{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

notes_11

# notes_11 - Notes on KAIL/15/06(B.F This is a very long...

This preview shows pages 1–2. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Notes on KAIL Chapter 10 11/15/06 (B.F.) This is a very long, technical, and difficult chapter. There is no hope of going through it all in one seminar. Instead, my comments (and Mike’s comments) will focus on a few issues that are central to TW’s discussion in the chapter. Let’s begin by recalling TW’s explication of “ E is evidence for H for S ”, which we have already discussed in our last meeting: (EV) E is evidence for H for S iff E is included in in S ’s body of evidence and Pr (H | E) > Pr (H) . Last week, we discussed the first component of (EV) — the “inclusion in S ’s body of evidence” component. This week, we will focus on the second component — the “support as positive probabilistic dependence” component. There are many important questions that must be answered in order to get a handle on the intended sense of this second component. Here are five: 1. What kind of probability function is TW’s “evidential” probability function Pr? • Pr (p) is clearly not S ’s actual degree of belief (credence) in p [Pr ≠ C a ]. For one thing, S ’s actual degrees of belief C a can be crazy (even incoherent ), and we want Pr to ground non-crazy claims about evidential support [although, it’s not entirely clear why we want this from a traditional epistemological point of view, which cares about full belief and knowledge — see question (3), below, for further discussion of that issue]. Moreover, as we mentioned last week, if S ’s credence in either H or E happens to be extreme (zero or 1), then E and H cannot be positively dependent on C a . And, surely, S can actually have extreme credences in either E or H , for any E / H . As such, identifying Pr and C a would basically trivialize TW’s “support” relation. Finally, TW wants his “support” relation to be objective , and identifying Pr and C a would make it patently subjective in a way TW doesn’t want. • Pr (p) is not even to be identified with the ideally rational or justified degree of belief (for S ) in p [Pr ≠ C j ]. TW says that this “comes closer” to what he has in mind with his “evidential” probability. For one thing, at least C j is objective in ways that C a is not. For instance, even if S happens to have extreme degree of belief in E or H , it might be that this is irrational or unjustified (given everything S knows). Moreover, C a might even be incoherent , whereas (presumably) C j cannot be. This gives C j a kind of objectivity (and also probative force) that seems to be required to undergird TW’s objective “support” relation. However, TW wants to think of “support” as an objective relation between E and H [and some “background corpus” K — see question (4) below]. But, if we identify Pr with C j , then we can still get undesirable verdicts about “support”, because an agent (albeit an ideally rational one) is “getting between” E and H . For instance, there may be cases in which C j (E) = 1, but we still want to say that...
View Full Document

{[ snackBarMessage ]}

### Page1 / 4

notes_11 - Notes on KAIL/15/06(B.F This is a very long...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online