nce upon a time, before science and society got into bed together, serious attention was given to the question of dangerous knowledge. First it was an issue between philosophy and the city (e.g., Athens against Socrates), later between science and biblical religion (e. g., the Church against Galileo). Keenly aware of knowledge's power to harm as well as help, even the great founders of modern science advocated self-censorship and practiced the art of veiled writing, not only to avoid persecution, but also to protect those who might be harmed by their "dangerous" truths.
Rene Descartes, mindful of the trial of Galileo but also eager to ward off unworthy followers, suppressed his radical scientific treatise, Le Monde, anonymously publishing instead a "mere autobiographical fable," The Discourse on Method, which subtly tells only the discerning reader the core of what Le Monde contained. And, in The New Atlantis, the ideal community ruled by his new practical science of nature, Francis Bacon has the scientists deciding among themselves which discoveries to make public (i.e., to publish) and which to keep secret.
Today, in our liberal, open, and "enlightened" society, we may worry about the bad effects of technology, but we believe wholehearted]y in the goodness--or at least the innocence--of knowledge. With us there is virtually no thought or opinion so dreadful or offensive that it cannot be--and is not--shouted from the rooftops. Thus, the issue of dangerous knowledge is now quite limited in scope and contested only by intellectuals. Yet they do so arbitrarily: the very same people who scream freedom of expression for Snoop Doggy Dogg or Robert Mapplethorpe, or who cannot understand why there is a contract out on Salman Rushdie, would be only too happy to silence--or academically lynch--Charles Murray and his late partner, Richard Herrnstein, for publishing their findings about the vexed subject of intelligence, genetics, and race. The vicious attacks on The Bell Curve[a] and its authors have been an ugly spectacle. But their enemies are right at least in this: there are vital matters at stake and the issue of dangerous knowledge is in this case well worth some attention.
Almost 25 years ago I witnessed up close two acts of censorship practiced by no less than the National Academy of Sciences, the American Olympus of scientific distinction. A report entitled Assessing Biomedical Technologies--I, as executive secretary, had drafted it for the National Research Council's Committee on Life Sciences and Social Policy--was censored by the Academy's Report Review Committee, largely on the grounds that its publication might frighten Congress into cutting off all funds for biomedical research. At the very same time, an ad hoc committee of the Academy, chaired (if I remember rightly) by the eminent geneticist, Theodosius Dobzhansky, decided that the Academy should not commission a study--called for by Nobel laureate and Academy member, William Shockley--into the relation between race and intelligence. Nothing such a study might discover, the ad hoc committee argued, could, if known, possibly do anybody any good.
Though I was appalled at the cowardly censorship of our, in truth, very bland document, and though I was amused to see this double-barreled suppression of thought and inquiry by a collegium that had only scorn for the Church's suppression of Galileo, I remember being very impressed by the prudent and statesmanly report of Dobzhansky's committee, with whose conclusions I then agreed. It seemed to me then that a society founded on the self-evident truth of human equality--the equal dignity of each human being--had no business ranking racial groups, especially on the basis of alleged "scientifically measurable differences" in the powers that most make us human.
Times have changed. Race-consciousness is now rampant, no longer condemned but instead insisted upon by the loudest partisans of equality. For us, equality stands no longer as a founding faith in the rights of individuals, but as a fanatically sought-for sameness of result in the wealth, status, and power of groups. Blocking the march to equality, however, is a new, electronically transformed world that increasingly penalizes those who can't keep pace (and their children), despite the well-meaning, but unsuccessful, programs devised for their advancement by the "best and the brightest," those who have smartly risen to the top. Perhaps a case can now be made that we do indeed need to know the truth about intelligence, its heritability and malleability, and its relation to our social dilemmas, including those connected with race. This, indeed, is the view taken by Herrnstein and Murray. Before judging the wisdom of their effort, one should try to understand what their book says and means, in its own terms.
Someone who has not read the book, but "knows" it only from the largely irresponsible things written and said about it, will be surprised to discover that The Bell Curve is not primarily about race. Neither does it teach that genes (fully) determine intelligence or that intelligence determines one's destiny. It remains loudly agnostic about whether, and to what extent, observed racial differences in IQ have any genetic basis. Also, its presentation of the scientific evidence is scrupulously separated from--and not driven by--its (limited) public-policy suggestions.
Herrnstein and Murray write with great clarity and unusual care. Immensely difficult and technical subjects are made accessible, often with the aid of marvelously apt examples or analogies. Premises are explicitly stated (for example, about which concept of intelligence they adopt and why), methods are explained, data are thoroughly, yet cautiously, interpreted, and arguments about their significance are presented explicitly and fully, yet with admirable recognition of the limits on what the data allow one to conclude. Evidence on all sides of controverted questions is always presented, and, in this reader's judgment, with remarkable even handedness and judiciousness. The tone throughout is sober, measured, concerned; there is nary a note of smugness or condescension. Whether its conclusions prove true or false, whether its interpretations are wrong or right, and, indeed, whether it should or should not have been published, The Bell Curve is an impressive work, written in the best academic (social) scientific spirit--animated entirely by the desire to know the scientific truth about these devilishly tricky and delicate subjects.
Well not quite entirely. The pursuit of the truth here is not simply disinterested. Herrnstein and Murray write also, if not in fact mainly, to address current disruptions of American social life and to rectify what they regard as the failure of social scientists, journalists, and politicians to diagnose or treat our social ills correctly:
They examine changes in the economy, changes in the demographics, changes in the culture. They propose solutions founded on better education, on more and better jobs, on specific social interventions. But they ignore an underlying element that has shaped these changes: human intelligence--the way it varies within the American population and its crucially changing role in our destinies during the last half of the twentieth century. To try to come to grips with the nation' s problems without understanding the role of intelligence is to see through a glass darkly indeed, to grope with symptoms instead of causes, to stumble into supposed remedies that have no chance of working . . . . IT]here can be no real progress in solving America's social problems when they are as misperceived as they are today.
The Bell Curve's science may not be contaminated by its public purpose, but its goal is not truth for its own sake but for the sake of social welfare and public good. Its publication even more than most scientific publication--always an act of making public--is an emphatically political act. This means that it must be judged by more than scientific criteria. Indeed, the authors invite us to judge their publication by the good that it might--in their view, will--bring: "What good can come of understanding the relationship of intelligence to social structure and public policy? Little good can come without it."
What are the main teachings of this book? The big picture is the emerging social stratification of American society largely on the basis of intelligence, rather than inherited wealth or social class of origin. At the top, a cognitive elite, a meritocracy of the smart (not necessarily the wise or the good), is increasingly affluent, powerful, and more than everyone else free to enjoy the privileges and delights of our increasingly complex and technological society. Educated together in our most prestigious colleges and universities, engaged in occupations highly screened for IQ, intermarrying more and more only with one another, and living and working apart from the rest of society, and especially from contact with the people and problems of the growing underclass, this class is comfortably but dangerously isolated from the mainstream of American life.
In contrast, at the bottom, a cognitive underclass is mired in poverty and especially subject to all the social ills that clamor for public attention. In the most impressive part (II) of the book, the authors analyze the relation between cognitive class and social behavior and show that low IQ, even more than low socioeconomic status of one's family of origin, is highly correlated with comparatively high risks of trouble: poverty, dropping out of school, being unemployed or falling altogether out of the labor force, divorce, illegitimate births, welfare dependence, criminal activity, "malparenting," and having children who are also intellectually handicapped. These findings have nothing at all to do with race; the evidence comes from data collected only for non-Hispanic whites.
Later, viewing the American population as a whole, the authors also show that low IQ is especially prevalent among people who have these problems: people from the lowest 20 percent in intelligence (IQ less than 87) account for much or most of the poor (48 percent), the high school dropouts (67 percent), the jailed and imprisoned (62 percent), chronic welfare recipients (57 percent), mothers of illegitimate children (52 percent), mothers of children living in poverty (63 percent), and mothers of children with IQs under 81, the lowest declie (72 percent). The disintegrating American family, often blamed for many of these ills, has crumbled most for people in the lowest cognitive strata.
The data are sobering, the conclusions disquieting, but, to this point, no one could have reason to object. Most people know that intelligence goes a long way in today's world; most readers of this magazine are probably where they are because they were dealt good intellectual equipment, which was then esteemed and cultivated. Most people, if they thought about it, would suspect what Herrnstein and Murray have convincingly shown: very low cognitive ability is a severe handicap in a high-tech world, especially when mores are loosened and when the collapse of families and neighborhoods compel people increasingly to fend for themselves. The Bell Curve also provides evidence to support clearly what any savvy employer already knows: on the job, there is no known substitute for intelligence. A high IQ predicts proficiency of job performance better than any other tested variable, including level of education, interview results, or college grades.
The fuss begins because Herrnstein and Murray also discuss genetics and IQ, the chances of making children smarter by social intervention, and the explosive subject of intellectual differences among racial (they call them "ethnic") groups. They argue, against prevailing public orthodoxy but in accord with most scientific opinion, that general intelligence ("g") is, for individuals, highly heritable--somewhere between 40 percent and 80 percent; they opt for an estimate of 60 percent. (This means, of course, that a sizable portion of cognitive ability depends also on environment and education--who could think otherwise?) They disappoint those who hope to raise cognitive ability in the very dull by arguing that, short of adoption at birth, there is--at least for now--no evidence that altering environment can significantly raise a child's general intellectual ability. And they summarize findings regarding "ethnic groups" that show Asian Americans to be somewhat more intelligent and African Americans significantly less intelligent than non-Hispanic whites.
Because they believe that individual IQ is largely tied to genetic endowment, and because they do dare to discuss the question of genetics and race, they have been (falsely) accused of teaching that these group differences in intelligence are also genetic in origin. In fact, they clearly declare themselves agnostic on this issue. However, they rather astonishingly claim that it does not matter if it should turn out that the black-white IQ difference is indeed genetic, even largely or wholly So.
The final part (IV) of the book includes a chapter exposing the "dumbing down" of American education, in consequence of lowered standards, competing (noneducational) agendas, and a virtual neglect of the gifted; another that assesses the gains and losses of affirmative action in higher education, pointing out costs of the extensive "race norming" now practiced by elite universities in their admissions policies, including increased minority dropouts, low morale, and the diluted "worth" of their college degree; and a chapter that reviews consequences of affirmative action in the work place, showing that, controlling for IQ, blacks have been hired at higher rates than, and for comparable salaries as, whites (of equal IQ) since the late 1960s, and arguing against racial preference in hiring, on grounds both of fairness and benefits to productivity. The book's last chapter, "A Place for Everyone, " presents policy recommendations for how we might live harmoniously, despite fundamental individual differences, in a world in which cognitive social partitioning will continue to increase.
What to think of all this? My reaction is mixed. On the one hand, I am impressed by the data regarding the reality of cognitive stratification and the contributions low intelligence no doubt makes to our most pressing social problems. I am made sensible of the need for realistic expectations about how much social change in the "underclass" we can accomplish, say, by income redistribution or by pouring more money into schools. I am moved not only by the courage and clear mindedness of the authors, but even more by their passion for the public good, which, in fact, informs the entire work.
On the other hand, I am deeply saddened and troubled by the book: by its reminder of the unfairness of it all; by the implication that- -if it speaks truly--there is little that hard work and the best will in the world can do to rescue many of my fellow citizens from a miserable life, which they not only suffer through no fault of their own but which they also will pass on to their children; and, worst of all, by the profound sense that all of this knowledge, because it is at best very partial knowledge, may not do us much good at all, indeed, may even make matters worse. Let me raise three kinds of misgivings. Each in its own way reveals a weakness in the social-scientific outlook and approach so well championed in this book.
First are questions about intelligence itself and the validity of attempts to isolate and measure it and to evaluate its role in life abstracted from other qualities of soul. Herrnstein and Murray partly address these questions (in the Introduction) and give practical (not psychological or philosophical) reasons why they adopt what they call the classical view of intelligence as a "structure"--"a general factor of cognitive ability on which human beings differ," most accurately measured by IQ tests expressly designed for this purpose. They claim, but make no attempt to show, that scores on these tests match what we mean, in ordinary speech, when we call someone intelligent or smart, dumb or stupid.
Sharing modern science's general aversion to questions of being or "essence," the authors care little for the question, "What is intelligence, " contenting themselves with having what they regard as its quantitative measure. But this quantified measure is readily reified and comes to be taken for the thing itself. However useful, this abstracted and reductive measure is not itself the mysterious and multifunctional power (or powers) of mind or soul.
What, really, is intelligence? How is it related to the powers of calculation? deliberation? inference? deduction? discernment? intellectual intuition? cleverness? awareness? imagination? aesthetic appreciation? inventiveness? shrewdness? insight? perspicacity? curiosity? reflection? inquiry? self-consciousness? conscience? moral judgment? Is intelligence as measured really the same as intelligence as lived? Are the intellectual gifts suitable for the higher mathematics related to the smarts necessary for prudent shopping or the moral intuition regarding right conduct? If so, why are so many people with high IQs "stupid" about human affairs? Why are people of average IQ often more sensible or reasonable--or better parents-than their "more gifted" peers? Should we really accept some numerical measure of some homogeneous "g" (general intelligence) as the common basis of the Apollonian gift of insight, the Promethean gifts of forethought and craftiness, and the Athenaean gift of sharp- eyed prudence?
Nearly 400 years ago Descartes invented the mathematical notion of quantity, usable everywhere, to make possible the grand project of modern science, ultimately useful for the conquest of nature. He asked readers to forget their interest in the being of things, for example, of color. What harm would there be, he asked, if we represented the differences among red, white, and blue as being like the differences of three geometrical figures? Plenty of harm: from there it was but a short intellectual step to reducing color colorlessly to wavelength, a move which does provide us predictive power and some control but only at the cost of a shrunken, abstracted, and distorted understanding of ourselves and the world in which we live.
Mysteriously, science's reductive approach works--both in general and here with IQ. Strangely, even though psychometry is deliberately indifferent to the psyche that it measures, its measurements do more or less accurately capture some abstractable features of intellect, whatever it is. Yet, the predictive power of knowing IQ for any individual is, as Herrnstein and Murray readily agree, very limited: everyone' s fate turns on much more than measurable "cognitive ability." And IQ's statistical predictive power for populations is at once not very surprising and not very deep, given that the dependent variables with which IQ is correlated are likewise mainly quantitative and abstract.
The numbers correlate partly because the world has been reconceived in terms of our reductive modern science, of which the science of psychometrics is itself a part. In a world conceptually and technologically re-created on the basis of mathematical physics, whose truths are found by reducing quality to quantity and manipulating it through equations, and in a world run by people who reduce all thinking to problem-solving, small wonder it is that the problem-solving sort of intelligence confers some selective quantitative advantage, other things being equal.
Other things are not always equal. Indeed, as the authors point out, success in life--even success as reductively measured by social scientists- -depends on many things, external and internal. Statistical correlation is not yet causation, certainly not in individuals, not even in groups. All things importantly human resist scientific simplification. Some aspects of intelligence may be assessed quantitatively, but what, for example, about character?
Character--the habits of the heart, the disposition of our loves and hates--surely counts a great deal for keeping a job, obeying the law, caring for one's children. Granted, a strong back and a willingness to work hard may not get one as far today as it once did. Granted, character may even depend somewhat upon intelligence: the ability to love and choose the right thing depends partly on the ability to discern it and largely on having been reared by people smart enough at least to provide adequate moral education. Still, one would have liked the authors to acknowledge the centrality of character as the cause of conduct and to confess the inability of quantitative social science to treat it properly. Where the authors do brush up against the topic, they are in over their heads. (See, for example, their silly suggestions about the connection between verbal skills or reading ethics books and acquiring "the habit of virtue.")
Second, some questions about policy implications. Herrnstein and Murray offer recommendations to "deal with the twin realities that people differ in intelligence for reasons that are not their fault and that intelligence has a powerful bearing on how well people do in life. " Their goal: to let each person--smart, dull, or in between--find a "valued place" in our society. Toward this end, they want us to cease treating individuals largely as members of ethnic groups, restore social functions to neighborhoods and municipalities, simplify the rules of life (largely) by reducing governmental interference, and make it "easier to live a virtuous life." For the latter, they make recommendations about crime and marriage: simplify the criminal justice system and return marriage to its formerly unique legal status. They want for everyone what they claim the cognitive elite now enjoys, life in community, "where being a good parent, a good neighbor, and a good friend will give their lives purpose and meaning."
The authors confess that their scientific analysis could just as well support opposite, egalitarian proposals, aimed at making up for unavoidable natural inequalities; their programmatic suggestions are based instead largely on personal philosophical preferences. I share these preferences and wish for the same goals, but I find the proposed remedies naive: much too weak for the problems, and, even then, very hard to come by.
The authors seem not to recognize that it takes more than the dismantling of the welfare state to restore neighborhoods, law-abidingness, civility, sexual self-restraint, strong marriages, devotion to children, and a valued place for all to a society in which the biggest rewards go for achievements of the mind rather than the heart. How can we hope to find a meaningful sense of place when the fabric of fundamental human relations is in tatters because of individualism, mobility, and the shallow artificiality of much of modern life? Rapid and life- transforming technological changes--especially in transportation and communication--have homogenized the country, obliterated the sense of natural place, weakened human ties, and undermined real civic participation (especially in our cities); people freed from mind-numbing toil are turning more and more to private affairs and mind-numbing amusements. Sexual license will be hard to reverse, given contraception, abortion, and the absence of public shame; home life will remain strained and voluntary charitable activities grossly understaffed if both husbands and wives are employed full-time. Real "places" are not constructed, they are bequeathed and inherited. The information superhighway is not a neighborhood.
To be fair, Herrnstein and Murray set out mainly to diagnose our situation, not to remedy it. But they claim that nothing good can come without understanding the relationship of intelligence to our social structure and public policy. Perhaps so. But the good they hope to accomplish neither depends on their analysis of intelligence nor can that analysis show us how to fix the problems associated with cognitive stratification, itself no doubt only a symptom of the cultural crisis of advanced technological society. Public policy--like rapid technological change- -can help destroy a culture; it has, to my knowledge, never yet re- created one.
Finally, then, was it wise to publish this book, and especially its discussion of racial differences in IQ and their possible genetic origins? The authors, who clearly take delight in speaking frankly (says Murray: "for once, there would be no euphemisms, no self-censorship, no ducking of tough questions"), have in fact considered this question. They acknowledge the risks, but have their reasons for speaking up. First, they wish to challenge the unexamined assumptions of our social policy concerning race: a belief in the genetic cognitive equality among the races and the possibility of equalizing existing socioeconomic differences by social intervention. Further, by breaking the taboo against public speech regarding genes, intelligence, and race, they hope to correct wrong and dangerous racist private opinions, which, they allege, are now widely held and wildly different from our (hypocritical) public orthodoxy of equality.
Insufficiently appreciative that hypocrisy is the tribute vice pays to virtue, they seek the same (alleged) benefits that came from breaking the Victorian taboos against public chatter about sex (to which they compare ours about racial differences in IQ). Herrnstein and Murray want to let it all hang out: "Taboos breed not only ignorance but misinformation." A richer social science, seeking wisdom and not just statistical correlations, might understand that taboos--including the Victorian ones--are often the embodiments of reason and goodness. Our taboo about race, genes, and IQ seems to me profoundly wise.
Scientists Herrnstein and Murray deceive themselves into thinking that the world can treat intelligence "as just a noun, not an accolade. " No one takes the term "stupid" as a neutral descriptive, much less as a compliment; as the authors themselves show repeatedly and emphatically, "smarts"--not virtue--are increasingly the prized coin of the realm. Also, almost no one is entirely able to keep statistically based stereotypes of a group from influencing their perceptions or expectations of its individual members, especially for those whose membership in the group is visible at a glance, knowable before anything else can be known about them. The authors claim, speaking, it seems, only for their highly rational selves, that it "matters little" whether the now-considerable black-white difference in IQ is environmental or genetic in origin: "We cannot think of a legitimate argument why any encounter between individual whites and blacks need be affected by the knowledge that an aggregate ethnic difference in measured intelligence is genetic instead of environmental" (my emphasis).
But life as lived is based not on "legitimate arguments" but on opinions and prejudices; and "need not be" rarely translates into "will not be." Precisely because most people do not--and probably cannot and will not--refrain from stereotypical thinking, and precisely because intelligence is so central to our humanity, it cannot be good for living together to go around broadcasting the low group IQ of blacks or Hispanics, or of Poles or Slovaks, for that matter. Few individuals (smart or dumb) belonging to such a group, "known" or "thought" to be intellectually inferior--even if through no fault of its own--are likely to live unencumbered by such opinions, once it becomes noised about and taken as truth.
Herrnstein and Murray are not responsible for our increasingly racialized political thought. On the contrary, it is much more their opponents- -partisans of affirmative action; teachers of the ruling importance of race, class, and gender; preachers of the divisive sorts of multiculturalism and ethnic tribalism--who have betrayed the American ideal of the dignity of the individual, and, alas, so tragically close to the time at which that ideal was finally being made available to all people, regardless of race, creed, or color. But in Herrnstein and Murray' s attempts to restore the focus on the individual, to replace the utopian drive for equality of outcomes with the dignified equality of opportunity, and to overcome tribalist thinking with a public-spirited concern for all our fellow citizens in the accepted presence of manifest inequality, they instead contribute--to be sure, unintentionally-- to the very sort of poisonous racial thinking they oppose.
The excesses of affirmative action can and must be opposed on moral and political grounds--as unjust, harmful, and un-American--without trying to show that it can never work because of the intellectual inferiority of blacks. The authors of The Bell Curve seek to overcome racialistic thinking in the long run by requiring it in the short run. In this respect, they do exactly as their misguided opponents have done in demanding minority set-asides and quotas--and they should be willing to accept the blame for the consequences that follow.
As these are not cowardly men, I rather think they (now, only Murray) will be willing to take the brickbats--if they are in fact deserved. They have been scandalously and unfairly vilified and falsely accused of saying things they have not said. But it is part of wisdom to understand that one is partly responsible for how one is misunderstood. And no true student and friend of the American republic ought to be surprised at the uproar.
Murray, in a Wall Street Journal article responding to his critics, defends the decision to publish against the charge of irresponsibility:
We said to ourselves that the question of irresponsibility must finally be determined by truth. As Mr. Herrnstein said shortly before his death: If what we say in The Bell Curve is not true, then there would be no responsible time to publish it. If what we say is true, there is no irresponsible time to publish it.
However admirable the sentiment, I can think of no more telling indictment of our value-neutral social science. Having banished virtue and justice from its field of study, it blindly confuses the factual with the good. A truer wisdom about human affairs knows when and how to speak, and when to keep silent. A high IQ and a Harvard education are clearly not enough.