Mulamadhyamakakarika

of
Nagarjuna
THE PHILOSOPHY OF THE MIDDLE WAY

lntroduction, Sanskrit Text, English

Translation and Annotation

DAVID]. KA.LUPAHANA

MOTILAL BANARSIDASS, PUBLISHERS

PRNATE LIMITED e DELHI

First Indian Edition: Delhi, 1991

ReprinL Delh~ 1996, 1999

C 1986 State UnivcrsiLy of New York,
All Rights Reserved
No part of this book may be used or rcproduced in any manner
whaLSoever \vithout wriuen permission excep t in the case of
brief uotations embodied in critical articles and reviews.

ISBN: 81-208-0774-x

Also availnbk at:

MOTILAL BANARSIDASS
41 U.A. Bungalow Road, Jawahar Nagar, Delhi 110 007
8 Mahalaxmi C hamoor, Warden Road, Mumbai 400 026
120 Royapettah High Road, Mylapore, Chennai 600 004
Sanas Plaza, 1302, Baji Rao Road , Pune 41 l 002
J 6 St Mark's Road, Bangalore 560 001
8 Camac Streel, Calcutta 700 O17
Ashok Rajpath, Pama 800 004
Chowk, Varanasi 221 OO I

Library of C!'ngress Cataloging-ln-Publication Data

Nagarjuna, 2nd cent.
Mufamadhyamakakarika : the philosophy of the middle way
Translation of : Madhyamakakarika
lncludes bibliographical refercnces and index
L Madhyamika (Buddhism)-Early works to 1800
1. Kalupahana, David J., 1933- II. tile
m. Lille: Philosophy of the middle way
BQ2792.E5K35 1986 294-.3'85 85-14757

PRINTED IN IND IA
BY JAINENDRA PRAKASH JIN AT SHRI JAINENDRA PRESS.
A-45 NA.RAINA, PHASE I, NEW DELHI 110 028
AND PUBLISRED BY NARENDRA PRAKASH JAIN FOR

MOTILAL BANARSIDASS PUBLlSHERS PRIVATE UMTTED,

BUNGALOW ROAD, DELHI I I O 007

To Profcssor N . A. Jayawickrcma
(yo) 11c11n'yo 'h osi mam' atthak.amo

U 5.~0)

PREFACE

Almost teo years ago, I undertook a new transation of Candrak.Irti's encydopaedic work the Pt4S(lnnapadi, a commentary on Nagarjuna's
primary philosophkal treatise, the Mu/41114dhyamak4kirikii. Meer 1 bad
completed nearJy ten cbapters, 1 'learoed through one of my scudents about
a similar attcmpt by Professor Marvin Sprung. I was about to give up my
project, when my student, who -had prviously srudied under Professor
Sprung, shared wltb me a copy of Ptofessor Sprung's ttanslatioo of the fitst
cbapter. Comparing his and my translations, I discovered that Professor
Sprung's translation was to some cxtent influenced by Stcherbatsky's wotk
(The Conception of Buddhist Nirvana , 1927). I fet then that my effon
would not be in vain, cspecially because l had expressed strong disagreement with Stcherbatsky's interptctation of the Buddhist philosophical tradicion (5ce my C4usality, The Cen/rai Philoiphy ofB11ddhism. 191)).
To my surprise 1 Professor Sprung's ttanslation) consiting of only seventeen chaptcrs (induding an incomplcte rendering of Chapter I), appeated in
1979. As I plodded along through my own labocious work, I began to
reaize how CandrakTtti was gradually leading me away from Nag~juna's
phllosophical standpoint. My suspicioos wece sucngthcned in 1981 when I
visited Jndla oo a Smithsonian gtant. Meeting with some scholai:s who were
brought up in the Vedantic uadition, I found them to be extremcy comfonable witll]'rgarjuna as intetpreted by Candrak.Trti and less impressed by the
tcachings -of early Buddhlsm as recorded in the Nikayas and the Agamas.
My suspkion tbat Nigarjuna and Candrakirti were ~pholding two differeot
philosophical standpoints compeUed me to take a fresh look at
Kumarajrva's Chung-lun, which is at least two ccncurics prior to Candrak.Ttti. Translating the entire Chung-lun into English an.d compating it
witb Nagarjuna's origioal Sanskrit text, I was pleasaotly surprized by their
similaries:. I found no justiflcation whatsoever in l~oking at Nagarjuna
through Candrakttti's eyes when there was a more faithful and doser disciple of Nagarjuna in Kumarajrva. This discovery diminished my enthusiasm
for deaning up my English ,:endeting of the Fraran1111p11dii for possible
publication.
Mter translating botb the Sanskric and the Chinese versions of Nigirjuna's tteatise, 1 proceeded to annotate both according to my understan.d ing
, of carly Buddhism as well as later Buddhist ttiditions before Nagirjuna.
The annotation of the Sanskrit tei alone turned out to be moce extcnsive

viii

THE PH!LOSOPH Y OF THE MIDDLE W/\ Y

than anticipated . FU.Cthermore, considcdng the di.fficulties that might arise

in publishing this work with Sanskrit and Chioese. texts side by side, and
also with the Chinese characters in the body of the annotation, 1 decided
once again to modify my project. The Sanskrit text is h~re presented with
annotation and inttoduction. The Chinese text with commentary wiH appear subsequently as a companion volume.
I am not unaware of the controversy this work may engender. Hoping
that it will be a healthy one, I intend to raise one mjor questioo regarding
Nagarjtina-, especially in the light of the more recent research in the bistory
of Bdclhism. Professor Hajime Nakamura's monumental work, Indian
Buddhis111. ( 1980). has provided more information regarding the bistory of
Buddhist literature tha any other work published so far. This carefully executed work not only deals with the cootents and authorship as well as the
chronology of most of the Buddhist texts, but also compares the different
versions available in Sanskrit, Paii , Chinese, Tibetan, and Japanese. After a
careful reading of this work, l canoor help recognizing an earlier stratum of
iterature that has so far been lumped together with all the literature that
carne to be called Mahayanistic. This indudes two famous pieces, the
Kayapaparivarta and the Vajracchedikii-prajnapiiramit?i (see Nakamura.
p. 159). I wondec whether the original versions of these t~ts can be approriately called Mahayanistic, even though they were preserved by the
Mahiiyiina schoos. This objection, indeed, is not very difficult from that
raised against considering the Nikayas and Agamas to be Hlnayiinistic because
they were preserved by the Theravadins, the Sarvastivadins, or any other later
uadition.
The major question that can be raised is: "Where would a philosopher like
Nagarjuna go in order to discover the Buddha's teachings?'' This historical
uescioo bas, to my koowledge, neither been raised nor aoswered. The
Sadriharmaput/anka-sutra that highlights the Hlnayana-Mahayana controversy was not yet written. That does not meao that the controversy was oot
known before Nagarjuna. Eveo if che controversy bad preceded Nagarjuna,
what were the canonical tcns, embodying the pure Mahayana phiosophical
standpoint, that Nagarjuna could have utilized in order to explain the
Buddha's message?
A careful reading of Nakamura's work shows it to be futile co attempc to
discover a pure Mahayana text that Nagarjuna might have been ab le to depend
upon. Before che compilation of the Saddharmapu??-(ianka, ope can bardly
expect to find a carefuJly executed tceatise chat would explicate the Mahayana
philosophy as it is presented by modern scholars. Since such sohistiated
Mahayana sutras wece not available to Niigarjuna, he could not help moving
on to the early discourses in the Nikayas and the Agamas in search of the Bud-

PREFACE

ix

dha's teachings, especially at a time when he realized that che probJems were
crcated not only by metaphysicians like the Sarvastivadins and the
Sautrantikas>, but also by morc popular religious ceachcrs like Avaghosa, who
over--emphasised the function of "faith" in the emerging bdief in a traosccndeoc Buddha. A careful rcading of Nagarjuna's rceatise will reveaJ that he
was. critical of both these trends. If Buddhagbosa wcre to be coosidered the
model of a Thcravadin and Candrakfrti or Santideva were to be looked upon as
idea MahayanistS, oeither the Buddha, nor Moggaupotta-tissa, nor Nigarjuna
would fit into thcir shoes.

The prcsent worlc may come as a surprise to many who arc familiar with my
prcvious publications, espccially bccause it repudiates many i:.hings chat I have
said about Nigarjuoa. Io those earlier works, my major eodeavot was to show
how the Buddhism of the Buddha diffcred from both Sthaviravada and
Mabayana, and the Jatter ioduded philosophers like Nagacjuna. My main contention with scholars Jike Stcherbatsky and Murti has been in regard to the_
manner in which the for.mer equated Sarvastivada with early Buddhism and che
latter portrayed the Buddha as a half-hearted rnctaphysician incroduciog a
theory of dcments char carne to be rejccted by Nagarjuna. I was preparcd co accept Murti's ioterpretation of Niigarjuna, white struggling to find ways in
which that interprl:tation could be justified without sacrificing the cmpiricism
of the Buddha. Amore detailed study of both Magarjuna and Candrak!rti has
convioced me that the former stili rcmains faithful to the Buddha, while the
lattec has moved more cowards a Vedantic inrerprctation, chercby initiating a
proces~ that culminated in the disappearance of Buddhjsm as a distinct..
ideology from the Indian scene a few cenruries later.

ABBREVIATIONS
(Consult Kenneth Inada's Niigiirjuna, for-a detailed
juna and the Midhyamika school.)

bibiography

on Nagar-

Anguttara-nikiiya, cd. R. Morris and E. Hardy, 5 volumes, London:

PTS; 1885-1900.

AA

Manorathapura1j'i, [Anguttara-a(fhtikathaj, cd. M. Walleser, London:

PTS, 1973.

Ahh11t

AbhidhammatthavibhaWii-(ika. cd. D. Pannasara and P. Vimaad

hamma, Columbo: Mahabodhi Press, 1933.

AD

Abhidharmadipa, see At;lv.

Adv

Abhidharmatlipa wilh Vibh'iifiiprabhii11rt1i, ed. P. S. Jaini, Patna:

K. P Jayaswal Research lnstirute, 1959.

AK

Abhidharmako1a, sec Akb.

Akb

Abhidharmakoa- bha!a,

ed. Pralhad Pradhao, Patna: K. P.Jayaswal

Research Institute, 1967.

'

Buddhay~as.

Ch'ang

Ch'ang A-han Ching (Dirghiigama) , te.

Tairho No . .1.

Chung

Chung A-han Ching (Madhyamiigama), tr. Gautama Sarghadeva

and Sargharalc$a, Tai.rho No.26.

Digha-flik41a, ed. T. W. Rhys Davids and J. E. Carpcnter, 3

volumes, London: PTS, 1890-1911 .

PIJ

Dhammappda, ed. V. Fausboll, London: Lu:zac, 1900.

DhsA

Atthaiiilim , Dhamma.rangan'i'-atthtikath'ii, cd. E. Muller, London:

PTS, !897.

DhsT

Abhidhammamula(ikii [Ttka oo Dhamma.raganl-affhakathaj , ed.

D. Pannasara and P. Wimaladhamma, Colombo: Mahabodhi Press,
1938.

It

ltivuttaka, ed. E. Windish, London: PTS, 1889.

Karika

See MKV{P), MKV(JQ .

Kp

Kiiyapapari1111rta

of the RPtnakuta:-sufra, ed. A. Stael-Holstein,

Shanghai: Commercial Press, 1926.

THE PHILOSOPHY Of THE

xu

MIDDLE WAY

Katha11at1h11, cd. A. C. Taylor, 2 volu.mes, London: PTS,

1894-1897.

Majjhima-nikiiya, cd. V. Trenckncr and R. Chalmers, 3 volumcs,

London: PTS, 1887-1901.

Mi/n

Milindapaflho , cd. V. Trcnckner, London: PTS, 1962.

MKV(P) Mlidhyamik.avrtti (Mmihyamakakiin'Jias), cd. L. de la Vallc Poussin,

Bibliotheca Buddhica 4, Sr. Petersburg: The Imperial Acadcmy of
Scicnccs, 1903-1913.
Madhyamakaiistra

of Niigiirjuna with the commentary; Prasannapt1'1ii by CandraRirti, cd. P. L. Va.idya, Darbhaoga: The Mithila

lnstiruce, 1960.

PEW

Philosophy East and West, cd. Eliot Deutsch, Honolulu: The

Univcrsity Press of Hawaii.

PTS

The Pali Text Society, London

Saf!Zulta-mlliiya, cd. L. Peer, 5 volumes, London: PTS, 1884-1904.

Sdv

Sphu(iirth'iibhidhaimakofa-11:/iikhy'ii, cd. U. Wogihara, Tokyo: The

Pubication

Sdmp

Association of Abhidwmakosavyakhya, 1932-1936 .

Saddharmapu'!<fe'ika-siilra, cd. H. Kern and B. Nanjio, St. Petersburg; The Imperial Acadcmy of Scieoces, 19 12.

Sn

S11tfii-nipiita, cd . D . Anderson and H . Smith, London; PTS, 1913.

Taisho

Tai.rho Shinshu Daizokyo, cd . J. Takakusu and K. Watanabe,

Tokyo: Daizo Shuppan Company, 1924-1934 .

Thag

Thera-thefi-gathli, cd . H . Oldcnbcrg and R. Pichel, London: PTS,

1883.

Thig

See Thag .

Tsa

Tsa A-han Ching (Saf!Zyukfiigama),

Tseng

Tseng-i A-han Ching (Ekottaragama),

Tai.rho No. 125.

Ud

Udiina, cd . P . Steinthal, London: PTS, 1948.

VbhA

Sammoha11inodanl, Vibhangatfhakatha, cd. A. P . Buddhadatta,

London: PTS, 1923.

tr. Gur,iabhadra,

Taisho No.99.

tr. Gautama Satghadcva,

ABBREVIATIONS

Vin

xiii

Vinay4 Pi{ala, cd. H. Oldcnbcrg, 5 volumcs, London: PTS,

18791883.

Vism

The Vi.s11ddh1~magga of B11ddhaghosa, cd. C . .A. F. Rhys Davids,

London: PTS, 1975.

CONTENTS

Preface
Abbreviations

vtt

xi

lNTRODUCTION

The Middle Path

Nagarjuna: The Myth
Nagarjuna: The Philosopher and Grand Commentator
Nagarjuna and .Kumarajiva
Nagarjuna' s Buddha
Analysis of the Kaccayanagotta-Sutta
Buddha's Conception of Language and Truth
Period of the ravakas
Moggaliputtatissa: The First Refotmer
Early Mahayana: The Second Reform Movement
Struture of the Kii:rikii.
Analysis of the Karikii
Nagarujuna' s Philosophical Enterprise
SANSKRIT TEXT, TRANSLATION AND ANNOTATION

Dedicatory Verses

1
2
5

8
9
11
16
20
23

24
26
31
81
99
101

CiAPTER

Examination of Conditions (Pratyayaparik$ii)

Examination of the Moved
and the Not-moved (Gatiigata-paif/fJii)
3. Examination of the Faculty of Eye .

1.
2.

(Cakfur-indriya-pafi~ii)

4. Examination of Aggregates (Skandha-paiilfJa)

5. The Examination of Elements. (Dhiitu-parilfJa)
6. The Examination of Lust and the Lustful
(Riiga-rakttJr-pari~a)'

7. The Examination of the Conditioned

(Sarp.skrta-parilfJii)
8. Examination of Action and the Agent
(Karma-karaka-pari~a)

105
118

132
140
147
153
159
180

xvi

THE PmtosOPHY OF TH E MrootE

WAv

9. Examination of the Por Entity (PUrva-pari/eya)

10. Examination of Fire and Fuel (Agnzndhana-pari/eyii)
11. Examination of the Prior and Posteor
Extremi ties (PUrvaparakoP,-paril~ii,)
12. Exatnination of Suffering (Du_ikha-parileyii)
13. xamination of Action and the Agent
(Sarriskara-parik.$a)
14. Examination of Association. (Sarrisarga-pariieyii)
15. Examination of Self-nature (Svabhiiva-pari~a)
16. Examination of Bondage and Release
(Bandhana-mo~a-parzl~a)
17. Examination of th e Fruit of Action
(Karma-phala-pari~a)

18. Exarnination of Self (Atma-pari~ar

19. Examination of Time (Kala-pari~ii)
20. Examination of Harmony (Sii:magri-pa/eya)
21. Examination of Occurrence and Dissolution
(Sa1!llihava-vibhava-paril~ii)

22.
23.
24.
25.
26.

Examination of the Tathiigata (Tathagata-pa~ii)

. Exrunination of Perversions (Viparyasa-pa~a)
Examination of th.e Noble Truths (Arya-satya-pari~ii)
Examination of. Freedom (Nirva?J.a-pan~ii)
Examination of the Twelve -Causal Factors
(D~iidasaga-pali~ii)
27. Examination of Views (D~ti-pa1i~ii)
.KARI.KA
Nomina Forms
Verbal Forms
General Index

INDEX TO THE

188
195

206
211
217

224
2.28
235
243
263
275
280
292
302
312
326
355
370
377
393

393
402
406

INTRODUCTION _ _ __

THE MIDDLE PATi

Myths of huge proportions have developed around the spirirua and
philosophicaJ stature of various personaJities in aJmost evecy scbool of Buddhism. Often these myths were inflated by sectarian rivarics tha~ continued to
plague the bistory of Buddhism , especially che rivary between the two major
schools, Theravada and 1.iahayaoa. These prejudices tended to polarize che
phiosophicaJ teachings of these two traditions chough. in face, they are similar
if oot idencicaJ. They are simiar in bcing faithful to the basie teachings of che
Buddha; they are aso comparable in the way in which chey rejected certain
metaphysical ideas that concinued to creep into the teachings.
The two aspeccs of the Buddha's teachings, the phiJosophica and the practicaJ, which arc muruaUy dependent, are clearly enunciated io ewo discourses,
the Kaccaya11agolla-1utta1 and the Dhamm111:akkappa11attana-sutta,2 both of
which arc heJd in high esteem by almosc all the schoos of Buddhlsm in spite of
theiI secrarian rivalries. The Kaccayanagotlll-Jutla, quoced by almost all the
major schools of Buddhism, deals with the philosophicaJ "middle path". paced
against che backdrop of two absolutistic cheories in Indian phiosophy, o:imely,
permanent exiscence (atthifii) propounded in the early Upant{ads and nihiistic
non-existence (na11hitfi) suggcsted by the Materialists. The middle position is
'expained as "dependent arising" (paficcasamuppada) which , when utilizcd to
explain the nature of the human pcisonalicy and the world of experience, appears in a formula consisting of twclvc factors (d11adaianga). The practicaJ middle pach is enunciated in the equally famous Dhammacakkappavattana-sulta,
respected by most Buddhists as the first sermon delivered by che Bud~ha . Herc
the middle path is becween the ewo extreflCS of self-indulgence
(kamaiuk.hallikanuyoga) and self-mortificarion (attak.ilamathanyoga) and consists of tbe noble eightfold pach (an'yo a!{hangiko maggo) leading to freedom
and happiness.
Throughouc che hiscory of Buddhism, Buddhists have endeavored to remain
faithful to the doctrines enunciated in 'chese two discourses, in spite of unforcunate divisions into Theravada and Mahayana and in the face of enormous
pressures. either from inside or from outside, either social or political, chat forced
them occasionaJly to deviated from the original message. For example, in the
sphere of philosophical specuations, one of the sects belonging to the so-called
Sthaviravada, namely, Sarvastivada, prescoted a theory of "sdf-narure" or
"substance" (svabhava) and some of the Mahayaniscs.admitted a conception of
" inherent thought of enlightenment" (bodhi-cilta), both of which, as .may be
indicated in tbe following d iscussions, are theories contrary to the fundamencal
1

TuE PHlLOSOPHY OF THE MIDDLE W AY

philosophical tenet of the Buddha, namely, "dependent arising" (paficca-

samuppida).
The practica middle path as enunciated in the famous Dhamffl(lcakkappavattana-sutta, which is complementary to or based upon the
philosophical middle path referrcd to above, was mote susceptible to varia
tions. The analysis of the wide variety of religous practices th;n emerged in the
two ttaditions, Theravada and Mahayana, which appear to be contrary to the
middle pach enunciated in the above discourse may require aseparate volume.
The present treatmenr will therefore be confined to the philosophical middle
path and its ~urvival, in spite of the heretical interpretations tbat occasionally
appeared in the Buddhist tradition. The survival of that middle position in
philosophy can be atttibuted to reformers lilce Moggaliputta-tissa (litde known
among the Western scholars of Buddhism, in spite of the important role he
played in the Buddhist council held during the re.ign of the Emperor Asoka of
Jndia) and Nagarjuna. Such personaitics bave emerged from time to time and
they have been responsible fot the continuacion of the Buddha's message. The
activities of such reformets have eitber been igoored, as in the case of
Moggaliputtatissa, or exaggerated, as in the case of Nigarjuna.
The present essay is not intended as an attempt to bighlight the conuibutions of the less ~nowo figures- like Moggalrputta-tissa.. whose famous
treatise, "The Points of Conuoversy" (Kathii.vatthu),3 awaits a ca.refu! and sympathetic treatqient by Buddhologists. On the contrary, this will be an attempt
to put into perspective the pbilosopbical and spiritual stature of Nagarjuna,
wh1ch has been cxaggerated beyond limit.S, more by modern scholars chan by
the classical Buddhiscs.

NAGARJUNA: THE MYTH

Nigarguna has been considered the second Buddha and has occupied a second position in the line of patriarchs in almost all schools of Mahayana Buddhism, primarily because the adherents of these schools refused to recognize
the spiritual status of thousands of Buddha's immediate disciples who, according to the Buddha's owo recognition, had attained the same knowledge alld
understanding (fiii'!-t1-dassana) as well as the mora and spiritual perfection attaioed by the Buddha. While the intellectua and spidtual attainments of the
immediate disciples are cleary portrayed in (excs like the Theragatha and the
Thengatha, no such info..tmation is available to us about Nagarjuna's spirirual
attainments exccpt the account of his conversion to . Buddhism and. his

lNTRODUCTION

scholastic activhies refecred to in a biogcaphkal account uanslated into Chinese

by KumarajTva. 4 Nagarjuna's stature as the second Buddha derives, therefore,
from his basie writings, which are generally looked upon as philosophicaJ 1nterpretations of the Mahayina suttas.
Kenneth lnada, who presented one of the most sympathetic analyses of
Nagarjuna's thought, has admitted that the veneration of Nagarjuna "at times
reached such ridiculous heights that his name was sanctified and stamped
everywhete with recldess abandon even for purposes of frauding scriprural
a.uthority."> He was probably referring to the attempt on the part of some of
the later Tantric writers to seek authority and sanctity for their ideas, which
were undoubtedly influenced by some later Indian religious practices. Even if
one were to ignore such excesses, it is possible to maintain that the exalted
position accorded to Nagarjuna yet reflects an uncritical and dogmatic attitude
of some of the later Buddhists toward the spi.rirual ideii.I of early Buddhism .
Such an attitude is reflected not only in some of the Mahiyana texts but also in
some of the Theravada commentaries. For example, in the later Theravada
commentarial liceratu.re, an exalted status is accorded to the Abhidhamma in
relation to the discourses, so much so that the Buddha had to ascend to the
world of deities (devaloka) and ps:each the Abh.idhamma to his "mothe.t''. who
was residing there. 6 Such an admission, though intended to provide authoriry
and santtity to a body of literature that emerged long after the passing away of
the Buddha, undoubtedly carried the implication that the Buddha's immediate
disciples were not capable of undersranding its contenrs. However, even
though such an irnplication was there, the Theravadins did not daborate this
story in such a way that it' would lead to the devaluation of the ear:ly idea of-.ui
arahant. Yet, when a similar need was felt by the Mahayanists ro provide
authotiry aod sanctity to somc of the latcr Mahayina textS such as the Ptajfiiiparamifii sutras, which were obviously later than the Abhidharma treatises,
they were not satisfied with merdy saying }hat they were "great dis_:ourses"
(11Pt'p11/ya-iiitra) grearer than those included in the Nikayas and the Agarnas.
They, in fact, proceeded to condemn the very idea of an arahant embodied in
those discourses and to criticize the spititual attainments of the early discipJes
of the Buddha.
In this particular movement, the Saddhartnapuf'!r/affka-s11tra leads the
fied.7 The motivation or even the fina goal of this rovement may have been
very noble. It was, in fact, one of the earliest atrempts ro unify all the conflicting ideas and ideals that were creating enotmous tifts among the Buddhist
community . Yer, the manner in which such unification was carried out led co
increasing conflicts rather than to their reconciliation or appeasement.
Even a superficiaJ glance at the rustory of Buddhism wouJd reveal the existence of "mooks'' (bhikfu) who deviated from the idea and who falsely claim1

THE PHILOSOPHY 01' THE MIDDLE WAY

c:d spirituaJ attainments white leading a form of life infetior to that of ordinary
lay people. Such monks were reported even from the time of the Buddha. The
Vinaya-pifaka8 as well as the Kif.Jyapararivarta9 generate no sympathy for such
miscreant monks, the latter branding them as a pack of dogs fighting cach
other for a morsel of food thrown at them.
Such sdfish and dishonorable behavior on th(' part of certain monks may
hve been counter productive. Self-sacrifice and absolute altruism could
emerg~ as noble ideaJs in sucha context. However, such actions and reactions
need not be a reason for condemning even the immediate disciples of the Buddha, l!rahanlr like Sariputta, Moggallana, and Kassapa, as people of "low
aspirations" (liin'iibhirafii), 10 and forcing them to discaim thc:ir attainmc:ncs in
order to accept a new idea, an idea] certainJy contrary to the "middk path"
enunciated by the Buddha in his very first discourse to the world. It is by
folowing a "middle path" avoidiog the rwo extremes of self.induJgence and
self-destruction that the discipes of the Buddha attained the state of freedom
called " the appeasement of dispositions" (sakhara-samatha) and continued to
work for the welfate and happiness of mankind. Very authentic records
avaitabe in the Thera- and Then-g'iithas bear ample testimony to the idea of
the ~arly discipes, and it is also an idea! recognized' by Nagarjuna, the champion of the ''middle way" (XVII. 1).
While ~he Theravad~ns elevated the Abhidhamma to an exalted position
without devaluing the ideas emhodied in the carty discourses, the S11ddhar~
mapu(llnka appears to have gone much funher in dealing with this eotire
Buddhist phiJosophical and religious tradition beginning with the Buddha
himself. It is responsible not only for condemning the early disciples, but aJso
for down-playing the value of the carty discours.es. The discoutses included in
the Nikayas and the Agamas were considered to be inferior in content. The
argument presented is that because the immediate disciples could not understand the deeper doctrioe the.Buddha had to preach an jnfecior and unsatisfattory doctrine to .suit their incelectual capacity. Such a statemeot, however, has
a hidden implication, nan1ely, that the Buddha lacked the capacity to teach the
deepcr doctrines in a way that would be intelligible to the people who were in
his presence. -In the Mahayana traditioo, the stage was thus made ready for a
philosopher li.kc Nagarjuna, who is supposed to have best expouoded the doctrine, to be eevated to the level of a second Buddha, nay, even to the starus of
a supreme Buddha more exalted chan Slikyamuoi. Thus 1 it is not surprizing to
find some modern commentators prodaiming the view that the locus bud that
appeated in the world with che binh of the Buddha grew up and blossomed
forth with the appearance of Nagatjuna. In fact, a scholar like T. R. V. Murti
has maiotained that che Buddha even suggested a "theory of elements'' (dharma), which carne to be be rejected by Nagarjuna when the latter presented his

lNTI\ODUCTION

cheory of "emptiness" (fiinyalii). 11 This uodoubtedly places Nagarjuna in a

more exalted position than that occupied by the Buddha. Mosc dassical and
some modern scholars have thus creaced an atmosphere where the incerpreta
don of Nagarjuna's philosophy wiJ ha_ve to assume a historical development
and unfolding of doccrines that wece merely suggested, not taught, by the
historical Buddha. Some writers on Buddhism, intoxicated by this conception
of the evolution of thought, havc shown reuccance to recognize che sophistication with which philosophical ideas were preseoted by the Buddha 2500 years
ago. Having miserably failed to perceive the philosophical ingenuicy of the
Buddha as reflected in the Nilcayas and che Agamas, as well as the subsequent
degeneration of that system in the !acer commenrarial tradition, followed by a
revival of the earlicr system by philosophers like Moggaliputta-cissa and Nagarjuna, these writecs arc insisting upon a gradua sophistication in Buddhist
thought comparable to what one can find in the Western philosophical trad!
tion. 12

N AGARJUNA:

THE PHCLOSOPHER AND GRAND COMMENTATOR

In the followiog pages, an anempt will be made to present Nagarjuna merely

as a grand commentator on che Buddha-word and to show that he did not try to
improve upon the ceachings of the: Buddha. His work will be: c:xplaincd as an at
tempc co destroy che weeds thac had grown around che Buddha's teachings as a
resulc of some of the ideas expressed by philosophers of both che Schaviravada
and che Mahayaoa tradicions. Ir will be showo that tne Mii/a-madhyamakakarika (hereafter abbreviated as Kiin'ka) is a superb commentary oo t he Buddha's owo KacciiyanagoJta-sulla, a commentary in which Nagarjuna upholds
cvery statemenr made by the Buddha in that discourse, bringing together more
materia from the ocher discourses as well, and chen clearing the watce muddied
by the speculations of some of che meraphysicians of the la ter Buddhist ttad ition. The continuation of certain seccarian prejudices among che faich.ful
adherencs of Theravada and Mahliylioa may be understandable. Critical
scholarship . on the contrary, has a responsibility co remain unsmeared by such
sectarian prejudices. Modern scholarship in Buddhism, which began with th('
recogoition of this sectariao rivalry as represencing a major split in Buddhist
philosophical and religious ideology. has come a long way in asserting ics
u n cenabiity. Howcver, scholars are now beginning to realize chat che
Theravada/ Mahayaoa discinccion is an cxaggeration and that the fuodamencal
reaching of the Buddha has remaiaed int~t throughouc the cenruries. Now it is

THE PH!LOSOPHY OF THE MIDDLE WAY

time to exorcize the terms Theravada and Mahayana from our vocabulary. A
major obstacle co the elimination of this distinction is the manner in which
Nagarjuna's phiosophy has com e to be expoundcd by a majority of modern
scholars. The presenc cranslation of Nigarjuna's Kiinka and commentary upon
cach of che vcrscs therein are inteoded as a correctivc to this intcrprctation.
A carcful study of the docuines in che cxteosivc corpus of Buddhis~ literature
indicates ve.cy dcarly how certain fundamental idcas have survived, in spite of
the occasional appearance of concepts chat conflict with the basie teachings of
che Buddha and thus produce controversies among the Buddhist thinkers.
Without undertaking a careful study of such insta.nces, scholars have rather uncriticaUy lumped together the early discourses of the Buddha and the summaries of their conten ts chat came to be preserved in the so-caJled Abhidharma,
togcther with all d1e interpret.ive cexts compiled by some of the later commentators, either in the form of vibhiif'ii or a!(hakatha, and criticized trus whole
corpus as being representacive of Theravada or Hrnayana. The same is clone
with some of the Mahayana discourses (sutra) and the treatises (fastra). The
contents of d1e discourses as well as of the Abhidharma literature arc examincd
only in the light of such commentarial explanations and not independent of
them. Modern scholarship has thus failed to extricate itself from commenta.riaJ
traditions. There seems to be oo justification for coosidering the discourses and
even the eady Abhidharma literature as sectarian works of the so-called
Thecavada. Theravada or Sthaviravada in genera , and Sarvastivada and
Saucrantika in panicular, may be considered sectarian, but their sectarian views
arc found not in the discourses and the Abhidharma but in the commeotarics
tbat carne to be compied on these two bodies of literacure. The elevation of the
Abhidharma to the level of a supreme body of litera tu re, moce exalted chan the
discourscs, is the work of these later commentators and not of the compiets of
chose Abhidharma texts. The Mahayanists themselves, bothered by the
substantialist thought of the Sarvastivada and Sautrantika schools, endeavored
to prcserve the eady teachings by emphasizing the negative aspect of the Buddha's doccrioe, especially the doccrine of emptiness (fUnyalii) . The
Kiifyapapari11arta as well as the early Pcajfiapararnita literarure represent tbis
reacrion co the substantialism of atce Buddhism and t.h.is literacure shouJd be
dissociated from the sectarianism chat cmerged as a result of the arcempt at
unificacion in treatises like the Saddh11rmapu11tfonka.
An attcmpt will be made in the followiog pages to show that even somc of
the moce prominent philosophers of Mahayana wece really trying to overcome
such sectarian ioterpretations and go back to the non-sectarian form of Buddhism as embodied in the early d iscourses, without rcjecting either the
canonical Abhidharma texts that embody posicive teachings or the early
Mahayana su tras that emphasizcd the ocgative aspect of the Buddha's docrrine.

[Nrn.OOUCTION

The present aoalysis will be confined co the v.:ork of Nigarjuna in India.

Once Nigarjuna's philosophy is critically and objectivey analysed, h will be
possible to see whether there is any substance to the rivalry between the two
major philosophical traditions, Madhyamika and Yogacara. That projecc has
co be postponed to a later occasion.
A.K. Warder was one of the first to .taise the question whetber Niga.r:juna
was a Mahayanist.n His reason for raising chat question was that in the KJirika,
which undoubtedly was Nigarju.na's most significant work, no reference what
soever is made to any one of the major discourses of the Mahayana tradition ,
not even to the fatnous P.rajff?iparamita-sutra.r. Wa.rde.r believed that the
discourse in the Samyukta served as a source for Nigatjuna's treatise, even
though he did not specify them. The only discourse referred to by name is Kiityay11niiva11ada,14 a discourse found both in the Pali Nikayas1' and the Chioese
Agamas. 16 This single most important fact has often beeo ovetlooked by mos.t
of the leading scholars who have written about Nagatjuna.
Even where this fact has beeo ooted, scholars have assumed that Nagarjuna
was merely referriog to the Buddha's rejection of the two extremes of existence
(atthitii) and non-existence (natthita) in that discourse and that was all. So far ,
no published work on Buddhism (available to the present author) has treated
the contcnts of the Buddba's cliscoucse to Kaccayana in detail before ptoceeding to analyse Nagarjuna's thought. Assuming that Nagarjuna was a
Mahayanist and, therefore, must have rejected any literature that carne to be
ptesctved by the Sthaviravadins, these scholars have proceeded to analyze the
Kiiriliii in the light of their own prejudiced understanding of Machyamika
philosophy. The ultimate .result is bafflement and coofusion. Not only are tbey
reluctant to accept certain positive statements of Niigarjuna in the Kiirik.a, they
are also ready to abandon some of the most imponant chapters in that work
eithet as later interpolations- or as having no relevance to Nagarjuna's thesis.
To asswne that Nigarjuna was a philosophyet: who would merely pick out
from the Buddha's statements only those that would support ot fit in with his
own preconceived nocions is to do him g.reat injustice. lf he was rejecting a
theory of elements suggested by the Buddha (as Murti seems to think). Nigarjuan, who was one of the most fearess critics of metaphysical views, would have
certainly said so. Nor is the.te any evidence to support the hypothesis that the
"Admonition to Katyayana" (KJity1iyanava11tida) that Nagarjuna was referring
to was a vetsion different from the. Kacciiyanagotta-sutta found in Pali and
Chinese. The Kiiyapaparif!aflfl of the Ratnakuta contains two discoutses, both
of whkh were addressed to Kasyapa and which deal with the middle path. One
of them refcrs to the middle path and explains it in terms of the twelve factors
of the human personality (d11adasaga), 11 while the other explains the middle
path in negative terms as "non-ceasing, non-arising, etc." 18 This latter version is

THE PHllOSOPHY OF THE MIODLE W AY

oot fouod in the Nikayas and the Agamas. Nagarjuoa was probably aware of
the existence of both these versions and he understood their implicatioos. His
Kiin"kti was an attempt to explaio the docuioe without rejectiog the contents of
any one of them. However, his refcreocc to Kacyayana, instead of Kayapa, is
extremely significant, in that be was conversaot with not onJy the contencs of
~e Ratnaku{a versions but also the version included in che Nikayas and che
Agamas. For this reason the analysis of Nagarjuna's philosophy as embodied in
the Kiirika will be undenaken with a view to locating the sources of those ideas
that arc accepted by him and those thar arc criticized and rejected by him.
lt is anticipaced that an analysis of the Kiin"k.a in relation to the "Discourse
to Katyayana" as well as ocher discourses iocluded in the Nikayas and the
Agamas will lead to a better understanding of the Buddha's philosophy
wichout exaggeracing the so-called Hlnayana-Mahayana discinctioo .

NAGA.ff.JUNA AND KUMARAJIVA

Recent scholarship in the bistory of Buddhisc thougbt has emphasized a
distioccion bctween Indian and Chinese forms of Buddhism. Indian Buddhism
is explained as an attempt to deal with causation throug!'i karma, white Sinitic
Mahayana is secn as advocatiog causation thcough dharma-dhatu. 19 Early Bud
dhism. according to che proponencs of this thesis, underwent a radical change
when it was introduced inro China. In order co deal with this question of transition, it would be necessary to provide a complete English translation of
Kumaraj"lva's rendition of Nagarjuna's Kiin"k'ii into Chinese. Richard Robinson
made the first systematk attempt to deal with this problem of transition. 20
However, that was done on the basis of an inadequate examioation of the first
chapcer of Nagarjuoa and Kumaraj"lva. A carefu1 scruriny of Kumarajiva's
work has convinced me that the form of Buddhism introduced into China by
him was not at all differenr from chat of Nagarjuna. Thus, if my contention
tbat Nagarjuna's philosophy is a mere rescatement of the empiricist and
pragmacic philosophy of the Buddha , the form of Buddhism incroduced into
China would also be the same as the origina teachings of the Buddha wich oo
paradigm changes. Aod this may account for che survival of Buddhism aloog
side of the equally pragmatic philosophy of Confucianism, wbereas it failed to
survive in lndia in the face of a very strong idealistic tradition . The prevalent
view - chat Buddhism, because of its idealistic character, goc absorbed inco the
ideaistic tradition in lndia and failed co remain as a separate enticy - needs

1NTRODUCTION

careful scrutioy, especialy when a leading philosopher like Nagarju na is not

seen to advocate such an idealistic view. This caHs for a detailed treatment of
the Yogacara tradition in Buddhism that has not yet been attempted .

NAGARJUNA'S BUDDHA
Nagarjuna's Buddha was no doubt Gautama (see KJirikii XXVIl.30), the
prince from che Sakyan coupcry (p(esently pan~ of Nepal) who attai.oed
enlightenment and rorned out to be the most formidable opponent of almost
every major philosophical idea that came to be presented by the lndians. In
fact, as will be explained below, the phiJosophical atmosphere was so confused
during the Buddha's day thac sometimes be was fo.cced ro coin new terms to ex-
press his thoughts.
Two of the major philosopbical theories that dorinated the Indian scene
during this time were (1) existence (sat, astitva), proposed and developed for
centuries by Indian chinkers since the time of the eady Vedas, and (2) nonexistence (a.sat, 11iiistit11a), presented by the Materialists reacting against the
craditional metaphysics . Existence or astitva was no ordinary empirical existeoce
b ut the existence ofa permanent and eternal substratum in man as well as in all
aspeccs of nature. In man. it was the immutable self (atman) that remained in
boodage to the impermanent psychophysical personaity and which returns to
its ultimate abode, the universal self (Atman), once it is freed from thar bondage and reaches its ultimate moral status (Brahman) . Attempting to eicplain
the origin of this reality in man as wel as in oature, some of the craditional
philosophers setcled for a conception of a creator god. As it is, this may not
have generated much protest from tte Buddha. However, the Indian
philosophers wece not satisfied with the simple aorion of a creator god. At a
very eary stage, they asserted chat this self (at1nan) was created by a god or gods
who dete.r mined that it beloogs to one or the other of the foursocial classes: the
priestly (brtihmatJa), the warrior (k1atriya) , che merchant (vaiya), and the ser\rant (iidra). 21 Thus, cach individual's :a arus was predetermined and unchangeable. It was th is particular idea of creation that elicited rhe most vehement criticism both from the Materialists as well as from the Buddha.
Denying such a metaphysi<ial self, the Materialists moved to the other extreme of advocatiog the annihilatioo of the human personaliry aftec death, and
then aso denied any mral responsiblicy for human actions. Instead , they propounded a theocy of the indestructibilicy of matter. 22

JO

THE PHILOSOPHY OFnlB MIDDLE WAY

Kaccayanagotta-Sutta
The Buddha's discourse to Kaccaya.na, a discourse wbose authority was
recognized by aJmost aH the major philosophica1 schools of BuddhJsm,
becomes mea.oingful only in the context of the philosophica.l views mentioned
abovc. Following is the complete text of the discour:se as reponed by Ananda:
Thus have I heard: The Blessed one was once living at Savatth.i,
in the monastery of Ana(hapiJ:l<;ika, in Jeta's Grove. At that time
the venerable Kaccayaoa of that dan carne to visit bim, and
saluting him, sat down at one side. So seatcd, he questioned che
ExaJted one : "Sir [people] spcak of 'right view, right view.' To what
cxtcnt is therc a right view?"
"This world, Kaccayaoa, is generally indined towards two
[views]: cxistencc and oon-cxisrencc.
To him who perceives with right wisdom the uprising of the world
as it has come to be, the ootioo of non-existence in the world does
not occw. Kaccaya.na, to him who perceives with right wisdom che
ceasing of the world as it has come to be, the ootioo of existeoce in
the world does not occur.
The world, for the most part, Kaccayana, is bouod by approach,
gra.sping and inclinacion. And he who docs not follow chat approach and grasping, that determinacion of mind, chat inclination
and disposition, who does not eling to or adhere to a view: 'This is
my self,' who chinks: 'suffedog chat is subject to arising ariscs; suffcring chat is subjcct to ceasing, ceases,' such a person does not
doubc, is not perplexed. Herein, his knowledge js not othcrdepeodenc. Thus far, Kaccayana, thcrc is 'right view.'
'Bverythiog exists,' - this, Kaccayana, is one excreme.
'Everything does not exisc,' - rhis, Kaccayana, is the second extreme.
Kaccayaoa, without approaching either extreme, the Tathagaca
teaches you a docrrine by the middle.
Dependent upon ignorancc arise dispositions; dependent upon
dispositions arise consciousness; dependent upon consciousncss
arises the psycbophysical personaJicy; dependent upon the
psychophysica.l persona.lity arise che six senscs; dependent upon the
six senses arises contact; dependent upon contact arises feeling;
dependent upon feeling arises craving; dependent upon craving
arises grasping; dependent upon ,grasping arises becoming; det>endent upon becoming arises birth; dependent upon birch arise oJd
age and death, grief, Jameotation, suffering, dejection and dispair.
Thus arises this encire mass of suffering. However, from the utter

lNTRODUCflON

11

fading away and ceasing of ignorancc, there is ceasing of disposi

cioos; from the ceasing of disposition.s, there is ceasing of consciousness; from the ceasing of consciousness, there is ceasing of the psychophysical persooality; ftom the ccasiog of the psychophysical personality. therc is ceasing of the six seoses; from the ceasing of the si.x
senses, there is ceasing of cootact; from the ceasing of contact, there
is ceasing of feeling; from the ceasing offeeling, there is ceasing of
craving; from the ceasing of craving, there is ceasing of graSpiog;
from the ceasing of grasping, there is ceasiog of becomiog; from
the ceasing of becoming, there is ceasing of birth; ftom the ceasing
of birth, there is ceasing of old age and death, gcief. amencarion,
suffering, dejeccion and dispair. And thus tbere is the ceasing of
this entire mass of suffering."23

ANALYSIS OF 1HE KACCAYANAGOTIA-SUTIA

The discourse is delivered in response to a fundamenta question in

epistemology: "What is a right view (sammadirtht)?" the Brahma1tila-suttanliJ
rcfers to sixty-two varieties of views prevalent during the Buddha's day .i4 After
his enightenment, the Buddha reaized that oone of these were sacisfactory.
He was not willing to subscribe to any one of them. For this reason, many
scholars of Buddhism have assumed that the Buddha did not have a view to
present. For them, he had oo sixty-third view co propouod. If that was the case,
the Buddha could have admonished Kacca yana not to be bothered by any view,
whether it was right or wrong, true or fase. However. that was not the case.
The Budha prceeds to enumerate tWO basie views that are prevaleot in the
world. The sixcytwo views referred to in the Brahmajala-suttanta represent, in
one way or fillother. a proliferation of these two basie views of permanent existeo.ce (atthitii, Sk. astitva) and non-existence (n'atthilli, Sk niistitva). He chen
provides reasoos for rejecting both these 'Views. The reasons arc epistemoogical
and therefore deserve detaied examinatioo. "For him, who perceives with right
knowledge , the uprising of the world as it has come to be, whatever view that is
i.o the world about non-existence will not be acceptable."
The two terms of great epistemological significance that occur in the above
statement are (1) "perceives" (passatz) and (2) "right knowledge" (sammappa;Jfiii). There could be no mystery associared with the implications of the first
of these two terms. Pa.ssali or "perceives" refers to simple, ordinary sense
perception, for what is perceived is not something chat is mysrerious bu~ simply
the arisiog arid ceasing of various phenomena in the world. lt does not, at least

12

THE PHILOSOPHY OF THE MIDDLE WAY

in th.is instance, refer to a special or unique form of insight not shared by the
ordinary people. What makes the difference is "r.ight knowledge" (sammappanfia, or simply panifii), and that difference is then explained in the next
paragraph.
The perceptions or sense experieoce of the ordinary person in the world are
geoerally "bouod by approach, grasping and inclination" (upaya-upadiinaabhinive.rt.i-vinibandha). They arc colored by one's prejudices, by one's likes
and dislikes. lf a person is ab le to avoid such approach, grasping-and inclinatioo, if he does not follow his dispositions, chen that person would not cake a
detecmined stand and say: "This is my [per.rnanent] self." He would perceive
phenomena in the world as arising and ceasing.
The perception of arisiog and ceasing of phenomena conditioned by various
factors is availablc even to ordinary people who have not been able to completely
free themselves from prejudices. Thus, there is a common denon1inator ber;ween the perceptioos of an ordinaryperson and those of the enlightened one.
H owever, the ordinary person continues to worry about a permanent and eteroal substance behind phenomena or about a supreme being who is the author
of all that happens in the world. He is assailed by doubtS about wbat he
perceives. One way of overcoming such doubts is to confine oheself to what is
given, chat is the causa! dependence ofphenomena, without rrying to look for
something mysterious. The Buddha realized that "Wheo phenomena (rlhamm'ii) appear befor<: the brahman who is ardenc and contemplative, his doubts
disappear, as he sees their causal natute. "2 >
According to the Kaccayanagotta-sutta, if a person does no_t make up his
mind that there is a permanent and eternal self and continue to look for it, but,
instead, merely uderstands things as they have come to be (yathabhUta'f(J), as
for example, understands suffering (dukkha) as something that arises depending upon conditions, then he does not fa11 into doubc. In other words, instead
of look.ing for mysterious causes one should start with whatever .cases one can
discover that are cootributory to each siruation. Of course, in most cases. past
experiences arc a good index. Thus, in order to attain such knowledge one does
not have to go around looking for a teacher who would transmit that
knowledge in a sec ret session or in some mysterious way .. His knowledge would,
in chat case, 'not be other-dependent (apara-paccaya fla1Jam ev 'as.ra etI ha hott).
On the basis of such knowledge and experience, one is said to have "right
view" (samm'ii-diftht). "Right view'' in early Buddhism is cot.rasted with
"wrong view" (mz~cha- rlitthz) . These ewo ace oot contrasted in the way truth
and falsehood are contrasted in the pre-Buddhist cradicion. In the latter case,
what is true is considered to be what exists (.rat). Whatever ex.ists, is real, and by
definition whatever is real caonot be othcrwise. According to this a pn'ori
definition, "truth" has to be something that exiscs dlway.r. Yet, what is given to

JNTRODUCTION

13

the sense expedences is nor available always. Hence it was assumed that what is
true is something other than what is given to the sense experiences, and that re
mains always (saisata) and in everything. lt is the ontological truth (atman) as
welJ as the mora.I truth (brahman) in evcrything.
As mentioned earlier, ::. theory of existence or a theory that says "everything
exists' (sahba1{1 pttht) is wrong, not becausc it can be proved to be false, but only
in the sense that it does not take into consideration the empii:ically gi'ven fact of
cessation (nirodha). Hence, it is characterized as "confusion" (musa, Sk. mrfii),
not as "non-existent" or "un-rruth" or "falsehood" (P-satya).
Avoidance of the theory that ''everything exists" docs not, according to the
Buddha, make the opposite view, narody "everything does not exist" (sab6P1{1
n 'fJttht), any tnore true. The reason for this is that this latter theory implies
much more than a simple denial of a pcrmanent and eteroal substance in man
(atman) or in the universe (btahman) . 1t implies complete discontinuii:y in.
phenomcna or their annihilation (uccheda), and this too is a wrong view, not
bec.ause, like the for mer, it can be proved to be false, but because it is partia! in
rhat som~ aspecrs of experience like arising (samudaya) cannot be accounted for
by such a view.
The Buddha did not worry about discovering strictly Jogical arguments to teject aoy one of these views. He merely avoided these two theories in his explanation of existence. Hence his statement: "Without approaching either excreme, the Tathagata teaches you the doctrine by the middle" (ete te ubho ante

anupagamma majjhena lathagalo dhammaf?I desett).

The Buddha dearly distinguished hjs philosophy from that of his contem
porary, Saojaya Bella~~hiputta, who refused to make any pronouncemeots
through fear that he would be found fault with.26 It makes no sense to assume
that the Buddba, after criticizing the rwo exueme views, avoided propoundfr1g
any view.or observed complete silence. Such an assumption would undetmioe
the autheocidty of almostaU the docttines attr.i buted to the Buddha and wouJd
stand in the way of appreciacing the greatness of this philosopher and spirirual
leader whose message clid not fade into oblivion, as .in the case of the skeptic
Sat1)aya Bella~~hiputta , but instead became a formidable world-view
throughout the last twenty-five ceoruries. For tbis reason, the fina concusion
of the Kaccifyangotta-sutta can in no way be ignoted as a later interpolation by
the so-called Theravadtns.
In this fioal statement, the Buddha was attempting to expfain the human
personality as well as its cicperien~es in the world in terms of the priociple of
dependence, without resorting to the two extreme views that be criticized
earlier. ln the first part of that explanation, he was describing the personality in
bondage, as it evolves conclicioned by "approacb, grasping and inclination."
This is the twclvefold formula (d11adasang11) presented in posicive terms,

14

THE PHILOSOPHY OF THE MIODLE W AY

describing the functions of ignotance, disposicions, and behavior prompted by

such disposicions in the maner of propeUing human beings into states of
uohappiness and suffecing as well as continued process of binhs and deatlS.
The negative statement e.xplains how, as a resut of the eliminacion of rhat igoorance and the dcvdopmeot of insighc, one comes co pacify one's dispositions
and ~hereby eliminate su.ffer'ing as well as the continued cycle of births and
deaths. Such is the conclnsion of che Kacciiyanagotla-sutta.
The Ka&cayanagotta-sutla, though brief, lays down in no unclear terms the
basie teachings of the Buddha. Funher elaboration of this philo~ophy is
available in the huge collection of discourses of the Nilcayas and the Agamas.
Although denying ex:istence and non-existence conceived of in such a
mctapbysical manner, the Buddha recognized existence and non-ex:istence in a
more empirical sense, such existence and non-existeoce being understood in
terms of the experience of coosequences or effects (attha , Sk. artha). Thus,
white bcing aware of the metaphysical implications of the nominalized forms :
"exist-ence" (atthi-ta) and "non-exist-ence" (n' atthi-fii), the Buddha continued to u~e che verba! fo.rms "exists" (attht) and ''does not exist'' (n 'atthz) to
explain his view of existence.11
The existence of things as wdJ as their arising and P3:SSing away arc clearly expressed in the famous formula:
When that exiscs, this comes to be; oo the arising of chat, this
arises. Wheo chat does not ex:ist, this docs not come to be; on the
cessation of chat, this ceases (/masmi'f!J sati idaf!t hoti, imassa up-

padii idaf!t uppa1j'ati. Jmasmif!t fi.fali idaf!t na hoti, imas.ra nirodha

idaf!l niru1j'hatt).2e
Yet the linguiscic conventions of his day did not provide che Buddha with
technica terms to express this idea. The notion of self-causation (.raya1(l kala1{Z,
Sk . .r11aya1'(l krtaf!l) was prevalent i.n the rradition of his day, but unfortunately
it carried witb ie the implication of a mctaphysical self (atman), permancnt and
eteroal (sassata), which he wanted to avoid. 2 9 The idea of external causation
(paraf'(l kataf'(l, Sk. para-krtatrJ) was not different from the Materialist view of
anoihilation, especially in its denial of mora responsibilicy ,Jo A combinacion of
these rwo views was also not satisfactory, for the Buddha was probably aware of
the implications of the Jaina theory of causation chat atteropted to combine
both.H Under such circumstances, i.t was almost impossible for him to express
his understanding of existence. This may also have contributed to his initial
reluctance to exlain his ideas after this enlightenmenc. However, he was equal
co the task.

lNTRODUCTION

I)

An attempt to explain rhe manncr in which a phenomenon gives rise to or

produces aootber phenomenon, how a cause gives ri.se to an effect, would have
involved bim not only in the taSk of unravelling the essence or substance in a
cause rhat produces the effecr but aso in the job of.predicting the effect ari.sing
from a cause with absolute cenainty, a job for which he did not have the
necessary empiricaJ mea.os. The Buddha decided to explain this process in terms
of "dependence." He was thus led to speak of an event that has occurred
(samuppanna) by uacing it back to a condition or set of conditions upon which
it depended (pa{icca) . Having analysed the proccss of becomiog (bha1Ja) in this
manner, he laid down a principlc that would cxplain future occurreoces. Thus,
from observiog "dependently arisen" (paficca-samuppanna) phenomena, he
asserted the principle of "dependent arising" (paficca-.ramuppada). That termioology is indeed conspicuous by its absence in the pre-Buddhist Indian
literature.
The old lndian term dhamia was retained by the Buddha to refer to
phenomena or things. However, he was always careful to define this dharma as
"dependeotly arisen phenomena" (pa#cca-samuppanna-dhamma) . Most of the
controversies of the ater Buddhisrs centered on this conception of dharma, and
thercforc the various uses of the term in the Budaha's discourses may be appropriatcly examined herc.
The term dhamma (Sk. dharma) has four rc!lated uses in the early discourses.

{l )

(2)
(3)

(4)

Dhamm'ii (in the plural) meaning phenomena or things.'2

These arc the dependently arisen phenomena refer~ed to
earlier. They may also be described as elements of experience.
Dhammo (in the singulac) meaning the uriifotmity o(
phenomena ot things as represented by rhe principle of
dcpeadence (paficca-sarnuppada).H
Dhamma (in the plural) referring to things or phenomeoa
cvaluated as good or bad in an ethical sense.H While good is
ofren designated dhamma , the notion of bad is expressed by
its oegation, (l-dhamma .
Dha??JmO (in the singular) expressing che uniformity of moral
phenomena, which aso represeated the idea or the standard
of morality derived from the mora! pcrfection attained by the
saint. Hence, nibbana oc freedom is also called dhammo. ~>

In order to djstinguish this notion of dharnma from the Indian concepcion

where the term dharma meant reality (atman), in an oncological sense, the .
Buddha utilized the cooception of resulc or coosequc:nce or fruit (attha, Sk. ar-

16

THE

Pl-IJLOSOPHY OF THE MIODLI!

W AY

tha) to bring out the pragmatic meaning of dhamma. For the Buddha.
whatever is true or real (bh'Uta, taccha) is alo what produces result (atthasa'f?lhita).36 This pragmatic definition of truth or reality was moce often used in
expJaining mora! phenomena. Hence the reference to the three types of results
or consequences or fruits:
1)
2)

3)

bad, evil-an-atthf!, correspooding to a-dhamma.

good, beneficiaJ-attha, corresponding to dhamma, and
ulcimate good, ultimate fruit-param' attha, correspooding
co nibbana.31

BUDDHA'S CONCEPTION OF LANGUAGE AND TRum

While the term dharma, in the four contexts referred to above, may be takeo
as implying empiricaJ truths, a more compreheosive use of the term is also
available. In rhis case, the dharma (or sad-dharma) expresses the ootioo of
"true doctrine," and without any hesitacion this may be explained as a "true
st~temeot ,'' a use that may be most appealing to the modern linguistic
philosopher who is generaUy averse to met.aphysics and insists that "truth" pertains to statements. The use of the term dharma in this sense at ooce renders
futile any attempt to speak of a lioguistically transcendent truth or reality in the
Buddhist cootext.

Dependent arising is the middle path presented by the Buddha between the
extremes of etetnalism and annihilationism, of strict determinism and chaotic
indeterminjsm, of absolute. reality and niliilistic unreality, of permanent ideotity and absoute difference. Considedag the mann.er in which he explained the
middle position between these extremes, no one could maintaio that this middle position is beyond linguistic descciption or uaoscends any forn1 of verba! expression. In fact, the two terms that are generally utilized in the absolutistic
sy~tems of Indian tbought to presentruch a standpoint, namely ''indescribable"
(avacya) and "indefinable" (anirv~antya), do not occur in the.early discourses
of the Buddha. The ter:m "undeclared" or "unexplained" (avyakata) occurs, but
it ~ used to refer to problems such as the duration and extent of the universe,
the identity of or the difference betweeo che soul and the body, as wel as the
status of the tathagata after death- these beiog pcobJems that could not be explaioed on the basis of any empirical evidence.~8 For the Buddha, whatever is
empirically giveo is also describable or defiaabte without having to Msume
!Uetaphysical standpoints.

INTRODUCTION

17

Thus in the Buddha's view language is not, in itself, an inadequate means of

expressing whac is cmpiricaly given. Y et modern incepreters of Buddhism
seem to assume chat the Buddha considered Janguage inadequatc to express the
truth about ex.istence that he d iscovered. The evidence for such an intcrprctation is racher dubious. No attempt is made to examine che Buddha's own
statements abouc his cnlighteoment, as recorded in such 9iscourses as "The Noble Quest" (Ariyapariyesana).l 8 Most books on Buddhism pubished in the
modern world wiU atuibute to che Buddha, as well as to his early disciples and
even to the ater ones ike Nigarjuna, a distinction between sammuti and
paramattha. Sammuti (Sk. samvrh) is explained by Candrakirti as language 0
and paramattha (Sic. paramartha) as ultirnace reality or absou te ttuth. 41 The
terms sammuti as well as voh'lira occut in the early discourses.41 Sammuti,
(derived from sam + V man ''to think") literaJly means "agreement" and
therefore, "conventioo" . Vohara (Sk. vyavaliiira) means "usage." A rather lop-
sided interpretation of rhese two terms as implying "language" only has caused
havoc in the expanatioo of the ceachings of the Buddha as well as of Nigarjuna. Tbere seems to be no justification for confining the meaoings of these
two terms to anguage only.
Conventions (sammult) are of various cypcs - linguistic, social, political,
mora! or ethica, or even reigious. Eveo a superficial glance at pre-Buddhist
literary uaditions would reveal the manner in which the Indians elevated
linguistic, social, political, mora , and religious conventions to the level of absolute reaitics, permanent and ecernal. The language of the Vedas became th e
absolute language, posscssing miraculous powers. In spite of the existence of
such languages as Dravidian and Chinese, which have nothing to do with the
Vedic language, some educated Indian schlars stil believe the Vedic is the
mothcr of a U languages. Therefore oot a single syllable of it is to be tampered
with. Another convention, the social order consisting of four castes, carne to be
considered absolute. Punishment awaited those who violated it or ignored ie.
Arjuna was to be rewarded for maiotaining chat social order. Political conveations derived their absoluteness from the absoluteness of the social order. A
rules (k1atnya) who ignores the advice of the spirirual leader and guide
(brahma1Ja)' was doomed to failure, since he was thereby ignoring the law
(dharma). The absoluteness of the moral and ethical conventions was equally
recogoized . No ocher form of moralicy except that which concributcs to the
preservation of the social system was permissible, Rcligious duties were specific
and unatcrablc.
When, in che Suttanipata, the Buddha spoke of sammuti, he was referring
to all these different kinds of conventions . 4~ According to him, these coovc;ntions have come co be depending upon specific condicions (puth111/ii). They
wece not absoute and ultlmace; rhcy were noc universally binding. The Buddha realized that when these convcncions were considered to be absolutc and

18

TuE PHILOSOPHY OP nm MIODLE WAY

ultimate (parama"!') they contributed to the worst form of dogmatism (dittht) ,

which evenrually led to all the cooflicts (ka/aha, viggaha) in the world. These
constituted the worsr forms of obsession, obstruction, constraint, or bondage.
Therefa re, ilie Buddha claimed that a wise man (vidva) does not approach (na
upei.1) such conventions.4) This does not mean that he ignores all conventions.
fnstcad, he understands their conditionality and as welJ as their fruitfulness
without clevating them to the level of ultimate realities, thereby ma.king them
absolutei, or simply ignoring them as absolutely unreal and therefore useless.
F,re"tdom (nibbana) could chen be interpreted more appropriately as freedom
ffom obsessions, obsessions for as well as against such convencions. The
elimination of such obsessions or constraints (pap4fica) turned out to be more
difficult chan abandoning pleasures ofsense, for if by freedom was meant only
the atter, the Buddha could have attained enlightenment during the time he
was practicing self-mortificacion.
Herc again, the difficulty lies in adopting a middle path without accepting
conventioos as being ultimate or rejecting them as being useess. The uniqueness of the Buddha's philosophy lies in the manner in which a middle path
can be adopted with regard to any conveotioo, whethet it be linguistic, social,
political, mora, or religious. Since the preseot aoalysis of the Buddha's
philosophy is undenakeo ooly as a prelude to the examination of Nagarjuna's
thought, and since the latter was more conceroed with the ,basic doctrines of the
Buddha, our attention at chis point will be focussed only on the way in which
the Buddha adopted the middle pach in dealing with linguistic conventioos.
As _poioted out above~~ term sammuti was used in the early discourses to
refer to all kinds of conventions: H.wever, chere were two terms that were.vCI'f
specifically employed to refer to linguistic conventions. They are nirt#ti-patha
(the way of etymology) and adhivacana.patha (the way of definitioo), The
Sarttyutt~nik'iiya contaios an important discourse dealing with linguistic con
ventions, which are neither to be clung to as absolute truths, nor to be ignored
as mere conventioos. The discourse called Nirutti-patha runs hus :

There are these three lioguistic conventions or usages of words or

terms, which are distinct, have been distinct in the past, are distinct
in the present and will be distinct w .the future and which are not
ignorcd by the wise brahmans and recluses. Whatever materia
form (rnpa) that has been, which has ceased ro be, which is past
and has changed, is c.alled, reckoned and termed 'has been' (ahost)
and it is not reckoned as 'lt CJtistS" (attht} or as 'it will be'
(bhavissatt) . .. . [This is repeated for the other four aggtegates:
feeling, perception, dispositions and consdousness.] Wbatever
. materia form has not arisen nor come co be, is called, teckoned or

lNTROOl)CTION

19

termed 'it will be' (bhavissatz) and it is not reckoned ;is 'ie eidsts (atthi) or as 'it has been' (ahost). , .. Whaeevet materia form has
arisen, and has manifested itself, is called , reckoned, or tecmed 'ie
exists' (atthi), and it is not reckooed as 'it has been' (aho!z) nor as 'it
wiH be' (bhavissatt).'6
The Buddha advised his disciples "not to eling to dialectical usage nor go
beyond the .imlts of linguistk convention" (janapada-nirultif!t nabhinivese.yya
i?imafJfzaf?l natidhaveyya). 47 Such being the middle position adopted by the
Buddha regardiog linguistic conventioo, it would be an extreme position to
maintain that language is either utimately real (as it was the case with the Indians who made vae a supreme god) or that it is useless when it comes to expressing ultimate reality.

For the Buddha, language derives its "meaoing" (attha) when it is able to
p.roduce results (attha). and thus what is true (bhuta, taccha) is that which bea.rs
results (attha-sarp,hita). 48 The Buddha did not recognize anything rhat is fase
to be productive of results. Truth in this sense can be equated with "me.aningful" language. Thus, linguistic expressions that imply permanence and annihilation would be "meaningless" (an-attha) in that they do not commu"nicate
anything thac is given in e.xperience (dhamma), where experience is understood
in terms of the felt results (attha) rather than in terms of an indefinable
ultimate reality.
Having thus rejected the two views, namely, the traditional Indian view that
che human personality consists of a permanent and eternal spirirual ent'ity
(atman) and the Matetialist view the denied such a spiritual entity and
recognized maccer (body) to be the only real~ty, the Buddha continued to speak
of the psychophysical personality (nama-~pa), referring to ie with such tetms
as "I" (aha'!IJ) and "you" (tvarp,) and even the term "self' (atta) when speaking of
that personality.
With the emergence of Buddhism as a formidable phiLosophicaJ and
spiritual movement that undermined the very foundations of che tradirional
Indian phjlosophy and religion . Indian thinkers reformulated their substaotialist world-view, presenting it in a morc subte and appeaJing form in the
Bhagavadgzta. The notion of dharma embodied in this t ext may be analysed in
terms of the three Buddhist categories presented above, namely, an-flrtha, ctrtha and pararrUirtha, Iastead of the pragmatic definitions of the Buddhists, the
Indian thinkers were presenting a moce substantialisr incerpretation where,

1)

an-artha = che psychophysical personality (nama-rnpa)

which is uoteal and whic.h is corHtasted with rhe real self
(atman).

20

2)

3)

THE Pl-l!LOSOPHY OF THE MIDDLE

attha

WAY

= the permanent and eternal self (atman) in man, the

so-called dehin (the dweller in the body), which is in bondage

because of attachmenc co che psychophysical personality, and
paramiirtha - = para11U1tman, which is the ultimate reality,
the universal self identified with God.

Thus was inaugurated an enormous controversy becween Buddhists and Indian philosophers that continued to rage for severa! cenruries until Buddhism
completely disappeared from the Indian soil as a philosophlcal and spiritual
force around the seventh and eighty centurie.s, only to survive and flourish in
the countries souch and southeast of India as well as in the Far East.

THE PERIOD OF THE SRAVAKAS

The survival of a pragroatic philosophy in the face of an extremely absolutistic tradition such as the one cmbodied in. the Bh4ga11adgT/a was not
easy . One of the ways in which the Buddhists responded to thac philosophy
was by compiling the now famous Buddhist classic, tbe Dha1tJmapad11. As
the citle indicates, it was an attempc to counter the Indian absolutist and
substantialist definition of dhamma. The Buddhist philosophers, confronted
by the onslaught of Indian thinkers asserting the reality of the self (atman),
spent most of their time analysing what they called dhamma in o.rder to
show that there was no permanent and eternal self. As Kenneth lnada has
rightly remarked , this represented 1'the most active, highly vibrant and compecidve age in Buddhist bistory known as the Abbidhaana period . ... lf there
arc high watermarks to be considered in Buddhist bistory, the Abhid.harrna
period certainy ta.ces a very high leveJ, a level of great fermentation and
flourishment pf Buddhist thought. IdeologicaJly speaking, no ocher period jn
Buddhist bistory, whether of the Tberavada or Mahayana, or even national
Buddhist developments such as in T'ang Dynasty China, could ever match, or
come up to the level of activiry as recorded during this period. " 49
Two complete sets of Abhidharma texts compiled during this period are
available to us. One is preserved by tbe Theravadins consisting of the foHowing
texts;
l. - Dhammasangani,
2
Vighanga,

lNTRODUCTION

3.
4.
S.
6.
7.

21

Dh111ukath'1i,
Puggalapanatli,

Kathavatthu,
Y1111111/a, and
Pauhana.

The other version was preserved by the Sarvastivada schoo and comprises
one major work and six ancillary texts. They are as folows;
1.
2.

3.
4.

s.
6.
7.

]fiangpraithana (atuibuted to Katyayaijfputra),

Sagttiparyaya

(attr. Maha Kau~~hila/Sariputra) ,

Prakaraf!apada (attr. Vasumitra),
Vijnanak'iiya (attr. Devaarman) .
Dhatukaya (attr. Pori:J ~ or Vaumitra),
Dharmaskandha (attr. Sariputra/Maudgalyayana), and
Prajaptifastra (attr. Maudgalyayaoa) .

That the Theravada and Sarvastivada schools preserved these two bodies of
literarure does not make them sectarian, any more chan t_he discourses, preserved by any school, could be branded as such. Athough the treatment of subjectmatter io these two sets of works differs widely, the subjecr-matter is pract.ically
the same. Both deal with the categories into which the human personal.ity as
well as human experience carne to be aoalysed in the early discourses where
tbey receive amore discursive treatment. The analysis of human experience into
aggregates, eleme.nts, and faculties , all of which were considered to be dharmas
~r elements of existence, seems to be the first and foremost concern of the
Abhidharmikas. Undoubtedly, the purpose of this analysis is co dcmonstrate
the absence of a self or substance in these phenomena. Exhaustive analyses of
the various types of relations tbat obtain among them wece also undertaken,
providing a sort of scholastic ad:v.ancernent in the study of such phenomena,
but sti11 not deviating fro.m the fundarnental teac.hings of eady Buddhisip. Examination reveals that these two processes in some way representan atte.mpt to
deal wich the same issues that the Buddha was concerned wich, namely.
"dependently arisen phenomena'' (pa{iccaiamuppanna-dhamma) and "dependent arising" (paficcasa1nuppada). The knowledge of these two processes was
looked upon as right understanding, which utimately leads to the attainment
of freedom (niniiif!a) .
Unfortunacely, ew.o of the schools thar were involved in this Abhidharmic
enterprise were driven roo far in tbeir academic study of the 4hfJN1UIS, probabJ.y
by the unrelenting criricisms evelled against Buddhism by the traditional

22

schoos

THE PHlLOSO PHY OF THE MIDDLE W AY

of Indian philosophy. These schoos carne to be knowa as Sarvastivada

and Sautrantika.
The Sarvastivada concluded their analysis of dharmas with the recognition of
ultimate discrete atomie elements which they wece unable to put together even
with a theory of four basie relations. The result was that they were compelled to
admir a singularly mec.aphysical conception of "self-narure" (.rvabhava) to account for the experienced concinuity of sucb discrcte phenomcna.>0 This selfnanae could not be Jooked upon as something impcrmaneot and cbanging, for
thac would be to defeat the very purpose for which it was formulated in the first
pJace. Therefore they insisted that this self-nature (svabhava, dravya) of dharmaJ remain during all cbree periods of time , the past, the present and the
future . No ocher conception could be moce beretical in the eyes of the Buddhisrs who were avowed non-substancialiscs (aniiJmaviidf ).
The view chat dhamJa.I have self-nacure had irs impru:t on the cooceptioa of
"dependent ar.iing" (prafiya-.ramutpada), the central coocepcion in early Buddhism as we as in tpe early Abhidbarma. Hcnceforward , "dependence" carne
co be expained on the basis of self-natute. l t curned out to be no mare different from the iden ticy theory of causation (.ratkarya-11ada) formulated with
such precision , usiog logical arguments, by the Sankhya school of thought.51
The celationship becween these two schools of thought is so close that one can
hardly discount mucual iofluences and borrowings.
One of the schools chat reacted against chis concepcion of "self-narure". other
than the tradition represcntcd by Moggauputcatissa referred to earier, was the
Sautrantaka school of Buddhism. As ics name implies, this school was openy
ancagonistic to the "treatises" (Jii.rtra) and iosisted upon rerurning to the
"discourses" (sU.tranta) as sources for the study of the Buddha-word .n lt considcred the notion of "self-nature" as a theory of "self' (atma-11ada) in disguise.
Howevcr, for some inexplicable reasons chey failcd to realize that neither a
theory of atoms (paramaf!u) nor a conceptioo of moments (k!a11a) was part of
the early Buddhist teachings, either in che discourses or in the Abhidharma
treatises. On che one.hand, chey p robably assumed that these two conceptions
were not the root cause of all the confusion among che Sarvastivadins. On che
other hand, they felt that these two concepcions were, afcer all. not incompatible with the docteioe of impermanence (anicca, Sk. anilya) in che early
discourses. Wichouc abandoning atomism and momentarioess, the
Sautrantikas proceeded to expJain "dependence" and eoded up recognizing a
sorc of non-identity theory of causarion (asatkiirya-vada) comparable to t he one
proposed by che Vaise$ika school of Indian phiJosophy.B
Evcn though the Saucrancikas were openly critical of the substantialist
conception of dharma advocated by the Sarvastivadins, their reluctance to
abandon the theory of momcnrs (k.fa1Ja) Jeft them with the difficut task of ex-

INTROOUCTION

23

plaining the experienced conrinuicy in the individuaJ person . The ernergence of

schools filce "personalists" (pudgala-1iiida) and "transm,igrationists" (samkrantt~
v'iida),> closely related to and sometimes identicaJ with the Sautraotikas, is indicative of the solutions that this school had to offer in order to overcome the
dtfficulties arising from the accep tance of a thcocy of moments.
The Satvastivlida and Sautraotika schools thus presented a rather complicated set of theories, all concribucing ro philosophica confusion. The farmer
percdved a "self-nature" (svabh'iiva) in the cause and emphasized the idenrity
(ekalfla) of cause and effect, while the latter, seeiog no such "self-oature" but
merely perceiving "other-oature" (para-bhava). insisced upon the difference
(niinatva) between cause and effect. The Sarvastivada cooception of self-nature
(svabhava) was extended to aJI phenomena, indudiog the human persooality,
while the Saucrliocikas, denying self-nature in pheoomena, surreptitiously iotroduced a conceprion of self or peISOn (at1nan, pudgala) in a human personality.

MOGGAllPUTIATI~A: THE FIRST REFORMER

The Russian-Buddhist scholar Tb . Stcberbarsky was one of the first among

Western scholars to ignore the vecy significant differeoces between early Buddhism and Abhidharma on the one hand, and Sarvastivada and Sautrantika interpretations of the "discourses" and Abhidharma on the ocher. For him, Sarvastivada was not only an interpretation of Abhidharma but was Abhidhama,
and the early Buddhism of the discourses as well." H owever, there was at least
one disciple of the Buddha who was not willing to accept either the Sarvastivada or the Sautrlintika as the correct intetptetation of Buddhism. This
was Moggalrputtatissa. Critical scholarship unforrunately has blindly dismissed his views without much serious consideratioo, even though they arc
presented with darity and logical acumen .
Almost 250 years after the Buddha and 300 yeacs before Nagarjuna, Mogga.lrputtatissa was responsible fot the "great purge" in the Buddhist tradition .
The Indian emperor Asoka, as he dcclared in his Minoc Rock EdictS, was actuaJly
instrumeotal in unifying the Buddhist Order (Saf!Jgha) by expelling the
miscreant and schismatist monks and getting them to don white (lay)
garmentS. Yet, the backgro4nd fot this great purge was prepared by MoggalTputtatissa wheo in his famous "Points of Conttoversy" (Kath'iivatthu) be
refuted the ideas presented by almost seventeen beterodox schools of Buddhism. "MoggalTpunatissa is said to have followed the mcthod of discourse .
adopted by the Buddha -satthar'ii dinna-naya-11asena-at the time he

24

THJ!

PHllOSOPl-IY OF THE MIDDLE WAY

established the matika, 'topics' of the ,K{111haJ11[11tthu]."'6 Th.is monumental

work is an attempt to go back to the early teachings, and in doing so ~he
author, for the first time in Buddhist history , utilizes even abscract logic.
Whether the use of such abstract logic. is appropriate is not a matter that coocerns us herc. What is importaot is that Moggafput tatissa was critical of certain
ideas which were incompatible with the Buddha's phllosophy,
The metaphysical theory of a person (pudgala), propounded by the
Sautrancikas and rheir allies, was the first of the mecaphysical views co be ta.ken
up for criticism in the Katha11atthu.>7 With great ingeouiry and logical precision, Moggalfputtatissa descroyed the concept of a person propounded by the
Sautrantikas and established what may be called the non-substantiairy of the
human pecsonaliry (pudgala-nairatmya).
With equally cogent argumeoc.s, he anaihilated the Sarvastivada, doctrine of
dhamza that implied substar:itial and eternal existence (sabba'f{J sabbadii
attht), 8 and established the non-substantiality of all dharma.J (dharmanairiitmya]. These two unconuovertcd achievements, recorded in one of the
most authencic texts, have been compJetdy ignored by those who atcributed a
substanrialist theory of elements to the eady Buddhist tradition.

EARLY MAHA YANA: THE SECOND REFORM MOVEMENT

The same text highlights anorher controversy that was beginning co ntl1e the
minds of Buddhist thinkers ducing the rhird cen rury BC and which became the
topie of a heated debate during the fust and second ccnmries AD. This pertained
to the questioo whether the Buddba is transcendent (lokutlara). Moggauput
tatissa rejected the view. gradually gaining ground in the Buddhist tradition,
which favored craoscendence.'9 The biographies of the Buddha, like the
Mahavastu 1 were probably not yet wricten . Mahayana, with its conception of a
coropletey transcendent Buddha, had not come into existence by that cime.
The Saddharmapu't/anka , which is resp onsible for condemning the miscreant
monks (bhik!u) as wdl as the ideaJ of a saint (arhant) in eiuly Buddhism, in"
dicates a gradua growth wich the fina version assigncd ro the third century
AD. In the earliest versions of some of the early Mahayana sutras, such as
Vajracchedika-prajnaparamita and the Kiiyapa-panvarta, there is no mention
of a bodhisattva. 60
.
What sort of Mahayana can chere be without the conception of a ctanscendeot Buddha and the notion of a borihiJattva? lt certainly ought to be differeot
from the kind of Mahaya na that one comes across in che avaiab le versins of

lNTRODUCTJON

25

the Saddharmapur!r;nka and other r:cxcs that indude a condemnation of the

early arhant-idea.
Iodeed the K.ayapa-parivarta, even in the Sanskrit version which is not the
earliest, will enable us to understand what that originaJ Mahayana was. lt was
not the Mab.a:yana that carne to be deeply prejudiced against early Buddhism as
weJl ~ Theravada, for, as pointed out eacJier, even though we find a suong
criticism of the moak (bhik111). a criticism that may be accepted even by the
Buddha and his disciples ike Moggalrputcatissa, this criticism is not extended
to the early idea of the "wonhy one" (arhant).
09e of the most important series of discourses ot instruction (dharma~
paryaya) referred to in the Klifyap~panvarta 'as "the great pinade of gems"
(maha-ratna-kufa), percains to che "middle path" (madhyam7i-prati'pat).6J
Hece we find a long list of middle paths, most of whicb are described in
negative terms. However, side by side with the negative descriptions, one also
discovers a positive description of the middle pach in ceans of the twdve faccor
of the human persona.ity (dvadasaizga). It is indeed an abbreviation of the
Kacaayanagotta-sutta ,"With Kasyapa as the interlocutor or the person to whom
the discourse is addressed.
The need for negative descriptions, especially at a time when Buddhahood
was nor yet looked upon as a transcendent state or as an AbsoJute, caJls for an
explanation. Available historica1 records indicate that soine of the canoaical
rexts chat emphasized the doctrine of non-subscanciality (naimtmya) , and
which included the work of Moggatrputtatissa, found a haven in Sri Lanka and
other South East Asian counrries after the third cenrury BC. However, some of
the early discourses as well as some of the Abhidharma teitts were still citculating in lndia and carne to be preserved in a Prakrit, slighdy different from
Pali. Yet, what carne to be popular after the tbird cenrury were the intCJpretations of the Buddhist metaphysicians, like the Sarvastivadins- and the
Sautrahtikas . The early Mahayana that did nor indude either the concept of
"ttanscendence" as applied to the Buddha or a notion of bodhisattva, but
which emphasized a aegacive doctrine while at the same time preserviag the
positive assertions of early Buddhism, was therefore a response to the Sar. vastivada and Sautrantika metaphysics, rather than a reaction to the early
Buddhism of the "discourses" or the Abhidharma.

N!igafjuna 1s Mifsion
Nagarjuna , wbo lived at a time when the Theravada-Mahayana conflicc had .
not degenerated to the level that is presented in the SadclharmapurJt/aiika,

26

THE

PHILOSOPHY OF TI-IE

MtDOLE WAY

therefore had an already different mission. It was indeed not the least different
from the mission chat lay before Moggailputtatissa, to expose the untenability
of certa.in heretical views chat were gradually becoming popular in the Buddhist
uaditioo . A careful reading of the Kiirikii will ccveal the fact that Nagarjuna
had all the help he needed to achieve this cask. As mentioned before, even after
the transference of the Pali canon to Sri Lanka, che discourscs and the
Abhidharma texts survived in lndia. Thus the discourses of the Buddha as well
as the traditioo of the disciplcs (fravaka) were available to Nagarjuua. The
humilicy with which Nigacjuna bows down to the BuJdha and the respect with
which he treats the Buddha's disciples (Jravaka) 61 are in complcce contrast to
the unsympathetic attitude of the later Mahayanists toward the eatlier Buddhist cradition.
In the foUowing analysis of che Kiirika, it will be shown char Nagarjuna actempced to discredit heterodox views, especialy those of the Sarvastivadins and
che Sautrantikas, and establish the non-substantialicy of all dharmas (Chapcers
III - XV) as wel as the non-substantiality ofpudgala (Chapters XVI - XXl) and
therc:after to explicate che positive doctrioes of che Buddha as embodied io the
early discourses like the Kaccayanagoita-sutta.
The present analysis is, tberefore, conuacy to the more popular incerprecacion of Nagatjuna espoused by commentators like Candrakirti who emphasized
the reductio ad abstJrdum (prasangika) metho~. lt will be moce sympathetic co
the interpretations offered by Nagacjuna's disciples like Bhavaviveka and the
moce positive thiokers of the Midhyamika school. While highlighting the iogenuity and philosophical marurity cf Nagarjuna, the preseot analysis wil at
the same time be unsympathetic toward the mych that Nagarjuna was a second
Buddha.

STRUCTURE OF THE

KARlKA

Selec'tions from the works of a major philosopher belonging to a tradition

may be hepful in introducing that tradition but not in providing a complece
view of that phiJosopher's thought. The reason is thac when a philosopher
presents his ideas io some form , he feels chat everything he has said in chat
wock is relevant co his thought. If anything chat he has said is ircelevant to what
he proposes to convey co his readecs, he would be not only wascing his time, buc
also the readcr's.
Actempts have often been made by modern scholars co pick out selections or
chapcers from the works of ernincoc philosophers of the East, hopiog thereby co

IN1'Rooucr10N

27

provide a complete and accurate picture of their ideas. Sometimes they are c.o nsidered to be essential secitons or cbapters, the implication being th.at the rest is
inessential. 6' This undoubtedly has conuibuted to a great deal of
misunderstanding and sometimes deliberate distortion of the author's ideas.
Coo.sidering the unsatisfaccoriness of such a method, the present analysis of
Nagarjuna's thought will be presented on the basis of an examination of his
Kiin'k'ii taken as a whoe, with ever,y word, every verse, and every chapter in it
created as in integraf part of that work. This is done in the absence of any concrete evidence that some portions of this work are not by Nagarjuna.
A supetficial reading of this work, with 448 vetses divided into 27 chapters,
could leave the reader with the impression chat the text is repetitious. This
wrong impressioo will disappear like a mirage if one keeps in mind the circumstances that led to the complication of this work, the motivation for writing
it, the background in which it was written, and the goal that was to be achieved,
Such considerations will enable one to see a carefully execuced plan or structure
in the Kiirika. In order to highlight di.is Stf\.lcture, the Kari/iii will be aoalysed
hece into four major .sections, withouc changing the sequence either of the individua! verses or of the cbapters.

Section I
This first seccion includes Chapters I and II. which deal with the most fundame.ntal doctrines of Buddhism , causatioo and change. The problem of causarion or "dependent arising" is taken up io the first of these. If this were a cext
writteo during the Buddha's day , this cbapter would undoubtedly have dealt
with theories of e.xistence presented by the traditional schools of Indian
philosopby advocating the reality of a permaneot self (atman) and the
Materialist school that denied such a self (an-atman) thereby denying the concinuity of the human person as wdl as his mora! responsibility. 'HoweveI,
Nagarjuna was living in the second century AD and his problems, as mentioned
earlier, were created more by the Sarvastivadins and the Sautraotikas than by
the non-Buddhist schools. Th.is is dearly evidencfrom the way in which Nagarjuna begins his first chapter.
The first verse in this chapter refers to four different theories of causatioo or
arising: (i) self-causatlon, (ii} external causation, (iii) both self- and external
causation , and (iv) arising out of a non-cause. After enumerating four such
theories, any further explanatio would naturally commeoce wi th an analysis of
the first of these four theories, sef-causatioo (svata.utpattt) . Thus , the four .
types of relations (pratyaya) referred co in the next verse shoud be takeo as ex-

28

THE PHILOSOPHY Of THE MIDDLE

W/I Y

arnples of the theory of self-causation (s,vata-uJpattt), even though the most

respected modern interpreter of Nagarjuoa, narody T,R.V. Murti, following
the footsteps of Candrak:Trti, took these four types of causa! relations to represent the theory of externa causation, and not self-causation.64 He assumed that
self-causation was ptesented by the Hindu schools and external causation was a
theory advocated by the early Buddbists only. As explained above, the early
13uddhisc theory of causation cannoc be placed und er ~he cacegory of eicher selfcausation or external causation. Nagarjuna was clearly aware of this and
therefo.re, even though in the tJJSt verse he denied the p0$Sibility of any one of
the four causal theories, in the second verse he recognizes four conditions
f.pmtyaya) without denying them~though to make his analysis more comprehensive he denies a fifth condition. The denial comes ony in the third
verse, but what is importaot to note is thar it is not a denial of che four conditions (prtltyaya) but of the rnanner in which the coodition is considered to be
relaced to the effect. When Nagarjuna said, ''The self nature of an existent is
not evident in the causa! condition, etc." (Na hi sfiabhavo bhavan"iit?i
pratyayadi;u11idyate, 1.3), he was not rejecting or denying condicions, but only
self-nature (svabhava) that some philosophers were positing in the condition
f.pratyaya) in order to account for the arising of the effect. This is a quite elear
iodication that Nigarjuna was not rejecting the Adhidharma theory of conclitions but only its interpretation by some of the metaphysicians, in this patticular case chat of the Sarvastivadins. As pointed out above, there is every
evideoce that it was the Sarvastivadins who interpreted the theory of conditions
(pratyqya) on the basis of a concepcioo of substancc (Ivabhfiva) . In the same
verse, Nagarjuna proceeds to deny externa causation (partlkl-utpattz) or, more
specifically, the conception of "other nature'~ f.par11--bliiiva) advocated by the
Sautrantikas. If this background is kept in rnind, the understandiog of Nagarjuna's ideas in the Kiirika is not as formidable as has been assumed.
Chap ter II deals with the problems created.- not by an empirical theory of
change and impermanence (anityata), for that was a fundamentaJ conception
of early Buddhism- by a more metaphysical theory of change and impermanence based on a logical or even a psychological theory of moments (k;af!a-

1iiida).

Sectzon II
The second section includes thitteen chaptets, begioning with an examination of sense faculcies (lndriya, Chapter ITI) and ending with an examination of

lNTRODUCTION

29

substance (svabha11a, Chapter XV) . The entire section is an attempt to establish

the doctrine of the non-substanriality of phenomena (dharma- nairatmya)
without having to get rid of aoy one of the categories such as the aggregaces
(skandha). spheres of sense (ayaJana), and eemen ts (dhatu). all of which were
part of the early Buddhist teachings erobodied in the discourses as well as in the
Abhidharma. Almost all the imporcant subjeccs dealt with in early Buddhism
arc takn up. once again not with the intentioo of rejecting them, but with a
determination co rid them of a.ny metaphysical explanation, espe~ially of existence (t11tit11a) and of non-existence (nastitva), implying permanencc (fii11ala)
and annihilation (uccheda) which the Buddha was openJy rejecting in the

Kaccayanagotttz.sutta.

Section III
The third secrion indudes eleven chapters from XVI to ~I. lt is a section
chat has caused confusion in the m inds of m any who wrote on Nagarjuna's
thought. First, many of the topics dealt with in the previous section arc again
analysed here. Thjs gave the impression that the text is repetitive and therefore
it is possible to ignore som e of its parts when presenting Nagarjuna's
philosophy . For example, the examination of action and agent (karma-karaka)
was attempted in Chap ter VIII , and a longer chap ter (XVII) on the examioation of the fruit of action (karma-phala) is induded in this seccion. Secondly,
rhis particular chaptcr (XVII) deals with the docttine of karma in a moce
positive way, asserting the existence of a moce appropriate view than the one
criticized at the beginning of the chapcer. Such ao assertion seems to go against
not only the doctrine of emptiness (uyata), as it is geoerally understood by
modern schola.rs, but also the view chat neither the Buddha nor Nagarjuna had
a view to propound.
H owever, reading the eleven chapters one cannot help comiog to the condusion that they were incended to establish the non-substanciality of the individual (pudgala-nairalmya) but not to eliminate the cooception of an individual or person a ltogether. The conception of the individual involves the
problems of bondage and freedom (bandhana-mo/qa) .ind, after defining
these , five chapters arc devoted to the problem of bondage. These five chapters
are undoubtedly commentary po a verse chat seems to have been excremely
popular among the Buddhists and which both Nagarjuna and Vasubandhu
wece cooversant with, for we find the latter composing a whole treatise caUed
Karm4Iiddhiprakara11a.6 > This verse is quoted by CandrakUci 'in his commentary, and runs chus:

30

n mPHILOSOPHY Of THE M IDDLE wAy

Actions a.re not . destroyed even by [the passage of} huodred

myriads of aeons. Having reached the harmooy of condicions and
the prope.r time, they bear fruits for the human beings .

Na pr_aalyanti karmil.t!i kalpako{ilatair api1

siimagrir(t prapya kala'!' ca phalr#nti khatu dehinar(t.66

The deoia of a permanent self (atman) by the Buddha prompted his critics
to insist that the Buddha coud not satisfactoriJy explain the problem of mora
responsibility. The present verse is only a summary of the Buddha's causa.I explanation of the doctrine of mora responsibility, not an admission of a permaoeot and eterna karma. In fact, the determinism that may appear with the
reading of the fust line, is immediately corrected with the conditionality
specified in the second line. Three mai n topics are dealt with here: the imperishability of karma (aviprat{iia) , the harmooy of t onditions (samagfi) and
the appropriate time (kala). It is therefore not surprising to see Nagarjuna inserting two chapters oo "harmony" (samagfi. XJX) and time (kala XX). after
stating the "imperishability" of karma (XVII) and denying the existeoce of a
permanent and eternal "self' (atman , XVIII). To eliminate any one of these
chapters as inesseotial co the underscanding of Nagarjuna's thought is,
cberefore, hlghly unwatranted.
Mter a clarification of the meaning of bondage (bandhana) in the context of
a human beiog who is without a permanent and eteroal self and who still concinues to wander along experiencing births and deatbs, happiness and suffering, Chapter XXII takes up a person who has attained freedom (mokfa), who
has "thus gone" (tathagata) without'having to waoder along as a person in bond~ge . No other issue in Buddhist thought has been as misunderstood and
misinterpret~d. not only by the non-Buddhists, but also by the Buddhists
themselves, as the conception of tathagata. Probably for this reason, Nagarjuoa
felt the need to begin his discussion of freedom witb an examination of the
conceptioo of tathiiggta. It will be shown that Nagarjuna's arui.lysis follows exactly the method of analysis given by the Buddha. The chapter that follows explaios the reasons for such misconcepdons (viparyiisa, XXJII).
Modern scholarship on Nagarjuna has empahsizcd the cooception of two
uuths to the c.oroplete oeglect of his explanation of the fout truths as enunciated by the Buddha. It will be shown that the two truths in Nagarjuna are not
an improvement on the four noble truths, nor a spccial insight on the part of
Nagarjuna, buc an understanding of a doctrine that is already clearly expressed
in the early discourses. This lengthy chapter (XXIV) concludes with a recogni-

INTROOUCilON

31

tion of the four noble truths and the docuine of dependent arising chat is the
foundation of the four noble truths.
The chapter on 111niii11a (XXV) is a refutation of the absolutisc interpretation
of the notion of freedom and a determined attempt to go baclc to the nonabsolucist form of Budchism enunciated in the early discourses.
Conuary to the view ot most modern scholars of Nigarjuna's thought,
Chapter XXVI (Dvadailingapankf1i) is hece presented as an eaboration of the
Buddha's owo conclusion in the KJUcayanagotta-sutta. lt deals with the Buddha's positive explaoation of how a human beiog in bondage cin free himself
from the mass of suffering.

Section IV
Nigarjuna could have concluded his ueatise with the previous secdon.
However, he was aware that his most favoritc discourse-che Kaccayanagottasutta- began with che queston regarding " right view (sammad1/th1) . Nigarjuna has already explained almost every aspect of the Buddha's doctrine and
sbown what conscirutes a "right view'' as against the "wrong or confused views"
(micchadiflht) chat appeared in the Buddhist tradition. Yet there wece some
views that che Buddha lefc aside without either asserting or denying them.
These penained to the cen, and sometimes fourteen, unexpJaioed or
undeclared questions (avyakata, Sk. avyakrta). A treatment of these questions
was nceded bcfore Nigarj_una could make a fina comme.nt about the attitude
of the Buddha, the "freed one" (nibbuta), with regard to aJI varic:ties of vicws,
whether they be right or wrong. One could hardly cxpect a better condusion co
a tcxt intended to bring about freedom from all obsessioos (prapancopaiama)
than this last chapcer. lt represcnts an expJanation of the higbcst form of
frecdom, chat is. freedom from ideoJogical cooscraints, thac che Buddha as well
as his disciples (Jravaka) had actained and wbich made Buddhism one of the
mosc tolerant religions ever to appear on earth.

AN Al YSIS OF 1li.E KARIKA

Part-I (Causa/ity and Change)

l. Conditions (pratyaya). The Buddha'~ main phiJosopbicaJ insight, as has
heco shown, was expressed by the term "dependent arising" (pa(i"asamup-

32

'THE. P!-111.0SOPHY OP 'TlE MIDDLE W AY

pada). It was presented against the four theories of self-causation, external

causacioo, both self- and external causation and non-causacion. Nagatj una, be~
ing a true disciple of the Buddha, is thetefore seen as initiating his prima.ty
philosophical treatise by categorically denying these four causa! theorjes (1.1).
According to the Buddha, the four theories imply the existence or nonexistence or both or neither, of a permanent existence, an eternal and substan.tial self (atman). That same impJication 'is assumed by Nagarjuna when he used
the term bl:iiiva (instead of bhava) to refer to the entities that are explained in
these four ways. Thus. in the very first stanza Nagarjuna's denial pertains to a
metaphysical existence (bhiiva) and not to the empirical notion of becomiog
(bhtt1N;) characterized by acising (utpada) and ceasing (vyaya).
In verse 2, Nagarjuna refers to four types of conditions (pratyaya), etnphaticaUy declaring that there is no fifth. However, therc is no categorical
denial of the four conditions, compared to the denial of the four causa! theories
in the previous ver.se. The reason for th.is is very evident. Nagarjuna, a 'ery
sophisticated philosopher, realized that the Buddha rejected the four causa!
theories mentioned in verse 1. He also knew that the later Buddh.ist disciples
(J.riivaka) attempted to eaborate upon the B.uddha's conception of "dependent
adsing'' (praiityPJarnutpada) by formulating a theory of four conditions
(pratyaya); these wece the eary Abhidharmikas. H e saw no reason for an
outright rejection of the theory of foUI condicions.
However, io verse 3, he immediately takes up a particular interpretation of a
conditioo (pratyaya) and oegates it: "The self-nature of existent:S is not found in
the condition, etc." lt is not difficult to see what is being denied here. To
Nligarjuna, it seems that some philosophers werc ioterpreting the Buddhist
(Abhidharmika) theory of four condicias (pratyaya) in terms Qf one or the
other of the theories mencioned in verse 1, whkh the Buddha himself had rejected. Nagarjuaa could not have been unaware of the face chat the
philosophers who spoke of conditions (pratyaya) at this early period in Indian
thought were the Buddhists and not the noo-Buddhists. Therefore, his attencion is now directed to these Buddhist interpreters and not to the nonBuddhists.
As meotioned earlier, among the Buddhists the only school that gave a
susbstantialist interpreracion of phenomena (dharma) during this early period was
the Satvastivada school. Their theoty of self-nature (S11abhava) carne
dangerously close to the Indian conception of self (atman). Nagarjuna was,
therefore, mercly criticizing the view of the Sarvastivada school of Buddhists,
who were suggesting a substantialist 1nterpretation of the four conditions.
'fhus, the categocical denial in stanze 3 pertains to the view of the Sar-.
vastivadins, who assumed a self-naturc or substance (svabhava) of the existent
(bhava) in the conditioos (pratyaya) .

33

lNTROOUCTION

What son of argument does Nagarjuoa present in order to deny the existence of self-nature? Murti and others who saw in Nagatjuna's method a
dialectic eomparable to that of Immanual Kant have considered self-nature
(111abhava) and other-nature (parabha11a) as antinomies. This may be true.
first refiitation of the
H owever, such a dialectic is not used by Nagarjuna in
nocion of self-nature. Nagatjuna rejects self-nature, not because ie is relative to
other-nature, but: because it is not evident (na 11idyate). The argument from
relativity is utilized'. to reject other-narure ony and not self-'natuce. ("In the
absence of self-nature, other-nature is aso not evident.") What is found hece is
a simpJe and sttaighdorwatd denial of self-nature on epistemologicaJ grounds,
even though h e does not elaborate upon tbat epistemology at this point.
Throughout the text, one finds Nagarjuna using the negated verb, na vidyata,
and somecimes the ptesent participle, avidyamiintJ. The former is often
rendeced as "not found ," and in our ttanslation preference is given to the more
epistemologically oriented rendering; "is not evident." This emphasis is dearly
evident from the manner in which he rejects "self-nature," as explained above.
In other words, Nagarjuna appears more as an empicicist than as a diaecti
ciao who merely utilizes reason. Thus, the text begin.s with a simple denia of
self-narure as something that is not evident. What Nagarjuna m eans by
evidence will be expained later oo in this essay. If this point ls kept in mind, it
becomes rather easy to understand the rest of Nagarjuna's analysis of condiuons.
Thus, in the verse that follows (1.4), Nagarjuna speaks of action (kriya) and
conditioo (pratyaya). In this case, neither the action nor the condition is
denied. What is denied is the sort of 1elatiooshlp that is assumed between
them, that is, ioherence which emphasizes identity. The denial of identity is
prompted by the fact that it is equated with "sdf-nanue'' (wabhava) which , in
its turo, was looked upon as a permaneot entity. Difference was likewise denied
because it was perceived as other-nature, which implied annihilation or lack of
any continwty.
Verse 5 takes up the definition of a coodition. A condition is such because
depending upon it others arise. However, the teason why Nagarjunarejects this
definition is not that it is not empiricaly valid, but that there j5 a rider attached
to the definition, That rider impUes that this dependence is eteroa and permanetlt. ln other words, that which is dependent and that upon which it depends
a.re substaotially coOI)ected through a telation of inherence. Hence,
Nagarjuna's queston "So long as it [the effect] does not arise, wby is it Ithe
cause] not coosidered to be a non-condition?"
Verse 6 completes Nagarjuna's genera criticism of eonditions. Most modern
translators have failed to bring out the significance of this verse, primarily
because the term artha ( = effect, fruit , result, rendered into Chinese as kuo)

his

34

THE PHILOSOPHY OF THE MlDDLE WAY

failed to attra.c t their attencion. What is denied is, therefore, not the arising of
an effect as ordinarily understood, but the arising .o fan effect that is already ex
istent (sato arthmy11) or one that is non-existent (asato arthtJJya). These again
represent the identity (1atkarya) or the oon-identity (a..ratkarya) theories of
causation presented by the Sarvastivadins and the Sautrantikas.
Verses 7-10 represent the criticism of the four types of conditions referred to
at 1. 2 as interpreted by the Sacvastivadins and the Sauttantikas. If the 1at ( existent) and a1at ( non-existent) qualifying dharma (phenomena) are undetstood
as implying "the substantia1ly (1vabha11ato) existent and oon-existent," a
quatification that Nagarjuna often make~. then it will be easy to understand
the nature of Nagarjuna's criticism. This is especiaJJy so in verse 10 where, on the
surface, it appears that Nagarjuna was criticizing the Buddha's owo statement:
"When that exists, this comes to be," (t1Jmi'f?'l .rafidam bhavatt). However,
Nigarjuna was very careful in dealing with this stacement, for in the first line
he was explicit with regard to the sort of existence he was criticizing, that existence being none other that "substancial existence" (sat-ta).
Verses 11-14 deal with severa! other aspects of the theo.ry of causacion such as
the arising of an effect from a combination of conditions. lt is indeed the coocluding line of the last verse ( 1.14) that possibly can give rise to all the
inisunderstanding regarding Nagarjuna's analysis of causa! conditions. "In the
absence of the effect, whence can thece be a condition or a non-condition." lt is
easy to interpret this statement to-meao that Nagarjuna did not accept either a
cause or an effect that is dependent upon a cause. To take it as a sirople deniaJ
of cause and effect would be to ignore evetytbing that Naga.rjuna has beeo t.ty
ing to say earliet in the chapter, regarding self-nature (s11abhava) or substantial
or permanent ex.isteoce (sal). Thete seems to be no tationae fot interpteting
this statemeot independent of the basie premises with which he set out on his
examinacion of conditions. To conclude; Whar is denied bere is not the simple
effect that depends upon the condltion or conditions for its acisiog, but an effecr rhat is either pre-existent, and therefore permanent, or oon-existent
because ie is not pre-existeot. It is also teasonable to assume that a siroilac
denial pertains not to a simple cause or condition but to a cause or condition
that produces an already existent or non-existent effect.
"Dependent arising" or casuaJity (pratftytZJamutpada) was, to reiterate , the
principle in terms of which the Buddha was able to explain the funccioning of
phenomeoa (dharma) without resorting to a conception of permaoent and eternaJ entity (nitya atman). In other words, dependent arislog explaios the impermanence (anityata) of phenomena that are dependently arisen (pratftytJJam1ttpanna) without which no identification of "dependent arising" is possibJe .
Bccause such phenomena acc dependently arise, they acc rmpecmanent
(anitya). Impermaneoce involves change and movement.

lNTRODUCTION

2. Change or movement (gat'iigata). Chapter Il of the Klin'kiis is an attempt to

reconsider this conceptioo of impermanence, i.e., change or movement. Such a
reconsideration, Jike the examination of causality, w.as necessitated not by a
de.sire to transcend it but by a desire to return to the original teachings of the
Buddha. Herc too the watets were muddied by the speculations of the Sarvastivadins and the Sautrantikas. In the first instance, Nagarjuna was compelled
to re-examine the conception of causality because these two schools wece confronting eacb other as a result of the former's recognition of a self-nature or
substance (svabhava). Why did the conception of self-na,ture emerge at all? As
we have seeo, the two schools had wrongly conceived of change and impermanence.
The Buddha descrihed time and temporality in a more empirical way when
he said tbat the arising of pheoomena, the change of what has come to be and
their cessation arc evident. 67 The three tempora I petiods of past (affta), present
(paccuppanna), and furure (anagata) in reatioo to phenomena were thus
tecognized. To refer back to the discoutse on "Linguistic Conventions" (Niruttip4tha) mentioned earlier, the Buddha even examined the three linguistic conventions (adhivacana) such as "existed" (ahost), "exists (attht) and "will exist"
(bhavissatt), pointing our that these should not be ignored.
However, in their enthusiasm to demarcate the boundaries of the three
periods of time, the Sarvastivadios and the Sautrantikas were led to an atomie
notion of time and temporaity ,68 unaware of the dangers chat lay ahead of
them. [n order to overcome the clifficulties they faced as a result of their acceptance of an atomie conception of time, the Sarvastivadins were bold enough to
admit an underlying substance that remaios unchanged, eveo though they did
not (ealize that sucha docttioe was incompatible with the Buddha's notion of
non-substantlality (anatman). The Sautrantikas, on the contrary, denying such
a substance and claiming themselves to be the fafrhful interpreters of the
discoutses, stili maintained the momeotary destructioo (h.a11abhanga) of
phenomeoa. They did not realize that their conceptioo of the momentary
desttuction of phenomena was forcing them to recognize a subtle
transmigracing entity. Hence, they carne to be characterized by their opponents
as "transmigrationists" (saf!tkrantivadin).
Nagarjuna's attempt, therefore, was to show that a speculative notion of
time and tempor~ity such as the one emphasized by the Sautrantikas was not
an empirically justifiable one. A modern critic of a simiJar conception of time,
Williamjames, has provided the following analysis:

In short, the practicaJly cognized preseot is no .knife-edge, but a

saddle-back, with a certain breadth of its own on which we sit perchcd,

36

TuE PHlLOSOPHY OF TiE MIDDLE W Y

and from which we look into two directions into time. The unit of
composition of our perception of time is a duration. with a how and
a stera, as it were-a rearward- and a forward-Jooking end. 69
The very first verse in Chapter II suggests the kind of movement or motion
that is under ctiticism: "What has. moved is not being moved."
For someone ro daim t~at what: has already moved ( =present), the underlying assumption is that eveo though there is a distinction in terms of temporality,
there is indeed the sense in which what is being moved in the present is not different from what, on an earlier occasion, was also in a state f moving. This can
easiy give rise to the view that phenomeoa are in a constant flux, a continuous
uninterrupted flow (santatz). While such an explanation may account for the
continuity of phenomena tha.t are analyscd into discrete events, it aL50 explains
the ideotity of cach inruvidual stream (santiina) . This latter idea, carded to its
extrernes, led to the roetaphysical notion of a subtle but substantial pcrsonality
(pudgala), neither 1dential nor differentfrom the a,ggregates (skandha). This is
the school of "pcrsonalists" (pudgalavada).
Thus, following the same method chat he adopted in criticizing the substantialist ootion of causality, Nagarjuna focuses his attention on the metaphysical
interpretacions of ideas of change and movement, without atternpting to deny
the cooccpts such as "the moved" (gata) , "the not moved" (agata), or "the present moving" (gamyamana) per se. After a detailed analysis of the unhealthy
consequences of such metaphysical interpretations, Nagarjuna, in che end,
specifies the sort of view he is criticizing when he maintained:
An existing mover does not carry out the movement in any of the
three ways. Nei~er aoes a non-existiog mover carry out the movement in any of the three ways. Nor does a person carty out a movement, both existing and non-existing, in any of the ee ways.
Therefore, neither the morion, nor the mover, nor the space moved
is evident. (Il.24-2~ . )

The 'existing mover' (stlflbhuto gania) is indeed similar to the 'exitcot'

(bha11a) possessing self-nature (s11abha11a) which was criticized in the previous
chaper. The denial of morion, the mover 1 and the space moved is thus not a
categodcal denial but the denial of a substancialist interpretation of these
phenomena.
Even though the refutation of che subscantiaiist view of existence (bhava,
s11abha11a) remains- the primaty coocern in Chapters I and II, the ideas examined
in Chaptet I seen1s to be predominantly those of che Sarvastivada school, while
Chapter II seems to _conccntrate more on the teocts of the Sautrantikas, who

INTRODUCTLON

37

were mote aligned with the "transmigrationists" (.raft'k.rantivadin) and the

"personalists" (pudgalavadin).
The m ethod developed in these two chapters provides a large framework oo
the basis of which the innumerable concepts are analysed in the chapters that
follow.

Part II (Dharma-nairatmya)
As mentioned earliec, Cbapters lll and XV deal with almost all the major
Abhidharma categories that are treated under general rubric "dharma." Of
these various categories, the most irnportant are aggregates (.rk.andht1) .
faculties/spheres (indriyalayatana), and elements (dhiitu). T~is is the order iJ
which these are normally eoumerated. However, Nagatjuna was interested in
epistemoJogy, and therefore ie is natura! for hlm to take up the faculties (indriya) for examination at the very outset.
3. Faculties (indriya). Chapter !Il, verse 1, refers to the six faculties and their
spheres. Yet, there is oo denial of any one of them. This roay be compared with
Nagarjuna's statement in Chapter 1, verse 1, chat refers to the four causaJ
tbeories, all of which were iostantly deoied. H owever, in verse 2, Nagarjuna
criticizes a parcicular defu1ition of ''seeing'' (darfatJa) and that definition involves "the perception of itself" (.rvat1!J(Z11af!J daranaft') . This undoubtedly is
the Indian version of the Cartcsian ':c;ogito" which led to the belicf in a perma
nent and eternal self during the period of the Upani!ad.r; 0 and continued to
flourish in the speculations of the later Indian philosophical schools.H lt rs the
defioition that p.coduced the most roetapbysical of ideas, such as the conception of the "inner controller" (antaryiimin) chat tums out to be the permanent
and eternal self or souJ (atman). Any form of perceptioo, fot them, involved
self-awareness as a necessary pre-condition, after which every ocher form of activity follow~. In fact, ater oo Nagarjuna devotes an entire chapter (IX) to an
examination of this notion of an antecedent self. Whether this view influenced
the Yogacara conception of "self-perceiving consciousness" (s11a.raf{1vedaka11ijfifina) remaios to be scen. For Nagarju.na, however, sucha defioition was not
sat:isfactory, sioce it implies the conception of a substantia entiry.
Here again. after ma.king a categorical denial of "eeing" as "seeiog itsef,"
Nagarjuoa procecds to draw the implication, as he did in his criticism of othcrnature (para-bh'iiva, 1.3) , that "if seeing cannot see itself, how caQ it see .
another?" Such a cridcism on the part of Nagarjuna would srill Jeave inract che
Buddha's owo expianatioo of perceptual experience in terms of the pciociple of
dependence (prafitya.ramutpada). Cn fact , it is for this reasoo that ater on

38

THE PHILOSOPHY OF THE MIDDLE Wt.Y

Niigiirjuoa was able to speak of visual perceprion (cak;ur-11ijflt1nt1) as a product

of causal dependence (sec XXVI.4).
The criricism of "seeing" (thrlt1na) in 111 .~ is similar to the criticism of causa!
condition in I. 5. Nigarguna assumes that the implicacion of the substantialist
notioo of "seeiog" is that "sceiog must always sec." Thus, if the Sarvastiviidins
wece to recognize a "self-natute" (wt1bhava) in "seeing" , then it could possibly
not be "not seeing" even on some occasions, for the very nature of sceing is to
~e . Therefore, when Nagiirjuna asserts that "tbere cannot be a non-seeing seeing'' (1111 tTj;afytlffliinaf!I- daanaf!J), he was merely stating the substantialist definition of the Sarvastiviidins. Hence the second statemcnt "seeing sees' (thrlt1n11f!1.
p4iyat1) becomes a merc tautology and, as such, is not appropriate. The rest of
the chapter deals with a criticism of all forms of perception conceived in the
above manner , indicating tt"lt "grasping" (upatliina), etc. wilJ remain inexplicabe in such a contcxt.
4. Aggregates (skandha) . Of the five aggregates into which the human personality carne to be anaysed in the Budd.Ust tcadition, Nagarjuna takes up only
the first, namely, materia form (ripa) . .After exp aining Niigarjuna's treatm~nt of materia form , Inada rightly remarks: "But all this does not mean that
neither rupa nor che elemencs cease to exist."'2 This confirms what we have said
about Nigarjuna's treatment of other concepts such as cause, effcct, morion, or
secing. However, lnada's expanarion of the reason for this needs to be
qualified. He maintains: "Nigarjuna is only trying to exhibit the face that any
cooception or thing cannot be describcd by rcfcrencc to a simplc cause-effect
rclationship in order to escablish its existentiaJ status.'' On the contrary, it
seems that Nagarjuna may not have aoy difficuty in maintaining that there is a
simpe cause-effect relationship betweeo d1e four primary clements (manbliuta) and material form (riJpa), so Jong as char cause-effect relatiooship is
understood as one of dependence, which was the Buddha's owo view.n Yet,
what is being introduced herc is not such a simple theory of dependence of the
effcct upon the cause.
The conception of k'iir1111ti chat Nigarjuna refers to hcre is one of the six
causcs (hetu) referred to in the Sanskrit .Abhidbarma texts and interpreted by
the Sarviistivadins as a "unique cause," that is, "anything other than itsef"
(s11ato 'nye Riiraahe111f?). H In other words, it is any cause whose self-nacure is
different from that of the effect. The four grcat element.s (mahabhuta) depending upon ~hich the material form (nipa) comes to be would be the kiir111!11 of
materiaJ form. Yet as a k'iira'fa of materia form it would be distinct from
_materia form. lt is this panicular definition of ~'iiraa that is criticized by
Niigiirjuna. His reason for denyiog it is stated in IV.2: "lf materia! form is
scparatcd from the unique-cause of materia form [i.e. the four great elemeots],

fNTRODUCrJON

39

it would imply thar macerial form is without a cause (ahet11ka) ." However,
Nagarjuna's empiricist and analytical approach does not allow him to recognize
an effect (artha) which is without a cause (ahetuka).
Existence (bh'ii11a), wbich Nagarjuna was ofreo criticizing iroplied selfexistence (s11ahha11a). The fact that the Sarvastivadins defined not only materia
form, but also the other four aggregates-feeliog, perceptioo, disposition, and
consciousness-as self-exiStent entities (bh'iiva) is evident from Nagarjuna's
statement at IV. 7. Thus, Nagarjuna's basie criticism of the Sarvastivadins in
this chapter is that they could not consistently speak of a unique cause
(karar,a), while at the same time recognizing a cause and an effect that arc
related by way of self-nature (svabhava). In other words, the notion of selfcausation (svatotpath) contradicts a "unique cause" (kara?Ja) , for it is anythlog
other than itself.
5. Elements (dhatu) . In the early Buddhlst tradition, the psychophysical personality was ~ysed into five agg,regates (skandha) in order to show that there
was no permanent spiritual entity or self (atman) as recognized by the traditional Indian phllsophers. Therefore, rhe psychic part of the personality was
analysed in detail. In order to refute the view of the Materialists that the eternal
entity is matter, not a spiritual or psychic entity, the Buddha once again analysed
the human personality in to six elements ( dhatu) with a detailed examinatioo of
the physical part of the persooality. 'rhus we have the category of eements consisting of earth {Prth111) , water (apm) , fue (tejm), air (vayu) , space (akafa) and
consciousness (vijiina).

While the cooceptlon of a "unique cause" (karafa) was iotroduced in the ex.
amination of the aggrcgates (skandha), the notion of "characteristics" (lakf al'Jtl)
is brought into the analysis of elements (dhatu}. Though the term
"characteristic" (Pali lakRhaf!a) occurs in the eady discourses, there it is not used
in the rnetaphysical sense in which it carne to be empoyed by the Sarvastivada
school. For t he Sarvastivada, a characteristic (lak,aa) represented the changing
. aspect of an entity (dhtlf'l1t4), while self-nature (svabhiiva) stood for the un
changing and eternal aspect. This particular notion of "charactetistic" needs to
be kept in rniod when analysing the cootents of Chap ter V.
A "characteristic" is cvaluated here in relation to an existent (bhiva) which
posscsses self-nature (svabhava) . For the Sarvastivaclins, this existent was a
dharma. Hence, very often we find Kumarajfva utilizing the tecm fa ( = dharma), in its restricted sense, to render bhava(yu), which is an indication that he
too was aware of the nature of the concept analysed by Nagarjuna. Nagarjuna's
major endeavor hece is to demonstrate the difficulties that ar.ise when speaking .
of charact.eristics (lak,a11a) in relatioo to eternaJ or absolute existence (bh411a} as
well as nihilistic non-existencc (abhava).

40

THB PHU.OSOPHY OF TiE MIDDLE

WA y

This analysjs becomes all the more 'important because of the way bhava and
abhava are treated hece. Nagarjuna's condusion is sigi:iificant: "Those people
of itde inrelligence who perceive the existence (t11tilva) as we11 as the nonexistencc (nastii-va) of existents (bhava) do not perceive the peaceful appeasement of the object (dratfavya)"(V.8).
In .the Buddhist text:S, including the Karika, we read more often about the
appeasement of obscssion (prapaflcopasama) . However, for the first time,
Nigarjuna introduces the nocion of the appeasement of the "object"
(dra1tavyopaiama). Why?
It was mencioned eadier that the Buddha's ruscourse oo the aggregates (skandha) was incended to refuce the nocion of a spiritual self (atman) and the
discourse on eemencs (dhatu) was meant to reject the notion of.a materia self
or eteroal matter. If this supposition is correct, we have no difficulty in
understanding the reasons fot Nigatjuna's introduction of the idea of appeasing the object. The objects of perception associated with the first five sense
faculties a.re materia!. If the Materialists wece looking for a self (atman) in matter and the Sarvascivadins wece looking fr .a seJf-narure (svabhava) in the
same, the best advice a oon-substantialist like Nigarjuoa could give such pe~
ple is to "vaporize or liquify" the object, and avoid grasping afcer ie. For Nigarjuna, there was no difference becween self (atman) and self-nature (svabhava).
While they carry the same philosophical impUcatioos, their praccical consequences are also similar, in that both lead to grasping and, therefore, suffering.
Abandoning grasping (upadana) for the object, one elin1inates the
metaphysical beliefs pettaining to eternal existence ("stitva) and oihilistic nonexistence (nastat1111). Hence the emphasis on che appeasement of the object. l ndeed, "the appeasement of the object" (drartavyopafama) is the means by
. which one can realize the ''oon-substantiatity of phenomena" (dharmanairii.tmya) and it does not mean the eiminatioo of the object.
Unles one were to keep in mind this particular cootext in which Nagarjuna
was emphasizing the "appeasement of the object," ie would be easy to assume
that here Nigarjuna was justifyiog idealism (11ijfiana11ada). Candr~rci's comments, unfortunately, lead to such unwarranted condusioos.n
6. Lust (raga) . The Buddha coosidered lust (raga) to be the cause of most of the
llls of life, the worst of these being bondage. Freedom (niroaf!a) was th'us defined
as absence of Just (vairagya). Not ooly did he speak of !ust and absence of Iust,
he also often spoke of people who are lustful (rakta) and free from lust
(virakta). Yet, all such statements were made with no assumpcion of a co.ocealed
substaoce (svabhava) or of a mysterious spiricual or materia! personality . The
analyscs offaculties (indriya), aggregates (skandh4) , and elements (dhatu) we re
. .intended to demonstrate rhe futility of such assumptions. However. the Sar-

INTRODUCTION

41

vascivadins conceived of a su.bstaoce in every element, while the Sautrantikas

posited a mysterious personality. Therefore, it became necessacy for Nagarjuna
to examine the concepts of lust (raga) as wel as the lustful (rakta). A varicty of
unsatisfactory implicatioos tbat arise out of the SarVastivada and Sautriiotika
interpretations is clearly presented in Chapter VI, with the cooclusion: "Tuus,
with or without the lustful, !ust is not established. Like lusr, aJI things. with or
without [accompanirnents] , arc not established."
Once again, we sboud not focget the fact chat the dharmas referred to herc
are those that were recognized by the Buddhist metaphysicians, not the empirical phenomcna as defined by the Buddha and the earJy Buddhists.
7. Conditioned(saft"krta). The carly discourses referred to three characteristics
of the conditioncd (sa1{Jskrta) . 76 Tbese wece arising (ul/iiida), change of what
has come to endurc (Jthitasya anyath'iil11a), and ceasing (vyaya). In a similiar
discourse, a definition of the "uncooditioned" (aiaf!Jskrta) is providcd and, in
this case, it is said chat the three characteristics mentiooed above arc not evident. It was, tberefore, easy for some of the later auddhists to Jeap to the
conclusion rhac the "uncooditloned" is aJso uncaused or independent
(apralityasatnutpanna). This, evidenty, was the intention of the Sarvastivada
commcntator, Ya.Somiua, when he stated thar the terms ''conditioned"
(sar.Mkrta) and "dependent" (prafityasamutpanna) are synonyms.n Yet, fcom
other statements in the discourses, it is elear that this was not the case. For example, white the threc teims 11ni&ca (impermanent), Ifl?!Ukrta (conditioned),
and pralilyasam111panna (dependent) occur togethcr ( though not as
synonyms), co explain the naturc of the wotld ,11 , of their negativc forms onJy
abhuta and asa'!Joskrta (together with t1Jala, and akrta 1?) arc used to
characte11ze nirvana. The negative form ofpra!ityaiamutpanna does not occur.
For this reasoo, it can be maintaincd that the term saftUkrta has the specific
meaning of "dispositiooally conditioned," and is not idencicaJ in meaning with
the term prafflya1a11'11tpan11a ("dependent").
Not only did the Buddhist mctapbysicians ignore this subde distinction and
considered the concepts of "conditioned" and ''dependent'' as being identicaJ,
they also explained the "conditioned" in terms of their metaphysicaJ notions of
substam.:e and theit speculative notion of temporality referred to carlier. Nagarjuna's Ieogtby chapter oo the subject of "The Conditioned'' (Sal'!Mkrta, VII)
draws out all the implicatioos of such metaphysics.
Mter rejecting the metapbysicaJly conceived notions of arising (utp1ida) and
along with ie all other related concepts such as "the present arising'' (ut
padyamiina) and "non-arising'' (an11tpada) , comparing all of them to the nocions of "the moved" (gata), "the not moved" (aga/a), and "the presenc moving" (gamyamiina)(Vll.14) which he had previously criticized. Nagarjuna

42

TuE PHILOSOPHY Of THE MIDDLE W AY

mak.es a vety significant statement at Vll.16: "Whatever that comes to be

dependently, that is inherently peaceful. Thecefore, chat which is presently
arising as we as arising itself arc peaceful."
In the fiest place, hece thece is no denia1 of arising (utp'iida) or that which
is presently arising (utpadyamtina). By implication, there is no denial of cessatioo (vyaya) either. Sccondly, there is no denia1 of "dependent arising''
(prafiyasam'utpada) or that which is dependently arisen (prafftya yad yad
bhavalt). On the contcary, there is a very significant assertion: "Whatever is
dependently ariseo is inherently peaceful." Explaining the "elements" (dhatu)
in a previous chaptet (V), Nagarjuna has shown how the belief in ett:cnal existence (astitva, bh'iiva) and nihilistic non-cxistence (n'iistitva, abhava) lead to
the unfortunate consequences such as grasping (upadiina) and, thecefore, suffering (dul/kha). The avoidance of such pcrspectives and the adoption of the
view that things arise and pass away dependently (prafftya) were considered by
the Buddha and the early Buddhists as well as Nagarjuna as being "nherently
peaceful" (svabha11atal/ iinla'f!").
The significaoce of the use of the term sflabliiifla in the above context should
not go unnoticed. The term wabha11a as weU as its adverbia! use, s11abhato asti,
especially when qualifying existence, was condemned by Nagarjuna
throughout the work. However, in the present verse he was willing to use this
very same term in an adjectival sense, qualifying anta (peaceful).
The concepts of "the conditioned" and "the unconditioned'', perceived in
terms of substancial cxistcnce, arc rejected herc, and the notions of arising,
ducacion, and ceasing, similacly conceived, arc also abaodoned as beiog illusory.

8. Action and agent (karma-k11mka). While "depenaently arisen (pratTtytJJamutpanna) phenomcna imply a process of natura1 occurence "uncondfrioned by dispositiooaJ teodencies1' (tJJaf!Jskrta) on the part of human
beings, "dispositiooally conditioned" (s(J1(Zskrta) pbenomena are the .results of
human deliberations (saf!Zskiira) or accions (karma). For this reason, after clari:fying the notions 9f the "conditioned" and the "unconditioned," it was natu.tal
for Nagarjuna to take a look at the notions of action (karma) and agent
(k"iirt1ka) . If these two wete found to be real in a substantialist sense, theo the
lengthy analysis of "conditioned" phenomeoa in the previous t.hapter would
appear faulty.
.
Therefore, Nagarjuna begins with a substa.ncial agent (sadbhuta kiiraka) who
performs a substantia11y existing action (sadbhuta karma) and his analysis
demonstrates chat such an agent as well as such an action, in face, logicaily lead
to a denial of action (kriy'ii) , agent (kartr) as well as a cause (kiiraf/a).

lNTRODUCflON

43

Verse Vlll.5 represents an u nequivocal assertion on the part of Na.g arjuoa

rhat such a substantialist view not only leads to the denial of action, etc., but
also co an.abandoning of discrimlnations and distinctions such as good (dht1rma) and bad (adha-rma) that are so relevant to the Buddha's conception of fruits
(phala) of life, both worldly ( = heavenly, svarga) and ultimate ( = freedom
from suffering, mok[a). Not only the goa or fruit of life, but aJso the pach that
leads thereto or all the accions that produce such fruits, would thereby be
rendered useless or meaningless (nairarJhakyaf!J). -.

The statement at Vlll.12 that both action and agent arc dependenty
(pralitya) acisen and that there is oo other perceivable mannec in which these
could be' estabished (nanyat pajiimaf? siddhikaraf!af!J) stands as an eloquent
testimony to Nagatjuna's vindication of the empirical standp.oint of the Buddha and of eary Buddhism.
'

9. Antecedent stale (of the self) (purva). If there were to be no subscantial ac:
cioo and agent, exc:ept the empirically gjven action and agent which are the
results of dependence, ho\V is it that metaphysicias carne to assume the existence of such an eteroal self or persooality?
Io Chapter IX, Nagarjuoa uodertakes to show how the belief in a pecmanent
and eternal entity arises as a result of the recognition of the exiscence of a person.ality prior (purva) to his experieoces such as seeing, heariog, and feelings
(IX.1). In other words, the Buddhist metaphysicians, foUowiog a method
similar to tbat adopted by Descartes in Western phiJosophy, were positjng a
substantiaJ eatity and theu p.rocecding to attriburc the funccions of seeing and
hearing to tbat entity. It is hard to believe that a philosopher like Nagarjuna
was unaware that the Buddha's notion of non-substaotiality (anat1114n) was the
direct result of his rejeccion of such a perspective, vety clearly expressed by the
Buddha in a passage in the Sutta-nipata: "Let bim' destroy the eocire root of
obsession, [oamely, the belief) 'I think, [therefore] I am," (mantii asm"itz). 80
This indeed is an unequivocaJ rejectioo of the ucogito ergo sum" (1114ntfi
asmz) wbich contrihuted to the substantialist thought of the Upani/ads as well
as later Indian thought. Nagarjuna's arguments shows how self-destructive
such an assertioo is. The implication of this assertion, as Nagarjuna perceives, is
that such a personality has to be separated from the experieoces that emerge
subsequently. Nagarjuoa wants to know how sucha personaity could be made
known (parjfJapyate) independent of such experiences (IX.3) thus implying
that the sum (aham a.rmz) is tlependenc. If rhese experiences can be separated
from the personaity, it follows that they could occur eveo witbout sucha persooaity (IX,4).
Having explained cettain ocher implications of this metaphysical position, aJJ
of which he considers to be uosatisfactory, Nagarjuna maiotains tbac with

44

THE PHILOSPHY OF THE MIDDLE W AY

regard to such a personality the concepts of existence and non-existence are not
validly applicable.
10. Fire and fuel (agritndhana). The futility of employing the example of the
fire (agnt) and fue (indhana) in order to ilustrate the relationship between a
substantial action and a substantial agent is shown in Chapter X. It is indeed
the. most important metaphor used by the substantialists to establish the conception of a metaphysical person. The manner in which che Pudgalavadias
utilized this metaphor is explained .in detail by Vasubandhu in the final
chap ter of his Abhidharmakoa- bh'?i!Ja (see annotation on Chapter X). Finally
he refuses to recognize those who admit the reality of a self (atman, pudgala) as
weU as those who uphold discrete substaorial entities (bhava), as people who
are conversant with the true meaning of the Buddha's teachings (X.16). It is
elear that the reference herc is not to the non-Buddhist metaphysicians, but
rather to the Buddhist metaphysidans who claimed themselves to be the tt:ue
interpreters of the Buddha-word, oamely the Sautrantikas ( =pudglavadins)
and the Sarvastivadins ( =svabhavavadins).
11. Prior and posterior ends (p'iirv'iiparako{t). The refutation of the prior existence of a substantial being or entity would stili leave open the question
regard.ing the beginning and end of things. Speculations regarding the beginning (purva-koti) wd the fina end (apara-ko) have occupied the atteation of
phiJosophers from the dawn of bistory. These speculatioas have given risc to a
wide variety of beliefs, one of which is the substantial cxistence of a being (such
as God) or an ultimate entity (such as pricnordial maner, pr~rti, sometimes
referred to as svabhava).81 Realizing the epistemological problems involved in
these speculations, the Buddha refraioed from mak.ing an.y state01ents regard~
ing such issues.
Chapter XI ,is i tended to explain the Buddha's attitude towards such ques
tions. Nagarjuna was aware that the Buddha refused to make any statements
about the prior end of the life-process. Hence his statement: "The Great Sage
has dedared that the prior end of the life-process is not known" (XI. I).
However, the Sanskritization of the Prakrit cerm anamalagga ( = "inconceivable
is rhe beginnmg") as t1na11aiiigra ( = "wirhout beginning and end") had already
appeared in the Buddhist texts that Nagarjuna was familiar with. 82 Taking this
lattet version of the Buddha's statemeot, Nagarjuna maintains that there is
neither a beginning nor an end, whereas the Buddha's own statement perta.ined to the epistemologica1 difficulties.
Yet, Nagarjuoa's ingeouity was such that he was able co indicate the Jogical
difficu1ties involved in any denial of either the beginning or the end. For he
finds that "no middle can be conceived of chat which is without beginning or

INnoouCTioN

45

end" (XI. 2). Because of such logical difficulties, he maintains that prior and
posterior as well as sjmultaneous srates (of SPf!l-s'iira) arc not appropriate, These
logkal difficulties arise primarily becauc entities are conceived of in. a rather
substantialist way. Thus, if birth were to be considered as being prior to old age
and death, and birtb as well as old age and death are substantial, thar is, existiog in their owo nature (svabhava), then there will be birth without old age
and death, which implies immorrality (XI.3). Moreover, if they were to be se!fex.istent, there would be no causa! connection between them (Xl.4). Similar
logical difficulties arise if they wete considered as being simultaneous.
While the Buddha was unwilling to cliscuss the absolute origin and end of
the life-process (saf!J-sara) and yet continued to speak of things arising and passing away on the basis of causa! dependence, Nagarjuoa bad to deal with che
notion of the life-process interpreted in a more substantialist way by the Buddhist metaphysicians. Thercfore, after ma.king the remark that the prior end of
s<1f!1-satt1 is not evident, a position upheld by the Buddha himself, Nagarjuna .
proceeds to maintain that "the pciot end of all existents is also not evident"
(sarve1am qpi bh'iivfinaf!J-pur:v'ii ko# na vidyate, Xl .8), theteby rejecting all the
views of the substanriaJists.

12. Suffen'ng (du'1kha). After analysing the nature of e.xistence and a.doping a
middle position between the two extreme views presented by his fellow Buddhist phi1osophers, Nagarjuna focussed his anention on the problem of human
suffe.ring (duf7kh11). Herc again, it is difficult to believe that Nigarjuna was
unaware of the statements of the Buddha as .recorded in the Nikayas and
Aga.mas. His analysis of suffering follows ex:actly the line that was followed by
the Buddha in the Acela-kau11pa-sutt11 of the Samyutttt-nikaya. 0 Herein, when
a disciple by na.me Kassapa quesd.oned che Buddha as to whether suffering is
self-caused (sayat?t kataf!l- dukkhaf!J-), the Buddha, without saying: " l t is not so"
(no h'etaf!l-), wbich is a formal negation, merely remarks that "he should not
speak so," or "shoud not put it that way" (ma h'evllf!J). Kassapa elicits the same
response from rhe Buddba when he questions him as to whetber "suffering is
caused by another" (paraf!l- kataf!J- dukkhaf!l-) or whethet it is "caused by both
self and other'' (sayaf!l- katan ca paraf!J- katan ca) or whether it is "caused neither
by oneself nor by another" (asayaf!J-k'ifraf!J- aparaf!l-k"iiraf!J-) and, therefote, of
"spontaneous origin" (adhiccasamuppannar?Z)
The reason why the Buddha d iscoutaged Kassapa from reflecting oo the
cause of suffering in this manrlet was that he felt thac the first two views led to
beliefs in permanence (sassata) and anojhiJation (11cch~da) respectively. In the
background in which the Buddha preached , to say that "one acts and the same
person experiences the consequences" (so karoti so pafis1111P11ediyat1) implied the

46

THE

PHILOSOPHY Ol' THE MIDDLE WAY

existeoce of an eternal soul or self, and to maintain that "one acts and another
expedences the coosequences" (afJo karoti ano parisa111vediyatt) was taken to
mean annihilation, that is, absence of any connection between act and consequence. This was tantamount to a denial of mora responsibility. Avoiding
these two extremes, the Buddha explained the rdation between action and consequences as one of dependence.
lt wouJd be necessacy to keep in mind that in the above context the Buddha
was not denying the four theories of the causation of suffering. H e was merely
stating that the theories as presented were not sacisfactory, because of the implications drawn by the metaphysicians. However, a.ter warn.ing that one
should avoid such implications and explaining the dependence f such
phenomena, the Buddha used similar linguistic expressions in order to explain
his view of the causation of suffering. Recognizing one's responsibility for one's
owo actions, he was even wiling to say: ''An action is performed by oneself'
(attan'li va kataf!t kammaf!t) 184 so long as one does not assuroe the existence of a
metaphysical agent or ignore any other factor chat contributes to the situation.
Following the same method, Nagarjuna (Xll. l), instead of denying these
possibilities, merely says that tbey are not proper (na yujyate). The reason for
this is that "if [suffetlngJ were to be considered self-caused, theo ie will not be
dependeotly arisen" (Xll.2). Here then is a distinction between self-causatioo
and dependence, a distinction based upon the assumption or the nonassumption of a metaphysical agent respectively . Therefore, Nagarjuna maintains: "These aggregates appear dependent upon these other aggregaces."
However, this latter view should not be takeo as meaning "enernal causation"
(parakrta).

Subsequently Nagarjuna proceeds co show the l ogical difficulties involved in

accepting either self-causatioo ot externaJ causation. And this criticism is then
applied to the self-causation or external causation of any other existent (bhava).
13. Dispositions (sa~kara) . The Buddha never claimed that all phenomena
(dharma!?) lead to suffering (duf?kha) . For him, all dispositions (sa7[>skaral;) or
everything conditioned by dispositions (sar.mkrta) are subject to suffering or are
unsatisfactory. 8 , Thus, after explaining the concepcion of suffering, Nagarjuna
deems it necessary to discuss the conception of dispositions (saf!tJkara). Chapter
XIII is devoted to this question .
The noo-absolutist standpoint of early Buddhism is clcarly manifest in the
Buddha's rejection of the sharp dichotomy between truth (ratya) and falsehood
(asarya) recogn.ized in the Indian philsophical tradition. lnstead of the
true/false dich'?tomy, the Buddha spoke of trutb (s{leca = satya) and confusioa
(musi = mrf'li), 86 indicating thereby that he was not advocating a notio1.1 of ab-

lNTilODUCTlON

47

solute or ultimate truth, comparable to the Atman/Brahman of the preBuddhist traditions.

Nagarjuna was faithfully following the Buddha and the carly Buddhists
w hen he bcgan his analysis of dispositions" (saf(tskiira) with a reference to chis
idea of "confusion" (mrfa). The relationship between "dispositions" and "confusion" needs to be carefully examined before any attempt to understand the
Budcha's as well as Nagiirjuna's disquisition on the nature of dispositions.
A careful reading of the early discourses will reveal that dispositions are an
ioalienable part of the human personality. In the case of. an ordinary
unenlightened person, they are not eliminated even at death. , Bence the
possibility of his being reborn. However, they arc completely eJiminated in the
tathagata when he attains parinirtia11a, that is, when he dies. Yet, there is no
mention of the dispositios being completely eliminated in the elightened
one (b1uldha, tathagata) while he is stili aive . What is achieved with the at
tainment of freedom (niroaf!a) is the "appeasement of dispositions"
(saf!Zskiiropafama). This very subtle distinction will become extremely impor
tant when we try to underscand Nagarjuna's treatment of "dispositions"
(saf!Z.rkara) in the present chapter.
William James explains human knowledge and nderstanding in the follov.'ing maoner: "The iotellectual life of man coosists almostwholly in his substitution of a conceptual order for the perceptual order in which "his experience
otiginally comes."8 7 Elaborating upon this statement, he says:

The subsrirucion of concepts and their connections, of a whole

concepcural order, in short, for the immediate perceptual flow ,
thus widens enormously our men tal panoram.a . Had we no concepts
we should live simply "getting" each successive moment of expericnce, as the sessile sea-anemone on its rock receives whatever
nourishment the wash of the waves may bring. With concepts we
go in quest ofthe absent, meet the remote, actively tuto this way or
that, bend our experience, and make it tell us wicher it is bound.
We change its order, run it backwards. bring far bits together and
separate near bits, jump about over its surface instead of ploughing
through its continuity, string its stems oo as many diagrams as our
rnind
frame.

can

Unless we attribute "omnisc~ence" (sproajata) to the Buddha, and chat is

knowledge of cverything that has occurred, is ocurring and will occur in this
world, a knowledge he refused to claim for himself,88 we may end up turning
him into a "sessilc sea-anemone," if we are to deny bim the need to conceptualize. In this process of conceptualizing, in "putting things together'' (which

48

THE PHILOSOPHY OF THE MIDDLE WAY

is the litera! meaning of sa'T!"Riira), our interests play a dominant role. lncerests
arc easily converted to likes and dislikes, and tbese lattet arc causes of most
human suffering. We are, therefore, in a double-bind. We need the S!Jf!Jsliiiras
in order to live. On the contrary, they can contcibute to most of out sufferLng

(du/?kha).
Realizing this face, the 'Buddha emphasized the need to pacify our dispositions rather than eliminate them competey. Thu~ , on the basis of expericnce
we come to know that things arc "depcndenty arisen" (prafftyasamutpanna)
and then adopt the view that in the dim past this may have been the case and
thac in the future it may be che case.
In the cyes of the Buddha this represencs a moce comprcheosive and,
thereforc, amore appropriate view (samyag-dr1ti) rathcr than the morc limited
views: (1) which says: "Suffering is self-caused" (svayaf{Jkfla'f?i), which is the
resut of our believing in a self (atman) to the exdusion of every other factor,
and (2) which says: "Suffering is caused by another" (dul/khaf!J parakrtaf!J),
which is, in somc sense, the rcsult of out reluctance to admit our own rcsponsibility. ln both cases, out likes and dislikes havc domioated our dispositions,
and heocc our pcrspeccivc~ push us in two cliffctent d.irections. Such dispositioos1 domioated by our lik.es and dislik.es, evenrually misead us regarding
many of our experjences and thereby contribute to out suffedng and frustta
tions. In the Buddha's Yiew, therefore, the cessarion of sufferiog is synonymous
with ''non-grasping" after views89 which comes about as a result of the appeasement of dispositions.90 Cessation of suffering is not synooymous with not having views or not having dispositions. Ratber, it is synonymous with the appeasement of dispositioos.
lt is very appropriate, therefote, that Nigarjuna decided to wcite f chapter
on the dispositions (sa1!1-Skara) after his analysis of suffering (duh.kha).
However 1 what is m oce important is that this chap ter is eotirely devoted to an
examination of "views" (dr'!tc) as we as of the condition that give rise to
"wrong-" ot "coofused views" (mithya dr1ft), narody, confusion (mrfii). In
fact, the term saf!Jskara ~curs in the first verse only.
The encire chapter is devoted to an examination of the notions of the "existent" (bhava), the ''non-existent" (abhiiva), "self-nature" (111abha11a), etc. and
the mannet in which these could be avoided by adopting the conception of
"emptiness" (Junyafii), wfrhout allowing that notioo of emptiness to be an
obsession. Heoce his condusion; "The Victorious Ones have announced that
emptiness is the reinquishiog of all views. Those who are possessed of the view
of emptiness are said to be incorrigible."
We have rendcred the tetro ni4sara1}a occurting in the above verse as "relinquishing" in order to bring out the specific meaning that Nagarjuoa probably
had in miod. Many ioterprcters of Nigarjuna havc explained "erriptiness"

INTRODUCTIO!ll

49

(fUnyafij) as a "provisional view," theteby impying thac the uJtimate truth is

beyond conceptualization.91 The foregoing analysis of che Buddha's as well as
Nag2rjuna's thoughts would mean that they inde~ did noc recognize a "noocoocepruaJ truth or reality." For them, thcre is oo way in which a ''truth" could
be understood oon-conceptually, because, as meotioned earlier, truth in its
most comprehensieve sense pertains to statements and thus involves conceptuaization. However, conceprualizations can be compreheosive and therefore
right (sa1nyak), or limited and confused and therefore wrong (mithyi), and
these depend upon the amouot of prcjudice thac has gooe into the formulation
of the conceptS. Thus, "emptioess" is a "view," a view not without identification, but which is identified wirh "the empty" (idaf!Z .'fitnya1p) . lt is a view that
hcps the indjvidual to attain freedom from views and upholding it as the absolute or ultimate truth without any reference to " the empty" would be the last
thiog either the Buddha or Nigarjuoa would advocace.

14. Association (Ia1'(Jsarga). The dispositioos arc inscrumental in our forming of

views on the basis of experience. It was also poinced out tbat if we wece not to
formulate such views we wouJd be no different from the sessile sea-anemone.
Dependent arising (prafftyasamutpiida), impermaoence (anityata), nonsubstaotialicy (an'iilmatii), emptiness (inyata), etc. arc all coocepts which
would be .rendered meaningless uness they wece to be identified with the
"dcpendeotly" ariscn," "the impermanent ," "the non-substantial," and "the
cmpcy" given to us in experience. They arc vicws or theories formuated by
stretching out our experiences into the dim past as wel as the future.
However, some of the Buddhist mecaphysicians, as explained earlier, had
complicated the siruation for Nigarjuoa by chcir analysis of experience into
discrete momentacy events. Such as analysis, which led to the formulation of
the metaphysicaJ oocion of self narure (S11abh'iiva), also created other problems
chat these metaphysicians were never able to solve satisfactorily. O ne of rhem is
che distioction t hey made becweeo mi nd (citta) and matter (mpa). In chcic
ultimate coruticution, chese were explained as having completely discinct
narures (s11abha11a). Mind was considered co be im-macerial (arnPa) and maccer
was looked upon as bcing non-mental (a-citta).
Avoiding such a reductive analysis, early Buddhism was able to maintaio tbat
depending upon the eye. the visible form, and consciousoess perception (saf(l
jfla) ariscs.92 However, follo:wing that reduccive analysis, the Buddhisc
mctaphysic.ians experienced difficulty in explainiog noc ony perceprual ex
perience, but also conceprual formuJation of such pen:eprual expericnce.
This accounts for the need to have a chapcer on "associacioo" (sa1'(1Sarga)
following che chapcer on the "dispositioos" (stlf(l.fkiilrl). How is it possibc to bring

~o

THE PHTLOSOPHY OFTHJl. MIDDLE WAY

together into association (saf?Zsatga) the objecc (drlJ.!favya) and the subject
(draf!r) in order to have a visual perception (darana)? Nigarjuna begios this
chapter with a denial of sucha possibility. Io doingso he is specifically denying
the possibility of an association of evencs that ace considered to be ucimacely
distinct, and this idea is being emphasized in Kumarajiva's translation of
XlV. 1. The problems of idencity and difference chat arise as a result of such a
reductive analysis are once again deatly brought our in this chapter.
15. Selfnature (s11libh1111a). The problem of associacion (saf'!JJarga), discussed
above in the specific context of seer, objecr of seeiog, and seeing, arose as a
result of admittiag a self-nacure (svabhava) in each one. of these phenomena.
This provided Nigarjuna with an opportunity to come into grips with the most
difficulc issue he had to deal with, self-nature or substance. In our earier
discussions we have shown how the Sarvascivadins utilized thls conception to
explaio the relationship between a cause and an effect, Nagarjuna's basie argument against this notion of self-nature js that it contradicts the conceptioo of
the occutrence (saf!Jbhava) of an event depending upon causes and conditions
(bett~, pratyay4). Nagarjuna's understanding of self- namre is that it is not
made (akrtaka) by anything clse. It is not dependent upon causes and conditions for its existeoce; hence independent. A "caused substaoce," according to
bim, is a contradiction in terms. This analysis should, therefore, be suppleme.nted by bis analysis in Chapter I. As we have pointed out there, Nagarjuna was not denying eithe.r dependently arisen pbenomena or dependent atising. He was merely showing the inconsistency in explaining causaly conditioned
phenomena in terms of self-natute. lt is in the present chapter tbat he is giving
a definition of self-nature that contradicts the ootlon of dependent arising or
causation. He says: "lndeed, an unmade self-nature is also non-contingent
upon another," (akrtn'm.al/ svabhavo hi nirapekfai{ paratra ca) (XV.2). The
argument in Chapter I is tben repeated to show that in the absence of selfnature, thete can.not be other-natute (parabhava). Buddha's faroous discourse
to Katyayana, discussed at length at the beginning of this lnrroduction, is then
quoted in order to reject the "existent" (bhiva) or "self-nature" (svabhava) and
the "non-existent'' (abhava) or "other-nature" (parabh1iva). These thcn arc
aligned with views regarding existence (astitva) and non-existence (nastitva) .
Existence (astitva) is funher defined as the original or primordia existence
(prakrtt), a conception developed in the Siinkhya school of Indian philosophy
which had close affinity, if not identity, with the Sarvastivada conception of existence . The empirical and logicaJ difficulties iovolved in this conception are
then laid bace. Reiteration of the face that the beliefs in self-nacure and other:nature, in the ccistent and the non-ex.isteot. in existence and oon-existence,

lNTRODUCTION

)1

lead to bcliefs in permanence and aonihilation respectively provides a co.ntlusion to this rather significanc chapter.
The topics taken up for discussion in this seccion (Chapters Ul-XV), as
pointed out above, deaJ with elemencs of experience (dharma) which were
originaly explained in terms of dependence , but which were complicated by
the introduction of the notions of self-narure (svabhava) and other-narure
(parabh'iit!a), of metaphysica identity and absohae difference, by some of the
. later Buddhists. Nagarjuoa's attempt in this section was mainly difected at get
ting rid of the cooception of sclf-nature or idcotity (the notions of other-nature
or differcncc falling apart as a result). This is the way in which he attempted to
estabish the noo-subscantiality of clements (dharma- nairatmya). The nonsu bstantiaicy of the human personalicy (pudgala-nairalmya) turns out to be his
next cooccrn.

PART 111 (Pudgalt1-nairal11J.ya)

16. Bondage and release (Bandhana-mokfa). The recognicion of a permanent
and eternaJ self (atman), even though it raised epistemologicaJ difficulties,
enabled the early Indian thinkers to expain many ideas ike karma a.od survival, bondage and release, in a more comfortable way. The Buddha's deniaJ of
sucha metaphysicaJ enticy gave rise to cnormous philosophical problcms. The
questioo was often raised: "Which self will be touched (or affected) by actions
performcd by a non-self. "9l So long as the docttine of dependent arising was
understood properly, the Buddhists coud consistently descirbe the manner in
which a person may be said to perform an action and reap its consequences.
However, as emphasized earlier, the Buddhist metaphysicians created moce
problems with theit interprecacion~ of dependence chan they solved. These
n1ecaphysical views were foremost in Nagarjuoa's miod when, after examining
the probems relating to suffering, etc., he proceeded to analyse the problems
of bondage and release (bandhana-mokfa), action and consequence (karmaphala), and so on.
Bondage (bandhana) can be of severaJ sorts. Human be.ings arc fettered by
the pleasant objects they perceive, the ideas they form, and finally the process
of becoming (bhava) itself. Craving for becoming (bhava-tr!?J'i) is looked upon
in Buddhism as one of the most troublesorne bonds. This craving for becomiog, while leading to suffering in the presenc life, kceps the individual wandering in samsaric existence, subjecting him to repeated births and dearhs.
Dcach, pcrsonificd as Mara, carries with it a snare (pasa) which very few

52

THE

PHILOSOPJY OF THE MIDDLE W AY

humans can escape. The present chapter, therefore, focusses its attention on
this bondage to the life-process.
Wh.ile the discussion of the Sarvastiviida conceptio~ of sclf-nature (wabhava)
1ingers along in the ncxt few chapters, the Sautrantika theory of a
transmigrating personality (pudgala) emerges into prominence in the present
section. When it is said that ''dispositions transmigrate" (.raf!U/iaral/ saf!Uaranft), what is assumed is chat there is a subtle essence in the dispositions that
enabe them to be perpetuated. Niigarjuna's argument herc is more diaectica.
lf dispositions arc permanent, then therc is oo point in speaking of their
transmigration. For, transmigration implies moving from one position to
anothcr, disappearing in one place and appearing in another. If something is
permancnt, it is always present and thcre- is no quescion of its ceasing and arising. On the conuary, if thiogs arc impermanent, in the sense of being compkcdy destroyed (11ccheda), they will never transmigrate. Nagarjuna,
therefore, .maintains that jf a human being is Jooked upon in the above manner. ir is not possible to speak of his 'cransmigration (XVI.1).
The impression one gets from the available uaoslatioos of XVI. 2 is that
Niigarjuna rejects the theories ofaggregatc:s (skandha), faculties (fiyata.na), and
elements (dhitu).94 Yet, what is <leady stated herc is che early Buddhist posi
tion: "lt may be assumed that a person transmigrates. Yet such a person,
sought for in the flvefold way, in the aggregates, spheres and dements, does
not exist. Who then will transmigrate?"
Thus, the transmigration chat is denied is chat of a subte personality.
However, if transmigracion is understood as the contiouation of the faccors of
the human personality on the basis of causal dependcnce, Nagarjuna may not
have any objection against it.
The notion of bondagc that is ,criticizc:d tutns out to be the bondage of a
substaotfaJ entity to such things as dispositions. Jt is similar co che ootion one
finds in the Indian tradition where the permaoeat "self' (atman) is said to be
in bondage co the psychophysicaJ personality which is impermanent. Hence
Nagarjuna's argurilent that anyth.ing that is of the nature of arising and passing
away (utpada-vyaya-dharmin) is ncither bound nor rcleased. What is being
criticized herc is not the simple notions of bondage and release but those that
takc into considcration a substancial subject and its attributes.
The cooclu~g v~rse could easily lead to much misuoderstanding if the
significance of the relativc terms "whcrc/ there" (yatra/ tatra) are ignored. The
context specified herc with these teative terms is what ca.me to be discussed
befre, namely, the assumption of a substantial su,bject and the attribution of
various attributc:s to it. Thus, in a context where,some substantial subject is at
tributed with something called freedom (nirvi~a samliropa) or is strippc:d of
the lifepr91=ess (st1f?Zfarapdar1a'!a), therei there is no sense .in mllkiog._a

lNTllOOUCTION

53

d.iscrimination (11ikzlpr1) bctween frecdom and bondagc, for what is real, namcly,
the substancial subject, will rcmain the same. Or else, where nirvana is "teified"
(sam'iirop11) and the life-process is eliminated (apakar1a1Ja), therein too there is
no sense in ma.king such discrimina.tions. These being totally different from
one another, the knowledge of one would have oo relevance to the understanding of the other.

17. Action and conse1111ence (karma-phala). Buddhism , in coiurast to the

theistic rcligioos of !odia, advocated human responsibility in the ease of boodage as well as release. Chapter XVII chat follows therefore deals with the notions of human action (karma) and its consequences or fruits (phala).
Ioada's analysis of this very lengthy chapter is very confusiog. He assumes
that verses 1-19 contain the popular vicws on karma, while in verse 20 Nagar-.
juna finally explaios the uue position of the Buddha who spoke of "emptincss"
(unyafii).9> Y et, verses 2 and 13 unequivocally attribute certain views to the
Buddha, as well as the Pratyeka-buddhas and the Sravakas. The contents of this
cbaptcr therefore deserve car:eful scrutiny.
The doctrine of karma is clearly statcd in XVU.1: "Sclf-resttaint as well as
bcncfitcing others-this is the friendly way and it constitutcs the seed that
bears fruit, herc as well as in the ncxt life."
lfNigarjuna was a Mahayanist, as many have poraayed him to be, then he
certainly could not deny this "friendly way" (mai/raf!' dha~). for ocherwise
he coud not qualify as a bodhisatt11a. Rcstraining oneself and benefirring
others arc actions chat necd to be recognized by a bodhisa1111a. These arc herc
dcscribcd as bcaring fruit (phala) in this world as well as in tbc next. Nagarjuna
was not naware of the fact that accord.ing to the Buddha, actions arc to be
dcfined in terms of volition (cetana) . Whic voition itself could be considered
an action, anything that is volitional also falls undcr the catcgory of action. This
is the implicat.ion of the Buddha's statcment inAgultara nikiiyfJ'J6 and Nagarjuna is scen to elaborate on this statement at XVll .2.
Nagarjuna asserts that ccording to the Buddha thcre are two main typcs of
karma: volition and volitional. These arc further analysed into a varicty of karmas that werc also recognized in the early Buddhist rradition . He gives no indication that all thcse karmas arc not real in the sense that they do not produce
fru.itS or conscqucnces. However, in XVIl.6 he raises a quescion which clearly
embodies the patticular form of inquiry carried out by the Buddhist metaphysi
cians with which he disagrecs. The inquiry is as follows: "Docs karma rcmain
even at the time it has not acta.ined maruriry? (Ti/fh11Ji apakakii/iit?).
Therc can be itde doubt as to who would raisc such a quescion. WhiJe a
pragmacic Buddhist may say that karma is what it is because it produccs consc-

THE PHlLOSOPHY OF THB MI DDLE W AY

qucnces or fruitS (phala), a substantialist could not resist asking the question:
"Yet, does not the effect (phala) pre-ex:ist, before karma reacbes its matur ity?"
Such metaphysical inquiries, as pointed out earlier, Jed to the belief in ao
underlyiog substance (wabha11a) , which was criricized and rejected by Nagarjuna in Chapter I. And hece Nagarjuna is ooce again asserting the view that
such inquiries lead to the beJief in permanence (niiya~) . If one were to reject
sucha notion of permanence, theo karma and effect are separated in sucha way
that once the karma ceases, ie will not producc any consequeaces (niruddhaf!J
sal kif!l phalaf!l janay#yatt) . Thus we are back again in the permanenceannihilation (fiif11ata-uccheda) syndrome.
Verse 7 introduces the notion of a scries (saf!ll'iina) upheld by the atomistic
Sautrantikas, and the difficulties this generates are chen examioed in che few
verses tbat follow .
lt is rather uofonunate that this new situation arisiog from the metaphysicalJy
oriented q ucstion raised in verse 6 came ro be igoored by those who dealt with
verse 12. The term efa (this, such) in XVll.12 refers specificaJly to the sort of
thinking (ka/pana) involved in XVII .6 and Nagarjuna maintains that such
thoughts engeoder a multitude of insuperable difficulties (ba4a11rli ca mahata
ca dofiif!). lt is this particular way of thinking chat is considered to be inappropriate (nopapadyale) .
lndeed, at XVll.13, Nagarjuna suggests anotber way of thinking (ka/pana)
which is more apptopriate and which was extolled by the Buddhas, the
Pratyeka-buddhas and the Sravalcas: Lilce an imperishable promissory noce, so
is debt as well as action. lt is fouold in terms of realms and indeterminate in
terms of prima! oarure. "
According to this, karma is imperishable like a promissory note. Ooe's debt
(rtJa) remains effective at least as Jong as the promissory note lasts. Even though
there is no continuiry of karma (and, it in this case, borrowing), that is, it does
noc contioue in any subcle or substantiaJ way, the responsibility for that karma
cannot be deoied once that karma is performed . The Buddha, the early Buddbists, and Nagarjuna were not prepared to say that the promissory note one
signs is uruea and therefore to be igoored. The responsibiliry and commitment
remains long after the docurnent is sigoed (maybe even if the document were to
be lost or desrroyed).
The idea that one is responsible for one's own actions has been emphasized
by the Buddha. A statement in the Dhammapada reads: "Neither in the sky
nor in the middle of the ocean nor haviog entercd into a cleft of the mountains
is tbere a place on earth scen remainiog where a person would be releascd from
his cvil actioos, "97 The cxistence and the popularity of a sirnilar statcment
amoog the Buddhists who preserved their literature in Sanskric has aJready
been referred to. The.re is itdc doubt that Nagarjuna was aware of this st:ttement. This conception of the imperisbablc nature of karma thus turns out to be

lNTRODUCTION

55

an imponant conception in Nigarjuna, primarily because he was not prepared

co accept the notion of substance (svabhi11a) or self (atman) to cxplain this pro
cess nor was he wiUing to deny the effectiveness of karma with che denial of
substance or self. As such lnada's statement that Nagarjuna "with equal force
condetnns any idea of an iodestruccible continuing action {avipraf!afa)" is SUI
prising.98 Neither the Buddha nor Nagarjuna nor even Vasubandu (who compiled the Karmasiddhiprakraf!ll) were saying that karma itself remalns m
destructible. Avipra1{iifa-karma or the imperishable action refers to the responsibility a person has for any karma that he pe.cforms and how that karma will,
dependirrg upon circumstances, bear fruit (phala). The simple notion of
human responsibility is what is upheld hece, not the metaphysical notion of the
fruit or result that lies hidden and gradually attains matudty, as was implied in
the question raised by the Buddhist metaphysiciao,
Therefore, without any hesitation Nigarjuna atttibutes the conception of
the imperisbabe (avipraf!ll4) karma to the Buddha hirnself: "Emptiness,
bowe-aer, is not annihilation; life-pt<x:ess is also not permanence; imperishability
is of ac~on- such is the doctrine taught by the Buddha." (XVII.34)
Themost significant assertion here is that the rejeccion of permanence and
annihilation and the acceptance of emptiness and saf{lsiira (or the life-process)
do not imply the rejection of the relationship between action (karma) and con
sequence (phala) . The imperishable nature (avipr4f!iifa-dharma) of action
merely implies the possibility of action giving rise to coosequences, and this
need not involve the notion of an underlying permanent substance in action.
The three vecs~ chat follow arc criticaJ of the concepcion ofkallna dut is based
on the recognition of self-nature. Such a conception, as explained at XVII.24,
conflicts with:' all che accepted conventions (vyavahara) and would imply the
denial of merit and demerit (pu11ya.piipa) and such other dis~inctioos
(pravihhiiga) .

Action (karma) , looked upon as something subsr.antial (sviibhavika) , not only

implies the production of a result (11ipiik11), which is already existing in marui:e
form (vipakva)(XVII.25) , but also goes against the admitted purity or impurity
of action. If action has its own nature, then defilement (k/efa) also will have its
owo nature and how these two natures could come together will never be
satisfactorily explained (XVIl.26-27).
Moving on to XVII.29 without keeping the above definition of action (karma) in mind , it is easy to assume that Nagarjuna rejects action as som ething
dependently ariseo and, therefore, there is neither action nor agent. On the
contrary, verse 29 simply rejects the possibility of an action being dependently
arisen, if that action were to be substaotlal (svabhavik.11) .
Thus the denia of action (karma) , agent (kart[), and consequeoce (phala), as
well as of one who experiences the consequences (bhoktr) . comparing them to
created forms (nirmilakiikiira) or to mirages and imaginary eotities, needs to be

56

TH!l PH!LOSOPHY OF THE MIDDLE WAY

understood on/y as refem'ng to the 1ub1111ntialconceptions of these phenomena

and does not reprcscnt an unqualified dcnial of such phcnomena (dharma).

18. Self(iitman) . The exa,minacion of self nature (svabhava) was undertaken by

Nigarjuna after a dctailed analysis1of all factors of experience (dharma), such as
aggregates, spheres, and elements. The subjects coming under the two previous
chapters (XVIXVII) pertained to bondage and freedom, action and consequence. Therefore, a close scrutiny of the notion of a personal self (atman) and
all other concepts wociated with it, such as selfishness, idencity, or pride, was
coosidered to be relevant. Hence the subject-matter of Chapter XVIIl.
The c.onception of a permanent and eternal self (atman) arose in relation to
the conception of the psychophysic.al persooality (n'iimariipa). The untenabiity
of the fortner has already been alluded to in Chapt.ers VIII-XI. In the present
chapter, howcver, a further question in tclation to the notion of self needs to be
examined, and that pertains to the mannet in which the notion of self (atman)
leads to bondage (bandhana).
.
Nigarjuna begins this chapter with the assercion that if the self (atman) is
identical with the aggregates (skandha). then it will be subject to arising and
ceasing (utpadtJvyayabhag ). If it is different from the aggregates, chen it will
not have the charactcristics of the aggregates. Having raised such qucscions
regarding the existence of the self, Nagarjuna proceeds to show that it is the
belief in a permanent and cierna/ self that gives rise to notions of possession
('iitf1Jlya). Absence of possessiveness (nirmama) and of pride (niraha'f!'kara) a.re,
therefore, the inevitable consequences of the appeasement (fama) of that beJief
in an eteroaJ self (XVIII. 2). For similar reasons, the belief ~at there is a person
who is without self1Shoess and pride is also not appropriate (XVIll.3).
The use of the z/i-formua at XVIIl.4 as aham-iti and mama-iti is imporcant
in thatit implies the denial of"theories" pertaining to "oneself' (aham) as well
as "self-possession" (mama), rathe( than the simple reflexive uses of these
. terms. When such metaphysical views arc abaodoned, gmping (upiidiina) as
well as rebitth (janma) arc avoided. The cessation of the defilements of action
(karma-klefa) is then declated to be release (moll!a). The vikalpa that leads to
such defilcmencs of action is, therefore, not any and every focm of conceptuaHzation, as some of the translations seem to suggcst, but only the
discrimination or thought of substantial (s11iibh'ii11ika) entities such as 'I'
(aham) , rcjected at the bcginning of this chapcer, and subscantial evenJS like
action (karma) and effect (p/Jala), criticized in the previous c.hapcct. The belief
in such substantial entitics and events gives rise to the feeling of "possession" as
"this is mine" (mama), whiclr in ruro produces obsessions (prapanca). Such
obsessions can be preventcd by the petception of emptincss (Unyatii) relating
to the notion of "self'' (itn;an) referted to above. To speak of"emptiness" apart

lNTRODUCTION

57

from this conteit is to make it not ooly "nihiliscic" (ucchedt1, XVIII.20) but
aJso absoJutistic and hence a metaph)'skaJ "view" (drffi, XVIII.8).
A super:ficiaJ glance at XVIII.6 may leave the impresion tbat Nigarjuna
viewed truth or reality as being beyond conceptualization. However, a more
careful consideration of the contents, especiaJly in the light of the teachings of
the Buddha as embodied in che ''discourses" would indicate that this is not the
case, Nigarjuna seems to have been fully cognisant of the Buddha's use of the
term "self'' (Pali, atla; Sk. 'iitmt1n) to explain individuality, and his atcernpt to
reject a metaphysicaJ entity when he spoke of "no-self'' (Pali, flmlllfl, Sk.
anatman). This does not involve two languages: a ptovisional ot ordinary and
philosophical.99 Tt is a question of two definitions. lf the "elf' ' is defined as a
permanent and subStaDtiaJ entity, the Buddha was ready to negate it with his
conception of "no-self'' (11n'iitman). lf it was not defined as such, he bad no difficulty in utilizing that conception in his discourses. Nigarjuna's understanding of the Buddha's intentions is clearly demonstrated in the first line of
XVUI.6, when he said: "The Buddha's have made known the conception ofself
and taught the doctrine of no-seJf." When, in the second line, Nigarjuna
maintained: "They have not spoken of something (kafcit) as the self or as the
non-self," he was certainly denying the conceptions of self-natute (svabhava)
and other-nature (parabhiva) of phenomena admitted by the Sarvastivadins
and the Sautrantilcas respectively.
With the above statement Nagarjuna couJd have concluded his chaprer on
the "self' (atman). However, there was one mote significant issue to be resolved. Up to this point he was discussing an embodied self, a self assoc.iated wirh a
psychophysical personality. The question regardiog the self that is freed from
the psychophysical personality aso had to be examined, for it was the belief of
the substantialists tbat when a person attains freedom his permanent and eternal self, dissociated Erom the psychophysicaJ personaUty, concinues to exist after
death. The two verses that follow (XVIIl. 7-8), therefore, arc iotended to oplain the Buddha's view regarding the nature of a person when be attains
parinirv'iir,a.
What happens to the freed person al death was clearly expressed by the Buddha. He ceases to exist, is not rebom, his birth has waned (khi'!'ii j'iitt), and there
is no further existence for him. 100 Yet, if someone were to ask him the question
as to whetherthat person exists in some form after death (param marmfa), the
Buddha was not willing to say anything, primatily because there was no
episcemological l>asis oo which'any predication can be made. 101 With verses 7
and 8, Nigarjuna is attempting to state this very same idea. "When the realm
of thought has ceased, that which is to be designated aso bas ceased."
"ReaJm of thought" (ciJltlgocara) that has ceased (nivrtta) can refer to the
pc~son who is frecd (Pali parinibbultl, Sk. pari-nirvrta) without his thought being (e-established (appati{(hitena cittent1). 102 The difficulties that would arise if

58

nrn PH1tOSOPHY OF THE MIDDLE WAY

someone were to ask the further question as to what happcns co him when his
thought proccss has ceased or is not re-established is thcn e.xplained. "That
which is to be designated has also ceased." This is ideocical with the Buddha's
statement: "That by which one shouJd spea.k of him does not cxist for him"
(yena naf!l 1)(Jjj"u laf!l tarsa n' aJtht). 0~ Indeed, the term dharmata is used in the
verse to refer to the narure of the freed one who has passed away. lt is not possible to ~en whether he has ariseo (utpanna) after death or whethcr he has ceased
to e.xist (niruddha) after death.
One of the epithets by wh.ich the "enlightened one" (buddha) carne to be
described is tathagata, mcaning "one who has thus gone." This term gave rise
to much controversy probably because of the metaphysical implJcations of the
term "thus'' (tatha). And it is interesting to note chat in the early discourses
whenevcr the qucstion regardiog the nacure of the freed one after death is raised
the term used to rc:fer to him is tathiigala.
The term tatha ("thus" or "such") involves the epistemologicaJ problem of
"reality," as opposed to no tatha ("not thus"). Hence the sccondary derivative
taccha (Sk. tathya) carne ro be used in the sense of what is true or reaJ. 04 Jt is,
therefore, not surprising to see the term tathiigata being utilized when ever the
quesrion regarding the de1ti'ny of the freed one is raised, 1os for in the eyes of the
ordinary man no ocher question would be more important thao the ultimate
de.rtiny of the freed one. AJ; meotioocd ear iec , for cpiscemologica rcasoos, the
Buddha cefused to make any assertions, either positive or negarive, ccgarding
this problem. Nigarguna's application of the fourfoJd negation (catu1kot1) to
the conccption of "suchoess" (tathya) comes immediately after his discussioo of
the nature (dht1rmal4) of the frced one who has rcached his fina destiny. It is
the same context in which the Buddha himself applied the fouod negation.
The four-corncred negation is always uscd by the Buddha to avoid metaphysics,
and the desciny of the lalhiigata was one of thosc popular mctaphysical isrues.
Nagarjuna could noc have been uoaware of the metaphysical assenions of the
prc-Buddhist thiokers who assumed chat a "freed self' ('iitman) becomes uoited
with the all-pervading universaJ self, the reality in cverything (saroattt). The
dangcrs involved in the Sa,-vastivada concepcion of substaoce (1vabbava) in
everything (saroaf?Z) and how such an idea could influence the intcrpretation of
a ta1hagata or tathya was, indeed, evident to Nagarjuna. Thus, we have ewo
metaphysicaJ issues combined herc-one of "everyching'' (1ar11"f!l) and the
othcr of tathya-aod the fourfold negation was the only reasonable soJucion
that Nagarjuna could provide. To explain the conception of "suchness"
(lathya) going beyond the context of the problcms of "everyching" and che
"thus gooe one" (talhiigata) after death would undoubtedly lead to a discortion
of the philosophicaJ standpoint of Nagarjuna.

1NTRODUC1JON

59

As mentioned befote, the Buddha was not williog to speculate on the nature
of the freed one afcer death (param-marafa) but that he was willing to speak
positively about what happens to him at death. In a passage in the Agutt4ra
nikaya the Buddha's understanding of this state is succincntly expressed in the
following verse: "This is the last body and is the essence of the higber life . In
regard to that this perfect knowledge has arjsen without depending upon
another." (Asekhafiii'f!ll'f(l uppana'l(J antimo ya'f!' samussayo, yo saro
brahmacariyassa tasmzftJ. aparapaccgya.) 106 This realization is elsewhere explained
in a stock passage: "Birth has been eliminated, the higher life has been lived,
clone is what needs to be done and there is nor aoocher of this flife) ." 107
This is the highes.r realization that one can attain and is referred to in the
discourses as anfUi. 10s lt is the result of the moraJ perfection one attains by be
ing freed from the three poisons: greed, hatred, and confusioo . lt is fina nirvana, and may jusdy be called the ultimate mora truth about the world (see p.
15 above). As such it is to be realized by oncsolf and is not a state to be known
by depencling upon another (aparapaccaya ).
When, therefore, concluding the discussion of the freed one after death,
Nagatjuna proceeded to speak of tattva ("reality"), he was not refeniog to an
"ultimate truth" per se but to the realization and attaioment of freedom from
birth. Hence he asserts that the knowledge of this truth is not dependent on
another (aparapratyaya). lt is peaceful (fiinta), uoobsessed by obsessions
(t1prapancita). and, hence, non-cliscrimioacive (niroikalpa) and noncontradictory (ananartha) . Nirvikalpa does not necessarily mean the absence of
t he subject-object discrimination. It means the abseoce of any discrimination
based upon one's likes and dislikes, one's obsessions. Conceptions of identity
and difference, permanence and annihiliation arc theri rejected as beiog part of
the Buddha's teaching. This is because thiogs arc recognized as being
dependencly areisen (prafitya ... bhavatt).
The conclusion of the chapter represeots Nagarjuna as one who remains
aloof from the so-called Hlnayana-Mahayana conflir.t. He asserts that in the
noo-emergeoce of the Buddbas and the waoing of the Sravakas, the knowledge
of the "truth" continues co be perpetuated by the Pratyeka-buddhas , even
without association with the Buddhas. The need to depend p,rimarily upon an
unbroken tradition or an uninterrupted line cf pacriarchs for the perpetuation
of the "true docttine" is discounted hete.
19. Time (kala). Io the discussion of the notion of imperishable action (atlipraf!'iifa-karma) discussed above, .two more conditions relevaot to tbe fruitioning of karma were mendoned, namely, time (kala) and harmony (fiimagff).

60

nrn PHllOSOPHY OF THE MIDDLE WAY

These two topics constitutc the subject mattcr of the next ewo chapters.
Time was conceivcd in a moce empirical way in the early discourses. Itnpermanence (aniccafii) was explained there in terms of the temporaHty of evcnts.
As meotioncd in the analysis of the conditioned (VII). such temporality was
understood as arising (uppiida), ceasing (vyaya) , and change of what cndures
(thitassa aflfJathatta). However, the problems created by the Sarvastivadins and
the Sauuantik:as by their analysis of time and temporaity have already been
aluded to (sec the discussion of the "Condicioned" above). There, the focus
was more on the substantiality of eventS (conceived as bhii11a) and the difficulties that arise as a result of attcmpting to place such "existents" (b/J4va) in
the context of tempol'ality. The present chapter is devoted to the conception of
cime itself, especially time as analysed by the metaphysicians into discrete
moments (k!M!tl). Nagarjuna's analysis brings out the disastrous implications
of such a notion of time and could approptiatdy be compaced with an analysis
provided by E. R. Clay and enthusiastically adopted by William James in
Westetn philosophy. Examining the ordioas:y notion of cime, Clay says:
The relation of cxpcrience to time has not been profoundly
studied. l ts objects arc givcn as bcing of the ptesent, but the part of
time referrcd to by the daturo is a very different thing from the conterminous of the past and future which philosophy denotes by the
oame Present. The present to which the datum refers is teally a part
of the past- a recent past - delusively given as being a time that intervenes between the past and the future. Let it be named the
specious present, and et the pas't, that is given as being the past, be
knowa as the obvious p ast. All the notes of a bar of a song seem ro
the Hstener to be contained in the present. All the changes of place
of meteor seem to the beholder to be contained in the present. At
the instance of the termination of such a serie~. no part of the time
measured by them seems to be past. Time, then, coilsidered
relative to human apprehension, consists of fout parts, viz,, the obvious past, the specious present, the real present and the future.
Omitting the specious present, it consists of three . .. noneotities- the past, which does not exist, the future which does n ot
exist, and their contetcnlnous, the present; the faculry from which
it proceeds lies to us in the fiction of the specious present. 109
Nagatjuna's analysis indicatcs, in a similar way, how a metaphysical notion
of time would lead to the abolition of the very notion of time as "specious."
Furthermore, the metaphysical notion of time, as propounded by the Satvastivadins and the Sautrantikas, also involved the conception of substantial ex-

61

lNTitOOUCTION

istence (bhtiva), the de nial of which would resut in the deniaJ of that panicular
conception of time.
20 . H4rmony (s/if!JtJgn). The anaJysis of the causal process in terms of discrete
momentaiy event.s eliminated the possibility of explaining harmony as part and
parcel of the eventS that combine to produce the effect. With such discrete.
events harmony becomes an attribute. Such a problem was not faced by the
Buddha wheo he expiained dependence of events because such events were
recognized as related events rather than discrete ones. Thus, the Buddhist
metaphysicians were compelled eithcr to accept an immediately contiguous
cause (samanantara.pratyaya} where each event is caused by an immediatdy
preceding event, or make barmony an attribute of such discrete events so that
their assemblage could provide.a rationale for the production of the effect. The
problem of causation recciYed the foremost attention of Nagarjuna, as is evi- .
dent from Chapter I. The difficulties involved in explaining the arising of the
fruit or effect (phala) on the basis of a metaphysical nocion of harmony _are further elaborated herc. Once again, what is denied is not the arising of the fruit
or effect, for that was the central philosophy of Buddhism, but only the mannet in which such arising is described by the metaphysicians. Hence
Nagatjuna's conclusion: The effect is not made by the harmony, nor is it made
by a oon-harmony. "Where can there be harmony without an effect?" (XX.24).

21. Occurence and dissolution (1an.zbhava-.vlbhava). This chapter condudes

Nagarjuna's examination of the nature of the human personality as it gradually
evolves or dissolves depending upon one's actions (karma). In the "Discurse on
the Knowledge of the Beginning" (Aggaflffli-suttanta). 110 the Buddha speaks of
the evolution and dissolution not only of the world, but also of the human personaicy . This discowse was intended primarily to refute the rather static cooception of the world and the social order presented in the Indian philosophica1
and religious uaditions. Without commirting hlmself to any notion of an absolute beginning, the Buddha spoke of a period of dissoJution (satp11at{a)
followed bya long period of evolution (11ivafta). Io spite of the Buddha's reluctance to get involved in the discussion of such theories, because of the
epistemological difficulties, be was compelled to do so by the unfortunate
moraJ and ethical implications of the Indian castesystem . Even though the
discussion of the world-systems is race in the early discourses, the evolution as
weU as the d.issolution of the human persona1ity through Jong periods of time
constiruted a popular subject. The process of the evolution and dissolution of
the human persooality carne to be designated sa'l(lfiira {life-process) or bha11a
;

62

l'HS PHILOSOPHY OF THE MIDDLE

W/\Y

(becoming) and was contrasted with the Indian notion of life as eternal ex
istence (atthit'ii, Sk. astilva) .
While the concepts of atlhiflz (existetlce) and n' atthitii ( non-existence) wece
. used in the pre-Buddhist literature, the Buddha, realizing the rnetaphysical
implications, avoided them and instead utilized the notion of bhava (becom
ing). No sooner than the Buddha explained the human personality as a process
of becoming, the metaphysidans of che traditional schools of Indian
philosophy began speaking of bhava (instead of astitva) and abha'Va (instead of
n'iistitva) when speaking about . existence and oon-existence respectivey. ewo
terms which wece not popular in the Indian tradition before rhe Buddha.
The Sarvastivadins and the Sautrantikas, who were lured into this substan
tialist trap as a result of their analysis of the process of becomlng (bhava) ioto
discrete moments, formulated che notion of a "secies of becoming" (bhavaIa'f?Zf4tt), instead of the "stream of becorning" (bhava-sota) referred to in the
early discourses. Buddhaghosa, who introduced these different interpietations
ioto the Thervada tradition in the South and South East Asian countries (circa.
sixth cenwiy AD), discinguished between three different nocions of the present:

,,
li

111

the specious present (addha-paccuppanna),

the momentary present (khaf!tJ-paccuppanna), and
the flowing pi:esent (sa'f?Ztati-paccuppanna). 111

He proceeded to identify che.se with the differetu;stages in the developmenr

of the Buddhist thought, maintaining that the "discourses" (suita) advocated
the first, that some othet Buddhists (probably the Sautrantikas) spoke of the
second, and chat the commentaries acceptcd the third.
Nagarjuna, compiling his treatise during the second century AD after the
Sarvastivada and the Sautrantika theories had come ioto prominence, could
not have been uoaware of these differences. In the present chapter he was
therefore critidzing the metap4ysical notion of a "series of becomiog" (bhavasa'f?Zfati), a series of disjointed or disconnected momentary (k!a11ika) existcnces,
rather than the conception of becoming as founulated ~y the Buddha. Thus.
after a critidsm of the notioo of bhava-sa'f?Ztati, in the present chapter, Nagarjuna is able to spcak of bhava at XXVI. 7-8 without rejecting it. He carefuJly
avoids the concepts of bhava and abhava as weU as svabhava and parabhava
throughout the text. Thus, che denial of bhava-sao/tati need not be consttued
as a de.nial of bhava or bhava-sotP which occur in the discourses of the Buddha.
The problems discussed so far relate primarily to the oature of the human
personality , its survival, and its mora responsibility, and Nigarjuna's endeavor
is to establish its non-substantiality (pudgala-nairlit1nya). The chapters that .

l NTRODUCTION

63

follow deal with the non-substantiality of the person who has attained
freedom .
22. "Thus Gone One'' (tathagata). A host of epithets wete used to desr..ribe the
attainments of the person who was able to understand the narure of human existence and overcome the , sufferin~ associated with it. Two of therp. stand prominent. These were buddha or the "enligbtened one" and tathigata or the
"thus gooe one," and even these two terms were used synonymously. The first
of these describes the ideal achieved by one who was aspiring for knowledge or
unde.rstanding, an aspiration cleady expressed in the famous Upani~adic
statement: "Prom darkness, lead me to light," (tamaso miir{J jyntit gamaya). 112
In the Upani/ads, this enlightenment was nothing short of"omniscience" (satvajfia),t19 even though this term was not used in a technical sense. For the llud- .
dha, enlightenment is attained as a result of the realization of the means and
limits of knowledge (see the above analysis of the Kaccayanagotta-sutta).
The second cpithct, ltTthagata, led to morc misunderstanding and rnisinterpret1nion in the Buddhist context. As the term impues, it cxplains the ultimate
goal to be achieved by such enlightenment. lt represents an attempt t o answer
the aspirations of the human beings who were looking ft>r immortality avoiding
the endless cycle of births and deaths. In the Upani.fads it is expressed in the
staternent: "From death, lead me to immortality," (mrlyor maf?t amrtaf?t
gamaya). 11 ~ According to the Upani.fads, the "enlightened ooe" who under
stands the nature of the real self (atman), upon the dissolution of the
psychophysical personali ty, is united with the universal idea! self (brahman)
and thereby enjoys eternal life. The Buddha, who was not willing to admit the
cxistence of sucha state, merely maintained that the "enlighteoed one" (b11dt/ha) is also freed from continuous becoming (bhava). Hence, the tathagata or
the "thus gone one" is conuasted with one who is subjected to becoming
(bhava) or re-becoming (punabbhava).
With the attainment' of enlightennleOt and freedom from grasping
(upadiina), the freed one leads a happy and contented life, white at the same
time not longing for a future existence. Enlightenment (bodht) is synonymous
w.ith waning of craving (laf!hakkhaya) , However, the unenlightened person,
bound by craving and grasping,, not only looks for eternaJ life beyond the grave
(param marat{a) , but a1so expects to sec something more mysterious and aweinspiring (acchanya-abbhuta) iri the life of a freed one who is alive. It is for this
reason that two major questions were raised in relation to the conception of a
tathagata. The first is the question as to whether the tathagata is identical or
differeot from the psychophysical pers0oality . 11-' The second is the question as
to whether the tathagata survives the destruction of the psychophysical per~
sonality. 11 ' This may explain why, in th.e early discourses, whenevet such ques-

64

TuE PHILOSOPHY OP THE MIDDLE WAY

tions wece taised they wece raised more in relacion to a iathagata thao in connection with a buddha.
Thus, after a detailed examioation of the problems of human existence,
problems sucb as action, mora responsibility, or becoming, it was natural for
Nagarjuna to direct his attention to the questioos pcrtaining to the cooception
of ultimate goal as envisaged in Buddhism. In doing so, he was compelled to
deal, first of all, with the meaning of the concepi tathagata.
Nagarjuna's examination of the lathiigala follows the line of analysis found
in the early discourses. While early Buddhism coofronted pcoblems relating to
the tonception of tathagata because it was understood 1n terms of the notion of
a permanent and eteroal self (atman), Nagarjuna is hece struggling with the
same conception as explained in relation to the notion of a real subsance
(s11abha114). Nagarjuna was not willing to consider the tathagata, conceived in
such a metaphysical way, as identical with the aggregates (skandha). Th.is was
the Sarvasdvada position which Nigarjuna was rejecting (XX.Il.2). Nor was he
prepared to accept the alteroative suggcsted by the Sautrantikas, who emphasized non-identity or difference~ The notion of anatman (no-self) referred
to at XXII. 3 is really the coo(eption of ''other-nature" (parabhava) that was
propounded by the Sautrantikas. loada'.s explanation that "the vse of the term
anatman hece is not to be confused with the cardinal Buddhist doctrioe by the
same term,"116 therefore needs to be qualified as the "cardinal early Buddhist
doctrine." for the Sautrantikas we re guilty of propounding a tbeory ofanatman
which emphasized real difference. Nagarjuna was therefore ready to assert that
a tathagala conceivcd in terms of cithcr "sclf-naturc'' (svabhiiva) or "othernacure" (parabhava) is not evideot.
The question whether the Jathagata is dependent is taken up next. Its
dependence or independence is then rejected prirnarily because once again it is
understood in terms of substantial dependence (svabhiivata upadanatp,) or
substantial independence (which is the implicacion of "otber-nanue" or
parabhava)(XXlI.9). Views of tathigata as "empty" (u11ya) or "not empcy"
(au11ya) are coosidered, and these again are particular view ( expressed in the
iii-formula) and are therefore rejected.
How the notion of a living tath'iigala, conceivedof io a rather metaphysical
way. leads to the belief in a tathagata after death is explicity stated by Nagarjuna at XXII. l3. Nagarjuna's argument is that jf the tathagata were to be conside.red empty in terms of self-nature (svabhavatal/) , any thoughr of his beiog
existeot or oon-existent after death (para'(J-nirodhat) is not appropriate. This,
indeed, is the view expressed by the Buddha in the early discourses. 11 1
The concluding- statement of this chapter is rather significant, especially in
view of the narure of the "freed one" (nibbuta) or "freedom" (nibbana) as
enunciated in the early discourses. We have already pointed out that freedom

INTRODUCTION

65

(nibbana) is a state unconditioned by dispositions (asakhata) . It is not a state

chat is uncaused (appaftccasamuppanna). The tathagata is, therefore, uncondiciooed by dispositions but not in-dependent. Elsewhere Nagarjuna insists that
there is nothing in the world that is in-dependent (XXJY.19). The tathagata is,
therefore, like the universe (jagat) wherein the principle of depcndence
(prafiJyasamutp'iida) functioos . He has no self (atman) or substance
(s11abhava), as it is in the case of the universe (jagat) .

23. Confusions (11iparyasa). The reasons for the misunderstaodings that prevail
rcgarding the narure of the enlighten.ed one (buddha) or the "thus gone one"
(tathagata) as weU as aoyching that takes place in the universe (jagat) are then
raken up for examination. Once agaio the nucleus of the chapter can be traced
back to the early discourses.
A discussion of the four types of confusions ( 11ipalliisa) relacing to perceptions
(sanna), thought (citta) , and views (diffht) is met with in the Aguttar.a
nikaya.118 The basie confusions rclatiog to thesc three differeot functioos are
given as follows:

1.
2.
3.
4.

Perceiving, thinking and viewing what is impermanent as

permanent.
Perceiving. thinking and viewing what is not suffering as suffering .
Perceiving, thinking and viewing what is non-substantial as
being substanciaJ.
Perceiving, thinking and viewing what is impure as pure.

The first three of these, in that particular order, arc generally .referred to as
the three characteristics (lakkha11a) of human existence, that is, impcrmanence
(anicca), suffec.iog (dukkha), and oon-substantiality (afllltta). Nigarjuna has
already dealc with these concepts at length (see Chapcers 11, Xll and XVlll).
Hence, he begins his analysis with the last, namely, the pure (Jubha) and impure (afubha). Thesc indced are value judgmeotS made by the human beings
and scrve as the foundacion of the reli.gious and spiritual life. However, in the
eyes of the substantialist phllosophers, these were ulrimate qualities, each having its owo narure (.rvabha11a) . With the first rwo verses, Nigarjuna sets the
tone of his criticism of tncse qualities, which aJlows hirn to move on to ocher
types of confusions subsequently. Whilc the interplay berween the qualities of
Jubha and aubha arc mentioned in verse 11, Nigarjuna's refuration of the

66

THE PHlLOSOPHY Of rnE MIODU! WA.Y

ultimatc reality of thcsc qualitics is bascd initially upon his rccognition of the
dependent arising of both.
The notions puricy and imputity, ike any other entities, have no ultimate
substaotial standing. Thcy are based upon lust (riiga), hatred (d11e1a) , and con.fusion (moha), wbich io turo arc the productS of thought or conccptualizatioo
(sarrz~pa) . The ccssation of Just, ha.tred, and confusion is generaly equaced
with "frcedom" (nifflaa). Hcncc, ie is geocrally assurocd that nirvt11!a is
beyond any conccptualization or thought. Tbis seems to be the manner in
which the notion of a "non-conceprual" and ultimatc reality (paramiirtha)
carne to be attributed to Nagarjuna and, hcncc, to all Madhyami.ka philosophy.
Early Buddhism refers to two forms of thought or conceptualization
(sa1r1kappa), che more comprehensive and, therefore, right rhought or conceptualization (sammii-sa1r1kappa) and the moce rcsuictcd and, thcrcfore, wrong
thoughc (micchii-sQtrJkappa). This is in no way different from whac Nagarjuna
discusscd in Chaptcr:XVIJ {12-13). wherc he distinguished thoughts(k.alpana)
chat arc pro~r (yojyale) from those chat arc inappropriate (nopapadyale).
In many instanccs, "thoughts" (sa1r1kalpa) and "dispositioos" (saf!l.Skii-ra).
two terms chat are semantically related, function in simiar ways. This is artcsted to by Nagarjuna's treatment of them. We have alrcady notcd how be
characterized "dispositions" as "confusions" (mr,r.i){XIII.1 - 2). This definition
was, in face, attribuced to che Buddha himself. In the presenc concext, speaking
of "confusions" (11iparyasa) , Nagarjuna introduces the notion of sa1r1kalpa and
proceeds to defrne it io terms of its consequenccs, namely, tbe generation of
Just, etc,
A carefuJ analysis of the notioos of sa~alpf1 and Jaf!l.Skiira in relation to
frecdom or ninili11a may clarify an importaot epistemological problem, l:roth in
early Buddhism and in Nagarjuna. We bave already referred to che fuoction of
dispositions (sankhara) in the context of cary Buddhism. They cannot be
eliminated exccpt at death, and arc, thcrefore, to be appeased (samalha,
upasa,na). This process of appeasement is to be acbieved by not clioging on to
any of che past disposirions whcn one has co deal with the problem of underscanding any situation . Hence, nibbana carne to be designatcd asankhata.
When speaking of the thoughts or concepts (sa1r1ktdpa), however, we are
prcscnted with two types, che right and che wrong. The right ones are to be
cultivated and the wcong ones eliminaced, a process not recommended in relatioo to dispositions. This dichotomy betwcen right and wrong thougbts could
have unsatisfactory implkations. Right choughts may be takeo as pointing to
true events ot phenomena, white the wrong ones ma.y indicate che absence of
such events or phenomena. This , indeed. was che subsrantialist trap which both
the Buddha aod Nagarjuna were anempcing to avoid. Therefore, without takiqg right choughts in the sense of absolutey uue ideas correspoading te

lNTROOUCTrON

67

ultimate facts, that is. as having substance or self-nature (s11abha11a), Nagarjuna

wants them to be treated as empty (Unya), not in the sense of absolute nonexistence (abhava) but in a more pragmatic sense of being abe to produce consequences. Thus, while wrong thoughts (mithy'ii sa??Jkalpa) are productive of
Just (raga), hatred (dvefa), and confusion (moha) , right thoughts (samyak
sa??Jkaipa) give rise t o fieedom froro Just (vairagya), compassion (karot{?i,
adve;a), and knowledge (praj'ifii).
The entire cha:pter on "confusioo" (11iparyaa). is therefore, devoted; not to
an ouuight rejection of the simple discriminations of putity and impurity, etc. ,
but to a vehement cricidsm of soch discriminations based upon the notions of
absolute existr.Qe (astitva) and absoute non-existence (nastiJva).

24. Truth (satya). Thus we are led to the most important discussions in Nagatjuna, namely , the conception of the fout noble uuths (arya salya). The contents of this chapters have generared much discussion as well as controverty during the centuries that foowed its compilation. t the same time, this chapter,
more than any other, will serve as a glowing testimony to the fact that Nagarjuna was simpy restating the ideas expressed by the Buddha in the eacly
discourses, rather tban bringing <.i.bout a Copernican revolution in Buddhist
thought.
In the early. discourses, the fou r noble truths were meaot to explaio the
nature of human ex.istencc., bth in bondage and in freedom, avoiding the extrems of permanent existence (atthitii) and nihiListic non-exisrence (n'
atthita). The difficulties encountered by che Buddha in ma.king this view of existence iot.etligible to the substantialist thinkers of India are evident from the
kind of criticism thcy levelled against the Buddha. Very ofteo they criticized
him as a nihilist (n. 'atthikt#ilida, ucchedavada), insiscing that he advocated the
annihilation and destruction of che conscious being (s11to sattaua uccheda?!J
viniisaf!i pannapett). 11 9 This was not because the Buddha denied the existence
of conscious human beings, but because he was notwillingto accept an cternal
and immutablc self (attnan) in them. For he often insisted th.at a conscious
human being is empty (sunna) of a permanent and eternal self (atta) as well as
anything penaining to or bclonging to a sdf (attan-iya). 1l 0 The <lenia of sucha
self or substantiality was not ony in relation to a human personali cy but also in
connection with any experienced phenomena (.rabbe dhamtnli artatt"ii). 121
Nararjuna was placed in an identkaJ situacion as a result of his rejection of the
Sarvasdvada conception of self-narure. (s11qbhat1a) and the Sautrii;ntika theory
of other-nature (parabhiiva). The theoty of nn-substantiality (anatman) or emptiness (funyatii) that he attempted to explain in the ,prev.ious chapters was not

68

THE PHILOSOPHY OF THE MIDDLE WAY

palatablc to the subst.antiaJist philosophers who raised the objeccion discussed

in the first sbc verse of the preseot chapter.
In the first instance, these substantialiscs, who understood empcincss as
"notbingoess," assumed that empti,ness leads to a denial of the four noble
truths, including a denial of the noQle fruits (arya-phaliint). These noble fruits
are elsewhcre referred to as fruitS of the ascecie life (1am4nath(l.Sa =
fr(ff1ii i11yiirtha ,22 fiimafiaphat 1 23) . Secondly, they felt that it also leads to a
denial of the fruits (phala) of ordinary human life (laukika), iocluding fruits of
all the ordioary mora and social conventions (sa1!J11ya11ahara) such as good and
bad (dharma-adharma) (XXIV.6).
Underlying this two-fold cricicism thcre seems to be a basie assumption.
Whcreas in the carly discourses the four noble trutbs could account for the
fruits of ordinary human existence (i.c . of the path, magga) as well as the fruics
of the higher life (i.e. , freedom or nibbina), it seems that when the above
cricicism of cmpciness by the substantialists of the Buddhist tradition was
presented , the four noble truths werc looked upon as referring primarily to the
higher life . Hence the need to present an additional criticism chat cmpciness
contradicts even the worldly (laukikan).
Nagarjuna's attempt here is to collapse these rwo As.mes inco one and treat
them undcr one rubric, namely dependent arising (prafilyasamutpada) (XXIV.40), which is the central philosophy of Buddhism.
For this reason, having made the inicial remark that his critics were not conversant with the use or purpose of emptiness and, therefore, are troubled by
both empciness and its meaning (XXIV. 7). Niigarjuna immediately proceeds
to explain the two truths, instead of the four truths.
Nagarjuna sees the Buddha as expouodiog two truths:
1.
2.

the truch of wordly convencion (Jok4-Sa1!Jvr ti), and

the truth in terms of ultimate fruit (paramiirthataf?).

Nagarjuna had aJready devoted twenty one chapters (I-XXI) to the cxplicatioo of the first of these truths. Causaliry, space, time, motion, the human personality, action, consequence, good and bad-all of chese have been deat wich
ac lcngth . Explanations of these in terms of absoluce existence or nihiiscic nonexistence were rejected in favor of dependence (pratityasamutpada) and ,
therefore, of emptiness (funyatii). Artha or fruit of existence, wh.ether that be
good or bad, was rccognized. Attempts oo che part of the metap\ys icians who
wanted to perceivc with absoluce certainty how a cause produces an effec;c were
abaodooed, sincc such attempts Led to the recognition of unacceptable entities
such as self (atman) oc substance (111abha11a). The abandoning of such attempcs

lNTRODUCTION

69

did not lead l'fligarjuna on to the other extreme of denying any connection between cause andeffect, action and consequence. The element of uncertainty involved in the cause-effect relationship made him moce cautious tban either the
Sarvastivadins or the Sauttantikas, and hence he was mote defeosive and
negative in his descriptioru. Yet in no way did he want to abandon that princi
ple of explanation, Chapter XVII on "The Examination of Action and Consequence" (K11mra-phala-paiikfa) being the most illustrative example.
The fact that a human being, having understood the nature and functioning
of phenomena (dharma/;), attempts to achieve various desired results (artha) by
manipuJating such phenomena, was well known to Niigarjuna when he spoke
of both Ja?(lsliiiras and sa'(Jkalpas, However, the possibility of achieving
ultimate freedom (ni1'1l'ii1Ja) or the ulcimate fruit of existence
(paramiirtha)(somecimes referred to by the Theravada tradition as aggaphala,
Sk. agra-pha/4124), has now been questioned by _his opponents. Again, withouc
falling ioto the extremes of existence and oon-exiscence and recogniziag the
emptiness of all dependently arisen phenomena, Nagarjuna had to explain the
fruits (artha) as well as the ultirnate fruit (ptTramiirtha) of eidstence, fa speaking
of these rwo truths, if he had assumed that the latter traoscended the former ,
ht> would be presenting the ideas atttibured to the so-called Mahayana, rather
than quoting the ea.tly discourses or refening to the teaehings of the Buddhas,
Pratyeka-buddhas and the Sravakas. This, however, is not the case, for his explanation of artha as well as paramiirtha is couched in the same anguage, and
that was the Janguage of dependeoce and empriness. Hence his famous dicrum: "Withour rdying upon convcotion , the u1timate ftuit is not taught.
Without understanding the ulcimate fruit , freedom is not anained"
(XXJV.10).
.
Artha as well as paramartha are tru.ths (satya)~ The for.mer is not presented as an
un-trurh (a-satya) in relation to the latter, as it would be in a.o absolutistic
uadition . Neithet is the form er sublated by t he latter. There is no indication
whatsoever thac these are two truths with different standing as higher and
Iower.
The fruits of ordinary human cxistence (artha), understood in terms of t:Jermanent existence (svabh?i11t1) conflicred witb everything in expe.deoce: "If you
perceive the existence of the existence i n terms of self-nature, then you will also
perceive these a non-conditions" (XXfV. 16). This would lt>ad to a denial of all
phenomena such as effect (karya), cause (kara11a), agent (kartr), doing
(karatfa), action (kriya) as well as arising (utpada), ceasing (n.irodha), and fruit
(phala) (XXIV. 17). This compelled h.im to make the most famous of his
staements: "We state that whatever is dependent arising, that is emptine.ss.
That is dependent upon conventioo. That itself is the middle path''
(XXIV.18). Everything is placed in one basket, the basket of "dependent aris-

70

THE PHlLOSOPHY OF THE MroDl.E WAY

ing'' (prafityasamutpada). "A thi:ng that is not dependently arisen is not evident. For that reason, a tbing that is non-empty is, indeed, not evidenr
(ibid. 19).
With that analys.is, Nagarjuna was ready to defend the four noble truths,
reconciling it with his cooception of emptiness (ibid . 20-21). 'Whoever
perceives dependent arising also perceives sufferiag, its arising, its ceasing and
the patb [leadiog to its ceasing]," (ibid 40).
Buddl:ta's statement in the Majjhima-nikaya: "He who perceives d~pendent
arising also perceives the dh111nma," could not bave received better confumation from a disciple who was removed from the Buddha by ar least six centuries.
2S . Freedom (ni1iiina). This is undoubtedly the most significant chapter in the
book. The interpretacion of the contents of this chapter by Stcherbatsky12' has
d~minated the Western understanding of Madhyamika thought for a considerable pedod of time. Elsewhere, we have disagreed with Stcherbatsky's interpretation of early Buddhism, allowing hi:m his interpretation of
Madhyamika phiosophy. Since Stcherbatsky's perception of Nagarjuna was
colored by his understanding of carly Buddhism, the rejection of the latter
shoud mean the rejection of the former too, if we are to consider Nagarjuna's
philosophy as a continuation of the ideas of early Buddhism. Without devoting
much time to an examination of Stcherbatsky's views at this point, we will present the conrents of Ch?-pter XXV in the light of the analysis of Nagarjuna's
ideas that we have already presented.
Other modern scholars have rejected Stcherbatsky's ideas and have presented
views about nirvana that do not appeac to accord with what Nagarjuna has said
in the previous chapterse.. One of these is the view presented by Inada that nirvana represencs the ''uncreated realm'' (as4'(Jskrt4). 126 The view that nirvana, as
tJS4f!lJkrta, belongs to the "uncreated realm," a view which is popular with both
the so-called Theravada and M9.Myana interpreters (especialy of the modern
world), may lose its tenab,ility if the cootents of this chapter are analysed in the
Jight of what went before rather than in isolation.
Th~ attempt to explain Nagarjuna's conception as one that is found in the
Mahayana tradition is based upon a complete misreading of Chapters XVI-XXI
of the Kiin'k'iis that deal with the notion of the human person~ity, human
behavior, and mora responsibility. Such a misreading compels Inada to reject
the va!ue of the two chapters (XXVI-XXVU) that follow the chapter on nirvana
(XXV), saying: 11With the discussion of Nirvana in the last chapter the treatment from the standpoint of Mahayana had basically come to a close. In this
chap ter and the fina one to follow , Nagarjuna goes into the analysis of the
Hlnayana 4ctrines."121

lNTRODUCTION

71

Our aaalysis of the Kiin'kii1 so far did not reveal any specific Mahayana doctrine presented by Nagarjuna chat may be contrasted with the so-called
HToayana, and we have therefore no reason co look at Nagaruna's cooception
of nirvaoa as chat of Mahayana or reject the last two chapters of the ueatise as
being cepresentative of the Hlnayana docuines. In fact, to assume that such an
outstanding philosopher as Nagarjuna, who presented the world with such a
beautifuly executed philosophical classic, could simply add two chapters utterly irrelevant to the basie theme of his work does not contribute either to the
understanding of his philosophy or an appreciation of his genius.
Nagarjuna begins his analysis of nirvana anticipating the same kind of objection that the substantialist raised againsc reconciling "empciness" with the four
noble truths. "If alJ this is empty, there exists neither arising nor ceasing. [As
such) through the relinquishing or ceasing of what does one expect freedom? "
(XXV.l).
l t is necessary to keep in mind herc the conception of a thing (dharma) that
the substantialist envisaged, which was the main subject of scrutiny on the part
of Nigarjuna. The substantialist had cliffi.c ulty with the conception of empcioess (unyatii) primarily because an existent or phenomena (dharma) for him
was one that posscssed sdf-oature (111tZbha11a). As emphasized earlier, if not for
that assercion of the substantialist, Nigarjuna had no reason or provocation to
compose the present treatise. This is dearly evideot from Nigarjuna's immediate response to the substantialist: ''If all this is non-empty , there exists
neither arising nor ceasing. Through the relinquishing and ceasing of what does
one expect freedom?" (ibid. 2).
For N:igarjuna, to say thac soroething ~ not empcy (fZfUnya) means that ie has
subscantial existence (s-v11bha11a) du ring the past, preseot and future, and if so it
would be meaningJess to speak of its arising and ceasing. This would certainly
rcnder "freedom" impossible.
Once again, keeping the substantialist view in mind, would it be possible to
speak of "freedom" (nirva'!a) as the relinquishing of something that is
substa.cial and the attainment of someching completely new or different. This
was another importanc assertion of the substantiaJist. For him, "freedom"
represented a ~otaly different state of cxistence (astilva), an existence that is
not only perrnaneot and eternal but also perfccdy blissful and happy. Freedom
or niroatfa tbus curns out to be a metaphysical notion, like the Hindu
brahman, uncaused, uocreatcq and, therefore, beyond aUspadaJ and tempora!
determination. Consideriog these two views, namely,
1.
2.

the substantialist conceptioo of ordinary existeoce, and

the substancialist notion of freedom,

72

THE PHTLOSOPHY Of THE MIDDLE WAY

Nagarjuna proceeds to define freedom (nirva,,.a). "Unabandoned and

unachieved, unincerrupted and impermanenr, unextlnguished and nonarisen- this is caUed freedom" (ibid. 3).
Here are three sets of negations, and unless we are careful .in analysing these
three as .ret.r, .it is possible to arrive at condusions not inteoded by Nagarjuna
himself. The follow.ing analysis of the sets will be made not only on the basis of
the coocepcioo of nirolif!a presented .in early Buddhism but also .in the light of
the substancialist views that Nagarjuna was confronted with,
1.

Unabando11ed 'and unachie11ed (aprahT.,,.am a.ramprap/411t): lt

(nirvat1a) is neither casting off nor reaching. These two activicies

arc complementary, and hence to be taken together. Casting

off something and reaching for something completely differenr represents the substatitiallst ot absolutist way of explaining freedom (A.'VI. 10). The attainment of freedom from
the threc poisons of Just (raga), hatred (dve1a), and coofusion (moha) by a person who is understood as ''beig in a process of bec;om.ing" (bhava) conditioned by various fattors (not
merely the three poisons) need not be explained .in terms of
the dual function of casting off and reachlng. While on the one
hand, one may be casting off the three poisons and not
everything, on the other ha.od there is noth.ing that is reached
for. lf there were to be seomthing to be Ieached for, chat
would again be a source of bondagc tathcr than frecdom.
2.

Uninlertupted and impermanent (anucchinnam afa11atam):

It is neithet interrupted nor eteroal. These again are complemeotary. Cutting off something completely and attaining
a state of permanent existence is once again part of the
substantialist conception offreedom. As explained in relation
to the previous chara<:teristics, a person who has attained
freedom cercainly curs off the three poisons. However, this
does not mean that "what is distinguishable is also separable"
(Humean cxplanation of distinction). ln the Buddbist cootext, a human being who has eiminated the three poisons,
that is, the Buddha or the arhant, still continues to be a
human being with a body as wel as the associated fedings.
perceptions, dispositions, and coosciousness, and this continuation is on the basis of "dependent arising" (pafic
casamuppada). Separation of "buddhahood" f.tom that
psychophysical petsonality led ro all the metaphysical issucs
that the Buddha as wel as Nagarjuoa were try.ing to deal with
(see Chapter XXII on "The Examination of Tathagata).
Hence the state ofBuddhahood, if sucha terminology can be

73

INTRODUCTION

3.

used in any meaningful sense, did not signify either a complete interruption or eternality.
Non-ceased and non-arisen (aniruddham anutpannam)~ lt is
neither ceased nor arisen. Considering this pair of com- .
plementary characteristics independently and in isolation
from everything that has so far been said by Nagarjuna, it
would be very easy to assume that nirvana, in his view,
trancends all descriptions and characterization. However, if
wbat has been said about arising and ceasing, espedally the
chapters on "The Examinacion of Causality" (Chapter I) as
well as in "The Examination of Dispositions" (Chllpter VII)
were to be tak.en seriously, one would refrain from such
generalizatlons. Instead, the conceptions of arising and ceasing, as weU as the arisen and the ceased, would be placed in
the context of a substantialist view of either exitence (astitva)
or oon-existence (nastitva) .
, Just as much as ordinary existeoce (bhava) and its fruits (art/Ja) cannot be defined utilizing the substaotialist conc;eptions of arising and ceasing. even so existence.(bha11a) and its
ultimate fruit (pata1'iiirtha) cannot be expla.ined on the basis
of a simiar conceptual framework.

in

This, indeed is what Nagarjuna wants to emphasize in the verse that

follows (XXV.4) where .he cakes up the notion of bhava ( =svabhava):
"Freedom is not an existent. (If i t were,] ie would foUow that it has the
characteristics of old age and death. Indeed, there is oo existent without old age
and death."
A substantialist speaking about the characteristics of the existent will have to
maintain that the existent, by its owo narure, is invariably associated with old
age and death. This woud mean that no one wiH be able to attain freedoro,
uriless he becomes a different sort of existent, an existent that is totaly different from what he is. This, indeed , is the absolutist's ootion offreedom . 1t is
a total freedom that has nothing to do with ordinary human existence
characterized by old age and death. And fot th.e Buddha as well as for Nagarjuoa
freedoro makes no sense in soch a context.
On the conrrary, if the existent (bhava) is defined as freedom (nirva'!a), and
an existent by definition is "clispositionally determined" (sa1!Jskrt<1) , freedom
jcself woud be "dispositionally determined'' (11irva11am sa7!J.Skrta1!J bhavet).
However, there is no existeot that is not dispositionaly determined (na
asa1!Jskrto hi vidyate bhaval?). Therefore, freedom coud not be an existent .
Having expressed his view that nirvana cannot be understood as an existent
(bhava) in a substantiallst sense, Nagarjuna, utilizing the argument from

74

TuE

PHllOSOPHY OF THE

Mloou:

way

relativity he used in Chapter I to refuce "other-nature" (parabhava), proceeds

to rejecr the view that nirvana is a non-existent (abh1ivt1).
Most of che confusion regarding che contents of this chapter can be cleared
up and the relationship betweeo early Buddhist and Nagarjunean conceptions
of freedom can be established by a careful examination of the following two
verses:
Whatever is of the nature of coming and goig , chat occurs contingenty or dependenty. However, freedom is indlcated as noncontingent and independent.
The teacher has spoken of relinquishing of both becoming and
other-becoming. Thetefore, it is proper to assume that freedom is
neither existence nor non-existence (XXV.9-10).
Nagarjuna is hece rciterating the exuemely important distinction made by
che Buddha between bhtWa and bh'iiva. Discussing the contents of Chap ter XXI
("The Examination of Evoution and Dissoution'') , it was pointed out chat in
the early discourses the term bhava (becom.i.ng) was utilized by the Buddha to
expJain the human life process. Human life as a wandering (Ia'f!Ziiira) was
characterized by two motivations or cravings (tan.ha), the first of wbich is craving for cootinued becoming (bhava-la1Jh'ii) or survival (punabbhava) . This is
one of the most dominant motivatioos for action. When continued becoming,
with its attendant suffering or frustrations, fails to satisfy a man , he desires to
''bc:come something else" (vi-bhava) , the attainment of permanenc and eternal
happiness in heaven r in the state of brah,,.,11111 being only one of them.
According to the Buddha this process of becoming something else, i.e the
dissolutipn of the present existence and the anainm~nt of a permanent state of
existence is anothcr form of craving (vibhava-laf!ha), which. instead of
leading on to the desired form of existence, contributes to further becomiog

(punabbhava).

Afcer having rcjected the conception of self (iitmart) understood as permaneot existence (atthifii. astitva), the Buddha used the term bhava to explainthe
proccss of becoming. The metaphysicans, as pointed out before, immcdiacely
brought back the nodon of atman or astitva into the Buddhist doctrine when
they began speculating on bhava, two species of which were "self~nature"
(s1111bhiiva) and "other-nacure" (parabhava). Nagarjuna seems to havc bcen
wel aware of the Buddh'as discourse on becoming (bhava) and other-becoming
(11ibh(1vt1). He realized that this was the life process or the wanderi.ng (sa'f!ZI'iira)
that the Buddha spoke of. In addition, he was also aware chat, while eocouraging the people to abandon both becoming and other-becoming, the .Buddha
did not preseot a per.mancnt and eterna life (bhava, M#tfla) or complete an-

INTRODUCTION

75

nihilation (abha11a, n'iistiffla) as "freedom'' (niroaa). This awareness is succincdy presented in X:XV. I quoted above.
lf Nagarjuna's analysis of bha11a-11ibhava and bhava-abhava is compacible
with the Buddha's own analysis of bhava-vibhava and atJhilii-n 'atthilii (astil11a11iistit11a), Nagarjuna will be confonted with the same set of problems that the
Buddha faced in explairung freedom Thus, after rejecting the explanation of
free.dom in terms of bha11a-11bha11a or a combination or denial of both
(XXIV.11-16), in the next rwo verscs Nagarjuna refuses to use such terminology
to cxplain the freedom attained by the enlightencd one, either while he is stili
alive (ti[fhamiina = sop'iidise1a-nibbana) or when he passes away (para<f!Z
nirodhad = anupadisesa-nibbana) (XXV .17-18) .
. This leads Nagarjuna to make a remark which elicited two polar inrerpretaooos:

The life-process bas nothing that distinguishes it from freedom.

Freedom has nothing that clistinguishes it from the life-process.
Whatever is the extremicy of freedom and the ei:ctremity of che
life-process, between them not even the subtlest something is evident (XXV.19-20).
The two polar interpretations to which these statements led arc as follows: The
adbereots and sympathizecs of Mahayana has ioterpreted thcse statements as
implying essential identity of sarp.sara and nirvaoa, which they rccognize as a
uniquely Mahayana view of nirva.na that goes beyond even the Buddha's owo
explanations. The so-called Theravadi.ns for whom such idenrification is racher
unpaJatable have condemned it by saying that this represents a cornplete abe.rrac.ion of the salient teachings of the Buddha as represented in the early
discouJses. However, a careful and sympathetic examinatjon of these two
sratements, placing them speci.ficalJy in the context in which they wece made ,
would certainly eliminate such conflicting views and lead to a better
understanding of both the Buddha and Nagarjuoa. Io order to do so, it would
be oecessa.cy to examine two of the key terms that occur in these ewo verses,
namey, 11ife1af!a and ko#.
Thesc two terms arc better understood in the philosophical background in
which Nagarjuoa was writing. Though the background in which the Buddha
presented his views about nirvana was nor as sophisticaed, ie was not different.
The distinction made by the essentialists of the pre-Buddhist traditioo benveen
ordinary human ex:istence (sa1!Jsara) and the state of freedorn (nirva1Ja) led
them to ewo cootraclictory condusions. First, a person has to completely abandon one in order to reach the ocher. It was not only attachment to the seoses or

76

THE PHJLOSOPHY OFTHB MIDDLE WAY

the objects of sense that had to be abandoned, but the se1lSes or the objects of
sense themselves. Freedom was thus reached on the basis of a non-sensuous insight, and the "freed one" (nibbuta) js one who has developed a form of knowing that transcends all forms of sensory perception, including the duality of
subject and object. This was the state of nzf'liiit!a enjoyed by the "freed one" as
long as hi~ body, together with the senses lasted . However, when tbat
psychophysical personali ty is destroyed at death, the "freed one" enters into the
stare of eternal and blissful life (brahman). Secondly, since such an absolute
distinction between sa'f!IS"iira and nirv"iif!a could not explain how one could
reacb a state that is qualitatively distioct, the essentialists also had to believe
that underlying sa'l(liiira is the reality (atman) that reaches nirvaf!a.
Oo the one h;uid, there is a point at whid, a ttansition is made from boodage
to freedom, a transition from one state of existence to something that is compJetel'y and absolutdy differeot, so much so that the one has nothing to do with
the other. Oo the otber band, there is a subtle personality (at1nan) that concinued from che time of the ocigio of existeoce and which lay concealed within
the psychophysical personali ty.
With the developmeot of metaphysical speculations in Buddhism, it was not
surprizing to see two similar conclusions reached by these metaphysicians,
especially the Sautrantikas. On the one hand, the Sautraorikas emphasized
distinctions (11iefa11a) in order to reject a permanent and eternal substance
(svabhava). Yet, when the oeed arose for an explaoation of the identity of
bondage andfreedom, they insisted upon a "seed of release" (mok1q.mja) (sec
anootation of X:XV.19-20). Thus, the first of these ewo verses (19) is intended
to reject the Sautrantika notion of distinction (11ie1af!aL while the second (20)
purports to deny tbeir concepcion of "the seed of re lease". which is not at a
differentfiom the Sarvastivada conception of substancc (svabhiiva).
Before and during Nigarjuna's day, uaditionaJ Indian phiJosophy was also
dominated by two similar essentialist enterprises, namely, detetcnioing identity
(sariipya) and differences (vie/an,a). The specu ations of the Siokhya school
concentrated on the problem of idencity-. The possible influence of this school
on the Sarvastivada theories has already been noted elsewhere. 128 The Vaise~ika
school, as its name implies, focussed on the distinccions (Yisep1~a) in the hope
that such a process would evenrually lead to the discovery of the nature of
ultimate reality. Some of the specuJations of the Sautrantika school reflect this
trend. The notions ofself-nature (svabhava) and other-narure (parabhava) were
the direct results of such an esseotialist search .
Thus, when Nagarjuna says: "The life-process has nothiog that distiogu.ishes
it from freedom. Freedom has notlllng that distinguishes it from the 1ifeprocess" (XXV. 19). to assume that he was presenting an identity of sa'l(liijra

lNTRODUCTION

77

and nir1iiif!t1 would be too hasty a cooclusion. In fact, such an assumption

would undermine all the atte.mpts he made to refute the notion of identity in
the pteceding chapters. lnstead. Nagarjuoa's attempt in this chapter is focussed
on a deniaJ of any ultimate substance, a dharma , that would make either.
sa'l(Jsara or nint1af!a, either bondage or freedom a unique entity and this, indeed, is what is emphasized in the conduding verse: "The Buddha did not teach
the appeasement of all objects, the appeasement of obsessioos and the peaceful
as something to someooe at some place," (XXV.24).
The method of criticism adopted herc is in no way different from that he
utilized in Chapter l in his rejection of the metaphysical theories of causality . ln
that chapter, there was oo denial of a cause (pratyaya) ot an effcct (artha) and the
arising of the latter depeodiog upon the former, so long as these two events and
thei.r mutual dependence is not explained on the basis of a theory of selfnature or other-nature. Similarly, in the present chapter, there is no denial of
the four noble ttutbs that include the fruit and the ultimate fruit
(param?irtha). so long as these are not cooceived of in the form of unique entities (bhava, svabhava), which indeed was the way in which the Satvastivadins
and the Sautrantikas defined dharma. The concluding verse ofthis chapter is
not properly explained except in the background of these two definitioos,
namcly, the Sarvastivada definition of dharma as "that whicb upholds the unique and genera characteristics" (sva-iiimanya-lak.yar;.a-dharaf!1it dharmaf?)129
and the Sautrancika definition that refers only to "unique characteristics" (sva-

lak.faf!a) .H~
26. Human persoMlity and its survival (di1'iidmaga) . Any reade.t who has
ploughed his way through the preceding chapters of Nigarjuna will certainly
be baffied by ~he conteots of Chapter XXVl on the "Examinatio.o of the Twelve
Factors" (Dvad4iii11ga-pank.ffi). By the time he ompleted {eadiog the first
twenty-five chapters, wherein t he conception of "emptiness" (iinyatfi) occurs
in aJmo.st every other verse, he would be imbued with that concept to such an
extent that it would become a dogma, a dr..f!i, rather than a merc method of
analysis. Therefore, Chapter XXVI would make no sense at all. Thus he would
get the impressioo that ir merely deals with the Hlnay~na docuine, having no
reevance to the basie teaching~ of Nagarjuna and, therefore, of Mahayana.
TUs is confirmed by the fact that there is no negative comment made anywbere
in the chapter and no mention of the famous doetrine of emptiness. Another
person can come to a roore drastic concusion. He may assume that this chapter
is ike an "illegitimate child" and could not be the work of Nagarjuna.
Contr.uy to all these widely held opinions, we teod to look upon this and
following chapter as the actual conclusioo of this most valuable treatise. They

78

THE PHJtOSOPHY OP THE MIDDLE W AY

arc intcgtal parts of the work, and without them one gets only a distorted view
of Nagarjuoa's thoughts.
It was mentioned earlier that the onJy d~coursc that Nagarjuna mentins by
name is the KiiJyayanavavada, and the significance of this fact cannoc be over
emphasizcd. Herc the Buddha was rcsponding to the question raised by Kaccayana as to what "right view" (sammii-dit1h1) is, compared with "wrong view"
(miccha-diftht). Nagarjuna has devoted most of his energy trying to darify
what "wrong views'' arc and occasionaUy spoke of "right views" (sec malysis of
chapter XVI!). If he bad concluded his treatise with Chapter XXV, he would
have rcad ooly a section of the Buddha's discourse to Kaccayaoa and ignored
the Buddha's owo ~ conclusioo in that discourse. The two extreme views of
existence and noo-existence were rejected by the Buddha, not because he
had oo views to propound, but because he had a betrer or more appropriace
one to offer. And this appropriate view is explained in the conclusion to that
discourse.
The appropriate view is the middle position specified as dependent arising,
which is iotcnded as an explanatioo of how a human beiog, conditiooed by
various factors, attempts to bccome this or that and wanders aJong in a ceaseless
process of birtbs and deaths. The cheory of personality consiscing of the twelve
factors cxplaining such bccoming thus curns out to be the philosophical middle
position, and the noble eightfold path (magga), avoiding the two exuemes of
behavior, represencs the practical middle patb inteoded to achieve the cessation
of that process of becoming (bhava) and sufferiog (dukkha).
If the Kiityayanavavada served as che foundation of Niigarjuna's philosophy ,
and there does not seem to beany doubt about it, it would have been impossible for him to overlook the conclusion of that discourse. This indeed is rhe rationale for a whole chap ter on the concept of a person explaioed in terms of the
twclvcfold formula of causation.

Part IV (Conclusion)
27. Views (diftht). Buddha's deniaJ of a permanenc and eccrna.I self (alman)
and his expla.oation of the human personality and its survival of death io terms
of the docuine of dependent arisiog have remained unpaJacable co most
philosophers ever since he presented them. Veridical memories of past existences being connected with prcseot cxpcriencc as a result of meditation, as in
the case of the Buddha and many othcr ascetics, and sometimes without any
such practice or effort, ui have received a cwo-fold interpretation in the hands
of these philosophers. Some have assumed che existence of a permanent and
eternal self or subscracum to account for such phenomena, even in the absence

INTRODUCTION

79

of so~called empirical evidence (empiricism as understood by the Buddha?) to

suppon che belief in a permanent and eternal self. These are the eternalistS
(1assakJ-vadin). Others have rejected not only the belief in Jl permanent and
eternal self but also the veridical, though spo.radic, memodes as mere
hallucioarioos of deranged minds. These are the annihilationists (ucchedavadin). The.se rwo views have prevailed in the world until the present day, just
as similar views regarding the relationship berween cause and effect have
survived. The reasons for the prevaleoce of such views were explained briefly
by the Buddha in the K11Ccliyanagott-111tta when he spoke of indinacions
(upliya), grasping (upadiina), and iovolvemenc (abhinit1e1a) on the part of
the human beings. More decailed explanations wece given 1n other discowses.
Wherher there were to be veridica1 memories of past lives oc not, the Buddha
recognized chat human beings were proae to ask tbree types of questions
regarding exisceoce:n2

1.

2.

"Did 1 exist in the past or no.c?" (Ahosif?Z nu kho afitam addhanatp,, na nu kho ahoJif?l afitam addhanaf?l). This, accordiog to the Buddha, pertains to the prior end of existence
(pubbanta), and is pcompted by a desire to know the first
beginn.ig of things, including oneself.
"Will I exist in the future or not?" (Bhavissami nu kho

anagakJ'I?' addhtina?!J, na nu /ho bha11is.rami antigtitdf?l addhana'f!J) . This i>ertains to the future and is prompted by a
3,

desire to k'11oW the fina} end of things, including one's own

destiny.
''Do 1 exi$t in the present, or do I nor exist in the pcesent?"
(Ahaf!J. nu kho asmi, no nu kho as1nt) , This pecrains to the
status of one's present existence.

The Buddha felt that such speculations 1ed to a wide variety of views (dt{(ht),
sixty-t:Wo as specified in the Brahmajiila-suttanta, 133 of which permaoent ~x
istence (althila, sassata-dt/!ht) and annihilation (n'atthita, uccheda-dif(ht) are
foremost. Even though the Suddha recognized veridica memocies of past existences, yet because of the absence of any empirical evidence to support the
hypotbesis of a permanent and eu:rnal substratum as wel as the difficulties involved in predicting future ev~nts with absolute cerrainry. the Buddha questiooed his disciples as to whether it is appropriare for them to " brood over the
past" (pubbarllaf?Z pafidhaveyyath,a) or long for the future (r1paranla'f!t
'iidhaveyjiitha) or be unnecessarily skeptical about the present (paccuppannaf?l
addhanaf?l a1jhattaf?Z kathaf?lkatlii assatha) . He advised them that instead,
with the resources available, they shou1d. try to understand rhings as they have

80

THE Pl-lllOSOPHY OF THE MIDDLE WAY

come to be (yathabhutatp) and work out freedom from suffering. Getting

enarnoured of aoy view, whether it is appropciate or inappropriate, would lead
to futther bondage and suffering. Hence the Buddha's advice to his disciples:
"Without gr~ing on to a view, being endowed with proper perceptioo and
motality, and haviog restrained ooe's greed for pleasures of sense, one avoids a
f urure birth. "134
1t is sigoificant to note that Nigarjuna's fina chapter on "Views" (dittht)
deals with the same issues. He begins his chapter with a refetence to the three.
epistemological inquiries of human beings in rdacion to the past, preseot, and
future, which give rise to various views on the narure of man and the universe.
The first three verses present in identical terms the Buddha's own expanation
of these views as stated in the passage from the Majjhima-nikaya quoted above.
These inuiries and graspings (upqt/ii11t1) lUe chen understood as the reason for
the variety of views discussed in verses 4-28. As the notes oo these verses would
iodicate, these twenty-five verses deal with the variety of views discussed by the
Buddha in the Brahmajiila-suttanta, thus providing underuable evidence that
Nagarju.na had access to most of the discourses of the Buddha that carne ro be
preserved in the Nikayas and the Agamas, and tbat he was merely restating the
origi.nal message of the Buddha rather than providing a philosophical justification of a sectarian view.
The variety of wro11g views (micch'if-di{!ht) was rejectcd by the Buddha
primarily on pragroatic grounds, that is, because they do not lead to freedom.
and happiness. u, They neither bring about worldy fruits (attha) nor they contcibute to the ulcimate fruit (pttramattha), i.e., freedom and happiness. lnstead
they lead to dogmatism, conflict, and suffering. The "rnidde position" as the
right view (samma-di{!IJt), whether it be dependent arising (paficcasamuppiida) or non-substantiality (anatia), or as Nagarjuna puts it, "absence of selfnarure" (nif!J11abhi11a) or emptiness (f.unyafq), Jeads to worldly fruits as well as
the ultimate fruit. However, if that right view wece to become another dogma,
it would cectainly cootribute to conflict and suffe1:ing, thereby Josing its
pragmacic value. In other words, a right view is one for which there cannot
be grasping, for if one wece to grasp it it would tum out to be a closed view not
an open one. This expaios why Nagarjuna condudes his chapter on "views,"
and along with it his famous treatise, with a salutation to the Buddha, a salutation that clearly reflects his knowledge of the Buddha's attitude towards
dogmative views (as embodied in the verse quoted prcviously):

I revereotly bw to Gautama who, out of compassion, has taught

the true doctrinc for th'e relinqu.ishing of all views. (XXVII .30)

INTRODUCTION

81

NAGARJUNA'S PHILOSOPHICAL ENTERPRISE

The above anaysis of the cootents ofNagarjuna's Kiinliii and the annotatioo
of individual verses d1at follow provide ample evidence to support the view that
his pcimary objective was to reject the substantialist or essentialist thoughr that
emerged in the Buddhist philosophical ttadition as a rcsuk .o f the speculations
of the Sarvastivadins and the Sautrantikas. The fact tbat he depended upon the
teachings embodied in the early discourses, or the fact chat che Kiirika is hece
looked upon as a grand commentary on the Kaccayanago1ta-111tta, in no way
minimize his contribution to the bistory of Buddhist thought. What is most
significant is the manner in which he proceeded to examine the subtle and
complex metaphysical issues that blinded che Sarvastivadins and the
Sautrantikas in a background in which speculative philosophy had reached a
high watermark, both among the Buddhists and the ttaditional Indian
philosophers. Ni.garjuna probed into almost every aspecc of thefr speculations,
whethet relating to epistemology, ontology, mora philosophy, or philosophy
of language. He linked disjointed concepts and dissolved the hardened and the
solidified. Concepts of identity and difference, substance and qualiry; selfnature and other-nature, permanence and annihilation, ven in thei( most subtle and imperceptible forms. never escaped his penetrating intellect. lt seems as
if he had read the Buddhist discourses, manuals, and commentMies, examining
every sentence, every word, and every syllable. Even if one cannot discover any
tangible evidence to provide that he was a "freed one". a nirvr.ta, the Kiin'kii,
indeed, bears ample testimony to .liis supreme intellectual starure.

Epistemological Jnvestigations
Nagarjuna was an empiricist par excellence. H owever, .the fundamntaJ
metaphysical assertion of most rationalists, and even the empiricists during his
day. was the cogito, the atman that sees itself before it comes to perceive
anything eJse. Nag;irjuna had oo hesitatioo in demolishing this metaphysical
idea at the vety outset. "Seeing oneself' (svatmiinaf!Z daranaf!Z) is rejected, not
on the basis of any di.alectical argument, but simply on the grounds of its nonavailability (flT.2). Comparable to the manner in which David Hume refused to
accept the notion of a cogito, Nagarjuoa proceeds to sl!ow that the so-called
process of "seeing oneself' is do more than "the arising of consciousness
depending upon the eye and visible form" (cak1u-rupe prafityaivam ukto
vijfziina-Jaf!lbhavaf?). III. 7). that is, a perception of some color, shape, etc.
Howevcr, Nagarjuna differs from Hume in not recognizing these ,perceptions

82

THE PHILOSOPHYOPTHE MtDDLE WAY

as momentary and discrete impressions on the basis of which we construct our

world-view. Momeotarjness, along with its philosophically unacceptable consequeoces, was rejected, especially in his treatment of "motioo'' (g11tiigata, Il)
and "time" (kii/a, XIX).
Thar Nigarjuna was rejecting sense experience i11 fo11or ofa special intuicion
is not ac all cvidenr from his treatment of sense expecience in Chapter III. His
relendess criticism of a mctaphysical cogito does not mean that he was evading
the problem. On che contrary, he was quocing a scacemeot from the Buddha co
show what a non-mecaphysical description of sense experience could be. That
descriptioo in terms of "dependence" is funbcr clabo~tcd in Chapter :XXVI
where he presented a quite positive explanation of the human personaJity as
weJI as its experiences.
Indeed, if "emptiness" (funyalii) were to be an "ultima te reality," there was
no reason wby Nigarjuna sbould noc have devoted at least one chapter of bis
work solely to explicate this conception and provide ioformation regarding its
epistemological basis. At least a chapter on "wisdom" (jna11a), explaining how
it penetrates into the ineffablc ultimate truth, abolishing aJI linguiscic conventioos in the process, would have established the basie philosophical standpoinc
aruibuted to the Mahayana schools by most classicaJ and modern scholars. No
such attempt is made in the Kiiriliii. On the coouary, the term that he most frequencly uses is payati, meaning, "perceives." He uses it in the same sense in
which the Buddha utilized it in th~ "Discourse to Katyayana." Often what he
claims not to perceive (na paiyatt) is self-nature or substance (svabhiva) or permaoeot existence (bh'ii11a, asli/1111). What he claims to perceive (pafyalt) is
dependendy arisen phenomena as vlell :s dependent arising. Such perceprioos
the
are not presented as che results of a special inruitioq, but primarily
absence of igoorance (4,virlya) or confusion (muii) created by onc's disposidons
or incli.nations for the extremes of substantiaJ existence and nihiliscic nonexisteoce.
Picking up the most impoctant epistemological theme from the Buddha's
discourse to Kacyayana, Nagarjuoa is insisting that when one perceives chrougb
widsom (jfiana) the arising and ceasiog of phenomcna, one abaodons the two
metaphysical explanatiooS' of that experience. Indeed, the theme chat is emphasized is not the perr:eplion,of a non-Wsing and nonceasiog ultimate truth,
bur rather the no11-perception of a metaphy~ical enticy that is oon-acisiog and
non-ceasing. Thus, for Nigarjuna, sense ex~Frience, explaincd as a process of
dependence, serves as the foundation of human knowledge. Concentrating his
attention on this foundation of human knowledge and understanding, Nagarjuna not only leaves out any discussion of speciaJ intuitions not related to sense
experience, but also avoids any refereoce to the SOcalled "extraordinary perceptioos" (abhijifii), probably because such perceptions had by this time come to be

of

}NlJlODUCl'ION

83

considered absolutely independent of sensory experience, even though this was

not the way in whic.b the 8uddha per~eived them . 1 '~ Nag.1i'rjuna may have been
awate that, even according to the Buddha, human beings whose six sensory
faculties arc not functioning properly could not develop such perceptions. For
example, in the early discourses, one canilot come across any reference to someooe who is blind by birth developiog "clairvoyance" (dwba-callkhu) or one who
is deaf evolving the capacity for ''clairaudience'" (dibba-sota).
Furthermore, Nagarjuna seems to have ditected rus attention more to the
fundamental problems that generated metaphysical theories. And the problem
of sense experience, indeed, was at the very top of that list.
As reiterated in the above analysis of the Kiirika as well as in the annotation
of the text that foUows, the empicist Nigirjuna continued to insist upon
evidence from experience for any idea before ie is accepced . The repeated us of
the pbrasses: na vidyata (meaning "not evident,'' "not known," "not
petceived," etc.) aad nopalabhyate (implying " not obtained" or " not
available") bears ample testimony to his predominantly empiricist attitude. In
the absence of any positive evidence that Nagarjuoa rejected sense experieoce
and accepted a higher "intuition," there is no reason to doubt chat he was cal
ing for evidence that is provided by sense experience. In fact, whenever the
phrase na upapadyate ("is not proper," "not appropriate") is used to describe a
situation, it wouJd often follow a statement implying chat it is not evjdent (na
vidyata). this means that rational or Jogical arguments atte.mpting to draw implications need to be strongly grouoded on empirically verified permises. A
thesis had lo be first found fault with before its ancithesis is faulced . "Selfnature is not evident" (na hi svabhavo. , .vidyate), not because "othernature" (parabhava) is not evident, as would seem to be the case if Nagarjuna
were to merely utilize the method of reductio ad absurdum (prasanga). Selfoature (svabhiiva) is noc evideot, because it is not available i.n expecience. It is
only aftet such a straightforward empirical statement that Nagarjuna proceeds
to reject "other-nature" (parabhiiva), for this latter would make no sense
without the conception of "self-nature." Indeed, it is the unknowabiHty of selfn.arure (avidyamiine svabh?ive) that destroys the very conception of othcrnature (1.3), not simply the relativity of the two differeatiated concept.s.

Onto/ogy
Just as much as the assercioo of a cogito led to the belief in a substantial
agent, a pudga/a, the lop-sided attempt to dissolve that coacept insisting that it
is dependent upon the elements of experience (dharma) led to a substantia!lst

84

THE PHILOSOPHY OF THE MIDDLE W AY

view reeating to the objective world. The conception of a person was replaced
by a substanciaJ worJd. The Buddha bad spoken of obsession (prapaflca) as che
reason for sucn world-views. Hence his emphasis on the "appeasement of obsessions" (prapacopaiama) as a means of overcoming metaphysic;s. Looking at the
philosophical background in which the objective world bad replaced the cogito
as an ultirnate reality, Nagarjuna was more specific. in insisting upon the "appeasement of the obje~ (drfl.!ftWyopaiama) as a means to attaining true
knowledge (V.8).
In fact, the Sarvastivadins and Sauuantikas, whiJe denying the substantiality
of che human person (pudgala), had rnoved to the other extreme of adroitting
the substantiality of elements . (dham111). This problem being foremost in
Nagarju na's mind, he devoted the second part of his treatise to its refutation.
Io this case, he reaJized that even ccitical philosophy bad fallen prey to the ot.
dinary human search for security and absolute certainty whea philosophers,
compe1led to give up something that provided them with some sort of certainty
(and in chis case the cogito), wece clinging like leeches to an objective wocld as
an ultima te reality.
1f Nigarjuna were to be an empiricist Jike the Buddha, he could not confine
hiroself to a world of abstract ideas . He knew that the Buddha was a ''verificationist" (ehipassika) and that this invoJved concrete "identificacion" (co use ct
term popular in modern philosopby). 1 n Thus, che conception of a person
begins with an identifiabJe and rc-identifiable "form" (rupa). Whilc the ud
dha recogni:zed the possibility of experiencing formless (artip11) states, it is not
very dear whether he bad accepted "forroless" persons. However, unlike some
modern philosophers who would consider the "body" or materia form to be
fundamental and, therefore, ulcimately real, leaving a1l otber conscituets as
being secondary, the Buddha wouJd merely look upon che "body" as a
necessary but not sufficient part of the human person. For him, feeling
(vedana) , perception (1afla) , dispositions (1ailkli?ir11), and coosdousness
(vinfiii'IJll) were as important as the materia body in making any identi.fication
or re-identification. Thus, the eliminacion of the oogito by expJaining it as
something that is dependently ai:isen (pa{icc4.Iamuppanna) did not mean the
recognition of the uJcimate reality of thesecondicions upon whic.h it depeoded.
Hence the Buddha's famous dictum: "Al1 things are non-substantiaJ" (1abbe

dhamma anatta).
For this reason, Nagarjuna's first major enterprise in the Kiirikii is to
escablish the non-subscantialicy of the elements (dharma-nairatmy11) . This cal
ed for a cririque of the Sarvastivada conception of substance (svabhava). As
mencioned earlier, che epistemologicaJ mechod by which he tried to achieve
this was "appeal to"experieoce" . In othet words, he was calling for identifi1tion
of substance, which nooc of his opponents were able co do.
However, Nigarjuna believed chat an ideutification of an eventcan be made

l.NTRODUCTION

85

on .the basjs of "fruit" or "effect" (arlha), for, according to bim, io the absence
of a fruit, one cannot speak of a condicion or non-condition (1.14). This is a
pragmatic theory of truth or reality. However, jf his substantialist opponenrs
substaoce can be identified through irs fruit ot effect,
were to insist tbat
Nagarjuna's immediate response is that the fruit or effect is dependent upon a
condition (or a set of condfrions), whereas substance is not. The definirion of
substance (s11abhava) as "having its owo (svo) existence (bhavo)" stood in the
way of Nigarjuna accepting his opponenes explanation. "How can a substance
be condicioned?" (s1111bh'iivaf? krklko nama bhavt~yali puna!? katha~ . XV.2)
grumbled Nagarjuna. For him, the conception of a "dependent substance" was
no mote t~an a ~elf-cootradictioo (XV. I).
Having criticized the substantialist theory of elements (dharma) , Nigarjuna
had to teturn once again to the cooccption of a person, evcn though he bad
begun his treatment of the non-substantiality of elcments by rejecting a.
substantialist conception of a person (Chaptei: Ul). The reason for this is not far
to seek. The Sautrantikas, who had themselves tejected the Sarvastivada on
ception of substance (s11abha11a), were surrepcitious1y iotroduciog a subtle personality (pudgala) to account fot human behavior, ~oraJ responsibility, bondage: and frecdom. Chaptets XVI-XXVI were, therefore, dcvoted to the expJanation of the human personality without falling into the substantialist trap.
Unlike some of his predecessors and most of his .modern day admirers,
Nigarjuna was indeed cogoisaot of the possible dangers involved in a "noosubstantialist discoutse ." He was aware that t he idea of non-substantiality could
eliminate even the empirical conception of a person (pudgala) and of elemcnts
(dharma) and enthrooe itself as the ultimate truth or reality. Hence his rather
bold dedaration: "Those who arc possessed of the view of emptiness are said to
be incorrigible." (Xlll.8)
An abso utistic view of emptiness would certainly contradict his empiricist
method that calls for identification as a test of uutb or reality. "Non
substantiality" (nairatmya) or "emptiness" (funyata) , taken in rhemselves,
would be as absuact and uoidentifiable as a substance (s11abha11a). lndeed , as
pointed out earlier, the notion of a substance was rejected because it could nor
be identified with anything in experience. Therefore, there was no excuse what
soever for replad ng "subsrance" wich an equally undefinabe or unidentifiable
coocepdon of "emptiness" or "nothingness. " "E.mptiness!' (flir!yata)
distinguished from "the empcy" (funya) , "non-sul>stantiality" (nairatmya)
separated from "the non-subst'antial'' (nairatmya-dharma) or "dependent arisi.ng" (pratityaramutpada) differenciated from "the dependently arisen"
(prafftyasamutpanna-dharrna) \\muld be as unidentifiable and therefore
nonsensical as any other metaphysical conception thar Nigarjuna was
endeavoring ro refute.
Thus, for Nigacjuna, emptiness (Junyata) was no more than what is implied

the

86

THE PHILOSOPHY OF nm MIDDLE

wAY

in the scacement: "All this is empty" (sllf'tlam ir/af?J Junyam) . The scatemeot,
"All this is empty ," is not idencicaJ with the scatement, "All is empty," (saroaf?1
funyam). In face, as pointed out in the anootacion, nowhere in the Kin'kii can
one come across an absolute statemcnt such as "All is empty." lt is iodeed
significant that even when malcing a universalized stacemeot Nagarjuna retains
the demoostracive "this" (idaf?J) in order to elimioate the absolutist scing.
The question whether one can speak of "empciness" (JUnyat'ii) of
"emptiness" (Junyat7i) is often raised in discussions of Nagarjuna's "middle
way." Por example, one could maintain chat "emptiness" itself is an idencifiable parcicular on the basis of which a universaJ "empciness" coud be idencified. The language udized by N:igarjuna docs nor permit sucha fligbt into
the realm of the Absolute. His conception of "the empty" (fUnya) is a particular. Yec this particular is not equivalent to a parricular "empciness"
( unyatii) abstractcd from a concrcte situacion. To movc from "the empty" to
"emptiness" is an altogether differeot process. The former is grounded in an cxpcrienced situation of an eveot with a charactcristic, while the later begins wich
a characteristic sans the event.
This careful avoidance of any absolutism or substantialism in reJacion to the
conceptioo of "dependent arising" (prafftyasarnutpada) as well as "emptiness"
(Jiinyatii) was dedared by Nagarjuna as the "middle path" (ptaJipat saiva
madhyamii, XXIV.18). That "emptiness" is a "dependent convencion"
(upadiiya prajnaptt), for it is dependent upon and, therefore, identifiable in
cetms of, "the empcy" (fUnya). Nagarjuna assercs that- ''emptiness" so identified would elimioate aoy dogmatism or obsession (adhilaya) and, aJong with
it, any erroneous views (do1a-pra.raga, XXIV.13).
This, undoubtcdly, is a beautiful rcscoration of the Buddha's conception of
"non-substantiality" (anatttJ). However, the modern interpretacion of Nagarjuna seems to move in a totally different directon. Modern scholars, favoring an
interpretacion by Candrakrti made knowo to chem by T. R. V. Murci. insist
chat Nagarjuna had no thesis of his own (svapak/a) to present. This Vedantic
incerpretation prescnts Nagarjuna as a critical or analyticaJ philosopher whose
sole funcicon was to crit:icize or analysc (vigraha) views presented by others
without having to recognize or uphold a view of his own. Such an interpretatioo has led to two more related theories being attribured to Nagarjuna. The
fi.i:st is the admission of the ioadequacy of conceptual thinking, and therefore
of language, to express the ultirnate truth. The second is the aruibution of a
coacept of ultimate truch in the form of "absolute emptiness" or "absolute
nothingness" inexpressible through otdinary human linguistk apparatus,
Thus, we are led to one of the most troublcsome quescions relating to Nagarjuna's philosopiicaJ enterprise.
Io che annotation of the dedicatory verses of thcKiirikii, we have suggesced a
different rcading which would make ie possible for Nagarjuoa to make the

lNTRODUCTION

87

claim that he is presencing a right view (1amy(lg-drf!t) when refuting the inappropriate views advocated by his opponents. Furd!ermore, Nigarjuna clearly
indicate that philosophicaJ enterprise consists not only of analysis (vigraha) but
aJso explanation (vj"iikhjiin11) (IV.9). This would provide legitimation for the
most positive explanation of the Buddha1s view in Chapter XXVI, in- addition
to ocher positive statements made by Nagiirjuna elsewhere (e.g Chapter
XVU).
However, the two most troublesome questions regarding ultimate re~ity and
the inadequcy of language still remain, primarlly because of the manner in
which the Vedantic scholars interpreted three related terms utilized by Nagarjuna. These arc praftiiirtha, nirvikalpa, and sa'l!,Zvrti.
For most scholars who have been nurtured in a predominantly substantialist
philosophkal or religious tradition, paramartha means ''ultimate reality," nirvikalpa iroplied "the non-conceptu1 ," and Jaf!lvrti stood for "lar:iguage." In
understanding these three terms in this maooer, did modern translators and interpreters impose theiI owo substantjafut outlook on Nigatjuoa's thought?
The first test of the validity of such translations would be a comparison of the
implications of these three traoslations with the Buddha's own conceptions of
"dependent arising" (prafftytJJamutpada) and "non-substantiality" (anatman),
all of which Niigacjuna ac<!epted with reverence.

Philosophy of Langu4ge
The term 1af!111rti (Pali, sammult) was never used in the earJy discourses
to refer el(clusively to language. Analysing the Buddha's philosophy on the
basis of the early discowses, it was pointed out that sammuti, voh'iira, and
pa~fiatli were terms used to refer co any convention, not merely linguistic convention. T)le specific terms used by the Buddha to refer to language are nirutti
(etymology) and adhivac4na (definition or St!mantics). Nagarjuna's use of the
term abhidheya, meaning "that which is to be designated,'' (XVIIl.7) would
provide us with a term that he may have used if he bad a need for refercing to
Janguage. Such :i term would be adhidhana and would not be semancically
much different from the tetm adhivacana used by the Buddha.
However, for Nagarjuna, the abhidheya or "that which is to be designated"
ceases with the cessatioo of citta-gocart1 ("the ojbect of thought"). Thus,
anything that is not the object of thought, that is non-conceptual (ni'.rvi.kalpa) ,
is also not describable. If so, Niigarjuna had no reason to compose more than
four hundred verses rryiog to explain the indescribablc. It would be a fruitless
attempt on the part of any philosopher, let alone one who is extoled as an
"enligbtened one" (budt/h4) .

88

THE PHILOSOPl"IY OF THll MI DDLE W AY

lf Nigarjuna was ttying to explain soroething and in that pcocess was utiliziog language, he would be dealing with the conceptual or the object of thought
(citta-gocara). Accordingly, anything that is conceprual would aso be the object of thought, and the noo-conceprual (nif"IJika/pa) could not be an object of
thought. A truth that cannot be thought of, let alone one chat cannot be
spoken of, would be as metaphysical as the conception of atman in the
Brahmanical speculatlons. Nirvikalpa would, thetefore, mean something else.
In the course of the analysis of the Kiirik.ii, it was pointed out that Nagarjuna
was critical of a spedfic form of discrimination, a discrimination that produced
polarities in human thinking. These coosisted of existence and non-existence,
substance aod quality, self-narure and other-nature, permanence and annih i a
tion. In such a context, nirvikalpa would refer co polar discriminations, not any
and every form of discrimination.
This lcads us to one of the mosr conuoversial discriminations that the
metaphysicians of the Buddhist tradition as wel as their couoterparts of other
substaotialist traditions made with regard to bondage and freedom . Chapter
XXV of the Kiin"k'ii is devoted to an analysis of tbis metaphysical issue. In
Chap ter XXIV, when Nigarjuna spoke of saf!'vrti and 11ya11ah'iira oo the one
band and paramiirtha on the other, be was paving bis way for a discussion of
the discrimination between bondage and freedom.
Nigarjuna, who provided every indicacion that he had read the early
discourses, could not bave been unawarc that the Buddha used the terms sammuti and 11ohara (11y1111ahara) moce ofcen in the sense of mora conventions.
Tbese mora convencions pertained to good (dharma) ~od bad (adharma) .
Thus, whenever he used the term 11ya11ahara, Nagarjuna was referring co the
mora convcncions of good and bad (dharma-adharma) (XXIV.36) or merit and
deroerit (p1''!-Ja-papa) (XVU.24). These moral conventions are accepced not
because they are mere conventions agreed upon by consensus but because they
work. They are pragmatically grounded. They produce fruits or conseuences
(artha). Such convencions provide a basis for idea~ conventions referred to as
paraiiirlha ("highcst fruit or consequence") . Yet, to safeguard the idea from
becoming a merc idea and not a fact, Nagarjuna insists upon the
dependence of the idea on the concrete. A quotacion from William J ames,
eveo though extensive, seems to be relevant herc.

If the cthical philosopher wece only asking after the best imaginable system of goods he would indeed have an easy task; for aU
demands as sucb are pnma facie respectable, and the best simply
imaginary world would be one in which every demand was gratified
as soon as made. Such a world would, however, have to have a
physical constitution entirely different from that of the one which

!NTilODUCTION

89

we inhabit. It would need not only space, but a time, of n-climensions, to include all the act:s and experiences incompatible with one
another here below, which would then go on in coojunction -such
as spending our money, yet growing rich; ta.king a holiday, yet getting ahead with our work; shooting and fishiog , yet doing no hurt
to the beasts; gaining no end of experience, yet keepiag our
youthfuJ freshaess of heart; and the like. There can be no question
chat such a system of things, however brought about, would be the
absoluteJy ideal system; and that if a philosopher could create
universes aprion', and provide all the mechanical conditions, that is
the sort of universe which he should unhesitatingly create.
But this world of ours is made on an entirely different pattero,
and che casuistic question is herc most tragically praccical. The acrually possible in this world is vastly narrower than all thac is
demanded; and there is always apinch betweeo the jdeal and the
actual wbich can only be goc chrough by leaving pan of the idea
behind. 1!
In a similar way. Nagarjuna, following the pragmatic teachings of the Buddha, could not clivorce paramarlha from Jaf!111rti (i.e., the ultimate fruit from
the fruit of cveryday life of a h11012n being). Just as much as "emptincs.5" is based
upon "the empty," even so paramiirlha bad to be based upon the Jllf!l11(1i.
Without any reference to the concrete concepcs of good, any notion of ultimace
or ideaJ good would be not only meaniogless but aJso "fruitess'' (an-artha) and
tcrribly harmful (as proven by many such instances in the .bistory of mankind).
Thus, for Nagarjuna, ultimate good is not one chat transcends ordinary notions
of good, but merely an cxtension of the so-called goodness rccognized in evcryday ue (11yavahara).
The sharp clichotomy between the orclinary ootion of good and the ideal
good is thus brokeo down. lt is significant to noce chat wheo speakJng of two
truths Nagarjuna utilized the terms sa?'(Jvr# and paramartha. Yet, when he
proceeded to explaio cheir reationship , he utilized the term 11yavahara, thereby
establishing the synonymiry of Jflf!Jvrfi and 11yariahara. The fact that Nagarjuna
was not prepared to create an unbridgeable chasm betwecn Jaf!111r1i or
vya11ahiirt1 on the one hand and paramarlha on the other is clcarly exprcssed in
his famous statement that without the farmer the latter is not expressed
(11yava/iiiram arziiritya paramiirtho na deyate , XXIV.10).
Sinl.ilarly, without understanding the ultimace fruit, freedom is not obtained
(paramiirtham anaga11zya niniiif!tJ'f?l nadhigamyate. ibid. ). This would meao
that frecdom (niniiif!a) itself is not something to be sharply distinguished from
Iaf!jsara or ordinary human life, even though they are not idencical. Freedom
would not be absolute freedom that has nothing to do with human life. lt is no

90

THE P HILOSOPHY OF THE MIDDLE W A'I

more than the absence of certain coostraints (such as greed, hatred, and confusion) in the life of a human being. It is, therefore, the life of an ordinary
human being that is gradually transformed , through the cutivatioo of mora!
precepts, into one of mora petfection. This transformation (ratber than
transcendence) is what is implied by nirrlii'fll.

Mora/ Philosophy
The mora life that leads to the transformation of the human personality is
cleady explained by Nagarjuna in Chapter XVII. The absolutistic interpretation
of Nagarjuna's conception of "emptiness" constrained many of his modern admirers from discussing his views regarding karma and survival,
even though these were part and parcel of the Buddha's teachings. The discussion of karma and survival in the Kiiriliii was thus considered to be
"HJnayanistic," having nothing to do with the so-called Mahayana. As such,
the Buddha's owo views regarding these issues turo out to be "J-ITnayanistic" or,
at least, were intended for those low-witted disciples who surrounded him.
Con traty to this view, our analysis of the contents of Chapter XVII , placed in
t he background in which Nagarjuna lived, shows that he was more positive
than his modern day disciples in his treatment of karma and survival. Nagarjuna's major endeavor in this chapter is to rescue the Budda's discourse on
mora responsibility from the ha:voc created by the substantiaist thinkers who
assumed karma to be either substantial or performed by a subsrantial agent.
His was not an attempt to dissolve the conception of karma in favor of an absolutistic notion of "emptiness,"
lndeed, the chapter begins with a reference to the Supreme Sage (paramar!t)
whose docuine he was about to expound. Speaking of the morally good life,
Nagarjuna uses the term dharma, instead of karma, and this may have confounded the modern interpreters. The term dharrna, as explained earlier in the
discussion of the Buddh~$ philosophy, was used both in an ontologica sense
and in an ethicaJ context. Nagarjuna h imself followed this P.ractice, as indicated in the annotation of the Kiirikii (see XXIV. 36). Thus, in the present
context too, dharma me.ans good karma and these are identified as (i) selfrestraint, and (ii) benefitting others. Nagarjuna's selection of these two types of
action as the foundation of mora behaviot is significant: They are an echo of
the Buddha's own first sermon to the world that advocated a middle path be
tween two extremes of self-indulgence and self-mortification. The ''Middle
Way" (miidhyamika) philosopher par excellence could nof have ignored the

1N1"RODUCI10N

91

ethical middle path of the Buddha. He knew the implications of that fi.est sermon. Self-restraint, but not self-mortification or sclf-immolation, c-0nstitutes
one of the foundations of mora life. In this regard, Nigarjuna was not advocating the extremist form of behavior sometimes extoled as the ideal of a
bodhisattva both in the Theravada and in the Mahiiyana. Sef-rcsttaint is a
nece$ary prercquisite for any akruistic activity, which .is the second of the virrues extoled by Nigarjuna as a "friendly way" (m11itrat?Z dharma'!J).
The irnplicacions of the maral Jife recommended herc should not go unnoticed.
In spite of the exceedingly popular theme emerging among the Buddhists during Nigarjuna's day that emphasized extreme altruism, Nigarjuna seems to be
playing a rather moderate tune recognizing the Buddha' own words in the
Dhammapada (166): ''One should not neglect one's own welfare through ex~
cessive altruism. Having understood one's own welfare, one should be devoted
to true welfare."
A reader of the early cliscourscs. cannot but be impressed by the idea of
human behavior advocated by the Buddha:. The noblest person according to
the Buddha is one who avoids suffering for himself as wel as others (attabyabadha; parabyabadha). 139 Thus, a noble action should be one that contrib1;1tes to one's own happiness as well as the happiness of others. This invoves

(he recognition that, while abandoning a belief in a metaphysical self, one has
ro cutivate compassioo for one's own person. At the same time such compassion should be extended to others as well. Nagarjuna seems to have picked up
th.is theme well when, unlike many Buddhist writers of his day, he emphasised
that self-restraint and benefitting others are both acts offriendliness (maittam) .
He was simply insisting: "Be a friend to yourself and be a fricnd to others." This
would certainly be opposed to the idea that caUs for complete and unqualified
self-sacrifice, including self-immolation . Thus Nigarjuna, the founder of the
''Middle Way" (Miidhyamika) school, could not have recommended amore
sober mora life t han one which avoids the rwo extremes of destroying onself
and destroyiog others .
. These two friendly way'5 are recommended by Nigarjuna because they arc
fruitfuJ not only in the present life (iha) but also in an after-life (prelyt1). There
could be no doubt tbat herc he was recognizing the possibility of human survival. The st1rruara or "life-process" referred to at XXIV.10 need not be confined
to this prcsent life ~lone. On the cont.rary, it refers to the continuity of te li.(eprocess through scveral births and deaths, refetred to as pu""1bbharJa in the early
discourses. That continuity, along with its attendant suffering, is to be
eliminated by the development of.wisdom (jfl'iina) which for Nigarjuna consists in the avoidance of all metaphysical views (dr1ft).

92

Tm PHILOSOPHY OF TH!: MIDDLE WAY

Knowledge Leading To Freedom

Knowledge that leads to freedom is not omniscience (sarvajflafii). Nowhere
in the Kiirika does Nagirjuna refer to omniscience, even though it w.Is a
popular theme among the 'fheravadins and the Mahayanists. In the absence of
omniscience, what form of knowledge could lead man from bondage to
fceedom? Nagarjuna refers to "a wise one" (vidriiin) who, _through his ~ercep
tion of the na.w re of truth (tattva-daran'iit), does not accumulate disposicions
(saf!Zskara) that lead to wandering (saf!Zsiira)(XXVI.10). Thus, for Nigarjuna,
as it was with the Buddha, the problem lies in the accumulation and put:su.it of
one's disposicions. Yet without following the dispositions a human being is
unabe to deal with the nther complex and cxcessive sensory input. The "big
blooming buzzing confusion" of experience has to be faced without the aid of
omniscience. The task is reodered exuemely difficult because the dispositional
teodeocies tbat are a neces.sary means of dealing with sucb experience also lead
to extremes, especially when these d_ispositions arc dominated by ne's ikes
and dislikes. When they are dominated by likes and dislikes, they produce
perspcctives oo the basis of which one looks at the world , two of these being
cternalism and anni.hilationism. In order to adopt a middle palli avoiding these
two extrcmes, one needs to eliminate the likes and dislikes and thereby appease
one's dispositions. A pe.rson who has achieved the state of the appeasement of
dispositions (Ia'!Jsliiiropaama) (and this would include the appeasement of the
object of perception (dra1tavyopaama), whether that o.bject be the cogito r
the real external world) is said to have attained erilightenment and freedom ,
Such a person is cnlightened beause he comes to perceive things as they iave
come to be (y11thabhutaji11na = t11tivadarfana), and he is free because he does
not adhere to any dogmacie view that rules out other possibilities. The difference between a metaphysical view cricicized by the Buddha as well as Nagar
juna and the "middle position" (madhyamiipratipat) accepted by both is that
t he former is a closed view while the latter is an open one. An open view does
not subscribe to an absolute discrimination as either I or. The very idea of opeoness imiplics non-grasping (a1111p"iitliina). Thus, whcn both the Buddha and
Nigarjuna emphasized the renunciation of all views (!arva-dr1t1~praha11a),
thcy were insisting upon abandoning all forms of dogmatism with regard to
views. For them, non-attaC.hment to views does not necessarily mean having
"no-views."
Inappropriate rendering of Sanskrit tet:1Il5' into English seems to have con
ttibutcd in some measure toward t he myth that Nagarjuoa bad no view to ex
press.140 The passage often quoted in support of this myth occurs in
Nagarjuna's Vigrah11vyii111Jrla1ii (29): "If I would make any commitment whatevd',

lNTRODUCTION

93

from that I woud iocur such error. Oo the coouary, I do not have a commiuneot.
Therefore, there is no error on my part."
The term that occurs herc is pratijfla, which has beeo translated as a simple
proposition or statemeot. lt is much more thao a simpe proposicioo or statement. lt is a commitment and should be contcasted with vyiik.hy1in(I, "explanation ," (IV.9). While avoiding the former, Nagarjuna continued to resort to the
fatter (see aso XVII.13, etc.). As such , it would be highly inapprop.riate to
compare Nagarjuna's philosophicaJ method with that of Ludwig Wittgenstein,
fot example by quoting bim as follows: "Philosophy simply puts everything
before us, and neither explains nor deduces anything. -Since everything lies
open to view, there is oothing to explain.''1 1 In fact, this quotacion
misrepresents Wittgenstein's owo approach to metaphysics: The most important part of the statemenr has been omitted . lt reads: "For what is hidden, for
example, is of no interest to us." If this c.tucial statement is retained, then Wittgenstcin's thought can certainly be comparcd with Niga.rjuna's or even the
Buddha's. This oroitted part of the statement makes it abundantly elear that
wbat Wittgenstein was not willing to cxplain is "wha~ is hidden," and this
"something'' is, indeed, comparable to what Nagarjuna was rcfcrring to as
kif!'Cit or kaicit, -that is, the hidden substance in phcnomena. Neither "the
empty" ("iinya) nor "emptiness" (Hi11yata), neithcr "the dcpendently arisen"
(ptalftyasamutpanna) nor "depehdent arising'' (prafityfl!4mutp4tla) represeot
a hiddeo something whkh Nagarjuoa was rcluctant to explain. On the con
trary, if it an be shown that Wittgenstein did not provide any explanatiorz of
experience, or did not attempt to formulate in linguistic terms w hat a tcue experience is, as opposed to a confuscd one, then he could certainly be enlightened
by the language of "emptiness" or of ''dependence" adopted by the Buddha
and Nagarjuna.

94

THE PHILOSOPHY OF THE MIDDLE W AY

NOTES
1. S 2.16-17; T.ra 12. 19 (Taisho 2.85c).
2. S 5.420-424; Tsa 15.17 (Taisho 2.103c).

3. For a detailed historical study of the Kathivatthu, see S. N. Dube. Cross

Cu"ents in Early Buddhism, New Delhi: Manohar, 1980. The study, of this
work will be greatlr facilitated by the recent PTS publication of a more carefully
edited version of Buddhaghosa's commcntary by N . A. Jayawickrema.
4. Lung-1h11-p 'u-1a-ch'u11n, Taisho 2047.
5. Kenneth K.. Inada, Niigfirjuna. A
Tramlation of his
Miilamadhyam11kakiirikli witl; an lntroductory Rrsay, Tokyo: The Hokuseido
Press, 1970, p . 3.
6. DhsA 76.
7. Sdmp p.43 ff.
8. Vin 2.241; see inoutnerable refereoces co the Chabbaggiya as miscreant.S
whose conduct was responsible for the formulation of many a rule of disciplioe.
9. Kp 105.
10. Sdmp p.45; see aJso MKV p .306.
11. T. R. V. Muni, Centrf/ Philosophy ofBuddhism, London: Allen & Unwin, 1960, pp.49- 50.
12, Mark Siderits, book review in Journal of Indian Philosophy, 8
(1980): 192 .

13. A. K. Wardcr, Indian Buddhism. Delhi: Motilal Banarsidass, 1970,

p .376.

14. Karikli XV.7.

15. 2.17.
16. Tsa 12. 19 (Taisho 2.85c).

17. Kp 60- 61.

18. Ibid. , 56.

19. See Whalen Lai, "Chinese Buddhist causation theoties: An analysis of

the si.nicie Mahayana uodecstandiog of Pr4/1!J1JJ11m11tpada," Philosophy East
and West, 27 (1977):241- 264.
20. Richard H. Robinson, &r/y Miidhyamika in lndia a11d China, Madison:
University of Wisconsin, 1967, pp .83- 88 .
21. Rg-11edo x.90; aJso Brhadiirayaka Upaniad 1.4,11- 13.
22. See David J . Kalupahana , Causality. The Central Philosophy of Buddhism, Honolulu: The University. Press of Hawaii, 197 5, p .39.
23. S 2.17. Cf. Tsa 12. 19 (Taisho 2.85c).
24. D 1.12 ff.; Ch'ang 14. 1 (Taisho l.88b ff.). For a detailed treatment of
these views, see K. N. Jayatilleke, &rly Buddhist Theory of Knowledge , London: Allen & Unwin, 1963, pp.23- 168
25 . Ud 1.
26. D 1.58-59; Ch'ang 17.1 (Ttlisho1.l07a).

INTRODUCTION

95

27 . Ud80.
28 . M 1.262-264; S 2. 28, 70, 96; Ud 2; Tsa 10. 7 (Tais/Jo 2.67a); 14.16
(Tai.Iho 2. lOOa), etc. See Kalupahana, Causality, p .9.
29. S 2.18.~ Tsa 12.20 (Taisho 2.86a); sec also Kalupahana, Ct1us4'ity,
p.5-6.

30. Ibid .
31. Ibid .
32 . S 2.25; 3.3-4; Tsa 12.14 (Taisho 2.84b).
33.
34.
35 .
36.
37.
38.
39.
40.
41.
42.

Ibid . Also M 1.167, 324, 325.

M 1.415-417; Chung 3.4 (Tais.ho l.436a ff.).

Thag 304.
M 1.395; also 3.237.
Sn 68.
See Kalupahana, Cau.sality, pp.177-183.

M 1.160-175; Chung 56.1(Taisho1.775c ff.).

MKV pp.264, 492.
Ibid., p.374.
D l , 202 (where all four terms iiimafJfla, nirotli, vohara and paflfJatli arc
used as synonyms); Sn 897.
43 . Sn 897.
44. Ibid., 903.
45. Ibid ., 897.
46. S 3. 70-73; Tsa 10. 5 (Tai.Iho 2.65c-66a).
47 . M 3.230; Chung 43.5 (Taisho 1. 70la f.) .
48. M 1.395; 3.237.
49. Inada, Niigarjuna, p .6.
50,~ See Kalupabana, Causality, pp.148-149.
51. Ibid . pp.149-150.
52 . Sak.11 p .11.
53. Kalupahana, Causality , pp.152-153 .
54. See Jicyo Masuda, 1'0rigin and Doctrines of Early Indian Buddhist
Schoos," (A translation of the Hsuan-Chwang Version of Vasumitra's
Treatise), Asia Major 2 (1925):17.

55. T. I. Stcherbatsky, The Central Conception o/Btt:ddhi.rm and the Meaning of the Term 'Dharn14 London: Royal Asiatic Society, 1923 , pp.37-38;
Buddhist Logic,Leningrad : Academy of Sciences of The USSR, 1930. I .4-5.
56. Jayawickrema, KtJth'iiv1111huppakarllf!t1a(!hakatliii1 London: ?I'S, 1979,
1 '

p.X'V.

57. K11u, p . l ff.

58. Ibid. p.IIS ff.
59. Ibid. p.456 ff.
60. See Hajime Nakamura, Indian Buddhiim. Osaka: Kansai University of
Foreign Studies, 1980, p.159.

61 . Kp 52.

96

THE PHtt:OSOPHY OF THE MIDDLE WAY

62. Karilia XVU.13; XVIII.12.

63. Marvin Sprung, The Lucid Exposition ofthe Middle Way. The Es1ential
Chapters from the Prasannapadii of Candrakzrli, Bouder; Prajna Press, 1979.
64. Central Philosophy of Buddhism, p .134.
65. Translated into French by E. Lamotte, in Melanges Chinois et Bouddhique, iv (1935-1936).
66. MKV pp.324, 390.
67. See Kalupahana; "The Buddhist Conception of Time and Temporality,''
in PEW, 24 (1974):181- 191.
68. Ibid.
.
69. The Pn'nciples of Psychology , New York: Henry Holt, vol. I, 1908,
p.609.
70. Brhadarar!yaka Upanzjad 1,4. 11-14.
71. See Swami Satptakashananda , Methods of Knowledge. According to
Advaita Vediinta, London; Allen & Unwin, 1965. pp. 41 ff.
71.. Niigarjuna, p .54.
73 . M 2.87 .
74. Ak 2.62cL
75 . MKV p .13 5.
76. S 3.38; Tseng 12. 15 (Taisho 2.607C).
77 . Sakv p.174.
78. S 3.36; Tsa 2.25 (Taisho 2. 14a); S 3.103; Tsa 2.26 (Tai..rho 2.14b-15a).
79. Ud 80.
80. Sn 916.
81. See J. A. B. Van Buitenen , "Studies in Sankhya," Journal of the
American On.enta/ Soci'ety, 76 ( 1956):153-157.
82 . See, for example, the Agama version at Tsa 34. l (Tai'sho 2.214b) .
83. S 2. 18 ff.; Tsa 12.20 (Tai.sho 2.86a).
84 . Dhp 165.
85 . lbid . 278.
86. Sn 886.
87 . Some Problems of Phzlosophy, Cambridge, Mass.: Harvard University
Press, 1979. p. 33.
88 . M 1.482.
89. See Sn 796-803.
90. M 1.111-112; Chung 28.3 (Taisho l.604b).
91. !nada, Niigar/una, p . 148.
92. M 1.111- 112; Chung 28.3 (Tais.ho l.604b).
93 . M 3.19; S 3.103; Tsa 2.26 (Taisho 2.15a).
94. !nada, p.102.
95. Ibid pp . 104-10~.
96. A 3.415; Chung 27.5 (Tai.rho l.600a).
97. Dhp 127.
98. Jnada, Nfigar,iuna, p .lo4.

INTRODUCTJON

97

99. Donald Mitchell, "Buddhist theories of causation-commentary ,'' in

PEW, 25 (1975): 102.
100. M 1.184; Chung 36.2 (Taisho l.658a).
101. Sn 1076.
102. S 1.120: Tsa 39.11 (Taisho 2.286a).
103. Sn 1076.
104. M 1.395; also 3.237; Chung 43.5 (Taisho l.703c).
105. M 1.426; Chung 60.6 (Taisho l.804a-b); M l.483ff.

106. 3.83.
107. M 1.184; Tsa 39.1 1 (Taisho 2.286a).
108. See Kalupahana , "Anna," in Encyclopaedia of Budt/h1:S1n, cd. G. P.
Malalasekera, Colombo: Government of Ceylon . 1961,- 1.693-696 .
109. James, Tlie Pn.nciples ofPsycho/ogy , vol. I , p. 609.
110. D 3.80 ff.; ChiJng 6.1 (Taisho l.36b).
1 l l. DhsA p.421 ; see Ka1upahana, "Schools of Buddhism in Early Ceylon,"
The Ceylon Journal of the Humanities, Peradeniya: University of Ceylon, 1
(1970):1 59-190.
112. Brhadaraf!yaka Upan1jad 1.3.28.
113. Ibid.
114. S 3.110; Tsa 5.2 (Taisho 2.3la-b) .
115. Ibid .
116. lnada, Niig'iitjuna , p . 132.
117. S 3.110; Ts. 5.2 (Taisho 2.31a-b) .
118. A 2.52.
11 9. M 1.140.
120. S 3.54: Tsa 9.4 (Taisho 2. 56b).
121. M 1.228; S 3.133; 4.401 ; A 1.286; Thag 678; Dhp 297; Tsa IO. 7
(Taisho 2.66b-67a) ; Tseng 32.4 (Taisho 2.668c).
122. The Giindhan Dharmapada, ed.]. Brough, London: Oxford Uoiversity
Press, 1962, 190. (also see p . 240).
123. D 1.51; Ch'ang 17. 1 (Taisho l.I07a ff.) .
124. Petavatthu, ed. Minayeff, London: PTS, 1888, 57; Apadana, ecl . M. E.
Lilley, London: PTS, 1925-1927, 395.
12 5. The Conception ofBuddhist Nirvana, Leningrad: Academy of Sciences

of the USSR, 1927.

126. Jnada, Niigarjuna, p .153.
127. Ibid p. 160.
128. Kalupahana, Causality, p.150.
129. Sak11 p.12.
130. See Poussin , L 'Abhidharmakoa, 4.140.
131. See Ian Stevenson, Twenty Cases Suggesti11e of Reincamation, second
revised and enlarged edition, Charlottesville: University Press of Virginia ,
1974.
132. M 3.94- 99; T.ra .2.25 (Taisho 2. 13c- 14a),

98

THE PHILOSOPHY OF THE MIDDLE WJ\Y

133. D 1.12 ff.

134. Sn 152.
135. D 1.191 ff.
136. Kalupahana, Buddhist Philosophy. A Hi1ton'cal Analysis, Honolulu:
The University Press of HawaH, 1976, pp.21-23.
_
137. P. F. Strawson, Indit1idual1, New York: Doubeday, 1963, pp.2 ff.
138. "The Mora Philosophers and the Mora Life," inE.rsays t'.n Pragmatism,
cd. A. Castell, New York: Hafner, 1948, pp.77-78.
139. M 1.89-90, 369-370, etc.; Chung 25.2 (Tgisho 1.586a).
140. See Chris Gudmunsen, Wittgen1tein and Buddhism, London: Harper
& Row, 1977, p.44.
141. Philo1ophical lnvestigations, te. G. E. M. Anscombc, New York: Macmillan, 1958, 126, as quoted by Gudmuosen.

SANSKRIT TEXT,
TRANSLATION AND
ANNOTATION _ _ __

DEDICATORY VERSES

Anirodham 11n11tpiidam anucchedam aiiifvalaf!',

anekartham aniiniirtham aniigamam anirgarm1.tp,
yal? pralilyasam111piidaf!' prapal'Uopaamaf!' ivaf!t,
desayamiisa sa'f!tbuddhaf? laf!t vande 11andafii'f!t vara'l!J.
I salute him, the fully cnlightened, the best of speakers, who
preached the non-ceasing and . the non-arising, the oooannihilatioo and che noo-permaneoce, the oon-idcntity and the
non-difference, the non-appearance and the non-disappearance,
the ~~pendent arising, the appeasement of obsessions and the
ausp1c1ous.

Madhyamakavrttif?, ed. L. de la Vallee Poussin [abbreviated MKV(P)),p.11 ;

ed . P.L. Vaidya, [abbreviated MKV(V? ,p.4.
These incroduccory verses appcir co be equivocal and therefore could accounc for
most of the co.nflicting views in the rwo major Madhyamika traditions: (1)
those of the Prasagikas, represented by Candraklrti and attributed by him to
the ~arlier Madhyamika philosophers like Atyadeva and Buddhapiilita, who
.recogniz.ed no views and merely utilized the reductio ad absurdum method to.
refute the views of their opponents; and (2) those of che Svatancrikas,
represeated by Bhavaviveka, who admitted a positlve thesis on the basis of
which they criticiz.ed the opponeots' views. In these verses, the contents of the
Buddba's discourse acc all referred to in the accusative case as anirodhaf!i, anut-

pada1p, anucchetlttf!t, aiivataf?l anekartha'f!}, ananartha'f!J, aniigama'f!},

anirgamar(l, prafityasamutpada'f!t, prapacopafamaf!t, fiPaf!t without giviog
any indication as to whether they tefer to one doctrine or s~veral.
Modern interpreters of Nagarj_una, probably following CandrakTrti, whose
commeotary, the PrasannaprJtlii.; is the ooly one availabe in its original Sanskrit, have assumed that all these tetms refer to one docuine. namely, dependent arising (parfilyasamutpada) . However. as will be shown below, it is also
possible to explain rhese verse as referring to several different concepts in the
Buddha's philosophy, thus providing justification for the standpoinr of the
'positivists' (svatantrika) of the Madhyamika cradirion.
l. Priisagika interpretation. Cand.rilitti's comments on these verses show

very defioitely how. he moves from a svatantrika interpretation ro a prasangikil

one. He begins his creatise, the Prasanntipadii, by emphasizing the significance
101

102

THE PHILOSOPHY OF rHii MIDDLE W AY

of dependent arising (pralityasamutp1ida). Accepting both the gene.tal or

distributive meaning and the particular usage of the term, he explains dependent arising as "the arising of things cootiogent upon causes and conditions"
(hetupratyafiipek.$0 bhatiiinam utpaehl? prafityasamutp1idaf?, pj), contns,ting it with the definition offered by those who accepted a theory of
momentariness (.!a'l!ill4-11aeh). Accocding to the latter dependent arisiog
mearu "the arising of those that arc repeatedly destroyed," (prali prali ityanit!J
vinafinat!J samulpida iii, loc. cit.). In fact, Candrak'frti seems to defend some
sort of ".tadical empiricism11 when he raises the questio : "How can one maintain that there is arising of that which has reached [another) without obtaining
a relacion?" (Ka1ham anenaiva praptel; sambhava ita yuktyanupidanena, p.9).
He continues to em1>hasize Nigarjuna's view that "wbatever that has arisen
reaching such and such, chat :~not arisen in terms of self-narre" (tat ;tal prapya
samutpannar{Z nolpannttr{Z tal svabh'iivatal;, pp. 9. 10). This certainly means
t~at the negations in the dedicatory verse are intended t deny that things are
arisen through self-nature (wabhii11atal?) and there seems to be no impication
that they are applicable to dependem arising itself.
However, a change of perspective appears when Candrakirti proceeds to cxplain the principle of dependent arising(prrJfity4111mu1pada). Instead of taking
dependent arising as the positive middle position of the Buddha, CandtakTrti
applies the negacions to dependent arising itself, as if the negatio~s arc
presented as adjectives qualifying dependent arising. Hence his statement:
"The entire trearise (i.e: the Kdn:a] purports to establish the absence of cessation, etc. ofdependent arising," ( nirodhidayo na sanli pralityasamutpidasy11,
p. 11).
Th.is eads Candrakrrti to a posicion of "no-vie\vs" whicb is then identified
with the "appeasement of obscssions" {prapaflcopaiama) or freedom (nirv'ii11a),
thereby emphasizing its uanscendence. Quotations fcom Nagarjuna as wel as
Aryadeva are presented as juscificacion for this identification of "dependent
arising'' and "appeasement of obsessions" (p. 16), even though Nigarjun,a
seems to discinguish betwcen utilizing right views (suc.h as dependent arising)
without grasping on to them as the absolute truth. With this.interpretation of
the negations, the dependent arising and freedom, Candtakfrij then moves on
to the contcnts of Nagacjuna's first chap ter on the examirtation of "conditions"
(#atyaya) and invoves himself in a lengthy discussion of the Prasarigika and
Svatantrika standpoints.
2. S1iiitantn'ka interpreialion. It is possible to interpret these eevcn
characterizatioos. as expressiog three major aspect:s of the Buddha's discourse:

a)

The eight negations may be ta.ken as a refutation of the faJse views

D EOICATORY VERSES

b)

c)

103

(milhya-dr:1t1), primlllily the theories of substantial existence (as#lva) and nihilistic non-cxistence (n'1islil11a), that is, the non-ceasing
(aniiodha'!') of a substantial entity and the non-arising (anutpadaf!J) of a non-existcnt cntity. Indeed, the non-ceased (aniruddhaf!') and non-arisen (an111pannaf?J) arc cquivalcnt to the nonempty (tliinya), a term used to refer to substance (wabh'iiva)
(XX.17). Hence these negations appear in couplets and could be
considcrcd as another w:ay of ptesentiog the non-substantiality
(anatman) not onJy of phenomena but al~o of those vicws. They arc
noo-substantial in the sense that thcy arc not absolute, as they were
assumed to be by their proponents.
''Dependent aris.ing" (prafityasamutpada) would then stand for the
midcJe position, which is the right view (samyag-dr:!ft) on the basis
of which the wrong views arc criticized. "Dependent arising" is considered to be the right view, not because it is an absolute trutb, but
because it allows for possible explanations of phenomcna not permitted by thcories of absolute existeocc and nihilistic nonexistence.
"The appeasement of obsessions" (prapalcopaama) and "the
auspicious" (fi11a) would be the result of adopting the middle position. This is freedom or nirvana.

lntcrpreted as such, thesc verses rcfer to a positive core of the Buddha's

ccachings, alongsidc of the negative aspcct which was intcnded as a rejection of
the hcrctical vicws. Such an interpretacioo would lcave dependent arisiog as the
position from which the Buddha rejccted the metaphysical or absolutc views
and clls would support the Svacaatcika understanding of Nagarjuna. If che
analysis of Nagarjuaa's philosophy is undenaken in the light of the "Discourse
to Kacyayana," as is done in the Incroduction to the present work, the above interprccatioo of the dedicacory verses may appear to be moce in confor.mlty with
the Buddha's own tcachings. Such an incerprcratioa would prcscat theKiirikas
as a compact and weU-organized compositioo and eliminatc the accd to prunc
poctions of the text as being irrclcvant or inessential to che main thcme.

CHAPTER

ONE
Examination of Conditions

(Pratyaya-partk.[ a)

I.

Na wato

napi parato na dvabhyar(l napy ahetutaf?,

utpann'ii /atu vidyante bh'iiv'iif? kvacana kecana.

No exi'stents whatsoever are evident anywhece that are arisen from

cheinsdves, from another, ftom boti, or: ftom a non-cause.
MKV(P] p.12; MKV(V) p.4.

The four types of events referred to here are comparable to those m~ntioned by
the Buddha at S 2.19-20, namely saya1pkata1{t para1{tkata'f! sayaf!Zkatafl ca
paraf!Zkatrzfl. ca, and asaya?[Zkaraf!Japaraf!Zkaram adhiccasamuppannaf!Z. Instead
of the term utpanna (arisen), which occurs only in the last phrase, herewe find
the occurrence of che term kata (Sk. kr ta). "done," primarily because in the
Upanifads, which served as background to che Buddha's teach ing, the substaotial self(atman) was 1ooked opon more as a "personal agent," than as a substantial principle (svabhava, pr4k.rti, etc.) . With the sophistication in philosophical
thinking in che later fndian schools., the "personal agent" wa5 gradually r5placed
by an "impersonal substance'' (svabha11a). The S;uvastivadins, who carne to ~c
cep t a conception of substance while at the sametime rejeccing a "personal
agent" (pudgala), failed to nocice the similarity if not the identity of their implications. In this verse, Nagarjuna present:s his oegative thesis, which io the
dedicatory verse. he expressed with the eight negations. lt is the thesis that he
undertakes to prove in the first twenty-five chapters. He has not provided any
arguments yet, except saying that these four kinds of events are not evident.
Pro,>ably he fet that these events needed furthcr ex:pJanatioo before he proceeds to refute them.
However, Candraklrti is all too impatient. His commentary on this verse is
more than one tenth of his eotire work (almost 65 pages), and it is a stupendous
contmentaty filled with lot of metaphysical trivia and diatribes, mosty directed
at Bhavaviveka and the Svatantrika uadicion. After assumi:ng that Nagarjuna
had "no position" (sec note on the dedicatory verses) with which to criticize
105

106

Tul!

PttlLOSOPHY OF nIB MIDDLE WAY

thcsc four theorics, Canc.rakitti: scttlcs down to juscify the reductio ad absurdum
by which the iaherent conuadictioos in a thcsis arc exposcd. He rcalizcs that
sclf-causation (s11ata-utpatl1) is based on the bclief in a permanent and cternal
self or substance (wahha11a). Quoting Buddhapalita, he maintains: '"Things
arc not atisen from self," bccause such arising is meaniogless, (lad utpada11a1yarthyat, p.14).' For, thcrc is oo purposc in the arisiog of things that
arc aJ~dr cxistent. This ccrtainly is Nigarjuna's cricicism of a substantialist
notion of a "condition" (pratyaya) at 1.6. However, in the prcsent verse and at
1.3. Nigirjuna appcars to usc the argument from empicicism to dcny the
su bstantialist view.
Thus, while Nagarjuna was saying that substance is n_ot cvident (na 111dyate)
and. thcrcforc, ioappropriatc (na yujyate), CandrakTrtl was maintaining tbat
substancc is not appropriacc (na yujyate) and , thcrcforc, not evident (na
11idyate). This indced is the diffcrcnce bctwecn cmpiricism and rationaJism, a
differcnce that is soon to lead to the conflict bctwccn the Svatanttikas and the
Prasangikas.

2.

Cat117ira'1 pratyajii hetu ciilambanam ananlartlf?j,

tathaivadhipateyaf!Z ca pratyayo nasli paicamaf?.

Thcre arc only four conditions, namely, primary condition, objectively

supporting condition. immcdiately contiguous condition, and dominant
condition. A fifth condition docs not cxist.
MKV(P) p. 76; MKV(V) p .26.

Caodrakini's commcnts on this verse bave mised almost cveryone who

analysed the cootents of this cbapter. He could not have been unaware
thac the theory of four condicions (pratyaya) was presented ~or the first time by
the Abhidharmikas. However, he falled to distinguish tie Abhidharma theory
(sec AK 2.61-62) from those of the ioterpreters of the Abbidharma , namely,
the Sarvastivadins (like Vasumitra) and the Sautrantikas (see Akb pp.98-100).
He simply assumcd that the cheory of conditions represents an instaoce of extcrnal causation. Heocc his statement: '1'hercfore, sioce things arise froro those
that are exter:tl. there is arising from another" (tasmiid ebhyaf? parabhutebhyo
bh'ii117inam utpaffir asli par11"1a-utpllllir iii, p. 77).

ExAMINATION OF CoNDITIONS

107

In th~ first place. such an intcrpretation would Jcave a rathcr tainted image
of Nigarjuna as an unsystematic philosopbcr, for baving spokeo of four causaJ
theocics beginoing with self-causacion (svata-utpaltt), Nigirjuna is here
representcd as elaborating upon the second, namcly, cxteroaJ causation
(part1fa-utp11t11), ignoring self-causation altogethcr.
Sccondly, wbile the four causal theorics mentioned in 1.1 are catcgorically
deoied by Nigarjuna, oo sucb denial is made of the four theorics of conditions
(pratyaya) . Thus, unlike Candnkrrti, Nigarjuna sccms to have accepted the
Abhidha.rmika theory of four conditions, without characterizing it cithci: as
self-causacion or as externaJ causation. Mter stating the Abhidharma theory,
Nigarjuna then proceeds to analyse .the views of the interpreters of Abhidharma, and, as the ve.tse that immediatcly follows (I. 3) seems to indicate, he found
chat thcse arc the ones who produced theories of self-causacion (wata-utpattt)
and cxternaJ causation (p11rata-11tpa111) out of the Abbidharma theory of conditions (pratyaya).

3. Na hi svabhavo bha11anam prgty,11.fiidi/u vidyate,

1111idyamiine s11abh'iive parabliii110 na 11idyate.
of existcnts is not cvidcnt in the conditions, etc. I.n the
absencc of self-oature, other-natute too is not evident.
The

scf-naturc

MKV(P} p. 78; MKV(V) p. 26.

These indccd arc the most significant statemeots of Nigarjuna in the present
chapter. The fiest statemcnt is nor a simplc but an emphalic denial (na ht) of
the view that the substancc or sclf-nature (s11a~h'ii11a) of an cxistent is found in
the condition (pratyaya). Since the tbeory of conditioos is primarily a Buddhlst
theory. and since amoog the Buddhlst schools the first to advocate a theory of
substance (Jt111bliiit1a, dravya) at tbis early stagt was the Sarvastivida scbooJ,
therc can be little disagreemcnt that Nigitjuna's statement rcprcscnts a
outright i:ejection of the Sarvastivada intcrpretation of the conditions.
This denial necds to be carefully analyscd. In the first place, as noted carlier,
the phr2Se use to express the deaial is n4 vidyate ("is not evident") and not na
Jujyate ("not propcr") or na upgpadyale ("not appropriatc"). Hence the deoiaJ
should be empirically grounded. Secondly, tbcre is no ouuight deniaJ of the

108

Tuli PHILOSOPHYOPTHB MIODLE W AY

"conditions" (pratyaya) but on1y ofsdf-nature (i11abhii11a) . Neither the Buddha

nor the early Abhidharmikas assumed that identity, defined as permanent
substaoce or self-narurc, is a necessary condit.ion for the explanation of coodicioos or of dependcnce (prafftyasamutpiida).
When no such absolute identity is perceived, is it the case that absolute difference is perceived? This would be the case only tf the perceptioos arc confined to the two exuemes (anla) , not otherwise. In the ''Discoursc to Katyayana,"
the Buddha maint.ains ch at he will teach a "middle position" withouc approaching (upagamma) the two extremes of cxistence and non-existence. This
means chat he was providing an explanation of existence without celying upon
this particular form of explanation. And chat middle position allows for an ex1>laoacion of experience or perceptioo of arisiog and ceasing in terms of
dependence. lt is indeed a similar view of existence and non-existence chat
Nagarjuna is denyiog, wirhout, at the same ciroe, denying the docuine of conditions (pratyaya) or of depcndence (prali.tyasamulpada).

4. Knya na pratyayavali niipratyayavali krijii,

pratyaya naknyavanlal/ kn/iivantai ca santy uta.
Acrivity is not coostituted of conditions nor is it not non-consrituted of
conditions. Conditions arc ncithcr constitutcd nor non-constitutcd of
activity.
MKV(P) pp. 79-81; MKV(11 pp.26-27. The focmer reads srmty atu,
which is correctc:d in the laccc:r as santy uta.

The term kn'yii, used io philosophical discourse, can convey two m eanings.
First, it ca-n refer to an inh erent activity, a power or potentiality (akti) in
something to produce an effect (llrlha) . Activity would then be an embodiment
of a condicion (knja praJyaya11ali) oc a condition would be an cmbodimeo t of
activity (praJyayfi knjiivantal/). In either case, the activicy or the coodition is
said to produce the effccc (artha). This, oncc again, is the substancialisr interpretation of causacion. lf the philosophical explaoation of experience is con. fioed to the two alceroatives. chen the conuasting view would be chat accivicy
is not an ernbodiment of a cooditin (apratyayavali knja) or thac a condirion is

EXAMINATION OF CONDlTIONS

109

not an embodiment ofactivity (pratyayii akriyavantaf;), And Nigarjuna says no

to both extremes.
The denial of the above extremes does not rnean che deoial of a second
meaning chat can be atuibuted ~o both krijii and pratyaya, namely, the
pragmatic view which defines borh in terms of the effect (artha). Nigarjuna
was not unawal:e of such an explanation of activity, for in che Buddhist context
kriya is generally identified with arthaknya (-karltra) or simply karitra.

~.

Utpadyate prafityeman iti '1ne pratyayiif! kila,

ya11an notpa4yata ime fiivan tiiipratyayalf kathaT(t.

These are conditio.ns, because depending upon them these [others] arise.
So long as these [others] do not arise, why are they not non-conditions?
MKV(P) p.81; MKV(V) p.28.

The first line of this verse presents a definition of a condition (pratyaya) that
would satisfy the pragmatic sense referred to eadier and therefo,re would be acceptable to the earJy Buddhist as well as Nigarjuna. However, Nagarjuna
wants to make sure that rhere areno metaphysical interpretations of this definition of conditlon. W oud someone assume chat for this.statement to be true the
dependence has to be invariable and eternal? Infact, the Sarvastivada notion of
sef-nature , in terms of which they defined a condition, implied such eternalism. In spite of the Sarvastivada assertion, no such guarantee can be given
on empirical grounds. [f so, it is appropriate to ask the question as to whet'hcr
the so-caled condition has to be caled a non-condition so long as the effect
does not arise. This means that it is inappropriate to say that a condition is such
by its own naturc (svabhava). Inscead , it bec0mes a condition depending upon
the arising of the effect.

6.

Na,iv'ii.rato naiva .ratai? pratyayo 'rthasya yujyate,

a.ratai/ pratyayaf? ka.rya JtJfa ca pratyayena kiT(t.

110

THE PHILOSOPHY OF nm MIDDLE WAY

A condition of an effect that is either non-existent or trostent is not pcoper. Of what non-existcnt (effect) is a condition? Of what use is a condition of the existent [effect)?
MKV(P) p.82; MKV(V) p.28.

Herc the condition (pratyaya) is exa mined in rdation to che effect (llrtha) . Even
though the criticism up to now has bcen directed on the metaphysical notion of
a subscantial condition, and not on a pragmatic definition understood in relation to the effect, the present verse is intended to clacify the nature of the effect. The question is: In terms of what kind of effect sbould a condition be
defined? An existent effcct or a non-existcnt effect? An effect cxistent in terms
of self-nature needs no suppon for its arising and, as such , a condition would
be meaningless. An effect that is non-existent in the sense of being absolutely
diffetcnt from the condition will not be .telated in any way.to a condicion.

7.

Na san nasan na sad asan dharmo nirvartate yadii,

kathaf!J nirvartako hetur evaf!J sati hi yujyate.

Since a thing that is existent or non-existent or. both existent and nonexistent is not produced, how pertinenc in that context would a- producing cause be?
MKV(P) p.83; MKV(V) p.28.

-This is an examination of the first of the four conditions referred to in l.2,

namdy, a primacy condition (hetu-pr11tyaya) . After examining the nature of a
condition (pratyaya) and the effect (artha) or the causally ariscn phenomena
(prafityasamutpanna dharma), in this and the oext three verses Nig:ujun is
dicecting his attention to the four specific condicions formulated by the
Abhidharmikas.
In defining the primary conditioo, the A.bhidharma tefers to five of the six
eauses (hetu)(AK 2.61) . They arc (1) a "co-operative cause" (sahabhii-hetu) or
factors that work together in producing aoother; (2) the "complementary
cause" (sabhliga-hem) which is a cause helping other causes of its kind; (3) the
"associated cause" (saf!Jpray11ktaka-hetu); (4) the "all pervading cause"
(sarvatrage-hetu) and (5) the "fruitioning cause" (11i/iiiktrhe1u). However, the

EXAMINATION OF CONDITIONS

llt

intetpreters of the Abhidharma defined a primary condition (betu-pratyaya) as

a producing (nirvartaka) or a root cause (mula(hena hetu upakaraka(fhena pa
cayo, Vi.sm p.~33). Nagarjuna, io the hope of clarifying the implications of this
definition, raises the question as to whether this condition is supposed to give
rise to an existent (sat) phenomeoa or a non-existent (11Jat) phenomena or
some-thing that is both eicistent and non-existent (sari asat). The eady
Abhidharmikas do not seem to have involved themselves in such speculations
regarding the absolute idencity or absolute difference betweeo a ~ondicion and
itS cffect; hence this criticism ofNigarjuna applies only to the later interpreters
of the Abhidharma cqoception of a primary condition.

8.

Analambana evaya~ san dharma upadiyate,

athanillambane dharme kuta alambanaf?Z puna/?.

A thing that exists is indicated as being without objective support. When

a thing is without objective support, for whac purpose is an objective
support?
MKV(P) P,.84; MKV(V) p.29.

The Abhidharma defines the objectivcly supporting condition as all

"phenomena" (sarvadhamiiil/) (AK 2.61). It was intended to explain the occurren_ce of all ideas of experience. Buddhism rec.ognized exterti)al objeccs as condit~ons for the arising ?f .ideas ~f ex~erie~~e (in co~trast to the ideas ~f.imagina
uon). For exampJe. tt Js admitted\tbat depending upon eye and vJS1ble form
arises visual consciousness" (cakkhufJ ca paficca riipe ca uppa1/ati zk
khu11inn"iif!af?Z, M 1.111-112), aod these serve as conditio.ns for the ideas of
tierceprual experience.
During the Abhidharma period these various conditions as wdl as the ideas
of e.xperience carne. to be categorized into miod (citta), mental concomitants
(caitta; cet11Jtka) , and matedal form (nipa). eventhough they were ootsharply
discinguished into substantially differeflt entities ~ miod and matter. Yet
those who defined these categories u1timatley ended up recognizing mental
substances and materia! substatices, the mental substances have the capacity to
perceive theit owo mental concomitancs (caitta). even though these concomitants are conditioned by materially constituted objects. The mental
substances thus became the subject, the .mental concomitants the contents of
perception and the materia form the objective condition. Such speculation not

112

THE PHILOSOPHY OF THE MIDDLE W AY

ony ed

to the belief in a cogit, thinkiogof its own subject-matter (cllitta) carcying with it all the metaphysical implicacions; but aso raised the question as to
the need for an external object (iilambana) as a support for the concomitants.
While the question regarding the cogito is tak.en up by Nagarjuna in Chapter
III, the need for an exrernal objective support is r~sed in the.present veJ:Se.
Thus . it is recognized by Candrakrrti that the question regarding objective
support is raised by Nagarjuna because those whose views he was criticizing admitted a cogito (athaivam aniilmbane dharme sviitmana prasiddhe kim
asyiilambanayogena pariktllpitena, (P) 84.; (V) 29) This is simHar to the refutation of a "materia! object" by the Western phiosopher George Berkeley. If the
object appears to the perceiving mind in its owo form (i.e. in the form of a
mental imptession) (s1111tma11a prasiddhe) 1 there is no necd for an objective
suppoct(iihmb4114). 1f it docs not appear to be the perceiving mind in its own form ,
itwill never be perceived, since the perceiving mind and the materia object are
of completey different natures (p~-ratman?). This is indeed not a rejection of
the notion of an objcctive support (til"1!Jbana-pratyaya) per se, but -an objcct
that 1s conceprualized in a metaphysical way, that is, as an object constituted
of a materia substance distinguished from a mental substance.

9. Anu1pa1111e1u dht1_mze1u nirodho nopapadyate,

niinantar.am ato yukfa'f!l niruddhe pratyayaf ca kal/.
When things arc tlot arisen [from conditions], cessation is not appropriate. When [a thing has] ceascd, what is [i~ that servcs as] a condition? Thereforc, an immediate condition is not propcr.

MKV(P) p .85; MKV(V) p.29.

The immediately contiguous condicion (s11ma114ntara-pratyaya) was first formulated by the Abhidharmikas in order to account for cenain kinds of relations
implied in the Buddha's statcmcnts such as: "In this way, monks there is the
immediate (anantara) waning of defiJements ,"' (S 3. 58). The Abhidharmikas,
therefore, specified the relation as one amoog mind and mcotal concomitants
(ciftacaitlii acaramii uppannii~ saman1111taraf?, AK 2.62). With the acceptancc
ofa theory. .of mments (kfaf!a), the interpretcrs of the Abhidharma were faced
wit~ sevcra questions: Does this relation obtain among cvents of a sirnHar
nature (s11aj1Jtt)? How can the cmergence of dissimilat events be explained?

EXAMINATION Of CONDITIONS

11 3

(Akb pp.65-66). The problems arc not different from thosc chat arc associaccd
with the conccpcion of a series (similar to the problems faced by empiridsts like
David Hume). The question as to how one momencar:y event can give rise to
anothet or how one series could pave the way for a different series was discussed, in relatioo to the'problem of knowledge , especially the knowledge of the
future . Some of the interpreters of the Adhidbarma recognized an unimaginable range of compreheosion on the part of the Buddha (acintyo hi
buddliiin'iif!J b11ddhi11i[aya iii, Akb p.66), which was probably a view ad
vocated by the Sarvastivadins, who admitted the possibility of knowing che ex
istence of everything (saffJam astt) bclonging to the past, the present and the
future . The Sauuintikas, who refused to accept such a position , maintained
chat the Buddha follows 'signs' (naimittdo) and that even in the absence of
direct pcrcepcion of future events (na sik/itkinl he is able to predict them on
the basis of "intention" (icchimatrena, ibid.).
Nagarjuna, realizing the difficulcies inherent in such speculations, raises
questions regarding the very cooception of 'arising' in such a cootext. Neither
the momentary eveots 1 nor the substances chat were posited to account for the
continulty of series of such events, according to Nagarjuna, can be describcd as
''arisen" (111panna). 1f they arc not arisen (anulpanna), their cessation is aJSo in
conceivable. Jf thcy were to cease momentarily, thcy could not scrve as con~di
tions (praJyaya).

10.

Bhivana'f!' n11/st1abhavan'iif(l na salta vidyate yataf!,

safidam asmin bhavaffty etan naivopapadyate.

Since the cxisteoce of existcnts devoid of sclf-naturc is not cvident, the

statcmeot: "When that cxists, this comes to be," will not be appropriate.
MKV(P) p.86; MKV(V) p.30.

The notion of dominance was understood in a very generaJ and broad way in
the carly discourscs. For example, oneself (alta), the world (loka). and
rightcousness (dhamma) were considcrcd dominant conditioos_('iirlhip111eyy11)
in the mattcr of refining one's mora li.fe (A 1.147-150). The .Abhidhatmikas
defined the dominant coodition as an active causc (uraf!a-hetu) and this dif.
fcred from the other fivc causes (sec note oo I. 7) because of the dominant cffcct
of this cause. Dominance, of cowsc, can be of diffcrent sort$. For example, a

ll4

THB PHILOSOPl-IY OF THE MJDDlll WAY

seed may be a dominant cause 0f the sprout, just as much as water is. The
Abhidhat.m.ikas cese(Ved the nocion of dominant condition (or active cause) to
explain the latter kind of relationship, namely, t bat becween water and the
sprout. Hence it was defrned as sometbiog "other than fr.self' (svato 'nye, AK
2..50). However, this distinction between self and otber ca.me to be "reified" to
such an extreme that the later interpreters of the Abhidbarma were left with
the notions of self-natute (.wabliiiva) and other-oature (parabha11a). lt became
almost iropossible to speak of 'chis' and 'that' without getting invoved in a
discussion of self-11ature and other-nanue. Therefo.re, when the Buddha's formulation of the genera causa] principle as: "When that tiists, this comes to
be," (asmin .rafidaf'{I bhavah) carne up for discussion, the. metaphysicians were
quick to interpret .rat (occu~ring in the locative absolute construction a.rmin sa)
as substantial existence of the two eocities rcferred to by asmin ( that) and idam
(this).
I
Here too Candrakrrti, in spite of his Jeanings towatds "no views," presents
dependent arisig as a position f.com which to criticize self-nal:!lre or selfexistence (svabhiivl'J). He argues: "Because existeilts are dependently arisen a,nd,
therefore, without sel{-nature, how can tbat statemcnt: 'When that.(exists),' be
intended as an active cause?"' (Bh'ii11an7i7t1- prafitya.ramutpannat1iiit
wabh'iiviibhave kutas tad yad a.rminn iti liiiratJalvena vyapadiyate , (P) 87; (V)
30) . Nagarjuaa's cdcicism, therefore, eaves the Buddha's general formula of
causation untouched, for it was not the Buddba's intentioo to reify either "this"
or ''that."

11.

Na ca vyasta-samaste1u pratyaye111 a.rti tat phalat?J,

pratyayebhyah. kathaf'{I tac ca bhaven na pratjaye1u yat.

The effect does not exist in the conditions that are separated or combined. Therefore, how can chat which is not found in the conditions
come to be ftom the conditions?
MKV(P) p .87; MKV(V) p.30.

Once again, the qucstion raised in the second line: "How can that whicb is
not found in the conditions come to be f.com che conditions?'' is grounded on
the asscrtion or premise menrioned in the first line. What is denied in the fust
line is chat the effect is found in the causa! condicions taken either sepatately or
togethet. It does not mean a denial of the statement that the effect comes to be

EXAMINATION OF COND!TIONS

11 5

tlepend;ng upon a coodition or a group of conditions, In other words, it is a rejection of the essentiaist method of looking for the effect even before it comes
to be. A truly empiricist approach would not be concerned with such ao enterpnze .

12.

Athasad api tat tebhyal? pratyayebhyal? pra1lartate1

apratyayebhyo 'Pi kasmtin nabhipravarlate phala1{l.

If that effect, being non-existent [in the conditions] were to proceed from
the conditions, why does it not proceed from non-conditions?
MKV(P) pp.87-88: MKV(V) p.30.

So far, most of the arguments wece directed against sdf-causation and the
substantial ~xistepce of the effect in the condition that gives rise to it. The t>resent verse is a direct refutation of the view chat the effect is different from the
coodicion, that is, the basie premise of the theory of external causation. As implied by Niigatjuoa, in sucha cootext, the term "condition" l'bses its roeaoing,
for jf the cause and effect were sharply distinguished, one could maiotain that
anything can come out of anyching. lt is in, fact, the contrary of the substantialist view: ''Nothing comes out of nothing."

13.

Phalaf!I ca pratyayamayaf!I pralyayaf ctisvayammayal/,

phalam asvamayebhyo yat tal pratyayamayatrt kathaf!J.

The effect is made of condition.s, but the conditions are themselves not
se1f-made. How can that effect made of conditions [arise] frotn what is
not self-made?
MKV(P) p.88: MKV(V) p.30.

The first line of this verse contains two assertions. First of these is chat "the effect is made of causal conditions" (ph4l41?1 pratyayamayatrt), wbich is already
negated .a t 1.4 (kriyti na pratyayavafi) where the term kriyti is equivalent in

116

T HE PHILOSOPHY OF THE MIDDLE WAY

meaoing to the term pha/41'(1 in the present context. So does the term
pratyayavafi convey the same roeaning as pratyayamayaf!I. lt is a statement
asserting the identity betweeo the condition and the effect. However, the next
statement impies difference between the condicions chat give rise to the.effect
and those ocher conditions that produce the conditions themselves, fot the
former arc not "self-made" (asvayamm1Jfiih). This, thcrefore, is a theory that
attemprs to accommodate both idencity and difference in the causa} process,
and N:igatjuna sees this as a self-contradiction. lt is indeed a rcfutation of the
third theory of causation negated at 1.1, namely, causation through both self
and other (dvabhya11J).

14.

Tasman na pratyayamayaf!I napratyayamaya'l!I phala'f!',

.ram.vidyate phaliibhavat pratyayapratyayal{ kutaf?.

An effect made eidier

of corulitioJlS or of ooo-conditions .is, therefore,

not evident. Because of the absence of the effect, whete could conditions
01 oon-conditions be evident?
MKV(P) p.89; MKV(V) p,31.

Thus, the ideotity of condition and effect (pratyayamaya?(l phala'f!'), as implied

in the identity theory of causatlon, as well as the difference between condition
and effect (apratyayamaya?(l phalaf!I), as envisaged in the non-idencity tbeocy
of causation, arc both not evidenl: (na .rllf(Jvtilyate). The second statemenc is, iodeed, the fina condusion of Nigarjuoa in this immensely significant chaprer.
A supetficiaJ .interpretation of this statement is bound to leave the impression,
generally popular among the intcrpreters of Nagarjuna, that he rejccted any
form of causation, induding the arising of an effect depending upon a cause or
coodition or a group of sutji causes or conditions (prafftyasamutpada) . Hence,
Nagarjuna is perceived as a trancendenralist who recognized an "absolute"
beyond all linguistic expression. (Following the prevalent inteipreration, the
presenr author bimself has taken that position, sec Buddhist Philosophy ,
pp.129-141). A morc careful cont?tual analysis would reveal that the effect
"fphala) Naprjuna was referring to in this verse, as well as in the entire chapter,
is one that is identical with the cause or different from it. Ir is only an effecr

ExAMINATION OF C ONDITIONS

117

understood in sucha mannet, as clearly iodicated in the present statcmcnt, that

he was categorically dcnying. If no such cffect is secn, why spcak of a conclicion
(pratyaya) that is idcntical with an cffect, or a non-condition (a-pratyaya) chat is
diffcrent from the cffcct?

CHAPTER

Two
Examination of The Moved and the Not-moved
(Gatagata-parz k.ra)

l.

Gata"!I- na gamyate tavad agataf!l naiva gamya1e,

gatagatavinirmuktaf!l ga,nyamana'f!I- na gamjate.

What has been movcd, in the first instance, is not being moved. What
has not becn moved is also not being moved. Separated ftom what has
been moved and has not been moved, present moving is not known.
MKV(P) p.92; MKV(V) p.33.

The positive stacement: "What has moved is being woved" (gata'f(I. gamyate)
does, indeed, carry the i.mplkation of a permanent substantial entity, an entity
with which movement was associated in the past and whlch is also presently
moving. N:igarjuna's negative statement is, therefore, a denial of such an entity. Yet, this denial may be interpreted as involving the opposite view, naniely ,
that an entity that was previously not associated with movement is at present
movig , that is, the entity that is presendy moving is completely different from
the previous entity.

s
(man moves)

VP

NP
i. (moving) man

moves

jj, (non-moving)man

moves

iii. (movlng and noo-moving) man

moves
118'

Ex.\M!NATION OF TuE M OVED AND 1"HE N OT-MOVED

119

If the two positive assertions: (i) "What has moved is being moved ' and (ii)
"What has not moved is being m011ed," ate accepted, then we have a present
moving which is with and without prior movemenc. This is io a way selfcontradictory. l t is like saying thac "a first cause is both caused and uncaused,"
or chat "a m over is both moved and urunoved." Nagatjuna would appear as a
traoscendentalisc if he had assumed the "present moving" (gamyamiina) which
has the characteriscic of both ''moved and not-moved." The substantialist
petspective was thus contribucory co three views, all of which were not acceptable.

2.

Ce!fii yatra gatis tatra gamyamiine ca sa yataf?,

na gate nagate ce1ra gamya111ane gati.s tata.

Where there is movement, there is morion. For which reason movement

is in the present moving, and not either in the moved or in the not movcd, for that reason motion is available in the prescnt moving.
MKV(P) p .93: iWKV(V) p.33 .

If a Cartesiao perspeccive wece to geoerate metaphysical views such as those

prcseatcd by the Sarvascivadins, one way of etiminating such mecaphysics is by
adopting the "Humean" perspective that emphasizes the "immediate present,''
without any reference to the past or the future. However, such an unrelated or
independent static present may once again lead to a subscancialist reductionism. The only way to gec rid of such ''essencialist" perspectivcs, both of the
racionalists and of some of the empiricists, is by adopc.ing a m oce
"phenomenologicaJ" explanation where, inscead of a "present," one speaks of
"presencing." If so, presenc movement (gamyamiina) could simply mean "motion" (gati, cejfa) which is not found either in the past or the future. Having rejected the substaotialisr implicatioos of the Sarvascivada in the previous verse,
Nigarjuna is here specuJating on the mcaning of che alteroative views of the
Saucrantikas, who wrcscled with the problems of atomie discreceness as welJ as
the experienced continuity. only to reject them in the verse chat follows.

3.

Gamyamana.sya gamanaf!Z kathaf!Z niimopapatsyate,

ga"Jyamiinatn hy agamana??J yadii naivopapadyate.

THE PHILOSOPHY OF THE

120

MIODLE WAY

How appcopriate would be the movement of the prescnt moving? For.,

the non-movement in the present moving is certainy not appropriatc.
MKV(P) p.94; MKV(v,l p.34.

As pointed out earlier, the statement "Man moves," can have two possible
metaphysical "deep structures." The same could be said f tbe assertion, "The

present moving moves."

s
present moving moves ( = maa moves)

VP

NP
(i) aoa-movlng ( = aoa-moviog man)

moves (involvcs aon-moveroent,

(agf1111Pna).

(ii) moving ( !: moving maa)

moves (involves two movements,

(dvigamnna).

Aftcr obsetving these two mecaphysical implications of the statcment:

"moveme.nt of the prcsnt moving," (gamyamiinflJya gamanaf(') , Nagarjuna
seems to take up the first altcrnative for e:Xa.m ination and maintains that a present movemc.nt, conceived in such a manner, is indeed a non-movement
(gamyamanatrt

hy agamanaf(') .

The Tibetan versions (texc as well as commentaries) seem to preseJVe this

original reading [sec MKV(P) p . 94, note 2] as do all the Chinese uanslacions
(sec Inada, pp.44-45), induding Kumarajiva's. As such, the available reading
in the S~nskrit version as gamyamiine dvi-gamanatrt could prove to be a scribal
error. This assumptioo is further strengthened by the fact that II.44 that follows
provides a cricicism of thefir1J metaphysical assumption, while ll.} cake che Ie
cond metaphysical assumpcion for cricicism.

4. Gamyamana.rya gamanaf?J yflJya tasya p rasa/yale,

rter galer gamyamanaf?Z gamyamiinaf'?l hi gam.yale.

ExAMINATION OF THE MOVED AND 11i.E N OT-MOVED

121

Fot hlm who assens the movement of the present moving; it follows that
there coud be present moving without motion. [However,] the p.resent

moving, indeed. tnea.ns being mo'Ved fi.e the present moving,

takes place] .
MKV(P) p .95;

ndeed ,

MKV(V) p .43,

In asserting the statement: ''The movement of the prcsent moving," one is

compelled to assume chat present moving is independent of mocion 'lUld that
the latter is something attributed ('iidheyahutaf!J) to the former . In chat case,
the present moving should be without motion. Hwever , Nigarjuna seems to
be ready to assume that "the present moving indecd takes place"
(gamyamlinaf!J hi ga.1nyale) without aUowing for such metaphysical implicatons.

5.

Gamyamlin11Sya gamane prasakta~ gamana-d11aya1!J,

yena tad gamyamiinll'f!J ca yac c'(itra ga'fft4naf!J puna!/.

A two-foki movetnent is implied in the movement of the J>resent moving: that by which thcre comes to be present moving and, again, the
movement itself.
MKV(P) p.95; MKV(V,l p.34.

This verse takes up che second metaphysical implicacion mentioned in the

analysis of ll 3, narody, "the movement of the present moving" involves two
forms of roovemenc (d11i-gamana1!J). The first is che movement through
dcsignation (11yapadea) and the. ocher is the movement in itself
(adhiktlrt1f!abhula). This seems to be the ristinction between the phenomenal
designation and the "thing-in-itself." Further metapbysical implications of this
understanding arc prescnted in che ncxt verse.

6.

D'llaf! gant?irau prasajyete pr11Sakte gamana-d11aye,

gantaraf!J hi tira.rkrtya gamanaf!J nopapadyate.

122

THE PHJLOSOPHY OF THE MIODLE WAY

lf two movements are allowed, it would follow that there would be two
movers. For, separated from a mover, a movement is not appropriate.
(MVP(P) p.96; MKV(V) p.35.

The assumption of two movemencs, as explained in the previous verse, will imply two movers (dvau gantarau). Hece, then, is a basie assertion of Nagarjuna
with regard to language, namely, that one can not speak of, say, movement, in
a vacuum. but ony in relarion to soroerhing tbat is moving (gantr) . This is the
noo-substantialist approach in Buddhism whlch refuses to recognize a sharp
dichotomy between substance (wabhava) and aruibute (lakfaf!a), a dichotomy
that has become an inalieoable part of the essentialist traditions in Indian
philosophy thar are in pursuit of a truy real or uJtimate entity.

7. Ganliir'"!J cet tiraskrtya gamanaf!J nopapadyate,

gamane 'sati ganta 'tha kuta eva bhavifyati.

If it is thought that a movement separ:ated from a movet is not apptopriate, then, wben no movementexists, how could there be a mover?
MKV(P) p.9.7; MKV(V) p.35.

This .represents a simple refutation of the essc.> 1tialist view involviog substance
and attribute. If there wece to be no movementseparated frorn the mover, then
in the absence of the movemeot there couJd be no mover. The emphasis here is
on the term tiraskrtya (separaced). When there is no such separation and whe.re
the movement is dependent upon (ptafitya) the mover and vice versa, Nagarjuna sees no difficulty.

8,

Ganfii na gacchati tiJ.vad aganla naiva gacchati;

4nyo gantur agantt ca kas trffyo hi gacchati.

As much as a mover does not move, a non-mover to does not move.

Other than a mover and a non-mover, what third party moves?
MKV(P) p .97; MKV(V) p. 35 .

ExAMINATION OF THE MOVED AND THE N OT-MOVJID

123

Nigarjuna's analysis now moves from the present movement to the present
mover. The possible metaphysicaJ implications he perceived in t he statement
"present rooving moves," are not l'ery different from those that may be involved in the assertion that a "present mover rooves" (gantii gacchatt). Yet, the
question raised is in a slightly differeot context from that embodied in Il. I
which refers to the past, that is, the moved or the one who has moved (gata),
whereas the preseot verse relates to a present mover (ganfii). It shows that
metaphysicaJ intetpretations can arise not only regarding che past and the
futuie but also in relation to the present.

9.

Ganili fjjv(ld gacchafiti katham evopapatsyate,

gamanena vina gant~ yada naivopapadyate.

lndeed, how appropriate will be the view that a mover moves? For, a
mover without movement is certainly not appropriate.
111KV(P) p.98; MKV(V) p.36.

This again is a positive a5se.ccion of Nigarjuna, who would be willing to speak

of a movement i.o a mover, without having to assume chat there are two entities
iovolved here, namely, a move.r and a tnovement. For him, the mover and
movement are dependent or contingent; one cannot speak of the one without
implying che other. The statement, "A m over without movement is cettainly
not appropriate," is the ultimate refutation of.a substantialist assumpcion rhat
there can be a real entity about whom or which certain attributes can be
predicared, the entity and
attributes being completely different.

the

10.

Pak10 gantli gt1ecbafiJi yarya tasya prasa1yate,

gamanena 11ina gania gantur ga1nafl.IJm icchatai{.

For bim who entertains the view; "A mover moves," and who Jooks for
the movement of a mover, it follows that tbere is a mover witbout movement,
(MKV(P) p .98; MKV(V) p.36.

124

TuB PHILOSOPHY OF THB MIDDLE

WA Y

A mover witbout movement is an entity without an attributc. For the esscntialist uadition, an enci ty is an cntity "in itself' without any atttibutcS, of which at
ttibutes arc predicated. An "cn~ity-in- itself'' thus bccomcs indefinablc and indescribable. Nigirjuna has no interest in such spcculations. No prcdication is
meaningful without the prcdicatcd. These arc mutually dependent (pr11fitya)
and not independent (a-pralitya).

11.

Gamane dve prasajyete gantfi yady uta gacchati,

ganteti cocyate yena ganta san yac ca gacchati.

If a movcr wcre to movc, then it would follow that thcre will be two
movcments; one in virtuc of which he is spoken of as a mover, and the
other in terms of which an cxiscing mover is said to move.
MKV(P) p.99; MKV(V) p.36.

This is the conversc of ll .5. The statement: "A movet m ovcs" would imply two
movcments: the first is a movc.r by designation (vyapade1a) and the second, the
really cxistent (sad) mover or " the movcr-in-itself." Whethcr Nigirjuna is
prepared to maintain that all designations arc so wild in thcir impications or
whethcr such is the case with ony some of them needs to be carefully examined, as in the case of concepts of causal depcnrl : nce.

12.

Gate tiiirahhyate gantu~ ganlllf!!. n'iirabhyale 'gate,

n'iirabhyate gamyamiine gantum firabhyate kuha.

Movemcnt is not bcgun in the moved, nor is it bcgun in the not moved.
Neithcr is it initiated in the prescnt moving. Whcreio is thcn movcmcnt
initiated?
MKV(P) p .100; MKV(V) p.36.

Nagarjuna now movcs on to the q uestioo regarding the origin of movemeot, a

question thac has Jeft a uail of metaphysicaJ speculacions from the vcry early

EXAM!NAUON OF THE MOVED AND 1HE NOT-MOVED

125

period of philosophical thinking. Mctaphysical speculations regarcling time,

leading to a theory of existence aoaJysed in terms of discrete momentary events
into the past, piesent and future, couplcd with the problem of cxplaining the
origin of each moment, a seoario cfeated by tb.e..Abhidharma intcrpreters, pro'vided Nagarjuna with the opportunity to utilize a letbod comparable to
Zeno's paradoxes in order to expose the meaninglessoess of such metaphysics.

13.

Na purvaf!1- gamanarambharl gamya11iiinart1 na va gataf!J.,

yatr'iirabhyeta gamanaf(I agate gamanaf!J. kuta4.

Por

to the commencement of movement. thete is neithet the present

moving or the moved fto.m which movement is i.n.itiated. How could
there be a movement in the not moved?
MKV(P) p.100; MKV(V) p.37.

Agate gamanll?!J /lula; ("How could there be movement in the not moved?;')

seems to be the tefutation of the idea of an "unmoved mover'' at a microcosmic

or phcnomcoal level. The interpreters of the Abhidharma were probably aware
of the Buddha's reluctance to cliscuss the absolute origin of the universc. Yet
their way of handling the Abhidharma aoalysis, cspecially thcir understandiog
ofchange as momeotary deruuctioo (k!111111-bhaga). left th~ sometimes with
four discrete rnoments (i.e. origio, stasis, decay, and destruction recognized by
the Sarvasti.vadins), sometirnes with three (Le., origin, stasis, and destruction,
as in the case of the TheraYadios) and sometimes with two (i.e., origin and
destruction, as it was the case with the Sautraantikas). In all these instances ,
cach preceding moment bad to account for the succeeding moment that is dif
ferent.

14.

Gata1!J. ktf!J. gamyamanart1 kim agata?!J kif!J. vikaplyafe.

adryamiina

arambhe gamanasyaiva sarvatha.

When the commeru:ement of movement is not being perceived jn any

126

THE P HILOSOPHY OF THE MIDDLE

WY

way, what is jt that is discriminated as the moved, the present moving, or

the not moved?
MKV(P) p . 101 : MKV(V) p. 37.

Unless the metaphysics referred to above is kept in mind , it would be easy to

assume that this verse cepresents a refutation of any form of discrimination of
events as past, preseot, or future . On the contrary, what is being emphasized
herc is that the commencement of movement, as explained in the previous
verses, is not perceived at aU anywhere (sarvathi). As such, a question is raised
regarding the validity of discriminacions or thoughts regarding the past, prefuture, which are based upon that particular conception of the comsent,
mencement of movement (gamtlnmya 'iirambha). Nagarjuna is not claiming
that there is only one way in which commencement of movement can be explained ; he is merely refuciog the metaphysical explaoation of movement and
1ts commencemeot.

and

15.

'

Gantti na titfhati lavad aganfii naiva ti.f fhati,

any gantur agantu ca kas trfiyo 'tha tinhati.

A"i much as a mover is not stationary, so is a non-mover not stationary.

Othet than a mover and a non-mover, what third party is s.tionary?
MKV(P) p.101 ; MKV(V) p.37.

To say that a mover is stationary (gat1.ta ti[thatt) is self-contradittory. To maintain thac a non-mover is stationary (agantii ttj{hatt) is tautological. Looking for
something or someooe (It)(rbhiiva, p11dg11/a) to whkh/ whom the characreristics
of motion. and stasis can be attributcd , one merely ends up conceiving of a
"hare's horn" (aa- vt!a11a) or "crow's teeth" (kaka-danta). Such is the essentialist enterprise. This certainly does not mean the rejeccion of the empirical notioo of telativity or depeodeoce of motion on stasis and vic,e pena, as in the case
of short and long. lt is indeed a simple refucation of the view chat there are independcnr enticies to which the characteristics of motio.n and stasis can be attributed .

EXMJNATION Ol'

16.

THE MOVED AND THE, NOT-MOVED

127

Ganfii tavat t#fhafiti katham evopapatsyate,

ga1nanena vinii gania yadii nai11opapadyate.

How appropriate would it be [to say]: "A mover. at the moment, is

statonary"? For, amover without movement is not appropriate.
MKV(P) p .102; MKV(V) p .38.

The argument in the previous verse is made very elear by the present. Empirically, a "mover" without motion is inconceivable, just '1S a pure entity
(s11abli1i11a) without function or characteristics is empirically meaningless. As
such, the statement: "A move(, at the moment, is stationary ," can be made
assuming that the entiry that previousy possessed the characteristic of motion
(gamana) has now abandoned it in order to assume a different characteristic,
narody, stasis. The Sarvastivada theory ofpiiipti and apr1ipti was formulated to
explain such "possessioo'' and "non-possession" after they assumed the
metaphysical notion of a substance or "pure being" (s11qbh'iivt1) (see Poussin ,
AK ii.36).

17.

Na ti!thati gamyamtinan na gafiin nagafiid api:

gamanaf(Z saf(Zpravrttil ca ni11r:1til ca gatel{ sa'!'Tiii.

One does not come to be stationary because one is either moving, or has
moved, or has not moved. Movement, commencement and cessacion {of
movement) arc all comparable to morion.
MKV(P) p .102-103: MKV(V} p.38.

The substance/ attribute distinction openly endocsed ~y the metapbysicians

cannot account for "stasis" in terms of motion, whether rhar morion relates io
the past, present or future . According to their aoalysis, stasis is disti.nct from
morion and therefore is indcpendenr. So arc conceprs of commencement and
cessauon.

128

18.

Tl-E PHILOSOPHY OF TH E MIDDLE WAY

Yad eva ga1nanart1 ganta sa eveti na yujyate,

anya eva punar gantti gater iti na yujate.

The view that movement is identica with the mover is not propcr. The
view that the mover is different from morion is also not proper.
MKV(P) p .104; MKV(v) p .39.

Haviog distinguished substaoce and attribute, the metaphysicians attempt to

sove the resutant'philosophicaJ issues either by assuming identity (sa evr1) on
the basis of an eterna substance (s11abhli11a) , theteby rendering the attribute
(lak!a'fa) an ephimera.1 or impermanent come-and go entity, as the Sarvastivadins did, or by emphasizing difference (anya eva), thereby denying the
substan.ce and accepting fleeting and momentary flashes of aruibutes without
. any real connections, as the Sautrantikas did . For Nagarjuna, both are inappropriate views. The two verse tbat follow provide speci.fic reasons for the rejection of these two views.

19,

Yad eva gamanaf!J gan tli sa eva hi bhavedyadi,

eliibhlival; prasajyeta ka.rtul; karmat1a eva ca.

If movement were to.be identical with the mover, it wouJd follow that
there is identity of agent and action.
MKV(P} p. 104; MKV(V) p . 39.

ldentity (ekTbhava) with regard to agent and action is here presented .as a
necessary implication of considering the mover and mocion to be the same. The
non-absoJutism in Naga:.rjuna's way of thinki~g woud leave the agent meaniagless independent of action and vice versa. Nagarjuna will have no difficulty
in speaking of either an agent or an action in an analycical way without reaching
che enremist position of recognizing distinct encities. For him, analysis (vigraha)
was meaningful and practical so long as the limits of such anaysis are observed.

EXAMJNA110N OF THE MOYEO AND THE NOTMOVED

20.

129

Anya eva punar ganfii galer yadi vikalpyate,

gamanaT?Z syad rter gantur gania syiid gamaniid rte.

If the discrimination is made th.at the mover is clifferent from morion,

chen there would be mo"Vement withouc -;;- mover, and mover without
movement,
MKV(P) p. 105 ; MKV(V) p.39.

This verse specifically lays down the limi t~ to which Niigarjuna was prepared to
go with his analytical method. That is, the analysis should not be carried our to
sucb an extent that Jeaves "mot ion" without a "mover" or a mover without
"morion." Empirical explanation does not allow for such "pure enticies" com
pletely independent of each other.

21.

Ekzbhavena 11ii siddhir n'iiniibliiivena v'ii yayol;,

na vidyate tayol; siddhil; kathan nu kha/11 vidyate.

Whose establishment is not evident either through identity or throqgh

difference, how is their establishment evident at all?
MKV(P) p.105: Hi.KV(V) p.39.

Herc again, Nagarjuoa is examini.og the concepts of substance and attribute.

For him, t:hese are not established either through identity or t:hrough difference. The question theo is: "How is thei.r establishment evident at all?" The
rejection of the substance/attribute distinction as admitced by the Sarvastivadins and the Sautranti.kas does oot mean the rejectioo o( all conceptualizatioos or d~scriminations (vikalpa). lt is this particular form of conceptualizatioo that is being questioned , not any form of cooceptualization.

22 ,

Gatyi yayocyate_ ganta gati'T!l tiif!l sa na gacchalt~

yanniin n4 gatipurvo / ti kaicit kif?lCidd hi gacchati.

130

THE P HJLOSOPHY OF TtE

MJDOt.E

wAy

Whatever motion in terms of wh.ich a mover is spoken of, he does not

move by that motion. Because he does not exist prior to motion, who or
what is it that moves?
MKV(P) pp. 105 106; MKV(V) p.40.

It is possible for one to speak of a mover dcpending upon motion. However,

when that mover is distinguished from. motion. then we re left with a pure entity or person (svabhava, pudgala) to which or whom the mot.ion is anributed
and, in that case, the encity or person should precedemovement. Such an entity or person is not evident in experience. Hence the statement: "He does not
move by that morion."

23.

Gatya yayocyale ganta tato 'nyatlJ sa .na gaccha11;

gati d11e nopapar/yete yasmiit eke prag~chali.

Whatever m~tion in terms of which a mover is spoken of, he does not

carry out a motion that is completdy differentfrom ir. A two-fold motion
is not appropriate, since it is only one person that moves.
MKV(P) p . 106; MKV(V) pAO.

The disticcioo becween the "mover" and "mocion" also does not mean ~hat the
"mover'' carries out a _morion chat is different from himself. If sucha distinction
is recognized, then, as explained eadier, there would be two movemeots, the
movement as a result of which one comes to be called "mover'' and the movemeot itself. No such dual motion is found. nor are there two movers correspondiog to the twofod motion . The /act is that it is only one person that moves.

24 . Sadbhuto gam(lnaf?J gantii 1nprakiira~ na gacch4t1;

n7isadbhulo 'pi gamanatlJ 1nprakart11(J sa gacchati,
25 .

gamanaf'(J sadasadbhutatlJ tnprakaraf!t na gacchati,

tasmarj gali ca gania ca g11nt11vyrJ??t ca 1111 vidyale..

EXAMLNA T!ON OP THE M OVED AND THE NOT-MOVEO

131

An existeot movet docs not carry out the movemeot in any of the thtee
ways. Neither does a non-exutent mover carry out the movement in any
of the thrce ways. Nor does a person carry out a movetnent, both cxistenr
and non-cxistcnt, in any of the three ways. Thcrefore, neithcr motion,
nor the mover, nor the space to be moved is evident.
MKV(P) p.107; ltf.KV(V) p.40.

The conclusion is very specificaJly stated in these two vecses. lt is not any kind of
movet or movement that is cejcctcd as being impossible. lt is the really or
substantially or indepcndencly existent (1ad) mover or movement that is rejected. Th.is is a criticism of etcrnalisrn (iaf11ala-11ada). The opposite view,
namely, a non-rea!, non-substantiaJ and non-independent existence (a1ad) was
the kind of irnpermanence advocated in annihilationism (uccheda-vada)
which, in the Buddhist context, ~ commensu.rate with momentary desttuction
(/qaf!a-bhanga). rather than the impermanence (anitya) advocated by the Buddha on the basis of "dependent arising" (prafftyasamulpada). The coml:?ination of the two metaphysicaJ views of cxistencc and non-existence docs not lead
co a happy synthesis. Change and impermanence undersrood in this
metaphysical way do not con tribute toward a reasonabe and cmpicical explanation of the motion, the movet, or even the space moved.

CHAPTER

THREE
Examination of the Faculty of Eye

(Cakfur-indriya-parz k!a)

1.

Dar(ln(lf!Z ravaf!llf!Z

ghrtif!an.z rasantJ'(J sparanaf!Z mana!;,

indriyar,i jaefetefiif!Z drllffavyadini gocaral;.

Seeing, heang, smelling, casting, touching, and mind ace the six
hculties. Their spbeces consist of the object of seeing, etc.
MKV(P) p.113; M.KV(V,) p .43.

Although the traditional Adhidharma dassification lists the "aggregates"

(skandha) ; "spheres" (ayatana), and "elements" (dliiilu) in that order, fot
Nagarjuna, the epistemology, the faculties (induded under spheres) were
more important, primarily because of the current controversies surrounding the
concept of existence (dharma) . Furthermore, even in the discussion of faculties,
Nagarjuna was not so much interested in the faculties per Ie, for therc was not
much controvcirsy regarding the eye, etc. No school doubted the existence of
these faculties . The controversies were centered more on the function of the
faculties, thar is, with regard to seeing, hearing, smelling, tasing, touch, and
thinking. For this reason, after providing a ci tle for the chapter as "Examination
of the Faculties" (Jndriya-paiikfii), Nagarjuna immediately moves on to an
analysis of the more complicated issues relating to tbeir funcrioos. Hence the
refereace to seeing, heariog, etc., especially as means of identificatioo of events
(see, e.g., P. F. Sttawson, lndividualI, New York: Doubkday , 1963, pp .35
ff.) that is so imporcant for an empiricist like Nigarjuna. Even among these
various faculties and their functioos , the most itnportant.epistemologkal issues
were connecced with sed~g. Hence Nagjrjuna's interest in the problem of seeing or visual perceptioo. Note that the term mana/; is used here to refer to the
function, even though manana woud be more appropriate in the context.
!his may have been done to p(eserve the metre.

2.

S11am atmanaf!Z c4rangf!Z hi tat taz eva na payati,

n4 payati yad iitmiinl1f!Z katha'f(l drak$y11ti tal paran.

EXMINATLON Of THE FACULTY OP EYE

133

Sceing does not perceive itself, its own form. How can that wbich does
not pecceive itself, see otbers?
MKV(P) p.113; MKV(V) p.43 .

Nagarjuna was dearly aware of the major contcoversy raging among the
.a dherents of the various "esseoti:is t'! schools regarding the problem of perception. Io their search for certaincy, these essencialist schools assumed char in any
act of perception the "most dear and distinct" is the perceptioo of "oneself."
(see Brhlkliiraf!yaka Upanifad 1.4.1). "I think, therefore, I am'' (cogito ergo
sum) was the premise with which the essentialist thinkers of pre-Buddhist India
began their exposicion of percep tion. The Buddha was himself aware of the difficulties involved in such an assumption when he advised his disciples not to
follow such speculations (mant/i ast1iitisabba1?1 uparundhe, Sn 916). White the
Buddha was willing to recognize consciousness or "sdf-consciou sness"
(11ifJfiiif!a, 11ijnana) as an imporcant constiruent of the human personality as
well as its experiences, he was not willi.ng co assume a metaphysical substracum
such as the "self' or "I" as beiog the objeot of such awareness. He was clearly
aware thac this Jacter episcemological method was the source of most obsessive
conceprions (mu/af!Z papancasakhaya, ibid.). However, the later Buddhist
rnetaphysicians, innocently unaware of the implicatioos of such a method,
seem to havc been led in that direction, thereby dragging themselves into the
quagmire of svabhava-metaphysics from wbich they could not ea.sily get out.
The resuit was the description of perception in the Vibha1aprabli?ivrtti (p.32):

The substance called the eye U.ofthe nature ofthatwhich sees. ln it

is p.roduced an action of seeing , when its power is awakened on account of the emerge.~ce of the rotality of its causes and condicions.
The eye does not apprehend independently of consciousness (vijfiiina), nor does eye-cons<:iousness know the object unsupported by
the active eye. Eye as well as eye-consciousness 1 with the help of accessories such ~ light, cooperate simultaneously toward bringing
che perception of an object. T he 9 bject, the eye, the eyeconsciousncss, and the light, cooperate simulta neously coward bringing the perception of an object. The object, the cyt\ the eyeconsciousncss, and the light , all manifest thir power, i.e., beco1ne
accivc and flash fortb simuJtaoeously. The object appears, the eye
sees, and the eye-conscoiusoess knows it. This is called the direct
knowk:dgc of an object. [Emphasis mine]
Reading through the present chapter of Na~rjuna , one can hardly miss the
tatget of his criticism if one were to keep io mind the above passage of the

134

rHE PHJl.OSOPHY OF THE MIODLI: WAY

Vibhafiiprabhavrtti. These problems will be discussed in their contextS.

However, bcforc taking up the meraphysical issues involved in the abovc sort of
description, Nagarjuna oeeded to eliminate the very source of such
metaphysics, namely, the cogito. For Nagarjuna, the method by which one arrives ar the cogito not only leads ro the belief in a "sva 1itman" , but also the
sharp dichotomy betweeo "self' (s11a attnan) and "ocher" (par.a atman).
Therefore, having statedpositively thl\t there is no "seeing of oneself' (svam
atmanatri daranatrt), Nagarjuna raises question as co the possibility of "seeing
an other" (paraf!J). Thus, the dichotomy bcrween self and ocher in a moce
metaphysical form is not only ethically unacceptable, but also epistemologically
unfounded.

3. Na pa1y'iipto 'gnidr1fiinto daranasya prasiddhaye,

Ja darfana'1 sa pratyukto gamyam'iinagatagatat''1.
The example of fire is not adequate for the establishment of seeing. That
[fire] together with sccing a.re refuted by [a rcfutation of] the prcscnt
moving, the moved and the not movcd.
MKV(P) p.114; MKV(V) p.43.

While those who accepted the cogito assumed that seeing oneself precedes any
act of seeing, their opponents seem co have used the example of the fire to
maiotain that, Jike fi're which burns cverything but itself, seeing perccivcs
everything else but itself.
The theory of moments (k/af}a) that led to meraphysics in the sphere of
causacion (Chapter I) and cha.age (Chapter Il) did not leave the problem of
perception untouched. lodeed, it was the problem of perception chat was most
affected by a theory of moments, as is evident from the variety of cootradictory
theories of perception presented by the Sarvastivadins, the Saucrancikas and
the Theravadins (sec Kalupahana, Buddhi.It Philosopl{y, pp.97-107).
Being aware of rrus fact , ~garjuna takes the easy route of referring to his
previous refutation of change (Chapter Il) . Hence Candrak:Trci's compositioo of
a verse com para ble to 11.1: What has bee o seen is not being seen; w hat has not
been scen is aso 1:1ot bcing scen. Apart fcom the seen and the not seen , dle prescnt seeing is also not seeo."
The ~ame can be said of that which is puent (dagdhvaf!J), etc. This is what

EXAM1NA110N OPTl-lll FACULTY OF EYE

135

Naga.tjuna means when he says: "That [fi.te] together with seeing are refuted by
[a refutation ofJ the present moving, the moved and the not moved."

4.

NiipaJyamiina~

bhavati yadii kifrzcana daranaf!Z,

darana?{' paiyafity evaf{' katham etat tu yu/yate.

When some form of seeing that is not perceiving does not exist, how percinent is the view that seeing perceives?
MKV(P) p.1 IJ ; MKV(V) p.44.

The essentialist definition of "seeing" as possessing the "nature of seeing"

(darana-wabhava) is tantamount to saying that "sceing always sees." This, indced, is the statement with which the Sarvastiv:ada (specificaly Vaibha~ika)
began its description of perception, as ind.icated by the quotation referred to in
the note to lll.2. In other words, it is not possible to recognize any form of seeing that is "not presenty perceiving" (na apaiyamiinaf!Z), or there cannot be a
"non-seeing perception." lf such be the definition, Nagarjuna raises a question
regatding the appropriateness of the statements, "Seeing per<:eives.''

5.

Payati darana?{' ntTiva naiva pa-yaty adarfanaf{',

vyakhyato daranenaiva dr(lffa cJpy upaga1nyafiif!Z.

Seeing does not perceive, nor does non-seeing perceive. One should admit that a seec is explained by [the aAAysis ot] seeing itself.
MJ...'V(P) pp.115-117 ; MKV(V) p.44.

After explaining the difficulties invoved in the statement, "Seeing perceives,"

here Nagarjuna insists thac it is not possible to assert that inon-seeing
perceives." Just as much as a substantial event like "seeing" cannot be appropriately expained, even so a substantial entity like a "seer" (dra.ff() also can-

136

THE PHlLOSOPHY OFT/-li MIDDLE W 11Y

not be established. The rejeccion of the latter follows from the refutacion of the
former. This point is further elaborated in the foUowiog verse.

6.

Tira1krtya dr(Zf(ii niisty atiraskrtyf' ca darlanaf!I-, .

dra!{avya'f!l. darana'f?l caiva dra;fary asati te kutai{.

A seer does not exist either separated or not separated from seeing. When
a seer does not exist, whencc can there be seeit1g and the opject of seeing?
MKV(P) pp.117 118; MKV(V? p. 45.

The metaphysical vie:ws discussed previously lead to two different conceptions

of a "seer" (dra;ft). namely, (i) a seer associated with seeing (a seeing seer),
which is based on a rheory of identity , and (ii) a seer di.ssoci.ated from seeing (a
non-seei'ng seec), which en1phasizes difference. Haviag denied both, Nagarjuna raises the question: "In the absence of a seer. whe.p.ce can there be seeing
and the objecc of seeing?" This question is raised not by a dogmacie
philosopher, but by a critical epistemology. Ir is simpy askiog rhe question,
'Just because you have come up with an unacceptable definition of a 'seer', are
we going to ignorc the fact that there is seeing and also the objects of such seeing? If there wece t be no seer, how can you account for che percepcion of objccts?" ln ocher words, Nagarjuna , in the way he formulates his question, is
asserting tbat seeing and the objects of seeing are mutualy dependent upon a
seer. This assertion leads Nagarjuna directly to the starement of the Buddha in
.che early discourses.

7.

Prafitya mata-pitarau yathoktal{ putr{l-St11?1bhavai{,

cak;u-rope prafttya11a1tz ukto vijfiiina-saf!Zbhavaf?.

Just as the birth of a son is said to be d~endenr upon the mother and the
father, cven so, the arising of (visual] <:onsciousncss is said to be depn
dent upon eye and materiaJ form.
MKV(P) p . 118; Ml(V(V) pA~ .

EXAMINATION Of THE FACUlTY OF EYE

i37

Being a competent and insightful philosopher, Nagarjuna immediatey

perceives the differeoce betweeo the Buddha's analysis and those of his
"substantialist" protagonists. fi,bandoning the miseading terminology of the
substantialist, Nagarjuna adopts the Buddha's owo terminology to explain the
process of perceptioo: "Depeoding upon the eye and visible form arises visual
consciousness," (Cakkhun ca paficca rupe ca' uppajjati cakkhu-vz"fJifii11a~) (M
1.111).
Even though the three terms used here correspond to the terms used by the
substantialists _:cakkhu stands for darana , rupa for 'dr(Jf(avya. and draftr for vijfiiina- tbe description itself is different. Here the explanation of visual perceptioo does not begin with the assertion of the cogito in order to end with the
perception of the external object, which was one of Nagarjuoa's cricicism of the
substandalist view. Neither is perceptioo defined in a more substantialist way
as in the quotation from the Sarvastivadios which says: "The substance called
the eye is of the narnre of that which sees," (sec nore IlI.2). No metaphysical
jargon is inuoduced herc at all. Without gettiog iovolved in the substancialist
terminology and concepcs, the preseot stacement of perception is based encirely
on the principle of dependence (prafityasamutparla) . Thus, visual consciousness is said to be dependent upon the eye ( = faculty) and visible form
( = object). Such an cxplanation immediately climinates the conception of a
substance (s1,1abhava, at1nan) and repfaces it with a principle of "dependence"
(prafity(l.Jamutpada) . In order to illustrate this process of perception, l'figarjuna utilizes the example of the birth of a son depending upon the rnother and
father
lt .would, therefore, be uofair to think tbat an illustrious . Buddhist
philosopher li.kc Nagarjuna failed to sec the importaat: philosophical differences between the Buddha's explanation of the causaity of perception and
that presented by the mecaphysicians. This verse, therefore, embodies another
of the more appropriace views (ka/pana yatra yojyate, sec XVII .13) that Nagarjuna has been elsewhere attributiog to the Buddhas, the Sravakas, and the
P.ratyeka-buddhas.

8.

Draftavya-daranabhiivad

vi.fniinat11:catu1!ayaf!t,
niisfity upiidiiniit/ini bhavzjyantz' punal? kathaf{t.

If it is che view that the four factots1 beginnirig with consciousness, do not

138

THE PH1LOSOPHY OF THE MIDDLE WAY

exist, bccause of the absence of seeing and the object of seeing, how then
can there be grasping?
MKV(P) p .1 19; MKV(V) pp.45-46.

Candtak1rti's negativist approach creates a problem in regard to the interpreta

cion of this verse. The use of che iti:formula needs to be carcfuly handled if we
arc to avoid a gross misinterpretation of Nagatjuna's thoughts at this point.
Nagarjuna has alrcady rcfuced the mecaphysical vicws pertaining to sceing and
seer. However, at 111.6, he was asserting that one cannot speak of seeing and the
scen withouc a seer. le is possible that somcone may insist: "If thete wece to be
no seeing (darana, cak!u) and tne seen (dra.f{avya, ropa), then the four factors
(namely, feeling (vedan'ii), perception (IattJjflii), disposicions (faf!Jsk'iir'lil?) and
consciousriess ('vijlana'f!I.), which conscitute the psycbic part of the
psychophysical personality and hence the equivalent of dr4!tr] arc also nonexiscenc (naslz)." the ih-formula cooverts th.is to a view or a scatement someone
could express. If so, Nagarjuna's counterqucstion wouJd be: "How then call
there be grasping?" Surprisingly, Candrakrrti interprets Nagatjuna's question
as implying a deniaJ of grasping (na santy upadii.n'iidimty arthal?). However, if
the s"tatem_e nt preceding ih is understood as the view of the opponent, then
Nigarjuna's answer is: "How can you explain grasping?" ln other words,
Nagarjuna seems to be saying: "Grasping existi, for that is what is eliminated
at the moment of enightenment and freedom (anupadii-vimukft). Grasping is
dependent upon consciousness ( vi.fnana) which is, in turn, dependent upon the
eye and visible form . Any other explanation of perception is unacceptable co
me."
Such an explanation is in perfect conformity with the contents of the two
preceding vetses both of which rep.tesent positive statements of Nagarjuna.

9,

Vyakhyalaf{J ravaf!am. ghiii'!af!I rasanaf!I sparanaf{J mana!/,

daranenaiva j'iintyac. chrotr-Jrotavyak.adi ca.

What has been expfained as hearing1 smelling, tasting, touching, and

ro.ind, as well as the bearer, the sound, etc. should be known i_n the same
.
way .as see1ng.
MKV(P) p.120; MKV(V) p.46.

EXAMlNATlON OP THE P,i;cutTY OF EYE

139

Thus, after pointing out the inappropriacness of certain metaphysical views perta.ining to visual perception and having stated the B';Jddha's owo explanation of
perception in terms of "dependent arising" (prafftyasamutpada) , Niga:rjuna
coocludes chat the other five faculties and their objects should be.understood in
the same way as "seeing." This indeed is a very positive explanation of sensory
experience and a faithful representation of the Buddha's teaching as embodied
in the "cliscourscs. "

CHAPTBR

FOUR

Examinacion of Aggregates

(Skandha-parik.fd)

1.

Rupa-karaf!a-nirmuktaf!' na rupam upalabhyate,

fiipeapi na nirmuktaT?Z drfyate fiipa-kara11af!'.

Materia

form, distinct from the cause of matecial form, is not obcaned.

Similady, a cause of materia form, disrinct from materia form, is also
not scen.
M.KV(P) p.123; MKV(V) p.48.

From amoog the five aggregares (skandha), Nagarjuna selects materia form
(riipa) , aod not one of the expicity psychologica aggregates such as feeling,
perception, disposicion, or coosciousncss. The rcasoo is dcar. He has already
examined the proccss of pcrccption and, rherefore, needs to aaalyse the objecc
of perccption, rather thn perceptioo itsef.
In the discourses, materia from (riipa) was anaysed into the four great
elemeots (catfaro mah<1bhUt"ii) and the elcments derived from these four
(catunnaf!' mahabhuliinarri uplidiiya rupaf!J) (M .87). In their attempc to
decermine whac these derived elementS were, the Abhidharmikas scanned all
the 'discourses looking for any element (dharma) that would be predominancly
materia and compiled varying lists. As a result of speculation on these different
lists, there carne
. ro be a distinccion becweeo gross matter (sthula-fiipa) and subde matter (Jiikpnza-fiipa). Yer, it was assumed chat the four great elements coosticuccd the foundatioo of alJ forms of matter.
When the question regarding the nature of the four great eementS was raised, the Buddha maintained chat chese consisc of hardness and rigidity (kakkhalaf(J' khangataf(J) whieh is earth (pafha11t), watery element (apogata) which
is water (apo), the fiery element (tejogalaf!') which is fire (tejo), and the airy
(11ayogala'f?I) which is air (viiyo) (M 1.421 ff.) . This expains the manncr in
which they are experienced. Howevcr, rhe intetpreters of the Abhidharma
began ro de.fine them as " the four elemencs that support self-narure as well as
derived form" (1va/akta11opadiiyarupad/iiira11ad dhatava/J, Akb p .8). An
~

140

EXAMlb/ATJO.N OF

ACGREGf\TES

141

almosc ide,icical definition was gradually being offered for the concepcion of
dharma (svasiimanyafak1a11adhara11ad dharma!?, Sakv p. 12) where svalak1a11a
refers to self-nature or substance and samiinyalak1a11a
to ' 'genera
characteristics" or ''quality," the atter corresponding in some way to the derived elements. Tbese ideas appeared in the Theravada uadition only in the
Abhidhamma commentaries and the later manuals and sub-commeotarics
(e.g., sabhav11-iiimanna-lakkha11a'f?l dhareliti' dhamma, Abhvt p.11). Thus,
with the Sarvastivada speculation, two new categories wece emerging-substance and charcteristics-whic.h utimately invoved a one-way relationship. The substances serve as the cause (kara1111) of characteristics, but not

vice versa.
Even though these specuations are recorded i.o treacises composed long after
Nagarjuna, they did not originate with such treatises but wece prevaenc during
his day r even before, as is evident from a careful reading of the Kan'kii. lt is
the above mentioned substantialist view of materia form (fitpa) thac is criticized in the present verse. For Nigarjuna (as well as for the Buddha), materia
form discinct from the cause of ma.terial form (ropa-kara11a = mahabhUta) is
not acceptable. Similarly, a cause of materia form distinct from mateti;tl form
is also not experienced . Here there is oo clenia! of materia form , but only a rejection of the idea tbat there is an invisib1e ground of 11111:terial form .

2.

Rupa-kara11a-nirmukte fitpe fitpa'f?l prasajyate,

ahetttka'f?l na casty arthaf? kaicid ahetukaf? kvacit.

When materia form is. [.considered to be] disrinct f.rom the cause of
matrial form, it follows that materia form is without a cause. Nowhere
is there any effect (arthal/) without a cause.
MKV(P) p . 123; MKV(V) p.48.

I t is the sharp distinction betwc:en materia form and its assumed cause that was
posing an epistemological problem for Nagarjuna. In fact, evidence from a
ater Theravada sub-commentary. sc:ems to lndicate that a school witb
Saucrantika leanings was trying to eliminate the disti'ne::tion between thcse two
ideas. Referring to the definition of dhamma mentioned.in reacion to fV.1. it
is said : "Therc: is no dha1nma over and above the narure of supponiog," [na fa
dhariyamiina.sabhav'ii afio dham1no n'iima atthi, DhsTp.21; see aJso my aniele, "Schools of Buddhism in Eatly Ceyon," in The Ceylon journ11/ of the

THE PHILOSOPHY OF THE MIDDLE W AY

142

Humanities, Peradeniya 1 (i970):78], a view that resembles the one presented

by George Berkeley during modern times.
Nagarjuna seems to be unwilling to accept such a solution . Considering the
philosophical issues a phiosopher like Berkeley had to face, one can understand Nagarjuna's unwillingoess to subscribe to such a view. Without aowing
the experienced elements (dharma) to hang loose, Nagarjuna was interested in
providing a causa! exp anation. Therefore, following the Buddha's explanation
of "dependent arising", Niigarjuna boldly asserts: "Nowhere is.there any effect
without a cause," (see also XXIV.19, aprafityasamutpanno dhaN!Ull/ kacin na

vid!Jate).

3.

RuPe11a tu vininnuktaf(l yadi sjiid rupa-kiirat/a?(i,

akiiryaka'l(i k'iiraf!41(l sy4t nasti ak-Pryaf(l ca kara11a'f!1.

H there were to be a cause of materia form distinct from materia form ,

there would then bC a cause without an effect. There certainly is no ineffect-ive cause.
MKV(P) p . 124; MKV(V) p.48.

The staternent in the previous verse: "Nowhere is thete an effect without a

cause," (na c'iisti arthaf? kaicid ahetukal/ kvacil) could lead ro the belief in an
invisible ultimate cause (like ~ubstance , or even God) that is ecernal. White th~
characteristics, perceived quaities , etc. could be looked upon as the experienced, yet variabc, effects, their ultimate grouod would be the substance , eternaly existing even when it is not producing the effects. Rcaizing that such a
view could emerge frorn his previous assertion . Niigarjuna immediately pto
ceeds to nip it in the bud when he insists: "There is no in-effect-ive cause."
Thus, IV.2 and 3, in combinacion should provide the interpreters of Nagarjuna wich the cearest evidence that he was upholding the theory of "dependent
atising" (prafityasamutpada) in the form in whjch it was fotmulated by the
Buddha in the early discourses.

4.

Rupe Jaty eva iiipasya kiiraf!af(l nopapadyate,

riipe 'saty eva ropasya kiirat}at?Z nopapadyate.

EXAMINATJON Of ACG REG ATES

143

When a materia form exists, a cause of rnaterial form is not appropriare.

When a materia form docs not exist, a cause of materia! form is also not
appropnate.
MKV(P) p. 124 ; M.KV(V) p.AS

Herc Nigirjuna is reverting back to che criticisms he made of the Sarvastivada

and Sautrantika views in Chapter I. The first is the ideotity theory of causation
and the second, che non-ideo city theory, and these criricisms are here app ied
to the causality of materia form (ropa).

5. Nifkaraf!af?J- puna rupa1p. naiva naivopapadyale,

lasmiit ropa-gatan klhrilcin na vikalpan vikalpayet.
Furthermore, a materia form without a cause is absolutely inapp ropriate.
Therefore, one should n_ot cliscriminarively think of anything confined to
materia! form.
MKV(P) p. 125: JlfKV(V) p.29.

A theory of an uncaused event, as reiterated , was cleady unacceptable co

Nagarjuna. Indeed, it is emphasized by the repctition of the oegarion, naiva.

Not accepcing the epistemology chat is gencrally and indiscriminarely at
tributed to Nagarjuna, we have avoided tra ns a ting the term vika/pa either as
"conceptualization" or "conceptual construction" or even "discriminacion," (see
Introduction). The importance of the term rupagafiin cannot ge overemphasizcd. The comments on the previous verses show to what extent
speculacion "confined" suictly to marerial form (ropa) led to all lcinds of weird
phi losophical theories. The analysis of materia form should be undercaken in
relacion to various other issues, especially language and epistemology . [n the
lasc few verses, Nagarjuna was engaged precisely in such an enterprise. Hcnce
his advice in the present verse. The suffix -gala is better understood in the
meaning in which ie occurs in phrases like kaya-gata ("confined to the body''),
ha11a-gata ("confined to che hand"), etc.

144

6.

Trm PHILOSOPHY OF THE MIODU! \V/ AY

Na k1iraf!uya sad(ia??J karyanJ ity upapadyate,

na karaasyasadram kiityam ity upapadyate.

The view that the effect is identical with the -cause is not appropriate. The
view that the effect is not identical with the cause is aJso not appropriate.
MKV(P) p. 125; MKV(V) p.49.

Unlike I\T.4 wruch sc.ates the lnappropriateness of identity and non-jdendty

theories of causation applicd specific;aUy co the subjecc matter under discussion,
narody, materia form (ropa), the present verse emphasizes t he inapp ropriatencss of identity and noo-idencity theories in generaJ .

7.

Vedan'?i-ciJta-Ja??J;mhiiim.- sarrutiiirar/ii'!t ca sarvaaf?,

sarvefiim evtT bhtivtinlif?J riipef!at'va s11maf? kramaf;:

The method of creatrnent of all existents such as feeling, thought, perception and dispositioru is in every way similar to that of materia form.

MKV(P) p . 126; MKV(V) p.49.

The previous comments on the contents of this chapcer would indicace that
Nagarjuna dld not dcny the reality of roateri.l form but only the method ofe:xplaining it. A similar treatment is requested of the othcr a~gregates tao, Note
the use of the term bhava in the present context, which prompts lnada to
fender it a5 "existential actiOflSH (compared wirh its usage at 1.3 wruch !nada
trans la red as "entities").

8.

Vigrahe ya4 pan'harartt krte Junyataya vadet,

sa1'1!1J.f?J ltnjtipanlJrta?p. samaf!1 sadhyana /iiyate.

When an analysis is made in terms of empti.ness, whosoever wece co ad-

EX.AMINATlON OF AGGREGATES

14 ~

dress a refutation, all that is left unrefuted by him will be equal to what is
yet to be proved.

9.

Vjiikhy'iine ya up'iilambhatp krte unyatafii vadet,

sarvarri tasyan1tpalabdha1?1 sama11J stidhyena jliyate.

When an explanacion in terms of emptiness is given, whosoever were to

address a censure, all chat is left uncensured by him will be equal to what
is yet to be proved.
MKV(P) p .127; ti!KV(V) pp.49-50.

These two rubtle and crypcic verses tan best be understood in the context jo
which chey appear. They are placed at the end of an extremdy important
anaysis of die metaphysics relacing co materia form (ropa). As.such they
should be consjdered the conclusion of that analysis.
The basie theme of the chapter has becn the rejection of any metaphysical
substance (svabhava) as the cause of materia! fo (rtipa-k'1ira,,.a). lndeed , there
was no rejection of the cause of materia form, only the criticism of the view
that this cause is an invisible permanent enticy distinct from the perceived
materia! form . This criticism if referred to ac IV.8 as "an ana.lysis in te~ms' of
empciness" (Siinyataya vigrahe krte). lnteresti.ogly, the term "empti1:_1ess"
(funyata) never occurred in char ana.lysis. The ana.lysis was made on the basis of
mutual "dependeiice" of lnaterial form (rupa) and the cause of material form
(rupa-karaf!a). Thus, materia! form and the cause are empty of substance
because they arc mutually dependent. At this scage, if someone wece to present
a refutation of Nigarjuna's view, chat refutatioo wou1d be. iotended as a
refucation of the "mu tual depeodence" of the materia! form and its cause.
However, such a refurarion does not aucomatically p.rove the valiclity of the
metaphysical idea, namely, svabhava chat is being rejecced by Nagarjuna on
che basis of "dependence." The argume~t in favor of dependeoce is experlence,
Hence, che person presenting .a refucation of this idea should be in a position
not only to negace "mu tu al dcpendence" bur also provide evidence for the
establishment of a metaphysical substance (svabhava). This has oot yet been
achieved : Thus, according to Nigarjuna, what stili remains ro be proved
(iiidhya) is the chesis regarding "substance" rather than "mutual dependence."

146

THE PHll.OSOPHY OF THE MI ODU! WAY

These two verses are very significant in that they seem to admit that the
Madhyamikas have a positive thesis, namely, "dependent arising" which is accepced on the basis of cxperience. The "emptiness" (fiinyalii) they advocate is
simply a challenge for the mctaphysician ro prove his own mctaphysics.
The tWo verses are idenricaJ except for the use of the rwo pairs of terms,
vigraha andpan"haru in the former, and vyakhyana and upala1nbha in the latter. As is wel known, Nagarjuna is also the author of a creatise called Vigrahavyavartai. The term 11igraha means "analysis." On the basis of this work,
modern interpreters of Nagarjuna have assumed that he was merely an
an:ilytical philosopher whose enterprise was confined solely to "analysis"
(vigraha) of opposing views utiiziog the conception of "emptiness" (f'iinyata).
However, IV.9, cast in the same mould as IV.8, raises doubts about the validity
of sucb an interpretation of rhe cbaracter of Nagarjuna's philosophical method .
The use of the term vyakhyana meaning "explanatioo" seems to indicate that,
in addition co "analysis io terms of emptiness," Nagarjuna was aJso providing
an "expanatioo ." That expanacion is once again said to be based upon emptincss (funyatay'ii vyiikhy'ii?Je krte). But as mentioned above, the term "emptiness" did not occur at all in the chapter. lnstead the explanation was provided
on the basis of "dependent arising" (pralityasamutpada). "Emptiness" being
che counterpart of "dependent arising," "explanation in terms of dependent
arising'' would, therefote, be the same as "explanation in terms of emptiness."
Thus, Nagarjuna, even when presenting his posirive theory of "dependent
arising," need not worry about someone censuring bim, for the theoi:y of
"dependence", likc "emptiness," was intended to reject the mecaphysicr of
subsrance, and the responsibilicy once again falls on his opponcnt co prove his
own substance-mecaphysics.

CHAPTER

FIVE
The Examination of Elements

(Dhatu-parz ksd)

l.

Niikafar?Z vidyate ki'f!Zcit purvam 'iikaa- lak!a'f!'iil,

alak!a'f!a'f!J prasajyeta syat purva?!l yadi lak1a11at.

No space is evident prior to the spaml cbatacteristics. lf it eicists prior to the

cbaracteristics, then it would follow that it is without characterisrics.
MKV(P) p. 129; MKV(V)

p.~l.

As pointed out earlier, the caregories of aggregates, spheres, and elements constituted an important pace of the Buddha's ceachings as well as of the
Abhidharma analysis (UL l ). This early classification was incended to accounc
for che human personality (skandha), its experience ( 12 ayatanas = 6 indn'yas
and 6 111/ayas.) and finally the dements to which this whole experience can be
analysed (18 elemcnts = 6 ayatanas, 6 v1jayas and 6 forms of consciousness).
Howevcr, Nagarjuna does nor cake them in that particular order. Th is is
because of the problems creaced by the ioterprcters of the Abhidharma. They
were more concerned with defining each one of che elements in eacb of the
catcgories without considcring them in the light of other elements within the
category to which ie belongs or in terms of ocher cacegories oucside of
themselves. This led to a wide range of mctaphysical speculations . For example,
one of the elemeots (dh'iit11) chat caused much misunderscanding and led to
many metaphysicaJ ideas is the conception of "space" (iik'iifa) . Although
"space" is not an item included among either che aggregatcs, spheres or
elements (in rhe earlier classificacion), ii indeed was pan of an analysis of che
human personality comparable co the analysis into five aggregates (skandha).
The counrerpart of the skandha-classificarion is the explanation of the human
personality io terms of six elcments (cha-dhatu or !afidhii,tu), one of which was
"space" (akafa) (M 3. 239). At S 2.150, the Buddha specifically recognized che
inrerdependence of material form and "the sphere of space"

( Yaya'f!Z .. . akaJaiJcayatanadhatu ayar!l dhatu rupa"!J pa{icca pailwyat1).

However, the Buddhist metaphysicians, creating cach one of the carcgoes and
147

148

lHE PHLLOSOl'HY Ol! THE MLODLE WAY

items independend y. assumed that "space" is "uncond.icioned" (asa1pskrta1?J,

Akb p. 3). the Jactet being unde1stood as "indepen dence" (see Sakv p .174;
DhsA p .193). Thus, space came to possess the unique narute of "nonobstruction,,- (aniivarana-svabhavam akiia'f!J) where mate~ial form finds its
locale (yatra ropasya gati/?, Akb p. 3).
This, indeed, was a very absolucistic conception of space , a conception chat
muddled u p the en tire Buddhisc epistemology. Ie posed the most significant
challenge for Nagarjuna. In order to eliminate such metaphysics, Nagarjuna
adopted the ingenious mechod of analysing the source of knowledge (i,e., seeing, darana, Chap ter lll), t he object of knowledge (i.e materia) form, rupa,
Chaprer IV) and its locale (i.e, space, ?ikafa, Chapter V) aod this he did on the
pretext of examining the Buddha's own categories of aggregaces, spheres and
elemen ts (skandha-ayatana-dhatu). Thus, the presenr chapter on the examina
tion of elem ents comes to be devoted not to the traditional category of
elements, but rather to the concep rion of space, hig bJ ~ghting its relativity,
especially co material form, discussed in the previous chap rer. With these three
chapters. Nagarjuna was thus able to give a rather com prehensive account of
the problem of knowledge.
Here there is no denial of space, but only the rejection of a particular way of
u nders tan~iing or conceiving it. If space were t be undersrood as the ult.imately
real pure enrity, a substance to which various characteristlcs ate attributed, in
wbich case space precedes tbe c:haracceristics, then Nagarjuna finds no
epistemologicaJ justificatioh for it. This is a c:riticism of the first aspect of the
metap hysical explanation of sp ace referred to above, namely, thac space has the
cbaracter of non-obstructon (anavaraf!a-svabhiiva) . The obsrruccion, in this
case, tefers to mare.rial form (i.e., pratigha).

2.

A/akf<1f!O na ka:icic ca bhavalJ Ja1rividyate kvacit,

asaty alak.rflf!e bha11e kramafiif!I kuha lak.raf!a'/'!J.

An existent tbat is without cbaracteristics is nowhere evident. When a n

existent without characteristics does not exist, where can cbaracteristic:s
appear?

MKV(P) pp.129-130; MKV(V) p .51.

The denial of pure "space" at V. I is here extended to all ttlemeo ts or entities

(bh'iiva). The question chat follows next is: lf there wece to be no pure entity or

TH E EXAMINATION OF El.EMl!NTS

149

a tabula rasa, then one also cannot account for adventitious elemeots through
which something comes to be koown. ihis tpresents the second aspect of the
metaphysical definition mentioned earlier-, namely, t he "unobstructed space"
serving as the locale for the appearance of materiaJ form (yq.tm iiiptJJya gatif? =

iiipasya kramata7(J).

3.

Niifak!af!e fak!af!tJJYa pravrtttr ha Jalak.faT!e,

sa/aktaniilak.fanabhyif!l niipy anyatr(I pnwartate.

The occurrence of a chacacteristic does not take place either in something

without chacactedstic or in something with charactecistic. Nor does itproceed from something other than those1 with or without characteristic.
trlKV(P) p.130; MKV(V) p .5 L

Here again, it seems inappropriate ro assume thac Nagarjuna was denying

either space or materia form , but only the mannet in which they were explained by the metaphysicians. It is the sharp dichotomy berween a thing and its
properties chat cannot account for either of them. The subscantiaJist mode of
speakir.g about entities (bhava) leads t two extremist views. oamey , identity
or difference. The second scatemeot rejects the view that the ptobem ofidentity and di.ffereoce can be solved by transcending both. The reasoning seems to
be that, rather than attempting co salve the problem of identity and dilfereoce
by follow-iag the method of traosceodence, it is mare appropriatge not to create
such a sharp distinction in the first place.

4. lak}"a1'faJa?!Jpravrttau ca na lakfyam upapadyate,

la~ya.syanupapattau ca lak!af!asyapy asa?!Jbhavai{,
When the characreristic does not occur, the characterized is not appropciate. In the absence of the characcerized, there is. no occurrence of
the characteristic.
1\ifKV(P) p . U I : MKV(V) p. 52.

L50

TuE PHILOSOPHY OFTllE MIDDLE WAY

The relativiry of concepts, or more spedfically the contextual meaning of concepts, is herc underscored. Definitions may be useful in clarifying the meaning
of terms. Yet these meani ngs are not derived independencly; they occur in conrexts and, as such, any reference to subsrance and attribute shouJd not be taken
co imply distinct or pure referents. What are denied here are not the concepts
of the characterized or of the characteristics, but merely their independent ex1stence.

5.

Tasman na 11idyate lak.fyaf!l lak1at/af!l nai11a 11idyate,

lak,ya-la/qana-nim1ukto 11aiva bhavo'pi 11idyate.

Therefore, the characterized is not evidenct. Neither is the characteriscic

evident. Distinct from the characterized and the characteristic, a n existent is certainly no t evident.
A!KV(P) pp. 131-132; MKV(V) p . .52

This is not a blanket denial of the cha.racterized and the characteristic. Rather it
is a denial of these two clements as expajned at V.4, namely, as substance and
attribure conscirucing independent entitics. Therefore, one cannot find any
event, any enci ty, any existence tbac is separated from the characterized and the
characteristic. An existent separated from the characterized as wel as the
charactcristic could turo out to be a "pure entity ," an idea clearly unacceptable
to Nigarjuna.

6.

A11tdya1niine bhave ca ka.syabhavo bhavqya11:

bhihiiibliiiva-vidharmii ca bhiivabh"iivam avaiti kaf?.

When an existent is not evident, whose non-existence can thcre be? W ho

could comprehend the cli~tincr things: ex.istent and non-ex:iste.11t as well as
existence and non-existence?
MKV(P) p. 132; MKV(V) pp.52- 53.

TH!! EXAM1NATION OF ELBMENTS

15 1

Herc again, if we arc to understand "cxistent" (bh7i11fl) wichout takjng inco consideracion the son of "existent" referred to at V. 5, we would be left wich a
universal statement regarcling all "existeots" (bh7i11a). Yet, it is not meant to
be taken that way. The "existent'' referced to herc is already defined at V. 5. It is
an "existent" chat is separated from either the characterised or the characteristic
or both. lt is a "pure existent." Such a pure existent could then be contrasted
with a pure non-existent, and i~ is this sharp distinction or dichotomy chat is
being questioned by Nagarjuna. Indeed , it is significant to note the use of the
term 11i-dharma (whose occurrence in any other text is not knowa to the present
author) in the sense of ''distinct things."

7.

Tasmiin na bh7ivo nabh7i110 na lak,ya?'!J -niipi lakfa!Ja'f?Z,

aliiiaf!t ak.aa-samli dhatavalf paca ye pare.

Therefore; there is neither an existent nor a non-existent, neither the

characterized nor the characteristic, neither space nor the other five
elements similar to space.
MKV(P) p.1 34; MKV(V) p.53.

Thus, neither eXJStence nor non-existence, the characcerized nor the

characcerjstic as envisaged by the metaphysicians exisc. The five
elernents= earth, water, fire , air and consciousness, which together with space
constitute . the personality (cha-dhaturoya?'!J punso)-do not exist if these
elements are conceived of in the same way as space. This does oot mean that the
way in which the metaphysicians conceive of the six elements is the only way in
wbich tbey can be understood and explained.

8.

Astiva'fP ye tu ptlyanti nastif1Jt1f!J calpabuddhayal;,

bh7ivanii?'!J te na pa.fyanti drfJ/{avyopa.fama?'!J fiva?'!J.

lhose who arc of little intelligence, who percdve the existence as weU as

152

THE PHilOSOPHY Of THE M IODLE W AY

che non-existence of existents, do not perceivc the appeasement of the

object, the auspicious.
MKV(P) p. 135; MKV(V) p. 54 .

The "appeasement of the object" (dra1fa11yopaa1na) is, no doubt, a synonym

for the ''appeasement of dispositions" (1a1!JSkaropaJama) or the "appeasement
of obsessions" (prapancopaama) . lt is indeed not the elimination of the object
(drfl!{avya-k1aya), implying the abandoning of both subject and objecc. lt is
mcrcly che appeasement of the object. What is implied by the appeasement of
the object is its non-reification . The concepdon of the non-existence of the object will emecge only if its existencc is understood in a substantial way. Existcncc (a1titva) and non-existence (ni1Jitva) arc correlave. Asscrtion or denial
of one involves the assercion or clenia! of the ocher respectivcly. Asserng cxistcnce in a mctaphysical way one is !cd to the denial of existencc. Denying cxistence in the same wa.y one is led to the assertion of non-existence. Instead of
eliminating both existence and non-existcncc and looking for a transcendent
reality, a reality that is beyond both exiscence and non-existence, it is possible
co appease. calm, or pacify one's cispositions (1af!1.Jkara) or obsessions
(prapaflca). Dispositions or obsessions, when foilowed to their positive extreme, lead to che belief in a permaneot exiscence (aititva); when com pletel)'
negated they rontribute t oward the belief in non-existencc (na1tit11a). Hence
the emphasis on rhcir appeasement rather chan cheir promotion ot elimioation.

CHAPTER

SIX
The Examination of Lust and the Lustful

(Ra ga-rakta-pari k.r a)

1.

Ragad yadi bhavet purva?(l rakto raga-tira1krtaf?,

fa?IJ pratiya bhaved rago rakte riige rago bhavet sati.

lf a lustful one, separated ftom lust, were to exist prior to Just, then
depending upon bim there will be Just. Lust exists when thete is a Justful
one.
MKV(P) p.138; MKV(V) p.55.
.
.

So far Nagarjuna was considecing the basie elemencs (dharma) lnvolved in an

explanation of the problem of perception, nmely . the "faculties" (indriya),
rhe ''aggregates" (sk1t11dha) , and "elemeots" (dhatu). However, the analysis of
perceptua experience, and rherefore of the elements (dharma) invoved in such
experience, is not confined to these. ln the Buddhist view, lusc (raga) is an im.
pottant constituent of perception, primarily because in Buddhism the analysis
of perception was not undertaken for its own sake, but for the sake of discovering the cause of bondage and freedom.
We have already referred to the confusioos created by Candraklrti, especially
in i:egard to the interptetation of the most crucial Chaptec I where he fails to
recognze the significaoce even of the ordec in-which Nagarjuna takes up the
variety of ideas for examinacioo, let alone the ideas themselves.
Another unfQrtunate misinterpretacioo emerges when Candrakrrti' assumes
that every ioicial verse in every chapter represc.nts a sratement of the opponcnt's
view which is to be repudiated. While this may be true in some chapters, thete
is no necd to universalize it . There is oo reason why Nagarjuna could not cake
up his or the Buddha's views first and then go on to repudiate what are considered co be inappropriate ideas.
In the Buddha's owo analysis of perception, opsessions (prapaca) appears in
a personality th~c isalready smeared with !ust. The Buddha consistendy avoided any speculatioo tegarding absoute origins. As such, he was not willing to
assurne either a pure personality, a tabula rasa which comes to be defiled by
153

1~4

TuE PHILOSOPHY OF THE MIODLE WAY

adventitious elemenrs (see A 1.254) or the existence of an element caed Just

(raga) in the external world independent of a conscious person who generates
lust on the basis of external objects (see S 1.22). For him. objects can be
classified as pleasurabJe (ubha) , loathsomc (aJubha) . or neuua only in rea
tion to a perceiving iodividual who is prone to make such distinctions. A person
who is prone to make such distinctions is one who is either dominated by Just
(raga) or aversion (dsa) or indifference.
Thus, Nagarjuna is able to assert tbat if any object were to cause any lusc
(raga). chen there must be a lustful one (rakla), noc simp ly a "pure person" who
is uotained by lust. For him, tbere is no difficulty in speaking of !ust, so long as
that lust is not distioguished as an entity which is then supposcd co infect a person who is pure and undefiled by !ust. lt is the same sort of relation that is exeropified by a compassionate one (maitra) and the recipient of compassion
(upagr'iihaka), a very apt example given by Candrakfrci ro illusuace the conception of "otheroess' (parat11a). but whicb has been misunderstood by many a
translator [MKV(P)p.78; Sprung , Lucid &position, p.66, who was probably
following Stcherbacsky's earlier translation]. There is oo sense in speaking of a
compassionate one or even compassion in a vacuum. A compassiooate one or
compassion bccomes meaningfu only in the concext of people who are recipiencs of such compassion. The same sort of celarionship exists among the
ustful (rakta), !ust (raga). and the objeccs chat generate !ust in the individuals.
The present verse, cherefore, is a elear scacement of Nagarjuna's own position, noc the statemeot of an opponent that is to be rejected.

2.

Rakte 'sati puna r'iigaf? kuta eva bhavifyati,

sati vasati v'ii r'iige rakte 'py e!a sama/; kramal/.

When a lustful one does not cxist, whence can tbere be lust? Wbether
lust exists or not, the method (of analysis) even of the lustful one would
be comparable.
MKV(P) pp. 138-1:39; MKV(V} p. 5~

Thus. Nagarjuna raises the question as to how there could be !ust in che
absence of a luscful one . This avoids the theoty of a tabula rasa and the adventitious impressions. Lec alooe the exisrence or the non-exiscence of lusc. even
the lustfuJ one has to be analysed in terms of dcpendeoce , not in terms of pure
entities having tbeir own sdf-narure (svabha11a) .

THE BxAMINATION OF LUST AND THE LUST~Ut

155

3. Sahawa puna udbhutir na yuklii r'iiga-raktayo/?,

bhavet'iir(l iiiga-raktau hi .nirapc/qau parmpara?('.

Again, the simultaneous occurrences of lust and the lustful one is not
proper. Lust and the lustful one would thcn be mutually non-contingent.

MKV(P) p.139; MKV(V} p.56.

To conceive of lust ot the lustful one in vacuity and then proceeding to explain
their simultaoeous arising (sahaiva udbhUtir) is not the purpose of "dependent
arising." In such a case, what is found is independent existence of lusc as well as
the lustful one, each having itS own nature. This, once agaio , represents Nagarjuna's rejeccion of the substance/ attribute relationship. "Cow-ness" implies the
fact of having "four legs;'' the latter is not an independent attribute of the
former . The dangers inherent in catrying out the analytical process initiared by
the Buddha to its extre.me i here indicated . Analysis is usefuJ in eliminating
metaphysiES, but useless when its leads to further metaphysics.

4.

Naikatve sahabhavo 'sti na lenaiva hi lat saha,

Prthaktve sahabhavo 'tha kuta cva bhavifyati.

In identity, tbere is no co-existence. That which is associated does not

arise together. In discreteness, how can ther:e be co-existence?
MKV(PJ pp.139-140: MKV(V) p.56.

The philosophical probleros created by an over-extended analycicaJ process cannot be resolved by either a cooception of ideotity or of discreteness, The relationship of co-existeoce (1ahabha11a) cannot be established once the analysis
leads to a sharp <;ichotomy becwee~ substance and arujbute.

5. Ekatve sahabhii11a1 cet sy'iil sahayatr1 11inapi sah.,

Prthaktvc sahabhavaJ cet syat .rahayar?f. viniipi salt.

156

THE PH!LOSOPHY 01' 'THE MIDDLE WA Y.

lf, in identity, there were to be co-existence, it coud oc.cur even without

association. If, in discreteness, there were to be co-existence, it could occur even without association.
MKV(P) p. 140; MJ(V(V) p.55

The Sarvastivada conception of identity in terms of substan tial existence

(s11abha11a), each substance being a part of a larger and mote pervading
substance (like Plato's "forms") did not need a concepcion of association. Coexistence comes to be overshadowed by idencity (just as much as the threc
Platonie forms ~ truth, beauty, and goodness-eould be found in amore eomprehensive form, i.e Good). The Sautrandka anaJysis of reality into diserete
momentary entiries (compaiable to the Humean impressions) led to an explanation of co-existence in terms of "associaiton." Io faet , such c;o-existence
cl.id not need any "association." Things could co-exist even wichouc association.
Thus, association i.s not a via.ble solution to problems either of ideotity or of
discrereness.

6. Prthaktve sahabhavaf ca yadi ki'f{Z f1iga-raktyo/,

st'ddhaf? Prthak-prth<Jg-bhaval/ sahabhavo yatas tayo/J.
If there were to be co-existence in discreteness, is it the case that lust and
the lustful one arc completely separaced, as a result of which their coex:istence is also established.
M.KV(P) p .140; MKV(V) p .56.

The problem of "associacion" arises more with the eoneeption of"discreteness,"

than with the nocion of identity. "Association" was not much of a con cern for
the Sarvastivadins. However, it was indeed a r:najo.r problem fot the
Sautraoti.kas. Hence their emphasis on the coneeption of " immediate contiguity" (1amananlara). Por Hume, with a similar atomie analysis of experience,
"association" was the- glue that bound togethet the discrete impressions. The
present question of Nagarjuna was a challenge co that Sautrantika (/Humean)
cheory of the ''ideas of association. ''

THE EXAMINATION OF L UST ANO THE LUSTFUL

7.

157

Siddhaf? Prthak-prthag-bhavo yadi va raga-raktayof?,

sahabhavaf?Z kirn arthaf!J tu pari/uz/payase tayof?.

If complete separation between ust and the lustful one is escablishcd , for
what pucpose do you conceive of their co-cxistence?
MKV(P) p . 141 ; MKV(V) p .56.

The uscfulness of appeaing to co-existence after analysing evenrs in co discrete

entitics is here questioned. Nagarjuna's approach here is to resove the very
idea that gave rise to the problem racher chan solving ir, namely, co avoid
creating sharp distinctions instead of trying to find solutions to problems
generatcd by such distinctions.

8.

Pr1hag na siddhyaffty eva'f!l sahabhatlat!J vikank,asi,

sahabhava--prasiddhy artha??i Prthaktva'f!J bhuya icchasi.

You fancy co-existence assuming that the cliscrete is not establishcd. You ,
again, look for cliscreteness for the purpose of establishing co-existence.
MKV(P) p . 141; MKV(V) p.57.

No better explanation of the dilemma of an analytical philosopher who

recognizes discreteness can be found than in the present scaceroent of Nagarjuna. The vicious ciecie in which one gees invoved when, after anaysing things
into discrete entiries, one tries to put things together is dearly explained here.

9. Prthag-bhiivaprasiddhef ca sahabhavo na siddhyali,

ka tarnas1nin Prthag b h'iive sahab hava'?"J salicchasi.
When discreteness is not cstablishcd, co-existence is not establishcd. In

158

T HE PHlLOSOPHY OF THE MIDDLE

WA Y

the presence of what kind of discteteness would you expect co-existeilce.

MKV(P}l p .142 Mk'V(VJ p.57.

Nagarjuna's question herc is specifically oo the nature of discreteness. If

discreteness is absolucc, then association of co-existent entities is not acceptable. If there is no such discreceuess, then there is oo need for co-existence, for
the lack of complete discreteness implies "dependence" (prafityasa111utpada).
Hence, the quescion pertains tq the kina of discreteness envisaged.

10.

Evaf!Z raktena raga.rya siddhtr na saha niisaha,

1liga11at saf'tlatlhamUif'!af!Z siddhir na saha nasaba.

Thus, with or without the; lustful one, there is no establishment of !ust.

Lilce lust, there is no establishment of anything with or without [accompaniments].
MKV{P) p. 142; MKV(V) p .57.

On the basis of the kind of discreteness referred to above, one caonot explain
either the association or the non-associa tion of lust and che lustful one. Avoid
the sha.cp distinccioo, the problems are not there to solve. The same can be said
in regard to all phenomena (d}Jarma). Fot this teason, neither absolute identity
nor absolute di.screceness can establish the nature of phenomena.

CHAPTER
SEVEN

The Examination of the Conditioned

(Sa'f!lskrta-parz kfa)

I.

Ylll"ii JtZf!Zskrta utpidas tatra yukta tri-lakfaf!t,

ath1is1Z1?1.skr:ta utpada~ katharp. sarp..slta-lalt;a11arp..

lf arising is co.oditioned, therein three chatacteristics are proper. lf arising
is unconditioned, how can there be charactcristics of the conditioned?
MKV(P) pp.145-146; MKV(V) p.59

The examinaiton of the "conditioned" (1arp..skrta) coming immediatey after

the analysis of lust (raga) and the luStful one (rmeta} brings out another important aspect of the B~ddha's conception of the pragmatic meaning of truth, a
conception that Nagarjuoa seems to be de:u:ly aware of.
Lust, as pointed out earlier, is one of the most important eements in the
Buddha's analysis of experience. Lust is operative in the perceprual process
espedaly in the formation of idcas derived from experience. Having rejected
"omniscience" (sarvajnat11a) as a source of. knowledge, the Buddha depcndcd
primarily on sense expcrience. However, fot him, sense experience was a "big,
bloomiog. buzzing confusion." One way of dealing with this confusing mass of
sense data is by concenuating upon nems that arc of interest to the indiv.idual
and then forming ideas. Such seection is generally based upon one's interest.
This is the signific.ance of "dispositions" (I(Z'tJSkira "' compounding of ideas).
For the Buddha, one's conception of truth is invariablybound up with such
dispositions: Therefore, all ideas are "dispositionally condicioned" (1af!Zskrl4).
The Buddha's fina statemcnt:before his death: "Dispoitions are subjcct to
change" (11ayadhamma sakhara, D 2.156), thetcfore , is an assection that,
since ideas are impermanent, there can be no absolutc truth.
However, the above conccption of the "dispositiooally cond.icioocd''
(Jaf?Ukrta) was to undergo a cadical change at the hands of the Buddhist
metaphysicians: Nagarjuna's examination of saf!JSkrta becomes meanirigful onJy in the background of chat change of perspeccive. His was. indeed, an auempt
to reject the Sarvastivada and Sautrantika intcrpretations of sarp..skria (and this

160

THE PH!LOSOPHY OF THE MIDDLE WAY

would apply ro the ideas that carne co be acceprcd by che later Theravadins)
and to reven back to the original teachings of the Buddha.
Once again, the Sarvastivada school was responsiblc for initiating this change
of perspeccive. In the eady discourses, the term Ja'T!Jskrta is used along with the
term pralilyasamutpanna, bur oot as synonyros (S 3.96, 103). While the
former implied "the disposicionally conditioned," the latter was used in the
moce comprehensive meaning of "the dependent." Thus, all dispositionaly
conditioned pheoomena are dependent, but not all dependent phenomena are
dispositionally conditioned. The equatioo of sa'f'(Mkrta and pralitya1am111panna
occurs for the first time in the interpretation of the Abhidharma. 1t was this latter traditioo that Vasubandhu was recording when he said: "Those chat arc made
by the conditions having come cogethcr arc caJled the samsk[tas" (sametya sambhuya pratyayail? krtii iti sa11J1krliill, Akb p.4; see also AA 2.252 , paccayehi
samiigantva kataua). The Sarvastivadins secm to have understood the concept
of sa'T!1Skr1a as a mere refutation of the bdief in the production of an event by a
single cause (eka-pratyaya-janilaf!#, ibid .). Yaomitra, commenting upon this
statement, goes on to say that Ja'f'(Mk('ta and pralityasamutpanna are, therefore,
synonyms (Sakv pp.171-172), thereby bliterating the se ma n tie difference between "made" (krta) and "arisen" (utpanna).
This erroncous simplc equation was to lead to further complicatioos ,
espccially in uoderstandiog the Buddha's characterizatioo of niniiit!a as
tl1t1"!1Jkrta. Eveo though the early discourscs prcsenred ni1'71af!a as an aJaT!JSkrta,
ir was nevcr coosidercd to be an apralityasamutpanna ("independent") . For
early Buddhism. borh sa'T!IJkrta and asaf!'skrta are pralityaJamutpanna.
However. the Sarvastivada equatioo led to the equation of their oegations as
wel . Asa'T!IJkrta secms ro have becn understood i.n the sense of t1prafitya1amutpt1nna, To what extent their perspcctive was dominated by an adherence to the
notion of self-nature (wabha11a), for which they gave no causal explanation at
all ocher thao merely maintaining chat it is permaneot, remains a surmise,
Ie is this Sarvastivada conception of sa'T!1Jkr1t1 as beiog identical wich
pralityaJamutpanna thac is beiog criticized at Vll.1. Nagarjuoa, as indicated in
the first line of this verse, had oo difficuty in assunilog that che Jflf!'Jkrta is
cbaracterized by arising, change, and cwing. This iodeed was a statemeot atttibuted to the Buddha in the early discowses, where it is said: " Monks, there
are these three characteristics of the disposiciooally conditioned . The arising of
that which is dispositionaJly conditioned is evident. l ts cessation is also evideoc.
Change of what has come to eodure is also evident," (A 1.152). Yct if, as explaioed by the Sarvastivadins. sarpskrta is ideotical with pralityaJamutpanna
(the latter also accouoting for arisiog, change and ceasing), then one cannot
peak of thcse tltee characteristics in the contcxt of the asa'T!IJkrta.
It may be of interest co noce that it is not only nirv'ii'!a that came to be included in the cacegony of asa11J1krta by these mecaphysiciaos. Thcy also admit-

THE EXAMINATION OF THE CONDITIONED

161

red "space" (akaa) as an asaf!Jskrta. The implications of that view was examined by Nagarj1,1 oa in Chapte( V.

2.

Utpadiidyas trayo vyasfa nala1!1 lak1a?Ja-kamzaf!i,

saf!lJkrtasya samastaf? syur 'ekatra ka1ham ekadii.

When t he triad consisting of arising, etc. are. discrete, they are not adequate to function as chara~ecistics of the conditioned. lf they were to be
combined, how can they be in the"same place at the same time?
MKV(P) p. 146; MKV(V) p.59.

Further objections co the Saivastivada and Saurra ntika rheories relaring to the
characteriscics of the conditioned are raised here. The d(screteness of the thtee
characteristics, as envisaged by these two schools, would not allow them to
function as che characteristics of the conditioned. The Sarvastivida theory of an
unqedying substance, which comes to be superficially characterized by rhe
three (or four) moments, as well as the Sautrantika theory that each moment is
inherent in the previous one in the form of pocentiality or seed (akti, bija),
were the direct rcsults of such a pcrspectivc. Furthermore, if rhcse
characcerisrics were disrinct in elation ro both time and space, thcy could not
occur in any one saf!Jskrta at che same time.

3.

Utpada-sthtti-bhaganam

anya1 saf!Jskrta-lakfarJfl'f!J,
asti ced (lnavasthaivaf!J nasti cet te na Ja'l'(lskrtalr

If there wece to. be a characteristic of the conditioned other than arising,

duration., and destruction 1 tbere would be infinite regress. lf there were
to be no such [charact~ristics], t~ese would not be conditioned.
MKV(P) p. t47; MKV(V) p .60.

lf arising , stasis, and ceasing are three distinc;t evcnts, chen each one of thcse
wiU require further characterisrics of arjsing, stasis, and ceasing to a.ccou~t for

162

THE PHllOSOPHY OF THE MIODLE WAY

themselves. Thus, the moment of arising will need three f\uther characteristics
of arising, stasis, and ceasing bcfore it could give rise to the next moment,
narncly, stasis. The same applics to stasis and ceasing. This wiJI lead to infinite
regress (ana11asJhi) . On the contrary, if cach of these m oments do not possess
further characteristics, then they cannot be defined as the conditioned.

4. Utpadotpada utpado mulotpadasya ke11ala'f(J,

utpadotpadam utpado maula janayate panaf?.

The asing of arising is exdusively the arising of pmary

the primary arising produccs the arising of arising.

asing.

Again,

MKV(P) p.149; MKV(V) p.60.

In order to avoid infinite regress, one may assume that the arising of arising is
che primary arisiog (mula-utpada) , and that this latter again causcs the arisiog
of arising. Such mutuaJ action on the part of primary atising and arising of arisiog coul~ climinate infinite regress.

5.

Utpadotp'iida utpado mulotpad~ya te yad1;

maulenajanit~ la'f!l te sa katha'f!l janayifyati.

lf asing of arising is the prim2ry arising, not being produced by the

pei.mary, how can it [the. former) produce that [the laner)?
MKV(P) p. l)O; MKV(V) p.61.

The questioo raised herc by Nigarjuna pena.ins to active causation . The

substantialist (Sarvastivada) view of causacion, which assumes the essence of
the effect to be already lateot in the cause,. does not permit mutual dependence
of cause and effect. In sucha case, the cause-effect relatiooship would be a oneway relatiooship. Nigarjuna perceives that the ioterdependence utilized in
VU.4 in order to avoid infinite regress would not be appropriate in the context
of a substantiaJist theory of causation.

THE EXAWNATION OF THl! CONDl'IJO~ED

163

6. Sa te maulena janito 1111JUla'f!J janayate yadi,

maulr!I? sa tenajanitas lam utjiiidtJyate katha'f!J.

lf, produced by the primary, it produces the primary, how can thar
primary, not produccd by it, produce h?
MJ..."V(P) p. 150; MKV(V? p.61.

Here again, Nagarjuna rcfuses to rccognize the mutuaJ dependence of causc

and effect in the contoct of an active or substantialist theory of causarion.

7. Ayam utpadyamanas te kamam

utpiidaye

tfna'f!J,

yadimam uipadayitum aj'iital? aknuyiid ayt1'1!J.

This, while arising, if it may so dcsire, producc tbat, so that it, bcing not
yet bom, will be able to ptoducc tbat.

MKV(P) p. 150; MKV(VJ p .61.

At this point Nagarjuna proceeds to state the Sautrantika theocy of causation,

whkh is described vcry succinctly by Vasubandhu (Akb pp. 76-77). The
Sautrantikas believed that the Buddha's discourse on the three characte.riscics of
existence-arising (utpiit:fr!) , change of what has endured (sthitasyanyathiitva) ,
and ceasing ( 11yayt1)-was intended for the folish people who arc blinded by
ignorance (avidyandhabaliif?). According to them, the rccognition of ~uch
charactcrisdcs iovolves the belief in the subs~ntial e.xistence (a1JittJa) of the
"conditioned'' (Jaf!lskrta) (sec ibid. , p. 77). Howcvcr, thesc three charactcristics
arc not part of reality which consists of momentary (kfaf!ika) cvents. Such
momentacy events or imprcssions coming one after another in rapid succcssion
(PUrvasya purvasyottarakfa?Jasyiinubandhaf?) produce~ the appearancc of
"change of what has endu.red" (Jthityanyathatva) , -vvbereas it is merely a serics
of cveots (pra11aha) resembling one anothcr (a11isadra). In sucha contcxt, arising is merely the immedidatc arising from a state of non-existence
(pratt'k!4'!am abhii.tva hhiiva utparlal(). A moment thus comes to be from
oowhere and ceases immediately. (Here one is naturally remindcd of the Humcan version of causal rdacioos.)

164

THE PHILOSOPHY OF THE MIDDLE WAY

In the presenc verse, Nagarjuna is providing his own description of the

Sauttanrika theory of change. As chis cheory is based upon the recognition of
momentarioess (k!aT!ikatva), Nagarjuoa insiscs that the preceding event has to
produce the succeeding event (in the series of similar events) even before the
form er is born, for the former does not endu re in order co prod uce the laner. In
other words, if there were to beany causal connection between two momencary
eventS, the moment chat serves as che cause should beready to generare the cffect before it passes away, that is, eveo beforc it is boro.

8.

Pradipal/ sva-par'iitmanau Ja??Jprakafayita yatha,

utpiidaf? svapaiiitmiinav rbh'iiv u1padayet tathii.

As a light illuminates itself as well as others, so does arising produce both

itself and others.
M.KV(P) p. 15 1; MKV(V} p.62.

Here, a meraphor is utili:zed to exp\a.\n che Sautrancika version of the causal

theory. The thrust of the argument i~ ~hac ie is noc necessary for an event to en.duce before it can produce some other cffect. A lamp or light does not have to
remaio for a while beforc it could illuminate itself as wel as others. This is che
same argument ~cd by the Sautrantikas and ther idealistic schools to ju'stify
the existence of a c'tigito discussed above (see also Masaaki Hatrori, Dignaga On
Perception , Cambridge: Harvard Uoiversity Press, 1968, p. 101). The
Sautrancikas feared chat any conception of duration or scatis (sthitt) will lead co
the unacceptable view of subscance (111abhiiva).

9. Pardipe niindhakaro 'sti yatra ciisau prafi!fhitaf?,

kif!1 prakiiayati dipaf? prak'iio hi tamo- vadhaf?.
There ex.ists no ~rkness ei.ther in the light or in whatever place it is
siruatcd. What does light illuminate? For, illumination is indeed the
dcstruction of darkness.
MKV(P) p. 151 ; .MKV(v,J p.62 .

165

lHE EXAMINATION OP THE CONDITIONED

Nagarjuna's criticism of the Sautrantika theory begins here. In order to get rid
of the metaphysics associaced with the Saucriinika cheory of change and
causation , Nligarjuoa begios by quescioning the meaning of the term "illumination" (prakiifa). For bim, light (a/oka) and darkoess (andhaliiira) are
relative ideas. This same idea is expressed by the Buddha at S 2. lSO, wherc it i.s
said: ''This so-called element of light is known through its dependence upon
darkness'' (yayaf(J . , , abhadhiitu ayaf(J dhatu andhakiiraf(J paficca paflfiiiyati;
see also Nagarjuna's discussion in his R.atniiva/i [Ratnava/i of Niigarjuna , ed.
G , Tucci,joumal ofthe Royal Asiatic Society, London, 1934, 1936, l.48]). The
relatonship herc is by way of conrrast; the absence of one means che presence of
che othcr. Thus, "i uminati on" is synonymous with the "destruction of
darkness" (lamo-vadhaf?). one is nor produced by the ocher in any substantiaJ
way.

10.

Katham utpadyamanena pradipena tamo hataf(l-,

notpadyamiino hi tama!? pradipa piiipn11te yadii..

H ow can darkness be desrroyed by the emergent light , when the e merging Light, indeed , does not teach darkness?
MKV(P} p. 152 : AfA.'V(V) p.62,

The example of light and darkness used to

i l us trate

che causal relationship berween two momenrary evencs is here showo to be inappropriatc, as
ie cvenrually leads to a rarher substancialist nocion of light reaching up to
d:ukness in order to destroy ie.

11.

Aprapyaiva pradipena yadi va nihatarrt tama!?,

iha-Slahal, sarva-loka-sthaf(J sa tamo nihani!yati.

On t he contrary, if darkess is destroyed by light witbout reaching ic.

then chat [light) rcmaining here will destroy the darkness present in all
the woclds.
MKV(P)

p . 15 :\ ~

MKV(J.? p .62

166

Tu PHlLOSOPHY oF THE

Mmo W Ay

The cootrary of the substantialist view referred to at VII.10 is here shown to

lead to furthcr confusion. If light can destroy darkness without reaching it, the
implication would be that any spack of light anywhere can destroy the darkness
present in the entire universe.

12.

Pradipal/ sva-paratmiin11u Saf!Zprllkaayate yal/1;

tamo 'pi svap11ratmii1111u cliiidayi/yaty a.raf!Jayaf'!'.

If light were to illminate both itself and otherS, then certain.ly darkness
too will .conceal itself and ot~ers.
MKV(P) p.154; MKV(V) p.63.

Taking the acgument of the subsrantialist at VII.8 that light illurninates itself as
well as others, Nagarjuna is herc showing that the same could be said of
darkness too, which would invalidate the meaningfuness of the very example
ilsed by the substantialist.

13.

Anutpanno 'yam utpadaf? svatm'iinaf!Z janayet kathaf!Z,

athotpanno janayate jiite kif!Z jany4te p11naf?.

How can this non-arisen arising ptoduce itself? Hit is the ariseo that pr<>duces, theo being boro, what is it that is produced again?
MKV(P) p .157; MKV(V) p.64.

The principle adopted at I.6 in criticizing the identity and non-identity

theorics of causation is herc applied to reject the notion of a self-creating arisIng. The metapliysical assumptions associated with the Sautrantika theory of
momentary arising is laid bare by an cxamination of the so-called "deep struc
tu res.''

167

THE EXAMINATION OF THE CONDITIONED

s
Arising produces itself.

NP

VP

i. ( non-arisen) arising

produces itself

ii. (arisen) arising

produces ltself

The first alternative is impossible. The second is meaningless.

14. N olp11dyaman11f?Z notpannaf?Z nanutpanna'f?Z kath(N!1cana,

utpadjate tathakhyata'f?Z gamyamiina-galiigataif?.
Neither the present arising, nor the arisen. nor the non-arisen, is being
arisen in any way. This has alteady been explained by mea!15 of [the concepts of] present moving, the moved and the not yet moved.
MKV(Pj p. 157~ MKV(V) p.64.

The anaysis of present arising (utpadyamana), the atiscn (utpanna), and the
noo-arisen (111'1Utpanna) conceived of in metaphysical terms compares wel with
the analysis. of motion in Chaptcr Il.

15.

Utp11dyamanam utpattav idaf?Z na kramate yadii,

katham utpizdyamanaf!' tu parlf.tyotpatti1!J ucyate.

When this present arising does not proceed from wichin arising, indeed,
how can the present arising be spoken of as dependent arising?
MKV(P) p. 158; Mk'V(V) p.6j.

168

THE PHILOSOPHY OF THE MIODLE WAY

The first line utllizes the concepcual apparacus of rhe subscantiaists in speaking
of the effect ( i.e .. che present arising. utpadya1nana) as issuing out of the c:ause
(i.e., the arising, utpattt). This is a theory of self-production. As indicated in
1. 7, Nagarjuna rejecrs this causa! explaoacion as meaningless, for there is acrually no production of anything new in such a cootext. lt is mere self-re
production.
If the substancialist were to adopt the cerminology of the Buddha ( utilized so
oftcn by Nagarjuna), theo he could say that the "presenc arising is dependent
(prafitya) upon arising. '' Even though this souods like the Buddha's (and,
therefore, Nagarjuna's) formulation of the causa! principle, yet Nigarjuna is
not willing to recognize it. This is because dependent arising does oot simply
mean the reproduction of the same thing. In che present case, it is an explanation of the present arising (utpadya1niina) on the basis of arising (utpallt),
which is simply tautological with no new informatioo provided . "Dependent
arising" accounts for the arising of something new or differenr, even though
such newness or difference is not emphasized to the complcte neglect of the
cela cionshi p of dependence.
The present verse is indicacive of che manner in which the Saurrancikas,
while trying to explain causation in terms of "associacioo" of discrere momeotary cntities, wece evencually led co a substantialist concepcion causacioo. Eirher
they had co accept self-causation or remain satisfled with rnere self-re
production, the latter providing no explanation of creacivicy at all. "Dependent
arising, on che conrcary. accounts for creacivity without falling in to the
substancialist trap.

16.

Parfitya yad yad bha11ati tat tac chiinla'f!l svabhiivataf?,

tasmad uJpadyamiinaf!l ca fiinJam utpattir eva ca.

Whatever that comes to be dependently, thac is inherently peaceful.

Therefore, that which is presenrly arisiog as well as arising icself are
peaceful.
!lf.KV(P) p. l ~9- 1 60 : ,W.K ll(V)

p. 6~ .

This singularly imporcanc statemcnt in the prcsent chaptet comes after Vll .-15
where Nagarjuna refused co recognize a form of dependent arising" chat also

T HE EX11MINA110 N Ot THE CONDIT10NBD

169

carries the implication of self-tausation (svata-utpattt). The nocion of depen

dent arising (prafitya bhavati = prafityasamutpada) is simply free from any
idea of sdf- causation involving substance or self nature (s11abliii11ata4 'iita'?!-)
and. in tbat sense, is " inhereotly peaceful." 1t does not con tribute to the notion
of "self1 (atman, svabhava) or "other" (para) and , as a result, does not generate
lusc (riiga) or hacred (dvefa), anachment or aversion . It is the more appropriate
view regarding the present arising (utpadya1nana) as well as arising (utpatt1)
and constitutes an understanding (jnana) or wisdom (prajifii) thac brings abouc
freedom (vimuk11: nirvaf!a). It is indeed not a negation of arising (utpada, utpattt), but simply a non-substantialist way of percciving such phenomena.

17.

Yadi kacid anutpanno bhava!J sa'f!J.vidyate kvacit,

utpadyeta sa kirtt ta1mirJ bhava utpadyate 'sati.

If a cectain non-arisen existent js evidcnt somewhere, theo that would

arise. When sucb a thing does not exist, how can.an exist.ent anse?
M.KV(P) p . 160; M.KV(V) p.66.

Here Naga(juna, seems to be sctting up the Sarvascivada theory ef ident ity

against the Sautrintika view of non-idencicy. The identity theory of causation
recognizes an ~nticy (bhava), permaoent and cternal, lying concealed
somewhere (kvacit), and therefore not yet arisen (anutpanna). If sucha thing
wece to exist, theo it could be said to arise. However, if sucha thing were to be
non-existent (lasmin asat1). how can one say chat a ching or eotity arises? This
is, once again, a criticism of the Sauttantika tbeory of the arising of a nonexisting enrity (abhutv?i bh?i11a utp?ida, Akb p. 77 ; sec also Kalupahana ,
Causa/ity , p. 151). Hete, Nigarjuna is maintaining that the Sarvastivada
cheory of causacion, though excessively metaphysical, may be inteligiblc in
some w... but that the Sautrancika theory of the arisiog of a non-existent entity
makes no sense ac all.

18.

Utpadyam4nam utpiido yadi cotpadayaty ayaf?l,

utpadayet tam u.tpadam utp'iida/f katamal; puna~.

170

J).rn PHILOSOPHY OF Tf!E MIDDLE

W /\Y

lf arising we(C to produce this prcsent arising, which arising would again
produce tbat arising of that arising?

MKV(P) p.161; MKV(P) p.66.

lnfinite Iegress becomes a n~essary corollary of the Saumntika theory of

momentary events. As mencioned eaclier, the Sauttantikas were compelJed. to
assume the notion of potentiality (fa/dz; hi/a) in order to avoid such infinite
regrcss. Nagarjuna was not ready to accept tbat solution.

19.

Anya utpadayaty enaf!t yady utpido 'navasthitif?,

athanutpacla utpannaf? sarvam uJpadyate tr;th'ii,

If this arising were to produce another, arising would turn out to be in-
finite regression. H the non-airising IS arisen, then i.t will produce
everything in tbis manner.
MKV(P) p.162;

MKv(fl?

p.67.

Not only does the Sautrantika theocy lead to infinite regress (anavastha), it also
contributes to the chaotic view that anything can come ~ut of anything, whicb
is the opposite of the Sa..cvastivada v.iew that notUng comes out of nothing._
Causa! unifocmity bas no place whacsoever in che Saucta.n tika scheme of
things. This is another point of comparisoa between the .Sauttaatika and Humean views of causatioa.

20.

Sata ca

tavad utpattir asataf ca nri yujyate,

na sata c'iisata ceti purvam evopapadita?JJ.

As such, neither the arisi.ng ofan existcnt noc the arising ofa non-existent
is pcopec. .Even se is the arising of that whlch is both existent and nonexistcnt,
this has been pteviously explained.

and

MKV(P) p . 162; MKV(V) p.67.

THE EXAMlNATION OF TI-IE CONDITIONED

171

Neither the identity thcocy of causation nor the oon-identity theocy nor cven a
cornbinacioo of these two will appropriatcly account for all t he issucs rclating to
causation. This idea continucs to be emphasizcd by Nagarjuna.

21 .

Nirudhyamiinasyo1p111Jir na bhavasyopapadyate,
ya c'iinir11dhyamiinas tu sa bh'ii110 nopopadyate.

The asing ofan existent tbat is ceasing is not appropriate. Whatever cxistcnt that is non-arising, that cxistent too is not appropriate.
MKV(P) p.163; MKV('1 p.67.

The theory discussed at VU.20 was the arising of a non-existent entity. The
discussion oow moves oo to the question rcgarding the arisiog of an event that
is ccasing. The Abhidharma interpreters recognized two typcs of causes (helu)
that provide a foundation for the causal efficacy of an cvent that is ceasing
(nirudhyamane k'iirilraf!!- dvau hefii kuruta4, AK ii.63; AKB p.100; also see
definition of k'iiritra at AD 321; Adv p.281). Nagarjuna is reluctant to acccpt
even the fact that an existebt that is on ics way to ceasing can arise. In other
words, cessation and arising cannot be describcd as evcnts taking place in relacion to the same entity. Or moce specifically, a changing substance is unacceptablc to Nagarjuna. Similarly, one cannot speak of an cntity that is not ceasing

(11nirudhy11miina) .

22.

Na sthita-bh1i11a.J li/thaty as1hi1a.bha110 na tifthatz~

na tifthati tirthamiina4 ko 'nutpanna ca f#thati.

An existent that has endured is not stationary, nor is an existent t.h at has
not endurcd. The presently cnduring is not stationary. What non-arisen
can stay?
A1KV(P) p.164; MKV(V) p.68.

172

THE PHILOSOPHY OP THE MIDDLE WAY

Tbc Sarvasdivadins (especiaJJy the Vaibha~ikas) argued thac un less "stasis"

(sthit1) of an encity (dharma) is not recognized , ie causa efficacy also cannot be
admirred (Adv p. 109). This is conrrary co the view (referred co in VTI.21) expressed by the Sautrantikas, namely, chat an entity, wheo it is ceasiog, can
generace causaJ efficiency. Thus, while the Sautraocikas rcfused to recognize a
moment of scasis fearing chat ie would lead co a belief in substance, the Sarvastivadiris insisted upon admittiog such a static momenc.
As such, the Sarvastivadins can maintain that a static moment cndures
(11hita-bhavaf? tiffhatt) , primariy because they assumed that scasis has its own
oacure (111abha11a). lf there was no such narure, chen phenomena could not eodure. Herc again, we have the discinccion between a phenomenon and its static
narure . Whac Nagarjuna is denying hece is not chat a phenomenon can endure,
but chat a static nature can endure (asthita-bhavaf? tiffhatt)? This woud appear
to be se f- concradictory to Nagarjuna. To say chat "presently enduring is enduring" {tt/(hamanaf? tt/ fhall) woud be taucologicaJ.
All the above mencioned views would pertain to something chat has lieady
arisen. Finally, Nagarjuna insists chat we cannot speak of a non-arisen entity as
having scasis.

23.

Sthitir nirudhyamiinasya na bhavaJyopapadyate,

yaf canirudhyamiinas tu sa bhavo nopapadyate.

Duration of an ~tent that is ceasing is noc appropriate. Whatever existenc chat is non-ceasing is also not .appropriate.
MKV(P) p.164; MKV(V} p.68.

While the Sautrancikas (a.S mentioned earlier) recogoized causaJ efficiency in a

phcnomcnon at the moment of its arising (utpada). and as such they did not
have co accept a static moment, che Sarvascivadins argued chat without a static
moment a phenomenon caaaot generate any causal efficiency (yadi hi dharmasya sthitir na sya11 tasyatmanya11a1thria1ya hetvakhjiif? faktiprabha11a11iefO
na.Jyat, Ad11 p. 105). A moment of stasis is ro be foJlowed by decay (jara) and
destruction (11yaya). If the Sarvascivadins were noc happy abour cecognizing
causa! efficieocy of a pheaomeaon at che moment of its ar-ising, che moment
being such a minute instant of time. they will be compelled to admic such
causaJ efficiency in a scacie moment as ie begins to disappear (ntrudhyamana).
Nagarjuna's argument, here is, thereforc , direcced agains che Sarvascjvadins

T HE EXAMfNAilON OF THE ( ONDITIONED

173

wheo he says: "The duration of an existent that is presently ceasing is not ap

propriate." H owever, someone may assume that Na~rjuna's C(idcism of the
Sarvativada theory of scasis would mean that he is compelled co accept che contrary view, oamely, t hat therc is a phcnomenon chat is not ceasmg
(anirudhyamana) . This lattet view he rejects in che second lioe.

24. jarii-mara11a-dharrrtef,u rarva-bha11e1u sarvadii,

t;,,thanti katame bhava ye jafii-1nara1Jan,z vina.

When all existents are always of the nature of decay and death, which existents that are without decay and death can stay?
MKV(P) p"l65; MKV(V) p .68.

This is a simple Iejection of any metaphysical idea pettain.ing to the real stasis
(sthitt) of phenomena. This rejection is based upon the empirical notions of
decay and death (jara-maraf!4) emphasized by the Buddha himself.
The interpretets of the Abhidharma seem to have experienced difficulty in
reconciling the Buddha's description of the "dispositionally cooditioned"
(san,iskrta) as having three characceristics (tn11i lak1a11am) (A 1.152; TsenB
12. 5 [Taisho 2.607c)) with chei.r owo theories of momeotarioess. While the Sarvastivadins recognized a fourth moment (caturtham atra vaktavya'f(l syat, Akb
p .75), splitting up "change of what has Iemained'' (thitassa aflfJathatta) into
ewo moments as sc.asis (sthiti) and decay (anyathiitva, jara), the Saucriinrikas admitced ooly two moments rejectiog both stasis and decay. The Theraviidins accepted a thcocy of three moments , once again omitting decay.
The need for recg.nizing stasis o.n the part of the Sarvastivadins and its
denial by the Sautrantikas is discussed at length by Vasubandhu (Akb
pp .75 76). Both schooJs assumed (and this is the case wich the later
Theravadins coo) chat the Buddha's definition of the "dispositiooally conditioned'' in tenns of three Gharacceristics (as arising, decay or change of what has
remained, and ~easing) is for the sake' of the unenlightened (11ineyajan'iirthaf?').
They argued chat it is for this reasoo chat the Buddba utilized the rerm ''appeats" (pannayati, prajnayate) when speaking of these three char.cceristics.
However, chis is not t he case with the Abhidharma theory of moments (na tu

kfa1'!1JJya).
The incerpreters of the Abhidha.rrna (hardly realizing chat the Abhidharma
did not have a theory of momencs) wece making a distinccioo between the

174

D-ffi Pl-ifLOSOPHY OF THE MIDDLE W AY

"discourses" (1utra) and Abhidharma, comparabe to the distinction made in

the Saddharmapu11 f1Jnka between the Agama and Mahayana discourses. They
assumed that the Abhidharma aoaysis (i.e. , in terms of moments) is for those
who realy understand the "meaning" (arthajiia) , not for those who merely go
after the "texts" (granthajfla) . Thus carne to be established the definition of
Abhidharma as concaining " taintless wisdom" (prafl'ii 'ma/ii J'iinucara 'bhidharmaf?, AK i.21).
Nigarjuoa soems to have remained undaunted by such claims on the part of
the Buddiist metaphysicians. He was probably aware that all three
metaphysical views presented by the Sarvasrivadans, the Sautrantikas and the
!acer Theravadins (the last was pcobably not knowa to bim) were incompatible
with the Buddba's own conception of change explaioed in terms' of decay and
death (jafii-maraf!a). Thcrefore, he perceives no existents (bh1iva) that arc free
from decay and death.

25 .

Sthityanyay1i Jthitef? sthanaf?l tayaiva ca na yujyate,

utpadasya yathotpado natmanii na paratmana.

The endurance of an cnduring tbing bascd. on the eodurance of jrself or

of another is not pcopet. lt is likc the absencc of arising of arising, cither
from itself or from anothe.r.
MKV(P) p. 165; MKV(V? p.68.

lf stasis (sthit1) we.re to be a distinct event, theo the metaphysicians who needed
to explaio such an eveot in terms of causality will have to maintain that it could
occur either depending upon itself ( = self-causation, svatotpatti) or based upon
another (externaJ causation, para/a JJtpattt) . Both are not appropriate. This is
simiar to the criricism made of arising ac VII .18- 19.

26.

Nirudhyate naniruddhaf?l na niruddhaf?l nirudhyate,

lathapi nirudhyamana11J kim aj1ita1J'} nirudhyate.

T~t which

has not ccased does not cease. That which has ceased also does

THE ExAMINATION OF THE CONDITIONED

175

not cease. Even so is that which is ceasing. Is it the urtborn that ceases?
MKV(P} p.i67; M.KV(V) p.69.

After analysing the metaphysical concepts of arising (utp'iida) and stasis (1thitt),
Nagarjuna settles down to an examination of the problem of cessacion or
destruction (nirodha, vyaya). 1his criticism follows the method adopted at II. l, ex
cepting the fina question: "Is it the unborn. that ceases?" lf cessation were to be
understood in the sense of momentary cessation, it cannot be explained.
However, if one wece to accept a permanent and eternal substance (111tJbha11a)
which would, at the same time, be unborn (a1izta), as the Sarvastivadins did ,
then that cenainly cancot cease to exist.

27 ,

Sthita.rya tavad bha11asya nirodho nopapadyate,

nasthita.ryapi bhava.rya nirodha upapadyate.

The cessation of an existent that has endured is not appropciate. The

cessation of an existent that has not endured is also not appropriate.
MKV(P) p. 168; MKV(V) p. 70.

As mentioned earlier, the Buddha's discourse p ertains to "change of what has

remaioed'' (!hitassa afJathatta), where ''what has remained" (!hita) would not
imply something that has remained static without any change , but rather "what
has become" (bhUta, bhutva). What has become can then cease to exist. Tuus,
"change of what has remained'' (namely. something that has undergone a process of change or uansfocmation-vzParif!ama) can come.to an end. Hence, itis
meaningful to speak of cessation (vaya) after "change of what has remained"
( fhilassa afJfJathatta),
However, this is not the case with stasis (sthitt) as explained by the
metaphysicians, primae.iy because. such stasis was disringuished from change
(anyathatva). As such , stasis implied that sornething remains static, without
change, for a while and then without any cause or reason it would suddeoly
start changing or would cease to exist. This is the implicaton of the rheory
refe.rred to in the first line which is not acceptabe co Nagarjuna.
,
The second line implics the equally metaphysical and unacceptable view of
the Sautrantikas who argued that an eveot ceases immediately upon arising

176

Tur! PHILOSOPHY OF THB MIDDLE WAY

without remaining cven for one moment (k/a11ikasya hi dharmasya 11ina sthitya
11yayo bhavet, Akb p. 77).

Tayai11a11aslhayii11aJtha na hi sai11a nirudhyate,

anyajii11aJ/haya11aJtha na canyai11a nirudhyate.

28.

lndecd, a certain state [of cxistence] does not cease Crom a state identical
with its owo. Nor docs a state [of existence] cease from another state dif.
ferent &om its own.
MKV(P) p .169; MKV(V) p .70.

The principles of self-causation (siJata-uljJatli) and external causation (paratautpattt), rejected in Chapter I, may have to be adopted in order co explain cessacion itself (nirodha), if the latter were to be uoderstood in the way the Sar
vascivadins and the Saucrancikas conceived of it. Eithcr ccssacion will havc to
occur on its owo, or on .the basis ofanother. Such discussions were rampant
ducing the scholastic period in ~n philosophy ( especially with the domination of the doctrine of mome,N;_armess) whcn it was argued as to whether
dcstruction is inherent in birth or whether,~ is brought about by extcrnal causcs
or conditions (sec Ad11 pp. 106-108). Nigarjuna's rejection applies to both
VICWS.

29.

Yadaiva sarva-dharniii111im utpado nopapadyate,

tadai11a rarva-dharm7h!i'f' nirodho nopapadyate..

Indccd, when the arising of all things is not appropriatc, then the ccssation of all things is also not appropriate.
MKV(P) p .169; MKV(V) p.70.

Note chat the term sart1a (all) is uscd as an adjective of dharma, but not utpida
(arising). This means thac Nagarjuna is not rejecting all forms of arising (sariiautpada) as being unsatisfactory. Ratber, he is critical of utilizing the concepcion

T HE EXhMINhTION OF IBE CONDITIONED

177

of acising-(111pada), discussed previously, to any or all pbenomena (dharma) . If

arising in a metaphysical se:ise cannot be applied to all phenome~a, then
cessaton (nt'rodha) too, similarly conceived, is inapplicablc.

30. Sataf ca tavad bha11asya nirodho nopapadyate,

ekatve na hi bliiivaf ca nabhavaf copapadyafe.

Furthermoret the ccssation of a real existent is not appropriate. lndecd, in

the context of identity, neither existence nor non-existence is appropriate.

31.

Asato 'pi na bhavasya nirodha upapadyate,

na d11iliyasya firasaf? chedanaf?' vidyaJe yatha.

The cessation of an unreal existent is also not appropriate, just as a second

beheading [of a person] is not evident.
MKV(P) pp. 169- 170; MKV(V) pp.70- 71.

Nagarjuoa's criticism so far has been confined to arising, stasis and ceasing
metaphysically conceived as events in themselvcs. In the present contexts, he
retums to the criticism of the ccssacion of real existcnts (satal; bha11asya), similar
to the criticism of the real cxistents in Chaptcr I. Neither existence or noncxistence ,- nor a combination of both. arc acceptablc to him.

32.

Na svatmana nirodho 'sli nirodho na pariitmana,

utpadasya yathotpado na~mana na para1mani.

There is no cessarion by i_tself or by anothcr entity, just as the arising of

arising is neither by itsclf nor by another.
MKV(P) p.lvl ; MKV(V) p.71.

178

TH E PH TLOSOPHY Of THE MIDDLE

\Y/ A Y

This is the fioaJ criticism of che identity and non-idencity cheories as applied to
the idea of cessation (nirodha). lt is comparable to the criticism of arising (utp'iida) presenced at
l 2.

vn.

33 .

U1p'iida-1thiti-bhaganiim asUlher

n'iisti Jaf!llkrtaf!J,
Sa'f!Jskrtasyaprasiddhau ca kathaf!J setsyaty asa1'(lskrtaf!J.

With the non-establishment of arising, duration and destruction, the

conditioncd docs not ex.ist. With the non-establishment of the conditioncd, how could thece be the uoconditioned?

MKV(P) p.1 76; MKV(V) p.73.

Arising (utpiida), stasis (s1hiti), and destrucdoo (bhanga), as explained by the

metaphysicians, would not establish their conception of che "conditioned"
(Jaf!llkrta). If the "conditioned", so conceived, is not cstablished , indeed chere
caooot be a similarly formulated conception of the "unconditioned"

(asaf!llkrta).
One very significant fact chat caonot escape a careful scrutiny of the contents
of this entire chapter is that , while Nagarjuna has spccifically criticizcd che
tuee kinds of events (arising, stasis, and ceasing), he has made oo criticism
whatsoever of decay (jara). In fact, at VII.24 , decay and death (jari-mararta)
were used as arguments for the rejection of stasis (1thit1). In short, the Buddha's
owo conceptions of arising (uppada), ceasiog (vaya, nirodha), change of what
has remained ([hita11a afifialhatta), decay (jar'ii), impermanence (anicattii) have
been Jeft intact. These have no room for a substantial entity (an atman, a
svabh'liva) and arc indeed compatible with the Buddha's famous doctrine of
noo-substantiaJity (anatta). There could be no reason why a philosopher of
Nagarjuna's caJibre could not distinguish the empiricaJ nature of the Buddha's
anaJysis of the "characteristics of the disposicionaUy conditioned" (sal!khatassa
sankhata-lakkha1{iint) from the metaphysical character of the ideas expressed
by the intcrpreters of the Abhidharma.

34.

Yatha mayii yatha 111apno gandharva-nagaraf!J yatha,

tathotp'iidas tath'ii 11hanaf(J tathi bhaga udiihrtaf(J.

T HE EXAMINAT!ON OF THE C OND!'rIONED

179

As an illusion, a dream, a city of the gandharvas, so have arising, en-

durance and destruction l:>een exemplified.

M.KV(P) p. 1'77 ; M.KV(V) p.73.

If the analys.is of the conteots of the present chapter is correct, then there
could be no question that the ilusory character as exemplified by "dream" and
"the city of the gandharoas'' is appplicabJe only to the metaphysical ideas of
arising, stasis and ceasing as presented by the interpreters of the Abhidharma,
and nt to any one of the conceprs of arising, ceasing, decay. change, and impetmanence formulated by the Buddha and accepted by Nagarjuna hirnself.
The Buddha utilized the similies of foam (phef!a) , bubble (bubbula) ,
mirage (manc1), trunk of a plantain tree (kadalt) , and illusion (maya) in order
w illustrate the non-substaotiality of the five aggregates (S 3.142). Nagarjuna
is here using comparable simi!ies for the same purpose . A careless applicatoo of
these similies to explain all forms of ideas, whether substantialist or nonsubstantialist, has given rise to enormous misunderstanclings relating to the
Buddhist doctrine.

CHAPTER
EIGHT

Examination of Action and the Agent

(Karma-karaka--partk.[a)

1.

Sadbhuta4 kiirakal? karma sadbhutarp na karoty ayaf!J,

k'iirtfko napy a.radbhutaJ? karmasadbhutam thate.
'

This really existent agent does not perform a realy existent action.
Neither is it intended that a really non-existent agent performs a really
'
,
non-existent action.
MKV(PJ p. 180; MKV(V) p.75.

The examination of the "dispositiona)Jy conditioned" (sa'f!Mktta) naturally led

Nigarjuna to another major philosophical problem that has been the subject of
much misunderstanding among the Buddhisrs as well as the non-Buddhists,
namcly, the _doctrine of karma.
ln the early discourses, karma and saf!Jskara (and, therefore, sa?!Jskrta) are
mutually related. Wbile karmas arc said to form saf!Jsk'iiras (see M 2.'121),
saf!Js/iara.r themselves arc decerminants of karma. The Buddha who denied an
eteroal soul or self (atman) was often confrooted by skeptics who raised the
qucstion as to how karras performed by a " no-self" can affect a person (anaflakatiini kammani ka'!' (Aatarp) atfina?!J phusirsanti, M 3.119;S 3.103). This is
not, howcver, the problem of mora responsibiity (karma--phala) , which
Nigarjuoa takes up later in Chapcer XVII. The present problem relates mostly
to the oature of an action and an agent (karma-karaka). When this lattec question was raised in the discourses, very often it is said that there is no substantial
agent who is the author of actions. What is generally understood as a substantial being (satta) or person (puggala) is nothi.og moce than a "lump of disposicions" (sakhiira-puija, S 1.134), which is another name for the five aggregates
(khandha), cach recciving irs form conditioned by the sakhiita.r (S 3.87; Tsa
2 .14 (Tairho 2.llc]). Neither the "lump of dispositions" nor the aggregates
were considered to be substantial, that is, having a reality of their own. The
"clispositions" (sakhara) , which arc acquired by the person and, therefore, im
perinanent, are also responsible for condiciooing that personali ty. Thus,
1RO

THE Ex.AMINA TION Of AcnoN AND TiiE AGENT

181

according to early Buddhism, neither the person nor die aggregate~ (one of
whicb is disposition, sankh'iira) have any substaotial reality. Io the fitst place,
the dispositions chemselves arc impermancnt (11ayadhamm4 sank~iira, D
2.156) and, secondly, all phenomena, includig all persons or aggregates, arc
non-substantial (sabbe dhammii anrJttii, M 2.228; S 3.133; 4.401; A 1.286;
Tsa 10.7 [Tairho 2.668)).
Yet, when the canonical Abhidharma tex:ts rejected a real self or soul (atta)
and listed the various physical and psychological factors that go to constitute
the human personali ty, the intcrpreters of these physical and psychological factors ttansgressed the limits of speculatioo and admitted the real existcnce of
these various factors . Thus carne to be the theory of the substantialiiy of
elemenrs propounded by the Sarvastivadins.
Thcreforc, in his attempt to establish the non-substantiaity of all elements
(dharma-nairatmya) , Nagarjuna was compelled to examine the concepts of action (karma) as well as the agent of such action (karaka) in the prescnt section
of his treatise (leaving the question regarding mora responsiblity for a. detailed
discussion at a later and more appropriate time). As explained previously,
Nagarjuna tealized chat the concepts of substantial existence (sat, astitva) as
well as a.bsolute noo-existcnce (asat, na1tit11a) werc two cxuemcs to be avoided
in any discussioo of action and agent.

2.

Sadbhutasya kn'yii nasli karma ca syad akartrkam,

sadbhula.rya krijii nasti karta ca syiid akarmakaf?.

A rcally existent cntity bas no' activity. Therefore, acton would be without
an agent. A really existent entity has no acriviry. Thcreforc, cvcn an agent
would be without action.
MKV(P) p.181 ; MKV(V) p. 75.

This is an cxtremely interesring analysis of the implicatioru as welJ as the conscqueoces of admfrring a rcally existing entity (sadbhuJa). a soul or a self (atman,
pudgala) or evcn a substance (!vabha11a). The concept of a self (atman) as envisaged in the substantialist traditions, like the Upanifads, is one of pure cnti
ty, permancot and eternal, unaffcctcd by the changes taking place in the
phenomenal world . lt is beyond all forms of duality and multiplicity. The con
cept of substance, even though rarely dcfined as a " pure enci ty," yet pactakes of
all othcr characteristics , namely , permanence, eternalicy, noo-duility, and

182

THE PHJLOSOPHY OF THE MIDDLE W AY

non-multiplicity. This means tha.t any activity, any change in quality, in quancity, in character, or in frm would b merely supe.rficial. The real or the essential is eternal. Such implications of the Sarvastivada theory of substance
(svabhliva) wece noted ear1ier (sec Chapter I). If the cause and effect were to be
idc:ncical in essence, then their diffe.reoce becomes superficial. A cause becomes
un-productive and wouJd be like a tree stump that .temains moti'onless (vanjha
ku{a!fha, D 1.14, 56; S 3.211 ; M 1.517; 11vicalit11-nity11tv11, sec Ka1upahana,
Causality , p. 28).
Hence Nagarjuna's statement: "A really existing encity is withoutactivity."
This assercion leads to two rather disastrous conseql.lences, especially for the
doctrine of karma; either an action (lu;muz) will be rendered agent-less
(a-kartru) or an agent (kartr) would be action-less (t1-karmakiz).

' 3. K11roti yady asadbhuto 'sadbhuta'f!J karma karakal/,

t1hetuka'f!J bhavet karota karfii ciihetuko bhavet.

lf non-existent age.nt w~e to perfor.m a nonexistent action, the action
would be without a cause, and the agent too would be without a cause.

'

MKV(P) p. 182; MKV(V) pp.75-76.

.

If one wece to accept the view which is contrary t o the one mentioned at VIU. 2,
that is, a non-existeot entity performing a non-existent action, then both agent
(kartr) and action (ka"na) would be rendered cause-less (a-hetuka).
VIII.2-3 thus turo out to be a elear warning against the universal and indiscriminate application of the examples of "illusion'' (maya) , "dream'' (sv(Zpna) and the "city of the gandbarvas" (gndharva-nagara) , espedally in the mattet of explaining "emptiness" (funyatli) at VIi. 34 and elsewhere.

4. Hetiiv asati liiirya'f!J. ca kliraf1af!l ca na vidyale,

tad abhlitie knyli karta karaf!a'f!J ca na vidyate.

'

When a cause does not xis(, both the effect and the sufficient condition

T HE EXAMlNATION OF ACTION AND THE AGBNT

18}

arc not evident. When thcse are ooo-existent, activity, agent and
performance of accion are also not cvideot.
MKV(P) p. 182; MKV(V) p. 76.

Furtbcr impications of the denia of a cause at Vlll.3 are highlighted herc. Not
only would it negate an effecr (karya) or a sufficient condition (k7ira1Ja), it
wouJd also lead to the abandoning of activity, agent as well as action, nooe of
which is accepcable to Nagarjuna.

5.

Dhamiiidharmau na vidyele knyadiniim aJaf(lbhave,

dharme ciisaty adharme ca phalf(l tajjam na vidyale.

With the non-occurrence of activity, etc., good and bad arc also not evident. Wben both good and bad do not exist, a fruit arising from these
would also not be evidcnt.
MKV(P) p . 183; MKV( V) pp.76-77 .

Inada's rendering of this verse clcarly indicates his faitbful adhereoce to the
transcendentalist interpretation of Nagarjuna offered by previous interpreters
like Stcherbatsky, Murti, Conze, and most of theJapancse scholars. In spite of
KumarajTva's very elear rendering into Chinese, lnada ttanslates dharma and
adharma as factors and non-facrors respecrively (p. 73). While ie is crue that the
term dharma is used in the Buddhist texts, both in an ontoogica sense (referring to "pheoomcna") and in amore ethical sense (meaning "good") , cbere is no
evideoce at aU that the negacive term ,a.dharma was ever used in the former
sense. A careful examinatioo of the contexts in which it occurs provides sufficient evidence that the term meant "bad" and, hence synonymous wich
akusala. Furthermore, of all che terms used co refer to an effecc, the term phala
occurring in this verse is invariably used in the sense of "fruit ," baviog a mora
connotatton.
The preseot verse. therefore, provides unmiscakable evidcnce that Nagarjuna was upholding che mora philosophy advocated by the Buddha in his first
discourse-the Dhammacappa111111an1rs11tla-wbich lays down a middle pach

184

THE PHILOSOPHY OF THE MIDDLE WAY

bctwccn the two cxtremcs , which in ics rurn is based upon the philosophica
middle position prcsented in the "DiscoMrse Io Kiityayana."
The belief in substancial cxistence or eternalisrn (a11hita, sassata) leads to
self-denial (at111-kilamathanuyoga) and the idea of nihilistic non-existence or
annihilationism (n'atthita, uccheda) coQtributes to self-indulgencc (kiimasukhallikiinuyoga) (sec Iotroduction). Verses VIIl .3-4 refer to che substantialist
and annihilationist views of karma, both of which lead to the abandoning of a
mora life (iocluding mora responsibility) as understood by the Buddha. Having spokcn of artha (effect, fruit , consequencc) in relacion to pratyaya (cause,
condition) in Chapter I, Nagarjuna is herc addressingrumself to the question
regarding the :fruit of action" (karma-phala), a detailcd treatment of this problem being reserved for a lacer occasion (Chap ter XVII). In the prescnt chapcer,
Nagarjuna's iotcncion is to cxplaio the non-substanciaity of the "agent'' and
the ''action," rather chan settlc the qucstion of mora] responsiblity. However,
he could not help rcferring to the "fruics of action," primarily bccause the
mctaphysical nocions of "agent" and "action" contributed to thcir vcry denial.

6. Phale 'sati na mokfiiya na s11argayopapadyale,

marga!? sarva-knya11am. ca nairarthak.yam, prasajyate.

When the fruit docs not cxist, the path of rcleasc or of hcavcn is not ap-propatc. This would imply the futility of all activity.
MKV(P) pp.183-184; MKV(V) p . 77 .

Not only is Nagarjuna interested in the :ultimate fruit" (paramartha), i.e.,

freedom from sufferiog (mo/qa), he is also concerned with the fruits (artha,
phala) chat actions can generate as human beings proceed aong the mora patb
(marga) gradualJy lcading to thac "ultimate fruit." Hence his interest in
"heaven" (s11arga) , so often discussed in the "discourses" as the "fruics" enjoyed
by the "wayfa.rer." As far as Nagarjuna is concerne<l, human actions arc
rendered "fruitless" or "meaningless" (nirartha), if one were co adhere ro either
a theory of eternalism or of annihiacionisn1 .

7.

Kiirakaf? sad-asad-bhutal; sad asat kurute na tat,

par1Upara-viruddha~ hi saccasacc-aikatal? kutaf?.

THE EXAMINATION Of A cnoN AND THE AGENT

185

An agent who is both existent and qn-exiscent does not perform an action that is both existent and non-existcnt, for they arc self-contndictory.
Wherc can cxistence and oon-cxistencc co-exist?
MKV(P) p. 185; MKV(V) p. 77 .

After criticizing the subscancfa)jst as well as the annihilationist views rclating

to agent and action, Nigarjuna examines the theory that atcempcs. to combine
them. This follows the criticism at I.13 . For Nigarjuna, contradictory
predicates such as existence (sat) and non-cxistence (a.rat) cannot be applied to
the same thing.

8.

Sala ca knyate niisan niisata kriyate ca sat,

kartra sarve prasajyante dofiis tatra ta eva ht'.

A non-existent action is not performed by a presently existing agent. Nor

is an cxistcnt action performed by a prescndy non-existent agent. Indeed,
if that wece to be the c.ue, all errots relating to the ageots [mentioned
earlier] would follow .
MKV(P) p .185; MKV(V) p .78.

Existence (sat) and non-ex.istence (asat} applied to the agent and action will
producc some other alternacive rheories than those mentioned earlier, e.g. :
i.

u.

A presently exiscenr agent (santa) performs a non-existent

(asat) action, and
a presencly non-existent agent (a.ranta) performs an existent
(sat) action.

Herc, the agent is described by means of a present patticiple (santa}.

9. Niisadbhutaf!J na stJdbhi1af/ sad-a.rad-bhuta1n eva va,

karoti k.iirakal? karma puN1ok11lir eva hetubhil/.

186

TuE PHllOSOPHY OF THE M IDDLE

W AY

For rcasoos stated above, an agent who has come co be existent does not
pcrform an action thac is non-existent or botb existcnt and non-existent.

10.

Niisadbhuto 'pi sadbhulaf?' sad-asad-bhutam eva va,

karoli k'iiraka4 karma.purvoktair e11a hetubhil,J.

For reasons -stated above, an agent who has come tobe non-existcnt does
not pcrform an action chat is exisrcnt or both cxistent and non-existent.

11.

Karoti 1ad-a1ad-bhuto na san nasac ca kiirakal,,

karma tal tu 11ifiinty'iit purvoktair eva hetubhif?.

An agent that has come to be botb existcnt and non-existent does not
pctform an action that exists and docs not exisc. This coo should be
undcmood in terms of the rcasons adduced above.
MKV(P) pp. 186-1.8 7; MKV(V) p.78-79.

W hHe VIII.8 utiized the p resent paniciple to refer to the agen t, the p reseoc
verses ernploy the past participle: (i) an agent who has come ro be existent(1adbhuta); (ii) an agent who has com e co be non-existenc (a.radbhula), and (iii) an
agent who has come to be bo~h existent and non-existen t (sadtUadbhuta) .
VIII.10 is not found in the Ch inese version.

12.

Prafitya karakal? karma taf(t pralitya ca karakaf{l,

karma p Mvarfale nanyal paiy'iimal/ siddhi-k'lrat!a'f!I>:

An agent proceeds d epending upon action and action proceeds depcncJing upon the agent. We do not pccceive any other way of establishing
(them] .
MKV(P) p.189; MKV(TI) p.79

THf

EXMvUNATION OF ACTION AND 1HE AGENT

187

After criticizing the metaphysical ioterpretations of agent (karaka) and action

(karma), Nagarjuna gives a positivc description of both in terms of the principle of "dependent arisint' (prafityasamutpada) . Indeed, a mote posirive assertion suh as, "We do oot perceive any other way of establishing [them]"
(ninyat siddhi-karaf!af!' payamal/), is rardy met with in the Kiirikiis.

13.

Evaf!t vidyad upadiinaf?Z vyutsargad iti karmaf!.al/,

k.arttJ ca k4rma-k.artrbhya1~ fe1a11 bh'iiv'iin vibhavayet.

Following this method of the rejection of agent and action, one should
understand grasping. The remaining existentS should be c.ritically examined in terms of the concepts of action and agent.
MKV(P) pp.189-190: MKV(V) p.BO.

Nagarjuna began the chaptcr with the problem of grasping (upadiin'ii). lf he

wece to accept the substantialisr notions of agent and action, he could not explain grasping. During the course of this chapter, he was able to anaJyse the
metaphysical (and, therefore, unacceptablc) implications of the substaatialist
views such as the deniaJ of mora tesponsibility and freedom. The abandoning
(vyulsarga) refers to the giving up of such metaphysical views. After giving up
such metaphysical views, he presented an aJccrnacive view, oamely , dependent
arisiog of both agent and action (VIIl.12). Such a view allows for a satisfactory
explanacions of "grasping" (11pidiina). A life with reduced grasping contributes
to worldly fruics such as the attainment of "heavenly" bliss. The complete
elimination of grasping culmioates in perfect freedom (mok1a) or the ultima.te
fruic.
This does not mean tbat Nagarjuna has no conception of an agent or an action or mora responsibility or freedom. It is mcreJy the renunciatioo of wroog
vicws and che adopcion of more appropriate explanations of these phenomena.
Such explanations, Nagarjuna concludes, should be extended to all
phcnomena.

CHAPTER

NINE
Examination of the Pri.o r Entity

(Piirva-pank1a)

l.

Darana-r11vt11{iidini vedanadini Clipy

bha11anli yasyt1 prag ebhyaf? IO

atha,
'sflly elee 11adanly utll.

"For whomsoever there eicists seeing, heacing, etc., and feeling, etc., he
exists prior to these." So do some declare. -

2.

Katharti hy avidyamiinasya dllr:ianidi bhavi.!yati,

bi11<1Iya tasmat prag ebhyaf? IO 'sti blfiivo t1yavmthitaf?.

How can there be sceing, etc. of an existent who is not cvident?

Thcrcforc, it is detetrnined that, prior t(l .these things, such an existent is.
MKV(P)

p. 192: MKV(V) p .81.

While the previous chapter is devoted to the refutation of the Canesian enterprise adopted lacer on by some of the Buddhist metaphysicians, the present
chapcer seems to be ca.king up specifically the Kantian project, namely, the
assumption of a primordial condition for all forms of koowledge , induding the
cogito. Heoce the questioo regardiog prior existcnce (purva) .
He could not have beeo unaware of the Buddha's attitude toward the problem of the past (pubbanta). For a r~dicaJ cmpiricist like the Buddha,
knowJedgc of the past (afilllf!J.IC n'iif!a) is as important as any other knowledgc
(D 3.275) . Indeed, knowledge of ''dependent arising" (pa;kcaiamuppada) is
invariably based upon such knowledge. Yet, when the pursuit of that
knowledge is attemptcd beyond its limi~. that is, w hen one tries to achieve absolute certainty w.ith regard to such knowledge, one ends up in specuations
regarding the past (pubbanfiinudittht) rathcr than knowledge (n'ii'!a). The
Brahmajiila-suttllnta refers to a whole hoot of mctaphysicaJ views presented by
188

189

EXAMINATION OF THE PR10R ENTITY

thosc who wece involved in such specuations (D 1.13 ff. ,pubbanta-kappik4

purriiinta-kalpikii).
Nagarjuna was thus convinced that the concepts of the cogito, the atman oc
s1111bh?i11a wcte fotmulated by the metaphysicians who wece actempc~ng to know
the past with ccrtainty. No ocher expl~ation would satisfy their yearning for
certainty regarding the past. Pec.tnancnt self, cternal substance, pcc-existing
cogito, a transcendental unity of appercepcion-tbese could account for
breaks or intcrruptions in. human experience as well as their continuity. The ~m
bolism of the chariotecr (Katha Upani1ad 3.3-6) as wcU as the example .o f the
two birds, one enjoying .the fruit, the other watching (Mu11fhka Upanifad
3.1.1), adoptcd SO cnthusiasticaly by the Brahmanical thinkers, were g1adually
making inroads into Buddhist philosphicaJ thinking. The SammitTyas (if not
identical with, at least relatcd to the Sau~antika school) we.re accused of permitting such a bclief into the Buddhist fod. So says Cand~rkTrci. Yet, Sarvastivada, witb its svabhava-metaphysics is no less culpable of this deviation
tban the Sammitfyas with their pudgala-mctaphysics or even the later
Theravadins with the bhavangametaphysics. "Nagarjuna begins the present.
chapter with a statement, not only of this theory, but also of the rationalizations of thosc who formulatcd such a theory.

wy

3.

Dar:fana-fravat}adibhyo 11edanadibhya eva ca,

ya}/ priig 11yavasthito bhaval/ kena prajnapyate 'tha sal{.

Whatever existent is detennined as existing prior to seeing, hcaring, et~.

and also feeling, etc., by what means is he [it] made known?
MKV(P) p . 193; .MKV(f1 p.81.

As usual, Nagarjuna's first objectioo against positing such an entity that exists
pdor to the experiences such as seeing. hearing, etc. as wdl as feeling, etc., is
epistemologicaJ. This objection should serve as warning against those who
believe chat Nagarjuna recognizes a special intuitive non-sensuous experience
through which_the so-called " ulcirarc reality" (paramiirtha?) is known. In fact,
this question on the pait of Nagarjuna is a elear incJiation of che fact that he
was quite aware of the sort of empiricism advocated by the Buddha , especialy
in his "DiscoUISe on Evetything (Sabba-sutta, S 4.15 ; sec also Kalupahana, "A
Buddhist tract on empiricism," in PEJP' 19 (1969):65-67).

190

THE PHlLOSOPHYOFTHE MIDDLE WAY

4. Vinlipi darlaniidini yadi casau 11ya11asthita4,

amuny api bha111jyan1i 11ina tena na sa1'!1faya4.
If he is determined as existing even without seeing, etc., undoubtedly
even these [i.e. , seeing, etc.] will exist without bim.
MKV(P) p.193; MKV(V) p.81.

After questioning the empirical validity of such an assumption, Nagarjuna is

herc raising a logicaJ objection . lf a prior existing entity can be determined
witbout depending upon its experiences such as seeing, theo ie shouJd also be
logically possible that such experiences as seeing can be determined without a
prior entity. Indeed, this logical conclusion seems to be so suqng that Nagarjuna is willing to use t he term asa~aya ("without doubt") co describe it.

5. Ajyate kenacit kalcit ktfrJcit kenacirl ajyate,

kuta!; ki'l'(lcid vina kafcit ki'l'(lcit kaf!lcid 11ina kutaf?.

Someone is made known by something. Something is m.a de known by

someone. How coul.d there be someone without something and
something without someone?
MKV(P) p.194: MKV(V) p.81

Herc chen is Nagarjuna's method of explaining (vyakhyana) the relationship

bctween substance and attribute . lt is a rdation of dependence and neither che
~bstance nor che atuibute can be understood properly if thcy were to be conceived of as .independent entities, cach having its owo narure.

6.

Sarvebhyo rlarianadibhyaf? kalciJ purvo na vidyate,

ajyaJe daranadinam anyena punar anyadii.

EXAMINATION OF THE PRle>R ENTl'IY

191

Sorneone is not evident prior to all of seeing, etc. Again, on different occasions, one could be made k.nown by things different from _seeing, etc.
'

MKV(P)

p. 194; MKV(V)

p .82.

While rejecting the view that there is an entity prior to all fortns of experiences
such as seeing, Nigarjun.a is, at the same time, uying avoid the other extrt!me of assuming two different encities when the experiences are di.fferent.
Avoidig .a metaphysical nocion of ideocity does not mean that one is invaciably
coinmitted to an equally metaphysica notion of difference. Just as much as
identity can be explained on the basis of an empirica notion of dependence,
Nagarjuaa seems to assen that differttnce can and need to be accounted for oo
/
an empirica basis.

to

7.

Sf)rvebhyo daranac/ibhyo yadi purvo na 11i4yale,

ekaikasmiit katha'!l pur110 daraniidef? sa vidyate.

lf someone existing prior to all of seeing, etc. is not evident, how can
someone existing prior to cach of seeing, etc. be evident.
MKV(P) p.195; MKV(V) p.82.

It seems that hece Nigarjuna is referring to an interestiog-assumption underlying an identity theory. The not1on of self (atman) or substance (s11abhii11a)
would geoerally be presented in order to account for the continuity in a large
number of dissimilar experiences. Taking that premise, Nagarjuna is arguing
that if it is not possible to discover someo.e or entity chat pre-exists all forms of
different experienc.es, then such a person or eotity would not be available even
in the case of individual experiential situations. A momentary cogito would be
as impossible as a permanent and eternal self (atman).

8.

Df'a.ftfi sa e11a sa frota sa eva yadi vedakaf?,

ekaikasmiid bhavet putva?!J e11afr.t cailan na yujyate.

192

THE P HILOSOPHY OF THE MIDDLE WAY

If a seer is, at the same time, a hearcr and fecler, then so1neone would ex-

ist prior to cach one [of the functions]. But this is not proper.
MK.V(P) p . 195 ; MKV(V) pp.8283.

Only if the seer, hearer, and experiencer ace absolutely identical. chen he will
prc-exist cach individual experiential situation . Nigarjuna righty denics any
ruch absolutc identity . He was probably assuming that even in the act of sceing
the same object at differcnt times, there cannot be absolute identity, lee alone
in the acts of secing di.ffLrcnr objccts . The reason is not chat human experiences
or even the objccts of expcricnce change every moment, but that the circumstances under which such cxperiences take place could vary. (For a discussion of the perccprual flux , sce William James, Some Prohlems ofPhilosophy,
Cambridge , Massachusetts : Harvard University Press , 1979, pp.31-32.)

9. Drtl!!finya eva frofjjnyo vedako 'nya!, punar yadi,

sari .ryad dnlffan frota bahutvaf!l citmanif!l bhavel.
lf seer and hearer and fceler arc different, then, when there is a seer,
there also would be a bearer, and as such there woud be a plurality of
selves.
MKV(P) pp. 196 197; MJ..'V(V) p .83 .

If absolute iden tiry is not a possibility , absolute difference also would be im

possible, for in that case within cach stream of experience chere would be
plurality of distinct sdves or entities corresponding to the different experiences.

10.

D(lrfana-fravaadini vedanadini cipy atha,

bhavanti yebhyas te/V e1a bhutepv api na 11idyate.

lt [i.e., the scH] is not evident in the elements from wtich seeing, hearing, etc., and feeling, etc. come to be.
MKV(P) p .179; MKV(V) p.83.

ExAMINATION OF 11-1E PRJ oR

ENnrr

193

In the Chpter V on the "elemeocs" (dhatu), Nigarjuna examioed the theorics

presented by the metaphysiciaos relating to the phy$ical foundations of human
experiences such as seeing, hearing. etc. Even though the Abhidharma
classified the fi~t five sensory organs as well as theiI objects under the category
of derived matter (upadii rupa), there was no indicacioa that this involved the
assertion of any self or subscance in the individual elemencs that go to con
stitute these sensory organs and objects. The interpreters of the Abhidharma, as
reitcrated earlier, were responsible for sucb metaphysics. Nigarjuna is herc going back to the analysis in the discourses as well as ia the Abhidharma where ,
not ooly in the combination of aggregates or elements, but aJso in each in
dividual element, there is oo recognition of such a meraphysicaJ self or
subscance (see aso Chapter IV).
11.

Dar1ana-fravaf!fid'ini vedanad'ini capy atha,

na vidyate ced yasya sa na 111dyanta iminy api.

lf he, co whom bclongs sceing, hcaring, etc. and feeling, etc., is not evidcnt, thcn even thcse would not be cvidcnt.
MKV(P) p . 198; MKV(V) p.84.

Nagarjuna starts with the negatioo of an opponent's view chat there is a prior
cntit:y to which the expcrienccs such as sedng and hearing belongs. The implication of the opponent's view is that the experiences of seeing, etc. arc independent elemenrs appropriaced by an equally independent prior entity. As
such, for Nigarjuoa, it is not merely the prior enricy that is unacceptabJe, but
also the experieoccs themselves as concdved of by the opponent. This,
tbcreforc, is not a simple denial of any and all forms of description of experience. Rather, it is a parcicular type of disccimination resortcd to by the
metaphysician chat is rejectcd .

12 .

Prak ~a yo darianadibhyaf? iii1!1-pralaf?1 cordhvam eva ca,

na vidyate 'sti nasffti nz11r:tt'iis tatra ka/pana.

Whcrcin someonc prior to, simultaneous with or posterior to, seeing, etc.
i.s not cvidcnt, thcrcin thoughts of existencc and non-existeoce are a1so renounccd,
MKV(P) p .199; MKV(V) p.84,

194

THE PMILOSOPHY OF TI-IE MIDDLE WAY

The cxistence of an independent enciry in the cxperientlaJ_prqcess, whether it

be prior to experience , simultaneous with it, or posterior to it, is rejected by
Nigarjuna. This does not mean that he was willing to accept the independent
occurrence of impressions co~ing one afrer another in rapid succession with oo
causaJ connectoins, as the Sautrantikas believed. In these particuJar contexts
(t111ra) , Nigarjuna is ready t.o abandon the thoughts (kalpanii) of cxistence and
non-e.xistence (11Jli nasffh) in the way they were understood by the metaphysi~
c1ans.

CHAPTER

TEN
Examination of Fire and Fuel

(Agm ndhana-pan k.ra)

1.

Yad indhanaf!l sa ced agnir ekat11af!Z kartrkarma11of?,

anyaf ced indhanad agnir indhanad apy rJe bhaveJ.

If fire wcrc to be fucl , thcn thcrc would be idcntity of agent and action. If
firc wcre to be d.iffcrcnt from fucl, thcn 1t would cxist evcn without the
fucl.
MKV(P) p .202; MKV(V) p .86..

as pointed out, was devoted to an examination of the cause or

foundation of the speculations that eveotually led to the belief in me'taphysica
notions such as "self' (atman) or "substance''. (wabha11a). Such specu1ations perta.ined to the "past" (piirva). Why should that anaysis be followed by an examination of che metaphor of "fire and fuel" (agncndhana)? What is its
relevance?
Only a glance at the controvcrsies goiog on during Nagarjuna's day can provi~e justification for this chapter. Once again, Vasubandhu comes to our
assistance. In his Abhidharmakofa-bh'ii.[ya, he was recording a conrroversy
among the Buddbists that was continued for centuries before rum. And that
conuoversy could not have escaped the attcntion of Nagirjuna.
The ninth chapter of the Abhidharmakofa is an appendix (Akb
pp.461-479; translated inro English by T. I. Stcherbatsky, The Soul Theory of
the Buddhists, in the Bul/eling de l'AC11demie des Sciences de Russie,
Petrograd, 1920, reprio red 1970, Bharariya Vidya Pra.kasan, Varanasi). lt deals
with the concroversial views of the YatsTputriyas (considered to be the same as
the Arya-Samm~tTyas , sec Sakv p .699) who propounded the view that there is a
"real person" (1anlaf!l pudgalar('). In the discussion that foll9ws, the
Yatslputriyas arc made to admit tha~. this "person" is neither a substance
(dravya), ike materia form (ritpa), etc., nor a mac designation (.prajnapti), a
merc name like "milk" (kfir11), this lacter being norhing more than an aggregate of subsrances (dravya). Whcn the Sautrancika V25ubandhu pressed his
Chapter

Jl(,

195

196

THE PHI LOSOPHY OF THE MI DDLE WAY

questioos as to whether the person is real or nominal. the Yatslputriyas fell back
upon the metaphor of "fire and fuel" (agnindhana) in order to illustrate their
point of aview (Akb p.461). This metaphor is then discussed at length and
seems to have been the most important means by whic:h the Yatslpuuiyas attempted to justify their concepcion of a "person."
Therefore, it is understandable why Nagarjuna. after dealing with the quescion regarding a substantia agent (karaka, Cbapter VTII) and aso the motiva
tion for such a theory (namey , the desire to rrac:e one's identity to the past,
purva, Chapter IX), would sette down to an examinacion of a singularly imponant metaphor used by the substantiaists to justify their conception of an
agent.
l t may be necessary to keep in mind that Vasubandhu, who wrote the
Abhidharmakofa, was a Saucrantika who believed that a "person" (pudgala) is
a merc designation (prajapt1) without any realiry. Hence his agreement with
the nominalist position (Akb p .461, a/ha prajaptif?, vayam apy evaf!J
brumaf?). The substancialist position (dravya-vada) was therefore identified
with the Sarvastivada view. Even though the Sarvascivadins did not actualy
propound a substantialist theory of a person. their conception of substance
(wabhavfl) could not escape such implications (sec Sakv p .362 , svabhavala iy

almalal/).
1

However, even if by implication, the Sarvasrivada theory were to be identified with che substantialist view of a person (pudgala), what sort of conception were the Yatslpucriyas upholding? They wece looking for a middle position between substantialism (dravya) and nominalism (prajfJaptt). The discussion in terms of the metaphor of "fue and fuc" was, therefore, intended to
overcome soch dualiry.
The argument follows thus: "Without fuel . there would be no designation of
fire. Yet, fire cannot be designated a.S sometbing different from fuel, nor as
something identical," (Akb p.462, Na hi vinendhanenagnif? prajapyale, na
canya indhanad agnif? iakyale prajflapayifuf?J niipy ananyal/). Similarly, a person is neirher identical nor different from the aggregates. (lf we a.re to accept
the oegacive interpretation of the Nigarjunian dialectic, as is ofren presented
by most modern scholars, tbe Madhyamika position would be no different
from that of the Yatslputriyas as described above.)
Whc.n the question was raiscd as co which of the six sense organs provide
knowledge of the "person," the Yatsputriyas answered: "By means of all six,"
(1abhir api, Akb p.463). They argued: "A 'person' is rccognized depending
upon visualy cognized materia form (ca./qur- vijeyani filPil'!t.). As sucha person should be declared as being visualy cognizable and visually not cognizable;
it is ,neither matecial form nor not materia form ."
Tbese and ocher arguments seem to indicate thac the YatsTputriyas were
following a dialectical method at arriving at a higher synthesis by avoiding the

EXAMINATION OF f IRE AND

Full

197

dicbotomies of thesis and antitbesis. This, unfortunately, is the view attributed

to Nigarjuna . The conteots of the present chaptet needs co be carefully
evaJuated in the light of the Yatsipuuiya conception of a person.
When the Yatsputriyas maiotained that firc and fuel arc neitber ident icaJ
nor different, thcy were acruaJly admitting both . Por them , both firc and fuel
arc cach constitutcd of four substanccs (ubhayam 11.f{a-dharmakarp), the only
difference is that in the case of firc the heac element (Uff!t1m) predominates,
whereas in the fuel it is latent. In this way , they ar both idcncical or noodiffercnt. Yet they arc different in terms of the difference in time (bhinnaktilatziiit). Thus, the negacivc description of the Yatsfputriyas is not meant as
a ncgation of metaphysical views, as it was in the case ofNigarjuoa (see above,
commentary on the dedicatory verses), but an assertion of both identity and
difference in order to arrive at a bigher synthesis. No such move is found in
Nigarjuna when be criticizes the idenciry of fire and fuel in che above verse.
For hisn identity of fire and fuel rneans identity of agent and action. If they werc
diffcrent, chen cach could be independent.

2.

Nitya-pradipla e1111 sjiid apradipana-helukaf?,

punar arambha-vaiyarthyam evarp ciikarmakafl sati.

A buming without a ause would be eteroally aflame. Furthecmore, its

commcnccment will be rendercd meani.ogless (usdcss]. Whcn that happcns, it will be without a function.
MKV(P) p .203: MKV(V) p.86.

The idcntity of fue and fuel recognized by the VitsTputriyas (as cxplaincd in
terms of the eight elements, alta-dharmaka, sec abovc) would lead to the view
that both firc: and fucl arc buroing all the time (nilya-pradipta), for the caloric
element (uma) is found in both, the difference being quantitativc rather chan
qualicativc . This fucthec lcads to the dcnial of the e.mpirical fact of starcing a
fire . lf fire and fuel arc always burning, thcn fire would be deprivcd of any
spcdfic function. Indeed, the Buddha's ust" of the metaphor of the fi.re at M
l.487 is non-substantiaJist in implication He wantcd to show rhat just as fire ~
not stoccd up anywhece when it is extinguished, even so a "frced one"
(talhiigata) does noc e.ist in ercrnal bliss afrer dcach (param maraf!"ii). Nagarjuna's criticism of the mctaphor of "firc and fuel" secms to follow doscly the
Buddha's own cxplanation of the phcnomenon of fuc .

198

T HE P~nLOSOPHY OF THE MIDDLE W AY

3. pt1ratrt1 nirt1pek;at11ad apradipana-hetuktll/,

p 1111t1r arambha-11aiyqrthyaf!l nitya-pradiptal/ prasajyate.

A burning without a c:ause, bec:ause it is not contingent on anothcr and,

therefore, etemally aflame, would imply the meaninglcssness of itS commencemeot.
MKV(P) p .203; MKV(V) p .86.

A substantialist v.iew of fire makes it iodepcndcnt of other clemcnts or things.

Hcncc, all the ,iinplications mcotioncd at X .2 will follow from such a view.

4. Tatrt1ilt1smad idhyamanam indhanaf!l bha11afiti tet,

k.enedhyat'lim indhanarp, lat tavan matra1n idarrt yadfi.
Hercin, if it is assumed that fuel is the present buroiog and, therefore,
that [i.e., buring] is mercly this [.i.e fuclJ, by what is fucl bcing
burnt?
MKV(P) p .204 ; MKV(V) p .87.

The substancialist point of view cxpressed at X.2 is further analysed herc. If fuel
already has the caloric element (umt1) and, thcrefore, is alrcady burning
(idhyam'lint1), one could not only raise the question as to when it started buroing but also joquite as to whac it is that brings about that burning.

5. Anyo na prapsyale 'priiplo na dha.!yaty adahan punal/,

na nirv'iisyaty anirvaa!, sthasyate 11a svaligavan.

[Fuel] that is <liffertnt is not reached; the urueached is not ignited. Furthermore, that wh.ich is n6t ignited does not cease. That which does not
cease remains, li.kc one that has its owo mark.
MKV(P) p .205; MKV(V) p.87.

199

EXAMJNATION OF FtRE AND FUEL.

lf it is assumed that firc and fuel arc different, a futther serics of questions
arise. Difference in the prescnt case implies absolutc distinction or in
dependcnce. As such, one entity cannot reach up to another. Fice canoot rcach
che fuel. Whcn firc cannot reach the fucl , it cannot burn. lhat which does not
burn reroains for ever and does not cease. This expla.ins the vicious circle the
metaphysical speculacions can lead to.

6. Anya evendhanad agnir indhanarrt prapnujiid yadJ~

stn sarrtpriipnoti puru1arrt j;urup1 ca striyar{Z yatha.
lf firc is diffcrcnt from fue it would reach the
would rcach for a man and a man for a woman.

fue,

just as a woman

MKV(P) p.206; MKV(V) p.88.

lf fire and fuel wece considered to be different yet complemcowy, theo Nagar
juna is willing to allow somc son of mutual relationship. He perceives such a
relationship bewteen a man and a woman.
Unfortunately, herc again 'we have a negacive interpretation from Can
dra.krrti. Without ta.king much trouble to cxamine the import of the metaphor
of "man and woman" uscd by Nagarjuna, Candraklrti assumcs that the rela
t iooship exemplified by it is mcaninglcss (dr1(iinta-t1aiy11rthyattz). On the con
uary, Nagarjuna is throwing the gauntlcc at his opponent asking him to show
that the sort of relationship chat cxists bctweeo a man and a woman can a..lso be
obtaincd between firc and fuel as the opponenl has conceit1ed of them .
Indccd, there could not be much difficulty in understanding the empirical
rclationship betweeo a man and a woman. Because of their complcmcntarity,
they arc attracted to cach other. Such a relationship is dearly expresscd by the
Buddha in the first two discourscs in thc Aguttara-.nikaya (1.1-2). Yct the
Buddha ncver attemptcd to go beyond chat cmpirical relacionship-to inquire as
to how it all happened. Such an inquiry was undoubtedly the foundatioo of the
speculations recorded in the Upanz'!ads. Describing the origin of the uaiverse
from a single unitary "self' (atman), the Brhadiira1}yaka Upanifad (I.4.3-4)
says:
He, verily, had no delight. Therefore he who is alonc has no
dclight. He desir~d a second. He became as large as a woman and a
man in dose cmbracc. He caused that self to fali into two parts.
'

200

T HE PHlLOSOPHY OF THE MIODLE W AY

From that arose husband and wife . .. . He became united with

her. From chat human bejngs wece produced .
She thought, "How can he unite with me after having produced
me from himself?" Well, !et me hide mysdf. She became a cow,
the ocher became a bull and was united with her and from thac
cows werc boro .... Thus, indeed , he produccd everything
whatever exists in pairs. down to the antS.
This is the son of answer char emerges from an inquiry chat is noc only
directed at understading the absolute origin of things, but also their substancialicy. The metapho,r of the fire and fuel wece utili.zed by che VitsTputriyas,
not merely co undersr:and the relacionship between two empirical evencs, but
also to justify the conception of a prior exiscent "person" (pudgala), no differeot from the pre-cxisccnc "self" ('iitman) assumed in the Upani/ads, which is
clearly exprcscd in the metaphor of "man and woman."
Percepcive Nagarjuna was chus aware of the motivations of those who
presented the metaphor. In the presenc verse he was therefore allowing the
possibiity of fire and fuel having a complcmcntary relationship . Howcver, if
any ocher implication is drawo from such a reJationship, Nagarjuna was ready
to expose its untenabiliry. This he docs in che verscs chat foUow.

7. . Anya evendhanad agnir irzdhanam. ka man apnuyat,

agnlndhane yadi .rjiit'iim anyonyena tiriukrta.
The fue that is different from fueJ may rcach the fud only if fue and fud
wece co exist mutually separated.
MKV(P) p .206; MKV(fi? p .88.

Here agaio, Nagarjuna is qualifying the sort of relatiooship chat may obtaio
between fire and fuel if they arc atuacced to one another. One of the first conditioos would be that they arc separate. Without such separation it would be
meaningless to speak of one reachiog for the ocher. It is an attempt to destroy
the belief in a myscerious underying unity , any substantial connection. Nagarjuna is wiUing to allow for the empirical differences and the relationship of
dependence among such cvents. However, he is not prepared to leave any room
for any speculation about underlying substances (.r11abhava).

EXAMINl.TION OF flRB AND FUEL

8.

201

Yadindhanam apek,yiignir apek,yignif!J yadindhanaf!I,

katara/ purva-nqpannaf!I yad apek,yagnir indhanaf!I,

If fire is contingent upon fuel and fuel upon fire, which of them is p~
accomplished so that fire could be contingent upon fuel?
MKV(P) p.207; MKV(V) p .88.

The motivation of the substantfalists who were ucilizing the metapbors of "fue
and fuel" (agnlnrlhana) as well as "woman and man" (stff-puru1a), as exlained
at X.6, arc brought to the forcfront hece. Let fire and fuel be related by way of
cootingence (apek,"ii) . Nagarjuna has no objection to it and this is clearly indicatcd by the manner in which Candrilirti himself ucilizcs this particularcon;
ceptioo of cootingence on numerous oa:asionions (sec MKV(P) pp.67, 189,
200, 202.. 345, 492, 527). Io the presenc case, Nagarjuna is oot rejcccing the
mutual contiogence (paraspar?ipek,'ii) of pheoomena, but ooly the inquiry
relating to the prc-accomplishmcnt or prior existence (purva--ni,panna) of any
one of them. Chaptcr IX made it abundantly elear that speculations regarding
the prior entity led to most metaphysical speculations. Having raised the ques-tion as to which one of thcse two thiogs -fue and fuel-is prior, a questioo
that a substantialist cannot resist asking, Nagacjuna procceds to analyse the
poss.ible answers and cxplain thei.r unsatisfactory implication.~ .

9.

Yadindhanam apek,fiignir agntl/ n'ddhasya sadhana1!J,

evaf!Z saffndhanaf!I capi bhavi,syati niragnika'f!'.

If ftre were to be contingcnt upon fuel, there would be proof of fire th.at
is already proved [to e:xist] . When that is the case, evcn fucl would exist
without ftr:e .
MKV(P) p.207; MKV(V) pp.88-89.

Frustrating any attcmpt on the part of the substantialist, Nagarjuna insists that
if fuel wece to be the prior entity (purva-nifpanna), and that firc is contingent
upon it, then what is to be established is alrcady cstablishcd. In other words,

TFE PHILOSOPHY OF nE M IODLE W AY

202

the substantialist interpretations of contingence (paraspar7ipekfi) would me2.0

that fire is aJready existenr in the fuel , which is itself the prior entity. If it is
already existent in the fuel, tben the implication is that it wiJI need no fuel for
its reaJ cxistence. If so, fue and fuel could not be murually contingenc and one
can conceive of fuel without fire (nir-agnikaftl) .

10.

Yo 'peh,ya Jidhyate bhavas tam etiiipeiqya Jidhyatt~

yadi yo 'pek1ita11yaf? Ja Jidhyaliif(J kam apekfya kal/.

MKV(P) p.208; MKV(V) p.89.

If events are to be truy concingent, then they should be murually contingent or

dependent. If any one of two murually contingent entities is to be fou o,d in a
substantia or essential way in the other, then the notion pf conringe11ce is
nullified. One becomes the essential and the orher the superficial . The problem chen would be: which depends upon what? The Wtsfpu-iyas . as mentioned earlier, werc not looking fot means of dissolving idcntity and difference.
They arc struggling to cetain both identity and difference. and the n-move on to
a higher synthesis. Nigatjuna's attempt herc is not to allow the metaphysici_an
the very concepts of identity and difference, which they were going to ucilize as
a springboard for leaping coward the notion of a mctaphyskaJ "person"

(pudgala).

1l .

Yo 'pek.tya Jidhyate bhaval, JO Jiddho 'pek.tate kathaf(J,

athapy 4pekfate Jiddhal? tv apek.tazya na yujyate.
1

Whatevct cxistent that is cstablished through cootingence, how Cln tha_t,

if it is not yet established, be conringent? Even so [how can,] that whi~ is
already establishcd be contingent? F~r, _i~. coot:ingence is not propec.
MKV(P) p.29: MKV(V) p.89.

EXAMJNATION OF flRE AND

FuEt

203

X.10 refers to the fire that is already established (.riddha), chat is, something
that is really existent (sad-bhuta). lf one were to move to the other extreme and
speak about events that are not established (a.ui:/dha), that is, those that are
really oon-existent (asad-bhuta), how can such events be contingcnt (apekfya

sidhyate)?
Thus, neithcr the already established (siddha) nor the unestablished (asiddha) can be relatcd by way of mutual contingeoce. Contingence (apek{a) is not
established in thls substantialist way.

12.

Apek,yendhanam agntr na nanapek!yagnir indhana1lJ,

apek,ycndhanam agni1lJ na n?inapek;y?ig~im indhana'f?t.

.Fire is .not conti.ngent upon fuel; fire is not non-contingent upon fuel.
Fuel is not contingent upon 6re; fue is not non-contingent upon fire.
MKV(P) p.209; MKV(V) pp.8890.

1f the ideas expressed in tl)e present verse were to be examioed independent of

wha.t went before, it would be. easy to eap to the conclusion that Nagarjuna is
here expressing the "inexpresible''; that the negations such as "oon-ceasing''
(anirodhaf!I.) ind "non-arising" (anutp?idaf(l) wece supposed to elear the way
for the ''non-cohceptual," lnon-dual" ultimate reality (paramiirtha?). On the
conttary, if these thoughts are placed in their proper cootext, namely, the
metaphysical explanations of "muial contingence" (parasparapek!a) , - then
'these oegative statements can be understood as outright rejections of such
metaphysical explanations. As is evident f rom several quocatioos from Candra.kl rti hiroself (see X.7), "contiogence" (ttpek!a) can be explained in a nonmetaphysica or empirica way, just as much as the not.ion of cause or condition
(pratyaya) or the idea of motion (gatt) can be elucidated without falling into
the quagmire of metaphysics (see Cbapters l ,Il).

13.

Agacchaty anyato nagnir indhane 'gnir na vidyate,

atrendhane efam ukta11'f gamyamina-gafiigatail;.

204

THE PHJLOSOPHY OF THE MIDDLE

WAY

Fi.te does not come out of something different nor is fire secn to be in the
fuel. Herein, with regatd to fuel, the rest is m.tcd .as i.n the case of present
moving, the moved and the not movcd.
MVK(P) pp.210-211 ; MKV(V,l p.90,

Mter dealig with che metaphysicaJ interpretations of "contingence" (apekf1i)

assumed betwecn fue and fueJ, Nagarjuna retu.ens co the more familiar
substancialist theories of self-causation (svata utpattt) and external causation
(para/a utpatti). Fi(e is not inherent in the fucl nor does it issue out of
something external. If any furcher explanations are to be provided which arc
themselves founded upoa sucb substantialist ideas, all such expla9ations can be
analysed in terms of the mechods adopced in Chapter II dealing with motion

(g_atz).

14. I ndhanaf?J punar agnir na ri1ignir 'anyatra cendhanat 1

nagnir indhanavan riiignav indhaniini na teru sal{.
Furthermorc, fuel is not fire. Apart from fucl therc is no fire . .Fire is not
possessed of fuel . Fuc is not in the fue, noc is it [i.c., fire] in them.
MKV(P) p.211; MKV(V) p .91.

The refutation of all metaphysical formulations of the nocion of identity as applied to fire and fuel is presented here. This is done in terms of the analyticaJ
methods followed in Chapter I.

15 .

Agnzndhanabhyiif?I vyakhyata atmopadiinayoh. kramal;.

s4rvo niravaefena sardhaf!J ghata-pa{iidibhil;.

Through the examples of fire and fuel, together with the examples of
pot, cloth, etc. evcry method of analysis of the self and grasping have
bcen cxplaincd without exccption.
MKV(P} pp.212-213; MKV(V,l pp.91-92.

EXAMINAT!ON Of fl RE ANO fUEL

20~

The metaphysicaJ ioterprecations of "self' (aJman) and "grasping" (uparliina)

arc exposed here , especially in relacion to the metaphor of fiie and fuel'
(agn'i'ndhana) . Does thls mean thac there could be ooo-metaphysicaJ expana
cions of both "self' and "grasping?" The answer would be in the posicive ,
especially in vew of whar Nagarjuna has said in reference to "graspiog" and the
"one who grasps" (uparliitr) or the action and che agent (ka"na-karaka).
Tb is analysis is not confined to the metaphor of "fue and fue" alone. lt applies to all other metaphors used during thls period of speculatioo, such as" cay
and the pt," and " thread and the doth."

16. Almamif ca satattvaf!Z ye bha11'iin'ii1?1- ca PrJhak Prthak,

nirdianti na fiin manye fiisanasyiirtha-kovidiin.
Those who posit the substantialicy of the self as well as of disaete existents- these I do not coosidcr to be experts in the meaning of the (Buddha's] m~ge.
MKV(P) p.214; MKV(V) p.92.

Here then is a elear and unequivocal assercion oo che pan of Nagarjuna. Those
who assert the substantial exi.Jtence (satat111am) , whether that be of a monistic
"self' (atman) or of distinct enticies (prthak PrJhak bha11ant1'f!') are not conversant with the teachings of the Buddha. When Nagaijuna makes that assertion
with the statement: "[I) do not consider" ([aha'!'] na manye), what sorr of self
was he recognizing? It certainly could not be anything like what he was
refutiog. The answer to this question will be provided ater.

CHAPTER
ELEVEN

Examination of the Prior and Posterior Extremities

(PUrvriparakoti-pank.ra)

1.

P"Urva prajfiiiyate kofir nety u11iica mahiimunil/,

J(lf!'J'iiro 'na11ariigro hi nasty adir napi pacimal/.

The Great Sage has stated that the prior end is not known. The lifeprocess is without beginning and end. There is neither a beginning nor
an end.
MKV(P) p .219; MKV{V) p.95 .

The criticism of the "self" (atman) as a substantial entity continued in the last
few chapters brings up more related issues for discussion. If there is no such entity, how can the life process consisting of tepeated births and deaths be explained?
Rebitth or rebecoming (ptfn(lbbhava) was an important element in the Buddhist doctrine,, even though the Buddha was concetned more with the problems of the present life than of the past. As mentioned earlier (see commentary on IX.1-2), for the Buddha, the knowledge of the past (afitat?Jse fJaf!a) was
an important means of understanding the present. Yet he did not encourage
speculation regarding the past as he feared that this would eventually lead to all
sorts of mecaphysical views. For this reason, he remained aloof from specuacing
on the absoute origin of things. That aloofness is cleary implied in his statement: "lnconceivable is the beginning of this life-process . The prior end is not
evident," (Anamataggo 1yat?J ... 1af{lJiiro pubbakofi na pannayati, S 2.178,
193; 3.144.151). There is here no denial of the prior end or the first beginning
of things, but ooly of its cooceivability or perceptibility. However, with the
problems that emerged duting the scholastic period In the matter of expaining
any form of origin (utpada), not merely of the first beginning (puN1a kop),
some Buddhists wece compelled to deny outrigbt any form of beginning.
Whilc the Sarvastivada conception of substance left no room for origin and
cessation, the Sautrantikas had difficulties explaining the origin and cessation
of momentary events. The difficulcies invoved in providing an explanatioo
206

ExAMlNATJON OF'fHE PRJO,R AND PoSIB.RJOR Ex~REMfflE

207

seems co have led the metaphysicians to assume the absence of a prior end,
rather thao of its iocooceivability. Henc;e the term ana'l!J4tagga ("inconceivable
is the begioning'') came to be replaced by anavqragra ("without end and beginning"). As such, Nagarjuna proceeds with the exa.mination of the view current
during his day, only to revert back to the Buddha's own approach to this problem in the cod. Cancraklr6. confining himself to .t he literary tradition of his
day and influenced by a transcendentalist approach in his interpretation of
Nagarjuna, does not seem to recognize the above mentioned change taking
place in the Buddhist conception of the life-process (.raf!liiira).

2.

Naivfigtaf'!I navara?!' y11Sya tasya madhya1?Z kuto bhavet,

tasmfin natropapadyante purvapara-.raha-kramaf!,

How could there be the middle of that which has neither a beginning nor
a_n end? Therefore, the methock of (distinguishing) the prior, the
posterior or both together (i.e., the middle) are not appropriate.
MKV(P) pp.220-221; MKV(V) p.96.

Startiog with the cunent assumption that there is neither a prior nor a posterior
end, Nigarjuna raises the quescioo as to how , in the absence of these two extremities, one can speak of a "middle'' (madhya). He then proceeds to apply
this cciticism to the conception of the life-process (.raf!J-.riira) as- understood and
interpreted by the metaphysicians.

3. Piiroa'l(I /iitir yadi bhavejjarf,i..maraTJam utt(JrM(l,

nir-jarii-marar{ii 1atir bhavej;ayeta camrtafl.

lf birt:twere to come fiist and decay and death were to follow, the n birth
wouid be without decay and death, and an immortal wouJd thus emergc.
MKV(P) p .221; MKV(V) p.96.

The difficuties cre~ted by the Sa.cvastivadios and the Sautrantikas in the mat-

208

THE PHILOSOPHY OF tHE MIDDLE

WAY

ter of explaining causa) continuity have beeo discussed at length. The linelU'
view of the causal process broken up into discrete events, one folowing the
ocher, left these evencs without any possible relations . lf the same model were
to be used in explaining evencs such as birth (jalt) decay-death (jafii-maraa),
then the implication woud be that birth has nothing to do with decay-death.
There is no necessary connection between them. If this argument were co be
carried to its condusio, then it could be maintained thac there is immoctaicy
(amrta), for there coll.d be sornething chat continues without decay.death.
Here, Nagarjuna is not insiscing that decay-death should be inherent in birth.
Rather, he is exposing the difficulties confrooted by the metaphysicians who
upheld a linear view of the causa) process.

4. Pa1caj Jalir yadi bhavejjara-mara1Jam adita/?,

ahetukam aj'iitasya syiij jarfi-mtlraf!a'l'(Z katha'l'(Z.

lf birth were to be posterio.r and decay-death anterior, chen the

acter

would be without a cause. How couJd there be decay-death of one who is

not bom?
MKV(P) p.222; MKV(V) p.97.

Pla:cing birth after decay-death in this lineat view of the lite-process, the logical
conclusion is inevitabe that birth cannot relace itself to anythiag prior and,
therefore, is uncaused. lf birth cannot be explained, .just as much as decaydeath could not be accounted foJ (as stated at X.1.3) , then we -are left with the
peak of decay-aeath of
tJOborn (ajiita). Nagarjuna considers it inappropriate to s_
>,'
something/ someone who is not born.

5.

Na jarli-mara'f!enai11a 1a1i ca saha yujyate~

mtiyeta /aJamana ca sfiic cahetukatobhayol(.

Indeed, decay-death as concomitant ofbirth is not proper. [In chat case,]

what is in the process of being boro will a1so be dying and both would be
rendered causeless.
MKV(P) p.223; MKV(V) p.97.

EXAMlNA'TlON Of THE PfllOR AND POSTBRJOR EXTREMITIE5

209

As mentioned in commentary on Xl.4, one way of expainng the problems

raised at XI.4-5 would be to assume that decay-death (jar'?icmaraf!a) is ioherent
in birth (/iitz). This celationship of inherence would mean chat decay-death
begios at the same time as bin:h, and as sueh ie would be appropriate to say that
one who is being boro is at the same time dying. If s0; neither one of them
could be the cause of the other. Being uncaused, their occurrence wou!d be
rather spontaneous.

6.

Yatra na prabhavanty ettl purvapara-saha-krarnal;,

pra:pancayanti fii?'(l jiili?'!l taj jara-rnarfJ'!4f!i ca kif!i.

Wherever such methods of (discriminating) the por, the posterior and

the $Unultaneous do not arise, why be obsessed by such birth -and such
decay-death.
MKV(P) p.224; MKV(V) p.97.

The speculation that is questioned herc is specifically related to the sort of succession discussed previously. lt is not every form of reflection that is rejected .
Even if the verb prapacayanti wece to mean "conceptualizing", as
understood by a majority of m odern uanslators of Buddhist texts, in chis context, it does not mean the emptying of the miod of all concepts of birth and
decay-death. lt is the particular form of conceptualization mentioned in the
previous statements as well as the earier verses that is co be avoided.

7.

Kiirya?'!l ca kartzf!a'!(Z caiva lak!Ya?'!l lak!af!am eva ca,

vedana vedtzkai caiva santy artha ye ca kecana.

Effett and cause as well as chara~terized and characteristic, together with

feeling and fe.e ter or whatever fruits there are.

8.

Pfirva na vidyate kofif? sa?'(ZSaraJya nil keva/af!J,

sarvefiim api bh'Qvari'ii'l'{J. purvli kofi na vidyate.

210

THE PHILOSOPHY OP THE MIDDLE .W AY

the prior end of thcsc is not evidcnt. Of the cntire life-process as well as of
all existcnts, the prior end is not cvidcnt.
MKV(P) p.224; MKV(V) p .98.

It is not suprizing to sec Nagarjuni concluding his analysis of extremities (kop)

of events, whether they be of cause and effect, or the charactcrized and the
char~cteristic, of experiences, of the life-process, and even of all existents
(bhava), with the assercion that thcy arc not cvident (na vidyate). He does not
mainwn chat these events arc without extremities.

CHAPTER

TWELVE
Examination of Suffering
(Du~kha-pan-k~a)

1. S11ayaf{' krtaf{' parakrtaf'?l dv'iibhyiif!I- krtam ahetukaf{',

duf?kham ity eka icchanti tac ca liiiryaf!t- na yujyate.

Some assume that suffering is sclf-c:aused, c:aused by anothcr, c:aused by

both or without a c:ause. [Suffering as] such an dfect is indeed not appropriate.
MKV(P) p .227; MKV(V} p . 100.

The discussion of the life-ptoccss (saf!t-fiira) leads Nagarjuna to an cxamination

of the problem of suffering (dul/kha). lt sccms that when compiling this
chapter Nagarjuna bad a copy of the Buddha's discoutse to Acda-Kassapa (S
2.18-22; T1a 12.20 [Tairho 2.86a]) in front of him.
The Buddha's discourse co Accla-Kassapa begins witb the fout cheories of the
causacion of suffering referred to by Nagarjuna in almost identical terms. The
on y difference is with regard to the explanation of the fourth. Where Nagarjuna has ahetukatrz, the Buddha's discourse refers to adhiccasamuppannaf{' (sec
Xll.9 below). Anotber difference is tha't while the Buddha's advises Kassapa
not to get involvcd in such speculacions. insisting, "Do not [say] so," (mah'
evaf!t-), Nagarjuna maincains chat thesc cheories arc not appropriace.

2.

Svayarp krtaf'?I yadi bha11et prafiya na tato bhavet,

skandh'iin iman ami skandhal/ sat?Jbhavanti prafiya

bi:

lf [suffering were to be] self-caused, thcn ie could not occur dependenty .

lndced, depending upon these aggccgates, these otber aggregates occur.
MKV(P) p.228; MKV(V) p. 100.

2 11

212

THE PHILOS0PHY OF THE MIDDLE WAY

The nocion of self-causation of,suffering is funher elaboraced by the Buddha as

the view expressed in the statement, "He acts and he [himself] reaps the consequences," (so karoti so pafisaf!l-vediy4tt). The Buddha's reason for not accepting
such a view is that. it leads to the belief in ecernalism (sassataf!I- eltl'f!I paretz) .
Nagarjuna certainly koew that the Buddha was utiliziog the cooceptioa of
dependence (paficcasamuppada) co avoid aay mecaphysical theory of eternalism. This was the basie theme, not oo1y in the cliscourse to Kaccayana, but
also in the dicoure to Acea-Kassapa , where it is ooce again presented as the
middle position berween the excremes of eternalism and annihilatiooism (S
2.20}. Ic is, therefore, noc surprising to see Nagarjunarejectingthe conception
of self-causation of suffcring on the grounds thai in sucha case suffering would
not be dependently arisen (na prafftya bht111et).
Furthermore, the eternalism that comes to be embodied in a theory'of selfcausation of suffering pertains moce to the individual self or soul (atman), and
not the substance (wabliiiv4). Thcrcfore, Nagarjuna insists that what is being
asserted as a self or soul is nothing other than the arising of a set of aggregates
dcpending upon (praffy4) another set of aggregates.

3.

Yr1dy amt bhya ime 'nye syur ebhyo vam'i pare yadi,
bhavet para-krtan;. duf?khaf!I- parair ebhir ami krtal;.

lf from these those that arc differeot were to come to be, or if from those
these different [things] wece to come to be, then suffering would be caused by another, for these are caused by those that are different.
MKV(P) p. 229; MKV(V) p .100.

Here we find a definition of "external causation" (para-krta) . Nagarjuna has

already explained the relationship between setf-nature or substance (svabliiiva)
and other-nature (parabhiiva)(see f.3). There he maintained that without selfnarure there cannot be other.narure (avidyamiine svabhave parabha110 na
vidyate). In the prcsent verse, Nigarjuna applies the same principle to explain
"causation by another. '' Thus, we find him utilizing the ablative case
(amTbhyal;, ebhyal;), exprcssive of "source" or "origin ," 'in order to explain the
arising of somcthiog different from within something chat preceded. In ocher
. words, even though the effcct is different frorn the .cause, it acises from the
cause; that is, exteroal causation is invariably related to self-causation.

EX/.MINATION OF SUFFERJNG

213

It is important to distinguish this statemeot from that found at XJI. 2 which ,

instead of ~ing the ablative case, empl0.ys the accusative (iman skandhan
pmliya) in order co highlight the principle of dependence. Indeed, the very term
idapp11Ccayafii (idaf!ZPraJyayaJ'ii} utilizcd by the Buddha to cxprcss the principle of dependence is couchcd in the accusacive case (idaf!Z). instead of the
ablative of soucce.

4. Sva-pudgala-krJa1{l dul;khaf!Z yadi dul;kllaf!Z punQr 11inii,

wa-pudgalal; sa katamo yena dul/kha?!f- svaya1{l krtaf!Z.

lf suffering is oused by ones owo person, thcn that own person can exist
without suffcring. Who is he by whom suffering is self-ouscd?
MKV(P) p .230; MKV( V) p. 101 .

As shown before, the Buddhjst mecaphysicians who adopted a thcory of

moments had difficuJty explaining personal continuity or identity in a more
empirical way. They were confronted with the problem of expaining good and
bad . suffering and happiness as pan of the persona continuity. Thus, the Sarvastivadins would maintain that suffcring. etc. arc mere qualities (lak,aa) thac
characterize the substance (dravya}, or they would, along with the Sautrantikas,
maincain rhat qualities are appropriated or become pan of the stream
(.rvasaf!Jfiina-patila, A.K 2.36; Akb p .62; Poussin, L'Abhidharmakofa, vol. 1.
p. 179). Such a perspeccive inevicably leads to a di.stinction between the person
or the stream of persona identity, on the one hand, and qualities like suffering
on the other. What Nagarjuna is atcempting to do in the present verse is to bring out the metaphysical implications of this cheory. According to this theory,
suffering is something cxternaJ to the individual. It is an enciry having its own
reality . lt is something caused by a person, as a carpenter wou d produce a piece
of furniture. As such, Nagarjuna questions the very nature of chat person who,
being independent of sufferig, causes suffering on its owo.

5. Para-pudga/11/af!l duf?khaf!Z yadi yasmai p radiyate,

pare71a kr111a lad duf?khaf!Z sa duf?khena vinii ku1af?.

214.

THE PHILOSOPHY 0F THE M IDPlB W Y

lf suffering were to be produced by one person and given over to anothet,

that suffering is caused by the former. How can the latter be ide~tified
without suffering?
MKV(P) p.231; MKV(V) p .l OL

If one were to accept theseparation of sUfferiog from the person who produces
ie (i.e. , the theory cticicized at XII.4), then it will lead to d.ifficulties not only
for one who adopts self-causation of sufferiog (svayaf!t Artaf!t du//khaf!t), but
aso for one who upholds the opposire view , namely, suffering is caused by
another (para-krtaf!t t:/uf?khaf!t). The latter will have to maintain that suffe.t ing
is caused by one person and passed on to another. But that other wouJd himself
be independen~ of suffering, in the same way as the person who caused it. The
question stili remains as to how that person can be identified .

6. Para-pudgalafaf!t dul;kha1'{1 yadi kal? para-pudgala/?.

11ina dul;khena yaf? krt1iii parasmai prahit1oti tat.

If suffering is caused by anothec person, who is that other person who,
himsclf without ruffecing, causes ie and bestows it on another?
MKV(P) p.231; MKV(V} p .101.

This is similar to the arguments presented before. The sharp dichotomy between the agent of suffering and su:ffecing itsef proinpts ~agarjuna to question the nature of that othet person (para-pudga/4) who is supposed to be the
author of suffering and who passes it oo to an.other.

7.

S11aya1!J. krtasyiipra.riddher du//kht111J para;krta??J kuta!/,

przro hi duftkha??J yat kuryat tat ta.rya syat svayarri krta??J.

With the non-establishment of self-causation, how can there be suffering cased 'by another? For, indeed, if another wcre to cause that suffer
iog, 'in rclaclon to bim it would be self-ca~.

ExAM1NATION OF SUPr'ER1NG

8.

21S

Na l'iival wakrtaf!l dul/khaf!I na hi lenai11a lat krtaf!l,

paro nii1makr1af cet sy'iid dul/kha'(l para,.krtaf!l kathaf!l.

So long as suffering 1s not self-caused, it is, indecd, not caused by oncself.

lf the ocher were not to do it by himself, how could suffcring be causcd
by anotbcr?
MKV(PJ p.232; MKV(V) pp. 101- 102.

The i.mplications ofXII.3 discusscd carlicr arc stated oncc moce ccarly and explicitly in thcsc two vcrses. If some othcr pctson wece co causc suffcring, theo
that suffering, in relation to that particular person, would be sclf-caused. Tuus,
if one pcrspcctvic is not valid then the other too would be invalidaccd . This, as
pointcd ou t earlicr, is the method adopted at 1.3 to reject both self-nature
(svabh'iiva) and othcr-narure (parabh'iiva) .

9.

Syiid ubh'iibhy'iif!l krkl?P dul;khaf!l sjiid ekaika-krtaf!l yadi,

pafiik'iir'iiwayaf!lk'iiraf!l dul/kham ahetukaf?I kutaf!.

If suffeng wece to be caused by both, it would be caused by cach individually. Whcnce can thece be suffering that is causcd neither by
another nor by oneself aild is without a cause?
MKV(P) p.233; MKV(V) p. 102,

The tbird chcory of the causation of suffecing refcued to in the discoursc co

Acela-l<2Ssapa is a combination of causation by oncself and causatioo by
anochcr (sayaf!l katan ca paraf?j katafJ ca). This is understood by Nagirjuna as
caus;1.tion by cach individual (ekaika-krta'!J) . Howevet, he has already rejected
both typcs of causation.
lf suffcting we re not caused by onesclf or another, chen accordiug to the
Buddha ie would "arise one top of another" (adhiccasamuppanna). Such arising
is contrary to "arising by moviog towards or depending upon another" (pa#cca.samuppanna). This implies arising wichout any causa! connectio n. Hcocc ,
atihiccasamuppanna becomcs a synonym for ahe111ka (cause-less) . For Nagarjuna, chcre is no such uocaused suffcring.

216

THE PHtLOSOPHY OF THE MIDDl.l! WAY

lO.

Na kevala?(t hi dul?khasya c~rvidya'f?J na vidyate,

bahyiiriiim api bhavana'f?J caturvidya'f!Z 11a vidyate.

lt is not that the foutfold theory applied exdusively to suffering is not cvident. Tbe fouold theo.ry pertaining to other existents too is not evident.
MKV(P) p .233; M.KV(V) p. 102.

Iha-<a's explanation

that these are the "four-fold pombie views" (caturvidhyanj)

(p ,88) seems to leave no toom for. a fifth view which both the Buddha (in his
discourse to Kassapa) and Nagarjuoa (at XII.2) wece very clearly and unequivocaUy upholding, namely. "dependent arising of suffering." Hence his
condusion that these four possible views can. equally be applied to demonstrate
the impossibility ofas.serting ele1nents of the external world. On the concrary,
Nagarjuna (as well as the Buddha) were mercly criticizing the futility of
adoptlog these f91J! .particular views io explaining suffering as weU as other
elements in the woi'td of experience. Indeed , Nagarjuna was dearly aware of
the facr that these four are not the ooJy views explicating the causation of suffering. Hence his statement in the very fust verse in this chapter, "Some assert"
(eke icchanti), wruch meaos that it is not evecyone that assects such theories.

CHAPTER

THIRTEEN
Examination of Action and the Agent

(Sa1'!Jskara-pctrtk!a)

l.

Tan mr1a mo1a-dharma yad bhagavin ity a'bhi1ata,

sart1e C11 mofa-dharmaf!al? sa'f!'Skiirif? lena le mrfi.

The Ble$ed One has said that whatever is of deccptive naturc, that is
delusion . All things that arc of deccptive nature involve dispositions.
Therefore, they are delusions.
MKV(P) p.237; MKV(V) p.104 .

A chapter dealing with disposicions (saf!Jskiira) immediately following an

anaJysis of suffering (duf?kha) need not create any confusion. Nor should the
fact the the title of this chapter is prescntcd in the Tibetan translation as lattva,
instead of sa'f!'Jkra (sec Inada, p .91), lead to difficultics in undcrstanding it.
There cannot beany doubt that the original chapter was named "Examinat ioo of the Dispositions." Even the most cursory glance at the statcments of the
Buddha in the discourses would reveal the naked fad: that he never looked
upon all phenomena (sabbe dhammi) as "suffering" or "unsatisfactory" (dukkha). However, the classicaJ Hindu philosophers who misquoted the Buddhist
textS, and some of the modern intccprerers who wece guided by such philosophm,
have been cesponsible for ponraying Buddhism as a pessim.istic religion by
misintecpreting the Buddha-word, especially the doctrine of "suffering" {sec
Kaupahana , "The nocion of suffering in early Budclliism, compared with som.e
reflectioos of early Wittgenstein ," PEW 27 [1977]:423-431.)
The three prominent characteristics, impermanence (iznicca), sufie.dog (dukkha) , and oon-substantiaHty (tJntta) have bcen recklessJy lumped together by
these interpreters and appied to all phenoffi:ena when the Bddha in innumbcrablc instances (M 1.228; S 3.133; 4.401; A 1.286; Dhp 277-279; T1a
10.7 [Tat'sho2.66b-67aj; Tseng 23.4 [Tairho2.668c) Ch'ang 1.1[Tairho1.9b))
and in every statement he made in regard to these three ch~racteriscics, always
'l.1

218

TH!l PHI LOSOPHY Ol' THE MIDDLE W AY

distinguisbed between Jankhara and dhamma. These duce characteristics a.re

always mentioned as follows:
"AU dispositions are impermaneot" (sabbe sankliiira aniccii) .
u . "All dispositions ase suffering" (sabbe sankh'iira dukkh"ii) .
iii. "AU phenomena a.re oon-substantial" (sabbe dhammii anat/ii).
1.

ln the first place, the elear distinction made by the Buddha becween
"dispositions" (sar.mkiira) and "phenomena" (dharma) and, secondly, the fact
tbat he specifically referred to the former being subject to suffering, could not
have escaped the penetracing and careful eye of Nagarjuna, a philosopher
whose wcitings have .influenced some of the best brains in the East throughout
the cenrur.ies. If all disposit.ions were coosideced by the Buddha to be subject to
suffecing, then there is no reason to doubt as to w4y Nagarjuoa sbouJd not concencrate bis attention on these "dispositions" aftcr his examination of the prob
Iem of suffering. Hence che reason for the present chapter.
Why the Tibet.an translators should considet this to be an examination of
truth (tat111a) is also nor a myscery. They were simply loolcing at the conclusion
of the chapter. Is there any connection between "disposit:ions" and "truth"?
As pointed out easlier (V.8), the "appeasement of dispositions"
(Ia'f!lskiiropafama) .is the ultimatc goal of Buddhism. Excessive lust (raga) is
supposed to lead to the strengthening or solidificacion of one's dispositions,
which in ruro conuibutes to grasping, no~ only for the objects of sense pleasure,
but also for ideas . The result would be the dogmacie grasping on to absolute
truth or truths. The elimination of lust wouJd then mean the elimination of the
dispositions too (Ia?!J.rkiira-R,aya) which would imply virtual death and oo
motivatioo for any action or even to continue with one's preseot life. Thus, the
Buddha hlmsclf was willing to characterize the death of a "freed one"
(tathagata) as the "cessation of dispositions" (sakharakkaya) (Dhp 383)
leading to the cessation of the stream of bccoming (bha11a-sota) .
Howcver, while recognizing the waning of lust (rag4k.khaya) as the way to
freedom , the Buddha did noc encomage the complcte elimination of dispositions which would mean su icide. Ie seems that che Buddha. did not recognize a
one-to-one relationship betweeo the waning of' lust and the cessation of
dispositions. Hence his emphasis on the appeasement of dispositioos while living and the cessacion of dispositions at the .time of death.
The sucngtheoing of dispositions, as meotiooed earlier, leads to dogmacie
bclicfs. These would pert:Un to personal immort.ality, conceived io the form of
a belief in an eternal soul or self (atman) or of a universal reality (loka,
brahf!14n). Any form of eternalism (fiivata) would be che consequence of such
strong dispositional tendendes. The opposice of ie would be annihilacionism

EXA.WNTION Ol' ACTION AND THE AGENT

219

(uccheda), and it is not difficult to understand why the critics of the Buddha
would refer to him as an annihilationist ( ucchedatiiidi) upholding the annihila
tion of re.lly ttisdng sentient being (M 1.140). They were p.robably referring to
the Buddha's advocacy of the "cessation of dispositions" (sankhtirakkhaya) at
death.
Yet, for th~ living human being the Buddha was nor prescribing the elimination of dispositions. Rather he adyocated their appeasement. This view has
significant epistemoidgkal implications. A living human beiog needs to act.
Action involves understanding. Conduct (caraf!a) is preceded by knowledge
(vidya). 'One needsJcnowJedge of oneself as well of the external world. "Omniscience" or knowledge of ever:ything :was not available to the Buddha. Hence,
neither the absolute oripn of things nor the absolute end of things were
discussed in Buddhism (see C)apter XI). Any theoty that attempts to explain
su<ih origins and ends, whether it._pertains to an eternal self oi: soul (atman) or a
substance (sva!Jliiiva) , was unacceptable to the Buddha.
Dispositions are invariably assodated with the knowledge derived from the
senses. The innumerable data provided by the senses cannot easily be handled
by the human being. As William James characterized sense experience, it is a
"big, b!ooming, buzzing confusion" (Some Problems ofPhilosophy p .32). Being unable to deal with such con.fusion, human beings are compelled to be
selective. They pick out the things that interest them, leaving out others. In
tbat process, they deveop dispositions and these disposfrions in turn conttibutes their share in gainiog knowledge of the world . As such, the world of
ordered experience is one that is construted, made, put together (sam
+ vkr. "to do, to make"), by the human being. This is the ptagmatic conception of truth (latt11a) that is prominent in the Buddha's teaching. The appeasement of dispositions thus contributcs to the e!imination of dogmatism, of
grasping after absolute truth or tru~hs. when all the time human beings are
creating truths. If a person is not aware of the process by which he construcrs
the truths about the world, be wiU not only be confused but also disappointed.
Dispositioos can thus rum out to be a great source of confounding and delusion
(mr!a), unless one understands their function in the formiJlation of uuths (lattva). The tram,ators of Nagarjuna's text into Tibetan probably pcrceived the
direction of the argument in this chaptet and na.med it accordingy.
Note that the dduslon (mrrii) is produced, not by all the phenomena (saf'tla-
dharma) but ooly by the way in which these phenomena are put togecher
(sa'J'!Mkarott) for purposes of understanding (see Websters' Seventh Collegiate
Dictionary, 1965, p. 219 where delusion is defined as "self-deception concerning facts or situations"). That putting together is the function of dispositions
~S<l'f?tskara) . Hence, for Nilgarjuna, as it was fr the Buddha, if anything is to
contribute toward delusion that would noc be all phenomena (H1rve dhr1mi1iJ,),
ratbei: it would be all dispositions (sarve sa'J'!Mliii.ra'1).

220

2.

THE Pl flLOSOPl-IY OF TifE MIODLE \'(/A Y

Tan rnriii mO!a-dharma yad yadi ki1?J tatra mufytlte,

etat tukta?(l bhagavat'ii IUnyata-paridipaka'f!'J.

II, whatever that is of deceptive nature is delusion, what is it about which

thece is deusion ? That too, namely, that which illuminates emptiness,
has been spoken of by the Blesscd One.
MKV(P) pp. 238-239; MKV(V} p.104.

If dispositions cause delusjons, what is 1t about which there arc delusions? The
answer would be: ''The world of e.xperience. " The Buddha bas spoken of thac
world coo. lt is the world that is non-substantial, is empty of any permanent
and eternal entity. All delusioos arise rega.rding that world which is
dependently arisen and non-substantial, but which is being uoderstood as being either eternal or absolutely unreal.

3. Bhav'iin'iif!l nil/svabha11at11a1(l anyatha- bhava-daran'ii f,

asvabh'iivo bhavo ,,,;Jsti bh'ii11'ii1iii'f!'l iinyat'ii yataf?.

Because of the perception of change, the absence of self-nature of existents is [recognized] . Because of the emptiness of existencs, there is no
existent without self-nature.
MKV(P) p . 240; hfKV(V} p .105

Tbis is a elear statement that truth or reaJ.ity (lattva) (there being no provisional
truth and ultim<tte reality) is neither substantial exisrence nor nihjlistic nonexistence . The perception of change or variacion (anya1habha11a) confirms the.
non-substantiaJity of phenomena (nzl1svabha11a). This is anothcr way of expressing the idea embodied in the discourse to Kaccayana that "ro hi.in who
percdvcs through right wisdom che cessation of the world as it has come co be,
the notion of existence (atthit'ii) in the world does not occur," (S 2. 17).
Cessacion (nirodha ) or change (anyathabhava) does not imply complete anruhilation. Hence Nagarjuna's view that there is no existent that is without
substance (a-svabhava), thac is, soroething that goes into complete oblivion
after Cl(tsting for a white (bhiitva prativigacchatt) leaving no trace at aJI. The

Ex.AMIN/\TJON OF ACTION AND 'Jlle. AGENT

22!

discourse to Kaccayana says that he who perceives arising (iamudaya) does not
hold on to the non-existence of the world . In other words, arising (Iamudaya or
.ramutpiida) contradicts nihjlistic non-existence (na.rtifii, a-.r11abha11a). Nagarjuna perccives this to be emptiness (!unyatfi) .
This, then, is the way in which "dependent arisiag" of phcnomena
(pra fitya.ramutpada) becomes a syoooym for "empciness'' (unyata) or "nonsubstantialicy" (nairatmya) whkh will be further elaborated in Chapter X.XIV.

4. Ka.rya .ryiid anyatha-bhava/J .rvabhav/J cen na vidyate,

ka.rya sy'iid anytha-bhavaf? .rvabhavo yadi vidyate.

Whose change would rhere be, if self-nature were not evident? Aga&n,
whose change would there be, if self-nature wete evident.?
MKV(P) p.24 1; MKV(V) p. 105 .

Nigarjuna is here conuasting identity and difference. If tbiogs are competely

different from one another, the n the re is no reason to $peak of the change of
things (ayathabhava). If, on the contrary, the.re were to be a substance
(.rvabhava) which is assumed to be permanent and eteroal, it could not change.

5.

Ta.ryaiva nanyatha-bhavo napy anyasyaiva yujyate,

yuvii na j'iryate ya.rmiid ya.rmiij/i"!O na jiryate.

Neither change of something in irself nor of someth.ing different is proper. The rea.son being that a youth does not age nor does an aged person
age.
MKV(P) . 241 ; 11r1KV(V) p. 106.

Nigarjuna cootioues to emphasize the view thac change (anyathabhava) is inexplicable in the contexr of identity or diffecence. "Of itself" ( ta.rya eva) means
"of somethffig chat has substantial exiscence;" "of anocher'' (anya.rya eva) implies " belonging co something completey differenc.'' As mentioned previously

222

TuE PH ILOSOPHY OF THE MIDDLE WAY

(VII.24), "decay" (jara) was not cejected by Nagarjuna. In the pcesent context,
what he intends to convey is that such decay makes no sense wheo applied to
explain a person who is metaphysically conceived either as possessing an eternal
self or as being dlfferent fcom moment to moment.

6. Tasya ced anyatha-bha11af? kfTram eva bh1J1Jed dadhz~

kftrad anyasya k.asyacid dadhi-bhavo bhavi[Jati.
If change were to be of sometbing in itself, then milk itself would be butter. Butter-ness would then be something other than milk.
MKV(P) p.242; MKV(V} p.106.

If change were to be applied to something recognized as exisiting in itself, i.e.,

asubstance, thcn the conception of change would be negated. Hece we find the
example of milk and butter (kflra-dadhi) utilized by the Vitsputriyas, along
with the metaphor of "fi.re and fuel" (Akb pp.432-433), to illustrate the relationship becween the aggregates and the self. If butter is considered to be
substantialy the same as milk, chen butter-ness wiU have to belong to
something different from milk. Otherwise we will be left with two differeot
substances having the same substantial constitution.

7.

Yady a:unyatrJ bhavet kt7?zcit syiic chunyam iti ktf?tcana,

na kitrJCid asty a:unyatrJ ca kutaft unyatrJ bhavzjy(lti.

lf there were to be something non-empty, there wou1d chen be

something called empty. However, there .is nothing d~at is non-empty.
How could there be something empty?
MKV(P) p .245; MKV(V) p. 107 .

The conception of "empciness" or "non-substantiality" is intended to eliminate

the belief in substance and attribute conceived in a me_raphysical sense.
However, if "emptiness" itself wece to be used in an attributive sense, that is as

'

EXAMINATI ON OF ACTION AND THE AGENT

223

a characteristic of something substantial, theo "em ptiness" itself becomes

"something" (ki'f!Jca114). A substaocial thing is a "non-empry-something''
(aUnya'f!' ki'l?Zcit). Such a thing does not exist. lf so, there cannot be something
called "empty'' (Unya1!l iti kitp.cana) .
This is the clearest warning from Nigarjuna against moving' towards the
mecaphysics of "empciness" (Unyata du!i. sec below).

8.

Unyalii sarva-dr1fina'f!J prokfii nil;saratJa'f!t jinail;,

ye1af!Z tu Unyafii-dr1fis tan asadhyan babha#re.

The Victorious Ones havc announced that entptiness is the relinquishio.g

of all vicws. Those who are possessed of the view of emptiness are said to
be incorrigible.
MKV(P) p.247; MA"V(V) p.1084.

The inevitable conclusion to be derived from XIII. 7 is that the conception of

"emptiness" (unyata) or "non-substaotiality" ( nairatmya), utilized by the
Buddha io order to free ooeself from all metaphysical views (dN!t), can turo out
to be. a n equa ly unsatisfactory view. if its application is to be carried beyood its
proper limits. lndeed , Nagarjuna perccives such activiry as leading to worse
forms of dogmatism.

CHAPTER
FOURTEEN
Examination of Assoc1atiQJJ
(Sa1!Jsarga-jJartka)

1.

DraftavJa'lt darana1,!t dra[{i.i Jiff!Y efiini dvio dviiaf?,

sarvaa ca na sa1,!tSargam anyonyena vrajanty uta.

ihe object ofseeing, the seeing and the seer-these threc do not function
in mutual association either in pairs or all together.

2.

Evaf?'l raga ca rakla! ca rafljaniya~ ca drfyafii~.

traidhena efiil{ klefi.i ca efany ayatanani ca.

Lust, the Iustful as weU as the object of ust shoud be scen in the same
way. The retrulining defilements as well ~s the remaining spheres of sense
should be seen in the tadic mocie.
MKV(P) pp.250 251; MKV(V) p.110.

The pragmatic theory i;if truth, that is, truth as somethiog put together according to human disp\lsitions (sa1,!tJkiira) depending upon something experienced
(dhamta), is not a very pala table one , especialy for some analytical philosopher
who wants to carry his analysis to the very extreme. In the present treatment of
Nagarjuna's philosophy, it has been repeatedly pointed out thar an extremist
analysis eft the Buddhist metaphysician with absolutely distioct entities. For
hlm (and this was the position accepted even by a philosopher like Hume),
1
'What is distinguishable ~ aso separable." Of coutse, these metaphysiciaos
would theo proceed to expl.in events in terms of"composition," of putting different entities rogether (sa1,!tSkaraf!a) accotc!ing to ooe's dispositions (.ra1,!tSkiira)
or, as Hume iosisted, in teons of one's imaginatioo. However, they will have to
cairy the burderi of explaining how only cettain things can be so puttogether
and not aoytfllog a.od everytbing. For example, one can insist thac it is possible
224

ExAM tNATION oi: Assoc1ATION

225

to bring togecher events such as the eye, color, and visual consciousness
togecher to produce the impression called "perception of color." Yet , one can
nor, either jo terms of dispositions or accordiog to any imaginarion, put
together the eye, sound, and guscatory consciousne~ and produce either a visual
im pression or an auditory im pression.
The ony way in which suci m etaphysicians can explain any possible association is by assuming a su bstancia re acion, an iohereot nature a.moog chose
events thac are so associated . This is how che anaJysis of events into abso u tely
different entities contributed co the recognition of mysterious subsrances. The
Sarvastivada notion of substance or self-nature (111abliii11a) was, therefore, an
inevitable answer to such extremist analysis, in the same way as Beruand
Russel's theory of relations, defined as neither mental or physical, was the
answer to the Humean analysis.
For such philosopbers, a pragmatic theory ofi truth, where truth is defined as
something "made" (ItJf!likrta), becomes a problem bccause their analysis has
deprived them of any empirical relations in terms of. which ch ings can be
associaced . ft is, therefore, not surpzing to see Nagarjuna raking up the quescion of associadon (saf!lsarga), in order to show chat ie does not work in the
background of the metaphysica assumpcions of certain ana ysts.
Thus 1t becomes necessary co keep in mind chat Nagarjuna's cricicism of
associacion is spcdfically rearcd co che associacion of eveots chat wece so
distinguished chat each was assumed co have its own nature (111obha11a). He
begins this chapter with a reference to the various categories he has alrcad y examined at the very oucset in this secdoo of the book, namely, seeing (darana),
the objcct of seciag (drQfavya) and the seer (drfltr) (Chapter III). XIV.2 refers
to another set of categories examined in Chapter Vl. This appica t ion is then
extended to aJJ occurreoces such as the defiements and facu lcics.

3.

Anyenanyasya sarruargai? tac canyatvaf!l na vi~yate,

drtl!{avya-probhrliniif!J yan na I IJT(Mflrgflf!l vrajanty ataJ?.

Association is of the mutually different [events). Such difference is not

evident in the ol:ijecrs of seeing, etc. Therefore, they do not function in
mutual association.
MKV(P) p.251; MKV(V) p.1 10 ,
Association, as mencioned above, becomes a philosophical p roblem only wheo
distinctions or differences are rendered absolute. Nagarjuna, basing himself o n

226

THE PHILOSOPHYOFTHE MIDDLE WAY

the pri ncipe of "dependent arising," insisrs chat such distinccions arc not
avai labe among objcccs of seeing, etc. lf these are discinguished or djffe1enciarcd in the way m etaphysicians do, then they cannot enjoy mutual harmony
or associaaon.

4. Na ca ke11alam anya111a1{Z drQ!fa11yader na 11irlya1e,

kasyacil kenacit iiirrlha'(J nanyatvam upapadyate.
lt is not only that the difference with regard to objects of seeing, etc. is

not evident; the possibility of something possessing difference jointly

with another is also not appropriate.
MKV(P) p.2)2 ; MKV(V) p. 110.

This is an interesting analysis of idenciry and difference. The dilemma of

substance is broughc out clea.rly in this analysis. ln order co relace things, differences need to be recognized. Once the differences a.re retognized as being
absoute, each thing is assumed ro have ics owo nacure. Ir becomes a subscaoce
different from any ocher substance. If cach subscance is ,differenc, ie cannot
have a substance that is shared by another. If it does .. the difference breaks
down. Eithcr rhere is differcnce or therc is idencity.

5. Anyarl anyat prafityfinyan niinyad anyad rte 'nya1af?,

yat prafitya ca yal lasmiit Jad anyan nopaparlyate.
Diffcrcnt things are de~ndc.nt upon cliffcrent things. Different things
arc not witbout different things. Because something depends upon
someching, a different thing is not appropriate.

6.

Yady anyad anyad anyasmat/ anyas11iiirl apy rte bha11et

tad anyad anyad anyasmiirl rte nasti ca nasty ataf?.

If a thing is different from anothcr bccause it arises from a different

thing, then it would ex.ist cven without that other thing. Howevcr, that

EXAMINA110N OF A SSOCIATION

227

other thing does not exist without the other, and therefore, it does not
exist.

7. N1inyasmin 11idyate 'nyat11am anany1111nit1 na vidyate,

a11idyamane C'iif!yatve nasty anyad va tad eva Va.

A diffcrence is not evident in reation to a diffcrent thing. Nor is it not

evident in a diffcrcnt thing. When diffcrence is not evident, rhcrc is
neither diHercnce nor ide.n tity.
MKV(P) pp. 252-255; MKV(V) pp. I I 1-112.

These verses seem to higblight the face that one cannot speak of dependence so
long as one .recognizes absolute difference among events. They arc a reminder
of the detaied treatment of the relacion of contingence (apek.fii) undenaken
previously in relation to the metaphor of "fire and fuel" (Chapter A.1.

8.

Na tena ltJJya I(Jf!MaTKO nanyen'iinya.sya yujyale,

saf?Zsr/yamiinaf!t saf!IS!l!af?Z saf?ZSr(ll{a ca na 11idyate.

The association of identical things or of diffcrent things is not propcr.

Neithcr the associating nor the associated nor even the agent of association is evident.
MKV(P) pp.255-256; MKV(V) pp. I 12-13.

ldencity and difference -assumed by the Buddhist metaphysicians cannot solve

the problem of truth, especiaJly its pragmacic version. The only solution
avai able to them is a recognicion of the Buddha's "omniscience" (saroajflatva),
which they unhesitatingly attributed to him, even without atcempting to
define what "omnis'' (saroaf?Z) stood for in the Buddhist context (see commentary on IX.3).

CHAPTER

FlFTEEN
Examination of Self-narure

(Svabhava-parzk.ra)

l.

Na Jt11!Jbhavaf? svabhavasya yuktal? pralyaya-hetubbif?,

hetu-pratyaya-sat!Jbhuta-1? .rvabhavaf? krtako bhavet.

The occurrence of self-oatnre through auscs and conditions is not proper. Sclf-narure that has occurrcd as a rcsut of causcs and conditions
would be something that is made.
MKV(P) p.259; MKV(V) p.1 14 .

Chaptcr XV is the condusioo to Part II of Nagarjuna's text. The maio thrust of

Pan 11, as cxplaincd in the Intioduction, is in the direccion of clarifyiag the
conccption of dharmas (in the plural), whcthcr they rcprescn tcd ideas, things,
events, or phenomena. As was evident from an examioacioo of the preceding
cwelve chapters, Nagarjuna's analysis of dharmas was intended to eliminatethc
metaphysical ideas relating to idenrity and d.iffereoce (ekiirtha-naniirlha). In
that proce~. he refuted the meraphysical notions of arisiog and ceasing
(utpada-nt'rodha) , of cternalism and annihilationism (Uivata-uccheda) and of
appearancc and dJsappearancc (agama- nirgama). These metaphysical ootions
werc the rc:sult of assuming a substancc or sclf-narure (svabhava) io
phcnomcna, an assumption that is muruaUy related by a conception of absolute
"otherncss" (parabliiiva).
Self-narurc or substance (111abhava) thus being the major ~ue , it is natura!
for Nigarjuna to concludc this scction with an cxamination of this particular
concepuon.
In rcfuting the conception of substance, Nigarjuoa relies heavily upon the
Buddha's own conception of a "middle position," nameJy, "dependent
arising." Because he was herc concecned maioly with refuting che mecaphysical
extremcs, Nigarjuna rcfers only to that section of his locus classicus (i.e the
discurse to Kityayana) that deals with the ewo extremes of exiscence (aslitva)
and non-existence (nastitva).
228

EXAMINATION OP SEl.l'-NAlURE

229

In the very first verse, Nigarjuna statcs his owo pragmatic view of truth as
something made (krtaka) depcncling upon causes and conditions (he111pra1yaya-sarpbhutaf?). subscance or sclf-nature, if it wece to ocist, couJd not
escape the principle of dependent arising (pralilya.Iamutpada) .

2. 'S11abha11af? krtako niima bhavi!Jali punal/ kathaf!I,

akrtn'maf! svabhavo hi nirapelqaf? paratra ca.

Again,
be a self-natute th2t is made? Indeed, an unmade self. how could there
.
nature is aso non;.contingent upon another.
MKV(P) pp.260-262; MKV(V) pp. 114-11 ~ -

.A.n artificial substance (krtakal? svabhival/) is not possible, for by defio.ition a

subscance is etcrnal and therefore not subject to acising and ceasing. Anyching
that is subjcct to arising and ccasing cannot be a substance and, bence, concrary
to dependent arising. lt is simply non-contingent (nirapelqaf?).

3. Kutafl s11abha11t1.Iyabha11e para-bhavo bhav#yatt;

svabhaval? para-bha11asya para-bhavo hi kathyate.
In the absence of self-nature, whence can there be other-nature? For, selfnaturc of othcr-nature is called other-nature.
MKV(P) pp.265-266; MKV(v,I p. 116.

This represeots a rcpetition of che argument used by Nigarjuna at I. 3 and

XIV .4 to indicate the rclativity of self-nature lld other-natu!C. lf one is not acceptable, the other too is not ~dmissible.

4. S1111bh'ii11a-para-bha11abhya1n rte bh'iivafl ku tal? punaf?,

svabhave para-bhave vii sali bhii110 hi sidhyati.

230

THE PHtLOSOPHY OF Tl-lE MIDDLE W AY

Without sef-nature and other-nature, whence can therc be an existent?

For, tbe existent is established only when there is self-nature or other
nature.
MKf/(P) p.266; M.k'V(V) p .ll6.

This probably could serve as evidence against the belief that Nagatjuna
recognized an ultimate reality beyond bothself-nature and other-nature. In the
case of the Yatslputriyas, the "ultimatdy real" emerges on the basis of an assertion of both idencity and difference, (see commentary on X.1, "Fire cannot be
designated as something differenc from the fuel, nor as something idencical.")
. The Japanese Buddhist scholar Y. Ueda perceives a unique logical principle
adopted in Madhyamika logic detived from the metaphor of "fire and fuel."
According to rum, ''There are inherent conditions in each sucb tbat their
ultimate rdatonship into a whole or unity entails a mutual dcnial of cach
other,'' (see Ioada, p.80, emphasis added). However , Nagarjuna is here raising the question; " Distinct from self-nature and other-nature, whence can
there be an existent?"
Thus, the eight negations are not intended to prove or establish the nature of
reali.ty, as it is oen and loudly asserted . They are primarily and solely intended
to eliminate metaphysical notions, not co characterize either an ultimate reali.ty
or dependent arising. Dependent arising is a totally differnt way of expressing
the truth or reality.
In the second statement above, Nagarjuna maintains that svtJbhav4 and
parabha11a are both dependen.t upon bhava. They represent a further bifurcation of bhava.

5. Bhavasya ced aprastddhir abhavo naiva sidhyatz;

bhavtJSya hy anyatha-bh411am abhavaf!l bruvate janiih.

When the existent is not established, the non-existent is also not
established. It is, indeed, the change of the existent that people generally
call the non-existent.
lef.KV(P) p.267; MKV( f1p. ll7 .

Whlle svabhava ' nd parabhiiva reprcsent a bifurcation of bhiiila, the atter is

EXAMINATION OP SELF-NATURE

231

itself dependent upon abha11a. Ordinarily peoplc speak of abliii11a as change in

bhava. The bha11a - abhava, chough moce comprehensive than che wabhava
-parabha11a dichotomy, carries the same implications as the lacter. Boch involve
the metaphysical nocions of idencity and di.fference.
Neither the Buddha's conceptioos of "dependent arising" (paf1cca.ramuppada) and "non-substantiality" (anatta) nor Nagarjuna's views on dependence
and emptiness (unyatii) should be understood as involving or crearing the
metaphysical notions of identity and di.fference. In face, the intcrpretation of
fUnyafii by some scholars as an "ultimate reality'' has brought about an immediate respoose from others who characterize it as "oothingness." The
dichotomy thac ordinary people assume is immediately fuought into play herc.
It was chis incorrigibilicy (aiiidhya) that Nagarjuoa was referring to at Xlll .8.
Hence. Nagarjuna's declaration that follows.

6. S11abhava1(' para-bhava?[I ca bha11t11(' cabhiivam eva ca,

ye pafyanti na payanli te tattvarri Buddha-fasane.

Those who pcrceive self-natutt as well as other-nature, existence as well as

non-existence, they do not perceive the tmth embodied in the Buddha's
message.
MKV(P) p .267; MKV(V) p .117.

lt is not merely self-oature and other-nature that arc rejected, but also existence
and oon-existence. The former pair covers a limited range of explanation, compared to the moce comprehensive notioos of ex:istence and non-cxistence.
An empirical definition of existence, as presented by the Buddha, would
mean some thing, some event, some phenomenon available to the sbc senses (see
S 4.15, Sabba-sulla). Such a phenomenoo is assumed to have come to be oo
che basis of cooditioos (pa.f1ccast1muppanna), to remain for a while showing
signs, ac the same time, of decay (!hita.rsa afJflathatta), and then cease to exist
(nirodha, 11yaya)1once agaio depeodiog upon conditioos. So thal even ordioary
unenlighcened people would say: "Change of what is existent is non:exiscence"
(bhiivasya hy anyathabhavam abha11a1(' bruvate janiil/, XV.5).
However, che metaphysicians can take over from this ordinary man's
language, especially with its use of the geoecive or possessive case (bhavasya).
He will assume that change is something possessed by the existenc (bhiiva)
which is always the same. Y et he cannot say the same about non-existence

232

THE PHLLOSOPHY OF THE MIDDLE WAY

(abhava). The metaphysicians part company here. one dogmaticaUy holding oo

to a theory of eternal existence (avatq), the other advocating absolute nonex.istence, which is annihilationism (uccheda).
Another metaphysician who is interested in explainiog the empirically fclt
"selfhood" and "others" wil attempt to bifurcate existence (bhava) as selfnature (svabhava) and otber-nanue (parabhava). Wheo applyiog this latter
dicbotomy to the explanation of causality, the metaphysician once again brl.ngs
the duali ty of existence and non-ex.istence into play.
Such metaphysics has no place whatsoever in the Buddha's explanation of existence and oon-existeoce. Hence Nigarjuna's statement that those who adhere
to these notions do not understand the truth or reality expressed in the
Buddha's message.

7. Katyayanavatiiide ca.stili n'iisliti cobhayaf?J.,

prati[iddhaf?J. bhagavat7i bliiiv'iibliiiva-vibhavina.
In the admonition to Katyayana, the two thcories Jimplying] 1exists' and
'does not exist' have been refuted by.the Blessed One who .is adept in existence as well as in non-existence.
MKV(P) p .269; MKV(V,) p. 11 7.

This, as mencioned in the Iouoduction, is the single most imporrant piece of

evidence avaiable in the work of Nigarjuna, which ca!1 rdate him to the Buddha as presented in the Pali Nikayas and the Ch.inese Agamas. Inada's note on
th.is versi:: is too brief and vague: "The Saoskrit Katyayaoavavada, either refers
to the sutra or the iosttuctions given to Katyayama [sic.) by the Buddha,"
(p.99).:Tfiis statement of Nagarjuna deserves much moce attention than has
ever beeo accorded to it.
The main theme of the dicourse is to expose th untenability of che two
metaphysical views .of existence (astitva) and oon-existence (ntistitva). This is
clone b.y appealing to the empiticaJ notions of arising (utpiida) and ceasing
(nirodha). With the fifteeo chapters (induding the present). Nigarjuna has
presented an outstanding explaoation of how the empirical conceptions of arising and ceasing, of irnpermanence and change;, can proficably be used to expose the futility of metaphysics. Hence, he is satisfied with merly referring to
that porrion of the "Discourse to Katyayana" whi~h deals wirh the rejection of

EJV.MlNATION OF SELP-NATIJRE

233

the two metaphysicaJ extremcs. A discussion of the remaining ideas expressed

by the Buddha in this discourse, especialy the positive description of the
human personalicy as well as its experiences , is reserved for a later occasion,

8.

Yady as1tt1111f!J pr11kr1ya syiin na bh1111ed asya nastilii,

prakrter anyatha-bha110 na hijatupapadyate.

If existcnce wece to be in terms of primal na tuce, then there wouJd not be

its non-existence. A change of pei mal nature is certainly not appropriate.
MKV(P) p.271 ; JHKJ/(V) pp. 118-1 19.

This and the oext three verses seem to constitute a digesr of the detailed and
meticulous analysis of the two extreme views presented by Nagarjuna so far .
lf existence isunderstood in the sense of primal narure (pr11kr11) , in the way
the Sakhya school clid, for. in fut, the Sakhya used the term 111abhii11a to refer
ro che primal oarure, then there cannot be its ooo-cxistcnce. The rea.son is that
change and primal nature or substance arc incompacible.

9.

Pf'flkrtau kasya casatyam anjathat11a1'{1 bha11ifya1i,

prakrtau kasya ca salyam anyatha111a1!J bha11i;yati.

When pei.mal nature is non-existent, whose change would there be?

When prima! nature is existent, whose change would there be?
MKV(P) p.271272; MKV(V) p.119.

Not only the existence of primal nature , but also its non-existence is iocompatible with change. Herc primal nature is understood as the substance and change
as the attribute. If the substance is nor available, the attributes cannot be applied to it. If the substance is present, the attributes become superficial . In
brief, the substantialist enterprise coosists of reconciling subscance and attribute aftcr creating a sharp and irreconciliable discinction between them (see
Chapter V).

234

TuB PHILOSOPHY Ol' 1'HE MLDPLE WAY

The above explanatioi1 would eliminace the need for assuming that the first
line of the verse represents a question raised by an opponent to which Nagarjuna gives his own reply in the second (see Inada p.99, who follows
CandrakTrti's interpreracion of this verse.)

10.

Asliti Saivata-graho n1islity uccheda-darfanat?J,

tasmiid astitva-nastitve naiiyeta vicak!at!tll/.

"Exists' implics grasping after eternalism. "Docs not exist" implics the
philosophy of annihilation. Therefore, a discerning person shouJd not rely upon either existence or non-existence.

11.

Asti yadd hi s11abha11ena na tan niistft fiivatat?J,

nasfidiintm abhut purva11z ity ucchedal; prasajyate.

"Whatever that exists in terms of self-nature, chat is not non-existent''

implies eternalism. "lt does not ex:ist now, but existed before" implies annihilation.
MKV(P) pp.272273; MKV(V) p.119.

These theories of existence and non-existence are not simple and harmless ones.
They contribuce to unforrunate consequeaces. The theory of existence kads to
the dogmacie grasping on to the belief in eternalism. The conception-of nonexistence leads to an equally dangerous view of annihilation, both o~which , as
will be pointed out later, are damaging to the moral life. Hence, a wise man
would not associate himself with such extreme views.
This seems to be a most appropriate condusion to an anaJysis chat was intended to establish che non-substantialicy of all phcnomcna (dharmanairatmya). Nagarjuna did not allow any toom for the reificarion of any one
single phenomenon that was refcrred to as being pan of human experience.

CHAPTER

SIXTEEN

Examination of Bondage and Release

(Bandhana-mok.fa-part k~a)

1.

Sa1p.skiiraf? sa'l!'saranti cen na nityal; sa1!'Jaranli te,

sa'l!'saranti ca ti4nityal/ sattve 'py e1a sam.aft kramaf?.

If it is assumed that dispositions transmigrate, they would not

transmigrate as permanent entities. Neither do they transmigrate as impermanent entities. This method (of analysis) is applicable even in the
case of a sentient being.
MKV(P) p.280; MKV(V) p .123.

Part Thrce, according to our analysis, consists of Chapter XVI-:XXVI; and is different from Part Two io its treatment of the subject matter, even though the
subject matter itself appears to be similar in them. While Part Two was concerned with the analysis of the elements of existence (dharma) showing how
t hey are lacking in any substance (dharma-11airalmya) and how tbey are
dependently arisen (parfitya.ram11tpanna), Part Three is concerned more with
the explanacion of the human personality (pudgala) without falling into
meraphysical uaps. The human persooality, both in bondage and in freedom ,
is analysed here. The problems of "self' (atman) , questioos regarding mora
responsibility and its associated concepts of time and fruitioning, are discussed
fust. Movfr1g therefrom, Nagarjuna takes up the problem of the person who
has anained f reedom, the question of rruths, of freedom itself, rrying to deal
ooce again with the metaphysica iotetpretations, until be reaches Chapter XXVI when he presencs the most positive explanation of that human p ersonality.
The present chapter beings with one of the most popularly held misconceptions about the Buddha's teachlogs pertaioing to rebinh (punabbhava) . Even
duriog the Buddha's day, when he spoke about rebirth being causally conditioned or ' 'depeodenty acisen" (patci;asamuppanna), and enumerated several
conditions that would contribute co ie (M 1.265), one of his disciples picked out
one among these conditions, namely, consciousoess (11ifJffiif!a), maintaining
235

236

T HE PHlLOSOPHY OF THE MI DDLE WAY

that it is "This very same coosciousness that transmigcates, not aoother" (:dam
eva viflfiifla?(I sandhavatt.' Ill1(1Saratt.' anaflfla?(I, M 1.256; Chung 54.2 [Tairho
1. 766c)). There certainly were many others, including some of his disciples,
who cootinued to uphold such views throughout che centuries. The widespread
p.tevalence of this view seems to indicate the adamant way in which people
believed that for survival to ta.kc place there must be a permanent and eternal
subscance:
The Buddha's answer co rhese believers in a permanent and eternal self or entity is that any factor that conrributes to human survival, whether ie is consciousoess (vififiafa) or disposition (sankhara), or action (kamma) or even
graspiog (upadiina), aJl these arc dcpendencly arisen. For the Buddha , contnuity can be explained in a more empiricaJ way by following the principle of
dependence of impermanenc fac cors of cxistence, where on leaves an impression on another, thus eliminating the need for assuming a permaoent entity.
Nigarjuna is herc referdng to rwo extremes, i.e., permanence (nitya) and
impermanence (anitya) , this latter being the momentary destruction (k!af!abhaga) advocated by the Buddhist metaphysicians. The. former cepresents the
Sarvastivada point of view; the lacter, the Sautrantika.
If the dispositions (Ia?(lskiira) arc presenred as being either pcrmanent or impermanent whcn thcy rransmigrate, and if there is no mention of causaJ conditioning of rhese dispositions as well as the ocher factocs, then the theories of
eternalism and annihilatiooism are inevitable. Furthermore, such excreme condusions arc especiaJly unavoidable when one factor or eotity is singled out and
shown to be the factor lnvolved in transmigratjon.

2. Pudgalaf? sa1(1Jaranti cel skandhayalanadhaltJ!U,

paflcadha mrgyama,,.o 'sau nasti kal? sa1(1Sari/yati.

lt may be assumcd tbat a person transmigrntes. Yet, such a person,

sought for i.o the fivefold way in the aggrcgates, spheres (of sense) and
elements, does not cx.ist. Who chen will transmigrate?
MKV(P) p.284; MKV(fl? p124.

lt is interesting to note that in the previous statement Nigarjuna rejects only

the view that dispositions cransmigrate. He did noc deny the disposicions
them$elves. However, in the present verse Nagarjuna maintains chat if a
transmigrating "person" (pudgaia) is sought for (mrgya11tano) in the ag-

EXAMINATION OF BoNDAGE ANO RELEASE

237

gregatcs, faculties, and elements (as Nagatjuoa tried to do in the last fifteen
chapters) one does not discovcr him. The "person" referrcd to bcre is no ordjnary person. This person shouJd possess the same characteristic which, according to the previous verse , made it impossible for the "dispositions" to
tranSmigcate, nam cl y, permanence .
!nada seems to miss the meaning of the term mrgyt1m'ii11a in his translacioo
of this verse.

3. Upadiinad up'iidiinaf!J sa'(lsaran 11ibha110 bha11et,

11ibha11a c'iinupadiinal/ kal? sa ki'(l Itlf!Jiari;yali.

Moving from one form of grasping to anotber, t~ere would be othcrbecoming. Who is this person who bas ceased to be and is [therefore]
non-grasping? Wherein does he transmigrate?
MKV(P) p .284; M.KV(V)
p. 124.
I

Uoderstanding the causa! process in a lioear way one runs into difficulties in expaining "grasping" (upadiina) as a reason, not only foc transmigration but also
for conceptualizing a person. White grasping was considered a.o impoctant
cause for the unhappiness and suffering (dukkha), as aJso the rebirtb of a
human bcing, "non-grasping" (anupadiina) was. a conditioo for happiness
(sukha) in this life and for not bcing reborn in a future life. In adclition, eveo
the vcry notion o'f a substantial "self' (alta) is supposed to be the result of
grasping on to the five aggregates (upadiinakkhandha).
However, if graspiog is singled out and explained in a lioear way. then moviog from one moment of grasping to another, one will be faced with otherbecoming (11ibha11a) . To explain this broken or inrerrupted series of graspings,
one. needs to assume that there is something to be grasped so that grasping can
continue. The: aggregates do not continuously provide a foundation for grasping. They arise and cease. With such arising and ceasing, grasping itself would
be ioter.rupted. This me.ans chat grasping that has come to be non-existent
(11ibha11a) would also be noo-grasping (anupadiina). If so, where is this socaled permanent cnticy and where does he transmigrate?

4. Saf'!J.Ikaiiif{iirp na n_iro'ii'arrt katha?!Jci'd upaparlyrrle,

sa1t11a.ryapi na niroar;a'!t katharp.cid upap11dyate.

238

THE PH!LOSOPl-IY OF THE MIDDLE WAY

The cessation of dispositions is somehow not appropriate. The cessation

even of a scntient being is also not appropate in a ny way.
MKV(P} p .288; MKV(V) p.126.

The Buddha left unanswered the quescion regarding the narure of a "freed person" (tath'iigata) aftec death (param1nara11a). Nagarjuna wi U return to this
question in his fina chapter, "The Examination of Views" (Dr1fi-pank1a). In
the present context, Nagarjuna is concecned mostly with the living person . The
previous verses referred to the mecaphysical views regarding a living person in
boodage. The present verse is, thercfore, devoted to an exarninatioo of the
mecaphysicaJ view percaining to a living person who has attained freedom.
Speakiog of chat fre.e dom, Nagarjuna doc:s not wanc co assume chat it is the
cessacion of dispositions, QI of a "person." He was probably aware that che Buddha spoke of the pacification of disposicions (sakhara-upasama) in relation to
a living person who has attained freedorn. H e was also aware that the Buddha
did not advocate the annihilation of a sentient being (1att11a, see commentary
of Xlll. l ). Heoce his present st~tement.

S. Na badhyanle na mucyanta udaya-vyaya-dharmit!af?,

.Iaf!IJliiiral? purvavat satto badhyate na na 1nucyate.

D ispositions that are of the nature of upsing and ceasing are neither
bound nor rcleased. A sentient being, like the foregoing, is neither
bound nor released .
MKV(P) p.290: MKV(V) p . 127.

According co the substanciaisc way of thinking , an erernaJ self or soul (at1na11)

is in bondage because it is bound co various ephimera1 factors such as che
psycho-physical personalicy (see. Bhaga11adgll'ii, Chapter Xlll). Such a self has to
break away from i-ts bondage in order to be free . H aving rejected a pcrmancnr
cntity Jike the self, if the Buddhisrs were to consider the disposicions as t he condition for bondage, such disposidons, being of the nature of arising and ceasing, could neither be bound nor freed . In ocher words. one canoot look at rhe
dispositions through the eyes of the substantialist. The same t.-an be said of a
sentient bcing (sanva).

239

EXAMINATION Ol' BONDAGE AND RELEASF.

6.

Bandhanaf?l cod uptirliinaf?l soptirliino na badhyate,

b11dhy11te n?inuparliinai/ kim avastho 'tha badhyate.

lf grasping were to be considered a bo.ndage, one who is with gcasping is

not being bound. Neither is one without gcasping being bound. A person in which state is then bou.nd?
MKV(P) 290; MKV(!'? p.127 .

The substance/anribute distinccion that emergcs from the statement such as

"one who is with grasping" (sa-uptiaano) militats against saying that he is being bound (b11t/hy11te). In this sense , the attribute is are.ady implicit in the
substance and there is no point in piling ,up another identical attribute on it. If
the substa0ce is without attribute (as in the case of an-upadiina), then there
.seems to be no way in which one can atuibute an attribute to it. They could
always remain independent. These difficulties relating to identi.ty and difference give rise to the questipn regarding the status of the person ..who is being
bound.

7,

Badhn'iylid bandhanaf?l kamaf?l bandhy'iit purvaf!J bhaved yadi,

na casti tal fe.fa'n uklaf!J g11v1yamana-gatiigalail/.

If it is assumed that bondage exists prior to the binding of that which is co

be bouod, that does not exist. The rest has been explained by [the
nalysis

of] present moviog, the moved and the not moved.

MKV(P) pp.291-292; MKV(V) pp 127-1?8.

lf, in answer to the question raised in the previow verse, it is said that bondage
exists prior to someone being bound, such bondage, according to Nagarjuna,
does not exist. The analysis in Chapter II as well as in Chapter X can be utilized
hete to refute the implications of a substantialist view. of bondage.

8.

Baddho na tntJcyate fiivad abaddho 114i11a mucyate,

s:f?itii'!l badtilhe mucyam?ine yugapad-bandha-mok/ane.

240

'fHE PHILOSOPHY OF THE

MJDDl..E W AY

One who is bound is not rcleased. nor is one who is o~t bound freed .
When there is rcleasing of one who is bound, then thcrc would be
simultaneous occurrcnce of bond.age and release.
M.KV(P) p.29) ; MKV(V) p.128.

One who is substantially bound (baadha) , i.e .. one who bas the sef-nature
(s11abhiiJ1a) of bondagc, cannot be frced . Similarly, it is meaningless to spcak of
somconc who is absoutely free (a-baddha). i.c . whose sef-naturc is frccdom,
as one beiag frccd . If one wcrc to speak of someoae who is idready bound and is
beiag freed , thca bondage and freedom would be simutancous.

9 Nirvasy'iimy 11nupadiino nir:iii1!1Zf(I me bh1111i./yati,

iti yefiif(J grahar tefiim upadiina-11111ha-grahaf?.
"Noi;i-graspiag, I sball be free. Freedom will then be mine." For whomsocver tbere is grasping in tbis manner, that will be a gigantic grasping.
MKV(P) p.295; MKV(V) p.129.

Nagarjuna is herc prescnting a fundamcntal idea expressed in the early

dtscounes (sec M 1. 145-151, Ratha11inl~sutta) as well as in the Praj1r\iparamita literaturc, espedally the Vajra&chediliii. It is the idea that one
Caruiot be frced and still eling to frecdom, let along braggiog about it.
Howcvcr, one nced not construe ie Buddha's statcment at M 1.171
1(Anyapariyesana-iutta) as an instance of such bragging. That statemcnt was
Jl)adc by the Buddha in responsc to a question raiscd by U paka at a time whco
. the only frced one in the world was the Buddha hirruelf. He was simply dcscrib/ ing to Upaka th~ abscnce of any one who bad attained freedom , in the sense in
which the Buddha understood the conccption of freedom. Whether the Buddha would make such a statcment subsequcnt to the prcacbing of the first scrJ mon and the attainmcnr of colightcnment and freedom by his fll~t fivc
disciplcs is extremdy doubtful.

ExAMINATION Of BONDAGE ANO.R!rLEASE

10.

241

N11 niNii'!11-s11miiropo 1111 sllf!ISt1rtip/Jkq11111111f!I,

yatr11 kas tatra Sllf[ISiiro niwii1111f!1 kif?Z 11ikalpy11te.

Wheccin there is neither the attribution of frecdom nor the elimination

of the life..process, what is it that is being discriminated as life-process or

as frcedom?
MKV(P) p.299; MKV(V) p.130.

This concluding verse provides a beautiful summary of the metapbysical views

reating to a life of bondage (st11[1Sir11) and frccdom (ninliif!t1) and can servc as a
key to unlocking the mystcries sucrounding the chaptcrs to.come. The constant
debates among modern scholars as to the implkacions ofNigatjuna's famous
chapter on Niroa11a (XXV,) tan easily be eliminatcd if we arc to keep in mind
the nature of the theories pertaining to bondage and freedom that Nagarjuna
h:od to deal with.

Presented in the substantialist moud, freedom becomes an attcibution

(samiiropa). while the life-process with its suffering requires elimination
(apak4r!tJ'!ll)_, No such frcedo~ or life proccss is accepted by Nagarjuna. Such a
description was too metaphysical for bim.
Attribution and elimination imply the existence of a neutral substance to
which freedom can be attt:ibuted or suung on to, while bondage in the form of
ife-process can be wrestled away from . If no neutral substance is recognized,
there couJd be two ocher ways of explaining freedom and bondage. First, it is
possible to say tbat the substance is inherenty free and that it is held in bondage by adventitious elements. Tuus, the Upani~adic or the Brahmaoical notion
of "self' which is pure and luminous is understood as something kept in bondage to the psycbophysical personality, like a sword kept in its sheath. The
originally pure mind (prakrti-pra6hiiwfll'a-~itta) of the Buddhist metaphysician
(sec Lank.avatiira, ed. Nanjio, 1956, p.358) resembles the Upan4aclic and
Brahmanical vie.ws of the ''self." Secondly, if such an originaUy pure encity is
not acceptablt:, then it is possi,ble to argue that wbat is called the life-process
(sa?!Mara) is completely annihilated and a completely new process of freedom is
initiated . The former has nQthing to do with the latter. lndeed, one cannot ig
nore the solutions offcred by the Buddhist me.taphysidans when they carne to
analyse che personal stream of becoming (bhava-sota) into discrete entities. Th
concepts of ''atcainment'' (praptt) and "non-attainment" (a-piiiptt) providcda

242

THE PHJLOSOPHY Of THE MIDDLE WAY

solution that is almost identical with those meotioned above. Niroa,,_a thus
becomes a prapti that fall into the stream (111a-Iaf!'t(ina-patita) and at that
point sa'!'Jara becomes at;i a-prap1i.
As pointed out in the lntroduction, the Buddha's conceptions of bondage
and freedom (and this woud also appJy to Nagarjuoa's views) have nothing to
do with aoy one of the alternacive explanations mentioned above.

CHAP'J'ER
SEVENTEEN

Examination of the Fruit of Action

(Karma-phala-partk!ti)

1.

Atma-sflr(tyamakartt cetafl paranugrahakaf?l ca yat,

maitrartt sa dht1rmaf! lad bl/af?l phalsya pretyt1 ceha ca.

Self-resti:aint as well as benefitting othets -this is the friendly way and it

constitutes the secd that bears fruit here as well as in the next life.
MKV(P) p .303; MKV(V) p. 132 .

The present chapter, unlike Chapter VIII , deals with the problem of mora
responsibility. It is an attempt to explain the "fruits" (phala) reaped as a result
of one's actions (karma). The docuine of the fruits of action or mora responsibilicy is pivotal to any explanation of human life , whether it is in bondage or
in freedom . However, in the present chapter, the idea of the accumulation of
merit and demerit (pu'!ya-papa) (for future benefit) is examined at length,
primarily because this parcicular idea is mostly associated with the Jife-process
(saf?Jsara) in .bondage. The Buddha insisted that such accumulatioo of merit
and demerit is abandoned (prahlf!a) by a person who is enjoying freedom (nirvaf!a), even though he does not transcend morals or is not unconcerned with
questions relating to mota! responsibility.
Inada assumes that verses 1-19 represent the popular explanation of karma.
This is questionable. Io fact, the popular, and therefore, a mistaken view of
karma is presented only in verses 6-12. Verse 13, as will be pointed out, refers
to a more sophisticated theocy of moral responsibility held by the Buddha and
his disciples.
The present verse deals with two importaot virtues-self-restraint and
benevolence-and these constitute the friendly way (maitraf?1 dharmafl) which
serves as the seed that fruitions herc as well as in the furure . !nada reads three
virtues-self-restraint, kindness towards others and benevolence. On the conuary, m(litrarrt seems co quality dharma, and Kumaraj1va undetstood it in this
latter sense.

243

244

ThB PHILOSOPHY OF THE MIDDLE

WAY

2. Cekina celayi111ii ca karmoklaf{l p11rama-r11f!'ii,

1asy'iineka11idho bhethl/ RaN1lllf!l1Q parikrrtitaf?.
The Spteme Ascetic has said tbat action is volition as well as volitio.nal.
Many distinct varieties of that action have aso been expounded.
MVK(P)

pp.30~- 306 ;

MKV(V) p.133.

At A 3.415 (Chung 27.5 (Tairho l.600a]), the Buddha idcntified karma witb
"volition" arid maintaincd I.bat kar-ia, whcther it b_c bodily, vcrbal, or mentaJ,
is to be ra:ognizcd as karma if it is volitional (Cetana 'haf?l bhikkhave kammaf?Z
11adiimi. Cetayil11'ii kllmmaf{l karoti kayena t1acaya manasii ca). ihe emphasis
oo volitjon was intendcd to eiminate the wrong belief that a person is responsible for any and every action be pcrforms, a view that was advocated by the Jain.a
thinkers of pte-Buddhist India (sec Kalupahana, Causality, pp.125- 126). The
distinct varieties of karma referred to herc arc the one's preseoted in the
canonical Adhidharma, and these arc based upon the discussions availabe in
the early discowses.

3. Tait" yac cetanety uklaf{l karma miinasaf{l smrtaf{l,

cetayitva ca yat tuk.taf{l tat tu kayik<1-11acikaf{l.
Herein, what is called volition is J"eminisced as menral action. Whacever is
called volitiot1al consists of the bodily and verbal.
MKV(P) p. 306; MKV(V) p.133.

A difference is noticcable between the Buddha's owo explanarioo of karma in

the statement from Aguttara quoced ac XVII.2 and the present description of
Nigarjuna. While cetana or volition is dcfinicely meotal , che Buddha seems to
assume that not all mental actioos are volitiooal. Hence his statement that all
, three forms of kar,na, bodily, verbal and menta , can be determined by volition. However, in the presen.t statcment, volition seems to have been identified
with mentaJ action, the voitional being confined strictly co bodily and vetbal.
This laner view may be a reflexion of the Buddba's own staccment at M 2.25
(Chung 47.2 [Tairho 1. 720)), whcrcio both bodiy and verbal actioos ace con
sidered co havc mind as a basis.

Exi.MINATION OP THE FRUIT OF ACTION

4.

Viig fli/Pando 'viratayo yli ca11ijntiptisaf!Z}lilii'1,

avijlaptaya e11'iinya'1 smrlii viratayt1.S taiha.

5.

Paribhoganvayaf!Z p11yam apuyaf!Z ca trJthavidha.f!Z,

cetana ceti saptaite dharma!? karmiijanal; smrtal;.

245

Whatever words and deeds that arc wociated with dclight and
dcsignated as nonintimation, and also tbosc others rctniniscod as nonintimation, but arc associated with non-delight; si milady, merit as well as
demerit consequent upon enjoymcnt, and finally , voliti?n - these arc
reminisced as the scvcn things that arc productive of action.

MKV(P) p.307; MKV(V,} p.133.

Some of the ter~ used here to define the various forms of karma appcar for the
first time in the Abhidharma (sec Akb iv.1-2; Akb pp.162 164; Adv
pp .118-119). They arc absent in the eady discourses. However, takcn in
themseves, they do not seem to crcate any philopsophical problems. The
philosophical problcms arose because of the manner in which these actions
wece interpreter/. These interpretations are then taken up for ex:aminacion by
Nigarjuna in the following verses.

6.

Ti!thaty apaka-k'ii/iic cet kt11'm4 tan nityafiim iy'iit,

niruddhatp cen nirudrlhaf!l- sat ki'f!Z pha/af!Z jpnayqyati.

lf it is assumcd that action remains during the time it is maturing, then it

will approach permanence. If it is assumed to have ccased, thcrt having
ceased, how can it produce a fruit?
MKV(P) p.311; MKV(V,} p .134 .

This is actually the point at which Nagarjuna begins his analysis of the
metaphysital assumptioos. Hece he immediately turns on to the theories of
idencity and difference. The assumption taken up for criticism in the :first line
needs to be examined carefully. UodoubtedJy, it is the problem ofpotentilJI ex-

246

THE PH!LOSOPHY OF rnE MIDDLE W AY

istcncc. Such cxistencc may be undcrsrood in various ways. Potcntialiry may be

takco as the cxistence of conditions that would eventually givc rise co some particular cvcnt. Nagarjuoa, as may be scen in a moment, does not sccm to objcct
to such an explanacion. Oo the contrary, if potentiality is undcrstood in the
sense of substantial existcncc of the panicuJar.evcnt (in this case, karma), even
when it has not marurcd or come to fmitioo, thcn that assumption lcads to
ctemalism. This lattcr view is cenainly not acccptabc to Nagarjuna. It also
leads to the contrary view, that is, if the evcoc is complcccly abscnt (and this
absence pertains to the very conditions that will cvenruaJJy give cise to the
cvent), chen thcre will be doubts about the occurrence of the cvcnt at all. This
is annihilationism. Thus, Nagarjuna is not dcoying the fruit of action but only
the manocr in whicb it is explaincd.

7.

Yo 'nkura-prabhrtir bl1at sa?r)tano 'bhipra11artate,

tata!? pha/af?Z rte b'ijat sa ca nabhipra11artate.

Wbatcvcr serics that begins with a sprout p rocccds from a sccd, and thcn
produccs a fruit. Howcver, without a sccd, such [a sccics] would not pro-

cced.
MKV(P) p.312; MKV(V,J p . 13) .

The mecaphysical assumptions of the Sautraotik.a doctrine of karma are under

review herc. The atomistic view of the life-process acccpted by the Sautrantikas
compelled them to analyse an event into a series (saf(Jt'iina) of moments. Their
major difficuty lay in explaining how one series (e,g. , a sprout-series) comes to
be tied up with another preceding scries (e.g., a seed-series), since thcy are different. If they are radicaly different, theo the sprout-series CUl occur cven in the
absence of the seed -series. Nagarjuna maintains chat thls docs not happeo. Io
other words, he is insisting chat philosophers like the Sautrantikas will have to
acccpc the view tbat no evcnt can come in co being unless therc wcre condicions
tbat give rise co ie; in this case it is the see_d -series.

8.

Bi/ac ca yasmiit saf{Jtanaf? rtZf?Zlaniic ca phalodbhavaf?,

brja-puroaf?Z phalaf?Z tasmiin nocchinnaf?Z napi f(i'l/ataf(J.

EXAbilNATION OF TI-IE FRUIT OF AcnoN

247

Since a series arises Crom a sd and a fruit ariscs from a series, a fruit t.hat is
prcceded by a seed is, thereforc, neither inteuupted nor etcrnal.
MKV(P) p.3 13; M.k'V(V) p. 135.

Lo answer to the question raised by Ni\garjuna, a Sautrantika can i:espood tbus:

lnstead of conceiving of these as two differeot series, et us say that a series is
produced from a seed. This series then gives rise to the effect (whk h is the
sprour). ln that sense, a fruit is preceded by a seed (btja-purvarp phalaftl).
Understood in this way, one does nor fall into either the eternaistic or the annibilationist view.

9.

Yas tasmlic cittt1-Jt1ftlfiinflf cetaso 'bhipmvartate,

lata? phaltlftl rte citfiit sa ca n'iibhipravarlate.

1 berefore, wbatever thought-series therc is, that proceeds from a thought

and ftom that fruit. That thought series would not proceed without a
thought.

10,

Citfiic ca yasmlit SflftJfiinaf? IflftJfiinac ca phalodbhaval?,

karma-purvaftl phalaftl tasmlin nocchinnaf!J napi ifvatrw.z.

Since a continuous series arises from thought and from the continuous
.series the uprising of \l fruit, the fruit that is preccded by action is neithcr
intcrrupted nor etemal.
MKV(P) p .313-314; MKV(V) p . 135.

The Sautrantika continues: Similary, a though series (cilta-sam,fiina) cmerges

from a chought (cela.raf?). From that se.ries arises the fruit. Thus, without a
thought, the fruit does not come to be. The thought followed by a thoughtseries thus gives rise to the fruit . The fruit is thus preceded by a thought which
is its cause. & such, it is neither permaoent nor interrupted.
The causa! connection envisaged by the Sautriintikas abovc is simple
antecedence.

248

11.

THE P !ill.OSOPHY OPIBE MIDPLli

w/.Y

DharmtHya iiidhanoplijiil/ 11klii'1 karma-path'ii ritJSa,

phalt11?1 kiimagu'!il/ paca dharmasya prety(I ceha ca.

The ten pure paths of action are the means of achieving good. The five
smnds of setlSC pleasutc represent the fruit of good, here as well as in the
next life.
MKV(P) p.314; MKV(V) p.136.

Following the causal pattern aid down above, namely, antecedence, the
Sautrantika would explain the tenfold path of action (karma-patha) as beiog
initiated by volition (cetan'il) which puts it together (tatliiibhisa7?1skilraal, A.kb
p.248). Volition WU$ becomes the karma, and the series of accions, bodily and
verbal (k'iiya-1iik), dctermined by that volition, becomes the vchicle of action"
(karma,,af ca pantliiinaf?, ibid.). The pure bodily and vernl actioos would
then be the mcans by which good is achieved (dharmasy11 fiidhanop'iiyiif?). The
five strands of pleasurc to be cnjoyed in a "heavenly" life (svarg11), cithcr herc or
in the next world, wouJd be the fruit of the god voJition. This is the manncr in
which the Sauttantlkas otplained the Buddha1s notloo of karma and its effect.

12. Bahavai ca mahantaf ca daiiil; syur api ka/pana,

yady eia tena naiflai. fii kalpanatropapadyate.

If there were ro be such a thought, there would be many a gren error.

Therefore, sucha thought is not appropriate hece.
llf.](J/(P) p.316; llf.](J/(V) p.136

Nagarjuna is not impressed by such an explanatioo. He perceivcs many and

substantial errors in such conceprualizations. Hence he considers them to be inappcopriate.
Inada's translacion, once again, skips an cxtremdy important qualific.ation
made by Niigarjuna. "If concepruaJi.zations arc permittcd, there will arise many
as weU as great errors," (p.107). The implicarion woud be that Nigarjuna rejccts all conc'Cptualizations. However, this is not the case. Niigatjuna is very specific
in bis reference when he says: e1a kalpa!Ui, "these concepcualizations," where

ExAMINATION OF THE FRUIT OP ACI'ION

249

"these'' refers to the prcceding conceptualizations or explanations. Indeed, it is

for this rcason that Nagarjuna can turo .around and speak of a more appropria te
thought or con~ptuaHzation in the next verse.

13.

lmiif!' puna!? pravqk/yami kalpanaf!t yatra yojyate,

buddhail, pratyeka-buddhai ca fravakai canuvaf'T!ittlf!'.

Moreovcr, I shall expou.nd the following thougbt which is appropriate

a.nd which has bcen extolled by the Buddhas, the self-enlightened ones
'
'
and the disciplcs.

MKV(P) p.317; MKV(V) p.13.

The preseot statement by Nagarjuna should.servc as an antidote to most of the
misunderstandings that have prevailed so far regarding his views about thought
and languagc. Nagarjuna is about to explain in no unclear terms a more appropriate thought or conceptualization (ka/pana}, a right thought (samyaksaf!'kalpa), a right view or perception (samyag-duft) relacing to karma and its
fruit (phala). It is one that is extolled not only by the Buddha, but also by his
disciples (fravaka) and the self-enlightened ones (pratyeka--buddha). If Niigarjuna had recognized a linguistically transcendent ttuth or reality, he could not
have made the above statement.

14.

Pattrat'(t yatha 'vlprar{iias tatha-r1111m iva karma ca,

ca1t1TT1idho dhatulai? sa prakrtya 'vyakt;taf ca sal,.

Li.kc an imperishable promissory note, so is debr as well as action. It is

fourfold in terms of rcalms and indeterminate in terms of primal nature.
Jrfl(V(P) p.317; jij}(l/(V) p.1)7

Herc, a .debt and karma are compared to an imperishable promissory note. The
metaphor is significant and needs to be carefully cicamined. It is used by
Nagarjuna to ilustratc the doctrine of karma as described in one of the most

250

THE PHII.OSOPHY OF THE MJDDLE WAY

popular and authoricative scacements in the Indian Buddhist craditioo. Two

centurics ater, Vasubandhu wrote a whole treatise - Karmasit/dhtprakara,_,a - attempting to explain this statement. Candrakirci quotes it
cwice in his commentary (pp.324,390). The fact that he does not quote iratthis
point, where it seems to be most relevant, indicates thac he was using it to
illuscrate a competely different point (sec bclow XVII.2 1). The staccment runs
thus: "Karmas do not pcrish cvcn after hundreds of millioos of aeons. Rcaching
the harmony of conditions and the appropriatc time, they producc conscqucnces for human beings."
The first statement, taken in isolation, will convey the impression that the
Buddhist theory of karma is detcrministic in an extreme sense. However, the second scatement provides sufficient qualificacions to take the dctcrminist sting
away. What seems co have compelled !nada to assume that herc, "With equal
force he condemns any idea of an indestructible continuing action (avipraafakarma) which gives the sense of continuiry or transirion in man's cvcryday life
and.deeds" (Niigarjuna, p.104), is his failure to cvaluate the first statement in
the light of the qualifications provided in the second statement, In the prcsent
chapter Nagarjuna is simply expaining the fust statement, i.e., "karmas do
not perish" (na pra,.,afyanfi karma'!t). His analysis, at this point, is confined to
it. He leavcs the second Statemeot to be exarruned in two ocher chaptcrs that
follow. After examiniog what an imperishable karma is, N:igarjuna wants to
keep any soul-theory out of the way, and this he docs with Chaptcr XVIIl . And
from thcrc, he immediately gets down to anayse the conteots of the second
statemeot by compiing two chapters: (i) Cbapter XIX on the "Examination of
Time" (Kii/a-pank.iii) and (ii) Chapter XX on the "Examination of Harmony"
(samagn-pankfii). As such, it would be inappropriate to come to any definite
concusions regardiog the contents of this chapter uoril the three followiog
chapters arc carefully cxamined. However, Nagarjuna's statement at XVIIl.13
thar he "will state thi's moce appropriatc view" (imaf!l kalpanaf!J pravak.$yami
yatra yojyate), which he thcn attributes to the Buddhas, his disciples, and the
Pratyeka-buddhas would certainJy seem to indicatc chat he is presenring an acceptabJe view, tather than one that should be rejected.
Furthermore, having made such a strong statement indicacing chat he is
presentiog "this'' (imiitp) appropriate view, if Nagarjuna were to follow it up
with a theory that he is condemoing, one will need to think twice before considcring Nagarjuna to be a second Bucdha.
To return to the metaphor of the prornissory notc (patra) that one signs when
borrowing money- this metaphor being Nagarjuna's owo - he is not speaking
of a permanent and eternal promissory note , but something that wilJ remain so
long as it is not redeemed. As long as a promissory noce is preserved, and u~css
one were to honor ooc's obligations, one will eveorualy, dcpeoding upon time

EXAMIN/\TION OF THli FRUIT OF ACTION

251

and cooditions, have to face the consequcnc;cs. The imperishability of the promissory note may aJso mean that even if th'c promissory note is descroyed there
is an obligation on onc's part to honor such an agreement.
If', in order to account for such an obligation, one wcre to assumc a substanda1 oature (prakrti = wabhava) in that act, Nagarjuna's reponse is chac sucb a
nacure is i.not determinate" or is "inexplicablc" (avyakr1a), an :11.nswer that the
Buddha himself gave when qucstioncd about metaphysical issues (sec Chapter
XXVII).
Finally, in terms of the realms in which the consequences may be reaped.
such actions can be foutfold. Candarkini rcfers to the fourfold realms as (i) the
sphere of sensuality (kiimiivacara), (ii) the sphere of materiaity (rupivacara),
(iii) the sphere of the formless-{ariipavacara) , and (iv) the state of freedom or
absence of influxes (aniifrtWa) .

15.

Praha,,,10 na prr1heyo bhavana-heya eva va,

tauniid avipra11a1ena 1ay111e kaf'lfJa1'1if!' phala'f!J.

That [i.e., the impcrisbablc karma] would not be relinquishcd by simplc

relinquishing. lt is to be relinquished on1y through cultivation. Thus,
through the imperishabe ;arises the fruit of action.

16. Prahaataf? praheyaf! ryat karmaf!af! sa1t1kramea 11a,

yadi do1af! prasajyeraf!IJ tatra karma-1111dhiyal/.
If it is to be relinquisbed through simple relinquishing or througb the
transformation of action, thcn therc would follow a vaety of errors soch
as the destructi0n of actions.
MKV(P) pp.319-320; MKV(V) p.137-138.

When spealcing of imperishable karma, naturally the question can be raised as

to how it can be gottern rid of. Is it possible to nullify the effect, say. of a bad
karma by simply not doing it again (prahaaJo praheyo)? The theory of prtipti
("attainment") and aprapti ("non-attairunent'') may impy such a sicuation.

2)2

THE PHlLOSOPHY OF THE MIODLE WAY

Preventing the momenwy sueam of life from appropriacing a bad karma

would mean the nulification of the effects of aH previous karmas. Nigarjuna
perccivcs this to be a negatioo of the docuine of karma (leaf'fllll-11adha). He
therefore insists that the abandoning of the fruits of karma c.a n be aclUeved , not
through simple abandonment, but through constant practice (bhiivanii), i.e.
constant performance or promotion of good accions and the constant avoidance
of cvil actins (cp. sabbapiipassa akargf!tlf'!Z kusfllassa upmampadii, D 2.49;
Dhp 183).

17.

Saf'IJefiif'!Z vi/abhiiganiit?J sabhiginaf!Z ca kArma'fl'iif!Z,

pratUaf!ldhau sadhatiinii'?I ekA utpadyate tu saf?.

Of all thcse accions, whether dissimilar or similar, bclonging to certain

realms, only one would arise at the moment of birth [of a being] .
MKV(P) p.321; MKV(V) p. 138.

Even though this statement of Nigacjuna is in perfect conformity with the

Buddha's owo explanation of the doctrine of karma, it may come as a surprise
to most Nigarjunian scholars, cspecially because it confliccs with most opinions
exp~d about Nigatjuna's philosophy. Herc is an cxplanation of rebirth (.punarbhavt1) cxamined in reation to past karma. In the eyes of most scholars, Nagar
juna could nevec make a statemeot Jike this. Herc, karma is presented a~ the
connecting link between two lives. lt is one of a myriad of karmas an individual may have petformed , whether they be similar or dissimilar. The face
chat only one (elea) among such actioos of a life-time may appear at the time of
the conception of a being (pratis'tl~ht) and which can influence the new lifeprocess is acknowledged by Nagarjuna. He could not have been unaware of the
statement. of the Buddha that consciousness (viflfii11a which is inextricably
bound up with volitional karma) was a possible connecting link between two
lives (D .3.105; Chang 12.2 [Taitho L 77bJ) and also of the ernphasis placed by
his fellow Buddhists on the last thought of the dying person (culi-citta) as hav
ing influence over a new life-process (pafisandhi-cilta, sec detailed discussion
at VbhA 15 5-160). Without falling back upon a mecaphysical theory of
moments, as some of the Abhidharma interpreters did, Nagacjuna is hece
recognizing the dependence of rebirth (pratisam,dht) oo at least one previous
karma. lt is a similar cecognition chat made the Buddh.a declare: "Beings have
karma as their own, karma as inheritance, karma as the source, karma as kin. Ir

ExAMJNATION OP 'fHE

FRuJT OF AcnoN

253

is karma that distinguishes human beings, for example, as inferior and

superior," (kammassa~a . .. satt'ii 'kammadiiy'iidii kammayoni kammabandhu,
kammaf!' salte 11abhaj<1ti yatlidatrt hln.appailllfiiya, M 3.203; Chung 44.1
[Tairho 1. 704c]). Unfortunately, Buddhaghosa's explanation, though retaining
the i.d eas expre~scd by the Buddha as well as Nigarjuna, is marreci by a thcory
of moments (see VbhA 156). Just as much as mcmory is bcing described by
most psychologistsas being "owned" . withouthavin~toassume its permanence,
bere we find a person's karma bcing perceived as something "owned'' by bim.

18.

Karma~111?

karman.o drrte dharma utpadyate tu .rab.,

d11i-prak'iirasya sarvasya 11ipakve pi ca tif/hati.

That [imperishable] arises in the present life, corresponcing to all the actlons having dual natures [similat and dissimilar, good and bad, etc.] and
stays so even when matured.
MKV(P) p . 321.; MKV(V) p.138.

A further explo.-atioo of the Buddha's docuine of karma is continued bere. The

phrase Jute dharme is a sanskritizatio.n of Pali dt{tha dhamma, which itself can
be traced back co drrta-Janman, meaning "the present life. " The fruitioning of
kanna .into good and bad consequences is admitted hece.

19.

Phalll~11jalikrafft4d 11a

sa marrJ'!ad va nirudhyate,
anafravatrt I tilra11aftt ca vibhagaf!' tatra lakrayet.

Tbat [imperishable} ceases as result of theintetruption of the fruit or as~

result of death. lierein, a clistinction betwcen one with influxes and the
one without inflm:es is to be signified.
MKV(P) p.322; MKV(V) p . L38.

The so-caJled imperishable action (avipraf!afa-karma) cao terminate as a result

of two 'evcnts: (i) the interruption of the fruit(phala-vyatikrama) , or ((ii) death

254

TuE PHlLOSOPHY OF THE M IDDLE

'iY/ AY

of the individual. ln the case of the latter, it is necessary to remtmber what was
said at XVII.17 , i.e., that all accioos performed during a lifetime are not continued . Only one of the myriads of actions performed can domioate the last
thought: moment of a person. This wouJd mean that many other karmas, even
though all of them have not matured, may become nulified at the time of
deat~.

20.

Unyata ca na cocched4// sa'!l-s'iiraf ca na iivata'!I-,

karma'o 'vipraTJaaf ca

dharmo buddhena defitaf?.

Emptiness, however, .is not annihilarion; life-pcocess is.also not eternal;

the imperisbability is of acrion-such is the doctrine taught by .he Bud-

dha.
MKV(P) p ,322; MKV(V} 138.

If what is said before is an approptiate explaoation of karma, then Nagatjuna

can maintain that "emptiness!' (JUnyalii) does not mean "annihilatioo" (uccheda). At the same time he can maintain that the life-process (saf'!JJara) is not
a permaoent and etemal (fafvata) process. In such a context, an irnpcrishabe
action (avipraf!aa-karma) simply means the continuity of that life-process conditioned by kanna until some of these karmas bear ftu.it or ~e lost on the way,
while others ike the threads of a web can contioue to iilfluence the future lifeprocess. In any case, the eotire process is one of depeodence- dependence
upon a whole composfre of factors .
Nagatjuna has no hesitacion in attributing such a doctrine of karma and personal idenity to the Buddha himself artd praising it as the Buddha-wotd, even
though the atuibution of such a doctrine co Nagarjuoa would be unacceptable
to some of the dassicaJ and modern foUowers of N""agarjuna himself.

21.

Ki11'f!la notpadyate kasmiit nil{svabhavaf?t yai(lI tata!?,

yasmac ca tad anutpannaf!Z na tasmiid vipraayati.

EXAMIN.<\TION 01' THE FRUIT OF ACTION

255

Why does action not a.rise? Because it is without self-bature. Since it is

non-arisen, it docs not perish.

MKV(P) pp .323-324; MKV(Jl) p.139.

Havjng <:oncluded the- explanatioo of the more appropriate view .of karma as
advocated by the Buddha, Nigarjuna does not leave the discussioo without
taldng a look at the possible metaphysical interptetacions or any m.i.sunderstanding of th.is doctrine. He has already spoken of the arising (utpaa'a)ofkar.nia at
the moment of rebirth (pratisaf!U/hi). That arising is understood iii rlacion to
the principle of dependent arising (prafityasamutpada). The i.mperishability
(avipra'fifa) is introduced in order to explain the continuity in the fruitioning
of action. It is not intended as a justification for the belief in a permanent and
eternal substance (svabhava). However, some of the Buddhists did utilize a notion of substance to account for the functioning of karma. It is this particular
notioa of substaace that is taken up for analysis.
If karma is "non-substantial" (nil/svabhava) in the way the Sautrantikas
understood it, i.e ., without any perceivable cootinuity, but ooly as something
that is continually inter.rupted, then the arising of such karma cannot be explained. If arising cannot be accounted for in such a metaphysical way , neither
can cessation be admitted. Imperishability (avi'prar;afa), as explained by
Nigarjuna, becomes the only othet alternative.
Unfortunatly, Candrakrni, who favored a rather absolutistic interpretation
of Nigarjuna (sec comments on the Dedicatory Verses), utilizes the conception
of inlpecishability in order to deny any form of arising. It is because he had such
.. an interpretation in mi'n d that he quotes the famous ve.rse elucidaciog karma at
this .point rather thao utilizing itwhen the i.mpecishabiliry was first meotioaed
by Nigatjuna at XVII.14. As has been shown alieady, Nagarjuna was n~t
critical ofany and every form of arising or ceasing. He was only rejecting the>
metaphysical ideas.

22.

Karma svabhavata cet sy?i chaftlata'f(l syad a.ra'J'!Zaya'f(l,

tJktttJT(l ca bhavet k.armt1 knyate nrz hi iifvafaf?t.

lf itis aSSwned that actio.o comes to be from self-nature, i-t certainly will

256

THE PHILOSOPHY OF THE MIDDLE WAY

be etccnal, and action would a1so be uncaused, for that which is crernal

is, indeed, not caused.

MKV(P) p .324; MKV(V) p.139.

If the arising of karma cannot be accounted for by following tbe Sautrantika

mcthod, can it be explained in terms of the Sarvastivada conception? Nagarjuna's answer is negative. He seems to know with great cenainty (asaf!Zfayao/,)
that the Sarvastivada solution docs not work. It implies permanence and as a
result karma would appear to be "un-done" or "uncreated" (akrta).

23.

Akrliibhyiigama-bh11y11f!Z syat karmiikrtakaf!Z yadi,

abrahmacarya-11asai ~a do!as tatra prasajyate.

lf an action werc not pecformed [by the individual], then there woud be
fcar ofbeing confronted by something not perfocmed [by him] . Anignoblc life as wcll as error would follow from this.
MKV(P) p .325; MKV(V) p . 140.
If actions were to be something noc performed, then a person would be
.b aunted by the fear (bhay11) or anxiety that he bas no hand in the organization
of his own lifc-process. Fears and anxieties, according to Na~rjuna's view, arc
one's own creations. Deny one's owo rcsponsibility, one does not have to justify
a life of maral purifY (brahmacarya-vasa). This is a elear and uncquivocal assercion of personal res}onsibility for onc's owo pufity and defilement.

24.

Vyavahara virudhyante saroa eva na sa1?1fayal/,

puf}ya-papa-krlor naiva pravibh~gai ~ yujyate.

Uadbubtedly, all conventions would tben be contradictcd. The distinction benveen the_performance of merit and evil will aJso not be proper.
MKV(P) p .325 ; MKV(V) p .140.

Ext.MINATION OF 'THE FR\JIT OF A cnoN

257

With the cejection of a life of mora purity resulting from the deoial of human
responsibility for act~ons, all mota conventions like merit and demedt (puyapapa) would be reodered 1Peaningkss. lt is interescing to not e thac Nagarjuna
is not referring to the good-bad (dharmt1-adhamz.a or kusala-akusala) distinc
tion , for this atter distincc~n is differctnt from the former in regard co ics value
as a convencion (11ya11ahar~). This is compacible with the Buddba's own explanation of 'karma. On 'pragmatic grounds, the Buddha recognized the
merit/ deme.cit distinction. lt was helpful in leading theordinary people coward
accepting a mora life. However, the notions of merit and demerit coud be
harmful in che long run. especialy because it involves the idea of accumuation
of merit for the sake of future enjoyment and pleasure. Funhermore 1 it is easily
assoc.iated with che notions of reward and punishment, a notion that the Buddha was not willing to relate to the doctrine of karma (see M 1.373; Chung 32. l
[Tairho l.628b]). As such, the Buddha insisted that a morally perfected person.
would eiminatc: the desire to accumuJate merit or demerit (punna-papapaht11a, Dhp 39; samiliitii paliiiya punnapiipaf!I, Sn 540). Yet, it does not mean
that suc~ a person also abandons the good/ bad (kusala-ak_usala) distinction
which is the very basis of merit and demerit. lod~ed , a morally perfect person is
expected to promote good, while eliminating evil or uomed.torious actioits (see
above XVU.l , 17). lt may be for this reason that even a ater Mahayana
philosophir like Dogen deemed it approprate to compile a whole treatise on
this subject.

25. Tad vipakva-vipiikaf!I ca punar ~va 11ipak1yatt;

karma 11yavasthita1!J yasmiit tasmat s11abha11ika1'!J yadi.

If actio.n wc:re to be detcrminc:d, because it possesses self-oature, thcn a
matury that has maturcd will again maturc.
MKV(P) p.326; MKV(V) p.140.

If an action wc:re to take pace without being performed by someone

(akrtakaf!I), theo it will occur on its owo. It will possess its own oature
(s11abhii11a). Such a substantiaJ action will have its own consequences (111jiiik11)
inhereot in it. In that case_, what is being described as the fruitining of kanna.,
namdy, a mar_iifestation of its consequences, wouJd merely be a re-fruitiooing.

258

THE PHILOSOPHY Ol'nfE

MrooL.E WAY

This is the same sort of criticism that Nigarjuna made of self-causation (st1atautpa1Jt) in Chapter I.

26. Karma klefiitmaka11J cedafl.J te ca kleii na tall11ataf),

na cet te tat111atal/ klefiil/ karma sjiit 1att11atal; kathaf{'.
lf thls action is associatcd with defilements, these dc6.1emcnts, in tum,
arc not found in themelves. lf deftlements arc not in themselvcs, .how
coud there be an action in itself?
MKV(P) p .236; MKV(V) p . 140.

The substantialist explanation of karma presented at XVII. 25 would lead to the

distinccion bctwcen karma and its quality or attribute. Qualities refcrred to as
dcfilements (klea), etc.. would be merdy incidental. A karma can then make
ddilement " ies owo" (kleiat11111ka) or it can be freed. from defilements (nil/kleiluif. Such a subscantialisc perspective, as mentioned so often by Nagarjuna, would rendcr the defilements unreal (na tatt11atal/). cspecially because
thcy come and go, arise and pass away, and hence without self-oacure. Karma,
in sucha case would be substantial, and the acuibutes non-substancial. Nigarjuna, the empiricist, sees no way in which sucb a substancial karma, divorced
from the acuibutes, can be identified.

27 .

KaTYTlll klefas ca dehiinaf{' pratyayif? samudiihrtal?,

kamtti klem ca te Junya ya_d i dehetu ka katha.

Action and dc61cmcnts arc specificd as the conditions of the [different)

bodies. Howcver, if these actions and defilemcnts arc empty, what could
be said about the bodies?
MKV.(P) p.327; MKV(V) p . 141.

Herc Nigarjuna moves on to a rugher generality. Botb karma and defilements

_arc geoerally considered to be the conditions that decermine the iodividual. As

ExAMINATION OF 1HE

FRurr OF ACTION

2~9

such, karma and defilments become attributes of the personaJity (de/.a) . In the
ptevious verse , Nigarjuna questioned the substantial reality of both karma and
defilmcots. lf they arc cmpty of such reality, what could be said about the per
sooaJity itself?

28.

A11idya-ni11r10 jantus tr111a-saf!J:fojanaf ca sal?,

sa bhoklii sa ca na kartur anyo na ca sa eva saf?.

A sentient being, beclouded by ignorance, ualso fettcred by craving. As

an experienccr, he is neither idcntica With nor different from the agent.
MKV(P) p .328; MKV(V) p.141.

The Buddha's discourse to Acela-Kassapa, the discoursc which served as the

fouodation for Nigarjuna1s treatment of "suffering" (Chapter Xll . Dukkhapanlqii), rejected the thcories ofself-causacion and exteroaJ causation of sufft.cing. Rejecting self-causation, the Buddha maintained: "Kassapa, co say that 'a
person acts and he himself expedences the consequences,' wbcre self-caused
suffering belongs to one . who has existed from the bcginning, implies etcrnalism" (so karali JO pafisaf(Jvediyaff ti kho KaJsapa adito salo sayaf(JRalafP
dukkhanti iii vadaf!J sassalaf!J elaf(I pareli, S 2.20; Tsa 12.20 [Taisho 2.86a]).
On the contrary, "To say thac 'one acts and another expcriences the conscqucoccs,' where the suffering causcd by another belongs to one who has been
afflicted with pain, implies annihilationism" (afllo karoli afllo
pafisaf!l.vediyaff ti .. . vedanabhitu'!'!fZSsa salo paraf(1kataf(1 dukkhanti iii
vada~ ucchedaf!# elQf(J pareti, ibid.). lt is inte resting to note that in the former
case, the Buddha refers to the bclief in a bcing who existed from the bcginning
(adilo salo), an idea that is generally considercd both by the Buddha and by
Nigarjuna as contributing to a belief in a permancnt entity (sec Chapter IX on
Piirv11-pank1ii).
In the present verse, Nigarjuna is faithfully following the Buddha'~ owo
argument to reject the identity as well as differeoce bctween a doer and an cxpcriencer. A person who believes either in ideotity or in difference is looked
upon a someone who is beclou'ded by ignorance (avidyii) and craving (trf'!ii).

29.

NQ pratyaya-samutpannaf(J niipratyaya-samullhitaf!J,
aJti ymf!iiitl idaf!J k.armll tasm41 kartipi niisty aJa.

260

T HE PHILOSOPHY Of THE MIDDLE W AY

Since this action does not exist as asen from a condition nor as issuing
forth from a non-condition, evco an agent does not exist.
MKV(P) p.328; MKV(V) p.141.

The causa! cxplanations of karma offcred by the metaphysidans, oamcly, selfcausation, cxternal causation, etc., or the ooo-causal cxplanations arc not accep~blc . If karma itscf cannot be expained in thls way. it would be futile to
attcmpt any such explanations of an agent of karma. Note the use of the term
i"'1fl.J (this) to ccfer to wma, similar to the use of the term efii at XVII.6, thus
spccifying the typc of explanation that is rcjected by hirn. Por this rcasoo, we
prefer to confine his cdticism only to the mctaphysical views mcntioned in the
verses immediatey preceding (XVII.21-28), lcaving the moce appropriate view
he mentioocd untouched. Tbis secms to be the only way in which one can
rccognize consistency in Nagarjuna's statements throughout this chapter.

30. Kllf1'/'la cen niisti karfii Cli llukll/ syi1 hlrmaja'f!' phalat?J,
maty 111ha phale bhokti kuta eva 'bh1111i1y11ti.
lf both action and agent are non-existent, whett could therc be the fruit
boro of action? When there is no fruit, where can therc be an ex
pericncu?
MKV(P) p.329; MKV(v,) p .141.

In the absence of either an aciton or an agent metaphysically coaccived , there

could be oo fruit or consequcoce boro of such action (hlrmafat?J phala?!Z) Herc
again, it is not a denial of fruit or consequcnce born of action, but only of those
that arc bora of such action as explained previously. If the (ruit or consequence
is not obtaiocd, its cxperiencer (bhoRfii) would also not be appropriate.

31.

Y111ha nirmilt1ka'f!' 1'iistli nirmimlta rddhi-111mpadii

nirmito nirmifliitlinya?!Z sa ca nirmitakal/ punal/,

EXAMINATION OF THI! FRUIT OF A cnoN

261

Just as a teacher, through psycho-kinetic power, were to create a figurc,

and tbis created figure were to create anothe.r, tbat in turn woud be a
created.

32 .

Tathi nirmitakaral; karla yat karma taf kr:tam.,

Ml yalha nirmiteniinyo nirmito nirmitas tatha.

In the same way, an agent is like a crcated form and his action is ike' his
creation. lt is like the created form created by another who is cr~ted.
MKV(P) p.330; MKV(Vj p . 142.

One metaphysical view lcads. to anothcr, that to a further mctaphysical view.

Such is the W1CAcling citdc. A metaphysical view is, indced fabricated by som~
one. Yet the fact that sucha mctaphysical view turns 9 ut to be empty does
not mean that the cxpcricncc dcpcnding upon which the metaphysical view
was formulated or the proccss of conception arc thcmselvcs non-existent. Expcriencc as well as concept arc availablc. Only that the conception is carricd
bcyond itS limits to assumc the existence of independent entities, .whcther they
be atman or svabhiiva. To understand the significancc of Nagarjuna's statement h~re it would be nc:cesary to cake a look at one of the rarc statemcnts of
the .Buddha recorded in the Siimafiaphala-.ruttanta (D 1. 76-77). In this
passage, which explains the fruit of redusesbip, the Buddha refcrs to two forms
of knowledge a contemplacive could develop before he directs his a~tention to
the so-called higher forms of knowledge (abhififiii}. The first is dcscribed as
fu~M:

.
0

With hiS mind thus sercne, made pute, translucent, cultured,

devoid of evl, supple, ready to act, firm and imperturbable ..he ap.
plies and bends down his mind to that knowledgc . and insight
whereby he gtasps the fact: "This body of mine has form, it. is built
up of the four clements, it springs from mother and father, is continually rencwed by so much ~oiled rice and .juicy foods, itS very
nature.is impcrmanence, it is subject to erosioo, ab~asion, clissou
tion, and clisintegration; and therein lics this consciousoess of
mine, too, boun1d up on that it oes depeod.

262

THE PHILOSOPHY OFTiE MIDDLE WAY

This represenu the Buddha's expeence and cooceptioo of his owo personality.
Having stated this, he i.mmediately proceeds to examine whether there is
aoytliing beyood this experience and cooceptioo. What he describes is no mote
than the funciton of man(IS, the faculcy which is responsible for the formulacion
of the metaphysical ideas of self (atman) and substance (s11abha1111). The statement reads:
With his mind thu serene, . .. firm and impercurbable, he applies and beods down his mind to the calling up of a men tal image.
He calls up from this body aoother body, having form, made of
mind (mano-maya), having all (his own body's) limbs and pans,
not deprived of any organ.

This is how manas becomes a sensus communis surveying the sensory fields of
other faculties (M 1. 295) and maiotaining a feeling of identity, which then gets
converte< to a permanent and eternal iitman. Could it not be pssible that
Nigarjuna was aware of the above statement of the Buddba? What could be
the 'differe.oce between the cteation of a non-existeilt form thtough the power
of psychok.inesis and th generation of a belief in a permanent and eternal self
tbough the activities of manas? It seems that human beings, with fa:culties
through which they arc able to perceive and conceive, a.re not the objects of
Nigatjuna's critidsm. lt is only the manner in which perceptions and.conceptions arc formulated that is under criticism.

33. Kiesa/; kaT'11'Ji11i deha.f ca karliiraf ca phaliini ca,

gandhaf"lla-111Jgarakara mtJiici-.wapna--sa??Jnibhal;.
Defilements, actioos and bodies, agents as well as fruits, all these aie similar to
the cities of the gandharvas, arc comparable to mirages and dreams.
MKV(P) p.334; MKV(V) p.143.

The metaphors usec at the end of Chapter VIl to illustrate the natutc of
m~tapiysical theories pe~taining to arising, stasis, and ceasing, arc employed
here to elucidate the character of similarly conceived theories relating to
defilements, actions, personalities, agents, and conscquences . .

CHAPTER
EIGHIEEN

Examination of Self

(Atma-par"ik.fa)

1.

Alma skandhii yadi bhaved udaya-vyaya-bhiig bha11et,

skandhebhyo 'nyo yadi bhaved bhaved askandha-lakfa'laf?.

If the self wcrc to be idcntical with the aggreptcs, it will .pattakc of upr:ising and ccasing. If it wcrc to be diffcrent from the aggrcptcs, it would
havc the charactcristics of the non-aggreptcs.
MKV(P) p.341 ; MKV(fl? p . 145.
The conception of an individual self (atman) was ' previously cxamincd in a
varicty of contc:xts. The prcscnt anaysis was occasioned by a occd to cxplain the
lifc-process as cooditioned by human actions (karma), the subjcct mattcr of the
preccding cfuptcr.

The Buddha's analysis of the human personaity into five aggrcgates (paflcllkkhandha) was intended to show that underlyiog the empirie.al factors constituting the human pcrsooality there is oo pcrrnaneot and eteroal self. The
Buddha's view was that chcsc five factors servcd .as the basis for any conceptuaization of a self or soul. Hence thcy arc aways referrcd to as aggregatcs of
grasping (upiidiinakkhandha) .
Nagarjuna bcgios his investigatioo into the conception of self by raising two
questions pertaining to the nature of the self, espccially in .its rcatiooship to
the fivc aggrcgates. If the self andthc aggrcgatcs werc identical, thcn the self
would havc to paru.kc of the characteristics of the aggregatcs. Thcsc latter bcing subjecc to arising and ccasing, the self cannot remain pcrmanent and etcrnal . On the contrary, if the self and the aggregates wece to be diffcrent, theo
the former could not have the $arrIC characteristics as the latter. Leaving the
argument at that, Nagarjuna is allowing the rcadcrs to come to thcir OWll conclusions.
So far Nagarjuna has not givcn any indication that he tecognizcs a spccia intuicivc faculty througb which one can sec beyond the world of change and impcrmancnce. Indecd, all that be has admiucd points to his i:ccognition of sense
263

264

THE PHILOSOPHY OP THE M IDDLE WAY

c:xpcrienccs as the foundation of human knowledge. The impcrmanent aggregates constitute not only the human personality, but also its experiences. If
the self is considered to be different from the aggregates, Nagarjuna is herc implying that it is unknowable, not merely inconceivable, for it will not have any
of the characceristics of the aggregates chat are all that we know through sense
cxpcnencc.

2.

Atmany 41ah ciitniiyaf!l kuta e11a bha11i.fyati,

nirmamo nirahaf!lkaraf? samiid iitmiitmaninayof?.

Jn the abscnce of a self, how can there be something tbat beJongs to the
self? From the appeasement of the modes of self and self-hood, one abstains from acating the notions of "minc'' and "I."
MKV(P) p . 34~ -347; MJ(V(V) pp.147 148.

If a permancnt entity does not exist, one cannot assumc the existence of
anything that bdongs to ie. The deniaJ of a permancnt entity does not mean
that Nigarjuna is committed to a rejcction of self-awareocss or selfconsiousncss . The rejcction of the lattec would undcrmine the very foundation
of his cpistemology. As mcntioned carlier (sec lll.17), Nagarjuna, following
the Buddha, recognized consciousness (and this includes sclf-awareness), not as
a preexistent cogiJo, buc as part of the human personaity condicioncd by facrors such as the sense organs and the objects of perception. Sclf-awarcncss or
self.instinct can be pursued to its extremc limit. The result is che "construction
of a self' (ahaf!l + kara) , which eventualy leads to the belicf in permanence.
The other cxueme is the complete rejetton of any form ofselfawareness, which
is tantarnount-to annihilation.
Without falling into these two exttemcs, Nigarjuna is herc prcsenting the
Buddha's owo "middle way" philosophy when he speaks of the appeasement
(f11ma), not the complere cradication , of the "self' (at.man) and ''whatever that
pertains to a self" (itmant) .. This is pecfectly in accord with the "appeasement
of dispositions" (1~kiiropaiama), the "appea5ement of the object"
(drf11(t111yopaama), ~nd the "appeasement of obsessions" (prapaficopafama),
discussed eatliec(see commentary on V.8).
Through the appeasement of the self-instinct one eliminates the
met2physical nod ons of a self (atman ), and through the appeasement of the
o bject (drtlf'WJopaam11) one is ablc to realize the non-substantiality of

26~

E:JW.lNATlON OF SELF

phenomcna and would not eling to them~ "ooe's own" ('iitmani). These
culminate in the absence of self1Shness (ni"'1'1ma) and the absence of egoism
(niraha'f(lkara).

The dual meaning of the term 11h11'f(l/dira isworth noting. Ah11f?Jkir11 (literaJly. "1-making") mcans both ego-centcredness and pride. )Vhile the term
aha'l.'kira has come to be so popula,r in ordinary languagc, it is interestingto
note that oo such term is construcied with the pluraJ of ahaf!I, namely, maya'f(I,
as ""'Jllf!IUra ("we-making1') would have expressed an idea which is eq'ually
unacceptable to the Buddha and Nigarjuna, for they were not willing to
e~te th~ notion of oneself (aha'f!I) altgether in favor of an equ~lly
mc~physical notion of a "social self" ,

3. Nirmamo nirahaf!luro ya ca .ro 'Pi na 11idya1e1

nirmama?'(J niraha'f(lk'iiraf!Z yal/ payati na payati.
Whosoever is free from selfisbness and egoism, he too is not evident.
Whoever perceivcs someone as lree from selfisbness and egoism, he too
does not eerceive.

MKV(P) p. 348; MKV(V) p . 148.

Nigarjuoa approvingly spoke of the appeasement of the ootioo of self and th~
conscquent elimination of sef1Shness and egoism. However, koowiog his contemporaries who were so prone to metaphysical speculacions, he was not willing
to rest satisfied with such a statement.
As a remind~r to those who have not achieved the "appeasement of the nocion of self," Nigjlrjuna pointS out' that someone who is assumed to havc got
ten rid of egoism 'and pride is als not available . The constant attempt by the
metaphysicians to reify things, entities, persons, etc. was kept in mind by
Nigarjuna whenever he roakes any positive assertion .

4. Me.mety aham iti '-!Tt!ebahirdhidhjiitmam e~a ca1

nirudhyata upadiinattz lat k!afiijjanmanal; R!ayal;.

Wben views pertuning to "mine" and "I" , whether

they arc assodated

266

THE

PHllOSOPHY Of rnE MIDDLE

WAY

with the interna!._ or the exterml, have wancd, ' chen grasping comes to
cease. With the waning of that [grasping], there is wani.ag of birth.

MKV(P) p.349; MKV(V) p.149.

The usc of the iii formula as mama-iii and 11ham-ili in the prcscoc scatcment
makcs it sigoific:antly differcnt &om the founcr statcrncnt at XVlll.2. In the
former statement, Nigarjuna spoke of 111m11n (self) and 'iilm"iy11 (that which
bclongs to the self) and cmphasizcd the nccd ro appcasc such awar~ncss or f~el
ing. I the prescnt verse, he refers-to t1h11m ("I") and mt1ma ("mine") using the
iii-formula and insists upon thcir complete elimination. (.,r11jt1) . Thus, it is not
che fact of self~awarencss that causes problems for the human beings but the
theorizing bascd upon such sclf-awarencss. The cogito may thus ruro out to be
harmless, so long as it is considered to be a product of the sensory process (sec
m. 7), but ergo sum is what is dangerous, episcemoJogically as wcU as ethically.
When such thcorizing has waned (kf.i11e), t}?cn there is cessarion of gtasping
(upadin11-nirodh11), which is frecdom whilc living. The cessation of grasping
cvcntually lcads to the waning of rebirth (j1111man11f? k,rayaf?).

5. Ka1"f1Ul-kle111--k1ayan mok,ra karma-kiera vikalpataf?,

te prapacat prapalcas tu filnyat'iiy'iif{i nirudhyate.

On the waning of dc6Jemen~ of action, therc is relcasc. Defilemcnts of
action bclong to one who discriminatcs, and thesc in turo result from
obsession. Obsession, in ics tuto, ccases within the context of emptiness.

MKV(P) pp.349-350; MKV(V) p.149.

The face that this statement of Nigatjuna immediately foUows his criticism of
theorizing relating to "mine" and "I," that is, speculation relatlng to subject
and object, becomes very valuable in deterroioing the meaning of the cerm
flWp11, a term that has caused much confusion and misunderst:anding.
Vikalpa can mean two differenc types of discriminatioo. One is the type of
discrimination made at the pheoomenaJ. levcl. lt is the discrimioation referred
to at XVIII.2, a discrimination that 'is empiricaJ y grounded , but which should
be kcpt under control (fama). The second ~ the type of discrimi02tion made at
a mecaphysical Jevcl. lt is the discrimination referrcd to at XVIII .4, which has

ExAMJNATION OF Sm.F

267

gone far beyond the cmpirica levcl and thus become theorecica or speculative.
For Nigarjuna, che defilements f action foJow from the atter, not from the
former. It scems that this is the primary reasoo for his rcfcrence ro and criticism
of 11ikalpa at XVIIl.5 rathcr than at XVIIl.3.
Wrong actions (mithya-karman111). i.e actions chat are defilcd, emanate
from wrong beliefs (milhy1i-dr1/I), which arc the results of wrong thoughts or
discriminations (mithya-sa~kalpa, mithjii 11ikalp11). Obsessions arc- the inevitable tesults of such wrong discr.iminations. When- such obsesioos arc appcased , then a person does not get involved either in a ootioo of a permanent
self or in a theory of complete annihilacioo. The reaizacion that self-awareness
is dcpeodently arisen (IlL7) is a realization that it is empcy of a permanenc
subsrance (svabhava-iunya). This latter is the middle path that avoids eter
naism and annihilationism.

6.

Atmety qpi prajflapitam an'iitmety (lpi deiita~,

buddhair niitrnii na caniitmii kaicid ity api deii.ta~.

The Buddha's havc malce known the conception of self and taught the
doctrine of no-self. At the same time, they have not spoken of somethiiig
as the self or as the non-self.
AfKV(P) p.355; MKV(V/ p.152

If the distinctions made in XVIII.2 and 4 are not recognized, it is nacuraJ for the
interprecers of Nigarjuna to run into diffiucies in explaining the contents of
tl:iis verse. Hece agin we have che use of the iti-formula, this time used with the
terms alman and anatman, as atma-iti and anatma-iti. However, the differeocc bctween the Buddha's specu ations and those of the metapbysicians in
this regard is thac the Buddha does not speculate on any ena ty (kaicid) as
alman OC as anatman. In other words, he does not reify either atman or
anatman. Reifying atman one eods up in eternaJism; reifying anatman one is
led to annihiJationism. Withouc resorcing to such reificacion, the Buddha has
indicaced the meaning of atman. (atmetz'prajnapitaf!I) and has spoken of the
implications of anatman (aniitmeti deiita~) . Both atman and an1ilman are explained by the Buddha in terms of dependent arising (.prafityasamutpada) .
This doctrine of dependent arising eliminates the need for posrulating either
permaoence or annihilation.

268

THE PHILOSOPHY OF THE MIDDLE WAY

As such, the distinccion that I nada attempcs to make bctween prajflapilaf!J as

"provisionally empoyed" and defitarrz. as "instructed '' becomes untenabe. 1f
" provisiona explanation" is to be achieved by the use of words, it r.ould not be
different from "instruction." Both need the use of language. Just as much as
the Buddha could instruct on the "true idea of anatfnlln," he could have instructed on the rrue idea of atman, without falling into mctaphysics in either
case. If cheuse of the cmpirical terms can be ucilized in one case, thece is no
reasoo to assume any difficucy in using them in the ocher. The problem then is
not with regard to language as such. but only in regard to the way in which it is
used .

7. Nivrttam abhidh'iita11yatp ni11(t1e ctlta-gocare,

anutpann'iiniruddh'ii hi nir11'iil!am iva dharmatii.
When the sphere of though w ceased, that which is to be dcsignated aJso w
ceased. Lilce freedom, the oature of things is non-arisen and non-ceased.
MKV(P) p.364; MKV(V) p .154.

"Whereof thought has ceased, thereof speech also has come to

cease."

Abhidhiilllllya meaos "that which is to be dcsignatcd." It rcfecs to the' Jb~Jd

of objects. lt ~ the same as abhidheya which, if we follow Nigarjuoa's roethod
of exposition, is mutuay relatcd to abhidhana, "designation." Bot&
abhidheya and abhidhana would thus cease to be aong w1th the ccssacioo of
the sphece of thought (citta-gocara).
"Wbeceof one canoot spea.k, thereof one must be silent."
lt is possible for the incerpreters of Nigatjuna to assume that herc is the idcotification of the "unspca.kabe" with "cmptiness" (Unyafii), generaly cefecred
to iq negative terms as "non-arisen, non-ceascd" (anutp(lnnaniruddha), which
is then identificd with both "true oature" (riharmat"ii) and frcedom ~nin.iii11a).
This is the easy route to the beief in the so-calcd "non-conccptua" (nir
vikalpa) ultiroate reality (paramiirtha, tallva).
Howcver, the fust lice of Nagarjuna's scatcment should prcvent anyone Erom

269

EXAM1NATION OP Sm..F

reaching such a conclusion. Nowhere has Nigarjuna asserted any form of

knowedge that transcenps the six senscs. For him, what is perceivable is aso
conceivable. He may, reflecting upon the conceptualization of the metaphysicians, admit that what is not perceivable is also conceivable . Yet, it woud be
improper to assume that he will recognize the non-conceptual (11ir-11ikalpa) as
being perccivable. Conceptualiziog and concciving arc not two different activicies. Hence his statement: "When the realm of thought has ceased, that
which is to be dcsignated also has ceascd.
As pointed out in the inuoduction, a similar statemcnt was made by the
Buddha in relation co a tathagata who has passed away, and the present context, in which the mctaphysical nocion of s~lf (atman) is discussed , is no more

differcnt from that in which the Buddha made that statement.

So far as Nigarjuna's analysis is concerned; it has become elear that his ncgations pcrtained primarily to m etaphysics, whether it be the nocion of a pcrmanent and eternaJ self (atman) or a substance (wabhava). Along with the negation of a permanent anq eternaJ self, Nigarjuna also rcjected absolute
"otherness" (para-bhava). Existence and noo-existence, in this metaphysicaJ
sense , wece rcjccted by him. As pointed out in the analysis of the Dcdicatory
Verscs, the "non-arising" (anutpada) and "non-ceasing'' (anirodha) wcre intendcd as criticisms of ruch m etaphysics. If one rccognizcs a substance, Nigarjuna would say it is non-ceasing: if one rocognizcs annihilation, Nagarjuna
would characterizc is as non-arising. In other words, if absolute arising and absolute ceasing wece to be acceptcd, thcsc would ncgate empirical arising and
ceasing whicb is the basis of "dependent arising" (pralityasamutpada). W hcn
both "the way of phcnomeoa" (dhtl1'mtzli1) and frccdom (nirva11a), which arc
expiaincd k terms of "dependent arising'' (pralityasamutjiida), arc referred to
as "non-arisen and non-ceasing", it is mor~ appropriatc to assume that herc
they arc to be distinguishcd from absolute arising and absoute ceasing. Arisiog
and ceasing in an absolutc sense represcot inappropriatc conceptualizations.
Neither che nature of phenomcna nor frccdom should be the subject of such
inappropliate conceptualizations.

8.

Saf'l!atrz tathya1'(J na va tathyaf?'l tathya1'(J c/jtathyam eva ca,

nai11atathya1'(J naiva tathyam etad buddhanufiisanaf!J.

Evcrythi.ng is such, not such, both such and not such, and oeither such
.and not such: this is the Buddha's admonition.
MKV(P) p.369:

MKV(~

p . 157.

270

THE PHllOSOl'HY OF 'Il-IE MIDDLE

W.AY

On the basis of this statement, Inada, following his predecessors in the interpretation of Nigarjuna, rcachcd the conclusion that "truth is non-relationaJ,
non-descripti:ve, non-diffcrential. ... it is thatness or thusness," (p.113).
Refercing specifically to XVIlI.8, he says 'that herc "Nigarjunaintroduces the
famed Four-corncred Logic, , , . i.e the pssible conditions of is, is not, both
is and is not, and ncither is nor is not, in order to exhibit the fact that fina!
truth uanscends all these possibilities; it is fUnyllt'ii per se" (ibid.)
o far, our analysis ofNagarjuna's statcmcntS has failed to reveaJ any form of
four-cornered logic that he uscd to establish an ultimate truth. Whenever he
ucilized it, he did s i.o order to rejcct metaphysicaJ assumpcions, rather than to
establish something or some theory.
Mter srating the fact that metaphysical views, especially those relating to a
tr1thagata after death, take us beyond the sphere. of thought
(citta-gocara)(XVUI. 7), Nagarjuna is here expanding the discussion to include
one of the most persistent problems of metaphysics, namcly, "everything" (sarVa'f!Z). lt is the problem that led the Sarvastivadins to uphold the view that
"everything exists" (sarvam astt) in .the form of substance (svabhava). Nagarjuna is simply allowing that metaphysica qucstion to be setded by the usc of
ihe fourfod proposicions tbat negate cach other.
Indeed, this is not presented as the "teaching" or "message'' (iisana) of the
Buddba, as Inada seems to understand. Quite on the contraty, it is an "admonition' ' or "advice" (anufiina) in regard to the manner in which spccuation
about "everything'' can be resolved, namey, by detnoristrating the inevitable
self-contcadictions. In other words, he is stadng that the question regarding
"everything" cannot be setted by aoy form of 4iscussion, a view clearly expressed by the Budcha in his famous "Discoutse on Everything;, (Sabba-sutt11, sec
commentary on IX.3), which Nagarjuna was probaby, conversant with.
0

9. Apara.pratyay4'f!Z anltJ'f!Z prapacair aprapancita'f{I,

nirvika/pam an'iinartham etat tattvasya lakfatJa'f!Z.

lndependently realized, peaceful, un'obsessed by obsessions, without

discriminations and a vaery of meanings: such is the charactedstic of

truth.
MKV(P) p.372; MKV(V( p. 158.

This is oneof the most iJllportant statcments ofNagarjuna, quoted often by his
dassical as wel as modern interpreters. The most receot translatioo and inter-

ExAMINATION OF SELF

271

pretttion comes from !nada and, as such, it could be consider~d a digcst of the
existing vicws. Hence, it is appropriate to quote Inada's rendering of this verse
in full. lt reads as follows :
Non-conditionally rclated to any cntity, q uiscent, nonconceptuafucd by conceptual play, non-discriminative and nondiffercntiated. Thcsc arc the charactcristics of reality (i .c. , dcscriptive of one who has gained the Buddhist truth) (p.115).
The term apara-pratyaya is herc taken to mean a special kind of rclation that
trarucends any form of empirical conditionality. iinta signifies "quicscence," a
quicscence resulting from the avoidance of any cooceptuality, probably the sort
of pcace and quiet experienced by one who has temporarily stoppcd the func- .
cioning of the six sense facultics by reaching the state of cessation (nirodhasamapatti, or sana-11edayita- nirodha). lt is where the dicbotomy of subject
and object is compJetcly dissolved (nir-11ikalpa) and where the variegated expedences of the world, the variety of m eanings or fruits (n'iinartha), is
climinated. Truth, in such a context, cannot be very different from that of
either the Upanifads or the Vedanta. l t is the flushing out of all conceprual
thinking (citta-vrtti-nirodha) , thereby transforming the empirical expecience
into one of absolute, utimate truth or reality.
However, a glance at the "Discourse to Katyayana," the primary source for
Nigarjuna's formulation of the "middle way," will reveal the untenability of
such an explanation.
Even though the verse seems to desccibe the characteristics of truth or reality
(tattvasya lalt1af!t1f!J) , every preceding scatement points to the means by which a
conccprioo of cruth is arrived at. Hence, it is one of the most important
scarements on epistemology. The most salient fearures of this epistemolgoy acc
already clearly embodied in the Buddha's discourse to K.atyayana, presented in
the form of an answer to t he question 'as to what "right view" (samma-dittht) is.
The concencs of the verse can be analysed in the light of this discourse in the
following manner.
The term apara-pratyaya does not refer ro a uuth chat is non-condicionally
rclated to any entity. Rather, it explaios the manner in which koowledge
(n'ii11a) is attained by one who has right view'' (samma-diftht). After rejectiog
the metaphysical views penaiolng to permanent existence (atlhifii) as well as
the belief in a permanent and substantial personal entity (alla .. . rne). a
belicf that will require the testimony of some ocher person who caims to know
the absolute bcginning of things, the empiricist Buddha daimed that
knowledge (of one who has right view) ~curs withouc having to dcpend upon
anorher person (aparf!-Pa&cay'ii flaf!am evassa ettha holi, S 2. 17). Ir is

272

THE PHtlOSOPHY OF THE MIDDLE W AY

knowledge for which one does not have to depend upon another, primarily
because it pertains to arising and ceasing of empi.deal phenomena. It involves
personal vcrification. a verlfication that can be accomplished by someooe
before one l:iegins to formulate any right view.
Man's search for a permanent entity, white he is equipped with limited
epistemological resources, leaves him with unresolved questions. He continues
co doubt. He is constantly perplexed and croubled. Looking for permanence,
he misses the empirically given. The ,search for the unseen "beauty queen"
(janapada-kaljiinT, D 1.193; M 2.40) makes him forget the immediatley relevaot questions (M 2 .40). As such, he has peace of mind. How.ever I if he were
to direct his attention to what is immediately given, .and understand the
human predicament in its context without being ioquisitive abouc
metaphysical entities (dukkham eva uppa1/amana'f!t uppa1jati dukkha??i niru/
jhamana?!J nirujjhatiti na kakhati 114 vicikicchatt; ibid.), his mind would be
peaceful (Siinta_
'f!l). The Buddha, in one of his very famous scatements, insisted
reflecting person understands the arising and ceasing of
that when
phenomena, all his doubts disappear (Ud 1). Such peace of mind is achieved,
not by ignoring what is relevaot in the human cootext, bur by igooring the irrelevant and irresolvable metaphysical issues.
Such a state of peace (flintt) cannot be achieved so long as one is bound by
one's prejudices (11piiyup'iidiinabhinivesa-vinibandho, ibid.,) This is the
"obsession'' (prapaca) chat Nagarjuna is referring to in the present cootext.
The discourse to Kat.yayana has no reference to any conceptua proliferation.
How such obsession~ have further stre.ngthened and encouraged the search
for utimate truths, contriburing to indiscriminate discrimioations,- soch as
those of existence and non-existence (bh'iiva-ab.h.ava)
or self--narure and other.
'narure (s11abhava-p11rabha11a), has akeady been explained by Nagarjuna. These
are the disctiminations that are to be avoided in the search for truth. Nirvikalpa .refers to the absence of such discrimioations.
A pluralistic view of the world is not incompatible with dependent arising
(pralityasamulpada). Pluralism in the context of dependent arising does pot
imply the existence of self-contradictory truths. It oeed not necessarily lead' to a
oorion of an Absolute that transcends such self-contradictory truths. The
critedon for deciding what is true io the cotllext of dependent arisiog is consequence or fruit (artha). When the Buddha maintained that "truth is one; chere
is no second" (eka'f!t hi sacca'f!t na dulijam aflhi, Sn 88'1), he was certaioy
referring to this pragmatic criterion of truth based upon the notion of dependent arising, not an absolute truth that uaoscends all forms of duality and
plurality. Nagarjuna's characterization of truth as "not having a v.ariety of
meanings" (aiinariha'l(Z) reflects more the Buddha's owo concepcion of truth.
To summarize, the conception of truth and the epistemoogical means on
the basis of which it is formulated all point to the face that the truth under con-

no

EXAMINATION OF SEl..F

273

sideration is the "dependend y arisen" (prafityasamutpanna). The principle of

dependent arising (pratfty4Samutptida) is merely the expression of the manner
in which "dependently arisen" phenomena are explained . They are rwo sides of
the same coin.

Yet, as reiterated befote. the greatest misunderstandings created by the

metap'hysicians were in regard to "dependently ariseo phenomena"
(prafityast1mutpanna-dharma). Therefore, after exam ining the epistemoJogical
means by which the conception of truth "as dependently arisen" is arrived at,
Nagarjuna, in the verse thac folJows immediately takes up the metaphysical
ioterpretations of the concept of dependence.

10.

Parfitya yad yad bhavati na hi t?ivat lad eva tat1

na canyad api tat tasmtin nocchinnaf!t niipi aala'f!J-.

Whatever that arises depencling upon whatever, that is not identical nor
clifferent from it. Therefore, it is neither annihilated nor eternal.
MKV(P) p. 375 ; MKV(V) p. 150.

Whatever is arisen dependent upon (pratftya) another, that is, the dependencly
arisen (prafityasamutpanna). is not appropriately expJained in terms of identity
or difference. As empahsi.zed so often by Niigarjuna, absoJute identity involves
permane.nce and absolute difference implies annihilation. Dependent arising is
the middle way adopted by the Buddha in eluc'idating change and causation.

11.

Anekdrtharr1- ananartham anucchedam aavata'f!J-,

etat tal lokanathanii'f!t buddhaniif(J fiisanamrta~.

That is without a variety of mean.ings or one single rneaning, it is not annih.ihlcion nor i~ it eternal. Such, it is reminisced, is the immottal message
of the Buddhas, the patrons of the world.

MKV(P) p .377 : MKV(V) p.160.

274

THE PHILoSOPHY OFTIIE MrooLE WAY

The metaphysics of identity or of difference, of one or of many. of permanence

or of annihilation, n1ay be provcd or not provcd. So far the cvidcnce has not
becn found thac would providc justifu::ation for any such notions. The only
known cvidcncc poi oes co che f~cc that things arc changing and arc depcodently
arisen. Such change and dependent arising do not fali wichin the duality of one
or many. of anni.hjlation or permaneoce. Dependent arising is what has bcen
koown so far. Unlcss a raclical change occurs in the conscicution of things, this
dependent arising will continue co be. Such is che immorcal ceaching of the
Buddha.
.iog.This, agUn, is Nagarjuna's rcflection on the cpiscemology of dependent aris-

12.

Sambuddhan'iimanulpiide iiivalii/niif!l punaf? kfaye,

jfiinaf!l prateyka-buddh'iin'iim asaf!Ut1rg'iil ptavartale.

Whcn the fully enlightencd oncs do not appcar; on the warung of

disciplcs; the wisdom of the sef-cnligbtcncd ones procccds without
association.
MKV(P) p.378; MKV(V) p .161

The Buddha's ceachings were perpccuated by a long line of disciples (ravaka).

lf that lineage were to be interrupted, stili ie is possible for his teachings to
reappear. Nagarjuna was probaby aware of the metaphor of che "ancient city"
(nagar11) whose discovery is compared to the discovery of "dependent arising"
by the histori~ Buddha himself (S 2.104-107; Tsa 12.5 [T11irho 2.80b]).
Therefore, concradicting many a cradition chac depended heavily on an uobroken concinuity as the primary source of the knowledge of che Buddha's
ceaching, Nagatjuna is herc maitaining thac such widsom (iflana) can occur
cvcn without any concact or association (a.saf!J.sargal) chrough the selfcnighccned ones (pratyeka-buddha).

CHAP'fER
NINETEEN
Examination of Time

(Kala-parTk!a)

1. Pratyutpanno

'niigata

ca yady afilam apek,ya hi,

prtJtyutpanno 'n1igalaf ca kale 'lite bh1J11qy1Jti.

If the present and the future exist contingent upon the past, ,then the pre~
sent and the future would be in the past time.

2.

Pratyutpanno
pratyutpanno

'niigala ca

na stas t(Jlra punar yadi,

'niigata ca sy'iitiim, kathaf!J 11pek,ya tam,.

Again, if the present and the future were not to exist therein [i.e. , in.the
past], how could the present an,d the future be contingent upon that?
MKV(P) pp. 382-383; MKV(V) p . 163 .

As mencioned at XVII.14, the conception of time was an imponant factor in

the explanation of the fruits of action (kllrma-phala). All actions do not bear
fruit immediately. Indeed , the problems were magnified by the Buddha's
recognition of .the possibility of the survival of the human pcrso.nality through
countlcss ives. The Abhidharn1a ists four categories of karma in terms of their
fruitioning, aJ of which can be trac;ed back to the discourses th~mselves. The
four categorics arc as folows: (i). krma that fruitions immediately~or in the present life (dit1ha-dhamma-11ed4nzya) ; (ii) kuma that fruitions in the next life
(upapa;ja-11ed4nlya); (iii) karma that fruitions in some after life (apartipariy11vedanlya); and (iv) karma that produces no fruitioning (ahosi-kamma)(Vism
p.601). Tbe use of the metapbor of "the imperishable promi$.$Oty norc'' by
Nagarjuna was necessitated by the fact chat some karmas ptoduce consequeoces
on a subsequent occasion.
.
Thus, the conception of time becomes invariably bound up with the notion
of karma. As such, whatever metaphysicaJ speculations that emerged with the
27.~

276

TH E PHILOSOPHY OF THE MIDDLE W AY

doctrine of karma, such as those mentioned at XVII. 7 10, also carne to be

assodated with the concepcion of time. Analysing time as a separate enticy, the
metaphysicians assumed that if there were to be any mutuaJ relaciooship between the present and the future on the one band and the past on the other,
then, since they are distinct entities, the present and the future wiU have to be
inherent in the past. In ocher words, the past produces the present and future
from within itself. This is the identity version of causation (s11atotpattz). A fur.
cher implication of chis is th~t if one knows the past, one also knows with ab
soluce cenaincy what the present and the future would be. These, indeed,
reprcsent some of the basie speculations of the Sarvastivadins.

3. Anapek,ya punal/ siddhir 11afi1a'!l vidyale tayol},

pr11tyutpa11110 '11aga1af ca lasmiit kalo na vidyate.

Moreover, non-contingent upon the past, their [i.e. of the present and
future] establishment is not evident. Therefore, neither a present nor a
future time is evident.
MKV(P) p. 383; M.KV(V) p.163.

The present verse embodies Nagarjuna's criticism of the ootion of cime referred
to in XIX.12. Taken away from tharcootext, this will appear co be a compJete
rejection of the very notion of rimc. However, Nagarjuna's cdticism pertains
only to the cootingencc (apek{a) uoderstood in the light of a theoty of inherence. On a previous occasion (X .-8-12), Nagarjuna has convincingly
demonstratcd the difficulties involved in explainiog conti~gence or reJativity in
the context of theories of ideotity (which is also implied in inherence) and dif.
ference.
Nagarjuna's argument seems to .rcad as follows :
1.

Major premiss:
The present and the future arc not scen to be cstablished non
contingent upon the past.

2.

Middle term:
Contingeoce of the present and the future on the past impLies
the substan tial existeoce of the present and the future in the
past, which is not evident.

EXAMINATION OP TTME

3.

277

Conclu1ion:
Therefore, the present and the future, as substantiaJ entities,
do not exist.

Ignoring the middle term, so cleary _defined at XIX.1-2, Nigarjuna's condusion will appear to be an absolute rejection of time. lt wouJd then read as
follows:
I.

Pint premisJ:
The present and the future are not contingeot upon the past.

2.

Second premiu:
The present and the future arc not non-contingent upon the
past.

3. Minor premi.u:
The present and the future cannot be both contingent and
non-cootingent upon the past.

4.

Conclusion:
The present and the future do not exist.

4. Etenaiv'iiva.ri/(tlu dvau krame'!a panvartakau,

uttamadhama-madhjiitlin ekatv'ii'rlif!t1 ca lak/ayet.
Following the same method, che remaining two pcriods of [time] as well
as related concepts such as the highest, the lowest and the middJe, and
also identity, etc. should be charactezed.

MKV(P) p.384; MKV(V) p . 164-.

Nigujuna is insistiog that the same argument be applied to the concept of the
present in relation to the past and future, and to the future in relatioo to the
past and present. In addition, he maintains that this analysis can be extended
to similar concepts like the highest, the owest, the middle, etc. In all chese
cases, the metaphysicaJ issues emerge as a result of the abso/ute distinctions that
are being made. Such absolutc distinccions arc being ofren made in logical
analyses, and are oot supporced by empirical evidence . Time, aJ experienced ,
cannot be analysed into three water-tight 'Compartmeots as past, present, and

THE,PHILOSOPHY OP TiE MIDDLE WAY

278

'

future . (See Introduction, for an explanation of the experienced time by a

modern psychologist.)

5.

Niistito grhyate kii/al? sthitaf? kilo na 11idyale,

yo grhyetagrl:iitaf ca kii/4'7 prajnapyate kathaf!J.

A non-static time is ,iot observed. A static time is not evident. Even ifthe
unobserved time were to be obse.rved, how can it be made knwn?
MKV(Pj p.385; MKV(V) p. 165 .

A non-static time is a tempoial flux. It is what the ioterpreters of the

Abhidharma referred to as the "flowing present" (santati paccuppana, DhsA
421), where the future cootinues to flow into the past througb the present. Any
attempt to grasp it would be futile, for by the time the attemp~ is.made the
present has disappeared into the past. In order to grast> it one has to stop the
flow. Hence the metaphysicians recgriized a static moment (sthitt~k/a'!a) .
Nagarjuna has already analysed the .implications of such t.heories in Chapter
VII. Thus, time understood in terms of distinct momentary entities could not
account for experience. This metaphysical theory viewed the present as the
momentary present (k.haf!a-paccuppanna, loc,cit.). Even if the non-graspable
time were to be grasped, Nag~rjuna's question is: "How can it be made
known?'' The; empiricist is. therefore, lcft with a specious time (prob;ibly an
addha-paccuppanna). lt is specious because, when aoy attemptis made to
gra.sp it independent of temporal events, it vanishes like a mirage. ft cannot be
made knowa. by any means. Absolute time mak:es oo sense for Nagarjuna.

6. Bhavaf!' prafity{I k.alai cet kalo bha.11ad rte kuta//, .

na ca k.aicana bhava 'sti kuta/? kalo bhavt!yati.

lf it is assumed that time exists depencling upon an existent, how can
therc be time without an existcnt? No existent whatsoev.er is found to exist. Whcf<' can tim.e be?

MKV(P) p.387 ; MKV(V)

pp.165-16~ ,I

ExAMINATION OP 'nM1!

279

An cxistent (bhi11a), since it represcnts a metaphysical encity baving its own

nature (s11abhiva) , has already becn rejected by Nigarjuna. Time (u/a) denied
by bim is an absolute time. lt Is time that is real as momenta.cy entities (k/a'!a)
or something that is flccting (ianta11) on its own . In the present verse, Nigiirjuna is maintainiog that two independent encicies- an cxistcnt ( bhii11a) and
time (u/a)-cannot be dependent upon one another. If they are dependent
upon one another (bhi11tlf'!l pralitya ula &et), then there cannot be time independent -of an existent. An existent as such is non-exisrcnt. Whence can
there be time? This is a rejection not of tempora! phenomena, but onJy of time
and phenomena as well as their mutua dependence so long as they arc perceived as independent entities.

CHAPTER

TWEN1Y
Examination of Harmony
(Samagn-parz k! a)

1. Hetof ca pratyay'finiir.n ca ilimag1yii ;ayate yadi,

phalam asti ca samagryaf!l iiimagrjii 1ayate kathaf!'.

lf the cffect were to arise Crom a harmony of cause and conditions, and if
it wcre to exist in the ruumony, how can it atisc from the hacmony?

2.

Hetof ca pratyayanaf!J ca samagrfii jiiyate yadi,

phalaf!' nasti ca iiimgryaf!' J1imagryii jiiyate kathaf!l.

If the effect were to arise fcom a harmony of cause and conditions and if ir
wcre not to exist in the harmony, how can it arisc from the harmony?
MKV(P) p .391 ; MKV(v,J p. 168.

The term uscd in the carly discourscs for "assemblage'' is sangali (M

1.111-112). For example, the eye, the visible form, and visual consciousncss arc
said to co~e together (sangatt) in contact (ph(Msa). Herc, the eye, visible form,
and visual coosciousness arc compatiblc factors, and not incompatiblc.
However, as evcnts carne to be distioguishcd in the Abhidharma, the intccprctcrs of the Abhidharma had to be concerned moce with "harmony''
(iiimagii) chan with simple "assemblage" (1agatt) . The concep tion of "harmony" thus carne to atuact Nigacjuna's attention, even though he will return
to the notioo of "assemblage;. later on in tbis chapter.
Nigarjuna's anerhpt at the beginoing of this chapter is to examinc the con
ccption of harmony and to pccvent any metaphysical interpretation of it.
Metaphysical speculacions wece further advanced by the S:uvastivada distinction berween cause (helu) and coodicioo (pralyaya). This lattec discinction,
supponed by the Sarvastivada conception of self-nacure (s11abha11a),
culminated in the idea of self-causacioo (111ata u1pattt).
280

EXAMINATION OF H A.RMONY

281

For this reason, in the very first verse, Nagarjuna takes up three ideas: (i) harmony, (ii) distinction betweeo cause and condicions, and (iii) the arising of an
eff.ectfrom a harmony of cause and cooditions. (Note Nagarjuna's use of helu
in the singu lar and pralyaya in the pl ural.)
Nagarjuna's critidsm is mainly directed at the idea of self-causation. If the
fruit arises from the harmony (siimagrya, abJacive case) of a cause and a set of
cooditions (helo!/ pratyayanaf!Z ca), then it is already existent in che ham1ony
(samagry'iif!Z, locacive case). H ow then can it arise from the harmony
(1amagrya)? The causal process presenced in this manner implies che idencicy
between "harmooy of cause and conditions" and the fruit or eff.ect that arises
from it. The ideotity theory of causatioo was already criticized in Chapter I.
Similarly, as stated atXX.2, if the fruit or effect atising ftom such a harmony
were not to be in the harmony, chat is, if the cffect is different from the harmony, it can never arise from that harmony. This is a criticism of the nonidentity theory of causacion discussed in Chapter I. These two vecses, therefore,
state the difficulties involved in accepting theories of identicy and difference.

3.

Heto ca pratyayanaf!Z

ca Iamagryam asati cel pha/af!Z,

grhyeta nanu iiimagryaf!Z iiimagryiif'(Z ca na grhyate.

If it is assumed chat the effect exists in the harmony of cause and conditions, should it not be observed i.n the hatmony? Howcver, it is not
observcd in the haunony.

4.

Helol ca pralyay'iin'iif!Z ca Iamagry'iif'(Z n'ii1ti cet phala1!1hetavaf? pratyaya ca Jyur ahetu-pratyayaif? ItJ'fn'iil/.

If the cffect were not to exist in the harmony of cause and conditions,
thcn the cause and conditions would be comparable co non-cause and
non-conditions. '
MKV(P) pp.392393; MKV(V) p.169.

lf the identity theory is valid, rhen the fruit could be observed (grhyela) in the
harmony itself, eveo before it is produced through such harmony. However,
Nagarjuna assumes chat ie is not observed or grasped .in this manner. Once

284

THE PHILOSQPHY OF TI-lB MJDOLE W AY

become the transformation of the cause, then it follows that tbere is a rebirth of a cuase that was already bom.
MKV(P) p.396; MKV(V) p .17L

Among the variety of causes that were formulated to account for a continuous effect (nif:;anda-pha/a) are complementary (sabhaga) and universal
(sarvatraga) causes (Akb p .94). Explaincd in the light of the theory of
moments, a universal cause would mean the contiouous acising of the same
cause in order to account for the cootinuous effect. The continuous effect
(ni.fyan4a-phala) thus turns out to be a cransformacioo (sarp,krama11a) of che
cause (helu). Nagarjuna refuse:i to recognize the rebirch of the same cause that
has ceased.

10. ]anqyet phalam utpanna?(l. niruddho 'sthaf!'gatal/ katharp,,

tz!thann api katha~ hetuf? phalena janayed vrtal/.

How can a cause that has ceased, has reached its end, give rise to an effect
that is already arisen? How can a cause, even though endung, produce
an effect, when it is sepatated from the latter?

M,KV(P) p.397; MKV(ll) p .171.

Throughout the present analysis we have indicaced that the philosophical

rnethod adopted in the Abhidharma did not create any metaphysical problems
until the interpreters adopted a theory of moments that were foilowed by
theories of identity and difference. The vadety of causes (hetu) and conditions
(pratyaya) , even though not presented in such detail and in identicaJ terminology in the discourses, is not incompatible with the teachings embodied
therein. However, even a cwsory glance at the manner in which the interpreters of the Abhidharma struggled with these different causes and conditions
. {see AK cbapter ii; Akb pp.38-tlO) is sufficient to indicate the magnitude of
tne probtems they were faced with.
Nagatjuna concinues his analysis showing how a cause that has ceased (nirurldha) or one that remains (tt!fhan), yet is distinct from the effect (phalena
vrtaf?), could never give rise to an effect.

EXAMlNTION OF HARMONY

11 .

285

Atliii11rta'1 phalenasau katarm1j janayed phala'f!J,

na hy adurva 11a dr!f11a 11a hetur janayaJe phalart1.

What causc, even if it wcre not separated from the effect, will give rise to
the effect? A cause does not produce an effect eitber impercepcibly or
pccccpribly.
M.KV(P) pp.398-399; MKV(V) p.1 72.

When the analytical process was carried to its extreme, many evenrs, whicb
under norma! cootexts would not have been quescioned , c:une to be doubced .
For example, inscead of a reJaced event, analysis produced two: a relarion and
an event. Wheo the norma! empiricist criceria wece adopted in the Jatter case,
one was compelled to assume chat the relation is not perccived in che same way
as the evenr is perceived . The metaphysician was thus compeUed to insist upon
the subscantial existence of the relatioo. "Birth is the arising of whac is to ~e
born and this does not take place wichout causes and conditioos," so says the
Abhidharmakofa (janyaJya Janik.ii 1atir na hetu-pratyayair vina, ii.49). The
bhii1ya on this passage concinues to argue about che narure of this "birth"
(jatt), insisring that the genetive case (!a/{ht) (as io che sracement, "arising of
what is to be boro") makes no sense if birtb (jiitz) is not perceived in the same
way as "that which is to be boro'' (Janiki.i) is perceptible. The Sautrantika
Vasubandhu argues against this position saying chat "umbers, Hmits, distinction, union, analysis, ochcrness, sameoess" are recognized as reals (salwa) in
the speculatioos of the herctics (llrthakara), and chat these arc oecded only to
establish the knowedge (but/dht) of the reality of "the one , the dual, the great.
the individuated , the uniced, the separated, the other, the same, etc. " To illustrate his point of view. he refers to the example of "the union of f o1m "
(iiipaJya sa<f!Jyoga) and maincains that the generive ca~e indicates the ownnature (svabhava) of "form ." However, in his owo Sautrantika view ie is a merc
designacion (pr(ljflaph~matram, Akb p . 79).
Ie is this conttoversy 'rega.rding the reality or urueality of numbers, conjunctions, disjuoctioos, etc. (a concroversy thac has contioucd to plague
philosophers in the modern world) that Nigarjuna is referring to in the present
verse when he uses the terms rlr!fva (seen) and adr11va (unseeo).

12.

Niifftmya hy affJena phalasya saha he111na,

naj'iitena na jiitena sa1!Jgatir 1atu vidyate.,

286

THE

PHILOSOPl-tY OF 11l1l MIDDLE W AY

Inceed, the assemblage of a past effect with a past or a future or a present

cause is not evident.

13.

Na jatasya hy a1atena phalasya saha hetun1i,

1iiifftena na 1atena saf?Zgatir jiitu 11idyate.

Indeed, an assemblage of the present effect with a furure or a past or a

present cause is not evident.

14. Niijiitasya hij'tena phalasya J(lha hetuna,

niijiitena na n(#fena Sll'f'!Jgatir fiitu 11idyate.
Indeed, as assemblage of the future effect :with a present or a future ot a
past cause is not evident.
MKV(P) pp.400-401 ; MKV(V) pp.172-1 73 .

After analysing the conceptioo of "harmony" (i1imagfi), Nagatjuna now retums

to "the eadier notion of "assemblage" (saf?Zgatt)(see commcnts n X:X.1). The
notion of assemblage may not have caused philosophicaJ problems for the early
Buddhists who adpted an empiricist theory of change and causation . But, for
Nagarjuna, the coaceptioa of assemblage, like the aotion of harmony , does
not work, so long as it is associated wich the metaphysical conception of time.
Vasubandhu , as a Sautrantika, himself raises the more radicaJ questlon as to
how the mind (1!l4nas) chat has alredy ceased cn assemble with future and
present concepts (dhaffl!a) and mental consciousness (mano-11ijf1ana) in order
to produce contact (Akb p. 143, manindriyasya punar niruddhasy'iinagalavarta1nanabhjii??Z dharma-rnanovij1in1ibhy1i'f'!J katha1'?1- sa'f'!Jnipataf?). He then
refe,rs to a variety of insights (bhedar!f. gata buddhayaf?) of different teachers,
and the controversy appears to be too complicated. He dismisses them saying:
"Enough of this argument" (ala1'?1- prasangena). Nagarjuna's reluctance to accept any one of these views is, therefore, the result of his realizatioo that they
are all metaphysical views not allowing for any definite answers or solutions.

15.

Asatjii'f?Z sa'f'!Jgatau hetuf? kathaf!J janayate phala?'(l,

saly41'?1- saf?Zgatau hetuf? katha'f'!J janayate p hala'f'!J.

EXAMINA TION OF H ARMONY

287

When an assemblage does not exist, how can a cause produce an effect?
When an assemblage exists, how can a cause produce an effect?
MKV(P) pp.401-402 ; MKV(V)p . 173

Here the causc (hel'll) is defined in terms of assemblage (Jaf?/,gatt) which is

either existent (Jat) or non-existcnt (a.rat). A cause defined in such a way turns
out to be substantialist or non-subsranrialisc. The former implies permanence
and the lattcr , annihilacion. Nagarjuna rejects boch.

16.

Hel'llf? phalena fUnyai cet kathaf?Z janayate phaiaf?Z,

hetul? phalenafunyai cet kathar?t janayate phalar?t.

lf it is assumed that the cause is empty of an effect, how ran ie produce an

effect? If it is assumed thac the cause is not empty of an effect, how can ie
produce an effect?
MKV(P) p.402; MKV(V) p.173 .

Turning atound, Nagarjuna now takes up che concepcion of the effecc or fruit
(phala). If che cause is empty (Junya) of che effecc, it can never produce an effect. Neither is it appropriate co assume that the effect is produced by the cause
if it is already in the caus, hence nor empty (afunyar?t) of the effecr.

17.

Phalar?t notpatsyate 'Junya1n aUnyar?t na nirotJyate,

aniruddha,n anutpannam aUnyarrt tad bhavzjyati.

A non-empry effen will not arise; a non-empcy effect will not ce-.tSe. For,
the non-ceased and non-ariseh will also be the non-en1pty.
MKV(P) p.402: 1HKV(V) p. 17-i.

This verse should clarify the meaning of the famous terms atnruddharri ("nonceased") and anutpanna1?1 ("non-ariscn") morc rhan any ocher scatement of

288

THE Pl-OLOSOPHY OF TI-IE MIDDLE

wAy

Nagarjuna. Whatever phenomenon (dharma) is characterized by Nagarjuoa as

"empcy" (unya) is without self-nature . 1f something possesses a self-nature,
chen it is not empty (aunya). As such, it can neither cease nor arise (anirodham
anutpadaf!t), Thus, an effect or fruir that is not empty and therefore is possessed of self-nature will nor arise (notpatsyate) nor cease (na flirotsyate) ,

18.

Katham utptTtsyate unyarrz katharri UnJll'f!Z nirotJyate,

funyam apy aniruddharp. tad anutpannarp prt1Sajyate.

How will the empty atise and bow will the empty cease? lf something is
empty, it follows that it is non-ceased and non-atisen.
MKV(P) p.403; MKV(V) p .174.

Taken by itself, this verse can be used to justify the view that accotding to
Nagarjuna "emptiness" (sunyaf(i) is the utimate truth beyond all forms of
description. Hence the negative d~scription: "non-ceased'' (ant'rttddha'f!Z) and
"non-arisen" (anutpanna'f!Z).
However , considered along with XX.17, whkh rcjects the notioo of identity
presented by the Sarvastivadins as self-nature (.rvabhava). which according to
Nagarjuna is "oon-empty'' (afunya), what is referred co as "empry" (Unya) in
the pi:eseot verse is mo,re apptopriately understood as a reference to the nonidentity theory of the Saucrantikas. It may be remembered that the first Buddbist school to deny the Sarvastivada theo.ty of self-nature was the Sautrantika
school. How che Sautrantika theory of "emptiness'' or ''absence of substaoce"
(nil/-svabhava) and their theoty of "momeotary destruction" (kfaf!a-bhanga)
Jcd to a denial of both arising and ceasing has already been pointed out (VIl.17
ff.).
As such, the present stacement of Nagarjuna, following upon his refutation
of ideotity, must involve a rejection of difference. the rwo extremes that he has
pe(Sistencly criticized. In other words, che Sarvastivada and Sautrantika
theories both rendcc acising and ceasing mcaningless.

19.

Hetol? phalasya caikatva'f!Z na hij111ii,papadyate1

hetol/ phalasya canyatvaflZ na hijat'Upapadyate,

EXAMINATION OP HARMONY

289

The identity of causc and effcct is indecd not appropriatc. The difference
becwcen causc and cffect is indeed not appropriatc.

20.

Ekat11e phala-hetvol/ sjiid aikyaf'!Z janaka-janyayof?,

Prthaktve phala-hel11ol; sjiit lu/yo hetur ahetuna.

ff thcre were to be identity of cause and cffcct, thcn thcre would be

onencss of produccr and the producc. lf there werc to be differcncc betwccn cause and effect, thcn the cause would be cqual to a non-cause.

21.

Phalaf'!Z svabhiiva-sadbhutaf'!Z kif!l hetur janayi/yati,

pha/af!l s11abhiivasadbhulaf'!Z kt?p hetur janayi,jali.

How is it chat a causc will producc an cffect which comes to be on its own
naturc? H ow is it that a cause will produce an effect which docs not come
to be on its own nature?
MKV(P) pp.403-404; MKV(J1 p.174.

The impicacions of the theories of idcntiry and difference discussed at

XX. 17 -18 are further elaboraced in these t:htee verse.

22.

Na cajanayamiinasya hetutvam upapadyate,

hetulvanupapattau ca phaktf'!Z kasya bhavi-ryatz:

Moreover, the causa! efficacy of something that is not producing is not

appropriate. In the abscnce of causa! cfficacy, co what will the cffcct
belong?
(MKV(P) p.405 ; MKV(V) p. 17).

Identiry and difference are thus shown co mii tate against noc merely arising,

290

THC: PHILOSOPHY OF THE MIDDLE WAY

bu t also causal or dependent arising. Causa! efficacy (hetut114f!J) is rendered

mcaningless if there were to be no production, chat is, if the arising of an effec t
cannot be accounted for by a cause, which is the result of the Sarvastivada and
Sautrantika theories. If causaJ efficacy cannot be accounted for . how can one
speak of an effect or fruit?

23 .

Na ca pratyaya-hetunam iyavz atmanam anatmanii,

ya samagfi janayate sa kathaf!J janayet phala1!t.

Whatever harmony of causes and conditions there is, ie is noc produced

by itself or by another. ff so, how can it produce an effect ?
MKV(P) p.406: MKV(V) p. 175.

After producing an exhauscive analysis of the relationship between a cause

(helu) and fruit or effect (phala) , Nagarjuna returns to the quest ion with
which he began che chapter, the harmony of causes and condfrions. Once
again, rcfuting the identity and non-idencity theories, he insists that whatcvcr
harmony there is, ft is not produced by the causes and conditions cither from
within rhemselves (atmana = svabhavala) or from outside (anatmana =
para1af?). Wben harmony cannot be explained in any of these two ways, chen it
is not possible ro assume chat an effect can arise from a harmooy so expJ.ained.

24.

Nasamagfi-krta1!} phala'f!l nasamagii-krtaf!J phala'f!l,

a.rti pratyaya-samagfi kuta eva phalaf{Z vina.

The effect is not made by the harmony, nor is it not made by a harmony.
Where iodeed can there be a harmony of condirions wichout an effect.
MKV(P) pp.406-407; MKV(V) p. 176.

The effect is not produced by a harmony of causes and condicions explained in

the above mann er. Nor is ie produced by a non-harmony. Nagarjuna's finaJ

EXAMINATION OF HARM ONY

291

qucstion is: Where can chere be a harmony without a fruic or effect? This fioal
conclusioo need not be understood as a deniaJ of fruit or effect. On the contrary, ie can be the assertion of a pragmatist who insists that chete canoot be a
harmony of causes and conditions without a fruir or effecr.

CHAPTER

TWENTYONE
Examination of Occurrence and Dissolution

(Sa'f(Jb hava-vibhava-part kffi)

1.

Vina 1111 Jaha va 11asti vibhaval; J111?Zbhave11a ""'~

vina va Jaha va nasti saf'!Zbhava/J vibh11ve1111 vai.

Dis.1olution docs not exi.st either without or with occuuence. Occurrencc

does not exist either without or with dissolution.

2.

Bh1111t!yati kathaf'!Z 11iima 11ibha11a'1 Jatr1bha11a1(l flina,

vinaiva janma maraf!a'fl 11ibha110 nodbha11af!Z 11in"ii.

How can there be dissolution without occunence, death without birth,

dissolution without uprising?
MKV(P) pp.4 10.411 ; M.KV(V) p.178.

The terms Sllf!lbhava and vibhava need to be cransJated keeping in mind the
purpose of this secrion. lt is an attempt on the part of Nagarjuna co cxplain the
life-proccss (saf!Mara) or the human pcrsonality without resorting to a theory of
self or soul (iitman1 pudgala) cooside1ed to be eternal. As mentioncd earlier,
this whole section is devoted to the cstablishing of the idea of nonsubsrantiality of the human person (pudgala-nairatmya) ..This has to be achiev,
cd not onJy by showing the untenabilicy of the theory of permancnce or eternalism , but also of a conccption of annibilationism.
Thus, in this particular chapcer satr1bhava, in the sense of ocCUirence, pcrtains to the "birth" (janma) of a human being conditioned by various factors,
without any underlying permanent entity passing from one life to another.
Vibhava, in the sense of dissoution . mcaos "death'' (mara11a), and herc the re'is
no implication of competc annihilation. In the life-proccss, binb is thus not
the absolute beginning , nor is death the absoute end.
292

EXAMINATION OF 0 CCURR!lNCE AND D ISSOLl!TION'

293

The first two verscs deny the occurrence of death wichout (t1inii) binh or at
the same time (saha).Just as much as death docs not occur without birth, there
is no up-rising (udbhava) witbout dissolution (vibhtJva). This is an empirical,
rather than a logical, analysis of birth and death. A rationalist can argue that
"all human beings arc not monal,'' for everyone who has becn born has not
died. Such a rational argument did not prcvent the Buddha from acccpting
mortality as a fact of life. For him, the evidencc lies in the fact that so fu all
human being who have died had been born. However, this does not lead the
Buddha to assert the metaphysical view that death is inherent ~ bitth.

3. Sam.bha11enai11a 11ibha11a'1 katha?(' saha bha11i1yati,

na janma-maraf!a?!J caiva?(' tulya-kiila??Z hi 11idyate.

How can there be dWolution aong with occurrence? Indeed,

simultaneous binh and death arc similarly not evi.dent.
Jl;U(V{P) p.411 ; JUJ(V(V) p. 178.

This is an explicit rejection of che metaphysical view chat death is inherent in

binh. lf the life-proccss (sa?('..riira) werc to be understood as a seties of momentary existences (sa11Jtana, sa'f!'talt), as the Saucrantikas believed, theo the sceds
of death should occur at the very moment of birth. This logi cal explanation was
not accep~able to che empiricist Nligarjuna.

4.

Bhavqyali katha'f!' nama It1'f!'bha110 vibht1114f'!t vina,

anityalii hi bhave!u na kadiitin na vidyate.

How can there be occurrence without dissolution, for the impermanenre

in existences is never not cvident.
'MKV(P) p.412;_MKV(V) p.178

Occurrencc (sa'!Jbha1111), as an absolutely new beginning. is rejccted herc, when

Nagarjuna affirms that without dissolution occurrence does not cake place .

294

THE PHrr.OSOPH\' OF THE MIDDLE

WAY

lndeed, the discourse to Katyayana utilized the cmpirical argument that the
belicf in a permaoent entity is abandoned wben one perceives the cessation of
the world (loka-nirodhaf!J ... yathabhufaf?1- sammappanfl aya passato yii loke
atthita sii na hotz). Nigarjuna's argument herc is not that "there is another
realm or aspect of beingwhich peope havc always overloo~ed. Thisis the realm
or aspect of bhiiva, [which] refers to the truy dynamie worldly existcnce" (as !nada seems to assumc, sec p .12 5). Rather, it is a rejection of bhava, piimarily
because impermanence (anityatfi) is incompatible with bhava, which impies
permanence. In fact, Nigarjuna was probably aware that the Buddha bad
always employed the term bhava to expain the process of "bccoming," instead
of the absuact term bhiva. Indeed bhiva, or its more resuicted form svabhava,
is equivaent to astitva (attht'tii) and more often Nagarjuna understood the
term in that sense.

5, Sa?'(Jbhavo vibhavenaiva kathaf!l saba bhavi/yflti,

na janma- maraflZf!' ctiiva tulya-kalaf!l hi vidyate.

How can occurrence be evident aong with dissolution? Indeed,
simutaneoos birth and death are similarly not evident.

At XXi.3, Nigarjuna uestioned the feasibility of asserting that death occurs

'
together with (saha) birth. In the present verse, he is questioning
the validity of
asserting the occurrence of birth together with (s11ha) death. In other wofds, he
is questioniog the feasibility of asserting an invariable connection between
death and rebinh, an iovariable connectioo that is never asserted by the Budcha. Io fact, what the Buddha asserted was that a dying person, depending
upon conditions, can be reborn. An invaribfo relacionship betweeo death and
rcbirth was admitted only by the substantialists.

6. Sahinyonyena 1ii siddhir vininyonyena va yayoi;,

na vidyate tayol, siddhil( kath11n nu khalu vidyate.

EXAMJNAIION OP 0 CCURRENCE AND D JSSOLUTJON

29)

'fhe occurrence of things, either together or separatey , is not evjdent: lf

so, how can their establishment be evident?
MKV(P) p.4l5 ; MKV(V) p. 180.

This cooclusion of Nagarjuna is related to his criricisms in the previous verses.

He is simply questioning the met aphysical assertioos relating to occurrence and
dissoution , birth and death.

7.

lqayasya St11{1bha110 niisti nfikfayasyasti Stl'f?'Jbhaval/,

k!t1Jfl5ya vibhavo niisti vibhavo nak1ayasya ca.

Occurrence of that which is waning does not exist, nor is there occurrence
of that whicb is not waning. Dissolution of that which is waning does not
exist, nor is there dissoution of the not waning.
MKV(P) p.415 ; MKV(V) p.180.

The term k!aya was used in the Buddhjsr texrs in the cootext where warung or
compJete extinction is implied. Whereas the term nirodha could mean ce.asing
that coud be followed by arising (utpada) and, as such, t hey could be used as
comp lementaries to explain change and impennanence as well as dependent
arising, the term k!aya had no such complementary term except irs negation ,
a-k!aya, wh ich impJies permanence. For t:his reason, Nigarjuna was able to
maintain chat there is neither absolute cessation (k!aya) nor perrnanence
(a-k!aya) of both occurrence (sa?'(lbhava) and djssolution (vibhava),

8.

Sa'f?'Jbhavo vibhavaf cai11a vina bhavaf?Z na 11idyate,

sa?'(lbhava?!Z vibhat1a7!1- oaiva 111'14 bh4110 na 11tdyate.

Without an cxistent, occurrence as weU as clissolution arc not evident,

Wicbout occurrence as well as clissolution, an existent is not evident.

296

9.

"
TuE PHllOSOPHY OF THE MIDDLE W Ay

St1f!Jbht111011ibhava c11i1111 n11 unyasyopapadyate,

It1f!Jbht1flo tJtbha11a caiva nafUnyasopapadyate,

Either occunence or d.issolution of the empty is not appropate. Either

occurrence or d.issolution of the non-empty is also not appropriate.

l O.

Sa'f!Zbhavo 11ibht111a caiva naika ity upapadyate,

Itlf?Jbhavo 11ibha11a ctlitJa na nanety upapadyate.

lt is not appropriate to assume tha.t occurrencc and d.issolution are identical. lt is not appropriate to assume that occucrence and d.issolution arc
diffcrcnt.
MKV(P) pp.416-4 18; MKV(l1 pp. 180-181.

lt is probab1y the use of the term bhava at XXl.8 that Jed !nada to assume chat
it represents a unique realm of existence recognized by Nigarjuna. However, if
we arc to keep in mind the two metaphysical schoos- Sarvastivada and
Sautrlintika-and their metaphysical doctrines of identity and difference, it is
possible to interpret the statement in verse 8 as wcll as the two verses chat follow
as straightforward criticisms of these two schools.
For example, a Sautriincika who denies a bhiva or .wabhava can maintain
that occurrence and dissolution can take plac:e without a permanent entity
(bhava, s11abha11a). Occurrence and dissolution woud then mean the absence
of continuity. Nagarjuna denjes this.
On the conuary, a 52.rvascivadin can maintain thar a bha11a oc wabhava can
exist without occurrcnce and dissolution, and this would account for continuity
but negate difference. Tbis too is rejected by Nagarjuna.
Similarly, occucrence and dissolution arc incompatible with the "empty''
(fUnya), as understood by a Sauuancika, or the "non-empty" (afUnya) as ex.plained by the Sarvastivadin (see XX.16-18).
Furthermore, occurrence and dissolution arc neither identica (naika) nor dif
ferent {niini) in a metaphysical sense.

ExAMINATlON OF 0 CCURR.ENCE AND D ISSOLUTlON

11.

297

Dryate

saf!JbhavaJ cai11a vibhavaJ cai11a te bhavett

dryate saf!lbhavaJ caiva mohad vibhava e11a ca.

lt may occur to yon chat both occurrcnce and dissolution are seen. How
cvcr, both occurrence and dissolution arc scen only through confusion.
MKV(P) p.419;,MKV(v,) p .181.

It would be a folly to assume that occurrcnce and dissolution arc perceived

(driyate). Neither the Buddha nor Nagarjuna would be willing to assert this.
Occurrence and dissoution in themselves are not perceivabe . They are seeo onJy in reation to phenomcna that occur and dissolve. The metaphysicaJ issues .
that arose during Niigarjuna's time as a result of the cxtremist aoaJysis of
phenomena 1nto cvents and relations have aready been referred tp (see
XX.11).

12.

Na bhaiiajjayate bhavo bhiivo 'bhavan na /iiyate,

nabhavaj 1ayate 'bhavo 'bhavo bhavan na ;ayate.

An existe.n t does not arise from an existent; neither does an existent arisc
from a non-existent. A non-cxistent does not acise from a non-existcnt;
neither does a non-existent arise from an existen't.

13.

Na svato j'iiyate bhaval; parato naiva jayate,

na wataf? parataJ caiva ;ayate jiiyate kutaf?.

An existent does not arise from iuelf, or from another or from both itself
and anothcr. Whcnce can it then arise?

M.KV(P) pp.419-421; M.KV(v,) p .182.

298

THE PHILOSOPHYOFTHE MIDDLE WAY

Bhiiva and abhava referred to herc may be compared with the sat and 11.fat in
the pre-Buddhist Indian philosophy. The unresolvable metaphysical questioos
that plagued Indian philosophy for centuries, uestions such as "Did existence
(sat) arise from non-existence (t11at)?" 01 vice versa, have once again been in~
ttoduced into Buddhist thought by the Sarvastivadins and the Sautrantikas.
The idencity and oon-idencity th.eories of causation, theories of self-causacion
and exteroaJ causation, and many other metaphysical views thatemerged in theBuddhist uadicion are repeatedly mentioned and rejected by Nagacjuoa.

14.

Bhavam .abhyupapannasya avatoccheda-darfana??Z,

prasajyate sa bhavo hi nitya 'nityo 'tha va bhavet.

For him who is engrossed in existence, eternatism or annihilarionism will

necessarily follow, for he would assume tbat it is either permanent or impermanent.

MKV(P) p.421 ; MKV{V) pp .182- 183.

Nigarjuna is here presentiog the inevitable conclusions chat a person involved

inspeculacions relacing to bhiiva will reach. If the bhava is assumed to be perrnanent , he will end up with a notion of eternalism . lf. on the contrary, the
bhava is looked upo as beiog impecmanent, then he will assert annihilation .
lt may be noted that the Sarvastivadins who were involved in the nocion of
bhfiva, carne up with the bdief in an eternal self-nature (1vabhava) or
substaoce (drr;vya). The same involvement led them to assume the momentary
destruction (kfa11a-bhaga) of impermanent qualities or characteristics of
bhi11fJ.

15. Bhava1n abhyupapannasya naivocchedo na afvattt1rt,

udaya-vyaya-sa'l'(tt?ina// phala-hetvor bhavaf? sa hi.

[On the contrary,] for him who is engrossed in existence. there would be
neither annihilationism nor eternalism, for, indeed , becoming is the
senes of uprising and ceasing of cause and effert.

EXAMINATION OF 0 CCURRENCli AND DISSO[UUON

16.

299

Udaya-vyaya-sa'f(tliinal? phala-hetvor bhaval; sa cel,

vyayasyapunarutpatter hetucchedaf? prasajyate.

lf ie is assumed that becoming is che series of uprising and ceasing of the

cause and effecc, tben with the cepeated non-arising of thac which ceases,
it will follow that there will be annirulation of the cause.
MKV(P) pp.422-423; MKV(V) p.183.

The Saucrantika position was no moce differeot. They assumed chat bhava is
oone other than bhava, the process of becorning represented by the series
(sarpt'iina) of arising and ceasiog of effect and cause (phala-hetvol?) As such ,
they believed chat cheir coocepcion of ex.istence docs not come under either
eternalism or annihilationism.
Howeve.r, Nagarjuna has a different perception. He 3:-SSum es that if the process of 'becoming (bhava) is analysed in to a series of arisi'ng and ceasing of effect
and cause, as the Saurrantikas did, one is compelled to recognize che nonrebirch (a-puJ'lllT-utpattt) of chat whicb bas ceased, and there would be compete annibiacion of the c.ause. This is simiar to the argument used at XXI. 7.

17.

SadbhavaJya svabha11ena nasadbhava ca yujyate,

nirva11a-kale coccheda~ praa111'iid bhava-sarptatef?.

The non-existence of thac which possesses existence lit tttmS of setf-nature

is noc appropriate. [On the contrary,J at the time of fceedom , there will
be annihilation as a result of the appeasement of the stream of becom.ing.
MKV(P) pp .423-424 ; MKV(V) pp.183-184.

Nagarjuoa's attcncion is now direcced at the Sauvaotika view , for it is chat

whfrh finally contribu ted to the "personalist theory" (p11dgala-vada) of the
Yatslputriyas. The Sauuintika will dismiss the Sarvastivada view, insisting
that there is no way in which nif'll'ii11a can be cxplained in terms of their nocion
of self-narure (svabhata) , especially because whac is really cJsting (sadbhava)
oo its own (svabhavena) cannot become a non-cxiscent (asadbhava).

300

THE PH!LOSOPHY OF THE MIODLE WAY

However, the Sautrantika will say that as a result of the appeasement of the
sedes of becoming (bha11a-saf!Jfat1), there is annihilation at the tltne of nirvaa.
Having stated this position, Nigjrjuna proceeds to analyse its implications.

18.

Carame na niruddhe ca prathttmo yujyate bhavaf?,

carame nanimddhe ca prathamo yujyate bhavaf?.

It is not propec to assume that thcre is first becoming whcn the last has
ceased . Nor is it proper to assume that thete is first becoming when the
last has not ceased.
MKV(P) p.425; MKV(V) p.184.
The Sautrantika tbeory of a series of momentary existences is under investigation here. As meotioned before, the Sauuancikas were often faced with the
problem of explaining arising (utpida) . Thus, Nagarjuna argues that the first
(moment of) becoming (prathamo bhava4) cannot occur when the last (carama)
has ceased, for there will be nothing to give rise to the former . Th is is what was
referred to as the cessation of the cause (hetuccheda) at XXI .16. The other
alternative is to assume tbat the enticy of the last moment has not ceased
(aniruddha) , and this, of course, makes it difficult for the first becoming to occur at all .

19.

Nirudhyaniiine carame pr11tha1no yadijiiyate,

nirudhyamiina ekaf? syaj1aya1nano 'jJaro bhavet.

If the fust were to be born when the kst is ceasing, then that which is
ceasing would be one and that which is being boin would be another.

MKV(P) p.426; MKV(V} 184.

Assum'.e that the first becoming occurs at the time wheo the last is ceasing
(nirudhyamana). NQ:garjuna insists that, in that case, whac is ceasing is one

EXAMINATION OP 0CCURRENCE AND DISSOLUTION

301

thing and what is arising is somcthlng com plctcJy diffcrent (aparo). Nagarjuna
is herc referring to the implications of the theory of mo.menrs, namely, the
recognition of absolute d iscinccions.

20.

Na cen nt'1Jdhyamana ca j'iiyamiinai ca yujyate,

iiirdha1p, ca mnyate ye1u leju skandhe!U1ayate.

Ii it is assetted th.at the ceasing is also the being boro, this would not be
proper. For, in th.at.case 1 whatever that is born in reation to the aggrcgates, would also be dying at the same time.

MKV(P) p.426; M.KV(V,J p.185.

Furcher complications will arise if it is assumed chat something chat ceases is
aJso arising. Nagarjuna is not prcpared to acccpt such an occurrence because
this would m can the simutancous death and birth of the aggtegates .
Thcrefore, an occurrence is not explained by a scrictly momentary theory of cxistence advocated by the Saucra ncikas.

21.

Evaf!l fri!v api kale!u na yuklii bhava-saf!ltatil?,

tn!u ka/e!U ya na.sti iii k.athaf!I bhava-SPT(Jlatil?.

Thus, the stream of becoming is not proper in the context of the three
pcods of time. How can there be a s.tream ofbecoming th.at does not aist during the three periods of time?

MKV(P) p.427; MKV(V) p .185.

The conclusion is inevitable that the series of becoming (bhavaSllf?llah) is not

appropriately explaiocd in ter ms of the three pcriods of time, a concept which ,
as shown in the previous chaptcr, is not empirically grounded .

CHAPTER
TWENTYTWO

Examination of the Tathagata

(Tathagata-pafik.ra)
l.

Skandha na n'iinya~ s~t1ndhebhyo n'iiJ?!Jin skandh'fi na tefu sal;,

tathagataf? skandhav'iin na kata1110 'tra tathagatal;.

The tathiigata is neither che ag.gregates noc different from them. The aggcegates are not in him; nor is be in the~ggregates . H e is not possessed of
the aggregates. In such a context, who is a tathagata ?

2.

Buddhaf? .fkandhan upadiiya yadi nasti svabhii11ataf?,

rvabhavata ca yo n'iisti kuta!? sa parabhavt1taf?.

lf a Buddha were co be dependent upon the aggregates, he does not exist

in tenns of self-nature. He who does not exist in terms of self;nature, how
can he exist in terms of othec na~re ?
MKV(P) pp.432-436; MKV(V) pp.187- 189.

ln addicion co some of the terms used by the Buddha to refer to his owo
achievements, his disdpJes used a vast array of epithets in extolling his virtues.
It is significant to noce that none of these epithets caused so much
m isunderscanding as the one under discussion in the preseot chapter. narriely,
tathagata. lt seems chat the very cooceptioo of tathagata iovited
misundetstanding.
l"he term can be rendeced inro English as "thus-gone" (tathii-gata) . The
conception of o ne who has "rhu>-gone" immed iatey brings to mind the idea of
an ''agent'' (sec Chaptcr 11). lt was. thefefore, ioevitable tliat when questions
relaring to rhe "destiny'' of r.he enlightened one were raisttd. they were always
ralsed in relation to a tath1ig11ta.
30~

EXAMlNATION OF TATHAGllTA

303

Nagarjuna was aware that questions pertaining not only to the final desciny,
but also to the nature of the living sa.int were raiscd during rhe Buddha's day . A
discussion between Sariputta and a monk named Yamaka, who had enten:a.ined rhe belief rhat a lathagala is annihilated after death (param maraf!a) is
repocced in rhe Saf!'yutla-niliiiya (S 3.109-115; Tsa 5.2 (Tairho 2.30c-3lcl).
Reading the first part of this discussion, one gcts the impression that Sariputta
was 'ieluctant to identify the tatliiigala with the five aggregates or to distinguish
the tath'iigata from the aggregatcs. On rhe basis of this, it is possible to come to
the condusions, rhat rhe tathagala is linguiscically cranscendeot.
However, if the discussioo is followed to the very end, one can clearly sec rhat
such a conclusion is not warranted. For Sariputta is not willing to admit a
talhagala in an absolute sense (saccato thetato), comparable to the
metaphysical "self'' (a/man) that was considered to be pcrmaoent and eteroal.
Indeed , toward the end of rhe discussion, Sariputta movcs from the conceprion
of talh'iigata to the notion of "self' (atta) and refuscs to admit a self rhat is identical with or different from the aggregates.
Nagarjuna, as if he had read this discoursc:, begins the present chapter in an
idencical way, first ma.inta.ining that tle tath'iigata is neither identical nor different from the aggregates, and then proceeding to questioo the existeocc or
non-existence of the talhagata after death. As in che Sa'1pyulta passage, Nagarjuna immediately qu,alifies his rcfereoq:: to ,the living tathagata, insisting that
the tath'iigata or the buddha under invscigation is one possessed of self-nature
(s11abha11a) and beoce similar to the nodon of "self' (a~ta) rejectcd by Saripuna.
Nagarjuna thereupQn uses his f'amous argumcnc that if the tathagata is not
found in cerms of self-nature, he cannot either be found in terms of ochernature (para-bh'ii11a).

3. Pralitya para-bh'iivaf!' yal/_ so 'natmety upapadyate,

yai canatm'ii sa ca kathaf!' bha11i;ya1i tath'iigataf?.
He who is dependent upon other narure would appcopriatelf be without
self. Y et, how can he who is without self be a tathiigata?
MKV(P) p .437; MKV(V) p. 189.

If self-oature (s11abh'ii11a) wece co be equivaJent co self (atman) as an enticy in

304

'THE

PH!LOSOPHY OF THE MIDDLE W AY

it.self, tben whaccver is of "other-nature" (para-bhiiva) will be "no-self'

(aniitma) . "No-self' in this parcicular sense, which implies absolute difference,
i.s not acceptable to Nagarjuna, for ie i.s a recognition of a difference between a
tathigata and the psycbophysical pcrsonalicy.

4. 'fadi niisti svabhii11a ca para-bhavafl kathaf{Z bhavel,

svabha11a-parabha11abhyaf{Z rte kal; sa tath'iigalal/.
lf thece exists no sclf-nature, how couJd thcrc be other-nature? Without
both self-nature and other-n.ature, who is this tathiigata?

MKV(P) p.437 ; MKV(V} p .190.

The idea that if there were to be no self-oaturc other-nature too would not be
evidcnt was already emphasized at I.3. A third metaphysical alte.?nacive that
uacscends both self-narure and other-natuie is herc denied.

5.

Skandhan yady 11nupadiiya bhavel kaicil tathagatal/,

sa idiin"im upiidadyad upadiiya tato bhavet.

lf there were to be a tathligata because of non-graspiog on to the aggrcptes, he should still depend upon them in the present. As such he
will be dependent.
MKV(P} p .438; MKV(V) p. 190.

In the early di.scourses, a person in bondage and therefore in a state of suffering

(dukkha) is explained in terms of the five aggregates of grasping (upadiinakkhandha) . A person who is freed is said to be withb uc gta.sping (anupadiina),
buc not wfrhout the aggregates. The gerund upadiiya was used in the discourscs
to express two different meanings, namdy, (i) ''clinging to" (sec an-upadiiya,
Vin l.14; A 1.162; 4.290, etc.) and (ii) "d epcoding upon" (D l.205, kii/an ca
I'"""Jafl ca upadiiya). lt was onJy in the fa rmer sense that a person was said ro

EXAMINATION OF TA11iAGJ\TA

30S

be freed from the agg~egates, not jo the latter sense. Yet, with the development of substantialism, that is, when a freed person carne to be looked upon as
being totally different from the one in bondage, cach having his own nature
(.wabhi11a)(see Chapter XXV), the freed one was pcrceived not only as one not
gr25ping (anupidiiya) on to the aggreg:ites, but also as one who is independent
of the aggregates. lt is this subsrantialist interpretation chat Nagatjuna is
criticizing when he points out chat if a tathagata were to exist without grasping
on to the aggregatcs (skandhan anupadiiya), he will still be dependent upon
(11padadyad) tbem ac the prcsent time (idiinl'?I), that is, as long as he is aJive.
Nigatjuna was thus going back to the Buddha's owo definicion of a freed one.

6. Skandhan capy anupadiiya nasti kacit tathagataf?,

yaf ca niisty anupadiiya sa upidiisyate kathaf!J.
There exists no tathiigata independent of the aggtegates. How can he
who does not cxist depcndenty be graspcd?
MKV(P) P.438; MKV(V) p .190.

Nigarjuna seems co be using the passive vcry upadiisyate in an

epistcmologicaJ sense. In the previous verse. he maintained thac a living
tathagata should be dependent upon the aggregatcs, even thougb he docs not
grasp on to them . Herc Nagarjuna is re-asserting the same position, wheo he
says: "There exists no tathagata who is jndependcnc of the aggregates." The
reason for this is that such an independent tathagata, being a mctaphysical entity like the atman, cannoc be grasped or known.

7. Na bhavaty anupadattam upadiina'!' ca ki'!'cana,

na c"ftsti nirupiJdiJnaf? kathrJf'!ICrJna tathagalaf?.
There is no sphere of non-grasping, nor is rhere something as grasping.
Neither is there someone who is without grasping. How can there be a

tathiigata?
MKV(P) p.439; MKV(P) p.190.

THE PHJLOSOPHY OF THE MIODLE W A't

306

The metaphysical speculacions regarcling identity and differcncc (svabhava,

parabha11a) leave the dependent or the gcasped (up'iidatta), depeodeoce or
grasping (up'iidiina) as well as the one who is free or independent
(nirupadiinaf?). wichout any basis. H ow can there be a tath'iigata? The fact that
it is only an explaoation in terms of identicy and difference that is rejected herc
is dcarly inclicaced in the verse that folJows.

8.

TatJ11'iinyatven11 yo nlirli mrgyamiif!a ca pancadh'ii,

upiidiinena sa katha~ prajflapyate tath'iigataf?.

He who, sought for in the fivefold manner, docs not exist in the form of a
different idcntity, how can tbar tathagata be made known throogh
grasping?
MKV(P) p.439' MKV(V) p.190.

The substancialiast explanation of a tath'iigata would imply that he has completely uaosformed himself in to a d ifferent entity, chat is, a tathagata having
his own-nature (wabh'iiva) with no relatiooship to the person in bondage .
However , examining the fivefold aggregates, no such encity can be dis<:overed.
Such a tath'iigata cannot be explained in terms of dependeoce (upadiinena) .
Thus, the conceptioo of depeodence (up'iidiina) is incompatible witi both
idcoticy a nd difference. What is denied herc is neither dependence nor a
tath'iigata, but merely the m ctaphysical approaches co both dependence and

tath'iigala.
lnstead of rranslating tatt11a and anyatva and as idencity and diliereoce, we
bave rendered che phrase as "different idenity" since it occurs in the singular.

9.

Yad aprdam uptliinatp tal wabhavan na vidyate,

s11abh'ii11ata ca yan nasti kuliJJ tat parabhavatal}.

This grasping is no t foond in terms of self-narure. How can that which

does not exist in terms of self-namre come 10 be in terms of
othcr-natwc.?

MKV(P) pp.439-440; MKV(J1 p. 191.

ExAM1NATI0NoF

TAmAcATA

307

Just as much as a tathagata cannot be e.xplaioed satisfactorily by reying upon a

theory of .identity (111abha11a) or difference (parabhiva), eveo so grasping itself
(upadiina) cannot be found in terms bf self-oature or other oature. In other
words, ie woud not be appropriate to cxplain grasping as an inevicable act or
tendency in human beings.

10.

Evarrz fUnyam upadiinam upadiita ca sarvafaf?,

prajnapyale, ca fUnyena ka1harrz funytlJ 1a1hiiga1af?.

Thus, grasping and gra.spec arc empty in every way. How can an cmpty
tathagata be made known by somcthing that is empty?
MKV(P) pp.440-441 ; MKV(V) p.191 .

Thus, grasping as well as the one who grasps are empcy in every possible way.
They are devoid of aoy substance.
For the substantialist, the lalhagala as well as the aggregates have substance
or owo-.nature, even though they arc clifferent fromone another. The substantialist can cxplain the tathigata, whose self-narure (svabhiva) is freedom, contrasting him with the one who is in bondage as a result of his being a "graspcr"
(upadiitr) and whose nature is discina (para) from that of the tathagata.
However, witb Nagarjuna's deniaJ of self-narure, thar sharp dichotomy aJso is
dissolved. Hence Nagarjuna's insiscence chat an empty tathagata cannot be
made koown in relation to an equally empty ''grasper" (upidiitr) or "grasping"
(upadiina) referred to in the previous verse.

11 .

U.nyam iti na vaktavyam Qfunyam iii va bhavet,

u6hayarrz nobhaya-f!J celi prajnapty artha'J?'J tu kathyate.

"Empty," "non-empty," "both'' or "neither" - thcse should not be

dedared. lt is expressed only for the,purpose of communication.
MKV(P) p.444; MKV(V) pp.192-193.

Note again the use of the tli-formula. Nagarjuna is rejecting any theonzing

308

iHE PHllOSOPHY OF THE MIDDLE WAY

rcgarding either the "empty" or the "non-empty" or both or neither. Neither

the cmpty nor the nonCmpty should be ceificd. These terms arc used onJy for
the sake of communicating or exprcssing an cxpe.rience which, being dependent (prmtyasamutpanna), has no static self-narure (svabhava), and as such
canooi: be demarcated and reified. The subtle difference between vaktavya
("should be declared") and kathyate ("is spoken of, isexpressed!') should not go
unnoticed . For Nagarjuna, dedaration and expression a.re two differcnt ac
tivities, the former caJling for unqucscioiled acceptance, a sort of cacegorical
1
imperative, the laner. lcaving room for modificatioo depending upon the contcxt.
. This stacemeot of Nagarjuna is better understood in the light of the analysis
of conceptS provided by a modern philosopher like William James who
recognizes a pragmatic view of language and truth, in contrast to substantialist
philosophers like Plato for whom "concepts" arc the incorruptible reals.

12.

Siifvafiifiivatiidy

atra kutal? rante calu!faya??J,

antananfiidi capy atra kuta/? 'iinte catu1faya??J.

Hw can the tettalemma of e,ternal, non-eternal, etc., be in the peaceful?

How can the tetralemma of finite, infinite, etc., be in the peaceful?
Mk1/(P-) p.446; MKV(V) p .194.

Siinta is .an appeased one. It is the tathagata who has appeased the dispositions,
'obsessions as we11 as the object (sec comments oo V.8). For him, the fourfold
metaphysical alternacives eicher cegarding the duration of phenomena such as
erernality (a11ata) and oon-eteroality (aiivata) or regarding the extent of
phenomena such as the finite (anta) and the infinite (ananta) do not exist. H e
has stopped brooding over the past and ruoning after the future.

13.

Yena griiho grhrtas tu ghano 's5ti tathagtrtaf?,

niis5ti sa vikalpayan nirortasyiipi kalpayet.

Discriminating on the bas.is of grasp.ing or the grasped, and firmly in-

EJw.JNATION OF TA m AGATA

309

sisting that a tathilgata "exists" or "does not exist," a person would think
similarty even of one who has ceased.
MKf/(P) p.447; MKV(V) p. 194 .

The Buddha's reasons for the rejection of the notion of tath'iigata in the
Sa11Jyutta passage refcrred to above (see comments on XXII.1-2) are clearly emphasized by Nagarjuna. The Buddha rejected a tath"(igata bccause his existence
was asserted in a real and absouce sense (saccato thetato): When Nagarjuna
spcaks of a dogmacie grasping (gt'1ih11) on to something as the real or substantial
(ghana) existence or non-existence (asfiti . . . niisfitt) of the tathagata, he was
expressing the sentiments similar to those of the Buddha. Fu!thennore, arguing in a.sim.ilar way as the Buddha did, Nagarjuna maintais that the same sort
of substantiaist specuations lead to the views regarding the existence and non'existence of the tath'iigata even after his death.
Hece chere is no denial of a talhag4ta, but only of a substantial entity. The
verse tbat follows is unequivocal in this regard.

14.

Svabhavata ca :fUnye 's1nif?I cinfii

naivopapadyate,
para'f{l nirodhad bhavati buddho na bhavatiti .Va.

When he is empty in terms of self-nature, the thought that the Buddha

exists or does no~ exist after death is not approptiate.
MK.V(P) p .447; MKV(flj p.194.

Hete the term asmin, in the locative absoute construction (asmin s11abbii11ataf
ca f unye), refers to the t11tliiigatt1. If the tpthiigata who is alivejs, empty of selfnature, then it ~not appropriate to assume that he exists or does not exist after
death. lt is only the substantialist thinking (cintii) that leads to the
metaphysical quescions which wece left undedared (avyakrta) by the Buddha.

15.

Prapancayanti ye buddha'f{l prapacafitam 4vyaja7[1,

te prapa11cah4taf? s(lf't)e na paiyanti tath'iigata'f{l.

THE PHILOSOPHY OF THB M IDDl.li W AY

310

Those who geo.erate obsessions with greac regacd to the Bpddha who has
gone beyond obsessions and is constant, all of them, impaired by obsessions, do not perceive the tathiigata.

MKV(P) p.448; MKV(V) p.19~ .

Here we radically differ from the prevailing cxplanations that prapaca means
thought distinct from reality (see !nada, p . 135). Having rejected the moce
widespread view that Buddhisrn recognizes an "unspeakable" (a11acya) and "indefinabJe" (anirvacanlya) truth or reality (tattva) , which Jeads co the above iq.
terpretatioo of prapaf!_ca, we have reodered the term as "obsessioo. ''
The Buddha remains aoof from obsessions (prapacali'ta"!t) . As such, he is
not understooc or grasped by those who are obsessed. A person who is obsessed
with the idea of identity will understand the Buddha in a way .different from
one who is obsessed with the idea of difference. One wilJ say that he "exists"
(astt) and the othec will insist that he "does nor exist'' (niistt) .
The Buddha who has overcome such ''obsessions" is "not so variable"
(a11yaya?fl) . His perceptions are not variable in the same way as those of the
unenightened ones who are dominated by obsessions. Nagarjuna could not
have been unaware of the defin:ition of the Buddha as "one who has become
stable and ste;JY' ( f hita1'{1- ane1jappatlaf!l, A 3. 37 7; f hita7?1 ciltaf!Z, S 5. 74).
T he term a11yaya in the present coutext expreses the same idea of stability and
steadfasroess achieved by a Buddha. This is not to assume his permanent ex
1Steoce.

16.

Tatliiigalo yat s11abh'ii11t,M tat s11abh'ii11am idaf!I jagat,

tathagato nil/s11abha110 n'i'1s11qbha11am idaT!l jagat.

Whatever is the self-naturc of the tathiigata, thac is also the self-nature of

the universe. The tathiigata is devoid of self-nature . This universe is also

dcvoid

of self-nature.
MKV(P) pp.448-449; MKV{V) p .195.

T he first statement of Nagarjuna may be takeo to mean thac there is a self. nature of the tathigata which is identical with ~hat of the universe. This wduld
be justification for the belief in a permanent entity wh.ich is identical witb the
reality of the universe, comparable to the atman and brahman of the Hindu

EXAMlNATION OP TAmAGATA

311

uadition. However, Nagatjuna irnmediatdy sounds the warning that there is

noc such self-natu re either in the lalliiigata or in the u niverse. Trus would mean
that non-substantiality (nairiitmya) applies not only to the individual (pudgala
= tathagakl) buc also co all phenomena (dharma = jagat).

CHAPTER
TwENTY THREE
Examination of Perversions

(Viparyasa-pan k! a)

1.

Sa~alpa-prabht1110

rago d11efo mohtli ca kathyate,

ubh'ii1ubha-viparyaiiin sa1{tbhavt1nti prafitya hi.

Lust, hatred, and confusion are said have thoug}tas their source. Perversions regatding the plea.sant and the unpleasant arise depending upon
these.
MKV(P) p.451; MKV(V) p .197.

Followiog upon a discussion of the tathagata in reJation to the world, a discussion thac avoided absolute identicy and ab$olute difference, Nagarjuoa rakes
up the questioo regarding perversions (11iparyasa) which makes a difference between a person in bondagc and one: who is freed .
The four perversions are discussed by the Buddha at Anguttara 2.52. Tbey
pertain to percepcions (saflflti), thoughts (citta), and views (diftht). Perversion
(vipalliisa) of perceptioo or thought or view occurs with the identification of
the impermaoeot with the permaoent (anicce niccan tt) ,
11. the not unsatisfactory with the unsatisfactory (aduk.khe dukkhan h),
UL the nonsubstantial with the substantial (anattani alta ti) ,
and
1v. the unpleasant with the pleasant (tliubhe ubhan lt).
t.

It is important to noce that perversion ii is. based upon perversion i , and

perversion iv is based upon perversion iii . While perversions li and iv relate to
subjective attitudes, perversions i and iii arc cognitive in nature.
The rccognition of subjective perversions (i.e. ii and iv) does not mean that
those experieoces 1elting to whith pervcrsions arise do nor exist. lndeed, experiences of the peasant (ubha = man'iipa) and unpcasanc (aubha =
3 12

EXAMINATION Ol' PERVERSIONS

313

amanapa) sensations occur even in the enlightened ones (Iti11uttaka 38). Perversion is the wrong identification of these exp~riences . However, the same cannot
be said of the cognitive peNersioos, for the Buddba would be reluctant to
assume that there is an experience corresponding to something that is either
permanent or substaotiaL
For Nagarjuna, perversions ii and lv arc not due to purely inteUectual or
cognitive incapacity on the part of the indlvidual. Tbey are due more to the
subjective elements dominating thoughts sur.h as lust and hatred, wbicb acise
along witb thoughts (sa1{Zkalpa-prabhavo). even though these lattcr are not
part of the wodd of experience. Th.is explanation of Nagarjuna is based u pon a
passage in the St11(1yulla wherein the Buddha identifies desire (kQ,ma) with
thoughts of lust (sa1{Zkappa-rag4) without confusing it with whatever is
beautiful (citrint) in the world (sec S 2.22; Tsa 48.20 [Taisho 2.345b]}.
As such, it is not surprfai.ng to see Nagarjuna beginnibg his analysis with the
last of the perversions, namely, the identification of the pleasant with the
unplcasant (a.ubhe ubhan Jt), a subjective perversion that is based upon the
cognitive. confusion (i.e., anattani atfii tz). Nagarjuna was probably interested
in beginning his analysis with the subjective perversion because the metaphysicians, who carried che analytical process beyond its limit, bad difficulty in accounting for the emergence of sensations such as the pleasant and the unpleasant.

2. Subhiubha- viparyaiiin saf!Jbht111anti prafitya ye,

te svabha11a11 na 11idyante tasmiit klea na tatl'llataf?.
Whatever perversions of the pleasant and the unpleasant that occur
dependendy arc not evident in terms of self-nature. Therefore, the
defiements acc not in themselves.
MKV(P) p .453; MKV(V) p.197.

The perversions of the pleasant and the unpleasant are dependcntly arisen. As
mentioned before, t hey are tht products of 1ust and hatred; they arise within
the individual . For this tea,son, they are not found in themselves (na
s11abhavataf?). Whatever defilments (.klea) occur as a result of such perversions
annot be pa.et of the expetiemced world. Rather they related to the way in
which the world isperceived by .the individual. lt is ooly in this sense that they
are looked upon as being unrcal (na.tattvata'f?).

Thra PHlLOSOPttY OF THE MIODLE W AY

314

3. Atmano 'JtittJa-n{istit11e na krJtha'f?Ztc aa sidhyatal;,

lam vinastitva-nastilve kleSiin'li'f?J sidhyatal; katha'f!'.

The existence or the non-existence of the self is not established in any
way. Without that, how can the existence or the non-existence of
defilemenu b-established?
MKV(P) p.453; MKV(V) p.198 .

The perversin regarding what is pleasant and unpleasant has been traced back
to the pecversion whcre the non-substantial (anatman) is .identificd with che
substancial (atman). The substantial and the non-substantial, in the eyes of the
mctaphysiciari, pertain to existence (flJtitva) and non-cxistence (n'1isti111a)
respectively (see V.8; XV. 71 11). The discourse to Katyayana .rejects both thesc
views as being metaphysi1=al.
Nagarjuna is herc insisting that neithc.r the 6cistence not the oon-existence
of a metaphysical self can be proved. lf there were to be no such self, then the
defilcments associated with such a self also cannot exist. In the absence of a
substantial self, if only the defilernents wece to exist, then they should have selfexistence, which was an idea denied at XXIII.2. Tuus, neither the substantial
existence of a self noc the substantial existence of defilrnents can be established.

4. KllIJacidt/ hi bhavanfime kiesa!? sa ca na .ridhyati,

kacid aho

11in1i ki'f?Z(;it s411ti k/eSii na kasyacit.

These defilements, indeed, bclong to someone. Yet, such a person is not

cstablished. In other words, in the absence of anyone, these defiments
seem o exist without belonging to anyonc.
MKV(P) p-453 ; MKV(V? p. 198.

Not onJy are the defilments not established as substantia elements, they can
not be proved to exist even as attributes. If defilements are considcred as
belonging to someone(.kacia) who issubstantially existing. the abscoc~of such
a substantial entity would mean the absence of de(llemcnrs as attr.ibutes.

Ex.AMlNATION OF PERVERSIONS

315

5. SvaA'iiya-dr1#11a1 kleiil? k/ifte santi na pancadha,

waliaya-dr1!i11at kii/ta'!' kleie1v api na pancadha.

The defilments arc like the view of one's own personality. Within the
defied, they arc not found in the fivefold way . The defiled is like the
view of one's own personality, for even within the defilements it is not
found in the fivefold way.

MKV(P) p.454; MKV(V? p. 198.

This is a utber cdcicism of the substantialist notions of "defilemenrs" (klefa)

as well as of the "defiled" (k.li$1<1). Such substantialist perspectives arc comparable to the views regarding an "embodied person" (111ak.aya-dr1/1), who is
not obtainabe when that personality is analysed into the five aggregates.

6. Svabhavato na vidyante fubhaiubha-11iparyajiil/,

prafitya k.atamiin kfefiil? iubhafubha-viparyayan.

The pcrvecsioN regarding the pleasanc and the unpleasant arc not evident from the standpoint of sef-nature. Depending upon which perversions of the pleasant and the unpleasant are these defilements?
MKV(P) p.455 ; M.KV(V) p. 199.

Just as much as botb the deft.lements (k.leia) and the defiled (k.li;Fa) arc not evident in substantial form, so are the perversions that give rise co dcfiements .
Tbey too arc not fouod in aoy substantial way. The question theo is: What is
the nature of the perverson relating to the pleasant and the unpleasant depending upon wbich defilements are said co arise?
Here there is no denial of perversions nor the deftlements. What has been
quescioneq is only the way in which theso are cooceptualized.

7.

Rupa-fabrla-rasa-sparf'ii gandha dharm'ii ea 11Z{ilidha~,

vas/fJ ragasya dve/aJya mohasya ca vikalpyale.

THE

316

PHILOSOPHY OF THE MIDDLE

WAY

Material form, sound, taste, touch, smeU and concepts- these are
discriminatcd as the sixfold foundations of lust, hatred, and confusion.
MKV(P) p .456; MKV(V) p.199.

Analysing !ust, hatred, and coofusion in terms of the four perversions, it may
be maintained that Just and hatred arc subjective attitudes, while confusion
refers to the cogni tive aspect of understanding (sec com.men ts on XX1lI. I),
even though the cognitive and artirudinal aspects are dependent upon one
another. The present statement of Nagarjuoa implies the dependence of the
attirudinal as well as the cognitive aspects of experience upon the sixfold objects
of sense experiencc.

8. Rupa-fabdarll.fa-spari"ii gandha dharma ca kevaliif?,

gandharvanagarak"iir7i

maficz~svapna-sa?rt,nibhaf?.

Material form, sound, taste, touch smell as well as concepts-all these are
comparable to the city of the gandharvas and resemble mir.ages and

dreams.
MKV(P) p.457; MKV(V} pp.199200.

The similes of the "dreamH (svapna) and the "city of the gandharvas"
(gandharva-nagttra) have already been employed, a.long with :ilJusion"
(maya), to refute the substantialist cxplanation of the dispositionally conci
tioned phenorncna (sa??ZskrtaXsee VII. 34). The six objects of experience referred to at XXIII. 7 arc indeed dispasitionally conditioned. They arc not objects
that arc found in themselves (s11abh411ataf?). Nor arc they absolutey noneXJstent.

9. AubhtTf{I va fubhaf(t vapi kutas teru bhavi[yati,

maya-puru!a-kalpe!U pratibimba-sametu ca.

ExAMINATION OF PERVERSIO"NS

317

How can the pleasant and the unpleasant come to be in people who are
fabrications of illusion or who are comparable to mirror images?
MKV(P) p .458; MKV(V) p.200.

What Nagarjuna has bee n criticizing so far is a substantiaist theoi:y of the pleasant and the unpleasaot (svabhtii/alo , .. fubhaJubha11iparyayan XXIII.6).
His argument in the prescnt verse is directed against the belief that such
substantial expeticnces can occur in thoughts of hwnan beings who are
themselves non-substa11tial, and who a.ce comparable to mirror images or reflections (see commentary on :XVII.31-32).

10.

Anapek!ya :fubhaf!Z nti.rti afubhaf!Z prajapayemahi,

yat prafitya fubha'f!t tasmtic chubhaf!Z naivopapadyate.

We make known that the unpleasant does not exist without being conti,ngent upon the pleasant, and that the pleasant, in its turo, is dependent upon that [i.e. the pleasant] . The.refote, the pleasant [in itself] is
not appropriate.

1I'.

AnapekfyiiJubhaf!Z ntisti :fubhaf!l prajfJapayemahi,

yat prafityaubhaf!Z ta.rmad aubhaf!Z naiva vidyate.

We make known that the pleasant does not exi$t without being contingent upon the unpleasant, and that the unpleasant, in its turo. is
dependent upon that [i.e. , the peasant] . Therefore, the unpleasant [in
itself] is not evident.
MKV(P) pp.458-459; MKV(V? p ,200.

This is a elear refutation of the substancialist views of both the pleasant and the

318

THE PHJLOSOPHY OF THE

MIDDLE WAY

unplcasant. Nagarjuna insists that thcy ace dependent upon one another. Thcy
arc not independent experiences wherc one is repaced by another, as in the
case of the theory of "attainmcnt" (praptz) and "oon-auainment" (apraptt)(see
comments on .XVI.10, XXIV.32).

12.

Avidyamiine ca fubhe kuto rago bhavi,yah~

aUbhe avidyamiine ca kuto dvefo bhavifyali.

When the pleasant is not evident, whence can there be lust? Wben the
unpleasant is not cvident, whence can there be hatred?
MKV(P) p.459; MKV(V) p.201.

Io addition to being mutuaHy dependent, the plcasant and the unpleasant provide a foundation for the subjective tcndencies such as lust and hatred. Thus,
lust would be non-ex.isrent if the pleasanc were not cvident. Similarly, hacred
would be non-ex.istent, if the unpeasant wece not evideot. This, indeed, is the
conclusion of the Sa'f!'yutta passage referred to at XXIII.1. In that context, the
Buddha was not denying the pleasant and the unpleasant expericnces in the
world. He was simply insisting chat w hen such experiences occur a wise man
resuainrs his ycarning (chanda) for it.

13.

Anitye nityam ity evaf'!J yadi graho viparyayaf?,

nanilya1p virlyate funye kulo graho viparyayaf?.

If tbere were to be grasping on to the view, "Wbat is impermanent is permanent," then there is perversion. The impcrmancnt is not evident in the
contcxt of the empty. How can there be grasping or pecversion?
MKV(P) p.460; MKV(V) p .201.

So far Nagarjuna has endeavored to show thac the perversion regarding the
pleasant and the unpleasant is che result of a cognicive confusion which led to
che belicf in a substance ot self (atman) . Wich the present statement, Nagar-

Exl.MINATION OF PERVERSIONS

319

juna begins co analyse the conception of impermanence (11nity11) in order to explain che lathagata's cognitive uoderstanding of the world of experieoce. This
understanding enables the tathagata co eliminate the perversion relating to
happiness and sufferlng .
Herc, the perversion pertains to grasping of the impermanent a:s the perma
nent. lf so, the perspectivc in wbich the lalhagata looks at the world should be
one of impermanence. Unforcunatcly, the notion of impermanence chat
dominated the Buddhist tradition during Nagarjuna's day was more a
metaphysical one (see commentaries on Chapter Vll) than an einpirical one.
For the metaphysician, the absence of permanence implied the reality of the
momeocary. Nagacjuna rightly believes that as much as grasping after permanence is a perversion. so is grasping aft er the reality of the momentary ( =
k1an.1'lla). The deoial of permanence does not commit oneself to the ocher ex
cceme of momentary.destruction (k/an.a- bhaa). Emptiness (unyafii) does not
imply any such momentariness .

14.

Anitye nityarn ity eva~ yadi graho viparyayaf?,

anityam i~y api gr'iihaf? iinye kif{J na viparyayaf?.

lf gcasping on to the view, "What is impermanent is pecmaoent," is

perversion, how is it that even the grasping after che view, "What is e1npty is impermanent," does not constitute a perversion?
MKV(P) p.462 : MJ...'V(V) p.202 .

It is possible for someone to raise the quescion as to whether grasping after the
empry would irself be a perversion, just as much as grasping after che permanenr or the impermanent would consritute perversioos. Nagarjuna is herc raising this question and chen proceeds to answer it in the following verse.

15.

Yena grhr;ah' yo grliho grahTt'ii yac ca grhyate,

"paantani saniiitJi tasmad gi?iho na vidydttte.

That througb which there is grasping, whatever grasping there ~. the

320

THE PHILOSOPHY OF THE MIDDLE

W/\. Y

grasper as well as that which is grasped - all these are appeased.

Therefore, no gcasping is evident.
MKV(P) p.465; MKV(V} p .20.3

The middle path enunciated in the discoutse to Kacyayana avoided the two extremes relating to philosophical understanding and explanation. This was
achieved thro1Jgh che appeasement of dispositions and the elimination of
grasping. Because of non-grasping, all che metaphysical qucstions relating to
the faculcies (yena grh11iit1), the proces.s of understanding (griiha), the person
involved in such understanding (grahtfii) as well as the object of uderstanding
(yac ca grhyate), come to be appeased. The appeased one (fanta) does not continuc to taise questions or doubts beyor.d a certain limit, not because he knows
evetything nor bccause he does not care to know, but because he is aware of the
conflicts generated by any pursuit of koowledge that gocs beyond experience.
The facr chat grasping for emptiness can conscitute a sort of petversion
(f'iinyatii dr1tz) has alrea.dy been e.xplained by Nagarjuna (XIU.8). This does
not mean that t he very conception of emptiness is invalidated in the same way
as the conceptS of permaoence and momentariness would be invalidated, The
reason ff!r this is thar the notions of permanence and substance as wel as of
niomentarioess arc not empirically grounded compared with dependent aiising, non-substantiality, or emptiness. Both the Buddha and Nagarjuna would
categorize the formcr under wrong thoughts (mithya sa1?Jka/pa), while tbey
would consider the latter as right or appropriate thoughts (saf!Jyak saf!Jka/pa,.
ka/pana yatra yojyate, XVll. 13) , The verse thar follows should be understood
in such a conteXC. This indeed is the fina conclusion of Nagarjuna in the
Kiirika (sec XXVII.30).

16.

Avidyamiine grahe ca mithfii va sa1nyag eva va,

bh11ved viparyayaf? kasya bhavet kasyaviparyaya/i.

When grasping; wrongly or righty, is not evident, for whom woud the.r e
be perversion and for whom woud chere be non-perversion?
MKV(P) p.466; MKV(V) p.204.

EXAMINATION OP PERVERSLONS

32l

With grasping gone, whether it be for the right thing or the wrong thing, the
enlightened one does not involved himself in any substancialist thinking
rdating to perversion or non-perversion. This is the state of the saint who has
anained freedom from grasping (anupiidaya vimuktt).

17.

Na capi vipafftasya sarrzbhavanti viparyayaf?,

na capy t1vipantasya 1ambha11anti viparyayaf?.

Perver.sions do not occut to one who is aready subjected to pcrversion.

Pervcrsions do not occur to one who has not been subjected to pcrvers1ons.
.

18.

Na 11iparyasyamiint1$ya sarrzbhat1anti t1iparyayal/,

1111rJrfasva 111ayaf!Z kasya sa1(Zbha11anti viparyayaf?.

PerveISions do not occur to one who is being subjected to pcrversioru.

Reflect on your own! To whom will the pervcrsions occur?
MKV(P) p.467: MKV(V) p.204.

Upon reaching sucha evel of mora and intellectual development (as referred
to in the previous verse), one does not get involved in mecaphyska spcculations such as whether pcrversions arise in one who is a1ready afflicted by pcrversions, or not afflktcd , or is being afflicated. The speculation thac is avoided
pertains not only to the past and the future buc also the present. As such he
avoids the metaphysics discussed in Chapter Il . When perveisioos themselves
are not per~eived as being substantial, how can one consider a perverse person
as a substantial entity.

f9.

Anutpannii katharrz nii1114 bhavi,yanti viparyayaf?,

11iparyayef11 a1atesu 11iparyayg-galfli? kuta/?.

How coud there be non-ariscn pcn;ersi~ns? When perversions are noc

322

THE P!llOSOPHY OF THE MIDDLE W AY

bom, whencc can there be a person who is subjccted to perversions?

MKV(P) pp.467-468; MKV(V) p.20).

Picking up the argument from the previous verse, Nigarjuna is bere specifically
rejeccing the su bstantiaist ootions of perversions as well as persons who arc
pervcrse. lf perversions are substancial or having self-narure (s11abha11a), they
cannot be considered as arisen (utpanna) . When they are non-arisen or unborn
(ajata), there cannot be a person in whom these would arise, for they cannot
serve as attribuces.

20.

Na svato /iiyate blia11af? parato naiva jiyate,

na s11ataf? parata ceti viparyaya-.gata/t kutal/.

An existent does not arise from itsef, nor does it arise from another, nor
both itsef and other. If so, whence can tbere be a person who is subject to
pcrversions?
MKVP) p.468; MKV(V) p.205.

This verse, which is not available in KumlirajTva's translation, represents U application of the analysis of substancial existence (bhava) in Chapter I to the problems of a substantial petsoo or eotity. It is aJmost identical with XXl.13.

21 .

A Jmi ca luci nityaf!Z ca sukhaf!Z ca yadi vidyate,

atmii ca luci nityaf'!l ca sukhaf!Z ca na 111paryayaf!.

If either the self, the pleasant, the permanent, or the happy is evident,
then neither the self, the pleasant, the pcrmanent, nr the happy constitutes a pcrversion.
MKV(P) p.468; MKV(V) p.205.

Nagarjuna began his examinacion of pcrversions (11iparyasa) utilizing rhe same

EXAMJNATION OF PF.RVERSIONS

323

terms used br the Buddha at A 2.52. In the preseot verse, he adopts a slightly
different terminology in referring to the perversions. Instead of fubha and
aubha, he uses the p~ of terms uci and .auci. and instead of adukkha be has
sukha. This latter terminology was indeed utilized by che interpreters of
Abrudharma when discussiog the perversions (see Akb p.286). As such, it is a
elear indication that N:igarjuna was conversan t withg the early discourses as
much as he was familiar with the Abhidharma commenrarial Hterature.
Nagarjuna's argument in the present verse coud lead tp misunderstanding
unless it is seen in the context of his previous sta.cements abouc the perversions,
expecially at XXIII.20. lf the self, the pleasant, the petmanenc, and the happy
arc evidenc as selfexistent entitzes, chen, indeed, they do oot constitute perversions. Unfortunately, Inada's uanslation does not seem to bring out this strong
sense of "exist" that Nagarjuna is implyicig here.

22.

Niit11Jii ca fuci nitya?!I- ca -sukha'f!J ca yadi vidyate,

anal1nii 'ucy anitya'f!J ca naiva du-1/Aha?!I- ca vidyate.

If neither the self, the p~easant, the permanent1 nor the happy is not evident, chen neitbec the non-self, the unpleasant, the impermanent, nor .
the $nffering would also be evident.

MKV(P) p.469; MKV(V) p.205 .

Having rejected the substancia existc;nce of the four perversions, Nagarjna is

here emphasiziog their relarivicy. 1-.owever, relativicy in this context need not
be understood as applying in an identical way to all the four perversions. As
pointed out earlier (see comments on XXIU .1), the four perversioos deal with
rwo different categoties, the cognitive and che attirudinal. The fact thar
anatman and anitya are conceptually related to atman and nitya does_not mean
that the latter are cognitivcly bascd . They arc relative only at the .cooceptual
leve . On the contrary, fuci and aiuci as wel as sukha and du/?kha arc part of
experiern;e, even though chere coud be confusioo regarding theit identificadon.

23 .

Evaf'{Z nirudhyate 'vidya r1iparyaya-nirodhanat1

avidjiiya?!I- tJiruddhayaf'{Z sa'f!JSkaradya?!I- nirudhyate.

THE PHILOSOPHY Of THE MIDDLE W/\ Y

324

Thus, with the cessation of pcrversions, ignorance ceases. When ignorance has ceased, the disposition.s, etc. come to cease.
ltfKV(P) p.469; M.KV( V) p.206.

The distinctioo betwcen the two kinds of perversions meotiooed earJier

(XXlll . t ,2i) seems to be juscified by the.present sratement. Wben the perversions cease, the immediate resut wouJd be the ccssatioo of ignorance (a11idyQ) .
This would pcn.a in to the more cognitivc pcrversions (i and iii). The cessation
of igoorancc would aJso mean the cessacion of the perversions relatiog to
dispositions; that is, pcrversions ii and iv .

Yadi bhutal/ 111abha11ena klefal/ keiidd hi lllZJyflcit,

kfllh.11rt1 n'iima prg/:iiyeran kall 111abh'ii11art1 prahasyati.

24.

lf, indeed, certain dcfile!bents of someonc have come to bt: un the basis
of sclf-naturc, how could they be rclinquished? Who cver cou1d relinquish self-nature?.

Y.ady abhutal/ svabhavena k/efal? kecidd hi lllZJyaciJ,

kathaf?Z fiiima pral:iiyeran ko 'sadbh'iiva?!J prahasyati.

25 .

lf, indced, ccnain dcfilcments of someooe havc not come to be on the

basis of self-oature, how could they be relinquisbed? Who ever could
rclinquish non-existence?
'

MKV(P) p.471; MKV(V) pp. 206-207 .

Nagarjuna's conclusion in this chapter is that the dcfilemeots (klea) , whicb arc
the rcsults of the pcrvetsions discussed abovc, cannot be elim.inated , and ,
therefoce, there could be no lalh'iigata, if the.se defilements are ooked upon as
being either substanciaJ; i.e,, something that has come to be (bhuta) having

EXAMINATION OF PERVERSIONS

32~

self-narure (s"Vabhiiva) or oon-substantial, i.e., sotnething that has not come to

be (abhuta) because of ics having no real nature (asadbhava). In either case,
there can be no abandoning of defieroe.nts and hence the achievement of the
state of tatliiigata.

CHAPTER
TwEN1YFOUR

'

Examination of the Noble Truths

(Arya-satya-parzkfa)

1.

Yadi unjam ida'f{Z sqrvam rtc/ayo niirti na vyayql{,

caJun{iitn iirya-satyanam abhaPas te prasa/yate.

If all tbis is empty, the n the.re exists no uprising and ceasing. These imply
the non-existence of the four noble truths.
MA'V(P) p.475 ; MKV(V) p .209.

Nagarjuna's discussion of the four trutbs was an inevitable consequence of his

analysis of perversions (Chaptet XX.lll) which, in its turo, was occasioned bya
need to explain the perspeccive of a tathagata (ChapterXXII). The foUI uuths
refer to the problem of suffering, the subject matter of two of the perversions,
namely, mistaking the unpleasant for the pleasant (aubhe iubhan tt) and nonsuffecing for the suffering (aduf?khe dul{khan tt).
Verses 1-6 In the pr~ent chapter pose one major problem faced by the Buddhist (or even by the non.Buddhist) in explaining suffering (duf?kha) and,
therefore, the fout noble truths, when placed in the context of "emptiness"
(fiinyata). Th~ important questions arc: Who are these Buddhists (or nonBuddhists)? Why arc they .faced with such a problem? .The answers to these
questions 'ate found in the Abhidharmakofa-bliii1ya of Vasubandhu.
Vasubandhu refers to a theory proposed by some: "Thete indeed is no fee ling
of happiness" (nasty eva sukha vedana) and "Evetything is suffering or unsatlsfactory" (duf?khaiva tu sarva'f!Z)(Akb p.330). There is very Little doubt that t.his
interpretacion of the Buddha's teaching emerged with and was continued until
modern times by thinkecs and scbolars belonging to the Brahmanical tradition.
By ignoring a simple yet extremely imponant pronoun (namely, idaf{Z, ''this"),
the Brahmanical interpreters transformed the Buddha's teaching from an em
pirical to an ab~outistic system. Thus, the Buddha's scatement: saroam ida'f{Z
dul(khaf?' ("all this is suffering'') turns out to be an unquaLified universal statement: saroa'f{Z duf?kha'f!Z ("everything i:s suffcring''). Similarly, a specific statement such as sunyam idaf!'J saroa'f{Z ("all ti.is is empty") turns out to be another

326

EXAMlNATlON OF 1HE NOBLE

Thurus

327

absolutisti<:: one, sunyaf(t sarvatp ("everything is empry ," a problematic not

noticed by Inada, p. 144).
I
Threatened by such a problematic situation , the Buddhist metaphysicians
wete driven to the extreme position of asse.tting absolute forms of happiness,
when they insisted: "Happy feelings do indeed ex1st in terms of unique
charactcr" ((IJty eva s-11alakfa11atal/ suk.ha vedanii, Akb p.331}. Yet, realizng
that the Buddha did oot prcsent the world as a ''bed of roses," they were compelled to admit that there. arc eonie experiences that arc inherently unsatisfactory (S11abhti11enai11a duf?kha, ibid., p .329).
Thus, not realizing that this was all due to a Brahmanical misreading of the
Buddha's discowses and assuming that there is a genuine problem, the Buddhist metaphysicians carne to admit both suffering or unsatisfactoriness
(duf?k.ha) and happiness (suk.ha) as substantially existing. Instead of correctiog
the Braimanical misreading, they struggled for an explanation:

When [the Buddha] declared: 'Ooe should perceive happy feelings as suffering,' [i.e. , che perception of one who has eiminated
the perversions,] both [hap~incss and suffering] arc available
thercin. Happiness is inherently so, because there is pleasantness ..
However, eventually there is suffering, because of its chaoging and
impeananent nature. When that [feeling] is perceived as hap
piness, it contributes to enlightenment, through its enjoyment.
Wheo it is perceived as suffering. it leads to release, by being non
attached to it. (Akb p.331, Yad uktaf(t "sukh'ii 11edanii dul;khe ti
dra/(a11ye li ubhayaf(t tasyam asti. Sukhat11a?tt ca svabhavato
1

'

ma1ilipat11iit, duf?ka?tJ ca paryaya~o 11ipari111imanitya-dharmat11iit.

Sit tu suk.ha.to drfyamana buddhaya kalpate, tad asviidamz.t,
duf?khato drfyam1in'i!_ mokfaya k.alpate, tad 11alragyad tli)

The substancett~rminology in the above passage should. make it cleat as to

which Buddhist school wa5 presenting tbis eicplanacion. There seems to be no
quescion that it was authored by the Sarvastivadins. Thus, it was their desire to
uphold the cnception of substance that made them uncomfortable with the
notioo of "the empty" (Unya) . The substantialists were prepared to wrestle
with the conceptions of substance and causation and, as shown earlier, they
produced substantialist theories of causatioo . However, they could not do the
same with the concepcion. of "the empty" (Unya) , even though "emptlness"
itself may be amenable to such substantia1ist interpretations (see XXIV.14).
It is in order to highlight this problem that Nagatjuna begins the chap.ter
with a reference to "the empty'' (funya, as in y,adi iinyam idaf(t saroaf(t) .rather
than to the abstract conception of "emptiness" (ftinyalii}. In Nagarjuna's

Tu1 PHILOSOPHY OF THE MIDDLE

328

W/\ Y

mind, to preserve the non-subscantialist and non-a bsoJucistic standpoint of the

Buddha, what needs to be restored is "the empty" rather chan "empciness.''
because the latter is dependent upon the former, whereas any emphasis on the
latter could wipe out the former , as it happened in the case of' the absolutists
and the substantialists. In presencing che metaphysicians dilemma in this man
ner, Nigarjuna was more concerned with the mistake of his Buddhist counter
pans than with the misreadings of the Buddhist texts by the Brahamincal
chinkers. For this reason, he makes no reference to Such misreadings as "sarva'f!t
,
"
sunya'f!t.

2.

Parijna ca praliii'!'f!t ca bhli11an'ii .r'iikfikarma ca,

caturtiiim iirya-saJyiinam abha11iin nopapadyate.

In che absence of the four noble truchs, uoderstanding, relinquishing,

cuJtivation, and reali~tion will not be appropriate.
MKV(P) p.477 ; ll'!KV(V) p.210.

Understanding (parijn'ii) pertains to the first of the four noble truths, namely,
chat suffering exists. Rclinquishiog (prah'ii1'!a) refers to the second, narody, the
cause of suffering, which is explained as craving (tr!t{ii). Realization (i?ik#karma) applies to the third, namely, the state of freedom from suffeting, which is
nirvana. Cultivation or practice (bha11ana) involves the fourth, namely, the path
leadiog to the cessation of suffering.

3. Tad ab/iiiv'iin na vidyante caw'iiry iirya-pl>a/iini ca,

phaliibhave phalasthii no na santi pratipannak'iif?.
In the absense of this [fourfold activity] , the four noble fruits would .not
be evident. In the absence of the fruit, neither those who have attained
the fruits nor those who have reached theway [to ~uch attainment] exist.
MKV(P) p.477 : MKV(V) p.210..

The non-absolutiscic standpoiot of eady Buddhism is clearly embodjed in the

EXAMlNATION OF THE NOBLE TRUTHS

329

doctrine of the four fru.its (arya-phtlla) . Even though freedom without grasping
(anupadii vimuktz) is sometimes referred to as the one goal (ek4y11na), yet a
deliberate attempt to assert degrees of attainments where some arc considered
to be inferior to others is conspicuously absent, Like the relay of chariots
(rathavinlta) , one segment of the path is as imporcaat as the other,
''dispassion" (vairagya) being one dominant attitude permeating all segments.
This has an imporcant bearing on the concepcion of ttuth to be discussed soon.

4.

Saf!lgho n'ihti na cet santi te 'f!tm puru!a-pudgaliif?,

abhaviit carya-satyanaf!I saddharmo 'pi na vidyate.

If the eight types of individuals do not exist, tbere will be 110 congregation. From the non-existence of the noble trnchs, the true doctcine would
aso not be evident.
MKV(P) p.478; MKV(V) p.210.

Inada was probably assuming that cet refers to thought (citta) when he
translaced pan of the first line as "the eight aspiracions of men do not exist,"
whereas it could be more ~ppropriatdy read as a cooditional particie.
The assoc.iation of the ttue doct.tine (sad-dharma) with the fout noble truths
(tiry11-1a1y11) to a point whe:te the abse.oce of th,e atter implies the non-existenc~
of the former should nacurally raise questions regarding the more popular
assigoment of doctrines to the var.ious sehools where the four noble truths are
assign~d to the . so-called H~nayana with the superior Mahayana having
somethiog more to offer, Such an assignroent. becomes really questiooable
when a greac Mahayana thinker like Nagarjuna himself admits of such a cor
relation (XXIV.30).
This equation is also signlficant for another very important reason: Dharma,
as ''teaching" also means a "statement" of doctrine. Some modern linguistic
philosophers would prefer to use the term "truth" primarily. in referring to
statements rather than to facts . Sad-dharma, .identified with truth, seems to accommodate such a perspective.

S.

Dharme casati sa'!l-ghe. ca kathaf'!l buddho bhavifyt#'i,

evaf!J. tnf!y api ratnani bru11lir1111f pratibadhase.

330

6.

THE PH!LOSOPHY OF THE MIDDLE WAY

funyaliif(Z phaltrsadbh'iivam adharmaf!l dharmam eva ca,

sarva-sa'{Z11ya1111har'iif!Jf ca laukikan pratibadhase.

Whcn the doctrinc and the coogregarion arc non-existent, how can there
be an cnlightcncd one? Spcaking in this .m anner about cmptiness, you
contradict the thrce jewels, as wcll as the reality of the fruits, both good
and bad, and all such worldly convenrions.
MKV(P) pp.478-489; MKV(V) 210-213.

At this point the metaphysidan brings up the notion of "emp tiness" (1unydta).
For him, emptiness is a problem, not because of any othcr reason, but because
it implies "the emp ty" (finya), and this latter lcaves no room for substantial existence (sad-bh'iiva) of the fru its (artha) of both good and bad as well as alJ the
worldly conventions-.
Inada's transation of dharma and adharma as prope and improper acts is a
couective to his earlier rendering of these rwo cerms as factor and non-factor at
Vlll.5 . However, his rranslation of saf!J11ya11ahiira as somcthing additional to
dharma and adharma can be quc:stioned in the light of the use of vyavahara at
XVII.24, wherc a similar, but not idential, set of categories (pu11y11-p'lipa,
"merit and demerit") is referred to as vyavah'iira. Jndeed, this particular use of
the term 11y1111ah'iira ( = Illf!lV[lt) will throw much light on an undc:rstandiog of
the distinction becween the ewo truths discussed at XXIV.8.

7. A tra bmmal; iunyatiiyfif(Z na tvaf(Z 11etsi prayojanam.,

iunyaliif!J iunyalllrthaf!J ca talli etJam. vihanyase.

We say that you do not comprehend the purpose of emptiness. As such,
you ar-: tormcnted by emptiness and the meaning of emptincss.
MKV(P) p.490; ltfKV(V) p.213.

Looking at the nacure of the objections raised by the Buddhist metaphysiciaru

against the notion of "emptiness," Nigarjuna is insisring chat they do not
undersund "the purpose of emptiness, " (iunyat'iiyam. prayojanaftZ). Does rhis
mean chat "emptiness'' has a functional or pragmatic vaJue, rather than beiog

EXAMINATION OF THE NOBIJi TRllrHS

331

an absolute concept? Does it need to be substantiated by "the empty," rather

than merely substantiating "the empty." Which is more important, "the empty'' or "emptiness," or arc they of equal importance? if "the empty" is considered to be important, "em ptiness" would be a merc idea with no grounding
in experience. If the ''emptincss'' is looked upon as being important, the empty
would ruro out to be a fabrication. If both arc considered to be equaUy important, there could be circularity in re~oning based upon them. However, if the
empty is considered to be the basis of experiencc, and emptiness is derived
frorn chat cxperience, this Jatter could be provisional and remain to be corrected in terms of future experience. In such a case, the circularity involved in
arguing about tbem could be easily eiminated. ln the next three verses, Nigarjuna proceeds to provide very specific answers to most of these questions.

8. Dve satye samuparitya buddliiinfi'T{Z dharma deana,

loka-.sa'f?Jvrli-satya'f?J ca.stttya?'p ca paramarthatal;.

The teaching of the doctrine by the Buddhas is based upon two truths:
tnith relating to woddly convention and truth in terms of ultimate fruit .
MKV(P) p.492; MKV(fl) lp.214.

This has turned out to be one of two most discussed verses in Nagacjuna's
Kiin'/ca. Modern disquisitions on the conception of two truths could perhaps fili
severa! substantiaJ volumes. Instead of plodding over trodden ground, the preseot discussion will be confined to a comparison of the conception of two tnlths
in early Budchism and the metaphysical vetsion presented by the interpreters
of the Abhidharma, with a view to clacifying Nagarjuna's position.
Nagarjuna's version will be ex:amined in the light of the problems posed at the
beginning of the pi:esent chapter, as wel as his refcrence to woddly conventions
in the previous chapters.
In the Su1ta-nipat41 the Buddha condemned any arcempt to uphold a view
(di!!ht) as the u!timate (paramaT?J)(Sn 796ff.). Refusing to recognize any
knowledge of "things as they really .a.re," and ma.king a more sober claim to
knowJedge of "things as they have come to be'; (yathabhuta), the Buddha was
reluctant to accept any notion of paramattha as "ultimate reality ." Instead, he
claimed to know "the dependently ariseo" (pa(iccasamuppanna) and, on that
basis . formulaced the cooceptioo of ''dependent arising" (pa{iccasamt1ppada).
In such a context, it was more n1eaningfu for him to speak of attha (Sk. artha)

332

THE PHJLOSOPHY OF THE MIDDLE WAY

and paramattha (Sk. paramartha) in the sense of fruit, consequence, or result

( = phala, as in s'iimafJa-phala, "the fruit of reduseship"). Tbus, a good action is one that is fruitful (attha-sa1?,Zhita) and a bad, unfruitfu (an-attha-

sa?phita).

The Buddha also recognized that views about good and bad, puricy and impurity, are in most cases relative convencions (Sn 878-894). According to the
same text, 'good (kusala) and bad (akusala) are conventions that vary depending
upon contexts (sammutiyo puthujja, Sn 897). These are che ways of the world
(loka) , often charactecized as sam1nut1~ 11ohara (D 3.232) or panfJatti (S
4.39-40). However, there is no indication that, sioce these are relative conventions, the Buddha advocaced the transcendence of both good and bad.
Having defined the good as the ftuitfu, the Buddha characterized the
ultimate good :as the ultimately fru i tfu!. The term paramattha was thus used to
refer to the "ultimate fruit'' rathet than ''u.ltimace realicy." Sammuti and
paramattha are therefore oot two contrasting truths. Theformer is a mixed bag ,
while the latter represents the ultim.ate ideal. As a result of anaining nibbiina.
which is the complete eradication of lust (raga), hatred (dosa), and confusion
(moha), the Buddha found himself enjoying an ultlmate fruit, withour hurting
hirnself or others (na attantapo na parantapo, D 3.232). Parama.ttha thus
becomes the mora.I ideal as reflected in the Buddha's own att.ainment df
freedom and happiness.
The theory of dependent arising (pra!f.ccasamuppada) expla.ining the
phenomena that arc dependently arisen (paficcasamuppanna) thus accommodates the four truths as well as the two truths. All ttuths being pragmatic,
there is herc no place for an "absolte or ultimate reality." The Buddha's
epistemological scandpoint does not allow for such speculations.
This, howevec, was not the case with the metaphysicians. l t is significant to
noce tbat when the Buddhist metaphysidans were faced with the problem of
reconciling the four truths with their conception ,o f substance (s11abha11a), they
were compelled to fall back on the conception of two truths (Akb p.33).
Howcver, their interpretation of the, two truths is totaHy different from the
Buddha's and , in fact, seems to be contrary to it. Here again, Vasubandhu is
our source.
As a Sauttantika metaphysician, Vasubandhu refers to the two truths (satya)
and surreptitiously moves on to a discussion of existents (sat): sa?fZvrtt~sat and
paramiirthasat (A,b p.334). He tbeo provides two examples to illustrate the
. nature of saf!lvrti-sat. The fust is a 'pot' (ghat) . There canoot be knowedge of
a pot when it is broken ioto pieces. The second is water (ambu). Water is not
knowa independent of the knowledge of "form" (riJpa). Contrasted with these
two is knowedge of the paramartha-sat. 'Materia! form' (ritpa) is an example of
such existeoce. Io this case, wheo an object is broken down inro its ucimacey
irreducible elemenrs, namely, ;uoms (paramii1'Ju) , there is knowcdge of the
"unique narure" (s11abhava = s11a/ak!a1Ja?) of such acoms. For the metaphysi-

ExAMJNAT!ON OF IHE N OBIE TRUTHS

333

cian, that unique nature, whether it is called svabha11a qt s11alakfa'la, has aways
remained an epistemological enigma .
This interpretation of the two truths is totaly abseot either in the eary
discourses or in the canonical Abhidharma. In the so-called Theravada ttadition, it appears for the first time in the ooo-canonicaJ tcxts (Mi/n 160), coouibutiog, as it did in the case of the Sarviiscivada and Sautraocika schools, to
insoluble problems such as the duality of mind and matter. Vasubandhu's
statement of the two uuths- .raf(lvrti and paramiirtha-in the above contcxt,
where he was r~cordiog the interpretation of the Sauuiintika view, remains in
sharp contrast to the spirit of the Buddha's teachlng on "noo-substantialism ."
This shouJd certainJy provide an interestiog background to the analysis of
Nigarjuna's owo version of the two cruths.

9. Ye 'nayor na vijananti 11ibhagaf!I- .ratyayad11ayof?,

te tatt11af(l na vijananti gambhiraf(l buddha-lisane.

Those who do not un'derstand the distinction becween these two truths do
not understand the profound troth embodied in the Buddha's mcssagc.
MKV(P) p.494 : MKV(V) p.215 .

Niigarjuna is, of course, criticizing his opponents for not understandiog the
Buddha's rnessage. If so, Niigarjuna's owo explanacion of tbese two trud~
should not come anywhere close to the one discussed by Vasubandhu, according to which the .raf(l11(ti reprcsents a merc designac:ion, not an ultimate reality
or paramiirtha. It is an ootologkal speculatioo whkh is not supported by the
Buddha's own conception of dependent arising.

10.

Vyavah'llram anaritya paramartho na defyate,

parav;.rrtham anagamya nirvaf!tlf!Z tiiidhigamyate.

Without relying upon convention, the ultimatc fruit is not taught.

Without uoderstanding the: ultimate fruit, frecdom is not attained.
MKV(P) p.494; MKV(V) p.216.

THE

PHILOSOPHY OFTHll MIDDLli

WAY

Unless one is committed to upholding a theory of linguistic or concepcual

transcendence of ultimate truth or reality, there seems to be no reason to
restr:kt the use of the terms saift'11rti, vyavaharai or prajapti to refer primarily
to linguiscic convention. These terms cou1d and did stand for any kind of conveotio, whether it be llnguistic, social, political, mora! , or religious. As menti.oned earlicr (sec comments on XXIV.8), all such conventions are pragmacic
a.ud contextual. As such, trutb rclati.ng to them would also be pragmatic and
C00textual.
The fact that in the two p.tevious instances when Nagarjuna used the term
11ya11ah'{ira (XVll.24; _X XIV.6), he used it in the context of a discussion of
morality, is of utmost significance. It is in this same conteitt char the Buddha
used the ceon vohara (D 3.232 an(lriya-vohara == muiavada, etc., and ariyavohara = musivlidii verama1fi, etc.) as well as the term sammuti ( = sa'f?1-vrtt~
Sn 897, 911). 'fhese, tberefore ; arc the mora convencions of the world (loka,
laukilut, sec XXIV .6) that are pragmatic, yet contextual. Howevet, an idea
mora! trutb caonot be suictly confined to a particular context. It needs to be
mote universa and compteheosivc. This, indeed, was the Kancian problem. If
Nagarjuoa was following the Buddha's solution to this problem (see comments
on XXIV. 7), he couJd not have ignored the pragmatic component in the
universal mora principle formulated by the Buddha. A mora law that is incapable of accommodating any exceptions can be uttcily useless and even
hatrnful. As William James once remarked, "There is aways a pinch between
the idea and the actual which can only be got through by eaving part of the
idea bchind" ("Mora Philosophies and Mora Philosophers," in Essays in
Prag1natiJm, ed. Albury Castell, New York: Hafner, 1948 . p . 78). This is
because the idea is an abstraction out of the coocrece and, thetefore, needs to
be modified in the light of new concrete siruations.
Nagarjuna's statetnent "Without relying upon the conventional, the
ultimate.fruit is not expressed" ( vyavahara1n aniin"tya paramarlbo na deJyate) ,
expiains ooly a one-way relation, not a one-one relationship. He is not saying:
"Without relying upon the ultimate fruit, the conventional is not expressed ."
Thus, he was rejecting a deontological mora principle cha provides an absolute source of all mora ideas with no concessions made for individual or con.
.
crete situations.
The second line emphasizes the need to have some undetstanding of that
mora! principle before one could think of attaining freedom (nir1iii11a) .
However, it is possible to argue that this mora principle issue:i out of the Buddha's attainment of nitvana, aod that without attai~g nirvar,a one wiU noc be
in a position to understaod wbat chat principle means . Neither the Buddha nor
Nagarjuna would advocate such a position. lf ooe wcre to first atta freedom
and then look for a mora ptinciple to account for ie, one c.ould sometim~ end
up in wayward fand es, utop1as, aod ballucina:tions. for this reasoo , . ~o

335

EXAMINA llON OF nrE NOBLE TRUTHS

underscanding of Jaf!Wrti (''worldly fruit, laukika artha) and, depending upon

that, gaining some knowedge ofparamartha ("utimate fruit ," lokuttara artha)
could serve as a guide for the attainment of freedom (niniiit1a). In the
Thefigatha we come across an iostance where a disciple claimed rhat he perceived che fauldess dhamma (addasat(' vira}a-f!l dhammat(') and the path co
freedom , while she was still !eading a household life (agarasmirp vasanfi 'ha11J).
Subsequently, she left the household life and attained freedom from all
dcficmenc.~ (Thig 97 ff.) . It is true that the Buddha attained enlighcenment
and freedom by sheer accident. This is why he was reluccant co tecognize any
teachcr (see M 1.171). Yet, before preaching about ie.to co the world , he spent
much time refleccing upon it, as a scientist continues to verify a discovery he has
made befoce making ie public. Nagarjuna probably assumed chat peopl.e could
have the beoefit ofthe Buddha's expcrience and not waste time experimenting
with practices that ~he Buddha hirnself fouod to be fruitess (an-artha).

11.

Vinaayati

durdu(ii 1Unyalii manda'fTJedhasat(',

sarpo yatha durgrmto vidya va du;prasadhita.

A wrongly perccived emptiness ruins a person of meager intelligence. lt is

lilcc a snakc th:at is wrongly graspcd or knowledge that is wrogly
culrivared.
MA.'V(P) p .495 ;

.!KV(V)

p .21 6.

Having explained the two truths, and establishing an important relationship

between the conveotional and the idea , the particular and the universal.
Nagarjuna returns to the conception of "emptiness" (Jiinyata) that gave rise to
the met.aphysicians' problems. The abstracr-conception of emptincss (JUnyata,
reprcsenting an abstract noun}, derived from tbe experience of "che cmpty"
(unya), could be as destructive andfatal as a wrongly grasped snake. How the
emphasis oo the idea and the universal to the utcer neglect of the panicular ot
the situational has wrought hav~c can be known tl)r9~gh a carcful and unbiased study of human bistory.

12.

Ata.f ca pratyudii11rtta'!' cittarp deayiluf!' munel/,

dharmatr1 mat11asytr dharmasya mandair duravagahalii-f!l.

336

THE PHTLOSOPHY Of THE MIDDLE WA Y

Thus, the Sage's (the Buddha's) thought recoiled from teachlng the doctne having reflected upon the difficulty of undemanding the doctrine
by peope of meager intelligence.
MKV(P) p.498: MKV(V) p.217.

This is a reference to the inddent related in the "Oiscourse on the Noble

Qu~st" (AriyapariyeJana-sutta, M 1.167-168) where the Buddba explains the
nature of his enlightenment as weU as the events immedia.r dy preceding and
succeeding his attainment of enlighteomeor. Niigarjuna could not have been
unawarc of the oature of the Buddha's eolighreoment as described in this
discourse and also the reasons for the Buddha's reluctance to preach the doc
trine. The Buddha's statemcnt reads thus:
lt occurred to me monks : "This dhamma won by me is deep, difficult to see, difficult to understaod, tranquil, excellent, beyond a
ptiori reasoning, subtle, intelligible to the learned. But these
human bdngs arc delighting in obsessions (alaya)', delighted by
obsessions, rejoicing in obsessions. So tbat for the human beings
who are rejoicing in obsessions, delighted by obsessions and rejoicing in obsessions, this were a mattec difficult to see, that is to say,
dependent arising. This too were a matter difficult to sec, that
is to say, the appeasement of all dispositioos. the renunciation of all
auachmeot, the waning of craving , the absence of lust, cessation,
freedom. But if I wete to tcach the dhamma and others were not to
understand me, that would be a weariness to me. that would be a
vexation to me.' (M 1.167)

This passage should dispel any doubts regarding the naru:re ofthe Buddha's
reasons for his ini~ial reluctance co preach. le was noc because of any intellectual
incapacity on the parr of the human beings, but mostly because of the emot ional difficulties chey would have in brcaking away frorn the accepted theories
in which they have fouod safety and comfott, a sort of 'mooring' (alaya),
especiaHy in accepting a less absolutistic and fundament?Jly non-substantialist
(anatta) philosophy and a way of life . As iodicated above, the phrase used by
the Buddha to refer to the attitude of the human beings is alaya-ratii
("delighcing in alaya," where alaya is derived from 4 + V li implying some
sort of mooring or obsession). Alaya represents an obsession for not only the
pleasure~ of sense, but aJso ideas, dogmas, theories, etc.

EXAMINATION Of THE N OBI.E Th.UTHS

13.

337

SUnyafiiyam adhilaya??t yaf!t puna'1 kurute bha11an,

dofa-prasago nasmakaf!t sa fiinye nopapadyate.

Furthermore, if you wece generate any obsession with regard to emptiness, the accompanying error is not ours. Thac [obsession] is not appropriate in the context of the empty.
MKV(P) pA99; MKV(V) p.21 7.

A.dhilaya has been rendered as "repeatedly refuced" (Inada , p.147). However,

the term is more closely related co alaya (see commentary onXXIV. 12) utilized
by the Buddha to refer to emotiooal attachment or obsession for views. Nagarjuna seems to be saying .chat the metaphysicians are generally fascinaced by
clear-cut and well-formed theories, abscract concepts, which they rcvere as
divine, while the concrete percepts are looked \.\pon as belonging to the brute
(see William James, Some Problems of Philosophy, p .34). This paves che way
fot an extremely important statement by Nagarjuna regarding unyalii a.d

funya.

14.

Sarvaf!t ca yujyate tasya Unyata yasya yujyate,

sarvaf!t na yujyate tasya fUnya??t yasya na yujyate.

Everything is pertinent for whom empriness is proper: Everytbing is not pertinent for whorn the empty js not proper.
MKV(P) p.500; MKV(fl/ p.218.

This terse stacement of Nagarjuna, represencing an extremcy valu_?.b~e assessment of the two concepcs of "emp tiness" (unyata) and "the empty" (funyaf!t),
can be comprehended only in t)le background of his previous statement (X:XIV.13). Ie is a cricicism of the absoucistic aod substantiaJist enterprise wherein
abstract concepts Jike "empciness" reccive a divine status thercby becorning
compatibe wich anything and everything in human experience, like the
Platonie "forms". whcrcas any emphasis on the concrete leaves everything .

338

THE PHILOSOPH Y OF THE MIDDLE WAY

banging in the air. Plato's "forms" . Spinoza's "substance", Vedanca "Brahman"

and Lao-czu's "Tao" - all these are fascinating abstract concepts that fit in with
everything. Even if one does not have to movc on to the ocher extrcme, as did
Hume with his "impressions". and the Sautrantikas with tbeir "atoms"
(paramaf!u). any recognition givcn ro the concrcte, to plwality , to the flux
tends to disrupt the sense of security one enjoys in a world of abscractions. The
Sarvascivadins raised objections against ''the cmpty" (fiinya)(XXIV. l-6) not
becaust there is an actual conflicc between the notion of "the empty" and the
theory of four truths, but because the cooception of "the empry" con.flicted
with their ootion of subsrance (s11abhava) which rhey were holding on to with
great enthusiasm. The faule, as Nagarjuna points out in the following verse, is
theirs.

15.

Sa

tva~

dofin atmaniyan 11Jmasu paripatayan,

avam evabhim"1ah
. . sann aam evasi 11ismrtah.
. .

You, acuibuting your own errors to us, ace like one who has mounted his horse
and confused about it.
MKV(P) p.502; A1KV(V) p. 218.

As mentioned previously (sec commenrs on XXIV.14), the problem faced by

the Buddhist metaphysicians was in regard to reconciling "the empty" (funja)
with che four noble truths. This was because they were explaining the four
truths in relation to the oorion of substance (svabhava) and the notion of
subsrance d early cooflicted with the concept of "the empty." lf Nagarjuna had
mercly reJied upon the abstra.cc concept of "empciness'1' (fUnyafii) , ignoring
"the cmpty" (fiinja) , the Buddhist metaphysicians w<uh: not have run into
difficuti es, for "cmpciness," being the abstract concept could easily be reconciled with the notioo of subscance. As such, the p.coblem was created by the
metaphysicians when they emphasized the non-empirical "substance'' to the
ncglect of the empirical phenomena, as is dearly evident from Nagarjuna's
statement that follows. Nagarjuna was, therefore, reminding them thac they
wece riding their owo horse wichout reaJizing it .

16.

Svabha11'(itl yadi bhav'iina~ sad-bhavam anupayasi,

ahtu-pratyay'iin bhiva~ 1vam evaf?Z sati paiyasi.

EXAM!NATION OF THE N OBLE TRUTHs

339

If you pccceive che exisrence of the existents in terms of self-nature, then

you will also perceive these existents as non-causa! conditions.
MKV(P) p.502 : MKV(V} p.2 19.

For Nagarjuna, any existcnt is a causally conditioned existcnt. Such an existent

is clearly imcompatible with a self-nature or substance, che at ter being permancnt and etefual by definition. This would mean that Nagarjuna's conception
of "cmptioess" (funyata) does not partake of any such characreriscics. Eveo
though "empriness" and "sclf -nature" are borh abstractions from conctete experiences, Nagarjuna woud continue to maintaio the pcimacy of such concrete
expericnces, which are "che empty," and insist that "emptiness" is dependent
upon "the empty." while the meraphysicians would consider sclf-nature to be
independent, thereby divorcing it from the concrete.

17.

KJirya'f!l ca karar,at?J C11111a kanara?'!' kara'f!tlf?J krifit?J,

utp1ida'f!l ca nzrodhat?J ca phalaf!J ca pratilfiidhase;

You will also contradict [the norions of] effccc, cause, agent, pctfoanance
of action, activity, arising, ceasing, as weU as fruit.
MKV(P) p.503: MKV(V} p.21 9,

How the recogrunon of self -narure (svabhava) contra4icts all forms of empiricaJ knowledge, knowledge recognized by the Buddha relating to dependent
arising, mora behavior (karma), and responsi biity (karma-phala), has already
been explained in detai l in the numerous chapters t hat precede. Herc Nagarjuna is simply summarizing all his conclusions.

18.

Yal; parfi1yasamu1p{idqf? f unyat1it?J laf!' praca/qmahe,

s1i prajnaptir upadiiya pratipat sai1Ja madhyami.

We state tbat whate"er is dependent arising, tfult is emptiness. Trun is

dependent upon convcntion. That itsclf is the middle path.
MKV(P) p.503; MKV(V) p.219.

340

T"HE PH llOSOPHY OF TiE MIDDLE

wAV

Thece see.ms to be no other staccmcnt of Nagacjuna more conuovecsiaJ than

this one. An cntirc school of Chinese Buddhism emerged as a rcsulc of the intcrprctatioo (or misintcrpretation) of this verse, [sec Hajime Nakamura, "Tbc
Middle Way and the Empcincss View," Journal of Buddhist Philosophy, ed .
Richard S. Y. Chi . (Indiana University, Bloomingtoo , Indiana,) l
( 1984):81-1 ll ] .
In the first line, Nagarjuna is prescnting an equation: Dependent arising
(pralityasamutpada) is "emptincss" (fUnyata). Jnada's rendering of this line is
an improvement on Nakamura's , since che lactcr assumes that herc there is a
reference to the evcnts (op. cit. , p .81), racher tban the principle in terms of
whicb the events arc explained. As emphasizcd earlier, "dependent arising"
and "emptiness" arc abstract concepts derived from concretc empirical evcnts,
"the dcpendently arisen" G"ralitya.ramutpanna) and "the cmpty" (1iinya)
respcctivcly . Unless this relationship between the abstract and the concrete is
clearly obscrved , the incerpretacion of the second line of the verse will remain
for evcr obscure , as scems to have happened since Nagarjuna composed this
ueat1se.
The second line refers to the middle patb (madhyama pratipat). The question is: In wbat way cao dependent arising and empciness, which arc
synonymous, represent a middle path? The answer to this question is in the
statement , sa prajnaptir upadiiya.
If this phrase wece to be transated as, "It is a provisional oame" (Inada,
p . 148) or as "That is a temporary dcsignation" (Nakamura, p .81 ), which arc
standard uanslacions offered by most schoars, chen in che explanacioo of
dependent arising and emptincss one will be committed either to an extreme
form of oominalism or a sirniliarly extremisc absolutism. Io such a case, dependent arising or emptincss would either be a mere descripcion wich no basis ln
cognitive expcricnce or it would be an cxperience that is ineffabk. The former
alternacivc would conflict wich everything Nagarjuna was t rying to esrablish,
narody, the reality of atising and ceasing, of human behavior, etc.
(XXIV16-l 7). The latcer alternative would rendcr any philosophic.al enterprise
meaninglcss and would uodermine Nagarjuna's asscrtioo ac XXIV. 10.
Taking Jaf?Z11fh: 11ya11ah"iira, and prajiiapti as synonyms, as was intended by
the Buddha bimself (D 3.202), and considering the contexts in which Naga rjuna was using the terms saf?Zvrti and 11ya11ahara (XVlI.24 ; XVTV.6,8) - namely, in relation to discussions of worldly conventions, the most importaot among
them being good and bad, dhartnal ariharma (and this sense is caprured by Inada's own translatioo of saf?Zvrti and 11yavahara, see p. 146)- it would be more
appropriate to chacacterize both "dependent arising" and "empciness" as being
the universal truths rather chan ahso/uJe truths. This would mean that both

EXAMJNA TION OF THE N OBLE TutrrHS

341

saf!Jvrti

(and, therefore, 11ya11ahara and prajflaptt) and paramartha arc

"thought conscructions" founded on experience. As such, chey are not absolutely rea.I or absolucely unreaJ. This, then , wou.ld be the midd.le position
(v1adhyama pratipat). Tb is middle path could be adopced in understanding all
forms of experience, whether they be Lioguistic, social , political, rnora1 , or
religious. Whie a1J convencions, whether chey be situational (saf!Zvrti) o r idea
(para11Jiirtha), are expained in terms of dependen t arising, the conceptioo of
"the empty'' (sunya) eliminates che absolutistic sting at the level of the situational , and "empciness" does so ac che level of the uoiversa. (See comments on
XXlV.19 that foJows.)

19.

Apraliya-samutpanno dharmaf? kafcin na vidyate,

yasmat /(ZJmad aiunyo hi dharmal/-kafcin na vidyate.

A thing chat is not dependently arisen is not evident. For thac reason, a
thing that is non-empty is indeed not evident.
MKV(P) p .505 ; MKV(V) p.220.

After clarifying the oaturc of che universal cooventions, Nagarjuoa recurns to

the situational. for it was this lacter chat caused problems for the metaphysicians. The metaphysicians realized that "the dependently arisen"
(pra/ity(ZJamutpanna), so long as it is considered to be "the empty'' (fiinya),
conflicted wich their nocion of subscance (svabhava). Therefore, they were looking for "the dependencly arisen" that is not empty of substance.
However, for Nag:trjuna, substance (svabhava) is not dependently arisen
(XV.2). A thing that is not depcndently arisen is not eviden r. Thcrefore , a
substance is also noc evident. A substance bcing non-empty, the absence of a
substance woud mean the absence of the non-empty (a-fUnya).
This is a elear and unequivocaJ d enial of substance (svabhava), of the nonempty (a-funya) and of the in-dependent (a-prafftyasamutpanna). and. the
assertion of the non-subscantial (nil/-svabhiiva), the empty (sunya) and the
dcpcndently arisen (praffty~amutpanna). In the hope of being an absolule
non-ahsolutist, if someone were to propose chat both alternatives - svabhava
and nil/svabhava, aiunya and Junya, aprafftyasamulpanna and prafityaramul
panna- are crue (or false), the empiricistN:igarjuna, following che Buddha (Sn

342

THE PHJLOSOPHY OF THE MIDDLE W AY

884) would insist that only one of these aternacivcs is crue (ekaf{l hi saccaT(J).
and not the second (na dutiya'!J). This means chat neither the Buddha nor
Nagarjuna could be characterized as absolutiscs.
However, if the two alternatives were to be two metaphysical altcrnatives,
then both the Buddha and Nagarjuna would be compelled to ncgacc them.
Takco out of context, wabhiiva and nzf/11111bhii11a could appear as mccaphysicaJ
alternacives. So could a!unya and filnya. Yct , apralitya.samutpanna and
pralitya.samutpanna, as understood by the Buddha and Nigarjuna, could not
be easily coo-.:ened to metaphysiall alceroatives. The reasoo for this is that the
term pralityasamutpanna, beiog a past parcic~ple, does not refer strictly to any one
tempora! experience, whcther it bclo_ngs to the past, present or future, nor does
it traoscend ternporality altogether. Whil_e it has a present connotacion, it is not
divorccd from the pasc. lt i~ . therefore, a term most appropriately used to
describe the events perceived by the Buddha through his "knowiedge of things
as they have come to be" (yathabhuta-fiii'la) . lt refers to the events experiencep
in the so-caled "spedous present" (see comments on XIX.4).
lt is this empiricist prajnapti that servcs as a corrcctive to any propos3.I which
would turn either pralityasaf!Jutpada or unyata into an absolute truth , ineffable and a-tempora!. lt is, indeed, the philosophical middle path referred co at
XXIV. 18 and the Buddha's discourse to Karyayana.

20,

Yady afunyam idatp. saroam udayo na.Ili na vyayal;,

catu"Jam iiryasalyiiniim abha11as te pra.rajyate.

If all this is non-cmpty, there exists no uprising and ceasing. Thcsc imply
the non-cxistencc of the four noble truths.
MKV(P) pp.505-)06; MKV(V) p.220.

By rescuing the philosophical middle pach from aoy absolutistic or substantialist incerpretation , Nagarjuna has sec the foundation for the explanation of
the Buddha's doctrine of che four noble truths. The facts of arising and ceasing
arc central to the fouc noble uuchs. lf tbere ever has been a permanenc enticy ,
an encity that is noc empty of a subsrance. then dependence as wdl as the four
noble truths would have been falsified . No such entity has been discovered so
far. Hence the four truths have remained valid.

Ex.J\MINATION Ol' THE NOBLE TRUTHS

21 .

Aprafftya-samutpannaf!Z kuto du/khat!' bhavir.Jati,

anityam uktaf!J dul,khaf!J hi :at sviibhiivye na vidyate.

How can there be suffering that is not dependently arisen? Suffedng has,
indeed, bccn described as impermanent. As such, it is not evident in
tcuns of self-nature.
MKV(P) p.506; MKV(V) p.220.

Nigarjuna is herc: returnjng to the interpretation of suffering by the substantialist metaphysicans (sec comments on XXIV . l), who held that suffering exists
in terms of self-nature. He is reminding che metaphysicians that the Buddha's
conception of suffering is founded not ony on the idea of dependent arising
but also the notion of impermanence. Io face, the perversion regarding the notion of impermanence is aJso the cause of che perversion regarding the concepcion of suffering (sec comments on XXIII. l).

22.

Svabhiivalo vidyamanaf!' Jur!J puna!? samude1yate,

tas1'!iiit samudayo n'iisti funyafiif!''P,r,atibadhatal/.

How can that which is evident ih ~euns of self-nature risc again?

Therefore, for one who contradicts cmptiness, there exists no [conception
of] arising.

M.KV(P) p .506; M.KV(V) p .221.

Throughout the treatise, Nigarjuna was refusing to recognize that the conception of self-nature or substance leaves any room for the recognition of arising
and ccasing. This was contcary to the attempr on the part of the metaphysicians. Nigarjuna is herc insisring that the conception of arising (samudaya)
makes no sense at all when applied to self-oature.
Self-nature is not something that comes and goes. lt is oot 20 occutreoce. lt
is thece for ever. As such. it contradicts the theory of emptiness which is intendcd to explain the occuuencc of cvents. Empciness, as ment:iooed earlier
(XXIV. 18), is synonymous with "dependent arising."

344

23.

THE PHJ (QSOPHY OF THE MI DDll

WAY

Na nirodhaf? 1vabhii11l!11a Jato duf?khasya vidyate,

1vabhavaparya11a1thaniin nirodhaf!t pratt'badhiise.

The cessation of suffering that exists in terms of self-nacure is not evidenr.

You conttadict ccssation by adhering to a notion of self-nature,
MKV(P) p.507; MKV(V) p . 221.

Just as the notion of self-nature conuadiccs the idea of arising, it also renders
che conception of cessacion meaningless. Note the cmphasis on the present participle , santa (left untranslated by lnada, p. 149), used in relation duf?kha ,
implying "suffering that exists" in terms of self-nacure. The notion of exiscence
defined in this manner has already been criticized by Nagarjuna (1.6) . In effect
Nagarjuna is saying thac the nocioo of exiscence (sat) may be compatible with
che conception of self-oature, but ie is not appropriate io the context of "chat
which has come to be" (bhuta) .

to

24.

Svabhavye sati m'iirgasya bhavanii nopapadyate,

ath'iisau bh'iivyate m1irgaf? 1viibhavya1?J te na vidyate.

When self-narure exists, the cuJtivation of the path is not appropriace.

And i.f che path wece to be cultivated, tben no self-nature associated with
it [i.e., the path] woud be evident.
M.KV(P) p. 507; MKV(V) p.221.

The priociple of sclf-nature (svii.bhiivya, noce the use of the absuact noun instead of svabhiiva) explaios the way in which self-nature funcri~ns . !hus, if
somechings are assumed co have self-nature , whac is assumed is that chings happen on cheir own (svo bhavo) without the support of anything else. la tbis
sense, chey arc independent. If rhe pach (marga) Jeading to che ccssacion of suffcring (duf?kha) were to be explained in cerms of self-narure, ie means chac che
way will work out on its own , wichouc any effort on the part of one who is expecced to cultivate it. If one is expected ro c utivate the pach, it means chat one
has to make an effort . Ie does not happen auromatically .

EXAMINATION OF 'rHE NOBIE TRUTI-S

25.

345

Yadii dul/ka'fP samudayo nzrodha ca na 11idyate1

miirgo duf?kha-ntrodhatvat k.atamal/ prapayz'!yati.

When suffering as well as its arising and ceasing are not evident, through
the cessation of suffering where will the path lead to?
MKV(P) p.508i MKV(V) p.221.

Insisting upon the recognition of self-nature, the metaphysicians were compeled to deny the fact of suffering as well as its arising and ceasing. lt would,
therefore, be meaningless to speak of a path leading to the cessation of suffering. Substancialist speculation leaves only two aternatives: either there is cessatio of suffering or there is no cessation of suffering. If the fact is that there is .
cessacion of suffedng (dul;kha-nirod/;atva). thtre is no room for a path. It happeos in any case. If the fact is that there is no cessation of suffering, then it is
meaningless to think of a path. FaraJism being the consequence of both standpoints, the notion of a path Jeading to any goal is rendered uttetly meaningless.

26.

Svabhavenapatijnanaf!' yadi tasya puna!/ kathaf!',

parijnanaf!' nanu kila svabliiivaf? samava.rthitalf.

lf non-undetstanding is due to self-nature, how can one come to possess

understanding subsequendy. Is it not the case that self-nature is fu:ed?
MKV(P) p .508; MKV(V) p.222.

lnada's rendering of nanu kzla svabhavaf? samavasthital/ as "self.nature, in

deed, oever re.mains fixed," seems to be based upon a rnisuoderstaoding of
the impon of the particie nanu which simiJly means: "is it not the case that,"
rather chan a simplc negacion. If self-oarure is something that "never re
ma.ins ftxed," what we ate presented here is the so-caled Spinozao solution
to the problem of substance. Nagarjuna was not willing to accept such an interpretation of self-nature, for if self-narure were understood as changing,
the whole purpose of formulating the notion of self-natute by the substan-

346

THE PHILOSOPHY OF THE MIDDLE W AY

tialists would be defeated. Therefore, he was simply asking the question; "Is
it not the case that self-narure is fixed?" He is, io fact, insiscing that if ie ~~
not fixcd, ie is not a self-nature. A change of subst.aoce was, io the eyes of
Nagarjuna, a self-conuadiction.
Hence he argues: If there is any inabili:cy to understand suffering because
such ioability is inherent, is through self-pacure, then there can never be ics
underscand:ing. Something chat is inherently uoknowable can oever be
subsequently known. Thi is the most telliog criticism of self-nacure

(wabha1111) as it is employed in the explanation of the path to freedom

(niniiif!a).

27.

Prahaa-iiikfatk.ara'!e bliii1111na caivam eva te,

ptmjfiivan na yujyante catv'iiry api phaliini ca.

As in the case of understanding, this [i.e. , the explanation in terms of

self-oature] is not proper in relation to the activities of relinquishing,
realizing as well as cultivating. And so would the four fruits be [improper).
MKV(P) p.509; MKV(V) p.222

An uoderscanding (parijfii) of the nature of suffering (duf?kha) is a necessary

prerequisite for its relinquishing , the reaJizaciotJ of freedom , and che cultivation of the path leading to freedom . In the previous verse, Nagarjuna ex
plained how a belief in self-nature would create difficulties in explaining
understanding (parijmi) or its absence (apanjfia).
The same di.fficulries are associated with the explanation of the relinquishing
(praha1'!t1) of suffering, the realization (s'iikfatkara11a) of frc~dom and che
cultivation (bhivanii) of the path. To highlight these difficulties, Nagarjuna
concentratcs on the fruits or consequeoces (phala).

28.

Svabhaveniidhigata'l'(t yaJ phala'l'(t tal puna!? kathar(t,

akya1'(t samadhigantu1'(t sy'iit svabhava'l'(t parigrh11ataf?.

How could it be possible for a person, who upholds a theory of self-

347

Ex.MINATlON OF THE N OBT.Jl TRUTHS

nature, to realize a f ruit that bas already becn realized through self-
nature?

MKV(P)

p . ~10;

MKV(VJ p.222 .

A person betieving in self-oatwe aJso admits that a fruit is obtained by that

means. lf a fruit has not been achieved through sdf-nature it can never be
achieved in any ocher way. lt is an either/or situation. Accept it as occurring on
its own or it will never occur. This is a substantiatist view of frcedom (nirvaf!a ,

which will be taken up for detailed treatment by Nigarjuna later on). This, indeed, is a fatalistic or a deterministic explanation of causality. Either the effect
exists in the cause or it does not (see discussion of artha [ = phala) at 1.6).

29 . Phaliibhave phalaslha no na stfntipralipannaktil?,

saf?i.gho nas# na cet santi te '1tau puntfa-pudgaliil?.

In tbe absence of the fnilts, tbere arc neither tbose who have attained the
fruits nor those who have reached the way [to sucb attainmeot]. If the
eight types of indivjduals do not exists, there will be no congregation.

30 .

Abh'?ivac carya-saty'iinaf!1- sadt/harmo 'pi na vidyate,

dharme casati .sa1{'ghe ca kathl#(J- buddho bhavi$yati.

From the non-existeoce of the noble truths, the truc doctrine would also
not be evident. In tbe absence of tbe doctrine and the congregation, how
can there be an enlightened one?
MKV(P) p .) 10; MKV(V) p.222.

Having shown that the explanation of the attainment of the frui~ (phala) of
life is rendered i mpossible or meaningless by the adhetence to the ootioo of
self-nature, Nagarjuna rurns the table on the metaphysicins, showing them
how the denial of emptiness (iiinyatii), rather thn itS assertion, leads to a
denjaJ of all that they wece trying to explain.

348

31.

THE PHILOSOPHY Ol' i HE MIDDLE

W AY

Aprafilfiipi bodh~ ca la11a buddhaf? prasafyate,

aprafi1yapi buddhatiJ ca ta11a bodhif? pra.JajytTte.

Your [conceprion of the] enlightened one implies an independent

enligbtenment. Also, your [conception of] enlightenment implies an independent 'enlighcened one.
MKV(P) p.510; MKV(V) p .223.

Explained in terms of self-narure, the enli~htened one would be so irrcspeccive

of conditions (apratyaya). Similarly, enl]ghrenment would be achieved irrespective of whether it is a person who makes an effort or not. This substantialist cxplanation of enlightenment and the eolightened one, in terms chat arc
strictly naturalistic or deterministic, wou ld rendet the activities of the one seeking enlighteoment and freedom utterly meaningless.

32.

Yaf cabuddaf? svabhavena sa bodhaya gha{ann api,

na bodhi.sat111a-caryayfi1!1 bodhi1!1 te 'dhigamifytTti.

Wh~oevet

is by sclf-nature unenlightened, even though he were to con-

tend with enlightcnment, woold not attain enlightenment thtough a

career off a bodhlsattva.
MKV{P)

p . 511~

MKV(V) p.223.

This is the one and only time Nagarjuna, the so-callcd pau on of Mahayaoa,
rcfcrs to the way (caryQ) of a bodhi.sall11a. Yet, the kind of criticism he is prcsenting herc offers no consolacion to those who accept cercain docrrines of
popular Mahayana. Any substaotiaJist theory, according to Nagarjuna's view,
militates against the ca.(eer of a bodhisattva.
To uphold che view chat a person is by narure u ncnlightencd is taotamount
to saying chat he can never attain the fruit of eolighteruncnt and frecdom by
excrting himself. Sir:Qiarly, to assert che view chat a person is by narure
enlighccncd, is a potential buddha, or possesses a bodht~citta, makes the attainmcot of enightenm.ent meaningess (XXIV.28) or impossible.

EXAMINATION OF THE N OBLE TRUTHS

This, indeed. is reminiscient of the long drawn controvcrsy between the Sar
vastivadins and the Sautrantikas regarding tbe oarure as wcU as the functioning
of good (kusa/a = dharma) or bad (akusala = adharma) thoughts (citta). The
metaphysical problems generated by an extremist analysis in rclation to
phenomena in genera have been explained in Chapters I and Il . The selfsame
metaphysical problems appeared in the explanatioo of moraJ phenomeoa. The
Sarvastivada-Sautrantika controversy relating to the "seed,s that arc of good
nature" (kusa/a-dharma-bija) reflects their iovolvement in sucb mt"taphysical
4

1SSUCS.

The atomiscic (and, therefore, noo-empirical) analysis of thought (citta) led

the metaphysicians to assume a sharp distinction not ooly becween good
thoughts (kusala-cilta) and bad cho_ughts (t1kusala~citta), but also betw~eo iodividual moments of thought in cach series. whether they be of good thoughts
or bad thoughcs. As such, severa] questions carne to be raised: How does one
good .thought moment givc risc to another similar thought mment? How can
two different thought moments, one which is good and the ocher which is bad,
co-exist in one moment, especially when they are distinct as light and darkness
(7iloka-Jamas, see Adv p .170)? If they cannot co-exist, how can a bad thought
moment give rise to or be followed by a good thought? While the Sarvascivadins urilized the cooceprions of prapti and aprapli to explain the manner of their occurrence, the Sautrantikas employed the notion of seed (liija) to
elucidate these problems.
The substantialist Sarvastivadins, who adhered to a oocion of self-nature
(svabhava) were compdled to maintain thac cach vari~ry of thought has its
own-nature, Yet, when they were called upon to explain the occurrencc or nonoccurrence of good or bad choughts in an iodividual person's stream of chought
(sa1{Jtana). they asswned that the non-attainment (apriiplt) of one kind of
thought could make room for the accainmeot (pr7ipti) of another kind . The
Sautrantikas, rejecting the noc:ion of self-nature, emphasized the idea of potentiality (Jaktt) exiscing in che form of seed (bija). Thus, when a good thought
moment occurs , the bad tbought moment can remain latent without
manifesc.ing itself. Tbus, we are left with two potencialicies, one manifescing
itself when the other is not. The terms iak11: bija, and 11iisafla werc all
employed to cefer to such pocenciality.
A discioction carne to be admitted not ooly between good (kusa/a) and bad
(akwala), but also among the good seeds themselves. Thus, chere arc defiled
good secds (sairava-kusala-blja) and non-defilcd good seeds (aniifrave-kusala/iija) , the former accou.oting for worldly goodness and the latter leading to
frccdom (11irva11a). These sceds werc considered to be extrcmely subte
(susuk/ma) and remain uocor~pted or undestroyed (na ramudghiila). They
were compared co god (kaflcana). A popular stacemenc runs thus:

350

THE

PHllOSOPHY OF THE MIDDLE W AY

I perceive his exuemely subtle seed of release lik~ a seam of gold

coocealed in a deft of elemeotal rocks. (Mo,~,fa-bzjam ahaf?I hy asya

S1'st1k1mam upa/alqaye, dhtitu-p1i1fi'!a-vi11are niliam va kaflcanam,

Sakv p. 644, see alsoJaini, Ad11 Introduction, p.116.)
Even though this substantialist view is attributed to tie Buddha himself (see
Jaini, loc cit.) on the basis of the Buddha's statement tigarding the "luminous
thought" (pabh(IJsar(,lf?Z citlaf?Z), we have already provided evidence to the contrary (see Inuoduction), where thought is comp;ued with "gold-ore" (iiita-!'Upa)
rather thao with god (suva1111a) . As such, oeither. .cJ.le theory of the "seeds of
celease" (moksa-bija) nor of the "originally pure mind" (prakrti-prabhtisvaractta), which is a predecessor, the Mahayana notion of a bodhi-citta, can be
reconciled with the Buddha's conception of\ non-substanciality (anatta) or
Nagarjuna's view of "emptiness (unyafii) . The present statement of Nagarjuna is a elear rejection of the substantialist standpoiot of the later interpreters,
which represents a recurrent desire to go back to a pn'ttzordial source.

33.

Na ca dharmam adharmaf?Z va kacij jiitu kan/yatz~

kim afunyasya /<.arta11ya7(t svab6aval/ knyate na hi.

No one will, indeed, do good or bad. What could the non-empty do?
foc, self-nature does not perform.
M.KV(P) p.511 ; MKV(v,i p.223.

Inada , once again. renders the terms dharma and adhamta as factoJs and non~
factors of experience, thereby reacling moce ontoogy than axiology inco the
statement of Nagarjuna. substantialist ioterpretation of good and bad a.llows
no room for their cu ltivation or performance. If someone; is good in terms of
self-nature, he does not have to 'perform the good; ir is simply there. SimiJarly
with bad. Self-nature (svabhava) implies absolute existence (sad-bhava), not
occurrcnce (sa'f(tbhava, XXIV.22) .

34.

Vinii dh11rmam 11dhPrmll'f!Z ca pha/af!> pi t411fl vidyate,

dhar~dharma-nimillaf!J "Ca pha"1'f(t lava na vdyale.

EXA.MINATION OF THE N OBLE TRUTHS

351

As for you, the fruit would be evident even without good or bad. Tbis
means that for you a fruit occasioned by good and bad would not be evident.
MKV(P) p.512 ; MKV(V) p.223 .

The consequence of upholding a nocion of self-narure in mora philosophy are

explicated herc. If good and bad were to be explained in terms of self-nature or
substance, then thete would be goodness and badness "in themselves." These
woud always be existent. A fruit, if it were ~o arise at aU. from a good action
will ahvays be good and. as such , chere would be identity of cause and effect
(sec Chapcer I) . If a good action were to lead to a bad consequence, ie wouJd
not matter at all, as ie is assumed to be the case with, for example, "good-wiU."
This ootion of inrriosic good or bad would render the concept of a cause (nimitta) almost meaningless .

35.

Dharmadharma-nimittatrt va yadi le vidyate phala.1?1,

dharmadharma-sa1nutpannam aiunya1{' te ka1ha1{' phalatrt.

If, on the contrary , a fruit occasioned by good or bad is evidcnt to you1

how can you maintain the fruit that has arisen from good or bad to be [at
the samctimeJ non-empty?
J.1.KV(P) p.512; MKV(V) p.223 .

Thus, Nagarjuna insists that rnora discourse and a subsrancialist world-view arc
incompatible. If something is good by nacure, good in terms of self-narure,
good in itself, chen it musc be always good. Ir cannot be otherwise. Such an absolutistic theory will fail to accommodate some individual or panicular siruations tbat arc in conflict with the theory, but which would cectaioly be valid in
certain concexrs. This was a serious defect in the absolutistic cheory. as explain
ed by Nagarjuna in the following verse.

36 .

Sarva-sa1{'vyt111ah'iir'iif?1f ca laukiliiin pratibiidha.re,

yaJ prafttyasamutp'iida-funyaliif!J pratibadhase.

3~2

THE PHlLOSOPHY OF TH E M IDOLE W AY

You will contradict all the worldly conventions when you contradict the
empciness associated with dependent arising.
MKV(P) p .)13 ; A-fKV(V) p .224.

This, ooce again, bighlights the significant reationship betwcen worldly conventions, that arc siruatiooal or contexrual and the prindple of dependent arising which, as explained earlier, is a univesal. The only way che universal can
accommodate the siruational is when the universal is not looked upon as corresponding co an "ultimace reaLiry." The subsrantialists who assumed that there
is an ultima ce reaJiry, rherefore, we re faced wich all the insoluable mecaphysics.
Thus the actual function of emptiness is the eiminarion of this substancialist
sring (see commcnts on XXIV . 18).

37.

Na kartavya'f!J. bhavet ki7{1-cit anarabdha bhavet kriyii,

karakal? sjiid akuniiirtalJ JUnyataf?Z pratibadhataf?.

For one who contradicts emptiness thece would be nothing that ought to
be done; accivity would be uninitiated and an agent would be nonacnng.
MKV(P) p.513; MKV(V) p.224.

The unfortunate consequences of upholding a subsrantialist theory and denying empciness arc listed herc. lncerestingly, all of them percain co human
behavior and moral responsibiliry. This says much abouc che concerns of
Nagarjuna, who by popular acclaim is more a logician primarily concerned with
language and truch and therefore with ucimate reality , rather tban a mora!
philosopher iotecesred in axioogy and, for thar reason, emphasizing the conception of "u timate fruit. " He is supposed to have scorned aoy specuation
about human behavior (karma), wbat human behavior oughr ro be (kartavya),
and who a responsibile human agent (kurvarJa) is.
In brief, a substantialist view leads co a denial of ~he human element fu ncrional in this w.ord, an element t hat is generally described as "disposition"
(saT(lskara). Not only does it negate the world condicioned by human disposirions (i.e., the saT(l.!krta), ie also denies any activiry and creativiry in the natura!
world (sec XXIV.38 that follows).

EXAMINATION OF TH!; NOBLE TRUTH~

38.

353

Afiitam aniruddhartt ca ku{astha11} ca bhavi!yati,

vicitr'iibhir avasthabhz}f svabhave rahita11} }agat.

In a substantialist view, the universe will be unborn, oon-ceased, remainlng immutable and devoid of variegated states.
MKV(P) p.513; MKV(V) p.224 .

The assumption of a substance (svabhava) would depri.ve the natura! world , the
world of dependent arising, of all che variety (vicitra) and freshness and leave it
sterile and unproduccive . This , indeed, is the Buddha's criticism of erernalism
(sassata-diffht) when he maintained t hat according to rhis view the self and the
world are sterile, immovable, and remaining stable like a pillar," (D J.14; S
3. 202, 21 1, etc yatha sassato atta ca loko ca vanjho ku{tiffho esikauhayz't{hito):

39.

Asarnpraptasya ca praptir duf?kha-paryanta-karma ca,

sarva-klea-praha17a11J. ca yady afunya11J. na vidyate.

If the non-empty [is evident] , then reaching up to what has not been
reached , the act of t'crminating suffeng as well as the reunquishing ofall
defilemenrs would not be evident.
MKV(P) p .51) ; MKV(V) p.225.

The substantialist view would oot ony negate the world decermined by h uman
dispositions (sa11}skrta) and che naruraJ world of dependent arising
(prafitysamutpanna), but also the idea world , che world of freedom . N iryana
would remain inexplicabe in the context of a substantialisc world-view. This is
che subject maccer of the next chapcer. Before proceeding to explain freedom,
Nagarjuna has a quotation from the Buddha (XXIV.40).

40.

Yaft prafityasamutpadarrt P<lyatida'f!l sa payati,

dul?kha11} sa1n.udaya1(l caiva nirodha'f?J margam eva ca.

354

THE PHJLOSOPHYOFTHE MIDDLE WAY

Whoevcr pcrceives dependent arising also pcrceives suffering, its arising,

its ceasing and the path (eading to its ceasing].
MKV(P) p .515; MKV(V) p.225 .

The Buddha maintained chat he who pcrceives dependenc arising pcrceivcs the
dhamma (M 1.190191 ; Chung 7.2 [Tairho l.467a}. Perceiving dependent aris
ing is not mcrely perceiving "things as they have come co be" (yath'iibhuta), but
also seeing how a human being placed in such a context of dependent arising
experiences suffering (dul/kha); what causes such suffering; what kind of
freedom can be attained and what the path is that leads to the attainmcnt of
freedom and cmancipacion. Dependent arising and the four noble truths arc,
therefoce, the foundations on which the whole edifice called Buddhism is built.
Any school of Buddhism chat refuses to recognizc the centralicy of these doc
uines would loose its claim to be a legitimate pan of that tradition. These con
stitute the truths chat Nigarjuna was attemptiog to explicatc in che pcesent
chapter.

CHAPTER
TwEN1YFIVE

Examination of Freedom
(Nirvar;,a~partk!ti)

l.

Yadi unyam idarp. sarvam udayo niisti na vy4Jal/,

prahaad va nirodhad va karya nirvaf!am i!ya.te.

If all tUs is empty, there ests neithec arising nor ceasing. [As such,]
through the relinquishing and ceasing of what does one expect freedom?
MKV(P) p.'.>19; MKV(V) p.227.

Even though someof the problems rdating to freedom (nittlaf'!a) have al.ready
been discussed in the context of a tathagata (Chapter XXII) and the four noble
truths (Chapter XXIV), Naga.rjuna singles out this topie for further analysis,
especially in view of certain grave misconceptions that emerged as a resuJt of a
substantialist world-view. Without making any attempt to determine what
these metaphysical views were, many a scholar has plung.ed himself inro assertions about Nagarjuna's coaception of freedom and produced views that arc no
less metaphysical than those rejected by Nagaijuna.
The present verse 6cplains the di.fficulties the metaphysicians, who acceptcd
substantial phenomena, cxpeticnced when they attempted to elucidate
freedom. While the ptcvious chapter dealt with the problems the substantialists faced when they accepted a notion of self-nature and tried to accountfor
che four noble t.ruths, especially the empirical fact of sliffering (duf?kha). the
present context Nagarjuna is primacily concerned with the question of freedom
from suffering. The substantialist dilemma herc would be : How can freedom
( niniiin.a) be explained , if the conception of empciness is ucilized not only to
explain the empiricaJ fact of~suffering but also to elucidate ninii'!a) which,
even though described as the cessation of sufferiog (du~kha-nirodha), was also
the uJtimace or absoluce reality.
After deoying any substance (111ahava) in the conventiona! or the contextua
(11yav11hara). if Nagarjuna had not proceeded to extend that denial to the idea
(p11ramiirtha), the substantialist could have remained silen.t. Hbwever, when
Nagarjuna univetsaJ.ized "the empty" (i#njll) by saying "aJJ this is empty" (IllT:
3)5

356

THE PHJtOSOPHY OFTHE MIDDLE WAY

vam idaf(J fUnyatrJ)- and that included the substantialiscs' conception of

"ultimate reality" (poramiirtha) understood in various ways, either as "atomk
reality" (parama11u) or the "seed of rdease" (rnok,a-liija) or the "originally
pure thought" (prakrt1:prabhasvarr1-citta)(see XXIV.32)- he was rhreatening
the very foundation of substantialist metaphysics . In the next cwenty-three
verses, Nagarjuna proceeds to demolish that foundation .

2.

Yady af unyam z'da17J saroam udayo nfisti na vyayaf!,

prahar(iid va nirodhiid va kasya ntNiiit?am l!Jdle.

'
lf all this is non-empty,
thete exists neithec arising nor ceasing, [As such,]

through relinquisrung a nd ceasing of what does one expect freedom?

MKV(P) p.521; MKV(V) p.227.

As in the previous chapter, it is because of his preference for an empiricai explanation of things in terms of arising and ce~ ing (udaya.vyaya) _that Nagarjuna is not willing to accept a mctaphysical substaQce. Hence his argument that
relinquishing (praha11a) a.nd ceasing (nirodha) are meaningless in the contcxt
of the ''non-empty" (afiinya). Thus, for Nagarjuna, the non-cmpiricaJ view is
the one expressed in the form , "All this is non-empty (sarvam idam afUnyatrJ) ,
rather chan .the view presented as "AU this is empty" (sarv(lm ida?(l fUn.:;af!l) . If,
on the i:ontrary , the substanti.ists were to accept the empirical phenomena to
be empty, but noc.ni rvaf!a, srill Nagarjuna could ask the question: lf nif'tl(Zf!a is
a permaoenr and eternal substance, why talk of arising and ceasing, relin
quishing and abandoning. for suffering (duf?kha) or defilements (kiesa) that
need to be relinquished actually do not affect the originally pure existence?
For Nagarjuna and the Buddha, neither the empirical events nor the
utimate fruit wece substantial. Neither samsara
nor nirvana
.
. were' absolute in
any sense. As such, what Nagarjuna is presenting is not different from the cmpiricaJ view of bondage and freedom explained in the Buddha's discourse. ro
Katyayana.

3.

Aprahzam asampraptam anucchinnam a?i:fvata?(l,

aniruddham anutpannam etan nirvaam ucyate.

EXAMINA.1.lON OF FREEDOM

357

Uruelinquished, not reached, unannihilated, non-eternal, no-ceased

and non-arisen - this is called freedom.
MKV(P} p.52 l i MKV(V) p. 228 .

Having rejected a permanent and eternal substance which, as explained earlier

(XXIV.38), "is unboro, unceased, remaining immovable and devoid of
variegatecl _situations," if Nagacjuna were ro explain freedom jn almost idenrial
terms, his philosophy would not be any differen.c from those of his substancialist opponents. Furthermore, it is the subscantialists chemselves who considered freedom (nirvaf!a) as well as the non-defiled good seeds (atiiira11a
kusalt:1-btja) oc the seeds of release (mo~a-bija) to be subde (susuk~ma) and
uncorrupted or undescroyed (nasamudghata)(see XXIV.32). This leaves t he interpreter of the present verse with only one escape-route, namely, to consider
the six negations presented hece to be simple oegations of metaphysical existence (sat) or its associate, non-existence (asat).
How relinquishing (praha11a) becomes probematic if one we.re to accepc a
self-nature (Jvabhava) has been explained at XiflV.27. The substantialists, as
mentioned in the ruscussion of XXN.32, were left with two distinct selfnatures: the good (kusa/a) and the bad (akwala). 'f.heit problem was how any
one of them, being permanent and eternal, could be relinquished .
If it is assumed that the "seed of release" (moh.abiJ~) is subtle.and incorruptible and produces freedom (niniiit?a) as the "ultimate fruit" (parama artha or
phala), Nagarjuna's refutation f ie is dearly stated at XXIV.28. He explained
how the reaching (sa~priipti = sama'dhigamana) of a f rui t (phalr;) is made impossible by the recognitioo of a perPianent and iocorrupcible selfcnarure . Thus,
when freedom (nirva11a) is explained as rcqlinquishing (prahat1a) and reaching
(sa1!Jpraptz), these should not be understood in a substantialist way. I/ they
were .ro understood, the oly way to explain the empirical concepcion of
freedom is by negating them . Thus, in freedom there is "oo r.hing1' (na
kacit)(XXV . 24) - a thing understood as being substantial or haviog sel:fnarure-that is either relinquished (prahT11a) or reached (Ja1!Jprapta) .
The other four negations were distussed in relation to the Dedicacory Verses
(as part of the eighr negations). There roo, it was pointed our tiat che eighc
negations-were intended to elim.inace the metaphysical nocions of existence
(astilva) and non-existence (naJtitva). The same could be said of the two
couplets in the present verse : anucchinna?!J/ afiifvatatrJ and aniruddhaf?l/ anu I

pannartz.
Thus. the sixfold acriviry- rclinquishing, reaching, interruption, remaining.
constant, ceasing and arising - as explained by the metpphysicians need to be
negated beforc an empiricist theory could cmerge. Such mecaphysical explana-

358

T HE PHlLOSOPliYOFTHE MIDDLE WAY

tions can easily be abandoned if their foundations, namely. the metaphysical

concepts of substantial existence (sat or t:!Jtit'lla, bha11a oc s11abha11a) and
nihilistic non-existence (a.rat or nastit11a, abh'ii11a or parabh'iiva), wece to be
given ~p. This , indeed, is the task that Nigarjuna sers up for himself in the
following verses.

4. Bhavas t'iivan na riinia'!af!' jara-maraa--lak!a1Ja7?1,

praJajet?isti bhavo hi na jfl1'a-maraf!af!' virfi.

Freedom, as a matter of fact. is not cxistence, for if it were, it would

follow that it has the characteristics of decay and death. Indeed, there is
no existence without decay and death.
MKV(P)

p . 524~

M.k"V(V) p.229.

Inada ~es bhava in the present context as the nature of ordinaty existence. Instead, the term is better explained as a reference to metaphysical existence
(a.rlitt1a), as has been the case with Naprjuna (see especially, XXl .15-16; also
XXV.10). The metaphysicians who admitted bliava or rvabhava as eternal and
perrnanent entities nevcr attempted. an empirkal justification of tbeSe eotities .
They seem to have assumed that these are known through "ominscience" (stlf'tlt1Jf4f11a) . Nigarjuna, realizing the Buddha's attitude toward such knowledgcclaims (sec discussion of Sabba-sutta at IX.3) and, thcrefore, making no
refcrcncc to sucb knowledge at all in the present text, simply noted the implkations of rhls metaphysical sp eculation and iosisted that such existence (bhava,
s11abhi11a) is invariably associated with charactcristics (/a/qaa) , Uke decay and
death (jtll''ii-maraf!a). Thus, freedom will not includc frecdom from decay and
death and this would contradict the Buddha's claim that he is freed from the
recurrent cyde of births and deaths. On the contrary, if the mctaphysicians insis.ted that there is no invariable connection between existence (bhava) and
characterisiics (la/qaf!4), then they <.:ould claim tha t after p11rinirvar1a a freed
peoon can retain the eternal existence and drop the characteristics. N:igarjuna was
qui~e certain that this is an extension of the metaphysical position and is in conflict with the non-substantialist teachings of the Buddha (sec X:XV. 17).

EixAMINATION OF fREBDOM

359

5. Bhii11a ca yadi nirvaf!tlf!I ni111ii1'Jtlf!I Jt1f!IJRrlt1f!I bht111et,

niiJaf!lskrto hi vidyate bha11al/ k11acana Racana.

Morcovcr, if frccdom wcre to be cxistencc, thco &cedom would be condicioned. Yct, an existence that is unconditioncd is not evidcnt anywbere.
MKV(P) p.526; MKV(v) p.230.

In rendering the terms Jaf!JSkrta and t1.It1f!llkrta as "condirioned'' and the "uoconditioned" rcspectively, we havc, in the present cootcxt, tried to ret.a.in the
interpretation of the metaphysicians, for it is this panicuJar interpretation that
is being questioncd by Nagarjuna. UndoubtedJy, it is the Sarvastivada theory
which cquatcd Jtlf!IJR[/4 and prafityt1.Iamulpanna that Nagat:juna has in mind
(see commcnts on VTI. 1). The implicacioo of this equation is that "the uocon
ditioncd" (asaf!1Jkr14) is also the "independent" or "uncaused''
(aprafityasamutpanna), an implication not acccptable to both Nagarjuna and
the Buddha.
In fact , XXV .4-6 highlight the sharp distinction the metaphysicians assurned
between frcedom and bondagc (sec Chaptcr XVI , Examinacion ofBoodagc and
Releasc). lt is this sharp distinction, whkh is the foundation of the substan
ciafuc cxplanacion of freedom and bondage, that is being analysed by Nagarjuna. Unlcss this metaphysical cxplanation is kcpt in view. speculatoins about
Nagarjuna's owo conception of frecdom can turo out to be as wcird as those of
the met.aphysicians.

6. Bh'ii11a ca yadi nirva11am anupadiiya taf kathaf!I,

nirv'ii'!tlf!l niiiulpiidiiya kaicid bhii110 hi 11idyate.

Furthctmocc, if frccdom werc to be cxisrcnce, how can tbat freedom be

independent, for an independent cxistcncc is ccctainly not cvidcnt?
MKV(P) p.526; MKV(V) p.230.

360

THE. PHILOSOPH Y OF THE M IODl.f. WAY

Hece agaio we have a metaphysical inte(pcetation of upadiiya (see IV.1-2)

whece existence (bhava), as the permapeot and the eterna , becomes not only
che "unconditioned" (XXV. 5), but also the absolutely "independent"
(anupadiiya)(see also x!I.5 ). Nigarjuna is not prepared to equate freedom
with such non-empirical existence.

7,

Y adi bh?ivo na nirziiif!a1n abhavaf? ki?p. bhavijyati,

nirvfirta?p. yatra bhavo na nabhavas tatra vidyate.

If freedom is not existence, will freedom be non-existence? Wherein

there is no existence, therein non-existence is not evident.
MKV(P) p. 527 ~ MKV(V) p.230.

The metaphysical notions of existence and noo-existence expressed in such

terms as aststva and nastitva are hece ccferred to as bhava and abhava. They
bcing back the pcoblems of identity ,and differences (see 1.3). Nigarjuna's argument, as at 1.3, is tbat these rwo are relative concepts.

8.

Y(ll/y abhava ca niniartam anupadiiya t4t katha?p.,

nir.va1!a?!J na hy abhavo 'stt' yo 'nupadiiya vidyate.

If freedom .is non-ex.istence, how etn freedom be independent? for there
exists no non-existence which evidently is independent.
MKV(P) p.527 ; MKV(V) p .231.

If there were to be no p ermanent existence, there could be oo permanent nonexistence . If chere wece to be no absolute existence , there could be no nihilistic
non-existence. Just as much as freedom is not absoluce existen~e, it is also not
nihiisLic noo-existence, Nigarjuna's atternpt in the present chapcet is directly
aimed at getting rid of sucha metaphysical explanation of freedorn . The Buddhists, throughout history, were prone to .make such distinctions, sornetirnes
ignoring the empirical middle position advocared by the Buddha. This is what
Nagar)una intends to explain in the next vr:rse.

EXAMINA1 lON OF FREEDOM

361

9. Ya ajavaf!Zfavz-bhava upadiiya prafitya va,

so 'prafity?inupadiiya nirvfirt-am upac/iryate.

Whatever is of che nacure of coming and going that ocrurs contingently

or dependently. Freedom is, rherefote, indicated as being noncontingent and independent.
MKV(P) p. 529; MKV(V) p.23 t.

Eveo though the Buddha empahsized the avoidance of metap,hysicaJ extremes,

there was indeed a tcndency to distiogulsh saf!Zs'iira and niniii'!a. The cycle of
binhs and deaths, "comings and goings" (iijavaf!Zfa'llif!Z), was looked upon as
having its owo nature (bhava) which is dependeoce. This is contrasted with the
narure (bhiiva) of freedom which is independent. The inuusion of the substantiaist notions in the form of saT!Jvrti (convention) and paramartha (''ultimate
reality") into the Buddhist doctrine during the scholastic period has aheady
beeo rcfened to (see comments. on XXIV.8). Even though most Buddhist
rhinkers were able to avoid such ontological speculacions in explaining the
nature of the world and freedom, the Buddhist metaphysicians had difficuty
in avoiding such a dichotomy. Therefore,
the next verse, Nagarjuna proceeds to remind the Buddhists of the Buddha's teaching on becoming and
other-becoming (bhava-vibhava) .

in

10.

Parhii11af!Z cabravzc chasta bhav"!ya vibhavasya ca,

tasmiin na bhavo nabhavo nirva11af!J iti yujyate.

The teacher has spoken of relinquishing both becom.ing and otherbecoming. Therefore, it is proper .to assume r.hat freedom is neither existence nor non-existence.
MKV(I() p.530; MKV(V) p.232 .

In addltion to the two concepts of Saf!Zbha11a and 11ibhava, discussed previousy

(Chapter XXI), Nagarjuna introduces another pair: bhava and vibhava. lt is
significant to note that when Nagarjuna rejected astitva and nastitva b~ was
quoting the Buddha (XV. 7). He does the same when he refers to bhava and
vibhava (XXV. 10). Yet, when he rejects bhava and abhava, rwo concepts that

362

THE PHltOSOPHY OF THE MIDDLE

WAY

do not appear in the early discourses, except as suffixcs, Nigarjuna mercly in

dicates that lhe Buddha realized their implications (bhav'libh'li11a-11ibhavina,
XV. 7) and does not present it as a statement of the Buddha. This supports our
previous contention tbat bh'iit!a and ilbha11a represent an attempt to reintroduce mtitva and riiistitva into philosophkaJ discussions by the
BrahmanicaJ and Buddhist metaphysicians.
One significant difference berweeo the concepts of bhiva and abhiva (astitva and riiislitva), on the one hand, and bhavaand vibhava, on the other, is that .
the former arc metaphysical ass\unptions and the latter dcscribc cmpirical
such, the reasons for their rcjection would be different. Bhiva and
events.
abhava arc rejected bccause they arc mctaphysicaJ and, as such, unvcrifiabe in
terms of the cpistemological standpoint adopted by the Buddha. On the cootrary, bhav~ and vibhava arc experienced processcs of life, namcly, becoming
and ocher-becoring, and the reason for theit being rejecced is chat they lead to
suffering (duf?kha) . B'hava is the process of becoming conditioned by dispositions. Hence .the Buddha spoke of "craving for becoming" (bha11a-l411h'ii) , a
constant thir'ting to btcomc this or that. Vibhava could mean either ''becoming something different" or annihilation of tbis process of becoming (namcly,
suicide). This was aJso referred to as a form of craving (vibhava-tat?hi).
Howeve.r, the process of becoming and becoming ocherwise could be explained
in terms of metaphysicaJ and, therefore, wrong beliefs (mithyii-drrti) in absolute. existence (bhiva) and nihilistic rion-existence (t1/?h'litit1), respeccively, in
stead of dependent arising (prt1t1tymt1mutpida) . Neither of the metaphysicaJ
explanacioos, according to the Buddha, could serve as the basls for frcedom.

As

11.

Bhaved abhavo bhiivaf ca nirva11af?'J ubhayaf!l yadi,

bhaved abh'iivo bhavaf ca mo/qas tru ca na yujyate.

If freedom were to be both existcnce and non-existence, chen rclease

would also be both existence and oon-existence. This too is not proper.
MKV(P) p.531; MKV(V) p.232.

It is possibe for someone to assume that "freedom" represents a special kind of

existence which combines both exi,stence
and non-existence. Jf
it is understood
.. '"
.
as some myscerious existcnce beyond exi~tence and non~existence, Nigarjuna
reminds his opponents that it is no more than simple "release" {mo/qa),and
this latter nced not be describeq
Release is
.' as both existence and non-existence.
.
'

ExAMINATON OF FlUlBDOM

363

no moce than the release from suffering and there is rtothing transcendental
about it.

12.

Bhaved abhiit10 bhiit1af ca nittlafam ubhaya'!I yadi,

niinupiidiiya nifV!ii'!am uP"iidiiyob/jaya'!J hi tal.

If freedom werc to be both existence and non-existence, freedom could

not be independent, for existencc and non-existeoce arc, indeed, dependent upon one another.
MKV(P) p.531 ; MKV(T1 p .232.

As pointed out eadiei:, the substantiaist definition o( free dom implied independence, rather than non-grasping or non-dinging (sec comments on XXIl.5). Thus, if the metaphysicians were to define freedoro as both existence and
non-existence, it could not be independent as they expect it to be. This is
because existepce and non-existence are dependent.

13.

Bhat1ed ahha110 bha11af ca nirva11am ubhya~ katha?"?Z,

tZSflf!1Jkrta~ ca nirvfif!tlf'?Z bh?i11abha11au ca sarrukttau.

How could freedom be both existence and non-existence, for freedom is

unconditioned while existence and non-existence arc conditioned?
MKV(P) p.531 ; MKV(V) p .233.

As in the .previous verse, here too Nagiirjuna is takiog the definition proffered
by the opponent and cdticizing his conception of freedom. This is similar to
the argument adduced by Nagarjuna atXXV .6 , except that in the present case
Nigarjuna is insiscing that if something is both existence and non-existeoce, it
could not approptiately be called an "unconditioned." 'The "uncondi tioned,"
by the meraphysician's definition, is the "inde(>endent" (apratftyasamutpanna),
and an element witb rwo eotities as parts of it will always be conditioned by
those rwo parts.

364

THE PHILOSOPHY O!' 'rHE MIODLE WAY

14. Bhaved abhavo bhava ca nirvaflam ubhayaf!' kathaf!' ,

na tayor ekatrastitvam (j/oka-tamasor yathli.
How could freedom be both existeoce and ooo-ex1stence, for theit
simultaoeious ~xistence in one place is not possible, as in the case of light
and datkness?
MKV(.P) p ..532; MKV( V) p .233.

This brings out anothcr aspect of the mecaphysicians' conception of frecdom ,

namely , the mooistic implication. Nirvfi11a, looked upon as the "ultimate reality" (paramartha), could not accommodate variety and multiplicity. For them ,
ie is the ooe ultimate truth, everything else beiog provisional. However, Nagarjuna is insisting that they could noc maintain that nirv'iif!a is both cxistencc and
'
noo-ex1stence.

15.

Naivabliiivo naiva bha110 nirviif!am iti fii 'njanii,

abhave caiva bhave ca s'ii siddhe sati szdhyati.

The proposition that freedom is neitber existeoce nor no-exiterice could
be established if and when both existence and non-existence arc
established.
MKV(P) p .532; MKV( V) 233 .

Anjanii from (yanj, ''to annoint") can mcan a symbol. Here Nagif.rjuna is not
speaking of logicaJ proof to decide the validity of a symbolism , or a proposicion. Rather, he is interested in the ernpiricaJ justification. Thus, unless both
existence and non-existence, as cxplained by the metaphysician, can be shown
~o have empirical validity, their oegations too would ruean nothing.

16.

Naivabh'iivo .naiva bhavo ninia??af?i yadi vi'dyate,

nai11abhavo naiva bhavo iti kena tad ajyate.

EXAMINATION Of FREEDOM

If freedom as neithec existence noc non-existeoce is evident, by means of

what is it made known as neither existence not non-existence?
MKV(P) p.533; MKV(V) p.233.

Thls seems to be a rather devastating cr1ticism of the more popular view that
freedom caooot be expressed either as existeoce or as non-existence. That is,
it is linguistically transcendent or is ineffable. If the statement, "Freedom is
neither existence nor nonexistence," is intended to express the ineffability
of nit;iii'!a. Nagarjuna is questioning the meaningfuloess of that very statemenc. This means that absolute negation is as metaphysial as absolute assertioo , ''absoute nothingness" is as meaningless as "absolute chingness," and
these arc che absolutes chat are expressed by terms like bh'iiva and abhava.

17.

Paraf!Z nirodhad bhagatiiin bhavafity eva nohyate,

na bhavati ubhayaf!t ceti nobhaya'l!Z ceti nohyate.

lt is not assumed that the Blessed One exists after deat h . Neither is it
assumed that he does not eist, or both, or neither.

18.

T1thamano 'pi" bhagavan bhavality eva nohyate,

na bhavaty ubhayaf!l- celi nobhaya1p celi nohyate.

lt is not assumed that even a living Blessed One exisrs. Neither is it

assumed that he does not exist, or both or neither.
MKV(P) p.534; MKV(f-9 p.234.

These, indeed, are the metaphysica assertions made regarding the tathagata
(XXII) which both Nagarjuna and the Buddha rejecied. The foundations of
these metaphysical views were the theocies of identiry and difference. This very
important problem is next taken up for examination by. Nagarjuna.

19.

Na s_a'l!Zsarasya nirv'iif'!al ki~cid asti 11iift11Jaf!l,

na nirv1if!asy a sartuarlit kif?tcid a.sti vie[f.l'!fl'f(t.

366

'f.HE PHILOSOPHY OF THE MIDDLE W AY

The iifc-pr~ has no thitig that distinguishes it from freedom. Freedom

has no thittg that distinguishes fr from the life-process.
MKV(P) p .535 ; MKV(V) p .234.

This statement of Nagarjuna has contributed to a major and wide-spread assertion regarding the uniqueness of Mahayana philosophy, namely. the ultimate
identity of sa1{'Jilira and nirvii11a. This assertion may appear to be correct, if we
arc to ignore all that has been said by Nagarjuna regardiog the metaphysical
doctrines of. identity (ekatva) and difference (n'iinatva) , especially in the
chapters dealing with the tathagata (XXII) and the four truths (XXIV).
Those who upheld the view that this statement is an assenion of the identicy
of Sllf!JSfJf'fJ and nirtiii11a do not seem t have paused for one moment to reflect
on the question regarding the nature of the identity they were implying; nor
have they attempted to place that conception of identity ( if there is one) in the
historical context.
The facr that chere wa's a great urge on the part of the Buddh.ists, especially
after the Buddha's death, to taise him to the level of a supreme being, having
no real connections with the ordinary human world, can easily be seen from the
mote popular Mahayana texts lilce the Lalitavistara and the
Satldh11rmapuf!t/4.fik11 as wel as the Theravada treatises like the}iilaka-nidiinakathii. Through thac popular perspective, substancialist views regarding the
nature of the Buddha began to emerge among philosophers. These
philosophical views were prominent with the early metaphysicians like the Sar
vastivadins and the Sauuantikas .
.Nirv'a11tJ, understood as the "ultimate reality'' (paf'l~martha), came to be
distinguished from sa1{'JS'iira, the unreal, he convention, the impermanent .
The Buddha's owo view chat niniiif?IJ is the "ultimate fruit' ' (paramattha), a
culmination of the fruit (111th11) of everyday life, the highest mora peection
invoJving the eschewing of aJl immoral cooduct (sabba-pap11ssa a.kartT!/111?") and
the promotin of good (kus11/111s11 upp,;11mplll/ii), was gradually being forgotten .
Dharma was gradually being distinguished from the dharma in che dharmaadharma context. The inapprop,riate equation of the pu;,ya-papa distinc:tion
with the dharma-adharma distinction (see commeots on XVJI.24) led to the
belief in a Dharma that transcends che dhamza..adhanna distinction.
However, this was not Nagarjuna's understanding. Vyavah'iira, the contex
tual or the conven,tional , 1s not provisional in a cheap pragmacic sense and,
therefore, unreaJ. lt is the concrete wichout which the ideal (paranzartha)
makes no sense at aU (XXJV.10). lt is the absolute distinction between
vyavahara and pararnartha rhat is denied in che present verse.
The fact that the Buddhist metaphysicians were involved in Jong-drawn

EXAMJNATlON OF FREEOOM

367

phiiosophical disputes with the Brahmanical schools of thought and. therefore.,

the possibility of mutual influences should not be discounted. For example, the
Saurranrika philosophical standpoint resernbles in many ways the standpoint of
the Vaise~ikas . This Jatter philosophy is founded on the basie methodology of
makiog "distinctions" ( vie1arJa). Nagarjuna's present statement can be properly evaJuated in the background of such a methodology.
Indeed, the most sigoificant part of the statement is: na ki1?1-cit asti
11iefa11a'!' , i.e., "there is no thing that can be taken as a distinctioo," or "tbere
is no distinct thing." (see XXV.20 and 24). The reference to somc metaphysical
entity as "some thing'' (kif?l-cit) is a popular feacure in the early discourses as
well as in the Prajflaparamita-sutras.

20.

Nirvaf!asya oa ya kofi/t koti/t saf?Zsara11(1.Jya ca,

na tayor antaraf?Z kif?Zcit susuk1mam api vidyate.

Whatever is the extremicy of freedom .and the extremicy of the lifeprocess, between them not even a subtle something is evident.

MKV(P) p. 5 3 ~ ; MKV(V) p .235.

H aviog rejected a sharp dichotomy between saf?Zsara and niroiif!a, Nagarjuna is

not willing to let the metaphysician discover something ~xttemely subtle
(kif?Zct't suiuk/maf!J), comparaele to the extremely subtle and incorruptible
seed of release (mokfa-blja)(see comments on XXIV.32), between saf?Z.Jara and
nirvii11a. The identity theories of both Sarvastivada and Sakhya school posited
such subcle encicies to explaio cootinuity. Nagarjuna's present stacement
should be understood in the background of the ideas expressed by these
schcols.
Therefore, the translation of ,he first line as "The limits of nirvatJa are t he
lin1ics of ItJf?Zsard' (Ioada, p.158), is not very appropriate. Furthermore, che
first line does noc make a complete sencence. The emphasis in the completc;,
sentence is oo the phcase ~if?Zoit susikpmaf?Z (see XXV . 19), which is
Nagarjuna's major concern hece.

21.

Para:?p nirodhad antlidya"f; favatlidya ca dofaya/f,

mrvlittam aparantat?'t ca purvanta'f!J ca sam'iifrita/?.

368

THE Pl-OJ.OSOPHY Of THE MIDDLE WAY

Metaphysical views relating to the finite, etc to the etcrnal, etc after
death arc associated with [the problems of] f.reedom as well as the
posterior -and prior exttetities.
1WKV(P) p.536; MKV(V) p.23).

Herein, reference is made to eight of the ten "undeclared'' (avyakrta) issues.

Nagarjuna perceives them to be associated with the question of freedom , a
freedom 1ooked at from two of the major concerns of.the human bcings, namely, the first bcginning and the ultima te end , the flrst eause and the final cause.
Human concerns like "Where did I come froro?" or "Where woufd I go after
this?" (as will be exlained in the fina Chapter on "Examination ofViews") have
influenced human thinking and often led to metaphysical views about freedom
and bondage.

22.

Silnyefu sarva-dharmeftt ki11J anantaf!Z kim antavat,

kim ananta1n antavac c.a n'iint#nta??t nfit111ntavac ca ki'!Z.

23.

Kir(I- tad eva kim anyat kir(I- afvalaf!l kim a1ivataf?1.,

a'iiwata1!l avata'f?Z ca kif?Z va nobhayam apy atai;~

When all things are empty, why [specuate on] the finite, the infinire,
both the finite and the infinite and neither finite nor theinfinite? \Vhy
[speculare on] the identical, the different, the etetnal, the non-eternal,
both or neither?
MKV(P) p.537; Ml-.'V(V) p.235 .

Empiricist Nagarjuna's fina! fina! qucstion, after presenting such a detailed

aoaiysis of all the ,netaphysical views to which he has devoted twenty-five
chapters, is "Why raise all these metaphysical issues, when aJl experienced
thlogs are empcy?" (iinyer11 sarva-dharmefu kif!l . .. ). As reiterated before in
the commentary, he is spea!dng about things that are empty (unya), wiich arc
the empirically given
which are seeo to be dependently arisen
(praityasamutpanna). H e is not emphasizing the absttact concept,
"emptiness" (iunyata).

and

EXAMINATION OF FR.BIDOM

24.

369

Sarvopalambhopaa11J11f? prapancopaamal? ivaf?,

na kvacit ka1yacit kafcid dharmo b11ddhena deital/.

The Buddha did not tcach the appeasement of all objccrs, the appeasement of obsession, and the auspicious as some thing to some one at some
pbcc.
MKV(P) p.538; MKV(V) p.236.

Inada has broken up the above inro two distinct statements and thereby lost its significance. The first line contaios qualificarions of dharma occurring in the second.
N:igarjuoa's emphasis is on the phrase: kacil dharmaf, as it was in the case of
the two important statementS made previously in the present chapter
(XXV. 19-20). When the Buddha spoke of freedom (nirva fla), which he
quaJified as the appeasement of aH objects (upaklmbha = al11mbana =
dra1ta11ya, cf. drfJ!(avyopaama at V.8) or the appeasement of obsessions
(prapalcopafama) or auspicious (Jiva, sec Dedicatory Verses). he was nor refeuing to them as "some thing,'' that is, some cotity having a specific distinction
(viiefaf!a) or subdety (1uiulqma111a). He was oot speaking of freedom in the
way the substantialist cooceived of it.
All these twenty-five chapters are, therefore, negative io characrer and tone.
They were devoted to a rcfutation of the two metaphysical but related views of
existence and noo-ex.istence (astitv'ri-ntistitva). Having cleared up the dust that
had gachered around all the doc~rines preached by the Buddha, Nagarjuna is
now ready to go on to the positive description of bondage and free dom as
enunciated by the Buddha in the discourse to Kacyayana. Nagarjuna's ttearment of the metaphysicaJ issues chat ernerged in relation to all these doctrinesdoctrines pertaining to causacion, change, the human personaJity, survival, karma, mora respo.nsibi ity, and freedom-is so cxhaustive and compe te that he
can proceed to explain the Buddha's conception. of bondage and freedom
without any fear of any one raising any question. Por he has already answered
them all. Herein Lies the greatoess of Nagarjuna as a philosopher.

CHAPTER
TwEN1Y SIX

Exarnination of the Tweve Causa Factors

(Dvadasaga-paffk!ti)

l.

Punarbhavaya sa??Zskiiran avidyii-nivrtas iridha,

abhiJa1(1Jk11rute fiit?ts tair gatt~ gacchati karmabhil{.

A person eovdoped by ignorance forms such dispositions in the threcfod

ways cading to re-b~oming, and through su.eh actions be moves on to
his destiny.
MKV(P) p.542; MKV(V) p.238.

This chapter is of itde significance to most Nagarjunian scholars. I nada argues:

With the discussioo of Niniii'!a in the last chapter the treatment
from the standpoinr of the Mabayana had basically come to a dose.
ln tbis chapter and the fina one to follow, Nagarjuna goes into the
analysis of the Hrnayanistic doctrines .... The discussion is
Hfnaya,nistic and it reveals that the source of trouble lies in ignorance which in turn initiates all kinds of meotal confotmations

(satpskiira) (Niigiirjuna, p. l60).

While disagreeing totaly with this interpretation, I propose to show chat thls
chapter represents the mosc positive explaaation by Na:garjuna of the Buddha's
teachings oo boodage and freedom as enunciated in 'the vast collection of
discourses, with the discourse to Karyayana as the pivot.al text. lndeed, without
the positive te.i,chiogs pl'esented hece coming after the negative analysis in the
last twenty-five chapters, it is rather difficult to see how Nagarjuna could claim
to be restating the Buddha-word, as he openly expressed it .i n severa! places
eaclier (see XV.6; XVII.6, 8, 11; XXIV.8-9). It is indeed surpri!ling to see how
most interpreters of Nagii;rjuna are loolcing for posidve statemeots of the doc
370

Ev.MINATION OF THE TwELVE CAUSAL FACTORS

37 1

trine primarily through negations {XXV .3 and Dedicatory Verses), while ignoring al the positive statements of Nigarjuna.
XXVI. l is a furt her cxplanation of the Buddh a's statement : "Depending
upon ignorance arc dispositions" (aviJJapaccaya sakhafii). lts d aborarion is
also the Buddha's and is presented in the Kukkuravatika-rutla (M 1.390).

2.

Vijffiinarp 1arpni11isa1e sarpsliiirapratyayarp gatatJ,

1arrznivi!fe 'tha vi/ fiiine niimafilparp nt! icyate.

Consciousncss, with disposition as i~ conditioo, cnters [the new] life.

Whcn coosciousnc:ss has cotercd, the psychophysical pcrsonality is i.nfuscd.

MKV(P) p.543; MKV(V) p.238.

This represents the cxplanacion of the relationships described in the statement,

"Depcoding upon dispositions is consciousness" (sakharapaccaya viffiif!arrz)
and "Depcnding upon consciousness is the psychophysical personality"
(vifllaf!apaccaya namafilparp). The explication of the first of rhese is found at
S 3.87 ( Tsa 2. 14 [Taisho 2. llc)) where it is said: "Disposition is so called
bccause it proccsses . .. consciousness that has alteady beeo dispositionally
conditioned, into ics present state." The dispositions arc thus responsible for
providing an individualicy to consciousness, and it is this individuated consciousncss chat comes to be established in a psychophysicaJ personali cy. Th is at
cer event is explaincd at D 2.63 (Chang 10.2 I Tairho l.6l b]), where it is scated
that t he psychophysical personalicy formed in the mother's womb will not grow
in to marurity (na vuddhi'?l viru(hirp 'iiP111/at1) if consciousness we re not to enter
it and get established therein. In fact, the Chinese translatioo of Kuma rajl va
refers to such attainment of marurity of the psychophysical pcrsonaJity.
Those who aJ"e unwilling to atcribure a doctrine of survival to Niga rjuoa may
interprer the relation becween consciousoess and the psychophysicaJ pcrsonaity
as oo more chan the explanation of the act of being conscious during ordinary
day to day experienccs. lf so, ir would have been possibJe for Naga rjuna to say
so at this point. Instead , he ptefers to sp eak of a life (.gah) , generally
understood as a fururc life or desdny, and proceeds to explain the process of
perception only after outlinjng the emergence of the six sensory faculties
associatcd with the psychophysicaJ personalicy.

372

THE PHILOSOPl-IY OF THE MIDDLE W Y

3. Nifikte niirnafiipe tu 1at!iiyatana-sa?!Jbha11a~.

1arfayatanam agamya saf!JSparfal/ sar(Jpravartate.

When the psychophysical personality has bee.n infused, the occurrence of
the six spheres (of sense) takcs place. Depencling upon the six spheres
proceech cont.act.
MKV(P) p.552-553 ; MKV(V) p.242 .

The Buddha's statemenr, "Dependiog upon the psychophysical personality are

che six spheres of sense" (niimarupapaccay'ii sa{'iiyatanaf!'), explains an import.ant pre-requisite for the emergence of human knowledge. If the
psychophysical personaity is not properly formed or if the sensory faculties
were to be. defective, the sensory spheres such as materia! form , sound , smell,
taste, touch, and concepts would not provide a foundation for contact. Thus,
the visual organ, whicb Is part of the psychophysical personality should be
unirnpeded (aparibhinna) and should function properly. The external object
should come into focus (bahir'li ca Nipa ipathaf!J agacchatt) and attention
(samannahara). which is a function of consciousness, shou.ld be avaiJable (M
1.. 190; Chung 7,2[Taisho1.467a]) . Theo only there will cmerge consciousncss
which is a necessary condition for contact. This . idea is theo elabotaced by
Nagarjuna io the next verse.

4. Cakfuh. pralitya rupa'!' ca sa1nan11aharam, eva ca,

n'iimarupt1f'?J p rafftyaivaf!J

vijiinaf!J

JIJ1(Jpravartatc.

5. Sa?!Jipatas tray'ii1{ii1(J yo ropa-vijflanuCa~tt!'li'f(t,

sparal/ sah. tas1niit sparfac ca vedanii Jat!Jpravartate.

Thus, depeoding upon the eye and mateal form, and attention too, and
depencling upon the psychophysical personality pcoceeds consciousness.
'Whatever is the harmonious coocurrence of the three factors: material
form, consciousness, and eye, is contact. .Feeling proceeds from such
contaet.
MKV(P) p,553554; MKV(V) p.242.

Ex.AMINA TION OF THE TWEL VE WUSAL FACTORS

The i:welvefold formula presents feeling as being conditioned by contact

(phassapaccaya vedanii). However, Nagarjuna feds the need to explain what
contact is all about. He, therefore, inserts the cxplanation of cootact provided
by the Buddha in the Madhupif!fika-sutta (M 1.111-112; Chttng 28.3 [Tairho
1.604b]). Contact is che harmonious concurrence (sagatt) of three factors:
materia form , (visual) consciousness, and eye. lt is tUs harmonious concurrencc chat provides a foundation for feeling (vedanii) .

6.

Vedanapratyaya trf'f!a vedanarthaf!I hi trf'yate,

trf~amlina upadiinam upad11tte catut'tlidha'(Z.

Conditioned by feeling is craving. lndeed, craving is feeling-directed.

The one who craves, gtaSpS on to the foutfold spheres of grasplng.
MKV(P) p..554-555 ; MKV(fl?p.243 .

Hece Nagarjuna has an interesting explanacion of craving (tr1n.?i) when he says:

11edan4rtharp hi t(!yate, i. e, "it i~ for feeling that one craves ." Ie is a reference to
the pleasant feelings (ubha1 maniipa, or sukha). The fourfold spheres of grasping are: grasping for desires (kama), for views (dil(h1). for rules and rights
(i1/abba1a), and for theories of self (attaviida)(M 1. 51).

7. Upadiine sati bhava upadii11,f; pravartqte,

JYtidd hi yady anup1idiino mucyeta na bhaved bhaval;.
When grasping exis(S, bccoming on the part of che grasper proceeds. lf he
were to be !l non-grasper, be would be released, and there would be no
further becoming.
MKV(P)

p . 5~6;

MKV(V} p.243.

lt is inceresting ro noce that at this point i"n the explaoarioo of the twelvefold
formula Nagarjuna speaks of freedom or release (mak.fa). It is a point at which
,

374

TuE PH!LOSOPl-I Y OF THE M IDDLE

W AY

the process of bondage can be interrupted. Nagarjuna is problably emphasizing this because he witnessed how his fellow Buddhists , in spite of their enormous intellectual capacities, constantly adhered to metaphysicaJ views and wece
caught in "Brahma's Net" (brahma-jii/11). Some of these intellecrual giants, who
wece able to write commentary after commentary, even though they had abandoned househoJd lives wece cnamoced with their views and invoJved jo ceaseJc~
diatribes. Realiziog the emphasis placed by the Buddha upon "non-grasping"
(anupadiina), Nagarjuna interrupts his exposition of the process of bondage co
explain freedom.

8.

Panca skandh'iil/ sa ca bha11afl bhav'iijj'iitil/ pravartate,

j ar'ii-maraf!a-duf?khadi sokiil/ saparidevan'iif?.

9. Daurmana.syam upay'iis'ii 1ater etat pravartate,

kevalasyaivam eta.sya dul/kha-skandhasya sa1tJbhaval;.
The fivc aggrcgatcs constitutc this bccoming. From becoming proceeds
birth. Suffering rclaring to decay and death, etc., grief, Iamcntation, dejcction, and ruspair- all tbese proce~ from birth. Such is the occurrence
of this e.nti.re mass of suffering.
MKV(P) p.556-557; MKV(V) p.243-244

The elimination of craving for becoming (bhava-Jaf!h'ii), as mentioned

cadier (sec XXV.10), was encouragcd by the Buddha. The rcason for this
is not merely because craving causes suffcring in this life, but aJso because
it could lead t o birth in another Jifc (jatt) , whicb would also involve the
repetition of all the unfortunate experiences of the present life. Tht<
phrase duf?kha-skandha is used in the sense of heap or mass of suffering
rathcr than "suffering attached to the skandhas," (lnada, p . 162).

10.

Sa?!Jsar11rmu!an sao/skiiran avidvan sar!l.skaroty atal/,

a11idvan kiirakaf; tasrniin na vidva??Zs tattva- daraniit.

Thus, che ignorant forms disposjtions that constirute the source of the

fx.AMJNt\.110 N OP THE TwELVE CAUSAL FACI'ORS

375

life process. Therefore, it is the ignorant who is the agent, not the wise
one, because of his [the atter's] perception of truth.
MKV(P) p .558; MKV(V) p.244.

The enlightened one who had attained che appeasement of dispositions

(S4'f!Uk'iiropaama) does not conttibute to the perpetuation of the life-process
(.ratrJsara). Hence, he is not an agent (kart.1kt.1f!) respoosible for such perpetuation. This seems to indicate Nagarjuna's awareness of the Buddha's famous
statement chat upon attainiog enligbtenment he perccived the builder of the
house (gaha-karaka, Dh 154).
Reading too much metaphysics into the phrase tattv11-darlt.1nt.1 can destroy all
that Nagarjuna attempted to do in the preceding cwenty-five chapters. Instead
of explaining it as "the perception of an absoute reality," it could be more appropriate1y understood as the perception of the empirical truth that "all this is
empty" (slZT'llam t'daf!J Unydf!J) of a substance (svabhavt.1tal/).

11.

Avidyay1i1fl niruddhafiif!J saf!Jskiiraf!am asa1{Jbht.1vaft,

avidjiiya nirodhas tu jniinenasyt.1t'va bhavanat.

When ignorance has ceased, there is no occurre.nce of dispositions.

However, the cessation of that igoorance takes pace as a result of the
practice of that [non-occurrencc of dispositions] through widsom.
MKV(P) pp.558-559r MKV(P) p .244.

The murual relacionship betWeen cessation of ignorance and the aon-arising of

disposicions is emphasized by Nagatjuna, keeping in mind the distinction between the appeasement of dispositions (sa1{Jskaropaama) in the pcesent life
and their waning (saf!JSk1ira-k1ay11) at the time of death, theccby not pcovidi:ng
another opportuniry for rebirth .

12.

Tasya ttZSya nirodhena lal tat1 nabhipra1111rttite,

du/;Jkha-skandht.14 kevalo 'yam evaf!J samyag nirudhyate.

376

T HE PHILOSOPHY Of THE M.JDDLE W AY

With the cessation of these, these othere factors [of the twelvefold formula] would not proceed. ln this way, this entire mass of suffering ceases
comt>etely.

MKV(P) p. 559; MKV(V) p .244.

The emphasis is probably on the cessation of ignorance and the non-arising of

dispositions. These two factors are taken to predominace the entire ife-process.
The attainmeot of e~ghtenment and the appeasemet of dispositioos through
non-graspiog (anupadiina) account for "freedom with substrate'' (saupadiefa
nirv'iif!a), whi.le the elimtnation of dispositions provides for non-re-becoming
(a-punar-bhava)(cf. XXVL l) and the waning of birth (iiiti-kfaya) , which is
freedom "without substrate" (nirupadi1efa-11irvii11a).

CHAPTER

'fwENTY SEVEN
Examination of Views
(Drffi-parz k~a)

1.

Dr1fayo 'bhuvaf?Z nabhuvaf?Z ki?!/> nv alite 'dhvanlti ca,

jiis tal? avatalokady7if? puniiintat?Z samuparitaf?.

Whatcver views asserting an eternal world, etc. based upon [the perceptio.nJ: ''Did I exist or not exist in the past?" are associated with the prioc

end [of existence].

MKV(P) p.571; MKV{V) p .249.

The first line of the verse has been reconstructed by Poussin on the basis of the
Tibetan ttanslation.
In che Pari(eyyaka-sutta (S 3.94-99; Tsa 2,.,25 [Tatiho 2.13c-14a]}, the Buddha maintained chat views such as "The self and che world are etetnal" (sassato
afta ca loko ca) are metaphysical because they are dispositional answers
(sankharo so) to querries about the past such as: "D id I exist in the past or
not?" As such, chey are not based entirely upon empirical facts. lndeed, they
are attempts to go back to the prior end of existence (pubbanta). While the
Buddha recognized them as futile attempts, mostly because of the limitarions
of human knowledge and understanding (see Chaptet XI), he was not uowill
ing to accept wbatever evidence available through veridical memories (sadi,
pubbenivasanussati). On the basis of such evidence, he characterized the past
existences as impecmanent (anicca), clispositionaly decermined (sankhata), and
dependently arisen (pa;iccasamuppanna), and not as permaoent (nicca) , immutable (dhuva) , and eternal (sassata) .
Nagarjuna's explanatioo of the metaphysical views of eternalism (a-vata),
etc. is, therefore, a vivid representacion of the Buddha's attitude regatding such
v1ews.

2.

Dpfayo na bhavi!yami kim anyo 'nagate 'dhvani,

bhavzjyatniti cantadya apf.1intam. sa1nupt1ritaf?.

377

378

THE PHlLOSOPHY OF THE MIDDLE

wAY

Whatever views asserti.ng the finite, etc. based upon [the perceptioo]:
"Would I not exist U1 the futu.re or would I become soreQne else?" arc
associ.ated with the posterior end [of existence].
MKV(P} p .573; MKV(Vj p.249.

The assumptions that the self and the world arc finite or iefinite ate similarly
based upon one's dispositions (sakhara). Those who arc enamored with life as
it is would insist upon the world being infinite. Those who are averse to life
would advocate finiteness, insisting that there will be no way in which life
could continue. This is tantamount to annihilation (uccheda-ditthi, S 3.99;
Tsa 2.25 [Tairho 2.14bJ) .

3. Abhum atitam adhvanam ity api etan nopapadyate,

yo hi janmasu purve!u sa eva ,na bhavaty ayaf!Z.

The view that I existed in the past is not appropriate, for whosoevet was in the
prcvious birth, he, indeed, is not identica with his person.
MKV(P) p .573>MKV(V) p.250.

This sratemenr may appear, at first sight, to mean a denial of rebirth or survival; that is, it would be inapproptiate to say: "1 existed in the past." Howevet:,
if one "is careful in observing the emphasis in the second line, nan1ely, sa eva
("he himself"), it will become apparent that what is denied is not ally relatiooship between two lives, but the identity of the two persons in different existences. Therefore, the statement, "I existed i11 the past," becomes
metaphysical only if an absolute identity is posited, which indeed was the prod ivity of the metaphysician.

4.

Sa evalmett' tu bhaved upadiina'f!l vii1yate,

rtpadiina-vinirmukta atmii te katamal? puna/;.

EXAM1NATIC5N OF Vtl!WS

379

If it were to occur [tosomeone): "He, indeed, is the self/' thcn grasping is

identified. Scparated f.rom grasping, what coostitutc your selt?
MKV(P) pp.574-575; MKV(V) p.250.

The Buddha explained self-consciousness expressed in such statemenrs as "I

am ... "(asmltt) as dependent upon (upiidiiy11) the five aggregates. lt is like
seeing one's own reflection on a clean mirror or a pan of elear water. In this
case, ooe cannot perceive one's refleccion unless there is a personality that is
rcflccted, and these constitute the five aggregates (S 3.105). However, convert~
ing cllls sdf-consciousoess to a cogito, a "self' that is permanent and eternal, is
also the work of upiidiina, meaoing "grasping or clinging. '' lt is sometimes
described as thirsting or craving (ta ha) and this could be for becoming (bhav11)
or other.-becoming (vibh1111a)(S 3.26; Tsa 3.23 [Taisho 2.19a].
Tbus, neither self-consdousness nor a "self' would be fouod independently
(upiidiina-vinirmukta). Howcver, the attaino;ient of freedom, described as
anupadii vimukti, does not imply the ncgation of self-consciousness, but only
of the "thirsting for becoming/' (bhava-t1J'!h7i), which otherwise could lead to
the belief in a permaneot and eternal self that is independent of the aggregates, and idea already rejected by Nagaijuoa .in Chapter XXV.

5.

Upadiina-vinirmukto nasty atmeti krte sati,

syad upadinam evat-rna tzasti catm.eti 11a!, puna}/.

W hen it is assumed that there is no self separated from grasping, g~g

itself would be the self. Yet, this is tantamount to saying that there is no

self.
MKV(P} p.575; MKV(V) p.251.

The assertion that thc:re is no .self apan from grasping (upadiina-vinirmukta)

provides the identity tbeorists with the opportunity of ideotifying the self with
grasping. Thus, when the Bllddha, .having denied a self, maintained that consdousness (vifl;;a'!a) prov,ides a link. between two ives (D 3.105; Chang 12.2

380

TifE

PHILOSOPHY .OF THE

Mrooi.E WAY

[Taisho 1. 767aJ) , one of his disciplies, Sari, immediately assumed that "tbls
consciousness icself t.ransmigrates, and ot anotber" (zdam eva vifiiiiif!a'f(l sandhavati sa'f(lHtrati ananfla'f(l, M 1.256; Chung 54.2 [Tairho l .767a]), chereby
atrributing ao identity theory to the Buddha. The Buddha's immediate
response was to. deny such identity by asserting that consciousness is depeodenty arisen (paficcasdmuppanna). Theo the Buddha proceeded ro show how
consciousness, when explained in terms of causa! dependence, leads to a denial
of such identity. This, indeed, is what Nagarju na eodeavors to do in the
present cootext.

6.

Na copiidiinam evatmii vyeti tcrt samudeti ca,

katha'f(l hi namop'iidiinam upiidiiiii bhavi.fyati.

Grasping is not identical with the self, for that ~i.e grasping] ceases and
arises. Furthermore, how can grasping be the gta$per?
MKV(P) p.576; MKV( V) p .251.

As was dooe by the Buddha, (see comments above XXVlI.5), Nagarjuna is rejecting the notion of self (iitman) on two grounds . First, it is assumed to be
permanent and eternaJ , whereas grasping that is identified with the self is sub- ject to ac:isiog and ceasiog. Secondly, the self is also assumed to be the ageo t
behind all human action and, rherefore, of grasping. lf so, how can tt be both
action and agent at the same time?

7.

Anyaf? punar upadiinad atrna 11t1ivopapad'yafe,

grhyeta hy anupadiino yady anyo na ca grhyate.

Furthermore, a self thar is different from grasping is oor~ppropriate. A

pecson who is without grasping can be observed. However, if he were to
be differeht [from grasping], he could not be observed.
MKV(P) p.577; MKV(V) p.251.

EXAMINATION OF VIEWS

381

For che Buddha as wel as for Nagarjuna, the denial of a substancial encity is
based upoo empirical evidem::e, namdy, the perception of an individualiry consisting of the five aggregates. The five aggregates do not appear individually as
insulated discrete elementS. They appear asa lump (puflja) wh.ich retains its individuality (based upon dispos.hions) and cootinuicy (founded tipoo consciousness). In such a conte.xt he is caUed a person ( hoti satto ti sammuti, S
1.135; Tsa 45.5 [Taisho 2.327a]). In the context of a society, he becomes a
social being with relarionships to others (as son, facher , or daughter , mother.
etc.) In the context of mora behavior, he becomes good or bad (see the explanation of empirical self by William James, The Pnnciples of Psychology ,
vol.1,291-293). H owever, if the self wete to be considered different from che
perceivcd individuaity, it woud turo out to be a rather mysterious en city and
bence unknowable through the availabJe means of knowledge.
CandcakTrti, who pcefers a transcendentalist interpretation of Nagarjuna, .
gives .a negative explanacioo of what is very cleary a positive statement of
Nagarjuna.

8.

Evarp, nanya upadiinan na cop'iidiinam eva sa~

11tma nasty anupadiinal; n'iipi' niisty e!a nicayaf?.

Thus, he is neitber clifferent from grasping nor identical with it. A self
does not exist. Yet, it is not the case that a person who does not grasp
does not ex;ist. This much is certain.
M.KV(P) pp.577-578; MKV(V) p.252.

CandrakTrti, who favored a uailScendentaJist interpretation of Nagarjuna,

makes this a negative statement. The positive reading given above is justified
on ~o grounds. First, ,iQ the previous instances; Nagarjuna was clearJy asserting an empirically known ( = grhyeta) anupadiinaf? (that is, a person f.reed from
grasping), white at the same time rejecting an atman different from both grasp
ing and non-graspiog. Secondly, che posicive readiog is clearly justified by
KumarajTva's Chinese rendering of this verse.
.
As such, what Nagarjuna is denying is both jdentity (upadiinatn eva sal;)
and difference (anya upadiina'f!J). A rejection of metaphysica identity and difference does not necessarily tnean that Nagarjuna was abandoning an empiricaJ
personali ty. For Nagarjuoa, Janguage need not be necessarily metaphysical in
character.

382

9.

TuE PHILOSOPHY OP THE M IODlf

W/\ Y

Nlibhum afftam adhvanam ity etan nopapadyelte,

yo hijanmaru purve1u tato 'nyo na bhavaty aya?!J.

The statement, "I did not exist in che past," is not appropriate, for this
person is not different from whosoever existed in the previous lives.
M.KV(P) p .578: MKV(fl?p.252 .

This . theo is a elear deoial of absolute difference. Just as much as rhe statement, "I existed n the past," is noc appropciate so long as it is interpreted to
meao absolute identity, even so the statement, ''I did not exist in the past," is
not appropriate as long as it is cakcn to mean absolute differeoce.

10.

Yadi hy ayatrJ bhfl'ved anyal? pratyakhyayapi tafl'.J bhavel,

tathaiva Ca Ia Iaf!Zti;fhet Jatra jiiyeta 11amrtaf?.

If this person wece to be different [from that person in the previous existence] , then he wouJd come to be even forsaking tbat person. In that
case he would remain the same and, in such a context, an immortal
wouJd emcrge.
MKV(P) p. 579: MKV(fl? p.252.

Absolute difference impies absolute ideocity. Nagarjuna has alJ:eady shown

that "other-nature" (para-bhava) means the sclf-nature (svabhiiva) of aoother
(XV .3. s11ahha11af? parabhavasya parabhavo hi kathyate). Absolute identiry as
wel1 as absolute diffetence, trus involves complece iodependence, and as such ie
woud constitute immortality. Notbing is to be achieved. Kumarajlva rende.rs
the phrase tatra /iiyeta v'iimrtaf? as "The atman will be self- cau~ed."

11.

Ucchedaf? karmaaf!' nalJ! Jatha 'nyena krta-karmatJaf!J,

anyena paribhogaf? sjiid evam adiprasajyale.

(1f that were the case,] there would be annihilation and destruction of ac-

EXAMINAT!ON OP VIEWS

383

tions. This implies that the fruic of action performed by one will be experiencecl by another.
M.KV(P) p.580; M.KV(V) p.253 .

The verse has been reconstructed by Poussin on the basis of the Tibctan uanslauon.
H ow a metaphysicaJ coocepcion of difference would lead not only to a denial
of survival but also a repudiation of moral cespoosibility has bcen aJready
discussed by Nagarjuna (Chapter XII). There it was shown that he depcnded
upon a discourse of the Budclha co explain this problem. The presenc is simply a
rcstatement of chat argument.

12.

Napy abhutva samudbhu10 rlofo hy atra prasajyate,

krtako vii bhaved atmii sa1?J-bhu~o 'Vapy ahetukaJ?,

Yet, in that context, the error of assuming a n emergent without prior existe nce docs not follow. Eithcr the self would be caused or, if it has occurred, it woud be without a c:ause.
1HKV(P) p.580; MKV(V) p.253 .

The concept of self (atman) was posiced in order to account for the concinuity
in the human personali cy which could also explain mora responsibility .
However, the denial of an eternal self led to the opposite view that there is no
continuity, but annihilation (uccheda). The denial of an eteraal self and the
consequeot assenion of nnihilarion do not imply (na prasajyate) che further
metaphysical view.; chat- stu:ticthing chat did nor exist before comes to be
(abhutva sqmhhotz), which carne co be interpreted as the abhutva bhava utpada in the Sautrantika school (sec Kalupahana, Causality , p .p . 152254). In
other words, rhere is no implication hece chat existence is completely independent, withouc any prior connections. As such, if there wece to be any self
(atman), ir would be "made" (krtaka, see karaka ar XXVT. 10). lf it is not, and
if it were assumed co arise , the self would be causeless.

13.

Evaf!J dufir a.lite y'ii nabhum aham abhu1n aha'f!J,

ubhayaf!J- nobhayaf!J ceti 11a1!a sa1nupapadyate.

584

THE PHILOSOPHY OF THE Ml DOLE WAY

Thus, whatever view there is such as, "I existed in the past; I did not exist;
both or neither," is not really appropriate.
MKV(PJ p.581 ; MJ(V(V) p.253.

The views (dr!ft) referred to here a.re the metaphysical views relating to idencity.
difference , both or neithe.r . lt should be ooced chat the reasons for rejecting
these views are empirical. Empiricism, in the Buddha as well as in Nagarjuoa,
allows for the recognition of concinuity without having to posit absplute identity or absolute difference.

14.

Adh11any anagate ki~ nu bhavzyiimzti darana~,

na bhav#yami cety etad afitenadhvana sa1na/?.

A view such as "Will I exist in the future?" or "Will I not exist in the
futu1e?'1 is comparable to those associated with the past.
MKV(P) p .5.82; MKV(V) p.254.

Here again, the views that rue inappropriate are those i nvolving identity and
difference, not aoy explaoation of what the future could be based upon one's
experience of the process of dependent arising.

15.

Sa de11af? sa manu1yaf ced eva??J bhavati fiifvata??J,

anutpanna ca devaf? syajjiiyate na hi favatam..

If it is tJtought truit a human is the same as a divine being, then there

woud be .the eternal. H the divine being were to be non-arisen, th61l he
would not be born and that would consritute the eternal.

16.

Devad anyo ma11u1ya ced afiivatam .ato bha'llei,

devad anyo tn4nu1ya ced S4??Jlatir nopapadyate.

EXAMIN AT!ON OF V 1ew s

385

lf i is thought that a human is the same as a divine being, then there

wouJd be the non-eternal. lf it is thought that a human is different from
a divine being, chen continuicy is not appropriate.
MKV(P) p.583; AfKV(V,) p.254.

ln addjcion to the attainment of the uJtimate goaJ of life (para1nartha) which is

freedom (nirvaf!a) , the Buddha allowed for the possibiJity of a human being
reachiog up to the status of a dei cy or a divine being (deva). assumed to be one
who is macerially as well as morally superior to ordinary human beings, yet not
coming anywhere close to the ultimate goal (see S 1.228; Tsa 40.l [Taitho
2.290b]). H owever, the Buddha refused co recognize these divinc beings as
eternal and permanent enticies (D I. 20, sec Chang 14. l (Tais ho l.90b-c]).
The recognition of che above possibility could prompt the su bsrantialiscs co
discover ''some-thiog" (ki?(lcit) in the human being that is identical with the
divine. Nagatjuna is herc rejecting any such iden ciry, as well as aoy alternace
theory that coud make diviniry uncaused.
Funhermore, the emphasis on absolute difference would not only lead to the
belief in non-eteroalicy (aafvata), which woud imply a denial of continui ry in
or the process of becoming (Jtt'!11afl), the.lacter being distinguished from the
meraphysical process referred to carlier (see XVI!. 7-10).

17.

Divyo yady ekade.fa~ sjiid ekadesai ca 1nanu!a4,

aiaf!flta'!1 fa:fvata'T(I ca bhavet tac ca na yujyate.

If a part wece to be divine and the otber part to be human, then there
wouJd be both the eternal and the non-eter.na!, and this tod would not be
p ropcr.
MKV(P) p.584; MKV(V,) p.255

The combination of two metaphysical views do~s not prov1de for a noometaphysical one. The Buddha's rejection of SJJch views is clearly srated in the
Brah11iajala-1uJlanta (D 1.21) . Nagarjuoa is simply reiteratiog that posiLion
herc .

386

18.

THE

PHILOSOPHY OF THE

MIDDLE W'AY

A:fii1vala1!J 1af.vata1!J ca prasiddham ubhayar(J yadi,

siddhe na fiiivala'f!l kama??J nai11afiiJ11ata11i ity api.

Supposing both the eternal and the non-eteroa arc established, then it is
not possibe to cithcr assert the etcrnal or the non-etetna.
MKV(P) p.585; MKV(V) p .255.

Nigarjuna is hece refusicg to accept a conclusioo which is only a pan of a conjunctive proposition.

19.

Kul41cid agataJ? kafcit ki'f!lcid gt1eche1 puna!/ ku(Uit,

yadi ta.smad anadis tu sa'f!J.fiiral/ syan na casti sal/.

If anyonc has come from somewherc and apin werc to go somewherc,

tben the lifc~process would be beginningless. Such a situation does not

cxist.
MKV(P) p.585; MKV(V) p.255 .

The m.ctaphysics rejected in Chaptcr II is refccred to in the first line. The second line cmphasizes che ideas cxprcsscd in Chap ter XI .

20.

Niisti cec chaivatal/ kafcit ko bha11i/yaty ai7ivataf?,

fiifutTIO 'fiivataf capi dvabhyam abhfii'f!l tiraskrtaJ?.

If it is thought that thcrc is oothing ctemal, what is it that will be noncrernal, both ctecoal and non-eteroal, and also what is separated from
tbese two?
MKV(P) p .586; MKV(V) p.256.

Hcre again, we havc Nagarjuna's analysis of "something" (kaicit) , an entity

WMINATION OF VIEWS

387

that he was determioed to get dd of oo previous occasions (XXV. 19,20,24),

The theories of identity, differeoce, both or neither, are associated with such
metaphysical eotiti:es. Hence Nagarju.na's refusaJ to accept such theori~ing.

21.

Anlariiin yadi /ok.al; sjiit paralokarp. katht1rp. bh1111e1,

athapy anantavarp.f lok.alf para/ok.alf ka1haf?Z bhavet.

H rhe world were to have a Jimjt, how could there be another wo.rJd?
Furthermore, if the world were to have no limit, how cou1d thcre be
another wodd?
MKV(P) p.587; MKV(V) p.;?56

Having discussed the rnetaphysical views relating to dw:ation, i.e eternality.

etc., of the world and the self, Nagarjuna moves on to a discussion of. the
theories. relating to the extent of the world.
The rejection of the metaphysicaJ notions of the finite and the infinite are
bascd upon a recognicion of the possibilicy of a future world (para-lokt/). According t0 Nagarjuna's analysis, a fioite (anta) implies abseoce of continuity (uccherla), and as such militates against any conception of a future world explained in terms of dependent arising (pralttyasamutpiida). The infioite (ananta)
implies permanence or eternalicy (fiif11ata) aod, as such, it woud be meaningless to speak of a future world as an "other world" (pf/f(l-/oka), for it would be
identical with the previo.us world.

22.

Skandhaniim efa sartzfiino yasmad dipiitci1am iva,

pravartate tasman tiiintiinanta11att1111~ ca yujyate.

As this scries of aggregatcs proceeds along like a flame of a Ja.mp,

[spcculation about] its finitude or its infinitude is not proper.
MKV(P) p.587.; MKV(V) p.256.

The simile of the lamp (patlipa) was popu]ar a.moog the Buddha and his
disciplcs, especialy in their expanation of freedom (nibban11). As the flame of

388

THE PHILOSOPHY OF THE M IODLI! W AY

a lamp is extinguished , because of che absence of the condicions necessary for

ics concinuous burning, so is a person's miod freed when the fuel chat feeds ics
continuous burning with anxiecy is cxhausted . Explaining this process in rerms
of dependencc, it would be most inap_propriate to speak of ics finfrude or infinicude, independent of any condicions that arc involved in its conrinuicy or
lack of continuity. This is a elear indicacioo chat an absolutiscic conception is iocompatible with an explanation based upon conditionalicy.

23.

PUrvc yadi ca bhajyerann utpadyeran

na capy ami,
skandhaf? skandhan prafityeman atha loko 'ntaiiiin bha11e1.

If the prior aggregates were to pe destroyed and rhese aggregates were also
not to arise depending upon these other [aggregates], then the wodd
woud be finice.

24. Purve yadi na bhajyerann utpadyeran na capy amT,

skandhaf? skandhan prafityemiin loko 'nanto bhaved atha,

If the prior aggregates were not to be destroyed and these aggregates were
aso

not to asc dcpcnding upon these other [aggregates], then the world
would be infinite.
MKV(P) p .588; MKV(V) p.256.

The above is a criticism of an explanation of the conrinuity in the aggregaces

which does not takc into consideracion thcir causa! depeodence. l f the aggregaces were looked upon as things that appcar and disappear with no causa!
relations, then only can one speak of a finite world. In other words, the cooception of a fin ite world involves one of the exueme views discussed earlier, namely, aonihilation (uccherla) .
The non-arising of the aggregates dependent upon other aggregates would
thcn imply permanencc, and this is what is involved when one speaks of an infini te world.

EXAMJNATION OF VJE\'(/S

25.

389

Antavan ekadea ced ekadeas tv anantavan,

syad antavan ananta ca lokas tac ca na yujyate.

lf the world we.re to be pa~dy finice and also pardy infinite, it would be
both finite and infinite, and this too is also not proper.
'

'

MKV(P) p .589; MKV(V) p.256 .

The rejection of both the finite and the infinite, the etetnal as well as the noneternal, as constitucing the reality, was prompted by the Buddha's refusal to
recognize two djfferent levels of reality: a changing and finite world contrasted
with an eternal and infinite ultimate reality or an Absollte. As such, neither it:i
the Buddha's philosophy, nor in Nagarjuna's thinking is thete any room for an
Absolute of any sort.

26.

Katha1(l tavad upadiitur ekadeo vinank,yate,

na nakfyate caikadefaf? eva~ caitan na yujyate.

How can it be possible that one part of a gr:asper is dest!oyed and the
othetpart is not destroyed. This too is not proper.

27.

Uptidiinaikadea

ca katha1(l n'iima vinank1yate,

na nakfyate caikadeo naitad apy upapadyate.

How can it be possible that one part of grasping is destroyed and another
part js not destroyed, This too is not appropri.ate.
MKV{P) pp.589 -~90 ; MKV(V) p.257 .

Nagacjuna leaves no room for the recognitioo of an ecernal and absolute entity
either in the person grasping (up'iidiitr) or in grasping (upadiina).

390

28 .

THE PHILOSOPHY 01' TH..E MIDDLE WAY

Antavac capy anantaf?'l ca prasiddham ubhayaf?'l yadi,

siddhe naivantavat kctmanJ naivanantavad ity apz:

Supposing both the finite and the infinite are e5tabiished, then it is not
possible to assert either the finite or the infinite.
MKV(P) p .590; MKV(Ji? p.257.

This rejeccion of the finitc and the infinite is comparabJe to the rejection of the
eternal and the non-cternal (XXVIl.18).

29.

Athava sarva-bhavananJ unyatvac chavatadayaf?,

kva kasya ~atamal? ka.rmat sambhavi,yanti t/ufayaf?.

Thus, because of the emptiness of all existenr.s, wbere, to whom, which

and for what reason vtews sucb as the etetnal could ever occur?
MKV(P) p. 591; MKV(V) p.258.

The recognition of depeodencly arisen phenomeoa (prafityasamutpanna dharma) meaos the acccpdnce of the non-substantiality or emptioess of all these
things. If things are non-substantial, how can thcrc be views about the eternal
and the non-ete,r nal , the finite and the infinite? The answer to this question
has been provided by the Buddh.a and clearly .rcstated by Nigarjuna. Tt is the
resut of an urge on the part of human beings to find absolute answers to ques.
tioos such as "Where did 1 come from?'' or ''Wherc do I go fcom herc?" These
qucstioos wouJd be raised and attempts to answer them be made so Jong as
human beings arc propeUed by a "craving for becoming" (bhava-tr1'!ti) o~
"craving for other-becoming" (vibhava-tu'!ii) . This was the riddle of human
existence faced by the Buddha. Understanding that riddJe of existence, the
Buddha attained freedotn (nirva11a) by spewing out craving (tr11iii)~ and
abandooing any grasping (upadiina) .

30.

Sarva-drtfi-pr11ha1faya yaf? saddharmam adeayat,

anukampa"' upadiiya ta1{'J naffl4Syami gautama~.

EXAMTNATION OF VtBWS

391

I revcrcntly bow to Gautama who, out of compassion, bas taught the truc
doctne in order to rclinquish all vicws.
MKV(P) p.592; MKV(V) p .258.

Th.is fina statemen't of Nagarjuna referring to the Buddha's.preachiog of the

doctrine (sad-dharma) is to be contrasted with his statement at the end of the
chaptcr on freedom (ntrviifa)(XXV.24) , whcre he says that the Buddha did not
"preach anything as .. ." (na ... kalcid dharmo buddhena deiital/). and
should open the cyes of those who stick to a completdy ncgative 'interprctation
of Nagatjuna. Nagarjuna's rernarks clearly show that he was aware thar the
Buddha did not spcak "metaphysically'' but only "empiricalJy ."

INDEX TO THE KARIKA

NOMlNAL FORMS
ag ni (firc) X. I, 6-9. 12- 1); nirogniAa X .9
ag111 (beginning) Xl.1. 2
11;1kura (sproul) XVIL 7
alj11ni (proposi tion) XXV. l 5
alita (pasc) XlX. 1-3; XX. 12; XXVll. l, 3, 9,
13- 14

adhama (low) XDC.4

adh1ga1t1 (attaincd, realized) XXJV.28
adh1la-ya (obscssion) XXIV. 13
11dhya1ma (imcrnaJ) :XVlll.4
adh111111 (time, pcod) XXVll.1-3, 9. 14
fl'!(ll!aragrtJ ( without prior end) Xl. l
anagata (future) XJX. 1- 3; XXVll 2, 14
anukampa (compusion. sympathy) X:XVJl.30
anu.van.11/a (excolkd) XVll. 13
anta (end, limit, finite) XXll.12; X:XV .21, 22;
imania XXJJ.12; XXV.22; X:XVJl.24
anantaval XXV.22; XXVll.21, 22, 25 .
28
amaval XXV.22 ; XXVIJ. 21- 23, 25 , 28
11paran111XXV.21; XXVIl .2
purvanta XXV.21 : xxv11.1
antanJ (becwccn, mcdiatc) XXV.20
atJantara (immedia te) 1.2, 9
1mdhakara (darknc$S) Vll.9
a,11ya (orbc1 , anothc1, differcnc) 11.8. 15. 18 ,
20. 23; VII . 19, 28; IX.6, 9: X. l , 5-7;
XJl.3: XIV.3. 5-8; XVLl.4, 28, 31. 32:
XVlll. l, IO: XXII.I: XXV.23:
XXVll .7-1 1, 16
QflJOfQ X. 13
a11onya (mutU:il, togctbcr) X.7: XIV. I;

XXl.6
a11y11111a (differtncc) XlV. 3-4. 7: X.X, t9;
XXIl.8
a11yathiit11a (change) .XV.9
anyathiibhiiva (change) Xlll.;}- 6; XV.5. 8
apalu1rfant; (climinacion) XVL. 10

apara (postcrjor) XJ. 2, 6

apekfii (cootingcnt, comingencc) XX. I I
nir- (non-) Vl.3; X.3; XV.2
apekfita11ya X .1 O
abhidh"iilavya (ro be de$ignated) :XVIJI. 7
abhin11J1a (mountcd) XXIV 15
abhutva (without prior exlsrencc) XXVU. t2
abhfiigama (bcing con!rontcd) :XVll.2')
abhyupapanna (cngr<>Mcc) XXl.14
amrt11 (immonal) Xl.3: xxvn_.10
11rcifilf1.1 (fl:unc) xxvn.22
artha (fruit, cffcct, purposc, mc:injng) 1.6;
IV.2; Vl. 7-8; VUl.6; X . 16: Xl.7

ekarJha (one meaning, identity)

Dedi~tory Vcrscs; XVIJl.11
'1a11ar1h11 (vllicry of mc:anings,
difference) Ocdicatory Vccses;
XVJU. l t
nairarthakyfl (futility) Vlll.6
paramiirtha (ultimnte fruit) X:XlV.8, 10
prajfJaptyartha (purposc of dcslgnatibn)

XXII. 11
11aryar1hya (meaningless) X.2-3
alarrz (adcquarc) Vll .2
a11agiihalif!1, dur- (difficulty of undcrscanding) XXIV.17. /
a11ara (end) Xl.1. 2
a11as1ha (state) Vll.28; XXIV.38;
avaslho XVJ.6
at1ava11hi, ana11a11hiti (infinite regress)
Vll.3. 19

aJva (horse) X:XfV. 15

a.uan;gala (rc:ichcd it's end) X:X.10
as/iii ("cxists'') XV. IO
flOJfiti ("docs nm cKist") XV.], 10-1 l
asti ni11fili lX. 12
a11i1va (existcnce) V.8: XV 8, 10
XXlll.3: XXV.14
11ii11i111a (non-existcntc) V.8; 10: XXll.3
nau11'ii XV.8

303

394

TuE PHILOSOPHY OF THE MIDDLE WAY

aham iii ("l am") XVlll.4

nirahaf!'Kartz (free from cgoism)

XVUl.2, 3

iiuJa (space) V.l, 7

iiRhyiila (explaincd) Vll. 14
iig11111 (come) XXVll. 19

71gl1fl'J4, an-(noo-appcarancc) DcdiC3co.iy Vcrscs

aJ11t11!1/tMbliiiva (oanuc of coming and going)

XXV.9
iltma (self, form) IX.9; X.15, 16; XVlll. l-2, 6;
XX.5; XXlll. 3. 21- 22; XXVIl.4- 8, 12
llniltma (oc>sclf) XVIll.6; XX23: XXIJ.3;

XXlll.22
111mal.r:111 (self..caused) Xll.8
i1ma-111f{Jy111'fl!lka (sclf-n:sttaint) XVll.1
ilmanly11 XXJV. I 5

xvm.

iilm1J11
2
p4iiil1TNI (anothcr self) vn. 25. 32
svilma (iaelf) m .2; Vll.13, 25. 32; :XX.24
$V11-pri1ma VIJ.8, 12

iitli (bcginning) Xl. l. 4

tmiitli (bcginn.inglcs.s) XXVII.19
iidhip111eyy11 (dominancc) I.2
iytilli1111 (sphercs of scruc) XIV.2; XVl.2
!4ff (six) XXVl. 3
ifmhdha (initilued)
1111 (non) XXIV.37
4111mbh4 (commencemcnt) ll.13-14: X.2-3
iilmnbtm4 (objcctivc ruppon) 1.2, 8
iiloh (light) XXV.14
i.fr111111 (inllux)
1111 XVIl.19
14- XVll.19

mdriya (faculty)
!Ilf/ (sile) m.1
Urtifuna (fucl) X.I. 4, ~9. U-15
idhm;ina (buming) X.4

uuheda (annihibtioo) :>W.11; XVll.20;

XXI.14-17; XXVll.11
an- (non) Dcdicatory Vcr.scs; XVlll. I I

ucchee/a.rillTfana XV.10; XXJ.14

ucchinna (annilulatcd) XVIl.8, 10:
XVIll.10
t111- (non-) XXV.3
1111i1ma (highest) XIX.4
11//(lm (follc>winu\
XI.3

.utpitla(arising) VIJ.1- 5, 8, 13, 18-19, 25. 29,

32-34; XXJV.17
t111 (non-) Ded.IC3tory Vc.iscs; XVUI.12
mulotpiitla (primary arising) VIJ.4, 5
.utpalli (arising) VD.15-16, 20-2 I
o-p11nar-11tpt111i (abscncc of rcpcatrd
arising) XXI .16
u1ptllly11mii174 (prescndy arising) VU. 7, 10,
14-16, 18

I. I; Vll.14; XX. 10
tm- (non-) l.9; Vll. IJ..: 14, 17, 22; XVU.21;
XVlll.7; XX.17- 18; XXlll.9: XXY.3
im-.uj>Qpatti (ab.sense of arising)) V.4;
XX.22
11p11J)iltli1t1 VU..20
uihy11 (uprising) XVJ. 5; XVIll. l: XXL 15-16;
XXN.l, 20; XXV. l, 2
uihyo-~111f!114na XXI. 1~16
udiihrto (cxamplified) Vll.34
utibhuti (occuuencc) Vl.3
1uibht11111 (uprising) XVU. to; XXl .2
upfllllmbha (objcct) XXV.24
uplli4"111 (appeasement)
uplllllmbhopaiama (appcasemcm of
objcct) XXV.24
tl171!{1111yop11ftlma (appeasement of
objcct) V.8
pr11paflcopaJam4 (appeasement of
obse.ssions) Dedicarory Verses; XXV.24
11ptlf11111a (appcased) XXIIJ. 15
.upiidiino (grasping) 111.8; Vlll. 13; X. U ;
XVl.3 , 6; XVllJ.4; XXII.7-10; X.XVl.~7 ;
XXVll.4- 8, 27
""" (non) XVl.3, 6, 9; XXVI. 7
nir-(witho11~) XXII. 7
upadaua (sphcrc of grasping) XXVl.6
:t1'(spherc of non) XXJJ.7
11pidii1r: (graspcr) XXlf.10; X:XVL 7;
XXVll.6, 26
upiyt1 (mcans) XVI!. I I
upiiyisa (dispair) XXVI. 9
upiiltlmbha (ccnsurc) JV .9
up4/abtiha (ccnsurcd), a1'111pilabdh11
(unccnsurcd) IV .~.
urdhva (abovc) IX. 12
11/jxmfltl (arisco)

r'!a (dcbt) XVCL 14

rli! (without) 11.4 1 20; X. 1; XIV.5, 6:

XVI!. 7, 9; XIX.6; XXU.4

XV.4;

lNDEX

fdrfhiIamp11tliJ (psycho-kioetic power)

XVll.31
r1i (3.Scciic)
parama- (supremc-) XVll.2

eka (one)
ek.al11a (idcntiry) VI.4 , 5; Vll.30; X . I ;
XIX.4 ; X:X.20
ek"tirtha (one meaning), an ekanha
XVlll. 11

eklbh1i11P (ideociry) 11. 19, 21

aika (identical) XXI. JO
aika111f? (co-exis1) Vlll. 7
aikahla (identity) XX.1 9
aikya (identity) XX.20
karma (action) 11.19: Vlll. l-3; 9-13; X. l ;
XVll.2-3, 5-6. 10, 14-18, 20-23, 25-27,
29-30, 32-30; XVlll.5; XXVI.1 ; XXVII.11
-ja (bom of action) XVIl.30

-patha (path of-) XVI!. I I

-11aa'ha (destruccion of-) XVll.16
a-karmaka (withouc action) Vlll.2; X .2
kamtJa (performance of action) XXN. 17
k.arta11ya (to be donc) XXJV.33. 37
kattr (agent) II. 19; VUJ.2-4, 8, 13; X. I;
XVIl.28-30. 32-33; XXIV.17
kartrka, li (withoUl agem) VUl.2

karaka (agent) Vlll.l, 3. 7, 9- 12;

XXIV.37 ; X.XVI. IO

ak11r111if!a (noo-accing) XXIV .37

kfira'!a (causc, sufficiem condicioo)
IV. 1-4, 6: Xl.7: XXIV.17
ni[- (withOUI) lV. 5
karya (effecr) V .6; Vlll.4; Xl. 7; Xll. l;

XXIV. 17
a-kiiryaka (in-effcn-ivc) IV. 3
kr1a (donc, causcd) Xll. 1-4. 7-9:
XVll.24, 32; XXVIl.5, Il
a- (not) XVIJ ,22, 23
krta.k.a (made) XV 1-2; XVIL23;

XXVll.12
J.ririma, 11- (unmade) XV,2
kriyfi (activity) 1.4; Vlll.2, 4-6;
XXIV . 17, 37
kalpa (fabricarion) XXll.9
kalpan?i (thoughr) IX,12, XVU.12, 13
11i-k.alpa (discrimination) IV.5

11U11;/pa1af? (one who disc.riminates)

395

XVlll.5

11lr-11ikalpa (without discrimination)

XVlll.9
11ikalpaya~ (discrimioatjng) XXII. H
!ll"J?lkalpa {thought) xxm :1
Kiityfiyanilvavada (Admonition 10
Katyliyana) XV.7
klima (plcasuce, dcsirc)
KUf!ll (strands of-) XVU.11
kfiyika (bodily) XVI13
ki/a(timc)XVll.6; XIX . I, 3, 5-6; XXl.21
ella- (comcmporaneous) XX. 7
111/ya- (nmc) XXJ.3
nirv1ir/a (-of frccdom) XXI. t7
kfl(.:utha (remaioing immutabe) XXJV. 38
ke11ala (cxclusivcly, cntirely. otll) VU.4; X.l.8;
Xll.10; XlV.4: XXlll.8: XXVI.9, 12
koti (cxtrcmiry, end) XI. I, 8; XXV.20
krania (mcthod, mode) fV.7; VI.2; X . n ;

XVI. I; XIX.4
puniaparasaha-krama XI.2. 6
klei (defilemcor, impurity) XVIl.26-27, 33;
XVU.5: XVlll.5; XXJll.2-6, 24-25
k/11fii1maka (defiling natwe) XVll.26, 27

-praha'!a (diminatioo, relinquishing)

XXIV.39

klifta (ddiled) XXJll.5

R,t:ya (waning) XVIU .4-), 12; XXI. 7
a - (not) XXl. 7
k[lr111 (waned) XVIJl.4
ki'ira (milk) XIIl ,6
gamana ( movemcm) 11.3-7, 9-11, 13-14 ,
16-20, 24-25
a- (non-) 11.3

gamyam1ina (presenr moving) 11. 1- 5,

12- 14, 17; ll.3; Vll.14; X. 13: XVI.7
gala (mo~ed) Jl . I- . 12-14, 17; lll .3:
Vll. 14: X.13 ; XVl. 7
fi (un) 11.1-2, 12-14, 17; 111.3; Vll. 14:
X. 13; XVl.7
gati (morion) 11.2, 4, 17-18, 20, 22-23,
25; XXVI .J-2
gat11411Jll (space to be: movcd) li.
gani(' (movec) 11.6-12, 15-16, 18- 20,
22- 25
a- (non-) 11.8. 15 , 20
gandha (smell) XXJJI . 7-8
gandharva11agara (city of the gandha.rva)

396

THE PHlLOSOPHY OF THE MIDDLE WAY

Vll.34: XVlt33: XXIU.8

gambhlra (profound) XX.IV .9
grhita (grasped) XXll t l3: XXlll.1.5
a- (non-) XIX ..5
d11r- (wrongly-) XXIV+li
gocara (sphcre) IfJ, l
c1iia- (-of rhought) XVIII. 7
Gautama XXVIl.30
graha (grasping, hold) XVI.9
a- (non-) XVI.9
griiha (grasping) XXJJ. l3 ; XX.ID. 13-16
ghatan (conrcnd) XXJV.32
ghana (substanda) XXII 13
ghfii11a (smdling) 111.1, 9

ca,t!u(cye) 111.7; X..XVJ.4-3

ttJrama (ase) XXI.18-19
ca1tirt1tdhya (fourfold mcthod) Xll. IO
ciita (thought) IV.7: XVll.9-10: XXIV.12
-gocara {spherc of-) XVIII. 7
-IDf?11'iina ( -scrics) XVU. 9
cettna/; (from thought) XVIJ .9
c1#ta (rhough1) XXli , 14
cet11na (voitioo) XVll.2,3,5
ce1fii ( efforc) li. 2
chedana (cuning) Vll.31
/agat (u nivcrsc) XX0 . 16; XXJV .38
janma (birth) XVfil.4; XXl.2-3. 5
punar (re-) XX.9
purva- (prcvious-) XXVIL3, 9
janaka (producer) XX. 7, 20
a-jamia (not produccd) Vll.5
janya (co~ produced) XX.10
janyamona, a- {not produdng) XX.22
jiit11 (bom) Vll.13; XX .6, 12- 14
a (un-, fururc) VU.26; Xl.4; XX.12- 14;
X:Xlll.19; XXJV.38
Jtili (blrch) Xl.3-6; X.XVl.8-9
jifyamana (being boro) XI.~; XXI.19-20
Jana (peopfe) XV.5
jrmt11 (scmient being) XVJI. 28
jara (dccay. -age) Vll.24: Xl.3-6; X.XV 4;
XXVJ.8
jit1a (vocwrlous one) XIII .8
J1rf!a (aged) XIll,5

jfliina (wisdom) XVlll.12; XXVl.11

(icsclf. 1ruth, idemi1y) XV.6; XV!lf.9;

XXII.8; XXIY.9
ta1111ala~ ("in icself") XVI!.26; XXlll. 2
/llttva-tlariana (perccpcioo of truth)
X.XVI.IO
lathya (sucb) xvnts
a-( not-) XVl!I.8
1111hiiga1aXX.Jl.1 , 3-8, IO , 13, 15-16
lama (darkness) Vll.9-12~ XXV.14
t1roJkr1a (scparaced) VJ .1; X .7; XXVJJ.20
ti!fhrJn (endu ring) XX. LO
l1[!hamona Vll.22; XXV.18
111/ya (cqual. same) XX.20; XXl.3
trfT{ii (craving) XVll.28; XXVl.6
f(!Jamana XXVI.6
la1J1111

darfan11 (secing. pcrccption) 111.1-6, 8-9;

l.X. 1-4, 6-7,
11-

1~12;

XIV.l

(not-) 111.5

darana (view,

mctaphysical view)

XXI. 14; XXVH; 14

du!a, dur- (wrongly perceived) XXJV. l l
dr!!i (view) XIH.8; XXVIL l-2 , 13.
29- 30; s11akaya XXllf.5
drfyamiina, a- (nor being perccived)
11.14

dra.ff( (scer) IJf.5-6; JX.8-9; XIV. l

41-a,rtavya (objcct of seefng) 111.l, 6, 8;
XJV.l,

3-4

dra(avyopt11ama (sec

upalam.a)

tlahana (burning), '1'" X .5

dipa (lamp) VJl.9; XXV!l.22
duqkha (suffedng. unsacisfacrorin~) Xll. J.
3-10; XXlll.22; XXIV.21 , 23, 25, 39-40:
XXVI.8- 9, 12
dofanta (example) 111.3
dew (divine being) XXVII .15-16
divya (clivioc) XXVIl. l7

defrma (1caching) XXIV.8

defita (caughc) XVll.20; XVJU.6;

x:xv. 24
dea (body) XVI!. 27. H
do1a (error, faulr) VJil.8; XVU. 12. 16.
23; XXJV.1.5; XXVll.12
do!a-prasai1ga XXlV, U
daurmanasya (dejccilon ) XXVI. 9
dviialJ (pairs) XIV. L
d11c1a (hacred) XXOI. l, 7. 12

397

INDEX

dharma (ching, nature) I. 7-9; XVll.5;

XXlV.19; X.XV.24
/ariimara'!a (nacurc of dccay and dca,th)
Vll.24
drffa- ( druaJanman, prcscm life)
XVll.18
fi ( distinct chings) V ,6
1arva- (all) Vl.10: Vll.29: X:XV.22
dharma (good) Vlll.5: XVU . l , il ; XXIV.6,
33-;5
a (bad) Vlll.5; XXIV.6, 33-35
dharma (doctrinc) XVll.20; XXJV.5, 12
sad- (true-) XXlV.4, 30; XXVU.30
-deiana XXJV.8
dharma!? (conccpts) XXUI. 7-8
dharma/ii. (nature of things) XVUl.7
dharmin, udaya-vyaya- (nature of arising
and ccasiog) XVI. 5
dhatu (element) XVl.2
panca (fivc-) V.7
far! XVll.17

mruddha (ceascd) VlJ.26: XVll.6;

XX.5-6; 10; XXl.18; XXUl. 23;
X.XV!. I I
a- (not, oon-) Vll.26; >.'VIIJ. 7;
XX.17-18; XXJ . 18; XXIV. 38; XXV.3
nirudhyamiina (ccasing) Vll.21.2},26;
XXl.1 9-20
a- (oon-) Vll .21, 23
'!irgama, 11 ( non-disappcar'lnct) Dcdic-.uory
Vcrscs
nirmila (crcared) XVI!. 31, 32
m'rmitaka XVJl.31
nirmuakakiira (crcatcd form) XVll. 32
nirmimita XVU.31
nirmukta (distinct from) IV. I, 2; V.5; X.X.8
11i- IV. 3
' (producing) I. 7
m"rllartaka
niffla'!a (frecdom, cessation) XVl.4, 9-10;
XVlll. 7; XXl.17; XXIV. 10, XXV. l- l6;

naya (mcthod, modc), a/mam- XVlll.2

rfi1nli (differcnt) XXl. 10
-bhli11a (differcncc) li. 21
nanlirtha, a (without a varicty of
mcaoings) XVlll.9
nimariipa (psychophysicaJ pcisooaliry)
XXVl 2-4
nilia (dcstrutcion) XX.VII. I I
''Offa (dcnroycd " past) X:X.14
riii1t1ifi (non-cicistcncc) XV.8
na1ffti ("does not exiJt'') XV.10-11 ;
X:Xll 13
flitya (pcrmancm. etcmal) X.2-3; XVI. I ;
XXl.14; XXIU.13- 14, 22

fli11rta (cnvclopcd) XX.VI. l

11ir1Jr1ta (rcnounccd) rx.12
11ivr11a (cc:asc:d) XVIII. 7
niJcaya (cercain) XXVll.8
nl!panna (accomplishcd) X.8
flil/1arana (1clinquishing) Xlll.8

a- X:Vl . 1; XXl. 14; XXlll.13- 14, 22:

XXJV.21
nityalii (pcrmancncc) XVll.6
a- (impcrmancncc) )QQ.4

nimitta, dharmiidharma (occasiohccl by good

and bad XXJV .34-n
mroa'ha (cessation, ccasing) 1.9; Vll.27,
29-32; XXJV. 17, 23. 25. 40; XXV. 1- 2;
XXVJ.11- 12
a- (non) Dcdicatory Vcrscs
parr11~ nirodhad (aftcr death) XXll.14;
XXV. l 7, 21
nirodhana (ccssacion) XXI II .23

19-21
a- (that wbich docs noi ccase) X.5
nmtrkl ( cc:asc:d) XX1 I. n

pa)a
pa~

(view, position) li. 10

(cloth) X .15

patlrrl (promissory notc) xvn.14

para (ocher) 111.2; V.7; Xll.3, 7-8
pain11grii/xJka (bcncfittiog ochcis)
XVll. l
-kara (ca.using by-) Xll.9
-krta (Cllused by-) X.Il. I , 3, 7- 8
-pudgala (-person) Xll. )-6
pamta~ (from anothcr) I.I ; XXl.1 3
paratra (dscwhcrc) X.3; XV.2
parabhava (othcr-oacurc) I.3: XV. 3-4 , 6:
XXll.2- 4, 9

paraspara (mucually) Vl.3

paramartha (ultimate ftuit) XXIV.8, 10
pankTr1i1a (cxpounded) XVIl.2
parig(h'!ata4 (one who upholds) XXIV.28
parljilii (undcrst~nding) XXIV.2. 27
panjfl'iia (under$tanding) XXIV .26
11 XXJV . 26

398

THE

PH!WSO!'HY OF THE MIDOLB WAY

paniliptda (1lluminating) Xlll.1

paride11ana, sa (Jamcncation) XXVl.8
pan/iiitaytm (attributing) XX'JV.15
paribhoga (experience, enjoymcnt) XVll.5;
X.XVU. ll
padvartal:a (1clacc:d) XIX,4
panhilra (rcfut:u.ion) TV.8
p arihrta, a- (unrefutcd) IV .8
parya11os1hina (adhering) XXIV.23
paryi1pta (adequa1c) lll.3
pa!ciiJ (posterior} XI.4
pafcima (cod) XI. l
ptiiya"1iina, 11- (noc pecceiving) 111.4
pika (m11.turing) XVIl.6
p utJya (merle) XVJL5 , 24

a XVII.5
putra (son) lU. 7
pudgala (pct!lon) XII.4-6; XVl. 2
puru1a XXTV .4, 29
purufa (man) X6
-pudgala XXJV.4. 29
purra (p rior) Jl.13; V. 1; Vl. t ; l'X.6-8; X.8;
XJ.1-3, 6, 8; XV. Il ; XVl.7; .XX.8-9;
XXVU .2)-24
PrthaJ: (discrctc. scparacc:) Vl.8; X . l6
lllfJ (discrcrcnC3s) Vl.4-7; XX.20
, -bhiiva (discretcness) VJ.6-7, 9
prakaia (illuminacion) Vll.9
prakrti (pramal narure) XV.8- 9; XVll.14
pr.1j1JajJl i (eommunicatio fl, &on 11en1ion)

XXJ/.l/

upiidiiya (dcpcndcn~) XXlV.18

pmtipa,, madhy@iii (mi6de path) XXlV. 18
pra1ipanflaka (rcachcd the way) XXlV. 3, 29
pmtibiidhatal, (one who concmdicts)
XXIV.22, 37
pralibimba (m irror image) XXIU.9
prat1u1thitP (sicuatcd) Vll .9
prPtisandhi (moment of conccp1ion) XVll. 17
prati!iddhP ( rcfu1ed) V. 7
prafi1yosamulpiida (depcndcnt u ising)
Dedica1ory Vcrscs; X.XIV . 18, 36, 40
pmlJllJfJ (condi1ion) l. 3, 5- 6 . !>. 11- 14;
XV. I; XVJJ ,27 . 29; XX, 1- 4. 8, 23;
X:XVl.2 , 6
a- (0011-) 1.5, 12, 14; XVll.29
t1/JPrt1 (indcpendcmly reaJized) XVJJI. 9
nhZJll (made of-) 1.1 3-14
-siimagn (harmony of-) XX.24

-11.zff (constinucd of-)

1.4

&alfliiml? (four) I. 2
pratyikhyiiya (fot!lak.ing) XXVll.10
pralyulua (ccfuted) lll. 3
pratyulpanna (prcscnt) XlX. 1- 3
prPtyudiivrtta ( cccoilcd) XXJV.12
pradipa (lighr) Vll.8-12
pradipta (aflamc) X.2- 3
propaflca (obscssion) XVIU.) , 9
-alila (gone beyond-) XXIl.15
-hata (impaired by-) XXll. 15
-opaama

(appe:tScment of) XXV.24

pmpancita, a- (unobsesscd) XV!ll. 9
prt1brti (bcgins wich) XV!l .7
prabaVtJ (sou rtc} XXI!l. I
pmyojana (purpose) xxrv.7
pravrlli (occuucoce) V.3
pnzfalua (implied) li. 5-6
pmsPnga (accompanying, following) XXIV.13

praiadhi1a, tiul (wrongly cultiva1cd)

XXIV .1I
prasiddha (cstablished) XXVl!.18, 28
pra1iddhi (cscablishmcnt) IJl.3; Vl.8-9
a ( non) VTI.33; XJl.7 ; XV. ~
prahli'!a (relinqu.ishiog) XVLl.15-16; XXlV.2,
21, 39; xxv.1- 2, 10; xxvu .~o
prah1'1a, P (unrclinquishcd) XXV.3
pr/Jh11ya (10 bt rdinquisncd) XVll.15-16
piiik (pl'ior) IX. 1- 3. 12
p raptP, a- (nor rc:ichcd) X. 5
prip1; (ceaching) X.XI V. 39

baddba (bound) XVl.8

bandhP (bondagc)_ XVl.8
bandbana XVl.6-7
bahirrihli ( cxternal) XVlll .4

bahu (many) xvn.12

lllP (pluraliry) IX.9
bahya (cxccrnal, Olhcr) XJJ .10
b/jP (sccd) XVIl. l, 7-8
b11ddh1Z XVll.I;, 20, XVlll.6, 8. 11; XXIJ.2.
L4- 15; XXlV.5, 8, 30-31 ; XXV.24
fiisana (Buddha's mcssagc) XV .6;

XXlV,9
a- (unenllghtcncd) XXJV 32
pmtyeliP (self-enlighic.-ned ) XVll. 13;
XVlll.12

sarn

(pcoctly

cnligh1enc.-d), Dc:dica1ory
Vcrscs-; XVIII. 12

399

INDEX

b1uldhi, alpa- (meager imdligcncc) V.8

bodhi (cnlightenment) XXIV.31- 32
bodhilatllla, -caryo (c21ccr of a bodhisanva)
XXIV.32

broh111acarya1 a- (ignoble life) XVll .23

bruv'iiTfa (speaking} xxrv

bhagaviin (blcsscd one) Xlll. 1- 2; XV. 7;

XXV.1 7-18

bhanga (dcstruction) Vll.3. 33- 34

bha11a (becoming) XXI.18; XXV. 10;
XXVl.7- 8
SllfJJlllli (sueam of-) XXl. 17, 21

vi- (other-, dissolution) XVl.3; XXI.1-5,

7- ll; XXV.10
punar- (re-) XXVl. l

bhava (cxistcnt, existence) l. l , 3, JO; IV.7;

V.2. S-8; VIL ,17, 21, 23-24. 27, 30-31;
Vlll. B; IX.2-3; X . 10-11, 16; XJ.8;
XJl.lO; Xlll.3; XV.4-7 ; XJX,6: XXl.4, 8,
12- 16; XXJll.20; XXIV. 16; X:XV.4-7,
10-16; XXVII .29
a (non, abscncc) V. 6- 7. 30; Vlil.4:
XV .3 , 5-7; XXl. 12; XXIV. l-4, 20,
29-30; XXV.8. 10-16
bhavaniJ (cultivation) XVll. l~ ; XXIV.2, 24.
27: XXVl.11
bhuta (clc:mc:nrs) lX.10
bhuta (come to be) XXlll. 24
a- XX!ll.25
bheda (distinct v:uicties) XVIJ.2
bhoktr (cxpericnccr) XVIJ.28, 30

rnadhya (middle) XJ.2

marlhyama XJX.4; XXIV. 18
'11111111 {mind) m.1, 9
mar1ufy11 (human) XXVU. l}- 16
mil#Ufa XXVll.17
mam/a (me:igcr incc:Wgcncc) XXIV. 12
-medhasa XXJV. 11
mama
mamefi ("mine") XVlll.4
nlr- (free from sdf1Shne.ss) XVUl ,2-3
ma!'tlf!11 (death) Vll.24; Xl.3- 6: XVl1:19;
XXl.2- 3, 5; XXV.4; XXVl.8
maffci (miragc) XVll.33: XX!Jl.8
mahan111 (grcat) XVII. 12
miiliipi1arau {mothcr and father) 111.7
miinaJll (mcnal) XVll.3

marga (path) VID.6; XXJV.24-25; 40

miiy'ii (ilusion) Vll.34; XXlll.9
milhyl (wrong) xxrn.16
mucyamiin11 (frccing) XVl.8
muni (sagc) XXIV. 12
maha- (great-) Xl.1
mu/4 (sourcc) XXVl. 10
mrgyamilf!11 (sought for) XVl.2; XXU.8
mrfil (ddusion) Xlll.1-2
medha1 (inrcUigcncc) XXJV. I I
maitro (fricndly) XVll. l
'
rno/qa (rtlca.$c) Vlll.6; XVlll.); XXV.11
mo1f11'!a XVl.8
mofa-dh11rma (dcccptivc nacurc) Xlll. l- 2
moha (confwfoo) XXI.Il; XXXIJl.l. 7

11111u/11 (prima.cy) VII.4-6

yukta (propcr) l.9; Vl.;; VII. I~ XV, I ; XXl.21
yugapad (sirnulcaneous) XVl .8
yu11a (young) Xlll. 5
rakla (lustful) Vl .1-3. 6-7, 10; XIV.2

ra1Jj11n1ya ( objcct of Lust) XIV. 2

flllna (jcw(ll)

XXJV .5

rasa (tascc) XXJll. 7-8

flUl11711 Ili. l , 9

rahita (dcvoid) XXIV.38

r.iga(lusc) Vl.1-3, 6-7, 10; XIV.2; XXJII.l,
7 . 12

riipa (m11ccria.I fonn) !11.7; N . l- 5, 7; XXIIL7;

:XXVl.4-5
liil:sana (characccristic/ V. 1- ). 7; VII. I; Xl.7;
XVlll. 9;

xxv.4

a (without) V.1-3
11Jkandha- (-of non-aggrcgatc) XVlll.2
sarpskrta (of the conditioocd) Vll. 1, 3
-karrnanl (function as-) VII. 2
/11kf11'!1, tri- (having duce charactcristirs)

VII. I
.
lakfya (ch:uutcriz.cd) V. 5, 7; XI. 7
liriga, SI/O (owo matk) X. 5
loka (world) X:XVll. 21, 23-2)
para (other-) XXVU .21
-ni1tha (pattoo of the-) xvm . 11
-sarrivrti (worldJy coovcntion) XXIV.8
/auAika (worldly) XXIV.6, 36

vak111vya (to be dcdatcd) X:Xll. 11

11a1111 (foundation) XXUI. 7

viik {word, speech) XVll .4

400

THE PHILOSOPHY OP nm MIDDLE

v'iicika (vetb21) XVII.3

'var/41i1(1 (spcak.cc), Dcdkatory Vcrscs.
Vtmlf!J (best), Dedicatory Vccscs
11iuk.f111!a (discerning person) XV.10
vigraha (analysis) N .8
11id1ra (vc:riegatcd) XXIV.38
11i}llllp1i, a- (n.on-mtimation) XVIJ.4
flijlliinfl (consciousncss) fil. 7-8:

XXVI.2, 4-)
vitijflrni11111 ( cvidcnt) XXIV. 22
fi (not) 1.3; V.6; IX.2; XI\1.7;
XXIIL 12, 16
11itlyii (knowledgc) XXIV.11
11- (ignonncc) XVU.28; XXIII.23;
XXVI.1 , 11
J1itl11il# (wise ooc:) XXVI.10
a (ignorant one) XXVL.10
11inii (without) Vl.5; IX.4-5 ; Xll.5-6;
XX.24; XXI.1-2, 4, 6, 8; XXlll>-4.;
XXV.4
11initmtd111 (separatcd) ll.1; XXVIL4-5
11ip111114 (sub)ect to perversfon) XXIU.17
a- (not ) XXIIl.17
11i.paryflya (pcrversiOJJ) XXIlI.6, 1}-14.
16-2 1, 23
a- (non-) XXIJL 16
flipary'iisa (pc:rversion) XXIII.1- 2
11ip1Jry4Sjaf!iiina (being subject to
pervcrsion) XXII!. 18
flipiil:a (maturiry) XVII. 25
l'iptmvfl (marurcd) XVII.18, 2~
11ijm1'!iiifl, 11- (impcrlshable) XVll.14- 15 , 20
vibha1111 (sec bh1111a)
11ibl/ilgt1 (distinction) XVIl.19; XXIV.9
prtwi (XVll.24
vibhaviil (adept) X,V. 7
111roli (non-dclight) X:VU.4
11- ( delight) XVII .4
vbuadha, par'1Jp11r11- (sclf-conrradictory)

VIU.7
11iie!"1!" ( distinguishing) XXV . 19

tiiglhatui11 (dced) XVll.4

11rl4 (separatc<l) XX.10
fi ( DOt) XX. U

vedako (feclcr) IX.8-9; XI. 7

vet/ana (feeling) JV.7; IX. I, 3, 10-11; XI.7;

XXVI.5-6

vdiyt1rlhya (see 11rlha)

tlyatiArat114 (incerruption) XVll. 19

wAy

11yaya (ceasing) XVI.); XVlil. l ; XXl.15-16;

XXJV. l , 20

a(comtant) XXII.15
vya1111Jthi'4 (dctermined) IX. 2-4; XVU. 25
vyflflahiinr(convencion) XVI1.24; XXlV.10
IO'!J XX1V.6, 36
vyasth4 (scparaced) l.11; VII.2
11jiillr1a. a (indcccrmlnate) XVll.14'
11jiillhya111 (explained) Ill.:>, 9; X.15
11jiillhyiin11 (cxplanation) IV.9
'JUtsarga (rejeccion) Vlll. H
i11bda (sound) XXlll. 7
ft1ma (appea.scmcnr) XVIlf.2

Pl'fl XXl.17
i inltl (pcaccful) Vll.16; XVlll.9; XXII. 12

1"iJiwll4 (ctcmii!, etemalism) XY.11; XVD.8, 10,

20, 23; XVlll.10; XXI.14-15; XXII.12;
XXV.21, 23; XXVIl.15, 17-18, 20, 29
fi

(not-, non-) Dedkatory Verses)

XVIIl.11; XXIL12; XXV.3 , 23;
xxvu. 16-18, 20

-lou (-"Vo.rld) xxvn.1

~gr11h11 (grasping

after-) XV. IO
Jiisl11'a (mcssage) X.16; XVlll.11
b11ddh11- (Buddha's-)XV.6; XXIV.9
anu (admonition) XVIll.8
I 'iisJr (teacher) XVll.31~ :XXV.10
Jir11Ja~ (head) Vll.31
fiw (auspicfous), Dcd.icatory Vc:rscs; V.8;
XXV.24
ftll/11 (pure) XVIJ.11
i11r;i (pleasant) XXlll.21- 22
11- (unpleasant) XXIII.22
finy11 (cmpty) XIII. 7; XVU.27; XX.16, 18;
XXl.9; XXll.10, 14; XXlll. 13-14;
XXIV. l, 13-14; XXV. 11 22
il (non) XIII. 7; XX. 16'-17; XXl.9;
XXIV.19-20, 33. 35, 39; X:XV.1
fiiny11m iri ("empty") XXIL 11
fil (cmptincss) IV.8-9; Xlll.2-3, 8;
XVIl.20; XVlll.5; XXIV.6-7, u,
B-14, 18, 22, 36-37
-14rtha (mcaning of>-) XXJV ,7

-wa (cmptiness) XXVU..29

i ubha(pleasant) XXIII. l- 2, 6, 9-12
a- (un) XXIll.l-2, 6, 9-12
iou (grld) XXVI.s
Jm111111 (hearing) lll. l, 9: IX.1. 3. 10-11

401

INDEX

iiilv11k.il (disciplc) XVD.13; XVlll.12

iro111t1111 (sound) 111.9
irotr (hcarc.r) IX.8-9
s1111,1yd (rightly, completcly) XXIJl.16;
XXVI. 12
s1111,11111'111U ( rcsuainc) XVU. l
11111,1yoj11n11 (fctter) XVll.28
s1111,111rti (convemion) XXJV.8
111'!J111Jt1 (doubt, uoccru.inty)
11- (undoubtcdly, c.cnainly Vll.12; IX.4;
XVIl.22
"" (wlchout doubt} XVU.24
Sllf!Ullrlln (moviog) XVl.3
Sllf!ISllfill (a5.50Ciacioo) XIV. I, 3, 8
a (w1iho11J.) XVlll.2
saff.11int (ifc-proccss) XI.1. 8; XVI. IO;
XVll.20; XXV.19-20; XXVl.1 0; XXVU.19
11111,ll.Ura (dispositioo) IV. 7; X.Jl!. I; XVJ. l,
4-); XXJil.23: XXVl.1-, 10--11
Stlff.'14{111 (conditioncd) Vll.l- 3, B;
XXV.), 13
li (un-) Vll.l . 33; X:XV.) , 13
saf!Jf/"ll{ll (agem of 2SS0Ciation) XIV .8
St1ff.1Jr.iJt1m4n11 (2.SWCiatlng) XIV.8
I"f!ISr/{il (associ11tcd) XIV.8
Ja~l'Omll (wuuformacion) XVU.16
st1f(lkntm11 XX.9

s11f?1g11ti (assemblage) XX.12- 14

s11rriJlli1 (pcrccption) IV. 7

J'"!f/llit11 ( dcslgmm~d) XVIJ.4
Sllf!1141i (sucam, contlnuity) XXVll. 16
bha1111 (-of bccoming) XXJ.17, 21
I11f!Jli1111 (serics) XVU.7-10; XXVJl.22
'"/6Jl1VJllJl1 (of acising and ccasing)
XXJ.15-16
1at(cxis1cnc1) l.6-8: Il.li; VU.20; 30: VlIJ.7,

XVll.6
11 (non) 1.6-7, 12: Vll.20 , 31;
Vlll.7-8, Il
-111111~11'!1 (substantiality) X.16
.Ja (existcoce) 1.10
-111'1 (scmiem bcing) XVI. I , 4-~
-bho11a (existcncc) XXJ.17; XXlV.6, 16
4bliii11a (nonexistencc) XXJ.17:
XXJll.n
-bhu111 (come to be , real) Jl . 24-2~;
Vlll. 1-2, 7. 9-1 I
1P-bhu111 Il.24-2) : VUJ. I, 3, 7, 9- ll

(locarivc case of 11111111, pr.p. of 1111)

I.IO; IV.4: Vl.1- 2, 9: IX.9; X. 2, 9;
XV.4, 9; XX . 1~ ; XXJV.16, 24;
XXV. 15; XXVl. 7: XXVJI.~
fi 11.7; 111.6: V.2; Vl.2; Vll.17;
Vtll.4-6: XV.9; XVll. 30; XX.15;
IRii

XXJV. ~

Jfl/J'1, iry11- (ooblc troth) XXJV. l-2, 4.

20, 30

ti11e (rwo) XXIV .8-9

lousllf!lrli (-rclating to wocldly
convention) XXIV.8
p11ramiirthal'11j (-in terms of ultimacc
fruit) XXIV .8
s11tir!11 (idcnc.ical) JV.6
fi (not) JV.6
Jll"!nipii111 (occurrrcnce) XXVJ .5
I llf!tnibha (rcscmble) XXJIJ.8
Jllf!1Prakiii11Yilii (illuminatc) VIJ.8
lllf!IP'""rlli (commeoccmcnt, occurrcncc)
u.17
'1 (non-) V.4
.
lflf!'Priip1111 fi (noc rcacbed) XXJV.39: XXV.3
Jt1f!Jbh111111 (birth, occurrcnce) 111.7; V.4:
XV. I; XXJ. 1-~; 7-11: XXVl.3. 9
111".'bhim (occurrcd) XV .1; XXVll.12
111bhaga (simila.r) XVJI. 17
11i (dissimiar) XVlI.17
s1m111 (comparablc, cqual) 11. 17; lV.8-9;
VJ.2: XVl.2; XX.4: XXIJl.9; XXVJJ.14
saman11ihiil'l1 (attcntion) XXV1.4
sam'1Jl'1 (combined) l.11 ; Vll.2
1am1111111thita (ftxcd) XXJV .26
sam4ropa (atuibutioo) XVJ. 10
JRmain'lll (associaterl} XXV.21
sam1111hi111, Rpratyay11 (issuing fonh from a
OOOCOndition) XVJl.29
11Jm111p11111111 (uisen)
prafftJ1J (depc:ndently-) XXIV.19. 21
praty11y11- (-from a condition) XVll.29.
tihllrm4tihimna (-from good and bad)
XXIV.3 ~

samutpatia (arising)
pr11fity'1 (dcpc:ndcot-)(sec prRl1/ya)
samutiaya (arising) XXJV.22. 25, 40

samudilhrt11 (sptcificd) XVU.27

Jflm11tibh'iita (cmcrgcnt) XXVll.12
111m11pii n'111 (a530Ciatcd) XXVll.1- 2
a (non-) VJU.5; XXVI.11

402

THE PHllOSOPHY Of THE MIDDLE WAY

(snalcc) XXIV .11

s-iJ (all, cvciything) IV.7-9; VTI.19, 29;

Jlll'Pll

VIIl.8; IX.6-7; X.U; Xlll.l; XIV.l;

XVll.17-18,

~4;

XVlll.8; XXll.15;

XXIl!. 15; XXIV. l, 6, 14, 20, 36, 39;

XXVl-2, 24
Ariji ( acrivlcy) Vlll.6
-Rufi (view$) Xlll.8; XXVlI.30
-dh11fl'J'UI (things) Vll.29; XXV.22
-bhi1111 (-cxisrcncc) VU. 24; Xl.8,
XXVU.29
-Iw (-in the wodds) VII'. 11
s~111hi (everyway) ll. 14
111ha (with, togcthcr, simutancous,
coocomitant) Vl.3, JO; XI.5; XXI.I, 3,
5-6

(without) VI.10
J11habl:la1111 (co-existence) VI.4-9
iiiltrirb/t111111 (rcalizatioo) XXJV.'27
. iiill[ilt11rm11 XXIV.2
rjitlhaflll (proof, achicving) X .9; XVll.11
iiitlhyll (10 be provcd) IV.8
a- (iocorrigiblc) Xlll.8
iii11J11gfi (harmo.ny) XX. l-4, 7-8, 23-:Z4
proty11ya- (-of conditions) XX. 24
ltrttJ (made by-) XX. 24
sif!tpru'~ (slmultancous) IX. 12
iiirdhaf?J, (togct:hcr, jointly) X. lS; XIV.4;
11

XXl.20

1it/</h4 (cstablislied, PtOvcd) VI.6-7; X.9. 11;

XXV.U; XXVll. 18, 28

11- (not) X.11

si<ldhi (csrablishmcm).II.21; VI.10; VIIf.12;
XlX.3; XXJ.6
a- (non-) Vll.33

pra- Vl.8
1u@T~(head)

Xlll.3
svay~ (self, oncself, own) X:Xlll.18
-ltiira ( -causing) Xll.9
krt11 (Ca\ISed) Xll,1-2, 4, 7-8
-maya (made of.) I. 13
11111rg11 (hcavcn) Vlll.6

hetu (cawc, ca11$al, rcason) 1.2, 7; VUl.4,

9-11; XV.1; ~ 1-6, 8- 16. 19-23
d (non-) I. I ; XX.20; XXJV.16
helultil X.2.-3; XX.5
a VIU.:3; XI.4-5; XII. l, 9; XX.6, 8;
XXVII.12
11he1111t11 IV. 2
tvll (causal efficacy) XX.22
heya {to be rcliaquished) XVII.15

VIJ.31

rulthll (happy) XXI11. 21-.22

111suAfm11 (subtlcst) XXV.20
slmdha (aggregaccs) Xll.2; XVI.2; XXI.20;
XXll.l-2, 5-6; XXVU.22-24

11- (non-) XVIII. 1

d11f?lt11- (ma.u
suffccing) XXVI.9. 12
p11flca (fivc-) XXVl.8
viin (posses.:d of-)"X:xU.l
slfi (woman) f'.6
slhlina (enduraiicc) vu.2~. 34
J1hit11 (codurcd, static) VlI.27

of

t1

slhifi (duratlon, stasis) Vll.3, 23, 2), 33

spall (roucb, roocact) XXIIl. 7-8; XXVl.5
Sllf!I XXVl, 3
spar11'!a ( toucb.ing) 111.1, 9
1mr1a (reminisccd) XVIl.3-5; XVUI. l l
vi- (co.ofused) XXlV. U
1v11 (own)
-/a(from irsclf) 1.1; XXl.13; xxm.20
-pudgtlla (-person) XIl.4
sv11pn11 (dream) VJI.34; XVll.33; XXIll.8
w11/:/Jii11a (sdf-nature) 1.3; XIIl.4; XV. 1- 4; 6,
11; XVll.22; XX.21; XXl.17; XXll.2 , 4,
9. 14, 16; XXlil.2, 6, 24-25; XXIV.16,
2Z-24, 26, 28, 32-33, 38
wrJbhii11111ilf? (inhctcntly) Vll.16
s11iibh'ii11ilt11 (posst'S$ing sclf-narure)
XVU.25
li (wichout) Xlll.3
11il?- (devold of) 1.10; XVll.21; XXll.16
11if?11111bh'ii1111tva (abscasc of sclf-narurc)

VII.27; XIX.5

-bhiva vn.22

VERBAL FOR.MS
ajyate (mll,dc knowa) IX.5-6; XXV.16
ade1ayal ( taught) XXVII. 30
adhigamifjyaJi (will anaia) XXIV.32
lldhigamyale (is act:aincd) XXJV.10
anupaiy111i (perccive) XXIV. 16
apekfya (bcing eontingcnt} X.8- 12; XIX.1-2
llfl X.12; XIX.3; XXIIl.10-11
11bhiifala (has said) XID .1
abhipravarlale (procced) 1.12;
7' 9;

xvn,

XXVl,12

TJbhitaf!JJkurule (forms) XXVl.1

INDEX
abhi~11m,

11hhum (I cxisccd) XXVU. I , 3,

9. 13
1111aiJi (comprchcnd) V.6
aga&&hati (comes) x.n
ag11rnj11 (dcpcncling upon) XXVl.3
iipnuyill (shoud reach) X. 7
arabhyate, ar11bhye111 (begun, initiated)
11.12- B
lfn'ty11, 1111 (not rcying) XXJV. IO
afffyeta (should 1cly) XV.10

1cch.ui (incli nc, cxpcct) Vl.8-9

idhyat'iifl,I (being burnt) X.4
lhate (Is intcndcd) Vill. 1

403

ll11rute XXIV . 13
ulpay11 (woud thiaJc) XXIJ. 13
llramate (proc~ds) VJI. l}
g11&&ha1i (gOC$, movcs) 11.8-11. 22-25;
XXVl.1
pra 11.23
g11&&he1 X:XVll.19
gamyate li. 1, 4
grhyalt' (is observcd, is graspcd) XIX.5;
XXJll.15
grh~ati XXlll .15

grhyeta XlX. 5; X:X.3: XXVJJ. 7

ce1ayi11111 (voitio112) XVU.2-3

ucyate (is spolcc of) U.11, 22- 23: Vll.15

111p11dyatt1 (u isc) 1.5; Vl. 14
111p11riyt1111 VII.17
111p1111ya1e XX:.17-18
111pariyer1111 XXVll.23- 24
111padayate (producc) Vll.6; XVll.17-18
u1padayati Vll.18-19
u1pada:ye1VII.7-8, 18
111padayi111m VII. 7
up11g11my11111m (shouJd adm ie) 111. 5
upaliisyale (shown) 1.8; XXV.9
upapaliyatt (is appropda1e) Jl.3. 6-7, 9, 16;,
IV.4-6; V.4; VII. 21, 23. 27, 29-31; Vll.6;
I.2; XIV.5: XV.8; XVI.4; XVll.12; XX.19,
22; XXl .9-IO; XXll.), 14; XXlll .10;
XXJV.2, 13, 24, XXVll.3 . 7, 9. 16, 27
upapatJyate U.3, 9, Hi
11papaliye1e 11.23
_ 111m- XXVIJ.1j
11plllabhy11te (is obtaincd) JV. 1
upadadyad (should be dependent) XXIJ.5
11p11riasyate (wlll be graspcd) XXlI.6
up11day11 (dependent) XXl1.2, ) ; XXV.9. 12;
XXVJI.30
1111- (not grasping. indcpendcm)
XXlJ.5-6: XXV.6. 8-9. 12

uhyate (is assumcd) XXV.17- Ja

chadayifyali (will conttal) Vll.12

1i1r1ayate' (produccs) Vll.4, 6. 13;
XX. U - 16

ja1111ye1 VII. U ; XX.10-11, 23

janayifyati VIL5; XVU.6; XX.21
ja11yaJe XX. 7
j1111iy111 (should be know n) Ili. 9
jay111e (be, arisc, be boro) IV .8-9:
XVll. 15;
XX.1- 2; XXlll.20; XXVU. I 5
j11ye1a Xl.3; XXVl1. IO
jiry11111 (age) XJll.5
1lr111lrtya

(having separated 11.6-7; lll.6

li[{hati (stays, is statiooary) 11.15- 17;
Vll.22, 24; XVll.6 , 18

riattva (having passcd on) XXV. 5

11 XX.6
drsyate (is sceo) IV, I ; XXI. U
dr1ya1am XJV. 2
drff11a, adrstva, XX. I I
riesyate (is taugbt, is prnched) XXJV.10
de1ay11mt1Ja, Dcdicacory Verses
de111yi111m XXJV . 12
droA,y11ti (sec) lll.1

ltathy111t1 (is callcd, is cxpressed, is said)

XV.3; XXII.li; XXlll.t
karoti (performs, doc:s) Vili . I, 3, 9-11
/luru te VIII, 7
ltnyale VIJ.8; XVll.22; X:XIV.33

urifyate XXIV.33
.uryat XJ I.7

dha/qyati (bums)

X .)

nanyate (is des1roycd) XXVU . 26-27

XXVll. 26-27
1111m111yami (I bow reverently) XXVll. 30
nllJti (docs not cxis1) 1.2; ll.22, ll . 6 , 8;
yj.

404

THE PHILOSOPHY OF THE MIDDLE WAY

lV. ~; Vl.4: Vll.9. 32-33; VIII.2: XID.3,

7; XlV.6-7; XVI.2 , 7; XVll.29; XIX.6;

XXJ .1, 7. 21; XXIl.2, 4, 6, 8- 9;
XXlll.10-11; XXJV. 1, 20, 22:
XXV. 1-2, 8, 19: XXVU.8, 19
111111111/i XXIV.3
ninl.tlh,11111 (ccascs} Vff.26, 18: XVll.19;
XVll.>; X:X.)-6; XXJII.23: XXVl. 12

nirudhyllllii XVllJ.4
11iro11111 XX. 17-18
n1r.:lir11nti (posit) X .16
11itW1r111111 (produces) 1. 7

nirvll!Jllli (cease) X. 5
ni,-,#{y11111i (J will be free) XVI.9
"i!i&yote (is infuscd) XXVl.2
lllhof'!i!Joti (will destroy) vn. 11
pori.lllp11y11Se (umcclvc) Vl.8

PlllJl11i (perccives) 111.2, 4--5 ; V.8; XVIll.3;

XXIV. 16, 40
posyimti XV.6; XXII. U
. ptt.syomllh VUI.12
Prdip11J11ti (illuminaccs) VIl.9
Pl&M,mllhe (souc) XXlV:tB
Pfllfa11pyau (makcs known ) IX.3; XIX. 5;
XXll.5. 10

prt1fa11P11Jemahi XXlll.10-ll
pra}noydle (is known) Xl. l
pra1i/J11dh11re (conmdicc) XXJV. 5- 6 , 17, 23,
36

Pf1111tyll (depending, dcpendcm, dcpendcocly)

1.5; UJ. 7; Vl. l ; Yll. 15- 16; Vlll.12; XII.2;
XIV.5; XVIII. IO; XJX.6; XXll.J;

x:xv

XXlll. l-2, 6, 10--11:

.9: XXVJ.4;
XXVll,2.3--24
11 XXIV. 19. 21, 31: X:XV.9
prlllliy11te (givcs over co) Xll.5
pr11p11c11Janti (obscss) XJ.6; XXIl. 15
pfllbh11V1111ti (arisc) XI.6
pro11~omi {I will expound) XVU. 13
Pfl1Vllrl41e (proceeds} 1.12; V.J; Vlll. 12;
XV!ll.12; XXVI. 7-9; .JO..'Vll.22
pr.u11jyo1e (wiJI , follow, fimply) Il.4, 10; IV.2;
Vlll.6, 8; X. 3; XV.1j; XVll.23; XX.9. 18;
XXl. 14, 16;
20, 31: XXV.4;
XXVll.11-12
I
ptrJ.J11jye1e ll.61 11, 19; V. I; XX. 7
prasojyeran XVII.16
pr.rhif'!Oli ( bcstows) XIJ.6

XXN.f.

pf'flh11.1y111i ( rdlnuish) XXJll.24- 2)

prahiyer11n XXllI.24-2S
pradurbho11e1 (would appcar) XX. 7- 8
pmp11yqya1i (will lead to) XX.IV. 25
propf!ule (rcach) Yll. 10
prapf!uy11d X .6-7
Pf"llPJll,

Vll. IJ

prop!J11le X.5
prety11 (having gooc bcyond, in the nat life)
XVll.1.

badhniyod (sbould be bound) XVl.7

bll/lre (ha.s said) IU.8
blldhyale (is bound) XVI.5-
badhyonle XVl,5
bn1v111e (ca.Il, say) V.5
brumah XXJV.7
b/:Jajyerrm (wece to be dcstroyed)
XXVll.23-24
bha11111i (comcsr to be, eKists, i.i) l.10; lll.4
Vll. 16; IX. I, 10; X.4; X:Xll.14:
X.XV.1 7- 18; XXVll.3. 9. 15
bh11~11nti XXlll.4
Bho11i!Jati U-7: UJ.8; Vl.2, 4; IX.2. 4;
X.9; XllI.6; XV.2-3. !>: XVl.9;
XVtl.30: XVUI.2: XJX. t, 6; XX,17;
XXI.2-4; XXIl.3; XXlll.9. 12;
XXIV.5. 30, 38; X:XV. 7: XXVJl.6
bh11vifyami xxvn.i' J 4
hl1llel l.11 : 11.1!>: VI. I; Vlll.3; JX.8-9;
X . 1; Xl . 2~ ; Xll. 3; Xlf!.6-7; XIV.6;
XV. I, 8; XVl.3. 7; XVJl.22: XVIII.I ;
xx.~6. B-9: XXI.u. 14, 19;
XX.ll.4-5, 11; XXlll.16; XXJV.37;
XXV.5. 11-14; XXVl.7; XXVU.4, IO,
12. 16, 23- 24
bhave1a1n Vl.3
bh11vy111e XXlV.24
(having rcfl~ctcd) XXIV . 12
manye (1 considcr) X.16
1'11J&ya1e (is releascd) XVI.' , 8
m1111111

m1"yan111xvr.5
m1Jcye111XXVl. 7
mn'yale (dies) XXl.20

mriy4l11 Xl.5

ytitiJ111e (proper. pcuincnt) J.6-7; ll.18; Ul.4:

Vll.20, 25; IX.8 ; X . li ; Xl.5; XID.5:

405

INDEX

XlV.8; XVII.24: XXlV.14: XXV.10-ll;

xxvn.22. 25-26
y11fj11nte XXXIV.27
:10Jy11te x.vn.13

"1k,11yet (should be indicatcd) XVIJ.19:

XJX.4

11"'11'li (dcdaic) IX.1

1111ek1 IV.B-9
'111'tle (l salutc) Dedicatory Vcrscs
111u/p11ye1 (should think) IV. 5
v1blpy111e 11.14, 20; XVl. lO; XXII!. 7
11ikil1'~11Ji (faocy) Vl.8
11ij'1111tl1'li (uodcm.and) XXIV .9
11ijaniy111 VIII. 11
11itly111t(iscvidcm)l.l, IO; 11.21 , 25; V. 1, 5:
VU.31: Vlll.4-5: IX.6-7, 10-13: Xl.8;
Xll.10; Xlll.4: XlV.~4 . 7-8: XVlll.3;
XIX.3, 5: XX. 12- 14; XXl.3-6, 8;
XXUl.11 , 13. 15, 21-22; )Q(IV.4, 19, 21,
23- 25, 30, .34-35. 39; xxv.~-8. t G, 20
111rn-

V.2

llirlyPntll lX. ll
11itly11nte XXlll.2, 6: XXJV. 3
11itlya1 Vlll.13
vinasay1m1i (ruin) )Q{IV, 11
11ip11~y11ti (matures) XVU.25
111pr11f!ll!Jl1ti (pcrishes) XVll.21
vihavayet (should be critically cxamincd)
VJil. 13
11~/PSlltl

(rdlcct) XXlU .18

virutihyante (arc c.ontradictcd) XVU.24
11iJ011te (is idcntified) XXVIl.4
vihanyase (arc tormcntcd) XXJV. 7
11et1i (comprchend) XX.IV.7
vy11t1' (ccases) XXVll.6

11rr1j11nli (function) XIV .1, 3

sakyam (possiblc;, :1blc) XXIV. 28

.std1111y11d VII. 7
saf!lfi/fhtt (would remain) XXVll.10

Jllf!Jnilli1a1e (cntcrs) XXV1.2

saf!11tilli!!e XXVI.2
santi (arc, ClCist) 1.4; Xl. 7
It1f!1PraR11Jt1JaU (illuminatcs) Yll.12
Jt1f1JPN1111rtate (procttds) XXVl. ~5
Sllf!1Pr11pf!oli (rcach) X.6
saff!bhilllanli (occur) Xll.2; XXIll, 1-2, 17-18

Jll'f?1b"411i!Jllli XXVll.29
s1111UU1hig11nl11f!J (to rcalizc) XXIV.28
J11m11<kti (uiscs) XXVU.6
s11m11tle!Jllle XXlV.22
111m11p111rity11 (bascd upon) XXJV.8
I11f!111irlyate (is cvidcot) 1.14; VIl.17
1ao/-111r111i (t1'11nsmigratc.s) XVI.2
St1f!Ullrt1trli XVI. l
JITf?11t1rify111i XVI. 2
saf?Uuroti (forms) XXVl. 10
sitlhy111i (is csrablishcd) Vl.B-9; XV.4-~ ;
XXJll.4; XXV.15
stilhyate X. l 0--11
sitlhy111ah XXIJl.3
1itftihyaum X .lO

Jel!)ati (could be) VII.33

1lh11SJ11le (rcmaias) X.~
sy111 (would be, wece ro be) U.20; JV. 3: V.l ;
Vl. 5: VJH .2: lX.9; X. l : XJ.4- 5; Xll. 7-9~
XIU.4, 7; XV.8; XVII.16. 23, 26, 30;
XX.20; XXl.19; XXIV.1&: XXVI.7:
XXVU.11, 15 , 19, 21, 25
Jy11111m X. 7; XVl.8; XlX.2
sy11r Vll.2; XIl.3; XVJl.12: XX.4

406

TuE PHILOSOPHY OF THE

Mroou;

wAy

GENERAL INDEX

219. 247, 56.

ar/Mnl 24 , 25, 72.

Abhidharrna, srarus of 3, 6 , 174; period of

20; ca.dy Buddhisrn and canonical 22- 26,
29, 3H; theory of conditions 28, 38, 106,
107. U0-114: catcgories of the 37, 132,
147 , 193. 244 , 245 , 275 ; interprctation of
the 6, 125, 140, 141 , 160, 17 1-173. 178.

1triyap11nyuona-111J1a 17 , 240, 336, 240, 332.

Aryadcva 102.
Asoka 2 , 2).
asscrtion, Abhidbarrna 193; absolute 365,
Buddha's 159. 379; Nigicjuna's 42, 43, 55 .

179, 252, 278, 280, 284, 323. 331.

Abhidharma litcrature a.n d ccxts, 6, 20, 25 ,
26. 38, 181. 323.
Abhidharrnika 32, 107.
Absolucc 2), 47. 116 , 272. 389 , For adjectival

usc, sec undcr cxisccnce. diffcrcnce,

ldcntity, truth, reality. etc.
absolut.ism 86, 128, 340.
absoludstic (vicws, sy5tems, traditions, etc.)
1, 16. 20. n. 69. 85, 90, 148. 2s5 , 326,
328, 336, 337. 342, 3H . 388.
Acela-lcou11pa-s11111J 45
urion (.orma. ltn]"ii, etc.) 29, 30, 33, 42-44,
46, 51, 53, 54. 55, 56, 59. 61, 64, ~8. 69,
74. 90. 91. 128, 133. 162, 180-187' 195.
197, 205, 218, 219 , 236. 243-2)1 ,
253-258, 260, 261, 263, 266, 261, 2n,
276, 332. 3}9. 3H- 353. 380, 383;
impcrisbability of 30, 55, 251 , 254, 255.
agem (ulf, karaka, etc.) 29, 42-44, 46, 55,
69. 83,.90. 105. 128. 180-187. 195- 197.
205, 214, 221, 259-261. 302, 339, n2 ,
37), 380. 195.
aggrcgatcs (1.11Jh11) 18, 21, 29, 36-40, 46,
)2, 56. 64, 132, 140, 144. 147, 148,
179-181, 193. 196. 211, 212, 222, 236,
237. 263, 264, 301-307. 315. 374, 379.
381, 387. 388.
altrui.~m (absolutc) 4, 91: altruistic 91.
analysis 21 . 22, 40, 49-50 , 61, 125 , l28m
129. 143-147, 1')3- 156. 159. 167, 174,
224-226, 228. 277. 285. 293. 297. 30.8',
322, 349, 368, 386.
anoihilacion (ucr:heda, etc.) 9. 13. 14, 19, 29,
33. 4), 46. 51. 54. 55. 59, 67. 75. 79, 81 .
88, 101 , 21 9, 220, 234. 238. 254, 264,
267. 269. 273, 274, 287, 292. 298-300,
362. 378, 382, 383. 357
aonihilationism (iaclJet/41iiid11, etc.) 16, 92.
131. 184. 212. 218, 232. 236, 246. 259.
267. 292 . 298. 299.
annihil:uionist (uccheri11vitli. etc.) 184. 185.

56. 92. 114, 122, 123. 136. 142, 205, 210.

2}6. 265. 291. 340, 341, 347; racion:i.list
81 ; Sarvisrivida 109; subscantialist 71, 120,
123. 182, 383; Vicslpurriya 197, 230.
attainment, nou-amunrncnt (pr#pli/
11priipl1) 127 . 241. 242. 251 , 318. 349.
amibutc 17. 43, 47, B. 62, 122. 124, 127129. 150, 155, 190, 222, 2H, 239, 258,
371.
bc:coming (bh111111) 10, ll. 15, 26, 32. H,
62-64. 72. 74, 78. 88, 218, 237. 241, 294.
298-301, 337, 361, 362. 370, 373. 374.
H6. 379. 385. 390: sucwn of 62, 218,
241 . 299. 301.
bcginning (absolutc) 44, 61, 79. 2o6. 207.
259. 271, 292, 293. 368.
Berkeley, George 112, 142.
Bhogavadglfii \9, 20, .238.
Bh:lvavivckll 26, 101, 105.
borihua111111 24, 25, 53. 91. 348.
bondage (b1mdhana. etc.) 9, n, 18, 20.
2.9-H. 40, 51-53. 56, 57 , 72, 76, 77, 80.
85, 88, 92. 133. 235. 238-243, 304-307,
3412, 356. 359. 368-370, 374.
Brahmajii/11-111/trmf.t li , 79. 80. 188. 385.

Candrakrrci 17, 26, 28. 29. 40, 86. 96. 101.

102, 105-107, 112, 114. 134, 138, 153.
154. 189. 199. 201, 203, 207, 234. 350,
251, 255. 381.
causallty 31 . 34-36, 68. 73. 77, 94 , 95, 97.
137. 143. 169. 174, 182 , 191 , 244, 347.
356; mctaphysical thco.cies 36. 77.
causation 22, 34, )0, 61, 78, 94, 97, 108,
114. 116, 134. 143. 144. 162, 163, 165,
166. 170, 171, 2ll. 215. 216, 273, 276,
282, 283, 286. 327, 369; ~ternal causation
14, 27. 28. 32. 46, 1o6. 107, 115. 174.
176. 204. 259. 260. 298; sclf-causatlon 14,
27. 28, 32', 39. 46, 106. 107. 115, 168,
169. 174, 176, 204. 212. 214, 258-260.
280, 281 , 298.

INDEX

cercaincy (absolute, etc.) n. 79, 84, 188,

256.
change 8, 27, 28; 31, 34-36 , 41, 32, 54,
102, 125 , 131, 134, 159, 160, 163-165,
173-175, 178, 179. 182, 192. 204, 220222, 23()-233. 263, 273. 274, 286, 295,
346', 369.
clairaudience 8}.
dalrvoyance 83.
cogito, sec under Dcsca([Cs.
co-existeli.ce 155-H8.
commitmeot (pralijn'ii) 92, 93.
conccptuafuatoo 49, 56, 57, 66, 129, 143,
209, 249, 263.
conditionalicy 18, 30, 271, 388.
Confucianisrn 8.
confusion (tnrfii, 11UJha, etc.) 46-48, 59, 66,
67, 72. 82, 90, 217, 219. 297, 312. 313,
323, 332; big bloomiog buzzfog 92, 159;
cognitive 316, 318.
consciousness (11ijffiin11, viftifii1!11) 10, 11, 18,
38, 49. 72, 81, 84, 111, 1H. 136-138,
140, 147, !)I, 225, 261 , 264, 280, 386,
371-373, 381; self 379; selfpt"tcc:iving 37;
tiansmigr.uing 235, 236, 252, 380.
contiogcnce (apek.fii) 198-204, 276,

convcntion (satr111rtil sammuti1 ryavahiiral

11oh'iiN, prajiiapti/paflfatti, nif'gtti, etc.)
14, 55, 69, 241, 256; absolute reality 17;
dependent 86; idea! 330-~32, 334. 340;
linguistic 18, 19, 35; mora 88, 89; woildly
68, 330-332, 334. 340, 352. 355. 366.
dcath(matll'fll) 9-ll, 16. 45. 47, 51,
57-59. 63, 64, 66, 73, 76, 78, 159, 173.
174, 178, 197. 207-209. 218, 219. 238,
253, 254, 270, 292-295, 301, 303, 309,
358, 365, 366, 368, 374, 374.
dcfilcmem (./efa) 55, 56, 112, 224, 225, 256,
258, 259. 262, 266. 278. 313-315, 324,
325, 335, 353. 356.
depeodence. principle of 12, 13. 15, 22, :B.
37, 38, 43, 45, 46, 51, 33, 61, 64, 65, 68,
69. 77. 82, 88, 93. 108, 109, 124, 126,
137, 145. 146, 154, 158, 162, 163: 165,
168, 190, 191 . 200, 212, 213, 227, 231.
236. 252, 254, 273. 279. 306. 316, 342,
361, 380. 388.

dependent uising (pr(l/fty11111mNJpiJfI,

pa(iccasarnuppida) 1, 2, 15, 16, 21, 22,

407

27, 31, 32. 34. 42 , 49-51, 66, 68-70, 72,

78, 80, 82, 85- 87, 93, 101-103, 114, 131 ,
139, 142. 146. 155, 167-169, 187' 188,
216, 221, 226, 228-231, 255. 267, 269,
272-274, 290. 295. 320, 331-333. 336,
339-341. 343, 352-354, 362, 384, 357.
dependendy arisen (prali1ya1amu1panna,
p11ficc4s11mupp11nnd) 15, 21, 34, 42, 48-SO,
55, 59, 69, 70, 82, 84, 85, 93. 114, 212,
220, 235, 236, 267. 273, 274. 313, 331,
332. 340, 341, 343. 368, 377, 380, 390.
Ocscancs, Renc (Cartesian) 37. 43. 119;
cogito ergo sum or ogiio 37, 43, 81-84,
92, ll:l. l33, 134, )37, 164, 188, 189,
191. 264, 266, 379.
dcterminism 16, 30; detcuninistic 235, 347,
}48,

Dhammac11kkappa11alta1111-1111111 l-2.
Dham11iap4d.rJ 20, 54, 91.
dharma (dham._1714), Abh[dhanna concepcioo
of. 20-21, dharnt4.(ihatu 8; docuioe,
discourse 25, 329. 331. 392; foUI different
uscs 15-16; ethical 43 , 68, 88, U3, 183,
243. 257. 330, 335, 336, 340, 349, 366;
metaphysical 18, 19, 22-24, 32, 39. 77.
84. 141. 172 , 176. 177, 369; life
( ~ janman) 2)3, 275 ; nacurc ( = dh11rrnali)
55, 58 , 268, 269; ontological 4', 8, 12, 19.
21, 22, 24, ~4. '!>7. 40, 51. 55, 56, 67, 70,
71, 83-85, 90~ llO, 11, 132, 140-142. Ul,
153. 181, 217-220, 224, 234, 235. 273,
286, 288, .Hl, 336, 354, 390, (See also
unde( demcnrs.)
diffcrence (absolute, real, etc.) 12, 16, 23,
33, 40, 50, H. 29, 64, 76, 81, 92, 101 ,
106, 108, 111 , 116, 128, 129. 136. 137.
149. 160, 168, 182, 191, 192, 197. 199.
202, 211, 221. 225-228, 230, 231, 23~ ,
244. 2ts . _259. 262, 261, 273. 274, 276,
281, 284. 288, 289, 296, 304, 306-308,
310, 312, 360, 362. 365, 366, 381-385 ,
387.
disdples (rii11aka, ecc.) 2-4, 17, 19, n , 26,
31, 32. 79, 80, 90, 133. 23~. 236, 240,
243. 249. 250, 274, 292, 380, 387.
discrimination H, 56, 59, 88, 92, 126, U9,
143. 193. 266.
dispositions (sarrz.skiira, sllkhar11) io-12, 14,
18, 40, 46, 49. 65. 66, 72, 73, 82, 84, 138 ,

408

THE PHILOSOPHY Of THE MIDDLE WAY

144, 159, 181, 217-220, 224, 225, 324,

352, 353, ~70, 371, 374-376, 378, 381 ;
appeasement of (-fllffUlth(J, t1Pfli11ma) 4,
47, 48, 92, 152, 335-238, .264, 308, 320,
336, ;15, 376; ccssation or waning of
(-k,rdyd) 218, 219, 238, 375; lump of
(-p11lj4) 180; uansmigrtc 52, 235, 236;
unconditioncd by 6,, 73.
discinccion (11i1e!llf!ll, etc.) 7~76, 114, 121,
127, 129, 130, 14(), 141, 149, 151, 158,
172, 173, 199. 233, 239.. 258, 280, 281,
283, 285 , 349, 359, 366, 367, 369;
Humcan cxplllnacion 72.
Dogcn 257.
dogmatism 18, 80, 86, 92, 219, 223.
dream (svapna) 179. 182, 316.

clcmcots, adveoticious 241; discrctc 22, 381;

of Cllpcdcncc 15, 21, 29, 37:-40, 52, 83,
132, 140, 142, 147-l H, 153, 154, 235237, 261; non-subscanciality of (-nt1iriitmy11)
24, 29, 37. 40, 5l. 84, 181, 235, 235;
subjcctivc 313; subsrantiaist thcory of 4,
7, 24, 84, 140-142, 147-151. 192, 193.
197, 198, 216, 332. (See ~lso dharma.)
cmpiricist 8, 33, 39. 81, 83-85, 113, 115,
ll9, 132, 188. 258. 271, 278, 282, 285,
286, 293. 341, 342, 357, 368.
cmptiness (h1.0yat1.) 5, 6, 29, ,48, 49, 53. 55,
56, 67-7 1, 77, 80, 82, 85, 86, 89, 90, 93.
144- 132, 182, 22(}-223, 231 . 254, 266,
268, 288, 319. 320, 326-328, 330, 331,
335. 337-341, 343, 347, 350, 352, 355,
368, 390.
empty (sun~) 14, 49, 64, 67, 70, 71, 85, 86,
89, 93. 103, 145 , 220, 222, 223. 258, 259.
261, 262, 267, 287, 288, 296, 307-309.
318, 319. 326-328, 330, 331, H5.
337-342, 350, 3H, 3B, 355, 356, 368,
375.
cnightened one 12, 47 , 58, 63, 65, n. 87,
30.2, 321, 330, 347' 348, 356. 249. 274,
313.
cnightnmcnt 1, 9, Il, 14, 17, 18, 63, 92,
138, 240, 327, 335, 336, 323, 375, 376.
cternal, life 63, 74; sdf 12, 20, 30, 37. 43,
H, 56, 57, 64, 67, 78, 79, 106, 191, 219,
222, 236, 238, 262, 263, 269, 298. 379.
383.

ctc:rnalism 16, 92. 109, 131. 184, 212, 218,

228, 234. 236, 246, 259, 267, 292, 298,
299. 353. 377.
cternalist 24 7, 79.
exiSrencc (absolutc, etc. bh'ii11111 11Jfti111, etc.)
1, 3, 7-11, 13. 14, 17 , 21, 24, 27, 29, 30.
'2- 34, 36, 39, 40, 42-46, 50, 51, 54, 56,
57' 61-65, 67-69. 71-76. 78. 79. 82, 85 ,
88, 103, 108, 113-115, 125, 131, 132,
1)(}-152. l.5~158 , 163, 164, 176, 177,
181, 184, 185, 188, 193, 194, 201, 202,
205. 220. 221 . 228, 231-234, 235, 236,
241. 246, 261, 264, 269, 272. 272. 276,
239 .. 294, 296, 298, 299, 301, 303, 309,
310, 314, 322-324, 326, 329, 330, 332,
339. 342, 344, 347, 350. 356-365, 369,
377, .378, 382, 383, 390~ non-existcnce
(artrl, 11bha1111, etc,) l, 7, 9- 11, 14', 29, 32,
39. 40, 42, 44, 50, 62, 67- 69, 73, 74, 78,
82, 88, 103, 108, 13 1, 15(}-152 , 154, 163,
177, 181, 184, 185, l93, 194. 220, 221,
228, 231-234. 269. 272, 298, 299, 303,
309, 314, 324, 326, 329, 342, 347, 357,
358, 360-365, 355.
cxpericocc: l, 12-15, 19, 21, 37, 47~9. 51,
56. 60, 69, 78, 82- 85, 89. 92, 93, 108,
ll l. 130, 139, 145-147. 153 , 156, 159.
189. 192-194. 216, 219, 220, 234, 261 ,
262, 264, 271. 278, 282, 308, 313, 316,
319, 320, 323, 331, 335. 337, 34'<>-342,
350, 384, suc:un of 192, 349.
finite (11n111) 292, 368, 378, 387-390.
fue (11gnt) 39, 140, 151, and fueld (i11dh11t111)
44, 134, 135 , 29$-205, 222, 227, 230.
freedom (11irvii~11, nibb'iin11) l, 4, 15. 16, 18,
21, 29-31, 40, 41, 43, 47, 49. 52, 53, 56.
n. 59. 63. 6~77. 80; 85 . 88, 89. 90, 92,
97. 102, 103, 138, 153, 160, 169. 184,
187, 218, 23S, 238, 24<>-243, 251, 266,
268, 269, 299. 300; 307, 321, 328, 329.
332-336, 346-349. 353, 354, 355-370,
373. 374. 376, 379, 385, 387, 390, 391.
Fruit (cffc"Ct, consequence, etc. 11r1h11/ 11tth11
pa/11, vipak11) 14, 15, 16, 19. 29, 33. 43,
5:3. 55 , 61. 68. 69. 73. 77. 80, 85 , 88,
89, 108-110, 183, 84, 187, 189. 243, 288,
290, 291, 35<>-335. 339, 347, 348, 3H,
352, 356, 357, 366, 383: ultimatc or

409

INDEX

highcst (p11rrzmiirtha, p11rrzm1111ha) 16, 19.

68, 69, 73, 77, 80, 88, 89, 184, 187, 33133), 352, 356, 3.57. 366.
grasping (upiidii1111) 10-13, 38, 40, 42. 48,
56, 63, 79. 80, 92, 102, 138. 187 ' 204.
205, 218. 219, 234, 236. 237, 239, 240.
263. 266, 304-309, HB-321, 329. 363.
373, 374, 376, 379-381, 389. 390,
Gudmunsen, Chris 98.

harmony (silmll~n. _ctc.) 30. 59. 61 , 226,

no.

280-283 . 286, 290, 29 1.

human personaliry 68, 70.ja2. 85. 369;
Abhldharnu analysis 21 , 181 ; analyscd into
aggrcpccs 38. 39, 263, 264: annihilatlon of
9; bondagc and &ccdom 29. 30, H-59.
dispositions and 47; mctaphysiw view 19;
non-subscanaliry 51, 52, 61, 62, 67 .
292; Sarvistivada and Sauuincika vicws 25,
84; sclf-consciousocss and 1.H ; six clcmcms
147; survival of 62, 77. 78. 90, 92, 236,
in , 369, 3n, 378. 383; cwclvc faccors 1.
7, IO, 11 . 13. Z5 , 77;
Hume, David 81, lH. 156, 224 , 338.
idea 3. 4. 15, 24,

25 , 63, 88, 89, 91, 332,

334, 335. 341 , 353, 355, 36~
idcncity 16, 22, 33, 34. 36. 50, 51 , 56. 59.

64, 75-77, 81 , 101, 105. LOB, 111. 116.

128. 129, 136, 143. 144. 149, 1S5, 156,
))8, 166, 169. 171 . 177. 178, 191. 192 ,
195-197. 202, 204, 213, 221 , 226-228,
230, 23'1, 239. 245 . 254. 259, 262, 273,
274, 276, 277, 281. 284, 288-290, 296.
298, 306, 307, 310, 312, 3)1, 360, 365367, 378--382, 348, 384. 387.
igno~ncc (a11iriya) 10, 11 . 14, 82, 163, 259.
324, 370, 37 1, Jn. 376.
illwion (nriyii) 179, 182, 316, H 7.
illusory 42, 179.
immortality (amr14) 45, 63. 208. 218, 382.
!nada 3. 20, 38, 53, 55, 64, 70, 94~97 . 120,
144, 160, 216, 217' 230. 232, 191 , 237'
243, 248. 250. 268, 270, 271, 294, 296.
310, 323, 327. 329. 330. 337, 340, 344.
345.
358, 367, 369. 370, 374.
infinitc 161, 162, 170, 308, 368 , 378, 387390; rcgrcss 161 , 162. 170.

no.

inhcrcnce 33, 109, 276.

lti11utlllkll 313.
Jaina 14, 244.
Jaini, P. S. 350.
James, William 35, 47, 60, 88, 97. 192. 219,
308. 334, 337' 381.
jayawidacma, N. A. 94. 95.
Kaa:iyana (Kiryayana) 7. 8, 10, li, 50, 78,
82, 103, 108, 184, 212, 220, 221, 228,
232. 271, 272, 294, 314, 320, 342, 356,
569. 370.
K4cliiya11agollil-1uffa (.Kiityiyanivavida) 1, 5,
1. 1~14 , 25 , 26, 29, :n. 63. 1s. 79, 81.
232.
Kant , Immanuel 33, 322.
karma (sec undcr action).
Karmmiridhiprwral!a 29, 55, 250.
Kii1yap11-p11n1'11rta 4, 6, 7, 24, 25.

K11(h11 Up1111i{llri 189.

K111h'iJ111111h11 2, 21. 23, 24.
Kityiyan (sec undcc Kaccayana),
Kil'jiiy11nii111111iria (sec undcr Kacc#y1111agotlll
SU/Ili).

lmowlcdgc, of the beginning 61; of

dependent arisiog 21, H 11 339; i odc
pendent or nor ocher depcndcoc (11p11r11pra1yay11) 10-11 , 59, 271, 272; empicical
knowlcdgc 381 ; going beyond expcricncc
320; foundation of 82, 264. 372: Jcading to
frccdom 92; of the future 159; of the past
206; hlghcr (abhiflii) 261 , 262; Jamcsean.
cxplanacioo 47; limications of 377; mctaphysical 285, 332; of objcct, etc. 133. 148;
not omniscicncc 159. 358: pragmatic
dcfinition 219: as prajflil or y11thiibhu111
jflina 2, 11 , 12, 67, 80, 274. 285, 294,
331. 342; sourcc of 84; transccnding che
&cnscs 269; of ulcimatc fruit 335:
Upani.~11dic view 63; ViiuTpuuiya dcfini.t.ioo
196.
KumirajTva 3, 8, 39. 50, 120. 18)243, 322,
371. 38 l , 382,
1."11i11111ir1111r1 366.

lifc-proccss (1111JJiira) 44, 45, 52, B , 55, 61,

74-76, 9 1. 206-208, 210, 211, 241 , 243.
246, 252, 254, 256, 263. 292. 293. 366,

410

THE PHlLOSOPHY OF THE MIDDLE

367 . 37). 376, 386.

M.1ulh11pi'! r!'4-f11/t11 373.

Mah'i111tJSt11 24.
Mahiylina 1-8, 20, 24. 25. 59, 69- 71. 75,
77. 82 , 90, 9l, 94. 174, 257. 329. 348,
350. 366, 370; Mahaylinist 7. 53.
Matc:riaJisu l, 9. l4, 19, 39. 40.
maucr, Abhid\larma concepcion 193; early
Buddhist view 39, 40, 14()-145; meta
pbysical view 9, t9, 39. 40; primordial 44;
Sarvlistiviid:l view 49, 111- 1}2, 140-145.
meraphysician(s), Buddhist 5. l5, 28, 41,
43- 46, 49. 51, H. 55. 60-62, 68, 74. 76, 88,
114. 127. 128, 133, 137. 146. 147. 149,
151 , 159. 160, 174, 175. 178. 189. 193.
194, 202, 207, 208, 213, 224-227 , 231 .
232, 236, 241, 260, 265. 267. 269, 273,
216. 218, 292, 313, 321, 328, 330, :n2,
H5, 337-339. 341, 343. 345, .349, 355,
357-359. 361-363, 366; r.caditional Ind.iw
62, 68, 189, 22~227 .
middle pach l , 2, 4, 7 , 16. 18, 25 , 69, 78,
86, 9()-92, 183. 267 , 320, 339323.
Mitchell, Dorutld 97.
Mogga!Tpurca-tissa 2, 5.
momc:nrs (theory of) 22, 23, 28, 35. 49, 54.
6()-62, 82, 112. 113. 125, 127 , 128, 131 ,
134, n6. 161-164, i66, t6s, 110. 112116. 191 , 192, 206, 213 , 222 . 236, 246,
2)2, 253. 278, 279. 282, 284. 288, 298.
300, 301, 319. 349.
monl, pc:rfcction U , 59, 90, 366; puri~ry 256,
257: rcsponsibility 9. 14, 27, 30. 46, 62,
64, 70, 85, 90. 180, 181. 184, 187, 235,
243, 352, 369, 383; truth 13, 59. 334.
motion 36, 38, 68, 82, 119, 121, 126-Bl,
167, 203, 204; movcmc:ot 3. 19, 24,
34-36. 118-131
multiplicicy 181, 182. 364.
Mutjlu UpflJ'lf!ad 189.
Muni, T . R. V. 4 , 7, 28, 33, 86. 94. 183.
Nabmura, Hajime 95, 340.
ncgatioo 45, 102, 143, 169. l!H. 197, 252,
269, 295. 345. 365, 379; fourfold 58, 270.
OOOCOOtCptuaJ 32, 66, 87, 88, 203, 268,
269.

objcct (cin!(1111y11,

11!f1J4.

c:tc.) 50, 59, 76. 84,

88, 112, 131, 133. 136-138. 140, 148. 154,

V/! AV

192. 224. 266, 271, 320, ;B2, 372;

appeasement of 40, 87, 264, 308.
obscssioo (prt1jJ11iica). 18, 48, 56, .59, 86. 153.
266, 267, 270, 272, 308, .BG. 337:
appeasement of 40. 84, 101- 103. 152, 264,
369; illoof from 310; c:Jimination of 18,
freedom from 31: root of 43.
omnisciencc: 47, 63. 92, 1.59. 219, 227, 358.
ontology 81 , 83. 323.
ot.hc:c-narure (p:ln-bhiva) 23, 28, 33, 37, .50,
H , .57, 64, 67, 74. 76, 77, 81, 83, 88, 107.
114, 212, 21.5, 229-232, 272, 302-304,
3o6. 307. 382.
Pirileyyaun#lll 377.

pc:rtn:lOCOCe 19, 29, 45, 51, }4, 5), 59, 81,

88, 101. 160, 236, 237, 24), 253, 256,
267 . 272- 274, 287, 292. 294. 295. 319,
320. 387. 388.
pecson (pudgala) 23. 24, 26, 29, 36, 44 , 4.5,
51 , 52 62, 83-85, 105, 126, 130, 180,
181, 189. 195-197. 200, 202, 21}-215.
221, 222 , 235-239. 299. 311. 312, c~cc

also under humllfl personalicy).

personilliru 23, 36. 37.
pc:rvcrsion ( 11ip11ry'ii.J11) 312- 31.5, 318-32 1,
322, 543, 312, 313. 3 15, 316, 319,
321, 322- 324, 326, 327.
po1eotial j1y 108, 161, 170, 246, 349.
Powsin, L de: la Valc:e 97, 101, 127, 213,
377. 356.

power (11h1) 81. 108, 133. 261 , 262.

pngmatic conccption, of conditions (pratya)'11)
109, 110; of emptincss 330; of language
308; of moraliry 89. 2 n, 334. 366; of
philosopby 8; of truth 16. 19, 67, 80,
85, 159. 219. 224, 225. 227. 229. 272,
332.
Pmj11'i1pirami'J7i-1iilr11J 3. 7. 367.
prii1angika 26. 101, 102.

Rath1111inlla-1111111 240.
reality 9 . U. 27, 44 , 49. 57- 59, 76. 82, 84,
85. 144, 152. 156. 163. 180, 181, 196.
213. 218, 220, 23()-232. 249, 2.59, 271,
n9. 310, 319, 330-334, 340, 352,
355. 37.5, 389; ultimatc: or absolutc
rcaJicy (p11TrJmii.r1ha) 16, 17. 19, 20, 66.
87. 189. 203, 268, 331. 3.H . 356. 361 ,

364. 366.

411

INDEX.

realization 59 , 63. 267, 286. 328, 346.

rebirth 56, 206, 235. 237, 252, 255, ;266,
284, 294, 299, 375. :n 8, cigl'n view 10-12,
31, 78, 80, 87. 103. 249. 271 , 272.
Robinson, Richard 8.
RusseU, Bertrand 225.
S11hb4-Iflll4 189, 231' 270, 358.
Sarlrlht1rmt1f1'f!tjliillasufrtl

3, 4, 6, 24, 25 ,

174, 366.
Sarvasrivada i . 6, :H -25, 32, 36, 39, 41. 50.
52, 58, 62, 64. 67, 76, 77, 84, 85,
107. 109. 119, 127, 135. 141, 143. 156.
159-162 , 169, 170, 173, 182, 189. 196.
206, 225. 236, 256. 280, 288, 290, 296,
299. H 3, 349, 359. 367; Sarvinivadins
22, 25-28, 32, 34, 35, 38-40,
44, 50, H, 60, 62, 69, 77, 81, 84, 105,
106, n;, 119, 125, 120, 129. 134. 137.
U6, 160, 1n-116. 181, 196, 201. 213,
270, 276, 288, 298. 327, 338, 349. 366.
Sauuaocika 6, 22-28, 34-36, 41 , 44, 52, 54,
57. Go, 62, 67, 16,
81. 84, 05 , 106,
IB, 119, 125, 128, 129, 134, 141, 143.
U6, 159. 161 , 163-166, 168-170. 172176, 189, 194-196, 206, 207, 213, 236,
246-248 , 255, 256. 286, 286, 288, 290,
293 , 296. 299. 300. 301 , 332, 333. 338.
349, 355, 367, 383.
'self (almaf) 9. 14, 19. 20, 27, JO, 32, 39.
40. 44. 48, 51, 55, 56, 63-65. 67, 68, 74.
78,, 105, 180, 181, 191, 218, 262-264,
269, 303. 318, 380, 383 .
sclf-indulgencc l, 4, 90, 184.
sclf-mortification i, 18, 90. 91.
self-l)atOIC (J1JiJbliiifJa) 1,
23, 28, 32- 36,
38-40, 48-52, 55-57, 64, 67, 69. 71, 74,
76. 77, 80, 81-83, 88, 102, 107-110,
113, 114, 140,, 141 , 154, 160, 169. 212,
215, 220, 221, 225, 228-282. 191 , 240,
255 . i57. 258, 272. 280, 288, 298. 299.
302~304, 306-311, 313, Jl), 322, 324.
325, 339, 343- 351, 355. 357. 382.
sclf-tcsuainc 53, 90, 91 , 243.
sense expcricncc 12. 82, 83. 159, 219, 264,
316, (sec also undcr expcricnce)
space (ii.iiia) 36, 39. 68, 89, 131, 147-149,
Ul, 161.
Spnmg, Maivin 96, 154.
Stchcrcabsky, T.l. 23. 70,' ?5, 154. 183. 19~.

n.

n,

Strawson, P. F. 98, 131.

substancc (s11abha1111, etc.) l, 7, 12, 13, 15;
21. 28, 29, 32 , 35, 40, 41, 50, 54. 55, 58.
64., 65 , 67. 68, 76. 77, 81, 82 . 84, 85, 88,
93. 103, 105- 108, 111 , 112, 122, 127-102,
133. 137, 141, 142, 145, 146, 148, 1.50.
155: Vi6. 161. 164, 169. 171, 172, 175,
181. 1S2, 189-191, 193. 195 . 196, 206,
212. 213, 219-222, 225, 226, 228, 229,
233. 235, 236. 239. 241, 255. 262, 267,
269, 270, 288, 298, 308, 318, 320, 327 ,
332. 338, 339. 341-34, 345, 346, 351,
3.53. 355-347, 375; non-subscantial 49, 65,
84, 85, 103. 131 , 181 , 218, 220, 255. 258,
312, 314. 317, 325, 341 , 390; AOO
subsuwtialiry 25. 35, 43, 49. 65, 67, 80,
178, 217. 220-223. 231, 320. 350, 390; of
agent/ action 184; of rhc :&ggrcg:ucs 179;
human petsonality (purlga/a- 11ir'ii1my11) 24,
26, 29, 51, 62, 63, 84-87. 181, 292, 3ll;
of phcnomcna (rlharmanairiitmya) 24, 26.
29, 40, H, 84-87, 181, 234, 264, .311; of
vicws 103.
sffcring (rl11~.h11) 10-12, 14, 30, 31, 40,
42, 43, 45. 46, 48, 51, 63, 65, 70, 74, 78,
80. 91, 184, 211-218, 237. 241, 259. 304,
319, 323, 32~328 , 343- 346, 3B, 354,
355 . 356, 362, 363, 374, 376.
S111ta-11tpila 17, 43, 331.
tabula r11111149, 153, 154.
T11.thigiita 10, 13, 16, 30, 47, 58, 65-65, 72,
197 . 218, 238. 269. 270. 302-312, 319,
324-326, 355, 365. 366.
Thef'llgithi 2.
Thcravada l-3, 5, 6, 20, 21, 25, 62, 69. 70,
91. 141, 333, 366.
Theifgat/111 2, 335.
time (ka/a)30, 35. 36, 59-61, 68, 82, 125,
161, 172. 197. 275-279, 286, 301.
transcendence 24, 25 , 90, 102, 149, 332, 334;
transcc:odeot 16, 24, 25, 152, 249. 303,
365; transcendeoral 363.
ttansmigntionists 23, 35 .
truih(s), absolute or ulcimate 47, 49, 82, 86,
103, 159. 218, 219, 270, 271, 272, 288,
342 , 364; a priori 12; and cofus1on 46;
Buddha's conccption of l7, 231. 232;
cha.tactecistks of 271; dependent :uising
230, 273; dispositions and 218; cmpirical
2

412

THE PHILOSOPHY OF TiiE

375; four noble 67, 326, 352; linguistically

transcendent i49; meaningful language 19;
metaphysic:al 88, mota 334; percepcion of
92; Platonie n9; pragmatic: dcfinicion 16,
85, 219, 225, 229, 272, 308. 334;
provisional 220; self-conuadictory 272:
statemcnt 3l9; two 68, 69, 330, 33 1, 333:
unspcalcable :HO; Upan~.rdic 11ohow1 J3.

MrDOLE WAY

Up1mi.f11dr l , 37. 43, 63, 105. 18l , 237 . 199.

200.
Valbh~1:111JJ ,
Vajr111;chetJi,;i.pr11jfliipir11milii

24, 240.
VitsTpuuiya 197, 195-197, 200, 202, 222,
230. 299.

Vigr11/J111ifiiw1r111irl 92.
Vinaya-pi/f1ll 4.

Ueda, Y 230.
ullimtlte retJllty (see umler re4'/1y).
undet/11reti questtons (1111yiilr1a-1111Jlu) 31.

volition 53. 244, 245, 248: volitional 53, 244,

252.

2.51, 309, 368.

uniformiry U, 1)9.
unity 2O, 230.
univcrse 13. 16, 65, 80, 89, 125, 166, 188,
310, 311, 353.

Warder, A. K. 7, 95.
Wittgenstein, Lud'l(ig 93, 217.
wrong view 12, 13, 78.
Yogacira 7, 9, 37.

Exr.erpts from reviews:

"The book represents a bold, new and independent
approach to the problem of Nagarjuna's philosophical
stance and provides a readable and stimulating addition
to the already existent literature on the subject.
The index to the Karika and the genera index
appended to the work increase its research value for the
serious investigators and students ofBuddhist thought."

Journal of the Oriental Institute, Vadodara,

"In this book, which is a new translation ofNagarjuna's
work Mulamadhyamakakarika. Prof. Kalupahana has
given a new interpretation of Sunyavada of Nagarjuna
who is the reputed scholar of Buddhism ... Sanskrit text
and its translation have precision in presenting the exact
nature of Nagarjuna as a truest interpreter of Buddha's
teachings. The author begins with a brilliant exposition
of middle path Madhyama Marga, for about hundred
pages, and th en presents the Sanskrit text of
Madhyamakakarika, translating all the 448 verses of the
text, covering 200 pages; and the last 20 pages are devoted
to index. The introduction is a discussion of all
metaphysical problems raised in 27 chapters of the text."

Pathway to God, Vol. xxvi, No. 4, 1992

"OTILAL

BANARSIDASS
PUBLISHERS PVT. LTD.
- - - Dellu (INDIA) - - E-mail: mlbdCvml.com
Wcbsite: www.mlbd.com

Rs.425

Code:0774x

I SBN 81 -20 1 -07 74 - X