Faculty of Arts and Social Sciences
A Graph-Theoretic Approach to Coherentism about Epistemic Justification
Wednesday, May 14, 2014
15:00 FASS 2034
Some of our beliefs depend for their justification on other beliefs we hold, which in turn depend on yet other beliefs. Where does such a regress end? This is the regress problem in epistemology. Fifty years ago, coherentism -- according to which each thread of support for a justified belief eventually loops back on itself -- was probably the dominant response to this problem, but in recent decades that view has fallen into disrepute. In my talk I will sketch a new way of thinking about coherentism, and show how it avoids many of the problems often thought fatal for the view, including the isolation objection, worries over circularity, impossibility results derived from probability theory, and concerns that the concept of coherence is too vague or metaphorical for serious theoretical use. The key to my approach is to take a familiar tool from discussions of the regress problem -- namely, directed graphs depicting the support relations between beliefs -- and to use that tool in a more sophisticated way than it is standardly employed.