Reading Eleonore’s massive work on the problem of evil, and in the middle of her defense of the claim that there is something special about knowledge of persons. She calls it “Franciscan knowledge”. She begins her characterization by making remarks about analytic epistemology, contrasting knowledge-that, what she calls Dominican knowledge with Franciscan.
She begins by claiming that “it is axiomatic in analytic philosophy that all (or virtually all) knowledge is knowledge that something or other is the case.” (p. 48) A startling remark, one that I don’t recognize as true. The parenthetical phrase is footnoted, so I thought, “maybe the footnote will make sense of this,” but the footnote says that the qualification is needed because of Russell’s distinction between knowledge by acquaintance vs. knowledge by description. But that doesn’t help: it’s simply false that epistemology of the last 100 years assumes that knowledge is either by acquaintance or de dicto.
Explanation below the fold.
What is true is that the kind of knowledge epistemologists tend to focus on is propositional knowledge. But it is not true that epistemology assumes that all knowledge other than knowledge by acquaintance is de dicto.
A better taxonomy here would begin by distinguishing practical from theoretical knowledge. For those linguistically-minded folk, we can distinguish, as the literature does, between knowing-that and knowing-how (which Eleonore notes as well), and between each of these and knowing-wh (knowing who what where when). Perhaps knowledge-wh can be analyzed in terms of propositional knowledge, but even given this reduction, there’s lots more in conceptual space than knowing how and knowing that.
Why? Because it is clearly a major part of epistemology of the last 100 years to distinguish between knowledge de dicto and other types of knowledge, such as knowledge de re and knowledge de se. While some have held that each of the latter two can be explained in terms of de dicto knowledge, that position looks mistaken to me and is not widely (or axiomatically) assumed to be true. In fact, I’d bet the majority of epistemologists think as I do, that there is no reduction possible here.
If we distinguish knowledge de dicto from knowledge de re and knowledge de se, we might try to fit Russell’s knowledge by acquaintance into the de re category. And once we get this far, we can see why a defense of Franciscan knowledge can’t be simply a matter of claiming that there is knowledge that isn’t propositional knowledge. Of course there is: there is objectual knowledge and self-knowledge, at the very least.
Categorized in this way, we can see what is required for a defense of Franciscan knowledge. We can think of de dicto knowledge as different from a linguistically-minded set of distinctions between first-, second-, and third-person awareness. De se knowledge is first-person knowledge, and de re knowledge is third-person knowledge. What is left is second-person knowledge, which is just the category it looks like Eleonore is targeting.
If so, however, it isn’t enough to rely on the variant of the Mary case that she uses to defend the possibility of Franciscan knowledge. (She changes the Mary case so that Mary has never had personal contact before, but only knows all the facts about persons; on leaving the enclosure, she meets her mother for the first time, and obtains knowledge she couldn’t have had by just knowing all the facts.) Even if persuaded by this variant of the Mary case that what Mary comes to know isn’t propositional in character, one also must show that it can’t be accounted for in terms of de se or de re knowledge (and one shouldn’t try to limit the de re category to what Russell describes as knowledge by acquaintance–that’s too restrictive).
Ruling out these other categories is much more difficult. It requires a defense of the claim that there is, in addition to de dicto, de se, and de re knowledge, something we might refer to as “de te knowledge”.