APRIL 1996

Editor and ESS Secretariate: Johan M.G. van der Dennen, Center for Peace and Conflict Studies, University of Groningen, Oude Kijk in 't Jatstraat 5/9, 9712 EA Groningen, The Netherlands, Fax: +31 50 3635635, E-mail:

Book Review Editor: Marcel Roele, Meeuwenlaan 111a, 1021 HX Amsterdam, The Netherlands, E-mail:

ISSN: 0929-0206. Published by Origin Press, Groningen, The Netherlands


by FRANS ROES, Lauriergracht 127-II, 1016 RK Amsterdam, The Netherlands

Ancient villages in some mountainous European regions are not located down by the river in the valley, but on top of mountains. This makes fresh water scarce. Years ago, hot July, I travelled by bike from one village to another in the south of France. I was puzzled by the preference, in times without modern technology and means of transport, to live at these high locations, as so many inconveniences seem associated with such heights. Medieval castles are often built on even more improbable places, such as the edges of steep cliffs.
    With Alexander's 'balance of power' theory in mind, it now seems clear to me that the inconveniences associated with high locations were out weighed by one essential advantage: An improved defensive position. Directed, of course, not against dangerous wild animals, but against other human groups.
    So far as I know no-one challenges the idea, suggested by many but nowadays largely attributed to Alexander, that human history has largely been the history of competition between different human groups. Some theoreticists, who call themselves 'group-selectionists', go much farther. They argue that the differential extinction of (human) groups - not necessarily due to competition - accounts for the evolution of important (human) traits. Here I will argue against this idea.
    Let me start with remarking that everybody agrees that groups of living organisms can go extinct differentially, and indeed sometimes do go extinct differentially. In other words, groups are sometimes differentially selected. So this is obviously not what sets group-selectionists apart from mainstream Darwinism. What is rejected by virtually all contemporary evolutionists (including Alexander), is the idea that the selection of groups played a major role in shaping the behaviour and traits of organisms. As the term 'group-selection' is sometimes used to describe the differential extinction of groups, while at other times it refers to the above mentioned theoretical idea, confusion seems inevitable. Therefore I propose the following definitions:
Group-selection is the differential extinction of groups.
Group-selectionism is the idea that behaviours and characteristics of living organisms can be explained as the result of such differential extinctions, that is, as the result of group-selection.
Finally, group-selectionists are those scholars who adhere to the idea of group-selectionism. Hopefully group-selectionists themselves agree with these definitions, as they surely want to avoid confusion over the issue whether a name refers to an empirical phenomenon (such as the elimination of a group of individuals), or to a theoretical idea.

Selfish individuals always out-reproduce group-altruists
"(...) consider an imaginary population of rabbits inhabiting an island. A mutant arises that grazes more efficiently - so efficiently that a population of such mutants will overexploit their resources and become extinct. The mutation is adaptive in the limited sense of causing its bearer to have more offspring than other rabbits, but maladaptive in the larger sense of driving the population to extinction."
"(...) it is easy to see that efficient grazers will evolve because they have more offspring than inefficient grazers. The negative consequences at the population level are irrelevant." (Wilson & Sober, 1994: 589).
This is the refutation of group-selectionism in a nutshell: Selfish mutants will invade and eventually dominate populations, even if the consequences are deleterious for the population and thus for the selfish mutants themselves. Individuals in a population who behave so as for the benefit of the species or the group will eventually - if they ever evolve at all - be outreproduced by selfish mutants or 'free riders', who profit from the altruistic behaviour of others, and hence will have more reproductive success than those others. Therefore, evolutionists assume that natural selection shaped organisms so as to maximize the reproduction of their own genes, and only their own genes - not those of their neighbours, not those of their group, and not those of their species.
In an article in defence of group-selectionism, David Wilson and Elliot Sober describe an archipelago of islands inhabited by rabbits. They conclude that even in this unrealistic model, "it seems that the parameters of the model must be very finely tuned for group-level selection to prevail against individual-level selection." (1994: 589).
    And if we assume that living organisms evolved in isolated and small groups, there is, except for the problem of the 'free rider', one more problem. Wilson & Sober assume that as a result of overgrazing, entire groups of rabbits will regularly go extinct - but will they? With Malthus in mind, I would argue that the history of life on earth looks like a history of overgrazing, overpopulation and starvation. Instead of entire groups regularly going extinct, thereby presumably favouring group- altruistic mutants in groups that survive, I would expect more often at least some individuals to survive, thereby disproportionally directing the evolution of the next generations. Which mutants will survive? Perhaps those that increase the amount of parental investment per offspring, (instead of investing in many offspring), thus better preparing the offspring to compete for the scarce resources found on the islands. I don't know if species which evolved on small, isolated islands show more evidence of K-selection than similar species that evolved elsewhere, but the idea seems testable. Another kind of mutant likely to be selected for on small islands is one that grows very large, because large size also in creases the ability to compete. If I am well informed, giganticism is in deed a common feature of inhabitants of small islands. The giant tortoises on the Galapagos islands could serve as examples.

The need for a convincing example
It is told of the late biologist Christian Vogel that he once rejected modern evolutionary theory or 'sociobiology'. Then he studied the grey Langurs and witnessed the now famous killings of young individuals by dominant males. This is a phenomenon hard to explain with the old concepts which assume that organisms behave 'for the good of the species', or 'for the good of the group'. But the modern individualistic approach not only turned out to be capable of explaining the infanticide, it also offered new hypotheses and predictions, for instance about the moment when female Langurs are likely to have spontaneous abortions. These events convinced Christian Vogel of the usefulness of the new approach, and he became (as I see it) one of the best and most enthusiastic lecturers on the subject of sociobiology.
This small story is very similar to what Thomas Kuhn describes several times in The Structure of Scientific Revolutions (1970): An old theory is capable of explaining several events, but some phenomena resist explanation, and even seem to refute the theory. Then a new theory emerges, and this new theory not only explains the unexplained phenomena and the phenomena already explained, it also offers new predictions. Very convincing, and if Kuhn ever writes a new edition of his book, I hope he includes a chapter on the scientific revolution caused by modern evolutionary theory.
    Group-selectionists sometimes suggest that their theory is capable of explaining phenomena which supposedly cause trouble to the usual individualistic approach. My problem is that I heard suggestions of broad areas where a group-selectionist approach might be useful, but I am unaware of the existence of even a single testable 'group-selectionist' hypothesis. Let alone new predictions. This is a major shortcoming, either of my knowledge, or of group-selectionism. Furthermore, if group-selectionism is needed because some phenomenon resists sociobiological explanation, then I would like to hear more about the exact nature of this troublesome phenomenon - and how group-selectionism copes with it.
We can only criticise a group-selectionist explanation, if such an explanation is clearly presented. Vague suggestions cannot be refuted by empirical data, actual explanations can. Group-selectionists should in my view present one clear explanation, and make it the focus of discussions. It would get real interesting if this explanation improves on work done by 'sociobiologists'.

A hesitating attack
Evolutionary theory assumes that organisms are expected to be selfish (in a particular sense). It appears that it is this fundamental idea about the 'selfishness' of individuals that group-selectionists want to remove from evolutionary theory, to be replaced it by some sort of group-altruism. Individuals do not behave so as to maximize their own reproductive success or inclusive fitness, no, they somehow restrain these efforts in order to favour their own group. I consider this idea an attack at the heart of evolutionary theory. Leaving creationists apart, group-selection ists are the only ones to challenge the fundamental ideas of evolutionary theory.
    At the same time, group-selectionists are downplaying their own at tack. For instance, Wilson & Sober (1994: 593) write that inclusive fitness theory "is not an alternative to the idea of group selection at all". This almost sounds as if group-selectionism and modern evolutionary theory are the same thing. If so, why bother to discuss group-selectionism at all?
Group-selectionists are either challenging basic ideas of evolutionary theory, or they are not. They can't have it both ways at the same time. I expect that ignoring or denying the differences between group- selectionism and sociobiology will only lead to confusion and boring discussions.

Alexander, R.D. (1979). "Group Selection". p. 36-43 in: Darwinism and Human Affairs. University of Washington Press.
Kuhn, T.S. (1970). The Structure of Scientific Revolutions. Chicago: University of Chicago Press.
Trivers, R. (1985). "The Group Selection Fallacy". P. 67-85 in: Social Evolution. Menlo Park: Benjamin/Cummings publ.
Wilson, D.S. & Sober, E. (1994). "Reintroducing group selection to the human behavioral sciences". Behavioral and Brain Sciences, 17: 585-654. (with 33 commentaries).


Harmon R. Holcomb III: Sociobiology, Sex, and Science. Albany, NY: State University of New York Press, 1993. Pp. 447. ISBN 0-7914-1259-8 (hbk), 0- 7914-1260-1 (pbk); US $73.50 (hbk), $24.95 (pbk).

by MICHAEL BRADIE, Department of Philosophy, Bowling Green State University, Bowling Green, OH 43403. U.S.A.

When philosophers or other critics render judgments about some activity or conception, it is always open to those who reject the judgment to challenge the grounds on the basis of which it was made. Is sociobiology a science or not? How one answers depends upon implicit or sometimes explicit adequacy conditions which serve to regulate the discussion. For instance, no characterization of science as we currently understand it could ignore the importance of testability in assessing empirical hypotheses. Any characterization which deemed testing to be irrelevant would be ruled inadequate. But this ruling is only as good as the claim that any adequate model of science must include a role for testing hypotheses. If, for some reason or other, we wanted to construct an image of science which did not deem testing to be significant it would be open to us to reformulate the adequacy conditions in terms of which alternative models are assessed. The controversy that would result would involve opponents talking at cross purposes to one another. Some would reject the new model on the grounds that a model of science which had no role for testing is inadequate. The innovators would respond by rejecting the adequacy conditions in favor of their new conception. There are no fixed ground rules for deciding these disputes. Sometimes the traditionalists win; sometimes the innovators. The scientific revolution in the seven teenth century was a time when both new scientific theories and new conceptions of science were being broached.
    Holcomb's book is in the spirit of this kind of dispute. His basic contention is that criticisms of sociobiology have more often than not relied on inadequate conceptions of what constitutes good science (p. ix). In order to do justice to the merits and demerits of sociobiological research, he proposes to develop a program for the evaluation of sciences which is adequate to the task. The key questions are: Is sociobiology a science? Can its results and methods be profitably extended to the analysis of human behavior? If so, what is its significance of any such results? This is a large and contentious subject. In order to keep the argument within reasonable bounds, Holcomb focusses on sociobiological accounts of human and animal reproductive behavior.
    The book is divided into two parts. Part I, "The Quest for Explanatory Completeness," contains two chapters. The main thrust of part I is that the evaluation of a science is a multidimensional task which involves three "units" or levels of analysis: theory evaluation, research program evaluation, and an evaluation of the conceptual frameworks associated with the science (p. 10, pp. 159). Chapter 1 argues for a conception of theory adequacy which rests on the notion of explanatory completeness. I shall return to this point below. Chapter 2 focusses on the criteria appropriate for assessing research programs. It includes a discussion of the limitations of falsifiability as an adequate criterion in this context, and the significance of reductionism and the importance of problem solving as evaluative tools. Part II consists of four chapters on "Conceptual Frame works as the Unit of Analysis." Holcomb distinguishes four kinds of conceptual frameworks. Chapter 3 deals with the conceptualization of problems. There is a discussion of the similarity between the "New Synthesis" proposed by E. O. Wilson and the Modern Synthesis, as well as the problem of the altruism and the problem of what are the proper units (or levels) of selection. Chapter 4 deals with the problem of how to conceptualize the explanatory domain of evolutionary theory. How do we determine what is or is not a legitimate question for evolutionary theory or one of its adjuncts to be asked to answer? Chapter 5, "Discipline Conceptualization," is a discussion of the different kinds of explanatory accounts which are to be found in evolutionary theory. Chapter 6, "Performance Conceptualization," addresses the general question of how a theory is to be judged with respect to its performance. The issues here involve the determination of the adequacy conditions which are appropriate for evaluating competing models of evaluation. Holcomb's position, which strikes me as basically correct, is that any set of evaluational criteria ought to emerge from the practice of the science in question and not be imposed upon that practice by some disassociated and abstract philosophical conception (p. 368f).
    Sociobiology, broadly construed, aims to uncover the biological basis of sociality in human and non-human animals. Narrowly construed, sociobiology is often identified with some particular theoretical attempt to explain social behavior in biological terms. Critics often wrongly conclude from the shortcomings of the views of particular theorists such as E.O. Wilson that the broadly construed project is ill conceived. Holcomb is careful to distinguish these two senses and to point up the shortcomings of both critics and supporters alike. The controversial part of sociobiology is the attempt to extend the analysis to human behaviors. Holcomb rightly construes this as the attempt to extend and apply a model or collection of models to a wider domain. When particular sociobiological accounts of particular aspects of human behavior fail this should be seen as "local" failures to extend a paradigm rather than as a "global" failure of sociobiology as such.
    At the programmatic level, Holcomb focusses the similarity between Wilson's call for a "New Synthesis" and the "Modern Synthesis." The "New Synthesis" is formulated as a dual hypothesis: "gradual social evolution can be explained in terms of genetic changes, recombination, and the ordering of this genetic variation by natural selection; and observed evolutionary social phenomena, particularly those involving cultural processes and primate evolution, can be explained in a manner that is consistent with known genetic mechanisms." ( p. 177) The key problematic feature here is what constitute "evolutionary social phenomena" and where and how we are to draw limits with respect to reasonable expectations for any theories formulated within this paradigm.
    As a methodological guide, Holcomb proposes that explanatory completeness is the overarching aim in terms of which theories and programs need to be evaluated. The notion of "completeness" is complex. Holcomb has many interesting things to say about this in remarks scattered throughout the book. The upshot is that while we want our theories to be complete, to explain all that we think they should explain, we should not hold them up to the impossible ideal of having to explain everything. They should explain only what is relevant. So, for example, if there are social phenomena which are not "evolutionary social phenomena", then we should not expect sociobiological theories to account for them. Of course, no general formula can be given for determining when some phenomena are or are not grist for the sociobiological mill. Each case has to be examined on its own merits. What complicates the issue is that, in line with his account of sociobiology as the attempt to encompass social behavior within the ambit of evolutionary theory, Holcomb endorses the view that explanation involves redescription. Part of the process of apply ing evolutionary theory to new domains is the reconceptualization of the phenomena of those domains in evolutionary or genetic terms. So what at first glance may appear to be phenomena outside the range of an evolutionary account - sexism in modern America, for example - may under appropriate reconceptualization become something that needs to be accounted for in evolutionary terms. These are the points at which critics and supporters will clash. Challenges to the applicability of particular theories to particular domains are subject to the same sort of tension we saw with respect to the confrontation between conceptions of science and adequacy conditions. When a theory confronts a recalcitrant domain, two outcomes are possible. [1] The domain is reconceptualized and the phenomena therein become subject to explanation by the theory. In this case, the theory remains incomplete until the relevant phenomena are explained. [2] The domain resists reconceptualization and the phenomena therein lie outside the range of what needs to be explained in order for the theory to be complete. What complicates this general issue in the case of sociobiology is that many of the arguments with respect to whether a domain is or is not to be reconceptualized are driven by philosophical commitments such as reductionism and determinism. In order to defuse these problems, Holcomb, following Arthur Fine, suggests that we adopt a "natural ontological attitude" or NOA. (p. 63f) The basic idea is to take the sociobiology as it comes and not to read into it any metaphysical commitments about human nature. As I understand it, this is a call for practicing science - that is, constructing models and trying them out - without taking them to entail philosophical commitments which may have normative implications. This is easy to say but hard to put into practice in the social sciences which are plagued by the intertwining of descriptive and normative concerns. So Holcomb's position, I fear, will not lead to the defusing of the problems of where and how and to what extent sociobiology can be applied to the explanation of human behavior.
    Holcomb's treatment is a rich and detailed analysis of the philosophical problems associated with evaluating sociobiological theory. He rightly emphasizes that the criticisms of popular or "vulgar" sociobiological accounts should not be taken to apply to the more solid work that is ever increasingly appearing in the specialized journals which have sprung up over the last 20 years. His analysis draws upon both genres for examples to illustrate methodological points. His treatment is balanced and even handed, criticizing both the critics and the supporters where he sees the arguments to be faulty. The tone is temperate in an area where the litera ture is often polemical. It is not easy reading and will tax the philosophically unsophisticated reader but there are sufficient nuggets to repay the patient reader. I wish I could say that this is the final word on the sociobiology debate that emerged in the 1970s. Unfortunately, I fear not. The debate has toned down in recent years but the underlying issue of the applicability of evolutionary theory to human affairs remains unsettled, dormant but not dead. In the meantime, a great deal of small bore research on particular topics is being carried out and will eventually either come to nothing or provide the basis for a solid evolutionary theory of human behavior. In the 1920s, when Quantum Mechanics was sorting itself out, Heisenberg reports that the physicists posted signs on their doors: Philosophers Keep Out. Perhaps sociobiologists ought to do the same at least for the time being. The problem is you can tell philosophers to mind their own business but since they often think that everybody's business is their business, they are not apt to listen. In that case, I suggest that the time has come for narrow focussed treatments of particular research programs. There is now enough sophisticated sociobiological research done to serve as the basis for several philosophical dissertations.


Mary Batten: Sexual Strategies: How Females Choose Their Mates. New York: Putnam, 1992. Pp. xix + 248. US$12.95 CAN$16.95 (paper) ISBN 0-87477- 757-7.

by EDWARD M. MILLER, Department of Economics and Finance, University of New Orleans, New Orleans, LA. 70148, U.S.A.

Mary Batten (a Los Angeles writer) reviews sexual strategy from a somewhat feminist perspective. While most of the facts and arguments will be known to someone familiar with sociobiology, the typical reader will still learn something. It would be good to suggest to a feminist friend who is somewhat antagonistic to biological arguments, since its basic message is that female choice has shaped human evolution.
    The book starts out by making the basic sociobiological point that because males need invest relatively little in mating, just a few sperm, they are genetically programmed to seek mating opportunities relatively unselectively. In contrast, because females must make relatively large investments, (especially in mammals) they are programmed to be choosy about their mates. Thus, it can be argued that evolution is driven by female choice.
    The longest chapter is the second, which deals with mate choice in non-human females. It is filled with fascinating details about how some females choose by the resources males control, and others by indicators of a male's genetic quality, such as robust health, which, in some species, is a cue to disease resistant genes.
    However, in an effort to emphasize female choice, the author may have extended the concept of female choice too far. For instance, it is described how in many species males compete to control territory, and the females apparently choose their territory and along with it their mates. This is called a case of female mate choice even though there is no evidence females actually chose mates, since the mate choice is just a byproduct of the choice of territory. After describing how the male blueheaded wrasse defends a territory, and making clear that the female choice is of site, not male, it is claimed (p. 28) that this situation "differs little from human societies that practice polygyny." Surely, the human situation is different, since the human female picks a male and associated resources, rather than attaching herself to a nesting site (or a home) and mating with all males that show up there.
    This chapter is filled with interesting trivia and research. New to me was the discussion of the sex choice gene in the fruit fly (Drosophila pseudoobscura) which leads to only daughters, and how females (if non- virgin) successfully avoid mating with the males carrying this gene. Also fascinating was Burley's discovery that red banded male zebra finches were more attractive to females, and how they experience twice as great reproductive success as green banded ones (apparently due to more polygyny), and how the less attractive males compensated by making higher parental investments. Even more striking, the sex ratio shifted so that birds with more attractive bands produced more offspring of their own sex. Naturally, this chapter is followed by a discussion of female mate choice in humans, recognizing how the choice is often decided by control of resources.
    The fourth chapter points out how sociobiology predicts that male and female behavior should differ, and that sex differences in the predicted direction have indeed been found. The results in this section, while likely to be of interest to most readers, will probably be familiar to those well read in sociobiology. These differences have implications for personnel selection and other policy problems, although she does not discuss them (Miller 1994b).     This is followed by a discussion of how males, both human and non- human try to subvert female choice. The discussion is from a feminist perspective of all the horrible things males do (although it is useful to see a logical account of why males have evolved to be the horrible things they are) such as killing a female's babies by other males, cloistering women, mutilating female genitals, and in general taking the most extreme measures to insure paternity certainty. At times she appears to go beyond the sociobiology. For instance, she interprets (p. 99) the antiabortion movement as "an extreme example of organized sperm protection with zealous groups, usually headed by men-campaigning to safeguard fertilization by any man's sperm." While sociobiology does predict that individual men will try to control women's reproductive potential, at tempting to explain anti-abortion by sociobiology is incorrect. Since the biology requires that all children have one, and only one, father, males as a group cannot benefit from restricting abortion. Thus, extending the actions individual men take trying to control the fertility of individual women to actions purportedly taken by all men to control all women is illegitimate. The author's feminism seems to have overtaken her logic here. The same charge is repeated later where after noting that some laws prohibit abortion even after rape, she states (p. 123) "Thus do some societies protect the rapist's sperm investment. What stronger evidence is there for the degree to which sperm competition among males influences social customs and law enforcement." Again, for the reason pointed out above, the existence of such laws provides no evidence about male sperm competition.
    Elsewhere (p. 102) it is claimed, "Muslim men define woman in terms of her vagina and believe that because of her voracious sexual appetite she should be circumcised - the clitoris surgically removed..." This is simply inaccurate, and an insult to Islam. Such a practice cannot be found in the Koran, and is not practiced by most Muslims (although possibly a majority of those that do practice female circumcision are Muslims, and some of the practitioners of the custom may believe it to be an important Muslim custom). She correctly points out that male rape is likely to be part of the male reproductive repertoire (either directly, or as a result of other evolved drives). She argues (p. 126) that "evolutionary biology would predict that men would be most likely to rape at the ages when the competition for mates is most intense. This corresponds to ages prior to the age at which men usually marry for the first time." Actually, it is not clear that evolutionary biology makes this prediction.
    Rape frequently has reproductive costs, either by antagonizing females (including potential mates), or by drawing retaliation from other males. Surely males who waited to rape until past the age when they had a realistic hope of marriage would enjoy greater reproductive success. Even then a strategy of purchasing occasional copulations might work better, since it would avoid being killed in an attempt to force a single copulation. More likely, rape is just a byproduct of males' strong drive to copulate and their tendency to take by violence what they want. If male rape is just a byproduct of other drives, raising the genetic cost of rape to above the genetic benefit is unlikely to eliminate the problem. Indeed, even if the probability of punishment is as low as she thinks, it still is high enough to make rape (except in war time) a poor reproductive strategy. However, penalties that impose a high cost of reproduction of those with drives that lead to rape should select for males less likely to rape (possibly because they have the self control to resist acting on their impulses). The discussion then moves to the standard sociobiological topics of monogamy and pair bonding. There is discussion (p. 137) of when monogamy should emerge, pointing out that it should emerge when cooperation in rearing offspring outweighs the advantage to either parent of seeking extra mates. However, she takes no position on the exact mechanism by which it emerged. What she says (p. 136) is "What is unclear is whether human monogamy arose out of some ecological necessity or whether it has been imposed by law and religion. Depending on the specific culture, a combination of factors may have been operating." Here she recognizes that reproductive strategies may have been affected by different factors in different cultures. The next logical step would be to suggest that the strength of the drives that lead to polygyny or monogamy would differ in various in regions of the world, whose environment and cultures had differed for tens of thousands of years. This step is not taken. Earlier, a very fascinating hypothesis had been put forward (p. 63) as "Indeed, it has been suggested that cultural differences grew out of the reproductive strategies used to attract mates in different environments. If this is the case, and there is increasing evidence that it is, female choice, ... played a major role... in determining how males look and act,..." If this line of reasoning was continued she might have developed the theory that where male provisioning was essential to female reproductive success (i.e.) where winters required male killed game for survival, the males would have been selected for willingness to provision, while in other areas (tropical) where male provisioning was not necessary, the females would have selected genes conducive to attracting and holding multiple wives (see Miller 1994 for a full development of where this reasoning goes). However, this line was not pursued. There is a discussion of r versus K selection (p. 139), but Rushton's extension of this the ory to different races (most recently described in his 1995 book) is not mentioned.
    Occasionally, the author seems to toss out her political opinions without really considering whether they are implied by, or even consistent with sociobiological knowledge. For instance, she states (p. 147) that "Research demonstrates that toxic substances can reduce the overall number of sperm, increase the number of abnormal sperm, and decrease the sperm's ability to move" and notes that radiation and smoking are known to reduce sperm quality, and then goes on to state, "Given the Earth's current population, such impairment of sperm poses no immediate threat to the species, but over the long term of thousands of years, reproductive capacity could suffer." Actually, as she is aware, it is the female reproductive capacity that is the limiting factor, not the number or mobility of male sperm. The characteristics of a male's sperm may determine whether his sperm fertilizes a specific woman, but women as a group are not threatened with an inability to obtain sufficient sperm of adequate quality to fertilize their eggs.
    Simple carelessness probably explains some errors. For instance she states (p. 164) that "For a man, a woman's display of fidelity ensures that he will father any offspring resulting from the union." Actually, by definition, he is the father of any offspring from the union. What the man really cares about is whether he is the father of the children born to the women he is united with. Likewise, saying "In Kenya," and then discussing Burkina Faso (p. 190), a country in West Africa, is probably some combination of carelessness, and lack of knowledge of African geography. Likewise, she casually states (p. 163), "Never before in the history of human evolution has it been possible for so many people to see for them selves that despite our many cultural differences, we have no biological differences." Taken literally, this is clearly wrong since there are biological differences in skin color, size, hair color, and in the frequencies of many genes. If it is taken as referring only to behavioral differences, it expresses at best only a desire, since there clearly are behavioral differences among the various peoples of the world, and no evidence that some of these do not reflect biological differences. Indeed, given that all known polymorphic genes exhibit differences in gene frequencies be tween populations (see the tables in Cavalli-Sforza, Menozzi, & Piazza, 1994), and that numerous behavioral traits are known to be affected by genes (Loehlin, 1992; Rowe, 1994), it is virtually certain that there are differences between populations in the frequencies of behaviorally relevant genes. Likewise, after describing some of the statistics on black male problems, such as their high homicide rates, we are told (p. 189), "In effect, the black male population is being systematically destroyed." No evidence is offered for this serious charge. The statistics show that not only is the number of black males increasing, but that their rate of population increase is higher than that for white males, who supposedly are doing the destroying.
    The last chapter of the book "Beyond the Reproductive Imperative: The Next Step in Human Evolution" discusses the policy implications. She is clearly concerned with overpopulation and the associated problems it brings. She proposes (p. 206) that we redefine reproductive success "in terms of quality instead of quantity". Unfortunately, redefining a word does not solve a real problem. One could imagine designing a society such that low quality offspring would not reproduce. Then producing high quality offspring, at the expense of offspring quantity, would indeed lead to reproductive success as it is measured in biology (many grand children and many descendants in latter generations). However, one suspects the author would not support such eugenic measures.
    Earlier, Batten correctly recounted how in most of history men who achieved power and high status have been rewarded with reproductive success, undoubtedly leading to the great striving for status and power found in today's male. However, she seems to confuse wish with fact when she states (p. 67) that today "For urban dwellers in industrialized societies, reproductive success is no longer a numbers game. More important is quality, the ability of offspring to compete successfully in their environment, mate, and pass on their genes. The family that has one or two children and provides high-quality parenting and education opportunities has greater reproductive success than a poor family that produces five children it cannot support and educate to become successful members of the society."
    The evidence (which she seems aware of) indicates that in modern societies (where virtually all babies survive) parents who have more children (even if these are poorly provided for and poorly educated) leave more grandchildren and have higher reproductive success as biologists define their terms, even if their success is not the type of success society approves of. Her error lies in not realizing that as survival to adulthood becomes virtually certain (with modern medicine) the number of surviving adult offspring is primarily determined by the number of babies born, not by the resources per capita. More resources may increase the ability of offspring to attract mates, but if it also (as with female education) reduces their willingness to produce grandchildren, more resources would not contribute to reproductive success. Thus, better educational opportunities probably do not contribute to reproductive success. This holds even if the additional education does not require reducing offspring quantity, which in practice it appears to do.
    She states (pp. 206-207) that "men could just as easily meet their reproductive needs by creating a new definition of competitive masculinity for themselves-one that encourages strength of character rather than strength of arms". However, if those that used strength of arms actually enjoy greater reproductive success, it is not clear how males might be induced to adopt the new standards. Possibly, the state's "strength of arm" could be used to penalize those that lack strength of character, but again one suspects the author would object, especially since such state action would probably end up being controlled by males. After all, their drives for power and status usually lead to them dominating. Thus, they would control the governmental processes that would determine how strength of character was defined, and who possessed it.
    For the serious scholar, a most infuriating character of this book is the failure to provide chapter headings for the footnotes section, while renumbering footnotes for each chapter. To find the source for a state ment the reader must determine which chapter he is in, find that footnote in the footnotes, and then go to the reference. Simply listing the chapter titles (instead of merely the chapter numbers) in the footnotes section would have reduced the labor required greatly.
    What is this book most useful for? It is an easy read that most sociobiologists will learn something from, although most of the arguments and facts discussed will be already known. The book does show that the vocabulary of sociobiology can be used to express left wing sentiments as well as right wing ones. Perhaps the greatest use will be educational, in exposing the arguments of sociobiology to feminists or others of what in the US is called a "politically correct" orientation.

Cavalli-Sforza, L. L., Menozzi, P., & Piazza, A. (1994). The History and Geography of Human Genes. Princeton: Princeton University Press.
Loehlin, J. M. (1992). Genes and Environment in Personality Development. Newbury Park: Sage.
Miller, E. M. (1994a). Paternal provisioning versus mate seeking in human populations, Personality and Individual Differences, 17, 227-255.
Miller, E. M. (1994b). The Relevance of Group Membership for Personnel Selection: A Demonstration Using Bayes Theorem, Journal of Social, Political, and Economic Studies, 19 (3), 323-359.
Rowe, D. C. (1994). The Limits of Family Influence: Genes Experience and Behavior. (New York: Guilford: 1994).
Rushton, J. P., (1994). Race, Evolution and Behavior: A Life History Perspective. New Brunswick: Transaction Publishers.


Richard Posner: Sex and Reason. Cambridge, MA: Harvard University Press, 1992. Pp. 458, US$35.95 cloth. ISBN 0-674-80279-9.

by MARCEL ROELE, Meeuwenlaan 111 A, 1021 HX Amsterdam, the Netherlands.

No longer are explicit markets the sole domain of economics. Gary Becker, Nobel laureate in 1992, has pioneered the application of economic analysis to issues of nonmarket behaviour, including marriage and the family. In Sex and Reason, Richard Posner attempts to expand the domain of economic theory even further to include sexuality. The fact that Posner is a judge of the US Court of Appeals for the 7th Circuit adds poignancy to his analysis. He expresses the hope (p. 5) that his analysis will shame his colleagues in the legal profession for ignoring the rich multidisciplinary literature on sexuality. Posner's wider ambition is "to present a theory of sexuality that both explains the principal regularities in the practice of sex and in its social, including legal, regulation and points the way to ward reforms in that regulation - thus a theory at once positive (de scriptive) and normative (ethical) (pp. 2-3)."
    Posner posits that sexual tastes are given, not chosen, and are extraordinarily resistant to change; thus, they are preferences in the sense used by economists. He further assumes that the actions people take to satisfy these preferences are governed by considerations of relative costs and benefits. Sexual behaviour is rational in the nonmentalist sense. Man responds appropriately to incentives, whether consciously or not. Consequently, the response of sexual behaviour to various forms of social regulation can be analyzed using the standard 'economic man' paradigm, producing a 'morally neutral' theory of sex.
    Before the costs and benefits of satisfying one's sexual preferences come into play, these preferences first have to develop. Different preferences may result from different (sex-limited) genetic propensities. Posner employs a sociobiological theory (largely based on the work of Donald Symons [1979]) to explain the evolution of different sex-limited genetic propensities for sexual preferences. The optimal sexual strategies for the two sexes are different. Posner stresses the similarities between the economist's cost-benefit analysis at the proximate level and that of the sociobiologist at the ultimate level: "Optimal here means, of course, from an evolutionary standpoint - the standpoint of spreading one's genes - because a strategy that was not optimal in that sense would tend to be weeded out by natural selection. And strategy carries with it no implication of conscious strategizing. It is a metaphor for the suiting of means to end; in this it resembles the economist's concept of rationality (p. 90)."
    Differences in potential reproductive capacity between the male and the female lead the male to cultivate the extensive margin; the female, the intensive. Therefore, the female has to be more discriminating than the male in her choice of sexual partner. Since the female is more discriminating than the male, the demand for sex with females exceeds the supply. Competition among men for women may result in the more power ful men asserting exclusive rights to a disproportionate number of women with high reproductive value, thereby decreasing the opportunities for the remaining men to have sex with young women. According to Posner, masturbation, homosexuality, fetishism, and voyeurism are safety valves which "enable male sexual desires to be satisfied when females are unavailable to particular men as sex objects (p. 99)."
    Of course, this begs the question of how, in the evolutionary past of our species, a genetic propensity for 'deviant' sexual preferences could have contributed to the actor's inclusive fitness. With the exception of masturbation (which is hardly a 'deviant' sexual preference), the safety valves Posner mentions are unnecessarily costly. Time and resources are spent and risks are taken in the pursuit of 'deviant' sex. Any genetic propensity for such costly 'deviant' sexual preferences would have been weeded out in the process of natural selection and replaced by a propensity to be asexual if women were unaccessible. Unless, of course, the costs of 'deviant' sexual behaviour were compensated by fitness benefits resulting from this behaviour, such as an increased chance of impregnating a woman or better survival prospects for one's relatives.
    Oddly, Posner acknowledges on page 101 that a gene for 'deviant' male sexual preferences could only have survived if it yielded inclusive fitness benefits to the individual, while on page 99 fitness benefits to the group are his sole explanation of the survival value of 'deviant' male sexual preferences in the process of natural selection: "the fact that some or even many men do not procreate will not limit the birth rate; so male deviance poses little threat to the survival of the community or of the race." However, the very opposite applies to female 'deviance', and therefore Posner states - applying the same group-selectionist logic - that in the evolutionary era, hereditary lesbian preferences would have tended to be selected out (p. 99). Of course, it cannot be denied that homosexuality among women exists, but, according to Posner, it is less common than among men, and it is relatively often opportunistic: "if circumstances make it advantageous for women to form close relationships with other women rather than with men, they are more likely to switch their sexual interest to women than would men be likely to switch their sexual interest to men" (p. 179).
    Posner interprets the type and frequency of different sexual practices as rational responses to opportunities and constraints. He claims "that much of the variance in sexual behavior and customs across cultures and eras is explained by a handful of factors, such as the sex ratio (which is far more variable than usually assumed), the extent of urbanization, and, above all, the changing occupational role of women (p. 86)."
    A sex ratio of 1 should not be taken for granted. For instance, "[i]n America today black women significantly outnumber available black men. This is partly because of higher fetal and infant mortality of black than of white males, which is due in turn to the interaction of the ten dency for pregnancy and neonatal care to be poor among blacks with the fact that male infants are more vulnerable than female ones; and partly because of the high death rate of black males from homicide and their high rate of imprisonment (pp. 137-138)." According to Posner, the surplus of available black women reduces the costs of heterosexual activities for black men, in turn fostering 'macho' attitudes in black men, such as disapproval of homosexuality and an unwillingness to invest in lasting relationships with females. However, from the data supplied by Posner on page 168, one might reach the conclusion that the skewed sex ratio is not the cause, but the consequence, of male reproductive strategy. Posner draws a parallel between African tribal society, where women combine motherhood with agricultural labour, and the poorest black neighbour hoods in the United States, where men are unemployed and the state finances single-motherhood. In both cases, men are not needed as providers. One could argue (Draper and Harpending 1988; Draper 1989) that in both cases the living conditions of men resemble an environment in the evolutionary past of our species, in which an r-selected male sexual strategy (opting for high-risk strategies to maximize opportunities for procreative sex, while minimizing paternal investment) optimized inclusive fitness. If a genetic propensity to opt for this strategy is activated, the consequences for males of acting out this strategy are high fetal and infant mortality, and high male criminality (including homicide).
    In his treatment of female infanticide - the cause of sharply skewed sex ratios in many societies - Posner does not seem to distinguish individual from common interests: "[I]n a poor society, the fewer children a woman has, the likelier they are to survive to adulthood. This is true even when most or all infants killed are girls, the 'efficient' form of infanticide because it limits the future growth of the population. (Reducing the number of males need not reduce the number of children in the next generation, since one man can fertilize an indefinitely large number of women.) ... [A] practice of female infanticide need no more bespeak hostility toward or a disvaluing of women than the thinning of trees in a forest signifies a dislike of trees. It would be irrational from a genetic standpoint for a father to be indifferent to the procreative potential, and hence to the survival, of his daughters (pp. 143-144)." If limiting the future growth of the population is the objective, female infanticide would be the best policy. However, if acting rational from a genetic standpoint (i.e. maximizing reproductive success) is the objective, male infanticide would be the best strategy for poor parents, as sons of poor parents have fewer children than their sisters (variance in reproductive success is larger among males than among females). Rich parents who act to optimize their inclusive fitness should not commit infanticide at all. As Philip Kitcher (1985:315-329) points out, rich parents have several strategies at their disposal (e.g. using wet nurses and avoiding marriage costs by allowing their daughters to marry 'down') that permit them to make the minimum investment required to enable their daughters to reach the age of marriage without substantially reducing the number of sons raised or the sons' ability to attract large numbers of wives.
    In my opinion, female infanticide among the poor is prompted by the fact that in traditional societies sons are much more valuable as labourers than daughters. In Bangladesh, the male offspring of peasant farmers become net producers from the age of ten. When they are fifteen, they have paid back all the costs their parents have made to raise them. As their parents continue to benefit from their sons' labour, sons can be regarded as a source of income for their families (Caine 1977).
    Female infanticide among the rich seems to be limited to stratified, preindustrial societies. In those societies, upper-class daughters cannot increase their parents' social status, but do cost them dowries (unless they marry 'down', in which case they do not cost money, but do decrease their parents' status). For the upper-class parents, these considerations seem to outweigh the contribution daughters can make to the parents' reproductive success. In other words, female infanticide among the rich, as well as among the poor, is a strategy to maximize socioeconomic success; it is not a strategy to limit population growth or maximize reproductive success (Roele 1993).
    Posner's treatment of the effects of urbanization and the changing occupational role of women on the type and frequency of different sexual practices allows him to make more conventional economic calculations of costs and benefits. He stresses that the main effect of urbanization is on the search costs for 'deviant' sex. Suppose homosexuals are initially no larger a fraction of city dwellers than of village dwellers. Search costs for a homosexual relationship will be higher for a homosexual villager, since they will include travel to other villages. So he is apt to move to the city. "And as a result of the migration of homosexuals to cities, the percentage of homosexuals in cities will be higher than in rural areas, which will make cities even more attractive to homosexuals by further enlarging their market (p. 126)." If homosexual acts are unlawful, the greater sense of anonymity that cities provide as compared to rural areas reduces the likelihood of detection, which in turn reduces the expected costs of punishment. In permissive societies, the sheer number of homosexuals in cities will enable the forming of visible homosexual subcultures, further reducing the costs of search. As a consequence of reduced costs, "[t]he total number of practicing homosexuals in a society will ... increase with the rise of cities, making homosexuality seem ... a by-product of economic development and modernity (p. 127)." In fact, urbanization reduces the costs of all kinds of unconventional sexual relationships and increases the costs of marriage, because a female city dweller is more likely than a female villager to have to forego career opportunities if she opts to marry. The combined result of these shifts in costs is that there is not only more homosexuality in cities, but also more prostitution, more couples 'living in sin', more abortion, more illegitimacy, etc. (p. 133).
    Sexual activities are also influenced by shifts in income. Sex is time-intensive, because time is an important component of search and search an important component of sexual activity, and the cost of time rises with income earned from work. For middle class men, an increase in income will increase the value of time and will therefore increase the costs of adultery, concubinage, and simultaneous or serial polygyny. They will be able to buy more food, better housing, better schooling for their children, etc., but will have to forego more of these goods if they opt for polygyny. Only the wealthy men can afford to buy leisure, enabling them polygynous leisure-time activities, without endangering their source of income and their purchasing power. In past times, when men were the sole earners, a rise in middle-class incomes would have reinforced puritanical values. In modern societies, the occupational role of women has changed. "That changing role is, in turn, a function of [the reduced] infant mortal ity, the [increased] value of children, the technology of contraception, the existence of labor-saving devices in household production (ranging from baby bottles to dishwashers), and the degree to which well-renumerated work not requiring great physical strength or stamina is available in the economy. All these are things that reduce the costs to women (including opportunity costs, that is, the costs of forgoing other productive activities such as household work) of working in the market (p. 86)." While a rise in middle-class incomes makes marriage more attractive when only men work in the market, it makes it less attractive the more important the occupational role of women is. They have less to gain from marriage (they do not really need a provider) and more to lose (career opportunities). Although the middle-class career woman as well as the underclass welfare-mother can raise their offspring independent of male providers, offspring are both more costly and more valuable to the former. Her time is more expensive, but there are very good opportunities for her off spring to be socially successful, provided that she invests heavily in them. Contrary to Posner, I would be inclined to speculate that middle- class and underclass women are not just faced with different opportunities and constraints, but also might have developed different preferences. While sexual preferences are probably relatively stable in adults, a father- present upbringing (still the usual setting among the middle classes) may activate a more K-selected sexual preference in the child than a father- absent upbringing (which is more common among the underclass, be cause of the higher incidence of single-motherhood as well as noncompanionate marriage/cohabitation). As a consequence, a girl from a father- present household may at a later age be more selective in her choice of sexual partners, regardless of the additional costs and benefits of her choice (Draper and Harpending 1982; 1987). If there is any truth in this, an evolutionary theory of socialization would enrich Posner's economic analysis of sexual practices. Also, there is the question of the relative strength of preferences. For instance, since in the evolutionary past of our species sex and pregnancy were always linked and pregnancy was more expensive for females than for males, we would expect that a preference has evolved in females to be more choosy in their selection of sexual partners than males. If the use of contraceptives uncouples sex and pregnancy and levels the costs of sex for females and males, would we expect this to render females as indiscriminate in their choice of sexual partners as males are, or to reduce but not eliminate the difference in choosiness between females and males, or to have no effect at all? Insight in the evolutionary psychology of sexuality is also required to make the correct analysis of the economics of sexuality.
    In the final part of the book, Posner demonstrates how his rational, morally neutral model of sexual behaviour can serve as a basis for nor mative analysis. The rational model can help identify the private and social costs and benefits of behaviours involved (e.g. rape or abortion) and can help determine how effective different attempts to modify behaviour will be. Of course, it is up to society to determine how the different interests at stake should be weighed against each other. For instance, a rapist may be regarded as a utility monster: someone whose pleasure from the coerciveness of rape - over and above all substitutes (such as sex with a prostitute who will, for a price, consent to the man's abusing her physically) - exceeds the pain suffered by his victim. However, quite apart from the fact that the fear of rape imposes social costs on all women, there is a near consensus in society that the interests of the victim outweigh the interests of the rapist to such an extent that rape should be prohibited (pp. 386-387).
    In the case of abortion, there is, according to Posner, nearly always a higher interest at stake for the fetus (namely, a 100% chance of losing its life) than for the pregnant woman. On the other hand, there is a near consensus in society that fetal life is less valuable than post-birth life. The orthodox Catholic position is that fetal loss as a side effect of some seriously indicated surgery demanded without delay for the sake of the mother can be justified, even when there is a less than 100% chance of the mother's dying if the pregnancy were to continue. Therefore, there seems to be a near consensus that the weighing value of the fetus's interest should be lower than that of the pregnant woman's interests (pp. 273- 290). However, there is no consensus on exactly how much lower the weighing value of the fetus's interest should be. Hence, as long as there is not a vital interest at stake for the pregnant woman, there is no consensus on whether her interests outweigh the fetus's interests.
    In this last part of the book, Posner only employs biology to determine the rationality of prescribing or proscribing certain behaviours. For in stance, he states that the fear of a large part of the American public that "young men and women will succumb to the blandishments of homosexual sex and the homosexual style of life" (p. 299) if legal and social inhibitors of homosexual activity are relaxed, is misplaced, since (1) the evidence indicates that homosexuality is to a large extent hereditary or congenital, and (2) insofar as it is developmental, it appears to take root early in life and independent of social attitudes (p. 296). However, Posner's utilitarian calculation of the interests of the different parties affected by a particular act could also have benefited from biological insights. For instance, if we incorporate the interest of the fetus's genes in the calculation of the optimal abortion policy, we would reach the conclusion that abortion may actually be in the fetus's interest in some cases. From this point of view, abortion is in the interest of the fetus if the positive effect of abortion on the reproductive value of the pregnant woman is at least twice as high as the reproductive value of the fetus.
    Although Sex and Reason has many weak points - of which I have mentioned but a few - my subjective assessment is that the private costs of time and energy spent reading Posner's book, increased with $35.95, are outweighed by the private benefits which can be derived from its contents. I expect that this book will also generate some social benefits, in the sense that it may have a positive influence on the development of bioeconomics and may inspire practitioners of the law to increase their knowledge of the life sciences. Purchasing and reading it should be required by law for all those active in these fields.

Caine, M.T. (1977) "The Economic Activities of Children in a Village in Bangladesh", Population and Development Review 201-227.
Draper, P. (1989) "African Marriage Systems: Perspectives from Evolutionary Ecology", Ethology and Sociobiology 10: 145-69.
Draper, P. and Harpending, H. (1982) "Father Absence and Reproductive Strategy: An Evolutionary Perspective", Journal of Anthropological Research, 38(3): 255-73.
----- (1987) "Parent Investment and the Child's Environment", in J.B. Lancaster, J. Altmann, A.S. Rossi and L.R. Sherrod (eds) Parenting Across the Life Span: Biosocial Dimensions, pp. 207-35. New York: Aldine de Gruyter.
----- (1988) "A Sociobiological Perspective on the Development of Human Reproductive Strategies", in K.B. MacDonald (ed) Sociobiological Perspectives on Human Development, pp. 340-372. New York: Springer Verlag.
Kitcher, P. (1985) Vaulting Ambition: Sociobiology and the Quest for Human Nature. Cambridge, MA: MIT Press.
Roele, M. (1993) "Religious Behaviour as a Utility- and Inclusive Fitness-Optimizing Strategy", Social Science Information, 32(3): 387-426.
Symons, D. (1979). The Evolution of Human Sexuality. Oxford, UK: Oxford University Press.

An abridged version of this review was published in Politics and the Life Sciences, August 1995, 14 (2): 286-288. ©Beech Tree Publishing.

Back to the top of this document
Back to the ESS homepage?
Purpose of the ESS?
Publications of the ESS?
Future activities of the ESS?
Recent ESS Newsletters
ESS Membership List
Websites of ESS members
ESS Associates and Links

Please send your reactions to