Sahorta Sarkar’s new book is a response to the latest strategy by creationists to combat the teaching of evolution in schools. The older strategy, of having openly religiously-inspired creationism granted equal time with evolution in science classes, failed dismally, because American courts ruled that such teaching violated the constitutional requirement for a separation of church and state. The response of the Reformed Creationists, or ID creationists, as Sarkar tendentiously but accurately calls them, is to deny that their views are religiously inspired at all. Instead, they present them as scientific hypotheses, supported by the best available evidence. The strategy was designed for the American context, but such is the prestige of science that it has proven of value elsewhere as well.

Reformed Creationists are advocates of ‘intelligent design’; the doctrine that there are features of the universe — usually of biological organisms — that show clear signs of design and that therefore could not have emerged ‘by chance’ (as they often say). This design argument is an old one; ID is novel only insofar as it attempts to impart biological and mathematical respectability to the argument. The mathematics comes largely from William Dembski, who provides an explanatory filter we can allegedly use to detect design. Roughly, Dembski argues that if a feature is not the product of regularity or chance, it is the product of design. Sarkar has no difficulty in showing that the filter is useless. For one thing, it generates plenty of false positives: events that are clearly not designed, but which pass through the filter. The problem with the filter, essentially, is that Dembski treats chance and regularity independently. But this is a bit of logical trickery, designed (!) to obscure the way in which natural selection actually operates. Adaptations do not emerge ‘by chance’, but through selection operating on chance.

More recently, Dembski has claimed to find evidence for design in ‘No Free Lunch’ theorems, which show, essentially, that no search algorithm outperforms any other over the whole set. This, Dembski points out, implies that random search cannot outperform a hill-climbing alogorithm (that is, an algorithm that preserves incremental increases in adaptiveness) over the set of all optimization problems. The problem is, as Sarkar points out, evolution by natural selection does not require an algorithm that outperforms all others. A successful algorithm will be one that is successful, given the problems it actually confronts; that is, given the actual environment in which evolution occurs. That it would perform relatively poorly in some alternative environment is simply irrelevant to whether it performs well in this one.

Dembski’s mathematics is an exercise in smoke and mirrors, meant to impress the mathematically unsophisticated and the credulous, and give ID the veneer of science. Of somewhat (but only somewhat) more value is Michael Behe’s attempt to give ID biological credibility. Behe attempted to show that certain biological features are ‘irreducibly complex’, and therefore could not have emerged by natural selection. A feature of an organism is irreducibly complex if performs some function through the coordination of several parts, and it would be useless at performing that function if it lost any of these parts. Behe points out that if a feature is irreducibly complex, it could not have become better adapted to performing that function by adding any of these parts: in the absence of any, it would have been useless. Behe’s challenge was a serious one, and helped spur serious biological work. As he must have known, though, the problems he regarded as intractable resembled others that had already been solved; since the publication of his book in which he announced the death of evolution, his own pet examples have followed suit. Some of these examples, such as the blood clotting cascade, have been shown not to be irreducibly complex at all: more primitive versions not only could have performed the same function (though less well), but actually do perform this function in living organisms. Others have been shown to have developed through exaptation, in which structures evolved for one purpose become co-opted to another. Behe’s initial challenge might have been serious, but his failure to concede that it has been adequately met reveals that it was never scientifically motivated.

Sarkar’s book is unusual in that it does not limit itself to the terrain of evolutionary biology. It also covers arguments from physics, to the effect that the universe is fine-tuned for life. The argument, roughly, is that given the narrowness of the band of fundamental physical laws that must hold for life to emerge, the probability that the universe would be hospitable to life is low, and this is evidence for a benevolent helping hand; the universe was designed to allow for our existence. The problem, as Sarkar shows, is that we have absolutely no way of beginning to estimate the probability that the physical laws should be as they are. The book ends with a defense of metaphysical naturalism — the doctrine that there are no non-physical entities — though as Sarkar notes, only methodological naturalism — the doctrine that in investigating the world we need postulate no non-physical entities — is needed for the practice of science, and some reputable scientists are not metaphysical naturalists.

Though I agree with almost everything in this book, I do wonder at its value. There is a perennially debated question among opponents of creationism: should we debate them at all? Debating them seems to give them a legitimacy they do not deserve. Now, though Sarkar is not giving them a platform, in quite the same way as he would were he to appear on the same stage as them, a book-length refutation is, in one way, far too much effort to expend on them and takes them with a seriousness they do not deserve. ID is an important political and sociological phenomenon, but its arguments are not substantive enough to warrant this refutation. Moreover, the would-be rebutter is caught in a double-bind. This is a short book, and Sarkar can therefore only has room to rebut creationist claims  briefly, and to examine only representative creationist arguments. He therefore leaves himself open to the claim that he has missed important arguments or failed to appreciate the significance of some obscure point. Given that his opponents do not argue in good faith, this is not a game that can be won. Despite the genuine achievement of this book, I wonder if it was an enterprise worth attempting at all.

© 2007 Neil Levy

Neil Levy, Ph.D., Senior Research Fellow Program Manager, Ethical Issues in Biotechnology, Centre for Applied Philosophy and Public Ethics, University of Melbourne

Comment on this review