On neo- (and post-) Gricean conditionals

This paper attempts to examine the foundations of a semantic–pragmatic theory that is able to accommodate an elegant analysis of indicative and counterfactual conditionals. Such an analysis is unlikely to be cut from the same cloth as truth-conditional analyses. The principal intuition is that conditionals are not part of fact-stating discourse: instead, conditionals express uncertainties and thus require a very different semantics and a very different notion of validity. Gricean, Strawsonian, Stalnakerian (and, by implication, Lewisian) and Lyncanian analyses are discussed, and neo-Gricean analyses are mentioned. A general preference for an account which exploits conditional probabilities is expressed.