Introduction To Indian Philosophy Christopher Bartley

An Introduction to
Indian Philosophy
Also available from Continuum:
Phenomenology, Michael Lewis and Tanja Staehler

The Philosophy of History, Mark Day
The Philosophy of Mind, Dale Jacquette
The Philosophy of the Social Sciences, Robert C. Bishop
Forthcoming:
Classical Chinese Philosophy, Im Manyul

Critical Thinking, Robert Arp and Jamie Carlin Watson
Ethics, Robin Attfield
Metaphysics, Jonathan Tallant
Philosophy of Language, Chris Daly
Philosophy of Law, Jeffrey Brand-Ballard
Philosophy of Science, Emma Tobin
Pragmatism, John Capps
An Introduction
to Indian
Philosophy
Christopher Bartley
Continuum International Publishing Group
The Tower Building 80 Maiden Lane
11 York Road Suite 704
London SE1 7NX New York, NY 10038
www.continuumbooks.com
© Christopher Bartley, 2011
All rights reserved. No part of this publication may be reproduced or transmitted in

any form or by any means, electronic or mechanical, including photocopying,
recording, or any information storage or retrieval system, without prior permission
in writing from the publishers.
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
ISBN: HB: 978-1-8470-6448-6

PB: 978-1-8470-6449-3
Library of Congress Cataloging-in-Publication Data

Bartley, C. J.
An introduction to Indian philosophy / Christopher Bartley.
p. cm.
ISBN 978-1-84706-448-6 – ISBN 978-1-84706-449-3
1. Philosophy, Indic. I. Title.
B131.B327 2010
181'.4–dc22
2010012597
Typeset by Newgen Imaging Systems Pvt Ltd, Chennai, India

Printed and bound in India by Replika Press Pvt Ltd
For my wife Loretta, with love
This page intentionally left blank
Contents
Introduction: Some Types of Indian Religiosity 1
1 Foundations of Brahminism: Vedas and Upanis.ads 7
2 Buddhist Origins 13
3 Abhidharma Buddhism 26
4 Sautrāntika Buddhism 35
5 Nāgārjuna and Madhyamaka Buddhism 56
6 ‘Mind-Only’: Yogācāra Buddhism 68
7 Sām.khya and Yoga 82
8 Nyāya and Vaiśes.ika 90
9 The Mı̄mām.sā Vision 117
10 Vedānta 134
11 Advaita-Vedānta 138
12 Viśis.t.ādvaita-Vedānta 168
viii Contents
13 Dvaita-Vedānta and Madhva 184
14 Tantra: Some Śaiva Philosophies of Kashmir 193
Bibliography 235
Index 241
Introduction: Some Types of
Indian Religiosity
This book attempts an overview of some of the topics, themes and arguments
with which Brahminical Hindu and Buddhist Indian philosophers were
concerned between the second and twelfth centuries A.D. It seeks to describe
a variety of very different world-views. It aims to explore a variety of different
mentalities, rather than to evaluate them or to ask whether they are true. It
begins with some general considerations about the background to the different
philosophical schools and tries to explain the origins of the fundamental
separation of mentalities into the enduring-substance ontologies propounded
by Brahmins and the event ontologies, repudiating real permanent identities
formulated by the Buddhists. In broad terms, we see a dialectic between the
two mentalities: one asserting the primacy of Being and the attendant concepts
of substance, universal property, essence and the individuality of entities with
persisting identities, and the other that understands the world primarily as
consisting of ephemeral beings in a temporal process of becoming: events,
complexes, change. According to the first view, states of affairs are produced
by interactions between stable continuing entities, including enduring selves.
On the second, the world is an ever-changing flow of events, and what we treat
as individual entities are convenient abstractions out of relational complexes.
We ourselves are no less conditioned than the things in the world with which
we are involved. The first outlook has it that one fundamentally is in some
sense a soul or substantial self, a further fact over and above one’s experiences.
In short, there is a difference between you and your life. This is precisely what
is repudiated by the second outlook, according to which there are just life
histories. There is no permanent ‘real me’. Terms like ‘Self ’ are convenient
abbreviations for the ways in which embodied persons function in the world.
The different ramifications of this mentality are explored in the chapters about
Buddhism.
2 An Introduction to Indian Philosophy
I am not apologizing when I say that this book is a survey of some of the
Indian traditions. The field is vast and there is so much to be explored. It is to
be hoped that readers are stimulated to consult more specialized works and
read the original sources so as to form their own views. Suggestions about
further reading are to be found at the end of each chapter.
***
In classical India, philosophy was understood as contributing to human
well-being by freeing people from misconceptions about themselves and the
world. Ultimate well-being was conceived as some sort of fulfilment outside
the conditions of space and time. Philosophies, as well as religious traditions,
understood themselves as paths to that final goal.
Where the religious contexts of those who engaged in critical, reflective and
argumentative philosophy are concerned we have to reckon with a tremendous
variety of beliefs and practices. Neither ‘Hinduism’ nor ‘Buddhism’ are really
homogenous. It is difficult to know where to begin: you can always go back
further. In the course of the second millennium B.C. the Aryan migrations
into north west of the sub-continent introduced the Vedic religious culture
and the four-fold hierarchy of varn.as (Brahmins, Warriors, Farmers and
Servants) that was superimposed on the indigenous system of jātis. It appears
that originally the ritual cult was concerned with the propitiation by offerings
of the many deities in the Vedic pantheon. Their favour thus secured would
yield mundane and supra-mundane rewards. Rituals performed by members
of the Brahmin caste were understood as yielding benefits for both the
individual and the community. But there developed an outlook that the con-
tinuation of the cosmos, the regularity of the seasons and the rising of the sun,
were not merely marked or celebrated by ritual acts but actually depended
upon ritual. What the rituals effected was too important to be left to the
choices of ultimately uncontrollable capricious divinities. So rituals came to
be thought of as automatic mechanisms, in the course of which the mention of
the deities’ names was but a formulaic aspect of the process. The relegated
gods existed only in name. The Brahmins unilaterally declare themselves the
gods in human form. From the point of view of the individual, the benefit of
the ritual was understood in terms of the accumulation of merit or good karma
that would be enjoyed at some point in the future, in this or a subsequent life
perhaps in a superior sphere of experience for those with sufficient merit.
The notion of karma is basically a recognition that one’s actions have future
consequences for one. But the consequentiality extends beyond this life. The
Introduction 3
idea is that a deliberately performed intentional action generates a residue
(good or bad) that remains with the agent until future circumstances occur
that are appropriate for its fruition in the agent’s experience. Karma does not
determine the future directly: rather, it lies in wait. It is karma that personalizes
and propels individuals through a series of births in the here and now.
The notion of repeated births (sam.sāra) is a presupposition shared by
Brahmins and Buddhists. Everyone agreed that the process of rebirth goes
on and on, is fundamentally unsatisfactory and is to be escaped from.
Whatever the quantum of good karma accumulated by an individual, it will
still become exhausted. Felicity is always temporary. Release or liberation is
always understood as irreversible freedom from rebirth. This freedom from
rebirth, the ultimate goal of religious praxis, is what is called by Brahmins
‘moks.a’ or ‘mukti’, and ‘nirvān.a’ by Buddhists, although as we shall see their
understanding of what it means is very different.
For there to be rebirth and the anticipation of future benefits, Brahmins
regarded it as essential that there be a permanent and stable identity (ātman)
to which the karma pertains, so that the instigator of the performance could
be he who enjoyed its consequences. The status of this ‘self ’ (ātman) in the
natural hierarchy of being was maintained, and hopefully improved, by the
spiritual purity of the persona with which it was associated. The system of
castes, whether the endogamous and commensal jātis that have a monopoly on
specific trades and professions or their interpretation in the varn.a framework,
is a hierarchy determined by spiritual purity. The hierarchy accords with and
expresses the cosmic order that is both natural and right (dharma). Each caste
has its own set of duties (dharma). The Brahmins insisted that it was better to
do one’s own dharma badly than that of another well. Dharma is not thought
of as a universal morality applicable and accessible to all. Rather it is a matter
of what F. H. Bradley called ‘my station and its duties’. The Brahmins’ purity
derived from obedience to the rules bearing upon every aspect of life that are
encoded in texts prescribing social and religious duties (the Dharma-Śāstras).
The orthodox Brahmin cannot choose his own values. The rules chart a safe
passage through a universe populated by dangerous forces that are looking for
an opportunity to occupy the body and mind of those who are negligent of their
observance. Spiritual purity is purchased at the price of moral heteronomy.
Daily ritual, as well as personal and social duties (dharmas), confers meaning
on the life of the orthodox ‘twice-born’ Hindu. It is clear that the mainstream
Vedic orthodoxy perpetuated by lineages of what are known as ‘smārta’
(traditionalist) Brahmins is more than a matter of personal or shared religious
allegiance. For much of its history India has been politically fragmented
into hundred of small kingdoms. In the absence of a centralized monarchical
institution, the integrating factor promoting harmonious coexistence, agreed
expectations, shared values and trust between different states throughout the
sub-continent was the Vedic religio with its established rituals, social norms
and Sanskrit language. Hence the anxiety occasioned by departures from the
common identity conferred by Vedism, which were potentially subversive of
good order.
Some aspects of the philosophical articulations and defences of this
mainstream orthodoxy (smārta), especially against the manifold Buddhist
articulations of their basic insights that there is no genuine permanence and
no persisting self, are described in the chapters on Mīmām.sā ritualism and the
Nyāya-Vaiśes.ika realistic metaphysical pluralism.
As well as ritual practice, there is the discipline of yoga, whose philosophy is
described in the chapter on Sam.khya-Yoga. Yoga aims at calming the mind so
that it may become fit for a non-discursive awareness of reality as it really is. This
begins with control of the body and develops mental discipline through medita-
tion, with a view to freedom from determination by natural causality. Obviously
such practice is consistent with participation in ritual acts, but it may be detached
from that form of religiosity in the case of those who have become convinced
that ritual practice is ultimately ineffective as a means to final salvation, under-
stood as freedom from repeated births in different spheres of experience.
The model of the human psyche is that what is called ‘mind’ (manas)
co-ordinates information received via the sense-faculties. For the most part
we are passive in relation to sensation and the feelings it evokes. In addition,
mental attention is often diffuse and not really focused. When mind and senses
are not controlled, we are living purely on the level of sensation. The ideal is
to discipline the senses by bringing the manas under the control of what is
called buddhi (usually not very helpfully translated as ‘the intellect’) and to
focus attention on one’s inner identity (ātman). Such a person is called ‘yukta’
or integrated:
When a person lacks understanding and his mind is out of control, he is subject
to the senses that are like bad horses of a charioteer. But when a person has
understanding and his mind is under control, his senses are subdued like good
horses. [Kat.ha Upanis.ad 3.5–6]
When the five sensory cognitions and the mind are stilled and reason (buddhi) is
steady, they call this the highest state. They call yoga a firm grip on the senses.
One is then free from distractions. [ibid. 6.10–11]
Introduction 5
He should sit still as a piece of wood. Collecting his sense-faculties, he should
focus his mind steeped in meditation. He should not hear with his eyes nor see
with his eyes. He should not long for the objects of the sense-faculties. He should
focus the mind, within for it wanders in and out of the five doors and has no
stable foundation. [Mahābhārata 12.195.5f.]
A quite different outlook says that understanding or insight alone is necessary

for salvation. This is the view of those individuals (including Gautama the
Buddha) who have chosen the path of renunciation (sam.nyāsa) of the every-
day social relations life of the adult male householder with his duties of wealth
creation, procreation and the performance and patronage of rituals. Renounc-
ers have come to the conclusion that the interminable performance of rituals
cannot conduce to ultimate well-being. They think that rituals and yoga may
be helpful preliminaries, but they are insufficient in themselves for freedom
from rebirth, which can only be approached through concentrated meditation
leading to insight into one’s true identity. Sam.nyāsa is formalized through a
special type of initiation ritual for the removal of inherited caste status. Ini-
tially a trend subversive of social order, renunciation became integrated into
the mainstream religion and was classified as the fourth stage of life (āśrama)
subsequent to those of celibate studentship, being a householder and retire-
ment or ‘forest-dwelling’. Renouncers are homeless wanderers committed to
celibacy. They possess only a saffron robe, begging bowl and staff. They are
dedicated to achieving liberation from rebirth through insight into the nature
of a soul that is a reality beyond space and time, and the ultimate source of all
reality (the Brahman).
Nevertheless, renunciation has always been controversial and for some
people renouncers are objects of fear and suspicion. There is a resistance to
the idea that anyone should quit the established social order and the rituals
commanded by the Vedas, and go it alone. So some traditions hold that
performance of social and ritual duty is mandatory: what one should renounce
is desire for the results of the rituals. That outlook is clearly expressed in the
Bhagavad Gītā.
The philosophical expression of the renunciatory path is to found in the
chapter on Advaita-Vedānta.
A quite different mentality has it that whole-hearted and deeply felt
devotion (bhakti) to a personally conceived deity elicits divine favour (prasāda)
and ultimate well-being in heaven. The self is understood as an essentially
dependent servant of God whose fulfilment is to be found in enjoyment of
the divinity. This is not a late development, but is apparent as early as the
Śvetāśvatara Upanis.ad: The Lord supports this whole world consisting of

the perishable and the imperishable, the manifest and the unmanifest that
are linked together. The soul who is not the Lord is bound because he is the
experiencer of the fruits of karma. Having known God, he is released from
all bonds. [1.8]. The one God rules over the perishable and the soul. By
contemplating him, by uniting with him and in the end sharing his way of
being, all the material world ceases. When one has known God, the bonds are
destroyed, when the afflictions (kleśa) have dwindled away, there is an end to
birth and death [1.10–11]. The articulation of the devotional outlook is the
subject of the chapters about Viśis.t.ādvaita Vedānta and Dvaita Vedānta.
As well as what is generally recognized as mainstream orthodox smārta
Brahminism, there are traditions that base themselves not on the Vedic
corpus but on divinely revealed scriptures called Tantras or Āgamas. These
are monotheistic ritual cults, whose praxis includes forms of yoga and
meditation. The deities are Śiva, the Goddess and Vis.n.u, who do not feature
prominently in the Vedic pantheon. Tantrism is the subject of the chapter
about the Śaiva and Śākta cults.
Further reading
A. L. Basham, The Wonder that was India: A Survey of the Culture of the Indian Sub-Continent Before
the Coming of the Muslims, is an interesting and comprehensive overview.
Alexis Sanderson, ‘Power and Purity’, (1985) is seminal, as are some of the articles collected in Halbfass,
Tradition and Reflection, (1991). For a translation of the Bhagavad Gītā, see Johnson (1994). For
the Upanis.ads, Olivelle (1998) provides an informative introduction, texts, translations and
illuminating annotations. What is called the Moks.a-dharma (composed while Buddhism was
developing) in the Mahābhārata Book XII offers an insight into the spiritual outlook of the
Brahminical renouncers. There is a text and translation in Wynne (2009).
Gavin Flood, Introduction to Hinduism, is easily the best of the countless introductions.
Louis Dumont (1980), ‘World Renunciation in Indian Religions’ has been very influential. Dumont
argues that the key to understanding Indian religion is to be found in the dialogue between the
householder and the renouncer. The latter, he says, represents the closest Indian equivalent to the
European notion of the individual as the bearer of values. The monograph, Homo Hierarchicus, is
a fundamental contribution to Indian sociology. Olivelle (1993) is about the four stages of life.
Olivelle (1996) translates texts bearing on Renunciation in Hinduism.
On devotional religiosity see Hardy (1983), which is unlikely to be superseded. J. A. B. van Buitenen’s
‘On the Archaism in the Bhāgavata Purān.a’ (Chapter XIX of Rocher (1988) applies M. N. Srinivas’
theory about the process of ‘Sanskritisation’ to bhakti religion.
Foundations of Brahminism:
Vedas and Upanis.ads 1
Chapter Outline
Further reading 12
Questions for discussion and investigation 12
There are four collections known as Vedas, composed during the period of the
Aryan migrations into northern India (1990–1100 B.C.). They are called the
Rig-Veda, the Sāma-Veda, Yajur-Veda and the Atharva-Veda and are regarded
by mainstream orthodox Hindus as an authorless, timeless and infallible
source of knowledge about religious and social duties in harmony with the
natural universal order (dharma). Their eternal sound-units are held to have
been discerned and composed by seven primordial ‘seers’. Priestly Brahmin
families preserve the different traditions of recitation of the sacred sounds.
The basic component of each of the four Vedas is its collection (sam.hitā) of
verses (mantra), evocative of the divinities whose sonic forms they are, which
are used in rituals. Attached to each collection are the ‘Brāhman.as’ which
prescribe, describe and elucidate the purposes of the sacrificial rituals
performed by Brahmin priests and their householder patrons. They include
explanations of the meanings of the ritual actions, and posit correspondences
between aspects of the rites and features of the cosmos including the social
structure and the human body. It was believed that ritual performance orders,
sustains and perpetuates the universe, creating new time and ensuring the
regular succession of the seasons. There are also the ‘Āran.yakas’ (‘Forest
Books’) that speculate about the ‘inner’ meaning of the rituals and are closely
associated with what are called Upanis.ads. They were composed by people
who had probably renounced the actual performance of ritual. Their view is
that the mental re-enactment of the meaning of the ritual is just as effective as
its overt performance. These fascinating compositions contain many reflective,
poetic explorations of fundamental metaphysical questions. The best way to
appreciate them is by immersing oneself in the originals, translations of which
are readily accessible.
Metaphysical questions are to be found in the hymns of the Rig Veda. These
hymns celebrate and propitiate gods mostly associated with natural forces.
Their cosmogonic hymns speak of the gods such as Indra and Varun.a
establishing inhabitable space and creating sustainable order out of chaos.
They were used by Brahmins in rituals whose aim was the perpetuation of the
cosmos. The questions about Being – Why is there something rather than
nothing? Where did it all come from? – feature there, if not prominently. One
line of thought says that there is an original One, beyond being and non-being.
Another idea is that Being (sat) arises from non-Being (asat). Since nothing
comes from nothing, ‘non-Being’ probably means a chaotic, undifferentiated
state in which are no things, no names and forms, no entities and kinds, no
structures or organizing principles. ‘Being’ would then be a cosmos of differ-
entiated, identifiable, organized realities. This is apparent from the somewhat
later Chāngogya Upanis.ad 3.19.1: ‘In the beginning this world was just non-
being. What now exists came from that. It developed and formed an egg . . .’
The speculation is rationalized in the sixth chapter (6.2.1) of the same work:
‘In the beginning, my dear, this world was just being (sat), one only without
a second. Some people say, In the beginning, this world was just non-being,
one only, without a second. From that non-being, being was produced.
But how could this be? How could Being be produced from non-Being?
On the contrary, in the beginning this world was just Being, one only, without
a second’.
In this intellectual milieu, the Being of beings is understood as the source,
basis, support and final cause what there is. It is the foundational ‘something
else’ out of which the world of entities emerges or unfolds. According to
this outlook, the cosmos has immanent order and purpose (dharma)
independently of any meanings that human beings might create for them-
selves. In the major Upanis.ads (composed between 800 and 400 B.C.) this
ontic support is called the Brahman – the Absolute, timeless, unconditioned
unlimited substance that needs nothing else in order to exist. Br.hadāran.yaka
Foundations of Brahminism 9
Upanis.ad (BAU) 1.4.10 says,
‘In the beginning there was only the Brahman, and it knew only itself (ātmānam),
thinking, “I am the Brahman”. From that, everything came into being.’
It is the imperishable principle (aks.ara), the thread upon which all realities are
woven. [BAU 3.8]
The Brahman lacks nothing, which is why it is sometimes said to be blissful.

The Brahman is the essence of whatever is real. The Brahman is the truth. This
Reality behind and underlying all contingent and finite realities is sometimes
characterized as the One, beyond the categories of being and non-being that
apply to what falls within the sphere of our experience.
Naturally, there was also speculation about what human beings basically are.
A possible view is that we are just parts of the physical cosmos, assemblages of
the same elements that make up the waters, wind, fire and earth. The original
meaning of the word ‘ātman’ (usually translated as ‘self ’ or ‘soul’ – but which
also often means the identity or essential nature of something) is ‘breath’
(prān.a), a word for the vital functions of seeing, hearing, speaking and
thinking. But there developed the view that we are not reducible to physical
factors and functions. The principle of reasoning is that if something can treat
a process as an object to itself, it cannot be something that emerges from or is
an aspect of that process. In this light, ātman is no longer the breath, or any
one vital function or the vital functions taken collectively, but that which is the
source of vital functions, that which underlies them and explains their reason
for being.
Soul’s secret name is, ‘the reality of what’s real’ for the real constitutes the vital
breaths and the soul is their essence. [BAU 2.1.2]
Ātman acquires a meaning approximating to ‘essential principle’ or ‘soul’. The

concept is that of an essence and is different from than what we ordinarily
mean by self (so we have to be careful about translating the term) or person,
that is to say, a subject of experiences or a thinking, willing, feeling, acting
individual. Crucially, soul becomes regarded as intimately associated with or
participating in the fundamental reality:
This soul is the honey of all beings, and all beings are the honey of this soul. The
radiant and immortal person in the soul and the radiant and immortal person who
is an individual self, they are both the soul. It is immortal. It is the Brahman. It is
the whole. This soul is the lord and king of all beings. As all the spokes are fas-
tened to the hub and rim of a wheel, so to the soul are fastened all beings, all the
gods, all the worlds, all the vital breaths, and all these people. [BAU 2.5.14–5]
The cosmos is thought of as a single whole that has essence and this is what is
called the Brahman. Individual selves, microcosms versions of the cosmos,
too have essence and this is what is called ātman. If essence is indivisible, the
Brahman equates to ātman. The Brahman and ātman come to be understood
as two sides of the same coin.
Another portion of the Br.hadāran.yaka Upanis.ad puts it like this:
Explain to me the Brahman (Reality) that is plain and not hidden, the ātman that
is within all. The soul (ātman) within all is this soul of yours.
What is the soul within all?
You cannot see the seer who sees. You cannot hear the hearer who hears. You
cannot think of the thinker who thinks. You cannot perceive the perceiver who
perceives objects. The soul within all is this soul of yours. . . . That is what is
beyond hunger and thirst, beyond sorrow and delusion, beyond old age and
death. When they know this soul, Brahmins cease to desire sons, wealth, other
worlds of experience and adopt the mendicant life. [BAU 3.4–5]
Shortly afterwards there is the question of on what the cosmos depends, the
thread running through everything, the inner controller (antar-yāmin) of this
world and the next, as well as all beings, who controls them from within. It is
said that when a person knows what the inner controller is, he understands
reality (brahman), he knows the worlds, he knows the gods, he knows the
Vedas, he knows the spirits, he knows the soul, he understands the whole.
This soul (ātman) of yours, who is present within but is different from the
earth [fire air, wind & other physical features], whom the earth does not know,
whose body the earth is, and who controls the earth from within – he is the inner
controller, the immortal one. . . . This soul of yours who is present within but
is different from all beings, whom all beings do not know, whose body is all
beings, and who controls all beings from within – he is the inner controller, the
immortal. . . . This soul of yours who is present within the breath but is different
from the breath [from speech, from sight & other sensory and cognitive functions],
whom the breath does not know, whose body is the breath, and who controls
the breath from within – he is the inner controller, the immortal. . . . That is
the seer who cannot be seen, the hearer who cannot be heard, the thinker
who cannot be thought of, the perceiver who cannot be perceived. There is
Foundations of Brahminism 11
no other who see etc. It is this soul of yours who is the inner controller, the
immortal. [BAU 3.7]
About this soul one cannot say anything. It is incomprehensible for it cannot be
grasped. [BAU 4.2.4]
This soul is the Brahman – this soul that constitutes perception, mind, breath,
sight, hearing, earth, water, wind, space, dharma and adharma – this soul that
constitutes everything. [BAU 4.4.5]
The breathing behind breathing, the sight behind sight, the hearing behind
hearing, the thinking behind thinking – those who know this perceive the
brahman, the first and the last. [BAU 4.4.18, cf Kena Up. 1.2]
It is the soul that should be seen, heard about, reflected upon and contemplated.
When the soul is seen, heard about, reflected upon and contemplated, the
cosmos becomes understood. [BAU 4.5.6]
It is recognized that there is a sense in which a person’s desires and actions

make him what he is. The person who acts in accordance with desires is caught
up in a chain of actions and consequences, and is repeatedly born in different
spheres of experience. But as for the person whose life is not determined by
desires and their satisfaction, because he lives for what really matters:
The Brahman he is and to the Brahman he goes. [BAU 4.4.6]
Such a one is at peace, in control, unperturbed, patient and focussed for he sees
the soul in himself and he sees all things as the soul. [BAU 4.4.23]
Finally, some passages from the sixth chapter of the Chāndogya Upanis.ad that
will exercise a considerable influence on later thought, especially on the
Vedāntic systematizations of the Upanis.adic teachings. Uddālaka Ārun.i offers
to teach his son Śvetaketu how one hears what has not been heard before, how
one thinks what has not been thought of and how one understands what has
not been understood. He says that the teaching is:
From one lump of clay one understands the nature of anything that is made of
clay – the modification being a name, a taking hold by speech, while the truth is
that it just clay. (Ch.Up.6.1.4)
In the beginning there was just Being (sat), one without a second. . . . Being
reflected, ‘May I become many. Let me become productive.’ It generated heat.
Heat generated water. Water generated food. (Ch.Up. 6.2.1 and 3)
That divinity thought, ‘Let me diversify names and forms by entering those three
creative principles by means of the living self, and make each of them threefold’.
(Ch.Up.6.3.2)
Being is the source of all these creatures. Being is their support and Being is their
foundation. (Ch.Up.6.8.4)
Bring me a fruit from the banyan tree.

Here it is, Sir.
Cut it up.
I’ve cut it, Sir.
What do you see?
Tiny pieces.
Cut one of them up.
What do you see?
Nothing.
You cannot see the finest essence here, but it is because of that finest essence that
this great Banyan tree stands.
Whatever is the finest essence, that is the identity of everything, that is authentic
reality. That is the soul.
That is what (tat) you (tvam) are (asi)’. (Ch.Up.6.12)
Further reading
For Vedic Hinduism and the contents of the Vedic corpus see Jamison and Witzel (2003). This is
accessible online via Professor Witzel’s website.
Rig Vedic hymns are translated in Doniger (2005).
The Upanis.ads can be found in Olivelle (1998). There is also an OUP paperback that just has the
translation.
Essays II, V–X in van Buitenen (1988) have all been influential. Chapters I and II of Halbfass (1992) are
thought provoking about the ‘question of being’ in India.
Questions for discussion and

investigation
1. Does it really make sense to suppose that the complex cosmos has a single source?
Is ‘being’ uniform?
2. Why might people find ritual pursuits unsatisfactory?
Buddhist Origins 2
Chapter Outline
No self 20
Further reading 24
Gautama Śākyamuni, who would become known as the Buddha, ‘the Enlight-
ened One’ probably lived in the period 450–400 B.C. Early Buddhism rejects
the notions of the Brahman and ātman, insisting that there are no enduring,
substantial realities. It teaches the essential temporality of beings. Things are
always changing. Nothing really lasts. There are no essences, no immutable
natures and universals, characterizing entities and running through reality as
a whole. It says that ritual religion is pointless and rejects the authority of the
Vedic scriptures. The attitude that self-advancement through ritual practices
is possible is a form of what the Buddha calls ‘greed’. It says that the caste
hierarchy, held by the Brahmins to be a natural fact, and the associated
deontological morality is nothing more than a system of social arrangements.
The Buddhist outlook denies that the cosmos has a single source and goal. If it
has a reason for being and an explanation for its dispositions, it is just so that
sentient beings may experience the fruits of their karmas.
Buddhists and Brahmins recognize that there are persons who remain
similar over time and whose futures are conditioned by their deliberate and
intentional decisions and actions. They differ about whether a basic principle
of identity is required to explain the continuity of such individuals. We might
depict the Buddhist as arguing that psychological continuity is enough for
personal identity. The Brahmin position is that psychological continuity

presupposes personal identity and therefore cannot constitute it. (Neither
seriously entertains a physical account – the body dies and is burned.) But
the disagreement goes deeper than questions about personal identity. It occurs
in the context of much broader opposition between two mentalities. The
Buddhists deny that there is anything that basically remains the same, that
there are any permanent stable identities at any level (nairātmya-vāda). There
are no real universals or kinds, no essences constituting enduring objects, no
eternal sounds or meanings. There are only processes of momentary events
(ks.an.ika-vāda). The Brahminical metaphysic posits Being as prior to and
underlying the sphere of beings and becoming. The Buddhists say that there
are just essentially temporal beings in an ultimately impersonal flux.
Buddhist philosophers adhere to an ontology of processes and events
(nairātmya – non-substantiality), rather than one in which enduring sub-
stances are the ultimate constituents of the worlds. They typically reduce
whatever is conventionally considered as a stable substance to a stream or
sequence of occurrences. The Buddhist claims that thoughts purporting to
be about persisting entities that undergo qualitative changes without loss
of identity can be reductively analysed into descriptions couched in terms of
sequences of instantaneous or momentary events. They reject any categorical
distinction between enduring substrates and the properties, essential and
non-essential, of which they are the bearers. Rather, they understand reality
in terms of temporary collocations of basic factors (dharma). Individual
personality is understood as a causally regulated flux of mental and physical
occurrences. What makes for the continuity of a person’s life is just the
occurrence of a series of suitably interconnected physical and mental events.
What is called ‘self ’ is not a substantial persisting entity distinct from the
contents of experiences but a construction, convenient but mistaken, out
of those contents. The Buddhist maintains that the human subject is an
essentially temporal (ks.an.ika) succession of phases that somehow imputes
permanence to itself.
So when the Buddhists say that there is no self or soul, they are doing
more than denying that there is a permanent and stable inner principle in
human beings, selfish obsession with which is the source of all our woe. The
anti-substantialist mentality denies that reality is constituted by stable
and reliable structures grounded in a single source. It denies that either the
totality or individual human lives have any intelligible purpose or meaning
outside themselves. There is no God who has a purpose for the world and for
Buddhist Origins 15
human lives. This may sound negative, but that was not the intention. Rather,
understanding life in these terms is held to be liberating. Much that one
had worried about does not really matter. One will not live forever. Perhaps it
enables people to ‘let go’ and enjoy life while they can. Above all it is meant to
empty our heart of hatreds, greed and bitterness and to engender a spirit of
compassion, friendliness and generosity.
Gautama Śākyamuni, who became known as the Buddha or Enlightened
One, lived around 450–400 B.C. in the far north east of India. Born into a
royal family, he like many others at the time became disillusioned with the
frustrations, superficialities and conventional expectations of normal social
life. Renouncing society and the ritual religion with its eye on future benefits,
he went from home to homelessness. He tried to live as an ascetic, practising
severe austerities, in the manner of contemporaneous Jaina renouncers, with a
view to acquiring spiritual insight. Finding that this did not work, he espoused
the ‘middle way’ – a path of morality and meditation, between comfort and
asceticism. The doctrine is also understood as the middle way between
eternalism and nihilism. The former says that identity is permanent. The latter
holds that universal impermanence and non-substantiality preclude ethical
consequentiality. The Buddhists thought that eternalism was amoral in that it
treats the soul (ātman) as outside the sphere of values. Moreover, the soul is
not what we normally mean by a person. Nihilism just denies that there are
bearers of moral responsibility and recipients of consequences. Buddhists
think that a morally significant stream of information and dispositions
(sam.skāras) continues after the death of the person. The stream’s future is not
the future of that very person. Still, we should care about its future and strive
for its ameliorization in the interests of general well-being.
The Buddha discovered the truths (dhamma) about the cosmos. The
Buddha’s teachings and spiritual journey are reported in the collections of
discourses found in the Sutta-pit.aka of the Pāli Canon of scriptures. After
centuries of development by oral transmission, the Pāli Canon was committed
to writing in Ceylon during the reign of King Vat.t.agāmanī (97–77 B.C) at the
Council of Ālokavihāra. These are the scriptures recognized as authoritative
by all Buddhists, whether belonging to the Theravāda traditions found in
Ceylon, Burma, Thailand and Cambodia, or to the Mahāyāna ones belonging
to parts of India (until the twelfth century A.D.), Tibet, China and Japan. Of
the differences between the two families (many of which are cultural) one
might mention that the Mahāyānists ascribe quasi-divine status to the many
Buddhas and accept texts additional to the Pāli Canon as authorities. It is
sometimes argued that some of the alleged doctrinal differences – e.g. the
Mahāyānists emphasize compassion and insight, worship the Buddhas and
hold that the path to enlightenment is long and difficult – are not really
differences at all but belong to the shared heritage.
The basic principles of the Buddha’s teaching are expressed in ‘Four Noble
Truths’:
1. All mental and physical phenomena are ultimately unsatisfactory (duh.kha/

dukkha). This is the eternal round of existences (sam
. sāra). Universal unsatisfac-
toriness is attributed to the impermanence and non-substantiality of all
conditioned entities, including human lives. In truth, there are no enduring
essences or identities. Selfhood is a fiction that the fluxes of thoughts and
desires conventionally called persons superimpose upon themselves. Attempts
to cling to what are in reality impermanent objects, states of affairs and
relationships are bound to end in suffering.
2. The second truth is that there is a causal explanation for the arising of
unsatisfactoriness in terms of ‘thirst’, a metaphor for desire and attachment,
and ignorance of the way things really. Unenlightened actions are motivated
by greed, hatred and delusion. Some fundamental causal relations are
formulated in the doctrine of interdependent origination (pratı̄tya-samutpāda).
This is basically a Humean view, holding that causation is just a matter of
Bs usually following As. There are no causal powers separate from what
actually happens in succession. Causal continuity within subjectless streams
of experiences does not require an internal principle but is organised by the
co-operation of twelve factors forming a circle rather than a chain, sometimes
put like this:
Ignorance conditions volitional actions. Volitions condition states of mind.

States of mind condition mental and physical phenomena. Mental and
physical phenomena condition the five senses and the co-ordinating
mental faculty. Those six condition one’s mode of interaction with the
external world. That mode of interaction conditions sensations. Sensations
condition desires. Desires condition attachments to objects. Attachments
condition the process of becoming (a life). A life conditions a new birth.
Birth conditions old age, death and suffering.
This is the arising of the mass of suffering.
This account of causation is held to account for the organisation of a stream

of karmically conditioned experiences so that it can be reborn in another foe-
tus as the start of another life. The idea seems to be that a bundle of selfish
energies coheres as a field of forces. To the question of whether the one who is
born is the same as the one who dies, the tradition replies that it is neither the
same nor different.
Buddhist Origins 17
Pratītya-samutpāda will be interpreted by Nāgārjuna as meaning that
everything comes to be in dependence on causes and conditions. Nothing is
self-sufficient. There are no essential forms reproducing their own kinds and
no causal powers over and above what actually happens.
3. There is potentially an end to suffering, called nirvān.a or nibbana. This is the

end of the series of unsatisfactory existences through the extinction of the fires
of possessiveness, antagonism and delusion that generate rebirth-causing
actions. The fires metaphor is no accident. It is more specific than the idea that
everyone is burning with desires that only generate more experiences and
more desires. It alludes to the three fires which the Brahminical householder
was obliged to keep burning and which symbolised his life, responsibilities
and attachments as a social being. It represents the endless repetitions that
characterise life of the householder. The tradition is reluctant to say anything
positive about nirvān.a as a post-mortem state. Whatever it involves, there are
no persons to experience it. It is perhaps sufficient to say that from the point
of view in terms of this life it is the extinction of the fires of greed, hatred and
delusion and the end of the afflictions (ignorance, selfishness, desire for sensory
gratifications, obsessive attachments and an exalted opinion of oneself).
4. There is a path to the cessation of suffering: the eightfold path of morality and
meditation.
The Four Noble Truths are the key to enlightenment. But hearing is not
enough. They have to be acted upon, and this is where the path of morality and
meditation comes in. The path is said to consist in: right views, right thoughts,
right speech, right action, right livelihood, right effort, right mindfulness and
right concentration. Right views and thoughts are mind-purifying wisdom or
insight into the fleeting and unsatisfactory nature of existence; right speech,
action and livelihood are moral conduct; right effort, mindfulness and concen-
tration are understood as meditation. Buddhists insist that virtue is necessary
for the cultivation of meditation and insight. Intention (cetanā) is understood
as determining the moral quality of an action. Morality consists in deliberate
abstention from murder, theft, sexual misconduct, false speech, slander, harsh
words, frivolous talk, covetousness, malice and false views. Right livelihood
would preclude such occupations as arms-trading, dealing in drugs and alcohol
and butchering animals. Meditative concentration (samādhi) is the achievement
of tranquillity through avoidance of distractions, and by suppression of sensory
activity. Emphasis is placed upon mindfulness or exercising control through
constant self-awareness of one’s physical, mental and emotional states.
The path is the ‘middle way’ between the self-indulgent and ascetic lives,
neither of which lead to release from the desire-fuelled series of existences.
What is needed is the elimination of the basic defects of craving, aversion and
delusion.
The Buddha insisted that it is the intention (cetanā) with which an action is
performed that determines its ethical quality. He tells his followers to cultivate
attitudes of non-violence, honesty, friendliness, gentleness, compassion and
generosity. The right sorts of actions will follow. He did not provide a rulebook
stipulating particular types of ethical actions (although his followers did for
the monks) or an ethical theory. There are some basic principles: don’t
kill, don’t steal, don’t tell lies or indulge in malicious gossip, avoid sexual
misdemeanours and intoxicating substances. These are universal – not caste-
specific – in that they have moral application to everyone: I do not want to
be assaulted and can reasonably conclude that no other sane person wants
to be. There is a trend of thought running through Hinduism that there
may be ways of acting that do not generate karma. Representative here is
the idea, promoted in the Bhagavad Gītā, that since it is deliberately purpo-
sive actions that generate karma, actions done for their own or duty’s sake
without a view to the advantage of the agent will not generate karma. The
Buddha thought that karma was inescapable by sentient beings. The eight-
fold path recognizes this. It is sometimes argued that the outlook is conse-
quentialist, holding that it is overall states of affairs that have moral value.
Some ways of life have better overall consequences than others. The view is
that although the future of the stream that is your life is not your future, you
still have reason to care about its future, as well as that of all other streams. The
Buddhist path aims to promote general happiness. Since there really are
persons, it is not open to the objection that it ignores the separateness of
individuals and their integrity.
Moral conduct attenuates afflictions (kleśa) that prey upon the mind. These
weaknesses are familiar ones: ignorance in the sense of indulging oneself
in self-serving fantasies and believing whatever it suits one to believe; the
sheer selfishness that sees itself as the centre of the world; desire for sensory
gratifications; neurotic obsessions that divert attention from what matters;
an exalted opinion of one’s significance in the scheme of things. The texts
catalogue other impediments to spiritual progress, two of which we shall
mention briefly. There is attachment to rituals, central to the Brahminical way
of life, and vows, the latter being personal undertakings and commitments
such as the harsh austerities that Gautama had found useless. I think that the
word ‘kāma’ often moralistically translated as ‘lust’ or ‘sensual desires’ was
probably more specific and refers to those rituals performed by Brahmins that
Buddhist Origins 19
will produce yield benefits in this world for those patronizing them. The
Buddha thought that all ritual performances were at heart self-interested,
not to say a waste of time and effort.
The texts are rich in descriptions of and prescriptions for meditational
practices, aiming at the control and ultimately the cessation of discursive
mental activities. The Buddha described his path to enlightenment as an ascent
through a hierarchy of four stages of meditations (dhyāna) which, far from
involving ascetic rigours, are pleasant experiences:
Then indeed, having eaten enough, having got my strength back, free from desires,
free from unhelpful matters, I reached the first stage that is accompanied by thought
and reflection, which is produced by discrimination and consists of joyful happiness
and remained there. But this pleasant state did not put my mind at rest.
Stilling thought and refection I reached the second stage that is inner tranquillity,
a focussing of the mind on one point, free from thought and reflection, consisting
of joyful happiness that is born of concentration and remained there. But this
pleasant state . . .
I reached the third stage when I became detached from joy, indifferent to
pleasures and pains, attentively mindful and knew physical pleasures. But this . . .
From abandoning bliss and abandoning pain and thanks to the disappearance of
cheerfulness and depression, I reached the fourth stage that is beyond pleasures
and pains, the quintessence of equanimity and attentiveness. But this . . . [Majjhima
Nikāya 1.247]
This is a typical account of meditation involving successive stages. It did not

put his mind at rest because it stopped short of revealing the fundamental
truth that our experience is unsatisfactory because we fail to realize that there
are really no individual identities and that everything is impermanent.
Meditation also has a positive goal, that is to say, the cultivation of the
helpful (kuśala) states called the Brahma-Vihāras: friendliness, compassion,
joy and equanimity. There is also what is called ‘meditation on the four infini-
ties’ that aims at the suppression of imagination and conceptual thinking.
There is a fourfold hierarchy of stages: the stage of the infinity of space, the
stage of the infinity of perception, the stage of nothingness and the stage
beyond conceptualization and non-conceptualization. Dwelling on notions
of the infinite emphasizes the limits of conceptual thought, and undermines
confidence in the capacity of our minds to grasp the nature of reality. I take
what is called meditation on no-thingness to refer to a state in which one is not
focused on anything in particular: a reflex of the realization that there are no
individual identities.
Finally, morality and meditation are not understood instrumentally, as

means to something different. Their complete realization is what enlighten-
ment means. That is the nirvān.a accomplished by the Buddhas. The person
who successfully follows the Buddhist path is liberated while still alive.
No self
The early tradition raises queries about the coherence of the notion of the soul.
Is it the same as experiences? Is it non-experiential in character? Is it the pos-
sessor of experiences? The first view reduces self to transitory states, but it is
meant to be constant. The second is literally self-defeating. Such an entity could
never have the awareness, ‘I am’. The third view treats experiences as contingent
properties that the self might lack. So in that case the self might sometimes lack
the awareness, ‘I am’ [MahĀnidĀnasuttanta in DIghanikĀya II 67].
As an example of the sort of understanding of selfhood that the Buddha
repudiates, let us look at Chāndogya Upanis.ad 8.7.1:
The soul that is free from evils, wrong, free from old age and death, from sorrow,
from hunger and thirst, whose desires and intentions are ever realised, that is what
is to be discovered, that is what is to be understood. When someone discovers that
and understands it, he obtains all worlds and all his desires are fulfilled.
According to this Upanis.adic outlook, the soul is the key to the meaning of
life and ultimate well-being. The Buddha thought that the mentality to which
such thinking belongs is basically self-interested. He undermines it by saying
that there is no such thing as the soul: all we can say is that there are temporal
streams of experiences and ethical consequentiality. There is no difference
between you and your life – you are the same as your life-history. You do not
go through phases: there is just a succession of phases. It makes no sense to
ask, ‘what if my early upbringing had been different?’ because the question
would be about another stream of experiences.
If what we call the person is the stream of thoughts, who is thinking them?
The Buddhist view is that intrinsically self-aware subjectless thoughts are
thinking themselves, as well as each other. They do not need illumination
by consciousness belonging to a persisting subject. They form a continuous
entity by knowing their immediate predecessors and successors. William James
characterized a ‘no-self theory’ as follows: ‘Each pulse of cognitive consciousness,
each Thought, dies away and is replaced by another. The other, among the
Buddhist Origins 21
things it knows, knows its own predecessor . . . saying: “Thou art mine, and
part of the same self with me.” Each later thought, knowing and including
thus the thoughts which went before, is the final receptacle of all that they
contain and own. Each Thought is thus born an owner, and dies unowned,
transmitting whatever it realized as its Self to its own later proprietor’ (The
Principles of Psychology, Volume I, p. 339).
The early Buddhist tradition reductively analysed what we understand as
persons into processes consisting of five impersonal components (skandha):
The body (rūpa)

Feelings of pleasure and pain (vedanā)
Sense-based perceptions of objects (vijñāna)
Conceptual thoughts (sam . jñā)
. skāra).
Volitions, inherited dispositions and habits (sam
Neither singly nor collectively do these transitory factors constitute an enduring

identity. There is no permanent self or person over and above the components.
There is nothing that ‘owns’ successively occurring experiences. Feelings,
perceptions, thoughts and volitions are all momentary events. There is no
mention of consciousness purely as such. They do not recognize what is
sometimes called ‘the unity of consciousness’. There is continuity within a
stream in that one event may cause the origination of another. The process
must be characterized as one of successive replacements rather than one of
change, because there is nothing that changes. A taste sensation may engender
a feeling of pleasure and each may be registered by distinct thoughts. This
complex may engender in the future a memory that causes a desire.
The term ‘person’ is but a convenient shorthand expression for talking
about a causally connected sequence of events. We unenlightened beings
make the mistake of supposing that what is merely a manner of speaking
designates an enduring substantial self, an inner controller, an irreducible
subject of experiences looking out on the world from a privileged perspective. It
is attachment to this misconception that is the ultimate source of human woes.
Buddhist praxis aims to eliminate the mistake. Belief in the soul leads to anx-
ious self-concern, narrowness of vision, defensiveness, antagonism to others
who are seen as barriers to the expansion of the ego and the perpetuation of
the process of rebirth. The Upanis.adic thinkers thought that the self was
ungraspable, beyond language, so subtle as to be undetectable. The Buddha
goes further. There isn’t anything there. We have here a radical solution to the
problem of human selfishness: there is no self to worry about.
It is apparent from the Pāli Canon that views about the nature of ‘the Self ’ –
what humans really are – were as multifarious during the Buddha’s time as
they are today. When fear of death perturbs the human spirit, we look for
something to hold on to that will exist in the future – or perhaps it won’t. Some
may think of ‘the self ’ in personal terms as the complex of experiences that
lasts for a while, others may understand it as a transcendental subject that is
exempt from worldly life and destined to exist for ever. It is this belief that
there is some sort of lasting self that is one of the varieties of grasping that only
lead to distress.
An early and popular scripture, the Snake Sutta in the Majjhima Nikāya,
echoes the Upanis.adic ethos:
The unenlightened person thinks of the body as his, as what he is, as his Self. He
sees feelings as his, as what he is, as his Self. He sees perceptions as his, as what
he is, as his Self. He sees thoughts as his, as what he is, as his Self. He sees
volitions and habits as his, as what he is, as his Self. He regards what he has seen,
heard sense known experienced, pursued and considered as his, as what he is, as
his Self. And there is the attitude, ‘The world and the soul are the same, and after
death this is what I shall be – permanent, enduring, eternal, immutable and I shall
exist like that forever.’ This view he regards as his, as what he is, as his self. . . . But
the enlightened persons does not think in these terms and so is not anxious about
what does not exist.
The point seems to be that identifying anything finite and transient, such as a
stream of embodied experiences, as a persisting personal identity (an obstacle
to enlightenment called ‘sat-kāya-dr.s.t.i’) is bound to lead to unhappiness and
anxiety. Thoughts and feelings just happen. In reality there is no one to whom
they happen. There is just a ‘bundle of perceptions’. The more sophisticated
understanding of oneself as a transcendental subject, really exempt from
worldly life and destined to exist forever, only leads to worry too. Will it really
continue? The enlightened person does not think in terms of either of these
egocentric frameworks, and that is the key to living without anxiety, without
obsessive self-concern. The key to happiness is not just altruism, but the
realization that there are no selves. The radical nature of the vision should
not be underestimated. Returning to the Snake Sutta:
When someone who does not have the view that the world and the soul are the
same, and that after death he will be permanent, enduring, eternal, immutable
and that he will exist like that for ever hears the Buddha’s teaching about
the abandonment elimination of theories, opinions and attachment to them, the
Buddhist Origins 23
teaching that aims at the suppression of clinging obsessive attachments, the
relinquishing of possessions, the end of craving, the cultivation of dispassion
and the extinction of greed, hatred and delusion – he does not think, ‘I shall be
annihilated. I shall be destroyed. I shall no longer exist.’ He is not distressed and
confused. He is not anxious about something that does not exist.
A much later work called the Questions of King Milanda relates a dialogue
about the no-self doctrine between Menander, the Greek king of Bactria in
what is now Afghanistan in the second century B.C., and a Buddhist monk
called Nāgasena. The text appears to be directed against a ‘personalist’ trend in
early Buddhism according to which the interactions of the five constituents of
personality (skandhas) produce a persisting individual that is reborn.
The monk says that he is called ‘Nāgasena’ but that is only a name, a label, a
conventional usage. It does not mean that there is a personal entity. The King
replies that this implies that in that case Nāgasena lacks parents, teachers
and superiors in the monastic order. It also rules out agency and moral
responsibility. At this point Nāgasena introduces the simile of the chariot,
which is made up of wheels axle, and chassis. The chariot is neither identical
with any one of its parts nor with their sum. The word ‘chariot’ is a conven-
tional designation for the collection. That is to say, if the chariot is dismantled,
we have collection of parts, not a chariot. Likewise with people. The name
‘Nāgasena’ is a conventional designation for a mental construction out of
the five constituents of personality. No person is found, as opposed to being
constructed. There is no further fact over and above the fleeting components
of the stream of experiences. (This perhaps overlooks the fact that the chariot
is the parts plus a structure.) Nevertheless, there is sufficient continuity within
the stream for us to make sense of agency and moral responsibility. The
Buddhist position is that moral responsibility does not require postulation
of a permanent self that is the subject of experiences. If we are to be happy,
we should be concerned about the future, even though it is not the future
of my self.
The tradition has it that the Buddha deliberately left a number of questions
unanswered. They include whether or not the world is eternal, whether or
not the world is infinite, whether self and body are the same, whether or not
enlightened beings (Buddhas) exist after death (a question that can be seen
as making the mistaken assumption that there are entities with determinate
identities).
In refusing to answer such questions the Buddha cannot be saying that
he does not know, for the Buddhas are omniscient. Rather, he is saying
that there is nothing to know. From the Buddhist point of view, such questions
reflect a resistance to the disturbing teachings of momentariness and the

non-substantiality of beings.
One Buddhist tradition says that thinking about such matters is not condu-
cive to the ending of suffering and enlightenment. One cannot lead a spiritual
life by virtue of believing that the world is eternal or infinite in space and time!
Another view is that the questions are themselves unanswerable because they
are posed in categories that are not ultimately real. The questions about the
cosmos presuppose that it is such that we are in a position to view it as a single
whole and that past, present and future are divisions within a single temporal
framework. It is difficult to formulate the question about whether the cosmos
has a finite age. As Sir Michael Dummett has put it, ‘Ascribing any finite age
to it depends upon calibrating time . . . given any such calibration, a new one
can always be derived from it so that the age of the universe becomes infinite,
although no new events have been postulated. Conversely, given a calibration
that brings out the age of the universe as infinite, it can always be revised so as
to render it finite’ [Thought and Reality p. 105].
Above all the problem is that the questions reflect a search for intellectual
certainty and conviction, an attempt to comprehend and perhaps control.
They imply that an absolute conception of reality, an Olympian perspective or
a totally objective grasp of truth, is accessible to us. The Buddha’s position
seems to be that is it not the case that such a perspective is available, but even
if it were, it would not help us along the path to enlightenment.
A much quoted traditional formulation distinguishes broadly between four
Buddhist philosophical positions:
The Vaibhās.ikas say that there are realities external to minds and that they are
directly perceptible. The Sautrāntikas say that there are realities external to minds
and that they are inferable from the occurrence of mental representations. The
Yogācāras deny that there are any realities external to minds. The Mādhyamikas
deny that there are any intrinsic natures.
The next chapters will look at these four schools of thought.
Further reading
Bechert and Gombrich (1984) is a collection of essays covering all forms of Buddhism, and is
beautifully illustrated.
Buddhist Origins 25
Rupert Gethin’s Foundations of Buddhism is comprehensive. It is now supplementary to his invaluable,
Sayings of the Buddha. Walpola Rahula, What the Buddha Taught, is a classic, written from a
Theravādin point of view. Steven Collins, Selfless Persons, is indispensable for early Buddhist
(Theravādin) representations of persons. Richard Gombrich, How Buddhism Began, puts many
things in context. Jayatilleke (1980) is informative about Early Buddhist Theory of Knowledge.
Edward Conze’s Buddhist Scriptures is a useful collection that contains the Questions of King
Milanda.
Bronkhorst (2000) contains translations of much original material and connects early Buddhism with
the Jaina renunciatory tradition. For the latter Jaini (1979) and Dundas (1992) are fundamental
contributions.
Mark Siderits, Personal Identity and Buddhist Philosophy, is a stimulating discussion of eastern and
western reductionist accounts of the self. So is Chapter Five of Paul Williams, Altruism and
Reality.

investigation
1. Can reductionism about persons make sense of human life?
2. If personal identity is an illusion what is making the mistake in the first place?
3 Abhidharma Buddhism
Chapter Outline
Ontology 27
Perception 31
Ethical consequentiality 32
The personalists (Pudgalavāda) 33
Further reading 34
It appears that after approximately 200 B.C., some philosophically minded

Buddhists set about reductively analysing our experiences, and the world-as-
experienced in ways that were more systematic and comprehensive than the
earlier reduction of the human person into the five impermanent components
of personality (skandhas): body, feelings, sensory perceptions, habits and
volitions and conceptual thoughts. They produced increasingly elaborate lists
and classifications of basic mental and physical elements (dharmas). The sys-
tematic organization of what was thought to be implicit in the Buddha’s teach-
ings, discursively presented in the Sutta-pit.aka, was called the Abhidharma-pit.
aka. Through the method of sub-division into minute detail and combination,
the catalogues proliferated. What motivated this enterprise? As Frauwallner
says, ‘The authors believed that they were accruing religious merit’. Discrimi-
native analysis (dharma-pravicaya) is an aspect of mental cultivation. This
means insight (prajñā) into the inherent natures of the elements. It involves
reducing familiar things and experiences to their constituents. It reinforces the
belief that there is no self understood as an entity in its own right. The skilled
Abhidharma Buddhism 27
practitioner dissolves the objects of attachment into their elements and
eliminates desires for them. The organization of dharmas into categories
also facilitated the drawing of distinctions between experiences that are
unwholesome and lead to more suffering from virtuous ones that lead to
liberation.
We know about the Theravādin and Sarvāstivādin (also called Vaibhās.ika)
Abhidharma sects (nikāya). It is the Vaibhās.ika view that will be described
here. Our most valuable source of information is a work called the
Abhidharmakośa-bhās.ya by Vasubandhu (350–400 A.D.), actually written
from a Sautrāntika point of view. A text called the Mahāvibhās.a, a com-
mentary on the Jñānaprasthāna, one of the seven treatises constituting a
collection called the Abhidharma-pit.aka of the Sarvāstivādins, is the source
of the designation ‘Vaibhās.ika’.
Ontology
Sarvāstivāda means ‘the theory that everything exists’. The questions ‘What
exists?’ or ‘What is there?’ cannot be answered by merely listing objects. We
need a classification of types of existents. Moreover, there has to be some
criterion or standard of judging what really exists and that criterion here is
irreducibility: nothing that is composed of parts is authentically real. The basic
elements in which all mental and material phenomena consist are called
dharmas. They are so called because they support (dhāran.ād < dhr.) their own
identity. Each dharma has a fixed essence or intrinsic nature of its own
(svabhāva). It is this possession of a permanent, fixed identity that differenti-
ates the dharmas from the macroscopic aggregates. Svabhāva means intrinsic
nature consisting in a specific inherent characteristic or power (svalaks.an.a).
The svabhāva of earth atoms is solidity, that of water atoms is fluidity and
that of fire atoms heat. It is the svabhāva of consciousness to apprehend
objects. The svalaks.an.a is not other than whatever it characterizes. An object’s
‘having’ svabhāva means that its identity is not determined by anything
else. It means self-sufficiency or independent subsistence. The presence of
svabhāva is held to permit the uniquely individuating definitions of the
basic elements proposed by the endlessly ramifying Abhidharma catalogues:
‘differentiation from the natures of others is in virtue of svabhāva’
(Abhidharmakośa 1.18).
Other Buddhist philosophical traditions see a problem here. The basic
elements are said to have intrinsic natures. But intrinsic nature is construed as
causal power, the capacity to do something. The nature of a white atom is to

contribute to a white surface and bring about a certain perception. The aim is
to identify the dharmas in virtue of their independent self-sufficiency, their
being as they are in themselves. But it seems that they are not characterized
in categorical terms but rather in dispositional ones, that is to say in terms of
their capacity to interact with other dharmas. Strictly speaking, each is charac-
terized in relational terms. We are not really specifying internal natures because
causal powers have to do with external relations. We are not being told what
the possessors of the causal powers are like in themselves. So the picture is one
of a giant causal flux, but with no explanation of the intrinsic natures of the
entities related in the flux. We shall return to this point in the next chapter.
The Sarvāstivādin catalogues distinguish conditioned (sam.skr.ta) and
unconditioned realities. The former are the constituents of causal processes,
being generated by co-operating causes and conditions (hetu and pratyaya).
The latter are whatever is exempt from causality: space (ākāśa), nirvān.a
understood as the cessation of the operation of dharmas due to knowledge,
and a range of unactualized possibilities.
The conditioned phenomena are:
i) Material phenomena (rūpa): bodies, sense-faculties and corresponding types of

objects
ii) Mind or thought (citta)
iii) Mental phenomena (caitta): feelings, sense-perceptions, intentions, volitions,
attitudes, memories, cognitions as well as a plethora of moral virtues and
weaknesses.
iv) ‘Factors dissociated from mind’ (citta-viprayukta-dharma) which cannot be
classified as either material or mental. These are: words and meanings; traces of
previous cognitions latent in the mind that supply the contents of dreams and
hallucinations; a force called prāpti that regulates the aggregation of particular
types of dharmas and locates them in a specific stream; the four characteristics
(sāmānya-laks.an.a) common to conditioned phenomena i.e. origination, temporal
extension, decline (entropy) and impermanence.
Dharmas, the elements of all mental and material phenomena, are manifest
as the cosmos. They momentarily exercise their power in causal complexes.
Since their existence does not depend upon anything else, they are basic and
irreducible. The Sarvāstivādins distinguish between the primary and irredu-
cible existence (dravya-sat) that belongs to the dharmas and the conventional
or nominal existence (prajñapti-sat) that belongs to their products. Dharmas
as primary realities exist in what we call past, present and future – but strictly
speaking, they exist timelessly or eternally. Their momentary occurrence as
the world of our experience is the exercise of their causal efficacy (kāritram)
that is also termed ‘svalaks.an.a’. A parallel distinction is drawn between
absolute or ultimate truth and reality (paramārtha-satya) and conventional
truth (sam.vr.ti-satya). The latter is the world as it is understood by finite beings
participating in conditioned causal processes. The former means reality as it is
itself, as understood by the Buddhas, who are omniscient beings.
Some assumptions lie behind the Vaibhās.ika view:
i) All mental acts, including memories and expectations, have existent objects
external to the mind. If past and future phenomena are cognized, they exist.
Memories and future expectations have real objects.
ii) The subject-object relationship in awareness requires two real terms.
iii) To be is to cause an awareness: anything that is the object of an awareness exists.
iv) We cannot escape the consequences of past actions.
[Past and future dharmas exist] because a cognition has a real object. When there
is an object, there arises a cognition. When there is no object, there arises no
cognition. If past and future dharmas did not exist, there would be cognitions
with unreal objects as their objective support (ālambana). Therefore there would
be no cognition of the past and future because of the absence of objective
supports. If the past were non-existent, how could there be future effects of good
and bad actions? For at the time when the effect arises, the efficient cause of its
actualisation (vipāka-hetu) would not exist. That is why the Vaibhās.ikas hold that
past and future exist. (Abhidharmakośa-bhās.ya 5.25ab)
The Sarvāstivādins say that everything – past, future and present – exists. By
contrast, the Vibhajyavādins say that only the present exists, as well as past
actions that have not yet yielded their consequences. What will be future, and
actions that have borne fruit do not exist. (Abhidharmakośa-bhās.ya 5.25cd)
(We shall be looking at the developed Vibhajyavāda view in the next chapter.)
A reality that is past has ceased due to impermanence. A reality that is future has
not originated. A reality that is present has originated and not yet ceased. When
basic realities (dharmas) exercise efficacy, this is called the present. If dharmas do
not yet exercise it, this is called the future. If efficacy has gone, this is called the
past. (Abhidharmakośa-bhās.ya 1.20)
The Vaibhās.ika view is that what we experience as the present is the exercise
of efficacy (phala-āks.epa-śakti) – the power of projecting effects that belongs
to a complex of dharmas. Svalaks.an.a is the same as specific function or
efficacy (kāritram). Efficacy (kāritram) may or may not be exercised. We

may think of the dharmas as they are in themselves as occupying another
dimension of reality, from which they briefly migrate into our world, and to
which homeland they return. The actualization, the temporal presence, of the
dharmas in and as our world, is the exercise of their efficacy in a causal complex.
Intrinsic natures are fixed (dharma-svabhāva is nitya), but the exercise of efficacy
is momentary (anitya) and circumstantial (kādācitka). The Vaibhās.ikas think
that it was to these momentary discharges of energy in the world of condi-
tioned phenomena that the Buddha was referring when he spoke of universal
impermanence. The exercise of efficacy when a dharma enters a causal com-
plex is not a change in a dharma, but just that element’s manifesting what it
permanently is. Each case of momentary actualization lasts just long enough
to cause its own following moment. A dharma attracts the manifestation of
another token of its own type. This explains stability and continuity. Efficacy
may be compared to the charge on the countless identical electrons. A dharma
is said to be in conditioned mode when it participates in a causal complex.
‘Conditioned’ means having the four characteristics of origin, duration, decline
and impermanence (sometimes expressed as impermanence, suffering and the
absence of essential identity). ‘Decline’, another basic reality, is what we may call
entropy, and explains why the flower fades and the grass withers.
Time is understood in terms of the exercise of efficacy. We see here a distinct-
ive understanding of time and a rejection of the view of the Vibhajyavāda-
Dārs.t.ānikas who held that time is a reality in its own right through which
the transient conditioned elements of existence move. They say that only the
dharmas that are present now exist, along with some retributive potencies of
past intentional actions. The Vaibhās.ika view is that what we call past is those
dharmas that have exercised their efficacy and what we call future is those
dharmas that have not exercised that disposition. Time is not a substantial
reality independent of subjects, objects and events. The dharmas do not ‘pass
through’ time as though time were a separate reality. Our experience of
temporal flow is in fact the replacement of dharmas. There is no real change
on the level of the primary realities, which exist timelessly and immutably.
(Some think that there is a problem in defining the present as the moment of
the exercise of causal efficacy. The latter is an activity, a process that requires
time. So the concept of temporal efficacy requires an independent grasp of the
distinction between past, present and future.)
Both macroscopic objects and experiences are built up out of the material
and mental elements of existence. Our world consists of entities which depend
upon primary existents. Both objects and subjects of experience are aggregates
or complexes, made up of parts. Anything that consists of parts, anything that
can be reduced to basic elements is said to exist by convention or nominally
(prajñapti-sat). If something can be dissolved into components, there is a dif-
ferent understanding of it. The atoms composing a clay pot are primary
existents, whereas the pot is a derivative ‘conventional reality’. This is not to say
that everyday things do not exist. But they are reducible to the basic realities of
which they are composed.
All causally conditioned things, spread out in space and time, and featuring
in our normal experience enjoy what is termed ‘conventional existence’. Our
‘life-world’ consists of interdependent, impermanent, temporary entities and
events that exist only, as it were ‘for the time being’. They are the referents of
our thoughts and language.
Perception
Where our interactions with the world about us are concerned, the Vaibhās.ikas
are direct realists. Macroscopic combinations of the atomic factors are the
causes and direct objects of awareness (an.u-sam.caya-vāda). Consciousness
is thought of as a blank page and is not affected by external reality. It reveals
objects without undergoing any change in its own constitution. This view
is called nirākāra-vāda and it means that our perceptions of objects are
unmediated by mental images, representations or ideas that fall as a veil
between mental acts and external reality.
Indriya Vis.aya Vijñāna

Sense Faculties Objects Experiences
Sight Colour/shape Seeing
Audition Sounds Hearing
Smell Odours Smelling
Taste Tastes Tasting
Touch Textures Tactile
Mental Thoughts Thinking
Items in column three are products of the interaction of the corresponding

pair (termed āyatana) in columns 1 and 2. They are described as perceptions
directed towards objects (prati-vis.aya-vijñapti). Mind (manas) is also treated
as a faculty. Just as we perceive physical objects by means of the senses, the
mind is that by which we grasp thoughts. This covers thinking about our
experiences of objects and states of affairs that are not present to the senses.
Ethical consequentiality
It is difficult to see how we could lead ethical lives if only the present were real.
The extension of causality to future consequences is a condition of ethical
being. Moreover, we all know that the past may return to haunt us. The
Buddhist insistence upon the importance of morality as the path to liberation
is apparently challenged by two of its own tenets: All conditioned things
(sam.skārā) are impermanent. All phenomena (dharma) are impersonal.
If ethically significant intentional actions are conditioned events, and thus
transitory and impersonal, how can they have subsequent effects in lives?
Buddhists believe that liberation is not possible for those who do not accept
that what is conventionally called (prajñapti) self is only a stream of compon-
ents of personality. Unenlightened people mistakenly believe in an identity that
is a further fact over and above the transient flux of embodied experiences.
From this attachment to the soul arise the afflictions such as grasping, aversion
and delusion. In fact, the notions of self and individual personal agency are
but conventional human constructs. The underlying reality of what we call
the person is a complex of components (skandhas) that admit of yet further
reduction into the real basic elements. Nevertheless, a causal chain of embod-
ied experiences is sufficient for the origin of the delusion that is personality.
It is the intentional actions of conscious beings alone that are responsible for
the arising and organization both of the sphere (loka) of creatures and of the
environments in which they may experience the consequences of their actions.
This is intended to exclude God, Time, the Soul or Prime Matter as causes of
the cosmos. Intention (cetanā), which is a mental phenomenon (dharma)
occurring in a stream, qualifies an action as ethically good or bad. But how do
actions have long-term consequences? The answer is that an intentional action
has both public expression (vijñapti-dharma) and a discrete feature called
avijñapti-dharma. The latter embeds itself in a stream of experiences and
remains there until circumstances are appropriate for the actualization (vipāka)
of its efficacy in that stream. This avijñapti-dharma is charged with the moral
quality of the public action from which it has arisen. The location of avijñapti
in an experiential series requires what are called ‘possessions’ (prāpti). The
latter belong to the class of those basic elements that are neither material nor
mental (citta-viprayukta-dharma) and are necessary if any phenomenon is to
adhere to the experiential stream in which it occurs. The avijñapti-dharmas
also account for settled commitments such as the Buddhist discipline and
dispositions of character.
The personalists (Pudgalavāda)

Vasubandhu says that what vitiates other doctrines is their false belief that
there is real personal identity: ‘There is no salvation from other religions,
because they are addicted to the false view that there is a permanent Self. They
do not understand that what they call self is only a label for a series of physical
and mental constituents. They imagine that identity is a distinct reality in its
own right. From this clinging to the conviction that there is a Self arise the
moral defects (kleśa).’
His commentator Yaśomitra quotes the poet Mātr.ceta:
As long as we think in selfish terms, the series of births does not cease. Selfishness
stays in the heart while there is belief in the soul. No other teacher in the world
propounds the unreality of the Self. So there is no path to peace other than your
teaching.
The ninth chapter of the Abhidharmakośabhās.ya is a critique of what was

called the Vātsīputrīya tradition, an offshoot of the Sarvāstivāda that held that
the Buddha actually did accept a form of personal identity, a self (pudgala) that
is the ground and support of changing mental states, as well as the substratum
of karma. This was, to say the least, controversial. It must be remembered that
Buddhism holds that acting unselfishly is not enough if it presupposes a belief
there is still a persisting individual identity. Buddhism goes deeper and insists
that not only is there no transcendental subject but also not the everyday
concept of stable relatively stable continuous selfhood is a misconception.
We do not know exactly what the Vātsīputrīyas taught because their views are
preserved only in the works of their opponents. I suspect that they thought
that something emerged from the combinations of the constituents that was
sufficient to account for moral responsibility. It could not be considered a
reality in its own right (a sixth skandha, as it were) because the Buddha did not
mention it. But it was more than merely a convenient designation (prajñapti)
for the aggregation of the interactive constituents of personality. Perhaps it was
thought of as something that emerges as part of a natural process and enjoys
stability for a while. Vasubandhu responds that if this ‘something’ were really
supervenient upon the constituents from which it emerges, it would be
knowable. If it is not a newly emergent entity, it is reducible to the constituents
and there is no point in positing it. Our experience can be explained as a
continuum of mental (and physical events) and there is no need to posit any
sort of self as the subject that owns the experiences.
Further reading
L. Pruden (1988) is an English version of Louis de la Vallée Poussin’s French translation of the
Abhidharmakośabhās.ya.
E. Frauwallner, Studies in the Abhidharma Literature, especially Chapter VIII, is a rich source of
information.
P. S. Jaini (2001), Section IV, contains some influential articles.
Alexis Sanderson, ‘The Sarvāstivāda and its Critics’, is illuminating. (This may be accessed via
Professor Sanderson’s website.)
James Duerlinger’s Indian Buddhist Theories of Persons, translates Vasubandhu’s critique of the
Pudgalavāda soul theory in the Abhidharmakośa.

investigation
1. Can the Abhidharma position make sense of a distinction between basic entities
and causal powers?
2. Can they make sense of the notion of change, rather than that of replacement?
Sautrāntika Buddhism 4
Chapter Outline
Ethical consequentiality 38
Dignāga and Dharmakı̄rti 38
Apoha: the exclusion theory of linguistic functioning 41
Self-awareness of mental events 42
The Ālambana-parı̄ks.ā 42
Extracts from ‘The Examination of Objective Supports of
Awarenesswith the author’s own commentary’
(Ālambana-parı̄ks.ā -vr.tti ) 43
Dharmakı̄rti 45
Metaphysics 46
Perception and thinking 47
The impossibility of permanence 50
Logic 51
The authority of the Buddha’s teachings 54
Further reading 54
The last chapter saw brief references to the Vibhajyavāda tradition. This out-
look became prominent and acquired the name ‘Sautrāntika’. They maintained
that only the Pāli Suttas, not the Abhidharma texts, were the authentic words
of the Buddha. They think that the Vaibhās.ikas have obscured the simplicity of
the Buddha’s original teaching and introduced the notion of permanence in
the guise of svabhāva. Hence they taught a doctrine of radical momentariness
(ks.an.ika-vāda), and simplified the ontology of the Vaibhās.ikas. They reject the
latter’s category of unconditioned phenomena, holding that space is just

the absence of extended objects. They say that nirvān.a is simply the non-
occurrence of suffering and not a reality or state of being. They reject the
category of the citta-viprayukta-dharmas. So there are no avijñapti-dharmas,
underpinning karmic causality and no prapti accounting for continuities
within a stream of experiences. Also rejected are the sāmānya-laks.an.as
(birth, continuation, entropy and impermanence) characterizing the brief
occurrence of dharmas in conditioned complexes.
The Vaibhās.ikas say that the atomic basic realities (dharmas) are permanent
essential natures (svabhāva) that may or may not exercise their causative
functions in and as the world. The Sautrāntikas retain the notion of dharma
but abandon that of intrinsic nature (svabhāva). We saw in the last chapter
that there may be a problem lurking behind the Vaibhās.ikas’ apparent inter-
pretation of intrinsic nature as causal power. The reason for positing intrinsic
natures for the dharmas is to enable their identification in virtue of their inde-
pendent self-sufficiency and thus to distinguish them from the conditioned,
fluid and transient macroscopic formations that they constitute. But in fact
intrinsic nature is characterized in causal terms as the capacity to interact with
other dharmas. So each dharma is actually characterized in relational terms.
We are not really specifying their internal natures in categorical terms but only
in dispositional ones. Causal powers have to do with external relations. We are
not being told what the possessors of the causal powers are like in themselves.
So the picture is one of a giant causal flux, but with no explanation of the
intrinsic natures of the entities related in the flux. This picture the Sautrāntikas
are happy to accept.
So the Sautrāntikas retain the notion of basic particulars but reject that of
essence (svabhāva). They understand the basic realities as instantaneous
unique particulars that are just moments of causal efficacy. They term the
basic realities ‘svalaks.an.a’, which was one of the Vaibhās.ika terms for intrinsic
nature expressed as causal efficacy. Such instants, dharmas minus svabhāva,
may perhaps be understood as flashes of energies forming fields of forces.
The Sautrāntikas insist that production by the co-operation of causes and
conditions obtains at every level, ‘all the way down’, and does not just apply to
macroscopic formations. All things are momentary in the radical sense that
they exist only for the moment at which they are produced. They argue that all
entities are inherently perishable, having no intrinsic tendency to continue in
existence. They reason that: Everything decays. Decay is non-existence. But
non-existence has no causes. So decay needs no external cause. It must be the
intrinsic nature of things to be perishable. Given that perishability is natural,
they cease spontaneously at the moment of their origination. Accordingly,
there is no sense in speaking of real past and future phenomena, as the
Vaibhās.ikas do.
While the Vaibhās.ikas hold that aggregates of atoms are the direct objects of
perception, the Sautrāntikas deny that the svalaks.an.as are directly perceptible.
We are acquainted with mental forms when clusters of svalaks.an.as, purely
causal potencies, somehow impart impressions of their forms (ākāra) to
conscious episodes. This theory is called sākāra-vāda. Clusters of unique
particulars are the material causes and objective supports of perceptions. Our
experience of external reality is mediated by mental representations caused by
the interplay of evanescent particulars. There is some sort of co-ordination
(sarūpya) between our mental images and the behaviour of the particulars. We
might consider here the case of rainbows. What we see are bands of colour in
the sky. But this phenomenal representation is caused by light waves refracted
off droplets of water. We do not directly see or feel the unique instants that
are, as it were, the raw materials of the world. Nevertheless they impress
themselves on episodes of awareness and are imperfectly grasped through the
filters of the mental images and concepts that they cause. Our concepts lead us
to reify the given and suppose that the reality basically consists of enduring
subjects confronting a world of objects, properties and structural principles.
The Sautrāntikas say that there are instantaneous realities external to
minds and that they are inferable as the causes of the occurrence of mental
representations (jñāna-ākāra). If the contents of awareness are just mental
representations, how do we know that there is an external reality? The reply
is that our representations do not occur at random, but are about definite
objects at specific times and places. Moreover, since we have no control over
much of which presents itself in our experiences, it is unlikely that mental
representations have been entirely generated from within a stream of experi-
ences. Surely they have causes other than the immediately preceding moment
(samanantara-pratyaya) in a mental stream. They conclude that the causes of
our mental representations are evanescent realities external to experiences.
Opponents are quick to point out that self-destructing instantaneous particulars
do not last long enough to enter into even short-lived formations that could
cause anything. Indeed the view that existence means spontaneous destruct-
ibility will attract the charge of nihilism. And we shall see that the Sautrāntika
representationalism may lead to idealism, which says that there is no need to
posit realities external to minds and that only ideas are real.
Ethical consequentiality
The Sautrāntikas account for karmic continuity by saying that an intentional
action, albeit instantaneous, is a ‘seed’ that initiates a transformation in an
experiential series (citta-sam.tāna-parin.āma-viśes.a). Its fruition is either
reward or punishment. The originating cause need not last until its effect is
realized since it is not a sustaining cause – like parents who are necessary for
the origination but not the continued existence of their offspring. They argue
that if an action continued to exist until its fruition, it would have to be eternal.
But if it ceased to exist, it could not produce anything. A seed initiates a series
beginning with germination. The fruit arises as the culmination of the series,
rather than directly from the seed. But it still needs the seed to start the process.
Although the series and the result depend upon the seed as the originating
cause, we do not say that the seed is either annihilated or that it is eternal.
Likewise an intention initiates a series of mental events from which the con-
sequence results. The series requires the initial intention and the consequence
arises from the series. The intention is neither annihilated, nor is it eternal.
Dignāga and Dharmakı̄rti

Dignāga (480–540 A.D.) was a Buddhist philosopher whose most important
work is the Pramān.a-samuccaya. He also wrote a work called ‘The Examination
of Objective Supports of Cognition’ (Ālambana-parīks.ā), extracts from which
are translated below. He belonged to the Sautrāntika tradition of thought, which
while admitting a real domain of instantaneous particulars external to minds
claims that what we know are only its reflections mediated by mental images and
discursive concepts. Dharmakīrti (600–660 A.D.) developed Dignāga’s ideas and
exercised an inestimable influence on subsequent debates. His works include the
Pramān.a-Vārttika, the Pramān.a-Viniścaya and the Nyāya-bindu. His treatise
called the Proof of Other Streams of Experiences addresses the problem of other
minds and argues that there is a multiplicity of streams of experience.
Prior to Dignāga most thinkers in the Buddhist tradition had accepted that
there are three means of knowing (pramān.a): sensory perception, inference
and reliable testimony (i.e. both human authorities and scriptural reports of the
Buddha’s teachings). Dignāga denies that testimony is an independent means of
knowing in its own right and subsumes it under inference. The rationale is
that we do not unquestioningly assent to what a person says but accept that
their words are true only when we believe that they are well-informed, reliable
and sincere. But above all what is apparent is a repudiation of the authority of
the Vedic scriptures, which are held to be a pramān.a in their own right by the
orthodox Brahminical traditions.
Dignāga divorces sensory perception (pratyaks.a) from thinking (kalpanā)
that always involves concepts and words. The former is pure sensation whereas
the latter always involves words, concepts and judgements. Dignāga calls it
‘inference’ (anumāna). Sensory perception never involves general concepts
(sāmānya-laks.an.a). Perception, always valid because there is no scope for dis-
tortion by the workings of the mind, is direct experience of external reality,
which consists of fluid clusters of unique, momentary particulars (svalaks.an.a).
Because they do not share any common features, particulars are indescribable.
We have here an instance of a reductionist and nominalist outlook: everything
truly real is individual or particular.
The expressions ‘svalaks.an.a’ and ‘sāmānya-laks.an.a’ are inherited from
the Abhidharma. There ‘svalaks.an.a’ means the characteristic activity of an
individual basic element (dharma) as it is in itself, and ‘sāmānya-laks.an.a’
means the features common to dharmas when their combinations produce
conditioned, macroscopic formations. Such generalities include non-eternity,
unsatisfactoriness and lack of persisting identity. As a Sautrāntika, Dignāga
accepts that reality consists of clusters of unique instantaneous particulars
(svalaks.an.a) and denies that each atomic factor has an unchanging and eternal
essence. General features are conceptually constructed by perceivers.
Moreover, he believes that the categories of things (padārtha) that the Nyāya-
Vaiśes.ika realists claim to be basic structures discovered in the world are in fact
imposed by the workings of our minds. He says that conceptual construction
(kalpanā) is the interpretation of what is given in pure sensation by means of
proper names, words for general features (jāti, sāmānya), words for qualities
(gun.a), words for actions (karman) and words for individual substances (dra-
vya). Our minds group unique particulars together and understand them as
continuing objects bearing types of properties. In other words, the construc-
tive activity of minds constitutes objects out of the flux of sensation. In reality,
there are no universals, no real stability and no entities with determinate
identities.
Dignāga thinks that thought and language are inseparable: conceptual thought
is born out of language and language is born out of concepts. Conceptual thought
and language deal in generalities (sāmānya-laks.an.a). But because there are

no objective general features, even simple concepts are at one remove from
reality. They are causally related to the realities and not just arbitrary fictional
inventions. But there is a gap between how our minds work and the way
things are. It is a mistake to suppose that the map is the territory. His view is that
concepts and thinking are interpretations that disguise rather than disclose.
So all inferential procedures are distanced from reality. Logic is not about
reality as rightly articulated in language, but is a set of rules governing the
moves in a conceptual language game. ‘All this convention involving inferen-
tial reason and properties to be established is based on the distinction between
property and property-possessor which is itself imposed by the human mind:
it is not grounded in anything existing outside the mind.’ But his contribution
to an empiricist theory of inference was very influential, although it must be
borne in mind that it is similar to ideas found in the works of Vasubandhu and
in a text called the Nyāyapraveśa. Dignāga proposed that in a valid inference,
we must have observed cases of an inseparable connection (avinābhāva-
sam.bandha) between the logical reason (hetu) and what it establishes (sādhya).
We observe wherever the reason occurs, there the sādhya occurs also. We
can be confident that we are reasoning reliably and responsibly (the three
conditions are not claimed to be of an absolute guarantee of truth) when
three conditions (trairūpya) obtain:
(i) The logical reason (hetu) must really be a property of the subject (paks.a) of the
inference.
(ii) The logical reason must be present in some instance (sapaks.a) other than the
subject of the inference, which is similar to that subject in that it too possesses
the property that is to be proved (sādhya).
(iii) Whatever lacks the property to be proved also lacks the proving property or rea-
son. There must be no instances (vipaks.a) where the proving property occurs and
the property to be proved does not.
Take the inference that sounds (paks.a) are impermanent (sādhya) because they
are products (hetu). [Invariable association: Whatever is produced is imperman-
ent.] Here the sapaks.a could be something uncontroversially impermanent
such as a pot that also exhibits the property of being a product. It is open to us
to cite an actual instance illustrating the joint absence of the property to be
proved and the logical reason. The atmosphere would be a negative example
because it both lacks impermanence and is not produced by effort.
Apoha: the exclusion theory of

linguistic functioning
The thinkers of the Nyāya-Vaiśes.ika realist tradition think that objective
general features (universal properties, natural kinds, qualities such as colours,
shapes and sizes) are the grounds for the repeated applications of general
terms. On this view we classify some individual animals as cows because they
form a natural kind. The single, real universal property ‘being a cow’ is itself
an entity common to all cows. But Dignāga is an anti-essentialist and a nomi-
nalist who denies that there are any objective generalities structuring and
causally regulating the world, which consists of instantaneous unique and
indescribable particulars. But if there are really neither shared properties nor
even resemblances, how do words and concepts function? They cannot all
behave like proper names because in that case we could not say anything
about things. Dignāga’s answer is that the word-meanings and concepts form-
ing the fabric of inherited and public conventional understanding ‘exclude
each other’. The idea is that words (and concepts) do not have meaning in
virtue of their referring to extra-linguistic realities. Rather, they are signs
whose meaning derives from their roles in a framework of significances,
where they stand in relations of opposition and complementarity. (Later
thinkers say that apoha means the mutually exclusive inter-relations of modes
of presentation (pratibimba) and concepts (vikalpa) that determine what we
deem objects and states of affairs.) Words and concepts normally have both
an inclusive and an exclusive role. Dignāga emphasizes the exclusive func-
tion. A word is expressive when it excludes other meanings belonging to the
same system. Dignāga denies that we need to posit a single real universal
property shared by all individual cows as the basis of the application of the
word/concept ‘cow’. It is sufficient that what those clusters of particulars that
we call cows have in common is just their being different from whatever is not
treated as a cow. We apply a meaning such as ‘cow’ just on the basis of the dif-
ference of cows from everything else. This ‘difference from non-cows’ is not a
genuine feature. Difference is purely relational. The word ‘cow’ does not stand
for a property or essence. We sort some particulars because it suits human
interests to do so. Language is a network of mutually exclusive meanings that
we have conventionally constructed in accordance with what matters for us.
To illustrate: We apply the word ‘analgesic’ to a variety of pills with totally
distinct pharmacological properties because they relieve pain. The concept
‘analgesic’ is a humanly constructed one because pain relief is a matter of

interest to us. It should be noted that we have here a theory about the func-
tioning of language and not a theory about the acquisition of language or the
acquisition of language and concepts. Were it the latter, it would be impossi-
ble to explain how anyone could learn the meaning of a word in the first
place, because they would have to exclude an infinite range of other things.
As for the former, there is no problem: we are born into a beginninglessly
established linguistic community.
Self-awareness of mental events

All Buddhists deny that there is any constant experiencing subject that is
distinct from the process of awareness. So how do we account for the phenom-
enon of subjectivity? Dignāga says that just as feelings are ‘self-aware’, each and
every perception and judgement is aware of itself (sva-sam.vitti): it is inherently
self-illuminating. A cognition simultaneously and in virtue of the same act
cognizes its own form as well as that of what it is about. If things were
otherwise minds would be like video-recorders receiving and recording
information and there would be no inner mental life. Reflexive mental events
follow each other so quickly that there is generated the illusion that there is a
persisting subject of experiences that we call a self.
The Ālambana-parı̄ks.ā
According to the Vaibhās.ika direct realists, the objective ground (ālambana-
pratyaya) of a thought is the reality in the world that it is about. According to
this theory, an objective ground is both the extra-mental cause of an idea
and the provider of its representational content. An hallucination is not an
objectively grounded thought in the sense that it has content but it is caused by
some defect in the perceptual system. Dignāga agrees that for something to
qualify as the objective ground of an awareness, it must be both the cause and
the representative content of that awareness, but he does not accept that such
causes have to be extra-mental realities.
In his Ālambana-parīks.ā, Dignāga argues against the Vaibhās.ikas that their
realistic atomic theory actually leads to the admission that the direct objects of
perceptual awareness are internal mental forms and not mind-independent
realities. The Vaibhās.ikas hold that we directly perceive structured masses
of real atoms of various kinds and that these cause our perceptions. Like
Vasubandhu, Dignāga questions the possibility of atomic aggregation. But
even granting that collections of atoms may cause mental representations, the
atoms do not figure in the subjective content of awareness. A compound of
clay atoms may cause the perception of a pot, but we do not see such a cluster
of atoms. The atoms do not enter into the content of the representation:
what we have is an experience of a solid, coloured extended object. It could
be argued that a conglomerate of atoms does constitute the representative
content. But the problem here is that conglomerates are not real and so fail to
qualify as causes of ideas. Dignāga concludes that only an idea (a mental rep-
resentation that appears as if it is about something external) can be the support
of another idea. The cause of a mental representation can be another represen-
tation: thoughts may arise from other thoughts rather than from external
objects. An idea may bring about another idea and be sufficiently like it to mir-
ror its representational content. Although it seems that the conclusion is an
idealist one, Dignāga is not an idealist. He believes that the momentary par-
ticulars exist independently of minds, but the direct objects of acquaintance
are their representations in consciousness. He wants to persuade us that our
shared, conventional framework of representations is just that, and that our
thoughts and concepts do not mirror reality as it is in itself. The goal is to
encourage us to realize that our everyday attachments, and our thinking in
terms of ourselves as persisting individual subjects confronting a world of
already established propertied objects awaiting our descriptive understanding,
are really just matters of conventional construction. Once such realization is
achieved we are in a position to detach ourselves from our basically self-
centred concerns and follow the path of insight and compassion leading
towards enlightenment.
Extracts from ‘The Examination of

Objective Supports of Awareness
with the author’s own commentary’
(Ālambana-parı̄ks.ā -vr.tti)
People who believe that external things are the objective support of sensory
perceptions suppose that either atoms are the support since they cause the
cognition, or that a conglomerate of atoms is the support because such a form is
the representative content of awareness.
Even if the atoms are the cause of sensory cognition, since the cognition
does not represent them, the atoms are not the intentional object (vis.aya) of the
cognition (just as the sensory faculties are not).
The expression ‘intentional object’ means the proper form of something as it is

grasped in cognition, since the cognition is manifest as having that form (ākāra).
Although the atoms are the cause of the manifestation of the thought-form, their
nature is not grasped by cognition (just as the nature of the sense faculties is not).
Thus the atoms are not the objective support of cognition.
As for the conglomerate of atoms, although it is what is represented in

awareness, it is not the objective support because
It is not the cause of the phenomenal representation (ābhāsa).
It makes sense that when a thing produces a cognition that represents it as it is,
that thing is the objective support. It has been stated that such is the originating
condition. But the conglomerate of atoms does not qualify as such
because it is not a reality – like the experience of the moon seeming double.
In the case of seeing the moon as double owing to a sensory defect, although
the moon appears double in awareness, it is not the direct object of awareness.
Likewise the conglomerate is not the objective support because it cannot be the
cause since it does not exist as a real entity.
Thus what is external is not the direct object of awareness.
The external things called ‘atom’ and ‘conglomerate of atoms’ are not the
supports of awareness, because although the atoms cause the awareness they do
not feature in its representative content, and while the conglomerate appears as
the content of awareness, it is not its cause.
Some hold that the aggregated form is the instrumental cause of the cognition.
The form of the atom is not the object of awareness, in the same way that its
solidity is not.
Everything is multifaceted. So it is perceived under some form or description. The

atoms have the nature of being the originating cause of awareness and are repre-
sented as conglomerates. Just as real solidity is not an object of visual perception,
likewise real atomicity is not.
While there are no differences between the atoms that make up objects, it
may be said that the differences between perceived objects emerge from the
different formations of atoms. But this is not the case because (according to
the opponent) the atoms that are the only true realities do not have different
dimensions. The differences between the forms of objects operate on the level of
human conventions. Conventional modes of differentiation do not apply to the
atoms. Everyday objects are posited by the human mentality.
Pots etc exist only as human conventions because if the atoms are taken away
the cognition whose form derives from them is lost. But this does not happen in
the case of what is truly real. Therefore it is intelligible that the direct objects of
sensory perceptions are not realities external to minds.
But the knowable internal form, which appears as if external to the mind, is the
direct object because it has the form of awareness and is its support.
Although there are no external objects, an internal reality, an idea appearing as if
external, is the objective support.
The internal mental form is the objective support of awareness since it both
supplies the manifest image in the cognition of the object, and produces the cog-
nition of the object. (The internal form both supplies the manifest image belong-
ing to the cognition (which the atoms do not) and is the cause (which the
conglomerate qua unreal, cannot be)).
Dharmakı̄rti
Dharmakīrti identifies belief in a substantial, permanent and personal self as
a form of ignorance, indeed enchantment. This ignorance is a genuinely
effective occurrence that brings about a complex of a specific mentality and
behavioural dispositions (sam.skāra). Belief in the self is inherited from mis-
taken constructions of selfhood in previous lives. From mistaken adherence
to belief in the self arise the moral defects (kleśa) such as desires, infatuations
and antagonisms. The settled condition that is attachment to the self is
expressed in first personal thoughts: ‘The innate belief in a personal reality
is expressed in I-thoughts such as: “May I be happy, may I not suffer.” ’
Attachment to self is inextricably associated with the notion of ‘mine’ and
this inevitably generates desires – we want things to go well for ourselves.
It automatically generates hatred and aversions towards whatever is felt to
be inimical to one’s own interests – often what are imagined to be other selves.
Striving for personal happiness conceals the true nature of the moral defects
so we do not see them for what they are. As long as there is clinging to
self, there is rebirth. But following the Buddhist path, including repeated
meditation on the unreality of the person and the way in which belief in the
self ’s reality causes suffering and frustration, on the impermanence and
non-substantiality of phenomena, gradually eliminates misunderstanding
and the consequent moral defects. Interiorization of and insight into univer-
sal non-substantiality, seeing the truth, are the proper functions of a purified
mind. Philosophy helps by revealing that nothing real can be permanent. We
are captivated by an inherited and shared web of conceptualization
(kalpanā). But the validity of the teaching that dispels our enchantment
(moha) is known by its fruits.
Metaphysics
Dharmakīrti agreed with Dignāga that in the final analysis objective reality
is a flux of momentary particulars (svalaks.an.a) that are inexpressible and
incommunicable as such. By virtue of their configurations in relation to the
cognizing subject, they determine the differences in the representative content
of cognitions. Each svalaks.an.a has its own causal effectiveness (arthakriyā).
Dharmakīrti develops Dignāga’s view. Instead of saying what is real by
specifying some type of entities (e.g. the svalaks.an.as or the dharmas), he uses
a criterion: only what is causally effective is real. Nothing permanent can be
causally effective either successively or in the present because it cannot engage
in processes of change. Each particular has specific location, time and form. If
something does not perform useful activity, then it is not a real entity, since it
satisfies the criteria for non-existence; but if it does perform useful activity,
then it is not permanent. This criterion of reality rules out the existence
of anything supposedly eternal and unchanging such as God, the soul and
its permanent consciousness, the sounds of the Vedic scriptures, the eternal
relation between Vedic words and their meanings, universals and primal
material nature (prakr.ti) that is inert prior to its plural manifestation as the
cosmos. The question about the reality of otherwise of permanence is central
to the dialectic between Buddhists and Brahmins, some aspects of which we
will be examining in later chapters.
Dharmakīrti follows Dignāga’s epistemology: only sensory perception and
inference are means of knowing (pramān.a). Knowledge is reliable cognition in
so far as it contributes to the successful accomplishment of some purpose. It
may also reveal something new: but to Dharmakīrti’s mind, the instrumental
function is primary and matter of fact truths are revealed in practice. Although
like Dignāga he refuses to accept that testimony and language can be epistemic
authorities in their own right – they are primarily thinking here of the absolute
authority that the Brahmins ascribe to the uncreated Vedic scriptures – in fact,
he says that language may be an instrument of knowing in the derivative sense
that it communicates what are already established to be useful truths about
what is to be sought after and what avoided. The Buddha, and the scriptural
records of his teachings, are sources of knowledge in that they reveal things
that would be otherwise unknown, and tell the truth about what should be
pursued and what eschewed.
As we have seen, according to the Sautrāntika outlook there are no permanent
realities. To be real is to be causally effective and that implies the capacity for
change. But they take it for granted that were there anything permanent
it could neither act nor change. Dharmakīrti says that there is no such thing
as a permanent means of knowing (nityam. pramān.am) such as the Vedic
scriptures or a divine intelligence because knowledge operates in a world of
changing realities. A means of knowing cannot be unchanging because it is
concerned with impermanent objects. Whatever happens as part of a process
cannot be permanent and unchanging. Something permanent could not be a
means of knowing about what is impermanent because it could not depend
upon assisting factors such as objects, subjects and instruments in the case of
the knowing process. The means of knowing are such because they enable us
to achieve our goals in a world of ever changing realities.
Perception and thinking

It is the particulars that are immediately given in sensory perception.
Dharmakīrti outlines his view of knowledge and perception in the first
chapter of his Nyāyabindu:
The accomplishment of human goals presupposes right cognition.

Right cognition is twofold: perception and inference.
Perception is free from conceptualisation (kalpanā) and is reliable (lit. non-errant)
Conceptualisation is cognition involving a representation that is capable of
beingexpressed in words. (He thus modifies Dignāga’s view that thought and
language always go together in suggesting that thought is prior to language.)
Cognition that is both free from conceptualisation and from defects is sense-
perception (pratyaks.a) [as a means of knowing.]
That has four aspects:
a) Cognition involving the sense-faculties;

b) Understanding (mano-vijñānam) produced by the sensory cognition that is its
immediate cause. This process takes the initial sensory cognition of the exter-
nal object as its object.
c) All minds and mental events are reflexively aware of themselves.
(The point here is that they do not require illumination by another conscious
principle such a conscious self). The reflexivity of awareness is used by some
Brahmin thinkers as a way of proving the permanence of the self.
d) Direct Yogic intuition into the atomic composition of reality that does not
require sense-faculties as intermediaries.
The momentary particulars (svalaks.an.a) are the objects of sensory perception.

They are what is real in the final analysis (paramārtha-sat), because reality is that
which has the capacity for causal efficacy.
Different from them is what falls under generalising conceptualisation (sāmānya-

laks.an.am).
That is the sphere of thinking (anumānasya vis.ayah.). [This is kalpanā]
Perceptual cognition produces knowledge when thought enables us to achieve or

obtain something.
Knowledge is a cognitive state that is in conformity with reality, because success-

ful activity follows when cognitions agree with realities.
Dharmakīrti’s theory about the relation between perception and the world can
be understood in terms of the triad: sensation – image – concept. A cluster of
svalaks.an.as has the power to produce the sensation of blue. Blue impressions
produce an awareness having two aspects: a blue mental image (ākāra) and the
blue mental image’s being aware of itself. The image copies the impressions.
Constructive mental activity (kalpanā), conditioned by traces of prior experi-
ences, interprets the image and produces the thought (vikalpa) that something
is blue. This thought enables us to think, act and communicate. The external
particulars are only the indirect objects of the thoughts that they cause. But
the unique instants behave in such a way that we can organize them under
unifying concepts. While our concepts, involving the association of names
and general properties with the given, do not copy the fluid play of the real
particulars, they represent it indirectly as a map does as territory. We do
not directly know reality as it is in itself because we are primarily aware of
images, some of which are converted into concepts, derived form sensory
impressions. In short, there is a gap between the way our minds work and the
way things work.
Sense perception on its own has no practical application because it does
not discriminate anything. Assuming that the senses are operating normally
and environmental conditions do not obstruct them, it cannot be either true
or false because truth and falsity apply only to conceptual mental states.
Perceptual sensation applies to reality as it is in itself (vastu) before we start to
thinking about it. But it is only when an aspect of reality has been mentally
discriminated in a perceptual judgement (adhyavasāya) that we can act in rela-
tion to it. Judgements using concepts enable successful activity (i.e. are reliable)
when they are causally related to the real particulars constituting events.
A vikalpa is a concept that the mind constructs out of the data given in
sensory awareness. Cognitions involving apparently shared features of objects
are conceptual interpretations based on experiences of particulars. Conceptu-
alization involves generalizations but there are no objective generalities.
Objective reality is strictly ineffable, since it includes no general features. Like
Dignāga, he wants us to realize that our conventional ways of understanding,
integral to which is the notion of individual subjectivity, in the final analysis
disguise the truth.
But to find our way around successfully, we need to make discriminations
using concepts and words. Some concepts, and elaborated conceptual schemes,
apply more adequately than others to objective reality: i.e. they work better for
us in leading to successful activity. Vikalpas interpret and organize the data of
sensation, making them intelligible and serviceable. The store of human
concepts, built up from impressions derived from a beginningless series of
previous experiences, is transmitted down the generations via shared language.
While some complex concepts ultimately derive from sensory impressions
and mental images formed from them, others, especially the idea that there is
a persisting soul, are just produced by the creative imagination.
A problem arises when people overlook the purely conventional nature of
what are only human ways of thinking and suppose that they correspond
to objective realities. It is a natural mistake to suppose that our concepts
are copies of reality or that our representations mirror reality as it is in itself.
Error occurs when conceptual thought takes its own forms to correspond
directly to reality. Since reality consists of momentary unique particulars,
general concepts cannot represent it as it is in itself. Moreover, stable concepts,
enshrined in language, encourage us to think that there are stable realities.
A much-quoted verse (354) from the Chapter on Perception of the
Pramān.avārttika reads:
‘Mind is really not diversified but it appears to be differentiated into objects

known or grasped (grāhya), perceivers (grāhaka) and cognitions because of
mistaken views.’
Later thinkers sometimes read this in an idealist sense, but that it not
Dharmakīrti’s meaning. He means that the oppositions between the perceiv-
ing subject, objects and thoughts are functions of the way our minds work and
not genuine realities. It is we who contrast subjects and objects, thinking them
external to each other. The differentiation of subjects, acts and objects of cog-
nition within the one mind appears because of inherited influences of previous
ideas in a beginningless and uninterrupted stream of experiences. Positing
oneself as an individual thinker facing a world of objects is a kind of selfishness
that Buddhist practice aims to eliminate. Since the polar notions of object and
subject are interdependent, by exposing the falsity of one, we can expose the
falsity of the other. Once a person has really understood that the conventional
view of reality as consisting of enduring objects existing independently of the
mind of the individual perceiver is false and that our thoughts are not copies of
reality, they can come to understand that selfhood and its attachments is an
illusory construct.
The impossibility of permanence

We have seen that Dharmakīrti takes dynamic causal efficacy as criterial of
reality. It follows that nothing can be permanent and static. Dharmakīrti’s
opponents recognize different forms of permanence (nityatā). By permanent
they mean both eternal and immutable. (Some realists distinguish two
varieties: absolute permanence and permanence compatible with change.)
Nyāya and Vaiśes.ika ascribe permanence to kinds, universals and some
relations. The Mīmām.saka ritualists say that both the basic sound units (varn.a)
of the Sanskrit language, the relation between a Sanskrit word and its meaning,
and the Vedic scriptures as a whole are all permanent realities. Others hold
that consciousness is the permanent nature of the soul. The Sām.khyas, for
example, say that the true nature of the soul is permanently static conscious.
(Dignāga had challenged this tenet with the consideration that if the self really
changes when a cognition occurs, it is impermanent. But if it does not change
when cognition occurs, it is not a knowing subject.) The consensus among
the various sorts of realists is that permanent realities are revealed and known
by their appropriate manifestations (vyakti) in space and time. Individual
entities are manifestations of universals. A specific usage of a word manifests
a timeless meaning. The evolutes of the material nature (prakr.ti) postulated
as the universal material cause by the Sām.khyas are its manifestations. Realists
argued that such manifestation does not compromise eternity and immutability.
Dharmakīrti says that manifestation entails mutability. If there are universals,
they could never be manifest in individuals. Similarly, if the sounds of the Vedas
are permanent, we could never hear their manifestations. If the soul is eternally
of the nature of awareness it would always know everything or nothing.
Dharmakīrti examines the notion of manifestation in the context of
the revelation of objects by the cognitive process. On a realist account, some
form of cognitive activity, accompanied by factors such as light, is instrumental
in revealing already existing realities. But according to Dharmakīrti, since
everything is momentary there can be no already existing objects. Objects
have to be constituted out of the data of sensation. He insists that whenever
such objects are cognized, this happens as an aspect of a process. Take a stream
of moments that someone may identify as constituting a vase. The stream is
occurring in a dark room. Open the door and switch on the light. You see the
vase. It becomes manifest. Dharmakīrti’s view is that the presence of the
observer (and the light) has introduced a change into the situation. The ‘stream
of vase moments’ is not what it was. It is now involved with the light and
the cognitions of the observer. The latter introduce additional factors that
render the ‘pot-stream’ capable of producing a cognition in the observer. The
manifestation or revelation of the pot is the product of the co-operation of a
variety of causes upon which it depends. Now consider the possibility of the
manifestation of permanent realities whose natures are supposed to be entirely
self-sufficient and independent of co-operating extra factors. If the essential
nature of an entity of that kind is such that it is productive, it will always
produce its characteristic effects. If its essential nature is such that it is not
productive, it will never produce effects. Let us apply this to consciousness:
if the nature of consciousness is to actually illuminate objects, it will always
illuminate everything. And in the case of universals, if being manifest as
individual instances is internal to their nature, they will always be manifesting
all their instances (and we will be aware of them). In the normal case of
cognition (of the vase), manifestation was dependent upon co-operating
causal factors and crucially upon the introduction of an extra-factor into
the situation that rendered the stream of moments capable of producing an
awareness. But this cannot apply in the case of allegedly permanent entities
that are supposed to be in principle knowable. When Dharmakīrti concludes
that their revelation is impossible, he is assuming the realist consensus that if
something is real, it is in principle if not in fact knowable.
Logic
As we have seen Dharmakīrti thinks that while immediate sensation relates
directly to reality that consists of unique instantaneous particulars, the mental
images (ākāra) and concepts (vikalpa) that they cause and in terms of
which we interpret what is given do so indirectly. But Dharmakīrti does
not think that our concepts are imaginary inventions, although some are.
The instantaneous actualities behave in such a way that we can organize
them under concepts. Although the natural regularity (svabhāva-pratibandha)
between smoke and fire or that between something’s being an oak and its
being a tree holds primarily between two concepts, it also reflects a real state
of affairs that causes us to make the connection between the concepts.
Dharmakīrti says that inference does not grasp the realities directly in that it
operates by determining the object in a mental representation that is not itself
the object. But because the representation of the object is causally related to
the real objects, we can make reliable inferences.
The key feature of a valid logical inference is the invariant association
(vyāpti) between the logical reason (hetu) and what is to be established
(sādhya). Dharmakīrti says that the invariant association of As with Bs (which
he also calls avinābhāva –‘sine qua non’) must be guaranteed by a natural
regularity (svabhāva-pratibandha). The theory of natural regularity attempts
to underpin some forms of inseparable connection in the absence of objective
universals. We know that the connection between the logical reason and the
property to be proved could not be otherwise when the connection is either
that between cause and effect (kārya-hetu: e.g. fire and smoke) or a case of
shared nature (tadātmya /svabhāva-hetu – if something is an oak, then it is a
tree). This necessary relation is the natural regularity. So we may infer from
the fact of something’s being an oak that it is a tree and from the presence of
smoke to the presence of fire. This principle is applied in characteristically
Buddhist arguments like, ‘If something is produced, it is perishable by nature’.
While Dignāga seems to have been content to allow that the idea of
invariable association is the product of a finite range of observed instances,
as well as the lack of counterexamples (adarśana-mātra), Dharmakīrti wants
to strengthen the basis of inference because the inductive approach is insuffi-
ciently general and leaves open the possibility of our discovering exceptions in
the future. His teacher Īśvarasena thought that our constant association of the
logical reason (hetu) and that which is to be established (sādhya) was based
merely on the fact of our not having observed any exceptions to the rule
(adarśanamātra). Dharmakīrti thought that this made the basis of inference
too fragile: why should we not discover an exception in the future? Moreover,
we have not surveyed every relevant instance. We do not know that there is no
instance where the hetu occurs and the sādhya does not. We just have not
come across one so far. This is why he argued that the inseparability of hetu
and sādhya had to be grounded in the natural order of things. This means that
the presence of the logical reason guarantees the presence of the property to be
proved. There can be no counterexample.
We saw that causal connection is one of the two forms of natural regularity
between the logical reason and what is to be proved. A causal relation is
understood through positive and negative perceptions. The causal connection
between smoke and fire is known when we find that smoke, which had not
been present, appears when fire is introduced and that when the fire is
extinguished, the smoke disappears.
Dharmakīrti applies this in a proof of the existence of streams of experience
other than one’s own. He argues that we have inferential knowledge of other
minds. Given that in one’s own case there is observation of the phenomena of
language and behaviour immediately after volition or intention, and given that
they are not observed in the absence of volition, one knows from one’s own
case that there is a cause-effect relation between volition and the occurrence of
actions. The causal relation is established purely because we are cognizant of
the relations between intention and action and know that where there is no
intention there is no action. Seeing that actions separate from us occur even
when we have not framed any intention, we infer intentions elsewhere to be
the cause of those other actions. Thus other minds are established. Just as I
know from my own case that certain actions are preceded by certain thoughts,
so I may analogously infer that similar patterns of speech and behaviour on the
part of other people show that they are separate streams of experiences.
According to Dharmakīrti, the non-apprehension (anupalabdhi) of an
entity that is in principle perceptible (dr.śya) establishes the absence or non-
existence of that entity. This applies to the problem of other minds: from the
fact that we do not perceive them it does not follow that they do not exist
because they do not fall under the category of the in-principle perceptible. But
the case of the Brahminical concept of the Self (ātman) is different. Those who
believe that the Self is a basic reality characterize it as something that should
be uncontroversially knowable. Dharmakīrti and other Buddhists focus on the
problems of disentangling the soul from the personality and its experiences.
They reason that it is never known, although it is described as the sort of
thing that is knowable. This non-apprehension proves its non-existence. The
same pattern of reasoning is applied to the notion of Prime Matter (prakr.ti),
which is supposed by Sam.khyas and Vedāntins to be the ultimate source
and underlying cause of all material products. But the fact that we do not see
supernatural entities, such as ghosts, does not prove that they do not exist
because they are by nature inaccessible to normal perception. This applies to
anything inaccessible to perception by virtue of space, time or its nature.
The authority of the Buddha’s

teachings
Dignāga had subsumed reliable testimony, including scriptural statements
about unverifiable matters, under inference as a means of knowing (pramān.a).
We infer that the Buddha’s teachings are valid because we know that he was
reliable and sincere, and above all the teaching works in practice. Dharmakīrti
is more radical. He denies that any scripture concerned with the supernatural
and suprasensible matters can really have epistemic authority. Human cogni-
tive possibilities are ordinarily restricted to objects that are actually percepti-
ble, in principle perceptible and hence inferable. We have no access to the
supernatural. But the religious person must be concerned with matters outside
the range of our ordinary cognitive capacities. Where ordinary human author-
ities are lacking, he must have recourse to some scriptural authority if he is to
pursue a way of life that conduces to well-being (purus.ārtha). So his situation
is that of one who must choose which of the scriptures and which form of
religious praxis to follow. The best we can do is to follow the Buddha’s advice
and adopt what works. We have no means of knowing about the supernatural.
All we can do is hope that if a body of scriptures is a reliable guide to living well
here, their teaching about what is unverifiable is trustworthy too.
Further reading
Satkari Mookerjee, The Buddhist Philosophy of Universal Flux is about both Dignāga and Dharmakīrti
and the many disputes between their school and the Naiyāyika, Mīmām.saka and Sām.khya realists.
M. Hattori, Dignaga, On Perception reconstructs some of the first chapter Pramān.a-samuccaya and
covers much more than perception.
R. Hayes, Dignaga on the Interpretation of Signs, has translations of chapters II and V of the
Pramān.a-samuccaya.
There is a translation and exposition of the Ālambana-Parīks.ā in Tola and Dragonetti (2004).
For the text of the Pramān.avārttika see Pandeya (1989). For that of the Nyāyabindu see Svami
Dvarkadasa Sastri (1994).
Kajiyama (1998) translates a work belonging to the Dignāga-Dharmakīrti tradition.
John Dunne, Foundations of Dharmakirti’s Philosophy contains detailed treatments of ontology,
epistemology, logic and the philosophy of language. There is a valuable Appendix of translations.
Siderits (1991) deals with the philosophy of language, especially the apoha theory, which
has attracted much scholarly attention in recent years. The Nyāya response is to be found in
Uddyotakara’s commentary on Nyāya-Sūtra 2.2.63. (Jha (1984), p. 1034 ff.)
The articles in Tom Tillemans, Scripture, Logic and Language combine philosophical acuity and
philological expertise.
Vincent Eltschinger, Penser l’autorité des Écritures does much more than that and is a mine of
information about Dharmakīrti’s intellectual context and religious concerns.
Bimal Matilal, Perception, relates Buddhist representationalism to modern concerns.
Claus Oetke (1994), Trairūpya, puts Dignāga’s logic in context, tracing its antecedents and relating it to
Nyāya and Vaiśes.ika parallels.

investigation
1. Does the Sautrāntika account of moral responsibility make more sense than that of
the Vaibhās.ikas?
2. Is it possible to reconcile atomistic impersonality and moral responsibility?
3. Is Dignāga’s theory of language a credible one?
4. Why does Dharmakirti think that truth is the same as successful practice?
5 Nāgārjuna and Madhyamaka
Buddhism
Chapter Outline
Verses from Nāgārjuna’s Ratnāvalı̄ 60
Refutation of Objections 65
Further reading 67
Nāgārjuna (c. 150–200 A.D.) was a monk trained in the Abhidharma tradition,
which tried to delineate the basic structures of reality as understood from the
ultimate point of view. He repudiated this enterprise that involved categor-
izing mental and material phenomena into types of basically real elements
(dharma) having essential natures (svabhāva). Sometimes when Nāgārjuna
says that he is not offering a theory of his own, he may have the Abhidharma
taxonomical activity in mind! Nāgārjuna holds that supposition that things
have unchanging and enduring natures, either at the fundamental or macro-
scopic level, only encourages us to become attached to them. He thought
that the Sarvāstivādins were in effect aspiring to make statements about
reality as a whole from a totally objective point of view. Nāgārjuna denies that
any such perspective is attainable by us unenlightened beings. There is no
point in our even attempting to distinguish between ultimate truth and
conventional truth.
His argumentative strategy is to list the possible propositions about some
subject-matter. He then examines them and shows that they are inconsistent
or lead to erroneous or unwanted conclusions. So he denies all of them. Using
this method, he tries to show that our theories and conceptual constructions
Nāgārjuna and Madhyamaka Buddhism 57
cannot capture reality. Since we are not Buddhas, we can attain no grasp of
reality as it is in itself.
His most significant philosophical works include the Madhyamaka-kārikās,
the Vigrahavyāvartanī and the Ratnāvalī. His thought is the subject of a long
exegetical tradition that continues to this day. He is sometimes difficult to
understand. This is because we do not always know the specific questions
to which his statements are the answer. The most important commentator
is Candrakīrti (600–650 A.D.) who wrote the Prasannapadā. Śāntideva’s
(700–750) Bodhicāryāvatara, describing the Bodhisattva’s path to final
enlightenment is another influential work.
According to the Abhidharma traditions, the basic elements (dharma)
have intrinsic natures or essences (svabhāva). Intrinsic nature was thought
to be timeless, self-sufficient, independent of all else and unchanging. (The
last predicate is crucial.) It is the possession of such permanent identity that
distinguishes the basic elements from the temporary conditioned aggregates
that are the objects of everyday thought and language and grounds the distinc-
tion between ultimate reality and conventional reality. Nāgārjuna insists that
neither everyday objects nor the dharmas can have intrinsic natures. If they
had, there would be no change. Reification, the investiture of states of affairs
and objects with persisting identities, comes naturally to us. But in truth,
everything is empty of intrinsic nature (svabhāva-śūnya). Emptiness (śūnyatā)
must always be understood as meaning ‘absence of essence’. It does not mean
non-existence. He often says that what we normally consider things, and the
concepts with which we carve up reality, are neither real not unreal. That is to
say, while our discursive conceptual schemes and the entities that they posit do
serve our purposes to an extent, they cannot be the whole truth.
There are no entities with having intrinsic natures that have arisen either from
themselves, or from other things, or from both themselves and others, or from no
causes. [MMK 1.1]
Whether in the cause, or in the conditioning factors, or in a complex of causes

and conditions or in something else, nowhere are there found intrinsic natures of
entities. This is what we mean by saying that all entities are empty. For instance
the sprout is neither in the seed that is its cause, nor in the conditions such as
earth, water and air taken singly or collectively, nor is it a separate reality distinct
from the causes and conditions. Since there is no intrinsic nature there, the sprout
lacks an intrinsic nature. Lacking intrinsic nature means that it is empty. Just as the
sprout lacks an intrinsic nature and is empty, so are all entities empty because they
lack intrinsic nature. [VV 1]
The origination of intrinsic nature from causes and conditions is not intelligible.
Intrinsic nature produced by causes and conditions would be created. But
how could intrinsic nature be created? For intrinsic nature is uncreated and not
dependent on anything else. [MMK XV, 1–2]
Since there is no entity (dharma) that has not arisen dependently on others, there
is no entity that is not empty of intrinsic nature. [MMK XXIV, 19]
Sometimes Nāgārjuna says that whatever is interdependently originated

from causes and conditions is devoid of essential nature. That is true. But
he does not mean that lack of essence follows just from the fact of being
interdependently originated. The basic point is that there cannot be any
essences in the first place.
A typical argument against essence is: suppose that seeing is the essence of
the visual faculty, the efficient cause in a visual awareness. The visual faculty
cannot see itself. Seeing, the intrinsic nature of the visual faculty, only operates
in the presence of a visible object and light, and is consequent to another state
of awareness. So the visual faculty’s characteristic way of being actually depends
upon the co-operation of a variety of conditions. Its nature is not intrinsic to it.
Another argument is that if everything is impermanent, things cannot have
intrinsic natures. If being young is the essence of youth and being old the
essence of the elderly, what undergoes the ageing process? There is also a
difficulty in formulating the relation between essence and that which has
the essence. If essence is a characteristic of its bearer, the bearer must already
exist. But if the bearer already is an entity, the notion of essential nature is
superfluous. And to return to a point made above, the Abhidharma thinkers
appeared to understand essential nature in terms of causal power. But this
leaves open the question of what it is that has the power. He thought that
they were not really identifying essences, but talking about the dispositions of
entities to behave in certain ways.
Nāgārjuna insists that the Abhidharma outlook is contrary to authentic
Buddhist teaching about momentariness and non-substantiality according to
which there are no unchanging, self-contained and self-sufficient realities.
Nāgārjuna says that the Buddhist theory of causation (pratītya-samutpāda)
means that everything is interdependent and empty of own-nature. He says that
entities with essential natures would have to be uncaused or self-created, which
is impossible. There cannot be any basic elements with essences or immutable
natures. If reality basically consisted of dharmas possessing essence, the
universe would be static and there would be no changes. If the unsatisfactory
round of existences (sam.sāra) had essence or a fixed nature, there would be no
possibility of nirvān.a and if nirvān.a had essence, there would be no sam.sāra.
The Abhidharma distinguished between what is genuinely and substantially
real (dravya-sat) and what is treated as real or true by convention (or nominally
existent) (prajñapti-sat). The possession of svabhāva was criterial for being
genuinely real. Nāgārjuna retains the distinction but does not see it as an
ontological one. For him, the distinction is between what we conventionally
take to be true (prajñapti-sat) and the absolute standard, the standpoint of
the omniscient Buddhas. There are not two worlds or two dimensions of
reality, one conditioned and the other unconditioned. There is one reality that
can be understood from the point of view of enlightenment or from the point
of view of some conventional pattern of thinking.
Unenlightened people think of the world as a system of interactions between
more or less stable entities that have determinate identities. This is reification
or commodification and it is a basic, inherited and shared mistake. Nāgārjuna
often compares our conventional outlooks to dreams, illusions and mirages that
cannot be classified as real or unreal. Nāgārjuna believes that he is recovering the
original teaching of the Buddha because the belief that things and selves have
stable, enduring natures only encourages us to become attached to them.
Analysis reveals inconsistencies and shows that the conventional world-view
cannot be true. The realization that everything is empty of essence puts an end
to conceptual construction and reification. We cease to believe that our con-
cepts are capturing an objective reality. Insight into emptiness, the realization
that discursive thought cannot reach the truth, leads to a compassionate
outlook and mental peace, as one is no longer disappointed by the search
for certainties. When we realize that the concept of svabhāva is incoherent,
we are on the path to enlightenment.
Not only does the Abhidharma distinction between conventional reality
and ultimate reality (dharmas with essential natures) collapse, but also we
cannot differentiate between sam.sāra as the conditioned realm and nirvān.a as
the unconditioned. Nāgārjuna insists that there is not the slightest difference
between sam.sāra and nirvān.a. But there is a difference between the ways in
which we may understand things.
Nāgārjuna recognizes that the Buddha’s teaching about interdependent
origination, which is one way of expressing emptiness, was expressed in
conventional terms. Indeed, all the Buddha’s teachings were. But it would be
a mistake to think that because the Buddha used the everyday categories of
commonsense, this somehow validates them. The Buddha, in seeking to point
us in the right direction by recommending a way of life that he had found led
to enlightenment, had to use some language. He did not say anything positive
about nirvān.a because it is not a state, thing or entity.
Paradoxes and inconsistencies in our ways of thinking show that we cannot
formulate a complete and correct description of reality ‘as it is in itself ’
independently of any particular perspective. In the Madhyamaka-kārikās,
Nāgārjuna subjects what realists take to be basic concepts such as those of
causality, motion, time, agency, self and substance to criticism. He shows that
they are cases of reification and conceptual construction. He thinks that it is
pointless to entertain the possibility that our limited conceptual capacities
and schemes can capture the ultimate truth. We cannot step outside the
world, look at it from the outside and make definitively true statements about
it as a whole. We cannot frame an absolute conception of reality and we are
wasting our time and spiritual possibilities in seeking to. We cannot formulate
a correct and comprehensive ontological theory from a totally objective
‘Olympian’ point of view, which is what the Abhidharma attempts to do. This
is one meaning of what he calls, ‘the emptiness of emptiness’. Emptiness is
not ‘a reality’ and there is no essence of emptiness.
Verses from Nāgārjuna’s Ratnāvalı̄

Chapter I
When we live well following the righteous path, the attainment of the ultimate
good follows. Those who practise the perfect life gradually achieve the ultimate
good. [3]
Living well is happiness, and freedom from rebirth is the ultimate good. Trust in
the Buddha’s teaching and insight are the means to that good. If he has trust, a
person may share the path. If he has insight [into emptiness], he knows truly.
Of the two, insight is the most important, but trust comes first. [4–5]
One who does not transgress the path because he is led by his own desires,
antagonisms, anxiety and delusions is to be considered trusting. [6]
(The wise person reflects upon the moral value of his actions. He avoids
violence and killing, theft and sexual misconduct. He controls his tongue to
avoid lying, cruel words and malicious gossip.)
The unenlightened person is frightened when he hears the teachings ‘I am not’,

‘I shall not be’ ‘Nothing belongs to me nor ever will.’ [26]
The Buddha has said that such fears are the result of a mistaken belief in personal
identity and possessiveness. [27]
In reality, it is a mistake to think in terms of ‘I’ and ‘mine’ because neither is a

reality when one has understood how things really are. [28]
The interactions of the five components (skandhas) arise from the sense of
personal individuality (aham. kāra). Personal individuality is not a genuine reality.
If the seed of something is unreal, how can its sprout be real? [29]
When it is seen that the components are unreal, the sense of individuality is given
up. From the giving up of that, the skandhas no longer function. [30]
Just as one sees the reflection of one’s face in a mirror, although it is not the real
thing, so one conceives individual personality on the basis of the components,
although it is not a genuine reality like the reflection of one’s face in the mirror.
[31–32]
Just as there appears no reflection of one’s face without a mirror, so without the
five components individual personality does not appear. [33]
While there is grasping at the components, there is the thought ‘I’. When there is
belief in personal individuality there is karma and rebirth. [34]
[Just as a mirage looks like water, but is neither water nor really anything, so the
components look like a persisting self but they are not really a self. [54]]
The individual person cannot be produced either by itself, or by another, or by

both itself and another. Nor is it not eternal. When one realises this, personal
individuality vanishes and thence karma and rebirth. [37]
The pattern of thought here is applied in all manner of contexts. Nothing

can bring itself into being. It would have to exist already in order to do so.
If something (in this case, the person) is like that it is a permanent, eternal
reality, neither beginning nor ending. But it makes no sense to say that an
entity is produced by another. This is because a cause is a cause only in relation
to an effect. But if what we are calling the effect does not exist, it is absurd
to speak of the cause. If the cause is non-existent, the effect will be too. The
formulation ‘not by itself and another’ follows from the first two.
When one understands the relation of cause and effect in this way, one realises
that the world as a whole cannot be considered as an entity that might or might
not exist. [38]
(If the notion of causation belongs to the sphere of our experience in that it is
to be understood in terms of relations between finite things, how can we apply
it to the cosmos as a whole and argued that it must come from someone or
something?)
Followers of the Upanis.adic tradition do not worry when they say that familiar
worldly life will not exist in the state of liberation. So why are they afraid when we
say that there are no absolutely real entities here either? [40]
In the state of release (moks.a) there are neither individual identities nor the five
components. But if such a state is dear to you, why do you resist the elimination
of the self and the components in this life? [41]
It is not the case that nirvān.a is non-existent. But what could constitute it as an
entity? Nirvān.a is beyond the concepts of being and non-being. [42]
(Nirvān.a is not to be understood as some state or place that is concealed by the

world. It is a mistake to reify nirvān.a and think of it as something that exists as
a sort of parallel universe. We return to the original message of the Buddha:
Nirvān.a is just the extinction of the fires of greed, hatred and delusion.)
In brief, the nihilist theory denies that actions have consequences. This false view
is immoral and leads to hell. [43]
In brief, the true view is that actions have consequences. This correct view brings
merit and rebirth in good states. [44]
When thanks to insight one has ceased to think in terms of what is and what is
not, one no longer thinks in terms of merit and demerit. The good say that this is
freedom (moks.a) that is beyond good and bad births. [45]
***
If a cause is produced before its effect or simultaneously with it, in reality it is not
a cause. The concept of origination is incoherent, either from the absolute or the
conventional point of view. [47]
(As we said above: since something is identified as a cause only when it

produces an effect, if the relevant factor pre-exists the effect, it cannot be
considered as the cause. It can only be considered as the cause after the time
when the effect has come into existence and in that case its causal function is
superfluous. Nāgārjuna is not saying that there are no entities. He is saying
that they are not essentially identifiable as causes and effects.)
Causal relations may be expressed like this: when A is present, B arises. For
instance when we have the idea of long, that of short arises. When something is
produced there is production of something else – such as radiance after the
production of a lamp. [48]
If there is no short, there can be no long. There can be no radiance if no lamp is

produced. [49]
When one understands causality like this, it does not lead to nihilism. [50]
This is a classic statement of interdependent origination (pratītya-samutpāda).

Regularity is just a matter of one thing following another. There is no need to
posit invisible and innate causal powers. The point here is that an attempt to
understand causality in terms of the transformation of essential natures (or
any other metaphysical account such as satkāryavāda or asatkāryavāda) – is
bound to fail and that disillusionment may lead to nihilism.
We move on to the problem of saying anything sensible about the ‘onto-
logical status’ of the cosmos or about reality as a whole ‘as it is in itself ’:
The world does not come into existence. It does not go out of existence. It does
not remain static even for an instant. How can we say that the world as a whole,
to which the categories of past, present and future do not apply, is real? [63]
In truth, since the temporal framework does not apply to either the world or to
nirvān.a, how can we specify a real difference between the two? [64]
Given that there is no duration, there is neither origination nor cessation. So how
can the world be produced, endure, and cease. [65]
Just as the concept of production cannot apply to the cosmos taken as a whole,
nor can that of time, and the correlative notions of origination, endurance
and cessation. The cosmos cannot have a starting point in time, if time is a
measure of change and times are relations between things in the cosmos.
Next we see a characteristic example of Nāgārjuna’s arguments against the
Abhidharma:
How can existence be non-temporal if things are always changing? If it is not the
case that things are always changing, how can we account for their variability? [66]
If everything is momentary, how do things get older? But if things are not
momentary, in the sense that they remain the same, how do they get older? [68]
We move on to a critique of the metaphysic of essential temporality – the view

that existence can really be reductively analysed into basic moments:
If an instant has an end, it must be supposed to have a beginning and a middle.
Given that the instant consists of three parts, it cannot be a basic reality. [69]
Beginnings, middles and ends must be considered like the instant (i.e. similarly
reducible, so there is an infinite regress). The condition of being a beginning, a
middle and an end does not exist from itself or from something else. [70]
No atom is simple since it has many sides. No atom lacks sides (if it did it could
not be connected with others). The ideas of unity and plurality are mutually
dependent, as are those of existence and non-existence. [71]
This anticipates criticisms of the notion of atomic aggregation into larger

entities. The point is that if atoms do not combine, they cannot be simple
or atomic, which is a contradiction. Nāgārjuna is saying that analytic reduc-
tionism as practised by the Abhidharmikas fails since we cannot identify
basic units. The later idealist argument will be that we cannot make sense of
physical matter.
Chapter II
As the Kadalı̄ tree and all its parts when split down the middle is not anything,
likewise with the person when it is analysed into components. [1]
Hence the enlightened ones have said that all dharmas lack intrinsic natures. They
have ascertained the real nature of the components and seen that they are not
substantial. [2]
It makes no sense to affirm or deny substantial identity. [3]
The Buddha has stated that what is observed and what is stated in scriptures is
neither true nor false. When there is an argument, there is a counter-argument
and neither is absolutely true. [4]
The universe is really beyond the categories of truth and falsity. In truth, we
cannot say ‘it is’ or ‘it is not’. [5]
How could the omniscient Buddha affirm of the universe, about which no true
statement is possible, that it has an end, or that it is infinite, or that it is really
plural or that it is non-differentiated? [6]
People ask how many Buddhas have been, will come and are here now. But the
notion of a limit on the number of beings presupposes the three-fold temporal
framework. [7]
There is no cause of the growth of the world. Decay is relative to the three-fold
temporal framework. [8]
In this consists the depth of the teaching that is a secret from ordinary people:
that the world is like a magical illusion is the essence of the teaching of the
Buddha. [9]
An elephant conjured up by magic may appear and it may seem to have a
beginning and end. But really it has no beginning and end. [10]
Likewise we see apparent beginnings and ends of things in the world. But in
reality there are neither fixed beginnings nor ends. [11]
As a magic elephant comes from nowhere and goes nowhere, being due to a
conjuror’s pretence, it does not last as a reality. [12]
Likewise the world is like an illusion that comes from nowhere and goes to
nowhere. It does not last as a reality since it is only mental delusion. [13]
What then is the meaning of this world organised by the three times? It cannot
be said to be nor not to be, except from the conventional standpoint. [14]
Therefore the Buddha did not say whether it is finite or infinite, plural or single. [15]
Refutation of Objections
Nāgārjuna wrote an important work called ‘The Refutation of Objections’
(Vigraha-vyāvartanī) in response to criticism levelled at his method by
followers of the Nyāya school. He envisages an opponent who says that
the proposition that all entities lack intrinsic natures itself lacks one and thus
cannot deny anything. If he admits that the proposition has an intrinsic nature,
he is contradicting himself [VV 1–2].
What does it mean to say that a statement lacks an intrinsic nature? When
a statement is true, it is an example of language operating as an instrument of
knowledge (pramān.a). Nāgārjuna is supposing that the opponent holds that
the essence of a pramān.a consists in its power (śakti) to be an instrumental
cause that establishes the truth about things. So the point is that if a pramān.a
lacks intrinsic nature, it also lacks that capacity.
Nāgārjuna replies that the opponent has not understood the meaning
of emptiness. Entities that are interdependently originated are empty of
intrinsic natures because they are dependent upon causes and conditions. His
proposition is indeed empty in this sense. But everyday objects, albeit empty,
perform their functions successfully. The same applies to his proposition.
Essence is not a precondition of functioning in an ever-changing world. The
statement has a therapeutic value for people who take it for granted that things
have essences.
There is also the objection from the Nyāya school that Nāgārjuna cannot
show that all things are empty, since such a demonstration requires that
there are valid means of knowing (pramān.as: i.e. perception, inference and
testimony). The objects of the pramān.as must exist too, because one cannot
negate what does not exist. The opponent says:
You deny the reality of things after you have apprehended them by perception,
but you also say that the perception by which entities are cognised is not a
reality. [VV 5]
Nāgārjuna replies:
If I apprehended an object by perception or inference or testimony, I could

then affirm or deny things about it. But I don’t do that so the objection is not
sound. [VV 30]
If you hold that objects are established by means of knowing (pramān.a), tell me
how you establish those means of knowing. [VV 31]
If the pramān.as are established by other pramān.as, there is an infinite regress.

There is neither beginning, nor middle nor end. [VV 32]
If you think that pramān.as are established without pramān.as, you have
abandoned your own doctrine. [VV 33]
A pramān.a cannot establish itself, because something cannot exercise its

characteristic activity upon itself. [VV 34–39]
If the means of knowing are self-established, they are established independently

of the objects known. Self-establishment does not require anything else. [VV 40]
If you think that the means of knowing are established independently of the
objects known, then those means of knowing are not means of knowing about
anything. [VV 41]
If the pramān.as are established only in relation to the objects known, the objects
known are not established by the pramān.as. [VV 43]
If the objects known are established independently of the means of knowing,

what is the point in seeking to establish the pramān.as? [VV 44]
If you hold that the objects of knowing are established by the means of knowing
and that the means of knowing are established by the objects of knowing, you
cannot establish either. [VV 46]
If the pramān.as are established by the objects known, and if those objects have to
be established by pramān.as, then, because the pramān.as have not been estab-
lished, the objects have not been established either. So how will the objects known
establish the means of knowing?’ [VV 48]
Further reading
David Burton’s Emptiness Appraised is the best starting point. Chapter III of Paul Williams’s Mahāyāna
Buddhism is helpful. His Altruism and Reality is mostly about the Bodhicaryāvatāra. See especially
Chapter V.
The Bodhicaryāvatāra is lucidly translated in Crosby and Skilton (1996).
Bhattacharya (1998) translates the Vigrahavyāvartanī. Lindtner (1982) contains much helpful
explanatory material, as well as texts and some translations. The most readable translations of the
Madhyamakakārikās are those by Jay Garfield (1995) and Frederick Streng (1967). Chapters VI–IX
of Mark Siderits, Personal Identity (2003), sees emptiness as a global form of anti-realism and offers
much food for thought.
Hahn (1982) has the text of the Ratnāvalī.
For engagement with Nyāya over the question of the pramān.as see especially Uddyotakara’s
commentary on Nyāya-Sūtra 2.1.8–19. This is translated in Jha (1984), p. 606ff.
Bimal Matilal’s ‘Logical Illumination of Indian Mysticism’ (Matilal, 2002) is stimulating.

investigation
1. Why does Nāgārjuna think that the Abhidharma distinction between the absolute
and conventional dimensions of reality collapses?
2. Is Nāgārjuna entitled to make any truth-claims?
6 ‘Mind-Only’: Yogācāra
Buddhism
Chapter Outline
Extracts from Vasubandhu’s ‘Twenty Verses Proving that only
Mental Phenomena are Real’ (Vijñapti-mātratā-siddhi) 72
Thirty verses on consciousness 74
Vasubandhu and Sthiramati on ‘The Construction of Phenomena’ 76
Illustrative extracts from Sthiramati’s Madhyāntavibhāgat.ı¯kā 76
Further reading 81
Some Buddhists espouse an idealist form of philosophy in that they deny that
there are any material or physical realities existing independently of minds.
There are neither selves nor an external world but only constructs of selfhood,
agency and objectivity arising from the flux of momentary self-aware thoughts
and feelings. We must bear in mind here that the philosophers were also
monks, practising profound meditation every day. The point is reflected in the
designation of this tradition as ‘Yogācāra’, which means, ‘the practice of yoga’.
(Other names include ‘Citta-mātra’, which means ‘mind-only’ and ‘Vijñāna-
vāda’ or ‘consciousness theory’.) Meditation often involves experiencing what
are purely thought-forms as if about external realities. It is not surprising
that such people should be especially open to the possibility that what we
ordinarily take to be external realities are but projections of consciousness.
The Buddhist idealists have a strong sense that the ways in which we experi-
ence what we unenlightened beings call the external world is conditioned by
personal and subjective factors. Our mind-set or world-view determines how we
see the world. What makes one person’s perception of a state of affairs different
‘Mind-Only’ Yogācāra Buddhism 69
from that of another is the moods, emotions and memories that one brings to
bear in the circumstances. This is illustrated by the point that when hungry
ghosts see a body of water, they see a mass of pus. Humans see it as a crystal
stream and drink from it. Such observations about the subjective constitution of
experience do not in themselves license any conclusions about the ontological
status of the physical world. But we shall see that these Buddhists present argu-
ments against the intelligibility of the concept of material substance.
The central figure here is Vasubandhu who lived during the period c. 350–
400 A.D. Trained in Sarvāstivāda methods of analysis and meditation, he
wrote a work called the Abhidharmakośabhās.ya which is a critical survey from
a Sautrāntika point of view of Buddhist realist schools. Another work is the
Karmasiddhiprakaran.a, which is a Sautrāntika critique of realist notions
of how karma operates, and an attempt to reconcile atomistic impersonality
with moral responsibility and consequentiality. It appears that he moved
from a representationalist to an idealist philosophical position and wrote the
Madhyāntavibhāgaśāstra, the Trisvabhāvanirdeśa, the Vim.śatikā and the Trim.śikā
from that point of view. His commentator Sthiramati lived around 550 A.D.
The Sautrāntika representationalists think that it makes sense to suppose
that most of our perceptions have external causes. Mental contents are
representations (ākāra) caused by a world outside the mind to which they
bear some relation. That relation is not one of mirroring or picturing since
they hold that the external world really consists of a flux of unique momentary
particulars (svalaks.an.a) and does not feature as such in the content of
awareness. They posit an external world on the basis of the inference that
there has to be something that causes those experiences over whose occur-
rence we have no control (bāhyārtha-anumeya), not as a result of direct
acquaintance (bāhyārtha-pratyaks.a). We can see how easily this may encourage
an idealist outlook. The gulf between what is supposedly given in experience
and its interpretation in concepts, thinking and judging (an interpretation that
conceals rather than discloses) is just too wide. If we are not directly acquainted
with objects in the world, and if conceptual and descriptive thinking does
not reach out to the world, and if the existence of that world can only be
certified by inference (itself a mental activity), why suppose that there is an
extra-mental physical dimension to reality? If the manifest content of experi-
ence is determined by our thoughts rather than by objects in the world, we
might wonder what sense can be given to the notion of a mind-independent
reality. Surely it is falling out of the picture. If experience of a hypothetical
given – the amorphous flux of ineffable particulars – does not express that
given as such but is posited as something that has a purely instrumental causal
role in the genesis of discursive thinking, actually concealing the true nature of
things, it is not obvious that experiences of such a given need be postulated at
all. Considerations like these (whose influence is apparent in Dignāga’s
Ālambana-Parīks.ā) seem to be the impetus behind Vasubandhu’s move
towards idealism or the view that only the mental factors of existence
(vijñaptis) are real. Vasubandhu came to reject the Sautrāntika view that we
can validly infer that there is an extra-mental reality as the cause of our percep-
tual sensations. We do not need to posit a material dimension of reality in
order to explain the character of experiences, whose occurrence can be
explainedby the revival of traces of prior experiences (vāsanā or sam.skāra)
within a stream. There is an argument from economy: it is always better to
assume one thing than to assume many. It is better to postulate potent mental
traces of experiences than to posit external objects. (Perhaps he thought that
the maxim that ‘when something is seen, there is no need to postulate the
unseen’ begs the question.) He argued against the very coherence of the notion
of material substance. He rejected the Vaibhās.hika view that we are directly
aware of objects made up of real physical atoms. Atoms are partless and indi-
visible. As such they cannot join together. But if we insist that atoms come
together, they must have parts. If they coalesce, there will be no increase in
extent: if they have no dimension, they cannot combine to form larger objects.
If they have dimension, they will be divisible and this undermines atomism. If
there are no atoms there cannot be any wholes distinct from their parts. The
Vaibhās.ika realists hold that all mental acts have existent objects external to
the mind. They say that to be is to cause an awareness: anything that is the
referent of an awareness exists. Vasubandhu argues that dreams and hallucina-
tions show that this is not true. Perceptions do not necessarily depend upon
mind-independent realities.
We might suppose that as well as objects outside the mind, there is also spatio-
temporal determination. Vasubandhu replies that experience of such determina-
tion also occurs in dreams. In response to the argument that it appears that
there are many minds experiencing the same objective environment, Vasub-
andhu appeals to the Buddhist notions of hells, which are shared hallucinations.
Vasubandhu thinks that individual events in a mental series are aware
of themselves. An awareness is simultaneously and in virtue of the same act
self-cognized, just as a lamp illuminates itself while illuminating an object.
This tenet of the reflexivity of mental events is central to the idealist outlook:
It shows that an idea can be the object of another idea and that there is no need
to posit physical objects as the causes of our thoughts. Moreover, the reflexivity
of each individual mental event dispenses with the need for an independent
consciousness, an observer-principle with a perspective on mental states.
Like all Buddhists, Vasubandhu believes that it is the intentional actions
of sentient beings that are responsible for the diversity and organization of
the cosmos, which exists to be the environment in which the consequences
of actions are to be experienced. The Abhidharma thinkers understand this
realm to be basically constituted by the material and mental elements of
existence. The idealists reject the category of material elements and hold
that think that the elements of existence are only the mental ones. Lives are
streams of ideas (vijñapti) ever emerging from a mental storehouse of vestiges
(ālaya-vijñāna) impressed by previous actions. These self-conscious ideas may
mistakenly conceive themselves as individual subjectivities, viewing ideas as
other than themselves and as constituting other streams.
Vasubandhu’s view is that unenlightened people lead an enchanted life con-
taminated by selfish attachments, aversions and delusions. Enlightened people
who are detached from the objects of sense realize that what we call the world is
a fabric of appearances. They are free from desires, aversions and delusions; in
particular the delusion that one is fundamentally an enduring, substantial soul,
a ‘further fact’ over and above the stream of one’s psycho-physical continuity.
Awareness of a mind-independent physical world is the product of habitual con-
struction by ideas projecting themselves as if external. People are individualized
not through relations to external circumstances but by a ‘mind-set’ consisting of
their inherited traits, attitudes, moods, emotions and memories. Deconstruction
of such factors encourages detachment from everyday experience.
Our everyday environment and ways of life are considered a mirage conceal-
ing authentic reality. That reality is consciousness from which arises phenomenal
reality, the environments experienced by sentient beings. For the first moment
of consciousness he adopts the tradition’s expression ‘the construction of phe-
nomena’ (abhūta-parikalpa), which means the dichotomization of conscious-
ness into subjects and objects of awareness (grāhya-grāhaka-vikalpa).
The unenlightened mind is veiled by moral, emotional and intellectual
defects (kleśa), foremost among which are cravings, antagonisms and failures
in understanding. Defects spring from seeds embedded in streams of con-
sciousness. They suppress the pure factors that are conducive to salvation
and whose cultivation promotes a transformation of mind and conduct. The
Buddhist path, understanding, meditation and morality, is intended to
counteract the impure factors. The aspirant to enlightenment must focus
attention upon eliminating impurities. Internalization of the teaching that the
elements lack fixed and enduring identities (dharma-nairātmya) produces
a non-discursive, direct intuition into the nature of unconditioned reality in

which subject-object dichotomy disappears. Intensification of this intuition
destroys all the defects, together with their seeds, in a stream of experience.
We have seen the basic tenet that what are ordinarily considered to be mate-
rial objects do not exist independently of awareness. The denial of the subject-
object-relation repudiates the outlook that we are individual subjects receiving
sensory impressions from a realm of material objects that are entities in their
own right independently of constructive consciousness. He thinks that no sense
can be made of the ‘realist’ view that the perceiving mind confronts an inde-
pendently existing domain of physical objects. This is not just because thinking
about physical reality may be contrary to genuine ‘spirituality’ but because
‘atomism’ as an account of a purportedly material domain external to minds is
incoherent. Like all Buddhist thinkers, he aims to provide a rationale for why
we should not be self-centred. A structural feature of the self-centred mentality
is that it thinks in terms of discrete subjects and objects. There is an internal
relation between this mentality and supposing that people and things have
timeless essences, permanent identities or unchanging natures. It does not mat-
ter whether what are considered objects are mental or physical. The real point
of ‘perceptions-only’ (vijñapti-mātra) is to help us to internalize and act upon
the truth that there are no individual subjects of awareness confronting things,
and each other as objects. The notion of the apprehending subject is relative to
that of there being mind-independent entities. Once it is realised that what we
think of as objects are not stable external entities, our everyday understanding
of cognition is transformed. Since subject and object are interdependent, the
subjective element is also eliminated. In short, he wants to undermine thinking
in terms of the pervasive subject-object polarity (grāhya-grāhaka-bhāva, liter-
ally grasped-grasper relation) which conditions our outlook on life.
Extracts from Vasubandhu’s ‘Twenty

Verses Proving that only Mental
Phenomena are Real’ (Vijñapti-
mātratā-siddhi )
According to Mahāyāna, it is established that the three realms are only mental
(vijñapti-mātra). The words cittam (mind), manas, vijñāna and vijñapti are syn-
onyms. Cittam includes all mental events (caitta). ‘Only’ excludes material forms.
If awareness of objects required causation by external objects, dreams and delu-
sions would not be possible. Consciousness appears in the guise of external per-
ceptible objects. Ideas of objects are sufficient. They arise from a sub-conscious
store of mental seeds and traces of prior experiences. The Abhidharmikas says
that all cognitions have real objects. The first verse responds to this by saying that
some thoughts have unreal objects. The point is that reference to an extra-mental
feature is not a necessary feature of awareness. If some thoughts can have signifi-
cance in the absence of any external object, the question arises whether we are
right to assume that there are really any extra-mental items.
1. All this is just awareness, because there is manifestation of non-existent objects –

as when a visually impaired person sees non-existent cobwebs etc.
Here the realist opponent says:
2. If ideas are not caused by external objects, they would not be determined by
time and place, there would be no shared experiences, and they would not
have effects.
If there could be awareness of colour and shape in the absence of an external

object with colour and shape, it has not been caused by an external object.
So why is it produced at a specific place and not anywhere? And why does it arise
at that place at a specific time and not always?
And why is it produced in the streams of experience of all who are present
at that time and place and not just in one, just as the illusory appearances
occur in the stream of the visually impaired and not in others?
And why are the hairs and bees seen by the visually impaired are not causally effect-
ive? Things seen in dreams do not perform the functions of their counterparts in the
waking state. Fictional cities don’t do anything because they do not exist.
Hence, in the absence of external objects, spatio-temporal determination,

sharing of experiences and causally efficacy are unintelligible.
3. Spatio-temporal determination is established in dreams; evil spirits (pretas) in

immaterial hells share experiences for they all see rivers of pus.
In dreams, things are seen at specific places and times. So spatio-temporal deter-
mination is established without external objects. Beings in hell, who are there
because of similar maturation of karma, all see a river of pus. Thus is there shared
experience although the objects of awareness are not externally existing.
4. [The real is the causally effective] and there is production of effects as in

wet dreams. Again in the case of hell, all see the hell-guards and are punished
by them.
In erotic dreams there is emission of semen without intercourse. All the denizens
of hell suffer, although the guards be unreal, as a consequence of the maturation
of parallel karma.
6. If you allow that the experiences of individuals in hell are the products of
purely mental karmic traces, why not admit that this applies to all
experiences?
10. The truth is that persons lack permanent identity. Put another way, the teach-
ing that everything is ideas leads to internalization of the truth that there are
no dharmas with permanent identities.
Material substance is impossible because:
11. The object in awareness is a not a single whole. It is not a multiple composed
of many atoms. Nor is it a conglomeration of atoms. This is because the
atoms do not exist.
12. Given its simultaneous connection with six other atoms, the atom would
have six parts. If the six occupy the same place, they would have the same
mass as one.
13–15 develop the refutation of the conglomeration theory.
Thirty verses on consciousness

Vasubandhu also wrote a short treatise, called the Trim.śikā or ‘Thirty Verses’,
which is a reflection upon the structure of consciousness and the phenom-
enology of experience. He begins by saying that the words ‘identity’ (ātman)
and ‘element of existence’ (dharma) are variously applied to what are modi-
fications of consciousness (vijñāna-parin.āma). There are three types of such
modification and they constitute an unenlightened mode of living.
The first is the fruition (vipāka) of ‘seeds’ of experiences, deposited in the store-
consciousness or receptive mind (ālaya-vijñānam) by previous actions occurring in
a stream. This is called the store-consciousness. It contains experiences in the form of
implicit ideas. The receptive mind is always associated with mental phenomena
(caitta) including sensations, perceptions, attentiveness, feelings, and intentions.
Enlightenment is a transformation of the receptive mind. [3–5a]
The second is thinking about or in terms of ideas (mano-vijñānam) belonging

to the inherited repository of ideas. This is always polluted and corrupted by
four defects: belief that there is a permanent self (ātma-dr.s.t.i), delusions about
oneself (ātma-moha), an exaggerated sense of one’s importance (ātma-māna),
and self-love (ātma-sneha). This egocentric mental mode is a function of the
mental phenomena. It is transcended in enlightened being. [5b–7]
The third modality is perceptual cognitions of objects (vijñapti-vis.aya). These

perceptions are conditioned by a range of mental phenomena: sensation, desire,
memory, and reflection. They may be affected by either virtuous factors such as
faith, shame, lack of greed, hostility and delusions, energy and non-violence or by
harmful ones. [8–14]
The transformations of consciousness are mental creations (vikalpa). What is

thus constructed does not really exist independently. Everything is really just
ideas. [17] The emergent transformations of consciousness influence one
another and generate a conceptual scheme or ‘worldview’. Thanks to the
ongoing revival of vestiges of prior actions, there is a constant supply of new
experiences. [18–19]
What commonsense regards as entities or objects (vastu) are in fact constructed

by the mind. Such constructed natures (parikalpita-svabhāva) do not exist in their
own right. [20]
Mental construction is produced by causes and conditions and hence its nature is
dependent or conditioned (paratantra). But in its original, pristine self-sufficient
state, mind is not conditioned by causes. This unconditioned mode is called its
perfection (nis.panna). [21]
It is rationally undecidable whether the unconditioned mind is the same as or

different from the conditioned mind. This is because the conditioned mind, the
process of mental construction, is just a feature, qua activity, of the unconditioned
mind. Furthermore, as long as the conditioned state is not understood as such, we
can have no conception of their being an unconditioned state. Immersed in the
unenlightened mode of awareness, there is no possibility of the conditioned
mind’s realising its limited nature out of its own resources. It simply lacks the
ability to attain an external, neutral perspective upon itself. [22]
What are called the three natures are not intrinsically determined. [23]
It is a mistake to attribute a self-sufficient intrinsic identity (svabhāva) to either the

products of mental creation, to the process of mental creation or to the uncondi-
tioned mind. [24]
The permanent and true nature of the elements is purely mental. [25]
As long as one does not realize that only the mental elements of existence are
real, the subject-object mentality persists. [26]
Just confronting an object and thinking ‘this is merely an idea’ is not to experience
the mind-only state. [27]
When thought does not apprehend any objective support (ālambana), then it is
established in the mind-only state. In the absence of objects of thought, there is
no grasping. [28]
This is supernatural direct intuition, beyond the mind (citta), beyond thinking.
This is the transformation of the receptive mind, immune from afflictions and
obscurations. [29]
Vasubandhu and Sthiramati on ‘The

Construction of Phenomena’
What we see here is the view that the world of our experience is nothing more
than the interplay of ideas projecting themselves as if referring to external
things. But there are no realities other than ideas and really no subjects of
experience to which those ideas belong. A line of idealist thought, traced back
to a revelation by the Bodhisattva Maitreya to Vasubandhu’s brother Asaṅga,
says that the conditioned realm of our experience manifests ‘the mental con-
struction of the unreal’ (abhūta-parikalpa). The word ‘bhūta’ means reality
or ‘a reality’. The negative prefix ‘a’ is probably being used in the sense of
‘mistaken for’. So abhūta actually has a more subtle meaning than ‘unreal’ or
‘non-existent’. What it means is ‘what is mistaken for reality’. On this reading,
the ‘mental creation of what is mistaken for reality’ means the same as
bhāva-kalpanā or ‘the imaginative construction of entities’ where ‘entity’
means a stable object (or subject) with intrinsic properties whose identity
is determined independently of its relations to others. So abhūta-parikalpa
means that what we ordinarily understand and treat as objects and subjects are
abstractions from a matrix of relations. Also constructed and not authentically
real is the dichotomy of subject and the subject (grāhya-grāhaka-vikalpa). The
notion of emptiness (śūnyatā) is taken to mean falsity of the subject-object
polarity that structures our unenlightened understanding. Sthiramati appears
to move towards a kind of absolute idealism, according to which everything is
a manifestation of an unconditioned (śūnyatā) fundamental reality that is the
ultimate substrate of the process called the ‘Construction of Phenomena’
(abhūta-parikalpa). The unconditioned reality is the precondition of the
experiences of subjectivity and objectivity, of minds and things. As such,
it transcends them and is neither mental nor physical while manifesting
itself as both.
Illustrative extracts from Sthiramati’s

Madhyāntavibhāgat.ı̄kā
The construction of phenomena (abhūta-parikalpa) exists. Duality is not found
there. The Unconditioned (śūnyatā) is real, and the construction of phenomena
depends upon it. [1.verse 2, page 9]
Vasubandhu says that ‘abhūta-parikalpa’ means the distinction between
cognising subject and objects grasped.
Sthiramati begins by saying that the stanza is directed against the view of the
Mādhyamikas that none of the elements of existence (dharma) are realities.
In order to repudiate this universal denial it is said that ‘the construction of
phenomena’ exists.
– But does not this contradict the Buddhist scriptures to the effect that all the
elements of existence are empty?
There is no contradiction because it says that ‘duality is not found there’. This
means that the construction of phenomena is empty of the distinction between
knowing subjects and objects known. It does not mean that the Absolute is void
of intrinsic nature.
– But if duality is never real, like the hare’s horn, the constructor of phenomena
will be the only true reality. This entails that the Unconditioned is not a reality.
No, because the scripture says ‘Emptiness is real’. Emptiness here means
the absence of the opposition between subject and object in the constructor of
phenomena, not the non-existence of the Unconditioned Reality. [page 10]
Sthiramati now turns his attention to the Sarvāstivādin view that in addition to
minds and mental acts, material objects are also objective realities. The statement
that the construction of phenomena really exists is intended to refute this. There
is no matter that is independent of this process. This is why the text says, ‘there is
no duality there’ meaning that the construction of phenomena is neither the
apprehender of anything, nor is it apprehended by anyone. Moreover, it is simple
reality, void of subjects and objects.
Material objects are not grasped independently of awareness. As in dreams and

hallucinations, consciousness represents material objects. If A is the cause of B,
it is illogical that B occurs in the absence of A. [If material objects are required as
the causes of ideas, in the absence of the former the latter would not occur. But
they do.] Hence consciousness does not require external objective supports
[to represent objects]. Consciousness represents objects as if external when
subconscious ‘seeds’ [in the storehouse of ideas] come to fruition.
If there are no objects, there are no subjects. There are no objects subsisting
independently of the construction of phenomena.
– But if there is nothing to be grasped, there can be no liberation because there

would be no transcendental reality.
This is why the text says, ‘the Unconditioned exists’. The Unconditioned is the
transcendental reality. It is free from the subject-object duality. It is the foundation
of the construction of phenomena. So liberation is possible.
– But if the Absolute is the basis of the construction of phenomena, why it is not
apprehended?
It is not apprehended because it is concealed by the construction of phenomena,

not because it does not exist.
The assertion that the construction of phenomena exists can be taken as meaning
that the elements of existence are modifications of consciousness.
The denial of duality can be directed against the Sarvāstivādin view that objects
both appear and exist in their own right independently of the construction of
phenomena. [page 11]
Some think that the denial of duality is just nonsense. Others think that the empti-
ness of the elements means just that there is no controlling inner soul.
To counter the denial that there is an ultimate reality, the scripture states that the
Unconditioned exists.
– But if the Unconditioned is the foundation of the construction of phenomena,

there is the entailment that liberation would happen for everyone without effort.
No – because the Unconditioned is concealed by the process of the construction

of phenomena. Liberation is not possible until the Unconditioned is disclosed
through the great effort involved in purification of the mind.
– But if the subject-object duality is unreal, why does the deluded world think that
it exists?
The duality is like a mass hallucination produced by the construction of phenom-

ena. The Unconditioned reality is defined as that which is free from subject and
object.
– But what is the construction of phenomena?
In general terms, it is mind and the whole range of mental acts in the three
spheres of existence, past, present and future, the complex of causes and effects
constituting beginningless sam . sāra and lasting until nirvān.a [page 12]. Specifically,
it is the subject-object polarity (grāhya-grāhaka-vikalpa). The object-pole is con-
sciousness representing things and people. The subject-pole is representations in
consciousness of a self and its perceptions. An example of an object apprehended
would be something with colour, shape and size. An example of a subject would
be a visual perception. The subject-object polarity is not intrinsic to the construc-
tion of phenomena. Unenlightened people do not understand the nature of the
Unconditioned Reality because it is concealed from us. But the enlightened being
(Bodhisattva) correctly discerns that the construction of phenomena is empty of
the subject-object polarity. After the experience of duality has been superseded,
the unconditioned reality and the construction of phenomena remain. The
Bodhisattva intuits them as they are without any mistaken superimposition.
Hence everything is taught as neither empty nor as non-empty. Because of exist-

ence, non-existence and existence, this is the Middle Path. (1.verse 3 page 13)
Vasubandhu: The Unconditioned and the Construction of phenomena are not

empty but the duality of subject and object is. Everything conditioned is called
the construction of phenomena: the Unconditioned is called Emptiness. The
construction of phenomena is real. The subject-object duality is not real.
The Unconditioned is the substrate of the construction of phenomena and
the construction of phenomena depends upon it.
Sthiramati: The construction of phenomena is conditioned when it is externally

related to causes and conditions. Emptiness is unconditioned because it never
related to causes and conditions.
Consciousness (vijnāna) generates projections (pratibhāsa)of objects, living beings,

selves and perceptions. Because the objects of these ideas are unreal, the ideas are
not real either. (1. verse 4. page 14)
Vasubandhu: The projection of objects means colours and shapes etc. The projec-
tion of living beings means the sense faculties in one’s own and other experi-
ential streams. The projection of self means the corrupt mind that posits a
permanent self. The projection of perceptions means the six modes of sense-
based awareness. The unreality of objects means that they do not exist inde-
pendently of consciousness. In this sense the ideas of them are false.
Sthiramati: It has been taught that the construction of phenomena that is empty
of subject and object is real. The verse explains how the sense-faculties, objects
and perceptions are related to it.
[page 15] By statements that the construction exists, we learn that it is a reality
but nothing about its nature. We do not understand the reason for our instinctive
adherence to the subject-object polarity despite the unreality of duality. That is
why the verse indicates that the intrinsic nature of the construction of phenomena
is consciousness. The basis of our instinctive adherence to the subject-object men-
tality is the projection of objects and living beings etc. Ideas of inanimate objects,
living beings, polluted minds and sense-based perceptions are maturations of
seeds, vestiges of prior intentional actions, in the storehouse consciousness (ālaya-
vijñāna). These specific transformations of karmic potencies create the different
modes of phenomenal existence.
– But why does mind represent things as if they were external to it? We would
never mistake a post for a man if there were no men.
When the mind forms representations of objects, unenlightened people habitually

assume that the objects exist independently of consciousness. It is like people with
cataracts seeing cobwebs. In order to free people from that instinctive attachment
to objectivity, it is said that the representations of objects are only subjective per-
ceptions and lack reality as do the cobwebs. Thus it has been said that the various
manifestations regarded as constituting an objective domain depend upon the
construction of phenomena.
Coloured things with size and shape, sounds, smells, flavours and textures are
really just internal to the mind. Likewise with our ideas of sentient beings and the
thinking self.
[page 16] Perceptions appear to grasp external objects but the truth is that there
are no external objects corresponding to them.
– But common sense says that the objects of perception and the sense faculties
are mind-independent realities. Why should we reject this in favour of idealism?
There are many cases of awareness in the absence of real external objects, for
example, dreams, hallucinations, projections in meditation. If the production of
awareness were causally dependent upon external objects, ideas could not occur
in the absence of the latter and their content could not be different from the
things that have produced them [so we would see water-atoms and light waves,
not rainbows].
Hence, we say that every perception representing things and living beings arises
without any external objects. If there are no real objects, the ideas of the self and
its perceptions are not genuine either, since the two are co-dependent. Although
the subject-object polarity is unreal, the consciousness that posits it is real.
***
The construction of phenomena can be understood under three descriptions:
It is called the Constructed nature when there are objects, the Dependent nature
because of the construction of phenomena, and the Perfected nature when there
is no duality. (1. Verse 6. Page 18)
Vasubandhu: The nature that is conceptually constructed (parikalpita-svabhāva) is

the world of objects. The dependent nature (paratantra-svabhāva) is the con-
struction of phenomena. The perfected nature (parinis.panna-svabhāva) means
the unreality of the subject-object polarity.
Sthiramati: It is meant here that the construction of phenomena is in itself devoid

of the subject-object polarity. Mentally constructed means that subject
and object are treated as if truly existing, although unreal because they lack
intrinsic natures as such. Dependent means what is produced from causes and
conditions. The perfected nature is empty of duality produced by the mind. It
is unconditioned and changeless.
The subject-object polarity does not really belong to the construction of phenomena,
which is called the dependent nature when it is subject to causes and conditions. It
is called constructed when it appears under the form of subject and object which
do not really exist there. It is perfected when those conditions do not obtain. In
this way, three natures are attributed to the construction of phenomena.
[pages 20–21] Idealist objections to the Sautrāntika representationalists:
They say that the objective support (ālambana-pratyaya) of awareness is some

instantaneous reality. It effects a representation of its form in awareness. They say
that the objective realities that we perceive are either particular atoms or clusters
of atoms. But this is unsatisfactory because there are no such atoms. But in any
case, what we perceive are macroscopic objects and not atoms or clusters of
atoms. There surely cannot be such a mismatch between the contents of our
mental representations and their objective supports.
Moreover, they themselves say that compounds or wholes are purely nominal
or conceptual existents (prajñapti-sat). How then can they be the causes of
perceptions if they are themselves constructs out of experiences of the given?
Further reading
Text and translation of the ‘Twenty Verses’ in Tola and Dragonetti (2004) and Wood (1991). For the
‘Thirty Verses’ see Wood. For Sthiramati’s Madhyāntavibhāgat.īkā, I have used Pandeya (1989). The
first book is interpreted with informative notes in Stcherbatsky (1970).
Chapter IV of Paul Williams, Mahāyāna Buddhism (1989) is enlightening.

investigation
1. Why do the idealists deny the reality of matter? Is it just because they want to
encourage us to be detached from things?
2. Has Vasubandhu succeeded in showing that there is no physical reality?
7 Sām.khya and Yoga
Chapter Outline
The Sām
. khya vision 82
Causal processes 84
The human condition: bondage to natural causality 86
The Yoga vision 87
Further reading 89
The Sām
. khya vision
Sām.khya is one of the six orthodox Brahminical Hindu systems of salvation
or ‘visions’ (darśana), and it is closely associated with the Yoga system of
spiritual development. Although this tradition is ancient, its basic text is
the Sām.khya-Kārikās of Īśvarakr.s.n.a (c. 400–500 A.D.) upon which there are
commentaries including the Yuktidīpikā (c. 650 A.D.) and the Tattvakaumudi
by Vācaspati Miśra (c. 841 or 976 A.D.). Sām.khya is basically a non-theistic,
world-renunciatory and gnostic outlook, rather than a religion for the person
immersed in daily life and ritual religion. Its goal is the elimination of
suffering by the eradication of its ultimate cause. Religious practices, such
as rituals and austerities, can only afford a temporary relief from suffering.
What is required is discriminative understanding of the difference between
the conscious subject, and material nature and its manifestations. In other
words, we need to understand that the active embodied person is alienated
from its true identity, which is but reflexive static conscious subjectivity. The
goal is ‘isolation’ or freedom from determination by natural causal processes.
Sām
. khya and Yoga 83
Sām.khya posits a dualism of souls and matter. There is an infinity of souls
(purus.a), which are self-contained and inactive self-aware conscious monads
whose true mode of existence is beyond space, time and matter. Souls are
merely disinterested observers, and most definitely not active participants
in the sphere of becoming. Somehow, some of these souls have become
entangled in the material environment, including individual personality and
the body. Sām.khya and Yoga aim to free the soul from this imprisonment
by matter and rebirth.
Souls have become confused with limited and basically material forms.
When there is an association between what is merely a static conscious monad
and the material mind (buddhi), the latter is illuminated, irradiated by the
light of consciousness and becomes as-if conscious [SK 20]. The confusion is
compounded when the activity of the buddhi is mistakenly attributed to the
inactive soul. Thus we have the origins of the individual person and the series
of births marked by suffering. But the souls are really always purely passive
spectators of human experiences, abiding in splendid isolation, each illumi-
nated only by its own consciousness. It is, however, a basic tenet of Sām.khya
that the experiences deriving from involvement with matter which bind the
soul also operates for the sake of its release (SK 21).
The other pole of the dualism is Primal Matter (pradhāna or mūla prakr.ti).
It is beginningless and ever-changing. The latter spontaneously transforms
itself (parin.āma) into the real cosmos of material and psychological phenom-
ena. The best we can say is that this just happens. There is no divinity initiating
or superintending the process.
Prime Matter is said to consist of three strands (gun.a): sattva (goodness and
light), rajas (dynamic energy) and tamas (heavy and dark). (The Yuktidīpikā
interprets the triad as standing for happiness, distress and delusion.) Before
the manifestation of the cosmos, they are in a state of equilibrium, cancelling
out one another’s properties. Their ‘mere existence’ is said to prompt the
transformation of material nature. Matter (prakr.ti) transforms for the sake of
the human souls so that they have experiences that lead them to realize the
difference between soul and matter. Opponents ask how an unconscious
cause can act for the sake of anything, let alone produce specific and organized
realities. The Sām.khya position, however, is that the existence of the cosmos
call for explanation. The world consists of active and complex realities made
of parts. Each has its own purpose and we should assume a purpose for the
totality. They espouse a principle that composite entities exist for the sake of
something else (parārtha) that is different in nature from them. So it is
concluded that physical entities exist for the sake of conscious souls.
Just as the unconscious milk functions for the nourishment of the calf, so matter
functions for the sake of the liberation of the soul (purus.a). [SK 57]
No purus.a is really bound or liberated or reborn. Only matter in her various

transformations is bound etc. [SK 62]
Through repeated meditation on the nature of the manifest world, there arises
the intuitive insight that the purus.a is not the individual personality and whatever
it identifies with. [SK 64]
Primal Matter evolves to produce the basic material and psychological realities
tattva – i.e. buddhi (mind/intellect); aham.kara (one’s sense of personality;
manas (the co-ordinator of the separate sense-faculties and their deliverances);
the five sense-faculties (indriya); physical organs; the essences of sounds,
touch, colours, tastes and smells; and the gross elements – space, air, fire, water
and earth which make up physical objects). These products contain the gun.as
in differing proportions and compose the world we inhabit.
Individual objects are collections of qualities (gun.a-sam.drava) such as
colours, shapes, textures, tastes and smells. The Sām.khyas reject the Nyāya
view that there is a separate property-possessor (dharmin) that is distinct from
the conglomeration of properties. They think that once we have listed, as it
were, all the properties of an entity, there is no extra factor called the substrate.
Such would be what is sometimes called a ‘bare particular’ or an entity without
properties, and that makes no sense. It has indeed been observed that the notion
that the ultimate subject of predication should be something without properties
is an idea so absurd that only philosophers could have come up with it.
All that is required for the substantial unity of entities over time is that
they be integrated in a suitable way. As the Yuktidīpikā puts it, ‘When an entity
without departing from its nature loses an earlier property and receives a
new one, that is called modification (parin.āma)’ (YD pp. 111 and 163]. This
is true to experience. People and things change all the time and still remain
identifiably the same. There can only be change, rather than replacement, if
something stays the same.
Causal processes
Sam.khya propounds a theory of causation termed satkārya-vāda which
says that future products pre-exist in a potential state in their underlying,
Sām
. khya and Yoga 85
substrative causes (upādāna-kāran.a) prior to their actualization or manifesta-
tion (abhivyakti) as entities identifiable by their specific names and forms.
Milk transforms into yoghurt. Milk is the underlying cause or substrate and
yoghurt emerges as a product (kārya) from it. Pots are transformations of
the clay that is their substrative cause and which their individual forms
have implicitly pre-existed. Here the causal process involves a modification
(parin.āma) of a stable underlying reality and not the generation of a totally
novel product. Hence there is a strong ontological link between the emergent
effect and its causal substrate. We shall see the importance of this emanative
model of cosmic causality for those forms of Vedānta that see the cosmos of
souls and matter as real transformations of the divine being.
The Sām.khya theory of causation develops in opposition to that of the
Nyāya-Vaiśes.ikas. That position is called asatkārya-vāda or ‘the production of
something new’. This says that prior to origination, the effect did not exist in
its underlying cause but is a totally new product, different from the already
existent basic elements out of which it is made. They reject the category of
potentiality, holding that only what is actual and concrete is real and can cause
something else. Causation is not the actualization of what was potential but
the generation, through re-arrangement, of new entities out of already existent
factors. A cause is defined as a necessary prior condition of an effect. There are
three factors in a causal complex such as the manufacture of a cloth by a weaver
out of threads: the substrative or underlying cause (samavāyi-kāran.a) which is
always a type of substance (dravya) – e.g. the threads comprising the cloth
(the new whole – avayavin); the non-inherent cause (asamavāyin) which is
always a quality (gun.a) or activity (karma) – e.g. the weaving and colour of
the threads; the efficient or instrumental cause (nimitta) – e.g. the shuttle
and other instruments. The weaver is the agent cause. The products of causal
processes are integrated wholes (avayavin). The whole is a new creation with
its own identity, over and above the sum of the parts in which it inheres. The
whole entity cannot exist without the parts, but the parts can exist without the
whole. It is distinct from the parts since it manifests a single specific universal.
An individual object must be the substrate of a universal; such as cowness
or potness – a collection of different parts will not suffice. That the whole is
not reducible to its parts is crucial to the Nyāya-Vaiśes.ika resistance to
the Buddhist reduction of objects to constituents and phases because they
explain endurance through space and time in terms of integrated natures
that are held together by the relation of inherence (samavāya). The Nyāya-
Vaiśes.ikas adduce a number of reasons for their view that prior to origination,
the effect does not exist in its underlying cause. Some of them are:
(a) The effect was not perceived in the causal substrate.

(b) If the pre-existent effect lacks specifiable properties, it is not identifiable and thus
its existence does not fall within the province of inference.
(c) The agent’s efforts would be superfluous if the effect already existed.
(d) A pile of threads is not called ‘cloth’ and vice versa.
(e) Difference in function of causes and effects: a lump of clay won’t carry water.
(f) Difference in form or shape of causes and effects.
(g) Number: threads are many, the cloth one.
The human condition: bondage to

natural causality
We said above that some souls have become entangled in and misidentify
themselves with aspects of the material environment, in particular psycholo-
gical faculties and events, and the body. The process occurs when the mind
(the buddhi), a material product, captures the reflection of the light of some
consciousness. The conscious spirit is then confused with some organic material
configuration. We only function as individual conscious agents and experiences
when conjoined with a body and psychological apparatus. We engage with the
world through the operations of the physical buddhi. Immersed in daily life,
where our natural drives and the acquisitive mentality encourages us always to
be moving on, satisfying our interests and achieving our own purposes, we
generate karma that necessitates further births in the here and now.
The Sām.khya distinction consciousness as the transcendental presupposition
of experience and consciousness as a stream of psychological events – cognitions,
thoughts, feelings and desires – will become influential. Consciousness is
constitutive of sentient beings, but sensory activity, perceptual cognitions and
consequent conceptual thoughts that come and go are psychological functions
that properly belong to the material mind and sense-faculties. Sām.khya-kārikā
5 says that perception is a judgement (adhyavasāya) about each of the sensory
faculties’ specific objects. Sām.khya-kārikā 23 says that judgement is a function
of the physical buddhi. The Yuktidīpikā commentary elaborates: definite
awareness is a conceptual apprehension involving a propositional belief
such as, ‘This is a cow’ or ‘this is a man’. Primary experience is a function of
the sense-faculties that assume the form of external objects. A thought such
as, ‘This white cow is running’ is a judgement based on sensory deliverances.
Sām
. khya and Yoga 87
Primary experience, mere observation or just seeing something, is restricted
to the present time, but perceptually based thoughts and judgements can
range over past, present and future. Sense-based primary experience is
not conceptual. Concepts belong to the buddhi, which is able to discover
generalities, and is unrestricted where its objects are concerned. But although
the mind apprehends the forms of objects once they have been grasped by
the senses, being physical it requires illumination by the consciousness that
it has borrowed if psychological events are to mean anything. Hence we need
to posit the conscious principle as the ultimate source of experiencing.
But that principle is merely an observer rather than an active participant
in experience. Such is the vision of the world-renouncer. This is of course
problematic because if the process by which the souls become enmeshed
in physical conditions is a purely mechanical and automatic one, and since
prakr.ti and its works are eternally active, it is hard to see why it should not
afflict the released soul again.
Liberation (kaivalya – ‘wholeness’, ‘isolation’) from the cycle of becoming
and rebirth (sam.sāra) results from the discriminating insight, presupposing
the discipline of yoga, that the purely conscious and inactive soul is distinct
from both the physical and psychological spheres that are the evolutions
of material nature. Prakr.ti then ceases to function in relation to the enlight-
ened centre of consciousness. Liberation occurs when the three gun.as are
reabsorbed into prakr.ti, whose functions cease. The spectator-soul recovers
its true form, detached from mental modifications and other features of
embodiment. Knowledge is enough to effect the soul’s disengagement from
the environments of experience.
The Yoga vision

Yoga accepts the Sām.khya metaphysic, but the Yoga tradition has its own
identity. The foundational text is the Yoga-Sūtra of Patañjali, variously dated
from the second century B.C. to fourth century A.D. The commentary by
Vyāsa is probably a work of the sixth century. Vācaspati Miśra (950–1000)
wrote a commentary called the Tattvavaiśaradi. The Yogasūtrabhās.yavivaran.a
is probably much later.
In Indian culture, any discipline of physical and mental self-cultivation and
self-transformation whose aim is that of freeing us from rebirth is called ‘yoga’.
The Yoga-Sutra defines its subject as the restraint and suppression of all mental
modifications – that is to say, all forms of thought and feeling, whose forms
the soul has assumed. The goal is a disengagement from the life of action in
which the soul recovers its true nature. This is achieved by constant contem-
plative practice and detachment, presupposing mental and moral cultivation.
Detachment is said to be self-mastery on the part of one who no longer
thirsts for perceptible objects or any of the transitory goals promised by the
Scriptures. Active yoga consists in austerities, the recitation of mantras and
the study of scriptures bearing on freedom from rebirth, and the direction
of the mind to Īśvara – an exemplary soul (purus.a) who has transcended the
mutually dependent factors of karma and what are collectively termed ‘the
afflictions’. The latter are ignorance (a failure to discriminate what matters
from what does not, the morally valuable from the corrupt, and what is one’s
true identity from one’s personality and everyday identifications), selfishness,
desires, animosity and attachments. The discriminating person has realized
that everything is unsatisfactory because pleasures turn into frustrations,
because of the weight of dispositions inherited from previous lives over which
we have no control, and because our minds are always restless and at war
within themselves.
The soul in its pure form is mere non-intentional awareness. When
implicated in the conditions of space and time, it has an observer’s perspective
on of the thoughts and feelings that are functions of the embodied mind with
which it is associated.
The eight stages of the physical, moral and mental discipline of classical
Yoga are:
Self-restraint; non-violence, honesty in thought, word and deed, sexual restraint and
lack of greed.
Discipline: interiorization, tranquillity, asceticism, mantra recitation, the study of texts
on liberation and attention to God.
Physical postures; exercising control over the psychosomatic complex.
Breath-control: regulation and reduction of the processes of inhalation and exhalation
that increase psychophysical control.
Withdrawal of the senses from their objects and direction of attention to the inner
self.
Attention: focusing the mind on a single point (i.e. an object of meditation).
Meditation: the uninterrupted continuity of awareness of the object of meditation.
Profound contemplative introversion in which there is no self-awareness.
Mental purification coincides with purification of the soul. The state of

liberation from rebirth is one of wholeness and isolation (kaivalya) where
Sām
. khya and Yoga 89
consciousness experiences only itself. It occurs when the constituents of
material nature no longer operate in relation to the individual centre of
consciousness. The soul recovers its true form, disjoined from mental and
physical modifications.
Further reading
For the Sām.khya-kārikas see Larson (1979).
Larson and Bhattacharya (1987) has a useful introduction and summaries of works, including the
important Yuktidīpikā, which is edited in Wezler and Motegi (1998).
For the Yoga-Sūtras and commentaries see Woods (1927) and Whicher (1998).
Chapter 11 of Halbfass (1992) is valuable for Sām.khya.

investigation
1. Can we make sense of there basically being more than one centre of pure
consciousness?
2. Does it make sense to hold that material nature operates for the sake of the
purus.as?
3. Can the purus.as be finally released from rebirth?
8 Nyāya and Vaiśes.ika
Chapter Outline
Nyāya 92
Vaiśes.ika 92
Metaphysics: the system of categories (padārtha) 93
Substances: the category dravya 93
The category Gun.a (quality) 98
The category Karman (motion) 100
The category general property (Sāmānya) 100
The category Viśes.a (ultimate particularity) 103
Samavāya (the inherence relation) 104
The category Abhāva (absences) 105
Epistemology: the Pramān.as 105
Knowledge by perception (pratyaks.a) 109
Anumāna: knowledge by reasoning or inference 111
Śabda: testimony and the transmission of true information 113
Words and sentences 114
Further reading 115
Nyāya-Vaiśes.ika is a form of direct realism about an objective mind-independent

world of enduring objects, properties and relations that shape our thought.
The tradition holds that when we are thinking rationally, reality is fully deter-
minate relative to our concepts and language. It is one of the Brahminical lines of
a defence against the Buddhist reduction of objects to temporal phases and their
rejection of permanent structuring principles such as universals, kinds and
Nyāya and Vaiśes.ika 91
enduring bearers of experiences. Their metaphysic is what is sometimes
called a ‘substance-ontology’ according to which the world consists of endur-
ing individual entities that are the bearers of universal and specific properties.
It is the interactions of these ‘basic particulars’ that generate processes and
events. By contrast, a Buddhist event or process ontology says that what is
basic is a system of relations and not individual objects. There, everything is
interdependent (pratītya-samutpāda) and causality just means that Bs follow
As. There are no stable persisting identities (nairātmya). Nothing really
lasts (ks.an.ikatva). No universals run through the whole of reality. Thinking
in terms of individual entities is mental construction out of what is in fact
a fluid process. It is to such an outlook that Nyāya-Vaiśes.ika is opposed.
For them, reality is an intelligible framework of stable structures and persist-
ing individual entities. For centuries Nyāya-Vaiśes.ika thinkers formulated
metaphysics and epistemologies that were designed as bulwarks against the
anti-substantialist Buddhist outlook.
Nyāya direct realism says that cognitions are informational states that
depend upon and disclose objects in the world (artha-prakāśa-buddhi). It is a
form of externalism – the view that mental events are necessarily, and not just
causally, dependent upon external objects. Our thoughts and representations
are not a veil intervening between cognizing subjects and the world. We are
always and already ‘outside ourselves’ with the entities that we encounter and
which belong to a world already discovered and invested with meaning. The
world is as it appears to commonsense: stable objects in a three-dimensional
spatial framework. The Nyāya-Vaiśes.ikas believe that their system of categories
captures the structure of reality as it is in itself. They accept a correspondence
theory of truth according to which there is a structural isomorphism between
true thoughts and states of affairs that obtain. Genuine universals (jāti,
sāmānya), qualities (gun.a) and relations determine the phenomenal, causal
and logical organization of the world of individual substances (dravya). They
fix the actual states of affairs that obtain as the world. Real properties that
qualify objects, and real relations (samavāya and sam.yoga) play a basic role
in the objective, non-arbitrary classification of objects.
Actually there are two traditions here, but they were never far apart. Nyāya
was originally concerned with epistemology and valid reasoning. It describes
itself as an investigation of matters established by means of knowing, including
inference that is based on perception and testimony. Vaiśes.ika was originally
more concerned with metaphysical questions about the basic constitution
of the cosmos and established the system of categories that we shall shortly
describe. By the time of Udayana (c. 1050–1100 A.D.) they had coalesced
and what follows is a synthetic overview of some of their concepts and
categories. Any attempt to delineate our world-view in a categorial scheme
will run into problems. Inconsistencies are bound to arise. The Navya-
Nyāya thinkers, the most influential of whom are Gangeśa and Raghunātha,
attempt resolutions of some of the problems and introduce clarifications
and innovations. Their writings are very difficult and I have not attempted
to say much here.
Some people and their works:
Nyāya
Gautama Aks.apāda (c. 150 A.D.), the author of the fundamental Nyāya-Sūtra.
Vātsyāyana (350–400 A.D.) author of the Nyāya-Bhās.ya on the Nyāya-Sūtra.
Uddyotakara (550–600 A.D.) author of the Nyāyavārttika.
Jayanta Bhat.t.a (850–900 A.D.) author of the Nyāyamañjarī.
Vācaspati Miśra (A.D. 950–1000) author of the Nyāyavārttika-tātparyat.īkā.
Bhāsarvajña (900–950 A.D.) author of the Nyāyasāra and Nyāyabhūs.an.a.
Udayana (c. 1050–1100 A.D.) author of the Laks.an.āvalī, Ātmatattvaviveka, Nyāyakusumañjali and
Kiran.āvalī.
Gaṅgeśa (c. 1300 A.D.), author of the Tattvacintāman.i.
Raghunātha (1475–1550 A.D.), author of the Padārthatattvanirūpan.a and Dīdhiti on the
Tattvacintāman.I.
Vaiśes. ika
Kan.āda, Vaiśes.ika-sūtras (first century A.D.?)
Praśastapāda (c. 500 A.D.), author of the Padārthadharmasam.graha.
Śrīdhara (fl.991 A.D.), author of the Nyāyakandali on the preceding.
Early Nyāya says that we reach the Highest Good by understanding the truth
about the reliable methods of knowing (pramān.a), the knowable objects
(prameya) and the various forms of argument and debate. In other words, the
truth (pramā) will set us free. Freedom here results from the elimination of
misconceptions, activities, rebirth and suffering.
We will begin by looking at the knowable objects and their organization in
a system of categories (padārtha) and then move on to a consideration of the
methods of knowing.
Metaphysics: the system of

categories (padārtha)
Anything falling under one of the categories is existent, and in principle
knowable and nameable.
The categories are:
Dravya – Substances or enduring objects

Gun.a – Qualities
Karma – Actions/Motions/Movements
The above three are all categories of particulars that fall under the universal
properties inhering them.
Sāmānya – General properties encompassing both real universal properties and

kinds (jāti), as well as certain imposed or imputed properties (upādhi). Only jātis are
genuinely real padārthas. By contrast, imputed properties or upādhis are just
concepts.
Viśes.a – Unique Particularities
Samavāya – The relation of inherence
Abhāva – Absences
Substances: the category dravya

Substance is defined as that which lacks the constant absence of qualities and
motions. Some substances are simple (e.g. souls) while others are integrated
wholes, consisting of parts. Each is a manifestation (vyakti) of a genuine uni-
versal property. They occur as natural or artificial kinds. Some are products
and thus perishable, while others are eternal. At the first moment of its exist-
ence, a substance has no qualities or actions. But it has a universal property. An
individual cow is nothing unless it is an instance of the generic property
‘cowness’.
Earth, water air and fire are basic substances. Combinations of their atoms
in varying proportions constitute non-permanent physical entities. Atoms
are uncreated, indivisible and indestructible. If there was infinite divisibility
a mountain would have the same size as a mustard seed! The cosmos
comes about when the godhead imposes repeatable structures upon these
raw materials.
The atmosphere is a substance. It occurs everywhere. It is the medium

through which sounds are transmitted. It is single and thus does not manifest
a genuine universal property.
Time and space are substances. In themselves they are single, eternal and
omnipresent. Space explains the structuring of our experience in terms of
right and left, up and down, east and west and distance and proximity. Such
experiences are grounded in stable objective structures independent of our
minds. Although time is single, we interpret it in terms of conditions (upādhi)
such as past, present and future, earlier and later and its passing slowly or
quickly. But these concepts do not really affect it.
Where these systems are concerned the word ‘Ātman’ might be best
translated as ‘principle of identity’ but I shall use ‘self ’ for short. It is a type
of substance that is the ultimate subject of cognitions, emotions, desires and
efforts, as well as the bearer of good and bad karma. The inclusion of selves
in the category of substances indicates that the concept was a naturalistic
one, radically different from the reflexive centres of awareness proposed by
the Vedāntins, the Sām.khyas and the Śaivas. Selves are eternal, everywhere
and essentially distinct from each other, but each is embodied differently and
has a series of life-histories through time. Selves are not essentially conscious,
although consciousness is their distinctive property. Cognitions, as well as
desires, feelings, intentions and personal characteristics are contingent prop-
erties that occur only when a self is localized in an objective environment.
Selves are non-physical and non-conscious principles of identity that must
be posited to explain our abilities to remember and to synthesize the present
variety of experiences into a unity. A life is a series of embodied experiences,
but something supra-experiential is required to impart unity and coherence to
the stream, indeed to constitute the manifold of experiences as one stream.
That something is the principle of identity.
The existence of such a principle of identity is established by inference. Not
being of the nature of consciousness, it cannot reveal itself and cannot be
known by introspection. The subject is not manifest in conscious acts, which
are entirely intentional or attentive to objects outside the mind. An argument
for the existence of the enduring self is that thoughts, desires and acts of will
are qualities and cannot stand alone: there must be a continuous subject to
which they belong. It is also argued that the pronoun ‘I’ must stand for some-
thing. It cannot be the body, because use of the expression ‘my body’ indicates
that it belongs to something else.
There are other arguments for the existence of an enduring principle of
identity in the face of the Buddhist reduction of personal identity to a causally
connected stream of embodied experiences. There are a number of the
background assumptions to be borne in mind. What one remembers depends
on what one perceived or learned earlier. One does not remember the experi-
ences of other people. The Brahmins insist that personal identity must be
a further fact over and above a stream of connected experiences. It is that
which underlies the stream of personality or individual character. It is
that which both relates experiences occurring at different times and is the
precondition of the coherence of the manifold of present experience. For
the Nyāya thinkers, this further fact is not essentially conscious. (Neither
the Brahminical nor Buddhist traditions think that the identity of a person
can be purely physical.)
One argument for persisting identity is that there must be a suitable vehicle
for the preservation of memory traces of previous experiences. Something has
to hold the stream together. Bundles of perceptions have to be bound by some-
thing. There is also the argument from experiential memory – remembering a
state of affairs in which one has participated. Sometimes I can recover the
sense of what it was like to enjoy a past experience: it is like being there
once again. For this to be possible, there has to be some factor that connects
present experience and recovered memory, namely the subject to which each
belongs. There is also the argument from recognition. I can only recognize
the building that houses the Liverpool University Philosophy Department as
the same one that I saw yesterday if the subject of yesterday’s experience is the
same as today’s. Also, the capacity to unify perceptions belonging to different
sensory faculties requires not just a single object but also a perceiver who
integrates them. I can simultaneously see, smell, taste and feel an orange. The
experience is given as a unity. But each sensory faculty has a distinct sphere of
operation. Some other factor must be unifying the perceptions. That some-
thing is the self, in association with the mind (manas). There is the argument
from the desire to enjoy again something remembered. I see an orange and
really want to eat it because I recall its delicious flavour. This complex mental
synthesis (pratisam.dhāna) must belong to a single subject who synthesizes
diverse experiential events (present cognition of the object, previous cognition
of the object, the recollection of the pleasure it gave, desire to enjoy it again).
They cannot synthesize themselves if each is momentary, confined to its
own sphere and is thus not cognizant of the others. But if it is allowed that the
mental synthesis holds merely between the cognitions, they would become
assimilated to each other and we would no longer have a series of discrete
individual factors constituting the stream of experiences. In a case like this
we need some further factor that explains unification across sensory faculties,
the connection through time between my present sight of the orange and
my recognition of the orange as similar to one previously tasted, and the con-
nection between the earlier experience, the present desire and the anticipation
of the experience of eating the orange. It is significant that we do not have
to connect these experiences with a single subject. The ‘I’ is given as implicit
in them. To turn Hume against himself, ‘I can never catch myself at any time
without a perception’.
The Buddhist may say that cases of mental synthesis arise in virtue of the
cause-effect relation between momentary cognitions in a single series and so
there is no need to posit a single cognizer. The response is that even if there
can be a cause-effect relation between different momentary mental events,
this would not explain the phenomenon of memory. For a mental event to
be experienced as a memory, there has to be an enduring subject. A prior
experience of pleasure in a stream would be precisely that. It would not
be remembered from the present perspective as what had been a pleasant
experience. Just as one person does not remember what was experienced by
another, so one discrete cognition cannot remember the content of another.
The Buddhist attempt to explain all of the above phenomena in terms of a
stream of experiences impersonally related as cause and effect is not really
satisfactory. There has to be something that holds the stream together, some-
thing that, as it were, underpins the causality. It is not clear that the Buddhist
account can explain synthetic experiences (pratisam.dhāna) if they understand
one mental event in a series as the cause of the next only in the sense that one
precedes the other. But such a relationship can occur between events occur-
ring in different series. Something stronger than causality is required – namely,
the mental events are all related to something else.
There has to be something in relation to which an experience is identified
as past, another present and another anticipated as future. Something has to
unify the manifold of current experiences. Experience flows: we are always
aware of it changing through time. But change is only possible if there is
some stable factor that undergoes it. The Buddhist is actually talking about
successive replacements rather than change. An enduring principle of
identity, the subject of experiences, is the most economical explanation of
the phenomena.
The reductionist theory that psychological continuity is nothing but a basic-
ally impersonal continuum of causes and effects has a persistent allure, despite
its physicalist and functionalist animus. But there are some considerations that
should make us think twice about the explanatory sufficiency of causality here.
It is uncontroversial that causal relations sometimes hold between experiences.
I burn my hand on a hot pan lid, cry out and my wife is alarmed (perhaps . . .).
But the experiences here do not belong to a single person. It may perhaps be
the case that a causal account can be given of the formation of personality
or character when this is regarded as in some sense an achievement. But a
problem is that a lot of fragmentation or disintegration of experiences over
time is consistent with their still being a single person. (Less dramatically, as
the reader knows, my mind flits from one thing to another quite at random.)
As Bishop Butler said to Locke, a person is more than what they remember.
It might be possible to give an objectifying purely causal description of the
psychological processes of an animal that is not a person – one that cannot
use the pronoun ‘I’. In the case of a personal individual stream of experiences,
it is plausible to say that sometimes a present experience (causally) elicits
a memory. But the reductionist account says that the relations between
experiences within a single stream (‘series person’) are all causal ones. This
is questionable. It is not clear that in every case of experience B following
experience A, the relation is causal. But the basic being of the subject (that
which is meant by ātman) is a given. It is not a product or an achievement.
This basic subject is expressed in first personal ‘I-thoughts’ and the relevant
connections here are not causal ones. The first-personal, autobiographical
continuity of human persons is not given in causal terms. If I think, ‘I was born
in Heswall’ and ‘I now live in Liverpool’, I can truly conclude, ‘I was born
in Heswall and now live in Liverpool’. That conclusion has nothing to do with
causation, but is a matter of inferential entailment. The inference is valid
because the two features are states of the same person. So it is not obvious that
the unity of a person’s mental life is to be explained in causal terms. Causal
connections between experiences are insufficient to constitute or produce a
continuum. The connections hold because of the character of the relation
between thoughts, and that relation derives from their being states of the
same person. The reductionist view is that personal identity is an illusion
that is constructed out of experiences. We then mistakenly invest this illusory
identity with an enduring character. The basic objection is this: how can
construction, mistaken or otherwise, happen if there is nothing capable of
doing the constructing in the first place?
Because having experiences is an accidental product of being embodied in

some environment, the released state is one of unconsciousness. Unappetizing
as this sounds, the Nyāya-Vaiśes.ika thinkers observe that at least you can’t want
anything, and that there is no such thing as pure happiness since happiness is
always pervaded by fear of its loss. Later writers, perhaps under the influence
of forms of Vedānta, mollify the position and say that the released state is,
somehow or other, blissful.
Mind (manas) is a faculty that is instrumental in our having sensorily-
derived thoughts, feelings, desires and intentions. Each embodied soul, the
vehicle of experiential continuity, is contingently connected with a mind.
The sensory receptors transmit a range of information about the objective
environment to the manas which operates as a central processor co-ordinating
that information and selecting what is relevant to the present state of the
organism. Since the operation of each of the senses is restricted to its own
proper sphere, another factor is required to explain the co-ordination of
sensory experience, so that we can simultaneously touch what we see.
The manas is instrumental in the conversion of some stimuli into feelings,
the translation of some items of cognitive input into conscious thoughts with
practical applications (storing some as memories), and the transformation of
some affective responses into acts of will. Thoughts, feelings and intentions
thus become temporary properties attaching to the permanent soul-principle
and a centre of knowing, agency and experience is created.
The category Gun.a (quality)

These are characteristics belonging only to substances or enduring objects.
Substances can undergo changes respect of their qualities and still remain the
same. The loss or gain of an integral part, however, results in a new substance.
Some are sensible properties: colours, tastes, smells, touch, sounds (each
exclusively related to the appropriate sense-organ). Other natural features
include: dimension, temporal and spatial relations of proximity and remote-
ness, weight, fluidity, numbers, as well as the relations conjunction and its
counterpart disjunction. Others are non-physical properties specific to
embodied selves: cognitions, pleasures, pains, desires, antipathies, conscious
efforts, spiritual merit and demerit and inherited character traits.
Gun.as are unrepeatable particular occurrences. The blue specific to my
shirt is a different instance from the blue specific to my pen, although
the two shades may be identical. Redness is not a quality but the universal
(jāti) common to all instances of red. Since a colour pervades the whole
substance in which it inheres, seeing an object’s colour implies seeing the
entire object.
Numbers are properties only of substances. But every entity has the number-
quality ‘one’, which is treated as a special case. In each pair of objects there
is the specific quality of two ness, although neither is double. (Duality is a
universal inhering in each case of two ness.) All larger numbers relate to
collections of objects in a special way by a relation called ‘encompassing’
(paryāpti). They are trying to do justice to everyday expressions like, ‘the
number of trees in the forest’. Whatever the number is, it does not apply
to each of the trees individually. They did not consider the possibility that
numbers apply to concepts.
If qualities only belong to substances and numbers are qualities, there can-
not be a number of numbers or of any other quality. Nor can we consistently
say that universals are one. In response to this, the Navya-Nyāya thinker
Raghunātha says that numbers belong to a separate category of their own.
Indeed Bhāsarvajña had said that numbers are not qualities, but relations of
identity and difference. What this means is that when we say that a jāti is one
or that the atmosphere is one, we are not attributing a property. Indeed, the
only attributive use of one is to say that something is integrated or that it is
identical with itself. When we say that a jāti is one what is meant is that it
is unanalysable.
Conjunction (sam.yoga) is a temporary relation of contact between two
separate objects.
Cognitions (buddhi, jñāna, upalabdhi) illuminate objects, external to minds.
They are functional psychological properties, playing a role in interpersonal
communications. The inclusion of cognition in the categorial schema is a fea-
ture of the naturalistic worldview, which denies that consciousness is abso-
lutely fundamental. Cognitions or informational states are particular episodes
targeted on the particular objects or states of affairs that cause them. As well as
current engagement with the environment (anubhava), there is memory or the
retention of information that is produced from stored traces of prior cogni-
tions of objects and is elicited by present ones. Their realism is what is some-
times called ‘externalist’ in that they say that the occurrence of a cognitive
episode is necessarily and not just causally dependent upon the subject’s being
placed in an objective environment. They recognize that there are many psycho-
logical episodes that pass unnoticed. Sometimes we are on ‘automatic pilot’.
That is to say, information may be received from the environment via the
senses, processed by the mind (manas) and stored in the memory without
being consciously registered. The introspective elevation to explicit awareness
of an item of informational input is termed anuvyavasāya. This happens when
one cognitive state becomes ‘telescoped’ by another. They do not accept that
cognitions about the environment are intrinsically reflexive or self-illuminating.
An objection to this is that if the elevation to consciousness of an informa-
tional state requires another psychological state, an infinite regress results.
A cognitive episode qualifies as a piece of knowledge rather than an infor-
mational state or mere true belief when it is produced by one of the reliable
methods of knowing (pramān.a). Present experience (anubhava) may be true
or misleading. Perceptual knowledge is direct contact with an object or
state of affairs. A true cognition is one in which the attribute that is possessed
by the external object is a feature of the content of the cognition. A false aware-
ness is a situation which the actually present external object does not have the
attribute that is a feature of the content of cognition, as when we think, ‘This is
silver’ in relation to a piece of shell. The awareness refers to real silver existing
elsewhere and not to a mental entity such as an hallucination or to a ‘mere
appearance’. Falsehood is exposed in practical failure. The tradition develops a
sophisticated form of a direct realist epistemology holding that there is a
structural isomorphism between the complex content of a true cognition and
an objective state of affairs.
The category Karman (motion)

Acts characterize substances and are always transient. They include the
varieties of motion, contraction and expansion. They cause conjunctions
and disjunctions.
The category general property

(Sāmānya)
This category includes objectively real universal properties (jāti – literally
‘kind’) that exist independently of their instances. There are also ‘imposed’
or ‘imputed’ properties, termed ‘upādhi, and these are understood as used
arbitrarily to group a plurality. They are just concepts. Genuine universal
properties (jāti) are manifest in what we would call natural and artificial
kinds. ‘Cowness’, ‘horseness’ and ‘being a Brahmin’ are examples of the former.
‘Potness’ is an example of the latter – individual pots manifest a permanent
and stable underlying structure. Jātis are manifested in their particular sub-
stances, qualities and motions but they exist eternally and independently of
their instances, which come into being, change and last for a while. Jātis exist
objectively, independently of human thinking. They feature in our modes of
thought but they are not thereby created. They are discovered not invented.
They are unitary and unanalysable (eka), eternal and permanent (nitya) and
occur in many manifestations (aneka-vr.tti). They occur in substances, qualit-
ies and motions by the relation of inherence (samavāya) but nothing inheres
in them. A genuine universal is the unitary property shared by all members
of a kind or class. The world consists of objects belonging to kinds. It is not a
collection of unique individuals. To identify an object, we mention what kind
of thing it is. (It is perhaps worth mentioning the view that there are objects
belonging to kinds and falling under concepts but there are no things. There
is no answer to the question, ‘how many things are there in this room?’ as
opposed to the question, ‘how many tables and chairs are there?’)
We explicitly recognize a real general property when we have seen a number
of objects belonging to the same kind. But even though we may have seen only
one elephant, we have still perceived the property ‘being an elephant’ even
though we do not yet know that the individual belongs to a natural kind. This
is a form of extra-ordinary intuition (alaukika-pratyaks.a) that relates to real
universals (sāmānya-laks.an.a). According to this theory, universals are not
innate ideas. Universals are not ideas abstracted from sensory impressions.
Nor are they general concepts formed by a process of induction. Universals are
objective realities, part of the furniture of the world that are discovered by
non-sensory intuition.
A universal is said to pervade the particulars in which it inheres. It produces
a concept of its own nature in respect to one or more objects. When we
have an inclusive cognition of a group of objects, some of which may be
remembered, it is the universal that causes the recurrent cognition.
Udayana says that universals regulate causality. Causal regularities hold
because causal relations obtain not merely between particulars (the Buddhist
view) but between particulars in so far as they belong to kinds. Kinds are
self-reproducing – dogs give birth to puppies and acorns produce oak trees.
Also, the regulative causality intrinsic to kinds imposes limits on the changes
that entities undergo. I can become heavier or more short-sighted, but I
cannot become a lion without ceasing to exist.
Universals explain our use of general terms, of which they furnish the stable
grounds for the repeated applications. The meanings of words are complex: a
word such as ‘cow’ signifies the concept cowness, it produces an image of the
form typical of cows, and it can denote an individual cow. Universals explain
why we understand entities as belonging to kinds. We can only identify
an entity as being similar to another entity if there is some objective basis
for that similarity. If everything were utterly unique, we could not relate the
contents of our cognitions to each other. It is an illusion to suppose that
we could re-identify anything if particulars did not manifest shared features.
Re-identification of a particular as belonging to a kind to which others
also belong presupposes that kinds are given and not merely conceptually
fabricated in accordance with human interests. The success of most of our
activities shows that the concepts and categories included in the padārtha
scheme are not more or less arbitrary inventions. There have to be universal
principles running through reality. It is not enough to say that we recognize
similarity of the basis of clusters of similar observable features. Particulars
differ and observable features are promiscuously distributed. An albino tiger
is still a tiger. Something else that combines the clustered features must be
posited and that extra factor is the unitary universal. Many universals are
manifested in concrete shapes (ākr.ti) that are specific and regular arrangements
of parts characteristic of a kind. When we see an individual cow that is a mani-
festation (vyakti) of the universal cowness we also see the universal by virtue
of the shape. The perceptible shape cannot be identical with the universal
because it is a collection of features that are integrated by the universal.
Not every characteristic that we understand as common to a group of
individuals is a genuine universal (jāti). We can divide the world up in all
sorts of ways, generating as many properties as suit us. We can speak of ‘the
community of cooks’, but ‘being a cook’ is not a jāti but an imputed property
(upādhi). ‘Being able to swim’ is an ability possessed by some in varying degrees
so it cannot be a unitary single property. Another example is ‘being a beast’,
which covers many different animals. This is a compound imputed property
that is a synthesis of a number of features in this case being hairy, having a
tail and having four legs. It is artificial in that it cuts across natural kinds and
violates the natural order of classification.
The genuine properties are objective and natural. They are intrinsic
features of the instances where they occur. They carve nature at the joints.
They are discovered, neither manufactured nor invented nor conceptually
constructed. The classes of their instances are not miscellanies. Some of our
concepts, such as ‘horse’ and ‘blue’ coincide with the genuine properties and
qualities immanent in the cosmos and regulative of its actual conditions. But
the imputed properties are just useful concepts.
Udayana formulated six impediments that prevented an upādhi from being
a jāti:
1) Unity: There is only one atmosphere, so ‘atmosphericity’ does not occur

in many.
‘Devadatta-ness’ (if attributed to the individual man called Devadatta) is also a
simple imposed property.
2) Different names and concepts may signify the same things or properties,
but they do not thereby generate different objective realities. There are at least
two words that mean ‘pot’ but they do not signify different kinds of pots.
Triangularity and trilaterality do not name two distinct universals. No two
universals can inhere in the same things.
3) No two universals partially overlap with each other in their instances. There
can be no ‘cross-cutting’. This rules out ‘beastness’.
4) No universal generates an infinite regress. Universals do not inhere in
universals. In other words, there is no cowness-ness inhering in cowness.
5) No universal can destroy the nature of that in which it inheres. Although there
is an infinite number of unique particularities (viśes.a) individuating atoms,
souls and minds, there cannot be a real universal ‘viśes.a-ness’ because the
categories jāti and viśes.a are mutually exclusive.
6) A universal must be capable of inhering in its instances. Every universal occurs
in its instances by the relation of inherence. There cannot be a universal ‘inher-
ence-ness’ common to all these relations. Inherence cannot be related to itself
by the relation that it itself is.
The category Viśes.a (ultimate

particularity)
These are unique features of the simple eternal substances (atoms, atmosphere,
time, space, souls and minds) distinguishing them from each other. Whereas
complex entities are differentiated by the different arrangements of their parts,
eternal substances are partless. So each must have its own individualizing
feature. They account for the unique identities of these kinds of entity. The
viśes.a belonging to a soul is what differentiates it from every other soul. So
released souls that have neither karma nor bodies remain different.
A problem here led to the ultimate abandonment of the category by Navya-
Nyāya. If the viśes.as are themselves distinguished by other individuators,
there is an infinite regress. If they do not need to be distinguished by other

individuators, they are unnecessary for individuation.
Samavāya (the inherence relation)

Nyaya posits a universe consisting of innumerable objects and structuring fac-
tors. Enduring substances, qualities, movements and universal properties are
all counted as entities. There has to be a sort of relation that can combine such
realities into complexes while preserving the differences between types.
Samavāya is the relation by which types are held together while retaining their
own identities. It is the cement of the universe. It integrates the constituents
comprising particular objects. It combines two items when one is inseparable
from the other in the sense that the breaking of the connection means the
destruction of one of the terms. It thus differs from conjunction (sam.yoga –
which links individual substances), where both terms survive separation.
Inherence obtains between qualities and the substances possessing them,
actions and substances to which they belong, real universals and their particu-
lar instances, and unique particularities and the permanent substances that
they individuate. A complex substance is a whole (avayavin) inhering in each
of its parts. A cloth cannot continue in existence without the threads, while the
threads may exist separately.
Qualities and action inhere in substances. The particular case of the quality
blue that is a feature of some lotus needs the lotus for its occurrence. A quality
is always a property of something. Although we never encounter a lotus
without some quality or other, at the first moment of its existence, the
substance has neither qualities nor actions. Quality presupposes substance
but substance does not presuppose quality. Qualities (and actions) only exist
in some substance that supports them.
There is a problem about treating inherence as a relation in its own right
that was exposed by the Advaita Vedāntin philosopher Śam.kara in his Brahma
Sūtra Bhās.ya 2.2.13. The inherence relation is just as real as the items that
it connects. This generates an infinite regress since it seems that further con-
nections are required to tie the relation to its terms. Further, the relation of
contact (sam.yoga) is a gun.a, so it is tied to the substances that it connects by
samavāya. To avoid this, the tradition holds that inherence relates itself to its
terms – it behaves like glue in that it is ‘self-linking’ (svarūpa-sam.bandha).
The Nyāya account of causation says that prior to origination, the effect
did not exist in its underlying cause but is a totally new product, different from
the already existent basic elements out of which it is made. They reject the
category of potentiality, holding that only what is actual and concrete is real
and can cause something else. Causation is not the actualization of what was
potential but the generation, through re-arrangement, of new entities out of
already existent factors. A cause is defined as a necessary prior condition of an
effect. There are three factors in a causal complex: the underlying cause which
is always a type of substance – e.g. the threads comprising the cloth; the non-
inherent cause which is always a quality or activity – e.g. weaving and colour
of the threads; the efficient or instrumental cause (nimitta) – e.g. the shuttle
and other instruments. The weaver is the agent cause.
The category Abhāva (absences)

Silence is the absence of sounds. Silences are real. Some absences are real. The
absence of coffee in my cup is a fact. When absence is the absence of something
it is classified as a reality. That which is absent is technically termed the coun-
terpart (pratiyogin) to the absence.
There are four varieties of absence. ‘Prior absence’ is the non-existence of
an entity before its production by the re-arrangement of other factors. Destruc-
tion is the dissolution of something that has existed for a period of time. Abso-
lute negation covers logical and physical impossibilities. Mutual absence is
another way of expressing difference. It is the denial of identity between two
things such as a pot and a cloth.
Epistemology: the Pramān.as

Classical Indian epistemological theory centres on the notion of pramān.a
or instrument of knowing. The number of such instruments differs according
to the different schools of thought. Here we shall just look at perception,
inference or reasoning and testimony as methods of knowing. The seeker after
truth wants to be sure that as many of his beliefs as possible are true. But he
cannot know this by checking them one by one. Obviously, false beliefs do not
admit the error of their ways. The truth-seeker wants to be in a position where
his beliefs are justified, that is to say, that they are not merely true by chance.
The best way to do this is to ensure that any beliefs that he acquires are
produced by a reliable method that generally leads to true beliefs. Pramān.as
are such methods.
The Nyāya outlook is basically one of commonsense or direct realism

according to which perception is direct acquaintance with an external world.
Although cognitive errors and hallucinations sometimes occur, for the most
part our perceptions are reliable, and vigilance against specious reasons
(hetu-ābhāsa) can ensure that inferences based upon them are sound. They
say that cognitions are primarily and directly about objects and states of affairs
belonging to an objective, mind-independent environment. Cognition, as we
saw above, is the illumination or discovery of objects. Cognition is direct in
the sense that no veil of representations, and no sense-data, fall between the
subject and what is given. The extroverted mind is a centre of interactions with
the physical environment. The mental is not a private inner arena isolated
from the world.
Their directly realist stance is illustrated by their view that memory is not a
means of knowing. This is because it does not reveal an object in the past,
but a representation of it. If the memory cognition is valid, that is because the
original cognition was valid. For the most part, what we see is the world about
us and not ideas in our minds. We are not, as it were, watching an internal
motion picture that has been constructed out of amorphous data supplied
from outside. Rather, perceptual cognition puts us directly in touch with
external reality.
A pramān.a is an epistemic capacity (śakti) or process that produces
knowledge rather than merely true beliefs, which may have been arrived at
accidentally or by chance. To do so it has to be functioning efficiently in the
absence of defects (dos.a) that may prevent its proper operation. Etymology
may be illuminating here. The word ‘pramān.a’ derives from the root ‘mā’,
which means measuring and ascertaining. A pramān.a is that by which some-
thing is measured. It is an instrument for ‘getting the measure’ of something –
understanding it as it really is in the face of some doubt or query that throws
the veracity of a belief into question.
The sorts of defects that can impede the proper operation of the perceptual
instrument include environmental factors such as poor lighting, haze or
remoteness, physical factors such as short-sight or jaundice which makes
things appear yellow, and psychological factors such as inattentiveness, a
tendency to jump to conclusions, greed in the case of shell mistaken for
silver or timidity in that of the rope seen as a snake. That we can identify such
defects and understand why we are sometimes mistaken underpins the basic
reliability of the perception as a means of epistemic access to reality.
Nyāya is a philosophy of direct realism according to which perceiving the
world is to be directly related to matters that are outside you – seeing things
that already exist prior to their being perceived as they really are. But the fact
that cognitive errors and hallucinations occur suggests that this cannot be
right. Surely in such cases what we are ‘seeing’ are sense-data or mental
representations (‘mere appearances’) of reality that happen to be deceptive.
Sometimes I look out of the window and think that it is raining. Sometimes
I am right and sometimes wrong. But the experiences are the same. From the
subjective or phenomenological point of view (‘how it feels’) really seeing
a real snake is the same as misperceiving a rope as a snake. Since both true and
deceptive experiences feel the same way to the subject (they are phenomeno-
logically or experientially indistinguishable), it appears that veridical experi-
ences too are mediated by mental representations and we are not directly
acquainted with the world. This may encourage the sort of idealist position
according to which all experience is just a matter of ideas occurring to minds,
and that there is no mind-independent physical world.
Nyāya has a response to these lines of thought. They say that we only
make mistakes when things are similar in some respects. We never confuse a
mountain with a mustard seed. When I see a rope as a snake this is because the
rope is similar to the snake in some respects. (Likewise in the case of seeing
a piece of mother of pearl as silver.) Under the influence of weaknesses such
as timidity and excessive caution, the appearance of the rope elicits memory
traces of a snake. The memory trace of the snake triggers the operation of
extraordinary perception (alaukika-pratyaks.a) that relates the subject
directly to some snake that is outside one’s field of vision. The appearance in
the deceptive case is not an inner mental item, such as a sense-datum or a
memory-image, but a real feature of the world. So the deceptive experience
relates to something real, even though it is not actually present to the mind
and senses. In other words, I am neither experiencing a purely mental
representation or ‘mere appearance’ nor am I merely imagining a snake out
of thin air. (Dreams, fictions and hallucinations are analysed as consisting
of remnants of previous experiences of realities to which they refer.)
It follows that although the true and deceptive experiences seem the same
to the subject, in fact they are different in themselves. This denial of a common
psychological process or mechanism shared by true and false perceptions
(a version of ‘disjunctivism’), despite how things seem to the subject, blocks
the fundamental representationalist and idealist objections to the direct realist
view because there is no need to invoke appearances or representations
intervening between the subject and the world.
We said that a true belief is a piece of knowledge as long as it is produced by
a properly functioning instrument. ‘Without a means of knowledge, there is
no understanding of objects. Without understanding of objects, there is no

successful activity’ (Nyaya-Bhashya 1.1.5). It is not a necessary condition for
being in a state of knowledge that the subject knows that he knows. Knowledge
does not require subjective certainty. Validity is not dependent upon confirma-
tion. What is required is that the cognitive state has been arrived at by an appro-
priate and reliable method. Nyāya philosophers hold that a cognition is true
when it corresponds to some reality external to the mind. They recognize that
we cannot check the truth of a cognition by stepping outside ourselves and
comparing our thoughts with objective reality. Just as false beliefs do not
declare their own falsity, so true beliefs do not display their own truth. Some-
thing more is required. Introspection (anuvyavasāya) may apprehend a primary
belief about the world. We can know that belief is veridical if successful activity
proceeds from it. So we infer on the basis of confirmatory evidence that a
belief is knowledge. When justification is required, we can appeal to successful
activity consequent upon a cognition.
Nāgārjuna had raised the problem of how we know that the pramān.as
themselves are veridical. Is there not an infinite regress here? Nyāya thinks
that the problem is specious. They admit that in principle there is a danger of
infinite regress but that in practice the problem never arises. They make the
points that we can see things before we know that we have a visual faculty and
that we all recognize that a lamp illuminates both other things and itself. They
also admit, in the light of the widely recognized principle that something can-
not exercise its proper function on itself, that while we are subjects of the visual
process of seeing an object, that process cannot itself be an object of observa-
tion. But that visual process can be scrutinized subsequently and assessed for
validity (‘Was the light bad? Did I have my glasses on? Did I have my mind on
something else?’). There is no contradiction is a means of knowing being the
object of another means of knowing.
Circumstances may arise when we want to check that what has been
assumed to be a piece of knowledge is indeed that, or when we want to justify
a piece of knowledge. In such circumstances, it is usually sufficient to appeal
to successful activity consequent upon the use of the pramān.as. This is what is
meant by notion that the validity of knowledge is extrinsic (paratah.-prāmān.ya).
The crucial point is that justification does not confer the status of knowledge.
If we had to justify everything before we could treat it as knowledge, the proc-
ess would have no end, and there would be no knowledge. It is worth noting
that Gaṅgeśa simplifies matters, quite consistently with the tradition, by
accepting that the validity of knowledge is intrinsic.
Knowledge by perception (pratyaks.a)

To begin with, it is worth observing that according to the Nyāya realist the-
ory, perception is basically a physical process. Cognitions are understood as
qualities residing in the soul-substance in the same way as the quality of blue
occurs in a jar. Cognitions are products of a perceptual mechanism based on
contact between sense-faculties and the external environment.
Nyāya-Sūtra 1.1.4 defines perceptual knowledge as a cognition produced by
contact between a sense-faculty and an object that is faithful (avyabhicāri),
focused (vyavasāya) and does not need to be expressed in words (avyadapeśya).
This definition is elaborated by Vātsyāyana and all subsequent commentators in
the Nyāya tradition. According to Annambhat.t.a’s Tarka-Sam.graha, ‘perception
is the instrumental cause of perceptual knowledge. It is knowledge produced by
contact between an object and a sense faculty. It is two-fold: non-conceptual
(nirvikalpaka) and conceptual (savikalpaka). The former is a cognition whose
content lacks specific components (nis.prakāraka), as when we say, “This is
something”. The latter includes specific components, as when we say “this man
Dittha is a dark-skinned Brahmin” ’.
Perception as a means of knowing is direct acquaintance, unmediated by
sense-data, with the objective environment. Their directly realist stance is
emphasized by the view that memory is not a means of knowing. This is because
it does not reveal an object in the past, but a representation of it. If the memory
cognition is valid, that is because the original cognition was valid. After the mid-
dle of the fifth century, all writers respond to the view of the Buddhist Dignāga
who says that perception is free from the imposition of concepts (kalpanā) on the
given. Writing from a completely different metaphysical perspective, he thought
that it is only perception that relates us to the real world of inexpressible unique
momentary particulars. According to this view, perception never involves
concepts. Any experience involving concepts and words he classifies as
‘thinking’ (anumāna) rather than perception. Thinking is at one remove
from what is immediately given. (The opposite view is articulated by the
grammarian and linguistic idealist Bhartr.hari, again from a different metaphysi-
cal perspective, for whom our world is the proliferation of meanings deriving
from a single absolute meaning. Bhartr.hari says, ‘There is no awareness that is
not accompanied by language. All cognition appears as permeated by words. If
the eternal identity of word and consciousness were to disappear, consciousness
would not illuminate anything: because that identity makes reflective awareness
possible.’)
The Nyāya response to Dignāga is to distinguish two varieties of perception.

They maintain that perceptual experience may or may not involve concepts.
Non-conceptual perception is termed ‘nirvikalpa’ and conceptual perception
is termed ‘savikalpa’. The latter is knowledge that comprehends the relation
between what is qualified (viśes.ya) and its properties (viśes.an.a) such as
name, universal and qualities. This may be understood as the distinction
between looking around the room without noticing anything in particular
and directing one’s attention to some particular feature and thinking ‘that is
a round table’. Alternatively, we may distinguish between mere seeing and
understanding what one sees. Although there are different formulations of
the distinction, they all posit a process of two stages. In the first stage (‘mere
seeing’) aspects of the perceptual field are subconsciously registered. They
may subsequently feature in the explicit content of conceptual perception,
which grasps the relation between a complex qualified subject and its prop-
erties. A non-conceptual perception is an informational state that grasps the
whole object as ‘a something’, but the generic features are not understood as
shared by others, and the specific features are not grasped as peculiar to it.
Although the generic feature is not grasped as such, the Naiyāyikas argue
that it must be perceived implicitly: it belongs to the informational content
of the state. We can only think that an object belongs to the same kind as
others if it has been perceived as belonging to a certain kind in the first
place. They think that it is a mistake to suppose that we can see any number
of ‘bare particulars’ and abstract a general form from these indeterminate
experiences.
A conceptual perception is a cognition of something that has attributes
as characterized by those attributes. It involves an explicit manifestation of
information that has already been received at the non-conceptual stage. So
conceptual thought is not fabrication, but the organization and conceptual
interpretation of what has been discovered at the non-conceptual stage.
This distinction surely makes sense. It has been argued by those who want
to repudiate the ‘myth of the given’ that genuine understanding involves the
capacity to make inferences and that this can only happen if the initial experi-
ential content has propositional and therefore conceptual form. But something
is going wrong here, because on this account of perception young children and
animals do not perceive anything. They would be confined to their inner
worlds of sensation. But they surely have experiences of an environment that
has significance for them and are in receipt of information about it, although
they do not think about it in words and concepts.
Anumāna: knowledge by
reasoning or inference
While perception is an instrument for the acquisition of knowledge about
what is present to the mind and senses, inference (anumāna) is a means of
acquiring knowledge about matters that are beyond the range of direct
acquaintance. The outlook is empiricist (although we must remember that
sensory experience is not restricted to particulars but includes universal
properties and relations): inference depends upon information supplied by
perception.
According to Nyāya, inference begins with a doubt, such as whether there is
a fire on a remote mountain. The relevant observation is that we can see smoke.
In this case, fire is termed the sādhya – that which is to be established. The
mountain is called the subject (paks.a) and the smoke is called the reason (hetu)
in the inferential process. We already know that there is no smoke without
fire (this invariable association is called vyāpti) and are familiar with other
instances where they co-occur, such as the kitchen. By way of corroboration,
we also know the truth of the contraposed version of the generalization: ‘no
fire, no smoke’ from cases like the lake. This negative example is intended to
show that we have investigated the matter thoroughly and have not confused
smoke with mist seen rising from a lake early in the morning. We apply know-
ledge of the general principle to the case in question and can safely conclude
that there is indeed fire on the mountain although we do not see the fire.
A demonstrative inference (prayoga) used to persuade someone else
(parārtha-anumāna) would be formulated by the Nyaya-Vaiśes.ikas as:
Statement of the position or uncertainty (pratijñā): ‘There is fire (sādhya) on the moun-
tain (paks.a).’
Logical reason (hetu): Because there is smoke on the mountain.
General principle (vyāpti): ‘Wherever there is smoke, there is fire’ that is supported by
examples (dr.s.t.ānta) – like a kitchen (sapaks.a); unlike a lake (vipaks.a).
Application: ‘There is smoke on the mountain’, which states that the subject under
consideration has the logical reason that is always associated with (pervaded by or
included in – vyāpta) the property to be proved.
Conclusion: Therefore, there is fire.
Clearly the notion of invariable concomitance or pervasion (vyāpti) is

pivotal. Knowledge of pervasion is said to be the instrumental cause of a piece
of knowledge arrived at by the inferential process. A is said to pervade B when

it occurs in all or more the instances where B occurs. Fire pervades smoke.
Having an agent pervades being a created product. Impermanence pervades
being a product. The factor of greater extent is called the pervader (vyāpaka)
and that of lesser extent the pervaded (vyāpya). This means that A invariably
accompanies B: smoke is always accompanied by fire and thus serves as a sign,
logical reason or proving property (hetu) of the presence of fire.
The presence in some cases and its absence in others are the definition of
the sign (liṅga) that is appealed to as the reason. In the background there is the
challenge from the materialist sceptics who deny the validity of inference
because we are not acquainted with all possible circumstances. We cannot
know that smoke is always accompanied by fire.
There was a range of views about the nature of the statement of pervasion:
some thought that it was just a generalization about many observed instances
to which no counterexample has been found. If the proposition that there is no
smoke without fire were just a generalization based on observations of instances
seen so far, we could not be certain that it will hold tomorrow. The developed
tradition resists scepticism by saying that nature is regulated in such a way that
there is an invariable association (vyāpti) between universal properties such
as being ‘being smoke’ and ‘being fire’ such that whenever ‘being smoke’ is
manifested so is ‘being fire’. The instances might be infinite, but the universal
is single. On this account, knowledge of the general proposition that there is
no smoke without fire is a type of non-sensory intuition of the invariable
association between the universal smoke and the universal fire, or of the
pervasion of smoke by fire.
Both the Nyāya and the Buddhist logicians devote much energy to discus-
sions of the many varieties of specious reasons (hetu-ābhāsa) in arguments.
Among the most common are:
Anaikāntika-hetu (inconclusive reason) of which are three varieties:
(a) the reason occurs in cases where what is to be proved is absent (i.e. it occurs in
vipaks.as).
Example: The village is holy, because it is close to the Ganges. But there are
unholy things close to the Ganges (called sādhāran.a-hetu).
(b) Where the logical reason only applies to the subject of the inference (asādhāran.a-
hetu). Example: ‘Sounds are impermanent, because they are audible.’ It is a
condition of an inference’s validity that we should be able to cite an instance
other than the subject of the inference were both the logical reason and the
property to be proved always occur together. But this is impossible here because
nothing other than sounds have the property of audibility.
(c) Where the subject of the inference is universal. For example ‘Everything is
nameable, because it is knowable’. The invariable association is, ‘Whatever is
knowable is nameable’. Distinct from the universal subject, there can be neither
a sapaks.a nor a vipaks.a, showing the invariable association between the logical
reason and what is to be proved. Since the inference begins from a question
about whether nameability applies to the subject, the invariant association,
‘whatever is knowable is nameable’ is itself doubtful.
The variety called Asiddha-hetu (unestablished reason) occurs when the

reason does not occur in the subject under consideration (svarūpa-asiddhi);
e.g. ‘sound is a property, because it is visible’ and in cases where the subject of
the inference does not exist or when its existence is controversial. Buddhists
apply this to what are held by Hindus to be proofs of the existence of the soul
and its properties. They say that no soul is perceived apart from experiences,
and since it is the sort of thing that ought to be perceptible, its existence cannot
be proved.
In a viruddha-hetu (contradictory reason) the reason is defective because it
never occurs where the property to be proved occurs: i.e. it contradicts what
one wants to prove. Examples: ‘There is fire on the mountain, because it is icy.’
‘Sound is eternal, because it is produced.’
An inferential argument for the existence of God might run, ‘The world has
an omniscient creator, because it is a complex product; like a pot’. Here the pot
example invites the accusation that the reason is contradictory because it leads
to the conclusion that the world has a creator of finite intelligence.
There is another classification, shared by Sām.khya and Mīmām.sā, of
inferences into three varieties. This is variously interpreted. Inference from
cause to effect (pūrvavat) is when we infer from the presence of clouds that it
will rain. Inference from effect to cause (śes.avat) is when we infer from a swol-
len river that it has rained. Inference from general observation (sāmānyatodr.s.t.a)
draws conclusions about the suprasensible. Observation that a heavenly body
has changed place implies movement, albeit unseen. If qualities belong to sub-
strata, and cognition is a quality, we can infer that there is some substance,
which we call self, to which it belongs.
Śabda: testimony and the transmission

of true information
Among the philosophical traditions, Vaiśes.ika and Buddhism admit that we
may obtain knowledge from words, but they say that testimony is a form of
inference and not a method of knowing in its own right. Nyāya, and virtually
everyone else, treat testimony as an independent pramān.a, largely because of
the problems attaching to assimilating it to either perception or inference.
At the same time, they are aware that of the things that we know, many are
known on the basis of testimony and not perception or inference. They say
that testimony is information supplied by someone who knows the truth and
wants to tell it. Speech often misleads but it is a method of knowing in so far
as the speaker is well-informed and sincere. As well as reliable information
about matters belonging to our world, testimony also includes the Scriptures
composed by the ‘Seers’ who originally heard the sound-units comprising the
Vedas. Later writers hold that the Scriptures are reliable because a benevolent
and omniscient deity is their author.
It is assumed that the normal situation is that what the speaker states when
he utters an assertoric sentence and what the hearer understands directly is a
piece of verbal knowledge (śābda-bodha) about a state of affairs in the world.
Let us bear in mind that it is possible to understand a proposition without
believing it to be true and that it is possible to assent to a false proposition.
So neither understanding not assent is sufficient for knowledge. The Nyāya
account of testimony establishes the conditions under which the hearer under-
stands a proposition, assents to it and obtains a piece of knowledge.
One knows a proposition expressed by a sentence (i.e. there is śābda-bodha)
when:
The hearer acquires a true belief about the world from hearing the sentence.
The speaker knows the truth and is reliable, sincere and competent.
The sentence expressing the proposition has these features:
(i) The words are uttered together or written together (āsatti).

(ii) It is grammatically correct (ākāṅks.ā). Just saying the word ‘a pot’ does not really
signify anything. We need another word such as ‘bring’.
(iii) It is semantically adequate (yogyatā). ‘He cuts with a knife’ makes sense, but ‘he
cuts with butter’, albeit grammatically correct, does not.
(iv) If the sentence is ambiguous, consideration of the speaker’s intention (tātparya)
in the circumstances should solve the problem.
Words and sentences

The Nyāya view about individual words is that while each word has its own
significative power (śakti), it is only in the context of a sentence that words are
used to really mean anything. In short, sentences, both fact-asserting and
action-guiding ones, are the vehicles of communication and understanding.
They think that the relation between a word and its literal or primary meaning
is created rather than innate. In some cases (e.g. Scriptural words and words
in common usage like ‘cow’) the word-meaning relation has been decreed
by God. The rationale is that we have no knowledge of any original human
stipulation that ‘pot’ should stand for pots. In other cases, the relation has been
fixed by human convention – this applies especially in the case of proper names
and in languages other than Sanskrit.
Words like ‘pot’ and ‘cow’ apply to innumerable individuals that share some
property. If the word pot just stood for an individual, the word-object relation
would have to be miraculously renewed every time someone mentioned a pot.
Communication would break down. But if the word just expressed a common
property, when someone said ‘bring the cow’ he would absurdly be saying
‘bring the universal cowness’. So they argue that words like ‘pot’ and ‘cow’ have
a complex signification. They express a universal feature common to a kind
(jāti), a perceptible shape (ākr.ti) and an individual (vyakti). A perceptible
shape, a configuration of parts, indicates the kind of which the individual is a
manifestation. In the case of a model or toy cow, it suffices for the application
of the word. While most forms indicate some jāti, not every jāti is indicated by
a form. For example, clay and gold have no specific configuration. They are
indicated respectively by their characteristic smell and colour. The variability
of forms led the Navya-Nyāya philosophers to deny that shape (ākr.ti) is an
ingredient in the meaning of words. Finally, since here is no real universal
‘cookness’, the general term ‘cook’ applies on the basis of participation in the
action of cooking. The signification here is just a matter of human agreement,
as in the case of proper names.
Further reading
An enjoyable starting-point is Jayanta’s play, Much Ado about Religion (Dezső, 2005) Act Three of
which sees a confrontation between Nyāya and Buddhist ideas.
Another way into the original material is Annambhat.t.a’s Tarka-sam.graha (Athalye 2003).
The Nyāya-Sūtras and the commentaries by Vātsyāyana and Uddyotakara are translated in Jha (1984).
There are summaries of works and a helpful introduction in Karl Potter (1977), which deals with
Nyāya-Vaiśes.ika up to Gaṅgeśa, and in Potter and Bhattacharya (1993) for the later period.
Bimal Matilal, Perception, treats extensively of Nyāya debates with the Buddhists. Despite the title, its
scope is wider than epistemology.
Chapters VI–XII of Matilal (2002) discuss the Nyāya realism in the light of some contemporary
philosophical interests.
Jonardon Ganeri, Indian Logic, is a useful collection. The same author’s Semantic Powers concerns
Navya-Nyāya philosophy of language, but says illuminating things about the earlier tradition.
Ingalls (1951) is a lucid guide to key Navya-Nyāya concepts.
Kishor Chakrabarti (1999), Classical Indian Philosophy of Mind, also deals with questions about the self
and the existence of God.
Scharf, Denotation (1996) has a section on Nyāya theories about linguistic meaning and reference and
translates Nyāya-Sūtra, 2.2.58–69.
Halbfass, On Being and What There Is (1992) and Tachikawa, The Structure of the World (1981) focus
on Vaiśes.ika.
The best, if not the only, work on the materialist tradition is Franco (1994), Perception, Knowledge and
Disbelief.

investigation
1. Is the Nyāya concept of self a physicalist one?
2. Are their arguments for selves as persisting principle of identity refute the Buddhist
point of view. Are the traditions talking about the same thing?
The Mı̄mām
. sā Vision
9
Chapter Outline
The authority of the Vedas (Veda-prāmān.yam) 119
Words and sentences 121
Kumārilabhat. t. a’s realism 123
The perceptual process and our experience of the world 124
The nature of ritual agents 128
Further reading 133
Mīmām.sā (called Pūrva Mīmām.sā when it is subordinated to its cousin Vedānta

or Uttara-Mīmām.sā, the Higher Inquiry) is one of the six visions (darśana) of
classical Indian philosophy. Its practitioners are specialists in the exegesis of the
Vedas as the sole authority for the ritual and social duties of twice-born caste
Hindus. It includes philosophical reflection upon matters connected with the
performance of rituals that are part and parcel of everyday life for orthodox
(smārta) Hindus. In opposition to Buddhism, Mīmām.sā thinkers articulate a
view of the world as sustained and organized by rituals that are performed by
persisting human agents who may enjoy their fruits in the future and in subse-
quent lives, and ultimately secure release from the series of births. A world safe
for ritual expectations is one in which there are stable subjects of experience
and an objective realm structured in such a way that it behaves predictably.
Here the most important figures are Śabara (probably second half of the
fifth century A.D.), who wrote the earliest extant commentary on the Jaimini
or Mīmām.sā-sūtras, which is the foundational text of the system; Kumārila
(600–660 A.D.) whose Ślokavārttika is an explanation of Śabara’s work; and
Prabhākara (c. 650 A.D.) whose works include the Br.hatī.
The central concept here is that of dharma, which comprises the notions of
natural law, right order and social and religious duty. Dharma is revealed by
the authorless, eternal and infallible Vedas. We act in accordance with dharma
when we obey the Vedic injunctions. What is to be done is dictated by scrip-
ture, not determined by human intellect and will. Such a mode of living relates
us to the highest good.
It is the dharma of grass to grow and of the sun to shine. It is the dharma of
members of the Brahmin caste to study and teach the Veda and the dharma of
Vaiśyas to engage in agriculture or commerce. Dharma would be unknown
were it not taught by the Vedas, which are explicated in texts called the
Dharma-Śāstras. As an ethical outlook, the concept of dharma is thoroughly
deontological. Consequentialist standards such as welfare, pleasure and pain,
the biddings of conscience, divine command or the cultivation of virtuous
character are all irrelevant to the determination of what is right and wrong.
Values are exclusively defined by Vedic injunctions and prohibitions, and are
manifested in the ‘conduct of the virtuous’ that derives from strict observance
of the Vedic rules separating the pure from the pollutant. Dharma is not a
‘universal’ ethic in that its demands vary according to one’s caste and stage
of life. One and the same type of action might be right for one person
(sva-dharma = ‘own dharma’) and wrong for another. There was a widespread
recognition of the principle that it is better to perform one’s own dharma
badly than that of another well.
It appears that originally the observance of dharma meant the performance
and patronage of elaborate and expensive sacrificial rituals generating
prosperity (bhoga) and temporary enjoyments in paradise (svarga). Its neglect
has all sorts of negative consequences ranging from personal misfortunes to
the collapse of the universe into chaos.
For the later Mīmām.saka theorists of ritual and social duty, the correct
performance of both the public sacrificial rituals by Brahmin priests and the
domestic rituals by householders of the highest three castes, in addition to
observance of the obligations appropriate to one’s caste and stage of life (varn.a-
āśrama-dharma) controls, maintains and perpetuates order and stability in the
universe. A properly performed rite automatically produces its result. It does
not depend upon any divine action. The gods exist only in name, that is to say,
only in so far as their names are mentioned in the course of rituals. There is no
belief in an absolute divinity unsurpassably great being. In that sense, the
Mīmām.sakas are atheists.
The Mı̄mām
. sā Vision 119
Mīmām.sā is primarily the hermeneutics and defence of those parts of
the Vedas that prescribe the performance of rituals and describe their results.
The primary concern is the correct performance of the rituals, including
the question of who is entitled to perform them and reap their benefits.
Mīmām.sakas are also concerned with questions about how language operates,
whether it is primarily referential and fact-asserting, or primarily action-
guiding, the nature and relationship of words and sentences and whether
words primarily signify individuals or express universal concepts. They debate
with the Buddhists over the question of the eternity of the phonemes (varn.a)
comprising the Vedas. They are also concerned with the question of what must
be true of the nature of individual sacrificial performers if they are to receive
the benefits of the rituals in the future.
Infallible Scripture (śruti) is classified in three ways:
Vidhi or Codanā: passages commanding the performance of specific rites

with specific results. Such injunctive statements are imperative in form (from
the technical point of view of Sanskrit grammar their verbs are in the optative
mood). They maintain that language is properly meaningful only when it is
injunctive or prescriptive – when it tells us to do something and how do go
about it – and not when it is descriptive or fact-asserting. The argument for
this is that children learn language by observing its application in various con-
texts of activities. The primary function of language is to produce some action
– something to be done (kārya) or brought into existence. The ritualists draw
the conclusion that those scriptural passages that appear to speak of already
established realities must be construed as merely supplementary to the all-
important passages enjoining ritual activity.
Arthavāda: Indicative statements describing the manner of performance of
rituals and providing explanations. As we have just seen, these are subordinate
to injunctions and not independently significant.
Mantras: Incantations invoking the presence of deities during the rites.
The authority of the Vedas

(Veda-prāmān.yam)
The Vedas are authorities that instruct us in matters – things to be done –
that are outside the scope of other means of knowing such as perception and
inference, and about which we would otherwise remain in ignorance. Dharma is

not something that we ourselves could know by means of our natural capacities.
Since it is something to be brought about, it is outside the scope of perception
and those means of knowing that are founded upon perception. A perception
is a cognition that is produced when the human sense-faculties are connected
with an already existing object. It cannot be the grounds of knowledge of
dharma since it only apprehends presently existing things. Only prescriptive
Vedic language has the capacity to inform us about dharma as something that
needs to be done.
Any religion that claims to know about the supernatural is going to need a
source of authority that is not of this world. For Christians it is the revealed
Word of God: both Jesus and the scriptural testimonies to him. Naiyāyikas,
Śaivas and Buddhists hold that scriptures are authoritative when they have a
promulgator (in the case of the first two it is Śiva) possessed of appropriate
virtues. For the Buddhists, the Buddha is himself a pramān.a – a teacher of
truth. His authority is indicated by the original and unprecedented character
of his salvific message when he tells us about what is to be sought after and
what avoided and the means of so doing. He teaches things that we could
not know otherwise, as the Veda is held to do. Because he is a reliable guide
(avisam.vādaka) in matters to do with human life, he can be presumed to be an
authority for non-empirical matters too.
For both Mīmām.sakas and their Vedāntin cousins, the Vedas are authorities
about what lies beyond the bounds of sense because they have no author,
either human or divine. They are called ‘apaurus.eya’, which literally means
non-personal. The Vedas are simply given, not created. The notion of the
uncreated, non-personal nature of the meaningful sound-units of the Vedas
is foundational for the authority of those scriptures.
The validity of the Vedas is intrinsic to them (svatah.-prāmānya). The notion
is an epistemological one, and it basically means that the truth of a thought
does not depend upon a later thought that verifies or establishes it. Were it to
do so, an infinite regress would result and we could never be said to know
anything. On this view, reliability does not depend upon verification or
confirmation and may safely be taken for granted. Validity just means not
being falsified. A statement may be presumed true if nothing falsifies it
or makes it suspect.
Error can arise from a lack of distinctness in the objects of cognition, or
it may occur because the complexity or subtlety of objects surpasses our
capacities for cognitive discrimination. But it may also arise because the
The Mı̄mām
. sā Vision 121
sense-faculties are somehow defective and thus fail to transmit information
to the mind. Kumārila specifies the nature of the possible sources of fallibility
which include such factors as greed, desire, hostility, pride, intoxication,
passion and shame. These are emotional and moral defects of finite beings.
It follows that a false speech-report is entirely the fault of the speaker. But the
Vedas have no author, so cannot be open to doubts about their reliability on
that score. Moreover, because the Vedas speak of what is absolutely imper-
ceptible and non-empirical, there is no possible cognition that could falsify
them. Mandates such as, ‘The person who desires paradise should perform
fire-sacrifices’ speak of things that can never be shown to be false.
The defence of the unquestionable authority of the infallible Veda is one
reason why the Mīmām.sakas argue against the existence of an all-knowing
and all-powerful divinity, or an omniscient Buddha. Kumārila argues that
neither perception nor inference can be used to establish an omniscient being.
Obviously, such is never observed. It would be difficult to frame a non-
question begging inference on a number of grounds, the most blatant of which
is that there are no relevant supporting examples that would validate a state-
ment of universal concomitance (vyāpti) between a reason and that which
is to be established. In addition, the arthāpatti-pramān.a (argument from
‘otherwise-inexplicability’ – ‘fat Devadatta does not eat by day, so he eats by
night’) would not be applicable because it is not clear that there is anything
that requires explanation by the postulation of an omniscient being. An
omniscient being cannot be established by scripture because if it were supposed
to be the author, the reasoning would be circular and if the author is not himself
omniscient, he cannot claim to know that there is an omniscient being.
Words and sentences

The Mīmām.sā view is that a Sanskrit word (pada) is a permanently fixed
sequence of timeless unproduced, and imperishable phonemes (varn.a), which
are manifested in audible sounds. It is the eternal sequence of phonemes that
is the conveyor of meaning or signifier (vācaka). Words are essentially and not
merely by convention, correlated with the extra-linguistic realities (artha) that
they express and that are their significations (vācya). The relation between
signifier and signified consists in the innate power (śakti/sāmarthya) of a word
and its reality is inferred from the fact that words have an informative capacity.
Kumārila distinguishes between this power in words and their application
(niyoga). A proper name such as Devadatta has a permanent meaning (‘Given
by a god’) and temporary applications to the men who are its bearers. But
in the cases of sortal and mass-terms such as ‘cow’ and ‘gold’, capacity and
application always coincide.
A given word stands for a limitless number of its objects: ‘We say, “The
single word ‘cow’ is uttered eight times and not that there are eight words
‘cow’ ” ’ The word-object connection (śabda-artha-sam.bandha) cannot be a
human creation. It cannot be established every time we use a particular word.
But we have no traditional recollection of anyone originally fixing the
references of words.
Each word expresses its own meaning and an uttered combination of word-
meanings is understood as a sentence. Sentences are not units of meaning over
and above the words comprising them. This is the mainstream view espoused
by Śabara and Kumārila and it is called ‘abhihita-anvaya’ or ‘the connection
of meanings that have been expressed’. (There is another view propounded
by Kumarila’s contemporary Prabhākara that is called ‘anvita-abhidhāna’
(‘expressing inter-related meanings’).) Prabhākara thinks that word meanings
are primarily understood when used to prescribe actions and to bring things
about rather than as referring to already existent objects. So the fabric of
linguistic understanding consists of sentences that include a verbal meaning.
Words have meaning only in the context of sentences. All meanings are
relative to particular situations. They argue that a child does not learn indi-
vidual words on their own but learns language through the insertion and
removal of words in sentences. She hears, ‘bring a cow’ and sees someone do
that. Then she hears, ‘bring a horse’. By the removal and insertion of words in
sentences, she learns the meanings of individual words. The theory of ‘the
expression of inter-related meanings’ may also be interpreted as the view that
it is only in the context of a sentence that a word has meaning: when we want
to know the meaning of a word, we should look at its use in context and not
scrutinize it in isolation.
Let us now return now to Kumārila’s view that the real relation between a
Sanskrit word and its extra-linguistic signification is inseparable and eternal.
If the word-meaning relation is to be permanent and unbreakable, words must
be primarily expressive of general concepts and the objects falling under
them. They do designate particulars when used in sentences in specific
circumstances, but this is not their primary signification. The word-meaning
relation obtains between a word and a natural or artificial kind (ākr.ti or jāti).
Words mean the underlying structures (ākr.ti) common to natural and artificial
kinds of things. The relation is fixed (niyama) and natural (autpattika). The
The Mı̄mām
. sā Vision 123
permanence of the word-object relation is of course consonant with the notion
of the intrinsic validity of the Vedas. If the Vedic words were like names and
primarily signified individuals, the word-meaning relation would be continu-
ally broken and reconstituted. But the form common to all cows is a structure
comprising the essential properties that make something a cow. It is basic to
the cosmos and is endlessly manifested as individual (vyakti) cows. The ākr.ti
is not just the visible appearance of cows. Cows come in all sorts of shapes,
sizes and colours. But each and every one shares the same internal biological
structure that we may call ‘cowness’ or ‘bovinity’. It is that to which we refer
when we speak of ‘The Cow’, meaning the species. The generic form of gold
(what we would call its atomic number or molecular structure) is likewise
common to all artefacts made of gold.
As Kumārila puts it, ‘the kind (jāti) is called the physical structure (ākr.ti)
because it is that by which the individual (vyakti) is formed. The generic
property (sāmānya) is the basis of a single concept under which individuals
fall’ [ŚV Ākr.tivāda 3]. The form is a generic property of many things. It is
not a configuration of parts (sam.sthāna) because there is no configuration of
parts in the case of actions, qualities and substances like the self. Because a
configuration perishes and differs for each individual, if that were the form, it
could not be the generic property expressed by a word. The generic property,
kind or form is a property constitutive of individuals and it is the object of a
simple cognition.
It is the form that is the primary signification of the word ‘cow’. That is why
it can apply to many. In everyday life and in Vedic usage, it designates an
individual in the context of a sentence. Kumārila says that in the case of
singular reference several factors are present: the kind (jāti construed as ākr.ti),
its individual manifestation (vyakti), their relation, the combination of
˘
those three, gender and causal role in an event (kāraka). In the expression
‘a white cow’, the quality-word (gun.a-śabda) ‘white’ behaves similarly. In
isolation it expresses a quality. In context it designates a particular instance
that is a part (am.śa) of an individual substance. A verbal root expresses
an action-type. In context, a finite verbal form specifies an activity.
Kumārilabhat. t. a’s realism

Kumārila holds that the uncreated and therefore eternal cosmos, structured
by universals and populated by individual knowers and agents confronting
a plurality of objects, is independent of human minds. This sort of realism

is integral to the ritualists’ world-view. People have to be confident that the
rituals are valid, that what they are doing matters and has real consequences,
and that they can expect to reap their benefits in the future. The objectivity of
the world is underwritten by the theory that cognitions are just acts, bearing
no internal content of their own, belonging to a self who is their agent. As acts,
they have an effect on objects, which already exist and are not brought into
being by thought. The effect that a cognitive act brings about in objects is the
temporary incidental property of being known (jñātatā) or made manifest.
This gives expression to the distinction, crucial to realism, between objects as
they are in themselves and objects as known. One knows that one knows
something not by introspection or from the internal luminosity of conscious-
ness, but as the result of a process of reasoning that observes the production of
an effect and concludes that it would not have occurred in the absence of
cognition. (Technically, this is arthāpatti, involving anyathā-anupapatti or
‘otherwise-inexplicability/impossibility. Of course, the theory here attracts
the objection that we are not knowing the world as such, but only the world
as modified by cognition.) Also, the fact that a cognition has occurred permits
us to infer that there is a self who is its agent.
The perceptual process and our

experience of the world
Kumārila maintains that both sensory perception and discursive, conceptual
thinking bring us into direct contact with the real external world. Direct
realists deny that perceptions are a veil of representations falling between
the subject and the given. Rather they disclose the actual structures of
reality. Kumārila confronts a range of Buddhist idealist and anti-realist
views, all of which maintain that perceptual experience somehow misrepre-
sents reality.
There is an extended treatment of questions about perception in the fourth
chapter of the Ślokavārttika.
The background here is the influential view of the Buddhist nominalist
philosopher Dignāga (400–440 A.D.) that sense-perception (pratyaks.a) and
reasoning (anumāna) are concerned with completely different spheres. This
appears to involve a radical severance between perceptual experience of the
given and its representation in thought. Dignāga defines perception as always
The Mı̄mām
. sā Vision 125
free from conceptualization (kalpanā). Perception is non-linguistic experience
of unique, ineffable momentary particulars (svalaks.an.a). The particulars are
the only genuine realities and they do not fall within the referential capacity of
language. Reasoning uses general concepts, which are mental constructs. There
are five types of conceptual construction (kalpanā): the association of a cluster
of particulars that have furnished a perceptual experience with a proper name;
the association of a cluster with a universal (‘this is a man’); the association of
a cluster with a quality or qualities (‘this is white’); the association of a cluster
with actions (‘this is running’), the association of a cluster with something else
that has been identified as a substance or individual entity (‘this man is wear-
ing glasses’). The meanings of words are the generalities (sāmānya-laks.an.a)
constructed by conceptual thought. He says that language is born out of con-
ceptual construction and conceptual construction is born out of language. The
picture is that as soon as we start expressing what we experience in words and
thoughts, we are distanced from the real world of momentary particulars.
Here is part of Kumārila’s response to nominalist anti-realism:
[The immediate background is that since inferential thought always involves
concepts, it cannot depend upon perception because the deliverances of the
senses are non-conceptual.]
Ślokavārttika IV, 111–120

[111] As for the view that the logical reason (and other elements in an inferential
sequence of thought) is not grasped by sensory perception because sensory per-
ception does not involve conceptualising: that view is false because concepts
(vikalpa) are implicit in that they enable us cognise the object.
The idea is that any perception that is about some object (in contradistinction
to a sensory impression that is of an object in the sense that it is caused by it)
must be to some extent informed by conceptual thought. That Kumārila
recognizes the non-cognitive character of sensation is apparent from verses
121–122 where he says that the sense-faculties are an instrumental cause
of cognition but they are not cognitive.
[112] In the first place here is cognition (jñāna) that is just seeing (ālocanā) and it
is free from concepts (nirvikalpaka). It is produced from the pure entity and is like
the cognitions of infants and the mute.
[113] Neither general (sāmānya) nor specific features (viśes.a) figure explicitly in the
content of awareness, but the individual that is their substrate is grasped.
The next three verses refer to an Advaita-Vedānta view that will appear in the
writings of Man.d.ana Miśra (c.700 A.D.) who says that, ‘Initially there is non-
conceptual perception relating to the bare reality of an entity (vastu-mātra).
The ensuing conceptual cognitions comprehend its peculiarities’ (Brahma-
Siddhi p. 71.1–2).
[114] Others say that there is an ultimate universal called ‘Substance’ or ‘Reality’
that is the sole object of perception.
[115] On the other hand, particularities (viśes.a) are known by conceptual

cognitions (savikalpa-buddhi). Some particularities are specific to an individual
and others are shared by many.
[116] But perception that arises without taking account of particularities as

either shared or specific does not differ whether it arises in relation to a cow or
a horse.
[117] That is false because we apprehend a distinct form in the case of each
individual entity. It is not the case that no differences are grasped just because we
cannot apply a word to the object.
[118] Even in non-conceptual awareness (nirvikalpaka-bodhe’pi), there is an

implicit apprehension of an entity as having both shared and specific features,
although only a simple form is grasped by the cognizer.
[119] The entity is not identified in its uniqueness because it is not distinguished
from others. A generic feature is not grasped because we do not notice any
similarity with other entities.
[120] A subsequent cognition by which an entity is grasped in terms of its proper-

ties such as its universal and its qualities is also considered a form of perception.
The denial by Buddhists such as Dignāga (and Advaita-Vedāntins) that

sensory perception is always non-conceptual is open to question. The same
can be said about the other extreme that perception always involves explicit
conceptual content: if that is true it follows that infants and animals lack
perceptual experiences. We see the latter view in the thought of the grammar-
ian Bhartr.hari, who says [VP 1.131–2], ‘In this world there is no thought that
is not associated with language. All cognition appears as pervaded by language.
If the eternal linguistic nature of awareness disappeared, consciousness would
not illuminate anything because it is that nature which makes identifications
possible’.
There is the question of explaining the relationship between sensation
and beliefs. Also, concepts are general but sense perception engages with
The Mı̄mām
. sā Vision 127
particulars. There is an argument that concepts and beliefs are abstractions
from the data supplied by sensory experience, but this is not really Dignāga’s
view. He insists on a radical disjunction between perceptual experience and
conceptualization because he wants us to realize that the everyday concepts
and categories of thought to which we are attached do not mirror reality as it
is in itself. Our concepts are at best impositions upon a given reality that is
unstructured in the sense that it is unarticulated and amorphous until we
organize it conceptually into persisting objects, kinds, properties and relations.
But if perception and thinking are two completely different modes of
experience, it is not clear how the raw data of sensation can be translated into
conceptual experience and perceptual judgements. If perception justifies
(rather than just causes) beliefs, a perceptual experience must yield a reason
to hold a given belief. But only an informational state with a least implicit
conceptual content can furnish such a reason.
Dignāga’s radical divorce of sensory perception and thinking provoked
opposition from thinkers belonging to the orthodox Brahminical schools.
They maintained that there is a single kind of mental activity called perception
and that it puts us in touch with the real world. Perception has two varieties:
non-conceptual (nirvikalpa) and conceptual (sa-vikalpa). As we have just seen,
Kumārila is party to this reaction to Dignāga when he says (as we saw above)
that a cognition that grasps an entity as a manifestation of a kind (a persisting
individual substance that possesses certain qualities and may be capable of
certain types of action) is also a form of perception. The point is that the con-
ceptual grasping of such structure is not a conceptual fabrication, imposition
or invention but rather represents a discovery. The informational states, which
a subject acquires on the basis of primary perception, are non-conceptual.
Judgements about the world based upon such states necessarily involve
conceptualization. For this to happen information about an object and its
properties must have been received and implicitly registered in the primary
informational state. According to this outlook, conceptualization or judgement
or belief takes the subject from an informational state with non-conceptual
content to a cognitive state with conceptual content. Although judgements are
based upon experience, this does not entail that they are about informational
states. They are about the world. When one wants to check that a judgement is
accurate, one looks again at the world and not at one’s states.
An illustration may help here: I may be looking in the direction of a green
expanse, although I might not even be explicitly registering that it is green
because my mind is on something else. This is a perceptual informational
state, but one that does not involve any specific thoughts about what is in the
field of vision. Its content is non-conceptual. But I may shift the focus of my
attention and register that the expanse is green and undulating. On closer
inspection, I come to understand that I am seeing a golf course. I am now in
an informational state whose content is conceptual. An animal or infant could
be seeing the same area, but could not believe that it is a golf course.
The problem with Dignaga’s position is that it cannot accommodate the
following distinctions:
a) Sensations, or sensory-impressions, are of objects in that they are caused by objects

but they are not about objects. In other words, they are not cognitive.
b) Some perceptual experiences are about objects without being discriminative or
involving judgements. They may just be informational states of the subject that
do not involve any commitment to beliefs about the object. But such states are
cognitive and informed by meaning.
c) Perceptual experiences that are explicitly discriminating (‘This is a black lap-top
PC’), and epistemic (‘I know that there is a book on the desk’).
The nature of ritual agents

Kumārila maintains that there is a subject or possessor (ātman) of cognitions,
(indicated by the pronoun ‘I’), which remains constant through all its
changing cognitive and affective states. It is an eternal reality, and our intuition
of it does not reveal any conditioning by space and time. Contact, mediated by
the psychological apparatus, with those conditions is a product of karma,
which propels the ātman through a series of embodied lives in a hierarchy of
contexts of experience.
For a ritualist like Kumārila, the primary significance of human individuals
consists in their being agents of sacrificial causality. The ideal person is simply
obedient to social and religious duty and free from the personal motivations
that generate karma. It is karma, purposive intentional action, that personalizes
the ātman. But truly disinterested action will not generate karma. Accumu-
lated karma will be exhausted over the course of lives dedicated to duty for
the sake of duty alone. A very long series of lives spent in disinterested
conformity to Dharma, in a spirit of ‘duty for duty’s sake’ and not for any
advantages that it might yield, will terminate in the release from rebirth of a
depersonalized, timeless and featureless entity without contacts or cognitions
of an environment.
The Mı̄mām
. sā Vision 129
But what about the selves that we are in the here and now? Fundamental to
the Mīmām.sā world-view is the conviction that the person who performs
an action is the same as the one who will enjoy its fruits. We would feel no
motivation to act if we did not believe that we will experience the consequences.
But if the ātman is inactive due to its permanence and omnipresence, and if
it is not really subject to pleasure and pain, how can it be the enjoyer of
the fruits of its actions? And if when it experiences suffering or pleasure it
undergoes a real change, how can it be permanent and immutable (nitya)
because that which is permanent is incapable of any sort of causal activity,
either simultaneously or successively? To questions like these Kumārila
responds that although the ātman is eternal, it can be connected with different
states, and it can be both an agent and patient. The distinction between
substantial entities and their states may be considered straightforward: I am
the same Christopher Bartley, although my thoughts, feelings and moods
are changing. I have a life history that is extended through time, but I cannot
be identical with the events comprising that history. If I were, it would make
no sense to say that on some particular occasion ‘I could have done otherwise
than I did’ because that would be to talk about another life history.
The Buddhists, echoed by Derek Parfit in Reasons and Persons, promote a
reductionist view of the self as a ‘series person’. They deny that we are persist-
ing individuals, enduring substances that are wholly present throughout the
course of their existence. What we really are is a causally related series or
stream of mental and physical occurrences (skandhas: body, feelings, sensory
perceptions, conceptual thoughts and inherited traits), which may be described
impersonally or from a third-personal point of view. Interrelated physical
and mental events constitute a person’s life. This is coupled with the thesis
of the essential temporality of beings (ks.an.ikatva): mental events and acts
are momentary and self-contained occurrences that somehow constitute
themselves as streams.
In the course of the eighteenth chapter of his Ślokavārttika, Kumārila
engages with this outlook and argues for a conception of the persisting Self
(ātman) as the ultimate subject of mental acts and experiences. Only such an
entity can be the guaranteed recipient of the benefits of ritual. People are only
going to be interested in ultimate release from rebirth if there is an identity
that will be released. His arguments are transcendental ones: continuity of
experience presupposes a single, enduring subject that is a further fact over
and above the stream of experiences and life-history. In fact, the resulting
picture of ‘the self ’ is a pretty minimalist one. There has to be a constant
subject if we are to make sense of the synthesis of experiences. This subject is

like Kant’s purely formal ‘I’ that accompanies all representations. It is not to be
confused with a personality or individual life.
The Buddhist interprets ‘person’ as meaning ‘a series of experiences’
(sam.tāna). Kumārila says that the expression ‘series’ may mean either a tem-
poral succession of momentary psycho-physical events, or it may mean a real
unity that is not reducible to its components but which emerges from them. In
the first case, the series is really impersonal and expressible in third personal
terms (‘there is the thought P’; ‘the feeling F is happening’; ‘the decision D is
being made’). As such the series cannot be considered as a genuinely personal
agent or subject. In the second case, the series might well be a single subject
compatibly with the diversity of its experiences. But now we are very close to
the notion of a stable self. He observes that identifying something as ‘this series’
presupposes that here is some sort of unity present and this involves abandon-
ment of reductionism about persons. Even if it were possible convincingly to
reformulate reports of mental states in third personal terms and still be true to
the character of our mental lives, we would need to be able to relativize those
reports to a ‘series-life’ because something has to own the experiences. We
would still need some explanation of why a series of physical and mental events
constitute one life. It is looking like the notion of a series-life is parasitic upon
that of a person as usually understood.
It is not enough to take the ‘stream’ metaphor literally and say that we
attribute unity to a succession to a flow of water when we call it a river. Rivers
are physical entities and they have banks. It is not clear what the equivalent to
the banks are where streams of experiences are concerned. If the self is really
just a bundle of perceptions, what ties the bundle together, if not a persisting
subject? (A physicalist may say that there is no problem here. Bodies, including
neurological events, individuate and underwrite personal continuity through
time. But the Buddhists are not physicalists, and they want to make sense
of the possibility of rebirth which of course presupposes the possibility of
disembodiment at death and re-embodiment at rebirth. This is not a problem
for the Indian materialists or Carvākas because they do not believe in rebirth.
Most Indian thinkers accept that the body cannot be the subject because (a) it
is not conscious and (b) it is a collection of parts and a number of parts only
form a system when they are subordinate to another principle that makes them
an organized structure.)
In an expression such as, ‘I know’, the pronoun ‘I’ expresses a constant
subject that is immediately given. Its support, the basis of its use in language,
The Mı̄mām
. sā Vision 131
is the persisting self. The crux in the debate between Kumārila and the
Buddhists is whether the subject is a thinking substance (an enduring thing
that thinks) or whether it is only an aspect specific to each successive mental
event and so different in each and every case. Kumārila argues that if moment-
ary ideas were the subjects of experience, as the Buddhists suppose, there could
be no experience of recognition in the form, ‘I previously saw this thing and
now I am seeing it again’. Kumārila recognizes that the flow of consciousness is
always both backward and forward looking, retentive and anticipatory. Con-
sciousness is fundamental to mental states in the sense that it can range over
them. The condition of connecting past and present mental states (‘I am look-
ing at something that I saw yesterday’) is a persisting conscious principle that
is not identical with those states. Successive distinct, momentary and
self-contained mental states cannot achieve this reconciliation. Now while
recognitional thoughts about external objects or other people may be false,
‘I-thoughts’ are immune to error by misidentification. They cannot relate to
anything other than the knowing subject. The subject is not identified in the
ways in which external objects are because there is no need for comparison
and the assignment of an object to a kind. We might mistake a piece of shell for
silver when, under the sway of avarice, stored subconscious traces (vāsanā) of
a prior perception of silver are revived by the shining surface appearance of a
piece of shell. The Buddhist holds that that what we call the recognition of the
self is a variety of this type of illusion resulting from confusion of similar
moments in a continuum. But Kumārila points out that strictly speaking there
is no recognition or memory of the self. It is not something contained in stored
traces of prior experiences. If I have an immediate sense of myself as the same
being over time (this is different from remembering what I had for lunch
yesterday), this does not involve a process of recollection like that of the
recollection of objects, facts or states of affairs. ‘I-experience’ is always simply
given: it does not have to be recovered. I never have to establish to myself that
I am now the same being that I was yesterday. The Buddhist may think that
the ‘sense of self ’ is a conventional misconception deriving from ignorance,
desire and attachment. This may be a reasonable account of the mechanisms
of selfishness. But ignorance, desire and attachment have to belong to some-
thing, and the best candidate appears some sort of persisting subject that
makes the mistake in the first place.
Kumārila’s epistemological stance is a strong version of commonsense
realism recognizing that for the most part, cognitions are valid. He thinks that
cognitions of a pre-existing objective physical world are always intrinsically
true. Errors and hallucinations sometimes occur, but they can be explained
as arising from identifiable defects in the perceptual apparatus, or from the
indistinctness of objects. Cognitions can be trusted as valid if they are not
contradicted by a subsequent perception. In short, there is no need to check
everything. Reliability may be safely assumed.
For a realist like Kumārila, the variety of experience depends upon differ-
ences between the objects grasped. The metaphor of ‘grasping’ is instructive,
conveying that objects already exist, independently of any thinker. They are
not created or produced by knowledge. There is no veil of representations
intervening between the knowing subject, the agent of the act of knowing
and the given. Consciousness is not a repository of forms or concepts, but is
more like a capacity for activity. When a cognition brings about the property
known-ness in an object, that property is accidental: its loss or gain makes no
real difference to the object.
We have said that intrinsically valid cognitions are true just in virtue of
their occurrence. The Vedic sounds are heard to command the rituals and
tell of the supernatural benefits accruing to their performers. (If what is
promised is something concrete and it does not happen, this can be blamed
on a mistake in the performance of the ritual.) In the cognition of the Vedic
mandates, there is no scope for falsehood, no possible standpoint from which
they might be criticized and countermanded, and no room for scepticism
about their authority.
Thinkers in the early Mīmām.sā tradition such as Śabara thought of dharma
as the same as the performance of prescribed actions. But this makes dharma
as transitory as those actions, with the consequence that we cannot credibly
establish a connection between an action and its consequences. The earlier
thinkers thought of the rituals as generating an unseen factor called apūrva
(‘something new’) or adr.s.t.a (something unseen) that transmitted the ritual’s
effect to the future. But its status and location were vague. Later writers favour
the view that Dharma is an eternal reality that is manifested in the rituals
and their consequences. The concept of manifestation is invoked in various
contexts to explain occasions where something eternal becomes perceptible in
certain conditions. For example, timeless phonemes are manifested in audible
sounds. A universal property (jāti or sāmānya) such as cowness is manifested
in individual cows (vyakti). Apūrva is treated by Kumārila as a power (śakti)
belonging to dharma that belongs to the sacrifices and to the identity of the
sacrificers. It activates the fruits of the rituals.
The Mı̄mām
. sā Vision 133
We began by mentioning that ritualism is sometime called the ‘earlier
Mīmām.sā’ by the Vedāntins who consider themselves in different ways to have
superseded the performance of rituals as leading the way to the highest good.
They maintain that at best it can only deliver a temporary state of well-being
in paradise, followed by a return to lesser incarnations. It is to the Vedāntins
that we turn in the next chapter.
Further reading
For the view that meaningful language is essentially prescriptive rather than fact-asserting (and the
Vedāntic response), see Lipner (1986), Chapter 1.
Chapters III, IV and IX of Halbfass (1991) are about Vedic orthodoxy, and ritualism and sacrificial
causality.
Olivelle (1999) and (2005) translate the texts that concern the practical applications of Dharma.
Act Four of Jayanta’s Much Ado about Religion (Dezső, 2005) sees a debate about the authority of the
Vedas.
Matilal (1990) Chapter X (‘Words and Sentences’) expounds abhihitānvaya versus anvitābhidhāna.
Scharf (1996) has a long section on words and meanings, accompanied by translations of typical
Mīmām.saka argumentation.
Eltschinger (2007) is a mine of information about Kumārila. For the latter’s epistemology, Taber (2005)
is invaluable and contains a richly annotated translation of the chapter on perception in Kumārila’s
Ślokavārttika.

investigation
1. Why do they put so much weight on the infallible authority of the Vedas?
2. Can the caste hierarchy with its implications that human beings belong to different
species be justified?
10 Vedānta
Chapter Outline
The interpretation of the Upanis.ads 134
The Bhedābheda tradition of Upanis.adic interpretation 136
Further reading 137
The interpretation of the Upanis.ads

The word ‘Vedānta’ literally means ‘the end of the Vedas’, where end means
the Upanis.ads. Vedānta is the systematic interpretation of the Upanis.ads
(collectively termed ‘śruti’ or ‘what has been heard’) either by direct comment-
ary upon them or by elaborate explanations of the aphoristic summaries of
their contents in the Brahma-Sūtras. The Bhagavad Gītā is also a key authority.
With the Epics (the Rāmāyan.a and the Mahābhārata) and Purān.as, the Gītā
is included in the category of traditional authorities called smr.ti or ‘what
has been remembered’. The latter has the function of elucidating and cor-
roborating śruti.
There are three antagonistic traditions of thought: Advaita-Vedānta which
is a metaphysical monism saying that fundamentally reality is undifferentiated
consciousness; Viśis.t.ādvaita-Vedānta which means the unity of a complex
reality; and Dvaita-Vedānta, which is a strict monotheism and a realistic
metaphysical pluralism. They all agree that eternal scripture (śruti) is the sole
means of knowing (pramān.a) about what is beyond the scope of sensory
perception and inference. Vedāntins hold that it is the Upanis.ads, the ‘know-
ledge portion’ (jñāna-kān.d.a) of the Vedas that reveal the truths that we need to
Vedānta 135
know about the Absolute Reality (Brahman), the soul (ātman) and the relation
between the two, the origin of the universe from the Brahman, the consequen-
tiality of actions (karma), transmigration (sam.sāra) as well as the means to
and nature of ultimate liberation from rebirth (moks.a/mukti).
All Vedāntins will agree with the grammarian Bhartr.hari’s observation that
even if something is inferred by clever logicians with a big effort, there will
always be cleverer ones who come up with another explanation [Vākyapadīya
1.42]. As Rāmānuja puts its: ‘A theory that rests exclusively on human concepts
may at some other time or place be refuted by arguments devised by cleverer
people . . . The conclusion is that with regard to supernatural matters, Scripture
alone is the epistemic authority and that reasoning is to be used only in
support of Scripture’ [Śrī Bhās.ya 2.1.12].
Vedāntins aim to construct a systematic and coherent interpretation of the
Upanis.ads, in accordance with a principle that they form a single body of
literature with a unified overall purport (tātparya). The Vedāntins follow
common exegetical norms and techniques in order to identify specific
coherent contexts of meaning (eka-vākyatā) and then demonstrate that these
contexts themselves fit together. As far a possible, the exegete must construe
the texts in their literal senses. The principal Upanis.ads were probably
composed over a very long period of time (roughly 800 B.C. to 300 B.C.) and
long before the first extant Vedāntic systematizations (c. 700 A.D.). Their
contents are diverse. They do not obviously teach a single coherent message.
Moreover, they are often obscure. The abbreviated summaries of the topics
of which they treat (the Brahma-Sūtras) are frequently ambiguous. So there
was ample scope for very different interpretations, and that is exactly what
we find.
Whereas the theorists of ritual performance (Pūrva-Mīmām.sakas) were
concerned with and insisted upon the primary meaningfulness of Vedic
action-commands (vidhi) bearing on ritual performance, Vedāntins focus
upon the fact-asserting or descriptive texts (arthavāda) referring to already
existent entities or states of affairs, rather than ‘things to be done’ (karya). Both
earlier and later Mīmām.sās developed sophisticated techniques of textual
exegesis and argued about whether ritual performance can be a path to
salvation with or without intuitive insight into the true nature of reality (jñāna),
devotion to God (bhakti) and divine grace (prasāda).
Most Vedāntins accept versions of the Sām.khya theory of satkāryavāda and
say that effects are emanations that do not differ essentially from their under-
lying or substrative causes. It follows that there is some form of ontological
nexus or parallelism of being (an analogia entis) between the world and the
Brahman that is its cause. Madhva is an exception here in that he denies that
there is a real continuity of being between God and the world. He maintains
(like the Śaiva Siddhāntins by whose outlook the Dvaita tradition is influ-
enced) that God produces the cosmos out of eternally real prime matter that is
distinct from him.
The Bhedābheda tradition of

Upanis.adic interpretation
Before turning to a consideration of some of the major representatives of the
different Vedāntic schools, we will mention a tradition of theological thinking
that is labelled ‘Bhedābheda’, meaning ‘difference and no-difference’. The idea
here is that the Supreme Being has two modes of existence: an unconditioned
mode that is the Brahman as it is in itself wherein all differentia have been sup-
pressed, and a conditioned mode that is the emanated cosmos. The cosmos is
understood as the real self-differentiation of the One, the substrative cause of
all finite existences. The cosmos emerges out of the Absolute Being. Freedom
from rebirth is achievable through a combination of works and knowledge. It
is the soul’s dissolution into the foundational reality.
We have here an attempt to hold together the transcendent unity of
divinity and the reality of the plural world. A version of this cosmography is
to be found in Ādiśes.a’s Paramārthasāra.
Rāmānuja radically modifies this outlook, but as a pupil of Yādavaprakāśa,
he is a reliable source of information:
Bhāskara and his followers say on the basis of the scriptures expressing unity
that the Brahman although having every excellent quality such as freedom from
evil is conditioned by a limiting condition (upādhi) and is bound and released, and
is the substrate of transformations (parin.āma) that are various imperfections.
[Vedārthasam . graha para. 8]
Because Bhāskara and his followers do not accept any realities other than the
Brahman and the limiting conditions, given the association between the Brahman
and the limiting conditions all the defects proper to the latter will apply to the
Brahman itself. [Ibid. para. 54]
Yādavaprakāśa and his followers, explaining the exact meaning of the scriptures
about unity say that the Brahman, an ocean of unsurpassable and immeasurable
noble qualities proper to its nature, is by nature both distinct and not distinct from
Vedānta 137
sentient beings, and the abode of many kinds of impure transformations.
[Ibid. para. 9]
Given their assumption that the individual self and the Brahman are both different
and non-different, it follows that if the Brahman is essentially the same as the
individual souls all the defects belonging to them will belong to it also. If God is
essentially constitutive of all the different creatures then he is the identity of each
and every one. Such being the case, all their pleasures and pains will belong to
him. [Ibid. para. 58]
In the course of his explication of Brahma-Sūtra 2.1.15 he attributes to these

thinkers the view that the Brahman is the primary cause – an entity that is
undifferentiated Being possessed of every potentiality. Prior to the emanation
of the cosmos it is self-luminous consciousness that is distinct from the insen-
tient and in which all experiences, pleasant and painful, are stilled. But during
the cosmic emanation it exists in tripartite mode as experiencers, objects of
experience and the controller. Because of the continuity between cause and
effects and the non-difference between the Brahman and the cosmos, all the
good and bad features of the produced cosmos affect the Brahman.
Further reading
Nakamura (1990), History of Vedānta is comprehensive.
For the Paramārthasāra see the text, translation and generous annotation in Danielson (1980).
11 Advaita-Vedānta
Chapter Outline
Śam
. kara 140
Authentic Being 144
The inexpressibility of the Brahman 149
Bhagavad Gı̄tā-Bhās.ya 13.12b 149
The path of active religious practices is insufﬁcient for
enlightenment 150
Śam
. kara and the Buddhists 156
Man.d.ana Miśra 162
The development of the tradition 166
Further reading 166
Discursive thought carries the mind here and there. Attention, extroverted or
introverted, is restless. Feelings and moods come and go. Most of life is
pervaded by the dualities of means-end rationality, the seeker and the goal,
actions and their results. But the meditator absorbed in profound contempla-
tion has neither thoughts nor feelings, nor experience of a world external to
consciousness. There is just motionless undifferentiated awareness that does
not seek to accomplish any purposes. This state is what Advaita calls ‘pure
consciousness’. Tranquil consciousness knows no fluctuations. It is not directed
towards nor about objects. It is not about anything. It has no specific content.
It is said to be blissful, for it nothing lacks. There is no sense of selfhood or
individuality. There is merely: being conscious.
Advaita-Vedānta 139
The Advaita tradition is inspired by certain Upanis.adic passages suggestive
of the identity of the soul and the Absolute Reality such as:
You are That. (Chāngogya Upanis.ad 6.8.7: ‘tat tvam asi’)
In the beginning, all this was just Being, one only without a second. (Chāngogya
Upanis.ad 6.2.1)
But many of the scriptures have a dualistic sense, some clearly suggesting a
difference between the Brahman and the individual souls and the cosmos,
others talking in terms of distinct agents, instruments and goals that are
aspects of external religious practices, and others obviously supposing that
the Supreme Principle is a being with glorious characteristics. Advaita-Vedānta
draws a distinction between the ultimate authority of texts teaching non-
difference and those that cannot possibly be construed in a non-dualistic
fashion. While this may appear controversial, it is not unprecedented for as
we saw above, the Mīmām.sakas had already distinguished between those texts
that prescribe actions (vidhi) and those that merely describe how and why to
do things (arthavāda).
Advaitins say that the essential teaching of the Upanis.ads is that what we
experience as the differentiated world of interrelated conscious and non-
conscious individual entities is really a complex, proliferated misunderstand-
ing superimposed upon the undifferentiated and inactive Brahman or Pure
Being. That foundational reality is nothing other than the coincidence of Being
and static consciousness. Liberation is just the cessation of the ignorance
or misconception (avidyā) that is responsible for our experiencing reality as
fragmented and our misunderstanding ourselves as individual experiencers
and agents. While religious activities, ritual and meditative may point one in
the right direction by purifying the mind and distracting us from immediate
selfish pursuits, they cannot produce enlightenment of liberation from rebirth
directly.
This is the tradition of those who deny that extroverted religious activity
can of itself deliver liberation from rebirth. Enlightenment arises from intuitive
insight unmediated by thoughts and words, into the identity of the ‘inner self ’
(pratyag-ātman) and the Brahman. This is the mystical realization of the
equation of Being and Consciousness. It is the manifestation of what one
always and already is. While insight obliterates all experience of differentiation
and individuality, vestiges of such experience persist in the life of the enlightened
one, who is ‘liberated while alive’ until his release at death. A possible response
on the part of the one who has seen the light is the renunciation of all ritual
acts as well as all everyday responsibilities and obligations. Śam.kara’s radical
vision is that of the world renouncer (sam.nyāsin). Man.d.ana Miśra is more
concerned with integrating liberating gnosis into the everyday life of the
householder. He recognizes that renouncing social ties and the shared religion
is not an easy option. He says that the Vedic rituals purify the mind and prepare
the way for realization of one’s true identity as the Brahman. Understanding of
that identity, conveyed by scripture, is intensified by ritual and contemplation
that counteract the still forceful traces of the pluralistic mentality. He recom-
mended the repetitive type of meditation called prasam.khyāna as a means of
removing moral defects and hindrances (kleśa) and as a way of internalizing
the Upanis.adic statements conveying non-duality. We shall see Śam.kara
rejecting this version of the view that liberation is the fruit of a combination
of works and gnosis.
Śam
. kara
Śam.kara, one of the founders of the tradition holding that differences are
unreal, probably lived around 700 A.D. His major work is a commentary on
the Brahma-Sūtras. He also wrote commentaries on the Bhagavad Gītā and on
individual Upanis.ads. Among the many other works attributed to him by
the Advaita tradition, the ‘Thousand Teachings’ or Upadeśa-Sāhasrī stands
out. His vision is that of the radical renouncer, which ultimately calls into
question the values of mainstream orthodoxy by denying that there are any
real individual thinkers, agents and acts.
There is no question that Śam.kara was an original genius, but it should
be mentioned that the Advaitic tradition traces itself back to Gaud.apāda
who probably lived around 450–500 A.D. and wrote the Āgama-śāstra about
the Mān.dūkya Upanis.ad. He likened the phenomenology of normal experi-
ence to that of dreaming and claimed that in both cases it is only the fact of
consciousness that remains constant. Individual entities (bhāva) are mental
constructs (kalpanā). The one supreme soul, the waveless absolute, imagines
itself as conscious individuals.
Gaud.apāda’s contemporary, the grammarian Bhartr.hari taught the ‘non-
dualism of meaning’ (śabda-advaita). The idea is that the diversified phenom-
enal cosmos (‘the proliferation of names and forms’) is the emanation from a
unitary sonic Absolute not of things but of meanings. It is the appearance
of the transcendent ‘meaning-reality’ (śabda-tattva), otherwise known as the
Brahman. The Absolute appears to transform itself through its innate powers
into meanings, words and sentences. Words and what they mean are identical.
The differentiated world of our experience is a product of diversification by
language. Reality is a matrix of differentiated meanings rather than things or
objects. Ignorance (avidyā), our default position as it were, is a function of
linguistic proliferation into individual words and propositions. It consists
in understanding the world in term of the individual entities that are the
referents of words and resting content at that level. Bhartr.hari’s linguistic
idealism exercised a considerable influence on Man.d.ana Miśra as well as on
the monistic Śaiva traditions.
In common with many classical Indian philosophers, Śam.kara’s soterio-
logical goal is the freedom of the authentic self (ātman) from rebirth. The
ātman cannot be captured by concepts and words. It will only reveal itself, and
that is something over which we have no control and upon which our activities
have no effect. Ātman here means something like ‘true nature’ or fundamental
identity. It is different from embodied individual personality, from caste-based
˘
social role, and from the psychological functions of thinking and feeling, as
well as the sense of being an agent interacting with the external world. All
of those involve a misidentification of what one really is with some aspect of
objectivity. Nothing that one can objectify, including thoughts and feelings,
can be the true self. We are approaching the notion of the ‘transcendental
subject’, the pre-condition of having any sort of coherent experience. The
notion of subjectivity here is neither individual nor personal. It is not a
particular perspective. We must remember that the state of release involves
no experiences of which anything might be the subject.
The word ātman is usually translated as ‘self ’ and sometimes as ‘soul’. But
the semantic range is broader. It is perfectly normal to speak of the ‘ātman’ of
an entity such as a pot (ghat.a). The compound ‘Ghat.a-ātmā’ does not mean
‘the self of a pot’, let alone its soul, but the nature or identity of pots. The
original meaning was something like ‘vital breath’. The word then came
to mean the spirit that is the essence of a sentient being. Thence developed the
meaning ‘the self ’ and ‘the soul’. Often the masculine singular forms operate
as reflexive pronouns such as ‘oneself ’, ‘herself ’ ‘itself ’ ‘themselves’. As such
it is synonymous with ‘sva’ (one’s/its own). With the rise of theistic forms of
Vedānta, a theological concept of self as The Soul, expressing the individual
as related to divinity, became prominent. But this concept is foreign to the
gnostic traditions, such as Advaita.
Śam.kara thinks that the ātman is revealed when what had thought of itself
as individual manifests its nature as one and the same as the Brahman, the
unconditioned and unchanging reality that is the coincidence of being and

pure consciousness. Such manifestation can never be a direct result of human
activity, although it may precede by religious practices that purify our con-
sciousness. Śam.kara’s vision of an absolutely impersonal and non-relational
state is an austere one, and it is mollified by other Advaita Vedāntins who say
that the ultimate state is ‘Being, Consciousness and Bliss’.
The Advaitin view is that we can neither satisfactorily explain why there is a
world of change and multiplicity nor why we find ourselves here. Relative to the
Brahman the diversified cosmos is less than fully real. All experience of indi-
viduality and plurality is somehow the product of beginningless ignorance or
misunderstanding (avidyā). The world is sometimes liked to a magician’s utterly
convincing illusions. But the working of avidyā is simply inexplicable. It is
neither being nor non-being. But aetiological explanations can be put on one
side. The important thing is to realize the truth and thus escape from rebirth.
Let us begin our exploration of Advaita-Vedānta with some extracts from
Śam.kara’s introduction to his commentary on the Brahma-Sūtras [BSB].
BSB 1 1.1
Given that it is impossible that there should be any real relation between Subject
and Object, whose spheres are the concepts of the first person and the third
person, whose natures are opposed like light and darkness, and that all the more
impossible is any real relation between their properties, then it is logical to hold
that it is mistaken to superimpose (adhyāsa) what is objective and its properties on
the first personal subject that is consciousness, and conversely to superimpose
subjectivity and its properties on the third personal objective order of things.
Nevertheless, there is natural everyday behaviour expressed in the thoughts,

‘I am this’ and ‘this is mine’, which combine the real and the unreal. The cause
is misunderstanding of the radically distinct subject and object, involving a failure
to discriminate the two, and the superimposition of the natures and properties of
each upon the other.
Superimposition is the manifestation in consciousness, in the form of memory, of

something previously seen in another place. Some say that superimposition is the
attribution to an object of properties seen in another case. Some say that it is an
error caused by a failure to discriminate between two things. Others say that it is
imaginative construction of properties that an object does not in fact possess.
But all the accounts agree that superimposition is the appearance to consciousness
of the features of one thing in something else. Such is the everyday experience of
mother of pearl’s looking like silver and the moon’s appearing double.
But how can there be superimposition of objects and their properties on the
Inner Self, the transcendental subject, that is never an object? One superimposes an
object on another object that is present, and you say that the inner self is never an
object.
We reply that the denial that the Inner Self is an object requires some qualification
because it may be intimated by the concept ‘I’ and because it is familiar in virtue of
its immediacy. It is not a rule that people only superimpose an object on another one
that is present: people superimpose colour on the sky. So there is no contradiction
in the superimposition of what is not self on the Inner Self. Superimposition as
defined above the learned consider as ignorance (avidyā). They say that knowledge
is the ascertainment of the true nature of an entity by means of discrimination.
The two-way superimposition of subjective and objective that is called ignorance

is the precondition of all religious and secular activities, of all behaviour involving
objects and means of knowing (pramān.a), and of all the scriptures whether con-
cerned with ritual injunctions, prohibitions or liberation.
But how can the scriptures and means of knowledge relate to what is infected by
ignorance?
We reply that the means of knowing cannot function unless there is an individual
knowing subject that has misidentified itself as ‘I’ and ‘mine’ in relation to the body,
mind and sense-faculties. There can be no perception without sense-faculties and no
operation of the sense-faculties without a body. And no one acts unless the body has
been superimposed upon the soul. In the absence of the two-way superimposition
of soul and not-soul, the Inner Self would not misconceive itself as an individual
agent of knowing. And without the agent of knowing, the means of knowing can-
not operate. In this way, the scriptures and means of knowing relate to what is
infected by ignorance.
Where scripturally ordained religious activities are concerned, the suitably quali-
fied individual is aware that he is related to higher worlds. But that qualification
for religious practice does not derive from a Vedānta-based understanding of the
Inner Self’s not being involved in rebirth and its exemption from caste-status. Prior
to knowledge of the true nature of the transcendental subject, scripture continues
to operate with regard to matters infected by ignorance. Scriptures such as, ‘A
Brahmin must sacrifice’ function only if there is superimposition on the transcend-
ental subject of such specifics as caste, stage of life, age and circumstances.
One superimposes external features on the Inner Self when one thinks, ‘I am doing
well’ or ‘Things are going badly’ with respect to factors such as one’s family. One
superimposes physical features when one thinks, ‘I am fat’ or ‘I am walking’. One
superimposes properties of the mind such as desires, intentions and judgments.
In these ways, the ego is superimposed upon the transcendental subject that is
really the passive witness of experiences. Conversely, the conscious nature of the
subject is superimposed on the mind and so forth.
Thus there is a beginningless and endless natural process of superimposition,

whose nature is misconception, which creates agents and experiencers, and is
directly known to everyone.
Authentic Being
For Śam.kara, the fact that awareness occurs as the same in all cognitions shows
that it is the basic reality. In his commentary on Bhagavad Gītā 2.16. (‘Of the
non-existent, there is no coming into being: there is no ceasing to be of
the existent. The difference between the two is seen by those who understand
the truth.’) Śam.kara formulates a substantial conception of ‘Being’ as the
changeless basis of all finite and transitory beings. ‘Being’ thus understood
is not involved in causal relations and is outside space and time. In the
background here is a reaction to Dharmakīrti’s espousal of dynamic causal
efficacy as criterial of reality, with its entailment that anything permanent,
in so far as it is static, would be unreal.
Unconditioned being is the foundational cause of the cosmos. Its genuine
features (jātīyaka-dharma), as opposed to those conditions (upādhi) that we,
or some scriptures, might superimpose upon it are universal presence, etern-
ity, omniscience, omnipotence and its being the true identity of everyone.
Nothing finite and nothing whose existence depends upon its relation to other
things can be truly real. The same applies to anything that is a product.
Products are transformations (vikāra) and a transformation (being a deriva-
tion) lacks a nature of its own. Thought expresses something truly real when it
is constant and unvarying, and when a thought expresses something that
comes and goes, then it concerns what is other than Being.
Bhagavad Gītā 2.17: Asks the question, ‘What is “Being” that exists
timelessly?’ Śam.kara replies that it is the condition of the possibility of
anything: Everything is pervaded by the Brahman, called Being. This Reality
never deviates from the nature that is its own because it has no parts. It cannot
suffer loss because it has no properties. It is the true identity (ātman) of every-
one. Commenting on Gītā 2.18, he says that this timeless, changeless and
indestructible identity cannot be determined by any of the means of knowing
(pramān.a). Dismissing the suggestion that it is determined by scripture and by
perception, he responds that transcendental subjectivity establishes its own
existence (svatah. siddha). It is only if such subjectivity, qua knowing subject,
is already a given that inquiry using pramān.as by one who seeks to know is
possible. We could not act with a view to understanding knowable objects
unless we had already understood ourselves as subjects. The transcendental
subject is a given, a presupposition of the sort of experience that we have.
Scripture is the ultimate pramān.a in the sense that it has its epistemic authority
about the transcendental subject only by stripping away properties that have
been mistakenly superimposed upon that identity, and not by providing new
information to us about something previously unknown.
From Śam.kara’s Upadeśa Sāhasrī (p.68):
Permanently unchanging consciousness, whose nature is self-luminosity, establishes
its own existence since it does not need a means of knowing for itself. [Everything in
the world requires consciousness to reveal or establish its existence. But conscious-
ness is unique in not needing anything external to establish it: it is self-revealing.]
Anything other [than consciousness] that is insentient exists for the sake of some-
thing else, since it functions in complexes. And in so far as that which exists for the
sake of the something else produces experiences of pleasure, pain and delusion, it
follows that it is what is not the self. Hence its existence (astitva) is not authentically
real (parama¯rtha). Just as the contents of illusions have no real existence apart from
consciousness, so everyday experience of differentiation has no real existence apart
from consciousness. And the permanently unchanging nature of the light of con-
sciousness, the authentic reality, follows from its uninterrupted presence. It is the state
of non-differentiation because it is unvarying in the midst of all the different ideas
presented to consciousness, while the different ideas come and go. Just as the differ-
ent ideas that come and go in dreams are said not to exist in reality, so the different
ideas inconstantly presented to waking consciousness must be unreal too. Because
there is no perspective on consciousness other than that of consciousness itself, it is
not the sort of thing that can be accepted or rejected, and there is nothing else.
Brahma-Sūtra-Bhās.ya 1.1.31
Scriptures such as ‘tat tvam asi’ (‘You are that’) and ‘aham . brahma asmi’ (‘I am the
Brahman’) show that what is called the ‘individual self’ is not ultimately distinct from
the Brahman. It is the Brahman that is called ‘individual person’ the agent and the
experiencer when it is regarded as diversified a result of properties (upādhi) super-
imposed upon it, such as body mind and sense-faculties. Some passages reveal the
true nature of Pure Being by negating differentia resulting from the superimposition
of features. They focus attention on one’s inner identity (pratyag-ātman).
Brahma-Sūtra-Bhās.ya 2.3.7 – ‘Where there is a modification, there is division – as

in everyday life.’
The plural world consisting of entities that are modifications (vikāra) is character-
ised by separations between entities, but what is not a product (avikr.tam) is not
found to be divided. The atmosphere is understood as differentiated from the
earth and so on. Hence it must be an effect. It follows that space, time, the mind
and the atoms are effects.
– But surely the Soul (ātman) is also divided from the atmosphere etc, so must it
not also be an effect?
The Soul is not a modification of anything because that would mean that it is an
effect.
We cannot deny the reality of the Soul, because of its very nature. The Self is not
a contingent property of anything else since it establishes its own existence and
that existence is not established by any of the means of knowing (pramān.a). One
uses the means of knowing to establish previously unknown objects of know-
ledge. But the Self, because it is the basis of the exercise of the means of knowing,
is established prior to their functioning. The rejection of a reality of that sort is
impossible. A person may deny the existence of some entity or another, but he
cannot deny the subject doing the rejecting.
In expressions such as ‘I know something present’, ‘I have known something past’

and ‘I shall know something future, the objects known differ as present past and
future, but the knower does not change because by nature it is wholly
present through the entire course of experience.’
The point here is that although one’s worldly experiences occur in temporal
sequence, it is a datum of consciousness that one’s core identity is not extended
in time: it is not constituted by temporal parts or phases.
Most followers of the Nyāya-Vaiśes.ika tradition hold that while we have
or are persisting identities, those identities are not intrinsically conscious.
They say that cognitions (and feelings, desires and intentions) are contingent
properties that occur when the embodied ātman is connected with a mind
(manas), the material faculty that co-ordinates sensory data. Śam.kara addresses
this view under Brahma-Sūtra-Bhās.ya 2.3.18:
[The opposing view]: If consciousness were a permanent feature it would persist

in the states of deep sleep, swoon and demonic possession. But people report
that they did not experience anything in such states. But in normal life people are
actively intelligent. Hence, given that consciousness is not a permanency, the
ātman is only contingently conscious.
We reply that knowing consciousness is permanent and it is one’s real identity

because it is not produced. Unchanging pure being exists as the individual person
when it is associated with superimposed properties. Scripture reveals that
consciousness is the true nature of pure being in definitional passages like, ‘The
Brahman is knowledge and bliss’ and ‘The Brahman is reality, consciousness,
infinite’. If the individual really shares the nature of the unconditioned reality, then
permanent consciousness is its essential nature.
As long as one’s identity is that of a transmigrator, as long as the sam . sāric state
has not been terminated by realisation of one’s true identity, association of the
soul with some mind and personality continues. And while there is a relation
to the superimposed property (upādhi) that is the mind then there is individual
personality and the series of births. But in reality there is no individual person apart
from the nature that is fabricated by relation to the superimposed property that is
the mind. This relation of the superimposed property that is the mind to one’s real
identity presupposes misconception (mithyājñāna) and misconception continues
until knowledge arises. So as long as there is no comprehension of Pure Being,
connection with the superimposed property that is the mind obtains.
Brahma-Sūtra-Bhās.ya 2.3.40 answers the question whether agency is innate

and natural (svābhāvikam) to one’s identity, or is it a superimposed property?
Agency does not belong to one’s true identity because it would follow that there
could be no liberation from rebirth. Were agency an essential feature of identity,
it could never be separated from it, just as heat is inseparable from fire. The
ultimate human good cannot happen unless one is free from agency because
agency always involves suffering . . . Agency is not essential because it is one of
the properties superimposed upon the fundamental identity.
The enlightened realise that there is no individual person; no agent and no enjoyer
of experiences, all of which are concepts superimposed upon the true self.
But surely if there are no real personal agents distinct from body, mind and sense-
faculties and distinct from the higher self, then it must be the higher self that is
the transmigrator, the agent and the enjoyer.
No – because agency and experience are presentations due to avidyā. Scriptures

reveal that being and agent and experiencer belong to the sphere of avidyā.
Agency is not an essential feature of the Soul.
It could be argued that agency must be an essential feature of one’s identity

because otherwise the scriptures enjoining actions would have no purpose.
We respond that scriptural injunctions teach that certain things are to be done and
thus presuppose an appropriate sort of agency. But it is not the case that this agency
is essential to the soul, because the scriptures teach that one is identical with uncon-
ditioned being. The scriptural injunctions presuppose the type of agency that is fab-
ricated by ignorance. Passages such as, ‘the agent, the person whose nature is
understanding’ (Praśna Up. 4.9) refer to the agency fabricated by ignorance.
The painful experience of the individual person is not something truly real but is
occasioned by a mistaken failure to discriminate between one’s true identity and
the superimposed properties (mind, body and sense-faculties) that are fabricated
by ignorance.
It is a cardinal Advaitin tenet that the notion of difference does not apply to the
Brahman or ‘pure being’. There is nothing else from which it could differ, and
it has no intrinsic complexity. It is just undifferentiated static consciousness.
Some scriptural passages refer to the Brahman as possessing differentia (sa-
viśes.am) while other say that it lacks differentia (nirviśes.am). Does this mean that
the Brahman has both a conditioned and an unconditioned nature?
Śam
. kara responds that the unconditioned reality cannot intrinsically possess
two natures because it is illogical that one and the same reality should both
intrinsically have and lack characteristics such as colour and shape. Relation to
superimposed properties (upādhi) does not involve a change in the real nature
of an entity. A brilliant crystal does not become dim by being related to a
projected red feature. And the superimposed properties in relation to the
Brahman are projected by avidyā. We must understand that of the two sorts
of characteristics, the one of the Brahman as void of every differentia and
beyond discursive thought is the true one.
Brahma-Sūtra-Bhās.ya 3.2.14–15
The Brahman has no forms (ākāra) such as colour and shape. As scripture says,
‘The Brahman is without before and after. There is nothing inside or outside
of it. The Brahman is the identity that experiences everything’ (Br.hadāran.yaka
Upanis.ad 2.5.19.) These passages refer to the transcendent (nis.prapañca) nature
of pure being, so it must be understood that it is formless.
Other scriptural passages that refer to the Brahman as having form are not primar-
ily about the Brahman, but are instructions to contemplate the supreme reality in
certain ways.
There is no problem about the fact that some texts teach meditation on the Brah-
man as having some specific forms. This sort of attribution of characteristics does
not compromise our view that the Brahman does not have a twofold nature
although properties are superimposed upon it. When something is due to a
superimposed property (upādhi), it cannot be a genuine property of an entity. And
the superimposed properties are fabricated due to ignorance. We have already
explained that primal ignorance is the precondition of all religious and secular
dealings.
Because the nature of unconditioned pure being is consciousness, void of
differentia, beyond mind and language and conveyed by the negation of all
finite characteristics, the scriptures teaching freedom from rebirth use the simile
of the sun reflected in water, meaning that the Brahman’s having different
features is not the real truth because those features are properties that have
been superimposed.
Being and consciousness coincide in the Brahman. They are not distinct properties.
The supposition that the Brahman has a mode of being that excludes conscious-
ness and another mode that has the form of consciousness that is other than
being implies that it is internally complex. Being is just consciousness and
consciousness is just being. They are not mutually exclusive, so conceptual analysis
(vikalpa) about whether the Brahman is either Being or consciousness or both is
groundless. Scriptural texts that speak of the Brahman under certain forms have
their own positive purpose: they do not merely have the significance of denying
that finite features of the cosmos pertain to the Brahman. When features of the
cosmos are mentioned in passages enjoining meditation – such as, ‘It is made of
mind: the vital breaths are its body; its appearance is light’ the text does not have
the purpose of suppressing plurality but that of enjoining meditation.
The inexpressibility of the Brahman

We have seen that one’s true identity as the Brahman, and the unreality of all
differences and individuality are revealed by some of the scriptures. But
the nature of the Absolute state cannot really be expressed by concepts and
words. Śam.kara understands Br.hadāran.yaka Upanis.ad 2.3.6 as referring to the
Brahman and reads the passage as, ‘Now there is the teaching, “It is not this.
It is not that”. There is no better expression than, “not this”. This is the designa-
tion of the truth about reality.’ The formula is meaningful by eliminating all
limiting conditions.
It expresses something that has no distinguishing features (name form, actions,

differences, class-property or qualities) that are the reasons for the application
of words. Brahman has no distinguishing features. Therefore it is not possible
to describe it as such and such. Brahman may be described by means of names,
forms and actions that are superimposed upon it. But when we want to express
its proper form that is devoid of every specific limiting condition, then it is not
possible to describe it in any way. There is only one way left – namely, the designa-
tion, ‘Not this, not this’ i.e. by the negation of all possible descriptions.
Bhagavad Gı̄tā-Bhās.ya 13.12b

Unconditioned Absolute is beginningless. It is not said to be Being or non-Being.
This text is tremendously influential in the Advaitin tradition. Among other

things, it was taken as stating that there can be no continuity of being
(analogia entis) between the Brahman and anything belonging to the cosmos.
The formula ‘Inexpressible as being or non-being’ was also applied to avidyā
and its works. What Śam.kara actually says here is that it means that the
Absolute is not the sort of empirical thing that either could or could not exist.
It is not knowable by conventional means adapted to our world, but only by
Scripture. He also says:
It stands to reason that the unconditioned reality (Brahman) cannot be directly

expressed by words such as ‘being’ and ‘non-being’ for words are ordinarily
used to reveal some object, and when they are heard they convey a conventional
meaning, by expressing the kind to which an object belongs, or some action,
quality or relation. Thus ‘cow’ and ‘horse’ express kinds, ‘he cooks’ and ‘he reads’
express actions, ‘white’ and ‘black’ express qualities, and ‘having wealth’ and
‘possessing cows’ express relations. But Brahman does not belong to a kind, so
it cannot be expressed by words such as ‘being’ or ‘non-being’. Because it is
without characteristics, it has no qualities such that it might be expressed by
words for them. Because it is unchanging it cannot be expressed by action-words.
Because it is non-relational, unique, non-objectifiable and the Inner Self of all, it
cannot be expressed by any word.
What about those Scriptures that speak of the Absolute in anthropomorphic

terms? The answer is:
That which is totally other than the cosmos is explained by the provisional attribu-
tion of features to it followed by a demonstration that they are inappropriate.
The path of active religious practices

is insufficient for enlightenment
All Vedāntins hold that the scriptures, meaning the Vedas and the Upanis.ads,
are the only means of knowing (pramān.a) the Brahman, the unconditioned
absolute reality, the Being of beings. Śam.kara thinks that some scriptural
passages generate an unmediated intuition into the identity of the true self
and the Brahman. Knowledge is enough: it is the necessary and sufficient con-
dition of liberation. This soteriology is resisted by those who think that the
scriptures are primarily about religious practices, rituals and types of medita-
tion, as the way to liberation. The scriptures tell us what do and how to do it.
They command us to bring about states of affairs. They do not provide infor-
mation or state facts about already established realities, except when such
information contributes to the accomplishment of religious activities. State-
ments of fact are meaningful only as supplementary to action-commands. The
scriptures do not teach about already established realities, since those are
within the province of pramān.as like perception and inference. Some Vedantins,
such as Man.d.ana Miśra, understood the function of scripture in this frame-
work. They thought that the scriptures bear upon the Brahman by treating it
as the object of meditation. That is, the scriptural statements primarily enjoin
the activity of meditation. The Brahman is the focus of meditation. So the
scriptures treat of Brahman only indirectly as something that is to be realized
by religious activity. Śam.kara rejects this outlook. Actions are always oriented
towards results. They presuppose duality and individuality. The framework of
means and ends and the associated instrumentalist mentality are part and
parcel of the perpetuation of sam.sāra. Liberation is simply the manifestation
of the soul’s true identity. It is not something to be produced or obtained and
thus is not connected with the performance of acts, be they ritual perform-
ances or meditations. The importance of this topic is reflected in the extent of
his commentary on Brahma-Sūtra-Bhās.ya 1.1.4.:
The Brahman, the omniscient and omnipotent foundational cause of the

origination, stasis and dissolutions of the universe, is understood from the
Upanis.ads because all their statements agree in conveying that reality as their
meaning. For instance there are:
‘In the beginning, this cosmos was being alone, one without anything else’ and
‘In the beginning, this cosmos was the one ātman.’ Once it has been understood
that the words in such passages cohere with each other in bearing upon the real
nature of the Brahman, it would be mistaken to assume another meaning, for
that would involve imagining what is not scriptural teaching and abandoning the
scriptural teaching. It is not the case that such fact-asserting passages are to be
understood as expounding the natures of agents [involved in ritual performances]
because there are scriptural texts that repudiate the fruits of ritual activities, such
as ‘Whom might one see and by what means?’
The Brahman, although it is a fully accomplished reality, does not fall within the
province of means of knowing such as perception and inference, because the
identity of the Brahman as one’s true self can only be known from the scripture,
‘That thou art’.
As for the view that the Vedāntic teaching is meaningless in that it is not
concerned with matters to be actively pursued or avoided – this is not a problem
for us who say that the ultimate good is achieved, after the destruction of all
afflictions, merely by knowledge of one’s identity as the Brahman, which is
not something to be pursued or avoided. It is true that there are subordinate
references to divinities and individuals in scriptural passages primarily enjoining

the practice of meditation, but this does not contradict our position. The Brahman
cannot be subordinate to injunctions to meditate because once one has realised
that one’s true identity is the Brahman, there is an end to thinking in dualistic
terms about rites and constituents of events and there is nothing to be pursued or
avoided. Once the dualistic mentality, which thinks of the Brahman as supplement-
ary to injunctions to meditate, has been eradicated by understanding identity [of
the Brahman and the self], it cannot arise again.
While some Vedic passages have authority by being injunctive, one cannot impugn
the authority of those scriptures that have knowledge of the self as their result.
The epistemic authority of scripture cannot be known by inference, because there
are no analogous instances that could be cited as part of the argument. Thus it is
settled that scripture is the means of knowing the Brahman.
***
An opposing view, from someone like Man.d.ana Miśra who thinks that liberation
is the fruit of meditation, rather than just knowledge:
The purport of scripture is not mere understanding but religious activity, in this
case meditation or contemplation.
Reply: Although scripture is the means of knowing about the Brahman, still
scripture teaches that the Brahman is something that should be meditated upon
(pratipatti-vidhi-vis.ayatā), in the way in which certain factors (e.g. the sacrificial
post) are taught as subordinate aspects of ritual activities. This is because scripture
is concerned with ritual actions and abstentions. Scripture is meaningful by
moving people to action or by restraining them. Anything else is relevant in so
far as it is supplementary to action-injunctions.
In the same way, the Upanis.adic statements are meaningful. Given that scripture
is injunctive, just rituals such as the Agnihotra are enjoined for the person who is
intent on paradise, so knowledge of the Brahman is enjoined as an activity for the
one intent upon immortality.
Preliminary reply: There is a radical distinction between two kinds of inquiry: in the
ritual portions of scripture, the ritual duty that one wants to know is something
that is to be brought into being, but our quest is for the Brahman that is always
and already fully accomplished. The fruit of knowledge of the Brahman must be
distinct from the fruit of the knowledge of ritual duty that depends upon
performances.
Response from the opponent: No, there is no difference because the Brahman is
taught as connected with injunctions to perform actions by texts such as ‘the Self
is to be visualised’ and ‘Everyone should meditate upon the Self’. The injunctions
stimulate the desire to know the natures the Self and the Brahman and the
Upanis.ads apply by teaching its proper form as eternally omniscient, all-pervasive,
pure knowledge, liberated, consciousness and bliss. By meditating on that proper
form there arises freedom from rebirth, unknown by ordinary means but known
from scripture. If the Upanis.ads taught established facts without reference to
injunctions to perform ritual actions, they would be meaningless because not
connected with things to be pursued or avoided.
A query: But a fact-assertive statement such as, ‘this is a rope, not a snake’ is seen
to be meaningful by removing fear produced by a misperception. Likewise, the
Upanis.ads are meaningful by removing the misconception about transmigration
when they teach the reality of the non-transmigrating Self.
Reply: this would be the case if the misconception about transmigration were
removed merely by hearing about the proper form of the Brahman, just as the
mistake about the snake is removed just by hearing about the rope. But it does
not cease. Although the Brahman has been heard about, features of transmigrat-
ory life are seen to continue as before. This is why there are injunctions that one
should meditate, after hearing about the Brahman.
Hence, it must be understood that scripture is a means of knowing the Brahman

in that it is the object to injunctions to meditate.
Śam. kara now replies: The above view is mistaken because of the radical difference
between the results of the knowledge of the Brahman and knowledge about rit-
ual actions. Actions called duty (dharma) are known from the scriptures. This is the
province of the Mı̄mām . sā Sūtras, which also tell us what to do and what not. The
consequences of right and wrong acts, success and failure, produced by contact
with the objects of the senses, are perceptible pleasures and pains experienced
physically and are known to apply to all creatures from the creator deity Brahma
down to inanimate things. Scripture teaches that there is a hierarchy of pleasures
amongst living beings and from these a hierarchy of dharmas is inferred. From the
hierarchy of dharmas there is a hierarchy of qualified practitioners. The latter
accords with what people aim at and their ability to pay. Those who perform sac-
rifices for the public good follow the higher path because of their special know-
ledge. The southern path is followed by those who perform rituals for themselves.
That there are gradations (tāratmya) in enjoyments in those superior realms is
known from the scripture, ‘dwelling there until merit is used up’. Likewise we
know that degrees of enjoyments amongst terrestrial beings (and below) are con-
sequences of dharmas indicated by Vedic mandates. The gradation in embodied
pleasures and pains, occasioned by the hierarchy of dharma and adharma, on the
part of those subject to defects such as ignorance is known from scripture and
reasoning to be the nature of transmigratory existence. ‘There is no end to pleas-
ure and pain for the embodied one’ refers to the nature of transmigration as just
portrayed. ‘Pleasure and pain do not touch the disembodied’ teaches that the
disembodied state called liberation (moks.a) is not the product of the right actions
specified by Vedic mandates. Liberation cannot be the product of the performance
of ritual duty, since liberation is the natural state of the soul. Scripture teaches that
the eternally disembodied state called liberation is utterly different from the fruits
of prescribed actions.
Some permanent things, such as earth and the qualities posited by the Sām . khyas,
may change and still retain their identity. But the permanence we are talking
about is absolute, all pervading like the atmosphere, free from every sort of
modification, self-sufficient, impartite and self-illuminating. Where merit, demerit,
their results and the passage of time do not apply, that is the disembodied reality
called liberation. It is the same as the Brahman. Were that being taught as
something subordinate to the performance of actions, and if liberation were to
be accomplished by prescribed actions, it would be impermanent. In fact, it would
be at the top of the hierarchy of impermanent states that are the fruits of action.
But everyone agrees that liberation is permanent. Thus it is illogical that the
Brahman is taught as subordinate to prescribed actions.
Many scriptures such as ‘He who knows the Brahman becomes that Brahman’
teach that liberation is the immediate consequence of knowledge of the Brahman
and rule out any action intervening. No prescribed action intervenes between
the vision of the Brahman and the realisation of pure consciousness as the nature
of everything. Other passages reveal that the sole result of knowledge of the
self’s identity with the Brahman is just the removal of obstacles to liberation. The
Nyāya-Sūtras say the same: release occurs immediately after the destruction of
misconceptions. . . .’ There is removal of misconception as a result of knowledge
of the identity of the self and the Brahman. . . . Knowledge of the Brahman does
not depend upon human activity. Rather, it depends upon reality, like knowledge
of mind-independent entities that are objects of pramān.as such as perception.
[‘knowledge of an entity as it is in itself does not depend upon human ideas
but only on the reality itself . . . knowledge of established entities depends upon
reality.’ Brahma-Sūtra-Bhās.ya.1.1.2]. We cannot rationally suppose that such an
Unconditioned Reality, or knowledge of it, has any connection with things to be
brought about. The Brahman cannot be something to be brought about as if it
were the object of the action of knowing. Scriptures declare that the Brahman is
not within the scope of knowing or contemplation.
But if the Brahman is not an object, how can scripture be the source of knowledge
about it?
We reply that scripture has the force of removing differences fabricated by

ignorance. Scripture does not intend to teach that the Brahman is a specific object
(belonging to a kind of things). Rather, scripture leads the mind away from
differentiation, such as that between objects, subjects and acts of knowledge,
that has been fabricated by ignorance, teaching that the Brahman is never an
object because it is one’s own inner reality. . . . Because scripture teaches the
proper form of the eternally released self by dispelling the belief, fabricated by
ignorance, that transmigration is a reality, we cannot be accused of holding that
liberation is a transitory state. It is logical for someone who holds that liberation is
something to be produced, or that it is a transformation of the self, to say that
is depends upon mental, verbal or physical effort. In either case the non-
permanence of freedom from rebirth is the certain conclusion.
Freedom from rebirth does not depend upon action as if it were something to be
accomplished. It is not something to be accomplished because it is the already and
always real nature of one’s identity. Even if the Brahman were different from one’s
true identity, it would not be something to be attained because it is all pervading
and always present to everyone.
Release is not the product of ritual purification, which would make it depend upon
human activity. Such purification comes about by the removal of defects or the
acquisition of virtues. The latter cannot pertain to liberation since it is the nature of
the Brahman to which no perfection need be supplied, and there are no defects to
be removed. If you say that release is a hidden feature of one’s self that is mani-
fested when the self purifies itself by action, we deny this since the self cannot be
the substrate of actions. Actions do not exist without modifying their substrates. If
the self were changed by action inhering in it, impermanence of the self would
result. Hence actions cannot inhere in the true self. The inner self cannot be puri-
fied by an action belonging to something external, because it is never an object.
The embodied self may be purified by actions but that which is purified is a self
that has been possessed by ignorance and confused with the body. It is this
personality that considers itself purified by ritual acts. All actions are performed by
the personality that understands itself as an individual centre of consciousness and
which enjoys the fruits of actions.
Release is not the product of ritual purification because it is just being the
Brahman. It has no connection with actions and is the fruit of knowledge alone.
But is not knowing a mental action?
There is an important difference. When an action such as contemplating is man-

dated, it is independent of facts and dependent upon human mental effort.
Although meditation and reflection are mental, whether they are performed or not
is a matter of human choice. But knowledge is produced by the valid instruments
of knowing. The instruments relate to entities as they are in themselves. Thus it is
not possible to create, not create, or change knowledge since it depends only upon
already established reality, and not upon Vedic mandates or human minds. To illus-
trate: when the Veda tells us to contemplate man or woman as fire, that is an
action since it is generated by an injunction, and it depends upon some human
choice. But the concept of fire depends neither upon Vedic mandate nor human
choice. It is a matter of knowledge not action because it depends upon an entity
that is the object of perception. The same is to be understood with respect to all
the entities that are objects of the valid means of knowing. This being the case, the
knowledge of the Brahman as it is does not depend upon Vedic mandates.
But what is the meaning of passages that look like mandates, such as, ‘The Self is
to be seen, to be heard about’?
We reply that they have the force of diverting attention from natural everyday
activities. A person preoccupied with externals, pursuing the objects of desires,
does not achieve the ultimate human good. Passages such as ‘the self is to be
seen’ actuate a person who seeks the ultimate good to direct his mind towards
the inner self, distracting his attention from mundane activities.
Śam
. kara and the Buddhists
Śam.kara is accused by some opponents of being a ‘closet Buddhist’. Buddhists
and Advaitins agree that the notion of selfhood is illusory because constructed
out of the interactions between our modes of consciousness and the world.
Moreover, like the Buddhists Śam.kara envisages the evaporation of personal
individuality once enlightenment dawns, and blames suffering on ignorance.
But the accusation is far from the truth. Śam.kara’s metaphysic is totally differ-
ent from the Buddhist temporalism that rejects the very notion of enduring
identities in favour of successions of phases. Śam.kara believes that ‘behind’ the
array of changing phenomena there is a single unconditioned reality: the static
co-incidence of pure being and consciousness. Relative to Unconditioned
Being, the world that we experience is less than fully real, not the genuine art-
icle, but there is an ultimate reality enjoyed by depersonalized consciousness.
In other words, the cosmos has a real cause (sat-kāran.a-vāda), even if we must
be agnostic about the ontological status of entities that ‘cannot be determined
as either real or unreal’. This is because every phenomenon is ultimately unreal
when considered as individual, but real in so far as it participates in the general
reality or the Brahman.
Let us see what he says about various Buddhist schools:
There are three traditions of Buddhist thought:
Those who say that both material and mental phenomena are real.
Those who say that only mental phenomena are real.
Those who say that there are no intrinsic natures.
To begin with, we refute those who admit the reality of all mental and physical
factors. By physical they mean both the four elements and the sense faculties and
their respective objects. They say that earth, water, air and fire are combinations of
four different kinds of atoms. The five constituents (skandha) that make up human
lives (body, perceptions, feelings, conceptual thoughts and inherited dispositions)
are internal and in combination form the basis of all interpersonal dealings.
Here we object that there are two different kinds of combinations: but the
reality of these sorts of combinations is unintelligible. This is because the atomic
components of the material combinations are non-conscious and the emergence
of sentience depends upon the prior existence of some compound. They do not
accept any other persisting conscious subject or director who could combine the
basic factors. It cannot be the case that the atoms and skandhas function sponta-
neously because that would entail that they would never cease from activity.
We see here the basis of one of the most significant objections to the Buddhist
reductionist analysis. It may appear economical, plausible and attractive but it
is hard to see how after completing the reduction of entities into their elements
there is any way back. It is easy to smash a glass, but impossible to put the
pieces back together again. It is not clear that Buddhism can account for the
emergence of entities, including the person, from the elements.
The Buddhist says that although there is no persisting subject of experience
or director who combines the basic factors, the world process is sustained by
the interactive causality of dependently originating factors such as ignorance
etc. Ignorance, habits, perceptions, name and form, the six types of sensation,
touch, feeling, grasping, birth, old age and death, sadness, pain, frustration and
discontent form a self-perpetuating circle of causes and effects. The reality of
those facts of life is accepted by everyone. The cycle of factors, each conditioning
the other as effect and cause, presupposes that there are real combinations [such
as bodies and minds].
Śam
. kara replies:
You are only talking about the originating causes of the elements in the series
and overlooking the sort of organisational causality that would account for the
formation of combinations. The latter is impossible if there are only momentary
atoms and no subjects of experience. Perhaps the factors beginning with ignor-
ance are the causes of the formation of aggregates. But how can they cause that
which is the necessary condition of their existence?
Moreover, you do not think that the combinations are formed in the interests of
enduring conscious subjects so that they might experience the fruits of their
karma (bhogārtham). Hence, experience is just for the sake of experience and is
not sought by anything or anyone else. So freedom from rebirth (moks.a) is just
for the sake of itself and there is no one by whom it is sought. A being with an
interest in both experiencing the fruits of action and gaining freedom from rebirth
would have to exist contemporaneously with those processes and such persist-
ence would conflict with your belief in the instantaneity of beings (ks.an.ikatva).
So while there may be a relation of originative causality between the members

of the series, this does not suffice to explain their organized combinations.
Śam.kara continues to establish the stronger claim that the Buddhist position
cannot even make sense of originative causality between members of the
series. He begins by arguing that a strictly instantaneous reality does not last
for long enough to bring about the existence of anything else.
The theory that realities are instantaneous implies that when the later moment
originates, the earlier one no longer exists. So it is not possible to establish the
relation of cause and effect between the two occasions. The claim that once
the prior moment is fully actualised it becomes the cause of the later one is
not intelligible because the hypothesis that a fully actualised entity has a causal
function (vyāpāra) means that it is connected to another moment [so it is not
strictly an instantaneous occurrence but an extended one].
Nor does it make sense to say that the causal function simply is the existence of the
prior entity. This is because origin of an effect that is not tinged by the own-nature
(svabhāva) of its cause is impossible. If one accepts that the effect is tinged by the
own-nature of its cause, the nature of the cause continues in the effect and that
entails the abandonment of the hypothesis of instantaneity. Nor can it be argued
that there could be a cause-effect relation without the nature of the cause affect-
ing the effect because if that were applied in all cases there would be chaos.
Moreover, what you understand as simultaneous ‘origin and cessation’ would be

either the same as the proper form of an entity, or two phases of its existence, or
something else. If they are the same as the proper form, the words ‘origin and
cessation’ and ‘entity’ would be synonymous. If the terms ‘origin’ and ‘cessation’
mean two phases that are the beginning and end of an entity whose existence is
what occurs between them, the acceptance of instantaneity of the entity would
be overthrown because it would be connected with three phases – beginning,
middle and end.
If origin and cessation are quite other than the entity, it would follow that the
entity is everlasting.
The theory of the essential temporality of beings (ks.an.ikatva) extends to

human personality which the Buddhists understand as just a series of causally
related experiential phases. There is no persisting principle of identity or
soul that is a further fact over and above the stream. Śam.kara challenges the
intelligibility of this proposal:
Moreover the nihilist (Vaināśika – literally the believer in the spontaneous
destructibility of all entities) accepting the instantaneity of everything must
apply instantaneity to the subject of experiences. But that is not possible because
of the phenomenon first personal memory [of the form, ‘I remember that I
did that’]. Such memory is produced by the reoccurrence of an experiential
awareness and it is only possible if the one who remembers is the same as
the original subject. One man does not remember the awareness of another.
How could there be the experience of the awareness, ‘I saw that and am seeing it
now’ if there were no single subject seeing the earlier and the later? We all know
that the experience of recognition occurs only when there is a single subject of
both seeing and remembering. . . . The nihilist knows himself to be the one
subject of seeing and remembering whenever he thinks, ‘I saw that’. He does not
deny that the past perception belongs to him any more than he denies that fire is
hot and light.
Since one and the same agent is connected with the two moments of seeing
and remembering, the nihilist must give up his acceptance of the essential
temporality of beings.
If he recognises that all his past and future experiences belong to one and the
same subject, and accepts that there is sometimes synthetic awareness of both
successive and simultaneous cognitions, how can the nihilist who asserts universal
instantaneity maintain his position?
The Buddhist may say that recognition and synthetic awareness derive from the
similarity of the momentary cognitions [and this generates the misapprehension
that there are persisting object and the illusion that there is an enduring self]. But
similarity is a relation between two different things. Someone who says that
although there is no single perceiver of two similar things, synthetic awareness is
based on similarity is talking nonsense. If he admits that there is a single perceiver
of the similarity between the earlier and later moments, he thereby grants that
there is one thing enduring through two moments and this contradicts the
hypothesis of instantaneity.
When a universally accepted reality is denied by philosophers, whatever they may

say in support of their own view or in finding fault in that of others, they convince
neither themselves nor others. When it is know that something is such and such,
it must be expressed accordingly. Their thesis about similarity fails to accord with
the facts of experience. The act of recognition is an understanding about one and
the same thing and not of something that is similar to something else. It may be
that sometimes there is a doubt about whether an external object is the same one
or something similar to it. But there is no room for doubt about whether the
perceiving subject is identical to itself or just something similar.
Śam.kara now turns his critical attention to the Buddhist ‘consciousness-only’

theory that reductively identifies what are usually taken to be extra-mental
realities with elements of awareness. Of the various ingredients of the cosmic
process posited by the Buddhist schools, they say that only the mental ones
are real.
Brahma-Sūtra-Bhās.ya 2.2.28:
The Buddhist pūrva-paks.in: the Enlightened One taught the theory that the
external world really exists in consideration of those followers who were convinced
about the reality of things external to minds. But this was not his own belief,
which was that amongst the five constituents of personal existence (skandhas),
only perceptions were real. According to the ‘vijñāna-vāda’, we can make sense
of everything to do with means of, objects and results of knowing if they are
purely internal to minds. Even if there were external objects, the process of know-
ing would not get under way without mind. But how is it known that this entire
process is internal and that there are no objects independent of perceptions?
Because there cannot be external objects. If external objects are accepted, they
would be atoms or combinations of atoms, such as pillars. But atoms are not
discerned in our awareness of pillars etc because they cannot be represented in
consciousness. External objects cannot be combinations of atoms because we
cannot determine whether they are the same as or different from the atoms.
Moreover, although cognitions share the same nature in that they are just con-
sciousness, they may express different objects. This would not happen unless the
differentia were internal to awareness, so it must be accepted that a cognition has
the same form as its object (vis.aya-sārūpya). Once this is granted, given that the
representation of the object is determined by cognition (and not the other way
round), the postulation of external objects is superfluous. Moreover, given that
the object and the awareness of it always occur simultaneously (sahopalambha-
niyamād), it follows that there is no difference between a cognition and its object.
It is not that the case that where the cognition and its object are concerned, there
is apprehension of the one when there is non-apprehension of the other. This
would not make sense if the two were different in nature – in which case there
would be nothing to stop the one occurring without the other. Hence there are no
objects external to the mind.
Perception of objects is comparable to dreams. Just as ideas in dreams manifest

the form of the apprehender and the apprehended although there are no external
objects, so in the waking state one must understand that representations of solid
objects occur without external objects. This is because from the point of view of
felt experience, there is no difference between the forms of awareness.
Now if there are no objects external to minds, how is the variety of representations
(pratyaya-vaicitrya) explained? It is explained by the variety of residual impressions
(vāsanā-vaicitrya) left by previous ideas. In the beginningless series of births, an
impression causes a perception which in turn leaves an impression and so on. This
explains the variety of representations. Moreover, we understand by reasoning from
positive and negative concomitance that the variety of cognitions is just caused by
residual impressions. We both admit that in dreams and hallucinations, the variety
of cognitions is caused by residual impressions in the absence of external objects.
But we do not accept that there can ever be any variety of cognitions without
residual impressions. Hence there are no objects external to awareness.
Śam
. kara now replies:
It is impossible to make sense of the non-existence of external objects because

they are perceived. External objects, corresponding to representations (prati-
pratyayam), are perceived. It cannot be the case that what is being perceived does
not exist.
The Buddhist may claim, ‘I do not say that I am not aware of objects. What I do
say is that I am not aware of any object apart from perception.’ But objects
independent of perception must be accepted simply because of the nature
of perception itself. No one perceiving a pillar or a wall is just conscious of his
perception. But everyone perceives pillars and walls precisely in so far as they are
the objects of perceptions. Even those who deny the reality of external things
implicitly grant their existence when they say that representations internal to
consciousness (antar-jñeya-rūpam) appear as if external. If we accept that reality
is as it is given in direct experience, it is logical to accept that it is precisely the
external that is manifested in consciousness, but not what is like the external.
The Buddhist argues that the ‘external-like’ is what is manifest because of the
impossibility of external objects. This can’t be right because what is possible and
what is impossible is ascertained by the means of knowing (pramān.a) and the means
of knowing do not depend upon independently arrived at ideas of what we might
imagine to be possible and impossible. What is possible and what not is understood
by the use of some means of knowing. External objects are apprehended as they are
in themselves by all the means of knowing. How can it be said that they are not
possible on the basis of specious argumentation, given that they are perceived. And
it is not the case that there are no external objects because of the conformity
between cognitions and objects. If there were no objects, conformity between
the representation of the object in awareness and the external object would be
impossible. And the object is represented as external. That is why the co-occurrence
of thought and object (sahopalambha-niyama) is due to the fact that a relation of
mode of presentation and object – presented obtains between thought and object.
It does not derive from the identity of thoughts and objects.
Moreover, consciousness remains the same although conditioned by different

objects, such as a pot and a cloth. This is parallel to seeing a black cow and a
white cow. The individuals differ in respect of their colours but the generic prop-
erty cowness is constant and immutable. The distinct identity of the one constant
factor is established in comparison with the two and the distinct identities of the
two are established in comparison with the single factor. Hence, thought and
objects are distinct.
Moreover, two successive but discrete thoughts with a thinker, self-contained and
confined to their own instantaneous occurrences, cannot be related as the appre-
hending factor and the apprehended. It follows that all the Buddhist teachings are
lost because they involve inter-related ideas.
Man.d.ana Miśra
Śam.kara’s contemporary Man.d.ana Miśra is the other founding father of the
monistic Vedanta vision. His Brahma-Siddhi was as influential as Śam.kara’s
commentaries. Vacaspati Miśra attempted to reconcile the outlooks of the two
thinkers in his Bhamatī commentary on Śam.kara’s Brahma-Sūtra-Bhās.ya.
Man.d.ana differs from Śam.kara in seeing positive value in religious
practices. He thinks that Vedic rituals are purificatory and predispose one
to the realization of one’s true identity. He has no quarrel with the path of
world-renunciation (sam.nyāsa) but observes that it is difficult. He thinks
that the enlightenment received from scriptural statements about the truth
of non-duality must be intensified by ritual and contemplation in order to
counteract still forceful residual traces of ignorance.
Śam.kara’s inspiration is selfless contemplative experience, which shows that
tranquil consciousness is the self-revealing and self-establishing true nature of
reality. For Śam.kara, the fact that awareness occurs as the same in all cognitions
shows that it is the basic reality. In pure consciousness there is neither differ-
entiation nor individuality. This is called the Brahman, where consciousness
and Being coincide. Experience of differences between knowers, thoughts
and objectivity are fabricated through misunderstanding. Man.d.ana’s position
is somewhat different. It is what Paul Hacker called a ‘radical ontologism’. Put
simply, Man.d.ana does not put so much weight on considerations about the
nature of consciousness as Śam.kara. If we say that Śam.kara’s vision arose from
looking within, then we may say that Man.d.ana’s began from looking outside.
There is something there, whatever it may be. The foundational scriptural text
here is the Chāndogya Upanis.ad’s ‘In the beginning this world was just being,
one without a second’ (Ch.Up.6.2.1). Being (sattā) presents itself universally.
Being is present everywhere. Not being something, just being. Being is not dif-
ferentiated. Primary awareness, non-discursive and pre-reflective, reveals this
non-predicative being. Being is apprehended prior to the identification
of objects in respect of their general and specific features. Being is the core
identity of entities when we abstract away their properties and relations.
Everything is experienced as sharing the undifferentiated form of Being that
is always the same everywhere. Hence we have a monism of Being, rather than
one of consciousness. The Brahman is already known in immediate experience,
even though we inhabit the sphere of avidyā (misunderstanding), having lost
sight of the true nature of Reality.
It is difficult to capture exactly what ‘Being’ means other than to say that
it is the foundational reality of beings. It is that which is the unconditioned
condition of there being anything at all. Man.d.ana also says what it is not.
For example, it is more than actual and concrete entities that exercise causal
efficacy, which Dharmakīrti treats as the criterion of reality. Moreover, it is not
the susceptibility for being connected with some means of knowing or mode
of evidence (pramān.a). Such an epistemic account of being would restrict what
is to what can in principle be known or identified. Man.d.ana’s view is that Being
transcends knowability. The conception is richer than that expressed by the
existential ‘is’ that we use to assert that some entity is numbered among the
objects that furnish our world (‘There is a table here’). It is certainly stronger
than Frege’s suggestion that affirmations of existence are just denials that the
number zero applies to some concept. It is also more than is captured by the ‘is’
of predication, which we use to say that something is such and such (‘The table
is black’). Finally, Being is not an entity because it does not belong to a kind.
The Being of beings is the Brahman, that which is unconditioned by
particular features. It is known as such from the advaita-śrutis, which also
teach the unreality of diversity. But if scripture is a means of knowing a
non-dual reality, it is in conflict with everyday perception that apprehends
differentiated entities. Man.d.ana denies that there is a conflict: perception
apprehends pure being. The rationale for the denial that difference and
individuality are basic realities and for the claim that perception does not
grasp difference is extensively elaborated in arguments that will be developed
by later Advaitins, especially Vimuktātman [fl. 950 A.D.] and Śrī Hars.a
[fl.1150 A.D.]. We shall briefly mention a few points here.
Perception identifies the proper form of an entity: it refers to the thing
just as it is in itself. Separating it from other things comes next. We cannot
differentiate unless we have identified something in the first place. If differen-
tiation or exclusion (apoha) were the nature and function of perception when
we discern a particular entity, we would perceive the difference of the object
from everything else whether or not present in space and time. This is
manifestly impossible and contrary to experience. Difference from other
things cannot be the very nature (svabhāva) of an entity. Difference is
relational. If it is constitutive of the nature of an entity, it follows that the
entity is the same as that from which it differs. But if difference is not of
the nature of entities, they are not essentially different.
Another consideration is that if difference is the very nature of entities,
given that it is a form of mutual absence or non-existence (anyonya-abhāva –
one thing’s not being another – a reciprocal absence of natures), it follows that
entities do not exist.
It is objected that Man.d.ana is treating difference as a real feature of entities,
when it is a best a boundary. Difference is not a thing in its own right; it is not
a mode of being of an entity but is only falsely presented by constructive
cognitions (vikalpa). There is no property ‘being different’ that belongs to
entities because difference does not really exist – an imagined nature does not
really belong to an entity. Man.d.ana responds that this is exactly what he is
saying: difference does not really exist but is projected by beginningless avidyā.
He also considers the alternative that difference means the interdependence
of entities and not their natures. A proper form is unitary but entities differ
with respect to each other. Man.d.ana denies that interdependence is a genuine
property of entities (i.e. a property whose loss or gain means a real change in a
thing) by which they are constituted. It is illogical to hold that the continuing
existence of entities depends on other entities when their natures are consti-
tuted by their own specific causes. Interdependence is a human concept and
not something that belongs to things as they are in themselves.
After Dignāga, it became a standard view among Brahminical philosophers
that there are two varieties of perception: non-conceptual and conceptual. The
former is reception of whatever is given: the latter is the explicit identification
of features, both general and specific. As Kumārila put it:
In the first place here is cognition (jñāna) that is just seeing (ālocanā) and it is
free from concepts (nirvikalpaka). It is produced from the pure entity and is like
the cognitions of infants and the mute. Neither general (sāmānya) nor specific
features (viśes.a) figure explicitly in the content of awareness, but the individual
that is their substrate is grasped . . . A subsequent cognition by which an entity
is grasped in terms of its properties such as its universal and its qualities is also
considered a form of perception.
Man.d.ana denies that there can be two varieties of perception. Perception is a
means of knowing that just refers to Being. Every perception reveals the
general form of being that is the Brahman.
His position is:
Initially there is non-conceptual perception whose sphere is just Being (vastu-

mātra). It is the ensuing conceptual cognitions that comprehend particularities.
What Man.d.ana means is simply that in the first instance we just register the
presence of something really there. (If reality is basically simple, features of
entities such as universal property, qualities and the relations between them
are just products of conceptual superimposition.) Judgements that involve
conceptual or constructive cognition come next. But that is not perception as
a means of knowing. Indeed it is not knowing at all.
The rationale is: perception is a means of knowing. What are called
non-conceptual and conceptual perception are different functions with com-
pletely different kinds of objects. The one refers to undifferentiated pure
being, the other to particularities. He has shown that differentiation is not
genuinely real, so cognitions of particularities must be false. They are cases
of avidyā. Avidyā here means error or misconception (vibhrama). Error must
be about something. There can be no apprehension of the non-existent. So
constructive cognitions must be misunderstandings about pure Being.
Avidyā is not genuinely real – if it were it could never be eliminated – but it
is not totally unreal in that it is a familiar phenomenon. That is why it is
called ‘illusion’.
We beings are alienated from Being. This is avidyā. Avidyā is responsible for
all plurality of individual selves, cognitions and objects, and the concomitant
process of rebirth. It is also connected with sorrow, delusion and passions.
It conceals one’s true nature; instead creating the illusion that one is an indi-
vidual agent subject to ritual and social duties and transmigration. While later
Advaitins will distinguish between subjective avidyā that affects individuals
and avidyā as a positive cosmic force that projects diversity and conceals the
true nature of the Brahman, Man.d.ana, like Śam.kara makes more modest
claims. He says that avidyā belongs to individual selves. This is one of the
reasons why the subsequent tradition will posit a causative avidyā. Something
has to constitute the individual as an individual in the first place if it is to be
the substrate of avidyā.
The development of the tradition

The early Advaitins attribute the illusion of multiplicity to beginningless
avidyā. The epithet ‘beginningless’ means that we cannot explain its occur-
rence. But it also encouraged the thought that avidyā is some sort of effectual
reality in its own right. Arguments about avidyā led to the development
establishment of two schools of thought. One is called the Bhāmatī school
after Vācaspati Miśra’s commentary of the Brahma-Sūtra Bhās.ya. They say
that the individual soul is the locus of avidyā. It distinguishes a fundamental
causal avidyā from everyday mistakes and ignorance. Avidyā has two powers:
it conceals the truth and it projects the illusion of diversity. What is called the
Vivaran.a school follows Prakāśātman’s commentary on Padmapāda’s
Pañcapādikā. Padmapāda (700–750 A.D.) said that avidyā is the cause of all
misconception (and superimposition), while Śam.kara tends to treat them as
the same. He described avidya as a material (jad.a) force that is the underlying
cause of the world-appearance. Avidyā veils the pristine nature of the Brahman
and in association with the workings of karma and memory traces of previous
cognitions produces the illusion of limited selfhood that is the substrate of
individual experience and agency. He thought that the limited self is a finite
reflection (pratibimba) of Brahman. This school says that the Brahman is both
substrate and the object of beginningless avidyā, which is the substrative cause
of plurality. Avidyā is an actual entity (bhāva-rūpa) that is the opposite to
rather than just the absence of knowledge.
Further reading
Śam.kara’s Brahma-Sūtra-Bhās.ya is translated in Thibaut (1904). Mayeda (1979) translates the
Upadeśa-Sāhasrī and has a useful introduction.
Suthren Hirst (2005) is highly recommended for Śam.kara.
Potter (1981) has an introduction and summaries of works by Śam.kara, as well as Man.d.ana’s
Brahma-Siddhi by Allen Thrasher. The latter’s Advaita of Vedānta of Brahma-Siddhi is stimulating,
as are Ram-Prasad (2001 and 2002).
Halbfass (1995) collects important articles by Paul Hacker.
Suryanarayana Sastri (1971) translates a classic of Advaita epistemology and metaphysics. It is a pity
that he repeatedly translates ‘vr.tti’ (mental function) as ‘psychosis’.
Some aspects of the debates between the Advaitins and theists about scriptural exegesis are described
in Bartley, Theology of Rāmānuja (2002).
The Summary of the Text in Acharya (2006) is useful for later Advaita.
See Granoff (1978) for Śrī Hars.a.
For Bhartr.hari, see John Brough’s classic essays on ‘Theories of General Linguistics in the Sanskrit
Grammarians’ and ‘Some Indian Theories of Meaning’ in Hara and Wright (1996). Also Matilal
(1991).
For Gaud.apāda, there is a text and translation in Karmarkar (1953). See King (1995), for connections
with Buddhism.

investigation
1. In what senses is Advaita subversive of mainstream orthodoxy? Does it represent
the implicit rejection of Hindu dharma?
2. Is a notion of a static absolute principle compatible with our experience?
12 Viśis.t.ādvaita-Vedānta
Chapter Outline
The religious context 168
Knowing God only from Scripture 172
Opposition to Advaita 173
The individual self 175
The soul-body model 178
Further reading 183
The religious context

After about 700 A.D. we see the rise to predominance within Hinduism of
the sort of personalist theism found in the Bhagavad Gītā and the Purān.as,
accompanied by a decline in the religion in which the Vedic rituals are central.
This is not to say that ritual practices disappear. Rather, they are assimilated to
modes of practice that are more concerned with worship of a deity conceived
personally whose grace or favour is accessible to those human beings who love
to respond to him in love (bhakti). God is thought to be concerned with the
destinies of finite beings. From this perspective, the cosmos is understood as
creation for the sake of sentient beings rather than as a hierarchy of spheres of
existence (tattva), through which souls may or may not progress as a natural
process in accordance with karma, understood as an automatic mechanism.
There develops theological concept of the soul and a notion of human beings
Viśis.t.ādvaita-Vedānta 169
as essentially lovers, enjoyers and knowers of God. According to this mentality
the individual is constituted by God and the meaning and fulfilment of its life
is to be found in relationship to God.
Viśis.t.ādvaita Vedānta is the doctrinal articulation of the theistic Śrī
Vais.n.ava religious tradition that still flourishes in Tamil Nādu. It is a pluralist
ontology and epistemological direct realism about a complex universe whose
basic constituents are kinds of property-bearing enduring substances. The
term ‘Viśis.t.ādvaita’, frequently mistranslated as ‘qualified non-dualism’, is held
by the tradition to mean ‘the integral unity of complex reality’. Vedānta is the
systematic hermeneutic of the Upanis.ads, the brief summaries of the teachings
of the latter in the Brahma-Sūtras, and the Bhagavad Gītā.
The Śrī Vais.n.ava tradition developed in interaction with the enthusiastic
devotion (bhakti) towards a personally conceived deity found in the hymns
of the Tamil Ālvārs, the temple and domestic rituals and theology of the
Tantric (i.e. non-Vedic) Pāñcarātra Āgamas, and a reflective Vais.n.ava smārta
orthodoxy. The latter found expression in the philosophical theologies of
Nāthamuni (980–1050 A.D.), Yāmuna (c. 1050–1125 A.D.) and Rāmānuja
(c. 1100–1170 A.D.). The latter was a creative genius who, adopting ideas
from Yāmuna and an earlier commentator on the Brahma-sūtras called
Bodhāyana, synthesized ideas current in the tradition into a realistic and
pluralistic philosophical and theological system.
Influential Viśis.t.ādvaitins and their works include:
Nāthamuni (980–1050): Nyāyatattva (known only from quotations).

Yāmuna (1050–1125): ātmasiddhi, Ī śvarasiddhi, Sam . vitsiddhi; āgamaprāmān.yam
(on the validity of the Pañcarātra cult and its scriptures).
Rāmānuja (1100–1170): Śrı̄ Bhās.ya (commentary on the Brahma-sūtras),
Vedārthasam . graha (précis of the former); Bhagavadgı¯tā-bhās.ya.
Paraśara Bhat.t.a (1170–1240): Tattvaratnākara (quotations from Vedānta Deśika).
Vedavyāsa (Sudarśanasuri) (1120–1300): śrutaprakāśikā (commentary on śrı¯
Bhās.ya), Tātparydı¯pikā (commentary on Vedārthasam . graha).
Vedānta Deśika (Veṅkat.anātha) (1270–1350): Tattvamuktākalāpa, Sarvārthasiddhi,
Nyāyapariśuddhi, Nyāyasiddhāñjana, Tātparyacandrikā (commentary on Rāmānuja’s
Gı¯tābhās.ya), Tattvat.ı̄kā (commentary on śrı¯ Bhās.ya), Pāñcarātraraks.ā.
Śrı̄ nivāsadāsa (1600–1650): Yatı¯ndramatadı¯pikā.
Where the formation of the Śrī Vais.n.ava tradition is concerned we have to

reckon with a confluence of various trends and factors. There is the intense
ecstatic devotional religion of the Tamil Ālvār poets, who thought of all
devotees equally as servants of God. They model the soul’s relationship with
God upon that between human lovers, and sing of the agony of separation and
the bliss of reunion. The theologians, at least in their prose works in which they
were concerned to demonstrate the concordance of their beliefs and practices
with the normative religion of social and religious duty (varn.āśramadharma)
that respects distinctions of caste, tended to suppress the ecstatic emotional-
ism and incipient social inclusivism of the Ālvār tradition. Nevertheless, they
belonged to a monotheistic devotional milieu in which one is encouraged to
delight in the awareness that one exists to be a servant of the divinity Vis.n.u-
Nārāyan.a. God is a person, a being with will, agency and purposes, upon
whom one is radically dependent and in whom one may take refuge. God is a
compassionate personal being who deserves praise and love. This entirely self-
sufficient deity creates and sustains the cosmos for no purpose other than his
own delight (līlā). He is immanent both as the inner guide, the innermost
constitutive element in people, (antar-yāmin) and as present in the consecrated
temple icon. The reconciliation of mainstream orthodoxy and devotionalism
is seen in the soteriologies of Yāmuna and Rāmānuja when they say that
performance of the duties appropriate to one’s caste and stage of life informed
by understanding of the natures of the individual and highest selves combined
with ritual worship and devotion invites the grace of the supreme person
[Vedārthasam.graha 3]. In the first verse of his Summary of the Meaning of the
Gītā (q.v. van Buitenen (1953), Yāmuna says that Nārāyan.a who is the supreme
Brahman is only accessible by devotion (bhakti) accomplishable by the observ-
ance of one’s social and religious duty, knowledge and dispassion. Bhakti is not
just a matter of feeling. It accords with the belief that if God is the foundational
cause of everything, everything one does is also an action of God. This does
not mean that one’s free actions do not flow from the will. It means that it is
thanks to God that the dependent soul is an entity in the first place.
A further factor constitutive of the Śrī Vais.n.ava tradition is the non-Vedic
or Tantric Pañcarātra system of theology and ritual informing the liturgies
enacted in South Indian Śrī Vais.n.ava temples. This tradition is probably as
old as the Christian era, but it is unlikely that any of the surviving texts were
composed before c. 850 A.D.
Pañcarātra sees the world as a real creation by a personally conceived
divinity. It emphasizes divine immanence and accessibility in the temple icon.
It understands people as individual souls. In his Āgama Prāmān.yam, Yāmuna
defends against attacks from Smārtas the orthodoxy of Bhāgavata Brahmins or
Sātvatas who perform Pāñcarātra temple rituals. Some of these who belong to
the Vājasaneyaśākhā of the white Yajur Veda are of unimpeachable Brahmin
status. There are others, who claim adherence to the Ekāyanasākhā and
identify the Pāñcarātra tantras as the fifth Veda (the Vedas are held to be
infinite in extent), but Yāmuna sees their activities as on a par with those of
Smārtas. Suffice it to say that Rāmānuja accepts the epistemic authority
(prāmān.ya) of the Pāñcarātra-Bhāgavata teaching, rejecting the claim that it
is opposed to śruti in that those tantras teach that the individual self has a
beginning. He denies that the system teaches this. Rāmānuja interprets the
Paus.kara Sam.hitā as teaching that the Supreme Brahman, called Vāsudeva, out
of kindness to his devotees continuously wills to exist in modes such as the
temple-icons so as to make himself accessible to those resorting to him. The
Pāñcarātra teaching was composed by God and conforms to that of the Vedas,
understanding of which it facilitates. The system teaches the nature of Nārāyan.a
and the proper way of worshipping him.
Finally, there is a lineage of learned Vais.n.ava teachers (ācārya) specializing
in Upanis.adic exegesis and adept in sophisticated śāstric traditions, Smārta
Brahmins who take refuge in Vis.n.u as their patron deity (is.t.a-devatā).
Nāthamuni, Yāmuna, Rāmānuja and their successors belong to this tradition
of realistic and pluralistic interpretations of the Scriptures instead of the
monism found there by the mystic renouncers of the Advaita tradition.
Viśis.t.ādvaita Vedānta represents a renewal of an ancient tradition of
realistic exegesis of the Upanis.ads. Rāmānuja’s sophisticated theological
formulation of the bhakti religion in opposition to the world-renunciatory
Advaitic gnostic tradition was not new. He cites [Vedārthasam.graha 92–93] a
long list of earlier teachers belonging to the Vedāntic tradition of Upanis.adic
exegesis who taught that bhakti alone, expressed in action and involving
profound understanding, is the path to god.
But Advaitic monism flourished after the seminal works of Man.d.ana Miśra
and Śam.kara (fl.c. 700 A.D.). They hold that ordinary experience articulating
a plurality of individual conscious and non-conscious entities is a beginning-
less global misconception (anādi-avidyā) superimposed upon an inactive and
undifferentiated Brahman characterized as non-intentional pure consciousness.
On this view, the liberation of the soul from rebirth is simply the cessation of
ignorance about the true nature of reality. It is the intuitive realization that
one’s true identity (ātman) is non-intentional awareness (jñapti-mātra) and
that one is not an individual agent subject to ritual duties and transmigration.
This outlook is obviously at odds with the Bhedābheda tradition of Upanis.adic
exegesis, which sees the real cosmos as an emanation (parin.āma) of the
Absolute and as the real self-differentiation of the Supreme Soul – the substra-
tive cause of all existents. The basic Viśis.t.ādvaitin doctrine that the actual, real
world of conscious and unconscious entities is an organic complex that is both
essentially dependent upon God, and intrinsically distinct from him belongs
to this realistic tradition of thought.
Knowing God only from Scripture

Like all Vedāntins, Rāmānuja assumes that the truth about God can only be
known from the timeless Vedic scriptures, primarily the Upanis.ads, the know-
ledge-portion (jñāna-kān.d.a) of the Vedas. Although only scriptural language,
and not perception and inferences based upon it, can reveal the truth about
the transcendent, Rāmānuja does say [Śrī Bhās.ya1.1.2] that we have natural
knowledge of God as that which possesses unsurpassable greatness, since this
is the meaning of the verbal root br.h from which the term ‘Brahman’ is derived.
Unsurpassable greatness includes powers such as omniscience and omnipo-
tence that are properties of that which is both the material and efficient causes
of the cosmos. Further content may be added to this concept by texts such as
Taittirīya Upanis.ad 2.1.1. ‘The Brahman is reality, consciousness, infinite’,
which he interprets as implying three distinct properties belonging to the
divinity, (in contradistinction to the Advaitins who think that the predicates
serve merely to differentiate the Brahman from all else).
Only the Scriptures, and not human reason, can reveal anything about the
nature and existence of God. Rāmānuja [Śrī Bhās.ya 1.1.3] uses philosophical
argument to show that argument cannot prove the existence of God. He
adduces a number of considerations against the standard inferences for the
existence of God, all of which rely on the general principle (vyāpti) that
products require an intelligent maker with the appropriate capabilities. For
example, although we can infer a producer from human artefacts, we have no
knowledge of the ultimate origination of natural features such as mountains
and oceans and their existence supplies no reason to suppose that they have
one all-powerful and all-knowing maker. The philosophical inferences cite
as examples the production by potters of objects such as pots that are unequi-
vocally single whole entities. But can we really treat the cosmos as a whole as a
single great big object? The complex world consists of many different types of
effects and as such cannot license the inference of a single maker. Assuming
that the universe can be treated as a single product, was it made at one time
or at different times? We have neither observational nor purely conceptual
grounds for supposing that it was made at one time. But if it is made at
different times, it is also possible that it has more than one maker. In our
experience agents, however capable, are finite and embodied. All our everyday
valid inferences about who produced what concern finite and embodied
makers. The Brahman is different in kind from everything else. Thus we
cannot infer that it is an agent in the sense in which we understand agency.
Opposition to Advaita
The Advaitic concepts of the Absolute as impersonal, static, consciousness
and of the non-individual soul (identical with that Absolute principle) as
utterly transcendental and detached from personal individuality are the
fruits of the mystic renouncer’s contemplative experience in whose light the
everyday world appears as less than fully real. But these visionary insights
are problematic when it comes to explaining the genesis of the finite universe
and its relation to the unconditioned reality that has nothing in common
with the world. The developed Advaita tradition attributes the plural universe
and our experience thereof to the operation of a positive force (bhāva-rūpa)
called Avidyā (as the substrative cause of the cosmos, it is obviously different
from everyday notions of ignorance and misconception) which projects
diversity and conceals pure being. The undifferentiated pure conscious real-
ity falsely appears as the plural world when it is obscured by avidyā. Avidyā
explains why we unenlightened beings mistakenly but inevitably think in
terms of different individual entities. Causative avidyā, and its product, the
cosmos, are indeterminable or inexpressible (anirvacanīya) as either real or
unreal. The Viśis.t.ādvaita tradition rejects the latter claim as incoherent: if
something is not real, it is unreal. If it is not unreal, it is real. If it is neither
real nor unreal, it is both real and unreal. If avidyā is different from the
Brahman, monism is compromised. If it is the same as the Brahman then it
exists either absolutely or never. They insist that ignorance cannot be a
subsistent entity with causal efficacy. It is just the absence of knowledge.
Moreover, the putative causal ignorance must have a substrate. It cannot
be the Brahman, which is pure knowledge. Nor can it be the individual self
which, according to Advaita, is itself the product of ignorance. Also, since the
scriptures, in common with all the pramān.as, belong to the sphere of avidyā,
their capacity to reveal the truth is undermined.
The Advaitic proposition that reality is unitary and undifferentiated quickly

attracted the objection that there is no means of establishing its truth (pramān.a).
There is already an instance of difference if the operation of the means
of knowledge presupposes a duality of act and object. Rāmānuja argues that
we cannot make sense of the notion that there is any sort of reality, ultimate
or otherwise, that is undifferentiated and non-intentional consciousness
(nirviśes.a-cinmātra). He is an epistemological direct realist, holding that all
cognitions are intrinsically valid (svatah.-prāmān.yam) just in virtue of their
occurrence. He holds that intrinsically reflexive consciousness is always
about something. In addition, apprehension is always of the real (sat-khyāti):
all perceptual cognition, even when misleading (bhrama), is in accordance
with what is the case (yathārtha) independently of our thinking. There are
no intrinsically false cognitions. Ultra-realism involves minimizing or explain-
ing away what are ordinarily viewed as perceptual errors and hallucinations.
There is an ancient theory that all material things are compounded out of the
same elements. When mother of pearl is mistaken for silver, we are actually
detecting traces of silver in the mother of pearl. Cognitions are intrinsically
formless (nirākāra) and assume the forms of their objects. The lack of
subjective contribution eliminates perceptual distortion. It is the extra-mental
environment, consisting of stable objects that endure through space and time,
that is responsible for mental variety. Truth is correspondence, understood
in a strong sense as structural isomorphism, between knowing and the
known. The subject-object structure of cognition is held to be self-evident and
encoded in normal language. There simply are no cognitionslacking an agent
and external object.
Implicit in Rāmānuja’s critique of the Advaitic idea that authentic reality is
featureless and non-intentional consciousness is an appeal to a principle
upheld by the Indian realist traditions: whatever is, is knowable and nameable.
All the means of knowledge refer to entities having properties. The notion of
an object without properties is unintelligible because thought and language
are possible only with respect to entities identifiable by their specific charac-
teristics (viśes.a). The language of the Scriptures, like all language, is composite
and relational, and its complexity mirrors that of its objects. This is a crucial
point. Scripture is our only means of knowing about the transcendent. If
its language is complex and language is isomorphic with what is expresses,
complexity must obtain at the ultimate level.
He continues with the argument that sensory perception (pratyaks.a), whether
non-conceptual (nirvikalpa) or conceptual (savikalpa) cannot establish the
existence of a non-differentiated reality. We have seen that thinkers of
the Nyāya-Vaiśes.ika and Mīmām.saka traditions tried to distinguish the two
types of perception in various ways. The problem is difficult: to what extent
can non-conceptual perception lack specificity while still referring to
something? Is it possible to apprehend a ‘bare particular’ devoid of specific
and generic features? Following Man.d.ana Miśra, Advaitin thinkers appeal to
the example of non-conceptual perception in arguing that there can be pure
uninterpreted experience that is the same as ‘pure being’ or featureless reality.
The Viśis.t.ādvaitins reject the mainstream view upheld by Naiyāyikas and
Mīmām.sakas that nirvikalpaka perception grasps a bare reality (vastu-mātra)
without reference to features such as names, quality, substance and generic
property. Rāmānuja says that the content of savikalpa-pratyaks.a (conceptual
perception) is complex since it explicitly refers to that which is qualified
by several categorial features (padārtha: i.e. jāti, gun.a, dravya and kriyā). But
the object of non-conceptual perception (nirvikalpa) is also complex since
such a prior sensational state is a condition for the comparison of already
experienced differentiated entities at the subsequent conceptual stage. Non-
classificatory perception is the apprehension of an entity as lacking some
differentia, but not of every differentiating feature since apprehension of the
latter kind is never encountered and is in any case impossible. Every cognition
arises in virtue of some differentia and is specifically verbalized in the form,
‘this is such and such a kind of thing’. The difference lies in the fact that in
non-conceptual perception a complex entity, analysable in terms of the
categories of substance, specific and generic properties, is cognized, but
what is missing is knowledge of the recurrence (anuvr.tti) of those features
in other entities of the same kind. In nirvikalpa-pratyaks.a we cognize an indi-
vidual and its concrete generic structure (jāti, construed as ākr.ti or sam.sthāna)
as distinct. But since the structure has been seen in but one individual we cannot
generalize and form a concept. The key point is that since non-classificatory
perception is complex and since there is a structural isomorphism between
knowing and the known, it cannot yield knowledge of an undifferentiated
reality. Finally, inference (anumāna) relies upon perceptual data and is thus
incapable of revealing featureless reality.
The individual self

The Advaitins think that the soul or one’s true identity (ātman) is inactive,
impersonal consciousness without any content because it is identical with
the Brahman. Everyday self-experience is a mirage, concealing the uncondi-

tioned being. But devotional theism requires that the self is an individual
thinking, acting and feeling entity, capable of responding to God. Advaitins
and Viśis.t.ādvaitins attach quite different meanings to the term ‘ātman’.
Viśis.t.ādvaita maintains that we really are individual selves, enduring
substantial entities that are really distinct from the bodies that they can
objectify, having experiences while not being reducible to streams of
experiences. Each soul has its own ineffable identity that is known only
to itself. Every subject of experience enjoys a sense of ‘mineness’ that is not
the same as the individuality of the physically, psychologically and socially
constituted person. It is this feeling that individuates the souls, whatever their
circumstances. The soul is a permanent principle of identity that underpins
the synthetic unity of experience in the present and through time.
Viśis.t.ādvaitins think that agency, which requires some form of embodiment,
is always an accidental and not an essential feature whether the soul is bound
or released. Nevertheless, embodied agency is a reality, not a misconception.
The Viśis.t.ādvaitin philosophers propound defences of commonsense realism
about the self as a persisting centre of reflexive awareness that is in contact
with mind-independent realities. Embodiment is crucial in that it enables
agency and sorts of experience that exhaust the accumulated karma of souls in
bondage to rebirth. The tradition does not understand karma as an automatic
mechanism. Rather it is an expression of the will of God. It is a perfectly just
mechanism, rewarding and punishing souls for their actions. It is karma that
contracts the soul’s intrinsically unlimited consciousness when we misidentify
ourselves with our embodied status that is a product of karma.
The Viśis.t.ādvaitins agree with the Advaitins that consciousness is unique in
that it does not require anything else to establish its existence (svatah. siddha).
But they differ from them in denying that the true self may be understood
simply as consciousness. Were that the case, selves could not be individuals.
Rather, reflexive and intentional consciousness is an essential property of
individual selves that are its agent and substrate. It is the true self that is
incorrigibly revealed as the ‘I’ in every conscious state. Our everyday sense of
selfhood is not a mistake concealing another authentic ‘inner self ’. Thoughts
are properties that require an enduring thinking subject. That consideration is
used by the Naiyāyikas as an inference for the existence of a persisting subject.
But Rāmānuja and his tradition think that such an inference is superfluous
because but we are immediately aware of ourselves as individual persisting
identities whose essential property is consciousness. Everyone always knows
who they are. The conscious subject is always a specific mental presence. The
self is not known in the same ways as objects and states of affairs in the world
are known because it is phenomenally given in direct intuition.
The Self is the agent of conscious acts that illuminate objects. Conscious
acts are both intentional and reflexive. The tradition defines consciousness in
terms of two factors: its intentionality – its being directed towards objects
other than itself – and its reflexivity or ‘self-awareness’. They hold that all
conscious states are intentional: they are acts on the part of a subject directed
towards some object or state of affairs. But Rāmānuja resists the externalist
view that experience is necessarily and not just causally dependent upon inter-
actions with an objective environment by insisting that all conscious acts and
states are always self-illuminating or intrinsically reflexive (svayam. prakāśa).
That is to say, when a subject cognizes something, simultaneously and in virtue
of the same act, he is aware of himself as cognizing that reality. Even in inten-
tional cognitive acts that are, as it were, absorbed in the object by being fully
attentive to it there is also an element of subjective awareness.
We have seen that the Viśis.t.ādvaitins rule out the possibility of non-
intentional blank awareness (nirviśes.a-cinmātra). Awareness is always com-
plex and always about something. Moreover, consciousness is in a state
of flux. Were it identical with the self, it would be impossible to recognize
something seen today as the same thing seen the previous day. The self is
not a bundle of fleeting experiences. It is the persisting subject that has the
experiences – a principle of continuity with a witness’ perspective upon the
states that it co-ordinates. The normal self keeps track of itself through time.
It is the agent of mental acts and its permanence as such and the momentary
nature of those acts are both directly perceived. Distinctively Viśis.t.ādvaitin is
the idea that the self both has the form of consciousness (cid-rūpa) and has
consciousness as its quality. They say that consciousness is both substance
(dravya) and an attribute (gun.a) and its nature is to render entities susceptible
of thought speech and action. As the essential property (svarūpa-nirūpan.a-
dharma) of the soul, it can be considered as substance, but as discrete mental
acts possessed by the self, it exists as an attribute.
The Advaitic tradition says that one’s everyday feeling of personal identity
(aham.kāra), the sense of oneself as an individual agent subject to religious
and social duties and confronting a world of objects, is a mask concealing
the identity of consciousness with the impersonal, inactive Absolute beyond
differentiation. The illusion occurs when the light of consciousness is con-
fused with the activity of the material intellect (buddhi). As we saw above,
Rāmānuja’s tradition denies that our everyday feeling of continuous personal

identity is an illusion or a case of mistaken identity. It is integral to theistic
bhakti that the self that understands itself as a servant and lover of God should
be the authentic self. The ‘I’ that thinks, intends and acts is the real self. The
pronoun ‘I’ ultimately stands for the inner self that is itself ensouled by God,
its inner guide and sustainer. As Nāthamuni puts it in the Nyāya-tattva, ‘If “I”
did not refer to the true self, there would be no interiority belonging to the
soul. The interior is distinguished from the exterior by the concept “I”. The
aspiration, “May I, having abandoned all suffering, participate freely in infinite
bliss”, actuates a person whose goal is liberation to study scripture etc. Were it
thought that liberation involved the destruction of the individual, he would
run away as soon as the subject of liberation was suggested . . . The “I”, the
knowing subject, is the inner self.’
The soul-body model

Rāmānuja’s basic ontology is a threefold hierarchy of three really distinct
categories: the personal God Nārāyan.a who is the first cause and sovereign
of the cosmos, also immanent in the individual souls; individual souls that
are the subjects of experience; physical bodies occupying the material environ-
ment that are the objects of experience. Rāmānuja’s basic conception, and most
significant contribution to the tradition, derived from śruti passages such as
Br.hadāran.yaka Upanis.ad 3.7.3–23 (e.g. 14: ‘This soul of yours who is present
within but is different from all beings, whom all beings do not know, whose
body is all beings, and who controls all beings from within – he is the Inner
Controller, the immortal one’) is that the cosmos of souls and matter is the
body of God (śarīra-śarīrī-bhāva). This is not intended as merely one possible
interpretation among others, one provisional model, a useful way of thinking
about God and the world. It is thoroughly Vedically based. It is what the
Upanis.ads, the sole source of Vedāntic theology, teach. While the development
of this idea is due to Rāmānuja, it was already present in his tradition. For
example, the final verses of Yāmuna’s Īśvarasiddhi say, ‘The universe obeys
the will of one person because it is insentient, like a body. All conscious
entities function under the governance of one, like touch, because functioning
depends upon connection with a body. The universe has one person as its
source because he animates conscious and non-conscious entities, like a
country with one king’.
The relationship between God and the world is parallel to that between an
individual self and its body. The key features are essential dependence and
difference. He defines a body as any substance (dravya) that a conscious entity
can control and support entirely for its own purposes and whose raison d’être
is to be an ancillary (śes.a) to that entity. All conscious (cit) and non-conscious
(acit) entities are the body of the Supreme Person since they are controlled and
supported by him exclusively for his own purposes and their raison d’être is to
be his ancillaries. The relation between soul and body is that between ontic
ground (ādhāra) and dependent entity (ādheya) that cannot exist separately,
between controller and thing controlled, between master and servant. The
‘body’ term is an essentially dependent mode (prakāra) incapable of existing
separately (apr.thak-siddhi). Souls and matter may exist potentially (the
Brahman’s causal state) or actually (the Brahman’s effected state). In either case
they are ensouled by God and cannot exist independently of him because
they are internally controlled by God and constitute his body. By being present
in the individual selves, God is present in matter. It is only because souls and
material objects are ensouled by God that they are entities. Just as a parcel of
matter only becomes a body when animated by a soul, so the souls are entities
only because immanent divinity sustains them.
Rāmānuja accepts the satkāryavāda theory of causation according to which
effects exist potentially in their causal substrata prior to their actualization as
entities with determinate name and form. An effect is a different state of the
substrative cause (upādāna-kāran.a) of which it is a transformation (parin.āma).
It is not a completely different substance. The Brahman is the material and effi-
cient cause of the cosmos, which is a single, organic and intelligible process.
There is an analogia entis, a continuity of being, between God and the world.
Rāmānuja interprets Upanis.adic principle that ‘by knowledge of the one, there
is knowledge of the whole’ as meaning that by understanding the nature of a
substrative cause one understands the nature of its effects. Understanding the
scriptural statements about the Brahman is the key to understanding the world.
It is God, ‘one without a second’ who is the substrative cause (upādāna-
kāran.a) of the cosmos and not an independent principle such as the pradhāna
of the Sām.khyas or the real māyā of the Śaiva Siddhāntins. God may be con-
sidered as both cause and effect: cause when cit and acit entities are in their
latent condition, and effect when they have evolved and acquired names and
forms. Cause and effect are ontologically continuous. The manifest cosmos is
a real transformation (parin.āma) in which the real three categories of God
souls and matter maintain their distinct identities.
‘Since everything forms the body of the Supreme Person, he is directly signi-
fied by every word.’ Rāmānuja has a semantic principle that the signification of
essentially dependent modal entities (e.g. bodies) extends to the mode possessor.
Whenever something is essentially in an attributive relation to something else
(this includes the relations between qualities, generic properties (sāmānya,
ākr.ti) and the individual substances (dravya) to which they belong as well as
that between souls and bodies), the terms for the attributes may also signify
the possessor. God has created the expressive power (vācaka-śakti) of words
together with the entities which they signify. Any self is a mode of the Brahman
since it is included in the Brahman’s body. Human bodies are modes of their
souls. Words for bodies signify in their primary senses both the conscious
entities ensouling them and God qua the inner controller and guide (antar-
yāmin) of the self. Thus God’s immanence as the soul of each embodied soul is
the basis of the literal reference of scriptural language to Him.
As we saw above (Chapter 10) thinkers of the Bhedābhedavāda tradition
such as Bhāskara and Yādava Prakāśa (1050–1125: originally an Advaitin, then
Rāmānuja’s teacher) formulated versions of pantheism according to which the
cosmos of souls and matter emanates from God. This was repugnant to
Rāmānuja’s tradition. According to the Bhedābhedavāda, the Brahman is the
all-encompassing category of being of which all entities are instances – the
emanations are actually instantiations of God. Their Absolute is originally
undifferentiated being, void of qualities, actions, kinds and individualities, but
becomes threefold as subjects of experience, objects and the controller. The
cosmos is its substrative cause in conditioned form. Effects are not really
different from their material causes and the world is non-different from the
Brahman. The individual self is but the Brahman affected by ignorance
(avidyā), karma and desires (kāma). Rāmānuja’s objection is that this view
converts the Unconditioned into finite reality, subject to transmigration,
imperfection and suffering. One reason for developing the ‘three-level ontol-
ogy’ in which the Brahman, souls and matter are essentially distinct is to avoid
the undesirable consequence that the Absolute is implicated in the vicissitudes
of finite existence. Thus he replaced the Brahma-parin.āma theory with the
idea that real transformation occurs only in the sphere of the entities that
constitute Brahman’s body (brahma-śarīra-parin.āma). He goes beyond an
emanative model of cosmic production by distinguishing between divine pri-
mary causality in constituting the cosmos of souls and matter, and the operation
of secondary causes in the created realms. The Brahman is essentially distinct
from its dependent modes: its essential being or proper form abides intact.
It is in the field of scriptural exegesis that the soul-body model comes into
its own. The interpretation of co-referential (samānādhikaran.a) statements
such as ‘Tat tvam asi’ (‘That thou art’ expressing a relation between the self and
the Brahman) and ‘Satyam. jñānam anantam. Brahma’ (‘The Brahman is real-
ity, consciousness, infinite’: Taittirīya Upanis.ad 2.1.1) is central to Vedāntic
theology. Samānādhikaran.a means the co-occurrence of two or more items,
for example an individual substance and its property, in the same locus or
substrate. In grammatical usage, it is ‘the reference to a single object by several
terms having different grounds for their application’. Such constructions
appear in scriptures expressing the relationship between God and the world,
God and the self or, in the case of ‘Satyam. jñānam anantam. Brahma’ as saying
something about the divine nature.
Vedāntins believe that the language of revealed śruti is our only means of
knowledge (pramān.a) about transcendent reality. We have no cognitive access
to God independently of the infallible and authorless scriptural authorities.
Rāmānuja is a realist holding that there is an isomorphism between knowledge
and the known. There is also a structural isomorphism between scriptural
statements and the reality of which they speak. It is not just the meanings of
words that are informative. Grammatical constructions reflect the nature of
reality. There is a sense in which a thinker’s theory of meaning determines
their metaphysics. Advaitins emphasize the singularity of reference and
construe co-referential constructions as identity-statements conveying an
impartite essence (akhan.d.ārtha). This usually involves attributing non-literal
senses (laks.an.ā-artha) to the co-ordinated words, and this is recognized as
an exegetical weakness. Rāmānuja’s tradition maintains that the grounds for
the application are differences belonging to what is signified. Co-referentiality
is thus the reference to a complex reality by words expressing its different
features. Rāmānuja says that it expresses a single entity qualified by its essen-
tially dependent modes. He interprets co-referential statements about God
and features of creation as expressive of the soul-body relation. In the case of
‘Tat tvam asi’ (‘That thou art’) the Advaitins attribute an extraordinary sense
to each term: ‘that’ stands for the impersonal Absolute, and ‘thou’, has to be
purged of its everyday connotations of individual personality so that it may
signify the Inner Self that is identical with the Brahman. The statement
expresses the identity of the two. But according to the Viśis.t.ādvaita exegesis,
the ‘that’ stands for the creator God, the inner guide of the soul, of whom all
entities are modes since they form his body. ‘Thou’ stands for an individual
self, an essentially dependent mode of God. ‘Tat’ denotes the Highest Self,
which is the cause of the universe, whose purposes are ever-actualized

(satya-sam.kalpa), who possesses every exalted quality and who is devoid of
every trace of imperfection. ‘Tvam’ denotes the same Brahman embodied
by the individual self, along with the body of the latter. The grammatical
co-ordination conveys the unity (not identity) of the two. The co-referential
terms apply in their primary senses.
The soul-body model enables Rāmānuja to interpret many scriptural
statements that had been treated as favourable to Advaitic monism in a
theistic way.
Here are some relevant passages from Rāmānuja’s works:
Because the body is a mode of the embodied entity and because the meanings of
words for modes extends to the mode-possessor, there is the principle that the
meanings of words signifying bodies extend to what is embodied. Whenever we
think, ‘This is such and such’, the mode is understood as the aspect that is ‘such
and such’. It is logical that the mode culminates in the mode-possessor since it
depends upon that entity and its being apprehended depends upon that entity.
Thus a word signifying a mode extends to the possessor. (śrı¯ Bhās.ya 1.1.13)
Because all conscious and non-conscious entities are modes of the Supreme Soul
in that they are his body, all words for such beings apply in their primary senses to
the Supreme Soul. (Vedārthasam . graha 75)
People unlearned in Vedānta do not see that all objects and all individual souls
participate in the nature of the Brahman. They think that the referential scope of
all words is restricted to the entities that they usually signify that are in fact only
part of their meanings. Once they have learned about the Vedānta passages, they
understand that all words signify the Brahman who constitutes his various modes
since everything participates in the Brahman in so far as he is the Inner Controller
and that everything is created by him. (ibid. 21)
The body may be thought of in co-ordination with the Self in that it is the essential
nature of bodies to exist in an attributive relation with selves since they would not
exist independently of them. This is comparable to the relation between generic
properties such as cowness and their individual instances. (Gı¯tā Bhās.ya 13.1)
We have interpreted the manifold scriptural statements consistently with each

other and without sacrificing their literal senses. The Scriptures saying that the
Brahman does not change retain their primary senses by rejecting transformations
of its essential nature. The Scripture saying that it is without qualities are estab-
lished as denying undesirable ones. Denials of diversity are preserved because
they assert that the Brahman has everything as its modes in that he is the soul of
all because all conscious and non-conscious things are entities only because
they are embodied modes of the one Brahman. Pass ages saying that the Lord is
different from his modes and has all perfections may be taken at their face value.
Statements that he is pure consciousness and bliss are taken literally because they
convey a self-luminous reality having blissful awareness as an essential property.
Statements of unity are well established because the co-referential constructions
in their primary senses express a soul-body relation.
Is the overall purport of the Upanis.ads difference or non-difference or difference-

cum-non-difference? It is all three. Non-difference is established because the one
Brahman has everything as its modes because he is embodied by everything.
Difference-cum-non-difference (bhedābheda) is established because the one
Brahman exists as multiplicity having conscious and non-conscious entities as
his modes. Difference is established because non-conscious entities, conscious
entities and the Lord differ in respect of their essential natures and activities and
are not intermingled. (Vedārthasam . graha 84–85)
Further reading
The Śrī Bhās.ya is translated in Thibaut (1904), the Vedārthasam.graha in van Buitenen (1956) and the
commentary on the Gītā has an English précis in van Buitenen (1953).
Svāmī Ādidevānanda’s edition and translation of the Yatīndramatadīpikā is useful primary source.
There is a classic monograph by Carman (1974), which can be supplemented by Lipner (1986). Bartley
(2002) sees the soul-body model as an exegetical device and dwells on controversies with Advaitin
interpretations.
For God as the ‘inner-controller’ see Oberhammer (1998).
For the devotional religious context see Hardy (1983).
For Pañcarātra, Schrader (1973) is still the standard work. There is interesting material in Sanderson
(2001). See also Sanjukta Gupta (2000), Laks.mī Tantra.

investigation
1. Does Rāmānuja have a pantheistic vision of the world as the body of God, or is the
notion better understood as an exegetical device?
2. Why is he so opposed to Advaita?
3. Does he have a basically theological concept of the self?
13 Dvaita-Vedānta and Madhva
Chapter Outline
Difference 188
Direct epistemological realism 191
The trouble with Avidyā 191
Further reading 192
Dvaita is a realist form of metaphysical pluralism. It is the philosophical

articulation of radical Vais.n.ava monotheism. Madhva and his followers
insist that the Scriptures are first and foremost revelations from and about
the one god called Vis.n.u. Madhva’s thinking is theocentric, rather than
anthropocentric or cosmocentric, to a remarkable degree. Anything that
we may know about God, we learn from the Scriptures. There can be no
natural knowledge of God. The tradition extends the canon of Scripture
beyond the revealed śrutis and traditional smr.tis that are accepted by all
Vedāntins to include many specifically Vais.n.ava sectarian compositions, many
of which appear to be very late compositions.
Madhva lived in Udīpi in Karnataka from 1238 to 1317 A.D. and wrote
commentaries on several Upanis.ads, the Bhagavad Gītā, the Brahma-Sūtras
and the Bhāgavata Purān.a, in addition to many compositions of his own
such as the Vis.n.utattvavinirn.aya. Other exponents of Dvaita-Vedānta are
the acute logicians Jayatīrtha (1365–88) and Vyāsatīrtha (1460–1539), both
of whom commented on Madhva’s works.
Basic to their ontology is the distinction between the only independent and
self-determining (svatantra) reality that is the unsurpassably great godhead,
Dvaita-Vedānta and Madhva 185
and the dependent sentient and insentient existences whose very being is sus-
tained by God. The universe is real creation but it has no intrinsic tendency to
continue in existence. The world only continues in existence beginninglessly
and perpetually because it is always known and sustained by God. Vis.n.u is
omnipotent, omniscient and blissful. Indeed he is possessed of the host of glo-
rious characteristics elaborated in the Pāñcarātra scriptures. All his attributes
are eternal and in no wise different from him. He is different from all other
beings, but this difference does not affect him. There is no plurality in the
divine nature parallel to that of the world. There are no differences intrinsic to
the divine nature, which is simple. God’s uniqueness consists in there being no
difference between his essence and his existence. In other words, it is his
essence to exist, whereas the existences of finite beings are dependent.
The cosmos is structured by five types of real difference: the difference
between God and the individual soul, that between material objects and God,
differences between individual souls, difference between individual souls and
material objects and differences between material objects. Souls and matter
are eternal realities. So the notion of divine creation should be understood as
an articulation of a relation of absolute dependence rather than as a process of
emanation from the divine substance.
The cosmos is real and has no beginning. If it has a beginning it would
perish. But it does not perish, and it is not constructed by mistaken thinking
(na bhrānti-kalpita). Were it thus constructed, it would cease. But is does not
cease. Only the mistaken could say that duality does not exist! The wise know
that the plural world is real because it is known and protected by Vis.n.u. There
are many individual selves and the complex physical world exists independ-
ently of consciousness. An attitude of commonsense direct realism pervades
his work. The basic principle is that in direct acquaintance (anubhava) with
environment that we inhabit, our minds are not creative and constructive. We
are confronted by a world outside us, before our minds get to work. He calls
this primary, pre-reflective and pre-discursive experience ‘pure knowing’
(kevala-pramān.a). Truth is correspondence between a judgment and an objec-
tive state of affairs. Correspondence here should be taken in the strong sense
of congruence or conformity, rather than just the sort correlation that we find
between a map and a territory. He says that knowing (pramān.a) is ascertainment
(niścaya) that conforms to its object (yathārthyam). The word ‘pramān.a’ can
be analysed as having two senses: a state or piece of knowledge (pramiti or
pramā); and an instrument of knowing – an intentional act by which the real
nature of an entity is ascertained or measured. Quite consistently with his
direct realist (anti-representationalist) epistemological stance Madhva departs

from the mainstream tradition in acknowledging that memory is a genuine
epistemic authority. He says that memory is produced by internal perception
(mānasa-pratyaks.a). It is pramān.a because it reports a stored informational
state (anubhava) that conforms to reality (yathārtha). Memory is pramān.a
when it is correct cognition that is faithful to an object or state of affairs that
has already been known and that has not been annulled by a correcting cogni-
tion – as in the case of veridical perception.
Madhva’s Vedanta is not only opposed to non-dualism, which denies the
reality of difference and maintains that the Absolute and the soul are identical,
but also to other theistic forms of Vedānta such as Viśis.t.ādvaita. The latter
posits links between God and the world of conscious and non-conscious entit-
ies. He thinks that his compromises God’s transcendent perfection. He disa-
grees with the Viśis.t.ādvaitin view that there is a continuity of being (analogia
entis) between God and finite existences. Conceiving of the relation between
God and the world in terms of a model such as that between soul and body is
a form of anthropomorphism. It posits too close a parallelism between the
divine and human orders of being.
Each entity is a unique individual (viśes.a). This individuality is directly and
non-discursively intuited (anubhavād jñāyate). The unique individuality of an
entity has to be known before we can apply words and concepts to it. This is
crucial: on this account entities are not constituted, constructed or individuated
by our thoughts. We immediately know entities, and ourselves, in their unique-
ness without having to compare them with anything else. The notion of viśes.a
eliminates the problem of explaining how released souls lacking personal indi-
viduality are distinguished from one another. There are three categories of
souls: those who are liberated; possible candidates for liberation; those who are
beyond salvation (those who explicitly reject the sovereignty of Vis.n.u).
Liberation, in which the soul retains its essential individuality, consists
in the realization of a state of innate consciousness and bliss focused on the
divinity and is unattainable without a combination of intense devotion (bhakti)
and divine favour (prasāda). Only Vis.n.u saves: liberation is not a human
achievement. Love follows acceptance of the unique sovereignty of the Creator
and Sustainer, and the understanding of the fivefold difference that character-
izes the cosmos in contrast to the absence of internal differentiation in God.
Madhva departed from mainstream Vedānta in denying that God is
the material or substrative cause of the world. God is only the efficient cause
or creator in that he actuates an independent material principle that is subject
to his governance. He creates by organizing an independent material principle
that is subject to his governance. The gulf between the absolute and the finite
is spanned by the divine will that sustains and supports finite beings.
Madhva’s doctrines are attractively and succinctly presented in a work
called the Vis.n.u Tattva Vinirn.aya, extracts from which we shall now look at:
If we say that there is no text without an author (apaurus.eya) the notions of right
(dharma) and wrong and the other matters of which the scriptures speak that are
accepted by all philosophical traditions lack any foundation. Someone who denies
right and wrong does not help the world and only encourages violence. There is
no point in his trying to serve mankind since he admits no supernatural reality.
Right and wrong cannot be established by human opinions because people are
liable to ignorance and dishonesty.
That the Vedas are not human compositions is self-evident because there is no
tradition testifying to their authorship. Positing an author when one is not known
is an uneconomical hypothesis.
The epistemic authority (prāmān.yam) of the Veda is intrinsic. Otherwise there is

the problem of an infinite regress.
Modes of knowing (pramān.a) like perception and inference have no epistemic

authority independently of scripture in matters such as what is right and wrong
because the latter are out of their range.
The Vedas are eternal and subsist in the mind of Vis.n.u. They are manifested, but
not originated, when uttered by God.
The Vedas are fact-asserting. They speak of already existing things as well as things
to be done. In ordinary language, meanings are primarily grasped in respect of
things that already exist. Language is primarily informative and descriptive. It is only
once one has understood that something is a means to an end that one acts
accordingly.
The overall purport of scripture is not the identity of the individual soul and God.
Passages stating that they are different are not uninformative repetitions of what
is already known by some other means because the existence of God is not estab-
lished without scripture.
God’s existence cannot be proved by inference because inference can also prove
the opposite.
The argument, ‘The world must have a creator because it is an effect, like a jar’ is
countered by ‘The world has no creator because it is not a single whole object’.
If the reality of difference is established by perception and inference within whose

province it falls, scriptures asserting non-difference must be false because they are
contradicted by those means of knowing.
Even if scripture is stronger than perception in some contexts, it cannot be valid

when it conflicts with it.
Self-evident experience establishes the difference between the individual soul and
God. Everyone knows that they cannot do everything. Scripture is not an authority
if it contradicts this sort of self-evidence.
If difference is established by a means of knowing other than scripture, it is

self-evident that it cannot be denied.
If it is not so established, scriptures teaching difference will still be authorities

because they inform us about something that would otherwise be unknown.
As non-difference is contradicted by all the means of knowing it is not the purport of

scripture. Rather, the purport of scripture is the unsurpassable greatness of Vis.n.u.
The goals of human life such as dharma have results that are ultimately transitory
and mixed with unhappiness (since one knows that they won’t last for ever). Only
freedom from rebirth (moks.a) is the supreme felicity to be sought by those
wandering in sam. sāra.
Freedom is not attained without the grace of God. God feels affection for those
who recognise his superior virtues but not for those who insist on their identity
with him.
Difference
The logical arguments against the reality of difference advanced by Advaitin think-
ers are unsound because difference is the proper form or essential nature (svarūpa)
of an entity.
Examples of such arguments are:
There is proof of difference by the existence of the relation between attribute and
substrate or quality and qualified. But relation between attribute and substrate
depends upon difference.
Knowledge of difference depends upon cognition of the counterparts to the

subject. But the cognition of those counterparts depends upon cognition of the
subject from which they differ.
These arguments are circular and lead us to conclude that we since cannot
properly formulate an understanding of difference it does not exist.
Madhva replies: But just because difference implies that there are counterparts to
the subject, it does not follow that it is not the proper form of the subject. That
number one is not the number two does not compromise its identity.
Difference is established whenever a proper form of an entity is identified. We see

the proper form of an entity as unique and other than everything else. The expres-
sion, ‘the difference of this’ is parallel to the expression, ‘the proper form of the
entity’.
If the proper form is not also the difference, the difference of the entity from all
others would not be known when the entity is seen. If such individuality is not
already known, it would be possible to doubt whether one’s self is not a jar.
Having cognised individuality and the general difference of the entity from
everything else, one might sometimes question whether the entity is in fact the
same as something else that it resembles. But one never doubts whether one is
the same person.
Cognition of a universal like potness occurs only once the identity of an object
has been cognised. Entities must be unique to begin with. They are not made
individual by their properties or modes because they must be distinct identities in
the first place if they are to acquire different features.
Individual identity is directly intuited.
When we say that something is different from a pot and that something is dif-
ferent from a cloth, the ‘being different’ is not the same in both cases.
The hypothesis that individual entities cognised by the means of knowing are
unreal (mithyā) is an obfuscation since it contradicts the means of knowing.
Reasoning (tarka) on its own cannot refute what has been established by the
means of knowing. What is directly perceived cannot be dismissed as error just by
reasoning.
We move on to the rejection of the Advaitin theory that just as illusions and
mistakes cannot be categorised as real (because subsequently corrected) or
unreal (because they have real effects) neither can the plural cosmos be real or
unreal because it is a product of avidyā.
[143] There is no means of knowing something that is neither real nor unreal.
When someone says, ‘We cannot be aware of what does not exist’ is he thinking
about non-existence or not? If he is not thinking about non-existence, he cannot
deny that non-existence was a real content of thought. If he is thinking about
non-existence, the same applies.
Without a concept of non-existence, the difference of something from the non-

existent cannot be known. When we mistake a piece of shell for a piece of silver,
it is not the case that the silver is neither existent nor non-existent because the
corrective experience is, ‘Non-existent silver appeared there’. We cannot say that
it was objectively real just because it was experienced. Illusion means thinking that
something unreal is real or thinking that something real is unreal.
In illusions there is a thought of something real that was not present in a certain
set of circumstances.
We do not need to claim that the content of illusions is neither real nor unreal.
Introducing that category introduces more problems than it solves. For instance, it
is real or not? The idea flies in the face of experience. Everyone thinks in terms of
things either existing or not existing.
Sections 160–308 provide dualistic and theistic interpretations of scriptural

passages to which Advaitins appeal as the foundations of their position. He
says that we cannot rationally reject the difference between the individual soul
and the Brahman when it is taught by many scriptures. The Advaitin distin-
guishes between scriptures teaching the truth of non-duality and the rest
whose authority is secondary. Madhva points out that this amounts to saying
that some scriptures are false. So why should we accept that the ones teaching
identity are true? We have no way of evaluating scripture apart from scripture.
The purport of the scriptures is the unsurpassable greatness of Vis.n.u. It cannot
be the identity of the individual and the Brahman when this is contradicted by
every means of knowing. Such identity is contrary to experience. No one
thinks, ‘I am omniscient’ or ‘I am the Lord of all’ or ‘I am free from sorrow’ or
‘I am perfect’.
Chāndogya Upanis.ad 6.8.7 says: ‘That which is the subtle essence, that is
the identity of the cosmos, that (tat) is the reality, that is the ātman, and that
(tat) is what you are’ (Sa ya es.o’n.imaitadātmyam. idam. sarvam. Tat satyam. Sa
ātmā. Tat tvam asi).
Advaita Vedāntins read ‘That thou art’ as an identity statement. Rāmānuja
understood it as expressing the relation of inseparable dependence between
body and soul, and between the soul and God. Madhva reads it as ‘You are
not that’. He contrives this by ignoring the natural break between ‘ātmā’ and
‘Tat tvam asi’ so that it becomes ‘ātmātat tvam asi’ which is analysable as
‘ātmā atat tvam asi’. According to the rules of Sanskrit morphology, the long
ā may indicate the coalition of an initial letter a with the letter ā at the end of
the preceding word. The form ‘atat’ means ‘not that’.
Sections 309–362 are devoted to a critique of a form of subjective idealism
which reduces to solipsism. As he puts it, ‘There is no way of proving that the
whole cosmos is a figment of the imagination of a single soul’.
The word ‘prapañca’ is a common expression for the plural cosmos.
Madhva analyses it by the nirukta method and finds that it means the five
kinds of differences. ‘Pañca’ means five and ‘pra’ abbreviates ‘prakr.s.t.a’ mean-
ing expansion. Because this prapañca is called ‘māyā-mātram’ it cannot
be unreal. What does he mean? Māyā, he says, means God’s consciousness.
That which is known (māna) and preserved (trān.a) by God’s consciousness is
māyā-mātram (mā + tra). ‘Since it is known and protected by God, the plural
world is not fabricated by misconception.’ ‘There can be no perceptual error
when one sees directly. Vis.n.u knows everything directly. He sees the universe,
so it cannot be unreal’.
Direct epistemological realism

[343ff.] If the world is fabricated by misconception, there would be two worlds
(one fabricated and another in which someone or something is doing the
fabricating). There is no mistaking a piece of shell for silver unless there is real
shell, real silver and similarity between the two. Even in dreams when traces of
prior experiences are active, a world appears to the mind as existing externally.
In cases like the conch seen as yellow and the sky as blue, the subject is related
to the properties yellow and blue. They exist elsewhere although not concretely
instantiated in those circumstances. There are no perceptual errors without
two similar real entities. This is why it makes no sense to speak of the superim-
position of what is not the self i.e. features of the world) on the self.
There are further considerations in defence of realism in verses 389–395:
The theory that something that is directly seen as real, is in fact unreal, needs the
support of stronger evidence than observation. But if there is no such evidence,
there is no need to suspect observation. What is known by perception cannot be
refuted by reasoning alone without other more authoritative perceptions. We
know on the basis of perception and reasoning that large objects appear small in
the distance. This is not a perceptual error. We can understand that this is the way
things appear to us. We can use perception to establish the scope of perception,
but there could be no way of establishing that all our perceptions are false.
***
The trouble with Avidyā

The Dvaita philosophers present a battery of arguments against the various
ways in which the Advaitin thinkers attempt to account for the appearance of
the many when the truth is that Absolute Reality is single and unchanging.
Advaita blames plurality and the ignorance on plurality on ignorance (avidyā).
The theory that ignorance is a cosmic force inexplicably connected with the
Brahman had become established as the canonical doctrine long before
Madhva’s time. The argument that if the plural universe were just a mistaken
mental construction it would cease to exist whenever anyone understood that
it was such recurs throughout his works. If the universe is but an ultimately
unreal construct and avidyā is always associated with the Brahman, there is no
genuine possibility of liberation.
Madhva applies the principle that illusions and cognitive errors only occur
when there is some similarity between two things. But this cannot apply
globally because there is no similarity between the Brahman and the cosmos.
The view of Man.d.ana Miśra and Vācaspati Miśra that avidyā belongs not to
the Brahman but to the individual self can quickly be dismissed because the
very concept of individuality is fabricated by misconception in the first place.
Moreover, if the soul that has avidyā is really identical with the Brahman, then
avidyā belongs to the Brahman too. Some Advaitins argue that the apparent
difference between the soul and the Brahman derives from some sort of
imagined feature or unreal qualification (upādhi) that becomes superimposed
upon the Brahman. But this is unconvincing. If the upādhi is constructed
(kalpita), this act presupposes ignorance in the first place, and the argument
becomes circular: avidyā produces the upādhi, and the upādhi is responsible
for ignorance. If the upādhi is not constructed and is a beginninglessly real
feature of the individual soul, it follows that there is something that originally
differentiates the individual soul from the Brahman. If the qualification really
belonged to the Brahman, it would compromise its perfect simplicity.
The idea that the Brahman is the substrate of ignorance was intended to
avoid these problems. But if this is true, the released soul will be subject to
ignorance too since the Advaitins suppose it to be the same as the Brahman. If
ignorance is somehow implicated in the very being of the Brahman, it must be
real. If such ignorance is responsible for plurality, then plurality is real and it
would be impossible to escape from ignorance-based sam.sāra.
Further reading
Sarma (2003) is a good start. It needs to be supplemented by Mesquita (2000). Gerow (1990) translates
a text from the subsequent tradition and is excellent on the details of the controversies with Advaita.
The Vis.n.u-Tattva-Vinirn.aya is in Raghavachar (1959) with a translation.

investigation
1. Is Madhva right to think that Rāmānuja’s soul-body model compromises God’s
perfection?
2. In what sense does he have a concept of God as a personal being?
3. Some people have suggested Christian influences. Do you detect any?
4. The Mādhvas were originally Śaiva Siddhāntins. Can you see any connections?
Tantra: Some Śaiva
Philosophies of Kashmir 14
Chapter Outline
Śaiva Siddhānta 194
The three categories: Pati, Paśu and Pāśa 196
Rāmakan..t ha on the enduring individual self and its experiences 199
Personal agency 207
Śākta Śaiva traditions 209
Utpaladeva and Abhinavagupta 210
Absolute Idealism 213
The essential dynamism of consciousness: prakāśa and vimarśa 216
Reinterpretation of Śaiva Siddhānta concepts 218
The Krama cult and the Pratyabijñā philosophy 220
Krama practice 221
Memory, cognition and differentiation 222
Further reading 233
Between 700 and 1100 A.D. Kashmir was home to an extraordinarily rich
and sophisticated religious and intellectual culture. Informed by their own
scriptures (called Tantras or Āgamas), monistic and dualistic schools of
Śaivism (the worship of forms of the deity Śiva, sometimes accompanied by
his female partner or Śakti), and to a lesser extent some Vais.n.avas (votaries of
the deity Vis.n.u), competed with Buddhism for the patronage of rulers and the
adherence of the populace. We shall look at the ritualistic monotheism called
Śaiva Siddhānta that understands the world as a real creation for the sake
of individual conscious souls, and also at some of the monistic Śākta cults
worshipping forms of the Goddess and the fearsome god Bhairava. The
latter follow scriptures that are quite separate from those of the Vedic
tradition, and involve practices involving the violation of taboos as means
of acquiring extraordinary powers, the expansion of consciousness beyond
conventional inhibitions and the propitiation of antinomian deities. They
think that freedom from rebirth is the recognition of one’s true identity
as nothing other than the dynamically conscious source of everything. Both
Śaivas and Śāktas accept the authority of a corpus of twenty-eight divinely
revealed scriptures called Āgamas or Tantras, but the Śāktas expanded the
canon significantly and of course claim finality for their own scriptures.
Most Śaivas accept that the religious observances, gnostic and ritualistic,
sanctioned by mainstream orthodox Brahminism can lead souls to contexts of
experience in higher levels of the cosmos. Indeed, it is of those traditions that
they claim to be superior versions. But they denied that orthodoxy was the
path to the highest attainable good beyond the cosmos of worlds. The ultimate
state is achievable only through the religious disciplines of the Śaiva cults.
Śaiva Siddhānta
This is a Tantric (i.e. non-Vedic) ritual cult teaching that there are three
permanently distinct eternal categories of reality: the godhead named Śiva,
individual conscious souls, and material and psychological realities. Its
scriptural authorities are 28 texts called Tantras or Āgamas that are believed
to be the word of God. It flourished in Kashmir between the eighth and
the eleventh centuries A.D. Important thinkers belonging to this tradition
include Sadyojyotis (c. 700 A.D.), Nārāyanakan.t.ha (925–975 A.D.), his son
Rāmakan.t.ha (950–1000 A.D.) and Aghoraśiva (c. 1150 A.D.). Sadyojyotis
wrote a work called ‘The Examination of God and the Soul’ (expounded by
Rāmakan.t.ha as the Nareśvara-parīks.ā-prakāśa [NIPP]), a commentary on the
Svayam.bhūva-āgama, as well as a number of shorter works. Nārāyanakan.t.ha
wrote a commentary on the Mr.gendra Āgama. Rāmakan.t.ha’s most significant
works, in addition to that just mentioned, include commentaries on the
Mataṅgapārameśvara Āgama [MPAV] and the Kiran.a Tantra.
Śaiva Siddhāntins believe that ritual worship of Śiva, consequent upon
initiation into the religion (dīks.ā) through the imposition of mantras by one
who has undergone a higher consecration (ācārya) and is held to be a human
expression of the deity, is the only means to the human soul’s liberation from
rebirth at death. Initiation and liberation are entirely thanks to the descent of
Śaiva Philosophies of Kashmir 195
Śiva’s grace (anugraha-śakti-pāta). Only Śiva saves. Initiation removes some of
the restrictions on the soul’s potentially infinite innate powers of knowledge
and agency. These, however, cannot be fully manifested in the context of
human life. So initiation does not wholly destroy all the limiting factors proper
to the human condition. The portion that remains is gradually eliminated
over the course of one’s life by the prescribed daily ritual and meditative
observances. Following the prescribed religious path for its own sake prevents
the production of personalizing karma that binds one to rebirth. The latent
accumulated karma that would otherwise have generated further finite
existences is wiped out in the initiation ritual.
The innate capacities for universal knowledge and agency of some souls
have been suppressed by an innate defect called ‘mala’. Mala is also responsible
for those souls’ subjection to bondage by karma and rebirth. The concept
of mala as a substantial and irreducible entity in its own right (vastu) is of
cardinal importance for the Śaiva Siddhāntins because it explains why the
souls undergo subjection to bondage or karmically bound experiences in
the sphere of materiality in the first place. There has to be such a primal and
irreducible defect obscuring the soul’s dual faculty (śakti) of knowledge and
action, because there is no other satisfactory explanation for the process of
transmigration. Originally pure souls would not become involved in rebirth.
(Rāmakan.t.ha discusses this in the sixth chapter of his commentary on the
Mataṅgapārameśvara-āgama [MPAV p. 208 ff.] and in the second chapter of
Kiran.atantra-vr.tti.) This original stain, the root cause of bondage to rebirth,
is categorized as a material substance (dravya) that attaches itself to souls.
Knowledge would be sufficient for liberation if bondage to rebirth were just a
misconception. Indeed, the monistic Śaivas identify it with ignorance and thus
say that it can be removed by knowledge. But knowledge of the presence and
nature of a material substance is insufficient for its removal. Such a substance
can only be removed by action – specifically, the Śaiva Siddhānta initiation
ritual (dīks.ā). Mala is like an ocular cataract, awareness of which does not
prevent its efficacy. Its removal requires the action of the surgeon’s instrument.
When Śiva decides that a human soul, who longs for liberation from rebirth and
accepts the Śaiva teachings, is morally fit for liberation, he induces that soul to
approach an ācārya and solicit initiation. That ritual weakens mala and enables
participation in Śaiva ritual life. The real nature of Śiva is revealed to the initiate
for the sake of the manifestation of his power of cognition. Thus illuminated, he
appears like Śiva and he becomes a Śiva at the death of the body.
Initiation leaves caste, understood as a physical property, intact. The Śaiva
Siddhāntin is thus able to fulfil his Brahminical social and ritual duties.
His exacting life of Śaiva ritual duty is thus compatible with the observance of
mainstream orthodox Brahminical duty and caste purity (varn.a-āśrama-
dharma). The tradition holds that the daily and occasional obligatory rites
must still be performed because there is still a danger of reverting to sam.sāra
if they are omitted. Indeed, one should not transgress the practices of one’s
caste and station in life. (Some of the monistic Śaiva traditions say that
there votaries are in everyday life Vedically orthodox (i.e. observant of varn.a-
āśrama-dharma), in religion a Śaiva (i.e. a Śaiva-Siddhāntin), but in secret a
Kaula (i.e. an initiate into an ecstatic visionary cult whose practices transgress
the boundaries of conventional orthopraxy).)
So the Śaiva Siddhānta is primarily a religion of ritual from initiation until
death. Mala obfuscates awareness that Śiva and the soul are equals (not, for
example, master and servant). Initiation enables the realization of this truth.
But it does not destroy every imperfection. Some karma (were it totally
obliterated, the initiand would die) remains and one is still embodied and
enmeshed in the impure cosmos. Post-initiation performance of ritual
eliminates the residual imperfections. Such observances are not mindless and
mechanical but an enlightened path of active gnosis or understanding-in-
action. Knowledge is only effective when acted upon, and action presupposes
understanding. Daily worship is preceded by a rite in which the practitioner
imagines himself as Śiva, sanctifying himself by the imposition of mantras
on his body and faculties, in accordance with the principle that only Śiva
can worship Śiva. Initiation marks the start of a new way of life and the
transformation of one’s being. Liberation, occurring at death, is understood
as equality with Śiva – meaning a state of qualitative identity in which the
soul’s innate capacities for knowledge and action are fully realized. (In order to
avoid a clash of purposes, the released selves choose not to exercise their
omnipotence.) In should be noted that the tenet that the capacity for agency is
an essential property of the selves in all their conditions is one of the factors
that demarcates the Tantric from those mainstream orthodox traditions that
treat agency as ultimately either illusory or a function of embodiment.
The three categories: Pati, Paśu

and Pāśa
Pati includes Śiva and released souls. The deity is the efficient (nimitta-
kāran.am), but not the substrative cause (upādāna-kāran.am), of the universe’s
cyclical emanation, stasis and reabsorption into its substrate. As well as the
cosmogonic functions, the deity also has the powers (śakti) of concealment
(tirodhāna-śakti) and grace (anugraha-śakti). The former is understood as the
compassionate provision of environments in which finite beings may experi-
ence the fruits of their karma, thereby exhausting its potency, and in which
they may work towards their salvation. Saving grace is primarily manifest in
the initiation ritual. Śiva is the efficient cause in that he activates the real and
beginningless substrative cause, called the māyā-tattva, out of which evolves
the cosmos of inferior worlds. This differs from the view, characteristic of
the Vedānta taught by Śam.kara and Rāmānuja (but not Madhva), that the
Supreme Being is the substrative as well as the efficient cause of the cosmos
that emerges from the divine being. The Siddhāntins reject this on the grounds
that the implicit ontological continuity between God as the material cause
and the world as the arena of effects would implicate the totally transcendent
divinity in the finite, imperfect and physical aspects of the cosmos.
The cosmos comes about so that finite beings may perform actions and
experience of their results. The worlds, or spheres of experience, are organized
in accordance with the accumulated karma of finite beings. World-production
is a compassionate act for the sake of bound souls who need spheres of
experience if they are to be freed from karma and mala. It follows that the
cosmos is ultimately friendly to human beings. The world is shaped by and for
human interests and there is the possibility for freedom and fulfilment of our
highest aspirations.
While the existence of the supreme divinity is revealed by the Śaiva
scriptures, Nārāyan.akan.t.ha and Rāmakan.t.ha hold that it can also be proved
by argument from effect to cause. The occurrence of an effect permits the
inference of its causal factors (sva-kārakam). Just as fire is inferred from
the observed presence of smoke, so the existence of God can be inferred from
the exercise of his creative powers. We conclude that an object such as a pot
has a maker because it is an effect or product. There is an invariable associa-
tion (vyāpti) between something’s being and effect and its having a maker, just
as there is between smoke and fire. Effects require makers with the appropriate
knowledge and power. The success of the causal inference for the existence of
God depends upon accepting that the cosmos as an integral whole can be con-
sidered as an effect. The Mīmām.sakas deny that the world is an effect because
it does not have a beginning (‘things have never been otherwise’). But
Rāmakan.t.ha claims that the world must be an effect because it is complex
(sam.niveśa/sāvayava) and composed of gross matter (sthūla). As such it cannot
be self-created but must have a maker with the knowledge and power appropri-
ate to its complexity. Thus we establish the existence of God.
There is a specific challenge at this point from the Buddhist Dharmakīrti
(600–660 A.D.) who argues that while we can infer in respect of specific cases
of composition that each has a controlling agent, we cannot infer that all effects
have a single maker. In other words, from the proposition, ‘every effect has a
cause’ we cannot infer ‘there is one cause of every effect’. Pots and mountains
are both effects, but they are effects of different kinds. Rāmakan.t.ha thinks that
this is a quibble that undermines inference: it is established that every sort of
effect is invariably concomitant with some kind of maker ([Kiran.atantra
(Goodall) p. 72]).
But perhaps the world just emerges from the material elements. So why try
to prove another cause called God that is absolutely unseen? Rāmakan.t.ha’s reply
is that a non-intelligent cause could not generate the regular and structured
diversity that the world displays. Without super-natural governance the
emergence of entities from matter would be chaotic. It is true that the world is
organized in accordance with the good and bad karma of sentient beings. But
karma is non-conscious, so such organization requires superintendence by a
single deity with the requisite understanding of the diverse karma of beings.
Paśu
Bound souls that are individual centres of reflexive awareness and agency
potentially capable of existing beyond space, time and the physical. Each
has the essential properties of being a knower and being an agent. While
potentially omniscient and omnipotent, some of them have become enmeshed
in inferior physical and mental existences in the realm of māyā where their
deliberate and intentional actions generate residues which personalize and
remain with the agent until circumstances appropriate for their fruition occur
(karma). Human souls are subject to mala, māyā and karma. Such souls are
equipped with five derivatives from māyā that are called kañcukas:
1. A limited capacity for agency (kalā-tattva) bestowed upon souls who would
otherwise be paralysed by mala.
2. A limited capacity for sensory perception and other intellectual acts (vidyā-tattva).
Aghoraśiva says that ‘vidyā is the means by which one knows the intellect (buddhi)
in its various aspects such as judgment, memory, imagination and concepts’
(Tattvasam. grahat.ika, verse 13).
3. A principle of causal regularity (niyati-tattva) ensuring that the results of actions
(karma-phala) accrue to the agent.
4. An interest in the objects of experience on the part of the otherwise apathetic
mala-afflicted selves (rāga-tattva).
5. Our experience of time and its successiveness (kāla-tattva). Time is a created reality
and plays no part in the lives of śiva and released souls.
Pāśa or bonds including māyā, karma, mala and Śiva’s power of concealment.
In addition to the five kañcukas, the products of māyā are prime matter
(prakr.ti-tattva) consisting of the three gun.as (sattva, rajas and tamas), intellect
(buddhi), mind (manas), the sense of ego (aham.kāra), the five sense-faculties
and the subtle forms of their objects, the organs of speech and movement, and
earth, water, air, fire and space. The mental apparatus, being inert and material
is not intrinsically conscious, but it may assume the form of awareness. Mental
faculties are purely instrumental, helping bound souls to find their way around
the world. Rāmakan.t.ha says that the instrument of knowing may be said to be
conscious only metaphorically.
Rāmakan.t. ha on the enduring

individual self and its experiences
Embodied human souls are self-conscious individual agents of knowing
(grāhaka) endowed with a psychological apparatus. They are neither merely
consciousness nor a just a stream of experiences. Their perceptual cognitions
grasp mind-independent realities. We can distinguish between objects as
known and objects existing mind-independently. Immediate experience, prior
to any conceptualization that may lead to sceptical doubts tells us that the
objects of awareness are external things that have effect upon us.
A Buddhist philosopher may argue that everything and everyone is really
undifferentiated consciousness variously expressing itself. He may say that
we do not apprehend any difference between the forms of awareness (ākāra) of
objects and cognitions because they are always co-experienced. A Mādhyamika
may argue for emptiness (śūnyatā): because there are no essences or intrinsic
natures (svabhāva), everything is relative. There are no absolute truths apart
from what humans may agree on and no possibility of an absolute conception
of reality independent of particular human interpretations. Rāmakan.t.ha points
out that either is going to need some means of substantiating his thesis.
Hence whatever we understand the world to be, that will have to become the
object of some means of knowing (pramān.a). There will thus be some sort of
relation between the means of knowing and the objects known. That relation
presupposes that there exists at least one kind of duality. Without a means of
knowing he cannot establish emptiness. The world is objectified (vis.ayīkr.ta)
whenever someone seeks to establish (vyavasthāpāyitum – identifying the
nature of something and discriminating it from others) anything by a means
of knowing. It is impossible to establish anything about that which has not
been made an object. Because of the reality of the process of objectification,
the world cannot be empty in the Madhyamaka Buddhist’s sense. That is to say,
there really are objective standards by which truths can be known.
Rāmakan.t.ha rejects the Buddhist idealist claim that there is non-apprehension
(anupalambha) of the difference between the forms of objects and our cognitions
of them. He says that this is contradicted by the fact that we recognize the
difference between establisher and established. Were it otherwise, we could
not establish anything. The Buddhist agrees that there are methods of estab-
lishing the nature of reality. But such methods cannot just be operating on
themselves because there is a contradiction in something’s performing its
proper function on itself. Hence if there are such methods, they must have
objects external to themselves [MPAV 154–155].
Let us remind ourselves at this point of the difference between the
Buddhist idealists (vijñaptimātra-vādins) and the Buddhist representation-
alists (Sautrāntikas). The latter hold that there is a significant difference
between the way things are and the ways our minds work. They say that
we have to infer (anumeya) the real mind-independent domain (bāhyārtha)
as the ultimate cause of the sorts of experiences that we have. But those
experiences are always interpreted in our ideas. While there is an external
domain consisting of instantaneous unique particulars (svalaks.an.a), it does
not figure as such in the contents of our thoughts. By contrast, the idealist
theory denies that there is a real domain independent of perceptions. Mental
variety derives from the accumulation of mental traces laid down by prior
perceptions. When Dharmakīrti says that there is no difference between the
colour blue and the cognition of blue because they always co-occur, he did
not intend the denial of the mind-independent domain. The argument
recommends agnosticism about its nature: an agnosticism that should wean
us away from our conventional mentality structured by the subject-object
dichotomy. The point was, however, taken by Buddhists and non-Buddhists
alike to be an idealist thesis.
Dharmakīrti had said that, ‘Although awareness is undifferentiated, it is
considered by the misguided to be differentiated into objects, perceivers
and thoughts.’ This can be interpreted in two ways. If it means that in reality
everything is internal to awareness, it is consistent with the idealist outlook.
But it is also consistent with the Sautrāntika representationalist’s anti-realism
if it is taken as meaning that we naturally understand the world in terms of
subjects, objects and experiences, although those categories do not mirror
reality as it is in itself. Rāmakan.t.ha follows the idealist reading. The idealist
will say that an experience of a pot is precisely that: just an experience. Believ-
ing in a world of mind-independent material objects is just such a matter
of experience. But there is a problem here. Rāmakan.t.ha observes that two
sorts of consciousness are pre-reflectively given in everyone’s experience
(anubhava-siddham): there is the awareness of oneself as the perceiving sub-
ject (grāhaka) and there is awareness of objects apprehended (grāhya). Also,
from the phenomenological point of view, we do experience a difference
between the sorts of awareness that we have of our cognitions on the one hand
and of objects on the other. Moreover, the subject is given in its internality as
the constant and uniform perceiver, but the contents of its awareness of objects
known are always changing. So Dharmakīrti’s claim that consciousness is uni-
form fails. Dharmakīrti thought that the notion of a constant subject was an
illusion, a product of mental construction. But Rāmakan.t.ha points out that
while all sorts of imagination and mental fabrications are possible, the basic
identity of the subject cannot be a construct because it cannot exercise the
process of conceptualization (vikalpa) on itself. The constructor can’t originate
itself as a result of its own constructive activity. It has to be there in the first
place. Rāmakan.t.ha argues that the Buddhists cannot make sense of the notion
of constructive superimposition (vikalpa) if everything is instantaneous. A
momentary awareness is no sooner come than gone. If there is no temporal
duration, there can be no mental synthesis of earlier and later. It follows that
recognition, memory and conceptual construction, all of which require both
duration and a single subject capable of uniting separate cognitions, are impos-
sible [MPAV 159–160].
Rāmakan.t.ha argues for a conception of the self as an enduring principle of
identity, whose essential properties are reflexively known consciousness and
agency. The self is that which always reveals objects. It is established by its own
self-awareness as the stable and continuous illuminator of objects. Embodied
human souls inhabit a structured environment consisting of kinds of persisting
objects that exist independently of minds. The world really is as it appears to
us. It is irrefutably given in pre-reflective experience that external things are

the objects of awareness in that they are causally effective [MPAV 155–156].
To say that the self is an enduring substance is to treat it as an entity is not a
property, state or feature of something else. It is numerically one and the same
at different times. Although some of its accidental properties may change, it
retains its essential character. Its fundamental stability does not preclude its
being involved in time and matter through the process of embodiment. We
may contrast this notion of substance with that of an event, which is a reality
that has temporal parts or phases. Examples are plays and cricket matches,
which are spread out in time. Substances and events can be conceived as
persisting through time in two different ways, endurance and perdurance.
Events perdure in virtue of their different phases, although no one part is
present at more than on time. Substances endure by being wholly present
throughout the course of their existence. This applies to the classical under-
standing of ātman and is stated explicitly by Śam.kara. The ātman is involved
in processes through the life of the body, mind, senses and public circum-
stances with which it is associated. Occurrences comprise the life-history
of such a continuant and it makes sense to speak of phases of this history.
But it is a mistake to suppose that such stages of the ātman’s history when
embodied are also parts or stages of that which has the history. On this under-
standing, events in one’s life are not parts of one’s essential identity but parts
of the life with which it is associated.
We have seen that Buddhist philosophers adhere to an ontology of
processes and events (nairātmya – non-substantiality), rather than one
in which enduring substances are the ultimate constituents of the worlds.
They typically reduce whatever is conventionally considered as a stable sub-
stance to sequential occurrences: the human subject is an essentially temporal
succession of phases.
Introducing his polemic against Buddhism (MPAV 150), Rāmakan.t.ha
mentions four traditions of Buddhist thought:
The Vaibhās.ikas and the Sautrāntikas who accept the existence of entities external
to the mind (bāhyārthavāda); the Mādhyamaka relativists who say that the
constituents of reality postulated by the bāhyārthavādins lack intrinsic identities
(svabhāva-śūnyatā); and the Yogācārins who hold that everything is dependent
upon minds.
They all agree that there is no entity called ‘soul’ which is distinct from transient
cognitions, because we have no knowledge of it.
The argument from non-cognition (anupalabdhi), characteristic of Dharmakīrti,
is that when an object or state of affairs satisfies the conditions for knowability
or perceptibility, its not being cognized allows us to conclude that it does
not exist. Dharmakīrti maintains that the soul is the sort of thing that would
be knowable by us (upalabdhi-laks.an.a-prāpta) as separate from essentially
temporal episodes of awareness. But it is not thus known and so we may con-
clude that it does not exist.
Rāmakan.t.ha attributes to his Buddhist opponents the view that personal
identity is just an essential temporal stream of experiences, continuously
subject to destruction:
Immediate experience proves that there is a perceiving consciousness, the basis

of personality and different from the impersonal external world, that is different
at each moment and in relation to each object. There is no further fact called
soul. It is hard to prove on the basis of proof by non-observation the reality
of an entity that is a possible object of cognition when it is never cognised.
[NIPP p. 8]
The Buddhist says that we see consciousness appearing in many forms such
as joy and despondency and concludes that we are a stream of impermanent
cognitions. He further argues that where the knowing subject is permanent, it
would be invariant and could not shift the focus of its attention from object to
object. As Dharmakīrti (to whom Rāmakan.t.ha refers frequently) puts it:
There is no permanent way of knowing because knowledge is authoritative when

it applies to real things. Given the impermanence of knowable objects, it cannot
be static. [Pramān.a-vārttika, Pramān.a-siddhi verse 10]
What is permanent cannot be causally effective either in the present moment

or successively. It follows that there is no permanent identities but only streams
of experiences in which cognitions differ from object to object. Moreover,
cognition and its objects are identical because a difference between them is
never known. It is not possible to know any object distinct from cognition,
because there can be no relation between consciousness and the insentient.
Quoting Dharmakīrti again:
Although awareness is undifferentiated, it is considered by the misguided to be

differentiated into objects, perceivers and thoughts.
All of this is anathema to Rāmakan.t.ha, who is adamant that we are directly

aware of the soul is an enduring substance:
The soul is defined as that which is established by its own reflexive awareness
(svasam . vedana-siddha) as a stable continuant (sthiratayā) in as much as it is always
the illuminator of objects.
The self, which is the ever-uniform stable conscious perceiver, is not a conceptual
construct in that it is reflexively given to each person as the observer of all objects.
[NIPP 166 and MPAV 158]
Given that perception proves the nature of the subject as the conscious agent of
the direct perception of all objects, it is not possible to establish the non-existence
of the self since it is self-evident. Being the subject of karmic experience means
being a cognizer. That is the true form of the soul and it is self-evident to every-
one. [NIPP p. 13]
That whose nature is to have knowing as its essential property is the soul that is
the subject of experiences. The soul is proved to exist for everyone because it is
manifest in one’s own direct experience. [Kiran.a p. 53]
This atemporal soul is constantly manifest as the same in all mental acts.
The individual consciousness that is an essential property of everyone is self-
manifesting or reflexive. The reflexivity of consciousness means that when
a subject is aware of some object or fact, simultaneously and in virtue of the
same act, he is aware of himself as the subject or possessor of the experience.
It is important to remember that in this sort of ‘self-consciousness’ the self
does not appear as an object. As Rāmakan.t.ha puts it: ‘It is not the case that
there are two cognitions: one of the object and another of the self. Rather,
when an awareness of an object is also aware of itself, the nature of the self is
given as that reflexivity’ [MPAV p. 157].
Rāmakan.t.ha has to reconcile the diverse and flowing character of our
mental life with the continuous integrity of the soul that is its subject. The
self is not reducible to the states that are its stream of consciousness. The self
cannot be the same as the states because it is the very condition of those states
occurring as a unified stream. The Buddhist takes the opposite position when
he argues that consciousness is always changing: we only find awareness
appearing in various modes such as joy and despondency but never encounter
a separate entity called self [MPAV p. 150]. He concedes that even if the cognit-
ive capacity of the perceiver is not momentary, it definitely is not permanent
because it comes and goes as expressed in experiences like, ‘I have a headache’,
‘this feels nice’, ‘my sorrow has gone away or it will pass’ [MPAV p. 172]. Given
that we experience cognitions as transitory, it follows that personality is in a
state of constant flux, and a bundle of perceptions is all we are.
Rāmakan.t.ha responds by distinguishing two modes of awareness: the
cognitive discrimination (adhyavasāya) and the permanent background
consciousness. The former is variable because it is a property of the essentially
material and mutable mind (buddhi-dharma). The latter is the awareness
that is a constitutive feature of the human condition (paurus.am). Its absence
is never experienced as it is always encountered as uniform subjectivity
[Kiran.atantra (Goodall) p. 54]. The distinction enables him to say that thoughts
and feelings may come and go, the succession of experiential states may indeed
be variable, but the enduring subject of experiences remains constant, always
revealing itself as the same.
It is undeniable that we experience a stream of consciousness, but this is
different from saying that we are only that stream. Consciousness is a unity
with a perspective upon its different states. The unity of consciousness means
that at any given moment I may be looking at something, feeling something,
thinking about something else, wanting something and deciding to do
something without falling in to schizophrenic morass. The different conscious
acts do not mean that my consciousness is fragmented. My awareness of
the different objects and contents of those states is unified. Consciousness
encompasses the range of mental operations.
Rāmakan.t.ha’s response to the Buddhist account of experience appeals to
two types of argument: one from the phenomenology of consciousness, and a
hypothetical inference (arthāpatti) from the intelligibility of activity that has
future goals:
Everyone knows that he is a knowing subject on the basis of immediate pre-

conceptual experience. The question is whether the perceiver is an appearance of
mere moments, differing in earlier and later experiences, new every moment and
in relation to each object, or is it something that never changes? The answer is
that the constant light of consciousness is given for everyone in reflexive aware-
ness. It knows no schism in itself despite the different imposed features that
are the objects of awareness. In past, present and future, it is exempt from prior
non-existence and destruction. Although experiencing the coming and going
of many mental events such as the various means of knowing, the sense that
one is the constant perceiver is unshaken. In the gaps between mental events,
the light of consciousness is uninterrupted. Self-consciousness is unbroken
in states such as deep sleep. It is called ‘ātman’ because it is always known as
self-illuminating.
It is in reliance on this unchanging and unfailing background consciousness

that people undertake actions with future consequences. Were it momentary,
all activity would collapse. Who would act, where and why if no experience
could extend to another moment, instantaneous awareness just ceasing? Activity
would be impossible for someone supposing that they lived only for a moment,
thinking that in another moment, ‘that is not me, it is not mine’. Everyone
would become inert, absorbed only in the luminescence of their own natures,
lacking conscience of right and wrong and devoid of the many types of
cognitions. This contradicts the immediate experience of everyone because activi-
ties presuppose the stability of the perceiver’s consciousness. (NIPP 13–14 and
MPAV 172–173)
Some Buddhists say:
It is undeniable that there is the appearance of a stable and uniform perceiver.

But that perceiver is not reflexively given. Rather, experience reveals a flow
of perceiver-moments and unity is superimposed upon the stream by misleading
synthetic cognitions due to perception of the similarity of the perceiver-moments,
just as we attribute unity to flow of water and call it a river. But that is a mistake.
This grasping at a stable identity (ātmagraha) is the root of all evil and it is what
the Buddhist teaching aims to suppress. [NIPP p. 14]
Indeed, belief in the soul derives from beginningless ignorance (anādi-avidyā) and
since it causes rebirth should be rejected by seekers after liberation who should
practice repeated contemplation of the non-existence of soul. [MPAV 151]
Rāmakan.t.ha replies that this cannot be right because we are aware of the inner
self as something different in kind from objects. If it were the product of
superimposition, it would appear like an object and as different from whatever
was performing the superimposition. But our experience is not like that.
Rather, given that the self is the illuminator of objects, its nature is that of the
internal perceiver. Any superimposer would have to be a stable subject of
awareness. Were it manifest as purely momentary, superimposition would
be impossible because that requires an enduring consciousness capable of a
synthetic cognitive grasp of earlier and later [NIPP p. 15].
According to Buddhists of Dharmakīrti’s school, an instant has no before or
after. A momentary thought is no sooner come than gone. Its origin coincides
with its destruction. If there is no endurance, how would memory and concep-
tual thought be possible since the agent of the synthesis of thoughts is the
consciousness that is proper to the self? Memory and the synthesis of experi-
ences involve conceptualization (vikalpa) which requires the mental synthesis
of earlier and later by a constant background awareness. But momentary cog-
nitions cannot perform those functions [MPAV p. 159]. Moreover, how can an
instantaneous perceiver objectify itself in such a way that it can mistakenly
impute permanence to itself?
Finally, Rāmakan.t.ha of course rejects the Nyāya-Vaiśes.ika view that the
existence of what for them is a non-experiential principle of identity has to
be inferred since it obviously cannot reveal itself. They posit the self as the
single principle that is necessary for the unification of diverse sensory expe-
riences, for example, touching, tasting and seeing the same thing. It also
explains the possibility of the synthesis of earlier and later experiences over
time. Because cognitions, volitions, pleasures, pains, efforts, merit, demerit
and inherited tendencies are qualities (gun.a) they need a substrate (āśraya)
that is a substance (dravya) and that substance is the self. Rāmakan.t.ha
simply does not accept the Nyāya-Vaiśes.ika ontology about the relationship
between qualities and their possessors. His view is that a substance or
basic particular (dravya) is a confluence of properties (gun.asam.drava or
gun.asamudāya). This is not the same as the Buddhist reductionist view
because substance here means a persisting substrate, an object that is an
integration of properties, where the whole is not a separate entity from its
properties or parts. Rāmakan.t.ha concludes that because cognition, feelings
and intentions are not separable qualities in their own right, we must estab-
lish on the basis of self-evidence, and not inference, that the capacity for
knowing belongs to the nature of the self as its essential property [NIPP
10–11; K53; MPAV153].
Personal agency
The self is not just a stable cognizer or detached observer (as the Sām.khyas
think), it is also a centre of free agency with causal powers (kriyā-śakti).
But the Buddhist cannot make sense of the phenomenon of action since if
the subject of experience were essentially temporal, it could not perform
actions. When something is done by a single instantaneous cognition, its
fruition would be another instant of cognition that would not occur in a later
life of the instigator. Since the instants differ, the experiencer of the fruits of
action will differ from the instigator. The experiencer of the fruit would be a
different entity. The enjoyer of the fruit would be other than the conscious
subject who was the agent of action [MPAV 165–166].
The Buddhist denies that when one person has done something, its fruits
are experienced elsewhere on the grounds that actions and their consequences
constitute different streams of experiences. Rāmakan.t.ha replies that the
question, ‘which stream is which?’ has no determinate answer if the streams
just consist of momentary entities where an earlier moment is followed by a
later one. That would not be sufficient to yield real distinctions such that we
could identify separate streams. In short, there would not be any individual
streams. The Buddhist claims that a relation of similarity between moments
individuates them into streams consisting of the five skandhas (the body,
feelings, sensory perceptions, habits and conceptual thoughts). Rāmakan.t.ha
says that there would have to be some sort of intimate connection between
the moments. It cannot be produced by space or time since the Buddhist
does not accept that they are realities in their own right. Nor can it derive
from the essential natures of entities. There are two points here. Buddhists
reject essences or unchanging natures. But in so far as anything may be
conventionally treated as having a nature, such a nature derives from its place
in a system of relations. So the notion cannot be appealed to as an explanation
of the generation of those systems. Rāmakan.t.ha concludes that here is no
proof of the existence of integrated streams of experiences because discrete
instants cannot produce individual identities [MPAV p. 166].
The Buddhist doctrine of the essentially temporal nature of all entities
involves a rejection of the theory upheld by mainstream Brahminical ortho-
doxy that actions are to be analysed in terms of specific factors (kāraka-vāda)
such as a fixed starting point, the autonomous agent, the recipient, the object
desired and means. They espouse a theory of causation according to which
there are just processes or sequences of events (kāran.a-vāda) in which
individuals, whether agents or patients, are just aspects of a causal event,
enjoying no special significance. The Brahminical view, according priority
to substances and agents as causal factors, is succinctly expressed in the verse,
‘The master of the factors in relation to action and inaction, whether it is
currently active or not, is the factor called the agent.’
It is soteriologically crucial for Rāmakan.t.ha that the persisting self-
conscious individual substances, embodied and enduring intact through time,
are autonomous ritual and moral agents spontaneously capable of initiating
novel sequences of events that follow from their decisions and that are
not wholly produced by antecedent causal conditions. That is to say, souls are
individual substances possessing innate causal powers that are dispositions
to act in certain ways in appropriate circumstances. Where events involving
human actions are concerned, souls are, as it were, the glue that holds the
members of a sequence of ephemeral events together as a causal process
and thus account for its continuity. As we have seen, the stable and enduring
ātman, the transcendental enabling condition of experience, which is given in
experience but not produced by it, is exempt from determination by time. Its
agency is not determined by sequences of events.
Personal agency, as opposed to behaviour which may be merely reactive,
instinctive and non-conscious, is necessarily connected with consciousness
and is introspectively manifest to oneself as the reason for one’s physical exer-
tions and movements. In the case of other people, it is analogously inferred
from their bodily actions. Personal agency, which has the nature of autonomy
(svātantrya) in that it is the rational basis for the function or non-function of
all the factors involved in events (kāraka), is directly experienced as being
responsible for effort and physical movements, the performing of religious
and everyday actions having seen or unseen results. It cannot be denied
because, like the state of being a cognizer, it is directly known to each as a form
of internality in that one is the inner instigator of the factors implicit in events
[NIPP pp. 95–96].
He claims that the Buddhist theory implies that it does not matter whether
Devadatta provides services for monks or kills them. Since service and killing
are equally treated as contributory causal aspects of an event, the merit and
demerit proper to each would accrue indiscriminately to Devadatta and to the
mendicants. Where could the difference lie if everything is just an aspect of a
causal process that is a sequence of events? The Buddhist reply that the causal
process is differentiated into streams that are the individual Devadatta as
the instigator, service as the intended purpose and the mendicants as the
beneficiaries then this amounts to acceptance of the kāraka-vāda which
involves the categories of agent, object and instrument and not the process
theory of causation (kāran.a-vāda). And if the kāraka theory is established, so
is the agency of the self [MPAV 171].
Śākta Śaiva traditions

As well as the Śaiva Siddhānta dualists there were many worshippers of
forms of Śiva and the Goddess who subscribed to a non-dualistic (advaita) or
monistic metaphysic. While believing that it is knowledge and not ritual that
is essential for liberation, adherents of these cults enjoyed a rich liturgical life.
Some rituals confer specific benefits and powers. But ritual practice may also
help to consolidate belief, deepen commitment and keep alive an original
inspirational insight by preserving a sense of enlightened deliverance from
the frustrations, changes and chances of daily life. Enlightenment is under-
stood as recovery of one’s true identity as the deity. Salvific realization may be
achieved by ritual informed by gnosis, or by gnosis alone, or it may simply
happen unexpectedly thanks to a purely fortuitous descent of divine grace.
While enlightenment and liberation, understood as the salvific expansion of
consciousness bestowed in initiation, are possible in the course of one’s life
(jīvan-mukti), most initiates have to wait for death, which is coterminous with
the exhaustion of the residual karma appropriate to this life, to experience it
fully. The life of ritual practice confirms and intensifies the original liberating
experience, purifying it of conceptual elements. Thus enlightened, one sees the
world in a new light.
Utpaladeva and Abhinavagupta

Utpaladeva (925–975 A.D.) follows Somānanda (900–950 A.D.), who was the
original theorist of the Pratyabhijñā school. Somānanda’s Śivadr.s.t.i expounds a
form of absolute idealism, the philosophical outlook that denies that physical
or material things have any reality independent of universal consciousness.
An aspect of the argument is that if there were a real difference in nature
between consciousness and material objects, knowledge of the world would
be impossible. Material things, whether atoms or concrete wholes, on the
one hand and consciousness on the other are utterly different categories
and cannot be connected. Relation is possible only when categories have
something in common. So a relation between thinking minds and objects
is possible if consciousness is the common factor present in everything. To
be is to be a manifestation of consciousness. All conscious subjects are
essentially the same. The universal consciousness is identified as the supreme
godhead named as Śiva, who is present everywhere. Everything is a manifes-
tation of the single divine consciousness and ultimately there are no real
individual identities.
Human problems start when we just think of ourselves as isolated individuals
with caste-based social identities confronting a separate material environment.
The point of religious practice is the recovery of one’s true identity as Shiva
through the expansion of one’s conscious energy.
Somānanda criticizes Advaita Vedantins who think that the differentiated
cosmos is an illusory manifestation (vivarta), due to mysterious ignorance,
of the static, featureless Unconditioned Reality (Brahman). He attacks the
type of idealism taught by the Buddhist Vijñānavāda that admits streams of
consciousness but regards our experience of the material world as merely
mental construction out of vestiges of prior experiences. Moreover, Buddhist
temporalists cannot allow that there is a stable subject doing the imagining.
Commonsense realists are criticized for admitting individual centres of
consciousness and agency, but distinguishing them from their physical
environments in such a way that no sense can be made of their relation to it.
Śaiva Siddhanta dualists are castigated for positing themselves as individuals
independent of godhead.
It is Utpaladeva (925–975) who provides a philosophical defence and
articulation of the sort of visionary spirituality that is central to the Krama
cult. His works include the Īśvarapratyabhijñā-kārikās [IPK] (with his own
commentaries), a treatise in the Nyāya style proving the existence of
God called the Īśvara-Siddhi, a work called the Ajad.apramātr.-Siddhi (about
the knowing subject) and a treatment of the topic of relations called the
Sam.bandha-Siddhi.
Abhinavagupta (975–1025 A.D.) was an influential theologian, philosopher
and aesthetician of remarkable profundity and intellectual sophistication and
erudition who belonged to the Tantric Trika cult. The Trika (meaning ‘triad’)
was a system of ritual originating and developing in Kashmir whose goal is the
acquisition by the votary, who has undergone a caste-obliterating initiation
ritual, of the supernatural powers of a triad of female deities. The latter per-
sonify the human-friendly as well as the terrifying and destructive aspects
of existence. Associated with this cult was that of the eight mother goddesses
and their expressions in families (kula – hence the term Kaula for forms
of Śiva-worship allied to the Trika) of female spirits called Yoginīs. They may
be invoked and pacified, in the impure cremation ground on the margins of
society, with offerings of impure and hence potent substances such as blood,
flesh, wine and sexual fluids. The cult adopted the horrific, all-devouring
Kālasam.kars.in.ī form of the goddess Kālī as the unifying form of the original
three goddesses. From 900 A.D. the Trika was in competition with the Śaiva
Siddhānta dualistic system of ritual and theology, according to which really
individual selves inhabit a physical world. Assimilating the sophisticated
Pratyabhijñā philosophy, the Trika was able to defeat the challenges posed
by dualism, Vedantic illusionism and Buddhism. Its explicitly sexual rituals
underwent a process of domestication and internalization. This trend appears

in the thought of Abhinavagupta where orgasm is understood as obliterating
one’s self-centredness and manifesting the expansion of that blissful self-
awareness that is the same as the universal consciousness projecting all
phenomena. The ritual use of impure substances, which had been understood
as sources of magical powers, is held to induce ecstasy, a sense of freedom
arising from violation of the taboo. Bondage to rebirth is understood as a
state of ignorant self-limitation that understands the orthodox values of purity
and impurity, as objective realities. Enlightenment presupposes the realization
that anxious concern about caste and related values such as one’s Vedic
learning, family’s status, prescribed conduct, conventional virtues and pros-
perity are aspects of an inauthentic identity. Liberation from rebirth consists
in the realization, typically through yoga and meditation, of oneself as a con-
tracted form of the universal consciousness. Freedom just is this awareness: it
is not a separate phenomenon produced by knowledge.
Abhinavagupta’s major philosophical works are commentaries on Utpaladeva’s
Īśvarapratyabhijñā-kārikās, on the Mālinīvijayottara Tantra, the Tantrāloka,
abbreviated as the Tantrasāra, which voluminously expounds the doctrines,
yoga and rituals of the Trika cult, the Parātrīśikā-Vivaran.a, the Bodhapañcadaśikā
(translated below) and the Paramārthasāra, a translation into Śaiva categories
of a Vais.n.ava work attributed to Ādiśes.a.
As well as demonstrating the coherence, soteriological value and finality
of the forms of Śaivism to which he was personally committed, these works
offered include sustained critiques of the dualism of the Śaiva Siddhānta ritu-
alism, the Vedāntic illusionism maintaining that all normal human experience
is infected by ignorance of the truth about reality and the Buddhist rejection of
persisting substances (nairātyma), including the soul. Indeed, there is a sense
in which here we find a polar opposite to Buddhism. While for the Buddhists,
the world is at base an impersonal process of events where we who mistakenly
think of ourselves as persisting centres of thinking and willing are in fact but
expressions of external forces over which we have no control, for these Śaivas
the world, pervaded by the divine consciousness, is not ultimately inimical to
our best interests.
Basic to his eirenic outlook is a belief that other doctrinal systems are not to
be treated as opponents but as aspects of the self-expression of the supreme
conscious reality. He formulates an inclusive hierarchy of belief systems in
accordance with how closely they approximate to the view that ultimate reality
is dynamic universal consciousness.
Abhinavagupta elaborates and refines the absolute idealism taught by
Somānanda and Utpaladeva. The world derives from a single universal,
autonomous and dynamic consciousness that expresses itself in an infinite
variety of subjects, objects and acts of awareness. The seeker after release
from rebirth is to meditate upon the nature of consciousness, oscillating between
the manifestation of objects (prakāśa) and reflexive awareness (vimarśa).
In ordinary individual awareness, the representation of what are taken to
be external objects, and hence duality, predominates. The adept should reflect
that what is experienced as the objective world that we inhabit is nothing
other than the transcendent consciousness, expressed as the union of Śiva
and Śakti, expressing itself. The subject-object polarity is understood as
internal to consciousness. Ritual, yoga and meditation enable us to decon-
struct that polarity, with a consequent collapse of all discursive thought.
Contracted self-awareness is thus dissolved, along with thinking about
the world in terms of external objects confronted by individual subjects
and agents.
Following Somānanda and Utpaladeva, Abhinavagupta believes that the
mind-independence of matter is impossible. Trans-individual consciousness
causes the manifestations that we experience appear as if distinct from the
subjects of experience. While Advaita-Vedānta understands the foundational
consciousness as static, for this school it is self-conscious activity and will
positing itself as apparently other than itself. It operates through projecting
objective Ideas (ābhāsa) that are the contents of our experiences. Against
Buddhism he argues for the self-conscious subject, the permanent background
to experience, persisting as a stable unity that synthesizes mental states.
Without such a principle, there would only be momentary, self-contained,
unrelated mental episodes.
Central to his philosophy is a critique of the notion that our thoughts
represent physical objects. He argues that consciousness can only represent
what is itself conscious. We only know our experiences. It makes no sense
to say that consciousness represents what is mind-independent because as
soon as something is represented by consciousness it is no longer mind-
independent.
Absolute Idealism
The philosophical articulation of the Śākta cults is a form of Absolute Ideal-
ism: the view that everything is a manifestation of a single trans-individual
consciousness, which is the only reality that has independent existence. (Since
there has been some controversy about whether some of the Indian thinkers
who have traditionally been characterized as idealists really are idealists,
I should say that I take idealism to mean the rejection of the possibility
of material substance that exists independently of some consciousness. On
this interpretation, Vasubandhu and his followers in the Vijñānavāda tradi-
tion, as well as Utpaladeva, Abhinavagupta and Ks.emarāja are idealists.) This
sort of idealism is not philosophical scepticism about the existence of the
external world. It is the affirmation of the real world as a partial expression of
the divine nature. The world really is independent of human minds. It is not
fabricated by individual perceivers. It is not surprising that people who devote
much time to mental purification by profound meditative exercises, in which
‘thought-forms’ sometimes appear to the practitioner as external, should
incline towards an idealist mentality.
For these ecstatic idealists the world exists only as its representation in
consciousness: it is not the case that there are two types of substance, the
mental and the physical. The theorists of these cults understand the sole
conscious reality as dynamic, projecting both finite centres of awareness
and the experiences of the kinds of objects that they enjoy.
In his Śivasūtra-vimarśinī, Abhinavagupta’s successor Ks.emarāja expresses
the idealist mentality like this:
Consciousness manifests itself both internally [as thoughts and feelings] and
externally [as things and events] in a variety of forms. Because objects only exist in
relation to consciousness, the world has the nature of consciousness. For entities
cannot be known without consciousness. So it is concluded that consciousness
has assumed the forms of entities. By contemplating entities, we can rationally
understand that knowable phenomena are conscious by nature. Consciousness
and its objects have a single nature because they are experienced simultaneously.
[ad Sūtra 30]
In addition to the difficulties about finding a coherent conception of matter

(that which keeps physics in business) and the philosophical problems of
explaining our perception of the physical world, there are a number of consid-
erations that encourage people to think that consciousness, not matter, is the
basic reality, and to adopt the idealist outlook. Try to imagine a cosmos in
which there never has been and is no conscious life. Remember that there are
there are no observers, no experiencers, no intelligences. What would it be
like? Your initial thoughts may be that it is all black, hard, soft, fluid, hot, cold.
But it isn’t even that. Any of those characterizations presupposes that there is
an observer or experiencer. As Abhinavagupta says,
Regarding the modes of thinking, ‘I know’, ‘I knew’, ‘I shall know’ which are
founded upon reflexive subjectivity, what else is there to know? If these did
not shine, the cosmos would be dense darkness, or rather it would not even be
that. ‘How does the knower know itself?’ If one denies the conscious subject,
what question and what answer would there be? [IPV p. 71]
We find that by strictly eliminating observers or intelligences, nothing has any

meaning. In the lifeless world there isn’t really any-thing: there literally are no
things. This is because the identification and classification of entities requires
conscious observers. In the dead world of chaotic matter, there cannot be
any intrinsic structures, organization, repeatable forms or natural laws. Such
organizing principles have to come from the outside. They cannot emerge
from within. That which imposes order cannot arise from the indeterminate
morass upon which it then imposes order. If natural laws impose regularities,
they must be different from that which they regulate. They have to be external
factors. Now our ‘bleak and blank’ chaotic world is by definition purely physi-
cal. So if there are to be external factors, structuring principles that necessarily
are external, they must be other than the physical. And what is non-physical is
conscious. Leaving imagination to one side, can we even conceive of their
being a world without consciousness? A conception of reality as it is in itself
not involving mention of human perceptual capacities would be a complex
mathematical structure, but that is not our world because human experience
has been left out of the picture.
Perhaps the best we can say is that there would be something there, but it
would not mean anything. Indeed, merely saying, ‘there would be something’
does not really mean anything. So some people think that it is senseless to say
that there could be anything at all if there were no consciousness, and this is
what idealists mean when they say that there can be no unexperienced reality.
Everything depends upon consciousness. Consciousness, not matter, is basic.
There is another sense in which everything may be considered to depend upon
consciousness. Let us not be afraid of the obvious and accept that there are
physical objects in space. If what such objects are is to be understood or deter-
mined, if their existence is to be established, if they are to mean anything, some
sort of intelligence different from those objects is required. But consciousness
differs in a crucial respect from physical objects. It reveals or establishes its
own existence. It knows its own nature simply by being what it is. It does not
need anything outside itself to do it for it. In this sense, it is consciousness, not
matter, which is basic.
The essential dynamism of

consciousness: prakāśa and vimarśa
People initiated into these Śaiva Śākta traditions accept the categories recog-
nized by their dualistic co-religionists, but claim that they are ultimately
aspects of the one purely conscious divine reality. One of the ways in which
they differ from the Advaita-Vedāntins is in their insistence that the uncondi-
tioned creative consciousness is dynamic, not inactive. The transcendent trans-
individual consciousness (parā-sam.vit) is the pure actuality of self-awareness.
It knows itself. It is fully realized, perfect and self-sufficient, ever and always
wholly present to itself. It is a state of peaceful repose (viśrānti) where actuality
and potentiality are in perfect dynamic equilibrium. Unconditioned by space,
time or form, this state of perfect balance lacks nothing and so has nothing to
accomplish. The divine sovereignty consists in perfect freedom (svātantryam).
This trans-individual consciousness generates the finite realm of multiplicity
and relations. Divinity, understood as the co-inherence of Śiva and Śakti,
spontaneously contracts itself to produce the matrix of individual knowers
and agents, all experiences and acts of knowing and all phenomena that appear
as if they were separate from it. We shall see how this activity is conceived
not as impersonal and unspecific surges of energy, but on the model of the
processes of finite intelligence. Unconditioned consciousness is the freedom of
self-determination. Creation is neither necessary to the Divinity, nor does it
require recourse to an independent material principle.
Śiva and Śakti correspond to prakāśa and vimarśa, two modes of the one
consciousness. These terms are rich and complex in meaning. Prakāśa liter-
ally means light, in this context, the light of consciousness. Vimarśa means
self-awareness or reflexivity, and representation. The divine consciousness is
an eternal dynamic and vibrant interplay between these two facets. To grasp
the difference we need to explore some aspects of the internal variety of the
notion of mind. We can distinguish psychological states and phenomenal con-
sciousness. We can think of perceptions and cognitions as sometimes purely
functional psychological states. Information is received by perceptions and
stored as memories. In terms of this outlook, a subject is in a perceptual state
whenever their psychological apparatus is receiving and processing data from
the environment. A subject can have a perception in this sense of being in an
informational state even in the absence of the corresponding subjective
awareness. This is what is sometimes called being on automatic pilot. My
field of vision at the present moment is greater than what I am seeing.
Sometimes I walk down the road, oblivious of my surroundings, on automatic
pilot, as it were. But what we call experiences are phenomenal conscious
states. There is something that it is like to have them or to be in the state. This
is sometimes called the subjective character of experience – what it feels like.
A state is phenomenally conscious if there is something it is like to be in that
state. This is one of the meanings of vimarśa. Abhinavagupta says that it is
the life, the vitality of awareness (prakāśa). Items of belief and knowledge,
when they are to be stored, are potentially conscious states that can be
brought to mind and made explicit in awareness. A thought or representation
is ‘alive’ if it is being used in reasoning and in the direct rational control
of action and speech.
The prakāśa of objects may mean their objective manifestation, their
capacity to become objects of explicit awareness. Prakāśa is like a light shining
in a dark room. But this in itself is insufficient for understanding. This is where
vimarśa comes in. Explicitly conscious representation of objects, their shining
in the consciousness of a perceiver is vimarśa. A subject’s explicit awareness
of itself as a centre of cognition and agency is called ‘I-representation’ (aham-
vimarśa). (Utpaladeva says that the reflexive awareness ‘I’ is not really a concept
because it is simple and does not require us to make any discriminations
because the ‘I’ is infallible and immediately given.)
Abhinavagupta says that consciousness without reflexivity (vimarśa) would
be blind. Reflexivity is both Śiva’s power (Śakti) of self-awareness and his
representation of objects (and perceivers) within and to himself. The latter is
called their being reflected in his own mirror.
We said that prakāśa is like a light introduced into a dark room. Although
there is illumination, this is insufficient for understanding. When the Nyāya-
Vaiśes.ikas say that cognition is just the illumination of objects (artha-prakāśa-
buddhi), they are treating consciousness as a searchlight illuminating objects
in the world. Take the case of immediate sensory perception and think of it
as pure observation. A video camera does something functionally analogous,
but it does not understand. The view of Utpaladeva and Abhinavagupta is
that vimarśa must be implicit even in immediate sensory perception if the
perception is to mean anything. For a perception to mean anything it has to
be raised to the level of explicit consciousness. There has to be a phenomenal,

experiential component (vimarśa). Also, for the perception to really mean any-
thing, it must be connected with other thoughts, and this sort of synthetic expe-
rience (anusam.dhāna) is also vimarśa.
We should bear in mind that it is not the case that conscious experience
consists of discrete cognitions of isolated objects. We do not see bare objects or
simple single things. The notion of present experience may be misleading.
In conscious experience we bring to bear a mentality informed by retained
memories and expectations for the future. I am looking at a cup on the table in
front of me. (I don’t think, ‘there’s an object above another object at such and
such a distance’ and then identify the objects and relations in the state of
affairs.) I know that it holds the coffee that I made earlier. I reach out, pick it
up and take a sip. I enjoy it. I know that it will be there later to be enjoyed
again. This ordinary experience is replete with acts of cognitive identification,
synthesis and separation. It is also pervaded by language. This is vimarśa.
The polarity of Śiva and Śakti, Prakāśa and Vimarśa is expressed in a text
belonging to the Krama tradition as follows:
Śiva’s tranquil state is the highest form of self-awareness. But there is an even
higher state that is ever so slightly distinct, and that is the abode of the Goddess.
The whole of reality comes from the creative light of consciousness (prakāśa), itself
deriving from the sheer delight that lacks nothing and which itself finds its rest
in the uncreated light wherein there are no traces of awareness of differences. The
Goddess is the unsurpassable tranquil state that has consumed the traces of
awareness of existences that had remained in the uncreated light. Śiva’s nature
is the tranquil state that devours time. The Goddess is the perfection of that
tranquillity. (Mahānayaprakāśa 3.104–11. Text cited in Sanderson (2007), p. 309)
Reinterpretation of Śaiva Siddhānta

concepts
Not only does the Divinity always know itself, but it also knows itself in and
through the creative process. Individual transmigrating selves are modes
of the Divinity that permits its own contraction by māyā, karma and mala.
Whereas the Śaiva Siddhāntins understand the latter as a material substance
attaching to the souls and restricting their powers of knowledge and action,
the non-dualists think that it is ignorance in the form of the unenlightened
acceptance that one is just a limited individual bound to rebirth, subject to
caste and social and religious obligations, in danger of pollution by objectively
real impurities that hinder spiritual progress and hence constrained by the
Brahminical value of purity, and one for whom freedom (moks.a) is a remote
possibility, difficult of attainment. This manifold Ignorance causes bondage.
Ignorance at base means the mistaken belief that there are individual entities,
including souls, which are capable of existing independently of the Divinity.
This ignorance is constitutive in that it establishes individuality.
Māyā, for the Siddhāntins the substrative cause of all objects, physical and
mental faculties and experiences in the impure levels of the cosmos that we
inhabit, is understood as the projection of the whole differentiated realm of
objectivity so that it appears as if it were other than the Divinity, and from the
many distinct and finite subjects of experience. It may also be understood
in terms of thinking of oneself as a limited individual who confronts a mind-
independent objective order. Similarly, karma is not a factor external to the
self. Rather, it is the conviction on the part of the self-limiting subject that
good and bad deeds really have a bearing upon one’s destiny.
Bondage to rebirth just is the belief that one is limited resulting from a
failure on the part of people caught up in the conventional dualistic outlook,
with its bifurcation of individual conscious subjects and material objects,
to understand that their true identity is pure, autonomous consciousness.
Liberation is the non-discursive (nirvikalpa), direct and fully expanded
experience (i.e. not merely a propositional thought that one is identical with
the Divinity) of being nothing other than the transcendent consciousness
(parā sam.vit), perpetually delighting in itself. In short, one becomes the
divinity: ‘I am Śiva and this whole world is my self-representation’. Śaiva
Siddhāntins may imaginatively identify themselves with the deity in liturgical
contexts, but they do not believe that they really are God. That is the crucial
difference between the dualistic and non-dualistic religious paths.
We have seen that the ultimate goal is to lose all sense of personal individu-
ality by recovering the awareness that one is not different from the Divinity.
This is sometimes expressed as immersion (samāveśa) in the realization of
the all-encompassing supremacy of Śiva, who is autonomous and undivided
consciousness uniting prakāśa and vimarśa, where the universe is experienced
as non-different from one’s identity. In its extroverted mode, trans-individual
consciousness projects all experiencing subjects and all phenomena in a
kaleidoscopic manifestation that is always aware of itself. In its introverted
mode, consciousness simply delights in itself. What we experience are the
multifarious self-representations of the trans-individual consciousness that
contains within itself all the projections. Experiencers, as well as phenomena,
are modes of consciousness contracted by time, place and forms.
The Krama cult and the Pratyabijñā

philosophy
The Pratyabhijñā philosophy is a sophisticated reflective articulation of
the Krama cult, central to which is the attainment of liberating gnosis in
the context of a form of liturgy known as the ‘Worship of the Twelve Kālīs’.
Freedom from rebirth is just re-cognition (‘pratyabhijñā’): the recovery
of knowledge of the truth that one’s real and ultimate identity is nothing
other than the trans-individual consciousness that projects all phenomena,
experiences and limited subjects on the part of a subject that had previously
considered itself as an individual thinking agent confined by space, time
and embodiment. The multiplicity of individual subjects in reality is the
one Subject and only this subject exists. This single real subject diversifies
itself into limited subjects and objects. The boundaries are termed ‘upādhis’
(‘superimposed conditions’ rather than genuine properties). The point is
that while seen real to us, any changes in and of them do not affect the uncon-
ditioned subject. This is an absolute idealism according to which everything
that is experienced by us as material and everything apparently individual
is projected by a single consciousness. It is argued that whatever causes the
physical world must be non-physical because the world needs a source that
is external to it and different in kind from it. Since individual centres of con-
sciousness are localized by matter, the ultimate source must be unconditioned,
creative universal consciousness. They argue that we can only make sense of
the coherence and continuity of our experiences, memories and interpersonal
communications if they belong to single persisting conscious subjects that
inhabit a stable world regulated by objective structures. This is extended
macrocosmically: the universe of subjects and objects holds together because
it has a single conscious source that preserves it in being. The co-ordination
of diverse subjects and objects is possible only if they are aspects of a single,
universal field of experience. Universal Consciousness causes objects of
awareness to appear as if distinct from the limited subjects of experience.
The forms figuring in our awareness express the Ideas (ābhāsa) projected
by the universal consciousness.
The goal of religious practice is the transcendence of limited individual
subjectivity, indeed of any subjectivity whatsoever. The state of bondage to
rebirth means thinking that one is a self, a person or personality. Real freedom
means the obliteration of petty selfhood. Enlightenment is the realization that
the subject that has mistakenly and selfishly considered itself as an individual
is identical with the universal transcendental conscious energy called the
Śiva-Śakti state. Freedom from rebirth is just the recognition that, ‘I am Shiva
and this whole world is my self-expression’. The authentic identity is already
present as the constant and undeniable self-awareness that is in the back-
ground of all experiences, but it must be re-cognized and reflected upon as the
ultimate conscious principle manifesting itself as all limited subjects, acts and
objects of experience.
What appear as external, physical objects depend upon consciousness.
Since causal agency is exclusively a property of conscious agents capable of
volition, creativity by the physical is impossible. Since only an idea can be like
and thus represent an idea, consciousness would not represent matter if matter
were something totally different from it. We cannot experience anything other
than consciousness.
Krama practice
The Worship of the Kālīs is a meditative sequence of twelve phases (each sym-
bolized by one of the twelve Kālīs) that effects an expansion of consciousness
from the confines of limited personhood to an enlightened form of awareness in
whose light the everyday world becomes transfigured. In other words, what had
been experienced as the merely mundane is recognized as the self-expression
of the Divinity. In the course of this worship, consciousness transforms itself
as it ‘devours’ both its own contents and its awareness of itself as individual.
One contemplates the emanation (sr.s.t.i) of the cosmos from its transcendent
source, its conservation in being or stasis (sthiti), and its withdrawal (sam.hāra)
into that source, followed by its repeated emanation and so forth. That
process of cosmic emanation is mirrored on the microcosmic level in the
sequential structure of normal cognition that reaches out to objects, focuses
attention upon them and absorbs them into itself. A clue to the nature of the
divine activity is found in such modalities of human consciousness, which
mirror creation, conservation and withdrawal. Our states may be more or less
self-aware. Sometimes we are in an extroverted state, totally absorbed in some-
thing and not really self-aware. But self-awareness brings consciousness to
life. The interplay of extroversion and introversion in our own minds is held
to be a microcosmic imitation of the divine nature. In the Krama ritual one
symbolically contemplates the cyclical process of cosmic emanation, stasis and
reabsorption as represented by the path of cognition from its starting point

as the initial state of the knowing subject (pramātā), via the internal mental
faculties and the extroverted faculties of will, cognition and action (pramān.a),
to its intentional object (prameya), and then back again as the object is inter-
nalised in the subject. The phenomenal representations of the objects are
withdrawn into the knowing subject, which is the terminus or resting point
(viśrānti) of the process ending with something understood. Worship cul-
minates in ‘the phase of the nameless’ (Anākhya-cakra), the unifying basis of
the process of projection, conservation in being and withdrawal. This is
the dissolution of all differentiated modes of cognition into the radiance of
consciousness that is common to all mental acts and states. The final phase
is symbolically expressed as the Goddess Kālī. She is beyond being and non-
being, an abyss of pure light in which the powers of knowledge and action have
merged, and where the distinctions between subjects, acts and objects of
knowing have collapsed. From her unconditioned nature the diversified cos-
mos is manifest. She is attainable only in mystic gnosis. As Abhinavagupta’s
successor Ks.emarāja (1000–50 A.D.) puts it in his Pratyabhijñāhr.dayam 19:
By means of the internal trance of the Krama he remains immersed in expanded

consciousness while still living in the world, and in this way achieves the final goal.
In this process, he enters in from outside. By the very force of this penetration, he
enters from within into his outer identity.
The language of penetration, emission, immersion and withdrawal may be

suggestive of sexual activity. Indeed, Abhinavagupta says that such is an appro-
priate context. In fact, any ecstatic experience will do.
Let us now look at what Utpaladeva and Abhinavagupta have to say in some
of the Verses on the Re-cognition of the Divinity (Īśvarapratyabhijñā-kārikā
(Torella, 2002)) and their explanations of them.
Memory, cognition and differentiation

Iˉśvarapratyabhijñā-kārika 1.4.1
The unrestricted consciousness that is the perceiver of the object previously
experienced and that is still in existence at the later time, realises that the present
object is the one previously experienced. This is called remembering.
By considering the operations of finite minds – which are microcosmic expres-

sions of the ‘Divine Mind’, we can achieve some insight into the nature of the
transcendent consciousness.
At the end of the third section of the first chapter of his Īśvarapratyabhijñā-
Kārikās (IPK), Utpaladeva mentions the three divine powers of knowing,
memory and differentiation (referred to in Bhagavad Gītā XV.5):
If there were not one conscious Divinity who contains the infinite universe within
himself and who has the powers of knowledge, memory and differentiation,
the harmonious functioning (sthiti) of the human world, which stems from the
synthesis by consciousness (anusamdhāna) of different and separate thoughts,
would cease’.
Most Brahminical philosophers accept the definition of memory in Yoga-

Sūtra 1.11: ‘memory is the retention of an object previously experienced’.
The alternative Buddhist account says that some perceptions leave a vestige or
trace (sam.skāra) in a stream of experiences. But while their theory may be an
adequate account of the mechanical transmission of information, it leaves out
the subjective, phenomenological component that is integral to memory. The
Buddhists say that the trace is revived at a later time when an experience calls
to remind us of something similar about the past. But there are problems here.
Experiential memory (as opposed to my recall of the stored information that
the Sanskrit word for horse is ‘aśva’) is both memory of the past object and of
the previous perceiving. If everything is momentary and in a flux, the stream
of experiences is in a different state and past objects and events have ceased to
exist. Memory cannot be the recovery of a past experience if that has not been
retained. The trace is neither the original experience nor the object as it was
cognized. Hence we are not in a position to know that the present perception
resembles the past one. The account also fails to do just to the phenomenology
of memory experience. We don’t just recall past objects and events: we often
remember what it was like for us to experience them and the account leaves
this out.
Experiential memory presupposes a unitary and active consciousness that
surveys different moments of time. This subject is the agent that can relate
cognitions occurring at different times. Given this, we can say that the past
and present perceptions are the same in their phenomenological aspects,
in their self-awareness (svasam.vedana), and that this provides the link
between them.
The idea is that if all minds and phenomena are aspects of a single conscious
field, we can explain how we can have knowledge of other minds, and commu-
nicate with one another. We can now overcome what is sometimes thought to be
a problem for types of mind-matter dualism: how can there be any sort of rela-
tion such completely different realities as consciousness and insentient matter?
Iˉśvarapratyabhijñā-Kārikā (IPK) 1.4.3–5

There would be no manifestation of the object being remembered if it appeared
as separate from the memory [i.e. were it not manifested in the present as
an idea internal to consciousness but as something external and distinct from
consciousness]. Therefore the unification of cognitions occurring at different
times presupposes that there is a persisting subject of experiences. In memory
the former experience does not appear like a separate external object, since
it appears as resting within the self and is expressed as, ‘I experienced this in
the past’.
In his introduction to the fifth section of the first chapter of his

Īśvarapratyabhijñā-Vimarśinī Abhinavagupta says that Utpaladeva begins
with the power of knowing (jñāna-śakti), and proceeds to say that the light
of consciousness is essence of objects. The text then establishes the existence
of realities independent of finite minds by refuting the Vijñānavāda view
that mental variety derives from stored traces or vestiges (vāsanā) of prior
perceptions. He then rejects the direct realist view that sense perception estab-
lishes the true nature of external objects. Next he rejects the view that the
existence of external objects is known by inference. Then he shows that we
know by reflective awareness that the true nature of external objects is that
they are included in the Absolute consciousness. He goes on to argue on the
basis of pre-reflective immediate perception (anubhava), scripture and logic
that self-awareness is the very life of the knowing subject’s consciousness. Then
he states that reflexive consciousness is foundational because it constitutes
the ideal types of knowable objects (jñeyam śuddham) and the form of the
knowing subject that contains them. Although consciousness is uniform, it is
diversified into acts and subjects of knowing. He then says that just as reflexive
awareness pertains to knower, it is also the very life of thoughts which may be
non-conceptual or conceptual in form.
IPK 1.5.1
The manifestation as external to consciousness of entities that are manifested in
present experience is possible only if they are internal to consciousness.
Abhinavagupta supplies the question to which the verse is an answer: Direct

experience (anubhava) is held to be the basis of memory and conceptualization.
If objects appear in direct experience as separate from the knowing subject, they
should appear like that in memory and conceptualization also. But they do
not. So memory and conceptualization cannot be the basis of memory and
conceptualization.
Abhinavagupta explains that the verse deals with the nature of direct experi-
ence, which is a form of the power of knowing (jñāna-śakti) and of the sense
in which objects are experienced as external to consciousness. He says that the
clear and distinct perception of objects as distinct from the finite knowing
subject is rationally intelligible only if they are one with the unconditioned
subject that is pure consciousness and which makes them appear as separate.
The divine jñāna-śakti effects the manifestation as different from the finite
subject of what is internal to unconditioned consciousness.
IPK 1.5.2
If the object were not of the nature of the light of consciousness (prakāśa),
it would remain unilluminated as it was before it was known. The light of
consciousness is not different from the object. The light of consciousness is the
nature of the object.
Abhinavagupta says: Objects have to be illuminated by an external source. If

they revealed themselves, every object would always be apparent to everyone.
The same applies if we understand cognition as illuminating objects external
to consciousness that were previously ‘in darkness’. This is Kumārila’s view
that consciousness introduces a new feature, luminosity, into objects. It would
be difficult to explain in these circumstances why the object does not appear
to everyone.
Without consciousness, nothing can be established and universal blindness
would follow. If objects are not constituted by prakāśa, they would be as
unmanifested at the rise of cognition as they were prior to that.
He rejects the dualist view that there are two categorically distinct
realms: material objects and consciousness. How could they be connected?
The problem does not arise if we accept that consciousness is the essence of
objects that are non-different from it.
If cognitions are unique in the case of each object, they could not be
synthesized because each discrete cognition would be confined to itself.
This is avoided if consciousness is unified awareness of which individual
cognitions, and the different conscious operations are modes.
If consciousness that is totally separate from objects is the illuminator
of objects we encounter another problem:
IPK 1.5.3
If the light (prakāśa) were intrinsically undifferentiated and different from objects,
objective reality would be confused. The object that is illuminated must itself be
of the nature of the light of consciousness because that which does not have that
nature cannot be established.
Utpaladeva explains: undifferentiated light that is different from objects would

illuminate every object equally. So there would be no basis for the specific
discrimination of individual objects.
The Absolute Idealist’s argumentative strategy for there being a single all-
encompassing light of consciousness that projects the variety of manifestations
has three stages:
The rejection of direct realism about objects perceived as occupying a mind-independent

domain.
The rejection of the Representationalist view, according to which the existence of
mind-independent domain is inferred.
The rejection of forms of subjective idealism that drift into solipsism.
Utpaladeva and Abhinavagupta hold that we can explain mental variety and
complexity only if it derives from a single conscious source. It is the Sautrāntikas
who think that the mind-independent realm of unique particulars must be
inferred as the cause of such variety, and to a statement of their outlook we
now turn:
IPK 1.5.4
If the light of consciousness is undifferentiated, it cannot cause a diverse and
complex manifestation. Because such a manifestation is inexplicable in these
terms, we must infer external objects as its cause.
IPK 1.5.5
[An argument against a Vijñana-vada Buddhist idealist]
A diverse revival of karmic residues cannot be the cause [of experiential variety]. In
that case there would be the question of what causes the variety in such revival.
Abhinavagupta summarizes the Representationalist position: the causes of

successive variety in intrinsically undifferentiated consciousness are the
reflections of external forms that correspond to the reflections. The external
is to be inferred but we still call it perceptible.
The reasoning is: awareness is intrinsically undifferentiated: its true form is
just light. Undifferentiated light cannot be the cause of different representations.
This leads us to infer an external domain separate from cognition that consists
of many different forms which successively cast reflections of their nature in
consciousness. The reflections are similar to what is reflected and there is a sort
of correspondence between them [like that between a map and a territory].
The Sautrāntika contends that the idealist’s explanation of experiential vari-
ety as deriving from the re-awakening of sub-conscious latent traces (vāsanā)
of prior perceptions does not make sense. We know that such re-awakening is
responsible for memory, but here we are looking for the cause of variety in
present experiences. Let us provisionally accept the theory that the latent
traces are powers enabling cognitions to produce ideas of objects. Their re-
awakening means their fitness to produce their own effects. In this way arises
the variety of ideas. The problem here is that although the representations
in our minds do not strictly mirror reality, still their causes must be real if they
are genuinely productive. If the traces are the causes of ideas, they must be
both different from consciousness and objectively real. So this theory is a ver-
sion of realism about the external disguised under a different name. Let there
be a variety of subconscious traces. In that case, given the idealist view that
there are no entities, space or time distinct from consciousness that could be
the cause of variety by activating the traces, if consciousness is uniform variety
would appear simultaneously. It is false that other cognitions occurring in a
stream are the causes of the awakening of diverse ideas because if all mental
states (pleasures and pains, cognitions of objects and awareness of places and
times) are just awareness and awareness is essentially just light, given the
indivisibility of essence, there would be no differentiation in awareness.
With respect to other mental states constituting other knowing subjects
(which the Buddhist calls ‘streams’) there is the same lack of individuation.
If we restrict ourselves to the point of view of subjective awareness we cannot
form an objective conception of there being more than one such stream. We
can only conceive of thoughts happening for that one stream. The Sautrāntika’s
conclusion is that we cannot account for experiential variety by appealing to a
sub-conscious store of latent traces. So it is established that if consciousness is
undifferentiated pure light, and so cannot cause different ideas, it is necessary
to infer an external realm.
Utpaladeva now states his own view:
IPK 1.5.6
That may be the case. But why posit the external on the grounds that we cannot
explain things otherwise, when all everyday activities can be explained if all things
are manifestations of the single divine consciousness?
Abhinavagupta comments that there is no need to posit an external domain

since all worldly life can be explained in terms of Ideas (ābhāsa). It is imposs-
ible to establish the existence of things independent of consciousness. We
cannot make sense of the notion of material substance. The Nyāya theory that
wholes inhere in their parts is invalid because there is no proof of inherence.
No sense can be made of the Buddhist theories about the composition of
macroscopic entities out of atoms.
The theory of the ābhāsas (Ideas) is crucial to Abhinavagupta’s metaphysic.
They are the primary realities that the trans-individual consciousness manifests.
They are objective thought-contents that can be grasped by everyone. Ābhāsas
are universal properties. They are instantiated as objects, and the states of
affairs arising from those objects’ interactions, by their mutual delimitation
and by their connection with space and time. The ābhāsas space and time are
particularly important in that they impart particularity, and suppress the
notions of eternity and omnipresence which produce the form of universality.
Synthetic mental faculty (anusam.dhāna) identifies the individual entities that
are constituted by the Ideas. According to this account of general Ideas as
objective types, when anyone has an experience of blue, the idea ‘blue’ is the
same whatever the relevant mental connection (which may be seeing, imagin-
ing, remembering or delighting in). Likewise the idea ‘seeing’ is the same
whether it is connected with a pot or a cloth. A round blue pot is the coales-
cence of the ideas round, blue and pot.
IPK 1.5.7
Like a Yogin, just by the power of will, the Divinity whose nature is consciousness,
manifests all phenomena lying within him as external without needing any
independent substrative cause.
IPK 1.5.8–9
We an only use inference if what is to be established has been perceived somewhere
before. The sensory faculties are only inferred in very general terms as causes.
Objects that are totally external to consciousness are never manifested to con-
sciousness at all. Thus their existence cannot be established through inference.
We can infer the presence of fire from the presence of smoke because we are
familiar with smoke, fire and their relationship. But inference cannot operate
when one of its terms is totally unknown (especially in this case where that
which is to be established is by definition noumenal and outside the range of
our cognitive capacities). This is not inference but pure speculation. It looks
like Utpaladeva may be going too far here. We can infer the existence of the
sensory-faculties from the occurrence of perceptions, although, ex hypothesis,
we never perceive those faculties. What he is in fact saying is that we do
not infer the actual natures of the sense-faculties but only their generic
characteristic of being something that has a causal function. So we are not
inferring and understanding concrete realities but only the abstract concept of
causal efficacy. Thus we have not left the sphere of thinking and entered the
territory of external reality as understood by the Representationalist.
IPK 1.5.10
It is true that there is manifestation of beings that already exist within the Lord.
Otherwise the act of reflexive awareness (āmarśa) that is deliberate willing (icchā)
would not occur [they would not appear unless the Divinity knew them and
desired that they be manifest].
IPK 1.5.11
Reflexive awareness (vimarśa) is the essence of manifestation by consciousness.
Otherwise the light of consciousness (prakāśa) although tinged by the objects
would be lifeless like a crystal.
[This is a response to the view that understands consciousness on the model of a

searchlight.]
IPK 1.5.12
That is why one’s real identity is consciousness, meaning the acts of awareness
and the state of being the agent of conscious acts. By that one is distinguished
from the insentient.
IPK 1.5.13
The act of awareness is reflexive, the repository of all meanings, and spontaneously
arising. This is real freedom, the sovereignty of one’s ultimate identity.
IPK 1.5.14
It is this that is the vibrancy of consciousness, unconditioned Being unlimited by
place or time. This reality is expressed as the essence and the heart of the
Divinity.
IPK 1.5.15
By virtue of this he makes himself the objects of awareness. But the field of cogni-
tion does not subsist independently of him. If he depended on knowable objects
independent of himself, his freedom would cease.
IPK 1.6.1
The reflexive awareness ‘I’, whose nature is the light of consciousness, although
expressed by a word is not a concept (vikalpa) because a concept is an act
of mental discrimination that presupposes the possibility of affirmation and
exclusion, and this awareness has no opposites.
IPK 1.8.1–2
Sometime the Ideas (ābhāsa) are grasped in present sensory experience but
at other times they do not depend upon present experience, as in the cases
of a blind person or in darkness. But there is no difference in the reality of
the Ideas of objects featuring in thoughts, whether they concern past, present
or future.
Abhinavagupta comments: When we say ‘I see this blue thing’ or ‘I imagine

it’ or ‘I remember it’ or ‘I make it’, the Idea ‘Blue’ is essentially uniform.
The same applies to the Idea ‘Seeing’ when ‘I see’ relates to something
yellow. The Ideas are joined or separated by the creative divine autonomy.
In this way we can make sense of the variety of everyday life in past, present
and future.
IPK 1.8.3–4
Even when feelings like pleasure and their occasions are real, and their manifesta-
tions are real conscious states, if they belong to the past their external conditions
are not given. Still, if feelings are intensely reproduced by imagination then they
are felt by the subject as if the past object were present since he experiences the
feeling so vividly.
IPK 1.8.5
Externality is not a genuine property of the Ideas about realities and non-beings.
Being experienced as external is not the essence of the Ideas. The Ideas, which are
internal, always exist [whether some finite subject thinks them or not].
IPK 1.8.7
The Ideas, in so far as they are of the nature of consciousness, always exist within
[the trans-individual consciousness]. Given that their manifestations as external
are due to the power of māyā, they are experienced as external.
IPK 1.8.9
Owing to the will of the Lord, mental representations and feelings of pleasure are
manifested as if relating to what is external to consciousness.
IPK 1.8.10
Without the unification of cognitions, there would be no worldly life. The unification
of cognitions is based on the unity of trans-individual consciousness. There is one
knowing subject common to all [called the Supreme Self].
IPK 1.8.11
It is he only that is the Divinity by virtue of his constant self-awareness and
representation of things to himself. Reflexivity (vimarśa) is the pure knowledge
and action of the deity.
IPK 2.2.1
The concepts (‘buddhayah.’, which is glossed as ‘satya-ābhāsās’) of action,
relation, universal property, individual substance, space and time which apply
in the sphere of unity and multiplicity, are considered true because of their
permanence and utility.
The theory of Ideas recognizes the categories and the pramān.a framework
that are recognized by the realist Nyāya-Vaiśes.ika school. Utpaladeva and
Abhinavagupta accept the validity of their conceptual scheme, as long as it is
construed within the overall framework of Absolute Idealism. This contrasts

with the argument of an Absolute Idealist such as F. H. Bradley who says that
ordinary concepts such as space, time, causation and personal identity, are
ultimately incoherent and do not apply to reality as it really is. This is because
we cannot identify them individually due to their relational nature. The con-
tradictions inherent in whatever is posited by ordinary thought show that they
are mere appearances and not realities. But the Theory of Ideas recognizes our
everyday concepts as valid in that they are contained in and projected by the
Divine Mind. They are not merely human conceptual constructs, useful in
helping us to find our way around the world. Unity and multiplicity are com-
patible because Ideas are joined and separated within the all-encompassing
divine consciousness.
Abhinavagupta composed a succinct overview of this theology called the
Bodhapañcadaśikā or Fifteen Verses on Consciousness.
That single principle which is both within and external, whose form is radiance
unlimited in light and darkness, that is the Divinity that is the essence of all beings.
Its sovereign śakti produces entities. [1–2]
The śakti does not desire to be different from its possessor. The shared nature of
the two is permanent, like that of fire and burning. [3]
This is the deity Bhairava who sustains the cosmos because by his śakti he has
made everything appear as reflected in his own mirror. [4]
The śakti is the transcendent Goddess who delights in contemplating his essence.
Her perfect state neither increases nor diminishes in relation to finite beings. [5]
The Divine Omnipotence eternally delighting in play with the Goddess simultane-
ously dispenses the emanations and reabsorptions of the worlds. [6]
His impossibly difficult unsurpassable activity is total freedom and sovereignty

whose nature is consciousness. [7]
The distinctive feature of what is insentient is its being a limited manifestation.

Consciousness is other than the insentient by which it is not limited. [8]
The emanations and reabsorptions of the worlds are established as fissions of the
autonomous innate śakti. [9]
In them there is infinite variety of spheres of experience and their regions, as well
as pleasant and unpleasant experiences. [10]
When the unconditioned divine freedom is not understood, there is cycle of birth
and death that terrifies the unenlightened. This too is his power. [11]
Divine grace is accessible for one who has gone to a teacher or from scripture.
[12]
God-given understanding of the truth is freedom from birth and death, and it is
perfection for the enlightened ones. This is known as being liberated while still
alive. [13]
Both bondage and freedom proceed from God. They are neither really different
from each other, nor different from God. [14]
In this way Bhairava exercising his three-fold śakti of will, action and knowing is
the true nature of all beings. [15]
Further reading
For the Siddhānta see Sanderson (1992).
Filliozat (1994) translates Sadyojyoti’s commentary on the Svayambhuvāgama.
For the context of Rāmakan.t.ha’s thought about the self, Watson (2006) is valuable.
Goodall (1998) provides a text and lucid translation of the first six chapters of Rāmakan.t.ha’s Kiran.a
Tantra commentary. The Mataṅga commentary is in Bhatt (1977). It has not been translated. Nor
has Rāmakan.t.ha’s Nareśvaraparīks.ā-prakāśa. A (rather doubtful) text is Shastri 1926. Watson
(2006) presents a critical version of much of Book 1.
Elaborate arguments for the existence of God were also formulated by Naiyāyikas beginning with
Jayanta (850–900 A.D.) They are mentioned in ‘Much Ado about Religion’ (Dezső (2005)). By
Jayanta’s time, personalist theism had long surpassed automatic ritualism as the dominant type
of religiosity in many parts of the sub-continent. A useful source here is Krasser (2002) that
discusses both Buddhist and Nyāya thought stemming from Dharmakīrti’s arguments against
an omnipotent and omniscient creator. Also very interesting is Frank Clooney Hindu God,
Christian God.
For the traditions that we have labelled Śākta, see Sanderson (1985 and 1992) to begin with and move
on to Sanderson (1990 and 1995 ‘Meaning in Tantric Ritual’). Professor Sanderson’s preeminence
in this field is evident from his ‘The Śaiva Exegesis of Kashmir’ (2007).
Ks.emarāja’s Pratyabhijñāhr.dayam is in Singh (1982) and his Śiva Sūtra-vimars.inī in Singh (1991).
For Utpaladeva’s Īśvarapratyabhijñākārikā with the author’s own commentary, see the text and
translation in Torella (2002). Abhinavagupta’ s Vimarśinī commentary is in Subramania Iyer
and Pandey (1986).
The Bodhapañcadaśika is in Shastri (1947).
Chapter III of Kahrs (1998) deals with Kashmiri scriptural exegesis. Padoux (1990) is a classic study of
the powers attributed to words in the Hindu traditions.

investigation
1. How does the idealism of Utpaladeva and Abhinavagupta differ from the Advaita-
Vedānta outlook? Can their views be characterized as theistic?
2. Are the Śaivas’ arguments against Buddhism convincing ones?
Bibliography
Acharya, D. (ed.) (2006), Vācaspatimiśra’s Tattvasamīks.ā: The Earliest Commentary on Man.d.anamiśra’s

Brahmasiddhi, Stuttgart: Franz Steiner Verlag.
Athalye, Y. V. (ed. and trans.) (2003), Tarkasam.graha of Annam.bhat.t.a, Pune: Bhandarkar Research
Institute.
Bartley, C. J. (2002), The Theology of Rāmānuja, London: RoutledgeCurzon.
Basham, A. L. (1967), The Wonder that was India: A Survey of the Culture of the Indian Sub-Continent
Before the Coming of the Muslims, 3rd edn, London: Sidgwick and Jackson.
Bechert, H. and Gombrich, R. (1984), The World of Buddhism, London: Thames and Hudson.
Bhatt, N. R. (ed.) (1977), Mataṅgapārameśvarāgama (Vidyāpāda) avec le commentaire be Bhat.t.a
Rāmakan.t.ha, Pondichéry: Institut Français d’Indologie.
Bhattacharya, K. (1998), The Dialectical Method of Nāgārjuna: Vigrahavyāvartanī, Delhi: Motilal
Banarsidass.
Bronkhorst, J. (2000), The Two Traditions of Meditation in Ancient India, Delhi: Motilal Banarsidass.
Buitenen, J. A. B. van (1953), Rāmānuja on the Bhagavad Gītā, The Hague: Smits.
Buitenen, J. A. B. van (1956), Rāmānuja’s Vedārthasam.graha, Poona: Deccan College.
Burton, D. (1999), Emptiness Appraised, Richmond: Curzon Press.
Carman, J. B. (1974), The Theology of Rāmānuja, New Haven: Yale University Press.
Chakrabarti, K. (1999), Indian Philosophy of Mind: The Nyāya Dualist Tradition, Albany: SUNY Press.
Clooney, Francis X. (2001), Hindu God, Christian God, New York: Oxford University Press.
Collins, S. (1982), Selfless Persons, Cambridge: Cambridge University Press.
Conze, E. (1959), Buddhist Scriptures, Harmondsworth: Penguin.
Crosby, K. and Skilton, A. (trans.) (1996), Śāntideva: The Bodhicāryāvatāra, Oxford: Oxford University
Press.
Danielson, H. (ed. and trans.) (1980), Ādiśes.a: The Essence of Supreme Truth (Paramārthasāra), Leiden:
Brill.
Dezső, C. (ed. and trans.) (2005), Much Ado about Religion by Bhat.t.a Jayanta, New York: New York
University Press.
Doniger, W. (trans.) (2005), The Rig Veda, London: Penguin Classics.
Duerlinger, J. (2003), Indian Buddhist Theories of Persons: Vasubandhu’s Refutation of the Theory of a
Self, London: RoutledgeCurzon.
236 Bibliography
Dummett, M. (2006), Thought and Reality, Oxford: Oxford University Press.

Dumont, L. (1980), ‘World Renunciation in Indian Religions’, in L. Dumont, Homo Hierarchicus,
Chicago: Chicago University Press.
Dundas, P. (1992), The Jains, London: Routledge.
Dunne, J. D. (2004), Foundations of Dharmakīrti’s Philosophy, Boston, Wisdom Publications.
Eltschinger, V. (2007), Penser l’autorité des Écritures, Wien: Österreichischen Akademie der
Wissenschaften.
Filliozat, Pierre-Sylvain, (ed. and trans.) (1994), The Tantra of Svayam.bhū: Vidyāpāda with the
Commentary of Sadyojyoti, Delhi: Motilal Banarsidass.
Flood, G. (1996), Introduction to Hinduism, Cambridge: Cambridge University Press.
Franco, E. (1994), Perception, Knowledge and Disbelief, Delhi: Motilal Banarsidass.
Frauwallner, E. (1995), Studies in the Abhidharma Literature and the Origins of the Buddhist
Philosophical Systems, Albany: SUNY Press.
Ganeri, J. (1999), Semantic Powers, Oxford: Clarendon Press.
Ganeri, J. (ed.) (2001), Indian Logic: A Reader, London: Curzon.
Garfield, J. (trans.) (1995), The Fundamental Wisdom of the Middle Way, New York: Oxford University
Press.
Gerow, E. (trans.) (1990), The Jewel-Necklace of Argument, New Haven: American Oriental Society.
Gethin, R. (1998), Foundations of Buddhism, Oxford: Oxford University Press.
Gethin, R. (trans.) (2008), Sayings of the Buddha, Oxford: Oxford University Press.
Gombrich, R. (1996), How Buddhism Began, London: Athlone.
Goodall, D. (ed.) (1998), Bhat.t.a Rāmakan.t.ha’s Commentary on the Kiran.atantra, Pondichéry: Institut
Français de Pondichéry & École Française d’Extrême-Orient.
Goudriaan, T. (ed.) (1992), Ritual and Speculation in Early Tantrism: Studies in Honour of André
Padoux, Albany: SUNY Press.
Granoff, P. (1978), Philosophy and Argument in Late Vedānta: Śrī Hars.a’s Khan.d.anakhan.d.akhādya,
Dordrecht: Reidel.
Gupta, S. (trans.) (2000), Laks.mī Tantra, Delhi: Motilal Banarsidass.
Hahn, M. (ed.) (1982), Nāgārjuna’s Ratnāvalī, Bonn: Indica et Tibetica Verlag.
Halbfass, W. (1991), Tradition and Reflection, Albany: SUNY Press.
Halbfass, W. (ed.) (1991), Philology and Confrontation: Paul Hacker on Traditional and Modern Vedanta,
Albany: SUNY Press.
Halbfass, W. (1992), On Being and What There Is: Classical Vaiśes.ika and the History of Indian Ontology,
Albany: SUNY Press.
Hara, M. and Wright, J. C. (eds) (1996), John Brough: Collected Papers, London: SOAS.
Hardy, F. (1983), Viraha-Bhakti, Delhi: Oxford University Press.
Hattori, M. (1968), Dignaga, On Perception, Cambridge, Mass.: Harvard University Press.
Hayes, R. (1988), Dignaga on the Interpretation of Signs, Dordrecht: Kluwer.
Ingalls, D. H. H. (1951), Materials for the Study of Navya-Nyaya Logic, Cambridge, Mass.: Harvard
University Press.
Bibliography 237
Jaini, P. S. (1979), The Jaina Path of Purification, Berkeley: University of California Press.
Jaini, P. S. (2000), Collected Papers on Buddhist Studies, Delhi: Motilal Banarsidass.
Jamison, S. and Witzel, M. (2003), ‘Vedic Hinduism’ in Arvind Sharma (ed.), The Study of Hinduism,
Columbia: University of South Carolina Press.
Jayatilleke, K. N. (1980), Early Buddhist Theory of Knowledge, Delhi: Motilal Banarsidass.
Jha, G. (trans.) (1984), The Nyāya-Sūtras of Gautama with the Bhās.ya of Vātsyāyana and the Vārttika of
Uddyotakara, IV Vols, Delhi: Motilal Banarsidass.
Johnson, W. J. (1994), The Bhagavad Gītā, Oxford: Oxford University Press.
Kahrs, Eivind. (1998), Indian Semantic Analysis: Cambridge: Cambridge University Press.
Kajiyama, Y. (1998), An Introduction to Buddhist Philosophy: An Annotated Translation of the Tarkabhās.a
of Moks.ākāragupta, Wien: Universität Wien.
Karmarkar, R. D. (ed. and trans.) (1953), Gaud.apāda-Kārika, Poona: Bhandarkar Oriental Research
Institute.
King, R. (1995), Early Advaita Vedānta and Buddhism, Albany: SUNY Press.
Krasser, H. (2002), Śaṅkaranandanas Īśvarāpākaran.asaṅks.epa, 2 Vols, Wien: Österreichischen
Akademie der Wissenschaften.
Larson, G. J. (1979), Classical Sam.khya, Delhi: Motilal Banarsidass.
Larson, G. J. and Bhattacharya, R. S. (eds) (1987), Encyclopedia of Indian Philosophies: Volume IV
Samkhya, Delhi: Motilal Banarsidass.
Lindtner, C. (1982), Nagarjuniana, Copenhagen: Akademisk Forlag.
Lipner, J. (1986), The Face of Truth, Basingstoke: Macmillan.
Matilal, B. K. (1986), Perception: An Essay on Classical Indian Theories of Knowledge, Oxford:
Clarendon Press.
Matilal, B. K. (1990), The Word and the World, Delhi: Oxford University Press.
Matilal, B. K. (2002), Mind, Language and World, Delhi: Oxford University Press.
Mayeda, S. (1979), A Thousand Teachings, Tokyo: Tokyo University Press.
Mesquita, R. (2000), Madhva’s Unknown Literary Sources: Some Observations, Delhi: Aditya
Prakashan.
Mookerjee, S. (1975), The Buddhist Philosophy of Universal Flux, Delhi: Motilal Banarsidass.
Oberhammer. G. (1998), Materielen zur Geschichte der Rāmānuja-Schule, IV, Der ‘Innere Lenker’
(Antar-yāmī), Wien: Österreichischen Akademie der Wissenschaften.
Oetke, C. (1994), Studies in the Doctrine of Trairupya, Wien: Universität Wien.
Olivelle, P. (1986), Renunciation in Hinduism: A Medieval Debate: Vienna: Universität Wien.
Olivelle, P. (1993), The Āśrama System: The History and Hermeneutics of a Religious Institution,
New York: Oxford University Press.
Olivelle, P. (ed. and trans.) (1998), The Early Upanishads: Annotated Text and Translation, New York:
Oxford University Press.
Olivelle, P. (1999), Dharmasūtras; the Law Codes of Apastamba, Gautama, Baudhayana and Vasistha,
Translated from the Original Sanskrit, Oxford: Oxford University Press.
Olivelle, P. (ed. and trans.) (2005), Manu’s Code of Law, Oxford: Oxford University Press.
238 Bibliography
Padoux, A. (1990), Vāc: The Concept of the Word in Selected Hindu Tantras, Translated by Jacques
Gontier, Albany: SUNY Press.
Pandey, K. C. (trans.) (1986), Īśvara-pratyabhijñā-vimarśin.ī of Abhinavagupta: Doctrine of Divine
Recognition, Delhi: Motilal Banarsidass.
Pandeya, R. (ed.) (1989), The Pramān.avārttikam of Ācārya Dharmakīrti, Delhi: Motilal Banarsidass.
Pandeya, R. (ed.) (1999), Madhyānta-Vibhāga-śāstra, Delhi: Motilal Banarsidass.
Potter, K. H. (ed.) (1977), Encyclopedia Of Indian Philosophies: Vol. II, Indian Metaphysics and Episte-
mology: The Tradition of Nyaya-Vaisheshika up to Gangesha, Delhi: Motilal Banarsidass.
Potter, K. H. (ed.) (1981), Encyclopedia Of Indian Philosophies: Vol. III, Advaita Vedanta up to Śaṅkara
and His Pupils, Delhi: Motilal Banarsidass.
Potter, K. H. (1996), Encyclopedia Of Indian Philosophies: Vol. VII, Abhidharma Buddhism to 150 A.D.,
Delhi: Motilal Banarsidass.
Potter, K. H. and Bhattacharya, S. (eds) (1993), Encyclopedia Of Indian Philosophies: Vol. VI, Indian
Philosophical Analysis; Nyāya Vaiśes.ika from Gangeśa to Raghunātha Śiroman.i, Delhi: Motilal
Banarsidass.
Pradhan, P. (1967), Abhidharmakośabhas.yam of Vasubandhu, Patna: K.P. Jayaswal Research Institute.
Pruden, L. (trans.) (1988), Abhidharmakośabhas.yam, Berkeley: Asian Humanities Press.
Raghavachar, S. S. (ed. and trans.) (1959), Śrīmad-Vis.n.u-Tattva-Vinirn.aya of Śrī Madhvācārya,
Mangalore: Sri Ramakishna Ashrama.
Rahula, W. (1969), What the Buddha Taught, London: Gordon Fraser.
Ram-Prasad, C. (2002), Advaita Epistemology and Metaphysics, London: RoutledgeCurzon.
Rocher, L. (ed.) (1988), Studies in Indian Literature and Philosophy: Collected Articles of J. A. B. van
Buitenen, Delhi: Motilal Banarsidass.
Sadhale, G. S. (ed.) (2000), The Bhagavad Gītā with Eleven Commentaries, Delhi: Parimal Publications.
Sanderson, Alexis (1985), ‘Purity and Power among the Brahmins of Kashmir’, in M.Carrithers et al.
(eds), The Category of the Person, Cambridge: Cambridge University Press, pp. 190–216.
Sanderson, Alexis (1990), ‘Śaivism and the Tantric Traditions’, in F. Hardy (ed.), The World’s Religions:
The Religions of Asia, London: Routledge, pp. 128–172.
Sanderson, Alexis (1992), ‘The Doctrine of the Mālinīvijayottaratantra’, in T. Goudriaan (ed.), Ritual
and Speculation in Early Tantra: Studies in Honor of André Padoux, Albany: SUNY Press,
pp. 281–312.
Sanderson, Alexis (1995), ‘Meaning in Tantric Ritual’, in Essais sur le rituel III, Louvain, Paris: Peeters,
pp. 15–95.
Sanderson, Alexis (1995), ‘The Sarvāstivāda and its Critics’, in Buddhism in the Year 2000: International
Conference Proceedings, Bangkok and Los Angeles: Dhammakāya Foundation, pp. 37–49.
Sanderson, Alexis (2001), ‘History through Textual Criticism’, in François Grimal (ed.), Les sources et le
Temps. Sources and Time. A Colloquium, Pondichéry: Institut Français de Pondichéry & École
Française d’Extrême-Orient, pp. 1–47.
Sanderson, Alexis (2007), ‘The Śaiva Exegesis of Kashmir’, in D. Goodall and André Padoux (eds),
Mélanges Tantriques à la mémoire d’Hélène Brunner, Pondichéry: Institut Français de Pondichéry &
École Française d’Extrême-Orient, pp. 231–442.
Bibliography 239
Sarma, D. (2003), An Introduction to Mādhva Vedānta, Aldershot: Ashgate.
Scharf, P. (1996), The Denotation of Generic Terms in Ancient Indian Philosophy: Grammar, Nyāya, and
Mīmām.sā, Philadelphia: American Philosophical Society.
Schrader, F. Otto (1916), Introduction to the Pāñcarātra and the Ahirbudhnya Sam.hitā, Madras, Adyar
Library.
Shastri, J. L. (ed.) (1980), Brahmasūtra- Śaṅkarabhās.yam, Delhi: Motilal Banarsidass.
Shastri, J. Z. (1947), Bodhapañcadaśika of Abhinavagupta, Srinagar: Kashmir Series of Texts and
Studies.
Shastri, M. K. (ed.) (1926), The Nareshvarapariksha of Sajyojyoti with the Commentary by Ramakantha,
Srinagar, Kashmir Series of Texts and Studies.
Siderits, M. (1991), Indian Philosophy of Language: Selected Issues, Dordrecht: Kluwer.
Siderits, M. (2003), Personal Identity and Buddhist Philosophy: Empty Persons, Aldershot: Ashgate.
Singh, J. (1982), Pratyabhijñāhr.dayam, Delhi: Motilal Banarsidass.
Singh, J. (1991), Śiva Sūtras, Delhi: Motilal Banarsidass.
Stcherbatsky, Th. (1970), Madhyānta-Vibhanga, Osnabruck: Biblio Verlag.
Streng, F. J. (1967), Emptiness: A Study in Religious Meaning, New York: Abingdon Press.
Subramania Iyer, K. A. (ed.) (1964), Vākyapadīya of Bhartr.hari (Kān.da I), Poona: Deccan College.
Subramania Iyer, K. A. and Pandey, K. C. (eds) (1986), Īśvarapratyabhijñā-vimarśinī of Abhinavagupta,
3 Vols, Delhi: Motilal Banarsidass.
Suryanarayana Sastri, S. S. (ed. and trans.) (1971), Vedāntaparibhās.a of Dharmarāja Adhvarin, Madras:
The Adyar Library and Research Centre.
Suryanarayana Sastri, S. S. and Barnett, L. D. (ed. and trans.) (2003), The Paramārthasāra of Ādi Śes.a
and the Paramārthasāra of Abhinavagupta, Fremont: Asian Humanities Press.
Suthren Hirst, J. G. (2005), Śam.kara’s Advaita Vedānta, Abingdon: RoutledgeCurzon.
Svāmī Ādidevānanda, (ed. and trans.) (1949) Yatīndramatadīpikā, Mylapore: Sri Ramakrishna Math.
Svami Dvarikadasa Shastri, (ed.) (1994), Nyāyabinduprakaran.am, Varanasi: Bauddha Bharati.
Swāmi Jagadānanda (1941), Upadeśa SāhasrI of Śrī Śam.karācārya, Mylapore: Sri Ramakrishna Math.
Taber, J. (2005), A Hindu Critique of Buddhist Epistemology: Kumārila on Perception: The ‘Determination
of Perception’ Chapter of Kumārila Bhat.t.a’s Ślokavārttika Commentary, London: RoutledgeCurzon.
Tachikawa, M. (1981), The Structure of the World in Udayana’s Realism, Dordrecht: Reidel.
Thibaut, G. (trans.) (1904), The Vedanta-Sutras with the Commentary of Ramanuja, Oxford: Clarendon
Press.
Thibaut, G. (trans.) (1904), The Vedanta-Sutras with the Commentary of Shankaracarya, 2 Vols, Oxford:
Clarendon Press.
Thrasher, A. W. (1993), The Advaita Vedanta of Brahma-siddhi, Delhi: Motilal Banarsidass.
Tillemans, T. J. F. (1999), Scripture, Logic and Language, Boston: Wisdom Publications.
Tola, F. and Dragonetti, C. (2004) Being as Consciousness, Delhi: Motilal Banarsidass.
Torella, R. (ed. and trans.) (2002), The Īśvarapratyabhijñākārikā of Utpaladeva, Delhi: Motilal
Banarsidass.
Watson, Alex (2006), The Self ’s Awareness of Itself: Bhat.t.a Rāmakan.t.ha’s Arguments against the Buddhist
Doctrine of No-Self, Wien: Publications of the De Nobili Research Library.
240 Bibliography
Whicher, I. (1998), The Integrity of the Yoga Darśana, Albany: SUNY Press.
Williams, P. (1989), Mahayana Buddhism, London: Routledge.
Williams, P. (1998), Altruism and Reality, London: Curzon Press.
Wood, T. E. (1991), Mind Only: A Philosophical and Doctrinal Analysis of the Vijñānavāda, Honolulu:
University of Hawaii Press.
Woods, J. H. (1927), The Yoga Sūtras, Cambridge, Mass.: Harvard University Press.
Wynne, A. (trans.) (2009), Mahābhārata Book Twelve. Peace. Volume Three. The Book of Liberation,
New York: New York University Press.
Index
ābhāsa-vāda (Theory of Ideas) 213, 220, apoha 41–2

228, 230, 231, 232 apūrva 132
Abhidharma Ch.3 passim, 35, 39, 56, 57, arthakriyā 46–7, 203
58, 59, 60, 63, 64, 71 arthāpatti 121, 124, 205
Abhidharmakośabhās.ya 27, 29, 33, 69 artha-prakāśa-buddhi 91, 106, 217
Abhidharma-pit.aka 26, 27 arthavāda 119, 135, 139, 151–4
abhihita-anvaya 122 asatkāryavāda 63, 85, 104–5
Abhinavagupta 210, 211, 212, 213, 214, atheism 14–15
215, 217, 222, 224, 225, 226, 227, 228, ātman 3, 4, 9–12, 13, 20, 22, 53, 74, 87
230, 231, 232 Advaita view 139, 141–2, 144, 145–6, 171
abhūta-parikalpa 76–81 Mīmām.sā view 128–31
‘absolute conception of reality’ 24, 56, 60 Nyāya view 94–8, 146, 176
adarśana-mātra 52 Pratyabhijňā view 220–1
adhyāsa 142–3, 147 Śaiva Siddhānta view 199–209
adhyavasāya 48, 86, 205 Viśis.t.ādvaita view 175–6
Advaita 5, 173–7, 186, 189, 190, 191, avidyā 139, 141, 142, 143, 147, 148, 150,
211, 213 163, 164, 165, 166, 171, 173, 180, 189,
agency (human) 128–32, 176–7, 196, 191–2, 212, 218–19
207–9 unexplainable 142, 150, 173, 189
aham.kāra 61, 84, 146, 155, 177, 199 avijñapti-dharma 32
ālambana 29, 42, 43–5, 75, 81 avinābhāva-sam.bandha 40, 52
Ālambana-parīks.ā 43–5
alaukika-pratyaks.a 101, 107 Being (sat) 1, 8, 14, 144, 162–3, 165
ālaya-vijñāna 71, 74, 77, 79, 161 Bhagavad Gītā 18, 134
Ālvārs 169–70 bhakti 5–6, 168, 170, 171, 186
analogia entis 135–6, 149–50, 179, 186 Bhartr.hari 126, 135, 140–1
anatta (no soul) 13–15, 20–3, 27, 33–4, 45 Bhāskara 136, 180
anumāna 40, 51–3, 111–13, 197–8 bheda 163–4, 174–5, 188–9
antaryāmin 10, 170 Bhedābheda tradition, 136–7, 171–2, 180
anubhava 99, 100, 201, 224 Bradley, F.H. 3, 232
anupalabdhi 53, 203 Brahman 5, 8–10, 11, 13, 135, 136, 137,
anusam.dhāna 218 139, 147–50, 151–5, 162–3, 165,
anuvyavasāya 100, 108 191–2, 211
anvita-abhidhāna 122 Brahma-vihāras 19
apaurus.eya 120, 187 Buddha , the 13–20, 30, 33
242 Index
buddhi (faculty) 4, 83, 84, 86, 87, 205 endurance vs perdurance 202
‘bundle of perceptions’ 22, 95, 130, 205 essential temporality 1, 13, 14, 158–9, 202,
203, 207, 208
caitta 28 ethics (Buddhist) 17–19, 33, 39
caste 2, 3, 5, 13, 18, 118, 143, 170, 195, 196, ‘event ontology’ 1, 91, 202, 208–9
210, 212, 218 ‘externalism’ 91, 99
causal efficacy see arthakriyā extrinsic validity of knowledge 108
cetanā (intention) 17, 18, 32
citta-viprayukta-dharma 28, 36, 37 ‘fivefold difference’ 185
cognitive errors 49, 73, 77, 80, 106–7, 174 flux, universal 28
consciousness 131, 138, 145, 162, 176–7, of sensation 39
201, 204–5, 214–16, 224 Four Noble Truths 16–17
psychological vs phenomenal 216–17 Frauwallner, E. 26
unity of 205, 231 freedom from natural causality 4, 87
‘Consciousness-only’ (Buddhist freedom from rebirth see moks.a
tradition) Ch. 6, 160–2, 200, 202, 211,
224, 227 Gaud.apāda 140
constituent of personality see skandha gnosis, liberation through 82, 87, 139, 210
‘construction of the unreal’ God
see abhūta-parikalpa inferential proof 113, 172–3, 197–8
linguistic reference to 180
devotional religion 168–9 revealed by scripture 134–5, 172, 181, 184b
dharma (natural right and order) 3–4, 8, grace, divine 168, 170, 186, 188, 195, 197,
118–20, 128, 132, 170, 196 210, 233
Dharmakīrti 38, 46–55, 144, 198, 200, grāhya-grāhaka-vibhāga see subject-object
201, 203 dichotomy
dharmas (basic elements) 14, 26, 27, 28, gun.a (constituent of prakr.ti) 83, 199
29, 30, 32, 33, 40, 57, 58, 59, 64, 75, 77 gun.a (quality) 39, 91, 98–9, 104, 207
Abhidharma classification 28, 36 gun.a-sam.drava 84, 207
differentiation (and individuality)
see bheda hallucinations see cognitive errors
Dignāga 38–45, 46, 47, 49, 50, 52, 54, 124, hetu 40, 52, 53
126, 127
dīks.ā 195, 196, 197, 210, 210 idealism 43, 49, 68, 69, 76–81, 107,
‘direct realism’ 31, 37, 90, 91, 100, 105, 213–14, 224, 227
106, 107, 109, 123–8, 131, 161, 169, absolute 210, 213, 220, 226, 232
174, 181, 185–6, 191 theistic 214
‘disjunctivism’ 107 inferential reasoning see anumāna
dravya 39, 85, 91, 93–8, 104, 175, 177, 179, initiation see dīks.ā
180, 207 intention see cetanā
dravya-sat 28, 59 interdependent origination
dr.śya-anupalabdhi 53 see pratītya-samutpāda
duh.kha 16, 22 intrinsic nature see svabhāva
efficacy see kāritram James, William 21–2

eightfold path (Buddhist) 17–18 jāti 39, 91, 100–3, 115, 123
emanation 136, 137, 171, 180 Jñānaprasthāna 27
emptiness see śūnyatā jñātatā 124, 225
Index 243
kaivalya 87, 88 nirākāra-vāda 31, 124, 132, 174
kalpanā (mental construction) 39, 45, 47, nirvān.a 17, 20, 37, 59, 62
48, 124, 140, 201, 204 Nyāyabindu 47–8
kāma 18 Nyāya-Vaiśes.ika 39, 41, 50, 65, 146, 176,
Kant, Immanuel 129 207, 232
kārakas 208, 209
kāritram 30 ‘objective support’ see ālambana
karma (fruit-yielding action) 2–3, 13, 18, 32, ‘other minds’ 53
61, 69, 86, 88, 94, 128, 157, 166, 168,
176, 195, 196, 197, 198, 210, 218, 219 padārtha 39, 92–105, 232
karman (movement) 39 paks.a 40
kārya-hetu 52 Pāli Canon 15, 22, 35
kinds see jāti Pañcarātra 170–1
kleśa 18, 33, 45, 71, 140 paratantra-svabhāva 75, 80–1
Krama 217, 220–2 Parfit, Derek 129
ks.an.ikatva 14, 35, 36–7, 91, 129, 158 parikalpita-svabhāva 75, 80–1
Ks.emarāja 214 parin.āma 84
Kumārila 120, 121, 122, 123–31, 164 parinis.panna-svabhāva 75, 80–1
permanence 46, 47, 50–1, 203
laks.an.ā 181–2 personal identity 13–14, 20–2
personality see aham.kāra
Madhva 136, 197 phala-āks.epa-śakti 29
Mādhyamakas 200, 202 prajñapti-sat 28, 31, 32, 33, 59, 81
Mahābhārata 5 prakāśa 213, 216, 217, 218, 219, 225,
Mahāvibhās.a 27 226, 229
mala 195, 196, 218 prakr.ti 46, 50, 53, 83, 87, 199
manas 4, 31, 84, 95, 98, 100, 146, 199 pramān.a 38–40, 47–8, 65–6, 92, 100,
Man.d.ana Miśra 125, 140, 141, 151, 152, 105–14, 143, 200
162–6 no pramān.a for featureless reality 174–5
Mātr.ceta 33 prāpti 28, 32
māyā 197, 198, 199, 218, 219 pratisam.dhāna 95, 96, 159
means of knowing see pramān.a pratītya-samutpāda 16–17, 57, 58, 63,
meditation 17, 19, 20, 68 65, 91
memory 95–6, 106, 186, 201, 206, 223 pratyaks.a 31, 39, 47–8, 108–10, 164
experiential 95, 159, 223–4 nirvikalpaka 109, 125, 126, 127, 164,
mental construction see kalpanā 174–5
mental synthesis (requires persisting savikalpaka 109, 126, 127, 174–5
identity) 94, 95–6, 129, 130, 201, Pudgalavāda 33–4
206, 207, 223 purity 3, 196, 212, 219
‘Middle Way’ 15, 17 purus.a 83, 88
Mīmām.saka 50
moks.a 62, 98, 128, 136, 139, 151–6, 186, Questions of King Milanda 23
196, 212, 219, 221
‘momentariness’ see ks.an.ikatva Rāmakan.t.ha 194–209
Rāmānuja 135, 136, 169–83, 197
Nāgārjuna 17 ‘reductionism’ (Buddhist) 95, 157
nairātmya 14, 21, 72, 91, 202, 212 reflexivity of awareness 42, 47, 94, 100, 124,
see ‘streams of experiences’ 177, 201, 204, 217, 221–2, 224, 229
244 Index
renunciation 5, 15, 82, 87, 140 ‘storehouse-consciousness’

‘representationalists’ see Sautrāntika see ālaya-vijñāna
ritual religion 2, 5, 13, 15, 19, Ch. 9 passim, ‘stream of experiences’ 14, 15, 16, 18, 20,
150–6, 168, 194–6, 209–10 21, 22, 23, 24, 32, 33, 37, 38, 39, 49,
51, 53, 70, 71, 72, 73, 74, 79, 94–7,
śabda 113–15, 121–3 129, 130, 159, 203, 206, 227–8
śabdārtha-sam.bandha 114, 121–2 subject-object dichotomy 43, 49, 71, 72,
sādhya 40, 52 76–81
sahopalambha-niyama 161, 200 ‘substance-ontology’ 1, 91, 202
sākāra-vāda 37 substances, individual see dravya
sāmānādhikaran.ya 181–2 substrative cause 85, 179, 187, 196, 197
sāmānya 39, 91 śūnyatā 57–8, 76, 199
sāmānya-laks.an.a (characteristics of svabhāva (intrinsic nature) 27–8, 35, 36,
conditioned phenomena) 28, 36, 40, 56, 57, 59, 75
39, 40 svabhāva-hetu 52
samavāya 85, 91, 104 svabhāva-pratibandha 52
samāveśa 219 svabhāva-śūnya 57–8, 65
Śam.kara 104, 140–62, 197, 202 svalaks.an.a (unique particulars) 27, 29, 36,
Sām.khya 50, 53 37, 39, 46, 47, 48, 69, 124, 200
sam.nyāsa see renunciation svasam.vitti 42, 47, 70
sam.sāra 3, 59, 87, 151 svatah. prāmān.ya (intrinsic authority of
sam.vr.ti-satya 29 Vedic meanings) 120, 187
sapaks.a 40 svātantrya (divine autonomy) 209,
sārūpya 37 216, 230
Sarvāstivāda 27, 28, 29, 33, 56, 69, 77
satkāryavāda 63, 84–5, 135, 179 taboo, violation of 212
Sautrāntikas 27, Ch. 4 passim, 69, 70, ‘tat tvam asi’ 139, 145, 181–2, 190
159–60, 200, 201, 202, 226, 227, ‘temporal parts’ 146, 202
228, 229 theism 168–9, 170, 185, 186, 214
ethical consequentiality 38 time, as exercise of efficacy 30
inferability of external reality 37, 69, 70, trairūpya 40
200, 226 Trika 211–12
perceptual representations 37, 43–4, 49,
69, 81 ‘Unanswered questions’ 23–4
scripture see śruti universal see jāti & sāmānya
self, the see ātman upādāna-kāran.a see substrative cause
self-awareness see reflexivity upādhi (Nyāya) 102–3
sense-perception see pratyaks.a upādhi (Vedāntic) 136, 144, 146,
‘series persons’ 129 147, 148
skandhas 21, 23, 26, 32, 33, 61, 129, Upanis.ads 8–12, 62
157, 208 Br.hadāran.yaka 9–11
Somānanda 210, 211, 213 Chāndogya 9, 11–12, 20
soul see ātman Kat.ha 4
Śrī Vais.n.ava tradition 169–72 Śvetāśvatara 6
śruti (testimony-human and Utpaladeva 210, 211, 212, 213,
scriptural) 38–9, 46, 50, 54, 119, 134 214, 217, 222, 223, 224, 226, 228,
Sthiramati 69, 76–81 229, 232
Index 245
Vaibhās.ika 27, 29, 30, 31, 35, 36, 37, 42, 70, vidhi (Vedic mandate) 119, 139, 150–6
156–7, 202 vijñapti (perception) 70, 71, 72, 74
vāsanā 70, 161 vijñapti-dharma 32
Vasubandhu 28, 33–4, 41, 43, 69, 70, 71, vikalpa 41, 48, 49, 51, 75, 201, 206
76–81, 214 vimarśa 213, 216, 217, 218, 219, 229, 231
Vātsīputrīyas 33 vipaks.a 40
Vedānta 53, 94, 117, 212 vyāpti 52, 111, 112, 197
Vedas 7–8, 13
authority of 39, 46 Yādavaprakāśa 136, 180
Vedic injunction 118 see vidhi Yāmuna 169, 170, 171, 178
‘veil of representations’ 31, 106, 124, 132 Yaśomitra 33
Vibhajyavāda 29, 30, 35 Yogācāra see ‘consciousness only’

Introduction To Indian Philosophy Christopher Bartley

Uploaded by

Copyright:

Available Formats

Introduction To Indian Philosophy Christopher Bartley

Uploaded by

Document Information

Original Title

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

Introduction To Indian Philosophy Christopher Bartley

Uploaded by

Copyright:

Available Formats

An Introduction to

Phenomenology, Michael Lewis and Tanja Staehler

Classical Chinese Philosophy, Im Manyul

© Christopher Bartley, 2011

All rights reserved. No part of this publication may be reproduced or transmitted in

British Library Cataloguing-in-Publication Data

ISBN: HB: 978-1-8470-6448-6

Library of Congress Cataloging-in-Publication Data

Typeset by Newgen Imaging Systems Pvt Ltd, Chennai, India

Introduction: Some Types of Indian Religiosity 1

1 Foundations of Brahminism: Vedas and Upanis.ads 7

5 Nāgārjuna and Madhyamaka Buddhism 56

6 ‘Mind-Only’: Yogācāra Buddhism 68

7 Sām.khya and Yoga 82

8 Nyāya and Vaiśes.ika 90

9 The Mı̄mām.sā Vision 117

13 Dvaita-Vedānta and Madhva 184

14 Tantra: Some Śaiva Philosophies of Kashmir 193

A quite different outlook says that understanding or insight alone is necessary

Śvetāśvatara Upanis.ad: The Lord supports this whole world consisting of

The Brahman lacks nothing, which is why it is sometimes said to be blissful.

Ātman acquires a meaning approximating to ‘essential principle’ or ‘soul’. The

It is recognized that there is a sense in which a person’s desires and actions

The Brahman he is and to the Brahman he goes. [BAU 4.4.6]

Bring me a fruit from the banyan tree.

Questions for discussion and

personal identity. The Brahmin position is that psychological continuity

1. All mental and physical phenomena are ultimately unsatisfactory (duh.kha/

Ignorance conditions volitional actions. Volitions condition states of mind.

This account of causation is held to account for the organisation of a stream

3. There is potentially an end to suffering, called nirvān.a or nibbana. This is the

This is a typical account of meditation involving successive stages. It did not

Finally, morality and meditation are not understood instrumentally, as

The body (rūpa)

Neither singly nor collectively do these transitory factors constitute an enduring

reflect a resistance to the disturbing teachings of momentariness and the

The next chapters will look at these four schools of thought.

Questions for discussion and

It appears that after approximately 200 B.C., some philosophically minded

causal power, the capacity to do something. The nature of a white atom is to

i) Material phenomena (rūpa): bodies, sense-faculties and corresponding types of

efficacy (kāritram). Efficacy (kāritram) may or may not be exercised. We

Indriya Vis.aya Vijñāna

Items in column three are products of the interaction of the corresponding

The personalists (Pudgalavāda)

The ninth chapter of the Abhidharmakośabhās.ya is a critique of what was

Questions for discussion and

latter’s category of unconditioned phenomena, holding that space is just

Dignāga and Dharmakı̄rti

and language deal in generalities (sāmānya-laks.an.a). But because there are

Apoha: the exclusion theory of

‘analgesic’ is a humanly constructed one because pain relief is a matter of

Self-awareness of mental events

Extracts from ‘The Examination of

The expression ‘intentional object’ means the proper form of something as it is

Thus the atoms are not the objective support of cognition.

As for the conglomerate of atoms, although it is what is represented in

It is not the cause of the phenomenal representation (ābhāsa).

Thus what is external is not the direct object of awareness.