Response to Camp

Abstract

Let S be an advocate of a substitution interpretation of quantifiers and R an advocate of a referential interpretation. S and R are bound to disagree about semantics on the subsentential level. For example, for S, the semantics of the predicate is a matter solely of the truth values of sentences of which is a part; for R, it is a matter of a relation between and red things. But it is not obvious that S and R disagree about semantics on the sentential level. It is at least possible that they share a project concerned with the concept of truth which can be described in terms of semantical and logical properties of whole sentences about which they agree. Put intuitively, the project is to explain why the sentences of our language have the truth conditions they have. It appears from some detailed examples (set out in papers to which Camp refers in [1]) that in attempting to carry out this project R would succeed but that S would either fail or call on semantical devices which make his subsentential semantics indistinguishable from R's. What can S do? His obvious first step is to ask that the truth project be clarified. Convention T shows us how to do this. The basic materials needed are an object language, a metalanguage which contains a system of structural-descriptive names of object-language sentences, and a translation of object-language sentences into metalanguage sentences. It is important here that a consequence relation be given along with the metalanguage, for the heart of Convention T is that all instances of