Indologica Taurinensia

Author(s): Colette Caillat, Siegfried Lienhard, Irma Piovano, Saverio Sani
Publisher: Comitato AIT
Catalog link: https://jainqq.org/explore/269196/1

Page #1

hin.
569 -
INDOLOGICA
TAURINENSIA
Official Organ of the International Association of Sanskrit Studies
VOLUME XXIX
2003
Proceedings of the Xith World Sanskrit Conference
(Turin, April 3d-81", 2000)
1* part
EDIZIONI A.I.T. TORINO (ITALY)

Page #2

INDOLOGICA TAURINENSIA Official Organ of the International Association of Sanskrit Studies
Founded and directed by OSCAR BOTTO
Editors Colette Caillat, Emeritus Professor of Indology, Université de Paris III. Siegfried Lienhard, Emeritus Professor of Indology, University of Stockholm. Irma Piovano, Supervisor of the scientific activities and cultural relations of the
CESMEO, Torino. Saverio Sani, Director of the Deparment of Linguistics, Università degli Studi di Pisa.
Scientific Committee
President Oscar Botto, Torino, Italy
Members Pierre-Silvain Filliozat, Paris, France
Minoru Hara, Tokyo, Japan Oskar von Hinüber, Freiburg, Germany
Romano Lazzeroni, Pisa, Italy Manfred Mayrhofer, Wien, Austria
Irma Piovano, Torino, Italy
R.K. Sharma, Delhi, India Jaroslav Vacek, Praha, Czech Rep.
Akira Yuyama, Tokyo, Japan
Indologica Taurinensia has been acknowledged as a Journal of High Cultural Value by the Italian Ministero per i Beni e le Attività Culturali, Copyright © 2003 Comitato AIT per la Promozione degli Studi sull'India e sul Sud-Est Asiatico, Torino (Italy) Apart from fair dealing for the purposes of research or private study, or criticism or review, and only as permitted under the current Italian Copyright acts, this publication may only be reproduced, stored or transmitted in any form or by any means, with the prior permission in writing of the Publisher.
Publisher: Comitato AIT
c/o Cesmeo, via Cavour 17 - 10123 Torino (Italy) Tel. +39 011 54 65 64 - Fax +39 011 54 50 31
E-mail cesme @tin.it Printer: Abaco Editori s.r.l., Torino (Italy) Annual Subscription (1 issue): € 30,00 Sole Agents: Herder International Book Center
Piazza Montecitorio 120 - 00186 Roma Tel. +39 06 679 46 28 or 679 53 04 - Fax +39 06 678 47 51
E-mail distr@herder.it - www.herder.it ISSN 1023-3881 Autorizzazione del Tribunale di Torino N. 4703 - 21/7/1994

Page #3

PIOTR BALCEROWICZ
ON THE RELATIONSHIP OF THE NYĀYÂVATĀRA
AND THE SAMMATI-TARKA-PRAKARANA*
. 1. For last few centuries the Jaina tradition came to uncritically attribute such works as the Sammati-tarka-prakarana (STP), the Nyāyāvatāra (NA) and a number of Dvātrimśikās to one and the same author, namely to Siddhasena Divākara. A number of scholars raised doubts as to the authenticity of such attribution and, at least as far as the number of Dvātrimśikās is concerned, there is no doubt that some of them could not possibly have been composed by one and the same person, and certainly not by the author of STP. Despite the necessity to update it', a valuable resume of research heretofore done on Siddhasena Divākara is offered by UPADHYE (1971), who writes: 'Its (= Nyāyâvatāra's - P.B.) constitution (whether it had 32 verses), its authorship by Siddhasena (the author of the Sanmati) and consequently it's date have to remain open questions for a number of
: * The present contribution is a slightly modified version of my paper that was presented at the XI World Sanskrit Conference, Turin, April 3d-84, 2000 and referred to in BALCEROWICZ (2001b: 377) as 'forthcoming'. The major changes are incorporated in the latter part, $$ 6 ff., which contains my analysis of some arguments brought forward by Bansidhar BHATT (2000). The present contribution is a part of my research subsidised by the Alexander von Humboldt-Stiftung.
1. Among recent contributions to the research on Siddhasena, we could list papers of: MOOKERJEE (1971), DHAKY (1981-82), GRANOFF (1989-1990), DHAKY (1995) and BHATT (2000), who while referring to various Jaina Prabandhas giving biographical accounts of Siddhasena, is silent on GRANOFF's (1989-1990) papers, as well BALCEROWICZ (2001a: iii-xl), (2001b).

Page #4

Piotr Balcerowicz
reasons 2.' Perhaps we will never know the true name(s) of the author(s) of STP and NA, but the question of the original number of verses of NA - viz. thirty-two, precisely as the text was rendered to us – seems to have been positively answered recently 3. In the present paper I intend to add some more arguments in favour of my thesis 4 that STP and NA are works of two different Siddhasenas, and their composition is separated by at least 150 years. Following my earlier suggestions that there is no substantial reason why to maintain the identity of the authors of STP and NA, I shall keep the name of Siddhasena Divākara for the author of the Sammati-tarka-prakarana. As a matter of fact, we can not even be sure that the Saṁmati-tarkaprakaraṇa was written by a Siddhasena and we might only rely on such a late statement as that of Abhayadeva-sūri, the commentator on STP, who - in the introductory part of Tattva-bodha-vidhāyini. (TBV) - explicitly mentions the title of the work (A Prabandha called Saṁmati') and the name of its author: Siddhasena Divākara. In order to distinguish this Siddhasena from the author of the Nyāyâvatāra, I will use the name of Siddhasena Mahāmati for the author of the Nyāyâvatāra, following Haribhadra-sūri in this regard 6.
1.1. Here, I would like to draw attention to certain discrepancies in the text of STP and NA that prove, in my opinion, that these two works were written by two different people and at different times. Both the genuine title of STP (whether Saṁmati, Sammati-tarka or Saṁmati-tarka-prakarana) and the name of its author are of secondary importance.
2. UPADHYE (1971: xxiii). 3. See BALCEROWICZ (2001a: xxxvi). 4. See BALCEROWICZ (2001b).
5. TBV.1.1 (introductory part), p. 1.17-18: ...Siddhasena-Divakara - tad-upāyabhūta-Sammaty-ākhya-prakarana-karane ...
6. On the name of the author of NA see: BALCEROWICZ (2001a: xxxiv-xxxvi). See also Pt. Dalsukhbhai MALVANIA (1979: 287-288), UPADHYE (1971: xxiv) and DHAKY (1995: 44). The supposition against the common authorship of NA and STP is enforced by what UPADHYE (1971: xxiv) was right to point out: 'Haribhadra, in his Astaka, quotes the Nyāyāvatāra 2, by referring to its author as Mahāmati. Elsewhere, however Haribhadra speaks plainly about the author of the Sanmati as Divākara and Śrutakevalin.'

Page #5

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
33
Most conspicuous differences, viz. Sanskrit of NA and Prakrit of STP as well as an overall stylistic dissimilarity, would certainly be a highly unconvincing argument against joint authorship of both works. Much less so is, however, the general outline and matters discussed in both works: apparently the single feature these two works share is their epistemological concern. With the excep-tion of two verses (NA 29-30), NA discusses the question of epis-temic validity (prāmānya) and the definition of pramāna as well as its divisions and their definitions, there being no single reference to the method of the seven-fold predication (sapta-bhangi), an important issue in STP. The text of NA - both in its rigid outline and vo-cabulary - clearly follows the epistemological tradition of Dinnāga-Dharmakirti school, especially the Nyāya-bindu and the Nyāya-praveśa of Sankarasvāmin ?. In its consistent structure, a definition is followed by its explanation and further exemplification; each thesis derives from the preceding. It is hard to find in NA typically Jinistic technical terms, in the sense that one does not have to be acquainted with the Jaina doctrine and creed in order to understand the contents of NA in its entirety. The contents of STP is, in it's turn, derives entirely from the Jaina tradition, with a detailed exposition of the theory of multiplexity of reality (anekānta-vāda) - that includes nayas (dravyāstika and paryayāstika, as well as the set of seven viewpoints: rju-sūtra, etc., especially STP 1.3-5, 7-18, 23, 31), niksepas (STP 1.6, 40 ff.), sapta-bhangi (STP 1.36-40) - and with a discussion of the cognitive faculties (upayoga) in Chapter 2, as well as ontological and ethical and soteriological questions of predominantly Jaina relevance. None of these topics is present in NA. Its outline seems to be entirely different from NA, much less rigid and consistent, almost chaotic, the same issue being discussed on several occasions. STP seems to be rather a plain exposition of some aspects of the Jaina doctrine, and a refutation of some contrary doctrines happens occasionally. STP constantly uses locutions and terminology that had been coined as early as in the Canonical literature and takes it for granted that the hearer / reader is well acquainted with Jaina doctrine and peculiarities of expression: certainly a non-Jaina would find it
7. See BALCEROWICZ (2001a: viii-xxix). Also BHATT (2002: 79-81) provides a useful list of structural and terminological similarities between NA and NB /NBT.

Page #6

Piotr Balcerowicz
rather difficult to follow the contents of STP without any prior knowledge of Jainism. Thus, one may have an impression that NA and STP derive from alto-gether different intellectual backgrounds 8.
1.2. Any direct comparison of NA and STP is indeed an unfeasible task in view of the fact that neither the contents nor the vocabulary of NA and STP overlap. Accordingly, we cannot pinpoint a single concept of expression in order to be able to see whether it is treated in the same manner in both texts.
Even a positive answer to the question (viz. the confirmation that a given term or concept is given precisely the same meaning both in STP and NA) would neither prove the same authorship for both works nor disprove the thesis that NA and STP were written by two different authors. Such overlapping could only have a supportive-corroborative strength, but would in no way be decisive; but its corroborative strength would be directly proportional to the uniqueness and singularity of the identical treatment of a given term / concept in both works and their divergence from the whole Jaina tradition in the case of the given term/concept.
To find a concept, however, that is explicated entirely differently in both works or that rests on completely different presuppositions would, on the other hand, prove that NA and STP could not have been written by one and the same person. Indeed, there seem to be indeed some points that presuppose an entirely different conceptual frame work in NA and STP.
2. One of them is the question of the cognitive faculties (upayoga) of the living element (jīva). It is problematic what exactly the relation between the two cognitive faculties (upayoga), i.e. jñāna (cognition) and darśana (insight / conation), on the one hand, and pramāņa (cognitive criterion), on the other, was for Siddhasena Divākara in STP. Commenting on STP, Abhayadeva-sūri - against the general Jaina tradition - apparently takes upayoga to bifurcate into
8. The above issues are discussed in more detail in BALCEROWICZ (2001b: 353354).

Page #7

On the Relationship of the Nyayâvatāra and the Sammati-tarka-prakarāṇa 35
pratyakṣa and parokṣa, and the five subvarieties (avadhi, manaḥparyaya, kevala and mati, śruta, respectively) to overlap with the Agamic varieties of pramāna".
Traditionally, pramāņa related only to the question of the validity of cognition (jñāna), viz. to one aspect of the cognitive faculties (upayoga). Thus, Tattvârtha-sutra in one chapter discusses five varieties of cognition, and only then adds that these are divided into two pramāņas 10. This opinion is further supported by Tattvârthadhigamabhāṣya 11. The question of the cognitive faculties (upayoga) is discussed separately in the next chapter of TS 12; the upayogas comprise all possible cognitive faculties irrespective of their validity. For instance, four varieties of erroneous cognition (ajñāna) as well as three varieties of erroneous conation / fallacious insight (adarśana) are still cases of upayoga 13. This is precisely the approach that overlaps with such Canonical works as the Pannavaṇāsuttaṁ 14, where the question of pramāņa is discussed separately.
There seems to be no indication, in my opinion, as regards the exact relation of the upayogas and the pramāņas in STP, and we must leave the question open.
9. See, e.g., TBV ad 2.1 (p. 457.6-7): ... upayogo 'pi paraspara-vyapekṣasāmānya-viseṣa-grahaṇa-pravṛtta-darśana-jñāna-svarupa-dvayâtmaka- pramāņam darśana-jñānaîkānta-rūpas tv apramāṇam ...; TBV ad 2.1 (p. 458.4-5): nirākārasākārôpayogau tûpasarjani-kṛta-tad-itarâkārau sva-viṣayâvabhāsakatvena pravartamānau pramāṇam na tu nirastêtarâkārau; and TBV ad 42 (p. 650.23-25): athava pratyakṣa-parokṣa-rupaḥ sankṣepato dvividha upayoga ātmanaḥ. tatra pratyakṣôpayogas trividhaḥ avadhi-manaḥ-paryāya-kevala-bhedena. ... paroksôpayogas tu mati-śrutarupo dvividhah.
10. TS 1.9-12: mati-śrutâvadhi-manaḥ-paryāya-kevalāni jñānam. tat pramāṇe. adye parokṣam. pratyakṣam anyat.
11. See, e.g., TBh 1.12: mati-śrutābhyam yad anyat trividham jñānam tatpratyakṣam pramāṇaṁ bhavati.
12. TS 2.8: upayogo lakṣaṇam, TBh 2.8: upayogo lakṣaṇam jivasya bhavati. TBh 2.9: sa upayogo dvividhaḥ sākāro anakaraś ca jñānôpayogo darśanôpayogaś cêty arthaḥ. TBh 1.3: jñāna-darśana-upayoga-lakṣano jiva iti vakṣyate.
13. See TBh 2.9.
14. Pann 9.1912-1914: ... duvihe uvaoge pannatte. tam jahā - sāgarôvaoge ya aṇāgārôvaoge ya. ... tam jahāḥmati-nana-sägarôvaoge, suya-nana-sagarôvaoge, ohinaṇa-sāgarôvaoge, mati-anaṇa-sagarôvaoge, suya-anaṇa-sāgarôvaoge, vibhanganaṇa-sagarôvaoge. ... tam jahā - cakkhu-damsaṇa-aṇāgarôvaoge, acakkhudamsaṇa-aṇāgarôvaoge, ohi-damsana-aṇāgarôvaoge ya.

Page #8

36
Piotr Balcerowicz
2.1. However, the second chapter of STP opens with the verse that distinguishes two kinds of the soul's cognitive faculties (upayoga), cognition (jñāna) and conation, or insight (darśana) 'S:
STP 2.1: jam sāmaņņa-ggahaņas daṁsaņam eyam visesiyam ņāņam / donho vi nayāņa eso padekkaṁ attha-pajjāo //
- 'Insight is the grasp of the general. Cognition is one, characterised by the particular. This modality of the object (viz. its general and particular aspect] is individually [the contents) for both viewpoints, [i.e. substance-expressive (dravyārthika) and the modal, or mode-expressive (paryāyārthika).
What the first hemistich of the verse states is that insight / conation (darśana) grasps the general (sāmānya), whereas the cognition (jñāna) grasps the particular (viśesa). Here the discrimination between the sāmānya and the višesa apparently does not pertain to the usual distinction of the universal (as related to the class notion, jāti, language and concepts, kalpanā) and the individual (vyakti, bheda). The dividing line is clearly the opposition between “general, indistinct, unclear' (for sāmānya) and 'particular, distinct, specific' (for visesa). Such a position was definitely liable to censure not only from the Buddhist side, in as much as it could easily be interpreted in opposition to perception (pratyakșa; called darśana), grasping the viśesa, and inference (anumāna), etc., grasping the sāmānya.
Besides, certain inverted, as it were, parallelism of formulations (sāmānya - darśana and višesa - jñāna) as compared with Dharmakirti's distinction of sva-laksana as a respective province (vişaya) for perception (pratyaksa) and sāmānya-laksana as a respective province (vişaya) for inference (anumāna) is likewise conspicuous. If the author of STP had been acquainted with Dharmakirti's ideas, he would not have failed, I expect, to enter into a polemics or elaborate on the issue, just the way Siddhasena Divākara defends his position as regards the distinction of darśana and avagraha (see § 4.2) against
15. I deliberately do not employ the term “perception' (reserved for pratyaksa) to render 'darśana', in order to preserve the distinct character of the two terms and to avoid hasty identification of pratyaksa and darśana, that in general epistemological discourse are often equated, but are often kept distinct in case of the Jaina theory of иpаyoga.

Page #9

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāna
37
possible misunderstanding (non-Jainas could have easily taken his darśana to be pratyaksa, or avagraha); by the same token one would expect him to have done the same in the case of the categorisation of the sāmānya related to darśana and the višeşa related to jñāna. And precisely in the same way as the author of NA anticipated a possible criticism from the side of his opponents (the school of Dharmakirti in the first place) in the case of svârtha and parârtha, while extending the distinction to both pratyaksa and anumāna, 16 as well as in the case of non-erroneousness (abhrāntatva) of both pratyakșa and anumāna "7.
I believe the author of STP did not merely ignore any possible criticism, for instance, from the Buddhist side, but he was not even aware that his statement could trigger such a criticism at all. This seems to be a serious indication that he flourished before Dinnāga and Dharmakirti.
2.2. On the other hand, in the text of NA we frequently find - as I have just mentioned – more or less concealed polemics directed against Dinnāga and Dharmakirti, e.g.: (1) in NA 5 18, the phrase stating that inference (anumāna) 'is non-erroneous because it is a cognitive criterion, just like perception', is a refutation of Dharmakīrti's thesis that inference, even though being a cognitive criterion, is erroneous 19, (2) NA 13 20 is reminiscent of the trairūpya doctrine and NB 3.1 in the sense that it states new and better condi
16. Cf. JACOBI (1926: iii, n. 1), VAIDYA (1928: xviii-xx), BALCEROWICZ (2001a: iv ff., xii ff.).
17. Cf. JACOBI (1926: iii), VAIDYA (1928: xviii-xx), BALCEROWICZ (2001a: v-viii, xvi-xvii). 18. NA 5: sādhyâvina-bhuno lingāt sādhya-niścāyakam smrtam/
anumānam, tad abhrāntam pramanatvāt samaksavat // - - 'Inference is regarded traditionally to determine the inferable property on account of the inferential sign, which is inseparably connected with the inferable property. It is non-erroneous because it is a cognitive criterion, just like perception.'
19. See BALCEROWICZ (2001a: xvi-xvii). 20. NA 13: sādhyâvinā-bhuvo hetor vaco yat pratipadakam /
parârtham anumānam tat pakşâdi-vacanâtmakam // - - 'Such an utterance that demonstrates the logical reason as inseparably connected with the inferable property is the inference for others, whose essence are propositions, like the thesis, etc.

Page #10

38
Piotr Balcerowicz
tions of validity 21. All these and similar passages prove that there can be no doubt that NA was written after Dharmakirti 22. Furthermore, NA reveals its author's general anticipation of possible criticism, e.g. verses NA 2-323 are meant to combat possible censure that the definition of cognitive criterion (stated in NA 1) is unnecessary. It is rather doubtful that the author of NA, so sensitive as regards possible criticism, might at the same time be the author of STP, so much oblivious of how controversial the verse of STP 2.1 could appear 24 .
3. Leaving aside a possible relationship between pramāṇa and upayoga in both works, let us take a look at the paradigms of valid types of cognition and their division into kinds of cognitive criterion (pramāna).
3.1. Let us begin with NA. We do not find any mention of cognitive faculties (upayoga) or insight / conation (darśana) there in the first place. After formulating the definition of pramāna in the first hemistich 25, NA lcd states that there are only two pramānas: *[Cognitive criterion is) two-fold: perception as well as indirect cogni
21. See BALCEROWICZ (2001a: xvii-xviii). 22. See BHATT (2000) and BALCEROWICZ (2001a: (2001a: iii-xl)). 23. NA 2-3: prasiddhāni pramāṇāni vyavahāraś ca tat-krtah/
pramāņa-lakṣaṇasyôktau jñāyate na prayojanam // prasiddhānām pramāņānām laksanôktau prayojanam/
tad-vyāmoha-nivettiḥ syād vyāmūdha-manasām iha // - - '[Rejoinder:) "Cognitive criteria are well-known and everyday practice is accomplished by them; [therefore) no purpose is known for stating the definition of cognitive criterion". [Reply:] The purpose for stating the definition of cognitive criteria, (although they are) well-known, should be the eradication of disorientation concerning that [definition of cognitive criterion) in the case of the disoriented-minded here.
On this see BALCEROWICZ (2001a: Xv-xvi: 8 4).
24. Another point of divergence is the way STP and NA approach logical issues and their different attitudes towards the question of reasoning and the proof formula (sādhana, pañcavayava-vākya). These issues are discussed at length in BALCEROWICZ (2001b: 360-362).
25. NA lab: pramānam sva-parâbhāsi jñānam, bādha-vivarjitam / - ‘The cognitive criterion is cognition revealing itself and something else [different from it and it is) free from subversion.'

Page #11

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāna
39
tion, corresponding to (the way of] determination of the cognoscible’ 26. Subsequent verses provide more information as to the character of the two pramāṇas. The understanding of the nature of the direct cognition, or perception (pratyakşa), radically diverts from the traditional strand in the Jaina tradition which regarded it to be supra-sensory, and coincides with the general Indian epistemic tradition that took it to be the cognition directly derived through and with the help of sense organs. .
Such an interpretation is not only confirmed by Siddharsi-ganin in his commentary 27 as well as by the whole subsequent tradition that followed NA, but also by the internal evidence. Indeed, the definition of pratyaksa and paroksa formulated in NA 428 is basically tautological in the logical sense, and henceforth brings no information 29. In
other words, the two definitions allow for all possible interpretations! · Fortunately, several other passages give us a clear idea that the main
division of the pramānas is along the lines of general Indian epistemological tradition, viz. the touchstone for directness of cognition (pratyakșa) are sense organs, whereas the indirect cognition (parokșa) comprises inference, verbal testimony, analogy, etc.
sabdo jivit indriya-was-sáksāt-kátne linguisticn the Canone lin
26. NA lcd: pratyakşam ca parokşam ca dvidhā, meya-viniścayāt //
27. NAV 1: pratyakşam cêty-ādi; tatra siddhânta-prasiddha-pāramārthikapratyakşâpekşayâkșa-sabdo jiva-paryāyatayā prasiddhaḥ. iha tu vyāvahārikapratyakșa-prastāvād akşa-dhvanir indriya-vacano grhyate. tataś câkşam pratigatam pratyakşam. yad indriyam āśrityôjjihīte 'rtha-sākṣāt-kāri jñānam tat pratyakşam ity arthaḥ - ... Perception, etc. Concerning that [issue), the linguistic unit "perceiving organ” - with regard to ultimately real perception well-known from the Canon - is well-known as a synonym of the living element. Here [in this verse], however, the linguistic unit “perceiving organ” is used - since [we) are dealing (here) with the conventional perception - as an utterance [denoting) the senses. And, therefore, that which has gone towards "the perceiving organ" is perception. The meaning is as follows: such a cognition perceiving directly an object which commences resorting to the senses is perception'. 28. NA 4: aparoksatayârthasya grāhakam jñānam idrśam/
pratyakşam, itaraj jñeyam parokşam grahanêkṣayā // - - 'Such a cognition that grasps an object not-indirectly is perception; the other one should be known as the indirect cognition, as far as [the manner of) grasping [an object) is considered
29. To put it plainly, 'direct is not non-direct' for pratyakșa and 'what is not direct is non-direct' for paroksa, or a = -(-a)' and '-a = -a', respectively. If we want to keep the sentential negation (-) distinct for the predicative negation (~), we can symbolise the two sentences respectively as: 'a= (-a)' and '-arna'.

Page #12

40
Piotr Balcerowicz
In the first place, subsequent verses illustrate two main divisions of the indirect cognition, viz. inference (anumāna) and the verbal knowledge (sabda) 30. They are indirect because inference is based on the inferential sign (linga), or the logical reason (hetu), and because the verbal knowledge is derived from a sentence (vākya). Besides, NA 5 and NA 6, while discussing the issue of validity and erroneousness, contrast perception (pratyakṣa) with inference (anumāna) 31. They are also contrasted in a few other passages 32. While providing a definition of the thesis in NA 14 33, we find the statement that 'the thesis is not revoked by perception etc.' (pakṣaḥ pratyakṣady-anirākṛtaḥ), viz. the soundness of the thesis is generally verifiable with ordinary means and common cognitive apparatus: by implication 'perception' has to mean here ordinary, or sensory perception, employed by everyone, and not extra-sensory perception - viz. either clairvoyance (avadhi-jñāna), mind-reading (manaḥ-paryāya-jñāna) or absolute knowledge (kevalajñāna) - not accessible to common people. The phrasing emulates formulations of a validity condition well-known form several other works, just to mention a few, NP 34, NM 35 and PBh 36. Similarly, the
30. NA 8:
drstêṣṭâvyahatad vākyāt paramârthâbhidhayinah/ tattva-grähitayôtpannaṁ mānam sabdam prakirtitam // -
- 'The cognitive criterion - arisen as grasping reality due to a [momentous] sentence, which is accepted as that what is experienced and which is not contradicted [and] which communicates the ultimate truth is declared [to be] the verbal knowledge.
31. NA 5:
sadhyâvina-bhuno lingat sadhya-niścāyakaṁ smṛtam/ anumānam, tad abhrāntam pramāṇatvāt samakṣavat // na pratyakşam api bhrāntam pramāṇatva-viniścayat/ bhrāntam pramāṇam ity etad viruddham vacanam yataḥ //
- 'Inference is regarded traditionally to determine the inferable property on account of the inferential sign, which is inseparably connected with the inferable property. It is non-erroneous because it is a cognitive criterion, just like perception. Neither is perception erroneous, because it is determined to be a cognitive criterion, since [to say:] a cognitive criterion is erroneous is a contradictory utterance."
32. E.g., NA 11ab pratyakṣenânumānena prasiddhartha-prakāśanat/ 33. NA 14: sadhyabhyupagamaḥ pakṣaḥ pratyakṣâdy-anirākṛtaḥ/ tat-prayogo 'tra kartavyo hetor gocara-dipaka // -
"The thesis is the acceptance of the inferable property; [it] is not revoked by perception, etc.; the pronouncement of it has to be made here as showing the domain of the logical reason."
NA 6:
34. NP(1) p. 1.5-7 (= NP(2) 2.1, p. 72): tatra pakṣaḥ prasiddho dharmi prasiddhaviseṣaṇa-visiṣṭataya svayam sadhyatvenêpsitaḥ. pratyakṣâdy-aviruddha iti vākya-seṣaḥ.

Page #13

On the Relationship of the Nyäyâvatāra and the Sammati-tarka-prakarāņa
41
cases which invalidate a thesis and turn it into a fallacious thesis (pakşâbhāsa) 37, enumerated in NA 21 38, overlap with formulations found both in Jaina texts posterior to NA 39 and in other epistemological-logical traditions in India 40. In other words, the meaning and implications of the term pratyaksa are so deeply interwoven in the epistemic system of NA in the very same manner as the term was
- "Out of these [members of the 'syllogism'], the locus is a well-known subject; it is intended by [the propounder) himself to be proved as qualified by a well-known distinctive feature. The following annotation to the [above] statements [is tacitly implied: the locus) is not contradicted by perception, etc.' 35. NM 1 (as restored by G. TUCCI):
pakşâdi-vacanānīti sādhanam; tatra hi svayam /
sādhyatvenêpsita - pakso viruddhârthânirākstaḥ // - - 'Proof (consists in) utterances (expressing) the locus, etc. Among these very [utterances), locus is what is intended by [the propounder) himself as the inferable property [and] which is not revoked by objects (that are] contrary [to it).'
. 36. PBh [266], p.49-50: avirodhi-grahaņāt pratyakşânumănâbhyupagata-svaśāstra-sva-vacana-virodhino nirastā bhavanti.
37. The sequence of NA 12 (on vacas / vākya and pratyaksa), NA 14 (pratyakşâdy-anirākrtah) and NA 21 (pratipadyasya yaḥ siddhah pakşâbhāso 'kşaliñgatah - it is the standard division of paksabhāsa) may prove that perception is sensory, hence shows that pratyakşa here is also indriya-pratyakşa (not only anindriyapratyakşa). 38. NA 21: pratipadyasya yaḥ siddhah pakşâbhāso 'kşa-lingataḥ/
loka-sva-vacanābhyāṁ ca bādhito 'nekadhā matah // -- - 'The fallacy of the thesis is what is [already proved for a (person] to be taught, (what) is subverted by the perceiving organ and by the inferential sign as well as by the opinion prevalent among people and by one's own utterances; it is known [to be) manifold.
39. See, e.g., PALV.6.15,16: /151 (pakşâbhāsaḥ] bādhitaḥ pratijñānumānågama-loka-sva-vacanaiḥ. /16/ tatra pratyakşa-bādhito yathâtuşno 'gnir dravyatvāj jalavat. Cf. NAV ad loc.: tathâkşa-lingato 'dhyaksa-hetubhyām loka-svavacanābhyām ca bādhitas tiras-krto yaḥ sa pakşâbhāsaḥ. ... pratyakşa-bādhito yathā: niramśāni svalaksanāni, paraspara-viviktau vā sāmānya-višeşāv iti. anumānabādhito yathā: nâsti sarva-jña iti. loka-badhito yathā: gamyā mātā iti. sva-vacanabādhito yathā: na santi sarve bhāvā iti.
40. See, e.g., NB.(2).3.49-53: 7491 (2) tatra pratyakşa-nirāksto yathā: aśrāvanaḥ sabda iti. /50/ (3) anumāna-nirākrto yathā: nityaḥ sabda iti. /517 pratiti-nirākrto yathā: acandraḥ śasîti. /521 (5) sva-vacana-nirākrto yathā: nânumānam pramānam. 153/ iti catvārah pakṣābhāsā nirākstā bhavanti; also to NP.(2).3.1: sādhayitum isto 'pi pratyakşâdi-viruddhah pakşâbhāsah, tad yathā: (1) pratyakşa-viruddhaḥ, (2) anumāna-viruddhaḥ, (3) āgama-viruddhaḥ, (4) loka-viruddhaḥ, (5) sva-vacana-viruddhah, (6) aprasiddha-visesanah, (7) aprasiddha-višeşyah, (8) aprasiddhô-bhayah, (9) prasiddha-sambandhaś cêti // tatra...

Page #14

Piotr Balcerowicz
understood in such systems as Nyāya, Vaišeșika or in Buddhist tradition, that it is thoroughly unfeasible to interpret it differently, in agreement with the Jaina Agamic tradition. The only exception to the sensory interpretation of pratyakṣa is provided in NA 2741.
The exceptional case of perception is thus the absolute cognition (kevala). But such a standpoint does not contradict the thesis of th general Indian epistemological-logical tradition as the background for NA; in fact, it corresponds to the idea of mystic insight (yogipratyakşa) as an additional, supra-sensory kind of perception, wellknown not only from the Buddhist tradition 42. Thus, in NA we find two subdivisions of pramāņa: (1) perception (pratyakşa), divided into sensory an supra-sensory (kevala), 43 and (2) indirect cognition (parokşa) that comprises inference (anumāna) and verbal testimony (śābda).
3.2. We find an entirely different world of epistemic ideas and notions in STP. In the first place, not only are the terms pramāna, pramiti, māna, pramā, or any equivalent, entirely absent from STP, but even the idea of cognitive criterion (pramāna) and the concept of validity (prāmānya) nowhere occur in STP. It is even more surprising in view of the fact that not only these questions are crucial in NA, but
41. NA 27: sakalâvarana-muktâtma kevalam yat prakāśate /
pratyakşam sakalârthâtma-satata-pratibhāsanam // - - 'That whose essence is freed from all veils, which shines as [something) absolute is perception representing constantly the essences of all objects.'
42. See, e.g., NB.1.11: bhūtârtha-bhāvanā-prakarsa-paryanta-jam yogi-jñanam cêti; or VS.9.13: ātmany ātma-manaso-samyoga-višesād ātma-pratyaksam.
43. Siddharşi-ganin confirms this in NAV 1: tataś ca sarva-jñānānām yat svarūpa-samvedanaṁ tad api pratyakşam ity uktam bhavati, tatrâpi svarāpasya grāhyasya sākṣāt-karana-sadbhāvād iti. aksebhyah parato vartata iti parokşam. aksavyāpāra-nirapeksam mano-vyāpārenâsākṣād-artha-paricchedakam yaj jñānam tat parokşam iti bhāvaḥ. - 'And, therefore, what it amounts to is the following: that which is a sensation of the intrinsic nature (of an object in case of acts of omniscience is perception, as well, because direct perception of the intrinsic nature (of an object] which is to-be-grasped is present also in (case) of these [acts of omniscience). That which operates aloof from the perceiving organs, [i.e., the senses), is indirect cognition. The intent is as follows: the cognition which determines an object indirectly by the operation of the mind, independent of the operation of the perceiving organs (sc. the senses) is the indirect cognition.'

Page #15

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāna
43
also it was Siddhasena Mahāmati who formulated the first descriptive definition of pramāņa in the history of Jaina epistemology 44.
e keywords, as it were, are in STP upayoga, five kinds of jñāna, and four kinds of darśana. The cognitive faculties are discussed at length especially in the second chapter, e.g. in STP 2.1-5, 18 ff., 30. The distinction between the two kinds is due to their respective provinces (STP 2.1, vide supra p. 4): conation, or insight (darśana), grasps the universal (samanya-grahanam darśanam), whereas cognition (jñāna) grasps the particular (viseșitam (=viseșa-grahanam) jñānam); this way they are related to two viewpoints: the substantial, or substance-expressive viewpoint (dravyârthika) and the modal, or mode-expressive viewpoint (paryāyārthika) respectively 45. However, none of the cognitive faculties is entirely bereft of the other viewpoint, STP 2.2 40: insight, being basically related to the substantial viewpoint, comprises secondarily the modal viewpoint, and the same rule applies mutatis mutandis to cognition. The difference between insight (darśana) and cognition (jñāna) remains on all stages, including telepathy (manah-paryāya), but their distinct character disappears on the level of the absolute cognition-insight (kevala), according to Siddhasena Divākara 47.
3.3. To prove this identity of jñāna and darśana in case of an omniscient person (kevalin) is one of the main points of STP, for which the work was subsequently criticised by many Jaina thinkers. The thesis that jñāna and darśana have their end in the absolute cognition (kevale sanidhane) was apparently so important for Divākara that he considered it imperative not only to restate the same idea in
44. See BALCEROWICZ (2001a: xiv-xv) and (2005, n. 31).
45. Another difference is mentioned in STP 2.11: jñāna is sākāra (distinctive) and vyakta (manifest, distinct), whereas darśana is anākāra (non-distinctive) and avyakta (not manifest, indistinct). See also STP 2.14 on the distinction sākāraanākāra. 46. STP 2.2: davvatthio vi heūna daṁsane pajjavatthio hoi/
uvasamiyāibhāvam paducca ņāne u vivariyam // 47. STP 2.3: manapajjava-nānasto nānassa ya darisanassa ya viseso /
kevala-ņāņam puņa damsanam ti ņānam ti ya samānam // - - "Up to the telepathy, cognition and insight are different; however, [in case of] the absolute cognition insight and cognition are the same.'

Page #16

44
Piotr Balcerowicz
STP 2.8 48, but also to introduce the pārva-pakṣa verse of STP 2.22 49 that claims the conditional difference between the absolute cognition (kevala-jñāna) and the absolute insight (kevala-darśana), which is subsequently refuted. This treatment of kevala is entirely with one exception 50 - different from its treatment in NA 27 (vide supra p. 8, n. 41). In NA there is not slightest trait of this highly debatable issue and the kevala knowledge is plainly stated there to be a special kind of pratyakṣa.
Furthermore, there is even an indirect indication as regards the real standpoint of NA. According to NA 7, the representation (pratibhāsa), or the mental 'mirroring' of an object, is a characteristic fea
48. STP 2.8: samtami kevale damsanammi nanassa sambhavo natthi/ kevala-nanammi ya damsanassa tamha sanihanaim // 49. STP 2.22: damsana-puvvam nāņam nāṇa-nimittam tu damsanam natthi/ tena suvinicchiyāmo damsana-ṇāṇāṇa annattam //
[The absolute] cognition is preceded by [the absolute] insight, but [the absolute] insight is not conditioned by [the absolute] cognition; hence we rightly conclude that there is difference between both [the absolute] cognition and [the absolute] insight."
This verse, in my opinion, does refer to the alleged distinction between kevalajñāna and kevala-darśana in view of the second hemistich of STP 2.21 that introduces. it: 'Just like [sensation], in the same manner, the difference between both the absolute [cognition and insight] is this much only...' (jaha ettha kevalāņa vi visesanam ettiyam ceva //).
50. This is the description of the absolute knowledge (kevala) in NA 27 and in STP 2.17. There is indeed some conspicuous similarity as regards the character of kevala (similarities are underlined):
(1) NA 27: sakalâvarana-muktâtma kevalam yat prakāśate/ pratyakşam sakalârthâtma-satata-pratibhasanam //
'That whose essence is freed from all veils, which shines as [something] absolute is perception representing constantly the essences of all objects.'
(2) STP 2.17: tamha cauvvi-bhāgo jujjai na u nāṇa-damsana-jiņāṇam/ sayalam aṇāvaranam anamtam akkhayam kevalam jamhā //
=
- "Therefore it would follow that [cognition would be] four-fold [not five-fold], but there [would be] no [separate] cognition and insight [in case of] Jinas, if the absolute cognition is without veils, eternal, imperishable."
The verse of STP 2.17 is a rejoinder of Siddhasena Divakara's opponent who draws the conclusion from Divakara's thesis about the identity of cognition and insight. It does not express Siddhasena Diväkara's own position! Hence the similarity of expression is only apparent and by no means conclusive since it may be due simply to a general, standard way of describing the nature of the absolute cognition..

Page #17

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
45
ture of cognition (jñāna) 51. At the same time, NA 27 defines the absolute cognition (kevala) as 'a perception that is a representation [revealing) constantly the essences of all objects' (p. 8, n. 41). Thus, the absolute cognition (kevala) is defined with the help of the characteristic mark of jñāna, and the only difference between kevala and ordinary jñāna is that the former is a representation of all things, whereas the latter is a representation limited only to some of their aspects. This can be interpreted as a proof that the kevala of NA is jñāna and the aspect of darśana is either not relevant or not important for Siddhasena Mahāmati, in so far as the author of NA does not consider it imperative to explain the nature of kevala explicitly. This would be surprising, if we assumed that the same person wrote also STP and was once in pains to prove that both jñāna and darśana become one single unity on the level of kevala.
4. In STP we come across the fivefold division of cognition: (1) mai-nāna = mati-jñāna (STP 2.6, 23, 27) that corresponds to ahinibohe = ābhinibodhika-jñāna (STP 2.32), (2) suya-ņāņa = śruta-jñāna (STP 2.16, 27, 28), (3) ohi = avadhi (STP 2.16, 29), (4) maņapajjava = manah-paryāya (STP 2.3, 16, 19, 26) and (5) kevala (STP 2.3, 5, 8, 14, 17, 20, 34, 36, 37). This is the typology well known from TS 1.9-12 (vide supra n. 10) as well as from Thāņ 60 52 (vide infra p. 11 ff.).
51. NA 7: sakala-pratibhāsasya bhrāntatvâsiddhitah sphutam/
pramānam svânya-niścāyi dvaya-siddhau prasidhyati // - - 'Since it is incorrect to assume erroneousness of all representation, cognitive criterion, which is patent [and] which determines itself and something different [from it), proves to be correct with regard to establishing the dyad.' Cf. NA 12: pratyaksa-pratipannârtha-pratipādi ca yad vacah /
pratyakşam pratibhāsasya nimittatvāt tad ucyate // - ‘And such an utterance that demonstrates an object recognised through perception is called perception, because it is the external sign for the representation.
52. This must have been the tradition from which TS evolved. In Thān 336 we come across another strand of epistemological tradition (vide infra, n. 56). In fact, the epistemological ramifications of STP are even closer to Thān 60 than to TS 1.9-12, inasmuch as both STP and the divisions found in Thān 60 lack the 'pramāna' element, in other words, pramāna does not occur in the classification at all. Clearly, this strand is not the tradition from which NA stems from. We do, however, find in the Jaina Agamas another strand - in Viy 5.4.26[3] and in Thāņ 336 (partly AņD 436] (vide infra, n. 56) - that goes back to the four-fold division of cognitive criteria, viz. pratyakşa-anumāna-aupamya-āgama, apparently derived from the Nyāya and

Page #18

46
Piotr Balcerowicz
What is now the relation between the sensuous cognition (matijñāna, ābhinibodhika-jñāna) and the opposition pratyakṣa-paroksa according to the author of STP? In other words: under which heading did STP classify the sensuous cognition (mati-jñāna, ābhinibodhika-jñāna)?
4.1. If we first take recourse to the Agamic tradition to decide whether the sensuous cognition (mati-jñāna, ābhinibodhika-jñāna) is to be classified as direct or indirect according to the Jaina tradition, we can easily find out that the Agamic tradition was not at all unanimous as regards the divisions and subdivisions of upayoga and jñāna. Accordingly, as far as classifications of cognitive faculties are concerned, we find at least five irreconcilable strands in the Canonical tradition prior to STP 53:
Model I: upayoga: (1) jñāna: (1) ābhinibodhika-jñāna, with its four
stages: (a) avagraha, (b) ihā, (c) apāya, (d) dhāraṇā, (2) śrutajñāna, (3) avadhi-jñāna, (4) manaḥ-paryāya-jñāna, (5) kevalajñāna, (II) darśana: (1) cakṣur-darśana, (2) acakṣur-darśana, (4) avadhi-darśana, (5) kevala-darśana. 54
Anvikṣiki tradition. These two traditions find their partial reconcilliation in AnD 435440 (p. 173-174). I do not find anything in AnD 435-440 that would essentially contradict the typology of NA, granting that NA subsumed under the paroksa head the categories of anumāna-aupamya-āgama found in AṇD and leaving aside the fact that NA is silent on the issue of the relationship of the pramana theory and the upayogas. Thus, it might have been the tradition from which the theoretical considerations of NA partly evolved.
53. Cf. e.g. SHASTRI (1990: 196-213, 'Divisions of Knowledge'), who gives a reliable account of various canonical positions regarding divisions of cognitive faculties (the tables on pp. 202-211 are generally quite handy). He distinguishes four strands, overlooking two models: my Model III, which is basically of non-Jaina origin, and Model VI. In the scheme of six models, Model I emphasises the subdivisions that could contribute to our case. Principally, I leave aside all sub-divisions of testimony (śruta), telaesthesia (avadhi) and telepathy (manaḥ-paryaya), since they are not directly relevant here.
54. Viy 8.2.22-23 (p. 336): ...pamcavihe nane pannatte, taṁ jahā – ābhinibohiya-nāṇe suya-nāṇe ohi-nāņe maṇapajjava-nāne kevala-nāne. abhinibohiya-nāņe catuvvihe pannatte, tam jahā - uggaho ihā avão dhāraṇā. Cf. NamS 6[28-29] and 8 (pp. 6, 9): ...nāṇa-damsana-guṇāṇam... ṇāṇam paṁcaviham pannattam. taṁ jahā abhinibohiya-nāṇam suya-nānam ohi-nāṇam manapajjava-ṇāṇam kevala-nāṇam.
Viy 2.10.9[2] (p.115): ...jive nam aṇamtānam ābhinibohiya-nāṇa-pajjavāṇam evam suta-nāṇa-pajjavāṇaṁ ohi-nāṇa-pajjavāṇaṁ maṇapajjava-nāṇa-pajjavāṇam

Page #19

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāṇa
47
Model II: jñāna: (I) pratyakṣa: (1) kevala, (2) no-kevala: (a) avadhi, (b) manaḥ-paryaya, (II) parokṣa: (1) abhinibodhika: (a) śruta-niḥ śrita (arthâvagraha + vyañjanâvagraha), (b) aśruta-ni – śrita (arthâvagraha + vyañjanâvagraha), (2) śruta 55. Model III: pramāņa / hetu: pratyakṣa, anumāna, aupamya, āgama 56. Model IV: pramāṇa: (I) jñāna: (1) pratyakṣa: (a) indriya-pratyakṣa
(śrotra, cakṣur, ghrāṇa, jihvā, sparśa), (b) no-indriya-pratyakṣa: (avadhi, manaḥ-paryāya, kevala), (2) anumāna: (a) pārvavat, (b) seṣavat, (c) sadharmya, (3) aupamya [...], (4) agama [...], (II) darśana: (1) cakṣur-darśana, (2) acakṣur-darśana, avadhidarśana, kevala-darśana. 57
kevala-nāṇa-pajjavāṇam mati-anṇāṇa-pajjavāṇaṁ suta-anṇāṇa-pajjavāṇaṁ vibhamga-nāṇa-pajjavā-nam cakkhu-damsana-pajjavāṇam acakkhu-damsana-pajjavāṇamohi-damsaṇa-pajjavā-ṇaṁ kevala-daṁsaṇa-pajjavāṇaṁ uvaogaṁ gacchati, upayoga-lakkane nam jive...
55. Than 60 (p. 14-15): duvihe nane pannatte, tam jahā - pacchakkhe ceva parokkhe ceva. paccakkhe name duvihe pannatte, tam jahā - kevala-nāne ceva nokevala-nāne ceva. kevala-ṇāne duvihe pannate, ...no-kevala-nāne duvihe pannate, tam jahā - ohi-ṇāne ceva maṇapajjava-ṇāne ceva. ...maṇapajjava-nāne duvihe pannate, ...parokkha-ṇāne duvihe pannate, tam jahā - abhinibohiya-ṇāne ceva suya-nāņe ceva. abhinibohiya-ṇāne duvihe pannate, tam jahā - suta-nissite ceva asuta-nissite ceva. suta-nissite duvihe pannate, tam jaha - atthoggahe ceva vamjanoggahe ceva. asuya-nissite vi emeva. suya-ṇāne duvihe pannate, tam jahā - amga-paviṭṭhe ceva amga-bahire ceva....
56. Viy 5.4.26[3] (vol.1, p. 201.1-2); pamāne cauvvihe pannatte, tam jahā - paccakkhe, aṇumāņe, ovamme, āgame. Thāṇ 336 (p. 149): ahavā heu cauvvihe pannatte, tam jahā - paccakkhe aṇumāņe ovamme āgame.
57. AnD 435-471 (p. 173-179): [435] se kim tam jiva-guna-ppamāṇe? ...tam jahā - nāṇa-guna-ppamāne damsaṇa-guna-ppamāṇe caritta-guna-ppamāṇe. [436] se kim tam nāṇa-guna-ppamane? ...tam jahā - paccakkhe aņumāņe ovamme āgame. [437] se kim tam paccakkhe?...tam jaha - imdiya-paccakkhe, noimmdiya-paccakkhe ya. [438] se kim tam imdiya-paccakkhe? ...tam jahā - soimdiya-paccakkhe cakkhurimdiya-paccakkhe ghanimdiya-paccakkhe jibbhimdiya-paccakkhe phāsimdiya-paccakkhe. se tam imdiya-paccakkhe. [439] se kim tam noimdiya-paccakkhe? ...tam jahā ohi-ṇāṇa-paccakkhe maṇapajjava-ṇāṇa-paccakkhe kevala-nana-paccakkhe. se tam noimdiya-paccakkhe. se tam pacchakke. [440] se kim tam anumane?...taṁ jahā puvvam sesavam dittha-sahammavam... [471] se kim tam damsana-guna-ppamāṇe?
-
tam jahā - cakkhu-damsaṇa-guna-ppamāṇe acakkhu-daṁsana-guna-ppamane ohidamsaṇa-guna-ppamāṇe kevala-damaṇa-guṇa-ppamāṇe ceva. cakkhu-damsane cakkhu-damsanissa ghaḍa-pada-kaḍa-raghādievsu davvesu, acakkhu-damsane acakkhu-damsaṇissa āya-bhāve, ohi-damsa-nam-damsanissa davva-davvehim savvapajjavehi ya. se tam damsaṇa-guna-ppamāṇe.

Page #20

48
Piotr Balcerowicz
Model V: jñāna: (1) pratyakṣa: (1) indriya-pratyakṣa (five sensory kinds), (2) no-indriya-pratyakṣa: (a) avadhi-pratyakṣa, (b) manaḥparyaya-pratyakṣa, (c) kevala-pratyakṣa, (II) parokṣa 58: (1) abhinibodhika: (a) śruta-niḥśrita (with four stages: avagraha, ihā, apāya, dhāraṇā), (b) aśruta-niḥśrita, (2) śruta 59. This classification is in so far inconsistent as it virtually classifies cognitive processes of mati-jñāna both as indriya-pratyakṣa and parokṣa. Besides, this classification does not know the notion of upayoga 60.
Model VI: (A) pramāņa (= jñāna): (I) parokṣa: (1) mati with four stages: (a) avagraha, (b) ihā, (c) apaya, (d) dhāraṇā, (2) śruta, (II) pratyakṣa: (1) avadhi, (2) manaḥ-paryaya, (3) kevala; (B) upayoga: (I) jñānôpayoga (sākāra): (1) mati-jñāna, (2) śrutajñāna, (3) avadhi-jñāna, (4) manaḥ-paryāya-jñāna, (5) kevalajñāna, (6) maty-ajñāna, (7) śrutâjñāna, (8) vibhanga-jñāna, (II) darśanôpayoga (anākara): (1) cakṣur-darśana, (2) acakṣurdarśana, (3) avadhi-darśana, (4) kevala-darśana. 61 This model offers two very similar sub-models (pramāņa and jñānôpayoga) that partly overlap.
58. The indirect cognition (parokṣa) is in fact said to be homogenous, its two subdivisions (abhinibodhika and śruta) being in fact identical (NamS 44).
59. NamS 9-61 (p. 9-27) expands the division of NamS 8: [9] tam samāsao duviham pannattam. tam jahā - paccakkham ca parokkhaṁ ca. [10] ...paccakkham duviham pannattaṁ. tam jahā - iṁdiya-paccakkham ca no-imdiya-paccakkham ca. [11]...imdiya-paccakkham pamcaviham pannattam. tam jahā - soimdiya-paccakkham cakkhimdiya-paccakkham ghanimdiya-paccakkham rasanomiya-paccakkham phāsimdiya-paccakkham. [12] ... no-imdiya-paccakkham tiviham pannattam. tam jaha - ohi-paccakkham manapajjava-paccakkham kevala-paccakkham... [43] se kim parokkham? parokkham duviham pannattam. tam jahā - abhinibohiya-ṇāṇa-pasrokkham ca suya-ṇāṇa-parokkham ca. [44] jaha 'bhinibohiya-ṇānam tattha suyanāṇam, jattha suya-nāṇam tattha 'bhinibohiya-ṇāṇam. do vi eyaim annamannam anugayaim taha vi puna ettha 'yariyā ṇāṇattaṁ - abhinibujjhai tti ābhinibohiyam, sunatîti sutam. 'mati-puvvam suyam, na mati suya-puvviya'. [46] se kim tam abhinibohiya-nāṇam? abhinibohiya-nanam duviham pannattam. tam jahā – suya-nissiyam ca asuya-nissiyam ca... [48] se kim tam suya-nissiyam mati-nanam? suya-nissiyam mati-nāņam catuviham pannattam. tam jahā - uggahe ihā avāe dhāraṇā. [49] ...uggahe duviham pannattam. tam jahā - atthoggahe ya vaṁjanoggahe ya... [NamS 61 ff. (p. 27ff) discusses at length the divisions of śruta-jñāna and the Canon].
60. The term upayoga occurs only once in 47 [ga. 66] (p. 21) in a different context. Similarly, the term darśana occurs only once in 71[1] (p. 28) in the context of divisions of the Canon (damsana-dhara).
61. TS/TBh 1.9-12,2.8-9.

Page #21

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
49
The idea of pratyaksa as the direct cognition occurs in Model II, Model IV, Model V and Model VI 62, it is still absent from Model I, while pratyakṣa in Model III corresponds to sensory perception of other systems.
Excluding Model III as genetically non-Jaina, Jaina divisions of pramāņa in the Agamic tradition could hardly accommodate the panIndian idea of pratyakṣa as sensory perception directly, perhaps with the sole exception of Model V and Model VI. The most frequently recurring point in these models (with the exception of Model I and III) is that avadhi, manah-paryāya and kevala are classified as direct types of cognition (pratyakşa), whereas śruta is catalogued under indirect types of cognition (paroksa). The most controversial - and in our analysis crucial - issue is, therefore, the proper assignment of the sensuous cognition (mati-jñāna). . Clearly, such a diversity of opinions points to the fact that there was no unanimity among Jaina thinkers and the author of STP could have either subscribed to any one of the Canonical models or could have devised his own model.
4.2. Since, as we have seen before, NA follows the general Indian epistemological tradition as regards the nature of pratyaksa as different from more advanced acts of conceptualisation, and the notion of the sensuous cognition (mati-iñāna) - with its four traditio stages, viz. sensation (avagraha), speculation (ihā), perceptual judgement (apāya) and retention (dhāranā) - does not fit into the framework of NA 63. If we analyse all the six models, the idea of the four
62. Kundakunda's standpoint follows the same lines, cf. PSā 1.57-58:
para-davvaṁ te akkhā neva sahāvo tti appano bhanidā / uvalddham tehi kadham paccakkham appaņo hodi // jar parado vinnānam tamṁ tu parokkham ti bhanidam atthesu /
jadi kevalena ņādam havadi hi jīvena paccakkham // - "These perceiving organs are (made) of different substance. Under no circumstances can they be said to be the essential nature of the cognitive subject (soul). How could possibly what has been grasped by them become direct cognition for the cognitive subject (soul)? As regards objects, what is the discernment through other (means (i.e. senses)] is called indirect cognition; for when cognition arises through the living element (soul) completely is direct cognition.' See also PSā 1.54.
63. See Model 2 in BALCEROWICZ (2005, $ 5).

Page #22

50
Piotr Balcerowicz
stages of the sensuous cognition (avagraha-ihā-apāya-dhāranā) occurs in all of them, with the exception of Model III. What these models have in common is their treatment of the four stages (avagraha-ihā-apāya-dhāraṇā) as indirect (parokşa). And it is not surprising, since only the first stage of sensation (avagraha) could correspond to the perception (pratyakşa) of other philosophical systems. The remaining stages – speculation (ihā), perceptual judgement (apāya) and retention (dhāranā) - involve the process of conceptualisation, rationalising, memorising, etc. Even if one admits that the notion of pratyakșa does not necessarily have to be taken strictly in the Dinnāgian-Dharmakirtian sense of being free of conceptualisation (kalpanâpodha) but it may allow, as the Jainas would have it, for pronounced conceptual component (savikalpaka), beside the non-conceptual variety (nirvikalpaka), nevertheless the set ihā-apāya-dhāraņā entails deep and extensive involvement of exclusively mental activities. Furthermore, a clear indication of the truly indirect, i.e. conceptual character of the sensuous cognition is a series of expressions synonymous to mati-jñāna from TS 1.13: matiḥ smrtiḥ samjñā cintâbhinibodha ity anarthantaram. Similar lists of synonyms can be found in other Jaina works 64.
4.3. In STP, Siddhasena Divākara clearly admitted of the sensuous cognition (mati-jñāna, ābhinibodhika-jñāna) himself, since he speaks of it explicitly, mentioning its name four times (STP 2.6, 23, 27, 32). Was this mati-jñāna for Divākara just the same kind of cognition as it was to the Jaina tradition prior to him only nominally, whereas in reality he took it to correspond to the sensory perception (pratyakşa) of non-Jaina traditions and considered mati-jñāna to be in fact 'direct' in the sense of direct sensory grasp? I am convinced that such a supposition would neglect the available internal textual evidence.
In fact, Divākara definitely subscribed to the notion of the four stages of the sensuous cognition (avagraha-ihā-apāya-dhāranā), since he refers to the first stage avagraha twice, namely in STP 2.21 65: [The
64. E.g. VABh 396: ābhinibohiya, ihā, apoha, vimaṁsā, maggaņā, gavesanā, saữna, sai, mai, paññã. 65. STP 2.21: damsaņam oggahametta 'ghado'tti ņivvannaņā havai ņāņa /
jaha ettha kevalāna vi visesanam ettiyam ceva //

Page #23

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
51
prima facie position to be refuted:) 'Insight is nothing but sensation, because it designates “[this is) a pot”, [hence it] becomes [the sensuous cognition. Just like (sensation), in the same manner, the difference between both the absolute [cognition and insight] is this much only...;' and in STP 2.23-24 66: [Rejoinder:) 'If you maintain that insight is nothing but [ocular 67] sensation, [or] a qualified cognition, (then), if it were so, it (would) follow that insight is nothing but the sensuous cognition, and such would necessarily be [the case) with the insight derived through the remaining sense organs. But this is not correct. [The opponent argues:) "If in [the case of] these [remaining senses] only cognition is understood , in the very same way in the case of] eyes (only cognition should be understood)” 69'
Avagraha is a well-known technical term and it unequivocally implies the acceptance of the remaining three members of the sen. suous cognition, viz. ihā, apāya and dhāraṇā. Such being the case, the use of the term avagraha not only indicated that Divākara subscribed to the tradition that subdivided the sensuous cognition (mati-jñāna) into the four steps, but he must have classified the sensuous cognition as indirect in view of the indirect, i.e. mental, or conceptual character of the triad: speculation (ihā)- perceptual judgement (apāya) - retention (dhāraņā). This is confirmed also by two rejoinders above (STP 2.21,23). quoting the position of a hypothetical opponent. The oppo
66. STP 2.23-24: jai oggahamettaṁ daṁsanaṁ ti mannasi visesiam nānam /
mai-ņāņam eva daṁsaņam evam sai hoi nipphannaṁ // evaṁ sesiṁdiya-daṁsanammi niyameņa hoi na ya juttaṁ /
aha tattha ņāņamettaṁ gheppai cakkhummi vi taheva // 67. In view of the phrase sesimdiya (sesendriya) in STP 2.24, the sensation here must refer to cakṣur-avagraha (*cakkhuggaha =ālocana).
68. The form gheppai corresponds to *Vghrp / *ghrpsyate, a root paralel to Vgrbh (see: PISCHEL (1981: 8 212, p. 182, $ 534, p. 434 and 8 548, p. 441). I would be inclined, nonetheless; to relate it to Vkhyā / khyāpyate ('to predicate'; cf. NA 19c: khyāpyate yatra drstānte).
69. The idea of the opponent is that if one accepts that there should be only respective sensory cognition, e.g., olfactory cognition (ghrāņa-jñāna), and the idea of an olfactory insight (ghrāņa-darśana) be rejected, the same rule should be applied to the sense of sight: one should accept only ocular cognition (cakṣur-jñāna) and reject the idea of ocular insight (cakşur-darśana). In the preceding section (STP 2.20) Divākara - following tradition - recognises cakşur-darśana as one of four subdivisions of darśana.

Page #24

Piotr Balcerowicz
nent's criticism jointly points to a seeming affinity between Jaina avagraha and darśana that could be understood to tally with non-Jaina pratyaksa. Divākara's reply shows his intention to prove that darśana is something more and above avagraha 70.
5. A separate question is the exact semantic relationship between the term daṁsaņa occurring in STP in a variety of meanings and the term pratyakșa employed consistently in NA, i.e. whether there is any sense in which the daṁsaņa of STP corresponds to the pratyakṣa of NA.
5.1. Indeed on some occasions one might have an impression that the daṁsaņa of STP is used in the sense of perception as such an operation of the cognitive apparatus (not necessarily sense organs in the case of STP) that provides some awareness of an external object, either indistinct (vyañjita) or non-conceptual (nirvikalpaka). This happens for instance in the case of STP 2.25 71: ‘The cognition as regards an object not [directly) touched [by senses) and beyond the province [of senses) becomes insight (perception?), with the exception of such [cognition that pertains) to the provinces of the future and the past through the [inferential?) sign.'
Similarly, the verses of STP 2.27-29 72: “[27] In (case of] a conditioned person (in the state of bondage) the comprehension of objects is occasioned by the sensuous cognition and testimony; there is no insight in any one of them; wherefrom (should there be) insight [in them]? [28] Since objects cognised through testimony are not amenable to grasping [them] directly, therefore the word “insight” (perception?) does not apply to the cognition through testimony at all.
70. More on the above passage (STP 2.21-24), comp. BALCEROWICZ (20016: 365-366, $ 6.2.2). 71. STP 2.25: nānam aputthe avisae ya atthammi daṁsanam hoi/
mottūna lingao jam anāgayāiya-visaesu // 72. STP 2.27-29: mai-suya-nāna-nimitto chaumatthe hoi attha-uvalambho/
egayarammi vi tesim ņa daṁsaņam daṁsanam kutto? // jam paccakkha-gahanam na inti suya-ņāņa-sammiyā atthā / tamhā daṁsaņa-saddo na hoi sayale vi suya-nāne // jam aputthā bhāvā ohi-nnānāssa homti paccakkhā / tamhā ohi-nnāņo daṁsaņa-saddo vi uvayutto //

Page #25

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
53
[29] Since entities not [directly) touched [by senses) become direct[ly cognisable] for the cognition through telaesthesia, therefore the word “insight” is (correctly] employed with regard to the cognition through telaesthesia.'
Of course, the verses explicate in the first place the three - out of four – subdivisions of darśana (insight / conation) as a subtype of the cognitive faculties (upayoga), viz. caksur-darśana, acaksur-darśana and avadhi-darśana. The issue of the proper interpretation of the term daṁsana in these contexts would, however, require a separate detailed analysis.
5.2. What concerns me here is the peculiarity of expression in STP 2.28 and its comparison with even more peculiar statement of NA 12 73: 'And such an utterance that demonstrates an object recognised through perception is called perception, because it is the external factor for the representation. This startling statement of Siddhasena Mahāmati should be viewed in the context of his thesis of parârthapratyaksa and his attempt to prove that the idea of efficacy for others (pārārthya) and efficacy for oneself (svārthya) is applicable both to perception (pratyakşa) and to inference (paroksa). What is important, NA explicitly accepts the idea that perception can be directly generated in other people also through verbal communication 74. In other words, objects are amenable to direct comprehension also on the verbal level and verbal statements can be classified as perception under special conditions.
This stands in contradiction with the statement of STP 2.28 quoted above that 'objects cognised through testimony are not amenable to grasping [them] directly' (paccakkha-ggahaņań na inti suyanāna-sammiyā atthā). And it is STP 2.16ab (pannavanijjā bhāvā samatta-suya-nana-damsanā-visao /) that links 'communicable entities' (prajñāpanīyā bhāvāḥ) with testimony (śruta).
73. NA 12: pratyakşa-pratipannârtha-pratipādi ca yad vacaḥ/
pratyakşam pratibhāsasya nimittatvāt tad ucyate // 74. Cf. NAV ad loc: pratibhäsasya nimittatvāt pratipādya-pratyakşa-prakāśa-hetutvād upacāreņôcyata ity arthah. - "because it is the external factor for the representation", which means that can utterance) is called metaphorically [perception] because it is the cause of revealing (an object] through perception to a person] to be taught.'

Page #26

54
Piotr Balcerowicz
6. It is well known that the author of NA is innovative in his attitude with respect to Jaina tradition and seems to go against the Canon, e.g. in his new division of the pramāṇas and in his new definition of pramāņa, even though everyone knows what it is' 75. Nowhere in the whole text of NA does Siddhasena Mahāmati takes recourse to tradition or the scriptures to corroborate his statements. He solely relies on the power of logic and argument.
This attitude largely differs from the 'Agamic approach typical of STP, viz. the attempt to remain faithful to the Jaina tradition and to the authority are the Āgamas, even though the 'Agamic tradition' is not always unanimous 76.
A good example of such an 'Agamic' attitude is provided by STP 3.10-12. In the first step, Siddhasena Divākara introduces the pārvapakṣa position: 'As a matter of fact, two viewpoints were distinguished by the Lord (the Jina), viz. substantial viewpoint (dravyarthika) and modal viewpoint (paryāyārthika); but since there is additionally the "quality" particular, [hence] also qualitative view-point (gunārthika) should be adopted 77.' This is rather a logical and consistent conclusion, if one accepts that the main divisions of view-points (naya) are derived from the main ontological aspects: the substance and its modes. Since there is also quality (guna) as the third aspect underlying the being, and it is even more fundamental ontologically, one would expect a separate viewpoint corresponding to it. Siddhasena Divākara's rejoinder to this argument is not based directly on logic or reasoning but takes first of all recourse to the stand of the
75. NA 2-3: prasiddhāni pramāṇāni vyavahāraś ca tat-krtah/
pramāna-laksanasyöktau jñāyate na prayojanam // prasiddhānāṁ pramāṇānām laksanôktau prayojanam/
tad-vyāmoha-nivettih syād vyāmādha-manasām iha // - '[The prima facie position to be refuted:) Cognitive criteria are well-known and everyday practice is accomplished by them; [therefore) no purpose is known for stating the definition of cognitive criterion. [Rejoinder:) The purpose for stating the definition of cognitive criteria, (although they are] well-known, should be the eradication of disorientation concerning that (definition of cognitive criterion) in the case of the disoriented-minded here.
76. For instance comp. above the lack of unanimity as regards the divisions of the cognitive faculties (upayoga). 77. STP 3.10: do una nayā bahavayā davvathiya-pajjavattiyā niyayā /
etto ya guna-visese gunathiya-nao vi jujjasto //

Page #27

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
55
scriptures and the authority: 'But because what the Venerable One (the Jina) distinguished and expounded to Gautama and other [disciples) in those sūtras is the notion of mode (paryāya), hence (we have] modal (viewpoints). Even though it is understood that "mode" and "quality" in many ways have the same meaning, nevertheless (we) do not say quality (viz., qualitative viewpoint), because [there is) the designation “modal viewpoint” 78.
7.1. Now, I would like to devote some space to recent arguments brought forward by Bansidhar BHATT (2000), who asserts that: 'we arrive at a certain conclusion that the author of NV (= Nyāyâvatara - P.B.) lived after Prajñākara / Dharmottara (both: 700-750 A.D.) - terminus a quo...' (p. 77). This 'certain conclusion' has, however, a
asis. In the first place, BHATT practically treats NB and · NBT jointly throughout his paper, from the very moment these two works are first mentioned (p. 70), and whenever he refers to the Nyāya-bindu, he consistently writes ‘NB/NBT', without making much distinction as regards the contents of NB and NBȚ. It is not surprising that via this petitio principii device one has to finally arrive at the con clusion that Siddhasena Mahāmati lived after Dharmottara.
Secondly, from the fact that NA follows the Buddhist texts on logic, e.g. Pramānasamuccaya, Nyāyapraveśa 79 etc. of Dignāga..., Pramāņavārttika, Nyāyabindu (NB) of Dharmakirti..., Prajñākaraguptabhyāsya..., but more often Dharmottara's commentary on Nyāyabindu (NBC)...', BHATT (2000: 71) comes to conclusion that NA must be posterior to all these works! Clearly, Prajñākaragupta in his PVA or Dharmottara in his NBT had to follow the pattern of the works which they decided to comment on, viz. Dharmakirti's PV and NB respectively. But there is nothing that would logically compel us
78. STP 3.11-12: jam ca puna arihayā tesu tesu suttesu goyamāiņam/
pajjava-sannā niyayā vāgariyā tena pajjāyā // parigamanam pajjāo anega-karanam guna tti tullatthā /
taha vi na guna tti bannai pajjava-naya-desaņā jamhā // 79. Wrongly ascribed by BHATT (2000: 71, 72 n. 16) to Dinnāga, instead to Sankarasvāmin, see MIRONOV (1927), Tucci, DHRUVA (1930: V-XIII), FRAUWALLNER (1961: 140), cf. also HATTORI (1968: 4) and STEINKELLNER-MUCH (1995: 16).

Page #28

56
Piotr Balcerowicz
to assume that NA, which did follow the pattern of NB 80, is posterior to NBT, which follows the identical structure.
His third and final argument to prove that Siddhasena was posterior to Prajñākaragupta, to which he devotes a few more lines, is based on the rather curious idea of 'perception for others' (parârthapratyakşa), most probably the innovation of Siddhasena Mahāmati himself and clearly modelled on Dinnāga's and Dharmakirti's distinction of two kinds of inference, but later on dropped by most, though not all 81, thinkers of the Jaina tradition. BHATT (2000: 73) maintains that: 'A new variety of pratyaksa, viz. parārtha pratyaksa in Indian logic is introduced by the NV [= Nyāyâvatara - P.B.] (v. 11). This represents a clear influence of Prajñākara who provided the NV-author with an idea of parārtha pratyaksa (cp. yady anumā-notpādanād vacanam anumānam, pratyakşotpādanāt, v. 1.: vacanam api parārthaṁ pratyakşam bhavet. PVB [= Pramāņa-vārttikälankāra P.B.) 3/4.1.1., p. 476; cp. also: NV v. 10cd with this PVB-statement).'
7.2. The first part of the argument about Siddhasena's authorship of the idea of parârtha-pratyaksa is indeed sound 82, however, its latter part is a result of some confusion. When we read the whole passage of Prajñākaragupta from which BHATT extracted just a line (vide infra, bolded), we realise that the quotation referred to by Bhatt is, in fact, an objection against Prajñākaragupta's position (!), which he subsequently refutes. In the excerpt, Prajñākaragupta explains why it is possible that genuine inferential process (svārthānumāna), which has been essentially carried out by the cogniser himself who witnessed the events that are the basis for generalisation, can be demonstrated ver bally to another person and evoke a similar cognition in the hearer; hence, even though the proof formula is, strictly speaking, merely a verbal statement, it is metaphorically called inference for others (parârthânumāna), because it generates the same inferential conclu sion in the hearer as it did in the speaker. Prajñākaragupta insists that
80. See: BALCEROWICZ (2001a: xii ff.).
81. E.g. by Vadideva-sūri in PNTAA 3.26 or by Gunaratna-sūri in TRD, p. 223 ff., see BHATT (2000: 76, nn. 27, 28).
82. See above $ 5.2; cf. also BALCEROWICZ (2001b: 367-368, § 8)..

Page #29

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
57
this metaphorical transference, does not, however, apply to perception. The complete passage of PVA, p. 467.15-26, reads as follows:
“However, the comprehension of the inferable property which has become the subject of the debate (is possible) by means of the demonstration of the triple-formed inferential sign, therefore [its) statement (sc. verbal demonstration of a proof formula) (which is inference for others (parârthânumāna)] is not pointless. Because inference is produced on the account of this (verbal statement], also it is inference due to metaphorical transference of the genuinely inferential character of the inference for oneself (svârthânumāna)].
[Objection:) “If the statement, (which is called the inference for others, is) inference because it produces inference, then (a verbal statement] is also perception for others, because it produces perception". No such quadruplet (is possible, viz. inference for oneself (svârthânumāna), inference for others (parârthânumāna), perception for oneself (svârtha-pratyaksa) and perception for others (parârtha-pratyakşa)].
[2] Perception never arises with respect to anything in the same way (sc. on account of the statement) as inference is produced: when one remembers the relation (sc. invariable concomitance) on account of the statement (of the logical reason).
[3] When one remembers the triple-formed inferential sign, there necessarily arises inference. However, when a statement (is formulated by one person which describes) only an object that one has cognised oneself, no one (else except for the speaker) cognises (this object] through perception (on account of the statement).
[In other words,] no perception arises on account of merely a statement.
[Objection:] “[When one person exclaims:] «Look! A deer is running!», (the other person) looks [in that direction), (and in this way, there arises perception (in that person)”.
No, also in this case (on account of the statement) there immediately arises inference (and not perception). For it is as follows:
[4] [A person) is enjoined to direct one's sight towards this object as follows: "Look!" (in the sense that:] "[This object which] I have (just) cognised (myself), has been demonstrated in the context”.
(So,] this is a statement of injunction (formulated by the speaker): "Direct your sight there!" in the sense of taking a look at the deer. And [further): "Just as I have directed my sight (towards this deer), so you, too, (look there]”. Then, when [the other person) is directing (one's) sight towards (that deer) in this way, there (takes place) the operation of logical reasons: remem

Page #30

58
Piotr Balcerowicz
bering (the logical ground) in this way, (the person) proceeds (with the intention to see the deer). So this is nothing but inference. Therefore, having considered, through inference, that there is a possibility of perception (sc. that one might be able to see a deer), one proceeds (sc. sees the deer). '83, 84
As we can see, there can be no doubt that Prajñākaragupta rejects even a slightest possibility of 'perception for others' 85. In his opinion, what the opponent - doubtlessly a Jaina thinker - considers to be perception for others is an inferential process, triggered by the verbal instruction: 'Look! A deer is running', which can be summarised as
[2]
[3]
83. PVA, p. 467.15-26: tri-rūpa-linga-prakāśana-dvāreņa tu vivādāspadibhātânu-meya-pratipattir iti na vacanasya vyarthatā. tato 'numānam udetîti tad apy anumānam upacārāt. yady anumanôtpadanád vacanam anumānam pratyakşôtpadanāt pratyakşam api parârthaí bhavet. nêdam caturasram.
yathā grhita-sambandha-smarane vacanāt sati/ anumānôdayas tadvan na pratyaksôdayah kvacit // tri-rūpa-linga-smarane niyamenânumódayah/ sva-pratitârtha-mātrasya vacane 'dhyakşavin na tu //.
na vacana-mātrād adhyakşam parasyôdeti. nanu "paśya mrgo dhāvatîti" drśyate darśanôdayaḥ. na, tatrâpy anumānasyânantaratvāt. tathā hi - [4]
tad-arthônmukhatāyām sa pasyêty evam niyujyate / mayā pratitam etat ca samarthyāt pratipăditam //
abhimukhi-bhava mrga-darśana iti niyoga-vacanam etat. abhimukhi-bhāvas ca yathā mama tathā tavāpi. tata evam abhimukhi-bhavane hetūnām vyāpāra iti smaran pravartata ity anumānam eva. tato 'numānāt pratyaksasambhavam alocya pravartate.
84. The same idea is elaborated also by Durveka Miśra in DhPr, p. 89: nanu ca parârthânumānotpadaka-vākyavad asti kiñcid vākyam yat para-pratyaksôpayogi. yathā "eşa kalabho dhāvati” vākyam. atah parârthânumānavat parârtham pratyakşam kim na vyutpadyata iti? atrôcyate - parokşârtha-pratipatter yā sāmagri - lingasya pakşa-dharmatā sādhya-vyāptiś ca - tad-ākhyānād vākyam upacarataḥ parârthânu-mānam ucyate. na tu tatra kathaṁcid anga-bhāva-mātrena, svāsthyâder api tathā prasangāt. idam punaḥ 'ayam kalabhaḥ' ity-ādi-vākyam na pratyakşôtpatter yā sāma-grîndriyâlokâdi tad-abhidhānāt tan-nimittam bhavat tathā vyapadeśam aśnute yena vyutpădyatām apy aśnuvita, kim tarhi? kasyacid didyksā-mātra-jananena. yathā kathancit para-pratyakşôtpattāv anga-bhāva-mātrena tādrāpye netrôtsave vastuni sannihite 'pi kathamcit parān-mukhasya parena yad abhimukhi-karanam sirasas tad api vacanâtmakam parârtha-pratyakşam vyutpădayitur vyutpădyam āpadyeta. etac ca kaḥ svasthâtmā manasi niveśayet. kim ca bhavatu tathā-vidham vacanam parârtham pratyakşam. kim nas chinnam?...
85. See also Manorathanandin in PVV ad PVI 1.3bc (p. 4.2-4): artha-kriyānirbhāsam tu pratyakşam svata evârtha-kriyânubhavâtmakaṁ na tatra parârthakriyâpekșyata iti tad api svato niścita-prāmānya.

Page #31

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
59
follows: 'He has seen a deer; if I look there the way he does, I will also see it.
Instead of finding in the above passage 'a clear influence of Prajñākara who provided the NV-author with an idea of 'parārtha pratyakşa', we see just the opposite: it is Prajñākaragupta who criticises the idea formulated in NA. This criticism cannot prove that the person whom Prajñākaragupta had in mind was Siddhasena Mahāmati and his NA. All we can with certainty say is that the criticism is directed against the same idea which we find in NA, and which may have been also maintained by some other thinker(s) who might have directly provoked Prajñākaragupta censure. We cannot, however, claim with absolute certitude - but with a high degree of probability - that it was indeed Siddhasena Mahāmati who was the inventor of the notion of parârtha-pratyakșa.
Additional corroboration for the above hypothesis is provided by Siddharsi-ganin, the commentator on NA. In his Nyāyâvatāra-vivrti, he refers to Prajñākaragupta critical remarks, and reasserts the genuineness of parârtha-pratyakșa:
'If perception were conjectured to be superior, on account of (the argument that) in certain cases the indirect cognition proceeds by the force of the relation which has [first] been grasped by perception, (then) the superiority of the indirect cognition would follow immediately on account of (the argument that) it is (also) an empirical fact that (1) a perception the province of which is a deer (that is observed] due to a sudden movement of the neck by the force of the speech element (sc. expression): “Look! A deer is running!” etc., [or] similarly [2] la perception) the domain of which is a forest or a temple, etc. (that are recognised) either due to the recollection (of the forest or the temple, etc.,) as such or due to grasping the linguistic convention [relating the word "temple" and the object temple) as such with curiosity, etc., to see objects not seen before, is preceded by the indirect cognition (in both cases). 86
86. NAV 1.8, p. 341: kvacit pratyaksa-parigrhita-sambandha-balāt paroksam pravartata iti pratyakşasya jyesthatva-kalpane "paśya mrgo dhävatîty” -adi-sabdabalāt krkâtikā-motana-dvāreņa mrga-vișayam, tathā smaraņāt sanketa-grahaņād vâpūrvâpūrvārtha-darśana-kutāhalâdinā vana-deva-kulâdi-gocaram parokșa-pārvam pratyakşam drstam iti parokşasya jyeșthatâsajyeta. A rejoinder to Siddharşi-ganin's position seem to have been formulated subsequently by Durveka Miśra, see n. 84.

Page #32

Piotr Balcerowicz
The injunction within the bolded phrase is exactly the same as the one found in Prajñākaragupta, and the detailed description of the whole process which finally leads to “perception for others’, triggered by a verbal statement, is to explicate the perceptual nature of this process, as against Prajñākaragupta's position.
The fact that it is Siddharsi-ganin who, in his commentary on NA, defends the idea propounded by Siddhasena Mahāmati in NA may lend some additional support to the claim that it was indeed the Nyāyâvatāra which Prajñākaragupta had in mind. And that may be taken as a corroboration of the hypothesis that the time of composition of the Pramāna-vārttikâlankāra is terminus ad quem for Siddhasena Mahāmati. As long as we do not prove that it was indeed Siddhasena Mahāmati who was the first to introduce the idea of parârthânumāna this remains only a very probable postulate. This does not have much bearing on the dating of Siddhasena Mahāmati in view of the fact that the date of Prajñākaragupta is to be assigned to the similar time, i.e. circa 800 87, as the date of Haribhadra-sūri 88, which is accepted as the upper limit for that Nyāyâvatāra.
8. Is has been frequently suggested that the structure of NA is not intact, or that some of the verses of the text are interpolations. As a matter of fact, if we analyse all secondary literature written on NA, we see that reservations have been occasionally expressed about the genuineness of perhaps half of all the verses of NA! Recently a similar suggestion has been voiced also by BHATT (2000), who drew our attention to a number of verses which may imply, in his opinion, that they were not originally an integral part of NA. On subsequent pages I shall try to evaluate such arguments and discuss the place of the spurious verses in the general structure of NA.
8.1. We should approach with special caution, as BHATT (2000: 73) warns us, the verses of NA 8-9: 'However, the original text of the NV does not appear to be intact. NV v. 8-9 describing śābda pramāņa have almost similar expressions, e.g. drsțeștāvyāhata- (v. 8a); adssteșta-virodha- (v. 9b); tattva-grāhita- (v. 8c); tattvopadeśa- (v. 9c)
87. See STEINKELLNER-MUCH (1995: 74). 88. Cf. BALCEROWICZ (2001a: xxxiv).

Page #33

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
61
etc. It is certain that v. 9, being confined furthermore to the sāstra and not referring to the sābda in general, is spurious in general, is spurious in the logical treatise, viz. the NV. Moreover, it is not in agreement with the scheme the NV-author certainly had in mind, viz. to encompass precisely the subject matter of logic in a small compendium.' As a matter of fact, in expressing his reservations about NA 8, BHATT revives the arguments brougt forward, among others, by P.L. Vaidya in his introduction to NA3 (p. xiii-xiv, xxviii) and MUKHTAR (1956).
The repetition of a phrase is hardly a conclusive ground to question the authenticity of any verse in NA. On the contrary, we can see that some verses in such a relatively short treatise reveal a repetitive character, just to compare three cases of pairs of adjoining verses:
2a, d: prasiddhāni pramānāni ... jñāyate na prayojanam and . 3ab: prasiddhānām pramāṇānām laksaņôktau prayojanam
22a: anyathânupapannatvas and 23b: yo 'nyathaîvôpapadyate 24ab: sādharmyenâtra drstânta-dosā nyāya-vid-iritäh 25ab: vaidharmyenâtra drstânta-dosā nyāya-vid-iritäh
By applying the suggested method to discard all verses that contain repetitions one would end with perhaps with a handful of verses The main point, as a matter of fact, concerns the occurrence of sastra, which might seem 'spurious in the logical treatise'. A closer analysis will reveal that the reference to śāstra /āpta not only is relevant to the discussion of epistemological issues, especially in the context of debate with Dinnāga and Dharmakirti, but also it is absolutely essential in the structure of NA, which offers a new model of pramānas. The sequence of the verses runs as follows:
NA 889: definition of verbal cognition (śābda),
NA 990: definition of a reliable source of verbal cognition (śāstram + āpta /āptôpajña),
89. See n. 30. 90. NA 9: āptôpajñam anullanghyam adsstêsta-virodhakam /
tattvôpadeśa-krt sārvam śāstram kāpatha-ghattanam // - 'Authoritative treatise is that which has been discerned by an authoritative per

Page #34

Piotr Balcerowicz
NA 109: definition of parârthânumāna as a special kind of verbal cognition, in which the reliable source is what the speaker himself experienced
Thus, Siddhasena's idea is to prove that there are only two pramāṇas: pratyakșa and parokșa, the latter comprising all categories of cognitions that are not pratyaksa. In this way, Siddhasena emulates Dinnāga's manoeuvre who first (PS 1.3) describes the nature of pratyakșa as being free from conceptual construction (kalpanâpodha) and then subsumes all other kinds of cognition under one general heading of anumāna, including inferences for oneself and for the others (PS 2.1ab and PS 3.1ab) as well as verbal cognition (śābda, PS 5.1: na pramāņântarań śābdam anumānāt), along with testimonial cognition derived from an authority (āpta, PS 2.5: āptavādâvi-samvāda-sāmānyād anumānatā). In the case of NA, all kinds of cognition other than pratyaksa, which is defined as first, are subsumed under the heading of parokşa. Just as Dinnāga specifically singled out sābda and emphasised that also verbal cognitions, including those based on testimony of authority, are comprised under anumāna, Siddhasena Mahāmati holds that verbal cognition (śābda, NA 8,9) and inference (NA 10), with its two subtypes 'for oneself' and 'for others' (svārtha-pratyaksa and parārtha-pratyakşa, NA 11), are likewise different varieties of parokșa 92. Therefore the place of NA 8 and 9 is justified in the whole scheme of cognitions. In other words, not only does Siddhasena do precisely what Dinnāga did when he comprised śābda / āpta-vāda under anumāna, but even terms used in both cases are almost identical!
son, which is not negligible, which does not contradict what is accepted or what is experienced, which gives the instruction about reality, which is for everybody (and) which obliterates errant paths. 91. NA 10:
sva-niscayavad anyeşāṁ niscayôtpadanam budhaih/
parârtham mānam ākhyātam vākyam tad-upacārataḥ // - 'A sentence which brings about the determination for others - just the way [it brings about the determination for oneself - is called by the learned the cognitive criterion for others because of the metaphorical transference of this (cognitive criterion onto the sentence].'
92. For further details see: BALCEROWICZ (2005, $$ 3, 4).

Page #35

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāṇa
63
In addition to that, Siddhasena's peculiar, as it were, usage of the term sastra in a logical-epistemological treatise finds its precedence again in Pramāṇa-samuccaya of Dinnaga. In the well-known opening verse, which is a homage to the Buddha, we read:
'Having paid to [the Buddha] who is a cognitive criterion, who strives for the welfare of the world, who is the teacher (śāstṛ), the well-gone, the res
93
cuer...
And further Dinnaga explains that 'the cause [why the Buddha is a cognitive criterion] is his perfection in inner disposition and in its application. The application [of being a cognitive criterion] is his being a teacher because he edifies the world.' 94 Instead of sastṛ, Siddhasena speaks of sastra, but in an active sense of sastṛ: 'authoritative treatise is that ... which gives the instruction about reality' (tattvôpadeśa-krt sārvaṁ śāstram). As we can easily notice, the idea, the wording and the epistemological context are very similar in both cases.
Thus, there remains nothing that could seriously disprove that either NA 8 or 9 are out of place or are some later interpolations.
8.2. Bansidhar BHATT (2000: 74) expresses his further reservations: 'The NV v. 27 defining the pratyakṣa as kevala appears all of a sudden between the final topic on the parartha anumāna (v. 26) and the conclusion of the entire thesis, viz. pramāṇa-phala (v. 28), without any specific hint of it at the initial stage (v. 1,4, 6 etc.). It is an interpolation.'
We should remember that the concern of the author of the Nyāyâvatāra, which is so closely tight to epistemological-logical issues, is not only to present a new model of epistemology, but also to present it in such a way that it may further serve as the basis of and proof for both Jaina ontology and soteriology, the latter being of paramount interest to the Jainas. This should, again, come to us as no surprise: it suffices to recall the opening verse of Pramāṇa-samuccaya, which correlates soteriological issues and epistemology, as well Dharmakirti's two introductory verses to his Pramāṇa-vārttika. Even
93. PS 1.1ab: pramāṇa-bhātāya jagad-dhitaișine praṇamya sastre sugataya tayine/, quoted in PV1 (Parisista, p. 518.26).
94. PSV: tatra hetur aśaya-prayoga-sampat. chāstṛtvam.
prayogo jagac-chāsanāc

Page #36

Piotr Balcerowicz
the chapter called Pramāna-siddhi of PV is formally a commentary on the idea of the Buddha's pramāna-bhātvatva. Therefore there should be nothing extraordinary if we come across any soteriological discus
on in NA. And precisely such is ultimately the relevance of NA 27 (n. 41), as I shall try to demonstrate in subsequent lines.
The verse NA 27 introduces the idea of perfect perception (kevala), after all 'mundane' varieties of cognition, direct and indirect, have been discussed in the preceding. Thus, the placement of the verse after the parârthânumāna section does not seem illogical or unjustified. On the other hand, it is hardly conceivable that NA could do without even mentioning the notion of kevala-jñāna, because it has always played paramount role in Jaina epistemology, ontology and soteriology. And the most suitable place to mention perfect perception, which is the consummation of all cognitive processes, is in the concluding portion on epistemological issues.
The emphasis on perfect cognition (kevala) was understood, for it fulfilled at least three cardinal functions in Jainism: ontological, epistemological and soteriological. According to Jaina ontology, the structure of the world was highly complex, in which all elements were related to the rest; consequently, the proper description of the reality, of its multiplex character (anekānta), would have to take into consideration all these intricate relations. A complete account of the multiplexity is possible on the level of perfect cognition, which can grasp all complexities. In such a way, only absolute perception (kevala) which perfectly reflects all relations in the world is the warrant for the idea of ontological premise of anekānta.
Besides, being staunch realists, the Jainas maintained that one of the proofs of the multiplexity of reality is the multiplex representation of the world as it is reflected in cognition. A typical argument ran as follows: since any piece of knowledge has a multiplex character, so must be also the world reflected through it 95. Absolute cognition was
95. See e.g. (1) SVIV 1.27 (p. 115.11 ff.): ata evânekānta-siddhiḥ. ... tad evas paramârthataḥ siddhih anekāntāt. - 'Precisely on the basis (of this act of grasping] the multiplexity sof data) is established. ... Thus, in exactly such a way, [we arrive at] a proof (of multiplex reality] on the level of the ultimate truth, because of multiplexity [of appearance)'; (2) NAV 29.1: iha yat pramānam tat parasparavinirlurhitânekadharma-parikarita-vastuno grāhakam, tasyaiva tatra pratibhāsamanatvād; iha yad

Page #37

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
65
considered the most perfect and full representation of the complexity of the world. For this very reason, extrasensory cognition of kevalajñana (NA 27), being the paramount representation of anekānta, is followed almost immediately by its mundane equivalent, viz. the naya-vāda (NA 29), that also proves anekānta, albeit in an imperfect manner.
The soteriological dimension of kevala was equally important. Absolute cognition served as a necessary link to prove that liberation is possible, and was used in the so-called “purification argument'. The full argument boils down to the following: 'You can purify yourself completely, because there is a method. Since your nature is consciousness, when you become absolutely pure, you are by nature endowed with absolute knowledge.' This kind of popular proof was possible only within Jaina ontology thanks to two crucial elements: (1) peculiar understanding of the soul's nature as intrinsically pure and omniscient and (2) the idea of karman as subtle matter, or dirt, that obstructs innate capacities of the soul. In Jainism ātman was conceived of as both the cognitive subject and cognitive instrument %.
yatra pratibhāti, tad eva tad-gocaratayâbhyupagantavyam; tad - yathā nirādinavanayana-prabhava-darśane pratibhāsamānam pātalatayā japā-kusumam tathaiva tadgocaratayâbhyupagamyate; parasparâvibhaktâneka-svabhāvākrânta-mārtikam ca bahir antaś ca vastu sarva-pramānesu prathata ity; atas tad eva teşām gocarah. - '[1. The thesis:) in this world, whatever is a cognitive criterion, it (also) grasps the real thing that is accompanied by multiple properties not detached from each other; (2. the logical reason:) because this (multiplex object] alone is represented in that (cognitive criterion); [3. the invariable concomitance accompanied by the example:) in this world, (if an object] x occurs in a cognitive criterion (cognition)] y, this (object) x alone should be accepted as the domain of this (cognitive criterion (cognition)] y; thus, [for instance]: just the way a (scarlet] China rose flower is being represented as roseate in perception having its origin in flawless eyesight, exactly as such (sc. as a roseate object) it is accepted as the domain of that (cognitive criterion]; [4. the application:) and (similarly, the real thing, both external and internal, endowed with a form that is under the sway, of multiplex essential natures not separate from each other, unfolds itself in all cognitive criteria; (5. the conclusion:] hence, this (multiplex object] alone is the domain of those (all cognitive criteria).' (3) NAV 29.9: tasmāt tasyaiva tatra pratibhāsanāt "sarva-samvidām anekântâtmakam vastu gocara" iti sthitam. - 'Hence, it is established that the real thing, whose essence is multiplex, [forms] the domain of all acts of awareness, because this (multiplex object] alone is represented in that (cognitive criterion).'
96. In its elaborate form it runs in two stages. The first stage has the form: 'The cognitive subject is such whose complete purification is possible, because the means

Page #38

66
Piotr Balcerowicz
Accordingly, a reference to this most important notion from the Jaina point of view finds its most adequate place in the structure of the Nyāyâvatāra: NA 27 concludes the discussion of pramānas, whereas the subsequent verse (NA 28 97) discusses their practical relevance and results (phala). After strictly epistemological issues have been dealt with, NA 29 describes the nature of the object of coghitions, NA 30 relates ontological concerns to epistemology and establishes the rela
etc., hence Pise purification t with the alkalihich the me
for [his) purification exists. In this world, whatever is such the means for the purification (of which] exists is (also) such the complete purification of which is possibly existent, like a particular gem for the purification of which the means exists, (namely] prolonged calcination in a clay furnace with the alkali, etc. And indeed the cognitive subject is such for whose purification the means exists, (namely] repeated practice of cognition, etc., hence (the cognitive subject is such whose complete purification is possibly existent.' (NAV 27.4; sambhavat-samasta-śuddhika ātmā, vidyamānaśuddhy-upāyatvād; iha yo yo vidyamāna-śuddhy-upāyaḥ sa sa sambhavat-samastaśuddhiko; yathā vidyamāna-kşāra-mrt-puța-pākâdi-buddhy-upāyo ratna-viseșas, tathā ca vidyamāna-jñānâdy-abhyāsa-śuddhy-upāya ātmātaḥ sambhavat-samasta-buddhika iti.) This first stage of the argument only proves that purification of the soul is possible, but it still does not prove that supernatural perception or omniscience is possible. In the second stage, the following equation is established: cogniser = cognition (because of the same nature): 'And the cognitive subject, (when) completely purified, is called the absolute, because there is no difference at all between cognition and cogniser.' (NAV 27.4: samastya-suddhaś câtmā jñāna-jñāninoh kathañcid abhedāt. kevalam abhidhīyata iti.) Both stages of this argument was formulated as early as in Kundakunda's works; SSa 278 recounts simile of a transparent crystal (= the knower) which is in its nature unaffected by colours (= passions) but is seemingly changing, the implication of the simile is that the crystal can be cleansed from colours that affect it. The idea that the soul can be omniscient by nature and the soul's knowledge can embrace everything is found e.g. in PSā 1.20, 28. The proof is formulated also by Hemacandra, who instead of the precious stone, speaks of clouds veiling the sun and the moon: 'The veiling of the self] of knowing essence is possible through cognitionveiling and other types of karman just like the moon and the sun (can be covered] by dust, fog, cloud, veil etc.; and like a blow of wind strong enough can remove [the veils obscuring) the moon and the sun, so can meditation and contemplation (remove veils obscuring the knowing self].' (PMIV 1.15 $ 50 (p. 12.20-22): prakāśa-svabhāvasyâpi candrârkâder iva rajo-nīhāhārâbhra-patalâdibhir iva jñānâvaranīyādikarmabhir āvaraṇasya sambhavāt, candrârkâder iva prabala-pavamāna-prāyair dhyana-bhavanadibhir vilayasy-ti.). 97. NA 28: pramānasya phalam sāksād ajñāna-vinivartanam /
kevalasya sukhôpekṣe seşasyâdāna-hāna-dhih // - 'The direct result of cognitive criterion is the cessation (sc. removal of nescience; [the result] of the absolute [cognition) is [both) happiness and indifference; [the result] of the remaining (ones) is the faculty of appropriation and avoidance'.

Page #39

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
67
tionship between cognition and the reality, NA 31 98 characterises the cognitive subject. The structure of the work in itself appears to be quite coherent.
Conspicuously, this structure of NA 27-28, 31 corresponds also to that of the last chapter of the Tattvârtha-sūtra. In TS, first the causes of the kevala knowledge are mentioned, viz. “the destruction of confusion (sc. delusive karman) as well as the destruction of [the karman) veiling cognition, of [the karman) veiling insight (conation) and of the obstructive [karman]', all of them infecting innate cognitive capacities of the soul. 99 The contents of TS 10.1 corresponds to NA 28ab: "The direct result of cognitive criterion is the removal of nescience' (pramāṇasya phalam sākṣād ajñāna-vinivartanam). Secondly, Umāsvāmin describes the result of perfect cognition, which is liberation, and defines it as a complete freedom (vipramoksa), or destruction of (ksaya) of all karmans, which are both the cause and manifestation of suffering. 100 Again, this corresponds to NA 28c: '[the result] of the absolute cognition is [both) happiness and indifference' (kevalasya sukhôpekṣe). The subsequent portion of TS is an account of the final journey of the liberated soul to the top of the world, the abode of perfected beings (siddha-loka), which has no relevance in epistemological context. Clearly, it is understandable that it finds no equivalent in NA. However, the final aphorism of TS describes the nature of a perfected soul 101, and is echoed by NA 31. The final verse of NA 32 is merely a summary.
As we can see, also final verses of NA do not only form a consistent whole, but even comply with the contents of the traditional Jaina
98. NA 31: pramātā svânya-nirbhāsi kartā bhoktā vivrttimän/
sva-samvedana-samsiddho jīvah kşity-ādy-anātmakaḥ // - 'The cogniser is the observer of himself and of something different, the agent, the experiencing subject, is subject to change, is well-established by self-cognition, is the living element [and is someone) whose essence is not of earth, etc.
99. TS 10.1: moha-ksayāj jñāna-darśanâvaranântarāya-ksayāc ca kevalam.
100. The Svetāmbara recension reads it as two separate sūtras TS1 10.2-3: bandha-hetv-abhāva-nirjarābhyām, krtsna-karma-kşayo mokṣaḥ; the Digambara recension reads it as one sūtra, with a slight modification TS2 10.2: bandha-hetyabhāva-nirjarābhyām krtsna-karma-vipramokṣo mokṣaḥ.
101. TSi 10.7 = TS2 10.9: kşetra-kāla-gati-linga-tirtha-cāritra-pratyeka-buddha-bodhita-jñānâvagāhanântara-saṁkhyâlpa-bahutvataḥ sādhyāḥ.

Page #40

68
Piotr Balcerowicz
textbook. In view of the above, it is most debatable whether one should consider any of these kārikās an interpolation.
8.3. In addition, Bansidhar BHATT (2000: 74) dismisses the authenticity of NA 27 on metrical grounds (“it is an interpolation'), whereupon adds: 'In contradistinction to v. 27, the authenticity of v. 26 cannot be questioned despite of its defective meter. ... Probably, this verse has been somehow disturbed.' No additional reasons are offered, why one of the two verses is accepted as spurious, whereas the other one is taken as authentic. Indeed, both the verses have defective meter and both have irregular number of syllables: NA 26 has only 7 syllables in pāda b, whereas NA 27 has 9 syllables in pāda a! Since NA 26 and 27 share the same fate of being abhinna-yoga-ksema (sc. are based on identical productive-supportive principle) in metrical terms, such a conclusions seem arbitrary. Either one should reject the authenticity of both, or accept them both as genuine constituents of NA, at least in terms of metrical analysis.
8.4. Concluding the main section of his paper, BHATT (2000: 75) briefly points to some terminological affinity between Prajñākaragupta's Pramāna-vārttikâlankāra and NA, which should, as far as I can understand, prove that NA is posterior to PVA: 'Some of its [of NA 32] expressions can be compared with those of PVB e.g. samvyavahārikam etad (= pramānam)..., and vyavahāratah ... pramāṇatva-vyavasthiti - (PVB 1.5.197, pp. 25-26).' Apart from the fact, that there is only a loose similarity in terms of terminology with the above-quoted expressions (NA 32 has: pramāņâdi-vyavasthêyam and sarva-saṁvyavahartřņāṁ), so it would not be easy to prove any direct relationship between PBV and NA on this basis only, the way the conclusion is reached is rather problematic. Indeed, when we compare expressions found in NA 32 102 with terminology found in other
102. NA 32: pramānâdi-vyavasthêyam anādi-nidhanâtmikā /
sarva-samvyavahartiņām prasiddhâpi prakirtitā // - 'The distinctive character of cognitive criteria etc., by nature with no beginning nor end, even though [it is) well-known to all (people) absorbed in everyday life, is [here) declared.'

Page #41

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
69
works, we discover similarities not only with Prajñākaragupta and his PVA, but with a number of other Buddhist works that precede Prajñākaragupta. The two expressions which BHATT has in mind bear also resemblance to some expressions found in the Pramāna-vārttika: PV1 2.58cd: arthakriyânurodhena pramānatyam vyavasthitam II. PV1 1.5ab: prāmānyaṁ vyavahārena sāstram moha-nivartanam 1, PV1 4.183: anumānânumeyârtha-vyavahāra-sthitis tv iyaṁ / This only shows that both Siddhasena and Prajñākaragupta were influenced by Dharmakirti. On this basis it would not be possible to establish any relative chronology between Siddhasena and Prajñākaragupta.
9. The authenticity of NA 16, viz. the illustration of a skilled archer, has frequently been questioned, most recently by BHATT (2000: 72). Indeed it is rather surprising to find, in such short treatise, a single explicit example which does not seem absolutely necessary at all: one could easily imagine the treatise without it.
Let us first take a closer look at the context in which this puzzling simile transpires, viz. NA 14-16:
'[14] The thesis is the acceptance of the inferable property; [it] is not revoked by perception, etc.; the pronouncement of it has to be made here as showing the domain of the logical reason.
(15) Otherwise, for a (person] to be apprised, who is confused regarding the domain of the logical reason intended by the proponent, the logical reason might appear to be suspected of being contradictory, just like...
(16) ...for a person watching an archer's skill, the archer who hits without the specific mention of the target [is endowed with both) skill and its op
posite. 103
Conspicuously, NA 16 is announced with yathā in NA 150, through which both verses are syntactically connected. One might,
103. NA 14-16 [14]
[15]
sādhyâbhyupagamah pakşah pratyakşâdy-anirākṣtah/ tat-prayogo 'tra kartavyo hetor gocara-dipakaḥ // anyathā vādy-abhipreta-hetu-gocara-mohinah/ pratyāyyasya bhaved dhetur viruddhârekito yathā // dhānuska-guna-samprekşi-janasya parividhyataḥ / dhānuşkasya vină laksya-nirdesena gunetarau //
[16]

Page #42

70
Piotr Balcerowicz
however, easily argue that yathā was a later modification after a new verse (NA 16) was inserted.
On the other hand, we can see that the three verses follow in a logical sequence: NA 14 defines the subject of the thesis (paksa) and, most importantly, verbalises the necessity to formulate it as a required member in a proof formula; NA 15 is a prasanga-type of argument: without a clear pronouncement of the thesis, the debaters and spectators may come to the conclusion that one's thesis is a fallacy (paksâbhāsa); NA 16 is a typical drstānta to illustrate the idea. So much effort (and space!) just to express the importance of paksa in the proof formula? That appears striking. Clearly, NA 15 and 16 must have been formulated against an opinion of someone who maintained that no pratijñā / pakșa is necessary at all.
In the Pramāņa-vārttika we find the context for this puzzle. Dharmakirti argues that statement of the thesis (pakşa-vacana) is not a necessary member of the proof formula, because it does not possess any capacity to prove anything, PV 4.18-22:
'[18] The assertion of the incapability (to prove anything on the part] of this (statement of the thesis (paksa-vacana, PV.4.16)] is made on the ground that (the statement of the thesis) has (merely) as its contents the object (sc. inferable property) of the logical reason.
[Objection:] "Also this statement of the thesis must] have the capability (to prove), because it facilitates the statement of the logical reason".
[19] [Rejoinder:] [Then,] for a person who wishes to know (the true state of affairs) due to his doubt as regards this (inferable property (sc. whether it is present or not)], there should (also) be a ground for an opportunity (to produce this doubt as a proving member of the proof formula). Also, when one accepts a counter-proposition, this (should be accepted) as equal (member of the proof formula, that proves the thesis). Thus, there would be infinite regress (sc. no limit to the number of efficient members that prove).
[20] However, the intrinsic efficacy (to prove the thesis) lies in three fea-tures (of the logical reason). Only the statement of these (three features] prevails as that which activates the memory as regards these (three features, and thus has the capacity to prove.]
[21] [Objection:] "For when (the logical reason) is established sto have its scope) only on account of the demonstration of the scope - because

Page #43

On the Relationship of the Nyayâvatāra and the Sammati-tarka-prakarāṇa
71
the operation of the logical reason would be impossible, if the scope [of inference (sc. thesis)] were not demonstrated - this [logical reason] is capable of proving the thesis]".
[22] [Rejoinder:] [A reply] to this has already been given [in PV 4.19], [namely that] also without this [scope being demonstrated (sc. without the statement of the thesis)], even when one asserts [only]: "Sound is [something] which has been produced, [hence] all [entities] like this are impermanent", [then] the comprehension of the impermanence of this [sound] should occur by implication.' 104
With the purpose to disprove Dharmakirti's position, Siddhasena Mahamati inserts the simile in order to show the proper role of the thesis. He does agree that pakṣa is not an integral part of the proof formula in the logical sense, inasmuch as it has no 'proving capacity'. Its role is rather didactic, to clearly demonstrate what the proof formula is intended for. But also the thesis corroborates the soundness of the
104. PV1/PV3 4.18-22:
hetv-artha-viṣayatvena tad-aśaktôktir iritā/ śaktis tasyâpi ced dhetu-vacanasya pravartanāt // tat-samśayena jijñāsor bhavet prakaranâśrayaḥ/ vipaksôpagame 'py etat tulyam ity anavasthitiḥ // antar-angam tu sāmarthyam triṣu rūpeṣu samsthitam/ tatra smṛti-samādhānaṁ tad-vacasy eva samsthitam // akhyāpite hi viṣaye hetu-vṛtter asambhavat/ viṣaya-khyāpanād eva siddhau cet tasya saktatā" // uktam atra" vinâpy asmat kṛtakaḥ śabda idṛśaḥ/ sarve 'nitya iti prokte 'py arthat tan-nāśa-dhir bhavet //
My translation differs in some crucial points from that of TILLEMANS' (2000:30-36). a. Cf. PV2 4.21cd: viṣaya-khyāpanād eva samarthyam iti cen matam // For variae lectionis see TILLEMANS (2000: 35, n. 130). It seems to me that the translation in TILLEMANS (2000: 35) slightly distorts the logical connection of the elements of the verse, which is as follows: The condition in the clause is siddhau, to be connected with hetu-vṛtter (primarily related with asambhavat, but here supplemented). Its justification is viṣaya-khyāpanād eva, as the necessary single condition; the ground for the fact the operation of the logical reason can be warranted 'only on account of the demonstration of its scope', is 4.21ab, which explains why and under what conditions the logical reason is not operational. When the condition is fulfilled ([hetu-vṛtter] siddhau), it is effective (tasya saktatā).
b. Cf. PV2 4.ab: vyāpti-pārve vinâpy asmāt kṛtakaḥ śabda idṛśaḥ / For variae lectionis see TILLEMANS (2000: 36, n. 131).
A similar idea is expressed succinctly in NB 3.34: dvayor apy anayoḥ prayogayor na avasyam pakṣa-nirdeśaḥ.

Page #44

72
Piotr Balcerowicz
proof formula by indicating that no fallacy is involved. The role of the simile of the skilled archer is not simply to point to some empirical instance of a contest, but to illustrate that in a well-defined context (e.g. when all onlookers see the archer's target) any explicit mention of the target can be easily dispensed with. In this way, not only the verses of NA 14-15 are necessary elements in the argumentative structure, but also NA 16.
10.1. The idea of 'perception for others' is referred to not only in Buddhist sources, but also in later Mimāṁsā literature, i.e. by Sucaritamiśra in his MSVT (III: p., 38.5-8) ad MSV 5.4.53-54: athânumāna-gocari-kṛtârtha-pratipadana-samartha-vacanapārârthyād anumānam parārtham ity upacaryate, tataḥ pratyakṣapratipannam apy artham bodhayad vacaḥ parārtham iti pratyakṣam api parârtham apa-dyeta. Interestingly, the context in which the distinction svârtha-parârtha is mentioned is the critical evaluation of the Buddhist concept of twofold inference. Both Umveka Bhatta and Sucaritamiśra reject the distinction into svârtha- and parârthânumâna in their respective commentaries on MSV 5.4.53-54, viz. ŚVVTT (p. 317.15-318.16, esp.: na tu parârthânu-mānam nama kimcid astîtyuktam) and MSVT (III: p., 37.11-40.21) respectively. Both argue that what the Buddhists call parârthânumāna is, in fact, just a verbal statement which communicates the result of an inference drawn by the speaker to the hearer; the hearer, on the basis of the utterance, subsequently draws his own inference, and there is no room for 'inference for others'. Interestingly, Sucaritamiśra avails himself, in addition, of the Jaina concept of parârtha-pratyakṣa and argues that in the same manner as one were to accept parârthânumāna one would also have to consent to parârtha-pratyakṣa, because both inference and perception can be indirectly triggered by a verbal
statement.
It would be quite natural that anyone who is willing to refute the idea of parârtha-pratyakṣa and is at the same time acquainted with the notion of parâtha-pratyakṣa, would use the latter to disprove the former. However, Umveka Bhatta does not seem to know the idea of parârtha-pratyākṣa at all; he is silent on it in his commentary on the same verses (MSV 5.4.53-54), which are an occasion for Sucarita

Page #45

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāna
73
miśra to speak of this kind of perception, see: ŚVVTT (p. 317-318) 105. That may be a hint (not a decisive proof!) that Siddhasena Mahāmati composed NA either after Umveka Bhatta or at the same time, but NA and the ideas contained in it did not reach any prominence outside Jaina circles by the time of Umveka. Alternatively, in case the idea of parâtha-pratyakṣa was not Siddhasena's own invention, but he borrowed it from some earlier Jaina source, one may likewise suppose that the inventor of the idea of parâtha-pratyaksa, who inspired Siddhasena, lived either after of contemporaneously with Umveka Bhatta. The date of Umveka Bhatta is uncertain, but can be roughly assigned to the first half of the gth century 106. That might mean that Nyāyâvatāra was not composed before 700.
10.2. We can be quite certain that NA was composed also after Pātrasvāmin (alias Pātrakesarin / Pātrakesarisvāmin), the author of the Tri-laksana-kadarthana. In his lost work Tri-laksana-kadarthana 107, Pātrasvāmin criticises Dharmakirti's concept of triple-formed logical reason (trairūpya) and the three restriction criteria of validity imposed on it (traividhya-niyama 108), and offers instead his own definition of valid hetu, viz. the relation of the inexplicability otherwise' (anyathânupapatti), which was meant to replace Dharmakīrti's definition. That clearly indicates that Pātrasvāmin was posterior to Dharmakirti 109. Both Jaina and Buddhist traditions regard him to be
105. On this discussion compare also Govardhan P. BHATT (1989: 248-249).
106. Either c. 700-750 (according to K. Kunjunni Raja in his 'Preface' (p. x] to ŚVVTT) or c. 710 (according to EIPHIL I: 8 371).
107. Anantavirya in SVIT (ad Svi 6.1, p. 371.19-372.6) mentions the title of Pātrasvāmin's (= Pārrakesarin's) work: Tri-laksana-kadarthana, and quotes a verse from it:
nânyathānupapannatvam yatra tatra trayena kim /
anyathânupapannatvam yatra tatra trayeņa kim // The verse is also quoted also in: TSa 1369 and SVR ad 3.13 (p. 521.5-6). Importantly, it is incorporated by Akalanka in his NVil 323 (p. 74.1-2) = NVil 2.154 (II: p. 177.22-23), and the fact that the verse was not a later insertion of the commentator Vadirāja-sūri is confirmed by the fact that Vadirāja-sūri comments on the verse in his NViV exactly in the same manner as he does in the case of other Akalanka's verses.
108. Cf. SHIGA (2003: 489).
109. Cf. also SHIGA (2003: 489): 'Patrasvāmin knew and criticised Dharmakirti as well as Dinnāga'.

Page #46

74
Piotr Balcerowicz
the inventor of this new notion of the logical reason 110. Furthermore, Pātrasvāmin's use of examples of invalidating Dharmakirti's trairūpya-hetu, which partly overlap with those used by Kumārila, shows that Pātrasvāmin was posterior to Kumārila as well 111. In his turn, Pātrasvāmin influenced Siddhasena Mahāmati, who uses the idea of anyathānupapatti as some-thing already well known"12.
11. To recapitulate, there are some points that speak in favour of the separate authorship of STP and NA, namely (1) the peculiar use of the concepts sāmānya and viseșa and their application to the terms darśana and jñāna that indicates STP was composed before Dinnāga and Dharmakirti (8$ 1.1, 2.1-2.2), whereas NA was definitely con
110. For instance, Vādirāja-süri (NVIV 2.160, II: p. 186.24-26: sa prasiddhah sa vā pātrakesarisvāminā nirāpitaḥ avinâbhāva eva sambandho hetu-sādhyayor na tādātmyâdis tasyâvyāpakatvāt) confirms that it was Patrasvāmin who introduced the notions of 'inexplicability otherwise' (anyathânupapatti) as the definition of valid logical reason and of relation of inseparable connection (avinâbhāva) as the single logical relation between the logical reason and the inferable property (sādhya). This is further corroborated by Vādideva-sūri in SVR (ad 3.13, p. 521.5-6: tad uktam patrasvāminā), see also NViV 2.171ab (II: p. 198.30-31). The same information is also supplied by Santarakṣita, who mentions Pātrasvāmin as the source of the idea, see TSal 1364 (p. 405.1: anyathêty-ādinā pātrasvāmi-matam āśankate...), and quotes a number of verses from the lost Tri-laksana-kadarthana in TSa 1364-1379. As regards the correct reading. of the verses, TSa 1365cd should be emended to: eka-laksanaka - so 'rthät caturlaksanako na vā II, instead of 'rthas, see STEINKELLNER (forthcoming).
Furthermore, Kamalasila quotes two more verses (TSaPad TSa 1386 (p. 409.1214)), which I believe (see BALCEROWICZ (2003: 359)) to stem from the same work of Pātrasvāmin:
vinā sādhyād adsstasya drstānte hetutê syate / parair mayā punar dharminy asambhāsnor vinâmună // arthâpatteś ca śābaryā bhaiksavāca cânumānatah/
anyad evânumānam no nara-simhavad isyate // [a TS1: bhaikṣavāś. For the emendation cf. PATHAK (1930: 156-7) and KUNST (1939: 26, n. 3).]
My ascription of these two verses to Pātrasvāmin is corroborated by Jinendrabuddhi's Pramāna-samuccaya-ţikā, as confirmed by Ernst Steinkellner in private communication: ‘PST B(i.e. the second manuscript) folio 54a2 says (after first quoting the stanza of TS 1365, and explaining anupapannatvam as aklptir asambhavah): ślokam apy āha: vinā sādhyād ...' = Tibetan translation of the verse ( 5766, 92a8f.): dpe la bsgrub bya med pa las/rtags ñid ma mthong gêan gyis 'dod / bdag gis 'di ni med par yang /chos can la ni mi srid pa'o/
111. For instance MŚV 5.4.64cd-65ab= TSa 1372, MŚV 5.7.46 = TSa 1377, MSV 5.4.67d = TSa 1378. See BALCEROWICZ (2003: 343 ff.)..
112. See BALCEROWICZ (2003: 343).
199 My aschis Pramana sai.e. the se

Page #47

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
75
ceived after Dharmakirti (§ 2.2); (2) the assignment of either sensory (NA) or strictly suprasensory (STP) character to pratyakşa, taken either as perception (NA) or as blanket term 'direct cognition' (STP) as well as the relation to concept of cognitive criterion (pramāņa) (98 3.1-3.2); (3) the (un)importance of the unity of iñāna and darśana at the kevala stage and different treatment of kevala ($ 3.3); (4) role of the four-phased sensuous cognition (mati-jñāna) in the epistemic schemes of STP and NA ($ 4); (5) divergent opinions on the direct, i.e. perceptual character (pratyakşa) of verbal utterances and on the thesis that things conveyed through language are grasped directly ( 5.2); (6) different attitudes to the Agamic tradition and to novel solutions (6). These points are additionally strengthened by a number of minor differences and incongruences (8 1.1) that by themselves are not only inconclusive but could probably be explained away.
Furthermore, the text of the Nyāyâvatāra does not seem to contain any serious interpolations, perhaps with the exception of some minor changes in the wording, conspicuous in the defective meter of NA 26 and 27 ($ 8.3).
'In view of the lack of any hint that that author of STP knew of any novel concepts introduced by Dinnāga, I would maintain that he must have flourished before ca. 500 CE, viz. at least about 150 years before the composition of NA.
Finally, considering Siddhasena Mahāmati's dependence on Pātrasvāmin (§ 10.1), the Mimāṁsaka evidence (§ 10.2) and Prajñākaragupta's reaction (7), we may suggest roughly the following relative chronology:
Siddhasena Divakara (STP): 450-500 Dinnāga: 480-540 Dharmakārti: 600-660 Pātrasvāmin: c. 660-720 Umveka Bhatta: c. 700-750 Siddhasena Mahāmati (NA): c. 720-780 Prajnakaragupta: c. 800 Haribhadra-sūri: c. 800

Page #48

76
Piotr Balcerowicz
BIBLIOGRAPHY
AņD = Aņuoga-ddārāim (Anuyoga-dvārāņi). See: Nams. BALCEROWICZ 2001a = Balcerowicz, Piotr: Jaina Epistemology in
Historical and Comparative Perspective. Critical Edition and English Translation of Logical-Epistemological Treatises: Nyāyâvatāra, Nyāyāvatāra-vivrti and Nyāyâvatāra-tippana with Introduction and Notes. 1-2 Vols., Alt- und Neu-Indische Studien
53, 1-2, Franz Steiner Verlag, Stuttgart 2001. BALCEROWICZ 2001b = Balcerowicz, Piotr: 'Two Siddhasenas and the
Authorship of the Nyāyāvatāra and the Samati-tarka-prakarana',
Journal of Indian Philosophy 29/3 (2001) 351-378. Balcerowicz 2003 = Balcerowicz, Piotr: ‘Is "Inexplicability
Otherwise" (anyathānupapatti) Otherwise Inexplicable?', Journal of Indian Philosophy 1-3 (2003) 343-380 (Proceedings of the International Seminar 'Argument and Reason in Indian Logic'
20-24 June, 2001 – Kazimierz Dolny, Poland). BALCEROWICZ 2005 = Balcerowicz, Piotr: 'Pramānas and language. A
Dispute between Dinnāga, Dharmakirti and Akalanka'. Journal of
Indian Philosophy (2005) [in print]. Bhatt 1989 = Bhatt, Govardhan P.: The Basic Ways of Knowing. An.
In-depth Study of Kumārila's Contribution to Indian Epistemology, Second Revised Edition, Motilal Banarsidass,
Delhi 1989. Bhatt 2000 = Bhatt, Bansidhar: 'A Study in the Nyāyâvatāra of
Siddhasena', Festgabe für Adelheid Mette, Indica Tibetica 37
(2000) 67-82 [Swissthal-Oldendorf]. DHAKY 1981-82 = Dhaky, M. A.: "Some less known verses of
Siddhasena Divākara’, Sambodhi 10 (1981-82) 169-173. DhakY 1995 = Dhaky, M. A.: "The Date and Authorship of
Nyāyāvatāra', Nirgrantha 1 (1995) 39-49. [Eds. M. A. Dhaky, Jitendra Shah, Ahmedabad).

Page #49

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
77
DhPr = Durveka Miśra: Dharmottara-pradipa. Pt. Dalsukhbhai
Malvania (ed.): Pandita Durveka Miśra's Dharmottara-pradipa [being a subcommentary on Dharmottara's Nyāya-bindu-tīkā, a commentary on Dharmakirti's Nyāya-bindu). Kashi Prasad
Jayaswal Research Institute, Patna 1955 [reprinted: 1971). DHRUVA 1930 = Dhruva, A.B.: 'Introduction', in: NP1, pp. V-XXXV. EIPHIL I = Potter, Karl H.: Encyclopedia of Indian Philosophies. –
Vol. I: Bibliography. Compiled by ... . Third Revised Edition. Princeton University Press, Princeton-New Jersey 1995 [Indian
Edition: Motilal Banarsidass, Delhi 1995). FRAUWALLNER 1961 = Frauwallner, Erich: 'Landmarks in the History
of Indian Logic', Wiener Zeitschrift für die Kunde Süd- und
Ostasiens 5 (1961) 125-148. GRANOFF 1989-1990 = Granoff, Phyllis: 'The Bibliographies of
Siddhasena - A Study in the Texture of Allusion and the Weaving of a Group Image'. Part I: Journal of Indian Philosophy 17 (1989)
329-384. Part II: Journal of Indian Philosophy 18 (1990) 261-304. HATTORI 1968 = Hattori, Masaaki: Dignāga, On Perception, being the
Pratyaksa-pariccheda of Dignāga's Pramāna-samuccaya, Edition of Tibetan translations and the Sanskrit text as well as the English
translation of the Chapter I. Harvard University Press 1968. JAMBŪVIJAYA 1981 = Muni Jambūvijaya: “Jainācārya-Śri
Hemacandrasūri-mukhyaśişyābhyām ācārya-RāmacandraGuņacandrābhyāṁ viracitāyāṁ Dravyālańkāra-svopajñatikāyām'. In: Studien zum Jainismus und Buddhismus. Gedenkschrift für Ludwig Alsdorf. hrsg. von Klaus Bruhn und Albrecht Wezler. Alt- und Neu-Indische Studien Band 23, Franz
Steiner Verlag, Wiesbaden 1981: 121-149. MALVANIA 1979 = Malvania, Dalsukhbhai: “Pariśista 1, Nyāyāvatara
ki Tulanā', 1979.

Page #50

78
Piotr Balcerowicz
Mironov 1927 = Mironov, N.D.: 'Dignāga's Nyāyapraveśa and
Haribhadra's tikā on it', Jaina Shasan, Extra (Divali) No., Benares 1911. (2) (Reprinted] in: Aus Indiens Kultur - Festausgabe für Richard von Garbe, ed. Julius von Negelein,
Erlangen 1927. MOOKERJEE 1971 = Mookerjee, Satkari: ‘A critical and comparative
study of Jain Logic and Epistemology on the basis of the Nyāyâvatāra of Siddhasena Divākara’, Vaishali Institute Research
Bulletin No. 1, Vaishali 1971. MŚVT = Sucaritamiśra: Mimāṁsā-śloka-vārttika-tikā. Sāmbaśiva
Šāstri; V.A. Rāmasvāmi Śāstri (eds.): Ślokavārttikam Sucaritamiśra-pranītayā Kāśikākhyayātīkayā sametam. 3 Vols.
Trivandrum 1926-1943. MUKHTAR 1956 = Mukhtar, Jugal Kishor: 'Samatisūtra aur
Siddhasena', Jaina Sāhitya aur Itihāsa par Visada Prakāśa,
Calcutta 1956: 538-543. NaṁS = Namdi-sutta / Namdi-sutta [Nandi-sūtra / Nandi-sūtra).
Muni Punyavijaya, Dalsukh Mālvaniā, Amritlal Mohanlal Bhojak (eds.): Nandi-suttaṁ and Aņuoga-ddārāim (Nandi-sūtra and Anyoga-dvārāṇi). Jaina-Agama-Series No.1, Shri Mahāvīra Jaina
Vidyalaya, Bombay 1968. NA = Siddhasena Divākara: Nyāyāvatāra. (1) See BALCEROWICZ
(2001a). (2) Bhagavāndās Harsacandra (Harakhchand) (ed.): Mahāvādi-sri-Siddhasena-Divākara-praņita-Nyāyāvatāraḥ śri Rājasekharasūri-viracita-Tippana samalankrta-ācāryaSiddharși-viracita-Vivrti-sahitaḥ. Hemācāndrācārya-Jaina-sabhā, Ahmedabad-Patan 1917. (3) P.L. Vaidya (ed.): Nyāyāvatāra of Siddhasena-Divākara with The Tippana of Devabhadra. Shri Jain Shwetamber Conference, Bombay 1928 [reprinted: Vaishali Institute Research Bulletin 1
(1971) 1-95). NAV = Siddharși-ganin: Nyāyāvatāra-vivrti. See: NA. NB = Dharmakirti: Nyāya-bindu. See: DhPr. NBT = Dharmottara: Nyāya-bindu-ţikā. See: DhPr.

Page #51

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
79
NM = Giuseppe Tucci: The Nyāya-mukha of Dignāga. The oldest
Buddhist Text on Logic, after Chinese and Tibetan Materials.
Heidelberg 1930. NP = Sankarasvāmin: Nyāya-praveśa. (1) 'Part One: The Nyāya
praveśa - Sanskrit Text with Commentaries' ed. by A. B. Dhruva, Oriental Institute, Baroda 1930; 'Part Two: Nyāyapraveśa of Ācārya Dinnāga – Tibetan Text ed. by Vidushekhara Bhattacharyya, Oriental Institute - Central Library, Baroda 1927. (2) Piotr Balcerowicz: 'Sankarasvāmin: Nyāya-praveśa - "Wprowadzenie w logike” (“Introduction to Logic”]’, Studia
Indologiczne 2 (1995) 72-77. NVi = Akalanka Bhatta: Nyāya-viniścaya. (1) Nyāyâcarya Mahendra
Pandita Kumār Šāstri (ed.): Śrimad-Bhattâkalarka-deva-viracitam Akalanka-grantha-trayam (Svôpajña-vivrti-sahitam Laghiyas-trayam, Nyāya-viniscayaḥ, Pramāņa-sangrahaś ca). Sanasvati Pustak Bhandār, Ahmadābād (Ahmedabad) 1996 [1st
edition: Ahmedabad-Calcutta 1939). (2) Mahendra Kumar Jain - (ed.): Nyāyaviniscaya-vivaraṇa of Śrī Vādirāja Sūri, the Sanskrit
Commentary on Bhatta Akalarkadeva's Nyāyaviniscaya. Vol. 1 & 2, Bhāratiya Jñānapitha Prakāśana, New Delhi, V 1949, 1955
[2nd edition: BhJP, D 2000). NViV = Vādirāja-sūri: Nyāya-viniscaya-vivaraņa. See: NVi2. PALV = Māņikyanandin: Pariksāmukha-sūtra. Pariksāmukha-sūtra
of Māņikya Nandi together with the Commentary Called : Parīkņāmukha-lagu-výtti of Ananta Virya, ed. by Satis Chandra
Vidyābhāsaņa. Bibliotheca Indica No. 1209, Calcutta 1909. Pann = Pannavanā-sutta. Muni Punyavijaya, Dalsukh Mālvania,
Amritlâl Mohanlāl Bhojak (eds.): Pannāvaņāsuttam. JainaĀgama-Series No. 9, Parts 1-2, Shri Mahāvīra Jaina Vidyālaya,
Bombay 1969-1971. PISCHEL 1981 = Pischel, Richard: A Grammar of the Prākrit
Languages. Translated from German by Subhadra Jhā. Second Revised Edition, Motilal Banarsidass, Delhi 1981. [Grammatic der Prakrit-Sprachen. Band 1. Heft 8. Grundriss der IndoArischen Philologie und Altertumskunde. Strassburg 1990.).

Page #52

80
Piotr Balcerowicz
PMi = Hemacandra-sūri: Pramāna-mīmīṁsā. (1) Sukhlalji Sanghavi,
Mahendra Kumar and Dalsukh Malvaniya (ed.): Kavikālasarvajña-Śri-Hemacandrâcārya-viracitā svopajña-vrtti-sahitā Pramāņa Mimiṁsā (with Bhāṣā Tippaņa of Pandita Sukhlalji Sanghvi). SPBh 1998. (2) Satkari Mookerjee and Nathmal Tatia (ed.): Hemacandra's Pramāna-mimāṁsā. Text and Translation
with Critical Notes. Tara Publications, Varanasi 1970. PMIV = Hemacandra-sūri: Pramāņa-mimāṁsā-svopajña-vrtti. See:
PMi. PNTĀA = Vādideva-sūri: Pramāņa-naya-tattvâlokâlankāra. Pt.
Dalsukhbhai Malvania (ed.): Pramānanayatattvāloka of vādin Devasūri with a commentary with the commentary Ratnākarāvatārikā of Ratnaprabhasūri, with Pañjikā of Rājasekhara, Țipppaņa of Jñānacandra. L.D. Series 16,
Ahmedabad 1969. PS = Dinnāga: Pramāņa-samuccaya. (1) See: HATTORI (1968). (2) In:
Dvādaśāram Naya-cakram of Acārya Sri Simhasūri Gani Vādi Kșamā-shramaņa. (Muni Jambāvijayaji), Pt. I, (1-4 Aras),
Bhāvahagar 1966: pp. 97-134. PSā = Kundakunda: Pavayana-sāra [Pravacana-sāra). A.N. Upadhye
(ed.): Sri Kundakundācārya's Pravacanasāra (Pavayanasāra), a Pro-Canonical Text of the Jainas, the Prakrit Text critically edited with the Sanskrit Commentaries of Amrtacandra and Jayasena. Bombay 1935. [reprinted: Sri Paramaśruta-Prabhāvaka
Mandala, Srimad Rājacandra Aśrama, Agās (Gujarat) 1984). PSV = Dinnāga: Pramāņa-samuccaya-vrtti. See: PS. PV = Dharmakirti: Pramāna-vārttika. (1) Rāhula Sānkrtyāyana (ed.):
Pramāņavārttikam Ācārya-Manorathanandi-kytayā vrttyā saṁvalitam / Dharmakirti's Pramānavārttikam with a commentary by Manorathanandin. Appendix to Journal of the Bihar and Orissa Research Society 24 (1938) Parts I-II: 1-123 / Patna 19381940. (2) See: PVA. (3) Yāsho Miyasaka (ed.): ‘Pramāņa-vārttika-kārikā (Sanskrit and Tibetan).' (Chapter 2 = Pramāna-siddhi, Chapter 3 = Pratyaksa, Chapter 4 = Parârthânumāna). Acta

Page #53

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāna
81
Indologica (Indo-koten-kenkyā] 2 (1971/72) 1-206 [Narita:
Naritasan Shinshoji]. PVA = Prajñākaragupta: Pramāna-vārttikälankāra. Tripitakāchārya
Rāhula Sānkstyāyana (ed.): Pramāņavārttikabhāşyam or Vārtikālankārah of Prajñākaragupta. (Being a commentary on Dharmakirti's Pramānavārtikam). Kashi Prasad Jayaswal
Research Institute, Patna 1953. PVV = Manorathanandin: Pramāņa-vārttika-vrtti. See: PV1. Shastri 1990 = Shastri, Indra Chandra: Jaina Epistemology, P. V.
Research Series No. 50, P. V. Research Institute, Varanasi 1990. SHIGA 2003 = Shiga, Kiyokuni: 'Jaina objection against trividha-hetu:
an opinion attributed to Pātrasvāmin', Journal of Indian and
Buddhist Studies 52.1 (2003) 491-488. SSā = Kundakunda: Samaya-sāra. (1) Manohar Varņi Sahajānand
(ed.): Parama-pūjya-śrīmat-Kundakundācārya-devena praṇīta[h] Samaya-sāra[h], parama-pājya-śrimad-Amstacandrasāriviracita-saṁskṛta-ţikā Atma-khyāti, Sapta-daśāngă-ţikā. Khemacanda Jaina Sarārph, Mastri Sahajānanda Šāstramālā,
Raņajitapuri (Sadar Merath) 1977. STEINKELLNER-Much 1995 = Steinkellner, Ernst; Much, Michael
Torsten: Texte der erkenntnistheoretischen Schule des Buddhismus - Systematische Übersicht über die buddhistische Sanskrit-Literatur II, Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-Historische Klasse,
Dritte Folge Nr. 214, Vandenhoeck & Ruprecht, Göttingen 1995. STP = Siddhasena Divākara: Sammati-tarka-prakarana. Sukhlāl
Sanghavi; Becardās Dośi (ed.): Saṁmatitarka-prakaranam by Siddhasena Divākara with Abhayadevasūri's Tattva-bodha-vidhāyini. Gujarāt-purā-tattva-mandir-granthâvali 10, 16, 18, 19, 21, Gujarāt-purā-tattva-mandir, Amdāvād 1924-1931 [reprinted:
Vol. I & II, Rinsen Buddhist Text Series VI-1,2; Kyoto 1984). STEINKELLNER (forthcoming) = Steinkellner, Ernst: 'An Old
Transmissional Mistake in Pātrasvāmin's Definition of the Logical Reason as Quoted by Sāntarakṣita and Jinendrabuddhi', forthcoming.

Page #54

Piotr Balcerowicz
SVIT = Anantavirya: Siddhi-viniscaya-tikā. Mahendrakumār Jain
(ed.): Siddhi-viniscaya of Akalarka edited with the commentary Siddhi-viniscaya-tikā of Anantavīrya. 2 Vols. Bharatiya
Jñānapitha Prakāśana, Vārāṇasi 1959. SVR = Vādideva-sūri: Syād-vāda-ratnâkara. Motilāl Lālāji (ed.):
Śrīmad-Vādideva-sūri-viracitaḥ Pramāņa-naya-tattvâlokâlankāraḥ tad-vyākhyā ca Syād-vāda-ratnâkaraḥ. 5 Vols., Poona 1926-1930
[reprint: 2 Vols., Bhāratiya Buk Kārporeśan, Dilli (Delhi) 1988). ŚVVTT = Umveka Bhatta: Śloka-vārttika-vyākhyā-tātparya-tikā. S. K.
Ramanatha Sastri (ed.): Slokavārtikavyākhyā-tātparyaţikā of Umveka Bhatta. Revised by K. Kunjunni Raja and R. Thangaswamy, Madras
University Sanskrit Series 13, University of Madras, Madras 1971. TBh = Umāsvāti: Tattvārthādhigama-bhāșya. See TSi. TBV = Abhayadeva-sūri's Tattva-bodha-vidhāyini. See: STP. TILLEMANS 2000 = Tillemans, Tom J.F.: Dharmakirti's Pramānavārt
tika. An annotated translation of the fourth chapter (parārthānumāna). Volume 1 (k. 1-148). Verlag der Österreichis
chen Akademie der Wissenschaften, Wien 2000. TRD = Gunaratna-sūri: Tarka-rahasya-dipikā. Luigi Suali (ed.): Şad
darśana-samuccaya with Gunaratna's Commentary Tarkarahasyadipikā. Bibliotheca Indica 167, Calcutta 1905-1914
[reprinted: 1986). TS = Umāsvāmin: Tattvārtha-sūtra. (1) M.K. Mody (ed.):
Tattvārthādhigama by Umāsvāti being in the Original Sanskrit with the Bhāșya by the author himself. Bibliotheca Indica Nos. 1044, 1079, 1118, Calcutta 1903, 1904, 1905. (2) Mahendra Kumar Jain (ed.): Tattvārtha-vārttika Rāja-vārttika) of Sri Akalarkadeva. Edited with Hindi Translation, Introduction, appendices, variant readings, comparative notes etc. Parts I-II. First edition, Jñānapitha Mārtidevi Jaina Grantha-mālā 10, 20
(Sanskrit Grantha), Delhi 1953-1957. [2. ed.: Delhi 1982]. TSa = śāntarakṣita: Tattva-sangraha. (1) Embar Krishnamacharya
(ed.): Tattvasangraha of śāntarakṣita with the commentary of Kamalasīla. 2 Vols., GOS, 1926 [reprinted: 1984, 1988). (2)

Page #55

On the Relationship of the Nyāyâvatāra and the Sammati-tarka-prakarāņa
83
Dvarikadas Shastri (ed.): Tattvasangraha of Ācārya Shāntarakṣita with the Commentary ‘Pañjikā' of Shri Kamalashila. 2. Vols.,
Vārāṇasã 1981-1982. Thāņ = Thānamga-sutta. (1) Muni Jambāvijaya (ed.): Thāņaṁga
suttaṁ and Samavāyāmga-suttam (Sthānânga-sūtra and Samavāyânga-sūtra). Jaina-Agama-Series 3, Shri Mahāvīra Jaina Vidyālaya, Bombay 1985. (2) Sāgarānanda Sūri (ed.): Thāṇamgasuttam And Samavāyāṁgasuttam with the Vrtti of Ācārya Abhayadeva Sūri. Lala Sundarlal Jain Agama
granthamālā 2, Motilal Banarsidass, Delhi 1985. UPADHYE 1971 = Upadhye, A. N.: Siddhasena Divākara's
Nyāyāvatāra (edited by the late S.C. Vidyabhusan with English Translation, Notes etc.) and with the Vivrti of Siddharşi as well as The Text of 21 Dvātrimśikās and the Sammai-suttam; Vinayavijaya's Nayakarņikā edited by... with and Introduction, Bibliographic Review, Indices etc. Jaina Sāhitya Vikāsa Mandala,
Bombay 1971. VABh = Jinabhadra-ganin Kșamāśramaņa: Višeșāvasyaka-bhāsya.
Dalsukh Malvania (ed.): Acārya Jinabhadra's Viseșāvasyakabhāsya with Auto-commentary; Part I-III, Lalbhai Dalpatbhai Series 10, 14, 21, Bharatiya Sanskrit Vidyamandir - L.D. Institute
of Indology, Ahmedabad 1966, 1968, 1968. Viy = Viyāha-pannatti Bhagavai-viyāha-pannatti. Bechardas J. Doshi;
Amritlal Mohanlal Bhojak (ed.): Viyāhapannattisuttam. JainaĀgama-Series No.4 Part I-III, Shri Mahāvīra Jaina Vidyālaya,
Bombay 1974, 1978, 1982. VS = Kanāda: Vaiseșika-sūtra. With the Commentary of
Candrānanda, ed. Muni Jambāvijayaji, GOS 136, Baroda 1961.

Page #56
_