How bacteria grew to love baseball

Stephen Jay Gould emphasises diversity and rejects the idea that evolution means progress. Just don't call him a postmodernist, writes David King. A small anecdote says a lot about Stephen Jay Gould. I ask him about the philosophical implications of his latest book, Life's Grandeur. He tells a story about reviewing one of philosopher/physicist Fritjof Kapra's books on cosmology: "He's a California mystical type, I mean a nice guy. What I said was that I'm a holist in some sort of philosophical sense, but I'm a New York holist, not a California holist, I believe in intellectuality, none of this California touchy feely stuff."

Although for the last 29 years his academic home has been the WASPish elegance of Harvard University, Gould remains a cosmopolitan New Yorker at heart. He was born in 1941 ("the year Ted Williams hit .400 for the last time" - baseball is an obsession which takes up a quarter of the latest book), in the borough of Queens, the son of second-generation Jewish immigrants. He picked up his appreciation of life's variety and his love of dinosaurs early, when he visited the New York Museum of Natural History with his father: "Journalists love that story, but to me it's banal, because it's so common among palaeontologists. More than any other profession I've ever come across, we were childhood enthusiasts."

He was educated in the New York public school system, and has fond memories of his Jewish and Irish teachers. In school he was also taunted as a "fossil-face", a less pleasant memory. His first degree was at Antioch College; his PhD at Columbia in New York. At Harvard since 1967, he is now professor of geology and a curator of the Agassiz Museum of Comparative Zoology. Despite his love of dinosaurs, his palaeontological speciality has been land snails, and he still goes out and chips away at bits of rock, although where he finds the time to do so is difficult to imagine.

For Gould is one of the most prolific essayists of our times. For the past 18 years he has written monthly essays for Natural History magazine, many collected in his books which invariably sell hundreds of thousands of copies. His recipe for such productivity is simple: "Pull out the phone plug and don't write drafts. I maintain the old fashioned skill of writing outlines, so I know what I'm going to write when I start."

His essays are extraordinarily eclectic, dealing recently, for example, with subjects that include an extinct species of South African antelope, the Nazi eugenics programme and 19th-century women naturalists. He delights and sometimes irritates his readers with his fascination for trivia, and the way he uses obscure examples to make his points. But he is an outstanding populariser of science, who seems able to convey scientific complexities to non-scientists without sacrificing accuracy to sloppy metaphor. His one stylistic fault is a tendency to prolixity. Critics sometimes quote passages where it is difficult to escape the conclusion that he is indulging his pleasure at his own cleverness. One reviewer even accused him of "vulgar cleverness".

In the academic sphere, despite universal respect for his scholarship, some critics have been even harsher. Evolutionary biology has always been a contentious subject, for the simple reason that so much is unknown and unknowable. That leaves plenty of space for disagreement over theory, and, as befits a Jewish intellectual, Gould is nothing if not contentious. The results of his efforts are usually productive, but his critics claim justifiably that he takes extreme positions, constructs straw men and is unnecessarily combative. He is careful to stress his areas of agreement, for example, with arch-opponent and rival populariser Richard Dawkins, but it is clear that he is sensitive to these charges.

The central debate of his career has been about the long-term dynamics of evolution. In 1972, together with the zoologist Niles Eldredge, he launched the theory of "punctuated equilibria", according to which species stay much the same over long periods of time, but then undergo rapid bursts of extinction during which new species are formed. It is in these rapid bursts that most evolutionary change takes place, rather than gradually over eons, as the prevailing view had it. Gould's evidence for these radical claims comes primarily from the fossil record, which, he says, contains no evidence for gradual changes within a species.

Punctuated equilibrium theory has various revolutionary implications for evolutionary biology, which have provoked a vigorous response from defenders of the previously prevailing consensus. The debate continues, although most evolutionary biologists have accepted much of what Gould says. Perhaps most shocking is his downplaying of the importance of natural selection - the theory that animals evolve largely as a result of adapting, over generations, to their environment, which "selects" those characteristics which best enable them to survive and reproduce. Gould insists, rather, that contingency has an important role in evolution. According to the received view, natural selection constantly creates gradual evolutionary change by favouring the tiniest selective advantages. But if the fossil record indicates long periods of evolutionary stasis, natural selection cannot be this kind of creative force. A further implication of the older paradigm is that if we knew what the selective forces were, we might be able to predict which evolutionary changes would occur. But Gould points out that evolution can only work with what it has to hand, so much of evolutionary change is accidental: if a successful individual happens to have a particular number of toes, for example, then all its descendants will have that number of toes, even though having that number of toes gives no particular advantage.

In pressing this point Gould sometimes goes too far, laying himself open to the charge that he does not believe in natural selection. But he insists that his work is an extension of Darwinism, not a challenge to it. "The overarching theme of my work is that you can't have an adequate evolutionary theory merely by looking at what's happening in local populations at the moment and extrapolating its adaptationist and gradualist style of change to the entire history of life. That's what Darwin wanted to do ... My examination of the fossil record requires that you develop a body of independent theory for large-scale events that take long periods of time ... Natural selection is a powerful, beautiful theory and it's correct. I just don't think it's fully adequate, or close to fully adequate."

Gould is also well known for his criticisms of sociobiology, which holds that the way humans behave and organise themselves socially can be explained as a biological adaptation to evolutionary forces. What about recent efforts to rehabilitate sociobiology, under the new title of evolutionary psychology, with its additional argument that modern human behaviour can be explained as a legacy of the kind of traits humans needed to survive two million years ago as hunter gatherers? Gould is gently scathing. "No one ought to have any objection to the proposition that evolution should be able to offer profound insights into the nature of behaviour. The problem with the old sociobiology and the new evolutionary psychology is that they're very naive pan-adaptationist theories, which rely entirely on the adaptationist component of Darwinism. The failure of the old sociobiology was that it tended to look at just about anything as adaptive now, or at least maintained for adaptive reasons. The evolutionary psychologist will say: 'No, we recognise that's wrong, there are many evolutionarily coded behaviours, like aggression, which are profoundly non-adaptive now, but, when they arose on the African savannas, they were adaptive.' "Now in a sense that is a more sophisticated insight, I'll grant them that. On the other hand, as a scientific proposition it's even worse than the old-time sociobiology, because at least the old sociobiological theories could be tested. If you say it's adaptive now, OK, go out and see if it does increase reproductive success. But the minute you say, 'it was adaptive on the African savannah', it's not a scientific theory any more. There's no way to test that proposition ... language doesn't fossilise, kinship doesn't fossilise ... how are you going to know what happened two million years ago to a band of hunter-gatherers on the savannah? You're reduced to speculative storytelling."

A related aspect of Gould's revised Darwinism is the lack of progress in evolution. This is one of the central themes of Life's Grandeur, in which he argues that the main mode of life on this planet has been and will remain bacterial. A large section is devoted to a rapturous appreciation of bacterial diversity, flexibility, sheer numbers and evolutionary success. But if life started with bacteria, and progressed through sponges and dinosaurs to homosapiens, surely that's progress? Only from a culturally determined viewpoint, and based on a Platonic misapprehension of the nature of reality, says Gould.

He fulminates at length against the evolution-as-progress interpretation of Darwinism, to which Darwin himself succumbed only reluctantly, which was imposed due to the 19th-century ideology of progress, and our self-importance. Evolution has always tended to be depicted as a ladder ascending to humans, or a series of "ages", always increasing in complexity. Yet the mode (the most common value, which is different from the average) of the distribution of complexity remains firmly anchored at the simple bacterial end. Increased complexity in some species arises randomly, because of the statistical effect of the "drunkard's walk". Because it is impossible to get less complex than bacteria, random motion will, over time, always produce an expansion at the other end of the the complexity distribution. But we should always remember that we are produced by random motion; we are not an inevitable result of natural selection operating in favour of increased complexity. For Gould this is the completion of Darwin's revolution, the dethronement of humanity from its self-centred view of its own importance.

What is new in Life's Grandeur is the author's excursion into philosophy. The book's American title is Full House, and it stresses the importance of apprehending the full range of variation in a system, rather than Platonic ideal abstractions, such as averages. This is not simply about a better approach to the truth, but because as an evolutionary biologist he loves variation, whether in snails or in human culture. The eclipse of the variety of the American diner by the uniformity of McDonald's, is, for him, an aesthetic issue.

Gould is, of course, part of the great tradition of liberal humanism, and proud of it. But his emphasis on diversity and rejection of the ideology of progress strikes resonances with postmodernism. It is a label which, true to type, Gould is reluctant to whole-heartedly embrace: "Of course, there are aspects of my work which are consistent with postmodernism. But no liberal scientist has any time for the notion of the relativity of truth. Of course, science is socially embedded, all science is done in a social context. But there is an external reality out there, and we do get a better approach towards it as we proceed through the history of science."