How to Refer to a Thing by a Word: Another Difference Between Dignāga’s and Kumārila’s Theories of Denotation

摘要

In studies of Indian theories of meaning it has been standard procedure to examine their relevance to the ontological issues between Brahmin realism about universals and Buddhist nominalism (or conceptualism). It is true that Kumārila makes efforts to secure the real existence of a generic property (jāti) denoted by a word by criticizing Dignāga, who declares that the real world consists of absolutely unique individuals (svalakṣaṇa). The present paper, however, concentrates on the linguistic approaches Dignāga and Kumārila adopt to deny or to prove the existence of universals. It turns out that in spite of adopting contrasting approaches they equally distinguish between the semantic denotation of a word and its pragmatic reference to a thing in the physical world. From a purely semantic viewpoint, Dignāga considers the exclusion (apoha) of others by a word as the result of a conceptual accumulation of the sense-components accepted in the totality of worldly discourse. Among the three characteristics Dignāga held must be met by universals, Kumārila attaches special importance to their entire inherence in each individual (pratyekaparisamāpti / pratyekasamavāya). This is because he pragmatically pays attention to the use of a word in the discourse given in a particular context (prakaraṇa) by analyzing a sentence into a topic and a comment.