Thursday, June 30, 2011

Epistemic luck (intro)

When I was about 10 years old, I had a philosophical discussion – or disagreement anyway, as we did not get very much farther than simply stating our positions (and then again, more firmly) – concerning the definition of knowledge. My interlocutor claimed that he could know that 2 + 2 = 5, even though that statement is false. I replied, that no, you can't know something unless it's true. You may believe that 2 + 2 = 5, but that's not enough: you can't know it (even though, in stating your belief, that may indeed be what you'll say), because knowledge must be true.

Contemporary philosophical consensus on the subject comes down on the side of my youthful self: knowledge must be belief which is true (so there, D___ O_____!). However, it is also generally accepted that a third condition is also required. And by "generally accepted" I mean "virtually universally simply assumed, on page 3 of just about every Epistemology 101 textbook, where no dissent is ever even imagined, let alone conceded actually to exist, let alone taken seriously." And yet not only are there more than one of us dissenters (for so I am), but we do so in various ways for various reasons. Analytic epistemology can be as dreary a subject as there is in philosophy, but as it turns out this is really, really important to get clear on, for reasons I hope to make clear as we continue.

The proximate stimulus for this current effort was a post (now two actually) on the New York Times's online philosophy blog The Stone. Distinguished Notre Dame philosopher Gary Gutting there deigns to enlighten the great unwashed about the importance of justifying one's beliefs, as opposed, apparently, to believing any old tosh that enters one's head:
Apart from its entertainment value, Harold Camping’s ill-advised prediction of the rapture last month attracted me as a philosopher for its epistemological interest.  Epistemology is the study of knowledge, its nature, scope and limits.  Camping claimed to know, with certainty and precision, that on May 21, 2011, a series of huge earthquakes would devastate the Earth and be followed by the taking up (rapture) of the saved into heaven.  No sensible person could have thought that he knew this. Knowledge requires justification; that is, some rationally persuasive account of why we know what we claim to know.  Camping’s confused efforts at Biblical interpretation provided no justification for his prediction.  Even if, by some astonishing fluke, he had turned out to be right, he still would not have known the rapture was coming. [my bold]
True belief, in other words, is not enough for knowledge: we need justified true belief (JTB).

I can already feel my eyelids drooping, so let me say this just the once and I'll be done with it. Since my dissertation (which went into it at some length, to little avail) I have resisted writing about this stuff, because to do so requires what seems to be the sort of academic nitpicking and intuition-mongering that makes so much contemporary philosophy look pointless and arcane. But if we get this right the payoff could be immense (for some of us at least, and part of my point is that you may not yet know who you are), so I hope I can count on the reader's patience as we wade together through the bog.

A relative newcomer to the current epistemological scene, Duncan Pritchard already has a zillion papers (available on his website) and an important book on the subject. As he explains on p. 4 of Epistemic Luck, the idea that true belief does not suffice for knowledge is very widely held indeed, to the point of invisibility:
[A]s befitting its status as a universal intuition—what these days we philosophers tendentiously call a 'platitude'—one finds this thesis both everywhere and nowhere at the same time. That is, whilst this line of thinking is clearly being presupposed in much of contemporary epistemological thought, the thesis itself is rarely drawn up to the surface of discussion, and even then it is left to stand as it is: a pure platitudinous intuition that is in need of no further explication.
As an example Pritchard points us to the SEP article on knowledge by Matthias Steup, who faithfully reflects the consensus of the discipline with only a gesture in the direction of argument:
Why is condition (iii) [i.e., the justification condition on knowledge] necessary? Why not say that knowledge is true belief? The standard answer is that to identify knowledge with true belief would be implausible because a belief that is true just because of luck does not qualify as knowledge. Beliefs that are lacking justification are false more often than not. However, on occasion, such beliefs happen to be true. Suppose William takes a medication that has the following side effect: it causes him to be overcome with irrational fears. One of his fears is that he has cancer. This fear is so powerful that he starts believing it. Suppose further that, by sheer coincidence, he does have cancer. So his belief is true. Clearly, though, his belief does not amount to knowledge. But why not? Most epistemologists would agree that William does not know because his belief's truth is due to luck (bad luck, in this case). Let us refer to a belief's turning out to be true because of mere luck as epistemic luck. It is uncontroversial that knowledge is incompatible with epistemic luck. [again, my bold]
Steup continues, naturally enough, given the purpose of mentioning it at all, to discuss the effectiveness of the justification condition for its purpose, that of ruling out lucky guesses ("mere" true beliefs) as cases of knowledge, which leads to the next section of the Gettier problem.

Now Steup does seem to give an argument here, with his cancer example. But when we look at it we see that there is no argument at all. Even though for some unknown reason only "most epistemologists" would agree here (who dissents and why need not concern us here, as we are busy men), our platitude is "clearly" true, and, again, "uncontroversial" for no actually stated reason. I should add that given the actual state of the discussion at this point Steup is entirely justified [!] in telling the story this way for this audience. No reason stretching out an already long SEP article with tracking down every last objection and quashing it.

Okay, let me just close for today with this last quote. Even after a couple of objections to his JTB account in the comments, which we will discuss later on, Gutting came back in his most recent piece with this:
Plato long ago pointed out [aside: I love that rhetorical tic! If he pointed it out, it must be true – so I need not argue for it, right?] that it is not enough just to believe what is true. Suppose [oh good, another example] I believe that there are an odd number of galaxies in the universe and in fact there are. Still, unless I have adequate support for this belief, I cannot be said to know it. It's just an unsupported opinion. Knowing the truth requires not just true belief but also justification for the belief.
Again, no argument at all, just an example. It's just obvious!

All right, so I've established that the consensus view is the consensus view. Next time I'll start distinguishing the various ways one might object to this "universal consensus."


Old 333 said...

Most amusing. Try page 9 of this book, if you have time for some more fun along similar lines.

And thanks for the read!

Daniel Lindquist said...

Wow, you weren't kidding about having something new up soon.

I still need to type up my notes from McDowell's visit last fall; at the time I remember feeling convinced that what he was saying there showed there to be utility in distinguishing knowledge and true belief after all (but not in a way which gets the Gettier train running, and not in a way which makes epistemic luck anything less than crucial -- those mistakes are still mistakes, and Levi's approach is still an attractive cutting of a gordian knot). But I now find I can't remember quite what the utility was.

I think it was something related to the explanation of action -- attributions of knowledge that p was supposed to function differently than attributions of mere belief that p for certain sorts of explanations of action, but he created examples where this showed up and it could only be an explanatorily irrelevant coincidence if the belief in question happened to be true. So a true belief that p and knowledge that p showed up as different sorts of states to attribute to someone in explaining certain sorts of action. Something like that. I can't remember the examples -- hopefully my notes are better than my memory. It was tied to his disjunctivism about explanations of action; he was replying to some of Dancy's stuff.

I'm curious to hear more about the Pritchard book; I've read a handful of his articles, but I only know his book through the NDPR review (which is by my academic adviser here at IU -- I should really read some of his work, beyond that review...).

Duck said...

Thanks Old 333, I'm a big Lem fan. Note that my boyhood interlocutor granted that 2 + 2 = 5 was false, but simply insisted that he could "know" it anyway.

Daniel: hold that thought about the difference between knowledge and merely true belief. My sense is that the utility you speak of is real (why wouldn't it be), but that McD's insistence on marking it that way keeps Levi's view from being visible as a possibility in other contexts. See what you think when we get there.

I've actually only read bits of the Pritchard book, but it looks really good (if you like that kind of thing). You might browse the extensive list of papers at his site for topics of interest (Moore vs. Wittgenstein for example).

Happy Independence Day, everyone!

Old 333 said...

I wonder if belief amounting to truth is a bit like flying - you have to be the kind of animal born with wings for it...

Or use a machine, I suppose. A God machine with mistaken beliefs!

I've stuck 3 Quarks in my reader, and will look for your post there. Thanks as always for the reading.

Ben W said...

So where's the next part?

JBS said...

"Plato pointed out" reminds me of "As seen on TV!".
I am less than your past 10 year-old...but for me 'knowledge' is that which no-one can take from me...and which I can not give...experiential memory.