Desktop version

Home arrow Psychology

  • Increase font
  • Decrease font

<<   CONTENTS   >>

Sociobiology and Human Development: Arguments and Evidence

Richard M. Lerner and Alexander von Eye

In 1975, E.O. Wilson published a book announcing the “new” scientific discipline of sociobiology. Wilson (1975a) contended that sociobiology would be the “master” synthetic discipline, a field enveloping all of behavioral and social science. In this article, arguments forwarded by Wilson and others in support of this synthetic role for sociobiology are evaluated, especially as these ideas have been advanced in regard to understanding human behavior and development. In particular we focus on sociobiological claims pertinent to features of behavior central to human reproduction, parenting, and child caregiving. Our assessment draws on and extends prior discussions of sociobiological ideas (Barlow and Silverberg, 1980; Caplan, 1978; Kitcher, 1985; Lewontin et al., 1984), particularly in our concern with issues relevant to human development and in our provision of an alternative to sociobiological views of the role of biology in human development, one based on a developmental contextual (Lerner, 1984, 1986, 1991) perspective regarding human development.

The analysis undertaken here is, we believe, timely. Wilson’s (1975a, b, 1980; Lumsden and Wilson, 1981) claims about sociobiology, as well as corresponding assertions made by others (Konner, 1982; MacDonald, 1988), have both found support and application (Rushton, 1987, 1988a, b) and evoked strong criticism (Barlow and Silverberg, 1980; Caplan, 1978; Kitcher, 1985; Lerner, in press), even among those associating themselves with the “new synthesis.” For instance, according to Dunbar (1987, p. 51):

Wilson created the impression that sociobiolog)' was on the verge of replacing most of the disciplines in the social and behavioral sciences. This, of course, is arrant nonsense since sociobiology does not, of itself, deal with much of the subject matter of these disciplines.

However, Dunbar (1987, p. 51) went on to defend the importance of Wilson’s views for social and behavioral science:

Wilson was, none the less, right to emphasize the importance of sociobiology in relation to these disciplines. What it in fact does ... is to provide a unifying umbrella under which these disciplines can interact on common ground.

It is the mechanism through which this unification is purported to occur that concerns us first. This mechanism bears on a key conceptual issue in the study of human development, the nature—nurture controversy.

Genetic Determinism as Sociobiology’s Key to Interdisciplinary Integration

In Wilsons (1975a) view, the “unifying umbrella" provided by sociobiolog)' is the ubiquitous influence of genes on all facets of individual and social behavior and development. Indeed, Wilsons (1975a, b) views exemplify a key conception of genetic determinism—thorough genetic reductionism. The complexity of all social behavior and development and, indeed, all human culture (Lumsden and Wilson, 1981) can be reduced to a few simple genetic principles. The key one is the idea of “gene reproduction.” To Wilson (1975a, 1980), Dawkins (1976), and other sociobiologists (Barash, 1977; Freedman, 1979; MacDonald, 1988), the essential, core purpose of human life is only to reproduce genes. As Konner (1982, p. 265) puts it, “A person is only a gene’s way of making another gene.” Similarly, Dawkins (1976, p. ix) sees humans as only “survival machines—robot vehicles blindly programmed to preserve the selfish molecules known as genes.”

Simply put, all of human development reduces ultimately to gene reproduction and, as such, a human organism “does not live for itself” (Wilson, 1975a, p. 3). Instead, the organism’s “primary function is not even to produce other organisms; it produces genes, and it serves as their temporary carrier” (Wilson, 1975a, p. 3). According to this view, humans have not evolved to produce other people, but only to replicate their particular complement of genes. Humans’ life spans represent only a relatively short period within the vast temporal span of evolution. During this time, they provide a temporary “house,” or a transport, for the genes carried within them. Given this machine-like view of the “human as transport,” it is clear why Dawkins (1976, p. 21) can see humans merely as “lumbering robots [housing genes that] created us, body and mind; and their preservation is the ultimate rationale for our existence.”

Given this core, “gene reproduction”principle, it is evident also why Dawkins (197(>) sees genes as selfish. Genes are “concerned” with nothing other than self-replication, with reproducing themselves over and over, as many times as possible. Simply, then, in this view humans are really just (seemingly complex) duplicating machines. Their mating rituals, their family relationships, and their cultural institutions are all inventions in the service of gene reproduction.

This core genetic principle of life leads to several other ideas about how genes influence individual behavior and the social world. The more copies of one’s genes one can send out into the world, then, in the terms of sociobiology, the more one is increasing one’s inclusive fitness. This concept derives from the sociobiological view that natural selection is the essential vehicle through which evolutionary change occurs (Dawkins, 1976; Wilson, 1975a). However, not all genes are able to compete equally in the face of the fierce and rigorous challenges imposed by the natural environment; only the most aggressive genotypes will succeed in this struggle for survival.

Sociobiology and Human Aggression

A parallel arises between the views of sociobiologists (Dawkins, 1976; Konner, 1982) and the ideas of Konrad Lorenz. Lorenz (1966) saw human aggression as both inevitable and inherent in the human genome—to the point of providing for innate “militant enthusiasm.” To Lorenz, not only do the genes of humans make them aggressive, but the “highest form” of humans should be the most aggressive—the most militaristic—since such action reflects the presence of genes that have succeeded most in the struggles of natural selection. Similarly, in the sociobiological world of selfish genes, and the robotic “survival machines” that house them, aggression functions to allow genes to enhance their inclusive fitness; “blind” (that is to say, unthinking, machine-like) aggression allows genes to eliminate anything in the environment that interferes with their reproduction.

Thus, to Lorenz (1966) and to sociobiologists (Konner, 1982), selfish— indeed ruthless—human aggression is the cornerstone of genes’ control over human functioning. The most successful genes—the “best” of evolution, if you will (Lorenz, 1940)—will be the most successful at ruthless aggression. They will be the most selfishly directed to maximizing their presence in the gene pool and, at the same time, to minimizing the genes of others. Similarly, and redolent of Lorenz’s (1966)) views in his book, On Aggression, Konner (1982) explains:

I believe in the existence of innate aggressive tendencies in humans [p. 203] ... if we are ever to control human violence we must first appreciate that humans have a natural, biological tendency to react violently, as individuals or as groups, in certain situations.

(p. xviii)

To Lorenz (1966), these innate violent reactions are elicited in a reflex-like manner among either individuals or groups. The reflex occurs when members of an in-group are threatened by members of an out-group. Similarly, Dawkins (1976)) contends that selfish genes impel humans to act aggressively against survival machines not of their own, close genetic group (i.e., a child or another close relative), and Wilson (1975a; see also Flohr, 1987, p. 199) likewise believes that fear of (and hatred toward) an out-group (xenophobia) is innate in humans. For example, in his book, Oil Human Nature, Wilson (1978, p. 70) writes of “hidden biological prime movers,” and contends:

In all periods of life there is [a] . . . powerful urge to . . . classify other human beings into two artificially sharpened categories. We seem able to be fully comfortable only when the remainder of humanity can be labeled as members versus nonmembers, kin versus nonkin, friend versus foe.

In sum, all genotypes must struggle arduously to include as many copies of themselves as possible in the gene pool. However, within sociobiological thinking all genotypes are not “created equal”; that is, whereas all genotypes strive to maximize their inclusive fitness, genotypes differ in what is termed in sociobiology gametic potential, that is, in the potential of a genotype to replicate itself. Differences in gametic potential are associated with the differences that exist between males and females.

Sex Differences in Gametic Potential

Within sociobiology, it is held that men and women differ in their potential for transmitting copies of their genes into the future. For example, as claimed by Konner (1982, p. xviii), “as now seems clearly demonstrated, there are biological reasons why women, like other primate females, have a weaker aggressive tendency than males.” Since aggression is the key, according to sociobiologists, to getting one’s genotype reproduced maximally, it follows that women, lacking an aggressive ability sufficient to compete with men, must evolve some other strategy to enhance their inclusive fitness. The strategy' that women use derives completely, sociobiologists contend, from the nature of the specialized cells used by women to transmit copies of their genes to future generations.

Genotype copies are contained in gametes, that is, sperms and ova. Both types of gametes function to maximize the inclusive fitness of the genotypes they carry. However, the two types of gametes have a different potential for such reproduction—due to the anatomical and physiological differences between the “lumbering robots”—men and women—housing these gametes. Men, who—it is of more than passing interest to note—were the founders and leading proponents of sociobiology, can generate a large number of genotype copies. Their gametes can be “sent forth to multiply” quite readily—millions can be sent out with each ejaculation. Thus, in the terms of sociobiology, their “gametic potential” is great, given that there is—at least theoretically—a ready, large pool of recipients of their gametes. Freedman (1979, p. 2) puts this idea as follows:

Since mammalian males produce many more sperm than females

produce ova, any given male has far greater potential for producing offspring. He is also more inclined to compete with other males over the “scarce” resource, females.

Simply put, then, any male has a greater potential for enhancing his inclusive fitness than any female, given males’ greater gametic potential. Moreover, males must have in general a more aggressive genotype than females, since they must compete for access to the female gamete, viewed as a “resource” for the deposit of the males’ sperm. Such competition is, of course, highly desirable in the view of sociobiologists, since it ensures that the most aggressive genotypes—those best suited to succeed against the struggles of natural selection—will reproduce most often.

In turn, the genotypes of females impel women to try to reproduce in quite another way. One might understand the origin and development of this “alternative,” female reproductive strategy if one asks these questions:

Given the vast difference in reproductive potential, and if the point of life is to actualize such potential, is it not reasonable to expect that on the average the male pattern of courtship will differ from the female? Might nature not have arranged it so that men are ready to fecundate almost any female and that selectivity of mates has become the female prerogative?

(Freedman, 1979, p. 12)

In answer to such questions, van den Berghe and Barash (1977, p. 813) note:

Human females, as good mammals who produce a few, costly and therefore precious, offspring, are choosy about picking mates who will contribute maximally to their offspring’s fitness, whereas males, whose production of offspring is virtually unlimited, are much less picky.

Gametic Potential and Social and Sexual Development

What does the sexes’ different gametic potential imply for understanding male and female social behavior and development? Given the selfishness of genes, and the single-minded direction of the duplicating machines housing them, men develop sexual mores dictating the acceptability (if not the appropriateness) of multiple sexual partners. Indeed van den Berghe and Barash (1977) argue that the different gametic potential of men and women explains:

the widespread occurrence in human societies of polygamy, hyper- gamy, and double standards of sexual morality. There is another related reason for the sexual double standard in such things as differential valuation of male and female virginity and differential condemnation of adultery: marital infidelity of the spouse can potentially reduce the fitness of the husband more than that of the wife. Women stand to lose much less if their husbands have children out of wedlock than vice versa [p. 813] ... In addition, a woman will, at a maximum, produce some 400 fertile eggs in her lifetime, of which a dozen at most will grow up to reproductive age, while a man produces millions of sperm a day and can theoretically sire hundreds of children. Not surprisingly, females tend to go for quality, and males for quantity.

(p. 814)

Moreover, given the large number of offspring they can potentially produce, a male’s parental investment in any one is quite small. Unfortunately for the recipients of males’ genetic copies—women, their gametic potential is quite different, and so too is their parental investment. They can replicate themselves at most every 9 months. Even with multiple births, a woman cannot replicate her genes as much in a lifetime as a man can in a short period of time. Therefore, a woman’s investment in her offspring is much greater than is a man’s. Moreover, since they cannot reproduce very frequently, women will not be motivated toward frequent copulation with multiple partners. Instead, women need to protect their offspring and assure their survival, and this need should motivate them to keep their impregna- tors bound to them. As a consequence, females develop monogamous sexual behaviors and a devotion to childbearing and rearing. Van den Berghe and Barash (1977, p. 813) argue:

For a woman, the successful raising of a single infant is essentially close to a full-time occupation for a couple of years, and continues to claim much attention and energy for several more years. For a man, it often means only a minor additional burden. . . . [MJost societies make no attempt to equalize parental care; they leave women holding the babies.

Lest anyone contend that the different moral, sexual, and social developments of men and women are merely products of socialization, Barash (1977, p. xv) argues that the sex differences in gene reproduction strategy explain “why women have almost universally found themselves relegated to the nursery, while men derive the greatest satisfaction from their jobs.” van den Berghe and Barash (1977, p. 815) note further that “ethnographic evidence points to different reproductive strategies on the part of men and women, and to a remarkable consistency in the institutionalized means of accommodating these biological predispositions.” They therefore conclude:

Men are selected for engaging in male-male competition over resources appropriate to reproductive success, and women are selected for preferring men who are successful in that endeavor. Any genetically influenced tendencies in these directions will necessarily be favored by natural selection.

(van den Berghe and Barash, 1977, p. 814)

Dawkins (1976) embellishes these ideas by contending that womens exploitation by men is biologically determined. He argues that the sexes’ behavioral developments are differentiated not only by the different number of sex cells that can be used for genotype reproduction but, as well, by the different size of their respective sex cells:

the sex cells or “gametes” of males are much smaller and more numerous than the gametes of females ... it is possible to interpret all the other differences between the sexes as stemming from this one basic difference [p. 1521 . . . Sperms and eggs . . . contribute equal numbers of genes, but eggs contribute far more in the way of food reserves: indeed sperms make no contribution at all, and are simply concerned with transporting their genes as fast as possible to an egg . . . Female exploitation begins here.

(p. 153)

Sex differences in the gametic potential and size of the gametes result not only in female exploitation in general but, in particular, in the legitimation of extramarital sex for males, but not for females, and of the use of violence toward wives who have extramarital sexual relations. To explain these sex differences, Freedman (1979) argues:

|WJe have to assume that cultural universals reflect those aspects of our species that were evolutionarily derived (evolved). Male promiscuity is universally winked at because there is nothing much we can do about it, and Kinseys [Kinsey et al., 1953] main findings appear to be descriptions of the species: males must have frequent “outlets” for sex, whether heterosexual or homosexual; whereas many females can go for long periods without copulation or masturbation. . . . And this difference appears to hinge on the difference in gametic potential that we have been discussing.

(p. 19)

As in the gelada baboon, in humans female jealousy is based not on the male’s sex act with another woman but on his potential attachment to the latter . . . Male jealousy is rather different. ... It does not make evolutionary sense for the male to invest in a child not possessing his genes, and the murderous jealousy exhibited by a cuckolded male is biologically sensible. . . . Furthermore, the cuckold’s retribution can strike either the female or the male cheater . . . and most legal systems (perhaps all patrilineal systems) wink at the ensuing violence

(pp. 20-21)

Dawkins (1976) extends across the life span the idea of the biological basis of men’s promiscuous sexual interests. He offers both “a possible explanation of the evolution of the menopause in females” (p. 136) and, at the same time, an account of the sociobiological basis of the existence of what are colloquially (and pejoratively) termed “dirty old men”:

The reason why the fertility of males tails off gradually rather than abruptly is probably that males do not invest so much as females in each individual child anyway. Provided he can sire children by young women, it will always pay even a very old man to invest in children rather than in grandchildren.

(p. 136)

Conclusions: Genetic Determinism and Human Development

In the sociobiology advanced by Wilson (1975a, b), men—impelled mechanistically by their genes—are oriented to seek sexual relations with as many women as possible, to achieve more and more copies of their genes, and to not be overly devoted to or concerned with any one or any few given “replicates.” Women, in contrast, are oriented to remain monogamous in order to maximize the probability that their relatively few replicates will survive. In essence, then, men and women are genetically impelled to differ in ways that are consistent with traditional, that is, stereotypic, sex role patterns.

As did Freud (1923), and later Erikson (1968), Wilsons (1975a, b) sociobiology in effect holds that “anatomy is destiny” regarding key features of behavioral development—ones involving reproduction, parenting, child caregiving, and sexuality. In other words, Wilson (1975a, b), Dawkins (1976), Freedman (1979), and other sociobiologists (Barash, 1977; Konner, 1982; MacDonald, 1988) have built a natural edifice encompassing the very core of all human behavior and development—the reproduction of men and women, the character of the family, and the survival of the species. Any notions of nurture or of nature—nurture fusion as sources of key features of human behavior are mere fictions if genes work in the way that sociobiology requires—that is, as selfish, goal-directed, intentional agents. According to this view, after other, more superficial “causes” of human behavior are stripped away (for instance, “causes” involved in an individual’s development, such as learning or social values), genes provide the ultimate basis for human functioning, the replication of genotypes.

According to this conception, the social world does not interact with humans’ genes, much less act as an alternative source for human development. Instead, to sociobiologists, our social world—human mores (e.g., regarding sexual permissiveness or monogamy), social institutions (such as marriage and the family), and indeed all of human culture—is nothing other than the outcome of strategies laid down by humans’ genes for their own replication. Sociobiologists have complete faith in the inevitable reducibility of human behavior to the functioning of selfish genes. Akin to Lorenz (1966), this genetic determinism view has necessarily xenophobic and ruthlessly (if not militantly) selfish implications for society. The faith in genetic determinism and reductionism maintained by sociobiologists is expressed by Dawkins (1976) in his claims that “it can be perfectly proper to speak of‘a gene for behavior so-and-so’ even if we haven’t the faintest idea of the chemical chain of embryonic causes leading from gene to behavior” (pp. 65—66) and that “Be warned that if you wish, as I do, to build a society in which individuals cooperate generously and unselfishly towards a common good, you can expect little help from biological nature” (p. 3).

To what extent is this sociobiological view of human development, and of society, supported by scientific evidence? Asked another way, what scientific evidence do sociobiologists draw on to legitimate their claims, and how adequate is this evidence? We turn now to examination of these key questions.

Evaluating Sociobiological Claims

Given Wilson’s (1975a, p. 4) original definition of sociobiolog)' as “the systematic study of the biological basis of all social behavior,” it may seem surprising, and perhaps contradictory, to learn that Wilson (1980, p. 296) also contends that “contrary to an impression still widespread among social scientists, sociobiology is not the theory that human behavior has a genetic basis.” Perhaps Wilson (1980) is just playing with words. Perhaps he means that sociobiology is not a “theory” but only a “perspective,” or merely a rather general framework within which to study systematically the biological and, therefore, ultimately, the genetic basis of all social behavior. Whether or not his statement pertaining to social scientists’ mistaken impressions about sociobiology rests on a difference in meaning between the phrases “the theory that . . .” and “the systematic study of. . .,” Wilson’s own words show that sociobiolog)' is the study of the role of the connection between genes and human social behavior. Wilson (1980, p. 296) in fact uses the term “sociobiology theory” to represent this linkage. He claims:

Real sociobiological theory allows no less than three possibilities concerning the present status of human social behavior: (a) During the rapid evolution of the human brain, natural selection exhausted any genetic variability of the species affecting social behavior, so that today virtually all human beings are identical with respect to behavioral potential. In addition, the brain has been “freed” from these genes in the sense that all outcomes are determined by culture. The genes, in other words, merely prescribe the capacity for culture. Or, (b) genetic variability has been exhausted, as in (a). But the resulting uniform genotype predisposes psychological development toward certain outcomes as opposed to others. In an ethological sense, species-specific human traits exist and, as in animal repertories, they have a genetic foundation. Or, (c) genetic variability still exists, and, as in (b), at least some human behavioral traits have a genetic foundation. Having identified these alternatives, and stressed the freedom of the discipline of sociobiology from the necessity of any particular outcome, I can now add that the evidence appears to lean heavily in favor of alternative (c).

In the case of each of the given options—(a), (b), and (c)—stress is given to the links among evolution, genetic variability, and human development and society. However, if sociobiologists have spent a good deal of time exploring the first two of the three options, such work has not found its way into the published literature. Hence, Wilson is correct in asserting that to the extent that “evidence” exists in support of any of the three options, it does so in regard to option (c). Yet, support for (c) does not exist because the three options have repeatedly been subjected to comparative scientific analyses. Rather, the preponderance of published sociobiological work—at least insofar as the human literature is concerned—has taken as its “working assumption" option (c). The “evidence” derived from such work constitutes not a test of competing hypotheses, but, rather, an attempt to bring empirical observations to bear on a demonstration of a guiding presupposition.

Indeed, given what are quite well-known facts of genetic variability' (McClearn, 1981), it would be nothing short of preposterous to conduct a scientific investigation predicated on the idea that genetic variability does not exist. As a consequence, we do not believe it plausible that either Wilson or other sociobiologists are not fully aware of this quite basic evidence about the existence of immense human genetic variability. As a consequence, it is equally difficult to envision that any serious scientific attention could be paid by' sociobiologists to options (a) or (b). Therefore, these two options cannot be, and, as we have indicated, are not, treated as viable counters to (c). Instead, this last conception is the only one actually pursued scientifically by sociobiologists. But, given that no alternatives are really comparatively tested, such pursuit is more a demonstration of how empirical phenomena coincide with a conceptual presupposition than a critical test of theoretical options.

How do such demonstrations proceed? Three types of evidence have been invoked. It is useful to examine each type separately.

Comparisons of Humans and Nonhumans: The Concept of Homology

One way in which sociobiologists demonstrate that human social behavior is constrained by evolutionarily shaped genes is to draw parallels between the behaviors of humans and of nonhuman animals. If the behaviors of distinct species can be described similarly, it is argued that there must be some evolutionary connection, or continuity, between them. A common evolutionary' pathway for a physical structure or a behavioral function in distinct species is termed a homology. Simply', then, sociobiologists argue that if the characteristics of two species can be described in a common way, evidence is present of homologous evolution. The positing of such homology is offered as proof that the characteristics in question are controlled, or constrained, by evolutionarily shaped genes.

The use of such “evidence” is exemplified in the writing of Freedman (1979). In attempting to document his views that human males’ gametic potential gives rise to sexually promiscuous behavior—in order to increase their opportunities to garner the “scarce resource” of females’ ova—while human females’ gametic potential makes them more monogamous, Freedman finds homologies between fruit flies, rhesus monkeys, and South American jungle-dwelling, polygynous humans. Freedman (1979, p. 13) argues that in all species

females tend to cluster about an average number of young whereas males form a greatly skewed curve, some very successful, many not successful at all. And, since most mammals are polygynous . . . this tendency may characterize the entire class Mammalia.

Freedman (1979, p. 14) carries his argument one step further. By again using what he regards as common behavioral descriptions across species, he attempts to provide an evolutionary and genetic account not only for inevitable human male promiscuity but also for the genetically preordained urge to seek sexual relations with other females even to the point of forcing oneself onto them, that is, committing rape (Freedman, 1979). First, citing the work of Grzimek (1972, p. 270), Freedman notes that “[i]n spring, when the gonads are at the peak of their development, there are attempts to ‘rape’ strange females in the mallard and pintail and a few other species.” Second, Freedman makes an inference about the “promiscuous, polygynous intentions” of ducks and, finally, draws a conclusion about the insatiable, continuous, and carnal search by human males for females with whom to copulate. Freedman (1979, p. 14) contends:

It would appear that if the mallard drake had his way his would be a polygynous species and, in fact, one does occasionally see a consortship of two females and a male. ... In our own species and our own culture, I am asserting nothing startling when I point out that with sexual maturity, most heterosexual males are in constant search of females, and if inhibited about sexual contact, they fantasize almost continuously and fairly indiscriminately about such contact. . . . [A]dolescent males in our culture frequently experience life as a nearly continuous erection— spaced by valleys of depression that accompany sexual disappointment.

Are these descriptions, and those by other sociobiologists (Barash, 1977; Wilson, 1975a), of purportedly comparable human and nonhuman social behavior, satisfactory proof of the evolutionary and genetic bases of human behavior? Does apparent descriptive similarity establish evolutionary homology?

The answer to both of these questions is no, for several reasons, not the least of which is the difficulty of accumulating sound scientific evidence of common evolutionary descent when only physical attributes are being considered (Atz, 1970; Gould, 1980). The task is even more problematic in the case of behavioral characteristics, as even very similar behaviors (a) may be manifestations of quite different processes, and/or (b) may serve different functions (Bitterman, 1965, 1975).

In regard to (a), it is a truism that one can describe similar behaviors across even vastly different species. For instance, insects, fish, rats, and humans all “learn”; that is, in members of each of these species systematic and relatively permanent changes in behavior occur in relation to experience. Nevertheless, the ways in which these species learn—the processes of learning—vary considerably. For example, it would be difficult to contend that thought processes play a part in the learning of insects at any point in their lives. In turn, it would be equally difficult to argue that cognition does not enter into human learning for anything other than the earliest years of the life span, and even in infancy cognition may play a role (Piaget, 1970).

Accordingly, although experience-based changes occur in all animals’ adjustment to the environment, this similarity is at best evidence for an analogy, not a homology' (Atz, 1970; Sclmeirla, 1957). In other words, different processes may subserve analogous functions. But to claim that such descriptive analogies are indicative of common evolutionary histories is, at best, naive, and, at worst, poor scholarship. Dunbar (1987) is frank in admitting this limitation in sociobiological scholarship:

[MJany of those who were influential in promoting the sociobiological perspective . . . (e.g., E.O. Wilson) . . . tend to be unaware of the more sophisticated nature of the behaviour of higher organisms and are apt to regard even advanced mammals simply as scaled-up insects.

(p. 53)

In turn, and in regard to point (b) above regarding multifunctionality, the presence of identical behaviors in different organisms does not constitute proof for even common function or purpose. To illustrate, the reasons that male mallard ducks might force copulation upon a female of their species are certainly distinct from those involved when a human male rapes a human female. Indeed, to label both the male duck’s behavior and the actions of the human male with the same term (rape) seems to trivialize, through biological reductionism, what is certainly a complex and violent human act, one that current scholars point out may not even be a behavior predicated in any way on sexuality or sexual feelings (Sunday and Tobach, 1985).

Can Freedman (1979), Barash (1977), or other sociobiologists who argue for homology on the basis of such cross-species descriptions, contend that the devaluation of women in many sectors of modern society, and the legitimation of violence as a means of exercising social (and political control), enter not into the primary causation of forced copulation by human males and/or enter as well into the basis of such behaviors in ducks? We think not. Simply stated, the mere portrayal of behaviors in two species as appearing comparable is no proof at all of their common evolutionary heritage. Nor is it any proof at all regarding the extent to which such behaviors are genetically constrained or produced.

Indeed, this conclusion seems to have been reached by Wilson (1980) himself. He notes that, “We cannot rest the hypothesis of genetic constraint in human social behavior on the indirect evidences of homology” (Wilson, 1980, p. 297). If the sociobiologists’ behavioral homologies do not constitute adequate proof for the genetic basis of human social behavior, what then does? Two other types of evidence have been offered, pertaining to the concepts of heritability and adaptation. We first consider heritability.

The Concept o f Heritability

Wilson (1980) has argued that the third of the three possible theoretical options upon which sociobiology rests is the one that current evidence favors heavily. This is the notion stressing that genetic variability exists and, as such, that at least some human behavioral traits have a genetic foundation. This seemingly straightforward statement evokes, in actuality, a thicket of conceptual confusion.

Sociobiologists wish to talk about behavioral characteristics—traits—that are common to a species. The task of the sociobiologist is to show scientifically that such traits uniformly and unequivocally characterize the subgroups of humans in question (e.g., males and females), and do so because of the possession of evolutionarily based genetic “directives” for genotype reproduction. Stated simply, sociobiologists wish to demonstrate that some human traits—that is, ones common to a given group and dealing with that group’s reproductive strategy and hence inclusive fitness—have a genetic basis.

To do so—to demonstrate the common, or invariant, inheritance of these traits—sociobiologists rely on the concept of heritability. The line of argument is that, if it can be shown that the trait in question is heritable, it must be the case that the trait is commonly inherited. The claim that if a trait is heritable it is therefore inherited seems so obvious as to border on a tautology or an assertion true by definition. However, nothing could be further from the truth: The demonstration of heritability says virtually nothing about the extent to which a trait is commonly inherited. In fact, evidence for heritability cannot be taken as evidence for the common possession of a particular set of genes.

Indeed, quite the opposite is the case. Heritability is “the percentage of variability attributable to genotype” (Pianka, 1978, p. 11). It is an estimate of variation in genes, not of commonality. In fact, we will show that if a human trait were underlain by genes common to all people in a group (as

292 Counterfactual Nature of Genetic Reductionism

sociobiologists need to establish), the estimated heritability index for that trait would be zero.

If this preamble to a discussion of heritability suggests that sociobiologists are using the concept in a confused and confusing manner, this inference is, from our perspective, entirely warranted. Heritability is a difficult, confusing concept. At first blush it would seem to pertain to the extent to which something is inherited, that is, is based in the genes. For instance, if one were told that “intelligence is 80% heritable” (Jensen, 1969), it might be reasonable to take this to mean that 80% of intelligence was genetically “determined”—that is, that a given person’s intelligence was largely (80%) shaped by his or her genes and that something else (environment) shaped the small percentage remaining.

This reasonable interpretation—one often made by some scientists, members of the media, and governmental policymakers—is completely incorrect. Technically, heritability pertains only to differences between people and has absolutely nothing to do with the extent to which anything—be it genes or environment—determines characteristics within an individual. What heritability does refer to is the extent to which differences between people in a specific characteristic can be summarized by genetic differences between these people.

For a technical explanation of this claim, consider the index of “broad heritability” (a term we discuss again later) as a sample case of the usual variance decomposition approach. The coefficient is

where H2 = broad heritability, s2(G) denotes the genetically determined variance and s2(P) the phenotypical variance. As is well known, the variance of a variable is defined as

From this expression we see that (1) relates the individuals’ genetic differences to their phenotypical differences. The coefficient H2 measures the portion of phenotypical variability that can be accounted for by genetic variability.

The coefficient in (1) has been criticized for many reasons such as, for instance, lack of inclusion of possible homogamy or the (dis)similarity of environments. Therefore, more sophisticated approaches have been proposed, such as Cattell’s (1960) multiple abstract variance analysis approach and, more recently, the structural equation models presented by Chipuer et al. (1990). However, these approaches have in common the expression of heritability in terms of variance components. Thus, the interpretational basis is, as for H2, a variability rather than a commonality measure.

To give an example of how misleading heritability interpretations can be, suppose a society had a law pertaining to eligibility for government office. The law was simply that men could be elected to such positions and women could not. Consider what one would need to know in order to divide completely correctly a group of randomly chosen people from this society into one of two groups. Group 1 would consist of those who had greater than a zero percent chance of being elected to a leadership post and group 2 would consist of those who had no chance. All that one would need to know to make this division with complete accuracy was whether a person possessed an XX pair of chromosomes or an XY pair. In the first case, the person would be a female (since possession of the XX chromosome pair leads to female development). In the second, the person would be a male. One could thus correctly place all possessors of the XY pair into the “greater than zero chance” group and all possessors of the XX pair into the “no chance” group.

In this example, then, all the differences between people with respect to the characteristic in question—eligibility for office—can be summarized by genetic differences between them, that is, possession of either the XX or the XY chromosome pair. In this case the heritability of “being eligible” would be 1.0. In other words, in this society eligibility is 100% heritable. But, by any stretch of the imagination, does this mean that the eligibility characteristic is inherited, or that the differences between men and women with respect to this characteristic are genetic in nature?

Is there a gene for “eligibility,” one that men possess and women do not? Of course, the answer to these questions is no. Although heritability in this case is perfect, it is social (environmental) variables—laws regarding what men and women can and cannot do—that determine whether or not someone has a chance of being elected. Indeed, if the law in question were changed, and women were now allowed to hold office, then the heritability of the eligibility characteristic would—probably rather quickly—fall to much less than 1.0.

Hebb (1970) offers another useful example, one drawing on a “modest proposal” put forth by Mark Twain:

Mark Twain once proposed that boys should be raised in barrels to the age of 12 and fed through the bung-hold. Suppose we have 100 boys reared this way, with a practically identical environment. Jensen agrees that environment has some importance (20% worth?), so we must expect that the boys on emerging from the barrels will have a mean IQ well below 100. However, the variance attributable to the environment is practically zero, so on the “analysis of variance” argument, the environment is not a factor in the low level of IQ, which is nonsense.

(p. 578)

In Hebbs example, environment had no differential effect on the boys’ IQs; presumably in all boys it has the same (severely limiting) effect. In having this same effect, environment could contribute nothing to differences between the boys. No difference—or variation—existed in the environment, and so the environment could not be said to contribute anything to differences between people. Yet, it is also obvious that environment had a major influence on the boys’ IQ scores. Even with IQ heritability equal to +1.0, the intelligence of each of the boys would have been different had he developed in an environment other than a barrel.

A third example is based on actual empirical research. Partanen et al. (1966) analyzed data from 172 monozygotic and 557 dizygotic male twin pairs. All participants were alcohol users. The aim of the study was to estimate the degree to which alcohol abuse is genetically determined. When measured by frequency of alcohol consumption, alcohol abuse seems to have at least a modest genetic component (heritability = 0.40). However, if one uses the amount of alcohol consumed on each occasion, the heritability estimate drops considerably (to 0.27). A third measure of alcohol abuse, the number of citations and other social conflicts resulting from drinking, yields a heritability estimate of 0.02.

Thus, judgments concerning heritability can depend largely on the definition and operationalization of the behavior under study. In addition, the confusion between commonality and variability can lead to misinterpretation. Accordingly, high heritability does not mean developmental fixity. A high estimate of heritability means that environment does not contribute very much to differences among people in their expression of a trait; yet environment may still provide an important (although invariant) source of the expression of that trait, for instance in determining the average level of a trait shown by people in a given group.

From Hebb’s example we see clearly that although heritability may be high, the characteristic in question may still be influenced by the environment. Even when environment contributes nothing to differences between people in a population, this fact does not mean that the population characteristic is fixed by heredity or that it is unavailable to environmental influence. As Hebb well points out, while contributing nothing to differences between people, environment can still be a uniformly potent source of behavioral development and functioning within each of the people in a group.

A related point has been made by Lehrman (1970). When geneticists speak of a trait as heritable, all they mean is that one is able to predict the trait distribution in the offspring of a group on the basis of knowing the trait distribution in the parent group. One can predict the distribution of eye color in the offspring generation merely by knowing the distribution of eye color among the parents. Thus, while geneticists may use the term hereditary or inherited as interchangeable with the term heritable, they are not, by such usage, making any statements about the process involved in the development of this trait. In other words, the geneticist is not saying anything at all about the way that nature and nurture serve as sources of a heritable trait. Thus, the geneticist is not saying anything about the extent to which the expression of the trait may change in response to environmental modification. In short, a geneticist would not say that a highly heritable trait cannot be influenced by the environment. Rather, the geneticist would probably recognize, as we now must, that even if the heritability of a trait is + 1.0, an almost infinite number of expressions (phenotypes) of that trait may be expected to develop as a result of an interaction with the almost infinite number of environments to which any one genotype may be exposed.

Those who equate heritability with genetic determination assume that as the magnitude of heritability increases from zero to + 1.0, less and less can be done through environmental modifications to alter the expression of the trait. Correspondingly, they assume that if heritability is low, more room is left for alteration of the trait by means of environmental manipulation. This argument is fallacious. As noted by Scarr-Salapatek (1971, p. 1128):

The most common misunderstanding of the concept “heritability” relates to the myth of fixed intelligence: if h2 [heritability] is high, this reasoning goes, then intelligence is genetically fixed and unchangeable at the phenotypic level. This misconception ignores the fact that lr is a population statistic, bound to a given set of environmental conditions at a given point in time. Neither intelligence nor h2 estimates are fixed.

In short, whatever the level reached by an estimate of heritability, environmental variation may be a (or the) key causative factor. In addition, it is clear that high heritability does not mean developmental fixity. If the social context changes, one cannot be certain if any information one has about heritability still applies.

Several other statistical and methodological problems associated with the determination of heritability are important to note (Hirsch, 1970, 1976, 1990a, b; Hirsch et al., 1980; Lerner, 1986; McGuire and Hirsch, 1977; Walsten, 1990). A key problem arises in regard to differences between the concept of broad heritability (H2), mentioned earlier, and the additional concept of narrow heritability (h2). To understand these concepts, we should recognize that the contributions of variation in heredity and environment to the variance in a given behavior such as general intelligence or personality might be expressed in several ways.

Hereditary and environmental variance can relate to behavioral variance separately and independently. In such a case, what one contributes is unrelated to what the other contributes. Each may be labeled as a main effect, in the conventional analysis of variance sense of the term. Alternatively, the contributions of hereditary variance and environmental variance may interact, again in the conventional analysis of variance meaning of the term. Finally, heredity—environment (or genotype-environment) correlation occurs when hereditary (genotype) and environmental differences covary. Together, the concepts of main, interactional, and correlational contributions of hereditary and environmental variation allow the concepts of broad heritability (H2) and narrow heritability (h2) to be distinguished and the statistical problems associated with them to be noted.

H2 is the proportion of the total variation in a given behavior that may be attributed to the sum of (a) the independent genetic variation (i.e., the main effect of hereditary variance), and (b) the variance due to genotype- environment correlation (McGuire and Hirsch, 1977, p. 46). H2 is of little interest in genetics (McGuire and Hirsch. 1977), primarily because it does not allow the contribution of genotype variance per se to be disentangled from environmental variance; and, of course, such separation is an objective of heritability analyses in the first place.

Narrow heritability (h2) is simply the proportion of variance in behavior that may be attributed solely to the main effect of hereditary (genotypic) variance; that is, h2 is the hereditary variance that exists independent of, and thus merely separately adds to, environmental variance (McGuire and Hirsch, 1977). It is the determination of h2 toward which most heritability analysis is aimed. But there, especially in the case of the analysis of such variation in human behavior, lies the rub. It is in human heritability' analysis, and the calculation of h2, that the statistical and methodological problems inherent in this work arise.

It makes sense to attribute through heritability analysis (and the calculation of h2) variation in some behavior to variation in heredity only if the contributions of heredity' and of the environment are additive (Walsten, 1990). If heredity and environment interact, the calculation of heritability is quite problematic (Bullock, 1990; Feldman and Lewontin, 1975; McGuire and Hirsch, 1977; Walsten, 1990). However, in assessing heritability among humans, the statistical techniques used to determine if there is evidence for heredity—environment interaction (analysis of variance or its equivalent, multiple regression) are not as sensitive to the presence of interactions as they are to the independent (and hence additive) influences of heredity' and environment. In other words, these statistical techniques cannot detect as readily' the presence of interactions as they' can the presence of main effects (Walsten, 1990).

This problem is especially' apparent when these statistical tests are used with relatively small numbers of observations, and it is unfortunately the case that such small samples are generally the rule in social and behavioral science studies involving the calculation of heritability (Walsten, 1990). Thus, the smaller the sample used in a study, the greater the likelihood of the statistical tests being unable to identify an interaction that is actually present. In such situations, the inference that an interaction between heredity and environment does not exist, and that in turn the contributions of these two factors involve only main, and therefore only' additive, effects, is incorrect. As a consequence, heritability estimates, and any conclusions based on them, are similarly misconceived.

If a large sample of observations exists, other methodological problems occur. If a scientist is interested in determining how much variance in a behavior is accounted for by variance in what people inherit, versus what they experience in their environment, it is imperative that the nature of the specific heredity involved in the group under study be completely certain. If one wants to determine the extent to which variation in children’s intelligence is accounted for by the genes provided to children by their mothers and fathers, as compared to the environments provided by these parents, one must be certain to measure the intelligence of children and their actual biological mothers and fathers. As Hirsch et al. (1980, p. 236) have argued, “A sine qua non for the study of heredity is proof positive of the presumed biological relationship, i.e., ascertainment of the biological validity of the designated kinships, such as parent-offspring, sibling, etc.”

Unfortunately, however, few studies of heritability have included controls for presumed biological relatedness. This absence makes one uneasy about presuming that in all studies involving estimates of the contribution of genes to behavioral resemblance between, for instance, children and parents, accurate designations of biological relatedness have occurred. Such concerns are magnified when, in the few studies including such controls, evidence emerges that substantial proportions of the people labeled as “biological parent” are in fact unrelated biologically to the children in question. For instance, Hirsch et al. (1980) were able to determine, through blood testing, the actual biological relationships between parents and children within a subsample of 38 of the 112 families they studied. In 13% of the families in this subsample, there were children who could not have been the biological offspring of at least one of the putative parents in the family.

Similarly, Philipp (1973; Hirsch, 1990a; Hirsch et al., 1980), reporting evidence derived from comparable analyses done on samples from the U.K., noted that there were data disqualifying as the presumed fathers 30% of the husbands within the families being studied.

When correct information regarding biological relatedness is unavailable to the researcher who is appraising heritability, literal miscalculations and inferential errors abound. Such concerns have led Kempthorne (1990, p. 139), a quantitative genetic analyst, to the view that “most of the literature on heritability in species that cannot be experimentally manipulated, for example, in mating, should be ignored.” In a similar vein, population geneticists Feldman and Lewontin (1975, p. 116B) have concluded that “Certainly the sample estimate of heritability', either in the broad or narrow sense, but most especially in the broad sense, is nearly equivalent to no information at all for any serious problem of human genetics.”

Thus, what heritability estimates at best provide (and only in the case of additive genotypic variation only, large sample sizes, compelling evidence for no genotype—environment interaction or correlation, and valid evidence for the relevant biological relatedness) is an estimate of the extent to which genetic differences (variation) within a given group are associated with differences (variation) in the scores for a trait measured among people in that group at a specific time in their lives.

In sum, then, heritability estimates describe only characteristics of a distribution of scores; they describe only a feature of differences between people. Such estimates say nothing about the trait itself. Such estimates, in particular, say nothing about the genetic and/or environmental determination (or cause) of the trait within any person in a group. Certainly, from such an estimate of/lefiiwii-people differences (which is what heritability is), one cannot legitimately make any statements about how humans have been selected for the homogeneous presence of a trait (Lewontin et al., 1984). Indeed, to the extent such homogeneity exists, heritability must be low, due to lack of variation.

Sociobiologists, in using the presence of heritability to support their claim for the genetic determination of universal characteristics of human social functioning, are, in effect, “shooting themselves in the foot” every time they argue that the presence of heritability supports this claim. As we have seen, heritability data, to the extent that they are useful at all, imply just the opposite. Any attempt to use heritability data to support a claim for the universal, evolutionarily based genetic determination of behavior is thus an argument based on a misunderstanding and misapplication of the heritability concept.

There is, then, only one line of argument left to support sociobiologi- cal claims about the evolutionarily based, genetic source of human social behavior. This line of argument involves the concept of adaptation and the view that the patterns of human development seen in society reflect evolutionarily based, and therefore naturally selected, genetically influenced adaptations to the pressures of humans’ context.

Arc Adaptations Everywhere?

A cornerstone of the sociobiological “method” is to offer explanations in the vein of Kipling’s Just-So Stories of how particular social behaviors, or differences among people in their social status or roles, came to be (Gould, 1980). As recounted by Gould (1980, p. 258):

Rudyard Kipling asked how the leopard got its spots, the rhino its wrinkled skin. He called his answers “just-so stories.” When evolutionists try to explain form and behavior, they also tell just-so stories—and the agent is natural selection. Virtuosity in invention replaces testability as the criterion for acceptance.

According to Gould (1980), this unacceptable scientific procedure led the biologist von Bertalanffy (1969, p. 11) to complain:

If selection is taken as an axiomatic and a priori principle, it is always possible to imagine auxiliary hypotheses—unproved and by nature improvable—to make it work in any special case. . . . Some adaptive value . . . can always be construed or imagined. ... I think the fact that a theory so vague, so insufficiently verifiable and so far from the criteria otherwise applied in “hard” science, has become a dogma, can only be explained on sociological grounds. Society and science have been so steeped in the ideas of mechanism, utilitarianism, and the economic concept of free competition, that instead of God, Selection was enthroned as ultimate reality.

According to both Gould (1980) and von Bertalanffy (1969), the key feature of sociobiological “just-so stories” is that these current arrangements in society are adaptations; that is, they are changes that enhance fitness, that have been shaped by natural selection over the eons of human evolution to have this function, and that are now represented in the human genotype. Yet, it is the key element in these arguments—the presence of an adaptation, of a change in fitness—that all too often remains a scientifically unverified, post hoc story.

Indeed, as admitted by Dunbar (1987, p. 50):

A simple statement that “X increases the fitness of those that perform it” explains nothing: it is strictly tautologous, for improving fitness is what every sociobiological explanation implicitly assumes. What we need to know—and this is the heart of any sociobiological explanation—is: How does it increase fitness? ... It is the transparent failure to answer this question that has left so many sociobiologists open to criticisms of “Just-So” story-telling and unscientific practice. Since we necessarily have to rely on comparative observations rather than experimental manipulation when tackling evolutionary problems, we are particularly exposed to this kind of accusation. The only way to avoid it is to provide as watertight a case as is possible by showing that proximate problems of survival or reproduction are in fact resolved when individuals behave in a specified way, and that efficient solutions to these problems will result in increased contributions to the species’ future gene pool. This will not always be easy, but, unless it can be done, sociobiological explanations will always be open to skeptical doubts, particularly where these doubts are fuelled by political or religious conviction.

Despite these explanatory difficulties, sociobiologists see adaptations— changes in fitness “designed" by (or, actually, “resulting” from) natural selection—as everywhere. And, in the view of sociobiologists, these changes in fitness, since they are adaptations, are optimizations. That is, as argued as well by 19th-century social Darwinists (Tobach et al., 1974), natural selection results in genetically based features that are the “time-tested,” best possible outcomes of humans’evolutionary history.

To sociobiologists, then, that which exists is an adaptation: Humans’ social behaviors and the niches they occupy in the social hierarchy have been shaped by natural selection to take their present form. As claimed succinctly by Konner (1982, p. 18), “An organism has characteristics; they must have been selected for or they wouldn’t be here now.” Given the centrality of the concept of adaptation in sociobiologists’ thinking, we may ask whether there is the direct, uniform, and singular pathway that sociobiologists infer from evolution, through natural selection, to adaptation and the present character of people and society. We may ask also precisely why presenting a story—which is a possible scenario of the way natural selection could have resulted in a given feature of human behavior—is not sufficient to establish scientifically that just such a history transpired.

The work of Gould and Vrba (1982) is quite relevant to these issues. They try to provide a new term in evolutionary biology in order to clarify some important, but confusing, uses of the term “adaptation.” Gould and Vrba note that one meaning of adaptation is the shaping of a feature of the organism (a physical attribute or a behavior, for instance) by natural selection for the function it now performs. A second meaning is a more static one, referring to the immediate way in which a physical feature or a behavior enhances the organism’s current ability to fit its context. This second meaning does not take into account the historical origin of the feature, but only whether the organism’s physical or behavioral characteristics help it to meet the current demands of its environment.

Gould and Vrba (1982) cite Williams (1966) as adhering to the first definition of adaptation. Williams (1966, p. 6) contended that one should speak of adaptation only when one can “attribute the origin and perfection of this design to a long period of selection for effectiveness in this particular role.” Bock’s (1979) views illustrate the second definition of adaptation. Bock indicates that “an adaptation is... a feature of the organism . . . which interacts operationally with some factor of its environment so that the individual survives and reproduces” (p. 39).

Gould and Vrba (1982) believe that a confusion therefore exists regarding a central concept in evolutionary theory, adaptation. This conflict exists because a single term has been used, despite the fact that different criteria for the historical basis of a given organism feature and for its current use are involved in the two meanings of the term. Darwin (1859, p. 197) himself may have seen this potential confusion:

The sutures in the skulls of young mammals have been advanced as a beautiful adaptation for aiding parturition, and no doubt they facilitate, or may be indispensable for this act; but as sutures occur in the skulls of young birds and reptiles, which have only to escape from a broken egg, we may infer that this structure has arisen from the laws of growth, and has been taken advantage of in the parturition of the higher animals.

In other words, while Darwin saw the necessity of unfused sutures in the skulls of young mammals, he was uncertain about labeling the unfused sutures as adaptations, because the unfused sutures were not built by selection to function as they now do in mammals (Gould and Vrba, 1982). But if the unfused sutures are not adaptations, if they were not shaped by natural selection, what are they and where did they come from? Clearly, a new term must be used to rectify the confusion, and Gould and Vrba (1982, p. 6) provide one. They suggest that such characters evolved for other usages (or for no function at all), and were later “coopted” for their current role. They term such characters “exaptations.” The characters are fit for their current role (i.e., they are aptus), but they were not designed by natural selection for this role, therefore, they are not ad aptus (pushed toward fitness by natural selection).

A clear implication of Gould and Vrba’s revised terminology is that not all instances of fitness are adaptations; that is, not all features of an organism s structure and function that are aptational have this character as a consequence of being shaped by natural selection. Such a possibility, if supported, would serve to weaken what Gould and Lewontin (1979) have labeled the “adaptationist program”—the position, reflected in the earlier quote by Konner (1982, p. 18) that a feature’s current aptational character implies historical shaping by natural selection for that character.

Lewontin (1981) has discussed the adaptationist “program” and its conventional use of the concept of adaptation. As do Gould and Vrba (1982), Lewontin (1981) sees problems with this view of adaptation; in essence, he sees the view as deficient because it ignores the active, constructive role the organism plays in its own adaptation. The organism shapes the context to which it adapts, and hence there is a reciprocal, multilevel (i.e., fused) relation between organism and context (Lerner, 1978, 1986, 1991; Lerner and Kauffman, 1985). Lewontin’s (1981) criticisms of the conventional use of the concept of adaptation lead to a view of the organism compatible within a developmental contextual conception of human development (Lerner, 1991). Specifically, Lewontin (1981, p. 245) notes:

Organisms ... by their own life activities determine which aspects of the outer world make up their environment. Organisms change the environment by their activities . . . they “construct” environments. The problem is that the concept of adaptation has been extended metaphorically from its valid domain of describing individual, short-term, goal-directed behavior to other levels . . . |I|t is pure metaphor, ideologically molded by the progressivism and optimalism of the nineteenth century, to describe numbers of chromosomes, patterns of fertility, migrations, and religious institutions as “adaptations” ... It is not simply that some evolutionary process can be described as nonadaptive, but that the entire framework is in question. Whether we look at the fossil record or at living species, we do not see them as “adapting,” but as “adapted.” But how can that be? How is it that, if evolution is a process of adapting, organisms always seem to be adapted? It may be more illuminating to see organisms as changing and, in the process, as reconstructing the elements of the outer world into a new environment that is sufficient for their survival.

Consistent with the position of Lewontin (1981) regarding the problems with the “adaptationist program," Gould and Vrba (1982) contend that recognition of the potential presence of exaptative features leads one to recognize that previously nonaptative (note, not preadaptive) features may be present and may be coopted for fitness—a recognition that provides a key for plasticity in evolutionary processes and for the role of individuals’ own organismic characteristics in their development. Gould and Vrba (1982, pp. 12-13) indicate:

Flexibility lies in the pool of features available for cooptation. . . . The paths of evolution—both the constraints and the opportunities—must be largely set by the size and nature of this pool of potential exaptations. Exaptive possibilities define the “internal” contribution that organisms make to their own evolutionary future.

The concept of exaptation leads to the understanding that the processes involved in evolution are plastic ones. The concept is consistent with a key theme in the developmental contextual alternative to a biological deter- minist view of the role of biology in human development (Lerner, 1991, in press). According to this alternative, processes exist that contribute to the plasticity of peoples functioning—that allow them to play a role in the development of their own flexible characteristics.


Hence, the third line of evidence relied on by sociobiologists—an adaptationist story line to explain what are purported to be genetically based differences in individual and social development—fails. “Just-so stories” (Gould, 1980) about human evolutionary history are used to substitute superficial descriptions for in-depth explanations (Piaget, 1979); alternative paths to current fitness (or aptation) are excluded from scientific consideration or analysis.

Equally serious problems arise in regard to the other two lines of evidence relied on by sociobiologists—involving the inappropriate postulation of homologies between nonhuman and human animals and the misuse of the concept of heritability. These logical and empirical problems reveal the weak scientific basis of the sociobiological viewpoint. The severity of these problems suggests that sociobiological thinking has little relevance for the understanding of human behavior and development in general, or individual differences (and, most specifically, sex differences) in particular.

The limited scientific utility of the sociobiological concept of genetic determinism is consistent with the shortcomings of other reductionistic views of the role of genes in human behavior (Lewontin et al., 1984; Tobach et ah, 1974). However, one need not eschew the role of genes in human behavior and development because of the failure of genetic reductionis- tic models such as sociobiology. Developmental contextual views (Lerner, 1986, 1991) and developmental systems models (Gottlieb, 1991b) incorporate gene structure and function within a “fusion,” or a reciprocal integration, of variables from multiple levels of analysis.

In developmental contextualism, the unit of analysis is the changing relation between organism (or gene) and context; it is not either element of this relation alone. There is compelling evidence that genes do not directly—that is, in and of themselves—produce any structural or functional characteristics of an organism (Gottlieb, 1991a, b). Genes are not independently acting sources of development. Instead, the action of genes (i.e., genetic expression) is “affected by events at other levels of the [developmental] system, including the environment of the organism” (Gottlieb, 1991b, p. 5), and all levels of organization within this integrated, developmental system “may be considered as potentially equal” (Gottlieb, 1991b, p. 6). The literature pertinent to developmental contextualism indicates that (a) intraorganism variables making up to the proximal context of the gene, and (b) extraorganism contextual variables, exist in a reciprocally influential relation with genes. If, as seems to be the case (Gottlieb, 1991a, b; Lerner, 1984, 1991), increasing theoretical and empirical attention is paid to synthetic viewpoints such as developmental contextualism, it may be that the future will bring greater understanding of the ubiquitous but dynamically interactive role of genes in human behavior and development.


Supported in part by NICHD grant HD23229.


Atz, J.W. (1970). The application of the idea of homolog)' to behavior. In L.R. Aronson, E. Tobach, D.S. Lehrman, & J.S. Rosenblatt (Eds.), Development and evolution of behavior: Essays in memory ofT.C. Schneirla (pp. 53-74). San Francisco, CA: Freeman.

Barash, D.P. (1977). Sociobiology and behavior. New York: Elsevier.

Barlow, G.W., & Silverberg, |. (Eds.). (1980). Sociobiology: Beyond nature/nurture? Reports, definitions and debate. Boulder, CO: Westview Press.

Bitterman, M.E. (1965). Phyletic differences in learning. American Psychologist, 20, 396-410.

Bitterman, M.E. (1975). The comparative analysis of learning. Science, 18, 699-709.

Bock, W. (1979). A synthetic explanation of macroevolutionary change—a reductionis- tic approach. Bulletin of the Carnegie Museum of Natural History, 13, 20—69.

Bullock, D. (1990). Methodological heterogeneity and the anachronistic status of ANOVA in psychology. Behavioral and Brain Sciences, 13, 122-123.

Caplan, A.L. (Ed.). (1978). The sociobiology debate. New York: Harper & Row.

Cattell, R.B. (1960). The multiple abstract variance analysis equations and statistics. Psychological Review, 61, 122-146.

Chipuer, H.M., Rovine, M.J., & Plomin, R. (1990). LISREL modeling: Genetic and environmental influences on IQ revisited. Intelligence, 14, 11-29.

Darwin, C. (1859). On the origin of species by means of natural selection or the preservation of favored races in the struggle for life. London: John Murray.

Dawkins, R. (1976). The selfish gene. New York: Oxford University Press.

Dunbar, R.I.M. (1987). Sociobiological explanations and the evolution of ethnocen- trism. In V. Reynolds, V. Falger, & I. Vine (Eds.), The sociobiology of ethnocentrism (pp. 48-59). London: Croom Helm.

Erikson, E. (1968). Identity, youth, and crisis. New York: Norton.

Feldman, M.W., & Lewontin, R.C. (1975). The heritability hang-up. Science, 190, 1163-1168.

Flohr, H. (1987). Biological bases of social prejudices. In V. Reynolds, V. Falger, & I. Vine (Eds.), The sociobiology of ethnocentrism (pp. 190-207). London: Croom Helm.

Freedman, D.G. (1979). Human sociobiology; A holistic approach. New York: Free Press.

Freud, S. (1923). The ego and the id. London: Hogarth.

Gottlieb, G. (1991a). The experiential canalization of behavioral development: Theory. Developmental Psychology, 21, 4-13.

Gottlieb, G. (1991b). The experiential canalization ofbehavioral development: Research. Developmental Psychology, 21, 35-39.

Gould, S.J. (1980). Sociobiolog)' and the theory of natural selection. In G.W. Barlow & |. Silverberg (Eds.), Sociobiology: Beyond nature /nurture (pp. 257-269). Boulder, CO: Westview Press.

Gould, S.J., & Lewontin. R.C. (1979). The spandrels of San Marco and the Panglos- sian paradigm: A critique of the adaptationist programme. In J. Maynard Smith & R. Holliday (Eds.), The evolution of adaptation by natural selection. London: Royal Society of London.

Gould, S.J., & Vrba, E. (1982). Exaptation: A missing term in the science of form. Paleobiology, 8, 4—15.

Grzimek, B. (Ed.). (1972). Animal life encyclopedia. Vol. 7. New York: Von Nostrand Reinhold.

Hebb, D.O. (1970). A return to Jensen and his social critics. American Psychologist, 25, 568.

Hirsch, J. (1970). Behavior-genetic analysis and its biosocial consequences. Seminars in Psychiatry, 2, 89-105.

Hirsch, |. (1976). Review of Sociobiology by E.O. Wilson. Animal Behavior, 24, 707-709.

Hirsch, J. (1990a). Correlation, causation, and careerism. European Bulletin of Cognitive Psychology, 10, 647-652.

Hirsch, J. (1990b). A nemesis for heritability estimation. Behavioral and Brain Sciences, 13, 137-138.

Hirsch, J., McGuire, T.R., & Vetta, A. (1980). Concepts of behavior genetics and misapplications to humans. In J. Lockard (Ed.), The evolution of human social behavior (pp. 215-238). New York: Elsevier.

Jensen, A.R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 19, 1-123.

Kempthorne, O. (1990). How does one apply statistical analysis to our understanding of the development of human relationships? Behavioral and Brain Sciences, 13, 138-139.

Kinsey, A.C. et al. (1953). Sexual behavior in the human female. Philadelphia: Saunders.

Kitcher, P. (1985). Vaulting ambition. Cambridge, MA: MIT Press.

Konner, M. (1982). The tangled wing. New York: Holt, Rinehart and Winston.

Lehrman, D.S. (1970). Semantic and conceptual issues in the nature-nurture problem. In L.R. Aronson, E. Tobach, D.S. Lehrman, & J.S. Rosenblatt (Eds.), Development and evolution of behavior: Essays in memory ofT.C. Schneirla. San Francisco, CA: Freeman.

Lerner, R.M. (1978). Nature, nurture and dynamic interactionism. Human Development, 21, 1-20.

Lerner, R.M. (1984). On the nature of human plasticity. New York: Cambridge University Press.

Lerner, R.M. (1986). Concepts and theories of human development (2nd ed.). New York: Random House.

Lerner, R.M. (1991). Changing organism-context relations as the basic process of development: A developmental contextual perspective. Developmental Psychology, 27, 27-32.

Lerner, R.M. (in press). Final solutions: Biology, prejudice, and genocide. University Park, PA: Pennsylvania State University Press.

Lerner, R.M., & Kauffman, M.B. (1985). The concept of development in contextual- ism. Developmental Review, 5, 309-333.

Lewontin, R.C. (1981). On constraints and adaptation. Behavioral and Brain Sciences, 4, 244-245.

Lewontin, R.C., Rose, S., Sc Kamin, L.|. (1984). Not in our genes: Biology, ideology, and human nature. New York: Pantheon Books.

Lorenz, K. (1940). Durch Domestikation verursachte Storungen arteigenen Verhaltens. Zeitscltrift fur angewandte Psychologic urtd Charakterkunde, 59, 2-81.

Lorenz, K. (1966). On aggression. New York: Harcourt, Brace & World.

Lumsden, C.J., Sc Wilson, E.O. (1981). Genes, mind, and culture. Cambridge, MA: Harvard University Press.

MacDonald, K.B. (Ed.). (1988). Sociobiological perspectives on human development. New York: Springer-Verlag.

McClearn, G.E. (1981). Evolution and genetic variability. In E.S. Golin (Ed.), Developmental plasticity: Behavioral and biological aspects of variations in development (pp. 3-31). New York: Academic Press.

McGuire, T.R., Sc Hirsch, |. (1977). General intelligence (g) and heritability (H2, h2). In I.C. Uzgiris Sc F Weizmann (Eds.), The structuring of experience (pp. 25-72). New York: Plenum.

Partanen, J., Brunn, K., & Markkanen, T. (1966). Inheritance of drinking. The Finnish Foundation for Alcohol Studies, 14.

Philipp, E.E. (1973). Discussion in Dtw and ethics of AID and embryo transfer (p. 66). Ciba Foundation Symposium 17 (new series). Amsterdam: Elsevier, Excerpta Medica, North Holland.

Piaget, J. (1970). Piagets theory. In PH. Mussen (Ed.), Carmichael’s manual of child psychology (3rd ed., Vol. 1, pp. 703-732). New York: Wiley.

Piaget, J. (1979). Relations between psychology and other sciences. Annual Review of Psychology, 30, 1-8.

Pianka, E.R. (1978). Evolutionary ecology. New York: Harper & Row.

Rushton, J.P. (1987). An evolutionary theory of health, longevity, and personality: Sociobiology and r/k reproductive strategies. Psychological Reports, 60, 539-549.

Rushton, J.P. (1988a). Race differences in behavior: A review and evolutionary analysis. Personality and Individual Differences, 9, 1009-1024.

Rushton, I.P. (1988b). Do r/к reproductive strategies apply to human differences? Social Biology, 35, 337-340.

Scarr-Salapatek, S. (1971). Unknowns in the IQ equation. Science, 174, 1223-1228.

Schneirla, T.C. (1957). The concept of development in comparative psychology. In D.B. Harris (Ed.), The concept of development (pp. 78-108). Minneapolis: University of Minnesota Press.

Sunday, S.R., tk Tobach, E. (Eds.). (1985). Violence against women: A critique of the sociobiology of rape. New York: Gordian Press.

Tobach, E., Gianutsos, J.. Topoff, H.R., & Gross, C.G. (1974). The four horses: Racism, sexism, militarism, and social Darwinism. New York: Behavioral Publications.

van den Berghe, P.L., & Barash, D.P. (1977). Inclusive fitness and human family structure. American Anthropologist, 79, 809-823.

von Bertalanfty, L. (1969). Chance or law. In A. Koestler (Ed.), Beyond reductionism. London: Hitchinson.

Walsten, D. (1990). Insensitivity of the analysis of variance to heredity-environment interaction. Behavioral and Brain Sciences, 13, 109-120.

Williams, G.C. (1966). Adaptation and natural selection. Princeton, NJ: Princeton University Press.

Wilson, E.O. (1975a). Sociobiology: The new synthesis. Cambridge, MA: Harvard University Press.

Wilson, E.O. (1975b). Human decency is animal. The New York Times Magazine, October 12.

Wilson, E.O. (1978). On human nature. Cambridge, MA: Harvard University Press.

Wilson, E.O. (1980). A consideration of the genetic foundation of human social behavior. In G.W. Barlow tk J. Silverberg (Eds.), Sociobiology: Beyond nature/nurture (pp. 295- 305). Boulder, CO: West View Press.

<<   CONTENTS   >>

Related topics