Saturday, December 13, 2008

On Gettier's Notion of Justification

I recently had cause to trudge over to the library and check out the Shope volume. Don't know it? It contains just about every Gettier and Gettier-like case that found its way into print between 1963 and, I don't know, 1988. A friend told me that the most boring author he's ever read told him that it was the most boring thing he had ever read. I don't want to knock it, the volume has its virtues. In research for a response piece I'm working on, I needed some references and I found a reference to a piece I hadn't read before, Catherine Lowy's "Gettier's Notion of Justification" (here). It's really an eye-opener. I thought that this was an interesting passage:
It has become customary just to refer to Gettier's counterexamples as counterexamples to the justfiied true belief analysis of knowledge whereas Gettier actually specifies the general form of such analyses in the following way:
S knows that p iff,
(i) p is true
(ii) S believes that p, and
(iii) S is justified in believing that p,
and it is to this analysis that he claims to present counterexamples. The locution 'justified true belief analysis' is misleading here for it distracts from an important clee to the notion of justification which Gettier has in mind. It is important to note that Gettier uses the locution 'S is justified in believing that p' (rather than, for instance, 'S has the justified belief that l') consistently, not only in the justification condition of the definition but in the formulation of the crucial counterexamples and in the statement of his points about justification.

She goes on to say, "It seems clear to me to be clear in the light of Gettier's consistent use of the locution 'S is justified in believing that p' that the sense of justification Gettier has in mind is the sense in which a person is justified in believing something" (106). Fancy that.

Then there's this little gem of a passage:
...some writers have maintained that Gettier's renowned examples are not in fact counterexamples to the traditional analysis of knowledge, on the grounds that Gettier's second point about justification is wrong, i.e., that it is not the case that for any proposition, p, if S is justified in believing p and p entails q, and S deduces q from p and accepts q as a result of this deduction, then S is justified in believing q, and it is not the case, at least in the case where p is false

Here's why I think this is interesting. The distinction between personal justification and doxastic justification is often thought to be some ad hoc device externalists use to dodge objections. It seems to have historical roots that predate the internalism/externalism debate. It's also interesting that there was a time when people seemed to think that false beliefs can't justify further beliefs. These authors were pretty hard core externalists. Their views seem far more radical than even simple minded reliabilism.

At any rate, the specific reason I wanted to look at this volume was that there was a suggestion that W was in trouble because by defending E = K he had to deny that Gettier cases are possible. It's interesting to note that there were contemporaries of Gettier that seemed to think that Gettier's cases weren't genuine precisely because they involved forming a justified belief on the basis of a false belief and it seems that what convinced everyone that Gettier cases were unavoidable was the (apparent) discovery that this was an accidental feature of Gettier-like cases. It suggests that the question about whether false propositions constitute evidence was shelved, not settled. I suspect that the views that formed the dialectical situation of the late 70's were not so much rejected for good reasons but ignored. Some will start to come back in vogue much to the chagrin of epistemologists who think that these are newfangled views.

4 comments:

jon kvanvig said...

Hi Clayton, a minor worry about your last paragraph. I don't think those who appealed to the "no false lemma" idea were rejecting the counterexamples. They just thought that the examples weren't that difficult, since they could be handled with a simple addition that one not reason through, or depend on, falsehoods.

But I may be misremembering, and perhaps you have some to cite who actually claim that the examples aren't counterexamples because of the falsehoods involved?

Clayton said...

Hey Jon,

Good point, there certainly was that reaction and I wouldn't want to deny that. But, there were articles by Almeder where he suggested that Gettier's examples weren't genuine because beliefs that did not meet the truth condition did not meet the justification condition. Then there was Robert Meyers and Kenneth Stern that similarly suggested that the examples weren't genuine suggesting (if I read Lowy right) that Gettier had confused justified belief with some other notion. (There might be others. She mentions articles by Margolis, Pailthrop, Richman, and Thalberg as authors that accuse Gettier of confusing justification with other related notions.) It's interesting that her response is that these writers miss the point because the point all along had to do with personal justification rather than doxastic justification. Rather than insisting that they were wrong about DJ, she points out that they were wrong about PJ.

Cathy said...

Hi Clayton
Nice to see that someone is still reading my graduate work. I shall have to think about it again after these many years of work on other problems.

Cathy Lowy

Clayton said...

Dear Cathy,

I tried to write you a note, but couldn't track down an email address. Anyway, thanks for the fantastic article. I wish that I had read it years earlier.

All my best,
Clayton