Belief, Credence, and Evidence

• Applicant: Dr. Dominik Kauss
• Subject Area: Theoretical Philosophy
• Term: from 2018 to 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 398821307

• Final Report Year: 2021

Final Report Abstract

It is a kind of received view in philosophy that the only sort of propositions one could rationally be certain of (i.e. have a credence of 1 in) are logical and mathematical truths, and perhaps phenomenological propositions describing one’s current experience. The first segment of the project led to the finding that singular existential propositions pose a threat to the generality of this view, since these propositions are neither logical in any standard sense, nor mathematical, nor phenomenological, and yet are such that the only credence one could rationally have in any such proposition is 1 (certainty). The second segment led to the finding that, if the received view is false, then, firstly, a prominent paradox about rational belief, called the Preface Paradox, can be generalized so as to apply not just to the notion of belief but to the notion of credence as well, and that, secondly, the standard solutions to the belief version fail when applied to the credence version. The third segment led to the finding that the Preface Paradox, as usually stated, is ambiguous between a strong and a weak reading of ‘belief’, and that solutions based on one particular interpretation while ignoring the other fail by way of failing to explain away the individual attractivity of at least one of the paradox’s assumptions. Finally, the fourth segment focused on the idea that the probability calculus is normative for credence, an idea enjoying widespread acceptance among philosophers, decision theorists, and economists. The segment led to the finding that this idea must be restricted so as to exclude cases where the agent possesses no evidence for or against the propositions in question. This result is motivated by the suggestion that the only pertinence the calculus has for questions of evidence and rational credence is its codification of structural rules for weighing the evidence. If true, this indicates that the calculus will remain silent on the question which credence to adopt in cases where the agent’s relevant evidence set is empty.

Publications

• ‘Context-Sensitivity and the Preface Paradox for Credence’. 2021, Synthese
  Kauss, D.
  (See online at https://doi.org/10.1007/s11229-021-03115-6)