Beliefs Old and New

Introduction Modern decision theory is full of talk about subjective probabilities and utilities. While these concepts fill something like the functional roles traditionally filled by belief and desire in philosophical discourse, the relationship between the old and new realizers is usually left unclear. A central piece of this puzzle is the relationship between subjective probability and belief. We shall defend a simple solution to this piece of the puzzle: X believes that p iff X’s subjective probability of p is 1. This view has been widely rejected, but there are telling responses to the standard arguments against it. Since probabilities are measures over possibility spaces, and the possibility space in question is determined by context, we may appeal to contextual features to explain away the absurd apparent consequences of our theory. Meanwhile, it is relatively easy to make a powerful case for our theory. Various functional roles are canonically associated with the concept of belief, as it figures in philosophical discussions. Belief is that psychological state which is a necessary condition for propositional knowledge. A belief that p is the psychological state that disposes linguistically competent people to assert p insofar as they desire to be sincere and believe that the audience would like to be informed whether p. Belief is a state that satisfies various conceptual truths connecting belief, desire and action, such as: