Good post, IMO, except for one part :
[ QUOTE ]
You are assuming that your wooden chair MUST have come to be a certain way because that's how it is now. But if you backtrack and think that the outcome was only one of many possible outcomes, you find that the occurance of a wooden chair is more likely. Simple Bayes.
[/ QUOTE ]
Actually, if we reverse in time, in order to examine the origins of the chair and we examine all the potential developments out of an original, no-chair situation, the eventuality of the chair coming up will NOT appear necessarily as more likely.
Bayes' got nothing to do with it. (Actually, some people use, even without knowing it, Bayesian logic, albeit erroneously, when they suggest that Man would appear anyway. They confuse chance for necessity. Cue for my monthly recommendation of the Ur-book on the subject,
Jacques Monod's "Chance and Necessity".)
If, by reversing back in time, we discover northing more about the chair-scenario that we do not already know (i.e if we do not "discover" circumstances that make the future appearance of a chair more likely), then we have no reason at all to assign a greater probability to a chair appearing.
As far as I know, the underlying randomness factor in biology has been established quite satisfactorily. (BTW, random does not mean equiprobable.)
--Cyrus