USING THE FIRST AMENDMENT TO SAVE RACE-CONSCIOUS COLLEGE ADMISSIONS Open Access

Two of the Supreme Court’s current inequality-producing lines of cases are in tension. The tension remains latent; it has not yet turned into conflict or inconsistency. But one area of doctrine provides a way for creative and courageous plaintiffs to weaken, and take advantage of, the other.

The Court ended its 2022 Term with bold, activist pronouncements in two lines of cases that appear unrelated except for their inequality-producing characteristics. On June 29, 2023, the Court ruled on challenges to the admissions policies of Harvard University and the University of North Carolina,¹ announcing the end of so-called affirmative action. Colleges and universities can no longer consider race in admissions, even as part of a holistic review of individual applicants.² One day later, the Court issued its ruling in 303 Creative LLC v. Elenis,³ announcing that a for-profit company can claim a First Amendment exemption from anti-discrimination laws requiring the company to speak a message against its beliefs.⁴ Though both rulings made new law, both flowed from the Court’s prior cases. The university cases built on earlier Court judgments insisting on a simplistic notion of color blindness as a touchstone for official action.⁵ For its part, 303 Creative built on prior cases enlarging the possibility of organizations and companies winning exemptions from laws because of political and religious scruples.⁶

Both of the new rulings will promote inequality. In the university cases, the Court’s insistence on color blindness may ostensibly be based on the notion that, in the words of Chief Justice John Roberts, “the way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”⁷ But little doubt exists that the ultimate impact of the Court’s ruling will be less diverse, and more unequal, college campuses.⁸ Colleges and universities that see themselves as having an obligation to take race into account to create a campus community that embodies their social, moral, ideological, religious, and educational commitments will be prohibited from doing so. Our educational system will be less reflective of a commitment to equality, and society will be less equal because of it. The Court’s insistence on facial equality will undermine the pursuit of substantive equality.

303 Creative, too, will undermine equality. Companies will now be able to assert First Amendment bases to be exempted from federal, state, or local laws protecting customers and other market participants from discriminatory treatment. The marketplace will now lack the consistent protection against discrimination that public accommodation law has long provided. Indeed, some companies—like 303 Creative itself, or Masterpiece Cakeshop⁹—can seek a market niche based on their religiously driven discriminatory perspective.¹⁰

This article is not about each of these cases in isolation, nor is it principally a critique of either—though both are fundamentally unsound in numerous ways.¹¹ Instead, this article proposes a way to lessen the harsh implications of the university cases by using the First Amendment doctrine on which 303 Creative builds and relies. There is a way in which these cases are in conflict, and a creative plaintiff could use the line of cases culminating in 303 Creative in a jurisprudential jujitsu move to weaken the strength and breadth of the university cases.

The conflict is between the Court’s insistence in the university cases that any attentiveness to race should be considered discrimination and the Court’s willingness to allow institutions to assert political and religious bases to discriminate. In the university cases, the Court explained that for public universities such as the University of North Carolina, attention to race amounts to discrimination that violates the Equal Protection Clause.¹² For private institutions such as Harvard, the Equal Protection Clause does not apply to limit their conduct. Instead, the Court held that private universities’ race consciousness amounts to a violation of Title VI of the of the Civil Rights Act, which applies to any institution receiving federal funds.¹³ Title VI prohibits “discrimination” by such institutions “on the ground of race.”¹⁴

Meanwhile, the Court has now afforded companies and nonprofits the ability to claim First Amendment–based exemptions from statutes that prohibit discrimination. Private universities and colleges with political, religious, or other First Amendment–protected bases to use race in admissions should therefore be able to claim a First Amendment–based exemption from any application of Title VI that would consider such use of race as “discrimination.”

A university’s assertion of a First Amendment right to use race in admission would indeed be an extension of the line of cases culminating in 303 Creative. But it is an extension that flows naturally from the cases already decided. Moreover, the First Amendment interests that many universities can claim are stronger, more persuasive, and more long-standing than those that formed the gravamen of the claims in 303 Creative and its forebears.

This article explores whether and how a private educational institution should raise a First Amendment claim in support of the use of race in its admission process. In part I, the article more deeply explains the conflict between the two lines of doctrines and describes how these cases would provide the framework for a private university’s First Amendment suit to protect its use of race in admissions. Part II provides a deep analysis of the First Amendment cases that would form the backbone of any claim by the universities. Part III explores how a claim by a university that it possesses a First Amendment interest in using race in admissions would flow from and extend existing doctrine; this part will analyze in particular the fact that a university claim would require the Court to extend the First Amendment cases to discrimination on the basis of race—a move the Court has not explicitly made to date. Part IV addresses the pros and cons of bringing such a case, including the possible cost of extending an already unfortunate line of doctrine to protect more nefarious behavior and the possible benefit of saving affirmative action for a good number of private universities.¹⁵ Whether one or more universities will choose to move forward will be a question for the universities themselves. This article provides the legal grounding for them to make that judgment.¹⁶

In the UNC and Harvard cases, the Court insisted that colleges and universities make admissions decisions that are color-blind—that is, in a way that is inattentive to the race of applicants. Chief Justice Roberts’s opinion on behalf of the Court ruled that color blindness is required by the nation’s commitment to racial equality and inconsistent with the principle that universities treat each applicant on their own merits. In the Court’s view, the universities’ desire to achieve diversity in their student communities is incoherent. Chief Justice Roberts described the goal of diversity as “standardless” and “amorphous”¹⁷ and thus insufficiently coherent to be justiciable.¹⁸ And if not justiciable, it cannot suffice as a compelling interest necessary to survive strict scrutiny.

Chief Justice Roberts’s opinion justifies serious critique and criticism, which has been forthcoming from numerous scholars and Court watchers.¹⁹ The opinion presumes that attentiveness to race and a dedication to merit are inconsistent, even though most universities do not in fact consider them so.²⁰ The opinion does not acknowledge that as a matter of substance it overrules cases—most notably Grutter v. Bollinger²¹ and Fisher v. University of Texas²²—in which the Court held that diversity could amount to a compelling governmental interest to justify a university’s limited use of race in an individualized, holistic admissions process.²³ The Court seemed to assert its own view of the facts with regard to both UNC’s and Harvard’s admissions practices, even though the plaintiffs’ claims had been subject to extensive fact finding by courts below,²⁴ to which the Court would normally defer.²⁵

Another valid critique would point out that notwithstanding the Court’s severe and apparently definitive language warning against any use of race, the Court allowed at least two exceptions to its ruling. The first was an allowance for students to raise their racial experiences in essays or other statements as a way to illustrate characteristics and traits that the university may find important, such as fortitude, resilience, and courage.²⁶ The Court said, “nothing in this opinion should be construed as prohibiting universities from considering an applicant’s discussion of how race affected [their] life, be it through discrimination, inspiration, or otherwise.”²⁷ But, the Court insisted, “the student must be treated based on [their] experiences as an individual—not on the basis of race.”²⁸ The Court did not explain the distinction between that allowance and what most universities were doing before the ruling, because even under Grutter, race was available only as a consideration as a part of an individualized, holistic review of each applicant’s file. What Chief Justice Roberts did say, however, was that “universities may not simply establish through application essays or other means the regime we hold unlawful today. … [W]hat cannot be done directly cannot be done indirectly.”²⁹

The second exception is even more difficult to explain, given the Court’s stated rationales. In a footnote, Chief Justice Roberts declined to apply its ruling to the military service academies, deferring to the academies’ defense of diversity on their campuses.³⁰ Thus, in an opinion announcing a robust dedication to ending discrimination by “eliminating all of it, ”³¹ the Court allowed that very “discrimination” to continue in the nation’s only federally run universities.

Even though the Court did not admit to overturning Grutter and Fisher,³² the nation’s colleges and universities were not deceived. Institutional administrators have almost uniformly interpreted the ruling to bar any use of race in their admissions practices outside the narrow allowance of students raising their racial identity in their essays.³³ Many universities have announced plans to increase their efforts to maintain some semblance of diversity on their campuses without the use of race,³⁴ and it is highly likely that admissions committees will increase the use of proxies that substitute for racial data, and it is highly likely that such proxies will be imperfect.³⁵ The extent of the ruling’s impact on the racial makeup of student communities remains unclear, as does whether the universities’ substituted proxies will be targets of follow-on litigation. Even so, there is little doubt that the ruling will have a serious impact on higher education, and it is highly likely that those universities using race in admissions before the rulings will see their campus communities become less racially diverse in the near future.³⁶

Looking ahead, some private schools may have a way to mitigate the ruling’s effects by building on the Court’s First Amendment doctrine. To understand how, it is important to recognize that the UNC and Harvard cases were differently situated as to the legal basis for the plaintiffs’ claims. The suit against the University of North Carolina was based on the Equal Protection Clause of the Fourteenth Amendment. UNC is a public institution and thus subject to constitutional constraints. The ruling’s insistence on color blindness was based on its constitutional judgment that the Equal Protection Clause imposes on public universities an obligation to prioritize formal over substantive equality.

Harvard, in contrast, is not a public institution, so it is unconstrained by constitutional limitations. Private parties are rights holders under the constitution, not rights obligors.³⁷ The case against Harvard was instead based on Title VI of the Civil Rights Act of 1964. That law bars racial discrimination by any recipient of federal funds.³⁸ The specific provision, 42 U.S.C. § 2000d, states that “[n]o person … shall, on the ground of race, … be subjected to discrimination under any program or activity receiving Federal financial assistance.”³⁹ But the Court, unconcerned about the distinction, explained in a footnote that the statute creates the same requirements for private universities that the Equal Protection Clause does for public ones.⁴⁰ As a matter of basic statutory interpretation, that assumption is highly questionable.⁴¹ It is unlikely that the Congress that passed the Civil Rights Act meant the term “discrimination” to bar any attention to race when remedial and reparative attention to race was widely acknowledged as necessary to address nearly a century of Jim Crow.⁴² But the Harvard Court did not engage in any kind of statutory construction of § 2000d. The footnoted assumption was, for the majority, enough.⁴³

In any event, it cannot be ignored that the Court had two legal bases for its ruling. As regards public schools, it is the Equal Protection Clause that constrains them. As regards private schools, it is Title VI of the Civil Rights Act that tells them that “discrimination” is unlawful. While the different bases could have provided a distinction between UNC and Harvard in the affirmative action cases, the Court treated them the same.

The difference between the bases for the two suits could have significant implications going forward. The most obvious one is that because the ruling in the Harvard case is based on a statute, it could be changed by statute. Congress could clarify that private universities and colleges that use race in a limited, targeted way—as they had for decades under the old interpretation of the Civil Rights Act—are not engaging in illegal “discrimination” under Title VI. Such a clarification would require Congress to pass a new provision and the President to sign it into law. Congressional action is highly unlikely, however, given the lack of political consensus on the issue—not to mention the institutional gridlock and chaos that characterizes the body in our current era.

Another implication of the difference is that the breadth of the ruling spreads beyond public actors such as state colleges and universities to all private actors receiving federal funds. Because of Title VI’s sweep—it covers any “program or activity”—the Harvard ruling would seem to impose the equivalent of constitutional obligations on any institution receiving federal funds. This would include most or all private colleges and universities, but also any other business, nonprofit group, hospital, research organization, or even law firm that receives federal funds.⁴⁴ The Court’s reasoning and analysis was limited to the university contexts of the two defendants, and the need—or, in the Court’s view, the lack thereof—for diversity in education. But the statutory basis for the claim provides a hook for suits against a broad range of potential defendants.

Another implication is the focus of this article. Because the requirement of private schools to be color-blind comes by way of a statutory requirement, those schools have access to a First Amendment argument to be exempted from that statutory obligation. The framework for such a claim would be the framework the Court has endorsed in cases such as 303 Creative: institutions have a First Amendment right to be exempted from anti-discrimination laws that force them to speak or associate in ways that are contrary to their beliefs.

This tension between the Court’s affirmative action cases and its First Amendment cases is easy to miss because the language and reasoning of the Harvard case sounded in constitutional values. If the source of institutions’ obligation to be color-blind were indeed constitutional in nature, no institution would have the ability to claim an exemption from it. After the Court’s ruling, for example, the University of North Carolina’s obligation to administer color-blind admissions is a constitutional obligation that cannot be avoided by reference to any oppositional constitutional right. Indeed, state actors such as UNC do not have cognizable free-speech rights vis-à-vis other government entities. To the extent that state actors such as UNC have interests to weigh in the balance, they are not constitutional in nature. Instead, they are only the kind of governmental interests to be weighed in the last stage of any constitutional rights–based litigation when courts apply the proper standard of scrutiny. In strict scrutiny, the court will ask whether the government has a compelling interest to burden the plaintiff’s constitutional interest and whether the government’s action has been narrowly tailored to pursue that interest.⁴⁵ And pursuant to the Court’s ruling in UNC, a public university’s interest in diversity is now insufficiently compelling to outweigh a plaintiff’s constitutional interest in being treated in a color-blind way.

With regard to private institutions, however, that analytical framework makes little sense. Harvard is not a government actor, and the strict scrutiny framework of “compelling interest, narrowly tailored” is thus incoherent. Private parties are not allowed to violate a statute by showing that they have a really good reason to do so and they violated it only as much as necessary. Instead, the framework for a statutory claim under Title VI is an allegation that a private college or university is “discriminating” “on the ground of race” by failing to be color-blind in its admissions policies. The university would not be able to assert a narrowly tailored compelling interest as a defense to the claim because it is not a public actor. But because it is a private actor, the university would be able to assert an even more powerful defense: a First Amendment interest in the continued use of race.

A college or university asserting a First Amendment interest as a defense to a claim arising under Title VI would fit into the analytical framework the Court has already recognized numerous times, most recently in 303 Creative. In that framework, the business or organization faces a statutory obligation to refrain from discrimination. The business or organization asserts that such an obligation burdens its speech or associational interests and that the First Amendment affords an exemption from the otherwise-applicable statute.

In such a framework, the Court typically takes as given the sincerity of the beliefs asserted by the organization as the basis of the First Amendment claim. The Court then considers the question of whether the anti-discrimination law imposes a burden on those beliefs. In 303 Creative, for example, the Court looked to whether the law coerced the business to speak views contrary to its beliefs.⁴⁶ But when private actors assert a First Amendment basis for a legal exemption, the beliefs on which that claim is based are not evaluated for their coherence, importance, or “compelling-ness.”⁴⁷ If they are the kind of beliefs that have First Amendment weight, there is no requirement that the beliefs be “compelling” in the view of the Court. They need to be beliefs, sincerely held, but they need not satisfy any importance standard.

This framework would thus be liberating for any university or college facing a Title VI challenge to its admissions policies and practices. Institutions would no longer be limited to the notion that attention to race is necessary to further diversity. The concept of diversity was important in the equal-protection framework because it was the only compelling interest the Court had recognized to justify a public school’s use of race in admissions.⁴⁸ But once the analytical framework shifts to private institutions asserting First Amendment interests as a basis to be exempted from otherwise applicable anti-discrimination statutes, diversity need not be the only basis for such an exemption. Private schools can assert any number of political, religious, or other expressive interests to support their use of race. For example, a religious school could assert that its consideration of race in admissions is commanded by their religious beliefs. Or a school could say that a color-blindness rule burdens its capacity to construct a student body that expresses its beliefs about the importance of pluralism. An institution could claim that its mission includes a goal to address systemic racism in America, that working toward that goal is an important part of how it expresses itself as an institution, and that a color-blindness requirement burdens its ability to work toward that goal. An institution could say that it has the same right to decide its membership as any expressive association, and that the obligation to ignore race in building its membership burdens its ability to build its membership as it chooses. Or an institution could claim that the insistence on color blindness burdens its ability to speak honestly about its dedication to race consciousness within its community. Or a college could say that an attention to race in admissions is one way it expresses its commitment to address, ameliorate, or repair the costs of its own past behavior.⁴⁹

The framework of these claims is different because the identity of the constitutional claimant is different. In the case of a public school using race, the student classified on the basis of their race is the constitutional claimant. The school is the government actor. In the case of a private school using race, the private school is not a government actor but rather the constitutional claimant. The government actor in the latter case would be the federal government, which is imposing an obligation of color blindness by way of Title VI enforced by the Department of Education.

In the latter framework, the compelling-interest standard would not vanish. But it would fit into the analysis only as the final question: does the federal government have a compelling interest to require an institution to abide by an anti-discrimination statute that burdens its First Amendment interests? In his opinion in 303 Creative, Justice Gorsuch ignored this last step of the analysis even though the dissent pointed out the incompleteness.⁵⁰ Perhaps he ignored it because once the Court decides that a burdened interest demands strict scrutiny, it only rarely decides that a narrowly tailored compelling interest justifies the burden.⁵¹ The application of strict scrutiny is usually “fatal in fact.”⁵² Nevertheless, any court evaluating a private institution’s assertion of a First Amendment interest in considering race in admission would have to weigh the government’s putative compelling interest in imposing color blindness on that school.⁵³

The First Amendment may thus provide a way for some private schools to avoid the Harvard ruling. While the Court has made clear as a matter of equal protection law that color blindness—formal equality—is required of public actors, it has meanwhile recognized First Amendment–based exemptions to otherwise applicable anti-discrimination statutes for private institutions and parties. The Court has not to date been forced to reconcile the tension between its insistence on color blindness in the affirmative action context and its protection of First Amendment exemptions to anti-discrimination requirements. But as soon as a private school asserts a First Amendment interest in the continued use of race in admissions, it will have to.

The strength of a university’s First Amendment claim to an exemption from Title VI’s new color-blind definition of “discrimination” will of course turn on how well the claim fits into existing First Amendment doctrine and precedent. This part analyzes that question, exploring the First Amendment cases that would form the backbone of any claim by private universities. The Court has, for several decades now, recognized the availability of First Amendment defenses to statutory anti-discrimination laws. This trend shows no sign of slowing. Indeed, the 303 Creative ruling expanded the doctrine, allowing for the first time a for-profit company to claim a First Amendment waiver from otherwise applicable anti-discrimination law. This part will describe the key precedents and the nature of the First Amendment interests the Court has recognized as being available to institutions wishing to claim an exemption from anti-discrimination laws.

The Supreme Court cases weighing the tension between the First Amendment and anti-discrimination laws have recognized several ways that First Amendment interests are burdened by such laws. Some are speech cases; some are association cases; some are hybrids. In 303 Creative, the interest was speech—the company claimed that the anti-discrimination law burdened its First Amendment interests by forcing it to speak in ways that violated its beliefs.⁵⁴ In Boy Scouts of America v. Dale, the interest was associational—the organization claimed its First Amendment interests were burdened by a statute that prohibited the BSA from terminating a scout master for his sexual orientation.⁵⁵ In Hurley v. Irish-American Gay, Lesbian, and Bisexual Group of Boston, Inc., the parade organizers’ claim depended on both associational and speech interests—the anti-discrimination statute would force the organizers to include a group (associational interest) proclaiming a viewpoint the organizers did not want to endorse (speech interest).⁵⁶ The Court distinguishes speech and association interests as an analytical matter, but it does not prioritize speech over association, or vice versa. The 303 Creative Court, in fact, described the First Amendment as protecting association and speech “equally.”⁵⁷ Both are core to First Amendment protections. The Court has been less consistent as to the level of constitutional scrutiny it applies in such cases and what kind of interests the government could successfully assert—if any—to satisfy the proper level of scrutiny. The following sections present the main cases.

Over the past several decades, the Court has wrestled with the problem of whether and when private groups may assert a First Amendment claim to be exempted from anti-discrimination statutes that would otherwise apply to them. While the issue lurked in the background during the Civil Rights era, when many businesses objected to the obligations of anti-discrimination statutes,⁵⁸ the Court did not articulate a framework for analysis of the question until its 1984 decision in Roberts v. U.S. Jaycees.⁵⁹ Until the suit, the Jaycees allowed women as “associate members” of the organization but permitted only men to be voting members.⁶⁰ The organization was sued pursuant to a Minnesota state statute that prohibited discrimination on the basis of sex, among other things.⁶¹

The Court explained that there are two kinds of associations that receive constitutional protection. The first, “intimate” associations, are those that pertain to those “highly personal relationships” deserving “a substantial measure of sanctuary from unjustified interference by the State.”⁶² These kinds of intimate associations are protected, the Court explained, because of their role in “safeguard[ing] the ability independently to define one’s identity that is central to any concept of liberty.”⁶³ They primarily concern familial and other close, intimate connections, and are much more a concern of the Due Process clause than of the First Amendment.⁶⁴

The second kind of association with constitutional relevance is the “expressive association,” which does not require intimacy but does further the collective’s interest in a point of view—a “collective effort on behalf of shared goals.”⁶⁵ The right to gather in a group with other like-minded individuals—the “right to associate with others in pursuit of a wide variety of political, social, economic, educational, religious, and cultural ends”—is protected by the First Amendment because such collective action is “implicit in the right to engage in activities protected by the First Amendment.”⁶⁶ This right includes both the right to invite into the group only those that the group desires to include⁶⁷ and the right to exclude those the group wants to exclude. “Freedom of association therefore plainly presupposes a freedom not to associate.”⁶⁸ The Court made clear that “forc[ing] [a] group to accept members it does not desire” is a paradigmatic example of “intrusion into the internal structure or affairs of an association” that the First Amendment finds problematic.⁶⁹

The Court recognized the Jaycees as an expressive association whose constitutional interests were “plainly implicated” in the case.⁷⁰ Nevertheless, the Court held that the group’s membership policy was subject to the state’s anti-discrimination statute and required it to admit women as full members.⁷¹ The Court reasoned that the inclusion of women into its voting membership would not burden the association’s expressive interests, since there was nothing necessarily inconsistent between the group’s stated viewpoints and the inclusion of women as voting (rather than “associate”) members.⁷² “The Jaycees has failed to demonstrate that the Act imposes any serious burdens on the male members’ freedom of expressive association.”⁷³ The Court emphasized that there was “no basis in the record for concluding that admission of women as full voting members will impede the organization’s ability to engage in these protected activities or to disseminate its preferred views.”⁷⁴

It is noteworthy that the Court did not defer to the group’s own view of whether the inclusion of women as full members would burden its interests. Indeed, the Court criticized the group’s exclusion of women as being based on “sexual stereotyping” that the Court had “repeatedly condemned.”⁷⁵ The Court’s lack of deference to the group’s own articulation of its need to exclude becomes even clearer when compared to later cases (Hurley, Dale, 303 Creative) in which the Court was more trusting of the group’s own explanation of its mission and the burden imposed on that mission by anti-discrimination laws.

The Court also remarked that the limit on the group’s associational interest was justified by “compelling state interests of the highest order.”⁷⁶ The Court explained that Minnesota’s anti-discrimination statute was based on its “compelling interest in eradicating discrimination against its female citizens.”⁷⁷ It emphasized that “acts of invidious discrimination in the distribution of publicly available goods, services, and other advantages cause unique evils that government has a compelling interest to prevent.”⁷⁸

This is the language of strict scrutiny, which requires that the government show a compelling interest. But the opinion is curious in that the Court never says it is applying strict scrutiny. Strict scrutiny is typically applied in cases in which a law or government act is aimed at speech and is content or viewpoint based.⁷⁹ But in Roberts the Court makes clear that the anti-discrimination statute was not aimed at speech. It was generally applicable, “not aim[ed] at the suppression of speech,” and did “not distinguish between prohibited and permitted activity on the basis of viewpoint.”⁸⁰ Moreover, the Court says, the law was not “applied … for the purpose of hampering the organization’s ability to express its views.”⁸¹ In comparable situations in which a law is aimed at conduct and not at speech, but which has incidental effects on speech, the Court applies a lower level of scrutiny. The Court announced in the 1968 case of United States v. O’Brien something akin to intermediate scrutiny: a regulation of conduct is “sufficiently justified … if it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression, and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest.”⁸² But even though the Roberts Court seemed to reference the O’Brien standard by saying the law was “not aim[ed] at the suppression of speech,” it did not say it was applying O’Brien intermediate scrutiny either.

The Roberts opinion is probably best read as hedging on scrutiny—that is, the lack of clarity was intentional. Analytically, strict scrutiny was not required given that the statute was not aimed at speech and was not content or viewpoint based. Moreover, strict scrutiny was not necessary given the Court’s view that the anti-discrimination statute did not impose any meaningful burden on the group’s associational interests, that the requirement to include women did not “impede the organization’s ability to engage in [] protected activities or to disseminate its preferred views.”⁸³ Even so, the Court thought it was important to say that the statute was justified by the “compelling interest” in fighting discrimination.⁸⁴ The Court wanted both belt and suspenders in upholding the statute: the law had to satisfy only O’Brien intermediate scrutiny but would meet strict scrutiny if it had to.

The Court’s lack of clarity on the proper level of scrutiny has implications for understanding later association cases and, indeed, the claims of universities that might bring First Amendment claims to be exempted from a color-blind Title VI. Instead of allowing either belt or suspenders to uphold a statute, the Court has since allowed the statutes to fall by ignoring the presence of both. The Court in subsequent cases has either ignored the level-of-scrutiny question or assumed that strict scrutiny applies. And when it applies scrutiny, the Court has not focused on the question of whether the government’s interest in fighting discrimination is important, much less compelling. In the First Amendment cases since Roberts, the Court has never said that an anti-discrimination statute is justified by a compelling interest. Indeed, in no case since Roberts has the Court held that First Amendment interests should yield to the government purpose of fighting discrimination.

The next time the Court grappled with the tension between anti-discrimination laws and the First Amendment was over a decade later in a case concerning the Saint Patrick’s Day parade in South Boston. The parade had long been a celebration of Irish tradition and community in “Southie.” One of the city’s most popular annual events, the parade had at times boasted as many as 20,000 marchers and 1 million spectators.⁸⁵ The case arose when an organization of gay, lesbian, and bisexual descendants of Irish immigrants (going by the shorthand “GLIB”) petitioned for permission to march, and the parade organizers—led by the memorably named John “Wacko” Hurley—refused to let them join.⁸⁶ The applicable statute prohibited “any distinction, discrimination or restriction on account of … sexual orientation … relative to the admission of any person to, or treatment in any place of public accommodation, resort or amusement.”⁸⁷ A state court ruled the parade was a public accommodation and held that the state anti-discrimination law required the inclusion of the organization, and the state supreme court affirmed.⁸⁸

The Supreme Court disagreed.⁸⁹ In an opinion by Justice David Souter, the Court described parades as paradigmatic speech. “Parades are … a form of expression, not just motion,” the Court said, noting the “inherent expressiveness of marching to make a point.”⁹⁰ The Court went further to say that parades do not lose their expressive quality even if they lack “a narrow, succinctly articulable message.”⁹¹ The parade need not have a single point to make and can include “multifarious voices.”⁹² Parades may claim First Amendment protection even if they “fail[] to edit their themes to isolate an exact message as the exclusive subject matter of the speech.”⁹³ That is, the decision to exclude GLIB and its message from the Southie parade did not need to be justified by an assertion of a message made incoherent by GLIB’s presence. All the organizers had to say was that it did not want to be the carrier of GLIB’s message. And because the state’s public accommodation law would “essentially requir[e] petitioners to alter the expressive content of their parade,” it had to yield to the parade’s constitutional interests.⁹⁴ “The fundamental rule of protection under the First Amendment,” said the Court, was that “a speaker has the autonomy to choose the content of his own message.”⁹⁵

The Court’s description and analysis nodded to both speech and association constructs. The parade is expressive, both by what the groups proclaim while marching (speech) and by what groups the parade includes and excludes (association). The choice of which groups to invite and refuse (that is, to associate with) is itself an expression, even if the choice is based on unstated, inexact, or inchoate points of view. The parade was distinct from the Jaycees because the parade did not exclude GLIB or individual gay or lesbian marchers because of their status alone.⁹⁶ Instead, the exclusion of GLIB was based on GLIB’s desire to engage in expression themselves. “The Council clearly decided to exclude a message it did not like from the communication it chose to make, and that is enough to invoke its right as a private speaker to shape its expression by speaking on one subject while remaining silent on another.”⁹⁷

By making the case about whether a group parading with a message must include another group with a conflicting message, the Court avoided the most difficult question of association claims: whether an association may claim a constitutional basis to deny membership solely because of an applicant’s status. In Roberts, the Court assumed that the First Amendment could support such a claim but also decided that the sex of the excluded individuals did not in itself carry a message that conflicted with the message of the Jaycees. It also decided that to the extent sex was the basis of the exclusion, the government had a constitutionally sufficient reason to require the Jaycees to include women. Hurley, in contrast, was a message-versus-message case. That allowed the Court to defer to the parade organizers’ assertion that the forced inclusion of GLIB’s message would conflict with the parade’s.

Moreover, understanding the case as a message-versus-message case allowed the Court to avoid the tricky question of the level of judicial scrutiny left hazy in Roberts. The government’s interest could not be that of fighting discrimination, as in Roberts, since the exclusion was based on message rather than status. The Court interpreted the state’s insistence on declaring the parade a public accommodation to be equivalent to declaring “the sponsors’ speech itself to be the public accommodation.”⁹⁸ The Court recognized that the purpose of the state law was

That might have sufficed in a case in which status was the basis of exclusion. But as applied to expression qua expression, the state’s “object is merely to allow exactly what the general rule of speaker’s autonomy forbids.”¹⁰⁰ And that interest would not satisfy the level of scrutiny whatever it was.

The Court seemed to acknowledge the uncertainty around the proper level of scrutiny. Citing several precedents, including O’Brien, the Court described the “threshold requirement of any review under the Speech Clause, whatever the ultimate level of scrutiny, that a challenged restriction on speech serve a compelling, or at least important, governmental object.”¹⁰¹ Justice Souter was wise, in all likelihood, to avoid the issue of whether the statute could be justified by a government interest in fighting discrimination against LGBTQ citizens. Hurley arose eight years before Lawrence v. Texas and twenty years before Obergefell v. Hodges.¹⁰² The Court had not recognized sexual orientation as a suspect class, and the lack of such recognition would probably have doomed any argument that the government had “a compelling, or at least important” interest in fighting discrimination against LGBTQ marchers.¹⁰³ It would have been quite difficult indeed to convince a majority of the Court to hold that the Massachusetts public-accommodation statute satisfied heightened scrutiny when applied to sexual orientation discrimination when other states still had operative criminal sodomy statues.¹⁰⁴

Hurley thus stands for an important and straightforward—but limited—proposition: an anti-discrimination statute cannot force an association to include within its own message a message with which it disagrees.

The most relevant case in evaluating the possible claims of universities to be exempted from a color-blindness rule is Boy Scouts of America v. Dale.¹⁰⁵Roberts and Hurley set the stage. Dale is the main act.

At issue in Dale was whether the Boy Scouts of America (BSA) could claim an exemption from New Jersey’s public-accommodation law when a local subdivision of the scouts dismissed an assistant scoutmaster who was openly gay.¹⁰⁶ The Boy Scouts placed central importance on inculcating its young members with the organization’s system of values and claimed that a sincere belief in the wrongness of homosexuality was among those values.¹⁰⁷ In general terms, then, the BSA’s claims fit the same form as those of the Jaycees and the South Boston parade organizers: the organization asserted a First Amendment interest to be excused from an anti-discrimination law that would be applicable otherwise. And as in Hurley, the Court held in favor of the organization claiming an exemption from an otherwise applicable anti-discrimination law.

As for the substance of the BSA’s asserted First Amendment interest, the evidence that the Boy Scouts had a long-standing policy against gay scoutmasters was thin. The organization could present no written employment policy on point that predated Dale’s dismissal.¹⁰⁸ The best the BSA could do was point to phrases in its “scout oath” urging scouts (and presumably scout masters) to be “morally straight” and in its “scout law” calling on scouts to be “clean.”¹⁰⁹

The New Jersey Supreme Court had held in Dale’s favor in part because it found the BSA’s assertion of beliefs unconvincing.¹¹⁰ Like the Roberts Court vis-à-vis the Jaycees, the state court took a hard look at the consistency of the organization’s purported associational interests. It concluded that the BSA’s exclusion of gay scouts and scoutmasters “contradict[ed] [its] overarching objective to reach ‘all eligible youth’” and “appear[ed] antithetical to the organization’s goals and philosophy.”¹¹¹ If the asserted belief was pretext, the BSA’s asserted interest would not bear the weight of the constitutional claim.

The United States Supreme Court, on the other hand, avoided a hard look at the group’s asserted beliefs, saying instead that such an inquiry was inappropriate.¹¹² In an opinion by Chief Justice Rehnquist, the Court deferred to the BSA’s own description of its interests and beliefs. In response to the New Jersey court’s review of the consistency of the group’s views, Chief Justice Rehnquist said that “our cases reject this sort of inquiry; it is not the role of the courts to reject a group’s expressed values because they disagree with those values or find them internally inconsistent. … We accept the Boy Scouts’ assertion.”¹¹³ This was an implicit mischaracterization of the Court’s precedent in Roberts, in which the Court analyzed whether the public-accommodation law posed a burden on an organization’s First Amendment interests. The Roberts Court did not hold that the presence of women in the Jaycees was required because the Court “disagreed with [the] values” of the Jaycees or even that the Jaycees had views that were “internally inconsistent.” The Court instead looked at whether the statute would burden the group’s associational interests by forcing the inclusion of someone who would make it impossible for the group to “disseminate its preferred views.”¹¹⁴

In Roberts, the Court did not defer to the organization’s own judgment as to whether the presence of the individual(s) in question would make such dissemination impossible. In Dale, however, the Court offered double deference: “as we give deference to an association’s assertions regarding the nature of its expression, we must also give deference to an association’s view of what would impair its expression.”¹¹⁵ The Court deferred to, first, the organization’s own description of its preferred view even in the absence of clear evidence with regard to the sincerity or even existence of that point of view. Second, the Court deferred to the organization’s assertion that an anti-discrimination statute burdened its ability to disseminate that view.

As Justice Stevens said in a robust dissent for four justices, the majority’s rationale amounted to a big change from Roberts.¹¹⁶ The views of the Jaycees about women were both clearer, more documented, and more long-standing than any views the BSA held about gay scouts and scoutmasters. The Roberts Court had more basis to defer to the Jaycees’ views as to whether the presence of women was inconsistent with the view it wanted to disseminate or embody than the Dale Court had to defer to the BSA. Indeed, if the Court had deferred to the views of the Jaycees as to whether the state law would burden or impair its First Amendment interests, then the Jaycees would have won on that issue as well.

The Dale Court did not admit it was changing the law and instead tried to distinguish Roberts on its facts. In Roberts, the Dale Court said, the Jaycees had “failed to demonstrate any serious burdens on the male members’ freedom of expressive association.”¹¹⁷ The New Jersey statute burdened the interests of the BSA; the enforcement of the state statute in Roberts, on the other hand, did “not materially interfere with the ideas that the organization sought to express.”¹¹⁸ This distinction is unpersuasive, as it depends completely on the level of deference the Court applies to the claimants’ descriptions of their interests, their asserted viewpoints, the asserted inconsistency between those viewpoints and the anti-discrimination law in question, and whether the inconsistency materially burdens First Amendment interests. As Justice Stevens argued, “this is an astounding view of the law. … It is even more astonishing in the First Amendment area, because, as the majority itself acknowledges, ‘we are obligated to independently review the factual record.’”¹¹⁹ Justice Stevens was snide: “it is an odd form of independent review that consists of deferring entirely to whatever a litigant claims.”¹²⁰ In the absence of an independent inquiry, Stevens said, the Court “turn[s] the right to associate into a free pass out of antidiscrimination laws.”¹²¹

Dale was thus not merely an application of the Roberts insight that expressive associations have a First Amendment interest in excluding members whose presence would be burdensomely inconsistent with the group’s expression. The Dale Court upped the level of deference in important ways, making it more possible to assert newly muscular First Amendment claims against anti-discrimination and public accommodation statutes.

Dale made two other important adjustments from the Hurley and Roberts precedents. The first went to the question of the speech/status distinction, the issue that Hurley had artfully avoided. The second went to the level of scrutiny, which Roberts had hedged. The Dale Court decided both in ways that put a finger on the expression side of the balance between expression and anti-discrimination.

Hurley had made clear it was not holding that the parade had a First Amendment right to exclude the LGBTQ marchers because of their status: “Petitioners disclaim any intent to exclude homosexuals as such. … Instead, the disagreement goes to the admission of GLIB as its own parade unit carrying its own banner.”¹²² In the oral argument in Dale, the Court appeared to be ready to base its decision on this distinction. Much of the questioning sought to nail down whether the BSA was excluding Dale on the basis of his status or because he wanted to use his position as a soapbox to espouse his viewpoint about sexual orientation.¹²³

But Rehnquist’s Dale opinion does not hand wring about the issue. Instead, the opinion implicitly equates status with viewpoint. “Dale’s presence in the Boy Scouts would, at the very least, force the organization to send a message, both to the youth members and the world, that the Boy Scouts accepts homosexual conduct as a legitimate form of behavior.”¹²⁴ And “the presence of Dale as an assistant scoutmaster would … interfere with the Boy Scouts’ choice not to propound a point of view contrary to its beliefs.”¹²⁵ If presence is communication, as the Court says, it becomes exceedingly difficult to distinguish status from viewpoint. The Court appears to go even further, implying that status can be more imbued with viewpoint than even pure communication of a viewpoint. The Court said that the Scouts would be within its rights to exclude a gay scoutmaster on the basis of status even if it did not choose to exclude a straight scoutmaster who espoused a message of “disagree[ment] with the Boy Scouts’ policy on sexual orientation.”¹²⁶ “The presence of an avowed homosexual and gay rights activist in an assistant scoutmaster’s uniform sends a distinctly different message from the presence of a heterosexual assistant scoutmaster who is on record as disagreeing with Boy Scouts policy.”¹²⁷

Where Hurley made clear that it was opining only about the message-versus-message case, Dale expands that holding by saying status can be message.¹²⁸ And—lest one forget—whether status is message in a particular case is something the Court will defer to the organization to decide.

As for the level-of-scrutiny question, Dale argued that expressive-association claims for an exemption from public-accommodation statutes should receive the intermediate scrutiny of O’Brien. As discussed above, the O’Brien test seemed to fit exactly. That test is used when a statute aimed at conduct—“unrelated to the suppression of free expression”—is applied in a way to impose an “incidental restriction on alleged First Amendment freedoms.”¹²⁹ Public-accommodation laws are aimed at conduct but impose incidental burdens on expressive associations. Under O’Brien, then, the statute should survive if it “furthers an important or substantial governmental interest.”¹³⁰

But Chief Justice Rehnquist distinguished O’Brien. “A law prohibiting the destruction of draft cards only incidentally affects the free speech rights of those who happen to use a violation of that law as a symbol of protest.”¹³¹ In contrast, Chief Justice Rehnquist said, “New Jersey’s public accommodations law directly and immediately affects associational rights, in this case associational rights that enjoy First Amendment protection.”¹³² This passage is as conclusory as it is cryptic. It seems to be saying that public-accommodation laws are not about conduct at all but about “associational rights” at their core. This is not, however, an understanding of public-accommodation laws that the Court had expressed before. Public-accommodation laws certainly have speech effects, just as the statute banning the burning of draft cards had. But the Court recasts the statutes as “directly and immediately” burdening expression, ruling that “O’Brien is inapplicable.”¹³³ Instead, the Court makes clear that strict scrutiny is the standard for public-accommodation statutes. To override freedom of association, regulations must be “adopted to serve compelling state interests, unrelated to the suppression of ideas, that cannot be achieved through means significantly less restrictive of associational freedoms.”¹³⁴

The Court then performs a most perfunctory inquiry into whether the state law could be justified by a compelling interest. “The state interests embodied in New Jersey’s public accommodations law do not justify such a severe intrusion on the Boy Scouts’ rights to freedom of expressive association.”¹³⁵ The Court recalled that the Roberts Court had decided differently on that score but characterized the holding there as recognizing “that States have a compelling interest in eliminating discrimination against women in public accommodations.”¹³⁶ Eliminating discrimination against gay scoutmasters, presumably, was less compelling. What’s more, the Court hinted that it did not consider as binding the Roberts statement that eliminating discrimination against women was a compelling interest. After quoting the Roberts passage, the Dale Court said, “[B]ut … we went on to conclude that the enforcement of [the statute] would not materially interfere with the ideas that the organization sought to express.”¹³⁷ Whereas Roberts seemed to use alternative bases (both belts and suspenders) to uphold the statute—it did not interfere with the interests of the Jaycees, and it was justified by a compelling interest—the Dale Court seemed eager to use the former justification to turn the second into dicta. And in a situation such as in Dale, where the Court ruled that the statute did burden the interests of the group, making the second of Roberts’s bases into dicta meant that the statute would fail in a conflict with the First Amendment.

For organizations wishing to claim an exemption from public-accommodation laws, Dale remains the central case in this line of precedents. From Dale, we know that organizations will normally receive a great deal of judicial deference with regard to the nature of the organization’s expressive interests and whether the obligations of the public-accommodation statute burden those interests. An organization can even claim a First Amendment interest in excluding someone on the basis of their status, especially if the person’s status is obvious and third-party observers could mistake the presence of the person as a message of acceptance. Finally, when the government asserts interests to balance against the organization’s expression, the government must satisfy strict scrutiny. And the need to fight against discrimination on the basis of sexual orientation is not compelling. Indeed, depending on how one reads 303 Creative, the Court may no longer consider the fight against any kind of discrimination as compelling.

The Court’s most recent ruling on the ability of claimants to use the First Amendment to avoid the obligations of anti-discrimination statutes came in the 2022 Term. Read carefully, 303 Creative v. Elenis¹³⁸ offers several doctrinal and analytical updates that are important to consider when evaluating the possible claims of universities asserting a First Amendment exception to a color-blind Title VI.

A Colorado woman named Lorie Smith created a website design company, 303 Creative, and described the business as including the creation and hosting of wedding websites.¹³⁹ She feared that the Colorado public-accommodations statute, which bars discrimination on the basis of sexual orientation among other things,¹⁴⁰ would require her to create websites for same-sex weddings, which she opposes because of her religious beliefs.¹⁴¹ She filed a declaratory judgment action before her business had any customers or produced any wedding websites and thus before the state could bring any action against her for discriminating on the basis of sexual orientation.¹⁴² She wanted a court to acknowledge that her company had a First Amendment right to refuse to build wedding websites for same-sex couples.¹⁴³

In an opinion by Justice Gorsuch, the Court ruled in favor of Smith and 303 Creative, saying that the operation of the statute would force Smith to express a message with which she disagreed.¹⁴⁴ Viewed narrowly, that particular holding was unsurprising as it flowed easily from the Court’s decision in Hurley as well as other precedents articulating the principle that the First Amendment protects people from government coercion to speak.¹⁴⁵ Where the Court expanded the doctrine past the previous holdings of Roberts, Hurley, and Dale was in its application to a for-profit company that voluntarily entered a marketplace that imposed on all merchants certain obligations of nondiscrimination arising from long-standing public accommodations laws. Rather than seeing coercion, the Court might have seen Smith as having made a choice to engage in the speech as a condition of entering into a marketplace in which nondiscrimination was an organizing principle. But the Court looked narrowly at the specific expression that Smith wanted to avoid making and asked whether the government was forcing her to engage in that specific expression.¹⁴⁶ Also, the Court implicitly expanded the doctrine in allowing the company, 303 Creative, to claim the expressive interests of its investor-owner, Smith.¹⁴⁷ That, too, had never been seen before in the precedents, but the Court did not seem troubled by the projection of Smith’s values onto a limited liability company she formed to insulate her from potential company debts and obligations.

Also worth noting is the profound level of deference the Court showed to Smith’s own description of her expressive interests and the putative burden on those interests imposed by the public-accommodation statute. Consistent with prior cases, the Court took as given the claimant’s assertions that her views on same-sex marriage were inconsistent with the state’s. But a website is not a parade, and the Court might have pressed her to show that the views implicit in a wedding website would be attributed to her rather than the couple featured. Or the Court might have asked her to show that curing any misattribution would be burdensome or unworkable. But the Court asked for neither.

Two aspects of the 303 Creative case are particularly relevant to the question of whether universities can bring a successful First Amendment claim to use race in admissions. The first is a series of questions posed in the oral argument as to whether the First Amendment right to discriminate would extend to race. Even though much of the oral argument revolved around that concern, the opinion does not offer any explicit answer. The second is the opinion’s exposition of the proper level of scrutiny to be applied in these cases. The Court returned to the hedging of Roberts and Hurley, but in a way that leaves the impression that “strict” would in fact understate the level of scrutiny applied. If so, while the opinion may not have offered an explicit answer to whether the First Amendment protects those who want to discriminate on the basis of race, the answer might be implicit and nevertheless fairly clear.

1. The 303 Creative Oral Argument and the Race Question.—Evident in the questioning by several justices was the concern that 303 Creative’s arguments would offer cover to businesses claiming a First Amendment exemption to statutes punishing discrimination on the basis of race. Several justices repeatedly pressed 303 Creative’s lawyer for a limiting principle to the slippery-slope problem lurking in the case.¹⁴⁸ If a business can claim a First Amendment interest in discriminating on the basis of sexual orientation, could they also claim an interest in discriminating on the basis of race?

Justice Ketanji Brown Jackson posed a hypothetical about a photography business offering pictures with Santa Claus in a mall around Christmastime. The business wants to create pictures representing nostalgia about a certain era, and for that reason wants to prohibit families of color from sitting with Santa.¹⁴⁹ “I mean, the objection, just like your client’s objection, is to expressions that violate their own views of what is being depicted, and so their view of what is being depicted is that a scene with Santa and a child on the lap and all of that in sepia tone, trying to harken back to the good old days, should only have white children in it. That’s their firm belief.”¹⁵⁰

303 Creative’s lawyer tried to avoid answering, but finally admitted that the Court had “protected vile, awful, reprehensible, violent speech in the past, and it has never compelled speech.”¹⁵¹ If the photograph was considered speech, “the Court would still have to protect the speech.”¹⁵² That is, when forced to answer, the lawyer was unable to draw a line between sexual orientation and race—a company engaging in speech would be able to claim an exemption from a statute banning discrimination on the basis of race if that statute coerced the claimant to express themselves in a way contrary to their views on race. To paraphrase, the Court “would … have to protect the speech” even if it was “vile,” “awful,” or “reprehensible.” A lawyer’s representations during argument do not amount to legal precedent, but these exchanges do make clear that the Court was well aware of the potential implications of a holding in favor of 303 Creative.

It is worth emphasizing what was not said, by advocate or justice, in answer to the slippery-slope questions. One easily accessible distinction between 303 Creative (and for that matter Dale and Hurley) on the one hand and the Santa hypothetical on the other is that the state has a compelling interest in fighting discrimination on the basis of race but not on the basis of sexual orientation. There are indeed cases that could have been cited to stand for the proposition that there is a compelling interest in fighting discrimination on the basis of race,¹⁵³ and one might have reminded the Court that it had recognized in Roberts a compelling interest to fight discrimination on the basis of sex.¹⁵⁴ These could have been compared to the Court’s holdings in both Hurley and Dale that held, implicitly or not, that the state did not have a compelling interest in fighting discrimination on the basis of sexual orientation.¹⁵⁵ None of the players, whether advocate or justice, seemed eager to rely on that distinction. The Court’s liberals may not have wanted to imply that discrimination against LGBTQ people is less problematic than discrimination against people of color. The conservatives may not have felt the need to make the distinction, believing that a win for 303 Creative was assured whether the distinction was made or not.

All of this is to say that the question of race—the question on which First Amendment claims of universities will turn—was front and center in the 303 Creative case. In trying to anticipate what the Court would do about it in a case brought by a university claiming an exemption from Title VI, one cannot say that the Court ignored the issue in 303 Creative. Instead, it was a central aspect of the Court’s oral argument in the case. It is indeed true that the Court did not resolve it explicitly in its opinion. But what the Court did say is revealing.

2. The Level-of-Scrutiny Discussion and the Race Question.—Justice Gorsuch’s opinion places Smith’s speech interests at the center of its analysis. She was engaged in “pure speech,”¹⁵⁶ and the Colorado public accommodations statute would operate “to compel speech Ms. Smith does not wish to provide.”¹⁵⁷ As a matter of typical analytical structure, the Court should have then turned to the question of whether the burden on Smith’s expressive interests was justified by a compelling interest, narrowly tailored. After all, the Court has reminded us numerous times that a content-based restriction on speech can survive only if the state can show the restriction is carefully tailored to further a compelling governmental interest.¹⁵⁸

But Justice Gorsuch did not apply that framework. In fact, he even conceded that public-accommodations laws play “a vital role” in realizing the civil rights of all Americans, and he admitted that “this Court has recognized that governments in this country have a ‘compelling interest’ in eliminating discrimination in places of public accommodation.”¹⁵⁹ Interestingly, in this section of the opinion, Justice Gorsuch cites not only the cases you might expect, including Roberts and Hurley, but also civil rights–era cases about race. In noting that the Court had “recognized … that public accommodations laws ‘vindicate the deprivation of personal dignity that surely accompanies denials of equal access to public establishments,’” the opinion quotes Heart of Atlanta Motel v. United States,¹⁶⁰ one of the leading cases upholding the power of Congress to ban racial discrimination in the marketplace. Heart of Atlanta is not a First Amendment case, but it certainly stands for the proposition that the Court cites it for—that public-accommodation statutes are vital in the marketplace. Then, Justice Gorsuch offers follow-on cites to Katzenbach v. McClung,¹⁶¹ the famous case upholding the ability of Congress to ban racial segregation in restaurants, and to Newman v. Piggie Park Enterprises, Inc.,¹⁶² a case that described as “frivolous” a restaurant’s claim that the ban on racial segregation violated its free exercise of religion.

This is the moment in the opinion where, if the Court had wanted to uphold the Colorado statute or had wanted to distinguish racial discrimination from discrimination on the basis of sexual orientation, it could have done so easily. The Court could have said, yes, Smith’s speech was burdened but the state’s public-accommodation law is supported by a compelling interest. Or, it could have said, yes, there are precedents standing for the principle that the state has a compelling interest in ending racial discrimination but there are none recognizing a compelling interest in ending discrimination on the basis of sexual orientation. Either would have worked analytically. But the first would have upheld the Colorado statute, which the conservative majority did not want to do. The second would have upheld the statute but limited the holding to companies claiming an expressive right to discriminate on the basis of sexual orientation.

But the Court did neither. Instead, the opinion seems to change the entire analytical framework of these cases. It does not do so explicitly, but a careful reading of the opinion suggests that the Court does not want to bother itself any longer with the strict-scrutiny step in these First Amendment cases.

After a discussion of the importance of public-accommodation statutes, even recognizing the “vital role” they play in furthering the “compelling interest” in eliminating discrimination, the Court says that these interests do not apply when “deployed to compel speech.”¹⁶³ The Court explains “that no public accommodations law is immune from the demands of the Constitution,” and “when a state public accommodations law and the Constitution collide, there can be no question which must prevail.”¹⁶⁴ That is, when a statute burdens expression, it must fall.

This is either careless, mushy analysis or disingenuous rhetoric meant to obscure what the Court is doing. Of course public-accommodation statutes, like all statutes, are subject to the “demands of the Constitution.” What the Constitution “demands”—at least until 303 Creative—is that statues burdening expressive interests be subject to strict scrutiny. That is, as the Court has explained countless times, the Constitution demands that the state show a compelling interest when it wishes to burden expression by way of a content-based regulation. Even assuming Colorado’s public-accommodation statute is best understood as a content-based regulation of speech—which is dubious at best¹⁶⁵—what the Constitution demands is strict scrutiny. The statute does not automatically fall; it can survive if the state can show it is carefully crafted to further a compelling interest. Under that framework, the Colorado statute should have prevailed, because—as the Court had just admitted—eliminating discrimination is a compelling interest.¹⁶⁶ The “demands of the Constitution” should have been met.

But the Court says the statute must fall because it collided with the Constitution. That must mean that the Court believes either that the last stage of the traditional analysis—does the statute further a compelling interest?—is no longer a part of the inquiry or that a compelling interest never exists. But the Court had just said anti-discrimination statutes are based on compelling interests. The only way to understand the Court’s opinion on its own terms, then, is to read it as holding that expressive claims such as 303 Creative’s are no longer subject to a strict-scrutiny analysis. The expressive interest will always win regardless of the strength of the governmental interest placed in the balance. “When a state public accommodations law and the Constitution”—that is, the constitutional claims of businesses whose expressive interests are burdened by anti-discrimination laws—“collide, there can be no question which must prevail.”¹⁶⁷

This reading of the 303 Creative opinion is bolstered by the fact that the Court never purports to apply strict scrutiny. Typically, any constitutional-rights case can be reduced to three steps: whether there is a constitutionally protected interest at issue; whether the interest has been burdened by official action; and whether the government can show a compelling interest, narrowly tailored, so as to justify the burden on the interest. The Court has used that framework for decades.¹⁶⁸ But in 303 Creative, the last step is omitted.

The Court does suggest that some applications of public-accommodations laws will nevertheless survive. “Colorado and other States are generally free to apply their public accommodations laws, including their provisions protecting gay persons, to a vast array of businesses.”¹⁶⁹ The Court recognized that states may “protect gay persons, just as [they] can protect other classes of individuals, in acquiring whatever products and services they choose on the same terms and conditions as are offered to other members of the public.”¹⁷⁰ But the Court implied a crucial limit: it noted that “there are no doubt innumerable goods and services that no one could argue implicate the First Amendment.”¹⁷¹ Parsed carefully, what the opinion seems to be saying is that while public-accommodation statutes can be applied to a “vast array of businesses,” if the First Amendment is “implicate[d],” all bets are off. In those cases, we return to the Court’s reminder that “[w]hen a state public accommodations law and the Constitution collide, there can be no question which must prevail.”¹⁷²

3. Implications.—One has to read 303 Creative carefully to discern its significance. On the surface, it is a mere application of the Hurley and Dale precedents: a company has expressive interests burdened by a statute that coerces it to say what it would like to avoid saying, and the Court provides it an exemption from that statute. But the opinion’s innovations are real and merit acknowledgment. Some, such as the fact that 303 Creative extends the protections offered to nonprofit associations in Hurley and Dale to for-profit companies voluntarily choosing to enter the marketplace, and its unquestioned projection of an individual’s beliefs onto a business form created to insulate that individual from the business, are not the focus of this article.

What is important for this article’s discussion is the opinion’s implicit judgment about race-focused First Amendment claims. The Court was squarely confronted in the oral argument with the problem of whether the First Amendment would provide an exemption for companies whose expression about race was burdened by a public-accommodation statute. In the argument itself, neither the 303 Creative lawyer nor any Justice offered the possible distinction that race-based First Amendment claims should be evaluated differently from the expression about sexual orientation that had been at issue in Hurley and Dale. And the opinion itself fails to make any attempt to avoid the all-too-obvious slippery-slope concern that race-based expression will win exemption from public accommodation laws. Instead, the opinion doubles down. Even “vital,”¹⁷³ compelling interests will now lose to First Amendment expression, even if the expression is—in the words of the 303 Creative lawyer—“vile”¹⁷⁴ or—in the words of Justice Gorsuch—“deeply misguided.”¹⁷⁵ Either strict scrutiny is a thing of the past or the level of scrutiny is something more than strict. In either reading, a compelling interest will be insufficient to protect a public-accommodation statute when it is applied to burden expression. And because, as Justice Gorsuch reminds us, the First Amendment protects expression and expressive association “equally,”¹⁷⁶ the same must be true for those associations and businesses that can show that a public-accommodation statute burdens their associational interests.

The opinion does not bode well, therefore, for those who worry that companies and associations could now successfully claim First Amendment exemptions from statutes barring race-based discrimination. The road ahead is now well mapped.

This road map is not available just to racist proprietors who want to exclude people of color from their businesses. It is also available now to universities and colleges that want to continue to use race in admissions. If the Court is ready to grant an exemption from public accommodation laws to the white mall Santa, it stands to reason that a claim brought by Howard, Berea, or Georgetown to be exempted from Title VI’s ban on “discrimination” should be successful as well. The next part of this article turns to a more precise description of a university’s First Amendment claim to continue to use race in admissions.

This part offers a fuller explanation of the First Amendment interests that a university could assert as a basis for an exemption from Title VI’s new color-blindness mandate. Institutions could assert both speech and associational interests, but associational interests will be the stronger. And those will be most persuasive coming from colleges or universities with long-standing and mission-driven associational interests that support the use of race in defining their student communities.

This part will also describe the ways in which a university’s First Amendment claim to be exempted from Title VI’s new color-blindness requirement would necessitate an expansion of the Court’s precedents. In some ways, the expansion is straightforward and unremarkable, fitting within the normal range of legal evolution typical to common-law reasoning. In one particular way, however, the expansion will be material and controversial. A university’s claim will require the existing precedent, which has allowed companies and groups to discriminate on the basis of sexual orientation, to be expanded to include an allowance to “discriminate” (under the Title VI definition) on the basis of race.

The cases of Roberts, Hurley, Dale, and 303 Creative are the primary tools in the precedential toolbox for a university wanting to assert a First Amendment right to be exempted from the new color-blindness requirement of Title VI. In terms of the framework of a claim, a university would mimic the stance of the Jaycees, the parade organizers, and the Boy Scouts. That is, a statute outlawing discrimination of a certain kind thereby forces an association to make decisions as to its membership that the organization would not otherwise make. The organization asserts that the statute thus burdens its associational freedoms protected by the First Amendment. For the statute to be upheld, the government would have to show a compelling interest to apply the statute to the university. Or, if 303 Creative means that even compelling interests will not outweigh claimants’ First Amendment rights, then the universities would win even if the government showed that Title VI furthered compelling interests.

The details of a university’s claim would of course be different from the precedents in several respects, and the question would be whether they would be similar enough to these prior cases to benefit from their overall precedential thrust.

What might a college or university assert by way of First Amendment interests?

Institutions of higher learning are associations of students, faculty, researchers, and others organized around the objectives of education, learning, and the furtherance of knowledge through research and inquiry. Even at their most basic, colleges and universities are organized around those goals. Most institutions go further, grounding their educational objectives within other core values. Many religious institutions, for example, situate their educational goals as applications of deep-seated religious values. Some schools explain their missions in moral but nonsectarian terms. Some explicitly claim a goal of serving society broadly; some assert goals that sound in an ideological register.¹⁷⁷ Some proclaim missions that contain implicit and often explicit critiques of the status quo and the need for change and progress.¹⁷⁸

While it may not be true for all institutions, a good number of private colleges and universities will thus be able to claim the kind of expressiveness inherent in their association that will merit protection under the Roberts understanding of expressive associations. The institutions have a point of view that they express by way of policies, procedures, curricula, research projects, areas of inquiry, public statements, events, hiring, promotion, and institutional and campus culture.¹⁷⁹

And—most centrally to this analysis—institutions may also choose to express their core values, commitments, and missions by way of choices as to whom to invite into their community.¹⁸⁰ Not all universities are selective in whom they accept into their student bodies, and even among selective institutions only a subset would choose to use race as one relevant criterion. But there is little doubt that colleges and universities are intentionally compositional in ways that merit protection as expressive associations.¹⁸¹ If anything, their decisions as to whom to accept into their student communities (not to mention as faculty, researchers, or administrators) are much more considered, sophisticated, and calibrated to fit the goals of the institution than were the decisions of the parade organizers in Hurley and the Boy Scouts in Dale.¹⁸² The very issue of whether race can be one criterion in admissions decisions depends on the fact that many universities are conscientious, careful, and intentional in deciding whom to admit as students.

The best illustration of this is to imagine a situation outside the racial context in which the government required an institution to admit (or deny) a specific student whom the college had decided to decline (or accept). Current controversies present an apt hypothetical.¹⁸³ If a college decided to deny admission to a student who, in her college essay, had called for the genocide of a particular group of people, a government requirement that the student be admitted nevertheless would clearly be a burden on the college’s associational rights. Or consider an institution that holds among its core values the amelioration of social injustice. If that institution refused admission to an avowed white supremacist, there is no doubt that the school’s right to expressive association would be burdened if the government forced the school to admit that student. To channel Dale, “the presence of [a white supremacist as a student] would … interfere with the [institution’s] choice not to propound a point of view contrary to its beliefs.”¹⁸⁴

This would also be true if exclusion or inclusion was on the basis of a non-expressive reason. Imagine a school that decided to cease its use of standardized tests as a criterion in the admission decision, and then the school is sued by an applicant who presents evidence indicating that she would have been admitted but for the new policy. Or imagine a school that decided to end its varsity fencing program and thereafter stopped giving preference in admissions decisions to students demonstrating skill with the epee.¹⁸⁵ Those decisions reflect the beliefs, values, and pedagogical judgments of the institution. If a government entity ruled that the institutions had no ability to exclude good test takers (or bad test takers) or swordsmen (or defensive linemen), the ruling would certainly burden the university’s interests in deciding the makeup of its own student body.

Another application would be a direct analogy to Dale itself. Imagine a private institution that professed a campus identity that proclaimed adherence to traditional mores around sexual orientation. Perhaps the institution dedicated itself to encouraging students to be “morally straight.” And assume further that the school had made clear that it considered same-sex sexual behavior contrary to its ethos.¹⁸⁶ (This would be stronger, then, than the claims of the Boy Scouts, which made explicit its anti-LGBTQ+ stance only after Dale was terminated.) And assume also that the school considers the student’s self-identification with respect to sexual orientation as a part of its admissions calculation. In such a situation, a public-accommodation statute operating to tell the school not to consider sexual orientation in those decisions would burden the school’s expressive association.

Is there anything different about an institutional decision to use race as one of the considerations in the admissions process?

Schools could have myriad reasons, based on their core commitments, to use race in admissions. Some schools might point to the diversity rationale long cited in the constitutional cases, such as Bakke and Grutter, up until the Harvard and UNC rulings.¹⁸⁷ A heterogeneous community of learners, a school could believe, is conducive to learning both in and outside the classroom and enhances the overall campus community. Schools could believe that racially diverse student bodies are instrumental in preparing their students for success in a pluralistic society. Many schools in fact continue to say this in various ways,¹⁸⁸ and scores of institutions made admissions decisions for these reasons for decades.

Notice that within the equal protection frame, the doctrinal battleground for the last fifty years, the question was whether the pursuit of diversity was a compelling interest sufficient to survive under strict scrutiny. Once the battleground changes to the First Amendment context, however, the question is instead whether a private institution could have a belief in diversity that serves as the basis for using race as a consideration in constructing the membership of its expressive association. This is a much easier test to meet, especially if given the level of deference to the claimants’ own descriptions of beliefs the Court afforded in Hurley, Dale, and 303 Creative. And the belief will have weight in the First Amendment context even after the pronouncement in Harvard that diversity is no longer a compelling interest in the equal-protection context. The requirement of presenting a compelling interest is no longer for universities. Instead, it is the government that needs to show a compelling interest to outweigh a university’s constitutional interest in the use of race in defining its membership.

What’s more, the desire to foster diversity is not the only rationale that schools could now assert. Once the legal context moves into the First Amendment arena, universities have the option to shed or downplay diversity as their core rationale.¹⁸⁹ A university is free to assert a belief that the use of race achieves additional objectives consistent with its values, goals, and identity. A school could argue, for example, that considering race in admissions is inextricably bound up with repairing and redressing past harms. Historically Black colleges and universities, for example, might argue that, in building its community, considering race in admissions is intrinsic to their core, historical missions to support and enhance educational opportunities for Black students.¹⁹⁰ Similarly, a religiously affiliated university might argue that faith-based moral imperatives to lift up the disadvantaged and right past wrongs compel it to consider race in admissions.¹⁹¹ The basis for these beliefs might be religious in nature, but that would not make it impossible to couch it in terms of speech/expression, as did the website designer in 303 Creative.¹⁹² Other schools might assert that the use of race is necessary for them to fulfill their mission to serve social justice. Berea College, for example, was founded by an abolitionist in the nineteenth century and continues to assert a core value of building a racially inclusive student body to serve its conception of social justice.¹⁹³ Other universities might assert the need to use race in constructing its student community in part to ameliorate their own past sins.¹⁹⁴

If the Court applies its precedents honestly, these beliefs should easily win recognition as the kind of interests protected by the First Amendment. Colleges and universities are driven by principles and values much more than the Southie parade organizers were. Their beliefs are more coherent and long-standing than those of the Boy Scouts. They are more intentional and discerning in whom they let into their association than the Jaycees, the parade, or the Boy Scouts.¹⁹⁵ The operation of a color-blind Title VI threatens to change campus communities more than the statutes in Minnesota, Massachusetts, and New Jersey threatened to change the Jaycees (who already accepted women), the parade (which generally accepted all comers until GLIB), and the Boy Scouts (which did not have a stated policy before the lawsuit). And if the Court followed its Dale precedent and deferred to the colleges’ own assertion that a color-blind Title VI burdened its ability to construct its expressive association, then there is little question that the first question in any constitutional case—is there a constitutional interest at stake?—would be answered in favor of the institutions.

One possible distinction is worth mentioning but easy to set aside. In Hurley the parade organizers sought to exclude a specific group; in Dale the BSA sought to exclude a specific individual. In a potential associational claim by a college or university, the institution may not be able to identify individual applicants that it wants to exclude. And even if an institution can identify applicants who would be excluded but for the color-blindness requirement, the individuals so identified would not likely personally embody a point of view that the university finds reprehensible.

But that is a meaningless distinction. In Hurley, the state was purporting to make it unlawful for the parade organizers to establish sexual orientation as a criterion for making decisions about participation in the parade. If the Court had ruled in favor of GLIB, the organizers would not have been able to use sexual orientation as a relevant criterion in deciding whether to allow still unspecified groups to march in the future. In Dale, the state was purporting to make it unlawful for the BSA to use sexual orientation as a relevant criterion in determining who could serve as scoutmasters. If the Court had ruled in favor of Dale, the unavailability of sexual orientation as a criterion would have made a difference in whether many other, unspecified individuals could serve as scoutmasters.

Moreover, in an actual case, these differences between Hurley and Dale on the one hand and university cases on the other would be even less. If the constitutional case begins when a rejected applicant sues a school saying they would have been admitted but for the school’s use of race, then the analogy to Dale is quite close. If, on the other hand, a constitutional case begins with a university bringing a declaratory action because it fears the color-blindness rule will burden its expressive interests, then the analogy to the procedural posture of 303 Creative is apt.

In the end, the interest question would thus be easy for colleges and universities to satisfy. That would not be sufficient to win their case, of course. Their victory would depend on extending the precedents to a context beyond previous cases: to allowing universities to assert a First Amendment interest to “discriminate” on the basis of race. That question is the focus of the next section.

For the reasons set out in the previous section, the potential claims of some colleges and universities will flow well from the Court’s existing precedent. Universities are expressive associations whose ability to construct their own communities are burdened by a statutory obligation to ignore a criterion for membership that the association wants to consider. The most salient difference, however, is that the universities would be wanting to claim a First Amendment interest to consider race, whereas the prior cases focused on sexual orientation. This section will analyze this extension.

1. Interest and Burden.—Looking only at the First Amendment precedents, the extension of First Amendment claims to cover race does not appear to be a big deviation from the current doctrine. Indeed, even if the Court considers race-based admissions policies to be “misguided” or even “vile,”¹⁹⁶ its existing precedent has articulated the principle that the Court will defer to constitutional claimants’ own views of whether an anti-discrimination statute burdens its expression and association. As the Court said in 303 Creative, the First Amendment protects even expression and association based on “deeply misguided” notions.¹⁹⁷

University claimants, for their part, would not of course think that the use of race in constructing their campus communities is deeply misguided. They would be basing their use of race on beneficial and remedial goals, as many universities have done for the decades since Bakke.¹⁹⁸ But universities would have to expect that, given the level of opprobrium the Court showed in the Harvard case for universities’ use of race, the Court will likely consider any use of race in the admissions process misguided. Even so, if we take the First Amendment precedents as given, the level of deference the Court affords to expressive claimants suggests that with regard to the first stage of the constitutional inquiry—does the claimant have a constitutional interest in resisting the obligations of the statute?—the answer has to be yes even with regard to race. A university claimant will be able to rest squarely on existing doctrine in asserting that it has a constitutional interest in using race in determining the membership of its community and that the application of a color-blindness rule burdens that interest.¹⁹⁹

At this stage of the analysis, the Court would likely consider one precedent relevant. The 1976 case of Runyon v. McCrary²⁰⁰ held that 42 U.S.C. § 1981, which ensures that all “persons … shall have the same right … to make and enforce contracts … as enjoyed by white citizens,”²⁰¹ prohibited private schools from excluding qualified children solely because they were Black.²⁰² The focus of the case was whether § 1981 should be interpreted to prohibit racial discrimination in the making and enforcement of private contracts and whether that interpretation was broad enough to support a suit by Black families who had been excluded from racially exclusive schools.²⁰³ But after holding that the statute should be so interpreted, the Court paused and spent two paragraphs addressing “[t]he question [of] … whether § 1981, as applied, violates constitutionally protected rights of free association.”²⁰⁴ The Court assumed that “parents have a First Amendment right to send their children to educational institutions that promote the belief that racial segregation is desirable” but said that the associational right did not extend so far as to allow schools to exclude “racial minorities,” because “the Constitution … places no value on discrimination.”²⁰⁵ The Court explained that while “[i]nvidious private discrimination may be characterized as a form of exercising freedom of association protected by the First Amendment … it has never been accorded affirmative constitutional protections.”²⁰⁶ In any event, the Court said, “there is no showing that discontinuance of [the] discriminatory admission practices would inhibit in any way the teaching in these schools of any ideas or dogma.”²⁰⁷

The doctrinal question would be whether Runyon changes the analysis springing from Hurley, Dale, and 303 Creative of whether colleges and universities may now assert associational claims to support their use of race as one factor in admissions decisions. It should not, for several reasons. One distinction is that modern-day colleges and universities are not using race as the solitary basis for inclusion or exclusion; they want to use race as one criterion among many. The kind of “discrimination”—using the Title VI language—the universities want to engage in is not the “invidious” discrimination that was at issue in Runyon. Then, schools were excluding children solely on the basis of their race. Now, universities want to use race as one criterion to address any number of social harms in ways that are core to their missions and principles.

Another distinction is that the foundational assumptions of the Runyon passage quoted above have been changed by subsequent First Amendment cases. The assertion that “private discrimination … has never been accorded affirmative constitutional protections” is now false. Roberts, decided eight years after Runyon, stated that “[f]reedom of association … plainly presupposes a freedom not to associate,”²⁰⁸ and Hurley, Dale, and 303 Creative each provided “affirmative constitutional protections” for expressive acts of discrimination. Moreover, the notion that “the Constitution … places no value on discrimination”²⁰⁹ is now in conflict with statements in both Dale and 303 Creative that the Court’s deference to associational claims does not turn on its substantive views as to whether the claimant’s beliefs are “misguided” or not.²¹⁰ The Court now professes agnosticism about the validity and reasonableness of the substance of constitutional claimants’ expressive interests.

Another reason to believe that Runyon would not control is that its treatment of the associational claim is much closer to Roberts with regard to deference than to the later cases of Hurley, Dale, and 303 Creative. The Runyon Court ended its short treatment of the association claim with the statement, “[T]here is no showing that discontinuance of [the] discriminatory admission practices would inhibit in any way the teaching in these schools of any ideas or dogma.”²¹¹ In other words, the inclusion of Black students in the schools would not burden First Amendment interests. This aligns with the Roberts opinion’s treatment of the Jaycees’ associational claims: “the Jaycees has failed to demonstrate that the Act imposes any serious burdens on the male members’ freedom of expressive association.”²¹² A careful reader will recall that the Dale Court used this statement from Roberts to turn into dicta the Roberts Court statement that the state has a compelling interest to eliminate discrimination against women.²¹³ It is thus possible to read Runyon’s entire association analysis as dicta because the Court determined that there was no burden on an expressive interest.

The best way to read Runyon, then, is that it would allow but not require a Court to distinguish between the claims at issue in Hurley, Dale, and 303 Creative and the claims of universities that they should be permitted to use race in admissions. If the Court wanted to draw a distinction between race-based expressive interests and those at issue in Hurley, Dale, and 303 Creative, it could use Runyon as a basis for making the distinction. But as a predictive matter, if the Court were predisposed to making this distinction, it could have done so in its 303 Creative opinion.²¹⁴ It did not, and no justice offered it as a possible distinction during the oral argument.²¹⁵

It is important to note that even if the Court declined to follow Runyon in a university case, it would not necessarily mean that it was overruled. The core holding of Runyon, that § 1981 governs private contracts, would not be weakened. The change would be in the narrower sense that a school that has an expressive or associational interest burdened by § 1981 might be able to claim an exemption from its application to them. And depending on how one reads the First Amendment cases, that doctrinal change may have already happened.

2. Compelling Interest.—Once there is a cognizable expressive or associational interest and once it is burdened, the doctrinal question then turns to whether the government can satisfy the proper level of scrutiny.²¹⁶Dale indicates that the proper level of scrutiny is strict, which means that the government could overcome the assertion of expressive interests only with a compelling interest, narrowly tailored.²¹⁷ This is where universities would have to push the Court beyond where it has gone before. The Court has said several times that the government has a compelling interest in fighting discrimination on the basis of race.

Most relevant is the 1983 ruling in Bob Jones University v. United States.²¹⁸ The Court considered whether nonprofit private schools that prescribe and enforce racially discriminatory admissions standards on the basis of religious doctrine qualify as tax-exempt organizations under § 501(c)(3) of the Internal Revenue Code.²¹⁹ Until 1970, the IRS granted tax-exempt status to private schools without regard to their racial admissions policies, but in 1970 the IRS changed its policy and announced that it would no longer accord tax-exempt status to private schools that discriminate in admissions on the basis of race.²²⁰ A private Christian university in South Carolina and a private children’s school in North Carolina that both discriminated in admissions on the basis of race challenged the IRS’s withdrawal of their tax-exempt status.²²¹ The Court rejected the challenge and upheld the IRS’s change in policy.²²²

The focus of the opinion was on federal tax law about charitable designations, and the Court held that the “entitlement to tax exemption depends on meeting certain common-law standards of charity—namely, that an institution seeking tax-exempt status must serve a public purpose and not be contrary to established public policy.”²²³ The Court observed that “there can no longer be any doubt that racial discrimination in education violates deeply and widely accepted views of elementary justice.”²²⁴ The Court used Congressional intent elsewhere as informative, saying that “Congress, in Titles IV and VI of the Civil Rights Act of 1964, … clearly expressed its agreement that racial discrimination in education violates a fundamental public policy.”²²⁵ The Court then concluded,

The Bob Jones Court then turned to an argument raised by the religious schools that even if the tax exemption was invalid as to nonsectarian schools, it could not be applied to them because they “engage[d] in racial discrimination on the basis of sincerely held religious beliefs.”²²⁷ Taking away their tax exemption would therefore violate the Free Exercise clause. The Court rejected the contention, explaining that “[t]he governmental interest at stake here is compelling. … [T]he Government has a fundamental, overriding interest in eradicating racial discrimination in education—discrimination that prevailed, with official approval, for the first 165 years of this Nation’s constitutional history.”²²⁸ The Court said that the governmental interest outweighed any Free Exercise claim pressed by the religious schools. “That governmental interest substantially outweighs whatever burden denial of tax benefits places on petitioners’ exercise of their religious beliefs. The interests asserted by petitioners cannot be accommodated with that compelling governmental interest, … and no ‘less restrictive means’ … are available to achieve the governmental interest.”²²⁹

For a Court considering a modern-day expression and association claim brought by a college or university seeking to be exempted from the new color-blind requirement of Title VI, the question would be whether, to hold for the school, Bob Jones would have to be overturned. The answer is no.

The most powerful distinction is the nature of the putative compelling interest. In Bob Jones, the schools were engaged in race-based segregation, the kind of “racial discrimination in education … that prevailed, with official approval, for the first 165 years of this Nation’s constitutional history.”²³⁰ That kind of anti-Black discrimination, meant to disempower and subordinate African Americans, is not the kind of policy that universities would be suing to protect. The kind of “discrimination” at issue in a modern university challenge to Title VI is different; it is not designed to suppress any group or marginalized minority; rather, it seeks to ameliorate conditions for African Americans and other people of color who have long been the victims of private and state-sanctioned discrimination. The government’s interest in eradicating segregation is much greater than its interest in telling a university it cannot use race as one criterion in a holistic review of admissions applications.

Indeed, we know that the kind of “racial discrimination” at issue in Bob Jones is not the kind of “discrimination” at issue in a potential university challenge to Title VI. The Bob Jones Court used Title VI as a reference for the congressional belief that “racial discrimination in education violates a fundamental public policy.”²³¹ But at the time of Bob Jones, colleges and universities were indeed using race both in a remedial way and to pursue diversity. Bakke had been decided in 1978, five years earlier.²³² Title VI was not color-blind at the time, and universities were using race as one criterion in admissions decisions. That must mean, then, that race-based admissions policies of the kind universities today would be suing to continue were not contained within the Bob Jones Court’s definition of “racial discrimination in education [that] violates a fundamental public policy.”²³³ To use Bob Jones to support a ruling that the government has a compelling interest to enforce color blindness would be to overread what the Court said and meant.

Even so, this entire discussion assumes that strict scrutiny is the standard and that a compelling interest on the part of the government would be sufficient to outweigh a university’s valid expressive or associational interest in using race in admission decisions. But 303 Creative seems to have changed the framework of the analysis, as discussed above.²³⁴ The 303 Creative opinion seems to say that expression (and presumably association) claims will win even in the face of a governmental compelling interest on the other side of the balance. If that is true, even if a modern Court equated the use of race as one criterion in a holistic review with the “racial discrimination in education … that prevailed, with official approval, for the first 165 years of this Nation’s constitutional history,”²³⁵ the constitutional claims of the universities would prevail.

This possibility raises its own difficult questions, which are less about doctrine than about whether universities should risk creating law that could be used for nefarious purposes. That is the question explored in part IV.

Before turning to that question, however, one additional issue is worth noting quickly for completeness. One difference between the Title VI context and the state public-accommodation statues at issue in the line of cases culminating in 303 Creative is that Title VI does not ban discrimination outright but rather imposes a condition that federal-funding recipients not discriminate. The public-accommodation statutes are enforced by way of fines or other civil penalties.²³⁶ A violation of Title VI is enforced by threatening the cutoff of federal funds to the offending institution.²³⁷

One might argue that a fine creates a different level of burden on constitutional interests than a funding cutoff, and some cases exist that would support that thinking.²³⁸ In Rumsfeld v. Forum for Academic and Institutional Rights, Inc.,²³⁹ for example, a coalition of law schools (FAIR) raised speech and association claims against a federal law that conditioned funding for universities on the requirement that they treat military recruiters (who discriminated against LGBTQ students at the time) the same as nondiscriminatory recruiters.²⁴⁰ The Court upheld the law and the condition on funding.²⁴¹

But FAIR can be distinguished. There, the Court did not hold that a cessation of funding was not a burden on expressive or associational interests. The Court held instead that the schools had no cognizable expressive interests at all in play. “The Solomon Amendment neither limits what law schools may say nor requires them to say anything.”²⁴² And the associational claims faired no better: “A military recruiter’s mere presence on campus does not violate a law school’s right to associate, regardless of how repugnant the law school considers the recruiter’s message.”²⁴³ The Court decided FAIR as a step-one case—is there an interest?—rather than a step-two case—has the interest been burdened?

When the Court does reach the burden question, the Court has said that conditions on funding can indeed be an infringement of protected interests. In FAIR itself, the Court recognized that “other decisions … recognize a limit on Congress’ ability to place conditions on the receipt of funds.”²⁴⁴ And it quoted a case that said that “the government may not deny a benefit to a person on a basis that infringes his constitutionally protected … freedom of speech even if he has no entitlement to that benefit.”²⁴⁵

This aspect of a potential university claim against Title VI would thus turn on the doctrine of unconstitutional conditions.²⁴⁶ While sometimes murky around the edges,²⁴⁷ the doctrine “vindicates the Constitution’s enumerated rights by preventing the government from coercing people into giving them up.”²⁴⁸ The doctrine in its current form is fairly robust and has been applied in a number of situations in which the burden on a constitutional claimant was the loss of a potential benefit rather than a fine or other punishment.²⁴⁹

Given this framework, then, assuming a university or college will be able to show a cognizable expressive or associational interest, it is highly likely that it will be able to claim persuasively that those interests are burdened by a threat of a funding cutoff. Universities and colleges depend mightily on federal funding, and a funding condition in Title VI that they be color-blind in their admissions policies will operate as a mandate in fact.²⁵⁰ The reality of this mandate is evidenced by the fact that no institution has publicly volunteered to give up funding following the Harvard ruling.²⁵¹ Indeed, the option to give up funding does not seem to have occurred to any of them. A condition on federal funding, in this context, operates as a requirement, because the loss of federal funding would be more impactful than any fine.²⁵²

The choice of whether to bring a constitutional challenge to a federal statute is always a serious decision meriting the most careful consideration. Similarly, the intentional choice to violate a federal statute to challenge its application as a defense to a civil action against the violator is a decision that demands a sober and honest analysis of the potential gains and probable costs of doing so. In the current context, a decision by a college, university, or association thereof to challenge or intentionally disobey Title VI’s color-blindness requirement will only be made after a frank assessment of the probable costs of litigation—financial and otherwise—balanced against an honest assessment of the likelihood of victory, the benefits of that victory, and the availability of alternatives. Campus decision-makers will have to weigh these considerations in the light of a thorough evaluation of the state of the doctrine and how well it would accommodate the expressive and associational claims of colleges and universities. This intent of this article is to offer that thorough doctrinal evaluation.

Several aspects of the cost/benefit analysis of a potential First Amendment claim to save race-conscious admissions are worth mentioning because they seem different from the typical constitutional litigation. Some of these cut in favor of bringing a suit (or being willing to defend against one); some cut in favor of demurring.

The status quo for institutions of higher learning in the United States is that they can no longer use race in admissions. After Harvard, no college or university can consider race without risking loss of federal funding to the entire institution. A loss in a First Amendment lawsuit will not change that status quo. In that respect, universities would be no worse off after a loss than they are at present.

The potential benefits, on the other hand, are significant. For institutions that have used race as one criterion in admissions over the past decades, the Harvard case effected a major change. Those colleges and universities will be worse for this change, assuming they are unable to discover a non-race-based proxy that will allow them to maintain the diversity and pluralism of their campuses. If such a proxy is available, then the cost/benefit calculation would take that into account. But if such a proxy is not available, a First Amendment claim may be the only way to save race-conscious admissions in the United States. It will not save it for all schools, since some institutions are public and some private schools will be unable to point to sincere expressive or associational interests that would support a claim. But it could save it for many, and that is no small thing.

This consideration will not weigh in the balance for everyone, but it may for some. The current conservative super-majority on the Supreme Court has been asserting its ideological commitments in a number of areas, including on race and on the First Amendment. On race, the current Court has been pushing hard to ensure that government judgments on race are color-blind, weighing in squarely on the facial equality side of the age-old facial-versus-substantive-equality debate. On the First Amendment, the current Court has been working to expand the ability of businesses and other associations to use the First Amendment as a way to gain exemptions from otherwise applicable statutes.

The tension between these two lines of doctrine may not yet be obvious to members of the current Court. As this article has argued, the First Amendment doctrine as developed by the conservatives on the Court provides a persuasive and strong basis for schools to exempt themselves from the new, conservative interpretation of Title VI. When faced with a First Amendment claim by a university to be exempted from Title VI’s color-blindness rule, the Court’s conservatives will be forced to confront the doctrinal conflict they have created. They will have two choices. The first will be to apply the First Amendment doctrine they have created to weaken their pronouncements on color blindness. The other will be to weaken or limit their First Amendment pronouncements to save their allegiance to color blindness.

This helps identify another potential cost. This cost is different from the usual ones weighing in the balance, because this one would flow from a victory. If universities were to prevail on First Amendment claims, their victories could give rise to claims on behalf of institutions or businesses seeking to invoke expressive or associational claims to discriminate in nefarious ways on the basis of race. In other words, the slippery-slope worries expressed by several Justices in the 303 Creative argument, and in Justice Sotomayor’s dissent, might come to pass.²⁵³ Moreover, to those who think that the entire line of First Amendment cases from Hurley through 303 Creative is horribly misguided, building on it, even for honorable reasons, should be avoided.

This is a serious concern that should not be minimized. The premise of a university suit would be that a university’s use of race to repair and redress past wrongs has expressive value. It is not hard to imagine a white-supremacist group asserting that its belief in racial superiority is every bit as sincere and genuine and every bit as protected. Any assertion that the former belief is in some sense “better” than the latter might itself be problematic under the First Amendment.

Having said that, there are some reasons to think that the downside risk of opening the floodgates to racist litigants is less than one might think at first glance.

The doctrine itself provides a constraint. For a business to be covered, its expressive claim would have to be based on a sincere political or religious belief against, for example, serving Black customers or hiring Black employees. While the Court provides a high level of deference to the claimants’ own descriptions of their beliefs, the claimants would have to be engaged in some kind of expressive or associational activity that is cognizable under the Free Speech Clause. Universities and colleges are distinctly expressive in ways that most businesses are not.²⁵⁴ To protect, for example, Berea College’s right to define who is invited into its campus community would not necessarily lead to a holding that, for example, Ollie’s Barbecue has a right to discriminate against Black customers. A college is not a barbecue joint, and the First Amendment need not ignore the difference.

This would be especially true if a holding in favor of a college or university claimant is rooted partially in the unique attributes of a university. If so, the slippery-slope worries could be cabined to the world of higher education. And in that case, one could make the coarse utilitarian calculus of comparing the benefits of protecting the ability of scores or hundreds of institutions wanting to use race to further remedial, reparative, or educational goals against the much lower number of universities that would seek to use race to further invidious ends. Few if any universities in America purport to follow an avowedly racist ideology; even Bob Jones University no longer subscribes to the views it once pressed.²⁵⁵

Finally, a potential university claimant would want to weigh in the balance the possible benefits of a loss. Admittedly, possible constitutional litigants do not usually find themselves pondering the upside of defeat. But because a First Amendment challenge to Title VI involves a conflict between two lines of cases that have both created harmful effects, a loss could be a doctrinal win. That is, a loss in a potential university case might mean that the doctrinal evolution of the First Amendment allowing more and more claimants to gain exemptions from public-accommodation laws had been arrested.

Said another way, there are two possible outcomes to a lawsuit by a university: A victory saves the ability of some institutions to use race in admissions. A ruling against the universities would likely be based on the Court’s unwillingness to extend the Hurley/Dale/303 Creative line of cases to cover associations and institutions that want to discriminate on the basis of race. This is the slippery-slope concern that has been present for years and that the Court has studiously avoided. A case brought by a university would force the Court to address it. If the university wins, Harvard is in effect overturned for private schools with an expressive or associational interest in using race. If the university loses, the potential expansion of Hurley/Dale/303 Creative is stopped.

In that respect, the universities cannot lose. They either take advantage of bad First Amendment doctrine to save race-conscious policies or they make the doctrine less bad.

This article has evaluated possible First Amendment claims that private colleges and universities could bring to win an exemption from Title VI’s new color-blindness requirement. For private schools with expressive or associational interests in defining their campus community with reference to race—for reasons of social justice, reparation, educational richness, or otherwise—they can bring claims that fit squarely within the line of precedents starting with Roberts and stretching to 303 Creative. The Court has long recognized that anti-discrimination laws and public-accommodation laws must yield to claimants whose expressive and associational interests are burdened by those laws.