WAS DARWIN WRONG?

The two great engines of evolution in the Darwinian conception are random mutations and natural selection. Drexler gives the example of crystals growing: by thermal motion atoms will randomly bump into each other, but stick selectively by stronger or weaker chemical bonds. An atom that sticks by a weak chemical bond gets bumped off by one that will stick for good; thus the crystal replicates itself and evolves selectively. This may also have been the mechanism that led to the formation of DNA chains in the primordial soup and to life itself. (I have here mixed in an example given by Richard Dawkin's in The Selfish Gene, Oxford U. Press 1976, which Drexler often quotes. I had read Dawkins' book some years ago; it makes a good introduction to Drexler's).

In the case of a crystal, I have no quarrel, for it is hard to argue with the facts. What I cannot swallow (as Dawkins and Drexler seem to do) is that the same mechanism applies to the evolution of more complicated systems such as, say, the human eye. Now before I go any further I must dash the hopes of some of my readers that I am a Creationist: I am not. The Creationist Theory, in my opinion, papers over our ignorance with words, and is therefore a theory of faith, not of science.

But the random-mutation part of Darwinism defies all results of probability theory. "If an infinite number of monkeys start playing with an infinite number of typewriters, one of them will type a play by Shakespeare," says Neville Chute in On the Beach (a novel of the 1950s, not unlike the Nuclear Winter fiction, that may have played its part in ushering in the age of apocalyptic thinking). But in the universe we live in, the statement is sheer nonsense. It can be shown by fairly elementary probability theory that to attain a 95% probability of this happening, the number of atoms in the solar system, and very probably in the entire universe, is insufficient to furnish the required number of typewriting monkeys.

And yet a play by Shakespeare (mathematically speaking, a finite string of letters taken from the English alphabet) is incomparably simpler than the human eye. And that is supposed to have come about by random mutations?

In Artificial Intelligence, there are programs that write programs. Do they try out a zillion duds before they luckily hit on a good one? No: they go for the best they can do straightaway, and then they debug it.

Surely in the evolution of complicated systems (as Drexler, but not Dawkins, seems to admit in places) there must be a strong element of purposeful self-organization.