Two central topics in epistemology are the nature of knowledge and the norm of belief. The first concerns what knowledge is, the second what we should believe. Infallibilism is a distinctive answer to both. On the first, it claims that you know p just if you infallibly believe p. On the second, it claims that you should believe p with certainty just if you know p. Below we will have much more to say about what “infallibly believe” means. But hopefully it is already clear that the two answers naturally fit together. Believing infallibly is, roughly, having a belief that could not be mistaken. Believing with certainty is, roughly, taking something for granted, without doubt or reservation, and disregarding alternative possibilities. It is natural to think that having a belief in p that could not be mistaken—and nothing short of that—warrants taking p for granted and disregarding the possibility that p is not so. So it is natural to think that if knowledge is infallible belief, then knowledge, and knowledge only, warrants certainty. Infallibilism has a bad press these days. It is said to be factually wrong because there are few things that we infallibly believe and few things that we should be certain of. It is said to be conceptually wrong because knowledge does not require infallible belief and is not enough for justified certainty. Budding epistemologists are warned: if you commit both errors, you will be excessively dogmatic like Descartes; if you commit the conceptual error alone, you will be excessively sceptic like Hume. Thankfully, the story goes, Reid, Peirce, Popper and others have led the way out of this madness and we are all past it now. Indeed, most epistemologists accept an outlook on which knowledge is in some important sense compatible with the possibility of error and on which there are few things we should be absolutely certain of. As Stewart Cohen (1988) put it, “the acceptance of fallibilism in epistemology is virtually universal.” The bad press is undeserved. There is a mistaken infallibilist view that we can ascribe to several past philosophers that leads to a false dilemma between excessive dogmatism and excessive ∗Thanks to Bob Beddor, Davide Fassio, Sandy Goldberg, Mikkel Gerken, Chris Kelp, Arturs Logins, Mona Simion, Levi Spectre, Jacques Vollet and the audience at the 2016 European Epistemology Network in Paris for useful feedback on the ideas of this paper. Special thanks to Charity Anderson for many fruitful discussions of this topic. This paper owes much to the work of Timothy Williamson and John Hawthorne though that is not to say that they endorse the views defended here.

中文翻译：

---

如何成为无误论者*

认识论的两个中心主题是知识的本质和信仰规范。第一个关乎知识是什么，第二个关乎我们应该相信什么。无误论是对两者的独特回答。首先，它声称只有当您绝对相信 p 时，您才知道 p。第二，它声称只要你知道 p，你就应该肯定地相信 p。下面我们将更多地讨论“绝对相信”的含义。但希望这两个答案很自然地结合在一起已经很清楚了。无误地相信，粗略地说，就是拥有一个不会出错的信念。确定地相信，粗略地说，将某事视为理所当然，毫无疑问或保留，而无视其他可能性。很自然地认为，对 p 的信念不会是错误的——而且仅此而已——保证将 p 视为理所当然，而忽略 p 并非如此的可能性。所以很自然地认为，如果知识是无误的信念，那么知识，只有知识，才能保证确定性。这些天，无误论受到了不好的报道。据说它实际上是错误的，因为我们绝对相信的事情很少，我们应该确定的事情也很少。据说它在概念上是错误的，因为知识不需要绝对可靠的信念，也不足以提供合理的确定性。萌芽的认识论者被警告：如果你同时犯这两个错误，你就会像笛卡尔一样过于教条；如果你单独犯了概念错误，你会像休谟一样过度怀疑。谢天谢地，故事是这样的，里德，皮尔斯，波普尔和其他人带头摆脱了这种疯狂，我们现在都过去了。事实上，大多数认识论者都接受这样一种观点，即知识在某种重要意义上与错误的可能性相容，并且我们应该绝对确定的事情很少。正如 Stewart Cohen (1988) 所说，“认识论中对易错论的接受实际上是普遍的。” 坏消息是不应该的。有一种错误的无谬论观点，我们可以归因于过去的几位哲学家，导致在过度教条主义和过度教条主义之间陷入错误困境*感谢 Bob Beddor、Davide Fassio、Sandy Goldberg、Mikkel Gerken、Chris Kelp、Arturs Logins、Mona Simion、Levi Spectre、Jacques Vollet 和 2016 年巴黎欧洲认识论网络的观众对本文的想法提供了有用的反馈。特别感谢 Charity Anderson 对这个话题进行了许多富有成效的讨论。这篇论文在很大程度上要归功于 Timothy Williamson 和 John Hawthorne 的工作，尽管这并不是说他们赞同这里所捍卫的观点。