Do religious “beliefs” respond to evidence?

Abstract

Some examples suggest that religious credences (or “beliefs”) respond to evidence. Other examples suggest they are wildly unresponsive. So the examples taken together suggest there is a puzzle about whether descriptive religious attitudes respond to evidence or not. I argue for a solution to this puzzle according to which religious credences are characteristically not responsive to evidence; that is, they do not tend to be extinguished by contrary evidence. And when they appear to be responsive, it is because the agents with those credences are playing what I call The Evidence Game, which in fundamental ways resembles the games of make-believe described by Walton's (1990) theory of make-believe.