Steven Goldberg: Logical Analysis of Social Questions

Many years ago William Morrow and Sons published my The Inevitability of Patriarchy. Open Court recently published a very-much updated edition under the title Why Men Rule, which provides much additional evidence, response to critics, and the like.

The central argument made in Why Men Rule (and in a summary article in my When Wish Replaces Thought ) is this:

Specifiable hereditary psychophysiological differences between males and females engender in males a more-easily-released tendency for dominance behavior. This is observed by a society’s population and is incorporated in all aspects of socialization that mediate the psychophysiological and the institutional. As a result all societies, without exception, exhibit patriarchy, male status attainment, and male dominance.

In other words, the social is a dependent variable that gets its limits and direction from the independent variable of the psychophysiological.

Demonstrating and explaining the universality of these institutions is the purpose of Why Men Rule and the summary article mentioned above.

I cannot here provide the mass of anthropological evidence of universality, nor that of the psychophysiological evidence supporting the theory presented in my book. My purpose here is merely to present an analysis of the logic infusing the theory and a compendium of the fallacies of arguments denying the role of psychophysiological differentiation in setting limits on social possibility.

It is worth mentioning that many scientists and other academic readers have posted admirable summaries of my book on Wikipedia, only to have these deleted within days. (Wikipedia’s concept of democracy seems to be that one can delete any material that he or she doesn’t like; ideologues seem to prefer deletion to debate.)

I apologize for the fact that no one could refer to reading the essay that follows as “curling up with a good read.” Responding point by point to logical fallacies does not permit this. I think that the reader will find my books much more easy-going.

Much of my career has been devoted to discovering, demonstrating, and explaining the universality of—the presence in every society that has ever existed—certain sexually-differentiated institutions.

A. (Patriarchy) The upper positions of the hierarchies of every one of the thousands of societies on which we have any significant evidence are overwhelmingly filled by men (patriarchy). A Queen Victoria or a  Golda Meir is always an exception in her society and is always surrounded by a  government of men. (There were more female heads-of-state, queens when no royal male was available,  in the first two-thirds of the sixteenth century than the first two-thirds of the twentieth. There has never been a “matriarchy” or “Amazonian society.”(There have been a very few, tiny societies with relatively little  hierarchy, but in all such societies an informal male dominance played a role  similar to that of patriarchy.)

B. (Male Status Attainment) The highest-status (non-maternal) roles are  occupied primarily by males. The high-status roles are high-status not primarily because they are male (ditch-digging is male), but because they have high status. This  high status elicits from males, more strongly than from females, the behavior  required to attain the status. Which roles are given high status and which  behavior is required to attain these roles is, let us agree for argument’s sake, socially determined. But the greater impulse to do whatever is necessary to  attain status—sacrifice health, safety, family, care of the infant, pleasure, and the like,-- for whichever roles are given high status is a function of male physiology (just as similar impulses lead men to more readily sacrifice these for hierarchical attainment and dominance in male—female encounters and relationships.).

C. (Male Dominance) Both men and women feel that the dominance resides in the  male and that the woman must “get around” the man to attain her way. Even when  male dominance is absent from law (as in the United States) or formal custom (as in  “chivalrous” societies), the expectation is still one of male dominance. This is  attested to in the U.S., for example, by the feminist’s detestation of male  dominance and her incorrect attempt to explain it in purely social terms. One might well argue that, in some families in all societies, some women have more power—the ability to “get around” men to “get their way” by cleverness, persuasiveness, etc,—but what is universal is the feeling—expressed in adages, songs, etc. of every society--that there is a male dominance that must  be “gotten around.” This reality may be abhorred  or favored. (A preference for men who “take the lead” was prevalent seventy years ago, but it is not the values or attitudes that are primarily causal of dominance.)

These institutions are universal, found in every society that has ever existed. In an Addendum to Why Men Rule I present quotations from the ethnographies of every society ever claimed (never by the ethnographer, but always by a third-hand source) to represent an exception, a society lacking one of these institutions. These quotations make clear that not one of the societies even begins to represent an exception. For forty years I have challenged professional anthropologists and archeologists to risk their reputations by specifying a society lacking any of these institutions. None has done so.

Of course there is variation in the manifestation of these three institutions.  They are different, for example, in the United States and Saudi Arabia. But all societies— whether Christian or pagan, etc.; capitalist or communal, etc.; “stone age” or  modern industrial, etc. exhibit the institutions.  In other words, whatever the variation in the institutions, they always follow the same direction and fall within the same limits.

And there are, of course, in every society many individuals who are exceptions with reference to the statistical behavior manifested in the universal institutions. For example, there are some women who manifest a more easily-released dominance behavior than do some men. But the universal institutions—and the  expectations relevant to them—are manifestations of a population’s observation of the statistical reality; a “social  law of large numbers” guarantees that, for example, dominance will be associated with males. This will tend to make the statistical observation absolute (so that the statistical reality[1] that males usually exhibit more easily-released dominance behavior may become the belief that “men are aggressive and women are passive.” This no doubt can lead to discrimination against the person who is an exception for his or her sex, but the discrimination gets its direction from the  fact that the behavior discriminated against runs counter to that which is observed in most males and females.

Even if there were no biological evidence of psychophysiological differences setting limits and direction on differing male and female emotion and behavior, this universality and parsimony would force us to posit a male-female psychophysiologial differentiation that explains the societal  universality, an innate difference between males and females that sets limits and gives direction to institutional possibility.   

But the fact is that there is an enormous amount of such direct evidence (which is summarized in my book), evidence that dominance behavior is more easily-released from males by the appropriate environmental stimuli (hierarchy, status, or member of the other sex)[2] and that institutions reflect and incorporate a population’s observation of the sexual difference in dominance behavior.

1. “Socialization explains our expectations of males and females and the male-female differences in behavior (cognition, emotion, and action).” There are two fatal problems with this claim:

(A) Socialization does not explain anything, but merely forces us to ask another question: why does socialization of men and women always work in the same direction (with reference to the behaviors and institutions we discuss? Just as the male's greater physical strength is not caused primarily by our telling little girls that the men are physically stronger than women, so, too, is the male's physiologically-based, more readily elicited dominance behavior---the factor relevant to the institutions we discuss-- not caused primarily by the socialization. Socialization may often increase sex differences, but it is not the primary cause of them. 

This point does not demonstrate that the socialization relevant to dominance behavior does reflect physiological factors. But that is not its purpose. The point is made merely to make clear that socialization is always a mediator, whether the primary causes be physiological, economic, or whatever, so that the presence of socialization in no way conflicts with the claim of the primary importance of inherent sex differences.  In other words, one must ask the question: why has the socialization of every society that has ever existed associated dominance behavior with males?

(B) The second problem with the explanation in terms of socialization is, as mentioned above, its implicit, incorrect, assumption that the social environment of expectations, customs, norms, institutions, and the like is an independent variable capable of acting as counterpoise to the physiological constituents that make us male and female.

If the association of sex and behavioral characteristic were a variable independent of a population’s observation of psychophysiological reality then, at least in principle, socialization could act as counterpoise to hereditary sex differences. For example, a society could, by having women lift weights throughout life and men remain sedentary, balance the male’s inherent strength advantage. (Note that physical strength is simply an analogy, not the relevant factor, which is dominance tendency.)

But in real life this can't happen because the social environment is a dependent variable whose limits are set by our psychophysiological construction and a population’s observation of the behavior related to it. In real life a population's observation of the relative physical strength of men and women precludes the possibility that expectation, socialization, and practice will balance the male’s greater inherent strength and will result in institutions rendering women as physically strong as men. Likewise, in real life a population observes the male’s dominance tendency and develops expectations and socialization concordant with this.

While all biologists responding to the theory I present granted the causal role of psychophysiological differentiation in sexual differentiation, one claimed that a change in social environment could balance the physiological differentiation. Using the analogy of phenylketonuria, a genetically-based disease whose symptoms are manifested in an environment in which some foods are eaten, but not in an environment in which other foods are eaten, the biologist claimed that social values could have an equivalent effect.

The problem with this analogy is that diet is independent of the genetic reality; the gene for phenylketonuria does not engender a "motivation" to desire phenylketonuria. There  is no gene militating against an individual's eating a diet that will prevent the phenylketonuria. But every society’s environment provides the social values that are the analog of the foods that potentiate phenylketonuria. They do so because they conform to the population’s observation of the physiologically-generated male behavior. In the case of the male and female differences we discuss, it is precisely the difference in "motivation"—more rigorously, the male’s lower threshold for the elicitation of dominance behavior by appropriate environmental stimuli--that is a function of the physiological differences. A society's norms and values could not, for example, reflect an equal male and female dominance tendency (or tendency to violence or physical strength or immediacy of sexual arousal); the norms and values must fall within the limits set by the psychophysiological differences between males and females and the population's observations of the differentiated male and female behavior.

In short, the problem with this analogy is central to even the best of the overly environmentalist analyses. While the analogy does “demonstrate” that, in principal, environment can overcome psychophysiological tendency (as it does our fantastical society in which only women do strengthening exercises), it fails to analogize the dependence of the variable in social situations. And it is this dependence that is the reality that precludes social values offsetting the psychophysiological”

Note that it is not so much that men (necessarily) limit women’s accession to dominance in hierarchies in any direct way. The limitation is primarily a side effect of the male’s greater “need” of dominance and the behavior this engenders (just as the absence of women from the best basketball leagues is not primarily the result of discrimination, but the inevitable rise to the top of the best players). In the case of hierarchies, it is not necessarily that males do the job better, but that they do what is necessary to attain the positions. And you can’t be a good or bad Senator until you become a Senator. To be sure, this reality is inevitably manifested in social values that increase sex differences in attainment and can lead to discrimination.  (Some societies preclude women’s entering the hierarchies altogether. But, as always, the question remains: Why in every society is it males who dominate the hierarchies? Why has there never been a matriarchy or “equiarchy”?

2. “The physiological theory of limits is “reductionist.” The problem with this criticism is that a scientific explanation is supposed to be reductionist (parsimonious), if by that we mean “capable of explaining the most empirical reality with the fewest hypotheses”. “Reductionism” is impotent as criticism unless the criticized analysis attempts to explain more than its explanatory mechanism is capable of explaining. “Reductionism” would be a legitimate charge if, for example, it were claimed that physiology explains the difference between women’s roles in the United States and Saudi Arabia. But my theory of constraints on social possibility is a theory of limits; it makes no claim of explaining any of the variation within the limits (i.e., any variation found from one actual society to another). A criticism of “reductionism” here is akin to one denying the physiological basis of the human need to eat (and the universality of institutions satisfying this) on the grounds that the explanation does not tell us why the French eat French food and the Chinese eat Chinese food, why societies have different numbers of meals per day, or why some societies associate food with religion far more than do others. The physiological explanation does not claim to explain this variation. In short, the criticism of “reductionism” is analogous to an accusation of “sophistry” that fails to specify any logical fallacy. It is mere name-calling.

3. “We define ‘patriarchy’ and/or ‘dominance’  differently from the way you do. In various anthropological writings, even those predating the ideologically-infused works of the past decades, one can find at least twenty varying definitions of “patriarchy”. The one I use is both that which is most often used and the common denominator of most of the other definitions.

But a far more important point is this: It is the empirical reality--not the word one uses to represent it--that is crucial. As long as one uses consistently the word he has chosen, the specific word chosen is unimportant. If you wish to call those gigantic gray animals with long trunks and skinny tails “toasters”, you can. But you can’t then claim that these “toasters” are good for making English Muffins.

Likewise: The empirical reality is that the hierarchies of every society without exception are filled primarily by males . If one objects to my terming this “patriarchy”--fine. Choose any other word, say. “toaster”. One must then explain why every society has “toaster”. One can’t make an empirical reality disappear with definitional fancy footwork. All of this can be said of the “redefinition of ‘dominance’.”

4. “What about this” non- “exception As I have mentioned, most attempts to provide a society lacking the institutions we discuss are invocations, based on third-hand sources, of societies for which recourse to the original ethnography demonstrates the claims to be risibly unwarranted. These usually simply assert exceptionality without giving any details (for obvious reasons).

Other attempts invoke a factor having nothing to do with the institutions we discuss. For example:

The “fact” that some societies have a highest god who is female. It is far from clear that this is a fact, that there has been such a society. But let us, for argument’s sake, assume that there is. All this would demonstrate is that that religion is of little importance to patriarchy (since such societies all exhibit patriarchy) and that the universality of patriarchy must now be explained without reference to religion.

 Often the claim that there is an exception refuses to specify the societal exception, but merely asserts that there is one. For example, the claim by  Richard Lewontin, Steven Rose, and Leon Kamin (in Not in Our Genes) that:

  "[Cross-cultural universals]   . . appear to lie more in the eye of the beholder than in the social reality that is being observed" is utterly untrue.

Indeed, to make their point the authors would merely have had to specify a society lacking one of the institutions I find to be universal. They can not do this because there is no exception

5. “The explanation fails to understand the “complexity of social life” and “the tremendous variation among societies.” AND “The explanation fails to understand the complexity of the mechanisms mediating neuroendocrinological differences between males and females, the behavioral differences, and the social differences to which the behavioral differences are relevant”. The ”complexity” and “variation” invoked here are irrelevant. No society is “so complex” that it lacks the universal institutions. There is not so much variation that any society manages to escape the constraints of the limits discussed here. The neuroendocrinological mechanism and its mediators to the behavioral and social are not so complex that they ever result in a society lacking the institutions whose universality’s explanation is our very purpose.

The neuroendocrinological explanation of universality is a sufficient explanation of the limits within which social variation and complexity take place. The issue of "complexity" is simply another version of the "Chinese food" attempt to obfuscate with irrelevant empirical realities that the theory presented here does not attempt to explain. (Similarly, many authors devote a great deal of space to arguing against the importance of  neuroendocrinological factors other than those that alone are sufficient to explain the universality; such arguments are irrelevant.)

6. “We have patriarchy for economic reasons.” This is a confusion of cause and function. The realities I discuss no doubt have important economic functions. But to ascribe patriarchy to economic factors is akin to ascribing the human need to eat to McDonald’s need to make a profit. At least with reference to sex differences, economies primarily exploit our natures, not cause them. That is why every economic system--communal, slave, feudal, capitalist, socialist, religious, etc.--works within the limits of patriarchy. This confusion of economic cause and economic function  is the explicit or implicit fallacy underlying  a host of social science works on patriarchy ranging from Gerda Lerner’s The Creation of Patriarchy to Nathalie Angier’s Women: An Intimate Geography. 

7. “Patriarchy is a result of the requirements of a hunting culture, or Christianity, or capitalism, etc.” If it is to be at all persuasive, an explanation of universality must be parsimonious; the explanation must invoke a causal factor common to the varying societies that exhibit the universal institution. Just as the explanation in terms of capitalism fails to explain patriarchy in all of the  non-capitalist societies, so do explanations in terms of any single factor other than the psychophysiological fail to explain the host of societies for which that factor does not apply. Non-hunting, non-Christian, non-capitalist, etc. societies are all patriarchal.

A single-cause theory of the limits constraining every society need not, of course, be the neuroendocrinological one I suggest. But the few alternative parsimonious explanations fail on empirical grounds. For example:

8. “Societies are patriarchal because women are tied down with giving birth and raising children and because men are bigger than women.” We can ignore here the fact that physiology accounts for the fact that women bear and raise children, because there are many societies in which women work harder and longer outside the home--doing objectively more important economic work--than do men, (though, as we have seen, however objectively unimportant the roles played by men, some of these roles will be given higher status than any other non-maternal roles). Whatever the non-maternal roles played by women, these never include primary responsibility for hierarchical position. Similarly, while males are everywhere bigger and stronger, all evidence from both human beings and experimental animals imply that it is the Central Nervous System  (CNS) difference relevant to dominance behavior, not physical size, that is primarily responsible. A one-generation, experimentally-created society stocked with infant daughters of very large parents and infant sons of very small parents would develop into a patriarchy of small men and large women.

9. “The myths of ‘ancient matriarchies’ and ‘Amazonian societies’ must be taken seriously as perhaps demonstrating that there were such societies.” Even ignoring the fact that such myths were refuted by writers of the time they were alleged to exist, assuming that there were matriarchies just because there are myths about matriarchies makes as much sense as assuming that there were cyclopses just because there were myths about cyclopses.

10. “We have patriarchy because we have ‘patriarchal values’.” All societies are patriarchal and all have patriarchal values. Just as a population’s observation of women’s physiologically-rooted maternal behavior explains why maternal values and expectations are always associated with women, so too does observation of male’s physiologically-rooted dominance behavior explain why dominance values and expectations are associated with men. Those who “explain” patriarchy in terms of “patriarchal values” must, without invoking physiological differentiation, explain why every society has patriarchal values.

11 “Attitudes have changed tremendously.” Yes, they have, at least those people claim to believe. (Behavior often belies the claims.) But the very point is that, with reference to the behaviors relevant here, attitudes are not all that much more causally important than they are to the sex difference in height.

For much behavior, of course, attitude is crucial. The decline of the sanction against premarital sexual activity has resulted in a lot more premarital sexual activity. But, while attitude is dramatically important in how people claim to feel about the sex differences of which we speak, there is no evidence whatever that they reduce those sex differences, certainly not to the extent that they imply the possibility of a society lacking the universal institutions.

12. ”Modernization and Technology Render Physiology Irrelevant.” There is not a scintilla of evidence that modernization renders likely the demise of the universals. To be sure, no modern society could preclude women’s playing any suprafamilial role as some non-modern societies did. But it is also true that no modern society is likely to give women the high status some other (matrilineal-matrilocal, but patriarchal) non-modern societies gave the woman’s maternal roles. In any case, even the Scandinavian societies often claimed to be “non-patriarchal”--called this despite the fact that they feel the need of cabinet departments to deal with the “inequality of women”--are, in fact, overwhelmingly patriarchal. (An interesting fact about the Scandinavian countries is that, some political scientists argue, the political plays a  less-important role than does the corporate, relative to other countries. While female membership of parliament is the highest in the world (though still far from equal), male control of the corporate world is absolute; there is no corporate “glass ceiling” issue because hardly any women rise high enough to see the “glass ceiling”. Perhaps the Scandinavian nations, which have before augured the future of bureaucratic societies, here also intimate coming realities in an increasingly global world.)

One can not, of course, prove that there will never be a non-patriarchal society. It is in the nature of science that one can never prove anything; such proof would preclude the requirement of prediction that defines science.

However, the likelihood that something will happen must be based on our knowledge, and our anthropological and physiological knowledge indicates the impossibility of a non-patriarchal society. It is not coincidental that no one argues for the possibility of, say, a society in which ninety percent of the males are celibate. Such a society is theoretically as possible as is a non-patriarchal society, but no one wishes for such a society, so no one makes a prediction so implausible given our knowledge of males physiology.

13. “Slavery was universal.” No it wasn’t. Many societies never had slavery and only one society lacking slavery is necessary to demonstrate that physiology does not render slavery inevitable. Had slavery been universal, this would not demonstrate slavery to be inevitable, but it would certainly make it likely

14. “We won’t know whether there could be a non-patriarchal society until we have one”(a fallacy first invoked by John Stuart Mill). We won’t have one if there could not be one. Must we refrain from saying that in any society composed of men and women it will be the latter who give birth until we have a society in which men give birth?

15. “Gender identity--our sense of our own maleness or femaleness--is purely determined by familial factors and socialization.” No it’s not, but let us assume that it is. Sex-associated behavior is not. Whether the hormonally feminized chromosomal male sees him/herself as an “aggressive” female or as a  male is irrelevant; it is the CNS-behavior correlation that is relevant here. For nearly all people, of course, there is a concordance of genetic, chromosomal, hormonal, anatomical, and social development with gender identity.

16. “Sociobiological and evolutionary theories are highly speculative and the ethological study of other primates encounters problems of anthropomorphism.  For argument’s sake, let us accept all this.  I offer no explanation why human male and female physiologies evolved the way they did, but take these as given. (However, it would be a strange evolution indeed that associated aggression with females. The loss of ninety percent of the males would be unimportant; the remaining lucky few would guarantee continuation of the population at its former size. Every lost female is a disaster.) Likewise, the theory presented does not invoke any ethological evidence--though such evidence strongly supports the theory.

Similarly, it is often argued that “environment determines physiology in evolution.” This is, of course, true over sufficiently long periods of time. In a sense, environment determined the physiological differences between an amoeba and a human being and, who knows, may eventually give descendants of humans beings the ability of fly. Should evolution eradicate the physiological differences between males and females—an eventuality likely to take quite some time, given the intimate connection of female physiology and the female’s ability to give birth-- then we will not be dealing with “males” and “females” as we know them. Every scientific hypothesis has some ceteris paribus limit and I don’t think this one is unreasonable.  (As noted, no one objects to a claim that it is inevitable that every society permits sexual activity far above that necessary for replacement of the population; sex is popular these days, while male dominance is not.)

In short, invocation of vague allusions to the (very) long-term causal effect of  environment in an attempt to minimize the importance of sex differences attempt to equate social environmental changes that take only hundreds years with physical environment changes that take millions and millions of years. This is not very persuasive.

17. “Boys and girls have equal levels of the male hormone, but boys are more aggressive. This shows that the behavior is a function of socialization.” No it doesn’t. The real, but less interesting, explanation is that it is simplistic to speak only of hormone levels; it is the fetal sensitization of the male CNS to the relevant properties of testosterone that is relevant.

But, even were this not the case, the implication would be that the socialization of boys and girls anticipates the pubertal physiological reality (when the male testosterone level is much higher). The reason men can grow moustaches is not that we tell little girls that facial hair is unfeminine. (And there is also the inevitable question:  the testosterone levels of boys and girls are equal in every society; why, in every society, is it the boys who are “more aggressive”?)

18. “Hormones are ‘suggestible’ and can have their behavioral effects determined by socialization.” Again, a tremendous oversimplification, but let that go. The more serious question is why every society without exception suggests male aggression. Once one requires a parsimonious answer to this question, the suggestibility criticism evaporates in the same way that the socialization criticism did.

A similar question is raised by the claim that environment can affect hormone levels (though not to anywhere near the extent that it equalizes the male-female levels or overcomes the male’s greater CNS sensitivity to the hormones); why is it always the hormone of male aggression that is suggested by the environment?

19. “You claim patriarchy is inevitable. Science never dismisses a possibility.” Of course it does, and should. Every hypothesis should specify things that won’t happen. It is only by doing this that we have any way of telling whether the hypothesis is likely to be correct. What science does not ever dismiss is an empirical reality that actually exists. Should a non-patriarchal, hierarchical society be found to have existed, presently exist, or come to exist, I will be the first to jettison the theory I present. But the hope that this will happen does not qualify.

(I originally titled my book, The Inevitability of Patriarchy precisely because I wanted to emphasize in the very title that the theory specifies the conditions under which it could be refuted: the discovery or development of a non-patriarchal society would immediately refute the theory. I still prefer this title, but Open Court preferred Why Men Rule.)

A similar criticism claims that I argue that “patriarchy is inevitable because it is universal.” No. Universality leaves open the possibility of inevitability (which an exception would preclude) and forces us to assess the likelihood of inevitability on the basis of the cause of the universality. Moreover, in a world of thousands of societies with unimaginable variation, universality demands that we consider the possibility that the universal is rooted in the biological nature of human beings or in the very nature of society, any society. When universality is complemented by an enormous amount of physiological evidence capable of explaining the universality, the likelihood that the limits are manifestation of the psychophysiological is overwhelming.

Cultural anthropology has given us an indescribably precious gift by demonstrating the astounding variation human societies have exhibited. In five hundred years, in a world likely to be far more homogenous than is the contemporary world, the ethnographies of all of the varied societies will act as counterpoise to our inherent ethnocentricity and will show that things can be different. 

This is, however, a two-edged sword. Given this societal variation, one must ask why certain institutions are invulnerable to the forces of variation, why they exist no matter how otherwise varied the societies.

 Indeed, the very variation in other institutions forces us to ask why certain institutions are never absent. The institutions we discuss here are found in all societies--be they past or present; primitive, pre-industrial, or modern; sun-worshipper or Christian; relatively communal or ruthlessly individualistic; slave or free; socialist or capitalist; primarily meat-eating or primarily plant-eating, etc, etc...

In short: Given the astounding degree of variation societies have demonstrated, the universality of the institutions we discuss must be explained and, as we have seen, the explanation must be parsimonious. And the only explanation of universality that is parsimonious, logical, concordant with the anthropological and physiological evidence, and plausible is one that understands that the institutions are not inevitable because they are universal; they are inevitable for the same reason that they are universal.

20. “There are studies that show that there is no difference in dominance behavior between men and women.” There is a common misconception that a study or experiment that finds something is refuted by one using a different methodology that doesn’t. This is roughly analogous to the argument that ‘You have five witnesses who saw my client commit the crime, but I have six who didn’t.’ If I measure the height of men and women with a yardstick that measures only to the closest yard, and you do so with a yardstick that measures to the closest inch, my conclusion that men and women are of equal height does not refute your conclusion that men are taller or your explanation that sees this as a result of heredity.

21. “Differences within-group (i.e., among males or among females) are greater than between-group (i.e., between males and females) and there are lots of exceptions (members of one sex that more strongly exhibit behavior associated with the other sex than do some members of the other sex).” This is, of course, true; the range within either of two groups is almost always greater than the difference in means of the two groups. (E.g., the five-inch difference in the mean heights of men and women is tiny compared to the difference in height between the shortest woman and tallest woman or shortest man and tallest man). Likewise, there are virtually always exceptions (i.e., the very tall woman or the woman with very easily elicited dominance behavior). But it is on the general statistical observation that expectations are based.

Another important statistical-empirical point is worth making here: For virtually every characteristic, variation among males is far greater than among females. Whatever the characteristic, males exhibit the most and the least of that characteristic. Thus, even when a characteristic is, on average, more associated with females, those most exhibiting the characteristic tend to be males, and even when a characteristic is, on average, more associated with males, those least exhibiting the characteristic also tend to be males. Why there is this greater male variation is unclear, but it is probably related to the fact that masculinization requires a fetal endocrinological stage that feminization does not. Socially, there is an asymmetry here: while the most and the least will be exhibited by males, a population notices only the former. Thus, for example, while women, on average, probably have a superior verbal aptitude, and certainly have greater psychological insight, than do men, a disproportionate number of the great novelists and psychologists have been male. The fact that a disproportionate number of the least literate or insightful people are male tends to go unnoticed, or at least is considered unimportant. (On the other hand, the male superiority at mathematics, chess, and composing music--hardly a macho role that boys are encouraged to pursue--is owing less to the variation discussed here than to a strong male superiority at the relevant aptitudes for spatial relations, mathematics, and logic, a superiority that manifests itself in a huge sex difference at the upper levels. 

22. “Female behavior X is increasing at a much faster rate than is Male Behavior X.” Tendentious discussions of sex differences often compare increases when these are unimportant for the purposes for which they are invoked. Tiny numbers will always increase much faster than will huge numbers. For example, one often reads about the “tremendous” increase in violent female crime”. This is legitimate if one’s interest is only in violent female crime. But the increase is often invoked in an attempt to cast doubt on the male-female difference in crime or the relevance of inherent importance of physiological differences to these. This use is illegitimate; the proportion of violent crime committed by females makes a drop-in-the-bucket seem like a drop-in-the-ocean.   

23. “The author is a sexist and the effects of his work will be politically bad.  The inadequacy of the ad hominem and ad consequentium arguments has been known for millennia. Even if these charges were true, they would be irrelevant. If biases infect an analysis, the effects on the analysis can be exposed. If biases do not infect an analysis, it does not matter how biased the author is. In neither case is any social or political consequence of the  analysis relevant to the correctness of the analysis.

A Note on Discrimination: If even such clearly hereditary properties as height permit many individual exceptions, we would expect--and find--even more such exceptions when characteristic in question can be caused also by, for example, an unusual familial psychological environment. But such factors are always exceptional and never sufficient to overcome the general statistical reality. (If they were, there would not be the general statistical reality.)

However, it is important to acknowledge that the individual exceptions often do encounter harmful discrimination and that society’s making the statistical absolute can generate greater sex differences than would heredity alone. But the point relevant here is that, with reference to sex roles, the discrimination is possible precisely because the exception is an exception and the exception is an exception precisely because physiology associates the expected characteristic with the non-exception. It is the very tall woman (or very short man) who encounters discrimination where the equally tall man (or short woman) does not.

A Note on Plausibility: Once the logic of a scientific explanation is demonstrated to be flawless and the empirical evidence for it is accepted, even then one’s acceptance of the explanation is a function of plausibility. And there is no irresistible method of demonstrating plausibility, no demonstration equivalent to exposure of fallacy or refutation by experiment.

Usually this is not much of a problem. The ability of one of two competing explanations to explain new empirical evidence smoothly, while the alternative must invoke increasingly unpersuasive, ad hoc arguments if it is to maintain its congruence with logic and experiment, renders the superiority of the former obvious. Soon all those not involved in the creation of the inferior explanation accept the greater plausibility of the superior explanation.

However, powerful psychological and ideological need can pervert one’s sense of plausibility and lead one to invoke increasingly bizarre arguments in order to maintain belief in the implausible explanation. Complementing the implausibility is often a demand that the more plausible explanation describe the mechanism at work down to the quantum level. Thus, for example, those who despise the causal explanation presented here will make arguments they would never make if the issue were the “sex drive”;  no one would argue that, despite the “sex drive”, we should expect a society in which no one had intercourse more than once a month. The implausibility of such a view would be instantly granted. But when the issue is the inherent differing hierarchy of motivations of males and females, they wish for an eradication of these that is as implausible as a wish can get.

Just as one can’t prove anything in science (the possibility of an empirical event that refutes an explanation must always be left open), one can’t demonstrate that a view is implausible. Plausibility must be decided by the reader.

I suspect that much of the impulse energizing such denial is the unwarranted fear that acceptance of the explanation of universality I offer would commit one to a moral or political view he or she finds abhorrent. This fear is unwarranted. No scientific explanation of how the world works can tell us how we should politically or morally act. Science knows nothing of "should." So, for example, one could agree with all that I have written here and argue that the theory presented here indicates the crucial importance of an equal rights amendment limiting as much as possible a male advan­tage in attaining positions, an advantage that often has nothing to do with performance in those positions. On the other hand, one could agree with all that I have written and argue that this indicates the need, in a time when role models are so hard to come by, for our emphasiz­ing differences between male and female tendencies and their ability to form the nuclei of strong roles and role models. "Is cannot generate 'ought". On the issue of good and bad, right and wrong, science must be silent.

[1] The height of men and women is always the model to keep in mind when considering male—female behavioral differences. There are many individual “exceptions,” many women who are taller than many men, but the mean height of men is always greater and no one can argue that exceptions belie the claim that “men are taller than women.” With all sex differences it is the modal difference that a population notices and which determines expectation and institutions reflecting them. Any discussion of sex differences that is not founded on an understanding of the statistical, rather than absolute, nature of individuals’ sex differences is doomed to incoherence.

[2] In general discussions of sex differences in hierarchy of emotions, desires, etc. it is tempting to use terms like “need,” “drive,” and “motivation. While these terms differ in connotation—“need,” for example, brings to mind vulnerability and “drive” brings to mind power, such terms are unnecessary and best avoided in rigorous work. Perhaps the best way to picture the role of psychophysiological tendencies is “threshold for the release of the tendency X when exposed to environmental stimulus Y. Think of iron and magnet. Iron does not have a “need” or “drive” to find a magnet, but a physical makeup that has it react in a certain way when exposed to a magnet.