Forgetting the Founders *

Introduction

Whitehead's dictum that ‘A science which hesitates to forget its founders is lost’ was quoted by Robert K. Merton at the head of his Introduction to the influential set of papers which he brought together under the title Social Theory and Social Structure in an enlarged and revised edition in 1957. But it is something which many of his colleagues of the second half of the twentieth century found it difficult to do. The agenda of Anglo-American sociological theory continued to be dominated by Marx, Weber, and Durkheim, and much discussion of their writings in the secondary literature took the form of engaging with them as if they were contemporaries rather than as predecessors whose ideas had been either absorbed or superseded and who were therefore of interest to historians of sociology rather than to practitioners at the forefront of current research. It was only to be expected that many of their conclusions would be invalidated by events which they could not foresee. But they also shared with many of their contemporaries two presuppositions of which both had, by the end of the twentieth century, been radically undermined.

To look back at the period in the history of Western thought when Marx, Weber, and Durkheim were in their heyday is to be reminded of the confident sense of rapid and cumulative advance then manifest across the whole range of Wissenschaft. Old orthodoxies were being toppled not only in physics, chemistry, geology, biology, geography, and engineering but in what were to become the 20^th-century disciplines of psychology, economics, and anthropology. What is more, practitioners of the traditional disciplines of history, linguistics, and archaeology were studying the past – including not least the past as depicted in the Christian Bible – with a hitherto unprecedented professionalism. There seemed every reason to expect that the hopes of the savants of the Enlightenment were on the way to being realized and that natural science, as what had previously been spoken of as ‘natural philosophy’ was by then called, would soon be complemented by a no less authoritative science of (as they thought of it) man.

The first presupposition which they shared was that this emerging science would lead to the discovery of regularities and trends from which the future of human societies could be predicted – not, of course, predicted exactly, any more than economists could be expected to predict exactly what stock exchange prices would be in twelve months’ time, but predicted in the way that Marx predicted the inevitable supersession of capitalism by socialism. Weber, to be sure, disagreed with Marx about that, and was always conscious of the contingencies of human history and the limited distance that our weak eyes can peer into the mists of the future. But he too saw human history as following an underlying pattern and proceeding in a definite direction – in his case, that of progressive rationalization – and Durkheim equally was, as Lukes (1973: 140) put it, ‘not free of an evolutionary perspective’. All three are now routinely criticized for a Eurocentrism which took it too readily for granted that progress, as they conceived of it, was the achievement of the West. But they were not mistaken in seeing around them a reversal in the direction of cultural and social influence which in earlier centuries had been from East to West rather than West to East. The presupposition which it did not occur to them to discard was that there must be a teleological process at work in this which sociologists should be able to understand. They would surely have been surprised, could they have known, that by the beginning of the 21^st century it would have become a virtual commonplace that there is no master narrative of human history, no law of social development, and no privileged vantage-point from which progress (or the lack of it) in social evolution can be assessed.

Their second presupposition was of a different, although not unrelated, kind. They saw themselves as not merely investigators but preceptors whose understanding of how societies function and change entitled them to tell other people how they should lead their lives. Weber, again, may seem an exception to this, with his well-known insistence that sociological research, although it cannot but be ‘value-relevant’, should nevertheless be ‘value-free’. But in his celebrated lecture on the vocation of Politik he left his student audience in no doubt that an ethic of responsibility is more to be admired than an ethic of absolute ends. Nor did he refrain from enlisting his academic expertise in the service of advocacy of policies which he believed to be in the best interests of the German nation. In Marx's case, the link between his analysis of capitalism and his injunctions to proletarians and right-minded bourgeois to hasten the overthrow of a system deservedly doomed to collapse is too familiar to need elaboration. Durkheim, too, was as much a moral didact as he was an academic analyst of the division of labour or suicide rates or primordial Australian religion. All three would have been as nonplussed by the rise of moral relativism in the late-20^th century as by the decline of teleological theory. The assumption that morality is a matter of ethnographically local conventions between which European sociologists have no special authority to arbitrate would have struck them as eccentric if not perverse. They did not, admittedly, assume that their own precepts would necessarily be followed, or their advice taken, by the governments of their day. In that, they were if anything less confident than their less theoretically-minded English counterparts who, as Laurence Goldman has pointed out, were heirs to a tradition which ‘from its very foundation, was premised on the use of social knowledge in the service of reform’ (2007: 434). But the difference between then and now is neatly captured in the contrast between L.T. Hobhouse's Morals in Evolution of 1906, with its account of gradual but steady progress in the direction of Hobhouse's own liberal convictions, and Richard Joyce's The Evolution of Morality of 2006, with its account of the moral sense as an adaptive trait of the human brain attributable to natural selection.

The Darwinian legacy

Nor can the founders be reproached for failing to foresee that by the beginning of the 21^st century the most pervasive influence on the human behavioural sciences might turn out to be Darwin's. Darwin himself, after all, did not, and could not, realize the magnitude of his achievement, since he did not, and could not, know how evolution through natural selection actually works. It was not until the ‘new synthesis’, as it came to be called, between evolutionary theory and population genetics was supplemented by the discoveries of molecular biology that the objections which he had been unable to answer in his lifetime could be definitively rebutted. Well into the 20^th century, self-styled Mendelians could still be anti-Darwinians. But by the last quarter of the 20^th century, when the ongoing debates among theoretical biologists were all being conducted within an explicitly neo-Darwinian paradigm, social scientists too were beginning to see in Darwin's conception of ‘descent with modification’ the starting-point from which they could best address the explanation of qualitative changes in human behaviour-patterns in disciplines ranging all the way from archaeology (Shennan, 2002) to economics (Hodgson and Knudsen, 2006). Sociologists remained for the most part either ignorant of or hostile to these developments. Marxists no longer saw Darwin, as Marx himself had done, as underwriting the idea of class struggle as the driving force of human history; the acrimonious controversies provoked by the publication in 1975 of E.O. Wilson's Sociobiology convinced many sociologists that neo-Darwinian theory had nothing to offer them; and the remark made in 1959 by John Maynard Smith, not yet known as the father of evolutionary game theory, at a conference organized by the Scottish Branch of the British Sociological Association to the effect that sociologists’ suspicion of the application of biological ideas to sociology might be justified by the ‘nonsense’ written in the name of Social Darwinism and the crimes committed in the cause of racial superiority (Maynard Smith, 1961: 83) continued to hold good for sociologists of all theoretical schools.

That suspicion is at last beginning to be dispelled, even if it is still found necessary on occasion to reaffirm the obvious proposition that ‘it is possible to recognize the importance of the mechanism of natural selection without thereby subscribing to ideas of selective breeding or ethnic cleansing’ (Studholme, Scott, and Husbands, 2007: 818). But the agenda bequeathed to 20^th-century sociology by Marx, Weber, and Durkheim is being revised under the influence of neo-Darwinian theory in two separate, although equally significant, ways: first, by the findings of biologists, geneticists, and psychologists which bear directly on naturally selected aspects of human behaviour; second, by the application of neo-Darwinian theory to the framing and testing of non-reductionist causal hypotheses about cultural and social evolution.

Sociologists may not unreasonably regard the findings of behaviour genetics and behavioural ecology as marginal to their concerns. Behaviour geneticists have done much to clarify and resolve the ‘nature-nurture’ debate in the study of individual development and to account for within-group differences in heritable susceptibility to both physical and psychological conditions of various kinds. But these differences have less to contribute to the explanation of the collective institutional processes which are sociologists’ concern. Similarly, the matching by behavioural ecologists of the reproductive strategies adopted by the populations they study to predictions modelled on maximization of inclusive fitness (eg Borgerhoff Mulder, 2000) have less to contribute to the explanation of the demography of populations of the kind with which sociologists are accustomed to deal, where the pursuit of inclusive fitness is frequently overridden by economic, ideological, and political influences (eg Banks, 1981). On the other hand, the burgeoning discipline of evolutionary psychology has generated an extensive and rapidly growing literature demonstrating how human behaviour-patterns are affected by innate cognitive mechanisms which have evolved through natural selection. Evolutionary psychologists have been rightly criticized for assuming too readily that the design of the human brain was fixed by the selective pressures which bore on our hunting and foraging ancestors in a supposedly uniform Pleistocene environment (Foley, 1996; Irons, 1999). But explanations of aspects of present-day human behaviour by reference to innate species-wide mental traits do not depend on knowing just when and how the design of our brains came to be what it is. Sociologists who feel threatened by the claims made by, or on behalf of, evolutionary psychologists (Rose, 2001; Jackson and Rees, 2007) will have to refute them rather than simply bemoan their influence. They need to show either that the observations which the contributors to the Oxford Handbook (Dunbar and Barrett, 2007) claim to be able to explain are incorrect, or that the proffered explanations fail to fit them. It can hardly be disputed that the human brain is as much a product of natural selection as human intestines and human teeth, and that an understanding of the resulting psychological capacities, dispositions, susceptibilities, and constraints common to all members of the human species bears directly on a whole range of topics with which sociologists are concerned. Nobody has overturned the study by Daly and Wilson (1988) of homicide, which found that in every society for which there is reliable evidence young adult males are overwhelmingly more likely to be the killers than either older men or coeval women. Research in this area has, moreover, been shown to have significant implications for feminist theory, not least because male-female differences in physical aggression are seen at too young an age for cultural stereotyping and gender labelling to account for them (Archer and Cǒte, 2005). It is true that evolutionary psychologists have been too ready to accuse any and all sociologists and anthropologists of being wedded to an outdated ‘Standard Social Science Model’ (Tooby and Cosmides, 1992) which assumes the human mind to be a blank slate on which culture inscribes whatever locally variable instructions it will. But collaboration between cognitive and developmental psychologists and neuroscientists is beginning to demonstrate in detail just how much further from blank the slate is than the majority of 20^th-century anthropologists and sociologists were willing to accept.

At the same time, it is being increasingly recognized that it is not only at the level of natural selection that Darwin's conception of ‘descent with modification’ applies, and that cultural and social evolution come about through a process continuous with and analogous, but not reducible, to it. Awareness that cultural evolution might be modelled as a system of heritable variation and competitive selection operating by different transmission rules from genetic inheritance has given rise to the ‘dual-inheritance theory’ influentially developed by Robert Boyd and Peter J. Richerson (Boyd and Richerson, 1985; Richerson and Boyd, 2005). Dual-inheritance theorists follow Cavalli-Sforza and Feldman (1981) in adapting mathematical models developed by population geneticists to the reproduction and diffusion of information transmitted not from organism to organism in strings of DNA but from mind to mind by imitation and learning. They are careful to distinguish themselves from the self-styled ‘memeticists’ (Blackmore, 1999; Distin, 2005) who have taken from Richard Dawkins the notion of particulate units of cultural inheritance which are replicated by analogy with genes and diffused by analogy with viruses. But the dual-inheritance theorists share with the memeticists an explicit recognition that cultural selection can work independently of natural selection and can either enhance or diminish (or, it may be, have no effect at all on) the inclusive reproductive fitness of the carriers of locally successful cultural mutations. Richerson's and Boyd's reluctance to adopt the term ‘meme’ leads them to talk in terms of ‘gene-culture’ interaction when a more logical formulation would be either ‘gene-meme’ or ‘nature-culture’ interaction. But they have produced not only deductive arguments but supporting empirical evidence to show that what they prefer to call ‘cultural variants’ evolve in recognizably Darwinian fashion. This would have been equally unacceptable to Marx, for whom culture was a superstructural reflection of underlying social relations of production, to Durkheim, who would have rejected outright the suggestion that cultural evolution comes about through the aggregation of individual psychological traits at population level, and to Weber, who explicitly considered the application of the concepts of selection (Auslese) and adaptation (Anpassung) to sociology only then to reject them. But 21^st-century sociologists have no reason not to welcome both dual-inheritance theory and evolutionary psychology for what they can contribute to topics of traditional sociological concern.

Natural, cultural, social

All three founders had their own preferred just-so stories of social evolution: Marx's of dialectical class conflict, Weber's of formal and substantive rationalization, and Durkheim's of the transition from mechnical to organic solidarity. But these are very different from neo-Darwinian stories of the path-dependent but open-ended evolution of societies from one to another kind through the heritable variation and competitive selection of information defining practices constitutive of institutional roles. Nor did any of the three distinguish explicitly between cultural evolution, in which change in collective behaviour-patterns comes about through the transmission of novel beliefs and attitudes from one person's mind to another's, and social evolution, in which it comes about through the institutional imposition of formal legal or customary practices as opposed to the interpersonal acquisition of informal habits or routines. In this they were at one with 20^th-century anthropologists who, as pointed out by Brown (1991: 40), have been particularly prone not only to contrast ‘cultural’ and ‘social’ jointly with ‘biological’ but to treat ‘a culture’ and ‘a society’ as synonymous. But sociologists studying the armies, churches, markets, businesses, bureaucracies, law-courts, and parliaments of large and complex societies cannot but be implicitly, if not explicitly, aware of a difference between behaviour controlled by formal institutional sanctions and behaviour which is the acting-out of beliefs and attitudes transmitted from person to person by imitation and learning. Indeed, the relation between the two is so integral to the explanation from almost any theoretical perspective of the behaviour of the members of collectivities like these that it may be thought by sociologists not to need any particular attention drawn to it in the course of normal research.

Until now, both evolutionary psychologists and dual-inheritance theorists have, like the anthropologists, made only the single distinction between the biological on one side and the ‘sociocultural’ on the other. They are not unaware of institutional as opposed to personal differences of power in human societies, but treat them as externalities diverting cultural evolution in a different direction from that in which it would otherwise proceed (Durham, 1991:443). This is partly because they tend to look for their empirical evidence either to psychological experiments or to small-scale inter-group comparisons, and partly because, like the behavioural ecologists, they tend to select for study in the field populations of hunter-gatherers or pastoralists rather than of large industrial societies. But there is no reason not to extend the same approach to large populations whose members are the incumbents of institutional roles to which there attach formal economic, ideological, and political sanctions. The way in which the practice of wage-labour, for example, can invade populations in which barter and customary exchange had previously appeared to have reached fixation is obviously analogous, but equally obviously not reducible, to the way in which a novel fashion in clothes or hairstyle or cookery is diffused across a population by imitation or learning. Imitation and learning still remain a part of the process: in an emerging capitalist economy, budding entrepreneurs will deliberately set themselves to imitate and learn from the practices of those of their competitors whom they see as the most successful. But, as James S. Coleman pointed out in discussing ‘authority relations’, we are now dealing with ‘an acting unit consisting of two individuals [in this example, the capitalist and the wage-labourer] in place of two separate and independent units’ (1990:146). Institutional role-dyads of this kind are as much the outcome of the continuous process of heritable variation and competitive selection as the distinctive mores and life-styles which are generated in the course of cultural evolution. But something quite different is going on in the two cases when a new behaviour-pattern displaces an old one in the population concerned.

When the neo-Darwinian approach is brought to bear in this way on the traditional sociological agenda, it may seem to imply that practices should be taken to be ‘the’ units of social selection, memes ‘the’ units of cultural selection, and genes ‘the’ units of natural selection. But genes are not the only candidate in biological evolution, where natural selection depends on other mechanisms besides the replication of DNA and ‘multi-level’ selection is recognized as commonplace. Sociologists can, and should, be open-minded about where selective pressure may be at work down the causal chain from the primary units of information, whatever they are, to their extended phenotypic effects. The benefit to be gained from such open-mindedness can be clearly seen in the study of the evolution of languages, where phonemes or syllables or words or even grammars can be taken to be the heritably variable objects which have outcompeted their rivals. There is nothing inherently misconceived in treating either ‘core traditions’ or ‘ephemeral assemblages of small units’ (Boyd et al., 1997: 365) as what cultural selection selects, or treating roles or even institutions as what social selection selects, if that will better explain what is going on in the culture or society concerned.

It is important also to dissociate neo-Darwinian theory from the imputation of genetic, cultural, or social determinism. Evolutionary biologists are well aware of the phenomenon of ‘niche construction’ whereby organisms themselves modify their environment and thereby affect the selective pressures acting on them. At the cultural level, memes are actively reinterpreted and reconstructed in the course of transmission from mind to mind. At the social level, the practices defining institutional roles are constantly being renegotiated within relationships of economic, ideological, and political power. The practical difficulty which this poses for comparative sociology is not that of identifying these ongoing reciprocal interactions but that of estimating the extent of their relative influence on subsequent population-level outcomes as biological, cultural, and social evolution proceed.

An illustrative example

It is still early days in the application of neo-Darwinian theory to the traditional sociological agenda, and any sociologist who attempts it is likely to be criticized by other sociologists as too Darwinian and by evolutionary psychologists and dual-inheritance theorists as not Darwinian enough. But one recent study, whose authors explicitly cite the principles underlying evolutionary psychology as providing the ‘metatheoretical base’ for research like theirs (Nisbett and Cohen, 1996: xviii), illustrates particularly clearly the value of a neo-Darwinian approach. Nisbett and Cohen's chosen topic is the ‘culture of honor’ in the Southern United States and the lethal interpersonal violence to which it gives rise. Unsurprisingly, the much higher incidence of homicide among young adult males than either older males or coeval females is fully confirmed. But Southern women are much more likely to kill than Northern women, and the consistent readiness of white, non-Hispanic Southern males to respond with lethal violence to personal insults or threats to their families or property is unmistakably attested in the homicide rates disaggregated by Nisbett and Cohen from U.S. Department of Justice records. The regional differences are, moreover, significantly more marked in ‘argument-related’ than ‘felony-related’ homicide. The regional correlations are supported by both survey data and experimental results. There is no evidence which would favour either an ecological or a genetic explanation. White Southerners differ as much from non-white Southerners as they do from white Northerners. Young Southerners acquire from parents, mentors, peer-groups, and role-models a tradition whereby insults must be punished, wrongs avenged, and recourse had to self-defence rather than public protection. This distinctive upbringing derives from the sociological difference that Southerners are descended from herders and Northerners from farmers, and herders all over the world are known to be more disposed to aggressive and retaliatory violence because of their vulnerability to losing the animals on which they depend for their livelihood and which, unlike land, are easy to steal. This therefore gives rise to the perceived need to establish and maintain a reputation for being willing to resort to violence in the face of attempted predation which does not arise in otherwise indistinguishable farming populations.

Sceptics of the neo-Darwinian approach to sociological questions may seek to dismiss this explanation as a just-so story constructed in hindsight. But all evolutionary explanations are just-so stories constructed with hindsight. Sociologists, like biologists, have to find out as best they can which of the alternatives on offer is the right one. Kipling's just-so story of how the elephant got his trunk is a comic answer to a question to which the serious answer is given by the theory of natural selection. Nisbett and Cohen's just-so story of lethal violence and the culture of honour in the American South cannot be tested as conclusively as that of the ecology and evolution of Darwin's finches (Grant, 1999). But that is because, as every practising sociologist well knows, even the panoply of methods deployed by Nisbett and Cohen cannot approximate the quasi-experimental contrasts on which explanations from within the theory of natural selection can be based. It is not because theirs is either trivial or circular.

Notice also how it is a story of selection at the biological and cultural and social levels. The culture of honour is handed down by imitation and learning and sustained by the sanction of personal reputation. But the lethal violence whose incidence is heightened by it depends on evoking an innate capacity differentially inherited by men and women, and it is in part the social-evolutionary product of a mode of production which differentially heightened the probability of cultural reproduction of its constituent memes. Nisbett and Cohen themselves suggest that if their explanation is well-founded, a similar behaviour-pattern is more likely to be found in horticultural societies than among either stable agricultural communities or hunter-gatherers. But they also point to the re-emergence of a sub-culture of the same kind in the environment of inner cities, which exerts strong selective pressure favouring norms of individual self-protection and vigilance against behaviour by others which threatens, or can be seen as threatening, the respect which the individual young male regards as his due.

Conclusion

One possible response to the suggestion that sociologists should be readier than they are to forget the founders might be to argue that the founders’ concerns are still our own. Don't Marx's views about class conflict and the social relations of production, Weber's about power and forms of legitimacy, and Durkheim's about anomie and the division of labour remain as relevant to 21^st- as to 20^th-century sociology? But although the concerns may be the same, the concepts and methods with which to address them are not. However far sociology may be from standing comparison with the advances which biology has made in the last hundred years, we now know very many things that the founders did not, and it implies no disparagement of their achievements to say so. Indeed, it can be argued that Weber, in particular, would have both understood and welcomed the support which evolutionary game theory can give to one aspect of his thesis about the part played by a distinctively Puritan Lebensführung in the evolution of American capitalism (Runciman, 2005).

The continuing resistance among present-day sociologists to neo-Darwinian theory is itself an interesting topic in the sociology of sociology. The overtly political hostility of Marxist critics fearful that it may be deployed in support of reactionary views and policies to which they object can perhaps be dismissed alongside the overtly religious hostility of Christian critics fearful that it may be deployed in support of atheistic views and policies to which they similarly object. But there remains a wilful disposition among neo-Darwinism's critics to attribute to neo-Darwinians opinions which they do not in fact hold. It is simply not true that neo-Darwinians are all covert reductionists who believe that cultural variation is controlled (as opposed to constrained) by the extent of its contribution to the maximization of inclusive reproductive fitness. Nor is it true that they are Panglossian optimists who think that evolution leads to a best of possible worlds as opposed to locally optimal trade-offs. Nor are they ‘pan-selectionists’ who think that every observed biological, culture, or social trait must be explained as an adaptation. Nor do they believe that there are single genes ‘for’ personal characteristics, or single memes ‘for’ life-styles, or single practices ‘for’ modes of production, persuasion, or coercion. Nor is it true that the population-level approach which is central to neo-Darwinian theory ‘systematically disrupts any attempt to understand the generative dynamics of developmental systems’ (Ingold, 2004: 219). The study of development and of adaptation are complementary, not mutually exclusive, and there is good reason to believe both that developmental plasticity is an adaptation and that some important elements of it are adaptive (Sterelny, 2003: 166). There is no lack of unresolved questions within the study of natural selection itself, let alone the study of its interaction with the concomitant processes of cultural and social selection. But polemical stereotyping of opponents’ views is not helpful to the development of research designs by which prospective answers to the unresolved questions can be formulated and put to the test.

For sociologists of the early 21^st century, the need is to recognize the extent to which neo-Darwinian theory, far from undermining their findings about collective human behaviour, can help to underwrite them. This implies no threat to the autonomy of sociology as an academic discipline. Study of the behaviour-patterns which distinguish different kinds of communities, institutions, cultures, and societies from one another involves both concepts and methods quite distinct from those involved in the study of individual human minds and bodies. But the increased understanding of the workings of the human mind which is emerging from current palaeoanthropological as well as psychological research holds out the prospect of grounding the findings of sociologists on a basis far more solid than Marx, Weber, or Durkheim could have conceived of. Then, perhaps, it may become possible to resolve what might be called Gellner's Paradox: ‘Question: how can a species, genetically granted by Nature such remarkable freedom and licence, nevertheless observe such restraint, such narrowly defined limits, in its actual conduct?’ (Gellner, 1989: 516).