In his Epistemology, Fumerton argues that propositional justification is more central to (applied) epistemology than is doxastic justification. Here is one of his arguments:
There is another reason that epistemologists interested in applied epistemology are probably well advised to focus on what there is [propositional] justification for people to believe rather than which beliefs are actually [doxastically] justified. If it is true that S’s belief is [doxastically] justified only if it is based on good reasons, and it is also true that basing is to be understood, even partially, in terms of causation, then it is not clear that philosophers, in their capacity as philosophers, are particularly well equipped to answer questions concerning which beliefs are [doxastically] justified. The causes of a belief are a more appropriate subject for the psychologist. Freud spent a great deal of time wondering what causes a belief in a God or in an afterlife. The epistemologist, qua epistemologist, should find such speculation utterly uninteresting. Whatever causes such beliefs, the epistemologist’s concern is with the question of whether we possess good reasons for believing the propositions in question. To answer that question we need not concern ourselves with what is actually causing our beliefs. (36-7)
I worry that this argument shows too much. If it shows that epistemologists, qua epistemologists, shouldn’t be interested in what we are doxastically justified in believing, then a parallel line of reasoning shows that epistemologists, qua epistemologists, shouldn’t be interested in what we have propositional justification or good reasons for believing. Here is the parallel argument:
Whether we have certain reasons to believe P largely depends on what beliefs or experiences we have. But whether we have certain beliefs and experiences is the sort of question that epistemologists, qua epistemologists, are not suited to answer. If you wanna know the answer to that question, go talk to our friends the psychologists and cognitive scientists.
The parallel certainly isn’t perfect, but it seems close enough to cast doubt on Fumerton’s argument. (The parallel is stronger if we assume (i) that one can be mistaken about which beliefs and experiences one has and (ii) that brain scans and/or psychological evaluation can provide information about what experiences or beliefs one is having. Fumerton seems to endorse at least (i).)