Cheaper than printing it out: buy the paperback book.

Out of Control

A famous image from Darwin's Origin of Species, written over a century before the dawn of the first computer, precisely embodies the task of evolution in computerese. Evolution, Darwin said, "is daily and hourly scrutinizing, throughout the world, every variation, even the slightest; rejecting that which is bad, preserving and adding up all that is good; silently and insensibly working...." This is the algorithmic search through the Library of forms. Is the Library of possible biological life forms a vast space with only a few sparse coherent works, or is it filled with many of them? How likely is it that a random evolutionary step will land on a possibility with real life? How closely bunched are functioning organisms in the space of possibilities? How isolated are viable lineages from each other?

If the density of possible life forms is sufficiently crowded with feasible beings, then the space of possibilities can be more easily searched by the chance-driven walk of natural selection. A space thick with prospects and searchable by randomness provides uncountable paths for evolution to follow through time. On the other hand, if functioning life forms are sparse and isolated from each other, natural selection alone will probably be unable to reach new forms of life. The distribution of functional units in life may be so scant that most of the space of possible organisms lies empty of workable cases. In this vast space of failure, viable life forms may be found lumped together in patches, or conglomerated onto a few crooked paths through the space.

If the space of functioning organisms is at all sparse, then it is clear that in order to proceed from one patch of viable creatures to the next, evolution needs something to guide it through empty wastelands. A trial-and-error walk, such as that which underlies natural selection, can only get you nowhere fast.

We know virtually nothing of the real distribution of life in the Library of realities. It may be so sparse and unpregnant with possibilities that there is only one living path through it -- the path we are currently on. Or there might be broad highways in the Library that channel a number of paths into a few bottlenecks that all beings must cross -- say, the resonant attractor of four legs, a tubular gut, five-digit hands. Or there may be a submerged bias in life's substrate, so that no matter where you start you eventually arrive on the shores of bilateral symmetry, segmented limbs, and intelligence of one kind or another. We just don't know. But with artificial evolution at work, we could know.

These fruitful questions about the constitutional laws of evolution are being asked, not in biological terms, but in the language of a new science, the science of complexity. Biologists find it most grating that the impetus for this postdarwinian convergence comes chiefly from mathematicians, physicists, computer scientists, and whole systems theorists-people who couldn't tell the difference between Cantharellus cibarius and Amanita muscaria (one of them a deadly mushroom) if their lives depended on it. Naturalists have had nothing but scorn for those so willing to simplify nature's complexity into computer models, and to disregard the conclusions of that most awesome observer of nature, Charles Darwin.

Of Darwin's insights, Darwin himself reminded readers in his update to the third edition of Origin of Species:

As my conclusions have lately been much misrepresented, and it has been stated that I attribute the modification of species exclusively to natural selection, I may be permitted to remark that in the first edition of this work, and subsequently, I place in a most conspicuous position -- namely at the close of the Introduction -- the following words: "I am convinced that natural selection has been the main, but not the exclusive means of modification." This has been of no avail. Great is the power of steady misrepresentation.

Neodarwinism presented a wonderful story of evolution through natural selection, a just-so story whose logic was impossible to argue with: since natural selection could logically create all things, all things were created via natural selection. As long as the argument was over the history of our one life on Earth, one had to settle for this broad interpretation unless inarguable evidence would come along to prove otherwise.

It has not yet come. The clues I present here of symbiosis, directed mutation, saltationism, and self-organization, are far from conclusive. But they are of a pattern: that evolution has multiple components in addition to natural selection. And furthermore, these bits and questions are being stirred up by a bold and daring vision: to synthesize evolution outside of biology.

The moment we tried to transfer the dynamics of evolution out of history and into a manufactured medium, the inner nature of evolution was exposed to scrutiny. Evolution pressed into artificial evolution within computers has passed the first neodarwinist test. It demonstrates spontaneous self-selection as a means of adaptation, and as a means of generating some initial novelty.