Apocalypse Now?

1.

The Bell Curve is the product of an obsession, or, more exactly, of two different obsessions. Richard Herrnstein—who died on September 24 of this year—was obsessed with the heritability of intelligence, the view that much the largest factor in our intellectual abilities comes in our genes. He was also convinced that there had been a liberal conspiracy to obscure the significance of genetically based differences in the intelligence of different races, social classes, and ethnic groups, and that all manner of educational and economic follies were being perpetrated in consequence. Charles Murray—who is energetically and noisily with us still—is obsessed with what he believes to be the destructive effects of the American welfare state.

The result of their cooperation is a decidedly mixed affair. The politics of The Bell Curve are at best slightly mad, and at worst plain ugly. Its literary tone wobbles uneasily between truculence and paranoia. Its intellectual pretensions are often ill founded. For all that, anyone who has an interest in the philosophy of science and a taste for public policy will enjoy much of The Bell Curve; it is full of interesting, if dubiously reliable, information, and it offers the always engaging spectacle of two practical-minded men firmly in the grip of irrational passion.

Richard Herrnstein’s passion was the conviction that each person has a fixed or nearly fixed quantum of “cognitive ability,” the intelligence whose quotient constitutes your IQ. Herrnstein began his career as a disciple of the behaviorist psychologist B. F. Skinner, and therefore as a devout environmentalist. Then he fell in love with “Spearman’s g.” Charles Spearman was a turn-of-the-century British Army officer and statistician who thought that people possess varying amounts of general intelligence—or “g”—and invented statistical techniques to discover which intelligence tests most directly tap into this basic ability.

Skeptics have always said that g explains nothing: the fact that the performance of individuals on different tests is closely correlated, and predicts their success in school work and some occupational settings, is important and interesting. Talk of g adds nothing to the fact of the correlation. Herrnstein, however, was no skeptic in this matter. At the first mention of g he confesses that “its reality … was and remains arguable.” But eleven pages later, he claims that g sits at the center of the mind’s capacities “as an expression of a core mental ability much like the ability Spearman identified at the turn of the century,” while eight pages on, after a further bout with the skeptics, he announces that it is universally accepted that “there is such a thing as a general factor of cognitive ability on which human beings differ.”

Does it matter? Only to the extent that it reinforced Herrnstein’s fascination with ethnicity. The more you think that talk of IQ is talk of a mysterious something that possesses the same reality as visible qualities like skin color or the curliness of the hair, the more obvious it will seem that ethnic groups that differ in such visible qualities must differ in intelligence too. By the same token, it bolsters the extraordinary fatalism that infuses The Bell Curve: once you discover that the average IQ of people in jail is 93, it’s easy to believe that people with too little g are more or less doomed to social dysfunction. How other countries of the same ethnic composition as white America manage to commit fewer murders and yet jail far fewer of their citizenry remains for ever inexplicable. Conversely, a certain skepticism about what there is to IQ besides being good at certain sorts of tests may make us less superstitious about its importance.

Charles Murray is intoxicated by an apocalyptic vision of the American future, nicely summarized as “The Coming of the Custodial State.” The anxieties about the widening inequality produced by the American economy are ones that Mickey Kaus and Robert Reich long ago familiarized us with, but they are here run through Murray’s wilder and darker imaginings to yield a vision of an incipient semifascist future that neither of them would recognize.¹ American society is increasingly partitioned into a high-IQ, ever more affluent, upper caste, a hard-pressed middle class, and a cognitively underprivileged underclass, whose criminality threatens the rest of us and whose unchecked breeding threatens to dilute the pool of talent, and so alarmingly on. The well-off migrate to enclaves of comfortable housing, which are walled-off, well-policed, and equipped with decent schools; the underclass are shut away in urban slums. The struggling middle class feels trapped.

The elite may hold liberal views and they may be willing to pay for help to the poor, but they will not live among them. The middle class have neither money to spend on the underclass nor tolerance of its ways. They will insist on coercive policing and a more punitive welfare system, and will want the underclass kept in whatever “hightech and more lavish version of the Indian reservation” it takes to keep them from preying on the respectable. The end result, Murray argues, is catastrophe: a version of the welfare state in which the incompetent have their lives managed without their consent.

It is difficult to imagine the United States preserving its heritage of individualism, equal rights before the law, free people running their own lives, once it is accepted that a significant part of the population must be made permanent wards of the state.

This is a eugenicist, and not (in the usual sense) a racist, nightmare, for Murray believes that the bottom 10 percent of the white population is headed for the degradation that already afflicts the black urban underclass. The people he affectionately describes as “white trash” will need as much looking after as their black counterparts.

So far as I can see, none of Murray’s anxieties about the direction of American domestic policy depends on the truth of Richard Herrnstein’s ideas about the ontological status of g, and none of Herrnstein’s claims about intelligence support Charles Murray’s ideas about social policy. Murray himself seems to recognize this: “Like other apocalyptic visions, this one is pessimistic,” he says, “perhaps too much so. On the other hand, there is much to be pessimistic about.” That statement is a bit casual when it is used as the basis of social prophecy; there always has been much to be pessimistic about, but not much of it licenses the expectation of the imminent extinction of American civil liberties. For all the scientific apparatus with which they are surrounded, Murray’s fears are closer to the ravings of Rush Limbaugh’s audience than to Tocqueville’s anxieties about “soft despotism.”

Herrnstein and Murray don’t explicitly contradict each other, to be sure, but Murray was hostile to the welfare state long before he encountered Herrnstein; and Herrnstein’s views on intelligence are in principle consistent with the politics of almost any persuasion from socialist to libertarian. Socialists might think that ineradicable differences in IQ should be met by making sure that the less clever were compensated with more education than the gifted, and with income supplements to make up for their difficulties in the competitive marketplace; libertarians might think we should treat such differences as the luck of the draw, no more worthy of treatment than the accident that makes some of us better baseball players than others. Between trying to obliterate their effects and letting them make whatever difference they make in the labor market, there are innumerable further alternatives.

Herrnstein and Murray have many common enemies—Head Start, open door immigration, unwed mothers, lax ideas about sexual morality, and the “dumbing down” of American secondary education—but The Bell Curve is very much not the work of one mind. Indeed, each of the authors is in more than one mind on more than one issue.

2.

For all its oddities, The Bell Curve is a fluent piece of work. It is a still more fluent piece of publicity-seeking. The authors have tried to have their cake and eat it, and they have succeeded in a big way. They—this is largely Murray’s achievement—claim to be frightened that they will stir up terrible controversy, but they have advertised their fears in such a way as to do just that. They insist that they have no urge to stir up racial dissension or give comfort to racists, but then say that their findings only reflect what people already think in their heart of hearts—which is, that blacks and white trash are born irremediably dumb, that black Americans have been overpromoted in the academy, that smarter white workers have been displaced by incompetent black ones at the behest of the federal government. A disagreeably wheedling tone is an unsurprising feature of such arguments.

There is a good deal of genuine science in The Bell Curve; there is also an awful lot of science fiction and not much care to make sure the reader knows which is which. What catches the eye of reviewers and reporters are Herrnstein’s gloomy predictions about the declining intelligence of the American population, and Murray’s prediction of imminent fascism. Fewer readers will notice the authors’ throwaway admissions that these predictions are highly speculative, and only loosely rooted in the data they assemble. Take the connection between the fact that illegitimacy rises as IQ declines, and Murray’s fears about the imminent collapse of the liberal state. It is, for a start, quite impossible—as is readily acknowledged by the authors—that the rising rate of illegitimate births in both the black and white American populations should in the first instance have had much to do with intelligence.

The rate remained almost stable between 1920 and 1960, at about 5 percent of all births, then took off sharply in the early 1960s to reach 30 percent in 1990. Herrnstein and Murray say, “If IQ is a factor in illegitimacy, as we will conclude it is, it must be in combination with other things (as common sense would suggest), because IQ itself has not changed nearly enough in recent years to account for the explosive growth in illegitimacy.” They then evade the obvious implication that their obsession with IQ is largely irrelevant. They say “some of these ‘other things’ that have changed in the last three decades—broken homes and the welfare system being prime suspects—interact with intelligence, making it still more likely than before that a woman of low cognitive ability will have a baby out of wedlock.” True, but largely beside the point; the social pressures they mention make it more likely that women of any degree of cognitive ability will have a baby out of wedlock. If the pressures operate more powerfully on women of lower intelligence, we want to know why this is so.

The interesting question is not one of genetics but one of changes in the culture; it is not what has happened to the intelligence of the mothers that needs explaining, but what happened in the early 1960s that so altered the incentives to have babies later rather than earlier and in wedlock rather than out. (It must mean something that divorce rates rose at the same speed during the same years.) That is the sociologist’s territory, not the psychometrician’s, and too often The Bell Curve relies on Herrnstein’s real distinction as a psychologist to prop up what is essentially armchair sociology. A sociologist would at least wonder why the welfare system should be one of the “prime suspects” in the rising rate of illegitimacy when it has been decreasingly generous over the past thirty years; and a sociologist would at least notice that other Western societies such as Britain and the Netherlands have experienced rising illegitimacy rates, too. None of this suggests we ought not to worry about the propensity of the less clever to get pregnant out of wedlock, but it does suggest that we ought to attend to the real complexities of the social environment in which all this takes place.

Again, all readers will grasp the authors’ insistence that Head Start programs haven’t worked; fewer will notice that those failures are more partial than the authors say, and that the failures provide a better argument for seeking programs that work than they do for The Bell Curve’s conclusion that we should abandon the attempt to raise the IQs of the disadvantaged and devote virtually all our attention to the highly intelligent. The fashion in which such programs have failed is not analyzed with the scrupulousness one might wish. In essence, The Bell Curve’s data suggest that Head Start and other preschool programs can raise children’s IQs quite sharply for a short period; once the children are in a regular school, their IQ scores drift back to something like the level they began at. For a believer in g, this is evidence that in the long run the quantum of cognitive ability, whatever it might be, simply reveals itself.

Someone who wanted to draw the opposite conclusion might think that the data only show that there is no cheap, one-shot environmental fix for deprivation. Environmental fixes are possible, but they take much longer to work, or where they work quickly, they need to be repeated so that they keep working. It may well be that a much more extensive transformation of the child’s environment than Head Start and preschool programs can offer is needed to effect lasting changes in intelligence. There are suggestive data about the impact of adoption on the children of low-IQ mothers that might make one believe that is the case. If it is true, however, it provides an argument for affirmative action that renders The Bell Curve irrelevant from start to finish; for it suggests that one purpose in creating a larger black (or whatever) middle class is to create a better environment for the next generation and its descendants. The true beneficiaries of affirmative action on this view would be the children and the grandchildren of the people promoted today.

That could be quite wrong; it might be that the only effective environmental fix would be a national health service that gave babies a better prenatal and perinatal environment. It might even be that Charles Murray’s “custodial state” would have to get into the act to insist that the mothers of children who are at risk should use such care. What one can certainly say is that the failure of Head Start to live up to its backers’ most extravagant hopes is neither a knock-down argument for genetic determinism nor any sort of argument for abandoning the disadvantaged. Herrnstein and Murray argue elsewhere in The Bell Curve that American secondary education has “dumbed down” bright children, and so imply—what they elsewhere admit—that bad environments at least have an effect. After several hundred pages of this, one begins to wonder just what Herrnstein and Murray do believe other than that any old argument against helping the disadvantaged will do.

3.

The sheer repetitiveness of its tables, graphs, and bar charts eventually dulls The Bell Curve’s impact for the conscientious reader; but Herrnstein and Murray do not expect—and perhaps do not really want—most of their readers to work their way through all 845 pages of their text.² They say they want to make the reader’s life easy. For readers whose minds go blank at the mention of multiple regression, they provide a wonderfully lucid appendix on “Statistics For People Who Are Sure They Can’t Learn Statistics.” For readers in a particular hurry, they summarize their claims in some forty pages of italicized text spread across their twenty-two chapters.

Their main claims can be boiled down further still. They are essentially these: America is today a “meritocracy” in the sense that the best predictor of success in life is IQ; the various institutions that pass children up the ladder to success increasingly select the brightest children to train for entry to the best colleges, the best professional schools, and the most rewarding occupations. Where once the alumni of Harvard and Princeton were socially rather than mentally smarter than their peers, the students of the best colleges are today almost off the scale—inside the top 1 percent of their age group. Nor does IQ represent the result of training, or parental advantage; the social standing of our parents is a less reliable predictor of our future economic success and failure than our IQ—it’s good to have well-off parents and brains, but if you can only have one, take the brains.

More intriguingly, most indicators of our ability to function successfully in society correlate to a significant degree with IQ. Very few students with an average or above average IQ fail to complete high school; conversely, the students who fail to complete high school usually do so because they find it intellectually beyond them; unsurprisingly, they have higher levels of long-term unemployment, both when they are able-bodied and because they are more likely to be sick. Men with lower IQs show up disproportionately in prison, and that is not because the dim crooks get caught, since self-reported but otherwise undetected crime is also largely committed by the less bright. Crime, of course, is mainly a male activity, and The Bell Curve duly acknowledges that what IQ explains is which men are more likely to commit offenses, not why men do and women (generally) don’t. Herrnstein and Murray’s interest in women is mostly an interest in their propensity to produce children out of wedlock, to go on welfare, and to have difficult children. As ever, the less bright have higher rates of illegitimacy and less amenable children, and to nobody’s surprise stay longer on welfare.

As for our relations with one another, the clever marry later, breed later, and stick together; the less bright marry in haste and repent in haste, or at any rate are twice as likely to get divorced within five years. One thing to remember in the face of all this—and usefully insisted on by the authors—is that IQ differences do not account for much of the difference in the fate or behavior of people; in statistical terms, IQ rarely accounts for as much as a fifth of the difference between one person and another, and usually for much less. The only thing with which IQ correlates very closely is our performance on tests that measure the same skills that IQ tests measure—which in a world full of lawyers and economists and scientifically trained professionals is surely likely to create a high-IQ caste of what Robert Reich labeled “symbolic analysts.” Intelligence tests test for just that kind of intelligence. To the extent that other personal characteristics are involved in what happens to us, the impact of IQ is less. The importance of any contribution of IQ to the causation of social problems, however, is that when we are dealing with very large numbers it makes a difference whether we think the population we are dealing with is averagely bright, especially bright, or rather dim.

For readers who are convinced that any discussion of the heritability of intelligence is fundamentally, if covertly, a discussion of the inferior mental capacities of black Americans, Herrnstein and Murray seem at first to provide some measure of reassurance. All these gloomy results about the damage done by having lower intelligence than the average come from an analysis of the experiences of white Americans in the 1980s. Most of the data which Herrnstein and Murray use come from the National Longitudinal Survey of Labor Market Experience of Youth (NLSY), a study of some 12,500 Americans who were between fourteen and twenty-two in 1979 when the survey began, and whose progress has been followed ever since.

Its usefulness to Herrnstein and Murray is that “only the NLSY combined detailed information on the childhood environment and parental socioeconomic status and subsequent educational and occupational achievement and work history and family formation and—crucially for our interests—detailed psychometric measures of cognitive skills.” The sample was used by the federal government to reassess its intelligence tests, so it fortuitously provides data on measured intelligence as well as on everything else that correlates with success and failure in the labor market. The NLSY covers all ethnic groups, but the first twelve chapters of The Bell Curve stick to the distribution of intelligence across the white American population in that sample. Only then do Herrnstein and Murray turn to the discussion of ethnic differences in IQ.

Of course, as the hubbub in the press suggests, the reassurance is less than skin deep; as soon as ethnic differences have been identified—the one that swamps all others is that the mean IQ of African Americans is 85 as against 100 for white Americans—the reader is in for two hundred pages of familiar complaints against affirmative action policies. Before we move on to these, some other findings are worth a brief look. The most familiar will be the fairly well-confirmed discovery that just as African Americans are one “standard deviation,” i.e., 15 percent, less good than white Americans at tests of analytical and spatial intelligence, so East Asians—especially the Hong Kong Chinese—are anything up to one standard deviation better. If the white American average is set at 100, the black American average is 85, and the East Asian average 111–115. Ashkenazi Jews have similar scores to East Asians, but the scores of Oriental Jews in Israel show an embarrassing contrast.

Herrnstein and Murray don’t dwell at length on the implications of their views for the social difficulties of black Americans, but they hardly need to. Once they have piled up the statistics on the disadvantages attendant on having an IQ much below 100, the case is made. Where they concentrate their attention is on the two related questions, whether we can do anything to raise IQ, and whether affirmative action policies in education and employment are worth the candle. In brief, their answer to both questions is no.

The greater part of the argument against remedial education is their argument against Head Start and analogous programs. But that argument, as we have seen, can be used to suggest that the programs should be more intensive, not abandoned. They acknowledge the possibility in principle of eugenicist programs, but flinch at the thought of putting into the hands of government the power to dictate such matters as who may and may not produce children—William Shockley gets a passing mention as someone who enjoyed shocking people by suggesting that we might pay the poor to be sterilized and might set up sperm banks to pass on the genes of geniuses (he contributed to a privately organized sperm bank: there is no record of the results). But Shockley is dismissed as excessively eccentric. Whether his proposal to pay the poor to be sterilized is more eccentric than Murray’s proposal to abolish welfare payments and face the short-term consequences for the hapless children on the receiving end of the change, readers will judge for themselves.

Affirmative action greatly preoccupies Herrnstein and Murray. Oddly enough in discussing it, they back away from an insistence on the genetic determination of IQ. All that matters is that IQ predicts performance at work and in the academy, and cannot be increased by short-term educational and environmental enrichment. In academic matters, they are much bothered by the probability that the SAT scores of black students at the best universities are anything up to 200 points lower than those of their white peers, with obvious consequences for the clustering of black students among the least successful and therefore least happy members of the college community. Herrnstein and Murray argue that we do such students no favor by putting them in a situation where they are anxious in school and possess an undervalued credential when they leave.

Nor are Herrnstein and Murray any happier about affirmative action in employment. They launch a two-pronged attack. The first is to demonstrate that although the raw income data suggest that black Americans earn less than white Americans, the picture changes when we add in the distribution of intelligence. At this point, we find that black Americans earn relatively more than white Americans—that is, relative to their IQs. What you might call “dollars per IQ point” comes out in favor of African Americans. If your notion of justice is that people should be paid according to their IQs, then this is unjust. On the other hand, you might think that what matters is overall efficiency; and Herrnstein’s other argument is that affirmative action damages efficiency. Given even halfway plausible assumptions, of course, it must do so; if IQ predicts competence, anything that makes us appoint people on some basis other than IQ produces some degree of incompetence. Old-fashioned class biases were denounced by British socialists precisely because they helped the incompetent to keep out the competent. Herrnstein advertises himself as an enthusiast for that view.

There is a lot to be said on both sides. In a highly competitive society like ours, it may be true that affirmative action causes anxiety in, say, the student who gets into a place like Princeton or Harvard with SATs well below those of his or her white peers. But this anxiety doesn’t seem to affect athletes or “legacies,” i.e., the children of alumni—groups whose presence at such places Herrnstein is surprisingly happy about—which suggests that even if this generation of black students does less well in strictly academic terms than their white peers, there are better ways of reducing their anxieties than refusing to admit them in the first place. The same thought applies in employment. It may be that there are many black Americans struggling with jobs they cannot deal with and many white colleagues muttering about them under their breath. It may also be that these are the labor pains of a different sort of society from the one we have had for the past three centuries.

4.

This, however, throws one back on the fundamental question raised by The Bell Curve—how seriously we should take its science. Is there an intelligence gap between black and white Americans that no passage of time and no social policy can close? If there were, would anything follow about the social policies a humane society should adopt? The answer seems to be that there is good reason to believe that there is a gap, but no conclusive reason to believe that it is unshrinkable; if there were, it would have a good many implications about the need to balance the search for efficiency against the desire for a more humane social order—but it would not dictate how we struck the balance and it would introduce no moral novelties into the calculation. In particular a belief in the importance of inherited differences in IQ need not encourage apocalyptic conservatism.³

It is an under-remarked feature of arguments over the inheritability of intelligence that an obsession with the presumed incapacities of the poor, the children of the slums, the bastard offspring of dim servant girls, and all the rest was once characteristic of reformers and sexual radicals as much as of anxious conservatives. The unwillingness of the contemporary liberals and the left to think eugenically has everything to do with racism being disgusting and not much to do with logic. In 1916 Bertrand Russell condemned the inner city as a site of “race suicide,” but meant only that the slums produced large numbers of undernourished, unfit, and ineducable children. It was a common hope of birth control pioneers that wanted children would be fitter and brighter children.

The label “eugenics” itself was coined by Darwin’s cousin, Sir Francis Galton, to describe a program for improving the British stock. Nor was the idea foreign to the Fabians. It is not an ignoble hope that as the welfare state improves the physical health of the citizenry it will also make them brighter, more alert, more interested in their surroundings and themselves. One could fear that the most likely end result would be Brave New World, with its Alphas, Betas, and Gammas, but that would not be because eugenic science was disgusting, but because the science would be used by the wrong people for the wrong reasons. It is surely true that an interest in the connections between heredity and intelligence need not be malign. After all, our interest in hereditary disease usually stems from the wish to help the sufferers. One can readily imagine a benign educational program that addresses the different strengths and weaknesses of students more intelligently than contemporary schooling does. But if a concern for inherited intelligence or the reverse need not be driven by panic and superstition, it usually has been.

In the United States, fear of new immigrants rather than optimism about the chances of raising the level of the whole population always seems to have driven the discussion; and fears of the “dilution” of the “pure-bred” white stock by Jewish or Negro blood were the common coin of academic discussion throughout the first forty years of this century. Herrnstein misrepresents this past and the complaint against it. He says that Stephen Jay Gould’s famous attack on psychometrics in The Mismeasure of Man was unfair to the military psychologists of World War I and to the psychologists on Ellis Island, whom Gould accused of announcing that on first testing 80 percent of the Jews, Hungarians, Italians, and Russians were feeble-minded, and that even on re-analysis, 50 percent were so. “The intelligence of the average ‘third class’ immigrant is low,” said H. H. Goddard, “perhaps of moron grade.” Gould, in essence, claimed that research into the supposed racial differences in intelligence was driven by panic and prejudice, and resulted in absurd findings. Herrnstein responds that the psychologists were looking only for mental defectives, and naturally reported cases of mental deficiency.

This, as Herrnstein knows, won’t wash in the case of Carl Brigham, the Canadian military psychologist who came to the United States in World War I and stayed to become professor of psychology at Princeton and a leading figure in the work of the Educational Testing Service and the development of the SAT. Herrnstein diverts the argument from the point at issue by claiming that Brigham’s book, A Study of American Intelligence, had less influence on the Immigration Restriction Act of 1924 than Gould supposed. But that is not the point. Brigham was a leading figure in World War I intelligence testing for the American Army, and Brigham held, and popularized, exactly the views that Gould complained of. Brigham knew that many people thought Jews were clever; having examined large numbers of Russian-born Americans in the war, he thought he knew better. His army sample was “at least one-half Jewish,” he thought, and they “had an average intelligence below those from all other countries except Poland and Italy.” Taking Negro draftees as his reference, he discovered that 39 percent of the Russian-born were below the Negro average, 42.3 percent of the Italian-born, and 46 percent of the Polish-born. This was not an eccentric’s vision of the world but the respectable psychometrics of the day.⁴

Brigham’s estimates of the cognitive inferiority of black Americans were, as this would suggest, much greater than Herrnstein’s—in which case, the unclosable gap has been closing. Even Brigham acknowledged that putting black Americans in a different environment made a difference to their tested intelligence; and other interwar studies confirmed Brigham’s finding that northern blacks did better on his tests than southern blacks. Some even found that northern blacks scored higher on many tests than southern whites. None of this proves that there is no inbuilt difference in cognitive abilities between different human groups, though it is hard to believe that anything of the sort would follow the haphazard lines of self-reported ethnicity. What it does suggest is that either relative cognitive abilities change more rapidly than Herrnstein believed or that our estimates of them are less reliable than he thought.

One other thing it suggests is that we should worry less than Herrnstein did about the danger that American intelligence is declining. Herrnstein was an old-fashioned “deteriorationist,” squarely in the Brigham tradition.

When people die, they are not replaced one for one by babies who will develop identical IQs. If the new babies grow up to have systematically higher or lower IQs than the people who die, the national distribution of intelligence changes. Mounting evidence indicates that demographic trends are exerting downward pressure on the distribution of cognitive ability in the United States and that the pressures are strong enough to have social consequences.

Other evidence, also quoted by Herrnstein, suggests that intelligence levels are generally rising. As usual where the evidence points in both directions, Herrnstein and Murray urge us to accept the more frightening scenario.

Herrnstein’s fears were partly those that recently alarmed Sir Keith Joseph, Mrs. Thatcher’s former education minister. In Britain as elsewhere, the cleverer members of the population have fewer children than the less clever. If g is handed down in the genes, there will be less to go round in each generation. Even if each clever woman had as many children as each less clever woman, there would still be deterioration; the less clever have their children earlier, producing three generations of less bright children, while their intellectual superiors produce two. But Herrnstein also shared Brigham’s more American anxieties: the wrong sort of immigrants have been flooding into the country. Small numbers of bright East Asians were no match for large numbers of less bright Latino and Caribbean migrants. Herrnstein knew that his critics would retort that all this was said about the Poles, the Russians, and the Italians a century ago; all he could say in reply was that this time the anxiety was justified.

The latent contradiction of The Bell Curve’s politics emerges when one contrasts Herrnstein’s enthusiastic defense of meritocracy with Murray’s final fantasy of a world in which we live in “clans” that are high on self-regard and cheerfully ignore the existence of cleverer and less clever people in the world. Herrnstein essentially wants the world in which clever Jewish kids or their equivalent make their way out of humble backgrounds and end up running Goldman Sachs or the Harvard physics department, while Murray wants the Midwest in which he grew up—a world in which the local mechanic didn’t care two cents whether he was or wasn’t brighter than the local math teacher. The trouble is that the first world subverts the second, while the second feels claustrophobic to the beneficiaries of the first. The authors are united only in their dislike of the mostly unnamed liberals who have been hostile to Herrnstein’s obsessions with race and to Murray’s obsessions with the welfare system. In short, The Bell Curve is not only sleazy; it is, intellectually, a mess. ⁵