On the Paradoxical Method

·期刊原文
ON THE PARADOXICAL METHOD OF THE CHINESE MAADHYAMIKA: SENG-CHAO AND THE CHAO-LUN TREATISE
Journal of Chinese Philosophy
Vol.19
1992
P.51-71
Copyright (c) 1992 by Dialogue Publishing Company,
Honolulu, Hawaii, U.S.A.

--------------------------------------------------------------------------------
.

P.51

I.

Opinions vary among scholars as to whether
Chinese Buddhists correctly understood the Indian
Madhyamika religio-philosophical system, and
equally, whether the Chinese understanding of
Buddhist religiosity as a whole was and has been
authentic. This question, whether or not the
Buddhist religio-philosophical system, comprising
the conceptions of dharma, 'suunyataa, praj~naa,
nirvaana, and so forth, was communicated across
barriers of cultural differences between peoples as
divergent as the Chinese and Indian, can be answered
in a variety of ways. I hold that the Mahayana
Buddhist insight of 'suunyataa (i.e. emptiness or
voidness), with which the Madhyamika method of
dialectic was exclusively concerned, is
transcultural; that by the time of the fifth
century, the Chinese Buddhists had come to
understand it in such precise terms that it became
the religio-philosophical foundation for their own
subsequent development; and that this achievement
was prompted by their intellectual or
religio-philosophical concern over the nature of
logic and language. The Chao-lun(a), which is
ascribed to Seng-chao (374-414) , is the best
evidence to substantiate the above views.
The Chao-lun or The Treatise of Seng-Chao, con-
sisting of four short essays and two epistles
written on Buddhist doctrines, is thought by modern
scholarship to be the earliest extant Chinese
Maadhyamika text composed by the native Chinese mind
and an imported land-mark in the early stages of
Chinese Buddhist thought formation. Once a student
of the

P.52

Taoism, Seng-chao had turned to Buddhism and was
trained under Kumaarajiiva, the greatest textual
translator. By assisting his master in the task of
translation and attending his teacher's lectures, he
became well versed in the Maadhyamika treatises and
developed a deep doctrinal understanding. The main
body of the Chao-lun essays consists of a series of
paradoxical arguments that parallel the
reducio-ad-absurdum arguments (prasanga-vaakya) of
the Maadhyamika treatises composed by Naagaarjuna
(ca. 2nd cent.) and his followers in India. The
primary purpose of this paper is to demonstrate why
Seng-chao's paradoxical method may have had a direct
bearing on Chinese logical and linguistic thinking;
and secondarily is to clarify both why the
Maadhyamika or Buddhist insight into `sunnyataa is
trans-cultural, and why Chinese logico-linguistic
concerns could have helped Seng-chao understand the
insights of `suunyataa.

II

In spite of a persistent scholarly interest in
Indian Maadhyamika Studies in the West, there has
not been a comparable interest in Chinese
Maadhyamika Studies. By "Chinese Maadhyamika
Studies," I mean the philosophical and doctrinal
investigation of expository works written by Chinese
Buddhist thinkers, among whom Seng-chao is to be
included together with other disciples of
Kumaarajiiva, and the writings ascribed to the
scholar monks who developed the San-lun School of
Buddhism in China. I believe the major reason for
this retarded interest in Chinese Madhyamika studies
can be attributed to the misconception that cultural
dissemination follows a general rule, that Buddhist
religiosity ought to be most authentic in the Indian
Buddhist cultural environment in which it was born,
and that the degree of authenticity will probably
diminish whenever it is transmitted to an alien or
different cultural environment. Since China and the
Indian sub-continent are historically regarded as
independent cultural zones, and despite the Chinese
acceptance of Buddhism, the aforementioned
misconception seems to have affected every
interpretation of Sino-Indian Buddhist transmission.
The study of the Chao-lun seems to have been
similarly affected; we have only three major
post-war works so far available to us: W.

P.53

Liebenthal's Chao-lun: The Treatise of Seng-chao,
comprising a full textual translation in English,
published in Peking in 1948 [a revised edition was
published by the Hong Kong University Press in
1968]; a group of Japanese scholars led by
Z.Tsukamoto published the Jooron no Kenkyuu (The
Study of Chao-lun), a full textual translation in
Japanese, published in Kyoto in 1955; and R.H.
Robinson's Early Maadhyamika in India and China, a
partial translation in English, published by the
Univ.of Wisconsin Press in 1967. I am, however,
somehow dissatisfied with the way each author has
approached matters, especially Seng-chao's
paradoxical method of argument. In spite of his wide
and deep knowledge of Chinese culture and
civilization, Prof.Liebenthal expressed skepticism
as to the degree of Seng-chao's understanding: "a
wall of misunderstanding separated the Chinese
understanding of Buddhism from that of Indians. "
(1) Robinson, however, dissented from Liebenthal's
conclusion and asserted that Seng-chao's understand-
ing of the Maadhyamika doctrine and its method is
fairly accurate, and yet, he too fell short of a
fuller understanding and failed to see the
implications of Seng-chao's paradoxical method: "
his argument is imitative and hence lesser in degree
than Indian masters." (2) I believe these scholars
may have believed that, as a general rule, cultural
dissemination is inevitably accompanied by a degree
of degeneration from the original form and quality.
The detailed textual and linguistic research of the
Japanese scholars is impressive, but their
translation seems to be over-whelmed by Taoistic
thought and jargon. This hinders the reader from
reaching a proper evaluation of Seng-chao's
treatise.
Having made these criticisms, I must say that I
have tried in several papers to develop a more
satisfactory approach to Seng-chao's religio-
philosophical insights into `suunyataa as well as
his dialectical method of paradoxical argument. The
underlying three premises have already been given in
my initial introduction. Along with these premises,
I intend here (1) to demonstrate why the paradoxical
method of argument Seng-chao uses in his exposition
of the insight of `suunyataa can be theoretically
ascertained as an authentic Maadhyamika method of
dialectic fit for a Chinese- speaking audience; and
(2) to show why this paradoxical method can be more
clearly analyzed in reference to the Chinese
language and the Mohist logic of ancient China. I
hold in hypothesis that Seng-chao knew the

P.54

Neo-Mohist system of logic, and that this knowledge
helped him formulate his own dialectical method in
terms of paradoxical argument.

III

In a series of research paper written in the
early 80's, I explored Chinese Maadhyamika thought
in parallel with that of its Indian counterpart. In
1983, I read a paper entitled "Is Seng-chao's method
of demonstration an authentic Maadhyamika? " (3) at
the 194th annual meeting of the American Oriental
society. I probed why Seng-chao's paradoxical
method has been regarded as an authentic Maadhyamika
dialectic. My thesis was that the Maadhyamika
dialectic as initially formulated by Naagaarjuna,
forefather of the school in India, shares the same
structural basis as an ordinary inferential process
of the mind, and that 'dialectical thinking' and
'logical thinking' derive from their respective
modifications of that common structural basis.
Moreover, the structural basis of inferential logic
formulated in ancient India is identical with that
which the Mohist logicians formulated in ancient
China, thus placing the workings of the human mind
beyond culturally different forms.
In India, the core operations of inferential
reasoning are seen as the dual principles of anvaya
(affirmative instantiation) and vyatireka (negative
instantiation). For instance, seeing a billow of
smoke over a hill, we infer the probable outbreak of
a fire there. Buddhist logicians, by minimizing the
extra-logical linguistic factors of syllogistic
inference, reduced essential mental operations to
three steps which are indispensable to valid
inference: (4)

(1) Position of a minor term "P" [ "bearing
smoke"] to a locus 'a' :(a) P
(2) Anvaya: Whatever similar locus 'x,' "bearing
smoke, " is likely to "bear fire" [="Q"]:(x)
[(x) P.(x)q]
(3) Vyatireka: Whatever dissimilar locus 'y,'
" bearing no fire, " is surely to " bear no
smoke" : (y) [ (y)-Q.(y)-P]

Since the initial step can be included in the second
step, the inferential process can ultimately be
reduced to the dual process of anvaya and

P.55

vyatireka operations. In an inference, these two
operations ought to be clearly separated in
reference to their respective, i.e. similar and
dissimilar, loci (sapak.sa and vipak.sa). This is
what I call 'Logical Context', which requires two
conditions: (1) a clear demarcation between the
class of things that can bear smoke as well as fire
and the class of things that can neither bear fire
nor smoke, for a logical fallacy otherwise ensues;
(2) the application of anvaya, on the one hand, by
verifying a given universal relation: "wherever
there is smoke, there is fire," in reference to the
class of similar loci, such as, a kitchen, and the
application of vyatireka, on the other hand, by
falsifying the same universal in reference to the
class of dissimilar loci, such as, a water tank,
which then ascertains the contraposition: "Wherever
there is no fire, there is no smoke."
As to the Mohist system of logic, it should be
sufficient to refer to such expositions as The
Development of the Logical Method in Ancient China
by Hu Shih (Shang-hai: 1928), The Later Mohist Logic
and Science by A.C.Graham (Hong-kong: 1978), and, in
addition, Janusz Chmielewski's series of articles:
"Notes on Early Chinese Logic (I- VI), " (5) Rocznik
Orientalistyczny, especially, No. III [ibidem 27, 1
(1963) 103-121]. Prof. Hu Shih translated the dual
operations as the t'ung (b) 'method of agreement'
and yi (c) 'that of difference' and explained that
the two methods combined can establish valid
inference which the Neo-Mohists regarded as the true
method of induction. (6) Graham's understanding of
these two methods does not fundamentally differ from
Hu Shih's. It was Chmielewski who compared
Neo-Mohist logic with its Indian counterpart and
demonstrated the important terminological
correspondences between Indian and Mohist
inferential structure.(7). Their difference is that
the principle of instantiation of Indian syllogistic
inference is absent from the Mohist system; the
Mohists apply the principle of hsiao (d8), a kind of
universal statement well established by induction.
The normative valid basis of inference, however,
does not affect the logic of inferrence process, and
is a correlative factor within the differences
between Sanskrit and classical Chinese. It thus can
serve as a commonality, between the reductio-ad-
absurdum and paradoxical methods respectively
adopted by the Indian and Chinese Maadhyamikas.

P.56

IV.

That the Buddhist insight of `suunyataa is trans-
cultural can be defended by the very fact that the
inferential logic of the mind transcends cultural
differences, and also that the dialectical method,
which is definitive of that insight, derives from
the same structural basis as that of logical
thinking. This, however, does not mean that the
dialectical method can arise directly from the
structural basis of the inferential process. AS has
been explained before, the inferential process
requires two conditios: (1) the clear dermarcation
of the two sets of loci; and (2) the separate
application of anvaya and vyatireka operations by
referring to the respective loci, a two-fold
condition which I called 'Logical Context.' Now, I
wish to call attention to the Dialectical Context,
in which the dialectical method has its immediate
basis, and which is radically different from the
logical context in two ways.
At the 31st Orientalist conference [CISHAAN] held
in Tokyo in the summer of 1983, my paper entitled:
"Buddhist dialectical methods and their structural
identity" (9) formulated the dialectical context in
terms of two criteria, namely: (1) that the
dialectical context is necessarily created in every
logical and linguistic process of the mind as
'simultaneous application of anvaya and vyatireka,'
and (2) that these operations in turn create a 'dual
natured reference' that can only be expressed by
metaphorical instances, such as, magical apparition,
sky flower, and so forth. Suppose we are listening
to on-going speech in which phonemes, words, and
sentences are incessantly coming and going. Catching
a series of rapid sounds, our mind spontaneously
configurates them into a word, a series of words
into a sentence, and a series of sentences into a
unified understanding. It is within this dynamic
flow of speech or thoughts that we have a glimpse of
the spontaneous juxtaposition of anvaya and
vyatireka, irrespective of whether it is a clear
awareness, at every linkage point where the two
consecutive moments of consciousness converge. This
juxtaposition is the first criterion of the
dialectical context, and is clearly recognized in
Naagaarjuna's favored reductio-ad-absurdum argument
(parasa.nga-vaakya) , for instance, over the
epistemic relation between the sense faculty and its
incoming stimuli, or between an individual self

P.57

(pudgala) and psycho-physical elements (dharmas).(10)
In applying his dialectical treatment to these
subject matters, Naagaarjuna used the metaphor of
light and darkness and that of fire and fuel
respectively. The pattern of his argument is
precisely identical. Conceptually, light and
darkness are incompatible by definition, and yet in
convention, these entities are taken for granted to
establish somehow their meeting so as to explain a
fact of illumination. In like manner, fire and fuel,
which are two different things, are taken for
granted to meet together for explaining combustion.
Through dialectical manipulation of statement,
Naagaarjuna repudiates this convention. The
following is a simplified translation for the
Kaarikaas 36 through 39 in the Vigrahavyaa-vartanii:
(11)

36. Wherever [ i.e. in reference to the meeting
point in space and time ] light illuminates
darkness, darkness also obstructs illumination
there. [ anvaya and vyatireka together ].
37. Wherever there is light, there should be no
darkness (because they are exclusive). How could
light illuminate anything. [ anvaya, while
vyatireka negated ]
38. Does light illuminate darkness at the moment
of its arising? No, it does not reach darkness
from the very beginning. [anvaya and vyatireka]
39. If light here illuminates darkness without
reaching it, this light here should illuminate
darkness of all the world. [anvaya, while
vyatireka negated]

Here's contact of light and darkness' refers to
the meeting point of the faculty of sight and its
incoming stimuli as a causal factor in vision. This
meeting point transcends linguistic convention, thus
Naagaarjuna points out two things: (12) (1) that the
spatio-temporal sphere to which such a contact is
referred is not a logical context, and (2) that in
that referential sphere, which is dialectical, light
and darkness no longer hold their distinct
self-identity, but reciprocally exchange their
functions, i.e., simultaneously light and darkness,
existent and non-existent, which I call the 'dual
natured reference.' From the Maadhyamika Buddhist
point of view, every

P.58

empirical cognition or entity is dualistic in
essence, related to another, and hence devoid of its
independent self-identity [ i.e. ni.hsvabhaava =
'suunyataa]. In short, the dialectical context
consists of two criteria: (1) juxtaposition feature
and (2) dual natured reference.

V.

The Chao-lun is regarded as a difficult text not
only because of its dialectical content, which it
shares, in common with the Maadhyamika treatises,
but also because Seng-chao's dialectical method is
not the so-called reductio-ad-absurdum argument used
by the Indian masters, but a paradoxical method of
argument of his own. That his paradoxical method is
authentic Maadhyamika dialectic can be proven in
reference to the two above-mentioned criteria,
namely: juxtaposition feature and dual natured
reference. Without these criteria, we would be
confronted with a formidable, perhaps unintelligible
form of paradoxical argument unprecedented in Indian
Madhyamika. Only after a careful analysis with the
dual criteria do we notice a set of basic patterns.
In the first essay : "The Immutability of Things
(i.e. dharmas) " (e), Seng-chao deals with the
causality of dharmas and sets forth a series of
paradoxical arguments in terms of very tangible and
concrete phenomena. As with Naagaarjuna, Seng-chao's
usual procedure is to match up a contrary conception
or phenomenon with one that is to be considered.
This essay provides an excellent example for
'matching up contraries': (13)

By motion, people mean that things are in
motion and not at rest, thinking that past
things do not reach the present. (f) For the
same reason, however, I claim that it is
stillness, meaning that they are at rest and
not in motion. (g) People think that because
things are in motion and not at rest, they do
not come to the present.( h ) I think,
however, that because things are at rest and
not in motion, they do not pass away from the
present.(i).

There follows Seng-chao's typical form of paradoxical
argument : (14)

p.59

Thus, things are said to be permanent and yet
not abiding (j) they are said to passing and yet
not changing. (K) Because they are not changing,
things are passing and yet always at rest. (l)
Because things are not abiding, they are at rest
and yet always passing. (m) Because things are
at rest and yet always passing, they are passing
and yet not changing.(n) Because they are
passing and yet always at rest, they are at rest
and yet not staying.(O)

Seng-chao then concludes the argument by applying
the principle of juxtaposition : (15)

When people say that things are abiding, I say
that they are gone. (P) When they say that
things are gone, I say that they are abiding.(q)
Although 'gone' and 'abiding' are different in
expression, what they mean is one reference.(r)

Given the foregoing type of argument, one anti-
cipates a spontaneous objection or skepticism from
the conventional point of view. Seng-chao states
such an objection in the third essay: " Praj~naa is
Not-Knowledge" (r1) in the second question-reply
section. The question which Seng-chao raises from
the common sense point of view reflects, I assume,
his own intellectual, logico-linguistic concern,
reads as follows:

Things cannot be used for comunication, (s) and
hence they are given names by which they are
referred to and communicated among men. (t)
Although things are not names, they become the
objects to be denoted by names in terms of
correspondence. (u) Hence, by a name, if we seek
for anything, we can obtain something that
corresponds to the name. (v)

Now, however, you assert that the transcendental
insight is 'not-knowing', (w) and yet you also
said that there is nothing that it does not
know. (x) If the Prajna is 'not-knowing,' it
must mean not to have ever known anything, (y)
and if it is 'knowing',

P.60

It must mean to have always been knowing. (z)
This is what the Doctrine of Names promulgates,
(aa)and it is the basic principle of stating
about things. (ab)

Contrary to this common sense, you are saying
that 'knowing' and 'not-knowing' are assumed to
be identical in the Praj~naa insight. (ac) Yet
you are differentiating the meanings of these
contrary faculties.(ad) Examining what is said
and looking for its reference, I have not found
anything that can substantiate what you have so
asserted.(ae) For if knowing should mean to
distinguish the nature of the Praj~naa insight,
then 'not-knowing' would not fit to distinguish
it (af) and vice versa.(ag) If neither does not
fit to do so, there is no sense for further dis-
cussion. (ah)

Seng-chao replies to this criticism through his
paradoxical method of argument, and in doing so, he
again applies the dialectical criterion of
juxtaposition. Initially he quotes some textual
passage as defining 'praj~naa' (17) in paradoxical
terms as 'neither namable nor expressible in
language,' 'neither existent nor non-existent,' or
'neither real nor unreal.' Then, after apologizing,
(18) "Although words cannot explain what the
Praj~naa is, since there is no other way of
communication, I shall try to explain it by sing
such words," Seng-chao confirms: (19)

The Praj~naa is subtle and formless and cannot
be said as existent. (ai) Yet, because it is
functionally extensive, it cannot be said
non-existent. (aj) Because the Praj~naa cannot
be said non-existent, it may have to be said
that the Praj~naa is (transcendentally)
existent. (ak) Because the Praj~naa cannot be
said as existent, no correspondence theory of
language can denote it by name. (ak1) Hence, it
follows that the Praj~naa cannot be (an
empirical) knowing.(al) If the comprehensive
awareness of the Prajna is wished to be
expressed, it must be said 'not-knowing' and yet
not 'not-knowing.' (am) If the function of
distinguishing (things') aspects, it cannot be
said as nothing. (an) The (state of)
comprehensive awareness cannot be said as

P.61

existent. (ao) Because it is non-existent, the
Praj~naa is 'knowing' and yet 'without knowing.'
(ap) Because it is not non-existent, the Praj~nss id
'without knowing' an yet 'knowing' (aq) It follows
that 'knowing' is identical with 'not-knowing,' (ar)
and 'not-knowing' is identical with 'knowing.' (as)
Although 'knowing' and 'not-knowing' are different
in expression, they are not different in the
Praj~naa insight. (at)

VI.

Admitting that Seng-chao's paradoxical method
conforms the dialectical criteria of 'juxtaposition'
and 'dual reference,' we are confronted with a
question as to why Seng-chao relied exclusively on
his distinctive paradoxical form of argument instead
of the reductio-ad-absurdum form which Indian
Maadhyamikas favored. In dealing with this curious
difference, I presented another paper at the 33rd
annual conference of the Japanese Association for
Indian and Buddhist Studies held in Tokyo, 1984.
(20) I identified two possible approaches to the
problem in question, namely, (1) by exploring formal
differences between (Sanskrit) and classical
Chinese, or (2) by exploring cultural differences
between the two peoples. On that occasion, I adopted
the first approach.

My reason for adopting this approach was that,
despite formal differences between classical Chinese
and Sanskrit, the human capacity for language and
symbol formation is generally accepted to be
universal, and the logical laws underlying it, as
noted above, are everywhere the same. On the
contrary, cultural forms, such as the conceptual
networks of Taoist and Indian Buddhist thought
cannot be immediately compared due to the lack of
clear criterion that the two systems share, unlike
the logico-linguistic rules underlying all forms of
language. For instance, the surface structure of
synatactical relations may be quite different from
one language to another; some languages have an
intricate case system, whereas some others a much
simpler one. Yet, the deep structure synatcical
relations are universal across languages, as is
confirmed by the fact of translatability from any
language to any other language. In any case, my
findings are as follows.

P.62

Sanskrit is equipped with an intricate system of
syntactical categories, i.e. Kaaraka Cases,
accompanied with case inflections, grammatical word
classes, and so forth, whereas classical Chinese is
mono-syllabic, lacking inflection and grammatical
word classes, and so forth. In addition, Chinese
ideographical symbols contrast sharply with the
phonetic symbols of Sanskrit. The formal aspects of
Chinese, however, are not necessarily a
disadvantage, because recent studies indicate that
certain characteristics of Chinese in question
facilitae logical thinking, because they bring
classical Chinese closer to the symbolic language of
modern logic than any tongue of the Indo-European
type. The action-centered sentential structure of
Sanskrit itself was restructured by Buddhist
logicians, such as Dignaaga,to attain the logically
calculational subject-predicate form. On the other
hand, the sentential form of Chinese (esp. Classical
one) is constructed over the form of the logically
fit topic-comment structure and with no grammatical
constriction, such as, copula, inflection, parts of
speech, etc., which are often logically ambiguous or
misleading. I here seek to demonstrate in that,
regarding the dialectical treatment of convention,
the reductio-ad-absurdum method is more suitable
vis-a-vis Sanskrit, whereas the paradoxical method
is more suitable vis-a-vis classical Chinese.
In Sanskrit, a sentence is action-oriented.
Syntactical categories (Kaaraka, case relations)
help a given action expressed by its main verb to
accomplish its goal. Thus, a simple statement "He
goes" [gacchati] is believed in convention to
acquire its symbolic power from the Kaaraka
framework, of which kart.r (agent, e.g. 'goer' or
gant.r) , karman (recipient of action, e.g.
'referential passage' or gamyamaana), and karana
(instrument, 'act of going' or gamana) and so forth
are theoretically abstracted as nomatives. From a
root verb gam (to go) are differentiated the forms
of gamyamaana (passive or reflexive progressive),
gamana (transitive, active progressive) , gantr
(agent of going). In the Muulamaadhyamikakaarikaa
II, Naagaarjuna applies his reductio-ad-abusurdum
method to these Kaaraka categories by focusing upon
the Dialectical Context in which an act of going or
motion (gamana) is concomitant with its referential
passage of motion (gamyaamana) as well as its agent
(gant.r). Although there are series of applications
to be found in the chapter, here it is sufficient to

P.63

give only two passage [ Kaarikaas 5 and 6 ] in order
to familiarize any one with the method in question.
(21)

If a gamana (an act of going) is to be found in
the gamyamaana (within its referential passage),
this assertion ipso facto commits itself to
another assertion, such that there are two
gamanas, namely: one gamana by which the
gamyamaana is produced, and another gamana which
is inherent in the gamyamaana.

If there are two gamanas, there should follow
two agents of going (gant.r), because no motion
is possible without its agent, and vice versa. A
similar predicament is also at hand by a simple
statement, namely: 'The goer goes" (gantaa
gacchati), because it implicates two gamanas,
one gamana which indicates its agent and another
gamana which the agent produces.

Now, it is clear why Seng-chao's dialectical
method did not resort to the reductio-ad-absurdum
argument over the aforementioned action context
(i.e. Kaaraka system) which is absent from classical
Chinese, and also the reason why, as has been shown
before, he availed the sets of contrary concepts or
phenomena, such as, motion and stillness, passing
and abiding, and so forth. Moreover in order to
repudiate the self-identifying nature (svabhaava) of
epistemic subject (pramaana) and object (prameya),
Naagaarjuna treated the relation over the metaphor
of light and darkness by means of
reductio-ad-absurdum method, whereas Seng-chao
instead matches up contraries such as 'knowing' and
'not-knowing' by means of his paradoxical method.

VII.

It is intriguing to consider whether a special
relationship exists between a language and a system
of logic developed within that linguistic
environment, especially in case of the Mohist system
of logic in ancient China. Indian logic, as the
terminology anvaya and vyatireka indicates,

P.64

evolved initially within the grammarians' inductive
discipline to organize the rules and principles of
Sanskrit. If Seng-chao's paradoxical method is more
attuned to Chinese language, it may very well be
questioned whether his method was at all based upon
the Mohist system of logic, as the Naagaarjunian
dialectical method was based upon Indian logic.
Charles Graham informed us that the Neo-Mohist text
was circulated during the third century.(22)
Seng-chao, who lived in the fourth century and was
trained in the Taoist classies might have been
exposed to the Mohist logical thought. In a paper I
presented at the 36th annual meeting of the Japanese
Association of Indian and Buddhist Studies in
summer, 1987, entitled " On the dialectical meaning
of instantiation in terms of maayaad.r.s.taanta in
the Indian and Chinese Maadhyamikas," (23) I give
evidence that Seng-chao may have known the Mohist
system of logic.
Unlike Indian syllogistic inference, the Mohist
system did not advocate the principle of
instantiation, but provided the principle of hsiao
(au) as essential to valid inference, which Hu Shih
interpreted to mean 'model thought' or 'standard
mode of thought,' and Chmielewski interpreted to
mean any universal statement of causality well
established by induction. (24)In short, hsiao was
conceived by the Mohists as an "all" statement
arrived at by an inductive procedure and accepted as
true, and hence regarded as a legitimatic general
premise for deriving any particular statement.
According to Sanskrit inferential rules
(anvaya-vyatireka), each inference must include an
affirmative as well as a negative instance, thus
combining inductive procedure with deduction. The
Mohist system theorized the dual rules as the method
of sameness and difference, fully parallel with
Sanskrit counterparts, but without the principle of
instantiation. This may be why Seng-chao does not
appeal to instantiation, except to quote the sutras
when necessary. For instance, in his second essay
"The Unreality of 'Suunyataa" he applies a maayaa
metaphor quite adequately, quoting: (25)

We want to say that dharmas exist, but their
existence is not a real production. (av) We want
to say that dharmas do not exist, but phenomenal
forms are already configurated. (aw) Phenomenal
forms cannot be said as 'identical with nothing,
(ax)

P.65

but we only say that anything unreal is not a
real existent.(ay) It follows that the meaning
of "Suunyataa of Whatever is Unreal" is thus re-
vealed.(az) Accordingly, the Pa~ncavi.n'sati-
saaasrikaa-praj~naapaaramitaa-suutra says:"Dhar-
mas are called metaphorically as 'unreal' (ba)
just as a magically created man is. (bb) For we
cannot say that there is no magically created
man, (bc) but that such is not a real man. (bd).

I am convinced that Seng-chao was acquainted
with Mohist logic because of the special context in
which he uses the term hsiao. Although he does not
use hsiao to designate an inferential universal or a
norm for valid inference, he uses it to designate
the dialectical context that validates all
dialectical statements. In his last eassay "Nirvana
is Name-less" he considers the relationship between
psycho-physical elements (dharmas) and an individual
self (pudgala) while trying to demonstrate the truth
of non-dality, by demonstrating that there is no
difference between the pudgala and the dharmas from
the point of view of the Praj~naa insight. Seng-chao
initially describes what the perfect man is,: (26)

The perfected man is all void without any form
whatsoever,(be) and yet there is none in myriads
of things that is not created by his self. (bf)
Whoever should cause myriads of things to be met
with himself and thus realize his own being,
would it be such perfected man alone?(bg) If a
man is not with the true nature of things, he
should not be a perfected man.(bh) Without a
perfected man, there should arise no true nature
of things(bi) Wherever there is the true nature
of things, there should arise a perfected man;
accordingly the perfected man is not different
from the true nature of things.(bj)

Seng-chao then quotes a doctrinal query and reply
between Indra and Subhuuti: (27)

Then, the lord of gods questioned: "In what
reference should I seek for the Praj~naa
insight?" Subhuuti replied: "You should

P.66

not seek for it in reference to a ruupa,(bk) nor
should you seek for it in reference to anything
separate from a ruupa." (bl) Subhuuti again
said: "Seeing the causality of 'Dependent
Origination" means seeing the Dharma; seeing the
Dharma means seeing the Buddha." (bm)

Seng-chao, then, referring to the above triad [i.e.
seeing pratityasamutpada, seeing dharma, and seeing
buddha], uses the term hsiao to designate it, and
says: (28)

This is but the ultimate causal basis for the
insight of non-difference between things
(dharmas) and an individual self (pudgala).(bn)

That Seng-chao deliberately uses the logical
term hsiao to denote the translogical context
indicates by itself not only his acquaintence with
Mohist logic but also his awareness of the
dialectical context underlying the Praj~naa insight
as the ultimate basis for realizing Nirvaa.na. This
hunch is supported by the immediately following
statements Seng-chao makes to substantiate his
quotation of the above tried: (29)

Thus it is said in the Sutra [?]:"One realizes
Nirvaa.na without separating himself from
dharmas" (bo) and also it is said:

"Because there is no limit with the (number of)
dharmas, there is neither a limit with
Enlightenment." (bp) Thus, the path of Nirvaa.na
is known as having no limit." (bq)....Hence, it
follows that things (dharmas) are neither
different from an individual self (pudgala).
(br) nor is the self different from things, (bs)
and that things and the self inexpressively meet
together.(bt)

VIII.

I deliberately avoid Seng-chao's expository
statements which contain Taoist jargon to avoid
making any interpretative commitment to one

P.67

cultural form or to another. I always try to confine
my investigation to the forms and functions of his
paradoxical method to see how sucessfully Seng-chao
created the dialectical context, which aone is
capable of revealing the Praj~naa insight of
'suunyataa.
Seng-chao's application of his paradoxical
method to terms, relations, and referential objects
indicates that he may have been familiar with Mohist
logic, and that his knowledge of that logical system
may have facilitated him in formulating his
dialectical method. Thd use of the Mohist terms
hsiao, by which he deliberately designates the
dialectical context, warrants careful study. The
logical meaning of 'hsiao' fits quite well in that
particular passage, preluding the suspicion that it
was mistakenly adde by scribes. If Seng-chao had in
fact used the term, it seems safe to assume that the
Mohist terminology was still used by intellectuals
of the day. My research on the relationship between
Seng-chao's paradoxical method and Mohist logic is
not complete. The origin and role of 'hsiao' as a
valid inferential criterion is still a mystery to
me.
The insight of 'suunyataa is not confined to the
linguistic and cultural form of Indian people but
transcends them. Fundamentally, it was only through
this term that Buddhist religiosity was accepted by
the Chinese. What Seng-chao teaches us is that,
unlike a cultural dissemination, paradoxically,
transcendental religiosity involves the application
of one's own method to transcend one's own
linguistic and cultural forms within one's own
cultural and linguistic medium. Seng-chao's
paradoxical method seems to vindicate the very fact
that he applied his own method in the confines of
his own linguistic and cultural forms in a way that
was fully parallel with the way Indian Maadhyamikas
accomplished the task in their own cultural and
linguistic environment.

Notes

(1). Libental, op. cit., p.22

P.68

(2). Robinson, op.cit.,pp.154-155.
(3). The article included in The Professor L.M. Joshi
Memorial Volume, Central Institute of Higher
Tibetan Studies, Samath, Varanasi, India
(forthcoming).
(4). 'Sankarasvaamin simplified the inferential rules
in his Nyayaprave'sakasuutram [GOS, 33 (1930),
p.1] pak.sadharmatva^m sapak.se sattva^m
vipak.se caasattvam iti.
(5). RO 26, 1(1962) 7-22; 26,2(1963) 91-105; 27, 1
(1963) 103-121; 28, 2(1965) 87-111; 29,
2(1965)117-138; 30, 1(1966)31-52.
(6). Hu Shih, op.cit., pp.103-105; Graham: op. Cit.,
p.102 and p.130. Here the principle of
agreement and that of difference are translated
as: "Having respects in which they are the same
is being of the same kind. Not having respects
in which they are the same, is not being of a
kind," namely: "having some thing by means of
which they are (judged to be) the same."
(7). Chmielewski, loc.cit., "Notes on...III." P.109.
(8). Ibid., " Notes on.. III ; " the entirety of the
article is devoted to the subject of hsiao.
(9). Ichimura: the resume of the article included in
Proceedings of the 31st International Congress
of Human Science in Asia and North Africa
[Tokyo: The Toho Gakkai] 1984, pp.149-150; the
full text included in The Dr. S.V. Sohoni
Felicitation Volume, JBRS, 1987
(10). For the dialectical treatment on the relation-
ship between pramaa.na and prameya, see the
Vigrahavyavaatanii, partially treated in this
article in the following. For the dialectical
treatment of the relation between pudgala and
dharmas, see the Muulamaadhyamikakaarikaa X:
Agniindhanapariiksaa, where Naagaarjuna uses
the metaphor of fire and fuel respectively for
pudgala and dharmas.
(11).yadi ca svaparaatmaanau tvadvacanena prakaa's-
ayaty agnih/
pracchaadayisyati tama.h svaparaatmaanau
hutaa'sa iva//36
naasti tama's ca jvalane yatra ca t.si.thati
paraatmanijvalanah/
kurute katha^m prakaa'sa^m sa hi prakaa'so
'ndhakaaravadhah//37
utpadyamaana eva prakaa'sayaty agnir ity
asadvadah/
utpadyamaana eva praapnoti tamo na hi hutaa'sa.h
//38
apraapto 'pi jvalano yadi va punar andhakaaram
upahanyaat/
sarve.su lokadhaatu.su tamo 'yam iha sa.msthito
haanyat//39
(12).Cf. Ichimura : " An analysis of Maadhyamika
dialectic in terms of the logical principle of
anvaya-vyatireka, in Studies in Buddhology: The
Professor P.V. Bapat Felicitation Volume, ed.
N.H. Santani [Delhi: Motilal Banarsidass,
1988], esp. Sect. IV. Also "An Approach to
Doogen's dialectical thinking and method of
instantiation (a comparative study of
Shoo-boo-genzoo), Journal of the Inteernational
Association of Buddhist Studies, IX, 2(1986)
79-80), in which the verses are put into
symbolic notation.

P.69

(13). Taisho., 45; No. 1858; p.151a(22-24).
(14). Ibid., p.151b(18-21).
(15). Ibid., p.151c(10-12).
(16). Ibid.p.153c(15-22).
(17). Ibid., (23-24); (bu)
(18). Ibid.,(26-27): (bv)
(19). Ibid.,p.153c(27-29) and p.154a(1-4).
(20). Ichimura : " A determining factor that differ-
entiated Indian and Chinese Maadhyamika
methods of dialectic as reductio-ad-absurdum
and paradoxical argument resepectively, "
Journal of Indian and Buddhist Studies, 33,
2(1985) 834-841.
(21). gamyamaanasya gamane prasaktam gamanadvayam/
yena tad amyamaana^m ca yac caatra gamana^m
puna.h//5
dvay gantaarau prasjyete prasakta^m gamanadvaye/
gantaara^m hi tirask.rt ya gamana^m nopapad-
yate//6
ganta tavad gacchatiti katham evopapadyate/
gamanena vinaa gantaa vada naivopapadyate//9
gamane dve praasajyete gantaa yady uta gaccahti/
ganteti cocyate yena ganta san yac ca agcchati
//11
(23). JIBS. 36, (1988) [forthcoming].
(24). Hu Sh ih : op. cit., p.95f; Chmielewski : loc.
cit., pp.106-109.
(25). Taisho., 45, p.152b (16-20).
(26). Ibid., p.161a(7-9).
(27). Ibid., (10-12).
(28). Ibid., (12).
(29). Ibid., (16-20).

P.70