MacRae v. Mattos

Opinion

SUPREME COURT OF THE UNITED STATES KARI MACRAE v. MATTHEW MATTOS, ET AL. ON PETITION FOR WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 24–355. Decided June 30, 2025

The petition for a writ of certiorari is denied. Statement of JUSTICE THOMAS, respecting the denial of certiorari.

Hanover Public Schools and two of its officials (collectively , respondents) fired petitioner Kari MacRae for her pre-employment political expression on the social-media platform TikTok. Through her personal account, MacRae had “liked, shared, posted, or reposted” six memes—images or other items that are “ ‘spread widely online’ ”—expressing her views that immigration laws should be enforced, that an individual’s sex is immutable, and that society should be racially color-blind. 106 F. 4th 122, 126–128, and n. 1 (CA1 2024). After her firing, MacRae sued respondents for “retaliating against her for exercising her First Amendment rights.” Id., at 130. But, the District Court granted summary judgment to respondents, and the First Circuit affirmed , finding that MacRae had not established a protected First Amendment interest under this Court’s framework for public-employee speech. Because her petition for a writ of certiorari does not squarely challenge the First Circuit’s application of that framework, I agree with our decision to deny it. I write separately, however, to raise serious concerns about the First Circuit’s approach.

Our precedents establish that “the First Amendment protects a public employee’s right, in certain circumstances, to speak as a citizen addressing matters of public concern.” Garcetti v. Ceballos, 547 U. S. 410, 417 (2006). Although “[g]overnment employers, like private employers, need a significant degree of control over their employees’ words

2

and actions,” they can regulate their employees’ private speech about “matters of public concern” only to the extent “necessary . . . to operate efficiently and effectively.” Id., at 418–419. Under the so-called Pickering-Garcetti framework , whether such speech is protected turns on a balancing test, wherein the employee’s speech interest is weighed against the government’s interest as an employer in avoiding workplace disruption. See Garcetti, 547 U. S., at 419; Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U. S. 563, 568 (1968).

This case turns on the balancing component of the Pickering- Garcetti framework. All agree that MacRae’s TikTok posts qualify as speech on matters of public concern, but the First Circuit concluded that the balance of interests favored respondents . That court first discounted the value of MacRae’s speech interest because her posts, which are reproduced below, at times spoke in what the court described as a “mocking, derogatory, and disparaging manner.” (See Figures 1 and 2.) 106 F. 4th, at 137; see Pet. for Cert. 7 (reproducing posts).

Figure 1. Kari MacRae TikTok posts

3

Figure 2. Kari MacRae TikTok posts

In contrast, the First Circuit explained that respondents —who fired MacRae out of “concer[n] about the potential negative impact [her] social media posts would have on staff and students”—had a “ ‘strong’ ” interest in avoiding disruption, and that they made a “reasonable prediction of disruption.” 106 F. 4th, at 130, 137–138. The court pointed to factors such as the public attention and news coverage MacRae had received in light of her position on a neighboring town’s school board, as well as the fact that at least some Hanover students and staff were aware of her posts. Id., at 139–141. It also cited the fact that “some of her Tik- Tok posts (at least arguably) conflicted with the District’s belief of ‘[e]nsur[ing] a safe learning environment based on respectful relationships’ and Core Value of ‘[r]espect[ing] . . . human differences,’ ” “given the potential to perceive some of her posts as transphobic, homophobic, or racist.” Id., at 139–140. The First Circuit concluded that, on balance , the risk of disruption outweighed MacRae’s interest.

The First Circuit’s analysis strikes me as deeply flawed.

To start, I do not see how the tone of MacRae’s posts can

4

bear on the weight of her First Amendment interest. “Speech on matters of public concern is at the heart of the First Amendment’s protection.” Snyder v. Phelps, 562 U. S. 443, 451–452 (2011) (internal quotation marks and alterations omitted). And, “[t]he inappropriate or controversial character of a statement is irrelevant to the question whether it deals with a matter of public concern.” Rankin v. McPherson, 483 U. S. 378, 387 (1987). “[H]umor, satire, and even personal invective can make a point about a matter of public concern.” De Ritis v. McGarrigle, 861 F. 3d 444, 455 (CA3 2017) (internal quotation marks omitted). Accordingly, we have declined to “affor[d] less than full First Amendment protection” even for speech that we have deemed “particularly hurtful,” such as the picketing signs used by the Westboro Baptist Church. Snyder, 562 U. S., at 454–456; see id., at 454 (listing, among other Westboro signs, placards reading, “ ‘God Hates the USA/Thank God for 9/11,’ ” “ ‘God Hates Fags,’ ” and “ ‘Thank God for Dead Soldiers’ ”).¹ Against this backdrop, I do not see how the First Circuit could discount the First Amendment value of MacRae’s comparatively mild posts, all of which reflected positions that represent “by no means an isolated segment of public opinion.” Noble v. Cincinnati and Hamilton Cty. Public Library, 112 F. 4th 373, 382 (CA6 2024).

The First Circuit’s analysis of respondents’ countervailing interest in avoiding disruption is similarly questionable . Although this Court has “consistently . . . given substantial weight to government employers’ reasonable

5

predictions of disruption,” the key word here is “reasonable .” United States v. Treasury Employees, 513 U. S. 454, 492 (1995) (Rehnquist, C. J., dissenting). The First Circuit accordingly should have discarded factors whose disruptive potential was purely speculative, such as the fact that “ ‘some students and staff . . . were aware of ’ [MacRae’s] posts” or that “students [were overheard] discussing her social media activity.” 106 F. 4th, at 139–140.

Even worse, the First Circuit compounded its reliance on speculative factors with consideration of illicit ones. We have made clear that the core First Amendment principle of viewpoint neutrality applies in the Pickering-Garcetti context as elsewhere. See Rankin, 483 U. S., at 384 (“Vigilance is necessary to ensure that public employers do not use authority over employees to silence discourse . . . simply because superiors disagree with the content of employees’ speech”). Yet, the First Circuit cited an arguable conflict between MacRae’s posts and institutional expressions of viewpoint such as Hanover’s “Core Value of ‘[r]espect[ing] . . . human differences’ ” as evidence of potential disruption. 106 F. 4th, at 139. It undermines core First Amendment values to allow a government employer to adopt an institutional viewpoint on the issues of the day and then, when faced with a dissenting employee, portray this disagreement as evidence of disruption. And, the problem is exacerbated in the case of an employee such as MacRae, who expressed her views only outside the workplace and before her employment.

Whatever the proper weight of respondents’ interest in minimizing disruption, the First Circuit failed to conduct a proper balancing inquiry because it improperly discounted MacRae’s First Amendment interest. To its credit, that court recognized that “[t]he government employer’s interest must be proportional to the value of the employee’s speech.” Id., at 136; see Connick v. Myers, 461 U. S. 138, 152 (1983) (“[A] stronger showing may be necessary if the employee’s

6

speech more substantially involved matters of public concern ”). But, because the court viewed MacRae’s interest as “weigh[ing] less than it normally would,” it did not hold respondents to their full burden. 106 F. 4th, at 137.

This case is the latest in a trend of lower court decisions that have misapplied our First Amendment precedents in cases involving controversial political speech. See, e.g., L. M. v. Middleborough, 605 U. S. ___, ___–___ (2025) (ALITO, J., joined by THOMAS, J., dissenting from denial of certiorari) (slip op., at 6–13). And, a concerning number of these cases have arisen in the context of the Pickering-Garcetti framework. See, e.g., Kennedy v. Bremerton School Dist., 586 U. S. 1130, 1132–1133 (2019) (statement of ALITO, J., respecting denial of certiorari) (explaining how “the Ninth Circuit’s understanding of the free speech rights of public school teachers is troubling”); Porter v. Board of Trustees of N. C. State Univ., 72 F. 4th 573, 586, 595 (CA4 2023) (Richardson, J., dissenting). If left unchecked, this number will likely increase: In many cases, government employers may find it convenient to attempt to “restric[t] . . . disfavored or unpopular speech in the name of preventing disruption.” Dodge v. Evergreen School Dist. #114, 56 F. 4th 767, 786 (CA9 2022). But, the Pickering-Garcetti framework plainly forbids using “the guise of protecting administrative interests” to “disfavor any particular view.” 56 F. 4th, at 785–787; cf. Mahmoud v. Taylor, 606 U. S. ___, ___ (2025) (THOMAS, J., concurring) (slip op., at 11) (recognizing , in the free-exercise context, that school claims of disruption must be scrutinized to avoid “giv[ing] schools a playbook for evading the First Amendment”).

Lower courts are bound to apply the Pickering-Garcetti framework as we have articulated it.² I have serious con-

7

cerns about how the First Circuit applied it here. But, rather than raise these broader issues, MacRae’s petition focuses on the discrete question whether the framework’s balancing test applies at all in the context of a public employee’s “unrelated, preemployment speech.” Pet. for Cert. i. Because I agree with the Court that this question does not independently warrant review, I concur in the denial of certiorari. In an appropriate case, I would make clear that public employers cannot use Pickering-Garcetti balancing generally or unsupported claims of disruption in particular to target employees who express disfavored political views.

Footnotes

1. 1Although Snyder was not a Pickering-Garcetti case, we grounded our analysis in caselaw from the public-employer context. See 562 U. S., at 451–455. And, our Pickering-Garcetti cases have not treated the tone or style of an employee’s speech as bearing on its First Amend- ment value. Cf. Rankin, 483 U. S., at 379–380, 386–387 (recognizing, without qualification, that a “remar[k], after hearing of an attempt on the life of the President, ‘If they go for him again, I hope they get him,’ ” “dealt with a matter of public concern”).
2. 2This obligation does not mean that the Pickering-Garcetti framework is necessarily correct as a matter of original meaning. Given that the historical rule was that “a public employee had no right to object to [em- ployer-imposed] restrict[ions on] the exercise of constitutional rights,” there is good reason to think it may not be. Connick v. Myers, 461 U. S. 138, 143 (1983). But, “unless and until this Court revisits it, [the Pick- ering-Garcetti framework] is binding precedent that lower courts must faithfully apply.” L. M. v. Middleborough, 605 U. S. ___, ___ (2025) (THOMAS, J., dissenting from denial of certiorari) (slip op., at 1).