Do Conditionals Have Truth Conditions?

Abstract

In the first part of this paper (§§ 2 and 4) I rule out the possibility of truth conditions for the indicative conditional 'If A, B' which are a truth function of A and B. In the second part (§ 6) I rule out the possibility that such a conditional has truth conditions which are not a truth function of A and B; I rule out accounts which appeal, for example, to a stronger-than-truth-functional "connection" between antecedent and consequent, which may or may not be framed in terms of a relation between possible worlds, in stating what has to be the case for 'If A, B' to be true. I conclude, therefore, that the mistake philosophers have made, in trying to understand the conditional, is to assume that its function is to make a statement about how the world is (or how other possible worlds are related to it), true or false, as the case may be. Along the way (§§ 3 and 5) I develop a positive account of what it is to believe, or to be more or less confident, that if A, B, in terms oi which an adequate logic oi conditionals can be developed. The argument against truth conditions is independent oi this positive account oi the conditional, as I show that any truth-conditional account has counterintuitive consequences, as well as clashing with my positive thesis. But the positive account prevents the paper from merely having created a paradox, or a vacuum. The paper is inspired by Ernest Adams' book, The Logic 01 Corulitionals? My positive thesis is a less technical variant