SOME REMARKS ON VERIFICATIONISTIC
THEORIES OF MEANING* I shall discuss some ideas and problems in theories of meaning that are based on a notion of verification. In recent years, Michael Dummett is the one who has most contributed to the discussion of such theories, and I shall start by outlining what I take to be his conception of a theory of meaning, what its main task is, and what general form it is to have.
1. DUMMETT'S PROGRAM FOR THEORIES OF MEANING
The main task of the philosophy of language is, according to Michael Dummett, to explain what makes a language language, i.e., what makes the utterance of sounds the communication of ideas. A reasonable first answer is that it is the speakers' knowledge of the meaning of the utterances which makes the utterances something more than mere noise. If this is accepted, the main significance of a theory of meaning is that it tries to explain what it is to have a language in terms of what it is to know the meaning of the expressions of a language.
The task of a theory of meaning understood in this way is something different from just explaining how language is used. The goal is not to give a causal theory of what the speaker says in different situations but rather to explain the possession of a language in terms of the mastering of the rules for how to speak the language correctly.
A theory of meaning for a specific language is conceived by Dummett as something that specifies the meanings of the expressions in the language in such a way that it constitutes what the speaker of the language (qua speaker) implicitly knows. However, what is of interest for a general philosophy of language is not always the particular content of such a theory of meaning but rather the general form that such a theory has to assume. Using Frege's distinction between sense and force, Dummett suggests that a theory of meaning is to contain two parts: one part that in