Joined: Dec. 2002
I'd like to quote Sober's concluding paragraph in it entirety:
My critique of the intelligent design movement has been based on the comparative principle I stated about evidence -- to say whether an observation counts as evidence against evolutionary theory and in favor of the hypothesis of intelligent design, one must know what each predicts about the observation. I have challenged intelligent design theorists to produce a theory that has implications about the detailed examples of "irreducible complexity" that Behe describes. However, there is another response that intelligent design theorists might contemplate. This is to deny the comparative principle itself. Dembski has seized the horn of this dilemma. If he succeeds in developing an epistemology of this sort (so far he has not), the way will be paved for an unprecedented result in the history of science -- the rejection of a logically consistent theory that confers probabilities on observations, but does not entail them, and its replacement by another, without its needing to be said what the replacing theory predicts.
Apropos of Sober's conclusion, I'd even more highly recommend Royall's Statistical Evidence -- a book that Dembski cites but shows little eveidence of having understood.
Paticularly revealing in this regard is Dembski's response to an earlier critique by Sober, Fitelson et al., in which he justifies his strict reliance on non-comparative (i.e., Fisherian) modes for the evaluation of evidence by pointing out the prevelance of this approach in applied statistics.
This is about as weak an argument as could possibly be made: "scientists prefer Fisher, therefore my argument is correct." Significance testing has dominated the statistical landscape primarily because it has been computationally easy relative to other approaches. Furthermore, the statistical landscape is changing: in many fields (medicine, ecology, computational biology) explicitly Bayesian approaches are being developed and routinely applied.