A Cartoon Elite
by Nicholas Lemann
From Young came the idea that not long after a society institutes mass educational sorting based on the results of IQ tests, a distinct high-IQ ruling class will begin to emerge. Because of the tendency of people in this class to marry fellow students at highly selective universities and pass on their IQ-rich genes to their offspring, over time the meritocratic upper class will more and more resemble a hereditary aristocracy. If this class absorbs the left-wing views that prevail in the universities, then once it is in power, it will resemble the arrogant Communist bureaucracy that was the subject of Djilas's book. The emotional charge of conservative attacks on organizations like the Corporation for Public Broadcasting and the National Endowment for the Arts, and on more-amorphous institutions like the news media, Hollywood, and the Washington establishment, comes from the idea that they are made up of superior-feeling, isolated New Class members who want to force their cultural mores on ordinary folks whom they despise.
Populism usually arises from a general discontent that precedes the identification of a specific villain. People feel that things are out of control, socially and economically. Men without higher education, especially, have dramatically lost ground over the past two decades. The idea that this is the fault of the meritocratic elite is an easy sell.
The pampered meritocrat has become the contemporary equivalent of the Organization Man of the 1950s -- a symbol of the age for purposes of middlebrow hand-wringing. A typical description of the type comes from an article by Michael Lind in Harper's Magazine.
The closer you come to the centers of American politics and society, the more everyone begins to look the same. . . . the people who run big business bear a remarkable resemblance to the people who run big labor, who in turn might be mistaken for the people in charge of the media and the universities. They are the same people. . . . most of the members of the American elites went to one of a dozen Ivy League colleges or top state universities. . . . They talk the same. They walk the same.
Christopher Lasch's last book, The Revolt of the Elites (published posthumously), limns the meritocratic elite -- which he called "the new aristocracy of brains," or "the knowledge class" -- but in a much more venomous way. In his previous book, The True and Only Heaven (1991), Lasch presented himself as the champion of a provincial lower-middle-class culture in which the highest good is community, not the fulfillment of ambition, and identified with a "petty bourgeoisie" of "small proprietors, artisans, tradesmen, and farmers," which he admired for "its moral realism, its understanding that everything has its price, its respect for limits, its skepticism about progress." In The Revolt of the Elites he did not change this position at all, but he directed his energies toward damning the people he disliked rather than praising the ones he admired. Also, although all of Lasch's books have a patchwork quality, The Revolt of the Elites especially does: it is a slim collection of reworked articles, and so careens from subject to subject. This, along with the strength of his animus toward his subjects, means that Lasch never patiently laid out a theory of who the members of the elite are and where they came from.
It's easy, though, to extract a picture of them from the book. They have a "growing insularity," they inhabit "an artificial world," they spend too much time talking, exercising, and going to restaurants, they are excessively mobile -- "at home only in transit, en route to a high-level conference, to the grand opening of a new franchise, to an international film festival, or to an undiscovered resort" -- and some partake of a sob-sisterish brand of social-issue liberalism,
a liberalism obsessed with the rights of women and minorities, with gay rights and unlimited abortion rights, with the allegedly epidemic spread of child abuse and sexual harassment, with the need for regulations against offensive speech, and with curricular reforms designed to end the cultural hegemony of "dead white European males."
The implication is that American life can be understood as a grand struggle between these people and the petit bourgeoisie: everything that helps the former hurts the latter. For example, "feminism's appeal to the professional and managerial class" is simply that it "provides the indispensable basis of their prosperous, glamorous, gaudy, sometimes indecently lavish way of life." Meanwhile, in the lower-middle class feminism doesn't bring gaudiness, but it does mean that the children wind up parked in front of the television set. The elite gets richer, and ordinary people lose ground economically. The elite migrates, and ordinary people remain in their deteriorating neighborhoods. The elite promotes cosmopolitanism, and ordinary people feel their steady, traditional, provincial civic life slipping away.
The origin can't be precisely pinned down -- it's part agitprop; part Robert B. Reich's influential book about the importance of "symbolic analysts," The Work of Nations (1991); part the ascent of super-meritocrats Bill and Hillary Clinton to the White House. The most coherent theory of the rise of the meritocratic elite can be found not in Lasch (he's too scattered) but in Richard J. Herrnstein (also now deceased) and Charles Murray's The Bell Curve (1994), which, in addition to the controversial material on race and IQ, contains a long, little-commented-on section delineating the rise of the meritocratic elite. Because it is difficult to evaluate a phenomenon that has not been precisely defined, the Herrnstein-Murray argument is worth laying out.
Until the midpoint of the twentieth century, Herrnstein and Murray say, people of high intelligence were scattered more or less randomly through the social structure. But during the 1950s a distinct "cognitive elite" began to emerge, because the highly intelligent embarked on an "invisible migration" to top universities and thence to positions of power. This happened because American society had become so technologically complex that it could no longer operate without a high-IQ leadership: "A true cognitive elite requires a technological society." The new elite's rise represents a dramatic change: "Cognitive stratification as a central social process is something genuinely new under the sun." Today, with the cognitive elite controlling so much of the economic apparatus -- and no wonder, because "intelligence is fundamentally related to productivity" -- we are seeing take shape "an unprecedented coalition of the smart and the rich."
To people who have been through the kind of cognitive-elite life trajectory that Herrnstein and Murray describe -- straight As, high SAT scores, admission to a selective college and graduate school, and off to the races -- the story makes intuitive sense. It fits Herrnstein and Murray personally; both came from obscure backgrounds to Harvard (college in Murray's case, graduate school in Herrnstein's) during its meritocratizing period. In the old days the Ivy League colleges and the institutions into which they fed -- upper academia, Wall Street, big law firms, research hospitals -- had a distinctly clubby, Episcopalian cast; they were filled with the kind of person Calvin Trillin (Yale '57) refers to as Thatcher Baxter Hatcher. Now they're run by people who feel they have earned rather than inherited their high positions. If you're at, say, McKinsey & Company, surrounded by people with personal histories similar to your own, the temptation to accept the cognitive-elite theory at face value is nearly overwhelming.
However, the theory has two major shortcomings: the relative unintelligence of all preceding elites is impossible to prove, and the degree to which the upper-income stratum is now made up of people with high IQs and prestigious educations is exaggerated. These are not just dry conceptual problems. They have distinct political consequences.
The idea that intelligence used to be randomly distributed across the class structure is problematic for Herrnstein and Murray. They contend that what IQ tests measure is general intelligence, or g in the impressive-sounding shorthand -- the most important human talent, which would have been found at above-average levels even among "the chief ministers in Cheops' Egypt." If you take this argument too far, you wind up saying that there was always a cognitive elite. How could it have failed to emerge earlier, if g is crucial? Herrnstein and Murray's answer is that only a technological society can coax a true cognitive elite into existence. But society has been becoming steadily more technological for centuries, so shouldn't the cognitive elite have taken form gradually over time, rather than overnight in the 1950s?
Herrnstein and Murray say that the mechanism for the creation of the cognitive elite was higher education. In The Bell Curve persuasive charts show people high in IQ becoming more likely to go to college over time and also bunching up in the elite colleges, but the underlying data are weak. For example, Herrnstein and Murray's figures on the (relatively low) average IQ scores at Ivy League schools in 1930, when pursued through the footnotes, seem to come from the first administration of the Scholastic Aptitude Test, to 8,040 students on June 23, 1926, and the subsequent conversion of SAT scores to IQs. The takers were not students at Ivy League colleges; they were high school students. What Herrnstein and Murray report as the average IQ of graduates of the Ivy League and Seven Sisters colleges appears actually to be an average of the SAT scores of high school students who told the test administrators they'd like their scores sent to these colleges.
By far the best statistical support for the idea of the scattered nature of the cognitive elite prior to the Second World War comes from the Pennsylvania Study, a large research effort conducted in the 1920s and 1930s by the Carnegie Foundation for the Advancement of Teaching. Ambitious as it was, the Pennsylvania Study was not universally accepted in testing circles, because its authors, William S. Learned and Ben D. Wood, were throughout the period of their research active proselytizers for the use of educational admissions tests to separate the cognitive wheat and chaff. Carl Brigham, the creator of the SAT, wrote to James Bryant Conant, the president of Harvard, in 1938,
Dr. Learned and I long ago agreed to disagree on the results of this Pennsylvania Study. . . . He finally agreed that his conclusions could not be justified from his data, yet he insisted that the conclusions were good propaganda for the educational world and he has continued to preach them. . . . Dr. Learned feels that certain desirable social results may be obtained even though the methods of obtaining them are wrong.
An alternate explanation for the emergence of the cognitive elite would be that the key variable was not so much the personnel demands of a technological society as the development and promotion of mental tests. Probably before there were mental tests the intelligent were overrepresented at the upper end of the social structure; the advent of testing -- not technological change -- made the fit tighter in certain professions. As soon as mental tests are used as a screen for entry into a field, by ironclad tautology that field will be dominated by people who score high on mental tests. With typical assurance Herrnstein and Murray report that "only people from a fairly narrow range of cognitive ability can become lawyers," but to the extent that that is true, it's because a cognitive measure, the Law School Aptitude Test, is the tollbooth through which all aspiring lawyers must pass, not because you have to be intellectually gifted to draw up wills and deeds of sale.
HERRNSTEIN and Murray claim that business, especially the corporate world, fits their model: formerly run by people who had the right connections and may or may not have been very bright, today it is controlled by the cognitive elite. Their evidence is pretty light. As at other key junctures in their argument, they depart from their statistics-wielding men-in-white-coats stance and simply present bold assertions as true. (My favorite of these is "Intermarriage among people in the top few percentiles of intelligence may be increasing far more rapidly than suspected" -- no footnote.) About the business elite they say, "Both common sense and circumstantial evidence suggest that people who rise to the upper echelons of large businesses tend to have high IQs and that this tendency has increased during the course of the century." To back this up they cite a 1976 article in Fortune saying that 40 percent of the chief executive officers of the top 500 corporations had a background "in finance or law, fields of study that are highly screened for intelligence."
A much more recent and comprehensive source is Forbes's annual survey of the chief executives of the 800 largest corporations, which lists their higher-education credentials. By now most of these executives are people who went to college in the 1950s and 1960s -- that is, during what Herrnstein and Murray identify as the beginning of the meritocratic era. The Forbes survey supports The Bell Curve to some extent but presents a much more nuanced picture. The college that provided the most CEOs is cognitively screened Harvard, with twenty-three, and Cornell and Princeton are second, with eighteen each. It must be said, however, that only a couple of the CEOs from these schools have ethnic-sounding names, and some of them leap out as belonging to the boss's-son rather than the cognitive-elite category, such as John Hess, of Amerada Hess, Harvard '75. Schools with much less cognitive screening are also important suppliers of CEOs, though: the University of Wisconsin (thirteen CEOs, 69 percent of applicants accepted), Purdue (ten CEOs, 90 percent), the University of Michigan (ten CEOs, 69 percent), and Vanderbilt (eight CEOs, 58 percent).
Herrnstein and Murray say that because college graduates on average have above-average IQs (they cite as proof an estimate from twenty-four years ago), because most business executives have college degrees, and because they have risen through the ranks in their companies, "in the neighborhood of 70 to 80 percent" of those in the ranks of management might have IQs of over 120. But of the 2,000 four-year colleges and universities in the United States, only sixty-six do not accept a majority of their applicants. The cognitive screening for a bachelor's degree is so much less severe than the financial screening for one that the proposition that college graduates are likely to come from comfortable backgrounds is much easier to support than the proposition that they are likely to be highly intelligent.
Sketchy estimates of the dimension of the new elite usually put it at about 20 percent of the population -- the "fortunate fifth," Reich has called it. It's impossible to get to any number this big -- about 50 million people -- without going far outside the bounds of the elite as it is usually described. Just a cursory glance at the statistics demonstrates that the top fifth of the income distribution is made up mostly of people who didn't attend highly selective universities and aren't politically liberal.
Every year the Internal Revenue Service publishes a booklet called Individual Income Tax Returns; the most recent one available covers 1992. The number of tax returns in 1992 that reported income of $100,000 or more was 3.8 million. It's hard to think of people who make less than $100,000 as members of "the elite," but returns reporting $100,000 and above represent only three percent of returns. To get an elite of more than 10 percent of tax filers, you have to drop down to an income level of $60,000 a year. (It is true that you get a much higher income for the "fortunate fifth" if you look at households rather than individual tax filers -- according to the Census Bureau, the mean income of the top fifth of households in 1992 was $91,110, no doubt because many of these households have two workers. Even this number, though, wouldn't put most top-fifth families into the familiar plummy picture of private school, gated residential community, and vacation home on Martha's Vineyard.) There are more than 23 million Americans with bachelor's degrees, and 12.7 million Americans between the ages of twenty-five and sixty-four, according to Herrnstein and Murray, are in the top IQ decile -- so the overwhelming majority of both groups must not belong to the over-$100,000 economic elite. In other words, there are a lot of highly educated and intelligent people who aren't especially affluent.
Also, as income rises above $100,000, the percentage of it that is derived from "salaries and wages" and "business or profession" declines, and the percentage derived from long-term capital gains steadily increases -- which hints at an upper end of the economic elite made up of inheritors, big-time entrepreneurs, and financiers, not high-IQ professionals. In other words, perhaps there are also still a lot of rich people who don't have high IQs and prestigious degrees. Finally, a consistent finding of polls is that the more money you make, the more likely you are to vote Republican. So there are a lot of elite people who aren't liberal either. The one public-opinion finding that supports the idea of a liberal elite is that more people with graduate degrees vote Democratic than do people with only bachelor's degrees.
Even if what was considered the meritocratic elite was trimmed back from a fifth to, say, a hundredth of the population, there would still be a big problem with the cognitive-elite theory, arising from the sociology of its members. Let's say that Herrnstein and Murray are right that the most selective colleges today efficiently "soak up" a majority of the American adolescents with the highest IQs, and that these students are plainly the most able members of their age cohort. Ever since the meritocratic apparatus in the United States became mature, around 1970, the undergraduate atmosphere in Ivy League colleges has been one of intense anxiety about pursuing careers outside those professions that follow from a law or medicine degree or an MBA (which for the best students generally leads to management consulting or investment banking and not, as Herrnstein and Murray imply, to corporate management). Because these fields are tightly screened on the basis of test scores and grades, they present the lowest-risk career options to people who have very high test scores and grades. One enters them as part of an extremely limited pool of people, most of whom are virtually guaranteed high incomes.
In fields that aren't so tightly screened -- from entrepreneurship to show business to corporate management -- the odds are much longer for members of the cognitive elite than they are in the professions. The stampede into the professions, driven by risk aversion, is the subject of many hand-wringing commencement speeches (and, years later, of the students' middle-aged longueurs). It would even be possible to take the perverse position that people with high IQs have become less powerful in recent decades, because they are so firmly channeled into advisory roles in the professions. This is why the Ross Perots and David Geffens of the world don't find themselves competing with many anointed members of the cognitive elite during their rise into the economic elite.
In his In an Age of Experts, Steven Brint sets out to create a more precise taxonomy of the professional classes than is available in the books that sweepingly assert the presence of a unitary new elite. Although Brint attempts to prove his points with somewhat woolly efforts at quantifying, such as enlisting a squadron of research assistants to code articles from sixteen publications numerically on each of a hundred variables, most of what he says has the ring of common sense. His main theory is that professionals are differentiated in ways that mirror wider divisions in the society. Professionals in "business services" (corporate lawyers, for example) and "applied science" (engineers) tend to be well-off and conservative. Professionals in "human services" (social workers, schoolteachers) are more liberal but most of them can't plausibly be presented as part of an economic elite. Professionals in "consultative" jobs are more liberal than those in "command-oriented" jobs. To the extent that there is a liberal strain among professionals in general, it is found in their positions on social rather than economic issues. To the extent that there is a New Class, it is found not here but in Western Europe, where there is a tradition of elite university graduates becoming career government officials.
Illustration by Maris Bishofs