Abstract

This thesis is an interpretation of the evolution of Wittgenstein's view of language. I distinguish three main phases in this evolution according to what Wittgenstein considers to be the basis of the distinction between correct and incorrect uses of language, between sense and nonsense. First, in the Tractatus, Wittgenstein held an essentialist view of language according to which the limits of significance are established by the logical form of the proposition. The challenge of this view is to provide a method of logical analysis that can uncover the logical form hidden in the statements of ordinary language. Wittgenstein could not find such method of analysis and, consequently, he rejected the idea of a hidden logical calculus underlying every possible symbolism. Second, in the intermediate period, Wittgenstein adopted a conventionalist view of language in which the distinction between sense and nonsense is drawn by arbitrarily stipulated rules of grammar. The central challenge of this view is to explain how symbolic conventions can create an objective distinction between correct and incorrect uses of language. Unable to meet this challenge, Wittgenstein concluded that the normative relation between rules and their applications cannot be forged by mere stipulations. Finally, in his mature philosophy, Wittgenstein defended a pragmatist view of language. On this view, it is our actual practices of language use that provide the requisite background for rule following. The main challenge of this view is to avoid a blunt identification of what is correct with what we count as such. I argue that Wittgenstein meets this challenge successfully by developing a practice-based account of language that does not reduce the normative to the empirical and psychological. On this account, what sustains the normativity of a linguistic practice is a consensus of action among practitioners created by processes of enculturation