Abstract

What is the property that makes knowledge when added to a true belief? This is one of the fundamental questions in epistemology. In this dissertation, I explore this property. Traditionally, epistemic justification, and, after Gettier, indefeasible epistemic justification has been considered as such a property. However, the discussions in the last four decades show that it is not at all easy to come to an agreement concerning epistemic justification really amounts to. Answers given by internalists often conflict with those of externalists and reliabilists. ;My general position is that reliabilism is correct as an account of the property that is needed for knowledge but has some difficulties as an account of epistemic justification. Thus, alternatively, I suggest a new term, "epistemic goodness," in order to isolate the substantial pre-theoretic property required for knowledge without identifying it as epistemic justification. Then, I propose my own account of epistemic goodness. I focus on the fact that a belief has two aspects to be assessed, namely the aspect related to belief acquisition and to belief maintenance. Using two separate criteria for measuring the goodness in each aspect, such as truth-conducivity and subjective correctness, I argue that an epistemically good belief is a belief that is reliably produced and is correct from the agent's point of view. Finally, after considering the cases where epistemic goodness can be undermined, I conclude that indefeasible epistemic goodness is the property that makes knowledge when added to a true belief. ;On the way to this conclusion, I critically investigate Alvin Plantinga's theory of warrant and Ernest Sosa's theory of intellectual virtue. They are chosen because they are claimed as improved adaptations of the basic reliabilist idea, and, for that reason, they are competing theories to my account. Although I accommodate many of the merits that I found in their theories in my account, I eventually reject both Plantinga's proper functionalsim and Sosa's version of virtue epistemology due to the various theoretic difficulties