DJK

Dennis’ last post generated lots of great discussion. In the comment thread, Declan Smithies proposed a modified version of principle JK that Dennis and others argued against. Declan’s version of JK, what I’ll call ‘DJK’, goes as follows:

DJK: If you have justification to believe P, then: you have justification to believe that you know P, if you believe P

I think the following is a counterexample.

But before presenting the example, a couple caveats to bear in mind. First, on Declan’s view, having justification to be confident, to whatever degree, that P does not entail that you have justification to believe that P. Second, Declan’s proposal concerns propositional justification, which does not entail that the belief is doxastically justified (or, as it is sometimes put, properly based or well grounded).

Now for the example. Suppose:

(1) S has justification to believe P.
(2) S believes P.
(3) S believes P on the basis of a very bad reason, one that does not in fact help justify P for him.
(4) S recognizes that his belief’s basis is inadequate, but cannot be bothered to do anything about it. (Perhaps S is presently psychologically incapable of doing anything about it.)
(5) S knows that knowledge requires that reason-based belief be properly based on adequate reasons.

In such a case, S is not justified in believing that he knows that P. (If anything, he’s justified in believing that he does not know that P.) Yet he satisfies the conditions set out in DJK.


Comments

DJK — 9 Comments

  1. This is interesting, John. But, aren’t the original examples ones of a different character (in which the belief that P IS based on the “good” reason)? As such, why not go for the following simple amendment to DJK, which seems more salient, given the nature of the alleged counterexamples to the original principle?

    DJKR: If you have justification (or reason) R to believe P, then: you have justification to believe that you know P, if you believe P on the basis of R.

  2. Branden,

    That’s a good suggestion. Two points in response.

    First, DJKR will face a similar c-ex, with this added twist: S improperly believes P on the basis of R. (I present a case like this in a paper I wrote on propositional justification.)

    This leads to the second point. Naturally, one might then suggest a further revision:

    DJKR*: If you have justification (or reason) R to believe P, then: you have justification to believe that you know P, if your belief that P is properly based on R.

    This is now a thesis about doxastic justification — basically, it says that if you have a doxastically justified belief that P, then you have justification to believe that you know that P. I don’t know if Declan would be satisfied with this revision.

    Finally, I think that even DJKR* is false. Some beliefs preserved through memory are doxastically justified. But the subject might presently lack any evidence that she has the memory-belief that P. In such a case, she is not justified in believing that she knows P.

  3. Thanks, John: I agree that your case is a counterexample to DJK and I much prefer Branden’s DJKR and your DJKR*. But as the JK-inspired principle becomes more and more complex, perhaps it’s worth considering whether there is a simpler way of formulating the intuitive idea that we’re after. How about this?

    IJK: If you have justification to believe P, then you have justification to believe that you’re in a position to know P

    Now DJK, DJKR, DJKR*, etc. can seen as increasingly more accurate accounts of what you need to do in order to transform a justification to believe that you’re in a position to know P into a justification to believe that you know P. You need to believe P, you need to base that belief on the justification you have, you need proper basing, and perhaps even further conditions need to be met. But I think we can abstract away from these questions and ask: IJK itself on the right lines?

    Your memory case might be a problem, but I need to think more about it…

  4. Declan, how about this kind of argument against IJK? When you have justification that you’ve met all the epistemic conditions for knowledge (feel free to interpose a different notion of “in a position to know if you don’t like this one), i.e., that you are justified in a way that isn’t gettiered, then you have justification for thinking that closure of further inquiry into the issue is legitimated. But lots of stuff that we have justification for believing isn’t like this: witness philosophical theses themselves. Williamson, for example, had justification for thinking that E=K, I presume, but not enough that it was legitimate for him to close off further inquiry into the matter. Or the same for the knowledge norm of assertion. In some cases philosophers will claim that they do have good enough reasons to close off further inquiry, but that seems more bravado than substance, doesn’t it?

    The argument for the claim that if you are justified in believing that you’ve met the epistemic conditions for knowledge, then you are justified in believing that closure of further inquiry is legitimate derives from the connection between knowledge itself and legitimated closure of inquiry. One way to sustain this argument would be by adopting a closure principle about justification, but I don’t think that’s necessary.

  5. Something interesting about the principles talked about here and in the other post is that they stand in a complex dialectical situation with respect to the kind of position advanced by Fantl & McGrath in their papers on epistemic purism. If, like them, one wants to be a purist about how much justification one has, but an anti-purist about how much justification is needed for knowledge, then these kinds of principles have to be false. Of course, F&M’s Modus Ponens…

  6. Jon, that’s very interesting. I haven’t thought much about it, but I think I’d be inclined to put some pressure on the connections between knowledge and closure of enquiry. Ram Neta discusses a case of Jessica Brown’s and he says that even if the surgeon knows which kidney needs operating on, it may be rationally permissible, or even rationally required, for him to double check which kidney needs operating on. That sounds right to me. If so, then presumably the surgeon is justified in believing that he knows which kidney needs operating on, but he is not justified in believing that further enquiry (so to speak) can be closed.

  7. Juan, I don’t know F&M’s stuff well enough, but are they purist about justification to believe outright or are they just purist about justification for degrees of confidence? I don’t know if they draw that distinction, but I take it that someone could be purist about justification for degrees of confidence, but anti-purist about justification for belief, and so anti-purist about the kind of justification needed for knowledge, since knowledge entails justification for belief. That’s a package I’m quite attracted to myself and I think it’s compatible with the various JK-style principles.

  8. Good point, Declan, I hadn’t thought about Jessica’s example. Two points, though. First, note that the stronger claim is what is needed to undermine my argument: it has to be rationally required to check further, not merely permissible. Second, the connection between knowledge and closure of inquiry should be restricted to epistemic permissibility, I would think. Restricting the connection in that way makes the proposal much more obscure than the unrestricted claim, though.

Leave a Reply

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