abstract. The Civil Rights Act was remarkably successful in fighting overt bigotry and discrimination, but much less so in combating the subtler, institutionalized disadvantages that are now the main sources of social injustice. The heroic idea of rights as protections from an oppressive state or oppressive powerful private organizations is misleading and distracts attention from the institutional reforms necessary to achieve real social justice. In fact, the very concept of discrimination is vague and contested—the conflict in contemporary civil rights disputes is not simply over the factual question of whether or not discrimination has occurred, but also over the essentially normative question of what should count as discrimination. The concept of discrimination itself has become a placeholder for ideological struggles over how to balance individual entitlements to fair treatment on the one hand against employer decision-making prerogatives and individual liberties of expression on the other. We should abandon unresolvable conceptual disputes over “discrimination” in favor of a focus on the extent of the employer’s affirmative duty to avoid decisions and policies that needlessly injure members of underrepresented or stigmatized groups.

author. George E. Osborne Professor of Law, Stanford Law School.

Introduction

Anniversaries are times to celebrate past glories, but they are also times to reassess and consider new directions for the future. The fiftieth anniversary of the Civil Rights Act of 1964 offers an opportunity to do both.

When it comes to outright discrimination and overt prejudice, civil rights have been remarkably successful. But today’s most serious social injustices aren’t caused by overt bigotry. For instance, in the context of race, they stem from segregation—a legacy of past racism but not by and large the result of ongoing discrimination—and the many disadvantages that follow from living in isolated, economically depressed, and crime-ridden neighborhoods. Civil rights litigation and activism have hardly made a dent in these formidable obstacles. Civil rights are an important part of many social justice struggles, but they are subject to the law of diminishing returns. Rights can offer limited improvements in a narrow set of circumstances, but the effectiveness of the civil rights approach diminishes and its costs increase as they are applied to more novel, complex, and elusive social problems.

In one sense, Title VII of the Civil Rights Act of 1964 is a prime example of such limitations. The Act relies largely on private litigation to enforce its mandate: the basic structure of anti-discrimination law is modeled on tort law. As a consequence, we have come to think of anti-discrimination law as a question of individual justice and private entitlement. The well-understood upside of this approach is that, at least in theory, every individual can assert his or her own rights without waiting for a cumbersome bureaucracy to implement comprehensive policy reform. But the downside of this approach will be familiar to critics of the tort system: private enforcement is chaotic and inefficient from a public policy perspective. Access to justice is limited by constraints of time, familiarity with the legal system, and resources. Incentives to sue are not closely related to the strength of the plaintiff’s claim or the culpability or social injuriousness of the defendant. As a result, enforcement of the law is spotty and arbitrary: disappointing to employees who often find pressing their rights in court too hard or too uncertain and frustrating to employers who face a constant risk of unexpected lawsuits.

But Title VII contains the seeds of an antidote to these ailments. The Act is a compromise between tort-like private enforcement and comprehensive regulation of the economy in the public interest. The Equal Employment Opportunity Commission (EEOC) represents the public enforcement side of the Act: an administrative agency with the authority to work out detailed rules for the implementation of a broadly defined congressional scheme. Unfortunately, the terms of the compromise created a defanged EEOC and subsequent amendments only slightly augmented the agency’s power: it gained the authority to bring suit directly in 1972,1 but it still cannot issue binding rules or impose fines or orders directly.

As Bruce Ackerman suggests in his fascinating history of the period, the successes of the civil rights movement owe more to the popular branches of government than to the courts.2 Ackerman’s account of the indispensable role of Congress and of President Johnson and the less-well-known role of President Nixon in advancing civil rights can help reframe our thinking about how the law of equality has worked and can work in the future. By placing the often glamorized role of courts in its proper context and by expanding our conception of constitutional history to include the seemingly mundane and often reviled business of administrative regulation, Ackerman’s work calls for a long overdue reassessment of our unfinished struggle for racial justice.

This short essay is a modest attempt to take up that challenge. In Part I, I’ll argue against what I will call the heroic idea of civil rights—the familiar idea that civil rights are inherent in each individual and should be understood standing stalwart and self-sufficient, without reference to collective public policy goals. In its place, I will argue for a more realistic or “disenchanted” idea of rights as a contingent decision to enforce public policy through private action—only a part of a larger approach that includes comprehensive regulation in the public interest.

The disenchanted idea of rights would let us reframe much of anti-discrimination law. In Part II of the essay, I’ll argue that we should think of Title VII not as guaranteeing an individual entitlement against discrimination, but rather as defining an employer’s duty to avoid decisions that cause inequality. I will suggest that this idea is already implicit in much of current anti-discrimination law, but it is at war with other parts of the law that privilege the heroic idea of rights as inalienable individual entitlements. Rethinking anti-discrimination law in terms of an employer’s duty of care could provide more effective deterrence, thereby benefiting more employees while at the same time offering employers clear direction on how to comply with the law.

I. heroic rights

The heroic idea of civil rights is that they protect the individual from a potentially oppressive state. A slight extension of this idea—especially congenial to liberals—is that rights protect individuals from an oppressive state and from oppressive private institutions that are large or influential enough to be “like” the state in some meaningful way. The first iteration of this extension appeared in cases like Marsh v. Alabama,3 which applied constitutional standards to private entities that served a “public function.” Shelley v. Kraemer4 involved a similar extension of constitutional rights to private action. There, the formal holding was that the enforcement of private racial covenants through the courts was a form of state action. But since this principle, taken to its logical conclusion, would transform the terms of any private agreement into state action the moment it required enforcement, it has long been supposed that a similar functional analysis explains the result: racial covenants mimicked the racial zoning invalidated by the courts in 1917’s Buchanan v. Warley.5And the idea also implicitly underwrites the growth of the regulatory state in areas such as employment and labor law and environmental regulation.

But underlying this imprecise functional analysis (no one ever quite defined “public functions” precisely except to say, tautologically, that they are functions typically performed by public entities) was a more coherent and more radical idea, advanced by American legal realists such as Robert Hale6 and Morris Cohen7: because the state enforces property entitlements and contracts, all private action is underwritten by state power, hence the public/private distinction could have no fundamental normative force.

It follows from this insight that rights are not a special protection against power; they are a political decision to assign power to one party or another. Rights are not a limitation on power; they are a way of distributing power and resources. I take this to be one of the implications of Wesley Hohfeld’s famous deconstruction of the concept of “rights,”8 in which he demonstrates that the term “right” may have many different meanings in practice, each of which involves a relationship of entitlement and corresponding obligation between two or more individuals. This also suggests (though this is a larger claim and demands more elaboration than I can provide here) that there is no moral or normative distinction between formal constitutional rights and the entitlements created by statutory law—both simply reflect a collective decision to assign a legal entitlement of some kind to one or another party. This is why, as Ackerman’s account suggests, the entitlements defined in the Civil Rights Act can be considered constitutional rights. I’ll call this the disenchanted idea of rights.9

Here Professor Ackerman’s idea of informal constitutional change offers a sociological and historical account of higher law. While I am less certain than Professor Ackerman that this account sharply distinguishes constitutional law from mundane statutory and common law, I find his nuanced and contextual account of constitutional law more convincing than its originalist or textualist competitors. That said, the disenchanted idea of rights does not require—and perhaps does not allow for—a theory of constitutional legitimacy. There is no notion that the legal entitlements we currently enjoy are the product of natural law, or are inherent in the very idea of democracy, or are necessary to meaningful citizenship or essential to a republican form of government. Rights have no justification other than that they are part of our political and legal tradition and emerge from the political and legal institutions and practices that most citizens accept. Nor do rights guarantee or underwrite the legitimacy of the social or political order; instead rights are a product of the social and political order.

Viewed from one perspective, this conception of rights is radically critical: there is no justification for the political order or the rights that flow from it; all political regimes impose a contingent form of social organization and law’s primary function is to control dissenters, deviants, and discordant elements. Here we might say, along with Michel Foucault: “Humanity does not gradually progress from combat to combat until it arrives at universal reciprocity, where the rule of law finally replaces warfare; humanity installs each of its violences in a system of rules and thus proceeds from domination to domination.”10 Hence rights can be another means of installing violence—a new type of domination in the form of the rule of law.

Viewed in another way, the disenchanted idea of rights has a more conservative, Burkean tenor. Rights are just one way that government does its work. Government is not made legitimate by rights; to the contrary, rights are made legitimate by government, and less directly by custom and tradition. Any government that has cleared the relatively low bar of being marginally preferable to anarchy is, by virtue of that improvement, legitimate; anything better than “tolerable” is icing on the cake. Once we have moved out of the state of nature and accepted the necessity of government, any question of rights is one of custom or convenience:

Government is not made in virtue of natural rights, which may and do exist in total independence of it; and exist in much greater clearness, and in a much greater degree of abstract perfection: but their abstract perfection is their practical defect. By having a right to everything they want everything. . . . Society requires not only that the passions of individuals should be subjected, but that even in the mass and body as well as in the individuals the inclinations of men should frequently be thwarted, their will controlled, and their passions brought into subjection. . . . The moment you abate anything from the full rights of men, each to govern himself, and suffer any artificial positive limitation upon those rights, from that moment the whole organization of government becomes a consideration of convenience.11

Viewed from the Foucauldian perspective, rights are dangerous and potentially oppressive. But short of anarchy, there is no way of avoiding this danger—there is no escape from power. The best one can do is to limit the risks and hope to shape the law to do more good than harm. Viewed from the Burkean perspective, rights can contribute to human happiness provided they are well considered and consistent with national culture, customs, and traditions. But when rights are derived from supposedly universal abstractions or first principles, they will almost certainly be neither well considered as policy nor consistent with tradition and hence will inevitably do more harm than good.

Of course, this does not mean that a Foucauldian leftist and a conservative Burkean would agree about which rights are good and which are dangerous! I offer a radical and a conservative version of disenchantment to demonstrate that the critique of heroic rights does not come with built-in ideological implications. Instead, the critique is designed to reveal the ideological stakes of controversies over rights. Once those stakes are brought into view, the real struggle begins.

But either way, no conclusion about the desirability of rights generally follows from the disenchanted idea of rights. One could be “disenchanted” and still support many rights for instrumental reasons or because they are the outgrowth of tradition and custom. But the disenchanted idea of rights allows for the possibility—indeed the certainty—that some rights may be the enemies of justice, freedom, and equality. It allows for rights-as-villain as well as rights-as-hero. Accordingly, my point here is not to suggest that rights are necessarily bad or to propose we do away with any specific set of legal entitlements. It is only to suggest that the heroic idea of rights is dangerously misleading—and to propose that we replace the heroic idea with the disenchanted one and then make context-specific judgments about when to extend legal entitlements and when to find other means to advance equality.

***

Since the 1960s, the ideas developed during the civil rights movement have dominated American race relations. In important ways, civil rights have been an astonishing success: race discrimination in restaurants, theaters, and hotels was quickly and thoroughly eliminated by the civil rights legislation of the 1960s. Discrimination in employment—while still a problem—has been dramatically reduced and is widely and roundly condemned. Public figures who make overtly bigoted statements typically suffer widespread contempt and often lose their jobs. As a result of these welcome developments, each successive generation is less bigoted than the preceding one.

But today some rights claims are doing more harm than good. Recently, the Fourteenth Amendment guarantee of equal protection has stymied sensible, if controversial, efforts to correct racial inequality, in direct contravention of its historical purpose. Most dramatically, in 2007 the Fourteenth Amendment was used to prevent racial integration in the public schools.12 The Supreme Court has used individual rights to undermine much of the practical work of the Second Reconstruction—from twisting employment discrimination law against itself in Ricci v. DeStefano13 to slowly choking off the life of affirmative action in Fisher v. University of Texas14 to gutting the Voting Rights Act in Shelby County v. Holder15—all in the name of equal rights. Today’s Equal Protection Clause works against equality more often than it furthers it.

It’s tempting to insist that we just need more of the same—that we’ve only been too timid in enforcing civil rights laws or too conservative in interpreting them. But too often rights in their heroic mode are a distraction from the real questions of how power and resources are distributed in our complex technocratic and bureaucratized society.

II. discrimination and duty

The essence of modern civil rights law is that individuals have an entitlement not to be discriminated against for certain forbidden reasons, such as race, sex, religion, etc. It’s tempting to imagine that the concept of discrimination is quite simple and straightforward, and the problem in applying the law lies only in proving when discrimination has occurred. But in fact we lack good and agreed-upon definitions of the relevant prohibited bases of discrimination (for instance, does race denote only inherited characteristics such as skin color and other physical features or does it extend to traits such as culture?). Worse, we lack a clear conception of discrimination itself. It is not obvious, for instance, whether discrimination is a decision made with a specific subjective mental state or whether discriminatory intent is simply evidence of objective disparate treatment. Likewise, it is uncertain whether evidence of an unjustified adverse decision affecting a female employee is relevant to prove a discriminatory motivation or whether it objectively demonstrates sex-dependent decision-making. Similarly, evidence of a statistical racial disparity between an employer’s workforce and the qualified labor pool might suggest either that decisions were made with discriminatory intent or that decisions were objectively race-dependent.

When the law insists that employers or other powerful actors must not “discriminate” on certain enumerated bases, what it often requires in practice is that the entities responsible for the ultimate decision meet a duty of care to avoid unnecessarily perpetuating social segregation or hierarchy. What are presented and fought over as questions of subjective mental state (discriminatory intent) and objective causation (was the disfavored mental state, if present, responsible for the challenged decision?) are, in practical terms, ideological struggles over the appropriate scope of the defendant’s duty of care.

The duty of care created by anti-discrimination law is no different in principle than the many other duties we all have as members of an interdependent society. We have duties to avoid common law nuisances and other torts, which can usefully be translated into comprehensive regulations involving environmental protection, land use planning, workplace safety, and standards of product merchantability. In this sense, racism, sexism, and other types of pervasive prejudice are social problems that we all have a duty to minimize in order to keep society productive and peaceable.

Of course, this is not how courts and litigants describe the controversies, nor do I believe it is how most consciously understand them. But because the explicit terms of controversies are conceptual and indeterminate, they cannot guide any specific resolution: something else must be at work behind the scenes.

The argument that follows is an attempt to excavate the practical stakes underlying anti-discrimination struggles. I will use the doctrine of Title VII of the Civil Rights Act to illustrate this point. I suspect many of the conclusions I draw here are more broadly applicable, but a robust argument to that effect will have to wait for another time and place.

Consider the controversy in 2011’s class action decision Wal-Mart v. Dukes.16Betty Dukes and her co-plaintiffs claimed that Wal-Mart systematically discriminated against women in pay and promotions. They claimed that Wal-Mart’s personnel policies, which gave almost complete discretion to store and district managers and encouraged subjective decision-making based on soft qualifications such as “teamwork” and “integrity,” were especially vulnerable to sex discrimination.17 But vulnerability to sex discrimination is not the same as discrimination itself. In its defense, Wal-Mart pointed out that the Dukes plaintiffs could not point to any specific company-wide discriminatory policy or practice—instead, they cited isolated and “widely divergent” anecdotes and advanced a vague hypothesis of a sexist corporate culture to conjure up the specter of a common pattern of discrimination.18 The absence of any centralized policy of discrimination doomed the class action against Wal-Mart. The Supreme Court found that because there was no common policy or practice of discrimination, there were no common issues of law and fact to justify class certification.19

Despite their references to the sexist corporate culture at Wal-Mart, the crux of the Wal-Mart plaintiffs’ claim wasn’t really that Wal-Mart, as a corporation,had encouraged sex discrimination. It was that Wal-Mart hadn’t taken sufficient care to prevent it. Wal-Mart’s policies were “vulnerable” to sex discrimination by individual managers, but Wal-Mart did nothing to change the policies to reduce the risk of discrimination. Why not? Did Wal-Mart’s management secretly want its managers to discriminate? There’s little evidence of such a motivation, and what’s more, there are obvious business justifications for Wal-Mart’s policies: in a service industry, subjective factors are relevant to job performance, but information about varying local conditions in such a large enterprise is costly to obtain and evaluate centrally. Decentralized decision-making is an efficient way of organizing personnel decisions; to be sure, there will be mistakes, local prejudices, and rogue managers who act on the basis of whim or bias, but—purely as a business matter—these costs are probably outweighed by the benefits and savings of a decentralized and discretionary system. And there is the added benefit that a decentralized system potentially contains any liability for unlawful practices to the level of the individual store: if there is no centralized policy or decision-making apparatus, there can be no company-wide liability. There was a conflict between protecting women from sex discrimination and Wal-Mart’s preferred personnel policies, which may well have been desirable for other reasons such as cost or ease of administration. Wal-Mart chose to retain the risky policies. That is the only common policy that joined the disparate incidents of sex discrimination that the Dukes lawsuit sought to litigate together. The common policy was one of nonfeasance or negligence: a failure to take due care.

Now, let’s consider the problem of “mixed motives” and the attempt of the courts to resolve it with an inquiry into whether the prohibited grounds “caused” the challenged employment action, squarely addressed by the Supreme Court in Price Waterhouse v. Hopkins.20

Ann Hopkins sued her employer for sex discrimination after being passed over for partnership. The evidence presented at trial showed that Hopkins was passed over because of shortcomings in her “interpersonal skills”: she was “overly aggressive, unduly harsh, difficult to work with and impatient with staff.”21 But it was also clear that sex played some role in Hopkins’s failed bid for partnership: some partners complained that she “overcompensated for being a woman,”22 several comments mentioned her sex in contexts both favorable and detrimental to Hopkins, and she was advised that in order to improve her chances for promotion she should “walk more femininely, talk more femininely, dress more femininely, wear make-up, have her hair styled and wear jewelry.”23

In Price Waterhouse, the pivotal question was not the existence of a discriminatory motive, but the question of causation: did the prohibited motivation “cause” the challenged action? According to the plurality opinion in Price Waterhouse:

Title VII meant to condemn even those decisions based on a mixture of legitimate and illegitimate consideration. When, therefore, an employer considers both gender and legitimate factors at the time of making a decision, that decision was “because of” sex and the other legitimate considerations—even if we may say later . . . that the decision would have been the same if gender had not been taken into account.24

But this definition of “because of” does not really involve causation at all. Instead, it prohibits a state of mind at a critical juncture. The plurality’s attempt at clarification in this respect is less than edifying:

To attribute this meaning to the words “because of” does not, as the dissent asserts, divest them of causal significance. A simple example illustrates the point. Suppose two physical forces act upon and move an object, and suppose that either force acting alone would have moved the object. As the dissent would have it, neither physical force was a “cause” of the motion unless we can show that but for one or both of them, the object would not have moved. . . . Events that are causally overdetermined, in other words, may not have any “cause” at all. This cannot be so.25

But the problem of simultaneous causes does not justify the plurality’s approach to the problem of mixed motives. The plurality would find liability whenever “an employer considers . . . gender . . . at the time of making a decision,”26 whether or not the legitimate considerations alone would have resulted in the decision and whether or not the prohibited consideration would have resulted in the decision absent legitimate considerations. This is not analogous to two forces, each of which would independently have moved an object. Instead, here the analogy would be to say that if a force, in and of itself too weak to move the object, acts in tandem with a force, strong enough to move the object by itself, the weaker force nevertheless “caused” it to move.

The plurality’s rule of decision dispenses with any notion of causation. Instead the rule is simply that any consideration of the forbidden ground at the time of the challenged decision triggers liability, regardless of whether or not the forbidden ground actually “caused” the decision. In short, the plurality’s rule requires employers to purify their personnel decisions of the forbidden considerations. This effectively imposes an affirmative duty of care.

Justice O’Connor insisted in her concurrence that a more rigorous requirement of causation is necessary to prevent Title VII from becoming a “thought control” law:

[When it passed Title VII,] Congress was attempting to eradicate discriminatory actions in the employment setting, not mere discriminatory thoughts. Critics of the bill that became Title VII labeled it a “thought control bill” and argued that it created a “punishable crime that does not require an illegal external act as a basis for judgment.”

. . . .

. . . Congress clearly conditioned legal liability on a determination that the consideration of an illegitimate factor caused a tangible employment injury of some kind.27

O’Connor’s own opinion suggested that the employer is liable if the prohibited ground played a “substantial role” in the decision-making process.28 O’Connor’s opinion differs from that of the plurality in its focus on “direct evidence” of sexism in the decision-making process. O’Connor insisted that the plaintiff must demonstrate by “direct evidence that an illegitimate criterion was a substantial factor” in the challenged decision.29 She insisted that “[neither] stray remarks in the workplace, . . . [n]or . . . statements by nondecisionmakers, [n]or statements by decisionmakers unrelated to the decisional process itself, suffice to satisfy the plaintiff’s burden.”30

O’Connor’s analysis in Price Waterhouse would make employers liable for prejudiced or stereotyping attitudes related to the challenged action but only when they were part of a formal or discrete decision-making process, as distinguished from more general attitudes and statements made in the workplace. This limitation would prevent Title VII from becoming a “thought control” law because employers would not be liable simply because their employees made sexist remarks; they would be liable only when employees with authority over the challenged decision made sexist remarks in direct relation to that decision. She used the image of the boardroom to illustrate her legal standard and to capture this idea of a discrete zone of decision-making:

It is as if Ann Hopkins were sitting in the hall outside the room where partnership decisions were being made. As the partners filed in to consider her candidacy, she heard several of them make sexist remarks in discussing her suitability for partnership. As the decisionmakers exited the room, she was told by one of those privy to the decisionmaking process that her gender was a major reason for the rejection of her partnership bid.31

For O’Connor, Title VII does not require employers to police their employees’ thoughts and expression generally, but it does require employers to keep sexism out of the figurative boardroom.

But this rule isn’t consistent with O’Connor’s concern with causation. The potential for sexism to cause adverse employment decisions cannot be so easily contained. Indeed, the decision challenged in Price Waterhouse was not made in such a closed-off room—the Price Waterhouse promotions process was a complex event that took place in multiple locations and involved several stages. O’Connor’s limitation does not change the fact that the prohibited motive may still have “caused” the decision. Ann Hopkins—like many employees—may well have been injured, not in any formal decision-making process, but long before a formal review or promotion occurred. Many Title VII plaintiffs suffer because of their sex at the hands of supervisors who give them “grunt work” instead of challenging assignments that offer the chance to impress, or by subtle comments and insinuations that will harm their reputations. This type of discrimination, much more so than overt bigotry in a formal process, is probably the most pervasive impediment to true equality of opportunity for women and people of color. Sexism quite removed from the formal decision-making process can have a powerful, if indirect, effect. Surely the private biases of influential nondecisionmakers, expressed in informal settings, can influence the decision-making process, even to the extent of “causing” adverse employment decisions. Why shouldn’t it be a violation of Title VII whenever racism or sexism are in the air in sufficient quantities to pollute the workplace atmosphere, potentially diminishing opportunities and creating a toxic environment for members of the potentially disadvantaged groups?

The answer is that it’s hard for employers to control bias in the air. Limiting liability to cases that involve prejudiced statements in a formal decision-making process does not impose liability in all cases where the discrimination “caused” the adverse decision. But it does define a discrete “danger zone” in which employers are on notice that they must aggressively police and counter discriminatory statements. In other words, it allows us to establish the scope of the employer’s duty of care. The figure of the boardroom in which employment decisions are made is important because it reflects the idea of a well-defined decision-making process for which an employer has a special legal responsibility.

The idea here is that employers should be liable only for discrimination they can prevent as institutions without overly draconian policing of the expressions of their employees (“thought control”). They should be liable only when they—as entities—could have prevented the discrimination from occurring. Hence, stray comments and sexism or racism in the air are not actionable, whereas prejudiced and stereotyping comments in a formal decision-making process are, even if they don’t actually affect the decision.

Here, Title VII does not even aspire to eliminate all intentional discrimination on the basis of race and sex. At most, it will regulate certain formal proceedings that are within the direct control of upper management and are fairly neatly sealed off from the more unpredictable give and take of the workaday world. The prohibition against intentional discrimination doesn’t offer a remedy whenever an individual is treated differently because of race or sex; instead, it seeks to limit the introduction of patently prejudicial assertions and bigoted stereotypes into a discrete formal decision-making process. For Justice O’Connor, “causation” is little more than shorthand for a policy analysis that balances the goal of reducing illegitimate workplace segregation and hierarchy with legitimate employer prerogatives, assigning responsibility to employers only in those circumstances that they can control at an acceptable cost.

Compare this approach with the Court’s opinion in Burlington Industries, Inc. v. Ellerth.32 In Burlington Industries, a male supervisor allegedly harassed a female subordinate, who eventually quit in response. She never informed senior management about the pattern of harassment but later sued her former employer for sex discrimination. The trial court found that the plaintiff had suffered severe harassment at the hands of the supervisor, but granted summary judgment for the defendant because the employer neither knew nor should have known about the harassment. Justice Kennedy, writing for the Court, noted that “Title VII is designed to encourage the creation of antiharassment policies and effective grievance mechanisms . . . [and] borrows from tort law the avoidable consequences doctrine . . . .”33 In order to encourage both employers and employees to take reasonable steps to avoid ongoing harassment, the Court held that when harassment does not take the form of a tangible employment action (such as firing, demotion, or undesirable transfer) an employer may

raise an affirmative defense to liability [which] comprises two necessary elements: (a) that the employer exercised reasonable care to prevent and correct promptly any sexually harassing behavior, and (b) that the plaintiff employee unreasonably failed to take advantage of any preventative or corrective opportunities provided by the employer or to avoid harm otherwise.34

Of course an individual may well be a victim of harassment even when the employer has a reasonable anti-harassment policy. In practice, Burlington Industries makes the employee’s recovery depend on whether she brought the problem to the attention of senior management—not on whether she in fact suffered discriminatory harassment. This is a scandal if Title VII liability is supposed to protect individuals from a discrete injury. From a more pragmatic perspective, what is important is that the law will reduce unjustified decisions affecting vulnerable groups in the run of cases, breaking down patterns of segregation and hierarchy.

As far as individual justice is concerned, the specific injury suffered by an individual treated adversely in violation of anti-discrimination law is the same injury anyone who faces the same type of unjustified adverse decision (rejection of a job application, termination, demotion, failure to receive a promotion, etc.) suffers. For the most part, we consider such disappointments to be among the costs of living in a free society: employers, like other private actors, are free to make bad decisions, and our solace is simply that they will have to suffer the consequences of their bad judgment along with those they unwisely decline to hire, retain, or promote.

The reason to prohibit a small group of unjustified adverse employment actions is not the nature of the specific individual injury, but that we expect members of some groups to suffer a disproportionate incidence of bad-cause adverse employment decisions in the absence of extraordinary intervention. That the challenged decision is made “because of” race or sex is relevant only because it signals that the target of the decision in question is likely to suffer directly—and indirectly as a member of an interdependent group—from a disproportionate number of adverse decisions made for bad reasons. This cumulative effect would be undone if the incidence of bad-cause adverse decisions were equal across social groups, even if some of the tolerated residuum of bad-cause decisions involved race or sex.35 The goal of anti-discrimination law, then, should be to make the predicted frequency of unjustified decisions roughly equal for all groups in society—not necessarily to eliminate them altogether.

One might object that this ignores the “stigma” and psychological injury associated with each individual case of discrimination. But only the relatively rare cases of overt classification based on race, unambiguous animus, or stereotyping are inherently stigmatizing. Today, most discriminatory decisions are ambiguous: the central problem involves determining whether the decision involved bigotry or not. If the decision is ambiguous, the stigma should be correspondingly weak. What strengthens and clarifies the stigmatizing effect of ambiguously “discriminatory” decisions is their frequency and pervasiveness. I may be in doubt as to whether race is the reason the first or second taxicab I hail passes me for another fare down the block, but after the sixth or seventh empty cab drives by, my doubt will be replaced by anger and shame. Consider Cornel West’s account:

I had an hour until my next engagement. . . . I waited and waited and waited. After the ninth taxi refused me, my blood began to boil. The tenth taxi refused me and stopped for a kind, well-dressed, smiling female fellow citizen of European descent. As she stepped in the cab, she said, “This is really ridiculous, is it not?”36

Taxi drivers refuse passengers for a host of reasons, of which race is only one. Black people will typically suffer mysterious rejections more frequently than whites because they will suffer the garden-variety rejections and the racial ones. But if the incidence of mysterious rejections were the same for people of all races, that stigma should be eliminated (with any remaining stigma the result of correctable misperception).

The policy goals underlying anti-discrimination law must compete with other potentially inconsistent policy goals. O’Connor’s “causation” analysis in Price Waterhouse is an attempt to balance anti-discrimination goals with both employment at will and the expressive liberties of other employees (hence the concern about “thought control”). The relevant question is whether this duty of care sufficiently deters adverse decisions affecting women. I think that Justice O’Connor’s analysis in Price Waterhouse too heavily favors expressive liberties and employer prerogatives over egalitarian goals, but this has nothing to do with whether Price Waterhouse rejected Ann Hopkins’s partnership application because of her sex; instead, it has to do with the relative weight Price Waterhouse’s promotional process accorded to gender equity as against the liberties of contract and expression.

So far, this essay has been descriptive; my claim has been that much of Title VII doctrine can be explained as a somewhat confused attempt to define the scope of an employer’s duty of care to avoid perpetuating unjustified social inequalities. The attempt is confused because the practical stakes of the disputes are not addressed openly—instead, they are addressed in code, in the language of conceptually ambiguous or practically meaningless terms such as “intent” and “causation.”

A prescription follows pretty easily from this diagnosis: the law should address the real stakes openly. Instead of an intractable controversy over intent and causation, the law should define an employer’s duty of care. An employer who fails to meet the duty should be punished in some way—either with liability for individual cases of unfair adverse decisions or perhaps through a more comprehensive administrative system of penalties. Conversely, an employer who meets the duty should not face liability for otherwise lawful adverse decisions.

I’m proposing what may seem to be a radical idea: that we put aside—if not abandon—the idea that every victim of discrimination is entitled to his or her day in court. This may seem unfair if you think that individuals have a right to equal treatment that no collective goal can outweigh. But the American legal system has never recognized such a broad and sweeping guarantee of individual fairness. In the United States, most employment is “at will”: employers can choose whom to hire, fire, and promote for any reason that isn’t explicitly forbidden by law. A lot of bad reasons—nepotism, favoritism, and idiosyncratic prejudice (imagine an employer who hates redheads or people with bad skin)—are legally acceptable. Employment at will reflects a sensible modesty about the capacity of government to identify and prevent or remedy unfairness: it can be very hard to tell the difference between unfair treatment and the kinds of difficult judgment calls that every employer must make. And if the labor market is working well, a good employee who is treated unfairly will be able to find employment elsewhere. In fact, many people who could sue their employers for discrimination prefer to cut their losses and find a new job.

And let’s face it—practically speaking, our current system of individual rights doesn’t provide every victim of discrimination a remedy either. Lots of people are victims of discrimination but don’t know it or can’t prove it. Plenty of others find that the costs of pursuing their rights—time and expense spent in litigation and a possible tarnished reputation in one’s profession—outweigh the benefits.

Individually driven civil rights inadvertently encourage weaker claims over stronger ones in other ways. For instance, people are much more likely to sue when they lose a job or are denied an expected promotion than when they are not hired in the first place. As a result, complaints of termination now outnumber failure to hire complaints by about six to one.37 Such lopsided enforcement misdirects enforcement resources, and creates perverse incentives. A well-crafted civil rights policy would reward employers who hire members of underrepresented groups and punish those who do not. But right now civil rights litigation does just the opposite: it makes it relatively safe to refuse to hire members of underrepresented groups (because there is little risk of being sued for failure to hire) and risky to hire them (because doing so opens one up to a much more likely lawsuit for discriminatory termination). As my Stanford colleague economist John Donohue has argued:

A worker who is not hired in the first place is obviously in no position to bring a future firing suit. . . . With the enormous increase in discharge cases, the probability that a worker will bring a discriminatory firing suit is now substantially higher than the probability that a worker will bring a failure to hire suit. Consequently, anti-discrimination laws may actually provide employers a (small) net disincentive to hire women and minorities.38

A requirement that employees meet a clear duty of care—enforced by liability and/or fines for those who fail to comply and encouraged by immunity from liability for those who do comply—would offer comprehensive improvements for those who currently cannot or do not have incentives to bring suit.

I can only sketch the outlines of this duty in this brief essay, but cases like Burlington Industries might provide a model. The Court in Burlington Industries drew on a practice that many employers were already using to define a legal standard for due care. By adopting the best practices of employers as guide, the Court was able to take advantage of real-world experience: employers would not voluntarily adopt a practice that is overly burdensome, nor would they likely continue to employ a litigation-avoidance policy that did not actually reduce the occurrence of legal violations. Employer practices are a good place to start in defining a duty of care, but they aren’t always where we should end: courts and administrative agencies such as the EEOC should study such practices to ensure that they are effective and should adjust them to strike the right balance between workplace equity and employer prerogatives.

Statistical measures offer another possible way of defining an employer’s duty of care. We might presume that an employer that has a workforce that reflects the demographics of the relevant pool of qualified potential employees has met its duty to avoid bias. One might object that such a standard would encourage quota hiring, but of course any statistical measure of equity has that potential. This alone is no more an objection to the use of statistics in defining a duty of care than it is in the context of statistical evidence of systemic disparate treatment or disparate impact.

Here I would return to Bruce Ackerman’s account of the civil rights struggle and his insightful emphasis on the centrality of bureaucracy or “government by the numbers.” The Second Reconstruction put the techniques of the New Deal administrative state at the service of racial justice. This was (and is) a perfectly sensible way to proceed. Indeed, it is indispensible if one admits that the social commitment to racial justice must be institutionalized in a form that is administrable by the bureaucracies that must actually implement public policy. The unrealistic heroic conception of rights makes this necessity a fatal fault: statistical approaches are condemned because they do not protect each and every individual victim of discrimination (and because they only protect individual victims of discrimination). But again, this is true of any comprehensive administrative rule. If we take the ultimate policy goal seriously, some reliance on statistical comparisons is unavoidable because, in many circumstances, statistics are the most reliable evidence of biased, unfair, or inequitable practices. An appropriately designed statistical standard for employer duty could be nuanced and flexible enough to prevent it from becoming a de facto quota, while still achieving the goal of broad improvement rather than individually tailored remedies.

Conclusion

To be effective, anti-discrimination law must do more than eliminate unambiguously evil or irrational practices—it must also curtail many arguably legitimate practices. The image of discrimination as a discrete evil or mistake obscures this necessity and hence leads to unrealistic aspirations and expectations. It leads the left to insist that simple fairness and justice justify the most assertive and ambitious egalitarian projects and inspires a profound sense of betrayal and frustration when the courts and popular branches of government are unwilling to go along. It leads conservatives to insist that only unambiguous bigotry justifies any corrective legal intervention and inspires self-righteous opposition to even modest egalitarian policies. Against both of these, I’ve tried to suggest a more pragmatic way of thinking about anti-discrimination law, which might better advance egalitarian goals while still protecting legitimate employer prerogatives and expressive liberty.