to see it as it would have been seen in the year in which it was made – as, that is, a piece of exciting, novel, up-to-the-minute, state-of-the-art technology. Can you do it? Try as I may, I’m not sure that I can. But the imaginative exercise required is the same as when an anthropologist tries to see a totem pole or a rain dance or an animal sacrifice or a Disney cartoon as ‘they’ see them. You don’t do it by tracing the sequence of causes and effects which has made the objects of your curiosity what they are. You do it by bringing to bear your knowledge of the cultural context in which they occur and the language employed by the native informants whom you have interrogated about their significance to ‘them’. And, once again, it’s no different for a Balinese anthropologist trying to understand (in the emphatic, descriptive, ‘what-is-it-like?’ sense) a baseball game at the Yankee Stadium than for Professor Geertz trying to understand a Balinese cockfight.
Suppose, however, that seeing things as ‘they’ see them entails acceptance of beliefs which ‘we’ know to be false (or at any rate think we know to be false – the point stands even if we later decide that we were wrong). Let’s go back to Evans-Pritchard among the Azande. He finds it quite easy to behave as if he shared their beliefs. But he can’t and therefore doesn’t actually share them, and he therefore can’t, whatever further enquiries he makes, see the poison oracle as they do, any more than I can see the wafer in the hand of the Catholic priest as the body of Christ. Does it matter? No, it doesn’t – not for the purposes of sociology. We don’t have to share their beliefs in order to grasp their meaning to them and convey it to you, our readers. You don’t, I assume, share any more than Herodotus did the belief of his Scythian informants that every member of the Neurian tribe is a once-a-year werewolf. But you can still grasp the concept (and enjoy the movie, too, if you don’t find it too scary). Indeed, think what would happen if sociologists and anthropologists did all come to share the beliefs of the people whose patterns of behaviour they had been studying. They could only explain the behaviour correctly if the correct explanation had already been arrived at by those whose behaviour it was. And how often would that be?
What’s more, this applies as much if not more when the beliefs in question are those of rulers, activists and decision-makers as when they are those of sociology professors. Rulers, activists and decision-makers all have explanations of their own of why the societies to which they belong are as they are as well as prescriptions of their own about how their societies ought to be changed for what they consider to be the better. But their memoirs are notorious for their unreliability. Only the most unsophisticated reader will be any more disposed to take them at face value than Evans-Pritchard to agree with his Zande informants that their misfortunes are due to the fact that one or more of their neighbours is a witch. But as you read the selective and tendentious reminiscences of important people, from Julius Caesar and before to Winston Churchill and since, and contrast them in your mind with the accounts of the same events given by uninvolved observers who have sought to test alternative possible explanations against one another, don’t you at the same time find the disjunction between the two entirely comprehensible? As a species, we are not only a compulsively social but a compulsively self-justifying animal, and the autobiographies of politicians need to be checked for their veracity and lack of misleading insinuations and omissions no less carefully than those of philosophers do (Bertrand Russell’s is a classic in this regard).17 But the disjunction between what it felt like to the autobiographer at the time, and how it is going to be explained by revisionist professors fifty years after the autobiographer’s death, is not a reason to question that that was what it felt like. The sociologist studying the societies in which the Great and Good (or Bad) occupied and performed their political roles may be as curious about the one as about the other, and increasingly struck by the irony inherent in the discrepancy between the two. But the discrepancy doesn’t of itself make it any more difficult to arrive at an authentic description or a valid explanation – or both. On the contrary, understanding the delusions of grandeur that led to the downfall of Croesus or Louis Napoleon or Margaret Thatcher may make the causes of it all the easier to see.
But explanation, in sociology or elsewhere, can mean several different things. Why, to go back to my earlier example, do I shake hands with you when I’m introduced to you? Because I don’t wish to seem impolite, because that’s how I was brought up, because it strengthens social ties within our community, because a mutual friend decided that we should meet, because in our culture that’s what we do instead of rubbing noses, or because in ruder and more violent times the symbolic meaning of a handshake was that neither of us held weapons in our hands?
That isn’t even an exhaustive list. But for the practising sociologist the important distinction is the threefold one between genetic, motivational and functional explanations. This difference does not, let me emphasize, correspond to the difference between evoked, acquired and imposed behaviour: explanations of each kind can be sought for all three. But sociologists are, typically, more likely both to be studying imposed behaviour and to be looking for functional explanations. Let me go back once more to the example of infantry drill in seventeenth-century Europe. If you want to know where it came from, the answer lies in a narrative account of the development by Maurice of Nassau, captain-general of Holland and Zeeland for forty years from 1585, of systematic routines for marching and countermarching, loading and discharging matchlock guns, and transmitting words of command down through co-ordinated tactical units. If you want to know what influenced the people involved, the answer lies in the careers and ambitions of the leaders of early modern European armies on the one side and the dispositions and responses of volunteer or conscripted foot-soldiers on the other – responses which may, as I’ve pointed out already, be explained as much by an unconscious bonding effect of co-ordinated movement to the sound of drums or music as by a cultural process of deliberate imitation or learning. But if you want to know why it came to transform the way in which wars were fought, the answer lies in the competitive advantage which armies so drilled enjoyed over their opponents and the function which drill performed in promoting discipline during training and garrison duty as well as on the field of battle.
The same distinction can be made on topics which fall more nearly within the domain of one of the specialized social sciences. If, for example, you are an economist studying the automobile industry, you may want to know about the initial commercial exploitation of the internal combustion engine, in which case you will need to find out about the cost – benefit calculations which showed it to be worthwhile. Or you may want to know about the appeal of the product to its potential purchasers, in which case you will need to find out about not only its utility as a mode of transport but also the effect of advertising in expanding consumer demand for it and the part played by peer-group imitation or rivalry in raising its priority as an item of household expenditure. Or you may want to know why some manufacturers have been more successful than others, in which case you will need to find out about production techniques, marketing strategies, tariff barriers, and rates of technological innovation and obsolescence. Indeed, you may well want to draw directly on models derived from the theory of natural selection, as a number of economists have done, in order to explain why some particular firms and their particular products win out over others in competition for market share.18
These examples can also be used (as the handshake example can) to illustrate the difference between the approaches of sociologists or anthropologists on the one hand and historians on the other. There is a familiar contrast, much discussed by philosophers of social science, between narrative explanations (‘because he couldn’t find a horse, the King of Ruritania lost the battle and therefore his kingdom’) and lawlike explanations (‘all monarchies, including the Ruritanian, depend on some kind of religious legitimation’). But the contrast mustn’t be overdrawn. Narrative explanations presuppose underlying regularities of certain kinds which must be true if the particular chain of causal connections is to hold; lawlike explanations are valid across the range of instances to which they are applied only if specific historical conditions are presupposed too. Lack of a horse only leads to the loss of a kingdom in a context to which implicit generalizations about certain forms of warfare apply, just as religious legitimation of a monarchy can only come about after a series of events which were contingently sufficient for it to do so. Sociologists, it could accordingly