Justification and Probability

在什么条件下我们有理由相信某事？对于这个问题，一个非常简单而自然的答案是，当一件事很可能是真的时候，我们就有理由相信它。以合理和负责任的方式管理一个人的信仰，就是要把错误的风险降到最低。尽管这个答案看起来简单而自然，但我认为它是错误的。假设有100人参加了一场音乐会，但我们知道只卖出了一张票。因此，我们知道音乐会中只有一人是合法参加的，其他99人都是不速之客。假设我们知道Joe是参加音乐会的人之一，但没有关于他的进一步信息。很明显，根据我们的证据，乔很可能是一个不速之客。但我真的应该得出乔是不速之客的结论并据此对待他吗？我是否应该断言乔是一个不速之客——我是否应该通知他的朋友、他的家人，比如他的雇主？乔应该被逮捕和定罪，并对他进行适当的惩罚吗？我们大多数人都会同意，这将是极其不公平的，根据现行的法律惯例，这类统计证据不会被视为定罪的合法依据。但是，如果我没有理由相信乔是一个不速之客，那么理由就不仅仅是一个可能性问题——肯定有比这更多的理由。有趣的是，如果一个目击者声称她看到乔翻过栅栏进入音乐会，那么我们通常会毫不犹豫地相信乔是一个闯入者，并采取相应的行动——这样的证据很可能被视为法庭定罪的合法依据。然而，我们都知道证词是不可靠的——目击证人可能在撒谎或弄错了。事实上，Joe是一个闯入者的可能性在一个证人作证说她看到他爬过栅栏的情况下不可能高达99%。我在这个项目中的主要目的是试图找到这个谜题的底部，并在这样做的过程中，对什么是相信某事的理由有了新的理解。这项研究不仅与认识论家和学术哲学家有关，也与任何想知道对给定主题应该相信什么或如何评估给定证据的人有关。正如上面的例子所表明的那样，这项研究也与如何最好地理解和合理化我们的法律实践的问题有关。我的目标是将这项研究成果以专著和研究论文的形式发表。这项研究将主要在苏格兰进行，并与苏格兰各地的一些哲学家合作。目前，苏格兰拥有许多知名的哲学系和研究中心，在认识论方面尤其有实力。

Under what conditions are we justified in believing something? A very simple and natural answer to this question is that we are justified in believing something when it is sufficiently likely to be true. Managing one's beleiefs in a reasonable and responsible manner is all about minimising the risk of error. As simple and natural as this answer seems, however, I believe it to be mistaken.Suppose that 100 people attended a concert but we know that only one ticket was ever sold. As such, we know that only one of the people at the concert attended legitimately and the other 99 were gatecrashers. Suppose we know that Joe was one of the people who attended the concert but have no further information about him. Clearly, in light of our evidence, it's very likely that Joe was a gatecrasher. But should I really draw the conclusion that Joe was a gatecrasher and treat him accordingly? Should I go about asserting that Joe was a gatecrasher - should I inform his friends, his family, his employer for instance? Should Joe be arrested and convicted and appropriate punishment applied to him? Most of us would agree that this would be exceedingly unfair and, under prevailing legal practice, statistical evidence of this kind would not be deemed a legitimate basis for a conviction.But if I'm not justified in believing that Joe was a gatecrasher, then justification is not simply a matter of likelihood after all - there must be more to it than this. Interestingly, if an eye-witness claimed that she saw Joe scaling the fence to get into the concert then we would generally have no qualms about believing that Joe was a gatecrasher and acting accordingly - and such evidence could well be deemed a legitimate basis for conviction in a court. And yet, we're all aware that testimony is fallible - the eye-witness could be lying or mistaken. Indeed the probability that Joe was a gatecrasher given that a witness testified that she saw him scaling a fence is plausibly not as high as 99%. My primary aim in this project is to try and get to the bottom of this puzzle and, in so doing, come to a new understanding of what it takes to be justified in believing something. The research is relevant not just to epistemologists and academic philosophers but to anyone who has wondered what to believe about a given topic or how to evaluate a given piece of evidence. As the above example suggests, the research also has a bearing on questions about how best to understand and rationalise our legal practices. I aim to publish the results of this research in a monograph and a research paper. The research will be conducted primarily in Scotland and in collaboration with a number of philosophers working throughout Scotland. Scotland, at present, boasts a number of high profile philosophy departments and research centres with particular strength in epistemology.