Experimental Philosophy discusses the following experiment. Participants were told a story of Tim, whose wife is cheating on him. He gets a lot of evidence of this, but tells himself it isn’t so.

Participants given this case were then randomly assigned to receive one of the two following questions:

• Does Tim know that Diane is cheating on him?

• Does Tim believe that Diane is cheating on him?

Amazingly enough, participants were substantially more inclined to say yes to the question about knowledge than to the question about belief.

This idea that knowledge absolutely requires belief is sometimes held up as one of the last bulwarks of the idea that concepts can be understood in terms of necessary conditions, but now we seem to be getting at least some tentative evidence against it. I’d love to hear what people think.

I’m not surprised – often people say explicitly things like ‘I know X, but I really can’t believe it yet’. This seems uninteresting from the perspective of epistemology. ‘Believe’ in common usage just doesn’t mean the same as what it means in philosophy. Minds are big and complicated, and ‘believing’ is about what you sincerely endorse as the truth, not what seems likely given the information you have. Your ‘beliefs’ are probably related to your information, but also to your emotions and wishes and simplifying assumptions among other things. ‘Knowing’ on the other hand seems to be commonly understood as about your information state. Though not always – for instance ‘I should have known’ usually means ‘in my extreme uncertainty, I should have suspected enough to be wary’. At any rate, in common use knowing and believing are not directly related.

This is further evidence you should be wary of what people ‘believe’.