Abstract

This is a new major and systematic monograph on the realism debate, written by a very skillful and sophisticated defender of anti-realism. The realism debate is conceived here as a primarily semantic controversy concerning the possibility of forming propositions or sentences about the world whose truth or falsity might be unknowable to us. Realists of various persuasions take it to be perfectly possible, while nonrealists of all sorts deny this. Tennant distinguishes two basic forms of nonrealism in his book: irrealism and antirealism. According to semantic irrealism, there are no such entities as propositions, or objective and determinate meanings for sentences to possess. Semantic antirealists generally do not agree with such a radical conclusion. They argue for the objectivity and determinacy of meaning, although at the same time they insist that both meaning and truth attributable to propositions or sentences has to be epistemically constrained, that is, grasping a meaning has to be manifestable and truth has to be in principle knowable. While developing in greater detail his own version of antirealism that respects those two constraints, Tennant combines the knowability requirement with the claim that truth need not be bivalent, which in turn leads him to replace classical logic based on bivalence by a nonclassical constructive logic, namely, intuitionistic relevant logic. Tennant attempts to show that this logic has much wider application than one might suppose. It suffices not only for mathematics but also adequately captures the basic inferential practices of empirical sciences. It would be hard to overestimate the significance of that attempt, since if successful it undermines the charge that anti-realism is a doctrine that cannot really be extended beyond the domain of mathematical discourse. The general strategy taken in this respect by Tennant is not, as has been common so far, to strengthen the notion of empirical verification or confirmation of a sentence in such a way as to make it a close analogue of the notion of mathematical proof. “Rather, one should attend to the main feature of natural scientific theorizing to which Popper drew our attention: our scientific theories can at best be falsified, not verified”. In other words, in the case of empirical discourse the perfect analogue of the antirealist notion of constructive provability is the notion of constructive falsifiability.