Under what circumstances should we understand government’s racist or otherwise hateful speech to violate the Equal Protection Clause? For example, should we understand the Equal Protection Clause to bar a government’s decision to adopt and display the motto “White Supremacy Forever” on the state seal or a state license plate? What if a governor or a president were to issue a proclamation or a legislature were to pass a resolution declaring that members of the Latino, Arab, or gay, lesbian, bisexual, and transgender (GLBT) communities are not worthy of respect? Many would consider such governmental messages of inequality to be offensive to equal protection values. Yet the small number of decisions in this area do not confirm that intuition.
To be sure, whether the government’s speech actually expresses animus on the basis of class status—that is, whether it actually communicates hatred or hostility—is often deeply contested. State and local governments’ decisions to fly the Confederate flag offer perhaps the most prominent contemporary example. Such symbolic use of the flag is, and is intended to be, expressive of the governments’ views, although the exact nature of those views remains bitterly contested. But such contests over meaning have largely obscured the theoretical question I seek to examine here: whether concededly hateful or otherwise facially discriminatory government speech that is unaccompanied by the government’s exercise of its coercive power violates the Equal Protection Clause. Note that here, by “government speech,” I mean the speech of a government agency or body—or the speech of an individual empowered to speak for such a body—rather than the speech of an individual legislator or other official expressing only his or her own opinion.
This Article1 starts by examining the various harms that might be caused by government’s speech that communicates hatred, hostility, or animus on the basis of race, gender, national origin, sexual orientation, or other class status. After describing lower courts’ failure to address most of these harms when considering equal protection challenges to government expression, it then contrasts courts’ more expansive understanding of the constitutionally salient harms of government’s religious speech in the Establishment Clause context—which is the only area other than the Free Speech Clause in which the Supreme Court has yet wrestled with the constitutional implications of government speech. The Article closes by exploring how coercion and endorsement analyses, which are often applied to government’s religious speech in the Establishment Clause context, might guide our thinking about government speech that expresses hostility on the basis of race or other class status for equal protection purposes. It concludes that, despite their imperfections, both analyses more accurately recognize those situations in which government speech may inflict harms repugnant to equal protection values than does the status quo approach, which largely ignores or dismisses those harms.
The Potential Harms of Government Speech that Expresses Animus on the Basis of Class Status
Government speech that expresses animus on the basis of class status can inflict harms that can take behavioral, psychological, and expressive form.
Under some circumstances, government’s racist or similarly hateful speech may sufficiently command or otherwise coerce its targets’ behavior to constitute the sort of regulation or “hard law” that clearly falls within the ambit of the Equal Protection Clause. In Lombard v. Louisiana, for example, the Supreme Court held that city officials’ speech commanding continued segregation by private restaurants was sufficiently coercive to violate the Constitution.2
Under some circumstances, moreover, government speech may encourage or facilitate discriminatory actions by private parties. Along these lines—in another decision long predating the emergence of the “government speech” doctrine—the Supreme Court recognized that government’s racially identifying speech on a ballot offended equal protection values because of its potential for encouraging voters’ private discrimination. More specifically, in Anderson v. Martin, the Supreme Court struck down, on equal protection grounds, Louisiana’s law requiring that political candidates be identified by race on all ballots because the Court recognized the danger that the government’s speech would facilitate voters’ discriminatory decision making.3
Government speech that expresses hostility on the basis of protected class status may also inflict behavioral harm by deterring targets from engaging in certain behavior. Consider, for example, the possible effects of a state capitol’s or courthouse’s racist proclamation or display on class members’ willingness to enter that building to petition the legislature or file a complaint.
Of course, government’s racist or otherwise hateful speech may also inflict psychological harms: these include fear, stress, feelings of inferiority, and depression. But even if the target suffers no such injury, government’s racist or otherwise hateful speech can inflict expressive harm simply by communicating the government’s view of class members’ unequal worth. In other words, the government may inflict expressive harm simply by sending a message of inferiority based on class status, regardless of whether listeners suffer resulting emotional distress or otherwise internalize the government’s message.
Courts’ Contrasting Equal Protection and Establishment Clause Analyses of Government Speech
Lower courts have largely failed to address the range of behavioral and expressive harms potentially caused by government’s hateful speech when considering equal protection challenges to such expression. To date, these equal protection issues have arisen only in a very small number of lower court cases, but there the courts have been unwilling to credit equal protection challenges to government speech. The few cases in this area have focused on various constitutional challenges to governments’ expressive display of the Confederate flag. There the lower courts first required plaintiffs to show that the government’s expressive choice had both a discriminatory purpose and a discriminatory effect. They then defined the requisite discriminatory effects narrowly, found none, and thus avoided inquiry into whether the government’s expressive choice was actually motivated by a discriminatory purpose.
For example, in NAACP v. Hunt,4 the Eleventh Circuit found that Alabama’s expressive choice to fly the Confederate flag above the state capitol dome inflicted no discriminatory harm to African Americans. The panel concluded simply that whites as well as African Americans were offended by the flag’s display, without exploring any other possible harms nor the possibility that the white and black experience of offense might be different. Several years later, and for similar reasons, the Eleventh Circuit again rejected an equal protection challenge to Georgia’s incorporation of the Confederate flag into its own state flag design.5 By quickly concluding that the flag caused no constitutionally cognizable harm to African Americans, the courts avoided having to determine whether the state’s choice to fly the flag was motivated by a discriminatory purpose.
Contrast the experience with Establishment Clause challenges to government’s religious speech. There the Court has held that government speech violates the Establishment Clause upon a finding of either impermissible purpose or effect; moreover, it has defined the requisite harmful effects comparatively broadly. As just one example, lower courts have consistently held that government speech violates the Establishment Clause if the government expresses its preference for the Christian faith by adorning its city or county seal with a cross.6
Some courts propose a coercion analysis for determining when, if ever, government violates the Establishment Clause through speech or other action. Under this view, government speech does not violate the Establishment Clause unless and until it coerces behavior, such as compelling or coercing participation or attendance at a religious activity. In short, courts should find government to violate the Establishment Clause only when its speech or other action coerces behavioral change, rather than when it inflicts expressive harm.
To be sure, divisions remain even among advocates of coercion theory about that test’s application to specific facts. For example, Justice Kennedy defines impermissible “coercion” relatively broadly to include government speech that influences onlookers’ behavior through peer pressure and other social dynamics—as he found to be the case with a prayer at a public high school graduation.7 Justice Scalia, in contrast, defines impermissible “coercion” quite narrowly to mean only the threat or imposition of government punishment.8
Endorsement analysis, on the other hand, identifies the key harm imposed by an Establishment Clause violation as expressive in nature. Under this view, the government offends a constitutional commitment to religious pluralism when it delivers a message that citizens’ statuses vary based on their religion or nonreligion. Justice O’Connor, often identified as the champion of endorsement analysis, considered this to be true regardless of whether the government’s message inflicted emotional distress or other psychological or stigmatic harm upon onlookers.9
Although not without its controversy, the Court’s Establishment Clause doctrine offers potentially instructive alternative approaches to the equal protection problem. Once we acknowledge that government’s religious speech can unconstitutionally establish religion in at least some contexts where it causes behavioral or expressive harm, we should recognize that government’s hateful speech may similarly violate the Equal Protection Clause in some contexts where it causes such harms.
Reconsidering Equal Protection Challenges to Government’s Hateful Speech
The Establishment Clause experience invites us to consider the ways in which coercion and expressive meaning analysis might apply to equal protection challenges involving government’s hateful speech.
First, adopting a variant on coercion analysis in the equal protection context would require that we focus our inquiry on whether and when government’s discriminatory speech coerces its targets’ behavior or facilitates others’ discriminatory behavior. In other words, it would recognize behavioral harms caused by government’s hateful speech as constitutionally salient. This approach best explains Anderson v. Martin; recall that the Court there found unconstitutional the government’s racially identifying speech on ballots because the government’s speech facilitated private voters’ racially discriminatory behavior. Along these lines, we might shape our understanding of behavioral coercion to include government speech that expresses hostility on the basis of protected class status in a way that would cause a reasonable person to alter her behavior—for example, by deterring a reasonable person from engaging in a particular activity such as pursuing a housing or employment opportunity, or petitioning the legislature.
In contrast, applying endorsement or expressive meaning analysis in the equal protection context would require us to emphasize the expressive, rather than just the behavioral, harms of government’s hateful speech. Under such an approach, we would focus on whether the government has communicated a message of hierarchy or supremacy based on class status, regardless of whether the message succeeds in inflicting psychological damage on class members or lowering public opinion of class members—and certainly regardless of whether the message carries behavioral effects like encouraging private discrimination against, or deterring certain activity by, class members. Such an expressive meaning analysis would identify the constitutionally relevant harm as the government’s imprimatur on a message communicating that class members are outsiders or something other than full members of the political community.
Both behavioral and expressive meaning analyses offer the potential for more completely recognizing the array of harms potentially posed by government speech that expresses animus on the basis of protected class status—although, of the two, expressive meaning analysis casts the wider net. To be sure, neither is without its flaws.
Both approaches, and especially expressive meaning analysis, for example, are subject to substantial concerns about indeterminacy. To be sure, if one recognizes expressive harm as constitutionally significant, the hypotheticals posed at the beginning—“White Supremacy Forever” as the state motto—should offer easy examples of government speech that impermissibly delivers the governmental message that class members are second-class citizens. But of course few cases will be so easy, and many will invite controversy over the precise nature of the message delivered; the Confederate flag cases are particularly prominent examples. For this reason, many will wonder whether assessing the expressive meaning of government’s allegedly hateful speech is unacceptably indeterminate and subjective, leading to inconsistent and unprincipled results. In the Establishment Clause context, for example, those critical of endorsement analysis often similarly object that the effort to characterize a governmental message as endorsing or disapproving religion exceeds courts’ institutional competence.
But although the exercise of determining expressive meaning can be challenging, it is not without parallel elsewhere. Endorsement analysis itself assesses the perceptions of a reasonable observer when determining whether a governmental message endorses or disapproves of a particular religion in violation of the Establishment Clause. Along the same lines, statutory antidiscrimination law looks to the perceptions of a reasonable person when determining whether harassing speech in the workplace is sufficiently severe or pervasive to alter the terms and conditions of employment based on protected class status. To be sure, such assessments are often hotly contested, but courts have a considerable body of experience from which to draw.
In sum, government speech can inflict behavioral and expressive harms on the basis of race, gender, and other class status just as it can on the basis of religion. Unlike current Establishment Clause doctrine, however, contemporary equal protection doctrine does not yet fully contemplate the range of harms potentially posed by government’s hateful speech.
The recent emergence of the Court’s government speech doctrine—which to date has emphasized the inevitability and value of government expression without yet fully addressing its potential costs—offers a new opportunity to consider the situations in which government expression may inflict harms repugnant to equal protection values. In so doing, we might draw from our experience in assessing the potential Establishment Clause implications of government’s religious speech. Please note that I do not mean to suggest that these are the only, nor necessarily the best, approaches for addressing these issues. Instead, I seek to start a conversation about how courts—and the rest of us—might think about the harms, and thus the constitutional implications, of government’s hateful speech.
Helen Norton is Associate Dean for Academic Affairs and Associate Professor at University of Colorado School of Law.
A version of this article appeared in the October 2012 issue of the William & Mary Law Review.
- For the full-length version of this Article, see 54 Wm. & Mary L. Rev. 159 (2012). ↩
- 373 U.S. 267, 273 (1963). ↩
- 375 U.S. 399, 402-03 (1964). ↩
- 891 F.2d 1555 (11th Cir. 1990). ↩
- Coleman v. Miller, 117 F.3d 527 (11th Cir. 1997). ↩
- See, e.g., Robinson v. City of Edmond, 68 F.3d 1226, 1232 (10th Cir. 1995); ACLU v. City of Stow, 29 F. Supp. 2d 845, 853 (N.D. Ohio 1998). ↩
- Lee v. Weisman, 505 U.S. 577, 593-95 (1992). ↩
- Id. at 640 (Scalia, J., dissenting). ↩
- See Cnty. of Allegheny v. ACLU, 492 U.S. 573, 627 (1989) (O’Connor, J., concurring in part and concurring in the judgment). ↩
Post a Comment (all fields are required)
You must be logged in to post a comment.