Abstract

The aim of this thesis is to demonstrate the deep and inherent inadequacy of any descriptionist or traditionally Fregean approaches to the semantics of indexical expressions. In the first study I argue that an illuminating truth-theory, capable of serving in a theory of meaning, must represent the semantic value of indexical sentences as relative to the satisfaction of explicitly pragmatic conditions. Introducing a notion of speaker's reference, I show that the semantic reference or denotation of demonstratives is straightforwardly determined by the speaker's acts of demonstrative reference in the introduced sense. To capture this requirement, I propose a condition of adequacy on any truth-theory for indexical languages containing the demonstrative 'that'. ;The second study is concerned to provide a more informative account of the required notion of speaker's demonstrative reference. I take as my starting point, an account of demonstrative reference proposed by David Kaplan. I show that a Fregean identification of the determinants of reference and the determinants of cognitive significance that Kaplan assumes is untenable for demonstratives. I argue that the actual cognitive significance of a given act of demonstrative reference is not a function of the semantics or "meaning" of demonstratives, and that the cognitive significance of a demonstrative claim for a speaker is not to be identified with the expressed propositional content of such a claim. Finally, I show that speaker's demonstrative reference cannot be reduced to or explicated in terms of any view of reference that understands referential intention as a speaker's intention to refer to whatever satisfies some set of "descriptive" conditions that he has in mind. I argue, instead, that it is necessary to recognize an irreducibly relational conception of a speaker's intention to refer to some object x. ;In my third study I argue that the considerations adduced by G. E. M. Anscombe to show that "I" is not a referring expression are inadequate, since they are based on precisely the inadequate theory of reference and referential intentions which I criticize. The referential status of "I" is defended and explicated