Cheaper than printing it out: buy the paperback book.

Out of Control

"It is totally wrong. It's wrong like infectious medicine was wrong before Pasteur. It's wrong like phrenology is wrong. Every major tenet of it is wrong," said the outspoken biologist Lynn Margulis about her latest target: the dogma of Darwinian evolution.

Margulis has been right about what is wrong before. She shook up the world of microbiology in 1965 with her outrageous thesis of the symbiotic origin of nucleated cells. To the disbelief of traditionalists, she claimed that free-roaming bacteria cooperated to form cells. Then in 1974, Margulis again rattled the cage of biology by suggesting (jointly with James Lovelock) that atmospheric, geological, and biological processes on Earth are so interconnected that they act as a single living, self-regulating system -- Gaia. Margulis was now denouncing the modern framework of the century-old theory of Darwinism, which holds that new species build up from an unbroken line of gradual, independent, random variations.

Margulis is not alone in challenging the stronghold of Darwinian theory, but few have been so blunt. Disagreeing with Darwin resembles creationism to the uninformed; therefore the stigma that any taint of creationism can bring to a scientific reputation, coupled with the intimidating genius of Darwin, have kept all but the boldest iconoclasts from doubting Darwinian theory in public.

What excites Margulis is the remarkable incompleteness of general Darwinian theory. Darwinism is wrong by what it omits and by what it incorrectly emphasizes.

A number of microbiologists, geneticists, theoretical biologists, mathematicians, and computer scientists are saying there is more to life than Darwinism. They do not reject Darwin's contribution; they simply want to move beyond it. I call them the "postdarwinians." Neither Lynn Margulis nor any other postdarwinian denies the true ubiquity of natural selection in evolution. Their disagreement is with the very sweeping nature of the Darwinian argument, the fact that in the end it doesn't explain much, and the emerging evidence that Darwinism alone may not be sufficient to explain all we see. The vital questions the postdarwinians raise are: What are the limits to natural selection? What can't evolution make? And if blind natural selection has limits, what else is operating within or beyond evolution as we understand it?

According to the ordinary contemporary Darwinian biologist, there is nothing we see in nature that cannot be explained by the elemental process of natural selection. In academic jargon this stance is called selectionism, and the position is nearly universal among biologists working today. Because this stance is more extreme that what Darwin himself believed, it is sometimes called neodarwinism.

In the pursuit of artificial evolution, the limits (if any) to natural selection, or to evolution in general, take on practical importance. We'd like an artificial evolution that generates neverending diversity, but so far, that isn't so easy to do. We'd like to extend the dynamics of natural selection to very large systems with many levels of scale, but we don't know how far natural selection can be extended. We'd like an artificial evolution that we could control a bit more than we control organic evolution. Is that possible?

Questions like these have prompted the postdarwinians to reconsider alternative theories of evolution -- many that existed before Darwin -- that were eclipsed by the dominance of Darwinism. In a kind of intellectual survival of the fittest, contemporary biology places very little importance on these "inferior" beaten theories, so they survive only in marginal out-of-print books. But the ideas of these creative theories are suited to a new niche called artificial evolution and are cautiously being resurrected for examination.

The most stellar naturalists, geologists, and biologists of Darwin's time hesitated (despite Darwin's constant badgering) to accept his general theory in full when it was published in 1859. They accepted his transmutation theory -- "descent with modification," or the gradual transmutation of new species from preexisting species. But they remained skeptical of his selectionist reasoning -- that tiny random improvements were all there was to it -- because they felt Darwin's explanation did not accurately fit the facts of nature, facts with which they were intimately familiar in a way that is rare today in this era of specialization and indoor laboratories. But since they could offer neither compelling disproof nor an alternative theory of equal quality, their forceful criticisms were buried in correspondence and scholarly disputes.

Darwin didn't offer a concrete mechanism by which his proposed natural selection would take place, either. He was ignorant about genes, for starters. The first fifty years following the publication of Darwin's tour de force were ripe with supplemental theories of evolution, until Darwin's dominance was clinched by the discovery of genes and later DNA. Almost every radical evolutionary conviction circulating today has as its source some thinker in the years after Darwin but before acceptance of his theory as dogma.

No one was more sensitive to the weaknesses of Darwinian theory than Darwin himself. As an example of trouble, Darwin volunteered the astounding multifaceted sophistication of the human eye. (Every critic of Darwin since has also used his example.) The exquisite design of interacting lens, iris, retina, etc., seems to defy the plausibility of Darwin's "slight, incremental" chance improvements. As Darwin wrote to his American friend Asa Gray, "About the weak points I agree. The eye to this day gives me a cold shudder." The difficulty Gray had was imagining how any portion of an unfinished eye, a retina without lens or vice versa, would be useful to its possessor. Since nature cannot hoard innovations ("Hey, this will come in handy in the Cretaceous!"), every stage in development must be immediately useful and viable. Breakthroughs have to work the first time. Even clever humans can't design in such a consistently demanding manner. Therefore nature appears superhuman in its ability to create.

Imagine, says Darwin, that we extrapolate the tiny microevolutionary changes we see in domesticated breeding -- a pea with extra -- large pods made larger, or a short horse bred shorter. Imagine if we extend those slight changes caused by selection over millions of years; we add up all the minute differences until we see major change. This is what makes coral reefs and armadillos out of bacteria, Darwin said -- accumulated microchange. Darwin asks that we extend the logic of microchange to cover the grand scale of Earth and Time.

The argument that natural selection can be extended to explain everything in life is a logical argument. But human imagination and human experience know that what is logical is not always what is so. To be logical is a necessary but insufficient reason to be true. Every swirl on a butterfly wing, every curve of leaf, every species of fish is explained by adaptive selection in neodarwinism. There seems to be absolutely nothing that cannot be explained in some way as an adaptive advantage. But, as Richard Lewontin, a renowned neodarwinist, says, "Natural selection explains nothing, because it explains everything."

Biologists cannot (or at least they have not) ruled out the role of other forces at work in nature producing similar effects in evolution. Therefore, until evolution is duplicated under controlled conditions, in the wild, or in a lab, neodarwinism remains a nice "just-so" story -- more like history than science. Philosopher of science Karl Popper said bluntly that neodarwinism is not a scientific theory at all, since it cannot be falsified. "Neither Darwin, nor any Darwinian, has so far given an actual causal explanation of the adaptive evolution of any single organism or any single organ. All that has been shown -- and this is very much [sic] -- is that such an explanation might exist -- that is to say, [these theories] are not logically impossible."

Life has a causality problem. Any coevolved organism seems to be self-created, making causality onerous to pin down. Part of the search for more complete explanations of evolution is a search for a more complete logical understanding of spontaneous complexity and the rules by which entities may emerge from a web of parts. The quest for artificial evolution -- so far done primarily in computer simulations -- is very much tied into a new way of establishing proof in science. Previous to the advent of ubiquitous computers, science consisted of two facets: theory and experiment. A theory would shape an experiment, and then the experiment would confirm or disprove the theory.

But computers have birthed a third way of doing science: by simulation. A simulation is at once both a theory and an experiment. By running a computer model, such as Tom Ray's artificial evolution, we are trying out a theory and also running something real and accumulating falsifiable data. It may be that the dilemma of ascertaining causality in complex systems will be bypassed by these new methods of understanding, wherein one studies the real by modeling working surrogates.

Artificial evolution is at once a theory and test for natural evolution, and something original in itself.