Saturday, October 14, 2006

"Justification?"

[This post is closely related to Trent's prior post, "Evidentialism for Everyone."]

So John (Kwak) and I tried to hash a little bit of this out yesterday in the grad office, and I think we arrived at conclusions something near to the following. (This is all very rough, and painting in some fairly broad strokes; I'm writing it in a hurry while out of town for the weekend. John, correct me if I’m mistaken or have missed something.)

First, one helpful way of framing the whole internalism/externalism debate – or at least of getting situated when it comes to remembering what we’re talking about – is to think back to Gettier’s article. Prior to that, I think it’s a fairly safe assumption that most people were thinking of knowledge as something like justified true belief, even if they weren’t entirely explicit about it. Gettier of course showed that something more is needed, and that’s more or less where it all began. One of the quests became that of discovering an analysis of knowledge that was not subject to Gettierization.

All of that should be fairly uncontroversial, but it seems like controversy lurks somewhere very nearby. It begins like this: some people ended up taking it that “justified true belief” is pretty clearly some central element of knowledge, but that it needs to be supplemented by some de-Gettierization condition (no false grounds, no defeaters, whatever). Others ended up taking it that “justified true belief” just is knowledge, but that the “justified” part of the analysis needs to be redefined in some way. (Hence we have Goldman’s “What is Justified Belief?” article saying more or less that reliability is central to justification.) Finally, another group apparently ended up keeping the “true belief” part of the analysis but replacing the “justified” part by something that was supposed to be entirely different – warrant, or perhaps “positive epistemic status” (whatever that involves), say.

At the same time, despite these differences, some fundamental similarities ended up being present in all of the above views. Aside from holding the “true belief” part fixed, each of the above views seems to hold that there is some crucial evaluative “thing” in the analysis of knowledge, and that satisfying that evaluative condition (along with true belief) either gets you pretty close to knowledge – you just need to add some “fillip” to prevent Gettierization – or in fact gets you knowledge (in that this evaluative element – once it has been redefined in some way different from the traditional analysis that Gettier criticized – is itself enough to prevent Gettierization). So, more or less, the crucial thing in the debate about knowledge and justification became figuring out what exactly this evaluative “thing” is.

But part of the problem should now be evident. Once the focus was shifted to this evaluative “thing,” it seems like the fact that one group took this evaluative “thing” to be all that you need in addition to true belief to get knowledge – usually, if not always, these are externalists – and that the other group took this evaluative “thing” to be not sufficient (in addition to true belief) for knowledge, was lost. And the worst part about it was that, even though they were apparently talking about different sorts of evaluative notions – because one was intended to be all that you needed in addition to true belief, but the other was never intended to be such – they all (at least initially, particularly with Goldman’s article) wanted to call these evaluative notions justification. Or, if these notions weren’t explicitly called “justification,” they were at least all taken to be that “thing” that fits into the analysis of knowledge where “justified” used to be (although, again, some took it that whatever this “thing” is is sufficient, along with true belief, for knowledge, and others took it to be something that needed to be supplemented by a fourth condition). It seems like that is precisely where all of the confusion really begins.

How so? Well, look at what’s going on here with (EJ). It’s probably should be platitudinous. Why would someone deny it? Well, perhaps it's because of the phrase “epistemically justified” contained therein. Some people read it and presumably think, ‘Oh, here we’re talking about that third, not sufficient condition (in addition to true belief), that needs to be supplemented by a fourth condition in the analysis of knowledge. Sure, fine; obviously that thing is characterized by (EJ).’ But then the others read it and think, ‘This is supposed to characterize that “thing” that gets us knowledge in addition to true belief? No way! Clearly we need something more than “fitting the evidence,” or we’re subject to immediate Gettierization!’ (I think that maybe this latter way of thinking could be exemplified by Plantinga, at least in certain places; hence Trent complains that he treats evidentialism as a theory of warrant when it was never intended to be such. He apparently takes it to be providing a characterization of this third and sufficient (plus true belief) condition, whereas the evidentialist (usually?) doesn’t mean that -- he (the evidentialist) wants for it to be a characterization of that third but supplemented condition in the analysis of knowledge.)

So here we see the rationale that may -- perhaps wrongly -- be underlying rejections of (EJ). And, further, it seems like the only way to sort these things out is to be almost drastically explicit about what we’re talking about when we say things like “justification,” and “epistemically justified.” Once we are drastically clear about that, I think we end up seeing that there’s a lot less disagreement than there initially appears to be. For, as I’ve now noted to Trent in e-mail correspondence, if we’re entirely clear that the “justification” characterized by (EJ) and (M) is something other than this “thing” that, in addition to true belief, is by itself supposed to get us un-Gettierized knowledge, then it seems entirely possible that a proper functionalist externalist could also be a mentalist (as characterized by (M)) about justification. He could be a proper functionalist about warrant, and perhaps even hold that justification is neither necessary nor sufficient for knowledge, but still agree that justification – now explicitly not intended to be that thing which gets us knowledge in addition to true belief – is correctly characterized by mentalism (in fact, for some comments suggestive of something near to this, see Plantinga's comments about Chisholm on justification at p. 45 of his Warrant: The Current Debate). But, of course, it seems that most people would think that “proper functionalist mentalist” – even as here characterized – is an oxymoron (I say this based only on informal polls of a few philosophy graduate students :) ). That tells me that many, many people – probably including myself until recently – have been using notions like “justification,” "warrant," and so on, much too quickly, without being clear about what’s being said.

This, I think, is why I’ve said that philosophy needs to hit the "reset" button, and why John continues to urge the creation of some Philosophical Standards Institute that establishes how such terms shall be used from now on. :)

So, briefly: As long as we’re clear that “justification” means “that in-some-way evaluative ‘thing’ that is not intended to be sufficient (in addition to true belief) for Gettier-proof knowledge” (maybe something like “evidence-sensitive rationality,” if that's not question-begging) then (EJ) does seem platitudinous, and it seems like even (M) is not nearly as controversial as it may otherwise appear. Externalists may accept both. On the other hand, if by “justification” we mean “that in-some-way evaluative ‘thing’ that is sufficient for Gettier-proof knowledge (in addition to true belief),” then “controversial” is the name of the game. Even (EJ) will likely be vehemently denied by some.

In any case, I think (in my current hurriedness) that that’s close to where things stand… Maybe. :) I'd love to hear further thoughts on this.

6 Comment(s):

  • Jason, nicely done! You've just summarized some of the key ideas I've been working on a paper for (hence, the "Justified Confusion" document I briefly showed you). Perhaps we should co-author? :-)

    One thing I might add to your discussion though is that I think that even when justification is fixed as a label for the third condition which when added to T, B, and de-G gets you knowledge, I think many would still debate whether it is platitudinous that justification is all about the evidence (i.e., EJ). What *is* platitudinous for all (I'd hope) is that internal rationality is entirely a function of internal mental state evidence. But I think many would question whether internal rationality is enough to get you justification, even when it is understood as just the third out of four necessary conditions for knowledge. The only reason why EJ *seems* platitudinous as a statement about justification is because it seems so clear to some internalists that internal rationality is all you would need for justification (again, even as the third out of four conditions for knowledge). But I think that itself is not clear to some. Case in point: Plantinga has a de-Gettierization condition (whether he incorporates that into his proper function account or not is immaterial in my view), but I don't think he'd say the remaining evaluative element will be had solely in virtue of internal rationality. He'd say there has to be more. In contrast, I think for Feldman and Conee, it just seems plainly clear that internal rationality *is* enough, and it kind of mystifies them that some think otherwise.

    By Blogger kdfkwak, at 10/14/2006 5:24 PM  

  • Jason,

    1. You write: "Gettier’s article. Prior to that, I think it’s a fairly safe assumption that most people were thinking of knowledge as something like justified true belief, even if they weren’t entirely explicit about it."

    I'm not so sure about that. See WCD, p. 6 regarding this. (Another interesting p. 6 is p. 6 of the 1st edition (1966) of Chisholm's ToK where he considers "having adequate evidence" as what Plantinga calls "warrant". (Furthermore, 3 x 6 = 18 and if you but a 6 on 18 you get 186 and it's on p. 186 of WPF that P says that he thinks Rich and Earl think justification is warrant (which is *manifestly* false). Yes, I'm co-authoring an epistemology text with Dan Brown)).

    2. You write: "it seems entirely possible that a proper functionalist externalist could also be a mentalist (as characterized by (M)) about justification."

    Those of you who've read my "Realizing Virtue" paper have seen an example of this very thing. PF isn't mentioned by name, but it's a pretty clear implication and easy extension. It presents an externalist theory of knowledge with a contingent relation to a mentalist-evidentialist theory of justification. If you've lost your copy, it's available on my Philosophy Page.

    The leading proponent of separating the theory of knowledge and justification is Richard Foley. See his "Trial Separation of the Theory of Knowledge and Theory of Justification" in Greco's _Sosa and His Critics_, his "Justified Belief as Responsible Belief " in the _Contemporary Debates_ volume, and the first couple of chapters of his _Self Trust_ book. I think he's on the right track in many respects.

    3. You write: "if by “justification” we mean “that in-some-way evaluative ‘thing’ that is sufficient for Gettier-proof knowledge (in addition to true belief),” then “controversial” is the name of the game. Even (EJ) will likely be vehemently denied by some."

    Well...not exactly, right, because one could just give, as Goldman sometimes does, an externalist theory of justification with an externalist theory of evidence as well, then EJ is platitudinous again, though seeing the consistency with M is harder to see (though I've argued for it by arguing that externalists are giving accounts of evidence *fitting* but I'll post that seperately).

    By Blogger Trent_Dougherty, at 10/14/2006 7:37 PM  

  • Jason,

    Oh, BTW, I actually don't think anyone means by "justification" that which can turn true belief into knowledge, at least not since Gettier. My take is that some people use it for that which turns true, de-Gettierized belief into knowledge, or synonymous with internal rationality. I think it'd be a challenge to see anyone since Gettier using the term "justification" for that which epistemizes mere true belief.

    By Blogger kdfkwak, at 10/14/2006 9:03 PM  

  • Quick clarification: In my last short post when I said: "My take is that some people use it for that which turns true, de-Gettierized belief into knowledge, or synonymous with internal rationality," I meant those as two distinct options. *Either* that which turns true, de-Gettierized belief into knowledge, *or* synonymous with internal rationality.

    By Blogger kdfkwak, at 10/14/2006 9:05 PM  

  • Note that if certain luck theorists in epistemology are right (Riggs, Greco, Dougherty), then there really is one thing that turns true belief into knowledge: sufficient credit.

    For details see Wayne Riggs: "Why Epistemologists are so Down on Their Luck" available on his website (probably has appeared by now), John Greco, "Knowledge as Credit for True Belief" in the Zagzebski and DePaul Virtue Epistemology volume, and my "Realizing Virtue" available on my website (plug, ahem, plug).

    By Blogger Trent_Dougherty, at 10/14/2006 9:17 PM  

  • The same should also go for some Safety theorists, especially Pritchard (who should have been mentioned among the luch theorists, but I really think his notion of luck is reverse-engeneered to be that-which-safety-eliminates, so I really think he's more of a safety theorist).

    Sosa's view is more complicated, but that might be a third example of a view with a unitary account of warrant.

    By Blogger Trent_Dougherty, at 10/14/2006 9:20 PM