ID and Probability
It seems that the probability arguments related to creationism don’t change much at the core, they just get more complex and verbose. The old “747 from a hurricane in a junk yard” argument just gets reformatted and reused about different things.
Ed Brayton has a response to DaveScot that is so good I need to link to it. It’s very helpful for all the non-statisticians/non-mathematicians amongst us.
DaveScot uses a deck of cards as his analogy, and the key thing to remember is that one can calculate the probability of a particular outcome because we know the precise process by which the cards get shuffled. All things being equal, each card has an equal probability of appearing in any particular position. Proteins are not like cards in that sense.
I’ve heard a few times that one has to be a mathematician to respond to Dembski on things like specified complexity and the “no free lunch” theorem, but I think the GIGO principle is more applicable. If you don’t know what the processes you’re analyzing actually are, it is silly to talk about how probable it is that they will occur. The problem here is logic. No matter how good your algorithm, it will not produce good information with bad input, and the input used by the ID proponents is bad.
Ed does a better job of analyzing it, though. Go read his.
I haven’t read Dembski’s stuff in a while, but if I recall correctly his argument was that, although evolution may produce complex functionality, it probably won’t be the complex functionality you were after (e.g. that matched your specification).
The obvious response is that this works fine… as long as you don’t actually pick your specification to correspond to the functionality.
I’d second the bit about needing to be a mathematician (I speak as someone who majored in maths). Dembski’s writing is heavy on the mathmobabble, but the actual mathematical content of it is very low. As far as I can tell, it’s only included because translating backwards and forwards between words and symbols introduces wonderful new opportunities for confusing the reader.