比較論理学研究 20 号
2023-03-25 発行

パーリ文法家モッガラーナと聖典

Some Canonical Examples in the Moggallānavyākaraṇa
矢崎 長潤
全文
322 KB
Ann-ResProjCent-CompStudLogic_20_63.pdf
Abstract
The Moggallānavyākaraṇa, a Pali grammar composed by Moggallāna (ca. twelfth century), is known to rely heavily on the system of Sanskrit grammar formulated by the Buddhist grammarian Candragomin (ca. fifth century). The Moggallānavyākaraṇa has, however, a remarkable feature naturally absent from Candragomin’s Sanskrit grammar: Moggallāna often cites and analyzes expressions or phrases from the Pali canon. Until now, this characteristic of his Pali grammar has not received enough attention. On the basis of materials currently being prepared for publication, the present study first identifies the sources of some canonical examples treated in Moggallāna’s grammar and then examines his reasons for citing them. The following cases are specifically discussed:
1. In M 2.4 (gatibodhāhārasaddatthākammakabhajjādīnam. payojje), corresponding to C 2.1.44 (gatibodhāhāraśabdārthānāpyānām. prayojye), Moggallāna prescribes in his rule a new item, namely bhajjādi “[the verbal roots] bhajj ‘to roast,’ etc.,” that is not provided in Candragomin’s rule. Accordingly, the Pali grammarian presents three additional examples such as aññam. bhajjāpeti “He prompts someone else to roast.” These examples are found in the Vinaya-pit.aka (IV, p. 264; pp. 274–275) in a slightly different form. The use of the second nominal case ending -am. in such expressions with añña (Skt. anya) would be ungrammatical in Sanskrit when occurring with forms derived from the verbal roots bhajj, etc. Moggallāna has modified Candragomin’s rule in such a way that these canonical expressions can be also considered, reflecting the fact that these specific expressions in Pali cannot be justified by Sanskrit grammar.
2. At the end of his commentary on M 2.2 (kamme dutiyā) relevant to C 2.1.43 (vyāpye dvitīyā), Moggallāna quotes the example sace mam. n’ ālapissati “If [he] does not talk to me ...” from the Saṃyuttanikāya (I, p. 177). Moggallāna first says that other grammarians have preferred to formulate a rule concerning examples of this kind and then points out that such a rule is actually not necessary. Hence, it appears that the intention of his citation is to criticize another Pali grammar, namely, the Kaccāyanavyākaran. a (Kaccāyanasutta 2.6.37/309: tatiyāsattamīnam. ca), where this example is indeed mentioned.
3. In the commentary of M 2.2, Moggallāna quotes furthermore the expressions paṭi bhantu taṃ Cunda bojjhaṅgā “Cunda, call to mind the limbs of wisdom!” and tassa na paṭibhāti “He has no inclination [to work or to eat]” from the Saṃyuttanikāya (V, p. 81) and the Majjhimanikāya (II, p. 106), respectively. The same examples are not cited in the Kāccāyanavyākaran. a, although there is a rule that deals with a comparable matter (cf. Kaccāyanasutta 2.6.36/308). Hence, it is unlikely that Moggallāna intends to criticize another Pali grammatical system. Interestingly, Sanskrit grammarians discuss a similar issue and disagree regarding whether the word prati in examples such as na Devadattam. pratibhāti kiṃ cit “Devadatta cannot think of anything at all” is merely a prefix (upasarga) or a grammatical element called karmapravacanīya (in the second case one would write prati bhāti instead of pratibhāti). Moggallāna opines that the latter interpretation, which is actually preferred by the Cāndra grammarians, allows us to properly understand the canonical expressions in Pali. He attempts to explain the Pali phrases on the basis of the interpretation given by the Sanskrit grammarians without providing any special Pali rules.
内容記述
広島大学比較論理学プロジェクト研究センター研究成果報告書(2022年度)