Giorgio Volpe

The importance of the comparative notion of versimilitude, or truthlikeness, for a realist conception of knowledge follows from two modest ‘realist’ assumptions, namely, that the aim of an enquiry, as an enquiry, is the truth of some matter; and that one false theory may realize this aim better than another. However, there seem to be two ways in which one (false) theory can realize this aim better than another. One (false) theory can be closer to the truth than another either by being preponderantly more accurate in its predictions or by providing more comprehensive information about the system (or class of systems) at issue. This paper presents a model-theoretic approach to the analysis of the comprehensiveness-related component of the comparative notion of versimilitude. The machinery of the ‘semantic’ view of theories is applied to the problem of providing necessary and sufficient conditions for the truth of sentences of the form, ‘B is truth-increasing with respect to A’, where A and B are taken to be sets of structures.

One often hears the claim that fact-based versions of the correspondence theory of truth face a disruptive dilemma: ‘if all true propositions correspond to the same fact, the notion is useless, and if every [true] proposition corresponds to a distinct fact, then the notion becomes idle’ (Engel 2002, 21). The assumption underlying this claim is that all conceptions of facts can be assigned to either of two categories. The first includes those conceptions according to which facts are so coarse-grained that they collapse into the One Great Fact that is the World itself. The second includes those conceptions that, by failing to individuate facts independently of the entities they are supposed to make true, end by regarding them as so fine-grained that they become identical to (the ‘tautological accusatives’ of) true propositions. The contention that these two alternatives exhaust the options available to the correspondence theorist is not, however, beyond suspicion. In this paper I side with those who are convinced that correspondence theorists can steer clear both of the Scyilla of the One Great Fact and of the Charybdis of the Identity Theory. The third way I shall endeavour to sketch is developed by bringing Stephen Schiffer’s theory of pleonastic entities to bear on the issue of the nature of facts. I shall suggest that by regarding facts as pleonastic entities whose principles of individuation are wholly determined by the hypostatizing practices that are constitutive of the possession of the corresponding concepts, one may hope to frame a (neo-Moorean) version of the correspondence theory that avoids the dilemma. Moreover, I shall argue that the ‘pleonastic’ version of the correspondence theory is in fact but a slightly inflated variant of the ‘conjunctive’ theory of truth defended by John Mackie, William Kneale and Wolfgang Künne – a variant whose main attractive lies in the fact that it is not afflicted by the problems of interpretation raised by the quantificational structure of its ontologically more parsimonious siblings.