Truth—Value of Ethical Statements: Some Philosophical Implications of the Model—Theoretic Definition of Truth

Abstract

By the model—theoretic definition of truth I mean Tarski’s definition of truth in its model—theoretic version. The definition has usually been deemed to be philosophically neutral. This has been regarded either as its advantage or as its shortcoming, depending on what one has expected from the definition. I shall question here this opinion by pointing out certain philosophical consequences which the model—theoretic version of this definition seems to imply. These are consequences which concern the following problem: What kind of expressions qualify as true or false statements? According to a widely held view, there is, in this respect, a fundamental difference between two kinds of sentences: descriptive and value statements. All descriptive statements are assumed to be true or false; all value statements are denied any truth—value. In what follows I try to argue that this qualification is untenable: the model—theoretic definition of truth seems to place both kinds of sentences in the same category. On the one hand, there are descriptive statements which are neither true nor false; on the other hand, there are true or false value statements. This conclusion follows from certain characteristic features of the semantical metalanguage in which model—theoretic definition of truth is couched, viz. from the set—theoretic ontology assumed by the model—theoretic semanticist.