Epistemic Risk and Epistemic Utility for Beliefs

Abstract

AbstractIn this chapter, we meet Thomas Kelly’s (2014) argument for permissivism about rational belief. It is inspired by William James’ (1897) permissivism about attitudes to epistemic risk. We interpret Kelly’s argument using an account of epistemic value due to Kenny Easwaran (2016) and Kevin Dorst (2019). According to this, we encode Jamesian attitudes to epistemic risk in our epistemic utilities. For the epistemically risk-averse, the disvalue of believing falsely is greater than the value of believing truly; for the risk-inclined, it is reversed. Kelly holds that rationality requires us to do whatever maximises expected epistemic utility from the point of view of our evidential probabilities. He notes that, since rationality permits different attitudes to epistemic risk, which we encode in different epistemic utilities, the same evidential probability might demand of one person that they believe a proposition and of another that they withhold judgment. This gives permissivism about rational belief.