Williamson on Skepticism and Evidence

Abstract

In his insightful paper Williamson is primarily concerned to cast doubt on the thesis that if one has evidence in support of one's belief then one knows what that evidence is. By casting doubt on that claim Williamson wants to argue that the skeptic cannot establish that the evidence one has for believing certain commonplace true propositions is the same as the evidence one would have for believing corresponding false propositions in phenomenologically indistinguishable skeptical scenarios. Despite the fact that one wouldn't be able to know that one is hallucinating when one is, it doesn't follow that one's evidence for believing that one's perception is veridical, when it is, is no different, and no better, than the evidence one possesses when one is the victim of hallucination. While I agree with much of what Williamson argues, I want to emphasize some distinctions that might help us think through some of the issues he raises and evaluate the implications of the conclusions he reaches. The first thing I'd like to point out is that any philosopher who embraces a non-redundant truth condition in the analysis of knowledge as justified true belief (plus whatever condition is added to deal with Gettier problems) is probably going to be forced to allow that there is the asymmetry Williamson points to at the start of the paper concerning epistemic possibility, at least on one understanding of epistemic possibility. With a non-redundant truth condition in one's analysis of knowledge, it certainly looks as if one must admit that it might be epistemically impossible that one is hallucinating the table (when one's perception of it is veridical) but epistemically possible that one is veridically perceiving the table when one is hallucinating it-and this is true even if one's evidence in the two situations is identical. If one can know P by having a true belief that P supported by evidence E, while one can fail to know P by having a false belief supported by that same evidence E, then it follows that there can be different hypothetical situations in which one has the same evidence E while radically different propositions are epistemically possible relative to one's being in the one situation rather than the other. If