An Appraisal of the Svātantrika-Prasaṅgika Debates

Author(s): Nathan Katz

Source: Philosophy East and West, Vol. 26, No. 3, (Jul., 1976), pp. 253-267
Published by: University of Hawai'i Press
Stable URL: http://www.jstor.org/stable/1397858
Accessed: 18/05/2008 01:32

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

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

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

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

http://www.jstor.org
Nathan Katz An appraisal of the Svatantrika-Prasangika debates

I. THE PROBLEM: ITS SCOPE AND OUR PROCEDURE

There were myriad issues in conflict between the Prasangika and Svatantrika
schools of Madhyamaka Buddhism: philosophic, formal, and practical. It
seems to us that the central focus of these interrelated problems is where the
limit of speakability is to be drawn. This issue is reflected in their understanding
of the relationships and levels of the relative (samvrtti) and ultimate (para-
mirtha) truths, the validity of syllogistic (formal) reasoning, as well as the
absolute or relative quality of the prasahga (reductio ad absurdum)methodology
of refutation.
Our procedure in this article shall be as follows: First we shall view Mad-
hyamaka scholarship in the West, demonstrating its Prasaiigika bias, and
pointing out some errors. We shall then consider the question of the two truths,
comparing the systematizations of Bhavaviveka and Candrakirti. It is by
understanding Bhavaviveka's pivotal position here that we may address the
question of the validity of his syllogistic reasoning, as well as to conjecture at
his purpose behind the syllogism, which we see as an attempt at (1) salvaging
the integrity of other Buddhist schools from the Madhyamaka critique, and
(2) pointing to the fundamental contradiction of samvrtti and paramdrtha,
which is not the mere accident of illogicality. Finally we shall consider the
debates held at bSam.yas. in Tibet between exponents of the Svatantrika and
Prasafgika positions, although the latter debator has long been erroneously
thought to have been of another school.
II. BACKGROUND: SCHOLARSHIPON THE QUESTION

Scholars of Madhyamaka have not been kind to the Svatantrikas. Edward

Conze expresses utter incomprehension at the Svatantrikas who, he claims,
"have upheld the well-nigh incredible thesis that in Madhyamika logic valid
positive statements can be made."' Bhikshu Sangharakshita chides Bhava-
viveka for holding "... the heterodox view that realization of Sunyata was
not indispensible for the attainment of Nirvana"2 which he sees as "... evidence
of ennervation and the precursor of decline."3 T. R. V. Murti more bluntly
states that the Svatantrika "... is against the correct standpoint of the Mad-
hyamika."4 L. de la Valee Poussin erroneously claims that "... the official
school of Tibet is called the Prasaiigika."5
Th. Stcherbatsky applauds Candrakirti because "He succeeds in driving
Bhavaviveka's school into the shade and finally settles that form of Mad-
hyamika system which is now studied in all monastic schools of Tibet and
Mongolia, where it is considered to represent the true philosophic basis of
Mahayana Buddhism."6 However, Richard H. Robinson keenly observed

Nathan Katz is a Ph.D candidate, University Fellow, and teacher in the Dept. of Religion, Temple
University, Philadelphia.
Philosophy East and West 26, no. 3, July 1976. ? by The University Press of Hawaii. All rights reserved.
254 Katz

that ".. . Nagarjuna's system has not been clearly distinguished from Can-
drakirti's,"7 an observation we would second.
Poussin shows greater insight when he writes of Bhavaviveka and Dhar-
mapala: "These two each represent an extreme; together they indicate the
Middle Way. They are in accord and not in contradiction."8 Frederick Streng
concurs: "We follow the lead of L. de la Valee Poussin... who suggests that
together the Prasafigikas and Svatantrikas show the 'middle way', one destroy-
ing the voidness of existence and the other destroying the existence of the
void."9 Although Streng nowhere does so, it shall be the purpose of this article
to indicate some of the vital contributions of Bhavaviveka.
III. HISTORYOF THE SVATANTRIKAS

Bhavaviveka (490-570 C.E.), or Bhavya (Tib. Legs.ldan. 'byed.pa. or Legs.

ldan.), is credited with having founded the Svatantrika school. His contem-
porary, Buddhapalita, is considered the founder of the Prasafigikas. Both
schools claim Madhyamaka self-identity and refer in their writings to Nagar-
juna's works.
The Prasaigikas derive their name from the methodology of reductio ad
absurdumrefutation (prasahga) employed by Nagarjuna in his Milamadhyama-
kakarikd. This methodology claims the refutation of all views (drsti) without
offering any view of its own. The Prasafigikas (and most contemporary scholars
of Buddhism) claim that the utilization of the prasahg methodology defines
one as a Prasaiigika.
The Svatantrikas see the question differently. According to their sources, a
Prasafigika is one who absolutizes the prasahga methodology. It was Buddha-
palita, say the Svatantrikas, and not Nagarjuna who was the first Prasafigika,
and he was followed by Candrakirti, Santideva, and others.
The Svatantrikas' name means "self-styled, or independent, argumentation."
This means that although a system of argumentation (for example, the prasahga)
may be valid on one level, it does not necessarily follow that it is valid on all
levels. One must adapt his argumentation to correspond to a particular level
of truth, which would seem to follow better Nagarjuna's claim that he has no
position of his own, but adopts his opponent's presuppositions and logic to
point to their self-contradictory nature.
Poussin sums up their initial positions as follows:

Buddhapalita founded a school, the official school of Tibet (sic), which is

called the Prasaiigika, and, one believes, is well within the lineage of the thought
of Nagarjuna. The Madhyamika, or 'man of the middle', with nothing to
affirm, nothing to negate; he has neither thesis, nor argumentation, nor exam-
ple. The only procedure available to him to destroy adversary doctrines, all
doctrines, is the prasahga, reasoning ad hominem to absurdity... Thus he
establishes emptiness. His position is pure criticism. Bhavaviveka demon-
strated the weakness of this methodology and replaced the prasahga of Buddha-
255

palita with formal reasoning, a svatantranumana:'From ultimate reality, per-

ception is not self-born...' ('En verite vrai, l'oeil ne nait pas de soi...') The
name Svatantrika is given by scholars who trace themselves back to him.10
Poussin's observation is formal: that whereas the Prasaiigikas content them-
selves with the negation of all views as their modality for the cognition of
sunyatd, the Svatantrikas want to demonstrate sunyata by positive argumen-
tation.
Yuichi Kajiyama sees the fundamental question separating the two schools
more philosophically:

Although yearning for the absolute truth is naturally accompanied by negation

of the relative and conditioned knowledge... (a) question should in this context
be reflected upon; that is, whether the system of the relative knowledge can be,
so far as the phenomenal world is concerned, recognized as valid or not, though
it is always delusive from the absolute point of view. This very problem seems
to have been a fork which divided... the Madhyamaka itself into the Prasafigika
and the Svatantrika.1
What is in question here is the very nature of the relative (samvrtti) itself, and
its relation to the ultimate (paramartha). Is the absolute to be found only in
the negation of the relative, or is the stuff of the relative somehow contiguous
to the absolute, allowing one to attain cognition of the absolute by means of
the relative? Bhavaviveka employs syllogism in order to state that the two
truths are contiguous, as we shall see later. However the question is not, as
some scholars would have us believe,12 of the implicit value of logic or syllo-
gism. The question, rather, is of the continuity of samvrtti and paramirtha.
There were divisions within the Svatantrika school itself. As dKon.mchob.
'jigs.med.dbang.po. writes in the Grub.pa'i.mtha'i.rnam.par.bzhag.pa.rin.po.
che'i.phrang.ba.:
If one divides it, there are two (that is) the Yogacara-Svatantrika-Madhyamaka
and the Sautrantika-Svatantrika-Madhyamakaamong them. The Madhyamika
who maintains self-awareness (rang.rig.; Skt. svisamvedana) while rejecting
the external object, has the character of the first, namely Acarya Santiraksita
... The Madhyamika who rejects self-awareness but admits the external object
established through its particular (svalaksana) has the latter characteristic,
namely Acarya Bhavaviveka. This is a further explanation of terms: As far
as the most basic points are concerned, by reasoning of maintaining agreement
with the Vijnanavadin, he is called a Yogacara-Madhyamika. By reasoning of
maintaining the external object as an aggregate of atoms, in the manner of
the Sautrantikas, he is called a Sautrantikacara-Madhyamika.13
The differences between the two subschools of the Svatantrika are not signifi-
cant for the purposes of our discussion. On fasc. 26a of the same work, we find
the conclusion that they differ only with regard to the status of self-awareness
(rang.rig.) and of external objects. Both of these categories apply to the level
of samvrttionly, and both maintain the ultimate unspeakableness ofparamdrtha,
and the role of logic and syllogism in general. As Poussin rightly notes, "... the
256 Katz

distinctionis not carried to the conception of ultimate truth, but to their

theoriesof relativetruth."14
Sgntiraksita,the founderof the Yogacara-Svatantrika-Madhyamaka school,
livedfrom705-762 C.E. andplayeda greatrolein theestablishmentof Buddhism
in Tibet. He was one of the first Indian iciryas to teach there, and it was at
his suggestionthat Guru Padmasambhavawas invited to Tibet. His disciple,
Kamalasila(713-793), representedthe Indian (Svatantrika)position in the
great debate of bSam.yas.(792-794), which shall be discussedin detail later,
and establishedthe Svatantrikanormfor Madhyamakaphilosophyamongthe
earlylineagesof TibetanBuddhism.
IV. THE TWO TRUTHS

The questionof the relationshipbetweenthe two levels of truth,samvrttiand

paramartha,is of utmost importancein understandingMadhyamakaphilos-
ophy in general and the Svatantrika-Prasangika debates in particular.Two
key versesof Nagarjunaunderscorethe vitalityof the question:
ye 'nayornavijanantivibhagamsatyayordvayoh/
te tattvamna vijanantigambhirambuddhasasane//
Those who do not know the distinctionbetweenthe two truthscannotunder-
stand the profoundnatureof the Buddha'steaching.
vyavaharamanasritya paramarthana desyate/
paramarthamanagamya nirvanamnadhigamyate//
Without relyingon everydaycommon practices(that is, relativetruths),the
absolutetruth cannot be expressed.Withoutapproachingthe absolutetruth,
nirvanacannot be attained.15
Thepoint whichwe wishto makein citingthesetwo versesis thatapparently
the Prasafigikaschool concentrateson the former, while the Svantantrika
school relies more heavily on the latter. For the Prasafigikas,paramarthais
utterly beyond constructed thought; they "... stress the contradictions between
absolute realityand the human attitudeof understanding,which constitutes
the groundof logic."16Becauseof this assumption,theyclaimthatparamartha
is the absolute negation of samvrtti.('Absolute negation' means negation
without counter thesis, or, the negation of A does not imply B. 'Relative
negation', on the other hand, means negation from a position, that not A
impliesB.)
The Svatantrikas,on the other hand, follow more closely verse 10 above,
the chargethatparamarthacannotbe expressedwithoutsamvrtti.Due to their
understandingof the contiguityof the relationshipbetweenparamarthaand
samvrtti,the Svatantrikasseekto establishparamirthanot only by the negation
of samvrtti,as do the Prasafigikas,but also by positive argumentationof the
syllogisticform. As HerbertV. Guentherobserves:"Whatthe Madhyamika-
Svatantrikaswanted to emphasizewas that all human experience,inasmuch
as it is experienceand not mere propositionsor the like, is an insight into
257

reality, an awarenessof coherence which is not its own authenticationof

reality, but reality itself."17This validity of all human experienceof which
Guenther speaks is a way of expressingthe contiguity of paramirthaand
samvrtti,a way of remindingus that paramirthais not somethingabove and
beyond our experience,but a way of experiencingreality directly (yathib-
hutadarsana),insofar as it is not 'mereproposition'(drsti).
Kajiyama offers a helpful, though somewhat misleading,analysis of the
problem. He writes that "... one of the most important subjects of the Mad-
hyamakaphilosophy... is if the absoluteis immanentin the phenomenalworld
or transcendsit, in other words, if the two worldsare absolutelydisparateor
can have any kind of logical connection."18We do not believethat any good
Madhyamikawould care to speculateabout any immanenceor transcendence
of worlds,but the point to be discussedis whetherparamirthais transcendent
to or expressiblein logico-linguisticconstruction;and if so, to what extent.
Both Bhavavivekaand Candrakirtiagree that the highest truth is beyond
speech.19But Bhavavivekaadds the most interestingcategoryin his system-
atization (vibhiga) of the two truths: he divides paramartha into aparyiya
(that paramartha which could not be inferred) and paryiya (that paramartha
which could be inferred), the former having a higher position than the
latter.20 This is done to express the contiguity of the two truths; it gives ".. .a
kind of logicality to the relation of the absolutenessand the ground of our
delusion."21
Some scholarssee this systematizationas a concessionto the Vijianavadins,
leading them to label the Svatantrikamovementas, unhappily,'synchronis-
tic'.22 But whereas the Vijiianavadinsare speaking of a metaphysicalim-
manence,Bhavavivekais merelytryingto call to our attentionthe value of the
teachingsof Buddhismfor our spiritualdevelopment,and is not formulating
yet anotherdrsti. ShotaroIida makesthis point quite clearly:

Bhavavivekagrades ultimate reality into two kinds, i.e., supremundane-

ultimate-realityand mundane-ultimate-reality.
The former has no attributes
(nirlaksana)and is inexpressible.However,the words and deeds of the arya,
who had some experienceof paramartha,differfrom those of the worldlings.
This is the meaning of 'prstha-labdha'(=parydya-paramartha).In other
words, the words and deeds of the arya based on ultimatereality should be
pure and true knowledgeof the world (tathya-samvrtti-jiiana).23
Despite Iida's unfortunatetranslationof paryaya-paramartha as 'mundane-
ultimate-reality',his point is well taken. Even Murti, who exhibits a strong
biasfor the Prasafigikaas the only truerepresentative
of Nagarjuna'steachings,
agrees that Bhavaviveka was trying to rescueBuddha's wordsfrom the relent-
less negation of the Prasafigikas:"In his MadhyamakdrthaSahgraha,he
accordsto the Absoluteof Hinayanaandhereticalsystemsthe statusof paryaya
paramartha.This makes him out as a liberal-mindedMadhyamikaunlike
Candrakirti.' 24
258 Katz

We feel that this attempt by Bhavaviveka is a well-taken one. Certainly

there is a difference, as Iida suggests, between the words of the sage and the
words of the common man. Whereas the latter may be said to be drsti, or mere
construction, the latter have aided devotees in liberation. If such is the case,
of what need is their negation? Would it not somehow be more true to speak
about sarvadharmasunyati than to make ontic-ontological claims about the
existence of dharmas? The difference is that of karma (bondage-producing
activity) and Buddhakarma (spontaneous openness). Bhavaviveka's criterion
for differentiating between the two types of words, or paryaya-paramdrtha
and samvrtti, is simply causal efficacy (kriya-kdra-sdmarthya).25When Can-
drakirti goes so far as to claim that only Madhyamikas could attain nirvana,
his position smacks of dogmatism rather than dialectic.26
Bhavaviveka, on the other hand, wants to affirm logically the veneration due
to the irya of another school. As dKon.mchog.'jigs.med.dbang.mo. writes:
"When the sravaka and the pratyekabuddhaare arhats, they integrate (them-
selves) by entering the path of the Mahayana, because they (= a school)
maintain the culmination in the one ultimate vehicle (conducive to enlighten-
ment)."27 For the Svatantrikas, then, the idea of ekayana means that nirvana
is possible regardless of school or formulations; for the Prasafigikas, this high
ideal is reduced to the dogmatic insistence upon the realization of sunyata as
the absolutely necessary condition for the attainment of nirvana.28
The Prasafigikas got into trouble because of the thoroughgoingness of their
negation. Although the Svatantrikas would agree with them that in the long
run negation must become absolute (that is, without offering a counter thesis),
as Bhavaviveka's aparyaya-paramartha-satyaand Candrakirti's paramartha-
satya are both beyond any specification, Bhavaviveka's category of paryiya-
paramdrtha rescues him from the charges of nihilism, which the critics of
Madhyamaka launch. The problem for the Prasafigikas (who, by definition
of the Svatantrikas, are those who absolutize the prasahga methodology) is
that consistent application of the prasahga to samvrtti leads to the denial of
samvrtti. If, as Nagarjuna tells us in MMK, XXIV, 10, samvrtti is our only way
of expressingparamartha, then are we not left at a loss in which we must negate
even paramirtha ? Although the Prasafigika would respond that we are negating

CANDRAKIRTI'S SYSTEMATIZATION
SATYA

Paramartha Samvrtti
(unspeakable
absolute truth) loka-samvrtti aloka-samvrtti
(real empirical (unreal empirical
truth) truth)
from Midhyamakivatira
259

BHAVAVIVEKA'S SYSTEMATIZATION

SATYA
I I
Paramartha Samvrtti

Paryaya- Aparyaya- Mithya- Tathya-

paramartha paramartha Samvrtti samvrtti
(speakable (unspeakable
ultimate truth) ultimate truth)

Jatiparyaya- Janmarodha- Sakalpa- Akalpa-

vastu-paramartha paramartha mithya- mithya-
samvrtti samvrtti

from Midhyamarthasahgraha

only views (drsti) and remaining nobly silent about reality, are we still not in
the dilemma of having no way of teaching?
This type of debate was carried on between Nagarjuna and Harivarman, the
founder of the Satyasiddhi school, who agreed that concepts cannot adequately
express reality but who taught that nirvianais cessation. C. D. C. Priestley
recapitulates:
Harivarman evidently thinks that the prasahga of the nihilist [sic] leads him to a
denial of conventional truth; and as Harivarman and Nagarjuna both realize,
conventional truth cannot consistently be denied, since the denial itself must
have at least conventional existence. The prasahga, then, seems to be too
wholesale in its effect: although it certainly can put an end to the depredations
of heterodoxy, it is liable to devour also the domestic concepts of Buddhism
which it was meant to protect. Nagarjuna is of course not unaware of this
danger; his Vigrahavydvartan7contains a detailed reply to what is essentially
Harivarman's objection. But even if Harivarman had seen and accepted
Nagarjuna's defense, he would still have been obliged to reject the prasahga.
For in trying to maintain simultaneously the reality of cessation and the reality
of non-existence, he involves himself, as we have seen, in precisely the kind of
inconsistencies that the prasahga is designed to expose.29
It was for Bhavaviveka to come and rescue the Madhyamaka from the dead
end of overzealously applied negation, to logico-linguistically detranscenden-
talize paramartha into the realm of the speakable. His methodology of doing
so was syllogistic argumentation, which he largely adopted from Dignaga with
some important revisions.
V. SYLLOGISM

According to Dignaga, syllogism is inference for others. It is not a source of

knowledge in the classic sense of perception (as distinct from recognition)
and inference. As Th. Stcherbatsky recounts: "When an inference is communi-
260 Katz

cated to another person, it then is repeated in his head, and only in this meta-
phorical sense can it be called an inference. Syllogism is the cause which
produces an inference in the mind of the hearer. Its definition is, therefore, the
following one-'A syllogism consists in communicating the Three Aspects of
the Logical Mark to others'."30 The Three Aspects are: minor premise (paksa-
dharmatva); major premise (anvaya); and the counterposition of the major
premise.31 Or, as expressed by Dharmottara: "Communicating the three
aspects of the logical mark, that is, (the logical mark appears here also in)
three aspects which are called (respectively) direct concomitance (or major
premise expressed positively) [anvava], its contraposition (or the same premise
expressed negatively) [vyatireka], and (the minor premise of) the fact of the
presence of that mark in the subject (of the inference, that is, the fact that the
subject of the inference is characterized by the logical mark) [paksa-dharmat-
va]."32
Karl Potter expresses the syllogistic paradigm thus:

Hypothesis: That mountain (is) fire-possessing (paksa)

Reason. (Because) that mountain (is) smoke-possessing (hetu)
Examples. (a) (as in) kitchen (sa-paksa)
(b) (unlike) lake (vi-paksa)33
The variant syllogistic formulations and controversies between the Buddhists
and Naiyayikas have been fully treated by Stcherbatsky,34 so we will confine
our discussion to the adaptations made by Bhavaviveka from this standard
form offered by Potter.
A. K. Warder states Bhavaviveka's syllogism as follows:

Ultimately (at the level of ultimate truth) the synthesized phenomena are empty
(of any own-nature), because of their conditioned origination (middle term),
as things illusorily created (example: that is, works of art, paintings, clay
models and the like do not have the real nature of the things they represent-
women, elephants and so on). At the concealing level, on the other hand, the
phenomena commonly accepted may be admitted. We do not contradict the
experience of the world but say that ultimately the phenomena of the experience
are not real.35

Iida represents the argument as follows:

Hypothesis: Earth, etc. (is) not own-being-possessing (paksa)

from the standpoint of ultimate reality
Reason. (Because) earth, etc. (is) (hetu)
(a) manufacture-possessing
(b) cause-possessing
Example. (like) knowledge. (sa-paksa)36
We at once notice some striking departures from Potter's formulation. In
the first place, we note the accretion of "from the standpoint of ultimate reality"
261

to the hypothesis, as we find in all of his syllogisms. This is because, as Kajiyama

writes, "His logic is logic of paramirtha, which criticizes the ground of logic
of the practical world, viz., the laws of identity, contradiction, excluded middle,
causality, etc., and the criticism has been done through the peculiarities of
his syllogistic form."37 Unlike the Prasanigikaswho criticize from the internal
inconsistency of the opponent's argument, Bhavaviveka criticizes by offering
his own hypothesis from the standpoint of ultimate reality. Kajiyama claims
that the Prasafigikas miss the essence of the contradiction when they restrict
themselves to internal criticism; that the real contradiction lies in the relation-
ship of samvrtti to paramartha, and that this is the true meaning of absolute
negation. He writes:

When we argue the transcendental contradiction arises not from logic itself,
but from the disparity between absoluteness and the ground of logic. The
transcendentality of the paramartha is nothing but the contradictory relation
of the paramartha and the samvrtti. If this is granted, cannot we speak of the
absolute reality through the logic of contradictions? The contradiction is not
merely illogicality but the unique method which can reduce to the absolute
reality our world, which is the human logicalization of the Absolute. Nagarjuna
and the Prasangikas can be accused for their negating logicality without strictly
showing the real contradiction. It is not efficient to condemn logic merely
standing on transcendence of the paramirtha. For Bhavaviveka to use the logic
of contradiction in the place where samvrtti and the paramirtha meet together
is methodological completion of the absolute negation of the Madhyamaka
philosophy. Bhavaviveka did not wildly fit in logical tendency of the age, but
he did, observing the traditional method, the same exertion in the Madhyamaka
theory as Dignaga did in the Vijnanavada.38
We quote this passage with the reservation that paramirtha is understood as
logico-linguistically transcendental to samvrtti in the Prasafigika school.
The fundamental contradiction of which Kajiyama speaks is not mere
illogicality on the part of the opponent but must be absolute negation, in the
sense of our modality of being in the world. Thus Bhavaviveka seeks to restore
teachings of other Buddhist schools as paryaya-paramartha,the meeting point
of paramartha and samvrtti, and his only method of establishing the fundamen-
tal (absolute) contradiction is by means of his syllogism.
We also notice in comparing Bhavaviveka's formulation of the syllogism
with Potter's standard form that Bhavaviveka has a negative hypothesis. Since
Bhavaviveka has unearthed the fundamental contradiction, any negation which
begins on the level of samvrtti must proceed to the level of paramartha. For the
same reason, his syllogism lacks the vi-paksa (negative example) component:
his negation is absolute, and, therefore, he cannot offer a contradictory exam-
ple. Although, following Potter's formulation, we can offer the negative example
of the lake which is not fire-possessing, we cannot offer any vi-paksa for Bha-
vaviveka's transcendental syllogism, which is not not-own-being-possessing
without invalidating the entire syllogism. After all, the syllogism must conform
to experience, and because Bhavaviveka begins with a negative hypothesis
262 Katz

from the standpoint of ultimate reality which he intends to demonstrate,

employing vi-paksa in this case would be absurdity.
In conclusion of this discussion of Bhavaviveka's syllogism, we may say
that his category of paryiya-paramdrtha served two purposes: (1) it rescued
early Buddhist teachings from the counterproductive negation, which in an
important sense is inconsistent; and (2) it pointed to the fundamental con-
tradiction of paramdrtha and samvrtti, which is not the accident of illogicality
(as with the Prasafigikas), but logically demonstrates the emptiness of drsti
inductively. In this specific sense, it may be said, as so many scholars have,
that Bhavaviveka sought to 'prove' isnyata logically.
VI. CONFRONTATIONAT BSAM. YAS

The great debates held at the monastery of bSam.yas. (792-794 C.E.) between
representatives of the Indian and Chinese Buddhist traditions are well-known
to scholars as shaping the form that Buddhism was to take in Tibet. We intend
to demonstrate that these debates were held between representatives of the
Svatantrika and Prasafigika schools, the former the Indian icdrya, and the
latter, the Chinese. After demonstrating this by consulting Tibetan historical
records, we shall review some of the issues at point, showing the practical
application of Bhavaviveka's principles. A word about the continuity between
the Svatantrikas and the Tantrikas will also be added.
Most scholars assume that the Chinese debator was of the Ch'an (Zen)
school, probably due to the popularity of that sect in some regions of China
during the eighth century. The bSam.yas. debates did take place about one
hundred years after the death of Hui-neng, during the rise of his Southern
School of Ch'an.39
This historical coincidence has misled such scholars as Warder into saying:
In due course he (Kamalasila) was invited to Tibet, whose Buddhists had
become divided, in fact because simultaneously with the missions of Indian
Buddhists there the Chinese Buddhists of the Dhyana (Ch'an, Zen) school
were spreading their own version of the Buddhist teaching.40

and caused Robinson to remark:

A Chinese Ch'an master named Mahayana came to the court and made many
converts. A debate was held before the king in 794 or thereabouts, in which
Santaraksita's school vigorously attacked the Ch'an faction and succeeded
in getting the king to banish the Chinese. Despite continual intercourse with
the Chinese, Tibetan Buddhism has ever since been based on that of India.41
Both Robinson and Warder are wrong, probably due to their lack of familia-
rity with Tibetan historical materials. The contestants in the debate were
Kamalasila, disciple of Santiraksita, the founder of the YogScara-Svatantrika-
Madhyamaka, and an anonymous representative of the "Chinese Fa-shang
Mahayana" school.42 "Fa-shang"is the Taoist term for monk, literally "home-
263

leaver"; "Mahayana" is not the name of the debator, but a designation of his
general philosophic stance. Bu.ston. is no more specific about this individual
than that.
However if we see what else Bu.ston. has to say about this "Chinese Fa-shang
Mahayana" school, we might be able to deduce their identity. All he says about
their beliefs is that "These favored nihilistic views and did not exert themselves
in the practice of virtue."43 We also know that their numbers had been
increasing among the Tibetan court. It is rather unlikely that a Tibetan chron-
icler would refer to the Ch'an school as nihilistic; this epithet had generally been
reserved in Buddhist literature as a perjorative against the Madhyamaka.
In fact, this charge was often levelled against the Madhyamaka by the
Sarvastivadins, who had long before this time established themselves in Central
Asia and had been exerting influence in Tibet. However, since Santiraksita and
Kamalasila were also Madhyamikas, and Bu.ston.'s text is unreservedly
laudatory of them, it is logical that this term would only be used against their
Madhyamaka rivals, the Prasafigikas.
Fortunately, Bu.ston. gives us more information than this. He also tells us
that the "Chinese Fa-shang Mahayana" were specialists in two texts: the
Shes.rab.kyi.pha.rol.tu.phyin.pa.stong.phrag.pa.(Satasahasrikasutra) and the
bSam.gtan.nyal.ba'i.'khor.lo. (Dhyana-svapna-cakra).44 While we have found
no reference to this second text elsewhere in Tibetan annals (possibly it was
composed by the Chinese themselves), the former prajnaparamita work is
repeatedly attributed to Nagarjuna himself,45 making it a Madhyamaka text.
Thus the primary texts of the "Chinese Fa-shang Mahayana" were Mad-
hyamaka, leading us to conclude that the debator himself was a Madhyamika.
It is quite clear that the Chinese version of Madhyamaka was Prasaiigika and
not Svatantrika. The founder of this Chinese school, which they called San-lun
(The Three Treatise School), was Kumarajiva (344-413 C.E.), who lived a
century before the formulation of the Svatantrika by Bhavaviveka. Their beliefs
are described by Junjiro Takakusu thus: "The truth can be attained only by
negation or refutation of wrong views within and without Buddhism and of
errors of both the Great and Small Vehicles.... Refutation-and refutation
only-can lead to ultimate truth."46 and by Wing-Tsit Chan thus:
To this school, refutation of all erroneous views is essential for and indeed
identical with the elucidation of right views. But when a right view is held in
place of a wrong one, the right view itself becomes one-sided and has to be
refuted. It is only through this dialectic process that Emptiness can be arrived
at, which alone is free from names and character and is 'inexplicable in speech
and unrealizable in thought'. The specific method in this dialectic process is
Nagarjuna's Middle Path of Eightfold Negations....47

Quite obviously these are clearly Prasafigika positions and not Svatantrika, as
it is a central theme of Bhavaviveka that one can establish paramartha by
264 Katz

positive argumentation since the fundamental contradiction is not the accident

of illogicality, but the relation ofparamartha to samvrtti. For these reasons we
conclude that Kamalasila's opponent at bSam.yas. was a Prasangika.
The issues in conflict between Kamalasila and this anonymous Chinese
Prasaiigika were fundamentally two: the negativistic attitude of the Prasafigika
and his insistence upon a sudden, rather than a gradual, path to enlightenment.
Kamalasila skillfully argued that the negation of samvrtti must, consistently
speaking, lead to a negation of paramdrtha, or a nihilism of sorts. Offering a
counterposition, he claimed that one realizes the nonsubstantiality of dharmas
(sarvadharmasunyata)by positive application of logic and intellect. This en-
counter is recorded by Bu.ston. as follows:
Then the Fa-shang spoke: If one commits virtuous or sinful deeds, one comes
to blissful or to evil births (respectively). In such a way the deliverance from
samsara is impossible, and there will always be impediments to the attainment
of Buddhahood. (The virtuous and sinful deeds) are just like white and black
clouds which alike obscure the sky. But he who has no thoughts and inclinations
at all can be fully delivered from phenomenal life. The absence of any thought,
search, or investigation brings about the non-perception (mi.dmigs.pa.;
anupalambha)of the reality of separate entities. In such a manner one can attain
Buddhahood at once, like a Bodhisattva who has attained the tenth stage.
To this Kamalasila himself answered as follows: Thou sayest that one ought
not to think about anything whatsoever. But this means the negation (or
rejection) of highest analytic wisdom (shes.rab.; prajfii) likewise. Now as the
latter represents the foundation of the divine wisdom of a saint, the rejection
of it necessarily leads to the negation of this sublime transcendental wisdom.
If analytic wisdom is absent, what meditator can come to abide in a state where
there is no constructive thought? If one has no thought concerning any of the
elements of existence (chos.; dharmas)and does not direct the mind upon them,
this does not mean that one can cease to remember all that one has experienced
and to think of it. If I think: 'I must not recall in my mind any element of
existence', such a thought will itself be an intense recollection and activity of
the mind. If the mere absence of (consciousness and) recollection is regarded as
sufficient, it follows that in a swoon or at the time of intoxication one comes to
the state where there is no constructive thought. Now, (in reality) without
correct analysis there is no means of attaining the liberation from constructive
thought. If we merely cease to reflect and have no discrimination, how can we
come to the cognition of the non-substantiality, it is impossible to remove the
obscurations. Therefore, the incorrect representation can be cast away only
by means of the correct analytic wisdom. For this reason, it is not proper to
say that one does not reflect, when in reality it is the reverse. Without recollection
and correct activity of the mind, how can one come to remember the place of
former residence and attain omniscience? But the yogin who reflects over an
object of correct analytic wisdom, cognizes all the external and internal elements
in the present, past, and future as non-substantial, has all thought-constructions
pacified within him, and rejects all evil doctrines. On this foundation he becomes
skillful in expedience and is the manifestation of highest wisdom.48
Kamalasila is presenting here, couched in terms related to religious practices,
many of the same ideas which Bhavaviveka presented three centuries earlier.
His is affirmingparydya-paramarthaagainst the Prasafigika negation. He asks
us to arrive at the cognition of the nonsubstantiality of dharmas by means of
265

discrimination,much as Bhavavivekaarrives at paramirtha syllogistically,

ratherthanby castingawaylogic, whichis the negationof samvrttiby meansof
unearthinginternalinconsistency(prasahgamethodology).He likens his op-
ponent'snegationto a swoon or intoxicationand reliesupon analysisbasedon
offeringcountertheses.He also points to the fundamentalcontradictionrather
than being satisfiedwith the accidentalcontradictionof the illogicalityof the
opponent: "... the incorrect representation can be cast away only by means of
correctanalyticwisdom. For this reason,it is not properto say that one does
not reflect,when in realityit is the reverse."49
VII. A WORD ON THE CONTINUITY FROM SVATANTRIKATO TANTRA

Theeffectsof thisdebateon the shapeof Buddhismin Tibetweremoreprofound

than many imagine. Santiraksita,who instructedKamalasilafor this debate,
also recommendedto the king that Guru Padmasambhavafrom U.rgyen.
(Uddayana;Swat Valley)be invitedto Tibet.Padmasambhava,of course,was
the greatTantricmasterwho was largelyresponsiblefor the popularizationof
Buddhismin Tibet and was held in specialreverenceby the oldest of the four
Tibetanlineages(the rNying.ma.pa.).
We offer that the link betweenthe Svatantrikasand the Tantrikasis more
than historical.One of the fundamentalprinciplesof the tantrasis that one
shoulduse the meansof sarimsrafor the attainmentof nirvana.We arereminded
here of Bhavaviveka'scontentionthat we can use the languageof the relative
to denote the ultimate, following from MMK, XXIV, 10. It seems that the
Prasafigikascouldnot allowfor this,contentingthemselveswithdemonstrating
that paramdrthais beyond all logico-linguisticconstruction.Thus a Tantrika
sayingthatwe can use the meansof sarhmsra to attainnirviinais but a jumpfrom
Bhavavivekasaying that we can use the language of samvrttito express
paramdrtha.

NOTES

1. Buddhist Thoughtin India (Ann Arbor: University of Michigan, 1967), p. 239.

2. A Survey of Buddhism(Bangalore: Indian Institute of World Culture, 1966), p. 346.
3. A Survey of Buddhism,p. 348.
4. The Central Philosophy of Buddhism(London: George Allen & Unwin, 1960), p. 132n.
5. "Bhavaviveka," (Brussels: Melanges Chinois et Bouddhiques,2, July, 1933, pp. 60-69), p. 66.
My translation. This error is understandable, since the dominant dGe.lugs.pa. lineage accepts the
Prasangika standpoint. However, at the time of these debates, this school was not in existence. The
earlier lineages were, and still are, inclined toward the Svatantrikas.
6. The Buddhist Conception of Nirvana (Varanasi: Bharatiya Vidya Prakashan, 1968), p. 67.
Stcherbatsky's error here needs no further comment.
7. "Some Logical Aspects of Nagarjuna's System," Philosophy East and West (5, no. 4, January,
1957, pp. 291-308), p. 292.
8. Poussin, "Bhavaviveka," p. 65.
9. Emptiness. a Study in Religious Meaning (Nashville, Tenn.: Abingdon, 1967), p. 36n.
10. Poussin, "Bhavaviveka," p. 66.
266 Katz

11. "Bhavaviveka and the Prasangika School," (Nava-Nalanda-Mahavihara Research Publica-

tion, vol. 1, n.d.), p. 291.
12. dKon.mchod.'jigs.med.dbang.po., Grub.pa'i.mtha.rnam.par.bzhag.pa.rin.po.che'i.phrang.ba.
(in S. Iida, "An Introduction to Svatantrika-Madhyamika," Ph.D. diss., University of Wisconsin,
1968), fasc. 22b-23a: de.la.dbyas.na. / rnal.'byor.spyod.pa'i.dbu.ma.rang.rgyud.pa.dang. / mdo.
sde.spyod.pa'i.dbu.ma.rang.rgyud.pa.gnyis. / phya.don.gas.mi.len.zhing. / rang.rig.khas.len.pa'i.
dbu.ma.pa.de.po'i.mtshan.nyin.mtshan.na.gzhi.ni. / slob.dpon.zhi.ba.'tzho.bu. /... rang.rig.khas.
mi.len.zhing. / phyi.don.rang.gi.mtshan.nyid.kis.grub.pa.khas.len.pa'i.mtshan.nyid. / mtshan.gzhi.
ni. / slob.dpon.legs.ldan.'byed.lta.bya. / sgra.bshad.kyang.yod.de. / gzhi'i.rnam.bzhang.sems.can.
pa.dang.mthun.par.khas.len.pas.na.rnal.'byor.spyod.pa.dang. / mdo.sde.pa.ltar.rdul.phra.rab.
bsags.pa'i.phya.rol.gyi.don.khas.len.pa.mdo.sde.spyod.pai.dbya.ma.pa.shes.brjod.do. // Vide also
Khai.dub., Stong.thun.bskal.bzang.mig.'byed.,Tsang ed., fasc. 37a sq.
14. Poussin, "Bhavaviveka," p. 67.
15. Nagarjuna, Miilamadhyamakakiriki, trans. K. K. Inada, (Tokyo: Hokuseido, 1970), XXIV,
9-10.
16. Kajiyama, "Bhavaviveka and the Prasangika School," p. 299.
17. BuddhistPhilosophy in Theory and Practice (Baltimore, Md.: Penguin Books, 1971), p. 125.
18. Kajiyama, "Bhavaviveka and the Prasangika School," p. 306.
19. Vide Bhavaviveka, Madhyamarthasahgraha,and confer Candrakirti, Midhyamakdvat a.
20. See charts, p. 259.
21. Kajiyama. "Bhavaviveka and the Prasangika School," p. 300.
22. See A. K. Warder, Indian Buddhism (Delhi: Motilal Banarsidass, 1970), pp. 474 sq., and
Sangharakshita, A Survey of Buddhism, p. 347.
23. S. Iida, "An Introduction to Svatantrika-Madhyamaka," Ph.D. dissertation University of
Wisconsin, 1968, p. 244, fn. 16.
24. Murti, The Central Philosophy of Buddhism,p. 98.
25. See lida, "An Introduction," p. 257.
26. Candrakirti, Madhyamikakirika- Vrtii, pp. 351-3: tad ayam aca yo yathaivamvidhe visaye
nacarya-matanuvarti tatha pratipaditam Madhyamakavatare 'diiranigamayiamtu dhiyadhika ity
atreti na punas tad dusane yatha asthiyate. Quoted in Murti, The Central Philosophy of Buddhism,
p. 96, n. 3.
27. dKon.mchog.'jigs.med.dbang.mo., op cit., fasc. 25a: nyam.dang.dgra.bcam.pa.yin.na.
theg.cen (sic, should read chen.) lam.du.'jug.pas.khyab.ste. / mthar.thug.theg.pa.gcig.tu.grub.par.
'dod.pa'i.phyir. //
28. See Murti, The Central Philosophy of Buddhism,p. 96.
29. "Emptiness in the 'Satyasiddhi'," Journal of Indian Philosophy (vol. 1, no. 1, pp. 30-37),
pp. 36-37.
30. Buddhist Logic (Leningrad: Biblioteca Buddhica, 1932), 1, p. 275.
31. BuddhistLogic. 2. p. 109.
32. Nydya-Bindu, 41.3, in BuddhistLogic,2pp. 109-110
33. Iida, "An Introduction," p. 246, n. 26.
34. BuddhistLogic, 1, 275-319; 2, 109-253.
35. Warder, Indian Buddhism,p. 475. From Bhavaviveka's Karalaratna.
36. Following lida, "An Introduction," p. 246n.
37. Kajiyama, "Bhavaviveka and the Prasaiigika School," p. 305.
38. Kajiyama, "Bhavaviveka and the Prasangika School," p. 300.
39. See Kenneth Ch'en, Buddhismin China (Princeton, N.J.: Princeton University Press, 1964),
pp. 355-357.
40. Warder, Indian Buddhism,p. 477.
41. R. H. Robinson, The BuddhistReligion (Belmont, Calif.: Dickenson, 1970), p. 110.
42. Bu.ston., Chos.'byung. (trans. E. Obermiller [Heidelberg: Materialien zur Kunde des
Buddhismus, 1931]), fasc. 142a.
43. Bu.ston., Chos.'byung., fasc. 142a.
44. Bu.ston., Chos.'byung., fasc. 142b.
45. See Bu.ston., Chos.'byung., fasc. 22a, fasc. 1la, and fasc. 131a.
267

46. TheEssentials of BuddhistPhilosophy (Honolulu, Hawaii: Office Appliance Co., 1956), p. 101.
47. A Source Book in Chinese Philosophy (Princeton, N.J.: Princeton University Press, 1963),
p. 359.
48. Bu.ston., Chos.'byung., fasc. 142a sq.
49. Idem.