Abstract

In this dissertation I argue that there are sound epistemic reasons for endorsing scientific realism. The realism I shall defend holds that epistemically reasonable belief is not limited to observables in van Fraassen's sense; that it includes an understanding of why classes of phenomena behave as they do; and that science has progressed by providing us with an ever more accurate and detailed picture of how and why the world behaves as it does. The argument proper consists of two main parts. The "defensive" part of the argument attempts to meet the multi-faceted and, I think, representative challenges advanced by Arthur Fine, Larry Laudan, and Bas van Fraassen, who deny one or more of the above claims. I argue that while their challenges push realism in all the right places, in the end they provide at most a pragmatic rationale for stopping short of realism and going the anti-realist route. The positive part of the argument defends an ontic account of explanation; argues that a theory is unlikely to exhibit explanatory success unless it is explanatorily adequate--unless, that is, it is correct about why the phenomena which constitute its explanandum behave as they do; notes that there are theories which exhibit the relevant kind of explanatory success; and concludes that realism as characterized above is epistemically respectable