By Professor Lani Guinier[i]








At the dawn of a new millennium, America’s quest to overcome the effects of centuries of racial inequality is on a collision course with its even more pervasive preoccupation with technology of all kinds, particularly quantifiable measurements of that all-American social good, “success.” Nicholas Lemann perceptively describes this impending train wreck in The Big Test: The Secret History of the American Meritocracy, when he writes, “in technology- and organization-worshipping” America, “two conflicting sets of numbers”[ii] are being generated that spell mutually inconsistent solutions for distributing high-stakes educational opportunity.

The first of those set of numbers is “everyone’s scores on standardized tests.”  Those who believe in education by the first set of numbers allocate places in colleges or universities based on scores on mental aptitude tests. Deciding who “deserves” to benefit from admission to selective colleges and universities occurs within a “testocracy” that sorts, evaluates and ranks measurable mental aptitude.[iii]  The resulting test scores are presumed to tell us all that what we can know in both real and relative terms about each applicant’s potential, which is then deemed the most important evidence of his or her “visible, rankable” “merit.”[iv]

The second set of numbers is “the share of good jobs and educational billets” held by African Americans and white women. Those committed to this numerical set believe just as fervently—in the name of fairness—in discovering and promoting untapped, and perhaps immeasurable, human potential when it lives in dark-skinned or female bodies.  In recent years claims of merit have trumped African Americans’ claims of fairness. Those dedicated to testing (the testocrats) have convinced many to share their faith in the importance of measuring mental aptitude—in advance and out of context—in part because they have successfully used the term “merit” to refer to their criteria. 

Yet, whether it is called a testocracy or the new meritocracy, certain principles of fairness are called upon to defend it. For example, some claim a social justice nirvana will arrive when we have firm knowledge about what all Americans know and are capable of learning. They believe that to measure in advance someone’s capacity to learn is to know exactly who, what and where to teach.[v] Reliance on standardized aptitude scores gives notice of what is expected and helps people prepare diligently and appropriately for the testing to come. Concerns about fairness and equal opportunity are presumed satisfied by awarding access to higher education in an open competition to those with the highest test scores (and hence the greatest “merit”), who can then gain access to jobs and influential positions in the upper reaches of society.

Related to principles of fairness is the claim of “desert.” Because aptitude testing opens opportunity not only to learn, but also ultimately to earn, one’s testable capacity to learn becomes the yardstick of his or her “general worth.”[vi] This faith in aptitude testing produces the confident assertion that those who are capable of knowing the most will rise to become part of a legitimate intellectual elite or mental theocracy.

The second set of numbers—often associated with a defense of affirmative action—seeks to redistribute opportunity to those who have traditionally been denied it. It assumes that the traditional “test-centered” approaches to merit fail to detect the actual potential of those who, because of the legacy of slavery and segregation, are ill prepared to excel on aptitude tests. In fact, it implicitly questions whether aptitude tests identify the merits of all who do excel; but this critique is often left unstated.  Whether the “winners” deserve to succeed is not the issue; the question is whether the particular losers of color deserve to lose.

Supporters of affirmative action often articulate this concern with fairness and compensatory justice in terms of the need to assure diversity among those admitted to elite universities. Proponents of affirmative action generally acknowledge that criteria to assure a diverse class may be at odds with the criteria testocrats think necessary to assure a maximally “qualified” class. They accept the general validity of the claim that standardized admission indices combining grades and aptitude tests generally predict a student’s capacity to learn and to do; affirmative action advocates concede the value of testing for everyone, other than people of color. Where these advocates take issue with the equation of merit and the testing hierarchy is their assertion that principles of fairness and diversity require compensating for the less impressive tests scores among minority applicants. For underrepresented people of color, defenders of affirmative action seek to supplement the testocracy with other criteria such as leadership ability, community service, and achievements that indicate ambition and determination.

            Unfortunately, affirmative action supporters have failed to go beyond the experience of people of color to challenge whether poor and working class whites, who also often lose out in the testocratic competition, deserve their lot based on their poor (relative to affluent whites) showing on high-stakes mental aptitude tests.[vii] They’ve expanded the idea of who “deserves” opportunity, but have not challenged the creation of such a category in the first place.  Nor have they explored the fundamental defects in the testocratic selection system, even for those who do well on entry-level tests.

            The inconsistencies between the positions of those supporting a colorblind testing hierarchy and supporters of a more racially diverse elite may appear irreconcilable. Yet, both sides share several things in common.  Trapped in the debate over individual “merit” and desert, both fail to connect admissions criteria for institutions of higher learning to the stated missions of those institutions.  This is because each allows the issue of the institutional mission to remain, for the most part, hidden or implicit.  Neither forces us to evaluate admissions protocols as a functional or integrated part of the whole institution.

If we are to move beyond the present polarization in a manner consistent with the commitments to fairness and equality that both positions endorse, we must more carefully explore how to measure and what to call merit, and what constitutes fairness for all, in a multiracial democracy. A first step is to view “merit” as a functional rather than generic concept, while keeping firmly in mind the democratic purposes of higher education and the specific mission of most institutions of higher education.

In other words, we should seek to reconfirm the democratic role of higher education in a multiracial society by re-connecting admissions processes to the public mission of both public and private schools.  In doing so, we confirm the benefits of affirmative action—but not simply to people of color—by re-casting merit as a practical term that is intimately connected with each institution’s specific mission.  That focus, in turn, allows us to reconsider the relationship between individual merit and operational fairness, between claims of individual desert based on past opportunities and individual contributions based on future societal needs.

I tentatively call this a process of confirmative action,[viii] because it takes lessons from both the testocracy as well as affirmative action to confirm a set of experimental and pragmatic actions that begin to link (ad)mission practices for all students to the broad mission and public character of higher education in a multiracial democracy.[ix]



             Both affirmative action supporters and the defenders of the testocracy share a commitment to democratic ideals, including a stated desire to promote competence and provide opportunity in a fair and equitable fashion.  These commitments also inform the stated mission of most public and private institutions of higher learning: a commitment to educating and serving the larger society.

1. The Mission and Public Character of Higher Education

Because it involves rationing access to power and influence, “the selection of young people to enter prestigious educational institutions amounts to a visible, high-stakes exercise in civic pedagogy,” says Professor Glenn Loury, “These ‘selection rituals’ are political acts, with moral overtones.”[x]

For Loury, as for many others, this sacred role stems from the obligations society vests in higher education to prepare young people for the responsibilities of citizenship (including employment),[xi] the challenges of public participation, and the chance for leadership. Mindy Kornhaber’s study of the mission statements of public and private institutions of higher education showed that the most important component of these institutions’ declarations is “service.”[xii]

This service commitment, training future leaders and citizens, is preeminent. She writes, “To train leaders, institutions educate people who will occupy positions of power and responsibility, people who will participate in their chosen discipline in ways that have lasting impact, people who will contribute to civic life and people who are committed to public service.”[xiii] In contrast to medieval institutions of Europe and the colonies, which trained future clerics and oligarchs, American institutions of higher education were established to develop leaders in a participatory democracy.[xiv]

Of course, this service commitment is not higher education’s only mission; preparing people for the workplace and, particularly among elite institutions, producing scholars who will advance knowledge also have high priority. Kornhaber notes, “American public colleges and universities were specifically constituted to serve the nation’s democracy and its economy.  They have been established to provide the society with knowledge and technical expertise.  Equally as important, they are intended to develop leaders and to hone the critical perspectives needed by citizens in a participatory democracy.”[xv]

It was this desire to identify competent students from public schools who would pursue public service that proponents of the testocracy used to spur the development of aptitude tests in the 1940s and 1950s.[xvi] The tests would presumably enable admissions officers at both public and private institutions to select good students from public schools by providing a uniform baseline of quality. For schools like Harvard and Yale, the tests would permit them to become national as opposed to regional institutions, recruiting students from the Midwest and the South who had not been groomed for admission to the leadership elite through the familiar private boarding school track. Those pushing for aptitude testing assumed that these students would not only work harder and do better, but would also, true to their middle class and public school origins, then become a public-spirited leadership cadre.

These democratic commitments undergird our present state subsidization of both public and private institutions of higher learning to affirm their sacred, civic nature.  These democratic aspirations also unify both the testocrats and supporters of affirmative action.

2. Both the testocracy and its progeny, affirmative action, profess democratic values

At first glance, the institutional practices of testocracy and affirmative action appear completely incompatible.[xvii] The test-centered approach argues for equal opportunity for all as measured by aptitude testing (with universities functioning as a union hall[xviii] or sorting mechanism to allocate jobs). It posits merit as measurable mental aptitude.

Affirmative action, on the other hand, contends that merit is not fixed or measurable by a simple test. Therefore, because the tests do not identify all who can succeed, testable merit should not be the pre-eminent value. Instead, society should provide opportunity to those who have been left out but who will perform competently if given the chance.

Both the testocracy and affirmative action were designed to promote a more democratic distribution of opportunity—continuing the democratic impulse behind adopting aptitude tests in the 1960s and 1970s that especially benefited Jews and recent immigrants previously excluded by the old boys network.[xix]

But it turns out that, over time, the testocracy has repeated a chief characteristic of the old system: it rewards those already privileged. In the old boys network, access to higher education and good jobs was passed from one generation to another through exposure to boarding schools and other institutions in which one’s merits (that is, proof of belonging) were observed and finely honed.

In the testocracy, or what some call “the new meritocracy,” privilege is passed on through a new kind of club: a testing system that allows those with resources to show that they too belong. They belong because they are able to learn the rules of the test through explicit coaching, private school, or an upper class, resource-rich suburban education. They then successfully play the testing game to their own advantage. Studies show that within each race and ethnic group, aptitude test scores rise with parental income.[xx]

The correlation between aptitude test scores and parental income should not surprise us, given the role that high-priced coaching techniques play in raising test scores.  But what may surprise some is just how weak the relationship is between high test scores and what the tests claim to predict (i.e., first year college or law school grades). Studies suggest that nationwide the aptitude test for law schools (LSAT) is between 9% and 14% better than random in predicting first year grades.[xxi]  In other words, what the testocracy promotes in the name of “merit” turns out to be based on privilege. What these studies find is that the testocracy tells us more about a student’s past than his or her future.

It soon became apparent that the test-centered approach to distributing opportunity was not meeting its democratic responsibilities. The expansion of opportunity to African Americans and Latino Americans was substantially incomplete.[xxii] Thus, techniques based on affirmative action were adopted to compensate for the deficiencies in the testocracy. Indeed, affirmative action became necessary to balance the serious flaws in the testocracy’s claims of democratic opportunity.  These flaws extended beyond people of color but remained mostly invisible because of the politics of race in the U.S.[xxiii]

In other words, a temporary peaceful coexistence between those included in the new meritocracy and those excluded by it was crafted when the rich testocracy was joined at the hip to its poor cousin, affirmative action.  Rather than threatening the testocracy or exposing its failure to meet its democratic promise, affirmative action camouflaged deficiencies in its test-centered assumptions. As long as it was supplemented by policies of affirmative action, the testocracy remained temporarily ensconced in its claims of equal opportunity.

In some ways contrary to Lemann’s vivid prediction of a huge collision, the testing system and affirmative action have actually been mutually interdependent. The argument is as follows: the flaws of the testocracy compelled affirmative action into existence and without it those limitations would have become more visible sooner. The faith in the capacity of scientific tests to measure and predict ability to succeed depended upon, for its legitimacy, the ability to predict success across the racial and class divide. Affirmative action helped disguise the elitist nature of the testocracy.

While advocates of testing and of affirmative action may argue about how to identify merit and realize equal opportunity, both seek legitimacy in a more democratic distribution of opportunity to those who are competent to use that opportunity wisely or well.  Unfortunately, what also unifies them is their failure to acknowledge the ways that the current selection system is not fulfilling its larger democratic mission. While the old boys network of yesteryear was an elite based on inherited privilege, and justified its privilege by acts of public service (noblesse oblige), this new elite claims to earn its place through the opportunity to demonstrate inherited aptitude, in a competition open to everyone.

The risk, however, is those individuals who demonstrate such measurable aptitude will feel they have earned the right to enjoy the fruits of their “merit,” free of any obligation to serve others or to give back to society.  These new meritocrats may presume that society will automatically benefit when those with high test scores have incentive to keep climbing upward—for whatever purposes they choose.  If they feel inclined to engage in public service, so much the better. But the danger comes when public service or contributing to society becomes a preference, not an expectation.

The testing advocates in particular have lost sight of the original connection between merit as competence and the expectation that competence will be deployed in significant ways to serve society. The issue of leadership and civic-mindedness seems to have dropped out of the merit formula. Instead, according to their calculations, the question of what we should be looking for has simply collapsed into performance on an aptitude test.


        In light of the democratic mission of higher education, three democratic values are at risk in the institutional practices of the testocracy, even when supplemented by affirmative action.  They are: inclusiveness, “transparency” (the condition that the procedures used to admit students are apparent and visible), and institutional accountability (linking the stated public mission of both private and public institutions with their practices). I will address each value in turn.

1. Inclusiveness

Inclusiveness not only means that no individual or group is arbitrarily shut out or excluded, but also that opportunity is distributed broadly (casting a wide net across many populations in society).  The testocracy asserts that because all citizens enjoy the same initial opportunity to prove their merit on objective tests, the resulting sorting and ranking will be democratic.  Even if this were true—and research suggests that it is probably not[xxiv]—the new meritocracy, while certainly more inclusive than the old boys network, still excludes many quite capable people.

It is not fair to rely on test scores to fulfill equal opportunity goals when some people are better prepared to do what is expected because of resources they have inherited from their parents and in most cases their grandparents.[xxv] Given the intergenerational nature of this wealth transfer, and its relationship to performance on standardized tests, those who are born to middle or upper middle class white parents are more likely to be included and those with black or poor white parents to be excluded for several generations.[xxvi]

That we have only noticed this relationship now reveals how profoundly we have permitted congealed privilege to continue to exclude many, even in the name of objective deserts. Thus, the testocracy is doing two things inconsistent with an equal opportunity mandate. It is “credentializing” those who are already quite privileged; and it is (mis)leading us into accepting the educational inequality in the K-12 system as “normal.” Confusing inherited privilege with inherited aptitude, it makes an inherited resource a pre-condition for opportunity—leaving out many people of color and poor and working class whites as well—based on the unfair accumulation of resources that some others bring to the test.[xxvii]

Furthermore, this exclusion from elite institutions is a continued exclusion from the paths to positions of leadership and influence within the society.[xxviii] “To understand this,” Mindy Kornhaber suggests that we “consider the features of institutions marked by higher SAT scores.   These institutions are more likely to provide better financial aid, more intellectual challenges for students, higher expectations, more mentoring, and more promise of reward for graduation.” She further argues that these characteristics are vital to surmounting many of the hurdles that students confront, saying, This holds true from elementary school through doctoral work: When students are challenged, high expectations are held for them, and there is a supportive social network, they are likely to achieve more.”[xxix]

Any system of sorting and ranking will develop “losers.” But a true democracy cannot countenance the creation of a class of permanent losers and permanent winners and the acceptance of the pre-democratic idea that only some people will rule all of the time. 


2. Transparency

The second value is transparency (what some people might call notice).  The testocracy claims to identify individuals who—based on abstract claims of fairness or excellence—deserve a chance to succeed.  It purports to do so in an “objective” system of sorting and ranking.  By giving people notice of what is expected, the testocracy, purports to give everyone an equal chance to get prepared to succeed.

But the assertion of fairness based on transparency here is false. To rely exclusively on “notice” to meet equal opportunity values is like supporting a voting system that announces elections will be held on Tuesdays, but only during the hours of noon to 5:00 p.m., and only at polling places accessible by private transportation.  Those who must work during those hours and those dependent on public transportation would have notice of what was expected, but simply could not meet the condition.[xxx]

Furthermore, the testocracy does not operate in a transparent way because it relies on asymmetric notice of the many exceptions from the announced set of expectations.  The testocracy has always made exceptions to the rule that students are admitted solely based on high scores.  But virtually none of the other exceptions are given the visibility that surrounds affirmative action.  The problem here is not just the existence of exceptions from the rule. The problem is the extent to which those exceptions are “announced” or made plain to the public.

For example, many universities acknowledge they reserve places for the children of alumni, or in the recent case of the University of Virginia, track applicants whose parents may be potential donors.  But colleges do not make this exception to the rule as visible as their affirmative action policies.[xxxi]

Students of color who are beneficiaries of affirmative action are highly “visible” throughout their college years, both inside and outside the classroom.  But alumni offspring, who are also considered outside the system of strict ranking and sorting, easily become anonymous. No one studies their performance as a group, or monitors their seating partners at lunch.[xxxii]

  Imagine what might happen if that were to be done.  An institution could publish the names of those whose parents also attended the institution, or circulate in some form the identity of those whose parents may be potential donors. It could require the “legacy” applicants admitted to wear a decorative ornament or a badge signifying their special status. One might argue that this requirement would stigmatize or unfairly target these students in ways that might interfere with their learning or their ability to feel welcome at the campus.  In response we might argue that everyone at the institution should then practice a form of “blindness” to the badge, refusing to draw any conclusions about the badge wearer or even literally to “see” the badge in order to allow the badge holders to avoid being stigmatized.[xxxiii] (You can almost hear the roommate comment, “I don’t think of you as a legacy admit; you’re not like the others.”)

            This idea seems preposterous. Yet, that is exactly what many beneficiaries of affirmative action experience. Their color is the badge they wear throughout their college years. The sense of being tagged may wear off as they get to know their well-intentioned, conscientiously color-blind classmates. But it always exists, likely to reappear when they first enter a classroom or when they ask questions of a professor who does not know them.

Of course, the point I am making is the opposite of suggesting that a democratic system of admissions stigmatize any of its choices.  A democratic system should be transparent in ways that avoid stigmatizing or announcing to the world that some students are “less equal” than others.  But here the current system fails.  Instead, it reflects the inexact science of determining who can benefit from the opportunity of an elite higher education—and it hides the built-in privilege that is connected to that judgment, calling it instead the “merit” of some admittees.


3. Institutional accountability

For an institution to be accountable, it needs to link its stated mission to its practices.  A recent study of the University of Michigan Law School offers a dramatic opportunity to assess the new meritocracy’s democratic vision as modified by the practice of affirmative action.  A study by Lempert, Chambers and Adams traces the post-graduate performance of minority and white alumni of the law school beginning with the Class of 1970, the first class with more than 10 minority graduates.[xxxiv] Thus it traces the careers of three generations of white students as well as students of color admitted to the school.

The Michigan study suggests that the new meritocracy has not only fallen short on its claim of elevating the most competent; it has also neglected to institutionalize its own democratic values of public service and rotating leadership.


A) Elevating the most competent

At Michigan, competence in lawyers is measured by their long-term career satisfaction, ability to earn a living, and community contribution.  While a modest correlation exists between entry-level criteria and institutional competence in the first year of law school, no apparent significant relationship exists between those criteria and what lawyers do when they graduate. The study found that for all graduates, the entry-level admissions index fails to correlate with post-law school accomplishments, including level of lawyers’ income and career satisfaction.  They found no relationship between admissions indices and income as an attorney.

The Michigan study reveals that traditional, “hard,” test-score based admissions processes are no better predictors of success after law school (whether success is measured by earned income, career satisfaction or service contributions) than are “soft,” “whole person” selection criteria.[xxxv] Michigan graduates tend to succeed,[xxxvi] whether they score at the upper levels of standardized admissions tests or whether their admission was due more to other strengths.  

The test-centered approach, in other words, allows efficient, but not necessarily “merit-based” decision-making.[xxxvii] This is an important distinction: even conceding the LSAT’s very modest ability to “predict” first-year grades,[xxxviii] such tests do not predict much beyond a student’s performance six months to a year after they take the test—and then, only in comparable (i.e. classroom examination) settings. 

Consistent with the Michigan findings, a Harvard study of three classes of its graduates over a thirty-year period found that two things predicted success as Harvard measured it: low SAT scores and a blue collar background.[xxxix] Harvard also measured success in terms of financial and career satisfaction and community contribution. Both studies challenge the widespread assumption that tests of aptitude, in the long run, assure competence or even quality among the high-scorers. Numerous other studies indicate a weak correlation between aptitude tests and actual performance either in the classroom or more importantly after graduation.[xl]

These findings suggest that institutions of higher education need to be more explicit about what, in fact, the school is attempting to measure or predict when it looks at an applicant. Usually we think a school is trying to measure a potentially “successful” applicant.  If we take “success” to mean anything other than high first year grades (including contribution to the diversity of the educational experience for all students, career satisfaction and community contribution), then we should seek more empirically accountable tools for achieving this.


B) Institutionalizing democratic values of public service and rotating leadership

At the University of Michigan Law School, it is its embrace of affirmative action—as an exception to the testocracy—that reflects the most significant effort to adhere to its professed allegiance to promoting public service or to training civic-minded, competent professionals.

The institutional mission explicitly includes a commitment to train lawyers who value and practice public service. Its mission statement declares that the school “looks for students likely to become ‘esteemed practitioners, leaders of the American bar, significant contributors to legal scholarship and/or selfless contributors to the public interest.”  The school also expects those it admits will, ‘have a strong likelihood of succeeding in the practice of law and contributing in diverse ways to the well-being of others.’[xli]

But if public service is a significant part of the mission, admitting the highest test-takers may actually undermine that goal.  The study found a negative correlation between high admission test scores and community service: those with high admission index scores are less likely to provide community or public-minded leadership.[xlii] 

In contrast, beneficiaries of affirmative action were more likely than their high test scoring white counterparts to mentor younger attorneys, serve on community boards, or assume leadership positions in their professional or public capacity. While years since graduation is the most important predictor of doing pro bono work, serving on community boards or providing leadership more generally, minority status is the most important of the other relevant variables.  Simply stated, minority graduates realized the expectations of the admissions committee that approved their applications. Perhaps this is because the committee actually looked at what they had accomplished in the multiple domains of their life given the opportunities presented to them. It may be that those who were leaders in their community before law school also do more relevant community and public service after they graduate.[xliii]

Whatever the admission protocol for minority applicants, the Michigan study reveals the importance of linking admission processes to long-term institutional goals. Rather than looking backwards to pick those who “deserve” to learn, institutions of higher education need to look forward in awarding opportunity. The study shows the importance and urgency, consistent with an institutional commitment to promoting public service and community leadership among its graduates, to give the opportunity to learn to those who will use that opportunity to benefit the larger society.  We need to place greater weight on what people do with their education after they graduate rather than whether they (or their parents) did something in the past to “earn” them an admissions slot in the first place.

Linking institutional mission to admission criteria seems a necessary element of any legitimate system. Certainly for any public institution, including state university law schools, excellence should be a function of larger societal needs rather than exclusively determined by individual or abstract claims.  As citizens, we should openly challenge the “merit” of those who benefit from access to educational institutions yet fail to give back to the society that enabled their elevation.  This is especially so in public universities where taxpayers subsidize the students, and in law schools, which are training “officers of the court.”[xliv]

Also important to the public character of private as well as public institutions of higher education is the idea that a permanent elite, whether built on principles of aristocracy or, like the testocracy, timed paper and pencil tests, is inconsistent with genuine democracy.  This is because democracy has at its essence the idea of rotating leadership, requiring leaders who serve in ways accountable to the people for a limited time.

Indeed, one could argue that the Michigan study suggests the ultimate danger to democratic values of a system of sorting and ranking that awards a privileged berth to those who test well and then enables these privileged souls to “reign” in peace—believing they have “earned” their success and have no obligation to give back.  This is exactly what Michael Young predicted by coining the term “meritocracy” in 1958 in a prescient satire.[xlv]  A meritocracy, he inveighed, would be a system for allocating opportunity that convinced both the winners and the losers that they each deserved their lot in life.

True to form, many advocates of a purportedly colorblind meritocracy are silent on the democratic idea that those who rule should do so in ways that are accountable to the people or to the values of the publicly-subsidized institutions (including private universities) that made their success possible. In both private and public institutions, they often fail to specify the ways their claims of fairness and excellence are based on criteria linked to the institution’s democratic and forward-looking mission.

The new meritocrats, relying instead on their individual “merit,” apparently feel no reason to justify their continued dominance in concrete and future-oriented ways from which the public directly benefits. Indeed, the richer, already-privileged beneficiaries of the testing system seem to be able pass on to the next generation, privileges intact, two important advantages: the opportunity to succeed; and the opportunity to succeed without guilt—indeed, without even the need to justify their somehow seamless and continuing intra-generational success.

As the Michigan study shows, the new meritocrats cannot claim that high test-scoring individuals are more likely to gain career or financial satisfaction because of their test-taking ability per se rather than as a result of the doors such test scores opened. Nor do they try to promote aptitude testing as a predictor of future job performance. Instead, their reliance on claims of competence is directly linked backwards to students’ aptitude test scores and high school or college grades. Similarly the new meritocracy’s claim of inclusiveness, while reaching beyond the old boys network, falls short of the democratic principles of accountability.  They pay lip service to public service but do little to enforce it as an admissions criterion for all students.[xlvi]

Indeed, the Michigan study confirms the need to re-think many of the qualifications for admission in order to link them to long-term institutional needs and values. A school, for example, that wishes to choose students who will have successful legal careers, true to the democratic aspirations of higher education and legal education, could read this study and conclude that affirmative action type admission processes as employed at Michigan are far superior to test-based procedures.[xlvii] Indeed, this study suggests the value of affirmative action as an experiment that should inform the way we admit everyone.

It follows, then, that the special scrutiny involved in affirmative action admissions policies should provoke a fundamental re-examination of conventional admission practices.  This is not as provocative a claim as might first appear.  It does not mean that race conscious policies should permeate law school or college admissions.[xlviii] It does not mean that conventional admissions processes are complete failures.[xlix] What it does mean is that we need to place the debate about merit in a more explicitly democratic framework.


Our commitment to democratic values benefits from studies like the one at the University of Michigan, which showcase the experience of people of color and many women, who carry a commitment to contributing back to those who are less fortunate. In this fuller accounting of the democratic values of publicly supported institutions, each of us is then obligated not only to succeed as individuals, but to “lift as we climb.”

Merit becomes a forward-looking function of what a democratic society needs and values rather than a fixed, quantifiable and backwards-looking entity that, like one’s family tree or family assets, can be chronicled with the proper instruments. Merit, in other words, becomes future-oriented and dynamic. Dynamic merit involves a commitment to distribution of opportunity not only at birth but also through one’s life. It is contextual and resistant to standardized measurement. It is changing and manifests itself differently depending on how you look at it. It requires modesty in our beliefs about what we can measure in human beings, even as it demands clarifying and explicitly stating our institutional objectives.

The goal, in other words, is to make good on our educational institutions’ commitment to democratic values in a multiracial society. The practice referred to at the beginning of this discussion, “confirmative action,” is to adopt admissions criteria that confirm that goal. This is not a plea for a specific magic bullet replacement for the testocracy. Nor is it an unqualified endorsement of affirmative action as the only tool to achieve coherence between what we say we want and what we are doing. Rather it is a claim that affirmative action is a valuable experiment from which we might learn. It is an experiment that teaches us to engage in a range of selection practices, each of which should be assessed in terms of its ability to realize the mission of the institution within the future-oriented and continuously changing framework of a multiracial democracy.

Confirmative action frames the question of allocating high-stakes opportunity by connecting it to a larger conversation about democracy.  It confirms the lessons learned under affirmative action that those who are given an opportunity to succeed, when motivated and when placed in an environment in which they can learn, do just that. Confirmative action affirms the value of identifying those who are not only likely to succeed but also to contribute, based on ability to overcome obstacles, to succeed in comparable settings, or to demonstrate character traits that motivate one to succeed not only in the classroom, but also in a place of employment and in a participatory democracy.  These qualities include (but are not limited to) leadership, emotional intelligence, critical and creative thinking, ability to seek help when needed, and ability to work and play well with others.[l]

Confirmative action openly experiments with whole person evaluations, deliberative committee evaluations in which a diverse group of informed participants make collective decisions in light of the individual’s potential and the needs of the institution.[li]  By confirming the need constantly to experiment, confirmative action uses the feedback of those who are left out to reevaluate and perhaps even change the systems in place. Confirmative action determines first the mission of the institution and only then seeks to develop a system for admission.

            However, confirmative action goes beyond affirming the experience of affirmative action for its specific beneficiaries. It decouples merit from testable mental aptitude for all applicants, thus freeing us to envision many more ways of experimenting with systems for allocating scarce resources. Some colleges, such as Mt. Holyoke, Bowdoin and Bard, might abandon the SAT entirely.[lii] Other universities might consider using the tests as a floor, below which no one in recent memory has succeeded in graduating from the institution. Above that test-determined floor, applicants could be chosen by several alternatives, including a lottery, [liii] portfolio based assessment,[liv] or a more structured and participatory decision making process.[lv]  If the SAT predicts only 17% of first-year college grades, then a college might consider limiting the use of the test scores to a comparable portion of its first year class.[lvi] Or at minimum it could choose to report to the U.S. News and World Report only an equally limited percentage of its admitted applicants’ test scores.  

             Alternatively, since past grades are in fact a better predictor than SATs of future grades, and if all that a college seeks to know is how well students will do in college, then offering admission based on grades alone makes some sense. For example, the University of Texas and the University of California have offered admission to high school students in the top 10% and 4% of their graduating classes, respectively.[lvii]  This has benefited not only students of color in Texas, but rural and poor whites as well who had been underserved by a test-centered admissions practice. From lottery or random selection above a floor to admission based on a geographic approximation of high school representation (the Harvard preference for geographic diversity, for instance), confirmative action promotes context-based decision-making.  

The relevant context is the mission of the institution in question that constantly seeks feedback in order to modify admission protocols as necessary.  Considering the democratic mission of higher education, at least three important questions should guide such an approach:  

1)      Are admission processes consistent with the purposes of the institution? Do they award opportunity broadly?  Do they admit people who demonstrate competence and potential?  Do they legitimate the institution as a public place?

2)      Are the decisions being made consistent with those processes?[lviii]  Is the institution acting transparently and fairly?

3)      Are the decisions and the processes accountable to democratic norms of leadership? Do they rotate leadership or beneficiaries in light of information and feedback?  Are those who succeed also leading and making efforts to contribute back to the institution and the society it serves?

Such an approach should be accompanied by a more explicit statement of the values an institution affirms and the way those values link up to the changing needs and future interests of a multiracial democracy.  

This approach has several potential benefits. It gives the losers hope in the future and at least removes the permanent stigma of either winning illegitimately or losing because one is judged stupid.[lix]  In addition to infusing new energy, ideas and perspectives into the admissions process, it has built-in feedback mechanisms that enable institutions to rethink their choices or their processes while simultaneously assessing whether the public institution is in fact accomplishing its own mission.  

It enables the institution to assess whether the groups being served represent democratic values or merely a narrow slice of the polity.  By setting in motion this larger conversation, it also enables us to question why such important resources as educational opportunity should be so scarce in a rich country like ours. It gives us a way to ask whether state budgets are being efficiently or wisely developed when they allocate so much more to penal institutions than institutions of higher learning.[lx]

Confirmative action not only asks us to confirm our values, it begins to answer the question: How do you get those with power to share their power voluntarily? It answers that question in several ways. It may persuade those who administer the testocracy to change their practices in order to preserve the legitimacy of their institutions.  It may also encourage current winners of the testocracy to acknowledge that those awarded privilege and opportunity should not be exempt from responsibility and citizenship. It encourages experimentation that showcases the strengths, as well as exposes the weaknesses of potential alternatives.

Finally, in the most ideal sense, it redefines power as a form of sharing rather than hoarding. If no system of admission is permanently enshrined, those who lose today may nevertheless organize to change the system tomorrow. It also does what every democracy ought to do: it continually questions both the right and the good in a way that reinforces social solidarity.

By locating merit in a democratic understanding, true freedom is redefined from its libertarian roots to its democratic potential, requiring  “the maximum power for all members of human society” to make the best of themselves.[lxi]  But true freedom also means linking the potential of the individual to ongoing societal needs and to the needs of particular public institutions making admissions or recruiting decisions. Freedom in a genuine democracy involves individual freedom in a collective enterprise. Although each individual’s potential for growth and development should be maximized, the needs of the general welfare cannot be ignored. Confirmative action in public institutions thus involves commitments both to the individual and to the group.

Toward this end confirmative action builds on the values of affirmative action but does not limit the practice to a historically disadvantaged group. Confirmative action awards opportunity more broadly, but without a one-size-fits all metric. It joins the idea of ‘opportunity for all’ that underpins the new meritocracy to the idea that no one-size-fits-all gauge can fairly allocate such opportunity.  It thus builds upon the lessons from the new meritocracy and affirmative action, but situates the practices of admission or selection within a clearer and dynamic understanding of institutional mission.

Confirmative action is a practice of social justice that is both experimental and contextual. It challenges the idea that those who are already privileged can be trusted to administer or justify a system for allocating opportunity. But it challenges that idea on behalf of all of those who have been left out, not just those who were deliberately excluded because of race, color or previous condition of servitude. It awards opportunity based on how people are likely to use that opportunity in the future. Most of all, confirmative action re-connects issues of selection and admission to the dynamic practices and democratic values of higher education.



© 2000 Lani Guinier


[i] Professor, Harvard Law School. Copyright with author. This article first appeared as an address delivered June 14, 2000 before the National Urban League’s State of America 2000 Conference. Steig Olson, HLS 2001, Jeannie Suk, HLS 2002, Samantha Bent, Harvard 2001, Alicia Johnson, Harvard 2001 and Angie Young, Stanford 1997 assisted in research related to this essay. Angie Young also provided superb technical and editorial support in the preparation of this essay. Mindy Kornhaber’s unpublished essay on the mission of public institutions of higher education was also invaluable. I thank Lee Daniels for his able assistance in helping me revise and shorten the original manuscript.

[ii] Nicholas Lemann, The Big Test (1999) at 164.

[iii] The term testocracy highlights the ways in which selection policies are heavily dependent on standardized aptitude testing (See Susan Sturm and Lani Guinier, The Future of Affirmative Action: Reclaiming the Innovative Ideal, Cal. L. Rev. 84: 953, 968 (1996)).  By testocracy we refer to test-centered efforts to score applicants, rank them comparatively, and then predict their future performance.  In this essay I limit the discussion of testing, however, to aptitude tests (as opposed to achievement tests).

[iv] I borrow this term from David Wilkins.

[v] Recent news accounts, however, suggest that at least the movement to introduce standardized tests has experienced a backlash when parents realized that it was their own children who would be held back based on test scores. See John T. McQuiston, Parents, Teachers and N.A.A.C.P. Sue School District Over Grouping of Students, N.Y. Times (January 14,1999). The realization that aptitude testing has an even more disparate effect has not yet percolated into national consciousness.  See William G. Bowen and Derek Bok, The Shape of the River (1998) at 276-286 (arguing that grades and test scores are not necessarily an indication of how hard students work, because they are based not only on effort but also on intelligence, which is influenced by factors like “inherited ability, family circumstances, and early upbringing.”  Test scores are also influenced by the quality of teaching students have received and by the ability students have to strategize at taking these tests.

[vi] Christopher Jencks and Meredith Phillips, The Black-White Test Score Gap (1998) (stating that for blacks as well as whites, cognitive achievement as measured by test scores largely determines professional success). See also Memo from Gavin Kearney (December 21, 1999) (stating that high educational achievement typically leads to better paying jobs, which in turn result in greater wealth accumulation).

[vii] Within each racial and ethnic group, SAT scores increase with income. See Sturm and Guinier at 989 (citing 1995 Profiles of SAT Scores by Income and Race from the College Entrance Examination Board).

[viii] I thank Jeannie Suk, HLS 2002, for naming this concept.  See also Sturm and Guinier in which an earlier version of this concept is set forth.

[ix] The challenge of embedding admissions into the institution’s mission should prompt us to reconsider issues of curriculum, learning theory and ways to motivate members of a diverse population to each reach his or her full potential. However, these issues are beyond the scope of this essay.

[x] Glenn C. Loury, Boston University, Draft Introduction for The Shape of the River, submitted for presentation January 7, 2000 (on file with author).

[xi] Some argue that the role of schools is to educate citizens for the market. They argue that having a job is the essence of fulfilling one’s civic responsibilities and that one’s identity as a citizen depends upon and is derivative of one’s job. For example, in President Clinton’s 1994 State of the Union address he declared, “We measure every school by one high standard: Are our children learning what they need to know to compete and win in the global economy.”

[xii] See Mindy Kornhaber, Reconfiguring Admissions to Serve the Mission of Selective Public Higher Education (January 14, 1999) (typescript on file with author).

[xiii] Kornhaber.

[xiv] Kornhaber writes: “These purposes originate with the nation’s founding fathers. Benjamin Franklin argued that the college curriculum ought to prepare citizens for practical affairs” (citing Brubacher & Rudy, 1976).  Thomas Jefferson asserted that public education, including higher education, was needed to develop both leaders and the citizens capable of monitoring them.

[xv] Kornhaber notes that while Jefferson’s desire to foster an enlightened citizenry was clearly hobbled by the exclusion from the university of women and African Americans, public higher education has gradually become available to ever broader segments of the population, guided by federal policies. In the 1860s, land grant colleges opened up higher education to the sons and daughters of farmers and working class people.  Just over a century ago, federal legislation helped to create public colleges for African Americans in historically segregated areas (citing Christy & Williamson, 1992).  Following World War II, the GI bill enabled millions of veterans, most of who were older and not economically privileged, to receive a college education.  Soon thereafter, the Supreme Court made it illegal to exclude African Americans from state colleges and universities (Sweatt v. Painter) or to segregate them once admitted (McLaurin v. Oklahoma State Regents).

[xvi] See Lemann at 140-154.

[xvii] I have oversimplified the values and commitment of each side because this article seeks to expose their commonality.  For a more nuanced version of the debate, see Sturm and Guinier.

[xviii] At a union hall jobs are often given to those who show up on a “first come first serve” basis. This may appear fair but is in fact limited to union members, a status in many trades historically limited to whites. Both union membership and notice of job openings were often limited to a network of friends and relatives.

[xix] For a history of the new meritocracy, see Lemann.

[xx] Sturm and Guinier at 989.

[xxi] The LSAT “explains” 14% of performance, for example, at the University of Pennsylvania Law School in the first year. See Lani Guinier, Michelle Fine and Jane Balin, Becoming Gentlemen: Women, Law Schools and Institutional Change (1997). This is about average for law schools. See Michael Selmi, Testing for Equality: Merit, Efficiency, and the Affirmative Action Debate, 42 U.C.L.A. L. Rev. (1995) at 1264 (correlation coefficients for the LSAT, which is intended to predict first-year law school grades, tend to hover around .35; correlation coefficient of .3 means that the test explains 9% of the variation in predicted performance). Or as Linda Wightman has said when she was at the LSAC, nationwide the LSAT predicts first year law school grades 9% better than random.

[xxii] See Sturm and Guinier (describing the impact of the testocracy on African Americans and Latinos).  See also Kornhaber who notes that recent enrollment patterns in California and Texas highlight that score-driven admissions practices exclude large portions of the black and Hispanic population from selective institutions (citing Bowen & Bok, 1998; Chapa & Lazaro, 1998; Kane, 1998; Karabel, 1998).

[xxiii] Unfortunately, the legitimate concerns of the other “losers” who are poor or working class and white were channeled by racism to scapegoat their poor cousins across the color line.  They seemed not to notice the almost automatic success enjoyed by richer, already-privileged beneficiaries of the testing system who were also white.

[xxiv] Sturm and Guinier at 980-987.

[xxv] See William Julius Wilson, The Role of the Environment in the Black-White Test Score Gap (discussing correlation between test scores and socioeconomic status of grandparents) in Jencks and Phillips (1998) at 501-510.

[xxvi] See Kearney (arguing that “Blacks and Whites with equal incomes possess unequal shares of wealth.  Furthermore, it is important to keep note that the accumulation of wealth for Whites is intimately tied to the lack of wealth for Blacks.  Just as Blacks have had cumulative disadvantages (lower salaries, Black businesses confined to a segregated Black market, FHA’s denial of loans to Blacks), Whites have had cumulative advantages. Bias and discrimination against Blacks produces positive gains for Whites.  The cumulative effect of such a process has been to sediment Blacks at the bottom of the social hierarchy and to artificially raise the relative position of some Whites in society”).  See Kearney for summary of the literature on wealth accumulation, where asset disparities between blacks and whites are considerably greater than income disparities.

[xxvii] The accumulation of resources brought to the test is unfair because 1) it is based on inherited access since the tests are administered so early in life that one’s own achievements have yet to be made, 2) it misjudges those resource accumulations by calling them “preparedness” rather than “privilegedness,” 3) it fails to make accommodations for those less privileged, judging them by the same single rubric that has a built-in bias toward resource accumulation and 4) it assumes that testing is synonymous with doing.

[xxviii] Selective colleges and professional schools not only provide students with opportunities to earn more money than those peers who attend lesser ranked schools; they also open doors to positions of prominence within the larger society. See, Bok and Bowen; see also note 36. I am making two related points here. One is that access to higher education is access to resources over which the society as well as particular individuals may have legitimate or even competing claims. Two, in a democracy that promotes ideals of equal opportunity, access to higher education should not depend on past access to inherited resources. At the very least, to the degree such resources are simply transferred from one generation to another, it is misleading and fundamentally unfair to call such inheritances “merit.”

[xxix] See Kornhaber.

[xxx] See, e.g. Harper v. Board of Elections, 383 U.S. 663 (1966).

[xxxi] See for example, Alisa Roth, Who’s your daddy? (January 14, 2000) at “When the University of Virginia acknowledged tracking applicants based on their parents’ donor potential, few people batted an eye. They were too busy flinching over affirmative action.”

[xxxii] Thus, one of the arguments against affirmative action–that it stigmatizes its beneficiaries–is less a function of affirmative action and more a function of the asymmetric notice of exceptions that accompanies affirmative action.

[xxxiii] Some might argue that these wealthy or well-endowed applicants would not be stigmatized even if their identity were public. In this sense they would be like athletes who are admitted with lower test scores but are recognized to have a talent or a skill that benefits the institution, both by building its morale as well as its endowment.

[xxxiv] Richard O. Lempert, David A. Chambers, and Terry K. Adams, Michigan’s Minority Graduates in Practice: The River Runs Through Law School, Law and Social Inquiry 25: 395-505 (2000).

[xxxv] Lempert, Chambers, and Adams at 468.

[xxxvi] See David Wilkins, Rollin’ On the River: Race, Elite Schools, and the Equality Paradox, Law and Social Inquiry 5:527-556 (2000) (emphasizing the importance of the “big wheels” of the Michigan Law School, what others call “branding,” that continue to “keep on turning” or open doors for graduates, both white and black).

[xxxvii] See Sturm and Guinier (describing ways in which SAT and LSAT tests correlate as much with parental income as first year grades). See also Lemann (describing theocracy of testing as a religion in which mental aptitude replaced inherited privilege as the means for identifying the new elite; testing is as much faith based as scientific, since the validity measures are much weaker than most people realize).

[xxxviii] See note 21.

[xxxix] David K. Shipler, My Equal Opportunity, Your Free Lunch, N.Y. Times (March 5, 1995) at E1, E16.

[xl] See note 21. See also Guinier, Fine and Balin at 8-9. See also Sturm and Guinier at 970-980.

[xli] Lempert, Chambers, and Adams at 396.

[xlii] They mentor fewer young attorneys, sit on fewer community boards and do less pro bono work. See Lempert, Chambers, and Adams at 401. If one compares the entry-level LSAT scores of Michigan graduates with their “service” index, there is a negative relationship that is statistically significant among 1980s graduates as well. See Lempert, Chambers, and Adams at 468-69.

[xliii] According to Lempert, the school’s affirmative action efforts were initially based on the assumption that if they put people in an environment like the University of Michigan Law School, they would flourish.  My impression from the study, and from my own experiences, is that the admissions process for minorities considered the “whole person” in deciding whom to admit among competing, qualified applicants.  To some extent, the school most certainly relied on test-based credentials for minority applicants; yet again, as is true at other institutions, they probably used them as a cut-off floor rather than an acutely sensitive ranking system.

[xliv] See, e.g., the American Bar Association Model Rules of Professional Conduct, 2.1, 2.2, and 2.3 that recognize the lawyers’ “special responsibilities for the quality of justice,” by contributing to the profession’s fulfillment of its responsibility to ensure that adequate legal services are provided to those who cannot afford to pay for them and by contributing to the profession’s fulfillment of its responsibility to enhance the capacity of law and legal institutions to do justice.

[xlv] See Michael Young, The Rise of the Meritocracy (1958).

[xlvi] What might happen, for example, were institutions to award degrees conditional on graduates fulfilling the promises they made in their admissions applications? For example, institutions could award provisional degrees that could only be certified if the graduate within ten years of graduation accomplished some of the public service, educational leadership, musical or artistic contributions that was the basis for their admission.

[xlvii] The study suggests that the dedication necessary to achieve a high LSGPA may result in long-term career dissatisfaction, because a person with such dedication is likely not good at balancing “separate spheres” of satisfaction: “Those likely to have concentrated most on getting good grades while in law school may be more likely than others to dedicate themselves to their jobs and to narrowly defined job responsibilities. The result is that they tend to earn more than others, but they also tend to do less service and to feel less satisfied because their jobs are so consuming.” See Lempert, Chambers, and Adams at 489-90.

[xlviii] I am arguing here that race as a ceiling or as a set aside is not the most important message of affirmative action. Race may in fact stand as a proxy for things we value, if for example, being part of an historically oppressed group gives individual members greater incentive to do community service, to mentor younger attorneys, or to become public minded leaders. But, even if race does function as a surrogate, it is useful as a floor not a ceiling. It is useful to identify potential applicants who then should be evaluated further. It is also useful as a source of feedback, if for example, the institution’s practices end up promoting a monolithic class without a range of perspectives, backgrounds or future commitments. This idea of “race” as a stand-in for access to power and attitudes toward social change is developed further in Lani Guinier and Gerald Torres, The Miners Canary: Rethinking Race and Power (forthcoming Harvard University Press 2001). It is also summarized in Sturm and Guinier.

[xlix] LSAT and UGPA scores do in fact correlate modestly with high law school grades first year and almost without exception, Michigan graduates who were admitted pursuant to such indices are successful, measured with respect to society at large. On the other hand, whether a particular way of admitting students is a failure turns on both: a) the baseline against which it is compared or the goal it is trying to achieve and, b) the potential of other alternative methods of admission. Thus, if all that Michigan Law School is attempting to do is admit students who will get good grades in law school, one could argue that the conventional indices are modestly successful.

[l] See Warren W. Willingham, Success in college: The Role of Personal Qualities and Academic Ability (1985) (finding that sustained evidence of leadership is best predictor of ultimate success).

[li] When I was on the admissions committee in the early 1990s at the University of Pennsylvania Law School, the process of admitting people who had some “special” quality to be considered (which included being a poor white chicken farmer from Alabama) was an openly deliberative process. It included students who knew more about the specific localities in which many of the applicants resided. The applications were redacted to eliminate personal identifying information but were otherwise available to the entire committee. The recommendations were read and considered (by contrast to the 50% of the class who were admitted solely on a mathematical equation based on their LSAT scores, their college rank, and the “quality” of their college as determined by the median LSAT score of its graduating class). In this process, the committee of faculty, students and admissions personnel had a sense we were admitting a “class” of students, not just random individuals. Thus, we might give weight to some factors over others, depending upon the “needs” of the institution to have racial and demographic diversity, but also upon our commitment to fulfilling the needs of the profession to serve the entire public and to train private and public problem solvers who would become the next generation of leaders. Thus, not all students were assumed to be as academically talented. Some might be better oral advocates, litigators, negotiators, or future practitioners of alternative dispute resolution practices, but all were judged qualified to do the work demanded at the institution.

[lii] Jodi Wilgoren, Mount Holyoke Drops SAT Requirement, N.Y.Times (June 7, 2000) at A22 (Mt. Holyoke, the nation’s oldest institution of higher learning for women, decided to stop requiring applicants to submit SAT scores, joining a “handful of highly selective colleges, including Bates and Bowdoin, both in Maine, Connecticut College in New London and Bard in Annandale-on-Hudson, N.Y., which have made standardized tests optional”). About 390 of the country’s nearly 1,800 four-year colleges and universities do not require such tests.

[liii] Sturm and Guinier at 1018, notes 271, 274 (describing a proposal to use a lottery above an admission-index floor to admit students to a magnet school in San Francisco so that random selection above a baseline of qualification becomes the principle for distributing high-stakes opportunity.  A lottery is a useful heuristic device because it at least exposes the arbitrary nature of allocating a scare resource; it might motivate those who “lose” to challenge why higher education is not more widely available since it is so important.

[liv]See Dennie Wolf, Portfolio Assessment: Sampling Student Work, Educational Leadership 46(7) (1989) at 35-39.  See also Samuel J. Miesels, Using Work Sampling in Authentic Assessments, Educational Leadership 54(4) (1996-1997) at 395-505.

[lv] See Sturm and Guinier at 1017-1019 (citing examples of instituting such processes in educational institutions). See also example in note 51.

[lvi] It is true that predicting the variability among a sample of those admitted does not predict the variability among the entire applicant pool. Nevertheless, if among those who are admitted the tests predict a limited amount of first year performance, that limited amount becomes an arbitrary yet quantifiable basis for pegging a limited use of the test.

[lvii] See Commission Denounces Affirmative Action Replacement Programs, The Associated Press (April 11, 2000) at

[lviii]Tim Scanlon helped me to see these two preconditions. The first, whether the processes are consistent with the purposes of the institution, generates claims by the polity to which the institution is accountable. The second, whether specific decisions are consistent with the announced processes, may generate claims by affected individuals. On the other hand, the first claim could conceivably be made by an individual that the institution’s purported mission is flawed and both claims could arguably be seen as a function of circumstances in which the institution fails to make its goals or its processes sufficiently transparent.

[lix] Of course, any system that promotes or acknowledges the likelihood of losing may stigmatize. My argument is that the stigma of losing is much worse when it is attached to things we cannot change, such as our inherited intelligence, or reinforces stereotypes about historically disadvantaged groups. The idea of losing because one is unlucky may lead to initiatives to expand what are now scarce resources, especially if the losing group includes many who are normally resource-rich and whose sense of entitlement may encourage them to fight to expand the resource base. I thank Robert Gordon for this insight.

[lx] In 1995, California spent more on prisons than it did on higher education. See Troy Duster, The New Crisis in Controls, Prisons, and Legal Structures, The American Sociologist (Spring 1995) at 20-24.  We should also consider how we are allocating resources when one of every three African-American men between the ages of 20 and 29 are under the jurisdiction of the criminal justice system (See Ira Glasser, Biennial Speech (June 1999)).

[lxi] Fareed Zakaria, Beyond Money, N.Y. Times Book Review (November 28, 1999) at 14 (quoting T. H. Green writing in 1881).