Metaphysical Anti-Realism?

Abstract

Michael Devitt describes the classical correspondence theory of truth, for sentences of a certain type x, as follows. 'Sentences of type x are true or false in virtue of: (i) their objective structure; (2) the objective referential relations between their parts and reality; (3) the objective nature of that reality.'1 By 'objective' Devitt means, at least, theory independent.2 Due partly to the influence of Dummett and Putnam, 'realism' or (Putnam) 'metaphysical realism' has in some circles recently come to mean, or at least to entail, this sort of theory of truth. Thus, being a 'realist' (about this class of sentences or that) has come to be contrasted, for example, with holding a redundancy theory, or a coherence theory, or a Peircean 'end of inquiry' theory of the nature of truth, or, perhaps, with holding that truth is rational assertability. Devitt demurs at this use of 'realism', and is surely right in saying that it tends to confound issues that should be kept separate.3 So the title of this paper should really be 'Metaphysical Anti-CorrespondenceTruth'. For the most part I will defer to Devitt's remonstrance, but I will occasionally lapse so as to avoid talking with too full a mouth. Putnam uses 'metaphysical' as an epithet to label correspondence theories of the sort Devitt described. It seems to connote non-empirical, verification-transcendent, empty, incoherent, bad. My title hints, and I will argue, that 'metaphysical', used this way, snugly fits Putnam's anti-realism. It does not fit the traditional correspondence theory. Two kinds of argument against correspondence truth dominate the contemporary literature. The first claims that it is incumbent on the correspondence theorist to demonstrate that we 'need' a correspondence theory in order to explain at least some phenomena, and that no such need has been demonstrated.4 The second claims that even if we do need a theory that affirms a correspondence between true sentences and affairs in the world, any such theory can correlate sentences only with theory-relative affairs and hence will not be a correspondence theory of the strong kind Devitt described; and that therefore correspondence of Devitt's strong kind could not possibly be the nature of truth. For example, Putnam has claimed that we must draw a distinction between two kinds of correspondence theories, one