Sing, prance, ruffle, bellow, bristle and ooze

Armand Marie Leroi

This article originally appeared in the London Review of Books, 17 September 1998

Over the past three years, the London School ot Economics has been holding a seminar series, or rather a salon, snappily titled Darwin@LSE. These seminars are always invigorating, and never more so than one evening this February when WG Runciman urged the necessity of refounding sociology along Darwinian lines. Weary of such pronouncements though they might be, even the most sceptical sociologists could not have failed to realise that here was a serious challenge to theoretical orthodoxy. For the speaker was no renegade entomologist, but the author of A Treatise of Social Theory, arguably the most deeply considered and formidable exposition of historical sociology in recent times and one, moreover, that rests on a Darwinian view of society. But as Runciman, the most courtly of men, argued his case with wit, clarity and the utmost intellectual candour, I found myself engaged with a nagging question: who is his tailor?

For he was not dressed as were the other academics present, what with the faultless lines of his suit, the softness of the fabric (visible at 40 feet), the dove grey shirt, the soberly elegant tie, and all of it without stain or crease, suggesting another dozen of each in a walk-in closet at home. Quietly and unmistakably, that suit spoke of money and influence. Why did Runciman wear it? Because he had evidently come to the LSE directly from his important day job in the City, and suits of understated perfection are the necessary uniform of such men. [Garry Runciman had something to say| about this.]

Why, though; are such suits necessary at all? This question was answered in 1890 by Thorstein Veblen in his Theory of the Leisure Class, where he had something to say about men's dress as about virtually all human artefacts:

Much of the charm that invests the patent leather shoe, the stainless linen, the lustrous cylindrical hat, and the walking stick, which so greatly enhances the native dignity of a gentleman, comes of their pointedly suggesting that the wearer cannot when so attired bear a hand in any employment that is directly and immediately of any human use.

Beautiful clothes are the most obvious form of what Veblen called conspicuous waste. It is not my intention, however, to pillory Runciman's taste in clothes, for, as Veblen shows at exhaustive length, conspicuous waste forms part of almost any aesthetic decision; it is almost, but not quite, the whole of beauty. A life without conspicuous waste is hardly imaginable, and certainly not one worth leading. The curious thing is that animals think so, too.

Take stotting. A troop of gazelles which has just detected the presence of a predator might be expected to slink quietly off into the undergrowth, or perhaps bound gracefully in the opposite direction. Not a bit of it. Instead, the antelopes bark, thump the ground with their hooves and stot, i.e. jump up and down while displaying their white rumps and whirling black tails to their aggressor. Only when the predator begins its strike do the gazelles pick up and run. This is the sort of thing that Amotz and Avishag Zahavi, two biologists from Tel Aviv University, call a 'handicap': a behaviour or structure which seemingly defies Darwinian rationality. A stotting gazelle wastes energy that might surely be better used in flight; it also draws attention to itself precisely when discretion would seem to be the better part of valour. Such handicaps are apparently ubiquitous when one creature attempts to convey information to another. The dawn chorus may enchant or irritate at 4a.m. but, if it irritates you, spare a thought for the singers, who are exhausting themselves in the effort. As is the kudu, when it wields its enormous horns, or the lyrebird wagging its fantastic tail, or the Great Bustard, in its elaborate dance to entice a mate or deter a rival; everywhere, when animals signal they are seemingly wasteful of their time and energy and running the risk of death. That, say the Zahavis in The Handicap Principle, is the whole point. When animals (and plants and fungi) sing, prance, ruffle, bellow, bristle or ooze, they are advertising their own Darwinian fitness, enticing those who desire to mate, deterring those who seek to subjugate, and dissuading those who simply see them as food. When stotting, the insouciant gazelle signals to the predator that it has vigour to spare, and will almost certainly elude an attempted strike; far better that the canny predator should try another antelope stotting with less conviction, or not at all. It's all pure Veblen. The elaborate signals displayed by so many creatures are conspicuous waste, the biological analogue of any luxury good: useless, even harmful, but an indispensable index to money in the bank.

The Zahavis are not the first to see meaning in animal signals, but no one before has seen as much meaning, or claimed with such conviction to understand it. In the 'handicap' principle they have found an interpretative tool that reveals significance in the most obscure phenomena. Even baker's yeast, when choosing a mate, is credited with subtle discrlminatory powers of a type that one would not normally expect from a fungus. The revolutionary import of the Zahavis' view is well illustrated by their discussion of that touchstone for evolutionary biologists, the peacock's tail. For the Zahavis, each feature of the peacock's courtship display - the shimmering colours of the tail, the posture in which it is held,the way it is shaken, the roaring and stamping - are all there to proclaim male vitality, for it is only the healthiest birds that can display a perfect tail and bear the costs of doing so. Even the eyespots are claimed to be uniquely suited to advertising male quality, for they will show imperfections to a degree that simpler geometries would not.

This is quite different from Darwin's own view. In the Descent of Man and Selection in Relation to Sex (1871) he argued that the peacock's tail, like all aesthetic traits, was essentially arbitrary in form. Females, perhaps because of some cognitive quirk, tended to favour males with certain attributes, say, a slightly more elaborate tail. Male tails and female choice would then evolve in tandem, mutually reinforcing each other, by a process that Darwin's successors called 'runaway sexual selection'. If the peacock's tail speaks of anything, it speaks only of the peculiarities of female psychology, a love of beauty for beauty's sake, and nothing of male quality. Indeed, it is just as likely that sexual selection would sometimes result in the evolution of very small tails.

Who is right? It is difficult to know. The Handicap Principle is a work of advocacy rather than a judicious review of the evidence. the problem is that discriminating between different evolutionary explanations for animal communication requires careful measurement of the costs and benefits to both those who are broadcasting the signal and those who are listening, and such measurements are difficult to make. Yet there is no denying that the Zahavis' view is compelling, drawing our attention to things that we would not expect to find otherwise in animals. That generosity can have earthly rewards is a commonplace. Veblen explained dissipation among journeyman printers in 19th-century America as a consequence of their 'treating' each other to endless rounds of drinks in order to gain prestige. More unexpected is that the babbler, a small bird that lives in hierarchical groups, does much the same thing. Members of a group vie for the right to feed nestlings, undertake guard duty, even attack snakes. None is more eager in the performance of these duties than the dominant bird; were it not, it would lose face and be at risk of a coup d'état. In short, 'conspicuous waste' and the 'handicap principle' represent a remarkable convergence between sociological and evolutionary theory. Even more remarkable is that for once the sociologists got there first.

The thesis of Runciman's new book, The Social Animal, is one that can be stated very simply. Sociology is both a science and an evolutionary science. How you take this will depend on who you are. If you are among the many theoretical sociologists with a distaste for 'positivist sociology', then the brisk common sense of this little book will have no appeal. But if you are, as I am, a natural scientist who has despaired of finding a sociologist whose thought is deeply infused with what is sometimes called (though not by scientists, who don't talk about such things) the 'scientific temper' then you will read this book with a sense of blessed relief.

By the 'scientific temper', I mean the taking for granted of those methods which natural scientists have used to such good effect, and the assumption that they can be applied to the study of human societies. Thus the statistical devices that might be used to study conflict and coercion within, say, the British Labour Party can, in large part, be applied with equal facility to a colony of naked mole rats. Many sociologists see it differently, however. Backbenchers, they might say, are self-conscious in a way that mole rats are not; they are capable of explaining their own actions. But so what? asks Runciman. 'The !Kung of the Kalahari are as aware as the professors and graduate students who study them of the function of meat-sharing in reinforcing their social ties. But the function would be the same even if they weren't.' Ah, replies the sociologist, but humans are also reflexive - they can respond to what others say or do - especially sociologists. But again, this makes the study of human society no more difficult than that of many natural systems in which the behaviour of some entity (atom, molecule, cell, tissue, a mole rat) depends on what other entities like it do. Nothing is more reflexive than the way we earn, save and spend our money, yet this doesn't seem to fuss economists, who accommodate such behaviour quite comfortably within systems of differential equations.

The problem seems to be that, for reasons peculiar to the history of their discipline, many sociologists take a rather provincial view of natural science, as a search for 'general laws'. General laws are, without a doubt, nice, if you can find them, but the business of scientists is to explain the particular as well as the universal; to understand things that are true for one moment in history, and other things that are true for all time; to supply facts that are beyond dispute, and disputes for which there are (as yet) no facts. A grab-bag of activities, but somehow all of it science. The methodological relativist may ask: what. then, isn't science? The answer is plain: most intellectual goods, and in particular, the goods purveyed by the Merchants of Attitude and Platitude.

Attitude Merchants are, in Runciman's words, those sociologists 'who allow their personal views about the behaviour they are studying to inform their conclusions about it to the point that they neglect or devalue uncomfortable evidence for the sake of those views.' They are legion, but Runciman singles out the current President of the International Sociological Association, Immanuel Wallerstein, for having devoted his life to understanding the relative wealth of nations even though he 'hates capitalism. . . like a Sunni Muslim hates a Shiite Muslim, or a Freemason a Jesuit, or a Rangers supporter a Celtic one.' That wouldn't matter, except that this hatred has, according to Runciman, blinded Wallerstein to facts which do not fit his Neo-Marxist theory of transnational economic exploitation.

It is the Platitude Merchants, however, who have been the true plague of 20th-century sociology: those sociologists 'who allow their conviction of the importance of some general truths about human behaviour to convince them that by rephrasing those truths in more impressive sounding words - or, sometimes, more impressive looking but nonetheless trivial mathematical equations - they are advancing sociological knowledge.' Here Runciman picks on Talcott Parsons, echoing C. Wright Mills's memorable skewering of the Harvard sociologist's mock profundities. But he might just as well have picked on a fresher target, Jürgen Habermas, say. Still, this surely hints that Runciman intends The Social Animal to do what Mills's The Sociological Imagination did nearly forty years ago: capture the hearts and minds of generations of sociologists yet unborn.

He intends to capture them for Darwin. He will, I think, succeed, and for a simple reason: the minimal requirements for Darwinian evolution are so few and are found everywhere. Consider a group of 'things'. If these 'things' vary in some way, reproduce more or less faithfully, and some variants reproduce more than others, then the 'things' will evolve. If the 'things' are DNA molecules, or genes, then we are in the familiar world of biological evolution. But they can be ideas as well - love of one's country; the refrain of Mandalay; the infallible cure for athlete's foot: all reproduce when passed from one person to another. Richard Dawkins has called such reproducing ideas 'memes'. Runciman's insight has been to see that Darwinian 'things' can also be whatever occurs when an individual finds him or herself to be a member of a group.

What then occurs are 'social practices', defined by Runciman as 'reciprocal behaviour[s] informed by mutual recognition of shared intentions and beliefs'. A nice example is the system of infantry drill invented by Maurits of Nassau, Stadtholder of Holland and Zeeland from 1585. This drill was a reciprocal behaviour in that officers commanded and soldiers obeyed, and all understood what was wanted: marching and counter-marching in lockstep to the rattle of drums. It was hugely effective and, along with a series of other military innovations, brought the young Maurits repeated triumphs against the Spanish in the 1590s. The innovations caught on and spread rapidly among European armies, displacing other ways of ordering soldiers in battle. In other words, the social practice of war evolved. If the argument seems obvious, that's because Runciman has simply formalised the way in which we normally think about such things.

What really drives social evolution? The answer to this rests on three other questions. First, how do novel social practices originate? Second, how are they transmitted among their 'carriers' (commonly known as people)? Third, what is the selective agent, or, why is one practice chosen over another? Runciman is coy about answering these questions but rightly so, for the practices are so diverse that they cannot be easily summarised in the kinds of simple rules that genetics gives to biological evolution, and the charm of his theory is its generality. Variation in social practices may arise as the conscious invention of Great Men, or as unthinking behavioural changes occurring in countless anonymous souls. Such variants may be entirely new or combinations of existing practices; the analogy is with genetic mutation and recombination respectively.

Their transmission, the analogue of Mendelian inheritance, is equally diverse. Some may be transmitted from parent to child, others may be disseminated by the LRB. An especially interesting factor, not shared by genes, is that their transmission can be influenced by inequalities in power. That is, in a hierarchical society, some people (employers, officers, politicians, bureaucrats) can make us adopt certain practices against our will, while other people can influence us in more subtle ways. To the degree that those born in liberal democracies are subject to what Tocqueville called the 'orderly, gentle, peaceful slavery' of the state, there need be no more potent force driving social evolution. This leaves open the question, if we are free up to a point to choose among social practices, what do we care about? Runciman's answer is power - economic, ideological, coercive or some combination of them. I am not sure that this is always so, however. Sociobiologists, who like to study such topics as polygamy, homicide and incest, might argue that certain social practices, like genes, are favoured for their influence on reproductive success; economists, on the other band; see us as maximising 'utility'. But it really doesn't matter: the logic of social selection is the same, and reproductive success, utility and the various forms of power are closely related; which is uppermost in our minds when we decide what to do remains a question for psychologists, and a very interesting one.

Just as Darwin was by no means original in viewing life's diversity as the outcome of descent by modification, so there are social evolutionists by the score behind Runciman. But Darwin's originality lay in providing a mechanism for evolution and exploring its consequences, and so does Runciman's. One such consequence is that there is no place in his account for the idea, dear to Comte, Spencer and Marx (to name but three), that social evolution necessarily proceeds in a particular direction, through particular stages, perhaps even to greater moral heights. Quite possibly societies do evolve as stage theorists say they do, but that is a matter of contingent history and does not stem from the logic of the selective process itself. As for the moral worth of one society or social practice as against another, 'that', as Runciman puts it, 'is up to you.' All this serves to damp the powder of his potential opponents, particularly the current Director of the LSE, Anthony Giddens, who has long insisted that there is no place in sociology for the concept of evolution. This, Runciman remarks, 'is as sensible as insisting that there is no place in physical theory for the concept of gravity', and after reading The Social Animal one can but agree. Indeed, it is easy to see how the social scientist who seeks to understand the causes of a social phenomenon, be it group fission in the Yanomami, state formation in Madagascar, the homicide rate of New York City, polygamy in the Mormons or the political radicalisation of British foxhunters seeks, in essence, to understand why one practice rather than another spreads in society.

Powerful though Runciman's ideas may be, it is also the case that evolutionary dynamics - whether biological or social -are far more complex and subtle in their causes and consequences than mere English can capture. Evolutionary biologists have known this since the Thirties, when Darwinism, at the time all but moribund, was reborn in mathematical form. in the so-called Neo-Darwinian Synthesis. This was, and is, a tool of great power and formal beauty which allows the consequences of evolutionary forces to be examined with precision. The theory of social selection now requires something like it. Runciman convinces us that societies have evolved, illustrates how they may have done so and provides the elementaIy logic of social selection. This is a grand scheme but far from a predictive and testable theoty of social evolution; for that, a mathematical account of the origin, transmission and selection of social practices is required. Founded on the established mathematics of natural selection, such a theory would have a good start in life, but since there are many more instances of the Darwinian dynamic in society than in organic life, the infant would swiftly outstrip its parent.. A successful theory would constitute nothing less than a universal formal language for sociology. Then we could truly say that Darwin lives @LSE.

Garry Runciman had this to say about his suit:

15 September. At last a review, and a flattering one, of The Social Animal in the LRB by Armand Leroi. But why does he use up precious column-inches talking about the suit I wore when I gave a seminar on the theme of the book at the LSE? I don't remember wearing a suit at all. But if I did, it will have been my nondescript Austin Reed loss-leader, not the dark blue tailormade pinstripe which I keep for telling the shareholders to forget about a dividend this year. Much ribaldry is caused among my family, by whom I am regularly pilloried for my tragic lack of sartorial sense, pig-headed indifference to fashion, and pitiful attachment to short socks and boring ties. I don't need a professor of semiotics to tell me that clothes send a message. But what are those of us who don't have a message to send supposed to wear? Perhaps performing academics should be dressed, like footballers, in the colours of their learned institutions, with their surnames in big capital letters on their backs.

(Diary, London Review of Books, 10 December 1998)