怀疑主义与认知价值

当代知识论越来越重视认知价值的研究，但从认知价值角度去重新思考传统怀疑主义问题的研究却很少。历史上的怀疑主义者既认为我们无法拥有任何知识，又主张我们应该悬置判断。因此，他们不仅预设了一个高的知识标准，而且预设了一个价值标准：任何达不到知识标准的认知状态都是没有认知价值的。但这个价值标准是错误的，所以我们即使无法达到怀疑主义者所设定的知识标准，也不必悬置判断。假设怀疑主义者认为我们无法拥有任何知识，同时认可“有些达不到知识标准的认知状态也有认知价值”，那么怀疑主义就没有很大的威胁性，因为即使我们无法获取知识，也可以获得具有认知价值的东西，取得认知进步。此外，摩尔主义、语境主义、溯因主义这几种对怀疑主义的主要回应是否成功，最终都依赖于对认知价值的讨论。

Skepticism and Epistemic Value

This paper argues that even if the skeptical argument is sound, skepticism does not pose a serious challenge to our inquiry, for we can still make epistemic progress. The skeptical argument comes in various forms, all of which rest on a set of demanding criteria of knowledge. Suppose the skepticscriteria of knowledge are correct. Then if we can meet the criteria, then we may achieve knowledge; if we cannot meet this set of criteria, then we can merely acquire things that fall short of knowledge. Knowledge (given the skeptics criteria) is epistemically valuable. But things that fall short of knowledge may also be epistemically valuable. Consider two individuals, S1 and S2. Both believe the truth that there is a table in the room. But both are unable to rule out the possibility that each of them is a brain in vat. Yet there is a difference between S1 and S2. If there were a chair or stone instead of a table in the room, S1 would believe that there is no table in the room, while S2 would still believe that there is a table in the room. According to the skeptics criteria of knowledge, neither S1 nor S2 knows that there is a table in the room. Thus, if nothing that falls short of knowledge is of epistemic value, then ceteris paribus, S1 is not doing epistemically better than S2. But clearly, ceteris paribus, S1 is doing epistemically better than S2. Therefore, there is something (e.g., S1's cognitive state) that falls short of knowledge but is of epistemic value. In addition, the skeptical argument, even if it is sound, does not show that S1's cognitive state is unattainable. If S1s cognitive state is attainable and epistemically valuable, then we can make epistemic progress even if we cannot know anything about the external world given the skeptics criteria of knowledge. Historically, the skeptics argue not only that we cannot know anything, but also that we should suspend judgment. They seem to implicitly assume that one should believe p only if one knows that p, for from the fact that one does not know that p, it does not follow that one should not believe p. This assumption is known as the knowledge norm of belief. The knowledge norm of belief seems to presuppose that any cognitive state that falls short of knowledge is epistemically valueless. For if there is a cognitive state that falls short of knowledge but is epistemically valuable, then we do not have to suspend judgment when we cannot acquire knowledge. If we can acquire something of epistemic value by believing a certain proposition, we may believe it even if we do not know it. Thus, skepticism is false. Finally, this paper shows that contemporary responses to skepticism cannot avoid value talk. I discuss three kinds of responses: Mooreanism, abductivism, and contextualism. It is argues that each of them appeals to certain value assumptions. Specifically, the core idea of Mooreanism is that we can know p even if we cannot prove p or rule out all alternatives to p. This criterion of knowledge is less demanding than the skeptics criteria. A less demanding criterion of knowledge is plausible only if it does not entail that those who know that p are doing no epistemically better than those who merely believe that p (as far as whether p is true is not about a trivial matter). So Mooreans must assume that knowledge that p (according to their account of knowledge) is epistemically better than mere belief that p. In addition, both abductivism and contextualism employs the inference to the best explanation. Abductivism directly appeals to the inference to the best explanation. Contextualists argue that contextualism is better than alternative accounts of knowledge because contextualism best explains both why the skeptical argument is forceful and why the daily knowledge attribution is appropriate. What counts as the best explanation is clearly a value question.