Are Humean Chances Formally Adequate?

This paper argues that the Humean theory of chance is formally adequate in the sense that it satisfies the axioms of finitely additive probability. To this end, firstly, a functionalist account of chance is defended, according to which chance is a functional property and the function relevant to chance is to constrain an agent’s credence in the manner prescribed by David Lewis’s Principal Principle. Chance thus understood is called an expert. Secondly, to specify the functional role of chance, it is claimed that primarily chance should be treated as a cognitive expert, i.e., expert that constrains an agent’s credence because it is predictively accurate. In this light, Humean chance is taken to be a possible physical realization of the cognitive expert. When Humean chance is understood in this way, it can be shown that it occupies the role of the cognitive expert if and only if it is formally adequate.