The Skeptics Society & Skeptic magazine


In Defense of the Bell Curve:
The Reality of Race and the Importance of Human Differences

Last week, we republished the cognitive psychologist Diane Halpern’s critique of Charles Murray and Richard Herrnstein’s book The Bell Curve: Intelligence and Class Structure in American Life. Dr. Halpern is one of the world’s leading scientists in the study of cognitive differences, including both race and gender, and she herself has encountered controversy for just daring to study the subject.

Below, we republish the late U.C. Berkeley anthropologist Vince Sarich’s article, “In Defense of The Bell Curve: The Reality of Race and the Importance of Human Differences.” Sarich was also a very controversial scientist for his research on race and group differences, from which he never backed down.

At Skeptic magazine, there are no sacred cows, no ideas too hot to touch. It is our belief that truth will win out in the bright light of open discussion, debate, and disputation. After reading both articles, judge for yourself how to think about the bell curve.

The Bell Curve and the many commentaries on it have brought several issues into an often uncomfortably sharp focus. Though race is by no means the most important of these, the historical baggage the term carries and the reality it symbolizes, require us to get past it before we are able to deal with more substantive matters. Yet that same baggage and those same realities often raise emotional barriers so powerful that they defy facts, reason, and logic.

Many commentators would have us believe that The Bell Curve is obsessed with race, and thereby provide a prime exemplar of pots, kettles, and blackness, evidenced in the following quote from the sociologist Alan Wolfe: “Murray and Herrnstein may not be racists, but they are obsessed by race. They see the world in group terms and must have data on group membership.” This is an interesting charge, says Charles Krauthammer (1994), “given the fact that for the last two decades it is the very liberals who so vehemently denounce Murray who have been obsessed by race, insisting that every institution—universities, fire departments, Alaskan canneries—must have data on group membership.”

We can begin this trip out of political correctness by noting that on genetic grounds alone there can be no doubt of the existence of a substantial number of human races. Races are, if you wish, fuzzy sets.

It is the liberals who have oppressively insisted that we measure ethnic “over-” and “underrepresentation” in every possible field of human endeavor. Here is a liberal establishment forcing racial testing for every conceivable activity, and when a study comes along which does exactly that … for SATs and IQ, the authors are pilloried for being obsessed by race.

No one who has actually read The Bell Curve could honestly document any such obsession. But, by the same token, no one even moderately conversant with the American society of the last 20 to 30 years could deny the accuracy of Krauthammer’s assertion that “it is the very liberals who so vehemently denounce Murray who have been obsessed by race.”

Further, it is these “very liberals” who deny that there is any significant genetic, biological, and evolutionary substance to race, and argue that it is, in effect, nothing more than a social and cultural construct. This view is epitomized in a recent story in Time (January 16, 1995), that carries the subtitle: “A landmark global study flattens The Bell Curve, proving that racial differences are only skin deep.” The reference is to The History and Geography of Human Genes, a recent, massive compilation and analysis of human gene frequency data (Cavalli-Sforza, et al., 1994). The story is an honest summation of that work—given that the genetic distances among human races are minimal, and that sections 1.5 and 1.6 of the book are entitled “Classical Attempts to Distinguish Human ‘Races’,” and “Scientific Failure of the Concept of Human Races.” One looks in vain, however, in both the Time piece and the book on which it is based for any definition of the term “race.” This omission is typical of race-debunking efforts. They never bother to define what it is that they are debunking. So let’s start there.

The Reality of Human Races

We can begin this trip out of political correctness by noting that there is a substantial amount of agreement on both a working definition of the term “race” and on the existence of races in species other than our own. Races are populations, or groups of populations, within a species, that are separated geographically from other such populations or groups of populations, and distinguishable from them on the basis of heritable features.

We can agree that we are all members of a single species—Homo sapiens—and that each of us is also a unique individual. The most basic evidence that races exist is the fact that we can look at individuals and place them, with some appreciable degree of accuracy, into the areas from which they or their recent ancestors derive. The process involved is illustrated by a thought experiment where one imagines a random assortment of 50 modern humans and 50 chimpanzees. No one, chimp or human, would have any difficulty in reconstituting the original 50 member sets by simple inspection. But the same would be true within our species with, say, 50 humans from Japan, 50 from Malawi, and 50 from Norway. Again, by simple inspection, we would achieve the same 100% sorting accuracy. Granted, in the second experiment fewer sorting characteristics were available, but not nearly so few as to produce any doubt as to the placement of any individual. Extending this look-see experiment to the whole of the human species would obviously give us a substantial number of such geographical groupings. The addition of direct genetic evidence—from blood groups to DNA sequences —would provide further resolving power. But there is a real problem here that goes well beyond ideology and political correctness.

The Nature of Categories

One might clarify the problem of defining groups by reference to the issue of color categorization. We know that speakers of various languages that have a term for “red” (and who also have a comparable number of basic color terms) will also show a remarkable degree of agreement as to the range of the spectrum to which the term applies, and as to which hues are better reds than others (Berlin and Kay, 1969). We look at a rainbow and we tend to see not continuity, but rather a small number of specific colors that we have no trouble naming. This example tells us that whatever may be going on with respect to cognitive processing of the visible light spectrum, we have no operational difficulties in at least this realm with the notion that categories do not have to be discrete. Red does shade imperceptibly into orange, and orange into yellow, but we have no difficulty in agreeing as to where red becomes orange, and orange, yellow. Thus, human cognition can handle categories that are not discrete. The flip side of that is that categories can be real without necessarily being enumerable— and that is the critical matter for this article.

In other words, we can easily forget that categories do not have to be discrete. If this were not so, then why should the notion of “fuzzy sets” been seen as so revolutionarily productive? Races are fuzzy sets.

How Many Race Are There?

One of the most commonly asked questions about race is: “How many races are there?” I contend that this is the wrong question. “How many” requires a precise integer as an answer— 3, 7, 15, whatever. But the nature of the category “race” is such as to make such an answer impossible, depending as it necessarily does on the degree of sorting accuracy required in a context where the categories involved are not discrete. Races, after all, are not species, since all humans are fully interfertile. Therefore, races must necessarily grade into one another. But they do not do so evenly. Even today, for example, to drive along the road north from Aswan to Luxor (a hundred miles or so) is to cross a portion of ancient boundary between, to use old anthropological terms, Caucasians and Negroes. These two large groupings have been separated for millennia by the Sahara Desert. The Sahara has caused the populations north and south of it to evolve in substantial genetic independence from one another. And that is all one needs for race formation— geographical separation plus time.

The race quantity answer depends on the degree of sorting accuracy with respect to individuals. If it is something close to 100%, then the areas involved could become smaller and more distant from one another, with at least 20 races easily recognized, or larger and less separated, in which case one would see the few “major” races that everyone has tended to see. If, however, the criterion were something more like the 75% which has often sufficed for the recognition of races in other species, then obviously the number would be very large. In either case, if we use a straightforward definition of race such as a population within a species that can be readily distinguished from other such populations using only heritable features, then there can be no doubt of the existence of a substantial number of human races. But, I hear you ask, don’t the races all blend into one another? Yes, they are supposed to blend into one another. That’s what races do. Nature’s categories need not be discrete. It is not for us to impose our cognitive limitations upon Nature.

The Cause of Racial Separation

If all that is needed for racial differentiation is geographic separation and time, then why have humans remained a single species? The answer almost certainly lies in the fact of glacial cycles throughout the existence of our genus. These have necessitated major movements of human populations at fairly frequent, if irregular, intervals throughout the million years or so that Homo has existed outside of sub-Saharan Africa and therefore been susceptible to differentiation into races. Thus, there would have been periods of relative glacial stability (such as the last 10,000 years or so) during which racial differentiation would have become more marked; and periods of glacial movement, such as the retreat which began about 18,000 years ago, during which gene flow would have pretty much obliterated the previously developed racial boundaries. This logic also leads to the conclusion that most existing racial variation must have developed since that last period of large-scale, world-wide gene flow; that is, over the last 15,000 or so years. There is extensive evidence at a number of disciplines—anatomy, linguistics, biochemistry, archeology—which is consistent with such a scenario. The most straightforward is the fact that Homo sapiens fossil skulls found in areas currently populated by “Caucasians” and ranging in age from about 15,000 to 30,000 years are not more similar to those of modern “Caucasians” than they are to those of other major racial groupings.

The question of the antiquity of human racial lineages remains one of the most controversial areas of human evolution. Basically two quite opposed views predominate, neither of which takes the fact of glacial cycles into account. (1) Regional Continuity or Multiregional Evolution. Homo erectus populations in different areas of the world are seen as having appreciable direct genetic continuity with modern populations in those same areas. This theory sees significant aspects of modern racial variability as having separate histories for the high hundreds of thousands of years. (2) Out of Africa or African Eve. Homo sapiens have a single, relatively recent (something around 100,000 years ago) origin in some limited area and are characterized by some novel adaptation which enabled them to expand out of that homeland, replacing the more primitive humans they found along the way. Racial differentiation then followed. Most people in the field have tended to see #1 as implying much more significant racial differences because they would have had longer to develop. This has also been a major factor contributing to its relative lack of support.

But, as the late Glynn Isaac (perhaps the most influential archeologist involved in studies of early Homo) pointed out to me in a Berkeley seminar many years ago, it is the Out of Africa model, not that of regional continuity, which makes racial differences more functionally significant. It does so because the amount of time involved in the raciation process is much smaller, while, obviously, the degree of racial differentiation is the same—large. The shorter the period of time required to produce a given amount of morphological difference, the more selectively important the differences become. The Out of Africa model in its earlier formulations envisioned perhaps 40,000 years for raciation of anatomically modern Homo sapiens. The current formulations would nearly triple that figure, and, thus reduce the implied significance of racial differences. Obviously the model I outlined above would do the opposite, increasing that significance well beyond anything contemplated in recent years. But that might not be all. During the last 10,000 years human cultures have differentiated to a much greater extent with respect to achievement than was the case previously. Thus, not only might the time involved for raciation have been brief, but the selective demands on human cognitive capacities might have differed regionally to a substantially greater extent than could have been the case previously (see Sarich, 1995, for an extended discussion of these matters).

How Large are Actual Racial Differences?

Current textbooks on human biology and human evolution go out of their way to deny either the reality, the significance, or both, of race in our species. Their efforts would appear to be based in the hope that if we can make races disappear, racism will follow. For example:

Race: In terms of biological variation, a group of populations sharing certain traits that make them different from other groups of populations. In practice, the concept of race is very difficult to apply to patterns of human variation.

The first sentence is fine. But the second implies that most human variation is not racially patterned. Which is certainly true. Most of the variation in our species, and in all other species, is found within and among individuals. But truth here has nothing to do with relevance. No one argues that race is the only dimension along which humans vary genetically. But, by the same token, there is more than enough heritable variation to produce human groupings which conform to any generally accepted definition of the term “race.” This fact tells us that a substantial amount of human variation is clearly racially distributed, and leads to the question of how different from one another human races are.

The answer is, it depends on what you are looking at. At the level of morphology human races are more strongly differentiated from one another than are any other mammalian species. I first became aware of this fact when considering the arguments in the anthropological literature as to the place of the Neandertals. There one would often see statements to the effect that “Neandertals are too different from us to be part of our evolutionary history,” but “too different” was never quantified. Quantifying it by using a standard set of measurements, correcting for size and calculating an average percent difference per measurement, gave some substance to the claim. Neandertals are, in fact, about twice as distant, on the average, from various extant human populations as the latter are from one another. But that exercise also demonstrated that (1) the anatomical distances among some modern races, for example, East Africans and Central Siberians, were much larger than those between Neandertals and the modern human populations most similar to them, and (2) racial morphological distances within our species are, on the average, about equal to the distances among species within other genera of mammals, as, for example, between pygmy and common chimpanzees. I am not aware of another mammalian species where the constituent races are as strongly marked as they are in ours.

The genetic distances are, in contrast, very small, and the no-races-in-our-species protagonists (such as Cavalli-Sforza) have seized on this fact to buttress their position. However, one needs to put the data into an evolutionary context to see what they really mean. The problem here lies in the fact that morphological evolution in our species has been extremely rapid, and this is not some sort of anthropocentric judgment. It can be demonstrated through two simple observations. We and our two closest living relatives—gorillas and chimpanzees— are about equidistant from one another at the DNA level with about 1.7% sequence difference seen in each of the three comparisons. Yet, morphologically chimps and gorillas are far more similar to one another than either is to us. This must mean that there has been much more morphological change along our lineage than along those leading to the African apes since the three genera last shared a common ancestor some 4.5 million years ago (the amounts of sequence change at the DNA level are the same). The current racial situation in our species is then entirely consistent with the history of our lineage: much morphological variation and change, little genetic variation and change.

Racial Differences in Athletic Ability

Another tack has been to acknowledge racial differences, but then argue that they are generally small with respect to differences among individuals within races, and, in any case, likely to be functionally irrelevant for any features of particular importance for the species. Consider the following example from sports. Every year perhaps 75 young men newly make NBA (National Basketball Association) teams. Of these, about 60 will be Black, and 15 White. (I am here using four years as the average length of a NBA career, and the current racial composition of the league as a source for these figures. “Black” means, in this country, that the individual has a substantial amount of obvious recent sub-Saharan African ancestry. “White” means no obvious ancestry other than European.) These numbers mean that the chance for a Black to play in the NBA is about one in 4,500; the corresponding figure for a White is about one in 90,000. We can then ask from how far out on their respective bell curves these 75 are drawn. Recourse to a z-score table tells us that 1 in 4,500 takes us about 3.4 SD (standard deviations) from the mean; 1 in 90,000 is about 4.3 SD from the mean. I submit that this almost one SD difference between populations in this suite of abilities based on a fundamental human trait is pretty substantial. In other words, it is simply not true that “bipedalism is such a critical aspect of the human adaptation that one would not expect to see great differences from either the individual to individual level, or between populations.” Bipedalism is certainly a “critical aspect of the human adaptation,” but it does not follow that therefore individual and group variation in what might be termed the quality of the bipedal adaptation would have been reduced. Indeed, it seems to me that, if anything, we might expect quite the opposite result. It took me a long time to figure this out, and thus it might prove useful to others to recount some of that process. The context is the relationship, if any, between brain size and cognitive performance.

Racial Differences in Brain Size

Discussing racial differences in athletic ability can get you into trouble, as some sportscasters have discovered. Discussing racial differences in brain size can be literally life threatening, as some psychologists have discovered. This issue ultimately divided Charles Darwin from Alfred Russel Wallace. Darwin was entirely comfortable with the notion that the human mind had evolved through natural selection, just as did the human body. Wallace, on the other hand, to the end of his much longer life, insisted that while our body had evolved, our mind must have been created. (See Michael Shermer’s essay on this subject in Vol. 3, #1 of Skeptic.) A century later the very influential book The Mismeasure of Man by Stephen Jay Gould also, in effect, denied that our brains had evolved. Gould spends the first two chapters telling us that brain size and intellectual performance have nothing to do with one another, without once bothering to remind us that our brains have not always been the size they are today. Nor is that awkward fact mentioned anywhere else in the book. You could never learn from it that in our evolutionary lineage brain size had increased from around 400cc to 1300–1400cc over the last four million years. Why this omission?

I think the answer is quite straightforward. That part of Gould’s psyche concerned with basic evolutionary biology knew that those large brains of ours could not have evolved unless having large brains increased fitness through minds that could do more. In other words, individuals with larger brains must have been, on the average and in the long run, slightly better off than those with smaller brains. How advantaged? Dare one say it? By being smarter. What else? If variation in brain size mattered in the past, as it must have, then it almost certainly still matters. And if you are going to argue that it does not, then you are going to have to explain why it does not. I do not think you can do this while maintaining your intellectual integrity. Thus Gould just ignored the demands of the evolutionary perspective by denying, implicitly, that our brains had evolved. I find it of some interest that no one has really challenged him on this point.

The evolutionary perspective demands that there be a relationship— in the form of a positive correlation—between brain size and intelligence. That proposition, I would argue, is not something that need derive from contemporary data (although, as we will see, those data do give it strong support). It is what we would expect given our particular evolutionary history; that is, it is the evolutionary null hypothesis, and, thus, something to be disproven. It seems to me that a demonstration of no correlation between brain size and cognitive performance would be about the best possible refutation of the fact of human evolution. It took me a long time to figure out what really ought to have been obvious: descent with modification by means of natural selection has been, and continues to be, the reality. It should be incumbent on those who would deny our evolutionary history to show that our biology is not involved. Otherwise there is an implicit creationism present in those who persist in ignoring the evolutionary perspective when they try to explain some aspect of our behavior (all too common in the social sciences). Brain size is an effective proxy for behavior, and it reminds us that evolutionary processes and evolutionary lineages are rather good data.

In other words, natural selection requires genetically-based phenotypic variation to work on; thus throughout the period of change in brain size, there must have been present a substantial amount of genetic variation for brain size, and, likely, the greater the advantage of larger brains, the greater the underlying genetic variation for brain size. I had long been frustrated by the canalization argument (the more important the characteristic, the less variation) with respect to human intelligence, my teaching experiences telling me that cognitive performance was one of our most variable features. Yet at the same time I was unable to refute the logic of the argument. This lasted until 1983 when I remembered Fisher’s Fundamental Theorem of Natural Selection: “The rate of increase in the fitness of any organism at any time is equal to its genetic variance in fitness at that time.”

This says it all. An earlier statement of the general argument was made by the late Bernard Davis in 1976:

Let me further emphasize that, even if no one had ever devised a test for measuring IQ, we could still be confident, on grounds of evolutionary theory, that our species contains wide genetic variance in intelligence. The reason is that natural selection cannot proceed unless it has genetic diversity, within a species, to act on; and when our species is compared with its nearest primate relatives, it is obvious that our main selection pressure has been for an increase in intelligence. Indeed, this change proceeded at an unprecedented rate (on an evolutionary time scale): in the past three million years the brain size of the hominid line increased threefold. Such rapid selection for increased intelligence could not have occurred unless the selection pressure had a large substrate of genetic variation to act on.

Brain Size and Cognitive Performance: Data Validate Theory

Any suggestion on one’s part that people with bigger brains are, on the average, smarter by virtue of those bigger brains leads the listener to doubt one’s intelligence, if not one’s sanity. The general belief is that this inherently sexist and racist notion died an ignoble death sometime in the last century. Its recent resurrection began with a 1974 article by Leigh Van Valen. In it he reviewed the literature and concluded that the published correlations between brain size and intelligence (as measured by standardized tests) were unrealistically low because they did not allow for the fact that external measurements of head size were an imperfect indicator of brain size. Correcting for this attenuation indicated that the actual value was probably about 0.3. (The Mismeasure of Man does not even mention Van Valen’s work.) A subsequent large scalestudy of Belgian army recruits, which also used a much wider variety of tests of cognitive function, gave figures consistent with Van Valen’s analyses (Susanne, 1979). Since 1987, there have been several studies on this subject in which the brain size of living individuals was measured directly and accurately using magnetic resonance imaging (e.g., Willerman et al., 1991; Andreasen, et al., 1993; Wickett, et al., 1994). These suggest that Van Valen’s estimate was, if anything, conservative— the consensus being in the area of 0.4 or a bit more. Although, as argued above, a positive relationship was to be expected on the basis of simple evolutionary considerations, the actual correlations found are higher than just about anyone would have predicted prior to Van Valen’s pioneering effort.

A correlation of 0.4 means that of the average of 17 IQ points separating two randomly chosen individuals (within sex and population), about 7 IQ points would derive from the differences in the sizes of their brains. The same would hold for populations, and existing human populations can differ in their means by as much as 2 SD in brain size. Thus, this variable alone could lead to close to a 1 SD difference in mean intellectual performance among them. With respect to the difference between American Whites and Blacks, the one good brain size study we have (Ho, et al., 1980) indicates a difference between them of about 0.8 SD; this could correspond to a difference of about 5 IQ points; that is, about one-third of the observed differential.

It should also be noted that these data strongly suggest that IQ tests are, in fact, measuring something that has been significant in human evolution, given that performance on them correlates so nicely with brain size. And what of the common accusation of circularity that intelligence is what the tests test? As Daniel Seligman notes, in A Question of Intelligence (pg 15):

[Herrnstein] said it was not at all intended as a put-down of IQ tests, certainly not as a complaint about circularity. It represented, rather, the perspective of a psychologist who believed (a) that “intelligence” needed to be anchored to some unambiguous operational definition and (b) that the cluster of abilities measured by IQ tests constituted a reasonable anchor. Fast analogy: You could define length…as “a distance or dimension expressed in units of linear measure.” You could also define it as the thing that tape measures measure.

Individuals and Groups

So far I have tended to go from group to individual and back again without addressing the fact that any number of commentators on The Bell Curve have argued that: (1) individual variation within groups is generally greater than variation between groups, and (2) the existence of functionally significant genetic differences among individuals (with which most of them apparently feel comfortable) does not necessarily imply such among populations (with which they, along with most people, definitely do not). But the obvious truth of these two assertions in no sense justifies the object lesson we are supposed to draw from them—that therefore group variation is not something that need particularly concern us. First, the fact is group differences can be much greater than individual differences within them; for example, hair form in Kenya and Japan, or body shape for the Nuer and Inuit. And even when the first assertion is correct, as it is for most human characteristics, the differences between groups can, as already noted, be quite consequential. There is a much weaker case to be made for the relevance of the second assertion. While a qualification such as “does not necessarily” makes it technically correct, the statement as a whole implies that we should expect a connection between individual and group variation to be the exception, rather than the rule.

The evolutionary perspective begs to disagree. Consider again the example of brain size. Wi thin sex and population, the coefficient of variation (standard deviation / mean x 100) is about 10%, a value typical for mass or volume characters. Two randomly chosen same-sex individuals within a population would then differ by about 12%, or about 150cc. But so can two populations. And this should not surprise us. Remember that our brain has increased in size some 1000cc in the last 3 million years. This is often termed “an explosive rate of growth,” yet it works out to only ¼ drop per generation. It could have gone faster, given what we know of individual variation and heritability for the character. That it did not implies that the huge advantages conferred by having more brain to work with must have been offset by (almost) equally large disadvantages. In other words, the adaptation here is best seen as a very slow moving compromise involving small relative differences between large forces. We should then have no expectation that those advantages and disadvantages would have balanced out in the same way in different populations at differing times and in differing ecological and cultural circumstances. But this same argument will apply to most aspects of individual variation. Given the number of characteristics in which functional variation is present, the ways in which they will balance out in two populations evolving more or less independently of one another are almost guaranteed to be different in the two. The balancing will take place at the level of individual phenotypes, and thus there is, in general, going to be a direct, inescapable connection between individual and group variation whenever evolutionary change is taking place.

Harmful Truths or Useful Lies?

Of all the thousands of words in print about The Bell Curve, about its data and arguments, perhaps none cut so close to the bone as those of Nathan Glazer in the October 31, 1994, issue of the New Republic (pp 15–16):

The authors project a possible utopia in which individuals accept their places in an intellectual pecking order that affects their income, their quality of life, their happiness. It may be true that we do not commonly envy the intellectual capacities of others—we allow Albert Einstein and Bobby Fischer their eminence—though I think even at this level the authors underplay the role of envy and rancor in human affairs. But how can a group accept an inferior place in society, even if good reasons for it are put forth? It cannot.

Richard Wollheim and Isaiah Berlin have written: “If I have a cake, and there are ten persons among whom I wish to divide it, then if I give exactly one-tenth to each, this will not…call for justification; whereas if I depart from this principle of equal division I am expected to produce a special reason.” Herrnstein and Murray have a very good special reason: smarter people get more and properly deserve more, and if there are more of them in one group than another, so be it. Our society, our polity, our elites, according to Herrnstein and Murray, live with an untruth: that there is no good reason for this inequality, and therefore our society is at fault and we must try harder. I ask myself whether the untruth is not better for American society than the truth.

And Bill Clinton, in a press conference of similar vintage, said:

I haven’t read it. But as I understand the argument of it, I have to say I disagree with the proposition that there are inherent, racially-based differences in the capacity of the American people to reach their full potential. I just don’t agree with that. It goes against our entire history and our whole tradition.

Are All Men Created Equal?

The issue here is not so much about “inherent, racially-based differences in the capacity of the American people to reach their full potential.” It is about inherent, racially-based differences in the potentials themselves. The “entire history and our whole tradition” is, of course, encapsulated in our Declaration of Independence, where Thomas Jefferson wrote:

We hold these truths to be self-evident, that all men are created equal, that they are endowed by their creator with certain unalienable rights, that among these are life, liberty, and the pursuit of happiness. That to secure these rights, governments are instituted among men, deriving their just powers from the consent of the governed….

Which takes us back to Glazer, and the real need to ask “whether the untruth is not better for American society than the truth.” The untruth in Jefferson is his first truth: “that all men are created equal.” We know Jefferson did not believe that to be literally true, or perhaps more fairly, that he could not have believed it true unless one word were added to his sacred text: “ ….all men are created equal, in that they are endowed by their creator with certain unalienable rights…” This addition in no way detracts from the power of the text (and only slightly from its rhythm), but does provide the advantage of literal truth—understanding, of course, that, ever since 1859, “creator” has had to be read as “the evolutionary process.” Reading it that way also has the virtue of ultimately leading us to an understanding of why “the evolutionary process has made all men equal” is no better than the original text. Also note that last right—it is not “happiness” but “the pursuit of happiness”—an opportunity, not a result.

Like most white Americans of my sex, class, and generation, I came into a world that soon made me a racist and a sexist. And then, like most well-educated people of that generation, as I grew up I repudiated both race and sex as explanations for differences in the behavior of human beings.
—Carl Degler (1991)

There have, in fact, been attempts to provide a justification based in evolutionary biology for a literal reading of “all men are created equal.” Gould, for example, entitled one of his essays “Human Equality is a Contingent Fact of History,” and summarized his argument as such (1985, p. 198):

Homo sapiens is a young species, its division into races even more recent. This historical context has not provided enough time for the evolution of substantial differences. But many species are millions of years old, and their geographic divisions are often marked and deep. H. sapiens might have evolved along such a scale of time and produced races of great age and large accumulated differences— but we didn’t. Human equality is a contingent fact of history.

The problems with this line of argument are many. First, it is strange to have one of the inventors of the theory of “punctuated equilibrium” argue that human races cannot be very different from one another because they are too young. Second, nowhere in the article does Gould give us an example of a species in which races are as strongly marked as ours. The reason very likely is, as I have already noted, that there isn’t any such species. Third, there are substantial racial differences present today—however they may have come about. I have already discussed two of these: athletic performance and brain size. Thus, Gould has it backwards. It is from the present that we obtain most of our knowledge of the past, and not, as most paleontologists would have it, the other way around. Finally, at least for our purposes, there is a strong tendency just about everywhere to extend the “there are no significant racial differences” argument to one which says “there are no significant gene-based differences between individuals.” And as more and more groups are seen as needing some sort of official recognition, this extension becomes more and more inevitable.

Nature, Nurture, and the Individual

The above extension is, and for a long time has been, the prevailing point of view in the social sciences and humanities. If one takes a course at U.C. Berkeley in, say, Anthropology 3 (Introduction to Social and Cultural Anthropology), or Sociology 1 (Introduction to Sociology), one will hear an enormous amount about individuals as constituents of groups, and precious little about individuals as individuals. There will be little discussion of genes, evolution, or biology. It then goes almost without saying that you are not going to hear anything about free will or personal responsibility. The willful development of this situation in this country is very nicely documented in Carl Degler (1991). His Preface begins (and I quote directly and at length because the statement is so representative):

Like most white Americans of my sex and class (the son of a fireman) and my generation (born in 1921) I came into a world that soon made me a racist and a sexist. And then, like most well-educated people of that generation, as I grew up I repudiated both race and sex as explanations for differences in the behavior of human beings. Indeed, I spent a good deal of my youth and adulthood arguing by voice and in print against biology as a source of human behavior, not only in regard to race and sex, but in other respects as well. How and why that sea change occurred in my thinking concerned me only peripherally. I knew there had been a time when biology was thought to be an important way of explaining why social groups differed, why some people were considered better than others. But that was another time. In my new outlook it was a given that the repudiation of biology had resulted from a penetrating, perhaps even lengthy scientific investigation of biology’s inadequacy in accounting for the ways in which human groups differed. In ruling out biology as a cause for human differences, I thought of myself as defending a truth as solidly established as the heliocentric universe. Human nature, I believed, was constructed over time, not inherited from time. I had no trouble accepting Karl Marx’s famous remark that man made his own history, not entirely as he pleased, meaning that history may limit us at times, but biology has little to say about our social behavior.

Today, in the thinking of citizens and social scientists alike the deeply held assumption is that culture has severed for good the link between human behavior and biology. The conviction is that human beings in their social behavior, alone among animals, have succeeded in escaping biology. The irony is heavy here. For that belief is accompanied by another deeply held conviction: that human beings, like all other living things, are the products of the evolution that Charles Darwin explained with his theory of natural selection. The irony is almost palpable as Darwin entertained no doubt that behavior was as integral a part of human evolution as bodily shape. And that is where Book III enters. It seeks to tell the story of how biological explanations have begun to return to social science… It is important to recognize that this “return of biology” is not simply a revival of repudiated ideas, like racism, sexism, or eugenics.

The problem here is that a “return of biology” means a return to the idea that sex and race will have consequences, and if you recognize this publicly, then you become, for many, a racist or a sexist. But the fact is, the evolutionary process cannot and does not produce equality either among individuals or groups.

Much of the furor surrounding The Bell Curve thus derives from a very real problem. Herrnstein and Murray are officially agnostic on the degree of genetic involvement in racial differences in intellectual performance, give gender differences one small paragraph on p. 275, and mention the implications of our evolutionary history not at all—but all that does not really help. The fact is that deep down all too many of us are aware of the reality of group differences, and of the virtual certainty that genes are somehow involved in producing some of those differences. But, as Ernst Mayr pointed out in 1963: “Equality in the face of evident nonidentity is a somewhat sophisticated concept and requires a moral stature of which many individuals seem incapable.”

Consider the treatment of E. O. Wilson after the publication of his masterful Sociobiology in 1975. Or, a more recent example, the June, 1993 issue of Scientific American. features a lengthy essay by John Horgan, one of their staff writers, entitled “Eugenics Revisited,” and teased on the cover as “The dubious link between genes and behavior.” This one was so egregious—especially given the venue—that I was moved to send a long letter to the editor, publisher, and other officials of the magazine. I had no illusions that it would be publicly acknowledged by them (and it wasn’t, though I did get a letter from the editor Jonathan Piel). Scientific American continued in this vein in its January and February, 1995 issues. In the first, Tim Beardsley, one of their staff writers, authored a piece entitled For Whom The Bell Curve Really Tolls, and subtitled: “A tendentious tome abuses science to promote far-right policies.” My thought is that you have to be pretty far left to see any of Herrnstein and Murray’s “messages” as “far right.” And Beardsley apparently has no compunction about penning flat-out lies, such as: “…numerous studies have demonstrated that early childhood surroundings have a large role in molding IQ scores—while very few studies have indicated a significant role for heredity.” Anyone who could write that those last 10 words presumably would also describe our national debt as composed of very few dollars. The February issue then contains a review of The Bell Curve by Leon Kamin, one of the authors of the 1984 book Not in Our Genes. His position can be inferred from the title, and from the fact that he and his coauthors were willing to state: “For all we know, the heritability (of IQ) may be zero….” And, in its final paragraph:

We should recall that the title of the article by A. R. Jensen…was “How Much Can We Boost IQ and Scholastic Achievement?” The answer, from cross-racial and crossclass adoption studies, seems unambiguous. As much as social organization will allow. It is not biology that stands in our way.

I submit that someone who could seriously entertain the notions that the heritability for any human performance measure could really be zero, and that our biology places no limits on a human performance, has thereby removed himself from serious consideration as a scholar of anything.

The Decline of Racism in Society

From an evolutionary perspective freedom can only mean freedom of opportunity, which, in the context of this article, necessarily leads to the question of how we are to recognize it among races and groups when we are living in a world where functionally significant, gene-based, racial and other group differences may well be the rule rather than the exception. It is here I think The Bell Curve makes its most meaningful single contribution (pp. 323–4). There we find the income data for young (average age = 29) year-round workers of three racial/ethnic groups: White, Black, Latino—with Latinos earning 86% and Blacks 77% as much as Whites. But when IQ is held constant (average = 100 for all three groups), both the Latino and Black figures climb to 98% of that for Whites. This result (which could be seen as remarkable only if one accepts the “this is a racist society” mantra) tells us about the degree of equality of opportunity in recent American society, and yet only one commentator of the more than 100 I have read or heard (including Murray and Herrnstein themselves) seems to have found it worthy of comment. This was Daniel Seligman, himself the author of the highly readable and most informative 1992 volume, A Question of Intelligence, who titled his brief column in the December 12, 1994, issue of Fortune, “News Nobody Noticed”:

Your servant has now read scores of reviews of The Bell Curve. Most have fiercely criticized the book’s thesis, which emphasizes the centrality of IQ in lives and careers, and most have dwelt insistently on race and the 15-point blackwhite IQ gap. But, oddly, we have yet to read a review noticing the racial news built into a table on page 324. In a rational world, the news would be on the front pages…. The news is about racial discrimination in America. As we all keep reading, blacks earn a lot less than whites, even when you compare workers of similar ages and educational backgrounds. This table confirms this finding. But it points to something else one has never before read: that when you control for age and IQ, the blackwhite earnings gap just about disappears….

Obvious implication: At least so far as younger workers are concerned, employers no longer engage in irrational discrimination based on race. They discriminate based on IQ—which is rational, given the avalanche of data linking IQ to performance in many different job markets. Fascinating question: How can it be, in a world where racial discrimination is (properly) an object of enormous concern, that we are ignoring powerful evidence of its decline?

I would add that Seligman’s comment that employers “discriminate based on IQ” has to be taken metaphorically. What they are doing is rewarding performance (as any rational employer would). The connection with IQ, as Herrnstein and Murray point out (pp 80–81), is that it is the best single predictor of performance—better than biographical data, reference checks, education, interview, or college grades. And as to his final question? The cynic in me cannot help commenting “So what else is new?” One does not really expect our media to report anything positive about this society, does one?

The Rise of Racism on College Campuses

No society has an unstained history. The treatment of individuals of sub-Saharan African ancestry is without doubt our largest and deepest stain, and that history, as are all histories, is beyond change. Given those truths, the worst thing we could do is to repeat that past in the name of producing an equality of results, by again allowing the treatment of an individual to be influenced by that individual’s race (or sex, or ethnicity, or any other grouping). Yet, increasingly over the past 30 years we have been doing just that.

What we are getting at Berkeley is two communities, separable on racial/ethnic grounds, and increasingly divergent from one another academically, socially, and in ethos—a result desired, presumably, by no rational soul.

My own direct experiences with such race-norming, quota-driven treatment of individuals has been at U.C. Berkeley, where, for the last 10 years or so, a substantial percentage of freshman admissions (up to about 40%) has been reserved for “underrepresented minorities,” and where race, ethnicity, and sex have become major factors in the hiring of new faculty. For students, what this has done is to produce two populations separated by race/ethnicity and performance who wind up, in the main, in different courses and pursue different majors. That is only to be expected when the SAT difference between the White and Asian students on the one hand, and Black and Latino students on the other, is about 270 points (1270 v. 1000; about 1SD difference). This is equivalent to about three to four years of academic achievement, and U.C. Berkeley is no place to play catch-up. And, as far as anyone knows (there are no published studies on the matter), no catching-up in fact takes place. I wrote of this situation in 1990:

The Berkeley administration has, in its admissions policies, especially over the past five years or so, ignored certain unpalatable realities, and given us an even more unpalatable set of results. They have given us a situation where the association between race/ethnicity and performance is real, obvious, and of ever-increasing strength. What we are getting at Berkeley is two communities, separable on racial/ethnic grounds, and increasingly divergent from one another academically, socially, and in ethos—a result desired, presumably, by no rational soul. It is, frankly, difficult to imagine policies more deliberately crafted or better calculated to exacerbate racial and ethnic tensions, discourage individual performance among all groups, and contribute to the decay of a magnificent educational institution.

The fact is that any group-based policies are bound to have effects of this sort. As I have already noted, the evolutionary necessity of individual variation is almost always going to lead to group variation, and statistical realities require that group differences get exaggerated as one goes toward the ends of the bell curves involved. Thus, when you look at group representations with respect to the high-visibility pluses (e.g., high-paying jobs) and minuses (e.g., criminality) in any society, one can virtually guarantee that they are not going to be equal— and that the differences will not be trivial. The problem is in recognizing and adapting to those realities, and not, as has so often been the case with responses to The Bell Curve, denying them. I noted this in a letter to a Berkeley Faculty Senate committee on “diversity”:

This current focus on “diversity,” if continued and “successful,” can only have the effect of rewarding individuals for making their primary allegiances to certain defined groups, and, thus, of tribalizing our society. It would require a mind completely closed to current realities, never mind historical ones, to remain ignorant of the disastrous effects of tribalization. One therefore has to suspect that anyone supporting policies that tribalize is either ignorant, or simply playing the very effective political game of “divide and conquer.” The number of different roles to be played in a society increases as the complexity of a society increases. Ours is a very complex society that will only become more complex in the future. The number of different roles to be played will thus increase, requiring a larger number of allegiances for individuals within the society, and selecting against those whose primary allegiance is to a particular group—be it one based on biology (race, sex, age) or culture (ethnicity, religion). If one of your roles is chemist, then one set of your allegiances is to the community of chemists and chemistry. You are then a chemist, period— and not a female, or White, or Catholic, or old. To the extent that you do not look at it that way; that is, to the extent that you see yourself as some sort of hyphenated chemist, you will necessarily reduce the effectiveness of your chemistry. And this is going to be true for each of the other roles you will come to play. To the extent that you see yourself as a hyphenated anything, your achievement in that “anything” will tend to be reduced. And to the extent that a society encourages and rewards individuals for looking at themselves in such a fashion, it necessarily reduces its total level of accomplishment.

Skeptic magazine 3.3

This article appeared in Skeptic magazine 3.3 (1995)

Buy the print edition

There are certain harsh realities in life. One of these is that groups, whether age, sex, race, ethnic, or whatever, are groups, and groups of anything are very likely going to differ from one another. If they didn’t, then they wouldn’t be groups, would they? I can then confidently guarantee that when we measure performance by groups, we are going to find group differences in performance. Some part of those differences will be nature-based, some part will be nurture-based, some will be will-based. No society has, or can have, the power to even things up. Societies are not omnipotent. They can provide opportunity; they cannot mandate individuals or groups making equal use of those opportunities, and, therefore, they cannot make either individuals or groups come out even. Individual and group variation are realities that they cannot will out of existence. They can try, and what happens then, is, unfortunately, no secret: a temporary leveling-down bought at enormous cost. They can in no sense make groups equal. They cannot level up—only down—and thus any such leveling is necessarily at the expense of individual freedom and, ultimately, that society’s total level of accomplishment.

Ending Racism Without Ending Race

There would appear to be a substantial consensus among some of the more “conservative” commentators on The Bell Curve as to its policy implications, and, for better or worse, I find myself in total agreement with them. Seligman, for example, closes his A Question of Intelligence with: “One major message of the IQ data is that groups are different. A major policy implication of the data, I would argue, is that people should not be treated as members of groups but as individuals.” Herrnstein and Murray give us the same message, but at much greater length, in their Conclusion (pp 549–552).

I opened with a quote from Charles Krauthammer. His conclusion says it better than I can:

I distrust all multiculturalism, liberal or conservative. The Balkans amply demonstrate the perils of balkanization. My answer is simpler: Stop counting by race. Stop allocating by race. Stop measuring by race. Let’s return to measuring individuals.

It seems hopelessly naive to propose this today. But it was not naive when first proposed by Martin Luther King and accepted by a white society that was finally converted to his vision of color blindness. Instead, through guilt and intimidation, a liberal establishment has since mandated that every study of achievement be broken down by race. ‘The Bell Curve’ takes that mandate to its logical conclusion.

Enough. As both Murray and Thomas Sowell explicitly state, knowing the group score tells you nothing about the individual. Well, we have seen the group score. Let’s go back to counting individuals. How many of Murray’s critics will agree to that?

Amen. Let’s go back to counting individuals. And how do we encourage such behavior? Simple. Just remove all reference to group identity from both statutory and administrative law. Period.

About the author

Dr. Vincent M. Sarich (December 13, 1934–October 27, 2012) was an American Professor Emeritus in anthropology at UC Berkeley. He taught courses in Physical Anthropology, Human Biology, Human Variation in an Evolutionary Perspective, and Evolution of Human Behavior, along with graduate seminars in comparative biochemistry. He was best known for his work in molecular dating in which he found that the accumulation of immunological differences among albumins occurred as a regular function of time, which lead him and Allan Wilson to propose a molecular clock of primate albumin evolution. They concluded that our closest living relatives were the African apes and that we had shared a common ancestor with them no more than about 5 million years ago, a revolutionary idea at the time. His later work centered on racial variation within the human species in which he suggested that while the species may be relatively old, races are young with most of the interpopulational variation having developed within the last 15,000 to 20,000 years. He was best known to skeptics for his many debates on evolution and creationism with Duane Gish and Phil Johnson.

  • Andreasen, N. C., M. Flaum, V. Swayze, D. S. O’Leary, R. Alliger, G. Cohen, J. Erhardt, and W. T. C. Yuh. 1993. “Intelligence and Brain Structure in Normal Individuals.” American Journal of Psychiatry 150:130–134.
  • Beardsley, T. 1995. “For Whom the Bell Curve Really Tolls.” Scientific American 272:1:14–17 (January).
  • Berlin, B., and P. Kay. 1969. Basic Color Terms: Their Universality and Evolution. Berkeley, University of California Press.
  • Cavalli-Sforza, L. L., P. Menozzi, and A. Piazza. 1994. The History and Geography of Human Genes. Princeton University Press. Princeton, N.J.
  • Davis, B. 1976. “Evolution, Human Diversity, and Society.” In Zygon 11:2:80–95.
  • Degler, C. N. 1991. In Search of Human Nature. Oxford University Press.
  • Gould, S. J. 1981. The Mismeasure of Man. New York: Norton.
  • ___. 1985. “Human Equality is a Contingent Fact of Evolution.” In The Flamingo’s Smile. New York: Norton.
  • Hardin, G. 1959. Nature and Man’s Fate. New York. Holt, Rinehart, and Winston.
  • Ho, K.-C., U. Roessmann, J. V. Straumfjord, and G. Monroe. 1980. “Analysis of Brain Weight.” Archives of Pathology and Laboratory Medicine 104:635–645.
  • Horgan, J. 1993. “Eugenics Revisited.” Scientific American 270:6:122–131 (June).
  • Kamin, L. 1995. “Behind the Curve.” Scientific American 272:2:99–103 (February).
  • Krauthammer, C. 1994. “Liberals, Obsessed by Race, Can Hardly Complain.” The News & Observer (Raleigh, N.C.) 23 October 1994.
  • Mayr, E. 1963. Animal Species and Evolution. Cambridge: Harvard University Press.
  • Sarich, V. M. 1995. Race and Language in Prehistory. In press. For copies write: 555 Pierce, Unit 730, Albany, CA. 94706.
  • Seligman, D. 1992. A Question of Intelligence: The IQ Debate in America. New York: Birch Lane Press.
  • ___. 1994. “News Nobody Noticed.” Fortune 12 December:255.
  • Smith, C. L., and K. L. Beals. 1990. “Cultural Correlates with Cranial Capacity.” American Anthropologist 92:193–200.
  • Subramanian, S. 1995. “The Story in Our Genes.” Time 145:2:54–55 (16 Jan 1995).
  • Susanne, C. 1979. “On the Relationship Between Psychometric and Anthropometric Traits.” American Journal of Physical Anthropology 51:421–424.
  • Van Valen, L. 1974. “Brain size and Intelligence in Man.” In American Journal of Physical Anthropology 40:417–424.
  • Wickett, J. C., P. A. Vernon, and D. H. Lee. 1994. “In Vivo Brain size, Head Perimeter, and Intelligence in a Sample of Healthy Adult Females.” Personality and Individual Differences 16:831–838.
  • Willerman, L., R. Schultz, J. N. Rutledge, and E. D. Bigler. 1991. “In Vivo Brain Size and Intelligence.” Intelligence 15:223–228.

Get eSkeptic

Be in the know!

Subscribe to eSkeptic: our free email newsletter and get great podcasts, videos, reviews and articles from Skeptic magazine, announcements, and more in your inbox twice a week. It’s free. We never share your address. Unsubscribe any time.

Sign me up!

Detecting Baloney

Baloney Detection Kit Sandwich (Infographic) by Deanna and Skylar (High Tech High Media Arts, San Diego, CA)

The Baloney Detection Kit Sandwich (Infographic)

For a class project, a pair of 11th grade physics students created the infographic shown below, inspired by Michael Shermer’s Baloney Detection Kit: a 16-page booklet designed to hone your critical thinking skills.

FREE PDF Download

Wisdom of Harriet Hall

Top 10 Things to Know About Alternative Medicine

Harriet Hall M.D. discusses: alternative versus conventional medicine, flu fear mongering, chiropractic, vaccines and autism, placebo effect, diet, homeopathy, acupuncture, “natural remedies,” and detoxification.

FREE Video Series

Science Based Medicine vs. Alternative Medicine

Science Based Medicine vs. Alternative Medicine

Understanding the difference could save your life! In this superb 10-part video lecture series, Harriet Hall M.D., contrasts science-based medicine with so-called “complementary and alternative” methods.

FREE PDF Download

The Top 10 Weirdest Things

The Top Ten Strangest Beliefs

Michael Shermer has compiled a list of the top 10 strangest beliefs that he has encountered in his quarter century as a professional skeptic.

FREE PDF Download

Reality Check: How Science Deniers Threaten Our Future (paperback cover)

Who believes them? Why? How can you tell if they’re true?

What is a conspiracy theory, why do people believe in them, and can you tell the difference between a true conspiracy and a false one?

FREE PDF Download

The Science Behind Why People See Ghosts

The Science Behind Why People See Ghosts

Mind altering experiences are one of the foundations of widespread belief in the paranormal. But as skeptics are well aware, accepting them as reality can be dangerous…

FREE PDF Download

Top 10 Myths About Evolution

Top 10 Myths About Evolution (and how we know it really happened)

If humans came from apes, why aren’t apes evolving into humans? Find out in this pamphlet!

FREE PDF Download

Learn to be a Psychic in 10 Easy Lessons

Learn to do Psychic “Cold Reading” in 10
Easy Lessons

Psychic readings and fortunetelling are an ancient art — a combination of acting and psychological manipulation.

Copyright © 1992–2022. All rights reserved. | P.O. Box 338 | Altadena, CA, 91001 | 1-626-794-3119. The Skeptics Society is a non-profit, member-supported 501(c)(3) organization (ID # 95-4550781) whose mission is to promote science & reason. As an Amazon Associate, we earn from qualifying purchases. Privacy Policy.