Equal Protection as Intentional Blindness
The Court that refused to see inequality … would be making the only kind of law that can be warranted outrageous in advance—law based on self-induced blindness, on flagrant contradiction of known fact.
Since the end of the civil rights era in the early 1970s, the emancipatory potential of the Fourteenth Amendment has been thoroughly undone. Today, its guarantee of “equal protection” no longer promotes reform but rather protects the racial status quo. This undoing has spanned four decades, with sharp plunges followed by ever-lower plateaus. These reversals have come in two principal areas: in how the Supreme Court evaluates claims of discrimination against nonwhites, an area governed by intent doctrine; and in how it regards affirmative action designed to ameliorate racial inequality, where the Court draws on colorblindness.
We find ourselves in a renewed period of rapid devolution. In the next few months and years, scholars will react to these changes, a few extolling and most bemoaning them, many focusing on distinct cases and some seeing overarching patterns. Yet almost all will accept as the foundation for their engagement the basic division in equal protection between intent and colorblindness. This chapter seeks instead to understand the present and near future by revising our understanding of the past. In this vein, it offers a wholly new conception of the relationship between discriminatory intent doctrine and colorblindness.
The pronounced tendency to regard these doctrines as fundamentally distinct is understandable, for not only do they operate in different domains, they seem to be mirrored inversions of one another. Colorblindness applies to affirmative action because it expressly uses race; intent doctrine sweeps up allegations of discriminatory treatment where race is almost never explicitly referenced. Colorblindness denies that the state’s purposes can be discerned; intent doctrine makes proof of malice indispensable. Colorblindness consistently imposes the most stringent form of scrutiny; intent cases always default to the most lenient form of constitutional review. Plaintiffs challenging affirmative action under colorblindness virtually always win; parties challenging discrimination under intent doctrine almost invariably lose.
Yet despite these dramatic dissimilarities, intent and colorblindness are profoundly connected. How could they not be? The contemporary versions of these doctrines were elaborated at the same time, by the same Justices, and to the same ultimate effect: to defeat challenges to, and remedies for, discrimination against nonwhites.
Moreover, for both doctrines, the basic method was and is the same: to deny that context matters. In declaring that only one form of intent counts—malice—discriminatory intent doctrine excludes most evidence of continued discrimination against nonwhites, whether rooted in history, contemporary practices, or social science. It thus blinds the courts to the obvious truth that forms of discrimination have evolved since the civil rights era, and no longer exclusively take the form of hooded bigotry. Likewise, in pronouncing that motives are unknowable and hence irrelevant, colorblindness similarly closes the courthouse doors to evidence of a changing context, in this case the rise of efforts by state actors to use race to break down inequality and to foster integration. Colorblindness fundamentally ignores the victories wrought by the civil rights movement itself, which moved the country from one seeking to enforce racial supremacy to one hoping for its eradication.1
We should no longer conceive of equal protection as divided between intent and colorblindness. Instead, we should understand it as unified under what might best be termed “intentional blindness.” Combining the names of the two doctrines, this portmanteau expresses the marrow of the Court’s racial jurisprudence—which seems intentionally blind to racial context.
This chapter first examines equal protection before the advent of colorblindness, when one doctrine—discriminatory intent—applied to both express and non-express uses of race. The chapter then details the heretofore unrecognized symbiosis between colorblindness and intent: colorblindness arose in response to intent doctrine, especially as applied to affirmative action; and in turn, colorblindness impelled intent doctrine toward a monocular focus on malice. It then traces the rise and application of malicious intent across the 1980s, reconsidering the actual operation of malice doctrine to offer a damning indictment heretofore missing from the literature. Virtually all of intent doctrine’s critics accept that today’s intent standard actually searches for illicit motives. This is evident in the two primary critiques they lodge. Either the critics reprove an exclusive focus on conscious animus that imperils attention to unconscious bias. Or they remonstrate that requiring proof of malice disables the courts from remedying entrenched disadvantage. In either event, they chide the Justices for not easily improving intent doctrine by updating its conception of cognition and broadening its focus from mindsets to structures. Yet in both cases, these critics assume that the malice test is in fact geared to searching for illicit motives.
Upon re-examination, however, the conservative Justices may have repeatedly exhorted proof of malice, but they showed no interest in parsing mental states. Instead, their demands seemed geared to closing the courthouse doors to evidence showing continued racial hierarchy. To make this point as powerfully as possible, this chapter asks how the malice test would resolve a challenge to Jim Crow segregation. It argues that the malice test would fail to arrive at the correct result even in a case like Brown v. Board of Education—the paradigmatic equal protection case of the twentieth century.
Finally, this chapter brings the analysis up to the present, demonstrating the consolidation of intent and colorblindness after the 1990s. This consolidation occurred in two senses: first, in terms of the emergence of a solid bloc of conservative Justices who repeatedly voted to ignore racial discrimination and to strike down affirmative action; and second, in terms of a striking convergence between the doctrines themselves.
A last introductory word: this chapter focuses on doctrine, but it is not a doctrinal article. Instead, it takes doctrine seriously in order to lay bare the racial politics of the conservative Justices who have reversed equal protection over the last four decades. This story could be told much more simply in terms of elections and judicial appointments. In this version, equal protection’s transmogrification since the 1970s follows most fundamentally from a backlash against civil rights manipulated by demagogic politicians, and resulted in the election of presidents, and in turn the appointment of Justices, hostile toward racial progress.2 Those Justices who consistently refused to find discrimination against nonwhites, or to support affirmative action programs, were prepared to reach those results without doctrine or evidence on their side. Crucially, though, they were also quick to craft doctrines and evidentiary standards that confirmed their favored outcomes, and it is here that the story becomes complicated, widely misunderstood, and critically important. True, malice doctrine and colorblindness were not engines for rolling back civil rights, so much as a dissimulating gloss. Over time, however, this gloss has hardened into a strong protective carapace. So long as colorblindness and discriminatory intent remain misunderstood as separate and plausible doctrines, it will be all but impossible to understand ongoing devolutions in the Court’s racial jurisprudence. Charles Black in 1960 described the refusal to see racial inequality as self-induced blindness. Colorblindness and intent doctrine are the bandages the Justices use to blind themselves to continued racism against nonwhites, and also to hide from public view their ferocious disfigurement of equal protection.
In the stock story regarding intent, the prime villain is a 1976 case, Washington v. Davis.3 In the 1980s, the leading treatise on constitutional law summarized Davis in the following terms: “In essence, Washington v. Davis announced that henceforth every lawsuit involving constitutional claims of racial discrimination directed at facially race-neutral rules would be conducted as a search for a bigoted decision-maker.”4 In so writing, Laurence Tribe accurately captured the state of intent doctrine at that date. But he was wrong to impute this structure to Davis, and wrong not once but twice.
To begin with, Davis did not require a search for a bigoted decision-maker, but instead envisioned a weighing of the totality of the circumstances. Rather than inventing intent doctrine, Davis helped formalize the Court’s basic approach to allegations of discrimination in place since Brown v. Board of Education struck down segregated schools in 1954. During the civil rights era, the Court evaluated the general purposes behind challenged government actions using a broadly inferential approach, rather than through a hunt for bigots. Sometimes this inferential process reflected little more than judicial notice of race relations. At other times, the Justices relied on a more focused examination of surrounding racial patterns, including through the invocation of social science. Whatever the combination, the point is that the Court did not demand direct proof of subjective mindsets. Instead, findings of discriminatory purpose reflected inferences drawn from the challenged action as well as the surrounding context—in a phrase popular with the Court, from the “totality of the circumstances.”5 Just so in Davis, where the Court held that “[n]ecessarily, an invidious discriminatory purpose may often be inferred from the totality of the relevant facts.”6
Next, Tribe erred in suggesting that, under Davis, intent applied to “claims of racial discrimination directed at facially race-neutral rules”—that is, that intent doctrine did not apply when the state expressly used race. On the contrary, up through Davis and just beyond, equal protection’s evaluative approach did not turn on whether the challenged government action expressly invoked race. Indeed, illustrating this, Davis itself presumed that the intent test applied to race-conscious state action. Immediately after announcing the need to show an invidious purpose, the Court explained: “This is not to say that the necessary discriminatory racial purpose must be express or appear on the face of the statute.”7 Davis thus assumed that the intent test applied to express uses of race, and clarified that it might also apply to race-neutral practices. By offering this clarification, Davis pointed back toward a basic fact that is now virtually forgotten: the Jim Crow cases expressly mandating segregation constituted the quintessential intent cases.
If Tribe’s treatise misrepresented Davis, it nevertheless correctly described intent doctrine as it stood in the late 1980s. To contrast the two distinct versions of intent doctrine, distinguish “contextual” from “malicious” approaches. Where contextual intent focused on motives only in the loosest sense and emphasized instead a broadly informed inferential approach to evaluating possible discrimination, today’s malice doctrine declares direct proof of injurious motives a prerequisite—and concomitantly declares contextual evidence irrelevant. Understanding the shift from contextual to malicious intent, in turn, requires understanding the strengths of contextual intent, and more generally the relationship between malicious intent and colorblindness.
Contextual intent, as its name implies, asked courts to consider the larger context in which challenged actions occurred. This allowed courts to discern racial discrimination even when it did not take the form of archetypal Jim Crow segregation. More strikingly, however, it also provided a ready means to distinguish oppressive uses of race from remedial ones—in other words, contextual intent provided a means to uphold affirmative action.
A year after Davis, the Court decided United Jewish Organizations of Williamsburgh, Inc. v. Carey,8 or UJO, involving a decision by the New York legislature to create two majority-minority districts in the Williamsburgh area of Brooklyn. UJO thus involved what Justice William Brennan characterized as the use of “race-centered remedial devices.”9 The decision was nearly unanimous, though various rationales were advanced. Two Justices applied Davis directly: “Under the Fourteenth Amendment the question is whether the reapportionment plan represents purposeful discrimination against White voters. Washington v. Davis.”10 Three others used an intent analysis, though without specifically citing Davis. Because the “plan represented no racial slur or stigma with respect to whites or any other race,” they concluded, “we discern no discrimination violative of the Fourteenth Amendment.”11 Justice Brennan concurred but hesitated to fully endorse the intent analysis, lest this suggest that merely rational review would suffice. For Brennan, even remedial uses of race carried sufficient risks to warrant a more searching, though not strict, review.12 Sifting through these rationales, a surprising result emerges: five Justices—and not just any five, but White, Stevens, Rehnquist, Stewart, and Powell, all of whom with the exception of White would soon support the rise of colorblindness—thought that race-conscious remedies should be treated just like every other use of race and evaluated for an intent to harm. To reiterate, in UJO those Justices poised to embrace colorblindness concluded that in the absence of any intent to harm, affirmative action was constitutional. What a brief, remarkable moment. In the first case to squarely address the constitutionality of a race-conscious remedy, the Court for the last time found affirmative action lawful by asking whether there was an intent to harm whites.
Colorblindness and Malice
UJO seemed to serve as a wake-up call to the racially conservative Justices. That decision warned (or from a different perspective promised) that race-conscious remedies, when reviewed for discriminatory intent, would be readily upheld. Betraying dissatisfaction with that result, the next year in Bakke five Justices voted to restrict affirmative action.13 Yet, likely reflecting their holding just the prior year, four avoided the constitutional question altogether. Standing alone, Lewis Powell experimented with a constitutional argument. Powell started by rejecting colorblindness, at least when defined as a per se prohibition on any use of race.14 In its place, though, Powell established the fundamentals of modern colorblind analysis: not automatic invalidity but its close cousin, the application of the highest level of constitutional hostility to all express uses of race—which is to say, to affirmative action.15
Though Powell initially attracted no other votes for this perversion of equal protection, two years later Stewart and Rehnquist exploited the opening when they dissented in a 1980 case upholding a minority set-aside in federal construction contracts.16 For the first time, conservative Justices declared that the “constitution is color-blind,” employing John Marshall Harlan’s famous aphorism from his dissent in Plessy v. Ferguson.17 Yet, this position, adopted by only two Justices, still seemed so bizarre that the majority opinion began: “As a threshold matter, we reject the contention that in the remedial context the Congress must act in a wholly ‘color-blind’ fashion.”18 It would take another nine years before a majority of the Justices consolidated behind colorblind logic as a basis for overturning an affirmative action program. In 1989 five Justices in City of Richmond v. Croson Co. used colorblindness to strike down a race-conscious remedy.19 Since Croson, colorblindness has become dominant on the Court, striking down remedial program after program, even if typically only with the votes of a bare majority of five Justices.
It is into the colorblindness timeline that one must place the origins of malicious intent. Just a year after Bakke, in Personnel Administrator of Massachusetts v. Feeney, five conservative Justices began to rework intent doctrine.20 Strikingly, they did so in direct reliance upon Powell’s colorblind reasoning—making Feeney, not Croson, the first case by a decade to assemble a majority behind colorblindness. Extending Powell’s analysis in Bakke, Feeney split equal protection into the separate domains now taken for granted, one governing affirmative action and the other discrimination against nonwhites. In turn, this schism contributed directly to the rise of the malicious intent rule requiring the nearly impossible proof of malice.
In Feeney, even the majority recognized that Massachusetts conferred on veterans “a well-nigh absolute advantage” in the competition for civil-service positions.21 For any available job, the state ranked applicants according to their score on a competitive exam but placed veterans above all others: the lowest-scoring veteran took his place above the highest-scoring nonveteran. As the Court observed, because 98 percent of veterans were male, “[t]he impact of the veterans’ preference law upon the public employment opportunities of women has thus been severe.”22 To challenge this system, Helen Feeney invoked the contextual intent approach formalized in Davis. She especially emphasized the concurrence by Stevens, who had urged attention to “objective evidence of what actually happened rather than evidence describing the subjective state of mind of the actor. For normally the actor is presumed to have intended the natural consequences of his deeds.”23
But the majority did not begin by addressing intent. Rather, Feeney used Powell’s equal protection story in Bakke to rewrite the Fourteenth Amendment in terms of a basic concern with the act of classification. “Most laws classify,” Stewart began, observing that equal protection generally required that distinctions among classes be subjected to no more than rational review—the standard asking only whether the government has a conceivable legitimate interest.24 “Certain classifications, however, in themselves supply a reason to infer antipathy. Race is the paradigm.”25 In this narrative, the simple use of race without more triggered constitutional suspicion. “A racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification.”26 Feeney pretended to reach back to Brown v. Board of Education for this radical proposition.27 Yet clearly the appropriate citation was Powell’s lone opinion in Bakke. Only a year after that case—and a full decade before Croson would see five Justices say the same to strike down an affirmative action program—Feeney assembled five votes for the rule that all uses of race, even if remedial, were “presumptively invalid.”
Only after announcing a mechanical distrust of any use of race did Feeney address intent doctrine. It did so by recasting intent doctrine as the flip side of colorblind doctrine’s automatic hostility to express uses of race: now, absent a reference to race, even government action that disproportionately harmed nonwhites would be presumptively constitutional. The express use of a racial classification became the crucial pivot. In effect, Feeney built on Bakke to bifurcate equal protection into two domains with two dramatically different presumptions of constitutionality.
In the first domain, all racial classifications merit the highest level of constitutional suspicion. Stewart cast colorblindness as skepticism of any “racial classification, regardless of purported motivation.” This put a universal spin on the point, as if colorblindness stood in opposition to every use of race whether propelled by invidious or benign motives, whether Jim Crow or affirmative action. But colorblindness—arising in the late 1970s—stood in opposition to Jim Crow only in its own mythos. In reality, opposition to every use of race meant hostility toward race-based remedies. Powell introduced colorblind logic in Bakke in 1978 primarily to evade the thrust of UJO. Stewart endorsed anticlassification reasoning in Feeney and then immediately seized upon colorblindness to condemn affirmative action in employment in 1980 and 1981. Once five Justices came to this position in Croson, colorblindness went on to strike down myriad remedial measures but applied in only one case involving the mistreatment of nonwhites—and that over the strenuous objections of Clarence Thomas and Antonin Scalia, two of the most vociferous proponents of colorblindness.28 The colorblind claim to oppose any government use of race is misleading, for in practice colorblindness opposes race-conscious remedies and nothing more.
Meanwhile, in the second domain, facially race-neutral laws merit almost complete constitutional deference. But there is obfuscation here too. After Feeney, the “neutral” laws presumed to be constitutional unless malice could be shown never involved government action that helped minorities; those laws were examined skeptically under colorblindness. And these “neutral” laws never involved the supposed mistreatment of whites. There have been no cases alleging the use of race-neutral devices to discriminate against whites. Instead, the fruitless search for invidious intent has only been required in challenges to government practices imposing disproportionate harm on nonwhites. Almost complete deference to government action applies only when minorities, not whites, allege racial discrimination.
Thus, in one domain of equal protection, extreme suspicion reigns, and in the other, virtually complete abdication prevails. Abstractly, these two domains correspond to facially racial and facially neutral government actions. In reality, they correspond to affirmative action and disproportionate harm to nonwhites. Affirmative action receives fatal-in-fact review; harm to nonwhites is treated with lenient indifference.
Feeney, of course, is not commonly known as a colorblindness case, but rather as the case that first announced, notoriously, that intent required a showing that the state pursued the challenged action “at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects.”29 That is, Feeney defined “intent” as acting not just in full awareness of impending harm but out of a desire to cause such harm. Reva Siegel, among others, understands the Court to have demanded proof of conscious antipathy, something approaching malice: “[I]n Feeney, the Court asked plaintiffs to prove that legislators adopting a policy that would foreseeably injure women or minorities had acted with the express purpose of injuring women or minorities—in short, a legislative state of mind akin to malice.”30
In reconsidering contemporary intent doctrine, we must recognize its close symbiosis with colorblindness. Colorblindness arose in response to contextual intent, in particular to its propensity to uphold affirmative action. In turn, the malice test developed as a mirror to colorblindness, deeming all facially race-neutral government practices presumptively innocent and imposing an egregiously high burden to prove otherwise. Bakke elevated affirmative action into a singular threat to equality. Feeney rendered discrimination against vulnerable groups a marginal problem. Once embraced as a way of thinking about equality, colorblindness both legitimized and impelled a highly restrictive approach to proving discrimination against nonwhites.
That said, Feeney remained a transitional case. Marking an abrupt rupture from past decisions, it endorsed the colorblind reasoning first proposed by Powell the year before in Bakke, and dramatically narrowed the meaning of intent to a mindset akin to malice. Yet harkening back to prior intent cases, Feeney used contextual evidence to decide whether or not the requisite intent existed. The next year, the Court would begin to impose on plaintiffs not only the onerous need to prove malice, but the impossible challenge of doing so through direct evidence regarding the thinking of identified individuals. It is important to note that Feeney did not take that second step, but remained open to inferences drawn from a totality of the circumstances. This bears emphasis not only because it confirms the settled nature of intent doctrine’s previous method, but also because Feeney’s continued reliance on contextual evidence marked the emergence of a remarkable double standard. Beginning the next year, conservatives on the Court would disparage and reject all but direct evidence to prove animus against nonwhites. But pursuant to Feeney, they would continue to draw contextually based inferences of innocence to defeat claims of discrimination.
Feeney reversed the logic of earlier cases, which had held that in some cases disproportionate harm to nonwhites was so extreme as to compel an inference of invidious intent. Feeney declared instead that a self-evidently legitimate purpose defeated evidence of discrimination.31 Ignoring the possibility of mixed motives, the majority declared that the presence of a single, obvious nondiscriminatory rationale insulated government action by foreclosing the possibility that discriminatory purposes also played some role. A defensible preference for veterans, Feeney reasoned, “necessarily compels the conclusion that the State intended nothing more than to prefer ‘veterans.’ Given this finding, simple logic suggests than an intent to exclude women from significant public jobs was not at work in this law.”32 In dissent, Thurgood Marshall offered the obvious retort: “That a legislature seeks to advantage one group does not, as a matter of logic or of common sense, exclude the possibility that it also intends to disadvantage another.”33 The malice cases that followed Feeney would demand that plaintiffs directly prove intent, as if they were deeply concerned with the actual—and necessarily multiple—motivations of individuals. But contrarily, these cases also drew on Feeney’s new rule that the presence of a single, legitimate government purpose, discerned generally from the context surrounding the challenged policy, would warrant an unimpeachable inference of innocence. The malice test developed as a rule banning contextual proof of discrimination, while at the same time allowing contextual exonerations.
1980–87: Malice Applied
The malicious intent test as a bar on claims of mistreatment came into full force almost immediately. This is not to suggest that, after Feeney, the Court demanded direct proof of animus in every case alleging discrimination against nonwhites. On the contrary, whether in terms of affirmative action or racial discrimination, a high level of doctrinal instability marked equal protection during the 1980s. Nevertheless, equal protection has since consolidated around the malicious intent test, warranting a special focus on the logic of the cases that developed this approach.
The year after Feeney, the Court heard Mobile v. Bolden.34 Mobile, Alabama, employed citywide elections for choosing commissioners. Given prevalent racism, although African-Americans constituted 35 percent of the city’s population, they were consistently outvoted—a pattern that resulted in no blacks being elected to the city commission since its establishment in 1911. The trial and appellate courts held that Mobile engaged in unconstitutional discrimination by maintaining an electoral system that virtually guaranteed the perpetual exclusion of African-Americans from local government—and, indeed, that operated to ensure the defeat of even those White politicians seen as more sympathetic to the Black third of the population.35
In reversing the trial and appellate courts, Mobile first reiterated Feeney’s logic regarding exculpatory inferences. The majority seemed reassured that the at-large electoral structure applied across the country.36 Mobile then weighed each bit of evidence adduced to show racial mistreatment, though piecemeal and seriatim rather than holistically. This technique of evidentiary disaggregation, which would become routine in malice cases, elicited a strong objection from Davis’s author, who correctly pointed out that it violated the spirit of the contextual approach: “By viewing each of the factors relied upon below in isolation … the plurality rejects the ‘totality of the circumstances’ approach we endorsed in … Washington v. Davis and Arlington Heights, and leaves the courts below adrift on uncharted seas with respect to how to proceed on remand.”37
The most lethal innovation came, however, in measuring each piece of the disassembled evidence for what it could prove regarding the existence of a subjective mindset of malice. Mobile neatly disposed of every probative iota by declaring it individually incapable of showing the actual thinking of an identified state actor. The district court had stressed continued Jim Crow patterns in Mobile. It documented at length the exclusion of African-Americans from municipal boards, the relegation of blacks to menial government jobs, the preservation of overwhelmingly White fire and police departments, and allegations of widespread police brutality that included complaints of mock lynchings.38 Waiving away this evidence, the opinion rejoined: “[E]vidence of discrimination by White officials in Mobile is relevant only as the most tenuous and circumstantial evidence of the constitutional invalidity of the electoral system under which they attained their offices.”39 Clarifying exactly why such evidence was “tenuous and circumstantial,” Mobile explained that the district court failed “to identify the state officials whose intent it considered relevant in assessing the invidiousness of Mobile’s system of government,” adding that “the actions of unrelated governmental officials would be, of course, of questionable relevance.”40 The case took a similar stance toward “the substantial history of official racial discrimination in Alabama.”41 It declaimed: “[P]ast discrimination cannot, in the manner of original sin, condemn governmental action that is not itself unlawful. The ultimate question remains whether a discriminatory intent has been proved in a given case. More distant instances of official discrimination … are of limited help in resolving that question.”42 Finally, Mobile