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
Unformatted text preview: Notes for Week 9 of Confirmation 10/31/07 Branden Fitelson 1 The CarnapPopper Controversy Almost all of the first-edition of Carnaps LFP is devoted to confirmation as firmness [which he explicates as Pr > (H | E) > r , for a suitable logical conditional probability function Pr > ( | ) see below for much more on the historical varieties of Carnapian logical probability functions]. But, in his discussion of Hempel, Carnap gives what he claims to be a counterexample to (SCC), which, as youll recall from Week 7, is: (SCC) If E confirms H relative to K and H K H , then E confirms H relative to K . This is rather puzzling, since (SCC) holds for confirmation as firmness (well return to this point when we discuss the conjunction fallacy in the final week of the seminar). Whats happening here is that Carnap presupposes a different probabilistic relevance conception of confirmation in his discussion of Hempel, which he (there) calls initial confrimation. This is very much like Mahers confirmation relation (from last week). It is something like probability-raising, relative to empty background corpus. Popper wrote a (nasty) series of criticisms of Carnap, accusing him of inconsistency (or worse) on this score. While Poppers discussions were not always the most charitable (or helpful), he did make a compelling case that the probabilistic relevance conception is often the more important of the two. One example (that is similar to an example Popper discusses) involves a person who is wondering whether ( H ) they have a very rare disease. They just received a positive result ( E ) for the disease from highly reliable test. In such cases, it seems wrong to say that E justifies or provides a reason to believe or supports H (for ). However, it is still reasonable all things considered (where this includes E ) for to believe that they do not have the disease ( H ). But, the right thing to say here is that it is despite having learned E that it is (still) reasonable for to believe H . Intuitively, E counter-supports H (for ) because E supports H (for ). Thus, in this case, the firmness conception says that E confirms H , but the increase in firmness conception says that E disconfirms H . And, Popper argues that in such cases the increase in firmness conception is the more salient (or important). I think Popper is basically right here. And, Carnap did not really disagree. Largely in response to Popper, Carnap issued a second edition of LFP . But, the second edition does not make any changes to the main text. It just adds a brief preface, explaining how he should have been clearer about the two distinct conceptions of confirmation that were floating around in the first edition....
View Full Document
This note was uploaded on 08/01/2008 for the course PHIL 290 taught by Professor Fitelson during the Fall '06 term at University of California, Berkeley.
- Fall '06