Response To Claus Obteks Paper

Author(s): Eli Franco
Publisher: Eli Franco
Catalog link: https://jainqq.org/explore/269436/1

Page #1
RESPONSE TO CLAUS OETKE'S PAPER by Eli Franco, Hamburg First, I would like to thank Professor Katsura for giving me the opportunity to respond shortly to Professor Oetke's paper "The Disjunction in the Pramanasiddhi." This response (as well as Oetke's response to it) seems worthwhile to me not only because this controversy is likely to clarify certain points and hopefully advance our understanding of the beginning of the Pramanasiddhi chapter, but also because Oetke does not represent my position faithfully. Oetke's main point, if I understand him correctly, is that Pramanasiddhi la and 5c should be interpreted as referring to either necessary and sufficient conditions or at least sufficient conditions for being pramana. This, he seems to believe, refutes my suggestion that Dharmakirti did not intend to define pramana in general. However, my contention does not depend on the assumption that la and 5c refer only to necessary conditions, and this is the main point in which Oetke misunderstands me. The basic question I raised was not whether the statements in la and Sc could be read as definitions; in the same paper I showed how Dharmakirti's commentators had done so, and in this respect Oetke beats the air. The question I asked myself was whether Dharmakirti intended to define pramana in general or whether he only intended to show that the Buddha is pramana. It is true that I arrived at my hypothesis that the latter is the case by reading la and 5c as formulations of necessary conditions (and this, I believe, is the more natural reading'). However, my resulting suggestion does not rest on the assumption that la and 5c refer only to necessary conditions. To make this point clear, let it be assumed, only for the sake of argument, that Oetke is right, i.e., that la and Sc refer to sufficient or necessary and sufficient conditions of pramana; does it follow from this that Dharmakirti intended to define pramana here? I This is not only my individual intuitive impression; cf. e.g., Pramanavarttikalankara 30.16: tathapy ajnaln/arthagrahanena grhitagrahipratyayah sakyah parihartum navisamvadigrahanena tatrapy avisamvidagrahanat. "Even so, (the second definition is not superfluous, because) by the employment of the word) 'unapprehended object' a cognition which apprehends an apprehended object can be rejected (from the realm of the definition, but not by the employment of the word 'non-belying because non-belying is apprehended even in the case of a cognition that apprehends an apprehended object)." Furthermore, the qualification of pramana as apprehending a non-apprehended object was also adopted by the Bhalta-Mimamsakas, and they too clearly understood it as a necessary condition; cf. Sastradipika, ed. Kisordas Svami (Varanasi, 1977). p. 71,10-12: etac ca visesanatrayain upradadanena surrakarena karanadosabadhakajnanarahitam agrhitagrahi jnanam pravanam iti pramanalaksanam sucitam. "And the Sutrakara, inasmuch as he appropriates these three qualifications, has indicated the following definition of pramana: pramana is a cognition which is free from faults in the causes and from a sublating cognition (and) apprehends a non-apprehended (object)." As an afterthought it occurred to me that my interpretation of the beginning of the Pramanasiddhi chapter is somewhat similar to Kumarila's interpretation of Minamsasutra 1.1.4 as not containing a definition of perception, whereas Oetke's interpretation resembles that of the Vrttikara.

Page #2
254 E. Franco think not. My main reason for assuming that Dharmakirti did not intend to define pramana in general is the absence of such a definition in the Pramanaviniscaya and Nyayabindu. In other words, if Dharmakirti had wanted to define pramana in general at the beginning of the Pramanasiddhi chapter, he would have repeated his definition, maybe in a modified form, in the systematic presentations of his epistemology. To further substantiate this hypothesis I pointed out that no philosopher before Dharmakirti had attempted to define pramana in general and that the topic of the general validity of cognitions had become central to Indian philosophers only after Dharmottara. In his conference presentation, as well as in the slightly enlarged version of it printed in the present volume, Oetke completely ignores these corroborative points in particular and the broader historical context of Dharmakirti's statements in general. Consequently, his treatment of Dharmakirti is, as is often the case with analytical philosophers, ahistorical and anachronistic. After his presentation I asked Oetke several times why, in his opinion, Dharmakirti did not define pramana in general in the above mentioned works. The only reply I have received so far is the mildly amusing statement that Dharmakirti was not a professor of Indology and therefore did not have to repeat himself. If Oetke remains satisfied with this reply, I shall prefer not to continue the debate on this point. I would like to turn now in some detail to Oetke's points of criticism in the order in which they appear. First, Oetke claims that my remarks on the logical properties of the particle va "obfuscate" the issue because whether va should be understood in an inclusive or an exclusive sense is irrelevant to the problem. I agree that this question becomes relevant only after the truth of at least one of the two propositions connected with va has been ascertained, i.e., after the two propositions have been examined as to whether they refer to necessary or sufficient conditions. Nevertheless, I cannot see why, in my introductory remarks on the general problem seen here by commentators and scholars alike, I should not also look at the two possible intended senses of va under the assumption that Dharmakirti indeed named two sufficient conditions here. Further, Oetke chastises me for not taking into consideration the distinction whether va connects verbal phrases of the same sentence or whole sentences. First of all, this distinction which seems important to Oetke ("must be observed") was not addressed by me because the former alternative is hardly justifiable on the basis of the grammatical structure of the present passage. Neither Oetke, nor I,2 nor, to my knowledge, anybody else ever assumed that la and 5c should be understood as parts of a single sentence. As long as the overall grammatical structure of the relevant passage does not warrant the possibility of interpreting its propositional structure in an alternative way the introduction of such a distinction seems pointless to me. Second, it is not clear to me why Oetke claims that my remarks against assuming an exclusive sense would be valid only as regards the first type of connection because very often, and also in this case where obviously no quantifier or negation is involved, disjunctive coordination of verbal phrases with "or" is logically equivalent to disjunctive coordination of clauses. Perhaps he would like to clarify this point in his response and also to explain why he does not illustrate the two alternatives in a parallel manner. Third, it has to be noted that some native speakers of English feel that the difference between "pramana is 2 Cf. the addition of "la means of valid cognition]" in my translation of 5c which makes my understanding of the propositional structure more than clear.

Page #3
Response to Claus Oetke's Paper 255 a non-belying cognition or the illumination of an unapprehended object" and "pramana is a nonbelying cognition or pramana is the illumination of an unapprehended object" is precisely that in the second case the disjunction is exclusive. Thus, my brief remarks are not completely disconnected from the issue at hand, as Oetke suggests to the reader. Oetke's next criticism concerns the aspect mentioned by me that, according to some logicians, va does not formally have an exclusive sense. Contrary to what Oetke implies I did not claim or even wish to imply that this formal aspect excludes the possibility that due to internal reasons Dharmakirti intended an exclusive sense in the relevant passage. Perhaps at this place a word should be said about Oetke's tone. While welcoming debates and controversies, I think we should not confuse sharpness of argument with sharpness of expression. In p. 246 Oetke refers to the suggestion proposed by Katsura and others to interpret va in a conjunctive sense. I too disagree with this interpretation, but I do not see the point of calling it preposterous." Oetke's propensity to use "strong words" leads him also to distortions which are, to be sure, unintentional, but nevertheless unfair. Thus, referring to Prajnakaragupta's interpretation of va, he says (p. 250): "It should be maintained against Franco that such a reading is not linguistically eccentric," but all I said is that this interpretation "is not really convincing (Dharmakirti on Compassion and Rebirth, p. 50). I still think that it is not convincing, and even Manorathanandin, who usually follows Prajnakaragupta, did not think it to be so. Further, although it is not my task here to defend Lindtner's paper it is clear that Oetke does not do it justice. Lindtner does not question the assumption that Dharmakirti intended to define pramana, and Oetke basically criticizes him for not answering questions that he (Lindtner) did not raise in the first place. Oetke claims (p. 247) that my "proposal invites us to assume an argumentation on the part of the writer of the Pramanasiddhi which is formally analogous to the claim that the Buddha is an Eskimo because he is a featherless biped." This statement could only be made by a logician who has lost sight of the way people use language. In many contexts it is not inappropriate to say that, to remain with Oetke's example, an Eskimo is someone who lives in an igloo and hunts polar bears, even though some non-Eskimos may also live in igloos and not all hunters of polar bears are Eskimos. One may use a property which is typical for Eskimos to characterize them even though it may not be a necessary and sufficient condition for being an Eskimo. In what other way could Dharmakirti be expected to show that the Buddha is pramana if no one before him had defined pramana in general, if he himself had no intention to define pramana, and moreover if he was looking for a formulation that could in some sense be acceptable not only to himself but also to his Brahmanical opponents? It is only natural that he should mention characteristics that were widely associated with and typical of the pramanas. Of course, Dharmakirti's position would In this connection I would like to rectify Lindtner's statement (p. 157) that according to Siddhasena "pramana is in need of no further definition." Siddhasena does define pramana; cf. Nyayavatara lab: pramanam swaparabhasi jnanam badhavivarjitam. The statement which is quoted by Lindtner says that the individual pramanas such as pratyaksa and anunana need not be defined. Siddhasena's definition of pramana seems to presuppose a knowledge of Kumarila's concept of pramana, and because the latter is a contemporary of Dharmakirti it is probable that Siddhasena is later than Dharmakirti as claimed by Steinkellner, but doubled by Lindtner.

Page #4
256 E. Franco have been stronger if he had named characteristics which are sufficient conditions or necessary and sufficient conditions; in this case no further problems would have arisen. However, he did not do so, or was unsuccessful in this attempt, because the more natural reading of la and 5c is to interpret them as referring to necessary conditions, and consequently all the commentators had to grapple with this problem and propose different solutions to it. Oetke's own interpretation of the text presents possible solutions to the problem, some of them inspired by the commentators' attempts to turn the necessary conditions into sufficient ones by implication, but this does not account for the final return to the assumption of necessary conditions and the ensuing rejection of the disjunctive meaning of va by Manorathanandin as well as the later Tibetan tradition. This does not mean that Oetke could not explain this turnaround in the tradition if he cared to do so. As regards the understanding of the commentators' solutions to the basic problem within the framework of my own hypothesis, I argued that the commentaries are not faithful to Dharmakirti's intention--which was not to formulate a definition-because after Dharmottara's work on general validity a new situation had arisen. This reasoning, of course, is not directly applicable to the case of Devendrabuddhi, but it nevertheless explains why Devendrabuddhi did not elaborate on this topic. The same could be said also about pre-Dharmottara texts which are not direct commentaries on Dharmakirti's work, such as the Tattvasangraha; and indeed Santaraksita is more concerned with the topic of svatah/paratah pramanyam than with that of general validity. In any case, whether la and 5c are referring to necessary or to sufficient conditions, it does not follow from either that Dharmakirti intended to define pramana in general. Oetke also raises a "philological problem" that rests on the interpretation of verse 5d-6d whose function he takes to be a restriction of the scope of the specification formulated in 5c. It is easy to see that Oetke's argument begs the question. If 5c aims at the formulation of a specification of pramana which is a sufficient condition, then 5d-6d could be interpreted as restricting the scope of the specification formulated in 5c. On the other hand, if 5c does not aim at such a formulation, then 5d-6d could just present an additional clarification. In this connection I would like to add that I have serious doubts about the usual interpretation of 5c-6d which I previously followed myself. 5d-6d reads as follows: svarupadhigateh param //5// praptam samanyavijnanam avijnate svalaksane/ yaj jnanam ity abhiprayat svalaksanavicaratah //6// The usual interpretation of this passage, which is based on the commentaries, divides it into an objection (svarupadhigateh param // praptam samanyalaksanam) and its reply (the remainder of verse 6). However, there is no indication in the passage itself of such a division, and the essential negation is strangely missing. If one attempts to read 5d-6d independently of the commentaries, one may translate it as follows: Accordingly 5d-6d may be translated as follows: "[Objection:] The cognition of the universal that is subsequent to the apprehension of the own form [of the particular] would be [a means of knowledge]. [Reply: No.] because [in Sc we] intend that [only] the cognition in respect to an unapprehended particular [is pramana], for the particular is examined [here]."

Page #5
Response to Claus Oetke's Paper 257 "The cognition of a universal (in an inferential cognition which has occurred as different from after the apprehension of the own form (of a particular is pramana) because in Sc wel intend that the cognition in respect to a non-apprehended particular (which also appears in an inferential cognition, albeit with a different form, is prainana), for the particular is examined (here in connection with its two possible forms). I interpret this statement in the light of Pramanavarttika III 53-54 which says that an inferential cognition which has a universal as its object is nevertheless a means of knowledge because it apprehends a particular with a different form, which is not the particular's own form (tasya (svalaksanasya) svapararupabhyam gater meyadvayam malam). However, Dharmakirti's mode of expression in verse 5d-6d is too laconic to allow a definitive interpretation. What about Oetke's own solution to the problem at hand? According to Oetke's first possi.bility outlined in section IV as regards la (1) avisamvadi jnanam represents the specification of a necessary condition only. Oetke attempts to solve the problem that la may thus not state a sufficient condition by itself by assuming that the entire textual passage 1-5b represents the first definition. This suggestion seems highly improbable, and Oetke does not advance a single argument to support it. In fact, with the exception of 3a he does not even attempt to show how 1b-5b are relevant to his purpose. Further, his suggestion that Dharmakirti's procedure could be "similar to that of utsarga and apavada in grammar" is pure speculation. I am not familiar with any case in which an Indian philosopher proposed a definition in such manner, neither of pramana nor of anything else. Perhaps Oetke can supply in his response some instances to substantiate his claim. Oetke continues by suggesting that (2) we may also follow the view of Dharmottara according to which la implies 5c, an interpretation which similarly to Oetke's first suggestion aims at providing a restriction of the specification named in la by use of subsequent statements. In this connection Oetke claims that Dharmottara did not solve all problems connected with the topic of the definitions of pramana; which problems he has in mind he does not say. Maybe he would like to specify them in his response. In any case, this interpretation is also improbable. Dharmottara's suggestion is a brilliant move of a commentator who attempts to read a later theory into an earlier venerated text. However, if we assume that the implication suggested by Dharmottara was already intended by Dharmakirti, we would also have to assume that Dharmakirti did not care to be understood by his readers and perhaps even purposely misled them. Besides, it should not be overlooked that Dharmottara's commentary is on the Nyayabindu; thus, great care should be exercised when one uses his comments to derive his direct, word-for-word interpretation of specific passages in the Pramanavartrika. Further Oetke proposes that (3) one should associate the words avijnate svalaksane yaj jnanam in 6b-c for the purpose of restriction (3a) with 5c, or even (3b) with la either in combination with SI proposed this interpretation in my summary of the Pramanasiddhi chapter prepared for Potter's Encyclopedia of Indian Philosophies. Although this summary was written in the early nineties it is still not published and, given the speed of publication of other volumes of the Encyclopedia, it will not be published in the near future. I therefore take this opportunity to repeat it here.

Page #6
258 E. Franco 5c or (3c) with la exclusively. However, I fail to see that, in interpretations (3b) and (3c), 6b-c can be associated "equally well" with la. No matter if one accepts the commentatorial interpretation of verse 50-60 or my suggested reading (cf. above), the remark in 6b-c refers first of all to 50-6a; next one could interpret it as referring back also to 5c which is the starting point of the clarifying discussion in 50-60. To connect 6b-c additionally, to say nothing of exclusively, all the way back to la seems to be far-fetched and certainly not "equally" possible. I also wonder whether Oetke took into consideration that the semantic stress on uvijnate svalaksane would have to shift from svalaksane to avijnate to make the remark meaningful in association with la. Moreover, the first Sanskrit sentence construed by Oetke to clarify his reading of la in association with the complete relative phrase 6b-c is syntactically awkward, if not impossible in regular Sanskrit prose. Next Oetke modifies his proposal (3c) (and (3b]?) and suggests that only 6b, i.e., avijnate svalaksane, should be connected with la. Oetke concludes his deliberations on this topic by suggesting that avijnate svalaksane could also be understood as a locative absolute and thus be associated, either alone or together with yaj inanam, not only with la (and there, too, as a locative absolute?), but also with 5c in order to provide a sufficient restriction also of the specification named there (3b). However, I gather that this latter connection of avijnate svalaksane also with 5c would only have to be assumed if one does not follow Devendrabuddhi's attempt to arrive at a sufficient restriction because at the beginning of section IV Oetke considers his interpretation to be "quite natural and plausible." Moreover, Oetke's explanation as to why Dharmakirti connected the two definitions with va rather than with a conjunctive particle such as ca is not at all convincing. It is simply not the case that if he had used, e.g., ca instead of va it would have been "natural" to interpret the passage as an "enumeration of three conditions of pramana-hood or of various types of pramana"; the third condition would presumably be Buddhahood, or the third type everybody who is a Buddha. Oetke argues as if 7a would be connected to the preceding words by a simple ca, and not by tadvat which first of all points at a comparison. Further, I fail to grasp why the very employment of va supports the assumption that Dharmakirti intended to provide two definitions of pramana. I could easily reformulate Oetke's argumentation to fit my own assumption that Dharmakirti intended to provide two characteristics typical of pramana. Now, the final question arises if Dharmakirti, according to Oetke, really proposed two definitions and not merely one. Oetke gives four different answers to this question. His first answer is that although the two definitions refer to the same range of objects they are not conceptually identical. Thus, the employment of va has the pragmatical purpose to indicate their equivalence which is not self-evident. This "enterprise," as Oetke grandly calls it, is theoretically possible, but it would be rather unusual for the historical period with which we are concerned. In Dharmakirti's time, in contradistinction to the Navya-Nyaya period starting with Udayana when proposing more than one definition for the same thing" had become a playful activity, it was not common to propose two definitions. I invite Oetke to inform us in his response about a parallel case in order make his suggestion more plausible, if indeed he thinks that it is plausible and not merely possible. Oetke's second suggestion basically repeats Prajnakaragupta's interpretation. I already mentioned above that I do not consider this interpretation "linguistically eccentric," but nevertheless it seems improbable to me. Neither here nor to my knowledge elsewhere in Dharma

Page #7
Response to Claus Octke's Paper 259 kirti's writings are there any indications to the effect that he intended the first explication of pranana to be applicable only to empirical reality, and the second (also) to absolute reality, and that he himself used the two explications with this distinction in mind. Oetke's third explanation suggests that the first definition is "tailored to Dharmakirti's own theoretical assumptions whereas the second alternative represents a more general explication," although with the addition of Dharmakiti's own epistemological assumptions the second alternative can become equivalent to the first. This partially dovetails with my own hypothesis that Dharmakirti formulated the two characteristics intentionally in such a way that they could be acceptable in one sense to himself, in another to philosophers of different persuasions; unlike Oetke, however, I argue that this applies to both characteristics. Finally, Oetke proposes that the two definitions are equivalent but that the first one is more suitable to indicate the fact that the one-place predication of something as pramana can be derived from a relational concept. Oetke himself admits that the last alternative is difficult to establish. Neither here nor in the case of the various interpretations of la and 5c as referring to necessary and sufficient conditions does Oetke attempt to narrow down the philologically and logically possible readings to such which are most plausible within the wider historical context of Dharmakirti's thought and within the conceptual framework of his works. Therefore, I can only concur with Halbfass' remarks on Oetke's Bemerkungen zur buddhistischen Doktrin der Momentanheit des Seienden (Wien, 1993): "Questionable, however, is the one-sidedness with which Oetke concentrates on the logical-analytical dimension of Dharmakirti's thought pursuing thereby an ideal of precision that is prescribed (vorgegeben) by modern logic, but not by Dharmakirti's own orientation and its historical context. Although rudimentary stages of wishing to understand historically and following attentively and carefully Dharmakirti's thought are certainly present they are again and again superseded and covered over by free exploration of intellectual possibilities which is not so much concerned with what Dharmakirti really has thought, but rather with what he could have or should have thought. One may be tempted to speak here of 'analytical overkill'. Greater precision pursued for its own sake leads by no means eo ipso to greater clarity and better understanding." Cr. W. Halbfass, "Arthakriya und Ksanikarva: Einige Beobachtungen," in Bauddhavidyasudhakarah: Studies in Honour of Heinz Becher on the Occasion of His 65th Birthday, ed. P. Kieffer-Pulz and J.-U. Hartmann (SwisttalOdendorf, 1997), p. 237.