Three reflections on Keith DeRose’s “latest contextualism paper”:http://pantheon.yale.edu/~kd47/OLB.pdf.

The first is probably trivial. The basic bank cases as Keith describes them prove precisely nothing about knowledge because the alleged knower in question doesn’t have the right _psychological_ states in order to know the bank is open. In particular, he doesn’t believe with something like subjective certainty that the bank is open. I’m not sure whether this is easily fixed by stipulating that he does indeed have such a belief, or whether this makes the case so unrealistic that we no longer have clean intuitions about it. Probably the first, so the worry is probably trivial, but I’m not quite sure.

Second, Keith spends lots of time talking about cases where the speaker is in a high stakes environment and the subject in a low stakes environment, but not much time talking about cases where the speaker is in a low stakes environment and the speaker in a high stakes environment. His basic theory appears to predict that knowledge ascriptions will go through fine there, though that isn’t my intuition. (My intuition about these cases is that I don’t have a clear intuition, for what it’s worth. I also don’t really have clear intuitions about the speaker high-subject low cases Keith discusses, but maybe I should have intuitions there.)

The final point is the biggest one. I half-agree with the spirit of the following paragraphs, but only half.

bq. Arguments from ordinary usage – appeals to what speakers of a language do or would say in applying the terms in question to particular situations (both positive and negative claims involving the term), appeals to which simple applications are or would be proper or improper for them to make, and appeals to intuitions as to the truth values of those claims in particular situations – don’t exhaust the considerations that need to be attended to in deciding among semantic theories, but do provide absolutely essential considerations.

bq. This importance is most forcefully impressed upon me when I consider some obviously false theories, and ask myself how (on what basis) we know that they are false, and the closely related, but perhaps even more important, question of what makes these theories false (in what does their falsehood consist)? So, to take one example of a theory that is obviously too demanding, and one that is obviously not demanding enough, consider, for instance, the crazed theory that I, following others, have discussed, according to which a necessary condition for the truth of “S is a physician” is that S be able to cure any conceivable illness instantaneously, and the theory about the meaning of bachelor like the traditional account, except that it omits the condition that S must be unmarried for “S is a bachelor” to be true, insisting that married men can be truthfully (if perhaps for some reason improperly) called “bachelors.” There are a couple of real clunkers! But in virtue of what is our language in fact such that these strange theories are not true of it? I’m of course not in a position to give a complete answer to this question, but, with respect to the strange theory concerning ‘physician,’ it seems eminently reasonable to suppose that such facts as these, regarding our use, in thought and speech, of the term ‘physician’ are centrally involved: that we take to be physicians many licensed practitioners of medicine who don’t satisfy the demanding requirement alleged; that we seriously describe these people as being ‘physicians’; that we don’t deny that these people are ‘physicians’; that claims to the effect that these people are ‘physicians’ intuitively strike us as true; etc. It’s no doubt largely in virtue of such facts as these that the traditional view, rather than the bizarre conjecture we are considering, is true of our language: The correctness of the traditional view largely consists in such facts. And these facts also seem to provide us with our best reasons or evidence for accepting the traditional, rather than the strange, hypothesis regarding the semantics of ‘physician’. Analogous points hold for why a traditional theory of ‘bachelor’, rather than the bizarre theory we are considering, is true, and for how we know it’s true.

As “I’ve said before”:http://www.brown.edu/Departments/Philosophy/homepages/weatherson/counterexamples.pdf, I think the spirit of this view needs to be pretty seriously qualified when we start working on the meanings of normative terms. If there’s a crazy view in philosophy, the view that an act is in extension of “good” if the vast majority of competent users of the term would judge that it is surely counts as crazy. And the same goes for all other moral terms. No matter how many competent and well-informed users of the term “wrong” endorse the sentence “Gay sex is wrong,” it does not follow that gay sex is wrong. And since gay sex is in the extension of “wrong” iff gay sex is wrong, it follows that speakers’ intuitions about the case don’t set the extension.

So while Keith is worried about crazy views about physicians and bachelors, I’m much more worried about crazy democratic views about evaluative terms. And since ‘knows’ is an evaluative term, as I think Keith agrees, I think my worries are more on point.

Here’s how to make that a little more targetted. I think when working out the extension of “good” we should systematically discount the intuitions of those who are driven by a false theory. And I think in some of Keith’s cases, the intuitions he’s looking for (particularly in the Thelma and Louise case, and also in the somewhat abstractly defined third-person bank cases) are driven by a false theory. In particular, they’re driven by the false theory that two people with the same evidence are in a position to know the same things.

Is that a false theory? Sure it is. It is inconsistent with SSI, and SSI is true, so it’s false.

How do we know SSI is true? It best systematises those intuitions worthy of systematising.

But aren’t you using SSI as the test for whether intuitions are worthy of systematising? Sure – it’s true so why wouldn’t I use it.

Isn’t this a little bit circular? No, it’s a lot circular. But I don’t see any way out of that. There’s a coherence/holistic/best balance kinda judgment to be made here, and those very rarely look good if you try and defend them on foundational grounds. But SSI does best on those tests.