Gone Fishing

I'll be in Alaska all next week, so probably no bloggage.

Here's a picture of where I'll be: (more pictures here)


See you when I get back.

Pragmatic Encroachment in Chisholm?

I'm working on a book project called _Chisholm's "Theory of Knowledge"_ (the mentioning rather than using indicates the fact that it's a history of the evolution of that book over more than two decades). So I'm reading them backwards and forwards (literally, 3rd edition, then 2nd, then 1st, then back again).

In the penultimate chapter of the 2nd Edition, on the concept of knowledge, Chisholm considers the following property of knowledge:

The Surety Thesis: If S knows that p at t, then at t S has the right to be sure that p.

He doesn't consider the option of rejecting the Surety Thesis as I would probably do. Rather, he provides a deflated interpretation avoiding the untoward consequence of knowing that p: that it seems that one thereby is in a position to know that all future evidence against p is misleading evidence and therefore can reasonably be ignored.

"There is another sense of 'right to be sure.' And if we take the expression in this second sense, then we can say...that knowledge confers this right..." (117).

What is the sense in question? In short, it is that we may conditionalize upon what we know. Chisholm:

"[W]e should be guided by the *absolute* probability of the proposition--the probability that the proposition has in relation to what we know.... If we *know* that a certain proposition is true, then we have the right to use that proposition...in calculating probabilities" (118).

That seems dangerously close to me to saying a la Williamson that our evidence is cooextensive with our knowledge (with that little caveat about "independent" evidence) as well as the Usability Thesis.

Then he seems to switch from conditionalizing to being sure that expected utilities obtained from such conditionalizing will yield reasonable decisions. Chisholm again:

"[If we know that a certain proposition is true,] [w]e have a right to be sure that, to the extent that we rely on *that* proposition as a basis for calculating the probabilities of other propositions, our decisions will be reasonable ones" (118).

In expected utility theories it sometimes suffices for a reasonable decision that it follow satisfactory expected utility, so it seems that it would suffice that we merely *justifiedly believe* that a proposition were true to yield a reasonable decision. So knowledge should secure this result by entailing the sufficient condition. In other theories, however, sufficient expected utility is merely a necessary condition for a reasonable decision. Decision theorists of the latter variety might balk at Chisholm's second statement here. (And surely the statement is only true modulo some caveat about having reasonable preferences or reasonable utility assignments.)

My concern, however, is that it seems that Chisholm has pre-endorsed a kind of "pragmatic encroachment" (Jon Kvanvig's mintage) similar to that of Williamson, Hawthorne, and Fantl and McGrath. All those guys endorse something to the effect of

The Usability Thesis: If S knows that p at t, then at t S can use p in practical reasoning.

I think it's pretty clear that the Usability Thesis is false. Even if obvious holes are patched I don't think the remaining principle has much hope. Sometimes I just can't reasonably act on what I know. The other day the bus was late and though I knew it would come fairly soon I couldn't risk it since I had a presentation in seminar. The rational thing to do was to *refrain* from acting on what I knew and go for the sure thing: having my wife drop me off. I don't think this is significantly different from the position I'd be in if I were in an airport and--do to extremely high stakes--need to be *certain* that I was at the right gate. Since knowledge doesn't entail certainty, knowledge is possibly insufficient for actionability here.

It will take more reflection for me to figure out if the thesis endorsed by Chisholm here is as objectionable-to-me as is the sort of wide-ranging pragmatic encroachment going about today. However, I have to say that I was a bit shaken when I read these passages.


Keywords: Epistemology, Chisholm, Kvanvig, Williamson, Hawthorne, McGrath, Fantl, pragmatic encroachment, evidence, decision theory.