Printer Friendly

The canary-blind Constitution: must government ignore racial inequality?

I

INTRODUCTION

For centuries, coal miners took canaries into the mine to alert them that atmospheric conditions were dangerously toxic. (1) If the canary showed signs of distress, it was time to get out. (2) The miner's canary has been used as a metaphor to describe the signal provided by observed disparities between racial groups along socioeconomic dimensions. (3) Black people experiencing disproportionate disadvantage suggests the presence of systemic underlying "toxic" conditions that correlate with race. The metaphor can be applied to infant mortality rates--a statistic long recognized as indicating a population's overall health and well-being. (4) A black infant in America today is up to three times more likely than a white infant to die in her first year of life. (5) Additionally, despite landmark gains in civil rights in the 1960s, the racial gap in infant mortality has consistently increased since then. (6) The disparity widens further as comparisons proceed up the socioeconomic ladder; that is, the ratio of black-to-white infant mortality rates is larger among the middle class than among members of lower socioeconomic classes. (7) In fact, infant mortality rates for college-educated black women are higher than those for white women without a high school diploma. (8) Moreover, for white women, infant mortality rates are higher for recent immigrants than for women born to several-generation families, whereas for black and other women of color, the opposite is true. (9)

The causes of racial disparities in infant mortality and related perinatal conditions, such as premature birth and low birth weight, are more complex and uncertain." (10) Immediate causes of infant mortality include congenital abnormalities, pregnancy complications, respiratory distress syndrome, and sudden infant death syndrome (SIDS). (11) As to why such maladies correlate with race, studies have found an association between perinatal problems and a range of socioeconomic factors disproportionately experienced by black Americans, such as living in segregated neighborhoods marked by poverty, unemployment, violent crime, and single-parent households. (12) Such conditions tend to intensify physical and emotional stress in pregnant women and contribute to maladaptive coping strategies, including smoking, substance abuse, and poor nutrition. (13) Low socioeconomic status is also associated with lack of access to prenatal care and infant parenting education. (14) Studies also find consistently that, even controlling for insurance status and income, minorities receive lower-quality care than whites. (15) One explanation is that racial bias or cultural ignorance on the part of health-care providers affects the quality of prenatal and other medical care provided to black mothers. (16) Still other studies point to beliefs common among some black communities that can deter accessing medical care, such as mistrust toward health-care providers. (17) More recent studies suggest that stress from discrimination, and the perception thereof, can also have a harmful effect on perinatal outcomes. (18) Time and generations compound these factors. That is, the longer a female lives in economically and psychologically stressful conditions, the more likely she will bear children with perinatal health problems; (19) in fact, the conditions under which a woman lives can adversely impact the health of not one but two generations. (20) For example, if a woman, due to life circumstances, bears an undernourished child with low birth weight, that child is more likely as an adult to bear an underweight child. (21)

On a positive note, research into racial disparities in infant mortality has helped to identify conditions that cause infant mortality generally, which has helped to bring infant mortality rates down for all women. At the same time, however, the racial gap--the ratio of black to white rates of infant mortality-- has increased, prompting some experts to describe racial disparities in infant mortality as a persistent "puzzle" and "mystery." (22) Despite uncertainty regarding underlying causes, broad consensus prevails that racial disparities in infant mortality is a concern warranting government attention. (23)

Such attention, however, may well be unconstitutional according to Supreme Court interpretation. (24) Equal protection doctrine increasingly demands that government institutions be "colorblind"; that is, give no effect to race in governmental decisionmaking. (25) Thus, it has been suggested that state action intended to benefit racial minorities arguably constitutes "suspect" racial discrimination subject to strict scrutiny even when implemented through raceneutral means. (26) Scholars continue to debate variations and implications of that claim, (27) and constitutional doctrine continues in the direction of colorblindness. (28)

This article follows the logic of the colorblindness trend further. It asks whether, for equal protection purposes, a state necessarily discriminates by race when it investigates a racial disparity, identifies its root causes, and addresses those causes without regard to race in purpose or means. Prior scholarship has considered a related, but different question. (29) The prior question was whether state action that is intended to benefit racial minorities through race-neutral means is racially discriminatory. (30) The current question is whether state action that is intended to benefit people regardless of race through race-neutral means is racially discriminatory if the state's decision resulted in any part from a prior investigation into the causes of a racial disparity.

To illustrate, consider again disparities in infant mortality. Assume that a state's investigation into such disparities reveals that black people disproportionately lack access to prenatal care. Three remedial responses suggest themselves, each illustrating a different role for racial motivation. First, a state could subsidize prenatal care for low-income black women in order to reduce the racial gap in infant mortality. Second, a state could subsidize prenatal care for low-income women of any race in order to reduce the racial gap in infant mortality. Third, a state could subsidize prenatal care for low- income women of any race in order to reduce infant mortality, period.

The first option would be subject to strict scrutiny because it employs a racial classification--black women--as means. The second option would arguably be subject to strict scrutiny pursuant to prior scholarship in the area (31) because, though it employs race-neutral means, it is arguably motivated by a racially discriminatory purpose: reducing a racial gap. The third option, which is the concern of this article, might also be subject to strict scrutiny even though it employs race-neutral means--low-income women--and pursues a race-neutral purpose--reducing infant mortality generally. The basis for applying strict scrutiny to this third option is that a racial motivation--to investigate the racial gap in infant mortality--played a causal role in bringing about the ultimate policy. In that sense, the policy to reduce infant mortality was because of, not merely in spite of, race. (32) If the state must be race indifferent in its motivations, then a state's pursuit of nonracial goals through nonracial means might still constitute racial discrimination if attention to a racial disparity motivated the investigative process that produced the ultimate policy.

Examining whether investigating racial disparities constitutionally taints any use of the information gained therefrom is important for at least two reasons. First, it might reveal that a significant amount of contemporary state action is unconstitutional. Every state and the federal government investigates racial disparities in an attempt to understand and address their root causes. (33) Health disparities alone require a significant investment of government attention and resources. (34) According to the Center for Disease Control and Prevention, stark racial disparities persist across a wide range of health conditions, such as cancer, heart disease, diabetes, and HIV-AIDS. (35) The racial gap in health, moreover, which has been documented for several decades, is an issue that has consistently raised concerns and calls for action from both political parties. (36) Beyond health, racial disparities in education, (37) employment, (38) wealth, (39) marriage rates (40) and nonmarital births, (41) juvenile delinquency (42) and punishment, (43) crime commission (44) and victimization, (45) arrest, (46) conviction, (47) and sentencing, (48) among others, are the subject of investigation by social scientists, policymakers, and organizations affiliated with the federal government, as well as every state government, regardless of which political party is in power. Even the government's role in simply compiling data on racial disparities may be problematic to the extent it encourages policymakers, public and private, to rely on that information in designing remedial policies. Indeed, the collection of racial data by itself, including by the Census Bureau, arguably constitutes the unlawful use of racial classifications inconsistent with the Equal Protection Clause. (49) If colorblindness is a constitutional imperative, then government cannot gather racial information about its citizens or respond to racial disparities because it cannot "see" race. A truly colorblind Constitution would require the government to ignore the canary in the mine.

Second, inquiry into this constitutional question may reveal that the relationship between race-consciousness and equal protection is more complicated than a literal understanding of colorblind constitutionalism admits. Equal protection doctrine tends to keep with the tide of prevailing societal views. (50) Societal consensus that government ought generally to be colorblind exists alongside substantial consensus that government may, and arguably should, investigate and ameliorate the root causes of racial inequality. This suggests that equal protection doctrine, as currently understood, can accommodate both propositions. Can it, and if so, how?

The article makes two principal claims about the government's authority to investigate and address the causes of racial disparities. First, a plausible interpretation of contemporary equal protection doctrine would subject to strict scrutiny and probably invalidate any state-sponsored attempts to reduce racial disparities. Second, and alternatively, a better interpretation of current doctrine could permit a state to act in response to racial disparities without discriminating by race, provided that the racial motivation is limited to investigating the causes of the disparities.

Developing the first claim, part II follows colorblind constitutionalism toward its logical end, which would invalidate or at least subject to strict scrutiny any state action directed toward addressing racial disparities in any context. The point of this part is neither to endorse nor deride an argument for complete colorblindness. Rather, the point is to take the colorblind trend of Supreme Court precedent seriously and demonstrate its implications for government-sponsored efforts to address the racial disparities that persist across a broad range of social and economic indicators. Part II.A synthesizes current equal protection doctrine as applied to racial classifications and other racially discriminatory state actions, delineating between those racial motivations that are per se invalid and those that are suspect and thus invalid unless they survive strict scrutiny. Part II.B applies these principles to the example of racial disparities in infant mortality, beginning with the most clearly suspect response to such disparities--reserving prenatal care for minority women--to the least suspect, awarding such care without any regard to race except in the initial impetus to investigate the racial disparity. Although the case for subjecting the latter to strict scrutiny is not irrefutable, it is within the range of plausible applications of colorblind principles already reflected in equal protection doctrine.

Part III develops the second claim, namely, that the investigative use of race need not be constitutionally suspect. Part III.A describes a range of contexts in which the government's attention to race is widely considered legitimate without warranting strict scrutiny. Those contexts include trial adjudication, legislative fact-finding, and research by government agencies. A common theme among these presumably legitimate uses of race is that race is taken account of for investigative, evidentiary, or otherwise informational purposes. Part III.B advances a positive account of why a state's investigation of a racial disparity and its use of the information obtained from the inquiry need not involve an illegitimate or racially discriminatory purpose. Applied to racial disparities in infant mortality, the state may legitimately inquire into such disparities, identify their root causes, and seek to ameliorate those causes on a race-neutral basis without needing to satisfy strict scrutiny.

The final two subparts of Part III flesh out the argument in two respects. Part III.C explains two corollaries of the investigative use of race; namely, that it may have the effect of benefiting white people and harming black and other people of color. Identifying these corollaries should make clear that the investigative use of race is not simply a form of affirmative action through race-neutral means. Part III.D considers whether this article's defense of using race to investigate racial disparities has implications for more controversial race- conscious state actions, such as disparate-impact liability for employment practices, electoral districting, police investigations, and state-sponsored efforts to remedy so-called "societal discrimination."

II

TOWARD COLORBLIND CONSTITUTIONALISM

A. Illegitimate and Suspect Uses of Race

Current equal protection doctrine establishes four propositions regarding the use of race by any state or federal actor, including legislatures, executive officials, administrative agencies, and courts. First, a law is per se unconstitutional, whether it employs a racial classification or is race-neutral on its face, if it is motivated by certain illegitimate racial assumptions, beliefs, or intentions. Second, any other law that employs a racial classification or which is administered with a discriminatory purpose is constitutionally suspect and, accordingly, invalid unless, under strict scrutiny, the government proves the use of race is narrowly tailored to further a compelling government interest. Third, this rule of suspectness applies to facially neutral classifications that are adopted with a purpose to discriminate against a racial minority. Fourth, a facially neutral law that is adopted with a purpose to discriminate in favor of a racial minority is arguably suspect, but the point remains unsettled.

At the core of impermissible racial classifications are those that exclude or segregate black people based on ideologies of white supremacy or racial animosity. Once the Supreme Court reached this position, most famously in Brown v. Board of Education, (51) it struck down all laws that segregated or excluded black people without subjecting such laws to strict scrutiny. (52) The Court then moved to invalidate, as per se illegitimate, racial classifications premised on certain unacceptable stereotypes, overbroad generalizations, or empirically unwarranted assumptions. (53) The peremptory challenge of prospective jurors, for example, was initially invalidated only if it reflected a prosecutor's belief that blacks were unfit for jury service in every case, (54) but in Batson v. Kentucky (55) and its progeny, (56) the Court decided that any peremptory challenge based on an assumption about a juror's likely perspective based on race was illegitimate because such an assumption is irrational and reflects the "very stereotype the law condemns." (57) A similar view of stereotypes has also informed the Court's skepticism toward drawing electoral districts based on race, describing the practice as reflecting the stereotype that people of a certain race "think alike, share the same political interests, and will prefer the same candidates at the polls." (58) Another illegitimate motivation is what the Court has called "simple racial politics." (59) By this, the Court seems to envision some sort of motivation to pass laws for the benefit of one's own racial group simply because the group has gained sufficient political power. (60) Finally, the Court has repeatedly condemned, as per se unconstitutional, "racial balancing"--the pursuit of proportional racial representation for its own sake. (61) Thus, if evidence demonstrates that a racial classification or other racially motivated state action is based on an illegitimate assumption or belief, racial politics, or a purpose to achieve racial balance, then the state action is unconstitutional per se, without subjecting the state action to strict scrutiny analysis.

Two kinds of racially motivated state actions are suspect regardless of which race is benefited or burdened by the action. The first are all racial classifications (other than those described above that are invalid per se). A racial classification refers to a law that expressly includes race as an element such that the government is required or authorized to take account of race in administering the law. (62) Despite disagreement on the Court for over a decade about whether racial classifications intended to benefit minorities should be subject to the same strict scrutiny as classifications designed to harm minorities, the Court decided in Richmond v. Croson (63) to treat both classifications the same, explaining that "the standard of review under the Equal Protection Clause is not dependent on the race of those burdened or benefited by a particular classification." (64) A second kind of state action that is subject to strict scrutiny regardless of which race is benefited or burdened is when a state discriminates by race in administering a law even if the law is race neutral on its face. (65)

Also suspect is a law that is race-neutral on its face and is administered without regard to race, but which is adopted for the purpose of discriminating against a racial minority. For example, if, for the purpose of reducing access by racial minorities to institutions, a government employer adopts a job qualification, (66) a public university implements an admission standard, (67) or a state enacts a voting requirement, including the shape of an electoral district, (68) such state action is presumptively unconstitutional.

Some uncertainty surrounds the constitutional implications of a law that is race-neutral in content and operation, but that was adopted for the purpose of benefiting minorities. The Supreme Court has analyzed the use of race-neutral means only for the purpose of benefiting minorities in the context of electoral districting. The Court subjected a majority-minority district to strict scrutiny in Shaw v. Reno, seemingly treating such districts as equally suspect as those drawn for the benefit of whites. (69) In Miller v. Johnson[TM] however, the Court announced a unique test for when a majority-minority district is suspect. The Court held that an electoral district drawn for the benefit of a racial minority is only suspect if race played a predominant role in shaping the district. (71) If, however, race is merely one among other factors of equal or greater influence, then a majority-minority district will not be evaluated under strict scrutiny. (72) In contrast, the Court will employ strict scrutiny whenever a state uses race- neutral means to burden a racial minority, including in drawing an electoral district, even when race was merely one among many factors. (73)

With respect to other contexts in which race-neutral means might be used to benefit racial minorities, the Court has not addressed the issue, but some justices have suggested in dicta that such policies would not be suspect. For example, in Parents Involved in Community Schools (74) Justice Kennedy indicated that he would join the Court's four more-liberal justices to permit, without strict scrutiny, the use of race-neutral policies to create racially integrated schools. (75) And in the context of government construction contracts in the late 1980s, a plurality of justices, including Justice Scalia, suggested that the use of race-neutral criteria to increase the number of contracts awarded to minority-owned firms would not be suspect. (76)

A more recent case suggests, in contrast, that the Court would treat race- neutral policies designed to benefit racial minorities as equally suspect to policies designed to harm minorities. In Ricci v. DeStefano, the Court held that a city's discarding of test results for firefighter promotions because the results would disparately promote whites, constituted discrimination against the white applicants in violation of Title VII. (77) The city's actions were arguably race-neutral in that they did not involve selecting particular black firefighters for promotion based on their race, but rather would presumably have involved redesigning the test and allowing all applicants for promotion to retake it. Although the Court ruled that discarding the test results was discriminatory under a statute, it is difficult to see why the Court would not also find it discriminatory under the Constitution. The Court intimated as much, (78) and Justice Scalia reinforced the sentiment emphatically. (79)

Despite the uncertainty over race-neutral policies designed to benefit racial minorities, the logic of current doctrine firmly supports subjecting such policies to strict scrutiny. (80) The Court has held that strict scrutiny applies to all racial classifications regardless of which race is burdened or benefited because, the Court insists, all racial distinctions are inherently suspect, even those designed to benefit minorities for ostensibly benign purposes. (81) The Court has also held, in earlier cases, that adopting a race-neutral policy with a discriminatory purpose is as suspect as an express racial classification. (82) Although it is true that those cases involved race-neutral policies designed to harm a racial minority, the Court's insistence that express racial classifications to benefit minorities are equally suspect strongly suggests that the justices would view race-neutral policies to benefit minorities with similar skepticism.

B. Colorblind Implications of Current Doctrine

Returning to disparities in infant mortality, the question is whether governmental investigation of such disparities renders suspect any resulting action directed at the causes of the disparities. Assume a state investigates the cause of racial disparities in infant mortality, discovers that blacks are less likely to receive prenatal care, and further determines that prenatal care reduces infant mortality. May the state pass legislation that provides prenatal care on a race-neutral basis? What if, by the time the legislation is adopted, lawmakers have no racial purpose behind their goal of reducing infant mortality?

Current doctrine could be interpreted to preclude such action or at least subject it to strict scrutiny. To see how, consider the following three laws, beginning with the most constitutionally problematic and concluding with the investigative use of race at the center of our inquiry.

Potential State Responses to Racial Disparities in Infant Mortality:

1. Prenatal care for black women to reduce the racial gap in infant mortality.

2. Prenatal care for all women to reduce the racial gap in infant mortality.

3. Prenatal care for all women to reduce infant mortality.

The first law above would be subject to strict scrutiny. By expressly targeting black women for prenatal care, the law employs a racial classification. Under current doctrine, all racial classifications are subject to strict scrutiny.

And strict scrutiny would most likely invalidate the law. Strict scrutiny requires the government to prove that the law is necessary or narrowly tailored to achieve a compelling state interest. The two most persuasive defenses of the law would likely fail. First, the state could claim that reducing infant mortality is a compelling interest. Surely it is, but the difficulty is that the Court would likely deny that favoring black women for prenatal care is necessary to reduce infant mortality. To the contrary, by limiting the benefit to black women, white women and other women of color would be left unaided.

The second potential defense of the prenatal care law for black women could be an argument that it compensates for the effects of racial discrimination. The Court has endorsed remedying racial discrimination as a compelling interest. Presumably, racial disparities in infant mortality reflect at least to some degree the consequences of historical and contemporary discrimination against black people. Allocating prenatal care to black women plausibly makes up for the detrimental conditions that black people have experienced as a result of discrimination that contributes to infant mortality.

The problem here is that, although the Court recognizes remedying racial discrimination as a compelling interest, it requires that the discrimination be identified "with particularity." (83) It is not enough to cite generalized statistical disparities--even if they plausibly result from historic discrimination, or what the Court calls "societal discrimination." (84) The state would need to identify, as much as practicable, the instances and perpetrators of discrimination and how that discrimination has contributed to black infant mortality. Such discrimination, however pervasive and intergenerational, is largely undocumented. It cannot be identified with the particularity required to satisfy strict scrutiny.

The second law, prenatal care for all women to reduce the racial gap in infant mortality, would arguably also be subject to strict scrutiny. The law is race-neutral on its face, but it is adopted with a discriminatory purpose--to benefit black women. Strict scrutiny applies to race-neutral policies if they are adopted with a purpose to discriminate against a racial minority and, as explained previously, such scrutiny logically applies to policies designed to benefit minorities. Intending to reduce the racial gap in infant mortality necessarily intends to benefit black women more than white women. To appreciate this point, consider if the goal were to increase the racial gap. That would most likely be viewed as discriminatory. The formal symmetry of the Court's equal protection doctrine suggests that it would view reducing the gap as equally discriminatory for the purpose of triggering strict scrutiny.

Whether this law would survive strict scrutiny is less clear than the previous law that employs an express racial classification. Laws that employ race-neutral means are considered more tailored under strict scrutiny than laws that employ racial classifications. The problems with the prior law, however, plague this one as well. For reasons similar to those concerning the first law listed above, the Court would likely conclude that purposely favoring black women is neither justified by a compelling interest nor necessary to achieve any such interest.

Finally, consider the third law listed above--prenatal care for all women to reduce infant mortality. This law is race-neutral on its face, administered without regard to race, and, at the time it was adopted, the state had no purpose to discriminate by race. At first glance, it is constitutionally unassailable. Recall, however, that a racial motivation played a causal role earlier in the legislative process, namely, the state's concern over the racial disparity in infant mortality motivated it to investigate potential causes, which led the state both to identify lack of prenatal care as contributing to infant mortality and to provide prenatal care to all women as a remedy. The question is whether that earlier racial motivation constitutes a discriminatory purpose attributable to the prenatal care law.

Whether this law involves a discriminatory purpose is debatable, an issue taken up in the next part at greater length. There is, however, a plausible argument that it does. To the extent the Court would view an intention to reduce racial disparities as a discriminatory purpose, then investigating racial disparities in order to address their underlying causes plausibly also qualifies as pursuing a discriminatory purpose. Consider, for example, the federal government's policy that calls for "working] with communities to reduce and eliminate health disparities between non-minority and minority populations experiencing disproportionate burdens of disease, disability, and premature death." (85) If the government's investigation into racial disparities is motivated, in any part, by the goal of reducing them, then such a purpose is plausibly characterized as a discriminatory purpose.

If the racial motivation behind investigating the racial disparity in infant mortality were a discriminatory purpose, there would still be the question of whether it would be legally attributable to the adoption of the prenatal care law if, as this hypothetical assumes, the legislators voting for the prenatal care law do not have a racial purpose. Does the fact that the racial purpose dropped out of the legislative process by the time the law was adopted preclude the law from being racially discriminatory?

The passage of time between the racial purpose and the law's adoption would not necessarily save this law. The Court has made it clear in striking down race-neutral laws adopted for discriminatory purposes that the time elapsed since the law was adopted does not necessarily support its constitutionality. In Hunter v. Underwood, for example, the Court invalidated an almost century-old Alabama constitutional provision that disenfranchised persons convicted of crimes of moral turpitude on the ground that the provision's original purpose was to disenfranchise blacks. (86) The Court rejected the state's argument that it had since retained the disenfranchising provision for legitimate purposes, stating that "we simply observe that its original enactment was motivated by a desire to discriminate against blacks on account of race and the section continues to this day to have that effect." (87)

In defense of the prenatal care law, the law in Hunter was adopted with a discriminatory purpose. With the prenatal care law, in contrast, the state no longer had a racial purpose by the time the law was adopted. This distinction is relevant but not dispositive. The Court appears to analyze purposes that are attenuated from the laws they affect as a question of proximate causation. (88) If a discriminatory purpose proximately causes a law's adoption or other state action, then that action will likely be treated as based on the discriminatory purpose. A recent case involving interpretation of a law against employment discrimination reflects this analysis. In Staub v. Proctor Hospital (89) the Court held that a discriminatory purpose by a supervisor, carried out through negative evaluations of an employee, and which proximately caused a later supervisor to terminate the employee, made the termination discriminatory even though the terminating supervisor did not know of the discriminatory purpose behind the negative evaluations. (90) Hunter also supports this analysis. There, the Court treated as irrelevant the state's claim that it had retained the disenfranchising law for nonracial reasons, suggesting that the retention, even if race-neutral in purpose, was tainted by its causal relationship to the discriminatory purpose behind the law's original enactment. (91)

Staub and Hunter indicate that if the state acts with a racially discriminatory purpose, and that action proximately causes the state's later adoption of a law, then that law can be considered based on the discriminatory purpose--even if the state did not act with such a purpose at the time the law is adopted. Thus, with the prenatal care law, if the motivation originating the investigation into the racial disparity was a discriminatory purpose, and if that investigation proximately caused the adoption of the law, then the law may be considered based on a discriminatory purpose even though the lawmakers adopting the law were not motivated by that purpose.

Under strict scrutiny, the law would probably be invalidated. Although reducing infant mortality is probably compelling, the goal of reducing the racial gap in infant mortality is probably not. The gap is plausibly caused by past discrimination that has caused black people to live in more "toxic" socioeconomic circumstances than whites. But the difficulty, again, is that to constitute a compelling interest, the past discrimination being remedied must be identified "with particularity." (92) Unless the state can trace the racial gap in infant mortality to identified discrimination, the Court would likely characterize the past discrimination at issue as "societal discrimination," the remedying of which the Court has rejected under strict scrutiny. (93) Indeed, the Court may well view the desire to reduce the racial gap as "racial balancing," a per se unconstitutional goal. (94) Thus, any attempt to remedy the racial gap in infant mortality would be vulnerable to invalidation. Whatever social and economic toxins cause the disproportionate rate at which black infants die are presumptively off-limits to government attention. The canary's distress must be ignored.

The foregoing analysis represents a logical extension of current equal protection doctrine's conceptualization of what constitutes a racially discriminatory purpose. The Court need not go so far, and the fact that many government institutions presume authority to reduce racial disparities suggests that the Court may identify a basis for upholding such policies. One approach the Court could take is to subject policies that address racial disparities through race-neutral means to strict scrutiny, but to uphold them if justified on grounds of remedying societal discrimination. (95) Alternatively, equal protection doctrine could be interpreted to permit government to investigate and address the root causes of racial disparities without having to satisfy strict scrutiny. Let us turn to that interpretation.

III

THE INVESTIGATIVE USE OF RACE

This part argues that a state can investigate the causes of racial disparities and address those causes without triggering strict scrutiny. Supreme Court precedent, read carefully, does not subject all racially motivated state action to strict scrutiny. Rather, strict scrutiny applies only if a challenger proves that a law employs a racial classification or that the law, if race-neutral, is adopted or administered with a discriminatory purpose. A discriminatory purpose requires that the state intends the law to have an adverse or beneficial effect on a racial group. The investigative use of race does not, without more, include a discriminatory purpose. Rather, it involves investigating racial disparities for the purpose of gaining information and, if the information is worth addressing for nonracial reasons, using the information without a racially discriminatory purpose.

To lay the groundwork for distinguishing investigative from discriminatory uses of race, part III.A describes various contexts in which governmental decisionmaking takes account of race without attracting demands, including by courts, that such decisionmaking be subject to strict scrutiny. If societal and judicial tolerance of the racially motivated state actions described below is consistent with current doctrine, then such actions, at least ostensibly, are not discriminatory. Part III.B develops the argument as to why they are not.

A. Apparently Nondiscriminatory Uses of Race

Race is routinely used by the government in a variety of contexts for investigative, evidentiary, or otherwise informational purposes without being subject to strict scrutiny. Such contexts include trial adjudication, legislative fact-finding, and government agency research. In the adjudicative process, trial courts frequently take race into account in ways that affect trial outcomes without the influence of race on the decision-making process being considered discriminatory or suspect. Perhaps the least controversial use of race is as evidence in adjudicating claims of intentional, racial discrimination. In such cases, the race of the party claiming to have been discriminated against, the race of the alleged discriminator, and often the race of third parties, such as other employees or applicants, is admitted as evidence for and against an inference of discrimination on the part of the defendant. (96) In such cases, race is not only accepted without being subject to strict scrutiny, it is sometimes required as part of the elements of a party's claim or defense. (97) This is especially evident in the adjudication of systemic or "pattern or practice" claims. Whether brought by the government or as a class action, such claims effectively require the plaintiff to present statistical evidence of racial disparities to prove her claim. (98) The defendant is also permitted, and effectively encouraged, to present her own racial statistics--ones that tend to counter an inference that the defendant engaged in discrimination. (99)

Discrimination cases challenging affirmative action policies also involve evidentiary uses of race. For example, Abigail Fisher recently sued the University of Texas and part of her evidence was racial. She pointed to the use of race in the University of Texas's admission policy and, furthermore, she emphasized the number of blacks and Latinos that are admitted by the university's alternative top-ten-percent policy. She also introduced the fact that she is white. (100) The Court has never suggested that the introduction of such racial information should be subjected to strict scrutiny notwithstanding the fact that the plaintiff provided that racial information to facilitate adjudication of her claim. (101)

Likewise, in disparate-impact claims under Title VII, the plaintiff must present evidence that an employer's hiring policy results in a significant racial disparity in order to shift the burden to the defendant-employer to show that the policy is job related. Evidence that a policy has a racial impact may also be offered by an employer as a defense to a claim of intentional discrimination. In Ricci, for example, the Court held that the city would need a strong basis in evidence that it would have faced disparate-impact liability if it had used the discarded test results in order to defend against the claim of intentional discrimination by the applicants who scored well on the test. (102) Presumably, the strong basis in evidence would include documentation regarding the test's racially disparate impact as well as its tenuous connection to job-related skills. Thus, far from being suspect, the presentation of racial data is required in persuading a court to dismiss a discrimination claim brought by disappointed applicants who scored well on a discarded test.

Justice Scalia's concurrence in Ricci is also noteworthy. Scalia was one of the staunchest proponents of colorblind constitutionalism on the Court and his concurrence in Ricci suggested that disparate-impact liability may violate equal protection doctrine by encouraging employers to alter hiring policies with a "racial thumb on the scale[]." (103) Yet, in the same opinion, Scalia endorsed the use of racial statistics for investigative purposes. He acknowledged that the racially disparate impact of an employer's policies could be used as an "evidentiary tool" in determining whether the employer acted with discriminatory intent. His objection was to using the racial impact of the test to establish employer liability rather than merely to support a rebuttable inference of discrimination.

Even in cases that do not involve allegations of racial discrimination, race has been admitted in the adjudicative process without any suggestion by any court that its use is suspect. For example, a crime victim, witness, or police officer may testify that a defendant resembles the observed perpetrator of the charged crime and may rely on race as a factor in explaining the resemblance. (104) Racial evidence has also been admitted to prove other facts, such as, in a paternity suit, whether a man is in fact the biological father of a child. (105) The attention to race in these cases is permitted to have a causal influence on a court's ultimate decision and yet neither the admission of racial evidence nor a court's reliance thereon is considered racially discriminatory.

Nor does the Court limit the investigative use of race to the judiciary. The Court has approved and even encouraged legislatures to use racial data for fact- finding purposes without suggesting that such use is subject to strict scrutiny. Consider affirmative action cases. The Court requires the legislature or other governmental body to document the compelling interest that purportedly justifies the preferential policy. If the government's interest is remedial, that is, to remedy prior discrimination, the Court requires that such an interest be demonstrated "with particularity," including by statistics of racial disparities that support an inference of prior discrimination. (106) For example, the Court upheld a federal race-based construction contract set-aside in Fullilove v. Klutznick (107) because Congress had made legislative findings that documented, through race-based statistics, evidence of nationwide racial discrimination in the construction industry. (108) In Croson, (109) although the Court invalidated the race-based set-aside program, which required a minimum of (30) percent of the amount spent on each contract to be paid to "Minority Business Enterprises," the Court did so because of the racial quota used to award contracts, not because the city took race into account in documenting the need for the program. In fact, the Court explained that the set-aside would more likely have been upheld if the city, when deciding whether to engage in affirmative action, had relied on more-particularized race-based statistical evidence of discrimination in the local construction industry. (110) Similarly, if a state pursues a nonremedial interest through affirmative action, such as the educational benefits of a racially diverse student body, the Court requires race-based evidence that having a critical mass of minority students produces pedagogical benefits and that using a racial preference is necessary to achieve those benefits. (111) The Court has never expressed any concern that the use of race to substantiate the government's interest is itself discriminatory and subject to strict scrutiny.

Ricci is also instructive for its support of the legislative use of race during the information-gathering stage of designing an employment test. By treating the discarding of test results with a disparate impact against blacks as discriminating against whites, (112) the Court employed a fairly strong colorblind understanding of discrimination under Title VII. Nonetheless, the Court seemed to accept, and even endorse, attention to the racial impact of a test during the test's design stage prior to its administration.
   Title VII does not prohibit an employer from considering, before
   administering a test or practice, how to design that test or
   practice in order to provide a fair opportunity for all
   individuals, regardless of their race.... We hold only that, under
   Title VII, before an employer can engage in intentional
   discrimination for the asserted purpose of avoiding or remedying an
   unintentional disparate impact, the employer must have a strong
   basis in evidence to believe it will be subject to disparate-impact
   liability if it fails to take the race-conscious, discriminatory
   action. (113)


At two points in the foregoing statement, the Court approves an employer's attention to race. First, the Court's assurance that Title VII does not prohibit an employer from considering whether a test is fair to individuals regardless of race must include the employer's consideration of whether a test has a racially disparate impact. Otherwise, the statement is trivially true. It goes without saying that an employer may consider a test's fairness in nonracial respects. The Court's assurance is only meaningful if it is referring to race-based considerations. Second, the Court's requirement that an employer have a strong basis in evidence of potential disparate-impact liability before discarding a test's results would presumably include a strong basis in evidence that the discarded test would have a racially disparate impact. The employer could not be subject to disparate-impact liability unless the test had a racially disparate impact. Thus, for an employer to discard a test's results lawfully, it must take account of race.

Executive agencies also collect, analyze, and use racial data. The most prominent of such practices is the Census, whose collection of racial demographic information is largely uncontroversial. Indeed, skeptics of colorblindness have cited the Census as an example of how absolute colorblindness is not a plausible interpretation of the Constitution because such an interpretation would invalidate the data-gathering methodology of the Census. (114) Beyond merely collecting racial data, countless governmental institutions, state and federal, as well as private institutions at the behest of government, routinely collect and analyze racial disparities across a broad range of contexts in an effort to diagnose their causes. (115) Such practices are widely accepted and, in fact, the government often requires various organizations to keep track of racial data, such as employers, educational institutions, and medical researchers. (116) Indeed, it is no exaggeration to observe that millions of hours are spent every year by researchers and policymakers at all levels of government, including public universities--and in a wide variety of private organizations, often with government funding--investigating racial disparities in contexts such as health, family, education, employment, criminal justice, and virtually all areas of the civic, economic, and social life of the nation. The societal acceptance of such broad-scale activity suggests that the activity is not racially discriminatory, at least not in the doctrinal sense that would require satisfying virtually-always-fatal strict scrutiny.

How the courts would analyze such investigative uses of race is difficult to assess; there has been scant litigation challenging such practices, a fact that itself suggests their unobjectionable status. The attention to race by the Census, however, has not gone completely without criticism. One court case, for example, challenged the Census as racially discriminatory, (117) but the court ultimately upheld the use of race by the Census Bureau on the ground the collection of demographic information is not in and of itself discriminatory. (118)

Supreme Court precedent also supports the collection of racial data and the use of such data for government decisionmaking. In Tancil v. Woolls, the Court affirmed a District Court ruling upholding the designation of race in divorce decrees. (119) The Supreme Court's opinion was per curiam and lacked explanation, but the District Court's opinion that was upheld reasoned that "the designation of race, just as sex or religious denomination, may in certain records serve a useful purpose, and the procurement and compilation of such information by State authorities cannot be outlawed per se." (120) Furthermore, the District Court explained, "[T]he securing and chronicling of racial data for identification or statistical use violates no constitutional privilege." (121)

In United States v. Armstrong, the Court cited with approval racial statistics published by the United States Sentencing Commission and relied on those statistics in deciding the case. (122) Black defendants challenged their prosecution for crack distribution on the ground that the federal prosecutor allegedly selected them for prosecution because of race. They introduced evidence in support of their allegation sufficient to convince the trial court to order discovery from the prosecution. (123) When the prosecution refused to comply, the District Court dismissed the case. (124) The Supreme Court reversed the trial court, reinstating the prosecution and holding that the defendants had not presented adequate evidence of similarly situated white crack dealers who were not prosecuted. (125) In concluding that the defendants had failed to demonstrate a likelihood of racial discrimination, despite evidence that all the defendants charged by the prosecutor's office were black, the Court cited the Sentencing Commission for statistics suggesting that crack distribution was committed disproportionately by blacks. (126) Judicial decisions that discriminate on the basis of race are subject to strict scrutiny, (127) yet the Court did not express any need to justify its decision to reinstate the prosecution because of race under strict scrutiny.

Armstrong is interesting for several reasons. First, the Court accepted the introduction of race-based evidence comparing white and black crack dealers charged by the federal prosecutor, faulting the defendants for not presenting more precise and reliable evidence of such. Second, the Court at least implicitly approved the legitimacy of the Sentencing Commission's compilation of data correlating race with crime commission--data that, in turn, incorporated conviction records containing racial designations. Third, the Court's reliance on race in reaching its decision implicitly approves a decision-making process--its own--that was influenced by attention to race without, presumably, discriminating on the basis of race.

This section has revealed a range of contexts in which government decisionmaking, whether by courts, legislatures, or governmental agencies, takes account of race in the process of determining facts and relies on those facts in making decisions. The next part develops a doctrinal justification for why certain race-based decisionmaking need not be subject to strict scrutiny.

B. Investigating Racial Disparities without Discriminating by Race

A state's investigation into the causes of a racial disparity, and the state's use of the information gained from the investigation in addressing those causes, need not be discriminatory nor subject to strict scrutiny. To most persuasively test this claim, assume a broad meaning of discriminatory purpose under current doctrine, namely, the goal of disparately impacting individuals or groups based on race, even if pursued through race-neutral means and, going beyond settled doctrine, even if race-neutral means are employed to benefit racial minorities.

The investigative use of race involves two stages: investigating the causes of a racial disparity and using the race-neutral information gained for race- neutral purposes. Regarding the first stage, a state's investigation into the causes of a racial disparity would not involve a discriminatory purpose if the state's motivation does not include a purpose to reduce the disparity or otherwise to affect people differently based on their race. If the inquiry's purpose were only to understand the causes of the disparity, it would be solely informational. The intent involved would be one of curiosity and concern, that is, a purpose to diagnose the cause of the disparity. Such a purpose is distinct, and it need not imply that the state intends to reduce the disparity.

For this distinction to be sustained, it should be acknowledged that, for the state's investigative motivation to count as entirely investigative, and not at all discriminatory, it must not include a precommitment to reduce the racial disparity regardless of what the investigation discovers. If a state's purpose to investigate underlying causes of a racial disparity were simply a means for advancing a purpose to reduce the racial disparity, then the investigation would arguably be tainted by its accompanying discriminatory purpose.

A second qualification is that the state's motivation to investigate a disparity must not be based on an illegitimate assumption or belief. A law motivated by an illegitimate belief or assumption is per se unconstitutional. (128) If an illegitimate belief played a substantial role in the state's investigation, and the investigation proximately caused the ultimate adoption of the law, then the law would arguably be unconstitutional per se. Assume, for example, that a state's investigation into a racial disparity in which minorities are worse off than whites is based on the illegitimate assumption that minorities should be worse off because they are innately inferior. The state expects that whatever is causing the disparity must be useful and worth reinforcing. Alternatively, assume that the state's investigation into the disparity is premised on the illegitimate assumption that any policy that disproportionately benefits white people, even unintentionally, is per se objectionable because all white people are unfairly privileged. Such inquiries, although investigative in nature, would be based on illegitimate assumptions and would arguably invalidate any use of the information proximately caused by the investigation.

A colorblindness advocate might object to the prenatal care law to reduce infant mortality for all women on the ground that, when the state earlier investigated the racial disparity in infant mortality, it improperly assumed that racial groups should always be proportionally represented in all contexts. That assumption may, in turn, reflect either a belief that any disparity is objectionable in itself or that any disparity must be directly caused by intentional discrimination. Such beliefs might be characterized, respectively, as an illegitimate interest in racial balancing or as an unwarranted belief that direct discrimination is always responsible for any disparity that exists. Alternatively, one might object that the state's investigation into a racial disparity is illegitimate simply because it classifies people by race and gives effect to race. It violates the proposition, in the words of the first Justice Harlan, that "our Constitution is color-blind" and does not permit the state to know the race of its citizens. (129) Concern with racial disparities improperly gives relevance to race in contravention of colorblind constitutionalism's premise that race is irrelevant.

Although some concerns over racial disparities would probably not be legitimate, some are. The Court has not clearly defined what beliefs about race are illegitimate, but the examples in case law of illegitimate beliefs either assume a racial group is inferior or posit generalizations about a racial group that are inaccurate, demeaning, or improperly essentializing. (130) A state's assumption that racial disparities might, though not necessarily, have an underlying cause worth addressing need not include a denigrating or essentializing belief about race.

To the contrary, a concern over racial disparities can be premised on the irrelevance of race. If race is truly meaningless, then statistically significant racial disparities suggest that something other than biology or chance is systematically at work. And, if it is legitimate to suspect that racial disparities may reflect harmful underlying causes, it should be legitimate to determine what those causes are. For instance, an investigation might uncover recent discrimination, conscious or unconscious, which colorblindness advocates accept as objectionable and worth correcting. Even if discrimination is not the immediate cause, discrimination may be an indirect cause of some harmful condition that disproportionately burdens racial minorities. And even if discrimination is not discernible as a cause at all, a harmful condition that systematically affects people of one race more than another is worth identifying in order to help all people who might be affected by that condition. It should thus be legitimate under such circumstances for a state to inquire into the cause of a racial disparity to determine if anything untoward is at play.

Assuming that a state's investigation into a racial disparity is legitimate, the next question is what, if anything, may the state do with the information gained from the inquiry? One possibility is that the inquiry identifies racial discrimination as the immediate cause of the disparity. In such a case, addressing the identified discrimination could satisfy strict scrutiny, (131) so it might not matter whether or not strict scrutiny is applied. But what if the immediate cause is something other than discrimination, such as a harmful socioeconomic condition with unknown preceding causes? May a state, after discovering that condition, seek to redress it on a race-neutral basis without having to meet the demands of strict scrutiny?

Ignoring the investigative process that revealed the harmful socioeconomic condition and assuming that the legislature genuinely wants to address the condition regardless of its racial effect, then no racial discrimination would occur. The law would simply involve the use of race-neutral means to address a race-neutral condition for race-neutral reasons. If the law is to constitute discrimination, it must be because of its causal relationship to the earlier inquiry that was motivated by the racial disparity. Even assuming that the state's prior investigation proximately caused the legislature to adopt the race-neutral law, the prior investigation did not include a racially discriminatory purpose. Rather, the state's purpose was only to diagnose the causes of the disparity, not to reduce the disparity regardless of its causes. The investigation's influence on the law would thus neither be suspect nor subject to strict scrutiny. Nor did the prior investigation involve an illegitimate assumption and thus its influence on the law would not per se invalidate the law.

The foregoing analysis could apply to racial disparities in infant mortality. First, a state may investigate such disparities if its interest is not based on an illegitimate assumption and if it does not include a racially discriminatory purpose, such as a precommitment to reduce the racial disparity regardless of what the investigation reveals. If, for instance, a state is concerned that the racial disparity in infant mortality might reflect discriminatory treatment by healthcare providers, (132) mistrust on the part of black women inhibiting seeking medical care, (133) or certain harmful socioeconomic conditions disproportionately burdening black women, (134) then the state may investigate the disparity to determine whether these or other factors are causing it.

Second, the state may take the information gained from the investigation and decide whether the underlying conditions causing infant mortality warrant redress regardless of the redress's racial impact. Assume, for example, that a state investigates racial disparities in infant mortality and discovers that lack of access to prenatal care causes infant mortality. If the state decides that prenatal care should be subsidized for low-income women in order to reduce infant mortality, it may do so without having to satisfy strict scrutiny, provided the policy it adopts is genuinely designed to benefit such women without regard to race and uses race-neutral means.

Beyond the issue of infant mortality, the investigative use of race could justify a range of other governmental decisionmaking that takes account of race. The adjudicative processes that relied on race, described in part III.A, may be understood as taking account of race for investigative, evidentiary, or other informational purposes. For example, when courts admit evidence of racial disparities in discrimination claims, the attention to race is to investigate facts in dispute, such as whether the defendant acted with a discriminatory intent against the plaintiff. Once that fact is determined, the court's ultimate judgment can be based on the facts in dispute rather than the race of the parties or other actors. In a nondiscrimination case, such as a criminal prosecution for theft, race may be admitted as part of eyewitness testimony identifying whether the defendant committed the crime. Once that fact is determined, either way, the decision to convict or acquit is based on whether the defendant committed the crime, and not on the defendant's race. In the context of research into racial disparities by scientists, by policymakers in governmental agencies, or by private organizations whose research is used by the government, such investigations need not trigger strict scrutiny if the intentions behind the investigations, and the government's use of the knowledge gained from the research, are both legitimate and nondiscriminatory.

C. Corollaries: Helping Whites and Harming Blacks

Two corollaries of the investigative use of race should help to clarify the argument, especially for skeptics. Some might suspect that the investigative use of race is simply a form of "race-neutral" affirmative action--a way to help racial minorities while avoiding strict constitutional scrutiny. Asking whether the theory necessarily benefits racial minorities tests such skepticism. If it does, then it may well be a form of discrimination for the benefit of racial minorities. If, however, the investigative use of race is not racially discriminatory, then presumably it could have the effect of disproportionately benefiting white people, and even harming black people.

The first corollary is that the investigative use of race can indeed have the effect of benefiting white people disproportionately. Consider, for example, that whites are more likely to commit suicide than blacks. (135) Why? Assume that an investigation reveals that black Americans tend to have more extended family and community relationships. (136) Such relationships likely reduce the extent to which blacks experience the kind of isolation that could contribute to suicide among people experiencing depression or mental illness. (137) Informed by this investigation, a state can take steps to encourage people of any race at risk of suicide to seek community support. Such a state policy may well benefit whites more than blacks, but if that racial benefit is not the purpose behind the policy, it should not be subject to strict scrutiny.

The second corollary is that racial disparities involving greater harm to blacks than whites need not be addressed if the underlying causes are determined to be unobjectionable in nonracial terms. For example, assume that a law school at a public university admits a disproportionate number of whites compared to blacks from its applicant pool. (138) Assume further that the law school investigates the disparity and determines that giving weight in the admissions process to the Law School Admission Test (LSAT) has a disproportionate effect against black applicants because their scores are on average lower. (139) May the law school give less weight to the LSAT in the admission process in order to reduce the racial disparity in admissions? To avoid strict scrutiny, the investigative use of race supports giving less weight to the LSAT only for nonracial reasons, not for the purpose of increasing minority admissions. If examining the LSAT reveals that it does not serve the law school's legitimate, nonracial admission goals, then the law school may give it less weight or stop using it altogether. If, however, the current use of the LSAT serves the law school's legitimate purposes, then the investigative use of race does not support discontinuing its use. Indeed, if the law school reduced reliance on the LSAT because of its racial impact, then that action would arguably be subject to strict scrutiny precisely because of its purpose to reduce the racial disparity between black and white matriculating students. (140)

D. More Controversial, Race-Conscious State Actions

This final section considers whether the distinction between investigative and discriminatory uses of race can help to assess other race-conscious state actions that have generated significant controversy. The state actions discussed below include disparate-impact liability, electoral districting, race-based policing, and efforts to remedy the effects of societal discrimination. The circumstances in each scenario differ in some respects from the investigative use of race discussed above, and thus the use of race might not as readily qualify as purely investigative. Generally, the more difficult it is to separate a state's attention to race from its ultimate decision, the more difficult it is to accept that the use of race is investigative only. The discussion merely sketches how the following uses of race might be analyzed, leaving to future work a more thorough examination.

1. Disparate-Impact Liability

Congress, through Title VII of the Civil Rights Act of 1964, (141) requires an employer to alter a job qualification or other hiring practice if the practice has a significant racially disparate impact and the employer cannot demonstrate that the practice is valid, that is, job related. (142) Is Title VII subject to strict scrutiny on the ground that Congress is acting with a discriminatory purpose by imposing a duty on employers to reduce the racial impact of their practices? (143)

The investigative use of race might justify disparate-impact liability. (144) First, when a court adjudicating a Title VII claim investigates the disparate impact of a job qualification, it is arguably acting for legitimate, informational purposes. Its concern, and that of Congress, is that since race does not affect one's ability to perform a job, a qualification that has a racial impact may either reflect intentional discrimination or some other factor correlated with race that does not predict job performance. If the investigation determines that the qualification is not valid, then the courts, implementing congressional will, have a race-neutral reason--the qualification does not predict job performance--to require the employer to modify the qualification in order to make it better select qualified applicants, regardless of race. If, however, the court's investigation determines that the qualification is valid, then Congress would lack a race-neutral reason to alter the qualification and thus the qualification should be upheld.

A few difficulties face the foregoing analysis. First, is it legitimate for Congress to investigate employment practices for their job validity only when they have an impact based on race--or other Title VII traits, such as national origin and sex--instead of requiring that all job qualifications meet a certain standard of validity? Is Congress thereby improperly prioritizing racial disparities over other effects of employment practices? The defense to this charge must be that qualifications with racially disparate effects raise greater concerns than qualifications with disparate effects based on nonracial factors. Such concern is arguably justified on the ground that a qualification with a racially disparate impact might reflect intentional racial discrimination, a practice that equal protection doctrine and antidiscrimination law recognize as more objectionable than other, nonsuspect forms of discrimination. Even if the disparate impact is not the result of intentional discrimination by the employer, the irrelevance of race, again a premise of colorblind antidiscrimination law, suggests that racial disparities reflect some systematic problem impairing access to employment unnecessarily--if the qualification turns out to be invalid--for people of all races--even if some racial groups are disqualified at a higher rate.

A second objection to disparate-impact liability goes to the degree of validity that a job qualification must satisfy. If a qualification must meet a high level of validity, that is, predict job performance with a high degree of empirically demonstrable confidence, then it might be inferred that Congress's goal is to reduce the racial impact of job qualifications, permitting only those qualifications that are so useful as to outweigh the harm of their racial impact. It would be a question of fact as to whether Congress, through Title VII, is using the racial impact of qualifications only for investigative purposes or whether Congress has the goal of reducing the racial impact of qualifications that would otherwise be unobjectionable.

A third concern with disparate-impact liability arises from the final stage of the proof framework under Title VII. If an employer demonstrates that a job qualification is valid, the plaintiff can still prevail by proving that there is an alternative employment practice, which the employer refuses to adopt, that would equally serve the employer's needs with less of a racial impact. (145) This step is in tension with the investigative use of race because it would require an employer to alter an employment practice in favor of another, not because the other would better serve the employer's needs (only equally), but rather because the other practice would have less of a racially disparate impact. The only way to avoid this discriminatory implication is if the employer's refusal to adopt the alternative practice indicates that its existing practice is actually motivated by a discriminatory purpose. If so, then requiring the employer to change the practice is simply correcting identified discrimination. That in fact seems to be how the Court views the purpose of this final stage of the disparate-impact proof framework. (146) To conclude that an employer discriminated in adopting its disparately impacting qualification, however, the circumstances should support such an inference from more than the racial impact of the qualification alone. Otherwise, a job qualification with a racially disparate impact could always be attributed to intentional discrimination, which is a move that the Court would reject.

2. Electoral Districting

The Court does not subject to strict scrutiny a state's majority-minority electoral district if the state relied on race in drawing the district, provided race was not a predominant factor. (147) The investigative use of race might justify this approach if the state's use of race is both legitimate and nondiscriminatory. The Court explains that a state may legitimately use race as one among many other factors, including districting principles like compactness, contiguity, and respect for political subdivisions, and other demographic information, in order to identify "communities of shared political interest" for inclusion in a district. (148) Perhaps this process could be understood, first, as using race to investigate the location of a community of interest and, second, drawing a district around that community of interest for nonracial purposes.

The foregoing account raises two difficulties. First, if the process of relying on race to identify communities of interest is investigative only, then why does the Court subject to strict scrutiny the districts that are identified through a process that uses race as a predominant factor? Why is that not equally investigative? Perhaps it is, but it might be objectionable on the ground that it involves an illegitimate stereotype. The Court seems to believe that a state's assumption that race is more predictive of shared political interests than all other factors, including economic status or shared political subdivision, is an impermissible stereotype about the relationship between race and political preference. (149) It improperly assumes that people who share the same race "think alike, share the same political interests, and prefer the same candidates at the polls," (150) whereas people who share other, nonracial traits in common do not share political interests as predictably.

The second difficulty is that, if a state's use of race as one of many factors is a legitimate use of race for investigative purposes, then why does the Court subject other decisionmaking that relies on race as just one of many factors to strict scrutiny? If a public university, for example, relies on race as just one factor in the admission process alongside other criteria such as grades, test scores, experiences, and reference letters, the admission policy would be subject to strict scrutiny. (151) Why is the university not, like a state drawing electoral districts, using race investigatively, merely seeking to identify applicants with distinctive perspectives and experiences with race as one of many factors that helps it to do so?

A possible distinction between the university's affirmative-action program and the state's electoral districting process is that the university is using race for the purpose of selecting applicants, whereas the electoral district drawers are using race-neutral criteria, such as census tracts and other geographic information, to draw around communities of interest after such communities have been identified by using race. This distinction is questionable, however, to the extent that a state's process of drawing electoral districts, especially large, congressional ones, probably takes account of racial data throughout the process. Perhaps the Court's nonapplication of strict scrutiny to race-as-a- factor districting is anomalous, as some justices claim. (152) Or perhaps voting is a context in which the Court recognizes, as some scholars have, the legitimacy, necessity and inevitability of race-conscious redistricting in order to protect the political interests of minority groups. (153) Put another way, electoral districting may well use race in a discriminatory, not-just-investigative, manner, but the Court permits it without strict scrutiny unless the state uses race to an excessive degree.

3. Suspect Descriptions

No court has held that law enforcement's use of an eyewitness's description of a suspect that includes race is discriminatory and subject to strict scrutiny, (154) although some litigation and scholarship have made such claims. (155) One might justify judicial tolerance of the practice on the ground that the police are relying on race only to investigate the crime, not to determine guilt or innocence, or to impose punishment. It is analogous to the use of eyewitness testimony in court--a use of race for identification and fact-finding, not as a discriminatory basis for rendering a verdict or judgment. On the other hand, subjecting a person to arrest is more burdensome to the individual than mere information gathering. This is a context in which the state action can fairly be characterized as both investigative and discriminatory. To the extent an arrest begins a process of bringing a suspect in for questioning, booking, or charging, it is arguably investigative in nature, and part of a process that will ultimately turn on nondiscriminatory determinations. At the same time, to the extent a police arrest is itself a coercive state action, arresting a suspect based on a race- based description is arguably a discriminatory action--at the point of arrest--despite the further investigation that will take place after the arrest. Accordingly, despite the widespread acceptance of police use of race-based suspect descriptions, it is difficult to characterize such a practice as wholly investigative.

4. Racial Profiling

If police use of race-based suspect descriptions is discriminatory, then arrests based on racial profiles of likely criminals would seem to be at least as discriminatory. But if the former is not discriminatory, is the latter still discriminatory? Despite their acceptance of race-based suspect descriptions, courts and commentators (156) typically view police use of race-based profiles as discriminatory. The Court in Whren v. United States (157) seemed to agree, stating in dicta that selective enforcement of the law based on race would violate the Equal Protection Clause.

But why is the use of race in a profile more discriminatory than its use in a suspect description? Both practices involve the police relying in part on the race of an observed individual in deciding whether to investigate him and possibly make an arrest. Sheri Lynn Johnson (158) and other commentators (159) have defended the distinction between race-based suspect descriptions and race-based profiles, while Richard Banks has argued that the distinction is tenuous. (160) A potential distinction between the two uses of race is that race in a profile reflects an impermissible assumption about the criminal propensity of different racial groups whereas race in a suspect description just assumes that an individual whose race matches that of an observed suspect is more likely to be that suspect. Profiles thus rely on illegitimate assumptions or stereotypes that suspect descriptions do not. That does seem to be the point of scholars who defend the distinction. Richard Banks argues, by contrast, that profiles need not involve assumptions or stereotypes about the criminal propensity of different racial groups. Rather, they can involve empirically reliable generalizations about the correlation, not causation, between race and certain crimes that police use to make statistically valid predictions in the process of their investigations, predictions no more inaccurate or essentializing than predictions made using race-based suspect descriptions. Whether or not the distinction between suspect descriptions and profiles is sound, both practices, although investigatory in law enforcement terms, involve the police selecting suspects to subject to coercive action based in part on a suspect's race. As such, they would seem to be discriminatory and subject to strict scrutiny.

5. Remedying Societal Discrimination

If a state seeks to remedy societal racial discrimination by intentionally benefiting racial minorities through race-neutral means, such action would arguably be subject to strict scrutiny. (161) If, however, a state remedies societal racial discrimination through race-neutral means, without intending to benefit one racial group over another, then strict scrutiny should not apply. (162) The reason is that remedying societal racial discrimination is a legitimate, nondiscriminatory interest. If that interest is pursued through race-neutral means without a racially discriminatory purpose, strict scrutiny should not be triggered.

The question arises, however, whether race may play any role in remedying societal racial discrimination without triggering strict scrutiny. May a state investigate the causes of racial disparities as part of a process of identifying the socioeconomic conditions that likely result from societal discrimination? May the state then address those conditions on a race-neutral basis without being subject to strict scrutiny?

The investigative use of race supports such an approach. Provided the state's purpose in investigating racial disparities is to determine whether societal discrimination is a plausible cause, and not a precommitment to assuming so, then such an investigation should be legitimate and informational only. If a state legitimately concludes, in combination with other indicia of societal discrimination, that some condition causing a racial disparity likely results from societal discrimination, then the state should be able to respond to that condition without its actions being subject to strict scrutiny.

The investigative use of race involves two stages: investigation and action. Remedying societal discrimination should be able to satisfy the second stage. If a state uses race-neutral means to address conditions that the state legitimately believes have resulted from societal racial discrimination, without selecting the race-neutral means in order to favor racial minorities, then such action should not be considered discriminatory.

Two objections may nonetheless be raised to the first stage--investigating racial disparities in search of the effects of societal discrimination. The question is whether that stage involves an illegitimate assumption or belief about race. The first objection made by the Court is that the state's interest in remedying societal discrimination is too amorphous. Second, the Court has criticized the assumption that racial disparities necessarily reflect discrimination.

Regarding the first objection, that remedying societal discrimination is too amorphous, the Court's concern is premised on relying on societal discrimination as a justification for racial classifications. (163) Under strict scrutiny, the Court requires that discrimination be identified with particularity in order to define the scope of a racial classification or an otherwise racially discriminatory policy. (164) Here, the state would not be investigating racial disparities in order to use racial classifications or in order to use race- neutral means to benefit racial minorities. Rather, the state would be investigating racial disparities for a completely nondiscriminatory purpose, namely, identifying effects of societal racial discrimination regardless of which racial groups experience those effects. The concerns that have led the Court to require a more particularized finding of discrimination would not be implicated.

The second objection, that it is illegitimate to assume that racial disparities are caused by societal discrimination, has force. The Court has clearly indicated that a state cannot properly assume that racial groups would be represented in proportion to their population in all facets of life absent discrimination. (165) To allow such an assumption to motivate state decisionmaking would come close to endorsing racial balancing (166)--the assumption that racial disparities are objectionable per se--only, the assumption here would be that all racial disparities reflect societal discrimination.

A permissible approach lies between two extremes. On the one hand, because the process of identifying the effects of societal discrimination does not by itself involve a racially discriminatory purpose, a state should not have to identify the discrimination with the same particularity required to justify racially discriminatory state action. On the other hand, a state cannot, under current doctrine, assume, without more, that all racial disparities reflect societal discrimination. For the inference of societal discrimination to be legitimate, it should be based on more than racial disparities alone. A middle ground would allow a state to infer societal discrimination if evidence in addition to racial disparities reasonably supports a conclusion that such discrimination is a likely cause of such disparities.

Thus, for example, assume that a state's investigation into racial disparities in educational outcomes reveals that especially low-performing school districts are predominantly populated by black schoolchildren. Assume further that other empirical evidence points to societal discrimination as a cause of the school district's condition. For example, social science research from such fields as history, economics, political science, psychology, and sociology may indicate that the school district's condition is likely a product of generations of state- and private-sponsored discriminatory policies, including residential and school segregation, low school funding, underinvestment in economic infrastructure and crime control, and discriminatory mortgage practices that both redlined the district and precluded its residents from accessing federal loans to purchase homes in the suburbs, among other practices. Based on such evidence of the effects of societal discrimination, the state could invest in the economic and educational improvement of the district on a race-neutral basis. The goal of such investment, to avoid strict scrutiny, should not be to benefit the district's black residents. Rather, the purpose should be to address the socioeconomic and crimogenic conditions of the district that likely result from societal racial discrimination regardless of what race might be benefited by the state's investment. Even if the state knew that a disproportionate number of beneficiaries of its investment policy would be black, awareness of racial consequences does not trigger strict scrutiny. Under Washington v. Davis (167) and Personnel Administrator v. Feeney, (168) a state's adoption of a policy despite, rather than because of, its impact on a racial group is not subject to strict scrutiny.

IV

CONCLUSION

This article began with the question whether government may investigate and address the causes of racial disparities. The example we focused on is racial disparities in infant mortality. The question applies, moreover, to health disparities more generally and to other aspects of American society in which racial disparities persist. The article does not criticize current equal protection doctrine. Instead, it takes that doctrine seriously and concludes that, notwithstanding its colorblind thrust, it can accommodate some attention to race for the purpose of addressing the underlying causes of racial inequalities.

The article concludes with the question of whether government should investigate and address the causes of racial inequality. Some may say no, and they need not be bigots. Some people may genuinely believe in good faith that racial inequality is simply an inevitable feature of American society--that racial inequality is normal, even natural. Justice Thomas may well believe this when he states that "the absence of racial disparities in multi-ethnic societies has been the exception, not the rule," (169) and that "[w]hen it comes to 'proportiona[l] representation]' of ethnic groups, 'few, if any, societies have ever approximated this description.'" (170) He may be right that some "racial imbalance can also result from any number of innocent private decisions." (171) But the racial imbalance across American society is not plausibly disconnected from America's long history of racial injustice. Racial disparities in America are not occasional, minor, or inconsistent. Rather, stark racial disparities in which blacks are consistently and substantially worse off than whites exist along virtually every indicator of social and economic well-being. Such disparities are at least a matter of sufficient concern to justify investigating and, if warranted, addressing their root causes.

Racial inequality in America is not normal, natural, or innocent. The basis of this claim is the same proposition that underlies the Supreme Court's suspicion of race-based policies, namely, that race is irrelevant. If race is irrelevant, then black infants should not be significantly more prone to die than white infants. Black children should not be more prone to fail out of school. Black adults should not be more prone to be poor, jobless, homeless, inmates, or victims of crime. If race is irrelevant, then racial disparities signal, like the canary in the mine, that something is amiss. Racial disparities in America signal that centuries of discrimination have entrenched inequalities that perpetuate themselves from one generation to the next. These inequalities, moreover, result in unequal access to the types of resources and opportunities that are necessary for people to have fulfilling lives.

If the Constitution protects equality, then it cannot logically mandate ignoring inequality. If anything, it should mandate addressing inequality, but if not, at a minimum it should permit the government to address inequality. The Constitution, properly interpreted, does not require us to ignore the canary in the mine. The Constitution is not canary blind.

Copyright[c] 2016 by Kim Forde-Mazrui.

This article is also available at http://lcp.law.duke.edu/.

Kim Forde-Mazrui, Professor of Law and Mortimer M. Caplin Professor of Law, University of Virginia; Director, University of Virginia Center for the Study of Race and Law. A.B. 1990, J.D. 1993, University of Michigan. I am grateful for comments I received on a draft from Regina Austin and Deborah Heilman. I also received helpful feedback from attendees at presentations I gave at Duke University School of Law, the University of Michigan Law School, the University of Virginia School of Law, and Wake Forest University School of Law, as well as from participants in the Mid- Atlantic People of Color Legal Scholarship Conferences at the University of Pennsylvania Law School in 2013 and at the University of Baltimore School of Law in 2014. I thank the University of Virginia Working Group on Racial Inequality for our constructive discussion and UVA Law librarian Ben Doherty for outstanding reference support. A special thanks to Jeremy Bennie, Evan Didier, Nina Goepfert and Peter Park for their first-rate research assistance. I welcome comments at kimfm@virginia.edu.

(1.) Lani Guinier & Gerald Torres, The Miner's Canary 1 (2003).

(2.) Id.

(3.) Id.

(4.) See, e.g., CENTRAL INTELLIGENCE AGENCY, THE WORLD FACTBOOK, https://www.cia.gov/ library/publications/the-world-factbook/rankorder/2091rank.html (last visited Nov. 12, 2015).

(5.) See Donald G. McNeil Jr., Broad Racial Disparities Seen in Americans' Ills, N.Y. TIMES (Jan. 13, 2011), http://www.nytimes.com/2011/01/14/health/14cdc.html?_r=0.

(6.) Marian F. MacDorman & T.J. Mathews, Nat'l Ctr. for Health Statistics, UNDERSTANDING RACIAL AND ETHNIC DISPARITIES IN U.S. INFANT MORTALITY RATES 1 (Sept. 2011), http://www.cdc.gov/nchs/data/databriefs/db74.pdf.

(7.) Paula Braveman, Racial Disparities at Birth: The Puzzle Persists, 24 ISSUES SCI. & TECH., no. 2, Winter 2008, http://issues.org/24-2/p_braveman/.

(8.) Ziba Kashef, Why African American babies have the highest infant mortality rate in the developed world, THE BLACK COMMENTATOR, no. 32, Mar. 6, 2003, http://www.blackcommentator .com/32/32_reprint.html ("College- and graduate-school educated black mothers have a higher infant mortality rate than white moms who didn't finish high school.").

(9.) Richard David & James Collins, Jr., Disparities in Infant Mortality: What's Genetics Got to Do With It?, 97 AM. J. Pub. Health 1191,1193 (2007); Braveman, supra note 7 (observing that black and Hispanic immigrants have better birth outcomes than their U.S.-born daughters despite the latter generally having higher socioeconomic status).

(10.) See Debbie Barrington, The Metaphor of the Miner's Canary and Black-White Disparities in Health: A Review of Intergenerational Socioeconomic Factors and Perinatal Outcomes, in SOCIAL Inequality and Public Health 83,83 (Salvatore J. Babobes ed., 2009).

(11.) Ctrs. for Disease Control & Prevention, Infant Mortality, http://www.cdc.gov/ reproductivehealth/maternalinfanthealth/infantmortality.htm.

(12.) See Barrington, supra note 10, at 84.

(13.) See id. at 84-85.

(14.) See Nat'l Ctr. for Health Statistics, Health, United States, 2011: With Special Feature on Socioeconomic Status and Health 46 (2012), http://www.cdc.gov/nchs/data/hus/ husll.pdf.

(15.) Mary Carmichael, Why Racial Disparities in Health Care Persist, NEWSWEEK, Feb. 14, 2010, http://www.newsweek.com/why-racial-disparities-health-care-persist-75409.

(16.) See Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Brian D. Smedley et al. eds., 2003).

(17.) Katrina Armstrong et al., Racial/Ethnic Differences in Physician Distrust in the United States, 97 Am. J. Pub. Health 1283 (2007); see also Barbara A. Noah, A Prescription for Racial Equality in Medicine, 40 CONN. L. REV. 675, 677-78 (2008) (stating that cultural barriers between doctors and minority patients contributes to the disparity in the quality of care).

(18.) Tyan Parker Dominguez et al., Racial Differences in Birth Outcomes: The Role of General, Pregnancy, and Racism Stress, 27 HEALTH PSYCHOL. 194 (2008).

(19.) Barrington, supra note 10, at 86.

(20.) See id. at 88 (citing studies that found a grandmother's educational attainment was linked to preterm births for the next two generation of women).

(21.) Id. at 86.

(22.) See Braveman, supra note 7; Kashef, supra note 8 (describing infant mortality rate disparities as "persistent peril" and "medical mystery"); Timothy Williams, Tackling Infant Mortality Rates Among Blacks, N.Y. TIMES, Oct. 14, 2011 (describing infant mortality disparities as a "mystery that has eluded researchers").

(23.) See Ctrs. for Disease Control & Prevention, CDC Grand Rounds: Public Health Approaches to Reducing U.S. Infant Mortality, Weekly, (Aug. 9, 2013), http://www.cdc. gov/mmwr/preview/mmwrhtml/mm6231a3.htm (describing state, local and national efforts to reduce infant mortality and racial disparities in infant mortality, including programs, funded by Congress through the CDC of forty-three states and New York City); Fern R. Huack, Kawai O. Tanabe & Rachel Y. Moon, Racial and Ethnic Disparities in Infant Mortality, 35 SEMINARS IN PERINATOLOGY 209,209 (2013).

(24.) This article's discussion over whether certain state actions are constitutional or not is limited to the application of equal protection doctrine as developed by the Supreme Court. The article does not intend to endorse or accept Supreme Court doctrine as a proper interpretation of the Constitution.

(25.) See Kim Forde-Mazrui, The Constitutional Implications of Race-Neutral Affirmative Action, 88 GEO. L.J. 31, 37-51 (2000); Helen Norton, The Supreme Court's Post-Racial Turn Towards a Zero-Sum Understanding of Equality, 52 WM. & Mary L. Rev. 197 (2010).

(26.) See id. at 95.

(27.) See, e.g., Brian T. Fitzpatrick, Strict Scrutiny of Facially Race-Neutral State Action and the Texas Ten Percent Plan, 53 BAYLOR L. Rev. 289 (2001); Alan Wendler Hersh, Keep It Quiet: How Facially Neutral Affirmative Action Passes Constitutional Scrutiny, 2011 U. ILL. L. REV. 1885 (2011); George La Noue & Kenneth L. Marcus, "Serious Consideration" of Race-Neutral Alternatives in Higher Education, 57 CATH. U.L. Rev. 991 (2008); Norton, supra note 25; Kimberly Jenkins Robinson, The Constitutional Future of Race-Neutral Efforts to Achieve Diversity and Avoid Racial Isolation in Elementary and Secondary Schools, 50 B.C. L. Rev. 277 (2009).

(28.) See Forde-Mazrui, supra note 25, at 37-51; Michelle Adams, Is Integration a Discriminatory Purpose?, 96 IOWA L. REV. 837 (2011); Richard A. Primus, Equal Protection and Disparate Impact: Round Three, 117 Harv. L. Rev. 494, 519 n.111 (2003) (citing Guinier & Torres, supra note 1, at 11-12).

(29.) Forde-Mazrui, supra note 25.

(30.) Id. at 34-35.

(31.) Id.

(32.) See Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 279 (1979) (explaining that a discriminatory purpose is one by which the legislature adopts a policy because of, not merely in spite of, its adverse impact on a racial group).

(33.) See infra text accompanying notes 37-48.

(34.) See, e.g., Healthy People 2020, HEALTHYPEOPLE.GOV, www.healthypeople.gov/2020/About-Healthy-People.

(35.) Ctrs. for Disease Control & Prevention, CDC Health Disparities & Inequalities Report 1 (Nov. 22, 2013), http://www.cdc.gov/mmwr/pdf/other/su6203.pdf.

(36.) Cf. Healthy People 2020, supra note 34 (suggesting a broad group of individuals and organizations support eliminating disparities in health outcomes); 2014 Health Disparities Legislation Nat'l Conf. OF St. Legislatures (July 11, 2014), http://www.ncsl.org/research/health/2014-health-disparities- legislation.aspx (listing proposed legislation from more than thirty states aimed at solving the health-disparities problem).

(37.) Ethnic and Racial Disparities in Education: Psychology's Contributions to Understanding and Reducing Disparities, Am. PSYCHOL. ASS'N, http://www.apa.org/ed/resources/racial-disparities.aspx (last visited Nov. 12, 2015).

(38.) Drew Desilver, Black Unemployment Rate is Consistently Twice that of Whites, PEW RES. Ctr. (Aug. 21, 2013), http://www.pewresearch.org/fact-tank/2013/08/21/through-good- times-and-bad-black-unemployment-is- consistently-double-that-of-whites/.

(39.) Rakesh Kochhar & Richard Fry, Wealthy Inequality has Widened Along Racial, Ethnic Lines Since End of Great Recession, Pew Res. Ctr. (Dec. 12, 2014), http://www.pewresearch.org/fact-tank/2014/12/12/racial- wealth-gaps-great-recession/.

(40.) R. Kelly Raley, Megan M. Sweeney & Daneille Wondra, The Growing Racial and Ethnic Divide in U.S. Marriage Patterns, FUTURE CHILD. 89, 89 (Fall 2015).

(41.) Carmen Solomon-Fears, Cong. Res. Serv., nonmarital Births: An Overview (July 30, 2014), http://fas.org/sgp/crs/misc/R43667.pdf.

(42.) Nat'l Conf. of State Legislatures, Disproportionate Minority Contact, http://www.ncsl.org/documents/cj/jjguidebook-dmc.pdf.

(43.) AM. Civ. Liberties Union, Racial Disparities in Sentencing (Oct. 27, 2014), https://www.aclu. org/sites/default/files/assets/141027_iachr_racial_disparities_aclu_submission_0 .pdf.

(44.) The Sentencing Project, Report of The Sentencing Project to the United Nations Human Rights Committee: Regarding Racial Disparities in the United States Criminal Justice System (Aug. 2013), http://sentencingproject.org/doc/publications/rd_ICCPR%20 Race%20and%20Justice%20Shadow%20Report.pdf.

(45.) Mark T. Berg, Accounting for Racial Disparities in the Nature of Violent Victimization, 30 J. Quantitative Criminology 629 (2014).

(46.) Nat'l Ass'n for the Advancement of Colored People, Criminal Justice Fact Sheet, http://www.naacp.org/pages/criminal-justice-fact-sheet (last accessed Nov. 12, 2015).

(47.) The Sentencing Project, Reducing Racial Disparity in the Criminal Justice SYSTEM, http://www.sentencingproject.org/doc/publications/rd_reducingracialdisparity.pdf .

(48.) Id. at 6.

(49.) See Morales v. Daley, 116 F. Supp. 2d 801, 812-16 (S.D. Tex. 2000) (entertaining though rejecting lawsuit claiming the census violates equal protection by inquiring into racial classification).

(50.) See Michael J. Klarman, From Jim Crow to Civil Rights: The Supreme Court and the Struggle for Racial Equality (2004) (arguing that the Supreme Court rarely deviates from the weight of societal opinion).

(51.) 347 U.S. 483 (1954).

(52.) See, e.g., Turner v. City of Memphis, 369 U.S. 350, 354 (1962) (invalidating segregation in airport restaurants and restrooms); New Orleans City Park Improvement Ass'n v. Detiege, 358 U.S. 54, 54 (1958) (invalidating segregation in city parks); Florida ex rel. Hawkins v. Board of Control, 350 U.S. 413, 413 (1956) (invalidating segregation in public law school); Gayle v. Browder, 352 U.S. 903, 903 (1956) (invalidating segregation in city buses); Holmes v. Atlanta, 350 U.S. 879, 879 (1955) (invalidating segregation in city golf course); Mayor & City Council of Balt. City v. Dawson, 350 U.S. 877, 877 (1955) (invalidating segregation in public beaches and bath-houses and swimming pools); Muir v. Louisville Park Theatrical Ass'n, 347 U.S. 971,971 (1954) (invalidating segregation in fishing lakes).

(53.) See e.g., Croson, 488 U.S. at 493 (justifying strict scrutiny as a test that "ensures that the means chosen 'fit' this compelling goal so closely that there is little or no possibility that the motive for the classification was illegitimate racial prejudice or stereotype." For a discussion on the illegitimate use of racial stereotype, see Forde-Mazrui, supra note 25, at 55.

(54.) Swain v. Alabama, 380 U.S. 202, 221-24 (1965).

(55.) 476 U.S. 79 (1986).

(56.) See, e.g., Miller-El v. Dretke, 545 U.S. 231, 272 (2005) (holding that a prosecutor's preemptory challenges based on racial stereotypes could be appealed, even though he provided race-neutral explanations, because the weight of other evidence could suggest the prosecutor had a racial bias); Miller-El v. Cockrell, 537 U.S. 322 (2003) (holding that a defendant was entitled to a certificate of appeal because the evidence could support a claim that the prosecution's exclusion on venire were racially motivated, which would violate the Equal Protection Clause); Georgia v. McCollum, 505 U.S. 42, 59 (1992) (holding that defendants cannot exclude jurors on the basis of race or racial stereotypes); Edmonson v. Leesville Concrete Co., 500 U.S. 614, 630-31 (1991) (holding that excluding jurors on the basis of racial stereotypes in civil cases violated the Equal Protection Clause); Powers v. Ohio, 499 U.S. 400, 415 (1991) (holding that defendant has the right to bring a third-party equal protection claim on the behalf of jurors excluded on the basis of their race, even if the defendant and the juror are of a different race).

(57.) Powers v. Ohio, 499 U.S. 400, 410 (1991).

(58.) Shaw v. Reno, 509 U.S. 630, 647 (1993).

(59.) City of Richmond v. J.A. Croson Co., 488 U.S. 469, 493 (1989).

(60.) Id. at 495-96.

(61.) Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 729-30 (2007) (citing Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 307 (1978)); see also Grutter v. Bollinger, 539 U.S. 306, 330 (2003).

(62.) See Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 269-79 (1978) (classifying an admissions program that took into account an applicant's race as having a racial classification).

(63.) 488 U.S. 469 (1989).

(64.) Id. at 494; see also Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227 (1995) (applying same rule to federal government).

(65.) See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266 (1977) (stating that judicial deference is not justified for race-neutral laws that exhibit a clear pattern of discrimination as applied).

(66.) Washington v. Davis, 426 U.S. 229,246 (1976) (assuming that, if police department adopted test in order to discriminate against black applicants, it would be subject to strict scrutiny, but concluding that such a purpose had not been proved).

(67.) United States v. Fordice, 505 U.S. 717, 739-40 (1992).

(68.) Gomillion v. Lightfoot, 364 U.S. 339, 348-49 (1960).

(69.) 509 U.S. 630, 653 (1993).

(70.) 515 U.S. 900 (1995).

(71.) Id. at 916.

(72.) Id.

(73.) See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266 (1977) (explaining in context of allegation of racial discrimination against blacks that a finding of an invidious discriminatory purpose as one "motivating factor" is sufficient to as trigger strict scrutiny); Gomillion, 364 U.S. at 347 (invalidating racial gerrymandering designed to disenfranchise blacks).

(74.) 551 U.S. 701 (2007).

(75.) See id. at 789 (Kennedy, J., concurring) (suggesting that school districts could engage in efforts to create racially integrated schools, such as site location and district boundaries that did not involve individually classifying schoolchildren by race without being subject to strict scrutiny).

(76.) City of Richmond v. J.A. Croson Co., 488 U.S. 469, 509-10 (1989); id. at 526, 528 (Scalia, J., concurring in the judgment); see also Forde-Mazrui, supra note 25, at 49-50 (quoting both the plurality opinion and Justice Scalia's opinion).

(77.) 557 U.S. 557, 563.

(78.) See id. at 576-77 (explaining that Court did not need to address potential equal protection claim because city's actions violated statute).

(79.) Id. at 594 (Scalia, J., concurring) (arguing that disparate-impact liability encouraged racially discriminatory employment policies in tension with equal protection doctrine).

(80.) Forde-Mazrui, supra note 25, at 46-49 (arguing that logic of equal protection doctrine supports subjecting race-neutral laws intended to benefit racial minorities to strict scrutiny).

(81.) Croson, 488 U.S. at 493 (majority opinion) (explaining that all racial classifications must be subject to strict scrutiny to "smoke out" illegitimate purposes).

(82.) See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S 252, 265-66 (1977) (explaining in the context of allegations of discrimination against blacks that race as one motivating factor triggers strict scrutiny); Washington v. Davis, 426 U.S. 229, 242 (1976) (assuming that proof of racial motivation behind race-neutral test with a discriminatory impact would trigger strict scrutiny, but holding that strict scrutiny was not triggered because the test was not motivated by a discriminatory purpose).

(83.) Croson, 488 U.S. at 492 (explaining that, to remedy past discrimination, the discrimination must be identified with particularity).

(84.) Id. at 501 (holding that remedying "societal discrimination" evidenced by generalized statistical disparities is not sufficiently compelling to justify racial quota but rather discrimination must be identified with adequate particularity).

(85.) Office of Minority Health & Health Disparities, Guiding Principle, http://www.cdc .gov/omhd/about/disparities.htm.

(86.) 471 U.S. 222, 233 (1985).

(87.) Id. (emphasis added).

(88.) See Staub v. Proctor Hosp., 562 U.S. 411, 416-22 (2011) (explaining that termination is discriminatory if earlier evaluation that was discriminatory had a reasonably foreseeable effect on the ultimate termination decision).

(89.) Id.

(90.) Id. at 416-22 (explaining that termination is discriminatory if earlier evaluation that was discriminatory had a reasonably foreseeable effect on the ultimate termination decision).

(91.) See Hunter, 471 U.S. at 231-33 (holding unconstitutional disenfranchising provision of state constitution that was motivated by a discriminatory purpose when adopted despite claim by state that it has since retained the provision for nondiscriminatory reasons).

(92.) City of Richmond v. J.A. Croson Co., 488 U.S. 469, 492 (1989) (explaining that, to remedy past discrimination, the discrimination must be identified with particularity).

(93.) See id. at 505-06 (rejecting the remedying of societal discrimination as compelling).

(94.) See Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 730 (2007) (explaining that using race to achieve proportionate representation of schoolchildren of different races constitutes "racial balancing," a per se unconstitutional purpose).

(95.) See Forde-Mazrui, supra note 25, at 51-81 (arguing that strict scrutiny should permit race-neutral policies designed to remedy societal racial discrimination).

(96.) See, e.g., Ricci v. DeStefano, 557 U.S. 557 (2009); St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502 (1993); McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973).

(97.) See, e.g., McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 (1973) (establishing elements of plaintiff's prima facie case, which includes that defendant "belongs to a racial minority"); St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 506 (1993) (noting District Court's finding that plaintiff had satisfied minimal requirements of prima facie case by proving, among other elements, that he is black).

(98.) Int'l Bhd. of Teamsters v. U.S., 431 U.S. 324, 339 (1977).

(99.) Id. at 340; see also Hazelwood Sch. Dist. v. U.S., 433 U.S. 299, 309-13 (1977).

(100.) Second Amended Complaint for Declaratory, Injunctive, and Other Relief at 3, Fisher v. Univ. of Tex., 645 F. Supp. 2d 587 (W.D. Tex. 2008) (No. 1:08-cv-00263-SS).

(101.) For the Court's ultimate decision in this litigation, see generally Fisher v. Univ. of Tex. at Austin, No. 14-981 (June 23, 2016) (affirming on equal protection grounds the Fifth Circuit's finding that the University of Texas at Austin's undergraduate admissions program is constitutional).

(102.) See DeStefano, 557 U.S. at 563, 584.

(103.) Id. at 594 (Scalia, J., concurring) ("Title VII's disparate-impact provisions place a racial thumb on the scales, often requiring employers to evaluate the racial outcomes of their policies, and to make decisions based on (because of) those racial outcomes.").

(104.) See R. Richard Banks, Race-Based Suspect Selection and Colorblind Equal Protection Doctrine and Discourse, 48 UCLA L. REV. 1075,1077 n.3 (2001).

(105.) See, e.g., Nolting v. Holt, 215 P. 281 (Kan. 1923); Damien v. J.G., 957 N.Y.S.2d 819 (N.Y. Fam. Ct. 2012). Such testimony is likely unnecessary today in light of paternity testing but, again, there was never any suggestion when it has been used that it represented racially discriminatory court proceedings.

(106.) City of Richmond v. J.A. Croson Co., 488 U.S. 469, 492 (1989) ("Richmond ... can use its spending powers to remedy private discrimination, if it identifies that discrimination with the particularity required by the Fourteenth Amendment.").

(107.) 448 U.S. 448 (1980).

(108.) Id. at 459.

(109.) See id.

(110.) Id. at 492, 509.

(111.) See Grutter v. Bollinger, 539 U.S. 306 (2003).

(112.) See Ricci v. DeStefano, 557 U.S. 557, 579 (2009) (stating that discarding test results because too many white candidates, and not enough black candidates, would be promoted, without more violated Title VII's prohibition against making employment decisions on the basis of a person's race).

(113.) Id. at 585.

(114.) Andrew M. Carlon, Racial Adjudication, 2007 BYU L. REV. 1151, 1158 (2007) (arguing that census race classification by itself and with any effect is not problematic as evidenced by jurisprudence and legislative initiatives).

(115.) Mary Bernadette Ott & Gary Yingling, Guide to Good Clinical Practice [paragraph] 350 (2015); Manav Bhatnagar, Identifying the Identified: The Census, Race, and the Myth of Self-Classification, 13 Tex. J. Civ. Liberties & Civ. RTS. 85, 86 (2007); David A. Harris, The Reality of Racial Disparity in Criminal Justice: The Significance of Data Collection, 66 L. & CONTEMP. PROBS., no. 3, 2003, at 71; Christopher Ogolla, Will the Use of Racial Statistics in Public Health Surveillance Survive Equal Protection Challenges? A Prolegomenon for the Future, 31 N.C. CENT. L. Rev. 1, 2 (2008).

(116.) See, e.g., 29 C.F.R. [section] 1602.7 and Equal Employment Opportunity Commission, 70 Fed. Reg. 71, 294, at 71, 295 (Nov. 28, 2005) (requiring all employers subject to Title VII of the Civil Rights Act of 1964 to submit race data on employees via the Employment Information Report (EEO-1); 34 C.F.R. [section] 303.721 (a) (2001) (requiring state agencies to report statistics on the number of children and toddlers receiving early interventions services in the state by race) and Final Guidance on Maintaining, Collecting, and Reporting Racial and Ethnic Data to the U.S. Department of Education 72 Fed. Reg. 59.266, 59,266 (Oct. 19, 2007).

(117.) See Morales v. Daley, 116 F. Supp. 2d 801 (S.D. Tex. 2000); see also James M. Balkin & Reva B. Siegel, Principles, Practices, and Social Movements, 154 U. Pa. L. Rev. 927, 937-43 (2006).

(118.) Morales, 116 F. Supp. 2d at 816.

(119.) Tancil v. Woolls, 379 U.S. 19 (1964) (per curiam) (affirming sub nom Hamm v. Va. State Bd. of Elections, 230 F. Supp. 156 (E.D. Va. 1964)).

(120.) Hamm, 230 F. Supp. at 157.

(121.) Id. at 158 n.5.

(122.) 517 U.S. 456 (1996).

(123.) Id. at 459.

(124.) Id. at 461.

(125.) Armstrong, 517 U.S. at 470-71.

(126.) Id. at 469.

(127.) See, e.g., Sheet Metal Workers v. EEOC, 478 U.S. 421, 479-81 (1986); Palmore v. Sidoti, 466 U.S. 429, 432 (1984).

(128.) See Kim Forde-Mazrui, Tradition as Justification: The Case of Opposite- Sex Marriage, 78 U. Chi. L. Rev. 281, 302 (2011) (explaining that government has no authority to pursue illegitimate interests or purposes).

(129.) Plessy v. Ferguson, 163 U.S. 537, 559 (1896) (Harlan, J., dissenting).

(130.) See, e.g., Batson v. Kentucky, 476 U.S. 79, 86 (1986) (stating that jurors cannot be excluded on the assumption that their race makes them less qualified).

(131.) City of Richmond v. J.A. Croson Co., 488 U.S. 469, 492 (1989) (explaining that legislation that was passed to remedy racial discrimination could survive strict scrutiny if the discrimination could be identified "with particularity").

(132.) See Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Brian D. Smedley et al. eds., 2003).

(133.) Katrina Armstrong et at, RaciaUEthnic Differences in Physician Distrust in the United States, 97 AM. J. PUB. Health 1283 (2007); see also Barbara A. Noah, A Prescription for Racial Equality in Medicine, 40 CONN. L. Rev. 675, 677-78 (2008) (stating that cultural barriers between doctors and minority patients contributes to the disparity in the quality of care).

(134.) See Barrington, supra note 10, at 84.

(135.) See Racial and Ethnic Disparities, SUICIDE PREVENTION RESOURCE CTR. (last visited May 13, 2016); Kelly Burns, Suicide Rate for Minorities Much Lower, Census Data Indicate, NEWS REPORTING & THE Internet (last visited May 13, 2016), http://students.com.miami.edu/netreporting/7page _id=1285.

(136.) See Burns, supra note 135 (suggesting that blacks may have greater access to community support than whites, which may contribute to lower suicide rates for blacks); Robert Joseph Taylor et al., Racial and Ethnic Differences in Extended Family, Friendship, Fictive Kin and Congregational Informal Support Networks, 62 Fam. REL. 609, 609 (2013) (reviewing research suggesting that black Americans tend to have greater extended family, friendship, fictive kin and religious support networks than whites, which serves as a protective factor against suicide).

(137.) See Taylor et al., supra note 136, at 609.

(138.) See Grutter v. Bollinger, 539 U.S. 306, 316 (2003) (revealing underrepresentation of black and Latino students at the University of Michigan Law School).

(139.) See Susan P. Dalessandro, Lisa C. Anthony & Lynda M. Reese, LSAT Performance With Regional, Gender, and Racial/Ethnic Breakdowns: 2005-2006 Through 2011-2012 Testing Years 2 (2012) (reporting that, between 2005 and 2012, "[ajverage LSAT scores were highest for Caucasian and Asian/Pacific Islander test takers. African American test takers and Puerto Rican test takers had the lowest mean LSAT scores").

(140.) Cf. Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 729-30 (2007) (citing Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 307) (suggesting that racial balancing was a per se unconstitutional goal).

(141.) Civil Rights Act of 1964, 42 U.S.C. [section] 2000e-2(k) (2015).

(142.) Ricci v. DeStefano, 557 U.S. 557, 577 (2009).

(143.) Justice Scalia appeared to believe so. See id. at 594-95 (Scalia, J., concurring); see also Primus, supra note 28, at 537-39 (explaining that disparate-impact doctrine is arguably subject to strict scrutiny); Deborah M. Weiss, All Work Cultures Discriminate, 24 HASTINGS WOMEN'S L.J. 247, 298 (2013) (suggesting that broad reading of Ricci would classify all employer attempts to avoid racially disparate impact as discriminatory under Title VII. while narrower reading of Ricci would not go so far).

(144.) Richard Primus suggests that disparate-impact liability might be justifiable along the lines of the miner's canary metaphor, as a warning signal that "the workplace is organized in ways that perpetuate unjustified power structures more generally, not just on the basis of race." See Primus, supra note 28, at 519 n.111 (citing GUINIER & TORRES, supra note 1, at 11-12). Deborah Weiss similarly suggests that an employer could use disparate impact as a "diagnostic tool" that helps the employer discover underlying problems with an employment practice which the employer could then address without discriminating by race. See Weiss, supra note 143, at 298.

(145.) Id. at 579.

(146.) See Ricci v. DeStefano, 557 U.S. 557, 578-79 (2009) (describing the three steps of a disparateimpact case, including the third step, at which an employee can "still succeed by showing that the employer refuses to adopt an available alternative employment practice that has less disparate impact and serves the employer's legitimate needs"); Albemarle Paper Co. v. Moody, 422 U.S. 405,425 (1975) (citing McDonnell Douglas Corp. v. Green, 411 U.S. 792, 801, 804-05 (1973)) (explaining that employee's success at third step of disparate-impact case serves as "evidence that the employer was using its tests merely as a 'pretext' for discrimination").

(147.) See Vieth v. Jubelirer, 541 U.S. 267 (2003); Miller v. Johnson, 515 U.S. 900, 916-20 (1995) (applying strict scrutiny to an electoral redistricting plan where it was found that race played a predominant factor in the district's design).

(148.) Miller, 515 U.S. at 916.

(149.) See Kim Forde-Mazrui, Jural Districting: Selecting Impartial Juries Through Community Representation, 52 Vand. L. Rev. 353, 383-85 (1999) (discussing Court's acceptance of using race as one of many factors in electoral districting while presumptively rejecting using race as a predominant factor in electoral districting).

(150.) Shaw v. Reno, 509 U.S. 630, 647 (1993).

(151.) See, e.g., Gratz v. Bollinger, 539 U.S. 244, 270 (2003) (subjecting the University of Michigan's admissions policy of taking race, among other factors, into account to strict scrutiny).

(152.) See Bush v. Vera, 517 U.S. 952, 999-1003 (1996) (Thomas, J., joined by Scalia, J., concurring in the judgment) (criticizing plurality's predominant-factor test as necessary predicate for strict scrutiny instead of applying strict scrutiny whenever race plays any causal role in a state's creation of a majorityminority district); see also Brian T. Fitzpatrick, Strict Scrutiny of Facially Race- Neutral State Action and the Texas Ten Percent Plan, 53 Baylor L. Rev. 289, 312-13 (2001) (observing that Justice Thomas's concurring opinion in Bush v. Vera, which was joined by Scalia, criticized the plurality's predominantfactor test for requiring that race be more than a "but for" factor in a redistricting plan before the plan is subject to strict scrutiny); accord Ken Gormley, Racial Mind-Games and Reapportionment: When Can Race Be Considered (Legitimately) in Redistricting?, 4 U. Pa. J. CONST. L. 735, 774 (2002).

(153.) See Forde-Mazrui, supra note 149, at 385-86; Pamela S. Karlan & Daryl J. Levinson, Why Voting Is Different, 84 Cal. L. Rev. 1201,1208,1215 (1996).

(154.) See, e.g.. Brown v. City of Oneonta, 221 F.3d 329, 338-39 (2d Cir. 1999).

(155.) See, e.g., Banks, supra note 104 (arguing that the use of race by police in an eye-witness description of a suspect would seem to be a discriminatory use of race despite courts not holding such).

(156.) See Whren v. United States, 517 U.S. 806, 813 (1996) (stating that Constitution forbids law enforcement based on consideration of race); Sheri Lynn Johnson, Race and the Decision to Detain A Suspect, 93 Yale L.J. 214, 242-43 (1983) (police reliance on race in deciding whom to detain is subject to strict scrutiny, except for use of race to identify a particular suspect); R. Richard Banks, Race-Based Suspect Selection and Colorblind Equal Protection Doctrine and Discourse, 48 UCLA L. Rev. 1075, 1090 (2001) (noting widespread institution that racial profiles are legally objectionable but race-based suspect descriptions are not).

(157.) Whren, 517 U.S. at 813.

(158.) See Johnson, supra note 156.

(159.) See, e.g., David Cole, No Equal Justice: Race and Class in the American Criminal Justice System 50 (1999); Randall Kennedy, Suspect Policy, 221 New Republic 30,34 (1999).

(160.) See Banks, supra note 104.

(161.) See Forde-Mazrui, supra note 25, at 46-49 (arguing that logic of equal protection doctrine supports subjecting race-neutral laws intended to benefit racial minorities to strict scrutiny).

(162.) It has also been argued that, although a state's interest in remedying societal discrimination is not sufficiently compelling to justify the use of racial classifications, it is arguably compelling enough to satisfy strict scrutiny when race-neutral policies are used to benefit racial minorities. See Forde-Mazrui, supra note 25, at 51-81. As the investigative use of race proposed in this article would avoid strict scrutiny, the discussion in the text is limited to non-discriminatory remedial efforts.

(163.) City of Richmond v. J.A. Croson Co., 488 U.S. 469,496-97 (1989).

(164.) Id. at 492.

(165.) Id. at 507.

(166.) Id.

(167.) 426 U.S. 229,246 (1976).

(168.) 442 U.S. 256, 279 (1979).

(169.) Texas Dep't of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, Inc., 135 S. Ct. 2507 (2015) (Thomas, J. dissenting) (quoting Donald Horowitz, Ethnic Groups in Conflict 677 (1985)).

(170.) Id.

(171.) Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 750 (2007) (Thomas, J., concurring).
COPYRIGHT 2016 Duke University, School of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:The Present and Future of Civil Rights Movements: Race and Reform in 21st Century America
Author:Forde-Mazrui, Kim
Publication:Law and Contemporary Problems
Date:Jun 22, 2016
Words:18404
Previous Article:Black contemporary social movements, resource mobilization, and black musical activism.
Next Article:Debt, merit, and equity in higher education access.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters