Lawrence Friedman*

Abstract:  Against the background of the debate over the constitutionality of state-sponsored higher education affirmative action programs, this Essay presents an account of the modern equal protection framework that balances a normative concern for the protection of individuals from discrimination and a structural concern for the counter-majoritarian difficulty. The author suggests that state-sponsored higher education affirmative action programs may survive strict scrutiny, and that proposals urging the judicial consideration of public opinion about such programs should be rejected. Even if public opinion could be accurately gauged, it should not influence the application of strict scrutiny to such programs, or play any part in equal protection review.


In Grutter v. Bollinger,1 the United States Supreme Court ruled that state-sponsored higher education affirmative action policies do not necessarily violate the Fourteenth Amendment’s commitment to equality. In Grutter, the Court applied strict scrutiny to the University of Michigan Law School’s admissions program, holding that the law school had sufficiently demonstrated a compelling interest in attaining a diverse student body,2 and that its admissions program was narrowly tailored to serve that interest because it did not “insulate” certain individuals “from competition with all other applicants.”3 In addition to establishing that state-sponsored higher education affirmative action [*PG268]policies may be constitutional, a principle in some doubt in the years since the Court decided Regents of the University of California v. Bakke,4 Grutter confirmed that the federal courts should apply strict scrutiny when faced with equal protection challenges to such policies.

Though the latter point may now appear unremarkable, a majority of the Court only relatively recently concluded that strict scrutiny should be applied to all racial classifications, including those associated with affirmative action programs.5 And, prior to Grutter, it was not at all clear that the Court would find that diversity supplies a sufficiently compelling interest to support the kind of preferential treatment based upon race that affirmative action necessarily involves, or that efforts like the University of Michigan Law School’s to limit the discriminatory effects of affirmative action would satisfy the narrow tailoring necessary for race-based classifications to survive strict scrutiny. Indeed, it has long been thought, to quote Gerald Gunther, that strict scrutiny in the equal protection context is “strict in theory and fatal in fact,”6 and recent judicial applications of strict scrutiny have not indicated that the Court, when it reached the issue of higher education affirmative action, would regard the matter any differently.7

Still, after Grutter, questions remain about whether the Court’s approach to the University of Michigan Law School’s policy will be relevant in settings other than higher education,8 and, more generally, what limits courts should respect when undertaking strict scrutiny equal protection review. In the case of higher education affirmative action, Professor Gail Heriot has suggested that courts, in determining whether policies satisfy strict scrutiny, should take account of majoritarian sentiment toward affirmative action by considering the views of the citizenry on the issue, as represented not by their formal expression—that is, the positive legal enactment itself, be it by statute, ordinance, or regulation—but by raw public opinion, as reflected in [*PG269]surveys, polls, and responses to ballot questions.9 She reasons that consideration of raw public opinion is appropriate because affirmative action represents a failure of representative democracy in respect to a policy that, surveys and polls show, people consistently oppose.10

Professor Heriot made this proposal before the Court decided Grutter. Though the Grutter majority did not consider public opinion on the issue of affirmative action in higher education in reviewing the University of Michigan Law School’s admissions program, Heriot’s proposal warrants examination, if only because it implicitly raises anew the question of what factors a court ought to examine when it entertains an equal protection challenge to governmental action. Heriot’s proposal raises this question not just in respect to affirmative action, but in any case in which the exercise of judicial review might result in the invalidation of an otherwise constitutional law.

In this Essay, I seek to explore the premises underlying the modern equal protection framework and the goals the framework achieves. Given the institutional role for the judiciary that the framework contemplates, I discuss the reasons why the courts should not in any case consider public opinion in determining whether governmental action comports with the constitutional commitment to equality. I begin in Part I by sketching an account of how courts approach judicial review in the equal protection context in view of both a normative concern to prevent political majorities from undermining the interests of “discrete and insular minorities”11 and the recognition that such a concern could lead to judicial overreaching and exacerbate what Alexander Bickel termed “the counter-majoritarian difficulty.”12 On this account, state-sponsored higher education affirmative action programs are not necessarily at odds with the Fourteenth Amendment, even though they contain racial classifications.13

[*PG270] Responding to Professor Heriot’s specific proposal, Part II addresses the argument that popular opinion should be a relevant consideration in certain cases involving suspect classifications. I conclude that, even assuming it could be accurately measured, popular opinion on an issue should be viewed as beyond the scope of the equal protection inquiry, even in respect to state-sponsored affirmative action, because it may undermine a court’s institutional role in equal protection cases. Finally, in Part III, I briefly address alternate ways in which to apply the equal protection framework so as to allow the courts to effectuate their institutional role, while at the same time respecting the legitimate limits of their authority in constitutional cases.

I.  The Modern Equal Protection Framework
and Affirmative Action

As the Supreme Court itself has observed, “[t]he Fourteenth Amendment’s promise that no person shall be denied the equal protection of the laws must coexist with the practical necessity that most legislation classifies for one purpose or another, with resulting disadvantage to various groups or persons.”14 In order to implement the constitutional commitment to equality without making nearly every governmental action a dispute that the courts must ultimately resolve, the Court has employed a now-familiar framework: laws that do not burden a fundamental interest or target a suspect class of individuals—that is, classifications based upon such characteristics as race,15 alienage,16 nationality,17 and, lately, sex18—will be upheld so long as [*PG271]the classification “bears a rational relation to some legitimate end.”19 Laws that do burden a fundamental interest, however, or single out a suspect class for differential treatment, will be subject to strict scrutiny, pursuant to which the classification must serve a compelling governmental interest and be narrowly tailored to promote that interest.20

The modern equal protection framework reflects a balance between competing interests. On the one hand is the courts’ legitimate concern to avoid the tension associated with judicial decision-making that is potentially counter-majoritarian. This tension may arise when an unelected judiciary exercises its authority to review and, perhaps, invalidate laws duly enacted by the people’s representatives in the legislative and executive branches of government.21 The counter-majoritarian difficulty, in other words, refers to the potentiality that a court, in saying what the Constitution requires in relation to a particular law—an act that necessarily involves some effort at textual interpretation—and then applying that understanding to the facts at hand, may invalidate a majoritarian preference, as formally expressed in the governmental action at issue, because it is inconsistent with constitutional principles.22

On the other hand is the institutional role the courts play in our constitutional democracy in respect to the Fourteenth Amendment’s commitment to equality: the judiciary serves a democracy-reinforcing function by, among other things, preventing political majorities from using lawmaking processes to discriminate against certain individuals.23 Through equal protection review, the courts can ensure that the dignity of those individuals who might be targets for discrimination is respected in the same way as that of other, similarly situated individuals. For democracy to function optimally, after all, “citizens from all walks of life [must] interact freely on terms of equality,”24 and equal protection review provides the courts a means to catalyze such interaction by invalidating those terms of inequality that serve to exclude certain individuals from the larger community.

The equal protection framework mitigates the potentially counter-majoritarian effect of equal protection review by cabining its scope. The framework assumes that there are bound to be instances in which the courts should not defer to the political branches because certain individuals, due to characteristics that identify them as members of a minority within a community, may suffer at the will of a political majority. The framework recognizes that some governmental classifications that emerge from ordinary political processes may reflect accommodations made at the expense of minority groups; after all, “ins have a way of wanting to make sure the outs stay out”25—more so when the outs may be defined by such immutable characteristics as their race or sex. Accordingly, we reasonably doubt that governmental classifications based upon such characteristics resulted from the benign operation of ordinary lawmaking, and we are not confident that the individuals negatively affected by such classifications could effectively remedy their situations through the political process.

Thus, the equal protection framework serves to focus attention on those instances in which we have doubts about the bases for certain classifications, causing us to suspect that representative democracy may have failed certain individuals in the community.26 The framework alerts lawmakers, judges, and lawyers to the particular features of a small number of classifications that relate to certain groups of individuals. In those cases, the government must show that a compelling interest justifies the discrimination at issue, and that the means to achieve that compelling interest are narrowly tailored.27 But, absent a suspect classification—in other words, in the vast majority of cases in the federal and state courts involving equal protection claims—a court will inquire only as to whether there exists a rational basis for the legislative classification, and whether the means of achieving that rational end are [*PG273][*PG272]reasonably related to the classification.28 This is a rather deferential standard, under which courts will often presume the validity of the classification at issue absent evidence to the contrary.29

Indeed, the key to the way in which the equal protection framework manages the counter-majoritarian tension is its emphasis on significant judicial deference toward governmental classifications in the mine run of cases. By focusing judicial attention on the question of whether a governmental classification is suspect, the framework functions to alleviate the counter-majoritarian tension by providing that, in most cases, the courts will review a governmental classification in a deferential light. The framework in this way seeks to maximize the reach of majoritarian impulses, as expressed through the political process, by delineating those very few instances in which a court will not show deference in assessing the constitutionality of a particular classification.

The equal protection framework accordingly defines the extent to which majoritarian impulses can be realized before a court will seriously question the validity of a governmental classification. That definition contemplates a broad plane of governmental discretion that may, in certain instances, contract, depending upon the circumstances of the case and the nature of the discrimination alleged.30 The framework thus legitimizes equal protection review by limiting the cases in which judicial deference toward the government’s classifications will be suspended. In particular, the framework evidences the judiciary’s ability to control itself by providing for the kind of consistency and predictability in judicial decision-making that allows the courts to fulfill their [*PG274]institutional role without anxiety in those instances in which, history has shown, their intervention is most needed.31

Notably, even in those few instances in which the courts suspend deference, the governmental classification may still survive equal protection review. Strict scrutiny need not inevitably be “strict in theory, but fatal in fact.”32 Though it may be difficult for the government to justify a suspect classification by demonstrating the existence of a compelling interest and narrowly tailored means with which to achieve that interest, the Supreme Court has made clear that such interests do, in fact, exist, and that narrow tailoring is possible. In the First Amendment context, for example, the Court concluded in Burson v. Freeman33 that a state had a compelling interest in protecting the right of its citizens to vote in an election conducted with integrity and reliability, and that a restricted, speech-free zone around polling places was narrowly tailored to serve that interest.34 The law at issue survived strict scrutiny, demonstrating that a compelling interest and the narrowly tailored means to achieve that interest are not merely theoretical constructs.

In light of the concerns underlying the articulation and application of the equal protection framework, affirmative action programs are not necessarily doomed under strict scrutiny review, even though they sanction preferential treatment based upon race. First, such programs may survive strict scrutiny because, while they establish classifications based upon race, they do not fall squarely within the central concern of strict scrutiny—namely, the exclusion of minorities by the majority in a community. As John Hart Ely noted some thirty years ago, “[w]hen the group that controls the decision making process classifies so as to advantage a minority and disadvantage itself, the reasons for being unusually suspicious . . . are lacking.”35 And, second, affirmative action programs may survive because the government may well be able to demonstrate compelling reasons that justify appropriately tailored affirmative action programs—as the decision in Grutter v. Bollinger demonstrates.36

[*PG275]II.  A Modest Proposal and Two Responses

The understanding of the equal protection framework sketched in Part I is premised on the notions that, in view of the counter-majoritarian difficulty, the courts should in most instances respect the results of the political process when entertaining an equal protection challenge, and that such deference should be suspended only when certain suspect governmental classifications are at issue. Professor Heriot, however, suggests that the case of affirmative action in higher education, which involves classification based upon race, is different. She argues that, in determining whether this kind of classification survives strict scrutiny, courts should show some deference to raw public opinion in opposition to higher education affirmative action because public opinion reflects a historically sound suspicion of racial preferences of all kinds.37

As an initial response to Heriot’s modest proposal, we might query whether it is possible, as a general matter, to ascertain an accurate representation of public opinion. Further, assuming a court could identify and quantify public opinion, there is still the question of whether consideration of public sentiment would upset the balance that the equal protection framework strikes between judicial overreaching and the judiciary’s institutional role in implementing the constitutional commitment to equality. I conclude that, from both an empirical and a normative standpoint, the courts should not consider public opinion in undertaking equal protection review.

A.  The Problem of Identifying and Quantifying Public Opinion

For purposes of judicial review, the relevant expressions of public opinion in our constitutional democracy are the result of the lawmaking apparatus of representative government: statutes, ordinances, and other like products of the political process. It is upon such articulations of majoritarian sentiment that courts traditionally have relied, for the reason, among others, that reliance upon any other source of majoritarian sentiment risks opening the door to judicial reevaluation of a great many laws that any plaintiff with standing could argue were not [*PG276]actually supported by a political majority within a given community.38 That is a recipe for undermining representative democracy, which is premised on the notion that elected politicians ought to be primarily and directly accountable for decision-making and for laws that are otherwise consistent with federal and state constitutional limits.39

Professor Heriot, however, maintains that representative democracy does not work, at least where affirmative action is concerned.40 And, she suggests that, in respect to higher education affirmative action, which involves racial classifications to which, she claims, most Americans are opposed, the courts should take account of the alleged failures of representative democracy in applying strict scrutiny equal protection review.41 Setting aside the question of whether the claim that representative democracy does not work in this instance is accurate or meaningful, Heriot’s proposal would create, at a minimum, an evidentiary issue—namely, how courts ought to identify and quantify public opinion, if not by relying upon the laws that result from the political process as indicative of majoritarian sentiment.

This is the stuff of which protracted litigation is made. Can we say that we know with any certainty the public’s opinion on a particular issue? Is public sentiment accurately reflected, for example, in the results of opinion surveys and polls? Heriot cites data showing that ninety percent of “whites” oppose affirmative action in hiring and promotion as well as in college admissions.42 The accuracy of such data may rightly be regarded as questionable. Professor Patricia Chang has documented the difficulties of obtaining data about public opinion from surveys and polls; as she notes, “survey results are rarely presented for public consumption with a reckoning of the response [*PG277]rate—a standard statistic that tells how reliably the results represent the target population to which the data is ascribed.”43

Further, there is authority to suggest that the results of surveys and polls may depend on who is doing the asking or the polling, and what the participants are being asked.44 Social science research indicates that the way in which questions are posed and hypothetical situations are framed affects the way in which participants assess value choices.45 These findings could also undermine the validity of using ballot measures, such as initiatives prohibiting affirmative action policies,46 as a gauge of public opinion, when the very way in which the ballot questions are posed may affect the way that voters respond. Of course, whether a ballot measure in one state can be said to represent the views of the citizens of other states on an issue, or the views of a majority of Americans, is, at the very least, debatable.

B.  Public Opinion and Equal Protection Review

Granted, because the issue of how best to ascertain public sentiment on a particular issue is essentially evidentiary, it may not be immune to resolution—perhaps by requiring that public opinion data offered in equal protection litigation survive a Daubert-type challenge.47 Nonetheless, the question remains whether, assuming opinion surveys and polls could be crafted to accurately reflect popular understanding of an issue and the views of the individuals who participate, courts should take that data into consideration when applying strict scrutiny in cases involving affirmative action. Professor Heriot contends that courts should do so because state-sponsored affirmative action represents an instance in which “the state bureaucracy favors racial discrimination while the people do not.”48 She argues that the courts, when [*PG278]undertaking equal protection review, should act to correct the failure of representative democracy by taking account of majoritarian sentiment opposing all racial classifications, including those associated with preferential treatment resulting from affirmative action.49

Though it may well be true that representative democracy, as currently configured, is not without its limitations, recognition of that fact does not necessarily mean the courts should ever consider popular opinion, in addition to the government’s proffered justification for a law, to be relevant when reviewing the law’s constitutionality. The judiciary is charged in our constitutional system with saying what the law is, and that responsibility requires courts to reconcile competing arguments about the validity of laws that classify individuals. Judicial reliance upon the government to make a case for the constitutionality of a particular classification, such as one that advantages minorities in higher education admissions, is, in effect, democracy-reinforcing because it serves to make transparent the work of the people’s representatives in the political branches of government. That is, it clarifies for the community the classifications that the people’s representatives have formally enacted into law, as well as the justifications that the government has put forward in support of those classifications.

After all, the nature of equal protection review, and especially strict scrutiny review, inevitably requires the government to articulate detailed justifications for the classifications contained in a particular law. As well, the reviewing court, in assessing the weight of the proffered reason for a classification, must undertake to explain the law and its effects. These are requisites of equal protection analysis, which are made public through a judicial decision. And, in the written opinion that accompanies the judicial decision, the people have a means by which to monitor the work of their elected representatives—a means by which they can learn, with great precision, of discrepancies between what they actually desired and what their representatives enacted into law. By relying exclusively upon the government’s justifications for a law, rather than upon assessments of popular opinion about the law, the courts effectively enable the people to hold their representatives accountable for misconstruing their interests, should they so desire.

Nonetheless, critics might argue, the reasons for a classification that the government’s lawyers put forward in litigation may not repre[*PG279]sent the true intent, if any, of the legislature in passing a law or the executive in promulgating a particular regulation or policy.50 While many justifications for classifications may be post hoc, at a minimum those justifications become operative as a result of an equal protection challenge to the law. At that point, the court must determine and explain what, exactly, the legislature or executive has wrought, and the government’s lawyers must articulate some justification for why the legislature or executive did what it did. That announced justification then becomes a benchmark against which the people can test the difference between their interests, as expressed to their representatives through political action, and the response of their representatives—if indeed their representatives responded at all.

In its opinion in Grutter v. Bollinger, for example, the Supreme Court described in great detail the University of Michigan Law School’s formal affirmative action program—its constituent parts as well as its effects.51 The Court also carefully explained the position of the state in the litigation, as well as the evidence the state’s lawyers offered in support of the constitutionality of the university’s policy.52 It is the quality of the state’s arguments in favor of affirmative action that the Court considered in rendering its judgment on the equal protection issue. Importantly, it is also the quality of those arguments, and the effects of the affirmative action program that the state supported, by which the people of Michigan can judge their representatives and hold them accountable for the choices made by the university. If, after Grutter, there are Michigan citizens who do not believe that the university’s affirmative action program and the announced justifications for that program accurately reflect the policy choices they would prefer their representatives—or the bureaucrats for whom those representatives are accountable—make, those citizens may seek change through the political process. They may band together to urge that their representatives pass appropriate laws or replace the bureaucrats responsible for the policy, or to urge that those representatives themselves be replaced in the next electoral cycle.

[*PG280] In this way, the decision in Grutter v. Bollinger respects rather than undermines the tenets of representative democracy, for it does not elide the fundamental importance, in a republic, of the people taking responsibility for holding their government accountable for its actions. This is not to say that it is in any way a simple matter for citizens who feel aggrieved by their representatives’ actions to effect change. Professor Heriot is correct when she notes that those individuals who seek to bring about changes in the law through the political process face numerous obstacles, including entrenched incumbents, the power of special interests, and general voter apathy.53 As an equal protection matter, however, these problems are not necessarily a court’s concern.54

III.  Conclusion: The Equal Protection Framework, Revisited

In this Essay I have presented an account of the modern equal protection framework that balances a normative concern for the protection of certain individuals in a community from discrimination, with a structural concern for the need to manage the tension to which the counter-majoritarian difficulty may lead. I have suggested that state-sponsored higher education affirmative action programs may survive strict scrutiny, and that consideration of public antipathy toward such programs, even if it could be accurately gauged, ought not to influence the application of strict scrutiny to affirmative action programs, one way or the other. This does not mean that the ways in which the equal protection framework seeks to implement the constitutional commitment to equality cannot be improved. The balance that the framework establishes between the judiciary’s institutional role vis--vis equal protection and the anxiety over the counter-majoritarian difficulty may still be profitably revisited.

Such reassessment is occurring in cases brought in state courts, raising issues under the equal protection provisions of state constitutions. In Baker v. State55 and Goodridge v. Department of Public Health,56 plaintiffs challenged the respective prohibitions in Vermont and Massachusetts on same-sex marriage, and the courts of each state took [*PG281]seriously the principle that non-suspect classifications must have a rational basis. The courts concluded, essentially, that prohibitions on same-sex marriage have no rational basis—at least not on the arguments presented to support the restrictions in each case. But the courts also declined to order an immediate judicial remedy—in Baker, the court directed the Vermont legislature to devise a means to provide same-sex couples with the benefits and privileges of marriage,57 and in Goodridge, the court stayed for six months the effect of its decision modifying the common-law understanding of marriage to include same- and opposite-sex couples, in order to give the legislature time to take such action as it deemed appropriate.58

In each case, then, the court exercised more intense review over governmental action than the history of rational basis review under the equal protection framework would suggest is the norm. At the same time, the courts declined to impose judicially-contrived remedies on the people, instead either leaving to the people’s representatives the task of crafting an appropriate remedy in view of the perceived constitutional violation, as in the Vermont case,59 or giving the legislature time to adjust, and perhaps react, to the ruling, as in the Massachusetts case.60 The decisions accordingly distinguish between the resolution of a constitutional dispute, which involves the declaration of constitutional obligations, a task that lies within the court’s competence, and the final articulation of a remedial policy to address the constitutional violation, a task that lies within the legislature’s domain.61 In effect, the courts struck a new balance between the judiciary’s institutional role in policing governmental classifications and the need to avoid overreaching and the attendant charges of judicial activism—a balance that springs from the equal protection framework, to be sure, but one that, at least in respect to rational basis review, pivots on a different jurisprudential axis.

[*PG282] Naturally, this kind of equal protection review raises its own questions. Though the approaches in Baker and Goodridge spurred public discourse by referring the issue of remedy to the political process, they risked delayed enforcement of constitutional mandates by reluctant legislatures. And, it’s unclear whether the more intense rational basis review that these approaches endorse will—or should—have any application in cases that do not affect such immensely personal matters as an individual’s interest in marrying a person of his or her own choosing. At a minimum, though, cases like Baker and Goodridge suggest that, apart from close adherence to the equal protection framework, there are ways in which courts can implement the constitutional commitment to equality that acknowledge both the judiciary’s institutional role and an appropriate understanding of how best to effectuate that role in a constitutional democracy.

?? ?? ?? ??