The strange survival and apparent resurgence of sociobiology

What is sociobiology?

Sociobiology is the ‘systematic study of the biological basis of all social behaviour’ (Wilson, 2000: 4), including behaviour occurring between interacting organisms of the same species, between parasites and hosts, and between predators and prey. It places a particular emphasis on the relationships between behaviour and evolutionary trends: behaviours are understood in relation to the question ‘How does doing this promote fertility, breeding, longevity, and so on?’ Animal behaviour is construed as the means by which genetic information is transmitted: animals (phenotypes) are the vehicles through which particular DNA ‘codes’ (genotypes) are transmitted, and their activities are therefore understood in relation to the ways in which they promote the spread of particular genes throughout a population. Genes, in turn – because they determine what behaviours will or will not be exhibited – are more or less ‘successful’ depending on whether the traits they control for make the organism more successful reproductively, thus passing those traits on to more offspring. Such genetically determined traits will spread through a population more than those which fail to promote reproductive ‘success’. In short, sociobiology makes the theoretical assumption that all behaviour is genetically determined in the last instance.

This assumption, however, opens up two problems. First, some highly social creatures do not breed, and therefore pass on their genes, at all. Female naked mole rats, for instance, are sterile throughout their lives unless their colony queen, which does reproduce, dies. In this event one or more may become fertile, leading to conflict and competition between those that do. This eusocial behaviour was noted by Darwin as being ‘one special difficulty, which at first appeared to me insuperable, and actually fatal to my whole theory’ (Darwin, 2008: 75). ² The discovery of DNA, and an increasing research consensus around the transmission of genetic material as the key driver behind evolution, rendered this ‘difficulty’ even more problematic.

Second, some – sometimes quite a lot of – animal behaviour does not seem to be self-interested in this way at all. Some animals, for instance, place themselves in jeopardy to protect one another, as when meerkats act as ‘lookouts’ for the rest of their colony by watching for predators instead of feeding, grooming, mating, and so on. Lives are sacrificed to save those of others, as when bees sting animals that pose a threat to their colony. How can such behaviour be in the interest of the animal exhibiting it?

Dawkins (2006a[1976]) sought to address these difficulties by redefining the meaning of the word ‘interest’. Rather than the organism having particular needs, requirements, preferences, and so on, ‘interests’ were defined in genetic terms. Genes, rather than organisms, seek to replicate themselves, and animals are the vehicles through which that replication occurs: animal behaviour, in short, becomes a by-product of the imperatives of genes. Both problems – eusociality and altruism – disappear if they can be shown to promote the interests of genes over and above the apparent requirements of their bearer organisms.

Dawkins’s work built on the work of W. D. Hamilton (1964a, 1964b), who provided the first modern (post-DNA) theoretical model of a ‘limited restraint on selfish competitive behaviour and possibility of limited self-sacrifices’ (Hamilton, 1964a: 1). Hamilton sought to comprehensively quantify the relationships between genetic benefit and social behaviour. This relationship was not limited just to eusocial or altruistic behaviour – indeed, it could not be, as the division of activities into categories such as ‘altruistic’ and ‘selfish’ is alien to biology – but rather to any interactions between related organisms. ‘Hamilton’s rule’, as interpreted by Wilson (2000: 3) can be summarised as rb > c (behaviour is exhibited where the relatedness of the participants, r, multiplied by the benefit to the recipient, b, is greater than the cost, c, to the acting organism). The implications of this are clear:

This means that for a hereditary tendency to perform an action of this kind to evolve the benefit to a sib[ling] must average at least twice the loss to the individual, the benefit to a half-sib must be at least four times the loss, to a cousin eight times and so on. To express the matter more vividly, in the world of our model organisms, whose behaviour is determined strictly by genotype, we expect to find that no one is prepared to sacrifice his life for any single person but that everyone will sacrifice it when he can thereby save more than two brothers, or four half-brothers, or eight first cousins. (Hamilton, 1964a: 16)

Hamilton’s rule provides for three types of social behaviour: altruism, selfishness and spite (Wilson, 2000: 118–9). Altruism accounts for the example of ants given above, while selfishness is the more routine pursuit of genetic self-interest. Spite accounts for those situations in which an individual deliberately reduces the fitness of an unrelated competitor without any personal gain (even sometimes incurring a personal loss) in order to increase the fitness of one or more closely related others. Social stability is handled with the concept of the evolutionary stable strategy, derived from game theory. ³ In an altruistic community, any one selfish organism would be able to accrue disproportionate advantages to itself. This, however, would lead to the gene for that selfish behaviour spreading throughout the population – as the selfish organism would breed more than its altruistic counterparts – reducing the proportion of altruistic to selfish actors. If left unchecked, selfishness would thus become the predominant mode of behaviour, leading, in turn, to small numbers of mutually co-operating altruistic individuals thereby accruing disproportionate advantages to themselves. Groups would potentially oscillate between altruism and selfishness, with the minority behaviour tending to be most advantageous to those organisms that exhibit it. In fact, according to Dawkins (2006a: 69), such oscillations tend towards stability.

Wilson and Dawkins together provided a comprehensive theory of social behaviour, based on Hamilton’s rule, which accounts for why individuals do what they do, motivated by the demands of their genes – and evolutionary stable strategies – to account for social stability and to provide social constraints on individual actions. ⁴

Sociobiology’s critics

Sociobiology was never uncontroversial: both Dawkins and Wilson argued that its tenets could account for all forms of animal behaviour, including human actions. ⁵ Its survival as a coherent perspective depended on its capacity to maintain this claim. Over the course of the late 1970s and early 1980s this claim was systematically demolished – or so it seemed. In fact, the early critics of sociobiological explanation failed to persuade Wilson, Dawkins or their followers that their perspective did not work. For the purposes of brevity four lines of criticism will be outlined: political, natural historical, anthropological and philosophical. ⁶

Sociobiology’s political critics focused on its congruence with free-market economics and Hobbesian self-interest. If all behaviour is ultimately selfish in motivation there are biologically driven limits on the possibilities for co-operative or progressive social change. Wilson’s work in particular was construed as just another attempt to ‘prove’ that such social change was impossible, making inequality and social conflict out to be ‘natural’ states of affairs. The most comprehensive criticism of this sort was collected in the book Not in Our Genes, written by the biologist Steven Rose, the psychologist Leon Kamin and the geneticist Richard Lewontin (Rose et al., 1984). They question the scientific precepts of sociobiology from a range of perspectives and find it wanting in all, concluding that its explanations depend on reductionism and ‘vulgar’ understandings of both genetics and evolutionary science:

Sociobiology is yet another attempt to put a natural scientific foundation under Adam Smith. It combines vulgar Mendelism, vulgar Darwinism, and vulgar reductionism in the service of the status quo. (Rose et al., 1984: 264)

Gould’s broader critique was that sociobiological ‘explanation’ depended on an inversion of observation and theory, whereby the latter was used to ‘make sense’ of the former rather than being derived from it. He argued that sociobiology was a form of ‘just-so’ explanation, little different to how fundamentalists invoke the concept of God:

When evolutionists study individual adaptations, when they try to explain form and behaviour by reconstructing history and assessing current utility, they also tell just-so stories – and the agent is natural selection. Virtuosity in invention replaces testability as the criterion for acceptance. (Gould, 1978: 530)

How sociobiologists handled these claims would determine whether the perspective could survive, and what explanatory form it would take.

Sociobiology’s strange survival

Sociobiological explanation still has a role in mainstream biology, which has allowed its adherents to claim it retains empirical validity. The fact that some animal behaviour can, very plausibly, be understood as promoting the selfish replication of genes does a lot of work. The behaviour of eusocial insects has been central to this line of argument: self-sacrifice among worker ants and the organisation of fertility and reproduction in an ant colony are perspicuous examples of where ‘gene-selfish’ accounts of behaviour provide a best fit to the empirical data. Thus, for example, queen leaf-cutter ants of the species Acromyrmex echinatior mate with multiple males, while genetically similar ‘social parasite’ queen ants of the species Acromyrmex insinuator have reverted to monogamy (Sumner et al., 2004). This behaviour can readily be accounted for using Hamilton’s rule: the benefits of polyandry –such as greater genetic diversity among sterile workers and the production of genetically varied daughter queens – are specific to genuinely social insects. Where insects are ‘socially parasitic’ (i.e. they take advantage of living in a social host colony but produce only sexually active offspring that take flight on reaching maturity) such benefits are less important, and so monogamous mating behaviour becomes more useful. Similar arguments have been used to account for, for example, the mating rituals of birds (Cronin, 1991).

These kinds of argument are relatively benign within natural history: Hamilton’s rule is used as a rule of thumb, to see in each case if apparently strange behaviour can be accounted for on the basis of genetic determinism. In some cases, as with the leaf-cutter parasite ants, a strong case can be made; in others (e.g. Barash’s mountain bluebirds) claims were weaker. Sociobiologists were always, however, able to claim that these ‘weaker’ claims could be strengthened as further data were obtained: in each case, the reason observed behaviour and gene-selfish theory could not readily be brought into alignment with one another was because more data were required to account for the behaviour. This line of argument originally derived from Hamilton himself, who argued that his model could account for Emperor penguins ‘fostering’ others’ eggs when their own chicks failed to survive incubation:

As regards the already mentioned fostering passion shown by Emperor Penguins that have lost their chicks, some doubt as to whether the observations have been correctly interpreted would seem to remain.…But taking the statements at their face value we might suggest for instance, that it has something to do with heat-conservation. Perhaps the parent penguin is so closely adapted to living with its offspring that it is, at the stage in question, at a positive disadvantage without a chick nestling in the brood-pouch. But such a situation would hardly come into being unless there were strong general relationship within the flock. We seem to need to postulate this in any case to explain some other social behaviour of penguins, for example, the way Adelie Penguins parents are said to leave their young in the care of only a few adults while they go off on long fishing expeditions. On the other hand, some apparently social behaviour such as the formation of the crêche in severe weather is easily interpretable as being almost entirely selfish. (Hamilton, 1964b: 50)

Second, sociobiology needed to retain its claim to be able to account for human as well as animal behaviour. To allow one species to be exempt from genetic determinism would undermine the project as a whole, as such an exemption would complicate the explanatory system (as with Dawkins’s memes) and potentially open the door to other exemptions. Accounting for human behaviour in Darwinist terms was therefore an ongoing priority, and was addressed in two ways: by reframing understandings of cognition and brain biology in evolutionary terms, and by advancing a ‘scientistic’ approach to the ‘irrational’.

Steven Pinker’s work on language acquisition is central to the first of these lines of attack. Pinker (1994) argued that language is not ‘learned’ but rather ‘acquired’ at a key stage in child development, during which humans are uniquely capable of recognising the linguistic rules of their community. No other animals have this capacity, and a failure to acquire linguistic skills at this stage of development means they will struggle to do so later on. Pinker believed that this capacity was hard-wired into the brain, in the form of a ‘meta-grammar’ capable of picking up the specific grammars of the community the child is a part of. He further argued that this part of the brain, and its ‘activation’ at a specific developmental phase, was driven by evolution: the capacity for language was selected as a means of solving the problems of communication in early hunter–gatherer societies.

These ideas are central for three reasons. First, Pinker locates complex human behaviour as something evolutionarily adaptive: it is not ‘social’ in the first instance but ‘selected’ because it – at some stage in the past – maximised the chances of survival as humans (and their social activities) evolved. Second, he makes behaviour like speech out to be organised in parallel with brain development and activity: what people do becomes the manifestation of neural activity, and being able to describe that neural activity will come to account for the behaviour in total. Third, he chooses language as his central area of investigation: if this quintessentially social and learned activity can be shown to be evolved, biologically based and innate, it will be far easier to bring other human activities under the auspices of Darwinist explanation.

Pinker’s work is also a key resource for the second sociobiological approach to human behaviour: a faith in ‘progress’. Pinker (2011) argues that there has been a steady decline in violence among humans, as ‘civilising’ forces have become stronger and more stable. This parallels the work of Wilson (1998), on the synthesis of the social sciences and humanities (under the auspices of evolutionary biology), and Dennett (1995), on the self-interested evolution of morality and ethics. The form of ‘progress’ advanced by these writers is one that typically points towards scientific rationality as the high-water mark of human evolution – particularly in its Darwinist form, of course – and is strongly critical of contemporary manifestations of ‘irrationality’, particularly postmodernism and other forms of relativist thought. The strongly secular bent of these thinkers is revealed in the anti-religious arguments of Wilson (Sarchet, 2015), Dawkins (2006b) and others.

The sociobiological account of human behaviour, then, has relied on two lines of attack: arguing that evolutionary adaptation accounts for complex social activities that appear to be learned, and reframing human history as the triumph of reason, science and order over irrationality and violence – where sociobiological explanations themselves are the key exemplars of reasonable, scientific and orderly thought.

Finally, sociobiology needed to show that genetic regularities were responsible for particular forms of behaviour. Similarly to how Pinker argued that neural development and brain structure underpinned language use, sociobiologists sought to argue that repetitive and characteristic behaviour must be the manifestation of the actions of genes which ‘code’ for that behaviour. As Dawkins (replying to Midgley) argued:

I am searching for a chunk of chromosomal material which, in practice, behaves as a unit for long enough to be naturally selected at the expense of another such fuzzy unit. I agree that there are difficulties in this way of looking at evolution, but I believe I have shown them to be less great than the difficulties inherent in any other way that has been suggested. (Dawkins, 1981: 569)

Put together, these are remarkably thin justifications for saying that sociobiology ‘survived’ its critics. Apart from (some of) the behaviour of eusocial insects, the perspective came to rely on a series of promissory notes: that natural historical evidence would start to reveal the evolutionary bases of other behaviour in other animals, that neurobiology would start to provide data showing the neural bases of complex human behaviour, that secular and rational social progress would continue across the world, and that the developing science of genomics would start to reveal genes ‘for’ behaviours in the same way it should show how there are genes for physical traits.

None of these things reliably or uncontroversially came to pass. Natural historians have continued to deepen their understanding of animal behaviour, but this increasing detail has not revealed a common, genetic, form of explanation underpinning them as a whole. Developments in brain scanning have apparently revealed some parallels between emotion and brain activity (e.g. Fisher et al., 2005), but these are controversial and have had little impact on the psychology, philosophy or sociology of emotion. The world is perhaps not getting better – although in evolutionary terms there has not been enough time to determine whether or not that is the case. Finally, most troublingly, genomics has not found genes ‘for’ behaviour. Indeed, as the discipline has developed, increasing numbers of geneticists have cast doubt on the idea that there can be genes ‘for’ anything at all (Barnes and Dupré, 2008: 56).

What these developments reveal is not the resilience of sociobiology but its character as an approach, a ‘correct’ way to study behaviour. It is a flexible and loose rubric for putting together different claims under the auspices of a heuristic-based theory of how the world works. To use Parsons’ (1937: 51) terminology, it is a ‘subsystem (or, perhaps better, an interrelated group of several sub-subsystems) of the theory of action’. It is utilitarianism.

Sociobiology is utilitarianism

Sociobiology has been accused of being a utilitarian perspective before, but this has generally been in the narrow sense of Hobbesian notions of self-interest (Sullivan, 1982). In a rather odd argument, R. R. Sullivan contrasts a ‘Hobbesian’ point of view in which all altruism is necessarily self-serving in the final analysis with a ‘Marxian’ perspective in which ‘altruism is man’s natural orientation’ (ibid.: 274). He concludes that the ‘sociobiological analysis of altruism is clearly Hobbesian and not Marxist’ (ibid.). The defining feature of this argument is its shallowness. Utilitarianism and Marxism are invoked as means of classifying selfishness rather than as means of examining more deeply the logic and grammar of the sociobiological perspective. In this sense, at least, Sullivan’s approach shares a problem common to many of sociobiology’s other critics: his analysis takes place on terrain largely chosen by sociobiology’s adherents. ⁷ It is, however, possible to go deeper by examining sociobiology – in the form it retained – with reference to Parsons’ more analytical description of utilitarianism as a programme.

Parsons’ work is particularly relevant in this regard for three reasons. First, he recognised that utilitarianism as a system of thought moved beyond the narrower arguments of Hobbes and became a way of construing the nature of social life and social action tout court. Rather than treating it as a philosophical perspective, therefore, he examined it as a persuasive and organised theoretical movement. Second, he considered it not from the point of view of the claims its advocates made but in relation to the grammar and logic of its arguments: he viewed utilitarianism as a particular way of looking at things that could be compared with competing ways and found to rest on particular, different, analytical decisions. Finally, and for our purposes, most importantly, Parsons viewed the movement as something with a history of its own. It emerged at a particular time, for particular theoretical and practical reasons, and its decline was the result of both internal contradictions (in particular an increasing concern with troubling residual matters that it was incapable of handling) and external pressures (the growth of positivism and the voluntaristic and normative theories of Durkheim and, in particular, Weber as both developments within and responses to that growth). Using Parsons’ analysis, therefore, allows sociobiology to be treated as a movement, with a particular theoretical grammar, subject to both internal and external pressures.

Parsons (Parsons, 1937: 51–60) defined utilitarianism as the theoretical action system characterised by four features: atomism, rationality, empiricism and randomness of ends. Atomism is ‘a strong tendency to consider mainly the properties of conceptually isolated unit acts and to infer the properties of systems of action only by a process of “direct” generalisation from these’ (ibid.: 52). Both Hamilton’s account of interactional behaviour, and the concept of evolutionary stable strategies, clearly fall under this rubric. The unit of action in sociobiology is that between interacting organisms – evaluated by the relative costs and benefits this interaction will bring each participant, in terms of the chances of preserving and distributing their genetic make-up – rather than, for instance, those actions being understood in terms of their relevance to the costs and benefits they might bring to the group as a whole. The group is no more than the sum of the actions of its members.

Rationality, for Parsons, is a slippier concept, but – in its simplest form – can be stated thus:

[a]ction is rational in so far as it pursues ends possible within the conditions of the situation, and by the means which, among those available to the actor, are intrinsically best adapted to the end for reasons understandable and verifiable by positive empirical science. (Parsons, 1937: 58)

Here, we can see a tension in how sociobiologists have addressed criticisms of their position. Gould’s argument that sociobiology is a ‘just-so’ explanation of behaviour illuminates the problem of behaviour that cannot readily be accounted for within the Hamilton–Dawkins sociobiological system, that is, behaviour that appears not to make sense in terms of the imperatives of selfish genes. Sociobiologists have to come up with explanations for this kind of behaviour – as non-selfish, non-rational acts are corrosive to their project – and have used Hamilton’s rule as their key resource for doing this. The question is whether this explains such acts or explains them away: Gould would argue the latter, as there do not appear to be any natural phenomena that cannot be so accounted for provided sufficient imagination is brought to bear. Hamilton’s rule rests on the creativity of the person applying it, and – because it is deemed to be correct, scientific and rational as an a priori matter – there are no ‘external’ criteria by which its application can be judged from the sociobiological point of view.

The (naïve) empiricism of utilitarianism, for Parsons, consists in its assumption that systems of action are no more than the sum of the acts that they consist in.

This is the simplest and most obvious mode of employment of this conceptual scheme – the assumption, often naïvely made without full realisation of what it implies, that the concrete action systems being studied are simply aggregates of such rational unit acts. (Parsons, 1937: 59)

This brings us to Parsons’ final characteristic feature of utilitarianism: randomness of ends. In many respects this falls out of the empiricism outlined above: to the extent that this empiricism is taken to describe what really happens in social action (as opposed to being an ideal typification, an abstraction, etc.) then there can be no meaningful co-ordination of ends without that co-ordination being in some respects teleological:

If the concrete system be considered as analysable exclusively into rational unit acts it follows that though the conception of action as consisting in the pursuit of ends is fundamental, there is nothing in the theory dealing with the relations of the ends to each other, but only with the character of the means-end relationship…[T]he failure to state anything positive about the relations of ends to each other can then have only one meaning – that there are no significant relations, that is, that ends are random in the statistical sense. (Parsons, 1937: 59)

In sociobiology, interests are defined with regard to ‘strategies’ and against a background of other organisms’ behaviours. Whether or not ‘spiteful’ behaviour serves self-interest depends on, among other things, whether other interacting organisms are also pursuing spiteful strategies. What turns out to be ‘rational self-interest’ can only be determined over time, with the assumption of an evolutionary stable state or strategy and the presumption of perfect knowledge on the parts of each organism. Social stability, then, is built into the system, but only insofar as the idea of co-ordinated ends or shared goals are ruled out a priori.

What Dawkins versus Wilson revealed

To the extent that sociobiology is simply a utilitarian way of redescribing activities in biological terms, its claims to be scientific are central to its survival. As we have seen, these claims have not weathered well. Claims about human behaviour have gained little traction outside the circle of already committed evolutionary determinists; genomics is moving away from the idea that there are genes for particular traits, let alone behaviours; and natural-historical research has not located the kinds of ‘forces’ Hamilton and others claimed ‘underpinned’ animal behaviour except in a restricted number of cases. ⁸

This is why Wilson’s apostasy so scandalised Dawkins. Wilson’s embrace of group selection over strict kin selection represents a move away from utilitarianism and towards a mode of explanation that recognises the importance of groups and their environments. Wilson, in short, has suggested that utilitarian explanations should be replaced by positivist ones. As Parsons points out, this is the logical conclusion of utilitarianism:

the end result, or the ultimate determinant factors are the same, adaptation to conditions through, in the last analysis, the influence of these conditions themselves. Indeed, in the very last analysis even the difference of process disappears, for in so far as the ‘conditions’ ultimately form the sole determinants of action the subjective aspect becomes merely a reflection of these ‘facts’; it is purely epiphenomenal. (Parsons, 1937: 120–1)

The dispute over Wilson’s application of group selection among biologists – the controversy played out in the journal Nature – is strictly about how best to account for the behaviour of insects. Indeed, the original paper concludes ‘[w]e have not addressed the evolution of human social behaviour here, but parallels with the scenarios of animal eusocial evolution exist, and they are, we believe, well worth examining’ (Nowak et al., 2010: 1062). The responses are not about humans at all, but about social insects. ⁹ The controversy is not about sociobiology but rather about how best to account for the ‘one special difficulty’ Darwin had to deal with, in a very restricted range of species. The dispute does not show that sociobiology is alive and well, but rather that Hamilton’s rule may not even apply in its home domain.

Sahlins (1976: 17–18) pointed out that sociobiology’s claims to account for human behaviour depend on the resilience of its theory of kin selection:

This is true for several reasons. One is the significance of kinship in the so-called primitive societies, from which may be inferred its importance throughout the earlier and greater portion of human history…But there is still another issue which makes the problem doubly critical…[i]f kinship is not ordered by individual reproductive success, and if kinship is admittedly central to human social behaviour, then the project of an encompassing sociobiology collapses. (Sahlins, 1976: 17–18)

Wilson may be right. Some biologists have adopted his position. Others do not see what the problem is, as selection can be construed as operating at several levels: close relatedness, environment and group-level adaptation all being relevant factors (Korb and Heinze, 2004). Others, of course, disagree. Dawkins’s review of Wilson’s book in Prospect, however, is not a defence of kin selection in insects but a defence of a way of describing activities tout court, a restatement of Hamilton’s rule and a bewildered objection to how Wilson of all people could doubt its veracity. Dawkins is reasserting utilitarian explanations in the face of a positivist challenge, and Wilson is threatening a further two volumes to follow his Social Conquest of Earth to show how his new (positivist) approach can be extended from ‘where we come from’ through ‘what we are’ and to ‘where we are going’ (Sarchet, 2015: 29). If Dawkins prevails, and the entomological consensus supports Hamilton’s rule, the final scientific strand supporting sociobiological utilitarianism will remain unbroken. If Wilson’s argument wins out, it will be used to make the kind of move from utilitarianism to positivism Parsons located in 19th century social thought, ironically starting around the time Darwin first published his Origin of Species. Whether sociobiology remains an 18th century mode of explanation or moves willy-nilly into the Victorian era, however, its reliance on a way of accounting for (some of) the behaviour of a (very) narrow range of species is now in plain sight.