Abstract

In a series of recent papers Stephen Kearns and Daniel Star argue that normative reasons to ϕ simply are evidence that one ought to ϕ, and suggest that “evidence” in this context is best understood in standard Bayesian terms. I contest this suggestion.