Gettier cases and the JTB analysis

Brian Weatherson has a new argument on his blog for why gettier cases are cases of knowledge. His first argument was something like this: an analysis of knowledge ought to be relatively simple, and gettier cases, if they are not cases of knowledge, conflict with this point. He now has given up on this argument, and uses three others in its place. The three arguments rely on the following claims:
1. Knowledge is a norm of belief.
2. Knowledge is central to practical reasoning, in the sense that if you reason using something you don’t know to be true, you’ve done something wrong.
3. Knowledge is the norm of assertion.

I won’t comment on these three premises further; I’ve already said a number of things here about each of them. But in a recent post, Weatherson adds another argument.

It begins by assuming that one can be happy about the truth in question in a gettier situation: you can be happy, for example, that someone in the room has ten coins in their pocket. The argument then relies on the following claim: if I’m happy that p, then I know that p. The plausibility of the argument from this claim is Williamson’s idea that knowledge is the most general factive mental state, so that if you are in some factive mental state, it follows that you know. Since being happy that p is a factive mental state, one can infer that knowledge that p underlies this state.

I feel outSmarted: isn’t this just embracing an obvious counterexample to the claim that knowledge is the most general factive mental state? It reminds me of reading Nozick when he claims that knowledge is not closed under conjunction elimination.


Gettier cases and the JTB analysis — 5 Comments

  1. I don’t see the role of the Gettierized belief here (maybe I should hop over to BW’s blog to read his post…well, maybe later). If I understand things, the argument just needs any old proposition which happens to be true. So, consider Serendipitous Sally. She has a belief (formed completely at random, or on the basis of anti-wishful thinking, or [insert your favorite bad way of forming a belief here]) that her favorite team lost last night’s game, she happens to be right. But she’s not happy about it: she regrets that they lost. The argument here seems to predict that she knows they lost. So true belief is sufficient for knowledge, so long as you subject such beliefs to other factive attitutdes like regretting, or remembering, or being happy about, or…. Something has gone awry, I’d think. The obvious thought, as Jon points out, is that it is Williamson’s “know is the most general FMSO” claim. (We had a reading group once upon a time at Texas, and I remember something like this being brandished as an obvious counterexapmle to this claim back then.)

  2. Thony, you’re right that the belief need not be gettiered for purposes of the argument. True belief will work well enough, and if you remember Hawthorne’s views, he’s willing to say that sometimes knowledge is nothing more than true belief (or more accurately, “S knows that p” is sometimes true when S’s belief is true and none of the other, ordinarily proffered conditions for knowledge are satisfied). Of course, that aspect of Hawthorne’s view is insufficient to accommodate the present argument, since the present argument requires that true belief is knowledge whenever one might be happy about the content, or sad, or regretful, etc.

  3. Thony, I agree with you on cases of true belief, but the argument extends more generally to all beliefs, and I think your example supports this. The day after Sally’s team plays, she believes they lost (even though she cannot justify that belief), and spends all morning depressed because of this belief (even though she has no way of knowing whether it’s true or false). If she finds out later that day that she was right, does this somehow retroactively validate her emotional state from before? She’s probably even more depressed, because whatever chance there was that her belief was wrong is now eliminated (assuming that beliefs we have for no good reason carry less weight than those based on good information). The fact that it is now known to be true doesn’t change the way she felt all morning. If it had turned out to be a false belief, would she then not have been sad? The answer is pretty obviously no, unless the belief’s _having been_ true earlier had some bearing on her emotional state, even though she had no way of knowing that at the time.

  4. I actually discussed this very question with Tim (Williamson) some time ago (probably back in 1995 or 1996, shortly after the appearance of “Is Knowledge a Mental State?”); Tim insisted that, strictly speaking, ‘S is glad that p’ cannot be true unless S knows that p.

    For my part, I simply lack firm intuitions on these questions. I am inclined to be quite suspicious of those who claim to have firm intuitions on these points.

    My own form of contextualism would allow that in particularly relaxed contexts, the “reliability” condition on knowledge amounts to nothing more than truth in the actual world — so in these contexts it would be true to call any justified true belief “knowledge” (at least so long as the belief isn’t in a particularly strong sense “based” on any false lemmas). So I might be able to accept a contexualist version of the idea that knowledge is the most general factive mental state. But I’m more tempted to follow Jon’s example and just deny that idea (e.g. perhaps it’s just a common conversational implicature of ‘S is glad that p’ that S knows that p?)

  5. Ralph, I think this difficulty is worse if the requirements for knowledge do not vary with context. Even with such variation, however, we get the following very strange result. We’re talking and I think I know that p is true, and it’s really good news for me. So I say that I’m ecstatic that p is true. We talk, and you show me that there’s evidence I haven’t taken into account, and so don’t know. You then show me why the evidence is misleading, so that we now both know that p is true. You then say, “so even though you weren’t ecstatic that p a moment ago, you are now.” I’d be dumbfounded by your remark…

Leave a Reply

Your email address will not be published. Required fields are marked *