The Contested Role of Time in Equal Protection

The Contested Role of Time in Equal Protection

This Essay examines how the Supreme Court has used conceptions of time and the passing of time to narrow the definition of racial discrimination and, ultimately, to constrain the very meaning of equal protection. The Essay challenges the common notion in equal protection that as time passes, discrimination and its harmful effects dissipate and eventually expire. Based largely on this notion, courts set artificial time horizons for identifying the continuing vestiges of past discrimination, which in turn rationalizes persisting inequality in the present. Using social science, the Essay explains how social conceptions of time discourage deeper inquiry into the relationship between past discrimination and present systemic inequality. It proposes that equal protection reject notions that discrimination naturally subsides so that courts and other constitutional actors may more freely explore these connections between past and present.


    1. Using Time to Erase the Past
    2. The Social Science of Time
    1. The Consequences of Time in School Desegregation
      1. Modes of Time Argumentation
      2. The Evolution of Intent Post-Brown
      3. “Remoteness in Time” and Systemic Harm
    2. Time and Discriminatory Motive




A Symposium on the life and legacy of Constance Baker Motley would not be complete without recognizing the instrumental role that her work at the NAACP Legal Defense and Educational Fund, Inc. (LDF) played in the evolution of civil rights and law’s eventual embrace of equality. 1 See Constance Baker Motley, Equal Justice Under Law: An Autobiography 61–86 (1998) (describing Motley’s work on Brown v. Board of Education). As Motley herself understood, however, the Supreme Court has not been kind to the body of law that she and her LDF colleagues con­ceived as a tool of freedom for African Americans and other margin­al­ized groups. In the decades since, the Court has defined equal protec­tion in ways that have undermined opportunities for redressing systemic wrongs. 2 See generally id. at 230–31 (describing the “derailment” of voluntary affirmative action—“the twentieth century’s most effective engine of change”—and analogizing this development to “separate but equal” in Plessy v. Ferguson).

This Essay argues that one way the Court has limited equal pro­tection is through interpretations that obscure the relationship between present-day racial inequality and past discrimination. It contributes novel insights about the Court’s use of time as a tool for narrowing equal protection’s meaning and scope. The Court achieves this outcome by advancing assumptions that the effects of prior discrimination expire, such that current inequality bears no cognizable relationship to discrim­ination from years past. Under the cover of these assumptions, courts set artificial time horizons for identifying the vestiges of past discrimination. 3 See Elise C. Boddie, Adaptive Discrimination, 94 N.C. L. Rev. 1235, 1266 (2016) [hereinafter Boddie, Adaptive Discrimination] (explaining that favoring simplistic solu­tions to issues of past discrimination results in “overlook[ing] the problem of time lags that inhere in adaptive discrimination due to active and passive resistance to anti­discrim­ination mandates”). Further, be­cause they imagine that these vestiges weaken naturally over time, courts discount the legitimacy of historical inquiry as a tool for identifying the connective tissue between present forms of inequality and prior discrim­ination. 4 Id. at 1267. As a result, current racial disparities enjoy a pre­sump­tion that they are unrelated to racial discrimination from long ago.

The Court’s thin conceptions of racial discrimination and how it op­erates help to account for these constraints on time, but the constraints themselves also play a circular function by entrenching these conceptions in doctrine. Courts presume that discrimination is linear and static—that it is attributable to specific “bad” acts with effects that can be dis­cerned only within discrete periods of time. They then deploy these time-bound notions to service narrow definitions of discrimination, enabling them to overlook the systemic, dynamic—and most importantly—chronic aspects of discrimination that spread racial harm across multiple do­mains. 5 Id. at 1239. By blunt­ing deeper inquiry into the continuing effects of past racial discrimination, courts justify shallow interpretations of that very discrimination. Time in this respect plays a role that is anal­og­ous to colorblindness doc­trine and the anticlassification rules in equal protec­tion that preclude broader exploration of the subordinating impact of gov­ernment policies and practices.

In a previous article, I introduced a theory of adaptive discrimina­tion—that racial discrimination adapts to law and to social norms as state actors reconstitute discriminatory behavior in forms that are legally sanc­tioned and socially acceptable. 6 Id. at 1248–63. Because discrimination is persistent, ra­ther than episodic, the theory of adaptive discrimination rejects no­tions in equal protection that the effects of discrimination invariably diminish with time. 7 Id. at 1266. The Court, however, conceives discrimination under equal protection as an anomaly—a stone thrown into a pond whose rip­ples will eventually subside, like a tort or breach of contract. Yet, discrim­ination is not static in the same way as other kinds of civil wrongs. It is part of a system that can maintain and rebuild itself, compounding and regenerat­ing the effects of original discrimination. Although we have achieved significant progress, that progress itself can evoke responses that recreate the old system in a new form. 8 I thank my colleague John Leubsdorf for this succinct formulation of the problem.

Arguments about the significance of time at one point played a particularly prominent role in equal protection through school deseg­rega­tion litigation. Following Brown v. Board of Education, the Supreme Court and lower courts used time considerations in school desegregation cases as a vehicle for advancing—and then resisting—broad conceptions of racial discrimination that did not require specific showings of discrim­inatory intent. 9 See infra section I.B. This Essay points to Columbus Board of Education v. Penick 10 439 U.S. 1348 (1978). as an important transitional moment in which the Court embraced the use of broad temporal inquiry to evaluate the constitutionality of existing school segregation. As the composition of the Court changed after Penick, however, time analysis assumed a different role by cementing equal protection’s turn toward animus-based intent and its related em­brace of colorblindness. Under the auspices of presumptions that the ef­fects of discrimination expire, the Court used the passage of time to con­clude that past discrimination was unrelated to ongoing segre­gation. 11 See Freeman v. Pitts, 503 U.S. 467, 498 (1992) (“A history of good-faith com­pli­ance is evidence that any current racial imbalance is not the product of a new de jure viola­tion, and enables the district court to accept the school board’s representation that it . . . will not suffer intentional discrimination in the future.”); Bd. of Educ. v. Dowell, 498 U.S. 237, 247–48 (1991) (emphasizing the “transition” period called for by Brown and Green and declaring that injunctions implemented for antidiscrimination policies were “not intended to operate in perpetuity”). Erasing the past from constitutional analysis also allowed the Court to ingrain formalistic interpretations of equal protection that ignored the broader social context for school segregation and the role that other forms of intent-based racial discrimination, in housing for example, played in perpetuating it. Emphasis on the “temporary” role of court ov­ersight and the use of narrow bands of time for evaluating intent made it easier in turn to justify the release of segregated school districts from ju­di­cial supervision. 12 See Dowell, 498 U.S. at 247 (“From the very first, federal supervision of local school systems was intended as a temporary measure to remedy past discrimination.”); see also Freeman, 503 U.S. at 498 (“We stated in Dowell that the good-faith compliance of the dis­trict with the court order over a reasonable period of time is a factor to be considered in deciding whether or not jurisdiction could be relinquished.”).

This use of time has been consequential. The Court has relied on the passage of time to absolve state actors of responsibility for their predecessors’ prior discrimination. Under this analysis, discrimination must not only be intentional; it must also be current. With litigants un­able to lay claim to the past, courts ignore discrimination that manifests over longer periods, disregarding policies and practices that entrench that discrimination’s earlier effects. These time considerations amplify the problems with intent doctrine by further disabling the legitimacy of systemic inequality as a focus of constitutional inquiry. 13 Allowing broader temporal inquiries into past discrimination could also limit state action doctrine by enabling courts to attribute current inequality to prior discrim­ination by governmental actors. This would usefully expand equal protection’s reach. See Charles L. Black, Jr., The Supreme Court, 1966 Term—Foreword: “State Action,” Equal Protec­tion, and California’s Prop­osition 14, 81 Harv. L. Rev. 69, 70 (1967) (“The amenability of racial injustice to national legal correction is inversely proportional to the durability and scope of the state action ‘doc­trine,’ and of the ways of thinking to which it is linked.”). For these rea­sons, considering the role that time plays in limiting equal protection’s scope adds new dimensions to existing debates about the propriety of in­tent as a measure of discrimination and about how to define and identify intent itself.

Section I.A explains how time is used to rationalize inequality by erasing the past. Section I.B explores social science to explain how social conceptions of time incentivize limitations on broader temporal inquiry. Section II.A examines the contest for the past through the lens of school desegregation cases. Section II.B discusses an influential article by Paul Brest that helped to embed time-bound notions of intent in equal pro­tection. Part III describes the continuing influence of time in equal pro­tection doctrine and other areas of antidiscrimination law. The Essay concludes by examining some of the practical challenges associated with more robust temporal analysis.

I. How Time Rationalizes Inequality

A. Using Time to Erase the Past

Law that seeks to improve the status of African Americans has long been contested. 14 See Kimberlé Williams Crenshaw, Race, Reform, and Retrenchment: Trans­form­ation and Legitimation in Antidiscrimination Law, 101 Harv. L. Rev. 1331, 1376–79 (1988) (describing the role of race consciousness in subordinating the status of African Amer­icans throughout American history); Reva Siegel, Why Equal Protection No Longer Pro­tects: The Evolving Forms of Status-Enforcing State Action, 49 Stan. L. Rev. 1111, 1111, 1119–30 (1997) (describing the “rules and reasons the legal system employs to enforce status relations as they are contested” in the area of race); see also Boddie, Adaptive Dis­crimination, supra note 3, at 1244–45 (arguing that the Supreme Court has relied upon the assumption “that discrimination ceases with the passage of time” to justify terminating judicial and legislative remedies to discrimination). Progress in civil rights is followed by retrenchment. 15 See Boddie, Adaptive Discrimination, supra note 3, at 1249–57 (describing how discrimination is reconstituted to evade and avoid law and to accommodate social norms). Social practices produce new racial hierarchies that are then rationalized by law, and the cycle starts anew. 16 See Crenshaw, supra note 14, at 1376–79 (“Racial hierarchy cannot be cured by the move to facial race-neutrality in the laws that structure the economic, political, and social lives of Black people.”); Siegel, supra note 14, at 1146 (“The body of constitutional law that disestablished slavery had to define the practice it was repudiating, and, as it did so, it simultaneously legitimated new forms of state action that perpetuated the racial strat­ification of American society.”). This process of progress, contestation, and retrenchment leads to what Professors Reva Siegel and Kimberlé Williams Crenshaw have variously described as the “preservation” of racial subor­dination through its “transformation” 17 See Crenshaw, supra note 14, at 1376–79; Siegel, supra note 14, at 1113. and to what I have described as “adaptive discrimination.” 18 See Boddie, Adaptive Discrimination, supra note 3, at 1239–40. As racial discrimination adapts, the legal sys­tem adapts with it, creating different rules and adopting, in Siegel’s terms, “justificatory rhetoric” that rationalizes the new racial order. 19 See Siegel, supra note 14, at 1113. We can also think of this time rhetoric as a type of “subconstitutional rule” that reinforces racial hierarchy. See Michael J. Klar­man, From Jim Crow to Civil Rights 457 (2004) (discussing the use of “subconstitutional rules” that limit constitutional equality to form alone, making formal legal status “vul­nerable to nullification by determined resistance”). Rac­ial inequality persists through the mutability of discrimination 20 See Boddie, Adaptive Discrimination, supra note 3, at 1239. and law’s willingness to accommodate its shifting forms. 21 As Professor Siegel has extensively argued, this dynamic is not just peculiar to racial discrimination. See, e.g., Siegel, supra note 14, at 1116–19 (discussing how “the effort to disestablish the common law of marital status” produced changes that “perpetuate[d] inequalities”). Consequently, law never fully invests in the dismantling of racial subordination but instead allows it to be recycled under different guises. 22 See Boddie, Adaptive Discrimination, supra note 3, at 1248–49.

Time considerations play a central role in rationalizing the process of adaptive discrimination. Courts invoke the remoteness of the original racial injury to justify conclusions that ongoing racial inequality is unre­lated to prior discriminatory intent. 23 See, e.g., Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 255 (S.D. Ohio 1977) (comparing defendants’ contentions that “intervening decades of racially neutral board policies” remove any element of illegality to plaintiffs’ contentions that certain actions and omissions allow for the inference of segregative intent), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979). This process further enables linear, static conceptions of discrimination—which lay blame for racial harm on a specific actor—to predominate in equal protection and to disregard existing forms of inequality that originated in a prior period. The pre­sumption in equal protection that the effects of discrimination “ex­pire,” therefore, both reflects and reinforces its normative focus on racial in­tent as the only recognized source of constitutional harm. The preoccu­pation with intent in turn allows law to ignore how discrimination is recycled through time. In this mode, current racial inequality has no ves­tigial, connective relationship to prior intent; 24 See infra Part II. it is written on an entirely clean slate. For purposes of constitutional analysis, it is as if that discrim­ination never existed, as if the past has been erased. 25 Cf. Amy Kapczynski, Historicism, Progress, and the Redemptive Constitution, 26 Cardozo L. Rev. 1041, 1044 (2005) (advocating for a conception of history as an “eternal present” in which “historical acts . . . coalesce now and then to render visible historical forces and contemporary dangers and possibilities”).

Abandoning the past limits our possibilities for diagnosing and treat­ing racial discrimination in the present. Because historical inquiry allows constitutional actors to detect patterns of social behavior across gener­ations, it provides a window into understanding whether racial problems in the present are isolated or reflect a deeper societal condition that requires more robust responses and corrective action. The courts should examine historical context for the same reason that one studies history generally: Understanding where the country has been—its mistakes and failures through time—is a template for identifying social tendencies that may be passed from one generation to the next. Awareness of this history allows constitutional actors to guard against the same kinds of future wrongs. As the old adage goes, “Those who ignore history are doomed to repeat it.”

Conditions of the present are forged in the past; therefore, we lose too much when we leave it behind. 26 See id. at 1086 (“Addressing ourselves to the facticity of the past . . . can help us disrupt narratives of heritage and progress that blind us to possibilities past and present.”). In her discussions of theorist Walter Benjamin’s concept of redemption, Professor Amy Kapcyznski writes that connecting ourselves to the past by drawing insights from history facili­tates actualization in the present. 27 Id. at 1113 (describing redemptivism as a process of “actualization” that “mea­sures its success . . . according to the degree of insight and kind of action in the present that the historical image enables”). When one confronts a problem, for exam­ple, a common tendency is to determine what happened that led to that problem—the idea being that understanding how a problem un­folded provides clues for how to solve it. Similarly, if law bars con­stitutional actors from taking account of the forces of past inequality, particularly those with a long historical trajectory, it precludes a deeper appreciation of discrimination’s depths and durability. 28 See id. (“[A] progressive who views the legacy of slavery and Jim Crow as truly surpassed must believe that only the elimination of de jure and intentional discrim­ination was required to overcome this legacy. He or she must write off the manifold ways that racial hierarchy continues to reproduce itself.”).

I am cognizant that the use of historical inquiry as a tool for diagnos­ing the effects of prior discrimination and assigning liability presents prac­tical challenges. Some of these challenges are addressed in the con­clusion of this Essay. Section I.B below focuses on one in particular: social pressures to forget the past. 29 See Owen M. Fiss, Gaston County v. United States: Fruition of the Freezing Prin­ciple, 1969 Sup. Ct. Rev. at 379, 433 (“Part of the pressure in society to forget the past no doubt generates from either those who discriminated or those who have little or no­thing to gain from the correction of past discrimination.”). These pressures are rooted in racial­ized conceptions of self-interest as well as in assumptions that past wrongs cease to be a basis for remedy as they recede in time. 30 See id.

B. The Social Science of Time

Section I.A discussed the presumption in equal protection that the effects of discrimination expire and how this presumption problematizes the relationship between present inequality and the past. This section de­scribes a different aspect of the problem: a social understanding that policies created to address the longstanding effects of discrimination are outdated and should be abandoned. 31 See Boddie, Adaptive Discrimination, supra note 3, at 1300–01 (describing Jus­tice O’Connor’s intimation in Grutter v. Bollinger “that affirmative action may no longer be necessary ‘in 25 years’”). For example, the familiar maxim that “time heals all wounds” plays to social intuitions that the passage of time helps to mend prior transgressions. 32 See, e.g., Johan C. Karremans & Paul A.M. Van Lange, Forgiveness in Personal Relationships: Its Malleability and Powerful Consequences, 19 Eur. Rev. Soc. Psychol. 202, 215–16 (2008) (citing research suggesting that forgiveness increases as time passes); Michael J.A. Wohl & April L. McGrath, The Perception of Time Heals All Wounds: Tem­poral Distance Affects Willingness to Forgive Following an Interpersonal Transgression, 33 Per­sonality & Soc. Psychol. Bull. 1023, 1034 (2007) (observing that “[t]he perception of time may be central to healing the wounds of the past”).

Social science research, however, suggests that the social dynamics of time are more complex—specifically, that the healing function of time depends on emotional distance from prior events. This means that an individual’s sense of time is different from how much time has actually passed. The significance that people place on time, moreover, is a func­tion of the baseline event’s relative importance. It may seem like “just yesterday,” for example, that one’s child was born or that one graduated from high school, even though both events objectively happened some time ago. How people remember these events, therefore, is contingent on what each individual deems important. Similarly, whether one remem­bers is also contingent. Events that stir up painful memories, like instanc­es of racial discrimination, may be the kinds of events that a person chooses to forget. 33 See Fiss, supra note 29.

This sense of self can bias perceptions of the past. Because it is easier to reject events “that reflect negatively on the self,” people often seek “to distance themselves from their own failures.” 34 Wohl & McGrath, supra note 32, at 1025. This is accomplished by “subjectively alter[ing] the temporal distance between the self and the [negative] event,” leading to perceptions that the event “occurred some time ago.” 35 Id. Thus, research shows that people tend to feel closer in time to experiences that make them feel good about themselves than to those that reflect negatively upon them. 36 Id. One is more likely to experience a bad event as having happened longer ago if that event undermines her positive self-regard.

How do the above explanations of time relate to the courts’ sense of time in equal protection? To my knowledge, social science does not ad­dress this question directly. We might infer from the literature, however, that most people are concerned about being associated with overt racial discrimination from the past, 37 See Rachel D. Godsil & L. Song Richardson, Racial Anxiety, 102 Iowa L. Rev. 2235, 2240–41 (2017) (discussing whites’ anxiety about being perceived as racially discrim­inatory); cf. Gregory W. Streich, Is There a Right to Forget? Historical Injustices, Race, Memory, and Identity, 24 New Pol. Sci. 525, 531 (2002) (observing efforts by “govern­ments and some historians [to] whitewash the past”); id. at 535 (discussing the role of guilt and inno­cence in white resistance to a government apology for slavery). leading to a general social reluctance to per­ceive connections between the present and past discrimination, espe­cially if doing so demands recognition of some complicity in that discrimination.

We can use the example of school segregation to explore this fur­ther. Let us assume that racial separation across public schools is uncon­sti­tu­tional if such separation is a vestige of prior unconstitutional seg­regation. This separation is not only legally wrong but also presumed to be morally wrong. Finally, let us assume that segregation is pervasive, such that it is very hard for people to deny that they are somehow as­sociated with segregation. Most people would prefer not to be connected with a system that is wrong. Thus, they are more likely to try to reject the association between the present racial separation and the bad segrega­tion from the past. In social science terms, they are temporally distant from the past, 38 Cf. Wohl & McGrath, supra note 32, at 1033 (“According to temporal self-ap­praisal theory, to maintain or enhance one’s feeling of positive self-regard, it is often necessary to distance oneself from past negative experiences and, at times, to disparage former selves.”). meaning that they are more likely to forgive themselves (as well as others, presumably) for the wrongs of segregation.

The point of this example is to illustrate that how we experience time is contingent on our perceptions of the underlying event, which are in turn shaped by our identity and sense of self. This perception of time can help to explain the courts’ reluctance to perceive a social status that seems innocent on its face—such as racial separation that is regarded as natu­ral or preordained—as unconstitutional racial wrongdoing. Declar­ing such segregation unconstitutional would mean condemning the ac­tions of a fairly large population. Without a bad actor to blame, the social repercussions of such a decision for a court could be significant. 39 Cf. id. at 1025 (discussing the tendency to reject past events that reflect neg­atively on the self). As one judge in an older case noted in describing the continuing impact of prior discrimination on black employees relative to their white co-workers: “It is undeniable that negroes . . . are at a disadvantage compared with white incumbents . . . . This is a product of the past. We cannot turn back the clock.” Whitfield v. United Steel­workers of Am., 263 F.2d 546, 551 (5th Cir. 1959).

What if, however, some would rather remember such prior wrong­doing because that memory connects with their identity and their ability to make sense of their present? 40 See, e.g., Kapczynski, supra note 25, at 1086 (“Without the past, . . . who am I? . . . Who are we? . . . Without a sense of our identity, how do we begin to make a case for anything? Without mining the past, where do we go for inspiration?” (internal quotation marks omitted) (quoting Frank Michelman)). Consider, for example, an African Am­erican who experiences overt racial discrimination. This experience may make little sense without a framework that places it in a historical and broader social context. 41 See Streich, supra note 37, at 527–31 (discussing the relationship between histori­cal memory and black racial individual and group identities); cf. Johanna Peetz et al., Crimes of the Past: Defensive Temporal Distancing in the Face of Past In-Group Wrong­doing, 36 Personality & Soc. Psych. Bull. 598, 608–09 (2010) (discussing temporal distanc­ing by Germans confronted with the history of the Holocaust). Indeed, without this connection between past and present and the awareness that it brings, she may be more likely to assume that there is something wrong with her. Rather than attributing her negative experiences to deeper social conditions and dynamics that are beyond her control, she may internalize the negativity as a reflection of her own ability and worth.

These subjective conceptions of time, in which people experience positive events as closer in time than negative ones, may help to explain the Supreme Court’s reluctance to attribute current racial inequality to prior discrimination. 42 See generally Streich, supra note 37, at 525 (placing the ongoing debate over slavery apology and compensation in the United States in the broader context of for­getting and remembering the past). Take, for instance, the Court’s 1883 opinion in the Civil Rights Cases. 43 109 U.S. 3 (1883). In striking down an 1875 federal statute that barred racial discrimination in public accommodations, the Court con­cluded that such discrimination was not sufficiently related to slavery to justify the statute’s enactment under the Thirteenth Amendment:

When a man has emerged from slavery, and by the aid of beneficent legislation has shaken off the inseparable concom­itants of that state, there must be some stage in the progress of his elevation when he takes the rank of a mere citizen, and ceas­es to be the special favorite of the laws, and when his rights as a citizen, or a man, are to be protected in the ordinary modes by which other men’s rights are protected. 44 Id. at 25.

The role of time is evident in the Court’s rush to declare the 1875 act unconstitutional. It is hard to imagine how the majority perceived a congressional civil rights statute enacted to prevent racial discrimination barely a decade after the end of the Civil War as not remedying vestiges of slavery. 45 See Kapczynski, supra note 25, at 1092 (“[T]ransition inaugurated by the [Thir­teenth] Amendment was less a transformation than a preservation of the old order.”). This conception of time, however, enabled the Court to char­acterize a law that sought to ensure equality for people who had been en­slaved—by providing a right to be treated the same regardless of race—instead as conferring a “special” right to be a “favorite of the laws.” 46 Civil Rights Cases, 109 U.S. at 25. For a discussion of the “boldness of the Court’s assertion,” see Boddie, Adaptive Discrimination, supra note 3, at 1286–87. The Court’s temporal distance from slavery thus blinded it to the permuta­tions of discrimination and how the vestiges of slavery could be mani­fested in other kinds of racial subordination.

Exploring these points helps us to appreciate that the debate about time “is not just about whether one should forget or remember, but also about how the past is interpreted.” 47 Streich, supra note 37, at 525; see also id. at 527 (noting that this question of whether to forget or remember may also apply to people of color who embrace a “right to forget where they came from”). Historical memory about race is so­cially and politically contested because it is interwoven with the racially fraught politics of guilt, innocence, and atonement for historical injus­tices 48 Cf. Aikaterini Vassilikopoulou et al., Product-Harm Crisis Management: Time Heals All Wounds?, 16 J. Retailing & Consumer Servs., 177–79 (2009) (describing consum­ers’ tendency to “forget” a company crisis and “to buy a new product when the company involved in the crisis is socially responsible”). and, therefore, challenges collective social identity and self-re­gard. 49 See Streich, supra note 37, at 531–36. On the other hand, a historical lens helps us to make sense of inherited social structures that “enhance[], restrict[], or distort[]” “[i]n­dividual choices and freedoms.” 50 Cf. Streich, supra note 37, at 537 (discussing how the individual choice of some light-skinned black Americans to “pass” into white society highlights the constraint on self-determination for other black Americans). As the discussion of school desegrega­tion cases below shows, examining the present in relation to the past helps to account for present racial inequality. 51 Cf. id. at 531 (“[S]urvivors and descendants of historical injustices have every right to argue that we—individually and collectively—must remember the past lest we fail to learn from it . . . .”). Martin Luther King, Jr. “rejected the myth that time heals all wounds” and claimed rather that “time is neutral: it can be used either destructively or constructively . . . . Human progress never rolls in on wheels of inevitability; it comes through the tireless efforts of men [and women].” Id. at 535 (alteration in original) (internal quotation marks omitted) (quoting Martin Luther King, Jr., Letter from a Birmingham Jail, in Why We Can’t Wait 86 (1964)). But these connections also require people to revisit suffering and own up to their possible complicity in past and present wrongdoing. This uncomfortable reality leaves some wanting to remember and others wanting to forget.

II. Time and the Contest for the Past

The next section introduces three modes of time in equal protection analysis, with a particular focus on school desegregation. This section then discusses Columbus Board of Education v. Penick 52 439 U.S. 1348 (1978). to illustrate the contest over time in school desegregation and how disputes over time masked substantive debates about whether equal protection applied to systemic forms of racial inequality that were tied to the past. In later cases, the Court invoked time considerations to conclude that ongoing racial separation in former de jure school systems was no longer consti­tution­ally cognizable, thereby justifying the termination of judicial over­sight and placing school segregation beyond equal protection’s reach.

A. The Consequences of Time in School Desegregation

1. Modes of Time Argumentation. — This section identifies three differ­ent ways that time is used and contested in equal protection analysis, with particular force in school desegregation cases. Each mode correlates with a different normative argument about the kind of racial harm that should be constitutionally actionable. I distill these arguments into the following categories: causative, systemic, and limited.

Causative mode is associated with the causation requirement in in­tent doctrine and the search for discriminatory motive. Under this mode, courts look to whether the challenged action or policy is sufficiently close in time to prior discrimination by the same perpetrator. 53 See infra section II.B (describing motive analysis). If so, a court is more likely to infer racial intent. This mode, which is analogous to causa­tion analysis in tort, 54 See Denise C. Morgan, The New School Finance Litigation: Acknowledging that Race Discrimination in Public Education Is More than Just a Tort, 96 Nw. U. L. Rev. 99, 114 (2001) (“The Court has required plaintiffs [in constitutional cases about education] to show that the defendant has engaged in wrongful conduct that proximately caused his or her injuries, and has sought to limit its remedies to ‘actual’ wrongdoers and ‘actual’ victims.”). But see Sanne H. Knudsen, The Long-Term Tort: In Search of a New Causation Framework for Natural Resource Damages, 108 Nw. U. L. Rev. 475, 497, 531 (2014) (noting that traditional tort causation standards focus on timing as circumstantial evid­ence of causation and are ill-suited to remedying long-term (environmental) injuries but arguing that substantial-factor causation analysis already used in toxic tort cases can be adapted for this type of injury). presumes that discrimination is linear and static—meaning that discrimination can be causally traced to a discrete agent, action, or policy—and that the site of the initial discrimination is the sole lo­cus for racial harm. Because the harm does not spread, it is easier to detect.

Systemic mode focuses on cross-generational inequality and is sen­sitive to dynamic shifts of racial discrimination. It is the opposite of causa­tive mode because it presumes that discrimination morphs through time and manifests across different social, economic, and political domains.

Limited mode is less descriptive and functions more like a statute of limitations. Courts that operate in limited mode may recognize the inter­generational nature of racial inequality but decide to disregard it as being beyond the proper reach of the federal courts, based most often on fed­eralism concerns about the role of the courts in relation to allegations of ongoing discrimination by state actors.

2. The Evolution of Intent Post-Brown. — The years following Brown v. Board of Education 55 347 U.S. 483 (1954). produced a dynamic that is typical of the intermittent nature of racial progress. 56 Cf. Derrick Bell, Brown v. Board of Education: Reliving and Learning from Our Racial History, 66 U. Pitt. L. Rev. 21, 22 (2004) (arguing that racial progress ensues only when it converges with the interests of whites and that progress invariably gives way when “policy­makers fear [that] the remedial policy is threatening the superior societal status of whites”). School districts throughout the country used a variety of tactics to create and maintain racial separation in public schools, 57 See Drew S. Days, III, School Desegregation Law in the 1980’s: Why Isn’t Anybody Laughing?, 95 Yale L.J. 1737, 1740–41 (1986) (book review) (discussing extensive segrega­tive practices of northern school boards). either in passive or open defiance of Brown. These techniques included use of neighborhood schools, manipulation of attendance zones, transfer policies, siting of new school construction, freedom of choice plans, and racially discriminatory faculty assignments. 58 See id. at 1740–41; see also Clemons v. Bd. of Educ., 228 F.2d 853, 857 (6th Cir. 1956) (finding the zoning resolution was adopted in order to perpetuate segregation); Reed v. Rhodes, 500 F. Supp. 404, 413 (N.D. Ohio 1980) (finding the com­plaint proce­dures in Ohio were deficient and not structured “to uncover instances of rac­ial segre­gation”); Berry v. Sch. Dist., 442 F. Supp. 1280, 1326 (W.D. Mich. 1977) (finding student transfers, boundary changes, and grade pattern changes were used to segregate schools); Hobson v. Hansen, 269 F. Supp. 401, 443, 500–02 (D.D.C. 1967) (holding ability grouping, optional zones, and discriminatory faculty assignment are inconsistent with the Equal Pro­tection Clause). In Seattle, the NAACP charged that the school district had manipulated boundary lines, school attendance policies, and school construction sites, in addition to discriminating in teacher assignments in order to create and perpetuate seg­regation. See Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 809 (2007) (Breyer, J., dissenting). Later, the U.S. Department of Health, Education, and Welfare claimed that the Seattle school board facilitated segregation by adopting school-transfer criteria and “a construction program that needlessly built new schools in white ar­eas,” in addition to maintaining “inferior facilities at black schools.” See id. at 810. In Pasadena City, California, a court struck down the district’s neighborhood school policy and its anti-busing policy upon a showing of significant racial imbalance in local schools. Spangler v. Pasadena City Bd. of Educ., 311 F. Supp. 501, 504, 507 (C.D. Cal. 1970). On appeal, the Ninth Circuit found it particularly instructive that the district, among its other practices, had altered attendance zones in ways that intensified black–white separation; rejected policies that would have increased integration; subsidized segregative transport­ation policies; “discriminat[ed] in the hiring and promotion of black administrators”; and granted transfer requests that exacerbated segregation. Spangler v. Pasadena City Bd. of Educ., 519 F.2d 430, 432 (9th Cir. 1975), vacated, 427 U.S. 424 (1976). In Columbus, Ohio, a district court found that the local school board was conducting “an enclave of separate, black schools on the near east side of Columbus” and that “[t]he then-existing racial separation was the direct result of cognitive acts or omissions of those school board members and administrators who had originally intentionally caused and later perpet­uated the racial isolation.” Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 236 (S.D. Ohio 1977), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979). School segregation was also perpetuated through other forms of systemic dis­crimination, such as residential segregation 59 See Paul R. Dimond, Beyond Busing: Reflections on Urban Segregation, the Courts, and Equal Opportunity 62 (2005) (“In all these respects, the board intentionally built upon the residential segregation to create, maintain, and magnify school segregation.”). and discriminatory voting and employment practices. 60 See generally Eric S. Stein, Attacking School Segregation Root and Branch, 99 Yale L.J. 2003, 2012–13, 2012 n.61 (1990) (discussing the relationship between discrimin­atory voting practices, school segregation, and other discriminatory government conduct). These mutually reinforcing policies and prac­tices allowed segregation to take root and to metastasize, not only in the public schools but also across social systems and governmental institutions. 61 See generally id.; see also Days, supra note 57, at 1740–41 (describing mutually reinforcing practices pursued by federal, state, and local officials that constituted school segregation).

With the backing of the federal government, 62 See Harrell R. Rodgers, Jr. & Charles S. Bullock, III, Law and Social Change: Civil Rights Laws and Their Consequences 81–84 (1972) (describing an uptick in deseg­rega­tion following the passage of the 1964 Civil Rights Act, which threatened to terminate federal funding to noncompliant schools, and the subsequent enforcement efforts by the U.S. Department of Health, Education, and Welfare). courts in the post­-Brown era became alert to these adaptive maneuvers and declared prac­tices that perpetuated prior segregation unconstitutional. 63 See supra Part I. Without the formal sanction of law, however, these practices often defied easy cate­goriza­tion. 64 See Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 820 (2007) (Breyer, J., dissenting) (“No one here disputes that Louisville’s segregation was de jure. But what about Seattle’s? Was it de facto? De jure? A mixture? Opinions differed.”); cf. Albert P. Blaustein & Clarence Clyde Ferguson, Jr., Desegregation and the Law: The Meaning and Effect of the School Segregation Cases 240–41 (1957) (noting that courts during this time defined “evasion” as noncompliant with Brown’s mandate and, therefore, unconstitutional); Andrew R. Highsmith & Ansley T. Erickson, Segregation as Splitting, Segregation as Joining: Schools, Housing, and the Many Modes of Jim Crow, 121 Am. J. Educ. 563, 566 (2015) (describing the “myth of de facto segregation” that was used to “de­fend racial segregation in the North and West”). Questions soon surfaced, therefore, about the extent of Brown’s reach in jurisdictions, especially outside the South, where the law had not formally effectuated or ratified segregation.

As Professor David Strauss has observed, the answers to these ques­tions turned on interpretations of the constitutional harm in Brown, 65 Strauss framed the uncertainty as such:
“While the principle of Brown seems clear to this extent, it was not clear, until Washington v. Davis, which conception of discrimination Brown embraced, or how far the principle of Brown extended. Did it reach only explicit segreg­ation? Did it extend to all actions that in some sense helped perpetuate the vices of the Jim Crow system, by stigmatizing blacks or keeping them in a subordinate posi­tion, even if those actions made no explicit reference to race? Did it require states to ensure substantive equality for black and white citizens, at least in areas like education, even if state-sponsored discrimination had not caused the inequality?”
David A. Strauss, Discriminatory Intent and the Taming of Brown, 56 U. Chi. L. Rev. 935, 947 (1989); see also James S. Liebman, Desegregating Politics: ‘All-Out’ School Desegreg­a­tion Explained, 90 Colum. L. Rev. 1463, 1662–63 (1990) (observing that Swann created a presumption that a showing of unaddressed, “[systematic,] comprehensive” intentional discrimination serves as proof that ongoing segregative effects are due to persistent racism).
in­cluding the cognizability of systemic racial discrimination and, by proxy, whether intent was required to establish an equal protection violation. Northern school districts wrestled with whether “racial imbalance” itself violated equal protection or whether such imbalance was constitutionally permissible in the absence of an identified discriminatory purpose by school officials. 66 See Note, Segregation Litigation in the 1960’s: Is There an Affirmative Duty to Integrate the Schools?, 39 Ind. L.J. 606, 606 (1964) (questioning whether the fact of racial imbalance resulting from the use of neighborhood schools in the context of residential segregation “is unconstitutional in and of itself”). In Green v. County School Board, the Court concluded that school officials were “clearly charged with the affirmative duty to take whatever steps might be necessary to convert to a unitary system in which racial dis­crimination would be eliminated root and branch.” 391 U.S. 430, 437–38 (1968). Professor Strauss has argued that the question was not settled until the intent standard was established in Washington v. Davis. 67 See Strauss, supra note 65, at 947. As the Davis Court stated:
“The school desegregation cases have also adhered to the basic equal pro­tection principle that the invidious quality of a law claimed to be racially discrim­inatory must ultimately be traced to a racially discriminatory purpose. That there are both predominantly black and predominantly white schools in a community is not alone violative of the Equal Protection Clause. The essential element of de jure segregation is ‘a current condition of segregation resulting from inten­tional state action. . . . The differentiating factor be­tween de jure seg­regation and so-called de facto segregation . . . is purpose or intent to segregate.'”
Washington v. Davis, 426 U.S. 229, 240 (1976) (citation omitted) (quoting Keyes v. Sch. Dist. No. 1, 413 U.S. 189, 205, 208 (1973)).
Indeed, in Davis, the Court, invoking school desegregation cases, clari­fied that the mere fact of “predominantly black and predominantly white schools in a community [was] not alone violative of the Equal Protection Clause.” 68 Davis, 426 U.S. at 240; cf. Milliken v. Bradley, 418 U.S. 717, 744–45 (1974) (con­cluding that interdistrict relief requires a showing of “constitutional violation within one district that produces a significant segregative effect in another district”).

Still, the meaning of intent had yet to be fully developed. The Court took this additional step in Personnel Administrator of Massachusetts v. Feeney. 69 442 U.S. 256, 279 (1979). Feeney further limited Brown by defining intent to refer to state actions taken “because of” their consequences, rather than merely “in spite of” them. 70 Id. (“‘Discriminatory purpose,’ however, implies more than intent as volition or intent as awareness of consequences. It implies that the decisionmaker . . . selected or re­affirmed a particular course of action at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group.”). Feeney was important for school desegregation doctrine because it threatened to undermine lower court decisions that school boards had unconstitutionally discriminated by taking actions having the “probable, foreseeable, and actual result of increasing segregation.” 71 Berry v. Sch. Dist., 467 F. Supp. 630, 686–87 (W.D. Mich. 1978); see also Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 255 (S.D. Ohio 1977), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979).

3. “Remoteness in Time” and Systemic Harm. — As courts struggled to discern the meaning of Brown and the scope of the intent doctrine, ques­tions surfaced about the significance of the passage of time for assessing seg­regative intent, foreshadowing the constitutional turn in later school deseg­regation cases. Penick v. Columbus Board of Education is instructive. There, the district court held defendant school officials in Columbus, Ohio, responsible for school segregation that “predictab[ly]” 72 Penick, 429 F. Supp. at 255. resulted from a constellation of policies: the combined use of school construction sites, 73 Id. at 241 (“The evidence supports a finding that the Columbus defendants could have reasonably foreseen the probable racial composition of schools to be con­structed on a given site.”). attendance zones, 74 Id. at 241–47 (discussing how attendance zones created and maintained segregation). and a neighborhood school policy implement­ed against the backdrop of residential segregation. 75 See id. at 241 (discussing how defendant built schools with a racially identifiable student population, notwithstanding a “neutral neighborhood school policy”). The opinion is use­ful for understanding the role of time, specifically how considerations of time were used to advance competing causative versus systemic interpre­tations of constitutional harm. 76 Id. at 255 (“Substantial adherence to the neighborhood school concept with full knowledge of the predictable effects of such adherence upon racial imbalance in a school system is one factor among many others which may be considered by a court in determin­ing whether an inference of segregative intent should be drawn.”).

The first issue for the district court was whether school officials could be held responsible for the acts of their predecessors. The defend­ant school board members had argued that their conduct was too “remote” in time to be actionable. 77 Id. at 252. The court concluded that these prior acts were rele­vant to the constitutional inquiry because the defendants had been sued in their official capacities. 78 Id. But the court then elaborated on the role of time considerations in the analysis, emphasizing that although the effects of prior bad acts theoretically could be “attenuated by time,” their “‘re­moteness in time’” alone was not dispositive. 79 Id. (quoting Keyes v. Sch. Dist. No. 1, 413 U.S. 189, 210–11 (1973)).

Here the district court drew on the Supreme Court’s decision in Keyes v. School District No. 1. 80 413 U.S. 189. Keyes is most notable for having imposed the intent requirement in school-desegregation cases. 81 See Liebman, supra note 65, at 1592 (describing the “special demands” posed by the intent requirement in school desegregation cases). For an excellent discussion of the evolution of the intent doctrine in school desegregation, see generally Katie R. Eyer, Ideo­logi­cal Drift and the Forgotten History of Intent, 51 Harv. C.R.-C.L. L. Rev. 1, 20–47 (2016). But it also advanced a systemic understanding of racial discrimination, finding that the pres­ence of “intentionally segregative school board actions in a meaningful portion of a school system” creates a rebuttable presumption that “other segregated schooling within the system” is not merely coincidental. 82 Keyes, 413 U.S. at 208. Such a showing establishes a “prima facie case of unlawful segregative de­sign,” the Court concluded, shifting to school authorities the burden “of proving that other segregated schools within the system are not also the result of intentionally segregative actions.” 83 Id. The Court’s analysis reflect­ed its understanding of the synergies of segregation—one that would be weakened in later cases—emphasizing that the presumption of segre­gative intent applied even if “different areas of the school district should be viewed independently of each other.” 84 Id. If discriminatory intent affect­ed one part of the school system, it was likely to have affected other parts as well. 85 Id.

Given the Keyes Court’s systemic interpretation of segregation, it was no surprise that it embraced a similarly systemic view of time and histori­cal inquiry for assessing constitutional harm:

We reject any suggestion that remoteness in time has any relevance to the issue of intent. If the actions of school auth­or­ities were to any degree motivated by segregative intent and the segregation resulting from those actions continues to exist, the fact of remoteness in time certainly does not make those actions any less “intentional.” 86 Id. at 210–11.

Keyes gave the district court in Penick an opening to extend its own temporal inquiry deep into the historical record, searching as far back as the late 1800s to document the origins of school segregation in Columbus. 87 See Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 234–36 (S.D. Ohio 1977), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979). The Penick court cautioned that its purpose was neither to “drag[] out skeletons of the past nor [engage in] a vindictive finger-pointing exer­cise.” 88 Id. at 234. Instead it explained that its “look to the past must be made to dis­cover whether past acts or omissions are in any degree responsible for the admitted current racial imbalance in the Columbus schools.” 89 Id.

From its review of prior history, the court concluded that the public schools were a dual, segregated system at the time of Brown. 90 Id. at 236. They were “the direct result of cognitive acts or omissions of those school board members and administrators who had originally intentionally caused and later perpetrated the racial isolation” in the system. 91 Id. It similarly conclud­ed that the defendants were complicit in perpetuating prior segregation in the twenty-year interval between Brown and the plaintiffs’ complaint due to their failure to take affirmative steps to desegregate. 92 Id. at 261–64.

The court’s use of time reflected its correspondingly broad concep­tion of constitutional harm, as not just purposeful discrimination but also as embracing the relevance of systemic inequality to inquiries about the constitutionality of persistent racial separation in public schools. This his­torical lens allowed the court to cast a wide net in examining practices that had created and exacerbated school segregation. It moved beyond an examination of specific school board practices to explore the local effects of past segregative practices in housing 93 See id. at 259. —including federal, state, and local governmental policies, “personal preferences of blacks and whites,” and private activity in the real estate industry. 94 Id. Recognizing the “substantial reciprocal effect” between neighborhood composition and local schools, the court concluded that school officials’ lack of “control” over the resulting housing segregation was immaterial. 95 Id. Their awareness of this relationship between neighborhood composition and local schools was enough to infer segregative intent. 96 See id.

In their briefs to the Supreme Court, the parties vigorously contest­ed the significance of time to the constitutional question. The board of education protested the lower court’s reliance on “remote” actions that, it argued, were insufficiently related to extant school segregation. 97 Brief for the Petitioners, Columbus Bd. of Educ. v. Penick, 443 U.S. 449 (1979) (No. 78-610), 1979 WL 200100, at *63 (“Although intentionally discriminatory actions by predecessor boards of education . . . may have had the immediate impact of causing the student bodies of five schools to be predominantly black, the racial composition of those schools at the time of trial cannot be logically attributed to the lingering effects of . . . [those] actions.”); see also id. at *67–70 (arguing that the lower courts should not have presumed “a causal connection between remote and isolated acts and a current condition of racial imbalance”). In contrast, amici who filed in support of the plaintiffs argued that the “cur­rent conditions of segregation” in the Columbus public schools had to be evaluated in light of the “historical creation and maintenance of the dual systems.” 98 See Brief for the United States as Amicus Curiae, Penick, 443 U.S. 449 (Nos. 78-610, 78-627), 1979 WL 200113, at *48; see also Brief of the Fair Housing Council of Bergen County, New Jersey Amicus Curiae, Penick, 443 U.S. 449 (Nos. 78-610, 78-627), 1979 WL 200110, at *21 (“Petitioners’ argument relies . . . upon claims that residential patterns and the passage of time have so attenuated any impact their actions could have had as to excuse their prior actions. Petitioners, however, do not demonstrate how these other factors have eradicated the effects of their actions.”).

The Supreme Court concluded that the district court’s decision complied with the intent standard in Washington v. Davis. 99 See Penick, 443 U.S. at 464. Most interest­ingly for our purposes, it rejected the time-limited analysis argued by the school board. 100 See id. at 455–56 (affirming “the judgment of the Court of Appeals, based on the [District Court’s] findings . . . , that the Board’s conduct at the time of trial and before not only was animated by an unconstitutional, segregative purpose, but also had current, seg­regative impact that was sufficiently systemwide to warrant the remedy ordered . . . .”). It is not at all clear, however, whether Penick would have survived the Court’s Feeney decision two years later, given its significant reliance on the foreseeability standard that Feeney explicitly rejected. 101 See id. at 464.

Then-Justice Rehnquist’s dissent in Penick foreshadowed the turn that the Court would take a decade later. He objected that the district court had failed to provide “any concrete notion of what a ‘systemwide viola­tion’ consists of or how a trial judge is to go about determining whether such a violation exists or has existed.” 102 Id. at 490 (Rehnquist, J., dissenting). The federal court’s “dis­place­ment” of “local autonomy” and control over the school system, Rehnquist insisted, required a showing of “discriminatory purpose and a causal relationship between acts motivated by such a purpose and a cur­rent condition of segregation in the school system.” 103 Id. at 490–91. In Rehnquist’s view, the majority’s decision in Penick was fatally flawed because it em­braced a “methodology [that] would all but eliminate the distinction between de facto and de jure segregation.” 104 Id. at 491. As to the role of time, Rehnquist argued that Penick threatened to “render all school systems captives of a remote and ambiguous past.” 105 Id.

Penick provides a useful point of comparison with the Supreme Court’s narrow treatment of both time and discrimination that would surface a decade later in its school-desegregation cases. 106 See infra text accompanying notes 139–146, 154, 162–172 (describing the resurfac­ing of the Court’s static, linear understanding of discrimination in cases such as Swift and Freeman). The district court recognized discrimination as systemic, dynamic, and chronic based on a historical record that spanned a century. Its expansive view of time similarly reflected its broad understanding of discrimination—as mani­festing not only in the public schools but also in the “reciprocal,” “mu­tually constitutive” 107 See Highsmith & Erickson, supra note 64, at 565 (“Rather than simply following the dynamics of residential segregation, schools have consistently helped define the boundaries of the segregated neighborhood. Schools and housing, far from existing in a hierarchical relationship, have long been mutually constitutive entities.”). sphere of housing segregation, itself the product of decades of discrimination by multiple governmental and private actors. The school board’s failure to adjust its policies based on these “fore­seeable” “social dynamics” 108 See Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 260 (S.D. Ohio 1977), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979). made it responsible for the resulting seg­regation in its public schools.

The competing interpretations of time in Penick—between the dis­trict court and Rehnquist in dissent—reflect different understand­ings of whether systemic racial discrimination violates equal protection. The district court focused its temporal inquiry backward to examine the con­nective relationship between segregation of the past and segregation in the present, and forward in finding that school officials had failed to take ap­propriate corrective action to prevent the segregation that would foresee­ably persist. Conversely, Rehnquist’s narrow, causal view of discrimination—as necessarily tied to discrete action, occurring during a discrete period of time—limited both his retrospective examination of the historical record and prospective inquiry into the likelihood that such discrimination would persist into the future.

As discussed below, Rehnquist’s view of time in Penick eventu­ally pre­vailed as the composition of the Court shifted. As a result, the Court down­graded its enforcement against school segregation by adopt­ing weaker standards that privileged local control of school districts over integration. School districts could be released from judicial supervision if they had acted “in good faith” to eliminate the vestiges of their prior dis­crimina­tion “to the extent practicable.” 109 Bd. of Educ. v. Dowell, 498 U.S. 237, 249 (1991). The Court’s emphasis on good faith and practicability meant that school districts that had abided by low­er court orders for a reasonable time could reclaim local control, the assumption being that any remaining “racial imbalance” was not attri­butable to the prior de jure violation. 110 See Freeman v. Pitts, 503 U.S. 467, 498 (1992) (“A history of good-faith compli­ance is evidence that any current racial imbalance is not the product of a new de jure violation, and [supports] the school board’s representation that it has accepted the princi­ple of racial equality and will not suffer intentional discrimination in the future.”). Under this limited mode of time analysis, school districts also could be released from judicial supervision incrementally, enabling courts to overlook the dynamic interactions of segregation across different facets of a school system’s operations. 111 See id.

By prioritizing time considerations in decisions to terminate judicial oversight, the Court facilitated and further entrenched school segre­gation. As school districts resumed control in the wake of these decisions, school segregation rose significantly in the South. 112 See Gary Orfield et al., Civil Rights Project, Brown at 62: School Segregation by Race, Poverty and State 3 (2016), (on file with the Columbia Law Review) (observing that after the Dowell decision “the share of intensely segregated nonwhite schools (which we defined as those schools with only 0–10% white students) more than tripled, rising from 5.7% to 18.6% of all public schools”); Sean F. Reardon et al., Brown Fades: The End of Court-Ordered School Desegregation and the Resegregation of American Public Schools, 31 J. of Pol’y Analysis & Mgmt. 876, 877 (2012) (noting that “after being released from court oversight, school districts become steadily more racially segregated”). Shifting racial demographics may be partly responsible. Id. at 1–2 (noting that the racial composition of public schools has also “changed dramatically, falling from 69% white to 50% white,” accompanied by an increasing share of Latino stu­dents, which “soared from 11% to 25%”). Brown’s declaration that segregation is “inherently unequal” 113 See Brown v. Bd. of Educ., 347 U.S. 483, 495 (1954). withered and then disap­peared from equal protection altogether. 114 Parents Involved in Cmty. Schs. v. Seattle Sch. District No. 1, 551 U.S. 701, 736 (2007). The resulting doctrine nor­mal­ized segregation as both natural and inevitable, 115 See, e.g., Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 233 (S.D. Ohio 1977) (“[A]s I view it, the real reason courts are in the school desegregation business is the fail­ure of other gov­ernmental entities to confront and produce answers to the many problems in this area pursuant to the law of the United States.”), aff’d in part, 583 F.2d 787 (6th Cir. 1978), aff’d, 443 U.S. 449 (1979). making it invisible to law. This shift allowed segregation to deepen and to spread, becoming entwined in daily life and placing it further from law’s reach.

B. Time and Discriminatory Motive

This section examines an early, influential article by Paul Brest that implicitly advanced intent-based or causative conceptions of time for di­agnosing discriminatory motive. 116 See Paul Brest, Palmer v. Thompson: An Approach to the Problem of Unconstitu­tional Legislative Motive, 1971 Sup. Ct. Rev. 95. It uses the lens of a contemporary prob­lem in school desegregation to unpack the influence of these conceptions of time. Part III below discusses how these considerations of time have been used to further limit equal protection.

Consider the following problem. In Time A, state and local officials design and advance policies for the express purpose of segregating stu­dents in public schools on the basis of race. These specific practices cease following court intervention and apparent changes in public attitudes about the acceptability of overt discrimination. In Time B, school officials adopt new practices that are facially race neutral but generate the same segregative effects. The governmental officials responsible for the Time B practices then move on. Several decades later, in Time C, the Time B practices are still in place, and the schools are still segregated.

How should courts respond in Time C to equal protection claims for continued enforcement and remediation in light of ongoing segregation, the roots of which originated in Time A?

In 1971, Paul Brest argued that “a decisionmaker’s motivation” should be “viewed in the context of antecedent and con­current events and situa­tions.” 117 Id. at 120–21. His contention was that the “juxta­position of a decision with some prior event or sequence of events often bears on the inference of illicit motivation.” 118 Id. at 122. Brest did not explicitly frame motive as a function of time. But as I have written elsewhere, 119 See Boddie, Adaptive Discrimination, supra note 3, at 1288–89. his emphasis on the relevance of the challenged event’s “chronological se­quence”—and whether the defend­ant decisionmaker had previously engaged in a discriminatory practice—implicitly incorporated considera­tions of time into the intent inquiry. 120 See Brest, supra note 116, at 122.

To illustrate, Brest offers the example of a conventional school de­segregation case immediately after Brown. 121 Id. at 99–102. A court enjoins a school dis­trict from assigning students on the basis of race. Immediately after the injunction is imposed, the state enacts a law providing tuition grants to students attending private schools, “or abandons a public school sys­tem, or engages in other practices that tend to maintain segregation.” 122 Id. at 123. Un­der Brest’s framework, the timing of these facially neutral laws and prac­tices indicates that their true purpose is to return to the prior system of segregation. 123 See id. at 122–23.

This interpretation of intent has a common-sense appeal. Indeed, at the time of Brest’s article, the sequence he described was similar not only to the evasive tactics of school officials but also to other responses to enforcement against racial discrimination. 124 See id. at 122 (providing a general timeline of regulation and backlash that sup­ports an inference of discriminatory motivation in various contexts). These el­ements of Brest’s analysis are similar to the Court’s test in Village of Arlington Heights v. Metropolitan Housing Development Corp., which em­braced chronological sequence as a factor for diagnosing discriminatory intent. 125 429 U.S. 252, 267 (1977) (“The specific sequence of events leading up to the challenged decision also may shed some light on the decisionmaker’s purposes.”); see also Boddie, Adaptive Discrimination, supra note 3, at 1290 (observing the influence of Brest’s theory in Arlington Heights). As discussed below, one can also see the influence of Brest’s ap­proach in the Rehnquist Court’s later school-desegregation cases.

Brest’s use of time, however, undermines broader inquiries for iden­tifying and rooting out systemic racial harms. 126 As Professor James Liebman writes:
“For this first [modern] pluralist-democratic political system of ours has been assiduously computing racist preferences for 200 years now, notwithstanding a civil war, three constitutional amendments, periodic more or less organized interracial violence, and the development of a largely racially defined under­­class that presents as big a domestic problem as the country now knows.”
Liebman, supra note 65, at 1636; see also Boddie, Adaptive Discrimination, supra note 3, at 1289–90 (noting that Brest’s “presumption that time inoculates us against racial discrim­ination is inconsistent with history and experience”).
An account that focuses strictly on chronology and sequencing misses discrimination’s adaptive, more dynamic elements and the subtleties of how it shifts across multiple domains through time. One can see this problem in the above query. Because Time C segregation is chronologically distant from the original violation in Time A, Brest’s framework might lead one to infer that it is not a vestige of past intentional discrimination and, therefore, is not con­stitutionally cognizable.

What if equal protection adopted a different framework for under­standing discrimination and how it operates—one like the district court’s opinion in Penick that embraced a broad historical inquiry based on its systemic interpretation of discrimination and correspondingly broad rac­ial harms? 127 See supra section II.A.3. An analogy here is useful. Assume that discrimination is can­cer and that public schools are the patient. Cancer manifests in the form of overtly discriminatory student assignment policies that lead to the segregation of black and white students in public schools. On this linear, static understanding of discrimination, the cancer is localized and does not move beyond the public schools. It can be surgically removed by excising the use of race from student assignment and then applying chem­otherapy to destroy the remaining cancer cells—that is, the residual effects of intentional discrimination. This “chemotherapy” consists of a period of judicial intervention and oversight of student assignment to ensure that the discrimination does not reappear. The patient is consid­ered cured if the cancer (the overt use of race in student assignment) does not resurface within a designated period.

Now let us consider a different set of facts based on the assumption that the cancer of racial discrimination is not linear and static, but sys­temic, dynamic and, therefore, adaptive. The treatment proceeds in the same way—by removing the use of race in student assignment and then applying the same chemotherapy of judicial supervision. We soon dis­cover, however, that the cancer is not localized. Instead it has spread to other ostensibly race-neutral local practices that achieve the same result. These include housing and zoning policies 128 Highsmith & Erickson, supra note 64, at 563–64 (“The array of forces that have divided people and landscapes in the United States, particularly since the turn of the twen­tieth century, is striking in its breadth: racial zoning, discriminatory home finance pro­grams, restrictive housing covenants, legally mandated segregation, and gerrymandered school zone lines, among others.”). that effectively block blacks from moving into white neighborhoods and, therefore, preclude blacks from attending white schools (because of other facially race-neutral, resi­dence-based student assignment policies). 129 Id. at 565 (observing that “segregated neighborhoods and schools that spread . . . during the twentieth century grew out of a combination of government policies and pri­vate acts of discrimination perpetrated in both educational and residential spheres” even though local actors and legal discourse attempted veiling practices in the language of de facto segregation). The metastatic nature of the discrimination not only affects school segregation but also spreads resi­den­tial segregation across communities and neighborhoods. 130 The problem is compounded because school segregation and residential seg­rega­tion are mutually reinforcing. Id. (“In reality . . . housing and school policies have al­most always worked together as part of broader networks of metropolitan segregation.”).

If we are focused on this latter, systemic discrimination, then the pas­sage of time tells us very little about whether the underlying problem of segregation has been remedied. Using time as a proxy for evaluating the effectiveness of the remedy distracts us from the true nature of the prob­lem. Returning to the above example, assume that a school’s student body in Time C has some black students now that the school’s prior dis­criminatory practices in student assignment from Time A have been reme­died. In Time D, however, the town in which the school is located adopts zoning policies that make it harder for the surrounding, lower-income black population to move in. The town becomes residentially segregated, as do its schools once the local population of black students graduates. If one focuses solely on the original Time A violation in stu­dent assignments, it would be easy to miss the segregative impact of the town’s housing practices in Time D that regenerate school segregation. Extending one’s horizons allows one to perceive segregative dynamics that persist across multiple domains through time. As a practical matter, it may be hard to prove that Time D policies were motivated by racial intent. On the other hand, folding the prior history of the town into such an inquiry could lead courts to be presumptively suspicious of policies and practices that reinforce previous segregative patterns.

Let us return once more to our earlier query about whether segre­gation in Time C is constitutionally actionable when unconstitu­tional dis­crim­ination occurred in Time A. Recall that Time A and Time C are separated by the interval Time B, in which government practices have changed and the offending state actors have left, but segregation never­theless per­sists. Whether the school board will be subject to contin­uing judicial super­vision in Time C will depend on the sufficiency of its correc­tive action in Time B.

The question of official liability in Time C turns on an additional set of inquiries. The first is whether the facially neutral Time B practices are in fact race neutral. The answer here is straightforward as a matter of cur­rent doctrine. Under Washington v. Davis, a showing of discriminatory ra­cial purpose, rather than discriminatory impact alone, is required to establish a constitutional violation. 131 426 U.S. 229, 239 (1976). Feeney further defines “purpose” to be deliberative, rather than simply the product of a foreseeable dis­parity. 132 Pers. Adm’r of Mass. v. Feeney, 442 U.S. 256, 279 (1979) (“‘Discrim­inatory pur­pose,’ . . . implies more than intent as volition or intent as awareness of consequences.”). Thus, the persistence of segregation by itself is not enough to establish an equal protection violation.

Assume, then, that the Time B practices are in fact race neutral and nonpurposive in Feeney’s terms. The next question is whether these prac­tices have been in place for a sufficient period of time to remedy the discrimination from Time A. The district court’s pre-Feeney opinion in Penick and then-Justice Rehnquist’s dissent in that case indicate the con­tested nature of the answer. Recall that their debate revolved around the prop­er time frame for identifying the vestiges of prior intent, which in turn depended on the kinds of discrimination that count in the equal protec­tion analysis. 133 See supra notes 102–105 and accompanying text.

Under the district court’s approach, Time A segregation (and per­haps even pre-Time A segregation) bears on our analysis of the potential vestigial elements in Time C segregation. We could also find it relevant that school officials failed to take proper cognizance of residential segre­gation and its systemic, synergistic relationship to segregation in public schools. Rehnquist’s approach, on the other hand, is much more con­strained by time. His decision to limit the retrospective analysis re­flected his narrow interpretation of discrimination as identified and dis­crete. As with Brest’s framework, Rehnquist focused on a tighter, linear chronol­ogy of events because he was looking for a linear (i.e., nonsystem­ic) kind of discrimination, one with a direct, causal connection to a clear bad actor.

How then does current equal protection doctrine address the Time C query? In Board of Education v. Dowell, the Court concluded that the defendant school district must implement and adhere to corrective action for some “reasonable” period of time. 134 498 U.S. 237, 248 (1991). If the school district achieves this good faith implementation and eliminates the vestiges of its prior intent “to the extent practicable,” then the court declares the dis­trict “unitary.” 135 Id. at 249–51. The determination of unitary status means that there is no constitutionally recognized (i.e., causal) relationship between the original constitutional violation in Time A and present segregation in Time C. 136 The Dowell Court attributed this judicial limitation to a recognition of the sep­aration of powers between judicial enforcement and local control of school districts:
“The legal justification for displacement of local authority by an injunc­tive de­cree in a school desegregation case is a violation of the Constitution by the local au­thorities. Dissolving a desegregation decree after the local authorities have oper­ated in compliance with it for a reasonable period of time properly recog­nizes that “necessary concern for the important values of local control of public school systems dictates that a federal court’s regulatory control of such systems not extend beyond the time required to remedy the effects of past intentional discrimination.'”
Id. at 248 (citation omitted) (quoting Spangler v. Pasadena City Bd. of Educ., 611 F.2d 1239, 1245 n.5 (9th Cir. 1979) (Kennedy, J., concurring)).
This racial imbalance is instead presumed to be the natural re­sult of private preferences unrelated (at least for purposes of equal pro­tection) to the state’s prior unconstitutional conduct. 137 See id. at 243 (describing the district court’s decision to vacate an injunction in a school desegregation case based on its conclusion that “residential segregation was the re­sult of private decisionmaking and economics and that it was too attenuated to be a vestige of former school segregation”). Thus, injunctive relief against the school district is no longer constitutionally permissi­ble. 138 After the declaration of unitary status, requests for relief must be justified by a new showing of discriminatory intent. See, e.g., Dowell v. Bd. of Educ., 782 F. Supp. 574, 576–77 (W.D. Okla. 1992). The touchstone of the analysis is the practicability of eliminating segregation, not whether segregation has in fact been eliminated.

There is another, related temporal consideration at work here. In as­sessing whether Time C segregation was sufficiently related to segre­gation in Time A, the Dowell Court had to address whether “compliance alone” 139 See Dowell, 498 U.S. at 246–48. was sufficient to discharge the remedial obligations of a district that had unconstitutionally segregated its schools. The Court rejected the more stringent standard from United States v. Swift for terminating an in­junction. 140 Id. at 247. Under Swift, a decree could not be lifted or modified absent “a clear showing of grievous wrong evoked by new and unforeseen con­ditions.” 141 United States v. Swift, 286 U.S. 106, 119 (1931). The Dowell Court held that this standard was not “proper” for “in­junctions entered in school desegregation cases,” 142 Dowell, 498 U.S. at 248. which “[were] not intended to operate in perpetuity,” 143 Id. but rather were imposed as a “tem­po­rary measure to remedy past discrimination.” 144 Id. at 247. The Court’s rejection of Swift revealed that the more pressing concern was not school segre­ga­tion 145 See id. at 247–48. but the school district’s loss of control over its operations. This infringement on local control threatened the “allocation of powers with­in [the] federal system” 146 Id. at 248. As Professor Richard Briffault has explained, the emphasis on local control in education itself has racial dimensions, as it privileges private preferences for homogeneity in public school systems. See Richard Briffault, Our Localism: Part II—Localism and Legal Theory, 90 Colum. L. Rev. 346, 384–87 (1990) (describing the relationship be­tween local control and the desire for “affluent homogeneity” in communities and pub­lic schools). and invited the possibility of “judicial tute­lage for the indefinite future.” 147 See Dowell, 498 U.S. at 249.

To emphasize the influence of time on the Court’s analysis, adjust the facts of our earlier query. What if one reconceptualized Time C seg­regation not as a distinct pattern of behavior unrelated to prior state ac­tion but rather as part of a continuous system of racial discrimination conditioned to endure by state policies and practices that began in Time A?

Here it is useful to compare the majority’s analysis in Dowell to Jus­tice Marshall’s use of time in his dissent. For example, the majority found it noteworthy that “the original finding of de jure segregation was entered in 1963, the injunctive decree . . . was entered in 1972, and the Board complied with the decree in good faith until 1985.” 148 Id. For the Court, thir­teen years of relief was enough.

Justice Marshall’s dissent began in a different time. Instead of start­ing in 1961, Marshall expanded the historical frame to sixty-five years prior and the origins of school segregation in the time of Oklahoma’s statehood in 1907. 149 Id. at 251 (Marshall, J., dissenting). By building out the record, Marshall emphasized discrimination’s systemic and dynamic character and its demonstrated capacity to persist through time. Echoing the approach taken by the dis­trict judge in Penick, Marshall observed how school segregation “pre­served” and “augmented existing residential segregation.” 150 Id. at 254. His point was that discrimination had a far longer trajectory and was more racially entrenched than the majority had acknowledged and that it also cross-fertilized other discriminatory systems. 151 See id. at 251. Thus, the injunction at issue in Dowell had not dislodged racial discrimination but merely “interrupted” it mid-“cycle.” 152 Id.; see also id. at 253–54 (citing Dowell v. Sch. Bd. of Okla. City Pub. Schs., 244 F. Supp. 971, 975–77 (W.D. Okla. 1965), aff’d in part, 375 F.2d 158 (10th Cir. 1967). Jus­tice Marshall’s dissent here resembles his dispute with Rehnquist’s majority opinion in Pasadena v. Spangler a decade earlier over the use of time. Spangler, 427 U.S. 424, 442–43 (1976) (Marshall, J., dissenting) (“According to the Court, it follows . . . that as soon as the school attendance zone scheme had been successful, even for a very short period, in fulfilling its objectives, the District Court should have relaxed its supervision over that as­pect of the desegregation plan.”). Removing the school district from federal supervision would only enable discrimination to return.

Marshall’s dissent in Dowell again highlights the contest over historical memory—not only over whether to remember but also how to interpret memory. While the majority embraced a narrower time frame that advanced a constrained interpretation of discrimination, Marshall con­structed a historical record that called for an acknowledgement of dis­crimination’s dynamism and adaptation. Because the “conditions” con­tinued to be ripe for the “threatened reemergence of one-race schools,” 153 Dowell, 498 U.S. at 251–53 (Marshall, J., dissenting) (“[T]o assess the full conse­quence of lifting the decree . . . , it is necessary to explore more fully than does the ma­jority the history of racial segregation in the Oklahoma City schools. This history reveals nearly unflagging resistance by the Board to judicial efforts to dismantle the City’s dual education system.”). Marshall favored a more robust standard for releasing the school district from federal oversight. 154 Id. at 252 (“I believe a desegregation decree cannot be lifted so long as condi­tions likely to inflict the stigmatic injury condemned in Brown I persist and there remain feasible methods of eliminating such conditions.”).

The Court’s static, linear understanding of discrimination resur­faced a year later in Freeman v. Pitts. 155 503 U.S. 467 (1992). Freeman addressed a highly tech­nical but important question of whether the district court could release its supervision over a school system in incremental stages before “full compliance had been achieved in every area of school operation.” 156 Id. at 490. The district had been under judicial supervision for the preceding twenty-year period. Id. at 471. Before Freeman, a district court’s decision to terminate oversight typically depended on whether school officials had achieved compliance across multiple facets of the system, including student assignment, faculty and staff assignment, facilities, curriculum, extracurricular activities, and transportation. 157 These factors are commonly referred to as “Green factors,” referring to the Supreme Court decision in Green v. New Kent County that first identified the components of a school system that would have to be remedied in order to eradicate segregation “root and branch.” 391 U.S. 430, 436–38 (1968); see also Keyes v. Sch. Dist. No. 1, 413 U.S. 189, 213–14 (1973) (holding that a school board must demonstrate “that its policies and practices with respect to school site location, school size, school renovations and additions, student attendance zones, student assignment and transfer options, . . . , etc., considered togeth­er . . . were not factors in causing the existing condition of segregation in these schools”). For a thorough discussion of school desegregation remedies, see generally John Leubsdorf, Completing the Desegregation Remedy, 57 B.U. L. Rev. 39 (1977). The reason for requiring systemic compliance prior to dissolution of any portion of a consent decree relates to the “intercon­nectedness” of these factors—that segregation in one aspect of a school system could have segregative impact on another, 158 See Green, 391 U.S. at 435–42; see also Brief for Respondents at 27–34, Freeman, 503 U.S. 467 (No. 89-1290), 1991 WL 521285. as the Court had recognized years earlier in Keyes. 159 Keyes, 413 U.S. at 208. Because of this synergistic relation­ship, withdrawing federal supervision incrementally could accelerate segre­gation throughout the whole system. Further, any segregation that fol­lowed dis­solution of a desegregation decree would not have the pre­sumption of unconstitutionality that existed under the prior, pre-unitary system. 160 See, e.g., Dowell v. Bd. of Educ., 782 F. Supp. 574, 577 (W.D. Okla. 1992) (ob­serving that upon dissolution of the desegregation decree and termination of the case, “the usual intent test” applies “for measuring the legality of school board actions,” rather than “the pre-dissolution, remedial effects test”). In­stead, plaintiffs would have to reestablish that such segre­gation resulted from discriminatory intent. 161 See, e.g., id. at 576–77 (stating that once a desegregation decree has been dis­solved, the school board’s actions should be evaluated “based on whether they are moti­vated by discriminatory intent and thus constitute a new equal protection violation”).

In holding that a school system’s unitary status could be declared piecemeal, the Court’s reasoning in Freeman again betrayed its limited understanding of the synergies of discrimination. The Court attributed segregation in the school system to significant shifts in the county’s racial demographics in the time since the district court’s 1969 order. 162 Freeman, 503 U.S. at 475–76. The Court emphasized that the percentage of black students in the schools had increased from 5.6% in 1969 to 47% nearly twenty years later. 163 Id. at 475. That “remarkable change[]” accounted for persistent segregation, not the school district’s prior discriminatory intent. 164 Id.

As in Dowell, however, the Freeman Court’s “inordinate focus on time clouded its ability to perceive segregation’s adaptive qualities and how easily it could spread across a school system.” 165 Boddie, Adaptive Discrimination, supra note 3, at 1295. What the Court missed was hiding in plain sight. The significance of the demographic changes was not the fact of increasing numbers of black students in the system but rather the racialized distribution of black and white students across the county system: The county’s northern half had become “predomi­nantly white” while its southern half had become “predominantly black.” 166 Freeman, 503 U.S. at 475.

As the Court accurately observed, this resulting segregation “com­pound[ed] the difficulty” of maintaining a desegregated system. 167 See id. But in diagnosing the source of this segregation, the Court ignored the possibil­ity that it was a vestige of the prior dual system, 168 See generally Elise C. Boddie, Racial Territoriality, 58 UCLA L. Rev. 401, 438 (2010) (explaining how race maps onto physical and spatial boundaries in the absence of conventional forms of discriminatory intent). as respondents had urged, and instead attributed it to supposedly neutral “demographic shifts.” 169 See Freeman, 503 U.S. at 476. Again, the influence of time and the correlative presumptions that time purges prior discriminatory motivation are evident. The Court pointedly observed that “[a]s the de jure violation becomes more remote in time and these demographic changes intervene, it becomes less likely that a current racial imbalance in a school district is a vestige of the prior de jure system.” 170 Id. at 496; see also Boddie, Adaptive Discrimination, supra note 3, at 1295. Justice Scalia in his concurrence was even more direct, stating, “[a]t some time, we must acknowledge that it has become absurd to assume, without any further proof, that violations of the Constitution dating from the days when Lyndon Johnson was President, or earlier, con­tinue to have an appreciable effect upon current operation of schools.” 171 Freeman, 503 U.S. at 506 (Scalia, J., concurring). The Supreme Court continued to project its narrow view of racial discrimination. In Missouri v. Jenkins, for example, the Court struck down a district court decision ordering salary increases to instructional and noninstructional staff in the Kansas City, Missouri School District for the explicit purpose of “attracting” white students from outside the school district. 515 U.S. 70, 92 (1995). The constitutional problem was that this remedy exceeded the scope of the violation. See id. While the district court in an earlier ruling found that Kansas City and the state of Missouri had segregated the local Kansas City schools, it rejected plaintiffs’ interdistrict claims against the surrounding suburban school districts for failure to prove an inter­district violation. See id. As a constitutional matter, the district court lacked authority either to order the redistribution of white students across district lines or even to order im­prove­ments to the Kansas City district in order to incentivize students to cross district lines. See id. at 76 (“Because it had found no interdistrict violation, the District Court could not order mandatory interdistrict redistribution of students between the [Kansas City Missouri School District] and the surrounding [suburban school districts].”). Compare id. at 100 (limiting remedial funding to once-segregated schools), with Dayton Bd. of Educ. v. Brinkman, 443 U.S. 526, 529 (1979) (upholding finding of system-wide segregation). But see Milliken v. Bradley, 418 U.S. 717, 744–45 (1974) (limiting multidistrict metropolitan desegregation when there is a failure to establish multidistrict violation).

Thus, as in Dowell, the Freeman Court invoked the passage of time to justify the termination of judicial oversight of school districts, emphasiz­ing the importance of restoring local control. 172 See Freeman, 503 U.S. at 489–90. The influence of Dowell and Freeman is apparent in the lower court decisions that followed. In these decisions, courts repeatedly emphasized the same “passage of time” con­siderations in terminating judicial supervision. 173 See, e.g., Monroe v. Jackson-Madison Cty. Sch. Sys. Bd. of Educ., No. 72-1327, 2010 WL 3732015, at *4 (W.D. Tenn. Sept. 24, 2010) (observing that “desegregation de­crees are ‘not intended to operate in perpetuity’” (quoting Bd. of Educ. v. Dowell, 498 U.S. 237, 248 (1991))); United States v. Bd. of Educ., 663 F. Supp. 2d 649, 652 (N.D. Ill. 2009) (observing the “twenty plus years of the Board’s commitment to the decree”); Hart v. Cmty. Sch. Bd., 536 F. Supp. 2d 274, 280–81 (E.D.N.Y. 2008) (evaluating “passage of time” in a motion to terminate judicial supervision and noting that “the last thirty years demon­strate substantial good-faith compliance over a long period of time”); S.F. NAACP v. S.F. Unified Sch. Dist., 413 F. Supp. 2d 1051, 1069 (N.D. Cal. 2005) (concluding that termina­tion of judicial oversight was entirely appropriate “[a]fter twenty-two years of judi­cial supervision and substantial compliance by the school district”); Hampton v. Jefferson Cty. Bd. of Educ., 102 F. Supp. 2d 358, 374 (W.D. Ky. 2000) (“Under Dowell and Freeman, ‘the phrase “to the extent practicable” implies a reasonable limit on the duration of . . . federal supervision.’” (quoting Coal. to Save Our Children v. State Bd. of Educ., 90 F.3d 752, 760 (3d Cir. 1996))); Davis v. Sch. Dist., 95 F. Supp. 2d 688, 696–97 (E.D. Mich. 2000) (observing time considerations in school desegregation doctrine and noting that “[t]he District has worked diligently throughout the course of the past three decades to adhere to the Court’s . . . directive”); Mills v. Freeman, 942 F. Supp. 1449, 1463 (N.D. Ga. 1996) (noting concerns about extending the court’s decree “in to perpetuity”); Stell v. Bd. of Pub. Educ., 860 F. Supp. 1563, 1580 (S.D. Ga. 1994) (“The hope of fuller compliance is insufficient to justify the court’s imposition of perpetual assignment orders.” (quoting Morgan v. Nucci, 831 F.2d 313, 324 (1st Cir. 1987))); cf. Dowell v. Bd. of Educ., 782 F. Supp. 574, 577 (W.D. Okla. 1992) (observing that granting relief from prior judgment to dissolve the desegrega­tion decree would “effectively leav[e] in force in perpetuity the pre-dissolution, remedial effects test”). Later studies con­firmed their segregative impact. 174 See Reardon, supra note 112, at 880 (citing multiple studies that unitary status determinations lead to resegregation). The Reardon study also found “that the districts released from court order were very similar to those not released in terms of their racial composition and segregation levels, suggesting that the process of release is not tightly linked to the success of the court order in producing integration.” Id. at 877. As one district court observed (with­out any apparent irony) upon declaring unitary status, “Even though segregation is ended forever, the Court must confront the reality that ab­sent the Decree some majority-black schools are likely to re-emerge.” 175 Hampton, 102 F. Supp. 2d at 373.

III. The Continuing Influence of Time

The previous Part described how the Court has used the passage of time to justify changes in school desegregation doctrine that transformed school segregation from an unconstitutional de jure system to a de facto status unrecognized by law. These contests over the meaning of time, which played out in struggles to define the scope of the historical record, also mapped onto important underlying debates about the kind of dis­crimination that equal protection would recognize. If courts conceived of school segregation as a systemic problem, they would extend the histori­cal record to capture its breadth. A longer record similarly helped to un­mask the trajectory of evasion and resistance that unfolded in response to Brown. Highlighting these past maneuvers provided a context for inter­preting later, segregative decisions by school officials as a vestige of dis­criminatory intent. Situating these later actions on a broader time contin­uum allowed courts to show a pattern of longstanding recalcitrance.

The Court’s decisions in Dowell and Freeman inverted this use of time. By emphasizing the length of time that school districts had been subject to judicial supervision, the Court painted a different picture of school segregation—not as the product of longstanding efforts by school officials to avoid integration but as the unavoidable result of private deci­sion­making and racially neutral governmental practices. 176 See Freeman, 503 U.S. at 496 (arguing that as time passes, “it becomes less likely that a current racial imbalance in a school district is a vestige of the prior de jure system”); see also Dowell, 498 U.S. at 248 (stating that a desegregation decree could be dissolved after “a reasonable period of time”). By giving segregation the veneer of inevitability, divorced from the historical record, the Court neutralized and then normalized school segregation.

This neutralization-normalization laid the foundation for the Court’s more aggressive turn in Parents Involved in Community Schools v. Seattle School District No. 1. 177 551 U.S. 701 (2007). While the Court’s decisions in Dowell and Freeman had in­voked concerns about the allocation of judicial power over school sys­tems and the resulting loss of local control, neither of those jus­tifications was available in Parents Involved. At stake was a deeper issue of whether two school systems with segregated public schools—one a for­mer de jure sys­tem that had been released from judicial supervision just seven years prior and the other a district that was segregated, though never formally by law—could use race to avoid entrenching patterns of residential segre­gation in student assignment. 178 Id. The case, which was brought by a white plaintiff, therefore, raised neither questions of local control nor federal judicial overreach, as both districts had voluntarily adopted the programs and wanted to keep them.

The Court held both policies unconstitutional on narrow tailoring grounds. 179 Id. at 748. Most remarkable was the plurality opinion, which equated the explicit uses of race for segregative purposes struck down in Brown with the race-conscious efforts by school officials in Parents Involved to promote voluntary integration. 180 Id. (“The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”). The plurality’s refusal to place the poli­cies in a broader historical context allowed it to focus on the districts’ use of race as the defining constitutional harm and to dismiss as inconse­quential the underlying segregation that the policies sought to address. 181 See generally Joel K. Goldstein, Not Hearing History: A Critique of Chief Justice Roberts’s Reinterpretation of Brown, 69 Ohio St. L.J. 791 (2008) (examining how the Par­ents Involved decision focused on the impropriety of race-based classifications rather than on the history of segregation in the United States). Parents Involved thus transformed Brown from a case that had been interpreted to challenge racial subordination in public schools to a decision about the harms of racial classifications that burdened white plaintiffs. 182 Id. at 794–96 (discussing how Chief Justice Roberts interpreted Brown’s holding as forbidding race-based classifications rather than segregation).

As previously mentioned, the Court’s focus on the passage of time is not limited to school desegregation cases. Recall the Civil Rights Cases, which relied on conceptions of time to conclude that racial discrimina­tion in places of public accommodations was not sufficiently related to slavery to justify the exercise of Congress’s enforcement powers under the Thirteenth Amendment. 183 See 109 U.S. 3, 25 (1883) (stating that the Thirteenth Amendment could not be used to prevent discrimination in places of public accommodation because there must come a time when a person freed from slavery “takes the rank of a mere citizen, and ceas­es to be the special favorite of the laws”). The Court’s 2013 decision in Shelby County v. Holder 184 133 S. Ct. 2612 (2013). relies on a similar time construct. 185 See Boddie, Adaptive Discrimination, supra note 3, at 1245. There the Court struck down a core provision of the Voting Rights Act—which had helped to block persistent, adaptive racial discrimination in voting—on the grounds that the nearly fifty-year old statutory remedy had not been framed to address “current needs.” 186 See Shelby County, 133 S. Ct. at 2628–52; Boddie, Adaptive Discrimination, supra note 3, at 1296–99. The Shelby County decision invoked the passage of time to deny the relevance of past voting discrimination to ongoing voting discrimination in the present. The same concern about the passage of time is at the root of what Professor Darren Hutchinson has described as the country’s perpetual “racial exhaustion” with reme­dies that address racial inequality. 187 Darren Lenard Hutchinson, Racial Exhaustion, 86 Wash. U. L. Rev. 917 (2009). The pervasive perception that racial wrongs are rooted too deeply in our past exacerbates social impatience and weariness with modern efforts to remedy racial discrimination. 188 See id. at 922–23.

Time considerations are evident in other areas of equal protection. In Regents of the University of California v. Bakke, the Court advanced the same causative, time-centered notion in striking down the use of race in a voluntary affirmative action program in student admissions. 189 438 U.S. 265, 307 (1978). The Court objected that the policy sought to remedy “societal discrimination,” which is “ageless in its reach into the past.” 190 Id.

These uses of time are deceptive. They lead to presumptions of clear delineations between past and present based on a “historically static con­ception of ‘discrimination.’” 191 Siegel, supra note 14, at 1113. In so doing, it becomes easy to assume that “bad” discrimination from the past has no relationship to the in­equality of the present. The temporal distance from these old practices—the desire to disassociate oneself from the past, universally condemned ra­cism—leads to the recasting of practices that have the same discrim­inatory effects as race neutral, while overlooking their subordinating consequences.


Our preoccupation with time has been consequential. In particular, it has allowed courts to overlook the dynamic and systemic nature of segregation and how it adapts to elude formal constraints set out by doctrine. Liberating ourselves from the constraints of time and expand­ing our historical record would allow us to probe the relationship be­tween past and present and to identify strategies that have been used over time to perpetuate racially subordinating conditions. This form of analysis would be useful not only in the context of racial discrimination cases but also for facilitating social understandings about the persistence of discrimination and its effects. Seeing that racial progress is subject to perpetual contestation could make us more alert to how our past is impli­cated in our present. 192 Id. Professor Siegel frames the central question:
“The body of equal protection law that sanctioned segregation was produced as the legal system endeavored to disestablish slavery; the body of equal protection law we inherit today was produced as the legal system endeavored to disestablish segregation. Are we confident that the body of equal protection law we inherit today is ‘true’ equal protection, or might it stand in relation to segregation as Plessy and its progeny stood in relation to slavery?”
Id. at 1114.

This Essay’s proposed use of historical inquiry to identify systemic patterns of racial discrimination, however, raises a series of practical questions. Chief among them is how a court would determine that a dis­criminatory system has been satisfactorily remedied. Is it certain that all racial inequality is necessarily a vestige of prior discriminatory intent? Or, more to the point, how would a court know whether the vestiges of prior intent have been remedied, justifying termination of judicial oversight? Must society be forever tethered to the past?

A full response to these questions is beyond the scope of this Essay, but it is possible to venture a few thoughts here. Some of these issues could be addressed by shifting burdens of presumption. For example, as in diagnosing a disease that manifests itself through ongoing symptoms, a long historical record of systemic racial inequality in a particular juris­diction could establish a rebuttable presumption of racial intent. Such presumptions might in turn incentivize public officials to break the cycle of discriminatory effects by taking more aggressive action against policies and practices that perpetuate inequality. Another possibility, of course, is to dispense with intent altogether and simply impose an affirmative obli­gation on public and private actors to root out racial inequality in juris­dictions with a significant prior history of racial discrimination.

Judicial inquiries here would focus on whether a system has achieved a stable equilibrium of racial equality within certain accepted (perhaps legislatively defined) parameters. As a practical matter, this would mean that a jurisdiction could not adopt policies that produce unequal effects and might be required to take steps to alleviate the discriminatory impact of policies imposed by government officials. Returning to the example of school segregation—a locality would be under an affirmative duty to re­dress segregation whenever it surfaces under this framework. It would re­tain local control of its schools as long as these conditions are satisfied but would be subject to enforcement measures if segregation resurfaced. 193 However, there are some useful comparisons to be drawn between this proposed scheme and Section 5 of the Voting Rights Act, which imposes a comparable preclearance remedy. 42 U.S.C. § 1973(c) (2012).

Of course, there are other concerns. 194 One issue is the evidentiary challenges of identifying the sources of discrimina­tion that originated long into the past. See Fiss, supra note 29, at 431 (noting that the “ev­iden­tiary difficulty of reconstructing the past limits the capacity to classify the effects in question as the ‘perpetuation of the past’”). Some might point to the potential strain on the judicial system and the risk that far-reaching inquiries could jeopardize the legitimacy of the courts’ factfinding processes and the delicate balance that judges have to strike as arbiters of their countermajoritarian powers. Id. Some might argue that the risks of possible windfalls for those who were not directly harmed by past wrongs—and the corresponding burdens on those who were not directly responsible—should limit systemic remedies that reach deep into the historical record. 195 See id. at 433–34 (noting the “inevitabl[e] error in identifying the victims of the past wrong” and the “margin of overinclusion in the class receiving the benefits of the cor­rective action”); id. at 434 (observing the risk that people would be “held responsible for the wrongs of another”). Tests and standards could be developed to determine which groups would bear the burden and the extent of that burden. 196 See id. at 429 (discussing the differential-impact theory, which does not take into account possible historical or causal explanations). On the other hand, it is possible that these tests and standards would just replicate the same racialized hierarchy of priorities that have dominated equal protection doctrine for the last several decades. At bottom, there­fore, there may be no satisfactory way to reconcile these tensions.

Returning to the purpose of this Symposium, what does all of this mean for the celebration of the life and legacy of Constance Baker Motley? Perhaps the best way to honor Motley is to acknowledge that the dynamic conditions of discrimination and racial subordination—the very conditions she fought during her lifetime—persist, and that redressing them will take considerable determination and patience. One can re­spect her legacy by recognizing the intergenerational nature of this chal­lenge, acting on it, and then passing on a more nuanced understanding of it to the generation that follows—all in the hope of achieving lasting ra­cial progress through time.