SOME ASPECTS OF DHARMAKIRTl’S ONTOLOGY RECONSIDERED

by L. Zwilling

New York

It is now almost fifty years since Th. Stcherbatsky published his monumental work Buddhist Logicé, and thereby “opened the way” (srol. 'bysd. pa / as the Tibetans wo-uld say, for the present-day study of the system of ontology, epistemology and syllogistic reasoning first established by Dignaga and given its final shape by his grand¬disciple Dharmakirti. Since its publication, a significant number of texts have come to light which have of necessity widened the range of inquiry into this field. In addi¬tion to the Nyáyabindu2 and the Tattvasamgraha3 , which were the most important of all the extant Sanskrit treatises on these subjects known to Stcherbatsky, and which were utilized by him as the prime sources for Buddhist Logic, scholars today have access to the complete Sanskrit text of Dharmaklrti’s Pramánavárttikam, as well as to a number of commentaries upon it.4 In addition, we now possess an important sub¬commentary on the Nyáyabindu, i.e. Durvekamisra’s DharmottarapradipaS , and in the area of Bauddha-pramánaváda in general the monographs of Jñánasii& and Ratna- kirtP , the Tarkabhásá of Maksakaragupta^ , etc.

However, when we turn to the work of subsequent researchers who have had the opportunity to utilize this great mass of newly discovered material, we find that Stcherbatsky’s idiosyncratic line of interpretation has continued to so influence them that they have been unable to correct the crucial errors in it, even when confronted with textual evidence which points to radically different conclusions than those drawn in Buddhist Logic. That such has been the case is a great tribute to the excellence of Stcherbatsky’s pioneering work, and is a consequence of both his forceful mode of expression and the seductive character of his views; nevertheless, we hope to demonstrate here that Buddhist Logic is greatly flawed respecting some of the keypoints of Dignága and Dharmaklrti’s system, specifically as regards ontology. Firstly, we shall deal briefly with Stcherbatsky’s own philosophical predelictions and the effect which these had upon his interpretation of Bauddhanyáya, and secondly shall try to delineate more clearly Dharmaklrti’s ontology with refer¬ence to the Pramánavárttikam and some of the Indian and Tibetan commentaries on this text.

It is to be regretted that Stcherbatsky, despite his vast erudition, viewed Buddhist doctrine through the Idealistic spectacles of 19th century academic philo¬sophy; among the Western authors most often cited by him are Bradley, Bosanquet, Lotze and Sigwart» , the second-rate successors to Hegelian obscurantism. But it is to Kant and Hegel themselves that we must look for the basic models which Stcherbatsky

used in his explications of the Buddhist doctrinar systems. In viewing the Mahayana, especially the school of the Madhyamaka, Stcherbatsky imposed an Hegelian pattern of radical monism upon it; i.e., he held that the Madhyamaka school accepted an Absolute Whole as the final reality (sünyatá), based on their dialectical reduction of all compound phenomena to unreality. Although this view had great tenacity, it has finally been overthrown by the more recent interpreto:" of the Madhyamaka such as De Jong, Jacques May, Robinson, et al. On the other hand, as regards the Hinayana systems, Stcherbatsky

imposed upon them a radically pluralistic outlook formulated along Kantian lines, which assumes only parts to be real while the wholes are unreal, thereby confounding the quite different points of view of the Vaibhasika abhidharmists and the Sautrántika logicians. Moreover the svalaksana, which is the keystone of Dharmakirti’s ontology, is taken by Stcherbatsky as beings analogous to the duration¬less thing-in-itself (ding an sich) which is considered by Kant to be tre cause of our sensations.1 la As we shall see when we examine the concepts of svalaksana and samñnya- laksana, nothing couid be farther from Dharmakirti’s intentions as expressed in his own works and in those of his commentators. But it is Stcherbatsky’s distorted view of this ontology which has enjoyed uncritical acceptance up to the present time.llb

The concept of svalaksana is the foundation of Dharmakirti’s ontology, since it alone is accepted by him as being actually real (paramárthasat).12 For Stcherbatsky the svalaksana is a unique point-instant, without either extention in space or duration in time13 which is known only through sense perception (pratyaksa) ahd which is wholly uncognizable though the intellect.14 He assumes that there is a strict dicho¬tomy between the two sources of valid knowledge (pramana) accepted by this school, that is, sense perception has for its object (prameya) only the real svalaksana and ina ference (annmana) has for its object only the unreal mental image (samányalaksana).l& Although Stcherbatsky gives assent to the universolly accepted position of the Sautran-

ti'ka logicians as regards existence, viz. “real existence, ultimate existence is nothing but efficiency. Whatever is qausualiy efficient is real”18 , the Idealistic bias of his pers¬onal philosophy renders this seemingly realistic and empirical ontology unacceptable to him, and he insists that this system really establishes “behind the veil of empirical reality the existence of its transcendental source; the world of things as they are in- themselves.”!? Thus for Stcherbatsky reality is in fact a transcendental absolute, and the real object (svalaksana) is not the efficient entity presented to us in

everyday per¬ception but a see^ningly uncognizable thing-in-itself, stripped of all sensible qualities.18 To buttress this interpretation he even goes to the length of citing without attribution Sanskrit eqivalents for such Kantian terminology as “pure object” (suddhartha), “pure reason” (suddhakalpaniO etc.¡9 , although such terms are not to be found in any Buddhistic philosophical work, and are actually neologisms coined by Stcherbatsky himself. Dharftoakurti. /

In fact, such a precipitous gulf between the objects of sense perception and inference, in which the former is a transcendental point-instant and the latter a fictitious mental construct, was never intended by Dharmakirti. Such an acceptance would have led him into insuperable difficulties, since the objects of our ordinary conceptual cognition (savikalpikajwana) would in that case be entirely cut off from those of our sense perceptions, and our everyday ideas about the world could not then be even indirectly related to things as they actually are. While Dharmakirti does accept two sources of valid knowledge, he does so merely from the point of view of the causal efficiency or non-efficiency of their prime direct objects (grahyavisaya); the prime

direct object of sense perception is the efficient object, while that of inference is the non¬efficient mental image.20 This however does not imply that the two pramana-s are completely dichotomous in respect of their cognizables (prameya), since the object of sense perception can indirectly be a non-efficient, as for example when we perceive that there is no elephant in our sitting roomSi ; even more importantly, the object of a judgement or an

inference can indirectly, be a svalaksana, as for example when we make the judgement “this is a pot”, or when we infer the presence of a real fire on the other side of a hill on the basis of the presence of smoke.22 Thus while the directly apprehended objet of the intellect is indeed a non-efficient mental image, these types of inferential judgements are connected with the world of reals indirectly, since the judge object (adhyavasayavisaya)23 -pot or fire etc-of a conceptual cognition which follows in the wake of a sense perception is held by Dhamakirti to be none other than the svalaksana.24 Although Stcherbatsky does accept an indirect relation between percepts and concepts, their relation actually is impossible given the dichotomous character of his epistemology.

The source of much of Stcherbatsky’s confusion as to nature of the svalaksana lay in his failure to correctly understand the Sautrantika view on the relationship bet¬ween a whole and its parts; for him “The parts alone are real, the whole is a fiction.”25 This position is actually that of the Vaibhasika, and not that of Dharmalqrti of any other Sautrantika logician, but it is erroneousiy ascribed to them by Stcherbatsky. The Vaibhasika position as expounded in the Abhidharmakosa in the context of a discussion of the two truths, is that conventional truth (samvrtisatya) is that which disappears either upon physical disintegration or intellectual analysis, and ultimate truth (paramarthasatya) is that which is capable of withstanding such

disintegration-or analysis. Thus for this school an efficient object is certified by common experience to perform a function, such as a car which is able to take us where we want to go, but such an object does not exist ultimately, since it can be broken down into its compo¬nent parts, which can in turn be reduced to ultimate atoms of color, tangibility etc. There are however certain irreducible components of things such as color, taste, odor etc. which really exist, since they are still present when analysed to the atomic level,

and their cognition does not cease when they are considered apart from other quali¬ties which occasionally accompany them.26 In the mental sphere as well there are certain irreducible experiential states, such as feeling (vedar.a), conception (samjna) and will (cetana)iwhich accompany every cognition (the dasa mahabhumika-s)27 ? which can be reduced to a partless moment of consciousness and will not disappear when separately analysed.28 It is in the Vaibhasika system then that the partless moment of consciousness (ksana) and the partless atom (paramana) really exist (dravyasat), while all compounds are merely nominally existent (prajwaptisat).29

Moreover, Stcherbatsky’s claim that the partless moment and partless atom alone are held by Dharmakirti to he ultimately existent is weakened by the fact that in none of the passages which he cites from the works of Dharmakirti or his followers do we find partless moments or atoms defined as ultimately real- it is always the efficient object (vastu) which is so defined.3» On the basis of the available textual evidence further¬more, we must adopt a wholly differnet conception of the relationship between a whole and its parts in this system. For Dharmakirti, it is only a whole existing seperately from its

parts which is rejected, but the empirical whole which is made up of its parts is real, because what is real in Dharmakirti’s system is that which is certified by every¬day right knowledge to possess efficiency, i.e. the entire car or pot. Not that our ordi¬nary cognitive mode of apprehension of these reals is correct, because they are mis¬takenly apprehended by us as perduring in time, and as wholes which encompass all of their parts. But although it is not exactly the same object which perdures in space arid timei, neither is there a universe of entirely discrete atoms and\ moments such as is posited by Stcherbatsky, since there does exist a continuum (santana) of moments of an object which is in itself a svalaksana and which is certified as being such in our comihon experience.31 And just as Stcherbatsky has mistakenly concluded that gross " material objects are ficta on the grounds that only their

constituent parts are real, so he has asserted that the Buddhist logicians believed'time to be necessarily a fiction, on the grounds that only the “sensible point-instants”, i.e. the shortest moment of time, '•is real.32 However, ksana is not always to be taken in the sense of “an indivisible time particle”,33 Two types of ksana-s or moments have been recognized by Buddhist philosophers : I) the theoretical smallest time such as was employed in the

Indian science of astronomy, and 2) the moment of everyday usage based on the time it takes for the completion of some activity (by a. rd^ogs. kyi. skad. cig. дал), the origin of which is undoubtedly to be found in the biological processes of pulse and respiration.3* This type of moment is known to all, and is reflected in our everyday speech, as for example when we say “wait a moment”, or “it takes just a moment”, and it is considered by . Buddhist logicians to be equally as real as the multitude of infinitesimal time particles which go to make it up. We may conclude then that a gross material object or a time-continuum are on exactly the same footing; i.e. these wholes are as real as the

parts whichgotomakethemup,sin.ce the grassmaterial object or time-cbntinua aie held to be the effeet oftheir parts^ and forDharmaklrtijtisarule thai^Whaleveiik either a cause or effect is necessarily real.35 ¿j. vrf

That the whole is real for Dharmakirti is confirmed by those passages in the Pramânavârtikam which deal with the nature of the âlambana, i.e. the object which serves as the cause of its cognition.36 According to thesè passages/the atoms of color etc., which are by themselves imperceptible (atrndriya), when brought tó^éthér give rise to a collection (sawjita) which is in turn the cause' fó‘r the cognition of gross form.37 And since, as we have mentioned above, whatever is a cause or effect is

Since we have shown that the svalak§ana, the ultimate object in Dharmakirti’s system, is actually defined in terms of its causal efficiency, and that both temporal con¬tinua and extended objects are included within its scope, we are in a position to declare that Stcherbatsky’s description of ultimate reality (paramarthasat) in this system as an “indivisible, transcendental, mathematical zero”38 is extremely wide of the mark, since it assumes that Dharmakirti was engaged in the sort of ultimate analysis which is per¬formed in the Yogacara and Madhyamika systems. In these two schools there is a search for some type of final mode of being (math. thug. pai. yin. lugs.) of the objects of our experience, which by necessity will stand opposed to the unanalysed

appearances of ordinary perception. But it is Dharmakirti’s explicit intention merely to formulate an ontology and epistemology in consonance with everyday right knowledge.39 For this reason we choose to render paramárthasat in the context of this system as “actu¬ally real”, i.e. real in practice, rather than as “ultimately real”, since the latter term implies the sort of ultimate analysis which is not employed by Dharmakirti. What is meant here by “actually

real” is that which is connoted in ordinary language by the expression “it really works”, as applied for example to a functional automobile or medicine etc. This is not to say that an automobile ceases to be actually real when it no longer performs its intended function, for even when lying in ttxe wreckers yard it is capable of generating the judgement “That’s really a piece of junk !” On the other hand, samvrtisatin this system means a purely

nominal existence. This refers primarily to mental images, as for example the ideas of a car or pot in one’s head. While these images as moments of consciousness are reaUO they are unreal in the sense that the idea of a car is incapable of taking us where we want to go, and the idea of a pot is incapable of carrying water, and so on.<i

Crucial to Stcherbatsky’s premise that the real is an entirely unique monadicali absolute, “the unrelated thing”, “the mathematical point-instant”4?is the stress which he places upon those texts which describe the svalaksana as unique in the three realms (trailokyavyávrtta), i e. as existing on its own apart from all other things as a separate

existence “which is something quite unique”.43 As we shall see, the uniqueness of which Dharmakirti speaks leads to no such consequences. To be “unique in the three realms” is taken by Tibetan scholars as referring to “uniqueness of space, time, and nature”. What is meant by this spatial, temporal, and essential uniqueness is simply a restatement of the law of contradiction-if X exists at place A, it doesn’t exist at place B; if it exists at time C it doesn’t exist at time D, and if it is of the nature of Y it is not of the nature of Z.*4 Stcherbatsky assumes however that there is the element of

necessity involved in these propositions, that is if X exists at place A it necessarily con- not exist at place B and so on.*5 it is from this idea of a uniqueness which entails ¡necessity that Stcherbatsky derives his idea that only the partless moment of conscious¬ness and atom are svalak§ana-s, the corollary of which is that all duration and exten¬sion are logical fictions. This theory of uniqueness as involving an element of necessi¬ty was anticipated by the Gelukpa acaryas mKhas. Grub. rJe. and rGyal. Tshab, and is considered by them as nothing less than a form of logical overkill, for as they have shown, the consequences of such a view is that only one specimen at most of any real •could possibly exist.*®

While the Indian and Tibetan commentators do agree that a particular instance of an object is unique as to place etc., this does not rule out the existence of a real general samanya made up of various spatial, temporal and essential instances,*7 the possibility of which is denied by Stcherbatsky* for whom all generals are unreal per se. Just as we have noted in Dharmakirti’s acceptance of a real, efficient whole (see above), he similarly does not assert that the general or universal is necessarily a fiction; in fact what is rejected by him is a general which is eternal, separate substance (dravyan- tara) possessed in common by all those individuals to which it is related, such as acce¬pted in various guises by the Samkhya, Nyaya-Vaise§ika and Mimamsa schools.4»

JDharmakirti divides generals into two distinct catagories: a general which is a substan¬tive, and a'general which is’a logical construct .49 if every instance of a general is a real <i.e. efficient) thing, then the general is likewise real, as are book, pot, car etc. In each individual instance of a book the particular and the general are the same entity, and are only logically different, because the particular book and general “book” have no other referent than the svalaksaqa, which is existent in its own right and not a mere imputation by such intellect (rtogs. pas. brtags. tsam. ma. yin. par. rang. ngos.

nas. grub. pa.)H> In such instances then the general and particular do not exist apart from each other; e.g. the svalak§ana of fire is heat, and since heat is general in respect of every instance of fire all of these individual instances receive a common name on the basis of their common contrast with those things which lack heat.51 However, this does not mean that they possess a common heat- the heat of one fire is the specific essence of that fire alone (svam tattvam asadharanam)-52 Unlike the Brahmanical schools and philosophies of a similar tenor in the West (e.g. the Platonic), Buddhist logicians never posited the existence of a real universal “cowness” apart from specific cows, such as-

Diarma&ïrti / 309

Elsie, Bossie, Spots or Blots.

Only when every instance of a general is not a real is that general necessarily a logical construct. For example, object of knowledge (jneya) and existence (sattâ) fall into this class, because there are instances of an object of knowledge which are unreal,

i. e. non-functional entities (anarthakriyâsâmârthya) such as space (âkâsa). Moreover non-functional relational terms— “one”, “many”, “relation” etc. fall into this class, although functional relations do not.53 This relegation of all non-functional entities and relations to the status of ficta (rtog. pas. brtag. tsam)^ is in accordance with the tendency of rhis school to reduce the number of reals as compared to the more naively realistic Vaibhâsika, Nyâya-Vaiéesika, Sâmkhya and Mïmâmsa.

For mKhas. Grub. rJe, who follows the commentarial tradition of Devendra- buddhi and Sakyamati,55 what is rejected by Dharmakïrti is only the permanent, part¬less, omnipresent universal which is accepted by the Brahmanical darsana-s, but “in the works of Dignaga, the Seven Treatises (of Dharmakïrti) and their commentaries, there is not one word to the effect that a general is necessarily not an efficient, causal entity.”56 As a direct collaboration of the above statement we can look to Dharma- kîrti’s discussion of the three aspects of a valid logical mark (trairüpyalinga). In the classic example : “Sound is impermanent because it is a product”

what we are concerned with here is the ontological status of the probans “product”. The fact of being a product is general to every instance of an impermanent, and since <as has been shown above), when every instance of a general is a svalaksana the general is a svalak§ana, “product” is both a general and an efficient particular. This is explicitly set out in the Pramânavârttikam, in which it is stated that product is an efficient parti¬cular,57 and in which it also is confirmed that it is a general, on the grounds that in this sequence of thesis and reason “Sound is impermanent” etc.—it pervades the pro- bandum (impermanence).58 Of course “product” in its function here as the probans is a logical construction,69 but the general to which it refers is a real, since every instance of its occurance is real, and hence it can be legitimately utilised to prove a proposition about the* external world, namely that

sound is impermanent. If the general were necessarily a fiction, inference and syllogism would be able to give us no information at all concerning the world of reals, a state of affairs which would obviously be unac¬ceptable to the Buddhist logicians who take these as their fundamental tools. It is apparent that Stcherbatsky’s interpretations of Dharmakirti’s ontology, which have by now assumed the status of idées reçues in the field of

Buddhology, are for the most part without solid foundation, and spring from a tendentious use of the available material. I have tried to clear up some of the misconceptions regarding a few specific aspects of Dharma

kirti’s system, but there is much that remains to be dealt with, especially regarding his epistemology, specifically in relation to perception and 3 io / Kaiiash

theory of meaning, both of which have not yet received the detailed treatment they deserve. There is no question then that with the great amoünt of textual data present¬ly at our disposal, both by Indian and Tibetan Scholars, the time is propitious for a thorough réévaluation of the philosophy of Dignâga and Dharmakirti.

Abbreviations

Ab. K— Abhidharmakosa and Bhâsya of Acharya(sïc)V'asubandhu with Sphutartha Com-mentary of Ach^ryaYa^omitra^ 4 Volumes, ed. Swami Dwarikadass Shastri, Bauddha Bharati, Varanasi, 1970-1973.

BL — Buddhist Logic by Th. Stcherbatsky, 2 Volumes, Dover Publications, Inc., New York, n.d. CD — Tshad. Ma. sDe. bDun. Gyi. rGyan. Yid. Kyi. Mun. Sei. by mKhas. Grub. rJe^ mLhas. Grub. gSung. ‘Bum Vol. Tha, edition ? n.d. (Personal collection of the author.)

NB — Nysyabindu of Dharmakirti contained in the Dharmottarapradipa of Pandita Durveka Misra, ed. Pandita Dalsukhbhai Malvania, Kashi Prasad Jayaswal Research Institute Patna, 1971.

PV — Pr&manav&rttikam of Dharmakirti. References to Chapter One are to TAr Pram&xpfarttikam of Dharmakirti, The First Chapter with the Autocommentary ed. Raniero Gnoli, Serie Orientale Roma XXIII, Roma, 1960. (Referred to in the notes as G).References to the remaining chapters are to Pram&riavarttika of Achâry a Dharmakirti (sic) with the Vritti of Achâry a (sic) Manorathanandin, ed. Swami Dwârikâdâss Shâstri, Bauddha Bharati, Varanasi, 1968.

Notes

(In the following notes Roman numerals refer to chapter number and Arabic numbers to sloka or sutra. Pada-s are indicated by small case letters. In the case of Ab. K and BL Roman numerals refer to volume number. In referring to PV the tradi¬tional ordering of the chapters is followed, i.e. I-Svârfhânumâna, II-Pramânasiddhi, Ïll-Pratyaksa, IV-Parârthânumâna.)

1. Stcherbatsky, Th. Buddhist Logic, Vols. I, II, Leningrad Academy of Sciences of the U.S.S.R., 1930

2. Nyayabiudu of Dharmakirti with Dharmottard* s fïkâ, ed. Th. Stcherbatsky, Lenin¬grad, Bibliotheca Buddhica VII, 1918

3. Tattvasangraha of Sântarakfta with Kamalasila’s Panjika, 2 Volumes, Baroda, Gaekwad Oriental Series, 1926

4. Pramânavârttikam : (A) Pramânavârttikasvavxtti, The Autocommentary on the First Chapter Only, Together with the Subcommentary Svavratifika by Karnakagomin, ed.. Râhula Sânkrtyâyana, Kitab Mahal, Allahabad, 1943; (B) G; (C) Pramwaav&rtti- k^lamkaraof Prajnäkaragupta, deciphered, and edited by Rähula Sänkrtyäyana, Jayaswal Reseorch Institute, Patna, 1933 ; (D) Pramänavärttikam of Dharmakirti with Vrtti by Achärya Manorathananandi, ed. Rähula Sänkrtyäyana, Appendix to J.B.O R.S , vol. XXIV/1938

5. NB

6. Jnänasrimitra-Niban.dhä.vall of Jnänasrimitra, ed. Anantlal Thakur, Kashi Prasad Jayaswal Research Institute, Patna, 1959
7. Ralnakirti-Nibandhävali of Ratnakirti, ed. Anantlal Thakur, Kashi Prasad Jayaswal Research Institute, Patna, 1957
8. T.arkabh&so. of Moksakaragupta, ed. H.R. Rangaswami Iyengar, Mysore, 1952
9. BL. I. 549, 550

10. Stcherbatsky, Th. Conception of Buddhist Nirvana, Bharatiya Vidya Prakashan, Varanasi, n.d. p. 68
Russell, Bertrand, A History of Western Philosophy, Simon & Schuster, New York, 1964, p. 731

11a. Russell, op. cit., p. 707

lib. Ganguli, Hemanta Knmar, Philosophy of Logical Construction, Sanskrit Pustak Bhandar, Calcutta, 1963, p. 4
Shastri, D.N., Critique of Indian Realism, Agra University, Agra, 1964, pp. 2, 3, 5, 187, 188 et passim
Potter, Karl H., Presuppositions of India’s Philosophies, Prentice-Hall of India (Private) Ltd., New Delhi, 1965, pp. 189, 191
Sharma, Dhirendra, 'Rhe Differentiation Theory of Meaning In Indian Logic, Mouton, The Hague, 1969, pp. 26, 37, 42, 67. 125
Matilal, B.K., Epistemology, Logic, and Grammar in Indian Philosophicol Analysis, Mouton, The Hague, 1971, p. 145
Dravid, Raja Ram, The Problem ofUuiversals in Indian Philosophy, Motilal Banarsi- dass, Delhi, 1972, p. 73

12. NB. I. 13-14: BL. II. pp. 35-37
13. BL. I. pp. 84, 87
14. BL, I. pp. 78, 184
15. BL I. p. 73
16. NB. I. 15; BL. I. p. 69; BL. II. p. 36
17. BL. I. p. 63
18. BL. I. pp. 70, 190 r . ... ..
19. BL. i. pp. 61,78 • ’/*• l. 1 ’
20. PV. III. lab; PT. III. $4

2L JSUT II. 12;'Ny&yabindutiksL of Vinitadeva, Sanskrit original reconstructed fi om the extant Tibetan version, with English translation and annotations by Mrnalkanti 'Gangopadhyaya, Indian Studies Past and Present, Calcutta, 1971, p. 131 n 22
312 I Kailash
22. NB. II. 17


23. According to the Tibetan Scholastic tradition, Dharmakirti distinguishes four types of objects; A-the apparent object (pratibhasavi§aya/snang. yul); B-the prima facie direct object (grahyavisaya/gzung. yul.) C-the judged object (adhyavavayavi§aya/ zhen.yul.); and D-the object of purposeful striving (pravrttivisaya/’jug. yul.) For a detailed discussion of the above see DD ff. 52b2-56b3. See as well the Tshad. Ma. Rigs. gTer. Gyi. dKa’. Bai. gN«. rNam. Par. bShad. Pa. sDe. bDuti. Rab. gSal. by Go. Ram. bSod. Nams. Senge in the Sa. sKya. Pai. bKa. ‘Bum , Toyo Buno, 1969, Vol. 12, p. 6 plate 4 line 5, to p. 8 plate 3 line 2

24. See G p. 39 lines 4-8. For a lucid explanation of this important point see the Grub. mPhai. rNam. Par. bZhag. Pa. gSal. Bar. bShad. Pa. Thub. bsTan. Lhun. Poi mD^es. iGyan. by ICang. sKya. Hu. Thog. Thu. Ye. Shes. bsTan. Pai. sGron. Me, printed and published by Lama Guru Deva, Sarnath, 1970, pp. 99-100

25. BL. I. p. 86
26. Ab. K. VI. 4
27. Ab. K. II. 24
28. See note 26
29. Ab. K. II. 46
30. BL. I. p. 69 n. 1; BL. II. p. 23 n. 2

NyayisrW»i7fltI^auji>/>4nl, ed. Tn. Stcherbatsky, Bibliotheca Buddhica 11, St. Peters- bourg, 1909, p. 11 lines 14-16
32. BL. I. p. 84

33. BL. I. p. 106
34. See note 29. Also the mKhas. Pai. Tshul. La. ’Jug.Pai. sGo. by Mi. Pham., Tashijong H.P., n.d., ff31a5-32bl for a good explanation of this point and of time in general.
35. PV. I. 172ab

36. Ab. K. II. 62c
37. PV. II. 86-92; PV. III. 194-196; PV. III. 321. Also the I shad. Ma. rNam. ’Grel. Gyi. gZhung. bShad. sNang. Pai. gTer. by Mi. Pham , Dehra Dun, UP, n.d., pp. 519ff
38. BL. I. pp. 182-183
39. NB. I. 1
40. PV. III. 9-10

41. NB. I. 15-16; PV. III. 3
42. BL. I. 104
43. BL. I. 103, 104, 402
44. DD ff.32b3-33a3; also Tshad. Ma. rNam. ’Grel. Gyi. Tshig. Leur. Byas. Pai. rNam. bShad. Tar. Lam. Phy in. Ci. Ma. Log. Pa. gSal. Bar. Byed. Pd. by rGyal. Tshab. rGyal. Tshab. gSung. Bum. Vol. Cha, ed.? n.d. ff. 44b2-45a4 (Personal collection of the author.) '
This point is discussed by Tibetan scholors in terms of “non mixing of place, time
Dharmaklrti /313

and nature.” (yul. dus. rang, bzhin. ma. ’dres. pa.)
Go. Ram. bSod. Mams. Senge, op. cit., refers to the following passages in PV. as the foundation for the accepted Tibetan interpretation;
PV. I. 153a4-c indirectly show yul. ma. ’dres. pa;

PV. I. 87ab & PV. I. 139ab directly show rang, bzhin. ma. ’dres. pa;
PV. I. 92b & PV. III. 487 directly show dus. ma. ’dres. pa. and indirectly dus. ’dres. pa;
PV. I. 68 & PV. I. 139cd directly show yul. and rang, bzyin. ’dres. pa;
PV. I. 98 directly shows dus. ’dres. pa;

PV. I. 136c directly shows yul. dus, rang, bzhin. ’dres. pa.
45. BL. I. pp. 86-87
46. DD ff23b6-24a3; f. 30b 1-2
47. DD f. 33a3-5
48. Pramavtavarttikavi'tti by ^akyamati, bsTan. ’Gyur., Peking ed, vol. Je, f. 226b2
49. PV. III. 5led. Although this text distinguishes three types of samanya, they are reducible to two without doing violence to Dharmakirti’s intention here.
50. This is tke standard definition of svalaksana accepted by the Gelugpa school. See DD f. 21b2; rGyal. Tshab. op. cit. f. 45a6
51. I allude here to the differentiation theory of meaning (apoha) which was first formu-lated by Dignaga in the 5th chapter of the Pramanasamuccajya and given its final form by Dharmakirti. It is discussed by him throughout the PV. but primarily in Chapter One verses 40-185 For a useful survey see BL. I. pp. 457-482

52. NB. I. 12 and the commentary by Dharmottara thereupon,
53. DD f. 31a6-31b6
54. DD f. 33b5-6

55. Tshad. Ma. rNam. ’Grel. Sogs. ¿Tan. Tshigs. R/g. Pa. Las. Byung. Bai. Ming. Gi. Grange. by KLong. rDol, contained in The Collected Works of Longdol Lama Parts 1, 2, reproduced by Lokesh Chandr from the Collections of Prof. Raghu Vira, International Academy of Indian Culture, New Delhi, Dec. 1973, plate 661 line 5 to plate 662 line 3. In this context see also footnote 48.
56. DDf. 33a5

57. PV. I. 172ab
58. PV. II. 16
59. DDf. 39al-6