Incomparability in epistemology

Abstract

Epistemologists are interested in what makes beliefs well justified. Even before considering competing theories of epistemic justification, however, we should ask what sort of valuational structure we are trying to explain. If, as far as epistemic justification is concerned, beliefs are like bank accounts, then all beliefs are comparable: just as in any bank account there must be more, less, or as much money as in any other, one belief must be better, worse, or as good as any other. Contemporary epistemologists take for granted the assumption that beliefs are comparable in the same way that bank accounts are comparable: bank accounts have balances, sprinters have personal bests for the 100-meter dash, and beliefs have degrees of justifiedness. Alternatively, we could understand the justificatory status of a belief to be more like the expensiveness of a restaurant. Consider a concept of restaurant expensiveness on which the expensiveness of a restaurant is determined by the range of prices for meals at that restaurant. If meals at Restaurant A are 25 to 50 and meals at B are 10 to 20, then A is more expensive than B. Restaurant C, however (30 to 40), is neither more nor less expensive than A. Nor is it equally expensive, for there are restaurants that are more expensive than C that are not more expensive than A (like D, 45 to 50). Some evaluative concepts, like this concept of restaurant expensiveness, do not reduce values to single numbers. If epistemic justification is like this, then there may be pairs of beliefs such that neither is better justified than the other, but nor are they equally well justified. Such beliefs would be incomparable. While incomparability is familiar in the ethics literature, it has not previously been explored in epistemology. I discuss the implications of allowing for incomparability in epistemology, both for theories of epistemic justification and theories of knowledge.