16 July 2007
Michael J Behe
Here I respond briefly to Richard Dawkins’ review of The Edge of Evolution in theNew York Times. I must admit I was surprised that he agreed to do it. In the past Dawkins has said that on principle he would not interact with proponents of intelligent design, because that would give us publicity. I guess when the New York Times offers writing space, principles can be reconsidered.
Other Darwinist reviewers have blustered; Dawkins is the only one who has dripped venom. I will pass on replying to that. He makes just two substantive points in his review. The first is that the success of artificial selection in things like dog breeding show the malleability of organisms, so why should Darwinian evolution be a problem? I already answered that point in my reply to Jerry Coyne. Briefly, it begs the question of what changes are occurring at the molecular level in those examples, whether simple ones or complex ones, and it begs the question of where the sophisticated molecular systems came from that we have learned control animal form and development. Dawkins seems quite reluctant to engage my argument at the molecular level; in his review he defers to other scientists for that. He himself gives the kind of argument that a 19th century naturalist might give, before the elegance of the molecular foundation of life was discovered by modern biology.
Dawkins’ second substantive point is that if I am right, then:
Behe’s calculations would at a stroke confound generations of mathematical geneticists, who have repeatedly shown that evolutionary rates are not limited by mutation. Single-handedly, Behe is taking on Ronald Fisher, Sewall Wright, J. B. S. Haldane, Theodosius Dobzhansky, Richard Lewontin, John Maynard Smith and hundreds of their talented co-workers and intellectual descendants.
It’s a flattering thought, but incorrect. If I am right it would overturn virtually no theoretical work, simply because theoreticians have not dealt with the sorts of complex, functional systems I write about. For the most part, models have considered one or two simple mutations at a time, conceptually isolated from the real-life complexity of an actual cell or organism. That’s necessary, because detailed models of complex systems would be intractable. Those (relatively) simple models can of course be very important and useful for things like predicting the spread of the sickle hemoglobin gene, or calculating from the number of neutral mutations the time since two species shared a common ancestor. But there is no theoretical evolutionary work on the production of molecular machinery.
(One of the luminaries Dawkins lists, John Maynard Smith, once briefly alluded to the kind of problem The Edge of Evolution deals with. In a letter to Nature in the early 1970s, Smith compared evolution of proteins to a word game where only one letter is allowed to be changed at a time, and misspellings are disallowed too. I cite Smith’s paper in the book.)
At the end of his review Dawkins chides me for lack of peer-reviewed publications. Talk about the pot calling the kettle black. If Dawkins himself has many peer-reviewed research publications in the last few decades, he must be writing them under a pseudonym. Dawkins’ hypocritical complaint makes a nice little example of Darwinian gate-keeping. The nebulous, wooly-minded scenarios Dawkins spins in his books, of the origins of bat echolocation, spider webs, and so on, have no real justification in peer-reviewed publications. Yet Dawkins is free to write trade books without howls of protest from the scientific community because his stories fit the way many scientists want the world to be. But if (ahem…) someone publishes a book critically analyzing the data from a different perspective, the reaction is dramatically different.