著者

丸井 浩

出版者

東京大学大学院人文社会系研究科・文学部インド哲学仏教学研究室

雑誌

インド哲学仏教学研究 (ISSN:09197907)

巻号頁・発行日

vol.19, pp.19-59, 2012-03-31

---

As the fourth of the Nyāya’s twelvefold prameya (the objects to be rightly known for the attainment of liberation), the “artha” is defined or simply described at the NS 1.1.14: gandharasarūpasparśaśabdāḥ pṛthivyādiguṇās tadarthāhḥ. According to Vātsyāyana, the sūtra seems simply to imply that smell, taste, colour, touch and sound, which are the qualities of earth, etc., are respectively the objects of the five (external) sense-organs. But the sūtra was interpreted differently by different scholars of Nyāya in later times. What attracts our attention, first of all, is a totally different way of analyzing the compound “pṛthivyādi-guṇāḥ.” Uddyotakara criticizes the interpretation of ṣaṣṭhītatpuruṣa (pṛthivyādīnāṃ guṇāḥ) as too narrow and strongly argues for dvandva (pṛthivyādīni ca gunṇāś ca), insisting that the sūtra should be taken to be the exhaustive presentation of all the perceptible entities as the artha or the indriyārtha, including three kinds of substances (pṛthivyādīni = pṛthivyaptejāṃsi), all the perceptible properties (the “guṇāḥ” in its proper sense) other than smell and the like, such as number, etc., along with motion, the higher or lower universals and even the principle of inherence, all of which are implied by the term “guṇāḥ” in its wider sense. Vācaspati clearly supports his view. On the other hand Jayanta definitely rejects the interpretation of dvandva and shows the relevance of tatpuruṣa analysis, whereas Bhāsarvajña tries to justify the dvandva interpretation. But it would be superficial to assume two lines of interpretation simply in terms of the two opposing ways of analyzing the compound. The fundamental difference lies in two divergent viewpoints of treating the concept of artha as a prameya. Both Jayanta and Bhāsarvajña emphasize the soteriological significance of valid cognition of the artha or the objects of sense-organs. Jayanta, in particular, refers to them as a cause of attachment (saktihetu) and states that those who seek liberation should contemplate them as the objects to be abandoned (heyatayā bhāvayitavyāḥ). Jayanta supports the tatpuruṣa analysis because the sūtra is meant to mention only the predominant causes of attachment. Bhāsarvajña, in contrast, insists on the dvandva interpretation because in his view it is necessary to read the sūtra as the exhaustive list of causes of “defects” (doṣa=rāgadveṣa-moha). In spite of the formal difference in the compound analysis, both of them share a soteriological viewpoint in the context of the artha as a prameya. It is difficult to conclude decisively the chronological order of their discussion, but Jayanta’s argument seems to represent a later stage. Uddyotakara’s explanation, on the other hand, concentrates on the epistemological aspect of the artha. He claims the irrelevance of tatpuruṣa interpretation on the ground that it would exclude, for instance, three substances (dravya) that should be admitted as perceptible by the authority of NS 3.1.1. He tries to justify the sūtra’s separate mention of five objects of sense from smell to sound by saying that it is specifically to show the particularity of the one-to-one relationship between each of them and its corresponding sense-organ. He never refers to the soteriological importance of the concept. Moreover, he evidently depends on the Vaiśeṣika doctrine of six categories for his enumeration of perceptible entities with a slight modification. Heavy dependence of his account of the artha upon Vaiśeṣika categories also provides a sharp contrast with Jayanta, who explicitly denied the relevance of six categories in the context of prameya, just as Vātsyāyana had done. It is relevant in this connection to examine the treatment of the artha in two representative compendiums of Nyāya System, probably written not long after Udayana (11th cent.), namely Varadarāja’s Tārkikarakṣā[-sārasaṃ graha] (12th cent.) and Keśavamiśra’s Tarkabhāṣā (around 1300 A.D.). Varadarāja defines the artha as “what can be grasped by senseorgans” (indriya-grāhya) and tries to make an exhaustive list of perceptible entities according to the Vaiśeṣika doctrine of categories without reference to the soteriological implication of the artha. Thus, like Uddyotakara, he concentrates on its epistemological dimension. But he goes a little further by introducing the detailed account of the Vaiśeṣika doctirine of seven categories as an appendix to the description of Nyāya’s twelvefold prameya. Keśavamiśra, on the other hand, goes so far as to extend the conceptual sphere of the artha up to the whole six or seven categories of Vaiśeṣika, thus totally abandoning the soteriological meaning peculiar to the Nyāya’s idea of artha. Therefore it may safely be said that Nyāya’s concept of artha as a prameya was shifted toward the direction of diminishing emphasis on its soteriological relevance and at the same time toward the direction of a closer and closer relation with the Vaiśeṣika doctrine of categories. Against this stream Jayanta tried to retain the traditional soteriological significance of the concept. This conclusion fits in well with the general observation that Jayanta’s account of Nyāya System generally represents its earlier phase. Incidentally, the Nyāyakalikā provides precisely the same soteriological account of the artha as we have seen above in the Nyāyamañjarī. The similarity in wording is also outstanding. These facts seem to speak for Jayanta’s authorship of the text. Even if we were to admit that someone else had composed it by extracting relevant sentences from NM with some modification, it is not likely that a compendium of such an archaic nature was written long after Udayana.