Beaver genes, ‘knowing’ about an external world of rivers, trees and dams, program bodies to exploit it. Like all mammal genes, beaver genes also ‘know’ about the internal world of mammal biochemistries and mammal bodies, and they build cells that transact the first and construct the second. Genes ‘know’ about their environment in the special sense that a key ‘knows’ the lock that it uniquely fits.

Where do genes gain their knowledge? All knowledge of the future must come from the past. Gene pools store knowledge of ancestral environments, and program future bodies to use it. To the extent that the future resembles the past, locks open and bodies survive to pass on the same genes. To the extent that it doesn’t, bodies die, and the genes inside them. In extreme cases, whole species go extinct.

But how is the information read out of the environment and into the genes? This is the indispensable role of natural selection, the stunningly simple yet powerful engine of evolution first discovered by Charles Darwin, although he expressed it differently. Neo-Darwinians speak of the nonrandom survival of genes in gene pools. The gene pool of a species is the set of genes that is available, through sexual shuffling, for making individuals of that species. With the exception of clones such as identical twins, every individual is unique. But genes are things you can count. As generations pass, good genes become more frequent in the gene pool; bad genes disappear. ‘Good’ means good at building bodies that survive to reproduce in the environment of the species: woodland, sea, soil, coral reef etc. Regardless of external environments, good genes are good at cooperating inside cells with other genes that have become frequent in the same gene pool and are therefore, by definition, also good.

As a sculptor shapes a statue by subtraction of marble, so natural selection chisels the gene pool towards perfection as generations go by. It isn’t only subtraction. New variation is added to the gene pool by mutation – random mistakes which occasionally turn out to be superior. The randomness of mutation is partly responsible for the widespread, ludicrous misconception that natural selection itself is a random process.

Nonrandom natural selection, automatically and without awareness or deliberation, funnels information about environments into the DNA of a species. This coded information fosters the illusion that organisms were designed precisely for their environments. Think of the uncanny resemblance of camouflaged insects to the background on which they sit. Think of the vertebrate eye with its high-res trichromat retina, variable focus lens, and light-metered fine-adjustment of the pupil. But think, too, of the strange fact that the vertebrate retina (though not that of the independently evolved octopus) is back to front. Light has to pass through a forest of connecting wires before hitting the photocells: exactly the kind of ‘mistake’ you would expect of an evolved, as opposed to designed, instrument.

Several factors conspire to make the natural illusion of design persuasive, complex and often beautiful. ‘Arms races’ between predators and prey, or parasites and hosts, drive the perfection of evolutionary adaptation to spectacular heights. Perfection is enhanced by large numbers of genes, each of small effect, cooperating with each other in cartels of long standing. The evolution of beauty is abetted by the principle that Darwin called sexual selection. The gorgeous colours of a male bird of paradise certainly don’t help it to survive as an individual. They do help the survival of genes that make them attractive to females.

Above all, the illusion of design depends upon the gradual accumulation of small improvements, escalating to levels of complexity and elegance that could not conceivably be achieved in a single lucky step. We are rightly incredulous of any suggestion that biological complexity could spring suddenly from primordial simplicity in one generation. But it is easy if each step of a gradual progression is derived from its immediate predecessor which it closely resembles. That, in a phrase, is why evolution can so brilliantly explain life, where neither chance nor design can.

Intelligent design works as a short-term proximal explanation of cameras and cars, prize roses and poodles. But it is fatally flawed as an ultimate explanation for anything, because it miserably fails to answer the $64,000 question: Who designed the designer? That is not a frivolous debating point. It looms menacingly and fatally over the case – such as it is – for intelligent design. And, by the way, there is nothing new about ‘Intelligent Design Theory.’ It boasts a slick, adman-crafted name but (aside from an irrelevant shift into cellular biochemistry) it offers no new arguments beyond those that Darwin himself demolished in his magnanimous chapter on ‘Difficulties’.

The central (and virtually only) argument offered in favor of intelligent design is the Argument from Improbability. Some biological feature – an eye or feather, biochemical pathway or bacterial flagellum – is claimed to be too statistically improbable (irreducibly complex, information rich etc.) to have evolved by natural selection (naive old-style creationists say ‘chance’). Therefore, by default, it must have been ‘designed’. Positive evidence for design is never even considered: only alleged failures of the alternative.

It is hard to imagine a more lamentably weak argument. The complex biological feature, in every case that has been examined in detail, always turns out to have a gradual ascent path leading to it. In any case, no attempt is ever made to show that the so-called alternative ‘theory’ of intelligent design fares any better. Ultimately, however statistically improbable, however irreducibly complex an eye or flagellum or anything else might one day prove to be, any intelligent being capable of designing it would have to be even more statistically improbable and complex.

Disingenuously, intelligent design advocates try to disguise their religious motives by claiming that the designer’s identity is left open. Not necessarily Yahweh, it could be an alien from space. Scientists would not object to that in principle, because the stellar alien, who might indeed be god-like from our humble viewpoint, presumably evolved by a gradual, cumulative process. You can roll the regress back if you wish, to a designer of the designer. But sooner or later you are going to have to foreswear what the philosopher Daniel Dennett calls ‘skyhooks’, and employ a solidly founded ‘crane’. The only natural crane we know is natural selection, and I have no doubt that if life exists elsewhere in the universe it will turn out to be, in the broad sense, Darwinian.

To the extent that creationists rely on the Argument from Improbability, they cannot get away with postulating an unevolved designer – who would have to be even more improbable. To the extent that they allow their unevolved supernatural designer to have sprung into existence ab initio, they should allow natural agents the same dubious privilege. Intelligent design is not only bad science; it is bad logic, bad philosophy and even – as my theologian friends point out – bad theology.

The United States is, by any standards, the leading scientific nation in the history of the world. Yet this unprecedented powerhouse of scientific achievement is being dragged down in derision, in the eyes of the entire educated world, by the preposterous antics now occurring in a Pennsylvania court, and threatening other boondocks of local democracy. A second rate mathematician, a mediocre biochemist, a born-again retired lawyer, and a Moonie have somehow succeeded in elevating themselves, in the eyes of influential but ignorant politicians, rich benefactors, and duped laymen, to near parity with the entire National Academy. How has it been allowed to happen? When will this great country come to its senses and rejoin the civilized world?