Expressive Discrimination: Universities' First Amendment Right to Affirmative Action — Part 4

The final post in my serial blogging on whether private universities can use a Boy Scouts expressive association theory to have race-based affirmative action.

Last week, I started blogging (and Monday and Tuesday, I continued blogging) about my new article Expressive Discrimination: Universities' First Amendment Right to Affirmative Action, just published by the Florida Law Review. In these days of federal attacks on private DEI, maybe some private universities might find this useful as a strategy for fighting back against the Trump Administration!

Monday's post gave the main thesis, laying out under what conditions a Boy Scouts expressive association right might be viable under current caselaw. Tuesday's post laid out some complications of the theory, answering four questions: (1) Does the race context make anything different? (No.) (2) Does the market context make anything different? (No.) (3) Does the Title VI conditional-spending context make anything different? (Yes, some.) and (4) Can public universities use this theory? (No.) Today's post asks: "What are the limits of this theory?", and how some lower courts have pushing its boundaries.

If you want to see the PDF with all the juicy footnotes, read the whole thing.

III.  What Are the Limits of This Theory?

A. Where the Doctrine Is

Let's recap. A lot is uncertain about the Boy Scouts doctrine. I've suggested how universities can look more like the Boy Scouts—make faculty and students into mouthpieces for inculcating university values. That might involve abandoning some academic-freedom norms (at least as concerns diversity values); many universities might not want to adopt this modest proposal, for good and obvious reasons.

Can a university take advantage of this theory without becoming a Boy Scouts clone? Perhaps, but it's hard to know for sure, because Boy Scouts is the only Supreme Court case that clearly shows how expressive-association norms can overcome antidiscrimination law.

Rumsfeld v. FAIR says, though, that we get no exceptions when the behavior isn't inherently expressive, i.e., when the behavior can be engaged in nonexpressively and you need words surrounding the conduct to explain that the behavior is expressive. This is what guarantees the constitutionality of antidiscrimination laws in most circumstances.

And the lesson of Roberts is that even when an organization engages in expression, it must show a substantial burden on its ability to express its message—which is difficult with an antidiscrimination law, except if the law alters the message or affects the choice of leaders or authority figures who speak for the organization.

B. Where the Doctrine Might Go

Where are some of flexibilities in the doctrine—i.e., where can we imagine that the doctrine might move in the future? Here, I discuss three areas: what constitutes a substantial burden, whether a governmental interest is compelling, and what activities count as expressive. The easier it is to find a substantial burden, the less compelling the governmental interest, and the more activities are expressive, the more likely an affirmative action program—even outside of universities—will be to survive under the expressive-association theory.

1. Substantial Interference

One area is what it takes to show substantial interference with an organization's expression. The Supreme Court held in Associated Press v. NLRB that a newspaper couldn't fire a reporter for his pro-union activity, but the Washington Supreme Court held in Nelson v. McClatchy Newspapers that a newspaper could fire a reporter for violating its policy against engaging in high-profile political activity. In both cases, the newspaper had asserted an interest in maintaining the appearance of unbiased and impartial reporting. Of course, the U.S. Supreme Court is supreme; but this shows that a lot depends on how willing a court is to accept that there's a real burden in a particular case.

For instance, in McDermott v. Ampersand Publishing, LLC, the Ninth Circuit held that a newspaper was entitled to a First Amendment exemption from the National Labor Relations Act, but this was against the background of a strike that was largely motivated by reporters' desire to assert their independence from the newspaper owner's editorial control. (Properly, this is somewhat less than a holding: under the procedural posture in this case, the court decided not whether the First Amendment protected the newspaper's activity, but just whether there was "at least some risk" of a First Amendment violation.)

This also relates to whether organizations can choose their listeners. Usually, it doesn't burden an organization's message to prevent it from discriminating between different groups of listeners. But what if the organization's message is tailored to a certain group in a mission-relevant way, as in the Nation of Islam cases? As I've noted above, there are various sources on point from a state court and a lower federal court, but none from the Supreme Court. If the Court decides to move toward more strong-form expressive-association protection, one way to do so will be to say that certain types of antidiscrimination laws are per se burdens on an association's expression.

2. Compelling Government Interests

Another potential flexibility is in how courts characterize the governmental interest that might overcome an associational claim. I've said earlier that overcoming discrimination is generally considered a compelling interest, regardless of the basis of the discrimination, and the market context may not make a difference.

But at least one lower-court case has questioned the government's nondiscrimination interest by characterizing it narrowly. The Gay Softball World Series had a rule permitting a "maximum of two Heterosexual players" on any participating team. This rule violated a Washington statute prohibiting sexual-orientation discrimination. However, the North American Gay Amateur Athletic Alliance asserted an expressive-association right to keep its teams predominantly gay, lest its pro-gay-athletics message be diluted.

The district court agreed with the expressive claim (see the next Subsection). But the court also examined "whether or not [the athletic league's] interest in expressive association outweighs the state's interest in eradicating discrimination." It acknowledged the Supreme Court's past endorsement of "the State's compelling interest in eliminating discrimination against women." But it went on to state that "[p]laintiffs have failed to argue that there is a compelling state interest in allowing heterosexuals to play gay softball."

Gay softball is admittedly different from a broadly intellectual organization such as a university. But the example illustrates a fact that is unsurprising and familiar in the constitutional law world: whether an interest wins depends on the level of generality at which a judge can be convinced to characterize it.

3. Inherently Expressive Activities

Another question is what counts as expressive. As the 303 Creative Court wrote, "Doubtless, determining what qualifies as expressive activity protected by the First Amendment can sometimes raise difficult questions." Though the Court denied that that "pure speech" case was one of the difficult ones, other cases might test the limits of the concept.

Consider, for instance, the gay softball case discussed above, where the gay athletic league won the right to discriminate in favor of gay players. Consider, too, these food distribution cases:

• Fort Lauderdale Food Not Bombs wanted to share free vegetarian food in a public park to communicate its message that "society can end hunger and poverty if we redirect our collective resources from the military and war and that food is a human right, not a privilege, which society has a responsibility to provide for all." The city wanted to regulate this under food-distribution and other regulations. The group asserted that its food sharing was expressive, and the Eleventh Circuit agreed.
• Adherents of the Krishna consciousness religion wanted to serve students vegan lunch at the University of California, Los Angeles (UCLA) in conjunction with literature distribution and religious activities. UCLA refused to grant permission to conduct Krishna Lunch. The group asserted that its vegan-food distribution, in context, was expressive and that UCLA's restriction violated its expressive-association rights; the Ninth Circuit agreed (in an unpublished opinion) that this claim was plausibly pleaded.

Consider also these donation cases, the second of which relates to affirmative action:

• Amazon's AmazonSmile program allowed a customer to direct 0.5% of the price of their Amazon purchase to "an eligible charity selected by the customer." Amazon chose to exclude charities that were designated as "hate groups" by the Southern Poverty Law Center. Coral Ridge Ministries Media, a Christian group classified as a hate group because of its opposition to homosexuality, sued Amazon under Title II for religious discrimination. Amazon argued that its "choice of what charities are eligible to receive donations" is "expressive conduct." The Eleventh Circuit concluded that forcing Amazon "to donate to an organization it does not wish to support" would "modify the content of [Amazon's] expression" and violate the First Amendment.
• The Fearless Foundation "seeks to bridge the gap in venture capital funding for women of color founders building scalable, growth aggressive companies. To bridge this gap, the Foundation operates the Fearless Strivers Grant Contest . . . , which awards $20,000 grants to small businesses owned by Black women." This contest was challenged for violating § 1981, but the Foundation asserted an expressive association right to donate to groups that are consistent with its mission. The Foundation won in district court, but lost in the Eleventh Circuit.

And here's an affirmative action case related to scholarship, though not at universities:

• Health Affairs, published by Project Hope, is a prestigious medical journal that has a "Health Equity Fellowship for Trainees, which provides mentorship and publication opportunities for health policy scholars." White applicants were excluded from applying for this fellowship. A would-be applicant argued that this restriction violates Title VI (since Project Hope accepts federal funding) and other antidiscrimination laws. But the journal asserted an expressive-association interest in maintaining race-conscious selection criteria. The fellowship "exists to increase the quality and quantity of health policy research focused on racial equity and authored by members of racial and ethnic groups that have historically been underrepresented in Health Affairs": "[d]iverse scholars produce research with unique insight and value as to racial equity issues," and the race-conscious criteria "cannot be separated from the overall expressive goals of Project HOPE and Health Affairs in the health equity field." The journal dropped its race-restrictive criteria, and the case was voluntarily dismissed.

The Health Affairs case is about policy research—i.e., writing. That's expressive. But . . . softball, food distribution, charitable donations? There's some lower-court willingness to find that various non-pure-speech activities are expressive enough to merit First Amendment protection, whether under a free-speech or an expressive-association theory.

Are these cases consistent with Supreme Court case law? On the one hand, if you can't make nonexpressive conduct expressive by saying it's ideological, softball and food distribution will probably have a hard case.

On the other hand, we should read the no-unilateral-characterization rule in context. In determining whether conduct is expressive, there's a subjective component—"[a]n intent to convey a . . . message"—and an objective component—whether "in the surrounding circumstances the likelihood was great that the message would be understood by those who viewed it." (Though "[a]n intent to convey a particularized message" certainly helps, the "particularized" component is optional, because we need to also protect Jackson Pollock, Arnold Schoenberg, and Lewis Carroll.) Even disregarding explanatory words, some conduct is already expressive just because of how it's socially perceived. "[C]ontext may give meaning to [a] symbol," from flags to black armbands.

Thus, flag burning's expressive nature is "overwhelmingly apparent," whereas "the point of requiring military interviews to be conducted on the undergraduate campus is not 'overwhelmingly apparent.'" Consider parades:

If there were no reason for a group of people to march from here to there except to reach a destination, they could make the trip without expressing any message beyond the fact of the march itself. . . . Parades are thus a form of expression, not just motion, and the inherent expressiveness of marching to make a point explains our cases involving protest marches.

How do we know that a parade is expressive and not just "march[ing] from here to there"? Even without reading the signs, we can observe that people are walking in a large group and holding signs—which invokes a social expectation of a protest, an expressive act.

This is a reasonable, moderate position: you can't make conduct expressive by talking about it, but conduct isn't necessarily nonexpressive just because there exist potential nonexpressive explanations. We look, objectively, to how it's likely to be perceived in the current social context—Justice Thomas's concurrence in Masterpiece Cakeshop, explaining why a wedding cake is expressive, suggests just how contextual such an analysis might be. These social expectations extend not only to particular acts but also to a particular medium's role—for instance, when a medium is forced to accommodate someone else's message. Newspaper owners and parade organizers are speakers, not conduits for other people's speech, but the answer is the opposite when it comes to cable systems or shopping mall owners.

The Supreme Court hasn't been very clear about precisely how to do this expressive-conduct inquiry. This gives courts leeway in finding expressive conduct in a variety of circumstances. It also suggests where litigants might want to act strategically to increase the chances that a court will find their conduct artistic or expressive. And it suggests where the Supreme Court may have flexibility in the "brutally difficult line-drawing exercises" this area of doctrine will require. Maybe the gay softball case went too far. Maybe food distribution is more arguable: distributing food in a park or on a university campus isn't itself communicative, but perhaps doing so in the context of handing out literature and doing religious observances would be perceived differently. (It's not that the government couldn't regulate it, but it would be evaluated under O'Brien intermediate scrutiny.)

What about cases involving giving money? Exchanging money for goods and services is usually nonexpressive: "[T]he First Amendment does not generally protect liberty of contract," and giving money for something of value is overwhelmingly likely to be perceived as merely one's attempt to get the thing. Thus, if a venture capital fund invested only in black-owned businesses—giving money for voting rights, cash flow rights, or both—this would violate § 1981 and wouldn't be considered expressive for First Amendment purposes.

But what about giving money for free, no strings attached?

Asking for donations is of course expressive, but that's because the asking is itself speech. Some cases have characterized donations as expressive, but that was where money was given to facilitate speech (e.g., campaign finance). This is consistent with the expressive-association right's limitation to people associating in order to engage in conduct that's itself already expressive.

Similarly, compelled contributions have been characterized as impingements on associational rights—but, again, that was in the context of compelled contributions to someone else's expression, such as required donations to a union's political activity.

That said, there is some support for the idea that donations can be expressive as such. Buckley v. Valeo said that "[a] contribution serves as a general expression of support for the candidate and his views" and spoke of "the symbolic expression of support evidenced by a contribution." This was of course in the context of political campaigns (i.e., speech). But the Court later reasserted the same idea in a more general context, involving donations to a variety of charities, not all of which were speech-related.

Ultimately, it comes down to how donations are socially perceived. Merely giving money to a charity is usually perceived as just wanting to promote the charity's work. Merely giving money to a black-woman-owned business could be perceived as just supporting that business's work, rather than as a broad social statement about the importance of black-woman-owned businesses in society. But giving out that money as part of an organized and widely advertised campaign can be perceived differently.

The lower-court post–Boy Scouts activity shows that courts have been flexible on what can be considered expressive. If that hurdle isn't cleared, then any expressive-association theory is a non-starter. But if a court does find expressive activity, a Boy Scouts–based expressive-association theory is more likely to prevail.

Conclusion

The primary moral of this story is: When one door closes, another door opens.

Because of Students for Fair Admissions, a lot of race-conscious affirmative action might now be illegal under Title VI and § 1981, but at least some private universities might have a constitutional right to do it anyway, based on the expressive-association right recognized over twenty years ago in Boy Scouts and recently beefed up in the compelled-speech context in 303 Creative. Some high points:

• The doctrine should carry over to the racial context.
• The market context should be irrelevant.
• The government should retain the ability to induce race neutrality using its conditional spending under Title VI. But the unconstitutional conditions doctrine should guarantee that the threat of pulling funding is limited to the specific units that discriminate, instead of affecting the entire institution. As a result, universities should retain substantial ability to engage in race-conscious affirmative action, at least in some units.
• This theory won't help public universities.

Many of those who like affirmative action may dislike the Boy Scouts doctrine and may hesitate to be very (303) Creative with it. But (to borrow an expression from the petitioner in Rumsfeld v. FAIR) you litigate with the case law you have, not the case law you wish you had.

The secondary moral of this story is: Mind the "at least some private universities might" that I italicized above. Despite fears that Boy Scouts would massively undermine antidiscrimination law, the actual Boy Scouts exception is likely fairly narrow. To most cleanly fit within it, those covered by an expressive-associational right to affirmative action should be authority figures who speak for the institution. The closest fits will be at universities that do have strong faculty- and student-governance norms but don't have strong faculty or student academic-freedom or free-speech norms, so their faculty and students can really be described as, in some way, leaders who speak for the institution.

That's not to say that other universities won't be able to qualify for the Boy Scouts exemption, but the looser the resemblance with the Boy Scouts, the less sure we can be that the doctrine will apply.

And a final moral is: Let's look, based on post–Boy Scouts lower-court cases, at where the doctrine is flexible and where it might be headed. There's some flexibility in when a burden on an organization becomes "substantial" and also in whether some antidiscrimination interests that the government might present to overcome the expressive-association right would be considered "compelling."

But the main question is whether certain activities are considered "expressive" to begin with. Because the test for whether an activity is considered expressive relies a lot on social expectations and how people interpret an activity, this element is key to whether any expressive affirmative-action right can be viable outside the educational context.