© 2012, Philip Yaure
In this paper, I consider the Problem of Old Evidence, which is meant to undermine the theory of confirmation Bayesianism uses to explain the role of evidence in science. The problem maintains that the Bayesian definition of evidence cannot include facts known before a theory is introduced (but whose relation to the theory is unknown at the moment of introduction). I argue that this problem can be diffused by the introduction of counterfactuals, which specify conceivable scenarios in which the fact is discovered after the theory is introduced. I consider several sorts of objections to this view, and contend that we have good reason to reject them in their own right, and that the other alternative solution in the literature does not offer a sufficient solution to the problem, further compelling us to face the objections, if we are to maintain a Bayesian confirmation theory.
Yaure, Philip (2012) "Counterfactuals and the Problem of Old Evidence," Res Cogitans: Vol. 3: Iss. 1, Article 2.