Syntactic semantics is a holistic, conceptual-role-semantic theory of how computers can think. But Fodor and Lepore have mounted a sustained attack on holistic semantic theories. However, their major problem with holism (that, if holism is true, then no two people can understand each other) can be fixed by means of negotiating meanings. Syntactic semantics and Fodor and Lepore’s objections to holism are outlined; the nature of communication, miscommunication, and negotiation is discussed; Bruner’s ideas about the negotiation of meaning are explored; (...) and some observations on a problem for knowledge representation in AI raised by Winston are presented. (shrink)

SummaryCorrectly understood, Wittgenstein's “picture theory of language” is remarkably similar to the basic ideas of a Tarskian‐type logical semantics, except for the crucial Wittgensteinian doctrine that semantical relations can only be shown, not said. This is an instance of the idea van Heijenoort calls “logic as language”.What happens in the transition to Wittgenstein's later philosophy is not that the picture idea is rejected but that a new view of the connections between language and reality is introduced. The basic representative relations (...) between language and the world, which are left unanalyzed in the Tractatus, are now assumed to be created and sustained by different language‐games. The primary function, if not the only one, of Wittgenstein's language‐games is thus to link language to the world.Wittgenstein's persistent attitude of “logic as language” nevertheless led him to claim that language‐games are as ineffable as the naming relations of the Tractatus. For instance, new language‐games cannot be defined by new rules, because games are primary with respect to their rules.The author's semantical games, especially the games of seeking and finding, provide examples to illustrate Wittgenstein's language‐games and also to cast doubt on their ineffability. (shrink)

We argue that the presence of a word in an utterance serves as starting point for a relevance guided inferential process that results in the construction of a contextually appropriate sense. The linguistically encoded sense of a word does not serve as its default interpretation. The cases where the contextually appropriate sense happens to be identical to this linguistic sense have no particular theoretical significance. We explore some of the consequences of this view. One of these consequences is that there (...) may be many more mentally represented concepts than there are linguistically encoded concepts. (shrink)