The ice is beginning to crack in another section of the cold, hard surface of modernity. The part of the frozen lake that is breaking up this time is Darwinism, or at least Darwinism as a worldview with implications for culture and social policy. And as with the breaking up of Marxism and Freudianism, we may not like what bubbles to the surface when all the ice finally melts.
The death of Darwinism, it now seems clear, will be chaos theory (or “complexity theory,” as the researchers in the young science of emergent order seem to prefer). The problem with Darwinism, as honest Darwinists have always admitted, is that it has nothing to say about how new features of living things arise or how new instinctual behaviors originate. It has a great deal to say about how natural selection can preserve and refine these things once they appear, of course. It inspired a search in the fossil record for the lineages of living creatures that has given our natural science a unique historical depth. Darwinism is not wrong; it just is not the final answer to the question of how living things originate. We have been looking in the wrong place for that answer-which must be not so much a matter of genes as of the order that arises spontaneously from simpler units (particularly the geometries of complex molecules).
Even children often notice that the paw and foreleg of a dog are like a strangely distorted human hand, with the thumb appearing as a useless claw partway up the foreleg. The skeletons of all vertebrates, in fact, have long been known to be variations on a few basic structural themes. Anatomical analogies show up among all classes of living things. These analogies were the basis in the eighteenth century for the comprehensive classification of species drawn up by Carolus Linnaeus. And these variations on common anatomical themes occur not only between species. Within individual organisms, sophisticated features grow from the variation of more primitive ones, as illustrated by Goethe’s still persuasive derivation of all the major parts of a plant from the basic form of the leaf.
A description of nature like this invites the search for common mechanisms in living things to generate the archetypical forms. If biology had maintained this perspective into the next century, however, the search might not have produced fruitful results, since the physics of the time was unable to address the question of the spontaneous generation of order. In any event, this way of looking at biology was reduced to a minor theme of scientific thought for several generations as more accessible avenues of research appeared.
In the nineteenth century, as Michel Foucault noted in The Order of Things, the structural diagrams of eighteenth-century biology became timelines, descriptions of the lineages of living things. This, of course, was only one of several areas in which knowledge was being given a genealogical twist. Much the same was happening in philology, as the comparative description of Indo-European languages became the description of the descent of language families. In the English-speaking world, however, the pervasive genealogical cast of modern Western thought was expressed primarily through the new, radically evolutionary biology that Darwin introduced. Darwinism became the governing mode of thought among biologists not because they felt it provided a foundation for early industrial capitalism or for some other ulterior motive, but because it produced results. It did not produce quite the results earlier scientists would have liked to see, of course. It led biologists to ask historical questions, to enquire “how” rather than “why” the living world came to be as it is, and evolutionary biology became a science of narratives rather than general principles. But the results Darwinism did produce were so important that only today do we perhaps see how the Darwinian perspective could ever be exhausted.
The problem is that the genes, the medium of natural selection, do not exhaust heredity. The DNA in the nucleus of a cell cannot even reproduce itself. It requires the special environment provided by the complex anatomy of the cell to do so. When a cell divides, many other bodies have to reproduce themselves besides the nucleus. What DNA does provide is information for the manufacture of proteins. What the proteins then do in turn, however, is a matter of structural chemistry. The British biologist Brian Goodwin, in his 1994 book How the Leopard Changed Its Spots, shows just how much autonomy proteins have in bending and forming themselves into the stuff of living things. There are unicellular organisms that will partially regenerate their quite complicated surfaces even if the nucleus is removed, guided by chemical geometries even more complicated than those that give us snowflakes and mineral crystals. You would never guess from looking at the formulae of these chemicals that they can form branches and other structures, quite without micromanagement from the nucleus. Researchers in complexity theory were not the first to notice that order can arise spontaneously. Thanks to some new mathematics and the number-crunching abilities of computers, however, we now have some understanding of how this is possible.
The unsettling thing about chaos theory is its casual dismissal of material reductionism. The stuff of which something is made does not necessarily determine its behavior. Rather, the behavior of material is guided by certain “shapes” that turn up throughout nature, in the living and the nonliving. It has long been known, for instance, that plants and animals often subtly incorporate familiar number sequences into their anatomies, such as those plants whose leaves are arranged around their stems in accordance with the dictates of the Golden Section, or with the Fibonacci series. Today we can see in detail how even simple, nonliving materials can also assume these forms. The mathematics that governs the “periodic” patterns formed by colonies of slime molds turns out to be the same mathematics that describes the function (and malfunction) of the heart. Like other complex systems, life manifests “emergent” properties, abilities of which there is no hint in the material that makes them up. Life’s special property, however, is to be able to regenerate itself, and by extension to reproduce itself. Goodwin argues that this self-sustaining ability is neither a miracle nor an accident, but a probable state into which the appropriate materials will fall, just as water running down a drain will assume the shape of a whirlpool.
The random mutation of genes explains neither life’s origin nor the complex structures of living things. Random accident simply will not provide the instructions needed to build an amoeba, much less an eye. If what the complexity theorists say is true, however, then the basic features of life are often-repeated variations on “accidents waiting to happen.” The formation of the eye, for instance, seems to be a recapitulation on a smaller scale of the folding and differentiation process that produces the whole nervous system. The fact that the formation of eyes-from the eyes of flies to the eyes of eagles-may be triggered by a genetic “switch” common to all forms of seeing life illustrates how a single piece of genetic information can produce widely different anatomical structures. Genes are important, of course. They define the field (called the “morphospace”) where the possible forms of an organism are found. The field is not flat, however. Irregularities do not arise from genes, but from general formal principles. For a normal developing organism, there are “potholes” in morphospace into which they fall. For an evolutionary line of organisms, morphospace is sloped in a way that guides evolution in some directions but not in others. Thus, for instance, there are no vertebrates with six limbs, useful though that arrangement might seem to be. Not just anything can happen in biology, time and natural selection notwithstanding.
Today’s Neo-Darwinists are aware of these things, and they are not unduly upset by them. Stephen Jay Gould, for instance, has noted that some biological features seem to be easier to evolve than others. He has made the point a part of his deconstruction of meaning in evolution. Not only does evolution have no particular direction, he says, but organisms do not even make the kind of sense they would if they were perfectly adapted to their environments. Their basic structures are even more arbitrary than we thought, for they seem to be guided in large part by nothing more than formal accidents. Natural selection, Gould says, simply makes organisms that are good enough to survive. It does not make them perfect.
While this argument has some merit, it can only deliver its intended frisson of existential emptiness if the easy-to-evolve forms are both completely arbitrary and very numerous. Gould thought he had evidence for these propositions from the analysis of the Burgess Shale fossils from British Columbia, which seemed to show that the biological world of half a billion years ago was inhabited by weird organisms that were not, for the most part, ancestral to the living things we see today. This biological dispensation was ended by one of the great, sudden die-
offs to which our planet is subject. The survivors survived merely because they were lucky, not because they were in any way superior.
This thesis, argued with great persuasiveness in Gould’s 1989 book Wonderful Life, was badly punctured by fossil finds made in China from roughly the same period. The Burgess Shale fossils had suffered a history that left them rather jumbled, so Gould was relying on careful but problematical efforts to piece them back together. The Chinese fossils were much better preserved. They showed that the strange monsters so imaginatively reconstructed from the British Columbian rocks were in fact mostly more primitive but ancestral forms of creatures with which we are familiar. This suggests that evolution had less choice about following the course it did than the Neo-Darwinian synthesis would lead us to believe.
None of this implies a radical break with biology as we know it. It does suggest something of a return to the more purely structural biology of the eighteenth century. The difference is that we will be able to go a long way toward describing just how anatomical forms arise in a given organism, rather than satisfying ourselves with an account of the organism’s lineage. Parallel evolution, the fact that very similar creatures can arise from quite different evolutionary histories, is likely to attract more attention now that we have a theory for it. Actually, a science of form suggests a return to a kind of biology even older than Linnaeus. How the Leopard Changed Its Spots does not even mention Aristotle in the index, but obviously complexity theory is well on the way to restating something like the notion of formal cause. The difference, of course, is that while Aristotle was very interested in the actual stuff in which forms were potentially present, complexity theory emphasizes the independence of forms from the materials they shape.
There was a strong Platonic streak in the biology of two centuries ago. The “archetypical” forms that supposedly defined the natural world were often incorporated into the “argument from design” for the existence of God. The forms described by complexity are far more abstract than anything conceived in the eighteenth century, however. An argument from design that attempted to make use of them would have to argue not from the design of actual living things but from that of the mathematical objects that govern their development.
If in fact the approach to evolution suggested by complexity theory is as important as its proponents say, one of the unforeseen side-effects is likely to be a noticeable demotion of the ontological status of the gene. Goodwin himself does not seem to appreciate this, since he ends his book with a tirade that, among other things, castigates genetic engineering as an impudent meddling with the forces of nature. But the whole point of complexity in biology is that genes just define the morphospace for other substances to do their work. They are not the “essence” of life. If life has an essence, it lies in the immaterial forms that govern its development. Furthermore, complexity suggests that genetic engineering may have surprisingly narrow limits. What the creature will be like cannot be computed from the DNA in the nucleus of a germ cell. This follows from the basic idea of emergent behavior: the activity of a complex system cannot be computed from the composition of its parts.
Genetics is a wonderful science, of course. We can alter the genetic makeup of a cell so that it produces useful substances, such as insulin, and we can spot mistakes in the genes of individuals. What we cannot do, however, is produce a super-race, or indeed any kind of radically new creature. Novelty in biology means a change not just in the genes, but in the whole organism. It is not in our power to create new archetypes to inform artificial creatures. Such archetypes might be waiting in morphospace for us to stumble on them in the course of doing something else, but that is hardly “engineering.”
Something else we may soon be able to forget about is sociobiology, at least in its evolutionist incarnation. Emergent behavior belongs as much to societies of living things as to the life of individual organisms. It is emergent behavior that permits insect colonies, each individual member of which is almost completely stupid, to act in a wonderfully coordinated fashion. (As Goodwin demonstrates, it is actually simpler to show the mathematical basis of social behavior than of physical morphogenesis, since whole organisms are much easier to observe than molecules.) Attempts to use evolution to explain why people act the way they do tend to take much the same form as evolutionary explanations of anatomy. If you believe writers like Robert Wright (author of The Moral Animal), it is all perfectly simple. Natural selection ensures that those genes that support life-sustaining behavior will be passed on to future generations, just as it ensures that genes associated with useful physical traits will be passed on. In Wright’s scheme of things, of course, communitarian and altruistic behavior are survival-enhancing, rather than the highly competitive and aggressive behavior emphasized by earlier Darwinists. And yet, while it is possible to show by studying primates why a particular behavior may be good for human genes, the only real evidence that a given human behavior may be life-sustaining is that some living people act that way today. In other words, their lineages survived because their lineages survived-which is true enough as far as it goes, but not really very helpful. Certainly it gets us not a millimeter nearer to understanding where these life-
sustaining behaviors came from in the first place. Natural selection is not false, of course: an unfortunate inherited disposition, such as an irresistible impulse to jump off high places, is going to diminish the survival prospects of any family line. Natural selection refines behavioral repertoires that have other origins. It does not explain them.
Yet while the theory of natural selection does not do everything we might wish, at least it does not do too much. Sociobiology is often criticized for suggesting that we are slaves to our genetic inheritances, to which sociobiologists reply that our genes compel us to do almost nothing, but simply give certain predispositions. An attempt to explain human behavior using complexity would actually be much more deterministic. Complexity would turn sociobiologists from looking for etiological myths in the anthropological literature and set them to trying to define archetypes of behavior. One can imagine a whole new sociology that is interested not in fundamental causes, but in the perception of ideal states, like discerning the hexagrams of the “I Ching” in daily life. Doubtless these states will be defined empirically, using the sort of computer models that have proven so helpful in explaining how all the ants in a nest can manage to start and stop in a synchronous rhythm. Post-Darwinian sociobiology, too, will be something best taken with a grain of salt.
Goodwin makes the point that Darwin’s model of evolution closely parallels what he, and others, imagine is the Christian model of history. Both deal with imperfect creatures slowly redeeming their natures as they struggle forward in time. Goodwin hopes that evolution as seen from the perspective of complexity will have an appeal beyond the Christian West. He sees it as an expression of the “original blessing” theology of Matthew Fox. The whole natural world participates in the same fundamental forms, and so trees should have rights like those of human beings, and people should be nice to each other and establish a cooperative, community-based economy. The last two chapters of his book are devoted to these ideas, but they are, I think, something of a non sequitur. Complexity theory is simply not a metaphysics of universal harmony. The more likely effect on popular culture of this view of evolution would not be communitarian, but apocalyptic.
There is not much difference between a Hegelian synthesis and the radically unpredictable emergent behavior of a complex system. The vitalist biologists of the first half of the twentieth century were criticized, doubtless correctly, for appealing to mysterious, unmeasurable organic forces. Complexity theory requires no new physical forces, but its emphasis on unpredictable leaps in nature is strongly reminiscent of Henri Bergson. It invites the creation of philosophies like dialectical materialism, but minus the material. Darwinism went well with a view of history characterized by slow human progress. The new view of evolution, on the other hand, suggests the possibility of sudden cosmic transformation. In New Age eschatology, it is not just mankind that will be made new, but the whole biosphere. Goodwin may have a point when he says that Darwinism bears a family resemblance to some Christian views of history, though not entirely for the reasons he cites. Like the Augustinian model of history, Darwinian time is linear, of indefinite duration, and essentially “forward-looking,” if not invariably meliorative. Both discourage the idea that we happen to live at a time of unique importance to the historical process. But another major Christian view of history is the millenarian, the daily expectation of unknown wonders. Like chaos theory, it is “nonlinear,” and so friendly to surprises and discontinuities. This mode of thought may be the mode of the future-and not just in biology.
John J. Reilly is Reviews Editor of Culture Wars magazine.