Epistemic Value

Tuesday, April 25, 2006

Survey Article on Epistemic Value

I'm currently in the process of writing a survey of recent work on epistemic value, and now have a rather rough-and-ready draft to hand. I've posted this on the 'drafts' side-bar below, and you can also download it here. All comments gratefully received!


  • At 7:59 PM, Blogger Albert Atkin said…

    Duncan, I don't know if people are e-mailing you comments rather than posting - but, I have a quick(ish) question (rather than a comment) about one of the points you make about Greco's Virtue Reliabilism and thought I'd post it instead of e-mail it.
    You question Greco's point that in Gettier cases the belief isn't formed because of the operation of the agent's virtues. In particular, you question why we would say such a thing in a Henry-and-the-Barns scenario. I agree that it looks as though the belief is formed because of the operation of the virtues, but why can't we just say that with that example, it merely goes to show that it isn't a Gettier case?
    I say this for all the usual reasons about diverging intuitions on whether or not Henry really knows it is a barn in that example. Teaching that case, I've found that most of students want to say that he knows - he did all the right things.

    Are there not other Gettier type cases more obviously repugnant to our intuitions, but where Greco's claims don't seem to hold? Or do you think Greco wouldn't/couldn't be inclined to say "Well, that shows this is not a real Gettier case afterall".

    For what it's worth, I was thinking of his analogy about athletes to draw out the difference between a beliefs being true and virtuously formed and true because it's virtuosly formed, and wondered whether the athlete's winning by her athletic virtues in a slightly deviant way might make space for a Gettier type problem.

    So,for example, her fellow athletes see her virtues and are de-motivated by them, thereby not running so hard. On the one hand she wins, because the athletes under-perform, but on the other, she wins because her athletic virtues de-motivate them. Has she not won, because of her virtues?

    An analogous case for knowledge might run (I guess) where I am a disillusioned teacher who systematically teaches my students falsehoods. However, one of my students impresses me with her epistemic virtues and instead of giving her the text book that teaches (say) geocentrism, I give her the textbook that teaches heliocentrism. On the one hand she forms the true belief that the solar system is heliocentric because I chose not to deceive her, but on the other hand she forms this true belief because her epistemic virtues inspire me to do my job. Has she formed a true belief because of her virtues?Don't know.

  • At 11:30 AM, Blogger Duncan Pritchard said…

    Thanks for this Albert. You're right that resisting the barn case is a possibility, though it's not one that I find appealing (Hetherington takes this line in his PQ paper 'Actually Knowing' as I recall, though he's suspicious of all Gettier cases of this sort I gather, and doesn't have the same anti-luck intuitions about knowledge that I do). It's interesting that your students are unsure about this one though (mine tend not to be, though maybe that's because of my influence!).

    More generally, I agree with you that the 'because of' relation in the virtue-theoretic account of knowledge is problematic--it's certainly something that has concerned me since it's hard to spell it out in an uncontentious fashion. Greco makes the best stab to my mind by appealing to explanatory salience--this might well deal with some of the cases that you mention, but I don't doubt that it will leave others undecided. Jennifer Lackey has some nice discussions of these issues in her forthcoming Synthese paper on knowledge and credit, which is, I beleive, posted on her webpage.

    Incidentally, let me take this opportunity to thank all those who have send me feedback on the paper by e-mail: Erik Olsson, Andrew Moon, Mike DePaul, and Don Fallis (I hope I haven't left anyone out!). An updated version will be posted soon, but in the meantime any thoughts readers have would be gratefully received, whether by e-mail or through the blog.

  • At 7:30 PM, Blogger Albert Atkin said…

    Thanks Duncan: on resisting the barns case, I was also thinking of the Gendler-Szabo and Hawthorne paper in Phil Studies where they talk about the instability of intuitions for these kinds of case - in particular, getting the time from the only honest person in a room full of liars. Such a case doesn't look different from the Barns case (in virtue-theoretic terms either) but intuitions differ (unless I taught you!).
    Glad you thought 'because of' was worrying too.

  • At 1:41 PM, Anonymous Martijn Blaauw said…

    Hi Duncan and Albert,
    I've been worrying about the 'because of' notion as well, and it seems to me that there is an additional problem. Greco explicitely models his virtue epistemology along the lines of epistemological contextualism (indeed, he calls his theory 'virtue contextualism'). But a well-known challenge for all contextualist accounts of 'knows' is that the contextualist should propose some sort of linguistic precedent for 'knows'; the contextuality of 'knows' shouldn't be a unique contextuality that is only associated with 'knows'. It seems pretty hard, however, to think of a linguistic precedent for a virtue contextualist treatment of 'knows'. That is, I'm not sure if there are any other verbs that are context dependent along the lines of 'explanatory salience'.

  • At 11:44 PM, Blogger Albert Atkin said…

    Martin, do you think that the task facing Greco is the very same task facing, say, Derose or Cohen? Or can he appeal to a wider range of context-sensitive phenomena?

    If he can, then here's a suggested approach, related to something I've been thinking about for a little while, though I'm not sure how strong it is or if it is properly in-line with John Greco's virtue-contextualism.
    Maybe Greco doesn't need to find a precedent based on the context sensitivity of adverbs, but can look for some other kind of context-sensitive expression, in particular, demonstratives. The reason I say this is that the notion of "explanatory salience" suggests (to me) the kind of context-sensitivity which we find in demonstrative pronouns. A use of “that” needs an object which is either self-salient, or which we make salient with a demonstration (or intention). And, part of the contextual variation we find in demonstratives is the variation in what makes an object salient (as well as the very object picked out). Maybe we can treat a virtue-contextualist "knows" as being made salient by whatever counts as an explanation in that context.

    Potentially, then, the way to treat "knows" on the virtue-contextualist account (with its emphasis on explanatory salience) is not after the contextualism of DeRose or Cohen who treat "knows" as a gradable adjective, but rather after demonstratives which do have fixed meanings, but inherits (as it were) its context sensitivity from whatever it is that makes it object salient.

    I think I'm being over-impressed with the word "salience" here, and that the parallel won't really run, but either way, is there any reason why Greco can't appeal to a wider range of phenomena?

  • At 1:29 PM, Anonymous Martijn Blaauw said…

    Hi Albert,

    It seems to me that the task Greco faces is the same as the task DeRose and Cohen face. I do not think, however, that the task of DeRose and Cohen is limited to finding gradable adjectives / indexicals that can serve as a model for 'knows'. Presumably, DeRose and Cohen could, in principle, appeal to a wider range of context-dependent phenomena. There do seem to be at least two constraints for what can serve as the strongest type of precedent for 'knows', or so it seems to me.

    1-The model should be a verb, just like 'knows' is.
    2-The model should be factive, just like 'knows' is.

    And 1 and 2 thus would also constrain what Greco can appeal to.

    As to your own (intriguing!) proposal, linking 'knows' to demonstratives would be out from the start, at least if we accept 1 and 2.

    What do you think?

  • At 2:10 PM, Blogger Albert Atkin said…

    Martijn, I agree with you! I think my idea clearly breaches your original point that 'knows' needed an obvious precedent - modeling 'knows' after a demonstrative would have made it a pretty peculiar beast. I'd just wondered whether Greco was as strictly commited to the linguistics for 'know'
    as other contextualists - and if not, could we find play from the idea of salience. I think if 1. and 2. are base-line requirements, then yes, he has the same task.

  • At 2:33 PM, Blogger Duncan Pritchard said…

    Hi Albert and Martijn,

    Thanks again for your input to this discussion. As a postscript to this issue of how best to understand the 'because of' relation on virtue-theoretic accounts, I briefly discuss this issue in my paper, 'Virtue Epistemology and Epistemic Luck, Revisted'. The latest version of this can be found on the 'drafts' side-bar, and can also be downloaded here.

  • At 2:36 PM, Blogger Duncan Pritchard said…

    That link didn't work, so let's try it again:



Post a Comment

<< Home