Contextualism, Intellectualism, and Ignorant Third Persons

It's a little bit natural to think that 'knows' contextualism and the shifty kind of invariantist that's sometimes called an 'SSI theorist' or an 'IRI theorist' come to a bit of an intuitive draw considering two kinds of third-person knowledge attributions. High Howie has whatever features you think makes it harder to know, or makes 'know' express a stronger relation: he's thinking about skeptical possibilities, it's really important to him whether p, or whatever. Low Louie is just the opposite: to Louie, p's no big deal, he's not worried about it, whatever. High Howie says things like "I don't know that p," while Low Louie says things like "I know that p," and both utterances look pretty good, even though in some sense Howie and Louie look to be in identical epistemic situations -- they have the same evidence, or something like that.

(Now I happen to think that it's not at all clear how to make sense of that last stipulation. This basically amounts to a worry whether there is any correct generalization characterizing the difference between the shifty SSI-like views from 'classical invariantism'. But I'm setting that aside for now, assuming, as is usual in this discussion, that the sense in which Howie and Louie are in the same 'epistemic position' is tractable -- and does not at least really trivially entail that they're alike with respect to knowledge. I'll here use 'epistemic position' technically to mean the stuff that traditional invariantists affirm, but shifty people deny, comprise a supervenience base for knowledge.)

Asserting Kp v p

Keith DeRose accepts something like the knowledge norm of assertion -- although as a contextualist, he can't have it entirely straightforwardly. He at least thinks this much: the assertability conditions for S for 'p' are the same as the truth conditions for 'I know p' in S's mouth. He takes it to be obvious that these are different than the assertability conditions for 'I know p'.

Now I'm one of those weirdos who is actually a bit sympathetic to the KK principle. So I'm interested in his argument. I find it pretty uncompelling. DeRose writes:

Both equations of standards -- (1) those for properly asserting that p with those for properly asserting that one knows that p, and (2) those for properly asserting that one knows that p with those for actually knowing that p -- are mistaken, as I trust the considerations below will show to anyone who has deliberated over close calls about whether one is positioned well enough to claim to know that p or should cool one's heels and only assert that p. (The Case for Contextualism, 103.)

I won't continue with his argument, because I'm already not on board. I'm pretty sure I've never deliberated over a close call about whether I was in a good enough situation to assert that I know p, or whether I should cool my heels and only assert that p. Indeed, that strikes me as a totally bizarre thing to do. DeRose himself says that to assert that p is, in some sense, to represent oneself as knowing p.

I know that there are strong theoretical reasons for denying KK, and for accepting the knowledge norm of assertion, and I see that those two verdicts together predict that there will be cases like these in which one can assert p but not that one knows p. But to take such cases as a clear starting point strikes me as bizarre; this, to me, is a cost that I'll accept if I'm forced to. But it's by no means obvious that this ever happens. "p but I don't know whether I know p" is not good.

Am I off base here? Do you ever consider whether Kp is warrantedly assertable, or whether to just stick with the safer p? I don't. (I know I don't.)

What is infallibility supposed to be?

This week I'm thinking about Laurence Bonjour's In Defense of Pure Reason. In §4.4, Bonjour offers what he takes to be a very straightforward argument against the infallibility of rational insight: just look, he says, at all the examples of alleged cases of rational insight that are false — some have been empirically refuted, and some been shown a priori to be incoherent, and some are just inconsistent with others in a way that guarantees that at least some are false.

He qualifies the charge of fallibility, recognizing that it's open to deny that such cases of seeming rational insight into something that ends up being fall are genuine rational insights at all; this, he says, is a "mere terminological or conceptual stipulation" and "fails to secure infallibility in any epistemologically useful sense".

I don't see what infallibility was ever going to amount to if it was to be something stronger than what Bonjour thinks is an uninteresting sense. Did or could anybody ever have thought that anything that seemed to be a rational insight thereby guaranteed its truth? Descartes, for example, recognized the possibility that human reason might be deceived by a God or demon, or imperfectly designed, such that it led into error.

What is the correct characterization of a strong form of infallibility?

Could there be Reductive Knowledge First?

Timothy Williamson has famously defended these two claims:

(1) Knowledge cannot be analyzed

(2) Knowledge can play lots of important explanatory roles all over the place

These two claims, if true, give us reason to think about the role of knowledge very differently; use it to explain things, instead, of as something we're trying to explain. Call this project -- the one recommended in the previous sentence -- 'knowledge first.' The question I'm wondering about right now is, what is the relationship between (1) and knowledge first? Does the case for knowledge first depend on the case for (1)?

(Cards on the table: I'm a guy who thinks that (1) and (2) are both true and that knowledge first is a good idea. So I'm engaging now in a fairly academic question about what depends on what.)

Surely (1) and (2) are consistent (modulo possible worries about whether it's possible to analyze anything). 'Prime number' can be analyzed if anything can, but this is no obstacle to our using the notion of a prime number to explain various phenomena in the world -- for example, in theorizing about encryption algorithms.

Suppose (2) is true. The case for (2) would presumably consist largely of examples -- cases in which we got good explanatory payoff by invoking knowledge. That's the sort of thing that makes up the latter two thirds of Knowledge and Its Limits. And suppose (1) were unestablished, or even known to be false. Wouldn't (2) all by itself make a strong case for knowledge first?