Full Program »
Inaccuracy and the Objection from VerisimilitudeAn extensional inaccuracy measure (EIM) is one which is a function solely of the truth-values of the propositions with respect to which the belief function is defined and the values qua credences which the belief function assigns to these propositions. Although EIMs are universally favoured in the context of formal epistemology, they are problematic: they cannot account for the verisimilitude of false beliefs. Call this problem the 'Objection from Verisimilitude' (OFV). Joyce (1998) first stated this problem and responded to it. However, Joyce's response is not satisfactory, or so I argue. I proffer, in the form of a conjecture, a response to OFV, which I take to capture the intent of Joyce's initial response. I call this the `Verisimilitude Conjecture' (VC). To solve OFV, VC must be shown to be at least plausibly true. But, VC is not plausibly true, I argue, in the context of science. Specifically, I consider cases of modeling the beliefs of scientists committed to competing physical theories and show that adopting a more verisimilar theory may increase the EIM score of the set of beliefs of a scientist committed to such a theory. This serves as a counterexample to VC. I conclude that as the OFV remains unsolved, EIMs are not tenable in underpinning a formal epistemology of beliefs.
California State University Los Angeles