There is no theory of truth that has not been ascribed to Wittgenstein. He has been 'credited' with a coherence theory, a pragmatic theory, a consensus theory. In many cases the motivation behind such attributions has simply been a desire to make Wittgenstein's views on truth look exotic and provocative. By contrast, I shall argue that Wittgenstein's reflections are an integral and important part of the loosely speaking realist accounts of truth that dominate mainstream analytic philosophy. Prima facie, the matter is straightforward. The early Wittgenstein developed a sophisticated version of the correspondence theory, while the later Wittgenstein, together with Ramsey, pioneered a deflationary theory of truth. My paper maintains that both claims are defensible, but that the matter is far from straightforward, and that a proper appraisal of the issue requires (and in turn contributes to) a better understanding of what correspondence and deflationary theories of truth amount to.
The paper starts with a brief sketch of different types of correspondence theories (sct. 1) and of why the Tractatus can be seen as solving some problems confronting such theories (sct. 2). In the remaining sections I argue for the following claims:
I. Standard correspondence interpretations of the Tractatus are mistaken, because they
treat the isomorphism between a sentence and what it depicts as a sufficient condition
of truth rather than of sense (sct. 3).
III. This is no coincidence. The official theory of truth in the Tractatus is an obtainment theory -- a sentence is true iff the state of affairs it depicts obtains -- which is close both to semantic/deflationary and to correspondence theories (sct. 6).
IV. The obtainment theory involves a relation of correspondence, yet it is the relation of depiction between a sentence and the possible state of affairs it depicts, not a truth-making relation between a true sentence and the fact it depicts. However, the idea of this kind of truth-making relation is incoherent, and it is not essential to the correspondence theories of Moore and Russell (sct. 7).
V. Semantic, deflationary and correspondence theories are not incompatible. They all pay heed to alethic realism, the idea that whether a sentence is true is independent of whether we say or believe that it is. Furthermore, they all combine a semantic explanation of the relation between a sentence and what it depicts or says, and a deflationary account of the equivalence between what the sentence says and what obtains or is the case if it is true (sct. 8).
VI. From this perspective, Wittgenstein's later remarks on truth do not constitute a radical break. What needs explaining is why he phrased his deflationary view in terms of disquotation (' "p" is true = p') rather than denominalization ('it is true that p = p'), like Ramsey.
VII. Wittgenstein's remarks on the role of consensus (PI §§ 240-2) do not amount to any kind of anti-realism, but only to a recognition that the concept of truth is intimately linked to the concept of what people say or believe, or might say or believe.