So, to get things started, here is something I've been kicking around for a little while. Presumably a primary motivation for the theory of knowledge is the notion that knowledge is a valuable thing to have. While the following may be a little too strong, a first stab at that notion may be something like this:
KIV: A rational agent should never prefer, for epistemic reasons, to merely have a justified true belief that p, rather than knowing that p.
But, at least if the standard view regarding knowledge and lottery cases is correct, KIV is false. So I think we should either abandon the idea that knowledge is valuable, and so perhaps reconsider whether having a good theory of knowledge is a really important goal, or admit that we do have knowledge in (some) lottery cases. Here is how it goes:
Consider the following two cases.
(1) I give you a lottery ticket. I tell you that the lottery has one and only one winner, and that there are n tickets in the lottery. You form a justified belief that you will lose. And let this belief be true.
(2) I give you a lottery ticket. On the ticket is a bar code that tells you whether the ticket is a winner or not. I don't tell you how many tickets there are in the lottery, but I do give you a nifty lottery ticket bar code detector. This device reads the bar code and outputs "winner" or "loser." This reader, while not infallible, is highly reliable, indeed, lets say the likelihood of it providing a false loser reading is less than the chance of a spontaneous visual hallucination. But, lets say the chance of it getting a false loser reading is less than 1-1/n. So the reader says that you will lose the lottery, you form a justified belief that you will lose, and again let this belief be true.
According to the received view in epistemology, you lack knowledge in (1). However, it seems fairly plausible that you do have knowledge in (2). Your belief is highly reliable, it is certainly justified, it is true, there are no gettier effects in play, it is a very sensitive belief, and so on. If the ticket was in fact a winner, than (unlike in the normal lottery case) you almost certainly would believe that it was a winner. And yet it seems that you should prefer being in (1) to being in (2). After all, your evidence that you've lost is better in (1) than it is in (2) because the chance of the reliable detector not making an error is less than your chance of being in error solely from considerations of how many tickets there are (and, of course, we can make the lottery a huge one, so the discrepancy can be fairly large).
And this situation doesn't arise only for lottery cases. Consider any pair of cases where in the first our evidence consists of the base rate for some outcome, and in the second our evidence consists of the output of some highly reliable test (but without knowledge of the base rate). If the base rate of the outcome is (much) higher than the reliability rate of the test, it seems like we should prefer being in the former position than the latter. After all, it certainly seems as if our evidence is better in the first type of case. But, according to the received view in epistemology, we will lack knowledge in the purely statistical case, while we may have knowledge as a result of a highly reliable test.
I can think of several ways of resisting this conclusion, but I'll hold off on addressing them unless they happen to get brought up in any comments. My question is whether anyone else does think that the merely statistical cases are preferable to the reliable-test cases, and if so whether this means we should give up KIV (and so undermine the motivation for the theory of knowledge generally), or allow that we do, after all, have knowledge in merely statistical situations.
Sunday, March 22, 2009
Saturday, March 21, 2009
About this blog
This blog is primarily intended as a space for Princeton philgrads to float ideas and elicit constructive feedback from each other (and from any other visitors, should this blog come to acquire a wider audience).
Fellow philgrads: email me to register as a member of this blog, if you haven't already. You can begin posting (and commenting) immediately. To keep track of new posts, use the sidebar options to 'subscribe' via email or your favourite rss feed reader.
Publicity Concerns
We've disabled search engine indexing, so this site shouldn't directly appear in google listings -- but note that it is publicly accessible nonetheless. For those who are wary of publicly attaching their name to half-baked ideas, I've created an 'Anonymous' account that anyone can use. (Email me for the login name and password.) Note that you're also free to choose a pseudonymous screen name for your Blogger account.
Having said that: (i) it's generally nicer to know who it is that we're talking to, and (ii) if anything, I would expect that prudential considerations count strongly in favour of getting your name out there and recognized by other academics / future employers. So I'd personally encourage people to post under their own names. But other options are available for those who prefer them.
Suggestions
- In addition to testing out possible ideas and arguments (e.g. from your work in progress), I'd also strongly welcome 'summary'-type posts on any neat ideas -- encountered in class, say -- that the rest of us might benefit from. (See, for example, my summary of Nomy Arpaly on 'Rational Akrasia'.) That could be an efficient way for those of us taking different classes to fill each other in. And of course the process of writing a summary helps to cement your own learning, in addition to helping others.
- As a general rule, I'd advise aiming for around 400-700 words per post. I've found that more elaborate issues can often be explored most fruitfully by breaking them down into a series of shorter posts. But it's up to you.
- For archiving purposes, it's helpful to 'label' your post with (i) an author-specific label, e.g. '-YourName', and (ii) the broad area(s) of philosophy that it falls under: ethics, metaphysics, ancient, etc. (Special labels include 'admin', for posts like this; and 'academia', for posts on broader issues relating to the profession, academic life, etc.)
Have fun!
Fellow philgrads: email me to register as a member of this blog, if you haven't already. You can begin posting (and commenting) immediately. To keep track of new posts, use the sidebar options to 'subscribe' via email or your favourite rss feed reader.
Publicity Concerns
We've disabled search engine indexing, so this site shouldn't directly appear in google listings -- but note that it is publicly accessible nonetheless. For those who are wary of publicly attaching their name to half-baked ideas, I've created an 'Anonymous' account that anyone can use. (Email me for the login name and password.) Note that you're also free to choose a pseudonymous screen name for your Blogger account.
Having said that: (i) it's generally nicer to know who it is that we're talking to, and (ii) if anything, I would expect that prudential considerations count strongly in favour of getting your name out there and recognized by other academics / future employers. So I'd personally encourage people to post under their own names. But other options are available for those who prefer them.
Suggestions
- In addition to testing out possible ideas and arguments (e.g. from your work in progress), I'd also strongly welcome 'summary'-type posts on any neat ideas -- encountered in class, say -- that the rest of us might benefit from. (See, for example, my summary of Nomy Arpaly on 'Rational Akrasia'.) That could be an efficient way for those of us taking different classes to fill each other in. And of course the process of writing a summary helps to cement your own learning, in addition to helping others.
- As a general rule, I'd advise aiming for around 400-700 words per post. I've found that more elaborate issues can often be explored most fruitfully by breaking them down into a series of shorter posts. But it's up to you.
- For archiving purposes, it's helpful to 'label' your post with (i) an author-specific label, e.g. '-YourName', and (ii) the broad area(s) of philosophy that it falls under: ethics, metaphysics, ancient, etc. (Special labels include 'admin', for posts like this; and 'academia', for posts on broader issues relating to the profession, academic life, etc.)
Have fun!
Subscribe to:
Posts (Atom)
Posts
