Truth

Deflationary theories treat the truth predicate as having only a logical or grammatical function, rather than as ascribing a property or relation to a truth bearer, as on correspondence, pragmatist, and coherence theories. Frank Plumpton Ramsey (1903–1930) proposed, contrary to Moore, that true generally makes no substantive contribution to what is asserted in a statement: "'it is true that Caesar was murdered' means no more than that Caesar was murdered" (p. 157). "Whatever he says is true" comes out "For all p, if he says that p, then p." This is called the redundancy theory of truth. It is deflationary in denying that true expresses a property of truth bearers or a relation of correspondence between truth bearers and the way the world is.

In Philosophical Investigations (1958), Ludwig Wittgenstein (1889–1951) rejected his idea presented in the Tractatus (1961, originally published in 1922) that "This is how things are" expresses the general form of a proposition. This form of words does express a proposition, but "To say that this proposition agrees or does not agree with reality would be obvious nonsense." Rather, it is a propositional variable the value of which is fixed by an earlier statement, in the way the referent of a pronoun is fixed by an earlier use of a name. This proposal, known as the "prosentential" theory of truth because it treats "That is true" as a "prosentence," was subsequently developed by Dorothy Grover, Joseph L. Camp, Jr., and Nuel D. Belnap, Jr. (1975).

Willard Van Orman Quine (1908–2000) proposed that the truth predicate is used for semantic ascent, which in certain cases is indispensable for expressive purposes: "If we want to affirm some infinite lot of sentences that we can demarcate only by talking about sentences, then the truth predicate has its use" (p. 11). If we wish to say only that Tom is mortal, we can say "'Tom is mortal' is true," but we need not do so. But if we wish to affirm each of the sentences of Euclidean geometry, we have no option but to say "All the sentences of Euclidean geometry are true." We are saying no more than we would say by uttering each of the sentences, but since there are infinitely many of them, we cannot utter them all. For this reason, the truth predicate is practically indispensable. This suggests a disquotational theory of truth on which the content of true for a language is given by all the equivalences: "p" is true just in case p (for all sentences "p" of the language). Presumably, occurrences of true in contexts like "What he said is true" or "That is true" would be either spelled out in the manner of Ramsey or implicitly defined by the T-equivalences in virtue of the fact that the subject of the sentence refers to a particular sentence. Paul Horwich developed a related minimalist theory of truth by taking as the axioms of the theory all the equivalences: the proposition that p is true if and only if p (1990). These deflationary theories have, in common with Moore's and Russell's correspondence theories, the disadvantage of entailing the principle of bivalence, which prohibits "truth-value gaps," as in borderline vague sentences or sentences with presuppositions that fail. A correspondence theory such as Field's can be formulated in a way that avoids bivalence (1972). Deflationary theories also differ from correspondence theories in making it impossible to ascribe truth to sentences in a language we can't understand. Field argued in a 1986 article that a deflationary notion of truth cannot be employed to account for how our tendency to believe truths explains our practical success in action.

Some have denied that truth is definable. Gottlob Frege (1848–1925) argued that if truth is definable as a property, then any judgment of whether an idea is true would involve judging whether the idea has the property, and the question would then arise whether it is true that the idea has the property, generating a circle (1956). The early Moore (1953, p. 262) and Donald Davidson (1917–2003) also denied that truth is definable. Since 1980, philosophers (e.g., Huw Price) have shown an interest in functionalist accounts of truth, which characterize truth in terms of its cognitive or social function, rather than define the concept in other terms.