The Social Animal. W. Runciman G.

but with the possibility of small but significant differences for which the conventional term is mutations. Second, these mutations must have the property of being capable of influencing their chances of reproducing themselves in turn – with, of course, the inherent possibility of yet further mutations. Whether mutations do in fact survive, spread and replicate then depends on how far the environment in which they emerge is favourable to their likelihood of doing so.

      This, let me emphasize as strongly as I can, is not the ‘survival of the fittest’ in the vacuous sense that the evidence for fitness is survival and survival the evidence of fitness. It’s a question to be settled by empirical research what mutations, whether in organisms, cultures or societies, have had their probability of replication and diffusion enhanced (or not) by what features of their environment. There is still room for argument about the value of the concept of evolution in the study of human societies – some of it prompted by the irrelevant fear that it will have implications which could be held to be ‘politically incorrect’, and some by the misguided suspicion that it implies that sociology is nothing more than applied biology. Anthony Giddens, who was for some years the professor of sociology in my own university of Cambridge, used to insist that there is no place at all in social theory for the concept of evolution, which to my mind is about as sensible as insisting that there is no place in physical theory for the concept of gravity. Nobody can deny that human groups, communities, institutions and societies are of many different kinds and that they all change sooner or later from one kind to another. So no sociologist, even those who start frothing at the mouth at the mention of Darwin’s name, can seriously dispute the proposition that something has to have happened to cause them to do so.

      The history of human social behaviour, accordingly, is inescapably ‘evolutionary’ in the sense that all new forms of it have evolved out of previous ones, but not – emphatically not – in the sense that change from one form to another is in the direction of some final state of affairs which can be specified in advance: that is precisely the mistake which rightly discredited nineteenth-century ideas about social evolution in twentieth-century eyes. The story goes all the way back to the emergence of organic matter out of the basic chemical ingredients of the universe as found on planet Earth, and forward all the way to the human mind and its ability to build and program computing machines which themselves have ‘mental’ capabilities. This doesn’t mean that the things which have emerged in the course of it are all things of the same kind whose workings can all be explained in the same terms. Our thoughts can’t be explained directly in terms of physics, even though our minds consist of nothing other than exceptionally complex molecular machinery. Nor can our institutions be explained directly in terms of biology, even though social behaviour consists only of what is done by individual organisms with minds in interaction with one another. Although evolution is, so to speak, seamless – God did not, one Sunday morning, decide suddenly to implant life into matter, and another Sunday morning decide suddenly to implant minds into living things – the changes which result from ‘descent with modification’ are of kind as well as degree. The important consequence, so far as sociology is concerned, is that what human beings do has to be analysed at three different levels which correspond to three different kinds of behaviour for which I shall from now on use the terms evoked, acquired and imposed.

      Suppose you are watching a baseball game at the Yankee Stadium. Provided you know the rules of baseball, you are in no doubt what is going on: the batter goes back to the dugout because the outfielder has caught the ball which the batter hit before it reached the ground, etc. But there are still three different ways in which you can look at it. From a biological (or ‘sociobiological’) standpoint, it’s an instance of human beings’ inborn propensity to enjoy sports and games in which the participants try to outrun each other, or throw or catch a ball of some kind, or wield an implement with which a ball can be hit. But from a cultural standpoint, it’s an instance of how psychologically gratifying leisure pastimes and the idioms, styles and fashions that go with them are popularized through imitation and learning among adjacent and successive populations. And from a sociological standpoint, it’s an instance of the workings of a capitalist economy in which professional sportsmen are hired by the proprietors of rival teams out of the proceeds of what the fans will pay to watch them (and the sponsors to advertise on the TV channels which show them).

      The direct response of players and spectators to the hitting of a moving ball is evoked behaviour: it is elicited by a stimulus to which we react as a result of those hundreds of thousands of years of natural selection during which those of our ancestors who had reactions like these were more likely to live long enough to replicate their genes than those who didn’t. The idioms, styles and fashions which attach to this particular sport, however, are acquired behaviour: they have been adopted by those who chose to do so from other people, whether known face-to-face or indirectly. And the hiring of salaried players by rival proprietors is imposed behaviour: although the contracts of employment are freely entered into, the transaction is conducted in accordance with institutional rules which, like all institutional rules except those framed at a time of constitutional choice, are not of the parties’ own making. To be sure, we can’t occupy and perform our roles without having learned the rules which govern the practices which constitute them. But although imposed behaviour presupposes acquired behaviour, just as acquired behaviour presupposes evoked behaviour, it is not merely an instance of it. A strike of professional baseball players is more than a matter of taste, just as their jargon and style are more than a matter of instinct.

      Although I’ve chosen a game as an example, I could just as well have asked you to suppose that you’re observing a religious festival, a court case, a stock market crash (or boom), an election, a battle, a strike, a revolution, or an office party. Whatever form of social behaviour it is, you will start by asking yourself what these people are doing – which means ascertaining what roles they are occupying and performing. But all three aspects of their behaviour will have to be covered before you can satisfy yourself why they are doing what they are doing – i.e., manifesting evoked, acquired and imposed behaviour of a specified kind. In practice, sociologists seldom observe directly the patterns of behaviour they are studying. But whether they are dealing with documents, eyewitness reports, tables of statistics, answers to questionnaires, or even monumental inscriptions or archaeological objects, the nature of their task is the same; and if they succeed in it, they and their readers will be left with a validated account of how the particular group, institution, community or society functions and how it has come to be what it is.

      Since all new forms of human social behaviour have evolved in one way or another out of old ones, the process which has brought about any particular form of it is by definition a selective process: to a sociologist, history is not just one damn thing after another, but one damn thing instead of another. But this immediately leads to the question of what it is, at each different level, that the ongoing process of selection selects. At the biological level, the objects of selection are genes. This has, as it happens, been disputed until very recently by biologists who have held that natural selection selects either the individual organism or the group; but neither organisms nor groups fulfil the conditions necessary for them to act as replicators in the way that genes do. At the cultural level, however, when instinct is supplemented by imitation and learning, the objects of selection are the units, or bundles of units, of information or instructions affecting behaviour which are passed from mind to mind. Some sociologists and anthropologists use for them the term ‘meme’ which was coined in the 1970s by the biologist Richard Dawkins,⁴ whereas others prefer to use the term ‘trait’ in order to allow for the replication not merely of units of information but of whole complexes of representation such as works of art, scientific theories, systems of myth and ritual, and so on. But it doesn’t much matter which term you use. The point is that to explain cultural evolution – i.e. changes in patterns of acquired social behaviour – you have to have a hypothesis about the features of the environment where the behaviour occurs which have helped the mutant ‘memes’ (or traits or bundles of instructions) to spread and replicate.

      At the social level, by contrast, the objects of selection are, as I’ve pointed out already, units of reciprocal action, since the rules which define the roles we occupy and perform are prescriptive for both parties