- 哲学 (ISSN:03873358)
- vol.2005, no.56, pp.113-129,6, 2005-04-01 (Released:2009-07-23)
The present paper considers the adequacy of an essential assumption in the philosophy of language in the twentieth century, which is that linguistic meaning is to be understood in referential/logical terms and that semantics is inherently truth conditional. Frege origi-nated this assumption. He defined Sinn (sense) as the determiner of Bedeutung (reference) . In contrast, linguistic semantics developed within the framework of the generative theory of grammar has no commitment to Frege's reference determining definition of sense. It defines sense as the determiner of sense properties and relations such as ambiguity, synonymy, analyticity, and entailment. Quine's argument that these sense properties and relations cannot be made objective sense of is critically reviewed.We claim that linguistic semantics is not truth conditional. Truth conditions are to be assigned to the proposition expressed by an utterance, which is captured by the theory of utterance interpretation, i.e., inferential pragmatics.Particular attention is paid to Yamada's argument that the content of illocutionary acts such as commands, promises, and the like cannot be identified with propositions. We claim that Yamada's argument is not convincing. We also comment on his idea of a general theory of content for illocutionary acts which generalizes Austin's theory of truth.