Facing Facts (review)

The correspondence theory of truth is based on the thesis that true sentences correspond to individual facts, and one of the tasks of semantic theory is to articulate an ontology of facts or states of affairs. Ontologies of facts have been proposed by Bertrand Russell, Ludwig Wittgenstein, and Clarence Irving Lewis, among many others. An alternative antirepresentationalist view, defended by Donald Davidson and Richard Rorty, considers the existence of facts as problematic and, consequently, rejects the idea that there are mental or linguistic representations of reality. Stephen Neale carefully reconstructs the case against facts and their representational surrogates by focusing on a formal argument that has been claimed to show that a viable theory of facts is untenable. This argument, which is a collapsing argument, has been called the ‘slingshot argument’ in deference to the minimal machinery or presuppositions required and its giant-slaying potential. Kurt Gödel hinted at a proof which seems to demonstrate that if a compositional semantic theory treats true sentences as standing for facts, then all true sentences must stand for the same Eleatic ‘Great’ Fact. Such a collapse would constitute a knock-down argument against approaches to the semantics of natural language postulating reference to facts. Nevertheless, the truth of the matter is that most incarnations of the argument are confused and inconclusive, with unclear or unnecessary strong commitments and assumptions.

N starts by examining the most prominent current antirepresentationalist program in the second chapter, namely Davidson’s view of the relation between semantics and ontology as well as his arguments against facts, representations, and the scheme-content distinction. In Ch. 3,N examines Gottlob Frege’s compositionality principle and his abandonment of ‘semantic innocence’. The fourth chapter is devoted to Russell’s claim that true sentences stand for facts rather than truth-values and to the details of his theories of facts and descriptions. In Ch. 5, Gödel’s version of the slingshot argument is introduced. The following five chapters spell out the details of N’s reconstruction of Gödel’s argument and different slingshot proofs that assume no particular account of reference. Interestingly, N shows that the slingshots used by Davidson and Willard van Orman Quine do not really demonstrate the impossibility of facts because they rely on certain logical equivalences and on a nonquantificational theory of descriptions, which is needed in order for those equivalences to obtain. N shows that an inconsistency results when one posits a nonextensional connective that freely allows the use of inference principles using descriptions within its scope. This inconsistency arises because descriptions contain formulae as proper parts. If the interchange of descriptions is allowed when they are satisfied by the same object, then one is basically allowing the interchange of formulae, which ultimately leads to collapse into one Great Fact. Russell’s theory ultimately does not face this problem, given that for him: (1) facts contain objects and properties as components; (2) the structure of a sentence mirrors the structure of the fact for which it stands; and (3) descriptions are quantificational devices. Thus, substitutivity collapses do not arise because two definite descriptions of the same object will not contribute the same descriptive properties to a fact in general.

N’s conclusion is that fact-based representational semantic theories survive the existing slingshot argument and would survive any conceivable, more powerful slingshot argument that could be constructed against them. This is because what a slingshot argument demonstrates is only that these theories have to satisfy nontrivial conditions in order to avoid logical and ontological collapse (the descriptive constraint). Given that current semantic theories based on situations or propositions take these to be complex or structured objects (as in situation semantics or in the theory of structured propositions), they also seem to survive N’s constraint (unlike the Strawsonian or the Austinian theories). Ultimately, N’s book can be read as a slingshot argument against most conceptions based on deficient collapsing arguments and against theories of facts, situations, or propositions that refuse to make...