In Santa Fe, New Mexico, a group of scientists, mainly from the Los Alamos National Laboratories, held a conference in 1990 on “Artificial Life.” The theme of the conference, which I attended, was that the essence of life lies not in biological substance but in patterned organization.
   If this idea is valid, the thinking goes, life forms should be able to set themselves up through many different types of material stuff. In particular, life should be able to exist as a pattern of electronic activity in a computer.
   The conference organizers, casually dressed, long-haired men in their thirties and early forties, say that artificial, computer-based life forms are developing even now—and may evolve to dominate the earth.
   According to this view, the evolutionary role of man is to give birth to silicon-based life patterns that will eventually look back on him as a primitive ancestor. The conference sponsors counseled a broad-minded attitude toward such evolutionary progress: we should transcend parochial anthropocentrism and welcome advanced life in whatever form it may emerge.
   However, some attending scientists doubted whether a program running on a computer could be properly thought to be alive. Philosopher Elliott Sober argued that when engineers make a computer simulation of a bridge, no one would think of it as a real bridge: the simulation merely shows a picture in which computations tell us something about bridges. In the same way, when a computer simulates an organism, we see a picture in which computations tell us something about life—we’re not seeing life itself.
   Tommaso Toffoli, a computer scientist from Massachusetts Institute of Technology, responded to this argument. Suppose, he said, that simulated people were driving in simulated cars on a simulated bridge. If the bridge were to collapse, the people would fall to their simulated deaths.
   The patterns in a faithful simulation match the patterns found in reality: the simulated people cross the simulated bridge just as real people cross a real bridge. And since these patterns, Dr. Toffoli proposed, are the essence of what is happening, we can think of the simulation the same way we think of the original.
   In principle, then, if a real material scene can exhibit life, so can a simulation. . . .

Copyright © 2004 by Richard L. Thompson