Abstract

This dissertation concerns rival philosophical approaches to the philosophy of language, the formal and informal. In chapter one I note that the formal approach has its roots in the study of formal languages, and that this study has devolved into the search for a theory of meaning, an account of semantic structure, of how it is possible to know the meaning of any of the indefinitely many sentences of a language. I introduce Davidson's theory of meaning, discuss how it is supposed to illuminate semantic structure, and preview its shortcomings. I then discuss the informal approach of the later Wittgenstein, why formal philosophers are dissatisfied with his work, and how an account of language understanding is forthcoming from his writings. ;In chapters two and three I bring the meaning-sceptical argument of Kripke/Wittgenstein to bear against a theory of meaning. I argue, by presenting a more revealing informal alternative, that such a theory cannot illuminate semantic structure. I outline a theory of truth for a formal language and give an informal alternative to such a theory, thereby highlighting the redundancy of the notion of truth in an account of language understanding. ;In chapter four I develop an informal account of the understanding of natural language, using French as my example. I show that from a finite number of 'M-sentences', such as: 'Je suis content ' means: I am happy. an M-sentence could be inferred for any French sentence. Such inferences are based on descriptions of the contribution that expressions, and syntactic constructions, make to the meaning of sentences, and such descriptions are based in turn on similarities between French sentences which have a relevantly similar meaning. ;In chapters five and six I address issues raised by this informal account, in particular the relation between language understanding and entailment. I argue that there is no reason why an account of language understanding should cover all the entailment relations that might be taken to hold between sentences. I show how the relations which are implicated in language understanding can be accounted for informally, without recourse to the notion of logical form.