Dharmakirti On Inference Of Effect

Author(s): Ernat Steinkellner
Publisher: Ernat Steinkellner
Catalog link: https://jainqq.org/explore/269403/1

Page #1
I did not receive proofs, the necessary corrections are added now.) Dharmakirti on the Inference or Erfect (karya) ansinin By Ernst Steinkellner. Vienna arm slitn. Several recent publications have shown how Dharmakirti deals with a set of inferences which use logical reasons Chetu) that are seemingly not included in any of the three kinds of reasons (trividhu netu) he proposes in the first presentation of his new theory of logical reasons, and moreover point to a functional connection for the most conspicuous of these, namely, the inference of effect (karyanumana). with the theory of non-perception as reason (anupalabdhihetu). Twats mainly interpreting the relevant PASSAges of PVT III where the discussion is already more comprehensive than in PV - PVSV shows that Dharmakirti explains how such inferences, 18 acceptable, can be reduced to either svabhava or karya as logical reasons. The rirst inference Dharmakirtl deals with is the inference of an effect (karya) from its cause (karana). Dharmakirtl shows how the reason Iri this inferorice must be understood in order for it to be acceptable. And that in this case the reason "Complete Cuse" (Sangra hetu) must be taken as a case or SVabhava as logical reason. Iwata further demonstrates how Dharmakirtl also uses this reduction to also reduce the reason "non-perception of something

Page #2
perceptible" (drsyanupalabdh to a case of svabhAVR as reason, taining a clear expression of a subsumption of the anupalabdhihetu This use of the reduction of the reason "complete cause to a case or under the sabhavahetu category only refers, merely in conclusion, to a later clarification, namely to PV IV 269. This sentence (ata svabhavahetu is also elaborated by Tani." eveyam.... vaksyamah. PVSV 105,1-3) must therefore be considered as a later Insertion belonging to the time when Dharmakirti was composing the Pramana varttika as a whole, but as not belonging to the text of the original 'Hetuprakarana. The systematical usage of the interpretation of the reason This motivation of a necessity for the anupalabdhihetu in order "complete cause as an essential property (svabhava) in an Inference to allow its reduction to a case of svabhavahetu - although already of the consequence (sadhya) fitness for the production of the effect" presented at a very early stage - cannot therefore have been for an explanation of the anupalabdhihetu as being included under Dharmakirti's original motivation for dealing with the structure of the category svabhavahetu, may look like a strong motivation indeed the inference from cause to effect (karyanumana). for Dharmakirti to deal with the problem of a possible interpretation of the Inference from cause to effect. However, we can see from the contextual situation in the "Het uprakarana, that is not very likely to be the original motivation II. There it is only in the later, more elaborate discussion of the anupalabdhihetu (PVSV 101,3-107.14) that this kind of reason is explained as a case of svabhavahetu because of the fact that the Inference of asaj nana)sabdavyavahara (PVSV 4,22; 104,27) from the reason upalabhyanupalabdhi (PVSV 104.28) is characterized as an Inference from cause to effect (karanat karyanumanalaksanatvat. PVSV 105,17.). And although this characterization is based on the preceding words (PVSV 104.28-105,1), where every effect for which the presence of its complete causes as capable for its production applies, is adduced as an example (drstanta), the sentence con When we look at the context of the introductory section of the Hetuprakarana, another, much more natural motivation becomes apparent. After introducing the theory of the logical reason Chetu, PV I 1 - 3) and explaining it in its formal aspects (trilaksanatheorem: PVSV 1,12-2.14) and in its material aspects (trividhatheorem: PVSV 2,14-6,21), Dharmakirti discusses the possibility of Inferring from a cause to its effect (PVSV 6,2277.). In this discussion he clearly distinguishes the alternatives as to whether the cause is complete (samagra) or incomplete (asamagra)". In treating the first alternative (PVSV 6,22-8.15), he shows how the "complete cause" as 712 713

Page #3
reason can be understood as being included among the three kinds of inference from reasons not directly based on causal connections with the inferred consequence, e.g. from taste to colour (PV1 9b = 11b). reasons he has just introduced This first section. Where Dharmakirti explains cases of from smoke to transformations of firewood (PY I 90 = 11d), or from a disturbance of ants or the changing behaviour of fish, such as Jumping, to rain (PVSV 8.13-15). These inferences are all reduced to cases where an effect (karya) serves as logical reason.1? acceptable reasons and shows how they can be fitted into his theorem of the three kinds of reasons (trividha hetu), is again divided into two parts: 1. PVSV 6,22-7,12, treating the Inference from the reason complete cause". and 2. PVSV 7.12-8,15, treating cases such as the Inference from the reason "taste" (rasa) to the consequence "colour etc." (rupid, where a causal relation cannot be relled upon directly, because reason and inferred consequence In this second part, the motivation for explaining the cases of reasons as cases to be subsumed under the category of karyahetu clearly lies in the fact that we are here confronted with a collection of inferences that are accepted in the mundane world'1' which was are simultaneous. entitled to hear from Dharmakirti to just which of his three kinds of reason the reasons used in these inferences would belong. Both systematically and because of the parallel Introductions, these two parts from the first section are of equal value as contextual alternatives: "inference from x to y as directly based on causation against "Inference from x to y as not directly based on causation". Thus the first alternative complete cause : incomplete cause" (PVSV 6,221f.: PVSV 8,1611.) is not decisive for the analytical However, the motivation for explaining the position of the first logical reason within his system of the three kinds, namely the reason "complete cause, can be seen in the absence of other indications in this context, and without any interpretation Influenced by the knowledge of its later application in Interpreting the anupalabdhihetu as a case of svabhavahetu - as not lying primarily I'mundane acceptance". It lies rather in a similar fact, namely an influential tradition with the right to hear from Dharmakirti how the reason under discussion fits into his scheme of structure of this part of the text and must be subsumed under a structure which can be inferred from what Dharmakirti actually achieves in these two sections, 1.e. an explanation of reasons that can be accepted within the frame of his theorem of the three kinds, and of reasons that cannot be accepted. In other words, while the the three kinds of reasons, and would not invalidate this scheme by cases of reasons in the first section are valid, the cases in the latter not fitting into it. are invalid.11 The difference to the mundane" reasons referred to above and The second part (PVSV 7.12-8.15) of the first section can now reduced by Dharmakirtl to cases of reasons as karya is striking. also be excluded from the present investigation, because it treats while we have actual instances of concrete inferences in this case 714 715

Page #4
IV. which are "accepted in the inundane world". we have here the general concept of complete cause as the reason from which the arisin This is how I understand Dharmakirti's words the effect" (karyotpada) is i rred and effect itselr (karyam eva) (Question:) IA] When the arising of an effect (aryotpada) is is said to be not inferable. This means, of course, that Dharmakirti Inferred from a complete (Samagra) cause, how is this [logical reason is not dealing here with "mundane reasons, but referring to the conception of an important logical theory which was so influential "complete cauce") included in the threefold logical reason? (Answer:) IB The arising of an effect (karyotpada) which is inferred from a completelo cause (as logical reason) is that it had to taken into consideration. III. This logical concept can in all probability be identified with a - for our historical context - prominent part of the theory of a trividha anumana as propounded by competitive traditions of logical called an essential property (svabhava) (of this complete cause), because this arising of the effect does not depend on (any) other things (for its occurrence. (v.79) IC) This arising of an effects, too. Is an essential property (svabhava) of the entity fconsisting in the complete causel as following only this (complete cause), because (t) does not depend on anything else but the (cause) as eodositulot (yathasamnihita). IDI Por in this case for the inference of the arising of an effect" only the possibility (sambhava)az of the arising (utpatt of an effect is inferred from the complete cause, (E) because the fitness of the complete (causes which constitute the causal complex) for a pro thought In Samkhya and Nyaya. It is the kind of inference called purvavat, when the term is Interpreted as a possessive adjective with the surfix matup.!! In this case the Inference from cause to effect" is well-established in the tradition as one of the possible meanings of the term.. And although this theorem - in its own historical origin - evidently has also the purpose of allowing for 'mundane Inferences, eg the inference from (rising) clouds to (Imminent) rain, it is in the form of a theorem belonging to rival logical schools that Dharmakirti deals with this kind of mundane Inferences from cause to effect. duction (ut padanayogyata) of the effect is inferred. [F] And the Otness (yogyata) follows only the complex lof the complete causes]. [G] Therefore it is inferred as being nothing but an essential property of the complete complex of causes). (Question:) (H) Why, on the other hand, is not the effect itself (karyam eva) inferred from the complex of causes a 3 717

Page #5
(Answer:) 111 If an effect fitself that is connected with trans Thus Dharmakirti explains just how the Inference from cause formations of the forces which result from earlier phases of the complex is to be inferred from a cause). (this cause) is inconclusive (as logical reason), because an Impediment (pratibandha) is possible. (v.8=10) to effect" has to be interpreted In order to be acceptable. An Inference of the effect itself is quite impossible, since Impediments to its production cannot be ruled out and the causal complex used as 1J) For the different) cause-substances (which constitute the complex of causes do not actually produce their effect just because of being (conceived of as) complete, (K) because the actual) arising of the effect depends on the (respective) transformations of the forces which arise from the respective preceding complex. IL) And since an impediment is possible within this period of the complex transformation), there is no inference of the effect itself). logical reason remains inconclusive (analkantika) in this case. This is shown in a digression ([HI-ILI) with the purpose of reinforcing the main argument (Bl-IG) and (M-N) by refuting an alternative conception of this inference: The main argument starts from a different Idea: the term 'arising of an effect" (karyotpada) is interpreted in ID as "the possibility of the arising of an effect" (karyotpattisambhava), and identified again in (E) as "fitness for a production of the effect" IM The fitness (yogyata) of the complete complex], however, does not depend on substances other than this complete complex). Therefore the following inference is not contradicted: "This complex of causes is capable for a production of its) effect on account of a Irinal] transformation of the respective later forces in its causally connected continuum of phases), because another I.e. cooperativel condition for the transformation of the cause's force, that the production of the effect) would necessarily depend upon, is lacking." [N] The fitness of the complex [for a production of its effect) is called (karyotpadanayogyata). In other words. If the arising of an effect" (karyotpada) is inferred from a complete causal complex, this can only mean - according to Dharmakirti - this complex' fitness to produce the effect" (karyotpadanayogyatd). And the latter, as dependent on this very complex alone, Is in itself an essential property (svabhava) of this complex. In this way, the Inference from an essential property (svabhava) "being a complete causal complex as logical reason to another essential property (svabhava) "being fit for the production of an effect as logical consequence is established as Dharmakirti's conception of the inference from cause to effect". And the reason "Independent of other causes)". because only & preceding similar (salatl-}3(complex) is the cause for the arising (prasuth of (this) force.25 718 719

Page #6
"complete cause of the initial question 1A is thus explained as a of the purvavat kind of Inference 29 Vrsagana explains purvavat as case of an essential property as logical reason (svabhavahetu) the case of an inference when from the observation of a complete cause (rgyn ma tshan ba med pa, avikalahetu) the Imminent arising (byun bar 'gyur ba nid, 'bhavisyattva) of an effect is cognized, like the imminent arising of rain from the observation of the presence of clouds".30 In this definition by Vrsagana," both terms used by Dharmakirti in discussing the inference from cause to effect", 1.e. "complete cause" and "arising of effect" are clearly present. The question nos arises of why Dharmakirtl starts this Interpretation of the Inference from cause to effect" with a formulation IA) that lends itself - because of its peculiar expressionsto - rather easily to be interpreted into an acceptable direction. This would, of course, be a good reason in itself. But there In conclusion: Dharmakirti's Inference "from a complete cause to the arising of the effect" is either entirely his own explanation of * possible meaning of the inference from cause to effect with only accidental" similarities to Vrsagana's formulations, or - and I think this is much more plausible - his acceptance of this inference in the form to be found with Vrsagana, Interpreting it as an Inference from an essential property as logical reason (svabhava 18 a strong case for seeing more in these expressions than just a technique of linguistic variation for polemical purposes alone. The clue lles in the expression for the effect. For the expressions "complete cause" (samagra hetu) or (causal) complex" (karanasamagri, samagri) are due to the fact that Dharmakirti considers an effect always to be produced by a complex of causes, and never by a single cause alone.27 in digressing on the possibility of Inferring effect itself" (karyam eva) in th). Dharmakirt seems to be referring to the formulation of the Nalyayika Paksilasvamin, who reports an older Interpretation of the purvavat kind of Inference in NBh 146,4 simply hetu), while simultaneously refuting a Nyaya formulation of the same Inference. Since Dharmakirti clearly says that the origin of the effect" as yatha karapena karyam anumiyate, yatha meghonnatya bhavisyati vrstir 1.2 interpreted as "fitness of the complete cause for producing its effect and conceived as the logical consequence (sadhya) is to be taken as an-essential property (svabhava) of this "complete cause', we are in fact facing nothing more than the case of an inference operating with an essential property as logical reason for the purpose of conceptual determination . In this case, it is the inferential determination of the concept "fitness for the production of the The expression arising of an effect" (karyotpada), on the other hand, can be found in the Sankhya teacher Vrsagana's explanation 720 721

Page #7
effect as an essential property of the causal complex on the basis sense of referring only to the last phase (antyavastha) of the causa! of its completeness ('samagrata). complex. In short: When Dharmakirti clarifies the possible meaning of an Inference from cause to effect in the Hetuprakarana, he refers to VII. 'cause in the sense of the continuum of the complete causal complex and states that due to the possibility of impediments (pratibandha) There is a systematical question that needs to be answered at the effect cannot be inferred from it. It is, however, possible to this point before we can continue to investigate the usefulness of Infer the concept to be fit for the production of the effect". In other this structural scheme. words, a causal complex which is called "complete" can also be called fit for the production of the effect". Strictly speaking, a causal complex (hetusamagri) can be called "complete" only when referring to its last phase (antyavastha, antyaksana). It seems, however, that Dharmakirti did not press the Issue that far in the 'Hetuprakarana.. Only with the introduction of the new category of a "non-perception of effect" (karyanupalabdh In explaining the only possibility of understanding this type of Inference, and systematically identifying it as an Inference from an essential property (svabhava), Dharmakirti in addition presents a structural scheme that he will later also use to explain the in the Pramanaviniscaya do we find the relevant rigorous formul anupalabdhihetu. ations. If we consider the complete cause of our context of the * Hetuprakarana to be meant as the causal complex in its last phase, an Inference with regard to neither an effect as such nor a fitness VIII. for the production of the effect would be meaningful because of the perceptible presence of the effect even before any possible This structural scheme amounts to a description of the conclusion of the inferential process.97 It is therefore clear that Dharmakirti refers only to the "continuum" (santana) of the causal complex (hetusamagri) when he possibilities and limits of conceptually determining the idea of "progressive causality": what can be said about causes as regards their necessarily producing effects? Or: In what way are causal processes entelechially" spontaneous or automatic? speaks of a complete cause" (samagra hetu) as logical reason in the 'Hetuprakarann. It is also clear that his usage of the attribute "complete" (samagra) in this context must not be taken in the strict As far as I can see. Dharmakirti did not put this scheme to any Interpretational use in his first work other than to explain the 722 723

Page #8
logical nature of the anupalabdhihetu. Nor did he do so later. It is nevertheless of considerable import for Dharmakirti's Buddhism. And, to be clear from the outset, I consider the true motivation for Innate nature (atyantasatmata) and thus perfection:** "It inerey which originates from its respective seed (bua) is not impeded by opposing forces (vipaksa) which (also) originate from their respective seeds. It develops to highest innate nature (atyantasatmata) in the continuum of the] Dharmakirti's treating this particular inference from cause to effect" in its possibilities to be not logical, but soteriological, the path towards Buddhahood itself being at stake. mind. 45 (PV II 129) Inami and Tillemans, in their introduction to dGe 'dun grub pa's In other words: If all opposing forces which could impede the arising of a cause's effect are eliminated from the mental continuum, analysis of the second chapter of the Pramanavarttika already have drawn attention to the fact that in PV 11 34-164a', where Dharmakirti discusses the four famous epithets of the Buddha from the "complete" or "sufficient cause or causal complex would give rise to its effect. Thus, although only the cause's fitness for the production of the effect would be inferable strictly speaking the mangala verse of Dignaga's Pramanasamuccaya "in direct order" (anulomena), a causal progression towards Buddhahood is shown according to our scheme, an additionally ascertained absence of where "each preceding epithet is said to be the cause for the one Impediments would also guarantee the actual arising of the effect. which foliows":41 Devendrabuddhi in explaining PV II 146 clearly states that this This additional moment is also already implied in the second part of the text examined above, namely PVSV 7,5-7, where Dharmakirti rejects an inference of effect as such (karyam eva) on is a case where the effects are inferred as "possible" (srid.pa) from the respective preceding causes.'? Inami and Tillemans, in line with account of the possibility of impediments (pratibandhasambhavat). some Tibetan opinions, therefore also draw attention to PV I 7=9 as of course, ascertaining the absence of possible Impediments remains providing the theoretical scheme for this argumentation." a crucial point in this extension of the original minimal scheme and must be considered as a necessary corollary to such an inference "from cause to effect". However, given the theoretical possibility of ascertaining the absence of impediments, we may deduce from of course, mere "possibility or fitness' to give rise to an effect would not be a sufficiently strong quality of the respective causes in this section to connect a causal chain in such a way that there is any certainty in the progression towards Buddhahood under discussion. But Dharmakirti gives a clue with regard to the crucial necessity of the effect's origination from a cause in his explanation Dharmakirti's complete explanation of this inference in PVSV 6,22 7.12 that not only a fitness for the production of the effect" can be of the possible development of mercy" (krpa) towards its highest Inferred from the respective, 1.e. complete cause, but also the effect Itself as actually, or rather necessarily? originating from it. An 725 724

Page #9
IX. "unimpeded continuum of a causal complex would then also be a It will have to be left to future research in what way Dharmakirti's followers and commentators made use to any larger possible logical reason for an Inference of its effect, I. e. the respective later phases of the same continuum. For - as Dharmakirti extent of this connection of a minor logical theorem with the Buddhist spiritual background. There are, at any rate, some says later - "the impediment to a causal complex in the production of its specific effect is defined as the production of something that does not agree with the arising of this specific) effect"." Indications for the fact that Dharmakirti's structural scheme for the Thus, here, in the 'Hetuprakarana, Dharmakirti provides the theoretical frame for the idea of - what I would tentatively call - an "entelechial causality", without which the whole "progression Inference from cause to effect' as presented in PV I 7=9 was later understood and used as an explanatory structure for other acceptable cases of "entelechial causality": I have come across two such Instances in post-Dharmakirti towards Buddhahood" as depicted in the anuloma section from the literature that merit some attention in this context and may serve second chapter of the Pramanavarttika would be inconceivable and Illusory as reason to look for similar examples: a few sentences in a Buddhist purvapaksa of Vacaspatimisra's Nyayakanika,9+ and the Tattve However, although Dharmakirti presents in this chapter a case siddhi ascribed to Santaraksita as a whole. Both are related to the where his structural scheme for an inference from cause to effect" theme of yogic cognition (yogMana) and seem clearly to support the Is actually applied, I cannot find any Indications in his writings that proposed hypothesis that Dharmakirti's religious presuppositions as a Mahayanistic Buddhist must be considered as the final motivation he intentionally elaborated this logical scheme in order to provide a logical structure for this particular application. The connection of his argument in the anuloma section of PV II with the structural for his elaborating the logical possibility of an Inference of effect (karyanumana). However, this is already a tople for another paper. Idea present in PV I 7-8 = 9-10 is nevertheless evident. And it is also evident that, as a Buddhist, he must have felt the need to conceive his new logical thought in such a way that Inferences "from cause to effect" were not altogether impossible. ABBREVIATIONS: TBV Tattvabodhavidhayini: Acarya-sri-Siddhasena-Divakarapranitam Sammatitarkaprakaranam ... Srimad-Abhayadevasurinirmitaya Tattvabodhavidhayinya vyakhyaya vibhusltam. Ed. S. Sanghavi and B. Dosl. 16 Vols.) Ahmedabad 1924 - 1931. 726 727

Page #10
NBh cf. NSu NBhus Nyayabhusana:Srimad-acarya-Bhasarvana-pranitasya Nyayasarasya svopariam vyakhyanam Nyayabhusanam. Ed. Svami Yogindranandah. Varanasi 1968. Yuk tidiplka. An ancient Commentary on the SamkhyaKarika or Tavarakrsna. Ed. Ram Chandra Pandeya. Delhi 1967. Samkhyakarika: cf. YD Nyayamartari of Jayantabhatta with Tippani - Nyayasaurabha by the Editor. Ed. K. S. Varadacharya, 12 Vols.) Mysore 1969, 1983. Syadvadaratnakara: Srimad-Vadidevasuri-viracitah Pramananayatattvalokalankarah tadvyakhyaca Syadvadaratnakarah. Ed. L. Motilal. (5 Vols. Poona 1926-1930. cf. NSu Hetubindu: Ernst Steinkellner, Dharmakirti's Hetubinduh. Tell I. Tibetischer Text und rekonstruierter Sanskrit -Text. Wien 1967. Nyayasutra. In: Nyayadarsanam. With Vatsyayana's Bhasya, Uddyotakara's Varttika, Vacaspati Misra's Tatparyatika & Visvanatha's Vrtti. Critically edited with notes by Taranatha Nyaya-Tarkatirtha and Amarendramohan Tarkatirtha. (2 Vols. Calcutta 1936-1944 (Reprint Kyoto 1982). PV I Pramanavarttika, Chapter 1: LITERATURE: Raniero Gnoli, The Pramanavarttikam of Dharmakirti, the First Chapter with the Autocommentary. Text and Critical Notes. Roma 1960. Acarya Dharmakirti krta svarthanumanapariccheda Ed. Dalsukh Bhai Malvaniya, Varanasi 1959. Frauwallner 1954: E. Frauwallner, "Die Reihenfolge und Entstehung der Werke Dharmakirti's", in Asiatica, Festschrift Friedrich Weller, Leipzig, 142-154. PV II.III.IV Pramanavarttika, Chapters II, III, IV: Pramanavarttika-Karlka (Sanskrit and Tibetan). Ed. Yusho Miyasaka. Acta Indologica 2 (1971/72), 1-206. [Chapters II, III, IV correspond to I, II, IV in Miyasaka's order.) Frauwallner 1958: Erich Frauwallner. "Die Erkenntnislehre des klassischen Samkhyasystems". WZKSO 2, 84-139 (Reprint in Kleine Schriften, Wiesbaden 1982, 223278. Iwata 1991: PVT Pramanavarttikatika (Sakyabuddhi): P 5718 Pramanavarttikasvavrtti (Dharmakirti): cf. PV Takashi Iwata, "On the Classification of three Kinds of Reason in Pramanaviniscaya III - Reduction of Reasons to svabhavahetu and karyahetu', in Studies in the Buddhist Epistemological Tradition, ed. E. Steinkellner, Wien, 85-96. PVSV PVSVT Pramanavarttikasvavrtticika (Karnakagomin): AcaryaDharmakirteh Pramanavarttikam (svarthanumanaparicchedah) svopa navrttya Karnakagomi-viracitaya tattikaya ca sahitam. Ed. Rahula Samkstyayana. Ilahabad 1943. Krasser 1991 Dharmottaras kurze Untersuchung der Gultigkeit einer Erkenntnis. Laghupramanyapariksa. Teil II. Obersetzung, Wien. Malvaniya 1959 Pin 1 Pramanaviniscaya, Chapter II: Ernst Steinkellner, Dharmakirti's Pramanaviniscayah. zweites Kapitel: Svarthanumanam. Teil 1, Tibetischer Text und Sanskrittexte. Wien 1973. D.Bh. Malvaniya, "Introduction" in AcaryaDharmakirti krta Svarthanumana-pariccheda. Varanasi, 1-27. Mimaki 1980 PVin III Pramanaviniscaya, Chapter III: P 5710. Katsumi Mimaki, "Sur le role de l'anatarasloka ou de samgrahasloka", in Indianisme et Bouddhisme, Melanges offerts & Mgr Etienne Lamotte, LouvainLa-Neuve, 233-244. PVINT Pramanaviniscayatika (Dharmottara): P 5727. 728 729

Page #11
Mookerjee/Nagasaki 1964 S. Mookerjee and Hojun Nagasaki. The Pramanavarttikam of Dharmakirti. An English Translation of the First Chapter with The Autocommentary and with Elaborate Comments [Karikas I-LI). Patna. Steinkellner 1967 Ernst Steinkellner, Dharmakirti's Hetubindub. Tell II. Ubersetzung und Anmerkungen. Wien. Steinkellner 1971 Ernst Steinkellner, "Wirklichkeit und Begriff bei Dharmakirt!", WZKS 15, 179-211. Steinkellner 1979 Ernst Steinkellner, Dharmakirti's Pramanaviniscayah, zweltes Kapitel: Svarthanumanam. Teil II. Ubersetzung und Anmerkungen. Wien. Tani 1991 Yalta 1984 Yalta 1985a Yalta 1985b - Tadashi Tani, "Logic and Time-ness in Dharmakirti's Philosophy Hypothetical Negative. Reasoning (prasanga) and Momentary Existence (ksanikatva)", in Studies in the Buddhist Epistemological Tradition, ed. E. Steinkellner, Wien, 325-401. Hideomi Yalta, "Hosho no 'hininshiki' [Dharmakirti's anupalabdhi". In Makio Ryokalhakushi shojukinen ronshu, Chugoku no shukyo shiso to kagaku, Tokyo, 35-45. H. Yalta, "On anupalabdhi, annotated translation of Dharmakirti's Pramanavarttikasvavrtti (1)", Taisho Daigaku Daigakuin Kenkyu Ronshu 9, 216 199. H. Yaita, "On anupalabdhi, annotated translation of Dharmakirti's Pramanavarttikasvavrtti (II)". Chisan Gakuho 34, 1-14. 730 I gratefully acknowledge the helpful criticism and remarks of Toru Funayama, Helmut Krasser, Monika Pemwieser and Prof. Tilmann Vetter on this paper. 1 PV 11-63-8 with PVSV. Iwata 1991. Iwata has also prepared a German translation of PVin III 310b7-313a3 (vv. 6467), the basis of his paper of 1991, to be published soon and which he has generously placed at my disposal. Iwata 1991: 87-89. This connection already has been elaborated in Yaita 1984: 3739 (cf. Yalta 1985a: 209 and notes 94-96) and Steinkellner 1967: 155. Tani 1991: 328ff. I use this hypothetical title, which was proposed by Frauwallner (1954: 152) for the text of PV I together with PVSV as being before its incorporation into the complete Pramanavarttika as its first chapter together with a commentary ("sva"vrtti) a work of Dharmakirti's youth, a work that was composed as an original unit of verse and prose (cf. also Malvaniya 1959: 11f.; Mimaki 1989: note 46). Translated in Yalta 1985a and 1985b. hetuna yah samagrena... anumiyate/ (PV I 7ab=9ab) and hetuna tv asamagrena... anumiyate / (PV I 11ab=13ab). The second alternative deals with those cases where a valid inference is Impossible according to Dharmakirti, because the resultant cognition remains doubtful. This part (PVSV 8,16-10,25) does not concern us here. 10 yas tarhi...... anumiyate, sa katham trividhe hetav antarbhavati (PVSV 6,221.) and ya tarhi ...-gatib, sa katham (PVSV 7,12f.) where the rest of the sentence trividhe hetev antarbhavati has to be supplied from the first question (cf. PVSVT 45,291.). 11 This is also supported by the analysis of, e.g.. 'U yug pa who distinguishes the latter as invalid (Itar snar) cases while he adds the former to the valid ones (yan dag). Cf. Y. Fukuda and Y. Ishihama, A Comparative Table of Sa-bead of the Pramanavarttika Found in Tibetan Commentaries on the Pramanavarttika. Tokyo 1986, la-2a. Dharmakirti finally chose to rearrange the texts involved here in accordance with their contextual structure in PVin III. There, the alternatives "complete cause : incomplete cause" serve to distinguish valid and invalid reasons in his discussion of a possible reduction to svabhavahetu (PVin III 310b7-311a8 has two parts. 310b7-311a4: "complete cause", and 311a4-8: "incomplete cause" as reason). 12 This theme of explaining certain well-known inferences as cases of karyahetu is later expanded by Dharmakirti in PVin III 311a8-313a3 (vv. 65-67) under Incorporation of the text and the examples of PV I and PVSV. Cf. Iwata 1991: 89-94. 13 Iwata 1991: 85. 731

Page #12
14 Cr. NSU 1.1.5.: atha tatpurvakam trividham anumanam purvavae chesavat samanyato drstam ca: and SK 5be: trividham anumanam akhyatam t al Ingalingipurvakam. Karnakagomi sveomin's (PVSVT 44.131.) commentaries. It does, however, occur in the parallel text of Plin III 311al. 24-S t -(cf. PVSVT 45,6) : ran gi rigs (-svrati-) Tib. (408b7), PVT 25al. 19 Cr. A. Wezler. "Dignaga's Kritik an der Schluslehre des Nykya und die Deutune von Nyayasutra 1.1.5., ZDNG, Supplementa 1, Teil 3. 1969, 836-842. Wezler gives a survey of the interpretations of the terms purvavat and sesavat as known to Dignaga and Paksilasvamin, and thus also to Dharmakirti. 1. This would then be the tradition present, e.g., in the first interpretation of the terms in Paksilasvamin's Nyayabhasya (NBh 146,4) and in Vrsagana's interpretation (cf. below note 30). 19 C. Mookerjee/Nagasaki 1964: 3211. A new translation of the whole text by Hayes and Gillon was also available to me (Richard P. Hayes and Brendan S. Gillon, "Introduction to Dharmakirti's Theory of Inference as Presented in Pramana varttika Svopajfavrtti 1-10. I am grateful to T. Punayama, Kyoto, for providing me with a copy of this manuscript). Parts of the text are translated in Steinkellner 1971: 1857 (PV I 7-9 and Vrtin: Yalta 1985a: note 95 (PV I 7=9), Krasser 1991: note 263 (PVI 8=10 and Vrton, and of the parallel text In PVin 111 in Iwata 1991: 86. vas tarhi samagrena hetuna karyotpado numiyate, sa katham trividhe hetav antarbhavati hetuna yah samagrena karyotpado 'numiyate / arthantaranapeksat vat sa svabhavo nuvarnitah // (v.79) As y apl ynthasamnihitan nanyam apeksata Iti tanmatranubandhi svabhavo bhavasya tatra hi kevalam samagrat karanit karyot pattisambhavo 'numiyate. Samagram karyotpadanayogyatanumanat. yogyata ca samagrimatranubandhinity svabhavabhutalvanumiyate. kim punah karanasamagryah karyam eva nanumiyate. samagriphalasaktinam parinaminubandhini / analkantikata karye pratibandhasya sambhavat // (v.8=10) na Msamagranity eva karanadravyani svakaryam Janayanti, samagruanmanin Saktinam parinamapeksatvat karyotpadasya, atrantare ca pratibandhasambhavan na karyanumanam. yopyatiyas tu dravyantaranapeksatvan na virudhyate numanam -uttarottarasaktiparinamena karyotpadanasamarthyeyam karanasamagri, Sakti par namspratyayasyanyasyapeksaniyasyabhavad Tu purvasalatima raheturvat Sakti prasoteh smagry yogyatananyapeksinity uyate. (PVSV 6,22-7,12) I use "arising" here to translate ut pada and utpatti, its paraphrase In PVSV 6.271., as distinguished from "production to translate utpadana 1. It is interesting to note that the attribute "capable" (nus pa, 'samartha) is substituted for complete" (samagra) in the parallel text of PVin III 64 (PVin III 310b7; cf. Iwata 1991: 86). 1. l.e. hetunk ... samagrena in the position of "cause, and karyotpadab in the position of "effect". IT cf. T. Vetter, Erkenntnisprobleme bel Dharmakirtl, Wien 1964. 181., and Steinkellner 1971: 1871. * Sakyabuddhi's explanations rgyu tshogs pa de ni (PVT 23b8) and 'di yan tes bya ba ni reyu tshogs pa'o/ (PVT 24al) must be a mistake, for he continues saying it does not depend on another cause but the complete cause as constituted" (rgyu'l tshogs I ltar ne ba las reyu glan la miltos te / PVT 24al). Cr. also PVSVT 43,20 karyotpadanayogyatamatranubandhitvat as the reason for karyor pada to be the svabhava of some karana, and PVSVT 43.221.: asav apiti karyot padah. 1. That this formulation is again not to be taken at its face value is then explained by Uddyotakara (cf. NV 147.5ff.). However, this need not concern us here. It is sufficient that karya has been offered as the form of the logical consequence in this Inference in early Nyaya. 11 I follow - with Iwata 1991: 86 - Dharmottara's interpretation of tatra, de la in PVin III (PVint 133b3) against Sakyabuddhi (PVT 24a2 = karnakagomin, PVSVT 43,28) and Steinkellner 1971: 185 who understand in this case (of the complete cause) thereby rather creating a minor redundancy. 1 cr. Prauwallner 1958: 124 and 128 (translation) (reprint 1982: 263 and 267). ** de la sna ma dan Idan pa ni gan gi tshe ngyu ma tahan ba med pa mthon nas 'bras bu 'byun bar 'gyur ba Mid rogs pa ste / dper na sprin byun ba mthon nas cher bA 'byun bar 'yur ba nid Ita bu'o // (Frauwallner 1958: 124, reprint: 263). 11 The terms karyotpattisambhava and karyotpadanayogyata below refer to the same fact (cf. PVSVT 43,29). The first term is expressed, however, from the viewpoint of the effect, while the second is expressed from the viewpoint of the cause. The slight redundancy thus resulting for the two sentences must have been the reason why Dharmakirti drops the content of the main clause when he re-writes this passage In PVin III 310b8f.: de la (D : de P) ni rgyu tshogs pa nams kyi 'bras bu bskyed per (D: pa P) run ba nid 'ba' (P : 'ga' D) tlg rjes su dag la/ (D : par P) 1" tatra hi kevalam samagranam karananam karyotpadanayogyatanumiyatel I owe the ideas for this note to T. Punayama. 01 That his formulation was known to Dharmakirtl can be assumed from the fact that Jinendrabuddhi's post-Dharmakirtian commentary on Pramanasamuccaya and Its Vrtli is the main source of information on Vrsagana's epistemological thought (cf. Frauwaliner 1958: 8511.) 1 karanasamagryah ed. Malvania * karanan samagryab ed. Gnoll. There is however, neither an equivalent to karana- in the Tibetan translation, nor confirmation to be gained for the word from Sakyabuddhi's (PVT 24a6) or 12 Especially when we adduce the formulations of the Yukidipika which look like a derivative from Vrsagana's definition: tatra purvavad yada karanam abhyuditam drstva havisyattvam karyasya pratipadyate, tadyatha meghodaye bhavisyattvam vrste....... yadi tarhi karanasaktim sahakarisaktyantaranugrhitam apratiyoginim drstva karyasya vyakt pratipadyate, ....... tada purvavat. (YD 38, 101., 12., 14). 732 733

Page #13
The fragment of Vrsagana's text referred to above (cf. note 30) was quoted by Jinendrabuddhi to give the context as a whole which Dignaga referred to in parts (cf. Frauwallner 1958 861., reprint: 225f.). It is not totally impossible, therefore, that Jinendrabuddhi may have changed the relevant terms in accordance with Dharmakirti's procedure. However, I find this unlikely. I prefer to think that Dharmakirti used an extant formulation from an opposing tradition to Interpret it according to his own understanding. Otherwise, the whole question of how a certain well-known inference fitted into his system would be pointless. That the explanation of the Yuktidipika (YD 38,121.) may also be influenced by Dharmakirti could be assumed on the basis of N. Nakada's statement that the author of the Yd, knows svabhavahetu" ("Word and Inference in the Yuktidipika", Naritasan Bukkyo KenkyuJo Kiyo 12, 1989, 47-74147). However, the passage adduced as evidence by Nakada (YD 39,331.) does not reveal a knowledge of Dharmakirti's svabhavahetu, but merely refers to Dignaga's explanation of linguistic cognition as & kind of Inference. The objection naitad asty udaharanam anekantat. na hi meghodayo 'vasyam vrsteh karanam bhavati, vyvadinimitta pratibandhasambha - vat (YD 38.111.) left out in the quotations above, may also give the impression of being influenced by PVI 8=10 and PVSV 7,5-7. Again, this need not be the case, because the pratibandha argument is fairly general and certainly not Dharmakirti's original Invention. and as such aims at that goal Independently, of its own. In this sense It Implies a artly teleological explanation in contrast to a purely mechanical-causal one. However, it would seem to be possible to use the adjective "entelechial" as an attribute of causality to refer to this Buddhist idea, although here a goal (telos) of merfection is not taken into consideration, since it cannot influence as being inexistent. the original causal complex. The Buddhist idea of a causal process that roceeds out of its own natural constitution" (svarasavahin) therefore does not involve a conflict between a teleological and a causal model, but rather reserves an area within its purely causal explanation for the possibility of proceeding towards perfection out of necessity. without having to consider that goal itself as an essential part of an existent being. It is of some import in this context, that Buddhist ideas of continuous causal processes are not developed by means of metaphors and examples from areas of mechanical causality, but rather of organic causality, e.. In the tradition of the Salistambasutra (cf. E. Frauwallner. Die Philosophie des Buddhismus, Berlin 1956, 49ff, Steinkellner 1967: 13611.) I owe the proposal to use the attribute "entelechial" for this purpose to Monika Pemwleser. Masahiro Inami and Tom T.F. Tillemans. "Another Look at the Framework of the Pramanasiddhi Chapter of Pramanavarttika", WZKS 30, 1986, (123 - 142)125-127. 41 Ibid. 125 42 cr. ibid.: 126 and note 10. 39 Dharmakirti does not hesitate to use also this simple formulation of the Inference when the meaning in his own terms is sufficiently clear and undisputed, e.g. karapat karyanumana- (PVSV 105.11.) and karanat karyasamsiddhi (PV IV 269a). 49 Cf. ibid.: 127 (note 10). ** PV II 129-130. cf. Vetter's description of the causality of mental properties (cajttah) in Erkenntnisprobleme bei Dharmakirti, Wien 1964, 25-27 and 84-87. ** The fact that this particular form of the inference was Dharmakirti's way of understanding the inference "from cause to effect" is also supported by respective statements to be found in the works of some later opponents. cf. e.g.. In Jayantabhatta's Nyayamartlari (NM I 336,13-337.3), Bhasarvalnia's Nyayabhusana (NBhus 300,8-13). Abhayadevasuri's Tattvabodha vidhayini (TBV 562,34-563,4). Vadidevasuri's Syadvadaratnakara (SVR 586.13-25). 49 krpa svabijaprabhava svabijaprabhavalr na cet/ vipaksair badhyate citte prayaty atyantasatmatam // (PV II 129) Cf. Vetter, ibid., 26. 33 cr. Steinkellner 1971: 199. ** Cf. the sentences (H]-IL) and the text in note 25. 56 cr. PVin II 13.17-20, 15.131., and later in the Hetubindu (HB 8.221.: 9.40.). 47 Ch. Nyayakanika 146, 10r. (ed. Kasi); 553,2-4 (ed. Stern): tatha ca satyam eva tasyam (kast: etasyam Stern) samagryam apratibaddhayam (Stern: apratibandhayam Kasi) niya mena phriam utpadyata it. (For thesef editions cf. note 51.) 37 This consequence is expressed in PVin II,15,14-18 (cf. Steinkellner 1979: 59). Cf. also PVSVT 43,16f.: ... tavata natyantavasthapraptena, tatra lingigrahanat prag eva karyasya pratyaksatvam. 4. PVin II 13,20-22: 'bras bu skye ba dan mimthun pa bskyed pa ni gegs kyi mtshan Nid yin pa'l phyir ro // 3. cf. PVSV 7.5, where he speaks of 'cause-substances' (karanadravya) as being called "complete (samagra). Cr. also PVSVT 43,25: santanapeksaya tad ucyate, na ksanapeksaya. On the use of the expression "substance" (dravya) by Dharmakirti cf. Steinkellner 1979: note 156. 49 I believe that the reason for this lack of any cross-reference is the situation that an obvious coherence of Ideas in the conceptual world of the Buddhist Dharmakirti never made such a cross-reference necessary. Only the modern Interpreter, belonging to another world, searches for one in order to understand as an idea in and from its context. 1 I would like to propose the term 'entelechial causality to refer to those causal processes where a "complete causal complex" necessarily proceeds towards perfection because of the ascertained impossibility of impediments. The substantive 'entelechy" in Western philosophical tradition refers to the fact that every being is directed to a certain goal on account of its constitution. 50 Hayes/Gillon (cf. their paper mentioned in note 17) do not adduce the Pramanavarttika material but aim at an explanation in a fairly similar spirit. I think, when they find the consideration of the cardinal principle of Buddhism. 735 734

Page #14
namely, that the presence of unwholesome mental properties will certainly give rise to sorrow in the future" (in their note B.5) to be the motif for this theorem in Dharmakirti's logic. si For Devendrabuddhi cr. the passage translated by Inami/Tillemans, loc.cit., 126. 52 In vidhivivekah, srimadacarya-Mandanamisra-viracitah puyapadatrimadVacaspatimisra-nirnitaya Nyayakanikakhyaya vyakhyaya samalankrtah. Ed. Tailangarasmasastrimanavalli, Kasi 1903-1907: 146,9-13. In Stern's new critical edition Vidhivivekah of Mandanamisrah with commentary Nyayakanika or Vacaspatimisrah and Supercommentaries Jusadhvarkarani and Svaditankarani or Paramesvarah, Critical and annotated edition. The purvapaksah. By Elliot M. Stern. 4 vols., PhD dissertation, University of Pennsylvania, 1988 (available from University Microfilms, Ann Arbor): 553,1-5. 53 Peking edition No. 4531. The edition of the Sanskrit text by Kameshwar Nath Mishra, Sarnath, is forthcoming. I would like to express my gratitude for receiving a copy of this text in 1986.