Abstract

The target of much work in recent epistemology is the specification, in terms of non-normative properties, of the nature of epistemic norms like justification. I'll call this the target-problem of epistemology. The target-problem is the problem of giving normative epistemic properties purely descriptive, or naturalistic, accounts. The target-problem itself gives rise to a deeper problem and one I think needs to be answered prior to any solution to the target-problem: What methods are appropriate in constructing solutions to the target-problem? I'll call this the meta-problem of epistemology. A solution to the meta-problem requires us to decide what methodology is best-suited to solve the target-problem. I will elaborate one negative and one positive answer to this meta-problem. ;First, a negative answer. The method of reflective equilibrium looms large in epistemology much as it does in ethical theory. Put simply, this approach tells us that the nature of epistemic justification is to be specified by stating, in purely non-normative terms, a set of general principles that accord with our common intuitive classifications of individual cases of justified and unjustified belief. I argue, however, that this method underdetermines a choice between the main theories of justification being defended in the literature today. That is, our shared epistemological intuitions do not justify one theory of justification over its contemporary rivals. Moreover, I consider two ways to try to overcome this underdetermination problem and I show that neither is effective. ;Second, a positive answer. I propose and elaborate a pragmatic-functionalist methodology for epistemology. Put simply, this approach tells us that the nature of epistemic justification is to be specified by picking out that property or set of properties that best promotes our cognitive goals . We want to find out which non-normative property does what epistemic justification does, which one plays the justification-role. When we find it, we will have found epistemic justification, we will have found what makes our justified beliefs justified. This approach is pragmatic because it gives considerable weight to what cognitive aims are being satisfied by epistemically justified beliefs. The approach is functionalist because it tells us that epistemic justification is identified by its causal role. Just to show that the pragmatic methodology gets results, I show that on certain empirical assumptions it will settle on a reliabilist account of epistemic justification