Friday, March 27, 2009

Darwin Bicentenary 16: Waving Flagella and Waving Hands

I’m rather getting ahead of myself here but I’ve just got to tag this post on Uncommon Descent. Here William Dembski tells us about an email he received from someone who was “thoroughly convinced” of ID theory but then went on to read “The Language of God” by Francis Collins. In the said email Dembski’s correspondent mentions Collins’ challenge to the claimed irreducible complexity of the iconic bacterial flagellum (see picture on left). Dembski’s inquirer also contacted Collins himself who forwarded a review article by Mark Pallen and Nicholas Matzke on the flagellum. Dembski quotes Pallen and Matzke who state “designing an evolutionary model to account for the origin of the ancestral flagellum requires no great conceptual leap.” Dembski responds thus:

“Of course it doesn’t — one can always imagine some way that natural selection might have brought about the system in question. ….Requiring no great conceptual leaps or being unable to find a case where Darwin’s theory could not possibly apply is not the same thing as providing evidence. Sure, the proteins in the flagellum may have homologues that serve functions in other systems. And we can imagine that the parts were co-opted over time by selection to produce the flagellum. But so what? We can imagine lots of things. Where’s the evidence that it happened that way? And why isn’t the exquisite engineering that we observe in the flagellum evidence for ID?”

This little dispute is something I have commented on before. At the time I asked on whom does the burden “proof” lie? Evolutionists demand that ID theorists show that there are no incremental paths through morphospace along which evolutionary shuffling can diffuse. Therefore, say the evolutionists, the case for ID is not proved. Ahh!, gasp the ID theorists, on contrary it is for evolutionists to show that those evolutionary paths through morphospace exist:

Evolutionist: “You ID theorists haven’t provided evidence that the bacterial flagellum is irreducibly complex”
ID theorist: “But you evolutionists haven’t provided evidence that is reducibly complex!”
Evolutionist: “No, it is you who should be providing evidence that the flagellum is irreducibly complexity”
ID theorist: “On the contrary it is you who should first provide evidence that the bacterial flagellum is reducibly complex!”

I suppose a methodological judgment could be made in favour of one party or the other; perhaps in favour of the evolutionist on the basis that the burden of proof is on the ID theorists because they are making an extraordinary claim….. or perhaps in favour of ID theorists on the basis that evolutionists are making an unfalsifiable existential statement. I think I’ll leave them to discuss this matter; last time I attempted to arbitrate in this dispute I got myself into hot water, so I won’t comment this time. In any case methodological decisions are likely to evaluated from the perspective of those with vested interests and preferred decisions are therefore likely to embody the value judgments of partisans.

What I want to draw attention to, however, is that William Dembski is conceding a possibility; namely that it is possible to imagine a flagellum that is not irreducibly complex. Dembski is alluding to the very conjectural structure of morphospace, so let’s forget all about evidence and ask ourselves if evolution is at least conceivable as a possibility if not as a reality? A condition of the latter, if not sufficient condition, is that the set of stable structures in morphospace form at least a partially connected set not unlike the connectedness of the Mandelbrot set. But, needless to say, unlike the Mandelbrot set there is no easy analytical way to handle morphospace or prove that it is connected. If evolution is a possibility then presumably the structure of morphospace is implicit in the laws and statistics of the cosmic physical regime. However, proving the connectedness of the set of stable structures is nigh on impossible. In morphospace we have an object of such complexity that it clearly confounds both evolutionist and ID theorist and leads to their polite hand waving exchange of “After you”… “No after you”.

Morphospace is a highly complex object that is unlikely to reveal its secrets easily; our theories about it are likely to be an intricate amalgam of succinct theoretical insights embedded in a very narrative intense body of knowledge - that and, no doubt, a lot of hand waving. One obvious complexity is that morphologies are only rated as stable with respect to the environment they are in and therefore environment is one of the dimensions of morphospace. Moreover, survivable morphologies are themselves part of the very environment in which they are placed thus leading to an analytically inconvenient feedback effect and non-linearities. Morphospace is undoubtedly one of those theoretically intractable objects that must take its place alongside God himself as an object beyond the human mind to handle succinctly and fully. However, as with theology one can but try.

In the absence of analytical answers to the questions of morphospace we depend on real life history having done the simulation for us. Hence paleontological evidence must have a strong showing in the evolution/ID debate. If sufficient missing link evidence comes to light it could weigh the argument in favour of evolution. But in the absence of evidence that is compelling to all parties methodological policy decisions will have to be made. But that takes us back to square one and the conflicts of the wild web.

Characters of the Wild Web 5: Would be Pat Garretts are gunning for Billy the DembskID.

William Dembski says he is revisiting evolutionary computation.(See here)


article345 said...
This comment has been removed by a blog administrator.
Anonymous said...
This comment has been removed by a blog administrator.