London Review of Books


John Maynard Smith

  • The Panda’s Thumb: More Reflections in Natural History by Stephen Jay Gould
    Norton, 343 pp, £6.95, April 1981, ISBN 0 393 01380 4

Pandas are peculiar bears, which spend most of their days munching bamboo. To do this, they strip off the bamboo leaves by passing the stalks between their flexible thumb and the remaining fingers. But how can a panda have an opposable thumb, when in bears the thumb lies parallel to the fingers, and inseparable from them? In fact, the panda does not have a proper thumb at all: it has five parallel digits just like other bears. The apparent ‘thumb’ is a modification and extension of a small bone in the wrist. For Stephen Gould, this is a particular and fascinating fact, but it is also an illustration of a general principle. The principle is that evolution proceeds by tinkering with what is already there, and not by following the canons of optimal design. Had the panda been designed by the Great Artificer, He would not have been constrained to make its hand by modifying the hand of a bear, and would doubtless have come up with a more elegant, if less entertaining solution to the problem of stripping bamboo.

This example comes from the first of this series of essays on natural history, the theory of evolution and the history of biology. It is characteristic of something which, for me, and clearly for Gould, is one of the major attractions of biology: the bringing together of particular fact and general theory. Darwin made it possible for us to see nature simultaneously with the eyes of a child and of a philosopher. The book is full of such examples. What deep lessons would you draw from the turtles that migrate 2000 miles to breed on Ascension Island, or the male mite that mates with all his sisters and dies before he is born?

Stephen Gould is the best writer of popular science now active, and this book the most enjoyable I have read for a long time. It has the two essential features of good science writing: it tells me of facts and ideas that are new to me, and it makes me want to argue with the author. Let me first choose some facts, almost at random. Although I am a professional biologist, I did not know about the panda’s thumb. I learnt a lot about the Piltdown forgery, and was delighted to find that my long-felt suspicion that Teilhard de Chardin had something to do with it is not entirely without support. I was strangely encouraged by the story of Randolph Kirkpatrick and his crazy theory of Nummulospheres. I defy anyone to read Gould’s account of the evolution of Mickey Mouse without feeling better for it.

I can perhaps best give an impression of Gould’s thinking if I describe two essays I mean to argue with him about next time we meet. It is a striking fact that, although Darwin and Wallace arrived independently at the idea of evolution by natural selection, Wallace never followed Darwin in taking the further step of asserting that the human mind was also a product of evolution. Gould has an interesting explanation of this difference. It arose, he suggests, because Wallace had a too simplistic view of selection, according to which every feature of every organism is the product of selection, whereas Darwin was more flexible, and recognised that many characteristics are historical accidents or the unselected corollaries of something that has been selected. Now there are features of the human mind which it is hard to explain as the products of natural selection: few people have had more children because they could solve differential equations or play chess blindfold. Wallace, therefore, was driven to the view that the human mind required some different kind of explanation, whereas Darwin found no difficulty in thinking that a mind which evolved because it could cope with the complexity of life in primitive human societies would show unpredictable and unselected properties.

This is an interesting idea, but it leaves something out of account. Darwin, as soon as he had become convinced that evolution had occurred, and before he had conceived of the theory of natural selection, opened a notebook[*] concerned with questions of psychology and metaphysics. The only explanation of this is that he felt at once that his theory must apply to man, and knew that this required that he develop a materialist theory of psychology. I do not know why Darwin so readily made the extension to man (although it was characteristic of him to push ideas to their conclusions), but I do not think it could have had anything to do with his views on selection, which had hardly been formulated at the time.

A second essay I want to argue about is called ‘A Quahog is a Quahog’. The question discussed in the essay is whether the names given by non-Western people to animals and birds do or do not correspond to the groups recognised by modern taxonomists. The answer seems to be that, so long as we confine ourselves to the species level, the correspondence is remarkably close. Species, to a modern biologist, are reproductively isolated groups – for example, Great Tits, Blue Tits and Coal Tits in Britain. At levels higher than the species (for example, the class Mammalia, the order Artiodactyla or the family Bovidae) there is virtually no correspondence at all.

I think this conclusion is essentially correct, and is an important proof that truth is not as relative as our cultural relativists would have us think. Species were made by nature, not by man. Why then do I want to argue with Gould? Only because of the last paragraph of his essay, which argues that, since species are real entities with discontinuities between them, it follows that evolution has proceeded in jumps and not gradually. The argument is persuasive, but it overlooks an important point. Primitive people who give names to animals do not travel about much. So long as one keeps in the same place, the distinctness of species is indeed a reality, but if one travels about, it largely disappears. Populations of a ‘species’ from different places are not identical, and are often so different that it is a matter of taste whether one regards them as belonging to one species or two.

I hope it will be obvious that my wish to argue with Gould is a compliment, not a criticism. Popular science should reflect science as it is practised: this means that it should reflect controversy and uncertainty. Anyone familiar with current debates in evolutionary biology will have noticed that my disagreements about Wallace and about Quahogs reflect a disagreement between Gould and myself about evolutionary theory.

The excellence of The Panda’s Thumb provokes me into making some more general remarks about popular science. I must make clear at the outset that I do not regard popular writing or broadcasting about science as an inferior or unimportant activity. My own education in science, to the age of almost thirty, depended entirely on reading the popular works of men like Julian Huxley, Wells, Haldane, Jeans, Eddington and lnfeld. Had I not been inspired by them, I would not later have become a scientist. More than that, the ideas I got from them were profound, not superficial.

The list of writers I have just given, together with a later generation – for example, Medawar, Dawkins and Gould himself – suggests that a successful writer of popular science must have two qualities. First, all except H.G. Wells were practising scientists, and the exception in a sense proves the rule, because Wells would dearly have liked to be. Second, and more difficult to define, all wrote from a highly personal, even prejudiced point of view.

To take the second point first, the need for a personal viewpoint, if harder to define, is easier to understand. Who wants to eat a boiled egg without salt? I have not yet recovered from the pleasure of discovering J.B.S. Haldane’s unique mixture of militant rationalism and paradoxical intelligence. Peter Medawar would be a duller writer if he didn’t hate all pompous writing, and all philosophy except Popper’s. Gould’s idiosyncracies are a passion for the quirks of history, and a conviction that a man’s science is part of his humanity, and not infrequently influenced by his political, sexual and racial prejudices. He also holds sadly misguided views about the mechanisms of evolution, and fails to share my prejudice that an ounce of algebra is worth a ton of words. These views, whether or not I share them, are an essential ingredient of his success as a writer.

But why should successful popularisers be scientists? Everyone knows that the prose written by scientists is constipated, full of jargon and altogether lacking in grace. Is it not better that science should be transmitted to the public by professional journalists and broadcasters? It is perhaps inevitable that it should be, because it is unhappily true that many of us do write and talk badly, and few are prepared to take the trouble to improve. However, the results are not always fortunate. The danger is best illustrated by that often excellent TV programme, Horizon. As the years go by, Horizon deals less and less with science, and more and more with the politics, the social consequences and the technical and medical applications of science. Watching Horizon, or reading the New Scientist, sometimes drives me to feel that the politics of science is an expense of spirit in a waste of shame, that technology is something Mrs Thatcher wants us to do, that the social responsibility of science is an honourable and necessary bore, that the philosophy of science is an entertainment for those who have passed the philosophopause. What matters is science. I suspect that the reason why, for me at least, the best popularisers have themselves been scientists is that, however interested they may be in politics or history or philosophy, their first love is science itself. Stephen Gould is deeply aware of the social setting in which scientists work, but he does really care about the science they do. Like me, he fell in love with dinosaurs when he was a boy. Often he infuriates me, but I hope he will go right on writing essays like these.

[*] Published in Darwin on Man by H.E. Gruber and P.H. Barrett (Wildwood). A two-volume paperback version is published on 24 September: Metaphysics, Materialism and the Evolution of Mind (228 pp., £6.95, 0 226 13659 0) and Darwin on Man (328 pp., £6.95, 0 226 31007 8).