Reverberations of Dharmakirti’s Philosophy
by Birgit Kellner | 2020 | 264,305 words
This page relates ‘Meaning of Bahyartha’ of the study on the philosophy of Dharmakirti (6th century) and his predecessor Dignaga (5th century). This collection of articles reflects philosophical currents in India, China and Tibet during their time and investigates the Buddhist theories of Pramana (“instruments of trustworthy awareness”).
[Full title: The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference by Kiyokuni Shiga]
This research was supported in part by JSPS KAKENHI Grant Number 24720029 and 25284014. I would like to express my gratitude to Prof. Kei Kataoka for having given me valuable comments after the conference.
1. Introductory remarks
In an earlier paper, “Remarks on the Origin of All-Inclusive Pervasion” (henceforth Shiga 2011b), I proposed the hypothesis that the theory of all-inclusive pervasion (sarvopasaṃhā-ravyāpti), considered to have been created by Dharmakīrti, in fact originated in Dignāga’s theory of inference.
The statement of Dignāga’s in question, which appears in the third chapter of the Pramāṇasamuccayavṛtti (henceforth PSV), is the following:
To be more precise, the co-existence [of a logical reason] with such an [ob-ject to be proved] is understood by means of two [types of] exemplification based on similarity or dissimilarity, [in] which [the property to be proved] is associated[1] with external thing[s[2] that respectively have the proving property] (bāhyārthopasaṃhṛtena).[3]
This statement coincides with Dharmakīrti’s formulation of all-inclusive pervasion in his Hetubindu4 (henceforth HB). Considering various factors, I concluded that Dignāga and Dharmakīrti shared the idea that pervasion is established or confirmed by taking individual cases5 into account.6
As I mentioned in Shiga 2011b, the term bāhyārtha also appears several times in the fourth chapter of the Pramāṇasamuccaya (henceforth PS) and the PSV in the expressions bāhyārthānupasaṃhāra, bāhyārthapradarśana, bāhyārthāpekṣā, as well as bāhyārtho-pasaṃhṛta. These examples of its use need to be examined in a new light to clarify the relationship between bāhyārtha and dṛṣṭānta. This paper aims to further investigate various issues that were not settled in Shiga 2011b, including the following: (1) What do Dignāga and Jinendrabuddhi mean by the term bāhyārtha used in logical contexts, mainly in the PS and PSV 4? (2) Does it literally denote “an external object,” that is, “a thing in the external world as an object of cognition,” or does it denote “something external to the subject or the property-bearer to be proved,” as Jinendrabuddhi understands it? (3) In this case, does Jinendrabuddhi deviate from Dignāga’s original intention by adopting Dharmakīrti’s theories when he interprets Dignāga, or are his comments on the term bāhyārtha in keeping with Dignāga’s intention? (4) What is the relationship between the terms bāhyārtha and bahirvyāpti?
2. Usage examples for the term bāhyārtha in the PS/PSV 4
2.1 Regarding the role of dṛṣṭānta7
First, we will observe an example of Dignāga’s use of the term bāhyārtha in PS/PSV 4, where he defines the role of dṛṣṭānta.[4]
Reconstructed text[5] of PSV on PS 4.3 (= Appendix [2]):[6]
evaṃ tarhi ghaṭo’nudāharaṇam. tatra hi yathā hetuḥ sādhyānugataḥ, tathā sādhyam api hetvanugatam iti cet. na, avivakṣitatvāt [yathā][7] kṛtakatvavi-śeṣasya hetutvena, tathā ghaṭe sādhyānugamasya. bāhyārthapradarśanaṃ hi nidarśye pradhānam.[8]
[Question:] If so (= if the example on the basis of similarity is defined as a thing in which a logical reason is accompanied by what is to be proved),[9] a pot would not [constitute a valid] exemplification, because in the [pot], just as a logical reason is accompanied by what is to be proved, what is to be proved is [also] accompanied by the logical reason.
[Answer:] This is not [right], because just as being produced as a particular property is not intended to be the logical reason, what is to be proved is [not intended to be] accompanied [by the logical reason] in a pot[, which is a particular thing, even though that is the case],[10] for the primary [role] of exemplification is to indicate external thing[s].[11]
Jinendrabuddhi comments on the term bāhyārtha as follows:
External thing[s] mean [those things] in general that are [similar] in kind to what is to be proved, which are [external] to a particular [thing, that is, a particular property-bearer] such as a pot. [The expression bāhyārthaprada-rśanam, i.e.] “to manifest external thing[s]” means to generally manifest the pervasion of a logical reason by what is to be proved in the [external things in the following way:] “Whatever is produced is all necessarily impermanent.”[12]
Here, Jinendrabuddhi interprets bāhyārtha differently from how he commented on the phrase bāhyārtha (-upasaṃhṛta) in the third chapter, where he interpreted bāhyārtha as “elsewhere than in a particular [property-bearer] that is made to be the subject, i.e., the [property-bearer] in general” (dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānye).[13] This gloss shows the contrast between the inside and the outside of the subject (pakṣa), and the contrast between particular and generalized property-bearers.
The above quotation, on the other hand, suggests that the term bāhyārtha does not mean a particular thing such as a pot, which is one of similar examples, but the generality of things that are similar in kind to what is to be proved (sādhyajātīyasāmānya). In this case, particularity (= an internal thing) is contrasted to generality (= external things). Jinendrabuddhi focuses on the division between an example as an individual thing and a generalized example, but not on the division between a thing inside the subject and things outside the subject.
Next, let us look into the compound bāhyārthapradarśana. According to the Pramā-ṇasamuccayaṭīkā, the first half bāhyārtha is to be interpreted in the locative sense, i.e. as “in external thing[s],” and the word vyāpti is complemented as an object of the last half-pradarśana. Thus, the compound as a whole means “indicating the pervasion in external thing[s].” Elsewhere in the same chapter, Jinendrabuddhi presents another interpretation with regard to bāhyārtha. He understands dṛṣṭānta as meaning an assembly of things (artha-rāśi), but not as a particular individual thing (vyaktiviśeṣa). And that particular individual thing is described as being included in the assembly of things.[14] For Jinendrabuddhi, the term bāhyārtha means “something external to a single and particular individual thing,” or “an assembly of similar things.”
It is worth mentioning in passing that an opponent paraphrases this role of dṛṣṭānta as: “This exemplification depends on external thing[s]” (*bāhyārthāpekṣam idaṃ nidarśa-nam).[15] Jinendrabuddhi paraphrases this as “pervasion that depends on external thing[s].”[16]
3. Examination of the instances in PS/PSV 1
As Katsura (2004) points out, there is still a possibility that the term bāhyārtha in bāhyā-rthapradarśana means “an object in the external world” and implies “some positive support in external reality.”[17] The expression bāhyārtha is used at least nine times in PS/PSV 1. An examination of these instances (PSV 1 4, 8; 18, 13; 18, 25; 19, 2; 19, 6; 19, 11; 19, 15; 19, 16; 19, 18) reveals that all of them mean “an external object” or “an object in the external world.”[18] It is easily understood that the counterpart of bāhyārtha is “an internal thing,” such as cognition, knowledge or the mind.
However, if the term bāhyārtha, which is used in logical contexts in PS/PSV 3 and 4, means “an external object,” we encounter certain difficulties. First, Dignāga uses the example buddhi in his formulation of this inference: “Sound is permanent, because it is intangible, like cognition.”[19] It is obvious that this buddhi is not a so-called ‘external object.’
Although the formulation that contains the example buddhi is presented as fallacious, this is not because buddhi is not “an external object,” but because buddhi is a pseudo-example that does not have the property to be proved (nityatva).[20]
Furthermore, if we take a close look at the expression bāhyārtha, the word artha itself can mean “an actual thing” with some factual basis.[21] So it is likely that the word bāhya is used in the sense of “outside” or “external [to the subject]” and artha is used in the sense of “actual example” or a thing that can be verbally expressed or named, i.e., in the sense of padārtha.[22]
4. The expression bāhyārtha-upasaṃhāra or -anupasaṃhāra
4.1 A usage example found in the criticism of the Naiyāyika definition of exemplifi-cation (udāharaṇa)
Next, we shall examine cases where the word upasaṃhāra or anupasaṃhāra comes after bāhyārtha.
[Even] if [an opponent states that] the example is qualified by the first half [of the definition of udāharaṇa in NS 1.1.36],[23] the similarity or dissimilarity to the [property-bearer] to be proved does not need to be mentioned [as a part of the definition of example], because:
[If] so, it (= the example) is not seen as what conveys the knowledge [of what is to be proved]. (PS 4.19b)
To wit, when [the example] does not associate [the property to be proved] with external thing[s that respectively have the proving property] (bāhyārthānupa-saṃhāre), the example does not convey the knowledge of the property to be proved. Alternatively, for that reason, [the necessity of mentioning similarity or dissimilarity] is not established, [because similarity and dissimilarity are implied] just by [the qualifier:] “having the property of the [property-bearer to be proved]” (taddharmabhāvitvena).[24]
This is found in the criticism of the Naiyāyika definition of the exemplification (udāha-raṇa) in the paramatas ection of PS/PSV 4. Dignāga asserts that if dṛṣṭānta “does not associate [the property to be proved] with external thing[s that respectively have the proving property],”[25] such dṛṣṭānta cannot prove what is to be proved. This suggests that bāhyārtho-pasaṃhāra is an indispensable condition for dṛṣṭānta to prove what is to be proved. Just before this assertion, he also states that if dṛṣṭānta is treated as separate from the content of the logical reason, a fallacy would occur: a certain thing (e.g. space) could become both a similar and dissimilar example because of its similarity (e.g. being existent) and dissimilarity (e.g. being inaudible) to the property-bearer to be proved (e.g. sound).[26]
Jinendrabuddhi comments on the use of the words bāhyārtha and upasaṃhāra in the above quotation as follows:
[The word] bāhyārtha in [the phrase] na hi bāhyārthānupasaṃhāre means another thing that is different from mere similarity or dissimilarity. Further-more, it (= bāhyārtha) [actually] means a pervasion or inseparable relation (avinābhāvitva). The association (upasaṃhāra) of it (= bāhyārtha = avi-nābhāvitva) means applying (upanayana), i.e. manifesting (prakāśana) that [bāhyārtha] is an object to be explained. When it (= upasaṃhāra) is not present, the example is not what conveys the knowledge [of what is to be proved].[27]
Jinendrabuddhi understands bāhyārtha as “another thing that is different from mere simi-larity or dissimilarity,”[28] then regards it as “pervasion” or “inseparable relation.” These interpretations are unique, in that he takes bāhya in the sense of “different” and then identifies it as pervasion.[29]
In any case, regarding the word bāhyārtha, Jinendrabuddhi calls attention to the fact that dṛṣṭānta does not consist of the combination of an actual example (e.g. a pot) and its mere similarity (e.g. being produced)[30] or dissimilarity (e.g. not being produced) to the subject (e.g. sound), but instead the combination of dṛṣṭāntadharmin and the inseparable relation of a logical reason (e.g. being produced) with what is to be proved (e.g. impermanence).
4.2 Usage examples regarding the Naiyāyika definition of the applica-tion (upanaya)
Here we will examine the uses of bāhyārtha and bahiḥ found in the criticism[31] against the Naiyāyika definition of the application (upanaya).[32]
On the other hand,[33] the application [of the property to be proved and/or the proving property] to [a dissimilar example such as] space [in the inference: “Sound is impermanent, because it is produced”] by depending on external thing[s] is correct, because, unlike space [and so on], there is nothing that is produced [and] permanent[34] outside [the subject, i.e. sound], whereas, like a pot [and so on], there is something [produced] and impermanent [outside the subject, i.e. sound]. Therefore [sound that is produced] is impermanent. Hence, [the application] is [valid only when it is made] with regard to certain external case[s].39 40
As we have already seen, both Dignāga and Jinendrabuddhi recognize “associating [the property to be proved or inseparable relation] with external thing[s]” (bāhyārthopasaṃ-hāra) as an essential factor of dṛṣṭānta for a valid inference. And here “external thing[s]” (bāhyārtha) are restated as being “outside” (bahiḥ), and space and a pot are given as actual examples. According to Jinendrabuddhi, common absence (vyatireka) is indicated in space, whereas common occurrence (anvaya) is indicated in a pot. And the word bahiḥ is mentioned “for the purpose of negating the dependence on the mere [fact that the property to be proved and/or the proving property] is present only in a pot and absent only in space,”[35] which implies that bāhyārtha does not refer to a single case, but plural cases. Indeed, on the basis of this passage alone, we cannot decide whether bāhyārtha means “an object in the external world” or “external things outside [the subject],” but at least Jinendrabuddhi’s comments consistently support the meaning of “external thing[s] outside [the subject].”[36]
5. bāhyārtha and the term and notion of bahirvyāpti
As I observed in Shiga 2011b, the expression anyatra, which means “in another place [than the location of the property-bearer to be proved],” is found in PS 2.11.[37] Dignāga himself comments on this word, presenting a remarkable idea. He introduces the concept of “the generalized substratum” (ādhārasāmānya)[38] .[39] Jinendrabuddhi glosses the word anyatra as “generally (sāmānyena) in all cases, e.g. a kitchen.”[40] He also uses the words anyatra and sāmānya when he comments on the expression bāhyārthopasaṃhṛta in PSṬ
3.[41] From this fact, we can safely say that Jinendrabuddhi, for one, sees the terms bāhyārtha and anyatra as referring to substantially the same thing: that is, property-bearer[s] of the example outside the subject. It is worth noting that he also states that a logical reason (e.g. smoke) is inseparably related not only with the property to be proved (e.g. fire), but also with the generalized substratum (e.g. any place that has fire), because a property-bearer depends on its property, and the property to be proved is particularized by the generalized substratum.[42]
If these observations thus far are valid, it follows that bāhya in the term bāhyārtha and bahir-in the term “external pervasion” (bahirvyāpti) have the same logical value. Dignāga is sometimes described as a “typical” bahirvyāptivādin by modern scholars, presumably because he constructed the system of inference, especially the theory of trairūpya, which requires presenting the similar (sapakṣa) and the dissimilar (vipakṣa) as separate from the subject (pakṣa).[43] Nevertheless, Dignāga does not declare himself a proponent of bahirvyāpti. In Buddhist treatises on logic and epistemology, the term bahirvyāpti first appeared in Arcaṭa’s HBṬ 62, 9; 62, 23; 62, 27.[44] Śāntarakṣita uses both bahirvyāpti and antarvyāpti in his VNṬ 5, 30–6, 7. It should be noted here that Arcaṭa and others understand bahirvyāpti as “pervasion that is indicated ‘only’ outside the property-bearer to be proved, that is, ‘only’ in a property-bearer of the example,” but this is different from Dignāga’s view. Dignāga does not state that pervasion is indicated ‘only’ outside the property-bearer to be proved. Rather, he devised the notion of “the generalized property-bearer” (ādhārasāmānya), which can implicitly include the property-bearer to be proved (sādhyadharmin, pakṣa).[45]
Pātrasvāmin (7–8 cent.),[46] a Jaina logician who is considered to have been the first to advocate the theory of the single characteristic (ekalakṣaṇa) of a logical reason, which is virtually identical to the theory of antarvyāpti,[47] states that the inseparable relation (avinābhāvitva) asserted by Buddhists is acknowledged outside sādhya or in dṛṣṭānta, whereas “being otherwise impossible” (anyathānupapannatva) is only acknowledged in the property-bearer to be proved.[48] Pātrasvāmin’s view was followed by other Jaina logicians such as Akalaṅka.[49] They regarded themselves as proponents of antarvyāpti and the Buddhists as proponents of bahirvyāpti.[50] Therefore, it seems reasonable to suppose that Arcaṭa and other successors of Dharmakīrti did not intend to criticize Dignāga’s view directly, but rather to point out that Dignāga was allegedly credited with creating the theory of the so-called bahirvyāpti.
6. Conclusion
We can safely state that, as long as we do not disregard Jinendrabuddhi’s annotations,[51] when Dignāga uses the term bāhyārtha in connection with upasaṃhāra, pradarśana or apekṣā in the context of inference, he does not mean “an external object” or “an object in the external world,”[52] but “external thing[s] outside the subject.” It is possible to suppose that Dignāga was well aware of the division between the inside and outside of the subject or property-bearer to be proved.
More specifically, the term bāhyārtha found in PS/PSV 3 and 4 refers to (1) “[external] things, that is, the generalized [property-bearer], which is other than a particular [property-bearer] that is made to be the subject” (PSṬ B146a2, Appendix [1]); (2) “the generalized [property-bearer] that is similar in kind to what is to be proved, which is external to a particular [property-bearer] such as a pot” (PSṬ B171b5, Appendix [2]); (3) “an [external] thing that is different from mere similarity or dissimilarity,” that is, “pervasion or inseparable relation” (PSṬ B185a2, Appendix [4]); (4) more than simply a single example such as a pot or space (PSṬ B190a3, Appendix [8]). Apart from (3), Jinendrabuddhi’s interpretations do not seem to deviate from Dignāga’s original intention.
Due to his trairūpya-theory, Dignāga is often regarded as the proponent of bahirvyāpti. What he intended to maintain, however, might not be the so-called position of bahirvyāpti that means, according to Arcaṭa, the pervasion that is to be grasped or observed ‘only’ in a property-bearer of the example, that is, outside the subject. Rather, it is probable that Dignāga, like Dharmakīrti, assumed some sort of generality or universality regarding pervasion and its substratum.
As for pervasion, on the other hand, Dignāga asserts the logical predominance of the common absence (vyatireka) over the common occurrence (anvaya),[53] and states that vyatireka can be grasped or understood by mere non-observation (adarśanamātra).[54] Indeed, the pervasion based on Dignāga’s system appears to be hypothetical, because pervasion would not be valid if even a single counter-example is found, but it is likely that he thought that pervasion should be universal once it was established.[55] We can assume that Dignāga thought that this universal pervasion should be applied to external cases outside the subject. In addition, since the subject is supposed to be implicitly included in those external cases, it is possible to apply the pervasion to the subject and infer what is to be proved.
Table 1: Dignāga’s usage of bāhyārtha in logical contexts, with Jinendrabuddhi’s comments (PS/PSV/PSṬ 3 and 4)
| No. | Dignāga | Jinendrabuddhi |
| [1] | PSV on PS 3.36b:62 tādṛksāhabhavyaṃ hi sādharmyeṇa vaidharmyeṇa vā bā-hyārthopasaṃhṛtena dṛṣṭāntadvayena gamyate. (PSV(K) P138a5–7; PSV(V) D51a6–b1; P54b5–7) | PSṬ B146a1–3: darmiṇaḥ pakṣīkṛtād vi-śeṣād anyatra sāmānya upadarśitenety arthaḥ. sāmānyasya ca viśeṣāparityā-gāt, sādhyadharmy api tatrāntargata eva. sādhyadharmiṇy evāvinābhāvitva-pradarśananirāsaparaṃ tu bāhyārtha- grahaṇam. |
| [2] | PSV on PS 4.3: bāhyārthapradarśa-naṃ hi nidarśye pradhānam. (PSV(K) P148b6; PSV(V) D60b2f.; P64a7) | PSṬ B171b5–6: bāhyārthapradarśa-naṃ hīti ghaṭāder viśeṣāt sādhyajātī-yasāmānyaṃ bāhyo’rthaḥ, tatra sāmā-nyena yan nāma kiṃcit kṛtakaṃ tat sa-rvam anityam eveti sādhyena hetor vyā-ptiprakāśanaṃ bāhyārthaprakāśanam. See also PSṬ B171b6–172a2: tasmāt sāmānyena sarvo yathoktadṛṣṭāntala-kṣaṇo’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva; PSṬ B171b6–172a2: bāhyārtha-pradarśanaṃ hi nidarśye pradhānam iti vacanam ekatraiva vyaktiviśeṣe pra-darśanasya prādhānyanirāsaparaṃ dra-ṣṭavyam. |
| PSV on PS 4.4: bāhyārthāpekṣam idaṃ nidarśanam. (PSV(K) P149b3; PSV(V) D60b6; P64b3) | No direct comment. Cf. PSṬ B173b6: na hi sadbhāvamātradarśane bāhyā-rthāpekṣatvaṃ nidarśanasya yujyate; PSṬ B173b6–174a1: tataś ca ghaṭe’ni-tyatvaṃ prayatnānantarīyakatvaṃ ca dṛṣṭam iti kṛte hetoḥ sapakṣe gamyata eva sadbhāvamātram iti kim atra bāhyā-rthāpekṣayā vyāptyaiveti ... | |
| PSV on PS 4.19: na hi bāhyārthānupasaṃhāre dṛṣṭāntena sādhyadharmo gamyate. (PSV(K) P153b7; PSV(V) D64b1; P68a7) | PSṬ B185a1–3: na hi bāhyārthānupa-saṃhāra iti sādharmyavaidharmyamā-trād anyo’rtho bāhyārthaḥ. sa punar vyāptir avinābhāvitvam iti yāvat. | |
| PSV on PS 4.20: yadi bhavatāṃ sādhyasādhanadharmavatā dṛṣṭāntenārthaḥ, na tūbhayadharmavatāpi bāhyārthānupasaṃhāreṇa sādhyo’rthaḥ śakyo bhāvayitum ity uktam. (PSV(K) P154a6; PSV(V) D64b6; P68b5f.) | No direct comment. Cf. PSṬ B186a3–4: na tv etāvatā vyāptiprakāśanam anta-reṇa sādhyo’rthaḥ śakyo gamayitum ity uktaṃ prāk; PSṬ B186a4–6: na hi vastudharmavatāpi dṛṣṭāntena vinā bā-hyārthopasaṃhāreṇa sādhyo’rthaḥ śa-kyo bhāvayitum. | |
| PSV on PS 4.20: na ca dṛṣṭāntasādhyayor ekīkaraṇād evaṃ anityatvaṃ siddham, bāhyārthānupasaṃhā-rāt. (PSV(K) P154b8; PSV(V) D65a6f.; P69a6) | No direct comment. | |
| PSV on PS 4.20: bāhyārthāpekṣayā tv ākāśasyopanayo yuktaḥ. (PSV(K) P155b4f.; PSV(V) D66a1f.; P70a1f.) | No direct comment. See the example [8]. | |
| PSV on PS 4.20: yasmād ākāśavad bahiḥ, na hi kiṃcin nityaṃ kṛtakam asti, anityaṃ cāsti ghaṭavat. tasmād anityam iti kvacid bahir asti. (PSV(K) P155b5f.; PSV(V) D66a2; D70a2f.) | PSṬ B190a2–3: kathaṃ bāhyā-rthāpekṣa ity āha: yasmād ityādi. kevalaghaṭākāśasadasattvamātrāpe-kṣatvanirāsaparaṃ bahirgrahaṇam. See also PSṬ B190a3–5: anvayavyati-rekaṃ cāpekṣya tatpradarśanārtham ākāśavad ghaṭavac cety ekadeśa udāharaṇamātram upanīyate. |
References and abbreviations
Primary Sources
HB Hetubindu (Dharmakīrti): Ernst Steinkellner, Dharmakīrti’s Hetubinduḥ, Teil I: Tibe-tischer Text und rekonstruierter Sanskrit-Text. Vienna 1967.
HBṬ Hetubinduṭīkā (Arcaṭa): Hetubinduṭīkā of Bhaṭṭa Arcaṭa with Sub-Commentary Entitled Āloka of Durveka Miśra, ed. S. Sanghavi and Jinavijayaji. Baroda 1949.
HBṬ(V) Hetubinduṭīkā (Vinītadeva): (Tib.) D4234, P5733.
NA Nyāyāvatāra (Siddhasena Divākara): Jaina epistemology in historical and comparative perspective: critical edition and English translation of logical epistemological trea-tises: Nyāyâvatāra, Nyāyâvatāra-vivṛti and Nyāyâvatāra-ṭippana with introduction and notes, ed. P. Balcerowicz. 2 vols. Stuttgart 2001.
NAVi Nyāyāvatāravivṛti (Siddharṣi): See NA.
NB Nyāyabindu (Dharmakīrti): See NBṬ
NBṬ Nyāyabinduṭīkā (Dharmottara): Paṇḍita Durvekamiśra’s Dharmottarapradīpa [Be-ing a sub-commentary on Dharmottara’s Nyāyabinduṭīkā, a commentary on Dhar-makīrti’s Nyāyabindu], ed. Dalsukhbhai Malvania. Patna[56] 1971.
NMukh Nyāyamukha (Dignāga): See Katsura 1981.
NP Nyāyapraveśakasūtra (Śaṅkarasvāmin): Nyāyapraveśakaśāstra of Baudh Ācārya Di-ṅnāga (The father of the Buddhist Logic). With the commentary of Ācārya Hari-bhadrasūri and with the subcommentary of Pārśvadevagaṇi, ed. Muni Jambuvijaya. Delhi/Ahmedabad 2009.
NS Nyāyasūtra (Gautama): Nyāyadarśanam with Vātsyāyana’s Bhāṣya, Uddyotakara’s Vārttika, Vācaspati Miśra’s Tātparyaṭīkā and Viśvanātha’s Vṛtti, ed. Taranatha Nyaya-Tarkatirtha and Amarendramohan Tarkatirtha. Calcutta 1936.
NV Nyāyavārttika (Uddyotakara): See NS.
NVTṬ Nyāyavārttikatātparyaṭīkā (Vācaspatimiśra): See NS.
PS Pramāṇasamuccaya (Dignāga): See PSV.
PS/PSV 1 Pramāṇasamuccaya/Pramāṇasamuccayavṛtti (Dignāga), chapter 1: See Stein-kellner 2005.
PSaṃ Pramāṇasaṃgraha (Akalaṅka): Akalaṅka Granthatrayam (Svopajñavivṛtisahitam Laghīyastrayam, Nyāyaviniścayaḥ, Pramāṇasaṅgrahaś ca) of Śrī Bhattākalaṅkadeva, ed. Mahendra Kumar Shastri. Ahmedabad/Calcutta 1939.
PSṬ Viśālāmalavatī Pramāṇasamuccayaṭīkā (Jinendrabuddhi): (Tib.) D4268; P5766.
PSṬ 1 Pramāṇasamuccayaṭīkā (Jinendrabuddhi), chapter 1: Jinendrabuddhi’s Viśālāma-lavatī Pramāṇasamuccayaṭīkā. Chapter 1. Part I: Critical Edition, ed. E. Steinkellner, H. Krasser, and H. Lasic. Beijing/Vienna 2005.
PSṬ 2 Pramāṇasamuccayaṭīkā (Jinendrabuddhi), chapter 2: Jinendrabuddhi’s Viśālāma-lavatī Pramāṇasamuccayaṭīkā. Chapter 2. Part I: Critical Edition, ed. H. Lasic, H. Krasser, and E. Steinkellner. Beijing/Vienna 2012.
PSV(K) Pramāṇasamuccayavṛtti (Dignāga) translated by Kanakavarman: (Tib.) P5702. See also Kitagawa 1965: 440–579.
PSV(V) Pramāṇasamuccayavṛtti (Dignāga) translated by Vasudhararakṣita: (Tib.) D4204; P5701. See also Kitagawa 1965: 440–579.
PV Pramāṇavārttika (Dharmakīrti): See PVSV.
PVSV Pramāṇavārttikasvavṛtti (Dharmakīrti): The Pramāṇavārttikam of Dharmakīrti. The First Chapter with the Autocommentary, ed. R. Gnoli. Roma 1960.
SVin Siddhiviniścaya (Akalaṅka): Siddhiviniścayaṭīkā of Anantavīryāchārya: The com-mentary on Siddhiviniścaya and its vṛtti of Bhaṭṭa Akalaṅkadeva, ed. Mahendra Kumar Jain. 2 vols. Varanasi 1959.
TSP Tattvasaṃgrahapañjikā (Kamalaśīla): Tattvasaṅgraha of Ācārya Shāntarakṣita with the Commentary ‘Pañjikā’ of Shri Kamalashīla, ed. S. D. Shastri. 2 vols. Varanasi 1968.
VNṬ Vādanyāyaṭīkā (Śāntarakṣita): Dharmakīrti’s Vādanyāya: With the Commentary of Śāntarakṣita, ed R. Sāṅkṛtyāyana. Appendix to Journal of the Bihar and Orissa Reseach Society: New Series 21–22 (1935–36).
YD Yuktidīpikā: The most significant commentary on the Sāṃkhyakārikā, ed. A. Wezler and Sh. Motegi. Vol. 1. Stuttgart 1998.
Secondary Sources
Frauwallner 1959 E. Frauwallner, Dignāga, sein Werk und seine Entwicklung. Wiener Zeitschrift für die Kunde Süd-und Ostasiens 3 (1959) 83–164.
Funayama 2001 T. Funayama, On the date of Vinītadeva. In: Le parole e i marmi, studi in onore di Raniero Gnoli nel suo 70゜compleanno, ed. Raffaele Torella. Rome 2001, 309–325.
Jambuvijaya 1966 Muni Jambuvijaya, Dvādaśāraṃ Nayacakram of Āchārya Śrī Malla-vādi Kṣamāśramaṇa. With the commentary Nyāyāgamānusāriṇī of Śrī Siṃhasūri Gaṇi Vādi Kṣamāśramaṇa. Part I. Bhavnagar 1966.
Kataoka 2012 K. Kataoka, Gengo-tetsugaku–Apoha-ron [Linguistic Philosophy–Apoha Theory]. In: Ninshiki-ron to ronri-gaku [Epistemology and Logic], ed. Sh. Katsura, A. Saito, M. Shimoda, and F. Sueki. Tokyo 2012, 189–226.
Katsura 1981 Sh. Katsura, Inmyō shōrimonron kenkyū (4) [A Study of Nyāyamukha]. Hiroshima Daigaku Bugakubu Kiyō 41 (1981) 62–82.
Katsura 2004 Sh. Katsura, The Role of dṛṣṭānta in Dignāga’s logic. In: The Role of the Example (dṛṣṭānta) in Classical Indian Logic, ed. E. Steinkellner and Sh. Katsura. Vienna 2004, 135–174.
Kitagawa 1965 H. Kitagawa, Indo-koten-ronrigaku no kenkyū –Jinna (Dignāga) no taikei –[A study on classical Indian logic –Dignāga’s system]. Tokyo 1965.
Matilal 1985 B. K. Matilal, Logic, Language and Reality. An introduction to Indian Philosophical Studies. Delhi 1985.
Pind 1999 O. H. Pind, Dharmakīrti’s Interpretation of Pramāṇasamuccayavṛtti V 36: śabdo’rthāntaranivṛttiviśiṣṭān eva bhāvān āha. In: Dharmakīrti’s Thought and Its Impact on Indian and Tibetan Philosophy, Proceedings of the Third International Dharmakīrti Conference Hiroshima, November 4–6, 1997, ed. Sh. Katsura. Vienna 1999, 317–332.
Pind 2011 O. H. Pind, Dignāga’s Apoha Theory: Its Presuppositions and Main Theoretical Implications. In: Apoha. Buddhist Nominalism and Human Cognition, ed. M. Siderits, T. Tillemans, and A. Chakravarti. New York 2011, 64–83.
Pind 2015 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamucca-yavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E. Steinkellner. Vienna 2015.
Shiga 2011a K. Shiga, antarvyāpti and bahirvyāpti re-examined. In: Religion and Logic in Buddhist Philosophical Analysis, Proceedings of the Fourth International Dharmakīrti Conference Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011, 423–435.
Shiga 2011b K. Shiga, Remarks on the Origin of All-inclusive Pervasion. Journal of Indian Philosophy 39 (2011) 521–534.
Steinkellner 2004 E. Steinkellner, The Early Dharmakīrti on the Purpose of Examples. In: The Role of the Example (dṛṣṭānta) in Classical Indian Logic, ed. E. Steinkellner and Sh. Katsura. Vienna 2004, 225–251.
Steinkellner 2005 E. Seinkellner, Dignāga’s Pramāṇasamuccaya, Chapter 1: A hypothet-ical reconstruction of the Sanskrit text with the help of the two Tibetan translations on the basis of the hitherto known Sanskrit fragments and the linguistic materials gained from Jinendrabuddhi’s Ṭīkā. 2005. https://www.oeaw.ac.at/fileadmin/Ins titute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf, last visited 10-12-2019.
Footnotes and references:
[1]:
As we will see below, Jinendrabuddhi understands the word upasaṃhṛta or upasaṃhāra as forming part of the compound bāhyārthopasaṃhṛta or bāhyārthānupasaṃhāra as upadarśita (PSṬ D185b3; P211a7; Skt. B146a2), upanayana or prakāśana (PSṬ D230a3; P260b5; Skt. PSṬ B185a2). Moreover, he interprets the word upasaṃhāra as “associating/bringing the [property] to be proved with/near all the proving [properties]” elsewhere in the same chapter (PSṬ D227b1; P257b7f.: nye bar bsdu ba ni med na mi’byung ba nyid de[ste P] / thams cad du sgrub par byed pa la bsgrub par bya ba nye bar bsdu ba ste / nye bar’dren pa’o // Skt. B182b6f.: upasaṃhāro ’vinābhāvitvam, sarvatra sādhane sādhyasyopasaṃharaṇam upanayaḥ). Therefore, it would be more appropriate in this case to translate upasaṃhṛta as “associated,” “applied” or “brought near” than as “summed up” (Shiga 2011b: 525 with n. 14) or “included,” as in the case of Dharmakīrti’s term sarvopasaṃhāra (HB 5*, 21), which could be translated as “including” or “summing up [the property to be proved in] all the [property-bearers that respectively have the proving property]” (HBṬ 62, 18–20: sarvasmin sādhanadharmavati dharmiṇi, na dṛṣṭāntadharmiṇy eva, sādhyadharmasyopasaṃharaṇam upasaṃhāro ḍhaukanaṃ …). Meanwhile, it should be noted here that Arcaṭa glosses upasaṃhāra as ḍhaukana (“bringing [the property to be proved] near [all the property-bearers that respectively have the proving property]”). Taking this interpretation into consideration, it is probable that the word upasaṃhāra, even in the compound sarvopasaṃhāra, could mean “associating” or “applying” rather than “summing up” or “including.”
[2]:
According to Jinendrabuddhi, bāhyārtha is to be understood not as a single and particular individual (PSṬ D215a4f.; P244b2; Skt. B171a2f. ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ draṣṭavyam), but as a whole assembly of things (PSṬ D215a3; P244a8f.; Skt. B171b7: sāmānyena sarvo yathoktadṛṣṭāntalakṣaṇo’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva) or the generalized property-bearer as opposed to the particular property-bearer to be proved (PSṬ D185b3; P211a7; Skt. B146a2: dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānye ).
[3]:
PSV(K) P138a6f.: de’dra ba’i lhan cig rgyu ba nyid ni phyi rol gyi don nye bar sdud par byed pa chos mthun pa dang mi mthun pa’i dpe gnyis po dag gis rtogs pa yin te / PSV(V) D51a7–b1; P54b6f.: de dang mthun pa dang lhan cig tu rgyu ba nyid dang / chos mthun pa’am chos mi mthun pa nyid kyis[nyid can gyis D] phyi rol gyi don nye bar bsdus nas dpe gnyis kyis go bar byed do // (Skt. reconstruction: *tādṛksāhabhavyaṃ hi sādharmyeṇa vaidharmyeṇa vā bāhyārthopasaṃhṛtena dṛṣṭāntadvayena gamyate.) See also Shiga 2011b: 524–526 with n. 11–18.
[4]:
See Shiga 2011b: 527. In PS 4.2 dṛṣṭānta is defined differently as follows: sādhyenānugamo hetoḥ sādhyābhāve ca nāstitā / khyāpyate yatra dṛṣṭāntaḥ sa sādharmyetaro dvidhā // (See Jambuvijaya 1966:133. This verse is quoted in Daśavaikālikasūtrahāribhadrīvṛtti 34b. Cf. NMukh 11.) “The example is a [thing] in which it is conveyed that a logical reason is accompanied by what is to be proved and that [the logical reason] is absent when what is to be proved is absent. This is in two forms: [the example on the basis of] similarity and the other (= the example on the basis of dissimilarity).” (See also Katsura 2004: 141 with n. 11.)
[5]:
Words in roman type are found in the PSṬ or in the fragments, whereas those in italics are reconstructed. The Sanskrit reconstructions of the Tibetan translations of PS/PSV 3 and 4 cited in this paper, other than a few passages, are part of the achievements of the PSṬ seminar organized and led by Prof. Shōryū Katsura at Ryūkoku University. I would like to specially thank all the participants of this seminar, especially Dr. Yasuhiro Okazaki, who made Sanskrit reconstructions and editions of PS/PSV/PSṬ 4; Dr. Toshikazu Watanabe, who checked and corrected Dr. Okazaki’s draft; Prof. Diwakar Acharya for his valuable comments and suggestions on the Sanskrit reconstructions; and Prof. Katsura for permission to use these results and other relevant materials. Any errors that remain are my own.
[6]:
See also Shiga 2011b: 527 with n. 21–24.
[7]:
No equivalent in PSV(K).
[8]:
PSV(K): P148b4–6: ’on te de ltar na bum pa dper mi bya ste / de la ni ci ltar gtan tshigs bsgrub bya’i rjes su’gro ba de bzhin du / bsgrub bya yang gtan tshigs kyi rjes su’gro ba yin no zhe na / ma yin te / byas pa’i khyad par gtan tshigs su brjod par’dod pa ma yin pa bzhin du bum pa la bsgrub bya rjes su’gro ba brjod pa ma yin pa’i phyir ro // phyi rol gyi don la bstan pa ni dpe la gtso bo yin no // PSV(V) D60b1–3; P64a6f.: gal te’di ltar bum pa dper brjod pa de lta na ni ji ltar gtan tshigs bsgrub bya dang ldan pa de ltar bsgrub bya gtan tshigs dang ldan par’gyur ro zhe na ma yin te / brjod par mi bya ba’i phyir te / bum pa ni gang bsgrub bya’i rjes su’gro ba can gyis[gyi P] byas pa nyid kyi khyad par du byas pa brjod par bya ba’i phyir ro // ngag gi don bstan pa ni nges par bstan pa’i don gtso bo yin no // (See also Katsura 2004: 155 with n.28.)
[9]:
PSṬ D214b5f.; P244a2f.: ’o na de lta na zhes pa / gal te gang du gtan tshigs bsgrub byas rjes su’gro ba yin pa de chos mthun pa nyid kyi dpe’o zhes pa mtshan nyid yin na / (Skt. B171b3: evaṃ tarhīti yadi hetoḥ sādhyānvayo yatra sa sādharmyadṛṣṭānta iti lakṣaṇam.)
[10]:
PSṬ D214b7; P244a4f.: ma yin te brjod par mi’dod pa nyid kyi phyir zhes pa bsgrub bya’i gtan tshigs kyi rjes su’gro ba yod kyang / brjod par’dod pa yod pa ma yin te / (Skt. B171b4: nāvivakṣitatvād iti. sann api sādhyasya hetunānugamo na vivakṣyate.)
[11]:
Cf. Katsura 2004: 143: “Dignāga clearly states that the main purpose of an example statement is to indicate an external object (bāhyārtha) as an example. This seems to suggest that as long as he is discussing logic and epistemology, he is assuming external reality.” Also cf. Katsura 2004: 155: “He (= Dignāga) further states that the main purpose in referring to a particular object like a pot is to indicate some positive support in external reality.”
[12]:
PSṬ D215a1f.; P244a6f.: bum pa la sogs pa’i khyad par las / bsgrub bya’i rigs can gyi spyi ni phyi rol gyi don no[to D] // de la spyir gang cung zad byas pa zhes bya ba de[de’i P] thams cad mi rtag pa kho na’o zhes bsgrub byas gtan tshigs la khyab pa gsal bar byed pa ni phyi rol gyi don gsal bar byed pa’o // (Skt. B171b5f.: ghaṭāder viśeṣāt sādhyajātīyasāmānyaṃ bāhyo’rthaḥ, tatra sāmānyena yan nāma kiṃcit kṛtakaṃ tat sarvam anityam eveti sādhyena hetor vyāptiprakāśanaṃ bāhyārthaprakāśanam.)
[13]:
PSṬ D185b2–4; P211a7f. (= Appendix [1]): phyi rol gyi don nye bar bsdus pas zhes pa / phyogs su byas pa’i chos can gyi khyad par las gzhan du spyi la nye bar bstan pas zhes pa’i don to // spyi yang khyad par yongs su mi spong pa’i phyir / bsgrub par bya ba’i chos can yang der nang du’dus pa kho na ste / bsgrub bya’i chos can kho na la med na mi’byung ba nyid ston pa bsal ba lhur len pa ni phyi rol gyi don smos pa’o // (Skt. B146a1–3: bāhyārthopasaṃhṛteneti. dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānya upadarśitenety arthaḥ. sāmānyasya ca viśeṣāparityāgāt, sādhyadharmy api tatrāntargata eva. sādhyadharmiṇy evāvinābhāvitvapradarśananirāsaparaṃ tu bāhyārthagrahaṇam.) “[The phrase] bāhyārthopasaṃhṛtena means ‘by [means of two types of examplification based on the similarity or dissimilarity, in which the inseparable relation] is indicated elsewhere (anyatra) than in the particular [property-bearer] that is made to be the subject, i.e. in the [property-bearer] in general.’ And because the general does not abandon a particular, the property-bearer to be proved is also included in it (= the property-bearer in general). The word bāhyārtha, on the other hand, is mentioned for the purpose of negating the indication of the inseparable relation only in the property-bearer to be proved.” See also Shiga 2011b: 525f. with n. 18. It is to be noted that Jinendrabuddhi paraphrases the word bāhyārtha to anyatra here.
[14]:
PSṬ D215a3–5; P244a8–3: de’i phyir spyis[spyi yis P] ji skad bshad pa’i dpe’i mtshan nyid can gyi don gyi phung po thams cad dpe ste / bum pa kho na[na ni D] ma yin no // bum pa ni de’i nang du’dus pas te de[om. P] kho na nye bar mtshan pa’i don du dper rjod[brjod P] pa’o // phyi rol gyi don ston pa ni nges par bstan par[par om. P] bya la gtso bo’o zhes pa’i tshig ni gsal ba’i khyad par gcig kho na la rab tu ston pa gtso bo nyid yin pa sel ba lhur byed par blta bar bya’o // (Skt. B171b6–172a2: tasmāt sāmānyena sarvo yathoktadṛṣṭāntalakṣaṇo’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva. ghaṭas tu tadantargatas tasyaivopalakṣaṇārtham udāhriyate bāhyārthapradarśanaṃ hi nidarśye pradhānam iti vacanam ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ draṣṭavyam.) “Therefore the example is a whole assembly of things in general that have the characteristic of the example as already stated, but not just a pot. A pot, on the other hand, being included within it (= the assembly of things), is given as an [actual] example for the purpose of implying the very [assembly of things]. It should be understood that [Dignāga’s] statement: ‘For the primary [role] of exemplification is to indicate external thing[s]’ is [made] for the purpose of negating [the view that] to indicate [pervasion] only [in] a single particular individual is its primary [role].” See also Shiga 2011b: 527, n. 23.
[15]:
PSV on PS 4.4 (K) 149b1–3; (V) D60b5f.; P64b2f.
[16]:
PSṬ D217a6; P246b7f.: ’dir phyi rol gyi don la ltos pas ci zhig / khyab pa kho nas zhes pa gang du bsgrub byas rjes su’gro ba zhes pa’i mtshan nyid las so // (Skt. B173b7–174a1: kim atra bāhyārthāpekṣayā vyāptyaiveti, sādhyenānugamo yatreti lakṣaṇāt.)
[17]:
See Katsura 2004: 143; 155.
[18]:
The expression bāhyārthopasaṃhāra is found in PSṬ 1 160, 7–9: yadi tāvad viṣaye pravartamānaṃ mana indriyavṛttikṛtam anugrahaṃ nāpekṣate, evaṃ satīndriyāṇāṃ sarvathaivānarthakyaṃ syāt, ma-nasaiva bāhyārthopasaṃhārāt puruṣasyopabhogasiddheḥ. In this case, bāhyārthopasaṃhāra means “to comprehend/cover objects in the external world.”
[19]:
PSV(K) P125b3f. (Kitagawa 1965: 474, 2–5): dper na sgra ni rtag[em.: mi rtag P/Kitagawa] ste / reg bya ma yin pa’i phyir blo bzhin no // de bzhin du mig gi gzung bar bya ba yin pa’i phyir mi rtag zhes bya ba’di yang bsgrub byar bstan pa’i phyir dam bca’ ba thob po // PSV(V) D41a5f.; P43b8–44a1 (Kitagawa 1965: 474, 2–5): dper na sgra rtag ste / reg par bya ba yin pa’i phyir blo bzhin no zhes bya ba dang / de bzhin du mig gis gzung bya yin pa’i phyir mi rtag ces bya ba’di yang bsgrub bya bstan pa’i phyir dam bca’ bar thal bar’gyur ro // (Skt. reconstruction: *tad yathā nityaḥ śabdaḥ, asparśatvāt, buddhivat, evam anityaḥ śabdaḥ, cākṣuśatvāc cety etad api sādhyanirdeśāt pratijñā prasajyate. Cf. NV 274, 4f.) Cf. NP 8, 9: sādhyadharmāsiddho yathā nityaḥ śabdo’mūrtatvād buddhivat.
[20]:
[21]:
Also cf. PV 1.26: tasmād vaidharmyadṛṣṭānte neṣṭo’vaśyam ihāśrayaḥ / tadabhāve ca tan neti vacanād api tadgateḥ // This is Dignāga’s view on the substratum (āśraya) of vaidharmyadṛṣṭānta, which is indirectly quoted by Dharmakīrti. They both think that it is not always necessary for a dissimilar example to have a substratum.
[22]:
Cf. Hetumukha (?) sarva evāyam anumānānumeyavyavahāro buddhyārūḍhena dharmadharmibhedena, [na bahiḥ sadasattvam apekṣate NVTṬ]. (Quoted in PVSV 2, 22–3, 1; NVTṬ 51, 11f.; 162, 28f.) See also Frauwallner 1959: 164.
[23]:
The entire Naiyāyika definition of udāharaṇa is as follows. NS 1.1.36–37: sādhyasādharmyāt taddharmabhāvī dṛṣṭānta udāharaṇam, tadviparyayād vā viparītam. “The exemplification is an [actual] example that [is supposed to] have [another] property (i.e., the property to be proved) of it (= the property-bearer to be proved) due to the [example’s] similarity (i.e., the proving property) to the [property-bearer] to be proved. The counter[-example] is [an actual example that is not supposed to have another property (i.e., the property to be proved) of the property-bearer to be proved] due to its opposite (= the similarity to what is to be proved) (i.e., the dissimilarity to what is to be proved).”
[24]:
PSV(K) P153b6–8: gal te dpe snga mas khyad par du bya ba yin na bsgrub bya dang chos mthun pa dang mi mthun pa smos par mi bya ste / gang gi phyir, de ltar go byed de ma mthong / (PS 4.19c) phyi rol gyi don rjes su ma bsdus pa la dpes bsgrub bya’i chos rtogs pa ni yod pa ma yin pa’i phyir de’i chos rtogs pa nyid du ma grub pa kho na yin no // PSV(V) D64a7–b1; P68a6–8: gzhan yang gal te dpes[dpe yis P] sngon du khyad par du byed na ni bsgrub par bya ba’i chos dang mthun pa dang / chos dang mi mthun par’dzin pa yin[ma yin P] te / go byed du de mthong ma yin // (PS 4.19c) phyi rol gyi don gyi rjes thogs su ma smos pa’i dpes ni bsgrub par bya ba’i chos la go bar bya ba yin no zhes chos de rtogs pa nyid du yang mi’grub po // (Skt. reconstruction: *yadi pūrveṇa dṛṣṭānto viśeṣyate, na sādhyasādharmyavaidharmyagrahaṇaṃ kartavyam. yasmāt, naivaṃ sa gamako dṛṣṭaḥ (PS 4.19b) na hi bāhyārthānupasaṃhāre dṛṣṭāntasya sādhyadharmagamaka[tvam] astīti taddharmabhāvitvenaiva vāsiddham.)
[25]:
Or “does not associate [the pervasion or inseparable relation of the proving property with the property to be proved] with external thing[s],” according to Jinendrabuddhi.
[26]:
PSV(K) 153b5f.: ’di la yang gal te gtan tshigs kyi don las dpe tha dad pa yin na dpe gtan tshigs kyi don la’brel pa nyid du gdon mi za bar brjod par bya bar mi’gyur ro // de lta na yang byas pa’i phyir mi rtag ste nam mkha’ bzhin zhes bya ba yang dper’gyur te yod pas yod pa’i phyir dang / mnyan bya ba ma yin pa’i phyir zhes bya ba’di la bsgrub par bya va dang chos mthun pa dang mi mthun pa yod pa yin no // PSV(V) D64a6f.; P68a5f.: de yang[de yang om. P] gal te gtan tshigs kyi don las dpe logs shig pa yin na ni dpe gtan tshigs kyi don dang rjes su’brel par brjod par mi bya bar’gyur ro // de bzhin du byas pa’i phyir na mi rtag pa ste / nam mkha’ bzhin no zhes bya bar yang dper’gyur ro // de’i phyir bsgrub par bya ba’i chos dang mi mthun pa dang / chos dang mthun pa ni yod pa dang mnyan bya nyid dag la sogs pa la yang yod pa[yod pa om. P] yin no // (Skt. reconstruction: *atrāpi yadi hetvarthāt pṛthag dṛṣṭāntaḥ, na dṛṣṭānto hetvarthānugata evāvaśyaṃ vācyaṃ syāt. tathā ca kṛtakatvād anitya ākāśavad ity api dṛṣṭāntaḥ syāt. asti hy asya sādhyena sādharmyaṃ vaidharmyaṃ ca sattvāśrāvaṇatvādi.)
[27]:
PSṬ D230a2f.; P260b4–6 (= Appendix [4]): phyi rol gyi don nye bar ma bsdus pa zhes pa chos mthun pa nyid dang chos mi mthun pa nyid tsam las gzhan pa’i don ni phyi rol gyi don te / de yang khyab pa ste / med na mi’byung ba nyid do zhes pa’i tha tshig go // de’i nye bar bsdus pa ni rtogs par bya ba’i yul nyid du nye bar sbyar ba ste / gsal bar byed pa zhes pa’i don to // de med na dpe la go bar byed pa nyid med do zhes pa ngag[dag P] gi don to // (Skt. B185a1–3: na hi bāhyārthānupasaṃhāra iti sādharmyavaidharmyamātrād anyo’rtho bāhyārthaḥ. sa punar vyāptir avinābhāvitvam iti yāvat. tasyopasaṃhāraḥ pratipādyaviṣayatvopanayanaṃ prakāśanam ity arthaḥ. tasminn asati nāsti dṛṣṭā-ntasya gamakatvam iti vākyārthaḥ.) Cf. HBṬ 62, 18–20: sarvasmin sādhanadharmavati dharmiṇi na dṛṣṭāntadharmiṇy eva sādhyadharmasyopasaṃharaṇam upasaṃhāro ḍhaukanaṃ ...
[28]:
This interpretation seems to reflect Dignāga’s preceding assertion that the part of the Naiyāyika definition of exemplification sādhyasādharmyāt is useless, because an example defined as such does not convey the knowledge of the property to be proved. Cf. PSṬ D232b1f.; P263a7: chos mthun nyid kyang mi rigs’gyur / zhes[shes P] khyad par med par brjod pa’i phyir / (Skt. sādharmyaṃ ca na yujyata ity aviśeṣaṇābhidhānāt.)
[29]:
In other words, Jinendrabuddhi here equates ‘the place’ where the association is made (= bāhyārtha) with ‘the object’ of the associaion (= vyāpti). And there is another passage in the PSṬ where bāhyārtha is interpreted as pervasion. PSV(K) P154a6f.: gal te khyod kyis bsgrub bya dang sgrub byed dang ldan pa dpe’i don yin na gnyi ga’i chos dang ldan du zin kyang phyi rol gyi don nye bar ma bsdus na bsgrub bya’i don rtog par nus pa ma yin no zhes bshad zin to // PSV(V) D64b6; P68b5f.: gal te khyod kyis bsgrub par bya ba dang sgrub par byed pa’i chos dang ldan pa’i dpe las don yin na ni gnyis ka’i chos dang ldan pas kyang ngag gi don mjug bsdus pa las bsgrub par bya ba’i don rtogs par mi nus so zhes sngar brjod zin to // (Skt. reconstruction: *yadi bhavatāṃ sādhyasādhanadharmavatā dṛṣṭāntenārthaḥ, na tūbhayadharmavatāpi bāhyārthānupasaṃhāreṇa sādhyo’rthaḥ śakyo bhāvayitum ity uktam.) PSṬ D231b1; P262a5: khyab pa gsal bar byed pa’di tsam med par bsgrub par bya ba’i don rtogs par nus pa ma yin no zhes sngar bshad zin to // (Skt. B186a4: na tv etāvatā vyāptiprakāśanam antareṇa sādhyo’rthaḥ śakyo gamayitum ity uktaṃ prāk.)
[30]:
[31]:
The outline of Dignāga’s criticism is as follows. According to the Naiyāyika definition of upanaya, the property as a logical reason, such as “being produced” (kṛtakatva), is applied to the subject, such as “sound.” In that case, it follows that the property kṛtakatva could be applied as either a common property or a particularized property. However, fallacies would occur in both cases. It is not possible to apply all kinds of common properties of the example to the subject. (PSV(K) 154b5, PSV(V) D65a4f.; P69a3–5)
[32]:
NS 1.1.38: udāharaṇāpekṣas tathety upasaṃhāro na tatheti vā sādhyasyopanayaḥ. “The application is to associate [the property of the example, i.e., the proving property] with what is to be proved in relation to the examples: [the subject] is like [the similar example] or [the subject] is unlike [the dissimilar example].” It should be noted in passing that Viśvanātha glosses the word upasaṃhāra as “placing near” (upanyāsa) in his Vṛtti 313, 26 on NS 1.1.38. Cf. YD 91, 3: sādhyadṛṣṭāntayor ekakriyopasaṃhāra upanayaḥ. “The association, i.e., the application is to equate an example with what is to be proved.”
[33]:
The preceding passage reads as follows: PSV(K) 155b3f.: mi rtag pa nyid kyi gtan tshigs lus can nyid la sogs pa yang med pa’i phyir mi rtag pa nyid’gog par yang ma gyur cig snyam nas / gang las de med pas rtag par mi’gyur ba ma byas pa kho na rtag pa nyid kyi gtan tshigs ni ma yin no // PSV(V) D66a1; P69b8–70a1.: ma byas pa nyid ni rtag pa nyid kyi gtan tshigs ma yin te / gang de med pa las rtag par mi’gyur ro // mi rtag pa nyid kyi gtan tshigs lus can ma yin pa la sogs pa yang med pa’i phyir mi rtag pa nyid’gegs pa yang ma yin no // (Skt. reconstruction: *mā bhūd mūrtatvāder apy anityatvahetor abhāvād anityatvapratiṣedho’pi. na hy akṛtaka eva nityatvahetuḥ, yataḥ tadabhāvān na nityaḥ syāt.) “Also regarding [the logical reason] having a fixed form (mūrtatva) and the like, it is not [possible] to negate impermanence either, because [mūrtatva] is not a [valid] logical reason for [proving] impermanence, for [the fact of] not being produced [would] never be a [valid] logical reason for [proving] permanence; otherwise (= if the fact of not being produced were a valid logical reason for proving permanence) what is permanent would not be present for the reason that it (= the fact of not being produced) is not present.”
[34]:
PSṬ D236a3; P267a4f.: de nyid kyi phyir gsungs pa / rtag pa nyid byas pa ni’ga’ yang yod pa ma yin no zhes pa ste / mi rtag pa nyid dang bral ba’i byas pa yod pa ma yin no zhes pa’i don to // (Skt. B190a3f.: na hi kiṃcin nityaṃ kṛtakam astīti na kiṃcid anityatvarahitaṃ kṛtakam asti.)
[35]:
PSṬ D236a2f.; P267a4 (= Appendix [8]): phyi rol smos pa ni’ba’ zhig pa’i bum pa dang / nam mkha’ yod pa dang med pa nyid tsam la ltos[bltos P] pa sems[sem pa P] lhur byed pa’o // (Skt. B190a3: kevalaghaṭākāśasadasattvamātrāpekṣatvanirāsaparaṃ bahirgrahaṇam.)
[36]:
Cf. PSṬ D185b2–4; P211a7f. (Skt. B146a1–3).
[37]:
PS 2.11: liṅgasyāvyabhicāras tu dharmeṇānyatra darśyate / tatra prasiddhaṃ tadyuktaṃ dhamiṇaṃ gamayiṣyati // See also Shiga 2011b: 528–532.
[38]:
This is paraphrased as “the generalized property-bearer” (dharmisāmānya) by Jinendrabuddhi.
[39]:
PSV(K) P112b5 (Kitagawa 1965: 461, 11f.): spyi dang’di bstan pa nyid yin te gang na du ba yod pa de na me yod do zhes rab tu bstan phyir ro // PSV(V) D30b1; P31b1 (Kitagawa 1965: 461, 11f.): gzhi thun mong ba nyid du bstan pa ste gang na dud pa yod pa de na me yod[de na med P] do zhes bstan pa’i phyir ro // (Skt. reconstruction: *ādhārasāmānyena tu pradarśita eva, yatra dhūmas tatrāgnir iti pradarśanāt.)
[40]:
PSṬ 2 45, 6: anyatreti sāmānyena sarvatra mahānasādau. Jinendrabuddhi glosses the word anyatra in the PSV as “the generalized place that has smoke” (PSṬ 2 45, 10: dhūmavatpradeśasāmānye).
[41]:
PSṬ D185b2–4; P211a7f. (= Appendix [1]): phyi rol gyi don nye bar bsdus pas zhes pa / phyogs su byas pa’i chos can gyi khyad par las gzhan du spyi la nye bar bstan pas zhes pa’i don to // spyi yang khyad par yongs su mi spong pa’i phyir / bsgrub par bya ba’i chos can yang der nang du’dus pa kho na ste / bsgrub bya’i chos can kho na la med na mi’byung ba nyid ston pa bsal ba lhur len pa ni phyi rol gyi don smos pa’o // (Skt. B146a1–3: bāhyārthopasaṃhṛteneti. dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānya upadarśitenety arthaḥ. sāmānyasya ca viśeṣāparityāgāt, sādhyadharmy api tatrāntargata eva. sādhyadharmiṇy evāvinābhāvitvapradarśananirāsaparaṃ tu bāhyārthagrahaṇam.)
[42]:
PSṬ 2 46, 4–9: ādheyenāvinābhāvitve darśyamāna ādhāreṇāpi darśitaṃ bhavati. ādheyatantratvād ādhārasya. atha vā yatra dhūmaḥ, tatrāgnir iti nānena dhūmasyāgnimātreṇāvinābhāvitvaṃ kathyate, kiṃ tarhi, ādhārasāmānyaviśiṣṭena. tasmād ādhārasāmānyenāpi tad darśitam eva. “When it is indi-cated that [a logical reason] is inseparably related with what is to be contained, it follows that [the inseparable relation of the logical reason] with the container (i.e. the generalized substratum) is also indicated, because a container depends on what is to be contained. Alternatively, it is not stated by this [formulation:] ‘Wherever there is smoke, there is fire,’ that smoke is inseparably related [only] with mere fire, but that [the smoke is inseparably related] with the [fire] that is qualified by the generalized substratum. Therefore, that (= the inseparable relation) [of the logical reason] with the generalized substratum is actually indicated.”
[43]:
We can describe the relationship between trairūpya and bāhyārthopasaṃhāra as follows: to confirm the three characteristics of a logical reason (hetu), i.e., its being the property of the subject, its being present in sapakṣa (which is similar to the subject in that it has the property to be proved [sādhyadharma]) and its not being present in vipakṣa (which is dissimilar to pakṣa because it does not have sādhyadharma), means to check whether the logical reason is valid, when or after the inference in question is made or formulated. This could be called “the process of applying the logical reason to those cases other than the subject” or “the process of inference from the perspective of hetu.” bāhyārthopasaṃhāra, on the other hand, means to associate/apply the property to be proved (sādhyadharma) with/to external things (= dṛṣṭāntadharmin), having the proving property (sādhanadharma) in the exemplification (dṛṣṭānta) as a member of proof, when or after the inference in question is made or formulated. This could be called “the process of applying the property to be proved to those cases other than the subject” or “the process of inference from the perspective of sādhyadharma.”
[44]:
Cf. HBṬ(V) D117a3f.; P145b2f.: ji ltar gzhan dag gis bsgrub par bya ba’i chos can yongs su spangs nas khyab pa gzhan rab tu ston pa lta bu ni ma yin te / thams cad smos pas[pa P] ni khyab pa gzhan dgag pa’i phyir ro // (For this passage, see also Funayama 2001.)
[45]:
See Shiga 2011b: 528–533 and Matilal 1985: 129, n. 1.
[46]:
For the detail of Pātrasvāmin, see Shiga 2011a: 423–426.
[47]:
See Shiga 2011a: 425f.
[48]:
TSP 500, 13f. and PSṬ 2 2, 13f.: vinā sādhyād adṛṣṭasya dṛṣṭānte hetuteṣyate / parair mayā punar dharmiṇy asambhūṣṇor vināmunā // PSṬ 2 2, 10f.: avinābhāvitvaṃ hi sādhyād bahir iṣyate, anyathā-nupapannatvaṃ tu dharmiṇy eva sādhya eveti; 2 3, 4: avinābhāvitvaṃ hi paraiḥ sādhyād bahir iṣyata iti, eṣo’dhyāropaḥ. Cf. HBṬ 62, 27–63, 2: na hi sa śyāmaḥ, tatputratvāt, paridṛśyamānaputravad iti tatputratvasya śyāmatvena sādhyād bahiḥ paridṛśyamānaputre vyāptipradarśane’pi sādhyasiddhir bhavati.
[49]:
See SVin 5.15cd: antarvyāptāv asiddhāyāṃ bahirvyāptir asādhanam // PSaṃ 32cd: avinābhāvasambandhe’py antarvyāptyāvatiṣṭhate // PSaṃ 50cd: antarvyāptāv asiddhāyāṃ bahiraṅgam anartha-kam //
[50]:
Cf. NA 20: antarvyāptyaiva sādhyasya siddher bahirudāhṛtiḥ / vyarthā syāt tadasadbhāve’py evaṃ nyāyavido viduḥ // See also NAVi 401, 10f. on bahirudāhṛtiḥ: bahir vivakṣitadharmiṇo’nyatra dṛṣṭā-ntadharmiṇy udāhṛtir vyāptidarśanarūpā ...
[51]:
Indeed Jinendrabuddhi’s interpretation may have been influenced by Dharmakīrti’s thoughts, but it is also problematic to think that Jinendrabuddhi always comments on the PS/PSV by adopting and following Dharmakīrti’s theories. There may be cases where Jinendrabuddhi’s comments conform to Dignāga’s original intention. (Cf. Steinkellner 2004: 227f. with n. 5.)
[52]:
We cannot find passages that positively support the meaning of “an object in the external world” in Dignāga’s statements on inference either.
[53]:
[54]:
See PS/PSV 5.34 and Pind 2011: 70–73.
[55]:
For the interpretation of adarśanamātra, see Kataoka 2011: 195 with n. 12f. and Pind 1999: 323–330.
[56]:
According to Jinendrabuddhi, bāhyārtha is to be understood not as a single and particular individual (PSṬ D215a4f.; P244b2; Skt. B171a2f. ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ draṣṭavyam), but as a whole assembly of things (PSṬ D215a3; P244a8f.; Skt. B171b7: sāmānyena sarvo yathoktadṛṣṭāntalakṣaṇo’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva) or the generalized property-bearer as opposed to the particular property-bearer to be proved (PSṬ D185b3; P211a7; Skt. B146a2: dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānye ).