while on the
other we see rituals and mantras derived from esoteric Buddhism widely
used at q popular level. A growing tendency since the mid-nineteenth
century has b.een pressure from reformist groups (often supported by
modernizing governments) to remove elements identified with esoteric
Buddhism from monastic practice, but there is some evidence to suggest
that it was a widely accepted part of normative Southern Buddhism before
the nineteenth century.
Research in this area has been spearheaded by the work of F. Bizot,
based originally on contacts with the still living form of this tradition in
191
Cambodia. He has collected a number of manuscript works of this kind

and translated some of this material into French. It is not yet entirely clear
where this literature fits in the overall history of Southern Buddhism, but
Bizot has, asJohn Strong puts it:
, .. managed to show that, where many had once seen only "corruptions" or
"popular aberrations", there may in fact be found a genuine tradition,
complete with its own history, ideology, ritual and soteriological endeavours, 11
At present there is no agreed terminology to designate this material.
Bizot tended originally to speak of the unreformed Mahanikaya tradition
and more recently of non-Mahaviharavasin currents. Bechert seems to
prefer to speak of'tantric Theravada', a term also sometimes used by Bizot.
Bizot sees the origins of this tradition in the ancient Buddhism of the
Mon. This may well be correct, but still leaves the problem of the ultimate
source, There are at least five possibilities, It is perhaps useful to look at these:
I.
TIle JvJantrayiina Buddhism of the later Indian Mahayiina.
This could be either a direct import from Bengal or via the Mantrayana
traditions which were at some points influential in, for example, Indochina
and Java. The objection to this is well-indicated by Bechert:
There is not the slightest hint at the influence of any Mahayanistic thought
nor are there any traces of the terminology of those forms of tantric
Buddhism which are known to have existed in India, In terms of its doctrinal
background, this "tantric Theravada" is based on the scriptures of
Theravada in Pali exclusively. The followers of this tantric Theravada,
however, discover a deeper meaning behind the obvious one in the doctrines
and texts of the Theravadins,12
2.
InfluencesJrom Saivite traditionsformerly current in Cambodia and elsewhere,
The objection to this is similar. There are a few concepts which perhaps
derive from brahmanical tantric traditions. Notably the terms for some of
the inner channels within the body seem to be Pali versions of those used in
various Hindu traditions. Overall, however, the resemblances do not seem
very great.
3, A home-g~own product which developed in Southeast Asia,

Bechert has recently put forward this view:
It seems that the esoteric teaching of tantric Theravada originated when new
methods of concentration and psychic cultivation which became known
under the name of "tantra" in India spread over the whole world ofIndian
religion ... , At that time Buddhism in mainland India was predominantly
IvIahayana, and thus tantric Buddhism arose on the basis ofMahayanism
there. The Theravada communities of Southeast Asia, however, have largely
192

remained HinaYanistic. Thus, the 'new methods were adapted to Hinayanistic
Theravada and the terminology of Mahayanistic Tantrism including the use
of the word tantra was avoided. 13
'
This hypothesis would certainly account for the emergence of esotericism

in Southern Buddhism and even possibly imply a fairly ancient date for
this. There are other possibilities:
4. A product qf the Abhayagiri school in Ceylon.

A similar process to the above might have taken place among the nonMahavihara schools in Ceylon or among the followers of that school in
mainland India. Unfortunately it has not so far proven possible to identifY
material which is clearly of this origin. 14 Moreover, the Abhayagiri school
in Ceylon seems to have adopted Mahayana teachings and their Sanskrit
literature without inhibition.
5.A product qfthe Mahiivihiira tradition.

We cannot as yet rule out the possibility of a development within the
bounds of the orthodox tradition itself Certainly of all schools this was the
. one most attached to the Pali language and the use of Pali terms to derive
mantras is a strong feature of these Esoteric Southern Buddhist texts. Such
a development, could have occurred at various points in the history of the
Mahavihara tradition. Indeed it might do much to explain that tradition's
successful resistance to the trends which led to the demise of most other
forms ofIndian Buddhism.
Buddhaghosa mentions secret texts (gUfJzaganthaT(l) in three ofhis works, in
the context of teachings which will not be received from a teacher if one
does not pay proper respect. 15 The author of the Abhidhamma commentary
and Upasena (c. 500 AD) refer to such secret texts, while discussing stinginess
as to teaching (dhamma-macchariya).16 There are also references to various
texts prefixed by the appellation 'secret', but these were certainly considered
heterodox. 17 The won;! gilfJza can hardly admit of any interpretation other
than secret. 18 (Gantha of course does not necessarily refer to written texts.) So
it is likely that trends towards esotericism were already developing before the
fifth century AD. There is obviously no way of knowing for certain the
content of Buddhaghosa's secret texts, but it is highly significant that the
idea of such works was present and the very existence of the term could
easily have been sufficient to open the door to an esoteric interpretation.
We may speculate that the content of these is likely to be similar to the
meditation manual which was found in Central Asia and subsequently
translated into German by Dieter Schlingloff. 19 Although this is presumably
of (Mula-)Sarvastivadin origin, it is very much based upon the application of
visualization techniques to the traditional topics for meditation. It is not
Mahayanist, although it could be seen as tending towards the Mahayana.
I93
Alexander Soper drew attention to various works preserved in Chinese with

a similar orientation towards visualization. 20 For our present purposes it is
significant that the Apadana, a late canonical Pali work, already contains a
marked tendency towards visualization.
DhammapaJa., closely followed by Sariputta, the h\1elfth century
commentator; explains Buddhaghosa's secret texts as: "Secret texts expound
emptiness [and] are connected with the truths, the taking rebirth !pa#Sandhz)
of beings and [the law of] conditionality". 21 The mention of the taking
rebirth of beings is interesting; in the mahiiflka to the Visuddhimagga, also.
attributed to Dhammapala, we find reference to the other items but not to
this. Instead Dhammapala refers there to the item of kamma![hanagantha'T{l.22
Since the flka writers normally stay rather close to one another, this is likely
to be a deliberate change. It may imply that he considered the notions of
kammatthana and taking rebirth of beings as equivalent in the context of
'secret texts'. It is tempting to suggest that he knew of an esoteric tradition of
the kamma![hiina (spiritual path,23 in this context) concerned especially with
the process of rebirth. The material discovered by Bizot could certainly be
. so described. 24
If this tradition was known to Dhammapala it perhaps tells us a little
about its origins. Clearly it was accepted by him as orthodox, since he is
prepared to connect it with the by then long-established notion of the 'secret
texts' to be taught to the meditation pupil. Unfortunately the date and even
the number of Dhammapalas is somewhat debatable. 25 Lily de Silva has,
however, established that by the twelfth century a single Dhammapala was
believed to be the author of most of the works traditionally attributed to
him. 26 H. Saddhatissa has advanced arguments which go far to identify the
Dathanaga mentioned in the introduction to the mahaflka to the
Visuddhimagga with a tenth century figure. 27 It is likely then that we are here
dealing with a writer or writers of the late tenth century, probably based in
South India rather than Ceylon, but in close contact with the Mahavihara .
tradition.
If such an interpretation did develop in Ceylon or South India, this
would most probably have occurred as part of the general growth of tantric
tendencies in Indian religion, as Bechert has suggested. Perhaps the most
likely period for such a development in Ceylon would be between the sixth
and the ninth centuries. At present this is definitely a possibility, but it is
certainly not yet conclusively established.
Which of these five alternatives is correct? I do not think this question
can be answered at the present time. Indeed they are far from mutually
exclusive. It is quite possible that present-day esoteric Buddhism contains
ideas and pra.ctices deriving from more than one of these sources.
Nevertheless it is certainly premature to assume that it has its origins in
unorthodox circles.
194 Indian Insights: Buddhism, Brahmanism and Bhakti
Part Two
Nature of Esoteric Buddhism
In the second part of this paper I shall present some of the principal ideas
of esoteric Buddhism as accessible at the present time. Let me begin with a
quotation to set the tone.
The tree grows as high as is nibbiina.
Its roots of crystal
number twenty six.
Set with diamonds is the precious trunk
which unfolds beyond
the four continents.
The first branches,
like six marvellous stars,
subdivide into nine smaller branches
whose thick foliage
very rarely
viithers.
The upper branches,
which number si.,teen,
with leaves that are
close-set and evergreen,
carry fruit that
last long before they fall.
The succeeding branches
-- there are nine most beautifulare very thick [with foliage]
ofleaves, some old, some new,
which, when their time has come,
fall one by one.
From the midst of the branching
which separates many times
in countless forkings
come forth four healing branches,
a permanent refuge
which protects all beings.
These are the four requisites Ipaccaya)
of man today,
the constant source of protection
which spreads good fortune,
happiness and success.
They bear leaves, flowers and ff).lit
which ripen and fall
195
in the network of the world,

and transform [on reaching] the ground
. to bring about the birth offigtrees
which possess five branches.
An .indriya [bird] keeps guard
-upon each of these five brandies
so as to devour human beings.
Such is the search for the crystal spheres, the fruits and flowers of the figtree. 28
Here we meet already the search for the crystal sphere to be found in
the figtree with five branches, i.e. our body with its five senses. This is
somehow connected with a cosmic tree which extends beyond the four
continents and reaches to nibbiina.
Perhaps this is enough to show already that this kind of Buddhist
esotericism is about mapping worlds, both visible and invisible ones. 29
Indeed the wider context of which it reminds one is a type of mysticism
which I shall call tantro-kabbalistic for the purposes of this article. By this I
mean a form of mysticism which utilizes a rather elaborate map of
correspondences- between the human body, the cosmos and some kind of
higher reality or knowledge. In the process it draws on the full resources of
the widely-dispersed traditions of magic and the occult - letter, sound and
number symbolism together with the use of structured patterns of shape or
gesture. Often these are applied in ritual. Typically too this is linked with the
,spiritual practice of one of the higher religions in a manner which integrates
the system of correspondences with a model of the spiritual path and with
various modes of spiritual practice. I do not mean to imply that this form of
mysticism is only to be found in the Jewish Kabbalah or in Hindu and
Buddhist Tantra. Similar traditions are certainly to be found within Taoism
and Islam, for example. I simply wish to take Tantraand the Classical
Kabbalah as loosely paradi",omatic. Needless to say, each such tradition has
its own unique features.
Let me now try to illustrate some of the above features by means of
quotations, mainly but not entirely, translated from the French of Fran<;:ois
Bizot.
I.
Correspondences with the bodJ

J.
2.
3.
4.
5.
6.
The fig-tree is first of all the trunk.

Its branches are the two arms and the two legs.
Its leaves are the two ears.
The flower of the fig-tree is the umbilical cord.
Its fruit is the child established in the womb.
Its roots are the three [parts]30 of the male sex organ which contain the
generations to come - that is to say, the precious water which flows and gives
birth in the realm with seven crystal walls. Hide so that no one may see. 31
I96
2.

Correspondence with the cosmos
Herewith some passages taken from another text (with omissions):

The land of the Rose-apple tree
It is located to the south of holy Mount Sumeru. Around lVIount Sumeru is
situated first lVIount Giri, then Mount Vulture Peak and. lastly the seven
ramparts of Mount Sevenfold Ring.
Mount Sumeru
In the five aggregates of our bodily form, our head is Mount Sumeru; our
chest Mount Giri; our pelvis Mount Vulture Peak; the two knees, the two
ankles and the two soles, the seven stages of lVIount Sevenfold Ring, the
ramparts of Mount Meru. The four lakes situated at the foot of Mount
Sumeru are the four elements in our bodily form.
The figtree
The five aggregates of our form are called the figtree. That which grows up
in the land of the Rose-apple tree is the two Children of the Heart32 who
are reborn in the maternal womb. 33
Here we see the traditional geography centred on Mount Meru now
linked to the system of correspondences. The form of the figtree, here and
elsewhere associated with the five aggregates and the four (or five) elements, is used for this purpose.
Another passage:
"0 Children of the Heart! In that place there is a fig tree with each of its
fruits concealing a precious crystal sphere. It casts a delicious scent over
countless world spheres. This is the holy34 Dhamma. The crystal sphere
placed in the flowers of the fig tree - magnificent and glorious - this is the
holy Buddha. The fig tree has four long branches, which extend in the four
directions beyond the limits of the world sphere. Its roots extend downwards as far as the lightless Unending (avlez) hell. Its topmost spike reaches
to the Great Brahma realm of the Eldest devas. 3s
In this second case the vertical cosmology of the heaven realms and
hell worlds is made use of in a similar way.
3. Correspondence with higher knowledge

A PA. NIA CU PA holy Vinqya.
DI
this is the quintessence of the five books of the
MAM SA1YI AlyI U36 -
this is the quintessence of the five books
of the holy Suttanta.
SAlyI VI DHA. PU KA YA PA seven books of the holy Abhidhamma.
this is the quintessence of the
197
A is the holy Adikamma; PAis the holy Pacittiya; MA is the holy

Mahiivagga; CU is the holy Culavagga; PA is the holy Panvara. These
are the names of the holy Vincrya.
DI is the holy Dfghanikiiya; MAlyI is the holy Majjhimanikiiya; SAly1 is
the holy SaTf1.yuttanikiiya; AM is the holy Anguttaranikiiya; U is the .holy
Khuddakanikiiya. These are the names of the five books of the holy
Suttanta.
SAM is the holy SangaTji; VI is the holy Vibhanga; DHA. is the holy
Dhatukathii; PU is the holy Puggalapafifiatti; KA is the holy Kathiivatthu;
YA is the holy Yamaka; PA is the holy Mahapatthiina. These are the
names of the seven books of the holy Abhidhamma.
MA ~ i.e. passiisa, breathing expels and draws in, firstjhiina;37
A-
i.e. assiisa, absence of breathing, ~econdjhana;
U-
i.e. nissasa, breathing expels, thirdjhiina. 38
The connexion made now is with the constituent parts of the Pali
Canon - the vehicle of the Buddha word itself This can be, and is, linked
to the idea of developing a dhammakiiya or body of D hamma. 39
In the Vimuttimaggadassana, a Pali text preserved in Ceylon,40 we have
the syllables of arahaTf1.linked with the three treasures (ratana) as follows:
A-
Buddha
RA -
Sangha
HAJ.Y1- Dhamma. 41
The same passage then goes on to relate assasa to Sutta-pifaka, passiisa to

Vincrya and nissasa to Abhidhamma. 42
4. Letter and number symbolism

9. Add together the virtues of the mother and father, which makes
thirty-three. This then is called the thirty-three letters which create
all human beings.
!D.
Add the thirty-three letters to the five aggregates, and one obtains
the thirty-eight virtues of the holy Dhamma. This is the holy
Dhamma in person.
II.
TakeNAMOBUDDHAYANA, the twelve virtues of the mother;

MO, the twenty-one virtues of the father;
BU, the six virtues of the king;
DDHA., the seven virtues of the family;
YA, the ten virtues of the teacher.
These are the five aggregates which give the fifty-six virtues of the
holy Buddha. This is the holy Buddha in person.
Ig8

I2.
Combine the three: passasa, assasa and nissasa, with SAM VI DDHA
PU KA YA PA and with the four elements NA MA A U. One
obtains the fourteen virtues of the holy sangha. This is the holy sangha
in person.
13. Adding all of these together gives ro8.

14. This is found in our body.
IS. This is that which all noble sons, all people, all men, must seek to find
and understand completely clearlyY

Obviously this passage could be discussed at great length. For present
purposes it suffices to note two aspects. Firstly the use of the number thirty-three seems to be deliberately intended to reinforce correspondences. It
is of course the number of the deities in the second heaven (cp. also the
Burmese thirty-seven Nats 44) and seems to have led to the expansion of
the traditional number for the parts of the body from thirty-two to thirtythree. 45 This is also connected to a version of the Pali alphabet with thirtythree letters. It could also be related to the thirty-two marks of a great
man (mahapurisalakkha7Ja).
Secondly the overall aim of the passage is to link the figtree to the
virtues (gur;a) of the Triple Gem. To illustrate, fifty-six is the number of the
syllables of the itipiso, often in fact referred to as the Buddhagur;a - an
extremely widely used formula in ritual, protective and meditative contexts. The overall figure of ro8 is the commonest number of beads in a
rosary as used in Southern Buddhism and indeed elsewhere. It may also
have astrological links as it is the number ofpadas in the yearly cycle. 46
5. Ritual
It is difficult to select a short passage to convey adequately the ritual
aspects of Southern Buddhist esotencism, since ritual texts tend not to be
very readable as coherent pieces. Suffice it to say that chants and practices
which derive from this tradition seem to be widespread at a popular level
in Thailand and Cambodia and probably elsewhere. Elaborate forms of
ritual are of course typical of traditional forms of Southern Buddhism. In
the past they seem to have contained a strong esoteric element, although
some of this has been removed as a result of recent reforms. Bizot has
made a substantial study of the texts associated with one form of ritual
performed for various purposes including the prolonging ofhleY
6. Application in daily lift

Whenever it is necessary to undertake 'a journey of some kind, take three
strides .. Crouch down on the heels and recite:
Aspects qfEsoteric Southern Buddhism
199
"MA is the holy Vinaya; A is the holy Abhidhamma; U is the holy Sllttanta.
MA dllkkhaTfl- sivaTfl nibbiinaTfl.

A aniccaTfl - sotdbbaTfl nibbiinaTfl.
U anattii - nirodhaTfl nibbiinaTfl. "48

Concentrate upon that. Fasten nissiisa in the heart. When nissiisa penetrates,
one obtains cessation. 49
This extract usefully demonstrates the thin line in this kind of religious
tradition between spirituality and superstition. Note the use of syllables
together with attributions to the tipipaka. The three syllables, although
obviously related to the Vedic ADM
OIyf, are commonly derived in
Thailand from a Pali stanzaSO :
Araha7J1. sammiisambuddho / Uttamadhammamqjjhagu / /
1'viahiisanghaTfl pabodheti / icc eva ratanattayaTfl / /

Yet we have here direct reference to the three signs (anicca, etc.) of
insight meditation as well as to the embryonic breathing (nissasa) of the
esoteric tradition. This brings us to:
7- Spiritual practice " meditation methods

The meditation methods of Buddhist esotericism have a strongly mantric
component, although breathing techniques and visualisation also have an
important role. There has been a tendency in scholarship to regard the
repetition of syllables and devotional phrases as something distinct from
meditation: a kind of prayer or mechanical repetition. In fact it plays an
enormous part in traditional Buddhist circles in all the Southern Buddhist
countries, even in Burma where it would originally have been regarded as a
type of samatha or calm meditation based upon concentration. sl
This kind of meditation is found quite widely in use among lay devotees.
Often it is no doubt rather minimalist, although in other cases it is
something more serious. It is quite long-standing. As an illustration we can
quote Robert Percival (in Ceylon from 1796-r800), who describes Buddhist
monks: "To their girdles they wear suspended strings of beads made of a
brownish or black wood; and mutter prayers as they go along". 52 (They are
described as wearing the robe with the shoulder bare and carrying a
painted cane in one hand with a palmleafumbrella in the other.) Later on
he describes the Ceylonese as wearing beads and muttering prayers "as
they count them and go along the road"." He understands these as
preventives against "the evil spirits which surround them".
One who enters the order and honours the three robes, must apply himself
to recollection of repugnance, to the ugly and the beautiful and perceive a
200
Indian Insights: Buddhism) Brahmanism and Bhakt:i,

noble altar within as without. One who enters the order and honours virtue,
must know the interior objects and observe the bodily postures (iriyiipatha) as
the Lord has taught in his holy word. Within and without are entirely alike.
If one receives dana from a_ do,yaka, let there be enough requisites to
correspond with the [different ele~ents in the] womb of the holy mother.
Then one will have followed the rules of the holy Vinl9'a.
All of you, brothers, pay heed. Let us reveal [what it is like] when one is in
the place of the embryo, that is in the belli of the mother, for ten months clinging to the umbilical cord of the holy mother, seated with the knees
drawn up, the eyes closed, the mouth practising the holy syllables: NA KA
Aty!. When one is expelled at birth, one is dazed, distraught, lost,
unconscious. One seeks in vain to remember the three holy syllables. This is
why the Lord has urged that one should prostrate oneself in order to practise
the exterior holy syllables: A RA HA!yL Those are the exterior things.
Prostrate oneself in order to practise as if-residing in the womb. The Lord
has one to fold the legs, set the right hand in the left hand and pronounce the
syllables of the holy A RA HA. If, for example, one prepares oneself to
receive an offering, one must conform to the posture in the womb. If one
receives the eight requisites, 54 the robes must conform to the objects of the
noble altar as in the place of the embryo in the belly of the mother. Then
one is in correspondance with the noble teaching. When one is established in
the place of the embryo in the belly of the mother, one possesses all the
objects. They are the colour of blood, red like a ball of gum resin !gamboge).
One possesses all the objects in full. Ifin the place of the embryo in the belly
of the mother, these objects are incomplete, our bodily form deteriorates and
cannot grow.
All of you, brothers, pay heed. If, having entered the order in order to pay
honour to .fila and to the precepts, you do not possess the seven pieces in full,
the sila and the virtues which you honour will deteriorate, and then
disappear.
Let us continue to reveal what exists within: the under-robe, the waist band,
the cloaking robe (dvara), the breast band, the inner robe (sanghiil:l), the sash,
the stole. These- are the seven elements. Know that they are the elements
within. The waist band is the cord at the umbilicus. The breast band is the
end which.is attached [to the placenta]. The sash is the part which remains
with the holy mother. The stole is the bag of waters which contains a little
blood. The cloaking robe is the amnion (innermost membrane). The inner
robe is the placental envelope. The under-robe is the pocket for excrement.
When the body is established in the womb of the excellent holy mother, the
waters are above our head af!d flow continually drop by drop upon our
bodily form. If this water does not irrigate our body, we cannot live. That is
Aspects qfEsoteric Southern Buddhism
20I
why one must know the within and the without alike and must observe
bodily postures in accordance with the laws of the within and the without. It
. must not have any colour there except red like a ball of blood. If another
colour is mixed with it, the form cannot be born. If sfla and the precepts are
not observed correctly, one decomposes like the body,55
Now we see a rather fuller exposition of some aspects of the assimilation
of the process of gestation to the experiences of meditation. At the same
time the environment of the embryo in the womb is compared to the seven
items of the traditional dress of the Indo-Chinese Buddhist monk (now
widely replaced by a simpler form of monastic clothing considered closer to
the canonical model).
8. Spiritual practice: the path

The wind descends from the nose down to the neck. This then is called the
holy Vinqya in five books. This is the path of the stream-enterer. The wind
descends from the neck down to the chest. 56 This is the path of the oncereturner. The wind descends from the chest down to the naveL This is the
path of the never-returner. The wind descends from the navel down to the
coccyx, without leaving or entering. This is the wind of the path of arahatship.
He attains cessation. Concentrate your attention correctly and you find
yourself as you were when you were inside your mother's womb. Do not let
the wind leave or enter. Following these instructions is to invite the eightfold
ariya-sangha [in order to enter the way to the attainment of cessation, by the
four paths].
The wind descends from the nose down to the neck. Practise correctly and
you will be able to contemplate the holy light that is like a firefly. This is
called obtaining the fruit of stream-entry.
The wind descends from the neck down to the chest. Practise correctly and
you will be able to contemplate the holy light that is like the morning star.
This is called obtaining the fruit of the once-returner.
The wind descends from the chest down to the navel. Practise correctly and
you will be able to contemplate the holy radiance that is like the full moon.
This is called obtaining the fruit of the never-returner.
The wind descends from the navel down to the coccyx. Practise correctly
and you will be able to contemplate the holy radiance that is like the rising
sun. This is called obtaining the fruit of arahatship.
One attains cessation. Leave cessation.
The breath leaves the coccyx and rises to the navel.
Practise correctly and you will be able to contemplate [the holy shimmering
radiance that is like the splendour of the sun and moon].
202

The breath lea~es the navel and ~ises to the chest. Practise correctly and you
will be able to contemplate the moon and the stars.
The breath leaves the chest and rises to the neck. Practise correctly and you
will be able to contemplate the holy radiance that is like the combined
splendour of the sun, the moon and the stars.
There appears a shining light which sparkles. One contemplates the ten
gates. One contemplates truly the realm of nibbiina.
The breath leaves the neck and rises up towards the fontanelle. Practise
correctly and you will see the holy light that opens the gate to nibbiina.
The mind, balanced and concentrated, should be kept like a hair belonging
to an inhabitant of the continent of Western Goyana.Then the mind will be
able to enter into nibbiina, thepeaceful and glorious realm of the highest
happiness. 57
This final passage should probably be considered as exemplifying

another general feature of Tantric religious forms. It is quite typical of
such traditions to utilize practices based upon the imitation of the higher
spiritual states which are sought. In the Buddhist context this has often
involved an attempt to Inimic the yuganaddha quality of the transcendent
(lokuttara) path i.e. the manner in which the experience of awakening
(bodht) unites and somehow harmonizes both the peaceful (samatha) and the
insightful (vipassana; aspects of Buddhist practice. The idea no doubt is that
imitation can ultimately induce the relevant mental states. Similarly in
visualisation practices the body of a Buddha is imitated by visualising the
marks of the Great Man.
In the present context the aim is to imitate the various stages of the
transcendent path (path of stream-entry and so on). By doing so, in a form
of meditative ritual enactment, conditions are created which can lead the
advanced meditator towards his goal.
Notes
1.
2.
Allchin, 1989,PP. 1-16; 1990, pp. 163-'73.

Coningham and Allchin, 1992, p. 166; Deraniyagala, 1990, pp. 149-68.
Lucet, 1974; Janice Stargardt, 1990, (reviewed by K: R. Norman inJRAS(1992) pp. 114-17);
Smith' and Watson, 1979; Wales, 1969, p. 1Of.; Allchin, 1989, pp. 1-16; Allchin, 1990,
.
PP 163-73
This at least is the implication of the account by the Chinese pilgrim Fa-hsien. However, his
contacts were with the Abhayagiri school and may therefore exaggerate its importance. In
particular the strongholds of the supporters of the Mahavihara seem to have been in the south
of the island and Fa-hsien may have been unaware of them or uninterested in up-country
Buddhism. Similar considerations may apply to the seventh century account of HSilan-tsang
who does not seem to be aware of the exi~tence of non-Mahayanist Theravadins, except in
Bengal and the Tamil country!
Aspects ojEsoteric Southern Buddhism

5.
6.
7.
8.
9.
10.
II.
12.
13.
14.
IS.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Holt, 1991. On the Mahayana in Ceylon, see also: Paranavitane, 1928, pp. 35/1; Mudiyanse
1967; Bechert, 1977, pp. 361-8; Dohanian, 1977; Ruelius, 1978, pp. 89-99; Norman, 1978,
pp. 28-47; Schopen, 1982, pp. roo-ro8; Smith, 1987, pp. 257-78, and citing: 'Donanian,
"Prolegomena tothe Study of the lVIahayana in Ancient Ceylon," in India's Contribution to
World Thought and Culture (Vivekananda Commemoration volume).
e.g. Cp-a 276-332.
Guillon, 1985; Strong, 1992.
Gombrich and Obeyesekere, 1988.
Bechert, 1995
Mostly in later portions of the tipifaka, but we should note that there is a tendency to date
material as late precisely because of the presence of .\Uch elements. There is an obvious
danger of circularity here.
Strong, op. cit., p. 185.
Preface to Bizot, 1988, p. 9.
Ibid., pp. II-2.
There is one passage related to the Vimuttimagga of Up atiss a, but it is uncertain whether this
was a work of the Abhayagiri school or not. Indeed, even if it was, it may have been written
at a date before there was significant doctrinal divergance from the lViahavihara.
Ps II 264 Mp V 97; Vism lI5. Dhammapala's Malzaflka explains the Visuddlzimagga reference
as: "kamma!!Jzana texts which are profound and concerned Witll such [teachings as] the truths
and conditioned arising or those which are connected with emptiness" (C" 1928: gilflza7(l
ganthan ti kammattfzanagantha7(l saccapa!iccasamuppadadiJ'ahita7(l gamblzira7(l, sUfifiatiipa!immyutta7(l va;
(B" 1977: ". cal. Vism-sn is slightly different. Compare also Pa,is-a 674 which links several of
these expressions: The 'secret meaning' (attha) or 'hidden g'oal' is the transcendent (lokuttara)
'because it is completely outside' (d Pa\:is-gp).
Dhs-a 374 (mt: gantlw ti pa!!); Nidd-a 1112.
The Secret Yesssantara, the Secret Vinqya, tlze Secret Ummagga, etc. These and others constitute the
'counterfeit saddlzamma'. The reference is probably to the early Mahayana in view ofSpk-p, II
(B" 1961) 171. See Sp I 232; IV 742; Sv II 566 (p,); Spk II 201; Mp III 160. According to
Sariputta's li:ka to the Vinaya (to Sp I 232), these are "suttas of the dwellers in the
Mallasanghika-nikaya" .
So Adikaram, 1953 [1946], p. 97ff. and Bechert (cited in note 12 above), p. II. A different
view: Collins, 1990, p. u61: n. 55. Collins correctly points out thatgilflwttha, 'hidden meaning',
used at a much later date in the title of certain workl (and at Pa,is ii 195), need not imply
esotericism. However, Buddhaghosa refers to texts (gantlla) as hidden or secret, not their
meaning (altha).
Schlingloff, 1964.
Soper, 1959, Chapter III lconograpJ,y; cp. Kloppenborg and Poelmeyer, 1987, pp. 83-95; Beyer,
1971, pp. 329-40; Harrison, 1990.
Ps-pt II (Be) 168 '" Mp-t (Be 1977) 369: Dhammakathiibandlwn Ii pave~i-agala7(l paki~~aka
dhammakatlziimagga7(l. SaccaJattappa{i.randhipaccqyiikarappap..l'a7(lyutta7(l .rufifiatiidlpana7(l guflzagantlzaT(1..
Some editions of Ps and Ps-pt read guyhagantlla7(l, but this probably does not affect the sense.
Vism-sn has kamma.l'thiina(-).l'atya-.
See my note on this topic in Hinnells, 1984, p. 179. The two senses of kamma!!hana are
distinguished clearly in Abhidh-s-mht (N' 1965: 236); cp. m, to Vibh-a 263.
Bizot, 1976.
See Cousins, 1972; Pieris, 1978; Jackson, 1990. A, I have pointed out in the introduction to
Narlamoli, Dispelir qf Delusion, Vol. I, p. x, n. 3, Dhammapala's commentary on the Udana
appears to cite the anuflkii to the Katthiivattlzu commentary. This poses difficulties for the theory
of two Dhammapalas, unless they are contemporaries in the tenm century (suggestion of P.
Jackson in a letter dated 31.1.92). If there is an earlier Dhammapala, author of various
allhakathiis to the J..7zuddaka-nikiiya, he cannot be before the sixth century at
earliest.
Sv-pt I Introduction pp. Ii-Iv.
Hammalawa Saddhatissa, 1965, p. 29
Bizot, 1976, p. II7f.
An earlier version of this paper was presented at a conference of' the Traditional Cosmology
Society (Lampeter, 1992) on the theme of "Mapping Invisible Worlds."
The two testicles and tlle penis.
me
26.
27.
28.
29.
30.
203
204
Bizot, 1980, p. 250.

Cittakumara and Cittakumati.
Bizot, 1976, p. II9[
Here and below the honorific is rendered by 'holy'. This no doubt overstates the case, but is
intended al a corrective to the common practice of not rendering honorifics in South-East
Asian languages which underplays the reverential element in some forms of traditional
Buddhism.
35. Bizot, 1976, p. 83. This was the first text translated Ii-om the Khmer by Bizot. A version orthe
same text in the northern Thai dialect of Lanna was discovered in 1983 with the title
Pavarabandha. Other versions have subsequently been found. See the comparison in
Lagirarde, 199+
36. As Bizot points out, this should be KHU. He suggests that this occurs frequently because of
the similarity of the writing of the two syllables. However, it may also have been due to a wish
at some point to make the five syllables Jor Vin~ya begin with A and the five for Suttanta end in
U i.e. two of the three parts ofMA AU.
37. Bizot points out that the normal rendering would be: passiLra, (i.e. pasviit) breathes in; a.rsiisa
breathes out; nirsiisa, absence of breathing.
38. Bizot, 1976, p. lOI.
39. For bibliographic references, see: Reynolds, 197740. On this text, see now: Bechert, 1989.
41. The beginning ofT. W. Rhys Davids, Yogiivacara's iVlanual, London, 1896, p. 2 identifies the
syflables of arahaT}'l with the three raiana al Joflows:
A ~ Dhamma
RA - Buddha
HAi'yI - Sarpgha.
This section is, howevel; missing in the Sinhalese edition given at the end of the Vimuttimaggo
referred to above. Both the Yogiivacara's lvlanual and the Vlmuttimaggadassana are texts of the
same (stif! extant) South-East Asian meditation tradition which is now treated in some detail
in Bizot, 1992.
42. 66b A-kilro Buddha-ratanaT}'l ra-kiiro Dhammam uttamaT}'l I
66c
IzuT}'l Sanglzattlzan tiyojetviijii~itabbaT}'l visuT}'l visUT}'l11
67a A.mil'o Sutlapi!akaT}'l p{Ltriiso VinayaT}'l tat/Iii I
67b
ni.s.riiso Ablzidlzammo IijiinitabbaT}'l visuT}'l viJuT}'l.11
(Ratan,yoti and Ratanapala, 1963, p. 114.) (uncorrected)
43. Bizot, 1980, p. 25 2 .
44. There are of course thirty-seven chief deities present in the sabhii of the second heaven, al
the Four King's are present as guardians of the directions. See the important series of articles
on this topic by Shorto, 1963, pp. 572-91; Shorto, 1967, pp. 127-41; Shorto, 1978, pp. 152-6+
45: The thirty second item of the later canonical sources - "the brain in the skull" - becomes
two items, brain and skull.
46. Conveniently there are four piidas in a lunar mansion and nine in a sign of the zodiac; so the
piida.r could function to link the solar and lunar lists (4 X 27 = 108; 9 X 12 = 108).
47. Bizot, 19 8 1.
48. Perhaps translate: MA is suffering, but nibbana is happy. A is impermanence, but nibbana
must be heeded. U is no-self; and nibbana is cessation.
49. Bizot, 1976, p. lO2.
50. A version of this stanza is the first two lines of the Ratanattayappablziiviibhiyiicanagiit/lii. ("Verses
invoking the Power of the Triple Gem"), now attributed to King Mongkut. Howevel; even if
the attribution is correct, they could be older as this kind of Pali composition sometimes
incorporates earlier stanzas of special importance. See e.g.: Anon, 1975, p. 121.
51. See Houtman, "990.
52. Percival, 1975 [1805], p. 143
53. Ibid. p. 150.
5+ According to Bizot these are the seven items of monastic dress together with the bowl.
55 Bizot, 1980, pp. 244-47
56. Lit. the xiphoid appendage.
57. Bizot, 1980, p. 248[
31.
32.
33
34.
205
Bibliography
Adikaram, E.w., Early History ofBuddhism in Ceylon, M.D.
Gunasena/Puswella, ColombolMigoda, 1953 [1946].
Allcllin, E R., "City and State Formation in Early Historic South
Asia," SAS, 5, 1989, pp. 1-16.
- - - , "Patterns of City Formation in Early Historic South Asia,"
SAS, 6, 1990, pp. 163-73Anon, The Book of Chants. Bangkok: Mahamakut Rajavidyalaya
Press, 1975.
Bechert, Heinz, "Mahayana Literature in Sri Lanka: The Early
Phase," in Prqjfiiipiiramitii and related systems: Studies in honour of
Edward Conz;e, ed. Lewis Lancaster. Berkeley: 1977, pp. 361-68.
- - - , "VimuttiInagga and Amatakaravar:u;tana," in Amala Prajfiii:
Aspects ofBuddhist Studies, P. V. Bapat Vol., eds. N. H. Samtani
and H. Prasad. Delhi: 1989.
- - - , "Buddhistic Modernism: Present Situation and Current
Trends," in Buddhism into the Year 2000. Bangkok: Dhammakaya
Foundation, 1995, pp. 251-60,.
Beyer, Stephan, "Notes on the VIsion Quest in Early Mahayana," in
Prajfiiipiiramitfi and related systems: Studies in honor ofEdward Conze,
ed. Lewis Lancaster. Berkeley: University of California, 1977,
PP3 2 9-4
Bizot, Franc,:ois, Le Figuier aCinq Branches. Recherche sur Ie bouddhisme
khmer, Publications de l'Ecole franc,:aise d'Extreme-Orient Vol.
CVIL Paris: Ecole franc,:aise d'Extreme-Orient, 1976.
- - - , "La Grotte de la Naissance. Recherches sur Ie bouddhisme
khmer, II," BEFEO, LXVII, 1980, pp. 221,3.
/
- - - , Le Don de Soi-m&rte. Recherches sur Ie bouddhisme khmer III, .
Publications de l'Ecole franc,:aise d'Extreme-Orient, Vol.
CXXX. Paris, Ecole franc,:aise d'Extreme-Orient, 1981.
- - - , Les traditions de la pabbajjii enAsie du Sud-Est. Recherches sur Ie
bouddhisme khmer, IV, Abhandlungen der Akademie der
Wissenschaften in Gottingen, Philologisch-Historische Klasse,
Folge 3, Nr. 169_ Gottingen: Vandenhoeck u. Ruprecht, 1988.
- - - , Le Chemin de Lll1ikii, Textes bouddhiques du Cambodge. Paris:
Ecole franc,:aise d'Extreme-Orient, 1992.
Collins, Steven, "On the very idea of the Pali Canon," ]PTS, Xv,
1990.
Coningham, Robin and E R. Allchin, 'Y\nuradhapura Citadel
Archreological Project: Preliminary Report of the Third
Season of Sri Lanka-British Excavations at Salgaha Watta,
July-September 1991," SAS, 8, 1992, p. 166.
206
CJ

Cousins, L. S., "Dhammapala and the tIka literature," Religion, 1972,
pp. 159-65.
.
Deraniyagala, S. u., "Radiocarbon dating of early Brahmi script in
Sri Lanka," Ancient Ceylon, II, 1990, pp. 149-68.
Dohanian, D. K., The Mahayana Buddhist Sculpture qf Ceylon. New
York: 1977- - - . , "Prolegomena to the Study of the Mahayana in Ancient
Ceylon," in India's Contribution to World Thought and Culture
(Vivekananda Commemoration Volume).
Gombrich, Richard and Gananath Obeyesekere, Buddhism
Traniformed. Religious Change in Sri Lanka. Princeton: 1988.
GuilIon, E., L'Armee de Mara. Au pied de L'Ananda (Pagan - Birmanie).
Paris: 1985.
Harrison, Paul, The Samadhi qfDirect Encounter with the Buddhas qf the
Present, Studia Philologica Buddhica Monograph Series V
Tokyo: The International Institute for Buddhist Studies, 1990.
Hinnells,]. R, The Penguin Dictionary qfReligions, 1984.
Holt,]. C., Buddha in the Crown. Oxford: 1991.
Houtman, Gustaaf, "Traditions of Buddhist Practice in Burma,"
Ph.d, London, 1990.
Jackson, P., '1\ note on Dhammapala(s)," ]PTS , Xv, I990,
PP 20 9-II .
Kloppenborg, Ria and Ronald Poelmeyer, "Visualizations in
Buddhist Meditation," in Effigies Dei: Essays in the History if
Religion, ed. D. Plas. Leiden: E.]. Brill, I987, pp. 83-95.
Lagirarde, F., "Textes bouddhiques du pays 1.<hmer et du Lanna: un
exemple de parente," in Recherches nouvelles sur le Cambodge, ed. F
Bizot. Paris: Ecole fran~aise d'Extreme-Orient, 1994, pp. 63-;7.
Luce, G. H., "The Advent of Buddhism to Burma," in Buddhist
Studies in Honour qf1. B. Horner, eds., L. S. Cousins, A. Kunst
and K. R Norman. Dordrecht: D. Reidel Publishing
Company, 1974, pp. II9-38,.
Mudiyanse, Nandasena, Mahayana Monuments in G!ylon, M. D.
Gunasena & Co. Ltd, Colombo, 1967.
Norman, K. R, "The role ofPali in early Sinhalese Buddhism," in
Buddhism in Ceylon and Studies on Religious Syncretism in Buddhist
Countries, ed. Heinz Bechert. Gottingen: 1978, pp. 28-47.
Paranavitane, S., "Mahayanism in Ceylon," CJSc. G., II, pp. 35-71,
1928.
Percival, Robert, An account if the Island of Ceylon, reprint of 2nd ed ..
Dehiwala: 1975 [1805].
Pieris, Aloysius, "The Colophon to the Paramatthamafijusa and the
Discussion on the Date of Acariya Dhammapala," in Buddhism
in Ceylon and Studies on Religious Syncretism in Buddhist Countries, ed.
Heinz Bechert. Gottingen: 1978, pp. 61-;7.
Aspects cifEsoteric Southern Buddhism
207
Ratanajoti, G. and K. Ratanapala, Vimuttimaggo. Colombo: I963.

Reynolds, Frank E., "The Several Bodies of Buddha: Refl.e~ons on
a neglected aspect ofTheravada tradition," HR, 16, I977,
PP374-'-89
Rhys Davids, T. W, Yogavacara's Manual. London: Pali Text Society,
1896.
Ruelius, Hans, "MafijuSrlbha~ita-citrakarmasastra," in Buddhism in
Crylon and Studies on Religious Syncretism in Buddhist Countries, ed.
Heinz Bechert. Gottingen: I978, pp. 89-99.
Saddhatissa, Hammalawa, Upasakqjanalankiira. London: Pali Text
Society, I965.
Schlingloff, Dieter, Ein buddhistisches Yogalehrbuch. Berlin: 1964.
Schopen, Gregory, "The Text on the 'Dhara.r.n Stones from
Abhayagiriya'," JIABS, 5, I, 1982, pp.roo-ro8.
Shorto, H. L., "The 32 Myos in the medieval Mon kingdom,"
BSOAS, XXVI, 1963, pp. 572-9I.
- - - . , "The Dewatau Sotapan: A Mon prototype of the 37 Nats,"
BSOAS, XXX, 1967, pp. 127-4I.
- - - . , "The planets, the days of the week and the points of the
compass: orientation symbolism in 'Burma'," in Natural Symbols
in South East Asia, ed. G. B. Milne. London: 1978, pp. I52-64.
Smith, Bardwell, "Varieties of religious assimilation in early
media::val Sri Lanka," in Buddhist Philosophy and Culture - Essays
in honour cifN. A. JC9'awickrema, eds., D. J. Kalupahana and W G.
Weeraratne. Colombo: 1987, pp. 257,8.
Smith, R. B. and W. Watson, EarlY South East Asia. Oxford: 1979.
Soper, Alexander Coburn, Literary Evidencefor EarlY Buddhist Art in
China. Ascona: Artibus Asiae Publishers, I959.
Stargardt,]anice, The Ancient Pyu cifBurma, Vol. I, PACSEA.
Cambridge: I990.
Strong,]ohn S., The Legend and Cult cifUpagupta. Princeton: 1992.
Wales, H. G. Quaritch, Dvaravafl; the earliest kingdom cif Siam, 6th to lIth
century AD. London: Quaritch, 1969.
Buddhist Nuns in Transition:

the Case of Burmese Thila-Shin
Hiroko Kawanarni, Lancaster University
Introduction
Despite the fact that the original lineage of the bhikkhuni Sangha has long
since become extinct in the Theravada Buddhist tradition, there has been
an ever increasing number of female renouncers (nuns) who shave their
heads and follow the lifestyle of religious ascetics, living on alms and
observing the five, eight or ten Buddhist precepts. 1 These women are called
thila-shin (keepers of the precepts) in Burma, and similarly mae ji (female
ascetics in white robes) in Thailand. In Sri Lanka, they were generally
known as upasikli (Buddhist laywomen), though a movement towards the
ten precepts propagated by a group of female renunciants called dasa sil
mlitlivo 2 (mothers of the ten precepts) suggests a slight change in the status of
Buddhist nuns. 3
Technically speaking, these female renouncers, whom I shall call
'Buddhist nuns' for convenience, are not imbued with any special religious
status in contemporary Buddhist societies. From an orthodox Buddhist
point of view, they are in the same category as uPlisikli, meaning 'pious
laywomen'. Buddhist nuns themselves, however, strongly identify with the
Buddhist community as far as their religious affiliation and mendicant
lifestyle are concerned, but the monks generally discount their importance
in everyday life by officially adhering to the doctrine that contemporary
Buddhist nuns are nothing more than uplisikli. In this respect, the authority
of Buddhist nuns has become diminished and they are downgraded in
religious significance. The standing of Buddhist nuns is further denigrated
by their association with the negative stereotype ofthe servile, tempting and
worldly woman. Moreover, most Burmese think the nuns' major role is to
serve the community of Buddhist monks.
2 IO
Indianlnsights: Buddhism) Brahmanism and Bhakti
This paper will focus primarily on thila-shin, the Buddhist nuns in

contemporary Burma. 4 I will explore their religious position, which is often
treated by scholars as 'ambiguous', in order to assess where these Buddhist
women actually stand in the light of the prevalent beliefs and symbolism,
social reality and spirituality that surround them. In addition to that, it
seems to me that their contemporary position is 'transitional', so I attempt'
to assess the direction in which changes are taking place. The opportunity
for full ordination may be closed for thila-shin, but the rising standards of
their Buddhist education, success in Buddhist exams and the general
recognition of their academic achievements seem to have contributed to a
renewed awareness and confidence on the part of Burmese Buddhist nuns.
The recent popularity among the urban middle-class of se.nding daughters
to nunneries to become temporary nuns, also suggests a new kind of social
appreciation and. changing attitudes towards' Buddhist nuns. This
phenomenon indicates that there ate channels other than the restoration of
the bhikkhunflineage that can be cultivated to allow Buddhist nuns to exert
their religious influence and work towards establishing a more respected
and secure religious itentity.
Studies on Women in Buddhism'

A number of substantial works on the position and role of women have
been produced, in the scholarly tradition of textual interpretation (Paul
19i9, Falk 1980, Kajiyama 1982, Kabilsingh 1984, Schuster.1985, Murcott
1991). These scholars, mostly in the textual tradition, have followed the
legacy of 1. B. Horner (1930) and investigated the orthodox views and
attitudes towards women as portrayed in the Buddhist literature. Although
their works contain valuable descriptive sources of information, they are
rather limited by dealing primarily with the portrayals of ancient nuns and
laywomen in early Buddhist texts, or with the interpretation of female
images in Mahayana Buddhist texts. Whilst Buddhist literature has
provided valuable insight into the official position of women, the gap
between the past and present, myth and reality, textual and sociological is
yet to be overcome. Consequently, ~e are left with an obscure picture of
'female renouncers' who may have existed at various stages in the history of
Buddhist monasticism.
In socio~anthropological work on Buddhist societies and monasticism,
interest has centred primarily on the monks and their relationship with lay
donors, in which the role of Buddhist women was accorded little significance
(Spiro 1970, Tambiah 1970, Gombrich 197I, Bunnag I973). It was only in the
1980's that scholars started attempting to understand the contemporary
socio-religious reality of Buddhist women, of both nilns and laywomen.
The work of scholars s1J.chas Cook I981, Khin Thitsa I983, Nissan I984,
Buddhist Nuns in Transition: the Case ifBurmese Thila-Shin
21 I
Bloss 1987, Havnevik 1989, Grimshaw 1992 and Bartholomeusz 1994 is

finally catching up with the actual position of women in Buddhism.
There is also a growing interest and influence from feminist scholars
who refer to the Buddhist religious tradition for an explanation of gender
relations in Buddhist societies. Some of them criticise the religious tradition
as the main culprit for limiting women's spirituality, and the radical work of
Khin Thitsa (1980) openly criticises Buddhist values for degrading women
to the 'realm of desire'. In the meanwhile, others say that Buddhist
ideology has had little impact on the traditional beliefs and customs of a
society to which Buddhism has failed to introduce its more egalitarian
principles to the nature of gender relationships (Klein 1985). Kirsch (1985)
has observed in Thailand that women are aCtually confined to 'more
worldly' roles and economic activities due to the significance placed on
merit-making and their tendency to lapse into such activities. Keyes (1984)
has depicted the image of woman as 'mother-nurturer' from his study of
popular Thai texts. A woman is more inclined to her 'nature'; which is
essentially 'maternal', thus her primary religious duty will be to nurture the
Sangha by providing her own son as novice. Gross (1993) suggests that
women's options are limited, and there are low evaluations of their spiritual
potential in Theravada Buddhist societies. Mi Mi Khaing (1984), however,
may not assent to her view, saying that, being a Burmese.woman does not
mean playing second fiddle. She sees women as influential and
'indispensable' in Burmese society, in which they complement their male
partners, participate actively in every aspect of society, and support and
maintain the whole religious system.
Restorati<;m of the Bhikkhunz Sangha

In the face of these discussions, there are scholars and nuns (Ayya Khema,
Gross, Lekse Tsomo) who have been active in their attempts to restore the
bhikkhunfSangha in the Theravada tradition. Kabilsingh (r991), argues that
the Theravada lineage has continued till today in the Dharmagupta sect of
the Mahayana tradition. 5 Therefore, just as the Sinhala bhikkhunf
introduced full ordination to the Chinese nuns in 434 CE (Senaveratne:
10,8), it is possible to restore the bhikkhunf order in Theravada Buddhist
societies through the assistance of contemporary Chinese bhikkhunf. 6 In such
an endeavour, they hope to raise the status of Buddhist nuns since the
extinction of the bhikkhunf lineage has been the consistent explanation for
their low religious status. In fact, although the reintroduction of the
bhikkhunf order seems technically plausible, the religious authority, the
Sangha, in Theravada Buddhist countries is bound to disparage such a
lineage as Mahayana, and therefore, see their tradition as inferior in its
authenticity. 7 Accordingly, it seems unlikely that such movement will uplift
212
Indian Insights: Buddhism) Brahmanism andBhakti
the consciousness of contempor;ry Buddhist nuns as most of them are

receptive to the 'authoritative' view of the monks. Since their sielicate
religious position rests entirely in complying with the existent authority of
the bhikkhu Sangha, it is unforeseeable that the Buddhist nuns themselves
would be particularly instrumental in the restoration of the bhikkhuni
Sangha, the movement of which may only be seen as an imposition of'
'Western', thus foreign, ideas. 8
Implicit Meanings
In the Buddhist teaching, it is well known that gender is ultimately irrelevant for one's higher spiritual pursuit. Thus being a woman is no barrier to
enlightenment. This 'truth' is frequently remarked upon by Buddhist
monks and learned scholars since divinity can be recognised in both sexes.
Moreover, both men and women are bound to mortality and social conventions which are not perceived as the ultimate reality. 'When one attends
the initiation ceremony of a thild-shin in Burma, Burmese nuns chant the
thirty-two parts of the physical body and bodily substances. 9 The parts of
the body and bodily substances are listed one by one in order to remind the
initiate of the transient nature of her physical existence. Their chant
continues while her hair is shaved, "there is neither male nor female ... no
one can escape the cycle of ageing, illness and death". In this context,
gender is described as almost inconsequential for the ultimate spiritual
attainment. Nevertheless, it has to be mentioned that this soteriological
ideal is not borne out in social reality.
The social reality of these women hinges on the fundamental dualism
inherent in traditional beliefs and religious symbolism. Feminist theologians
in theJudao-Christian tradition, for example, have frequently criticised the
body and soul dichotomy, the ideology of which has associated women
with the physical realm of flesh in contrast to the spiritual supremacy associated with men. In their view, this ideology has led to the negative evaluation of women, thus historically relegating women to a symbolically 'inferior' position (Ruether 1972, Daly 1973). Among women, this dualism is further applied to separate their inherent nature between the seductive whore
and the holy mother or obedient wife. Paul (1979) has described how
women are either discredited or idealised according to the different contexts in NIahayana Buddhist texts. The female body has been treated as
their weakness as well as their strength; a vessel to reproduce and cradle life
as well as an instrument to tempt and corrupt the opposite sex.
In the Burmese context, there is a traditional belief that 'one is born into
a woman due to one's bad karma or lack of sufficient amount of merit accumulated in previous lives'. In addition to that, Burmese women in general
are referred to as the inferior sex because of their physical weakness.
Buddhist Nuns in Transition: the Case ifBurmese Thita-Shin
2 I3
Men can go and meditate alone in the deep forest but women cannot do so
because they are more likely to be attacked by bandits and 'bad spirits'.
Women are weak-minded, easily frightened, dependent, prone to illness and
vulnerable to spirits. In addition to all that, they lack the physical stamina to
tolerate the severity of ascetic training.
Such was the response of Burmese male meditators when they were
asked whether women could attain a higher spiritual stage similar to what
they were striving for. Burmese Buddhists, furthermore, believe that
women have more suffering, dokklui,10 than men because of their corporeal
state of being which is suggested as inferior to that of men. The female
body is regarded as a vessel to communicate physical and psychological
suffering such as pain, agony, grief, labour and frustration. It is also
perceived that women represent physical reproduction, being caught in the
endless cycle of thanthara or saTJlsiira, in opposition to men who can aim for
the ultimate termination of rebirth. The reproductive power of women,
which is essentially threatening, is denigrated by reference to the Buddhist
ideology, as yet another beginning in the endless cycle of rebirth.
In contrast to the laypeople, monks and nuns are religious ascetics who
have renounced their reproductive power and sexuality by their commitment to celibacy. Burmese nuns refer to themselves as having to suffer lesser
dokklza compared to laywomen who have little control over their own body.
A laywoman in general is thought to endure all of the aweinikti dokkha nga-ba
(five aspects offemale suffering). These are I. menstruation, 2. pregnancy, 3.
childbirth, 4. separation from her parents, 5. having to attend her husband.
Moreover, once a woman becomes a mother, she is considered to develop a
strong thanyawzin, which is a deep emotional attachment to her family and
children. Burmese informants have often commented that everyone, both
man and woman, has thanyaw;:;in, but the strongest feeling is experienced by
a mother. Her thanyawzin is appreciated by family members but is regarded
as rather negative since it binds and confines her to a stronger degree of
suffering. Thus the stronger the attachment is, the more the notion of aneiksa
(impermanence) is stressed to counterbalance the burden of heavy emotions.
The nuns have already left their parents and are not married, so they are at
least spared four aspects of such female suffering, considered to be the cause
of miseries and pain. The only burden they have to carry is the dokkha of
menstruation. Thiia-shin actually say that it is their good karma that they
have chosen a chaste life, and now that they do not have to go through the
pain of childbirth or tolerate a drunken husband. The moment a woman
shaves offher hair, it implies that she'has become celibate; in other words,
she has taken full contr'ol of her own sexuality by giving it up altogether.
This rejection of sex, rejection of reproduction, rejection of the endless cycle
of rebirth, and rejection of suffering, come across as a powerful religious
statement from the Burmese Buddhist nuns.
2I4
The Buddhist nuns have transcended what are perceived to be the fundamentallim,itations of womanhood and have embodied religious ideals.
Yet they have left revered roles in society as mothers and wives; a rebellious
act which will attract its own share of challenges in the form of criticism
and suspicion from society and the religious authority. After all, female
renouncers .are not seen to be entirely relieved of their biological capabilities. Therefore, they are seen as implicitly dangerous, thus innately threatening to the spiritual ideal of Buddhist asceticism in the eyes of the monks.
Consequently, these women are seen as challenging the existing order and
disrupting the religious and ideological division between physical women
and spiritual men by their very presence in the monastic community. The
social and moral pressure of their ambivalent position manifests itself in
various ways .. Most nuns admitted to praying, sometime in their career, to
be reborn as amale person, hopefully a monk, in their next life.
Informal Power
In order to fully comprehend where Buddhist nuns stand and the actual
roles they fulfil, we have to look at the informal power they exert in
everyday monastic life. I have suggested that their lack of formal religious
status and the. attribution of negative images do not necessarily imply that
the nuns are treated as inferior and downgraded in their daily activities.
On the contrary, thila-shin have a strong presence in Burma/ 1 in which
they take initiatives, make decisions, and exert considerable influence in
their relationship to the monks and to the lay devotees. Some of the roles
they actively engage in help distance the holy monks from secular forces.
The nuns tell the lay visitors where to take off their shoes, how to prostrate
themselves in front of the monks and how to use religious honorifics
properly, and by doing so, make sure that the sanctity of the monks is well
maintained. The nuns' identity thus rests with the monastic community of
the monks, and they are enthusiastic about policing the boundary and
protecting it from corrupting influences. Such functioning as a buffer zone
is probably facilitated by their ambiguous religious position, or it may be
the result of their standing between the religious and the secular. However,
their ultimate motivation behind such role-playing has t9 be fully explored
if we are to truly understand what they are trying to achieve.
One of the ways that thila-shin make themselves useful is through the
handling of money. As the Vinrrya prohibits the monks from the handling of
money or fire, the nuns make themselves almost indispensable to the monks
in these areas. These functions give Buddhist nuns a certain level of influence as they often come to be in charge of the kitchen and monastery
finances. It is common for them to welcome lay visitors, look after young
novices, organise ceremonies and in many cases, take an active part in
Buddhist Nuns in Transition: the Case qfBurmese Thila-Shin
215
running the monasteries. However, it has to be pointed out that such economic power does not imbue them with any more power in the religious
domain. In other "livords, the ability to handle money is regarded as opposite
to the spiritual potency which carries the highest prestige. This ability, there~
fore, assures the nuns of lower rather than higher prestige, and forces them
to take money, the 'hot substance', full of worldly pollution on behal(ofthe
monks who do not touch it. This ability adds to the perception that they are
'too attached' to the secular world: in. other words, spiritually inferior.
Yet the majority of thila-shin are more than keen to offer help and, serve
the monks in their belief that they are accumulating spiritual merit. Such
activities involve various kinds of menial work and domestic chores, such as
cooking, cleaning and attending the Sangha. All such services are regarded
as meritorious and good for their moral balance, but it is generally
accepted that merit-making activities are inferior to meditation and textual
learning to which true Buddhist renouncers should aspIre. Buddhist nuns
play the symbolic role of daughter to the Buddha, and symbolic mother to
the monks. It is a plausible way in which female renouncers pursue their
religious gDals within culturally valorised roles. Having said that, it is ironic
that Buddhist nuns who have left their families and renounced the traditional roles of mother and wife, should lapse into a similar domestic
situation and even find it their major raison d'etre to serve yet another set of
dependants, the monks. 12 Buddhist nuns, however, are in the front-line as
the most pious among faithful devotees who uphold the religious tradition.
They are keen on altruistic self sacrifice and virtuous acts for the monks,
and simultaneously they are easily influenced by the monks since such acts
of devotion and loyalty provide them with a secure religious identity. Yet
this may all be an expression of their insecurity in the religious domain. In
other words, it may be their lack of formal religious standing that obliges
the nuns to handle money on behalf of the monks, and by acting as minder
to the monks and the monastic community as a whole, the Buddhist nuns
are trapped between the secular and the religious world.
Religious Roles
As far as the religious roles and functions of thila-shin are concerned, they
act as important ritual specialists, educators and propagators of Buddhism.
Officially, however, most important ritual roles such as officiating at I
religious ceremonies are closed to the nuns. 13 When the monks officiate,
their role accords with the superior religious status of the Sangha. The
nuns, being competent ritual specialists equipped with detailed knowledge
of every ceremonial procedure, complement the monks in Buddhist
ceremonies. They normally sit in front of the lay congregation facing the
monks, firstly in their position as the front -line of the pious, and secondly as
2i6
ritual specialists. The ceremony only starts when the nuns invoke the
monks by initiating the mandatory prayer glorifying the 'Three Jewels': the
Buddha, the Dhamma and the Sangha. After the prayer, the chief monk
starts the ceremony. The monks and nuns remain in ritual dialogue
throughout the ceremony, at times in verbal acknowledgemerit or by using
bodily gestures such as nodding or prostrating themselves towards the
monks, which occur at the end of each prayer sequence or at small intervals
of the sermon. Symbolically, the monks give and the nuns receive on behalf
of the laity, by which act they lead the whole congregation. By knowing
every detail of the ceremonial sequence, the nuns facilitate and make sure
that the ceremony is completed without inconvenience. Having noted the
nature of their actual role in a ceremonial situation, it has to be mentioned
that it is not technically impossible for a nun to officiate. This involves the
giving out or the conferring of the five or eight precepts as well as the
reciting of major payeik 14 prayers in the Pali language. The problem does
not seem to concern the ability of the religious person concerned so much
as her religious status and official qualifications. 15 If a thila-shin were asked
to conduct a Buddhist ceremony, it is most likely that she would refuse for
the simple reason that it implied a major public performance. In this
respect, it seems to me that the issue is about social approval for granting
more religious authority to women, and this depends on whether the
monks, laity and the nuns themselves are willing to accept such a precedent.
On this point, Burmese people expressed apprehension about allowing
important ceremonies to be handled by nuns who observed only eight
precepts (ten at the most). The monks obviously displayed their displeasure
at such an unacceptable suggestion, and the nuns themselves, whose
primary interest rests in consolidating their existing authority by upholding
the sanctity of the monks, seemed perplexed by such a possibility.
Preaching is another area of religious importance. A Buddhist nun, in
theory, is not obstructed from preaching on the grounds of being a woman.
But despite stories of famous female preachers in ancient Buddhism,16 the
incidence of contemporary Buddhist nuns preaching in public is still rare. In
a country like Burma where the media is not highly developed, preaching is
a powerful way of communicating and popular monk preachers attract large
crowds, exerting considerable influence on their audience. In Sri Lanka, it is
reported that many nun preachers such as Sister Sudharma are becoming
accepted. The increase in number of these nuns show that they are moving
into areas which have "always been a male prerogative" (Gombrich &
Obeyesekere: p. 285).17 Compared with Sri Lankan nuns, preaching nuns in
Burma have not made their way in public. 18 Having said that, thila-shin are
actually keen on the idea of preaching and propagating Buddhist teachings.
The notion of thathana-pyu, meaning 'to spread the religion', appeals to them
as one of the most meritorious deeds. However, actual preaching takes place
mostly in informal and private settings: inside the confinement of nunneries
Buddhist Nuns in Transition: the Case qfBurmese Thild-Shin
2 I7
and in people's houses. They preach to villagers during tours where village
girls who have few options other than getting married and bearing children
are persuaded to join the monastic community. Thita-shin also visit prisons
and preach to ferriale criminals, or go to hospitals to console the patients,
but these are limited activities in restricted circumstances. Most learned
nuns are experienced at public speech since oration is part oftheir religious
training. They are an enthusiastic audience of the monks' sermons, wellversed and knowledgeable, with a good number of Buddhist anecdotes to
draw from, So again, they are well qualified to propagate the Buddhist
teachings to the general public. However, most of them are reticent as the
idea of preaching in public, especially in front of a large audience, is seen as
a male prerogative. Furthermore, the nuns' first and utmost concern will be
that such public display oflack of modesty may taint the purity of their self
image. And such exceedingly bold behaviour can even jeopardise their
relationship with many oftheir lay donors on whom they are completely
dependent. Some said, "a nun must be modest and shy. It is bad for a
woman to be bold". And most of them agree that "it is the role of the monks
to preach". Thus, most Burmese nuns whom I interviewed were not willing
to risk their prudent image and challenge socio-religious conventions lest
they should undergo unpleasant experiences in public, which they
anticipated might happen, and lose the support of their lay donors. In this
respect, personal merit such as educational qualifications, which is becoming increasingly important in contributing to improve the religious standing
of the nuns, is yet to show its influence in the people's perception of the nuns
or in their general image. However, such changes may be on their way.
Education as an Asset
In Burma, where Buddhist education was traditionally promoted and
encouraged by the kings, the state support of monastic higher education
remained until the present and has upheld the standards of Buddhist
learning among the monks and nuns. There were periods under the British
rule when national exams were suspended, but by the end of the I9th century, annual Buddhist exams were officially reinstated, continued until I941
and then resumed in 1946. The intervention of the government in setting
academic standards to maintain the quality of Buddhist teachers and standardising text books, as well as the notion of having their religious knowledge examined, met with a degree of resentment from some quarters of the
Sali.gha. However, advantages brought about by achieving degrees
awarded by the government and a strong tradition of learning in the
Sangha counterbalanced any such hostility. This was combined with a
reward system!9 which gave grants and prizes to successful students and
therefore provided them with strong incentives to study the Buddhist texts
'218
(Mendelson, 1975, pp. '250-'2). It is in this historical background that

Buddhist nuns have been given the opportunity to show their ability and
competence by learning the Buddhist scriptures. 2o
Around the turn of this century, Buddhist nuns were first given permission to sit for the state exams. When the first nun passed the national exams
in 1915 with flying colours, more were encouraged to compete. An increas- .
ing number of nuns studied for exams despite hardship and obstacles. By
1947, the first nun from Sagaing passed the Damasariyd 21 exams (Nandamala Ven, 1980, p. 160), and since then the desire to establish oneself as a
Damasariyd degree teacher has become one of the primary motivations for
a Burmese Buddhist nun. Nowadays thild-shin are allowed to sit up to the
fifth level in the national exams. 22 If she passes up to Patama-gyi (advanced
level), which is the fourth level, she is qualified to teach the Buddhist scriptures and start her own nunnery. vVhen she passes the final exam and is
given a Damasariyd title, she is granted not only an official credential to
teach but also prestige and a general acknowledgement of her religious
worth."'
The pursuit of higher education, however, has not been an easy path for
the nuns since there were customs and conventions that discouraged them.
There were also procedural problems. It was assumed that all candidates
were resident students of the institution, monastery or nunnery, from which
they were presented to the examinations and that they pursued their studies under lecturers residing in the same institution. Many nuns, however,
whilst residing in their nunneries had to commute long distance to receive
tutelage from learned monks (Mendelson, ibid., pp. '250-1). Besides, there
were not many monks who agreed to teach the nuns, or even if they did,
nuns were confined to the back of the classroom or to the other side of the
curtain so that they remained invisible in the eyes of the monks. The hardship and disadvantage endured by them were hardly acknowledged. They
were not given special financial support for their study and books were
hard to come by. The biggest hindrance came from the monks who refused
to give them tuition at all, declaring that nuns should be satisfied with the
basic prayers. In their quest for learning, nuns in the pre-war period sought
out learned 'ex-monks' (hpongyi ludwet) who had left the Buddhist Order,
and studied under their tutelage. 24 In the post-war period, there were more
monks who came to accept the need to educate the nuns. In this context,
Musawyin Hsadaw, a highly respected abbot in Mandalay, who had been
conferred the title Agga Maha Pandita lVIaha Thera (Monk of the Highest
Learning), played an important role. He was also politically influential as a
member of the Supreme Sangha Council during the U Nu regime in the
I950s. He publicly stated his support for the uplift of the standard of education for the Burmese nuns. Since then, many senior nuns commented that
the Sangha and society have become much more supportive towards their
education.
Buddhist Nuns ,in Transition,' the Case
ifBurmese Thild-Shin
2Ig
In contrast to what one may achieve in patipatti (meditation), the level of

achievement in pariyatti (learning the scriptures) can be easily recognised
since the distinction between those who are learned and those who are not
is clear. The results of private and state exams are openly discussed, awards
are given to the students with honours and lay donors give support to those
who have achieved better degrees. As far as academic qualifications are
concerned, it must be emphasised that they are gender-neutral and provide
both monks and nuns with special channels to acquire regular patronage
and to upgrade their careers in the religious community. As more thild-shin
emerge as teachers and scholars, their contribution to Buddhist scholarship
in areas of literature, scriptures, treatises, commentaries, philosophy and
Pali studies, has become widely acknowledged. Consequently, the rise of
the nuns' educational standards and contribution have not only helped to
upgrade their socio-religious status, but also had a positive effect on people's previously ambivalent attitudes towards Buddhist nuns.
Alongside changes in general attitudes and more positive images of
Buddhist nuns emerging, one of the curious social phenomena is the
growing popularity among urban middle-class girls to shave their heads
and take temporary ordination. The 'temporary' status for a Buddhist nun
or a monk is referred to as dzlldbd. All Burmese Buddhist boys become
d1ildbd, or temporary novice monks, at least once in their lifetime but it was
never common until recently for girls to become nuns even for a temporary
period. Whilst the experience as a monk is highly prestigious, a similar
experience as a nun has not been seen as a means .of acquiring equivalent
respect and reward. 25 However, the recent phenomenon seems to suggest
an apparent change in the general perception of Buddhist nuns. Nowadays,
nunneries in Yangon and NIandalay are crowded during the summer
holidays by freshly shaven young nuns between the ages of five and twenty.
They go through an initiation ceremony and take the eight precepts which
oblige them to fast daily after midday and to abstain from sex, alcohol and
other kinds of worldly pleasure. They also have to live as alms collectors
and experience the hardship and humility of being at the mercy of people's
goodwill. The duration of their nunhood is arbitrary as it depends on the
convenience of every temporary nun, but an average period will be
between a week and ten days. Although few of them actually stay on to
become permanent nuns, their basic aspirations are different. NIost of the
girls I interviewed emphasised that it was their own decision to become
temporary nuns and their aim was to acquire spiritual merit. As it has
become almost fashionable to become a temporary nun in Yangon, the
phenomenon presents a noteworthy shift from the traditional negative
image of old kin-less women living on charity and in poverty, to a new
glamorous image of young Buddhist teachers exemplifying spiritual purity
and religious worth.
220
Indian Insights: Buddhism, Brahmanirm and Bhakti
Towards Higher Spirituality

The Buddhist nuns maintain their.ambiguous position by juggling two
identities, one secular and the other religious. Their lack of formal religious
status may well foster a sense of insecurity, but there is an attempt to
overcome this by creating an image of piety, purity and spiritual worthiness
in their capacity as 'pious laywomen'. One possible new direction for their
religious standing may be found in the movement towards taking the ten
precepts, as in the case of in Sri Lanka. The dasa sil matilvD are prominent
teachers in vipassanii meditation who strive for moral purity and observe
precepts which free them from the handling of money and mundane
matters. Bloss has suggested that they are neither a part of the Sangha nor a
lay order, but their final aim is to become arahat (1987, p. 18). SchusterBarnes notes that their movement has earned much respect for these
women as true renunciants (1994, p. 154). Unlike dasa sil matilvD, Burmese
thila-shin; who are mostly eight precept observers, do not aim to detach
themselves from the mundane realm. It has been mentioned that mediation
between the secular world and the Sangha is an essential part of their
religious identity and it will not serve their interests to become marginalised
in their quest for more religious influence. But the mediating role through
which the nuns seek empowerment, particularly in the handling of money,
also limits their prestige and worth as renouncers. So if they are given the
chance, most of them would probably opt for becoming ten precept
observers. At present, there are about six ten precept nuns in Burma who
are much respected and considered to be higher than ordinary eight precept
nuns on the spiritual ladder. Most of the eight precept abiding nuns cannot
afford to become ten precept observers, but they still make a point of not
going on alms and refraining from cash transactions on days of rest, or
designating a special day or a length of time during which they can become
part-time ten precept observers.26 By doing so, they attempt to achieve a
higher spiritual state of detachment and tranquillity. In their quest for
higher spirituality, however, it is difficult to anticipate the Burmese nuns
becoming independent of the monks or from their immediate kinship ties.
We have seen how merged their identity was in the actual running of the
monastic community. The nuns derive their status from association with the
monks and from the part they play in enabling the monks to separate from
the worldly. Therefore, equality and independence may not be an attractive
proposition for them but rather threatening and confusing to their basic
sense of religious identity.
It seems to me that the most plausible direction for the strengthening of
the nuns' religious position is in the area of Buddhist education. Renewed
religious awareness ancf confidence, and the growing popularity of temporary nuns may suggest increased opportunities and increasing respect and
Buddhist Nuns in Transition: the Case ofBurmese Thita-Shin
221
recognition for the nuns' religious position. Education for them has become
their 'symbolic capital' (Bourdieu, I977). The rising academic standards of
the nuns may allow them to secure a higher and more equal position in the
religious domain without becoming marginalisedor treated as 'anomalies'.
Notes
1.
Five precepts observed by lay Buddhist,:,

I. Abstention from killing.
2. Abstention from stealing.
3. Abstention from adultery (sexual mi,demeanours).
4. Abstention from telling falsehoods.
5. Abstention from the taking intoxicant, (alcohol and drugs).
Eight precept' observed by Buddhist nuns and laypeople while taking special vows: precepts
I, 2,
2.
3.
4.
5.
6.
7.
8.
g.
ro.
II.
12.
13.
'4.
4, 5, are the same as above.
3. Celihacy.
6. Abstention from taking solid food after midday.
7. Abstention from dancing, singing, music and shows, from .garlands, perfumes,
cosmetics and adornments.
8. Abstention from sleeping on high and luxurious beds.
Ten precepts observed by Buddhist nuns: precept' I t6 8 are the same as above.
g. The sending of loving kindness to all sentient beings (strictly speaking,. this i, not a
precept. I have described it elsewhere (lgg0, p. 24)).
10. Abstention from handling gold and silver (money).
BlOSS,lg87.
In the late Ig80s, there is said to be more than 25,000 full-time Buddhist nuns in Burma,
8,000 to 10,000 in Thailand and 5,000 in Sri Lanka.
Burma officially changed its name to the Union of Myanmar in June Ig8g. The capital
Rangoon became Yangon. For this paper, however, I do not adopt the Burmanisation of
place-names since it only changes the English spelling and the pronunciation remains the
same in Burmese.
Gombrich & Obeyesekere confirm that the differences in the two traditions "have no bearing
on monastic traditions" since they concern only the doctrine (lg88, p. 274).
Dr H. Gunatillaka who visited China in 1985 to study the Chinese bhiJddzunf order claimed
that contemporary Chinese blzikklzunffollow the same rules and regulations prescribed in the
Dharmagupta Vin'!Ya: 8 Pariijika, 8 Sangha-dis.sa, 17 Nzssagiya, 30 Piicittrya and 7 Adlzikarana
Samatha prescribed in the Theravada bhikkfzunf Vm'!Ya (Daily News of Sri Lanka, '4 September
Ig8g). The Theravada has 166 Piitimokkha rules for blzikkhunfwhile the Dharmagupta has an
additional 12 in the section of minor offenses, making the total 178.
Ven. Vorami is the first Thai nun to have received full ordination as blzikkhunfin the Chinese
Dharmagupta lineage. She is, however, only a maeji to the general public. "To the Thai
Sangha, her status i, at best that of a Mahayana blzikkhunc' (Kabilsingh, Iggl, p. 52).
See Nissan, Ig84, p. 45.
The cha~t is called tllOTl?e-hnik ko-hkanda, literally meaning, 'thirty-two part, of the body' in
Burmese. It is recited in the Pali language, which starts from kesa (hair) and ends in mulla
(urine).
Dokkhd is a Burmese-Pili word commonly used in colloquial Burmese meaning dukklla
(suffering) as in the PaJi language.
I am referring to the Buddhist community of Sagaing Hills in Upper Burma, in which
approximately 10,000 monks, novices and nuns live side by side.
Having said that, not all nuns engage in domestic chores'all the time. The nuns who teach
and learn are usually exempt from menial work.
Although the monks are seen to be the 9fficiaters of rituals due to their superior religious
position, there are no regulations to prohibit the nuns from officiating.
Paritlii.
222
IS
16.
17
18.
19

In theory, it is conceded that those who keep more precepts can confer to those whose status
implies fewer. I have witnessed a young novice, a ten precept abider, officiate and confer five
precepts to the lay audience. If that is acceptable, it should also he posssible for a Buddhist
nun, an eight precept abidet; to perform the same role.
Early Buddhism hal produced many nun preachers, out of whom 'Dhammadinna' is the
most well-known (Altekar, 1978, p. 209).
One educated nun, Ma Vicari, gave 2,530 public sermons in twenty years (ibid., p. 281). .
Thila-shin Daw Yusanda is one of the rare Buddhist nuns who preaches at the Shweidagon
Pagoda in Yangon. She hal been preaching during the Buddhist Lent, July to October, for
almost twenty years. Her topics !liainly concern the social behaviour of women, morality and
other general religious themes. She attracts a wide audience but the majority of them are
women and children.
In the post-war period, apart from national exams conducted by the Burmese Government,
an increaling number at local townships and private Buddhist associations have become
active in organising Buddhist exams.
20.
21.
In Thailand nuns learn ordinary chants in the vernacular form, the same as that uttered by
the lay congregation. Thai nuns have little access to the Buddhist knowledge which is
believed to endow monks with magical powers.
The term means, the teacher of the Buddha's Dizamma, the highest level ofPali scholarship or
the name of the examination.
22:
25
The five levels are lvluLa-ddn (elementary level), Patama-nge (lower level), Patama-La (middle
level), Patama-gy! (advanced level) and Damasariya. Buddhist texts such as Dizammapada,
Ablzidlzammattha Sangalza (a twelfth-century Sinhalese compendium of Ablzidhamma philosophy
known a< 171ingyo in Burma) and sections of the Vzn'D'a (depending on the level being taught)
are essential for all levels. Diziitukatlza (the third book of the Abhidhamma) is taught from the
second level, Burmese grammar at all levels and Pali grammar from the third level onwards.
At the Patama-gy! level, the study becomes much more difficult as students have to tackle the
hardest book of the Ablzidlzamma, Pattlzana Kusalatika. At the Damasarfva level, they study
Parajika-kanda, Diglza Nikfiya and Dlzammll.langa1Ji (Mendelson 1975, Appendix G, Report of the
Ministry of Religious Affairs, Burmese GoY. 1990).
There are more than two hundred nuns who hold the Damasariva title in the Buddhist community of Sagaing alone. Sagaing has a population of about 3,~00 nuns which is about one
tenth oftl1e total nun population in the whole of Burma.
In the autobiography ofDaw Malayi, it is described how she went with other nuns to live in
the 'illage of a famous ex-monk, Htayanga Hsaya-gyi, 'and attended his intensive lectures.
He was one of the most learned monks at that time who had disrobed due to political
intrigues. He spared much of his time and knowledge for the nuns and devoted his later life
to improving their education, so he later came to be called 'father of the nuns' education' in
Burma.
An exception was reported from the Tenassarim Division of Lower Burma where it was traditionally compulsory for girls from Buddhist families to become nuns temporarily while
boys became novices in the slzin~Y71 ceremony (1986, Beik-myo thathana-yei hnin, Beik-myo
thiIa-shin-mya.lvya Mingala, 6: 2;-8, 66/).
As far as the taking of the ten precepts is concerned, in Thailand, Kabilsingh (1991) reports
that there were female renunciants called siLacarini who observed the ten precepts at Wat
Chanasongkram (Bangkok) in the 1960s. She also mentions sikhamat who are the ten precept
nuns at Santi Asoke's ordained community. The practice is often combined with vegetarianism, both of which the practitioners see al a move towards higher spirituality.
Bibliography
Altekar, A S. The position of women in Hindu Civilization:from prehistoric
times to the present day. New Delhi: Motilal Banarsidass, I956
[rep. I978J.
Buddhist Nuns in Transition: the Case ofBurmese Thila-Shin
223
Barns, N. "Women in Buddhism" in Women in world religions, (ed.) A.

Sharma. Albany: State University of New York Press, 1987Bartholomeusz, T Women under the Bo tree: Buddhist nuns in Sri Lanka.
Cambridge: University Press, 1994.
Bloss, L. W "The female renunciants of Sri Lanka: the
Dasasilamattawa" in Journal of the International Association OJ
Buddhist Studies IO, pp. 7-31, 1987.
Bourdieu, P Outline of the theory ofpractice. (trans. ) R. Nice.
Cambridge: University Press, [1977J 1987.
Bunnag, J. Buddhist monk, Buddhist layman: a study of urban monastic
organization in central Thailand. Cambridge: University Press,
1973
Cook, N. M. "The position of nuns in Thai Buddhism: the
parameters of religious recognition". MA Thesis, Australian
National University, 1981.
Daly, M. Beyond God the Father: towards a philosophy of women's liberation.
Boston: Beacon Press, 1973.
.
Falk, N. A. "The case of the vanishing nuns: the fruits of
ambivalence in ancient Indian Buddhism" in Unspoken worlds:
women's religious lives in non- Western cultures, (eds.) N. A. Falk & R.
Gross. New York: Harper & Row, 1980.
Gombrich, R. F Precept and practice: traditional Buddhism in the rural
highlands of Ceylon. Oxford: Clarendon Press, 1971.
- - - . & G. Obeyesekere, Buddhism Traniformed: religious change in Sri
Lanka. Princeton: University Press, 1988.
Grimshaw, A. Servants of the buddha: a winter in a Himalayan convent.
London: Open Letters, 1992.
Gross, R. Buddhism after patriarchy: afeminist history, analysis, and
reconstruction ofBuddhism. Albany: State University of New York
Press, 1993.
Havnevik, H. Tzbetan Buddhist nuns: history, cultural norms and social
reality. Oslo: Norwegian Univ. Press, 1989.
Horner 1. B. liVomen under primitive Buddhism: laywomen and almswomen.
London: George Routledge and Sons, 1930.
Kabilsingh, C. A comparative study qf bhikkltunT Patimokkha. Varanasi:
Chaukhambha Orientalia, 1984.
- - - . Thai women in Buddhism. Berkeley: Parallax Press, 1991.
Kajiyama, Y "Women in Buddhism" in Eastern Buddhist 15,
pp. 53;0, 1982 .
Kawanami, H. "The religious standing of Burmese Buddhist nuns
(thila-shin): the. Ten Precepts and religious respect words" in
Journal of the International Association qfBuddhist Studies, 13, pp.
17-39, 1990.
224 Indian Insights: Buddhism, Brahmqnism and Bhakti

Keyes, C. F. "Mother or mistress but never a monk: 'Buddhist notions
of female gender in rural Thailand" in American Ethnology, I I,
pp. 223-41, 1984.
Khin Thitsa Providence and prqstitution: image and reality for. women in
Buddhist Thailand. London: Change International Reports,
1980.
- - - . "N).lns, mediums and prostitutes in Chiengmai: a study of
some marginal categories of women" in Women and development in
South-East Asia (I). Canterbury: University of Kent, 1983.
Kirsch, A. T "Text and context: Buddhist sex roles/culture of gender
revisited" inAmerican Ethnology, 12, pp. 302-20, 1985.
Klein, A. C. "Primordial purity and everyday life: exalted female
symbols and the women of Tibet" in Immaculate and powerfol: the
female in sacred image and reality, (eds.) C. WAtkinson, et al.
Boston: Beacon Press, 1985.
Mendelson, E. M. (ed.)]. P. Ferguson Sangha and state in Burma: a sturfy
qfmonastic sectarianism and leadership. Ithaca: Cornell University
Press, I975.
Mi Mi Khaing The world qfBurmese women. London: Zed Books, I984.
Nanda-mala Biwuntha, Ven. Thameikdaw-daya hsaya-gy1. Daw Malayi
. (The venerable teacher qfThameikdaw-daya nunnery, Daw Malayi).

Rangoon: Nylin Sapei-daik, 1980.
Nissan, E. "Recovering practice: Buddhist nuns in Sri Lanka" in
South Asia Research, 4, pp. 32-49, I984
Paul, Y D. Women in Buddhism: images qf the feminine in Mahiiyiina
tradition. Berkeley: University of California Press, [I979] I985.
Schuster, N: "Striking a balance: women and images of women in
early Chinese Buddhism" in Women, religion and social change,
(eds.) Y Haddad & E. Findley. Albany: State University of New
York Press, I985.
Senaveratne,]. M. "Some notes on the Chinese references" in Journal
qfRoyalAsiatic Society (Ceylon Branch), 24, pp. I06-II, I915.
Spiro, M. E. Buddhism and society: a great tradition and its Burmese
vicissitudes. Berkeley: University of California Press, 1970.
Tambiah, S.]. Buddhism and the spirit cults in North-east Thailand.
Cambridge: Up, 1970.
Willis,]. D. "Nuns and benefactresses: the role of women in the
development of Buddhism in Women, religions and social change,
(eds.) Y Haddad & E. Findly. Albany: State University of New
York Press, 1985.
Ruether, R. Liberation theology. New York: Paulist Press, 1972.
Wijayaratna, M. Buddhist monastic ljft: according to the texts qfthe
Theraviida tradition. (trans.) C. Grangier & S. Collins. Cambridge:
Note on Contributors
Dermot Killingley
Dermot Killingley was born in Liverpool in 1935. He studied Latin, Greek
and Sanskrit at Oxford, and later studied at the School of Oriental and
African Studies, London. He has taught Sanskrit and Indian Culture at the
University of Malaya, Kuala Lumpur, and is now Senior Lecturer in the
Department of Religious Studies, University of Newcastle upon Tyne,
where he runs an annual seminar on the Sanskrit Tradition in the Modern
World.
He is interested in the Sanskrit tradition and its interaction with modern
India and the modern world, and his publications in this field include
Rammohun Roy in Hindu and Christian Tradition (The Teape Lectures 1990),
Newcastle upon Tyne, Grevatt & Grevatt, 1993, and Hindu Ritual and Society
(with Werner Menski and Shirley Firth), Newcastle upon Tyne, Grevatt &
Grevatt, 1991. He is also interested in the presentation of Hinduism in
religious education, and has written a book for teachers, Approaches to
Hinduism (with Robert Jackson), London, John Murray, 1988. He has
written a descriptive sketch for linguists, Sanskrit (with Siew-Yue Killingley),
Munich, Lincom Europa, 1995, and his teaching book, Beginning Sanskrit, is
to appear shortly from the same publisher.
Peter Connolly
Peter Connolly studied for his BA, MA, and Ph.D. degrees in Religious
Studies at the University of Lancaster. Since 1980 he has held a post as
Lecturer in Religious Studies at the Chichester Institute of Higher
Education, Chichester, England. His recent publications include Vitalistic
Thought in India (1992), Buddhism (1992), "Phenomenological exegesis and
Patanjalis Yoga Sutra" in Masefield P and Wiebe D. (eds.) Aspects ofReligion:
essays in honour ofNinian Smart (1993) and "Hypnotic dimensions or religious
worldviews" diskus Vol 3, No.1, 1995. Dr Connolly's current research is
mainly focused on the hypnotic aspects of a range of religious experiences.
226 Indian Insights: Buddhism, Brahmanism and Bhakti.
Mark Allon
Mark Allon first studied fine art in Sydney, majoring in drawing,
lithography and sculpture. He took an Honours degree in the languages of
Buddhist texts and in Western philosophy at the Australian University,
Canberra, translating the Buddhist Sankrit Mahiiparinirvii7JiLSutra for his
Honours thesis. He completed his Ph.D. in Buddhist Studies at the
University of Cambridge in 1994, and since then has taught and worked as
a resea~ch assistant at the School of Oriental and African Studies, London.
He is currently involved in the publication and indexing of the collected
papers of John Brough. He will soon take up a Bukkyo Dendo Kyokai
Fellowship.
Sally'Mellick Cutler
Sally Mellick Cutler was born in Australia in 1957. She moved to the UK in
1979 and began her first formal study in the fieldin Indology at the School
of Oriental and African Studies in London in 1982. In 1994 she obtained a
D.Phil. from Oxford University and was blessed with a son. She lives in
Oxford where she is the manager of the Pali Text Society Office.
Lynn Thomas
Lynn Thomas is lecturer in Hinduism at the Roehampton Institute,
London. She received a BA and MA from the University of Lancaster and
a D. Phil from the University of Oxford in 1987 for a thesis on theories of
cosmic time in the M.ahabharata. Her publications include articles on
various aspects of this topic and other research interests include the
. portrayal of women in Hindu and Greek mythology.
David Smith
David Smith teaches South Asian Religion and Art at Lancaster
University. A Sanskritist by training, he has published Ratniikara's Haravijqya:
An Introduction to the Sanskrit Court Epic (OUP' 1985), and The Dance if Siva:
Religion, Art and Poetry in South India (CUP, 1996). He has visited South. India
many times, and is currently writing a book on South Indian Temple
Painting.
Note on Contributors
'2'27
Anthony Tribe
Anthony Tribe received his doctorate in Classical Indian Buddhism from
the University of Oxford in I995 and is currently teaching in the
Department of Philosophy at the University of Montana.
Robert l\!layer
Robert Mayer received his Ph.D. from the University of Leiden in the
Netherlands. His special interests are ideas of canonicity in Tibetan
. Buddhism, and the Ancient Tonga (rNying-na-pa).
Theodore Gabriel
Dr Theodore Gabriel was born in Badagara, Kerala State, India. He did
postgraduate work in Social Sciences in the department of Religious
Studies and Sociology at the University of Aberdeen and took his M.Litt in
198'2 and Ph.D. in 1986, studying under Professor Andrew Walls. He taught
Islamic Studies at the University for one year and later on moved to the
College ofSt Paul and St Mary (Now Cheltenham and Gloucester College
of Higher Education), Cheltenham, as Lecturer in Religious Studies where
he continues to teach Islam and Indian Religions.
Dr Gabriel's research interests are in the area of Islam in South and
Southeast Asia, Hinduism in South India, tribal religions and interreligious relations. His book on the Muslims of Lakshadweep came out in
1989. Two more books will be published shortly, Hindu-lvIuslim Relations in
North Malabar 1498-1947 (E. Mellen Press) and Christian-Muslim Relations in
Sarawak, East Malaysia (Ashgate Publishing House). He has also contributed
several articles to academic journals in India and the United Kingdom.
Gavin Flood
Gavin Flood teaches in the Theology and Religious Studies Department,
University of Wales, Lampeter. He has published a number of articles and
two books including An Introduction to Hinduism (CUP, I996).
228
Indian Insights: Buddhism, Brahmanism and Bhakti.
L. S. Cousins
L. S. Cousins - formely Senior Lecturer in Comparative Religion at the
University of Manchester, where he taught various courses on Buddhism
and also one on Mysticism and others at different times on various topics
connected with Indi,an religion, includingJainism and Pali. Educated at St
John's College, University of Cambridge (History and Oriental Studies).
Author of the chapter on Buddhism in: J. R. Hlnnells, Handbook qf Living
Religions, Penguin/Viking, I984 (revised and enlarged version to appear in
I996) and of around twenty articles on Buddhist meditation, the history of
the early Buddhist schools and Abhidhamma. Co-editor of two felicitation
volumes, an Index volume and a two volume translation - all work
connected in some way with the Pali Text Society.
Hiroko Kawanami
Hiroko Kawanami completed her undergraduate studies at Sophia
University in Tokyo and.received her Ph.D. (I99I) in social anthropology
from London School of Economics. She has done extensive fieldwork on
Buddhist practices in Burma, Thailand and Chittagong (Bangladesh).
During her first fieldwork in Burma, she stayed as a Buddhist nun for 16
months. Her interest focus on gender and Buddhism and new Buddhist
movements in Southeast Asia. She is currently lecturer at the Department
of Religious Studies at Lancaster University.
Sue Hamilton
Sue Hamilton is lecturer in Indian Religions at King's College London.
Having. taken an M. Phil in Classical Indian Religion at Oxford
University, she went on to do a D.Phil in early Buddhism. Her thesis was
published in 1996 by Luzac Oriental entitled Identity and Experience: the
Constitution qf the Human Being according to Earfy Buddhism. She is currently
working on a book about the philosophical implications of the use of
metaphor in the Buddha's teachings. She is Hon. Secretary of the Pall
Text Society.

Indian Insights - Buddhism, Brahmanism and Bhakti

Uploaded by

Copyright:

Available Formats

Indian Insights - Buddhism, Brahmanism and Bhakti

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Indian Insights - Buddhism, Brahmanism and Bhakti

Uploaded by

Copyright:

Available Formats

INDIAN INSIGHTS:

BUDDHISM, BRAHMANISM AND BHAKTI

Papers from the Annual Spalding Symposium on Indian Religions

Luzac Oriental I997

All rights reserved: no part oj this publication may be reproduced, stored in a

British Library Cataloguing-in-Publication Data.

Printed and bound in Great Britain by

The Vitalistic Antecedents of the

The Oral Composition and Transmission

Sally Mellick Cutler

Still Suffering Mter All These Aeons:

The Nature of the Repetition

Vi~nu and the Dance of Siva

Indian Insights: Buddhism, Brahmanism and Bhakti

Caskets of Treasures and Visions of Buddhas

Indian Insights: Buddhism, Brahmanism and Bhakti

essays demonstrate both the perennial interest the ancient material

The Paths of the Dead and the Five Fires

Dermot Killingley, University of Newcastle upon Tyne

Indian Insights: Buddhism, Brahmanism and Bhakti

Dispersal is not thought of as destruction here; it is prayed for as something

The Paths qf the Dead and the Five Fires

The dialogue ofYajfiavalkya and Artabhaga (BrhUp 3, 2, I3), which is well

Indian Insights: Buddhism, Brahmanism and Bhakti

and ChUp 5, 3. A similar story occurs as KauUp I, I; here, however, the

The Patizs ifthe Dead and the Five Fires

Despite his father's teaching, he is unable to answer five questions put to

The Five Fires pericope

Indian Insights: Buddhism, Brahmanism and Bhakti

man (BrhUp)/embryo (ChUp)

The Paths of the Dead and the Five Fires

The Two Paths

The Two Paths pericope is structured round the well-known opposition

Indian Insights: Buddhism, Brahmanism and Bhakti

inadequate. Upasana is a mental operation in which something is identified

bright fortnight l9 uttarayal).a20 c1eva-Ioka 21 sun

woman - Kal).va text only)

The Paths of the Dead and the Five Fires

The differences are not really significant; both versions agree in

Parallels to the Five Fires pericope

Indian Insights: Buddhism, Brahmanism and Bhakti

after advisin.g his companions not to engage in debate withJanaka in case

The Paths of the Dead and the Five Fires

Indian Insights: Buddhism, Brahmanism and Bhakti

The Paths qf the Dead and the Five Fires

space wind fire

water earth:m plant' fc.lOd

Parallels to the Two Paths pericope

Indian Insights: Buddhism, Brahmanism and Bhakti

hright fortnight day

The Paths of the Dead and the Five Fires

day bright fortnight northward path year

moon lightning Brahman

Indian Insights: Buddhism, Brahmanism and Bhakti

The Paths if the Dead and the Five Fires

Reat, oj,. cit., p. 164.

(HiidoxtNtLrk, eh. 2); the l<:xl is translated hy lVI. I'doj(: in Yoyotte

Indian Insights: Buddhism, Brahmanism and Bhakti

can be <tpplied to Upan~l'adic texts.

Gonda, 1965, pp. I~22-5'