Referential Use and Polysemy

Abstract

What is the relationship between the meaning of a word and its referential uses? Katz (1979) argues that a word's meaning determines its referential uses--but only in "the null context ". Putnam (1977), on the other hand, denies that meaning determines referential use. The present paper will focus on a third, and somewhat different, attempt to answer this question: namely, that of Nunberg (1978, 1979). Nunberg (1979, p. 177) espouses a relationship between meaning and referential use that results in the conclusion that it is possible for linguistics to give a proper account of the way language is used and understood "without having to say that speakers know what words mean." This conclusion is equivalent to the claim that there can be no coherent semantic theory because, as Nunberg puts it, "The semantics /pragmatics distinction cannot be validated even in principle; there is no way to determine which regularities in use are conventional, and which are not." (Nunberg (1979), p. 143) Nunberg calls this position "radical pragmatics" and it is clear that much of contemporary speech act theory would have to be drastically revised if Nunberg's claim about ward meaning were valid. Nunberg also claims that his arguments for this conclusion are based, in part, on arguments to be found in Wittgenstein. In Section 2 of this paper, we will sketch Nunberg's arguments. In Section 3, we will show that these arguments are not valid, and that, therefore, his conclusion regarding knowledge of word meaning is false. In Section 4, we will briefly discuss Wittgenstein on this and a related matter. We do so partly to clear up misunderstandings of Wittgenstein to be found in Nunberg, but more importantly to lay the foundation for our own outline of the relationship between meaning and referential use to be found in Section 5.