Saturday, April 25, 2009

An argument with very limited appeal!

The following three propositions constitute an inconsistent triad:

1. Foundationalism: All knowledge and justified belief rest ultimately on a foundation of noninferential knowledge or justified belief
2. Knowledge Account: All justified beliefs constitute knowledge.
3. Knowledge from Falsehood: It is possible to derive knowledge from falsehood via inference.

In support of (3), consider:

Santa
Cooper’s parents tell him that Santa is going to leave presents under the tree and he believes he will have presents because he believes Santa is bringing them. We can rightly say he knows he will have presents even if we stipulate that given his background beliefs he would only believe he will have presents if he believed Santa was going to bring them.


We should give up Foundationalism.

There you go. The argument with very limited appeal!

Myself, I'd reject the knowledge account, but since I think that there are no false, justified beliefs and think that knowledge entails justification, I'd have to say that some justified beliefs rest on a foundation of unjustified beliefs. Or, something like that.

4 comments:

Lewis said...

It seems like your reply would involve giving up Knowledge Account and Foundationalism.

Kevin said...

We can rightly say he knows he will have presents even if we stipulate that given his background beliefs he would only believe he will have presents if he believed Santa was going to bring them.Isn't that an example of a belief that is not well-formed? We might say that the child really doesn't know. Perhaps because his causal chain is misguided, his understanding of what will constitute the event of having presents the next day is not equivalent to what the actual event will be. For example, his criteria for identifying that event (it was made possible by Santa) are not actually at play in the event (it was made possible by the parents, or some such).

I do think we should give up foundationalism, and I think there are grounds for rejecting the knowledge account. (So, all three propositions, while inconsistent, might be plausibly rejected anyway.) I'm still not clear why some justified beliefs need a foundation of unjustified beliefs. Although I think it would be plausible to say that justified beliefs can't be the end-all, because they presuppose perhaps irreducible conditions of rendering something like justification intelligible.

Definitely an interesting argument with very limited appeal!

Mike Almeida said...

Clayton,

Wouldn't almost any Gettier case get you knowledge from false belief, even given your rejection of justified false belief? I have evidence E for the belief that p = S owns a Ford. You infer q = someone own's a Ford. E justifies q pretty clearly, as the standard case goes (it cannot justify p, given your reasons). So, you derive q from a false, unjustified proposition.

But then take any old false proposition, say, p = it is raining on the cat. I might have evidence E = I observe the pouring rain on the animal (though, as it happens, it's a rat, not a cat). I infer q = it's raining. Clearly q is justified, and it's derived from the false, unjustified proposition, p.

Clayton said...

Mike, I suppose that's right. That, however, would be an argument with far less limited appeal! It should work for J-foundationalism, but I'm not sure the G cases work for K-foundationalism unless we have an inferred belief at some point that constitutes K and not just a JB.

I suppose the interesting q's are how K from falsehood is possible and what to say to regress arguments that suggest that it's not. But, that's for another time. I'm running off to Austin for the weekend to see the quasi-inlaws and for an engagement party (not mine).