Two Circles Or Parallel Lines

Author(s): Eli Franco
Publisher: Eli Franco
Catalog link: https://jainqq.org/explore/269437/1

Page #1

TWO CIRCLES OR PARALLEL LINES?
by
Eli Franco, Hamburg
The opinion that Dharmakirti's reflections on the authority of the Buddha in the Pramanavärttika involve a logical loop, a vicious circle, or even a benign circle, has been asserted several times by renowned Dharmakirti scholars such as Tilmann Vetter, Masatoshi Nagatomi, Ernst Steinkellner and Richard Hayes. According to the first three scholars named the circle consists in the fact that perception and inference legitimate the authority of the Buddha, and that the authority of the Buddha, in its turn, legitimates the validity of perception and inference. This circularity was asserted for the first time, I believe, in Vetter's pioneering study Erkenntnisprobleme bei Dharmakirti. Thus, Vetter says: "Er gilt als Erkenntnismittel, weil seine Autorität die Maßgeblichkeit von Wahrnehmung und Schlußfolgerung legitimiert. Denn der Buddha gibt Ziel und Anweisungen des Handelns, welche unsere Wahrnehmung und Schlußfolgerung nicht geben könnten, und auf welche nur eine flache Aufklärung verzichten würde. Daß er hierfür Autorität ist, muß erst bewiesen werden. Denn bloß auf Treu und Glauben ist man nicht Buddhist. Das ergibt einen historisch-sachlichen Zirkel, dem Dharmakīrti mit Recht nicht aus dem Weg gegangen
ist."
Steinkellner translates this passage as follows: "He (the Buddha) is considered as a means of valid cognition, because his authority legitimates the authenticity of perception and infer
As usual I am indebted to my best reader, Priv.-Doz. Dr. K. Preisendanz, for reading this paper with her usual thoroughness and improving it with her thoughtful comments.
Cf. T. Vetter, Erkenntnisprobleme bei Dharmakirti (Wien, 1964), p. 27.
2 CC. E. Steinkellner, "The Spiritual Place of the Epistemological Tradition in Buddhism." Nanto Bukky 49 (1982): 13.
Steinkellner claims to give only the gist of Vetter's explanation, but this does not seem to be the case. Even if one disregards the question whether Vetter's Massgeblichkeit" should be rendered as "authenticity" rather than "validity," they do not seem to describe the same circle. According to Steinkellner ("Spiritual Place," p. 11) the circle consists of three elements: pranana, avisamvadana and the Buddha's authority, which has to be justified in its turn by pramana: "The structural scheme of these ideas of Dharmakirti turns out to be a true circle: The decisive defining characteristic of a means of valid cognition (pramāna) is the demand that it must stand the test of meaningful (my emphasis) practice (avisamvadana), and this decisive characteristic?) connects it (pramana?) with the Buddha as the one on whose authority one knows what meaningful practice is. The reciprocity then is brought about by the need to prove this authority of the Buddha."
More important, however, is the difference between Vetter's and Steinkellner's ideas concerning the extent of the Buddha's legitimation of perception and inference. According to Vetter, op. cit., pp. 27, 32, the Buddha legitimates only those perceptions and inferences that operate on areas that are beyond the normal realm of perception and inference; cf. the reason above ("For the Buddha gives a goal and guidance for action, which perception and inference cannot give"), as well as p. 32, referring to PV Pramānasiddhi lab: "Erkenntnismittel ist Wissen, das sich in (wir können ergänzen: auf die Erlösung gerichtetem)(my emphasis) Handeln bewährt." According to Steinkellner.

Page #2

64
E. Franco
ence. For the Buddha gives a goal and guidance for action, which perception and inference cannot give, and which would be disposed of only by a superficial rationality (flache Aufklärung, lit.: shallow enlightenment). That he is an authority on this has yet to be proven. For nobody is a Buddhist simply on trust. This results in a historical-factual circle, which Dharmakirti has rightly not avoided."
Even though Nagatomi also discerns circularity in Dharmakirti's reasoning, his position is different from Vetter's. Unlike Vetter who maintains that "nobody is a Buddhist simply on trust" Nagatomi perceives the faith in the words of the Buddha to be decisive: "We must note, however, that the final authority by which they [viz. Dignaga and Dharmakirti] claim the validity of their pramāṇa system was none other than the Buddha's words which they accepted as authentic by faith. Thus, the Buddhist pramāna system and the authenticity of the Buddha's words stood, in reality, in a reciprocal relation: the structuring of the former was done within the limits of the latter, and the latter was meant to be supported by the former."4
What I find surprising about these statements is that their authors have not taken the trouble to point out to their readers the textual basis on which their interpretations rest. It is as though we were reading a crime story in which the scene of the crime is never mentioned. And one can well understand the puzzlement of Tom Tillemans when he says: "Frankly, I must confess that I fundamentally do not understand what is meant when Prof. Vetter says that '[the Buddha's] authority legitimates the authenticity of perception and inference'." Tillemans then suggests that (p. 24) "Steinkellner, in his 1983 article [i.e., "Tshad ma'i skyes bu: The meaning and historical significance of the term"], may have given us some clues as to what is being meant. In his article the Buddha was characterized as 'the final source and judge of any validity and usefulness in any kind of cognition' (1983, p. 276)."
In my opinion, however, this is not what Vetter (or Nagatomi) had in mind when claiming that the Buddha is the ultimate source for the validity of the other pramānas. I am not, of course, able to read the minds of Vetter or Nagatomi. Nevertheless, being relatively well acquainted with the Pramāṇasiddhi chapter to which both scholars refer, I think that there is only one passage they could have had in mind. The crucial statement occurs in verse 283a-c according to Vetter's enumeration" [= PV 283cd-284]:
on the other hand, the Buddha legitimates all valid cognitions, not only those that are soteriologically relevant. Thus, Steinkellner and Vetter describe two different circles. Furthermore, in the discussion that followed the presentation of this paper Steinkellner kindly informed us that the circle he described is not based on the interpretation of any specific verse, but is due to meta-textual reflections. In this respect too the two circles differ. Nevertheless, I will not discuss Steinkellner's circle here because I do not know what could substantiate or refute such a hypothesis. As far as I can see, Steinkellner does not adduce any evidence to support his hypothesis; lack of textual evidence, on the other hand, would presumably not refute such meta-textual considerations. I fail to see why, according to Dharmakirti, e.g., a cognition of water by means of perception or of fire by means of inference derives its validity from the Buddha's authority. Furthermore, how would one ever be able to ascertain whether such cognitions are true or false?
* Cf. M. Nagatomi, “Mānasa-pratyakṣa: A Conundrum in the Buddhist pramāna System." in Sanskrit and Indian Studies: Essays in Honour of Daniel H. H. Ingalls, ed. M. Nagatomi et al. (DordrechuBoston/London, 1980). p. 246.
Cf. T. J. F. Tillemans, Persons of Authority. Tibetan and Indo-Tibetan Studies 5 (Stuttgart, 1993), p. 22. Cf. T. Vetter, Der Buddha und seine Lehre in Dharmakirtis Pramāṇavārttika, WSTB 12 (Wien, 19902).

Page #3

Two Circles or Parallel Lines?
upadeśatathābhāvastutis iadupadeśatah / pramanatattvasiddhyartham... // "The praise fof the Buddha) by means of the teaching being so (i.e., being valid or non-deceptive) (is uttered by Dignāga) in order to establish the true characteristics (Vetter: Wahrheit) of the means of knowledge from his teachings."
As is often the case with Dharmakīrti's statements, this statement is obviously open to several interpretations. Indeed, its interpretations by Vetter and Nagatomi can be accommodated within the wording. Yet, at least three questions should have been raised by them before arriving at their conclusion that the verse implies circularity or reciprocity. First, does this interpretation fit within the immediate context of the verse? Second, was the verse interpreted in this manner by any of Dharmakirti's commentators, and if not, why is the traditional interpretation to be rejected? Third, does Dharmakirti actually attempt to establish the validity of perception and inference by relying on the authority of the Buddha?
As a matter of fact, the immediate context of the verse does not support Vetter's and Nagatomi's interpretation because Dharmakirti argues here and in the following two verses that the Buddha used perception and inference, not that they are valid because of him. Nor does the commentators' interpretation support Vetter's or Nagatomi's hypotheses, for none of Dharmakīrti's commentators detects here the circularity perceived by Vetter or the reciprocity claimed by Nagatomi.
Vetter's interpretation of the verse (Erkenntnisprobleme, p. 33) is bewildering and differs radically from that of all the commentators. All the commentators interpret the praise as referring to the praise of the Buddha in Dignāga's mangalasloka. Vetter, however, claims that the praise is directed at the teaching and is constituted by the section on the four noble truths in the Pramänasiddhi chapter: "Das Sosein der Lehre wurde deswegen gepriesen das heißt: die ganze Beschreibung der buddhistischen Erlösungslehre wurde deshalb unternommen , um aus seiner Lehre die Wahrheit über die Erkenntnismittel zu erweisen." Vetter is perhaps led to this interpretation because he interprets the compound tathābhāvastuti as a genitive (rather than instrumental) tatpurusa. Consequently, the fact that Dignāga praises the Buddha and Dharmakirti praises his teachings may have led Vetter to dissociate the two statements. In my opinion, however, Dharmakīrti's shift from considering the teacher as pramāna to considering his teachings as pramāna is trivial and therefore needs no further justification; this shift is obviously conditioned by the present context, because the characteristics of pramāna are not derived from the person, but from his teachings. Even if the traditional interpretation by the commentators is slightly problematic. Vetter's own interpretation, inasmuch as it lacks any justification or substantiation, remains unconvincing. In Der Buddha und seine Lehre in Dharmakirtis Pramānavürttika (p. 172), however, Vetter has tacitly changed his interpretation: "[Dignāga) preist (im Einleitungsvers zum Pramāṇasamuccaya) (den Erhabenen) mittels des Soseins (d.h. Erkenntnismittelseins) seines Lehrens, ..." The translation is not accompanied by any comments, but according to the introduction (p. 16) Vetter does not seem to have changed his mind about the circularity in the Pramāṇasiddhi chapter because he speaks of "Beweiszirkel" and maintains that one can rely on perception and inference because the authority (i.e., the Buddha) used them.

Page #4

66
E. Franco
The commentatorial explanation of this and the following verses' is simple, and it agrees with a natural reading of the verse which appears immediately after the conclusion that the Buddha is a means of knowledge, and which forms, so to say, the conclusion of the whole chapter. After the Buddha has been proven to be a means of knowledge, the question arises as to why Dignāga chose to praise the Buddha precisely with the epithet pramanabhūta because there are many epithets to choose from. The answer indicates that we have to understand this crucial verse within the context of the self-understanding of the pramana school within the Buddhist tradition, that is, as clarifying the school's relation to the teachings of the Buddha. Dignäga praises the Buddha as a means of knowledge, which amounts to praising him because his teachings are valid. If his teachings are valid, we can establish from them the true characteristics of the mcans of knowledge because perception and inference are used in the Buddha's teachings, as shown in the last two verses. The framework in which these statements are made is clearly apologetic, as Dharmakīrti aims at anchoring his own epistemology in the original message of the Buddha. Thus, according to Dharmakirti, Dignāga praises the Buddha for being a teacher not only in religious matters, but also in matters of epistemology. These apologetics, however, are to be distinguished from the actual procedure by which Dharmakirti reaches the conclusion that the Buddha is a means of knowledge. Even if the usage of perception and inference has been observed in the Buddha's teachings and serves as a model for establishing their true characteristics, one can test and independently use them to establish, among other things, the validity of the Buddha's teachings, and from it the authority of the Buddha himself. Perception and inference within this context are certainly not established from the outset just because of the Buddha's authority. Nevertheless, one is justified to claim later on, in an apologetic context, that the means which have proved so successful were already discovered by the Buddha and taught in his teachings.
Moreover, even before Dharmakīrti's time Dignāga's logic was attacked for being a heterodox or "external" science that has its source in the Nyayasūtra of Aksapāda. We know of several attempts to somehow anchor this new science in the Buddha's word in order to provide it with legitimacy. Uwe Frankenhauser reports one such attempt (at the beginning of Kuiji's Yin ming ru zheng li lun shu) which bears a remarkable similarity to Dharmakirti's endeavour: even though there is no sūtra that deals directly with yin ming, the method of argumentation was used by the Buddha. An interesting attempt to legitimate Dignāga's logic without depriving it of its heterodox origins is found in Hötan's Zuigenki 1/26–3a: The yin ming lun existed since beginningless time. The Rşi Aksapāda proclaimed it at the beginning of the kalpa. Later on, when the Buddha appeared in the world, he considered it as the basis of his transformation.
The answer to the third question has already been given by Tillemans: Dharmakirti never attempts to establish the validity of perception and inference by appealing to the authority of the
The four direct commentaries on this verse are quoted below. * CF. U. Frankenhauser, Die Einführung der buddhistischen Logik in China. Wiesbaden 1996. pp. 137-138. Cr. the translation in Frankenhauser, op.cit., p. 139.

Page #5

Two Circles or Parallel Lines
67
Buddha, neither in the Pramanavārtika, nor in any of his other works. Onc has to conclude, therefore, that the assertions by Vetter and Nagatomi are based on nothing but the doubtful interpretation of a single verse.
*
*
*
Another kind of circularity in Dharmakirti's exposition in the Pramānasiddhi chapter was suggested by Richard Hayes in his paper "The Question of Doctrinalism in the Buddhist Epistemologists." The paper examines the stance of the Buddhist epistemologists Vasubandhu, Dignāga, Dharmakirti and sāntaraksita on agama, namely, whether they take the body of canonical writings as a source of knowledge and whether the Buddhist logicians should be characterized "as champions of reason or rather as champions of dogma" (p. 646). In other words, Hayes raises the question whether in developing their principles the Buddhist epistemologists were decisively influenced by their partisanship to canonical doctrines.
According to Hayes (p. 661), Dharmakirti must address the following two questions in order to establish that the Buddha was a credible person: 1) How do we know that the Buddha was in possession of the truth? 2) How do we know that the Buddha did not lie to us? Hayes claims that "[i]n dealing with the first of these questions, Dharmakirti falls into the familiar trap of circularity. We can be certain that the Buddha was in full possession of the truth, says Dharmakirti, because it was the Buddha who taught us the four noble truths and the eightfold path to salvation. In other words, we know that the Buddha is trustworthy because it was he who taught us all the things that we Buddhists believe. The vicious circularity of Dharmakirti's argument is disguised somewhat by the fact that it takes him some two hundred couplets to make the circle, but it is undeniable that he does eventually end up in a logical loop on this first question."
Now, strictly speaking, there is no circularity in what Hayes describes. However, if I understand what he meant to say rather than what he actually said, then he claims that, according to Dharmakirti, the Buddha is considered to be trustworthy because he taught us the four noble truths, etc., which we consider to be true; and we consider the four noble truths, etc., to be true because the Buddha, who is trustworthy, taught them to us. However, just like Vetter and Nagatomi, Hayes fails to provide a single reference as to where Dharmakirti says that "we" believe in the four noble truths, etc., just because the Buddha taught them to "us." And I submit that no such passage can be found in Dharmakārti's writings. However, contrary to the circularity claimed by Vetter and Nagatomi, the circularity alleged by Hayes is not due to the misinterpretation of a single verse or even a group of verses, but to a misunderstanding of the structure and proof strategy of the Pramānasiddhi chapter.
This misunderstanding is also seen in Hayes's answer to the second question, namely, how we know that the Buddha did not lie to us. Here again Hayes asserts a vicious circle: "...for Dharmakirti begins by trying to show that the Buddhist doctrine is justifiable, because it was taught by a trustworthy authority. But in showing that the Buddha was trustworthy, Dharmakirti
1° C. Tillemans, op. cit., p. 22. "Cf. Journal of the American Academy of Religion LIV4 (1984): 645-670.

Page #6

68
E. Franco
ends up appealing to information about the Buddha's career through several rebirths. But the only source of information about the Buddha's career is the very body of scriptures the authority of which Dharmakirti is trying to defend" (p. 662).
Now, it is wrong or at least highly misleading to say that Dharmakirti appeals to information about the Buddha's career through several rebirths. Hayes is surely referring to the Buddha's compassion, but the Buddha's compassion is inferred (indirectly) from his teaching of the four noble truths, and not known from any canonical information about his past lives. Thus, as far as I can see, the circularities observed by Hayes have no foundation in Dharmakirti's text. In a way it can be said that Dharmakirti's reasoning moves from the teaching to the teacher and back to the teaching, but this does not involve a vicious circle. In fact, it does not involve a circle at all, because Dharmakīrti does not understand the Buddha's teaching to be a single whole. For him the teaching has a main or essential part (pradhāna) and a secondary or subordinate part. The main part, which consists of the four noble truths, is established independently of the authority of the Buddha by means of perception and inference. However, the Buddha's teaching has also a secondary part that cannot be established by means of perception and inference. The truthfulness of this part is inferred indirectly from the exceptional qualities of the Buddha. Thus, Dharmakirti demonstrates a five-step inference from the Buddha's essential teachings to his non-essential teachings via his personal qualities: First, the essential part of the Buddha's teaching, namely, the four noble truths, is established independently of the Buddha's authority through perception and inference. Second, from this established main part of the teaching the Buddha's knowledge is inferred. Third, from this knowledge the Buddha's compassion is inferred. Fourth, from his knowledge and compassion taken together his authority and trustworthiness are inferred.12 Fifth, the truth of the non-essential part of the Buddha's teaching is inferred from his trustworthiness. Only if one thinks of the Buddha's teaching as a homogeneous corpus and does not distinguish between the essential and non-essential parts of his teaching can one conclude, as Hayes does, that Dharmakīrti ends up in a loop from the teaching to the teacher and back to the teaching.
Furthermore, it is not the case that to use Hayes's words-it takes Dharmakirti some two hundred couplets to make this circle. In the Pramānasiddhi chapter Dharmakirti concludes his line of argumentation with the establishment of the Buddha's authority. He does not proceed, at least not explicitly, to the final step, namely, the inference from the reliability of the teacher to the truthfulness of his teaching. I suppose that this step is not taken for two reasons: first, because it is not thematic in the context of Dignāga's mangalasloka which deals only with the Buddha's properties, and second because it is trivial—if the Buddha is a pramāna, then it goes without
12 The correspondence of the different elements of this reasoning to the Buddha's epithels in Dignāga's mangalaśloka should be obvious. Dharmakirti interprets tāyitva as "revealing the four noble truths,'t sugatava as referring to the Buddha's knowledge and jagaddhitaisitā to compassion (karuna). The epithet śästriva has no correspondence in the above reasoning because, according to my understanding of Dharmakirti, it plays no direct role in the proof of the Buddha's authority (pramanabhūtarva); cf. Franco, Dharmakirti on Compassion and Re. birth, WSTB 38 (Wien, 1997), p. 22.
In fact, Dharmakirti is willing to reduce the essential part of the Buddha's teaching even further into the fourth truth alone: cf. v. 145a: täyah swidrstamärgoktih. I hope to return to Dharmakirti's reductionism on a different occasion.

Page #7

Two Circles or Parallel Lines?
saying that his teachings are true.
However, the fifth step of Dharmakirti's line of reasoning is made explicit by him in the Svārthānumāna chapter v. 217:13
heyopádeyatattvasya sopayasya prasiddhitah/ pradhanarthavisamvadad anumanam paratra va II "Or because (the correctness of what is to be avoided and what is to be appropriated, as well as (of the means (thereof), is well established, the main content (of the Buddha's teaching) does not belie (its promise). Therefore, it is justified) to infer (also in respect to other (parts of the Buddha's teaching that they do not belie their promise.)"
This verse and its interpretation by the Tibetan writers mKhas grub rje and Ngag dbang bstan dar were discussed by Tillemans, the only scholar so far to have argued strongly against the assumption of circularity in Dharmakīrti's writings. Although I want to acknowledge here my indebtedness to his important contribution, I would like to suggest a somewhat different interpretation of the purport of this verse. According to Tillemans, Dharmakirti's method of transferring the speaker's credibility from one area to another is not significantly different from the usual Nyāya-Vaiseșika view, as we find it for example in Nyāyasūtra 2.1.69 and Vātsyāyana's Nyāyabhāsya thereon, where the exactitude of someone's statements on matters of medicine, incantations and such things is said to serve as grounds for inferring his trustworthiness in other areas" (Persons of Authority, p. 16).
This view, I believe, misses two crucial elements in Dharmakīrti's argument. Before pointing out these two elements, let me emphasize that the argument in this form is not valid: Just because someone is trustworthy in matter x he is not necessarily trustworthy in matter y. Indeed, Tillemans makes it clear that both mKhas grub rje and Ngag dbang bstan dar considered such a statement to be false (op. cit., pp. 16-17). And, indeed, Dharmakirti's argument differs on two points. First, Dharmakirti does not simply draw an inference from trustworthiness in any part x to trustworthiness in any part y; he allows such an inference only when one moves from the main part of a teaching to its secondary part. Second, when we look at Dharmakīrti's own commentary on this verse it becomes obvious that the logical reason used in the inference is not—as in the Nyäyabhäsya—the sameness of the author, Dharmakīrti includes the motivations of the speaker in his reasoning, and this is what allows the inference of trustworthiness in one area from that in another:
"What has to be avoided, what has to be appropriated and the means (thereof) that are taught by that (trustworthy person)" do not belie (their promise), i.e. are not contrary to reality): just as the four noble truths in the manner which will be stated in the Pramănasiddhi chapter]: because this very same (main content of what was taught by the Buddha, namely, the four noble truths), which serves the (ultimate) purpose of man (namely, nirvana, and thus) is capable of application (i.e.,
13CL, PVSV 109. This verse is translated twice in V. A. van Bijlert, Epistemology and Spiritual Authority, WSTB 20 (Wien, 1989), pp. 124, 142. The entire discussion of agama was translated by H. Yaita; cf. next note.
14 The discussion on agama in PVSV was translated by H. Yaita in "Dharmakirti on the Authority of the Buddhist Scriptures (agama)—an annotated translation of the Pramanavärttika-svavrtti ad v. 213-217." Nanto Bukkyo

Page #8

E. Franco
repeated practice), does not belie sits promise), it is apprehended (that he or his teaching) is so (i.e., non-belying) with regard to another object too; she does) not (say anything) in order to deceive, because (his statements are not obstructed by any other pramāna) and because it (would) be futile for a speaker to make) false statements without (any) motivation."
It is clear, I believe, that Dharmakirti's logical reason for the reliability of the secondary part of the teaching would not be "because the same author stated it," but "because the same reliable person stated it." And one knows that he is reliable because one has ascertained that he has nothing to gain by telling a lie, etc., not simply because he is known to have told the truth on a different matter.
Therefore, the dGe lugs pa view that the authority of the teaching is a sufficient condition and a valid reason for the authority of the teacher in the Buddha's case is, I believe, correct, inasmuch as it represents Dharmakīrti's thought (cf. Tillemans, op. cit., p. 20).16 Tillemans somewhat tentative suggestion that the authoritativeness of the teaching is only a necessary condition for the authoritativeness of the teacher, if accepted, will have precisely the opposite effect to what Tillemans intends." Tillemans attempts to find an appropriate place for the epithets of the homage verse in the proof of the Buddha's authority. However, if the authority of the teaching is only a necessary condition, this will invalidate the successive proof of the epithets, and consequently also the proof of the Buddha's authority. On the other hand, the authority of the Buddha's main teaching as a sufficient condition does not render this proof of the epithets dispensable. Rather, the Buddha's teaching is a sufficient condition for the establishment of the epithets, which are, in their turn, a sufficient condition for the establishment of his trustworthiness.
58 (1987): 1-17. Apart from minor points of detail, our interpretation differs on the logical subject of this passage According to Yaita the scriptures (agama) are referred to by the anaphoric pronoun tad-whereas I think that this passage refers primarily to the reliable person, i.e., the author of the scriptures, and more specifically to the Buddha. Thus, Yaita translates this compound as a locative tatpurusa (p. 9): "which were taught in the lāgama)." Cl., however, PVSVT 395,14... tadupadi stånām tenäptenopadistānăm ... This is also supported by Karnakagomin's comment on this verse in PVSVT 394,27: hevopädevalarivasya sopayasya bhagavaddarsitasya... here too Yaita interprets differently: "... the causes (upaya) (of both, all of which are the teachings of agama)." The above translation also differs from the one in my Dharmakirti on Compassion and Rebirth, p. 36, which I now consider to be mistaken on some points.
ISCI. PVSV 109.15-19: heyopädeyatadupāyanām tadupadistānam avaipariryam avisamuādah yarha caturnām aryasatyānām vak syamānanitya. tasyasya purusarthopayogino 'bhiyogárhasyavisamvādad visayuntare pi tathārvopaganah, na vipralambhāyla]nuparodhal, nisprayojanavitathābhidhänavaiphalyac ca vakruh.
16 or course, as noted above, it is indirectly sufficient. However, the relation of being a sufficient condition is transitive. If a is sufficient for b, and b is sufficient for c, then a is sufficient for c.
"In the discussion that followed the presentation of this paper Prof. Tillemans has kindly told me that he no longer holds this view.
Is On the successive proof of the Buddha's epithets culminating in the proof of his authority, cf. Franco, op. cit.. chapter 1

Page #9

Appendix: Dharmakirti's commentators on Pramāṇavärttika II 283a-c
PVP Peking 141a5-7= Derge 121b1-3: bstod (P: ston) pa'i tshig gcig ma gtogs pa lhag ma ni tshad mar sgrub par byed pa yin no // ci'i phyir de las gzhan pa'i yon tan mnga' na yang tshad mar gyur pa zhes bya ba nyid bstod par mdzad pa yin zhe na/ston pa de lta'i dngos bstod ni ll zhes bya ba smos te/tshad ma nyid du bstod pa gang yin pa de de'i (P: de'i sa) bstan pa nyid las ni / tshad ma'i de nyid grub don yin //
PVA 166,1-7: tata eva bhagavato 'nenaiva gunena stutih/pramanabhūtatvalakṣaṇena/tad aha/.../kasmad upadeśasya tathābhāvaprāmāṇyalakṣaṇ/a/stutiḥ / tadupadeśataḥ pramäṇatattvasiddhir yatha syaditi.
†The text is slightly corrupt. Unfortunately the Tibetan translation seems also corrupt; cf. PVA(Tib) Peking 183a2-3 = Derge 154b6: ci'i phyir tshad ma'i mishan nyid de lta bu'i dngos po nye bar ston pa ce na.
Considering Prajñākaragupta's introduction to the verse, as well as the context, one expects the Buddha, rather than his teaching, to be the object of praise. In this case one could conjecture tathābhāvena prāmānyalakṣaṇena stutih. Such conjecture is corroborated by Manorathanandin's text quoted below: tena stutiḥ. However, Ravigupta's text does not support this conjecture (cf. below:... tshad ma nyid can gyi...).
PVV(R) Peking 265b3-6= Derge 398b1-3: 'on na nye bar bstan pa bden pas (P: pa yis) bstod pa la dgos pa ci yod ce na/smras pa/de yis bstan pa nyid las (D: la) ni // ston pa de lta'i dngos bstod pa // tshad ma'i de nyid grub don du //...// zhes bya ba la/nye bar bstan pa de lta bu'i dngos po ishad ma'i mishan nyid can gyi bstod pa bcom ldan 'das kyis nye bar bstan pa las ni tshad ma nyid grub pa yin no zhes bstan pa'i phyir ro //
PV
PVV(M) 96,26f.: kasmat punar anekagunasambhārasambhave 'pi prämäṇyenaiva bhagavatah stutih? ity aha-upadeśasya tathābhavaḥ samvadakarvam pramanyam, tena stutir acāryena krta/tasya bhagavata upadeśataḥ pramäṇasya tattvam lakṣaṇam, tatsiddhyartham bhagavaddeśanayaḥ pramāṇaviniscayaḥ, notpreksämätrenety akhyatum ity arthaḥ
PVA
Two Circles or Parallel Lines?
PVA(Tib)
PVP
PVSV
PVSVT
71
Abbreviations
Pramāṇavärttika (Dharmakirti): Yusho Miyasaka, ed. Pramäṇavärttika-kärikä (Sanskrit and Tibetan). Acta Indologica 2 (1971/72): 1-206.
Pramāṇavārttikālankāra (Prajñākaragupta): Rahula Sänkṛtyāyana, ed. Pramāṇavārtikabhashyam or Vārtikālankāraḥ of Prajñākaragupta: Being a commentary on Dharmakirti's Pramāṇavārtikam. Tibetan Sanskrit Works Series 1. Patna: Kashi Prasad Jayaswal Research Institute, 1953. Pramāṇavārttikālankara (Prajñākaragupta) (Tib.): P 5719, D 4221.
Pramāṇavārtikapanjikā (Devendrabuddhi) (Tib.): P 5717, D 4217.
Pramanavärttikasvavṛtti (Dharmakirti): Raniero Gnoli, ed. The Pramanavarttikam of Dharmakirti. the first chapter with the autocommentary, text and critical notes. Serie Orientale Roma 23. Rome, 1960.
Pramāṇavārtikasvavṛttitikā (Karnakagomin): Rahula Sänkṛtyāyana, ed. Acarya-Dharmakirteḥ Pramāṇavärttikam (svärthanumanaparicchedaḥ) svopajñavṛttyä, Karnakagomiviracitaya tarikaya ca sahitam. Allahabad, 1943.

Page #10
72 E. Franco PVV(M) Oramanavartikavrti (Manorathanandin): Swami Dwarikadas Shastri, ed. Pramanavartika of Acharya Dharmakirti with the commentary. "Vritti" of Acharya Manorathanandin. Bauddha Bharati Series 3. Varanasi: Bauddha Bharati, 1968. Pramanavartikavrtii (Ravigupla) ad PV Pramanasiddhi (Tib.): P 5726, D 4224. Wiener Studien zur Tibetologie und Buddhismuskunde. PVV(R) WSTB