©Ronald de Sousa

Now published in International Studies in the Philosophy of Science 4(1):271-283 (1990)

ABSTRACT

This paper turns the tables on the criticisms of sociobiology that stem from a sociological perspective: many of those criticisms lack cogency and coherence in such measure as to demand, in their turn, a psycho-sociological explanation rather than a rational justification. This thesis, after a brief exposition of the main ideas of sociobiology, is argued in terms of four of the most prominent complaints made against it. Far from embodying tired prejudices about the psychological and sociological implications of biology, sociobiology actually reverses a number of naive assumptions about the consequences of natural selection. I surmise that what really provokes the critics of sociobiology is a certain philosophical relevance of sociobiology both in the broad sense (the application of natural selection principles to behaviour) and in the narrow sense (the insistence on the centrality of certain mechanisms, such as gene selection). In both cases, taking biology seriously affects our philosophical vision of the nature of human beings. At the deepest level, however, the distinction between the level at which rational criteria apply and those where we must have recourse to psycho-social explanations probably breaks down.

I.  Introduction: the Attraction and Repulsion of Sociobiology.

Loathing is not too strong a word. There has been a litany of invective directed against sociobiology. Let me begin, to give you a little of the flavour of the controversy, by quoting the defensive protestation which the editors of a recent book on applied sociobiology (Reynolds et al., 1987) feel called upon to make in their introduction to a recent book of essays in applied sociobiology:

Sociobiology is not a politically biased form of pseudo-science. It is not a movement of the far right--or of any other political position. It is not politics dressed up as science. It is not naive science playing into the hands of willing political ideologues.... (p. xv)

A field in which such declarations need to be made at the outset is obviously a field ideologically beleaguered. And in fact sociobiologists have been accused of many intellectual and political crimes, not least of being ideologues whose biological views are actually mere images of their (right wing) political views. Their application of those biological ideas are therefore politically vicious as well as viciously circular.

Here is another quotation, this time from Philip Kitcher, whose relatively moderate and extremely thorough critique of sociobiology is contained in a book over four hundred pages long:

Even in the mollifying terms of [E.O. Wilson’s] On Human Nature, we are told that there may be ‘unmeasurable costs’ involved in trying to implement certain ideas of social justice..... Pop sociobiology... revives that familiar flop, the tawdry drama of genetic determinism. (Kitcher, 1985, p. 21).

I must begin by acknowledging that many proponents of sociobiology as applied to human beings have indeed earned the charges of sloppy thinking and hasty leaps to socio-political dogmas. Documentation for this will not be rehearsed here, since it is easily found in Kitcher’s book, as well as in many of the feminist critics of sociobiology such as Bleier (1984), Hubbard and Lowe (1979), Longino (1989).

Moreover, I am aware that to comment favourably on sociobiology may seem especially irresponsible at this time and place. I write this at a dramatic moment of history in which some of the most ambitious experiments in ‘trying to implement certain ideas of social justice’ are being ignominiously scrapped. In this contest, the claim that there were biological reasons for the failure of those experiments is bound to be pressed with renewed vigour. There is too much obscene gloating in America today, and, in Eastern Europe’s now formerly socialist countries, too ready an assumption that everything that stems from the ideals of equality, social justice, or public welfare is inseparable from tyranny. The temptation will prove irresistible to many of the cruder exponents of sociobiology to shout ‘We told you so.’ (See Masters, 1985). In this mood, hasty conclusions from half baked arguments may seem good enough, and biology may once again be dragged in to justify ideological bias on the grounds of biological destiny.

Some might find in all this political reason enough to join the chorus against sociobiology, but for my part I find that unsatisfying. I want to insist that the real source of tyranny lies less in any particular doctrine that in the intellectual climate which condemns an intellectual entreprise for extra-intellectual reasons. The failure of Stalinist versions of ‘socialism’ bears only remotely on the validity of the ideals of socialism; just so, the brash sins of particular exponents of sociobiology has little bearing on the scientific worth of the research programme of sociobiology.

My method will be to examine some of the best known charges against sociobiology as a research programme. I will try to show specifically, with the aid of a couple of specific examples, that sociobiology tends to undermine rather than to confirm naive presumptions about the bearing of Darwinian biology on our ideas of social and psychological life. I hope to make plausible the view that apart from exaggerated claims by particular exponents of sociobiological ideas, there is little in the ideas themselves to justify the intensity of the attacks to which they have been subjected: their virulence, then, requires just what they accuse sociobiology of being driven by: extra-scientific motivation.

But before we come to the controversy itself, it might be useful to have a reminder of what Sociobiology purports to be.

II.  What is sociobiology?

In the narrow sense, sociobiology is a specific set of doctrines, based on gene selectionism, or at least rejection of group selection. In this sense its originator is not so much E.O. Wilson (1975) as George C. Williams (1966). In this more specific sense, sociobiology has been associated with a number of technical concepts, such as inclusive fitness, evolutionarily stable strategies (ESS), parental investment, and so on. To give the flavour of this work to anyone not already familiar with it, I will sketch two of these concepts: inclusive fitness, and ESS.

Most famously, sociobiology has been concerned with the solution of the problem of altruism, in terms of W.D. Hamilton’s (1964) ‘inclusive fitness’. Very briefly, the problem of altruism arises from a superficial reflection on the idea of natural selection: if the fittest are those that leave most genes behind, then surely natural selection will always favour selfishness, defined as action that promotes one’s own survival and reproductive success. So how could altruism ever arise?

Hamilton’s solution is quickly apparent when we reflect more carefully on the mechanisms that are supposed to cause the problem in the first place. What needs to be preserved are not individuals but genes of certain types. But since all sexually reproducing diploid organisms share half their genes with parent, sibling, or offspring, that gene type will be advantaged, if, roughly speaking, any of those organisms sacrifices itself for any two or more of the others. Four cousins, or eight cousins once removed, can also be worth a sacrifice. Moreover, since these calculations are in any case probabilistic, ultimate sacrifices are not necessary. It suffices to increase my risk by a factor of f, if it results in diminishing the risk to my only sibling by a factor of at least 2f, or to a lone cousin by a factor of at least 4f, and so forth. The same will be true mutatis mutandis for sacrifices made on behalf of other degrees of relationship weighted by the probability of the risk. This means in effect that there is some degree of probability of benefit that will rationally justify sacrifices of any organism for any other of the same species (and even beyond), since we are all in some degree related.

This evidently speaks to the powerful tendency of parents to take risks on behalf of their children. It also allows us to predict (when we look at the probabilities a little more minutely) both that children will also sacrifice themselves for their parents and that they will do so less than the other way around. It may still seem inadequate, however, to explain acts of altruism that extend beyond the small circle of those who have a substantial interest in the altruist’s genes.

But the logic of this idea, it turns out, can be deepened again. Robin Russell (1987) has pointed out that what matters about kinship is not kinship as such, but the fact that it acts as an indicator of the probability that the other organism will have a copy of my genes.

Relatedness is simply a clue to the likelihood of a potential recipient possessing the same genes as the altruist. The information gained by knowing the degree of relatedness is only probabilistic. Furthermore, there are other potential sources of information about the genetic constitution of potential recipients. To take an obvious example, one may observe whether or not they behave in an altruistic fashion. Thus, the emphasis on kin selection in current evolutionary theory may be misleading. (Russell, 1987, p. 123)

Russell’s speculation is supported by some computer simulations, which seem to be akin to the findings about ‘tit-for-tat’ reported by Axelrod (1984).

The notion of an evolutionarily stable strategy (ESS) (Maynard Smith, 1978) represents a second way in which the apparently obvious consequence of Darwinism may turn out to be false. The apparently obvious here is this: if a trait is adaptive, it will spread, and ultimately one will find it (almost) universally distributed among the population. Thus we all have a mouth, two eyes, etc.--except for the odd thalidomide or radiation induced mutant whose difference can surely not be said to be adaptive.

But from a more careful consideration of gene selectionism one can again see that this is a mistake. The advantage conferred by a particular gene G exists only in the context of other genes in the gene pool. As it spreads, the advantage G confers may wane, until an equilibrium is reached, which could be anywhere between the point where the percentage of G in the population is 0 and the point where it is 100. Richard Dawkins’s (1976) example involving ‘Doves and Hawks’ makes the point clear: on the basis of a plausible set of assumptions about the costs and benefits of fighting a hawk to the death, wasting time in harmless confrontation, or just giving in, and on the basis of the probability of having to do one or the other of these things if one is in a population where the proportion of Hawks to Doves is P, we can infer that the ratio of the two in the population will stabilize somewhere without allowing either type to spread in the whole population.

Assume:

     1 When Dove meets Dove, each will have .5 probability of intimidating
the other; the winner will gain 20U (utiles), the loser will lose 10U.  The
net expected utility for Dove Dove interaction Udd is therefore 5U.

3 When Dove meets Hawk, the hawk will win always (U_hd=20U) the Dove always withdraw, so lose, but without sustaining any loss (U_dh=0U)

Now the probability of encountering a Dove or a Hawk depends upon their frequency in the environment. In an all dove environment, Hawks will invade, as they will all contests without risk. But the converse is also true: in an all Hawk environment, Doves will invade, because although they lose all contests, most contests will be between Hawk and Hawk, and the expected utility of those is lower than zero.

We can easily see, therefore, that if p is the probability of encountering a Hawk, (and hence 1-p, the probability of encountering a Dove), an equilibrium will be reached when the expected utility of being a Hawk U(h)is equal to that of being a Dove U(d).

But U(h) = U_hh x p + U_hd x (1-p) = p x -40 + (1-p) x 20

U(d) = U_dh x p + U_dp x (1-p) = p x -0 + (1-p) x 5

Solving, for this particular value of the parameters, we would expect the ratio of Hawk to Dove to be 3:8.

III. Rationalism and Sociologism.

The last decade or two have seen a number of ‘sociological’ objections to rationalistic history of science. To take account of some of these objections, some rationalists have suggested that explanations of science must involve two components: one focusing on the rationality of science, the other on the irrationality of science. Only in the second case do we properly require a sociological explanation external to the internal logical of science itself (Brown, 1989). I agree with this view, but I must admit that it is a lot more difficult to apply this insight to contemporary science, as opposed to the past history of science. It exposes one to something akin to Moore’s paradox of the person who affirms ‘p, but I don’t believe that p’: we must make separate judgments of what rationality demands of us, and of what social forces are causing us to think. It isn’t quite Moore’s paradox, since it is not a priori impossible to acquire the necessary self-conscious perspective. But one’s efforts are obviously threatened by self-deception. In the case of sociobiology, this means that there may well be motivations and social pressures of which we are not at the moment aware (though some are all too obvious).

I don’t wish to claim that the this modified rationalist strategy is definitely the best for the historian of science. As philosophers, however, I believe it would be disingenuous to ignore the distinction just made between rational and irrational episodes in science. I believe this not on the ground that I have some sort of absolute criterion of rationality (though I believe the relativity of criteria of rationality has been much exaggerated of late) but because in science as in life I can only tell whether to look for ulterior motives by making a preliminary judgment as to whether the advertised reasons are adequate to support the proffered conclusions. It is only a preliminary judgment, because only the competent scientist can ultimately decide on the adequacy of the scientific arguments. But on the basis of my preliminary judgment, I must decide whether I can legitimately get involved, as a philosopher-sociologist, in explaining why the scientists believe what they do, and that, it seems to me, requires that I first ascribe a measure of irrationality to the working scientist.

In the present case, I hope to lead you to agree that the discussion has indeed got off the scientific rails, and that the reason for this, in a word, is none other than the fact that sociobiology is philosophicallyinteresting, in a way that I shall explain.

The charges against human sociobiology have fallen mainly into four classes:

(i) the ‘spectre of biological determinism’;

(ii) the charge of anthropomorphism;

(iii) the alleged impossibility of establishing the values of the parameters involved; and, in a different vein,

(iv) the complaint that one should not do sociobiology even if it were true because it is likely to have a bad effect on society.

A brief look at the first three of these charges, I will argue, suggests that whatever can be said in favour of them does not justify the opprobrium poured onto the whole research programme. This will be the task of the next section. The fourth charge will call for a somewhat more extended discussion, which will be the topic of section 5.

IV.  Three Standard Charges against Sociobiology.

Those (on both sides of the dispute) who characterize sociobiology as a thesis about biological determinism omit this crucial consideration (Jacob, 1977): evolutionary change always represents some sort of departure from something that already exists, and all such departures involve some sort of cost. This has at least two consequences. One is that the we can expect a certain number of Rube Goldbergish biological mechanisms: devices that might have been designed more simply had they been designed, as it were, from scratch. The other is that whatever determinism is involved in genetic causation is necessarily going to be only one factor in a complex net of causation in which the environment must play a crucial part. This is quite obvious when you think about it, yet it’s quite incompatible with the charge of ‘genetic determinism.’ The point is that to program something to happen regardless of environment is impossible. Insofar as genes can indeed be said to program anything, they inevitably require that certain material conditions for the execution of the program be supplied: minimally, energy and matter. More concretely it means the degree to which the environment is relied on to provide the necessarily conditions will always be a result of a certain trade-off between the advantages of wiring in complex behaviour and the evolutionary cost of doing so.

Donald Symons (1979) provides a neat example. In the mallard duck, the requirement that individuals recognize its conspecifics is apparently satisfied differently in the male from the female: in the female, what is programmed is the information adequate to recognize the male without learning. In the male, on the other hand, what is programmed is only the disposition to learn to distinguish conspecifics by imprinting. The reason seems to be that males are more easily recognizable, and therefore releasers can be coarser and their programming simpler. (One could test this explanation, though I don’t know whether this has ever actually been tried, by seeing whether it is easier to fool a female than a male with roughly designed decoys.)

The whole idea of genetic determinism is therefore entirely bogus. So it becomes, in accordance with Brown’s principle, particularly interesting to ask what extra-rational reasons people might have had for focusing so heavily on this issue.

(ii) I shall consider the second charge, of anthropomorphism, in just one specific form: the complaint that sociobiology uses models from economics (and, of course, specifically from capitalist economics) to describe the biological world; it then imports back those same ideas to try and show that capitalist economics conforms to the natural order. (Montagu 1980, Sahlins 1976)

In fact, however, economic models apply more directly to biology than they do to the standard subject matter of economics (i.e. people considered as rational agents). The reason is that when economics is applied to its usual domain, a number of dubious psychological assumptions need to be made about the ‘rational self-interested economic agent.’ No such assumptions need to be made when the models are applied to biology, since we can find, in frequency of a gene in the long run, a trivial (and literal) interpretation of the basic terms of both advantage and probability presupposed by economic theory. The probability of this gene’s turning up in the successor population can simply be taken as the actual frequency (in some run considered long enough) of the gene, and the benefit can be interpreted as the difference between this frequency and the corresponding future frequency of its alleles (or some other acceptable measure of fitness). So this second line of criticism, which stems from within a perspective of sociology of science, is quite without force. It is without force precisely because it is a sociological criticism, in that it looks only at the intellectual origin of the model instead of looking at how the model actually fits the biological case.

(iii) The third form of criticism I shall consider cannot really be adequately discussed without technical discussions of individual instances; but the general idea can be summarized easily enough. It is the charge, pressed most fully in Kitcher’s (1985) that for any given claim about the likely consequences of the principles of natural selection, (say, on sexual differences, homosexuality, incest, infanticide, etc.) the values of the crucial parameters cannot be determined independently of the conclusions that the sociobiologist wants to establish. If that is so, then all sociobiological arguments are strictly vacuous.

But if that is so, surely it applies as well to animal as to human sociobiology: for there is nothing is this argument that relates uniquely to the human case. Besides, it is hard to see how one could be sure in general that any investigation undertaken as a part of the sociobiology research programme would suffer from this defect. There are surely some cases in which a plausible circumstantial case could be built in favour of assigning a certain range of values to a given parameter, if only enough to obtain a qualitative prediction, clashing with those derived from rival hypotheses, which might or might not be borne out by the evidence.

As an example, consider some of the data on homicide examined in Daly and Wilson’s (1988). Sociobiology obviously makes no quantitative prediction about the incidence of infanticide or patricide. It does, however, lead to fairly precise qualitative predictions about the relationship between the probability of a child being murdered by a parent and the probability of a parent’s being murdered by a child: in particular, the figures in question should reflect the fact that the reproductive value of a child for the parent rises with age until reproductive maturity, and that of a parent for a child diminishes with age. These predictions conflict both the Freudian assumptions (based on the idea of Oedipal conflict) and on those of ‘any theory that treats patricide and filicide as alternative products of a common set of stressors.’ (Daly and Wilson, 1988, p. 104) In the latter case,

‘ we would.. expect that the risks of parricide, filicide, and child abuse would be related to demographic variables in directionally similar ways.... And yet we find that the effect of parental age upon the risk of parricide is precisely the opposite of the effect upon the risk of filicide and child abuse.’

In the case of Freudian theory, we would expect that because of Oedipal conflict

the victims of males [would be] more often male than the victims of females--at all ages. [Sociobiology,] by contrast, suggests that there should be no contingency between the victim’s sex and that of the assailant when young children are involved, but that such a contingency should appear--due mainly to a preponderance of male-male rather than female-female cases--after the child reaches maturity.’ (121)

That reason for the sociobiological prediction is that on the basis of selection thinking,

there is indeed a conflict between father and son over the wife/mother. There is furthermore a genuine ‘sexual’ conflict between them, or at least a conflict over the timing of the son’s accession to a potentially reproductive status... But the former conflict is not sexual and the latter is not over the mother.’ (Daly and Wilson, 1988, p. 115)

The former conflict has to do with the competition for the attention of the caretaker (mother), and therefore should have nothing to do with the sex of the child. The latter, however, has specifically to do with sexual rivalry between fathers and sons.

As it turns out, the sociobiological hypothesis is strongly supported over the rival hypotheses: when we look at the incidence of homicide, there is no significant correlation between the victim’s sex and that of the assailant where the victim is younger than 15, whereas over 16 there is a correlation such that the probability of its being due to chance is less than .001. Now no doubt some sociologically oriented theory might be concocted to explain this fact, but since it is a genuine and not antecedently plausible prediction of sociobiology not made by any existing sociological theory, it will seem very much as if it is sociology, not sociobiology, which in this case has done the ad hoc cooking.

The argument about the triviality of sociobiological predictions is closely related to Gould’s (1977) complaint in a famous review of Sociobiology: The New Synthesis:

Unless the ‘interesting’ properties of human behaviour are under specific genetic control, sociology need fear no invasion of its turf. By interesting, I refer to the subjects sociologists and anthropologists fight about most often--aggression, social stratification, and differences in behaviour between men and women. If genes only specify that we are large enough to live in a world of gravitational forces, need to rest our bodies by sleeping, and do not photosynthesize, then the realm of genetic determinism will be relatively uninspiring.

Let us ignore the issue of genetic determinism which, as I have already argued, is a red herring. Surely it would be a miraculous coincidence if genes only affected the uninteresting properties of human beings! What mechanism can we possibly imagine that would guarantee that genes would keep ‘hands off’--or ‘helix off’--all and only those properties which we, after a few million years of evolution, happen to find interesting?

Once again, an attack against sociobiology by an outstanding biologist and philosopher has far overshot the mark. It is not legitimized by the logic of the argument itself, and therefore itself demands to be viewed in a psycho-sociological perspective.

I have so far suggested that the arguments offered against sociobiology cannot justify their virulence on the basis of their epistemic warrant alone. In the remainder of this paper, I want first to make a few rather unsystematic remarks about the direct clash of epistemic and other values, and then to suggest that sociobiology arouses such passions not so much because of its clash with those other values as because of its perceived philosophical implications.

Assuming that there is a truth of the matter, should the search for such truth be subordinated to certain non epistemic values? By asking this question, one has already stepped outside the context of scientific rationality: we are in the realm, therefore, in which we have in effect agreed to play the game of the sociologist/pragmatist.

Again, I need to step backwards for a moment for a couple of methodological remarks. In his paper opening the present conference, Jim McAllister (1990) characterized the debate between the rationalist and the sociologist in terms of the explanatory role of rationality: is rationality a genuine causal factor, or can it be effectively eliminated from our understanding of science, in favour of factors more properly studied by the sociologist than by the epistemologist? McAllister’s proposed dissolution of the dispute was pleasantly irenic, urging us to treat both classes of factors seriously. But it did not address an issue that arises more naturally for the sociologist (henceforth for convenience ‘him’) than for the rationalist (henceforth for convenience ‘her’.) This is the question of the value of science, or of the values served by science, in the context of a field of competing practical values.

The reason why the issue arises more naturally for him than for her, is that she can more readily argue that there is something about epistemic value that can be radically distinguished from other values. One can debate whether self-deception is ever a good policy, given the complexity of competing human values; but for a rationalist the debate is engaged against the background of a common an assumption that from a strictly epistemic point of view self-deception is wrong. Analogously, in the context of the philosophy of science, the type of value in question may plausibly be taken to be exclusively epistemic, and the issue of the importance of epistemic values in the general field of competing human interests is a political one. (In practice this means that we can continue to argue, with a clear conscience, that as scientists we must be allowed to do pure science however we understand that notion, because the price of censorship by politicians is too high even for politicians to countenance.)

If, on the other hand, the acquisition of scientific beliefs is viewed, as it is by the sociologist, as being caused by factors that have no privileged link to a special class of canons of rationality, then there will be no prima facie reason for insisting that epistemic value should be insulated from open competition with other values. In this context, I can explain why I have called this paper ‘the Sociology of Sociobiology,’ despite my manifest lack of any qualification to speak as a sociologist. I have an ad hominem excuse for arrogating to myself the right to use ‘sociology’ to mean the study not so much of the facts about certain social institutions but of their evaluation. My excuse is this: If the sociological critics of science are right, their own activity can make no claim to scientific status in the traditional sense, since no science can. But then their critique of science is not scientific but ethical. So if I recommend a certain ethics of science, it seems I must be in the same business after all: they are really doing a kind of ethics; and so am I. But if they are wrong (as in fact, after all, I think they are) then I am content to think of what I am doing as philosophy. After all, that’s what I’m paid to do. In any case,I shall return to philosophy in a more conventional sense in a moment.

V.  The Argument from Consequences.

In looking at the value of scientific research one must not taken an exclusively epistemic point of view, but rather a pragmatic one.

The argument from consequences can usefully be considered with reference to Arthur Jensen’s arguments about race and intelligence. Jensen, though not a sociobiologist, is among those researchers whose work has become practically taboo in North America. He cannot appear on a North American campus to give a lecture without there being large demonstrations, mostly consisting of people whom have never read his work, dedicated, sometimes successfully, to the aim of preventing him from speaking.

Now Jensen has (or might have) defended the publication of his research on the political level--precisely the level on which he has been attacked--by pointing out that if the IQ of blacks is lower than that of blacks, then equity demands that they be given certain educational advantages.

And surely he would be right here, in the sense that the [putative] fact about race and intelligence can be used with equal force to support both the claim that we should spend less and the claim that we should spend more on the education of blacks. The missing premise on which the contrary conclusions depend are, of course, evaluative or political premises, commitments that need to be made by a society to certain norms of fairness.

The counter might be that it is a social fact that people are not going to use the results in question in this way. Instead, they will likely argue that educational resources are wasted on blacks, since they are unable to benefit equally from them, and therefore that less money should be spent on their education rather than more.

The operative principles in this area appear to be confused. In North American society, for example, it is now taken for granted that certain physical handicaps ought not to stand in the way of a person’s access to public places, such as movie theatres or University classrooms. On the other hand, it is not at all taken for granted, so far, that people suffering from the mental handicap, of, say, math anxiety, are entitled to become nuclear physicists.

In terms of policy, then, the argument again seems to me inconclusive. And it if is inconclusive, the vehemence of the attacks against sociobiology (and related research such as Jensen’s) is understandable more as an indication of insecurity than of correctness.

Worse, on a philosophical level, the argument is likely to be counterproductive in its own terms. If we confuse practical consequences and scientific truth, we will in fact be less likely to deal with the practical consequences than if we make the distinction clearly at the outset and deal with the consequences separately. The reason is simple: One may not be in a position to know just what the consequences in question might be until we know whether the doctrine in question is true. The consequences of ‘most people believes p and p is true’ might well be quite different from the consequences of ‘most people believes p and p is false’. Suppose, for example, that p is the proposition that blowing cigarette smoke into a child’s ear will cure head colds. This belief, if widely shared, would have widely beneficial consequences--but only if happened to be true. Consequently, if a certain doctrine is to be promulgated for the sake of its consequences, we should first discover whether or not it is true. In such cases, it is precisely in order to be faithful to his commitment to the importance of social consequences, that the sociologist must first attend to questions of rationality.

VI.  The philosophical dimension of sociobiology.

One definition of philosophy that has always appealed to me is that philosophy aims at change of vision brought about by argument. Now there are two aspects to the ‘vision’ proposed by sociobiology, corresponding respectively to the broad and narrow meanings of sociobiology.

Corresponding to the narrow interpretation of sociobiology, the competing visions concern the primacy of groups, individuals, or genes. Whether the group or the individual is more important is a dispute that can readily be given a political sense, which, though somewhat vague, is powerfully divisive. It is closely connected to another philosophical dispute: whether there is something natural about cooperation, or whether psychological egoism is correct. It makes some people very anxious if someone suggests that psychological egoism is true for biological reasons. Actually it’s perfectly obvious that psychological egoism is false, as a matter of empirical fact, and it makes no difference at all what theoretical biology has to say about it. Surely the discipline we should be looking at if we are interested in what is humanly possible is not biology, but anthropology and history. The proper business of (socio)biology is the search for the mechanisms underlying the possibilities revealed by those human studies.

Corresponding to sociobiology in the broad sense, the vision is of humans as natural beings, biological organisms whose motivation is ultimately to be explained in ways that must conform to (though not necessarily be exhausted by) the imperatives of selection theory. That vision competes most sharply with the religious vision, in which human life has a meaning deriving from some purpose outside itself, but many people who would disavow the religious vision would want to resist the biological one, in the name of some form of autonomy of the human phenomenon, whether by reason of reason or by reason of the emergence of irreducibly social features.

Evolutionary biology must at least act as a constraint on what is possible, biologically, psychologically, and socially. This doesn’t just mean that for biological reasons we can’t, say, jump to the moon: constraints of that sort are those dismissed by Gould as too obvious to be interesting; as for those that are equally general but less obvious, they are also likely to be too controversial to be of much use. The constraints in question are theoretical, in the sense that if, say, some physiological theory (let alone psychology or sociology) actually contradicted evolutionary biology, then this would count against the physiological theory (or psychology or sociology). Freud’s death instinct is a good case in point: selectionist thinking shows it to be completely absurd, as are all versions, still to be found in the literature, of the attempts to give an ‘evolutionary explanation’ for death. (One favourite old chestnut is this: ‘if organisms did not die evolution could not take place.’ This is logically on a par with explaining our need to eat by pointing out that if we did not eat we would not excrete,-- so there would be no fertilizer with which to grow food, and we would die of hunger....)

   VII. Conclusion: philosophical myths.

This is not to say that nothing can change our visions in this area of myths; though it is also possible that they have to do with temperamental dispositions which are so deep that we might as well consider them to be innate. Still, as philosophers, we keep trying, and the challenging of these assumptions in oneself is perhaps the most exciting and difficult task we ever set ourselves. But it is at best a matter of time and a process of ‘re-gestalting,’ not anything that will be settled by any given battery of arguments.

If we try to reapply Brown’s two-level scheme on the level of philosophical myths, we shall not really know in which class to place them. Do we hold them rationally, since there is no compelling argument against them? or irrationally, since there is no compelling argument in their favour?

At this level, the psycho-sociology of science might come into its own, without actually conflicting with the rationalist scheme. For we would indeed like to know why we hold to one set of myths and not another, and we know that we cannot expect an answer in strictly rationalist terms.

What then are the myths that sociobiology threatens?

Some of the myths that seem to do battle behind the overt disputes about sociobiology include the myth of freedom and merit, which for some reason I don’t quite understand seem to get associated with the idea of culturalrelativity; opposed to those is the myth of predestination, which in the present context is associated with biological determinism. They are also related to our deepest myths about our place in nature. In what measure are we akin to the animals? Or is out world somehow entirely defined on the plane of some purely human attribute such as meaning, society, or language? This last view seems to me the modern inheritor the claims of pure mind favoured by some of the Judeo-Christian tradition, whose defenders, often found among anthropologists such as Sahlins (1976) and Geertz (1980), envisage social science as entirely separable from biology.

To my mind, the relation between the biological and the realm of meaning is a genuine scientific problem, not some sort of ideological mirage. Sociobiology, in the background of detailed arguments by means of which cognitive science attempts to work out the details, can actually contribute to the task of devising a fruitful approach to the problem. In my view, Darwinism represents the most important of all scientific revolutions; but it is far from complete, and a surprising number of people have not yet really understood its import. Sociobiology as a research programme, even if it is in error on specific points, can help us to see the consequences of the Darwinian revolution.

But these convictions are already substantive metaphysical ones, in which I espouse one set of philosophical myths as against another. So it is fitting that this is where my argument should end.

Axelrod, R. (1984) the Evolution of Cooperation. New York: Basic Books.

Bleier, Ruth (1984) Science & Gender: A Critique of Biology & Its Theories on Women Athene Series. New York: Pergamon.

Brown, James (1989) ‘Learning from the Past’, in Brown, J.R. and Mittelstrass, eds, An Intimate Relation.

Daly, Martin and Margo Wilson (1988) Homicide, Hawthorne, NY: Aldine de Gruyter.

Dawkins, Richard (1976) The Selfish Gene. Oxford: Oxford University Press.

Geertz, Clifford (1980) Review of Symons (1979). New York Review of Books 26(2):2-5.

Gould, S.J. (1977) ‘Biological Potentiality vs Biological Determinism’, reprinted in Ever Since Darwin, New York: Norton, pp. 251-259.

Hamilton, W.D. (1964) ‘The Genetical Evolution of Social Behaviour’ Journal of Theoretical Biology, 7, pp. 1-52.

Hubbard, Ruth and Lowe, Marian, eds. (1979) Genes & Gender Two: Pitfalls in Research on Sex & Gender New York: Gordian.

Jacob, François (1977) ‘Evolution and Tinkering.’ Science 196, pp. 1161-66.

Kitcher, Philip (1985) Vaulting Ambition: Sociobiology and the Quest for Human Nature. Cambridge, MA: MIT Press, A Bradford Book.

Longino, Helen (1989) Science as Human Knowledge: Values and Objectivity in Scientific Inquiry. Princeton: Princeton University Press, 1989

Masters, Roger D. (1975) ‘Politics as a Biological Phenomenon.’ Social Science Information 14, pp.7-63.

Maynard Smith, J. (1978) ‘The Evolution of Behaviour’, Scientific American, 239 (3), pp. 176-93.

McAllister, James (1990) “Rationalist vs Sociological Views of Science.” This volume, p. 000-000.

Montagu, Ashley, ed (1980) Sociobiology Examined. New York: Oxford University Press.

Reynolds, V., Vincent S.E. Falger, and Ian Vine, eds (1987) The Sociobiology of Ethnocentrism: Evolutionary Dimension of Xenophobia, Discrimination, Racism and Nationalism. London and Sydney: Croom Helm.

Robin J.H. Russell (1987) ‘Genetic Similarity as a Mediator of Interpersonal Relationships,’ in The Sociobiology of Ethnocentrism: Evolutionary Dimensions of Xenophobia, Discrimination, Racism and Nationalism, ed. V. Reynolds, Vincent S. E. Falger, and Ian Vine. London and Sydney: Croom Helm.

Sahlins, Marshall (1976) The Use and Abuse of Biology. Ann Arbor: Univ. of Michigan Press.

Symons, Donald (1979) The Evolution of Human Sexuality. Oxford: Oxford University Press.

Williams, George C. 1966. Adaptation and Natural Selection. Princeton: Princeton Univ. Press.

Wilson, E. O. 1975. Sociobiology: The New Synthesis. Cambridge, MA: Harvard Univ. Press.