Thursday, December 27, 2007

Why I don't like Possible World Semantics

I've previously expressed unease with possible worlds (roughly, sets of propositions), as a way of understanding what "possibility" means, partly because of the sets (which like numbers may well be explicable only in terms of possibilities) and partly because of the propositions (which seem too artificial to cover all possibilities) but mostly because far from helping us to understand possibilities, they just seem to make it easier for us to get them mixed up.
......The first time I met PWS, it was in a reply (I forget whose) to Cartesian scepticism: The idea was that, since we know that we're not BIVs, hence BIV-worlds are remote, whence they can be ignored by whatever truth-tracking (across close possible worlds) criteria is required by our concept of knowledge, which seemed circular and to entirely miss the point.
......More recently the following—Gale and Pruss (2003, eds. "The Existence of God": xl-xli) describing their New Cosmological Argument (1999, Religious Studies 35: 461-76)—seemed to confuse subjective or epistemic possibilities (e.g. Goldbach's conjecture might be false, or true, for all I know) with logical or metaphysical ones (for us platonists, if it is true then it is necessarily true, and not possibly false):
Using the semantics of possible world for modal propositions, according to which a proposition that is possible is true in some possible world, and the weak version of [the Principle of Sufficient Reason], they show that there is some possible world in which there is an explanation for [the conjunction of all contingent propositions that are true in the world, say P] about the actual world. But since a possible world is individuated by its [P], it follows that this possible world is identical with the actual world.

No comments: