It has been an enormous change, more profound than the mere circulation of elites that often follows political revolutions. Since 1945, elite institutions in the United States that were once almost entirely the preserve of white high-church Protestant men have opened their doors--first to Jews and Catholics in the early postwar years and then, in a second opening, to blacks, Asians, and Hispanics, and to women, in the 1960s and 1970s. During the same period, higher education has also become far more important as a way to the top of American society, and even to the middle class. Scarcely anyone doubts, at least in public, that these changes have been for the better. They have made America a more just society and contributed to the nation's prosperity, power, and cultural vitality.
But is this transformation incomplete because the criteria for admissions and advancement are imperfect or biased or corrupt? Or, worse yet, is it a fraud? Has one elite favored by birth merely given way to another, supposedly based on merit, but no more deserving and just as self-interested? Criticism of the meritocracy dates from the 1960s, when the new radicalism identified higher education as simply the latest "mask for privilege." (The phrase "meritocracy" came from Michael Young, a British sociologist and social democrat, who worried that the new educated ruling elite would have so powerful a basis of legitimacy as to make inequality even harder to reduce.)
The opening up of the universities to blacks and Hispanics also brought out a previously hidden tension. The first postwar opening of the elite institutions to the children and the grandchildren of European immigrants coincided with the introduction of standardized educational testing. Although selective colleges continued to pay attention to other characteristics of students, performance on the Scholastic Aptitude Test (SAT) began to count heavily in admissions. At that time, testing worked in favor of the new groups, as it did later for Asians; but blacks and Hispanics didn't do as well. For blacks and Hispanics, therefore, the tests amounted to an obstacle to admission, and accordingly the tests came under fire for alleged bias. And many people besides advocates of affirmative action have also worried about the limits of the SAT (or of any multiple-choice test) as a measure of ability, much less of merit more broadly conceived. If high-status positions depend on admissions to selective colleges, and admissions in turn depend on tests that are, in some measure, arbitrary, is the system fair and reasonable?
In a society with a strong populist streak, it would be asking too much to expect any system of elite selection to be universally approved. The very idea of an elite is suspect, and there is no agreed-upon ranking of different abilities. If the Congress had ever expressly debated whether America ought to have a national system of testing to determine which people would end up occupying much of society's upper tier, the system could probably never have been enacted. Still, as a result primarily of private decisions, we do have such a system, and for the past half century it has occupied a central place in the experience of nearly everyone growing up in America with aspirations not just to high status, but also to the good life.
Nicholas Lemann's book tells how this system came about. The book is a high achievement. It provides a rich account of the people and the institutions involved in the rise of educational testing and the turn toward meritocracy in major universities, as well as the later struggle over affirmative action. One value of historical inquiry is that it can lay bare long-forgotten choices that created the seemingly settled world that we have today. The standardized testing system confronts every student and teacher as a fact of life, yet as facts of life go, it is a recent one: just five decades old. Who decided things should be this way? What were they hoping to achieve? Why did this particular system prevail over others? Since most of the critical decisions about testing and admissions were not governmental, they were not matters of public record. Before Lemann's book, the full story was little known, even by people who work on these issues.
Lemann has clearly put years of research into his book, no less than would be expected of a full-scale academic study, though his book is more entertaining. And while he is highly critical of testing, he has enjoyed the cooperation of key institutions, notably the Educational Testing Service (ETS), the maker of the SAT. Like any good historian, he has trolled their archives for pertinent and lively material; but he has also interviewed many of the people who figure prominently, or even only incidentally, in the events that he is relating. The Big Test is a pleasure to read not least owing to Lemann's ability to tell the life stories of these people so well while developing his larger narrative. But the book is a narrative rather than an analysis, and so it leaves some important questions unanswered. In the last section of the book, when he turns to the battle over affirmative action in California, Lemann gets too caught up in side stories, and he fails to bring the narrative threads to a satisfying conclusion. Still, this is a major work of contemporary history that ought to be read by anyone interested in understanding the development of American society during the past half-century.
As Lemann describes it, the rise of testing and the meritocracy in America began with what might be called a status revolution from above. The key instigators came from the heart of the Protestant establishment, though they were in some ways detached from it. Max Weber could not have dreamed up a more perfect figure to be the historical agent for rationalizing the choice of the elect than Henry Chauncey, who became the first president of ETS in 1945. (In his mid-nineties today--"a miracle of geriatric robustness," as Lemann calls him--Chauncey served as a direct source for the book.)
The Chaunceys trace their lineage back to a Puritan minister who emigrated to Massachusetts in 1637 out of dissatisfaction with the lax rules of the Church of England, only to find that the Puritans in the New World were also, in his eyes, too lax. Another member of the family, a leading Boston minister in the mid-1700s, believed in the need for a disciplined, trained elite. It was as if Henry Chauncey distilled the faith of his ancestors into a modern scientific form when early in his career he became an enthusiast of mental testing as a means of improving the organization of society. Although Chauncey's immediate family was not wealthy (his father was also a minister), he went to Groton and Harvard, where he became an assistant dean and a protege of Harvard's president, James Bryant Conant.
It was Conant, a chemist by training, who publicly articulated the grand vision for reforming higher education in America, not by enlarging it (he opposed the GI Bill) but by admitting to selective colleges the most intelligent students and preparing them for positions of public service and influence. This group would be the modern fulfillment of an ideal that Conant derived from Jefferson of a "natural aristocracy" of virtue and talent, yet they would somehow have no social standing above anyone else. For Lemann, Conant's ideas are the founding illusion of the meritocracy, and they are certainly a fat target for criticism.
But how change happened, or at least got started, may not explain why it happened. Conant's vision does not explain why all the private universities, some sooner, some later, began recruiting high achievers outside the old WASP elite, or why law firms, banks, and many other organizations went through similar changes. Larger forces were at work that Lemann's narrative does not always disclose. The open, competitive structure of American society would have imposed substantial, long-run costs on any institution that kept its doors shut. A university that limited its faculty and students in Cold War America would have lost grant money, prestige, and ultimately endowment; the opening up of the old elite institutions is what saved them from decline.
But why did ETS and the SAT triumph over other organizations and other tests? In some industries such as software, early advantages enable one organization to gain market share, spread fixed costs over a large customer base, and lock in dominance regardless of whether the company has the best product, much less the ideal one. This appears to have been the pattern in educational testing. ETS benefited from early sponsorship and cost advantages that it used to lock in a monopoly position. As Lemann tells the story, Conant and other university presidents, as well as foundation executives, intervened in the late 1940s to give ETS the central role in test-making, while the College Board became the owner of the tests. In 1951, at a difficult point in its history, ETS won a contract to administer a test for the Selective Service that proved to be an enormous financial bonanza.
From the start, low cost was a critical advantage for the SAT over the College Board's earlier examination, which was more expensive to score. Moreover, the SAT didn't threaten high schools, because it purported only to measure an innate characteristic, scholastic aptitude; ostensibly, aggregate test results could not be used to assess school performance. The huge expansion of college enrollment in the postwar era created a growing institutional demand for a screening mechanism of some kind; and the SAT was a low-cost solution to this problem that did not challenge any of the institutions whose cooperation it required.
Early in the history of testing, critics raised objections to the tests as a measure of merit. As Lemann shows in one of the most revealing aspects of the history, Chauncey himself long tried to measure other abilities through tests of personality, but none of those tests could command the same consensus support as the SAT. The weight given the SAT partly reflects the difficulty and the high cost of obtaining more complete assessments. The elite private colleges, which enjoy not only greater resources but also more discretion, do give more weight to subjective judgments of students' potential (as well as less legitimate factors, such as their family's wealth). It has been the big state universities, with their more formalized, quantitative criteria for admissions, that have relied most on SAT scores as well as grades. And partly because the link between testing and the meritocracy is actually greater in the public sector of education than in the private sector, the controversy over the use of test scores in minority admissions has been more acute and explicit there.
As Lemann's narrative proceeds, therefore, the scene of the story moves from the elite private colleges to the state system that by the 1960s had become the very model of what a great modern university ought to be. This was the University of California; and its chancellor, Clark Kerr, becomes the third major figure, after Chauncey and Conant, in Lemann's tale of the making of the new social hierarchy. A labor relations expert, Kerr formulated the Master Plan for higher education in California. Unlike Conant's vision, Kerr's plan accommodated the demand for vastly expanded popular access to college but still reserved a more privileged tier of campuses for the high-achieving elite.
In the nuanced and sympathetic portrait that Lemann draws, Kerr emerges as a tragic figure. Just at the moment in the 1960s when his ideas about higher education were triumphant, he fell victim to the conflict between the emerging radical movement at Berkeley and conservative backlash in the state. One of Ronald Reagan's decisions after taking office as governor was to remove Kerr. While Kerr fell, however, the design that he established for higher education endured; life narratives do not always mirror the history of institutions.
This lack of fit between individual biographies and the larger history becomes a serious problem in the last portion of Lemann's book, as he takes up the battle over affirmative action in California, culminating in the passage in 1996 of Proposition 209, which banned the use of race in university admissions. Once again, he tells the stories of people whose lives cross in the struggle; but while their stories are interesting, they do not clearly advance the overall narrative. Worse yet, it is not clear what Lemann expects the battle over 209 to show. When the book ends with the apparent defeat of affirmative action, he doesn't draw the conclusion that the majority of the public in California had affirmed the legitimacy of the SAT and the meritocracy, though that might seem to be the obvious inference. The book does not really have an ending; it just comes to a stop.
And, of course, the story does go on. The subsequent decisions in Texas and California to link college admissions to high-school class rank, effectively undoing race-blind mandates, suggest that there is actually more political support for minority admissions than the California vote indicated. Recently, a controversy was touched off by news reports--misleading news reports, as it turned out--that ETS was going to introduce a Strivers' Score, adjusting SAT results for students from disadvantaged backgrounds. As Lemann shows, this is an idea that has been stirring at ETS for a while, generating strong internal opposition. ETS says it has not made a decision about the proposal, which would not adjust SAT scores but rather provide separate data about test-takers' backgrounds. This would scarcely mean the end of meritocracy; it would be, at most, a minor aspect of a system that still overwhelmingly affirms the importance of the SAT and kindred tests.
Is the SAT biased? Not in the sense of asking questions that give an unfair advantage to whites. But it is true, as critics point out, that test-prep courses and other coaching can raise SAT scores, and that families with low incomes are less able to afford such help. In that sense, there is indeed a socio-economic and racial bias in the results. But it is not at all clear that some other test would eliminate this problem. Moreover, the far greater deficit arises long before test-prep courses, early in a child's education and development. If the gap in scores between blacks and whites is to be remedied, this is surely where the principal effort should go.
In an afterword, Lemann calls for a more democratic educational policy, focused on improving the "bad bottom tier of public schools" and sending "most people all the way through college." Those are both sensible aims. But to get more people through college, Lemann suggests that we need "greater national authority over education," a "nationally agreed-upon curriculum," and tests for college admission on mastery of that curriculum. If, by "national authority," Lemann means federal control, this is certainly not going to happen any time soon, as the recent experience with curriculum standards suggests. There is scarcely a pedagogical issue that is not an ideological issue in America today. We do not enjoy the kind of consensus that a single national curriculum would require. Moreover, localism is deeply embedded in the history of American education; and with suspicion of the federal government still running high, this seems unlikely to be the time for centralizing decisions about what is learned in American schools.
It is conceivable, however, that a national curriculum might evolve through the kind of process that originally produced the SAT. High school students already take curriculum-based tests, such as the ACT, the SAT II tests, and tests for advance placement (AP) courses. Colleges might demand more tests of that kind instead of the SAT. But would that matter? Empirical research shows that reliance on curriculum-based tests would not select a substantially different group of students.
According to Wayne Camara, the executive director of research for the College Board, the correlation between the ACT and SAT is .88; the correlation between AP and PSAT scores ranges (depending on the AP subject) between .6 and .8, which indicates a strong relationship. In other words, much the same students do well on both tests. Relying on curriculum-based tests would not significantly change either the ranking of individual students or the ethnic composition of high and low scorers.
Still, greater reliance on curriculum-based tests could affect the incentives of teachers, students, and schools. On the positive side, a national curriculum test might help upgrade the performance of the worst schools, but it would also discourage curriculum innovation and diversity. American education is already in a testing frenzy, as business-minded leaders try to achieve greater teacher accountability and school accountability by tracking measurable results. Lemann's proposal risks going so far in this direction that the facets of education least amenable to testing, such as the arts, would suffer even more than they do now, and the country's entire educational system would be tuned to a single instrument.
It is ironic that Lemann should end his book by calling for a new testing system that would be more restrictive than the current one. In his afterword, he says that we should not have an anointed meritocracy at all, but it is hard to see why his alternative is less likely to produce one. His proposals do not grow out of his earlier historical analysis, and Lemann does not spell out who or what might actually bring about the change he wants. In his ideal of a "true meritocracy," the biggest rewards "would devolve to people only temporarily, and strictly on the basis of their performances ... The elite would be a group with a constantly shifting, rather than stable and permanent, membership." Actually, with constant shake-ups in the corporate world, things have already been moving in the direction of greater job insecurity, which should make Lemann happy. The biggest rewards today do seem to be flowing not to the meritocracy but to the entrepreneurs and the celebrities whom Lemann dubs "Talents," and membership in that world is constantly shifting. Even the elite graduates working in law firms and corporations whom Lemann calls "Mandarins" continually need to prove their worth. Most of them are scarcely coasting through life on their SAT scores. In fact, Lemann's term "Mandarins" does not capture the pressures and the uncertainties of contemporary professional lives (except, I am happy to say, for tenured university professors).
Fortunately, the value of The Big Test does not rest on Lemann's ideas about alternatives. The book's lesson may be the opposite of what its author intended: that the basic idea of social mobility through performance in education has become thoroughly established in America. Much of the leadership class in America will find other routes to the top, particularly through success in the marketplace, but in a society in which advanced skills and education count as heavily as they do in our society, the meritocracy is not going away. The house that Chauncey, Conant, and Kerr built looks like it is going to stand for a long time.
PAUL STARR is professor of sociology at Princeton University and co-editor of The American Prospect.