Articles by alphabetic order
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0

Derrida and Madhyamika Buddhism: From Linguistic Deconstruction to Criticism of Onto-theologies

From Chinese Buddhist Encyclopedia
Jump to navigation Jump to search

A COMPARATIVE STUDY of Jacques Derrida's deconstructive philosophy and the Madhyamika Buddhism-which was founded in India by Nagarjuna
(c. 100-200) and established in China by Seng-chao (374-414) and Chi-tsang (549-
623), and flourished in Korea from the sixth to the fifteenth century, in Japan from the seventh to the twelfth century, and in Tibet from the eighth century to the present-probably no longer needs to be preceded by an apology.1 The remarkable parallel between the Derridean logic and the Madhyamika prasaizga (reductio ad absurdum) has already been carefully examined by a number of specialists in comparative philosophy.2 In the Derrida-Madhyamika studies, there are still many other important parallels awaiting our discovery and exploration. Here, we propose to consider the relationship between linguistic deconstruction and ontotheological criticism in these two (anti)philosophical traditions. In the course of our inquiry, we will discover four important parallels in Derridean and Madhyamika theories: (1) Both Derrida and the Madhyamika thinkers develop deconstructive theories of meaning based on the similar ideas of differance and differentiam, and seek to nullify the logos and the Name of Non-Existence reified by Western idealists and Buddhist Essentialists; (2) both apply the same theories of meaning to deconstruct Matter and Existence, reified by Western materialists and Buddhist Realists; (3) both conceive of their double negation as an exercise of neither/nor logic and set forth their deconstructive formulas in similar terms of "tetrapharmakon" and "tetralemma" (catuskoti); and (4) both abolish their own tetrapharmakon and tetralemma, and embark on their self-deconstructive course along an aimless "supernumerary" and a linear "hexalemma." While we examine these four parallels in the following sections, we shall also pinpoint the fundamen­ tal differences between the Derridean and the Madhyamika theories and consider how these two deconstructive traditions lead to the end of philosophy.

1For a succinct account of the origin , major figures, and scriptures of the Madhyamika Buddhism, see C. W. Huntington and Geshe N. Wangchen, The Emptiness of Emptiness: An Introduction to Earl y Indian Madhyamika (Honolulu: Univ. of Hawaii Press, 1989), pp. 25-67; Hsueh-li Cheng, Empty Logic: Madhyamika Buddhism from Chinese Sources (New York: Philosophical Library,
1984), pp. 9-32.
2See Robert Magliola, Derrida on the Mend (West Lafayette, IN: Purdue Univ. Press, 1984), pp. 3-
129; David Loy , "The Cloture of Deconstruction: A Madhyamika Critique of Derrida," International
Philosophical Quart erly 27 (1987), 59-80.



Bagan, Burma.jpg

Both the Derridean and the Madhyamika theories are cast in terms of disagree­ ments over language with the idealists and the materialists. Derrida holds that Western idealists from Plato to Heidegger operate through a false conception of language. They all invoke the logos, a linguistic sign, as an intermediary between the transcendental and sensible, the divine and the human. Socrates in the Phaedo, Derrida observes, "tells of his fear of being blinded by looking at things directly.... And he tells how, instead of turning directly to things, he turned rather to logoi in order to examine there the truth of beings." 3 To vindicate the logos as an embodiment of truth of beings, Plato and later idealists adopt a two-fold strategy-to banish its corporeal "gram" (gramme , in the French) and to reify its intangible "phone."* Although the term "logos" contains the meaning of language as a whole, they do not employ it pertaining to the graphic form of language. This banishment of the gram from the logos stems from the fear that the visibly corporeal gram will contaminate the phone, the transcendental signi­ fied. Plato unequivocally expresses such a fear when he denounces the gram as "the intrusion of an artful technique, a forced entry of a totally original sort, an archetypal violence: eruption of the outside within the inside, breaching into the interiority of the soul, the living self-presence of the soul within the true logos, the help that speech lends to itself.4 While Plato and later idealists banish the gram as "an orphan or a bastard," they reify the phone as "the legitimate and high-born son of the 'father of logos' "5-on the ground of the "proximity of voice and being, of voice and the meaning of being, of voice and the ideality of meaning" (OG 39). As a result of such a phonocentric reification, the logos has assumed the ontotheological significance of "the Word, the Divine Mind, the infinite understanding of God, and infinitely creative subjectivity , and closer to our time, the self-presence of full self-consciousness" (OG lxviii).
In the opinion of Derrida, the phonocentric reification of the logos presents a paradox. While such a reification is intended to forestall "the eruption of the outside within the inside," it actually exposes the outside within the inside of all Western idealist metaphysics. For Derrida, the reason cannot be simpler. The privileged phone is as much a linguistic sign as the denigrated gram and is subject to the same rules of linguistic signification governing the gram. So, the logos is by necessity a linguistic sign external to the transcendental absolute. As such, it inevitably inscribes all metaphysics within a space of externality and precludes the possibility of any intrinsic absolute presence.
To expose how Western metaphysics "finds inscribed, rather than inscribing itself, within a space (the externality of the logos] which it seeks but is unable to

3John Sallis, Deconstruction and Philosophy (Chicago: Univ . of Chicago Press, 1987), p. xi.

  • Editor's note: Phone is Derrida's and his translators' convention.

4Jacques Derrida, OfGrammat ology , trans. Gayatri Chakravort y Spivak (Baltimore: Johns Hopkins
Univ. Press, 1974), p. 34; hereafter OG.
5Jacques Derrida, paraphrasing Plato, in Positions , trans. Alan Bass (Chicago: Univ. of Chicago
Press, 1972), p. 12; hereafter P. See also Derrida's discussion of this issue in OG 39.


control,' '6 Derrida coins the word differance. This neologism constitutes, on the one hand, the nominal form for the French verb differer, which means both "to differ" and "to defer," and on the other, a dissimulation of the French noun difference (difference). Simple and playful as it seems, this neologism spells out Derrida's deconstructive theory of linguistic signification. First, differance calls into question the privileging of the phone by the idealists, because it is not the phone but the gram of the word that makes its meaning understood. If heard but not read in French, differance is bound to be confused with the noun difference. Second, differance underscores the prerequisite for linguistic signification. A sign cannot exist unless it differs spatially and is deferred temporally from the signified. This ever receding gap between the signifier and the signified disproves the alleged fusion of the phone and the ontotheological essence in the logos. Third, the Latin root of differance (differre, in the sense of "to scatter, disperse") denotes the necessary play of opposing referents (the spatial versus the temporal in differance) in linguistic signification. A name must contain its disputant meaning(s) in order to exist as a name (i.e., A cannot be called A unless A also signifies the existence of non-A). So, a name signifies absence as well as presence. This being the case, what a name signifies cannot be the pure signified of presence, but another signifier which in turn signifies absence as well as presence. 7 This goes on and on to infinitude. It follows that the very possibility of the transcen­ dental signified or the ontotheological presence is to be denied. Given this operation of differance, Derrida contends that logocentric concepts-"eidos, arche, telos, energeia, ousia (essence, existence, substance, subject), aletheia, transcendentality, consciousness or conscience, God, man, and so forth"8-are all caught in an infinite circularity of signifiers and will never be able to presence the transcendental absolute.
Like Derrida, the Madhyamika thinkers seek to demolish the ontotheological
arguments of the Buddhist essentialists by exposing their false conception of language. Whereas Derrida invalidates the intrinsic nature of the logos by dem­ onstrating the phone as a conventional sign in differance, the Madhyamika philosophers nullify the "intrinsic identity" (svalak$ana) of the Name of Non­ Existence by showing its conventionality.9 Candrakirti, a great Indian Madhya­ mika thinker of the seventh century, repudiates the essentialist claim of the intrinsic nature of the Name of Non-Existence. In the opinion of a Buddhist Essentialist, that the phrases "body of a statue" and "head of Rahu" exist­ although an inanimate statue has no body and Rahu (a demon) has no head­ attests to the non-representational, intrinsic identity of language. In response to such an essentialist view, Candrakirti writes:

6Jacques Derrida, "The Time of a Thesis: Punctuations," in Philosophy in France Today, ed. Alan
Montefiore (Cambridge: Cambridge Univ. Press, 1983), p. 45.
7 See Murray Krieger, "Poetics Reconstructed: The Presence and the Absence of the Word," New
Literary History 7 (1976), 347-76.
8Jacques Derrida, Writing and Difference, trans. Alan Bass (Chicago: Univ. of Chicago Press,
1978), pp. 279-80.
9See G. C. Nayak, "The Madhyamika Attack on Essentialism: A Critical Appraisal," Philosophy East and West 29 (1979), 467-490; Peter G. Fenner, "Candraki:rti' s Refutation of Buddhist Idealism," Philosophy East and West 33 (1983), 251-56.


When the words "body" and "head" normally occur in grammatical connection with companion entities like "hand" or "mind," the thought produced on the basis of the words "body" and "head" alone carries an expectation of the companion entities in the form, "Whose body?" and "Whose head?" ... Furthermore, the terms "statue" and "Rahu," which are the qualifiers, actually exist as part of conventional usage, and are accepted without analysis, as in the conventional designation "person." Therefore your example is incorrect. 10

Here, Candrakirti considers the two phrases as no more than ordinary words that can convey a meaning only through conventional association established by grammatical connections. He believes that "whatever meaning they [words] had was acquired by a process of mutual dependence (paraspariipek$ii siddhi/:t), with one word depending for its meaning on the network of those that were used before it."11 These remarks of Candrakirti are reminiscent of Derrida's theory of differ­ ance. Like Derrida, he conceives of linguistic signification as an interplay of signifiers, and argues on that ground against the notion that a word has an intrinsic essence within itself.
It is important to note that Candrakirti's analysis of linguistic signification is based on his commonsense observations while Derrida's is developed from modern theories of semiology. For a more systematically developed Buddhist theory of language, we must turn to the differentiation theory of meaning (apoha) developed by Dignaga (c. 480-540), a phenomenal Indian Buddhist logician, under the influence of the early Madhyamika school. 12 Whereas Candrakirti explores the mutual dependence of words within a grammatical construction, Dignaga looks into the mutual dependence of opposing elements within a single word and yields a fresh insight into the nature of linguistic signification: "Indeed the name can express its own meaning only by repudiating the opposite meaning, as for instance the words 'to have an origin' designate their own meaning only through a contrast with things having no origin or eternal." 13 This statement seems at first sight to be an affirmation of the self-present meaning of a name. A closer examination of it will reveal that Dignaga is actually arguing the opposite because he sees the very condition for any name to establish a meaning is the existence of a disputant meaning. This necessary dependence of a meaning upon its contrary is sufficient disproof of the alleged intrinsic meaning of a name. The deconstruc­ tive significance of this statement becomes much clearer in the following commen­ tary by Jinendrabuddhi: "Indeed the aim of the text of Dignaga is that the word
'expresses per differentiam' its own meaning.... (The words express only
negations, only differences!), because a pure affirmation without any (implied) negation is senseless." 14 This commentary assures us that Dignaga indeed con­ ceives of linguistic signification as a process of negation and difference akin to


10Prasannapada , in Bibliotheca Buddhica , ed. Louis de Ia Vallee Pousin (St. Petersburg: Akad. Nauk-Izd. Vostochnoi Lit-ry, 1913), IV, p. 16.
"Malcolm D. Eckel , "Bhavaviveka and the Earl y Madh yamika Theories of Language," Philosophy
East and West 28 (1978), 325.
120n the influence of the early Madhyamika on Dignaga's theory of apoha , see F. Th. Stcherbatsky,
Buddhist Logic (New York: Dover, 1962), vol. 1, pp. 27-31. ya , V. 1 ; Stcherbatsky , vol. 1 , p. 459. ya-vrtti ad V. 11 ; Stcherbatsky , vol. 1, p. 463.

that of differance.15 Considering all this, it seems no accident that Jinendrabuddhi uses differentiam to characterize Dignaga's differentiation theory of meaning, just as Derrida uses differance to typify his own. Moreover, like Derrida, Digmlga and other Buddhist logicians go from linguistic deconstruction to ontological nega­ tion (arthiitmaka-apoha). While Derrida annuls "eidos, arche, telos, energeia, ousia ..." through a demonstration of differance in those sacred names, they disprove the essence of language through a revelaton of differentiam in ontotheo­ logical names.


Both the Derridean differance and the Madhyamika differentiam illuminate the mutual dependence of signified and signifier, referent and non-referent , presence and absence. On the ground of this mutual dependence, both Derrida and the Madhyamika thinkers believe that intrinsic identity cannot be claimed for either the signified-referent-presence side or the signifier-nonreferent-absence side. They hold that all onto-theologies are necessarily false insofar as they ascribe intrinsic identity to one or the other side of this paradigm. For Derrida and the Madhyamika thinkers, this paradigm of mutual dependence not only exposes the erroneousness of all onto-theologies but also provides a very convenient way to deconstruct them. All one has to do is to overturn the "hierarchization" of these two sides by a philosophical system. We have already seen how Derrida and the Madhyamika thinkers launch such a deconstructive attack against Western ideal­ isms and Buddhist Essentialisms. Now, let us see how they return to deconstruct Western materialisms and Buddhist Realisms.
While Derrida seeks to overturn the superiority of the phone at the first phase of his deconstruction , he re-marks the gram at the second phase to prevent the re-institution of logocentrism in the form of materialism. He writes: "Nothing would be more ridiculously mystifying than such an ethical or axiological reversal, returning a prerogative or some elder's right to writing" (P 13). He perceives the "counter-privileging" of the gram, the ostensibly corporeal signifier, as emblem­ atic of a metaphysical reification in the form of materialism. Such a reification is not dissimilar to that of the phone as the transcendental signified by the idealists:

Borobudur 2008.JPG

... the concept of matter has been defined as absolute exterior or radical heterogeneity. I am not even sure that there can be a "concept" of an absolute exterior. If I have not very often used the word "matter," it is not, as you know , because of some idealist or spiritualist kind of reservation. It is that in the logic of the phase of overturning this concept has been too often reinvested with "logocentric" values, values associated with those of thing, reality, presence in general, sensible presence, for example, substantial plenitude, content, referent, etc. Realism or sensualism-"empiricism,"-are modifi-

1lDignaga, however, does not pursue his apoha to the point of a total denial of all ontotheological positions. It is probably for this reason that Stcherbatsky puts Dignaga on the side of the idealists in his schema of the three .Phases of Indian Buddhism (vol. I , p. 14). Dhirendra Sharma discusses how Dignaga's followers pursue the apoha to different ontological conclusions in his The Differentiation Theory of Meaning in Indian Logic (The Hague: Mouton, 1969), pp. 19-46. See also Bimal K. Matilal and Robert D. Evans, eds., Buddhist Logic and Epistemology (Dordrecht: D. Reidel, 1986), pp. 77-
87, 185-91, 229-37.

188 CAl

cations of logocentrism. (I have often insisted on the fact that "writing" or the "text" are not reducible either to the sensible or visible presence of the graphic or the "literal.") In short, the signifier "matter" appears to me problematical only at the moment when its reinscription cannot avoid making of it a new fundamental principle which by means of theoretical regression, would be reinstituted into a "transcendental signified." (P 64-

In this passage, Derrida aims to show how easily the rehabilitation of a signifier
-be it "thing, reality or presence in general"-may go overboard and result in
the reinstitution of logocentrism. Derrida mentions the materialist texts of Marx and Lenin as typical cases (cf. P 72, 74-76). There, the signifier "matter" has become the absolute cosmological and socio-historical principle. In other words, the signifier is reified and turned into a transcendental signified no less fictitious than the logos subscribed to by most idealists from Plato to Heidegger. To prevent his own grammatology from being "re-invested with 'logocentric' values," Der­ rida advocates a simultaneous deconstruction of the gram and the phone and calls such a practice "biface or biphase," "double seance," or "double register in grammatological practice" (P 42, 45 passim , 35).
The Madhyamika double negation bears close resemblance to the Derridean double seance. At one phase, they seek to reduce ontotheological Names "to mere conventional negative signs of differentiation and thereby deconstruct all Essentialist schools "whose valuation of Speech and of Names [Non-Being] had all the character of religious veneration-for whom the Word was an eternal positive Ens existing in an eternal union with the things denoted by it."16 At the other phase, they seek to reduce the Physical Phenomenon to a mere language­ thought construct and thereby destroy all Realist schools which valorize Exis­ tence as the eternal Ens-what Derrida calls thing, reality or presence in general."
The Madhyamika thinkers pursue double negation more even-handedly than Derrida does. Derrida concentrates his attacks on idealisms and seldom takes on idealisms and materialisms simultaneously, as he claimed. By contrast, the Madhyamika thinkers almost always seek to negate Essentialisms and Realisms in the same breath. For instance, Seng-chao, founder of Chinese Madhyamika, exploits the mutual dependence of a "name" and a "thing" to expose the illusory nature of both. He writes: "A name does not correspond to an actuality. An actuality does not correspond to a name."17 On that ground , he goes on to criticize the "name"-valorizing schools-the School of Original Non-being prop­ agated by Tao-an (312-385), its Variant School of Original Non-being headed by Fa-shen (286-374), and the School of Non-being of Mind led by Fa-wen (ft . 374). 18
He considers all their theories "nothing but a talk partial to non-being." 19 Then, he takes on the "thing"-valorizing schools like the School of Matter As It Is,


16Stcherbatsky , vol. I , p. 480.
17"The Emptiness of the Unreal ," collected in A Source Book in Chinese Philosoph y , ed. W. T. Chan (Princeton: Princeton Univ . Press, 1963), p. 356. For comparison , see also Chao Lun: The Treatise of S eng-chao, trans. Walter Liebenthal, 2nd rev. ed. (Hong Kong: Hong Kong Univ. Press,
1968), p. 56.
18For a brief introduction to these three Buddhist Essentialist schools before Seng-chao, see Chan ,
Source Book , ch. 20, pp. 336-42; and appendix I of Liebenthal, pp. 133-50.
19Chan, S ource Book , p. 352.

represented by Chih Tao-lin (314-366). He contends that these schools do "not understand that matter [including its conditional existence) is really not matter at all."20 Such a relentless double negation abounds in all Madhyamika schools. After all, it is in reaction to Essentialisms and Realisms that the Madhyamika arose and developed as a deconstructive philosophy. Its exercise of double negation is the very raison d 'etre of its existence.


Both Derrida and the Madhyamika thinkers describe their double-register deconstruction in terms of "neither/nor." In Positions, Derrida double negates a host of conceptual opposites endowed with ontotheological significance, and then sums up the principle of his biphase deconstruction as a practice of neither/nor.

. . . the pharmakon is neither remedy nor poison, neither good nor evil, neither the inside nor the outside, neither speech nor writing; the supplement is neither a plus nor a minus, neither an outside nor the complement of an inside, neither accident nor essence, etc.; the hymen is neither confusion nor distinction, neither identity nor difference, neither consummation nor virginity, neither the veil nor unveiling, neither the inside nor the outside, etc.; the gram is neither a signifier nor a signified, neither a sign nor a thing, neither a presence nor an absence, neither a position nor a negation, etc.; spacing is neither space nor time; the incision is neither the incised integrity of a beginning, or of a simple cutting into, nor simple secondarity. Neither/nor, that is simultaneously either or; ... (P 43)


Likewise, Seng-chao characterizes the Madhyamika double negation of the
"Name" and "Thing" as an exercise of "neither-this-nor-that" logic:

The Chung fun (Treatise on the Middle Doctrine, Madhyamika sdstra by Nagfujuna] says, 'Things are neither this or that.' ... Thus 'this' and 'that' do not definitely refer to a particular name, but deluded people would believe that they necessarily do. This being the case, [the distinction] between 'this' and 'that' is from the beginning nonexist­ ent, but to the deluded it is from the beginning not nonexistent. If we realize that 'this' and 'that' do not exist is there anything that can be regarded as existent? Thus we know that things are not real; they are from the beginning only temporary names.21

Not only do Derrida and the Madhyamika thinkers theorize about their decon­ structive logic in the same terms of neither/nor, they also seek to distinguish their deconstructive formulas from other modes of philosophical thinking by invoking the same symbolic number of Four. In Dissemination, Derrida self-consciously defines his deconstructionism against other philosophical schools "through its insistence upon squares, crossroads, and other four-sided figures ... [and its] violent but imperceptible displacement of the 'triangular' -Dialectical, Trinitar­ ian, Oedipal-foundations of Western thought.' '22 First, Derrida restates the different symbolic numbers adopted in ontotheological discourses. Numbers One

21Chan, Source Book, p. 356.
22Jacques Derrida, Dissemination, trans. Barbara Johnson (Chicago: Chicago Univ. Press, 1981), p. xxxii; hereafter D.

and Two represent the classical opposition of Being and beings and its attendant duplicities-' 'remedy/poison, good/evil, intelligible/sensible, high/low, mind/mat­ ter, life/death, inside/outside, speech/writing, etc." (D 24-25). Number Three emerges as a resolution of the duplicities-in religion as the Trinity (Hegel's word), in Kant as a lifeless "triadic" form (Triplicitat, Schelling's word), in Schelling as the quasi-dialectic triplicity, and in Hegel as the living triplicity.23
Then, Derrida compares his deconstructive enterprise to "a pharmacy in which it
is no longer possible to count by ones, by twos, or by threes" (D 24). There, all the twos "can be neither reduced to unity, nor derived from a primary simplicity, nor dialectically sublated or internalized into a third term" (D 25). Similarly, all the threes no longer give us the ideality of the speculative opposition but rather the effect of a strategic re-mark" (D 25). While destroying the dualistic and trinitarian horizons, Derrida envisions his operation of differance or textual dissemination as "a fourth term" and "the supplementary four (neither a cross nor a closed square)" (D 25). Indeed, Derrida is so fascinated with the idea of a fourth term that he renames "pharmakon"-one of his principal examples of differance-as "tetrapharmakon" (D 350). To elaborate on the significance of this fourth term, he cites the following passage from the Philippe Sollers' Nombres: "Even though it is only a triangle open on its fourth side, the splayed square loosens up the obsidionality of the triangle and the circle which in their ternary rhythm (Oedipus, Trinity, Dialectics) have always governed metaphysics. It loosens them up; that is, it de-limits them, reinscribes them, re-cites them."24
That a fourth term marks off Madhyamika Buddhism from other philosophical
systems is self-evident in the very name by which Madhyamika deconstruction is best known-catu$koti, rendered as tetralemma or the four-cornered method of argument. Like Derrida, the Madhyamika thinkers believe Number Four repre­ sents a negation of the preceding three numbers representative of all ontotheolog­ ical positions. To nullify the three existing kinds of ontotheological claims, Nagarjuna introduces the neither/nor as a fourth term:


The world is finite. The world is infinite.
The world is both finite and infinite. The world is neither finite nor infinite.25

Like Derrida, Nagarjuna and other Madhyamika thinkers aim to destroy not only the fundamental opposition of Being and Non-being, but all its attendent duplici­ ties and triplicities. Indeed, just as Derrida rules against counting "by ones, by twos or by threes" in his pursuit of tetrapharmakon, Nagarjuna consistently disposes of the ontotheological ones, twos, and threes in his exercise of tetra­ lemma. He casts out-among numerous other sets-the ones, twos, and threes

2lFor Derrida's critiques of these trinitarian concepts, seeD 20-25.
24 Quoted without any documentation in D 25 (Derrida's italics).
25Based on Nagarjuna's MU!amiidhyamakakiirika (hereafter MK), XXVII, 21, 25, 28, R . D. Gunar­ atne presents this form of catusko{i in "Understanding Nagarjuna's catusko{i," Philosophy East and West 36 (1986), 219. Cf. David J. Kalupahana, Niigiirjuna: The Philosophy of the Middle Way (Albany: SUNY Press, 1986), pp. 387-91.


regarding the extension of the world, the form of soul, and the finality of death, and other important philosophical issues.26


When Derrida and the Madhyamika thinkers reach the fourth term in their deconstructive process, they face the danger of getting trapped in a new dualism between the three preceding terms and their own fourth terms. Unless this new dualism is disposed of, their deconstructive terms themselves will become a fixed ontotheological thesis. To overcome such an ontotheological re-inscription, both Derrida and the Madhyamika thinkers undertake self-deconstruction. Derrida sloughs conceptuality off his deconstructive terms and sees to it that those terms get "imprinted and fractured" by their own logic. He writes:


The motif of differance, when marked by a silent a, in effect plays neither the role of a
concept, nor simply of a word. This does not prevent it from producing conceptual effects and verbal or nominal concretions. Which, moreover-although this is not immediately noticeable-are simultaneously imprinted and fractured by the corner of this letter, by the incessant work of its strange logic. (P 40)

[Difjerance] cannot be elevated into a master-word or a master-concept ... it blocks every relationship to theology .... (P 40)

In the final analysis dissemination means nothing, and cannot be reassembled into a definition.... If dissemination, seminal differance cannot be summarized into an exact conceptual tenor, it is because the force and form of its disruption explode the semantic horizon. (P 44-45)

In grappling with the same problem of re-inscription, the Madhyamika thinkers, too, attempt to turn their deconstructive terms against themselves. For instance, they use the term sunyatti to deconstruct itself and develop a self-deconstructive doctrine of sunyata-sunyatti ("the emptiness of the emptiness"). From the writings of Nagaijuna and Candrakirti, we can find the following elucidations on sunyatti-sunyatti:

"Empty" (sunya), "non-empty" [asunya], "both" (sunya and asunya], "neither" (sunya nor asunya]-these should not be declared. It is expressed only for the purpose of communication.n

This statement (viz. that nothing has self-existence) is not self-existent. ... Just as a magically formed phantom could deny a phantom.... Just so (is) this negation.28

Emptiness is not a property, or universal mark, of entities, because then its substratum would be nonempty, and one would have a fixed conviction ({irsti) about it. In fact, it is

26 See also the following important studies of catu$ko{i: R. D. Gunaratne, "The Logical Form of Catu koti," Philosophy East and West 30 (1980), 211-40; Ives Waldo, "Niigfujuna and Analytical Philosophy," Philosophy East and West 25 (1975), 281-90, and "NiigaJjuna and Analytical Philoso­ phy, II," Philosophy East and West 28 (1978), 287-98; and Richard H. Jones, "The Nature and Function of NiigaJjuna's Arguments," Philosophy East and West 28 (1978), 485-502.
27 NagaJjuna, MK XXII, 11; trans. by Kalupahana in his Niigdrjuna, p. 307.
28NagaJjuna, Vigrahavydvartanf, 24, 23; trans. by Frederick J. Streng in his Emptiness: A Study in
Religious Meaning (Nashville, TN: Abingdon, 1967), p. 226.

Buddhism 2.jpg

a mere medicine, a means of escape from all fixed conviction.... It is not a positive standpoint, but a mere turning away from all views and thought-constructions.29

Like Derrida, Nagaijuna and Candrakirti stress the importance of abolishing their arguments and positions. They also believe that they must treat catusko!i, their deconstructive apothesis, as "mere medicine" and a "magically formed phan­ tom," rather than a positive entity. Through this exercise of sunyata-sunyata, they seek to avoid getting trapped by a fixed conviction while undoing "all views and thought constructions."
Neither Derrida nor the Madhyamika thinkers believe that they can truly abolish their own arguments merely by disclaiming them. They both launch into a sustained , rigorous process of self-construction. For Derrida, a true abandon­ ment of positions, whether his own or others', must be achieved through a kinesis of mutual negations. To distinguish this deconstructive kinesis from the teleologi­ cal kinesis, particularly the Hegelian one, Derrida characterizes it as a process of infinite regress. For Derrida, the term "trace" best captures the infinitude and the drifting nature of his deconstructive kinesis: "The trace is in fact the absolute origin of sense in general. Which amounts to saying once again that there is no absolute origin of sense in general. The trace is the differance" (OG 65; Derrida's italics). Like the "trace" described here, the Derridean deconstructive kinesis will reach no destination-always drifting amidst the phantoms of "the absolute origin of sense." It does not at any stage bear fruits comparable to the synthesis born of the Hegelian kinesis. To emphasize the baren nature of his deconstructive kinesis, Derrida compares it to "a sowing that does not produce plants, but is simply infinitely repeated," and to "a semination that is not insemination but dissemination, seed spilled in vain, an emission that cannot return to its origin in the father" (OG lxv).
The Madhyamika thinkers, especially later masters like Chi Tsang, also seek to subject their own tetralemma to a deconstructive kinesis. But they follow a deconstructive kinesis in a totally different direction. Whereas Derrida views it as a random drift, they treat it as a decidedly directional operation leading to an ultimate stasis, if you will, beyond language and conceptuality. We can discern this distinguishing trait of the Madhyamika deconstructive kinesis in the Doctrine of Two-Fold Truth on Three Levels (erh-t' i-san-kuan), developed by Chi Tsang, the greatest Chinese Madhyamika thinker.30 This theory divides all philosophical positions into mundane truths and absolute truths and stratifies them on three levels of spiritual maturity. A close look at the following scheme drawn by Fung Yu-lan ,31 a leading scholar in Chinese philosophy, will provide some insight into the directional development of Madhyamika deconstruction and self-deconstruc­ tion:

Buddhist stupa.JPG

29Candrakirti, Prasannapadii , 12; trans. by Edward Conze in his Large S utra on Perfect Wisdom
(Berkeley: Univ. of California Pres, 1975), p. 144, n.4.
30For a discussion of this doctrine, see Aaron K. Koseki, "The Concept of Practice in San-Lun Thought: Chi Tsang and the 'Concurrent Insight' of the 1\vo Truths," Philosoph y East and West 31 (1981), 449-66.
11A History of Chinese Philosoph y, trans. Derk Bodde (Princeton: Princeton Univ. Press, 1973),
vol. 2, p. 295.

�Three levels of Double Truth


1. Affirmation of being
2. Affirmation of either being or non-being
3. Either affirmation or denial of both being and non-being

1. Affirmation of non-being
2. Denial of both being and non-being
3. Neither affirmation nor denial of both being and non-being


The first lemma (affirmation of being) represents mundane truth; and the second lemma (denial of being), absolute truth on the first and the lowest level. The third lemma (affirmation of either being or non-being) reconciles the first two lemmas and registers as mundane truth on the second level. The fourth lemma (denial of both being and non-being) negates the third lemma and reaches absolute truth on the second level. Then, just as the third lemma reconciles the first and second lemmas, the fifth lemma (either affirmation or denial of both being and non-being) reconciles the third and fourth lemmas on the second level and becomes mundane truth on the third level. Significantly, this fifth lemma represents the initial stage of Chi Tsang's deconstruction of his own catu$koti (tetralemma). The sixth lemma (neither affirmation nor denial of both being and non-being) completes this self­ deconstruction through the negation of the fifth lemma, and rises to absolute truth on the third level. For Chi Tsang, a deconstructive kinesis is a means of cleansing the mind, "not limited to three levels but to be employed progressively to infinite levels until one is free of conceptual attachment.' ' 32 In his Profound Meaning of Three Treatise, he elucidates this view of a deconstructive kinesis: "San Lun (Three Treatise) doctrine teaches that each thesis that may be proposed concern­ ing the nature of truth must be negated by its antithesis, the whole process advancing step by step until total negation has been achieved ... until everything that may be predicated about truth has been negated."33 Chi Tsang himself, however, does not pursue the deconstructive kinesis any further than the sixth lemma. This is probably because he believes that the human mind will be rendered incapable of language-thought formulations when the mind-taxing deconstruction reaches its sixth lemma. For Chi Tsang, how far one should pursue the decon­ structive kinesis depends on how effectively it renders one speechless and thoughtless. Once that state of enlightenment is achieved, the pursuit of the deconstructive kinesis ends of itself. Given this ultimate soteriological goal, the Madhyamika thinkers would not set their deconstructive kinesis adrift among traces of traces, less would they delight in an infinite play of deconstructions, and still less would they compare their deconstructive enterprise to a fruitless dissem­ ination.
As Derrida and the Madhyamika thinker Chi Tsang perceive their deconstruc­
tive kinesis in radically differently lights, they naturally describe it with different numeric symbols. In his last section of Dissemination, Derrida invokes Sollers' "supernumerary" and then concludes the book with this mathematical formula:

32Hsueh-li Cheng, Empty Logic, p. 51.
33The Taisho edition of the Chinese Tripitika, no. 1852, p. 10; trans. by Cheng, Empty Logic, pp. 51-

194 CAl

"(1+ 2 + 3 +4)2 • • " Here, with a squaring formula, Derrida seems to indicate the non-linear drift of his deconstructive kinesis. With the supernumerary "..." after the squaring formula, he seems to symbolize the perpetual delays of an outcome and hence the fruitlessness of his deconstructive kinesis. In contrast to Derrida's aimless, fruitless "supernumerary," Chi Tsang's "hexalemma" can be conceived of as a linear progression: 1-2-3-4-5-6. This simple formula betokens the reverse of what Derrida's formula represents. It stands for the directional character, the finite duration, the fruitful promise of the Madhyamika deconstruc­ tive kinesis.


Having examined the four important parallels in the Derridean and Madhy­ amika deconstructive theories, we can now reflect upon these two deconstructive enterprises in the light of their different directions and goals. The two deconstruc­ tive kineses reveal that the Derridean and Madhyamika deconstructive programs differ from each other in their fundamental aspects. In their rigorous pursuits of deconstruction and self-deconstruction, their philosophies both will inevitably reach a point where philosophy ends and "non-philosophy" begins. In the case of Derrida, "non-philosophy" takes the form of infinite textual proliferation beyond comprehension. This is the consequence Derrida knows he will have to bear: To risk meaning nothing is to start to play, and first to enter into the play of differance ..." (P 14). It would be wrong, however, to think that Derrida delights in such a consequence of non-sense, because he confesses that " 'mean­ ing-to-say-nothing' is not, you will agree, the most assured of exercises" (P 14), and he regrets that his deconstructive discourse has provoked "resistance or out­ of-hand rejection even on the part of the best informed readers" (P 68). His is a dilemma between "true to his own word" and "meaning-to-say-nothing." If he practices what he preaches, he would fall to "meaning-to-say-nothing"; but if he tries to avoid "meaning-to-say-nothing," he would go against his own word and become a half-hearted deconstructionist. While we can sense a frequent wavering between these two choices in Derrida's early writings, we feel that he seems more inclined to risk "meaning-to-say-nothing" in his later works. For instance, his Glas seems to exemplify a work "entangled in hundreds of pages of a writing simultaneously insistent and elliptical . . . carrying off each concept into an interminable chain of differences, surrounding or confusing itself with so many precautions, references, notes, citations, collages, supplements" (P 14). When Derrida's faithful execution of his deconstructive theory results in such a verbi­ age, we have to doubt his claim that he does not believe in what today is so easily called the death of philosophy" (P 6). We are left to assume that he has jettisoned, along with onto-theologies, philosophy itself, and led himself into the prison house of language, or even the abyss of nihilism.
In the case of the Madhyamika Buddhism, "non-philosophy" is synonymous with religious enlightenment beyond language and conceptuality. For the Madh­ yamika thinkers, it is not a consequence to be dreaded, but to be celebrated. The Madhyamika deconstruction is geared to none other than this dawning of Nirval)a upon the death of language, concept, and philosophy. In this light, we can very


well see the transformation of Chinese Madhyamika into the non-philosophical Ch'an (Zen)-"a practical, anti-intellectual, irrational, unconventional and dra­ matic religious movement"34-as a doctrinal fulfillment of the soteriological promise of Madhyamika Buddhism.35

34Hsueh-li Cheng, Empty Logic, p. 55.
JjCf. Hsueh-li Cheng's observations on the historical and doctrinal relationship of the Madhyamika and Zen in his Empty Logic, chapter 3.