Permissivism, The Value of Rationality, and a Convergence-Theoretic Epistemology

Epistemic permissivism says that sometimes there are multiple rational responses to the same body of evidence. A recent argument against permissivism says that this view is incompatible with a plausible understanding of the accuracy-conduciveness of rationality, according to which rationality is accuracy-conducive because rational credence is more expectedly accurate than irrational credence. This is called ‘the value problem for permissivism.’ In this paper, I propose a new response to this problem. I defend a convergence- theoretic epistemology: Rationality is accuracy-conducive not because rational credence is more expectedly accurate than irrational credence, but because rational credence performs better with regard to convergence to truth. Drawing on recent developments in formal learning theory, I argue that this ‘convergence-to-truth’ understanding of the accuracy-conduciveness of rationality has many attractive features, and I argue that the convergence understanding is compatible with and even favors permissivism.