Abstract

I argue that the analysis most capable of systematising our intuitions about coherence as a relation is one according to which a set of beliefs, A, coheres with another set, B, if and only if the set-theoretical union of A and B is a coherent set. The second problem I consider is the role of coherence in epistemic justification. I submit that there are severe problems pertaining to the idea, defended most prominently by Keith Lehrer, that justification amounts to coherence with an acceptance system. Instead I advance a more dynamic approach according to which the problem of justification is the problem of how to merge new information with old coherently, a process which is seen to be closely connected with relational coherence.